I've created a calibration curve for a phospholipid sample analysis that works very well. Except when I take samples from my batch reactor, which are slightly more dilute than a "neat" sample prep would be, I get inflated results even when I take the batch reactor dilution factor into account. I see this effect in all of my samples, but the higher PL content samples have been impacted more so than the lesser PL content samples. My gut tells me this must be a density effect but how, and how do I account for it in a way that doesn't limit the samples I can run on the curve, or over complicate my sample prep process?
Maybe your calibration is in error. Which analytical technique(s) and/or specific method (in detail) are you using to make these measurements? Sometimes a very simple assumption such as "the results are linear" is false, skewing the answers.
Hilic method on CAD with an applied power function. The results are consistent. They make sense. Except that the neat samples look fine bit samples from the reactor are slightly inflated. And they have a dilution factor of .70 relative to the neat samples. This is being taken into account when quantifying. The two grades of PL oil show different degrees of inflation in tve reactor samples vs neat preparations.
Actual samples often present difficulties not seen in pure standards. CAD's are very troublesome in general to use (best for advanced users only with lots of experience), not linear and a "power factor" formula fit may not be enough to compensate for the matrix involved "actual" samples. You need curve fits with lots of points that provides a very accurate description of the response (e.g. sigmoidal, polynominal fits etc). Hopefully your software has many options in this regard. CAD also outputs a different signal response for different samples. They are not universal, hard to reproduce and often require many more calibration standards then normal per order of magnitude (i.e. where 5 stds might be normal, but for CAD/ELSD maybe 9). Use of an internal standard may also prove valuable too.
Another common application issue is solubility. PL's must be FULLY dissolved in solution (both in the method used and in the injection solvent). HILIC methods are generally poor for this. Please provide your exact HPLC method (with details) for more info.
For help with your question, please share the HPLC method used (basic details include info such as the column name/brand/type, column dimensions, particle size, flow rate, column temperature, inj volume, inj solution, mobile phase composition etc). Is there someone local to you with experience in HPLC that could help you?
BTW: Specialized detectors such as CAD and ELSD provide a mostly non-linear output, different for each sample, different for each concentration range and the response is not "universal" (as suggested by some sales dept's).
More specifically, my mp is ACN and a 0.5% DEA buffer with formic acid, pH 3.
There are two grades of oil I test, the difference btwn them being the PL:TG ratio. I prep the neat oil in MeOH, and for both oils, I get good results that are confirmed by an external source. But when I prep samples from tbe reactor, where the oil is slightly more diluted, I observe that the expected results are slightly inflated, in a consistent manner, and the degree of the inflation is greater in the higher PL sample than in the lower. And again, I am accounting for the additional dilution in the processing method.
Sara Kuhl: No need to post the same thing 4 times. If you want help, please provide the requested info. BTW: You can delete posts you made in error on RG by clicking on the delete button.
As requested, could you provide your HPLC method to us? When discussing HPLC questions, the basic HPLC method should be provided so we can understand your question.
I have HPLC experience.
I havent used this method or worked with PL samples previously. But I have HPLC experience enough to know I’m not doing it “wrong” but theres some phenomena I’m missing here. Ive been working with an applications chemist at Thermo and we are still trying to perfect the method.
My gut instinct is that it has to do with the PL:TG ratio vs the matrix solubility. But thats a particularly difficult thing to google. Additionalky, not many people are experts in krill oil chromatography. So finding someobe to help equates to polling the literature experts, or asking on a reputable forum. Case in point.
In our lab, we are using CAD for the detection of PC and the IS which we use is Dimyristoylphosphatidylcholine (DMPC)
Kindly check...
Article Revisiting the dilution factor as vital parameter for sensit...
- Badges
- Science method
- Similar topics
- Chemistry
- General Biochemistry
- Biochemical Techniques
- Analytical Chemistry