Fluid inclusions 9781501508271, 9780939950164

Citation preview

REVIEWS in MINERALOGY Volume 12

FLUID INCLUSIONS

An introduction to studies of all types of fluid inclusions, gas, liquid, or melt, trapped in materials from earth and space, and their application to the understanding of geologic processes.

EDWIN ROEDDER

UNITED STATES GEOLOGICAL SURVEY RESTON, VIRGINIA 22092

Series Editor: PAUL H. RIBBE

Department of Geological Sciences Virginia Polytechnic Institute & State University Blacksburg, Virginia 24061

MINERALOGICAL SOCIETY OF AMERICA

COPYRIGHT

1984

M I N E R A L O G I C A L SOCIETY of A M E R I C A PRINTED BY BookCrafters, Inc. Chelsea, Michigan 48118

REVIEWS in MINERALOGY (Formerly: S H O R T C O U R S E NOTES) ISSN 0275-0279 Volume 12: FLUID INCLUSIONS ISBN 0-939950-16-2

ADDITIONAL COPIES of this volume as well as those listed below may be obtained at moderate cost from:

Mineralogical Society of America 2000 FLORIDA AVENUE, N.W. WASHINGTON, D. C. 20009 Volume 1

SULFIDE MINERALOGY, P.H. Ribbe, editor

2

FELDSPAR MINERALOGY, P.H. Ribbe, editor

3

OXIDE MINERALS, Douglas Rumble III, editor

4

MINERALOGY and GEOLOGY of NATURAL ZEOLITES, F.A. Mumpton, editor

5

ORTHOSILICATES, P.H. Ribbe, editor

6

MARINE MINERALS, R.G. Burns, editor

(1974) (1975; revised 1983) (1976) (1977)

(1980; revised 1982) (1979)

(1980)

284 p. 362 p. 502 p. 232 p. 450 p. 380 p. 525 p.

7

PYROXENES, C.T. Prewitt, editor

8

KINETICS of GE0CHEMICAL PROCESSES, A. C. Lasaga and R.J. Kirkpatrick, editors (1981)

391 p.

9A

AMPHIBOLES and OTHER HYDROUS PYRIBOLES - MINERALOGY, D.R. Veblen, editor (1981)

372 p.

9B

AMPHIBOLES: PETROLOGY and EXPERIMENTAL PHASE RELATIONS, D.R. Veblen and P.H. Ribbe, editors (1982)

390 p.

10

CHARACTERIZATION of METAMORPHISM through MINERAL EQUILIBRIA, J.M. Ferry, editor (1982)

397 p.

11

CARBONATES,

394 p.

R.J. Reeder, editor

M

(1983)

Reviews in Mineralogy — Volume 12

- FLUID

INCLUSIONS

FOREWORD "FLUID INCLUSIONS"

is the first single-author contribution to be published by the

Mineralogical Society of America in its decade-old series, begun as SHORT NOTES

but since 1980 issued under the title REVIEWS

COURSE

in MINERALOGY.

This is

Volume 12, and it is the largest single volume (644 pages) with the most references (2000), diagrams (131), and photographs (387). Edwin Roedder, who assembled the encyclopedic work after more than thirty years' study of fluid inclusions in minerals, is eminently qualified for the undertaking. He has been the editor and primary contributor to FLUID PROCEEDINGS

INCLUSION

RESEARCH

-

of C O F F I, since its inception in 1968; these volumes provide

citations and abstracts in English of the 800-900 items published each year on fluid inclusions. In 1965 Roedder edited an English translation of a massive work by N.P. Yermakov and others [see Ermakov, 1950], and in 1972 he published COMPOSITION FLUID INCLUSIONS,

of

U.S. Geological Survey Professional Paper 440JJ. As retiring

President of the Mineralogical Society of America, he delivered his presidential address, The Fluids in Salt, exactly six months ago at M.S.A.'s annual meeting in Indianapolis.* The organization of this book is somewhat different from previous volumes in that the Table of Contents lists only chapter titles: detailed contents are given on the first page of each chapter. Other unique features (made possible by the fact that there was no time pressure from an impending Short Course) are the Subject Index and the Locality Index at the end of Chapter 19, followed by an extensively cross-referenced bibliography. Titles of other REVIEWS in MINERALOGY

are listed on the opposite page. Paul H. Ribbe Series Editor Blacksburg, VA

l/V/84

Note for the Second Printing:

ERRATA

A substantial number of corrections have been made in this the second printing of FLUID INCLUSIONS.

For those who own a copy of the first printing, a list of the

errata is given on pages 645 and 646 of this volume to facilitate locating the changes. * It was announced in late 1985 that Edwin Roedder will receive from the Mineralogical Society of America its highest award, the Roebling Medal, at its annual meeting in 1986.

iii

CONTENTS COPYRIGHT; LIST OF PUBLICATIONS FOREWORD PREFACE CHAPTER 1* I n t r o d u c t i o n to f l u i d i n c l u s i o n s 2 The o r i g i n of i n c l u s i o n s 3 Changes i n i n c l u s i o n s a f t e r t r a p p i n g 4 Nondestructive methods of determination of i n c l u s i o n composition 5 D e s t r u c t i v e methods of determination of i n c l u s i o n composition 6 I n c l u s i o n sample s e l e c t i o n , p r e p a r a t i o n , petrography, and photography 7 I n c l u s i o n measurements - - h e a t i n g , c o o l i n g , d e c r e p i t a t i o n , and c r u s h i n g 8 I n t e r p r e t a t i o n and u t i l i z a t i o n of i n c l u s i o n measurements — compositional data on l i q u i d and gas i n c l u s i o n s 9 I n t e r p r e t a t i o n and u t i l i z a t i o n of i n c l u s i o n measurements - temperature, pressure and d e n s i t y at t r a p p i n g 10 I n t e r p r e t a t i o n and u t i l i z a t i o n of i n c l u s i o n measurements - metastability 11 Sedimentary environments 12 Low- to medium-grade metamorphic environments 13 Medium- t o high-grade metamorphic environments 14 I n t r u s i v e rock and pegmatitic environments 15 Ore d e p o s i t i o n environments 16 E x t r u s i v e rock and v o l c a n i c environments 17 Upper mantle environments 18 E x t r a t e r r e s t r i a l environments 19 Future of i n c l u s i o n s t u d i e s SUBJECT INDEX LOCALITY INDEX REFERENCES *A chapter o u t l i n e i s given on the i n i t i a l

Page ii iii v 1 11 47 79 109 149 181 221 251 291 305 337 361 381 413 473 503 533 571 585 591 595

page of each chapter.

ABBREVIATIONS USED IN THIS BOOK T P V Th, Ph T t , Pt Td, Pd Tm Te Tn XCO2 kbar atm 6

-

equiv. Salinity -

Temperature Pressure Volume Temperature and pressure of homogenization** Temperature and pressure of t r a p p i n g * * Temperature and pressure of d e c r e p i t a t i o n * * Temperature of m e l t i n g * * Temperature of e u t e c t i c * * Temperature of n u c l e a t i o n * * Mole f r a c t i o n CO2 ( s i m i l a r l y , XCH4, XH2O, e t c . ) K i l o b a r s pressure Atmospheres pressure In i s o t o p i c r a t i o s , the d i f f e r e n c e , stated i n p a r t s per thousand (per m i l ) , between the unknown and a standard ( u s u a l l y SMOW f o r H and 0, and PDB f o r carbon), Equivalent Expressed as wt % NaCl e q u i v . ; the concentration of NaCl needed to achieve the same depression of the f r e e z i n g point.

* * F o r d e t a i l s on d e f i n i t i o n s and m o d i f i c a t i o n s , see Table 7 - 3 , page 198. iv

PREFACE and ACKNOWLEDGMENTS This book has been written mainly to help the newcomer in f l u i d - i n c l u s i o n work learn how to use f l u i d inclusions and to avoid many of the p i t f a l l s and blind a l l e y s that beset anyone s t a r t i n g in a new f i e l d of research. Of course, i t i s impossible to avoid all such diversions. However, too often, writers of s c i e n t i f i c papers (and some editors) seem to believe that i t i s undesirable or even demeaning to report experimental details and the various problems that had to be overcome in the work. I do not agree with t h i s approach. Why should subsequent workers be frustrated and waste much time solving problems that others have already solved? Give them the benefit of previous experience so that they can get on with new work; in so doing, they will encounter enough new problems of their own. One d i f f i c u l t y in presenting a subject such as f l u i d inclusions i s the surp r i s i n g degree to which the chapters are interrelated. I have t r i e d to s t r i k e an appropriate compromise between repeated referral to other chapters and excessive repetition, because everything cannot be put into logical sequence without redundancy. Chapters 11-18 attempt to discuss the many applications of f l u i d inclusions to the study of and understanding of geologic processes and the geologic environments in which they acted. For the reader's convenience, I have categorized all environments from which f l u i d inclusions have been studied into these eight chapters. The arbitrary d i v i d i n g l i n e s between such environments are never sharp, nor generally acceptable, p a r t i c u l a r l y i f more than one geologist i s asked, so I hope the reader w i l l forgive me i f my semantics disagree with his or hers; the differences are of no real consequence to the points being made. Although some of the data and ideas in t h i s book are new, other parts come from e a r l i e r papers of my own or from those on which I have been a coauthor. I make no apology for t h i s , as I see no point in using quotation marks or trying to rephrase one's own words. Only about a third of the text i s taken more-orl e s s d i r e c t l y from these e a r l i e r works (with modifications). S i m i l a r l y , many but not a l l the photomicrographs have been used e a r l i e r . I n the choice of examples, I have leaned heavily on those from my own experience and papers, mainly because t h i s procedure i s l e s s prone to errors from misquotation, and because I have a l l the negatives of the photomicrographs I made in these studies. In a petrography c l a s s , in 1939, my teacher, Dr. Donald M. Fraser, showed me some inclusions in Precambrian quartzite in which the bubbles were rapidly bouncing around in their tiny c e l l s , as they presumably had been for more than a b i l l i o n y e a r s . This so intrigued me that after completing graduate work (more than 30 years ago) I started studying f l u i d i n c l u s i o n s . I hope that some aspect of t h i s book may, in the same way, intrigue others. I have t r i e d to help the reader by including chapter outlines and a det a i l e d index, and in the References I have l i s t e d the page(s) where each item i s cited, as t h i s also can help the reader to become acquainted with the rather large and scattered l i t e r a t u r e and some of i t s applications. The overall organization i s somewhat of an adaptation of the news reporter's outline — "who, what, when, where, and why": what kinds of information inclusions provide, when and where i n c l u s i o n s form, how they change, how to prepare material and make microthermometric measurements!/, how to interpret these data, and then what has been found in applications of f l u i d - i n c l u s i o n studies to each of a series of d i f f e r e n t geologic environments.

i / The use of trade names In t h i s publication I s for I d e n t i f i c a t i o n only and does not imply endorsement by the U.S. Geological Survey.

v

As i n most developing areas of s c i e n c e , numerous erroneous concepts, procedures, and statements have been published ( i n c l u d i n g some of my own). I have a f i l e of several hundred of these e r r o r s , but most do not merit a t t e n t i o n and hence are not mentioned i n t h i s volume, except where they may have led to more than occasional c o n f u s i o n or misunderstanding by l a t e r workers. Caveat emptor. I have been helped by many people in the preparation of t h i s volume. Indiv i d u a l s from a l l over the world have sent data and r e p r i n t s , and others have s u p p l i e d o r i g i n a l p r i n t s of some of the published i l l u s t r a t i o n s that are used here. Most important, over the y e a r s I have p r o f i t e d from d i s c u s s i o n s with many of my c o l l e a g u e s ; hence, the acknowledgements f o r t h i s book could well be a t a b u l a t i o n of a l l those people. I am p a r t i c u l a r l y indebted to my coauthors where I have used our j o i n t e f f o r t s i n t h i s book; e . g . , R . J . Bodnar was coauthor of a paper used as the major source f o r part of Chapter 9. In 1981 the M i n e r a l o g i c a l A s s o c i a t i o n of Canada held a wel1-attended short course on F l u i d Inclusions: A p p l i c a t i o n s to P e t r o l o g y , and i s s u e d a handbook ( H o l l i s t e r and Crawford, 1981) that covers some of the same material presented here. Although the planning of present volume l a r g e l y preceded my a c t i v i t y with the short course, I p r o f i t e d g r e a t l y from the experience gained i n the preparation and p r e s e n t a t i o n of my c o n t r i b u t i o n s to the c o u r s e , and from the many i n t e r a c t i o n s with other i n c l u s i o n i s t s i n v o l v e d . Many o t h e r s , too numerous to l i s t , have helped i n a v a r i e t y of ways. S p e c i a l mention i s needed, however, f o r H.E. B e l k i n and E . L . L i b e l o , who s k i l l f u l l y made many of the photographic p r i n t s used here, and, with the cooperation of the USGS l i b r a r y s t a f f , located many obscure references. Several i n d i v i d u a l s helped with the typography of some of the f i r s t chapt e r s , but s p e c i a l thanks are due N. Teed f o r most of t h i s work. I am a l s o thankful t o A. Sangree f o r e d i t i n g , and to my wife Kathleen f o r frequent e d i t o r i a l c o n s u l t a t i o n s — as well as f o r her c o n s i d e r a b l e patience. Parts of the manuscript were reviewed by too many i n d i v i d u a l s to l i s t comp l e t e l y , but the f o l l o w i n g have c o n t r i b u t e d thoughtful reviews of one or more chapters: E.C. Alexander, J r . , A.T. Anderson, J r . , L . D . Ashwal, C.E. B a r k e r , M.D. Barton, P.B. Barton, J r . , H.E. B e l k i n , S . C . Bergman, P . R . L . Browne, I . - M . Chou, M.L. Crawford, C.G. Cunningham, P . J . Eadington, N.K. F o l e y , J . Guha, D.M. H a r r i s , I . Haapala, 0 . 0 . Hayba, P. Heald-Wetlaufer, R.T. H e l z , R. Henley, L . S . H o l l i s t e r , S . S . Howe, R. Kreulen, S . D . Ludington, P.T. L y t t l e , H.O.A. Meyer, S . Morasse, R. P e t r o v i c h , N.M. R a t c l i f f , T . J . Reynolds, W.I. Rose, J r . , L . P . Rowan, R . L . Rudnick, R.O. Rye, E . T . C . Spooner, T.G. Theodore, G.C. Ulmer, and R.W.T. W i l k i n s . I owe a s p e c i a l debt to J.W. Hedenquist and R . J . Bodnar, who reviewed a l l of them. These colleagues should not be held accountable f o r f a i l i n g to catch any of the many e r r o r s and omissions that I am sure remain. I would appreciate having a l l such shortcomings c a l l e d to my attention. I am e s p e c i a l l y g r a t e f u l , and the members of the M i n e r a l o g i c a l Society of America should be equally g r a t e f u l , f o r the c o n t i n u i n g hard work and s e l f l e s s c o n t r i b u t i o n of P.H. Ribbe, E d i t o r of "Reviews in M i n e r a l o g y " . Only authors i n the s e r i e s r e a l l y know how large and t h a n k l e s s a task he has. Edwin Roedder 959 U.S. Geological Reston, VA 22092 17/11/84

vi

Survey

Chapter 1 INTRODUCTION to FLUID INCLUSIONS CONTENTS

Page "T

GENERAL NATURE OF FLUID INCLUSIONS BACKGROUND

3

Historical Perspectives

3

Literature Sources and Summaries Literature o r i g i n a l l y published in E n g l i s h Books from the USSR Symposium volumes Abstract volumes

4 4 4 5 6

DATA OBTAINABLE FROM FLUID INCLUSIONS

6

Temperature

6

Pressure

7

Density

7

Composition

8

APPLICATIONS OF FLUID INCLUSIONS

9

GENERAL NATURE

FLUID INCLUSIONS

"A F l u i d I n c l u s i

Speaks I t s Mind" You want to know my past. Was I homogeneous? Was I heterogeneous? These are the questions I see in your eyes.

Do you l i k e me? Why do you look At t h i s tiny thing That I am? I see your eyes And you want to sing to me To discover my soul

Can you hear me? I am trying to say something. I know you can love me. And when you do I will t e l l i t to you a l l .

D o n ' t t e l l me That you want to know me J u s t to learn my secrets. Caught you there With your soul bare.

Jayanta Guha Chicoutimi, March 6, 1982

With the notable exception of those c r y s t a l s in metamorphic samples that have grown in the s o l i d state, a l l c r y s t a l s in a l l t e r r e s t r i a l and extraterrest r i a l samples have grown from some kind of f l u i d . The minerals of many meteori t i c and lunar samples, and of t e r r e s t r i a l igneous rocks, have grown from f l u i d s i l i c a t e melts; some of the minerals from these sample groups have grown in the presence of an additional low-density vapor phase. New c r y s t a l s in many sedimentary and some metamorphic rocks, and in almost a l l ore deposits, formed from an aqueous f l u i d containing various solutes. After c r y s t a l l i z a t i o n , the minera l s of p r a c t i c a l l y a l l t e r r e s t r i a l sedimentary, metamorphic, and igneous rocks 1

have been fractured one or more times, and the fractures have healed in the presence of liquid or gaseous fluids. During these processes of crystal growth and fracture healing, small quantities of the surrounding fluid medium are commonly trapped as fluid inclusions in the host crystal. If solid materials are also present in the fluid, they may become enclosed by the growing crystal as solid inclusions; often some fluid is also trapped .as the host crystal surrounds the solid inclusion. In much of the inclusion literature, the term fluid inclusion has been used only for those inclusions that trapped a fluid that remains in large part fluid at surface temperature, and the term "melt inclusion" has been used for those that have become essentially solid at surface temperature. As will be seen, however, a continuum exists between these two extremes, and because the processes involved in trapping, the methods of study, and the problems in interpretation of the data are basically the same for all such inclusions, I use the term fluid inclusion for all. The term is thus used here to refer to the state of the trapped material at the time of trapping (i.e., a liquid or a gas) and not to its condition as we observe it now. During cooling to surface temperatures, the fluid in many inclusions has formed crystals, and the fluid in inclusions that have trapped a silicate melt may have cooled to form a glass. The term liquid inclusion (or gaseous inclusion) can be used when it is necessary to differentiate between melt inclusions and those that remain fluid at room temperature. Further ambiguity can be eliminated by the use of compositional terms such as gaseous, aqueous, CO2, oil, or melt inclusion. Minerals from certain metamorphic rocks, particularly those recrystallized in a water-deficient environment (Yoder, 1955), may be free of all fluid inclusions (gas, liquid, or melt); if any inclusions are present, they are most apt to consist of CO2. Most materials believed to have formed deep in the earth, such as diamonds and the minerals of some eclogites, are essentially free of fluid inclusions. The most notable exceptions are the CO2 inclusions in the minerals in the dunitic and peridotitic xenoliths or "nodules" brought to the surface with many alkalic basalts (Roedder, 1965a). Liquid inclusions have even been reported in certain meteorites (Yasinskaya, 1967; Fieni et al., 1978; Warner et al., 1983). Inclusions are seldom larger than 1 mm and commonly go unnoticed. On the other hand, museum specimens that have single inclusions containing tens or even hundreds of milliliters of fluid are known (Hidden, 1882; Prikazchikov, 1959; Prikazchikov et al., 1964; Rankin and Greenaway, 1978). The number of inclusions in any given sample is usually related inversely to inclusion sizel', for very small inclusions are much more abundant than large ones. In most samples, inclusions in the range 1 to 10 ym outnumber all inclusions larger than 10 ym by a factor of 10 or even 100, and electron microscopy has revealed large numbers of inclusions as small as -0.02 ym (2 x 10" 6 cm; Green and Radcliffe, 1975). Presumably, a size continuum exists down to single water molecules (~2 x 10"8 cm) trapped along grain boundary dislocations or bound in the structure (Spear and Selverstone, 1983). Ordinary white quartz or calcite is usually white from the presence of perhaps 109 inclusions/cm^. Most studies of fluid inclusions are made on populations in the range 10-100 ym, although some have been made on inclusions averaging but radiogenic Ar formed by the decay of 40K is pure « A r . Bakhanova et al. (1976; also Naydenov et al., 1978) have examined the Ar isotopic ratios from an Au deposit in Kazakhstan, USSR, and have found that inclusions in the richest ores were highest in 36/\r> suggesting the presence of more surface waters containing dissolved light atmospheric Ar. In the future, we may also expect to see the use of ^He/^He in a somewhat similar manner. Here, the heavier isotope is also from radioactive decay, but the light isotope is derived mainly from deeper in the earth and comes, at least in part, from original nucleosynthesis. The large mass difference introduces new problems, however, as a result of significant differences in isotopic diffusion rates. Beryl has been found to contain excess ^He and ^ k r , i.e., ^He and ^ A r in excess of that from radioactive decay since crystallization (Damon and Kulp, 1958, p. 449). This excess is probably present in two forms, as fluid inclusions, and in the relatively large channels in the structure, along with H2O (Ar from fluid inclusions has been reported for ultramafic xenoliths from Hawaii (Funkhouser et al., 1965; and Funkhouser and Naughton, 1968) and for hydrothermal fluorite (Lippolt and Gentner, 1963). Dalrymple and Lanphere (1969) reported >50 apparently anomalous ages attributed to excess ^ A r , only a few of which were specifically assigned to ^ A r from fluid inclusions. In spite of the problems from possibly large initial ^ 0 A r contents in the fluid inclusions, Zentilli and Reynolds (1 977) obtained a reasonable age on svlvite-bearing inclusions in quartz from a porphyry Cu orebody, using the ^ A r / •"Ar age spectrum method. They used the behavior of the inclusions on heating under the microscope to evaluate their 40/\r/39/\r s t e p heating data. Recent developments have greatly reduced the sample-size requirements for the method (Currie, 1982), and York et al. (1980, 1982a,b) used a laser probe mass spectrometer for dating of opaque ore minerals. It seems reasonable to suppose that a large part of the K and Ar involved in this determination is present in solution in fluid (or solid) inclusions rather than in the host mineral structures. York et al. used the isochron approach, separating various parts of the sample with varying K/Ar ratios; by this means, they circumvent the initial Ar problem to some degree. Release, Extraction and Analysis of Liquids Many quantitative analyses of liquid inclusions have been reported, using a wide range of combinations of methods of release, extraction, and analysis, making comparison difficult. Furthermore, many of the published analyses contain obvious internal inconsistencies, such as gross cation/anion imbalance, or do not agree with optical observations (e.g., low total concentrations yet NaCl daughter crystals reported to be present). Others present analytical results that seem impossible to obtain with the stated methods, or do not state the methods, making evaluation impossible. References giving the several thousand published analyses up to 1972 are cited by Roedder (1972) and hence are not reviewed here. Also, the various qualitative and semiquantitative analytical procedures based on microscopy (Chapters 4 and 8) are not discussed here. Most reported analyses of fluid inclusions are ratio analyses, in which the ratios of two or more ions are determined, commonly on solutions obtained by leaching a crushed sample. Ratios such as F/Cl, K/Na, or even those for the 128

bulk of the s o l u t e s p r e s e n t i n aqueous s o l u t i o n s , such as K/Na/Ca/Mg/S04/Cl, are o b t a i n e d , but s i n c e the amount of H2O i s not determined ( o r i s only c r u d e l y e s t i m a t e d ) the c o n c e n t r a t i o n s of these s o l u t e s i n the i n c l u s i o n f l u i d s are unknown. In c o n t r a s t , a r e l a t i v e l y few i n c l u s i o n analyses are q u a n t i t a t i v e a n a l y s e s , i n which both s o l u t e s and s o l v e n t are determined. R a t i o a n a l y s e s . Most of the s e v e r a l thousand i n c l u s i o n a n a l y s e s r e p o r t e d i n the l i t e r a t u r e have been made on composite or " b u l k " samples, by the water leach procedure. In t h i s procedure, some of the m i l l i r f n s or b i l l i o n s of i n c l u s i o n s i n a sample ( o f perhaps 100 g) are opened by f i n e g r i n d i n g ( o r d e c r e p i t a t i o n ) , and the sample i s then leached w i t h "pure" w a t e r . The l e a c h a t e (or " e x t r a c t " ) i s then analyzed f o r such s o l u t e s as Na, K, Ca, Mg, SO4, and CI, and (mainly i n the S o v i e t r e p o r t s ) , a l s o f o r pH and even Eh. Although the i o n a n a l y s i s i s perhaps a v a l i d procedure 1 on a few very c a r e f u l l y s e l e c t e d samples, the r e s u l t s are so e a s i l y i n v a l i d a t e d e i t h e r p a r t l y or completely by any of a s e r i e s of problems t h a t the r e s u l t s must g e n e r a l l y be viewed w i t h s u s p i c i o n . The b i g g e s t problem l i e s i n the sampling, as m u l t i p l e g e n e r a t i o n s of i n c l u s i o n s are p r e s e n t i n many i f not most samples. Even i f the i n c l u s i o n s i n a given sample are products of only one p r o c e s s , i t i s e x c e e d i n g l y d i f f i c u l t t o e x t r a c t the f l u i d w i t h o u t gross l o s s or c o n t a m i n a t i o n , or both (Roedder, 1958, 1972). Too f r e q u e n t l y t h i s l e a c h i n g and a n a l y s i s i s t r e a t e d as a simple r o u t i n e operat i o n , and as a r e s u l t of numerous p o t e n t i a l a n a l y t i c a l p i t f a l l s along the way, much of the e x t e n s i v e e a r l y work i s of dubious v a l i d i t y , e n t i r e l y a p a r t from the e v e r - p r e s e n t problem of the assignment of the i n c l u s i o n s i n the sample t o a s i n g l e epoch of o r i g i n , and the added p o s s i b i l i t y of f l u i d changes d u r i n g the f o r m a t i o n of a given sample. Sample c l e a n i n g i s a p a r t i c u l a r l y important and o f t e n n e g l e c t e d s t e p . Most o r d i n a r y samples c o n t a i n only perhaps 0.1% i n c l u s i o n f l u i d , c o n t a i n i n g perhaps 10% t o t a l i o n s i n s o l u t i o n , and i n many s t u d i e s , an ion of p a r t i c u l a r i n t e r e s t may c o n s t i t u t e only 1% of the t o t a l i o n s p r e s e n t . T h i s i o n thus c o n s t i t u t e s o n l y one p a r t per m i l l i o n of the whole sample. S i g n i f i c a n t c o n t a m i n a t i o n (and/or l o s s ) of such a small amount of m a t e r i a l i n the sequence of processes t h a t must be used i s the r u l e r a t h e r than the e x c e p t i o n , and the v a l i d i t y of the a n a l y t i c a l r e s u l t s o b t a i n e d w i l l be a d i r e c t f u n c t i o n of the care used i n m i n i m i z i n g (and e v a l u a t i n g ) the several sources of e r r o r . Two types of sample c l e a n i n g need to be c o n s i d e r e d : (1) The e l i m i n a t i o n of a l l m i n e r a l s but the one of i n t e r e s t , and (2) The e l i m i n a t i o n of s u r f a c e impurities. The s m a l l e r the sample to be used, the e a s i e r i t i s t o handpick c l e a n g r a i n s under the microscope, f r e e of o t h e r contaminating m i n e r a l s . L a r g e r samples, p a r t i c u l a r l y of t r a n s l u c e n t or opaque m a t e r i a l s such as v e i n q u a r t z , are f r e q u e n t l y s e l e c t e d by simple v i s u a l i n s p e c t i o n of the p i e c e s to e l i m i n a t e those c o n t a i n i n g o t h e r m i n e r a l s (on t h e i r o u t s i d e s u r f a c e s ) . Such s e l e c t i o n , although h e l p f u l , i s f a r from s u f f i c i e n t . Grains of mineral contami n a t i o n can be b u r i e d w i t h i n the fragments, and f i n e l y d i v i d e d i m p u r i t y phases w i l l almost always be found i f a t h i n s e c t i o n of such m a t e r i a l i s examined. I f a 100 g sample i s used, c o n t a i n i n g 0.1 wt % of f l u i d i n c l u s i o n s , w i t h 1 wt % of element X i n s o l u t i o n i n the i n c l u s i o n f l u i d , o n l y 0.001 g of X w i l l be p r e s e n t i n the l e a c h a t e . I f even j u s t a few m i l l i g r a m s of a mineral c o n t a i n i n g X are p r e s e n t as a b u r i e d c r y s t a l or as f i n e l y disseminated s o l i d i n c l u s i o n s i n the sample and are exposed d u r i n g g r i n d i n g ( e . g . , s e r i c i t e f l a k e s i n q u a r t z , when X i s K, or c a l c i t e , when X i s Ca), contamination from such sources can be s e v e r e . Even w i t h o u t d i s c r e t e contaminant m i n e r a l s , t r a c e c o n s t i t u e n t s i n the s t r u c t u r e s may be e x t r a c t e d and contaminate. The standard r e b u t t a l t o suggest i o n s of t h i s p o s s i b i l i t y of c o n t a m i n a t i o n from s o l u t i o n of o t h e r m i n e r a l s i s t h a t repeated l e a c h e s p r o v i d e l i t t l e more of the element of i n t e r e s t ; t h i s i s a n e c e s s a r y , but f a r from s u f f i c i e n t p r o o f . Such a d r o p o f f i n s e q u e n t i a l l e a c h e s i s c h a r a c t e r i s t i c i n s o l u b i l i t y d e t e r m i n a t i o n s of most f i n e l y ground pure m i n e r a l s . 129

400

Figure 5-7. Typical plot of electrodlalysls cell current flow vs time (diagrammatic) during the cleaning of a sample. The vertical lines at A, B, and C represent the times at which the solutions In the electrode chambers were removed and replaced with fresh deionized water. From Roedder (1958).

Surface impurities present a problem that is generally less obvious but that can be of similar magnitude. Included here are ions from soil or groundwater (or fingerprints) that have either adsorbed onto the mineral surfaces or have precipitated as minute grains of gypsum or other compounds in the surface cracks. I found that electrolytic cleaning was very simple yet effective in eliminating these contaminants (Roedder, 1958). The mineral fragments were placed in the bottom of a large Pyrex glass "U" tube with deionized water and Pt electrodes in each arm, and a constant DC voltage (e.g., 90 V) applied. An ammeter in the circuit provided control of the operation, as it monitored the conductivity of the fluid. The current flow increased rapidly as ions from the sample spread into the originally very low-conductivity water in the arms of the tube, and then declined with time (~1 day) as these ions clustered around the electrodes, leaving the main part of the water cleaner (Fig. 5-7). The contaminated water around the electrodes was then sucked off first, followed by the remainder of the water. New water was added to repeat the process, until there was essentially no change in conductivity with time, and the conductivity was approximately that of the deionized water used, modified by the solubility of the minerals. This took several days and three or four changes of water. (If slightly soluble minerals such as calcite are exposed somewhere on the sample, the conductivity will stay relatively high.) Once cleaned in this manner, it is safe to assume that the sample is free from any ionizable impurities except those in sealed solid inclusions or within completely closed and sealed fluid inclusions. The process has been subsequently been made more convenient by the addition of a water inlet at the bottom, permitting the contaminated water to be flushed over both tops simultaneously by the new water (D. Pinckney, pers. conm.). Many investigators have used acid treatments to eliminate both extraneous mineral grains and surface contaminants. This treatment for acid-insoluble samples such as quartz will eliminate solid contaminants, but has some serious drawbacks. First, the acid does not dissolve extraneous mineral grains that are completely embedded. Second, on contact with the sample, the acid moves into the inevitable surface cracks in the grains, dissolving the impurities. This contaminated acid must subsequently be removed from these cracks, and even the clean acid must be removed quantitatively if it contains an anion that is to be determined. A sequence of water "washes," as generally described in the literature, will remove most of the acid on the outside surfaces of the grains, but when such a "clean" washed sample was subsequently put in the electrolytic cleaning cell, I found that considerable acid was still present in the cracks, and required days of cleaning. P. Eadington (pers. comm.) avoids some of these problems by using HBr, followed by HNO3 and then washing. 130

Once the sample e x t e r i o r i s c l e a n , the i n c l u s i o n f l u i d i n i t must be r e leased. Two methods have commonly been used, c r u s h i n g (or ball m i l l i n g ) , and thermal . d e c r e p i t a t i o n . Some have ground the samples under water i n an agate m o r t a r . O t h e r s have used b a l l m i l l i n g , but r a r e l y report blank runs. I have found t h a t standard high temperature p o r c e l a i n m i l l s and g r i n d i n g media y i e l d g r o s s contamination, and even the use of a s i n t e r e d alumina ball mill and alumina g r i n d i n g media y i e l d e d s i g n i f i c a n t contamination i n g r i n d i n g 1000 g samples of -10+48 mesh quartz (Roedder, 1958). Goguel (1963) used a special vacuum b a l l m i l l i n g procedure on very small samples, p a r t i c u l a r l y to obtain the gases f o r a n a l y s i s . The ball m i l l e d samples were given a quick leach with water, f i l t e r e d with a membrane f i l t e r and analyzed. Goguel (1964) r e v i s e d some of h i s techniques to reduce contamination and obtained considerably lower values f o r B and C I . These CI values are a c t u a l l y the sum of C I , B r , and I ; t h i s i s probably t r u e f o r many of the " C I " analyses given i n the l i t e r a t u r e , but i t i s g e n e r a l l y not s t a t e d . Some workers ( e . g . , Savel 'yeva and Naumov, 1979) have used thermal decrepi t a t i o n to open the i n c l u s i o n s , but I had no success with the method, at l e a s t f o r the ions i n s o l u t i o n (Roedder, 1958, p. 263-266). I heated an 815 g sample of gold quartz from Grass V a l l e y , C a l i f o r n i a , f o r 23 hours at 488°C i n a fused s i l i c a f l a s k i n a stream of pure N2, and obtained 0.19 wt % H2O, and U.006 wt % CO2 from the d e c r e p i t a t i o n ( a f t e r f l u s h i n g at 110°C). A f t e r d e c r e p i t a t i o n , I used the e l e c t r o l y t i c c l e a n i n g c e l l (see above) t o c o l l e c t the ions remaining from evaporation of the f l u i d s on d e c r e p i t a t i o n . I obtained only 7 mg t o t a l K+Na, corresponding t o only 0.4 wt %, whereas another e x t r a c t i o n procedure on a part of the same sample y i e l d e d nearly 25 times more. Several explanations are p o s s i b l e . F i r s t , when i n c l u s i o n s d e c r e p i t a t e , they frequently do not f l y apart, but simply l o s e t h e i r contents through v i s i b l e or i n v i s i b l e f r a c t u r e s . This could leave part of the n o n v o l a t i l e s a l t s i n s i d e the i n c l u s i o n s and hence r e l a t i v e l y i n a c c e s s i b l e to l a t e r l e a c h i n g . Second, d i f f u s i o n of a l k a l i e s in quartz i s r e l a t i v e l y rapid at the temperature used; any that d i f f u s e d i n t o the quartz would a l s o be e f f e c t i v e l y l o s t . A much more rapid heating could p o s s i b l y minimize the second of these two problems. Some workers have heated samples to the d e c r e p i t a t i o n temperatures, and then crushed them before l e a c h i n g ; t h i s might minimize both problems. Thompson et a l . (1980) have used the d e c r e p i t a t i o n i t s e l f to i n j e c t an uncontaminated, d i s p e r s e d sample i n t o the detection device (see below). E x t r a c t i o n of the i n c l u s i o n ions from the ground host mineral might seem to be a simple f i l t e r i n g o p e r a t i o n , but i f the c r u s h i n g has been f i n e , as i s necessary to open small i n c l u s i o n s , i t i s not simple.®./ B a l l m i l l i n g can y i e l d f i n e s that c l o g f i l t e r s . Thompson (1981) used c e n t r i f u g i n g to provide a clean filtrate. Contamination from a l l handling steps i s a l s o a constant problem. Thus even h i g h - g r a d e a n a l y t i c a l f i l t e r papers were found to add f a r too much CI (from the " a c i d washing" p r o c e s s ) , and asbestos f i l t e r s had to be used (Roedder et a l . , 1963). To keep the blanks low, very high q u a l i t y water and reagents are needed. More important than such contamination, however, i s l o s s by a d s o r p t i o n on the l a r g e amount of new mineral s u r f a c e i n a f i n e l y ground sample (~10^ cm^/g). An example of the p o s s i b l e magnitude of both contamination and l o s s from s u r f a c e phenomenon i s seen in the advertisements f o r a commercial laboratory glassware cleaner that make a s a l e s point of the fact that a f t e r a

1/ Agate i s s l i g h t l y porous, and one i n v e s t i g a t o r found that his inexplicably high published results on CI i n h i s i n c l u s i o n s came about because a coworker in the laboratory was using concentrated HC1 to clean the mortar. I n c l u s i o n s in cinnabar are the e a s i e s t , as they permit combining the release and extraction steps. cinnabar can be simply sublimed away at low temperatures, thus leaving the i n c l u s i o n solutes behind. 131

The

glass surface has been cleaned with it, and rinsed four times with distilled water, only 0.6 mg/m^ of the cleaner remained on the surface. Electrodialysis with organic membranes can eliminate surface adsorption problems completely (Roedder, 1958), but introduces a few new problems. Others have used an acid leach, which probably removes adsorbed ions satisfactorily, but can result in gross contamination from exposed impurity minerals. Obviously, all these various problems will be minimized if the amount of new surface is small relative to the amount of ions released. Thus, coarse crushing of samples selected to have a large amount of fluid, as large inclusions, would be optimum. In a number of earlier papers, the leach solutions were evaporated to dryness and analyzed by qualitative or semiquantitative spectroscopy, but analysis of the extracted ions from samples in the 100 g range represents a relatively minor problem with the methods now available. Flame photometry is perhaps the most commonly used, but many others, such as atomic absorption spectroscopy (AA), ion chromatography (particularly for F, CI, Br, and SO4; see Thompson et al., 1983), ion-sensitive electrodes, direct-current plasma spectrometry (DCP), etc., have been tried and may be particularly suitable for a given application. Cation analyses are generally much simpler and more accurate than anion analyses. CI presents relatively few problems, in large part because it is generally the major anion. Liquid chromatography has been used to detect amino acids (Kuznetsova et al, 1983). /

The use of inductively coupled plasma (ICP) emission spectroscopy of elements released into a carrier gas during decrepitation has been reported in a series of papers from Imperial College, London (Alderton and Rankin, 1981; Thompson, 1981; Thompson et al., 1980; Alderton et al., 1982; and Rankin et al., 1982; see also Walsh and Howie, 1980). The method has very low detection limits for most metals ( < 1 0 0 g), as well as for B, C, S, and P, but F, CI, and Br cannot be determined at all. As is common, the major uncertainties in the procedure may well lie in the sample extraction step. Chryssoulis (1983) has shown that the presence of feldspar and mica in the sample as decrepitated is of minor consequence, but I believe a more basic fallacy remains unresolved. When an inclusion decrepitates, part of the fluid is dispersed into the Ar carrier gas, and part evaporates, leaving a solid residue around the pit in the surface. These deposits have also been analyzed, by electron microprobe (see "Electron microprobe" section below). As the various solutes in inclusion fluids vary widely in their volatilities, these two portions of the original fluid, that in the carrier gas and that in the residue, will almost certainly differ significantly in composition, but in the analytical procedures used, each one is tacitly assumed to approach a representative sample of the solutes originally present. The various S species that might be present in the original inclusion fluid, particularly S0|~, HS", and S , should be determined separately, but as even total S is frequently rather low, and much is normally in the form of sulfate, the analyses are generally of "total S as SO4.'1 Unfortunately, most of the analyses for S that have been reported, particularly in the Soviet literature, list "S0^ " without either giving the method used or indicating whether the analysis is that of total S as SO4 or of actual sulfate. As crushing and leaching have generally been done in an air environment, most such analyses may amount to total S as sulfate. Also, as small amounts of sulfide minerals are generally present and can oxidize quickly, most S determinations actually represent maximum values for the inclusion fluids. Only a very few workers have crushed their samples in an inert atmosphere and leached them with carefully deoxygenated solutions. The interpretation of the analysis can be far more frustrating and difficult than the analysis itself. Most of the older analyses list Na, K, Ca, Mg, CI, and SO4. Generally, these six ions do not add up to electrical neutrality. If we assume that the analyses themselves have been made correctly, this unbalance can come about from any (or all) of a variety of reasons. Other (unanalyzed) 132

ions are important c o n s t i t u e n t s i n some i n c l u s i o n s , and may cause unbalance. The most common of t h e s e are A1, Fe, B, P, F , S i , and OH. "Excess" c a t i o n s can come from s e v e r a l s o u r c e s . Thus excess a l k a l i e s can e a s i l y be leached from crushed f e l d s p a r , mica, and c l a y i n the sample, and excess a l k a l i n e earths can come from exposed c a r b o n a t e s . Excess CI i s u n l i k e l y , but excess SO4 can be r e l e a s e d by o x i d a t i o n of s u l f i d e s , which can proceed at an amazingly f a s t r a t e . I f so, t h e r e may w e l l be s i g n i f i c a n t amounts of other c a t i o n s i n s o l u t i o n t h a t are not normall y determined, such as Fe or Cu. S i l i c a t e anions of one s o r t or another are f r e q u e n t l y r e p o r t e d ; whether these are a r t i f a c t s from the l e a c h i n g process or r e a l c o n s t i t u e n t s of t h e f l u i d i s seldom obvious. Large amounts of s i l i c a have been moved around by f l u i d s i n the e a r t h ' s c r u s t - - probably more than any other m a t e r i a l - - but t h e e x t r a p o l a t i o n s from the amounts of s i l i c a detected by a given a n a l y t i c a l procedure i n a leach s o l u t i o n , t o the f l u i d s i n the unopened i n c l u s i o n s , and on t o t h e s e same f l u i d s at the temperature of t r a p p i n g , may be r a t h e r long s t e p s . The l a r g e s t s i n g l e source of ambiguity i n i n c l u s i o n analyses by f a r i s a r e s u l t of COg i n i t s v a r i o u s forms. Most a n a l y s e s , i f CO2 i s even mentioned, simply l i s t HCOj", t o be taken at f a c e v a l u e . Free CO? i s a very common major c o n s t i t u e n t of i n c l u s i o n f l u i d s i n a wide range of g e o l o g i c a l environments. On c r u s h i n g under water i t w i l l form H2CO3. This w i l l add t o the c a t i o n / a n i o n imbalance i f HCOg" i s determined. A simple c r u s h i n g t e s t t o v e r i f y the absence of f r e e CO2 under p r e s s u r e i s an a b s o l u t e l y e s s e n t i a l p r e l i m i n a r y t o any a n a l yses of f l u i d i n c l u s i o n s . I f s o l i d carbonates are p r e s e n t , f r e e CO2 w i l l d i s s o l v e them t o add c a t i o n s and t w i c e as much carbon as H C Q 3 " t o the s o l u t i o n as the o r i g i n a l CO2 t h a t was p r e s e n t . In a d d i t i o n , CO3 and/or H C O 3 may be present i n the o r i g i n a l i n c l u s i o n f l u i d s . I f H C O 3 " i s h i g h , i t can d i s s o c i a t e d u r i n g dry c r u s h i n g t o C O 3 " and f r e e CO2. In a d d i t i o n t o these sources of a m b i g u i t y , some of t h e a n a l y t i c a l procedures used f o r H C O 3 " or CO3 are not always capable of d i s t i n g u i s h i n g these two, and, i n a d d i t i o n , the s p e c i f i c methods used are seldom s t a t e d . An added problem i s i n t r o d u c e d by daughter m i n e r a l s when p r e s e n t , and by the s o l u b i l i t y of the s o l i d s p r e c i p i t a t e d on dry c r u s h i n g . The l e a c h i n g medium should take i n t o s o l u t i o n a l l daughter phases, and a l l s o l i d s p r e c i p i t a t e d on opening (and n o t h i n g e l s e ) , but t h i s may not always be f e a s i b l e . Thus i f the i n c l u s i o n s i n a sample from an ore d e p o s i t have v i s i b l e c h a l c o p y r i t e daughter c r y s t a l s , any l e a c h s o l u t i o n t h a t w i l l d i s s o l v e them would a l s o d i s s o l v e the "minute amounts" of s u l f i d e contamination commonly present i n the sample t o y i e l d g r o s s l y erroneous values f o r Cu, F e , and S. Loss of gases on opening may a l s o p r e c i p i t a t e s o l i d s t h a t are r e l a t i v e l y i n s o l u b l e i n pure leach water, and even without such l o s s , as the l e a c h water i s v a s t l y more d i l u t e i n other s a l t s than t h e i n c l u s i o n f l u i d s , i t may not r e d i s s o l v e them. A v a r i e t y of s p e c i a l techniques have been used f o r s p e c i f i c problems, and many more may have t o be d e v i s e d , as t h e q u a n t i t i e s of ions a v a i l a b l e from w e l l documented samples commonly are toward the lower l i m i t s of c u r r e n t l y a v a i l a b l e t e c h n i q u e s . Thus t h i n - f i l m anodic s t r i p p i n g voltammetry was used f o r Pb i n i n c l u s i o n s ( M i l l e r and Shepherd, 1984), and a h i g h l y s e n s i t i v e method r e c e n t l y developed f o r one r a t h e r d i f f i c u l t element, A l , may be a p p l i c a b l e t o f l u i d i n c l u s i o n s as w e l l (Brady and F r a n t z , 1980). Even a simple r a t i o of o n l y two elements can be u s e f u l . Thus H o l s e r (1963) and Kramer (1965) drew c o n c l u s i o n s from the r a t i o s Br/CI and F / C l i n i n c l u s i o n s from s a l i n e d e p o s i t s . Kozlowski and Karwowski (1974) used the r a t i o C I / B r t o a i d i n understanding the c o n d i t i o n s of formation of some g r a n i t e - g n e i s s m a s s i f s , and Kozlowski (1978) suggested t h a t the r a t i o F / C l might be u s e f u l as a p r o s p e c t i n g t o o l . Many s t u d i e s have reported Na/K r a t i o s f o r i n c l u s i o n s from ore deposi t s , as these are r e l a t i v e l y easy t o determine (by flame photometer), and can be of value i n understanding the sequence of w a l l r o c k a l t e r a t i o n stages i n ore d e p o s i t s , as w e l l as the temperature of rock-water e q u i l i b r a t i o n (Montoya and Hemley, 1975). 133

Neutron a c t i v a t i o n a n a l y s i s (NAA; a l s o c a l l e d INNA, f o r instrumental neut r o n a c t i v a t i o n a n a l y s i s ) i s a p a r t i c u l a r l y e f f e c t i v e method f o r c e r t a i n c o n s t i t u e n t s . Analyses of Na, K, Rb, Cs, C I , B r , Cu, Mn, and Zn have been made on i n c l u s i o n s by neutron a c t i v a t i o n ; Touray (1976) added A s , I , Dy, and W to t h i s l i s t (see a l s o reviews by L a u l , 1979; Muecke, 1980). Czamanske et a l . (1963) determined the Cu, Mn, and Zn contents of large i n c l u s i o n s i n a transparent f l u o r i t e from southern I l l i n o i s , and i n a t r a n s l u cent quartz sample from Creede, Colorado. They determined the volume of the i n c l u s i o n s with a vacuum c r u s h i n g procedure ( F i g . 5 - 2 ) , leached the crushed fragments f i r s t with H2O and then with cold d i l u t e HNO3, and the p a r t l y evaporated leachates were i r r a d i a t e d f o r 30 minutes at lO'^n/cm^/sec. Induced a c t i v i t i e s were counted a f t e r chemical s e p a r a t i o n procedures. The r e s u l t s f o r Cu and Zn i n the f l u o r i t e were s u r p r i s i n g l y h i g h , p a r t i c u l a r l y i n the a c i d leach (they c a l c u l a t e d to 0.9 and 1.1 wt % i n the i n c l u s i o n f l u i d s , r e s p e c t i v e l y ) but the v a r i o u s p o s s i b l e sources of contamination seem t o have been e f f e c t i v e l y excluded. Subsequent analyses of both s i m i l a r and d i s s i m i l a r samples by other methods and i n v e s t i g a t o r s have shown that the Cu and Zn values obtained by Czamanske et a l . (1963) were high but not r e a l l y e x c e p t i o n a l . Puchner and Holland (1966) a l s o used neutron a c t i v a t i o n , but before openi n g , t o determine Na, Cu, Mn, and Zn. The i n c l u s i o n s they used were i n quartz from a Pb-Zn ore deposit at P r o v i d e n c i a , Mexico, and were much s m a l l e r (6 770 u g ) . They used o p t i c a l measurements t o o b t a i n the volumes. During the i r r a d i a t i o n (5-25 h at ~ 1 . 3 x lO^n/cm^/sec) s y n t h e t i c l i q u i d s t a n d a r d s , sealed i n fused s i l i c a c a p i l l a r i e s , were run along with the unknowns. Na and Mn were found i n a l l four samples but Cu and Zn were below t h e i r detection l i m i t s , which they estimated to be x: r«* ifl m x a» kû Z -i- 0 4JV » — C O fi ci .o ai -—"a • O r-. 3UJû.a>—

E cuœ il IO CN r m o o E— co _e «r X o u o o >o oo o — i û£

o

3

a

o

£ O C M 10 ZJ F — > C O CSI UJ

S r— 3 *

> a» a> r— o o

158

{_ r— O «

U O U O ig u i g « >— 4J U 3 • i J O t/1

4) O « O 0-(T> lA O) 3 O IQ C

^

E

u 4>C JZ +> 4 O 18 X UJ l/l

4) 4-> I— IO — *

t_

1— IO -O Ï

>t O >? U O M -t-> « io 41 «

4)

SI

OJ TD > 0) O t-

4Jlrt.CC u. i/i en £ O 4) -i- -p

i/I O C L O .O

Si

IO IA c e

V) o ai E

.

41

4-> O. U E « m w

:8S

•

O « O. f o IO i*- e >—

*-> 0 O o o. m

l/> O. l/l Vi . E 4) 4> 41 4» 41 3 r—

C i— i— O O O C lA

c o « o CJ , C

C •r I£ i/l

Q. E O u 41 £t

I - >> 1 rT3 J 4-> 01 • 4>

«

IO 4) • 4) JZ 4J . >- O O 41

u r 41 > , 0 OI C r— — Wl O - d 4flflV J C7> ^ 4) 4> C CT C Z " T C 41 i *-» O 4-> C I- X • t S P 4 I T ) 41 O tO 41 -Ç5 E •*->•• io ^ t_ • > , L. tO O >) io 41 » - I - ' O OI •o 1T3 — > C ^ O CÌU") « O IO I 4) C 1 . - . L , -

>> O O

«- Ï C O O t/1 r— +J *J O) t l Ol u to >,

O 41

C £ C7>»— 1. 41 «/> "O ©."O 4-> C

ínuo

«

X "O •f~ 41 E

O C 4-> 4J , U r—

i— I > C , CO, H->, and CH< w i l l

drop i t , and

and another smaller "bubble" within the l i q u i d CO2 (gaseous CO2). phases homogenize as described above for pure CO2 inclusions.

The two CO2

Water solution plus s o l i d crystal plus vapor bubble. This type i s also sometimes ambiguously termed simply a "three-phase inclusion." If the crystal i s an i s o t r o p i c cube, i t i s normally NaCl, but i t s composition can be v e r i f i e d by several procedures (see Chapters 4 and 8). The presence of a crystal of NaCl indicates that the f l u i d is saturated with NaCl, but t h i s saturation does not mean the solution must contain 26% NaCl, as in the pure system NaCl-I^O; generally i t w i l l contain less than 26%, and sometimes much less (see Chapter 8). Selection by Elimination of "Divergent" Inclusions A problem is presented by occasional inclusions that d i f f e r visually in their phase ratios from the bulk of the inclusions, either in gas/liquid ratio or in the presence (or nature) of "daughter crystals" precipitated from the l i q u i d after trapping. (It i s important to keep i n mind that the apparent gas/liquid r a t i o is strongly affected by inclusion shape, so inclusions that appear to be different may have identical contents and homogenization temperatures; e . g . , see Bodnar and Beane, 1980, p. 881.) Generally, such visually divergent inclusions (sometimes called "anomalous" or "abnormal") w i l l be found to have grossly different Th. Avoiding them, as recommended by Ermakov and Kalyuzhnyi (1957), and others, represents an attempt to eliminate those i n c l u sions whose Th data would not be valid. The problem l i e s in deciding whether they are v a l i d samples of preexisting f l u i d s (e.g., formed from rapidly changing, inhomogeneous, or immiscible f l u i d s ) or samples that have been made divergent by some l a t e r (secondary) process and hence should be eliminated from the group to be run. I f they are v a l i d but are eliminated under the assumption that they have been altered, valuable information i s l o s t ; conversely, grossly misleading numbers w i l l confuse and d i l u t e good data from v a l i d inclusions. The l i n e is sometimes d i f f i c u l t to draw, and the subjectivity involved has caused some doubts about the v a l i d i t y of the method ever since the 1850s, when Sorby was accused of introducing bias into his samples by this selection procedure. The main secondary processes that must be considered when one finds such divergent inclusions are: (1) leakage and stretching (naturally or laboratory induced); (2) l a t e r r e f i l l i n g ; and (3) necking down (see also Chapter 3). Leakage. Leakage occurs in nature for a variety of reasons but i s usually from cracking related to some deformation of the host crystal. These cracks are not always v i s i b l e , as t h e i r width must be an appreciable f r a c t i o n of a wavelength of the l i g h t to affect i t . Leakage from such fracturing i s f a i r l y evident when a series of primary-appearing, but divergent, inclusions (often empty of a l l l i q u i d ) occurs as a planar array through a group of otherwise uniform l i q u i d - r i c h primary inclusions. In thin sections, inclusions within a few micrometers of the top or bottom surfaces are frequently empty, particularly i f t h e i r contents had been under high pressure. Overheating of inclusions beyond t h e i r Th, whether from later dikes (Lokerman, 1962, 1965) or in the laboratory (Roedder and Skinner, 1968; Larson et a l . , 1973; Bodnar and Bethke, 1980, 1984), can build up high pressures and cause decrepitation or stretching, p a r t i c u l a r l y of the larger inclusions. Freezing can also cause similar damage i f the vapor bubble is small. Later r e f i l l i n g . Later r e f i l l i n g i s a special case of leakage in which fracturing and rehealing of the inclusions occurred in the presence of a later f l u i d , under a different set of conditions. The opened inclusions may retain their o r i g i n a l shape and spatial arrangement but w i l l y i e l d a different Th. Necking down. Necking down i s the process whereby a long thin tubular or f l a t inclusion spontaneously reduces i t s t o t a l i n t e r f a c i a l energy by selective solution and redeposition, to y i e l d two or more separate, smaller, more equant 167

i n c l u s i o n s of the same t o t a l volume (see Chapter 3 and F i g . 3 - 9 ) . Nothing i s l o s t or gained from t h e system as a whole, but i f a phase s e p a r a t i o n has o c c u r red previous t o t h e necking down, t h e separated i n c l u s i o n s w i l l d i f f e r i n phase r a t i o s and Th. Necking down of homogeneous, one-phase i n c l u s i o n s w i l l have no such consequences. Necking down occurs commonly i n nature but i t g e n e r a l l y takes p l a c e near the o r i g i n a l t r a p p i n g temperature. I t i s probably r e s p o n s i b l e f o r much of the s c a t t e r of r e s u l t s t h a t i s apparent i n any c a r e f u l thermometric study. In the l a b o r a t o r y one must be a l e r t t o r e c o g n i z e the e x i s t e n c e of necking down. When i t has been i n t e r r u p t e d w h i l e i n p r o g r e s s , the long t a i l s between i n c l u s i o n s are easy t o see ( F i g . 3 - 1 8 ) . A f t e r s u f f i c i e n t r e c r y s t a l 1 i z a t i o n , however, i t may be recognized only by the adjacent occurrence of two or more i n c l u s i o n s t h a t are d i v e r g e n t i n o p p o s i t e d i r e c t i o n s ( e . g . , too much and too l i t t l e vapor r e l a t i v e t o o t h e r nearby i n c l u s i o n s ) , and sometimes by abandoned daughter c r y s t a l s , l e f t surrounded by host c r y s t a l . S e l e c t i o n Based on D i f f e r e n c e s i n Otherwise S i m i l a r

Inclusions.

Several other c r i t e r i a f o r s e l e c t i o n or r e j e c t i o n of i n c l u s i o n s e x i s t i n a d d i t i o n t o the above. Thus, a l l the i n c l u s i o n s of a given apparent o r i g i n may not be the same. The primary i n c l u s i o n s i n a given c r y s t a l may, and f r e q u e n t l y do, show s i g n i f i c a n t d i f f e r e n c e s between the s e v e r a l zones of a zoned c r y s t a l . I f zoning i s v i s i b l e w i t h i n a c r y s t a l , i n c l u s i o n s from each zone should be run, i f possible. Zoning t h a t i s not v i s i b l e i n a c r y s t a l might s t i l l be revealed by examining i n c l u s i o n s from core and rim. S i m i l a r l y , i f the host occurs i n two d i f f e r e n t c r y s t a l h a b i t s or s i z e ranges, these two may have formed at d i f f e r e n t t i m e s . I n c l u s i o n shape, although a t r a n s i e n t f e a t u r e of l i t t l e d i a g n o s t i c value f o r the d e t e r m i n a t i o n of o r i g i n (Roedder, 1968b), can r e f l e c t d i f f e r e n c e s i n composition or c r y s t a l h a b i t at the time of t r a p p i n g . Thus, the shapes of the primary i n c l u s i o n s i n some zoned f l u o r i t e changes from negative t e t r a h e d r a i n the core ( t h a t grew as an octahedron) t o negative cubes i n the rims (that grew as a cube; Roedder, 1977a). Secondary i n c l u s i o n s i n d i f f e r e n t planes i n a given c r y s t a l may d i f f e r i n time of o r i g i n and hence i n composition and shape. The behavior of the vapor bubble i n an i n c l u s i o n when s u b j e c t e d t o a s l i g h t thermal g r a d i e n t can p r o v i d e u s e f u l i n f o r m a t i o n on the c o m p o s i t i o n a l equivalence of two i n c l u s i o n s . The bubble can respond i n f o u r ways: (1) i t can move up t h e g r a d i e n t , toward the source of heat (most common b e h a v i o r ) ; (2) i t can f a i l t o move at a l l (even though f r e e of the w a l l s ) ; (3) i t can move down the g r a d i e n t ; or (4) i t can o s c i l l a t e w i t h a p e r i o d i c i t y complexly r e l a t e d t o the ambient temperature. The b a s i s f o r a l l these phenomena appears t o be v a r i a t i o n i n the s u r f a c e t e n s i o n over the bubble s u r f a c e , c a u s i n g flow of the s u r f a c e t h a t drags f l u i d with i t . Although t h e phenomena are s t i l l not f u l l y e x p l a i n e d , the motions of the bubbles can be detected very q u i c k l y and simply by hot or c o l d probes (Roedder, 1965c; 1966; 1967c; 1972, p. 19). These motions are u s e f u l i n i n c l u s i o n s t u d i e s i n a v a r i e t y of ways, only one of which i s p e r t i n e n t i n the present context. The s p e c i f i c behavior i s o b v i o u s l y c o n t r o l l e d by t h e composition of the f l u i d . Even though the nature of the e f f e c t s of these compositional parameters i s complex and not understood, d i f f e r e n c e s i n the behavior of the bubbles i n adjacent i n c l u s i o n s t h a t o t h e r w i s e seem t o be i d e n t i c a l i n d i c a t e d i f f e r e n c e s i n composition and presumably i n o r i g i n , and these d i f f e r e n c e s f r e q u e n t l y c o r r e l a t e w i t h d i f f e r e n c e s i n f r e e z i n g and Th. (Swanenberg, 1976, b e l i e v e s that c e r t a i n bubble motions can be assigned t o the presence of CO2; although t h i s may be t r u e i n some i n c l u s i o n s , I have found e x c e p t i o n s . ) I use a s i m p l e , homemade hot w i r e probe ( F i g . 6-4) t o check these movements. As the r e a c t i o n i s almost i n s t a n t a n e o u s , and may have t o be viewed w i t h o b j e c t i v e s of short working d i s t a n c e , the probe heater must be very t h i n and should be at room temperature when f i r s t put i n p l a c e above the p l a t e but i n the f i e l d 168

Figure 6-4. Diagram (approximately f u l l s i z e ) of e l e c t r i c a l l y heated probe f o r d e t e r m i n a t i o n of the thermal response of i n c l u s i o n b u b b l e s . Power i s s u p p l i e d t o the 0.019-nm Pt w i r e heating element from a v a r i a b l e v o l t a g e t r a n s f o r m e r v i a a foot s w i t c h . The probe body i s a piece of 2 - h o l e p o r c e l a i n " s p a g h e t t i . " The leads i n the s p a g h e t t i are a l a r g e r diameter w i r e , so h e a t i n g i s concentrated i n the h e a t i n g element.

of view through the microscope. Once the probe i s in place and i t s p o s i t i o n r e l a t i v e to the i n c l u s i o n i s seen, the foot switch i s tapped b r i e f l y , sending a pulse of current through the probe heating element. Any " a t t r a c t i o n " or " r e p u l s i o n " of the bubble w i l l g e n e r a l l y be i n s t a n t a n e o u s . Number of I n c l u s i o n s t o be Selected No f i x e d number of i n c l u s i o n s can be set as the appropriate number f o r any given i n v e s t i g a t i o n . I f v a r i a t i o n s are found to e x i s t i n composition and/or density w i t h i n apparently cogenetic groups of i n c l u s i o n s , t h i s f a c t may be important i n the i n t e r p r e t a t i o n and i t s f i r m establishment may require many more data. Many s t u d i e s of metamorphic terranes have shown that about 200 i n c l u s i o n s are adequate t o e s t a b l i s h a pattern and provide a b a s i s f o r a p r e l i m i n a r y i n t e r p r e t a t i o n of the r e s u l t s . Smaller numbers of measurements may be adequate in ore deposit s t u d i e s , where a s i n g l e generation of i n c l u s i o n s , a l l primary, may be a v a i l a b l e , but even here, s i g n i f i c a n t v a r i a t i o n s may be found i n primary i n c l u s i o n s from d i f f e r e n t zones in a s i n g l e c r y s t a l .

RECORD KEEPING The problem of "bookkeeping" i n f l u i d i n c l u s i o n s t u d i e s — simple recording of the data — may seem t r i v i a l at f i r s t , but i t i n v o l v e s a s u f f i c i e n t l y complex matrix of v a r i a b l e s that i t can i n t e r f e r e s e r i o u s l y with progress of the work and even with the r e c o g n i t i o n of important major trends or features i f not done p r o p e r l y . The design of appropriate procedures w i l l o b v i o u s l y vary with the nature of the i n v e s t i g a t i o n , but, as changes in the procedures made i n the middle of a p r o j e c t , f o r c i n g t r a n s c r i p t i o n or improvement of o l d records, can be time consuming, i t i s important to design w e l l . Two aspects should be considered separately: sample i d e n t i f i c a t i o n and coding, and data r e c o r d i n g . Sample I d e n t i f i c a t i o n and Coding Sample i d e n t i f i c a t i o n and coding includes a l l p e r t i n e n t o b s e r v a t i o n s concerning a given i n c l u s i o n other than the actual thermometric data derived from i t . The obvious parameters i n c l u d e the host mineral, the o r i g i n a l sample number, the i n c l u s i o n s i z e , shape, and vapor bubble diameter, the apparent phase assemblage and estimated phase volume percentages, and the assignment of o r i g i n (and the b a s i s f o r assignment). But i t must a l s o permit recovery of such important f a c t s as the i n c l u s i o n ' s p o s i t i o n i n and r e l a t i o n s h i p to i t s host c r y s t a l and the host c r y s t a l ' s p o s i t i o n and r e l a t i o n s h i p t o the host rock. As an example that i s not extreme, primary i n c l u s i o n s from v a r i o u s zones i n each of a s e r i e s of zoned c r y s t a l s from each of several d i f f e r e n t samples may be selected f o r s t u d y , as well as a s e r i e s of secondary (or p o s s i b l y pseudosecond a r y ? ) i n c l u s i o n s from each of several planes c r o s s c u t t i n g each of these c r y s t a l s . Superimpose on t h i s matrix the need to cut up the o r i g i n a l c r y s t a l s i n t o fragments that w i l l f i t i n the apparatus used, and the common requirement of a f i e l d of view of only a 3 mm c i r c l e during any run, and i t i s obvious that sample coding can become cumbersome.

Recording of Thermometric Data Data r e c o r d i n g i s r e l a t i v e l y simple compared w i t h sample i d e n t i f i c a t i o n but i t has some problems of i t s own. The v a r i o u s thermometric measurements, as d e t a i l e d i n the next c h a p t e r , g e n e r a l l y r e s u l t i n a s p e c i f i c temperature (or range) f o r each phase change observed i n a given i n c l u s i o n . A g e n e t i c a l l y r e l a t e d group of many i n c l u s i o n s ( e . g . , a c l u s t e r of a given or estimated number of p r i m a r i e s , or s i m i l a r l y , a plane of s e c o n d a r i e s ) may behave so nearly the same t h a t temperatures or ranges can be given f o r the group. Repeat measurements are f r e q u e n t l y made on p a r t or a l l of the i n c l u s i o n s , f o r s e v e r a l reasons. Each such measurement, or group of measurements, should have an assigned o b s e r v a t i o n a l u n c e r t a i n t y ( q u i t e apart from the i n s t r u m e n t a l u n c e r t a i n t y ) . The above i n v o l v e s s t r a i g h t f o r w a r d , simple data r e c o r d i n g . However, some l e s s obvious items of data may prove t o be e i t h e r e s s e n t i a l or d e s i r a b l e and are seldom s u p e r f l u o u s . The p o s s i b l e need f o r each of these i s d i s c u s s e d i n Chapter 7; they are simply t a b u l a t e d here because they are an i n t e g r a l part of the problem of record keeping: (1) P r e v i o u s h e a t i n g and c o o l i n g h i s t o r y of the p l a t e , i n both sample preparat i o n and runs; maximum and minimum temperatures achieved, and d u r a t i o n at each extreme. (2) Heating (or c o o l i n g ) h i s t o r y d u r i n g t h e s p e c i f i c run y i e l d i n g a given temp e r a t u r e of phase change; o v e r a l l r a t e s of temperature change from o r i g i n a l ambient. (3) Heating (or c o o l i n g ) r a t e at t h e moment of the determined phase change. (4) Date of d e t e r m i n a t i o n and o b s e r v e r . (b) Other unknown or c a l i b r a n t samples i n t h e stage d u r i n g the same run. (6) Equipment v a r i a b l e s such as s t a g e , window t h i c k n e s s , thermocouple, microscope o b j e c t i v e , instrument mode, e t c . Some of these items may be i d e n t i c a l f o r many i n d i v i d u a l i n c l u s i o n s or groups of runs, and hence need t o be recorded only once, but they should be recorded. Thus, on numerous o c c a s i o n s , I have learned t h a t some seemingly i n c o n s e q u e n t i a l change i n o b s e r v a t i o n a l technique or equipment v a r i a b l e i n the past i s p e r t i n e n t t o the i n t e r p r e t a t i o n of the r e s u l t s obtained before or a f t e r the change. Such records are p a r t i c u l a r l y important i n t r y i n g t o unravel the mass of apparently i r r e p r o d u c i b l e and c o n f l i c t i n g data t h a t may a r i s e when some of the data represent s t a b l e e q u i l i b r i a , and other data represent metastable e q u i l i b r i a , and the observer does not know which are which. PHOTOGRAPHY OF INCLUSIONS Photography i s a n e a r l y i n d i s p e n s a b l e p a r t of most i n c l u s i o n s t u d i e s . Photographs p r o v i d e an e x c e l l e n t record of the i n t e r r e l a t i o n s h i p s of the i n c l u s i o n s and the host c r y s t a l s ; they p r o v i d e documentation of the steps i n d i s a g g r e g a t i o n of a p o l i s h e d p l a t e i n t o the c h i p s t h a t are a c t u a l l y run; they provide a r e f e r e n c e a g a i n s t which changes i n t h e i n c l u s i o n s d u r i n g runs, such as d i s s o l u t i o n of the w a l l s , r e c r y s t a l l i z a t i o n of daughter c r y s t a l s , or changes i n bubble d i a m e t e r , can be measured; and some are used as i l l u s t r a t i o n s f o r p u b l i c a t i o n s . Most of the procedures d e s c r i b e d above f o r i n c l u s i o n petrography are u s e f u l f o r i n c l u s i o n photography. Commonly the eye "sees" more than the f i l m , however. No s i n g l e best procedure f o r photography e x i s t s ; the f o l l o w i n g s e c t i o n d e s c r i b e s some of the procedures I have found u s e f u l i n my own work. Depth of Focus As most i n c l u s i o n f e a t u r e s are t h r e e d i m e n s i o n a l , the procedures used i n normal photography of p é t r o g r a p h i e s e c t i o n s must g e n e r a l l y be m o d i f i e d . This 170

problem of the t h i r d dimension i s the major hurdle i n i n c l u s i o n photography and commonly the r e s u l t s are d i s a p p o i n t i n g , e s p e c i a l l y at high m a g n i f i c a t i o n . The eye i s capable of i n s t a n t l y i n t e g r a t i n g s e v e r a l l e v e l s of focus through an i n c l u s i o n i n t o a t h r e e - d i m e n s i o n a l mental image, but the camera s t u b b o r n l y i n s i s t s on r e c o r d i n g only what i s a c t u a l l y i n focus at a s i n g l e s e t t i n g . Similarly, a p l a n a r group of i n c l u s i o n s t h a t i s almost p a r a l l e l w i t h the plane of focus may seem very photogenic, but only one small zone through the plane i s a c t u a l l y i n focus at any given s e t t i n g . Such problems are most apparent at high m a g n i f i c a t i o n , where the depth of focus i s at a minimum. The obvious s o l u t i o n of i n c r e a s i n g the depth of focus by " s t o p p i n g down" the a p e r t u r e s i n t h e microscope i s seldom e f f e c t i v e . The more c o l l i m a t e d l i g h t i n g from s t o p p i n g down r e s u l t s i n widening of the black shadows from t o t a l r e f l e c t i o n around each i n c l u s i o n (and each i n t e r n a l phase boundary), thus e l i m i n a t i n g many of the f e a t u r e s t o be i l l u s t r a t e d . Stopping down at very high magn i f i c a t i o n g r e a t l y i n c r e a s e s d i f f r a c t i o n e f f e c t s , s e r i o u s l y reducing r e s o l u t i o n . The most common undesired r e s u l t of s t o p p i n g down, however, comes from the f a c t t h a t i t emphasizes the many i n c l u s i o n s , c r a c k s , or other i m p e r f e c t i o n s i n the p l a t e above and below the i n c l u s i o n of i n t e r e s t . Once a g a i n , the eye conveni e n t l y ignores these dark, o u t - o f - f o c u s smudges, but the camera i s honest. Except f o r l o w - m a g n i f i c a t i o n photography (see below), few o p t i o n s f o r r e s o l v i n g t h i s dilemma are open. One i s simply t o compromise, by a d j u s t i n g t h e l i g h t i n g t o minimize the u n d e s i r a b l e f e a t u r e s . Another i s t o look f o r a more photogenic f i e l d t o photograph, one i n which one or more i n c l u s i o n s are f l a t t e n e d p e r f e c t l y p a r a l l e l t o the plane of f o c u s , so t h a t a l l phase boundaries are v i s i b l e and not superimposed. Such a f i e l d should a l s o be i n a c l e a r area, w i t h no d i s t u r b i n g i m p e r f e c t i o n s above or below, thus p e r m i t t i n g more s t o p p i n g down. This i n t e r f e r e n c e from superimposed images i s p a r t i c u l a r l y d i s t u r b i n g i n i n c l u s i o n - r i c h areas. I t can be g r e a t l y reduced by making a t h i n n e r p l a t e . Still another o p t i o n i s t o photograph at lower m a g n i f i c a t i o n on a h i g h - r e s o l u t i o n f i l m (such as P o l a r o i d type 655) and then e n l a r g e . Perhaps someday m i c r o h o l o graphy w i l l p r o v i d e the u l t i m a t e s o l u t i o n . Photography at Low M a g n i f i c a t i o n Documentation and f i n d i n g photographs. Photographs of each e n t i r e p o l i s h e d p l a t e are u s e f u l f o r s e v e r a l purposes. They document the s i z e , z o n i n g , and i n t e r r e l a t i o n s of the host c r y s t a l s , and they are the f i r s t , i n d i s p e n s a b l e step i n r e l o c a t i n g p r e v i o u s l y s e l e c t e d i n c l u s i o n s . Although a sketch w i l l sometimes s u f f i c e f o r the l a t t e r purpose, p a r t i c u l a r l y i f the p l a t e has few i n c l u s i o n s or i s so shaped t h a t r e l o c a t i o n i s s i m p l e , a photograph becomes almost e s s e n t i a l i f the rock i s f i n e grained or the i n c l u s i o n s abundant. Such a photograph w i l l represent an enlargement of only perhaps 3x t o lOx, a low range t h a t has not been common i n p e t r o g r a p h i c l a b o r a t o r i e s , ! . / but i s becoming more s o . I have used s e v e r a l procedures, depending on a v a i l a b l e equipment: (1) Normal photography. Here a n e g a t i v e would be made and subsequently e n l a r g e d t o an a p p r o p r i a t e s i z e . The r e s u l t a n t p r i n t can p r o v i d e e x c e l l e n t r e s o l u t i o n of d e t a i l , and the p r i n t i n g can be c o n t r o l l e d t o p r o v i d e optimum contrast. The two-step process i s slow, however, and the r e s u l t s are not immediately available. C o l o r transparency f i l m provides an a l t e r n a t i v e . Ektachrome 64 f i l m can normally be developed i n one day, and only t h e best frames need be e n l a r g e d . R e l o c a t i o n diagrams can be made by p r o j e c t i n g t h e f i l m and t r a c i n g t h e d e s i r e d f e a t u r e s and i n c l u s i o n s . (2)

I n s t a n t photography.

Although most p e t r o g r a p h i c microscopes do not

! I Even the IX objective available for pétrographie microscopes does not usually suffice to 90 this low. 171

permit viewing of l a r g e areas at low m a g n i f i c a t i o n , i f a P o l a r o i d camera a t t a c h ment i s a v a i l a b l e , an o v e r l a p p i n g mosaic of photographs of a d j o i n i n g areas can be composited. A mechanical stage w i l l help i n s u r e complete coverage. A u s e f u l v e r s i o n of t h i s c o m p o s i t i n g procedure t h a t permits almost any d e s i r e d degree of m a g n i f i c a t i o n i n v o l v e s p u t t i n g the sample p l a t e i n an o r d i n a r y photographic e n l a r g e r and p r o j e c t i n g onto P o l a r o i d f i l m i n a h o l d e r t h a t i s indexed a g a i n s t the normal enlargement e a s e l . Care must be used i n o r i e n t i n g the p l a t e . The p l a t e must be i n s e r t e d i n the e n l a r g e r w i t h the c o r r e c t s i d e up, depending upon whether the r e s u l t a n t f i n d i n g photograph i s t o be used i n connection w i t h l a t e r o p t i c a l systems (on microscope or microprobe) t h a t do or do not i n v e r t the image. Most s i g n i f i c a n t i s t h a t the r e s u l t s are immediately a v a i l a b l e f o r examination, but the cost per exposure i s h i g h e r . (3) E n l a r g i n g xerography. Some x e r o g r a p h i c equipment designed t o enlarge standard 2" x 2" (5 cm) photo t r a n s p a r e n c i e s w i l l permit d i r e c t enlargements of a mounted sample p l a t e i n the d e s i r e d range of m a g n i f i c a t i o n . The r e s u l t i n g p r i n t s do not have the r e s o l u t i o n or c o n t r a s t c o n t r o l of photography, but are obtained q u i c k l y and i n e x p e n s i v e l y . Photographs f o r i l l u s t r a t i o n s . Photography f o r s c i e n t i f i c i l l u s t r a t i o n s has much more s t r i n g e n t requirements than simple record photography. When the i n e v i t a b l e degradation of photographic q u a l i t y upon reproduction i n a j o u r n a l i s superimposed on the problems of depth of f i e l d d e s c r i b e d above, s e l e c t i o n of the most a p p r o p r i a t e f i e l d becomes i n c r e a s i n g l y d i f f i c u l t , and v a r i o u s p r o cedures need be used t o make the best photograph of the a v a i l a b l e i n c l u s i o n s . I f the c o l o r zoning i n the host i s p a l e , i t i s normally d i f f i c u l t t o r e c o r d . I t can f r e q u e n t l y be i n t e n s i f i e d by p l a c i n g the a p p r o p r i a t e c o l o r f i l t e r i n the l i g h t p a t h . The e a s i e s t procedure i s t o use a s i m p l e , i n e x p e n s i v e , hand-held "wedge i n t e r f e r e n c e f i l t e r , " which p r o v i d e s the e n t i r e range of the v i s u a l spectrum. Merely a d j u s t the wavelength by t r a n s l a t i n g i t i n the l i g h t beam u n t i l a maximum c o n t r a s t i s seen. C o n v e r s e l y , i f the c o n t r a s t between c o l o r e d zones i s so l a r g e t h a t d e t a i l i n one or another zone i s l o s t ( e . g . , a f l u o r i t e w i t h deep blue and c o l o r l e s s zones), simply a d j u s t the wedge u n t i l the f i e l d i s bathed i n blue l i g h t t o mini m i z e the c o n t r a s t . Q u i t e commonly, the d e s i r e d f i e l d of view v a r i e s g r e a t l y i n transparency from s i d e t o s i d e . The eye a d j u s t s e a s i l y t o t h i s on moving from one p o i n t t o another but the f i l m has f i x e d c h a r a c t e r i s t i c s . This problem can be minimized by s h i f t i n g the substage condenser o f f - a x i s , or by the standard darkroom "dodgi n g " procedure, moving a card i n the microscope i l l u m i n a t i o n system t o cut down the exposure on the b r i g h t e r part of the f i e l d . A poor p o l i s h can be p a r t l y c o r r e c t e d by adding a matching index l i q u i d and a g l a s s p l a t e or c o v e r g l a s s w i t h care t o a v o i d t r a p p i n g bubbles. Dark p i t s from p l u c k i n g of g r a i n s d u r i n g p o l i s h i n g w i l l a l s o be minimized by t h i s procedure. Both top and bottom s u r f a c e s should be t r e a t e d . Large samples can be immersed i n a bath of the f l u i d ( e . g . , see Rankin and Greenaway, 1978). Even i f the sample i s not immersed, d i s t r a c t i n g black bars from t o t a l r e f l e c t i o n on cleavage c r a c k s can be minimized by f i l l i n g the cracks w i t h the a p p r o p r i a t e index l i q u i d and then c l e a n i n g o f f the e x c e s s . Inadequate l i g h t i n g on a s p e c i f i c f e a t u r e , such as the contents of a l a r g e i n c l u s i o n at low m a g n i f i c a t i o n , can sometimes be c o r r e c t e d by the use of s u p p l e mentary l i g h t from one or more f i b e r o p t i c i l l u m i n a t o r s , from above or below the sample. Sometimes s e v e r a l exposures on the same f i l m , w i t h the a u x i l i a r y l i g h t i n g i n d i f f e r e n t p l a c e s , y i e l d good r e s u l t s .

172

Figure 6-5. T r i p o d f o r t i l t i n g samples i n a f k i i d bath f o r optimum photography of noncoplanar inclusions. The sample (S) on g l a s s s l i d e (G) i s held over h o l e i n t r i p o d p l a t e .

Polycrystal 1 i ne plates sometimes y i e l d much more i l l u s t r a t i v e photographs i f they are taken with s l i g h t l y crossed polars. This permits both the crystal boundaries and the internal details to be seen in a l l grains in a single picture. The position of the two polars relative to each other and to the sample can be adjusted to provide the optimum degree of contrast in the most important grains. I f a group of otherwise photogenic inclusions l i e in a plane that i s s l i g h t l y inclined to the polished plate, the focus can be improved by tipping the s l i d e appropriately. The simplest method i s to mount the plate or s l i d e on a larger glass plate, with lumps of modelling clay to provide the needed t i l t . More control i s possible with a simple homemade t i l t i n g device such as that shown in Figure 6-5. Optical problems from the t i l t , particularly at larger angles, can be eliminated by immersing the t i l t i n g device plus sample in a bath of l i q u i d with an index s i m i l a r to that of the host. Waves in the rather large upper surface of t h i s l i q u i d from the inevitable vibrations can blur the images. The simplest remedy i s to minimize the effective surface area through which the photograph i s made by e n c i r c l i n g i t with a metal ring that s i t s on the top surface of the sample and extends above the l i q u i d surface. Better results can be obtained by photographing through a flat-bottomed glass immersed in the l i q u i d over the sample. This sample t i l t i n g procedure turns out to be much less effective in pract i c e than in theory. The problem i s in that the magnitude of the t i l t needed i s frequently much larger than the s l i g h t difference in levels of focus would seem to indicate. The amount of t i l t needed can be estimated quickly as long as the microscope has been calibrated both horizontally and v e r t i c a l l y (see above, under "Selection by s i z e " ) . Photography at Intermediate to High Magnification Most of the procedures detailed in the previous section are just as appropriate here and need not be repeated. Some additional procedures that I have found useful are given below. When the bubble in an inclusion i s in an inconvenient position, i t can sometimes be moved into a more photogenic position by application of a thermal gradient with a hot wire probe (Fig. 6-4) or even by putting a shadow across part of the f i e l d but out of the camera's f i e l d of view (Fig. 6-3). Similar results can be gotten by adjusting the substage condenser. I f the bubble is partly obscured in a dark corner, f i b e r - o p t i c l i g h t from above or below will generally help. Daughter c r y s t a l s sometimes can be moved into a better position by a heating and cooling procedure. (However, some return only as a confused tangle of c r y s t a l l i t e s . ) For most of my own photomicrography, I use an instant f i l m (Polaroid P/N 665) that provides both a positive (print) and a negative. The contrast of such prints i s normally rather low, but the high-resolution negative can then be printed on higher contrast paper. This procedure permits immediate v e r i f i c a tion that a s a t i s f a c t o r y photograph has been obtained and yet permits later enlargement, reduction, and cropping of prints made with appropriate density and contrast. These can then be trimmed and mounted together into composites that provide the maximum of photographic i l l u s t r a t i o n in a minimum of space. The scale of photographs i s a subject that i s sometimes treated rather 173

too l i g h t l y . I object s t r o n g l y to the ( s t i l l too p r e v a l e n t ) use of a simple m a g n i f i c a t i o n value ( e . g . , "x250" or " 2 5 0 x " ) on i l l u s t r a t i o n s . Such an express i o n i s u s u a l l y ambiguous. Does i t refer to the m a g n i f i c a t i o n on the o r i g i n a l negative, t o the p r i n t that was sent to the j o u r n a l , or to the photograph as f i n a l l y p r i n t e d ? On checking c a r e f u l l y , I have found examples (but not so s t a t e d ) of each of these three usages. In a d d i t i o n to t h i s ambiguity, such m a g n i f i c a t i o n values become d i f f i c u l t to use when d e a l i n g with m i c r o f i c h e , microfilm, or other photocopies of j o u r n a l s . A simple bar s c a l e on^ the photograph, i t s length l a b e l l e d on i t or i n the c a p t i o n , i s the obvious s o l u t i o n . T h i s bar s c a l e can be e s t a b l i s h e d several ways. Some workers c a l i b r a t e t h e i r equipment so that a known combination of lenses represents a given area. The number of p o s s i b l e lens combinations i n microscope and camera, and the usual enlargement or reduction and cropping of the p r i n t s in my own procedures, makes t h i s process unwieldy. I n s t e a d , I note, on the o r i g i n a l photograph, the length of some e a s i l y measured f e a t u r e , as determined u s i n g a c a l i b r a t e d ocular r e t i c l e s c a l e . When the f i n a l p r i n t i s made, the length can be used to e s t a b l i s h the length i n m i l l i m e t e r s of an appropriate bar s c a l e on the photograph. (This i s e a s i l y cut from a t h i n s t r i p of black p r e s s - o n tape stuck on a metric r u l e r . ) Most o p t i c a l microscopes are corrected to y i e l d maximum r e s o l u t i o n and minimum a b e r r a t i o n in green l i g h t . A l l photographs should be taken with such green l i g h t , u s i n g an a p p r o p r i a t e narrow-band green i n t e r f e r e n c e f i l t e r made f o r the purpose. T h i s i n c r e a s e s exposure times but helps improve r e s o l u t i o n , p a r t i c u l a r l y at high m a g n i f i c a t i o n . Longer exposures may require photography l a t e at night or on weekends, when most b u i l d i n g s are much l e s s a q u i v e r . I f b i r e f r i n g e n t daughter phases are present, p a r t l y crossed p o l a r s u s u a l l y provide f a r better r e s u l t s than completely crossed p o l a r s . I f the b i r e f r i n g e n c e of the host permits, the p a i r of p o l a r s (or the sample) can be rotated f o r optimum c o n t r a s t . (See Chapter 4 f o r other recommendations.) Optical microscope lenses are normally corrected to provide optimum r e s o l u t i o n when there i s a standard c o v e r g l a s s between the object and the o b j e c t i v e . As a r e s u l t , most i n c l u s i o n photographs must be made under less-than-optimum o p t i c a l c o n d i t i o n s . As u s u a l , the problem i s most acute at high m a g n i f i c a t i o n . Frequently, the i n c l u s i o n i s too deep i n the m i n e r a l ; t h i s depth e f f e c t i s exacerbated by a high index of r e f r a c t i o n f o r the m i n e r a l , s i n c e the "optimum" c o v e r g l a s s i s assumed t o be normal g l a s s of index - 1 . 5 1 . L i t t l e can be done about t h i s except to t r y i n v e r t i n g the plate (or g r i n d i n g i t t h i n n e r ) . At the other extreme,' i f one looks at an i n c l u s i o n very near the surface of a p o l i s h e d p l a t e , there w i l l be i n s u f f i c i e n t apparent c o v e r g l a s s t h i c k n e s s . In such cases, adding a drop of o i l and a c o v e r g l a s s can improve the image. The i n t e r r e l a t i o n s h i p s of v a r i o u s photographs and the r e l o c a t i n g of photographed areas can be very simple i f an adequate s e r i e s of numbered photographs are taken at i n c r e a s i n g m a g n i f i c a t i o n s . On each photograph i s drawn the approximate o u t l i n e of the area covered by subsequent photographs at higher m a g n i f i c a t i o n , with t h e i r photograph numbers i n d i c a t e d (see a l s o " R e l o c a t i n g I n c l u s i o n s , " above). F a s t moving bubbles, as in very small i n c l u s i o n s and p a r t i c u l a r l y in CO2 i n c l u s i o n s , may be b l u r r e d i n ordinary photography. Using a good IR f i l t e r , and reducing the temperatures, may help, p a r t i c u l a r l y f o r CO2 i n c l u s i o n s . Increased l i g h t i n t e n s i t y cuts exposure time but speeds up the movement. I f f l a s h microscope i l l u m i n a t i o n i s not a v a i l a b l e , use of high speed f i l m g r e a t l y cuts exposure time. As the pseudo-Brownian movement i s very e r r a t i c , sometimes one of a s e r i e s of exposures w i l l happen to have occurred during a time of l i t t l e movement. Such bubble movement can sometimes be made use o f , to permit photography of the bubble in the optimum p o s i t i o n .

174

CUTTING UP PLATES FOR MICROTHERMOMETRY A f t e r the i n c l u s i o n s t o be s t u d i e d have been s e l e c t e d and a l l necessary sketches and photographs made f o r documentation, f i n a l p r e p a r a t i o n of the p l a t e s f o r t h e h e a t i n g / c o o l i n g stage can be s t a r t e d . Although two h e a t i n g / c o o l i n g stages (the U . S . G . S . and t h e Linkham s t a g e s ) w i l l accept l a r g e samples, only one (the U . S . G . S . ) w i l l permit viewing of the whole sample; the others have a very l i m i t e d f i e l d of view (only a few m i l l i m e t e r s ) . Thus, unless the U . S . G . S . stage i s used, the doubly p o l i s h e d p l a t e s , whether on a g l a s s s l i d e or not, must g e n e r a l l y be broken or cut i n t o s m a l l e r pieces f o r microthermometry. To avoid loss of v a l u a b l e data, each stage i n d i s a g g r e g a t i o n of a p l a t e (sawing, breaking, e t c . ) should be w e l l documented, so t h a t i n t e r r e l a t i o n s h i p s of i n c l u s i o n s t o each o t h e r , t o growth zones, and t o the whole c r y s t a l can be r e c o n s t r u c t e d . If the p l a t e i s t h i c k enough t o be handled w i t h c a r e , i t i s best t o remove i t from the g l a s s s l i d e . T h e r m o p l a s t i c cements can be s o f t e n e d and the p l a t e s l i d s i d e ways, f o l l o w e d by c l e a n i n g w i t h an a p p r o p r i a t e s o l v e n t and a s o f t swab ( e . g . , Q-tips). The g e n t l e s t procedure i s simply immersion i n s o l v e n t . A f t e r removal, u l t r a s o n i c c l e a n i n g i s best i f the p l a t e w i l l not break up, as i t helps t o get the mounting r e s i n s , which might s p o i l the run, out of cracks and pores. Methyl a l c o h o l i s the best s o l v e n t f o r Lakeside cement; acetone tends t o convert i t i n t o a s o f t gum and e v e n t u a l l y d i s s o l v e s most of i t but leaves a r e s i d u e . Complete removal of a l l r e s i n i s e s s e n t i a l and may be s u r p r i s i n g l y t e d i o u s , as any small p o r t i o n s l e f t tend t o coat the p o l i s h e d s u r f a c e d u r i n g the f i n a l c l e a n i n g . After c l e a n i n g and sawing (or b r e a k i n g ) the p l a t e , the s u r f a c e s of t h e chip above and below t h e s e l e c t e d i n c l u s i o n s should be checked w i t h the microscope, as nominally " i n v i s i b l e " f i n g e r p r i n t s or r e s i d u e s from d i r t y s o l v e n t can wreak havoc with the o p t i c a l image at high m a g n i f i c a t i o n . Thin p l a t e s , p a r t i c u l a r l y of minerals t h a t are f r i a b l e and those t h a t cont a i n only r e l a t i v e l y low-temperature i n c l u s i o n s , can be l e f t on the g l a s s s l i d e . Mark the p a r t s on which runs are t o be made ( e . g . , w i t h I n d i a i n k ) , and then remove them by c u t t i n g out wedges of sample (and g l a s s s l i d e ) w i t h a diamond c i r c u l a r saw, h o l d i n g t h e s l i d e i n the hand r a t h e r than i n a clamp. A diamond w i r e saw i s even more g e n t l e than a c i r c u l a r saw f o r t h i s o p e r a t i o n but i s slow. The cuts are made so as t o p l a c e the d e s i r e d i n c l u s i o n s near the p o i n t of the wedge, t o permit s e v e r a l chips t o be p l a c e d i n t h e h e a t i n g stage s i m u l t a n e o u s l y , s t i l l mounted on the g l a s s wedges, a l l w i t h t h e i r s e l e c t e d i n c l u s i o n s w i t h i n a radius of s e v e r a l m i l l i m e t e r s . To keep the thermal mass t o a minimum, a c o v e r glass can be s u b s t i t u t e d f o r t h e o r i g i n a l glass s l i d e (R.E. Bennet, J r . , p e r s . comm.).

ARTIFACTS AND THE PROBLEMS THEY CAUSE Henry Baker, i n an e a r l y book "The Microscope Made Easy" (1743), that c a u t i o n was needed i n microscopy:

indicated

"Beware of d e t e r m i n i n g and d e c l a r i n g your Opinion suddenly on any O b j e c t ; f o r Imagination o f t e n gets the the S t a r t of Judgment, and makes People b e l i e v e they see Things, which b e t t e r Observations w i l l convince them c o u l d not p o s s i b l y be seen: t h e r e f o r e a s s e r t nothing t i l l a f t e r repeated Experiments and Examinations i n a l l L i g h t s and i n a l l P o s i t i o n s . When you employ the Microscope, shake o f f a l l P r e j u d i c e , nor harbour any f a v o u r i t e O p i n i o n s ; f o r , i f you do, ' t i s not u n l i k e l y Fancy w i l l betray you i n t o E r r o r , and make you t h i n k you see what you would wish t o see. Remember t h a t Truth alone i s the Matter you are i n search a f t e r ; and i f you have been mistaken, l e t not V a n i t y seduce you t o p e r s i s t i n y o u r M i s t a k e . " 175

During optical examination, a variety of spurious images and apparent or real optical artifacts are commonly seen; the microscopist soon learns to ignore them, but they can cause considerable confusion and wasted time when first encountered. These effects become most notable at higher magnifications, particularly at lOUUx and above. The following are a few of the more common examples that I have encountered either in my own work or in working with students. Inclusions that are intersected by a polished surface appear very different from those fully within the plate, and hence such "strange" inclusions are commonly spotted by beginners and professionals alike. Tiny droplets of oil from fingerprints or former mounting media on the surface are also common sources of confusion, as are polishing pits and abrasive grains or other debris within them. Cork dust and even dandruff flakes, particularly in an oil mount, provide some odd apparent "inclusions." Mounting media, both impregnating resins and immersion fluids, yield a separate set of artifacts. Thus if an index liquid or an embedding medium or cement flows into the cavity of an opened fluid inclusion or surface pit, particularly if incomplete filling leaves a bubble in the fluid, very realistic but spurious inclusions can be formed. Leaching with solvents before observation, particularly in an ultrasonic bath, can avoid some but not all such artifacts. Opened inclusions that have become completely filled with such fluids may look like normal single-phase inclusions. Comparison of the index of refraction of the "inclusion" fluid with the host will sometimes help clarify matters, but only if one has the data on the specific mounting medium used. If the recognition of such artifacts presents a serious problem, a coloring dye may be added to the impregnating or mounting resin. All such problems are easily eliminated once the observer realizes his plane of focus is at the upper or lower polished surface, or in the mounting medium above or below. At low magnification, and when the sample is poorly polished, this is simple. But in a well polished sample of a clear mineral, at high magnification in a medium of matching index, this upper (or lower) surface is sometimes difficult to recognize. Usually there are a few imperfections in the polish, dust specks on the surface, etc., that mark it, and any crack or grain boundary can be followed by focussing up (or down) to the surface. A spherical bubble in liquid (or even a roughly spherical fluid inclusion) acts as a negative lens in the optical system and may bring another lower inclusion (or even a substage diaphragm) into sharp focus, simulating another bubble within the first, and this frequently leads to an erroneous conclusion that liquid and gaseous CO2 phases are present (Fig. 6-6). (Simple heating of the slide to above 31°C, where a liquid-gas meniscus is no longer possible in a CO2 system, can remove this ambiguity.) Gemologists frequently use dark-field microscopy when examining gems for inclusions. Although darkfield illumination does show the inclusions brilliantly, it has some shortcomings. Thus, these illumination devices generally result in an optical artifact of an additional apparent bubble inside each real bubble, and one type results in two such apparent bubbles (Fig. 6-7). Total reflection from a somewhat planar, inclined phase boundary can simulate the appearance of an opaque phase, until the sample plate is sufficiently tilted to place the phase boundary more nearly perpendicular to the line of sight (Fig. 6-8). Very small colorless objects (+10°C), after refreezing down to about -50°C. Finally, Th CO2 L-V can be measured (if above Tm clathrate). In some samples, Th CO2 L-V will be found to be at a lower temperature than Tm CO2 clathrate, and occasionally both Tm CO2 clathrate and T m ice can be measured. After the freezing determinations are complete, homogenization determinations at above room temperature can proceed. The mode and sequence of homogenization must always be recorded; far too often in the literature these observations have been left ambiguous. Most ordinary G-L inclusions containing small bubbles homogenize in the liquid phase (i.e., by expansion of the liquid to eliminate the bubble). Low-density aqueous inclusions and many C O 2 inclusions homogenize in the vapor phase (i.e., by evaporation of the liquid). It is important to record also the behavior on heating of all solid phases originally present -- whether they seem unaffected, get smaller (or larger), or dissolve completely, and, if so, at what temperature(s). If homogenization occurs at a critical point (i.e., by fading of the meniscus), the behavior of this meniscus just before homogenization should be noted; if the meniscus moves toward the "gas" phase, the density of the inclusion was slightly greater than the critical density, and vice versa (see CO2 in Chapter 8). Because daughter crystals may not renucleate on cooling after homogenization, T m dms should be determined carefully on the first try; note also that Tm dms may be above or below Th L - V . Similarly, to avoid the effects of stretching from overheating beyond Th L - V , particularly in soft minerals, those inclusions having the lowest values for Th should be determined first. Any inclusion first seen after the sample has been heated may have been stretched. Even if stretching is not a problem, decrepitation of overheated low-temperature secondary inclusions may ruin the sample before the highest temperature data are complete. In order to avoid overheating any inclusion to >Th before Th is determined, visual estimates of the gas/liquid ratios are normally used to guide the sequence in which the inclusions are run. If the inclusion having the smallest ratio looks as though it would homogenize at 150°C, one may be tempted to set the heating stage to level out at ~125°C and to creep up from there. However, because the coefficient of expansion of water solutions varies inversely with the salinity, an inclusion filled with pure water at a given temperature will have a larger bubble (at room temperature) than one filled at the same temperature 200

with s a l i n e b r i n e . Hence, a l l the s e l e c t e d i n c l u s i o n s should be checked p e r i o d i c a l l y , even when the concentration i s on one. I n c l u s i o n s having wide dark borders present s p e c i a l problems, because small bubbles can disappear i n these areas of t o t a l r e f l e c t i o n . The use of a f l e x i b l e f i b e r - o p t i c s i l l u m i n a t o r from above or below can h i g h l i g h t the bubble i n some but not a l l such i n c l u s i o n s . Because some h y s t e r e s i s ( i . e . , d i f f e r e n c e ) normally e x i s t s between Th L-V on h e a t i n g and Tn V on c o o l i n g , a quick c y c l i n g of temperature near Th with an i n c r e a s i n g maximum temperature on each c y c l e can s t i l l permit r e l a t i v e l y accurate determinations of Th (see "The USGS s t a g e " ) . Because the heating i s very slow, the magnitude of the h y s t e r e s i s i s mainly a f u n c t i o n of the rate of c o o l i n g . This procedure permits an accurate determinat i o n of Th on i n c l u s i o n s f o r which only an estimate could be made otherwise. The small thermal mass of the USGS stage makes such c y c l i n g r e l a t i v e l y f a s t (Woods et a l . , 1981). The p r e s e n t a t i o n of i n c l u s i o n data must be adapted to the nature of the problem and the data. Histograms of Th or Tm ice on which s p e c i f i c i n c l u s i o n types or samples are separately i n d i c a t e d , sequences of histograms arranged i n the order of time of formation or sample depth, p l o t s of Tm i c e vs Th f o r the same i n c l u s i o n s , e t c . , may be a p p r o p r i a t e . Far too o f t e n , only ranges are given f o r several measurements, such as Tm and Th, without the actual data on i n d i v i d u a l i n c l u s i o n s , which may show a p o s i t i v e or negative c o r r e l a t i o n between the two v a r i a b l e s f o r i n d i v i d u a l i n c l u s i o n s . One or more groups of i n c l u s i o n s with s i m i l a r c h a r a c t e r i s t i c s w i l l sometimes be evident i n h i s t o g r a m s . If this grouping i s not a r e s u l t of sample s e l e c t i o n b i a s , i t can provide important evidence of s p e c i f i c pulses or stages of f l u i d movement. U n f o r t u n a t e l y , i t i s very d i f f i c u l t to avoid sample s e l e c t i o n b i a s . Great care i s needed to avoid the i n t r o d u c t i o n of bias i n t o the data by the s e l e c t i o n of the i n t e r v a l s to be p l o t t e d , p a r t i c u l a r l y with small numbers of random(?) p o i n t s . As Wise (1982, p. 887) has so a p t l y noted ( i n a d i f f e r e n t but very c l o s e l y p a r a l l e l c o n t e x t ) : "Unwanted peaks on bar-graph histograms are e a s i l y removed by s h i f t i n g the boundary between two b a r s . This s p l i t s the peak i n t o two s m a l l e r bars and r e l e g a t e s the data to ' n o i s e 1 . Conversely, new peaks can be generated." A continuum of data p o i n t s in a histogram can mean that a continuum of changing f l u i d s was involved; i t can a l s o r e s u l t from poor p r e c i s i o n i n the measurements or other procedural problems. Television options. Commercially a v a i l a b l e t e l e v i s i o n cameras, monitors, and videotape recorders can be used with a t r i n o c u l a r head t o record and to demonstrate phase behavior i n f l u i d i n c l u s i o n s to students and c o l l e a g u e s , and a l s o to preserve tapes of the phase t r a n s i t i o n s f o r l a t e r use ( e . g . , H a l s o r et a l . , 1983). Commercially a v a i l a b l e " t i t l e g e n e r a t o r s " (or a second camera and s c r e e n - s p l i t t i n g device) may be used to allow the temperature readout to be displayed along with the optical image. Audio t r a c k s on the tape allow verbal comment. For f l u i d - i n c l u s i o n work, cameras, monitors, and video tapes having more than 700 lines/frame r e s o l u t i o n (monitors having 1000 or even 2500 l i n e s / frame are a v a i l a b l e ) should be used, because many d e t a i l s cannot be seen in f l u i d i n c l u s i o n s at 500 l i n e s / frame. Color cameras and monitors are l i m i t e d to about 400 l i n e s / f r a m e . Black and white cameras are much l e s s expensive and generally adequate f o r teaching and l a b o r a t o r y record purposes. The r e s o l u t i o n of a l l the recorders i s l i m i t e d ( i n p r a c t i c a l terms) to about 500 lines/frame, so better r e s o l u t i o n i s obtained i n l i v e time than when played back from the recorder. Cali bration The sources of e r r o r to be corrected.

201

The c a l i b r a t i o n of

heating/freezing

Sample C 310 ^ 0£ O
Tt). Similarly, inclusion (f), which trapped liquid plus a small vapor bubble, homogenizes in the liquid phase at (F 1 ) (also too high). In my experience, trapping of primary gas, although common in geothermal samples, certain types of ore deposits, and some other environments, is relatively rare. Where it does occur, large numbers of inclusions may show evidence of it. In studies of those ore deposits in which inclusions with widely variable gas/liquid ratios are found, it is important to verify whether or not "boiling" has actually occurred, since both true boiling and particularly effervescence can cause ore deposition. In addition to "boiling," variable gas/liquid ratios can be caused by trapping at different times from fluids under different P-T conditions, by leakage of part of the inclusions, or by necking down (discussed in Chapter 3), but such processes can usually be recognized by microscopy. In the simplest case, "boiling" will result in two types of inclusions, representing the trapping of either the liquid phase or the vapor phase. Sometimes the latter are called "steam" inclusions, solely on the (inadequate) basis of low density. These inclusions contain a very little liquid and a very large bubble at room temperature. If they trapped only the vapor phase, on heating they will homogenize in the vapor phase by evaporation of the liquid (i.e., Th L + V(V)), at the same temperature as the liquid inclusions homogenize by expansion of the liquid to eliminate the vapor bubble (i.e., Th L + V(L)). (The determination of Th L + V(V) will be accurate only when the steam inclusion has a narrow reentrant into which the last bit of fluid phase has been concentrated by capillarity, e.g., Fig. 9-2.) It is apparent from studies of fluid inclusions that small amounts of liquid are more commonly trapped with the steam than are small amounts of vapor (i.e., a small bubble) trapped with the liquid. It is possible, although hazardous, to use the minimum temperatures determined on a large number of presumably coeval inclusions (e.g., a,b,c, and f on Fig. 9-la), as a maximum value for the true Tt, as such inclusions are likely to have trapped pure liquid or pure vapor.

256

F i g u r e 9 - 2 . Two pseudosecondary i n c l u s i o n s of "steam" (S) and 4 of l i q u i d , trapped i n q u a r t z from a b o i l i n g geothermal system. I f the steam i n c l u s i o n i n the center d i d not have the narrow r e e n t r a n t i n which the l i q u i d phase (L) i s c o n c e n t r a t e d by c a p i l l a r i t y , the l i q u i d would form a t h i n f i l m on the w a l l s and be i n v i s i b l e . Sample Ka-19 from Kawerau, New Zealand, c o u r t e s y of P . R . L . Browne.

Experimental Problems In addition to the normal experimental problems of sample selection, temperature control and measurement, c a l i b r a t i o n of stages, etc., discussed in e a r l i e r chapters, several problems may make the experimentally best of measurements grossly inaccurate. One such problem, caused by surface tension, can y i e l d Th values that are far too low (see Chapter 10); i t becomes serious only with very small i n c l u s i o n s ( i . e . , 2-3 um). A s e r i e s of special problems in melt-inclusion studies are covered in appropriate l a t e r chapters. Two other problems, much more general in occurrence, are stretching and leakage. The pressure within a f l u i d i n c l u s i o n can vary greatly, depending upon the composition and density of the contents and the temperature. As an example, a pure-water i n c l u s i o n containing l i q u i d and a vapor bubble has an internal pressure of only 24 mm (-0.03 bars) at 25°C ~ in other words, the bubble i s a moderately good vacuum. Figure 8-5 shows that i f the inclusion homogenizes at 215°C, the vapor pressure at that temperature i s ~20 bars. The pressure in the inclusion on heating above Th will depend upon the bulk density and hence on the way the i n c l u s i o n homogenizes. I f i t homogenizes at 215°C in the vapor phase (by evaporation of the l i q u i d ) , the density i s 0.01 g/cm3, and the pressure in the i n c l u s i o n will increase very s l i g h t l y with temperature along the isochore in Figure 8-5 marked " . 0 1 . " I f , however, the inclusion homogenizes at 215°C in the l i q u i d phase (by expansion of the l i q u i d , eliminating the vapor bubble), the density i s 0.85 g/cm3, and the pressure w i l l increase rapidly, at about 14 bars/ °C, along the ".85" isochore. Hence even a small amount of "overheating" ~ heating beyond Th — can y i e l d high internal pressures. Substances other than pure water have different rates of pressure increase (isochore slopes), and some isochores are strongly curved, but most behave as shown in Figure 8 - 5 . The i n i t i a l pressure of l i q u i d CO2 inclusions i s higher, e.g., 70 bars at 31°C, but the slopes of the isochores are much lower than those for pure water. The slopes of the various isochores are discussed in more detail l a t e r in t h i s chapter, under "PRESSURE OF TRAPPING." As the internal pressure builds up during continued heating, four different events may take place. The inclusion may stretch, i t may undergo partial decrepi t a t i o n , i t may leak, or i t may decrepitate completely. The f i r s t two events are discussed in t h i s section; the second two are covered in the next. Stretching and partial decrepitation. Stretching refers to a permanent deformation of the host crystal around an i n c l u s i o n , generally without v i s i b l e 257

100

H2L»

Figure 9-3. I n c l u s i o n A/4 from f l u o r i t e sample ER 65-115, from L a r s o n et a l . ( 1 9 7 3 ) , photographed i n t r a n s m i t t e d l i g h t at room temperature before ( l e f t ) and a f t e r ( r i g h t ) o v e r h e a t i n g . A cleavage crack (arrow) i s v i s i b l e as a r e s u l t of the expansion on o v e r h e a t i n g , but such c r a c k s f r e q u e n t l y are not v i s i b l e . Note a l s o the l a r g e r bubble a f t e r h e a t i n g . A l t h o u g h the p r e s s u r e i n t h i s i n c l u s i o n at the moment of c r a c k i n g may have been as h i g h as 1300 b a r s , i t has not cracked to the s u r f a c e , as i t (and many other s i m i l a r i n c l u s i o n s ) continued to g i v e c o n s i s t e n t almost r e p r o d u c i b l e Th d e t e r m i n a t i o n s , 40-70°C too h i g h . A small i n c l u s i o n i n lower r i g h t (not run) a l s o -»shows a f a i n t crack a f t e r the h e a t i n g .

¡¡f^

cracking, that may occur in soft minerals such as f l u o r i t e i f the internal pressure exceeds a certain f i n i t e l i m i t , e.g., by overheating beyond Th (in the laboratory or in nature), or by expansion of ice on freezing an inclusion that had a bubble so small that i t was eliminated on freezing (Lawler and Crawford, 1983; see also Chapter 10). This expansion or stretching may be evidenced only by a r i s e in Th in subsequent runsi/, and once the s p e c i f i c overheating limit i s exceeded (this l i m i t will vary with i n c l u s i o n s i z e as well), the increase in Th i s a s u r p r i s i n g l y regular function of the amount of overheating and hence internal pressure (Bodnar and Bethke, 1980, 1984). As a result of such stretching, either in nature or the laboratory, such inclusions can y i e l d reproducible but erroneously high values for Th that may well go undetected. Partial decrepitation i s a closely related phenomenon. A very few of the overheated inclusions may show a small fracture out into the surrounding crystal but not to the surface. The volume increase represented by t h i s crack partly relieves the 'pressure. The crack may subsequently heal, trapping numerous i n c l u sions of t h i s expanded f l u i d in the form of a halo of tiny secondaries around the larger central original i n c l u s i o n , which now has a lower density. Recognition of such partial decrepitation, f i r s t described by Lemmlein (1956a), i s relatively simple and seldom causes errors (Fig. 3-22). Touret (1977) described "exploded" inclusions from g r a n u l i t e s , and Swanenberg (1980) found numerous examples, which he called "decrepitation c l u s t e r s , " in the high-grade metamorphic rocks (granulites) of southwest Norway. Other than the i s o l a t i o n of some of the inclusion f l u i d in new secondary i n c l u s i o n s , I believe that no real difference e x i s t s between stretched and p a r t i a l l y decrepitated inclusions. However, the major mechanisms involved in the deformation may be different and may range from d i s l o c a t i o n creep to open fractures. Without an obvious halo of small secondaries, or a v i s i b l e fracture, the stretched inclusions present a much more serious hazard to thermometry. Stretching obviously i s related to the amount of overheating and to inclusion s i z e , but many other factors may well be involved, and hence i t s effects may be variable and e r r a t i c . The concept of stretching was introduced by Larson et a l . (1973) when i t was discovered as a result of an interlaboratory standardization. In t h e i r work, they found that inclusions in f l u o r i t e from east Tennessee had Th raised by as much as 71°C (from 118 to 189°C) as a result of unspecified amounts of overheating during early reconnaissance Th runs. Only later were a few such inclusions found to have tiny cracks (Fig. 9-3). Cracks in general are i n v i s i b l e i f t h e i r T h e o r e t i c a l l y , the s t r e t c h i n g should be e v i d e n t in an i n c r e a s e i n the diameter of the vapor bubble at room temperature a f t e r the run. Although such bubble measurements s h o u l d always be made before and a f t e r a run, p a r t i c u l a r l y to detect leakage, note that the s t r e t c h i n g that r e s u l t s i n a 10°C r i s e i n Th of a water i n c l u s i o n from 215 to 225°C w i l l cause an i n c r e a s e i n the diameter of the bubble at room temperature of only 0 . 5 1 , an amount f a r too small to detect by most bubble-measurement p r o c e d u r e s .

258

width i s a small fraction of the wavelength of the l i g h t used, and t h i s lower limit i s largest i f the difference in index of refraction of host and the f l u i d f i l l i n g the crack i s small (e.g., f l u o r i t e with r> = 1.43 and aqueous brines of n = 1.36). The few cracks seen were far too small to form the halo of secondary inclusions characteristic of partial decrepitation, and many inclusions showed no v i s i b l e cracks. However, the walls of the inclusions had expanded enough to increase the volume of the (normally small) bubble measurably (Fig. 9-3). Stretching represents a permanent deformation of the w a l l s , permitting expansion of the f l u i d contents, until the pressure i s reduced to the point that e l a s t i c deformation of the surrounding undisturbed host prevents further propagation. Probably several mechanisms are involved; small cracks are obvious in some examples (Fig. 9-3), but p l a s t i c deformation may be involved in others. Presumably some of t h i s expansion i s reversed on cooling, which closes the cracks appreciably but not completely (as indicated by the increase in bubble s i z e at room temperature). Repeat determinations of the new higher Th values on the f l u o r i t e from east Tennessee yielded results that were either identical or increased one or a few degrees. The smallest inclusions showed the smallest amount of stretching. Inclusions in sphalerite from L a i s v a l l , Sweden, were also believed to have stretched. The high index of refraction of sphalerite causes heavy black borders on the i n c l u s i o n s . Roedder (1968d, p. 394) noted during the o r i g i n a l Th runs that "...Some inclusions in sphalerite . . . in which the very small bubble had seemingly disappeared (into the dark borders) at about 150°C suddenly showed a very tiny bubble in rapid Brownian [ s i c * / ] movement at 180-190°C; these homogenized at temperatures up to 223°C, and the process [ i . e . , homogenization] could be repeated." This behavior was misinterpreted at the time. I now believe that the reappearance of the bubble at ~180°C was due to stretching and that the true Th for these inclusions should have been about 150°C (Larson et a l . , 1973). One of the most extreme examples of stretching of i n c l u s i o n s i s seen in halite (Roedder and Belkin, 1979a, 1980a). Such inclusions decrepitate i f they are heated rapidly to temperatures above Th. However, s a l t i s so p l a s t i c , part i c u l a r l y in the presence of water, that decrepitation w i l l not occur i f the heating rate i s slow; instead, once Th i s reached and the internal pressure starts to increase rapidly (~14 bars/°C), the host s a l t simply expands by permanent p l a s t i c deformation. Thus, inclusions having t i n y bubbles at room temperature and an original Th = 40°C, after heating to 250°C, w i l l have large bubbles at room temperature and will have a new Th near 250°C (Fig. 3-23). This expansion can happen even at low temperatures: a large i n c l u s i o n containing a tiny bubble, homogenizing at 20°C, was overheated to 40°C and cooled; on redetermination, Th had risen to 39°C. Obviously, s a l t samples must not be heated above Th before Th i s determined. However, I do not believe that measurements of Th on inclusions in s a l t have much meaning anyway. I f i n c l u s i o n s in s a l t can expand that e a s i l y under a few hours of internal pressure in the laboratory, I must assume that they have also expanded (or contracted) in response to natural d i f ferences in internal and external pressure, due to changes in P and T. Therefore, the inclusion volume we now see represents some rather unknown and a r b i trary "quench" conditions after a long period of sequential "anneals." Stretching can occur either in nature or in the laboratory. Sabouraud et a l . (1980) suggested that some f l u o r i t e samples from Pb-Zn ore deposits in France have been stretched by natural overheating, and Bodnar and Bethke (1980, 1984) believe that natural stretching i s a potentially serious problem. The latter investigators made a systematic laboratory study designed to determine how much overheating i s needed before stretching s t a r t s , the amount of stretching (in

S i n c e shown to be pseudo-Brownian (see Chapter

7).

259

°C change in Th) for any given overheating, and the c r i t e r i a that might be used to recognize stretched i n c l u s i o n s . This study showed the value of the USGS gasflow stage (see Chapter 7). The combination of very high precision and very rapid thermal response of t h i s stage permitted the necessarily large number (1300) of precision measurements for s t a t i s t i c a l validation of the complex r e l a tionships studied. This task would have been enormously time consuming i f the older conduction-heated stages had been used. Bodnar and Bethke (1980, 1984) found that f l u i d inclusions in f l u o r i t e and sphalerite stretched systematically and somewhat predictably. The amount of overheating needed to i n i t i a t e stretching depends upon the properties of the inclusion f l u i d , the inclusion s i z e and shape, and the physical properties of the host mineral. E s s e n t i a l l y n£ stretching occurred in f l u o r i t e on overheating 400°C. On Figure 9-4, the spreading of the isobars toward higher temperatures show that the pressure corrections for Th >400° are going to be s t i l l l a r g e r than at 400°C. Figures 8-2 and 9-1 show that the pressure c o r r e c t i o n s for pure-water i n c l u s i o n s with Th >350°C become very large. Thus an i n c l u s i o n of e s s e n t i a l l y pure water that formed at 700°C and 1200 bars pressure (point X on F i g . 9-1 a) would have Th = 375°C ( i . e . , at Cp. on F i g . 9 - l a ) and a pressure correction of 325°C. Such pressure corrections are very s t r o n g l y affected by composition, so i t i s very hazardous to try to estimate t h e i r magnitude unless the composition i s known. F i n a l l y , the pressure corrections of Potter (1977) as well as those of Lemmlein and Klevtsov (1961), although not so stated, are only v a l i d f o r i n c l u sions homogenizing i n the l i q u i d phase. Figure 9-1 shows that the pressure corrections f o r i n c l u s i o n s homogenizing in the vapor phase, p a r t i c u l a r l y those trapped at r e l a t i v e l y low pressure, w i l l be v a s t l y greater than those with the same Th and Ph, but homogenizing in the l i q u i d phase. Thus, an i n c l u s i o n homogenizing in the l i q u i d at 320°C and trapped at 250 bars would have a pressure correction of ~20°C ( i . e . , i t was trapped at ~340°C), whereas a s i m i l a r i n c l u s i o n also trapped at 250 bars but homogenizing in the vapor phase at 320°C would have been trapped at 700°C, y i e l d i n g a pressure correction of 380°C. Decrepitometry Considerable d i s c u s s i o n in the l i t e r a t u r e concerns the t h e o r e t i c a l s i g n i f i cance and p r a c t i c a l usefulness of the decrepitation method. Decrepitation can r e s u l t from a variety of f a c t o r s , such as the buildup of internal s t r e s s in a s i n g l e grain between two minerals that have d i f f e r i n g rates of thermal expansion; most commonly, however, i t i s a r e s u l t of the internal pressure w i t h i n f l u i d i n c l u s i o n s exceeding the t e n s i l e strength of the b r i t t l e host mineralZ/ at that temperature, causing i t to break. I n c l u s i o n s having a high degree of f i l l i n g (and hence low Th) have a very abrupt increase in A P / A T at Th (as much as 14 bars/°C). This abrupt increase i s the theoretical b a s i s for the method, since i t i s assumed that Td Th, or at l e a s t that Td i s a known function of Th. However, Td i s also a function of so many v a r i a b l e s other than Th that any agreement of Td with Th i s considered by some to be e s s e n t i a l l y only c o i n c i d e n t a l . In

U

D e c r e p i t a t i o n data have even been p u b l i s h e d f o r g r a i n s o f d u c t i l e m e t a l l i c

264

Au.

addition, the change in AP/AT is much less abrupt or even-negligible at Th for lower density, higher temperature inclusions, and AP/AT actually decreases (but stays positive) above Th in those inclusions containing less than the critical density of filling (i.e., Th L-V(V); see Fig. 8-5). Most of the objections to the decrepitation method are founded on this lack of a theoretical basis for at least some inclusions, the commonly observed lack of agreement with actual Th data (e.g., Little, 1955, 1960), and the decrepitation in samples apparently free of visible fluid inclusions (e.g., Kennedy, 1950a; Stephenson, 1952). Also, some degree of subjectivity is involved in eliminating "anomalous" decrepitation (Stephenson, 1952; Smith and Little, 1953) and in selecting the appropriate spot on the decrepigram for the "start of decrepitation" of a given generation of inclusions (Kennedy, 1950a). In addition to instrumental problems, scatter in the results can stem from the irregularity in the size of the inclusions (large ones have lower Td); their shape (jagged irregular ones decrepitate more easily than smooth equant ones); variations in composition (e.g., NaCl:H20:C02) that cause gross differences in internal pressures; abundance of inclusions (if abundant they can act in concert); arrangement of inclusions in planes, which makes decrepitation easier; variations in heating rate; grain size of the host mineral; and variations in the toughness or brittleness of the mineral (i.e., the "snap" of Smith and Little, 1953) and their changes with temperature. These factors all result in large variations in the amount of overshoot (heating above Th) before decrepitation. The literature contains many records of overshooting of 50 or 100°C without decrepitation or leakage, and Roedder (1970b) reported some in quartz that could be overheated 600°C. Certain liquid-C02 inclusions in olivine may be heated 1200°C above their filling temperature without decrepitation (Roedder, 1965a). Inclusion size is a particularly important variable, as shown by studies of inclusions in quartz, which found that inclusions - 3 5 ym in diameter will decrepitate at ~850 bars internal pressure, but inclusions ~1 y m in diameter can withstand - 6 0 0 0 bars (Fig. 3-21). Even at a given size of inclusion, the wall thickness (i.e., the position of the inclusion within the grain) is an important parameter. Mechanical stress from thermal gradients in the grains can also cause decrepitation and hence must affect the temperature at which inclusions decrepitate. In a study of the evolution of gases from decrepitating CO2 inclusions in olivine, Killingley and Muenow (1975b) found an increased rate of decrepitation during cooling. Apparently the sample thermal gradients in their apparatus during cooling were larger than those during heating, and hence the stress was greater. Attempts to quantize the many variables and phenomena involved have been only partly successful. Khetchikov and Samoilovich (1970) have examined the difference between Tt and Td of synthetic quartz grown at known temperatures and pressures and have found that Td can range from 100°C below to 160°C above Tt, depending on Pt. Wilkins and Ewald (1982) have reexamined the decrepitation process in an attempt to develop a theoretical basis for it and suspect that at least four distinct mechanisms of decrepitation appear to be involved. The above discussion shows that the decrepitation method has serious problems in both theory and practice, that interpretation of its results is complex and difficult, and that the method probably will never be truly calibratable to yield a valid measure of Th (or Tt), free of error and subjectivity. The only possible exception may be runs made on rather uniform sample suites that have been previously standardized by the homogenization method (e.g., Khoteev, 1980). However, the method is rapid, inexpensive, requires little training, and integrates the results of many hundreds or thousands of inclusions. It probably is most useful for purely empirical screening, to recognize differences between samples (e.g., Burlinson et al., 1983; Wilkins et al., 1983), or the presence of several different generations of inclusions, particularly in the opaque minerals 265

and in the low-temperature range. Since the method can readily be adapted to f i e l d use (Ermakov, 1966b, reported handling 90 samples per working day for a f i e l d u n i t ) , i t has been used most effectively in exploration. Some examples from the l i t e r a t u r e have been given by Roedder (1977b), in which a "steam halo" of high-temperature inclusions surrounding a deposit provided a larger target than the deposit i t s e l f and revealed blind deposits (see also Demin, 1970; Ermakov and Kuznetsov, 1971; and numerous other references in Roedder, 1968a). Workers in the People's Republic of China have recently expanded the use of decrepitation in ore-deposit studies (e.g., see many of the 54 papers in a recent symposium - Academia S i n i c a , 1981), and they have prepared a standard sample for decrepitation work (same reference, p. 115). As a warning to the reader, I must admit my personal bias against the use of decrepitation data in any quantitative connotation. More than 400 papers have been published, mainly Russian, in which decrepitation data are reported, but very few of these are referred to in t h i s volume. However, not a l l Russian workers accept decrepitation temperature data at face value (e.g., see Khetchikov and Samoilovich, 1970; Dmitriyev, 1970; Butuzov et a l . , 1971; Pal'mova, 1972; Bobolovich, 1972; Barsukov and Sushchevskaya, 1973; Pashkov and Piloyan, 1973; Sharonov et a l . , 1973; and other entries in indices in Roedder, 1968a). PRESSURE OF TRAPPING (Pt)£/ Introduction The use of f l u i d inclusions as geobarometers, to determine the pressure of past environments, i s intimately related to thei r use as geothermometers. Since Sorby's 1858 c l a s s i c paper on the use of f l u i d inclusions for geothermometry, inclusions have been the subject of numerous papers, mostly in the Russian l i t e r ature, and have provided a s i g n i f i c a n t part of the data in many more. In these many papers, two aspects are most commonly mistated or misunderstood: (1) the effect of the hydrostatic pressure at the time of trapping of a given inclusion on the thermometric results obtained, and (2) the use of inclusion data themselves to obtain an estimate of t h i s pressure. In many, i f not most, inclusion i n v e s t i g a t i o n s , the pressure i s not determined from the i n c l u s i o n s . Most i n c l u sions have trapped f l u i d s at pressures higher than t h e i r vapor pressures ( i . e . , in the one-phase " f l u i d " f i e l d above the two-phase curve shown in Figure 9 - l a ) . Generally, the pressure i s estimated from independent evidence of the depth of cover at the time of trapping (e.g., from geologic reconstructions of the t h i c k ness of material since removed by erosion or f a u l t i n g ) ; then, t h i s pressure i s used, along with P-V-T data on supposedly appropriate s o l u t i o n s , to calculate the pressure correction. The following section shows why some commonly used procedures for evaluating pressure from inclusion studies ( i . e . , inclusion geobarometers) are wrong and can y i e l d very erroneous pressure values. I t also reviews the variety of valid inclusion geobarometers, along with their precision, accuracy, l i m i t a t i o n s , and applications. Evidence of b o i l i n g and data on the pressures e x i s t i n g during geologic processes are of far more than academic interest. B o i l i n g can cause ore deposition, and because pressure data provide some evidence of the depth beneath the surface at which the process occurred, they can provide the exploration geologist with valuable information on the amount of cover that has been faulted or eroded away and the possible nature of the deposit. Pressure differences may eventually also y i e l d information on the direction of flow of the ore-forming f l u i d s .

8/ T h i s s e c t i o n taken w i t h some m o d i f i c a t i o n from Roedder and Bodnar

266

(1980).

General P r i n c i p l e s and Nature of Available Data Relationship to inclusion geothermometry. The s i x assumptions that form the basis of geothermometry using inclusions (see beginning of t h i s chapter) are equally pertinent to geobarometry. The f i r s t assumption, the trapping of a single f l u i d phase, i s essential to geothermometry, but as will be shown l a t e r , evidence of the trapping of a nonhomogeneous mixture of two f l u i d phases, part i c u l a r l y l i q u i d and vapor, can provide an excellent geobarometer. The second assumption, no change in volume, i s of r e l a t i v e l y minor concern in most geobarometry because the most common volume change, that due to precipi t a t i o n on the walls during cooling, i s generally not only small but also e a s i l y reversed during heating in the laboratory, except in s i l i c a t e melts. Very s i g n i f i c a n t errors may occur, however, i f care i s not used in the determination of Th, because of permanent deformation (stretching) of the host-mineral walls from overheating. The errors from d i l a t i o n (see "Changes i n Volume," Chapter 3) are well below the errors from other sources, and they can generally be ignored at t h i s time. Assumption number three, that nothing i s added or l o s t , may be pertinent to geobarometry in several s i t u a t i o n s . The l o s s of H from inclusions by d i f f u sion through the host mineral has been suggested as a possible explanation for some puzzling daughter minerals that do not homogenize (Roedder and Skinner, 1968). As the hydrogen presumably comes from the d i s s o c i a t i o n of inclusion H2O, the volume change would be effectively controlled by the chemical behavior of the oxygen l e f t behind. This oxygen may oxidize components of the f l u i d , such as s u l f i d e to sulfate, or, in hydrous s i l i c a t e - m e l t i n c l u s i o n s , i t may diffuse into the host (Anderson and Sans, 1975, and pers. comm.). The volume changes within the inclusion in these examples might be d i f f i c u l t to predict. Assumption four, that the effects of pressure are i n s i g n i f i c a n t , or are known, refers to the fact that Th of the vapor and l i q u i d phase of an inclusion establishes only a minimum value for the Tt. All i n c l u s i o n s of a given composition trapped along the isochore o r i g i n a t i n g at the Th point on a P-T plot ( F i g . 9 - l b ) will have the same homogenization behavior, so either P or T must be known to determine the other. This interrelationship i s examined in more detail below. The f i f t h assumption, concerning the o r i g i n of the i n c l u s i o n , causes problems in many inclusion studies. This ambiguity, f r u s t r a t i n g to the student of i n c l u s i o n s , does not affect the v a l i d i t y of most of the geobarometers presented here, but i t i s of major concern to some. P r a c t i c a l l y all estimates of either the temperature or the hydrostatic pressure at the time of formation of the host mineral from inclusions require two types of data: (1) the composition of the f l u i d phase (or phases) trapped, and (2) the phase behavior and P-V-T-X properties of that composition in the range involved. Unfortunately, these two main requirements place severe cons t r a i n t s on the accuracy of a l l pressure determinations based on i n c l u s i o n s . Obviously, the d i f f i c u l t y in estimating the depth of cover, plus the uncertainty of hydrostatic vs l i t h o s t a t i c pressure described above, and the errors in estimations of the composition of the f l u i d , and hence i t s P-V-T characterist i c s , can cause r e l a t i v e l y large uncertainties in the pressure correction. Fortunately, absolute geothermometric accuracy i s sometimes only of minor concern, e.g., in current studies in ore deposTtion, where the relative temperature values for various samples may provide the most valuable data. When the pressure corrections are large, the limitations are such that most commonly the geothermometric or geobarometric data from f l u i d inclusions are compared with independent evidence from other sources in an attempt to a r r i v e at a consensus and to recognize which data or assumptions are i n v a l i d , and why. 267

Composition of the f l u i d phase(s) trapped. Many inclusion studies involve only a qualitative identification of the phases present (e.g., liquid-water solution, halite or other daughter c r y s t a l s , and vapor bubble) and a crudely quantitative estimate of the relative volumes and hence masses of these various phases. As seen in earlier chapters, these volume estimates can be precise but seldom accurate, and quantitative data, from heating and cooling studies, are absolutely necessary for pressure estimates. The f l u i d s present around a growing or healing crystal are not always homogeneous. In many places, two immiscible f l u i d s , such as H2O and CO2, or water and steam, were present. Fluid inclusions forming in such a heterogeneous f l u i d environment may trap only one of the f l u i d s , or both of them in random ratios. Such compositionally divergent inclusions provide several of the geobarometers described below, but compositional data are required on both f l u i d s . The use of data from more than one kind of inclusion to provide a geobarometer must be based on the assumption (often unstated) that the several inclusions used are truly contemporaneous ( i . e . , cogenetic). Validation of this assumption i s a far from t r i v i a l matter and has led to many dubious or erroneous data in the 1iterature. Phase behavior and P-V-T-X properties of f l u i d phase(s) trapped. Any determination of pressure based on study of a single inclusion requires compositional data on the inclusion contents and P-V-T-X data on such a composition. The available data on the phase behavior and P-V-T-X properties of appropriate compositions in the range involved in f l u i d - i n c l u s i o n studies are meager, and even at best require considerable extrapolation. P-V-T data are most complete on the two pure compounds that are major components of many inclusion f l u i d s , H2O and CO2 (for details and references, see Chapter 8), and these data are frequently used in interpreting f l u i d inclusions. Natural f l u i d s are not pure, however, because the water inclusions often are at least one molal in ionic solutes, and CO2 inclusions frequently contain appreciable CH4 and N2, so the properties of the pure compounds are useful only as an approximation. For water solutions, the most extensive experimental data available are for the system NaCl-H20, but even t h i s system has not been studied in the entire range of interest. In addition, most natural f l u i d s are not simple solutions of.NaCl and H2O but contain s i g n i f i c a n t amounts of other solutes, such as K + , Ca , S0|", and H C O 3 . Although Potter and Clynne (1978b) showed that many of the solutes present in natural inclusion f l u i d s result in f l u i d s that have thermodynamic properties rather close to those measured for simple NaCl-H20 solutions that have the same value for the depression of the freezing point ( i . e . , Tm ice), extrapolation to fluids containing significant quantities of CO2 i s hardly warranted. The available experimental data on the system NaClH2O-CO2 are even more limited. Depth vs pressure; hydrostatic vs l i t h o s t a t i c . Most pressures determined from f l u i d inclusions are stated to represent either the " l i t h o s t a t i c " or the "hydrostatic" pressure, or some intermediate value. The local pressure at the s i t e of the inclusion at the time of trapping i s actually hydrostatic, in any case, because i t i s a f l u i d pressure, but the two terms are used to indicate the source of the pressure on the f l u i d . " L i t h o s t a t i c " i s generally used to refer to the pressure from a column of country rock of density and height appropriate for the depth of the sample below the surface at the time of trapping; "hydrostatic" usually refers to the pressure of a column of fresh water above the sample, of the appropriate temperature and hence density, corrected for depth to water table. Although the pressures are probably between these two limits in most natural situations, the actual values in nature may span a considerably wider range at both ends, and even within t h i s range numerous factors may affect the specific pressure. 268

GROUND SURFACE

A R U B BELOW BOILING nun THROUGHOUT mACTURE

B STRATIFIED

R-Uios eaow B0WNG POINT THROUGHOUT fflACTURE

C FLUID JUST AT BOIING POINT THROUGHOUT FRACTURE

0 FLUID BOIING ABOVE INCLUSION SITE

E NEARLY HYDROSTATIC PRESSURE AT INCLUSION POINT

G

F OPEN TENSION CAVITY. FlUO BOILING

HTERMEDIATE PRESSURE (SUM Of h, md

H UTHOSTATC PRESSURE

- HOST ROC* FUSTIC -

Figure 9-5. Diagram showing range of p o s s i b l e p r e s s u r e in a crystal growing freely into a f l u i d in a vein. See Bodnar (1980).

c o n d i t i o n s on a f l u i d text for discussion.

inclusion trapped From Roedder and

In Figure 9-5, diagram A represents the simplest p o s s i b i l i t y , in which the hot f l u i d moves up an open fracture. The f l u i d w i l l expand as pressure decreases, but the cooling will generally be much greater than adiabatic, as the f l u i d i s l o s i n g heat to the cooler adjacent rock before flowing out on the surface as a hot spring. The pressure at the inclusion would be that from a column of f l u i d of appropriate composition and temperature. The actual integrated density of the f l u i d column would be higher than that of the f l u i d trapped in the inclusion because the density increases toward the surface. A modif i c a t i o n of t h i s i s shown in diagram B, where the upper part of the column i s ground water. Because density increases with both s a l i n i t y and cooling, the pressure in A and B might be identical i f the integrated density of the cool dilute ground water i s the same as that of the hotter, more saline f l u i d s over the same interval of the upper portion of the vein length. I f the f l u i d in the vein above the inclusion i s at the point of b o i l i n g throughout i t s length ( d i a gram C), and the s a l i n i t y i s known, the data of Haas (1971) permit an unambiguous pressure determination. Unfortunately, however, there i s a fourth and probably common p o s s i b i l i t y , shown in diagram D, which may r e s u l t in pressures at the inclusion that are considerably l e s s than simple "hydrostatic" (as in diagram A), depending upon where in the column the switch from l i q u i d to vapor occurs. A vapor-dominated system, in which l i q u i d i s above vapor, would be treated similarly. The four p o s s i b i l i t i e s described above a l l involve a vein system assumed to be open to the surface. In diagrams E through H ( F i g . 9 - 5 ) , a t i g h t , almost closed vein or throttle (Toulmin and Clark, 1967) i s assumed. Where the host rocks are r i g i d , as shown in E, the pressure of a flowing f l u i d at the i n c l u s i o n s i t e will be the hydrostatic head, h, plus an unknown overpressure limited only by the vein c o n s t r i c t i o n . This overpressure could r e s u l t from the presence of 269

an igneous body below or from a hydrostatic load not d i r e c t l y above the vein, such as an artesian-type regional system. Diagram F i l l u s t r a t e s an extreme condition, in which fault movements have increased the volume occupied by a r e l a t i v e l y fixed volume of f l u i d ; t h i s increase causes b o i l i n g . The b o i l i n g would cease when the pressure in the cavity reached the vapor pressure of the f l u i d at that temperature, even though t h i s chamber might be deep in the earth. This low-pressure condition could p e r s i s t as long as the surrounding rock remains r i g i d . The same mechanism has probably been operative in many ore deposits, but only enough to cause migration of f l u i d s and not enough to cause b o i l i n g . Thus, the well-known concentration of ore or vein matter in the apices of folds i s frequently ascribed to the lower pressures there, due to the mechani c s of folding; such low pressures may well be reflected in the inclusion densities. In diagrams G and H, depths are assumed to be such that the country rocks are p l a s t i c . In G, the pressure would be the sum of the hydrostatic pressure of f l u i d column, h-|, plus the l i t h o s t a t i c pressure of rock column, h2; in H, i t would be simply l i t h o s t a t i c for column h. In any l i t h o s t a t i c pressure environment such as that shown in diagram H, the true pressure at the inclusion can be considerably less than l i t h o s t a t i c i f the country rocks are not completely p l a s t i c during the time involved. Also, the pressure in H can be more than l i t h o s t a t i c i f the horizontal extent i s small relative to the depth h, or i f these country rocks are under compression from regional forces. In such cases, the pressure in the chamber could be greater than l i t h o s t a t i c without breaking out to the surface. The only s i t u a t i o n in which the pressure would be t r u l y " l i t h o s t a t i c " would be the rather rare h o r i zontal vein having a large lateral extent relative to i t s depth, so that a f l a t slab of rock i s effectively " f l o a t i n g " on f l u i d under pressure. "Overpressures" can also be caused by the increase in vapor pressure during c r y s t a l l i z a t i o n of magmas; when roof rocks y i e l d suddenly, explosive volcanism may r e s u l t . Similar overpressures in sedimentary rocks (Barker, 1972) can result from the heating of formation waters trapped under impermeable beds. Owing to the large lateral extent, such f l u i d s a re not l i k e l y to be pressurized much above the l i t h o s t a t i c pressures. The pressure environment within the pores of a rock, e . g . , during metamorphism, i s much more d i f f i c u l t to quantify than that within a vein or open f r a c ture. Obviously, the pressure within disconnected pores between c r y s t a l s , or even within single c r y s t a l s , w i l l be close to the regional l i t h o s t a t i c load i f the host rocks or c r y s t a l s are completely p l a s t i c during the time involved. Thus, in the interpretation of inclusions in deep-seated rocks that have been unloaded over a s i g n i f i c a n t period of time, an important but unanswered question i s the following: what amount of change, i f any, has occurred in the inclusions during the slow decrease in pressure and temperature from the maximum values? This problem i s equivalent to that of the apparent "quenching temperature" during the annealing of various compositional differences between minerals used as geothermometers. Such annealing of inclusions apparently does occur in rocksalt (Roedder and Belkin 1979a); does i t occur in quartz? (See discussion in Chapter 3). Although low pressures within tension fractures in veins (Fig. 9-5F) may be r e l a t i v e l y rare, such low pressures may be the prevalent environment during the healing of the small disconnected tension fractures that have yielded the abundant secondary inclusions in many metamorphic rock samples. I f so, t h i s mechanism could explain some otherwise discrepant data, but apparently i t has not previously been suggested. D i s t i n c t i o n between low-density i n c l u s i o n s formed at r e l a t i v e l y great depths by t h i s mechanism and similar low-density inclusions formed by b o i l i n g in effectively open fractures at much shallower depths would be impossible from the inclusions alone and would require additional data. 270

Geobarometry based on vapor pressure of solution. If a group of two-phase, liquid-plus-vapor inclusions i n a given sample a l l have the same composition and Th, the general presumption i s that a homogeneous f l u i d was trapped. Thus, we can safely assume that the hydrostatic pressure at the time of trapping was greater than the vapor pressure of that particular solution at that temperature, or the f l u i d would have been b o i l i n g ( i . e . , two-phase). Such simple inclusions provide no evidence concerning how much the pressure might have exceeded the vapor pressure. Simple two-phase, liquid-plus-vapor inclusions are the most common by far; hence, this minimum (vapor pressure) vaTue i s generally the only constraint on the pressure that can be obtained from the inclusions themselves. For ore deposits formed at low temperatures, this vapor pressure may not result i n much of a constraint. Thus, only 37 m of hydrostatic head of cold water (at density = 1.0 g/cm3) i s required to prevent b o i l i n g of the t y p i c a l l y 150°C f l u i d s that formed the Mississippi Valley-type ore deposits (Roedder, 1976a). I f there are gases in solution, the vapor pressure of the mixed solution would be higher than that of the pure solvent l i q u i d , but appropriate P-V-T-X data are generally lacking. I f such inclusions are studied on the crushing stage, however, an estimation of the internal pressure, at room temperature, can be obtained. Such an estimate must be less, of course, than the pressure at Tt. Geobarometry based on comparison of Th with an independent geothermometer or geobarometer. It the composition and Th of a simple two-phase, liquid-plusvapor inclusion are known, and another independent geothermometer i s available to determine the (higher) temperature of formation ( i . e . , Tt) of the host or associated minerals, Pt may be determined from P-V-T data on the f l u i d of the inclusion. The uncertainty in the values derived from any such geobarometer i s obviously at least equal to the sums of the uncertainties in the precision and accuracy of the determinations by the two geothermometers used, as well as other f a c t o r s . The independent thermometer must be very accurate in order to provide useful results since a r e l a t i v e l y small pressure correction can be equivalent to a rather large pressure. Thus, for low-temperature ore deposits, an error of 25°C in the temperature obtained from the independent geothermometer i s equivalent to the hydrostatic pressure from >3 km depth of burial (Roedder, 1971d). Some published geobarometry data have been based on such poor "thermometric" data that the pressure values obtained (and the geologic speculation based i n turn on them) are v i r t u a l l y meaningless. The l i m i t a t i o n s imposed by inadequate P-V-T data on the inclusion f l u i d can also be severe. E s s e n t i a l l y , the Th for the f l u i d inclusion, along with knowledge of i t s composition, l i m i t s the possible conditions of trapping to a single isochore on a P-T plot for the appropriate composition f l u i d . I f another independent geothermometer-geobarometer can also be used on the same samples, and i t y i e l d s a l i n e having a different slope on a P-T p l o t , the point of intersection of these two lines w i l l be the Pt and Tt; because of experimental uncertainties in the determination of both l i n e s , the intersection becomes an area. Bethke and Barton (1971) showed that the d i s t r i b u t i o n of Mn between sphalerite and galena could thus be combined with f l u i d inclusion Th to y i e l d both Pt and Tt. Fortunately, the slopes of the data on Mn d i s t r i b u t i o n and the inclusions on a P-T plot are strongly inclined to each other, thus reducing errors from this source. The accuracy of this determination would be limited not only by that of the experimental measurements involved but also by the v a l i d i t y of the necessary assumption that the sphalerite and galena and the f l u i d inclusions were a l l t r u l y cogenetic. Validation of this assumption i s not a t r i v i a l matter (Barton et a l . , 1963). Coveney and Kelly (1970) described a similar technique for use on inclusions trapped in quartz that permits some l i m i t s to be placed on the P-T conditions at trapping. Quartz has two modifications, with a rapid inversion, at 573°C (at 1 atm). Quartz that c r y s t a l l i z e d as the high-temperature (g) form can be 271

recognized as such (by crystal morphology, etching, e t c . ) . even though i t has inverted on cooling to the low-temperature (a) form. The inversion temperature i s raised by pressure, but only 28°C/kbar. Most isochores for f l u i d inclusions in quartz are much f l a t t e r on a P-T p l o t ; hence they intersect the a/g line at a large angle. I f a given inclusion were trapped by growth of an a-quartz c r y s t a l , the conditions of trapping would s t i l l have to l i e along the isochore defined by i t s homogenization but below the intersection of t h i s isochore with the a/3-quartz inversion l i n e . I f the host crystal can be shown to have grown as 3-quartz, the growth conditions must l i e along a different isochore at a higher T than that of the intersection. However, at the instant of inversion, the internal pressure in the inclusion w i l l decrease owing to the -1% volume increase on going from a to S. Because the thermal expansions of a- and e-quartz are quite different, the P-T path for an i n c l u s i o n trapped in quartz will consist of two lines having a discontinuity at the a/3 inversion, and hence i s not isochoric. Geobarometry based on simultaneous trapping of two immiscible f l u i d s . Where two e s s e n t i a l l y immiscible f l u i d s are present, each with known P-V-T properties, and separate inclusions of each f l u i d were trapped simultaneously, both P and T can be determined from the values for the Th of the two i n c l u s i o n s . In addition to the requirement that the P-V-T properties of both f l u i d s be known, the slopes of the appropriate isochores for the two f l u i d s on a P-T plot must be s i g n i f i cantly different, or the accuracy of the determination w i l l be poor. Most of the large number of reported geobarometric determinations (mainly in the Russian l i t e r a t u r e , see Roedder 1968a) are based on the use of the two immiscible f l u i d s , CO2 and H2O. Unfortunately, these two f l u i d s are not t r u l y immiscible (see below), and severe problems exist in e s t a b l i s h i n g the contemporaneity of t r a p ping. Furthermore, many of the published reports do not make clear the a l l important point, whether the CO2 and H2O phases studied were in separate i n c l u sions or were two immiscible phases in a s i n g l e i n c l u s i o n . The only pair of f l u i d s that give promise of providing good geobarometric data by t h i s method are the o i l and brine inclusions found in some M i s s i s s i p p i Valley-type ore deposits (Roedder 1963, p. 176-177), but the P-V-T data on the o i l phase are unknown and can only be guessed at the present. Oil that shows postentrapment degradation or maturation w i l l not provide good data. Evidence of such changes i s obvious in some o i l inclusions from these M i s s i s s i p p i Valleytype ore deposits (Roedder, 1972, Plate 9), but no such v i s i b l e evidence i s found in other o i l inclusions in the same deposits. When coeval inclusions of brine and o i l are found, Th i s always higher for the brine i n c l u s i o n s . Some hydrocarbon f l u i d s have P-V-T properties s i m i l a r to H2O (R.C. Burruss, pers. comm., as quoted by Narr and Currie (1982)), but many are very different (see Fig. 11-25), thus explaining the different Th values ( i . e . , the pressure corrections d i f f e r ) . Geobarometry based on simultaneous trapping of two partly immiscible f l u i d s . I f two partly immiscible f l u i d s , each saturated with respect to the other, are present at the time of trapping, inclusions of these two f l u i d s can provide some constraints on the pressure, so long as the appropriate P-V-T-X data are available and the mutual s o l u b i l i t i e s decrease from the conditions of trapping to those of observation. The method was o r i g i n a l l y proposed by Smith and L i t t l e (1959). Guilhaumou et a l . (1981) suggested use of the immiscible pair brine-(C02,N2) mixture. The nature of the f l u i d - i n c l u s i o n evidence on the phase condition (homogeneous vs heterogeneous) at the time of trapping i s crucial. I f , for example, a series of inclusions a l l have the same ratio of the two f l u i d s , presumably a homogeneous phase was trapped, and the P-T conditions at trapping are limited to the appropriate one-phase area of the pertinent diagram. I f different ratios are found in different coeval i n c l u s i o n s , the f l u i d s were presumably immiscible at the time of trapping, but a variety of problems make i t rather d i f f i c u l t to determine the pressure from such data. 272

Problems exist even when the inclusions are proved to be coeval. If necking down occurs in inclusions that contain more than one phase (from either the trapping of immiscible fluids or subsequent phase separation), the resulting inclusions can be very misleading. Similarly, the changes in mutual solubilities between Tt (and Pt) and those at the time of observation can be rather complex. Such problems may have invalidated many of the published geobarometric data sets where this method was used. Geothermometry based on trapping of boiling fluids. The boiling of a fluid is merely a special example of immiscibility under the prevailing conditions, in which the composition variable is eliminated. If a boiling liquid and its coexisting vapor phase are trapped separately in a pair of inclusions, these two inclusions will homogenize in the liquid and in the vapor phase, respectively. These two homogenizations must be at the same temperature, and if the boiling curve is known for that fluid, the pressure can be determined from this Th. Experimental difficulties are involved (e.g., inaccurate Th values may be recorded because the small amount of liquid phase that condenses in the vapor inclusion after trapping usually coats the walls as film and hence is difficult to see), but, more importantly, individual inclusions may have trapped a mixture of two phases, rather than a single homogeneous phase (Fig. 9-la). As the immiscibility "solvus" generally closes at higher temperatures, this trapping of a mixture will result in inclusions with higher Th than would be obtained on inclusions that trapped only liquid or only gas. Thus, where a group of such inclusions can be proven coeval, the minimum Th for inclusion homogenization in vapor and in liquid is provided by inclusions trapping pure end members and should be equal and represent Tt. All other inclusions, from the trapping of mixtures, would yield higher, and spurious, Th values. This method is, however, subject to many serious pitfalls. The identity of these two Th values is not only a necessary requirement, but also can be taken as relatively unambiguous proof of boiling. Because the pressure environment under which boiling can occur in nature may be rather variable and transient, and such pairs of inclusions are seldom exactly coeval, small differences may be expected. Of course, two separate fluids, one liquid and one vapor, could be trapped at different times and at a fortuitous combination of P and T values that would yield such similar Th values. Not uncommonly, however, individual inclusions grown from heterogeneous mixtures have trapped only the dispersed phase (which occurs as isolated droplets in the other, continuous phase). This is particularly expectable when most of the host crystal growth occurs from the continuous phase, e.g., from a water solution that contains dispersed steam bubbles. In theory, it makes no difference whether the dispersed phase is the same composition as the continuous phase (as in true boiling of a one-component system) or whether it consists merely of bubbles of a minor, more volatile constituent in a solution. The geobarometry method should work in either case, but solubility reversal and observational problems effectively preclude the latter. Evidence of boiling provides us with some of the most accurate and unambiguous geobarometry data available and has been reported in numerous ore deposits. However, the most important part of the evidence, the proof of contemporaneity of trapping of the two types of inclusions, is frequently poor or lacking. The mere existence of the two inclusion types is inadequate, as this can also come about by sequential trapping of different fluids at different times, by necking down, and by leakage (Roedder, 1979b). Careful microscopy is necessary to minimize or eliminate these sources of ambiguity. The distinction is far more than merely academic, because several varieties of rich ore deposits ("bonanzas") are widely believed to have formed as a result of the gross change in the chemistry of the ore fluids upon boiling or effervescence. 273

/ /

/

/

WATER 10 WT.%NaCI 25 WT.%NaCI CfHTICAL POINT 10)

200

400 TEMHRATUBE (°CI

500

GOO

700

Figure 9-6. A p o r t i o n of the phase diagrams f o r water and 10 and 25 wt % NaCl s o l u t i o n s , showing the l i q u i d - v a p o r curves and s e v e r a l i s o c h o r e s (g/cm^). Data f o r the 10 wt % i s o c h o r e o r i g i n a t i n g at Th 400°C are from Urusova (1975); those f o r the 25 wt i i s o c h o r e o r i g i n a t i n g at Th 450°C are e x t r a p o l a t e d from P o t t e r and Brown (1977). Other data from S o u r i r a j a n and Kennedy ( 1 9 6 2 ) , Burnham et a l . ( 1 9 6 9 ) , Keenan et a l . ( 1 9 6 9 ) , and Haas (1 976). From Roedder and Bodnar ( 1 9 8 0 ) .

Not uncommonly, one may find inclusion evidence of b o i l i n g at a given pressure, along with other i n c l u s i o n s , in the same deposit, indicating possibly higher pressures. In t h i s s i t u a t i o n , the low-pressure l i m i t established by the boiling inclusions i s most informative, because i t places an upper (hydrostatic) depth l i m i t on the crystal in a vein open to the surface ( i f we ignore the interval of unsaturated ground above the water table). The higher pressures are not as meaningful because of mechanisms for generating l i t h o s t a t i c or greater pressures at shallow depths ( F i g . 9 - 5 ) . The only way in which such an interpretation of the evidence of b o i l i n g could y i e l d misleading depths i s shown in the r e l a t i v e l y rare example F in Figure 9-5. Geobarometry based on inclusions containing daughter minerals. Many f l u i d inclusions contain one or more s o l i d phases in addition to l i q u i d and vapor; i f the P-V-T-X properties of that particular s a l t - ^ O system are known, the phasedisappearance temperatures may be used to determine a Pt for the f l u i d i n c l u s i o n . A f l u i d inclusion containing s o l i d s a l t , saturated l i q u i d , and saturated vapor at ambient temperature might follow any one of three possible paths to homogenization, indicating three different trapping pressures. I f the vaporbubble-disappearance temperature (Th L-V) i s higher than the temperature of solution of the s a l t (Tm NaCl), the inclusion trapped an unsaturated s o l u t i o n . The minimum Tt and Pt are Th L-V and Ph for a solution of the s a l i n i t y determined from Tm NaCl. I f Tm NaCl and Th L-V are the same, the temperature and pressure on the s o l i d - l i q u i d - v a p o r curve at that point are also minimum values for Tt and Pt. Furthermore, where one of these inclusion types coexists with i n c l u s i o n s that homogenize at the same temperature but in the vapor phase, then both were probably trapped from a b o i l i n g solution, and Th = Tt and Ph = Pt. F i n a l l y , i f Th L-V occurs below Tm NaCl, the l a t t e r provides a minimum value for Tt. The minimum pressure i s the pressure along the s o l i d - l i q u i d - v a p o r curve at Tm NaCl. Examples of pressure determinations on t h i s basis are given in the next section. 274

Application to Inclusions of Aqueous Salt Solutions Fluid-inclusion-pressure determinations are based on the volumetric properties of the inclusion fluid. Therefore, the composition of the fluid must be known and experimental P-V-T-X data for that particular fluid must be available before the results of heating/freezing runs can be used to calculate Pt. This section describes the procedures that can be used for given compositions and some of the errors that have been made in geobarometric studies in the past. Pure-water inclusions. Inclusions of essentially pure water are relatively rare. They are most commonly found in samples from low-pressure, near-surface hot-spring environments (Roedder 1977a) but are also found in samples from geothermal systems formed at considerably higher pressures. In the latter, they may permit some reconstruction of the former P-T regime of the geothermal system (Browne et al., 1976). If the salinity is low (less than a few percent), Figure 9-1 provides the basis for interpreting the data. (To a first approximation, each 0.1°C depression of the freezing point (i.e., Tm ice) corresponds to 5850 ppm of salts.) CO2 in geothermal fluids can significantly influence both the apparent salinity and the pressure, as discussed below. Low- to moderate-salinity inclusions.j/ To extend the methods of pressure determination described above to solutions of low or moderate salinity, we need to determine only the composition of the inclusion fluid from Tm ice and then use the volumetric properties of that fluid to determine the correct isochore. Combining this with an independent mineralogical geothermometer will provide Pt. The isochores for NaCl-H20 solutions differ from those for pure water (Fig. 9-6). Note that, contrary to some published statements of mine, the isochores for pure water at low temperatures are steeper than those for NaCl solutions, but this relationship is reversed at higher temperatures. High-salinity multiphase inclusions. If a homogeneous fluid of sufficiently high NaCl concentration is trapped as an inclusion, the saturation limit may be exceeded as the fluid cools; the cooling causes precipitation of a daughter crystal of halite. When this multiphase inclusion is reheated, it once again becomes homogeneous. Ph depends on the temperature, salinity, and order of disappearance of the phases, as indicated by the three inclusions shown in Figure 9-7, all of which are assumed to be in the system NaCl-hfeO and to homogenize completely at 400°C. Inclusion A follows the solid-liquid-vapor curve until the halite dissolves at 158°C (Tm NaCl), corresponding to a 30 wt % NaCl solution (Benrath et al., 1937). With continued heating the inclusion follows the liquid-vapor curve for a 30 wt % NaCl solution (Haas, 1976; Urusova, 1975) until the vapor bubble disappears at 400°C (Th L-V). At this Th, Ph, and thus the minimum Pt, is 222 bars (Urusova, 1975). This inclusion may have been trapped at any higher P and T along the isochore originating at A (Fig. 9-7). Inclusion B follows the solid-liquid-vapor curve until, at 400°C, both the halite and the vapor bubble disappear simultaneously (Tm NaCl = Th L-V). At this temperature, the inclusion contains a 46 wt % NaCl solution (Benrath et al., 1937) under a pressure of 182 bars (Keevil, 1942; Sourirajan and Kennedy, 1962), corresponding to the minimum Pt. The inclusion could have been trapped at any higher P and T along the isochore originating at B (Fig. 9-7). Inclusion C follows the solid-liquid-vapor curve until the vapor phase disappears at 310°C (Th L-V) at only 66.4 bars (Haas, 1976). From this Th L-V

U For simplicity, low- to moderate-salinity inclusions are arbitrarily d e f i n e d here as those that are u n s a t u r a t e d at room t e m p e r a t u r e , i.e., less than - 2 6 wt % NaCl e q u i v a l e n t .

275

Figure 9-7. P r e s s u r e s at homogenization f o r three h a l i t e - b e a r i n g f l u i d i n c l u s i o n s (A, B, C ) , from Roedder and Bodnar (1980). A l l are assumed to be i n the system NaCl-H20 and to homogenize at the same temperature (400°C) but to e x h i b i t d i f f e r i n g modes of homogenization, r e s u l t i n g in d i f f e r e n t p r e s s u r e s at Th. Data are from S o u r i r a j a n and Kennedy (1962), Urusova ( 1 9 7 5 ) , Haas ( 1 9 7 6 ) , and P o t t e r and Brown ( 1 9 7 7 ) . The i n s e t shows the p r e s s u r e s obtained f o r i n c l u s i o n C u s i n g the v a r i o u s methods d e s c r i b e d i n the t e x t .

to Tm NaCl at 400°C, the inclusion follows a s o l i d - l i q u i d curve ( l i q u i d u s ) , shown schematically in Figure 9-7. The pressure at t h i s point, which i s the minimum Pt, i s calculated to be ~650 bars, as shown below. Other ions in s o l u t i o n , such as could reduce t h i s minimum even farther (Stewart and Potter, 1979). The inclusion could have been trapped at any higher P and T along the isochore originating at C (Fig. 9-7). Although f l u i d inclusions homogenizing by h a l i t e disappearance are common, much confusion concerns their o r i g i n and interpretation. Assuming that t h i s behavior i s not a result of k i n e t i c effects (Chivas and Wilkins, 1977; Eastoe, 1978) or that the inclusions have not trapped the halite as s o l i d phase, as suggested by some data (e.g., Bodnar, 1978; Nagano et a l . , 1977; Wilson, 1978), the most common explanation i s that the i n c l u s i o n s were trapped at "high pressure" (e.g., Piznyur, 1968; Dolgov et a l . , 1976; Kamilli, 1978; Milovskiy et a l . , 1978; and Andreyev and Shvadus, 1977), on the basis of what i s generally called the "Lemmlein and Klevtsov method." As detailed below, t h i s method i s found to y i e l d spurious r e s u l t s . The "Lemmlein and Klevtsov method" was apparently f i r s t used by Klevtsov and Lemmlein (1959a) to determine minimum Pt for f l u i d i n c l u s i o n s homogenizing by h a l i t e disappearance. The pressure was obtained by conducting experimental P-V-T studies on aqueous solutions thought to correspond to the composition of the inclusion f l u i d s and then plotting a pressure-correction diagram from the 276

data. However, Roedder and Bodnar (1980) found several errors in the experimental and theoretical data of Klevtsov and Lemmlein (1959a) that i n v a l i d a t e their r e s u l t s . The net result was that Klevtsov and Lemmlein did not have the phase assemblage ( i . e . , with s o l i d NaCl) in t h e i r pressure vessel that they thought they had, and the isochore they measured was for a single-phase ( f l u i d ) system without the volume change from a d i s s o l v i n g crystal of h a l i t e ; the r e s u l t ing isochore was therefore much steeper than i t should have been. The actual path along the s o l i d - l i q u i d surface cannot be determined because of the lack of volumetric data in t h i s region. However, i f the apparent molar volumes of NaCl are negative in t h i s region, as Urusova's (1975) data indicate, the actual pressure change between Th L-V and Tm NaCl might be very small, though the temperature difference may be very large. Lyakhov (1973) presented a different method of calculating pressures from inclusions homogenizing by h a l i t e disappearance, which i s also based on the P-V-T-X properties of NaCl-HgO s o l u t i o n s . Unfortunately, t h i s method assumes a constant volume but allows the mass of water in the inclusion to vary as the density of the water changes. This results in the erroneous conclusion that above 320°C, the s o l u b i l i t y of NaCl decreases rapidly. In Lyakhov's method, the effect of t h i s erroneous conclusion i s that an inclusion following a path to Th described by inclusion C (Fig. 9-7) would have been trapped at a f i c t i t i o u s minimum pressure of 5000 bars. Roedder and Bodnar (1980) concluded that the "Lemmlein and Klevtsov" and "Lyakhov" methods of determining pressures are both i n v a l i d , not only for the theoretical reasons given above, but also because the pressures obtained from these methods are not consistent with other data. Thus, using the "Lemmlein and Klevtsov method," Kami 11i (1978), indicates that the inclusion data "require pressures much greater than any reasonable l i t h o s t a t i c load," and Bodnar and Beane (1980) reported pressures an order of magnitude higher than those obtained from other f l u i d - i n c l u s i o n data and from geologic reconstruction. A l s o , pressures of 5000 bars (Lyakhov 1973) and 6000 bars (Dolgov et a l . , 1976) in quartz and 2700 bars in f l u o r i t e (Erwood et a l . , 1979) are in gross disagreement with measured pressures required to decrepitate inclusions in quartz (850 bars; Naumov et a l . , 1966) and to stretch f l u i d inclusions in f l u o r i t e (100-700 bars; Bodnar and Bethke, 1984). Roedder and Bodnar (1980) showed that a good approximation of Ph in such multiphase inclusions at Th can be calculated . They i l l u s t r a t e d t h i s method by calculating the pressure of the same inclusion C (Fig. 9-7) in which Th L-V i s 310°C and Tm NaCl i s 400°C. To simplify calculations, they assumed that the inclusion contains 1000 g of H2O and that the f l u i d properties are adequately represented by those of the NaCl-HgO system. Assuming that NaCl s o l u b i l i t y i s independent of pressure (Adams, 1931), they obtained the bulk composition (46 wt % NaCl) from Tm NaCl (400°C) and the s o l u b i l i t y data of Benrath et a l . (1937). Hence they needed to calculate only the density of the homogeneous f l u i d and to refer t h i s value to the P-V-T-X measurements of Urusova (1975) to obtain the pressure at f i n a l homogenization (Tm NaCl). At 310°C, the inclusion volume i s simply the volume of saturated aqueous solution plus the volume of s o l i d h a l i t e remaining. The solution volume i s the total mass of solution divided by i t s density. A saturated solution on the liquid-vapor curve at 310°C i s 10.8 molal NaCl and has a density of 1.07 g/cm3 (Haas 1976), which, when combined with the previous assumption of 1000 g of H2O and a molecular weight of NaCl of 58.44 g/mole, gives a solution volume of 1524 cm3. The volume of h a l i t e remaining can be calculated from the difference in s o l u b i l i t y between 310° and 400°C (Benrath et a l . , 1937) and the molar volume at 310°C. When a value for the s o l u b i l i t y of NaCl at 400°C of 46 wt % or 14.6 molal i s used, calculated from Benrath et a l . (1937), the difference i n s o l u b i l i t y i s 3.8 moles/1000 g H 2 0. The molar volume of h a l i t e at 25°C i s 27.018 cm3 277

(Robie et a l . , 1966), which i n c r e a s e s 3.92% upon h e a t i n g from 25°C (Skinner 1966) to give a molar volume at 310°C of 28.08 cm 3 . Therefore, the volume of h a l i t e at 310°C i s 106.7 cm 3 , and the t o t a l i n c l u s i o n volume at 310°C i s 1631 cm 3 . The i n c l u s i o n volume at homogenization i s the volume at 310°C plus the i n c r e a s e due to thermal expansion upon h e a t i n g and d i s s o l u t i o n of the i n c l u s i o n w a l l s . Where the host mineral i s q u a r t z , the i n c l u s i o n volume w i l l i n c r e a s e 0.54% or 8.8 cm3 from thermal expansion when heated from 310 to 400°C ( S k i n n e r , 1966). The i n c r e a s e i n volume due t o d i s s o l u t i o n of quartz from the w a l l s when the i n c l u s i o n i s heated from 300 to 400°C was found to be i n s i g n i f i c a n t and was ignored i n the c a l c u l a t i o n s . The two d i l a t i o n e f f e c t s (see Chapter 3) were i n c o r r e c t l y assumed to cancel each other, but the e r r o r introduced by t h i s misassumption i s s m a l l . From the above, the d e n s i t y of the homogeneous f l u i d at 400°C c a l c u l a t e d from the mass of s o l u t i o n (1853 g) and the i n c l u s i o n volume (1639.8 cm 3 ) i s 1.130 g/cm 3 . E x t r a p o l a t i n g U r u s o v a ' s (1975) data at 400°C along i s o c h o r e s from 45 wt % NaCl (her maximimum concentration at 400°C) to 46 wt % NaCl provides a p r e s s u r e at f i n a l homogenization of 640 bars. This p r e s s u r e i s compared with pressures at homogenization obtained by the "Lemmlein and K l e v t s o v " method (~1025 bars) and the "Lyakhov" method (5000 b a r s ) f o r t h i s same i n c l u s i o n ( F i g . 9 - 7 ) . These p r e s s u r e s are a l l minima; Pt could be at any h i g h e r value along the appropriate i s o c h o r e . The accuracy of p r e s s u r e determination by t h i s method i s d i f f i c u l t to e s t a b l i s h and, of course, depends on the accuracy of the P-V-T-X data used. In t h i s respect, Roedder and Bodnar (1980) noted that where the data of Urusova (1975) and S o u r i r a j a n and Kennedy (1962) o v e r l a p , U r u s o v a ' s data are always h i g h e r by ~25 b a r s , and S o u r i r a j a n and Kennedy's data are thought to be high because of the presence of H2 gas derived from c o r r o s i o n ( L i u and L i n d s a y , 1972; J . Haas, J r . , p e r s . comm., 1979). A complete a n a l y s i s of the accuracy and p r e c i s i o n of the method must await experimental volumetric data along the l i q u i d - s o l i d curve of the NaCl-H20 system at high P and T. Although h a l i t e i s by f a r the most common s o l i d phase i n f l u i d i n c l u s i o n s , s y l v i t e i s often found along with h a l i t e , e s p e c i a l l y i n porphyry-type ore d e p o s i t s . Combining data on the N a C l - K C l - ^ O ternary diagram with the data of Ravich and Borovaya (1949) (see F i g . 8 - 2 5 ) , these i n c l u s i o n s bearing s y l v i t e and h a l i t e may be used t o estimate f l u i d compositions and minimum pressures attending entrapment. Other ions i n s o l u t i o n , such as would s i g n i f i c a n t l y lower the minimum. The vapor p r e s s u r e s of such " h y d r o s a l i n e melts" are s u r p r i s i n g l y low. Thus, Stewart and Potter (1979) showed that a f l u i d saturated with respect to NaCl, KC1, and CaCl2 at 350°C has a maximum vapor pressure of only - 2 5 bars. Most other s o l i d phases are more r a r e l y found, but t h e i r s o l u b i l i t i e s and volumetric p r o p e r t i e s so poorly known that i n c l u s i o n s c o n t a i n i n g them provide no useful data on P t . A p p l i c a t i o n t o Carbon D i o x i d e - B e a r i n g

Inclusions

Carbon d i o x i d e - b e a r i n g f l u i d i n c l u s i o n s occur in rocks from a wide range of g e o l o g i c environments, and many methods have been suggested f o r determining Tt and Pt from these i n c l u s i o n s . Considerable confusion e x i s t s , however, mainly from f a i l u r e to c o n s i d e r the t r a p p i n g c o n d i t i o n s . The three most commonly used methods are described below. The f i r s t r e q u i r e s separate i n c l u s i o n s of CO2 and H2O; the other two are based on mixed H2O-CO2 i n c l u s i o n s . I n t e r s e c t i n g i s o c h o r e s i n pure CO2 and H2O systems. 278

Kalyuzhnyi and Koltun

F i g u r e 9-8. Combined P - T diagrams for CO2 and H^O, illustrating the Kalyuzhnyi and K o l t u n (1953) method of geothermobarometry using separate CO2 and H^O inclusions trapped at the same T and P, from R o e d d e r and Bodnar (1980). Point A represents the comnon trapping t e m p e r a t u r e (237°C) and pressure (1180 bars) for a CO2 inclusion and an H2O inclusion h o m o g e n i z i n g at 11°C and 167°C, Data are from Kennedy (1954). B u r n h a m et al. (1969), respectively, both in the liquid phase. Keenan et al. (1969), and Weast (1980).

(1953) applied a method of geobarometry, first described by Nacken (1921), which is applicable to separate CO2 inclusions and H2O inclusions trapped at the same P-T conditions, either at the same location at different times or at the same time but at different locations. By plotting the P-V-T diagrams for H2O and C02 in the same plane, Pt and Tt are defined by the intersection of the isochore corresponding to the CO2 inclusion with that corresponding to the H2O inclusion (Fig. 9-8). Thus, a CO2 inclusion homogenizing in the liquid phase at 11°C and an H2O inclusion homogenizing in the liquid phase at 167°C would both have been trapped at 237°C and 1180 bars (A, Fig. 9-8). This method of obtaining trapping conditions is valid if, and only if, the two inclusions were separately trapped as essentially pure components at the same temperature and pressure. Pressure determinations from C02-bearing fluid inclusions are much more complex if the carbon dioxide and water were not physically separated at the time of trapping. The addition of CO2 to water causes the critical locus to migrate from that of pure H2O to lower temperatures and higher pressures, reaching a minimum temperature of 266°C at 2450 bars and 41.5 mole % CO2 (Todheide and Franck, 1963). Upon further addition of CO2, the critical temperature slowly rises, reaching 268°C at 43.5 mole % CO2 and ~3600 bars; this rise results in a large two-phase field spanning a wide range of P-T-X conditions (Figs. 8-19, 20; 9-9). Naumov and Malinin graphical method. Naumov and Malinin (1968) proposed a graphical technique for determining pressure that uses the Th and Td of inclusions containing both liquid-water solution and CO2. The method is based on a straightline extrapolation, in P-T space, from the point representing partial homogenization (liquid and gaseous CO2, i.e., Th CO2 L-V), through Td (assumed to be at 850 atm for inclusions in quartz), to Th, and thus Ph. The validity 279

of this technique is in question because there is no basis for assuming that the pressure increases linearly with temperature between Th CO2 L-V and total Th. In fact, because the slopes of the "isochores" and mutual solubilities (see below) of the two phases are constantly changing as the system within the inclusion travels through P-T space, it would be quite surprising if the pressure did vary in such a simple fashion with temperature. Gehrig (1980) has shown that isochores in the two-phase field (C02-rich vapor and (H2O + NaCl)-rich liquid) for the system H20-C02-NaCl are strongly curved. Further, the assumption that decrepitation begins at 850 atm is based on experimental studies (Naumov et al., 1966) using synthetic quartz crystals that normally contain very large fluid inclusions relative to natural quartz and that do not contain lower temperature secondary inclusions. As Naumov and Malinin (1968) themselves pointed out, pressures necessary to decrepitate inclusions in synthetic quartz range from 850 to >3000 atm, depending on inclusion size, and Swanenberg (1980) reported pressures of 6000 bars without decrepitation. Therefore, if the smallest inclusions in synthetic quartz (decrepitating at >3000 atm) correspond to the largest inclusions studied in natural quartz, as is possible, the onset of mass decrepitation of natural quartz samples will be >3000 atm. As a result of the extrapolation procedure, the difference in the assumed pressure at Td is magnified in the estimate of Ph. Finally, and most important, at many of the P-T-X combinations obtained by this technique, that were reported in the literature, water and CO2 form two immiscible phases (Chapter 8), which raises the possibility that the inclusions did not trap homogeneous fluids. Mixed H2O-CO2 inclusions. At ambient temperatures, inclusions consisting purely of CO2 and H2O commonly contain three phases -- liquid and gaseous CO2, and liquid H2O. These phases are essentially pure (relative to each other), so by measuring the volumes of the three phases at a known temperature and using the density data for C O 2 and H2O, the mole % C O 2 in the inclusion may be calculated. Such a procedure was used to obtain the mole % CO2 (and mole % H2O) in three hypothetical inclusions, shown in the inset of Figure 9-9. In practice, volume measurements are generally very imprecise, for reasons previously mentioned, and caution is needed when compositions are calculated by means of this technique. The complexities of pressure determinations from inclusions containing both CO2 and H2O might best be illustrated by examining the phase changes upon heating of three hypothetical pure H2O-CO2 fluid inclusions and referring these observational data to the CO2 solubility diagram shown in Figure 9-9. First, consider a fluid inclusion assumed to have trapped a homogeneous fluid containing 25 mole % CO2 and having the volume percentages of phases at 25°C shown by inclusion B (Fig. 9-9). On heating to 27°C, the CO2 liquid and vapor phases homogenize in the liquid phase, which then has a density of 0.672 g/cnr (Quinn and Jones, 1936; Kennedy, 1954; Newitt et al., 1956). Continued heating causes the mutual solubilities of CO2 and H2O to increase until, at 275°C, the inclusion contains a homogeneous fluid phase composed of 25 mole % CO2 and 75 mole % H2O (B in Fig. 9-9). These data require that the internal pressure,!^/ and thus the minimum trapping pressure, Pt, is ~1000 bars. Note that homogenization at 275°C and 1000 bars would also occur in an inclusion containing 45 mole % CO2. If the pressure on this fluid dropped below 1000 bars, CO2 and H2O would no longer be completely miscible. Thus, at 275°C and 575 bars, a C02-rich fluid containing 40 mole % H2O and an H20-rich fluid containing 11 mole % CO2 coexist (Fig. 9-9), and two separate inclusions trapped at these conditions would have phase relations as shown by inclusions A and C in the inset on Figure 9-9. Upon heating from 25°C, the CO2 phases in inclusions A and C would homogenize at

— ' The high internal pressure developed during homogenization of many CO2-H2O inclusions is the reason that decrepitation is common before Th is reached.

280

VOLUME PERCENT C0 2 (AT 25°C|

F i g u r e 9-9. P-X p l o t o f isotherms showing c o m p o s i t i o n s o f c o e x i s t i n g phases i n the system H2OCO2, u s i n g data of Todheide and Franck (1963) and Greenwood and Barnes (1966). The upper a b s c i s s a shows volume percent CO? at 25°C along the CO2 l i q u i d - v a p o r curve (64 b a r s ) , assuming d e n s i t i e s of C0 2 l i q u i d , C0 2 vapor, and HoO l i q u i d of 0.71, 0.24, and 1.0 g/cm 3 , r e s p e c t i v e l y (Newitt et a l . , 1956; Keenan et a l . , 1969). The i n s e t shows the two-dimensional appearance at the stated conditions for three c y l i n d r i c a l i n c l u s i o n s having compositions as given ( l i q u i d CO2 shaded), which are a l s o shown on the diagram. The 250°C isotherm for a 6 wt % NaCl s o l u t i o n from Takenouchi and Kennedy (1965b) i s shown for comparison. From Roedder and Bodnar (1980).

26 and 28°C, respectively, and complete homogenization would occur in both inclusions at 275°C and 575 bars. In most f l u i d - i n c l u s i o n work, the composition i s unknown or i s imprecisely known from volume measurements. In t h i s s i t u a t i o n , the c r i t i c a l pressure along the 275°C isotherm determines a maximum Pt of 1080 bars. Ypma (1963) pointed out, however, that noncoeval CO2-H2O i n c l u s i o n s , trapped under different P-T conditions, that can homogenize ( f o r t u i t o u s l y ) at the same temperature. Most CO?-bearing f l u i d inclusions contain an aqueous s a l t solution rather than pure H2O, adding further complexity to pressure determination. When NaCl i s added to the CO2-H2O system, the c r i t i c a l s o l u b i l i t y of CO? at a given temperature and pressure decreases (see Chapter 8 ) , and the m i s c i b i l i t y gap i s widened. 281

The 250°C isotherm for CO2 s o l u b i l i t y in a 6 wt I NaCI s o l u t i o n , from Takenouchi and Kennedy (1965b), i s shown in Figure 9-9 for comparison with pure H2O. Note that at 250°C, a 10 mole % CO2 f l u i d at 750 bars would be j u s t within the onephase region i f we assume a pure H2O-CO2 system, but well into the two-phase region i f the H2O phase actually i s a 6 wt % NaCI s o l u t i o n . I f a 10 mole % CO2 i n c l u s i o n that had a Th of 250°C were trapped as a homogeneous f l u i d , Figure 9-9 shows that i t would have had a minimum Pt of 750 bars, i f we assume that the i n c l u s i o n c o n s i s t s of pure CO2 and H2O. However, i f the H2O phase contained 6 wt % NaCI, the minimum trapping pressure would have been ~2500 bars, determined by extrapolating the 250°C, 6 wt % NaCI isotherm of Takenouchi and Kennedy (1965b) to higher pressure. Methane i s completely m i s c i b l e with both l i q u i d and gaseous CO2 and i s a common component of C02-bearing f l u i d i n c l u s i o n s in metamorphic rocks. Its presence i s usually implied when the t r i p l e point for the CO2 phases i s found to be at a lower temperature than that for pure CO2. H o l l i s t e r and Burruss (1976) suggested that the addition of CH4 to the H2O-CO2 system r a i s e s the top of the solvus to higher temperatures, and the m i s c i b i l i t y gap i s therefore widened. Swanenberg (1979) indicated that data from CO2-CH4 i n c l u s i o n s may be used in pressure determinations i f the f l u i d density i s expressed as an "equivalent CO2 d e n s i t y . " This method i s based on the observation that, over the temperature range 200-800°C, the slopes of the isochores for the CO2-CH4 system are s i m i l a r to those of the pure CO2 system. Application to I n c l u s i o n s in Igneous Rocks S i l i c a t e - m e l t i n c l u s i o n s . Although many determinations of Th have been made on s i l i c a t e - m e l t i n c l u s i o n s (generally they f a l l in the range 800-1200°C), Ph determinations based on such i n c l u s i o n s normally are not made. In large part, t h i s stems from the fact that s i l i c a t e melts are r e l a t i v e l y incompressible, so there i s l i t t l e need for a pressure correction to be applied to most such Th determinations. Murase and McBirney (1973) determined the adiabatic compressib i l i t y of a s e r i e s of s i l i c a t e melts from t h e i r d e n s i t i e s and longitudinal-wave v e l o c i t i e s and found them to f a l l i n the range of 2 to 7 x 10 1 2 cm2/dyne at 1000-1200°C. Combining these data with thermal-expansion data on c r y s t a l s and glasses (Skinner, 1966), one f i n d s that the e f f e c t of 1 kbar at the time of trapping would correspond to ~20°C, an amount far smaller than the probable experimental error alone on most determinations of s i l i c a t e melt Th (Roedder, 1979a). Most determinations of melt Th are made on g l a s s y i n c l u s i o n s , from rocks formed under intermediate or shallow depths, so the pressures are seldom >1 kbar. Silicate-melt/C02 i n c l u s i o n p a i r s . I f the s i l i c a t e - m e l t i n c l u s i o n s give evidence of having been trapped from an immiscible mixture of s i l i c a t e melt with another, more compressible f l u i d , pressure estimates become f e a s i b l e . Here the s i l i c a t e - m e l t i n c l u s i o n s , containing a r e l a t i v e l y incompressible f l u i d , permit a f a i r l y close estimate of the true Tt ( i . e . , the necessary "independent geothermometer"). This value can then be used, along with P-V-T data on the other, more compressible f l u i d ( i n a separate i n c l u s i o n ) , to obtain Ph. The major l i m i t a t i o n in the practical a p p l i c a t i o n of the procedure l i e s in the d i f f i c u l t y of proving contemporaneity of the two i n c l u s i o n types. Many of the published pressure determinations from i n c l u s i o n s in igneous minerals are based on such an assumption of i m m i s c i b i l i t y but lack the necessary evidence to prove contemporaneity of trapping. The abundant CO2 i n c l u s i o n s found along with s i l i c a t e - m e l t i n c l u s i o n s in the o l i v i n e of o l i v i n e nodules from b a s a l t occurrences a l l over the world (Roedder, 1965a) provide an example of the a p p l i c a t i o n of the method. The s i l i c a t e - m e l t i n c l u s i o n s in these samples, and actual observations on l a v a s , show that these i n c l u s i o n s were probably trapped at ~1200°C. I f we assume t h i s 282

Tt, and obtain the density of the CO? (0.85 g/cm 3 ) from Th C0 2 (10°C; see Fig. 8-9), the P-V-T data on COg (Fig. 8-TO) permit an estimate of 6 kbar for Pt, corresponding to the hydrostatic pressure from -20 km of liquid basalt. This is not a minimum or maximum value, but the actual pressure, under the given assumptions. Unfortunately, this pressure estimate must be based on extrapolations from experimental data (Fig. 8-10) in which maximum temperatures and pressures are 1000°C and 1400 bars (Kennedy, 1954), or 707°C and 8000 bars (Shmonov and Shmulovich, 1974). Other inclusions in these samples were-presumably trapped at similar temperatures but greater depths; these inclusions had such high internal pressures that they have generally decrepitated upon eruption at the surface at ~1200°C. Therefore, the range of application of the method is limited (Roedder, 1965a), but most important, the decrepitation of those inclusions formed at higher pressures will certainly result in a biased sample. Bilal and Touret (1977) reported Th in the liquid phase as low as +10°C on presumably pure CO2 inclusions in phenocrysts from a basalt, and estimated Pt = 5 kbar. However, using the molal volume data on CO2 of Shmonov and Shmulovich (1974), this density CO2 at 5 kbar would indicate a trapping temperature of only 900°C. Extrapolation to an assumed Tt of 1200°C indicates a trapping pressure >>6 kbar. Lower pressure CO2 inclusions in basaltic glass from submarine flows were studied by Moore et al. (1977), who showed a constant relationship between the pressure of CO2 in gas vesicles (i.e., bubbles) and the known depth of water in which the eruption occurred. Using this method, one can estimate the pressure at the time of eruption of unknown basalt samples simply by piercing the bubbles under a liquid in which CO2 is not soluble, and measuring the volume expansion (Chapter 16). The above examples are based on the relatively immiscible pair of fluids, CO2 and basalt. Inclusions from somewhat less immiscible fluids, such as silicate melt and water, can also be used. If the second fluid phase is a hydrosaline melt, containing >50% of NaCl and other salts, as in the Ascension Island granites (Roedder and Coombs, 1967) or in the various hypabyssal granites reported in the extensive Soviet literature (see many entries in Roedder, 1968a, particularly those by A.I. Zakharchenko), inclusion data could, in theory, provide an estimate of pressure. Such an estimate requires, however, that the composition of the aqueous fluid phase be known and that P-V-T data on such fluids be available; neither of these requirements can be met satisfactorily at present. Where the concentration of salts in the aqueous phase is low, the P-V-T data for H2O can be used, and the accuracy of the resulting geobarometric values will increase. The simplest case involves silicate-melt inclusions containing H2O but having no evidence of a second, volatile-rich phase. These assumptions require that the pressure at the time of trapping was above that of the vapor pressure of water over that melt. Thus Anderson (1974a) and Anderson and Sans (1975, and pers. comm.) devised several methods for estimating the H2O content of magmas from melt inclusions and found some high values, as much as 12 ± 2% H2O (Roedder, 1979a). Such inclusion data can be combined with experimental data on the vapor pressures of hydrous melts, when they become available, to provide at least a lower limit for Pt. Application to Siting of Nuclear Reactors One interesting and important new facet of fluid-inclusion barometry, one that may well have more extensive application in the future, is its use in evaluating a site for a nuclear reactor. Cunningham (1974) provided such an evaluation, based on a study of fluid inclusions in euhedral crystals protruding into vuggy cavities along a fault that cut a proposed site. The important 283

question was to date the l a s t movement on the f a u l t and hence t o evaluate the p o s s i b i l i t y of renewed movement. The c r y s t a l s must have grown s i n c e the l a s t movement on the f a u l t . An elevated pressure at t r a p p i n g , shown by thermometric data on the i n c l u s i o n s i n the c r y s t a l s , e s t a b l i s h e d a minimum depth below the surface at the time of t h e i r formation. Subsequent e r o s i o n had removed t h i s overburden, so when an estimate was made of the rates of denudation, a minimum time s i n c e the l a s t movement was obtained. Numerous other s i m i l a r a p p l i c a t i o n s have been published subsequently (Chapter 13). Summary of the Present S t a t u s of Geobarometry Most f l u i d - i n c l u s i o n s t u d i e s today i n v o l v e the determination of Th and a search f o r g e o l o g i c evidence concerning depth of b u r i a l i n order to estimate the pressure c o r r e c t i o n t o be added to Th to o b t a i n Tt. Assuming that the composition of the i n c l u s i o n can be determined, the four i n t e r r e l a t e d v a r i a b l e s here are Th, Ph, T t , and Pt; i f any three are known, the f o u r t h can be obtained. Th can be determined with r e l a t i v e l y high accuracy at p r e s e n t , and, assuming experimental data are a v a i l a b l e , t h i s determination a l s o f i x e s Ph. Estimates of Pt, however, have very l a r g e e r r o r bars r e s u l t i n g from the n e c e s s a r i l y large u n c e r t a i n t i e s of the g e o l o g i c r e c o n s t r u c t i o n and the problem of h y d r o s t a t i c vs l i t h o s t a t i c pressure mentioned i n a previous s e c t i o n . Independent geothermometers f o r o b t a i n i n g Tt are not always a v a i l a b l e and are seldom very accurate. However, I believe that a good p o s s i b i l i t y e x i s t s of u s i n g them in the future as these other geothermometers are r e f i n e d , and hence of determining Pt from (Tt-Th) and the P-V-T-X p r o p e r t i e s of the f l u i d with an accuracy at l e a s t comparable with that from g e o l o g i c r e c o n s t r u c t i o n . As i n a l l s c i e n c e , we can expect that more accurate thermometric determinat i o n s on i n c l u s i o n s w i l l c e r t a i n l y become a v a i l a b l e i n the f u t u r e . Will they help i n geobarometry? In t h e o r y , i t i s p o s s i b l e t o obtain both P and T from careful s t u d i e s of the small d i f f e r e n c e s that should e x i s t between otherwise i d e n t i c a l coeval i n c l u s i o n s i n two host minerals with d i f f e r i n g thermal expans i o n s . Such an e f f o r t seems doomed, however, by the requirement of coeval i n c l u s i o n s ; whenever d e t a i l e d s t u d i e s are made, i t i s apparent that the f l u i d s present i n most g e o l o g i c environments have varied i n P, T, and X even during the format i o n of small p a r t s of a s i n g l e c r y s t a l ( e . g . , Roedder, 1977c). Such data emphasize the d i f f i c u l t y of proving two i n c l u s i o n s to be t r u l y coeval, and show that even small e r r o r s i n such an assignment of o r i g i n can r e s u l t i n major e r r o r s in pressure by any method r e q u i r i n g coeval i n c l u s i o n s . Although immiscible f l u i d p a i r s now provide us with some of our best geobarometers, t h i s same problem of f i n d i n g t r u l y coeval i n c l u s i o n s i n material formed in a changing environment makes future refinements i n i n c l u s i o n thermometry on such i n c l u s i o n s of r e l a t i v e l y l i t t l e value f o r geobarometric determinations. More accurate determinations of the composition of i n c l u s i o n s are a l s o forthcoming, as a r e s u l t of the a p p l i c a t i o n of a v a r i e t y of new techniques. Although such improved data on the composition of the former f l u i d phase are i n v a l u a b l e in other i n v e s t i g a t i o n s , such as the causes of ore d e p o s i t i o n or the nature of the v a r i o u s r e a c t i o n s that have y i e l d e d metamorphic rock assemblages, they are of rather l i m i t e d immediate value t o geobarometry, because experimental P-V-T-X data are g e n e r a l l y l a c k i n g f o r such c o m p o s i t i o n s . Even the simple systems, such as NaCl-H20, are not known to the P, T, and X l i m i t s necessary, and as the complexity of the composition i n c r e a s e s , the a v a i l a b l e data decrease very r a p i d l y . After such P-V-T-X data become a v a i l a b l e , however, we may expect that the accuracy of geobarometry based on i n c l u s i o n s that have trapped a s i n g l e , homogeneous phase may exceed that from the t w o - f l u i d methods because the d i f f i c u l t requirement of coeval i n c l u s i o n s i s not i n v o l v e d . The e f f e c t s on an i n c l u s i o n of events a f t e r i t i s trapped, although gene r a l l y a source of p o s s i b l y major e r r o r in both geothermometry and geobarometry 284

io Figure 9-10. P h o t o m i c r o g r a p h s o f t h e same f a c e t t e d t e t r a h e d r a l i n c l u s i o n i n f l u o r i t e a t two l e v e l s of f o c u s . The b u b b l e i n t h e p h o t o g r a p h on the r i g h t has a diameter c o r r e s p o n d i n g to 63% g r e a t e r volume t h a n t h a t on t h e l e f t . Presumably, i f v i e w e d p e r p e n d i c u l a r t o one o f t h e t e t r a h e d r a l f a c e s , the c o r r e c t diameter (whatever i t i s ) would be s e e n . Sample from S e r r a S ' l l i x i , S a r d i n i a , c o u r t e s y o f B . De V i v o .

of the original environment, may provide useful data on the pressures during t h i s post-trapping h i s t o r y . Although I do not believe that decrepitation of inclusions in the laboratory provides much quantitative thermometric data, natural decrepitation can prove useful. The rocks containing f l u i d i n c l u s i o n s , formed deep in the earth at elevated P and T, have dropped to present-day surface P and T via a generally unknown route on a P-T diagram for the appropriate f l u i d . I f the path taken i s such that the pressure within an inclusion exceeds the external pressure, the inclusion may deform or rupture i t s host c r y s t a l . Evidence for such natural decrepitation, reported by numerous i n v e s t i gators (e.g., see Voznyak and Kalyuzhnyi, 1976, and various entries in Roedder, 1968a), may provide valuable information on depths of burial and rates of u p l i f t and denudation; i t should be looked for routinely. Perhaps the most serious problem l i e s in the development of petrographic c r i t e r i a for assigning relative ages to the various generations of secondary inclusions that may reflect such decrepitation in many metamorphic samples. One aspect of natural decrepitation that may hold considerable promise for c l a r i f y i n g the time sequence in certain geologically complex terranes has not been adequately applied in the past. This i s the recognition of natural decrepitation from the pressure r i s e due to local heating, e . g . , at the intersection of veins and dikes (Roedder, 1977b). S i m i l a r l y , some constraints on the P-T path taken by magmas during ascent and eruption may be obtained from data on those high-pressure CO2 inclusions that have decrepitated and on those that have not (Roedder, 1965a). A study of the systematics of stretching of overheated f l u i d i n c l u s i o n s (Bodnar and Bethke, 1984) suggests the p o s s i b i l i t y of recognizing naturally stretched i n c l u s i o n s . I f so, we can use such naturally stretched inclusions as geobarometers. Bodnar and Bethke (1984) showed that the internal pressure necessary to i n i t i a t e stretching of f l u i d inclusions in f l u o r i t e varies in a systematic manner as a function of inclusion s i z e but that, once begun, the amount of stretching i s a linear function of the internal pressure. When a f l u o r i t e crystal has been overheated by some later thermal event, i t s large inclusions might have been stretched but other smaller inclusions might not have been, and heating tests would reveal an increase in f i l l i n g temperature with inclusion s i z e for inclusions exceeding a certain minimum s i z e . Then, knowing the pressure necessary to i n i t i a t e stretching of the smallest observed stretched i n c l u s i o n , and assuming that t h i s pressure represents the difference between the internal and external pressures e x i s t i n g when stretching occurred, the external pressure might be calculated. This pressure i s the external pres285

7

6 5 O o

4 ^

o

0

0

10

20

30 40 WT. % No CI

50

60

70

Figure 9 - 1 1 . D e n s i t y o f v a p o r - s a t u r a t e d NaCl f l u i d s i n the system N a C l - ^ O , c a l c u l a t e d by Bodnar (1983) u s i n g a s t e p w i s e m u l t i p l e r e g r e s s i o n from l i t e r a t u r e data and p l o t t e d on h i s Figure 4. References used are S o u r i r a j a n and Kennedy ( 1 9 6 2 ) , Potter et a l . ( 1 9 7 7 ) , and K h a i b u l l i n et a l . (1 9 8 0 ) .

sure at the time of s t r e t c h i n g and not n e c e s s a r i l y the p r e s s u r e during t r a p p i n g of those p a r t i c u l a r i n c l u s i o n s . Furthermore, evidence suggests that f o r a given pressure d i f f e r e n c e , the s t r e t c h i n g i s dependent on the c o n f i n i n g pressure (Poland, 1982). The above d i s c u s s i o n shows that of the numerous i n c l u s i o n geobarometers used i n the p a s t , some are wrong in concept and can y i e l d g r o s s l y inaccurate p r e s s u r e s , and others need an independent geothermometer to y i e l d an estimate of the p r e s s u r e . S i n g l e i n c l u s i o n s can y i e l d only data that are f u n c t i o n s of both P and T, but p a i r s of i n c l u s i o n s trapped from immiscible f l u i d s ( e . g . , l i q u i d water and steam, or water and CO2) can y i e l d both P and T. Inclusions of an immiscible water (or CO2) phase along with a s i l i c a t e - m e l t phase are a s p e c i a l example: the s i l i c a t e - m e l t i n c l u s i o n s provide, i n e f f e c t , an independent thermometer, and hence the p a i r can provide a geobarometer. Any determination of the pressure of formation from i n c l u s i o n s requires the following: 1. Good evidence of the time of t r a p p i n g of the i n c l u s i o n r e l a t i v e to the g e o l o g i c process being s t u d i e d , from careful microscopy. 2. Good evidence of freedom from v a r i o u s secondary e f f e c t s on the i n c l u s i o n s i n c e t r a p p i n g , both i n nature and the laboratory (necking down, leakage, decrepitation, stretching, etc.). 3. Good thermometric data (Tm i c e , Tm daughter m i n e r a l s , Th) on the phases i n the i n c l u s i o n (and sometimes volumetric measures of the i n d i v i d u a l p h a s e s ) . 4 . Good compositional data on the i n c l u s i o n from thermometry and other procedures. 5. Good experimental P-V-T-X data c o v e r i n g the necessary range of c o n d i t i o n s , on appropriate systems f o r the composition found. The most s e r i o u s handicap to accurate geobarometry at the present time i s the lack of good experimental P-V-T-X data. 286

IO

-

s o (*~E3)

t-

o

(•D

3 0 (• •) 158

(•D

200

ra

m

70

( Q 100

CD CD m

m «

®

440

€D A 500

(•D œ> 350°C) may have bubbles, whereas inclusions as large as 20 pm in some minerals formed near 100°C seldom show bubbles. Aqueous inclusions formed at 70°C may be as large as 100 urn and still not nucleate a vapor bubble (Roedder, 1967a).

1/ The converse of this statement (I.e., inclusions without bubbles must have formed at surface t e m p e r a ture) is not generally valid, as d e t a i l e d below.

292

A

,p

F i g u r e 10-1. P-T plot of the phase diagram for w a t e r (not to scale) showing the stable phase boundaries (solid lines) and m e t a s t a b l e extensions (dashed) in the immediate vicinity of the triple point B (0°c and 4 ran pressure). (See F i g s . 8-6 and 10-4 for true-scale p l o t s . ) A homogeneous inclusion cooled from point (p) should nucleate vapor at (q); if it does not, it continues on the m e t a s t a b l e extension of J s o c h o r e (p-q) to some point (r) or (r'), w h e r e vapor nucleates in the m e t a s t a b l e - s t r e t c h e d liquid and the inclusion returns to the L + V line. The several types of metastability commonly e n c o u n t e r e d on cooling inclusions are illustrated by paths 1 through 5. See text for details.

P

0

The explanation for the lack of a vapor bubble is usually assumed to be simply metastability from lack of nucleation. Thus, an inclusion trapped at point "p" on Figure 8-1, on cooling, would follow isochore "p"-"q". At point "q," it should nucleate a vapor bubble of specific volume "q 1 ," but if it does not, it follows "pq" extended along the (hypothetical) metastable curved surface until finally at some point "r" it nucleates a bubble and hence rises to the surface ABCcp. Figure 10-1 shows the behavior of the same inclusion on a P-T plot. The point "r" at which nucleation occurs can fall anywhere along the line pq extended and may even fall in the area of negative pressure ("r" 1 on Fig. 10-1; discussed below). The full explanation probably involves several superimposed phenomena. First, surface-tension forces result in higher pressures inside bubbles, in inverse ratio to the radius of the bubble. As the solubility of gases in liquids increases with pressure, and surface tension causes high pressures in small bubbles, very small bubbles in a free liquid or foam collapse instantaneously by dissolution of the gas in the liquid. Somewhat larger bubbles dissolve rapidly, effectively transferring their gas to other, still larger bubbles.?/. Fluid inclusions are essentially isochoric (fixed-volume) systems, however, and hence yield very different results. Whether the bubble is undissolved gas or merely the vapor of the surrounding fluid, the only way surface tension can cause it to collapse under isothermal isochoric conditions is to stretch the liquid, perhaps even into the range of negative pressures (Roedder, 1967a). The size division between inclusions with and without bubbles thus reflects the balance between surface tension tending to collapse the bubble and shrinkage (internal tension) of the liquid trying to form it. As the surface tension varies inversely with the radius of the bubble (and hence inversely with the cube root of the bubble volume), but the internal tension is independent of volume, there is a bubble volume (and, hence, an inclusion volume for a given composition and total density) below which the fluid will remain stretched indefinitely (i.e., without nucleation of a bubble), as the stable configuration. Even if a bubble did

y Such phenomena are very easily o b s e r v e d u n d e r the m i c r o s c o p e in soapsuds or in an effervescing chemical reaction.

293

micro-

Inclusion

radius /
200°C >270°C

P >1,200 bar >1,700 bar

Frey et al. (1980b) combined these inclusion data with measurements of the 353

c r y s t a l 1 i n i t y of i l l i t e ( I C ) and coal rank (by means of r e f l e c t i v i t y , R) from the same s e r i e s of very low grade metamorphic rocks s t u d i e d by M u l l i s . They found a general i n c r e a s e i n IC and R from t e c t o n i c a l l y h i g h e r to lower u n i t s and from external t o i n t e r n a l parts w i t h i n the same t e c t o n i c u n i t ( e . g . , a given nappe), as might be expected as an i n d i c a t i o n of i n c r e a s e i n metamorphic grade. I n the lowest grade zone, the "deep d i a g e n e t i c zone," the f l u i d s contained >1 mole % of higher hydrocarbons and had Th ~5 kbar and >~450°C to magmatic temperatures). High-grade metamorphic rocks are much lower in total volatiles than are the lower grade rocks, as the devolatilization process has gone much further. Their porosity is lower and i s often close to zero, the volatile-bearing phases they have are lower in volatiles and/or in amount, and they generally have relatively few f l u i d inclusions. Ambiguity about inclusion origin in these rocks can be as great as or greater than that for the lower grade rocks. Some rocks may be essentially free of inclusions. Most of the inherited inclusions (e.g., in detrital grains in metasediments) will have been swept out by recrystallization, along with most of the original depositional textures. Furthermore, as a result of higher temperatures, and perhaps longer times at those temperatures, the minerals may be much closer to equilibrium with each other and with the small amount of f l u i d phase present as pore f l u i d s and inclusions, in terms of both chemical and isotopic composition. The porosity and permeability are near zero, and even the microcracking appears to be a retrograde phenomenon (Padovani et a l . , 1982), yet differences 361

in fluid composition as a result of differences in local host-rock mineral assemblages over an outcrop are smaller than those in the lower grade rocks, or even immeasurable. Over distances of kilometers, however, the differences can still be striking. Commonly superimposed on this picture are the effects of retrograde reactions, which may involve the introduction of new fluids from external sources. Thus, adjacent planes of secondary inclusions may have grossly different compositions or densities, just as they do in the lower grade rocks. In some terranes equilibrated under high T-P vapor-absent conditions, the inclusions may only be from late-stage events, but even such data can constrain the uplift path in P and T (Selverstone et al., 1983). Lastly, the uncertainties in estimates of P and T are generally larger than those in the lower grade rocks, mainly because the fluid-inclusion data must be extrapolated further to the higher P and T values. As in Chapter 12, I have excluded contact-metamorphic rocks, most ore deposits in metamorphic rocks or those formed by metamorphic fluids, and kimberlites and ultramafic nodules, even though some of these may have formed in this same range of P and/or T; they are discussed in Chapters 14, 15, and 17, respectively. The dividing line between high-grade metamorphic environments and igneous processes is much more nebulous than that between medium- and high-grade metamorphism (Chapters 12 and 13). Even if an arbitrary division based on temperature is chosen, the data to which it needs to be applied are seldom accurate. The selection of a significant upper dividing line, e.g., in the series gneissgranite or granulite-granite, is even more difficult and risks a surfeit of semantic arguments as well. Touret (1981) considers the first formation of "mobilisâtes" — migmatitic to pegmatitic stringers — as a convenient boundary. Touret (1981) has pointed out that the study of inclusions in such highgrade rocks started a relatively few years ago, with the publication of three important papers: Dolgov et al. (1967) on dense CO2-H2O inclusions in kyanite from metamorphic rocks and pegmatites in the USSR; Touret (1971) on Norwegian granulites; and Hollister and Burruss (1976) on the Khtada Lake complex in Canada. Since then, still other major studies have been made, but much is still unknown. FLUID COMPOSITIONS VS MINERAL ASSEMBLAGE Nonmineralized Rocks The common achievement of equilibrium between fluids and solids at higher temperatures, or at least a close approach to it, in contrast to the lack of phase equilibrium in lower temperature environments, permits detailed comparison among results from experimental metamorphic petrology, theoretical predictions of equilibria, and the natural mineral and inclusion assemblage. Eugster (1981, 1982) provided excellent reviews of the present status of this comparison (see also Loomis, 1983; Bowers and Helgeson, 1983b). In a study of fluid inclusions in calc-silicate rocks from Prince Rupert, British Columbia, Canada, Crawford et al. (1979a) showed that not only did the CO2/H2O ratio of the inclusions generally agree with the calculated ratio for the mineral assemblage but that the aqueous inclusions in quartz grains in the calc-silicate rocks were brines containing a very high proportion of CaCl2 (some had Te as low as -60°C), whereas quartz in nearby carbonaceous semipelites had inclusions containing dense CO2 and CO2-CH4 mixtures. This difference suggests internal control of the fluid composition. Yardley (1979) and Yardley et al. (1983) reported that the dominant fluid in inclusions from the upper -si 11 i manite zone (temperature maximum ~640°C) in some Dalradi'an metasediments is CH4. These inclusions had Th near -110°C. Bodnar and Connolly (1983) showed that fluid pressure could be derived from integration of fluid inclusion and mineral phase equilibria, particularly for assemblages in equilibrium with CO2/H2O mi xtures. 362

In addition to CO2/H2O, the ratio K/Na is particularly useful. Poty et al. (1974) calculated the K/Na ratio in the fluid phase that should be in equilibrium with two feldspars (adularia and high albite); they showed that analyses of K/Na in fluid inclusions in quartz from such assemblages could be used as a geothermometer. Luckscheiter and Morteani (1980a) applied this geothermometer to analyses they made of inclusions from quartz from Alpine veins in gneiss, amphibolite, and mica schist. Their results (435-490°C) are lower than estimates made by others on the basis of 1 8 0 / 1 6 0 studies (500-600°C). They ascribe the differences to a combination of sampling problems (primary vs secondary inclusions) and the influence of CO2 on the equilibria (because this geothermometer "is valid only for C02-free aqueous inclusions"). They also concluded that the very high Ca/Na and Mg/Ca ratios they obtained on extraction and analysis of fluid inclusions in quartz are probably spurious, as a result of contamination from solid inclusions. Poty et al. (1974) got much lower ratios in otherwise equivalent quartz samples from the Western and Central Alps. Touret (1981, p. 190) reported that the aqueous inclusions in high-grade metamorphic rocks (in which most inclusions are CO2) range widely in salinity from halite-bearing inclusions to almost pure water in a single specimen. Apparently this range cannot be explained by necking down. In large part it may be caused by multiple sources of introduced fluid, presumably at different times, or it might result from fluid immiscibility, but because no experimental data on the system H20-NaCl-C02 are available for these P-T conditions, quantitative evaluation is not possible. For many years before the abundance of CO2 in metamorphic fluids was recognized (as evidenced by the fluid inclusions), the common assumption in metamorphic dehydration reactions was that P(H20) = P(total). We now know that H2O and CO2 each act as major diluents for the other in such fluids, and many current studies have concentrated on the calculated H2O/CO2 ratio and its many important implications (e.g., Jacobs and Kerrick, 1981b), but several other constituents can be examined by similar procedures. Any mineral whose structure can accept several different volatile species has the potential of providing a record of the ratio between those species in the fluid, to be compared with that obtained from the fluid inclusions, provided the necessary partitioning data are available. The substitution of F for OH in micas and amphiboles provides an example. Petersen et al. (1981) have been able to estimate the fugacities of H2O, HF and F (as well as CO2, C H 4 , CO, H2, S2, and O2) in the fluids during metamorphism of Grenville marbles near Balmat, New York, USA, on the basis of the mineral assemblage. They showed that failure to consider possible F substitution can lead to large errors in estimated P, T, and fluid compositions. Rozen et al. (1977) suggested that the composition of scapolite and apatite can be used as indicators of the composition of the volatiles during metamorphism of the Kola Peninsula granulites, USSR. Cordierite, with its well-known channels, can accept both CO2 and H2O (as well as organic compounds), and the CO2/H2O ratio in it has been investigated by several groups (Armbruster and Bloss, 1980; Johannes and Schreyer, 1981; Zimmermann, 1981). The interpretation is complicated by variation in the partition coefficient with variables other than merely the ratio of the surrounding fluid. Thus, Coolen (1980) showed that the S/0 ratio in scapolite from Tanzanian granulites is a sensitive pressure indicator under S-saturated conditions. The presence or absence of NaCl-rich scapolite has been used as an indicator of the nature of the fluids. The presence of highly saline fluid inclusions in or associated with sodic plagioclase suggests the possibility of reaction to form scapolite, perhaps during cooling. Such reaction products have been looked for but not found at Ascension Island (Roedder and Coombs, 1967). Vanko and Bishop (1982) showed that this reaction proceeds, but only at 700°C and above in very strong brines [X(NaCl) >0.64 or >85 wt % NaCl].

363

2 Oum

Figure 13-1. Photomicrographs of inclusions in emeralds from Chivor and Muzo emerald mines, Colombia, taken in plain transmitted light. Samples courtsey of Banco de la República, Bogota, Colombia, (a) Photomicrographs taken at +27.5°C (left) and +in.0°C (right) of a multiphase inclusion in emerald (E), showing a relatively small vapor bubble (v) and a large daughter crystal of halite (h) in saturated brine (lw). The bubble contains CO2 gas at moderately high pressures, as shown by a crescent-shaped fillet of liquid CO2 (1c, right photo) that evaporates into the bubble when warmed to room temperature. Sample ER 63-139b, Muzo, (b) Photomicrograph of inclusions in emerald (E), showing an isotropic daughter crystal (h) of apparently hexagonal outline (actually it is an octahedron) that is believed to be halite, plus vapor (v) and liquid (1). Why this particular halite daughter (and some others) should be an octahedron rather than a cube is unknown. Sample ER fi3-139b. Muzo, (c) Photomicrograph of a cluster of multiphase inclusions in emerald. The amoeboid mass in the center right is all one large inclusion containing a black vapor bubble (v) and a large daughter crystal of halite (h; enlarged in photo (e)). Several tubular inclusions with the same phases are shown at bottom of photo; in photo (d) they are shown enlarged. Sample ER fil-6, from an unspecified locality in Colombia (probably Chivor). (d) Enlargement of the group of tubular inclusions shown at bottom of photo (c). Each contains one or more crystals of halite (h), a small anisotropic grain (b), a gas bubble (v) with considerable CO2 under pressure (see also (a)), and a very strongly saline solution (1) containing other ions in addition to NaCl. In the two partly overlapping very thin, acicular inclusions (1) and (2), the darkest band is vapor, liquid is lighter, and the NaCl crystal (h, in contact with the emerald walls), is bright and almost invisible (this is shown particularly effectively in inclusion 1). Inclusion 3 is an all-liquid inclusion, presumably formed from necking down of the larger inclusion visible in photo (c). (e) Enlargement of part of the halite daughter crystal visible in photo (c), showing many fluid inclusions in the daughter crystal, some containing liquid and vapor in various ratios (arrows), as a result of growth at various stages during the cooling of this emerald or of necking down of a larger amoeboid daughter crystal, or both. Scale bars in urn.

Although of

former

soluble

reference

evaporite

chemical

rocks

has

of

evaporite

an

2000

been

m.y.,

Precambrian The able of

hydrous

Sabine

this

made

and

sequence anhydrite

in

Chapter

during

constituents,

reported

an

occurrence

is s u g g e s t e d

(Leake is

et

one

12

al.,

of

to

the

metamorphism,

the

to

of

As

earliest

to

the

anhydrite

represent

1979).

obliteration

and

in

the

these

of

evidence

dispersal

metamorphosed rocks

evidences

of

of

the

granulite-facies are the

remains

older

than

salinity

of

oceans.

amount

during

was

sequences

of

H£0

high-grade minerals

(1973)

showed

is

the

largest

metamorphism.

present, that

such

a series

as of

unknown

in

In p a r t

this

hornblende nominally

364

the

composition

will

be

of

evident

or mica.

However,

anhydrous

silicates

fluids in t h e

availamounts

Wilkins such

as

and

kyanite, andalusite, andradite, pyrope, diopside, rhodonite, andesine, adularia, and olivine, all have OH contents equivalent to at least 0.008 wt % H2O (specifically excluding H2O as liquid inclusions). A water content of 0.008% may seem insignificant but amounts to 8xl0 5 tons of H2O in every cubic mile of such rocks. Perhaps even more significant to fluid-inclusion studies was that these values, obtained by IR absorption spectra, were verified by observing the changes when H was replaced by D after treatment with 100 bars D2O pressure at 750°C for 15 vol %), a small amount of liquid CO2, and a moderate-sized vapor bubble (~ll%).i/ These fluid inclusions provide excellent illustrations of a variety of aspects of inclusion study (see also Figs. 3-16, 4-8). As Te for these inclusions ranged from -63 to -58°C, and Tm ice was — 3 4 ° C (Roedder, 1963), major amounts of Ca (and probably other ions) must be present. These inclusions were exceedingly difficult to freeze, and some never froze (see Chapter 10, p. 297). Although liquid CO2 is present (e.g., Fig. 13-la), the water phase showed no dissolved CO2, HCO3", or 003^" by Raman spectroscopy (Rosasco and Roedder, 1979). The absence of such spectra suggest that the concentrations of these constituents are probably each less than a few thousand ppm. The cubic daughter crystals (and more rarely, octahedra; Fig. 13-lb) of halite are so common in these emeralds-that they have been used to establish that (1) the stone is natural and not synthetic, and (2) that it probably came from the Colombian emerald mines (Gilbelin, 1953)..?/ The larger daughter crystals themselves very commonly contain fluid inclusions (Fig. 13-le).

U A small b l r e f r i n g e n t d a u g h t e r crystal commonly found 1n these Inclusions w a s originally thought to be parisite but was shown by Raman spectroscopy (Chapter 4) to be some phase (unidentified) other than parlsi te. U I have found NaC1 crystals in primary inclusions 1n Rhodesian (Sandawana) emeralds, however, and they have also been reported i n emeraIds from W e s t e r n A u s t r a l i a and India (Roedder, 1982c)• There have been several reports of "three-phase inclusions", presumably liquid plus gas plus NaCl crystal, in some s y n t h e As some of the procedures used to grow synthetic emeralds are tic emeralds (Webster, 1952; Wells, 1953). still s e c r e t , such inclusions could be important, but the true source of any given stone is difficult to verify, and some s y n t h e t i c emerald is c r y s t a l l i z e d on natural beryl seed p l a t e s . See also R o e d d e r (1982c, p. 501).

365

M o s t of the inclusions in C o l o m b i a n e m e r a l d s d e c r e p i t a t e d b e f o r e h o m o g e n i z a t i o n , but t h o s e that did not had Tm NaCl at ~ 3 3 0 ° C and still had a v a p o r bubble when the run was stopped at 355°C ( R o e d d e r , 1 9 8 2 c ) . The p r e s s u r e c o r r e c t i o n to be a d d e d to these values m a y be l a r g e . Escobar and Mariano (1976) s u g g e s t e d 6000 m of s e d i m e n t a r y o v e r b u r d e n , requiring a pressure c o r r e c t i o n of 1 6 3 ° C if the fluids are pure N a C l - H 2 0 s o l u t i o n s (Potter, 1 9 7 7 ) , and y i e l d i n g Tt > 5 1 8 ° C ; even greater depths of o v e r b u r d e n h a v e b e e n p r o p o s e d by o t h e r s . The high salt c o n c e n t r a t i o n s in t h e f l u i d may be f r o m t h e s o l u t i o n of s e d i m e n t a r y beds and domes of salt t h a t O p p e n h e i m (1948) r e p o r t e d n e a r b y . M e t a l l i c Mineral

Deposits

A l t h o u g h n u m e r o u s o r e d e p o s i t s have been e i t h e r f o r m e d by, or s u b j e c t e d t o , h i g h - g r a d e m e t a m o r p h i c e v e n t s , a genetic link is d i f f i c u l t to p r o v e (see also p. 4 6 1 ) . Many of t h e m a s s i v e s u l f i d e d e p o s i t s of t h e w o r l d have been s u b j e c t e d to m o d e r a t e - to h i g h - g r a d e m e t a m o r p h i c c o n d i t i o n s , but very few s t u d i e s of t h e m have i n v o l v e d any i n v e s t i g a t i o n of t h e f l u i d i n c l u s i o n s . Many of t h e ores t h e m selves do not a p p e a r very a p p r o p r i a t e for i n c l u s i o n s t u d y , but t h e host rocks m i g h t y i e l d useful d a t a . W i l k i n s (1977) c o n c l u d e d t h a t at B r o k e n Hill, New S o u t h W a l e s , A u s t r a l i a , i n c l u s i o n s f r o m t h e p e r i o d of o r e f o r m a t i o n h a v e b e e n e l i m i n a t e d by repeated d e f o r m a t i o n and r e c r y s t a l l i z a t i o n . He f o u n d t h a t t h e i n c l u s i o n s p r o v i d e d only a record of t h e s e q u e n c e of m e t a m o r p h i c fluids s i n c e t h e p e r i o d of h i g h - g r a d e m e t a m o r p h i s m , and s u g g e s t e d (p. 187) t h a t t h e r e is no easy way to p r o v e that " . . . e a r l y - f o r m e d i n c l u s i o n s h a v e not d e v e l o p e d t h e i r p r e s e n t c o m p o s i t i o n s by d i f f u s i v e loss of w a t e r or o t h e r c o m p o n e n t s d u r i n g t h e long p e r i o d of retrograde metamorphism. This a p p l i e s e s p e c i a l l y to t h e highly s a l i n e i n c l u s i o n s . . . . " He n o t e d , h o w e v e r , a c o n s i s t e n c y of c o m p o s i t i o n w i t h i n individual h e a l e d fractures and even o v e r d i s t a n c e s of s o m e k i l o m e t e r s . F l u i d flow a f f e c t e d t h e c o m p o s i t i o n , h o w e v e r , as c e r t a i n s h e a r zones w e r e c h a r a c t e r i z e d by a d i s t i n c t a s s e m b l a g e of inclusion compositions. T h e c o m p o s i t i o n s range f r o m h i g h - s a l i n i t y aqueous to high-density CO2 ±CH4. Both (1978) has s h o w n t h a t s o m e of t h e d e p o s i t s at B r o k e n Hill have been r e m o b i l i z e d d u r i n g r e t r o g r a d e m e t a m o r p h i s m by fluids h a v i n g Th = 1 8 7 - 2 0 4 ° C and s a l i n i t i e s - 2 3 wt %. S i m i l a r l y , K o n n e r u p - M a d s e n (1979b) d e s c r i b e d i n c l u s i o n s f r o m m o l y b d e n i t e m i n e r a l i z a t i o n a s s o c i a t e d w i t h a z o n e of a m p h i b o l i t i c h i g h - g r a d e b a n d e d gneiss in N o r w a y . These f l u i d s r a n g e d f r o m nearly pure C O 2 to e n t i r e l y a q u e o u s and m o d e r a t e l y saline. T h e l a t t e r he b e l i e v e s r e p r e s e n t f l u i d s p r e s e n t d u r i n g repetitive e p i s o d e s of m i c r o f r a c t u r i n g and p r o g r e s s i v e i n t r o d u c t i o n of m e t e o r i c w a t e r d u r i n g u p l i f t and c o o l i n g . T h e h i g h - d e n s i t y , C 0 2 - r i c h f l u i d s are c o n s i d e r e d t o a p p r o x i m a t e m o s t nearly t h e f l u i d s p r e s e n t d u r i n g peak c o n d i t i o n s of m e t a m o r p h i s m , at 3.5 to 4.5 kbar and ~ 6 5 0 ° C . The high CH4 fluids so c o m m o n l y found in t h e lower grade m e t a m o r p h i c t e r ranes may well be i n v o l v e d in t h e f o r m a t i o n of g r a p h i t e vein d e p o s i t s . Frost (1979) has s h o w n , from c a l c u l a t e d e q u i l i b r i a in t h e s y s t e m C - O - H , t h a t graphite' veins can form by o x i d a t i o n of a CHz^-rich f l u i d p h a s e , thus p e r m i t t i n g t h e t r a n s p o r t of C f r o m one p l a c e to a n o t h e r (e.g., f r o m a g r a p h i t i c g n e i s s to a g r a p h i t e vein).

FLUID

I N C L U S I O N S , M E T A M O R P H I C Z 0 N A T I 0 N , AND

UPLIFT

T h e c o n c e p t of m e t a m o r p h i c f a c i e s does not i n v o l v e the c o m p o s i t i o n of the rocks i n v o l v e d but refers only to t h o s e rocks that have e q u i l i b r a t e d w i t h i n a given P - T range, usually as i d e n t i f i e d by s o m e s p e c i f i c (index) m i n e r a l s . As a result, one s h o u l d not e x p e c t t h e fluids in e q u i l i b r i u m w i t h each of the various a s s e m b l a g e s w i t h i n a given facies to be t h e same. H o w e v e r , if a given t y p e of s t a r t i n g material is f o l l o w e d t h r o u g h a p r o g r e s s i o n of m e t a m o r p h i c f a c i e s , as long as e q u i l i b r i u m is o b t a i n e d , t h e c o m p o s i t i o n of t h e f l u i d s s h o u l d be relatively c o n s i s t e n t f o r e a c h mineral a s s e m b l a g e . This p r o g r e s s i o n will be 366

Figure 13-2 Map of metamorphic zones and f l u i d - i n c l u s i o n data from f i s s u r e quartz c r y s t a l s along the "Geotraverse" of Frey et a l . (1980a, their F1g. 2) 1n the Swiss Alps. E, e c l o g i t e s .

i l l u s t r a t e d with two examples that may well be t y p i c a l . Central Alps of Switzerland Several of the most detailed studies of the relationship between f l u i d i n c l u s i o n s and other c r i t e r i a for metamorphic grade in medium- to high-grade rocks have been made in the Swiss Alps. A detailed study of a belt 50 x 100 km in size ("Geotraverse") across a series of metamorphic zones in the Central Alps has been made by a large number of i n d i v i d u a l s , over several decades, as summarized by Frey et a l . (1980a). The study extended from very low grade rocks in the north to g r a n u l i t e , eclogite, and garnet peridotite in the south (Fig. 13-2), covering the entire range of Grubenmann (1904), from his non- to 367

Table 13-1.

Fluid field (see F i g . 1 3 - 2 )

F l u i d f i e l d s , f l u i d composition and metamorphic grade of e a r l i e s t f l u i d i n c l u s i o n s i n the "Geotraverse" (see F i g . 1 3 - 2 ) . From Frey e t al . (1980a). Fluid composition**

Metamorphic grade

HHC*

- 1 to >80 mole % HHC (CH 4 , H 2 0 , C0 2 , NaCl )

CH4

- 1 to >90 mole % CH4, 99 mole % H 2 0 , 60 mole ï C0 2 (CH4, H 2 0 , NaCl) * HHC = higher

Nonmetamorphic zone

NaCl)

Low- and medium-grade anchizone H i g h - g r a d e anchizone and epizone Mesozone

hydrocarbons

* * In parentheses: a d d i t i o n a l s p e c i e s found i n i n c l u s i o n s . (90 mole % H2O) to CO2r i c h ( - 1 0 to >60 mole % CO2) at about the s t a u r o l i t e i s o g r a d (the boundary between the e p i - and the meso-metamorphic zones). Frey et a l . (1980a) l i s t e d the three p o s s i b l e o r i g i n s f o r t h i s CO2 put forward by Hoefs and S t a l d e r (1977): decarbonation r e a c t i o n s , o x i d a t i o n of g r a p h i t e , or j u v e n i l e gases. A l l h i g h grade metamorphic rocks such as g r a n u l i t e s c h a r a c t e r i s t i c a l l y have C 0 2 - r i c h i n c l u s i o n s (see l a s t part of t h i s c h a p t e r ) , and Frey et a l . a l s o reported some e c l o g i t e s having the h i g h e s t estimated temperatures and pressures (800°C and 25 k b a r ) . These p a r t i c u l a r P and T data were obtained not from i n c l u s i o n s but from element p a r t i t i o n i n g between phases. Frey et a l . (1980a) c a l c u l a t e d mean geothermal g r a d i e n t s f o r these areas and found that they decrease southward, from ~40 to ~25°C/km. Attempts to e s t a b l i s h the dip of the isotherms at the peak of metamorphism on the b a s i s of these data are subject to numerous problems. For example, as shown by England and Richardson (1977), the maximum i n P i s probably reached e a r l i e r than the maximum i n temperature, so there may be e r r o r s inherent in these gradient estimates. A l s o , metamorphism, and the t r a p p i n g of i n c l u s i o n s , was probably c o n t i n u i n g d u r i n g the t h r u s t i n g and f o l d i n g (and p o s s i b l y even d u r i n g the sedi mentat i o n ) . Khtada Lake Complex, B r i t i s h Columbia, Canada H o l l i s t e r and Burruss (1976) presented an e x t e n s i v e study of f l u i d i n c l u s i o n s i n matrix quartz from a high-grade metamorphic g n e i s s complex, the Khtada Lake complex, B r i t i s h Columbia, Canada. T h i s complex i s composed of v a r i o u s 368

hornblende and/or biotite gneisses (some almost "plutonio") and migmatites, the most l e u c o c r a t i c of the gneisses being t o n a l i t i c in composition. Field evidence suggests a q u a s i b a t h o l i t h i c environment. Abundant sillimanite and absence of primary m u s c o v i t e , coupled w i t h the occurrence of K - s p a r , suggests temperatures >650°C at 6 kbar. Some graphite is present. Although the inclusions contained only t h r e e phases (H20-rich fluid, C02-rich f l u i d , and gas, each from 0-100 vol %), t h e studies permitted an internally consistent interpretation of the t h e r m o m e t r i c results in terms of the geologic work in the area. Only a few of the m a j o r conclusions are discussed here. Most important, the work showed that the inclusion contents consisted essentially of H2O, CO2, CH4, and "NaCl." A l t h o u g h H o l l i s t e r and Burruss have collated widely scattered data from the literature on parts of the compositional range of this four-component system, in p a r t i c u l a r over parts of the P-V-T range of interest here, considerable physicochemical reasoning was needed to interpolate or extrapolate to fill t h e gaps. By a p p r o p r i a t e simplifying a s s u m p t i o n s , and by the existence of inclusions whose c o m p o s i t i o n s permitted treatment as a system of less than four components, they w e r e able to interpret the phase behavior of these inclusions both at low temperatures (melting phenomena) and at higher temperatures (homogénization t e m p e r a t u r e s ) . T h e basic data they used w e r e : estimated vol % liquid CO2, T m CO2, T m ice, T m c l a t h r a t e , Th CO2 L - V , Th CO2-H2O, and the phase in w h i c h homogénization took p l a c e . T h e s e measurements provided not only compositional data on the inclusions but also density data, thus defining t h e appropriate isochores. T h e lowest temperatures found (-56.5 to - 6 9 ° ? C ) w e r e for equilibria among CO2 solid, C02~rich liquid, and CH4-rich gas, in inclusions essentially in the system CO2-H2O (see Fig. 8-23 and discussion of it). Since CH4 is miscible with both t h e gas and liquid phases but not w i t h the solid, the three-phase assemblage changes the invariant t r i p l e point for CO2 to a line, w h i c h shifts to lower t e m p e r a t u r e s w i t h increasing CH^.i.' For such inclusions, the t e m p e r a ture of this t h r e e - p h a s e a s s e m b l a g e , in conjunction with the G/L ratio upon m e l t i n g , provides a measure of the mole fraction of CH4. The next higher t e m p e r ature w a s T m ice, observable in only eight inclusions. It ranged from -1 to -7°C in m i x e d CO2-H2O inclusions, suggesting moderately low salinities. (The actual salinity may be less than indicated, due to clathrate crystals removing H2O f r o m the s y s t e m , as pointed out by C o l l i n s , 1979.) A very few aqueous inclusions w i t h o u t visible CO2 showed higher salinity. At higher temperatures (+6.5 to + 1 7 ° C ) , t h e incongruent melting temperature of CO2-CH4 hydrates (clathrates) provides some constraints on inclusion composition, but only>with considerable ambiguity. Not only was the measurement necessarily of low accuracy, but these e q u i l i b r i a are affected by the nature of the gas m i x t u r e , the pressure, and the salinity (Fig. 8 - 2 2 ) . T h e next higher temperature observation was Th CO2 L+V (L). This homogénization o c c u r r e d at - 3 5 to +30°C, corresponding to CO2 densities from 1.09 to - 0 . 3 5 g / c m ' (Fig. 8-7); two inclusions h o m o g e n i z e d in the vapor phase and hence c o n t a i n e d lower density fluid. Th H2O + CO2 was next. This occurred, in either the C 0 2 - r i c h or t h e H20-rich p h a s e , at temperatures from 159 to 339°C (for those inclusions that did not decrepitate first). Interpretation of the pressure of t r a p p i n g of such CO2-H2O inclusions was based on an estimation f r o m t h e solvus at 1 kbar for t h e pure system CO2-H2O (Todheide and Franck, 1963) and the w a t e r rich leg of the solvus for 1 kbar and 6 wt % NaCl in H2O (Takenouchi and Kennedy, 1965b), as shown on Figure 13-3 (see also next section). T h e m a g n i t u d e of the effects of CH4 on raising this solvus could only be surmised.

U The a l l i t e r a t i v e memory aid: "If the low-temperature form has less, the 'inversion' temperature is lowered" is useful here. Solid CO2 is the low-temperature form, and hence the "inversion" (solid * liquid + vapor) is lowered by the addition of CH4.

369

Figure 13-3 1-kbar solvus in the HjOCO2 system. Data points are observed high-temperature i nc1u s i on-h omo gen i z ation phenomena. Note cluster of data points near the experimental curve for CO2 solubility in 6 wt X NaCl solution at 1 kbar. Since this cluster includes both types of homogenization, it implies that the crest of the solvus for the inclusion fluids is very near the H2O side, at -10 mole I CO2. From Holltster and Burruss (1976).

Although q u a r t z grains from numerous rock types w e r e examined, the intrasample composition range w a s sometimes as large as the intersample range. Some distinctions c o u l d be made in the nature of the inclusions vs their composition: T h e highest density CO2 inclusions w e r e small (400°C and p r e s s u r e s » 3 . 5 kbar (Todheide and Franck, 1 9 6 3 ) . The l i m i t s on t h e f i e l d of i m m i s c i b i l i t y , or s o l v u s , a r e w e l l known f o r t h e pure system CO2-H2O (see F i g s . 8 - 1 9 , 2 0 ) , but the changes i n s o l u b i l i t y of CO2 i n H2O w i t h v a r i o u s s o l u t e s ( i n c l u d i n g o t h e r g a s e s ) are known f o r very few c o m b i n a t i o n s of P, T , and X (see a l s o Bowers and H e l g e s o n , 1 9 8 3 a , b ) . A c t u a l l y , i n view of t h e p a u c i t y of experimental d a t a , the evidence f o r the extent of the f i e l d of CO2-H2O i m m i s c i b i l i t y i n t h e " s y s t e m " H2O-CO2-CH4NaCl at v a r i o u s p r e s s u r e s and temperatures l i e s mainly i n t h e p e t r o g r a p h i c evidence from the i n c l u s i o n s t h e m s e l v e s . Ypma and Fuzikawa (1980) recognized evidence of i m m i s c i b i l i t y between a C 0 2 - r i c h phase (some c o n t a i n i n g >H- «J O c O

I : 4» S O C O

T C 3 o se

427

made all these values too high by 2.9°C, Tm ice would range from -7.6 to 0°C, corresponding to s a l i n i t i e s of 11.3 to 0 wt % NaCl equiv., but Smith et a l . reported a s a l i n i t y range of 13.8 to 1.9 wt %. Several of the epithermal f l u o r i t e deposits seem to have formed from essent i a l l y heated surface waters, as they show Tm ice to be e s s e n t i a l l y 0.0°C. These inclusions commonly show metastable superheated ice, however, so many y i e l d i n v a l i d determinations of Tm ice. Veins of coarsely c r y s t a l l i n e f l u o r i t e plus minor opal, c a l c i t e , barite, etc., at Browns Canyon, Chaffee County, Colorado, USA, provide a good example of the problems presented by the inclusions in such deposits (Roedder, 1977a). Th, measured on 179 inclusions that had a bubble at room temperature, ranged from 119 to 161 °C, and averaged 141°C, but many l i q u i d , one-phase inclusions were found ( i . e . , no bubble). These one-phase, l i q u i d - f i l l e d i n c l u s i o n s are believed to represent metastable equilibrium (at room temperature); i . e . , they were trapped at ~140°C but now contain stretched l i q u i d under negative pressure, from f a i l u r e to nucleate a vapor bubble since the original cooling ~7 m.y. ago. Three facts support t h i s contention: First, no two-phase i n c l u s i o n s were found that had intermediate Th ( i . e . , 50 wt reflected in as many as 10 daughter c r y s t a l s . Nash and Cunningham (1973) reported (by optical study) h a l i t e , s y l v i t e , f l u o r i t e , hematite, anhydrite(?), c a l c i t e ( ? j , and 9 unknowns (Fig. 15-9). Detailed SEM and microchemical studies were made by Metzger et a l . (1977) on these daughter c r y s t a l s , which may constitute >50 vol % of the inclusions (see Fig. 5-14). The major phases i d e n t i f i e d were gypsum, a calcium aluminum s i l i c a t e (probably a n o r t h i t e ) , barite, c e l e s t i t e (normally c e l e s t i t e shows retrograde s o l u b i l i t y , but apparently i t i s prograde in these strong s o l u t i o n s ) , h a l i t e , s y l v i t e , thenardite [Na2S04], and possibly arcanite [K2SO4] and g l a s e r i t e [ ( K . N a ^ S O ^ . Other phases found include ferroan rhodochrosite and phlogopite. Metzger et a l . (1977) could not confirm the presence of hematite, anhydrite, or c a l c i t e . Nash and Cunningham also noted that some daughter c r y s t a l s grew when the i n c l u s i o n s were heated to 150-180°C, and others did not dissolve in 3 or more hours of heating. Whether the c r y s t a l s that grew were a phase that switched from retrograde to prograde s o l u b i l i t y with temperature increase or a new phase from some incongruent s o l u b i l i t y i s not clear, but they did report a different new phase that formed at ~120°C and redissolved by ~185°C. Bluebell mine, B r i t i s h Columbia, Canada Ohmoto (1968) and Ohmoto and Rye (1970) made very extensive and detailed mineralogical, geochemical, inclusion and isotopic studies of t h i s Pb-Zn-Ag limestone replacement ore deposit. Although most (-90 %) of the ore i s massive, the l a t e s t part i s represented by euhedral c r y s t a l s , including quartz, in vugs, suitable for i n c l u s i o n study. Only one specimen of quartz and l i g h t sphalerite from the massive ore was s u i t a b l e . In >5000 i n c l u s i o n s studied, no evidence of b o i l i n g or NaCl daughter c r y s t a l s was found in any sample. Small carbonate and possibly c h l o r i t e and/or d i c k i t e c r y s t a l s are present in some i n c l u s i o n s , and CO? hydrate c r y s t a l s were observed on cooling in most of the larger i n c l u s i o n s . The few inclusion data on the massive ore showed Th < 410°C (Tt estimated to be < 445°C and the pressure correction was estimated at . 3 a. o •«*•

cm O o

*

"O O) "O -M

» oc «e ^ oc X 4> i ; 3C en o i l c V > WC 1/1 •e E «i «e i C C S » o» o 3 o 3 » »u. c L U C a. co "O ^•e «.-•>, o o k evi _c cu i « O » — . « >tO flj T E » e. « — a) . o « o • • - .e EJ3 E V • 4» •«- 4» E * .C < J

3 3 «B O •s «e a. o u o >> E E 4i c -C s « • o EE » C ai C p" -ic

• c o " 41 « e evjfj (_ O •*-*-> »Ö o >

l i

T> en oí - C •!- -I- IO ID

c o o o c o CCICCU10 u") a? in ic v £ >

o c © o o o ID C C V O

e o o o i j n

o o o o o o o o LT io V C 00

C C T >I •r- (/> t ati c o. c

- O 3

c •«3 O

a. 4) W Q. « 5 Q. +J - «-H O U

9 4)* c or m (A M 41 (/I ÍO o < _ > JZ C -.-CL O 41 •«*A V « E -C. O T3 L ».C O o. >, c co 41 m 4) 4) » "O 3 W JZ 0> O E U O c 41 t_ O C i- — I L. r « « i c ig 3 Z > N l/l >•

«•—• '>- O. X .C ->41 O- — . » —

•o 3 « «a o. 3 3 CL •r

o

-o

ifl o 3 3 "O 4) 4J O •> L — >>

z -o >> c« C D 4) O O C 1Ä O 1«3 « T IA C •or n C C 3 C _

o o V O o «í ID

r-< (Tl

z

N

OÎ

n i f j r s if i -S,S|S£- T «Í U

Uli j

fj

!|

.1

8

'Jüjé

SiisiWiS m i o W

ihh




I: si

1 i;

. g W I

lîr-i

'

lîiiïjili

a

.

r- U

W i I

a

. . . Ü W W W a>

s 'c

X

-g,

S*

1 ! Sis

¡p^lfps

iff

s

í

•§

S S »

S

x ' 5 §

o

fitti

o ®

o s i

.

w

¿

« -

'S S

g

l

«:-¡ Ils «pi!-::-! Ilifói!

j g

i

siili

I

443

Figure 15-12. Polyphase i n c l u s i o n s i n v e i n quartz from Chorolque porphyry Sn d e p o s i t , B o l i v i a . From Grant et a l . (1977, 1 9 8 0 ) . The i d e n t i f i e d phases a r e : 1 - v a p o r bubble; 2-hydrous Fe c h l o r i d e ; 3 - h a l i t e or (NaK)Cl; 4 - f l u i d ; 5 - u n i d e n t i f i e d . F i g u r e 1 5 - 1 3 . Polyphase i n c l u s i o n i n q u a r t z of q u a r t z - t o u r m a l i n e a l t e r e d vent rock from Chorolque porphyry Sn d e p o s i t , B o l i v i a . From Grant et a l . (1977, 1 9 8 0 ) . Phase i d e n t i f i c a t i o n as i n F i g u r e 15-12, 6 - p r o b a b l y s y l v i t e , and 7, 8, and 9, a l l u n i d e n t i f i e d . I n s e t shows b i r e f r i n g e n c e of hydrous Fe c h l o r i d e i n c r o s s e d p o l a r s .

c h a r a c t e r i s t i c s of most porphyry deposits, very high temperatures (~400-800°C), very high s a l i n i t i e s (~40->60 wt % s a l t s ) , and " b o i l i n g " sets them apart from most other ore deposits as completely as do the geological c h a r a c t e r i s t i c s . Noticeable differences in the f l u i d - i n c l u s i o n c h a r a c t e r i s t i c s among the Cu-Au, Cu, Cu-Mo, Mo, Mo-Sn, and Sn-W porphyries certainly have some genetic s i g n i f i cance but are not completely understood at t h i s time. Regardless of the physical and chemical s i g n i f i c a n c e of these three attributes of the f l u i d s , the presence of i n c l u s i o n s indicating such conditions can apparently be used as an empirical " f i n g e r p r i n t " for porphyry Cu^ deposits, and hence can be of value as an exploration tool for t h i s one large and important class of deposit (Roedder, 1971b, 1977b; Cunningham, 1975; Nash, 1976). In contrast to the above data, strongly s a l i n e , but apparently somewhat lower-temperature i n c l u s i o n s were found in a s e r i e s of barren intrusions having similar chemistry and level of emplacement in Colorado (Bradbury, 1 975); Steve Ludington and T.G. Theodore (pers. comm.) also have found such inclusions in other barren i n t r u s i o n s . The relationship between such h i g h - s a l i n i t y inclusions in barren i n t r u s i o n s and apparently similar inclusions in productive porphyry Cu i n t r u s i o n s i s an open question. R.J. Bodnar (pers. comm.) has suggested that there may be £0 s i g n i f i c a n t difference in the major c h a r a c t e r i s t i c s such as s a l i n i t y , Th, b o i l i n g , composition, e t c . ; only the metal content of the f l u i d s may d i f f e r , for reasons s t i l l to be established. Heavy-Metal Content of Inclusions The heavy metal content of the inclusion f l u i d s in the porphyry deposits has been estimated by various methods, including chemical analysis of leachates (Hall et a l . , 1974), and by calculations based on the volume percentage of various heavy-metal-bearing daughter phases, p a r t i c u l a r l y those i d e n t i f i e d by SEM studies of opened i n c l u s i o n s (see Fig. 10-2). Fe i s most commonly recognized in the form of hematite^/, generally corresponding to 1000°C (Bell et a l . , 1975). Possibly similar intergrowths of vermiform spinel plus pyroxene(?) are abundant i_n the o l i v i n e of some mare basalts (Fig. 18-23), as well as in o l i v i n e from t e r r e s t r i a l basalts (Roedder, 1976b, F i g . 5). These i n c l u s i o n s , which sometimes appear to decorate growth 553

layers in the host c r y s t a l , are almost opaque rounded masses 5-15 pm in size that consist of curving radial rods ~0.2 pm thick of chrome spinel (?) in an unidentified host (pyroxene?) that i s not the same as the surrounding o l i v i n e . As possibly s i m i l a r features, decorating dislocations in o l i v i n e , precipitated on heat treatment (Fig. 18-11), I suggest that both the natural symplectites in the o l i v i n e and those formed by heat treatment originated from exsolution of trace elements in the host o l i v i n e , but others disagree with t h i s hypothesis (Bell et a l . , 1975). Symplectite inclusions in plagioclase. Possibly similar inclusions of spinel were found in the plagioclase of lunar highlands t r o c t o l i t e 76535 (Bell et a l . , 1975) and in numerous plagioclase grains from various lunar s o i l s and breccias (e.g., Roedder and Weiblen, 1973, their F i g . 8; James and McGee, 1979, t h e i r F i g . 3). In some of these g r a i n s , the core and rim of the plagioclase have very different inclusion population densities. Some of these inclusions in plagioclase appear to consist not of spinel but of pyroxene plus a vapor phase; t h i s result i s exactly that to be expected i f a melt whose composition was e s s e n t i a l l y that of a mixture of pyroxene and plagioclase were trapped in plagioclase and cooled slowly. However, some inclusions of t h i s type in Luna 24 soil fragments that were thought to consist of pyroxene seem instead to cons i s t of o l i v i n e , thus suggesting a t r o c t o l i t i c melt, i f they are indeed melt i n c l u s i o n s . Sobolev et a l . (1980) performed heating experiments on possibly similar inclusions in o l i v i n e from Luna 24, and reported homogenization at 1300-1360°C. S i g n i f i c a n c e . The d i s t i n c t i o n between the two modes of o r i g i n proposed for these particular types of symplectite inclusions — c r y s t a l l i z a t i o n of melt i n clusions or exsolution of minor elements out of the host structure — i s of considerable importance but i s s t i l l an open question. In many samples, these inclusions, both in o l i v i n e (e.g., F i g . 18-23) and in plagioclase, appear to be present in high concentration, but when such materials are examined in reflected l i g h t , the concentration i s found to be « 1 % (400 ppm by volume in one Luna 24 feldspar). Hence, i f they are a result of d i f f u s i o n of trace constituents, the s t a r t i n g concentrations in the host need not be high. Also, i f these inclusions have formed by d i f f u s i o n and exsolution, we should ask what similar diffusion processes might have affected the composition of small true melt i n c l u s i o n s ? Inexplicable Inclusions In the pressure of examination of the lunar samples as fast as they were returned, a s u r p r i s i n g number of inexplicable inclusion types and some i n e x p l i c able data were reported. Only a few of the more s t r i k i n g types of inclusion problems can be mentioned below, in the hope of encouraging additional study to resolve the problems. Additional puzzles are found throughout the literature on lunar melt inclusions (e.g., Roedder and Weiblen, 1971, p. 519; 1972b, p. 278). Inclusions in o l i v i n e with high Kp values for Fe/Mg. The equilibrium d i s t r i b u t i o n of Fe and Mg between melt and o l i v i n e c r y s t a l s has been the subject of many studies. Equilibrium values of the d i s t r i b u t i o n coefficient Kp [Fe ( c r y s t a l ) - M g ( l i q u i d ) / M g ( c r y s t a l ) « F e ( l i q u i d ) ] , determined from laboratory data (Roeder and Emslie, 1970; Longhi et a l . , 1978; Walker et a l . , 1976; L i p i n , 1976; Grove and Bence, 1977; Takahashi, 1978), are about 0.3, whereas most of the Luna 24 inclusions (Roedder and Weiblen, 1978) and many of those from the Apollo samples (e.g., Roedder and Weiblen, 1971, p. 515), have analyses y i e l d i n g KD values between 0.5 and 1.5. Why? Several ad hoc (and hence not very s a t i s f a c tory) disequilibrium mechanisms can be visualized to explain these high values, invoking different d i f f u s i o n rates for Fe and Mg, either in the main melt adjacent to the growing crystal before the inclusion was trapped, or in the crystal adjacent to the i n c l u s i o n . The measured KD values also do not necessarily reflect the local equilibrium values at the actual inclusion/host interface. 554

Low-K i n c l u s i o n s i n i l m e n i t e . Many of the l a r g e i l m e n i t e c r y s t a l s in the lunar h i g h - T i mare b a s a l t s have 50-100 ym melt i n c l u s i o n s ( F i g . 1 8 - 2 4 ) . Some of these are merely the normal immiscible K - g r a n i t e melt ( i . e . , h i g h - S i melt, now a g l a s s ) c o n t a i n i n g - 7 6 wt % S i 0 2 and 6-7% K2O. Other g l a s s i n c l u s i o n s , even i n the same i l m e n i t e g r a i n s , a l s o have ~76% S i 0 2 , and may look i d e n t i c a l ( F i g s . 18-25 to 2 9 ) , but i n s t e a d of 6-7% K2O, they contain an average of 0.04% K2O, and many have ru) detectable K2O by e l e c t r o n microprobe. Most of the d i f f e r e n c e i s made up by CaO. About 70% of the 440 analyzed i n c l u s i o n s are low-K, and 30% are h i g h - K , and only very few have intermediate compositions (Roedder, 1979e). When these g l a s s e s were f i r s t described (Roedder and Weiblen, 1975), 13 suggested mechanisms of formation were presented, and each in t u r n was shown to be i n v a l i d . To my knowledge, no one has come forward s i n c e then with a fourteenth and workable h y p o t h e s i s . S t i l l other g l a s s y i n c l u s i o n s i n i l m e n i t e are high i n K (-2% K2O) but i n e x p l i c a b l y low i n S i 0 2 (~35%). A very few i n c l u s i o n s i n i l m e n i t e have apparently been cooled s l o w l y enough to c r y s t a l l i z e l a r g e amounts of t r i dymite or c r i s t o b a l i t e ( F i g . 1 8 - 3 0 ) . Gas i n c l u s i o n s in s i n g l e c r y s t a l s . Many of the igneous rock fragments i n the lunar b r e c c i a s and s o i l s contain i n t e r g r a n u l a r gas i n c l u s i o n s ( e . g . , from Luna 20, Roedder and Weiblen, 1973a, t h e i r F i g s . 9 and 1 0 ) . S i m i l a r i n c l u s i o n s are found i n many t e r r e s t r i a l samples. In hypabyssal rocks they are c a l l e d m i a r o l i t i c c a v i t i e s and represent the volume o r i g i n a l l y occupied by a residual H20-rich phase that f i l l e d the i n t e r s t i c e s between e a r l y c r y s t a l s . As the lunar magmas were extremely low in v o l a t i l e s , however, I believe that most of the gas i n c l u s i o n s i n them are e s s e n t i a l l y shrinkage c a v i t i e s . Once a magma has c r y s t a l l i z e d to the point at which there i s very l i t t l e i n t e r s t i t i a l melt, the c r y s t a l mush can support i t s e l f , and f u r t h e r volume reduction w i t h i n the mass (mainly during c r y s t a l l i z a t i o n of the remaining melt) w i l l r e s u l t in low i n t e r n a l p r e s sure on the l i q u i d phase and the formation of s h r i n k a g e c a v i t i e s between the existing crystals. I n these same s o i l samples, however, a few primary gas i n c l u s i o n s were found e n t i r e l y w i t h i n s i n g l e c r y s t a l s of ~An90 p l a g i o c l a s e ( F i g . 1 8 - 3 1 ) , and three small (2-4 ym) primary gas i n c l u s i o n s were found i n A p o l l o 11 o l i v i n e (Roedder and Weiblen, 1970b, p. 833). The " s a n c t i t y " of most of these samples prevented any c r u s h i n g - s t a g e s t u d i e s , but no v i s i b l e l i q u i d condensed when some were cooled to — 5 ° C . A primary gas i n c l u s i o n w i t h i n a s i n g l e c r y s t a l i n an igneous rock may form by several mechanisms. The most commonly considered mechanism requires the p r e s ence of a d i s t i n c t gas phase at the time of s o l i d i f i c a t i o n . Such a phase may form under a wide range of c o n d i t i o n s (Roedder, 1972), so the i n t e r p r e t a t i o n of the p h y s i c a l s i g n i f i c a n c e of the r e s u l t i n g i n c l u s i o n i s d i f f i c u l t . A gas i n c l u s i o n in a s i n g l e c r y s t a l ( F i g . 1 8 - 3 1 ) , could form by e x s o l u t i o n of gas from the o r i g i n a l melt, as a r e s u l t of the c r y s t a l l i z a t i o n of an e s s e n t i a l l y g a s - f r e e phase i n c r e a s i n g the concentration of gas i n the remaining melt, e i t h e r at the growing c r y s t a l s u r f a c e or throughout the melt. Primary gas i n c l u s i o n s of i d e n t i c a l appearance can a l s o form i n s i n g l e c r y s t a l s by an e n t i r e l y d i f f e r e n t p r o c e s s . I f the c r y s t a l ( p l a g i o c l a s e i n F i g . 18-31) grows from an e s s e n t i a l l y monominerallic melt ( i . e . , a n o r t h o s i t i c ) , and some growth i r r e g u l a r i t y causes the t r a p p i n g of some of t h i s melt, subsequent c r y s t a l l i z a t i o n of p l a g i o c l a s e from t h i s melt on the w a l l s of the c a v i t y w i l l form a gas i n c l u s i o n . The " g a s " w i l l then be a shrinkage bubble r e s u l t i n g from the volume decrease on c r y s t a l l i z a t i o n , s i n c e t h i s takes place w i t h i n a r i g i d container (the host c r y s t a l ) . As such, the bubble can be e s s e n t i a l l y a vacuum. Any components i n the melt that have r e l a t i v e l y high vapor p r e s s u r e s , such as a l k a l i e s and S compounds, w i l l , of course, p a r t i t i o n s t r o n g l y i n t o the vapor phase and may condense to s o l i d s at room temperature, once again l e a v i n g e s s e n t i a l l y a vacuum i n the i n c l u s i o n . 555

556

Figure 18-31. Single crystal of lunar p l a g i o c l a s e (P) from Luna 20 fragment 506.3 (~An90) enclosing a small area of p i g e o n i t i c pyroxene (Py) and several gas inclusions (G). Inclusions G1 and G2 are at the surface and hence could be merely c a v i t i e s from plucking of pyroxene grains during s l i d e preparation, but G3 and G4 (14 pm in s i z e ; enlarged in i n s e t ) are entirely surrounded by p l a g i o c l a s e c r y s t a l . Two c i r c u l a r areas (8) are a r t i f a c t s (bubbles in the mounting media). Transmitted p l a i n l i g h t . From Roedder and Weiblen (1973). Figure 18-32. Early melt inclusion in lunar o l i v i n e (0) from mare basalt 10045,32, now containing e p i t a x i a l daughter c r y s t a l s of ilmenite ( I ) and p l a g i o c l a s e ( P I ) , and a g l a s s (Gf) from which numerous l i t t l e blebs of very low-index, h i g h - S i g l a s s (Gs) have separated. An opaque sphere (presumably s u l f i d e , S) 1s out of focus. The shrinkage bubble (over the p l a g i o c l a s e c r y s t a l ) Is apparently packed with abrasive compound from sample p o l i s h i n g . Scale bar in pm. From Roedder and Weiblen (1970b). Figure 18-33. Melt i n c l u s i o n in o l i v i n e crystal (0) from 1965 Makaopuhl lava lake, Hawaii, USA (sample M-13-12), showing s p l i t t i n g into immiscible l i q u i d s , one h1gh-Fe (Fe) and one high-S1 ( S i ) . Scale bar 1n um. From Roedder and Weiblen (1971).

D i s t i n c t i o n between these two processes would be desirable but i s d i f f i c u l t , in the second process, c r y s t a l l i z a t i o n of plagioclase from the trapped, essent i a l l y monominerallic melt would concentrate the nonplagioclase constituents into a small amount of residual melt that might c r y s t a l l i z e to form daughter c r y s t a l s of o l i v i n e or pyroxene. Unfortunately, no such residue could be resolved on the walls of the small inclusion found (Fig. 18-31 i n s e t ) , nor on the walls of several other s i m i l a r i n c l u s i o n s . Even i f found, however, t h i s residue would not prove o r i g i n by the second process. Identical results could form by the f i r s t mechanism i f , as i s not uncommon, some of the surrounding melt was trapped along with the bubble. At present, I have no evidence enabling me to choose Figure 18-24. Completely i s o t r o p i c glassy high-K (4.37t K2O) melt inclusion (G) in lunar ilmenite ( I ) grain 75075,88,P8. The i n c l u s i o n i s viewed in transmitted plain l i g h t plus a small amount of reflected p l a i n l i g h t to show upper surface of ilmenite. Note shrinkage bubble (v) in g l a s s . Another s i m i l a r inclusion i s at center bottom. F rorn Roedder and Weiblen (1975). F i g u r e 18-25,26. Low-K melt i n c l u s i o n i n i l m e n i t e ( I ) of lunar sample 71175,34,P7 (0.1% K2O) viewed i n reflected ( F i g . 25) and transmitted ( F i g . 26) p l a i n l i g h t . I t includes a faceted s u l f i d e mass (S) and v a p o r ( ? ) bubble ( v ) . Note e x s o l u t i o n lamellae (arrows) and chrome s p i n e l (Sp) at contact of s u l f i d e with wall of i n c l u s i o n . F i g u r e 26 shows t i n y c r y s t a l s from d e v i t r i f i c a t i o n . Scale bar i n ym. From Roedder and Weiblen (1975). F i g u r e 18-27. Adjacent low- and high-K melt i n c l u s i o n s i n lunar ilmenite 71175,34,P2. Inclusion A i s pyroxene of the same composition as that outside and i s probably a reentrant. I n c l u s i o n s B and C are high-K (5.68% KgO) and low-K (0.009% K 2 0 ) , r e s p e c t i v e l y . Reflected p l a i n l i g h t . Scale bar i n um. From Roedder and Weiblen (1975). F i g u r e 18-28. Adjacent high-K and low-K melt i n c l u s i o n s i n a segment of a s i n g l e t a b u l a r ilmenite c r y s t a l i n lunar sample 70135,61,P3. I n c l u s i o n A contains 5.90% K2O; i n c l u s i o n B contains only 0.033% K ^ » The 55 urn of ilmenite between them shows no compositional g r a d i e n t s . Reflected p l a i n l i g h t . From Roedder and Weiblen (1975). F i g u r e 18-29. Adjacent high-K and low-K i n c l u s i o n s i n s i n g l e ilmenite g r a i n from lunar sample 75075,82,P26, i n r e f l e c t e d l i g h t . The low-K i n c l u s i o n (A) contains an immiscible s u l f i d e bleb, and the high-K i n c l u s i o n (B) has a vapor bubble (dark) and a p a r t i a l rim (arrow) of pyroxene(?). From Roedder (1979e). F i g u r e l 8 - 3 0 a , b . Melt i n c l u s i o n s , in lunar ilmenite ( I ) , now c o n s i s t i n g of brown g l a s s and s i l ica c r y s t a l s . The g l a s s of the i n c l u s i o n on the l e f t contains only 49.3% SiO2 but contains a large dendrite of t r i d y m i t e (sample 10047,31); that on the r i g h t contains many s l i g h t l y euhedral c r y s t a l s of s i l i c a that may be tridymite or c r i s t o b a l l t e (sample 10044,7652). From Roedder and Weiblen (1970a).

557

between these two p o s s i b i l i t i e s , and, u n f o r t u n a t e l y , problems e x i s t with both. L i t t l e evidence has been found of s u f f i c i e n t v o l a t i l e s in lunar rocks to permit the f i r s t p r o c e s s , and the second requires the u n l i k e l y t r a p p i n g of an almost pure a n o r t h i t e l i q u i d . Sobolev et a l . (1980) cooled 38 i n c l u s i o n s (1 to 6 um i n s i z e ) that had anomalously l a r g e vapor bubbles, from o l i v i n e from Luna 24, to -198°C; no evidence f o r phase change i n the bubbles was.seen, i n d i c a t i n g l i t t l e i f any H2O or CO2 could be p r e s e n t . High-Si glass inclusions in olivine. Many of the l a r g e r melt i n c l u s i o n s i n e a r l y p h e n o c r y s t i c o l i v i n e i n the lunar mare l a v a s have, i n a d d i t i o n t o e p i t a x i a l daughter c r y s t a l s of i l m e n i t e and p l a g i o c l a s e and an immiscible s u l f i d e bleb, small beads of very low-index g l a s s (n probably c II !ï -i «s h r yI.- :i tö |»i -â =s Iss ¡I LIS s. « il i SIÍ ï 3115 II ! W ! Ï! 1! a ii H ï ! i r i=i¡j r : 2^13a;¡ïî.irpyi yfiplf if 1« ¡i 3 Ifiliîliï fj ¡I M Ij M H. M i !V I îfii ! H !% 3 S

S

Í

s-

S =Si

«S

" .

Jif-l^Srir^ il : • * (i i M n I« 624

625

1 1 I I I

ì l i I I ! s:

Ii

t f î i

3

m

m

ix£ir.£ri ss

e

" E

Mìe

m

m

m

s •sii :f"aïfs£iï

S u j s ^ L-sìs3;^

ìfZW

• • • ' ' i ^ K w i

m

m

n

r

•i Pal** s o'SË

'

ir--

s

i i i. ; J i _ I i |S S í s . li " o n i * i ii ZU* 5? I ïîRlI'l ¡ H S } s «¿SI* I « I » ^ : i s i ! ôs=n:sa » . • « i s ¿s

m

lì s, ¡ r

H

M1

i « -¡sis

- « s

|=|lp.¿fll|:¡

ti

E?

íSÍÍ M f

U

»

2 ¡Ss

O O

P

H

y L t ^ i

iî

!¡

M I I i

i

S

r

" lì

I s ! ¿ s s¿ I , I S~ii*. « S -

i

1 1

ii!¿f¿isiiiii¡y=i'

^

II I ! i

1 1 626

i

i

. % .. tal 1 ?! i: Ms IW 9 I ir 111^5=1 sir 1 ÈSU&5ÉÌ if= tem'úi 3 m ss" 1° Ü 1"P ffiMï! ií tSfi*J IsIS^ssSIs P iîiïk¿i im i P Ä i uhm1 l i l i = = C Se C " .S îi «1« P WIE 1» M i n ¡ippilipppipsrrii W W ® I il ië

Pa Pe

i . i l¿ i í i is ü si i l í | I„ i"II!is11¿If i*|¡ A ! Î!;l iHp ï1«P isiîsls*í :¥V*l'PP ! ¡i i » í fi MM* f Ml «S £?ë sZ fiH i Ir lililí íes* J ^ . . S S s l 3 s-? s ¡ Sinnui"* UaUlhïn=ïîSssi i njiÄii iiifipyfeB^ ¿p^PlíiKirflpPI:^ 1 IM H

s s ^ ¿s s2=< s =5¿fi J SI^Í ¡ Il I i H i

s 627

628

629

ii É H iî y

il JKs tï Il 1IÜI i If1

i

11 i I 1 i

un

¡S SS S - S ^ ü ^ í ¡ = ë ^ £ " S

630

.VSJ

S s

Í

I

S. 'S

lì

ïs" s - "S_i I 1 1 -I Hé l Ik ñ^.kl »J ï ii5ii=-f„S ìA ìi ? » J r.£ u p il mm ì; E g fi s ï nx s " «s^-s» fi C

55

'«.s

»S-

Ï"1

ïsaja^i o

I 4->

- - n e -

HlllV.îiûïïéW. «SSSSST: J i

lïllfjlls îllll^lfâ jïfpiiliPljîi: iSiSMfijf

titilli ili

631

Ro,

li J

^ fi li IH 1 Mi : á si |°s i ¡s» Il II Ii8! ì K

... IS al J 1 I =: sr. |S8| ! Il

lis lid í r !!§ s?« i JS sis 11: zìi IF.lzì Ki ¡ m fi m mí u i." m iumt m htm

Il SiîïMrsf J p y

s« S

il là'

632

il- l i î ; fi î I l i l i } ï !i 5 R

U U

îl

! I h | | iâ h - s V. s : s hs ï 5; s * I l s ÌÌ •!• sa ¡ 1 s * i=£

I! p i 1 J ti : l i i ! U i p f e =

Î I ^ é i w . Î ;

633

fi « P .si - i -i l f i -

J

1»

« c i i :-

Ro Sa

634

635

JMW! «líK'lHi

636

ï W l L

FfffF

637

638

2