Can you just ignore a coelution?

Discussions about IC and related topics

47 posts Page 3 of 4
She did it AGAIN. This time she manipulated the instrument so it wouldn't take the integration settings for the calibration standards. She showed me her calibration and I saw that she had to take the high level nitrate and sulfate standards out of the curve. I saw that was because the low level calibrators were integrated as drop perpendicular separations and the high ones were valley to valley. The samples were also integrated valley to valley so in my opinion the analysis was invalid. Interrogations for calibration standards are supposed to match the samples. I didn't think it was possible to make the instrument do it any other way but she found it. I was trying to look at what she did but she kept badgering me and saying "I didn't ask you to touch anything!" I told er there was something seriously wrong with the way the instrument was working, but she said she liked it that way. So I walked away and e-mailed the chief and told her I would stay late to find out what happened.

After the ex-boss left I looked and saw how she did it. Instead if updating retention times with a new column she manually assigned the peaks. Then afterwards she updated the retention times. Somehow that allowed the instrument to ignore the programmed integration parameters for the calibrators. I brought this up in the morning meeting and said I could not sign off on the data unless it was reprocessed. My ex-boss was not happy. Someone else said to her that I knew more about the instrument than her and she said 'I disagree." The chief finally came down on my side and told the ex-boss to reprocess the run.

Then late in the afternoon I was reviewing some data at my desk and the ex-boss comes over and asks when I'm leaving. I said 'In a few minutes.' Then she gave me a nasty glare and said "Because I'm not leaving until you leave." I said I would in a few and after she went away I left. I still have not seen any raw data she has generated. I looked in our data management system and see some results that have a failed sulfate spike, so I suspect some integration parameters are still wrong. With a new column and a low level sulfate it should not fail a spike.
I'm working in pharma testing. When staff are caught repeatedly ignoring SOPs or perpetrating fraud, they generally get fired. Is anything about your anion analysis covered by SOPs tightly enough that you can lean on them to defend the correct way of doing things?

For example, manual integrations are "strongly discouraged" in our SOPs and QC shemists are required to discuss same with QA before reporting anything that has been manually integrated. Part of QA's data review specifically requires that the chromatograms are examined for consistency of integration.

All I can say is that if it looks fishy, don't sign off on it!

PS - yes, we're still reading. :salut:
Thanks,
DR
Image
The problem is that SHE would just rewrite the SOP on a whim and she didn't even care if it didn't comport with the method. We must follow the methods. At an ethics training I asked what if our procedures don't match the method? The guy said tell the bosses to change the SOP. I asked what to do if they refuse to change the SOP to match the method and he said he didn't know what to say.

Our SOP just says manual interrogations are allowed on occasion, but no specifics. I'm pushing to disallow manual interrogations unless changing integration parameters have been tried first. The software has a lot of parameters that can be tweaked to take care of baseline and co-elution problems and if you do it that way you assure that all samples are being treated the same way as calibrators which is required 'as much as possible'. I have not manually inegrated any Qc in years. Back when we were running at 27 mM (at my ex-boss's instructions) and before I was good at using parameters I manually split nitrite and carbonate a few times. I finally moved to 23 mM and solved that co-elution problem Just manually integrating a QC sample to make it pass looks bad, especially the way she did it to include an obvious second peak. And there are so many ways to work with parameters I think it might never be necessary. The problem is that I'm the only one who knows how to work with the interrogation parameters on the instrument. I've tried showing another analyst but she seems flummoxed.

Our data review specifically requires that the chromatograms are examined for consistency of integration too. We had to put that in a several years ago when we failed a proficiency because the ex-boss did it and didn't look at the baseline and see that it was wrong. After that we wrote into the SOP to check, but she wants to ignore that. She thinks the instrument is just like a big pH meter where the correct answer just pops out. She says just put it on auto detect and let the machine think for you because that avoids user bias. But we want bias in favor of correct results. I validate every change I make to parameters.
Now my coworker is so stressed out that she says she's going to retire. She was on the verge of tears this afternoon after work and said she can't take the stress anymore. I said we could complain to HR about the situation. It's a dysfunctional and hostile work environment. She said she didn't think that would help.

She's worried about what happens next time we get audited if the fraudulent co-eluted data gets found. Her name is on some of it. But we were under strict orders to obey or get fired and the chief was informed and did nothing for a week. If someone is going to get in trouble for what happened it should be the chief and the ex-boss, not some poor bullied worker who is afraid of losing her job. I told her to hang on a while because it can take a long time to get these things worked out. The exec at my last job finally got fired after years of incompetence and corruption. It's government work. It's hard to fire people from government work. But it's still no fun when they try.
I tried talking to the conflict resolution guy a couple of weeks ago but I said I was constrained by threats of being fired for talking about ethics because of confidentiality. I was told that talking to him is confidential and told to call him again. I called him again. I said the boss told me to talk to him and he said that put him in a difficult position. I said I'm just trying to do my job. He said there are different ways of doing jobs that can be mediated. I said we are scientists and we have to obey strict protocols and methods so that leaves little room for negotiation or mediating. It sounds to me like the chief just doesn't know how to do her job. This guy can't do her job for her.
"There's a right way to do this that a certain person is refusing to follow.
Every other way of doing it will eventually be bad for the company. You now know what you have to get done."

I'd explain the specifics of the issue, and than say something like the above to the chief and the resolution guy in a meeting that excludes the problem person. After that, document the meeting and walk away from the problem. You've done what you can and should.
Thanks,
DR
Image
The convict resolution guy doesn't know anything about chemistry. He's useless. The exec and QA manager told me they were going to hire an expert in ion chromatography months ago after the initial ethics complaint I made. They never did. And now they won't let us optimize the integration parameters. I have the metals manager on my side now too, but that's not helping.I can't walk away from the problem without quitting the job. I'm not going to be goaded into signing off on something improper. It could end up in conflict with QA. They have never done any chromatography so why are they telling me how to do it?

The other week the fluoride reporting limit failed for a run because the nearby formate peak shrank by the end of the run. All that needed to be done was lower the minimum peak area so it would see the formate in the reporting limit the same way it did in the calibrator. It took a few seconds. But we had a meeting about it with the QA people and even with THREE chemists saying optimizing integration parameters is ok it was denied. All the fluorides had to be rerun after another calibration. I said what if this happens with time sensitive analytes like nitrate and nitrite? Then we would have to call the customers and ask them to send in more samples to rerun. That would be a disaster. People would just send their samples to a lab that knows what it is doing. I said that at the end of the meeting and said "I know what I'm doing."

The good news is that we had a meeting with no QA people and we agreed to change the concentrations of the calibrators to make it more logical. The ex-boss created the calibration concentrations years ago so that there are 3 separate solutions we have to run for reporting limits. I came up with something better. She was trying to criticize my calibrators, but the metals boss was there and said that it was settled and we were only deciding whether to weight the curves. That got the ex-boss mad and she stormed out of the meeting. I gave the metals boss a big thumbs up.
More good news: the ex-boss is out of chromatography for good. She's actually out of the entire accessioning-anions -physical chemistry team. Since the physical methods and anions involve time sensitive samples there is a 3 person team for that and we rotate duties. Now it's a 2 person team with some help from a radiochemist. We don't get audited for another year. The ethics investigation should be on record and should get reviewed. I wonder what will happen.
More news. My ex-boss retired. She's gone. But now my co-worker who was in tears made it official - she is retiring. Now I don't have enough help. If I take days off samples won't get tested. The radiochemistry guy isn't always around. They've posted a job and hopefully we can get someone to replace her. Very sad to see my co-worker leaving.

And we have a new executive director. I don't know if I should tell her what happened. The woman who had been filling in hired her and went back to her division director job. But that makes her the exec's boss. So it's still an office politics problem not really a chromatography problem.
They qualified a result for a failed spike even though the spike passed. The guy transferred from rad was running it. He had a sample with pretty high Cl and the spike level went over the calibration range. That biased it a bit low and it failed. The method says to dilute results over the calibration range and when he did it passed. I even pointed out in the SOP where it said to dilute results over the calibration range, but the chief still insisted on qualifying the result. My head almost exploded. I would call my retired coworker but I think she might not want to remember that the lab even exists.
We got audited by NELAP. No mention of any ethics investigation came up. An auditor looked at one current chromatogram. She saw in the SOP where it said the IC can be run at 27 mM provided that nitrite and carbonate separate. I told he I run at 23 and showed how that puts carbonate in a large gap between chloride and bromide.

Now I am wondering what my duty to data integrity is. And whether I could be in legal jeopardy at some point for not getting amended reports sent to those people who got invalid results. It was a slow week so it was only 12 samples that needed retesting. Why couldn't the bosses just send out 12 letters telling people their samples needed to be retested??

I took a close look at the data integrity SOP and it says that if laboratory personnel feel unable in reporting the suspected unethical conduct behavior to the area Program Chief/Supervisor, Ethics and Compliance Officer, or Laboratory Director, they should contact HR. I already reported everything to the Lab Director and nothing was done so it looks like my next duty is to report to HR. Beyond that? Has anyone ever had to report something to the EPA Office of Inspector General? Is anyone still reading?
I'd just keep covering your own tail if I were you. Don't sign stuff that is clearly wrong. Leave that to the chief.

If the NELAP auditors didn't flag your results (they're evidently incompetent or just did not happen to look in the right place), I suspect that HR won't spend a lot of time on your issues with the lab practices. Check that box, but don't expect anything.

I think a customer audit will be more likely to turn up something.

Still reading...

If nobody with experience will return HR's calls about openings in the lab, call the placement offices of colleges with ACS certified chemistry programs (Truman State in Missouri, for example). This way, you get to train them yourself - none of your ex-boss' bad habits.

Best of luck
Thanks,
DR
Image
DR wrote:
I'd just keep covering your own tail if I were you. Don't sign stuff that is clearly wrong. Leave that to the chief.

If the NELAP auditors didn't flag your results (they're evidently incompetent or just did not happen to look in the right place), I suspect that HR won't spend a lot of time on your issues with the lab practices. Check that box, but don't expect anything.

I think a customer audit will be more likely to turn up something.

Still reading...

If nobody with experience will return HR's calls about openings in the lab, call the placement offices of colleges with ACS certified chemistry programs (Truman State in Missouri, for example). This way, you get to train them yourself - none of your ex-boss' bad habits.

Best of luck
I left plenty of notes on data packets and sent plenty of e-mails to people complaining about the situation. In one I even used the phrase lab fraud. They are all searchable public documents. I have not signed off on bad data, but I am not sure I did enough to rectify the situation. There is some ambiguity in the ethics procedure but it looks like I may be required to report to HR. Nothing in my training prepared me for a situation where supervisors would order me to do things that are unethical or not respond properly to complaints of ethics violations. We are supposed to send out amended reports if we just misspell a name. These 12 sample were not even tested properly. That's far worse than just a misspelling.

Our data integrity procedure says HR should bring in an outside neutral party to assess data integrity complaints. They did that to one of the organic chemists several years ago when someone complained. They found that he had not done anything wrong. I think NELAP didn't find the problem because it was hidden from them. I don't know if they ask about any ethics complaints. I would think they would and the people in charge at the lab could easily have lied and said nothing.
In analytical chemistry, coelution refers to the situation where two or more chemical species in a sample cannot be completely separated from each other during a separation process, such as chromatography.

Whether or not coelution can be ignored depends on the specific analytical question being addressed. If the analyte of interest is well resolved from other species in the sample and coelution with a minor interfering species is occurring, then it may be possible to ignore the coelution and accurately measure the analyte. However, if the interfering species is present at high concentrations and coelutes with the analyte of interest, ignoring the coelution may lead to inaccurate results.

In general, coelution should be minimized as much as possible in analytical methods, and if it cannot be avoided, appropriate correction factors should be applied to account for the interfering species.
AlanPearman wrote:
In analytical chemistry, coelution refers to the situation where two or more chemical species in a sample cannot be completely separated from each other during a separation process, such as chromatography.

Whether or not coelution can be ignored depends on the specific analytical question being addressed. If the analyte of interest is well resolved from other species in the sample and coelution with a minor interfering species is occurring, then it may be possible to ignore the coelution and accurately measure the analyte. However, if the interfering species is present at high concentrations and coelutes with the analyte of interest, ignoring the coelution may lead to inaccurate results.

In general, coelution should be minimized as much as possible in analytical methods, and if it cannot be avoided, appropriate correction factors should be applied to account for the interfering species.

It's easy to eliminate this coelution by running the eluent at a lower level. That's what I am doing now. When the column ages I lower it to 22mM and then down to 21 mM. Running higher and letting carbonate and nitrite merge could hide low level nitrite and bury the peak under a large carbonate peak. I showed that early on by spiking a sample with nitrite. The carbonate peaks in samples range in size fro 0.1 uS-min to 0.3 uS-min. The low level nitrite standard peaks are only 0.05 uS-min. The 0.5 mg/L nitrite standard peak area is 0.3 uS-min. I showed the bosses and they ignored it. They just let a week of corrupted data get reported like that. I finally reported to the head of HR in December and I characterized it as criminal lab fraud. They finally had an investigator call me 2 weeks ago but he knows nothing about chemistry. No outside neutral party to assess my data integrity complaint has been contacted as far as I know. I don't see how than can count someone who knows nothing about chemistry as a neutral party.
47 posts Page 3 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