Third party testing results are all over the place

Posted by titrationclub in COA and Purity Testing

Man I am so frustrated right now. I sent out samples from two different vendors for third party purity testing. Same compound, same batch numbers, everything should be identical. But the results came back wildly different. Like, one report says 98% pure and the other says 92%. How am I supposed to make sense of this for my research? It is not like these are cheap tests either. I am trying to be diligent and ensure good data for my studies, but this just throws a huge wrench in everything. My protocol relies on knowing the exact concentration. Has anyone else experienced such a big discrepancy with their third party COAs? I am at a loss for how to proceed. Do I send it to a third lab? That is just more time and money.

7 Comments

kineticmind

Yeah, I have definitely seen this. Sometimes it is the testing method itself, different labs use different techniques. Or even just how they prepare the sample.

grayowl0

Could it be a calibration issue with one of the labs? That happened to me once, a lab had some equipment that was not properly calibrated and it skewed all my results.

bench2146

I had a similar situation where the purity was off by a few percentage points, but 6% is a pretty big difference. I would definitely question the lab that gave the lower number.

biohackguy72

This is why I always get a small sample first and send it for testing before committing to a larger order. It is an extra step but saves a lot of headaches later.

ambernotes

I always try to use the same lab for all my testing if possible. It helps control for some of those variables. Consistency is key for research.

titrationclub

That is a good point about calibration. I did not even think of that. I will reach out to both labs and ask about their methods and calibration schedules. Thanks for the idea!

cyclegal27

Also check the dates on the COAs. Sometimes vendors will provide older ones that do not reflect the current batch. Not saying that is what happened here, but it is worth a look.