Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Electrum tarnish by sulfur : a study of an irreversible process Francis , Donald Michael 1971

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
831-UBC_1971_A6_7 F73.pdf [ 3.6MB ]
Metadata
JSON: 831-1.0053130.json
JSON-LD: 831-1.0053130-ld.json
RDF/XML (Pretty): 831-1.0053130-rdf.xml
RDF/JSON: 831-1.0053130-rdf.json
Turtle: 831-1.0053130-turtle.txt
N-Triples: 831-1.0053130-rdf-ntriples.txt
Original Record: 831-1.0053130-source.json
Full Text
831-1.0053130-fulltext.txt
Citation
831-1.0053130.ris

Full Text

ELECTRUM TARNISH BY SULFUR: A STUDY OF AN IRREVERSIBLE PROCESS. BY DONALD MICHAEL FRANCIS B.Sc. McGill University, 1968 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE In the Department of Geology We accept this thesis as conforming to the required standard: THE UNIVERSITY OF BRITISH COLUMBIA April 1971 In presenting t h i s t h e s i s in p a r t i a l f u l f i l m e n t of the requirements f o r an advanced degree at the U n i v e r s i t y of B r i t i s h Columbia, I agree that the L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r reference and study. I f u r t h e r agree t h a t permission for e x t e n s i v e copying of t h i s t h e s i s f o r s c h o l a r l y purposes may be granted by the Head of my Department or by h i s r e p r e s e n t a t i v e s . I t i s understood that copying or p u b l i c a t i o n o f t h i s t h e s i s f o r f i n a n c i a l gain s h a l l not be allowed without my w r i t t e n p e r m i s s i o n . Department The U n i v e r s i t y of B r i t i s h Columbia Vancouver 8, Canada Date ABSTRACT The tarnishing of electrum (40 atom % Ag) at 750°C and a sulfur fugacity of -2.8 to -2.9 1og1Q atmospheres is heterogenous and irreversible. This is the case even though these conditions are less than 37°C and .4 log-jQ atmospheres above the point at which tarnish is first observed on the electrum. The tarnish forms initially on the grain boundaries and lattice defects of the electrum. With time, the interiors of the electrum grains become completely masked by tarnish except for narrow untarnished borders adjacent to the grain boundaries. Microprobe analysis indicates that the untarnished borders are depleted in silver and thus cannot tarnish at 750°C. The conclusion is drawn that the tarnish forms as an external layer which destroys the original composition of the adjacent electrum by preferentially removing silver. The silver depletion near the grain boundaries is thought to be caused by rapid diffusion of silver along grain boundaries to the electrum-tarnish interface.. i . CONTENTS I. INTRODUCTION 1 II. THE ELECTRUM-TARNISH METHOD FOR THE MEASUREMENT OF SULFUR FUGACITIES 1 III THE METHOD OF STUDY 5 IV. OBSERVATIONS AND RESULTS 6 A. MICROSCOPIC EXAMINATION OF TARNISHED ELECTRUMS 6 B. ELECTRON MICROPROBE ANALYSIS 16 i . Analysis of Tarnished Regions .. .. 16 i i . Untarnished Borders of Electrum Grains 22 i i i . Analysis of Pyrrhotite 22 V. DISCUSSION . . . . 25 A. THEORETICAL CONSIDERATIONS OF UNTARNISHING 25 B. A MODEL FOR THE TARNISHING OF ELECTRUM .. 27 i . Nucleation of the Sulfide Tarnish .. .. 27 i i . Growth of the Sulfide Tarnish .. 28 C. UNTARNISHED BORDERS .. .. .. .. .. .. .. ..30 D. PSEUDO-GRID NATURE OF THE EARLY TARNISH ON THE INTERIORS OF ELECTRUM GRAINS .. 31 E. EFFECTS OF PRE-TARNISH TREATMENT 32 VI. CONCLUSIONS 33 VII. REFERENCES .. 35 i i • VIII. APPENDIX .. 37 A. RUN DATA ON PYRRHOTITE USED IN THE STUDY 38 B. ELECTRON MICROPROBE ANALYSIS OF PYRRHOTITE 40 C. ELECTRON MICROPROBE ANALYSIS OF TARNISHED ELECTRUM GRAINS 41 / i i i . LIST OF ILLUSTRATIONS i . FIGURE 1 Electrum Tarnish Curves 3 i i . PLATE I A tarnished Electrum Grain Boundary on the El ectrum used in Run 17 9 i i i PLATE II A Reflected Light Image of the Electrum used in Run 29 11 iv. PLATE III A reflected Light Image of the Electrum used in Run 30 11 v. PLATE IV A reflected Light Image of the Electrum used in Run 31 .. 13 vi. PLATE V A Reflected Light Image of the Electrum in Plate IV at Higher Magnification .. .. ..13 vii. PLATE VI A Tarnished Electrum Grain from the Electrum used in Run 33 .. 15 v i i i . FIGURE 2 Variation of Ag/Au Across Tarnished Electrum Grain (Run 30) .. .18 ix. FIGURE 3 Variation of Ag/Au Across Tarnished Electrum Grain (Run 31) 19 x. FIGURE 4 Variation of Ag/Au Across Tarnished Electrum Grain (Run 33) 20 xi FIGURE 5 Variation of Ag/Au Across Tarnished Electrum Grain (Run 32) 21 iv. xi. FIGURE 6 Composition of the Untarnished Borders (Run 31) 23 xii. FIGURE 7 Composition of the Untarnished Borders (Run 33) .. .. ..24 x i i i . FIGURE 8 The System Gold-Silver-Sulfur 26 ACKNOWLEDGMENTS Thanks must go to Professors H. J. Greenwood, E. P. Meagher and J. A. Gower who have contributed much in the way of advice and criticism to this study. The author is also indebted to J. Harakal and A. Lacis for the technical aid that they willingly rendered. Financial support for this work was supplied in part by two National Research Council Bursaries for the periods 1969-70 and 1970-71. / 1. I. INTRODUCTION Attempts by the author to measure sulfur fugacities over natural pyrrhotite using the Electrum-Tarnish Method (Barton and Toulmin, 1964) failed because none of the achieved tarnishes could be untarnished. Although the Electrum-Tarnish Method has been used successfully by its innovators, there is no report of an investigation of the kinetics and mechanism of the tarnishing process. This study is an attempt to analyse the tarnishing process and to discover why reversibility could not be achieved. II THE ELECTRUM-TARNISH METHOD FOR THE MEASUREMENT OF SULFUR  FUGACITIES The basis for the Electrum-Tarnish Method is the univariant relationship between sulfur fugacity and temperature for the assemblage: silver, silver sulfide (Ag2S tarnish) and sulfur vapour. Diluting the silver with gold reduces the activity of the silver. Therefore at any given temperature, a higher sulfur fugacity will be required to tarnish a gold-silver alloy (electrum) than will be required to tarnish pure silver. As a result, any electrum of a specified composition is characterised by a unique log sulfur fugacity versus temperature curve of tarnishing. This relationship is complicated by the fact that a significant amount of gold can enter the tarnish as Au?S. This amount 1 2. FIGURE I The curves for the electrum indicators were calculated from data published by Barton and Toulmin (1964). The curve for the pyrrhotite buffer used in these experiments was drawn from experimental data. (Appendix: Runs 17, 20, 21 and 24.) 3. ELECTRUM TARNISH N AG L 0 G S U L F U R F U G A C I T Y -21 -41 -6 -8 -10 -12 CURVES y y y y y y y y\ y y y y y y / y .1 S -35 ^ 1 y // / / / // / / ELECTRUM INDICATORS . . PYRRHOTITE BUFFER 200 400 600 TEMPERATURE 800 1000 FIGURE 1 is fixed by the composition of the electrum involved; with the ratio of silver to gold of the tarnish always being greater than that of the coexisting electrum. Figure I shows the tarnishing curves for a variety of electrum compositions calculated from data given by Barton and Toulmin (1964). A pyrrhotite is stable at a given temperature only at a specified sulfur fugacity. Thus any pyrrhotite of a specified composition is characterised by a unique, univariant temperature versus sulfur fugacity curve. The initial objective of this study was to determine points on the log sulfur fugacity versus temperature curve of a natural pyrrhotite using electrum indicators. For a measurement, pyrrhotite and an electrum of specified composition are heated in an evacuated silica glass tube. A sulfur fugacity arises which is characteristic of the temperature and the composition of the pyrrhotite. At any temperature the slope of the tarnishing curve for the electrum is less (Figure I) than the slope of the log fugacity versus temperature curve for the pyrrhotite. Thus by increasing the temperature of the assemblage a point will be reached at which the respective curves for the pyrrhotite and the electrum intersect. At this temperature, a tarnish should form on the surface of the electrum. A measure of the sulfur fugacity at this temperature can be obtained from the calculated tarnishing curve of the electrum. Performing this operation with a variety of electrum compositions gives a series of points which define the log sulfur fugacity versus temperature curve of the pyrrhotite. 5. For practical and theoretical purposes complete reversibility of the tarnishing process is required. This is necessary not only to pinpoint the temperature at which tarnishing occurs by a series of telescoping temperature brackets, but also to establish that the initial tarnishing could have been a stable reaction. Equilibrium requires that the tarnish have a silver-gold ratio that is stable with the coexisting electrum. In addition, the tarnish layer must be thin enough that continuous equilibration can take place between the electrum, tarnish and sulfur atmosphere. If the ratio of silver to gold of the tarnish becomes greater than the ratio which is specified by the composition of the coexisting electrum for equil-ibrium conditions; then a lower temperature (and thus sulfur fugacity) would be required to untarnish the electrum than was required to tarnish the electrum. This would cause the tarnishing process to be irreversible. In the experiments reported here, no electrum once tarnished could be untarnished. III. THE METHOD OF STUDY A special method of preparing the electrum was used to facilitate later microprobe and reflected light study. Barton and Toulmin's (1964) procedure involved the use of electrum chips simply cut off a stock rod. In the experiments reported here, the electrum chips were flattened to round disks of a thickness slightly less than one millimeter in a cold press. One surface of each disk was polished with tin oxide abrasive. The natural pyrrhotite used in this study was obtained from the 6. Bluebell Mine in southern British Columbia. The °f this pyrrhotite was determined to be 2.0666 ± .0002 A° on a Philips Diffractometer using a Ni filter with Cu Ko<radiation. Three oscillations were made at 1/8 of a degree 29 per minute using potassium bromide as an internal standard. Combining this with data given by Barton and Toulmin (1964) the composition of the pyrrhotite was calculated to be .4736 ± .0002 atom fraction iron. Runs 17, 20,. 21 and 24 (appendix) determined the log sulfur fugacity versus temperature curve of the pyrrhotite (Figure I). However, equilibrium in these runs could not be proved as the tarnishes achieved could not be reversed. "Five identical disks of electrum of composition .4000 atom fraction silver were tarnished for successively longer periods of time at 748°C to 753°C using the above pyrrhotite. It had formerly been determined that this electrum would tarnish between 713°C and 735°C (Runs 17 and 20, appendix). These tarnished disks of electrum were then studied micros-copically in reflected light and analysed with the JXA-3A Electron Microprobe located in the Department of Metallurgy at the University of British Columbia. IV OBSERVATIONS AND RESULTS A. MICROSCOPIC. EXAMINATION OF TARNISHED ELECTRUMS Plates II through VI show the development of tarnish on the electrum with increasing time. 7. The polycrystalline nature of the electrum (not seen in the untarnished specimens) was brought out by preferential tarnishing of the grain bound-aries. These grains ranged in size from .1 mm. to .4 mm. The width of the tarnish on the grain boundaries was found to be roughly proportional to the time of exposure to tarnishing (Table I). Soon after the grain boundaries began to tarnish, a light tarnish formed on the interiors of the electrum grains as irregular blotches and discontinuous rectangular networks (Plates II and III). With time, the tarnish spread and darkened, completely masking the interiors of the grains. Table I Duration of Width of Grain Run No. Tarnishing Boundary Tarnish (hrs.) (microns) 29 .5 3.75 30 1 5-5.5 31 4 8.75 - 12.75 33 8.17 16.25 - 21.25 32 22.75 23.75 - 25 The degree of tarnishing on the interior of any one grain was not diagnostic of the duration of the experiment. However, the percentage of grains exhibiting extensive tarnishing of their interiors was roughly proportional to the time of exposure to tarnishing. Even following 24 hours of tarnishing the electrum grains were never completely covered with tarnish. There always remained a border of untarnished electrum between the tarnished grain bound-aries and the tarnish on the grain interiors (Plates I, IV, V and VI). 8. PLATE I. A tarnished electrum grain boundary on the electrum used iri Run 17. - electron microprobe absorbed electron image. - 25 KV - magnification: 1020 X - field width: 87 microns . - light regions on photo are tarnished and enriched in silver - dark regions on photo are untarnished electrum which is depleted in silver. 9. 10. PLATE II. A reflected light image of the electrum used in Run 29. - run duration: .5 hrs. - field width: 1.25 mm. - grey regions are tarnish PLATE III A reflected light image of the electrum used in Run 30. - run duration: 1 hr. - field width: 1.25 mm. - grey regions are tarnish. 1 1 . 12. PLATE IV A reflected light image of the electrum used in Run 31. - run duration: 4 hrs. - field width: 1.25 mm. - grey regions are tarnish PLATE V A reflected light image of the electrum in Plate IV under a higher magnification. - run duration: 4 hrs. - field width: .6 mm. - grey regions are tarnish 14. PLATE VI. A tarnished electrum grain used in Run 33. Line indicates traverse analysed by the microprobe. - electron microprobe absorbed electron image - run duration: 8 hrs. - 40 KV - magnification: 646 x - field width: 136 microns - light regions are tarnished and enriched in silver - dark regions are untarnished electrum which is depleted in silver. J 15. 16. B. ELECTRON MICROPROBE ANALYSIS The tarnished electrums were analysed by a JXA-3A Electron Microprobe at a voltage of 40 KV. With this instrument the surface to be analysed is perpendicular to the electron beam and the resultant x-rays are read at a take off angle of 20°. Gold and silver counts were taken simultane-ously so that silver-gold ratios could be used, minimizing error due to drift in the instrument. Gold and silver counts from pure element standards, electrum of .4000 atom fraction silver, and background were used to construct calibration curves for each set of readings. The series of readings taken over untarnished electrum before each analysis in the appendix show that the electrum was homogeneous to the electron probe. i . Analysis of Tarnished Regions Figures 2 through 5 show the results of microprobe analysis of tarnished electrum grains. Traverses were run across grains from electrums that had experienced successively longer durations of tarnishing. Regions of electrum covered by tarnish always yielded significantly higher silver-gold ratios than were obtained from untarnished electrum of the same original composition. The tarnished grain boundaries showed the greatest increase in silver-gold ratios. The magnitude of this increase was found to be roughly proportional to the exposure time to tarnishing (Table II). 17. The tarnished interiors of electrum grains showed similar increases in their silver-gold ratios, though not of the same magnitude as the increases found over the tarnished grain boundaries. There was a direct correlation between the degree of tarnish development, as visually - ; Table II ^ ^ Ag/Au Tarnished Grain Duration of Boundary Run No. T a™shing Ag/Au Untarnished electrum 29 .5 1.25 - 2.05 30 1 1.96 - 4.35 31 4 4.45 - 20.16 33 8 16.8 - 39 estimated, and the magnitude of the silver-gold ratio. Tarnishes that appeared light and patchy gave sporadic silver-gold ratios which were two to three times the silver-gold ratio obtained over untarnished electrum. Dark, heavy appearing tarnishes gave uniform silver-gold ratios which were four to five times the silver-gold ratios obtained over untarnished electrum (compare Figures 2, 3 and 4 with Plates III, IV and VI respectively). Weight percentages of gold and silver were not calculated for tarnished regions because the effects of the sulfur could not be accounted for, and because microprobe readings are thought to be an average of the compositions of the tarnish layer and the underlying electrum. VARIATION OF ACROSS TARNISHED ELECTRUM GRAIN R U N 3 0 1 H 0 U R AG/ 'Mi L 0 G .1 S C A L E < 2 cc o X Z —. Q O CC O O CQ U N T A R N I S H E D E L E C T R U M .01 25 50 MICRONS 75 100 19. A VARIATION OF A G / 'AU ACROSS AG/ 'AU L 0 G S C A L E J .1 \ \ TARNISHED ELECTRUM GRAIN RUN 31 I / I I I >-or < ? § rr o O CO cc UJ Q o m 4 HOURS A \ w X CO < N ' co z si a cc z O 3) CO I \ >-cc < < 3 cc o e> co UNTARNISHED ELECTRUM .01 25 50 75 MICRONS FIGURE 3 'AU A / \ / \ / \ VARIATION OF ftj ACROSS TARNISHED ELECTRUM GRAIN RUN 33 8 HOURS \ \ L 0 G S C A .ir L E y \ \ f \ \ UNTARNISHED ELECTRUM -A .01 25 50 75 MICRONS 100 125 150 21. VARIATION OF /fa ACROSS TARNISHED ELECTRUM GRAIN cr UJ a cr o m co re a cr O oo >-cr < < => cr o O co UNTARNISHED ELECTRUM .01 25 MICRONS 50 75 FIGURE 5 22. i i . Untarnished Borders of Electrum Grains Figures 6 and 7 show the detailed variation in the weight percentages of gold and silver across the untarnished borders within tarnished electrum grains. The weight percentages were calculated from gold and silver counts (appendix) obtained from microprobe analyses across tarnished electrum grains. Grains from electrums that had experienced less than four hours of tarnishing showed no evidence of silver depletion in the untarnished borders (Figure 2). All grains analysed from electrums that had under-gone four or more hours of tarnishing had untarnished borders which were enriched 3-4 weight percent in gold and depleted 1.5 to 3 weight percent in silver as compared to untarnished electrum. More confidence should be placed in the gold figure because of the large absorption effect of the gold on the silver readings. i i i . Analysis of the Pyrrhotite The pyrrhotite used in these experiments was analysed for iron and sulfur by the electron microprobe. The iron-sulfur ratios that were obtained on a random walk across a 3/4 inch specimen are listed in the appendix. These figures indicate that the pyrrhotite was homogeneous to the microprobe. The microprobe could detect no trace of antimony or bismuth. A qualitative analysis of the pyrrhotite was run with a Philips X-Ray Flourescence Spectrometer in the Department of Geology at the COMPOSITION OF THE 79 UNTARNISHED BORDERS MICRONS 79 W E I G H T P E R C E N T G 0 L D 77 75 73 71 69 BORDER 1 \ \ /—/ \A I 67 L / / / !_/ I i i I I ! I I COMPOSITION OF THE UNTARNISHED BORDERS RUN 33 8 HOURS UNTARNISHED ELECTRUM BORDER \ \\ \ \ \ \ \ •\AG \ AU \ AU \ AG f I I I / / / \ \ \ \ / \ \ 22 24 26 28 30 32 34 W E I G H T P E R C E N T S I L V E R 10 MICRONS 10 25. University of British Columbia. Traces of lead, zinc and copper were detected. No trace of arsenic could be found. V. DISCUSSION A. THEORETICAL CONSIDERATIONS OF UNTARNISHING The composition of a tarnish which forms on an electrum is characteristic of the composition of that electrum. The gold-silver ratio of the tarnish which formed in the experiments reported here is not accurately known because microprobe analyses represent an average of the compositions of the tarnish and the underlying electrum. However, it has been observed that tarnished regions are enriched in silver with respect to untarnished electrum. Therefore, the closest approximation to the composition of the tarnish is given by the most silver rich analyses. Point "B" (10.1 weight percent gold) in Figure 8 represents such an analysis taken from a tarnished grain boundary on the electrum used in Run 33 (appendix). Figure 8a is a schematic diagram of the system gold-silver-sulfur for conditions such that electrum of composition 40 mol. percent silver ("A") will react with sulfur to form a tarnish of composition 10.1 weight percent gold ("B"). Because silver is preferentially taken up by the tarnish, the composition of the electrum surface will become more gold rich and shift from point "A" to some point "D" along the gold-silver join (Figure 8a). 26. THE SYSTEM GOLD-SILVER-SULFUR FIGURE 8 27. / The procedure to effect untarnishing is to lower the temperature (and thus the sulfur fugacity) below the point at which tarnish first began to form on the electrum. At such a temperature the phase relations for the system gold-silver-sulfur would be given in Figure 8b. The original tarnish ("B") would break down to form a small amount of electrum of composition "E", sulfur, and a relatively large amount of tarnish of composition "C". The only way that tarnish will completely disappear is for the newly formed electrum of composition "E" to equilibrate with the electrum surface (composition "D"). The intermediate electrum so formed would react with the tarnish to give sulfur plus more electrum of composition "E". This latest electrum could then equilibrate with the surface electrum returning it to its original composition ("A", Figure 8b). If the tarnish forms as an external layer on the surface of the electrum, then this complex equilibrium process may be very slow. As a result the tarnish would persist metastably. B. A MODEL FOR THE TARNISHING OF ELECTRUM i . Nucleation of the Sulfide Tarnish Little has been published about the tarnishing of metals by sulfur. However, the theory of the oxidation of metals has received considerable attention (Bernard, 1962; Kofstad, 1966) and this process provides a useful analogy. When a metal is brought in contact with an oxygen atmosphere, a thin layer of adsorbed oxygen is formed on the metal surface. Under conditions 28. of high temperature and low oxygen fugacity metal oxide nucleii form on grain boundaries and on surface irregularities such as point defects and dislocation walls within the metal grains. It is thought (Kofstad, 1966) that the formation of these oxide nucleii results in a surface diffusion of the adsorbed oxygen towards the nucleii. This results in a depletion of adsorbed oxygen in the areas surrounding the nucleii and so prevents the formation of additional metal oxide nucleii. The mechanism described above can explain the observations made during the tarnishing of the electrums in the experiments reported here. The sulfide tarnish occurred ffrst on the grain boundaries, followed by the formation of grid like tarnish patterns on the interiors of electrum grains probably related to dislocation walls (Section D). The fact that c a border adjacent to the tarnished grain boundaries never formed any tarnished nucleii may be the result of two phenomena. A migration of adsorbed sulfur to the tarnish nucleii on the grain boundaries might have depleted the border region of available sulfur. In addition, it is possible that any lattice defects in the border region which could have served as nucleation sites were adsorbed by the grain boundaries during heating. It is suggested here that the nucleation of tarnish on electrum involves the same mechanism that is held to be operative in the oxidation of metals. i i Growth of the Sulfide Tarnish It is proposed that two mechanisms were involved in the growth of the sulfide tarnish. Lateral growth is thought to have been effected by the reaction of silver with adsorbed sulfur at the edges of the tarnish nucleii. Vertical growth is thought to have occurred by the reaction of the sulfur atmosphere with silver atoms that had diffused through the tarnish layer from the underlying electrum. Lateral growth seems to have been more important in the growth of the tarnish on the interiors of the electrum grains than on the grain boundaries. The tarnish nucleii on the grain interiors quickly coalesced and formed a continuous tarnish layer that grew outwards towards the grain boundaries. The tarnish on the grain boundaries however grew to a maximum measured width of only 25 microns (Table I). The maximum Ag/Au ratio over tarnish on a grain interior was about five times the Ag/Au ratio for untarnished electrum. Over the tarnished grain boundaries, however, the Ag/Au ratio rose to a maximum of 39 times the Ag/Au ratio over untarnished electrum (Table II). If the increase in the Ag/Au ratio is taken as a measure of the thickness of the tarnish, then it must be concluded that the vertical growth mechanism was more important in the growth of the grain boundary tarnish than in the growth of the tarnish on the interiors of the electrum grains. The different growth characteristics of the two types of tarnish can be explained by the manner in which silver is supplied to the electrum surface. At 700°C, the self diffusion coefficient of silver along grain boundaries is TO4 to 106 times greater than the coefficient for volume self diffusion (Shewmon 1963). Therefore one would expect a localized influx of silver atoms into the tarnish just above the grain boundaries 30. "' / -• • ; " of the electrum. This would produce a chemical gradient in the tarnish layer, forcing silver to diffuse to the surface of the tarnish. There-fore more silver might be available on top of the tarnish layer than at the edges of the tarnish. Thus the grain boundary tarnish would grow preferentially in the vertical direction. In the grain interiors, however, most of the silver would be supplied to the surface by volume diffusion. After the formation of tarnish over a particular spot the underlying electrum would be depleted in silver and volume diffusion would decline. This would not be the case at the edges of the tarnish. Therefore i t would be expected that less silver would be available on the tarnish surface than at the tarnish edges. Thus lateral growth would be the dominant growth mechanism. C. UNTARNISHED BORDERS Two questions arise concerning the untarnished borders that have been observed in tarnished electrum grains. First, why have these borders remained untarnished while all other parts of the electrum grains have tarnished; and second, what is unique about their position just inside the tarnished grain boundaries? The answer to the first question lies in the anomalous composition of the electrum in the untarnished borders. It has been previously shown that after a run duration of more than four hours, the gold content of the untarnished borders was at least three weight percent more gold rich than the untarnished electrum. This corresponds to an electrum of 31. composition .3625 atom fraction silver. Such an electrum would tarnish between 755°C and 780°C with the pyrrhotite charge used in the experiments reported here (Figure I). However, the above electrums were subjected to only 750°C. Therefore these gold enriched borders could not have tarnished. It has been noted that the untarnished borders always occurred just inside the tarnished grain boundaries. It seems reasonable to suppose that rapid tarnishing on the electrum grain boundaries could cause signif-i icant depletions of silver in the electrum below. This would set up a ! diffusion of silver from adjacent electrum, leaving gold enriched borders on both sides of the grain boundaries. It this were the case, then the tarnish on the grain interior would grow outwards towards the grain boundary until it encountered electrum which was too gold rich to tarnish at 750°C. The result would be the untarnished border that has been observed in tarnished electrum grains. i -! \ I i D. PSEUDO-GRID NATURE OF THE EARLY TARNISH ON THE GRAIN INTERIORS 5 11 The density of edge dislocations increases from 10 /cm to 10 /cm when metals are plastically deformed (Friedel, 1964). This results in an increase in the stored energy within the crystal lattice. Annealing deformed metals at high temperatures enables these dislocations to migrate, seeking a lower energy configuration. During this treatment many of the dislocations are destroyed. Those remaining tend to align 32. parallel to each other, forming planes or walls of dislocations. These walls of dislocations, called sub-boundaries, orient themselves so as to divide the metal grains into many sub-grains. This process is termed "polygonization". These sub-grains range in size from 1-10 microns and differ from each other by slight misorientations of their crystal lattices. Self diffusion occurs preferentially along the length of edge dislo-cations as compared to volume diffusion within the grains themselves. Thus a sub-boundary, being a row of dislocations, is a path of rapid self diffusion in the direction parallel to the dislocations. The electrums is this study were first subjected to plastic deformation (flattening) and then tarnished at a temperature near 750°C. Therefore, polygonization would be an expected phenomena. The nearly rectangular grid nature of the original tarnish on the grain interiors probably marks sub-boundaries formed by this polygonization process. The first sulfide nucleii formed on the intersections of these dislocations walls with the electrum surface because of the resultant surface irregularities and the ready availability of silver diffusing up the edge dislocations. E. EFFECTS OF PRE-TARNISH TREATMENT The effects of the cold working (flattening) of the electrum chips on the tarnishing mechanism must be considered. The fact that the grain boundaries of the electrum were so heavily tarnished could indicate that they were areas which had stored strain energy. However, Plate I, which is a photograph of a tarnished electrum chip that had not experienced 33. - • • J ' • flattening, shows all the tarnishing features that have been observed on flattened electrum chips. Of course, even the chipping of an electrum chip from the stock rod is a form of cold working. However, grain boundaries are characterise cly regions where gas-solid surface reactions begin, whether the material considered has been Cold worked or not. VI CONCLUSIONS Successful application of the Electrum-Tarnish Method requires that the tarnishing process be reversible. For a tarnishing process of a binary alloy to preceed in equilibrium it should be an internal process so that metal and tarnish will be in intimate contact (Kofstad, 1966). The formation of an external "scale" of tarnish results in the removal of the tarnish from the alloy, discouraging equilibration. The large increase in the silver-gold ratios over tarnished electrum indicates that the sulfide formed an external layer on the electrum, with silver being preferentially drawn out of the electrum. The original composition of the electrum near the tarnish-metal interface was destroyed. Following Barton and Toulmin's (1964) procedure of viewing unflattened electrum chips in silica glass capsules with a binocular microscope, the tarnishing of an electrum could not be detected unless the electrum had experienced at least four hours of tarnishing. Data presented in this paper indicate that after a run duration of four hours, the tarnishing was irreversible. 34. Barton and Toulmin (1964) have used the Electrum-Tarnish Method with success. Impurities in the natural pyrrhotite used in the experiments reported here might have affected the tarnishing process. However, no volitile elements besides sulfur were detected in the pyrrhotite. Barton and Toulmin (1964) must have been able to detect the formation of tarnish early enough to reverse i t . It the experiments reported here, the tarnish was invisible until the point of reversibility was passed. / 35. VII. REFERENCES i . Barton, P.B. and Skinner, B.; 1967; "Sulfide Mineral Stabilities"; Geochemistry of Hydrothermal Ore Deposits; Holt, _ Rinehart, and Winston, Inc.; 236 - 333. i i . Barton, P.B. and Toulmin, P.; 1964; "The Electrum-Tarnish Method for the Determination of the Fugacity of Sulfur in Laboratory Sulfide Sustems"; Geochemica and Cosmochemica Acta; vol 28, 619 - 640. i i i . ; 1966; "Phase Relations Involving Sphalerite"; Economic Geology; vol. 61, 816 - 849. iv. Benard, J.; 1962; Oxydation Des Metaux; Gauthier-Viliars et C i e Editeur-Imprimer. v. Birks, L.S.; 1963; Electron Probe Microanalysis; Chemical Analysis vol. 17; Interscience. vi. Bokshtein, S.Z.; 1965; Diffusion Processes, Structure, and Properties of Metals; Consultants Bureau; Russian trans. vii . Friedel; 1964; Dislocations; Peragon Press. 36. ' / - • • I ' v111. Hirth and Lothe; 1968; Theory of Dislocations; McGraw Hill Book Co. Inc. ix. Kofstad, Per; 1966; High Temperature Oxidation of Metals; J. Wiley and Sons. x. Newkirk and Werwick; 1962; Direct Observations of Imperfections in Crystals; Interscience. xi. Shewmon, P.G.; 1963; Diffusion in Solids; McGraw Hill Book Co. Inc. xii. Toulmin, P. and Barton, P.B.; 1964; "A Thermodynamic Study of Pyrite and Pyrrhotite"; Geochemica and Cosmochemica Acta; vol. 28, 641 - 671. x i i i . Wager, C.; 1953; "Investigations on Silver Sulfide"; Journal of Chemical Physics; vol. 21, 1819 - 1827. xiv. White, J.L.; orr, P.L.; and Hultgren; 1957; "The Thermodynamic Properties of Silver-Gold Alloys"; Acta Meta.; vol. 5. 747 - 760. 37. i j VIII. APPENDIX i : • i 38. A. RUN DATA ON THE PYRRHOTITE USED IN THIS STUDY. Runs 17, 20, 21 and 24 were held at the temperatures indicated below for eight to twelve hours and examined two to three times during that time interval. If no tarnish was observed (binocular microscope) on the electrum of a run at the end of such a period, the temperature was increased and the procedure repeated. Once tarnishing was observed, the temperature of a run was reduced below the highest temperature for which the electrum remained untarnished. The run would then be held at such a temperature for up to five days in an attempt to untarnish its electrum. i . Run 17 electrum indicator; Nag = .4000, unflattened. charge: BB#2 pyrrhotite. temperature (°C) tarnish no tarnish 562 590 608 617 634 650 660 665 680 700 713 735 762 727 x-x X X X X X X X X X X X X X 39. 1 i i . Run 20 - electrum indicator: Nag = .4000, unflattened. - charge: BB#2 pyrrhotite. temperature (°C) 590 612 632 648 665 736 705 i i i . Run 21 - electrum indicator: Nag = .3000, unflattened. - charge: BB#2 pyrrhotite. temperature tarnish no tarnish (°C) 815 x 825 x 832 x 800 x tarnish x x no tarnish x x x x x iv. Run 24 - electrum indicator: Nag = .5000, unflattened. - charge: BB#2 pyrrhotite. temperature tarnish no tarnish (°C) 515 x 528 x 486 x (5 days) 40. v. Runs 29 - 33 - charge: BB#2 pyrrhotite Run electrum temperature duration indicator (°C) (hrs.) 29 .4000 750 - 753 .5 30 ,4000 753 1 31 .4000 748 - 749 4 32 .4000 748 22.75 33 .4000 748 - 753 8.17 B. ELECTRUM MICROPROBE ANALYSIS OF PYRRHOTITE 25 KV S counts Iron counts Sulfur counts pe COunts 104526 21691 .2075 104218 21991 .2110 103611 21736 .2097 104084 22173 .2130 101668 20322 .1999 103145 21983 .2131 104497 21187 .2028 101340 25183 .2485 102687 21796 .2123 104306 22477 .2155 103786 21773 .2098 103740 21358 .2059 104305 21541 .2065 104120 21409 .2056 103905 21741 .2092 103830 21563 .2076 98469 20633 .2095 102485 25097 .2449 Sb count nil Bi count nil 41. C. ELECTRON MICROPROBE ANALYSIS OF TARNISHED ELECTRUM GRAINS The variation in the analyses for untarnished electrums associated with the following runs results from the fact that runs were analysed on different days. It should be noted that similar discrepancies appear in the analyses of the pure element standards. Because of this lack of reproducibility in microprobe counts, analyses of tarnished electrum chips could not be meaningfully compared with analyses of the same chip before it was tarnished. Therefore each analysis of a tarnished electrum is compared to a coeval analysis of a separate untarnished electrum chip. i . Run 30 Au counts Ag counts Ag counts Au counts untarnished electrum: 32562 32622 32580 32378 32843 32922 32694 32326 32781 32828 2870 2828 2840 2856 3010 2881 3079 2973 2885 2810 .0881 .0867 .0872 .0882 .0916 .0875 .0942 .0920 .0880 .0856 average: 32658.4 2903 .0889 tarnished electrum: 1.25 micron steps untarnished border 32390 32253 32265 32180 33231 33549 30066 3450 3967 4047 3948 3675 3449 3897 .1065 .123 .1254 .1227 .1106 .1028 .1296 42. Run 30 (cont.) Au counts Ag counts Ag counts Au counts grain boundary untarnished border tarnish partial tarnish tarnish 5 micron steps 1.25 micron steps tarnish untarnished border 26746 6218 .2325 27680 7109 .2568 30484 5598 .1836 31154 5366 .1722 32516 4222 .1298 32662 3433 .1051 32214 3620 .1124 32240 3520 .10918 32067 3552 .1108 32102 3595 .112 32178 3618 .1124 30288 5113 .1688 29611 5906 .1995 28679 6451 .2249 29118 6329 .2178 29538 5575 .1887 30088 4671 .1552 30968 4059 .1311 32073 3825 .1193 33036 3300 .0999 33127 3062 .0924 30063 3829 .1274 29716 5070 .1706 31724 3806 .1200 32399 •4024 .1242 31262 4784 .1530 30648 5783 .1887 29886 5980 .2000 29288 5280 .1803 29426 6063 .2060 31204 5118 .1640 31376 4012 .1279 31054 5566 .1792 31675 5117 .1615 29611 6887 .2326 28691 7062 .2461 29089 5870 .2018 31586 5099 .1614 31760 4353 .1371 31889 4107 . .1287 32025 3690 .1152 33157 3316 .1000 33341 3346 .1004 43. Run 30 (cont.) Au counts Ag counts untarnished border grain boundary untarnished border 33101 32788 33009 32684 32559 33429 34247 32485 22940 23608 27675 29661 30541 31371 3049 3101 3227 3269 3447 3435 3174 3614 8920 7594 5507 4136 3354 3245 Ag counts Au counts .0929 .0946 .0978 .1000 .1059 .1028 .0926 .1113 .3888 .3217 .1990 .1394 .1098 .1034 element standards; average: background 45679 45423 45803 45805 46321 45627 45776.3 553 556 553 523 31181 30745 30809 30974 31328 31242 31397 31089 31134.4 301 299 284 292 Run 31 untarnished electrum: 32479 32101 32378 32645 32097 32088 32483 32369 32488 32641 32552 32263 3071 3143 3042 3178 3006 3048 2979 2922 2990 3086 3062 2950 .0946 .0979 .0940 .0974 .0937 .0950 .0917 .0903 .0920 .0945 .0941 .0914 average: 32382 3039.75 .0939 44. Run 31 (cont.) tarnished electrum: 1.25 micron steps grain boundary untarnished border light tarnish Au counts Ag counts Ag counts Au counts 31727 4249 .1339 31072 4090 .1316 29582 4870 .1647 25744 7039 .2734 20495 10414 .5081 15371 11776 .7661 10955 13325 1.216 10725 14805 1.380 16293 13797 .8468 24036 11422 .4752 31317 6288 .2008 34458 3835 .1113 34213 3351 .0979 33664 3338 .0992 33042 3315 .1003 32769 3304 .1008 32414 3581 .1105 32500 3727 .1147 32252 3865 .1198 32040 4272 .1333 31633 4346 .1374 31861 4163 .1307 32313 4065 .1258 32262 4216 .1307 31834 4338 .1363 32575 4087 .1254 32197 4207 .1307 32323 4184 .1294 31652 4546 .1436 31364 4743 .1512 31247 4813 .1540 31752 4454 .1402 31949 4199 .1314 30431 5243 .1723 30121 5244 .1741 31097 5052 .1625 29162 6885 .2361 28858 7000 .2426 29881 6280 .2102 30067 5912 .1966 29594 5805 .1962 30864 5268 .1707 30689 5303 .1727 31453 5184 .1648 32279 4343 .1345 32320 4123 .1276 45. Run 31 (cont.) Au count Ag count Ag count Au count untarnished border grain boundary untarnished border untarnished electrum: average: element standards: background: 33586 3516 .1047 33872 2812 .0830 34260 2666 .0778 34220 2646 .0773 33878 2670 .0788 33524 2886 .0861 32604 3387 .1039 30926 4129 .1335 26642 — 6777 .2544 20887 9362 .4482 16583 11314 .6823 13887 11746 .8458 15146 10594 .6995 21766 8223 .3778 25940 6674 .2573 32238 5061 .1570 34001 3848 .1132 34126 3022 .0886 33904 2837 .0837 32979 3076 .0933 31686 4314 .1361 30324 5424 .1789 32562 2870 .0881 32622 2828 .0867 32580 2840 .0872 32378 2856 .0882 32843 3010 .0916 32922 2881 . .0875 32694 3079 .0942 32326 2973 .0920 32781 2885 .0880 32826 2810 .0856 32653.4 2903.2 .0889 see Run 30 see Run 30 46. i i i . Run 33 Au count Ag count Ag count Au count untarnished electrum: 27348 3383 .1237 — 27361 3247 .1187 27239 3283 .1205 27538 3276 .119 27541 3260 .1184 27486 3191 .1161 27566 3261 .1183 27601 3257 .1180 27164 3276 .1206 27242 3270 .1200 27131 3310 .1220 average: 27371.91 3274.0 .1196 tarnished electrum 25705 5152 .2004 1.25 micron steps 25588 4355 .1702 26305 4482 .1704 26828 4022 .1499 27625 3661 .1325 28270 3336 .1180 28878 2657 .0920 28920 2580 .0892 29206 2557 / . .0876 29081 2514 .0864 untarnished border 28973 2560 .0884 28398 2560 .0901 28045 2656 .0947 27562 2687 .0975 26949 2924 .1085 25952 3377 .1301 grain boundary 24204 4036 .1667 20878 6690 .3204 15643 9771 .6246 10994 12458 1.133 7720 12785 1.656 6497 12917 1.988 7930 12787 1.6125 10663 12089 1.1337 14166 12827 .9087 19253 11526 .5986 23206 7661 .3301 26678 4930 .1848 28640 3143 . .1098 untarnished border 28728 2888 .1001 28518 2921 .1024 27748 3254 .1173 47. Run 33 (cont.) tarnish 5 micron steps 1.25 micron steps tarnish untarnished border grain boundary Au count Ag count Ag count Au count 26778 4094 .1529 25561 5151 .2015 24940 5666 .2272 24065 5833 .2424 23674 6250 .2640 23313 6905 .2961 22329 7560 .3386 20784 8101 .38977 20553 7980 .3882 21722 8498 .3012 21327 9088 .4261 19760 8599 .4352 19451 9690 .4981 19123 8867 .4636 18821 8928 .4744 19932 8492 .4260 22335 8250 .3694 25211 5294 .20998 25385 4916 .1937 26523 4303 .1622 26571 4198 .1580 26647 4482 .1682 26661 4372 .1640 27468 3617 .1317 28559 2912 .102 28369 2734 .0964 28540 2811 .0985 28445 3122 .1098 27510 3318 .1206 26920 3500 .1300 25990 4124 .1587 23962 5249 .2191 18882 8172 .4328 13969 10381 .7431 10354 11805 1.140 7835 13269 1.694 5933 14148 2.385 4829 13934 2.887 4359 13496 3.096 4700 13473 2.8665 5414 12287 2.269 6554 11309 1.726 7639 10733 1.405 8871 11165 1.259 10281 11555 1.124 48. Run 33 (cont.) grain boundary untarnished border Au count Ag count Ag count Au count 11847 11276 .9518 13615 11075 .8134 15603 10357 .6638 17099 10145 .5933 18989 9387 .4943 20675 8284 .4007 21738 7707 .3545 23018 7043 .3060 24192 6468 .2674 24757 5456 .2203 25369 5326 .2099 26146 4692 .1794 26843 3745 .1395 26876 3314 .1233 27109 3336 .1231 element standards: average: background: 39500 28491 39286 28355 39589 28058 39650 28139 39938 28483 40023 28417 39684.3 28323.8 633 284 588 277 616 231 581 258 iv. Run 32 untarnished electrum: 27110 3086 .1138 26722 3088 .1156 26683 3007 .1127 26746 3181 .1189 26579 3245 .1221 26520 3011 .1135 26404 3114 .1179 26699 3024 .1133 26866 3072 .1143 average: 26703.2 3092 .1158 49. 32 (cont.) tarnished electrum; 1.25 micron steps grain boundary untarnished border tarnish untarnished border grain boundary Au count Ag count Ag count Au count 14101 11820 .8382 19244 9320 .4843 23299 5450 .2339 26694 3460 .1296 27738 2963 .1068 28200 2652 .0940 28273 2531 .0895 28448 2515 .0884 28415 2900 .1021 26474 5061 .1912 23821 6530 .2741 21864 7083 .3240 22872 6800 .2973 20765 7937 .3822 19210 8810 .4586 19044 9176 .4818 18147 7103 .3914 17616 7099 .4030 17432 8501 .4877 18118 9149 .5050 17370 8628 .4967 17224 7969 .4627 17974 8917 .4961 18430 8204 .4451 20235 9388 .4639 21075 9511 .4513 21635 7789 .3600 21661 8855 .4089 20477 8400 .4102 20775 8218 .3956 21434 6839 .3191 24208 5831 .2409 23814 6402 .2688 26643 4355 .1635 27458 3380 .1231 27428 3267 .1191 27691 2990 .1080 27840 2800 .1006 27799 2805 .1009 27992 2814 .1005 25458 5644 .2217 22881 7163 .3131 20393 9359 .4589 16943 11376 .6714 50. Run 32 (cont.) Au counts Ag counts Ag counts Au counts grain boundary untarnished electrum: average: 12848 10735 .8355 11284 12117 1.0738 11376 12455 1.0948 12610 11521 .9136 14987 9845 .6569 15989 12399 .7755 17834 10689 .5994 18428 9690 .5258 17660 11094 .6282 18849 8588 .4556 20704 8993 .4344 24093 5753 .2387 25443 3885 .1527 25901 3593 .1387 26151 3504 .1340 26252 3152 .1201 26300 3224 .1226 26427 3041 .1151 27455 3137 .1143 27949 3111 .1113 27799 3243 .1166 27734.3 3163.7 .11407 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0053130/manifest

Comment

Related Items