Wednesday, February 18, 2009

High metal concentrations in fluid inclusions?

I've got a question for any readers with expertise in ore deposits. There's a paper in the Feb. 6 issue of Science (Wilkinson et al; discussion of paper here; warning: paywall) about anomalously high concentrations of metals found in fluid inclusions in ore minerals. I'm curious what you think, and whether the conclusions have practical implications for mineral exploration.

Here are the basics, as far as I understand them. Hot water traveling through rock is responsible for depositing many of the ore deposits that we use for metals. (There are exceptions - iron, for instance.) In previous work, the concentration of metals in the hot water has been estimated from the tiny bits of fluid trapped in quartz or other minerals in ore deposits, but not in the ore minerals themselves. (The quartz has been interpreted as having formed at the same time as the ore minerals, but the study authors argue that it's difficult to be certain.) The concentrations of the metals in those inclusions are fairly low, which means that a lot of water would need to have traveled through the rock in order to form the ore deposits.

This study measured the amount of metals in fluid inclusions in the minerals quartz and sphalerite (zinc sulfide) from two different lead-zinc deposits (one from a Mississippi-Valley-Type deposit in Arkansas, and one from a higher temperature deposit in Ireland). In both cases, the concentration of metals (especially lead) were one to two orders of magnitude higher in the sphalerite. That's a huge difference, and would mean that much smaller amounts of water (and shorter amounts of time) may be necessary to form economic mineral deposits. It also means that the processes that collect and concentrate dissolved metals may be more important in making an ore deposit than the processes that cause them to be precipitated.

But the commentary said that both experiments and theoretical models (I'm not sure which, exactly - thermodynamic models?) predicted lower concentrations than those observed, so I'm curious exactly how big of a problem these observations are for ore geochemistry. (This study dealt with lead-zinc deposits, but the commentary mentioned that other studies had found similar results for copper and gold.) This paper talked about ways that lead and zinc could be concentrated, by evaporation of brines, but I'm not sure what processes might control the concentration of copper or gold in deeper deposits. Does it change the way one might explore, or are the targets the same regardless of whether we understand the source of the fluids or not? (Would understanding this problem make it easier to guess which sites wouldn't have economic concentrations of metals?)

References:

Wilkinson, J.J., Stoffell, B., Wilkinson, C.C., Jeffries, T.E., and Appold, M.S., 2009, Anomalously metal-rich fluids form hydrothermal ore deposits: Science, v. 323, p. 764-767.

Perspective: Bodnar, Robert J., 2009, Heavy metals or punk rocks? Science, v. 323, p. 724-725.

3 comments:

Mathias said...

I can't even access an abstract of the article mentioned. So whatever. I am a bit sceptical if gangue and ore minerals really form at the same time. In Pb-Zn deposits the ore deposition releases a fair amount of acids that dissolve the host-rock of MVTs creating the space needed to precicpate ore. Then again the ore itself my directly replace the host-rock or even earlier generations of gangue or ore minerals. The ore fluid needs to be brought of out balance usually by encountering either different chemical conditions within the host-rock or by fluid mixing.

MVTs usually have a small to not-recognisable alteration halo making them hard to spot. The more fluid is needed the bigger this halo I assume. Most common are dolomitization, Ca being flushed out into country rock or silification. If the amount of fluid needed for the ores to form is smaller, so would, I assume, be the chances to form large alteration halos.

I do not think the general exploration approach to MVTs is changed but it might lead to the question: how do we recognise a large hydrothermal system in the field? Is it still viable to disregard small scale alteration as unimportant? Perhaps ore deposits can form in such short time not to leave any impression in terms of alteration.

It might explain giant deposits like the Silesian ones. These are known to have stalagtites and stalagmites growing up and downward from the walls of cavities. They are consist only of ore minerals! Imagine what high concentration you might need for that. It might be easier to explain now.

Well, I am not an expert though. Just a student with an interest.

Silver Fox said...

I think these are kind of complex questions, and after scanning as much as I could online (mostly abstracts), I'd have to note that this study was done on one MVT-type deposit and one other zinc deposit. MVT deposits form at very low temperatures; I'm not familiar with the other deposit studied or how much higher its temperature of formation was (the Ireland deposit could be MVT-type or volcanogenic massive sulfide - VMS-type). MVT deposits also form from highly saline brines, with higher salinity than fluids from many other hydrothermal systems.

Fluid inclusion in non-gangue minerals are sometimes analyzed for metal concentraions, but contamination from the surrounding material may be a problem. Possibly that's why they analyzed lead in a zinc-iron sulfide. I'm not sure if they ruled out any galena intergrowths in the sphalerite, or if that would be a problem.

There have already been suggestions from some other studies (in MVT deposits and in larger, higher-temperature hydrothermal deposits) that during the long ore forming period, there may be pulses of fluids that are higher in metals than the usual relatively metal-poor fluids. Those results have been seen from gangue studies.

If this study holds up for MVT-type deposits, it could change the way one looks for those type deposits in particular. Before doing much with it, however, I'd really want to talk to a fluid inclusion expert and even an MVT expert. Possibly one could hypothesize a new deposition-formation model that would lead to new discoveries.

I have a problem extrapolating this study to other types of hydrothermal systems like porphyry copper deposits where the alteration halo is huge and the temperatures are much higher, or to deposit types where boiling and gas transport of metals is present, like in many volcanic-hosted gold deposits. Metal concentrations are also studied in active hydrothermal systems rather than just from fluid inclusions, and can be applied to these other kinds of (non-MVT) systems.

I hope that helps!

Mathias said...

Silvermines in Ireland is a high-temperature MVT deposit that have been given their own class of so called Irish-type deposits because of their relatively high formation temperatures. There is some speculation(??) of an underlying magmatic heat-source which is rather uncommon for MVTs.

I think the conclusion, even if correct, should be considered with care. MVTs especially are very varied and with some deposits the only overlapping criteria is their variety. Still not having really read the articles (thanks for them again Kim!) I would assume that the results might be more secure if not conducted on MVT deposits but on a deposits type that is generally not so extremely varied. Nonetheless, it is interesting stuff.