non-aqueous fluid inclusions

Recently there have been a number of papers inferring the transport and deposition of gold by pure CO₂ fluids in mineralizing hydrothermal systems. These conclusions have been based on the observation of fluid inclusions which lack a visible aqueous phase and the failure to observe (or document) aqueous fluid inclusions in the same sample of quartz. These authors usually refer back to several previous publications to justify their own inference of such fluids as being dominant in the system under study. However all these papers fail to deal with the issue of the origin of the quartz in which all the studies were carried out.

Many authors refer back to the paper "High CO₂ content of fluid inclusions in gold mineralisations in the Ashanti Belt, Ghana: a new category of ore forming fluids?" by Schmidt-Mumm et al in mineralium Deposita, 1997, V32, p107-118. But there are serious issues that need to be considered and which are not dealt with in that work.

The real problem with this and with all subsequent work purporting to prove that Au is transported in (essentially) non-aqueous fluids is the complete failure to observe that the inclusions being studied are within quartz. Everyone is so intent on seeing the microscopic fluid inclusions and discussing the ppm levels of gold in the system that they have failed to see that the system is in fact about 99.999% quartz (with 10 ppm gold). It must be remembered that the fluid inclusions being observed are trapped entirely within quartz and that silica deposition must have been essentially contemporaneous with the presence of the non-aqueous fluid under study in order for the inclusion to seal and trap that fluid. However, silica is not known to be transported by or deposited from non-aqueous CO₂ fluids. Based on the "like dissolves like" guidelines for solubility, it is unlikely that silica could be transported by super-critical CO₂, because CO₂ is a non-polar molecule and would not be expected to dissolve silica which is soluble in polar water. Silica dissolves in water primarily because of hydrogen bonding with the protonic solvent, water, but non-protonic CO₂ lacks the capability to form such hydrogen bonds and it is quite unlikely that super-critical CO₂ could dissolve silica. It is disingenuous and probably misleading for authors to infer the transport of silica in super-critical CO₂ as a mechanism to try and justify their conclusions.

Although most authors make a comment that quartz is not known to be transported in non-aqueous CO₂ fluids, the all too obvious conclusion is that there really was a dominant aqueous fluid present in all these systems and that the aqueous fluid transported and deposited the overwhelmingly dominant quartz mineralization and quite possibly also the gold. And for some reason the aqueous fluids failed to give rise to identifiable fluid inclusions, or that the aqueous phase inclusions present have been ignored.

The real problem here is to understand why the dominant aqueous fluid is not being recognized and reported in the research results.

Microscope observation of fluid inclusions is a tedious and often boring task and there is a strong tendency to focus ones attention on the unusual observations. So large inclusions, or ones with unusual compositions invariably attract attention. The myriad of small inclusions are often too small to allow the contents to be seen clearly and are ignored. Or they are considered to be secondary inclusions and unimportant. In many cases the paragenesis of the small inclusions is unclear and established protocol is to ignore inclusions which do not meet the very strict observational criteria to confirm their origin. So these inclusions are routinely ignored simply because they cannot be classified. But that does not mean the fluids they represent are inconsequential! It means the research is incomplete!

If a fluid inclusion study were done on our favourite sugary, fizzy soft-drink, we would see that the "inclusions" (bubbles) present are almost entirely CO₂ with a very small water content. So should we conclude that the "mineralization" (sugar) present must have been transported by pure CO₂ fluids? Of course not, the host medium (water) is not irrelevant in understanding this system, even though the water itself is rather common and boring. Similarly, the host quartz medium for the fluid inclusions and gold in a mineralized quartz vein is not irrelevant. To understand the system one must explain the origin of all the species present and silica (as quartz) is the dominant component which must not be ignored merely because it is common or boring. It is a key part of the mineralizing system and it was not transported by pure CO₂ fluids. The ubiquitous existence of the quartz suffices to assert the dominant involvement of aqueous fluids in the system.

Observations of CO₂ rich fluids are not wrong. They probably demonstrate clearly that at some stage during the millions of years of activity of the hydrothermal system there was at least one CO₂ rich fluid phase event. But for how long did this last, how important was it in the overall formation and was it actually the phase that carried the economically interesting mineralisation? Or was it just a sideshow to the main aqueous fluids which actually transported the silica and other minerals of interest?

Hydrothermal systems are not simple stable fluid events. Most studies describe multiple different fluid inclusion assemblages and there are clearly different fluids present at different times. But there is still a reluctance to understand just how variable such systems are as would be witnessed by visits to the geyser fields at Yellowstone park or other surface outcrops of present day hydrothermal systems. The fluids can change in mere minutes. And typical mineralizing quartz systems were active for millions of years, so there was ample opportunity for numerous fluid compositional changes. Our observations are hardly representative of the immense amount of variability that really does occur in hydrothermal systems. It is naive to therefore think that observation of some interesting CO₂ rich, non-aqueous inclusions in a quartz vein is adequate to prove that an aqueous phase was never present at all and that the economic mineralization (and the quartz itself) must therefore have been transported in this probably fleetingly transient fluid event.

Observations of non-aqueous CO₂ fluids in quartz veins do not prove that aqueous fluids were absent. Until there is evidence (currently lacking and unlikely) that silica can be transported in and deposited from such non-aqueous fluids, then the mere fact that the observations are being made on fluid inclusions hosted within quartz is enough to be quite certain that aqueous fluids were not only present but were actually dominant as they transported and deposited the quartz which comprises some 99.999% of the mineral matter present.

In my own baro-acoustic decrepitation analyses of samples the results reflect the entire assemblage of inclusions present in 1 gram of sample. Using this method at the Brusson gold mine in Italy I identified 3 distinctly different inclusion assemblages and their spatial distribution at differing depths within the vein system. (Burlinson, 2010) Because one of these assemblages terminated at the same mine level as the gold mineralisation terminated it was possible to develop a depositional model to explain the high level cutoff of both this quartz type and the gold mineralization. Understanding the change in quartz deposition at this level was crucial in realising why the gold mineralisation terminated at the same level. Previous studies of this mine had been unable to provide anything more than a vague explanation of this mineralisation cutoff despite several PhD doctorates, numerous fluid inclusion measurements and extensive stable isotope work. It is always appropriate to understand the quartz vein itself when studying the genesis of a minor mineral phase within a quartz vein.

It is unacceptable to ignore the overwhelming presence of quartz, merely because its market value is less than that of some minor metallic constituent with which it is associated. And given the presence of quartz we can be pretty certain that aqueous fluids were not only present but also important, or even essential!

An explanation of these gas dominant inclusion assemblages which lack aqueous inclusions is that they are caused by disproportionate inclusion trapping from a heterogeneous parent fluid system, as discussed here.

Reference: Burlinson, K. - Gold exploration using baro-acoustic decrepitation (abstract). International Mineralogical Association meeting, Budapest, July 2010.

Gold exploration using baro-acoustic decrepitation - full oral presentation.

Depositional model for the Brusson Au deposit, Italy

Return to main CONTENTS