manipulated HPLC spectra?

Discussions about HPLC, CE, TLC, SFC, and other "liquid phase" separation techniques.

10 posts Page 1 of 1
Hi there,
As an editor, I got a set of very "strange" HPLC spectra: Y-axis all started from 0mV, the baselines have no noise at all, the peaks are sharp (without any slop at the bottom). Since I am no HPLC expert as you guys are, I just want to know what kind of data should I ask from the authors to find out if the spectra were manipulated. If they were manipulated, how did they do it?

Thanks for the help in advance,

ASC
It's hard to comment without seeing the chromatogram, but it would be unethical to post a submitted figure to a site like this. I'd recommend you use your reviewers and make sure you choose a reviewer with good analytical experience in as close a field as you can find to the paper (you'd probably be doing that anyway!). They can, of course, see the full data and are free to comment if they think any likelihood of foul play.
The fact the chromatograms all start at exactly zero isn't necessarily a problem. Many detectors can be set to autozero at the start of a run. Perfect baselines are rare, but possible if the compounds are present in large amounts, and the detector is selective and doesn't respond to background chemcials etc (look in the column manufacturer's catalogues; their chromatograms are about as good as it can get).
You are, however, right to worry. For example, some years ago I encountered an LC-MS method that gave perfect Gaussian peaks in a chromatogram. The reason was that the mass spec was quite slow at scanning, and was, in fact, only seeing the target compound in a single scan. The analyst had used a Gaussian smoothing filter, with the result that one single value was expanded into a nice Gaussian peak! This wasn't deliberate cheating, it was merely a misunderstanding by a relatively inexperienced operator.
Some older (and I mean like 1970's vintage!) integrators would basically show only "0" until the signal got above a threshold level -- which generates chromatograms exactly as you describe.
-- Tom Jupille
LC Resources / Separation Science Associates
tjupille@lcresources.com
+ 1 (925) 297-5374
Thanks lmh.

This was what I guessed:
possible if the compounds are present in large amounts, and the detector is selective and doesn't respond to background chemcials etc


Then I found out that some spectra using racemic samples were just normal with baseline noise and the Y-axis were even from 150mV to 1000mV, instead the "strange" ones (the corresponding chiral samples) had Y-axis only from 0mV to 300 mV or 500 mV. Does these Y-axis' comparable when the same column and same conditions with basically same compounds (racemic substance and chiral substance) were used?

I asked about more information such as apparatus, column, eluent, program and will then contact one of reviewers for the reevaluation. But before I do this, I want to build up my own opinions by asking you guys :wink:

Thank you again for the help.

ASC
tom jupille wrote:
Some older (and I mean like 1970's vintage!) integrators would basically show only "0" until the signal got above a threshold level -- which generates chromatograms exactly as you describe.


Thank you very much, Tom. As I wrote to lmh, the authors had some normal spectra too and all the spectra were taken from the same apparatus. So I think that probably won't be the possibility. One reviewer found the problem first and asked to ask for "raw data" from the authors, but what is "raw data" of a HPLC spectra? The authors just provided a full X-axis print. Could this be called "raw data"?

Thanks again for the help.

ASC
lmh wrote:
The fact the chromatograms all start at exactly zero isn't necessarily a problem. Many detectors can be set to autozero at the start of a run.


Dear lmh,

the author responded and said that the students "pressed the “A/Z” button to adjust the “zero” of the HPLC machine." Will this action produce those "strange" spectra?

Bests,
ASC
Pressing an autozero will indeed make the start value zero. One way to generate chromatograms that look like those you describe is when the zero of the instrument is well below the zero of the recording device, you then only see the tops of the peaks, with a dead flat baseline. This is easy to do with simple chart recorders and old-fashioned paper-based integrators, I have never tried it with current software based data processing.

Forgive for saying so, but your use of "spectra" for what are actually chromatograms suggests that your journal is one where the analytical techniques and their results are just tools to solve larger problems. If the analytical results are crucial to the validity of the findings, then you need analysts as referees, just as you would use a specialist statistician to evaluate statistics.

Peter
Peter Apps
Peter Apps wrote:
Forgive for saying so, but your use of "spectra" for what are actually chromatograms suggests that your journal is one where the analytical techniques and their results are just tools to solve larger problems. If the analytical results are crucial to the validity of the findings, then you need analysts as referees, just as you would use a specialist statistician to evaluate statistics.


Thanks Peter! You are right about the journal: we are organic chemists and the HPLC is just a tool to get ee or yields. We just decided to reject the manuscript because the newly measured chromatogram showed much lower ee and other byproducts than the previous one. I appreciate the help from all of you very much!

ASC
Ask the authors to provide the acquisition parameters used for the analysis. This includes information about the instrument settings, such as column temperature, injection volume, mobile phase composition, and flow rate. wordle game
You wrote: "I got a set of very "strange" HPLC spectra: Y-axis all started from 0mV, the baselines have no noise at all".

A few comments:
-A 0 mV starting point is normal.
-You refer to "spectra" (which is specific to UV/VIS analysis), but your comments seem to imply a chromatogram. Very confusing.
-Baseline noise can only been observed when the signal shown is "zoomed" in to show it. If the signal is large, noise will often appear flat and absent (again, normal). Sometimes this is simply due to the copy/paste of the data for formatting inro the paper/document.
A person who does not have professional experience (e.g. many years in an industrial lab) can not be used to evaluate any HPLC method for validity.
-As an editor, you should have professionals in the field who act as reviewers. They should have many years of experience in HPLC analysis so they can understand what the proposed article describes. If you are an editor for a journal, then send the paper for review and all questions will be answered.
-My boss used to be a key reviewer for many of the main journals and he said that many of the articles submitted did not include many of the basic details needed for the methods used. Lots of missing info as the authors did not even know which parameters were needed. Many did not follow good chromatography fundamentals and were invalid as received.
10 posts Page 1 of 1

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