DECREPITATION SAMPLE SPECIFICATIONS
The following items should be carefully considered in collecting samples as part of a decrepitation survey. Each major consideration is discussed in detail below.
- Choice of mineral phase
to be used.
- Preference for
- Zoning of vein quartz.
emplacement metamorphism & recrystallization.
- Sample quantity and
- Contamination by trace mineral phases. (Carbonate & Sulphide)
GENERALThe decrepitation method can be used on opaque, translucent and transparent minerals with equal ease and provides information on the abundances, formation temperatures and the fluid compositions of the fluid inclusions. Although it is sometimes possible to infer absolute formation temperatures, this involves making several assumptions about the fluid composition and so it is generally preferable to interpret the results empirically in a manner similar to interpreting a soil geochemical survey. It may often be of advantage to examine some of the samples in thin section in order to gain a better appreciation of the nature of the fluid inclusions involved. If concomitant microscope work is planned, great care should be taken to ensure that the sample sectioned is exactly the same as the one decrepitated. The decrepitation sample in such cases should therefore be the offcuts from the thin section preparation to ensure that both samples are from the same growth zone. (Growth zoning in quartz is more common than many people care to admit!)
CHOICE OF MINERAL PHASE TO BE USEDGenerally, quartz is the preferred sampling medium as thin section work can be done to confirm and complement the decrepitation results if necessary. Quartz is also resistant to weathering and can retain it's fluid inclusions despite lateritization. It also has low ductility, which is theoretically favourable for generating counts in the decrepitometer. The quartz may be either as ordinary veins or as silicification of host rocks. Silicification haloes give low level responses and it is suggested that more samples than usual be collected when using this medium as some 20% of the samples typically give responses too weak to interpret. Many other mineral phases can be used and the following have been tested successfully to date:- haematite, magnetite, pyrite, fluorite, barite, calcite, dolomite, olivine, feldspars (MUST be fresh), galena and sphalerite. Unsuitable minerals include chlorite, muscovite, sericite (and presumably all layer silicates), kaolin (and presumably all clays). Note that these hydrous minerals do NOT produce decrepitation counts due to dehydroxylation. In theorey, ductile minerals like galena, fluorite and the carbonates should not decrepitate, although in practice they decrepitate very well. (Theorey has yet to catch up with the facts on this!) However, it would be best to avoid these minerals if possible until we develop a better understanding of them. Many other minerals could also be used although several trial samples of other minerals should be done first to check their efficiency.
MONOMINERALIC SAMPLESDifferent minerals with different mechanical properties require varying excess pressures to cause decrepitation. Hence, to ensure consistency, it is preferable to avoid mixing minerals either within a sample or across a single survey. If it is impossible to collect monomineralic samples, mixtures can sometimes be used (such as quartz-feldspar in granites) or mineral separations can be done. Please discuss such problems before collecting such mineralogically diverse samples. Inclusion of clays and layer silicates within the sample is not a serious problem (except inasmuch as it dilutes the active mineral), but inadvertant inclusion of carbonates and some sulphides can cause serious problems. (See below) (Acid washing of the samples can be used to remove carbonate contamination.)
ZONING OF VEIN QUARTZMany quartz veins have pronounced growth zoning developed inwards from the vein walls. Often this is readily visible, but sometimes it is not. In general, veins which posess vugs and cavities with well formed quartz crystals are growth zoned. Uniform, massive milky veins tend not to be zoned. If they don't look completely homogenous from wall to wall, they probably aren't! Failure to recognize the possibility of growth zoning could render the results uninterpretable, so veins in unfamiliar areas should be sampled at several positions across their width to check for growth zoning. Variation along the length of a quartz vein does occur within limits. Depending on the objectives of the survey it may be necessary to ascertain the amount of such variation by collecting samples at various points along strike in the same vein. Variation down dip in quartz veins is theoretically likely, but rarely observed. Unless the thermal gradient up the vein was exceptionally high, a vertical section of some 500 metres would be necessary to observe this effect. Only rarely would it be necessary (or possible) to collect samples to check this effect in a survey. Variation between different vein sets, even when they spatially overlap, is usually far more pronounced than variation due to any of the above effects. Test samples should be planned and collected to assess whether any of the above variations interfere with the variations you plan to interpret.
POST EMPLACEMENT CHANGESMetamorphism and recrystallization during folding subsequent to emplacement can alter or erase the fluid inclusions from the minerals of interest. The effects are quite variable and in a great many cases useful decrepitation data can still be obtained despite these processes. Metamorphism to greenschist facies rarely affects quartz veins, but at higher grades, new inclusions might be introduced and swamp the original inclusion signature. Recrystallization typically reduces the numbers of inclusions present but, even when severe, sufficient inclusions usually remain to allow interpretation. These effects do need to be considered carefully, but they are not usually as serious as they theoretically might be. But be careful if you are in high grade metamorphics.
SAMPLE QUANTITY AND FORMThe amount of sample actually analysed is 0.5 to 2.0grams. The form of the sample analysed is as grains, pre-sieved to -420+200 microns (-40+80 mesh). In finer grain sizes, many of the inclusions have already been mechanically broken open during sample comminution, so pulverized samples are unsuitable. Either lump samples or samples passed through a jaw crusher should be submitted for analysis. A minimum of about 10 grams is preferred to allow for losses into fine grain fractions during sample preparation. (Smaller samples can be used by special arrangement.) Lump samples some 2 to 4 cm. across are convenient and are cheap to freight. Note that because there are typically billions of fluid inclusions per gram (in vein quartz), there is no need to send large samples for reason of obtaining a representative 0.5 gram split. The samples should be as monomineralic as possible (see above) unless mineral separations have been arranged.
CARBONATE AND SULPHIDE CONTAMINATIONCarbonate minerals all decrepitate violently (well below their pyrolytic decomposition temperatures) and almost always give exceedingly high count rates. (They can saturate the decrepitometer counter circuits.) It is quite possible for contaminant amounts (say 5%) of carbonate in a quartz sample to completely swamp the quartz result. If this happens, the decrepigram cannot be interpreted. (The presence of carbonate is immediately obvious). Consequently, unless you are deliberately using carbonates for your samples, you should scrupulously avoid them. If necessary, acid washing can be arranged to remove carbonate contamination. Sulphides can give rise to strange peaks on the decrepigrams. Generally these do not swamp out the result from the dominant phase, but they can make the interpretation more difficult. Some sulphides oxidise within the heating range of the instrument and this generates many decrepigram counts. Such oxidation is exothermic and thus self sustaining and the decrepigram peaks resulting are of little interpretative use. Some sulphides melt and some have been observed to form molten native sulphur during the analysis. These are definitely undesireable as they mess up the sample tubes. Unless you are deliberately using sulphides it is best to try and exclude them from the sample, although their effect is nowhere near as bad as carbonates and trace amounts can be tolerated.
K.Burlinson April 87 V1.2, July 2000, 2009