Jump to content

A Basic Intro To Xrf Guns For Prospecting


Recommended Posts

Thanks for all the comments everyone, I'm glad there is some interest in this!

The LIBS are interesting units, they are starting to appear on Ebay now used too, but more expensive than XRF. I'm waiting for that tech to develop some more. I was looking at a trying to build LIBS type system at home for lighter stuff like lithium and beryllium a few years back. Specifically lithium because of the big electronic vehicle battery rush. But I think I found a way to detect it with optical spectroscopy that is cheaper so I haven't paid much attention lately.

Yep, the measurement is only a small fraction of an inch deep. And the spot is quite small in area too. For ores it seems the best technique is powdering them in a crusher and pelletizing them to avoid the "nugget effect" in even smaller scale. One thing I am thinking about is simply grinding down rock with a diamond wheel on a battery powered angle grinder, and collecting the powder for analysis, instead of crushing.

That's interesting you are getting very close to your assay results Gold Hound. I hope to compare some XRF measurements to actual assays next winter. I am looking forward to trying this out on some gold ores, but right now everything near me is under 6ft of snow and my project in Arizona is 16 hours drive away. 

This XRF I have uses a rhodium anode, for the reasons you mentioned, I wanted a bit of increased sensitivity to gold. However, I also am looking at prospecting for rhodium itself and associated things like iridium, so I'll have to see how this anode effects that. I would like to buy some rhodium and iridium ore in proper matrix to test on before I go out to the field, but the stuff on ebay all appears fake (and way too expensive).

  • Like 1
Link to comment
Share on other sites


Since I don't have any gold ore here at home and there is 6ft+ of snow in the mountains here, I was trying to think of something sitting around the house with a fairly complex composition that would still have relevance to prospectors. So I decided to take a look at Gold Basin meteorites. I'm not a meteorite guy and honestly other than iron and nickel I was pretty unclear about what exactly a stony meteorite is made out of since I've never read into it so I thought this might be interesting.

On a side note, I just realized another thing stolen out of my RV during the theft out there - a big tupperware storage box full of meteorites. Probably 10lbs+ worth, I had been tossing them in there for a few years. Luckily I have more at home, and some thief probably has no clue what he even stole. Oh well. So anyways, these meteorites I have here were mostly recent and dug deep as they were stuff I had missed with the Z14 but later hit with the various X Coils this season, and as such they do not have strong fusion crusts since they have weathered heavily being buried in the ground where moisture remains much longer than on surface.

Sample 1, 110grams, not much fusion crust, small stress fracture

image.thumb.png.e277f0f0c8124040404102cec94940e6.png

I shot this one as close to the middle on this face you see in the photo here, off to the left side of the stress crack. I am running in Mining_LE_FP mode, which means "light element" and "fundamental parameter". Iron is not light, but what it's referring to is the matrix material (the meteorite itself) because this mode is actually calibrated to detect heavy elements too, and I'm unsure what the matrix is so I'm using LE mode to start since it's a common mining mode for general ores including those with iron.  I am using a fundamental parameters mode because I want to see if it can calculate elements independent of the matrix. This does not require known samples to calibrate to, and clearly my XRF is not calibrated to a Gold Basin meteorite. I am just doing 15 second scans as we don't need extreme accuracy here, we're just spitballing the basics without many concerns for accuracy yet.

(note: mining modes are generally in PPM, but I have changed it set to % since this is just an intro guide and % makes more "human" sense)

Let's look at the results:

image.png.44d4b7f42552af33fbaa4c9cf64d343a.png

Wow...ok. Not what I expected, but again I don't know much about meteorite composition. I really thought I'd be above 50% iron and maybe 15%+ nickel. And far less silicon. But, these are stony meteorites and not iron meteorites, so maybe these results do make sense. 

The graphs are showing the spectral results for each element in the meteorite. The top graph is from the 40kEV beam and the bottom graph is from the 15kEV beam. You can see that there isn't much in a GB meteorite that a 15kEV beam misses, other than a bit of info around 20kEV. I cannot figure out how to get the machine to export the element lines themselves with the graphs yet, or to zoom in around a specific energy, only on the viewscreen on the machine itself can I do those things. So, to save space I won't include the spectrums in the other meteorite results.

Notice the results do not add up to 100%, as I explained would happen in the first post. This is because the XRF cannot detect elements lighter than magnesium, so the missing mass is locked up in rock building stuff like oxygen, hydrogen, carbon, etc.

Let's shoot it again, this time using the general Mining Mode, which lets the machine automatically select the best calibration. This is the default mode, and what most people would use starting out.
 

image.png.8ac9a1d5b868fccacb62f457486c4246.png

Ok hmm, what's changed here? Why does the meteorite no longer have silicon, aluminum, or magnesium in it?

So first, this is a chrondrite. Which means it's very un-homogeneous from point to point. One measurement could hit a chondrite, while the next could be in all matrix material. This is why samples really need to be crushed/powdered, mixed, and then analyzed for accurate results.

But look at the "Class". The XRF has automatically selected "mining_fp", notice the LE (light elements) is missing from that calibration? This is what I meant in the first post by the gun itself could be calibrated for all elements, but if the specific calibration (in this case, mining_fp) is not calibrated for those elements, they will not show up. This is why buying a gun with only "alloy" calibrations will require sending the gun to the manufacturer for significant new calibrations, and significant more $$'s.

The subtleties are not often explained, and I wanted to show a good real life example of why people will buy a unit off ebay thinking they can use it for some specific application like gold prospecting, only to discover their unit needs another $8k in work at the manufacturer. Luckily my XRF has basically every calibration the factory offered, already installed. And this is why it took me 5 years of searching for one like this that was also affordable.

Sample 2, 170 grams, some fusion crust remains

I wanted to to see how the fusion crusted edge looked compared to the weathered away part of a meteorite. Like I mentioned earlier, most of these were dug deep where the soil stays moister, so the ones I have here mostly lack any kind of very strong crust, but this one definitely has it visible. I took a picture outside where it's snowy because the white balance on my phone seems to get thrown off and not show the crust very well inside.

image.png.305e4d689e5124fe9c37da7d7c1e0813.png

I tried centering the measurement on that good sized area of fusion crust above my thumb. Accuracy is not exactly easy to achieve though, but I'm sure I got it onto the crust. Here are the results:

image.png.110105b9a59b23767575d3ac9f648982.png

Hmm again, not what I expeted. Being the rim, I expected to see some iron depletion due to weathering. But we actually see a higher concentration of iron. Looking closer though, you can see that the crust appears to be largely iron and in some places it's turning into limonite or rust (brownish red spots o the top and right of my thumb). 

Oddly though, there is no magnesium in this reading even though I am in an "LE" configuration. Why? I don't know, but my guess is that iron melts and forms the majority of the crust while the other minerals in the meteorite, especially whichever one contains magnesium, ablates away as the stone melts and falls. So, this is one example where it's important to understand the calibration, but also to be able to interpret the results. The answer depends largely on wether I can interpret correctly or not (in this case, I honestly have no idea, just guessing).

Also, in this case the relatively shallow sample depth of the XRF wasn't a hindrance as much as it helped us. Because all we wanted to look at was the thin veneer of fusion crust.

So anyways, it appears that the fusion crust is enriched in iron, whatever the reasons may be. It is possible this could affect how detectable a meteorite is too, or what it sounds like on some metal detectors, but that's another guess.

Sample 3 (sample 1 reshoot)

I decided to reshoot sample 1 again except on a different spot, intentionally trying to avoid a chondrule and just into the matrix if I could. My main goal was to see what the elements with less concentration than aluminum look like and how they change. I am doing this because after I write this post I am going to see if I can find actual XRF spectra of GB meteorites (if any meteorite guys have a university or lab result, please share it) and then to see if I can identify what likely other mineral components make up a GB meteorite based on the results we see here.

image.png.034745be1ecfff7e6c3c926e3b5ae2a1.png

Ok, so the new reading is at the top and the old reading is at the bottom. It does seem like I got more matrix in this reading, but you can see how this particular meteorite also has a somewhat unique "fingerprint" (I hope to look more into that in a future post when I get some time to think about the results and do some internet sleuthing), and it also seems to have 3 unique trace elements - Cr, Co, and Ti - which might be able to be used to differentiate a GB meteorite from another identical looking chondrite. But I'm just guessing because I haven't looked at any other meteorites to see if those elements are also present or not.

So, looking at contituent mineral ID's, here's my thoughts right now before I've done any googling - Fe and Ni are self explanatory, but I feel they are a bit low and maybe my XRF needs calibrated or a different method used instead of Mining_LE_FP. Si, Mg, Al, and Ca are very common rock building elements. Right away, being a physics trained guy and knowing how old meteorites are, my first thought is the Si and Mg is coming from fundamental, old, and very common rocks like olivine. The Ca could be a number of things but the small concentration leads me to guess terrestrial caliche remnants after falling to Earth (if accuracy were a concern I would have done the measurement on a fresh cut face). And the Aluminum is the big mystery to me since the minerals I can think of generally wouldn't be solar system-level old and are largely terrestrial, so that will require Googling.

Also of note is that Iron and Nickel both decrease, showing they may be alloyed together somehow or physically together in some way and not dispersed through the stone homogeneously. Which again is likely due to this being a chondrite.

So, while we don't technically prospect for meteorites - this ended up being a great example to show how to use the modes, what to look for, the importance of interpretation and some basic geologic/chemistry knowledge, and how you might use a tool like this to ID minerals or mineral constituents in ores.

More later on figuring out whats inside these meteorites... And even though the XRF really doesn't have good calibrations for looking at solid gold nuggets, that will be the next thing I play around with. Maybe see if we can tell the difference between a Quartzsite and a Gold Basin nugget, even though the geology  and gold itself of the areas can be similar looking. 

  • Like 1
Link to comment
Share on other sites

So I wasn't able to find a university or lab produced x ray spectra of a Gold Basin meteorite, I was hoping I could to use as a control measurement since it's possible my XRF could be miscalibrated or not in the best mode. So, I'll wing this one and take my best guesses based on the data, and anyone with such a report can correct me later.

What I did find was a report stating the University of Arizona has classified the meteorite as an L4 Stone Olivine Hypersthene Chondrite containing Olivine, Pyroxene, Kamacite, and 0.09% weight Cobalt.

So, my initial guess about olivine being a likely constituent mineral was correct. Why did I immediately guess that? Because olivine is a fundamental rock forming mineral on Earth (and thus, likely in other places too), probably the most common, and I knew it had a good bit of magnesium in it. Simply put, olivine is always one of my suspects when I'm trying to ID minerals that have high magnesium, especially in cases the rocks are very, very old.

Olivine (from webmineral):

     Magnesium  25.37 %  
     Iron       14.57 %  Fe  
     Silicon    18.32 %  Si   
     Oxygen     41.74 %  O

Also, my XRF reading matches fairly close with UA's cobalt measurement, indicating these are both Gold Basin's and not some other stray meteorites in the strewn field.

What I'm seeing in various reports is that the iron content of these meteorites is somewhere around 24%, which also matches fairly closely to my readings, understanding that the iron various widely millimeter by millimeter in a chondrite. The iron and nickel come from what UA is calling Kamacite, which has a very low nickel to iron ratio, which also matches what my XRF readings have shown (keeping in mind we are also pulling in Iron from the Olivine)

Kamacite (from webmineral):

Iron    89.54 %  Fe
Nickel  10.46 %  Ni

The aluminum is still a mystery to me. Pegmatitic materials would be unlikely on a meteorite I think. I'm not very familiar with the Pyroxenes, so I checked Wikipedia and I see the first sentence:

"Pyroxenes have the general formula XY(Si,Al)2O6"

So there is our aluminum constituent. Reading further the last mystery of this meteorite is solved:

"...Y represents ions of smaller size, such as chromium, aluminium, iron (III), magnesium, cobalt, manganese, scandium, titanium, vanadium or even iron (II)."

Wiki says the mantle of the Earth is composed of olivines and pyroxenes, so I probably should have guessed this one along with olivine to begin with, but I'm learning.

Remember the strangely appearing Cr and Ti? It would be reasonable to assume it's coming from the included pyroxene minerals.

Lastly, looking at melting points for Olivine and Iron, it appears to me the most likely theory for the iron enrichment in the fusion crust is simply that Olivine is melting first and ablating or even vaporizing, then later at a higher temperature the Iron melts but the heat is not enough to vaporize it so it remains. Oddly, the silicon remains too from the olivine though while the magnesium appears to vaporize. I don't know why and I don't have a guess because I suspect it has to do with chemistry and I suck at actual real chemistry.

So, I know this was a lot to take in, especially for people just casually reading this. But this is the sort of information that can be really useful when using an XRF. These are the ways you can solve various mysteries, or at least how I approach doing so, and I'm wrong a lot too. These same mysteries appear daily when I prospect for terrestrial minerals, and thinking about and then solving the questions that arise can lead to some understandings of an area of interest that literally no one else on the planet has except you. That is an advantage, and a part of prospecting is finding advantages and putting them to work and turning them into discoveries. If this is the type of stuff that is useful to you when prospecting, then an XRF may be a tool that is worth every penny.

 

  • Like 1
Link to comment
Share on other sites

A very handy tool for prospecting.  Arsenic is the main indicator element in our area.  To save money on assays I zap every soil sample then only send the ones away with elevated As.

They are a great toy to play with, my mate and I end up zapping everything in sight.  Would never eat fish again if you seen some of the hits we got on them!

Another interesting one, Caterpillar wanted $8k for some head bolts, being tight arses we got the $600 after market ones.  Once we zapped both the composition of them was identical!

  • Like 2
Link to comment
Share on other sites

I believe Arsenic is the main tracer element in Gold Basin soils too. I think lead maybe too, but I'll have to test it next winter. We have a commercial scale exploration project going on there right now with a ton of assays and drilling but that stuff is mostly mum at the moment. But I definitely look forward to doing a bunch of soil sampling next winter on my own, like you are doing. I had the same thoughts as you - zap everything with the XRF and then just send in the ones with tracer elements for assay when I'm just trying to outline ore bodies at first.

That's crazy on the fish. I hadn't thought about looking at them. Mercury over there or something? I'm kinda curious to see what trout have in them over here now, but being so far from the ocean I'm guessing they are fairly clean of chems?

I'm mostly using my XRF in Wyoming to identify rock types that evade tradition ID means since those odd rock types can themselves be tracers to things I'm looking for, just like Arsenic to gold, but I'm not really looking for gold here. In AZ next winter I should get some good field time with gold prospecting on this unit.

Link to comment
Share on other sites

We are mainly zapping the quartz reef around areas that I have traced by years of metal detecting or panning.

We mill the samples down to 20 micron and peletize them which makes a more homogenous sample and lessens the nugget effect like you were talking about. But then I leach the samples with two different types of solutions that I developed. I then dry the leached material and re-mill it, pelletize it and scan it again. This takes about 15min per sample in the field. And you can easily perform the leaching part of the sampling concurrent, its the drying, re-milling and cleaning of the milling gear prior to the next sample that takes up the time.

Because we are using a larger sample and concentrating it the results are much much more accurate, so much so that we can rely on them to make in field decisions.

Link to comment
Share on other sites

Ok, last one. This is on solid gold nuggets. Unfortunately, I tried every single calibration and when I zap a nugget the gold lines seem to overwhelm the copper. I have to assume some of these nuggets have a small percentage of copper in them at least? But it isn't showing up no matter what I do or try. I zapped a solid copper nugget and it read 100% copper, so it's detecting copper fine. So, if I want to look at solid nuggets, I probably need to have a calibration that can ID copper in high Au% matrix. Or alternatively, I suppose some of these might just not have trace copper? Anything I send to the refiner is always a ton of stuff melted together so I'm not exactly sure the assay composition of individual nuggets.

Mining is limited to 75% gold concentration, so I went with precious metals package, also wanted to see if any platinum, palladium, etc were present, so any quartz (Si) will not show up here is about the only caveat.

This is far less interesting to me since there aren't any real mysteries here to solve, but there are a few and some things to learn from it. But I did want to see if there were any hidden platinum group metals in any of these nuggets. A bit of rhodium or platinum here and there could add up quick! And I honestly have no idea if PGM's can be found alloyed in nuggets or not. Here are nuggets I tested:

image.png.285522764fdbb21fb899e16dd3840d9f.png

Sample A: 10 grams, Quartzsite slug & Sample B: 7 gram Quartzsite tangle w/small quartz grain inclusion

image.png.9db3ff47eb4d8c8af0654765587949e1.png

Hmm, not much interesting happening in Quartzsite. Just some very pure gold with few inclusions. I've always suspected the nuggets here were high purity, and it's why I don't sell these ones in my 10% under spot bulk lots.

A has more indication of travel than B does. The general theory is that nuggets enrich in gold at the surface with travel and weather, and you can see nugget A with more travel is exhibiting potential silver depletion at surface.

Sample C, GB 7 grams with many angular pyrite casts & Sample D: GB 36 gram slug 

GB nuggets commonly have a black coating or inclusions that are resistant to acids. I have assumed they were manganese but a geologist looked under a microscope and told me he thought they might be cryptocrystalline versions of iron oxides, and that the crystalline structure of the oxide is preventing the acids from reacting vigorously. So, I can finally figure this mystery out.

C has almost no indication of travel and many sharp, angular edges indicating pyrite or similar was once included but has weathered out. D is a large slug exhibiting signs of transport, and this piece is also older than sample C as it was found embedded into solid caliche about 6" deep in a layer of paleo gravels that I estimate are at least 800,000 to 1.5 million years older than the gravels which sample C was found within.

image.png.dba61ff1ddbf0ca4efea51662f6cb612.png

So it seems the geologist was right! The black, acid resistant inclusions are in fact iron based, and not manganese based as I had assumed.

Also, as with the Quartzsite nuggets, here in GB the nuggets displaying the most transport seem to be enriched in gold at the surface, and silver has decreased with weathering and transport.

Sample E: octahedral crystalline NV nugget, Sample F: typical crystalline NV nugget, Sample G: NV paleoplacer nugget

Sample E has several octahedral "grapes on a vine" (it's easier to see in person) and had distinct chevron patterns on the back. It was found on bedrock in a wash in Northern Nevada. It displays minor signs of transport. Sample F was found about 4ft downhill from a hardrock pocket of crystalline gold I discovered on a hillside and shows no signs of transport. Sample G was found very close to sample E on bedrock, but on an uplifted portion of bedrock and exhibits heavy levels of transport, and I am sure is a remnant from an uplifted paleoplacer even though the gravels have eroded away and left no evidence except occasional scattered quartzsite or granitic cobbles. All 3 nuggets came from within a 10 mile radius.

image.png.d739a731ff7d7f4d510d37e9f260fe30.png

There is an immediate correlation again, as with Q and GB nuggets, between gold enrichment on surface and evidence of transport. The inclusion of iron in these nuggets makes me wonder if the XRF is pulling over some copper results and interpreting them as iron. But I'm unsure with a control nugget with a known measurement to compare with.

Looking at the results, I'm even more sure now that Nugget G, the hypothesized paleoplacer nugget, actually comes from a paleoplacer that has simply eroded away. Transport alone is likely not enough to account for the much higher concentration of gold in this nugget, and it's likely it came from another area altogether, especially since it appears to lack Sn (tin). Though, the inclusion of iron makes me think it might not have travelled as far as I thought (the Sierras or maybe Idaho), but instead might have a more localized source within a radius of 50 miles or so within NNV itself. 

Nuggets E and F both came from a decent distance apart, but the inclusion of Sn seems to indicate they came from the same mineralizing event. 

So - these results are enough to give me a good idea of exactly how to use this XRF for prospecting this specific area of NNV for gold though. I'll be zapping soil with an eye for Sn as a potential pathfinder. And looking at more nuggets along the way to try to find more correlations or clues, as well as analyzing the rocks themselves where I've found pockets and looking at what accessory elements are in them, and using those as soil pathfinder elements as well. And this is how I plan to personally use this XRF for gold prospecting this summer in NNV.

Thanks for reading, congrats to anyone who made it through these walls of text. :cool:

  • Like 1
Link to comment
Share on other sites

Also, it seems that simply looking for increasing concentrations of silver in nuggets is the XRF equivalent of panning up a hill until you find the source. If you start finding a line where every nugget is increasing in silver concentration, then it may be a good bet that the source is or was at one time located somewhere up that line/direction.

  • Like 2
Link to comment
Share on other sites

15 hours ago, jasong said:

I tried every single calibration and when I zap a nugget the gold lines seem to overwhelm the copper. I have to assume some of these nuggets have a small percentage of copper in them at least?

Could it be a surface selection effect?  Are there chemical leaching processes that selectively remove the copper from the surface?  It would be interesting if you could cut a nugget and zap the resulting inside surfaces (but of course that would devalue your nugget, I realize).

 

16 hours ago, jasong said:

I did want to see if there were any hidden platinum group metals in any of these nuggets. A bit of rhodium or platinum here and there could add up quick! And I honestly have no idea if PGM's can be found alloyed in nuggets or not.

I suspect at least platinum would alloy with gold since mercury does, but I really don't know, either.  However, I think Pt will be hard to distinguish even with the high resolution of the gun's detector.  Note that atomic numbers of Pt, Au, and Pb are respectively 78, 79, and 82.  Your original post had a spectrum that showed the Pb L lines (about 8-9% separated in energy) being on the shoulder of the Au L lines.  The Pt L lines are only 2.5% to 3% separated from their Au counterparts.

 

  • Like 1
Link to comment
Share on other sites

All good questions - that I don't know answers to. :smile: But I think you are right that any Pt in a nugget may kinda "blend" into the Au results if it's present. For testing it's sorta a chicken or the egg the problem - I can see if it'll ID Pt in Au if I had a Au/Pt alloyed nugget, but I can't find out if I have a Au/Pt alloyed nugget unless I ID it first. So, I'll probably start acquiring some odd samples over time which already have a certified assay or XRF or XRD results as I run across them for sale, just so I can use them as control measurements for my gun.

It occurred to me last night before bed that to test the ability to detect Cu in gold alloys, I could just buy a ring, or actually I might already have some rings made of Black Hills Gold (high Cu) that I've metal detected over the years in some boxes. I presume the silver depletion is due to oxidation (or maybe reaction with environmental chlorine or acids from rains or leeched from chems in gravels/nearby ore?) and it's possible copper could undergo similar processes. But chemistry is my seriously weak point. 

I have no problem cutting into general nuggets, and I have a thin kerf diamond blade I could use for minimal waste. I just sold all the gold I had which I didn't specifically want to keep undamaged though, so I don't really want to cut into any nuggets I have right now since it's just a small handful of type specimens from each location which I'm going to eventually put in a display case. But I think I will do that next time I find a large-ish nugget 1/2 oz or so, big enough to take readings across from the center to the rim. See if I get a "tree ring effect" of increasing silver and maybe copper towards the center of it?

When I get a big handful of nuggets again with different compositions I think it might be interesting to find like 5 with the same weight and geometry more or less, but different alloy compositions. See how different gold alloys affect detection depth since the conductivity of them generally decreases. 

  • Like 1
Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...