Hi all, just wanted to see if anyone out there is having this problem.

We have several different methods that utilize SPE followed by HPLC analysis. We are running all methods on a Shimadzu component system and we get drug peaks in every blank that we run. The drugs we use vary from very hydrophobic to very hydrophilic. We have isolated every part of the procedure (i.e. SPE blanks, post-extraction blanks, mobile phase blanks) and have been unable to come up with anything. And to make things worse, the problem seems to come and go. We have even gone to breaking down our SPE manifold and washing and sonicating every part and this doesn't help. We run mp blanks and sometimes they have a drug peak and sometimes they don't. I hate the instruments and have readily blamed them but the data doesn't support this when we inject mp blanks after high-concentration samples and don't see any peaks.

I know this seems full of contradictions but this problem has been one contradiction after another. Just wanted to see if anyone else is having/has had similiar problems. Prior to my working on Shimadzu, I have done SPE for 10 years and have never had this problem.

Any ideas????????

You have a Shimadzu component HPLC. What is the injector? How do you do SPE? Individually or in a 96-well plate? Do the blanks go through the SPE method as well, or are they just blank injections? Please clarify!

There appear to be many people who "solve" the problem by not running blanks.
More seriously and generally:
Check "carryover" in keyword search.
Yesterday I worked with a 123I labelled compound and couldnīt completely remove radioactivity in the injection syringe and in the Rheodyne (very common observation). Its really a matter of how much carrover you can tolerate or "see". Remember: Identifyable peaks are only observed if some unequilibrium is perpetrated ahead of the column, usually via mechanical action like an injection. Look first at movable parts and fittings, ie, unswept (by mp) places.

The injector is a SIL-10ADvp, and as far as the SPE procedure, they are all different but are done individually.

As for the blanks they are as follows:
1. SPE Blank - Mobile Phase instead of serum taken through entire SPE (always has a drug peak)
2. Serum Blank - Blank serum taken through SPE (always has a drug peak)
3. Post-Extraction Blank - Elution solution is evaporated then re-coned with mp (occasionally has a drug peak)
4. Mobile Phase Blanks - Aliquots of mp injected (sometimes has drug peak)

We have in the past run an experiment where we did 20 injections (alternating between high std and mp blank) and have not seen peaks in mp.

This would seem to clear the instrument and indicate that it is not carryover, although we occasionally see peaks in the mp blanks during analysis (they always follow the QC std)

These peaks are very small (usually slightly lower than our LOQ) and analytically can easily be subtracted out and not cause huge problems, but I would rather get rid of them although at this point it seems impossible. Like I said earlier, I have alot of SPE experience and have never had this problem, ever. So it has to be something going on here, either instrument or otherwise. And Shimadzu has been very little help.

I haven't used the SIL-10 vp autosampler but here are some ideas based on experience with the SIL-10A. Have you tried creating a pretreatment file to rinse the autosampler thoroughly before the sample is aspirated? Have you tried increasing the excess volume for the injection to more thoroughly rinse the lines to the 6-port valve with your sample solution before the loop is loaded? If the vp syringe drive is the same as the SIL-10A, you might check to make sure there is no liquid leaking from the pressure bypass at the bottom of the syringe valve. During the rinse the syringe speed is increased and if there is a slight blockage, the rinse solvent might get pushed out this bypass instead of through the sampler lines. During an injection the syringe speed is lower so an over-pressure problem is not as likely. Obviously if it isn't getting rinsed properly, carry-over could be a problem. However, I still can't explain why the problem ALWAYS shows up for the SPE samples. I am assuming you condition the cartidges before use.

I am not sure what kind of detector you are using, but it could be that your detector isn't specific enough. I even had a similar problem doing SPE with GC/MS analysis of pesticides. The SPE cartridge contained an interferrent with one of my peaks so even distilled water run through the SPE resulted in a peak. The more solvent I used to condition the cartridge, the lower the peak became. It took more than 30 mL of methanol to condition the cartridge, so I ended up just using another cartridge.

Thanks for everyones input. We have tried a more thorough rinse and this seems to have alleviated the problem. But we have seen the problem disappear before and then show up later. Thanks to all for your input

Try Symbiosis from Spark Holland!