Oil extraction and analysis: critical issues and comparative studies [1 ed.] 1893997782, 9781893997783

Citation preview

Oil Extraction and Analysis Critical Issues and Comparative Studies

Editor D.L. Luthria U.S. Department of Agriculture Agricultural Research Service Beltsville, Maryland

PRESS Champaign, Illinois

Copyright © 2004 AOCS Press

AOCS Mission Statement To be the global forum for professionals interested in lipids and related materials through the exchange of ideas, information science, and technology. AOCS Books and Special Publications Committee M. Mossoba, chairperson, U.S. Food and Drug Administration, College Park, Maryland R. Adlof, USDA, ARS, NCAUR, Peoria, Illinois J. Endres, The Endres Group, Fort Wayne, Indiana T. Foglia, USDA, ARS, ERRC, Wyndmoor, Pennsylvania L. Johnson, Iowa State University, Ames, Iowa H. Knapp, Deaconess Billings Clinic, Billings, Montana A. Sinclair, RMIT University, Melbourne, Victoria, Australia P. White, Iowa State University, Ames, Iowa R. Wilson, USDA, REE, ARS, NPS, CPPVS, Beltsville, Maryland Copyright © 2004 by AOCS Press. All rights reserved. No part of this book may be reproduced or transmitted in any form or by any means without written permission of the publisher. The paper used in this book is acid-free and falls within the guidelines established to ensure permanence and durability.

Library of Congress Cataloging-in-Publication Data Oil extraction and analysis : critical issues and comparative studies / editor, D.L. Luthria p. cm. Includes bibliographical references and index. ISBN 1-893997-78-2 (acid-free paper) 1. Oils and fats, Edible--Analysis. 2. Food--Fat content--Analysis. I. Luthria, D. L. (Devanand L.) TP671.O45 2004 664′.3--dc22 2004005649 CIP

Printed in the United States of America. 08 07 06 05 04 5 4 3 2 1

Copyright © 2004 AOCS Press

Preface

During the past two decades, there has been remarkable advancement in the field of biotechnology. The first phase of genetic engineering in crops was centered on agronomic traits, whereas the current trend focuses on value-added traits such as oil or protein content, modifications, or yield enhancement. This biotechnology advancement in conjunction with the globalization in trade has resulted in the development of new opportunities and challenges for the industry and society. Appropriate valuation and differentiation of these value-added quality products around the globe pose a major challenge faced by large number of industries and other grading organizations in different regions of the world. This is caused by differences in the technologies and procedures approved by various official agencies for the assay of value-added traits. Accurate determination and proper assessment of value-enhanced products are critical for the success of the biotechnology industry in the global market place. There is a crucial need for harmonization of assay procedures among different official agencies around the globe. This books attempts to address these issues by using crude fat as an example of how this approach could be extended to other valueadded products. The topic of accurate determination of oil content in oil seeds is of significant interest to the members of the AOCS and is strongly supported at the organization level. A symposium entitled “Critical Issues, Current and Emerging Technologies for Determination of Crude Fat Content in Food, Feed and Seeds” was held at the AOCS Annual Meeting in Kansas City, MO in May 2003. This book contains represented papers from this symposium. The book is divided into five sections: Section 1 deals with the economic significance of accurate determination of crude fat and the need for harmonization of procedures among different official agencies around the world. Section 2 describes in detail the different extraction technologies and their principles that have been used for crude fat content determination. These technologies can also be extended to other products. Section 3 provides a comparison of different primary extraction technologies and identifies the importance of sample preparation and issues related to crude fat analysis. Sections 4 and 5 depict current and emerging secondary rapid nondestructive technologies (e.g., NIR and NMR) used for crude fat determination. The topics covered give a broad perspective of the challenges and issues of the value-added enhanced products. Addressing assay of quality and product differentiation is vital if the maximal potential of biotechnology is to be fully realized. In bringing out this book, the editor realizes that the contributors to the chapters have not written the last word; indeed, some of this work is in its infancy. It is sincerely hoped that this book will be of interest to biotechnology professionals, processors, scientists, nutritionists, economists, new product development and business profes-

Copyright © 2004 AOCS Press

sionals, official agencies, and others actively engaged in development and marketing of value-added products. Contributions by all of the authors are gratefully appreciated. The author is also thankful to family, friends, and colleagues at Monsanto and USDA-ARS for their encouragement and support. D.L. Luthria

Copyright © 2004 AOCS Press

Contents

Preface

Chapter 1

Section I Introduction Technology is one aspect of today that is truly fresh and burning with new tunes and story turns. So there is and can be content in technology— new tunes we’ve never heard before because they’ve never been possible before. (Francis Ford Coppola)

Chapter 1

The Commercial Significance of Oil Content Analysis: The Position of Official Methods Richard C. Cantrill

Section II Primary Reference Methods for Crude Fat Determination Science and technology multiply around us. To an increasing extent they dictate the languages in which we speak and think. Either we use those languages, or we remain mute. (J.G. Ballard)

Chapter 2

Soxtec: Its Principles and Applications Shirley Anderson

Chapter 3

Accelerated Solvent Extraction Devanand Luthria, Dutt Vinjamoori, Kirk Noel, and John Ezzell

Chapter 4

Evaluation of the Rapid, High-Temperature Extraction of Feeds, Foods, and Oilseeds by the ANKOMXT20 Fat Analyzer to Determine Crude Fat Content L.R. Rudnick

Chapter 5

Analytical Supercritical Fluid Extraction for Food Applications Tracy Doane-Weideman and Phillip B. Liescheskii

Section III Comparative Evaluation of Primary Reference Methods and Issues Related to Oil Analysis The higher we soar on the wings of science, the worse our feet seem to get entangled in the wires. (Anonymous)

Copyright © 2004 AOCS Press

Chapter 6

Oil Content Analysis: Myths and Reality V.J. Barthet and J.K. Daun

Chapter 7

Effect of Moisture Content, Grinding, and Extraction Technologies on Assays of Crude Fat Devanand L. Luthria, Kirk Noel, and Dutt Vinjamoori

Section IV Secondary Methods for Crude Fat Analysis Science may be described as the art of systematic oversimplification. (Karl Popper)

Chapter 8

The Rapid Determination of Fat and Moisture in Foods by Microwave Drying and NMR Analysis Bobbie McManus and Michelle Horn

Chapter 9

Simple Methods for Total Oil Content by Benchtop NMR P.H. Krygsman, A.E. Barrett, W. Burk, and H.W. Todt

Chapter 10 Internet-Enabled Near-Infrared Analysis of Oilseeds Ching-Hui Tseng, Kangming Ma, and Nan Wang

Section V Emerging Technologies Invention breeds inventions. (Anonymous)

Chapter 11 High-Resolution Nuclear Magnetic Resonance and Near Infrared Determination of Soybean Oil, Protein, and Amino Acid Residues in Soybean Seeds I.C. Baianu, T. You, D.M. Costescu, P.R. Lozano, V. Prisecaru, and R.L. Nelson Chapter 12 Near Infrared Microspectroscopy, Fluorescence Microspectroscopy, Infrared Chemical Imaging and High-Resolution Nuclear Magnetic Resonance Analysis of Soybean Seeds, Somatic Embryos and Single Cells I.C. Baianu, D. Costescu, T. You, P.R. Lozano, N.E. Hofmann, and S.S. Korban

Copyright © 2004 AOCS Press

Chapter 1

The Commercial Significance of Oil Content Analysis: The Position of Official Methods Richard C. Cantrill AOCS, Champaign, IL 61821

Introduction There are many anecdotal claims that “the error in the measurement of proximate X is costing/losing the industry millions.” Such a charge has been heard in the methods and commodity committees of many national and international organizations. On the other hand, contractual specifications between suppliers and consumers of raw materials are being written much more tightly than ever before. The result is that the precision and accuracy of the methods of analysis used to support these contracts are routinely being questioned. Such circumstances have led the Federation of Oils, Seeds and Fats Associations Ltd. (FOSFA International) to study the contractual method for sunflower seed oil content and modify it to include the determination of moisture, both before and after grinding before the oil extraction step. The original FOSFA Contractual Method was previously adopted by ISO/TC 34/SC 2 (Oleaginous Seeds and Fruits and Oilseed Meals) and developed as ISO 659; it is also reproduced as AOCS Am 2-93. Other standards development organizations (SDO) such as AOCS, AOAC International, CEN, ISO, and Codex Alimentarius are faced with similar problems as the globalization of world standards follows the need to open up world trade. The existence of many versions of the same analytical method in the standards arena is complicated by the routine practice of translating these standards into company standard operating procedures (SOP) and the existence of more variant methodologies. Differences in regional customs, training, and language also contribute to the diversity of analytical methods. All of these considerations have a large effect on both the trade of oilseeds and the introduction of new or modified, value-added crops into the specialty and niche markets and the acceptance of improvements to existing commodity oilseeds. Oil Markets Trade in oils and oilseeds depends on the purchaser being able to determine the yield and subsequently the price of value-added products. The oilseed crushing industry depends on the sale of crude oil and meal for its income stream. In commercial trade,

Copyright © 2004 AOCS Press

commodity prices are set by the major grain exchanges (e.g., Chicago Board of Trade) and by contractual agreement. Discounts and premiums may be paid on contract specifications depending on whether the purchased lot fails to meet, meets, or exceeds these specifications. The acceptability of the lot depends on the analytical results of samples collected according to specific sampling protocols, the price paid, and the anticipated yield of products. Soybeans have a nominal composition of 38% protein, 15% soluble carbohydrate, 18% oil, 14% moisture and ash, and 15% insoluble carbohydrates (dietary fiber). Soybean crushers (processors) buy beans on the expectation of making a profit from the production and sale of soybean meal and oil. The soy crush margin can be calculated as follows: Margin = [(price of soybean meal × pounds of meal per bushel) + (price of soybean oil × pounds of oil per bushel)] – price of soybeans per bushel (1). According to USDA statistics, the margin has rarely been above $1/bushel and has generally been ~$0.75 over the last 15 y (2). Crush margins assume that 1 bushel of soybean (60 lb) yields 48 lb of meal with 44% protein and 11 lb of oil. Because soybean quality varies among varieties and from region to region, the actual yield of oil and meal differs from the assumption above. A calculator to determine the Expected Processing Value (EPV) may be found at www.stratsoy.uiuc.edu. This tool, based on a publication by Brumm and Hurburgh (3), determines the value of soybeans on the bases of their protein and oil content. It uses these values together with commodity prices quoted on the Chicago Board of Trade for soybean meal and oil to determine the processed value. Although there are some assumptions made in the calculation, valuable potential revenue information can be determined (Table 1.1). Profitability is further related to operational efficiency, transportation logistics, and capacity utilization. All of these factors play an important role in the final margin realization. Any number of potential scenarios can be run using EPV to determine the magnitude of the effect of protein and oil content on profitability. At present, the value of the soybean is limited by the values of protein and oil content. Limits are set by the available commodity seed stocks. However, because greater profitability exists in sourcing higher-quality raw materials, soybean buyers are aware of this opportunity and search out these stocks. TABLE 1.1 Example of Output from an Expected Processing Value (EPV) Calculator Calculation inputs

$/bu

Soybeans Calculation results

4.40 $/bu

Meal Oil Hulls Subtotal Theoretical margin

Copyright © 2004 AOCS Press

3.74 1.44 0.14 5.32 0.92

Quality parameters also affect the international trade in oilseeds. The recent acknowledgment of genetically modified soybeans in Brazil removes some of the competitive advantage of soybeans from that area. However, increased yields and increased production in South America are driving U.S. producers to look at the quality of U.S. soybeans to determine a competitive advantage. Production and consumption of oilseeds generally go hand in hand. Reviews of the production of oilseed and oils and fats are regularly produced by world experts Frank Gunstone and Thomas Mielke and may be found in the AOCS membership publication inform (4–6). Development of a Quality Assurance Program Evidence is available from the USDA, the United Soybean Board (USB), and other sources detailing the effect of location within the United States on soybean quality. Each industry would, within the limitations of economic feasibility, source oilseeds with the highest quality. In a commodity-based industry in which the farmer is paid on yield, and profits are determined after the fact, sourcing suitable oilseeds for crushing is more of an art than a science. However, the oilseed trade could be revolutionized if oilseeds with higher levels of protein and oil or other desirable constituents were available on the U.S. market on a regular universal basis. This latter goal is the driving force for the USB Better Bean Initiative. For further details of this program visit www.unitedsoy.org. The program promotes the development of soybeans with enhanced quality traits such as increased oil and protein content, enhanced amino and fatty acid composition, and low phytate content while maintaining yield. Because differences in the precision and accuracy of methods of analysis will hinder the introduction and identification of new varieties of commodity oilseeds, this program has also recognized the need for analytical performance. In 2002, USB sponsored a 4-y program with AOCS to introduce the Soybean Quality Traits Analytical Standards Program (SQT) (Table 1.2). With the initiation of a proficiency testing program to support SQT, AOCS is currently (end of 2003) proceeding with phases 3 and 4. In parallel, the development of a program to support the use of near infrared (NIR) spectroscopy for the analysis of soybean quality traits is underway. In an ambitious and highly cooperative program, the seed breeders, NIR equipment manufacturers, analytical laboratories, consultants, and academics have come together to foster the development of a library of diverse soybean samples to support NIR calibrations for protein, oil, and fatty acid composition. In future developments, it is anticipated that phytate, amino acid composition, and other analytes will be included. Factors Determining the Use of Globally Recognized Standards When considering how complex the soybean industry appears, how integrated the activities and the wealth of expertise available, it is easy to question the need for an analytical system to ensure confidence in the identification of crops with enhanced

Copyright © 2004 AOCS Press

TABLE 1.2 Features of the AOCS-USB SQT Program Phase 1

Phase 2 Phase 3

Phase 4 Phase 5

Phase 6

Identify analytical methods for Protein content Oil content Fatty acid composition analysis Develop and validate methods of analysis including the evaluation of secondary methods Identify users and their requirements Seed companies Referee and private laboratories End-user laboratories Elevator and crop handling facilities Establish a core group of expert laboratories Develop Soybean Quality Traits Laboratory Program, including use of proficiency testing and standards Implement laboratory quality assurance Standard methods Certification Proficiency participation Results monitoring Encourage incorporation of SQT methods of analysis into ISO 17025 certification and quality audits

traits. In fact it is the complexity of the current model that drives the need for standardization. Both internal and international trade in oilseeds require the assurance of analytical values. When contracts specify the expected level of an analyte, they may also specify the method of analysis. This recognizes the variability among different methods of analysis and their performance characteristics, and many disputes are averted because of this understanding. However, these differences in the methods of analysis are inherent from their empirical nature. AOCS has methods for the determination of oilseed fat content dating from the 1930s. At the time of their introduction, they were considered state of the art and an industry standard. Still in use today, they remain the methods of choice in arbitration, and closely related versions of the same methods can be found in the collections of many SDOs. Fats, Oils, and Lipids Methods Standardization There is a long list of developers of national and international standards. Although national standards bodies generally adopt standards methods that have been developed through international cooperation, they may also be developed internally in response to the needs of the trade or other organizations. These standard methods often form the basis of future international methods. In the United States, national standards are published by ANSI, although the developmental work may be carried out by any of a large number of professional organizations. However, in this arena,

Copyright © 2004 AOCS Press

AOCS is recognized as an international nongovernmental organization (NGO) with wide global representation. Most standards organizations publish their own standards and rely on volunteer committees to carry out the developmental and technical work. This generally requires the provision of precision data that are based on collaborative studies as defined in either ISO 5725:1994 (7) or the IUPAC/AOAC Harmonized Protocol (8). Both standards allow the user to calculate the number of laboratories and samples required to provide a good estimate of precision. In most cases, results from 8–15 laboratories for 3–5 samples covering the level of expected values are required for each analyte and matrix covered by the method. The IUPAC/AOAC protocol requires the participation of 5 international laboratories, whereas ISO stipulates that 5 countries agree to participate in the development of the method. There are also minor differences in the statistical treatment of the data to determine outliers, although this does not generally affect the outcome of the analysis of the collaborative study. ISO (International Organization for Standardization; www.iso.org) standards for the fats and oils community are developed by Subcommittees 11: Animal and Vegetable Fats and Oils (TC 34/SC 11) and 2: Oleaginous Seeds and Fruits and Oilseed Meals (TC 34/SC 2) of ISO Technical Committee 34 Food Products (ISO/TC 34). ISO uses a 6-step consultative process to ensure consensus and participation of interested parties. One of the characteristics of ISO standards is that the standard itself may contain only references to other appropriate ISO standards. In recent years, however, ISO has recognized that the source scientific literature is relevant to ISO standards and may be included in Appendices to methods. CEN (European Committee for Standardization; www.cen.be) publishes standards that meet the specific needs of European countries in response to the needs of European industry and the regulations of the European Commission. Fats and oils are handled by TC 307 “Oilseeds, vegetable and animal fats and oils and their by-products—Methods of sampling and analysis.” Under the Vienna Agreement, ISO and CEN agree to cooperate in the development of standards without duplication. Work started by CEN may be transferred to ISO, and CEN adopts ISO standards wherever possible. Methods are developed by a process similar to the ISO process. The International Union for Pure and Applied Chemistry, IUPAC (www.iupac. org) has had a long history of developing methods of analysis for fats and oils. Many of these have formed the basis for harmonization efforts in the fats and oils arena and have been adopted and refined by sister organizations. Although the members were a very active group of dedicated scientists, the Fats and Oils Commission was incorporated first into the Food Chemistry Division, and the latter group was absorbed into the Division for Chemistry and the Environment as IUPAC moves onto a grant-based/project-based program. AOCS (www.aocs.org) supports and maintains an active standards development program and publishes methods for the fats and oils industry in the Official Methods and Recommended Practices of the AOCS. This compendium contains >400 methods of analysis for oilseeds, oils, fats, and their derivatives. Additions and Revisions

Copyright © 2004 AOCS Press

are published annually and the whole volume reviewed on a 5-y cycle. In line with the expectations of ISO 5725 and the IUPAC/AOAC harmonized protocol, new methods are studied collaboratively before publication as Official Methods. Recommended Practices are those methods either with a limited scope, incomplete validation, or urgently needed by the fats and oils industry. The methods program is supported by an Editor-in-Chief, currently David Firestone, the Uniform Methods Committee (UMC), and a number of technical subcommittees of experts in particular methods of analysis or matrices. AOAC International (www.aoac.org) has a long history of providing methods of analysis. Its method validation programs comprise the AOAC® Official MethodsSM Program®, a Peer-Verified MethodsSM Program®, and the AOAC® Performance Tested MethodsSM Program. In addition, AOAC International is in the process of developing an online methods resource (eCAM) that will be a compilation of analytical methods from many organizations. A method published under the AOAC® Official MethodsSM Program® requires collaborative study data from a minimum of 8 independent laboratories. Peer-Verified MethodsSM provide a rapid way for methods to be recognized by a standards writing body at an entry level of validation. National Professional Associations IUPAC, AOCS, and AOAC International are some of the international professional associations with standard methods development programs; however, there are also many national associations that publish standard methods either in the national language or because they meet specific regional needs. Trade Associations Industrial trade associations may require special methods to be used by their members to support the exchange of goods. Although many methods have been proposed for adoption as national or international standards, for others, there is an insufficient amount of precision data available or a more generic method is already available on a national or international basis. Some trade methods have been retained to ensure the continuity of trade because a newer international standard may give slightly different results. In the area of fats and oils, several trade associations play a role in developing analytical methods for their members. FOSFA International (www.fosfa.org), a world leader in the area of fats, oils, and oilseeds, maintains a technical manual that lists methods that are to be used as part of trading contracts among its members. The manual lists methods developed by international groups such as ISO, IUPAC, and AOCS, and maintains FOSFA methods when no official methods are available. Process for AOCS Approval of Official Methods Methods submitted for inclusion are first screened by the Technical Department of AOCS and then evaluated by one of the subcommittees. The response of the sub-

Copyright © 2004 AOCS Press

committee is relayed to the author or proposer, and the revised proposal is considered by the subcommittee. If more validation data are required, the method may be considered as a Recommended Practice and forwarded to the UMC or a collaborative study will be proposed. The Technical Department is available to conduct the trial or analyze the data. Once the method is approved by the subcommittee, it is passed to the UMC for consideration and voting. With the use of electronic manuscript transmission, the whole process can be achieved within 6 mo to 1 y and a new method can be incorporated into the next set of annual Additions and Revisions. In 2001, in response to consumer requests and the trend toward the purchase of individual methods brought about by the requirements of ISO 17025, AOCS introduced Methods Online, which allows users to search the AOCS Methods of Analysis and select individual methods on the AOCS website and receive them in electronic format. AOCS harmonizes and maintains it methods through active participation in ISO, CEN, Codex Alimentarius, IUPAC, AOAC, AACC, and IOOC. Process for the Maintenance of Standard Methods Most standards writing organizations have instituted a process for review of methods with the aim of confirming, revising or deleting them. These reviews usually take place on a regular schedule of 3–5 y. All ISO standards must be reviewed once every 5 y. At this time, the voting committee members are asked to decide whether a standard should be confirmed, revised, or withdrawn on the bases of its relevance, usage, and technical merit. Interactions Between Standards Developing Organizations The last 15 y have seen a major increase in the harmonization of methodologies among SDOs. IUPAC, ISO, AOAC International, and AOCS have actively pursued a program of harmonization of methodologies. The provision of lists of methods of analysis in many Codex Alimentarius documents has done much to foster this process. Codex Alimentarius, a governmental organization set up under the auspices of FAO (Food and Agriculture Organization of the United Nations) and the WHO (World Health Organization) sets guidelines for food safety and trade in food. The activities of the Codex Committee on Fats and Oils (CCFO) and the Codex Committee on Measurement, Analysis and Sampling (CCMAS) are of particular importance to the harmonization activities of AOCS and its partners. As a recognized international NGO, AOCS participates actively in these activities and provides comments in areas in which agenda items affect the interests of the AOCS membership. AOCS methods of analysis can be found listed in relevant Codex documents. The Interagency Meeting (IAM), a subcommittee of CCMAS, currently comprises more than 30 NGOs listed in the CCMAS Directory of Organizations, “known to be active in the field of methods of analysis and sampling” or that have had a method adopted by the Codex Commission. Harmonization is also accomplished by the inclusion of a significant number of representatives from many dif-

Copyright © 2004 AOCS Press

ferent organizations on several of the major committees. An impetus for harmonization is the desire to avoid holding more than one expensive collaborative trial for new methodologies (9). AOCS continues to focus on the needs of the fats and oils community and is seen as an advocate for fats and oils methods. Although there is much to be gained from the harmonization of methods of analysis, there remains a regional preference for methods of analysis from different organizations. It is not clear whether this is based on language, style, the experience and training of laboratory staff and assessors, or habit. These questions are frequently addressed by those concerned with the market share of the methods of a particular organization. Issues Related to Differences in Analysis Procedures The list of reasons for the establishment of several different methods for the determination of the same analyte can be very long. Historically, different techniques grew up around a specific matrix (oilseed) and location. The development of general methods has been one of trial and error. Even among the AOCS methods, where many simple methods are repeated for single oilseeds, it has proved to be a difficult task to develop generic methods and then compile the special considerations for each application. Sampling and laboratory sample preparation are central to the performance of any method and are important considerations when determining the precision of an analytical technique. In the development of oilseed extraction technology, the choice of solvents and the type of equipment play an important role. Many methods committees have devoted considerable time to the discussion of the philosophical question “what is oil?” Because the degree of lipid extraction is highly solvent dependent, answering this question is at the center of determining the equivalence of different methods whether they are standard methods or equipment-based or semiautomated systems. Up to now, standardization procedures have been developed to consider only generic public-domain methods in a recipe-like prescriptive manner. Comparison with proprietary methods has not been performed on a large scale. This book describes one attempt to conduct such a study using the same samples in parallel (see Chapter 3 of this book, Accelerated Solvent Extraction). It takes considerable time, money and goodwill to carry out successful comparisons. Without the help of major corporations and equipment vendors, it is not possible to reach statistically sound conclusions. AOAC has addressed this need through its Research Institute and Peer-Verified Methods program, whereas AOCS has established AOCS Approved Procedures as a category of official methods. Need for Harmonization of Procedures In the face of such a large number of different methods and technologies to determine the oil content of oilseeds, why should there be methods harmonization? The answer is clearly financial. If a seed company develops an oilseed with enhanced quality traits, then to make a return on its investment, it must have market penetra-

Copyright © 2004 AOCS Press

tion and be able to assure the buyer that these traits exist and represent a benefit to the value chain. Without supporting analytical data, these claims cannot be substantiated. The use of different methodologies to determine the trait can easily lead to loss of confidence. Hence, there is a need for consistent methodology. In the case of a dispute between the buyer and seller, the dispute may be settled by arbitration; in that case, a referee laboratory will be selected to perform the analysis according to an agreed reference methodology. In recent meetings, the Codex Committee on Measuring, Analysis and Sampling has proposed the use of performance-based criteria in the establishment and use of methods of analysis. Although this appears to be a challenge to the adoption and use of official methods and international standards, the list of required performance criteria is long and onerous. The degree of validation required may convince the laboratory to adopt and use official methods. The determination of performance criteria or the publication of precision data from a collaborative study in an official method allows the user to determine whether the method will fulfill his needs and whether it is fit-for-purpose. Indeed, the development and interpretation of performance criteria may be considered another way of looking at fitness-for-purpose. Challenges for Harmonization For an SDO and its volunteer members, there is no difficulty in deciding whether the standard developed by another organization is similar enough to its own to consider harmonizing with it. The challenge is to convince current method users that there is an improvement in performance if certain changes are made. When industries are consolidating and looking for ways to streamline and economize, they are less likely to accept the advice of standards organizations. They are more likely to implement vertical integration within the different business units, thereby making official methods secondary to in-house methodology. This may be seen as a problem for the SDO in the short term, but the streamlining of company activities is also a fertile area for invention through miniaturization, high-throughput, real-time, and novel in-line technologies. In the future, these technologies will require recognition and validation through collaborative trial, a role that many SDOs have been performing for the last century. Acknowledgments This chapter was based on a presentation made at the AOCS Annual Meeting 2003 in Kansas City, MO, entitled: The Commercial Significance of Oil Content Analysis, Richard Cantrill. Thanks are given to John Hancock, FOSFA International, London, UK and Mark Matlock, ADM, Decatur, IL for their discussions and loan of materials.

References 1. University of Illinois, Stratsoy, Grain Market Analysis, http://web.aces.uiuc.edu/faq/ faq.pdl?project_id=9&faq_id=677 (November 2003).

Copyright © 2004 AOCS Press

2. United States Department of Agriculture, Oil Crops Situation and Outlook Yearbook, http://usda.mannlib.cornell.edu/reports/erssor/field/ocs-bby/ocs2002.pdf (November 2003). 3. Brumm, T.J., and Hurburgh, C.R., Jr. (1990) Estimating the Processed Value of Soybeans, J. Am. Oil Chem. Soc. 67: 302–307. 4. Gunstone, F. (2003) Early Forecasts for World Supplies of Oilseeds and Vegetable Oil in 2003–2004, inform 14: 668. 5. Mielke, T. (2003) The World Outlook for Major Oilseeds, inform 14: 712–713. 6. Gunstone, F. (2003) Soy—Beans, Oil and Meal, inform 14: 720–721. 7. ISO 5725 (1994) Accuracy (Trueness and Precision) of Measurement Methods and Results, Parts 1–6, International Organization for Standardization, Geneva. 8. Horwitz, W. (1995) Protocol for the Design, Conduct and Interpretation of Method Performance Studies, Pure Appl. Chem. 67: 331–343. 9. Daun, J.K., and Cantrill, R.C. (2003) Process for Development of Standard Methods for the Analysis of Fats, Oils and Lipids, in Advances in Lipid Methodology—Five (Adlof, R.O., ed.) pp. 273–299, The Oily Press, Bridgwater, UK.

Copyright © 2004 AOCS Press

Chapter 2

Soxtec: Its Principles and Applications Shirley Anderson Foss North America, Eden Prairie, MN 55344

Abstract The classical Soxhlet method provides the fundamental basis for a modern-day solvent extraction system, the SoxtecTM. Using the Randall modification, sometimes called the submersion method, the Soxtec provides a faster approach to solvent extraction for the gravimetric quantitation of fat and oil. Typically, the Soxtec methods require only 20–25% of the time required for traditional Soxhlet extraction. Sample preparation, general extraction procedures, method considerations, and optimization are addressed. By definition, the procedure to determine “crude fat” is an empirical method in which the result is determined by the conditions of the procedure. Several aspects of the extraction process, such as solvent type, time, and temperature, are explored. Several standardized Soxtec methods are discussed, including the recently approved AOAC method for determining crude fat in feeds, cereal grains, and forages. Many Soxtec applications are routinely used in food, feed, industrial, and environmental laboratories for the measurement of fats, oils, semivolatiles, and other solvent “extractables.” For the determination of crude fat, descriptions are given for various sample pretreatment and extraction procedures. Practical guidelines for handling challenging samples as well as general suggestions are presented.

History The foundation of today’s automated solvent extraction systems can be traced to 1879 to a German chemist, Franz Von Soxhlet. He devised a liquid/solid extraction apparatus in which a sample is placed in a cellulose thimble and stationed over boiling solvent. Condensed solvent would then drip into the sample, solubilizing extractable material and then siphon back into the boiling solvent, where this cycle would then repeat. After several cycles over many hours, the apparatus is disassembled and the solvent, now containing extract (fat), is evaporated off, leaving the residue for further analysis. The Soxhlet procedure remains the most exhaustive extraction technique, and today it is still widely used. Over the years, there had been some improvements to the basic technique but the procedure remained long, tedious, and prone to variability. In the early 1970s, Edward Randall (1) developed an accelerated extraction technique that cut the extraction time to as little as 30 min. In the Randall method, the sample is lowered

Copyright © 2004 AOCS Press

and totally immersed in the boiling solvent. The simple principle is that the material to be extracted, in this case, fats and waxes, is more soluble in hot solvent than in cold or room temperature solvent. His procedure included this new boiling step followed by a rinsing step to flush residual extract from the sample (Fig. 2.1). Demonstrating excellent agreement with Soxhlet and improved precision, the Randall method has become the basis of many automated extraction systems. In 1975, Tecator AB of Höganäs Sweden acquired the rights to what had become known as the “Randall modification” of the Soxhlet method. This was first commercialized as the RaFaTec and later became the Soxtec systems of today. Figure 2.2 is a photograph of the SoxtecTM Avanti, from Foss-Tecator, an automated Soxhlet extraction system using the Randall submersion technique.

Procedural Overview of Soxtec, Automated Soxhlet for Crude Fat Automated extraction methods using the Soxtec have gained widespread acceptance and have a number of regulatory agency approvals worldwide. The Soxtec method has been used on literally hundreds of different sample types for many extracts. For our purposes, this discussion will be limited mainly to crude fat extraction. Crude fat by solvent extraction is classified as an empirical method (2). This means that the final result can be arrived at only according to the terms or variables of the method. It therefore becomes critical that all aspects of the procedure be followed strictly. Sample Preparation. Proper handling of the sample and attention to detail are extremely important parts of the analytical process. An improperly or sloppily preFIG. 2.1. Original Soxhlet (left)

and Randall Extraction Apparatus. (a) Condenser (b) sample thimble (c) solvent flask (d) siphon tube (e) solvent vapor tube (f) thimble positioning mechanism (g) heater (not shown on the Soxhlet). In the original Randall method, the thimble is positioned by use of the slide rod (f). Lowering the thimble (b) into the boiling solvent for the boiling step, then raising it out of the solvent for the rinsing step. In both stages, condensed solvent is flowing continuously through the sample and thimble back into the boiling solvent.

Copyright © 2004 AOCS Press

FIG. 2.2. The SoxtecTM Avanti 2050 automated extraction system.

pared sample can invalidate even the most carefully performed extraction procedure. Depending on the type and nature of the sample, the sample preparation may incorporate different procedures. For grinding and weighing, the sample should be homogenous and finely ground, usually to pass through a 1-mm sieve (~18 mesh). Particular attention should be paid to the type of grinding mill used. The mill or milling process should not contribute to any loss of moisture or fat from the sample. Samples should be weighed using a calibrated 4-place analytical balance and in most cases, can be weighed directly into the cellulose thimble. The weight of the sample is dependent on its approximate fat content. Table 2.1 can be used a guideline. Because only a few grams of sample are normally used for the analysis, it is critical that this small sample be representative of the larger sample lot. Pretreatment Drying. Most samples should be predried to optimize the fat extraction. Water in the sample can decrease the efficiency of the solvent extraction, resulting in low fat recoveries. Conversely, water-soluble components in the samples such as urea, carbohydrates, salts, and glycerol can be extracted with fat yielding falsely high recoveries. Samples are weighed into the extraction thimbles and are then typically dried at 102 ± 2°C for 1–2 h (3–5,7). Because the samples are weighed before drying,

Copyright © 2004 AOCS Press

TABLE 2.1 Expected Fat Content and Sample Weights Fat content (%) 0–10 10–25 >25

Sample weight (g) 2–3 1–2 0.5–1

the results are on an as-is basis. Results expressed on a dry matter basis must be calculated from a separate moisture determination. For very moist samples, sand may be mixed with the sample before drying. This prevents the sample from becoming caked during drying and improves the solvent flow for optimal extraction (see below: Crude fat in meat and meat products). Hydrolysis. Samples that have been processed, cooked, or extruded often have fat that is bound to proteins, carbohydrates, and/or minerals, making it unavailable for solubilization. Acid hydrolysis, in which a sample is boiled with hydrochloric acid, breaks these bonds, allowing the fat to be solvent extracted (see below: Total fat). Water Rinse. Samples that contain a large amount of water-soluble components may exhibit poor solvent extraction efficiency. A preextraction with water, followed by a thorough drying step can be used to obtain an acceptable recovery by removing these water-soluble components. The procedure specifies washing the weighed sample with 5 aliquots of 20 mL of deionized water. The sample is dried and the extraction procedure is carried out as usual. (3,4) Solvent Extraction Once the samples have been properly prepared and pretreated, they can now be placed in the Soxtec for fat extraction. The weights of clean and dry extraction cups must also be obtained for later use in the final calculation. The samples and extraction cups are then positioned in the extractor. The solvent is added through a closed-loop addition process and the extraction begins. The steps of boiling, rinsing, and evaporation/solvent recovery then proceed in an automated manner. At the end of the cycle, an alarm signals completion. Boiling. In this step, the sample and thimble are lowered and totally immersed in the boiling solvent contained in the extraction cup. The solvent vapor refluxes against a water-jacketed cooling column and the condensed solvent flows back continuously through the sample returning to the boiling solvent (Fig. 2.3). The boiling step is the key to accelerating the extraction process compared with the Soxhlet method. The solvent simply solubilizes the extract faster in hot solvent, thus decreasing the time required for extraction.

Copyright © 2004 AOCS Press

FIG. 2.3. Three-step extraction procedure. The Foss-Tecator Soxtec Avanti automated extraction system is based on the Randall modification of the Soxhlet technique. In the boiling and rinsing steps, solvent is refluxed within the condenser. During evaporation, solvent flow is blocked from returning to the extraction cup and flows out tube (a) into a collection tank (not shown).

To ensure optimal extraction, the level of boiling solvent must be higher than the sample in the thimble. A plug of defatted cotton is frequently placed on top of the sample to keep it in the thimble during extraction. With the Soxtec Avanti system, 70–90 mL of solvent is used. Typical extraction times range from 20 to 40 min depending on the solvent and sample characteristics. Immediately after the boiling step, the rinsing step begins. The sample is raised and suspended over the boiling solvent. During rinsing, residual traces of the extractable material are flushed out of the sample and are retained in the extraction cup. This step is usually 10–20 min longer than the boiling step to ensure complete extraction. The last step in the crude fat extraction process is evaporation/solvent recovery. The condensed solvent continues to boil and evaporate and, using an internal valve, the condensate is redirected out of the condenser. The evaporation step is complete when all solvent is driven from the cup, concentrating the extract. This usually requires 7–10 min depending on the solvent. Excessive drying may oxidize the extract, causing weight changes and erroneous readings. The Soxtec Avanti stores the evaporated solvent in a common collection tank for reuse. The Soxtec Avanti offers an optional fourth, cup predrying step, i.e., the extraction cups are raised a few millimeters off the heating surface allowing radiant heat to complete the drying cycle. This step is used in applications in which the extract is extremely heat labile. Postextraction. Once the extraction process is completed, the cups are taken off the Soxtec and placed in a drying oven at 103°C for 30 min to drive off any moisture or solvent residuals. Extended drying, especially at higher temperatures, should be avoided because it can cause oxidation of the fat extract and falsely high results. Extraction cups are cooled completely to room temperature in a desiccator before final weights are taken.

Copyright © 2004 AOCS Press

Calculation of Results. Crude fat by solvent extraction is a gravimetric method. The final result is calculated from the original sample weight and the weights of the extraction cup before and after the extraction. % Fat = (W2 – W1)/W3 × 100 where

W1 = weight of the extraction cup W2 = weight of the extraction cup + extract W3 = weight of the sample

All weights should be recorded to 0.1 mg (0.0001g).

Optimizing the Extraction Process The Soxtec/Soxhlet extraction method for crude fat relies on separating sample components on the basis of physical and chemical (solubility) properties. There are several factors that influence the extraction; the most significant of these is the specific solvent that is being used. Nevertheless, the influence of sample preparation, extraction timing, and general Soxtec operating conditions is important. Keeping in mind the empirical nature of the analysis, consistency with all aspects of the procedure is strongly recommended. An often overlooked aspect of a fat extraction method is the predrying of the sample. As mentioned earlier, water in the sample can contribute to error in two ways, i.e., it can act as a physical barrier preventing dissolution of the fat into the solvent, thus generating low fat recoveries; it can also contribute to falsely high apparent fat recoveries by allowing water-soluble components such as urea or carbohydrates to be co-extracted with the fat. Unfortunately, in the interest of time and productivity, many laboratories do not predry samples. In these instances, the error potential for each type of sample should be investigated fully by carrying out extractions both with and without predrying. Figure 2.4 illustrates apparent fat recovery on dried vs. undried samples. Moisture in the samples ranges from 5 to 25%. Some samples show a “water effect” more than others. Samples such as the texturized feeds, which contain molasses, and the feedlot concentrate, which contains urea, are examples in which failing to predry the sample can have a marked effect on recovery. Note that this is also dependent on the solvent used. [%Recovery is defined as (%crude fat from the undried sample/% crude fat from the dried sample) × 100]. Solvent Choices. The versatility of the Soxhlet/Soxtec extraction method allows for the use of various classes of organic solvents. These include ethers, aliphatic, aromatic, and chlorinated hydrocarbons, as well as alcohols. Due to the different solubility characteristics of various solvents, a sample extraction will have somewhat different fat yields depending on the solvent. For crude fat extractions, diethyl ether and petroleum ether are most commonly used. The peroxide-forming

Copyright © 2004 AOCS Press

Undried vs. Dried

Diethyl ether

Corn High Oil

Broil er Sta rter

Swin e Fe ed

Calf Feed , Med icate d Meat Meal/ Hulls Mixtu re

Feed lot C onc. (Urea ) Calf Start er, M edica ted

Med icate d Go at Fe ed

Fat S upple men t

Textu rized F (Mola eed sses )

Mixe d Bir d Se ed

Corn Silag e

Dehy drate d Alfa lfa

% Recovery

Hexanes

FIG. 2.4. Apparent fat recovery from dried vs. undried samples with moisture content ranging from 5 to 25%.

nature of diethyl ether causes it to be a less than desirable choice for routine use in the laboratory. This has caused many laboratories to look for an alternative solvent. Commonly, petroleum ether is directly substituted. However, petroleum ethers or ligroin, are not true ethers but mixtures of aliphatic hydrocarbons and can be purchased in various formulations and boiling point ranges. Further complicating the “pet ether” issue is that solvents are often recycled and reused in the Soxtec. This can cause a change in the properties of petroleum ether because its more volatile components may be driven off. This can cause a change or drift in the fat results. Considering the innate variability of petroleum ether and the relative lower recovery of plant-based lipids, it is not a suitable substitution for diethyl ether. An experiment was undertaken (3,4) to compare the recovery of three common solvents, petroleum ether, hexanes, and pentane, to that of diethyl ether in terms of crude fat recovery. The objective was to find a solvent that is safer and has recovery statistically equivalent to diethyl ether. The results are shown in Table 2.2. From these results, it can be seen that hexanes yield a recovery equivalent to that of diethyl ether with an R2 of 0.9925. It should also be noted that for meat and bone meal, petroleum ether also yields a good recovery. This is consistent with the use of petroleum ether in the AOAC method 991.36 (7) for crude fat in meat and meat products. Extraction Times. In Soxtec extraction procedures, the timing for the boiling and the rinsing steps is important. If the boiling or rinsing step is too short, the extraction will likely not adequately recover the fat in the sample. Most method development protocols will define the extraction times at which the results closely match

Copyright © 2004 AOCS Press

TABLE 2.2 Crude Fat Recovery of Four Common Solvents Diethyl ether Sample Alfalfa hay Beet pulp Meat/bone meal Cattle protein supplement Corn Average R2

Petroleum ether

Hexanes

Pentane

(% Crude fat) 1.29 0.30 10.52 3.10 3.59 3.76 —

1.00 0.24 10.40 2.65 3.06 3.47 0.9878

1.36 0.25 10.69 2.88 3.16 3.67 0.9925

0.97 0.19 10.54 2.54 3.00 3.45 0.9880

those obtained by classical Soxhlet methods using suitable reference materials. The automation of the Soxtec offers consistent extraction timing for each batch of samples. Extraction Temperature. The temperature of the extraction system should be set to the recommendations provided by the manufacturer. This helps ensure optimal condensation or reflex rates. Ideally, this is typically 3–5 drops/s coming off the condenser. Condenser Temperature. The temperature of the condenser cooling water plays an important role in establishing the condensation or reflux rate of the solvent. Cold tap water, 5% carbohydrates, >15% glycerol, lactic acid, or amino salts, or >10% of other water-soluble

Copyright © 2004 AOCS Press

TABLE 2.4 Evaluating the Robustness of the Crude Fat Method Using Cattle, Swine, and Mixed Feeds from Three Different Laboratories Laboratory 1

Laboratory 2

Laboratory 3

(% crude fat) Feed type

Cattle

Swine Mixed

Cattle Swine Mixed

Cattle

S T U V W X Y Z Predry 2h 4h Difference Boil time 20 min 40 min Difference Ether Diethyl Petroleum Difference Rinse time 30 min 60 min Difference Sample wt 1g 3g Difference Cup dry 30 m 2 hr Difference Drop rate 2/s 6/s Difference

11.61 11.30 11.79 11.02 10.87 11.11 11.62 10.73

2.45 2.10 2.46 2.17 2.60 2.00 2.55 1.91

11.43 11.08 0.34

Swine Mixed Average

10.50 10.24 10.60 9.88 10.67 10.00 10.60 9.82

11.82 11.63 11.98 11.76 11.61 11.28 11.49 11.19

2.41 2.35 2.81 2.70 2.52 2.10 2.64 2.10

11.16 10.25 10.84 10.78 9.86 10.28 10.53 10.01

11.99 11.55 11.84 11.57 11.87 11.24 11.94 11.74

2.99 2.49 2.93 2.46 4.13 2.38 2.82 2.51

10.97 10.64 11.05 10.50 10.94 10.43 10.91 10.73

2.29 2.26 0.03

10.30 10.28 0.03

11.80 11.39 0.41

2.57 2.34 0.23

10.76 10.17 0.59

11.74 11.70 0.04

2.72 2.96 –0.24

10.79 10.75 0.04

0.16

11.22 11.29 –0.06

2.29 2.27 0.02

10.35 10.23 0.13

11.59 11.61 –0.02

2.35 2.56 –0.22

10.39 10.54 –0.15

11.66 11.77 –0.11

3.00 2.68 0.32

10.75 10.80 –0.05

–0.02

11.47 11.04 0.43

2.52 2.04 0.47

10.59 9.99 0.61

11.73 11.47 0.26

2.60 2.31 0.28

10.60 10.33 0.27

11.91 11.53 0.38

3.22 2.46 0.76

10.97 10.58 0.39

0.43

11.32 11.20 0.12

2.25 2.31 –0.06

10.29 10.29 0.00

11.53 11.66 –0.13

2.38 2.53 –0.16

10.49 10.44 0.05

11.81 11.63 0.17

2.70 2.98 –0.27

10.81 10.73 0.08

–0.02

11.31 11.20 0.11

2.20 2.35 –0.15

10.23 10.35 –0.12

11.57 11.62 –0.06

2.36 2.55 –0.20

10.57 10.36 0.22

11.70 11.73 –0.03

2.70 2.98 –0.27

10.80 10.75 0.05

–0.05

11.06 11.46 –0.40

2.28 2.28 0.01

10.22 10.36 –0.14

11.60 11.60 0.00

2.43 2.48 –0.04

10.45 10.48 –0.02

11.79 11.64 0.15

3.02 2.66 0.37

10.79 10.76 0.03

0.00

11.34 11.17 0.17

2.29 2.27 0.02

10.24 10.33 –0.09

11.59 11.60 –0.01

2.46 2.45 0.02

10.69 10.24 0.45

11.69 11.75 –0.06

2.66 3.02 –0.35

10.70 10.84 –0.14

0.00

components, the sample is washed with 5 aliquots of 20 mL deionized water. The sample is dried at 102°C for 2 h. Extraction is performed with either diethyl ether or hexanes using a 20-min boil and 40-min rinse cycle. The Soxtec method was compared with the AOAC Soxhlet method on data from 90 AAFCO check samples. Regression analysis generated an R2 correlation coefficient of 0.9946, slope

Copyright © 2004 AOCS Press

TABLE 2.5 Average Recoveries and Relative Standard Deviations (RSD) for Meat Samples with Petroleum Ether Extraction Average recovery

RSDr

Average recovery

Soxhleta Meat sample 1 2 3 4 5 6 Average

RSDr

Soxtecb (%)

4.6 28.35 28.21 34.98 34.10 26.81 26.18

2.63 8.75 5.52 2.09 2.25 1.74 3.83

4.34 27.29 27.95 34.51 33.57 26.20 25.64

2.44 1.95 2.32 2.21 1.01 1.55 1.91

aSoxhlet: bSoxtec:

4-h extraction, 2-h drying time. 55-min extraction, 30-min drying time.

of 1.00062 and a y-intercept of 0.137. On the basis of these data, the methods appear to be comparable (3,4). Environmental EPA 3541, SW 846. Of note, the Soxtec extraction procedure is used widely in environmental laboratories to extract organic compounds such as pesticides or PCB from soils, sludge, sediments, and hazardous waste samples (9,10). In these applications, the Soxtec is used as a sample preparation device. The sample is weighed directly into a cellulose or fritted glass thimble and placed in the Soxtec where the extraction of semivolatile organics is done using a mixture of hexane and acetone (1:1). The process is stopped during the evaporation/solvent recovery step while there is still 15–20 mL of solvent (containing the semivolatiles) left in the cup. The extract/solvent mixture is further concentrated and the final analysis performed by GC or GC/MS techniques. The Soxtec method with a 2-h extraction replaces the traditional Soxhlet, which takes 8–24 h. Total Fat. In samples that are baked, extruded, or with some commercial processing, the fat becomes bound to other components in the sample such as proteins, carbohydrates, and minerals. An acid hydrolysis before the solvent extraction step is needed to “free” the fat in the sample, making it available for solvent extraction. Typically 1–2 g of sample is boiled with strong hydrochloric acid solutions. The sample is then rinsed, dried and then extracted. The SoxCapTM (Fig. 2.5) for acid hydrolysis from Foss Tecator enables the acid hydrolysis step and extraction step to occur in the same sample vessel, thus eliminating any sample transfer errors. Industrial Applications. Soxtec extraction methods were found to be suitable for use in industrial applications. A summary of some of these applications appears in Table 2.6. Application Sub Notes (ASN) are available from Foss Tecator.

Copyright © 2004 AOCS Press

FIG. 2.5. The 2047 SoxCapTM hydrolysis system.

TABLE 2.6 Summary of Soxtec Extraction Methods Used in Industrial Applicationsa ASN ASN ASN ASN ASN ASN ASN ASN ASN ASN ASN ASN ASN ASN ASN ASN ASN ASN ASN aApplication

3516 3602 3603 3604 3605 3606 3607 3608 3611 3612 3613 3614 3615 3616 3617 3618 3619 3622 3700

Extraction of aromatic hydrocarbons in soil Extraction of resins from paper pulp Extraction of finish from textiles Extraction of starch containing finish from textiles Extraction of surfactant from detergents Extraction of paraffin from detergent Extractable matter in leather Extraction of petroleum source rock Extraction of explosives and propellants Extraction of plastics and polymers Extraction of rubber and rubber compounds Extraction of finish oils from textiles and synthetic fibers Extraction of migration components in plastic packaging Extraction of organic dyestuffs Extraction of leather Extraction of core material in petroleum exploration Extraction of tobacco Extraction of solubles in paper pulp Extraction of fecal fat

sub notes (ASN) are available from Foss Tecator.

Copyright © 2004 AOCS Press

TABLE 2.7 Common Approaches to Help Extract Crude Fat from Difficult Samples High-fat samples that melt during drying

• Place the thimble containing the sample into a preweighed extraction cup. Any sample that melts out of the thimble will be retained in the cup.

Incomplete fat recovery, high-fat seeds

• A two-phase extraction protocol is used: After the first extraction boil and rinse, the samples are removed from the thimbles, ground with a mortar and pestle and returned to the thimble for a secondary extraction. The results are added together.

Sample becomes impacted during extraction

• Mix equal volumes of acid-washed sand or Celite and sample in the extraction thimble. This allows for better solvent flow through the sample.

Solvent overboiling

• Use 3–5 boiling beads in the extraction cups. • Decrease the temperature setting on the extractor.

Moist samples

• Mix sample with sand and predry. • Mix sample with equal weight of sodium sulfate to bind water.

Semisolid

• Depending on the nature of the sample, mix with either sand and dry or mix with sodium sulfate.

Nonhomogeneous samples • Optimize sample preparation step. • Use larger sample size to obtain a representative sample. • Do replicate analysis to generate reportable results. Low-fat samples

• Use larger sample weights.

General practice

• Place a plug of defatted cotton on top of the sample to ensure that the sample is retained in the thimble. • Wear gloves during handling of thimbles and cups to avoid errors. • Weigh cups at room temperature. Weighing errors will result from warm cups. • When using recovered petroleum ether, supplement with fresh ether to help maintain desired boiling point range. • Diethyl ether can be purchased with stabilizers to minimize the formation of peroxides. Such ether should be used with the label guidelines. • Test strips are available to check for peroxide formation in diethyl ether.

Difficult Samples. Samples that represent special challenges and handling are frequently encountered. Table 2.7 summarizes some common approaches to aid the analyst in extracting crude fat from these types of samples. Quality Control. Evaluating the performance of the extraction process is normally achieved by running a reference or check sample. Many commercial check sam-

Copyright © 2004 AOCS Press

ple services are available for different sample types. Results should be compared only to those from similar instruments running the same extraction procedure and sample protocol.

Conclusion The long history of solvent extraction has led to automated Soxhlet extraction systems, such as the SoxtecTM Avanti. They offer convenient and useful tools with which to improve productivity in the laboratory. Modern instrumentation provides application flexibility and improved economy as well as enhancement of the precision and recovery of the extraction. References 1. Randall, E.L. (1974) Improved Method for Fat and Oil Analysis by a New Process of Extraction, JAOAC 57: 1165–1168. 2. Codex Alimentarius Commission (1986) Procedural Manual, 6th edn., p. 139, Food and Agricultural Organization, Rome, Italy. 3. Thiex, N., Anderson, S., and Gildemeister, B. (2003) Crude Fat, Hexanes Extraction, in Feed, Cereal Grain, & Forage (Randall/Soxtec/Submersion Method): A Collaborative Study, JAOAC Int. 86: 888–898. 4. Thiex, N., Anderson, S., and Gildemeister, B. (2003) Crude Fat, Diethyl Ether Extraction, in Feed, Cereal Grain, & Forage (Randall/ Soxtec/Submersion Method): A Collaborative Study, JAOAC Int. 86: 899–908. 5. Official Methods of Analysis of AOAC International, 16th ed., Chapter 4, p. 25, section 4.5.01. AOAC Official Method 920.39, Fat (Crude) or Ether Extract in Animal Feed, AOAC International, Gaithersburg, MD, 1997. 6. Foster, M.L., and Gonzales, S.E. (1992) Soxtec Fat Analyzer for Determination of Total Fat in Meat: Collaborative Study, Kansas State Board of Agriculture, JAOAC Int. 75: 288–292. 7. Official Methods of Analysis of AOAC International, 16th ed., Chapter 39, p. 3, section 39.1.08. AOAC Official Method 991.36, Fat (Crude) in Meats and Meat Products, AOAC International, Gaithersburg, MD, 1997. 8. Youden, W.J., and Steiner, E.H. (1975) Statistical Manual of the AOAC, Association of the Official Analytical Chemists, Arlington, VA. 9. Lopez-Avila, V. (Beckert, W., Project Officer) (1991) Development of a Soxtec Extraction Procedure for Extracting Organic Compounds from Soils and Sediments, EPA600/X91/140. U.S. EPA, Environmental Monitoring Systems Laboratory, Las Vegas. 10. Test Methods for Evaluation of Solid Waste, Physical/Chemical Methods (1996) SW-846, Method 3541. U.S. Environmental Protection Agency, Office of Solid Waste, Washington, DC.

Copyright © 2004 AOCS Press

Chapter 3

Accelerated Solvent Extraction Devanand Luthriaa,*, Dutt Vinjamoorib, Kirk Noelc, and John Ezzelld Beltsville, MD 20705; bMonsanto, St. Louis, MO 63167; cMonsanto, Ankeny, IA 50021; dDionex Corporation, Salt Lake City Technical Center, Salt Lake City, UT 84119

aUSDA/ARS/FCL,

Abstract Extraction of solid and semisolid samples using liquid solvents is a common practice in nearly every analytical laboratory. Years of empirical testing have resulted in rugged and reproducible methodologies for a wide range of analyte classes. However, recent concerns regarding the volumes of organic solvents used (with the associated human exposure), along with increased purchase and disposal costs, have emphasized the need for more efficient sample extraction methods. In response to these concerns, accelerated solvent extraction (ASE®, Dionex Corporation, Salt Lake City, UT) was introduced. Since its introduction in 1995, ASE has grown rapidly as an accepted alternative to traditional extraction methods. Accelerated solvent extraction takes advantage of the enhanced solubilities that occur as the temperature of a liquid solvent is increased. Increasing the temperature of solvent results in a decrease in viscosity, allowing better penetration of the sample matrix. In addition, analyte diffusion from the sample matrix into the solvent and overall solvent capacity are increased. In traditional Soxhlet extraction, the solvent that comes into contact with the sample has passed through a cooling condenser, and is therefore close to room temperature at the point of contact. The time required to complete Soxhlet extractions ranges from 6 to 48 h. Semi-automated Soxhlet systems that immerse the sample into boiling solvent are available. This increase in the temperature of the contacting solvent shortens the required extraction time to ~2 h. Using these systems, a further increase in temperature beyond the boiling point of the solvent is not possible due to solvent loss because these systems operate at atmospheric pressure. However, a continued increase in the temperature should continue to enhance the extraction process. This can be accomplished by applying pressure, which maintains the solvent in its liquid state beyond its atmospheric boiling point. This is the theoretical basis for ASE technology and represents the next step in liquid solvent extraction of environmental samples. There are, of course, limits to which raising the temperature is feasible, due to thermal degradation concerns. However, as evidenced by data published to date, there is room to continue raising the temperature, thereby improving the extraction efficiency, without risking analyte degradation in environmental samples. As the extraction efficiency is *The research work was done at Monsanto.

Copyright © 2004 AOCS Press

increased, the time required to perform extractions and the amount of solvent needed is reduced. When performing ASE, a 10-g sample can be extracted in ~12 min using 12–15 mL of solvent. As the sample size is increased (33 mL volume maximum), the amount of solvent used increases proportionally (45–50 mL maximum), but the total extraction time remains unchanged. The short extraction times and small solvent usage make this technique amenable to automation. Samples loaded into stainless steel extraction vessels (11, 22, or 33 mL internal volume) are extracted sequentially into standard 40- or 60-mL glass collection vials. After extraction, the spent sample remains in the cell, whereas the extract is immediately ready for processing. The system is designed to extract up to 24 samples unattended. Because existing solvent-based extraction methods can be readily transferred to ASE technology, methods development is greatly simplified. Existing sample preparation and postextraction processing steps remain in place because the extracts generated by ASE will be of very nearly the same composition as those generated with the existing solvent-based extraction technique. The large range of polarities and solvent strengths available when using liquid solvents, including solvent mixtures, allows a high degree of flexibility and selectivity when developing methods for new sample matrices.

Introduction Accelerated solvent extraction (ASE), also referred to as pressurized fluid extraction (PFE) and pressurized liquid extraction (PLE), is a liquid solvent extraction technique that uses aqueous and organic extraction solvents at elevated temperatures and pressures. Although the initial applications focus of this technique was the environmental area, the versatility and ease of use of the approach have proven useful for laboratories performing extractions in the food and polymer industries, as well as in the pharmaceutical and consumer products areas. Traditional reflux-based extraction techniques such as Soxhlet extraction can take from 4 to 48 hour to perform, and 24-h extractions are common. Other liquid solvent-based extraction techniques such as wrist shaker, hot plate boiling, and sonication require copious amounts of solvent and often involve extensive labor steps such as filtering or concentration before extract analysis. One thing that they all have in common is operation at ambient pressure. An increase in temperature beyond the boiling point of the solvent is not possible due to solvent evaporation. Accelerated solvent extraction is performed by using the same solvents as in the traditional approaches, but at higher temperatures than is possible in these techniques. This increase in temperature improves the kinetics of the process, resulting in more efficient extractions (faster and using less solvent) compared with traditional approaches. The solvents are used under pressure so that their liquid state is maintained under heated conditions. For example, solvents such as water, methanol, acetone, or hexane are routinely used in ASE at temperatures ranging from 75 to 150°C. The solvents are maintained as liquids under pressure, normally at 1500 psi (10.4 MPa). ASE is performed, therefore, using very hot liquids to expedite the extraction process.

Copyright © 2004 AOCS Press

The flow-through design of the technique results in extracts that do not require the extended labor of filtration as a means of separating the sample matrix from the extracted analytes. In further contrast to traditional extraction approaches, all of the basic steps are amenable to automation, freeing the analyst from the labor-intensive nature of most sample preparation protocols. Automated ASE systems can extract up to 24 sample cells, and have built in the necessary safety considerations for unattended operation. Instrumentation A schematic diagram of an ASE system is shown in Figure 3.1. The extraction procedure consists of a combination of dynamic and static flow of the solvent through a heated extraction cell containing the sample. These cells must be capable of safely withstanding the pressure requirements of the system, and are normally constructed of stainless steel, with frits in the end caps to allow the passage of solvent while maintaining the solid sample within. Cell sizes range from 1 to 100 mL. The pore size of the frit should not allow passage of the matrix particles (5–10 µm is typical). The sample cell is interfaced to the solvent flow path, where it is filled with solvent. It is important to ensure that all of the void volume has been filled with solvent to have good contact between the sample matrix and the solvent, and to avoid possible analyte oxidation, which may occur in the presence of air at elevated temperatures. The sample cell is then heated by direct contact with a heat source (heating the cell before solvent introduction can result in the loss of volatile compounds). To maintain the extraction solvents in their liquid state, a pressure source must be applied. The system pressure must be above the threshold required to maintain the liquid state of the solvent at the set temperature and be able to move the solvent through the sample cell in a reasonable time period. This is normally accomplished with an HPLC-type pump, which can

Fig. 3.1. Schematic diagram of an ASE system.

Copyright © 2004 AOCS Press

maintain a constant fluid pressure of 1000–3000 psi (6.9–20.7 MPa). Once thermal equilibrium has been reached, the sample cell is maintained at the set temperature for an additional time period of 5–10 min. During this static phase, analyte diffusion from the matrix into the solvent is believed to occur. After this static hold step, the outlet valve is opened and a measured volume of fresh solvent, usually 40–60% of the cell volume, is allowed to flush over the sample, discharging the previous volume into the collection vial or bottle. Last, compressed nitrogen gas is used to force all of the solvent from the lines and cell into the collection vessel. It is important that all of the liquid solvent used in the extraction process be collected for analysis. The collection vessels normally used are standard 40- or 60-mL vials, or 250-mL bottles, sealed with Teflon-coated septa. This allows the extracts to be collected and maintained in a sealed, inert environment (under a nitrogen blanket) to prevent sample loss while waiting for quantification. Sample Preparation Proper sample preparation is essential to obtain efficient and reproducible extractions. The ideal sample for extraction is a dry, finely divided solid, through which the extraction solvent can easily flow and thoroughly penetrate the matrix particles. Whatever can be done, within reason, to make samples approach this definition will be beneficial to the extraction process. Generally, samples should be prepared for ASE extraction in the same manner as traditional extraction techniques. Samples with large particle sizes (>1 mm) should be ground so as to increase the surface interaction of the solvent and matrix. Wet or sticky samples should be mixed with drying agents such as sodium sulfate or pelleted diatomaceous earth, or with dispersing agents such as Ottawa sand before extraction. Typical sample sizes used in ASE are 1–50 g of solid or semisolid material. Sample Extraction Parameters Extraction Solvent. As extraction parameters, solvent choice and temperature have the greatest effect on extraction efficiency with ASE. An extraction solvent that would solubilize the target analyte(s) but leave the majority of the sample matrix intact should be chosen. This is normally done by matching the polarity of the solvent and target analyte. ASE extraction can be performed with the entire range of aqueous and organic solvents, with the exception of strong mineral acids (hydrochloric, nitric, sulfuric), which will attack the stainless steel flow path of the system. In those cases in which an acidic pH is required, small amounts (1–5%) of acetic, phosphoric, or other weak acids can be used. The choice of solvent should also be considered in light of the postextraction analysis technique to be utilized. Solvents such as methanol and acetonitrile are suitable for direct HPLC injection, whereas solvents such as hexane, methylene chloride, or acetone are more suitable for complete evaporation, or concentration and GC analysis. If the target compounds are easily oxidized, solvents should be degassed before use. It has been observed that solvents that perform only marginally

Copyright © 2004 AOCS Press

well at ambient temperature will often perform quite well at elevated temperature. This increases the range of solvent choices available to the analyst considering ASE because the use of more than one solvent may result in good recovery of target analytes. The selection of the appropriate solvent can then be made on the bases of selectivity of extraction, solvent cost, safety and exposure factors, and compatibility with postextraction processing steps. Solvent mixtures should also be considered in cases in which minor adjustments to polarity are desired. Automated sequential extraction of the same sample with multiple solvents is also possible. This approach can be used for fractionation of analytes based on polarity from the same sample matrix. This approach is very valuable for isolation and separation of bioactives from different natural sources. Extraction Temperature. ASE extraction can be performed from ambient temperature to 200°C. Increased temperature will increase the efficiency of the extraction process, and this should be optimized short of the point at which analyte degradation or excessive co-extraction of matrix components occurs. Many applications are performed in the 75–150°C range, with 100°C as the recommended starting point for new methods development. In this temperature range, significant increases in extraction efficiency are observed without breakdown of target compounds. If an extraction is to be performed on a compound with a known thermal degradation point, then the method should be developed to operate below that point. Extractions performed at low (40–70°C) or ambient temperatures may be sufficient for analytes that are weakly or only surface-bound to the matrix. The extracts generated using ASE will be similar in composition to those produced by other techniques using the same solvents. If a postextraction clean-up step is required after a Soxhlet extraction, the same process will most likely have to be performed after ASE. Extraction Pressure. Although essential to the process, pressure is not generally considered a critical parameter. Normal operating pressures of 1500–2000 psi (10.3–13.8 MPa) are well above the threshold pressures required to maintain the solvents in their liquid states at ASE operating temperatures. The main purpose of using pressures in the ranges indicated is to provide rapid filling and flushing of the extraction cells. Typical extractions are performed in 12–20 min, although this time can be extended for difficult samples. In addition, multiple static cycles can be used to periodically introduce aliquots of fresh solvent during the extraction process. Method Development and Optimization When developing a new method, the following approach has proven useful. A representative sample should be prepared as outlined above. Select an extraction cell size that most closely matches the desired sample size. The extraction cells do not have to be filled completely; however, a full cell will use less solvent in the extraction process than a partially filled one. Select the extraction solvent using the considerations listed above, although normally the same solvent or solvent mixture

Copyright © 2004 AOCS Press

used in a traditional liquid extraction method is used. Extract the sample starting with the following standard conditions: pressure, 1500 psi (10.3 MPa); temperature, 100°C; heat time, 5 min; static time, 5 min; flush volume, 60% of cell volume; purge time, 60 s; static cycles, 1. Extract the same sample multiple times to assess the efficiency of the method. If there is significant analyte present in the second or third extracts, adjust the following parameters (one at a time), and repeat the validation process: (i) Increase the temperature (use 20°C steps). (ii) Add a second or third static cycle. (iii) Increase the static time (use 5-min increments). If these steps do not result in a complete extraction, reexamine the sample preparation steps and/or the choice of extraction solvent.

Applications ASE extraction technology has been used extensively in various industries. Some of the applications of ASE extraction technologies are summarized briefly below. Extraction of Crude Fat (Oil) from Soybean Seeds Ground corn and soybean seeds were placed in the extraction cell and crude fat was extracted with ASE. Details of extraction parameters and solvent used are listed in Table 3.1. The total crude fat content was determined by collecting the extracts in preweighed vials followed by evaporation of the solvent under a nitrogen stream. The percentage crude fat content was determined gravimetrically. Comparison of Crude Fat Content Determination by ASE with Standard Butttube and Soxtec Procedures. The total crude fat content (dry matter basis) was determined in three soybean samples by three different extraction methods. Each sample was ground with a ball grinder mill and the ground sample was mixed to ensure homogeneity. The same ground sample was used for analyses to reduce the effect of particle size on extraction efficiency. Results are reported on a dry matter basis (DMB) to eliminate the effect of moisture content on crude fat analyses. Six replicate analyses were performed on each sample for each of the three methods: ASE , SoxtecTM (Foss North America, Eden Prairie, MN), and Butt-tube. Table 3.2 compares the percentage of crude fat, standard deviation (SD) and the relative stanTABLE 3.1 Operating Conditions for Accelerated Solvent Extraction Preheat Heat Static Flush Purge Cycles

0 min 6 min 5 min 50% volume 90 s 3

Copyright © 2004 AOCS Press

Pressure Temperature Solvent compartment Petroleum ether Average run time/sample

1000 psi 105°C A 100% 30 min

TABLE 3.2 Comparison of Three Crude Fat Extraction Procedures from Three Soybean Samplesa

Sample ID SOY 1 SOY 2 SOY 3 Soy average aResults

Soxtec 1 g/2 g Average

Soxtec 1 g/2 g SD

Soxtec 1 g/2 g RSD

ASE 2g Average

ASE 2g SD

ASE 2g RSD

Butt-tube 2g Average

Butt-tube 2g SD

Butt-tube 2g RSD

19.68 21.86 23.70 21.75

0.20 0.23 0.18 0.20

1.02 1.05 0.74 0.94

21.15 22.66 24.61 22.81

0.07 0.05 0.12 0.08

0.32 0.21 0.50 0.34

19.54 21.22 23.32 21.36

0.08 0.11 0.11 0.10

0.41 0.54 0.46 0.47

are average of six replicate analyses.

Copyright © 2004 AOCS Press

dard deviation (RSD) from the three soy samples by three different methods. The results in Table 3.2 indicate that the percentage crude fat extracted by the ASE procedure for soybean samples were 1.1 and 1.4% higher, than the Soxtec and Butttube methods, respectively. The higher extraction yield may be due to differences in extraction conditions or passage of very fine particles through the frit, or passage of moisture during the flush cycle because analyses were carried out on “as is” basis and the results were converted to DMB after analysis. Effect of Sample Size. To evaluate the effect of sample size on crude fat determination, a single ground soybean sample was extracted in six replicates with sample sizes varying from 20 mg to 2 g. The average crude fat extracted from the various sample sizes ranged from 20.3 to 21.3% (Table 3.3). The SD and % RSD gradually decreased as sample size increased from 20 mg to 2 g. Reproducibility. Table 3.4 depicts the ruggedness data of the percentage of crude fat extracted from 125 soy samples with two ASE instruments by multiple operators over a period of time. The SD and % RSD from 125 replicate analyses of crude fat extracted from the soybean samples were 0.32 and 1.56%, respectively. The results indicate that single seed or partial seed analysis is feasible using ASE. In particular, this technique should be very helpful for the analysis in F1 and F2 stages of plant breeding and for screening rare and elite germplasm lines in which sample amounts available are always limited. Extraction of Tocopherols from Soy and Corn Analysis of tocopherols in soy and corn is of considerable importance from the nutritional perspective. Although there are several HPLC methods reported in the literature, few reliable sample preparation/extraction techniques exist that ensure the integrity and stability of tocopherols with quantitative recoveries. Addition of an antioxidant such as pyrogallol or ascorbic acid to the extraction solvent usually helps in achieving quantitative recoveries. However, no such antioxidant need be used if the ASE technique is used because the extraction is performed under nitrogen atmosphere and samples are collected in sealed vials. Figure 3.2 illustrates the comparison between the manual tissue grinder extraction with ethanol containing pyrogallol and ASE extraction with EtOH alone (ASE conditions are the same as those stated in Table 3.1). Defatting of Soy Samples for Isolating Soy Protein–Enriched Fractions The soy protein–enriched fraction is currently used for different nutraceutical formulations. Preparation of soy protein isolate is usually accomplished by the stirring and/or soaking approach with hexane. ASE extraction offers a much better alternative because similar results are obtained more quickly, with reduced solvent usage (Table 3.5).

Copyright © 2004 AOCS Press

TABLE 3.3 Effect of Variation of Sample Size on the Percentage of Oil Recovery from Ground Soybeans Vial number

Vial weight

Expected sample size

Actual sample size (mg)

Vial + Oil

Oil

% Oil

Average % oil

SD

RSD (%)

1 2 3 4 5 6

21900.58 21751.64 22114.1 21936.47 21901.18 21852.14

20 20 20 20 20 20

20.12 20.06 20.45 20.22 20.73 20.48

21904.67 21755.86 22118.39 21940.73 21909.59 21856.41

4.09 4.22 4.29 4.26 4.41 4.27

20.33 21.04 20.98 21.07 21.27 20.85

20.92

0.32

6.6944

13 14 15 16 17 18

21879.49 21877.48 21993.32 22170.74 21847.3 21892.28

50 50 50 50 50 50

50.42 50.4 50.56 50.43 50.26 50.24

21889.76 21887.62 22003.76 22181.61 21857.48 21902.47

10.27 10.14 10.44 10.87 10.18 10.19

20.37 20.12 20.65 21.55 20.25 20.28

20.29

0.53

10.7537

1 2 3 4 5 6

22043.77 21982.11 21654.63 21922.27 22017 21646.67

75 75 75 75 75 75

75.5 75.31 74.95 75.26 75.24 75.23

22059.1 21997.59 21669.62 21937.41 22032.14 21661.94

15.33 15.48 14.99 15.14 15.14 15.27

20.3 20.56 20 20.12 20.12 20.3

20.23

0.2

4.046

7 8 9 10 11 12

21799.92 22152.1 21916.85 21838.45 21769.81 21932.01

100 100 100 100 100 100

100.67 100.02 100.16 100.77 100.82 100.5

21819.99 22172.21 21937.03 21858.88 21790.12 21952.14

20.07 20.11 20.18 20.43 20.31 20.13

19.94 20.11 20.15 20.27 20.14 20.03

20.11

0.11

2.2121

(Continued)

Copyright © 2004 AOCS Press

TABLE 3.3 (Cont.) Vial number

Vial weight

Expected sample size

Actual sample size (mg)

Vial + Oil

Oil

% Oil

13 14 15 16 17 18

21848.53 21847.81 21732.02 22029.23 21855.07 21984.08

250 250 250 250 250 250

250.32 250.51 250.24 250.52 250.45 250.32

21900.57 21900.02 21784.43 22081.57 21907.02 22036.51

52.04 52.21 52.41 52.34 51.95 52.43

19 20 21 22 23 24

21956.02 21887.51 22427.7 21715.06 22307.78 21969.67

500 500 500 500 500 500

500.17 500.62 500.68 500.53 500.08 500.35

22062.05 21993.18 22533.39 21820.67 22413.38 22075.50

25 26 27 28 29 30

21750.43 21799.76 21771.48 22017.33 21919.8 21796.91

1000 1000 1000 1000 1000 1000

1000.26 1000.17 1000.81 1000.55 1000.28 1000.38

31 32 33 34 35 36

22122.96 21917.83 21956.1 22179.94 21554.21 21686.5

2000 2000 2000 2000 2000 2000

2000.62 2000.51 2000.05 2000.76 2000.2 2000.09

Copyright © 2004 AOCS Press

Average % oil

SD

RSD (%)

20.79 20.84 20.94 20.89 20.74 20.95

20.86

0.08

1.6688

106.03 105.67 105.69 105.61 105.6 105.83

21.2 21.11 21.11 21.1 21.12 21.15

21.13

0.04

0.8452

21964.14 22013.21 21983.71 22229.52 22130.69 22008.87

213.71 213.45 212.23 212.19 210.89 211.96

21.37 21.34 21.21 21.21 21.08 21.19

21.23

0.11

2.3353

22550.82 22344.12 22380.74 22607.11 21980.69 22112.87

427.86 426.29 424.64 427.17 426.48 426.37

21.39 21.31 21.23 21.35 21.32 21.32

21.32

0.05

1.066

TABLE 3.4 Reproducibility Data: Percentage Oil Recovery from 125 Replicate Soy Samples Sample (n =125) Soya

Average % fat weight

SD

RSD (%)

20.38

0.32

1.56

aExperiments

were conducted with multiple operators with two instruments over a period of time. Extraction conditions were the same as those in Table 1.

Extraction of Fecal Sterols Determination of excreted fecal sterols is of prime importance to evaluate the efficacy of some of the pharmaceutical and nutraceutical products used for cholesterol reduction. The most common approach for quantifying the amount of fecal sterol is the saponification of the fecal matter followed by extraction with dichloromethane. Figure 3.3 shows the comparison of extraction of fecal sterol by a standard procedure vs. ASE procedure. The extracted sterols were isolated and analyzed by GC. Extraction of Fat from Infant Formula and Meat Samples Determination of total fat in powdered infant formula is performed using a solvent mixture of hexane/acetone (4:1) at 100 or 125°C. Three 5-min static cycles are used in the method. Milk-based formulas are prepared by mixing 1 g of sample with 3 g of HydromatrixTM (Varian Sample Preparation Products, Palo Alto, CA) before cell loading and extraction at 125°C. Soy-based and hydrolyzed milk–based formulas are mixed with wet Hydromatrix (3 g + 0.4 g water) and extracted at 100°C. ASE extraction of these samples can be performed without the aggressive alkaline pretreatments required by some methods. Extraction results were compared directly with results obtained using alkaline pretreatment followed by Mojonnier extraction with a mixture of petroleum ether, diethyl ether, and ethanol (AOAC Method 932.06). The results obtained for the ASE extracts averaged 99.7% of the Mojonnier results for six differ250

ppm

200 150

Tissue Grind

100

ASE

50 0

Alpha

Gamma

Delta

Fig. 3.2. Correlation of ASE and tissue grinder extractions for tocopherols in soybeans.

Copyright © 2004 AOCS Press

TABLE 3.5 Defatting of Soy Samples for Isolation of Soy Protein–Enriched Fractions

Amount of solvent per extraction, mL Number of extractions, n Total amount of solvent, mL Total extraction time, min Amount of fat left over, %

Classical stirring/soaking

ASE approacha

15 6 90 60 2

10 1 10 10 2

aNote: ASE extraction conditions were the same as those in Table 3.1 except for the number of cycles and static time.

ent formula types, including a certified reference material (SRM 1846) available from the National Institute of Standards and Testing. The fat content was determined gravimetrically, and verified by fatty acid methyl ester analysis. Fat extraction from a variety of meat samples is performed by mixing 3–4 g of a homogenous meat sample with 6 g of Hydromatrix. Moisture can be removed from the samples by drying in a microwave oven before extraction. Up to five samples can be dried at once in an 800-W oven at full power for 3 min. Samples are then extracted using either petroleum ether or hexane at 125°C, with two 2-min static cycles. Extraction results were compared with a 4-h Soxhlet extraction with petroleum ether (AOAC Method 90.39). Results for a variety of samples are shown in Table 3.6. The ASE method used here was shown to be useful for both low- and high-fat meat samples and saved considerable time compared with the traditional approach. 35000 30000

Control ASE

25000

Amount

20000 15000 10000 5000 0 Stigmasterol

Stigmasterol

Sitosterol

Fig. 3.3. Extraction and GC analysis of fecal sterols.

Copyright © 2004 AOCS Press

Sitosterol

TABLE 3.6 Extraction of Fat from Meat Samples by ASEa Sample (n = 3) Beef Chicken Ham Bacon Sausage

Average % fat weight

SD

Soxhlet

2.85 0.82 1.82 46.66 33.80

0.046 0.025 0.069 0.820 0.280

2.81 0.75 1.72 46.83 33.54

aConditions: 1-g samples, 125°C, 1500 psi, 6 min heat up, 10 min static, 60% flush, 60 s purge, hexane, 2 static cycles.

Other Applications Table 3.7 depicts the percentage of crude fat extracted, the SD, and RSD of the crude fat extracted from different matrices (potato chips, corn chips, cheese snacks, tortilla chips, snack chips) using ASE. Five replicate analyses were carried out with each sample. Table 3.8 shows the comparison of crude fat extracted from dog biscuits by ASE and Soxhlet procedures.

Summary Compared with conventional extraction times ranging from 4 to 48 h in length, ASE extractions are normally performed in 12 to 20 min. Although the decrease in extraction time is favorable for most laboratories in general, it can be critical for those industries in which laboratory data are used in feedback control of production cycles and manufacturing quality control. The volume of solvents used can be as much as 10–20 times less than traditional extraction methods. When factors such as safety and analyst exposure, as well as solvent purchase and disposal costs are considered, the benefits of ASE can be quite substantial for most laboratories. In a direct comparison with traditional extraction techniques, the recoveries generated by ASE normally equal or slightly exceed the comparative method. The ability to use the same liquid solvents used in traditional methods allows for rapid converTABLE 3.7 Extraction of Fat from Snack Food by ASEa Sample (n = 5)

Average % fat weight

SD

RSD (%)

Potato chips Corn chips Cheese snacks Tortilla chips Snack chips

34.0 32.8 33.3 21.5 19.2

0.11 0.08 0.17 0.07 0.10

0.33 0.25 0.34 0.34 0.53

aConditions: 3-g samples, 125°C, 1000 psi, 6 min heat up, 10 min static, 100% flush, 60 s purge, chloroform, 3 static cycles.

Copyright © 2004 AOCS Press

TABLE 3.8 Extraction of Fat from Dog Biscuits: Comparison of Results by Soxhlet and ASEa

aConditions:

Method

Solvent

Soxhlet ASEa

Petroleum ether Petroleum ether

Average % fat

SD

RSD (%)

8.80 9.12

0.50 0.15

5.7 1.6

7-g samples, 125°C, 1000 psi, 6 min heat up, 25 min static, 60% flush, 60 s purge, 1 static cycle.

sion to this technique, without much effort spent in methods development. Once an ASE method has been developed for a class of compounds, that same method can be applied successfully to a variety of matrix types without adjusting the extraction parameters. This lack of matrix dependency has allowed a very small set of standard methods to be applied to a large number of sample types. Because the entire extraction is carried out under nitrogen atmosphere, oxygen-sensitive species such as vitamin E, isoflavones, phospholipids, and unsaturated fatty acids are well preserved without the addition of an antioxidant to the extraction solvent. Further Reading Ezzell, J.L. (1999) Extraction Methods in Organic Analysis: Pressurised Fluid Extraction (PFE) in Organic Analysis, (Handley, A.J., ed.), pp. 146–164, Sheffield Academic Press, Sheffield, England. Richter, B.E. (1999) The Extraction of Analytes from Solid Samples Using Accelerated Solvent Extraction, LC/GC 17: 6S–32S. Richter, B.E., Jones, B.A., Ezzell J.L., Porter, N.L., Avdalovic, N., and Pohl, C. (1996) Accelerated Solvent Extraction: A Technique for Sample Preparation, Anal. Chem. 68: 1033–1039. Schäfer, K. (1998) Accelerated Solvent Extraction of Lipids for Determining the Fatty Acid Composition of Biological Material, Anal. Chim. Acta 358: 69–77. Luthria, D., and Cantrill, R. (2002) Evaluation of Five Different Methodologies for Determination of Oil Content in Ground Corn and Soybean Seeds, Inform 13: 893–894. Official Methods of Analysis AOAC International (1999) 16th edn. (Cunniff, P., ed.), Gaithersburg, MD.

Copyright © 2004 AOCS Press

Chapter 4

Evaluation of the Rapid, High-Temperature Extraction of Feeds, Foods, and Oilseeds by the ANKOMXT20 Fat Analyzer to Determine Crude Fat Content R.J. Komarek, A.R. Komarek, and B. Layton ANKOM Technology Corporation, Macedon, NY 14502

Abstract The process of extraction for the quantitative separation of fat/oil is the basis for the majority of official methods. The extraction process, which separates the sample into two fractions, permits two approaches to quantitative measurement. The analysis can be performed by either weighing the fat/oil fraction directly, or indirectly by measuring the loss of weight due to extraction. Acceleration of the extraction process has been achieved by elevating the temperature of the solvent. This chapter discusses a recently developed primary method called the Filter Bag Technique (FBT). This technique utilizes temperatures of up to twice the boiling point of petroleum ether to accelerate extraction. High sample throughputs are accomplished by batch processing of samples encapsulated in filter media formed in the shape of a bag. The extraction is performed automatically in an ANKOMXT20 Fat Analyzer, an instrument that can process 20 samples in 20–60 min. The fat/oil percentage is calculated indirectly from the loss of weight from the sample in the filter bag. Various studies related to the extraction and gravimetric measurements of these fractions are discussed in this chapter for both the conventional method and the FBT. The accuracy of the FBT depends on effective predrying and proper weighing of the sample. Studies of the conventional method suggest that samples containing polyunsaturated fatty acids are sensitive to oxidation particularly during the solvent evaporation step when the oil is heated in the presence of oxygen. Various studies of the ruggedness of the FBT indicate that the method is not sensitive to small changes in analytical conditions. The ruggedness of the method was confirmed in an experiment utilizing Youden’s Ruggedness Test. When the accuracy of the FBT was compared to that of the conventional method with a wide variety of samples (n = 22) in a regression analysis, the two methods were highly correlated (R2 = 0.9996). There was essentially no bias (–0.046 intercept) and no distortion over the range of the samples (slope 1.001). Two collaborative studies with laboratories from five countries provided similar evidence of the accuracy of the FBT. The second collaborative study, designed to evaluate the FBT as an AOCS official method, was conducted with 28 samples presented as 56 blind

Copyright © 2004 AOCS Press

duplicates. Twelve international collaborating laboratories used the FBT for the analysis, whereas three AOCS certified laboratories utilized the official methods. This study resulted in a similar highly significant R2 of 0.9990 compared with the official methods, with an intercept of 0.046 and a slope of 1.005. The average repeatability within laboratories was Sr= 0.31 and reproducibility among laboratories was SR = 0.46. These studies indicate that the FBT is an accurate and precise method capable of analyzing large quantities of samples in an efficient and automated fashion.

Introduction Knowledge of the fat content of food and feed, or the oil content in oilseeds is of critical importance when evaluating the value of these materials. The oil content of oilseeds determines their commercial value, whereas the fat content is important in gaining an understanding of the nutritional value and energy metabolism of a diet. Both fat and oil represent the fraction of lipids generally associated with triacylglycerides and compounds of similar solubility in nonpolar solvents. In this chapter, the terms “fat” and “oil” will be used interchangeably. The quantitative analysis of “Oil” as it is termed by American Oil Chemists’ Society (AOCS) (1) or “Crude Fat,” as designated by Association of Official Analytical Chemists (AOAC) (2), is based on separating the fat/oil from the sample matrix by extraction with nonpolar solvents. The amount of oil is determined either by directly weighing the extracted oil (Direct Method, AOAC Method 920.39a) or by measuring the loss of weight from the sample (Indirect Method, AOAC Method 920.39b, 948.22a). This process is described in the flow diagram in Figure 4.1. Each step in the process affects the accuracy and precision of the analysis. There are several critical drying, weighing, extraction, and evaporation steps. The process terminates with two fractions, i.e., the residue extracted by the solvent, for which the percentage can be calculated directly, and that portion of the sample not soluble in the solvent for which the percentage can be calculated indirectly. Because both values can be determined on the same sample, their agreement verifies the accuracy of the analysis. Nonpolar solvents such as diethyl ether, petroleum ether, and hexane dissolve fats and oils and leave behind proteins, carbohydrates, and other compounds insoluble in these solvents. This fractionation is the basis for most of the “Official” analytical methods established by AOCS, AOAC, International Organization for Standardization (ISO) (3), German Fat Science Society (DGF) (4), and Federation of Oils, Seeds and Fats Associations (FOSFA) (5). These methods utilize either the Soxhlet extraction apparatus, developed by Franz Von Soxhlet (6) in 1939, the Butt-type apparatus (2), or the Goldfisch apparatus (7). All of these methods boil the solvent and utilize the condensed solvent to extract the sample. The Soxhlet apparatus allows the sample chamber to fill and periodically siphon off into the boiling flask; the others simply allow the condensed solvent to pass through the sample as the solvent is refluxed. The

Copyright © 2004 AOCS Press

Fig. 4.1. A diagrammatic

representation of the analysis of fat/oil by solvent extraction.

sample is therefore extracted with solvent at a temperature below the boiling point of the liquid, requiring extraction times from 4 to 16 h. The rate of extraction has been increased by immersing the sample in the boiling solvent (8), thereby extracting the fat/oil at a higher temperature and reducing the extraction time. Further improvements in the kinetics of extraction have been achieved by performing the extraction in a sealed chamber at elevated pressures that permit extraction to be performed at temperatures well above the boiling point of the solvent [ANKOM (9) Dionex (10) and supercritical fluid extraction (11)]. This results in a further reduction in the extraction time. A recently developed method that utilizes high solvent temperatures in an automated batch process is being evaluated as an Official AOCS Method. This technique responds to the need for a rapid, efficient, high-volume process for the analysis of fats/ oils that is equivalent to a primary method using petroleum ether. The method is entitled, “Rapid Determination of Oil/Fat Utilizing High Temperature Solvent Extraction.”

Copyright © 2004 AOCS Press

This method is performed by the ANKOMXT20 Fat Analyzer (XT20) and can also be performed by the ANKOMXT10 Extractor (XT10) (9). Batch processing is accomplished by encapsulating each sample in a special filter medium, preserving its quantitative identity while performing the high temperature extraction of multiple samples in a common extraction chamber. The filter media is made in the shape of a bag and is heat sealed after the introduction of the sample. This method of analysis will be referred to as the Filter Bag Technique (FBT) and has the capability of high sample throughput (>200 sample/d). This chapter will discuss the background of the extraction process and the evaluation of the precision (reproducibility among different laboratories in a collaborative study), accuracy (comparison with standard methods), and ruggedness of the FBT in laboratory and interlaboratory collaborative studies.

Materials and Methods Conventional Method. The Goldfisch Method, conducted on a Labconco Goldfisch Fat Extraction Apparatus, was used in a number of studies as the conventional standard for comparison with the FBT (7). The apparatus functions essentially the same as the Butt-type apparatus, continually refluxing solvent over the sample during the extraction. The method can follow both paths, i.e., direct analysis and indirect analysis of fat/oil (Fig. 4.1). Extractions were performed over a 4- to 5-h period and the solvent was partially evaporated and recovered in a glass beaker. In earlier studies, the residual solvent (~10 mL) was evaporated above the hot plate on a holder in the apparatus. In subsequent studies, with sensitive samples, the residual solvent was evaporated on a steam bath under nitrogen. The analysis was conducted by weighing the sample in a tared thimble, drying the sample at 100°C for 3 h, and weighing it at ambient temperature from a desiccant pouch. The thimbles in these studies were made from the hydrophobic filter medium used for the filter bags. Typical cellulose thimbles are very hydroscopic and are difficult to weigh. The thimbles containing the samples were inserted into the apparatus and a tared glass beaker with 50 mL of petroleum ether was attached to each reflux unit. The cycle was started by turning on the hot plate. When the extraction was completed and the solvent evaporated, both the residual sample in the thimble and the fat/oil in the beaker were dried at 100ºC for 30 min, cooled to room temperature in a desiccator, and weighed. Both direct and indirect analyses were performed on the same sample as a check for accuracy. Filter Bag Technique. The FBT follows the path in Figure 4.1 of the indirect analysis and was performed in the XT20 (9). The sample was weighed in the filter bag, heat sealed, dried at 100ºC for ~3 h, cooled in a desiccant pouch, and weighed. Samples (n = 20) were placed in a carousel in the extraction chamber. The temperature (90ºC) and time of extraction, usually from 10 to 60 min, were selected and the instrument was started. The XT20 automatically processed the

Copyright © 2004 AOCS Press

samples in the following fashion: sealed and purged the chamber, inserted and heated the solvent, rotated the bag carousel, and emptied when the extraction time was complete. Solvent was then added for the first rinse, emptied after 3 min and refilled with fresh solvent for a second rinse. After the solvent was emptied, the residual solvent was evaporated and the chamber was purged with nitrogen. When attached to an ANKOMXT Recovery System, the instrument automatically distills and recycles the solvent. A similar process is performed by the ANKOMXT10 Extractor. The samples were then dried at 100ºC for 30 min, cooled to room temperature in a desiccant pouch, and weighed. The desiccant pouch was developed to more conveniently handle the filter bags during the weighing process. The pouches were made from resealable polyethylene bags containing desiccant and were used in all of the FBT studies. Filter bags were removed from the oven and placed directly in the desiccant pouch. The air was pressed out and the pouch was sealed. The samples rapidly equilibrated to room temperature and were effectively protected from ambient moisture by the limited head space in the pouch. The introduction of moist air during the removal of each bag was reduced by minimizing the size of the opening and pressing the pouch flat. Solvents. Although other solvents can be used, petroleum ether is the preferred solvent for the FBT because of its safety, cost, and ease of recycling. Petroleum ether was used in all the studies reported in this chapter. The boiling point range of commercial petroleum ether is specified by the supplier as 35–65ºC (12). The distribution of the solvent components over the temperature range was investigated in a fractional distillation study of both new and recycled petroleum ether (distilled to remove fat). Fractions were collected within 5°C increments from 36 to 80ºC. Sample Preparation. The objective of sample preparation is to provide a sample that accurately represents the “population” being studied and sufficiently disrupts the matrix to permit more efficient extraction. Meat samples were ground to a uniform consistency with a food processor and mixed thoroughly. For shipping convenience and sample uniformity, the meats in the international collaborative studies were dried for 3 h at 100°C and then ground in a cyclone mill to pass through a 2mm screen. The feed samples were ground in a cyclone mill to pass through a 1-mm screen and mixed thoroughly. The food samples were processed with a food processor or cyclone mill to produce a representative sample of uniform consistency. Soybean samples were first dried at 130°C for 30 min and then ground in a cyclone mill to pass through a 1-mm screen. Other oilseeds were ground in a cyclone mill to pass through a 1- or 2-mm screen, depending on the level of screen occlusion. The effects of grinding were demonstrated in a study with soybeans by processing them three ways. In the first treatment, soybeans were ground through a 2mm screen and extracted. In the second treatment they were processed according to the AOCS procedure (13) by first heating the soybeans in a 130°C oven for 30 min and then grinding through a 1-mm screen followed by an extraction. The third

Copyright © 2004 AOCS Press

treatment involved regrinding the soybean samples from the second treatment through a 1-mm screen and then extracting a second time. Conventional Method Weighing Procedures. The weighing procedure is critical to the gravimetric analysis of fats/oils. Accuracy of the analytical balance was verified and checked each day that weighing was performed. Accurate weighing of dried samples requires rapid processing directly from a desiccating environment, limiting exposure to moist ambient air. The glass beakers used in the conventional method were hydroscopic and can, under certain circumstances, carry a significant static charge. The effect of static charge was investigated in an experiment with samples of a pig diet. Samples were extracted for 4 h with petroleum ether, and the residual oil in beakers from six replicates was dried at 100°C for 30 min. After equilibration to room temperature in a desiccator, the beakers were weighed. The oil was then transferred with a small amount of petroleum ether to tared aluminum pans because they do not retain a static charge. After evaporation of the solvent, the samples in the aluminum pans were dried in the oven, equilibrated in a desiccator, and weighed. Oxidation. A study designed to evaluate the relative accuracy of the direct and indirect measurements was conducted on duplicate samples of ground beef, hot dogs, potato chips, high-energy horse diet, pig diet, corn, oats, and soybeans. Both direct and indirect determinations were performed on the same sample using the conventional method. The oil was evaporated using the holder on the Labconco apparatus, which holds the beaker above the hot plate. Due to the lack of agreement of the direct and indirect measurements with certain samples, studies were conducted to evaluate the role of oxidation in the elevated values of samples containing unsaturated lipids. An experiment was conducted with a corn sample that in previous studies had shown elevated direct values relative to the indirect values. A series of treatments were designed to first limit oxidation and then incrementally increase the opportunity for oxidation. It was observed that the bulk of the oil/fat was extracted at the beginning of the extraction period and that the oil/fat dissolved in the solvent was subjected to the boiling temperatures for hours during the refluxing of the solvent. Because the system was not anaerobic, there was a possibility that these conditions could present an opportunity for oxidation. In this experiment, extracted oil was removed from the apparatus during the extraction process in the first two treatments at 1.5 and 3.0 h, and continued with fresh solvent to complete the 5-h extraction. The remaining treatments were refluxed for 5 h without the removal of the first fraction. The last 10 mL of solvent was evaporated in several ways. For treatments 1, 2, and 3, solvent was evaporated on a steam bath with a nitrogen stream directed on the surface. In treatment 4, the solvent was evaporated on a steam bath without nitrogen. In treatment 5, solvent was evaporated above the hot plate in the Labconco holder. In treatment 6, the solvent was evaporated directly on the hot plate until all the solvent was observed to have been removed. After extraction, the samples for treatments 1 and 2 were dried in a desiccator and purged with nitrogen for 4 h.

Copyright © 2004 AOCS Press

The remaining treatments were dried in the oven at 100°C for 30 min. When samples were removed from the oven they were equilibrated to room temperature in a desiccator purged with nitrogen. The vacuum in the desiccator was returned to atmospheric pressure with nitrogen. Oil recovered from treatments 1, 5, and 6 was analyzed by thin-layer chromatography (TLC). Samples were chromatographed on silica gel plates with methylene chloride and visualized with bromo thymol blue (14). This procedure separates the sterols, triacylglycerides, and the less polar fractions. FBT Predrying. Before extraction, all samples were dried at 100°C for 3 h for both the conventional and the FBT methods. It is particularly important to remove the residual moisture from samples analyzed by the FBT because the moisture is removed during the extraction process, causing erroneously inflated values. A study was made of the effects of predrying on ground beef, a high-energy horse diet, corn, soybeans, and a pig diet for different periods of time and at different temperatures. Samples were weighed in a filter bag and dried at 100, 105, and 110°C. The samples were analyzed at intervals of 30 min up to 180 min and each treatment was replicated three times. FBT Sample Size. The effect of sample size (1.00, 0.50, and 0.25 g) on the precision of the analysis of six corn and three soybean samples was investigated in a study with the FBT. The samples, analyzed in triplicate, were finely ground and had a uniform consistency. Because of the sensitivity of the analytical balance (capable of weighing to 0.1 mg) and the relatively small tare weight of the filter bags (0.5 g), it was expected that weighing errors would be minimized and the variance associated with this study would be related to sample handling and sample homogeneity. FBT Extraction Temperature. Because elevated solvent temperatures enhance the extraction kinetics, the effects of extractions at three temperatures, 85, 90, and 95°C were studied. Samples were extracted in 15-min intervals over a 60-min period. The FBT analyses were conducted in triplicate on ground beef, soybeans, potato chips, and a high-energy horse diet. FBT Postextraction Drying. After extraction and solvent evaporation in the XT20, samples can absorb weight from exposure to ambient moisture and can contain traces of solvent that must be removed. Postdrying periods of 10 and 20 min were studied. Samples were weighed directly upon removal from the XT20 and then placed in an oven at 100°C for two consecutive 10-min periods and weighed after cooling in a desiccant pouch. Samples (n = 10) were analyzed in duplicate (oat meal, brownie mix, crackers, dog food, pig diet, ham, turkey, corn, soybeans, and canola). A second study was conducted to determine the effect of drying at 100°C for intervals of 20, 40, 60, and 80 min. The FBT analyses were conducted in triplicate on soybeans, canola, potato chips, and horse feed.

Copyright © 2004 AOCS Press

FBT Youden’s Ruggedness Test. Youden’s Ruggedness Test (15) was performed to evaluate seven variables in the method and the effect of modest changes in these variables. The variables were sample size (0.8–0.9 g vs. 1.2–1.3 g), predry time (2.5 vs. 3.0 h), predry temperature (98 vs. 102°C), extraction time (25 vs. 35 min), extraction temperature (89 vs. 94°C), postdry time (25 vs. 35 min), and postdry temperature (98 vs. 102°C). Nine sample types were analyzed in triplicate, including ground beef, chicken thighs, hot dogs, corn, soybeans, potato chips, cattle feed, poultry feed, and dog food. Comparison of the FBT with the Conventional Method. The relative accuracy of the FBT was evaluated by comparing the results of this method with those of the conventional method. Samples (n = 22) were analyzed by both methods; each was replicated five times to compare the relative precision. The samples included a range of samples encompassing meats, grains and oilseeds, feeds and foods. The data were analyzed by Regression Analysis. Multilaboratory FBT Study. A study was designed to evaluate the precision and accuracy of the FBT by analyzing five samples in quadruplicate using the same protocol in 13 laboratories and completing the analysis within a 3-wk period. This study provided an opportunity for the laboratories to familiarize themselves with the FBT protocol to be used in the more extensive collaborative study. The laboratories were located in the United States, Canada, and Europe. The samples used were ground beef, cheese curls, soybeans, corn, and a horse diet. The conventional analysis was performed by ANKOM Technology. FBT Collaborative Study. A collaborative study, performed in conjunction with AOCS, was designed to evaluate the precision and accuracy of the FBT with a wide variety of samples that represented foods, feeds, meats, and oilseeds. Samples (n = 28) were sent to 12 laboratories in the United States, Canada, and Europe in the form of 56 blind duplicates. Each laboratory was given a detailed protocol and had an opportunity to become familiar with the method in a preliminary study. These samples were also analyzed by three AOCS Certified Laboratories using the relevant official methods.

Results and Discussion Reusing Solvent. The results of the solvent fractionation study of petroleum ether (Fig. 4.2) indicated that the majority of the solvent (~70%) distilled in the range of 36–40°C with no other fraction >8%. The distributions of all of the fractions were similar for both recycled and purchased solvent. This study indicates that petroleum ether can be recycled without significantly changing the distribution of the solvent components. Sample Matrix Disruption. Fats and oils that are not hindered by the sample matrix or by various types of binding rapidly dissolve in fat solvents. Oils trapped

Copyright © 2004 AOCS Press

Start

36–40

41–46

47–50

51–55

56–60

61–65

66–70

71–75

76–78

79–80

Boiling point fractions (°C)

Fig. 4.2. The boiling point distribution of new reagent grade and recycled pretroleum

ether. The recycled petroleum ether was recovered by distilling waste solvent from fat extractions.

in plant cell matrices are particularly difficult to extract due to the cell wall. This microstructure can act as a semipermeable membrane where larger molecules have limited access to exit the structure even though the smaller solvent molecule can penetrate the structure. Plant matrices are difficult to disrupt on a cellular basis, and this has led to the development of extensive grinding procedures. The grinding and regrinding procedures required in the AOCS and FOSFA methods for certain oilseed samples attest to the difficulty of preparing these samples for analysis. The grinding study with soybeans illustrates the problem of sample preparation for complete extraction of the oil (Fig. 4.3). The drying of the whole soybean at 130°C for 60 min before grinding improved the yield by ~3%, whereas regrinding after extraction improved the recovery by another 2%. In both treatments, it would be expected that more extensive fracturing of the cell wall had occurred, enabling greater extraction of oil. Unfortunately, the oven treatment and extensive grinding

Copyright © 2004 AOCS Press

Treatment 1

Treatment 2

Treatment 3

Fig. 4.3. The effect of three grinding treatments on the quantity of oil extracted from

soybeans. Soybeans were ground through a cyclone mill before extraction (Treatment 1), ground after drying at 130°C for 1 h before extraction (Treatment 2), and ground after drying at 130°C, extracted, ground again, and reextracted (Treatment 3).

increase the chances of oxidation of the unsaturated fatty acids in the soy lipids, potentially increasing the weight of the oil extracted. However, there may be sufficient protection within the matrix afforded by tocopherols and other antioxidants to retard this oxidation. Weighing Errors. It is necessary in all gravimetric procedures to pay particular attention to factors that affect the weighing process. When samples are oven dried, water molecules are driven off binding sites on the sample and on the sample container. These active sites are rapidly refilled by ambient moisture if given the opportunity. Desiccators provide protection but care must be taken not to compromise this protection and to limit the exposure time during weighing. When glass vessels are dried, they can hold a static charge that can interfere with the weighing process. This phenomenon is illustrated in Figure 4.4 with the conventional analysis of a pig diet. The erratic weights of five glass beakers containing the residual oil from replicate extractions were greatly improved by eliminating the static charge. This was accomplished by transferring the oil sample to aluminum weighing pans and reweighing. The SD of the oil value was reduced from 0.33 to 0.05. This effect can also be controlled by using an ionizing source to dissipate the static charge on the glass beakers.

Copyright © 2004 AOCS Press

Fig. 4.4. The effect of static charge on glass beakers was examined in five samples of a pig diet by first weighing the fat in the beaker and then transferring the fat to aluminum pans and weighing the fat again.

Oxidation. During a series of experiments with the conventional method, it was found that for certain samples, such as hot dogs, ground beef, and potato chips, the direct measurements of fat (the weights of the fat recovered) were in good agreement with the indirect measurements (weight lost due to extraction) (Fig. 4.5). By contrast, Figure 4.5 shows that the direct measurements of fat/oil were considerably higher than the indirect measurements in oats, corn, soybeans, and a pig diet. The distinguishing characteristics of this group include their plant origin and higher concentrations of polyunsaturated fatty acids compared with the meat and potato chips group. Similar studies with corn and oats also showed higher values for the direct compared with the indirect analysis when solvent was evaporated on the Labconco holder. Oxidation increases the weight of the oil (16), thereby increasing the direct measurement of the oil. The extracted sample is not subject to the same effect, and no distortion of the indirect measurement would be expected due to oxidation. In the experiment designed to investigate variables in the method that would enhance or avoid oxidation (Fig. 4.6), the indirect measurements of the oil content were in excellent agreement across all six treatments. This was not true for the direct measurement of the oil. Incremental changes in the time the oil was boiled in

Copyright © 2004 AOCS Press

Indirect

Direct

Fig. 4.5. Comparison of the direct and indirect analysis of samples containing unsaturated oils (a pig diet, corn, oats, and soybeans) with samples containing predominantly saturated fat (ground beef, hot dogs, and potato chips) (n = 2).

the solvent during the reflux resulted in slight but inconclusive increases in the direct value (Treatments 1–3). In Treatment 5, the solvent was evaporated using the Labconco holder, which positions the beaker above the heater and allows the temperature of the oil to rise above 100°C. The direct measurement of oil yielded a value that was 4% higher that the indirect value. In Treatment 6, in which the solvent was evaporated on the hot plate in the Labconco, the oil was subjected to temperatures of 200°C for ~1 min. This resulted in a direct value that was lower than the indirect value. In the TLC chromatogram of Treatments 1, 5, and 6 (Fig. 4.7), the triacylglyceride spot (Rf 0.45) was the dominant spot for Treatment 1 where the direct value closely agreed with the indirect value. In Treatment 5, with an increase in oil weight, a large spot (Rf 0.67) developed above the triacylglyceride spot (Rf 0.45). Only a trace of that spot (Rf 0.67) was detected in Treatment 6 in which the oil received the highest heat treatment and had the lowest weight. Presumably, hydroperoxides were formed in both Treatments 5 and 6; some of their breakdown products (aldehydes and carbonyls) were observed on the chro-

Copyright © 2004 AOCS Press

Post Drying Solvent Evap. Reflux Time Treatment

Fig. 4.6. Effect of a progressive increase in oxidative conditions on the weight of oil

recovered from a corn sample.

matogram for Treatment 5 but were essentially absent for Treatment 6. Although the results of this experiment may represent a special case, they support the conclusion that when the samples were exposed to air at elevated temperatures, oxidative formation of hydroperoxides occurred. In Treatment 5, the hydroperoxides decomposed but were not volatilized, whereas in Treatment 6, the breakdown products were volatilized (16) by the higher temperatures. These experiments indicate that

Copyright © 2004 AOCS Press

Solvent front

Origin

1

5

6

Fig. 4.7. TLC chromatogram showing the separation of oil samples with different heat treatments. Sample 1 was analyzed under the mildest conditions; Sample 5 was heated on the Labconco holder and Sample 6 was heated directly on the hot plate. Triacylglycerides migrated to an R f of 0.45 and suspected oxidation degradation products migrated to an R f of 0.67. Samples were separated on silica gel G TLC plates with methylene chloride.

care has to be taken to avoid oxidation when measuring the oil fraction directly, particularly with plant samples containing significant quantities of polyunsaturates. Sample Predrying. During the refinement of the FBT, critical steps in the protocol were investigated and optimized. The requirements of predrying were investigated for a variety of sample types. A ground beef sample provides an example (Fig. 4.8) of the relationship of moisture removal and the fat percentage. The percentage of dry matter decreased for the first 120 min and then leveled off. The percentage of fat followed the same pattern starting off high and leveling off after 120 min. The moisture that was not removed in the oven was removed during the extraction and resulted in elevated fat values. The three oven temperatures (100, 105, and 110°C) produced similar results with ground beef. The same experiment with a high-energy horse diet (Fig. 4.9) indicated that the lipids in this diet were sensitive to temperature and that time in the oven increased the effect. The horse diet, starting with 1 g/mL and employed several depressurization steps during the extraction process. They suggested that the freezing-thawing mechanism that occurred in the extraction vessel resulted in the modification of the protein structure, thus leading to the destruction of lipid-protein aggregates. They theorized that the PL become readily available to the SCCO2, and thus more extractable. Postextraction analysis and identification of lipid classes included HPLC-ELSD and TLC (38). Other lipids may be contained within cell wall components such as cellulose and not be accessible to SCCO2. Studies have shown that the SFE may cause chemical alterations when sufficient quantities of oil are removed. This alteration allows subsequent extractions to remove additional oil (39). During sample preparation for oilseeds, such as grinding, the surface-to-volume ratio is increased, thus exposing more oil to the extraction solvent. During the process, seeds are unintentionally milled to different particle sizes and the amount of oil extracted from each fraction (based on size) produces different results (40,41). These studies all indicate that proper sample preparation before SFE has an effect on the quantity of analyte recovered. In 1996, the method using SFE was approved by the AOCS. Method Am3-96, SFE Determination of Oil in Oilseeds, was subsequently adopted by AOAC in 2000 as method number 999.02. The method is based on gravimetric analysis from a set of oilseeds determined to be representative of the oilseed industry and encompasses soybeans, canola, sunflower, safflower, and sunflower. This method references AOCS sample preparation methods as determined for each type of oilseed and allows the user to choose between two variations of SFE (CO2 alone or CO2 with a 15% EtOH modifier). The SFE parameters are 100°C, 7500 psi, with or without a 15% modifier at 2 mL/min with a total extraction time of 30 min (CO2 only) or 45 min (CO2 + 15% EtOH) (42). The application of SFE for the determination of total fat in meats was investigated by numerous researchers in response to the Nutrition and Labeling Education Act (NLEA) of 1990, which defined fat as the sum of all fatty acids obtained from a total lipid extract expressed as triglycerides (43). Snyder et al. (44) investigated the possibility of determining fat content for the purpose of nutritional labeling of meats and meat products using SFE (50°C, 12.2 MPa with