THE FLUID INCLUSION TECHNIQUE

IN EXPLORATION FOR GOLD MINERALISATION

K. BURLINSON February 1982

BURLINSON GEOCHEMICAL SERVlCES PTY. LTD.

PREFACE - April 2013

This old report from 1982 was written at a very early stage in the decrepitation research when an understanding of the decrepitation response of gas-rich CO₂ inclusions was only just being recognized and long before a full appreciation of its significance in exploration for gold. The current full understanding of the importance of CO₂ rich inclusions is detailed elsewhere on this website.

	Page #
Conclusions
Introduction	1
Literature survey
1) Temperature of formation	2
2) Boiling	3
3) Salinity	4
4) Carbon Dioxide	4
5) Fluid Chemistry	5
6) Gold Transport & Precipitation Mechanisms	5
7) Zonal Patterns	6
8) Vein Characterization	6
9) Literature Summary	7
Relevance of Decrepitation analysis to Exploration	8
Decrepitation studies at Australian gold deposits	9
References	11

CONCLUSIONS

A review of the literature on fluid inclusion (FI) studies at gold deposits shows that there is no single temperature. salinity or fluid condition which is universally characteristic of gold mineralisation. However. although no systematic relation between mineralisation and FI data can be discerned on a regional scale, there is considerable evidence in both the western and Russian literature that such relationships do occur on a local scale. The Russian literature in particular shows many examples of the use of FI decrepitation data to define aureoles both around deposits and across entire mining districts.

FI data is used in exploration for deposits in Russia but has received scant attention in the Western world. Hence. because of the difficulty of accessing the Russian data, there is no satisfactory documentation available tn the western world either to support or discredit the usage of Fis as an exploration tool.

Some preliminary work towards filling this void has been undertaken on several Australian gold mining areas.

This has given some surprising results in suggesting that at least some (but not all) gold deposits show a characteristic decrepigram not observed either on barren quartz or at deposits of base metals.

While this preliminary work seems to substantiate the local effects described in the literature, it also implies a regional relationship between gold and FI decrepitation which has not previously been documented.

It is concluded that the use of FIs and the decrepitation method in particular, warrants more serious consideration as a tool for gold exploration than it is currently being given.

INTRODUCTION

This discussion is specifically directed towards aspects of fluid inclusion (FI) data considered to be of relevance to exploration for gold deposits. A previous report (6) has explained the general concept of FI techniques (particularly decrepitation) and their application in exploration for a wide variety of ore deposits including gold. The reader is referred either to that report or to Roedder (32) for a general overview of FI techniques and decrepitometry in exploration.

This application note includes both a literature survey of the available data on FIs in gold deposits and some selected decrepitation results from my own investigations on Australian deposits during the past year. A number of decrepigrams are enclosed and these have been printed on a less opaque paper so that they may be overlain on a light table for relative comparisons.

Gold deposits of many diverse origins including sedimentary, volcanic hydrothermal, granitic hydrothermal. skarn and metamorphic occur in a wide range of geological environments. However the fluids involved are only rarely distinct between these types and so I have subdivided this discussion on the basis of fluid parameters rather than the mineralisation origin in order to avoid undue repetition.

The conclusion is drawn that existing FI work has been strongly biased toward the detailed study of individual deposits and that there is a serious lack of data pertinent to exploration situations. However this previous work does show the importance of the characteristics of the ore forming fluid during the mineralising process. Hence, by obtaining information on these fluids by taking measurements on FIs it should be possible to substantially improve the efficiency of gold exploration procedures.

LITERATURE SURVEY

Temperature of formation

The homogenization temperature (Th) of inclusions within the quartz gangue of many gold deposits has been determined in order to define the temperature of deposit formation. The formation temperature is derived from Th by making a correction for pressure, which in the case of must gold deposits (being emplaced at shallow depths) is only small (up to 30°C). The Russian literature includes numerous determinations of decrepitation temperature (TD) which is less directly related to the formation temperature and in this section only the more reliable Th data is discussed. The task of measuring the formation temperature is complicated by the complex parageneses of most gold deposits. Up to 7 stages of deposition are recognized in many deposits, these being based on the occurrence of characteristic mineral assemblages (quartz-adularia, quartz-carbonate, etc.). In general only 1 or 2 of these stages carry gold in any single deposit and it is necessary to determine the formation temperatures of each stage in order to deduce the mineralisation history. Most authors report that temperatures range from about 400 C to 200 °C over all stages with occasional references being made to higher temperatures of 500 °C or more in the mesothermal type deposits. However the temperature range of formation of the gold mineralised stages is typically much narrower and frequently the gold deposition occurs only over a narrow temperature range of some 20 to 50 °C. (20, 24, 27). Despite this there is no general consensus on the absolute temperature during gold deposition and this may be anywhere from 200 to 350 °C in most deposits. Although each mineralisation stage commonly shows a restricted range of temperatures, adjacent stages generally overlap in temperature. Some studies have shown a general decline in temperatures towards the later mineralisation stages (11). but in general the temperature remains fairly constant for all except the last stage which is often of markedly lower temperature.

Overall there does not seem to he a regional relationship between Au mineralisation and temperatures of formation for the hydrothermal type deposits, as is concluded by Nash in his study of 9 are deposits in Nevada (25). However within an individual district a correlation between mineralisation and formation temperature may occur and Nash (25) states that in such cases "the temperature range (of mineralisation) is rather small".

The only deposit types which show a distinctly different temperature pattern are the disseminated and stratabound types. Data from the Carlin deposit shows that the likely temperature of formation is 150° - 225°C (25,31), which is significantly lower than the gold stages of typical hydrothermal deposits. Fripp (11) proposes a formation temperature of 160° to 280 for stratabound deposits in Rhodesia but no actual FI data seems to be available to confirm this postulate.

Practically no data is available on the formation temperatures of barren veins in the vicinity of gold mineralised veins and it is impossible to say whether the two types could be distinguished by temperature measurements.

Boiling

Numerous studies have found evidence of boiling in association with gold mineralisation (31, 26, 29, 8, 16, 28). The gold deposition usually accompanies such a boiling stage although Casadevall et al (5) noted that boiling at the Sunnyside mine, Colorado, occurred only in the last (Au barren) mineralisation stage. Kamilli (16) presents a convincing case that the gold-silver bonanza at the Finlandia Vein, Peru, is due to Au precipitation accompanying boiling and suggests that many other such bonanzas mined in the past may also have been due to local boiling. The point is made that veins showing evidence of boiling in present outcrops may contain concealed bonanza deposits at the base of the zone of boiling.

However not all deposits show evidence of a boiling stage during formation despite specific efforts in some cases to find such evidence (25, 33). As no data on barren quartz veins in mineralised areas is available it is impossible to say whether evidence of boiling would be a useful exploration aid.

Salinity

Salinity measurements have been made on many deposits and in most cases the ore fluids contain from 0.4% to 4% NaCl equivalent. There do not seem to be any major systematic variations of salinity between mineralisation stages although a weak trend toward reducing salinity in later stages occurs at the Finlandia vein, Peru (16). Nash (25, 27) notes that significant local variations of salinity occur although these cannot be correlated with mineralisation. He also proposes that fluid density (closely related to salinity) could be used as an exploration guide in Nevada based on the marked difference in salinity between porphyry copper deposit fluids (frequently greater than 30% NaCl equivalent) and the gold deposit fluids. However no data is presented on the salinities of unmineralised veins to consolidate this suggestion.

In the absence of data showing distinctive salinities or salinity trends either through the various mineralisation stages or spatially near the ore bodies, the measurement of FI salinity does not seem to be of use as an aid to gold exploration.

Carbon Dioxide

The presence of CO₂ rich inclusions has been noted in a number of gold deposits including the Oriental (9) and Sunnyside (7) mines in the U.S.A. and deposits in South America, Canada, France and Russia (30, 4, 23). Machairas (23) has observed a direct correlation between the presence of CO₂ rich inclusions and the occurrence of gold grades of 30 - 40 grams/tonne for South American, Canadian and French vein deposits. However an inverse correlation between gold and CO₂ has been observed by Kalyuzhnyi et al at Chukotka, USSR, (15) who conclude that the presence of CH₄ and a low concentration of CO₂ are favourable gold indicators.

These contradictory results are probably due to different mineralisation styles as CO₂ could be expected in skarn deposits or strongly metamorphosed areas (37). The occurrence or absence of CO₂ rich inclusions could be a useful regional guide to gold mineralisation. Again there is a lack of background data available on barren areas.

Fluid Chemistry

In this study I have not attempted an exhaustive survey of data on the chemical composition of the fluids because of the difficulty of applying such information in exploration. (Chemical analyses of FIs require considerable effort because of the exceptionally small quantities of fluid available.) However practically all deposits have low chloride contents (salinities less than 4% NaCl equivalent) with high Na/K ratios.

High concentrations of bicarbonate ions have been reported in many Russian deposits (19,2). Although data presented by Anufreyev et al (3) confirm these high bicarbonate levels compared with various Sn, W and base metal deposits, bicarbonate levels in barren veins were similar to those in mineralised veins!

Again. no clear pattern of relationship between fluid chemistry and gold mineralisation is apparent. This would also seem to indicate that the Au can be transported in complexes with various anions in different situations.

Gold Transport and Precipitation Mechanisms

In accord with the wide range of deposit styles, temperatures and fluid compositions reported for gold deposits there is a similarly wide range of suggested transport and precipitation mechanisms. Transport as chloride, sulphide or bisulphide complexes is generally proposed but most work concentrates not on the mode of transport but on the mechanism of precipitation. Temperature changes, pressure changes, oxygen fugacity changes. boiling, wall rock interactions, CO₂ effervescence and pH changes have all been proposed in roughly that order of frequency. Some workers (14) invoke several mechanisms acting together while others are able to distinguish a single principal cause such as oxygen fugacity (13) or boiling (16). Obviously each deposit needs to he considered separately and no overall conclusion of immediate relevance to exploration methods can be drawn.

Zonal Patterns

Very few studies have collected background samples or attempted to sample outside or near the mineralised zone in order to test for any zoning of FIs around the mineralisation centre. The Russian literature is the best source of data on the development of FI "aureoles" around mineralisation. Th zonation has been observed at the Darasun deposit where palaeoisotherms concentric to s porphyry intrusive have been plotted and both the horizontal and vertical temperature gradients determined (22). Other workers (10, 21, 1) have also shown the existence of Th zoning both around individual gold deposits and across entire orefields and they have used this information to determine the direction of ore fluid movement, thereby inferring the ore source.

Boyle (5) conducted an extensive decrepitation survey in the Yellowknife area, Canada, in an effort to define fluid flow direction within the mineralised system but his results were inconclusive. Unfortunately no samples were collected of nearby barren veins for comparison with the mineralised veins. Decrepitation measurements have been widely used in Russia to outline TD and decrepitation activity zoning patterns around ore deposits (1, 34, 18, 12).

Much of this work is only available to the western world as abstracts in which actual maps of the zonation patterns are not given, however contours of palaeoisotherms and profile plots of decrepimetric activity over non-gold mineralisation are commonly shown in translations of Russian papers. Several such plots are shown by Roedder (32).

The observation of such zonal patterns seems to he of considerable relevance in exploration for gold deposits.

Vein Characterization

The ability to distinguish between ore bearing and ore barren quartz veins in the same district would be of great value in exploration. Normal geochemical techniques are difficult to use in this way due to the nugget effect of gold (which necessitates unreasonably large samples for Au analyses) and the lack of reliable geochemical pathfinder elements. At Murantau (36) decrepigrams enabled the distinction of 3 genetic types of quartz, one of which was mineralised. This was then used on a district scale to identify the various quartz types. Konovalov (17) also outlined differences between the gold bearing and barren veins in the Lena goldfield and considered such differences to be of use as an exploration tool. At the Aldan complex (35) mineralised quartz was found to show a single peaked decrepigram whereas barren quartz gave a bimodal decrepigram enabling ready distinction of the mineralised veins. In my own work I have also been able to distinguish mineralised from barren veins by using this technique on deposits in the Pilbara and Yilgarn blocks in W.A. (discussed fully later in this report). This method has also been of use on some Australian Sn and Mo deposits and it is now being applied in their exploration.

Literature Summary

The fluids associated with gold mineralisation are of quite diverse character and do not exhibit any consistent temperatures, salinities. chemical compositions or phase effects such as boiling ON A REGIONAL SCALE. However temperatures and phase effects do seem to show patterns related to mineralisation on a district scale. Most FI studies done to date have focused on the mineralised bodies themselves with very little attention to determining the local backgrounds. Only the Russians have looked at this aspect in detail.

Particular aspects of relevance in exploration are:-

The occurrence of boiling.
Possible temperature or decrepitation activity gradients around individual deposits or across a mineralised district.
Characterization of veins into either potentially mineralised or unmineralised by using decrepitation techniques.

In view of the gold potential of Australia and the importance attached to FI determinations in investigation of gold deposits in the USSR and the USA, the scarcity of published FI data on Australian deposits is enigmatic.

Relevance of Decrepitation analyses to Gold Exploration.

Because the collection of FI data by conventional microscopic methods is slow and expensive, decrepitation measurements are considered to be the most promising means of using FIs to advantage in exploration for gold. While decrepitation data is relatively quick and inexpensive the technique does not produce exactly equivalent data to the microscope methods. Decrepitation temperatures are rather less reliable than homogenization temperatures for estimating the deposit formation temperature. In decrepitation work there is only a limited ability to distinguish between primary and secondary inclusions and it is not possible to obtain any information on the fluid chemistry. However decrepitometry does give an indication of the abundances of inclusions and by measuring a large number of inclusions a statistically more meaningful result is obtained for each sample.

Despite these differences from the conventional microscopic methods, decrepitation data is still applicable in exploration in the following ways.

Palaeoisotherms of TD around individual ore bodies and across orefields can be obtained. Within any given area, assuming limited fluid composition changes and a constant host mineral phase, contours of TD would delineate formation temperature patterns although the absolute temperatures would not be accurate. However palaeotemperature anomalies and gradient directions could be defined.
Decrepimetric activity contours are easily produced and have been shown to be useful in exploration in many areas in Russia.
It may be possible to detect boiling conditions, given some microscope control data. This would be expected to give a very broad decrepitation maximum due to the FI heterogeneity which occurs under boiling conditions.
It is possible to characterize various vein stages and systems and relate them to mineralisation on a local scale by using their decrepitation signatures.

This technique is currently being exploited in exploration of some Australian deposits.

Decrepitation Studies at several Australian Gold Deposits

Over the past year Burlinson Geochemical Services Pty. Ltd. has investigated a number of gold deposits in Australia using a micro- processor controlled instrument especially constructed for the purpose. This is one of only a very few decrepitometers outside Russia and is thought to be the only computerized decrepitometer in existence. With it, reliable results, completely free of subjective effects can be obtained.

The work done to date has concentrated on distinguishing between multiple generations of quartz veins. At the Linden area, 200km NE of Kalgoorlie, W.A., 15 samples were collected for decrepitation as well as chemical analyses. Figures 1, 2 and 3 show typical results from 3 of the 4 different types of quartz veins in this area. Most of the auriferous veins (5 samples) shew a moderate activity decrepigram with a well defined peak (Figure 1), whereas the barren veins show low activity without any distinct peaks (Figure 2). A second type of auriferous quartz occurs in a brecciated quartz vein and this shows a well developed low temperature peak on the decrepigram (Figure 3). This survey successfully distinguished the 2 auriferous quartz systems from the barren quartz veins nearby.

Another survey at a small. unnamed prospect east of Port Hedland. W.A.. also successfully distinguished between auriferous and barren quartz veins only 10m apart. Figures 4 and 5 are of samples 40m apart in the auriferous quartz vein and show similar decrepigrams with distinct low temperature peaks whereas a barren quartz vein nearby completely lacks this peak {Figure 6).

Several samples from the Carida mine in the Menzies-Sandstone area, W.A.. also show a distinct low temperature decrepigram peak, albeit of lower amplitude than either of the above areas (Figure 7).

Samples of veins in the Pine Creek region, N.T., also show a similar, distinct low temperature peak on auriferous veins (Figure 8), while barren veins lack such a feature (Figure 9).

A sample from the Peak Hill mine, NSW, also showed some low temperature decrepitation activity (Figure 10), although this was much less prominent than seen elsewhere. No suitable barren vein samples were collected at this locality for comparison.

This low temperature feature on the decrepigrams does not occur at all gold deposits. Some auriferous veins at Linden (above) and auriferous "siliceous lode" material from the Fimiston mine, Kalgoorlie, W.A., (Figure 11) do not show such a peak. However in several hundred decrepigrams of other mineralisation types, including tin, molybdenum and copper deposits, only rarely have similar low temperature peaks been observed.

This curious apparent relationship between decrepigram shape and some types of gold mineralisation is based on insufficient studies to be convincing at present but is obviously worthy of further investigation!

REFERENCES

NOTE: References to COFFI indicate an abstract in "Fluid Inclusion Research - Proceedings of COFFI" Editor E. Roedder, University or Michigan Press, Ann Arbor, USA.

l. Andrusenko N.I. 1973 Temperature zonation of gold-silver deposits. Internat. Geol. Rev. V21 no.7 p815

2. Andrusenko N.I. & Shchepot'yev 1974 Temperature conditions and stages of formation of subvolcanic gold-silver deposits of central Kamchatka COFFI V7 p5, Geochem. Internat. VII no 1, p130

3. Anufryev Yu. N, Moskalyuk A.A., Pokrovskii P.V. & Purtov V.K., 1974 Mineral Forming solutions of hydrothermal deposits of Ural. COFFI V7 p7

4. Boyer F., Touray J-C, Vogler M. 1967 Presence if liquid CO₂ inclusions in quartz from the auriferous district of Salsigne. COFFI V2 p31

5. Boyle R.W. 1954 A decrepitation study of quartz from the Campbell and Negus_Rycon shear zone systems, Yellowknife, Northwest Territories. Geol. Surv. Canada Bulletin 30

6. Burlinson K. 1980 Fluid inclusion procedures for exploration - a literature survey. Burlinson Geochemical Services Pty. Ltd.

7. Casadevall T., Ohmoto H. 1976 Sunnyside mine. Eureka mining district, San Juan County, Colorado : Geochemistry of gold and base metal ore formation in the volcanic environment. COFFI V8 p32

8. Casadevall T., Ohmoto H., Rye 1974 Sunnyside mine, San Juan County Colorado: Results of mineralogic, Fluid inclusion and stable isotope studies (abst.) Econ. Geol. V69 p1178

9. Coveney R.M. Jr. 1973 Fluid inclusion studies at the Oriental Gold mine COFFI V6 p31

10. Davidenko N.M., 1973 Genetic classes and zoning of gold ore deposits of mesozoic Chukotka folded belt (based an inclusions in minerals) COFFI V7 p43

11. Fripp R.E.P. 1978 Stratabound gold deposits in archaean banded iron formation, Rhodesia Econ. Geol. V71 p58

12. Goncharvo V.I., Sidorov A.A & Shapovalov 1973 Decrepitophonic survey as a method of determination of hidden veins in the ore fields of volcanogenic regions. COFFI V7 p69

13. Hattori K. 1975 Geochemistry of ore deposition at the Yatani lead-zinc and gold-silver deposit, Japan. Econ. Geol. V70 p677

14. Henley R.W., Norris R.J. & Paterson C.J. 1976 Multistage ore genesis in the New Zealand geosyncline, a history of post-metamorphic lode emplacement. COFFI V9 p56. Mineralium Deposita V11 p180

15. Kaluzhnyi V.A., Davidenko N.M., Zinchuk I.N., Svoren' I.M.& Pisotskiy B.I. 1975 Role of CO₂-H₂O and CH₄-H₂O fluids in forming ores at Chukotka COFFI V8 p82

16. Kamilli R.J., Ohmoto H. 1977 Paragenesis, zoning, fluid inclusion and isotopic studies of the Finlandia Vein, Colqui district, Central Peru Econ. Geol. V72 p950

17. Konavalov l.V. 1975 Temperature of formatonn of gold ore deposits in the Leno Field as a function of metamorphic facies. COFFI V8 p93

18. Korobeynikov A.F.& Matsyushevskiy A.V. 1973 Application of methods of mineral thermometry for prospecting and evaluation of one necks in endogenic gold deposits. COFFI V6 p81

19. Korobeynikov A.F. & Matsyushevskdya L.B. 1973 Geochemical types of gold bearing hydrotherms from data on gas-liquid inclusions in minerals. COFFI V6 p82

20. Lazko E.M., Doroshenko Y.P., Koltun L.I., Lyakhov Y.V., Myaz N.I. & Piznyur A.V. 1971 Processes of hydrothermal minerogenesis in the East Trans baikalain deposits / on gas- liquid inclusions in the minerals COFFI V4 p40

21. Litvinov V.L., Lyakhov Y.V. & Popivnyak I.V. 1970 Palaeotemperature zoning of the Kariyskoye gold deposit. COFFI V7 p126 lnternat. Geol. Rev. V14 no. 9 972

22. Lyakhov Y.V. 1975 Temperature zoning of Darasun deposit. COFFI V8 p110

23. Machairas G. 1970 the association of fluid inclusions and gold particles in auriferous quartz. COFFI V3 p43

24. Naiborodin V.I. & Sidorov AA. 1973 Some properties of the dynamics of the mineral forming process in volcanic Au-Ag deposits. COFFI V7 p151

25. Nash J.T. 1972 Fluid Inclusion studies of some gold deposits in Nevada. USGS Prof. Paper 800C pC15

26. Nash J.T. & Cunningham C.G. Jr. 1973 Fluid inclusion studies of the Fluorspar and gold deposits. Jamestown district Colorado. Econ. Geol. V68 p1247

27. Nash J.T. 1975 Fluid Inclusion studies of vein, pipe and replacement deposits. Northwestern San Juan Mountains, Colorado. Econ. Geol. V70 p1448

28. O'Neil .J.R. & Bailey G.B. 1979 Stable Isotope investigation of gold bearing jasperoid in the Central Drum Mountains, Utah. Econ. Geol. V74 p852

29. Petrovskaja N.V. & Vasiliev V.I. 1973 Evidence from electron microscopy on gaseous inclusions in quartz from the Balei deposit as an indication of the boiling of hydrothermal solutions. COFFI V6 p123

30. Popivnyak I.V. 1975 Role of CO₂ in forming of deposits of Muyskiy gold ore region COFFI V8 p146

31. Radke A.S. & Dickson F.W. 1974 Controls on the vertical position of fine grained replacement type gold deposits COFFI V7 p177

32. Roedder E. 1977 Fluid inclusions as tools in mineral exploration. Econ. Geol. V72 p503

33. Sawkins F.J., O'Neil J.R. & Thompson J.M. l979 Fluid inclusion and geochemical studies of vein gold deposits, Baguio district, Philippines Econ. Geol. V74 p1420

34. Shevkalenko V.L. & Tsoi A.V. 1972 Use of the decrepitogram of quartz for the study of zoning in gold-silver deposits of Shkolnoe (northern Tadzhikstan) COFFI V5 p100

35. Shany G.K. & Levitskiy Y.F. 1976 Gold ore mineralisation and metasomatic processes in alkaline rocks (exemplified by Aldan complex) COFFI V9 p127

36. Veres G.I., Kasavchenko G.V. & Trankvillitskaya I.A. 1973 On evaluation of gold bearing features of quartz Fran Muruntau by thermobarometric analysis. COFFI V6 p162

37. Wilkins R.W.T. 1977 Fluid inclusion assemblages of the stratiform Broken Hill Ore deposit, NSW Science V198 p185