Issues with ThermoScientific consumables

[Jdaniels]

I run method 314.0 for perchlorate in groundwater. For the last year our results have been irreproducible including standards and laboratory reagent blanks made in DI. I recently talked to others experiencing the exact same issue.

We all run different IC systems, columns and methods. We are all seeing this issue from inject to inject out of the same vial. Now the large contract labs that we ship samples to are also experiencing the same problem. None of us can get accurate or precise results (which goes over really well with the regulators). We often send samples to multiple labs for analysis, and they rarely match. Sometimes off by an order of magnitude. I believe that Thermoscientific has changed how they make their consumables. We are all going through columns, suppressors and elugen at much higher rate than when Dionex manufactured the items. Is anyone else seeing a change since ThermoScientific took over?

[Chris Pohl]

At first glance it might seem plausible to guess that something is changed with Dionex consumables since Dionex was acquired by Thermo Fisher Scientific. However, I can assure you that nothing has been changed with regard to any of the consumables you mention. The same people are following the same manufacturing processes at the same location using the same materials and adhering to the same quality standards. The only thing that has changed is the name on the box.

Perhaps if you supplied more specifics, I can help troubleshoot your application.

[iceyuser]

I can attest that the AS11 column doesn't work at all any more for the organic acid test that we had previously run for over a decade. Column heating helped, but only so much.

[Chris Pohl]

iceyuser,

Could you be a bit more specific as to what the issue is that you are seeing with the AS11?

Chris Pohl

[iceyuser]

Chris,
Specifically, we've been measuring mandelic acid and 2-methylhippuric acid. As I said, this test worked OK for years. Then one lot of columns came in and the separation of these analytes was terrible. You couldn't discern two peaks at all when injecting 50 uL of 0.4 mg/L. We were in touch with customer service. Everyone said nothing had changed. But we waited around for a new lot of columns to come out and the problem persisted.

[Chris Pohl]

iceyuser,

OK. As they said, we haven't changed anything (at least not intentionally) but still your report suggests something is amiss. Our lot validation process involves more than 30 analytes but it doesn't include either of your analytes so it's possible we've missed something in our process. We'll check to see how the selectivity compares to manufacturing reference columns from earlier lots. Can you supply the serial number of the last column that worked and the two lots that didn't? If you like you can email me directly at chris.pohl@thermofisher.com

Chris

[iceyuser]

The last of of columns that worked OK for this analysis was 011-15-061 (SN 019577), the lots after that that didn't work were 011-33-189 and the others I only recorded the last three digits 189A and 129, but were from the same time period.

[Chris Pohl]

OK, I'll have someone check into this and let you know what we find.