Calibration Curve Advice

[GCgirl26]

Hello,

Thank you for clicking on this post. I have a question for any chromatographer that is more experienced than I.

I have just taken up GC-MS as of the past three months. I have been relying on the methods set up by the guy who preceded me. It is to my understanding that a five point calibration curve is pretty standard, however, I am not sure that the concentrations he chose are appropriate.

I am measuring the concentration of active flavor compound in an oil. The spec range for the analyte is roughly 28-32 percent. The supplier sends a COA, but we check it because of a past incident. The calibration curve per the aforementioned method uses standard concentrations of 2,5,10,25, and 50 percent. I don't have the appropriate pipettors to measure the first two standards accurately, so I have just been using three points for the calibration curve.

I need to decide if I want to buy a new pipettor, or bracket my spec range with standards that are closer to the analyte concentration.

Advice?

[Consumer Products Guy]

I am measuring the concentration of active flavor compound in an oil. The spec range for the analyte is roughly 28-32 percent. ... The calibration curve per the aforementioned method uses standard concentrations of 2,5,10,25, and 50 percent.

I need to decide if I want to buy a new pipettor, or bracket my spec range with standards that are closer to the analyte concentration.

1. Just because "a method" exists or is supplied by the manufacturer - doesn't mean that it's a good method. After our QA Director = who I didn't like - "bragged" about his 10 years experience as a QC chemist in manufacture of detergents for a competitor, I responded "but no one said that you were a GOOD chemist". Difference between a QC chemist and one who develops and validates cGMP methods is night and day.

2. Specification range is 28-32 percent, so anything out of spec is "who cares how much out of spec". That said: if me, I might do a two point calibration, low point at 25 to 28 percent equivalent, the other at 32 to 35 percent equivalent.

3. I might even do a single-point calibration at 30% equivalent because your spec range of 28-32 is narrow.

4. Going back to #3, I would likely explore GC-FID at 30% equivalent calibration as well. GC-FID is quite linear, and expecially for such a small range; do you really need GCMS in this instance??? In most of our house-developed GC and HPLC cGMP methods, we demonstrated/confirmed linearity using a 5-point calibration curve, then used a single point calibration standard.

5. If you're able to identify the analyte and the matrix, maybe we could offer more-detailed help. For example, if the analyte is UV-active, maybe HPLC could be a better choice.

[GCgirl26]

Hi,

Thank you for your reply. I completely agree with the QC chemist vs. method developer comment. I am just very new, so apprehensive.

The QC check is on allyl isothiocyanate in commercial mustard oil by a liquid injection method. Mainly I do headspace analysis on Horseradish. I work for a relatively small family owned company, and I am the only chemist. GC-MS what I have to work with for now.

May I ask what your background is? You sound like you know your way around a chromatograph. What is your motivation for answering a post like this, sales?

Thanks

[Consumer Products Guy]

May I ask what your background is? You sound like you know your way around a chromatograph. What is your motivation for answering a post like this, sales?

I worked over 4 decades for a consumer products/food products company in R&D; I did reverse engineering of products, test method development, cGMP and GLP method validations, instrument repair, and manufacturing troubleshooting. So I was working when fused silica capillary columns, reverse phase HPLC, and benchtop GCMS were all developed.

We had GC-FID, GC-TCD, and GC-MS available. And HPLC with UV/DAD, conductivity, refractive index, and ELSD. And UV-vis spectroscopy, FTIR. So we had a range of instrumentation so I could use the simplest technique that would do the job. Our manufacturing site QC did not have as much instrumentation, so sometimes that affected our decisions, but most times we just made our manufacturing buy same equipment that we used.

My motivation? Well, I eventually made a good living doing chromatography, so I guess my motivation now is to help others learn and apply their learnings, and to keep my brain moving ! We didn't have computers and Google back when I started !!! Actually, my department shared a single desktop calculator...

If all available to you is GC-MS, then not the time/place to investigate papers published for this using HPLC.

[GCgirl26]

Well what an amazing contribution to make. Thank you!

I am going to focus on becoming a skilled GC-MS chromatographer before I ask for any more expensive instrumentation yes.

That is a fantastic story about the calculator. I often comment that I can't fathom doing research in a world with no google or YouTube. Despite all of the resources, chromatography is still really challenging.That is why I find myself here. Nothing replaces experience.

I will definitely be bothering you with more questions at a later date.



Take care.

[Consumer Products Guy]

That is a fantastic story about the calculator.

About 5 years later, I "borrowed" my girlfriend's Radio Shack calculator, then 3 decades later the on-off switch went out, and the screen cover broke. Anyway, Photobucket has now blurred my picture (holding for ransom) but I engineered fixes for that.


I'm listening to Court TV online live right now, why I'm on my laptop. I'm up to Windows7 !!

[James_Ball]

Hello,

Thank you for clicking on this post. I have a question for any chromatographer that is more experienced than I.

I have just taken up GC-MS as of the past three months. I have been relying on the methods set up by the guy who preceded me. It is to my understanding that a five point calibration curve is pretty standard, however, I am not sure that the concentrations he chose are appropriate.

I am measuring the concentration of active flavor compound in an oil. The spec range for the analyte is roughly 28-32 percent. The supplier sends a COA, but we check it because of a past incident. The calibration curve per the aforementioned method uses standard concentrations of 2,5,10,25, and 50 percent. I don't have the appropriate pipettors to measure the first two standards accurately, so I have just been using three points for the calibration curve.

I need to decide if I want to buy a new pipettor, or bracket my spec range with standards that are closer to the analyte concentration.

Advice?

Already good advice on this one, but a tip, if you really do want to get those other two low points in the calibration you could simply do a 1:10 dilution of the 50 and 25 calibration solution to get to a 2.5 and 5 concentration.

I actually did a little method setup on the same compound on GC/MS looking at it in mustard and broccoli seed extracts. We never got into the testing so don't have experience on it long term with actual samples.

As above I have about 30 years in GC/MS, GC/MS/MS, LC/MS/MS and ICP/MS doing environmental testing ranging from drinking water to waste water to soil to just about anything else nasty someone wanted to send us. We have done a little work on nutriceuticals also. Mass Spectroscopy is definitely a good and interesting field to work in. If you get a chance try to attend a few conferences along the way, you can meet some great contacts there and here also.

[GCgirl26]

Thank you James,

That is a good point about the dilutions. I mainly wanted to draw on someone else's experience to confirm that it was dumb to calibrate at such different concentrations than the analyte would ever be present in.

Thanks for replying.