Developing An Industrial Chemical Process: An Integrated Approach [1 ed.] 0849313600, 9780849313608

Citation preview

DEVELOPING an INDUSTRIAL CHEMICAL PROCESS An Integrated Approach

Copyright © 2002 by CRC Press LLC

DEVELOPING an INDUSTRIAL CHEMICAL PROCESS An Integrated Approach

Joseph Mizrahi

CRC PR E S S Boca Raton London New York Washington, D.C. Copyright © 2002 by CRC Press LLC

Library of Congress Cataloging-in-Publication Data Catalog record is available from the Library of Congress

This book contains information obtained from authentic and highly regarded sources. Reprinted material is quoted with permission, and sources are indicated. A wide variety of references are listed. Reasonable efforts have been made to publish reliable data and information, but the author and the publisher cannot assume responsibility for the validity of all materials or for the consequences of their use. Neither this book nor any part may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, microfilming, and recording, or by any information storage or retrieval system, without prior permission in writing from the publisher. The consent of CRC Press LLC does not extend to copying for general distribution, for promotion, for creating new works, or for resale. Specific permission must be obtained in writing from CRC Press LLC for such copying. Direct all inquiries to CRC Press LLC, 2000 N.W. Corporate Blvd., Boca Raton, Florida 33431. Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation, without intent to infringe.

Visit the CRC Press Web site at www.crcpress.com © 2002 by CRC Press LLC St. Lucie Press is an imprint of CRC Press LLC No claim to original U.S. Government works International Standard Book Number 0-8493-1360-0 Printed in the United States of America 1 2 3 4 5 6 7 8 9 0 Printed on acid-free paper

Copyright © 2002 by CRC Press LLC

Preface This book presents a detailed discussion of the issues that have to be addressed, in most cases, in the development and the first implementation of a novel industrial chemical process. These issues start with the “whys” and “wheres,” then address the working organization and all the different steps, activities, and reviews in the process development program, and finally in the implementation, design, construction, and start-up of a new plant.

Why is such book needed at all? This specific field of activity is constantly occupying many thousands of managers, scientists, engineers, chemists, specialists, economists, and technicians. These professionals work in industrial corporations, research organizations, universities, engineering companies, equipment suppliers, statutory public functions, to name a few, in many countries around the world. The result of their activity has been hundreds of new processes and new plants in the chemical industry every year. Nevertheless, at present, there seem to be no recognized professional standards, no generally accepted written procedures, or even a book covering this professional field. Quite different working practices are implemented in different corporations and in different countries. Thus, any professional who encounters some of these issues for the first time in his job can only rely on the direct teaching of his boss and colleagues. And in that lottery some have more luck than others. Strangely enough, up until now, the knowhow in this important professional sector has been transmitted only by “apprenticeship.” Somehow, novel processes have been finally developed and used in new plants that have been built and operated, most of them successful. But, on the other hand, many case stories are widely spread in the profession about all the associated problems, serious waste of time and resources, start-up troubles, and occasionally complete failures. These problems have been generally attributed to personal errors in specific situations, possibly to the individualistic characters of the inventors and promoters, and to the opportunistic demand for quick results in new processes. Such explanations could only be true for the initiation stage

Copyright © 2002 by CRC Press LLC

(possibly 5% of the efforts invested), but cannot hold for all the development and implementation work. So, a systematic study of the common aspects to most projects can be instructive. This book is intended primarily for those professionals who are already on the job in real life, to help them, hopefully, to do a better and more efficient job, to be happier by understanding more about what is going on around them, and to reduce the frustrations associated with this line of work. It is assumed that the readers will be graduates with some professional experience, who have access to all the textbooks, handbooks, and publications available, to Chemical Abstracts and to the Internet, and who know how to use these. So, this book will not be competing with these sources and will not copy what is readily available. At most, it will refer the readers to the more useful sources, in this author’s opinion. The suppliers of commercial services have essential contributions to such projects, and the general issues connected with the selection of such suppliers are discussed, but no particular reference is given as far as possible. The other references direct the readers, who may be interested in any of the example cases mentioned, to more detailed sources. Also, in this book, with due apologies to the chemists, a chemical process does include any physical or mechanical transformation or separation which is necessary to obtain the final products. On the face of it, the development and implementation of a new chemical process may appear to be a matter of chemistry, materials, equipment, control, etc., but it should be recognized that this is a very complex endeavor, and its success depends, in fact, mostly on the interactions and organization of many different people in various positions. In each such project, hundreds of professionals are concerned, full-time or part-time, with the research organization, the various functions in the corporation, the engineering company, the equipment suppliers, patent attorneys, specialist consultants, and civil servants with statutory functions. These professionals are mostly chemical engineers, but all the related professions are also involved: managers (in particular in finance, production, and marketing), different fields of engineers, research and analytical chemists, various specialists, patent attorneys, lawyers, economists, and supporting technicians. The first need in a new project organization is to establish a common communication and reference system in which every participant in the project will understand the point of view, the priorities, and the “jargon” of the others. This aim can require both patience and goodwill from everyone concerned and should be motivated by the example of the management. It is hoped that this book can be used for such purposes. The author has been occupied in this field of activity all of his professional life in many different positions. He strongly believes that a project involving the development and implementation of a new chemical process can be done better and more efficiently if: Copyright © 2002 by CRC Press LLC

• All the issues and all the interactions were discussed and understood from the beginning by all the participants • The limits of responsibility were clearly defined • A proper organizational structure and adequate programs were used The detailed recommendations in this book can be readily integrated, without any contradiction or competition, with the latest trends in corporate research and development (R&D) management procedure, such as the “Stage Gate” system and similar tools, which recently have been introduced in many large corporations. These detailed recommendations can assist the “Gate Keepers” in defining the “deliverables” and “criteria” to be achieved in the next “Stage.” All the engineers, scientists, and managers concerned with the development of a novel industrial chemical process, and/or with the implementation, design, construction, and start-up of a plant based on this process, can use this book to assist them in their work. The book will give them a general overview of all the issues ahead, and also provide them with checklists to draw up their own working programs, or at least understand the logic of the instructions given to them by their boss. Friends with experience have remarked that the scope of this book may appear to be very complex and its “message” may be confusing for rapid readers sampling here and there. Therefore, it was decided to add at the end of each chapter a short recapitulation of the issues that can be worth an additional thought and possibly further reading or discussion. At least, the core team of a project would benefit from a systematic study. Evidently, not everyone would be interested in all issues at one specific time, but it is nice to know that they can come back and consider more intensively any pertinent issue whenever they might face the need. Professionals with a few years of experience in this field, who may recognize some of the issues discussed from personal exposure, should benefit more. Part of the material in this book can also be used as a basis for an overall course for graduate students who are intending to start their work in industrial R&D, equipment development, process engineering, plant design, and managing functions in industrial corporations. It also can be used for workshops of continuing education for these working professionals. Obviously, one could have filled the book with examples from actual projects, but it is debatable whether more such particular examples would have helped illustrate the points or distract attention from the complex issues. Furthermore, most of the examples are covered by commercial secrecy and cannot be published. So, the compromise chosen here by the author may not satisfy every reader. The author will be pleased to receive any comment or suggestion that can help expand the usefulness of this book. Copyright © 2002 by CRC Press LLC

The author Dr. Joseph Mizrahi was born in 1933 and lives in Israel since 1951 at 27A Einstein Street, Haifa, 36014, phone (972-4) 824-4431, office phone (972-4) 826-0737, fax (972-4) 826-0797, email [email protected]. He holds B.Sc. and M.Sc. degrees in Chemical Engineering and a D.Sc. in Mineral Engineering from the Technion, Israel Institute of Technology in Haifa. In addition, he received the Diploma of Imperial College, London, 1965, and the professor-equivalent grade of Research Institutes Scientists. He also taught and was a postgraduate supervisor part-time at Technion from 1956 to 1979. Dr. Mizrahi has published 14 papers for international scientific conferences, 29 papers in international journals, has received 20 patents, and 24 communications to various professional conferences. He worked at the IMI Institute for Research and Development in Haifa from 1958 to 1974, first as a research engineer, then as head of the Chemical Engineering Department. His work included basic engineering design for process implementation, engineering aspects of licensing agreements, analysis of new processes, economic evaluations, surveys, worldwide liaison with engineering companies, piloting of new processes, run-in of new plants in foreign countries, and development and testing of new industrial contacting equipment. In addition, fundamental research was done under his supervision and published in the fields of mixing and separation of liquids and of hydrochloric acid technology. From 1974 to 1978, Dr. Mizrahi was Managing Director of Miles-Israel Ltd. in Haifa, a subsidiary of a multinational corporation in food, pharmaceutical, and speciality chemicals. This work included the completion of new plants, the introduction of new products to the world markets, and the stabilization and diversification of operations. From 1979 to 2001, he provided independent professional consulting services to corporations worldwide in the fields of organization and streamlining of R&D programs; consolidation, evaluation, and transfer of knowhow; initiation, organization, and evaluation of projects; process design of new plants; troubleshooting and expansion of existing plants; and analysis of corporate development strategy.

Copyright © 2002 by CRC Press LLC

Acknowledgments This book is dedicated to my wife, Sara, for a lifetime of motivation and support. I would like also to acknowledge: • The influence of Professor Avram Baniel from whom I learned very much in various forms of collaboration in many projects over more than 4 decades, since he founded and managed the pioneer team at the IMI Institute for R&D where I spent the first 16 years of my professional career. • The friendly and helpful reviews of the draft of this book by Ari Eyal, David Gonen, Chanoch Gorin, David Meir, and Tuvia Zisner. • The long and productive interaction over all my professional life with a large number of my friends and colleagues in many countries, the names of whom I cannot list in this limited space.

Copyright © 2002 by CRC Press LLC

Contents Chapter 1 Why a new industrial chemical process could be needed? 1.1 Changing world 1.2 A better quality product 1.3 Lower cost of production 1.4 Different raw material 1.5 Ecological pressure 1.6 New products for the corporation 1.7 Newly available industrial technology 1.8 New functions for new products 1.9 Corporate public image 1.10 Worth another thought References Chapter 2 Starting the development of a new process 2.1 Driving forces 2.1.1 Backing of a large corporation 2.1.2 Promoting group 2.1.3 The second part 2.1.4 Public authorities 2.2 How a new process is born 2.2.l Normal research and development activity 2.2.2 Personal motivation 2.2.3 Corporate function 2.2.4 Financial and commercial rewards 2.2.5 False starts 2.3 Explicit definition of the development project 2.3.1 Objectives and purposes 2.3.2 Patents 2.3.3 Possible industrial framework 2.3.4 Timetable 2.4 Different stages of a typical program 2.5 Corporate management procedures for new projects 2.6 Worth another thought

Copyright © 2002 by CRC Press LLC

Chapter 3 Essential resources needed for the development project: preceding implementation 3.1 Introduction 3.2 Specific managerial skills 3.3 Core project team 3.4 R&D laboratories and pilot installations 3.4.1 Company’s own laboratory and pilot installations 3.4.2 Outside laboratories and pilot installations 3.4.3 Analytical laboratories 3.5 Experts on marketing and on potential users 3.5.1 Particular terminology 3.5.2 Clients’ needs 3.5.3 Competition 3.6 Support from experts on hardware 3.6.1. Plant engineering and operation 3.6.2 Equipment design 3.6.3 Corrosion in construction materials 3.6.4 Operation and process control 3.7 Support from experts in software 3.7.1 Publication search and analysis 3.7.2 Intellectual property and secrecy 3.7.3 Patent application 3.7.4 Process modeling 3.8 Safety, public regulations, and waste disposal support 3.8.1 Safety 3.8.2. Public regulations 3.8.3 Waste disposal 3.9 Support of specific codes relevant to plant design and operation, and product quality 3.10 Economics 3.11 Development expense budget 3.12 Worth another thought References Chapter 4 Actual case examples 4.1 Nature and man: the Dead Sea 4.2 Magnesium chloride-based industries 4.3 Economic uses for the HCl by-product solutions 4.3.1 Strategic policy 4.3.2 Coupling of HCl-producing and consuming plants 4.3.3 Timing of implementation 4.3.4 Production of pure phosphoric acid 4.3.5 Technological difficulties 4.3.5.1 Materials of construction 4.3.5.2 Safe, stable conditions for solvent extraction in large mineral plants Copyright © 2002 by CRC Press LLC

4.3.5.3

Clean starting solution for solvent extraction 4.3.5.4 Recovery of the residual solvent from different exit streams 4.3.5.5 Large-capacity liquid–liquid contacting equipment 4.4 Phosphoric acid diversification processes 4.4.1 Different quality specifications 4.4.2 Solvent extraction opening 4.4.3 IMI “cleaning” process 4.4.4. “Close-cycle” purification process 4.4.5 Mixed process 4.4.6 New proposals 4.5 Citric acid by fermentation and solvent extraction 4.5.1 Conventional lime sulfuric acid process for citric acid 4.5.2 IMI-Miles solvent extraction process for citric acid 4.5.3 Newer solvent extraction process for citric acid 4.6 Preparation of paper filler by ultra-fine wet grinding of white carbonate 4.7 Worth another thought References Chapter 5 Process definition and feasibility tests 5.1 Translation of the idea into a process definition 5.1.1 Scope of the preliminary process definition 5.1.2 Comprehensive literature survey 5.1.3 Block diagram 5.1.4 Quantitative definitions of the different sections 5.1.5 Process calculations for the preliminary process definition 5.1.6 Presentation of one feasible implementation formula 5.1.7 Possible industrial implementation framework 5.1.8 Timetable 5.1.9 Important note 5.2 Critical and systematic review of the process definition 5.2.1 Review forum 5.2.2 Fundamental process issues 5.2.3 Patent situation 5.2.4 Profit potential 5.3 Design and execution of the feasibility tests 5.3.1 Purposes of the feasibility tests 5.3.2 Equilibrium conditions 5.3.3 Scale up of reactors 5.3.4 Physical separation operations 5.3.5 Scale-dependant and dynamic flow operations 5.3.6 Extreme conditions Copyright © 2002 by CRC Press LLC

5.3.7 Actual raw materials 5.3.8 Analytical difficulties 5.4 Analysis of the results from feasibility tests 5.5 Second review of the process definition 5.6 Worth another thought References Chapter 6 Experimental program 6.1 Basis 6.1.1 Experimental program purposes 6.1.2 Different sections 6.1.3 Quantitative data needed for process design 6.1.4 Format 6.1.5 Representative raw materials 6.1.6 Classification of missing data 6.2 Chemical equilibrium data 6.2.1 Vapor–liquid equilibrium system 6.2.2 Gas–liquid equilibrium system 6.2.3 Liquid–liquid equilibrium system 6.2.4 Solid–liquid equilibrium system 6.2.5 Reversible and nonreversible equilibrium 6.2.6 Chemical equilibrium laboratory tests 6.2.7 Experimental difficulties in chemical equilibrium tests 6.3 Dynamic flow conditions 6.3.1 Design data required 6.3.2 Simpler processes 6.3.3 Theoretical models 6.3.4 Special test rigs 6.3.5 Indirect methods 6.4 Scale-dependent operations 6.4.1 Vertical driving force depending on the hydrostatic height 6.4.2 Wall effect 6.4.3 Crystallizer 6.4.4 High-temperature equipment 6.4.5 Failure to recognize the wall effect 6.5 Reporting results from the experimental program 6.5.1 Frequent partial reports 6.5.2 Complete reports on the experiment part 6.5.3 Implications of the results 6.6 Worth another thought References Chapter 7 Preliminary process design for a particular proposal 7.1 Process team Copyright © 2002 by CRC Press LLC

7.2 7.3 7.4 7.5 7.6 7.7 7.8 7.9 7.10 7.11

Process flow-sheets Preparation of an overall detailed description Listing of all the main process streams Material and heat balances Material handling operations Summary tables for all required services Major pieces of process equipment Main operational and control procedures Listing of required staff Worth another thought

Chapter 8 Economic analysis of the specific proposal 8.1 Purpose 8.2 Preliminary estimate of the Fixed Capital investment (revision 0) 8.3 Estimate of operating costs 8.4 Expected net sales income estimate 8.5 Profitability calculation 8.6 Optimistic evaluation of the profit potential in other applications 8.7 Possible synergetic effects with other production facilities 8.8 Comprehensive report for the justification of the specific proposal 8.9 Contractual agreements 8.10 Worth another thought References Chapter 9 Working program toward a first implementation 9.1 Patent protection 9.1.1 Revised or additional applications 9.1.2 Extended geographical coverage of the patents 9.2 Detailed process design 9.2.1 Piping and Instrumentation Diagrams 9.2.1.1 Piping lists 9.2.1.2 Valves 9.2.1.3 Instruments 9.2.1.4 Control loops 9.2.1.5 Flanged manholes and hand-holes in closed pieces of equipment 9.2.1.6 Provisions for possible future connections 9.2.1.7 Non-conventional drives 9.2.2 Examples of portions of piping and instrumentation drawings 9.3 “Major” equipment packages 9.4 Pilot testing of specific process operations 9.4.1 Multiple-effects evaporator Copyright © 2002 by CRC Press LLC

9.4.2 Liquid–liquid contacting battery 9.4.3 Main problems for piloting 9.5 Modeling 9.6 Complementary bench-scale testing program 9.6.1 Detailed specification of the industrial equipment 9.6.2 Pilot installations 9.6.3 Process modeling 9.6.4 The design of instrumentation 9.6.5 Corrosion tests 9.6.6 Clarification of waste disposal issues 9.6.7 Clarifying process safety issues 9.7 Preparation of product samples for market field tests 9.8 Clarification concerning any formal permits needed 9.9 Worth another thought References Chapter 10 First implementation plant design: compromises and optimization 10.1 “First implementation” policy 10.1.1 Expected start-up problems 10.1.2 Design policy 10.1.3 Identifying probable causes of problems 10.1.4 “Guarantees” for reasonable plant performance 10.2 Modeling and optimization 10.2.1 Composition of raw materials 10.2.2 Effects of impurities 10.2.3 Changes in the kinetics of mass transfer 10.2.4 Changes in specifications for the final product 10.2.5 Normal fluctuations around the designed average 10.2.6 Differences in the performance of equipment 10.3 Critical pilot testing 10.4 The process package 10.5 The role of the engineering company in the first implementation of a novel process 10.5.1 The interests and limitations of the engineering company 10.5.2 The engineering company and the project manager 10.5.3 Specialization 10.5.4 The chemical process engineering department 10.5.5 Timetable 10.6 Detailed engineering documents 10.7 Final review and approval for construction 10.8 Worth another thought References

Copyright © 2002 by CRC Press LLC

Chapter 11 Running in and adjustments in the new plant 11.1 The plant construction period 11.2 Assembling and training the operating team 11.2.1 Recruitment 11.2.2 Maintenance 11.2.3 Training 11.2.4 Safety 11.2.5 Functional organization 11.3 Preparation for start-up 11.3.1 “Dry runs” 11.3.2 The plant manager 11.3.3 The construction manager 11.3.4 The project manager 11.4 Preparation with real materials 11.5 Strategic options for the running-in of the new plant 11.5.1 Possible causes of problems 11.5.2 Unsatisfactory results 11.5.3 Start-up strategies 11.6 Stabilization of production 11.7 Demonstration run and project success report 11.8 Optimization of operating conditions 11.9 Worth another thought Chapter 12 Consolidation of the new know-how 12.1 Updating the process know-how 12.2 Final revision of the Process Package 12.3 Updating the Operational Manual 12.4 Feedback from users in the market 12.5 Additional patent applications 12.6 New publications 12.6.1 Information on the competition 12.6.2 Publications on the new process and plant 12.7 How can this accumulated specific know-how be used again? 12.8 A final note: what have we learned? 12.9 Worth another thought Appendix 1 Typical organization and contents of a Process Package A1.1 General A1.2 Definition of “black box” objectives A1.3 Division of the process into sections as illustrated in a block diagram A1.4 Separate discussions for each section A1.5 Material and heat balances

Copyright © 2002 by CRC Press LLC

A1.6 A1.7 A1.8 A1.9

Equipment choices Services Materials of construction: options and preferences Safety aspects

Appendix 2 Functional organization structure of a typical development project A2.1 Successive stages A2.2 The invention and promotion stage A2.3 The process development stage A2.4 The construction and running-in period

Copyright © 2002 by CRC Press LLC

chapter 1

Why a new industrial chemical process could be needed? 1.1 Changing world The development of a new chemical process is a major technical, economical effort that can be justified only if it fills a definite need of an industrial corporation. The present chapter discusses the various situations in which such a need could be defined. This review allows one connected to the chemical industry to evaluate the probabilities that his/her corporation would need a new chemical process in the foreseeable future. There are basic reference books that can be used as sources for this initial information.1–5 The chemical industry has always been operated in a changing world with expanding markets, a need for better products at lower prices, change in raw materials, addition and removal of political barriers, great jumps in the technology available for industrial application, higher ecology demands, etc. As time goes on, the dynamic rate of such changes seems to be increasing exponentially. In the past 3 decades, in particular, it requires an open attitude from any corporate management towards possible process revision. In such a changing world, an operating chemical corporation could require a novel process for a certain product, if and when one (or more) of the objective situations discussed below becomes dominant and is recognized, at least inside the organization. Let us consider first the situation in which a corporation is already producing and selling the product, but now needs process changes for: • • • •

Obtaining a better quality product Reaching a lower cost of production Using different raw materials Responding to ecological pressures

Copyright © 2002 by CRC Press LLC

A different situation occurs when a corporation is considering making a new product. The company will need a new industrial process for: • • • • •

Producing according to a soon-to-expire patent “Bypassing” an existing patent Using a newly available industrial technology Creating new markets with a product fulfilling new functions Expanding its public image

1.2 A better quality product The need for a better quality product could be felt in one of the corporation’s existing markets and reported by the marketing organization. Such a need could arise from the persistent requests or complains of clients or from the pressures of competitors’ products, and it could be reflected in the presentation of more stringent standard purchasing specifications. Furthermore, an upgraded product could open the way to other market segments. This situation is quite common in the process industry, as a chain result from changes in the downstream uses of the products. It generally motivates a continuous effort in limited research and development (R&D) projects, resulting in gradual changes in the existing production technology, in an attempt to improve the product’s quality as requested. Such an aim could possibly be obtained, for example, by the addition of purifying operations to the production line, such as distillation, recrystallization, active-carbon decolorization, ion-exchange purification, and the like, or by compromising on the product’s yield in order to remove more impurities in the wastestreams. However, in many cases, a point is reached when further improvement would no longer be possible with the existing process or with the raw materials presently used, or when such quality improvement would become too expensive. At this point, the need for a significant process change will be recognized and defined inside the corporation, and such need could also be made public in the market segment. This significant process change would preferably be limited to the core production process, while almost all of the expensive infrastructure could most likely be maintained with minimum adjustments.

1.3 Lower cost of production Lower cost of production is, of course, always desirable in any existing plant, either to increase the profits or to allow lower and more competitive prices. In practice, in all operating plants, this objective is dealt with continuously by small and gradual ad-hoc steps, which do not impair the regular flow of production. There is not always a direct link between the production cost and the sale price, and there are even examples of plants that have been supplying an essential strategic corporate need while losing money. However, many Copyright © 2002 by CRC Press LLC

operating plants are living under the shadow of the possible development of a more efficient, completely new process with drastically lower production costs. This process may become available to the competition and may endanger the basic economic existence of the plant. Thus, corporations must always devote a continuous effort to keeping up to date with all the developments that could lead in this direction. These include higher yields, lower energy consumption, shorter route, revaluation of byproducts, etc. This could evolve into a full-scale process development effort, whenever a company intends to build a new plant to replace an old installation or when stronger protection is required against the perceived competition.

1.4 Different raw material In some cases, different raw materials may become available that could have definite technical or cost advantages. In other cases, a significant change could be expected in the future quality or in the cost of the raw materials that are presently used, or even in the continuation of their future supply. The changing situation concerning the raw materials’ supply has always characterized those industrial chemical processes that start with natural raw materials, i.e., mineral ores, agricultural crops, or petroleum fractions for the petrochemical industries. The situation could be even more sensitive when the raw materials from a plant are byproducts or waste products from the main production of another plant that is using such natural raw materials (i.e., grain hulls, molasses, mineral concentrate fractions, hydrocarbon streams, etc.). A similar situation relates to the use of some waste products from the combustion in large power plants (fly ash from coal, soot, solutions from ecological scrubbers, etc.) as the starting raw materials. For example, the world’s main supply of zirconium oxide (and zirconium compounds) for many decades came from a byproduct (Baddelayite concentrate) mined in South Africa. It has been known from the 1990s that this unique source was progressively and irrevocably being depleted6 and all the suppliers and users of zirconium oxide had to urgently look for new processes. The acute need directed the users’ attention to options for extracting zirconium oxide from the mineral Zircon (zirconium silicate), which is plentiful worldwide as a heavy-sand concentrate. Unfortunately for the developers, however, it also has a very stable mineralogical structure. To overcome this inherent stability, some proposed the use of brute force, such as fusion in an electric arc furnace at 2700˚C, followed by volatilization of silica fumes and other impurities (some of it radioactive) that had to be collected, or thermal dissociation by a shock treatment at very high temperatures in a plasma torch, followed by a wet treatment. Other proposals were based on sophisticated chemical detours by additive reactions with calcium or sodium oxide at relatively lower temperatures.7–10 The recently patented process, developed by Chanoch Gorin and Joseph Mizrahi9 for that purpose, presents an efficient novel route and will be discussed in Chapter 5 as an illustration of several development steps. The possibility of getting some Baddelayite supply from Copyright © 2002 by CRC Press LLC

a mine in Russia’s arctic Kola region, along with the rather small world market (in tons and in sales volume) also represent limiting factors in the development of these new processes. In an opposite situation, the exclusive and efficient production of highgrade synthetic potassium nitrate, according to the 1967 IMI solvent extraction process,11 has been a profitable operation for several decades as the principal worldwide supplier, despite the well-known existence of large natural deposits of nitrates in South America. Since the mining and refining operations have finally been established in Chile, the situation in this market changed throughout the world. Different grades of potassium nitrate are now available to different users at different costs and the consumption of the highest quality synthetic product has decreased. All of these changes called for drastic process reconsideration in the plants using the synthetic route. Such options for change had been available for at least 10 years,12–13 but there was no pressing incentive for a development effort. In the last few decades of the twentieth century, the fluctuations in the quality as well as the cost or the availability of many raw materials have often reflected the changes in international trade, as many political and customs barriers were added or removed. Examples of such changes are the decolonization of many countries, the European Union and other regional unions, the decentralization of the former Eastern block into separate countries and the accelerated privatization of their industries, as well as the increasing role of The Republic of China in all economic areas. All of these geopolitical changes have seriously affected the way in which many older chemical plants have been operated for generations, and have forced companies to reconsider their production processes and possibly how to develop alternative processes more related to the new situation. For example, raw (brown) cane sugar could be produced somewhere in Asia, transported to a European city to be refined and recrystallized, and then reexported around the world. Such activity could only have been developed in the past generations under the cover of heavy custom tariffs, which have finally affected the European consumers. But the gradual reduction of this practice in the future also will affect a series of downstream industries, which are linked to the byproducts of the sugar refinery in Europe (i.e., molasses or low-grade sweeteners). There are many similar examples in other fields and in other parts of the world. In addition, the new “global village” economy has led to many international corporate mergers and other “arrangements” that have affected the distribution of raw materials in different areas. This presently accepted practice constitutes a drastic change from the anticartel laws that were taken very seriously until recently in the American sphere of operation (at least in open references).

1.5 Ecological pressure Such pressures have been systematically applied in the last generation by public organizations and/or by statutory regulations in developed countries, to reduce Copyright © 2002 by CRC Press LLC

as much as possible the environmental damages caused by some existing chemical plants. In many cases, serious cleanup operations have been successful and all concerned, including the employees of these plants, were much relieved. In other situations, the response of the chemical industry to such pressures has been to “do something” that is not too expensive (mostly downstream effluent treatments), and to claim to have done “everything possible,” except for the ultimate closing of the plant, which is generally not desired by the community. In this continuing struggle, both sides are progressively improving their knowledge as more experts are called in. An underlying menace, however, is the occasional threat to move an industrial activity to another part of the world where ecological pressures are less demanding. In many situations, a mutually acceptable solution would evolve from a change in the source or quality of the raw materials. This would require a significant change in the main process, while retaining the plant’s entire expensive infrastructure. In such a case, the development of the new process has to be done within strict boundaries, but the know-how developed could eventually be applied in future plants. Another aspect of the ecological pressure relates to the combustion gases from fuel burning, either in cars or power stations. The effluent gases from cars have been dealt with more efficiently, in particular by auto industry improvements and through the supply of cleaner fuels from the petroleum refining industry. This necessitated the development of many new chemical processes (most of them still not published). This solution is not feasible for power stations, which are using mostly coal and the residual “dirty” petroleum heavy fractions. There an additional treatment must be done on the effluent gases on the way to the chimney to separate the SO2/SO3, NO/NO2, particulate matters, and possible poisonous metallic traces. Such treatment is complicated (from the chemical and technology points of view) and expensive, because gases need to be cooled and then saturated with water vapors. The resulting heavy white “plume” from the chimney would be much more visible and of concern to the surrounding population. This could also be corrected with the use of more heating and pressure, which would result in more energy and higher costs. If the chemical industry participated in such efforts, they could recover part of the costs from the marketing of, for instance, valuable ammonium sulfate and nitrate of fertilizer grade produced from the treatment of effluent gas. Many processes were proposed along these lines and are actively being considered, however, actively but slowly by the power station operators. (No references are given here, considering the actual commercial interests.)

1.6 New products for the corporation Let us consider now the situation in which the corporation has not been producing and selling the product, or a new corporation that is organized for such project. A corporation may have been prevented from entering into a specific production line that was well protected by a competitor’s existing patent. Such Copyright © 2002 by CRC Press LLC

patents could cover either the nature (analysis, specification) of the product or a specific production process for such product. These are different issues. If the existing patent covers the nature of the product, a process development effort would be required as soon as it is established that such patent would expire in a few years, or if a way to by-pass such protection can be proposed (e.g., by a small change in the formula that does not affect the performance). Note that the patent law prevents only the selling of the product covered by the patent, not the study or the preparation for its eventual production or even its production for storage. This situation has been typical, in particular, to the pharmaceutical industry, as so-called generic medicines are sold in the marketplace at reduced prices as soon as the basic patent covering the trademark medication has expired. This same tactic relates to the fine chemicals industry, producing patented chemical specialties, additives, resins, catalysts, etc. A patent covering a specific production process can generally be extended on and on, by additional filing of complementary patent applications based on the specific practical know-how that has been accumulated during the plant’s operation. This technique is not always effective, but it is widely used, mostly as a deterrent toward weaker, would-be competition. On the other hand, if such a competitor has a strong incentive and a good R&D team, a serious effort could possibly indicate some ways to avoid the formal definitions in the claims of these complementary patent applications. This would collapse the whole patent protection. (See the case of citric acid production discussed in Chapter 4, Section 4.5.)

1.7 Newly available industrial technology Generally, whenever a new industrial technology has become available from an external source supplying other industries, typical opportunities for new process developments should be investigated. Such new technology could be applied to the potentially profitable production of desired products, which previously could not be produced economically. The timely recognition and exploitation of such opportunity is one of the main challenges of industrial R&D. As a classical example, the solvent extraction technology has been researched, applied, and refined as an industrial separation/purification tool in the 1940s and 1950s. This was due to the urgency nuclear applications at the time; however, on a relatively small scale. When the essential basis of this technology became publicly available in the 1950s, it was recognized as a powerful separation tool by many of the best R&D leaders in the chemical scientific profession. Its potential uses were intensively and competitively studied by many faculties and institutes and discussed in successive international conferences. The various proposals for processes and contacting equipment then were developed further and patented in an all-out race by those in the fields of chemical processing, pharmaceuticals, petrochemicals, fertilizers, and hydrometallurgy, resulting in dozens of highly profitable industrial processes and enterprises by the late 1970s. Copyright © 2002 by CRC Press LLC

The so-called “energy crisis” of 1973 prompted many fundamental studies on the more efficient production and use of energy, and particularly in the chemical industry. Many old-fashioned processes and equipment were then condemned as utterly inefficient and, after intensive scientific and technological development, were replaced eventually by new solutions. Many new equipment models and designs were developed and introduced in the following 15 to 20 years, and most of these are now considered “standard practice.” A similar international effort at the time was devoted to the desalination of seawater in order to supply potable water to arid areas at a reasonable cost. Such an intensive effort resulted in improved industrial equipment and technologies, which are now available on a wide and diverse scale, although the industrial investments (dependent mostly on public funds) apparently are still not catching up with the demand. These technologies include, for example, multistage flash evaporation, multiple-effect distillation with different heater combinations, vapor recompression, reverse osmosis membranes, etc. (See the excellent review of Rafi Semiat in Reference 14.) However, it is important to remember that these technological developments should not be classified for a limited “specialized” application. They could also be the critical key for many new processes in the chemical and biotechnology industry that has involved a significant evaporation load, or that operates sections at widely different temperatures and requires large heating/cooling exchanges. Later on, the use of advanced membranes as separation tools, of nano-structured catalysts, of extraction at “supercritical conditions,” of the high vacuum technology, of lasers and plasma as focused heat sources, of micro-systems, (to name a few), have added many new, potent processing possibilities. Today, the advances in industrial biotechnology are notable and already offering industrial ways to replace many old chemical synthesis processes and to produce economically some of the large-scale organic chemicals. This is a direct link to the ongoing progress made by the corn sweetener industry (mostly in North America) in the industrial uses of enzymes (in particular, the immobilized enzymes) for producing very pure, defined compounds from starch or cellulose by chemical and physical processes. (See some basic references in 15, 16, and 17.) Many very important applications in the pharmaceutical industries for very expensive products were handled as a “lot of small-scale batch production units.” The simpler large-tonnage fermentation processes were for a long time limited to the smaller molecules (ethanol, acetic acid, etc.) and in direct competition with the petrochemical processing industry, except for food applications. The large-scale production of citric acid by fermentation opened the way to more complex products. At present, the biotechnology R&D handled by the largest corporations aims mainly to large tonnage, relatively lower cost, and intermediate chemicals for the polymerization of industrial plastic materials, such as lactic acid as just one example.18 Of course, any such research project starts with the fermentation biology in order to select the organism and the conditions in which the desired Copyright © 2002 by CRC Press LLC

compound can be reliably produced. However, one should note that any such fermentation can only be operated in relatively dilute conditions compatible with the life (osmotic pressure?) of the microorganism. Thus, the desired compound can only be obtained in a concentration range of 1 to 8% (very rarely up to 12 to 15%) in the fermentation broth, together with unavoidable residual contamination from the fermentation media. A quite expensive concentration installation will be needed downstream, together with specific separation and purification processes, to obtain the final 100% product. And this fact-of-life brings us back to the solvent extraction and/or desalination technologies mentioned above. Finally, the electronic computer process control technologies, which became widely spread in the past few decades, did allow the practical reconsideration of some processes that were studied theoretically, but were previously rated as difficult or even hazardous to control manually (i.e., based on the operator’s decisions and responses). These are mainly in the petrochemical field, but also in the classified chemical industry for military applications.

1.8 New functions for new products A new product could also be needed in the market to fill a new function at the users’ end, resulting from some parallel technological development in other industries. Whenever the need for such a product can be defined, a process development and evaluation effort will be justified. Of course, the silicon chip industry jumps to mind, but there are many more prosaic largescale products. For example, the production of citric acid by fermentation was handled for many decades as a pharmaceutical product on a small scale. However, the expanding industry for soft drinks and packaged food required more and more citric acid, until it was treated as a commodity and produced in larger tonnage in continuous plants by a completely different technology. In a different field, the way in which fertilizers are used in more sophisticated and intensive farming by many developed countries, under ecological control, has continuously changed. This has called for the supply of more concentrated, cleaner, multicomponents mixtures, mostly water-soluble, with less residual contamination of the soil and underground water layers. The same principle applies to products in the insecticide and fungicide fields, as the toxic metals were removed from the formulae and replaced by very specific, biodegradable, organic components. The purchase specifications of many of the fine chemical intermediates used in the mass production of plastic, refractory, and ceramic materials have also changed significantly to meet the users’ demands. The term “advanced material” is more and more fashionable these days (although not always justified) and are interesting and profitable markets that the chemical industry is expected to supply. This would require a significant innovative effort. For example, a young entrepreneur named Steff Vertheimer started nearly 40 years ago to study the preparation of small bits of very hard and Copyright © 2002 by CRC Press LLC

tough solid material, by sintering tungsten carbide powder with various chemical additives, mechanical pressing, and heat treatments. These products improved continuously and now the cutting tools produced by his companies throughout the world have a sizable portion of a billion dollar market. Unfortunately, due to the climate of terrorism there also are increasing markets developing today for shock-resistant ceramic protectors and bulletproof glass panels. However, there should be a real need or demand for such new products from the potential users, and not just the desire from the suppliers to sell more or to respond to a passing fashion. When this author was starting in R&D, he was given a project (with his tutor, A. Mitzmager) to develop applications for the use of tetra-bromoethane (TBE), a heavy, stable organic liquid containing 88% bromine with a specific gravity of about 3. The wishful purpose of this development was to increase the limited markets that existed for the company which was (and still is) making and selling bromine compounds. TBE was used then only in mineralogical laboratories for bench-scale, “sink-float” separations between solid particles of different densities after the controlled dilution of TBE with a solvent. For example, a mixture of particles is slurryed in a liquid of specific gravity 2.83. All the “reject” particles with a lower average density will float while the heavier particles with valuable metallic content will sink. So, why can’t similar separations be obtained on an industrial scale? This R&D project was a very interesting challenge and within a couple of years several possible industrial applications became focused. A continuous separation technology with liquid cyclones was developed and piloted, and methods for the recovery and recycle of the TBE were designed and tested. The economics looked good on paper and the know-how (with full technical assistance) was offered practically free of charge to any user willing to buy the TBE.19–26 However, despite all the sales efforts, nothing really happened in the industry. A basic difference had been ignored; that separating a hydrometallurgical plant (which is basically a chemical plant, using acids or cyanide or similar materials) from a mineral beneficiation plant, where, at most, small quantities of chemical reagents could be handled. This difference is not reasonably objective, but it relates to the people, organization, management, and staffing. Apparently, no manager of a mineral plant was willing to have a separation unit with thousands of tons of a bromine compound in his backyard, and all the potential objective advantages and profits could not change that fact. This manager may be convinced that nothing would go wrong as long as the plant would be operated according to the instructions. But he also knew his staff and that, somehow, someone could make a mistake, and he had enough worries to keep him awake at night. This lesson was painful but clear; the developing team should try to put themselves in the place and the mentality of the potential user of the new product. They should ascertain that they would like to have such a new supply or means as this before convincing themselves that there should be a need and a market. Copyright © 2002 by CRC Press LLC

1.9 Corporate public image The development of a novel high-tech chemical process technology has often been used to enhance the public image of a chemical corporation as a progressive factor, particularly by those companies operating old plants in crowded areas. Of course, this cannot be the main reason for a new development project, but it could be a contributing factor. Although it is quite difficult in these cases to separate publicity from fact, this factor has often been used effectively by interested parties to gain the good will of upper management so they will invest in a novel process development, in particular, in this hightech generation. Another related aspect, which is recognized inside the profession but hardly ever discussed publicly, is the importance of the professional selfesteem of the engineering and R&D staff of the corporation. Their involvement in a pioneering development should boost their interest, loyalty, and efficiency. Upper management does not always appreciate this effort and often act as if employees are disposable. In many cases, temporary pressures and false economy considerations have led upper management to drastically reduce, or even eliminate altogether, the R&D and new project budgets. Such decisions could have an immediate effect on the yearly profit statement, but it generally leads to a serious loss in the corporate market position in the future, as available know-how becomes obsolete and the more qualified individuals leave the company.

1.10 Worth another thought • The development of a new chemical process is a major technicaleconomical effort that can be justified only if it fills a concrete need of an industrial corporation. • All operating plants are living under the shadow of a possibly more efficient, completely new process with drastically lower production costs that may endanger the company’s basic economic existence if it ever becomes available to the competition. • All the geopolitical changes have seriously affected many older chemical plants, forcing the owners to reconsider their production processes and develop alternative ones. • If an existing patent covering the nature of a “valuable product of interest” is due to expire, or if a way to by-pass it can be proposed, a process development effort is justified. • Whenever a new industrial technology has become available, opportunities for new chemical process developments should be envisioned. • The biotechnology R&D handled by the largest corporations aims mainly at large-tonnage, relatively lower cost, and intermediate chemicals for the polymerization of industrial plastic materials.

Copyright © 2002 by CRC Press LLC

• Any industrial fermentation can only be operated in relatively dilute conditions, and a very expensive concentration installation will be needed downstream, together with specific separation and purification processes. • A new product could be needed to fill a new function at the users’ end, resulting from parallel technological development in other industries. Such a need will justify a process development and evaluation effort, if there is a real need from the potential users and not just the desire from the suppliers to sell more. • The developing team should try to put themselves in the place of the potential users of the new product and ascertain if they would like to have this new supply, before claiming that there should be a need and a market. • The professional self-esteem of the engineering and R&D staff is very important to a company, and their involvement in a pioneering development should boost their interest, loyalty, and efficiency.

References 1. Hawley, G.G., The Condensed Chemical Dictionary, 10th ed., Van Nostrand Reinhold, New York, 1981. 2. McKetta, J.J. and Cunningham, W.A., Encyclopedia of Chemical Processes and Design, Marcel Dekker, New York, 1983. 3. Meyers, R.A., Handbook of Petroleum Refining Processes, 2nd ed., McGraw-Hill, New York, 1996. 4. Bickford, M. and Kroshwitz, J.J., Concise Kirk-Othmer Encyclopedia of Chemical Technology, various eds., John Wiley & Sons, New York, 1999. 5. Comyns, A.E., Encyclopedic Dictionary of Named Processes in Chemical Technology, 2nd ed., CRC Press, Boca Raton, FL, 1999. 6. Skidmore, C., Review of World Baddelayite Production and Future Outlook, presentation to the Zircon 1995 Conference, Munich, May 1995. 7. Poleatev, I.F., Krasnenkova, L.V., and Smurova, T.V., Manufacture of zirconium oxide for fusion cast, Tsvetn. Met. (Moscow), 12, 56.8, 1988. 8. Tan Guoca et al., Preparation of zirconium oxide from zircon by slaked lime sintering process, Faming Zhuanli Shemqing Gonkai Shuomingshu CN, 1, 063, 268, August 1992. 9. Mizrahi, J. and Gorin, Ch., Process for the manufacture of substantially pure zirconium oxide from raw material containing zirconium, Israel Patent. Application 127,848, December 1998; PTC/Il 00/00125, March 2000. 10. Schoenlaub, R.A., Method for Manufacturing Zirconium Oxide and Salts, U.S. Patent 3,832,441, July 1973. 11. Araten, Y., Baniel, A., and Blumberg, A., Process for the manufacture of Potassium Nitrate, Proc. of the Fertilizer Society, No. 99, 1967. Also U.S. Patent 2,902,341, 1959. 12. Eyal, A., Mizrahi, J., and Baniel, A., Potassium nitrate through solvent extraction of strong acids, I&EC Proc. Dev., 387, 1985.

Copyright © 2002 by CRC Press LLC

13. Mizrahi, J., Improved process and apparatus for the production of potassium nitrate, Israel Patent Application 9347HA1, 1993. (Assigned to Haifa Chemicals, Ltd.) 14. Semiat, R., Desalination, present and future, Water Int., 1, 54–65, 2000. 15. Vogel, H.C. and Todaro, C.L., Biological Engineering Handbook Principles: Process Design and Equipment, Noyes Publishing, Park Ridge, NJ, 1996. 16. Blanch, H.W. and Clark, D.S., Biochemical Engineering, Marcel Dekker, New York, 1997. 17. Johnson, A.T., Biological Process Engineering: An Analogical Approach to Fluid Flow, Heat Transfer, and Mass Transfer Applied to Biological Systems, John Wiley & Sons, New York, 1998. 18. Baniel, A., Eyal, A., Mizrahi, J., Hazan, B., Fisher, R., Konstad, J., and Steward, B., Lactic Acid Production, Separation and/or Recovery Process, U.S. Patent 5,892,109, 1997. (Assigned to Cargill, Inc.). 19. Mitzmager, A. and Mizrahi, J., Pre-concentration of flotation feed with TBE, Min. J., 7, 481, 1961. 20. Mitzmager, A. and Mizrahi, J., Improvement in the Sink-Float Classification of Solid Granular Material, Israel Patent 18,108, 1962. 21. Mitzmager, A. and Mizrahi, J., Method for the Sink-Float Classification of Wet Granular Material, Israel Patent, 18,230, 1962. 22. Baniel, A., Mitzmager, A., Mizrahi, J., and Star, S., Concentration of Silicate Minerals by tetrabromoethane, Trans. Am. Inst. Min. Eng., 146–154, 1963. 23. Boskovich-Rohrlich, E., Mitzmager, A., and Mizrahi, J., Structure and beneficiation of a low-grade iron ore, Min. Mag., 325–331, 1963. 24. Schachter, 0., Mitzmager, A., Mizrahi, J., and Brillianstein, A., Classification and jigging with heavy liquids, Trans. Am. Inst. Min. Eng., 91–96, 1964. 25. Mitzmager, A. and Mizrahi, J., Correlation of the pressure drop through small cyclones operating with dilute pulp of various liquids, Trans. Inst. Chem. Eng. (London), 42, 152–159, 1964. 26. Mizrahi, J., Separation Mechanisms in Hydro-cyclone classifiers, Brit. Chem. Eng., 10, 686–692, 1965.

Copyright © 2002 by CRC Press LLC

chapter 2

Starting the development of a new process 2.1 Driving forces 2.1.1

Backing of a large corporation

It is evident at the onset that the development and implementation of a novel industrial chemical process is a very expensive project; that only the backing of a sizable corporation can carry it to completion in the final instance, and then only if and when it fits into its corporate framework. Thus, this backing is a necessary condition for the completion of the project.

2.1.2

Promoting group

However, in most cases, such development projects can be initiated by a group of promoters, who could be a part of one or more of the following functions: an individual scientist, an academic department, an industrial research organization, or an engineering company. Lately, certain “risk capital” funds are involved in such promotions as well. In certain cases, this promoting role could be carried out inside the corporation by its own R&D section, by its new business department, or even in many cases by able production engineers. (One may also mention that in certain large corporations, some “secret” development projects are actively encouraged by certain executive managers, who report only to them with entirely separate budgets.). This promoting group could be formally organized as a legal partnership and raise a limited investment, in order to manage and carry on the first part of the development project, which includes the following elements: • The “invention” (in fact, a proposal for a new industrial process) with its justification compared to the existing situation, its basic chemistry and mode of operation, and its implementation logic. Copyright © 2002 by CRC Press LLC

• A sufficient basis for the formal claims in a patent application, which can derive from a novel reasoning and/or of newly-discovered factual evidence. • A bench-scale experimental demonstration of the novel aspects of the proposal, which could convince, or at least impress, experienced scientists. • A preliminary technical, economical study of the proposal, which indicates conclusively that its potential profitability should justify the necessary investment in the development program. • The promotion, i.e., the location of potentially interested corporations, contacts and presentations, and negotiations of a commercial contract, until the project is sold and transferred to a corporate organization.

2.1.3

The second part

The second part of the project follows the transfer of the management and the associated responsibility of the project to the corporation. The transfer, from the promoters in the first part (or period) to corporate management in the second period, changes drastically the vision and rules of the game. This transfer could be a delicate procedure with many pitfalls, as completely different driving forces are operating during the development and implementation of a novel industrial chemical process. In the first period, the promoters are mainly interested in all the principal issues that could affect the elaboration of the rationale of the project and the choices of possible implementation objectives. Such issues could determine the decision-making process of each of the prospective corporate candidates, and result in their buying and implementing the project. Obviously, the promoters, as a small group, have not the means nor the time, and possibly not the ability to pursue in detail all of the possible options. In the second period, on the other hand, the corporate project manager is taking over the decision-making, with the concrete task of optimizing the novel process in one particular context, and building and operating a viable plant. The manager has to cover every significant aspect of the development and implementation, but in a definitely limited scope.

2.1.4

Public authorities

Public authorities also are actively pushing or helping such industrial development in many countries. For example, funds are made available as grants, loans, or subsidies (i.e., tax credits) for industrial R&D budgets and/or for risk capital companies, and these funds could facilitate the promoters’ initiative or corporation incentive. However, this procedure could also introduce significant restrictions concerning the location and ownership of the plant. Copyright © 2002 by CRC Press LLC

2.2 How a new process is born The objective need for a new process and its potential application must first become identified in one of the situations listed in Chapter 1, and become known to the professionals in the field. Only then will the subjective motivation for an industrial invention be actuated in one or more of the following routes listed below.

2.2.l

Normal research and development activity

Normal R&D activity creates a situation in which a better basic scientific understanding of the limitations of the existing industrial processes is systematically associated with the study of similar developments, and with new available data or technology in parallel fields. When scientists are saturated with this information, an idea may come to someone in the form of a proposal: “Why can’t we do it better another way?” This “click” is part of the functions normally expected from any industrial R&D group, albeit in a corporation, an academic department, or an industrial research organization. Nevertheless, the mechanism of its occurrence is not well understood, and it is generally attributed to individual characteristics. (Despite much interest, most of the studies and dissertations devoted to this idea-generating psychology are related to artistic creation and apparently there is still no accepted theory as regard to scientific/industrial inventions.) But not all such ideas are actually pursued. Many (one would say most?) are impractical, premature, or incorrect in some aspect. There is no discredit in that, since a more fundamental study of the limits of the problem can only be reached by raising these proposals. Many potentially interesting ideas could also be stopped just for lack of follow-up by the initiator, who, for example, could be too busy. One of the main challenges of any R&D organization is to have a proper forum and a routine procedure for the systematic recording and review of such ideas, which would then avoid any possible bias due to personalities, positions, and past records.

2.2.2

Personal motivation

The main driving force for a successful innovation (the invention, the promotion, and the first steps) is without a doubt the personal motivation of the more-talented R&D scientists. In addition to their genuine scientific curiosity and drive, a series of successful innovations is generally considered as a key for their personal advancement, their public recognition, and their personal satisfaction. It could also be linked to a financial bonus or other incentives in certain organizations. Since these more talented scientists could also be successful and happy in an academic position, a major challenge for the management Copyright © 2002 by CRC Press LLC

of the industrial R&D organization is to create conditions in which their scientists would be interested in continuing to work there, effectively and for a prolonged period. Of course, this motivation is a delicate matter, which could concern nonscientist personalities as well. There are no easy shortcuts.

2.2.3

Corporate function

The managers of the dedicated corporate departments (R&D or “new business”) have the role, the staff, and the budget to generate new projects, and they are generally looking for new ideas that may be worth promoting. These new subjects could be found internally by a continuous and systematic covering of their defined territory, or from the outside by promoters who are familiar with their corporate business field. On the other hand, once their hands and means are more or less full (as decided in advance by the yearly plans and budget), they have to find ways to delay additional new proposals without causing too much ill will with the promoters who are offering a golden opportunity. More flexibility in this matter could give better overall results.

2.2.4

Financial and commercial rewards

Each participant in the promoter/developer group (external to the corporation) is normally motivated by some financial reward, expected from a successful implementation, including buy-in at an early stage, development, and re-sale when ready. But some of the participants in this group also could have additional commercial considerations related to their other activities, such as the supply of engineering services, the sale of proprietary equipment, the assignment of marketing rights, exclusivity in certain services, agent’s commission, and so forth. Unless all of these interests are clear from the beginning, they could lead to conflicts between the partners. Such unpleasant cases are not uncommon; therefore, it is advisable to have a clear picture of the situation at the onset of a joint venture to help promote an innovative process development.

2.2.5

False starts

It is generally recognized that, due to the pressures stated above, a very large part of these “would-be inventions” eventually will be false starts and dropped sooner or later. This situation could also happen to exceptional R&D scientists who, following reappraisal, will readily pull back their proposals (for the time being) and find other avenues for their efforts. There is no shame in such a decision, as this is an integral part of R&D work. Copyright © 2002 by CRC Press LLC

Unfortunately, some of these false starts may take a long time to die, wasting precious time. The general efficiency of an industrial R&D organization depends on the routine screening procedure for new ideas, preferably by a peer review that is more readily accepted than a manager’s ruling.

2.3 Explicit definition of the development project It is essential, at the beginning of every development project, to detail explicitly what the project will try to achieve and what would be considered a successful implementation. This clearly written definition may be critical for the success of the entire project, and the promoting group should give it utmost attention. The first benefit will be that thorough discussions will force the group to focus its proposals exactly toward objectives and procedures that are feasible in this real world. This definition should include the following components listed below.

2.3.1

Objectives and purposes

A quantitative definition of the actual objectives and purposes of the development project, as compared with the known existing situation, may include, for instance: • • • •

2.3.2

Minimum specification of the new product or products Maximum acceptable production cost Minimum recovery of the valuable component Acceptable waste disposal, etc.

Patents

There is no point, however, in starting a significant development effort unless there is a reasonable prospect for an eventual patent protection in case of positive results. An adequate patent search and strategy should be discussed and decided at an early stage, after consultation with the relevant experts. This analysis should start with a clear statement and definition concerning: • Extent of effective patent protection needed for the increasingly large investments in industrial research and the potential profits • The need to avoid some constrains in an existing patent

2.3.3

Possible industrial framework

A projection of the eventual (or possible, probable) industrial implementation framework of the new process is needed to help cement the technological factors Copyright © 2002 by CRC Press LLC

specific to that framework. This projection, which will be continuously updated with a compilation of more available details, generally includes: • Scale of production, which affects the equipment size and function • Different options of raw materials; availability of critical services • Possible synergetic coproductions, local regulations, etc. In some cases, the initial projection of such framework may only be wishful thinking in the eyes of the promoters, as the corporation concerned may not have been approached in the early stages. But, at least, there should be a reasonable assumed framework since, without it, the process development would be mere speculation.

2.3.4

Timetable

In industrial reality, once the need for a new process has been recognized and a feasible idea or proposal has been advanced and approved, the results of the development effort should be delivered reasonably fast, despite the many complex issues and decisions that need to be resolved. An often-cited goal, before the detailed engineering of a new plant can be started, is between 12 and 24 months. A detailed time-table — desired or imperative — should be worked out and included in the particular project definition, listing all the different projected activities (see Section 2.4 Different stages of a typical program), the periodical review points, the change points in project management (passing the torch), the requirement for introduction of additional support teams, and the emphasis on specific efforts. Note that the change in project management will generally require a few months for systematic transmitting of know-how and of periodical summary and review of the process package.

2.4 Different stages of a typical program The different stages of a typical development and implementation program are listed below relating to the author’s own experience. Each of these different stages will be discussed in detail in the remaining chapters of the book. Of course, there could be many different situations relating to specific case histories. One should emphasize that in many new processes, the requirement for a comprehensive pilot work is essential to ensure a thorough understanding of the effects of the different recycle streams. Note that this is not a simple procedure and there could be at least five reviews at different levels of responsibility and authority. Each of these reviews should be well prepared and be concluded either in a “no” (closing the project) or a “maybe” (okay to proceed to the next step) decision . In certain cases, additional facts and information are required before a particular review can be concluded. Copyright © 2002 by CRC Press LLC

Definition of the Objective Need for a New Process • Study of the existing process limitations, yielding the idea • First review of the idea in the Okay to proceed promoting group Definition of the Development Project • Grouping of the core project team • Transformation of the idea into a process-working definition • Critical and systematic review Okay to proceed Feasibility Tests and Analysis • Literature survey • Review Okay to proceed Promotion • Patent application • Experimental program and reporting of the results • Preliminary process design for a particular proposal • Economic analysis of the specific proposal Negotiation, Agreement, and Transfer to the Corporation • Management review Okay to proceed Working Program Towards a First Implementation • Complementary R&D • Piloting and modeling • Patent updating • Process and equipment detailed design • Market tests of the products • Formal permits required Decision to proceed with the first plant • Final review Process Package and Plant Design • First plant design • Modeling and optimization • Critical piloting Approval for construction • Final engineering review Construction and Running-In • Personnel training • Running-in and adjustments in the new plant Consolidation of the New Process Know-How Package • Patent updating

2.5 Corporate management procedures for new projects In recent years, a number of management procedures have been adopted in most large corporations for the control of strategy, choice, and cost of development programs. These procedures resulted mostly from the large number Copyright © 2002 by CRC Press LLC

of new product developments aimed at the consumers in an affluent society, such as electronic hardware and software, travel, household products, toys for all ages, etc. Large R&D budgets have been geared in this direction and the different management schools have stepped in with recommendations and procedures for “doing it better.” Among the more known, commercially available management tools, for instance, is the “Stage-Gate” system, propagated by Dr. R. G. Cooper for new product development. This system combines certain strategic principles and procedural steps for choosing the right product to develop from an assumed large number of proposals, and how to control the different steps of the program with “Gatekeepers,” all from inside the corporation, starting with the “invention.” There is no doubt that such management tools could be very useful to the extent that they would force, step-by-step, the preparation of orderly documents, analyses, and reviews of all different aspects of the project. After such preparation, the case will be better based, but the value of such decisions will still depend on the decision-makers. Although the approach described in this book for development of a new chemical process to respond to a recognized need (as discussed in Chapter 1) have different emphasis, it is also based on stages and successive reviews. Therefore, the employees and consultants of corporations that have already adopted one of the above mentioned R&D management procedures, such as Stage-Gate, will find it easier to understand and assimilate the message in this book, and to use the detailed recommendations within their corporate directives.

2.6 Worth another thought • The “invention” is, in fact, a “proposal” for a new industrial process with its justification related to an existing situation, its basic chemistry and mode of operation, and its implementation logic. • The transfer of decision-making from the promoters to the corporate management can be a delicate procedure, as completely different driving forces are in power. The promoters are mainly interested in the “principal” issues affecting the project rationale and the possible implementation objectives. Later, the corporate project manager has to cover every significant aspect in a definitely limited scope, since he has a concrete task of optimizing the novel process in one particular context: building and operating one viable plant. • The R&D systematic activity associates a better basic scientific understanding of the limitations of the existing industrial processes with the study of similar developments and with new available data or technology from parallel fields.

Copyright © 2002 by CRC Press LLC

• One of the main challenges of any R&D organization is to have a proper forum and a routine procedure for the systematic recording and reviewing of proposed ideas. • There is no point in starting a significant process development effort unless there is a reasonable prospect for an eventual patent protection (in case of positive results.) • Without a projection of the eventual industrial implementation framework of the new process (scale of production, options of raw materials, availability of critical services, local regulations, etc.), the process development could be merely speculative. • The different stages of a typical development and implementation program would include at least five comprehensive reviews at different levels of responsibility and authority, each ending with a No/Maybe decision.

Copyright © 2002 by CRC Press LLC

chapter 3

Essential resources needed for the development project: preceding implementation 3.1 Introduction A new industrial chemical process is concerned, in the final analysis, with chemistry and technology, plants and products, and markets and finances. But the successful development and implementation of a project depends mostly on the interaction and cooperation between many critically important human factors. This basic statement was not realized at the onset by all concerned. When this author suggested it in a paper in 19721 after a year of struggling with a very difficult new plant start-up and after long nights thinking why it went wrong, the thesis apparently touched a nerve, as an overwhelming number of colleagues from around the world responded to the idea. Academic research is done mostly in small groups at universities and institutes. Until the final product (the thesis, the paper) is sent out, any interaction with other colleagues on the subject of research is done purely on a voluntary basis. Apart from his/her personal scientific curiosity and drive, the external interests of each of the researchers are also obvious, i.e., personal advancement and recognition, or the next research grant. (At least it was so before the epidemic of “start-up” ventures.) Applied R&D toward a new industrial process is very different, as the timely contribution of many professional specialties is essential and critical to its success (after the first inventive steps). In many cases, this interaction is not well understood by some professionals coming from academic research and this has often been a major source of problems. Therefore, it is important to discuss this “fact-of-life” in detail in this chapter, together with the essential resources needed for the process development, up to the decision to build a new plant.

Copyright © 2002 by CRC Press LLC

3.2 Specific managerial skills A qualified and efficient manager for a process development project incorporates certain personal qualities and professional experience, since he/she has to deal with a different and special management challenge. The manager of such a process development project needs to: • Mediate the essential work and the temperamental egos of individual personalities (inventors, promoters, experts), as well as the orderly coordination and interaction between many different disciplines and functions, and of proper formalities, records, and communication. • Report upwards and find his way through the internal politics of a large corporation, in which every director may have his own vision. • Have an extensive and diversified background in the basic sciences, in the engineering disciplines, in project control, and in plant operations. • Be willing to learn something new every day from every new situation. • Assume his first project management responsibility preferably after his participation in several similar projects, as a professional engineer and as assistant project manager. Managing a project long term is generally an exhausting experience, so a successful project manager expects after that and generally gets a promotion to a less-demanding job. The scarcity of qualified managers is generally recognized as a critical bottleneck in many organizations. A notso-qualified individual also may succeed, but he/she should be ready to ask for advice when needed and have adequate support from management and external consultants.

3.3 Core project team The core project team consists of all the members reporting directly to the project manager and working full time (or at least most of their time) on the project. This core team generally includes, in addition to the project manager’s executive assistants, people from other departments and organizations who are temporarily delegated and integrated into the project team for this particular project. For example: • Inventors and researchers from the R&D promoting team who are continuing to work with the project team as long as they are needed, bringing with them their scientific knowledge of the subject and helping in the coordination of future R&D activities, along with the process engineers who are taking over the continuation of the process design. • A specialist from the products’ marketing organization who is assisting in pinpointing the market needs and supervising the product’s testing. • A number of process chemical engineers from the engineering department (or division or selected company) who are in the interim

Copyright © 2002 by CRC Press LLC

delegated to lay down the essential process flowsheets, prepare balances and economic spreadsheets, equipment comparison, engineering and optimization studies, budgets, etc. This group is expected to work as a team so that all its members have access to all the documents and are aware of all the facts, and each can contribute his opinion freely inside the team. (Communications outside the team are, of course, subject to the manager’s instruction.) Therefore, it is important and it should be accepted that external credits are given to the team as a whole and not to individual member’s contributions. All of the team’s outgoing documents are approved and signed by the project manager or by his functional representative.

3.4 R&D laboratories and pilot installations 3.4.1

Company’s own laboratory and pilot installations

In many industrial companies, the promoters and project manager have to use mostly the company’s own laboratory and pilot installations in order to decrease costs and preserve confidentiality. This may be helpful on one hand since these laboratories should be generally familiar with the materials, analytical methods, and modes of reporting. On the other hand, this could also be a limiting factor in that conflicting lines of duty and priorities are placed before the laboratory managers with the representatives of several projects putting pressure on them.

3.4.2

Outside laboratories and pilot installations

In other cases, the promoters and the project manager are allowed to contract parts of the testing program to outside laboratories and pilot installations, i.e., universities, public institutions, or private specialists. As such laboratories are generally limited in their specialization, each would be doing only a specific part of the overall job. Another fact of life is that the better labs tend to be over-booked and may not always be available when needed. Thus, the choice, motivation, and efficient use of these laboratories require experienced coordination and complex planning of all details so that results will be received in time and be relevant to the specific conditions. Unfortunately, the importance of this detailed coordination job is not always appreciated and is often delegated to the younger, less experienced engineers on the team.

3.4.3

Analytical laboratories

The larger part of the man-hours (and cost) of an R&D program is generally devoted to the performance of the chemical and physical analyses. Yet, in many cases, it has been seen that some researchers and engineers relate to the analytical laboratories as they would to automatic machines — bring in Copyright © 2002 by CRC Press LLC

the samples, press a button, and collect the results. This approach is most unfortunate and often backfires. Any analytical test work, and in particular for R&D, involves complex decisions and good judgment. The managing and principal chemists of the analytical laboratories are, in most cases, highly trained professionals with extensive and varied experience and interests. It is strongly advisable to seek their cooperation at the onset by telling them about the aims and scope of the project, inviting them to meetings and reviews, and discussing with them the significance of the results. The contribution of these chemists has been found to be very fruitful in many cases. The exact definition and limitations of the analytical methods often present a difficult area, since there is generally a direct link between the accuracy of the results and their unit cost and time delay. The highest level of accuracy is not always justified and affordable, especially in the exploratory stages of the R&D where a fast procedure is preferable. For example, the early process development and promotion of the transformation of solid potassium chloride into solid potassium nitrate was based on the direct examination under a microscope with polarized light. The black potassium chloride (cubic crystals) could be seen as they transformed into brightly colored product (monoclinic crystals) when subjected to a specific solution. Thus, many different compositions of such solutions could be screened rapidly. This was also an impressive demonstration to visiting dignitaries to the laboratory, which helped promote the sale of the process. It should also be noted that the accuracy and significance of the result from any chemical analysis could not be any better than the sampling procedure that was used to procure the sample for analysis. Sampling of multiple-phases from small test vessels can often be a complicated task requiring a delicate touch.

3.5 Experts on marketing and on potential users 3.5.1

Particular terminology

A contribution of experts could start with the trivial aspect of the particular terminology and the sets of units used for centuries in certain market segments, for example, P205, Baume, Avoirdupoid, grains, ounces per metric ton. There are also specific analytical tests, such as the citric soluble P205 used in fertilizer, which is supposed to quantify the process in which phosphates are absorbed by the plants’ roots, and also some very particular wording of specifications and legal references. These arbitrary, often nonsensible names and units can be infuriating for scientists who are newly exposed to them, but they cannot be changed in practice, so one should accept them and make use of a translation sheet or program. Before being “accepted into the family,” the project team is often instructed to demonstrate a thorough familiarity with these terms in all contacts with potential partners and clients.

Copyright © 2002 by CRC Press LLC

3.5.2

Clients’ needs

On the more substantial level, the process development team should understand from these marketing experts which details and features of the products under consideration are really desirable and important to their final users from their point of view, and what these final users would be ready to pay for these results, if they were given the choice between different qualities. (Consider the classical example of instant coffee. Should one pay more for freeze-dried than for spray-dried? It’s a matter of taste.) In many real cases, the most desired features can be technically achievable, but this result can increase the final production cost too much, therefore, a compromise should be reached. So, the practical sale price structure of the particular products line must be well understood at an early stage, although the preferred marketing aims may not always be explicitly announced outside the core team in order not to alert the competition.

3.5.3

Competition

In other cases, existing competitors operating in this market may appear to be closer to approaching these final users’ needs. Such a disturbing situation should be recognized and extrapolated by marketing experts from their sources of information in the markets, so that the project team can focus on their work and possibly supply attractive improvements in time.

3.6 Support from experts on hardware 3.6.1. Plant engineering and operation When implementing the new process at an existing plant site, cooperation between the process developing team and the senior technical staff of the plant should be established early. This will be beneficial to both sides. Admittedly, such cooperation may often cause personal problems, mostly due to the differences in priorities, point of view, and style of communication between the two groups (i.e., what do they know about running a plant and about R&D science?). Here, the personality of the project manager should bridge these differences, and it is always better to address and solve them calmly and in a timely manner than under decision-making pressure. The plant’s management also can contribute some very good ideas and propose effective and practical design solutions from their point of view, and should be given credit for that. This cooperation can become critical anyway if it is decided to install and run a pilot plant at the plant’s site and connect it with “live” streams. On the other hand, several cases have been known of new processes that have been developed “in secret” in the corporate R&D facilities and that were later bluntly opposed and rejected by the plant’s operating management who felt that it was forced on them without their consultation. Copyright © 2002 by CRC Press LLC

Whenever possible, the process developing team should get a clear and early picture of the eventual implementation conditions of the new process in connection with an existing facility, including its infrastructure, existing services, and waste disposal possibilities. These specific conditions can pose objective limitations that have to be taken into account in the early stage of development, rather than making changes later. For example, the design temperature of the cooling water supply depends on the average climatic conditions in the area and can be critical when designing an installation for evaporation/condensation under high vacuum.

3.6.2

Equipment design

In many cases, the design of a novel process section can be critically linked to one particular piece of equipment or specific technology. Thus, the process’ results will depend not only on the process chemistry, but also on a particular combination of equipment design factors and operating conditions. Furthermore, it may be that this particular piece of equipment or specific technology can be supplied only by a very small number of specialized companies, each of them with their particular know-how, or at least their claims of such know-how. For example, this situation can apply to industrial crystallizers, special dryers for hygroscopic solids, industrial plasma heat torch, and the like. The process developing team may be feeling “cornered” if they are operating in a corporation committed to the “purchase-by-bid-only” procedure, since such formal link with a specialized supplier may cause the following problems. • From the beginning, in order to get enough information from any would-be supplier for evaluation and preselection, a mutually binding secrecy agreement should be negotiated. This is not a simple proposition, but requires at least that the novel process has already passed the patent application stage and that the equipment supplier is not already signed up with competing corporations. • The pilot tests should be done with one particular supplier in mind (most probably with his pilot equipment) and the cooperation of his staff after a basic commercial framework has been established. • This procedure would give the selected equipment supplier a clear advantage in the final price negotiations, which would include some remuneration for his know-how and past experience, and for his guarantees and assistance in start-up. • It would be logical to include the engineering company staff at this stage of “prepilot” equipment survey and contract negotiations, and use their experience and services also in the pilot testing. However, this participation would need to advance the decision on the contract bid for the choice of the engineering company more than it would be generally anticipated. Copyright © 2002 by CRC Press LLC

3.6.3

Corrosion in construction materials

In many cases, the novel chemical process conditions can introduce unknown corrosion aspects, which have to be clarified as early as possible. These aspects relate to the reliability of the materials of construction that will be used for the equipment and for the piping. This reliability bears first on safety considerations deriving from a possible accidental failure (particularly in pressurized and/or high-temperature systems), but also on the estimate of the lifetime, supply cost, maintenance schedule of each piece of equipment, or the possibility of contamination of the product with metallic traces. The orderly and reliable testing of the corrosion rate for each combination of one particular choice of construction material and one particular set of process conditions is a relatively long procedure of many months, starting with obtaining reliable samples of unusual materials of construction. Furthermore, the exact and final conditions for such a corrosion test might be known only after the process development has firmed up (compositions, additions, temperatures, etc.). Therefore, the tendency is to be safe and to test the worse possible conditions. But this choice can also lead to an expensive overshooting. Even if the choice of just-in-case better/safer materials is available, it may result in a significant increase in investment costs and reduce the calculated profitability. The presence of certain “trace elements” impurities in certain streams can affect seriously the corrosion properties. A classical example is the presence of copper cations in a solution, which can “cement” on a steel surface, create a corrosion cell, and (quite surely) a hole. When such possibility is defined and confirmed, the need for certain pretreatments or a side-stream treatment becomes an essential part of the process or, in certain cases, the need for bleed streams to avoid accumulation of such impurities. This is a highly specialized field, and it is advisable to engage, from an early stage, the support of an expert consultant with relevant industrial experience, who can recommend the options and procedures to arrive in time at the optimum specifications for materials of construction. Furthermore, the public authorities and the insurance company representatives often insist on receiving written recommendations from an expert, at least in relation to the risks and damages that could result from a possible accidental failure.

3.6.4

Operation and process control

Nowadays, automatic process operation and control are taken for granted for nearly every new chemical plant. The design techniques and the hardware selection are well advanced; however, the correct design is critically dependant on the input of process experts with industrial experience on the following issues. • What are the more efficient procedures for starting and stopping of the plant? These transient procedures are not obvious, and they are not always well covered in the basic designs. They have to be careCopyright © 2002 by CRC Press LLC

fully thought out for every new process and for every installation in particular, as they determine the internal inventories of the buffering tanks, or the need for recycling certain streams, or for reprocessing some bulky intermediate streams. These transient procedures are based mostly on kinetic response data, which may have to be measured experimentally, or inferred from previous reliable industrial experience in similar situations. • How to assure a safe response to any possible failure of some equipment or a possible error in the action of an operator, limiting the risks and damages. • What is the better choice for reliable probes and instruments in direct contact with the process streams, which are made of suitable materials and can be available and supplied off-the-shelf?

3.7 Support from experts in software 3.7.1

Publication search and analysis

Obviously, the cheapest and fastest part of the R&D effort is to retrieve practically everything that has been published on all the different factors relevant to the proposed process. This publications search can be subcontracted to specialists or academic libraries where it is done by computer screening of large databases, according to agreed key words. The search output is a long list of published items with titles, address, and, in some cases, abstracts. In order to keep this output in a manageable volume, one should be careful in the choice of these key words or, preferably, start exploring with a trial-and-error iterative procedure. The resulting references can be first sorted and divided by the R&D team into two categories: those that were published because the authors considered the new information to be of considerable scientific interest (of course), but with no commercial value, and those that were patented. The collection of workable copies of the preselected publications can be subcontracted to specialized organizations, and sometimes can be lengthy and expensive, since most of the comprehensive experimental data was published long ago when the competition for journal space was not so intense. Unfortunately, in more recent publications, new experimental data are more often presented as small figures that can hardly be used as sources for numerical data correlation. If a set of such data appears to be important enough, one could try to locate the authors and ask them for a copy of their original numerical data. It was observed that most authors were more responsive when such a request came from an academic researcher than from a commercial company. Some publications also need to be translated. The senior R&D team should devote an analytical effort to the study of these publications and of their results, and possibly to the numerical correlation of the included experimental data, if needed. In addition to the factual information on the data, such analysis may give the senior R&D team some interesting hints about:

Copyright © 2002 by CRC Press LLC

• Reasons that initiated such previous research work • Strategic aims of the authors and their supporters • Why they didn’t succeed in this aim on the industrial scale up until now After the distribution of this survey and analysis to the core team and their relevant consultants, a thorough discussion can be very useful before deciding and starting on any significant experimental program.

3.7.2

Intellectual property and secrecy

If one accepts for a fact that the novel industrial process under consideration is important and needed on the market, one must also assume that the competition is also looking at new processes of their own, which could be quite close to the one proposed in this project. Everyone in this activity is trying to keep his own programs secret for as long as possible. (“Confidential” may be a nicer word) Every corporation has its own secrecy procedures and all its employees, consultants, and contractors are generally signed to extensive formal secrecy obligations. But it should be recognized that such signed undertaking is mostly of a moral nature, since it would be practically impossible to enforce, particularly with dissatisfied participants. Thus, some corporations sometimes add restrictions on the internal flow of process information, on a need-to-know basis (limited distribution lists), which is a normal practice in certain business departments. Such restrictions have been shown to be very detrimental to a process development team, where all members should be talking freely among themselves while contributing to a common goal. These “limited distribution lists” can have clearly negative effects on the motivations and efficiency of certain members who may feel personally insulted, such as, “they don’t trust me.” In the final instance, there is no alternative in this kind of activity but to work with reliable, well-motivated, and satisfied individuals.

3.7.3

Patent application

If another corporation has already filed a patent application that could prevent the implementation of the proposed process, this should be known as early as possible, so that the claims of such an application can be avoided and possibly by-passed. The real problem is that patent applications are made public only 2 to 3 years after their filing date, and most patent attorneys can use perfectly legal tactics to extend this period. The issues affecting the patent filing strategy are complex and all derive from the potential menace of “the competition:” 1. The promoters (inventors) need the patent award as proof of the novelty of their proposed process and of their intellectual property, partly for their self-satisfaction, but mostly as a necessary condition and instruCopyright © 2002 by CRC Press LLC

ment which will allow them to sell and transmit the process to the implementing corporation. Note that the patent office’s checking and awarding means only that no public information or previous patent claim has been recorded, but it does not state that the process will work as claimed or that it has any practical usefulness.The filing date of the application is an important asset, but it is also a limitation, since it sets the procedural mechanism in motion. When the patent is awarded, the inventors have to decide within a short fixed period about the other countries where they need to apply for and record the application at their own significant expenses. Most importantly, a patent application represents the best extrapolation of what was known to the inventors at that particular filing date. Such extrapolation includes the purposeful addition, into the claims, of conditions that have not yet been proven, but are expected to give, more or less, similar results, e.g., a larger group of reagents or solvents, a wider range of operating conditions, and so forth. As the inventors will probably continue to work after filing their application and could arrive at “better” claims later, they would have to decide whether it is worthwhile to cancel the previous application (and lose their priority date) and file a new application with the better claims. 2. The management of the implementing corporation needs a strong patent to justify and protect the company’s significant investment. They also are interested in securing the widest possible coverage, both in substance and in countries worldwide. But if the corporation managers were not part of the decision-making on this subject at an early enough stage, they will have a complex choice when they take over, between filing an additional patent application, or canceling the previous application and reapplying, or accepting what has already been done. 3. The consultations with expert patent attorneys are concentrated mostly around the legal procedural options and the exact wording in the application. There is a specific professional “jargon” which is apparently mandatory in all relations with patent offices. However, there is an element of risk as to the extent of coverage in the claims, which is the exact definition of what is claimed to be new and exclusive to this patent. A larger coverage in the claims may weaken their exclusivity position and could be more difficult to secure (that is to convince the patent examiner). A narrower coverage may be held stronger, but could be easier to by-pass. This coverage has to be decided by the inventors, possibly with the input of the project management (if there is any at this point) with respect to the possible competition.

3.7.4

Process modeling

Process mathematical modeling has been one of the main advance fronts in chemical engineering research and development since the 1980s, with excellent theoretical books and computer programs available. Today, this mathematical Copyright © 2002 by CRC Press LLC

modeling can be a useful tool in process development in many cases. More specifically, it can be useful if the relevant numerical data can be made available or can be reasonably inferred from precedents. This possibility can make a lot of difference in the entire program. Thus, the contribution of an expert in this field is needed, at an early stage, to draft a model from the principal elements of the process definition and to define exactly which numerical data will be required to make the performance and results of such a model significant. (This is discussed further in Chapter 9, Section 9.5.)

3.8 Safety, public regulations, and waste disposal support 3.8.1

Safety

Most chemical plants present some form of known safety hazards which are kept well under control. All of the chemical industry is living with this inherent characteristic. The implementation of a new industrial chemical process can introduce a different safety hazard that was not previously known in the operation practice of this particular corporation, although it is probably familiar to other parts of the chemical industry. This potential safety hazard can be due to the composition of the raw materials or of the reagents (i.e., metallic impurities, organic solvent, and acids) or to the operation conditions (i.e., flash point, pressure, etc.). For example, an engineering requirement in the early large-scale implementation of solvent extraction processes in the chemical, mineral, hydro-metallurgy, and food industries specified the “explosion-proof” safety standards, which were already well developed and used in the petrochemical industry for years. There are very experienced safety consultants around who are available to inform and reassure the project team in this field and to prepare the different manuals needed for the lab, the pilot operation, and the full-scale plant. Systematic surveys and consultations with one of these safety experts are needed to identify such potential safety issues early in the development program and to document them in detail. Any public regulations relevant to such hazards in the area of implementation should also be well understood. This information should allow to include any necessary requirement into the process definition and the plant’s future design and control, and to assure a safe operation. For instance, certain parts of the process and of the plant can be declared as “explosion-proof” areas, separated and designed/operated accordingly, while other parts can be located far enough and avoid such additional expenses. Needless to say, the insurance company will also be inquiring as to when an investment will be decided.

3.8.2. Public regulations In present days, any industrial or commercial activity is subject to a score of public regulations (laws, taxes, custom duties, permits, etc.) which are changing Copyright © 2002 by CRC Press LLC

quite often, particularly in the democratic countries. A manager in industrial R&D cannot expect to know all of these regulations just from reading the newspapers or from personal experience. Lawyers and specialists should be commissioned for the task of collecting up-to-date information that is relevant to the project and organizing it into different files from which project management can decide what needs to be included into the working program.

3.8.3

Waste disposal

A critical aspect of any chemical project is the definition and quantification of all the possible waste streams and of the general options for their disposal within the framework of the particular region considered. This should be addressed quite early in any development program. This is again a specialized activity including technical, commercial, and legal aspects. This should be dealt with in collaboration with suitable consultants to find at least one (but preferably more) acceptable and affordable disposal procedures for each waste stream.

3.9 Support of specific codes relevant to plant design and operation, and product quality Many products have to conform to a client’s own purchasing specifications. However, certain groups of clients are buying on express condition that the product is fulfilling the requirements of specific “official” codes controlled by a suitable governmental regulation, e.g., a food-grade, reagent-grade, or pharmaceutical-grade product (i.e., the FDA in the U.S. sphere of influence). Such codes are issued, controlled, and maintained generally by public organizations (mostly manufacturers, but also consumers), and they regulate not only the final composition and packing of the product, but also the raw materials and additives used, and the conditions prevailing in the execution of most stages of the production. For instance, the food-grade code specification dictates not only that all the raw materials and additives introduced in the process should be of foodgrade quality, but also that the conditions in every process stage should be designed and controlled to prevent contamination, oxidation, microbial activity, etc. This code also details the routine quality control with very detailed analyses and formal reporting and recording. External quality control also is often required. If such specific code can be relevant to the new process and/or to the new product, it should be studied from the beginning by the marketing and analytical professionals and well understood by the core team. In addition to the general principles of the code, its practical implications may relate in some detail to the selection of raw materials, the plant design, or the analytical control. For example, it took one producer of food-grade phosphoric acid many years to drop from 2 ppm arsenic in the product to less than l ppm, as required by the food-grade code. Copyright © 2002 by CRC Press LLC

3.10 Economics The ability to do economic evaluations of investment and of operation costs on the whole process or on a defined part of it is needed from the beginning in order to assist in the choice between alternative options. These economic evaluations can start with simple “order of magnitude” calculations, but can become increasingly complex as the project’s scope gains substance. Therefore, one should assure from the onset of the project the contribution of a specialized cost engineer, who will lay down standardized spreadsheets adapted to the particular framework of the project and collect the exact unit costs that should be used by the corporation at the intended site. As the process and the implementation framework begins to firm up, the standard tools and the experienced staff of an engineering company should produce investment budgets and operating cost estimates with an accuracy (plus/minus) margin starting at 30%. This can be progressively reduced to 15% in the final report.

3.11 Development expense budget Last but not least in the list of essential resources needed is an expense budget to cover all the development costs (salaries, transfers, suppliers, consultants, materials, patents, special equipment, etc.). At the beginning, this budget has to come from the promoters’ own sources. At a later stage, if the project can be incorporated in the form of a limited responsibility shares company, those investing in risk capital funds can possibly be convinced to buy a certain portion if it looks promising and if the promoters have good personal records. In certain countries, public funds can be procured as a partial contribution to specific industrial or scientific developments, mostly in the form of loans repayable in case of economic success, with many conditions attached. This is generally a very lengthy procedure as always with public funds. When an implementing corporation takes over the project, it covers the past and future costs from its own financial resources, through one of many different financial formulas. However, as every project is different and past records can only be indicative, “the future is no longer what it used to be.” The ability to predict logically a future process development budget has always been a weak point, although this fact of life is not always admitted or even recognized. Most promoters naturally tend to be rather optimistic in that regard. The only practical method, while consulting with all experienced participants around, is to: • Divide the development program into a number of functional periods with specific aims. • Define and estimate separately every possible cost item in each period. • Draw up a detailed list of every possible cost item. • Then add generous safety factors. Copyright © 2002 by CRC Press LLC

3.12 Worth another thought • The success of the development and implementation of a new chemical process depends mostly on the interactions and cooperation between many critically important human factors. • The core project group is expected to work as a team, so that all its members have access to all the documents and are aware of all the facts, and each can contribute his opinion freely inside the team. • There is generally a direct link between the accuracy of the results from analyses and the unit cost and time delay. The highest level of accuracy is not always justified and affordable for all results, particularly in the exploratory stages of the R&D where a fast procedure is preferable. • The accuracy and significance of the result from any chemical analysis cannot be any better than the sampling procedure used to procure the sample. • The process development team should understand which features of the product are really important to the final users and what the final users would be ready to pay for these results if they were given the choice between different qualities. • The process developing team should get a clear and early picture of the eventual implementation conditions in an existing facility, which could impose objective limitations that need to be taken into account. • In many cases, the design of a novel process section may be critically linked to one particular piece of equipment or specific technology, and the process results will depend not only on the process chemistry, but also on a particular combination of equipment design factors and of operating conditions.

References 1. Mizrahi, J., People, organization, and process implementation, Chem. Tech., 459–464, 1972.

Copyright © 2002 by CRC Press LLC

chapter 4

Actual case examples The following cases are given here only to the extent needed to illustrate the general principles that are discussed in this book. Obviously, the detailed account of each of these cases could have filled a book (assuming that such details were allowed for publication). The particular cases chosen here come mostly from the author’s direct experience, but are “old” cases, not any with relevance to an ongoing operating corporation.

4.1 Nature and man: the Dead Sea The Dead Sea is one of the world’s natural wonders; the deepest and one of the hottest places on Earth. During milleniums, it has accumulated chlorides and bromides of magnesium, calcium, sodium, and potassium. The average composition of the sea brine reached a steady state, as the average yearly amount of fresh water that was brought by the Jordan River into the Dead Sea brine was equaled by the amount of water that had evaporated from this brine. In the first half of the 20th century, processes were investigated to recover the potassium chloride from this brine as a vendable product (potash). Chemists at the Hebrew University in Jerusalem (M. Novominski, M. Langoski, and others) studied the entire relevant physical–chemical solids/liquid saturation system. They found that when the Dead Sea brine evaporated and gradually concentrated in a solar pond, salt (sodium chloride) reached first its saturation point and precipitated. Then carnallite (a hydrated double salt of magnesium and potassium chloride) was crystallized together with some more sodium chloride. The scientists also found that the mixture of the carnallite crystals and sodium chloride, obtained from solar ponds, could be leached at ambient temperatures with a large amount of water to leave a number of fine potash crystals with a rather low yield. Or the mixture could be leached alternatively with a limited amount of water at a higher temperature to decompose the carnallite and dissolve all the magnesium

Copyright © 2002 by CRC Press LLC

chloride, allowing to separate by filtration the remaining solid sodium and potassium chlorides. These can be hot-leached and then the hot filtrate brine can be cooled and concentrated under vacuum in conventional equipment to crystallize the potash. The remaining brine can be recycled back into the solar pond to repeat the process.1 This straightforward process was eventually developed and an industrial plant was built to produce potash at the southern end of the Dead Sea near the biblical site of Sodom. The plant included the following successive sections (see the excellent description by J. Epstein, Reference 2): large solar ponds for salt, solar ponds for carnallite, wet harvesting of the crystals from the carnallite, solids–brine separation, decomposition of carnallite in two countercurrent stages, hot leaching of the solids in a circulating brine, hot filtration of the salt, vacuum cooling crystallizers, potash washing and drying, and all the adjacent services required for a desert location. In this first venture, the solar ponds with wet-harvesting were, indeed, the critical new element essential to efficiently handle millions of tons per year of corrosive slurry. This was done with floating dredges that crisscrossed the ponds, slurrying the crystals from the bottom, and pumping the slurry into a floating pipeline to the shore. But as we said at the beginning, this is a changing world and two humaninduced changes, which were outside the control of the operating Dead Sea Potash Company, occurred later and required new process developments. First, starting from the 1950s, all the fresh water from the Jordan River was diverted for agriculture development in Israel and Jordan, and practically no water was allowed to drain anymore into the Dead Sea. The age-old steady-state was ended and the concentration of the Dead Sea brine started to increase, slowly but inexorably with more salt precipitated at the bottom of the sea and, consequently, the carnallite production of the existing solar ponds increased. The trend of these changes could be followed, analyzed, and predicted exactly from the 1960s, and since it was imperative that all crystals produced in the carnallite solar ponds be removed to avoid clogging the whole system, it meant that the potash production capacity should be increased accordingly. Finally, this additional raw material was available almost for free, so why not build more potash production in the 1970s? But when expansion plans were prepared and approved and their execution was about to start, the so-called “energy crisis” of 1973 happened and the cost of thermal energy jumped almost overnight by a factor of four to five times. With the new production costs and the uncertainty concerning the future situation in this regard, the additional potash production by the hot leach process could possibly lose money. So the processes basic concepts had to be urgently reconsidered, including the potential use of some elements which were known but not considered essential in the previous economic context. It was known, for example, that in the crystals mixture produced in the solar ponds, the salt and the carnallite were precipitated practically in separate crystals, and that the size distribution of the carnallite crystals was relatively coarser than that of the salt crystals. It was possible to separate Copyright © 2002 by CRC Press LLC

about 25 to 30% of the carnallite, in rather pure form, from the feed to the potash plant just by coarse wet-sieving of the slurry. It also was known for a long time that, when a controlled quantity of water is added to carnallite crystals at ambient temperature, all the magnesium chloride would go into a solution with about 20 to 25% of potassium chloride, leaving the remaining potassium chloride as fine solids. There were no incentives to make use of this information up to that point, since it would have only complicated the straightforward “hot leach” process. But when severly pressed by the energy crisis, its reconsideration allowed a corporate task force to develop a “cold crystallization” process, which required almost no thermal energy. From a carnallite stream with a relatively small salt content, a cold crystallization system would produce reasonably coarse and clean potash crystals without heating and cooling. This process was analyzed in detail and its implementation depended on the development of a novel type of continuous industrial reactor–crystallizer, in which the rather pure carnallite solids were fed and dissolved in one part, while the potash was crystallized in another part, solids were decanted in quiet zones and brine was circulated between the different parts.3 This new design was piloted and demonstrated in an intensive program. The crystal mixture pumped from the carnallite ponds to feed the existing hot leach potash plant was first wet-screened to separate as much as possible the coarse carnallite fraction. A first cold crystallization plant was successfully build for several hundred thousands ton per year of potash. This new development allowed a few years of breathing time in the race against nature and the oil “lords.” Then, the gist of the problem was proposed to the physical mineral separations scientists. Given a crystal mixture of carnallite (specific gravity 1.6) with salt (specific gravity 2.1) in a slurry with end brine, a residual solar pond by-product from the potash production of Dead Sea brine (in fact, a concentrated solution of magnesium chloride with a specific gravity 1.35), how can one separate a greater part of the carnallite in a reasonable pure form, in millions of tons per years, and at very low cost? This physical separation did not have to be completed, since the remaining mixture could still be treated by hot leach, but the content of the pure carnallite fraction should be above 95%. This challenge was again solved by a novel technology: by centrifugal jigging on a tumbler centrifuge equipped with a conical wedge-wire screen with a rather large aperture. This type of centrifuge was developed earlier in Germany as a large-capacity screening device to produce low moisture coarse salt cakes. It was found that the pulsations in the expanded fluid bed of crystals, flowing on the inside of the conical wedge-wire screen, caused the heavier salt crystals to concentrate nearer the screen and, thus have the priority of passage through, leaving most of the lighter carnallite crystals behind. The large-scale application of this technology allowed another expansion of the “cold crystallization” plant and more breathing time in the continuing race against the clogging of the solar ponds. Copyright © 2002 by CRC Press LLC

Finally, the separation of the salt from carnallite in the finer size fractions was obtained by adaptation of the conventional froth-flotation technology for salt used in other lands, to the particular conditions of the Dead Sea chemistry. Today, the multimillion tons per year production of potash from the Dead Sea is using all of these originally developed technologies in an optimum combination.

4.2 Magnesium chloride-based industries In the early 1960s, it was apparent that end brine was a raw material very suitable for the production of magnesium oxide (MgO = periclase). This material is widely used for refractory bricks. Up to that time, a part of small deposits of natural magnesium carbonate, all the existing world production of this material was based on precipitation of magnesium hydroxide from sea bitterns. Such production is done in a very dilute system with hydrated calcined lime, which is an energy-intensive raw material. The application of a similar technology to the Dead Sea end brine would deny any advantage of its higher concentration, leaving only the disadvantages of a desert location. On the other hand, it was known that the thermal decomposition of such end brine can produce solid magnesium oxide and a vapor phase with a water/hydrochloric acid mixture. The detailed conditions required to conduct continuously such thermal decomposition processes were studied by Dr. J. Aman from the Hebrew University in Jerusalem, who developed and patented in the 1950s the direct-contact continuous “Aman reactor.”4 This is a sort of spray dryer (vertical cylindrical/conical chamber) in which the end brine is sprayed at certain locations from the top while hot flue gases at 800 to 900°C are introduced tangentially at the middle height, creating a definite internal flow pattern. The solid impure MgO particles remaining from the liquid drops are settled and removed from the lower conical outlet and the gases exiting from the top are directed to a direct-contact absorption column, producing a 18 to 20% HCl solution (somewhat below the 22% azeothropic concentration). This novel process was piloted in the 1960s, and its enormous industrial potential was then demonstrated. However, its implementation remained critically dependent on the economic utilization of the HCl by-product and, thus, it was delayed until a proper combination could be organized. Other issues were connected to the presence of smaller quantities of magnesium bromide in the brine, which would produce elementary bromine in the gases, and this needed to be dealt with. This started a solvent extraction process for separating a stream of pure magnesium bromide from the end brine, but this is a different story. Note that the same Aman process technology was also licensed and applied successfully in other countries by the Ruthner Company for the decomposition of the iron chloride solution resulting from steel pickling plants, where the recovered HCl solution could be recycled and reused on site in the pickling plant. Copyright © 2002 by CRC Press LLC

4.3 Economic uses for the HCl by-product solutions 4.3.1

Strategic policy

In the 1960s, the managing team of the IMI Institute for R&D, directed by Dr. A. Baniel, created a corporate strategic policy defining the need of developing economic uses for by-product HCl solutions. As part of the Aman process for magnesia, a number of promising “acid-salt” double decomposition processes were under consideration aimed at upgrading the value of the chlorides of potassium, sodium, and magnesium (available in very large quantities at low cost) into the relevant sulfate, phosphate, or nitrate salts. Implementation of any one of these potential processes would also yield HCl as a by-product solution (as indeed, the IMI process for potassium nitrate when it was implemented by Haifa Chemicals Co.).5 Up to that time, HCl was mainly a traditional by-product of organic chlorination reactions and of small-scale chemical industries. In most of these cases, the HCl was wasted, neutralized with lime and/or limestone, and disposed of as CaCl2 in the sea. Only in very large organic chlorinating installations could the HCl by-product be collected as a solution and recycled to an electrolysis section to regenerate elementary chlorine, or collected and recycled by the Kellog’s Kel-Chlor process.6 This route was hardly more economical, but it was possibly less problematic than the neutralization route.

4.3.2

Coupling of HCl-producing and consuming plants

Some industrial uses with economic justification were developed within this strategy (see discussion below), but the basic problem that remained for several decades was the critical coupling in the implementation between the plant producing the HCl and the plant consuming it, in their geographical location, in quantity, and in timing. It should be remembered that the HCl–water system is dominated by an azeotrope at 20 to 22% HCl, so that every ton of HCl generated below the azeothrope is accompanied over the fence by 4 to 5 tons of water, and the transportation of such solutions would be impractical over any significant distance. “Breaking the azeothrope” (i.e., obtaining more concentrated solutions or even 100% dry HCl) is possible, but complicated and expensive, both in investment and in energy consumption; for example, by using a cycle of CaCl2 brine. This was a wide field of creative process design, aiming at a better use of the energy and expensive heat exchangers, and for possible synergetic utilization of sources of low-temperature waste heat.7 (See also Chapter 6, Section 4.)

4.3.3

Timing of implementation

As an acid reagent, HCl could be used to replace sulfuric acid in several mineral industries. Some new processes in hydrometallurgy and mineral refining were studied and a few of these could have been developed and Copyright © 2002 by CRC Press LLC

used if a reliable HCl source could have been made available at the right time; for example, the cleaning of sand for the glass industry, the purification of different sorts of clays, the reprocessing of nonferrous scraps, etc.

4.3.4

Production of pure phosphoric acid

The main novel process that was actually developed and used on a large scale in several plants was the production of phosphoric acid by hydrochloric acid leaching of (calcium) phosphate rock. The conventional process with sulfuric acid gives a solid gypsum residue, which is separated by filtration from the impure “wet” phosphoric acid (WPA) solution. There exists also processes based on nitric acid. When using a hydrochloric acid solution to dissolve the phosphate rock, the water-soluble residual CaCl2 remains in the same aqueous solution with the phosphoric acid. A new separation process, therefore, was required to isolate the phosphoric acid from the CaCl2 (and all the other soluble impurities). This result was provided in a pioneering breakthrough by A. Baniel and R. Blumberg, by way of solvent extraction. The first IMI “standard” phosphoric acid process was quite complex with six different multistage, countercurrent batteries. A comprehensive description of all the issues related to its development and implementation was presented at an international scientific conference,5 as an IMI staff report prepared by a dozen senior staff members, each one in his/her specialty. This unique approach started in this book, but unfortunately it was not well understood and was not pursued in further scientific publications. The new process also accomplished a thorough purification of the phosphoric acid product, which could aim at the higher value markets. Such markets were traditionally supplied by “thermal” phosphoric acid, obtained via elementary phosphorus (see below). This novel process was actually licensed and implemented first in Japan, Brazil, and Spain, where some existing sources of by-product HCl already existed, before it was used in Israel in large plants fed with HCl by-product from potassium nitrate and periclase productions.

4.3.5

Technological difficulties

After the basic chemical research and the bench-scale demonstration of the new process, the developing team at the IMI Institute for R&D had to face some difficult technological issues on the way to implementation.

4.3.5.1 Materials of construction HCl is a well-known, “nasty” component to work with, as it attacks practically any metal. Previously, it could be handled in industry only in small glass equipment (i.e., Pyrex), or small glass-lined steel (i.e., Pfaudler), or in some cases, in rubber-lined steel (limited to the lower temperature range of less than 60°C). As all of these options for materials of construction were

Copyright © 2002 by CRC Press LLC

very expensive, sensitive, and quite limiting in large volumes, it was obvious from the start that plants for the large-scale production of relatively cheap materials could not be built exclusively from expensive materials. Fortunately, during the same period, the technology for the design and erection of large equipment and piping made from plastic was being developed in several advanced industrial countries. This technology used sheets and tubes of thermoplastic PVC and polyethylene (and, later, polypropylene) with possibly the external reinforcement of layers of glass-fiber/polyester setting mixtures and of steel members. Later, the fabrication of equipment made exclusively from reinforced polyester or epoxy setting mixtures was established. This new technology was the critical engineering basis for any HCl-based industry at that time. Thus, the process development group had to take a very active role in locating the best know-how available worldwide, in establishing specialized companies and workshops in Israel, in creating design standards and testing procedures, and in merging all of these into a practical working system. Another limitation was that the use of plasticizer materials in the thermoplastic material was strictly prohibited for all vessels containing solvents, as these plastisizers would be leached out by the solvent. (Today, this fabrication technology is essentially available widely as a standard engineering choice, but there are continuing new improvements in materials and in design techniques that have to be evaluated.) At the same time, it was obvious that the use of these thermoplastic materials would limit the process temperature to below 60°C (at most). Thus, any solvent stripping operating at higher temperatures would remain mostly with conventional glass-lined equipment, although thermo-setting resins could sometimes be used for limited functions, and should be minimized as possible. Another prosaic but important limitation was that, for structural strength design considerations, all plastic vessels needed to be round (vertical cylinders) and this affected both the internal functional design and the plant’s general layout considerations.

4.3.5.2 Safe, stable conditions for solvent extraction in large mineral plants At the beginning of the project, the “explosion-proof” conditions associated with the handling of relatively large quantities of organic solvents with rather low ignition points (i.e., butanol, pentanol, and the like) were well known in petroleum refineries and petrochemical installations, but rather unfamiliar in the mineral/chemical industry. The process development group had to recruit experienced consultants in this area and make a special effort to study, assimilate, and adapt the explosion-proof codes to these particular projects, even for such simple items as the venting of excess gases. In addition, the composition of the solvent stock circulating in the plant could hardly be taken as a constant, as it undertook various chemical degradations and additional reactions, mostly with the unavoidable impurities flowing through the plant streams. For example, most phosphate ores contain some organic matter soluble in acidic leach solutions, which are Copyright © 2002 by CRC Press LLC

partly extracted and accumulated in the solvent, and require specific cleaning procedures.. Such reactions can produce contamination in the product or even change some of the solvent’s properties. Thus, a surprising amount of sophisticated R&D in organic chemistry was needed for such mineral process development.

4.3.5.3 Clean starting solution for solvent extraction One of the main enemies of industrial solvent extraction is the crud consisting of fine solid precipitates, which accumulates at the interface between the two liquid phases and may prevent their separation and cause emulsions. This crud may also clog lines and build up in equipment. When dissolving, for instance, a typical phosphate ore into a hydrochloric acid solution (with minimum acid excess), most of it goes into a solution. The resulting slurry is degassed under vacuum and the solid residue (consisting of sand, clays, dirt, etc.) is then flocculated and separated by countercurrent decantation, and the overflow is polished by filtration. This is the easy conventional part. However, when the clear filtrate solution comes in contact with the organic solvent, the solubility conditions change, and it was often found that crud would precipitate. For example, part of this crud can be organic colloidal material originating from the natural phosphate ore, which was maintained completely in solution in the strongly acid solution, but can be precipitated when part of the acid is extracted and flocculated by the organic reagent. Other parts of this crud can be mineral or metallic salts, which were initially kept in solution by the strong acidity. As these elements were defined, the process development team had to devise ways to avoid this crud, or at least reduce it to proportions that could be handled with periodical cleaning schedules. This aim could be achieved by either changing the properties of the starting phosphate ore, if there was an affordable choice (e.g., by using a more expensive calcined phosphate concentrate with no organic matter), or by adding pretreatments (such as ion exchange) to the solution before its transfer to the solvent extraction section.

4.3.5.4 Recovery of the residual solvent from different exit streams All the effective solvents were partially water-soluble, and their saturation solubility in the exit aqueous streams was of the order of a few percents, depending on the temperature and on the other solutes presents. In principle, these residual solvents could be stripped down to the allowed and affordable level of, say, less than 100 ppm in a distillation column with sufficient number of stages and reflux. Again, this may seem a trivial question of engineering, but it was rapidly apparent to the process development team that the investment on the glass-lined equipment, the possible attack of fluoride anions on the glass lining, the possible deposition of calcium fluoride from waste solutions whenever heated, and the associated thermal energy and cooling water would be a critical load on the economics.

Copyright © 2002 by CRC Press LLC

Thus, every possible way to decrease these costs had to be considered in the process development. The overall technical–economical optimization could recommend a different solvent, which possibly may have been less effective in the separation, but cheaper to recover. Other practical questions also needed to be addressed, such as the possible fouling of the higher temperature stripping equipment with solid incrustations and, in particular, on the heat exchanger surfaces.

4.3.5.5 Large-capacity liquid–liquid contacting equipment The implementation of the new processes required large-capacity liquid–liquid contacting equipment8–23 for multiple-stage countercurrent batteries made of suitable materials, i.e., plastic (see above). The concept of the mixerssettlers was already established on bench scale and used in pilots and relatively small industrial installations (i.e., for uranium extraction). However, the design of efficient large-scale equipment was not established at the time and some issues had to be solved. First of all, hydraulic heads for the flow of liquid streams from stage-tostage in both directions had to be worked out. The simplest solution for maintaining such hydraulic heads is to install two interstage pumps for each stage (with the associated sumps and level controls, but with very low heads), which, in addition to the mixer, amounts to three explosion-proof motors per stage. No problem. One has only to multiply the number of stages by three, but the result is not a simple number, but more likely a snowball. The cost of an explosion-proof motor is 2 to 3 times that of an ordinary one, but its installation can cost 10 times more, and the level control loop will double that total. One alternative can be to design with only one transfer pump, plus a difference in height, but this difference would accumulate and certainly complicate the vertical layout for large multistage batteries. Therefore, to keep all the mixers–settlers on the same level with only one motor per stage, there was a clear and imperative need to use each motor for more than one task. This prompted from the beginning a hydraulic research and development program as an integral part of the new chemical process implementation on a large scale. Figure 4.1 illustrates the principle of the patented IMI “pump-mix” concept, with a vertical pump (between two static baffles) on the same shaft and above the mixing axial propeller. This pump design has a very steep {Q @ H} curve, so that the level in the mixer is self-regulated without any level control hardware. A manual weir for each ratio of liquid densities fixes the level of the apparent interface in the settler. This design was successfully installed in a large number of industrial installations for those cases where two liquid phases had relatively close densities and low viscosity and could be easily dispersed and circulated in the liquid–liquid mixer. A reasonable mass transfer was obtained with the large range of droplet sizes. But at a later stage, when implementing such processes involving the contact and equilibration of a heavy aqueous brine with a light organic solvent, the above design could no longer give the adequate hydraulic heads, mass Copyright © 2002 by CRC Press LLC

vent light in

light phase

heavy in

weir mixed phase

heavy phase

apparent interface

heavy phase

Figure 4.1 Mixer-settler with pump mix.

transfer, and phase separation rate. So, as an integral part of these new process developments, a new liquid–liquid mixer had to be designed and tested, resulting in the IMI “turbine pump mix” design (see Figure 4.2), which produced a controlled droplet distribution when operated with a variable speed drive. Finally, the use of plastic materials of construction also necessitated the functional design of a relatively new round settler (instead of the conventional “shoebox” design), with a central inlet of the two liquid-suspension from the mixer, and radial flow of all separated phases. This design required a fundamental study of the basic hydraulics connected to the settling coalescence process and of quantitative design procedures for such a settler. This study resulted in a later stage to the invention of the patented “compact settler” design, making use of racks of inclined partitions to save the largest part of the area and of the internal volume, and reduce the expensive solvent inventory. (See also Chapter 6, Section 6.4.1 and Figures 6.5 and 6.6.)

vent

light in heavy in

stator

light phase

heavy phase

mixed phase turbine

heavy phase

tangential connection

Figure 4.2 Mixer-settler with turbine pump mix. Copyright © 2002 by CRC Press LLC

light phase

weir apparent interface

4.4 Phosphoric acid diversification processes 4.4.1

Different quality specifications

The different users of phosphoric acid require different quality specs, which are listed below in order of decreasing purity and purchase cost per unit of P2O5: 1. 2. 3. 4. 5.

Chemically pure/pharmaceutical grade (CP or PG) Food grade — FGPA Technical grade — for different phosphate salts Animal feed grade — for cattle and poultry feed supplements Detergent grade — mostly for sodium tripolyphosphate (STPP and similar) 6. Liquid fertilizer grade — giving a clear aqueous solution after neutralization 7. Solid fertilizer grade — lowest acceptable grade (almost anything goes)

4.4.2

Solvent extraction opening

Up to the introduction of the solvent extraction processes in several countries,24,25 there were only two grades of phosphoric acid available: • The wet process acid (WPA) that was adequate only for the solid fertilizer grade, as it contained a few percents of sulfuric acid, some F, Ca, Mg, Fe, Al, etc. • Expensive “thermal acid,” which should be used for any of the other grades. The production of thermal acid (and these markets) was limited not only by its cost (2 to 3 times more than WPA), but also by the serious ecological hazards related to the elementary phosphorus. The unfulfilled potential and the needs were clear to the whole industry. Various chemical treatments were started in various places in connection with specific partial neutralization processes of WPA. As soon as the solvent extraction technology got established worldwide and the phosphoric acid extraction and purification was demonstrated,26–27 there was a worldwide rush by R&D units in this industry to establish new processes, to patent different related issues, and to build producing plants.28 The solvent extraction technology allowed for producing different quality grades of phosphoric acid at varying production costs, starting with the merchant qualities of WPA, which could be produced on site or be purchased. But one should also note that any one of these processes would leave a more impure residual stream containing between 30 and 70% of the starting phosphate values. This should be downgraded and compounded into a solid fertilizergrade by-product or mixed, if possible, with merchant WPA. This meant that their implementation could only be in proximity to a large solid fertilizer plant. Copyright © 2002 by CRC Press LLC

There was a very intensive worldwide struggle in the 1960s and 1970s until the novelty appeared to be more or less exhausted and the worldwide market saturated; this despite a very fast increase in the demand for the products, mostly for the detergent and liquid fertilizer uses. (The Tennessee Valley Authority [TVA] point of view is summarized in Reference 29.) Following are some of the processes developed by the IMI team in this particular field.

4.4.3

IMI “cleaning” process

The IMI “cleaning” process30 was implemented in 1974 in a large plant that is still functioning in the port town of Coatzacoalcas in southern Mexico, near a large WPA producer. The solvent extraction process is extremely simple and flexible and is based on the extraction of phosphoric acid from a 53 to 54% P2O5 WPA feed with a diisopropylether (IPE) organic stream at a lower temperature and its back extraction at a higher temperature as clean product. This is at a concentration of 50% P2O5, which is used either for detergent grade or for liquid fertilizer grade. The novelty of this process, which is quite unique, is that each operation is conducted in a temperature-controlled invariant system, in which three liquid phases with fixed compositions coexist in the zone delimited by the points: A. Light phase, almost only IPE B. Intermediate zone with a relatively high solubility of phosphoric acid C. Heavy aqueous solution with very little IPE, as shown in Figure 4.3 on a triangular equilibrium diagram for the tertiary system phosphoric acid-water-IPE IPE

A

B

C H2O

WPA

Figure 4.3 Triangular equilibrium diagram H3PO4-water-IPE.

Copyright © 2002 by CRC Press LLC

H3PO4

Without this particular feature, IPE would be a quite inefficient solvent. But the WPA is mixed with the recycled solvent in a weight ratio such that the total mixture composition fell into the three liquids zone, as close as possible to the line BC. An extract B is separated and most of the impurities remain with the residual C. Some water is then added to the extract to separate it along the line AC, which will give the Clean Acid(c) and the recycled solvent (A). All the mass transfer and the final results are obtained in a single equilibrium stage for each operation (for a fixed number of components, more phases at equilibrium = less degrees of freedom = simpler process, as Gibbs would have said), although a second mixer-settler was provided in the plant as backup and energy optimization. However, small amounts of the cationic impurities and sulfuric acid entering with the feed WPA (more components) are co-extracted and can be reduced to the extent needed by a countercurrent, backwash reflux battery. About 60% of the phosphoric acid is recovered as “clean acid,” while the “residual acid” containing 40% P2O5 (with most of the impurities) is returned and back-mixed into the fertilizer plant. The traces of the volatile solvent are removed from the two exit streams in two steam-stripping distillation columns.

4.4.4

“Close-cycle” purification process

The IMI “close-cycle” purification process31,32 to produce a quite pure phosphoric acid from WPA was a modification of the “standard” process in which the CaCl2-rejected solution was concentrated, roughly cleaned, and recycled to be mixed with the WPA feed. The rest of the process was similar. This allowed by-passing the situation where HCl was not available. The process worked well and the product was very pure, but the process was quite complicated. Several studies by large corporations showed that it could be justified economically only if it was implemented on a very large scale. Such scale exceeded the demand for such pure product in most geographical areas.

4.4.5

Mixed process

A mixed process is practiced in Israel by mixing a certain portion of WPA (at a 28 to 30% P2O5 solution before its final concentration) into the HCl leach operation in which the phosphate concentrate is dissolved. Such addition increase the average concentration of P2O5 in the leach solution and makes use of the acidity of the sulfuric acid in the WPA. The sulfate precipitates as gypsum with other impurities before the filtration of the solid residue. Straight sulfuric acid can also be used, but this would increases the load of gypsum on the filter, which needs to be HCl-resistant in such cases. The “mixed” process is then operated in the same way as the “standard” process, but in a more concentrated and cost-efficient way. It also avoids the restrictions caused by the limited supply of HCl and the low concentration of its solution.

Copyright © 2002 by CRC Press LLC

4.4.6

New proposals

New processes for phosphoric acid published since the 1990s included oen for obtaining phosphoric acid from phosphate rock and hydrochloric acid via ferric phosphate, which was patented in 1997 and published at the International Solvent Extraction Conference (ISEC) in 1999.33 It aimed at cutting drastically the volumes and number of contact stages involved in the standard IMI process, at the cost of a couple of solid–liquid separations by using the very low solubility of ferric phosphate (see also Chapter 5, Figure 5.1).

4.5 Citric acid by fermentation and solvent extraction 4.5.1

Conventional lime sulfuric acid process for citric acid

Citric acid is an expensive but widely used food additive, giving acidity and lemon flavor to industrial food and soft drink products. Sodium citrate is also much used as a detergent component for domestic laundry and in many pharmaceutic and fine chemical products. Citric acid is produced by aerobic fermentation in deep tanks, starting with a carbohydrate solution containing different additives and seeded microorganisms. After the practical completion of the fermentation and the filtration of the suspended material, the citric acid contained in the fermentation “broth” needs to be separated from any residual carbohydrates and from all the various impurities and byproducts, in a form suitable to produce pure citric acid crystals. The chemical separation route used by most producers consisted, in the addition of hydrated lime (“liming”), the precipitation of calcium citrate, the filtration and washing of the solids, then the decomposition of the filter cake in a sulfuric acid solution in a strictly controlled ratio to liberate the citric acid and precipitate the calcium as gypsum. After filtering and washing the gypsum, the solution is concentrated and the citric acid crystallized. The product crystals are washed and the wash solution returned to the concentrator. The remaining mother liquor is bled and recycled back to the liming. While all producers are keeping confidential the details of their operating procedure, it is probable that they are also using additional purification steps on different streams, such as active carbon, adsorbing filter aids, ion exchange, etc. This chemical separation route was used for many years, but is rather delicate to operate, as it had three solid–liquid separations with washing, resulting in a relatively low citric recovery yield, a high consumption of many reagents, a costly waste disposal, and the occasional dumping of contaminated batches. This was an obvious place for a better separation process.

4.5.2

IMI-Miles solvent extraction process for citric acid

With the increasing understanding of the temperature effect on the reversible extraction/separation of mildly strong acids with tertiary amine Copyright © 2002 by CRC Press LLC

extracting agents, the IMI team, under Dr. A. Baniel, proposed in 197034 a new process to replace the chemical route described above. This was rapidly developed, demonstrated, patented, and licensed to one of the major producers in those days, Miles Laboratories, Inc. The new process was then piloted and implemented in close cooperation with the Miles technical team, under Dr. Toby Wegrich, in an existing large plant in the U.S., affecting only those sections that were to be replaced. The satisfactory operation resulted in a much increased citric recovery (that relates to the plant production capacity) and in lower production costs. This process was then used in other plants of the company, giving them a strong advantage over any competition worldwide.

4.5.3

Newer solvent extraction process for citric acid

About 20 years later, another large American corporation (Cargill, Inc.) decided to get into the citric acid business from the start. The company had developed its own fermentation technology and knew, of course, about the IMI patent licensed to Miles (which was still in force), and were looking for a similar solution. Dr. A. Baniel and David Gonen provided this result in a very creative but simple way35 (which should be very instructive for future similar cases). The first process intended to use the fermentation broth as it was produced then in the existing Miles plants, so the attention of the developers and the original patent claims were referred to this particular range of citric concentration. But the citric concentration can be increased 2 to 3 times by evaporation, if needed, before the separation process. Such change allowed not only to avoid the formal wording of the claims in the original patent, but also to take advantage of the higher concentrations to get a more compact and efficient separation process with relatively smaller equipment and less solvent inventory. This novel concept was rapidly developed and demonstrated in close cooperation with the designated corporate task force. A large plant was built and operated very successfully on this basis. Why wasn’t this increase in concentration thought of and introduced in the IMI-Miles process from the beginning? Only because the exact framework of allowed changes (in the existing and producing plant) were defined from the start as a precondition for the novel process design, to limit the risks that the implementing corporation would be ready to accept. Why wasn’t this increase in broth concentration studied and patented later by the operating company after they had a working plant and complete control of the technology? Because then, the common rule in industry was applied: “If it works and makes a nice profit, don’t touch it.” Twenty years later, starting with a confident new team and a blank sheet, this preconcentration was a perfectly normal option for consideration. This lesson can be applied to many other processes. Copyright © 2002 by CRC Press LLC

4.6 Preparation of paper filler by ultra-fine wet grinding of white carbonate White paper, made with “neutral/alkaline sizing,” contains between 20 and 35% by weight of white filler powder, which is mostly precipitated calcium carbonate in the form of crystalline needles. This filler gives to the finished paper its whiteness, opacity, and weight. Such precipitated calcium carbonate is not difficult to make, but it is energy-intensive and has to be produced on a relatively large scale. Thus it is relatively expensive, particularly if it has to be dried, packaged, and transported for long distances. So, the economic need prompted the question: Why can’t it be replaced by finely ground, white calcium carbonate? A new process technology was developed and implemented on a moderate scale near a large paper mill. This dedicated exclusive user was receiving the slurry in accordance with his own specification, ready to mix into the feed going to the paper machine, naturally flocculated with a consistent size distribution, characterized by an average size of 1 micron, with most smaller than 2 microns, and with a minimum content of minus half a micron.36 The novelty and the particular features of the process technology, which was needed to obtain such final particles, were in the operating condition of a regular iron ball mill, such as the pulp density, the residence time, the temperature and certain chemical additives. However, since the final size specification cannot be obtained in a single pass, an extensive external circuit was needed for fine-size classification, separating the product’s fine particles from the recycled coarser particles. This circuit represented the main process challenge, considering the requirement that the product particles should retain their natural flocculation. Generally, in the technology of “fine particles,” dispersing agents would be used to achieve such size classification, but they were not allowed in this case. The process solution was derived from the previous research work done by one of the developers in the mechanism of hydrocyclones,37–40 which allowed the design of batteries of microhydrocyclones in a countercurrent arrangement, handling large flows of diluted slurry (Figure 4.4). In addition, an original automatic control scheme was designed to handle the natural fluctuations in the raw material and mechanical system, based on the continuous measurement of the size distribution in the product slurry, which operated a number of flow splitters affecting the recycle cycles. This lesson can be applied to other similar microparticle systems. These filler particles were more or less round, whereas the usual precipitated calcium carbonate generally consists of elongated needles and this affected somewhat the “usual appearance” of the finished paper, although most users were unable to perceive the difference. This ultra-fine grinding plant was happily operated by the Polichrom Company for about 12 years, but then the trading conditions in the area were changed, forcing the paper mill to modify its line of products and its operating procedure. Copyright © 2002 by CRC Press LLC

water recycle

water centrifuge fine product slurry

split

solid feed

split

Ball Mill

Figure 4.4 Principle of an ultra-fine wet grinding and classification process.

4.7 Worth another thought • The use of the by-product from one process in one plant as a raw material for another process in another plant creates a critical coupling between the two plants, in geographical location, in quantities, and in timing. • The implementation of a new process can also require new solutions as regards materials of construction, design standards, and new functional equipment. • The introduction of a new process to replace part of an existing plant is generally preconditioned into the existing conditions of the remaining sections. However, once it is well integrated into the production, an overall optimization should be studied for further improvement or future plants.

References 1. Kenat, J., The production of potash from the Dead Sea, Second Symposium on Salt, Cleveland, OH, 1965. 2. Epstein, J.A., The recovery of potash from the Dead Sea, Chem. Ind., 572–576, July 1977.

Copyright © 2002 by CRC Press LLC

3. Tzisner, T., The Maklef plant for cold crystallization of potash, Comm. Israel Soc. Chem. Eng., personal communication, October 1989. 4. Aman, J., British Patent 793.700, 1950; Israel Patent 8722, 1956. 5. IMI Corporation, Development and Implementation of Solvent Extraction Processes in the Chemical Industries, staff report at the Int. Solvent Extraction Conference, The Hague, 1386–1408, 1971. 6. Van Dijk, C.P. and Schreiner, W.C., Hydrogen chloride to chlorine via the KelChlor process, Chem. Eng. Prog., 69, 57–61, 1973. 7. Mizrahi, J., Barnea, E., and Gottesman, E., Production of Concentrated HCl from Aqueous Solutions Thereof, Israel Patent 36,304, 1972. 8. Mizrahi, J. and Barnea, E., A Gravitational Settler Vessel, Israel Patent 30,304, 1968. 9. Mizrahi, J. and Barnea, E., A Liquid–Liquid Mixer, Israel Patent, 43,692, 1973. 10. Mizrahi, J. and Barnea, E., A Gravitational Settler, Israel Patent, 43,692, 1973. 11. Barnea, E. and Mizrahi, J., Compact settler gives efficient separation of liquid–liquid dispersions, Proc. Eng., 60–63, 1973. 12. Barnea, E. and Mizrahi, J., A generalized approach to the fluid dynamics of particulate systems, I: General correlation for fluidisation and sedimentation in solid multiparticle systems, J. Chem. Eng., 5, 171–189, 1973. 13. Mizrahi, J., Barnea, E., and Meyer, D., The Development of Efficient Industrial Mixer-Settlers, paper presented at the Int. Solvent Extraction Conference, Lyon, France, 1, 14l-168, l974, 14. Barnea, E. and Mizrahi, J., A generalized approach to the fluid dynamics of particulate systems, II: Sedimentation and fluidisation of clouds of spherical liquid drops, Can. J. Chem. Eng., 53, 461–468, 1975. 15. Barnea, E. and Mizrahi, J., Separation mechanism of liquid-liquid dispersions in a deep-layer gravity settler (four-parts series), I: The structure of the dispersion band, II: Flow patterns of the dispersed and continuous phases within the dispersion band, III: Hindered settling and drop-to-drop coalescence in the dispersion band, IV: Continuous settler characteristics, Trans. Inst. Chem. Eng., 53, 61–69, 70–74, 75–80, 83–93, 1975. 16. Barnea, E. and Mizrahi, J., A generalized approach to the fluid dynamics of particulate systems, II: Sedimentation and fluidisation of clouds of spherical liquid drops, Can. J. Chem. Eng., 53, 461–468, 1975. 17. Glasser, D., Arnold, D.R., Bryson, A.W., and Vieler, A., Aspects of mixers settlers design, Min. Sci. Eng., 8, 23–31, 1976. 18. Barnea, E. and Mizrahi, J., On the “effective” viscosity of liquid-liquid dispersions, I&EC Fundam., 120, 1976. 19. Barnea, E. and Mizrahi, J., The Effects of a Packed-Bed Diffuser Precoalescer on the Capacity of Simple Gravity Settlers and on Compact Settlers, paper presented at the Int. Solvent Extraction Conference, Toronto, 374–384, 1977. 20. Barnea, E., The Application of Basic Principles and Models for Liquid Mixing and Separation to Some Special and Complex Mixer-Settler Design, paper presented at the Int. Solvent Extraction Conference, Toronto, 347–355, 1977. 21. Barnea, E., Meyer, D., and Wahrman, D., Logical Design of Mixers, paper presented at the Int. Solvent Extraction Conference, Liege, France, 6–12, 1980. 22. Harel, G., Kogan, M., Meyer, D., and Semiat, R., Mass Transfer Characteristics of the IMI Turbine Pump-Mix, paper presented at the Int. Solvent Extraction Conference, Denver, 26–27, 1983. 23. Cusack, R. and Karr, A., Extractor Design and Specification, Chem. Eng., 113–118, 1991.

Copyright © 2002 by CRC Press LLC

24. Toyo Soda Manufacturing Co., Japanese Patent 7,753, 1964. 25. Albright and Wilson Ltd., German Patent Application, 2,320,877, 1973. 26. Baniel, A. and Blumberg, R., in Phosphoric Acid, Slack, A.F., Ed., Vol.1, Part II, Marcel Dekker, New York, 1968. 27. Blumberg, R., Industrial extraction of phosphoric acid, Solv. Extrac. Rev., 1, 93–104, 1971. 28. Blumberg, R., Meyer, D., and Mizrahi, J., Development and implementation of solvent extraction processes in the chemical industries, paper presented at the Int. Solvent Extraction Conference, The Hague, 1386–1408, 1971. 29. McCullough, J.F., Phosphoric acid purification: comparing the process choices, Chem. Eng., 101–103, 1976. 30. Mizrahi, J., IMI Technology for Cleaning Wet Process Phosphoric Acid by Solvent Extraction, paper presented at the Symp. Am. Chem. Soc., 1973. (The data in Figure 4.1 was included by Blumberg, R. in a communication to Isr. Chem. Eng. J., September 1973.) 31. Blumberg, R., Miscellaneous Inorganic Processes, in Handbook of Solvent Extraction, Lo, T.C., Baird, M.H.I., and Hanson, C., Eds., John Wiley & Sons, 827, 1983. 32. Slack, A.V., Phosphoric Acid, Part 2, Marcel Dekker, New York, 721, 1968. 33. Mizrahi, J., New Process for Phosphoric Acid from Phosphate Rock and Hydrochloric Acid Via Ferric Phosphate, paper presented at the Int. Solvent Extraction Conference, Barcelona, 1999; also Israel Patent Application, 120,963, 1997. 34. Baniel, A., Bkumberg, R., Haidu, K., U.S. Patent 4,275,234, 1971. 35. Baniel, A. and Gonen, D., European Patent 91304805, 28.5.91. 36. Hirsch, M., Hirsch, I., and Mizrahi, J., Production of white carbonate paperfillers by a new ultra-fine wet grinding technology, Ind. Miner., 67–69, 1985. 37. Mizrahi, J., Separation mechanisms in hydro-cyclone classifiers, Brit. Chem. Eng., 10, 686–692, 1965. 38. Cohen, E., Beaven, C.H.J., and Mizrahi, J., The residence time of mineral particles in hydro-cyclones, Trans. Inst. Miner. Met. (London), 129–138, 1966. 39. Mizrahi, J. and Cohen, J., Studies of factors influencing the action of hydrocyclones, Trans. Inst. Miner. Met. (London), 318–330, 1966. 40. Mizrahi, J. and Goldberg, M., Computer simulation of unflocculated hindered settling, Isr. J. Tech., 318–392, 1969.

Copyright © 2002 by CRC Press LLC

chapter 5

Process definition and feasibility tests The first review of the proposed idea was done inside the R&D group (see Chapter 2, Section 2.2.1). It was shown in that review that the process can “make sense,” did correspond to a real need, and, on the face of it, was not scientifically incorrect. As a result of these early consultations, a “green light” was given to the promoters’ group for the commissioning of the literature survey, for the preparation of this preliminary process working definition and for their formal presentation for a second review in a larger forum.

5.1 Translation of the idea into a process definition 5.1.1

Scope of the preliminary process definition

An essential starting point for any development program is a preliminary process definition, which will allow: • Bringing everybody concerned to an explicit common reference basis • Illustrating a concrete venture in order to develop the interest of the hard-nosed decision-makers in the continuation of the work • Outlining a proper experimental program and starting its detailed design This field of synthesis and design of chemical processes has been the subject of a number of excellent theoretical textbooks.1–4 These manuals can be useful mostly for the analysis and understanding of the fundamental principles, and for the definition of the data which would be needed for the use of the sophisticated models available. Unfortunately, at the beginning of a new development program, most of such data would have to be assumed. Therefore, this preliminary process definition should be assembled and presented by an experienced process engineer who, in addition to the general knowledge published, would use: Copyright © 2002 by CRC Press LLC

• • • •

Personal interaction with the inventors and promoters Past experience in similar cases Some reasonable assumptions (which will always be presented as such) More specific considerations, which are detailed below

The written preliminary process definition document also will include: • Results from a comprehensive literature survey describing what is generally known in this particular field • Division of the process into defined sections and interconnecting streams, as shown on a block diagram • Calculation of the first process material and heat balances (the socalled “revision 0”) • Definition of at least one feasible implementation scheme • Projection of an industrial implementation framework and timetable • A detailed list of critical feasibility tests

5.1.2

Comprehensive literature survey

The inventors and the promoters have probably already done the best literature survey they could with the means available to them on the core aspects of their proposal. Now that the field of interest has been both enlarged and more focused with the participation of additional experienced professionals, a renewed literature survey can be commissioned, in parallel to the other work described below (see also Chapter 3, Section 3.6.1). This publication’s search can be subcontracted nowadays to specialists or to academic libraries where it is done by computer screening of large databases, according to agreed “key-words.” The first result of such screening is generally a very long list of items, including titles, authors, journal, date of issue, language, and possibly a couple of lines of abstract. A first manual selection has to be made from the computer’s output, according to some criteria to be agreed upon. However, the ordering and collection of workable copies of the selected publications (and their translation, if needed) could be sometimes lengthy and expensive. The senior process team should devote a continuous effort to supervising such screening and to the study of these copies/references as they arrive, looking in particular for any factual information, and for any possible numerical correlation of the included relevant experimental data. In addition, such analysis may give some interesting hints about the reasons for any previous research work on this subject and about their potential projection on the industrial scale up to now. The continuous recording and distribution of the results and the analysis from this survey to the core team and to the relevant consultants had often provoked important practical responses and proposals concerning the work at hand. In the end, all such findings have to be summarized and included in the material submitted to the second review.

Copyright © 2002 by CRC Press LLC

5.1.3

Block diagram

The overall process will be separated, as far as possible, into different sections and represented in a block diagram with numbered interconnecting process streams. This division is very important for all the following work and, therefore, it needs to be carefully devised so that each block ( = section of the whole process) would contain, as far as possible, only one well-defined operation. In this context, a section is a definite part of the process in which the flow rates and compositions of the exiting streams are determined uniquely by: • Flow rates and composition of the entering streams • Operating conditions that can be controlled by the operator, such as temperature, pressure, residence time, velocities, reflux ratio, and the like The presence of recycle (reflux) streams between certain sections and the exact location of their return point are very important aspects in many processes. The two typical examples given below have been chosen in order not to trespass into any actual process or new technology handled by an operating company. First Example — A typical illustration of a block-diagram is given in Figure 5.1 as an example describing a new process, which was not completely developed, for producing diammonium phosphate (DAP) from phosphate rock, HCl solution, phosphoric acid, and ammonia following a patented solvent extraction process.21 However, this proposed process also incorporates a new process concept, which is hereby offered to the consideration of the readers, as it may have applications in many other fields. Instead of an organic solvent cycle circulating inside the plant, there is an internal cycle of ferric ions (in various forms) kept inside. In short, the incoming HCl solution (stream 1) encounters a ferric hydroxide cake (stream 2, with some solid impurities originating from the phosphate ore) and dissolves it, giving a FeCl3 solution. The solid impurities are taken out and the hot FeCl3 with some HCl excess (stream 3) is used to dissolve phosphate ore (4). FePO4 is precipitated and separated (6) from the resulting CaCl2 solution (5). The washed FePO4 cake is dissolved in WPA (7) — “wet” phosphoric acid — as a mixture of soluble mono- and diferric phosphate (8). Ammonia (9) is added to that mixture and to a mother liquor DAP recycle (10); more DAP is formed and ferric hydroxide is precipitated. The later is filtered and recycled, the DAP solution (11) is cooled and crystallized, and the DAP crystals are separated and dried (12). In total, half of the phosphoric acid in the final product originated from the reaction of HCl with phosphate rock. The ferric ions are acting as a separation tool between phosphoric acid and the resulting CaCl2 and other impurities, including most of the impurities from the WPA, which precipitated in the higher pH section. This proposal is also illustrated as an example of a “black box” in Chapter 10, Section 10.4.3 and Figure 10.3.

Copyright © 2002 by CRC Press LLC

insol. waste

HCl solution

waste solids separation

Hydroxide dissolution

filtrate

2

solid / liquid separation

slurry

DAP filtrate

3

Phosphate leaching

4

9

motherliquor

Ammonia

centrifuge 10

8

FePO4 separation

6

FePO4 dissolution

FePO4 cake

CaCl2 brine

11

coolingcrystallyzer

DAP Reactor

phosphate

5

ferric hydroxide cake

1

12

DAP crystals

7

Phosphoric acid

Figure 5.1 Process block diagram for a DAP process.

Second Example — This example of a process block diagram (Figure 5.2) is the Gorin-Mizrahi patented process8 for the recovery of zirconium from the natural zircon mineral. (See Chapter 1, Section 1.4. A more detailed discussion of the process choices and issues in the high-temperature sections can be found in Chapter 6, Section 6.4.4 and in Figure 6.9.) A commercial grade of zircon heavy sand is finely ground and mixed with a concentrated solution of CaCl2, granulated and dried at 180 to 300°C. The process flow sheet of that section is illustrated in Figure 7.5, Chapter 7. In these free-flowing aggregates, the CaCl2, with up to 6% water, is distributed in intimate contact with the solid surfaces. The granules are heated and calcined successively in two rotating kilns in series, at different temperatures, in direct contact with combustion gases. In the first kiln, the calcium chloride melts at 782°C, reacts with the zircon solids and with the water vapor, and decomposes, giving very active CaO while releasing HCl into the exit gas stream. This HCl is then absorbed adiabatically to give an HCl aqueous solution while scrubbing the exit gases. The overall reaction is completed in the second kiln at 1400°C. The reacted clinker is quenched in water and ground, then partially attacked with the recycled HCl solution. The calcium silicate and other impurities are dissolved in the waste solution and only zirconium oxide remains in the solid phase, which is filtered,

Copyright © 2002 by CRC Press LLC

ground zircon

CaCl2 solution

slurry mixing and granulation

granules drying

water

hot gases

adiabatic absorption flue gases

first kiln hot gases second kiln

HCl makeup HCl solution

quenching and grinding

product drying

wash water leaching and filtration

waste stream

Figure 5.2 Process block diagram for a zirconium process.

completely washed, and dried. The basic technology of all these operations is more or less conventional, and the novelty of the process resides in the exact conditions in the calcination, which give a liquid–solid reacting front and an intermediate double salt of calcium silicate and calcium zirconate. More details on the high-temperature chemistry are discussed in Chapter 6.

5.1.4

Quantitative definitions of the different sections

For each section, the quantitative definition should consist of two parts, which have to be detailed in the textual description and presented together with the block diagram, in addition to the available data or the “agreed assumptions.” • Formal characterization of the prevailing mechanisms in the generally accepted terminology (and in more detail if there is any doubt) • Quantification of the aims to be achieved This physical-chemical mechanism can be, for example: • • • •

A chemical reaction11–16 A heat/mass transfer operation17,19 A separation operation (see Chapter 6, References 1 to 9) It also can be a conventional material handling or a storage operation

Copyright © 2002 by CRC Press LLC

For example, such mechanisms can be defined as: • Homogeneous (one-phase) mixing and equilibration of certain streams • Neutralization reaction resulting from mixing and reacting different phases • Multiple-stages, countercurrent, liquid–liquid extraction battery, which involves in every stage the mixing of liquid streams in order to achieve mass transfer of specific components and approach to equilibrium, followed by the separation of the resulting liquids • Similar batteries for different process functions, such as “extract wash” or “back extraction” in solvent extraction • Concentration of a solution by evaporation and the subsequent cooling of the resulting solution and of the evaporated condensate • Filtration and washing of crystals on a wedge-wire screen centrifuge • Drying of the solids from a wet filter or centrifuge cake • Separation of a solids stream into (different, defined) size fractions, etc. The aims to be achieved in each separate section also should be given quantitative indexes, such as the minimum concentration, the specification in the exiting stream, the acceptable upper energy consumption, the minimum recovery of a valuable component, an acceptable waste composition for disposal, and so forth.

5.1.5

Process calculations for the preliminary process definition

The chemical engineering calculations can follow well-established procedures, which are listed and detailed in some of the basic reference books.6–12 As far as possible, these calculations will be done in parallel and will include: The correlation and analysis of all the relevant data already available, either from the previous work of the inventors/promoters, of the process engineer, or from previous publications on related subjects, which may have been obtained from an extensive literature search (see Section 5.1.2). These correlation formulas can be carefully extended by extrapolation, if needed, as long as this is clearly recorded as a provisional mean. The formulation of the quantitative relations (known or assumed) that can affect the process mechanism for each separate section (e.g., the yield of reaction, the solubility). These relations sometimes can be based on the theoretical thermodynamics or the physical chemistry knowledge, as given in basic reference books.18,19 But it is seldom that in the first stages of a development effort, these sources could be useful or justifiable. Therefore, quite often at this stage, some of these assumptions have to be based on known analogies with other processes from the process engineer’s own background (although he may have to keep some of these references secret, and personal trust will be essential).

Copyright © 2002 by CRC Press LLC

The evaluation of the effect that each of the different operating variables may have (for instance) on distribution, solubility, recovery, heat effects, and their combinations … for each separate section, on the basis of the above quantitative relations. The preparation of computer spreadsheets for each separate section with the preliminary design material balance and with the heat balance (if the heat effects are important in such sections, which is not always the case), on a reasonable arbitrary basis, e.g., 1000 units of the main raw material. These balances are prepared by conventional chemical engineering calculations, and this task should give the process engineer a good insight into the importance of each of the different factors in the play-offorces inside such process (leverage). Of course, there are interactions between the different spreadsheets, since each section starts where another ends. These spreadsheets can be easily converted at a later stage to any other basis needed. A list is also prepared of the critical feasibility tests that should be done to define or confirm some of the high-leverage assumptions taken in these calculations before these are presented for the review (see Section 5.3).

5.1.6

Presentation of one feasible implementation formula

This description is, in fact, a concrete, possibly optimistic, illustration of the implementation of the concept, if it could be made to work as intended. The starting point is the integration of the above quantitative assumptions into one possible implementation case, describing an operating plant with the specifications of the raw materials and the products, the material and heat balances, the process control, the recoveries, the disposal or treatment of the resulting waste streams, the relevant safety aspects, the choice of materials of construction for the contact equipment, and so on. Typically, if one of the raw materials have to be transported in large lots, the material handling facilities and storage can be significant. For example, it may be projected that the new process would need large heat exchangers made from expensive materials (graphite, glass-lined, tantalum), or that the rapid scaling of such heat exchangers should be expected, considering the composition of the solutions being heated. In such case, one design option could be to resort to the technology of organic “heat carrier” (stable liquid or vapor hydrocarbons) and, if this is considered a practical design possibility, it should be studied, defined, and included in the experimental program (see Chapter 6, Section 6.2.1)

5.1.7

Possible industrial implementation framework

A projection of one possible industrial implementation framework (known or assumed) is presented for the proposed novel process with its specific features, such as the possible site, the scale of production, the equipment size and function, the different raw materials available, existing connections

Copyright © 2002 by CRC Press LLC

to critical services, any possible synergetic coproduction. Such projection should help specify the technological factors which will have to be solved. For example, the maximum supply temperature of the cooling water that could be reliably procured or produced in this site depends on the recorded climatic conditions. It often has been found in warm countries that such temperature can have a critical importance for the design of a new process based on evaporation or distillation under vacuum, or involving materials with low boiling points. If the normal cooling water at this site isn’t cold enough, the cost of supplying artificially chilled water would have to be included or the process scheme radically changed.

5.1.8

Timetable

A reasonable projected timetable for the whole development and implementation project prior to the plant’s start-up and to the market penetration is also essential to the decision-making forum, in order to: • Evaluate the availability of the required resources • Coordinate the assistance of the many different support groups and experts • Connect with the market projection studies

5.1.9

Important note

Obviously, in the beginning, the above preliminary “process working definition” would have to be based, in a large part, on weighted and explicit assumptions and on previous professional experience. Its purpose is to focus the team’s attention as well as to plan any future work on the limited scope of application that is of interest in real life, and to allow for a more effective allocation of the available industrial R&D resources. It is agreed that this process working definition will be progressively changed, enlarged, and refined in future numbered revisions, as more information will be gathered and analyzed, and more promising avenues will be defined. However, such a working method has not always been accepted by all. In many cases, it has been resisted and even ridiculed by senior scientists, who were used to the open-ended academic research approach, claiming, “What do we really know for sure? Let’s collect some data first and then we will see what kind of process will result.” Bluntly speaking, their noncommittal approach represents the more serious danger to the development of any novel chemical process (like gambling without knowing the odds). The least damage can be that a large part of the experimental data collected would be outside the scope that is relevant for implementation, resulting in a loss of time, resources, and good will. More serious damage could be caused if wasting of time and repeated abortive trials would erode the corporate management’s interest and promising ideas would be lost.

Copyright © 2002 by CRC Press LLC

5.2 Critical and systematic review of the process definition 5.2.1

Review forum

This second critical review aimed at reaching operative decisions with regard to the next steps (“no or maybe”) is generally called in by the decision-maker who can be, for instance, the managing director of the R&D organization, the corporate vice president for new business, or the director of the funding committee, and comparable in function or in title. This discussion is conducted in a larger forum with ranking colleagues of the inventors/promoters (“peers”) and with outside experts who are invited, if and as needed. It should cover all the essential elements, such as the technology and patents, the corporate strategy and markets, the profit potential, and the capability to handle the proposed development program with the resources available. In a larger organization, there can often be a competition for the priority in allocations between different projects. The raising and discussion of the more “difficult” subjects in such a review can sometimes be unpleasant, if it is seen as a personal criticism among working colleagues with complex human relations. Therefore, it can be useful to appoint a merciless “devil’s advocate” to present the pros and cons of the problematic aspects, in advance and in writing. His contribution would then avoid wasting time in rhetoric and personal maneuvers. As a result of the first part of this review, a number of specific activities should be approved for immediate execution, and an additional meeting of the critical review forum generally would be reconvened by the “decisionmaker” a few months later. This additional meeting will then study and review the different reports prepared on the feasibility tests, the patent discussions, and the clarification of the critical economic factors.

5.2.2

Fundamental process issues

In this review, all the fundamental process issues should be raised and focused on, and a list prepared all information requested to arrive at the final clarification of these issues in the future. • Any apparent reason why “this cannot work?” Have we forgotten anything? • Could any “wishful thinking” bias be included in the conceptual reasoning? • Does any quantitative factor have a leverage large enough that might turn the balance critically away in the wrong direction? • How to deal with any relevant existing patent claim (see below). • Is the process flexibility sufficient to accommodate some changes that could be needed to bypass such typical claims, should they appear in the future? • Is there a sufficient profitability potential (see below)?

Copyright © 2002 by CRC Press LLC

As a result of this review, a list of critical tests for process feasibility should be defined and agreed upon (see Section 5.3).

5.2.3

Patent situation

Relevant patent claims granted to another party, which may have been found in the first survey, will be discussed in this review. In many cases, such previous claims may still be avoided quite fairly, on the basis of their exact formal definition, but such constrain can dictate some changes in the development program. Unfortunately, patent applications are not made public for 2 to 3 years after their filing and many inventors are using perfectly legal tactics to delay their publication, so one also may have to look for hints from professional circuits and to prepare for eventual “surprises” from this direction. The promoters will also draw up and present to this forum a very detailed list of every possible patentable claim for the new process. After this review and the approval to proceed farther, this list will be used for additional discussions with the patent attorneys and the specialists, and after an elimination and selection procedure, for the drafting of a comprehensive patent application (see Chapter 8). Note that all the formal procedures of patenting are now quite rigid and very few procedural decisions are really needed, apart, of course, from the delicate issue of the names listed as the inventors. In particular, since the establishment of the PCT (Patent Cooperation Treaty), all international applications can be based on the examination done in one patent office (i.e., in Washington, D.C.). Therefore, the main deliberations with the patent attorneys on any new patent and the resulting decisions will be related to the exact formulation and wording of the claims in the application. Even among patent attorneys, there is a certain degree of specialization, since only a professional with a real knowledge of the particular scientific technological field can contribute effectively to such formulation. However, in some corporations, any talk about a patent application would immediately involve their top lawyers (who usually are very busy) and, therefore, the patenting procedure could become very slow and very expensive without any real additional contribution. This frustrating situation is often a pitfall. As a curiosity, in the U.S., a corporation is legally bound to pay the inventors cash in return for their assignment of the patent rights, even if the inventors are their own employees or their regular consultants under contract. This cash payment is often done in the form of a brand new $1 bill, which is handed over to each inventor with his signature, and which is often framed together with the invention certificate. Copyright © 2002 by CRC Press LLC

5.2.4

Profit potential

It also should be shown and agreed in this review that the profit potential of the proposed process could be attractive enough to justify the estimated costs of the next stages of a development program. There can be many different definitions of the profit potential, which are used by different corporations and industrial sectors in various countries and tax situations. In general, this profit potential should quantify the expected increase in ROI (Return On Investment) above what would be the ROI for an available, safe, no-risk investment, over a 10-year period. The profit potential, as an absolute number for a particular proposal (i.e., in dollars per year) is obviously getting higher as the contemplated scale of production or the sales turnover are higher, while the costs of the process development (the risk?) are much less affected by the size, if at all. Thus, a new process could be brilliant and sophisticated, but if its final product has only a small market potential, the straight prospects of approving funds for its development will be dim. Such a proposal then will be generally presented as a strategic investment, opening the way to … (as the popular joke says, “One can save more money by running behind a taxi than by running behind a bus). At this early stage, the standard economic calculations can only be rudimental and based on reasonable assumptions. The bottom-line results will generally not be clear-cut either way, but they will indicate mostly orders of magnitude (so-called “back of an envelope” calculation type). The debatable issues should focus on the degree of confidence that can be attributed to some high-leverage factors, e.g., sale prices, cost of certain raw materials, possibly transportation, taxes and customs duties of export/import, commissions, royalties, etc. As a result of this review, a fact-finding program will be defined to confirm or correct the required quantitative assumptions for such critical economic factors, in addition to the publicly available information. According to their specific strategic considerations, most corporations would be ready to “gamble” a certain percentage (say between 1 and 20%) of the yearly profit potential on the net costs of a comprehensive development program. Obviously, such a budget would only be released progressively, in installments, as certain objectives are achieved with positive results.

5.3 Design and execution of the feasibility tests 5.3.1

Purposes of the feasibility tests

Feasibility tests should convince the decision-makers and demonstrate that the results from the new aspects can be achieved more or less as expected in each stage of the process in order to justify a more extensive experimental program.

Copyright © 2002 by CRC Press LLC

As for the “more or less” qualification in this context, it is generally agreed that the better results cannot be obtained in the first attempts, but should be achieved more likely in the more favorable conditions that will be defined later, after an extensive process optimization. It is also realized that such demonstrations can generally be attempted only with severe limiting conditions, such as: • Small, bench-scale, batch tests • Standard or improvised laboratory equipment and analytical facilities • While starting with synthetic “clean” mixtures and reactants “from the bottle” The results from each stage can either be shown by the direct analysis of the phases obtained after the test or calculated indirectly on the basis of those analyses by accepted chemical engineering methods. For example, a wet filter cake can contain some impurities, which are dissolved into the layer of liquid that is retained on the solids. This layer can be washed out almost completely on a conventional industrial filter, but a similar washing operation cannot be done conveniently on a small-scale laboratory batch filter with the same results. In this case, the level of these impurities related to the retained filtrate can be calculated on the basis of the retained water (or of another soluble component) and then deduced from the level found in the unwashed filter cake.

5.3.2

Equilibrium conditions

In the limiting experimental conditions mentioned above, a more convenient feasibility demonstration can be achieved for those process operations that are based on equilibrium conditions. For instance, a particular vapor–liquid equilibrium system can be governing some distillation, rectification, or stripping operations in the proposed novel process. A reliable calculation of the results from these operations can be obtained on the basis of the correlation of the composition of the vapor phase with respect to that of the liquid phase at equilibrium. The theoretical background for the calculations is well established and several correlation formulas were published on the subject. A limited number of experimental points on the particular system under consideration can be interpolated quite safely and used for such process calculations. All such “points” connect the two compositions of the phases at equilibrium in specific external conditions of temperature, pressure, and the partial pressure of inert gases present. Similarly, a liquid–liquid equilibrium system can be relevant to a proposed solvent extraction process. The composition of the two liquid phases, found in a specific equilibrium test in certain defined conditions, can be translated into a distribution coefficient for each of the components of interest. The correlation of this distribution coefficient with the operating conditions can

Copyright © 2002 by CRC Press LLC

be used for calculation of a multiple-stage countercurrent process, in which the two exit liquid phases from every stage are assumed to be at equilibrium. Solids–liquid equilibrium data can regulate certain dissolution and precipitation processes, which are widely used in the inorganic salts industries and mineral treatments, and also in the production of organic crystals. In all of these processes, the determination of the quantitative relations for the solubility at equilibrium, in several conditions in the projected range, can be sufficient, at least for the preliminary process design and feasibility demonstration.

5.3.3

Scale up of reactors

The scale up of batch reactions in mixed vessels is well established in chemical engineering using the reaction kinetic curves and the definition of the mixing regimes. Continuous mixed reactors can also be designed reasonably well from small-scale batch tests, or upscaled from small continuous mixed reactors, using correction factors. Thus, the feasibility demonstration of such reactions can be based on straightforward batch reaction tests. Similarly, the mixing and the reaction between gases flowing in a pipe reactor are also relatively easy conditions for the process feasibility demonstration on a quite small scale, and for reliable scale up based on hydrodynamics conditions and residence time.

5.3.4

Physical separation operations

The scale up of solid–liquid separation equipment has been well established.21 Many continuous separation operations (solids from liquid or liquid from liquid) can be demonstrated, sized, and upscaled quite well from batch tests made in standardized conditions. In this context, the term “standardized conditions" involves the definition of a particular set of conditions which are recommended for a batch test in order to obtain applicable results. For example, the separation obtained in an industrial decanter or thickener — from a “feed” suspension into a more concentrated “underflow” slurry and a clear liquid “overflow” — can be demonstrated and quantified from a standardized settling test in a 1-liter glass cylinder starting from a well-homogenous slurry (in a thermostatic bath, if necessary). The settling curve of the upper limit of the concentrated slurry obtained from each test (see typical example in Figure 5.3) depends on the initial solid concentration, on the differences in density between the solids and the liquid, on the liquid viscosity, and on the degree of flocculation. The plotting of an empirical settling curve allows the calculation of the maximum solid concentration in the underflow, the level of entrained fines in the overflow (if any), and the horizontal area of the settler needed, per tonhour of solids, by the well-established Kinch method following Coe-Clavenger (see p. 4.121 in Chapter 6, Reference 1). This experimental procedure is used also for studying the effects of the addition and dozing of flocculating agents. Copyright © 2002 by CRC Press LLC

initial

settling curve of solids front

point of most rapid change of slope

height final

time

Tx

Figure 5.3 Slurry settling curve — Kinch procedure.

Similarly, the separation of a cake of solids (including the washing of the cake) from a particular slurry in a filter or in a centrifuge can be demonstrated and quantified on a small scale with the same type of results. Note, however, that a continuous filter is, in fact, a batch filter that happens to be moving on a belt during the filtration cycle. The difference obtained in the rate of filtration derives from the operating variables: the pressure differential on the filter or the G-forces in the centrifuge, and the hydraulic resistance of the formed cake. Flocculation, however, does not affect significantly these operations.

5.3.5

Scale-dependant and dynamic flow operations

In contrast with the operations discussed in the above three sections, a feasibility demonstration cannot be readily performed for such operations in which the results are scale-dependant, such as, for instance, the crystal size distribution obtained in a continuous crystallization (see Chapter 6, Section 6.4). The feasibility demonstration also cannot be readily performed if the mechanisms are based on dynamic flow conditions concerning mostly separations between phases (see Chapter 6, Section 6.3) In such cases, any feasibility demonstration has to be connected to a particular equipment choice, and some specific form of piloting is necessary to determine the dimensions and results. Copyright © 2002 by CRC Press LLC

A shortcut sometimes can be found if a logical analogy can be established to another known process, which is in actual operation and accessible to the R&D team or their consultants. Then, if it can be established that the new proposed process and the operating process behave more or less similarly in simple bench-scale tests, this analogy could justify further piloting. For example, the feasibility demonstration of many proposed waste stream treatments, based on dynamic flow conditions, is very problematic, but these treatments generally fall in a small number of categories.

5.3.6

Extreme conditions

It is not a simple proposition to improvise on laboratory bench-scale a feasibility demonstration for a process operation which has to be done in extreme conditions of temperature, pressure, electrical fields, etc. If such an extreme operation is an essential element in the new process, small testing equipment could probably be specially designed and operated, but this would be expensive, require a long time and expertise, and would divert the team’s attention. In certain cases, small-scale testing equipment with associated services (and valuable advice) can be rented from one of many suppliers of furnaces, kilns, autoclaves, electrostatic and magnetic separators, plasma torches, etc. These suppliers can also provide experienced engineers to perform these tests, since they are interested in promoting good will towards their knowhow. The main concerns against such services are the unavoidable secrecy leaks and possibly the geographical distances. If the extreme operation is a self-contained side element in the new process, it should be well defined, isolated, and subcontracted to one of these specialized suppliers.

5.3.7

Actual raw materials

In certain processes, it can be very important to perform such feasibility tests with the actual raw materials, since a synthetic mixture cannot duplicate exactly the complex phases structure and/or the compositions of these materials in which a large number of impurities can sometimes be involved. In many actual projects, the use of certain raw materials in later tests did result in serious nonexpected problems; for example, the precipitation of solids causing incrustation on the walls of the equipment and pipes, or a tendency to emulsify, or the precipitation of colloid suspensions, or a coloring phenomenon, etc. In other cases, the reaction kinetics with the actual solid raw materials was much slower (by orders of magnitude) than the reaction kinetics with the synthetic mixtures. If a practical solution to such troubles cannot be provided in time and included in the proposed industrial process design, the project will be “killed” sooner or later, at least in its initial form. Therefore, it is important Copyright © 2002 by CRC Press LLC

to discover these problems as early as possible by using representative samples of the actual raw materials in the feasibility demonstration tests. Unfortunately, such representative samples cannot always be procured at the start. This difficult situation was typically encountered when the new process involved the treatment of mineral concentrates from new deposits that, as of yet, have not been fully explored, or the down-stream treatment of some material that was expected from certain “future” operations.

5.3.8

Analytical difficulties

In some situations, the available analytical laboratory personnel may not have previous experience with the exact type of analyses required for these feasibility tests and they will have to learn, introduce, and calibrate new methods. This can be a lengthy procedure, and the time needed can possibly be reduced with outside help. The allocation of priority in this area, or the need to compromise on a “second-best” method, had often caused delays and personal tension.

5.4 Analysis of the results from feasibility tests When these results become available, it is advisable that the promoting team prepares and presents two separate reports, which will be studied and discussed in different contexts and review meetings (sometimes again many years later). 1. A report on the results of the feasibility tests, which should be presented in the normal format for R&D experimental reports with a complete description of all the laboratory procedures and equipment, all data collected, and, in particular, any observation about unusual features. 2. Another report discussing the significance of the experimental results and observations from these tests, as regards the process feasibility demonstration and the design issues of the proposed novel process. This report may also include further chemical engineering calculations or economic evaluations. The discussion in this report should also define exactly a limited range for each of the variables to be covered in any future experimental program as basis for the process design.

5.5 Second review of the process definition The forum of the second critical review is generally reconvened by the decision-maker along with participating colleagues (“peers”) and the calledin experts to study and review the different reports that were distributed in advance and in writing. The reports include: Copyright © 2002 by CRC Press LLC

• Feasibility demonstration tests • Patent discussions • Clarification of the critical economic factors This second review can end up in one of the following ways: A. In most cases, the analysis and discussion of these reports would give the “green light” for proceeding with the experimental program (see Chapter 6), the preliminary process design (see Chapter 7), and the economic analysis (see Chapter 8). If this program is agreed upon, this would be the appropriate time to formalize a contract transmitting the implementation rights from the promoters to the corporation. The corporation would then confirm the appointment of a project manager to carry on the responsibility of the future program. This manager has most probably already been a member of the process evaluation team up to this point. B. In other cases, the results and calculations could indicate a need to correct or readjust some of the initial process working definitions. The file would be given back to the promoters for the repeat of certain tests, additional reports, and back for another similar review. C. In certain cases, the continuation of the project could not be logically justified in the present framework, and it would be terminated at this point, at least until the promoters’ prayers for some surprising development are granted.

5.6 Worth another thought • An essential starting point is a preliminary process definition to bring everybody concerned to an explicit common reference basis, illustrate a concrete venture, outline a proper experimental program and start its detailed design while focusing on the limited scope of application that is of interest in real life, and allowing for a more effective allocation of the available industrial R&D resources. This process working definition will be based, in a large part at the beginning, on weighted and explicit assumptions and on previous professional experience, but it will be progressively changed, enlarged, and refined. However, such a working method has often been resisted and even ridiculed by senior scientists who were used to the open-ended academic research approach. • A novel process could be brilliant and sophisticated, but if its final product would have only a small market potential, the straight prospects of approving funds for its development will be dim. Such a proposal will then be generally presented as a strategic investment, opening the way to new potential products. Copyright © 2002 by CRC Press LLC

• The feasibility tests should convince the decision-makers and demonstrate that the results from the novel aspects in each stage of the process can be achieved more or less as expected to justify a more extensive experimental program.

References 1. Biegler, L.T., Grossman, I.E., and Westerberg, A.W., Systematic Methods of Chemical Process Design, Prentice Hall, New York, 1997. 2. Douglas, J.M., Conceptual Design of Chemical Processes, McGraw-Hill, New York, 1988. 3. Duncan, T.M. and Reimer, J.A., Chemical Engineering Design and Analysis: Introduction, Cambridge Press, London, 1999. 4. Seider, W.D., Lewin, D.R., and Seader, J.D., Process Design Principles: Synthesis, Analysis, and Evaluation, John Wiley & Sons, New York, 1999. 5. McCabe, W.L., Smith, J.C., and Harriot, P., Unit Operations in Chemical Engineering, 5th ed., McGraw-Hill, New York, 1993. 6. Hicks, T.G., Ed., Standard Handbook of Engineering Calculations, Section 6. Davidson, R.L., Chemical Engineering, McGraw-Hill, New York, 1972. 7. Clarke, L. and Davidson, R.L., Manual for Process Engineering Calculations, McGraw-Hill, New York, 1975. 8. Branan, C.R., Rules of Thumb for Chemical Engineers, 2nd ed., Gulf Publishing Co., 1998. 9. Meyers, R.A., Handbook of Petroleum Refining Processes, 2nd ed., McGraw-Hill, New York, 1996. 10. Perry, R.H. et al., Chemical Engineers’ Handbook, various editions, McGrawHill, New York, 1999. 11. Froment, G.F. and Bishoff, K.B., Chemical Reactor Analysis and Design, 2nd ed., John Wiley & Sons, New York, 1990. 12. Smith, J., Chemical Engineering Kinetics, 3rd ed., McGraw-Hill, New York, 1990. 13. Schmidt, L.D., The Engineering of Chemical Reactions, Oxford University Press, Oxford, 1998. 14. Fogler, H.S., Elements of Chemical Reaction Engineering, 3rd ed., Prentice Hall, New York, 1998. 15. Levenspiel, O., Chemical Reaction Engineering, 3rd ed., John Wiley & Sons, New York, 1998. 16. Butt, J.B., Reaction Kinetics and Reactor Design, 2nd ed., Marcel Dekker, New York, 1999. 17. Honig, J.M., Thermodynamics Principles Characterizing Physical and Chemical Processes, 2nd ed., Academic Press, New York, 1990. 18. Klotz, I.M. and Rosenberg, R.M., Chemical Thermodynamics: Basic Theory and Methods, 6th ed., John Wiley & Sons, New York, 2000. 19. Coulson, J.M. and Richardson, J.F., Chemical Engineering Fluid Flow, Heat Transfer, and Mass Transfer, different editions, last 6th eds., Butterworth-Heineman, Oxford, 1999. 20. Rohsenow, W.M. et al., Handbook of Heat Transfer, 3rd ed., McGraw-Hill, New York, 1997. 21. Purchas, D.B., Ed., Solid/Liquid Separation Equipment Scale Up, Upland Press, Croydon, U.K., 1977.

Copyright © 2002 by CRC Press LLC

22. Mizrahi, J., New process producing phosphoric acid from phosphate rock and hydrochloric acid – via ferric phosphate, paper presented at the Int. Solvent Extraction Conference, Barcelona, July 1999. Also Israel Patent 120,963, June 1997.

Copyright © 2002 by CRC Press LLC

chapter 6

Experimental program 6.1 Basis 6.1.1

Experimental program purposes

Main Purpose — At this stage of the process development, the main purpose of the experimental program is the collection, the correlation, and the presentation of the design data that is specifically needed for the design and optimization of the new process, as defined and in the limited range of variables of practical interest. It is important to note that the scope of the investigation can depend also on the variability of the particular function under consideration. It can be found, after the first series of tests, that the already mentioned range of specific interest happens to be located in one part of the function where sharp changes can be seen from a first plotting of all available information. In such case, it is advisable to enlarge the scope of investigation in order to assure a reasonable reliability when interpolating between experimental points. The experimental program, therefore, is formulated in relation to the perceived needs in one particular situation. Unexpected Problems — Another important purpose includes the observations of possible, but unexpected problems, that can occur and that should be dealt with. These can be, for example, difficulties in the separation between phases, slow rates of reaction/mass transfer, colloidal precipitates, unwanted color, etc. (see below). The experimental staff should be instructed to observe carefully and to call their supervisor whenever anything seems unusual. The “top” R&D managers are often seen circulating between the benches when such tests are done to obtain a personal “appreciation” of the behavior of the reacting or separating mixtures. Preparation — In addition, the preparation of relatively large representative samples of certain products or of certain intermediate phases are often needed for further specialized tests, for market surveys, or just to “show around” in the promotion contacts for the project. Therefore, generally all the materials resulting from these tests should be well packaged, labeled, and stored for future use.

Copyright © 2002 by CRC Press LLC

6.1.2

Different sections

The separation of the process into the different sections was already represented in the process “map” (block diagram), with the formal definition of the prevailing chemical mechanism, of the separation between phases, and of the process results expected in each separate section (see Chapter 5, Section 5.1). Before starting the experimental program, these definitions should be discussed systematically, well understood, and agreed upon among the inventors, the process engineers, and the senior experimental staff. The calculation methods that will be requested for the process design of each operation will be formulated and agreed upon by the whole process engineering team. For this purpose, use the manuals and textbooks listed as references in Chapter 5, as well as for those concerned with the separation processes listed in this chapter.7–10 Note that many of the operations in any chemical plant can possibly be designed on the basis of conventional know-how, with the specific input of only a few specific physical properties; for instance, all the sections concerned with material handling, liquid flows, blending, packaging, spraying, gas compressing, steamboiler, cooling tower, etc. Thus, the experimental program will not be concerned with such operations at this early stage of process development.

6.1.3

Quantitative data needed for process design

Guided by the definitions, the process engineers also should prepare a list of all the quantitative data that will be requested for the process design of each operation. Some parts of this design data may be already available in the files from the previous analysis of the results of the feasibility tests, from the promoters’ own sources, or from earlier publications in related fields. The systematic organization of what is available will allow the delimitation of the missing data that should be generated in the experimental program presently considered. Discontinuities are often found between the sets of data obtained from different sources, as it may be seen when these are plotted on a common graph, due to the differences in experimental or analytical techniques. Therefore, in this analysis and determination, it is preferable to allow for significant overlapping to arrive at a reasonably reliable common function. (In the author’s considered opinion, there is not much point in designing and starting any significant experimental program without performing first this process engineering analysis.)

6.1.4

Format

At the same occasion, it would be useful if the process engineering team can specify exactly the preferred format for the results on these data to be used in the experimental reports to allow their direct application in the Copyright © 2002 by CRC Press LLC

process calculations. As there can be many parameters in each stage, the primary variables should be indicated in the order used in the calculations (see Section 6.2.1). This early specification of the format can avoid or reduce the communication problems and the waste of time devoted to clarifications and recalculations, which often happens between the process engineering team and the experimental group. In many cases, these two units can also be situated in geographical locations far apart and may not be able to meet frequently face-to-face. It would also be useful to decide as far in advance as possible the preferred order for the generation and transmittal of partial data to the process engineering team in separate, successive, numbered reports. This demand can appear to be trivial, but this has been, in fact, a sore point in many projects. If certain parts of the process design work can be started and advanced before all the data is transmitted in one big bound report, some of the pressure can be relieved from a serious bottleneck in process design. It generally happens that as soon as the final experimental report is issued, everybody wants to know all its implications on the new process, the plant under consideration, the economic parameters, etc. On the other hand, the preparation of serious answers takes time and experienced process engineers are scarce.

6.1.5

Representative raw materials

As far as possible, these R&D tests should use representative samples of the actual raw materials, fuel, water, reagents and additives, filter aid, active carbon, and IX resins that would be expected to be used in the final plant. As discussed previously, a synthetic mixture of pure laboratory chemicals from the bottles on the shelf cannot duplicate in many cases the exact complex physical structure and/or the large number of impurities in these raw materials. Similarly, whenever it is intended to use combustion gases in direct contact with the process streams (e.g., in a calciner or a dryer), the exact composition of the fuel can be significant. Such combustion gases can contain ash particles or gaseous impurities that can contaminate the products or would need to be treated in the waste streams or can accumulate in the plant. It has often been found in real cases that the use of certain raw materials in such tests did result in serious, nonexpected problems. For example, in solvent extraction processes, some impurities can precipitate as fine solids causing the liquid–liquid mixture to emulsify and, thus, preventing the normal operation of the process. In one particular case, the raw material contained an impurity with oxidizing power, which attacked and destroyed the organic extracting reagent used. In hydrometallurgy and salt processes, the precipitation of solids that stick or build on the walls of the equipment and pipes can stop a plant. Some natural streams can, when heated, release some dissolved noncondensable gases, which may Copyright © 2002 by CRC Press LLC

disturb the vapor–liquid equilibrium, and/or (at least) need to be collected and vented properly. Other problems can be the precipitation of colloid suspensions, coloring phenomena, etc. Such troubles may be serious enough to “kill” sooner or later the proposed industrial process at least in its original form, so it is very important to discover and diagnose them as early as possible. Possible solutions can include pretreatments or even changing the source of the problematic raw material. Unfortunately, it often happens that, despite all reasonable efforts, representative samples of all the actual raw materials cannot always be readily procured from the beginning of the experimental program. First Situation — This difficult situation has been typically found, for instance, when a new mineral deposit was being explored and small samples of the expected raw materials can only be separated and reconstituted from small drilling rig cores in quantities hardly enough for the analytical and preliminary bench tests. Second Situation — Another typical case has been the development and demonstration of a proposed process, which was intended to handle an effluent stream expected from a “future” operation that was still at the design or construction stage. Third Situation — This situation has been also encountered now and again in the development of large-scale biotechnology processes, which typically consist of two separate parts: 1. The fermentation section, which is producing a broth containing a valuable product (e.g., a carboxylic acid) 2. The recovery section, intended for the separation of such valuable product from the broth into a pure, concentrated, marketable form These two sections are generally developed and even designed separately by two specialized groups, and they are often built in different plots “across the road.” Obviously, all the characteristics of the new recovery process are derived from the exact composition of the expected fermentation broth, as defined at the time of the project justification. But it often happened that while the “recovery” group was developing, designing, and building their processing unit, the “fermentation” group was continuing their efforts to improve their part. They would aim at a better productivity (which is the average production rate per unit volume of fermentor) and/or a better yield on the raw materials. This can be quite natural from their point of view, but the resulting changes in the broth composition (mostly with regard to the associated impurities) can have serious (negative) effects on the recovery process, as developed. A mutual understanding and coordination between these two groups should be obvious, but often can be delicate in real life and may have to be imposed from above. To be fair, the fermentation group is not always informed of the downstream development (“What do these biologists know about our separation processes?”). But, at least in one case known to the Copyright © 2002 by CRC Press LLC

author, when the fermentation group was informed that one of the impurities generated by the microorganism constituted a very serious separation problem, they found a genetic “trick” to prevent this particular impurity and replace it with a less problematic one. In such cases, where representative samples of the actual raw materials cannot always be readily procured from the beginning, an experimental program on “synthetic” mixtures can only be done as an exploratory work aimed at the preliminary process design. The results need to be clearly marked as such, and this situation reflected in the safety factors included in the economic analysis. Repeated tests should be scheduled for later, in the exact selected conditions, as soon as the actual representative samples can be obtained.

6.1.6

Classification of missing data

The data missing at the beginning of the experimental program can be divided into three main categories according to the testing techniques that will be required in obtaining the results, as discussed below, and the design methods for: • Operations based on chemical equilibrium data, (as reviewed in the comprehensive book by Henley and Seader2 • Operations based on dynamic flow conditions • Operations that are scale-dependant, i.e., the results depend on the absolute size of such equipment

6.2 Chemical equilibrium data (See basic reference books on separation processes and, in particular, ones on the equilibrium stages, which are the most useful tools in new process development.1–7)

6.2.1

Vapor–liquid equilibrium system

Vapor–liquid (reversible) equilibrium systems are used in unit operations, such as distillation, rectification and stripping, evaporation, and condensation. (Note that gas–liquid equilibrium systems, which are relevant in unit operations dealing with scrubbing, cooling, etc. of a gaseous stream, have some similarity. However, these will be discussed separately, in the next section.) Data needed for process design are obtained by correlating the compositions of both phases at equilibrium in certain conditions of temperature, absolute total pressure, and the partial pressure of noncondensable gases (inert, nonreactive) that may be present (assuming that such partial pressure does not exceed 70 to 80% of the total). The pair of compositions for both phases can be obtained from a total reflux test, where the vapors from a boiling liquid phase are totally condensed at the same absolute pressure and all the condensed liquid is returned to the

Copyright © 2002 by CRC Press LLC

boiling liquid. As equilibrium is established, samples are withdrawn from both the boiling liquid and the refluxed condensed liquid, and completely analyzed for all components. If there are only two components, the plotting of the results is straightforward. But since, in most cases, there are more than two components present in each phase, one has to decide from the start which two components are the variables under study and which other components are to be considered as parameters for the purpose of the present process design, together with the obvious physical parameters, such as the temperature, absolute pressure, and partial pressure of noncondensable gases. All the parameters have to be kept constant in each series of tests, to obtain a cross section for two variables. One may see that the experimental program for a typical system with four to six components can become very complex unless one limits from the beginning the ranges of practical interest (see Chapter 5, Section 5.1). Once this chosen range is covered with a limited number of experimental “points,” the numerical results can be interpolated quite safely into a mathematical function by using one of the published theoretical correlation formulas. The correlated function then can be used for the process calculations of the multiple stages equilibrium system in one of its forms: countercurrent, cocurrent, or crosscurrent. The process design can be done using the “theoretical stages” concept, and then translated into an equipment design, by relying on the correlation linking the “height of theoretical unit” with the operating conditions and the details of the chosen equipment. Such correlation has been published for several basic designs or may be obtained from the suppliers of more specialized equipment. When there are many condensable components from the beginning (as in petroleum processing, as an extreme case), one may have to “cut” the mixed feed by a coarse separation into two or three ranges (“heavies” or “lights”) and to treat each range as a separate problem with recycles at the starting point. A special case is the concentration of a solution containing nonvolatile solutes by evaporation of water (or another solvent), leaving the nonvolatile solutes in the concentrated solution. The vapor phase contains only one component, but the concentrations of the solutes into the liquid phase increase gradually, decreasing the partial pressure of the water. In such case, the important data are the quantitative link between the absolute pressure and the boiling temperature of the solution and the concentrations of the solutes below their saturation limit. These data are essential, for example, for starting the design of energy-efficient, multiple-effect evaporators, which are a critical element in many processes (e.g., salts and sugars). An equally important result of such tests can be any observation about the precipitation of certain solids from the liquid, and the form and behavior of such solids, in particular as to their incrustation inside equipment and pipes, or on heat exchangers surfaces (for their composition, see Copyright © 2002 by CRC Press LLC

Section 6.2.3); as well as the conditions of the release of any noncondensable gases dissolved in the feed solution. Another important field of process development is concerned with the separation between two volatile components that cannot be obtained directly due to the presence of an azeotrope or another particular feature of the equilibrium curve. (As a reminder, at the azeotropic point, the compositions of the liquid and of the vapors are identical.) A well-known case is the system HCl-water (already mentioned in Chapter 4, Section 4.3.2) which is dominated by an azeotrope at 20 to 22% HCl (the exact number depends on the absolute pressure). Every ton of HCl generated “below the azeotrope” is accompanied by at least 4 to 5 tons of water at its maximum practical concentration and this feature may prevent or limit its use in other processes. “Breaking the azeotrope” means obtaining a more concentrated solution that can be handled at ambient temperature (say about 30 to 40% HCl) or even a 100% dry HCl gas, if needed. Such a result is possible by using a cycle of CaCl2 brine in a close cycle, as the brine absorbs the water and releases the HCl, but this is a complicated process with many reflux streams. It is also an expensive process, both in the investment in the HCl-resistant equipment and in the energy consumption. Another commonly found complication can be the presence of nonvolatile components in the starting HCl solution, which can accumulate in the circulating brine. In such case, the starting solution should be completely evaporated upstream and the heat loads should be redistributed. This problem was at the time an open field for creative process design, aiming at a better use of the energy and the expensive heat exchangers, and of any possible synergetic utilization of sources of low-temperature waste heat.11–14 Different aqueous solutions were used, including MgCl2 and LiCl. It was also proposed to replace the expensive heat exchangers by direct contact heating with organic “heat carriers.” (See below and also Chapter 5, Section 5.1.5.) Direct contact heating technology, with organic “heat carriers” (stable hydrocarbons, liquid, or vapors) — Certain processes need large heat exchangers made from expensive materials (resistant to corrosion, such as graphite, glass-lined, tantalum) to introduce heat into the process streams and evaporate certain components, and similarly for removing heat in condensers. In other cases, heating of such solutions in a regular heat exchanger would precipitate solids and cause the rapid scaling of such heat exchangers. One can resort to introducing very hot organic liquid or vapor “heatcarrier” in direct contact with the process stream to be heated. After heat transfer and equilibration, the organic liquid is separated, removed, washed, and reheated in a separate boiler made of cheaper materials. Although the heat carrier material would have a boiling point much higher than any of the components present, it can have a definite vapor pressure in the hotter parts of the equipment, which should be taken into account and included in the experimental program. Copyright © 2002 by CRC Press LLC

6.2.2

Gas–liquid equilibrium system

Gas–liquid (reversible) equilibrium systems are relevant in unit operations dealing with cleaning or cooling of a gaseous stream in contact with a liquid phase. In such operations, the concentrations of certain components that exist in both the liquid and the gas phase are related by a definite reversible equilibrium function, for a definite set of parameters. For example, water and ammonia can be both in a gas stream and in an aqueous solution (or water and HCl, or water and methanol, and the like). The ammonia can be recovered from the gas stream into an aqueous solution in a packed column where the gas stream (say with a few percents of ammonia) will be introduced from the bottom and will flow upwards (exiting with, say, 0.02% ammonia), while water is introduced at the top and flows downwards, countercurrently. The liquid also may contain nonvolatile solute components, and the greater part of the gas stream would consist of “inert” noncondensable gases. In principle, such a system can be studied as a vapor–liquid equilibrium system, and a certain number of theoretical equilibrium stages can be defined to obtain a certain result. But there is a quantitative difference compared to a regular vapor–liquid equilibrium system. The kinetics of reaching such an equilibrium are much slower, as they depend on the diffusion of the relatively few condensable molecules in the bulk of the gas phase until they reach the liquid interface, and possibly also on the “resistance” to mass transfer of the layer adjacent to such interface. So, the contact conditions are often more important than the theoretical equilibrium and the height of a theoretical equilibrium stage must be determined experimentally for each of the exact sets of operating conditions and for the exact geometry of the packing in the column. It becomes a completely empirical design and the position of the equilibrium curve has, in fact, little practical importance.

6.2.3

Liquid–liquid equilibrium system

Similarly, a liquid–liquid reversible equilibrium system can govern a solvent extraction process, which is intended to separate, concentrate, or purify a particular component from a mixed solution, such as a fermentation broth, a leaching solution from a mineral acid reaction, or a waste stream from other operations. One can refer to the basic reference books15–18 and note that the “official” denomination is “liquid–liquid extraction,” but most people in the field keep calling them “solvent extraction processes.” Generally, one is considering two liquid phases, but there also exists invariant systems with three liquid phases at equilibrium, according to Gibbs’ “phases law.” At least one of these systems was used in the IMI “cleaning” process for separation of clean phosphoric acid from wet phosphoric acid (see Chapter 4, Section 4.4.3, Reference 26).

Copyright © 2002 by CRC Press LLC

Again, in almost all processes of practical importance, there are many components in each phase. One has to define all these components and their ranges of concentration and to choose, on one hand, the variable of major interest and on the other hand, the ranges of parameters (such as the concentrations or ratios of the other constituents and the physical conditions) that can be covered in a reasonable experimental program. If one is not careful in his choice, the number of tests required can easily shoot up exponentially. The “distribution coefficient” of this major variable can be correlated and used for multiple stage process calculations in the defined ranges of parameters. The major difference from the vapor–liquid physical equilibrium systems is that in most liquid–liquid extraction processes, the major variable of interest in any particular process can be either an ionic or a molecular entity, according to the chemical extraction mechanism. Once this procedure is well understood, the bench-scale experimental program for the development of a separation process based on solvent extraction can be relatively straightforward. The technique of the so-called “separation funnels” tests is based on equilibration in defined conditions, sampling and analyses, and it can be carried out routinely by laboratory technicians. This fact of life was probably one of the main reasons for the successful development of many dozens of new solvent extraction processes in the years 1960 through 1980 in various countries. Very promising R&D programs in this field are continuing nowadays inside some of the large industrial corporations, although not much is published about that at international conferences. In this connection, it is important to stress the experimental technique called “limiting conditions,” which makes it easier to study the effect of one variable at a time. If, for example, 50 ml of a starting aqueous solution are mixed with 50 ml of a solvent phase, the concentrations of the component of interest after equilibration will probably change significantly in both phases. If, for example, a series of tests are done at different temperatures, the quantitative results can be “all over the place” and a lot of tests will be required to find a working hypothesis to explain the results. But if 100 ml of the starting aqueous solution are mixed with 1 ml of solvent, the chosen composition of the aqueous phase will change very slightly, while that of the small organic phase can change very significantly. Thus, three to four tests at different temperatures should give a clear indication of the effect of that variable for the particular chosen aqueous composition. The experimental procedure is also simpler for processes in which the solvent added is composed of a single component, such as butanol, pentanol, methyl iso butyl ketone (MIBK), and so on. But, for other processes, the composition of the solvent phase added can be quite complex by itself and may present a large number of additional components and parameters, such as the nature of the extractant (i.e., one particular tertiary amine from the dozens of tertiary amines commercially available), of the modifiers (i.e., one of many long-chained alcohols available for such duty), and of the diluent (i.e., a light, saturated hydrocarbon), and the relative weight ratios of these three classes of components. Copyright © 2002 by CRC Press LLC

In the typical practice of solvent extraction process development, one would generally start with a screening procedure. (Even my granddaughter knows by now that, in real life, the Princess would have to kiss many Frogs before she would find, maybe, her Prince.) This “screening” procedure is generally started with one effective composition “formula” found in previous publications, or in the suppliers’ recommendations. This formula may not necessarily be the optimal composition for the specific case studied, so that some “exploration tests” with moderate changes outside this range should be done preferably before any specific commitment. But the prevailing attitude has often been: “Let’s start with the composition that works, and we will optimize later.” But as often happens, everybody is too busy “later” to look back at this exact composition. This is a well-known pitfall. It has also been observed that the chemical behavior of some extractants (in particular tertiary amines) does change as the “new” reagent (straight from the bottle) is “aged” after a few dozens cycles of loading/regeneration. This change, which may include a significant shift in the equilibrium curve, can be due most probably to the elimination of some traces of impurities which remained in the “new” reagent from its synthesis, and possibly also to the oxidation of unsaturated bonds in the experimental manipulations. Since the plant will be working eventually with an aged extractant, the testing conditions and results should reflect that change from the beginning. An additional form of aging occurs in functioning plants due to accumulation of certain impurities in the solvent cycle. Although a continuous purification procedure is generally used on a side stream, there is an economic limit to such purification and any plant has to live with a certain level of impurities in the solvent. This effect is difficult to reproduce from these early tests, but has to be accounted for in the design safety factors.

6.2.4

Solid–liquid equilibrium system

Solid–liquid reversible equilibrium data are regulating many processes. All the metallurgical transformations relate, in fact, to this field, but the high temperatures of more than 1000ºC are considered as a “far away situation” by most chemists and chemical engineers. Some chemical industries are touching these high temperatures, but most remain below 200ºC and, in this range, these systems relate to solid dissolution, precipitation, and crystallization, which are widely used in the various inorganic industries, mineral treatments, and also in the natural sugar and sweetener industries. In all these processes, the saturation concentration of one variable component depends also on the concentration of the other components present, in addition to the physical parameters of temperature, absolute pressure, and non-condensable gas. The experimental determinations of such saturation concentration can be rather simple, in principle, if only one component is precipitating or dissolving, while the other components are remaining as parameters each in its respective phase. (Note that the “limiting conditions” technique mentioned above should also be used in such experimental determinations).

Copyright © 2002 by CRC Press LLC

However, in many other cases, two components may change phase simultaneously in certain ranges and the experimental program becomes more complex. Some solids are also precipitating in the hydrated form, taking with them some of the water into the solid phase and contributing to further changes in the concentrations in the solution. Furthermore, there are important cases of double-salts precipitation at certain concentrations. For example, the conditions for the precipitation and for the decomposition of Carnallite crystals (hydrated double-salt of KCl and MgCl2) are the key features of the potash industry from the Dead Sea brines as mentioned in Chapter 4. Kainite and Langbeinite crystals are important hydrated double-salts found in mined potash deposits containing magnesium sulfate and, therefore, they have been investigated in great detail to optimize potash recovery processes. Various degrees of supersaturation are always of major concern (see below). It is important to note that, in such cases, many of these solid precipitates can be clearly identified and quantified by established mineralogical techniques, in addition to the usual chemical analysis methods. Thus, the collaboration of a mineralogical laboratory equipped with all the usual microscopes and x-ray diffraction equipment can be a great help in such R&D programs. There was a time when radioisotope tracers were also used in such research in many research labs, but this technique can be risky if it is not done with all the special equipment, and it apparently went out of fashion.

6.2.5

Reversible and nonreversible equilibrium

In all the processes mentioned in this chapter, whenever a reversible equilibrium is expected, the determination of the quantitative relations of the concentrations at equilibrium (or at saturation), in various conditions and in the chosen range, is sufficient at least for the preliminary process design and demonstration stage. However, in many other cases, the reaction involves first a nonreversible change (such as one or more strong, one-sided reactions, i.e., neutralization) and then the equilibrium is established between the resulting phases. This order of reactions should be taken into account, and it may complicate further the experimental design. Therefore, one can appreciate the importance of the process working definition discussed in Chapter 5 in order to keep the experimental program within affordable limits.

6.2.6

Chemical equilibrium laboratory tests

As discussed in Chapter 5 in relation to the feasibility tests, most of the chemical equilibrium data can be determined in standard laboratory conditions, in rather small batches (in the hundred grams range) unless a larger quantity of one of the resulting phases is required for further tests or evaluation. Each test consists of bringing into contact, in the specified conditions, proportional quantities of the different “inlet streams” at the assumed Copyright © 2002 by CRC Press LLC

compositions. After the different reactions and/or mass transfer have occurred, the equilibrium is established and the phases are separated (if needed); the different phases are sampled and the compositions and the physical properties of the different phases analyzed. But it should be realized that with “real systems,” the total data collected from each such test, representing one “equilibrium point,” amounts to possibly 10 to 20 numbers. The recording and presentation of all the different numerical data sets in the experimental report can be only in the form of systematic tables. However, the correlation of this data for all the points collected is not straightforward, as it depends mainly on the way in which the data will be used in further calculations (sometimes many years later). This important fact of life should be recognized. Unfortunately, in many cases encountered in the past, certain parts of the data were lost on the way, as their future use was not clear to those editing the experimental reports.

6.2.7

Experimental difficulties in chemical equilibrium tests

Possible experimental difficulties can derive from the following causes. Establishing Equilibrium — Absolute equilibrium is, by definition, never reached, as its approach is asymptotic. Chemical engineers work with practical equilibrium, which can differ slightly from the absolute value by a few tenths of one percent, relative. In almost all cases of industrial interest, a practical equilibrium should be reached in a matter of minutes. Therefore, a contact time of 10 to 15 min in a test, before sampling, should assure a practical equilibrium. In some particular cases, wherever a doubt exists, the tests can be repeated with different contact times, say 3, 8, and 15 min, and the results compared and interpolated to determine the contact time required in this particular contacting mechanism. A slower mass transfer rate can also result from the adsorption, precipitation, or collection of impurities on the interfacial area, or in the adjacent “double layer”; this situation should be recorded and, if possible, corrected. At a later stage, when the actual plant design will be considered, the exact contact time required to obtain the desired result will need to be determined and optimized in relation to the contacting conditions in the chosen equipment (see Chapter 10, Section 10.5). For some items of equipment that are handling very large flows, this contact time can be an important factor, as every second would be cost significant. But such optimization can only be done in direct relation to the type of equipment chosen and to the designed conditions for the contact and for the subsequent phases’ separation. Supersaturation — This is often creating additional complications in equilibrium operations involving solids, by biasing the solubility’s levels. Although the physical causes of “natural” supersaturation are not really known, there are effective empirical ways to “break” the supersaturation and reach practical equilibrium (such as “seeding,” for example). Before

Copyright © 2002 by CRC Press LLC

undertaking an extensive testing program, these techniques need to be tried and confirmed in every case to obtain absolute data. On the other hand, one should remember that a controlled level of supersaturation is an essential factor in the design and operation of continuous crystallizers and that it can also be put to use for other process separations. In one specific case, a high level of “natural” supersaturation was found and exploited for an interesting process separation. A double decomposition reaction yielded two solids products, one of them (A) precipitated immediately and while the other (B) was maintained in solution by the supersaturation for a period of time sufficient to separate the (A) solids by centrifugation. Sampling Problems — Every multiple phase contact/equilibration should be followed by a complete phase separation before the resulting phases are sampled. Imperfect phase separation (entrainment of small quantities of one phase into another) is a common cause for serious problems, first in the reliability of the data obtained from the experimental work and later in operation of the continuous industrial equipment. When small-scale tests at nonambient conditions are done in closed laboratory containers, it is not always possible to separate and sample the phases inside. If a mixture is taken and separated outside, i.e., in a centrifuge tube, the contact conditions may change (temperature, pressure) and the equilibrium can be shifted before the phases can be separated. This experimental problem is not always recognized and may result in erroneous results. In such cases, the size of the test rig may need to be increased. Analytical Issues — By the time the samples have reached the analytical laboratory, the temperature/pressure conditions have changed and a sample can separate into a nonhomogeneous mixture of phases. This possibility should be suspected and checked in the analytical laboratory before a small aliquot is taken out for analyses. If this happens, the situation should be reported, as this can have other implications in the plant. Of course, the whole sample should be rehomogenized (by heating or dilution in a solvent) before the aliquot is removed.

6.3 Dynamic flow conditions Continuous reactions or separations, which are dependent on dynamic flow conditions, are generally much more complicated to study or even to fully understand. For example: • A one-phase stream containing a mixture of components is flowing through a packed bed of solids with a catalytic action, causing reactions between components in the flowing stream. • A countercurrent contact between a gas stream and a liquid stream, which allows reactions to take place, and a small amount of a certain component is changing phase in either direction (see Section 6.2.2). This operation can be done in a packed bed column or in other Copyright © 2002 by CRC Press LLC

types of contact equipment and can be used either to “wash” an exit gas stream or to “strip” a liquid stream from an undesired volatile contaminant. • Some thermal energy is introduced into a reaction mixture in the form of the combustion gases from a direct flame, from a plasma, or by induction microwaves. • A multiple-phase mixture, resulting from a reaction in a very short mixing zone, is separated continuously while flowing through a separator, such as a gravity decanter, a cyclone, or a centrifuge.

6.3.1

Design data required

The flow conditions (velocities and paths) determine both the residence time and the contact conditions affecting the interfacial mass transfer, such as the turbulence or the shearing forces, the differential gravity or G-forces, or a temperature gradient. The design data required to link quantitatively these flow conditions with the final results obtained can be either: 1. Rates of reaction and of mass transfer, which determine the chemical composition of the different phases in the resulting stream(s). 2. Physical separation between the different phases in the exiting streams.

6.3.2

Simpler processes

Fortunately, many of these mechanisms of industrial interest were straightforward enough and have been extensively studied. For example, many of the catalytic reactions in a gas mixture flowing in a packed column, or the changes in a solution flowing through an ion-exchange resin column, can be simulated in relatively small continuous test equipment. The scale up of the performance of a gas cyclone or a liquid cyclone can be predicted from smallsize continuous tests (see Chapter 4, References 37 to 39). The separation results in a continuous centrifuge (of most types) can be evaluated from simple tests with a small, bench-scale machine.

6.3.3

Theoretical models

Theoretical models allow the simulation and data generation from “standardized” batch tests in some other widely used mechanisms that have been extensively researched. For instance, the mass transfer occurring in the continuous solid–liquid and liquil–liquid contacting inside mixed vessels can be reliably designed from the kinetic batch reaction curves obtained in benchscale tests in well-defined mixing conditions. Batch Aerobic Fermentation — A particular case of increasing industrial importance is the batch aerobic fermentation involving the mixing of a liquid solution with dispersed microorganism particles, chemical additives, and air Copyright © 2002 by CRC Press LLC

bubbles. In such a process, from the chemical engineering point of view, oxygen from the air bubbles is continuously dissolved and consumed by the microorganisms (the “bugs” in operators’ jargon), CO2 is generated and evaporated, carbohydrates are reacted and consumed, a soluble valuable (desired) component is produced, and a lot of other side reactions may be occurring, all simultaneously with the release of heat. A significant cooling capacity is critical. The flow rate of the air, the pH, and the temperature of the mixture are generally maintained and controlled by external means and the excess gases are vented. Mixing is very important in maintaining the aqueous solution more or less uniform; it can be internal or external and is generally combined with the cooling system (jacket or heat-exchangers). The batch period is a matter of days. Considering a large tonnage plant (say, 50,000 metric tons/year [MTY]) with a batch turnover of, say, 4 days, and a product concentration of the order of 10% in the fermentation broth, the net internal volume inside all the fermentors is considerable, about 6800 m3 or 34 fermentors of 200 m3 each. Therefore, any improvement in the average batch period or in the final broth concentration can have a serious economic effect. The hydrostatic pressure at the bottom of such a large fermentor (say, 5 m diameter and up to 10 m in height or even more) is an important operating parameter. It affects not only the supply pressure and, therefore, the cost of the compressed air, but also the solubility of the different gases in the solution and possibly also the biological functioning of the microorganisms. Different models of large fermentors are used in industry, each with its apparent advantages and disadvantages. (Only apparent since many of the important features relevant to their operation have not been released for publication by the corporations operating them.) In any case, the internal inspection and periodical cleaning is essential. Figure 6.1 illustrates the principle of a draft tube circulation with a cooling jacket, which is using the inlet air in the draft tube to promote the circulation of the media. The air/liquid contact inside the draft tube is short, but at a higher turbulence regime. Such type of fermentors have limited size and limited cooling capability and, therefore, they were used mostly for smaller production capacities, since their upscaling is estimated to reach a limit around 60 to 100 m3. Figure 6.2 shows the main features of a fermentor in which the compressed air is sparged from the bottom and an external pumping circuit takes the media around through the heat exchangers. These features have more design options and scale up possibilities, but the passage of the microorganisms through a (positive flow) pump and through the heat exchangers has been hotly debated. The composition of the solution in the batch fermentor is changing all the time and is monitored by the operator to detect any unexpected trend. As the final trend in composition is asymptotic, the main operating issue is how and when to stop the “reaction” (“dropping” the fermentor), since in many cases, the later period of operation produces little valuable component but a lot of impurities, which can complicate the recovery. Copyright © 2002 by CRC Press LLC

gases out gas separation

froth

CW in

draft tube

cooling jacket

CW out

air in

Figure 6.1 Fermentor with a cooling jacket and draft tube circulation.

gases out gases separation

CW in cooler

air in

Figure 6.2 Fermentor with sparging air and external cooling cycle.

Copyright © 2002 by CRC Press LLC

In most new implementations, a batch fermentor is a prudent starting choice, but it is generally expected that when the industrial process will be well in hand, a number of such fermentors (four to eight) would be connected and operated in series in a continuous fashion. This possibility should be a basic condition for the study and that option should be provided in the plant design. Note that an industrial setup should also include a smaller special “inoculum” fermentor and one or more nonaerated “drop-tanks” into which the content of a finished fermentor is transferred to stop the fermentation, in addition to means for pasteurization (“steaming”) of all incoming streams and all equipment and piping, an acceptable waste disposal treatment for the “bugs,” and, in many cases, the supply of chilled water. Once a particular process is defined and a model of fermentor is chosen, the study and design of quite large industrial units can be done from a straightforward quantitative model based on the data generated in a pilot fermentor of 10 to 100 L. Such a pilot is often made in the form of very high vertical glass pipes of 7 to 10 cm diameter, with induced circulation to duplicate the changing hydrostatic pressure effects. Separation of Solids — The rate of separation (or concentration) of solids from a slurry in a continuous solid–liquid thickener depends on the “filtering” velocity of a liquid flow through a dilute solid bed. It has been modeled by Kinch18 long ago and can still be calculated from a standardized slurry settling curve in a 1 L graduated glass cylinder. This method is used routinely for the study of the effects of flocculating agents or other pretreatments on the settling rate of the slurry and on the capacity of the thickener. (See Chapter 5, Figure 5.2.) Vacuum Filters — Large industrial continuous vacuum filters can be designed from standardized, bench-scale, batch-filtering tests. Rate of Continuous Separation — The specific rate for the continuous separation in industrial liquid–liquid settlers can be predicted from standardized batch tests following the Barnea-Mizrahi model.19–23

6.3.4

Special test rigs

Despite the above examples of the better-known technologies, there are still many cases in which the study of a new system can only be done in a small specially designed continuous test rig. In such installations, the contact parameters and the flow conditions (velocities and paths) should be changeable and controlled exactly and the resulting streams should be separated and sampled, then analyzed. The design of this special rig should be based on a theoretical model that will allow to separate, as far as possible, between the different assumed physical–chemical mechanisms, such as: • Mixing and dispersion of phases • Flow of the continuous phases near or around the surfaces of dispersed solids or liquid particles

Copyright © 2002 by CRC Press LLC

• Forces causing the physical segregation of dispersed phases, and/or the coalescence of liquid droplets, etc. The results of such tests should give a good holding on the scale-up and sizing of conventional industrial equipment, such as mixed tanks, settlers, or centrifuges. Flash Dryer — An important example for producing powders from solution would be a “flash dryer.” A more complex case is the drying of vegetable protein and similar organic concentrates (particularly wheat gluten) that has to be done in industry in a set of severe limiting conditions. The high starting moisture gives a messy sticking consistence to the feed in the dryer, the so-called “dry” powder must retain a relatively high minimum moisture in order to maintain its “activity” for later use, and a maximum operating temperature and a maximum residence time should be maintained in the heated zone, as any overheating would damage the product. Such performance can be done, for example, in a so-called “ring-dryer” in which a very large flow of air is circulated around at a controlled temperature (Figure 6.3). This air stream passes through a mechanical “disintegrator” into which the protein concentrate is injected together with a stream of hot combustion gases. The small wet particles formed in this very short shock classifier

wet gases outlet

fan thermal insulation bag filter wet feed material

hot gases supply powder product

disintegrator

Figure 6.3 Main elements of a “ring dryer.”

Copyright © 2002 by CRC Press LLC

treatment are coated with recycled drier dust and are entrained by the hot air flow into the ring for an average number of turns (residence time). A small side stream is removed and directed to a classifier that concentrates the dried particles collected from the air bleed stream in a bag filter. In view of all these essential preconditions, such processes can only be studied and demonstrated in a special pilot rig, which should be flexible enough to reproduce the different sets of operating conditions, while using real vegetable protein concentrates. Most suppliers of this type of equipment are equipped with this pilot rig and it is preferable to work with a selected expert supplier. Cleaning of Waste Combustion Gases — Another example of the need of special test rigs, which have a widely declared importance, but on which relatively little process information has been seriously published, is the cleaning of the waste combustion gases from power plants and large kilns before discharge to the atmosphere. These combustion gases are discharged from the boiler systems in very large volumes as hot and corrosive, at a very low positive pressure, and they contain, mainly, a variable concentration of acid sulfur oxides with some nitrous oxides and various fine ashes and other impurities. Any treatment should handle the very large volumes efficiently, and neutralize and eliminate the a.m. impurities without creating any back pressure that can affect the boiler systems. From a technical point of view, the problem can be solved, but at a cost! Various scrubbing systems with slurries of limestone, lime, and dolomite were proposed and offered commercially, but all had to face the direct correlation between the cleaning efficiency and the bottom line cost. One patented route25 to decrease this overall cost proposed to use ammonia as the neutralizing agent and to recover the ammonium salts and use them as fertilizers. This route required an efficient scrubbing system that could assure that: • • • •

Objectionable impurities would be completely eliminated. No significant ammonia would remain with the exit gas. A concentrated solution of ammonia salts is obtained. No back pressure is affecting the boiler system.

A new scrubber design26 had to be developed to answer these demands, which is illustrated in principle in Figure 6.4. This is a multichamber, lightweight, horizontal scrubber. In each chamber, there is a large flow of liquid sprayed across the gas flow, maintained at a different concentration. The net flow of the solution is countercurrent to that of the gases. Ammonia gas is introduced into the hot gas and water is fed from the other side. An auxiliary fan is used before the chimney. The exit gases can also be reheated, if needed.

6.3.5

Indirect methods

When studying a complicated dynamic mechanism, indirect methods may sometimes be used quite successfully, but the “convenience” of obtaining a lot of data by an indirect method may cover a basic difference in the mechanism.

Copyright © 2002 by CRC Press LLC

flue gases out

water makeup

flue gases in ammonia in first

second

intermediate compartments

fan last

option heater

fan premixer

Am Sulfite Solution out cooling tower

Figure 6.4 Flue gases scrubber.

When this author started a M.Sc. graduate research in Chemical Engineering a long time ago, he was directed by his tutor, David Hasson, to study the mechanism of the creation of liquid drops by pressure spraying, by using a convenient but indirect method. The conventional experimental method used at the time consisted of a complicated sampling procedure in order to collect drops on “targets,” followed by lengthy manual sorting under the microscope. (All of this was, of course, before the automatic computer sorting available today.) The proposed idea was to use hot molten wax as the liquid, so that the drops would solidify as spheres, and the powder could be sampled and separated into size fractions in a conventional laboratory-sieving machine. This seemed a good idea, and after the usual literature search and study, an experimental setup was prepared and tried, but a prosaic problem was immediately encountered: the sample of wax powder was warmed and softened by the friction on the sieve deck, which became rapidly clogged and useless. This graduate researcher almost despaired when, by chance, an “old hand” visitor passed through the faculty. Hearing of the problem, he said, “Yes, we once had something similar and we solved it by adding 'dry ice' to the screens.” This dry ice (solid CO2 powder) can be easily produced in the lab from an inverted pressure bottle of liquid CO2, and then sublimized while cooling its surroundings without leaving any traces. This was tried and it worked perfectly, and we started to get nice reproducible results. So, this author received a very useful lesson — do not keep your problem to yourself, go and consult experienced professionals. A lot of results were collected, which correlated nicely with the operating variables under study. However, such correlation was completely different

Copyright © 2002 by CRC Press LLC

from the partial data published by several previous researchers, indicating a much smaller drop size and a narrower distribution. So the gist of the thesis was shifted to the explanation of such basic difference, by getting into a more fundamental account of the different mechanisms that occur successfully when the liquid is sprayed out under pressure and flows away while forming filaments and drops. The molten wax method “froze” literary the dynamic process in its first stage, while with the normal liquids, the larger drops would be catching up with the smaller ones, collecting and joining them, and reaching a larger size distribution (which can be relevant to the spraying of paint or inside gas–liquid contactors/scrubbers). So, the “convenient” research technique gave very good results,26 but on a completely different situation that can be relevant to certain other applications, such as the direct spray into a reaction/combustion zone. Thus, this author got his second lesson in R&D. Before investing a lot of work, time, and energy, one should be reasonably sure that the results will remain in the range of interest for the further application considered. Such a typical error is still seen today all around.

6.4 Scale-dependent operations These are operations in which the reaction rate, the mass transfer rate, or the phase separation rate, obtained per unit volume of equipment, will depend on the actual size of the equipment.

6.4.1

Vertical driving force depending on the hydrostatic height

The effect of the height of a chimney on the resulting draft is well known. The complex effect of the hydrostatic pressure on the operation of industrial aerobic fermentors was already mentioned above. In a continuous industrial liquid–liquid settler (Figure 6.5) operating under gravity forces, there are two layers of separated liquids and a layer of “mixed phase” between them consisting of a dispersion of drops from one liquid to another. The specific rate for the continuous separation, in m3 per m2 of horizontal surface, increases with the thickness of the “mixed phase” layer (since a higher combined hydrostatic force increases the pressure on the drops and accelerates the drop-to-drop coalescence). It is therefore advantageous to operate with a settler of maximum height which can accommodate a thicker “mixed phase” layer. However, there is a diminishing return since the quantitative function of the separation rate in m3/(m2 h) is proportional to the mixed phase thickness at the power (0.40 to 0.45). This mechanism and optimization of such settlers was studied extensively, including the procedure for scaling up from relatively small batch tests.19–23 It was then observed that the separation efficiency of the “mixed phase layer” per unit volume increased as its thickness decreased. One would have thought that a “flat” settler would be the most efficient, but this was, of course, impossible to realize without counting on the minimum volume

Copyright © 2002 by CRC Press LLC

vent nitrogen blanket option

light liquid phase

passive interface

mixed phase layer heavy liquid drops sight glass feed in apparent interface active interface heavy liquid phase

Figure 6.5 Liquid–liquid continuous settler.

taken by the separated layers. This observation led to the invention and development of improved “compact” settlers, in which sets of inclined partitions, made from thin PVC plates, were installed with 7 to 10 cm distances between them (Figure 6.6). The volume between two partitions constituted a “flat” settler fed from one direction while the separated phases were collected on the inclined partitions and drained into vertical “chimneys” left between the stacks. This addition of stacks of inclined partitions increased the overall volumetric separation efficiency of large industrial settlers by a factor of about 3. This reduction in the solvent inventory in the plant was obviously very important when working on large scale with expensive solvents.

6.4.2

Wall effect

This dependence of the results on the absolute size of certain types of equipment is explained in most cases by a “wall effect” and is related to the nonhomogeneity caused by the inside flows near the walls or to significant larger heat losses through the walls and the like. A larger wall effect can be quantified by a relatively larger ratio of the internal wall surface to the volume of the equipment. This effect can be more serious whenever the operation depends on a metastable dynamic balancing (“walking a tight

Copyright © 2002 by CRC Press LLC

mixed phase passive interface

active interface

vent

nitrogen blanket option

light liquid phase

feed in

heavy liquid phase

Figure 6.6 Liquid–liquid settler with sets of racks of inclined partitions.

rope”), in which a relatively short change can cause a collapse, for instance in a fluid bed solid gas contactor (see below).

6.4.3

Crystallizer

For example, it is well known that the average size of the crystals obtained from a continuous industrial crystallizer increases generally with the size of the equipment, up to a certain maximum related to the flow mechanism and residence time. Furthermore, this average crystal size distribution in the product can be very important as it determines critically the design and the daily operation of all the downstream operations involving the crystals, such as their filtration or centrifuging, washing, drying, screening, marketing, etc. The driving force for the crystallization is always a certain degree of supersaturation which is created purposely by a chemical reaction (addition or decomposition), or by a change of temperature (mostly by cooling), or by a change in concentration caused by the evaporation of water or of another solvent. Such supersaturation strives to decrease by precipitation on any existing crystal surface. Copyright © 2002 by CRC Press LLC

An additional important factor in the design and the daily operation of an industrial crystallizer is generally a size classification system between the larger crystals, which are ready to be removed as product, and the smaller crystals that should be left inside for some additional growing time. This size classification can be either on an internal or an external flow cycle, and it has to be done in unfavorable conditions, such as a concentrated, heavy and viscous “mother liquor.” In some decomposition crystallyzers, such as in the potash industry, there is a further complication as the feed is in the form of larger, lighter crystals, which have to be kept from mixing with the product until they are decomposed. So, there are many types of crystallyzers in use, but their basic principles are similar and relatively simple, and for each practical case the logical choice can be reduced a priori to two or three possibilities to be studied further in detail. Some typical illustrations of the principle of a “draft tube” crystallizer are shown in Figures 6.7a and 6.7b, and of a dense slurry “Oslo-type” crystallizer for larger crystals in Figures 6.8a and 6.8b. Coming back to the “wall effect” on the product’s size, the prevailing explanation from experts in this field is that the circulation flow of the slurry inside the crystallizer is slower near the walls. As a result, a relatively larger number of crystal nuclei does precipitate from the part of mother liquor that is passing in these regions than in the part remaining in the main cycle flow. There are also more “fines” generated in a smaller equipment by attrition with the higher RPMs of the impeller. This larger number of nuclei translates into a smaller average size of the crystal product for a fixed production tonnage. But this is only part of the story. On the other hand, in most well-designed crystallizers, the amount of “fines destruction” is an effective (although somewhat expensive) operating tool for increasing the average crystal size by reducing the number of new crystals that are allowed to develop. Fines destruction is obtained on purpose and on demand by the dissolution of a part of the nuclei by using either local dilution or heating of the circulating “clear” mother liquor. In most processes, the amount of fines destruction can be enhanced in the test unit to balance the negative wall effect and to get larger and nicer crystals. The prediction of the final crystal size distribution in the final industrial unit from such small-scale tests does require a lot of experience (and perhaps some guessing) to balance between the contributions of the wall effect and the fines destruction. Large-scale piloting with “real” streams, of course, would be much preferable, but in most projects, such piloting is not practical until an advanced stage (if at all). The final result of this dilemma is that any crystallizer installed in a first plant with a new process is generally oversized to allow more flexibility in the level of fines destruction and to be safer as regards the final crystal size distribution. As a matter of fact, many industrial crystallizers designed by reputable suppliers for new processes were finally operated at higher capacities, up to twice their nominal specification, after the plant’s experience was Copyright © 2002 by CRC Press LLC

vent

external heater or cooler circuit

(a)

product slurry feed

vent

waste solution

(b)

water

product crystals slurry

Figure 6.7 Draft tube crystallizer.

Copyright © 2002 by CRC Press LLC

feed crystals slurry

to vent or condenser or vacuum unit external heater or cooler circuit

product slurry

bleed

feed

Figure 6.8a Crystallizer with external loop.

optimized and stabilized. (As the manager of one supplier said once, “You’ve got a good deal ... why should you complain?”)

6.4.4

High-temperature equipment

The study of processes based on high-temperature reactions and transformations is generally done quite effectively batchwise, in small crucibles in a laboratory furnace, indicating the effects of the reactants and of the temperature vs. time curve in a controlled environment. This study is much easier when it involves a gas reacting with solids or liquid surfaces, since the contact is generally good and the diffusion inside and outside of the solid or liquid is a matter of time. It is also reasonably effective for a solid–liquid reaction if the solids have been finely ground so that the specific contact surface is large enough. The main process problem is encountered when the reactants are all solids, despite the pretreatment of fine grinding and mechanical compaction. In this case, the addition of a fluxing compound is required that melts and supplies a film of liquid phase between the solids in which the reaction can progress. But such fluxes can generally create other complications downstream. Copyright © 2002 by CRC Press LLC

to vent or condenser or vacuum unit external heater circuit

product slurry

bleed feed

Figure 6.8b Crystallizer with external loop.

As a typical example, the high temperature reaction of zircon (zirconium silicate) with calcium oxide to liberate zirconium oxide was mentioned in Chapters 1 and 5. Several proposals were published for solid–solid reactions with small additions of different fluxes, but apparently none of these routes can guarantee the complete elimination of the silica from the zirconia. The Gorin-Mizrahi patented route mentioned in Chapter 5 involved the reaction of molten CaCl2 in intimate mixture with fine zircon powder and water vapors (see Figure 6.927). Previous proposals have been to react, at higher temperatures, CaO and zircon in a molecular ratio of more than 3:1 to obtain a mixture of CaO,ZrO2 and 2CaO,SiO2, which will have to be treated by wet acid to separate the ZrO2. Some published trials to operate directly in a 1:1 ratio to obtain directly ZrO2 were not successful in meeting the desired purity as regards the very low requirement of residual SiO2. The Gorin-Mizrahi route involved the use of a reacting film of liquid molten CaCl2 which decomposed at a relative low temperature (less than 1000ºC), to liberate all the HCl gas and active CaO, and gave an intermediate complex, but well defined, solid double salt of CaO,ZrO2 and CaO,SiO2, on a remaining

Copyright © 2002 by CRC Press LLC

ZrO2 2715 C.

ZrO2,SiO2 -1675 C.

CaO,ZrO2 - 2340 C. Ca Zirconate zone

ZrO2 zone

CaO zone

SiO2 1723 C.various calcium silicates zone

CaO, 2 CaO, SiO2 1540 C. SiO2

CaO 2570 C.

Figure 6.9 Triangular diagram ZrO2-SiO2-CaO.

core of zircon. In a second kiln at a somewhat higher temperature (about 1400ºC), this intermediate complex was decomposed with the remaining zircon to give solid zirconia and a CaO,SiO2 melt. After quenching and leaching in a dilute HCl solution, only pure ZrO2 remained in the solid form. The industrial large-scale processes are done in continuous rotating drum kilns, calciners, and dryers in which the amount of heat losses through the walls is generally quite significant. Large velocity gradients exist along the radius affecting the residence time. Due to these two causes, large temperature gradients are found both along the axis and the radial dimension. Thus, the scale up and design of a piece of high-temperature equipment from the results obtained in a small continuous test unit gets more sensitive as the overall residence time is shorter. On the other hand, over sizing of this type of equipment is generally not a practical option, considering the scale of production and the control of the unit cost.

6.4.5

Failure to recognize the wall effect

In certain cases, the wall effect can become an essential component of a reactor’s operation. Failure to recognize that fact in the scale up can be very serious. When this author was very young, he witnessed such an “error” in a project managed by one of the world’s largest chemical companies. A process was developed in which a decomposition reaction was done at a high temperature in a fluid bed reactor, which was maintained in a fluid condition by a stream of combustion gases introduced from below while the feed solution was sprayed on the bed from above. The solid product from the decomposition was collected on the fluid bed particles, which kept growing Copyright © 2002 by CRC Press LLC

on the upper layers of the bed, and were taken out with a bleed stream of the coarser particles accumulated in the bottom part of the bed. The chemistry was simple, but the local conditions inside the reactor were quite heterogeneous and required a systematic vertical circulation to take the grown coarser particles downwards. This circulation was obtained in a pilot reactor of 2-ft diameter, most probably by the wall effect, since the upwards gas velocity near the wall is always much greater causing a vertical displacement in both directions. This pilot fluid bed could be maintained quite stable and the test results obtained were reasonable. However this essential effect was ignored when the reactor was scaled up and built into a 50-ft diameter tower. This catastrophic error led to a complete failure to perform, as the fluid bed was basically unstable with particles growing and growing in the upper layers while smaller particles were removed from the lowest layer. If this issue had been recognized in time, a modified design with induced circulation could have been successful, but a new plant was left to rust and a very large investment went down the drain. Needless to say, the reasons for this failure have never been publicly explained to the profession, but following such a shock, all the R&D projects in the chemical industry in the area were shelved for quite a few years.

6.5 Reporting results from the experimental program 6.5.1

Frequent partial reports

It has been noted (in Section 6.1.4) that one of the main bottlenecks of any development project is always the calculation of the practical implications of the results obtained from the experimental work and their presentation by the process engineering group. (The professional joke refers to the period in which a process engineer is requested to work 24 hours a day and then continue the work through the night.) If these results can be made available in a series of successive self-contained reports, each dealing with one section of the process block diagram, the process engineering group can start to correlate and work out these results as they come, making better use of their limited resources. It is, therefore, important to agree on a transmittal procedure, which can include eventually the transmittal of draft reports (with due reservations) if certain details are still not available. It is also important to identify clearly these reports as any other project documents with the code number, revision number, date of issue, and name of the responsible person for further contacts.

6.5.2

Complete reports on the experiment part

In many projects, it has been seen that such experimental reports were handled and written as internal memos of current value only. The authors of these documents seemed to assume that the limited number of readers

Copyright © 2002 by CRC Press LLC

should remember all the details of “last week’s discussions” and, thus, there was no need for further detailing. Such practice often caused serious misinterpretations. But more importantly, these experimental results have been often retrieved a few years later for further studies in order to improve or expand the plant’s operation. In many cases, unfortunately, they could not be used for lack of critical factual information. It is therefore very important that all these experimental reports are written as self-contained complete scientific reports, which can be used also by a “new guy” who has just arrived on the project. They should include full details on the purposes, the procedure, the materials, the sampling and analytical methods, the numerical results, the calculations procedure, any reference documents, the names of the responsible personnel and all the participants, and any observation or reservation or recommendations as regards the value of the results. The few extra hours required for a complete report would be well invested and would be recovered, in any case, when the “process package” is prepared (see Chapter 7), although possibly by a different team.

6.5.3

Implications of the results

Finally, it is important as well to prepare and present a comparison of the numerical values from the experimental results actually obtained in the tests with the assumptions or extrapolations used by the process engineering group in the preliminary process working definition (see Chapter 5, Section 5.1). Reasonable differences can be expected and the overall effect can be evaluated readily in a recalculation of the balances with the already available spreadsheets. But if these differences or their implications are larger and more significant, a review meeting should be called to decide on any change in the program.

6.6 Worth another thought • There is not much point in designing and starting any significant experimental program without performing first the process engineering analysis and being reasonably sure that the results would remain in the range of interest for the further application considered. • The main purpose of the experimental program is the collection, correlation, and presentation of the design data that is specifically needed for the new process design and optimization in the limited range of variables of practical interest. Another important purpose is to observe possible, but unexpected, problems that can occur and that should be dealt with. • If representative samples of the actual raw materials cannot be readily procured, an experimental program on “synthetic” mixtures can only be done as an exploratory work for the preliminary process design. Copyright © 2002 by CRC Press LLC

• Since, in most cases, there are more than two components present in each phase, one has to decide from the start which two components are the variables under study while the other components are to be considered as parameters for the purpose of the present process design. • Very hot organic liquid or vapor “heat carrier” can be introduced in direct contact with the corrosive process stream. After heat transfer and equilibration, the organic liquid is separated, removed, washed, and reheated in a separate boiler made of cheaper materials. • The experimental technique called “limiting conditions” make it easier to study, specifically, the effect of one variable at a time. • Many of the solid precipitates can be clearly identified and quantified by established mineralogical techniques in addition to the usual chemical analysis methods. The collaboration of a mineralogical laboratory can be a great help in a R&D program. • With “real systems” testing, the total numerical data collected representing one “equilibrium point” amounts to possibly 10 to 20 numbers, and its recording and presentation can be only in the form of systematic tables. • A 50,000 MTY fermentation plant with a batch turnover of 4 days and a product concentration of the order of 10% in the fermentation broth, needs 34 fermentors of 200 m3 each (5 m diameter and up to 10 m high). The hydrostatic pressure at the bottom is an important operating parameter.

References 1. Schweitzer, P.A, Ed., Handbook of Separation Techniques For Chemical Engineers, McGraw-Hill, New York, 1979. 2. Henley, E.J. and Seader, J.D., Equilibrium-Stage Separation Operations in Chemical Engineering, John Wiley & Sons, New York, 1981. 3. Davis, G.A., Separation Processes in Hydrometallurgy, Society of Chemical Industry, Ellis Horwood, London, 1987. 4. Rousseau, R.W., Handbook of Separation Process Technology, John Wiley & Sons, New York, 1987. 5. Wankat, P.C., Equilibrium Staged Separations, Prentice Hall, New York, 1988. 6. McCabe, W.L., Smith, J.C., and Harriot, P., Unit Operations in Chemical Engineering, 5th ed., McGraw-Hill, New York, 1993. 7. Humphrey, J.L. and Kelier, G.E., Separation Process Technology, McGraw-Hill, New York, 1997. 8. Seader, J.D and Henley, E.J., Separation Process Principles, John Wiley & Sons, New York, 1998. 9. Khoury, F.M., Predicting the Performance of Multistage Separation Processes, 2nd ed., CRC Press, Boca Raton, FL, 1999. 10. Moriyama, T. and Sakaki, M., Vapor liquid equilibrium of hydrochloric acidcalcium chloride-water systems (in Japanese), kogyo kagaki zasshi, 64, 18771878, 1962, (see also French Patent 979,790, 1965).

Copyright © 2002 by CRC Press LLC

11. Tyshotskaya, O.V. and Grinstein, I.M., The system HCl-H2O-CaCl2, sbornik trudy vese. nauchn, issled inst gidro sulfitn prom. (Russian), 13, 184–202, 1965. 12. Mizrahi, J., Barnea, E., and Gottesman, E., Production of concentrated HCl from aqueous solutions thereof, Israel Patent, 36,304, 1972. 13. Lotzch, P. and Scherz, G., System HCl-H2O-MgCl2, Chem Technol. (German), 339–340, 1973. 14. Kolek, J.F., Hydrochloric acid recovery process, Chem. Eng. Prog., 69, 47-50, 1973. 15. Lo, T.C., Baird, M.H.I., and Hanson, C., (Eds.), Handbook of Solvent Extraction, John Wiley & Sons, New York, 1983. 16. Ritcey, G.M. and Ashbrook A.W., Solvent Extraction, Principles and Application to Process Metallurgy, Vol. 2, Elsevier, Amsterdam, 1984. 17. Rydberg, J., Musikas, C., and Choppin, G.R., Principles and Practice of Solvent Extraction, Marcel Dekker, New York, 1992. 18. Godfrey, J.C. and Slater, M.J. (Eds.), Liquid-Liquid Extraction Equipment, John Wiley & Sons, New York, 1994. 19. Barnea, E. and Mizrahi, J., Compact settler gives efficient separation of liquidliquid dispersions, Proc. Eng., 60–63, 1973. 20. Barnea, E. and Mizrahi, J., Separation mechanism of liquid-liquid dispersions in a deep-layer gravity settler (4-part series), Part 1: The structure of the dispersion band; Part 2: Flow patterns of the dispersed and continuous phases within the dispersion band; Part 3: Hindered settling and drop-to-drop coalescence in the dispersion band; Part 4: Continuous settler characteristic, Trans. Inst. Chem. Eng., 53, 61-69, 70-74, 75-80, 83-93, 1975. 21. Barnea, E. and Mizrahi, J., The effects of a packed-bed diffuser precoalescer on the capacity of simple gravity settlers and on compact settlers, paper Proc. Int. Solvent Extraction Conference, Toronto, 374–384, 1977. 22. Barnea, E. and Mizrahi, J., A generalized approach to the fluid dynamics of particulate systems, Part 1: General correlation for fluidisation and sedimentation in solid multi-particle systems, J. Chem. Eng., 5, 171-189, 1973. 23. Barnea, E. and Mizrahi, J., A generalized approach to the fluid dynamics of particulate systems, Part 2: Sedimentation and fluidisation of clouds of spherical liquid drops, Can. J. Chem. Eng., 53, 461-468, 1975. 24. Clue, A.S., POB 1723, 5816 Bergen, Norway, e-mail [email protected]. 25. Mizrahi, J., A scrubber for the treatment of flue gases, Intern. PCT Patent WO 99/20371, Washington, D.C., Appl. 22.10.97, assigned to Clue, A.S. 26. Hasson, D. and Mizrahi, J., The drop size of fan-spray nozzles, Trans. Inst. Chem. Eng, 39, 415-422, 1961. 27. Qureshi, M.H. and Brett, N., Phase equilibria in terniary systems containing zirconia and silica, Proc. British Ceramic Soc., 67, 1968.

Copyright © 2002 by CRC Press LLC

chapter 7

Preliminary process design for a particular proposal 7.1

Process team

From this stage on, the process engineering team consists typically of: • The process engineers from the promoting group, who have been involved in the two previous reviews described in the chapters above and who will still lead the effort • A delegate from the newly appointed corporate project manager, who reports directly to him and will gradually take over the lead of the process engineering team • A number of process engineers from a chosen engineering company, who are introduced into the project at this point • Experienced consultants, or “freelance” specialists, as needed The different creative and critical tasks described below have often been referred to as the “basic engineering” of the new process. They will be based on the preliminary process definition (Chapter 5) and on the reported results of the experimental program (Chapter 6). All these tasks should be done in parallel, since there are strong interactions among them. Their respective results will add up to the preliminary process design package. All the basic engineering documents will be reviewed, discussed, and agreed with the entire project team as soon as they are produced, before being issued and distributed outside this team. This procedure is important, because there will certainly be some differences of opinion among experienced professionals, differences that are normal in any working group. However, outsiders could misunderstand these differences of opinion, as they are only partly informed, and could attribute them to a possible lack of confidence in the whole project. Also, in many cases, the results from the first round of calculations and evaluation could indicate the need for changing

Copyright © 2002 by CRC Press LLC

some of the initial choices. Another round should be done to produce revised documents, with improved results. The final results of the preliminary process design described in this chapter, together with the economic evaluation discussed in the next chapter, will be reviewed with the relevant corporate management. Hopefully, they will justify a “maybe” decision from this management, authorizing the continuation of a working program towards a first implementation, possibly with some preconditions. In some cases, a negative decision could result from the review, putting an end to the proposal.

7.2 Process flow-sheets The preparation and drawing of the first set of process flow-sheets (produced at this stage as revision 0) represent the translation of the process block diagram and the chemical mechanisms concepts used by the R&D group into the usual chemical engineering methodology and practice of the chemical plant. This bridge between these two disciplines is an important creative task, requiring a fundamental scientific understanding of the process, as well as extensive experience in the design of chemical plants and their operating practice. These process flow-sheets will represent a reference basis for every professional related to the project, and they should be clear also to engineers and technicians who do not have an extensive scientific background (or may have forgotten it since college). In these process flow-sheets, the different functional sections are presented in separate drawing sheets, each with its defined conditions. These drawings also include the systematic tag-numbering of all pieces of equipment and all the main streams. These tag-numbers will then be used for reference in all future project documents. Some corporations have their own practice, but practically any one of the many numbering systems can be used, as long as the system chosen is clear and consistent. These process flow-sheets include only such details on the piping, valves, and instruments that are essential for an understanding of the description of the process operation; there is no need to include at this stage all the valves, bypasses, drains, sampling points, instruments, and automatic loops that will be needed eventually for the convenience of the plant operations. Most of these items are not essential at this stage, and they may be introduced later into the P&ID (piping and instrumentation) drawings, at the detailed engineering stage. With progress of the development project, it can often be decided (or at least considered for alternative study) that certain pieces of equipment, streams, or control instruments should be added or removed, or that the routing of certain streams should be changed. Therefore, each of these process flow-sheet drawings will probably be revised several times at later stages. Three or four revisions are quite common in most projects. The following five examples of small portions of (real) process flowsheets illustrate different typical possibilities: Copyright © 2002 by CRC Press LLC

Interchanger H-03

Trim heater H-02 steam

Trim cooler H-04 cooling water

10

T-01

T-02 T-03

T-04 11

15 16

steam TK02

TK04 to DC-02

H-01 P-02 1

P-04

3

To DC-01 2

TK-01

P-01

TK-03

P-03

TK-05 P-05

Figure 7.1 Extraction/back extraction scheme at two different temperatures.

• Figure 7.1 shows a liquid–liquid extraction/back extraction scheme operated at two different temperatures. Each of the two stages consists of a liquid–liquid mixer, a phase separator (a settler or a centrifuge; it may not have been finalized at this stage), two collecting vessels for the separated phases, and two pumps. The solvent cycle (streams [10] and [11]) passes through three heat exchangers, one for heating with steam, one for cooling with cooling water, and an interchanger between them for energy economy. The feed (stream 1) is first heated then extracted at the higher temperature, and its residual stream exits as (2). Water (15) and possibly some reagents (16) are added to the hot extract (11), and result in the back-extraction aqueous product (3). • Figure 7.2 illustrates a single-stage, liquid–liquid contact pilot installation, presented for actual use in an experimental program. Thus, it includes much more relevant details that need to be referred to in the operational procedure and measurements, such as sampling, temperature indication, liquid interface location, venting, etc. • Figure 7.3 shows a process flow-sheet for a distillation section under vacuum, in which small quantities of a residual volatile organic solvent are eliminated from two aqueous streams (the product solution and the residual solution) in two stripping packed columns. The Copyright © 2002 by CRC Press LLC

mixer VSD closed vent

TI

solvent in

overflow

sample aqueous in

aqueous out

TI

overflow

liquidliquid settler

LI

sample sample

solvent sump

LI

aqueous internal recycle

solvent out aqueous sump

LI

solvent pump

sample

Figure 7.2 Pilot single-stage liquid–liquid continuous contact.

aqueous pump positive VSD

condenser light vapors

product solution

raffinate solution

TI

TI

light vapors rectifier

CW TI

vacuum reflux LI

S

TI

LI

A

TI

TI

LI

LI

live steam S

water to waste

S S

Figure 7.3 Recovery section for vacuum distillation of residual solvent.

Copyright © 2002 by CRC Press LLC

live steam

to raffinate tank

to product tank

to solvent tank

aqueous solutions are fed from the top of the columns and live steam is introduced at the bottom. The vapors from both columns, containing solvent and water, are mixed and sent to the middle of a packed rectification column, which separates the solvent (top) from the water (bottom). The solvent vapors are condensed and part of the liquid is refluxed. Live steam is fed below the packing layer and water is removed from the bottom. The condenser is connected to a vacuum system through vacuum traps (not shown here). Four pumps are needed to remove the four exit streams and transfer them to their respective tanks. Since the operation of such pumps that are sucking liquids from a system under vacuum may be quite problematic, an experienced designer will do everything possible to replace them with “barometric legs” connected directly to the tanks. Therefore, such systems under vacuum are often found in the higher towerstructures in the plant. • Figure 7.4 illustrates a process flow-sheet for the energy-efficient separation, on a relatively large scale, of an extract stream containing a major proportion of a “light” water-soluble organic solvent, together with water and nonvolatile, water-soluble impurities. The extract (stream 1) is first clarified by passing through a pressure filter. In the appended textual description, it is explained that this operation is very important and that only one of two pressure filters in parallel is shown. One of these filters is in operation while the second is being washed and refurbished with filter-aid, but this standard feature does not have to appear on this flow-sheet at this stage. The stream is then fed first to a vapor-recompression evaporator, in which a great part of the volatile organic component can be evaporated in such conditions in which the vapor pressure of the water is still very low. The condensed solvent (10) flows to the recycled solvent tank and the remaining solution (2) passes into a double-effect co-current evaporator, heated with indirect steam in the first effect. Each effect is a forced-circulation evaporator, consisting of a vertical heat exchanger, a vapor–liquid separator, and a circulation pump. The organic vapors from the first effect (12) at the higher temperature are condensed in the heat exchanger of the second effect, and the organic vapors from the second effect (13), at the lower temperature, are condensed with cooling water. The remaining aqueous solution (4) still contains some dissolved organic and, thus, it is reheated and sent to the middle of a packed distillation column, with a steam-heated jacket at the bottom and a condenser with reflux at the top. All the solvent recovered from these four successive operations are mixed in the recycled solvent tank. Such a complicated flow-sheet is justified, or in fact dictated, by the allowable cost of the energy consumption in this process. • Figure 7.5 illustrates a process flow-sheet from a different field, the preparation of dry granules of zircon and CaCl2, which constitute the starting section for a process described earlier (see Chapter 5, Copyright © 2002 by CRC Press LLC

pressure filter

extract

1

double-effect cocurrent evaporator vapor recompression evaporator

12

10

steam

CW

13

22 1

2 3

3 2

10

CW

15

4

condenser 17

reflux

12

recycled solvent tank

steam 4

solvent stripping column steam

Aqueous residual solution

5

Figure 7.4 Complex separation of a volatile solvent from an aqueous solution.

Figure 5.2). It starts from the periodical reception of the merchant zircon sand concentrate in “big-bags” and its transfer into a silo. From there, it is fed at a controlled rate (1) to a wet ball-mill, where it is mixed with a carefully controlled stream (2) of a concentrated solution of CaCl2. In the wet grinding mill, the zircon sand is reduced to very small particles in a concentrated slurry, which is passed through a wet magnetic separator into a holding and blending mixed tank. From there, this concentrated slurry is fed to a fluid-bed agglomerator/dryer, in which hot combustion gases are introduced from the bottom, below the “grid,” and part of the gases from the top are recirculated to maintain the FB and the partial pressure of the water vapor. In this case, complete evaporation of the water is not wanted; on the contrary, a certain concentration of water must be left in the granules, to prevent the beginning of thermal decomposition of the CaCl2 and the liberation of gaseous HCl in the granulator. A partial Copyright © 2002 by CRC Press LLC

crane M-08 big bag 40% CaCl2 tank

Zircon silo S-01 feeder M-09

MT-01

1

2

M-01

fan-01

P-01 wet magnetic separator M-12

wet grinding ball mill

to stack

MT-02 mixed slurry tank P-02

M-02

FB aglomerator/dryer

3

fuel 1

combustion chamber

dried granules

fan-02

Figure 7.5 Preparation of dry granules of zircon with CaCl2.

elimination of the water is sufficient to produce hard granules that can be transferred to the next stage of the process.

7.3 Preparation of an overall detailed description This is a written document that describes the chemical and physical mechanisms of the various sections of the process, its operation, and its control. It should include all the information that is available at this stage, organized in a useful format, so that any new member of the team can catch up on the reference data basis. This presentation should start by explaining the origin and nature of the raw materials and additives that are entering the process. It should include any particular reasons for their choice; their average nominal compositions, which will be entered in the calculations that will follow, and their possible or expected variations (natural or seasonal), which have to be accounted for in the plant’s operation and process control. This description should then systematically cover every piece of equipment chosen (at this stage); the reactions and/or the separations expected in it, its functioning and operating control, and all the possible problems that should be taken into consideration in its final design. The document should conclude by detailing the properties required from the products, in order of importance, and also the requirements and the (possibly) acceptable methods for disposal of the waste streams. Copyright © 2002 by CRC Press LLC

7.4 Listing of all the main process streams This list includes all the main process streams and also the main service streams, as these appear in the process flow-sheets, each with a specific name and number. One should remember that these names and numbers will be used from now on, hopefully by hundreds of individuals and by the plant’s operating staff, for many years, long after the designers’ job is finished. Therefore they should be chosen carefully, to be as clear, descriptive, and easy to remember as possible. (In situations where this requirement has not been met, this can became very annoying to users, who then start to “nickname” these features with their own private code.) Each stream may retain its number as it goes in and out of a piece of equipment, although its temperature or pressure may change, but it should get a different number wherever its nominal composition is expected to change. Note that while a certain stream is passing through a buffering storage vessel, its nominal (average) composition should not change, but its instantaneous composition may change. Since the extent of such changes will probably be analyzed later in the process control modeling, it would be a good practice to give a separate number to the exiting stream.

7.5 Material and heat balances Once the first revisions of the process flow-sheets, the description, and the lists of streams are more or less in workable condition, the next step will be the preparation of spreadsheet tables for the design material balances and for the heat balances, including all main components in all main streams (process and services). First, the bases of the detailed process calculations have to be chosen and agreed upon, for the record. These are generally the nominal plant yearly capacity and the compositions of input streams, although other bases are used occasionally. The nominal plant yearly capacity can be based either on the raw materials input, or on the products output. Note that the second alternative may appear neater, with nice round yearly production figures (some would say too nice!), but it could require more trial-and-error iterative runs, as the process results are optimized. These design balances are generally calculated for one hour of steadystate nominal flows, while the economic calculations (see Chapter 8 below) will be done on a calendar year basis. The average number of production hours expected per year, at the steady-state nominal average rate should be decided and recorded at the beginning. In the chemical industry, this average would generally be in the range of 7,500 to 8,500 hours. This number should be estimated for each case, depending on: • The nature of the process • The extent and frequency of stoppages expected for maintenance work (for example, how often would internal cleaning be needed?)

Copyright © 2002 by CRC Press LLC

• Any “annual shut-down” legally required for certain high-temperature/high pressure processes • The local working habits (annual leave and holidays) The detailed process calculations needed for these balance sheets use well-known chemical engineering tools and are based on: • The general chemical functions (stoichiometric and/or thermodynamic) • The numerical information available on the raw materials • The empirical correlation resulting from the preliminary design data collected in the experimental program (see Chapter 6), or from data that may be available from other sources; such empirical correlation could involve, for example, the reaction or separation yields, the distribution factors for mass transfer operations, and the vapor partial pressures of various components in different conditions • Various “reasonable” assumptions, as needed and agreed, which will always be clearly marked as such; these assumptions can provide for either (1) quantitative relations, such as the percent recovery (split) of a particular component or the percent of entrainment of a phase into the wrong stream, or (2) numerical values, such as the heat transfer coefficient or the weight percent of residual liquid in the cake from a filtering centrifuge The sensitivity of the final results to the value chosen in any of these assumptions (its “leverage”) should be checked before proceeding further. The stronger this leverage is, the higher the danger of a significant error becomes. In this case, additional cross-checking and better procedures need to be devised and carried out urgently, in order to arrive at safer assumptions. The results of the process material balance calculations will be tabulated in a spreadsheet for all main streams, as illustrated for a typical example in Table 7.1 for a solvent-extraction battery: • The nominal flow rates per hour on a weight basis, and sometimes also on a volumetric basis, if needed for design • The compositions in weight and percentage, including all the major components and the minor components of importance • All relevant properties of the stream, such as the temperature, pressure, specific gravity, viscosity, specific heat, and possibly also some other properties relevant to the particular technologies used, such as the wetting contact angle, the dielectric coefficient, and the porosity If any stream does contain more than one phase (a “plurality” of phases, as the patent’s jargon would say), the flow rates, compositions, and relevant properties should be given for each phase separately, as well as for the mixture as a whole, indicating the weight ratio between the phases. Copyright © 2002 by CRC Press LLC

Tablet 7.1 Typical process material balance for a solvent extraction battery Aqueous Feed Stage AAA BBB CCC Water Total Specific gravity Flowrate m3/hr

6,112 509 720 7,951 15,292 1.116 13.703

E1

E2

6,000 499 712 7,951 15,162 1.100 13.784

4,800 468 708 7,951 13,927 1.084 12.848

E3

E4

Aqueous phase kg/hr 3,400 2,400 448 428 708 708 7,951 7,951 12,507 11,487 1.065 1.054 11.744 10.898

E5

E6

E7

1,400 418 708 7,951 10,477 1.040 10.074

700 408 708 7,951 9,767 1.031 9.473

300 398 708 7,951 9,357 1.019 9.183

Aqueous Residual E8 112 318 708 7,951 9,089 1.009 9.008

Organic phase kg/hr

AAA BBB CCC Water Solvent Total Specific gravity Flowrate m3/hr

Copyright © 2002 by CRC Press LLC

Extract 6,050 211 30 7,000 90,750 104,041 0.971 107.148

5,938 180 22 7,000 90,750 103,890 0.965 107.658

4,738 160 18 7,000 90,750 102,666 0.945 108.641

3,338 140 18 7,000 90,750 101,246 0.939 107.823

2,338 130 18 7,000 90,750 100,236 0.934 107.319

1,338 120 18 7,000 90,750 99,226 0.926 107.156

638 110 18 7,000 90,750 98,516 0.912 108.022

238 100 18 7,000 90,750 98,106 0.906 108.285

Recycle Solvent 50 20 18 7,000 90,750 97,838 0.904 108.228

AAA BBB

39.969 3.329

Aqueous phase % w/w 27.185 20.893 13.363 3.582 3.726 3.990

7.167 4.177

3.206 4.254

1.232 3.499

Organic phase % w/w 4.615 3.297 0.156 0.138 6.818 6.914

2.332 0.130 6.984

1.348 0.121 7.055

0.648 0.112 7.105

0.243 0.102 7.135

6.337

5.729

5.315

4.951

5.079

26.946 9.89

30.762 10.65

34.542 11.31

38.094 11.76

34.325 12.02

39.573 3.291

34.465 3.360

AAA BBB Water

5.815 0.203 6.728

5.716 0.173 6.738

AAA distrib. A/O BBB distrib. Net flowrate ratio O/A

6.805

6.030

5.891

16.228 7.77

19.395 8.38

22.984 9.25

Copyright © 2002 by CRC Press LLC

0.051 0.020 7.155

It is important to realize that different rows may be calculated by different methods and from different sources. As the whole picture could be quite complex, it is a good practice to include in each such spreadsheet an automatic overall balance check for the different recognizable components in the inlet and outlet streams. In any case, these overall balances will be needed for the “black-box” presentation (see below). Differing from the overall approach needed for the material balances above, the process heat balance calculations should be started and presented separately for each of the different operations in which: • Heat can be generated or consumed by a reaction (including combustion), or by a significant input of mechanical energy • Heat is transferred between different streams, either directly by contact between different process phases, or indirectly in heat exchangers For each such operation, a quantitative heat balance must be produced in any convenient format, which can be checked or modified if needed. Such balances should include assumptions related to heat loss to the surroundings and the nature of any heat insulation system assumed (i.e., the area and transmission characteristics). A typical example is presented in Table 7.2. One should note that the heat resulting from the input of mechanical energy is generally small and neglected by chemical engineers in most cases (in pumps, agitators, etc.), but this could be a major item in certain operations, such as gas compression or expansion, or solid grinding. When all the heat balance results are available for the different operations, they should be compared and analyzed to see if any synergetic combination could be possible in order to minimize the overall energy consumption and/or its cost. Finally, the total heat requirements are calculated and translated into a consumption of services (fuel, various grades of steam, cooling water) as discussed below in Section 7.7, and tabulated as in the typical example in Table 7.3 below. It is also advisable to join to these tables a detailed description of the methodology that was used in these calculations and any literature reference for the functions used, to allow independent checking by other users. In the choice, preparation, and description of methodology, one should take into account that it will most probably be used again in the future, possibly by other engineers, when the process optimization will require a systematic screening of the effects of the different variables (see Chapter 11).

7.6 Material handling operations In large chemical plants, these material-handling operations, needed to bring in the raw materials supply and to send out the products of the plant, will generally require special arrangements, extra space, roads, railway tracks, storage volumes inside the plant, etc. All these material-handling Copyright © 2002 by CRC Press LLC

Table 7.2 Typical Process Heat Balances (metric units) Stream Example 1 FB Dryer Granulator M-02 Solids in heating from 25 to 200 Solution in heating from 25 to 200 Water conc. + evap. Heat losses loss 30% from above Combustion gases cooling from 1350 to 250 Fuel combustion

kg/hr

m Diff

Heat

kcal/hr

1000 1516 809

175 175

0.23 0.74 590

2621

1100

0.24

40,250 196,322 477,310 214,165 691,465 692,000

8,500

692,000 314,088 288,135 59,590 132,363 794,176 794,150 351,552 442,598 442,598

81.4

Example 2 First Kiln K01 Granules in heating CaCl2 endothermic decomposition Water conc. + evap. Heat losses loss 20% from above

1707 607 101

800

0.23 475 590

Combustion gases cooling from 1200 to 400 Combustion gases recycled from kiln 2 Combustion gases from combustion Fuel combustion

4136 1831 2305 46.6

800 800 800

0.24 0.24 0.24 9,500

1321

400

0.23

1831

400

0.24

121,532 13,688 40,566 175,786 175,776

10,000

175,776

0.18 0.243

282,096 282,175

Second Kiln K01 Solids heating from 1000 to 1400 Endothermic reaction Heat losses 30% loss 20% from above Combustion gases cooling from 1600 to 1200 Fuel combustion Clinker Cooler K03 Clinker cooler from 1400 to 200 Air heating from 25 to 75

17.6 1306 1792

1200 648

operations are not different for a new process from those required for a conventional process, but the design options and choices can be wider when starting anew. A detailed description of the objective needs should be presented and distributed to the team and the consultants to invite creative proposals. In addition, the requirements and available options for the disposal of each waste stream should be described in relation to the local environmental and ecological regulations in the area (with any available dumping pond or any means for its transfer into some existing waste treatment installations in the affordable vicinity). Copyright © 2002 by CRC Press LLC

Table 7.3 Typical summary table for the services required Stream

Description

1 1-a 2 3 4 5 6 7 8

Process water Process water Boiler-feed water Steam Cooling water Heavy fuel no. 6 Fuel oil no. 2 Compressed air Electricity

Copyright © 2002 by CRC Press LLC

Characteristics Municipal Municipal Own 7 ata In 28°C Out 40°C 10,000 kcal/kg 10,000 kcal/kg 8 ata Purchased

Unit m3 m3 m3 MT m3 MT MT Nm3 kw-hr

Consumption per Hour Year 749 487 24 167 34,415 15 2.7 3,360 780

5,843,000 3,800,000 187,200 1,300,000 268 MM 117,000 21,000 26.2 MM 6.0 MM

Notes Without recycle With recycle Make-up Medium pressure Own tower For steam For dryer

Typically, the inventors and R&D personnel consider all these materialhandling operations very “trivial,” and these are often neglected or ignored altogether in their process presentations (“the engineers will take care of that later”). However, although generally conventional mechanical devices, these are in fact quite expensive and visible as the facade of the new plant. The direct cost of all these material-handling installations can be a major consideration in many cases, or even a decisive factor for the future of a project, or at least for its particular location. The preparation of a clear quantitative description of the operational options and their relative costs, at an early stage, could allow all the team to think in concrete frameworks and possibly create better or cheaper solutions.

7.7 Summary tables for all required services The nominal (average) flow rates for all services will derive directly from the material and heat balances calculated above and will be used for the economic calculations. But the design quantities for the supply of services that should be available to the plant generally include higher instantaneous rates (i.e., for starting, stopping, or emergency), and possibly also a significant reserve for eventual increase in production. An important design decision involves the maximum delivery rate designed for each service. This data is tabulated, as shown in a typical example in Table 7.3. There could also be several different possible options for the supply of each service. They are essentially major cost factors and a wide field is open for optimization studies, to achieve the cheapest and most convenient solution. Again, these issues are not different for a new process from those considered for a conventional process, but the choices and options may certainly be wider before the process is finalized. The services generally considered in most chemical plants are: • Fuel, either for direct use in a combustion device incorporated into the process, or for indirect burning for the production of steam or another heating medium (oil, brine) in the new plant. Different types and qualities of fuel could be available and considered, from coal or liquid petroleum fractions to natural methane gas. In many cases, internal waste streams are also burned. In addition to considerations of cost and convenience, the impurities in the flue gases discharged from the stack (SO2/SO3, nitrous oxides, metallic dust, and so on) or fly ash could be a decisive factor in the choice of fuel. There could also be local ecological restrictions, which may require intensive cleaning installations and cancel the advantages of a cheaper fuel. • Several types of steam (high, medium, or low pressure) or other heating media, which may possibly be purchased from the site’s central services or from adjacent producers. With indirect steam functions, the condensed water is returned to the boiler circuit, while when direct

Copyright © 2002 by CRC Press LLC

•

•

• •

•

steam is used in the process, the condensed water stays with the process streams and has to be replaced as “boiler-feed-water” quality. Cooling water, generally produced in the new plant’s own cooling tower, but sometimes received from a nearby sea or river. The maximum supply temperature of the cooling water should be clarified at an early stage, since it could affect the process operating temperature, or at least the size of the cooling heat exchangers. In some cases, chilled water (in the range of 4 to 15°C) could be needed and should be prepared in a separate installation. Electricity is generally purchased. Only in few cases would the new demand be large enough to justify the purchase of a generator. Even then, a back-up connection to the external electrical supply grid would also be needed. Some process operations may require an emergency electrical provision for safety or damage control (installed and automatically started). If this situation is expected with the new process, it should be emphatically stated and a separate list started for the electrical consumers that will need to be connected to the electrical emergency supply. Compressed air, generally produced in the new plant’s own station. Oxygen for certain reactions and nitrogen as inert gas, if needed, can be purchased in certain cases, or have to be produced on site by an air separation installation. Fire-fighting water supply and rig.

The availability at the site in the quantities required, and the unit cost of each of these services or possibly the need for new additional installations for self-supply, should be investigated, presented, and discussed.

7.8 Major pieces of process equipment The items of equipment in the process flow-sheets may be divided and listed separately into four broad categories (see a typical example in Table 7.4): 1. Small standard equipment, which can be selected from catalogs from a relatively large number of suppliers, such as pumps, fans, blowers, agitators, standard heat exchangers, etc. 2. Custom-built standard equipment, which is made to order from the engineering drawings in fabrication workshops. The costs are estimated mostly on an empty weight basis, or even on a volume basis. For example, tanks, silos, large separation vessels, etc. 3. Major process equipment, which is detail-designed and made to order by specialized suppliers. The preliminary selection of type, model, and size of each of the major pieces of process equipment will be presented on the basis of a functional analysis of its duties and quantified in the above-mentioned balances (average only).

Copyright © 2002 by CRC Press LLC

Table 7.4 Typical List of Major Process Equipment Tag T-01 T-02 T-03 S-01 MT-01 MT-02 MT-03 MT-04 MT-05 MT-06 M-01 * M-02 * M-03 C-01 * K-01 * K-02 * K-03 * M-04 * M-05 * M-06 M-07 M-08 M-09 M-10 M-11 M-12

Name HCl solution tank Fuel tank Water tank Raw material silo CaCl2 soln mixed tank Slurry mixed tank Leaching first mixed tank Leaching sec. mixed tank Leaching third mixed tank Lime mixed tank Grinding ball mill FB Aglom. dryer Clinker hammer mill Adiabatic absorption tower Kiln 1 Kiln 2 Clinker cooler Belt filter Product dryer (solids) Cooling screw conveyor Wet cake elevator Overhead crane Solids feeder Bagging machine Dry granules elevator Wet magnetic separator

Size 50 100 300 70 50 20 1.9 1.9 1.9 20 1.80 ID 1.48 ID 1.5 1.8 ID 0.87 ID 0.87 ID 1.0 ID 32 700 0.1 ID 1.5 5 0.4 1 2 0.8

3

m m3 m3 m3 m3 m3 m3 m3 m3 m3 3.2 m. L 3.0 m. L tph 8.0 m. L 12.4 m. L 8.62 m. L 10 m. L m2 kg/h 2 m. L tph mt m3/h tph tph m3/h

MOC** FRP MS MS MS MS MS FRP FRP FRP MS RLS MS MS FRP + RL BL BL SS / BL FRP + RL SS SS FRP + RL MS MS SS SS

Kwatt inst. 0 0 0 0 5 10 1 1 1 10 150 0 15 0 25 25 25 50 3 1 1 3 1 0 1 1

Service needed HCl soln

200 mt.

HCl soln HCl soln HCl soln for waste neutral incl. classification circuit fuel +CaCl2 soln incl. quenching water fuel - max. 1000 C. fuel - max 1400 C. with air CC cake washing rotary trays -fuel CW jacket

50 kgs bags

continued

Copyright © 2002 by CRC Press LLC

Table 7.4 (continued) Typical List of Major Process Equipment Tag P-01 P-02 P-03 P-04 P-05 P-06 P-07 P-08 P-09 P-10 Fan-01 Fan-02 Fan-03 Fan-04 Fan-05 Fan-06

Name CaCl2 brine pump Slurry pump Clinker slurry pump Waste stream pump Wash water pump HCl solution pump A HCl solution pump B Fuel pumps (1+1) Water pumps (1+1) Milk of lime pump From aglom. dryer Air to combustion 1 Air to combustion 3 Air to clinker cooler To stack Air to combustion 4 Total

* = complete package ** = material of construction

Copyright © 2002 by CRC Press LLC

Size 1.3 2 1.5 5 3 2 3 0.3 10 0.5 20 2 1 0.7 2.5 1.4

3

m /h m3/h m3/h m3/h m3/h m3/h m3/h m3/h m3/h m3/h m3/sec m3/sec m3/sec m3/sec m3/sec m3/sec

MOC** MS MS SS PP PP PP PP MS MS MS MS MS MS MS FRP MS

Kwatt inst. 1 1 1 3 2 1 2 1 5 1 5 1 1 1 1 1 357

Service needed

4. In certain cases, the new process may require the development and design of a new or modified type of reactor or separator, which cannot be procured readily from established suppliers. Of course, this will create an additional load on the development effort and on the investment, time, and talents needed, but it could also increase the intrinsic value of the new know-how generated. If this is the case, this new equipment will obviously be at the center of attention of the project team. For each major piece of equipment, a specification file will be opened in which this selection should be recorded, together with: • A detailed explanation of all the possible options and the reasons for the recommended choice of type, model, size, and any other important specification • The selection of the materials of construction • The estimated electrical supply • The plant space needed for this major equipment can also be indicated, for the lay-out studies (area and height, free space access needed, rigid connections) A list of all known potential suppliers of the recommended type can also be included, to allow further inquiries, as needed. In certain cases, one preferred supplier can be recommended. This critical dependency could simplify, but could also complicate the situation, and many corporations oppose such situations as a matter of principle.

7.9 Main operational and control procedures At this stage, a detailed operational and control manual is not needed and, in any case, it cannot be delivered before the detailed engineering stage, when the P&IDs are well advanced. However, a first write-up of the main operational and control procedures should be prepared in relation to the process flow-sheets, including: • Start-up and stoppage requirements • Any particular safety aspect of this installation (for example nitrogen blanketing) • Waste disposal methods Emphasis in this write-up should be on those procedures that may be unusual at the particular location, may require a major effort to develop or implement, or may have a significant cost attached. Specialized consultants and corporate staff should be consulted on the operational aspects of the plant. This first draft of the main operational and control procedures should be intended mostly as a reference basis for the comments of all members of Copyright © 2002 by CRC Press LLC

the preliminary process design team. It will be gradually developed, by adding different contributions, and presented for review with the first P&IDs (see Chapter 10, Section 10.4).

7.10 Listing of required staff An experienced operation manager or consultant will prepare a first list of all the personnel positions that will be needed for operations (mostly shift-work), management, supervision and control, quality assurance, the generation of services, plant maintenance, and any special requirement of the project. For the different positions, the availability, extent, and cost of training (part of the investment), and the yearly cost of this staff (fixed operating cost) should be estimated. This listing and discussion will be related to the general industrial experience in the specific area under consideration and the expenses (as a total) will be included in the economic calculations in Chapter 8. In many cases and remote locations, the availability of adequate staff for particular positions cannot be taken for granted, and it may have to be relocated with special effort and extra costs.

7.11 Worth another thought • The preparation of the first set of process flow-sheets represents the translation of the process block diagram and the chemical mechanisms concepts used by the R&D group into chemical engineering methodology and practice. This bridge between these two disciplines is an important creative task, requiring a fundamental scientific understanding of the process as well as extensive experience in the design and operation of chemical plants. • The overall detailed description of the chemical and physical mechanisms in the various sections of the process, as well as its operation and control, includes all the information currently available, in a useful format so that any new member of the team can become familiar with the reference data basis. • The new process may require the development of a new or modified type of reactor/separator that cannot be procured readily. This creates an additional load on the development and the investment, time and talents needed, but it could also increase very much the intrinsic value of the new know-how generated. In such case, this new equipment will be at the center of attention of the project team.

Copyright © 2002 by CRC Press LLC

chapter 8

Economic analysis of the specific proposal 8.1 Purpose At this stage, the only purpose of this preliminary economic analysis of the proposal is to supply information on the profit potential of the final plant, which is needed to justify the continued expense for its development. This is definitely not the kind of economic analysis that will be needed later to approve the investment of tens or hundreds of millions of dollars in an industrial plant (see Chapter 10). The preliminary analysis can be done within the project team, if an experienced “cost engineer” is available, or by a specialized consultant, or it may be subcontracted to an engineering company. It is based on: • The process data contained in the preliminary design package, which was described in the previous chapter • The cost data available in the files from previous projects • The input from the marketing experts • The non-committing, up-to-date quotations for the major equipment, obtained by direct contact with potential suppliers (not as formal tenders) To save time, this economic analysis could be started on partial drafts of the preliminary design package, which could be supplied informally, but of course, the “bottom line” recommendation can only be completed after the preliminary design package is completed and ready to be formally presented.

8.2 Preliminary estimate of the Fixed Capital Investment (revision 0) The working methodology for the preparation of fixed capital investment (FCI) estimates is well known from textbooks.1,2 It is also practiced in

Copyright © 2002 by CRC Press LLC

most engineering groups with good results and need not be discussed here in detail. Just to recall some “fact-of-life”: For a preliminary estimate, the purchase, delivery, and installation costs of the major pieces of equipment are estimated separately, from the company’s own records, or from published cost correlations, but mostly from up-to-date quotations obtained “on-the-wire” from suppliers. Note that such preliminary quotations can be obtained at this stage from only three or four suppliers, as it is not intended to make a bid comparison, but only to select a reasonable average cost for this task. But if one of the quotations received from the suppliers is “way out” from the average, this could indicate a problem in the concept or the wording of the request specification, which should be clarified. Table 8.1 illustrates a typical example for a preliminary small project for the production of 5,000 tons per year (tpy) of zirconium oxide according to a novel process (Gorin-Mizrahi, described in Chapter 5, Figure 5.2). The list of equipment is taken from the preliminary process flow-sheets and the estimated installed costs for each are based from recorded data from other projects, with the necessary conservative adaptation and updating. One should note that 90% of the installed equipment cost can be attributed to eight complete packages from specialized suppliers (kilns, dryer, ball-mill, agglomerator, etc.), including their design and operating know-how. The sum of the installed costs of all the major pieces of equipment is then multiplied by different relative statistical factors, representing the cost contributions of minor standard equipment, buildings, piping, electrical, instrumentation and control, services connections and infrastructure, engineering and management, start-up, and miscellaneous. These different statistical factors are individually chosen by an experienced cost engineer on the basis of the recorded analysis of previous projects, and are adapted to the specific characteristics of the present case (such as its location, request for explosion-proof conditions, large flows of gases, etc.). For the case described in Table 8.1, it was estimated that a factor of 3.0 would be sufficient, considering a lower need for detailed engineering and electrical hardware. Separate safety margins (reserves) are then chosen to fit the uncertainty built into the present state of project definition. These safety margins are added to the total sum obtained above, but they will have to be reconsidered in each of the future revisions of this FCI estimate, as the process and implementation conditions will be more focused and their uncertainty range will be decreased. For the case under discussion, a safety margin of 30% was added to reflect the early status of the estimate, bringing the total FCI preliminary estimate to $16,510,000. This does not include the promoters’ own expenses, or the cost of additional testing. This is the usual methodology for a preliminary FCI estimate. The format used is not critical, as a completely new format will be used later for the “real” investment budget, when it will be prepared for approval of the plant (see Chapter 10). Copyright © 2002 by CRC Press LLC

Table 8.1 Typical Preliminary Fixed Capital Investment Estimate Tag T-01 T-02 T-03 S-01 MT-01 MT-02 MT-03 MT-04 MT-05 MT-06 M-01 * M-02 * M-03 C-01 * K-01 * K-02 * K-03 * M-04 * M-05 * M-06 M-07 M-08 M-09 M-10 M-11 M-12

Name HCl solution tank Fuel tank Water tank Raw material silo CaCl2 solution mixed tank Slurry mixed tank Leaching first mixed tank Leaching second mixed tank Leaching third mixed tank Lime mixed tank Grinding ball mill FB Aglom. dryer Clinker hammer mill Adiabatic absorption tower Kiln 1 Kiln 2 Clinker cooler Belt filter Product dryer (solids) Cooling screw conveyor Wet cake elevator Overhead crane Solids feeder Bagging machine Dry granules elevator Wet magnetic separator

Number 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

MOC**

Size 50 100 300 70 50 20 1.9 1.9 1.9 20 1.80 ID 1.48 ID 1.5 1.8 ID 0.87 ID 0.87 ID 1.0 ID 32 700 0.1 ID 1.5 5 0.4 1 2 0.8

3

m m3 m3 m3 m3 m3 m3 m3 m3 m3 3.2 m. L 3.0 m. L tph 8.0 m. L 12.4 m. L 8.62 m. L 10 m. L m2 kg/h 2 m. L tph mt m3/h tph tph m3/h

FRP MS MS MS MS MS FRP FRP FRP MS RLS MS MS FRP + RLS BL BL SS / BL FRP + RLS SS SS FRP + RLS MS MS SS SS

Installed Cost 20,000 15,000 32,000 30,000 17,000 20,000 20,000 20,000 20,000 25,000 900,000 1,000,000 50,000 100,000 500,000 500,000 100,000 500,000 200,000 7,000 20,000 10,000 15,000 25,000 10,000 18,000 continued

Copyright © 2002 by CRC Press LLC

Table 8.1 Typical Preliminary Fixed Capital Investment Estimate Tag P-01 P-02 P-03 P-04 P-05 P-06/07 P-08 P-09 P-10 Fan-01 Fan-02 Fan-04 Fan-05

Name CaCl2 brine pump Slurry pump Clinker slurry pump Waste stream pump Wash water pump HCl solution pump Fuel pumps Water pumps Milk of lime pump From aglom. dryer Air to combustion Air to clinker cooler To stack total

Number 1 1 1 1 1 2 1+1 1+1 1 1 4 1 1

MOC**

Size 1.3 2 1.5 5 3 2 0.3 10 0.5 20 2 0.7 2.5

3

m /h m3/h m3/h m3/h m3/h m3/h m3/h m3/h m3/h m3/sec m3/sec m3/sec m3/sec

MS MS SS PP PP PP MS MS MS MS MS MS FRP

Installed Cost 2,500 2,500 4,000 5,000 5,000 10,000 3,000 6,000 2,500 5,000 6,000 2,000 2,000 4,162,500

* = complete package ** = material of construction FRP = fiberglass reinforced polyester; MS = mild steel; RLS = rubber-lined steel; BL = brick-lined; SS = stainless steel; PP = polypropylene

Copyright © 2002 by CRC Press LLC

But there is an additional item that is often neglected, in the experience of this author. That is the purchase cost of the internal inventory that is needed to arrive at the steady-state operation of the plant. At most, the relatively small value of the “work-in-progress” is recorded, that is, the cost of the different materials having intermediate compositions between the raw materials and the products, contained in the different pieces of equipment piping and storage. The cost of the “work-in-progress” depends on the unit costs of the materials and on the various residence times (as an extreme case, consider for instance a solar pond in a salt operation, which takes several years to fill and concentrate). Many plants also require auxiliary purchased materials. For example, the solvent stock in a solvent-extraction plant may account for a quite significant part of the FCI, depending on the unit cost of the particular solvent used and on the internal volumes required. A similar situation exists with mercury cells, resins columns, circulating active-carbon, thermal oil, etc. There is definite room for optimization in this area, which is not always appreciated. This “omission” is often attributed to the fact that many certified accountants do not accept the value of this internal inventory as part of the FCI, which is used to calculate the tax-allowed depreciation and profitability of the investment, and also to evaluate indirectly the cost of maintainance. Despite the bookkeeping formalities, however, almost all of the value of this internal inventory is a one-time expense, which cannot be recovered, even if the plant is terminated, and which, furthermore, will require some periodical make-up. Another controversial issue is how to handle past and future development expenses, in relation to the FCI of the first plant, which could be build on a modest scale. Again, many certified accountants do not accept these development expenses as part of the FCI. There are different practices in this regard. Most corporations have probably already included the (recorded) past expenses in their yearly balances, and they are now “forgotten.” Other corporations may have capitalized these past expenses as part of their investment in “special subsidiaries” (daughter companies, joint ventures) and these sums have to reappear in a new plant’s investment. The same issue will be related to the treatment of (expected) future development expenses, to the point when a decision to build a plant is reached. After that point, all development expenses are generally included in the “engineering” budget, a definite part of the FCI. Furthermore, the governments of certain developing countries encourage the establishment of new industries (declared of national interest) by contributing a grant of 20 to 40% of the FCI, subject to certain conditions. Such a grant could change much of the “rules-of-the-game.”

8.3 Estimate of operating costs These operating costs are generally divided into fixed costs and variable costs, to allow studies on the effect of different production levels. (See a typical example in Table 8.2.) Their estimate is a standard compilation of all the different operating cost categories, that is delivery unit costs and the consumption of: Copyright © 2002 by CRC Press LLC

Table 8.2 Typical Preliminary Operating Cost Estimate Fixed Cost per Year

Units

Management / sales employees Operation / shift employees Other employees Overhead on employees Indirect taxes and insurances Spare parts Waste disposal R&D Total fixed costs per year

Man-year Man-year Man-year Estimate Estimate Estimate Estimate Estimate

No. 3 12 8

Unit Cost 80,000 50,000 40,000

Interest on working capital per year Variable Costs per Ton of Product Raw material A Ton 1.51 Raw material B Ton 0.916 HCl (100% basis) Ton 0.076 Lime Ton 0.075 Heavy fuel Ton 0.39 Water m3 15 Electricity Kwatt.h 425 Packing and transportation to Estimate CIF Total variable costs

• • • • • • • • • •

Total 240,000 600,000 320,000 250,000 100,000 100,000 50,000 180,000 2,000,000 263,000

435 100 35 100 35 0.32 0

656.85 91.60 2.66 7.50 13.65 4.80 25.50 44.00 846.56

Raw materials and different materials additives Services (see below) Any disposal cost of the waste streams Any packaging needed and the shipping of the products The maintenance of the plant, including property taxes Any contribution to maintenance of the site, taxes to the city, county, etc. The yearly cost of the plant staff and contractors Sales expenses, with any duty and taxes (if relevant) Financial costs, i.e., depreciation and interest on the operating capital “Miscellaneous” other minor factors

To each significant cost category, a separate safety margin (reserve) is added to reflect the present insufficient state of knowledge on consumptions and actual delivery costs. These reserves can be added either in the units needed (i.e., number of kilowatts per hour) or in the unit costs. These should be noted for the record and should be reconsidered in each future revision of the operating cost estimate, as more reliable information is progressively accumulated. Contrary to the FCI estimate, the detailed format used for this estimate of operating costs will probably be used later in many revisions, by other engineers and managers, for repeated economic studies and for routine

Copyright © 2002 by CRC Press LLC

production budget planning. So it is worthwhile to consider from the beginning the most convenient and detailed working format.

8.4 Expected net sales income estimate The expected net sales income is generally the “weaker point” of the preliminary economic study, since it has a very large “leverage.” It can only be estimated, with the active contribution of the market experts and consultants, from the expected sales of product, quantity and invoiced prices, after deducting any possible expense related directly to sales, such as the various freight expenses to bring the containers from the plant to the final destination, agents’ commissions, bank transfer fees, customs duties, and so on. The confidence level in this estimate could be very different, if: • The product is intended to be sold in an already established market, with a recorded market price history, or within an exclusive sales contract to a wholesale distributor • The product is relatively new or improved, and its expected sale price can only be based on what it should be worth to users In addition, the interest on working capital or the cost of the standing credit at the bank should be estimated, from data on the percent of the product in store, in transit or payment delayed as per sales conditions, on the sales revenue and on the expected level of banking interest. For example, in the above case, 3 months of credit at 7% gives an expense of $262,500 per year for 100% production.

8.5 Profitability calculation The expected profitability is calculated following one of several standard methods, which are used in different countries and industrial sectors. This profitability is generally expressed as the percent of return on investment (ROI), before any taxes on corporate income or profits, or as the present value of the operation of the implemented project over a period of time, for example 10 years. Table 8.3 gives an example of the presentation preferred by this author, at this stage of the project review. It includes: • A project cash flow for a period of 12 years, in which the first 2 years are for construction and the following 10 years for production at increasing rates. For example, 50% of the nominal production in the third (start-up) year, 75% in the fourth (consolidation) year, 100% for the next 5 years, then a slight but gradual increase of production to 110%. (Note that this last assumption has almost no financial consequence in a healthy project; it is only included as an expression of confidence in the future.) Copyright © 2002 by CRC Press LLC

Table 8.3 Typical Return On Investment — Based on Cash Flow and on Present Value of Yearly Cash Flow (in $1,000) Total Investment Average Sales $/Ton CIF Pecent of Design Production Return on Investment Year

16,770 3000 100 31.1% 1

Fixed Capital Investment FC Investment grant Production % of Design Capacity Total production sale value, CIF Fixed Production expenses Variable production expenses Interest on Working Capital Royalties Net Cash Flow Disc. Net Cash Flow PV (10% rate) project's present value at 10% ROI Disc. Net Cash Flow PV (ROI rate) Cum Disc. Net Cash Flow PV

(6,708)

Copyright © 2002 by CRC Press LLC

2 (9,224)

3

4

5

6

7

8

9

10

11

12

(839) 50

75

100

100

100

100

100

105

107

110

7,500 (2,000) (2,141) (131)

11,250 (2,000) (3,211) (197)

15,000 (2,000) (4,281) (263)

15,000 (2,000) (4,281) (263)

15,000 (2,000) (4,281) (263)

15,000 (2,000) (4,281) (263)

15,000 (2,000) (4,281) (263)

15,750 (2,000) (4,495) (276)

16,050 (2,000) (4,581) (281)

16,500 (2,000) (4,709) (289)

(300)

(1,000)

(7,008) (7,008)

(10,224) (9,294)

2,390 1,975

5,842 4,389

8,457 5,776

8,457 5,251

8,457 4,773

8,457 4,340

8,457 3,945

8,979 3,808

9,188 3,543

9,502 3,330

(7,798)

1,390

2,593

2,863

2,184

1,666

1,270

969

785

613

483

24,828 (7,008) 9

• The projected capital investment is divided according to a reasonable pattern, such as 40%, 55%, and 5% in years 1, 2, and 3, unless there is a specific reason to differ. Any expected FCI grant is deducted. • The average net sales return per ton is taken as constant, for lack of more information. • The calculated fixed production expenses (see above) start in year 1 as 15% of the average, and 50% in year 2. • The calculated variable production expenses (see above) are assumed to be proportional to the production. • Royalty payments are generally not relevant at this stage, but interested parties could use this format in their eventual negotiation, to see the impact of various royalty payments on the profitability. This can be either a yearly fixed sum, a percentage of the net sales, or a percentage of the profits according to some formula. • The net cash flow (negative or positive) is calculated for each year. If it is discounted at 10% (say, as a “normal safe” investment), the present value of the project is obtained from the sum at year 1, which represents the potential contribution of the new proposal/know-how (nearly $25 million in Table 8.3). • In addition, a discount rate can be calculated by simple trial-and-error, which results in present value = zero. This is in fact the ROI (31.1% in Table 8.3). Once prepared, this spreadsheet can be used easily to survey the effect of various factors on either the ROI or the present value of the operation, such as for instance, different values of FCI, net sales returns, raw materials cost, and so on. Different corporations use different standards to judge the attractiveness of new process proposals, according to their prevailing strategy considerations. As a general order of magnitude, however, we can say that in the free economy of the Western world, the profitability test at this stage should probably show an ROI of, at least, 25% before taxes, to justify the continuation of an intensive development and implementation effort. But if this development opens new, promising avenues to the corporation, it could well be that a lower ROI would be accepted for the first plant (see also Sections 8.6 and 8.7 below). Of course, the degree of taxation varies in different locations and situations. Certain countries promote the establishment of new industries by providing them with an investment grant (say, as a fixed percentage) or with a period of reduced (or no) income-tax payment.

8.6 Optimistic evaluation of the profit potential in other applications Together with the profitability study, the promoters could also develop and present to the decision makers the possibility that the proposed process may have a larger potential for profitable applications, once the first implementation Copyright © 2002 by CRC Press LLC

is proven to be successful. Such larger profit potential can be in one or more of the following forms: 1. The simplest formula is increased production volume at the same site in the future. This extra product can be obtained practically with the same management and services, by making use of the built-in oversized facility and of the experience gained by the operating staff, and could be sold in developing markets. Thus, instead of the usual cash flow and profitability spreadsheet based on the present nominal capacity for 10 years, an alternative spreadsheet could be prepared in which the production would be increased by 10% (for example) every 2 to 3 years and the span increased over 15 years, reaching 150%. This is a frequently used format. Table 8.4 illustrates this change with regard to Table 8.3 above. It can be seen that on this basis, the project’s net present value at 10% discount rate increased from $25 to $46 million. Note that the increase in ROI is not as spectacular, only from 31.1% to 32.8%. This is typical of the cases with quite high ROI, where the contributions of the later years are felt less and less. For another case with an ROI of 15%, this increased production could have raised the resulting ROI to 25%, and this change would have made a different impact. 2. Another commonly considered possibility involves future “repeat” plants built in other locations or countries, on the basis of the experience learned in the first plant. 3. A more complex possibility is the adaptation of the novel process technology to the production or improvement of (a line of) similar new products. The presentation of such potential applications could change the perspective of the decision makers from short-term cash flow into a wider corporate strategy. Of course, the access and exclusivity of these options would need to be secured by adequate patents.

8.7 Possible synergetic effects with other production facilities In many cases, the promoters may also gain the goodwill of the decision makers by pointing out synergetic (that is, mutually profitable) effects between the proposed project and some other existing or planned industrial facility of the corporation. For example, the proposed project may: 1. Use or upgrade the value of a by-product or waste stream. For example, the recovery and profitable use of valuable acids from a waste stream, instead of neutralizing them, or the utilization of excess concentrated thermal energy, instead of dispersing it into the surroundings. Copyright © 2002 by CRC Press LLC

Table 8.4 Typical Return On Investment with increased production volume and on Present Value of yearly Cash Flow based on Cash Flow Total investment Average sales $ / ton CIF % of design production Return on investment Year Fixed Capital Investment F C Investment grant ( 20% ) Production % of Design Capacity Total production sale value, CIF Fixed Production expenses Variable production expenses Interest on Working Capital Royalties Net Cash Flow Disc. Net Cash Flow PV (10% rate) project's present value at 10% ROI Disc. Net Cash Flow PV (ROI rate) Cum Disc. Net Cash Flow PV

16,770 3000 100 32.77% 1

2

(6,708)

(300)

(7,008) (7,008)

(9,224)

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

(839)

50

75

100

100

110

110

120

120

120

130

130

140

140

150

150

7,500

11,250

15,000

15,000

16,500

16,500

18,000

18,000

18,000

19,500

19,500

21,000

21,000

22,500

22,500

(2,000)

(2,000)

(2,000)

(2,000)

(2,000)

(2,000)

(2,000)

(2,000)

(2,000)

(2,000)

(2,000)

(2,000)

(2,000)

(2,000)

(2,000)

(2,141)

(3,211)

(4,281)

(4,281)

(4,709)

(4,709)

(5,137)

(5,137)

(5,137)

(5,565)

(5,565)

(5,993)

(5,993)

(6,422)

(6,422)

(131)

(197)

(263)

(263)

(289)

(289)

(315)

(315)

(315)

(341)

(341)

(368)

(368)

(394)

(394)

(10,224) (9,294)

2,390 1,975

5,842 4,389

8,457 5,776

8,457 5,251

9,502 5,364

9,502 4,876

10,548 4,921

10,548 4,473

10,548 4,067

11,593 4,063

11,593 3,694

12,639 3,661

12,639 3,328

13,685 3,276

13,685 2,978

(7,700)

1,356

2,496

2,721

2,050

1,735

1,307

1,092

823

620

513

386

317

239

195

147

(1,000)

45,791 (7,008)

4

Copyright © 2002 by CRC Press LLC

2. Make use of idle production capacity in certain operations, in packaging or in utilities generation, or exploit the significant cost advantage of larger installations. 3. Utilize available developed land, roads, warehouses, and similar facilities. 4. Participate in a combined marketing effort aimed at the same users, for example in the compound fertilizers market. Such interactions would involve the management of the existing operation and, obviously, their essential cooperation and good will should be secured by the promoters before the above claims are presented to a larger audience.

8.8 Comprehensive report for the justification of the specific proposal The expected profitability is the “bottom line” of a comprehensive report presented to the relevant corporate management for a detailed and exhaustive review leading to a “no/maybe” decision, which should include, with an executive summary: • • • •

The preliminary process design detailed in Chapter 7 The economic estimates listed above in this chapter The profit potential for other applications, described in Section 8.6 Some possible synergetic effects, described in Section 8.7

Many proposed projects have met their end at this stage (“NO”), as the calculated profitability level was considered, in the eyes of the decision makers, definitely not good enough and with no reasonable prospects of improvement. If the profitability potential does look promising (“MAYBE”), a go-ahead will be given for the next stage of the development program, as discussed in the next chapter. In this case, the above preliminary economic study will be also used to emphasize those cost items that are “really heavy,” for which improvements during the development program could result in a significant positive effect.

8.9 Contractual agreements Any authorization given by the corporate management for the next stage of development will probably be conditional on the finalization of two contractual agreements. First, relationships between the inventors/promoters and the implementing corporation need to be finalized at this point by a formal contract. Prior to this, an exclusive option agreement may have been signed for a Copyright © 2002 by CRC Press LLC

limited period, conditional on the decision by the corporation to build a plant, by a certain date. The details to be included in such a contract depend very much on the particular situation and cannot be discussed here, but the basic interests of each party are clear: • The inventors/promoters really want the project to succeed, and they generally believe that they can contribute to that success by having a say in any major decision making in future. They also want clear public recognition and, of course, the maximum financial remuneration possible, in relation to their optimistic profit potential. • The implementing corporation wants to secure the full cooperation and contributions of the inventors/promoters in the future, including the assignment of (present and future) patents, exclusivity on their past know-how and on their future work in this field for many years. The corporation would like all this, of course, at a minimum cost without conceding any part of its decision-making position. To assure that position, the financial remuneration may be divided into progressive installments. Once this contract is signed, the overall responsibility of the working program will be transferred to the project manager appointed by the corporation. The inventors/promoters will generally continue their contribution as consultants. Second, a suitable engineering company should be selected and engaged by a service contract that, although quite standard in nature, always includes many specific clauses (see also Chapter 10, Section 10.5). The engineering company will provide many of the professionals needed for the working program, mostly from their permanent offices, but some of them may also be delegated to the project manager’s team or to the different pilot sites. The project manager will likely chose the engineering company based on past experience and will generally include, as a basic condition, a list of eight to ten key leaders, employees of the engineering company, who will be assigned to work most of their time on the project.

8.10 Worth another thought • The only purpose of the preliminary economic analysis of the proposal is to supply information on the profit potential and to justify the continued expense for its development. • The presentation of a larger potential of profitable applications, once the first implementation is proven to be successful, could change the viewpoint of the decision makers from the short-term cash flow to a wider corporate strategy. Copyright © 2002 by CRC Press LLC

References 1. Peters, M. S. and Timmerhaus, K., Plant Design and Economics for Chemical Engineers, 4th ed., McGraw-Hill, New York, 1990. 2. Chauvet, A., Leprince, P., Barthel, Y., Raimbault, C., and Arlie, J. P., Manual of Economic Analysis of Chemical Processes, McGraw-Hill, New York, 1981.

Copyright © 2002 by CRC Press LLC

chapter 9

Working program toward a first implementation Following approval by corporate management to proceed on the basis of the preliminary process design and economic estimate (the “green light,” see Chapter 8, Section 8.8), the work program will expand considerably and will require a larger number of professionals, organized in several different groups and possibly in different locations. The core team, which will work with the project manager at the center of the campaign, will also have to be consolidated at this point. This chapter deals with those parts of the working program that can be done in parallel and handled as “separate jobs,” with proper directives, under the direct coordination of the project manager. Chapter 10 will be concerned with the consolidation of all the results from these different jobs into a single, final plant design.

9.1 Patent protection 9.1.1

Revised or additional applications

Following the experimental program and further conceptual shaping in the preliminary process design, one should ask: “if we knew at the time of the first application what we know at this present stage, how would we formulate the patent’s claims?” In many cases, analysis of the results from the experimental program may point out specific features, or particular ranges of variables, that appear at this stage to be clearly essential for successful and profitable implementation, which were not obvious from the start, even “to a person versed in the art” (in the patent’s archaic jargon). The patent should therefore be reviewed in order to determine if any of the claims included in the first application should be changed, or new claims added. The senior project staff will now examine with the patent experts whether:

Copyright © 2002 by CRC Press LLC

• Such specific novel aspects can be presented convincingly as discoveries that are essential to practical industrial implementation and, if so, • What is the best procedure and timing for preparing and submitting revised or additional applications to cover these specific novel aspects? Separate applications are often recommended to increase the legal protection from different angles, but these additional applications also require investment of time and some significant expense (see below). The conclusions of this review should probably also be relevant within the terms of the contractual agreement between the inventors/promoters and the implementing corporation (see Chapter 8, Section 8.9). These generally define the nature and extent of the intellectual rights for which exclusive licensing should be provided in the contract, including any patent assigned by the inventors to the corporation, and (most important!) the option for the corporation to award recognition for substantial contributions, by adding other “names” to the list of inventors, in the revised or additional applications.

9.1.2

Extended geographical coverage of the patents

The procedure of the international Patent Cooperation Treaty — PCT — allows a patent applied for in one location, on approval, to be filed in any one of 70 to 80 different countries with no further examination. However, the extent and locations of additional filing have to be decided within a relatively short time after the PCT approval, and filing in each country can cost up to thousands of dollars in registration, attorneys’ fees, translation, and other similar expenses. This cumulative cost is for every separate application, so that the decision to file many separate applications in many different countries can become very costly.

9.2 Detailed process design 9.2.1

Piping and Instrumentation Diagrams

The process engineering team will now prepare, with the active participation of the engineering company’s staff, “revision 0” of the P&ID drawings. When ready, these drawings will then be distributed to all members of the project team and consultants, for their review and comments. These drawings will be revised several times in the future, as more detailed information and comments become available, until they are finalized as “approved for construction.” These P&ID drawings, as can be seen in typical examples in Figures 9.1, 9.2, and 9.3, include all major and secondary equipment items, with their formal names, tag-numbers, and main specifications, as also collected in the equipment list. Copyright © 2002 by CRC Press LLC

• The major equipment items have already been defined and characterized in the preliminary process design (see Chapter 7, Section 7.8) and are further discussed in Section 9.3 below. • The secondary equipment items are those that can be selected from standard models found in supplier’s catalogs. This selection or recommendation is generally made by the sales engineers or specialists from suppliers, in response to specification sheets prepared by the engineering company. These equipment items consist mostly of pumps, fans, compressors, standard heat exchangers, solid handling equipment (conveyors, elevators), agitators, and so like. • In addition, in a new process, there may be some equipment items that cannot be attributed from the start to either group. For instance, a particular heat exchanger duty could present unusual features (such as in the flow conditions, in some safety hazards, in the possibility of fouling, or in feed-back control) and thus a thorough discussion is needed with the potential suppliers, before a choice can be made between a standard model or a special modification. Similarly, the choice of certain pumps can be critical in particular situations, if a “normal” leakage of a process stream could become a safety issue or if solids could accidentally find their way into the stream. The impeller of an agitator may have to be specially redesigned in order to avoid the creation of emulsions in certain liquid–liquid mixing operations.

9.2.1.1 Piping lists These lists include all the piping lines with their standard diameter and “schedule” (wall thickness), material of construction, and tag-numbers. A piping line is defined and tagged as it connects one piece of equipment to another. If thermal insulation of the piping line is needed, its thickness is also noted at this point. The pipe diameters are calculated from the material balances tables, including a chosen “reserve,” and from acceptable “line velocities” (which can be quite arbitrary), and rounded to the next upper standard diameter (that can also add a significant reserve!). The routine specification/sizing of piping in a conventional plant design is generally processed automatically by technicians, but for a new process where different “unknown” factors could be relevant, the process engineering staff should review these piping specifications carefully and repeatedly. This task may seem trivial, but many of the problems in start-up can generally be traced back to an error-of- judgment in piping specification. Any change in the piping in an operating plant can be very complicated. Therefore, greater care should be devoted to those “added reserves” for a new process, to allow for the possible increase in some flow-rates during the plant’s start-up, to solve unexpected problems, to arrive at process optimization, and for the eventual increase of production after de-bottlenecking. Note that the added cost to increase the size of small diameter piping is generally insignificant. Copyright © 2002 by CRC Press LLC

However, in certain cases, a larger diameter may be counterproductive. For example, in pipes handling streams that may contain solids, a decrease in the stream velocity could induce their settling and accumulation in certain parts.

9.2.1.2 Valves All the valves needed for different functions are listed and tag-numbered, each with its location, standard, material, and size. The types of functions are: • “On-off” for complete opening or closing of the flow, leaving the pipe upstream full of the process stream. Is this acceptable process-wise? Not always. • Throttling to impose a back-pressure and reduce the flow-rate. Different types are available for specific stream characteristics, i.e., the expected degree of erosion. How to choose the best model? • By-pass: a set of three “on-off” valves that allow the flow to be detoured from its normal route, through a particular piece of equipment. • Drain: on-off valve at the lowest point of a piece of equipment or a pipe. Does it open directly into the open air or into a draining pipe? Could solids clog it, and in such cases, would a flush-back washing arrangement be needed? Would there be a hazard if some leakage did occur? • Venting: on-off valve at the highest point of a piece of equipment or a pipe. Is it open directly to the atmosphere or into a venting pipe? Would a leakage present a safety hazard? • Sampling: specially designed on-off valve, to remove samples for inspection or analysis, while avoiding any dead space between the running stream and the outlet, or providing a flush-back arrangement to arrive at the current material. • Flushing: certain pipes need flushing with water to prevent accumulation of solids. In a new plant with a novel process, more valves are generally provided than for a conventional process, to allow for more flexibility, for easier inspection and cleaning (i.e., if solid precipitation could occur), and for the extensive collection of data during start-up and optimization. This is important in particular for the flushing and sampling valves.

9.2.1.3 Instruments All the instruments for local indication and for the control loops are marked and tagged on the P&ID drawings, then listed according to their expected basic process duty. Their functional scope and general policy are agreed within the core team, before transmission to the detailed engineering. In this rapidly changing technological field, the specification sheets and the final choice of these instruments have to be done by specialists, updated with the latest Copyright © 2002 by CRC Press LLC

developments, and with the feedback comments from users. This choice is generally done in personal meetings between a process engineer and an instrumentation designer from the engineering company and the sales specialist from the supplier. In a new plant with a novel process, a much larger number of local instruments are installed to allow for initial calibration, collection and checking of indicating data during the optimization stage, together with some degree of cross-checking on important points. These extra local indicators will be removed eventually, when the production is stabilized and there is no need for them anymore.

9.2.1.4 Control loops Control loops, with their standard definition and control valves, are essential to control flow-rates, liquid levels, stream temperatures, pressures, pH, or other analytical features. The further integration of these control loops into the computerized control system is discussed further in Section 9.6.4 below. The selection of the hardware equipment is similar to the previous section.

9.2.1.5 Flanged manholes and hand-holes in closed pieces of equipment These openings are needed in any plant, for inspection and cleaning of the interior or access to carry out adjustments or modifications, without having to dismantle the piping and instruments connections or the drive and upper connections. It is always advantageous to have more of these openings, but they add significantly to the installation cost, and they can eventually cause trouble, by leaking out process streams or leaking in air. In a new plant with a novel process, a much larger number of flanged manholes and hand-holes are generally designed than in a conventional process, to allow for better inspection during start-up and optimization and eventually, for handling of unexpected situations.

9.2.1.6 Provisions for possible future connections Since changes in the piping of an operating plant are always very complicated, it is a good practice to design and provide certain connection points (flanged nozzles in the equipment and flanged “tees” in the piping), so that piping additions can be hooked up with only brief interruption. In a new plant with a novel process, a much larger number of such provisions is worthwhile, to allow for more flexibility in optimization and de-bottlenecking. Their extra physical cost is very small compared to the possible gains. It is important, however, that their locations be very carefully planned according to likely scenarios. Another use for these extra nozzles in a new plant with a novel process has often been to “unplug” solid deposits in certain lines that often appear, sometimes with no clear cause, and have to be dealt with. Copyright © 2002 by CRC Press LLC

9.2.1.7 Non-conventional drives Non-conventional drives are also marked on these drawings. Although generally, the electric motor drive is not tagged separately but considered as part of the equipment, certain motors are different. The special process duty of such drives could be specified as: “variable-speed,” “high-torque,” “with feed-back control,” or “direct steam turbine,” etc. If certain electrical drives need to be connected also to an emergency electrical supply, this is emphasized on the P&ID drawings and on the lists. Lists of all equipment items, piping lines, instruments, control loops, and electric drives are prepared in a suitable format (revision 0), in addition to the P&ID drawings. These spreadsheets will be used and revised further in all the detailed engineering work.

9.2.2

Examples of portions of piping and instrumentation drawings

The following three typical examples are given to illustrate that certain very simple process concepts, which are almost taken for granted, can become quite complicated to design and operate in the plant, and require careful attention to many details to be applied successfully. Figure 9.1 illustrates a very simple statement in a process description: “the overflow (stream 6507) from stage 1 is cooled to 45°C and transferred to stage 2.” In the plant, this statement translates into three pieces of equipment and three control loops, plus pipes and valves and local instruments. Of course, most chemical engineers know this but quite a few process developers have no definite idea of the translation of such a simple statement (“the engineers will take care of that”). This means that the overflow process stream PI-6507 is collected in a buffer tank TK-1054 and pumped by P-1054 FY 09

3"-PI-6508

6'-PI-6507 FC 09

8"-V-4320 3"-N-4367

FT 09

FV 09

FEM 09 TT 08

TC 08

3"-CWS TK-1054

LC 06

Cooler E-1054 2"

LT 06 TI 07

TY 08

1"

TT 07

2"

3"-CWR TV 08

P-1054

Figure 9.1 P&ID example of overflow cooling and transfer duty.

Copyright © 2002 by CRC Press LLC

through a plate heat exchanger E-1054. The flow is measured by FEM-09 and the liquid level in TK-1054 is kept constant at a desired level by LC-06, which cascades on FC-09, which operates the control valve FV-09 on the outgoing stream to stage 2. The temperature is monitored in TK-1054 by TI07 and the final cooler temperature is measured and controlled by TC-08, through the control valve TV-08 on the return flow of the cooling water circuit. Of course, TK-1054 has to be vented and (in this particular process) kept under nitrogen blanketing. A bypass is provided for the process stream around the plate heat exchanger to be able to continue operation while maintenance operations are done in this cooler. Sampling valves and other stand-by valves are also provided. Figure 9.2 represents a portion of a P&ID for a process making pure dry hydrofluoric acid (HF), by distillation from an intermediate process stream containing HF, water, and a third component (needed to decrease the vapor pressure of the water). In principle, this is a very simple stripper/rectification column, with a reboiler, a condenser, and condensate reflux, and extensive physical data has been published on this system. The bottom stream is recycled to the process backwards. But in fact, there are quite a few complications that require experienced decisions. First, the atmospheric boiling point of HF is about 20°C. Operating the condenser HF vapors

to vent

freon vapors

TI TE

to vent

E-102

reflux

TI

PI

PI

CWS

D-101 mixed solution

freon liquid M

HPS

TE

TRC

Refrig. system HS-101 to vent

TK-105

TRC PI

LT PCV

LIC

PC

TRC

P-102

E-101

P-103

PI TE

bottom to recycle P-101

Figure 9.2 P&ID example of a distillation section for dry HF.

Copyright © 2002 by CRC Press LLC

CWR from to product storage

with cooling water would require maintaining the whole system under pressure and raising the temperature in the reboiler, thus requiring a very high steam pressure and mostly very expensive materials of construction. This was ruled out and an operating pressure around atmospheric was opted for; therefore, the condenser was designed with a dedicated mechanical refrigeration unit. The column is operated with a temperature gradient; its upper section is kept quite cool by the reflux of cold HF, which serves also as a direct contact cooling medium, as a large part of the reflux is just evaporated and returned to the condenser. This means that the column cannot be operated, or even started, without a significant amount of reflux, and therefore a stock of HF must always be kept in the receiver TK-105 to bridge temporary interruptions in operation. After longer stoppages, HF may have to be brought back from the product tank into TK-105 to restart this unit. The ultimate irony is that this plant cannot be started for the first time without buying some product from the competition! Of course, the whole unit must be close-vented and all the noncondensable gases sent back to the scrubbers operating in another section of the plant. Thus the amount of instrumentation and control shown in Figure 9.2 is in fact only the starting minimum for review, and careful designers may decide to add more means of operational flexibility and safety. These problems are typical in many cases of new process design and development. Figure 9.3 illustrates some typical complications that have to be taken into account in the P&ID for such a simple operation like a thermal evaporator for large-scale preconcentration of a relatively diluted aqueous solution going to crystallization. Significant amounts of water have to be evaporated at the lowest cost. Mechanical recompression is generally one of the best choices in connection with a falling-film evaporator operating under vacuum, which seems to be simple enough and there would be a number of specialized suppliers always eager to make an offer. In principle, the solution is circulated and distributed as a film on the inner wall of the vertical tubes; it falls down while it is heated by the condensing steam in the chest outside the tubes; part of the water is evaporated, separated in a side vessel. After that, the water vapors are compressed and sent into the chest. A centrifugal compressor is generally the best choice for such duty (compression ratio), energy-wise. But such a compressor is very sensitive to the presence of solid particles, or even liquid drops in the vapor, considering the very high shearing forces. Thus the vapors from the evaporation have to pass through a series of treatments: 1. Separation from the main concentrated liquid, which may be frothing, into a side vessel. 2. Passed through a mesh entrainment separator, equipped with a periodical washing system, actuated by a differential pressure controller. 3. Mixed with recycled hotter vapors to “dry” any possible microscopic droplets remaining, before entering the compressor. This recycle is set by a down-stream temperature controller. Copyright © 2002 by CRC Press LLC

LP steam makeup

vacuum pump M

TI

CW

TIC TE

PI

compress.

PIC PE

cold water

PDI

M

concent. soln.

hot well LE

DC

LC

condensate FR

LE

M

TI TI

LIC

TI TI

LE

feed

condensate

Figure 9.3 P&ID example of a falling-film, vapor-recompression evaporator.

4. After the compressor, the vapors are desuperheated by a spray of condensate water in excess, since for a change, such excess is not detrimental. 5. Then, the make-up stream of low-pressure steam is mixed in. There could be different control schemes, according to the characteristics of the system and the quality of the thermal insulation. The mixed vapors are distributed in the chest and flow downwards while condensing on the tubes. 6. The chest is a closed vessel and any noncondensable gases present would accumulate and prevent further condensation. Thus these gases have to be continuously removed by a side-vacuum system, under the control of a pressure-control system. The standard vacuum system requires a direct condenser stage, discharging through a “barometric leg” or “hot well” and a water-ring vacuum pump with a cold water stream. Copyright © 2002 by CRC Press LLC

7. The amount of noncondensable gases in a system under vacuum depends mostly on leaks inward of air, due to faulty installation or maintenance, and many plants experience difficulty as a result. Thus, an oversized vacuum system could help in many cases. One can therefore see that the smooth operation of such a conceptually simple evaporator can depend on many detailed issues requiring decisions, and the developers cannot just relegate these details to the expertise of the suppliers. Instead, they should at least understand exactly what is involved in the new process. We have not discussed proprietary technology that every supplier is claiming for himself such as, for instance, the exact distribution of the liquid films inside the tubes, the eventual cleaning of these tubes, various internal baffles and vapor routes, etc.

9.3 “Major” equipment packages Major equipment packages are groups of equipment items that are pivotal in the plant and should be designed or procured together, in process-compatible materials. These packages could be, for instance, a multiple-effects evaporator, a distillation section, a crystallization system, or a mixers-settlers battery for solvent extraction. The preliminary process design gave a functional analysis of the process requirements for this equipment package, as quantified in the material and heat balances, with a draft specification sheet and a list of potential suppliers (see Chapter 7, Section 7.8). Some of these suppliers have already been contacted for the FCI estimate. At this stage, more extensive discussions should be conducted by the senior process team with each of these suppliers, preferably in face-to-face meetings, if this is possible, to explain the particular needs of this project and to clarify the following aspects: • Which possible design options would these suppliers consider for this particular case and what is their preference? Each option will be detailed, with advantages and disadvantages. • What is the extent of their previous experience on similar projects? Are they willing to disclose details and allow direct contact with references? • Who are their designing experts and what is their theoretical background, in particular for designing a first-plant case? • What process data is absolutely needed for their design? • Do they have the piloting equipment and the staff to run the necessary demonstration and optimization tests, which could be used in this project at their place or which could be shipped to another pilot site? When these open discussions reach a more detailed stage, a mutually binding secrecy agreement will probably have to be signed. A good procedure with suppliers is to cross-check their claims with independent consultants. Copyright © 2002 by CRC Press LLC

After the first round of discussions, a preferred supplier will usually emerge for each major equipment package, on the basis of the confidence, the cooperation and the facilities that they can provide. For a critical major equipment package, the actual purchase cost is probably a secondary consideration of the project team, as long as it remains in the reasonable range (although this fact of life will never be admitted openly!). The proposed equipment description and the numerical data obtained from this supplier will be used to proceed with the engineering work at this stage. However, to maintain the formal procedures and to retain a fallback option (in any case), this preferred supplier will be generally included in a short-list of two or three other possibilities, which will be maintained fully in the picture until the final bid is allocated. Such “rules-of-the-game” are generally well known to everyone concerned in this field! Apart from the “packages,” the configuration of some other major equipment may have to be specially developed (or more exactly modified) and designed to obtain the particular performance duty that is specified for the new process. The eventual cooperation of a specialized supplier, who is “close enough” to the desired technology, could be advantageous but it could also raise delicate issues concerning the future exclusivity and ownership of this new know-how, once the actual plant results become available.

9.4 Pilot testing of specific process operations Pilot testing of some process operations may be required to confirm detailed quantitative specifications for particular pieces of equipment items, when these operate with the exact streams of the new process. Examples are given below.

9.4.1

Multiple-effects evaporator

The basic design of a multiple-effects evaporator for budgeting purposes could be based on bench-scale equilibrium data, but the reliable detailed design of an industrial installation will require the experimental determination of certain quantitative factors that could be still unknown, such as, for instance: • What heat transfer coefficient can be obtained with different concentrations of the solution (density, viscosity) and with different velocities in the tubes, and what will remain from the heat exchanger’s performance after a few hundred hours of operation and the resulting deposits (coating) on the heat exchange surface? • What would be the possible effect of any noncondensable gases or soluble impurities dissolved in the feed solution on the behavior of the solution boiling inside, such as frothing, splashing, precipitation, and possible encrustation of solids?

Copyright © 2002 by CRC Press LLC

• What would be the frequency, method, and ease of internal cleaning and how would that affect the average number of working hours for design? • What would be the “external” behavior of the concentrated solution, once it is removed from the evaporation conditions (depressurization, cooling)? Furthermore, larger quantities of concentrated solution may be required for testing of the downstream operations relative to this evaporator, such as a crystallizer, a flaker, a spray-dryer, and so on. These requirements necessitate the continuous operation of a pilot evaporator, equipped with all the instrumentation for collecting the necessary data, while storing the resulting concentrated solution in suitable containers. The test period would be relatively long (a few weeks, for example) and sufficient quantities of the starting solution needed, of the actual composition or as close as possible (with due reservations). Such piloting can be done in a specialized R&D institute, or in cooperation with a potential equipment supplier, who could rent a portable pilot installation and operate it.

9.4.2

Liquid–liquid contacting battery

Another example of a major package is a “multiple-stage, counter-current, liquid-liquid contacting battery” for a solvent-extraction process. For designing a “horizontal” battery of mixers–settlers, in which each stage is assumed to be practically at equilibrium, all the process aspects can be calculated reliably from the results of bench-scale equilibrium tests, such as the number of theoretical stages, the concentrations, the mass transfer rate in a mixed vessel, and the liquid–liquid separation rate (see Chapter 6, Sections 6.2.3 and 6.4.1). This choice of equipment was therefore popular for implementation of new processes (Chapter 4, Reference 13) and it would probably still be the best choice for a small number of stages (say three to five). However, when a larger number of stages is required, with larger flowrates and more costly solvents, the option of a mixers–settlers battery could present significant disadvantages as compared to a continuous vertical column-contactor, or to a set of centrifugal extractors. These disadvantages could be, for example, a bigger internal inventory of solvent, a larger horizontal area in the plant’s layout, the need for more intermediate pumps, and so on. These issues have been discussed extensively in international conferences and some recent papers relevant for industrial equipment are listed in the references.11–14 Cusack et al.11 presented the development of a new high capacity column design from an analysis of previous models (Koch). Axial mixing in large-scale packed extractors is detailed in Reference 12. Lo13 reported on an experimental comparison among three different existing models of columns, his conclusions for one particular case and the scale-up procedures to be used for an industrial project. Movsowitz et al.14 reported on a rather exceptional case in which a uranium plant in Australia worked

Copyright © 2002 by CRC Press LLC

for two years in two parallel lines, one with four mixers-settlers and the second with two Bateman columns and then it was decided to use (of course) more columns for their new expanded plant. Several suppliers offer vertical columns/contactors, each with their own proprietary design and know-how. None of these columns can be reliably sized without extensive pilot testing for each case with the actual materials, in order to determine: acceptable velocities, height of a theoretical contact stage, behavior of the phases mixture (observations), starting procedure until a steady-state operation is reached, and so on. Therefore, each supplier is organized with its own portable pilot installation and expert staff, which can be hired by a prospective client to conduct such tests with his own materials, for process demonstration and equipment sizing. In many cases, the hiring fee for the pilot is deducted from the purchase price of the industrial equipment, if a deal is reached. But for the process engineering group, the main issue before ordering such equipment for a novel process is to understand the internal mechanisms, which are generally not entirely published. For example, how the performance is scaled-up and what can be modified if the results obtained during start-up are unsatisfactory?

9.4.3

Main problems for piloting

The above typical examples emphasize the two main problems related to the piloting of specific process operations, from the point of view of the implementing corporation: • The investment in a new, owned pilot would be expensive, require in-house expertise and a relatively long time to start and, therefore, would be justified only for a long-term continuing R&D program in this particular field. Otherwise, after the conclusion of these series of tests, this pilot installation could remain unused for a long time. On the other hand, these pilot tests could be possibly done in cooperation of a pre-selected supplier, as most suppliers of specialized equipment have their own pilot installations. However, such preselection could impose many formal limitations, which should be clear and acceptable from the beginning. • The procurement of a sufficient quantity of representative feed solution may be difficult and may need to be produced in another pilot, according to the upstream operations (before the one under consideration). This condition may require a more comprehensive and lengthy program.

9.5 Modeling The methodology and technique for the development of a dynamic mathematical model that can simulate a specific process have been occupying the

Copyright © 2002 by CRC Press LLC

attention of the chemical engineering scientific community for the last two or three decades, and have evolved rapidly with the advancement of computerized resources. This is one of the most popular (fashionable?) academic fields in chemical engineering faculties and many commercial programs are also offered to the professionals. There are good textbooks and publications and there is therefore no need to recapitulate them here.1–4 However, any such model can only be as good and accurate as the numerical data in its data base and this weak point has discouraged many developers of novel processes. It is therefore important to recognize the importance and the need for such a tool, by considering it a long-term investment in the process development program. Thus, the model can be started and run first on the basis of “reasonable assumptions” and the significance of the results can be studied and understood. Then, after the first runs, a list should be prepared of certain data with significant leverage, which should preferably be confirmed and completed by additional bench-scale tests (see Section 9.6.3 below). The model will therefore be progressively improved. A dynamic mathematical model, as the quantitative basis of this specific process, will be used first for the important, but not critical, task of the engineering design of the instrumentation and control systems and for decisions concerning the volumes of buffer tanks. (This design is not critical because it is dealing with relatively wide ranges.) However, at a later stage of the plant’s design (see Chapter 10, Section 10.2), this dynamic mathematical model should be used for a critical task, in order to evaluate the consequences of any change in: the composition of the raw materials, the concentration of possible impurities, the kinetics of mass transfer, or the quality requirements from the new products. Finally, after the plant’s start-up, this model should be expanded to correlate and interpolate the operating plant’s results. This expansion will hopefully culminate in the achievement of new process know-how for the developing corporation.

9.6 Complementary bench-scale testing program Following the experimental work described in Chapter 6, which served as a basis for the preliminary process design described in Chapter 7, specific additional experimental work will generally be needed to generate some specific and important missing data. This complementary program of bench-scale tests can probably be done in the same R&D laboratories that were used before, in parallel with the other tasks discussed in this chapter. It can be divided, according to their level of urgency for the overall effort of this working program, into seven different tasks to obtain additional data needed for: • Detailed specification of the industrial equipment • Design of pilot installations or interpretation of their results • Process modeling Copyright © 2002 by CRC Press LLC

• • • •

9.6.1

Design of instrumentation The final choice of materials of construction (“corrosion”) Clarification of waste disposal issues Clarification of process safety issues

Detailed specification of the industrial equipment

Following consultations with the equipment suppliers, a detailed list will emerge of all the specific factors that may have a significant effect on the choice, sizing, design or expected performance of the different items of major equipment (see Section 9.3 above). A fact of life common to almost all projects is that, from the moment of their request, the availability of these quantitative results becomes an urgent requirement for the continuation of effective engineering work, either by the contractors participating in the bids, or by the engineering company. (This urgency will be strongly and repeatedly emphasized by the engineers.) Examples of quantitative data that may be required are: • Simple physical properties, such as the density or the viscosity of a particular stream, at a specific composition and temperature. Less frequently, more complex physical properties may be needed, such as the wetting contact angle between different phases, thermal conductivity, dielectric coefficient and other electrical properties, optical characteristics, and so on. • The density, size distribution, and characteristic shape of resulting solids (i.e., crushed rock or crystals produced), which could affect the bulk density of a loosely packed bed of such solids in a silo, or their transfer by pneumatic conveying. In addition, the shape and size of these particles can affect their flow properties (“bridging”) and their rate of dissolution or of settling. • The reaction rate and/or the mass transfer rates of certain reactions, under specific driving forces and in specific contact conditions, such as the mixing regime, differential velocity, temperature and pressure, etc. • The ion-exchange rate with specific commercial resins and process streams, in certain flow conditions. • The results from standardized technological tests, such as the grinding rate of solids, the settling rate or the filtration rate of a slurry, the compaction of a powder under pressure, etc.

9.6.2

Pilot installations

As discussed in Section 9.4 above, pilot-plant installations can be very expensive and lengthy operations, and they can “get stuck” if they are not properly designed to cover the specific function of the novel process. However, a pilot must obviously be designed without having all the necessary information. It may be worthwhile, therefore, to increase its chances of success by getting

Copyright © 2002 by CRC Press LLC

some of this information from rapid bench-scale testing, if possible before completing the design of the critical parts of the pilot (for example, the reaction rate curve in a mixed reactor). These bench-scale tests suddenly become of the highest priority on the critical path of the whole project.

9.6.3

Process modeling

As discussed in Section 9.5 above (and later in Chapter 10, Section 10.2), the methodology and technique for the development of a dynamic mathematical model, simulating the specific process under consideration, require accurate numerical data at its data base. Most of this data could probably be obtained from textbook “laws,” published scientific information, or previous test work done on this subject or on similar projects. However, some assumptions are probably also needed to close the cycle and start the model running. Thus, after the first few runs that are needed to get a good feeling of the system, certain “insecure” cause-to-effect data with significant leverage can be defined. For instance, the cause could be a change in temperature, or a change in concentration of a certain variable that could be caused by dilution, or by the ineffective dispersion of an added reactant stream. The related effect could be in the rate of precipitation, in the level of supersaturation, or in the solubility concentration. Obviously, in any real process, the number of such theoretical cause-to-effect relations could be enormous “on paper,” but fortunately, only a relatively small number of these relations would generally have the kind of leverage to justify their inclusion in the dynamic model, and these relations should be carefully selected and defined, in the limited range of practical interest. Therefore, additional bench-scale tests should be arranged to confirm and complete the data needed on these selected relations. Such tests could probably be combined with those described above in Section 9.6.1, since they are of the same general type.

9.6.4

The design of instrumentation

The choice of instrumentation hardware for chemical plants, from standard catalogue items, can be critical. Most of the modern instruments are based on physical characteristics of the stream under surveillance, such as its electrical conductance or capacitance, its magnetic density, and its optical properties under various wavelengths. Entrained impurities in the actual process stream, or even simple dissolved components such as water or atmospheric gases, may affect some of these physical properties. Fortunately, the specialized companies supplying such hardware have their own extensive data bases and their instruments cover wide ranges so that, in most cases, they are able to complete their recommendations and offers on the basis of the nominal analyses of the streams concerned, with the reservation that their final calibration needs to be done during the plant’s start-up. However, there may also be some reservation when a new process

Copyright © 2002 by CRC Press LLC

is considered and their guarantee conditional on certain assumed values of these characteristics. These companies will probably recommend performing special tests to supply such information or perhaps delivering representative samples for testing in their laboratories.

9.6.5

Corrosion tests

The tests needed to confirm the choice of materials of construction were discussed above in Chapter 3, with respect to any unknown corrosion effect on the materials of construction used for the equipment and piping. This is important first for establishing safety measures to prevent accidental failure, but also for estimating the lifetime of each piece of equipment, its supply cost and maintenance schedule, or the possible contamination of the product with metallic traces. The orderly testing of the corrosion rate is a long procedure that was (hopefully) started at the beginning of the process development, for each combination of the most probable construction materials and some typical sets of process conditions. An expert consultant with relevant industrial experience was probably engaged at that time to recommend options and procedures to arrive in time at the optimum specifications for materials of construction. But at that time, the exact process conditions for corrosion testing (compositions, chemical additions, or temperatures) were not available. Now, since the confirmation of the final choice has to be included in the purchasing specification of the equipment and the matter is again on the critical path of the project, additional tests may have to be done very urgently with the final materials and conditions. Furthermore, the public authorities and the insurance company representatives may insist on receiving written certification from an expert, at least in relation to the risks and damages that may result from a possible accidental failure.

9.6.6

Clarification of waste disposal issues

The definition and quantification of all the possible waste streams and options for their disposal within the framework of the particular region considered, are critical features of any new chemical implementation. This specialized field of activity includes various technical, commercial, and legal aspects. During the new process development stages, at least one acceptable and affordable disposal procedure has been defined for each waste stream and included in the project’s scope. But now, this proposed disposal procedure should be presented to the relevant authorities, with all the supporting data to obtain their formal authorization. Urgent tests may now be performed to produce any additional data that may still be needed for the final design of the treatment operation and the convincing evidence of the results. These tests are generally of a specialized

Copyright © 2002 by CRC Press LLC

nature and may have to be subcontracted to a suitable laboratory or to an institution with the relevant experience. For example, in one case, it was intended that most of the organic waste stream would be incorporated into a commercial cattle-feed mixture and thus had to meet certain specifications. In another case, the solid residue was intended to be integrated into a building-blocks production line and had different requirements. In still another case, the residue of microorganisms from a fermentation process was found to have a very beneficial effect on the operation of a municipal sanitary-waste bio-sludge installation. Each case is different and there are many problematic situations.

9.6.7

Clarifying process safety issues

Most chemical plants could present some form of known safety hazard, which has to be kept well under control. (See Chapter 3.) The implementation of a novel industrial chemical process can introduce an unknown safety hazard, unfamiliar to the corporation and not taken into account in its operating practice. A systematic survey and consultations with a safety expert should have been started early enough in the development program to identify such potential safety issues, and to document them in detail, in different safety manuals, for the lab, pilot operation, and plant. Relevant public regulations in the area of implementation should also have been surveyed, and this information included in the plant’s design and control, with the necessary requirements to ensure safe operation, to the best of the project manager’s judgment. Now, with the finalization of plant design and the application for the necessary permits, certain standard laboratory tests may still be needed (possibly from a statutory organization such as a standards institute) to certify certain key issues, such as for example, the flash-point for a particular mixture of organic solvents, or the handling of a certain radioactive or poisonous impurity present in any of the raw materials or fuels, which should be monitored and kept under surveillance.

9.7 Preparation of product samples for market field tests The marketing experts of the corporation generally conduct these field tests, either directly or through their regular geographical distribution channels, while the plant is being designed. The usual procedure is to contact a number of randomly chosen end users, show them the samples, the analyses, and the specification of the expected products, and ask for their comments. The feedback from such contacts should be available as soon as possible, as it could be very important for: • Confirmation of the final form designed for the products • Specification of any change needed

Copyright © 2002 by CRC Press LLC

• Confirmation of the estimated sales revenue from the products But this simple standard procedure can only start when significant quantities of representative samples of the products are made available, of the order of tens of kilograms. Thus, an important but difficult task, at this stage of the working program, is to prepare rapidly such representative sample products, without a “production line.” This problem should be emphasized to the whole team, as creative thinking and past experience are often needed, for example, to: • Integrate as much as possible of such production into the piloting or the testing programs mentioned above, such as the crystallization, evaporation, and liquid–liquid extraction. • Improvise practical batch or manual methods, with any available equipment, to prepare large quantities of starting material and to bridge over the missing intermediate operations, such as acid leaching, filtration, centrifugation, solid drying, and screening.

9.8 Clarification concerning any formal permits needed Any new plant projected will require, most probably, a number of formal permits from different public authorities, as well as comprehensive insurance policies. These permits differ from country to country, but they are concerned with different aspects of: • The plant’s construction, in relation to possible adjoining operations, local building planning, residences, roads, etc. • The plant’s operation, in particular the transportation of materials, ecology, the disposal of possible accidental leaks or gases, etc. • The safety of the plant’s personnel, including fire fighting (see References 9 and 10) • The marketing of the product, where public regulation is concerned, such as for food, pharmaceutical, animal feed, or building materials • Disposal of waste streams and possible poisonous or radioactive effects The technological background is detailed in several basic reference books.6, 7, 9, 10 Early clarification with the authorities as to proper procedure is best conducted by corporate specialists or consultants, to indicate exactly what factual information should be provided by the corporation to secure these formal permits, possibly in the form of an “environmental impact statement,” required in certain regions. Then, the project manager will determine if such data is already available in a convincing form, or if its preparation would require any additional testing or engineering studies. The preparation of such document can require a great deal of work from the professional members of the core team and their consultants, especially for a new process or product, considering the lack of exactly similar references. Copyright © 2002 by CRC Press LLC

9.9 Worth another thought • If we knew at the time of the first patent application what we know at this present stage, how would we formulate the patent’s claims? • Certain very simple process concepts that are almost taken for granted can become quite complicated to design and operate in the plant, and will require careful attention to many details, to be applied successfully. • Major equipment packages are groups of equipment items that are pivot in the plant and should be designed or procured together, in process-compatible materials. • For a critical major equipment package, the actual purchase cost is probably a secondary consideration of the project team, as long as it remains in the reasonable range. • The configuration of other major equipment may have to be specially developed (or more exactly modified) and designed to meet the specifications of the new process. • A pilot plant that is owned by the corporation is expensive, requires in-house expertise and a relatively long time to start, and therefore, would be justified only for a long-term continuing R&D program in this particular field. • Any dynamic mathematical process model can only be as good and accurate as the numerical data in its data base and this weak point has discouraged many developers of novel processes. It is important to recognize the importance of this tool and to consider it as a longterm investment in the process development program. • At a later stage, this dynamic mathematical model should be used for a critical task, in order to evaluate the consequences of any possible change in the composition of the raw materials, the concentration of possible impurities, the kinetics of mass transfer, or quality requirements for the new products.

References 1. Bequette, B. W., Process Dynamics, Modeling, Analysis, and Simulation, Prentice Hall, New York, 1997. 2. Law, A. M. and Kelton, D. M., Simulation, Modeling, and Analysis, 3rd edition, McGraw-Hill, New York, 1999. 3. Edgar, T. F. and Himmelblau, D. M., Optimization of Chemical Processes, McGraw-Hill, New York, 2000. 4. Turton, R. et al., Analysis, Synthesis, and Design of Chemical Processes, Simon & Schuster, New York, 2000. 5. Corbitt, R. A., Standard Handbook of Environmental Engineering, 2nd ed., McGraw-Hill, New York, 1998. 6. Meyers, R.A., Encyclopedia of Environmental Pollution and Cleanup, John Wiley & Sons, New York, 1998.

Copyright © 2002 by CRC Press LLC

7. Tedder, D. W. and Pohland, F. G., Eds., Emerging Technologies in Hazardous Waste Management, ACS Symp. Series, American Chemical Society, Washington, D.C., 1990. 8. U.S. Department of Health, Education, and Welfare, Air Pollution Engineering Manual, Washington, D.C., 1967. 9. Steinback, J., Safety Assessment of Chemical Processes, John Wiley & Sons, New York, 1998. 10. Kletz, T. A., Process Plants, A Handbook for Inherently Safe Design, Taylor and Francis, London, 1998. 11. Cusak, R. W., Glatz, T. J., and Holmes, T. L., The AP column, the development of a high-capacity extraction column, Trans. Int. Conf. Solvent Extraction, 1999, 427. Cox, M., Hidalgo, M., and Valiente, M., Eds., Society of Chmeical Industry, London. (Koch Process Technologies). 12. Becker, O., Lewis, C., Hardy, R., and Seiber, F., Axial mixing in large packed extractors, Trans. Int. Conf. Solvent Extraction, 1999, 475. Cox, M., Hidalgo, M., and Valiente, M., Eds., Society of Chmeical Industry, London. (Koch Glisch) 13. Lo, T. C., Process development, design and scaleup using a large Scheibel extraction column, Trans. Int. Conf. Solvent Extraction, 1999, 1503. Cox, M., Hidalgo, M., and Valiente, M., Eds., Society of Chmeical Industry, London. 14. Movsowitz, R. L., Kleinberger, R., Buchaliger, E. M., Grinbaum, B., and Hall, S., Comparison of full-scale pulsed-column versus mixer-settlers for uranium solvent extraction, Trans. Int. Conf. Solvent Extraction, 1999, 1455. Cox, M., Hidalgo, M., and Valiente, M., Eds., Society of Chmeical Industry, London. (Bateman-Israel)

Copyright © 2002 by CRC Press LLC

chapter 10

First implementation plant design: compromises and optimization The detailed engineering design of a plant generally follows well-known procedures that need not be detailed in this book. Most engineering departments and companies are well staffed and knowledgeable in that area; however, not all of them are experienced in, or even aware of, some of the specific issues involved in “first implementation of a new process.” This chapter emphasizes only the additional features that derive from the fact that the plant being designed is the first implementation of a novel process.

10.1 “First implementation” policy 10.1.1

Expected start-up problems

Any new chemical plant, even when it is based on a well-established industrial technology, involves a certain degree of uncertainty and may result in some start-up troubles. These problems may be due to failures in equipment or workmanship, dirt inside equipment or pipes, error of an inexperienced team operating under pressure, etc. Such start-up problems are expected, and they are normally corrected during the first weeks of operation of a new plant. The first implementation of a novel process can obviously also present these difficulties but, in addition, one should generally expect more serious problems that may require physical changes. The greater degree of uncertainty can be attributed to the following: • It may have been impractical to test everything in advance, for a sufficient period of time. • High expectations influenced the decision making on the project. • A few years may have passed between the final process package decisions and the plant start-up, and certain quantitative aspects may Copyright © 2002 by CRC Press LLC

have changed (i.e., in the raw materials, or in the services supplied); somewhat different factors may have been introduced, either by the team doing the detailed design or by the equipment suppliers.

10.1.2

Design policy

Thus, in the first implementation of a novel process, it would be reasonable to specify, from the beginning of the detailed engineering design of the plant, a design policy that should allow, in general, for: • An over-designed capacity of most of the individual functions. For example, the electrical motors’ drive power, or the diameter of the smaller pipes, or the capacity of solid feeders could be increased at very little cost. • A greater degree of flexibility in operation, exceeding what is generally accepted in conventional design. For example, the following typical decisions should not affect the budget much but could be appreciated in the process tune-up: installation of variable-speed drives in some of the agitators and positive-displacement pumps, larger buffer tanks between sections, and manually set variable-level overflows. • Built-in preparations for possible changes and additions of hardware. For example, more “blind flanges” installed in the piping at the correct places (avoiding dead-end traps) could allow for easier future additions. • Additional engineering effort with very careful attention to any possible cause of problems, according to a detailed list prepared and agreed in advance (see below). This is important in particular whenever wellknown, conventional tools (“old horses”) are used in new applications. For example, in the early days of industrial solvent extraction, a new plant could not be operated due to a severe emulsification, which had not been seen in the previous pilot development work. A detailed inspection showed that the impellers installed in the liquid–liquid mixers had very rough edges, which mashed the liquids at high velocity. These impellers were returned to the workshop, their edges were ground and polished and this problem disappeared in the plant. The supplier of the impellers had extensive experience with mineral plants, mixing solid–liquid slurries, and had used the same fabrication technique for this new application. After this experience it became standard procedure to include in the specifications for liquid–liquid mixers, this finishing procedure, with reminders on the drawings and in the inspector’s checklist. But this lesson should also be applied systematically to every specification for adaptation of conventional equipment to “first-time” applications. Management should accept that the implementation of such design policy may increase the final investment in the plant by about 10 to 20% (relative) over standard practice. This extra margin should be accounted for in

Copyright © 2002 by CRC Press LLC

the investment budget, but of course this reserve may be an easy target for budget cuts during construction, unless its importance is clearly understood and agreed to, as a matter of management policy.

10.1.3

Identifying probable causes of problems

The project managers have been expected to display a large degree of selfconfidence when appearing before the “higher authorities” to obtain their approval of the investment (“nothing can go wrong, everything is under control”). Then, a few weeks later, they are expected to sit with their engineering staff, make a list of every possible cause of problems (the worst case) and define in detail the features that would be needed to prevent or minimize any resulting damage. This situation is very difficult and the review is often “delayed.” This psychological pitfall has been seen over and over again in many projects in different countries and it could be very detrimental. This problematic situation can be by-passed if the role of the pessimistic “devil’s advocate” in this review is delegated in advance to an experienced external consultant, who has no psychological commitment to any previous claims and no concern that this role may damage his or her career. Identifying potential problems and ways to prevent them should have contributions from the entire team. Specific professionals will be assigned to be responsible to follow up on these the detailed design process and the purchasing specifications, and to report on any deviation.

10.1.4

“Guarantees” for reasonable plant performance

These guarantees for reasonable performance of the whole plant (or of some part) can sometimes be provided, for a price, by the engineering company and/or by the equipment suppliers. But as an insurance policy, they are generally so loaded with formal conditions and reservations that their effective coverage is practically worthless. For example, it is generally stated that the “performance test” should be carried out within a fixed period, in the exact conditions that are specified in the initial specification sheet. Such guarantees are sometimes used by a project manager in order to build confidence and possibly also to reduce personal responsibility. It is preferable that the project team ignores these guarantees on the practical working level.

10.2 Modeling and optimization The basic development of a dynamic mathematical model for the process has been discussed above (Chapter 9). As a starting exercise, it was based mostly on theoretical considerations of the chemical and physical mechanisms and on the process data available at that time. It has been used for a “first stage process simulation” in order to decide on the volumes of buffer tanks and for the basic design of the plant instrumentation and control system.

Copyright © 2002 by CRC Press LLC

As the implementation project now reaches the concrete design stage, the process team should expand the study of the different control aspects, as allowed by the limited data available. There are basic reference books that cover all the fundamentals, but refer mostly to “known and wellbehaved” processes.1–7 For a first implementation of a novel process, the design team must identify all the possible changes in the plant’s daily operation, by asking all the relevant “what if…” questions, as illustrated in the situations discussed below, and then by calculating, or at least evaluating, their resulting effects. All these possible changes are then built into this model and a comprehensive series of balance spreadsheets are run to evaluate the possible changes. The results of these simulation runs are then analyzed to check if the proposed engineering design would also cover such changes.

10.2.1

Composition of raw materials

Most raw materials used in the chemical industry are subject to a normal range of fluctuation in their compositions, which is to be expected and handled in the plant’s routine operation. For example, merchant raw materials of agricultural origin have seasonal changes, as they are stored for different periods. Raw materials coming directly from a mine or quarry depend on the geology and working program. The grade of a mineral concentrate can change with the daily operation of the plant. For example, merchant phosphate concentrates can be in the range of 28 to 32% P2O5 (mostly 29 to 31%). Although the buyers would pay only according to the “official” analysis of the material lot that they receive, they cannot dictate such analysis in advance, and they have to handle what they receive. The same issue exists for petroleum fractions from a refinery, and so on. The material balances and the operating model should be run for many different compositions of raw materials in the “reasonable” range, covering all the combinations (“ratios”) of possible extreme conditions. The effect of such different compositions on the flow rates and the compositions of the different streams should be noted and analyzed. With these results in hand, the project management team must make some hard decisions, early enough in the design work: • If they can afford to design the whole plant on the “worst possible” case within this “reasonable” range, they will change the basis of design accordingly. The result will be a larger plant, capable of producing more whenever better raw materials are available. • If such “worst possible” cases could occur “reasonably” only for relatively short periods, they could accept that the plant would operate at a lower capacity during such periods. They would slightly increase the design capacity of the plant to produce more for the rest of the year, in order to make up the final yearly production.

Copyright © 2002 by CRC Press LLC

A decision such as this can have very heavy consequences and must be reached at an early stage, even with a more rudimental model. In some cases, extensive storage and blending facilities for raw materials should be considered.

10.2.2

Effects of impurities

Many impurities enter the plant with the raw materials that were not accounted for to any extent in the basic material balances. These have generally been given little attention in the main process development. However, some of these impurities may have a very significant effect on multiple-phase systems. For example, in crystallization, chemical compounds could affect the growth of crystals by adsorbing onto the fresh crystalline surface. Impurities could cause emulsification of liquid–liquid systems, by absorbing on the interface of the smaller drops. Other impurities can find their way into the final product or increase corrosion on the materials of construction to an unacceptable level. Such effects should have been observed, identified, and possibly quantified in the previous experimental work done with “real” raw materials. Possible solutions for these problems could include the following: • Changing the raw material source to one that does not contain the impurity, if this choice is available and affordable • Adding purification steps to the process, such as an active carbon treatment or an ion exchange column (on side streams) • Deciding to “live with it,” while increasing the size of some of the equipment and of the bleed streams

10.2.3

Changes in the kinetics of mass transfer

Any contact operation involving mass transfer between different phases will be sized on the basis of the design material balance and on an average coefficient of mass transfer. This coefficient, which is based either on a definite surface basis or on a volumetric basis for a particular packing geometry, is generally determined experimentally in a pilot test, specific to the particular equipment chosen, and scaled up according to the know-how of the supplier. However, one should ask what would happen if the mass transfer coefficient actually obtained in the plant is smaller than the one taken for design, for some reason? • Can this deficiency be corrected by a change in conditions (i.e., turbulence) or covered in some other way? For example, in certain processes, the driving forces for mass transfer could possibly be increased to compensate for a lower coefficient, i.e., by a change in temperature or pH. • Is it possible to maintain at least the concentrations, at a lower rate of production? • Is there any built-in oversize reserve or the possibility of adding stages and residence time in the plant, by adding contact equipment in series? Copyright © 2002 by CRC Press LLC

10.2.4

Changes in specifications for the final product

The specifications for the final product have been the basis of the entire process development and the goal of the project preparation was to assure its quality. However, market studies have continued since (see Chapter 9) and the competition may also have been active. So it would be reasonable to assume that by the time this new product is to be distributed, there may be a demand for further improvement, for example in the maximum concentration of a particular impurity. Therefore, some typical questions could be asked at this early stage of the detailed design: • Can some options for improvement be provided now by including provisions in the present plant design; for example by changing some of the purification conditions, or by adding an extra purification stage? • Alternatively, can the physical preparation for a possible addition in the plant be provided, in case it is needed in the future?

10.2.5

Normal fluctuations around the designed average

A basic feature of the plant operation is that conditions in one stage depend on the results from the previous “upstream” stage. There can be, on the one hand, normal routine fluctuations in these results, which can be accounted for, and on the other hand, some differences that could derive from presently unknown causes. An example of so-called normal fluctuations is the gradual clogging of filtering screens, which require periodical cleaning. The liquid content remaining in a filter cake, from a continuous filter or from a filtering centrifuge, directly affects the evaporation load in the downstream dryer. If the moisture content actually obtained in the plant is higher than the average moisture content that was taken in the process design, this dryer could become a bottleneck for the whole production, unless additional features have been built in to allow the increase in its drying capacity. Consider, for instance, a cake of (water-soluble) crystals from a continuous wet screening centrifuge, with a so-called “wedge-wire” conical basket, which is treating the slurry product from an industrial crystallizer. A typical range of its moisture content could be about 3 to 5%, which means that when the screen is clean, in the first hour of operation, the cake may contain 3% moisture or less. However, any slurry feed likely contains crystals having the exact size of the slot aperture of the “wedge-wire” screen, and these get stuck and progressively clog the free draining area. Thus, the moisture content of the cake gradually increases and when it reaches (say) 5%, the feed stream is stopped, and the screen is washed with a close-circuit of hot water to dissolve all these obstructions. Then the cycle is resumed. In a large installation, where four to five such screening centrifuges are operated in parallel to process the total tonnage required, the operating Copyright © 2002 by CRC Press LLC

practice is generally to arrange a rotating cleaning schedule, where for instance, each machine is operated for 4 hours and then stopped for 1 hour for cleaning. The mixture of cakes from the four operating centrifuges should give a more or less steady average around 4% moisture and this can be taken for designing the downstream dryer. In smaller installations, there will be larger fluctuations, unless a buffer arrangement is installed on the cake stream to the dryer (i.e., wet solids blender and storage). Centrifuge suppliers are very experienced and if they are given the full picture of the upstream and downstream conditions in the tender specification, most of them will recommend an efficient rotating cleaning schedule and give very good advise in the detailed design. However, unfortunately, many cases have been seen in which these suppliers were asked in the tender to compete only on the moisture content and on the cost of one machine. It was not surprising that many offers only state that the tests show that 3% can be obtained (with probably a reservation in small letters saying “with a clean screen”). This typical and straightforward case illustrates again the need for complete presentation and expert advice. Similar situations are encountered with other process equipment that requires periodic treatments (regeneration), such as fixed-bed ion-exchangers or active-carbon columns. Other types of routine fluctuations are often created by the control loops that are programmed to start at a certain level (reading) and to stop at another level.

10.2.6

Differences in the performance of equipment

In the process design, if there are no such “normal” fluctuations, the results passed over from the upstream stage have to be assumed, either from “previous experience” or from the conclusions of the experimental tests done so far. At this point, the quantitative model should be operated to determine “what would happen if” in real life, there was a deviation from these assumptions. How would the plant react and how could these reactions be corrected with minimum damage? This systematic check could involve a lot of work and some timedelay, and thus, unfortunately, is not always done in projects. The three typical cases below may be instructive, as illustrations of this task: For example, whenever a continuous thickener or decanting centrifuge is used ahead of a pressure filter, the solids concentration in the underflow slurry obtained directly affects, with a high leverage, the filtrate load and the size of the downstream filter. It has often been seen that such solid concentration in a flocculated underflow slurry could decrease with the presence of small quantities of certain soluble impurities, with so-called “dispersing” properties (they adsorb on the surface of the particles and create repulsion forces between them). If this effect is a known possibility in the novel process, but it could not be controlled and corrected in a reliable and positive way, the filter should be over-sized (at this stage!) or an additional thickener may have to be added (possibly later!).

Copyright © 2002 by CRC Press LLC

Another field illustrating this problem relates to the multistage countercurrent decantation scheme, CCD, for washing of solids from soluble components. These CCD processes were developed long ago for the hydrometallurgy of copper and other base metals. In such applications, the ore is finely ground then treated with acid to leach out the metals and other soluble components. Then the solids must be separated from the solution, while aiming at a maximum recovery of the valuable solubles from the waste solids, and the solution is then further treated to recover the valuables. This separation cannot be practically achieved on a filter, due to the fine nature of the solids and to the large tonnage involved. It is usually done by a series of countercurrent stages, each involving a dilution and decantation in a thickener, as illustrated in Figure 10.1. The underflow stream is pumped, while the overflow stream could either be pumped or transferred by gravity if the plant’s layout allows. In the last decades, similar CCD processes were also developed and used extensively in an opposite way, for the extensive cleaning and purification of various “high-tech” minerals, which are intended to be used as fillers or for fine ceramics, or of “special” powders which are generated by chemical precipitation of very fine particles. In such cases, the process efficiency is measured by the very small amount of residual soluble impurities remaining with the solids. The efficiency of this separation, or the recovery of the solutes, is directly dependent on the ratio of the liquids contained in the underflow and overflow streams and can be calculated exactly by a rather straightforward mathematical model with a 0.0000 format. (This would be a good exercise for those chemical engineers who believe that all numbers in a chemical engislurry to CCD

washing water

washed solids

wash n

wash (n-1)

wash 2

wash 1

solution from CCD

Figure 10.1 Countercurrent decantation process for washing of solids. Copyright © 2002 by CRC Press LLC

neering calculation should have no more than three significant figures! This was not true even in those days when most calculations were done on a slide-rule and CCD calculations were done slowly by solving a set of six to seven equations with six to seven unknowns with “determinants”!). The amount of wash water (overflow) used per ton of solids is limited, and generally fixed by practical considerations, such as the dilution of the valuable in the resulting solution, and the size of the thickeners. Therefore, the separation efficiency depends critically on the solids concentration in the underflow slurry stream. Table 10.1 illustrates the relative part (in percentage) of the entering solutes that remain with the solids after four, five, or six countercurrent stages, using 3 tons of wash water per ton of solids, with different solids concentrations in underflow slurries. One can also see the very striking effects of these two factors in the plot in Figure 10.2. Table 10.1 Percent of the Inlet Solubles Remaining with the Solids Percent Solids in the Underflow Slurry

6 Stages

5 Stages

4 Stages

40 42 44 46 48 50

0.7874 0.5157 0.3365 0.2188 0.1418 0.0915

1.5873 1.1261 0.7959 0.5604 0.3932 0.2747

4.7619 3.5725 2.6719 1.9927 1.4821 1.0989

5

Perc. solutes remaining

4

3

2

1

0

35

40

45

50

Perc. solids in U 6 stages 5 stages 4 stages

Copyright © 2002 by CRC Press LLC

55

Figure 10.2 Residual solute with the slurry, as a function of the percent solids in U slurry for four, five, or six stages

In real life, the solids concentration in the underflow slurry stream could change daily in the plant according to the pH, or with the nature and dosing of a flocculating agent, or due to some excessive attrition of the slurry upstream. One can see, therefore, that even a rather simple “old” process still requires a large degree of advance care regarding the design conditions and the effective process control, starting from the identification of the critical parameter and the factors that could cause it to change, and of the means to correct such changes. In the third example, each of the exit aqueous streams from a solvent extraction plant is steam-stripped to eliminate and recover the remaining solvent “dissolved” in it. The design of the steam-stripper is therefore based on the solubility and vapor pressure data determined in the lab tests. However, there could also be some entrainment of microdroplets of the solvent phase into these aqueous streams in the plant. This effect could be due to the imperfect operation of the equipment’s settling zones, or caused by occasional impurities unaccounted for. Since such entrainment is not a “normal” constituent of the process, the plant’s designers would generally not concern themselves with this complication, unless the possibility has been specifically introduced and quantified in the process package. Then, the designers of such operations should ask themselves: • What would happen with the microdroplets of solvent in the stripper and how can one assure the standards requested for the exit streams? • Should the option be introduced in the design “just in case” for increasing (how much?) the steam rate, or the temperature, or the vacuum, with the resulting increase in equipment size? • Would it be preferable in a large installation to have a decanting supercentrifuge stand by ahead of each stripper, to catch eventual droplets?

10.3 Critical pilot testing In Chapter 9, the pilot testing of specific process operations was discussed. This piloting was required to confirm expected results and the detailed quantitative specifications for particular equipment items, operating with the exact streams of the new process, before the whole process is proposed for implementation. In addition, larger quantities of intermediate streams (or final products) were required for further testing of the downstream operations. These requirements probably necessitated the continuous operation of one or several pilots, equipped for collecting both the necessary data and the resulting streams, for relatively long test periods (for example, a few weeks each). The typical examples discussed there included a multiple-effects evaporator and a liquid–liquid contacting battery for solvent extraction. The reliable detailed design of a multiple-effects evaporator industrial installation required, for instance, experimental determination of the heat transfer coefficient for different concentrations of the solution and different Copyright © 2002 by CRC Press LLC

velocities in the tubes; the long-term performance of the heat exchangers; the behavior of the boiling solution inside, such as frothing, splashing, precipitation, and possible encrustation of solids; etc. For designing a “horizontal” battery of mixers-settlers, in which each stage is assumed to be practically at equilibrium, all the process aspects can be calculated reliably from the results of bench-scale equilibrium tests. However, if a vertical column liquid–liquid contactor is preferred, it cannot be reliably sized without performing some pilot testing for each case with the actual materials, in order to determine: the acceptable velocities, the height of a theoretical contact stage, the behavior of the phase mixture (observations), the starting procedure until a steady-state operation is reached, and so on. Before the whole process was proposed for implementation, such a pilot would have been organized with a chosen supplier, with their own portable pilot installation and expert staff, for process demonstration and agreement on the equipment sizing. Now, after the approval of the implementation and concurrently with the detailed design, some additional critical piloting may be necessary on very specific issues.

10.4 The process package The need for a process package, for the implementation of a newly developed process, may seem obvious but, in fact, has not been universally accepted. It is still not used in many projects, and its absence does create many misunderstandings and problems. The process package is the essential basis for the design of the plant. When working on a large project on many different fronts, important process items should not be decided under the day-to-day pressure of detailed engineering. All the decisions that could affect the new process operation and results should, as far as possible, be included in the process package, after they have been well thought out by all the scientists and managers who participated in the novel process development and in the project definition, and who should therefore be well aware of the possible implications of such decisions. Of course, there will still always be a need for further consultations, but on relatively few specific points. The content of a typical package, described in detail in Appendix 1, is presented mostly to give an indication of the general scope, and it should be considered as a checklist to be adapted to each specific project. Its main sections are: • Definition of the “black box” objectives, raw materials, and products • Division into functional sections, as illustrated in a block diagram, with definitions of their function, interconnecting streams, recycles, closed loops, and buffering • Separate discussions for each of the sections, with the process flowsheet, the operating variables and the design data used (sources) • Material and heat balances, with any modeling already available Copyright © 2002 by CRC Press LLC

• Major items of equipment — functions, choice, considerations, and preferences • Services required — options, sources, and costs • Materials of construction — options, “least expensive but reliable” • Safety aspects • Disposal of waste streams The first version of the process package should be prepared and reviewed by the process development group, then approved by the project manager and transmitted to the engineering company, the operating group, and anybody else involved in the project (managers, consultants). Some engineers from the process department of the engineering company may have already participated in the early preparation of the process package, as consultants or service providers. This key procedure (approval and transmission) is often called the “freezing” of the process for design. Still, this process package will be further revised two or three times during the detailed engineering work, following possible significant decisions, or necessary changes in the equipment or in the control scheme. The process package will be finally completed with the plant’s operating results in the consolidation stage (see Chapter 12). The bulk of the process package is a compilation of written material from many different sources, which can probably be found in the working files of the different people who participated in the project from the beginning. The checklist of a typical package, given in Appendix 1, could serve as a working tool for organizing the compilation of the package. Thus, most of the man-hours in the preparation of the process package are devoted to the orderly organization of all these documents according to a preset schedule, leaving clearly indicated space for whatever is still missing. Some editing will probably be needed to make the compiled document as uniform, friendly, and accessible to new users as possible. This can also be an instructive task for new members of the team. Most of this material should be prepared and reviewed in draft form well in advance, as early as possible before the detailed engineering is begun. Engineering companies usually present to the client’s project manager, in the first few weeks of the contract, all the usual engineering “design criteria” that they propose to use for the different disciplines: civil, mechanical, electrical, instrumentation, material handling, etc. The review and approval of all these books takes a lot of time and could well distract the attention of the project management from important decisions on process issues. Advance preparation therefore minimizes the delay in completion of the process package and its approval for distribution. As described further in Appendix 1, the “black box” representation defines the streams entering the plant (raw materials, streams and services from adjacent plants, chemicals and additives) and streams exiting the plant (products, waste streams, gaseous emissions). In other words, this is a definition of what is done inside the “black box,” without describing how it is done. Copyright © 2002 by CRC Press LLC

Ammonia 26,000 18% HCl soln. 390,000

WPA 50% P2O5

HCl 70,000

P2O5 27,000

Phosphate 31% P2O5- 87,000 P2O5 27,000 impurities

insolubles

CaCl2 soln. Waste

50,000

50,000

Product DAP 100,000

Figure 10.3 Black-box illustration of a process for DAP (all figures in MTY).

This “black box” is a very useful tool that allows accurate description of the nature and extent of the project to those who need not be bothered with “the technical details,” or outside people who are not allowed into the confidential aspects of the novel process. With the widening of the external front, this tool will be very much in demand from this stage on. A typical example is illustrated in Figure 10.3, for the process for DAP that was described in Chapter 5, and presented as a block diagram in Figure 5.1. Note that, despite its simplicity, this “black box” representation can also be devised to emphasize the two main “sales points” of the proposed process, which should be appreciated by people operating in this field, namely that: • Half of the P2O5 used comes from phosphate rock concentrate and hydrochloric acid, and should be much cheaper than in the concentrated WPA, which constitutes the second half. • At least some of the impurities in the WPA raw material are also eliminated, resulting in an upgraded DAP product.

10.5 The role of the engineering company in the first implementation of a novel process 10.5.1

The interests and limitations of the engineering company

The engineering company may, in certain cases, be a corporate engineering branch but in most cases is an external independent company, with satisfac-

Copyright © 2002 by CRC Press LLC

tory relevant experience and work record. The engineering company has an essential role in the implementation, and it is important that everybody on the implementing team understands exactly the objective interests and limitations of any engineering company. In most cases, a few representatives of the engineering company selected may have already participated in certain aspects of the development effort, in direct collaboration with the promoters and the corporate team, as pilot designers, consultants for equipment, or in preparing economic calculations. As these individuals contributed to reaching the implementation decision, it is quite normal that the developers and corporate teams would consider them, on the human level, as good “friends and partners,” with close working and personal relations. However, once a comprehensive contract has been signed for the supply of design and some other services (including possibly their assistance in procurement activities and construction supervision), a lot of money is now involved. Thus, any engineering company will normally behave as a separate organizational entity, with its own characteristics and legitimate interests. The extent of the services contracted from the engineering company could derive also from the availability of an experienced construction manager (for chemical plants), whose contribution can be critical to the whole project. Such individuals are very scarce, and to recruit a good construction manager for the project, who would be available for the right period, it may be well worth making exceptional organization combinations.

10.5.2

The engineering company and the project manager

The engineering company operates in direct contact with the project manager, who is the client’s representative authorized to make decisions within the contract. The private and public aspects of this relation and the attribution of responsibility are generally quite complex and require great care from both sides. There are certain characteristics in their relations that must be publicly recognized and appreciated, to avoid pitfalls. The first is obviously payment for the engineering services. Although all the financial terms of the engineering contract are supposed to be settled in advance, as with any external contractor, the contract can hardly foresee everything. There are generally more “extras” in a first implementation of a new process. There is a give-and-take situation in which the two sides are not free: the project manager has to deliver a working plant in time and the engineering company has to maintain its good name, and possibly other jobs with the corporation. But if “money” becomes a central issue in the relationship, it could affect negatively the effectiveness of the engineering function. The engineering company is expected to assist publicly the project manager and often acts as his or her representative in relation to other contractors and suppliers, mostly on technical matters, but occasionally also on financial aspects. But this is a small world, and one should recognize that the engineering company may have its own interests, from other long and complex

Copyright © 2002 by CRC Press LLC

interactions with these contractors on other projects. On the other hand, if the engineering company action for the project manager results in a better deal, this may justify their requests for “special” financial remuneration.

10.5.3

Specialization

The work of a typical engineering company is characterized by specialization in almost all its departments, which consists mostly in the adaptation of previously successful designs. This basic feature is encouraged in mechanical, civil, piping, or electrical engineering, since it should lead to “safe and sound” designs. In these disciplines, innovations are generally limited, unless the client specially requests some new feature in the design, such as an unusual “span” between columns, or pipe support, etc. The rule is “take no chances” to avoid failure and penalization. Whenever new materials, equipment, or design methods are essential to the success of the project, this requirement should be specified clearly in the contract. Even then, a formal “guarantee” will generally be obtained by the engineering company from the supplier and transmitted to the owner. All such formal blankets should be well recorded and double-checked with independent experts. In many cases, it is specified from the beginning in the contract that a well-known independent expert will give the necessary instructions on a particular issue to the engineering company (which is generally happy with such arrangement).

10.5.4

The chemical process engineering department

The chemical process department of the engineering company is generally characterized by very heavy fluctuations in workload. In a new project, a large number of specialized man-hours is invested in the first 3 to 6 months, to receive and assimilate the essentials of the new process, organize the P&ID flow-sheets, the balances, the lists, the process specifications of the main equipment, the preferred suppliers, etc. All this work is needed in a hurry and under pressure, to allow the engineering company to “deploy all its forces.” After that period, the participation of the process engineering department is reduced very much, mostly to “checking and polishing.” One should also note that on the formal side, the engineering company always disclaims responsibility for any process aspect that has not been specifically and emphatically stressed by the process developers, and included in particular in the process package. However, how could they undertake to design and build a plant without really understanding the process? Therefore, their chemical process engineering department is assigned to assimilate the essentials of the process, without spending too much time and without going in depth. As a result of these limitations, the type of chemical engineers who are genuinely interested and professionally trained in new process development are generally not induced to work for a long time in this function in an Copyright © 2002 by CRC Press LLC

engineering company, where their competence is not used all year round (unless they are in one of the few leading positions). Many of the younger process engineers spend a few years there in order to “learn the ropes,” on the way to a corporate management function. In other cases, they may feel that their creative contribution to the final result was not really appreciated by the process developers, and this feeling could lead to tension between them on the personal level. If such counterproductive tension develops, it should be recognized at an early stage by the project manager and corrected by demonstrative recognition steps, to maintain their professional stature. The active participation of these process engineers in the “critical piloting,” discussed in Section 10.3 above, could help in this matter.

10.5.5

Timetable

Another important aspect is that the engineering company is always working within a rigid and critical timetable, imposed by their formal obligations. Thus, from their point of view, decisions must be made at fixed dates, whether all the information required is available or not. Some of these decisions may involve only expenses (i.e., over-design on the safe side), while others may affect the results of the process’s operation. The intergroup relationship should be focused on getting all the relevant aspects well understood and documented at the time of the decision. But here, a conflict could derive from the project manager’s requests to work out more alternatives, more optimization and more checking procedures, according to his management judgment. Since most engineering companies are working generally for a fixed fee, such extra work is not welcome, unless the project manager has specifically requested it during the contract negotiations. These pressures could lead to shortcuts, which could be the cause of many problems that will appear later during the plant’s start-up.

10.6 Detailed engineering documents Detailed engineering work always produces a very large number of documents, specifications, drawings, tables, purchase requests, summaries of meetings, budget accounting, etc. All these documents are dated, numbered, recorded, divided into disciplines, and reviewed daily by the different members of the project manager’s team, each in his field. This working procedure is commonly used in any detailed plant design and need not be expanded on here. However, the following remarks should be noted for a new process. The main concern is that between these thousands of decisions, some features may be introduced that may adversely affect the operation of the new process. But it is practically impossible for the inventors or the process developers to review all, or even most, of these detailed engineering documents. They would not have the time, or the competence in the technical details and jargon. Their process review can only be effective on those decisions concerning the main equipment specifications and mostly on the Operational Manual.

Copyright © 2002 by CRC Press LLC

This document describes, in detail and step by step, how the plant is intended to be operated and controlled, from its empty start to its steadystate operation, through every conceivable cause of problems (each with its diagnostics and its remedy), including the occasional stoppage. This manual will include periodic jobs and routine maintenance functions. The main management issue is the time scheduled for preparation and review of the operational manual. It is generally much more convenient for the engineering company to delay the preparation of this manual until the finishing stage, after the vast majority of their work has been done. Then, their key engineers will be less loaded and the manual can refer exactly to the concrete details, as finalized in the last revision of the P&IDs. But if this manual will be reviewed with the process developers and the operating staff only at this late occasion, most of the changes that may be then requested would need serious rework and revisions, and some of these changes could be too late altogether, after the purchase of the major equipment. This situation has been often seen, leaving no choice, forcing more compromises and resulting in personal tensions and accusations. Therefore, it is advisable to provide from the start, in the engineering contract for the first implementation of a new process, that a first draft of the operational manual be prepared and transmitted with the first revision of P&IDs. This first review will give to the process developers the opportunity to stress all the important aspects, in their opinion, at that early occasion. A revision of the operational manual should then accompany every later revision of the P&IDs and should allow for continuing review. It will be the responsibility of the engineering manager and his team to see that the items that have been stressed by the process developers will be maintained throughout the detailed engineering documents.

10.7 Final review and approval for construction With the conclusion of the detailed engineering, the project manager will distribute a complete set of the final revisions of the plant’s design documents to all concerned with the final review. These include: • The P&ID drawings • The operational manual • The purchase specifications for all the major process equipment, with the definite offers negotiated • The revised investment budget The final comprehensive review often takes a few days of quite intensive meetings. As a result, the formal approval for construction is released, possibly with some modifications calling for the reworking of some items. Copyright © 2002 by CRC Press LLC

10.8 Worth another thought • In the first implementation of a novel process, a specified design policy should allow for over-designed capacity of most of the individual functions, a greater degree of flexibility in operation, built-in preparations for possible changes to or addition of hardware, and careful attention to planning for possible problems. • The task of defining possible problems and planning to minimize damage could be delegated in advance to an experienced external consultant. • Specifications for the final product have been the basis of the process development, but by the time this new product is to be distributed, there may have been demands for further improvement, for example in the maximum level of some impurity. • The “process package” is the essential basis for the design of the plant. In a large project, important process items should not be decided under day-to-day pressure. All the decisions that affect the new process operation and results should be included in the process package, after they have been well thought out by all the scientists and managers who participated in the novel process development and are well aware of the possible implications of such decisions. • The work of a typical engineering company is characterized by specialization in almost all its departments and innovations are generally limited. Whenever new materials, equipment, or design methods are essential to the success of the project, this requirement should be specified clearly in the contract. • The chemical process department of the engineering company is generally characterized by very heavy fluctuations in workload. On a new project, a large number of their specialized man-hours is invested in the first 3 to 6 months, to allow the engineering company to “deploy all its forces.” After that, their participation is reduced very much, mostly to “checking and polishing.” • The engineering company is always working within a rigid and critical timetable, imposed by formal obligations. Decisions must be made at fixed dates, whether all the information required is available or not. Some of these decisions may affect the results of the process’s operation. • In the first implementation of a new process, a first draft of the operational manual should be prepared and transmitted with the first revision of the P&ID, to give the process developers the opportunity to stress all the important aspects that should be included. A revision of the operational manual should then accompany every later revision of the P&ID and continuing review should be allowed for. Copyright © 2002 by CRC Press LLC

References 1. Seaborg, G. E. et al., Process Dynamic and Control, John Wiley & Sons, New York, 1989. 2. Stephanoloulos, G., Chemical Process Control: Introduction to Theory and Practice, Prentice-Hall, Englewood Cliffs, NJ, 1990. 3. Coughhanover, D. R., Process Dynamic Modeling and Control, 2nd ed., McGrawHill, New York, 1991. 4. Ogunnaike, B. A. and Ray, W. H., Process Dynamics, Modeling and Control, Oxford University Press, New York, 1994. 5. Luyben, W. et al., Plantwide Process Control, McGraw-Hill, New York, 1999. 6. Svrcek, W. Y, Young, B. R., and Mahoney, C. P., A Real-Time Approach to Process Control, John Wiley & Sons, New York, 2000. 7. Cluett, W. and Wang, L., From Plant Data to Process Control Design, Taylor & Francis, London, 2000.

Copyright © 2002 by CRC Press LLC

chapter 11

Running in and adjustments in the new plant 11.1 The plant construction period The plant construction period may be between 10 and 20 months, depending on the size and complexity of the plant, the site location, and conditions. There is no need to detail here the various activities concerned with construction of the new plant from the detailed engineering documents that were approved. These activities are adapted to the particular case and include the purchasing, procurement, and inspection of hardware and services, supervision of field contractors, and quality control of materials and equipment. The practice of the construction site management is dominated by the authority and experience of the construction manager and supervising team. As mentioned before, a good construction manager is always in great demand and should be “booked” well in advance for the project. In the case of a first plant for a novel process, the practice is not basically different but extra care should be taken to avoid “unseen” deviations and shortcuts from the documents that were approved for construction, which could ultimately affect the process operation. While the construction manager will be busy handling suppliers and contractors, the project manager and his team will still be fully occupied with the following tasks: • Selecting the new plant management team, within the relevant corporate organization. This is of course a critical choice. Once this team is selected and available, the project manager will introduce them into the project, while establishing a gradual program for thorough teaching of all the features of the new process and its implementation, and for handing over the plant management to them. • Assembling and training the operating personnel, together with the new plant management team (see Section 11.2 below). Copyright © 2002 by CRC Press LLC

• Completing R&D activities, and summing up and consolidating R&D reports, while maintaining links with the promoters/inventors, relevant corporate R&D managers, and any external organization concerned (i.e., funding committees). • Summing up the contractual procedure with the engineering company so far, except for their participation in the work of the construction manager’s team or in the start-up effort. • Handling the formal interactions with statutory authorities to secure the necessary permits and establish procedures. • Coordinating the work of the marketing team, which is preparing distribution channels and should be ready to sell the new products as soon as they are available.

11.2 Assembling and training the operating team 11.2.1

Recruitment

In the 6 months preceding the expected start-up date, the members of the operating team will be recruited, designated, or transferred, according to the company’s organization framework, starting with the key personnel (plant management team) and with the plant’s maintenance team. This assembling is very important, quality-wise, but different procedures may be used, depending on local conditions, in particular whether the new plant is on a new site or if it is part of a larger existing plant. Most companies insist on individual health and psychometric testing and on confidentiality contracts, and these procedures take time. For a first plant with a novel process, the previous experience and psychological profile of this personnel may be very important, in particular their patience and perseverance.

11.2.2

Maintenance

If possible, the newly assembled plant maintenance team is temporarily attached to the construction manager and put to work on the close supervision of the site contractors during the advanced stages of the installation of the equipment and the assembly of the piping. This proximity will give to the maintenance team more extensive knowledge of the hardware than can be obtained from the drawings or from verbal explanations only (“seeing is understanding”).

11.2.3

Training

An extensive training program will be conducted to familiarize the operating team with the site, the different sections of the new plant, the details of the process, the raw materials and products, and the operating procedures. The members of the process engineering team, who designed the plant, and the Copyright © 2002 by CRC Press LLC

different consulting specialists (controls, safety, quality assurance) will do most of the training, in the form of lectures and written material, and comprehensive feedback tests. The inventors/developers should also be invited, if possible, to participate in this training program, in order to emphasize the more important aspects of the process, from their point of view, and be reassured that these aspects have been well assimilated. Their participation is also important for their personal acquaintance and appreciation of the plant’s team. This personal contact will “defuse” in advance possible accusations of inadequate selection or training, which may burst out later under the pressure and excitement of the usual start-up problems.

11.2.4

Safety

The general training will include also all the safety aspects relevant to the new plant, in their basic substance as detailed in the operational manual (see below) and in the forms that may be required by the local law or by the insurance company. Generally, such training is given or supervised by a specialist, including the detailed inspection of the relevant hardware. For example, in certain locations and depending on the classification of plant, an individual examination/accreditation on safety issues may be required at the end of the training.

11.2.5

Functional organization

After the general training described above, the plant personnel will be divided into the relevant functional groups and each group will be given more detailed training on the equipment and procedures within their field of responsibility, under the direct supervision of the plant management (their “new bosses”). These functional groups would consist, for example, of the operators, the instrumentation and control technicians, the QA (quality assurance) and laboratory staff, the material handling and warehouse personnel, the office and communication team, etc. The inventors/developers will normally participate more in the training of the groups concerned with process control, laboratory analyses, data reporting, and quality assurance methodology. In particular, they will interact more with the process (or shift) engineers who will have to take the routine decisions, and should be preparing for the process optimization (see below in Section 11.6).

11.3 Preparation for start-up 11.3.1

“Dry runs”

In the last stages of the plant assembly, “dry-run” tests are conducted under the supervision of the construction manager’s staff, with mixed teams of the Copyright © 2002 by CRC Press LLC

contractors and the new plant operating staff. In the professional jargon, dry runs indicate those tests that are done before the actual raw materials or auxiliary chemicals are introduced in the plant. During the construction and installation of the equipment, close supervision and measuring were done to ensure that the work was executed properly, as designed. These dry run tests are now intended to confirm the functional aspects. The field contractors expect to get their formal “certificate of completion” according to their contract, after execution of the finishing touches. The plant’s operating team will get their first hands-on feeling of the live plant. The dry runs start generally with the internal cleaning of all the equipment and pipes, closing of all seals, checking all electrical connections, filling with water, conducting hydrostatic tests and pressure tests with water and/or air, to detect leaks. After that, water is pumped around, while commissioning proceeds of the steam and compressed air systems, heating and cooling, checking fans and blowers, burners, etc. Then, the more delicate field instruments are installed and calibrated, and all the connections to the central control room are checked for response and accuracy. The control computers are connected and their output is checked.

11.3.2

The plant manager

With the completion of the dry runs and any necessary corrections, the main responsibility for the physical plant will be formally transferred from the construction manager to the plant manager (or supervisor). With this formal change, the contractors’ staff leaves the fenced area with all their belongings, and if they are needed in the future, they will be considered “visitors” within the framework of the safety procedures. The plant’s maintenance staff will then closely supervise any finishing job required from a contractor during such visits.

11.3.3

The construction manager

After formal handing over of the management of the plant, the construction manager and construction team will remain on the site for a few more weeks, mostly in their offices, to finish the paper work, formalities and accounts with all the contractors, but they will still be available if needed for consultation and any ad hoc assistance.

11.3.4

The project manager

The project manager and his or her team will operate, from now on, only at the corporate level and the timing of their phasing out will depend on their other projects and their relative priorities. The project manager should still be in charge of the consolidation of the new know-how, as detailed in Chapter 12.

Copyright © 2002 by CRC Press LLC

11.4 Preparation with real materials After internal draining and drying all the equipment and pipes, any required internal flowing inventory is prepared and installed. For example, in a solvent extraction plant, mixing of three to five components may be necessary to make a solvent stock. In other plants, mercury is filled into electrolytic cells, or resins into columns or fluid-bed reactors, or scrubbing solutions into a flue-gas treatment unit. In addition, boiler-feed water is filled into the steam boilers circuit, soft water into the cooling water circuit, and so on, as needed. In many cases, this filling has to be accompanied by some pre-pretreatment (transfer, formulation-mixing, neutralization, filtration, ion-exchange, etc.) and this hands-on operation gives to the staff their last occasion to practice such procedures and all the relevant analytical controls, without the pressure of the running production. The tanks and silos for the raw materials and other additives are filled with real materials and these operations allow also practicing the material handling procedures, including the records, weighing, sampling and analytical control.

11.5 Strategic options for the running-in of the new plant At this point, it would be nice if the operating staff could start running the plant according to the written operating instructions and at the designed capacity, and begin making products for sale and profits for the corporation. This may well be possible in certain cases, but it cannot be taken for granted. Therefore, it would be prudent to expect that, in most cases, some problems will be encountered and to make the necessary mental and organizational preparations to handle those situations.

11.5.1

Possible causes of problems

• Errors in the detailed design or execution of certain features. These errors may occur in any plant, despite extensive checking and supervision. The effects should be apparent in a short time, but identifying the causes and making corrections may be complicated. Therefore, quite often, it is decided to delay correction until the whole picture is clarified, or “until convenient,” and this situation may cause temporary constraints and limitations in the normal operation. • Occasional human errors of the new operation staff. Experienced managers reduce these errors with additional close supervision, in the early period. • In the new process design, certain operations have been purposefully over-designed, or provided with a larger range of options for additional flexibility, due to incomplete information available at the time. For instance, variable-speed drives on the mixers, or flow Copyright © 2002 by CRC Press LLC

control on recycle loops affecting hydraulic loading or reflux. The final optimization was referred to the running-in stage and now a systematic series of runs must be done in controlled conditions, before the optimum is chosen. This procedure may delay passage into regular production. • The detailed composition of the raw materials may have changed, or at least different options may now be available, since the basic R&D was done. The process conditions must be re-adapted to the new raw materials, and this may require a series of closely controlled runs.

11.5.2

Unsatisfactory results

These problems may cause, during the start-up period, one or more of the following, hopefully temporary, unsatisfactory results: • A product of lower quality that cannot be marketed to the intended clients, who are expecting to receive the promised quality. The quality deviations could be in the chemical analysis (concentration or impurity) or in the physical characteristics, such as the particle size distribution or color. Such “off-spec” product may have to be recycled, or sold if possible as a cheaper lower quality or, in certain cases, it may have to be dumped as waste. • A lower than expected recovery, as a large part of the valuable components end up in waste streams. This may be due to incomplete reactions or transformation, or to unsatisfactory phase separations (entrainment of one phase into another stream). If the final products are of acceptable quality, production may continue, possibly at a loss, while the problem is being identified and solved. • No production at all, due to internal problems preventing normal plant operation. Examples of such problems could be severe leaks, the emulsification of one liquid into another, the encrustation of solid precipitates on heat exchanger surfaces or valves, the clogging of pipes, filters or centrifuges, or settling of coarse particles from a slurry at the bottom of a mixed tank. Obviously, this dramatic situation will call for a mobilization of everyone who could contribute to finding a solution, and executing the necessary changes.

11.5.3

Start-up strategies

One of the following start-up strategies could be tried, in order to reduce the damage caused by such problems: • Starting at a low production rate (for example 25 to 30% of the nominal) in order to make salable products, if possible. The lower production rate would increase the residence time in every operation, improve Copyright © 2002 by CRC Press LLC

the reaction yields and the phase separation results, and allow for the increase of recycle rates (in those processes in which such recycles are built in). The lower rate also allows better operator and instrument response to any unforeseen change. Then, once the plant is onstream and making “on-spec” products, the production rate could be gradually increased, for example by 10% every week under full analytical process control, while recording the systematic trends in qualities, reactions, and separations. If or when these trends appear to exceed the allowable range, a step is taken backwards into the “safe range,” and the identification and correction of the problem made while the plant is working and producing. This strategy is generally preferable and the financial penalty is limited, as probably the marketing front may not be ready to sell all the nominal production from the beginning. But it is not always possible, as certain processes cannot be operated in this way. • Allowing a lower-than-expected recovery while starting near the nominal input of raw materials, if salable “on-spec” products can be assured at a lower production rate. That scenario assumes a lower degree of completion of the reactions or a lower separation efficiency. This strategy is generally acceptable in the running in of large mineralbased plants that are using cheaper “off-the-mine” raw materials. This may not be possible for all processes, and in any case would increase the content of “valuable” or intermediate compounds in the waste streams. • In some cases, the new process can be operated with a better quality of raw materials, although this particular plant was justified on the basis of cheaper, more impure raw material. (This situation is typical for instance for processes starting with various grades of phosphate rock or concentrate.) One could start up the plant with the better, cleaner, and more expensive raw material, to avoid surprises caused by impurities in the raw material and establish that the whole plant is working satisfactorily. Then, the cheaper impure raw material could be gradually mixed in increasing proportions (20%, 40%), with full analytical process control, while recording the systematic trends in the qualities, reactions, and separations that may be caused by specific impurities. Whenever the allowable range appears to be exceeded, the identification of the problem can be focused. A step can be taken backwards into the “safe range,” and a solution devised by changing some operating conditions or adding a separation step, while the plant is working and producing. • On-off operation. If the plant can be separated into independent sections, one could work each section separately, using and filling the intermediate buffer tanks and silos, and then stop for analytical control and consultations. If an adequate product is obtained, it is transferred into the shipping storage. If the product is not satisfactory, it is recycled and the process started again. This strategy is often Copyright © 2002 by CRC Press LLC

opted for new plants based on batch operations, such as fermentation or organic syntheses. • In the worst case, if the plant gives an unsatisfactory product from the beginning, the following questions should be asked until this is corrected: • Could the “non-spec” product be recycled in the plant? • Could the lower quality product be unloaded into some other markets? • Would the intended clients wait? • How could the cash flow deficit be handled?

11.6 Stabilization of production After start up, the main short-term aim of the plant management is to stabilize the production at some level that is consistent with both the plant’s present capabilities, on one hand, and the present marketing possibilities of the products, on the other hand. As regards the production capability, in addition to the lack of experience of the operating team, some physical bottleneck may have been identified and a program devised to correct it. Until this correction is effective, the practical production capability of the plant is lower than the nominal. This limitation may not be really hurting yet, since it most likely had been anticipated in the project presentation that the full marketing volume anticipated would not be realized in the first year of operation. There is no point in producing only to fill the warehouses.

11.7 Demonstration run and project success report In order to call the whole project “a success,” the project manager must be able to report that the plant is capable of performing all its anticipated functions, in nominal quantities and quality. In addition to the personal career and status of the project manager, there may be some contractual clauses involving other parties that are dependent on this statement. An often-used solution for this situation is to organize carefully a 24hour demonstration run in which the whole plant is operated in “favorable” conditions, and everybody would be fully mobilized to assist in order to obtain a good report. Of course, all concerned are well aware that the plant does not operate in that way all year-round, but it is rightly claimed that the accumulated experience and improvements will compensate in future for the present extra attention.

11.8 Optimization of operating conditions Optimization of the operating conditions is one of the last cooperative studies performed by a group that includes process engineers from the plant, the project team, the engineering company, specialist consultants, and possibly Copyright © 2002 by CRC Press LLC

also the promoters. It will be based on the existing mathematical models and on the additional data collected during the plant’s start up and operation. The aim of this study is to prepare a working model and to leave it with the plant’s process engineers and managers, who will use it for decision making and for determining the optimum conditions needed to obtain each of the following goals: • The lowest production cost per unit of product, at different levels of overall production obtainable in the present plant or for different options of raw materials, with no restriction in supply or sales. Apart from the straight costing aspects, the different scenarios may involve different recoveries and different waste disposal costs, etc. • The maximum production rate obtainable in the present plant for different options of raw materials, and the associated costs (increasing production could reach a minimum unit cost and then rise again). • The maximum overall profit obtainable in the present plant, consistent with the other restrictions of the corporate strategy, with no restriction in supply or sales.

11.9 Worth another thought • It would be prudent to expect that some start-up problems will be encountered that may cause one of the following results: a product of lower quality, a lower recovery, or no production at all, and to make the necessary mental and organizational preparations to handle these problems. • One of the following start-up strategies could reduce the damage caused by such problems: • Starting at a low production rate in order to make salable products, if possible, then increase gradually. • Allowing a lower-than-expected recovery while starting near the nominal input of raw materials, if salable “on-spec” products can be assured at a lower production rate. • Start up the plant with better, cleaner, and more expensive raw material, to avoid problems caused by impurities and establish that the whole plant is working satisfactorily. Then, the cheaper impure raw material can be gradually mixed in increasing proportions, with full analytical process control. • On-off operation. • After start up, the plant management should stabilize production at some level, consistent with the plant’s present capabilities (some physical bottleneck may have been identified and a program devised to correct it), and with the present marketing potential of the products

Copyright © 2002 by CRC Press LLC

(it was probably anticipated that the full marketing volume would be realized in the first year). • Optimization of operating conditions should be one of the last cooperative studies performed by the plant’s process engineers, project team, engineering company, specialist consultants, and promoters. The aim of this study, based on the existing mathematical models and on the additional data collected from the plant’s operation, should be to prepare a working model, so that the plant’s process engineers and managers can use it for determining the optimum conditions and for operating decisions. • The project manager must be able to report that the plant is capable of performing its anticipated functions, in nominal quantities and quality. An often-used solution is a carefully organized 24-hour “demonstration run” in which the whole plant is operated under favorable conditions and with everybody fully mobilized to assist. It is rightly claimed that the accumulated experience and improvements will compensate in future for this extra attention.

Copyright © 2002 by CRC Press LLC

chapter 12

Consolidation of the new know-how 12.1 Updating the process know-how The new specific process know-how, which typically has been developed throughout the project outlined in this book, represents a valuable asset of the corporation that can be used again and again in different frameworks. Surprisingly enough, this fundamental fact is not always realized. In many cases, the higher levels of corporate management appear to be content that this one plant is finally working and leaves the “technical details” of how to manage this specific process know-how to the plant supervisor. Needless to say, in the first few years, the attention of this manager will be highly focused on the immediate problems in his production, product, and market segment. A few years later, when it may be necessary to improve and enlarge this plant, with probably a new managing staff in charge, this basic know-how could be missing. On the basis of past experience, it is highly recommended that all the time and resources necessary for the consolidation of the new know-how be invested during the first year after the start-up. This consolidation can be done effectively under the project manager, as his or her last task before being transferred from this function, before the members of the project team are dispersed and their attention become occupied elsewhere. This consolidation can be done in parallel by the different functional groups and then reviewed by all concerned. The documents resulting from this effort would then be entrusted to the corporate or divisional technical manager function (which may have different titles in different corporations). These documents will consist of: • • • •

Updating the process package and operational manual Analysis of feedback comments from users of the products Review of any new publications or information on the competition Review of the need for additional patent applications and initiation of controlled publication on the new process, products, and plant

Copyright © 2002 by CRC Press LLC

12.2 Final revision of the Process Package The initial process package, as described in detail in Appendix 1, was prepared and reviewed to be the essential basis for “freezing” the design of the plant. This process package was then further revised two or three times during the detailed engineering work, under the project manager’s leadership. The final revision/consolidation will incorporate all that was learned from the plant’s construction and from operation, up to a certain date. The principle of the process package remains, but it is expanded to include all decisions that could affect the process operation. Results reported should, as far as possible, be based on all the known facts and on the best analysis of all the people who were concerned with the process development, plant design, equipment design, plant operation, and maintenance. Therefore, updating the process package is necessary whenever significant additional facts and experience have been collected. The typical definition of “black-box” objectives may remain, unless some changes have been introduced in the definition of the raw materials or of the products, or in the acceptable recovery, as a result of the optimization study described in Chapter 11, or as a response to feedback from the market. The final division into functional sections may reflect the practical experience accumulated during the start-up, as embodied in a revised block diagram and in the definitions of the functional sections, interconnecting streams, recycles, closed loops, buffering, etc. The separate discussions for each of the sections may be based on the last revision of the process flow-sheet, of the practical ranges of the variables in the plant’s operating practice, and on the evaluation of the adequacy of the chosen design data, with specific recommendation for future R&D, if necessary. The updated material and heat balances, tables, and modeling calculations may now be based on new correlation from the plant’s logbooks (organizing the daily data and records). A critical analysis may be prepared of the original choices, considerations, and preferences for the major equipment and packages, of the working relations with the suppliers, and of the actual results and learning from the plant’s operation. This critical review will also extend to the issues of “materials of construction,” with the options for “least expensive but reliable” choices and the observations after the first year of service in the plant. Original considerations on issues of safety will be reviewed and the practical plant experience of the first year on the subject will also be included. The actual requirements for the different services will be updated as shown in the plant’s actual experience (options, source, cost).

12.3 Updating the Operational Manual For the first implementation of a new process, the first draft of the operational manual was prepared and reviewed with the first P&ID drawings. Further

Copyright © 2002 by CRC Press LLC

revisions of the operational manual were reviewed with successive revisions of the P&ID drawings. During the training period of the operating staff, further clarifications were probably introduced in this manual, in response to the queries by the new members of staff, each of them arriving with his own background and understanding level and representing a different point of view. During the start-up, this operational manual was put to the “acid test.” In most cases, changes have been proposed by the operating staff, some of these deriving from necessity (such as the acceptable human speed of response) and others motivated by the convenience of the people who will have to live with it, day in, day out. The impact of these proposed changes on the expected process results should be evaluated independently, then incorporated, if possible, in the final revision of the manual that will be enforced by the plant manager. These changes will also be included in the revised process know-how, which will describe in detail, step by step, how the plant has been operated and controlled, from its empty start to steady state, including the problems encountered, their diagnostics and their remedy, and the plant’s occasional stoppage. This manual will also include all the practiced safety instructions and the periodical and routine maintenance jobs. The cost of its final editing should be properly budgeted.

12.4 Feedback from users in the market The revised process know-how should also incorporate results from the first year of marketing, which could affect the process conditions and the plant’s operation. The project was started on the basis of specifications for products that were already on the market, or on changes that should be preferred by the clients, or on responses from market surveys done with samples of the intended products. As the production has stabilized and the products are actually sold, there would be a steady feedback of response from the users, which is relayed through the marketing channels. This information should be carefully analyzed and coordinated with the optimization studies described in Chapter 11. Such analyses could prompt short-term action, if needed, or define the desirable medium- to long-term trends for a follow-up program. This task should be part of the know-how consolidation program.

12.5 Additional patent applications At this point, the original patent applications on the new process have probably been granted and released for publication. The project’s team should now consider carefully if the experience of the plant’s design and operation have revealed any additional novel aspect. If this novel aspect appears to be essential or favorable to the application of the new process, even if it is only quantitative (such as some specific optimum ranges of operating conditions), it could be covered by an addiCopyright © 2002 by CRC Press LLC

tional patent application. A new application could also enlarge the circle of process inventors and could allow giving public credit to the outstanding contributions of certain participating professionals.

12.6 New publications 12.6.1

Information on the competition

As mentioned before, if there is a market demand for a particular product, there would also be, most probably, some potential competition. This competition was specifically identified as part of the initial project presentation. An information program was put in place, under the project manager, to collect all relevant details about the activities of those competitors, either from open publications (papers or patents), or through the trade “gossip” channels in the market. The distribution of the products from the new plant, and possibly also the publication of the original patents, represent step-changes that will probably cause some response from the competition, which should be monitored and evaluated. With the stabilization of the plant’s operation, this aspect should be recapitulated and analyzed, then added for future reference to the know-how consolidation program.

12.6.2

Publications on the new process and plant

With the publication of the original patent, and actual production from the new plant, the need for secrecy has changed, although there are operating details that are still kept confidential. Generally, the suppliers of the equipment packages cannot also be prevented from publicizing their contribution to the plant as a positive reference. The public image of the corporation would probably also benefit from an organized program of publications glorifying the successful pioneering effort of the new development. The public credit can also be extended to different individuals and this personal gratification would generally be very important to them.

12.7 How can this accumulated specific know-how be used again? The circumstances in which the corporation (or its global partners) could possibly profitably reuse the accumulated specific process know-how should also be analyzed and detailed in this consolidation effort. Such potential circumstances could be in one or more of the forms described below (see also Chapter 8). 1. Increasing the production volume at the same site, in order to match demand from developing markets. Expansion of 10 to 30% within the first few Copyright © 2002 by CRC Press LLC

2.

3.

4.

5.

6.

years is normally attainable with the existing resources, on the basis of this know-how consolidation. Sophisticated engineering will also be required, probably by the same company, for the de-bottlenecking of the process and the equipment, with the addition of small marginal equipment and clever utilization of the built-in reserves. Future “repeat” plant(s) in another location or country, on the basis of the operating experience learned in the first plant, as consolidated in the available package. Generally, this repeat project will not be a mere copy, but it may require some extensive design change for adaptation to the new local conditions and possibly to different raw materials and services. This design effort will be based mostly on the knowhow consolidation, but it may probably also include new “ideas” on how to improve the process and/or the product, which could not be realized in the presently operating plant. Adaptation of the novel process technology developed in this case to similar new products. For example, there is a whole range of organic carboxylic acids with similar chemical properties. The study of their use started from those that are more common and could be extracted from natural vegetation or simple fermentation. The development of biotechnology is expanding rapidly and more such acids can be produced industrially in fermentation broth. The downstream purification processes can use the same general principles, although they will have to be adapted to the reactivity of each particular compound. A group with access to a proven technology for one acid has a valuable starting base for development of similar processes. Synergetic effects between the new plant and some other existing or planned industrial facilities of the corporation, in order to use and/or upgrade the value of a by-product or waste stream. Typical examples are: • Recovery and use of acids from waste streams or gaseous effluents instead of neutralizing them • Conversion of organic wastes into animal-feed products • Utilization of gypsum waste from different processes for production of cement or other building materials • Separation of sweeteners for human consumption out of industrial molasses • Use of concentrated thermal energy, instead of dispersing it into the surroundings Synergetic effects from making use of eventual idle production capacity in certain of the corporation’s operations, available developed land, roads, warehouses and similar facilities, in packaging or utilities generation, and for exploiting the significant cost advantage of larger installations. Participation in a combined marketing effort to the same users. For example, in many markets, the users need a number of different products, and it is convenient for them to purchase them already mixed in the correct form, and this form may be also

Copyright © 2002 by CRC Press LLC

profitable for the marketing corporation. Such products could be in compound fertilizers, herbicides, fungicides, animal feed supplements, etc. If the products from the new plant are incorporated into this kind of marketing effort, this will probably require a change in some of the production steps, to adapt its final form to the combined mixed package.

12.8 A final note: what have we learned? “If we had to do it again, we would…” As has been shown in this book, the development and implementation of a new process is a rather complex endeavor, involving many people, a large number of decisions, and many unavoidable compromises. It should be quite normal, at least to the inquisitive scientific minds of the team’s leaders and eager younger professionals, that some of these decisions and compromises may be seen in retrospective as inappropriate choices in the light of their less than satisfactory results. The public, objective analysis of such “errors” could teach interesting lessons to all concerned, and prepare them for higher achievements in their next jobs. The problem is that, in the internal politics of (most?) corporations, there seems to be no place for any analysis of this type. In “real life,” one is generally expected to glorify the success and bury/forget any mistake. From the personal experience of this author, every one of these professionals would gain by summarizing at the end of each project, at least in his own papers and for his own learning and conclusions (and for those around him whom he could trust), the question: if we had to do it again, which tasks would we do differently (or at least try to)? It is hoped that the personal notes and checklists included in this book will facilitate such self-examination.

12.9 Worth another thought • Past experience indicates that it is worthwhile to invest the time and resources needed for the consolidation of the new know-how, during the first year after the start up. • During the start-up, the operational manual was put to the acid test. Changes proposed by the operating staff were derived from necessity or motivated by the convenience of the people who will have to live with it, day in, day out. Their impact on the expected process results should be evaluated independently. • The distribution of the products from the new plant and the publication of the original patents represent step-changes and the response from the competition should be monitored. Copyright © 2002 by CRC Press LLC

• The accumulated specific process know-how could be used again profitably in the following frameworks: increasing the production volume at the same site; future “repeat” plant in another location; adaptation of the novel process technology to similar new products. • Synergetic effects between the new plant and other existing or planned industrial facilities of the corporation may be used for upgrading the value of a by product or waste stream or to utilize idle production capacity and available facilities. • Every professional would gain by summarizing at the end of each project the question: if we had to do it again, what would we do differently (or at least try to)?

Copyright © 2002 by CRC Press LLC

appendix 1

Typical organization and contents of a Process Package A1.1 General This appendix is a detailed development of the concepts presented in Chapter 10. The typical process package described below should be prepared and reviewed by the process development group. After approval by the project manager, it is transmitted to the engineering company, the operating group, and to anybody else concerned, as the essential basis for the design of the plant. This procedure is often called the “freezing” of the process and detailed verbal explanations and discussions accompany it. Some engineers from the process department of the engineering company may have participated in the preparation of the process package as consultants or service providers. The principle of the process package is that, as far as possible, important process items should not be decided under the day-to-day pressure of a large project. All the decisions that could affect the process operation and results should be well thought out by all the professionals who were concerned with the process development and the project definition and who should be well aware of the possible implications of such decisions. Of course, there will always be changes and surprises and the need to consult further on specific points but, in general, with the distribution of the approved process package, the engineering design and plant preparation can proceed. The typical content of a process package described below should be considered more as a checklist, to be adapted to each project. Particular cases or local situations may be different and justify a different content, with omissions and additions. The usual procedure is that an experienced engineering company, when starting a new project, presents to the client’s project manager for approval, all their usual engineering “design criteria” for civil, mechanical, electrical, instrumentation, material handling, etc. that they propose to use in their detailed design. All these decisions may be seen as trivial to the process Copyright © 2002 by CRC Press LLC

group, but some could be very substantial budget-wise. The review and approval of all these takes a lot of time and could well distract the attention of the project management from the process issues. Therefore, it is important that the process package should be prepared and approved well in advance.

A1.2 Definition of “black box” objectives The “black box” representation (see a typical example in Figure 10.3) is a very useful tool, which defines the streams going into the plant (raw materials, streams and services from adjacent plants, chemicals and additives) and the streams exiting from the plant (products, waste streams, gaseous emissions). In other words, this is a definition of what is done inside the “black box,” without describing how it is done. Why is it important? First, the “black box” representation allows accurate description of the nature and extent of the project to people who need not be concerned with the “technical details” (such as those in higher corporate management levels: financial, purchasing, and marketing managers), or to persons who are not allowed access to the confidential aspects of the novel process (outside public or statutory authorities). In addition, this has been found to be a good introduction for personnel who will have to study the new process and work on the new plant. It gives them an overall view of what is involved and of the basic material changes, before they get too confused with the details. The definition of all the streams in and out of the “black box” should include all the significant components on a weight basis and in weight percentage. These figures derive directly from the design basis of the plant, as this has been formally defined, or on an arbitrary round basis if the design basis of the plant has not yet been finalized. This data also allows the recipients to become familiarized with the orders of magnitude and to relate to the main components and traces of impurities that could be critical. It also provides for a quick check of the overall material balance; it is surprising how many errors have been discovered at this stage, due to inconsistency in the basic assumptions made by different people at the various points of their interactions while the project was put together. The essential quality requirements for each of the product streams should be discussed in detail at this point, since this is one of the main objectives to be achieved. These requirements could be in the chemical composition (main components or maximum levels of specific impurities) or physical properties (color of solution, crystal size, solid microstructure), or even the final packaging. The importance of each of these items for marketing and sales should be emphasized and understood. The possible options and variations (random or seasonal) in the source of each of the raw materials and of any fuel used should also be discussed, to explain or justify the choice made in the basis of design. In many industrial cases, important parameters in the raw materials composition change in the two to three years that may have passed between the process freeze and the plant start-up. Copyright © 2002 by CRC Press LLC

The detailed description of each of the waste streams (if any), as it leaves the plant, after any compulsory waste treatment included in the plant’s scope, should be an important part of the process package. The temporary level of objectionable impurities in such streams may possibly jump by orders of magnitude due to operational errors, and such fluctuations should be evaluated and taken into account. There should be at least one acceptable form of disposal of each of the waste streams. If there are several disposal options, the final choice is a matter to be worked out by the project team during the detailed engineering phase, in relation to the local conditions and regulations and to the associated costs.

A1.3 Division of the process into sections as illustrated in a block diagram The next step in the preparation of the process package is the division of the “black box” into functional sections, connected by numbered streams. These different sections and interconnecting streams can be usefully represented in a block diagram (see typical example in Figure 5.1). Each section is also given a formal name, defining its prevailing chemical mechanism (i.e., gas–liquid reaction, liquid–liquid extraction, evaporation, crystallization, drying, etc.). These formal names will be used in all the project documents and later by the plant staff for many years, so one should think carefully before freezing them. The exact definition of each section is important for efficient process design, but this can be complex, as there may be different acceptable divisions. In any case, a careful analysis is needed to arrive at a reasonable number of sections, as many people may have to work with these definitions for many years. In this context, a section is a definite part of the process in which the flow-rates and compositions of the exit streams are determined uniquely by: • The flow-rates and compositions of the entering streams • The operating conditions controlled by the operator (i.e., the temperature, pressure, residence time, reflux ratio, circulation velocities, etc.) Process streams do not always pass from one section to another in a forward direction only. In many equilibrium-controlled processes, there are great advantages to recycling some streams in the backward direction, and sometimes this is an absolute necessity. A well-known example of this principle is the reflux stream from the condenser to the top of the rectification/distillation column, but the same principle can be applied effectively to most equilibrium-controlled processes. The exact return point of each recycle stream could be critical and could determine whether there needs to be division into more sections. In certain other processes, the “black box” does include an internal stream circulating between the different sections, which hardly gets outside,

Copyright © 2002 by CRC Press LLC

except for unwanted losses. This is typical, for example, in solvent extraction processes, in which a relatively large solvent stream circulates in a closed loop. Other examples are the mercury loop in a chlorine-soda plant and the mother-liquor loop in a salt purification plant. A buffering tank volume may be needed for averaging the fluctuations of certain streams passing from one section to another. In semibatch processes, which, despite their old-fashioned connotations, are still necessary and useful in specific cases, some of the sections fluctuate on-off and require buffering “before and after,” so that the other sections can be operated continuously, in more or less steady state. In some other processes, the composition of a raw material may fluctuate and despite all process control efforts, the output streams from certain sections receiving such raw material need to be blended and averaged before proceeding. All these aspects should be discussed explicitly at this stage of the process package preparation, in connection to the block diagram, to bridge between the theoretical steady-state ideal and the real-life necessities.

A1.4 Separate discussions for each section 4.1 There should be a detailed description of the prevailing conditions, successive and overall chemical reactions, and the physical changes or separations obtained, with particular emphasis on the critical items. In this description, it is important to convey exactly: • How each of these reactions or separations is controlled • Which are the controlling parameters (for example, temperature, pressure, reactant ratio, velocity, residence time) • How a change in the magnitude of any of the controlling parameters would affect the final result 4.2 A process flow-sheet drawing, which is the most recent “frozen” revision of the preliminary flow-sheet discussed in Chapter 6 , with all the numbered equipment, pipes, valves, and instruments essential for the understanding of the operation of the section. This process flow-sheet drawing is the translation of the process block diagram (referred to above) and the chemical mechanisms concepts into the usual chemical engineering methodology and into the chemical plant’s practice. This document should be clear also to the “nonchemical” engineers and technicians who will be working on the detailed design and plant construction and should include also the systematic tag-numbering of all pieces of equipment and all main streams, which will then be used as references for all future work. 4.3 A list of all the operating variables, which can be controlled by the operator in order to obtain the desired results. 4.4 Any design data available, from tests, publications, or internal correlation, on the process behavior or equipment operation. This design data should also include the physical properties of each stream Copyright © 2002 by CRC Press LLC

(specific gravity, specific heat, viscosity, vapor pressure) in the specific operating conditions and the kinetics of the reaction as a function of the operating variables.

A1.5 Material and heat balances 5.1 Following the discussion of the process design for all the sections, any modeling that may have already been done should be presented and discussed, together with all assumptions and data sources. Ideally, a computer model is available and can be readily used to deliver reliable tabulated material and heat balances. But generally for a new process under development as discussed above, the preparation of such a model is generally not of the highest priority and it may be available only if, by chance, a modeling specialist is on the team. 5.2 Preliminary material and heat balances of all the significant components in all the streams should be prepared on spreadsheets by the usual trial-and-error methods, and included in the process package (together with all assumptions and data sources), clearly marked as “preliminary” and placed “on hold” (in the professional jargon). It is very important to emphasize that, due to the possible gap in the transfer of the project’s working leadership from the promoters to the engineering company, the inclusion of such numerical tables in the process package should be considered for illustration purposes mainly. It will be the responsibility of the engineering company, and one of their first tasks, to check and confirm the numerical accuracy of these tables before making further use of them. The process modeling effort, which is no longer on the critical path, should then be continued either by the engineering company (if it is within the scope of their contract), or by a specialist consultant. The resulting comprehensive model will be used later in the final optimization of the process operation.

A1.6 Equipment choices 6.1 A preliminary list of all major pieces of equipment, with their tagnumber, formal name, budget installed cost, and possible suppliers, should be prepared on a suitable spreadsheet and included in the process package (see typical example in Table 7.4). This preliminary equipment list is a very important working tool, which is started by the development team on the basis of the different sections of the process flow-sheet, but which will have to be worked out by the engineering company in many formats for many purposes: • For different types of equipment • For different suppliers Copyright © 2002 by CRC Press LLC

For different geographical areas in the plant For piping connections For different materials of construction As a basis for the investment cost sheet To sum-up the electrical drives, etc. Note in this regard that a separate list should be started for those electrical consumers that need to be connected to the electrical emergency supply, as this information becomes available. The preliminary selection of the type, model, and sizing of each major piece of equipment should be presented in the process package, based on a functional analysis as quantified in the average material balances. A specification sheet would be opened for each major piece of equipment, in which this selection should be recorded, together with all the facts related to its function, a detailed explanation of all the possible options for its type, model, size, and any other important specification item, for the selection of the materials of construction, and the estimated electrical requirement, and the reasons for the recommended choice. A preliminary list of potential suppliers for the recommended equipment type. This list is by no means exhaustive at this stage. In certain cases, when only one preferred supplier can be recommended for a major piece of equipment, the situation will be simplified but also more complicated: this creates a critical dependence and many corporations are opposed to such a situation, as a matter of principle. In certain cases, the new process may require the development and design of a new or modified type of reactor or separator that cannot be procured readily from an established supplier. This requirement has been identified before, but has constituted an additional load on the development effort and has possibly already been dealt in a pilot program. In this case, a large part of the process package should be devoted to the analysis of: • The function of the new equipment • The pilot results • The design principles • The sizing calculations • The exact recommendations for the final industrial design As an additional result of this presentation of the major equipment, the plant space needed for the recommended choices (area and height) can be indicated for the preliminary layout studies.

• • • • •

6.2

6.3

6.4

6.5

A1.7 Services 7.1 These services are essential and major cost factors, although they are often considered by the R&D scientists as trivial. The options available for each service are not basically different for a new process Copyright © 2002 by CRC Press LLC

than for a conventional one. However, the choices and the options are much wider before the “freezing” of the novel process and/or of the implementation site. Those generally needed in most chemical plants are: • Electricity for drives and sometimes also for heating • Cooling water • Saturated steam at several pressures (live or condensing) • Compressed air • Fuel of different kinds • Occasionally, heating oil, nitrogen and/or oxygen are also needed Optimization studies to achieve the cheapest, most convenient solution could make a decisive difference for the economic value of the new process. Often the development team is able to start such studies but not complete them, as the economic factors have not been clarified before the process package is presented. If there still are attractive prospects, these should be described clearly in the process package. 7.2 The nominal consumption rate of each service needed for steady-state operation can be calculated directly from the material and heat balances presented above, and from the equipment list in the previous section for the electrical power. Those average consumption rates are used for the (annual-basis) economic calculations, but higher design quantities should be provided to cover the instantaneous rates (i.e., for starting or stopping, or for emergencies). The process package can only provide general guidance on these design quantities, and they may only be finalized after all the detailed information is obtained from the various equipment suppliers. So, one of the first assumptions (placed on “hold”) in the detailed process engineering would relate the maximum delivery rate needed for each service. These assumptions are generally based on past experience and the intuition of the leading process engineer, but they should be confirmed as soon of possible so that the services supply can be finalized. 7.3 Fuel could be needed either for direct use in a combustion device incorporated into the process, or for the dedicated production of steam or other heating medium in the new plant. Several types and qualities of fuel can be considered, including coal, liquid petroleum fractions, or natural methane gas. In addition to the obvious considerations of delivery cost and convenience, a decisive factor in the choice of fuel will be the impurities in the flue gases discharged from the stack (SO2/SO3, nitric oxides, metallic dust, and so on) and/or of fly-ashes. If local ecological restrictions require intensive cleaning installations, this may cancel the advantages of a cheaper fuel. 7.4 Condensable saturated steam (at different pressures), or another heating medium (oil) is used in heat exchangers. In some cases, it may be purchased from the site’s central services, or from an adjacent producer. If not, a steam system should be installed with all the ancillaries, such as the production of boiler-feed water. In many cases, Copyright © 2002 by CRC Press LLC

7.5

7.6

7.7

7.8

7.9

when large quantities of lower temperature heating are needed, there could be a decisive advantage in a synergetic combination with a plant that has a large excess of waste heat. This is of course in addition to any possible internal saving, for example by the use of multipleeffects evaporators or vapor recompression. Cooling water is generally produced in the new plant’s own cooling tower. The minimum supply temperature (usually of the order of 26 to 30°C) depends on the climatic conditions of the location, and it could be an important limitation in the basic design of the process. For example, in many processes that include large-scale evaporation and condensation under vacuum, if the cooling water is not “cold” enough, it becomes necessary to use much more expensive, artificially chilled water, and this can make a significant difference. It may sometimes be received from a nearby sea or river at a lower temperature and this can be an important asset. Electricity is generally purchased, unless the new demand will be large enough to justify the purchase of a generator. Even then, a back-up connection to the external grid will generally be needed. Some processes also require an emergency source of electric power for safety or damage control, as mentioned in Section A1.9 below, on safety issues. Process water is used in all plants in relatively small quantities and in many different specifications (quality, purity). Generally this supply is not a significant consideration, but it could become very significant in certain hydrometallurgical or mineral projects, and in desert areas. In most plants, a fire-fighting water supply and rig must also be supplied from a reliable source, with an adequate back-up to meet possible emergencies. Compressed air is generally produced in the new plant at different supply pressures. A small quantity is handled separately for pneumatic control at assured pressures, but the larger quantities are generally at a lower supply pressure for large aerobic fermentors, air mixing of pulps, direct-contact drying in closed vessels, etc. In certain processes, it could be a major consumption and production cost. Oxygen and/or nitrogen as inert gas, if needed, can be purchased in certain cases, or produced on site by an air separation installation.

A1.8 Materials of construction: options and preferences The choice of the materials of construction that are in direct contact with each of the process streams, which must resist any corrosion or erosion action, can be critical to assure a long plant life. A large choice of sophisticated materials is now available, i.e., different kinds of metallic alloys, polymers, glass, ceramics, refractory bricks, etc. One of the main considerations is that some of these materials are quite expensive and involve a large investment. Options for each of the streams should be indicated in the process package, together with any relevant factual information (previous experience, tests, and Copyright © 2002 by CRC Press LLC

expert’s recommendations). However, the choice of the “least-expensive-butreliable” option should be an essential part of the project manager’s responsibility and this should be indicated clearly in the process package, even in those cases when it may be considered trivial and well known. Should there be any doubt on choice of construction material for a particular stream at the time of reviewing the process package, this reservation should be indicated (“hold”). Thus, the engineering company will not make any binding commitment on this item, until it is further clarified and confirmed with experts or by corrosion tests, and the “hold” removed by the process manager.

A1.9 Safety aspects Those who work in the chemical industry routinely encounter potential safety hazards, including fire, explosion, burning, poisoning, radioactivity, thermal or visual radiation, and air or water contamination. Nevertheless, it is the responsibility of the process developers to indicate clearly in the process package if there could be new or unusual hazards in the novel process. They should also provide any available data relevant to the evaluation of the extent of the known safety hazards, such as data on the ignition point, flash point, explosive ratio, volatile components or gaseous emissions, poisonous or carcinogenic effect on humans, for different streams. Engineering companies that design chemical plants are generally experienced and have their own experts in this field. The design and specification of the provisions needed for preventing such hazards (whether “conventional” or emphasized in the process package) in the new plant, and eventually for controlling them, is a definite part of the detailed engineering work. This work may be guided by specialist consultants, within the framework of external statutory regulations and insurance requirements. This specification is also generally linked to the ecological permitting procedures in effect in the particular area.

Copyright © 2002 by CRC Press LLC

appendix 2

Functional organization structure of a typical development project A2.1 Successive stages The functional organization structure of a typical process development project is critically important, but is also complex and constantly evolving, depending on local and personal circumstances. For that reason, this organization has been, more often than not, left to the “play of forces” or to the wisdom of the manager in charge, and the results may or may not have been effective. This appendix considers the objective demands on the organization, in order to increase its chances of success. The resources discussed in Chapters 2 and 3 are organized here in a more conventional hierarchic structure, in which every leader is responsible for four to six functions. In this regard, one should recognize the basic differences in the three main stages of the project, as these relate to the demands on the organization. 1.1 The invention and promotion stage, which may have one of two contexts, depending on whether: • It is done inside the implementing corporation, in its R&D department or “new business” department, on the basis of the corporation’s already established position in the field and of its accumulated know-how • It is done by external promoters, prompted by published information about the potential need for such process, and possibly also by their desire to promote sales of new technology, equipment, or services 1.2 The process development stage, with the financial support of a corporation and under a designated project manager, until a decision is reached to build a plant Copyright © 2002 by CRC Press LLC

1.3 The construction and running-in of the plant, under the project manager, until the responsibility is transferred to the plant manager

A2.2 The invention and promotion stage See Figure A2.1. 2.1 The inventors (there are usually two or three co-inventors) or corporate R&D scientists have generally limited executive resources of their own, and they typically form collaborative links with promoters, which can take different organizational forms, depending on local conditions and personalities. From the working organization’s point of view at this stage, the inventors should be controlling the following functions: the literature and patent search, the process engineers, and the laboratory feasibility tests. They should also be in working contact with the patent attorney. 2.2 The promoters, or the corporate managers in charge of R&D or “new business” have the function of defining a favorable implementation

literature and patent search

inventors

promoters

process engineering

consulting and costing engineering

feasibility lab. tests

business consultant

patent attorney

lawyers

negotiation and agreement with corporation

Figure A2.1 Invention and promotion stage.

Copyright © 2002 by CRC Press LLC

2.3

2.4

2.5

2.6

2.7

scheme, contacting the corporation at the suitable level, promoting and negotiating an agreement concerning the development program. They have to prepare the proposal in a presentable and attractive form, on the basis of the information relayed by the inventors. They will be using for that purpose specialist consultants, costing engineers, business advisers, agents, and lawyers. At this stage, an exploratory literature and patent search begins using material already on file, probably performed by the inventors’ assistants in an academic library and on internet data bases, with trialand-error and direct feedback. This exploratory work could also be subcontracted, i.e., to graduate students. Many patent attorneys are also organized to supply such services at affordable rates. In any case, the analysis of the results from this search will require the personal attention of the inventors. Preliminary feasibility tests are very important and are generally done under the direct supervision of the inventors, first, to reassure the inventors that they are working on a reasonably firm basis, then to supply concrete exploratory results to be compared to the expected/predicted results. In addition these preliminary tests will provide observations on the behavior of the reacting and resulting phases. In many cases, such tests may have to repeated later, as a demonstration to the delegates of the prospective implementing corporation, so that special attention is often devoted to obtain a show-case impressive format. Engineering consultants advise the inventors and promoters of what can and cannot be done in the way of implementing the new process. This input can have a critical effect on the focusing of the basic features of the new process. Then, the engineering consultants will prepare the preliminary process definition and flow-sheet, the basic balances and cost estimates, and will present these in a preliminary engineering report, in an acceptable and “friendly” format, which will be used by the promoters in their future presentations. The patent attorney is either a free practitioner or a full-time employee of the organization connected to the inventors/promoters. At this stage, he or she will advise the inventors and promoters about the procedure and the wording of the first patent application, and file it with the patent office. In later stages of the project, the patent attorney will help to formulate the scientific aspects of the claims. There are business consultants who specialize in selling and buying industrial intellectual properties. They advise the inventors and promoters about the acceptable procedure and the criteria for selecting and contacting a prospective corporation, which may be interested in the new development. In many cases, they also provide personal introductions from their previous records. They will also prepare a preliminary economic and market analysis, in a conventional format, and (hopefully) with attractive bottom lines, which shall be used by the promoters in their presentations.

Copyright © 2002 by CRC Press LLC

A2.3 The process development stage See Figure A2.2. 3.1 The project manager is nominated by, and reports to, the corporation’s relevant manager. Apart from his executive assistants, he should in direct working contact with the inventors/promoters, and in direct control of the leaders of the following functions : • Engineering deputy • Senior process engineer • Know-how management • Marketing specialist • Coordination with site management 3.2 The inventors/promoters continue to participate in the project’s core team on a consulting basis, making their basic know-how available, mainly for the experimental program. Their main daily contacts are with the senior process engineer and with the “know-how management” function (see below). 3.3 The engineering deputy of the project manager (in fact “number two” on the team) manages the following functions (see Figure A2.3): • Cost engineer, in direct daily contact, who also handles the engineering files and contacts with outside suppliers corporation management and services

project manager

inventors/ promoters

senior process engineer

engineering deputy

see Figure A2.3

see Figure A2.4

know-how management

see Figure A2.5

coordination with site management

Figure A2.2 Process development stage I.

Copyright © 2002 by CRC Press LLC

marketing specialist

project manager

engineering deputy

equipment designers/ suppliers

cost engineer

plant operation specialist

plant safety expert

economic studies and analysis

market specialist

Figure A2.3 Process development stage II.

• Plant operation specialist, generally a part-time job, who is coordinating all issues related to the future plant’s operation, staffing, and safety procedure, and should be coordinating efforts to find acceptable solutions • Plant safety expert, also a part-time job, possibly an external consultant • Economic studies and analysis, possibly with the participation of another corporate department and/or of an engineering company and based on the input of the market specialist • Designers and/or suppliers of major equipment, who provide the necessary information before any formal bidding, and participate in piloting. 3.4 The senior process engineer, working in direct contact with the project manager and the inventors/promoters, and managing the following functions (see Figure A2.4): • Process engineers — a number of full-time process engineers who prepare the process flow-sheets, studies of alternative options, balance spreadsheets, equipment specifications, correlation of experimental data, etc. Copyright © 2002 by CRC Press LLC

project manager

inventors/ promoters

senior process engineer

process modeling specialist

process engineers

R&D laboratories

pilot installations

corrosion and material specialist

Figure A2.4 Process development stage III.

• Process modeling specialist — who prepares and updates the mathematical model and analyzes the results of various runs. • R&D laboratories — the senior process engineer also manages contracts with the R&D laboratories and coordinates the experimental work ordered, by supplying detailed instructions, materials, and additional personnel as needed, in addition to reviewing the reports and approving the accounts. • Pilot installations — a similar organization relates to contracts for work ordered with external pilot installations, except that the process engineers on the project’s team are likely to participate personally and closely in these tests and in the analyses of the results. • Corrosion and materials specialist — who coordinates the testing and collection of information required for determining the construction materials to be used. 3.5 Know-how management (see Figure A2.5): most large corporations have specialists who can fill this function on a part-time basis. In less complicated cases, this function may be filled by the inventors or by the senior process engineer. It involves a considerable amount of paperwork for the orderly management and up-dating of the intellectual propriety, in direct contact with the relevant patent attorney and publication search specialist. Copyright © 2002 by CRC Press LLC

project manager

inventors/ promoters

know-how management

patent attorney

publication search specialist

Figure A2.5 Know-how management.

3.6 The marketing specialists of the corporation who conducted the field tests (see Chapter 9) are expected to report back as soon as possible with findings and recommendations concerning the details of the products that should preferably be changed, or adapted, in order to get a better sales return or market share. Such feedback is also required for the final economic studies, to confirm the estimated sales revenue from the products. 3.7 Coordination with site management: if the new plant is erected within a larger industrial site owned by the corporation, there should be a lot of coordination with the site management. Generally, this is time well spent, as the help obtained is of great practical value.

A2.4 The construction and running-in period All the functions described above in Section A2.3 are continuing during this period, under the project manager, with some changes in emphasis and with the following additions (see Figure A2.6): 4.1 The engineering company staff, which is doing the detailed plant design and issuing drawings and specifications for approval by the project manager or his delegate. This staff is generally drawn from different departments in the engineering company. 4.2 The construction manager and his staff manages all the activities related to procurement, construction, equipment erection, etc., and all the actual contractors on site. While formally under the supervision of the project manager, the construction manager has generally a wide operating authority to organize the work on the site. 4.3. The new plant manager and operating staff, who are trained to receive and operate the plant in coordination with the existing site management (see Chapter 11). Copyright © 2002 by CRC Press LLC

corporation management and services

project manager

site management

senior process engineer

new plant manager

engineering company

new plant staff

construction manager

Figure A2.6 Plant construction and running-in stage.

4.4. On the project team, the following tasks are emphasized in this period: • The process engineers work mainly on checking and coordination of the design and assisting in training and preparation for start-up. • Complementary R&D tests may be needed to ascertain certain details of the design. • Cost engineering personnel are still fully occupied with economic studies and analysis of alternatives (this line of work seems to be never ending). • Financial coordination and control need to be maintained, according to the established corporate criteria, with the relevant departments of the owning corporation. This becomes very important and time-consuming, with the total amount of money spent.

Copyright © 2002 by CRC Press LLC