Can you just ignore a coelution?

Discussions about IC and related topics

47 posts Page 1 of 4
Carbonate and nitrite coelute on the ICS-2100 I operate. With 27 mM KOH they can be separated at low levels but at high levels they merge completely. I've been running it at 23 mM for years and that separates them well. But now my new boss told be me to run it at 30 mM so they would merge completely in both low and high concentration standards. That would work if there was no carbonate in the samples, but some samples have quite a bit of carbonate.
Google Dilbert's pointy-haired boss.
One of the guys who used to work there said her arrogance is only surpassed by her incompetence. So naturally they made her a supervisor. I complained to her boss, the chemistry chief. But she was a radio chemist and she knows nothing about chromatography. She's just ignoring it. It will give bad data but we would be very unlikely to get caught. The federal inspector never looks that closely at chromatograms.
We had an R&D reorganization in the mid-1990s, and a woman who had been in charge of sensory evaluation was made manager of Analytical Chemistry. When this was announced at an R&D meeting, I struggled to refran from falling to the ground and feigning a fake heart attack.

She met with us and actually stated "I don't know a darn thing about analytical chemistry". She lasted two years, very tough time for us in the department. Several were sent to outside "counseling" for which she did a pre-emptive strike by talking to the counselor !! I was one of those folks who "needed" counseling; HR told me I'd never win against management, advised me to go.

I was cured !!!
Image
start reporting your results as the sum of nitrite and carbonate... then at least you're not giving away wrong numbers. Ask the QA for a correpsonding change of the specifications because of this method change.
Maybe your boss will find a way to convince them too
That's an idea. I can put an attachment on the results. My boss will see it and hit the roof though. But she can take it off and then I'm covered ethically. We had a brutal situation at the lab 10 years ago where a couple of the toxicologists were having trouble with an instrument and blamed their lab tech. They even went to the local papers and accused him of unethical behavior. He got the boot. You can still find it when you google his name.
Ignore it? No, that would add false positive area counts to Nitrite.
You need a CRD, this is a common problem

https://assets.thermofisher.com/TFS-Ass ... 441-en.pdf

You can also right click and choose split peak to integrate correctly.
Also may be time for a new column
The instrument already has a CRD. It takes out a bit more than half of the carbonate. It's not needed at all if we run at 23 mM though because everything separates so well. It's the EG concentration recommended. The certificate that comes with the AS-18 column shows what the chromatogram looks like at 30 degrees, 1 mL per minute and 23 mM. It's the same as on pg 33 of this instruction manual.

https://assets.thermofisher.com/TFS-Ass ... 878-en.pdf

At 30 mM there is no way to accurately split carbonate and nitrite. I spiked a couple of samples with 0.1 ppm nitrite and the peak was a long blob of merged carbonate and nitrite. At a 1 ppm spike it looks a bit more like a normal peak but recovery is 130 - 140% due to the added carbonate. My boss has even come up with an unethical way of dealing with the failed spikes. She wrote into the SOP that when spikes fail you make a dilution of the sample and spike that dilution. A 1:10 dilution reduces the size of the carbonate so much that the added area will not cause the spike to fail high.
anionman wrote:
The instrument already has a CRD. It takes out a bit more than half of the carbonate. It's not needed at all if we run at 23 mM though because everything separates so well. It's the EG concentration recommended. The certificate that comes with the AS-18 column shows what the chromatogram looks like at 30 degrees, 1 mL per minute and 23 mM. It's the same as on pg 33 of this instruction manual.

https://assets.thermofisher.com/TFS-Ass ... 878-en.pdf

At 30 mM there is no way to accurately split carbonate and nitrite. I spiked a couple of samples with 0.1 ppm nitrite and the peak was a long blob of merged carbonate and nitrite. At a 1 ppm spike it looks a bit more like a normal peak but recovery is 130 - 140% due to the added carbonate. My boss has even come up with an unethical way of dealing with the failed spikes. She wrote into the SOP that when spikes fail you make a dilution of the sample and spike that dilution. A 1:10 dilution reduces the size of the carbonate so much that the added area will not cause the spike to fail high.


Is there a justification for going to 30mM? If you are making your own eluents then I would complain to higher up that doing so cost more per sample due to increased reagents to make the higher concentration.

Sometimes when you can't win the fight with logic or reason, getting the bean counters involved can help :)

Also a good C.Y.A. folder to store any emails where you brought up the problem helps protect you if there is ever an audit. Tell her you will only do it if she puts it in an email.
The past is there to guide us into the future, not to dwell in.
Her justification was that it separated Nitrite and Sulfate better. But at 23 mM they are separated by a minute and that has been good enough for years now. I did have her put the order to go 30 mM in an e-mail before I would do it. My office-mate and I went to the Chemistry Chief and complained that reporting results with coeluted carbonate-nitrite would be lab fraud that we would have to report to the QA specialist and trigger a full investigation. Our ethics policy states that lab fraud accusations constitute an emergency. It didn't go that far. The Chemistry Chief told us to to go back to 23 mM....

But now she insists on using auto detect for all sample integrations. I have shown in the past that doing that leads to high recoveries due to baseline issues. I might be able to program a low rider peak threshold into the software to compensate. I wrote a report on that and gave it to the Chief 3 weeks ago but it is still sitting in her mailbox...
Update: My boss is now threatening to have me fired for disobeying orders. I told her I am ethically obligated to push back against unethical directives. Then she told me that ethics are arbitrary. Really. I reported that to the QA Specialist and program chief. I don't know if our QA Specialist knows anything about chromatography. She has only been on the job for a few months.
anionman wrote:
Update: My boss is now threatening to have me fired for disobeying orders. I told her I am ethically obligated to push back against unethical directives. Then she told me that ethics are arbitrary. Really. I reported that to the QA Specialist and program chief. I don't know if our QA Specialist knows anything about chromatography. She has only been on the job for a few months.


Where I work ethics are strictly enforced. Ignoring them is serious enough for dismissal. A mistake is forgivable, blatant disregard will see you to the door.

If you are regulated by FDA or EPA or any of your results can be reported to any government agency, then ignoring ethics can not only result in dismissal but also jail time.

Manual integrations are allowable if you have a SOP for manual integrations and you strictly follow it and have a sign off to show it has been reviewed and meets policy. It is better though if you can adjust the auto integrate parameters to consistently pick up the peaks correctly, as it leaves less to question, but never change them between samples as that can be construed as fraud just as bad manual integrations are.
The past is there to guide us into the future, not to dwell in.
The lab is EPA certified. I had a meeting with the QA specialist and Chemistry Chief the other day where I told them that my supervisor doesn't seem to understand the ethics of the situation. A couple of weeks ago I saw her at the instrument trying to figure out a way for it to narrow the coeluted nitrite-carbonate peak without manual integration so the nitrite spike would pass. I mentioned that in the meeting. The spike will pass if you SEPARATE them I said. That's the point of chromatography - to separate analytes.

We have a manual integration policy, but I haven't had to do many because I set the integration parameters to compensate for problems. I did experiments where I spiked low level standards with high amounts of chloride and sulfate to see how it affects the nitrate and nitrite peaks. Autotetct wasn't integrating those properly so I tweaked it. There are a few ways to do it. But now my boss insists on using the basic autodetect parameters because she thinks that more defensible ethically. It will lead to more failed spikes and qualified results, but at least it's not lab fraud.
anionman wrote:
The lab is EPA certified. I had a meeting with the QA specialist and Chemistry Chief the other day where I told them that my supervisor doesn't seem to understand the ethics of the situation. A couple of weeks ago I saw her at the instrument trying to figure out a way for it to narrow the coeluted nitrite-carbonate peak without manual integration so the nitrite spike would pass. I mentioned that in the meeting. The spike will pass if you SEPARATE them I said. That's the point of chromatography - to separate analytes.

We have a manual integration policy, but I haven't had to do many because I set the integration parameters to compensate for problems. I did experiments where I spiked low level standards with high amounts of chloride and sulfate to see how it affects the nitrate and nitrite peaks. Autotetct wasn't integrating those properly so I tweaked it. There are a few ways to do it. But now my boss insists on using the basic autodetect parameters because she thinks that more defensible ethically. It will lead to more failed spikes and qualified results, but at least it's not lab fraud.


Well, you also have to be careful just using the default parameters. If the default parameters will give a bad integration (such as adding two peaks together) that will make the results pass when they should fail, that can be considered fraud because you can make adjustments to make the integrator more accurate. But it is only fraud if you know the default is working in your favor but not doing anything about it because it makes things pass.
The past is there to guide us into the future, not to dwell in.
The default parameters definitely aren't best and I have run test that prove it. But I'm not sure the results are far enough off to be fraud. I remember seeing a result that measured 0.7 ppm Nitrate that was really a bit under 0.5 ppm when integrated properly. On the other hand letting Nitrite and Carbonate coelute leads to ridiculous results. A 1 ppm Nitrite result could read 2 ppm or more because of the high carbonate we have in some samples. A low level Nitrite around 0.1 ppm could be totally buried under carbonate. We could say it's just carbonate and report no Nitrite or report 1 ppm Nitrite if the carbonate peak is large.

The news is that I have been taken out of chromatography for now and put on lab tech duties. Sticking labels on samples and doing hardness titrations. I'm not confident. At my last job where I was running a cafe I reported a couple of people for payroll fraud and I was the one who got kicked out and then laid off...
47 posts Page 1 of 4

Who is online

In total there is 1 user online :: 0 registered, 0 hidden and 1 guest (based on users active over the past 5 minutes)
Most users ever online was 1117 on Mon Jan 31, 2022 2:50 pm

Users browsing this forum: No registered users and 1 guest

Latest Blog Posts from Separation Science

Separation Science offers free learning from the experts covering methods, applications, webinars, eSeminars, videos, tutorials for users of liquid chromatography, gas chromatography, mass spectrometry, sample preparation and related analytical techniques.

Subscribe to our eNewsletter with daily, weekly or monthly updates: Food & Beverage, Environmental, (Bio)Pharmaceutical, Bioclinical, Liquid Chromatography, Gas Chromatography and Mass Spectrometry.

Liquid Chromatography

Gas Chromatography

Mass Spectrometry