AN ANALYTICAL XRAY SERVICES LABORATORY
Feel free to call us: 940-784-3002

Clay minerology is a fascinating subset of the common powder XRD applications we work with. It seems like every lab has their own unique way of handling the various challenges presented by clays and hence the need for myriad solutions. Most recently we worked with a lab which processes large volumes of clay samples in their various Bruker XRD system and they needed a sample holder that would accommodate their 26 x 26mm glass slides on which their clays are mounted. The solution was a custom holder with a square recess for the plate which is simple enough, but in cases like this, one must take into account the real-world challenges of machining as well as usage. The edges and corners of the glass slides are the least precise part of the plate so the holder has a “moat” around the perimeter to eliminate binding or shifting due to those issues. The center is also relieved as this area does almost nothing to improve precision of the mount, but will cause dramatic displacement errors if any debris gets between the holder and plate in that area.

These were a little time-consuming to make, but the end result worked beautifully. The client was very happy and we’re happy to have another design to offer to other labs with similar needs.

 

Our lab sees a very wide variety of sample types and they are almost always 90% unknown materials. So while we’ve always offered elemental analysis by WDXRF, sample preparation has always been done by the least destructive method. That means powders and solids get crushed, ground, and pressed into pellets and liquids either get run as they are under a He flush environment or dried and filtered for particulate matter. These techniques work well for qualitative analysis and can also be extremely effective for quantitative work assuming proper preparation and a good calibration curve.

However, we’ve been working on a comprehensive calibration curve for geological materials recently and the simple fact is that the very nature of these materials makes them very difficult to measure in their natural state. The fact that the various elements may be combined in any number of different minerals as oxidized, reduced, or metallic phases makes analysis by XRF, even WDXRF very difficult. The solution to these problems is Li-Borate fusion.

I’ve resisted getting into fusion for a number of reasons, not the least of which is the substantial cost of entry if you’re going to be running much volume at all. It’s not hard to spend $70-90k on the tools for an automated preparation setup. After researching all the major manufacturers we came out with a few front-runners that are worthy of note. These are only the models we looked at so please don’t consider omission from this list as a negative comment on any company or model.

Katanax was our first on our list as a few of our clients are running these systems. The electric oven style heating is very convenient and the fully-enclosed design was also quite attractive. They also have three different models with single, up to 3, and up to 6 positions for simultaneous fusion. The touch-screen interface is large, well-laid out and easy to work with. While all this sounds great, the best part about Katanax in my estimation is their support. Talking with their applications experts was a great experience and I could tell this was not just a job, but something they truly enjoyed. I recommend Katanax for anyone new to fusion who might need some help. The only things we didn’t like about the Katanax models was the “swirling” motion style which is a simple tilt back and forth and the electric oven heating. With most crucibles, the swirling motion won’t have any effect on results, but I was partial to the 360-degree rotation style. As for the electric oven, this works great for labs with higher volumes, but we aren’t expecting large volumes of samples any time soon and the idea of heating the entire oven for just a couple samples would have ended up being time to consuming. If you know your volumes will always be low, the single-stage K1 Prime.

 

Another model recommended by several of our clients was the Phoenix sold by Premier Lab Supply. Premier deals with several different models of fusion systems with gas and electric options, but we were focused on gas by this time. The Phoenix had a few unique features like O2 injection and a mechanism by which a non-wetting agent tablet could be introduced very late in the fusion step. This is necessary with some non-wetting agents and a very nice feature to have. The Phoenix is also expandable and we liked the fact that one can specify which positions are to be used for a particular fusion so all the burners don’t need to run if we’re just running a single sample. One of our clients at USGS has been running the original Phoenix models for many, many years and I’ve heard nothing but glowing praise for it. Again, reliability and the availability of support are extremely important to us and I believe Premier would have been a great choice. The age of the Phoenix 1 was a small issue and the interface is a little dated, but it really came down to physical size. This is a big machine and would have taken up a fair bit of our sample prep area. I’m quite confident that any lab would be well served by one of these.

 

No discussion about Li-Borate fusion would be complete without a nod to Claisse. They have options for just about everyone, but we were looking at the Claisse M4. The M4 is an older model, but offers some very nice features. Their swirling mechanism is easily my favorite as the crucible rotates continuously. The machine is also remarkably space efficient for being a 3-position system. The interface was the weak link for us compared to the newer models with touch-screens. Another minor issue that I’d never considered till we actually started running a fluxer in-house was that it’s very convenient to fill the crucibles from weighing boats while they’re installed in the machine. This would be difficult to impossible on the M4. I would expect a very high level of experience and support from Claisse, but didn’t actually get that far in our conversations. I should mention that we’ve seen several of the Claisse Ox machines out and about. They seem to be a great solution for high throughput labs.

 

 

Last, but certainly not least, we have the Nieka G4 series (sold by Chemplex in the US and pictured at the top of this article). This is the unit we eventually settled on though I had never heard of it until recently. It was simply the right machine for us with a 4-position configuration, bright and relatively easy to use touch-screen and an open design. We actually ended up with a propane version which was another last-minute choice that we were very happy with. This allows for very easy installation, portability (hard to believe that’s a feature, but in our evolving lab, it’s great to know that we can move it anywhere we want with ease), and a higher energy density fuel than natural gas. The swirling motion on the G4 is a circular motion which works very well for mixing up hard-to-dissolve powders. The integrated software offers a very high degree of control which may or may not surpass the others shown here, but we really appreciated this while we refining out fusion recipe. Extending the time for a certain step, adding steps, etc were all very easy.  No PC or other interface was needed or even wanted. The open design makes working with the crucibles and loading flux/powder a breeze. If there is a caveat for the G4, it’s that while Chemplex has never failed to get us an answer to any of our applications questions, this definitely seems to be a new branch for them. I assume it will take some time to build up an in-house knowledge base, but we didn’t need too much hand-holding to get started. For users with more consistent materials, I would recommend you take them up on their offer to develop a fusion method for you. This can be done entirely remotely and would probably have been a great option for us if we didn’t have such a large variety of materials coming through.

We’re probably about 100 fusions in and the machine has not missed a beat or given us the least bit of trouble. The amount of control available is likely a large part of this. We were able to fine-tune the oxidation, fusion, pouring, and cooling phases with multiple steps for each which has allowed us to create a recipe which covers a wide range of geological materials from shales to silicate rocks.

 

Update 23-MAY-2018:

We actually had our first interaction with support for our G4. One of our crucibles cracked during the heating stage. I’m not sure what caused this and neither was Chemplex, but it appears to have been some type of stress as it was right on top of a bulge. Getting a replacement was effortless and fast though. Having good, local support is another factor in this decision that was very important to us and I’m happy to say that our first issue was taken care of very quickly. Thanks Chemplex!