This reads to me like the HPLC retention time is shifting after extraction with SPE? What is your HPLC method? How much are you injecting? Have you run your standard through the SPE methods and see if it behaves the same?

The consistent shift in retention time and ion enhancement of Diclofenac after SPE may be due to matrix effects, pH variations, sorbent characteristics, and instrumental factors. Optimizing SPE conditions and standardizing procedures can help improve reproducibility and accuracy.

You should always spike your matrix with the analyte in order to avoid matrix effects.

If the molecule is sensitive to pH change or cation exchance there will be shift in retention time in LC

Which salt of Diclofenac are you spiking?

Are you using LC-MS/MS?.

Best

If I read it correctly, you are spiking water with this compound and test SPE recovery by loading, washing and eluding with different solution to get the best recover without sample matrix. If this the case, sample matrix is not the issue. Your compound has carboxylic side chain and it will change retention in different pH environment. If it is in acid, (mobile phase or sample solution, it will be a neutral compound and retain well in a C18 column (you don’t provide the run condition). If the pH is higher, it will be less retentive as it will be ionized. I need moredetail of the procedure before I can fix the problem.

Thank you all for your replies. Narong Chamkasem The retention time shift is only in SPE samples (pH=2 and pH=10) not in the standard, so I think it should be due to the matrix I am using LCMS and my mobile phase is A) 0.1%FA and B) MeOH. (c18 Column). Additionally, there is considerable ion enhancement as well. I am analyzing 20 compounds and I encountered this issue in Diclofenac and Erythromycin. For Erythromycin I even lost the pseudomolecular ion peak (ESI +ve) and got this at M+H-Water, and that too at shifted retention time compared to the same peak in the standard. I am Spiking Diclofenac sodium Eleuterio Umpierrez

Also to resolve this issue I'm planning to determine the calibration curve of Diclofenac by spiking the standards and recovering them after spe so that I get a consistent retention time in Standards as well as the samples. Is this approach correct?

It is not clear to me why you have two pH values in the same sample. The way you make the std curve with SPE is cumbersome and should be avoided. Can you use the small injection volume to 1 uL in both sample and std ?

Sorry I didn't said clearly. I am working on optimising the SPE method for highest recovery so for that some of the SPE method involves diclofenac spiked to a pH 2 solution and some to the pH 10 solution and then different combinations of solvents for conditioning, equillibriation, and elution (different pH as well as solvents). I have been using 10uL injection volume throughout the study (for all compounds).

At the method development stage, you don’t have the final method to get the highest revovery, so the RT shift is not the important issue. Your compound is acidic in nature and a t pH 2, it will stay in the C18-type SPE. At pH 10, it should ionize and not retain. If you load the sample on the C18 SPE at pH 10, it will pass thru. Once you get the final cleanup method, then you can focus on the RT shift.

You can test the sample that the injection volume is not the issue by injecting a relative high concentration in the final sample with 1,10 and 20 uL and see if you have RT shift

Narong Chamkasem Thanks for your reply. I am using the OASIS HLB for selected compounds with a range of pKa and logKow values. In both acidic sample loading and pH 10 sample loading cases, I observed that DCF has a retention time of approximately 1 minute (36 minute method) shorter than the standard (200 ppb in an 80:20 water-methanol solution), and there's noticeable ion enhancement.

It is difficult to troubleshoot your issue without a full pic of your method. Please send me a whole protocol.

Narong Chamkasem PFA the SPE protocol. Earlier I was using this method then I observed a very small peak at the retention time of standard but larger peaks shifted 2 minutes earlier. However now I tried a combination of ethyl acetate and methanol as eluent recovery has improved for almost all of my compounds and also I got a recovery of about 62% for DCF (at RT of standard) . Thanks for your help.

From the protocol, did you prepare a sample twice to get acid fraction and basic fraction? I assume that you have retention time shift only in the acidic fraction? If so, we can disregard the basic fraction and focus on the acidic one. need to troubleshoot if what causes the retention time shift by isolating the step. 1) spike std in matrix and process 2) spike std in water and process thru the method and inject 1 and 2 to HPLC and compare with std alone. Which one of the two has retention time shift, or both. This will show if it is caused by matrix or the SPE itself

I just spiked to the miliQ water now and I am comparing the recovery for the MilliQ water only as of now. I am spiking the compounds in MilliQ water adjusted to pH 2 and pH 10. after this stage, I will spike to the matrix (Soil and wastewater) and check the recovery. I got retention time shifts in both cases (acidic and basic SPE). My colleague analyzed the diclofenac in the surface water earlier this year and she found the same issue. (These are not recovery studies; they are analyses of samples for quantification purposes.)

https://www.iomcworld.org/articles/development-of-a-rphplc-method-for-the-estimation-of-diclofenac-sodium-vitamin-b1-vitamin-b6-and-vitamin-b12-insoft-gela.pdf

https://www.agilent.com/cs/library/applications/5991-9077EN_diclofenac_hplc_icp-qqq_application.pdf