pXRF for Agricultural Soil/Plant Tissue

[Chris Peterson]

Good Morning Everyone,

Long time reader, first time writer here. I currently work at an agricultural lab that tests soils, water and plant tissues (mainly Corn/Soy Beans) in the US. My question is, are there any others in this group that do this, either as their main focus or on a contract basis?

We have the Tracer 5i and 5g and I am the sole operator. I started last year, being unfamiliar with XRF or agricultural testing as a whole, so have delved deep into publications, XRF Guru guides and other materials to educate myself. We use a Bruker provided Plant Tissue calibration and a custom Soil calibration I created on my own using EasyCal and I would love to be able to exchange PDZs to verify the accuracy of my calibrations with similar users. 

We use ICP-OES in house, but we do not do the full digest, rather an abbreviated version that I cannot use for standard creation/validation. 

If it would help, I do have NIST standards for Soils and Plants (2710a/2711a/1646a and 1515/1573a) which have Certified concentrations that I have used to create my calibrations. 

I understand most in this group work in archaeology/conservation, so just putting this out there. 

Thank you all in advance, 

Chris 

[Lee Drake]

Hi Chris,

Sending some publications and contact info over for folks I've collaborated with who may be able to help:

• Dr. Erick Towett (consultant): eto...@gmail.com. Has worked extensively with XRF & AI in plant, soil, and fertilizers with multiple publications for all of the above
  
• Dr. Marian Hamitlon (University of Northern Colorado): marian....@unco.edu. Uses the Tracer 5g to reconstruct ecology using plant tissue measurements
  
• Dr. Robert Brent (James Madison University): bre...@jmu.edu. Developed "wet" calibrations for measurement in living environments. 
  
• Dr. Yadav Sapotka (US Army): yadavs...@gmail.com. Used machine learning to calibrate for properties like moisture in fertilizers. 
  
• Dr. Lee Kalcitis (Washington State University): lee.ka...@wsu.edu. Used XRF to estimate nutrient content in live fruits. 

Hope this helps! The big challenges I see are a) measuring live vs. prepared samples and b) sticking to simple empirical cals (ala Easy Cal) or going wild with AI. There's a lot of navigable space here, but the limiting factor is the development of calibrations. I suspect this is best done at the community level, because right now it seems most people are having to re-invent the wheel for each application. 

- Lee