A Basic Intro To Xrf Guns For Prospecting

[jasong]

XRF's hold sort a mysterious place on the shelf of semi-unobtainable prospecting equipment. 99% of prospectors don't need one. Maybe this post will help clear up some of mystery around these devices, and show where they can actually be worth the outlay of capital. And why for almost all recreational/hobby prospectors, they are not worth the money.

What does an XRF do? In very simple terms you point it at an object and it will tell you what elements are in that object. More on this, and why it isn't this simple, momentarily...

After sometime over 5 years of searching, I was finally able to find a used XRF I could afford to finance recently. These are not tools for recreation. They are expensive and require understanding how they work, what tasks you need to accomplish, and understanding the limits of XRF. The trick with these units is to find one with the proper calibrations already installed as they can be many thousands of dollars to send to the manufacturer to get configured correctly for mining/prospecting uses and to add/subtract elements or to calibrate for certain matrixes (silicates/iron/etc). X ray tubes and X ray detectors are about $6k each to replace, and recalibrations are about $1500 a pop, so even maintenance is crazy expensive. It's a tool you need be certain you need or can put to good use before buying one. And buying used, it's probably best to find one with as few hours use as possible to delay the inevitable tube replacement, as well as with a recent calibration certificate.

My unit is an XMET 7500 made by Oxford (now Hitachi). The more common units people generally see are the Olympus and Niton guns.

This unit has basically every mining calibration Oxford offered on it in addition to soil and other specialized mining related modes, which is very valuable and very useful for prospecting. It also detects down to magnesium without any fancy helium purge techniques. The guns sold on ebay with only alloy calibrations are pretty useless for prospecting without spending a lot of $$$ on additional calibrations. 

Some other things to consider are the machines themselves vary greatly between model numbers and some models may be unsuitable for specific uses in prospecting. A few things to educate yourself on are:

• Beam energy and detector type (determines if certain elements can be detected at all, and how accurately)
• Electrode composition (Gold electrodes have lower sensitivity to gold in ores, for instance)
• Calibration to light elements, or ability to detect certain elements

I don't think an XRF is particularly useful for people who are only looking for gold. Due to the electrode limitations, the PPM minimum to detect gold in ores can often be above what would be an economic (and thus desirable) concentration in gold ores. But, looking for tracer elements (stuff like Pb, Cu, As, Zn, etc) can be quite useful. It can also help outline buried ore bodies which can then be explored mechanically via drilling or other methods.

For prospectors branching out beyond just gold however, an XRF can be even more useful. And that's when one needs to understand the elemental limitations and what your application specific uses are. Any affordable XRF today will not detect lighter elements than magnesium. Some will detect to magnesium, but then do not contain calibrations to allow it (extra $$) and some require helium purging to measure light elements.

Elements like hydrogen, carbon, oxygen, and sodium are very common "rock building" elements. But XRF readings will lack these measurements. So, when a looking at a rock your readings will often give fractional (less than 100%) results. This is why - the missing mass is tied up in atoms lighter than magnesium.

Fortuantely, a lot of common rock types have unique fingerprints still in elements such as Mg, Al, Si, P, S, Cl, K, Ca, and Fe. But some don't. This is why it's important to understand what you are looking for first in the field, and then find a tool that is going to match your needs. Further, a lot of minerals in certain locations but not other locations will also have further fingerprints in other elements such as Cr, Co, Mo, Nb, certain compositions of rare earths, etc.

To make it more complex (this part took me a while to wrap my head around), each calibration within the machine may or may not be configured for some of these elements - even if they are within the range of detection of the machine! Like, an alloy calibration will have little use for silicon or calcium. Conversely, a mining calibration without magnesium or calcium may be next to useless depending what you are looking for. Of course, it costs extra money to add elements and even if you have for instance a precious metals calibration that includes platinum, the mining mode may not itself include platinum and that's more $. That is why the matrix matters, each mode can be calibrated to a specific matrix. Like mining modes are generally going to assume that the sample is mostly silicon, whereas precious metals mode might assume the only things that exist in the universe are metals. So if you analyze solid metal with mining mode it may misidentify elements thinking they have to be metals when they aren't, same as if you analyzed a piece of gold ore in precious metals mode where it will try to assign certain non-metallic spectra in the ore to something like gold or platinum, giving you false positives. This is why calibrations available and elements assigned to that calibration is so very important when it comes to XRF and accurate results.

Why else is XRF bad for gold-specific uses? (I emphasized this because this is primarily a gold prospecting site, even though I prospect for many other things myself).

First one needs to understand how XRF works - simply put it kicks a few electrons out of a few different orbitals around an atom at discrete energy intervals (these are spectral "lines"). When another electron falls into the empty orbital to replace the vacancy, another X Ray is emitted at this discreet energy. Unfortunately, some elements have some very close to identical spectral lines. Look here at some lighter elements and see the overlaps on this visible spectra chart that we use to ID elements in stars? Some might be familiar with these from astronomy or high school. Well, the same happens in the X Ray realm.

This is coincidentally why ionized gases look a certain color to us and how "neon" signs can be different colors (different elements inside the tubes). The same thing happens in the X ray spectrum, just not visible to our eyes.

Except when the X ray spectra is reaaaaaaally crowded around the gold lines. Making it hard for specific ID's when other elements with similar lines are also present in ore, and unfortunately some of the elements are also commonly found with and around gold mineralization. Combine this with the anodes on many affordable XRF's being gold which itself interferes with really precise Au measurements, and you can see why an XRF isn't the best tool for specifically gold prospecting.

Here is an actual XRF spectrum. You can see how very common accessory gold ore elements populate and crowd the gold spectral lines at various orbitals. And also how you might be missing critical lines if your X ray tube only goes to say 15kEV instead of 40kEV (EV stands for electron-volts), you might miss some Ag, Ru, Cd, or Zr fingerprints in this specific case.

Now notice how iron stands all alone? That's why some elements (iron) are easier for an XRF to ID than others like gold.

So for some such tracer elements in soils and ore, and identifying certain minerals which really can only be accurately identified via spectroscopy or thin sections as for some gems, an XRF can save months of time and thousands of dollars for in field qualitative assays to do first stage determinations, ie, wether a resource is simply present or not, ignoring actual concentrations.

This is why it's so important for anyone considering one of these units to know exactly what they are looking for first, to know the limitations of XRF, and to know if a unit will meet their application specific needs. Almost every company I spoke with had a story about a prospector, or even a few cases some junior mining companies, who purchased an expensive unit only to find it wouldn't work at all for what they needed to do. So hopefully this clears up a little mystery about XRF's and maybe saves someone from making an expensive $15k mistake.

I am by no means an XRF expert and everything I know is just self taught. So if I've included an inaccuracy then please correct me. This is not intended to be definitive, but just to share what I've learned over the years in a few pages of simpler to understand jargon for those prospectors interested in these devices.

More later with some actual measurements...

[GotAU?]

Great introduction, Jason.  I was familiar with using microprobe analysis with electron microscopes to identify elements in microscopic targets such as plankton shells for research back in the 80’s, (it was basically the same as XRF, but not a very portable version), and am amazed at seeing how portable the technology has become!  I don’t need a hand held XRF gun, but when those GPXRF 8000 mono coils are finally released to the public for doing lead discrimination, I will be one of the first wanting to pre-order it! 😁

[jasong]

I've been thinking about trying to build a homebrew scanning electron microscope for like 15 years. One of those things that will probably always be a month off in the horizon and never get around to though. Like cleaning out my garage. 

I was in school in the early 2000's and even then XRF's were totally beyond the range of affordability even for the university so we never could play with them, though we did have a SEM which was cool. The last 5 years or so have really brought units that could be used for prospecting onto the used market finally. It really feels like a Star Trek tri-corder type device to me in many ways, sorta unreal.

Discrim only works to a whopping 1/100th of a inch or so though unfortunately on the GPXRF 8000. 

 

[GotAU?]

51 minutes ago, jasong said:

I've been thinking about trying to build a homebrew scanning electron microscope for like 15 years. One of those things that will probably always be a month off in the horizon and never get around to though. Like cleaning out my garage. 

I was in school in the early 2000's and even then XRF's were totally beyond the range of affordability even for the university so we never could play with them, though we did have a SEM which was cool. The last 5 years or so have really brought units that could be used for prospecting onto the used market finally. It really feels like a Star Trek tri-corder type device to me in many ways, sorta unreal.

Discrim only works to a whopping 1/100th of a inch or so though unfortunately on the GPXRF 8000. 

 

Oh man, you want to make a SEM too?!  I saw that project on hackaday and it was on my bucket list too, but have you seen the scanning tunneling microscope? That would be a really cool build, looks easier than the SEM also.  Pretty cool to be able to get a image of atomic structures up close like that, hu?

 

You have to see the author's image gallery, too - examples of gold atoms and crystals there too:

https://dberard.com/home-built-stm/image-gallery/

[jasong]

That's cool, I saw a similar one except it was on a forum back in 2003 or so that dealt with building coilguns and railguns and a guy posted a SEM DIY tutorial. These newer ones look like better designs, and easier. I actually have most the stuff in the SEM guide other than the HV Arduino shield, I even have that finite element analysis program they are using, it's not really needed though.

I've been thinking about making videos again except not prospecting/detecting since so many people are making those nowadays. I thought it'd be fun to build amazing stuff like this that few have heard of, out of stuff laying around. My first project is going to be building a laser out of common household trash I find laying around McGuyver style (semi cheating since I've done this before out of 50% trash basically so I have a rough idea ahead of time what to look for), then I might look at a microscope.

Those crystalline gold STM photos are cool. I wonder if cryptocrystalline structure pattern analysis could be used to track/match placer gold to it's lode source. I don't think anyone has ever studied that, mostly they just do elemental fingerprinting.

[Steve Herschbach]

Great post Jason, Thanks! 

[Valens Legacy]

Great post and some really good information with the reading of this subject.

At one time I had given it some thought about one of those, but now see the reasons not to get one. The way I have banged up and dropped some of my tools, I would have it out of calibration in no time.

I think I will just learn these detectors my grandfather left me and be happy once I can find time to get out and hunt.

[rvpopeye]

Verrrrry interesting......thx for posting it up.

[Gold Hound]

Hi Jasong

Our company owns an olimpus vanta. I have also used the Nitton gold3+ too. It costs $50 000 aud + extras so it a bit out of reach of the hobby prospector.

For gold exploration you need the Rhodium xray tube as the accuracy with the other tubes (tungsten or silver) is very low on Au and PGM's. And a lot of the gold areas we are chasing have mostly free gold and little of the pathfinder elements accociated with them. Although the pathfinders are always elevated near the deposit in relation to the backgound norm for the areas, they may only have a small halo and this may only be slightly elevated. This is probably due to the large amount of water we get in such a little time here with our tropical wetseason. Which can render the pathfinders of little use to the explorer up here in parts of northern Australia. But in the more arid areas they are of much greater use. Our xrf has the gold and pathfinders suite but we also got the rare earth and base metals suite's too. But each program is expensive and an added extra.

A suitable xrf is an essential tool in a modern technologically advanced exploration company. We have developed 2 special ways that we prepair our samples which increase the accuracy of the scanner and allow it to accuratly detect down in the lower end of its recomended minimum ppm. Our results closely mirror our assay results since developing these methods. Which further increased the usfulness and reliability of the xrf results for field use.

The scanner saves the savvy user 1000's in assays and weeks in waiting time for assays which can be a real pain if you are in a remote location, as you may have to return after good results to resample a hot area further. It is much easier to be able to make on the spot decisions on the viability of further samples and when a hot area is discovered you can concentrate your efforts on that area rather than just taking grid samples and sending them to assay. 

There is also another technology that is of use to the modern gold explorer it is called LIBS. LIBS as an Emerging Analytical Tool for Mineral Exploration - SciAps
https://www.sciaps.com/newly-published-research-libs-as-an-emerging-analytical-tool-for-mineral-exploration/

I am are looking at ways of increasing libs usefulness in gold exploration. Its limiting factor is that it only scans a very small area which can give you inaccurate results but Im trying to addapt the same technology but from a different sector where they use it in a different way that should prove very useful when addapted to exploration.

 

[GB_Amateur]

Nice writeup, Jason.  The one thing you didn't emphasize (at least I didn't see it) is that for typical materials the depth the measaurements are made is quite shallow -- effectively surface effects for many materials.  It varies greatly with atomic number of the material as well as the energy of the impinging X-ray.

Most people are familiar with X-rays due to their medical diagnostic applications, particularly in dental use.  Those tend to be X-rays at  higher energy (approaching 100 keV).  Also skin and bone are composed of low atomic number elements and thus easier for X-rays to penetrate.  If you recall seeing dental x-rays, the metal fillings (e.g. silver amalgam from the 'old days') show up clearly but nothing behind them.  Another 'advantage' of medical x-ray diagnostics is that they are monitoring transmission only.  (That's why they put the detector -- used to be film -- in your mouth!)  The XRF guns work by transmission + absorption/excititation + de-excitation + retransmission, with each step robbing the process of efficiency.  And then they have to sort out signal from noise and measure the energy of the few (relatively speaking) relevant signals.

There are parallels between how these guns work and how a standard metal detector works.  The strength of the magnetic fields from the target in the ground (what the detector relies on to sound off) is tiny compared to the magnetic fields the detector transmitted initially.

It's pretty impressive how well these hand-held devices work, given all the sophisticated components (including software) it takes to run them.  The progress made in the last decade or so is analagous to the evolution of the hand-held computer (aka cell phone) from the desktop computer.