Shield RP18 problems

<p>After several months of trial and error, I settled on a Shield RP18 Acuity 2.1x100 mm column for an anticoagulant analysis.  It worked beautifully, but over a couple hundred injections the resolution has fallen apart.  I was told that it was probably because I was storing the column in 9:1 MeOH:water between uses or because I was running it at 60C, so I bought a new column and resolved to flush with ACN at the end of every run and keep the temperatures down.  The new column worked just fine with standards but, after about 40 injections, its starting to fall apart.</p><p></p><p>Does anyone have any idea of what I'm doing wrong?  I really like the method and column - it replaces a conventional HPLC method that took more than an hour to run, and relied upon a column that is no longer available, but I can't afford to replace columns at this rate.</p>

Comments

  • Hello,

    Could you please post your detailed chromatographic conditions? Other than just guessing, it is difficult to accurately suggest any causes for premature column failure without knowing how the column is being used/treated.

    Thanks,

    --Doug

  • A - 0.1% Formic Acid in LC/MS water

    B - LC/MS grade MeOH

    2.1 x 100 mm RP18 Shield column @ 60C

    Flow 0.6 ml/min

                             %A    %B     Curve

    Initial  95.0  5.0

    1.00   40.0  60.0      6

    3.00   34.0  66.0      6

    7.00   34.0  66.0      6

    8.00    5.0   95.0      6

                   10.00   95.0    5.0     11

    I conditioned the column when I received it with 50:50 MeOH:water before using it, wash it with 5% water:ACN turn the column heater off as soon as a run is finished.  So far, I haven't run any "real" samples (tissue extracts), just standards in 50:50 MeOH:water.  At 10 injections, I thought that some of the later peaks were looking a little fuzzy.  By 30 injections, its obvious that something isn't right.  Retentions have shifted and peaks that used to be 10 secs wide are spread to 40-50 seconds.

    I spent a long time developing this method, and the RP18 seemed like a godsend when I tried it.  Nice, sharp, symmetric peaks in a short run.  However, we can't afford a new column every 200 injections, let alone every 20.

  • Thanks.

    What is the maximum column pressure that you observe during the gradient? Does the pressure change as the column ages?

    What is the maximum number of injections that you have ever realized? How long have you been developing the method? In other words, have you ever obtained long column lifetimes during your method development process or did this just occur? On how many ACQUITY UPLC BEH Shield RP18 columns has this occurred? Can you provide serial numbers?

    I am asking these questions since the BEH Shield RP18 column has been on the market since 2005 and is a solid, robust stationary phase. The columns SHOULD provide long column lifetimes, especially under the conditions that you describe. Something odd is occurring and I do not know what it is based upon what you have told me thus far. I am trying to determine if you have observed shorter than expected column lifetimes DURING the method development process, or now, which is AFTER the method development process.

    So, I would like to know how many columns you've tested, their serial numbers and the lifetimes observed during the met dev process.

    The problem is either chemical (bonded phase/endcap hydrolysis or substrate dissolution) or mechanical/physical (voiding).

    --Doug

  • I haven't been recording column pressures routinely (I do have some data, but it will take a while to winnow from the data directory), so I can't comment other than that this mobile phase skirts the 15,000 psi cut off for the instrument.  At the beginning of the run, its typically 7800 - 8,000 psi, then I believe (don't usually sit there and watch) increases to a peak of 13,000 - 14,000 somewhere in the middle and comes back down as the gradient finishes.

    Most of the development time was spent with other columns, especially the BEH C18 and some non-Waters columns.  This is why I love the RP18:  it worked practically right out of the box, whereas I'd spent days and weeks getting nowhere with the other columns I'd tried.  I'd have to say that the "development process" with the RP18 consisted of less than 50 injections and the remainder production samples.  The new column has had nothing but standards on it, yet.

    I was told that the problem was likely the fact that I stored the original column in MeOH:water, however the new column is stored in ACN.

    Off the top of my head, I have about 150 - 200 injections on the first column, and less than 40 on the new one.  So far, I've used the same mobile phases and similar temperatures and gradients for hundreds of injections on the BEH C18 column(s) with no apparent loss of function, so I have to believe that there's something unique about the RP18.

    I'll see if I can dig out the pressures and serial #'s when I get a free minute during the next couple of days.

  • Hello,

    While waiting for your response, I input your method conditions into the ACQUITY UPLC Columns Calculator and got a pretty high backpressure (>12000 psi). That is why you are operating at 60oC, correct (to lower the operating pressure at the flow rate you selected)?  Did you select these same conditions for any other UPLC or UHPLC column? If so, how did they last?

    At this point, I suspect that you are mechanically voiding the columns. One way to find out is to open up a column at the inlet and see if there is a void. It's easy to do and I can send you a procedure if you wish.

    Please let me know what the serial numbers are for the columns in question.

    Lastly, have you notified your local Waters person about your troubles? Just curious.

    Thanks for your patience,

    --Doug

  • Actually, someone in tech support suggested 60C @ 0.6 - 0.8 ml/min for a similar mobile phase in the BEH C18 column.  It typically gives me 12,000 - 13,000 psi (less with ACN than MeOH) and some beautifully sharp peaks.  Prior to that time I'd been running 30C and maybe 7000-1000 psi but the higher pressures and temperatures spiffed up the chromatography so much that its kinda become my default conditions.  I've been using similar conditions in the BEH C18 columns (1000's of injections) for roughly a year and haven't noticed any deterioration, yet.

    I'll scrounge around for the pressure data when the instrument is free tomorrow, but I'd be interested in your method for taking the column apart.  I'd really like to get to the bottom of this.  If I can safely use the RP18's at lower pressures, I'll gladly trade a little resolution and speed for longevity.  I just need to know before I blow another $500 on a column.

  • Hello,

    How many BEH Shield RP18 columns have failed under these operating conditions?

    This may just be a 'one off' column failure. It happens. We will certainly replace the column(s) for if you contact our Technical Service group. That's why I inquired about contacting a local Waters person. He/she can do that for you if you'd like. Don't worry about having to buy a new column.

    If it's a chronic issue (multiple columns) then something else is afoot.

    I've attached a troubleshooting procedure that we suggest if people run into column lifetime issues. Usually the issue is column plugging. This occurred frequently back in the early days of UPLC. At the end of this presentation is a frit replacement procedure. Please take a look at the step-by-step slides that I have created. Opening up a UPLC column is actually quite simple to do since they are so well/tightly packed. Just make sure the column is at room temp before opening it up.

    Please let us know how things turn out. Also, let's get your dead column replaced.

    Regards,

    --Doug

  • Two.

    I contacted Phil Kim (chromatography guru) and Bill Hobbins (salesman) about the first column going downhill.  That's where it was suggested that the problem was storing the column in MeOH:water for prolonged periods.  They also suggested some things to try to resurrect it, but none worked, so I bought the second column.

    I dug out some of the old data files.  Peak pressures in the early runs, when I was getting the amazing chromatography, were between 11,800 - 12,500 on both columns.  More recent pressures seem to be slightly lower, maybe 11,500 - 12,000.  Don't know if the difference is significant, but its a data point.

  • Hello,

    I emailed Phil and asked him to follow up with our Technical Service group to get you replacement columns.

    I do not feel that those pressure differences are significant.

    Please open up your column(s) and see if it/they are voided. If for some reason you do not feel comfortable doing so, please let me know and I will arrange for the column to get sent to me for internal evaluation.

    --Doug

  • I opened up the column last night and examined the area under frit with a hand lens.  I can't see any voids, but then would they necessarily be visible at that level of magnification?

  • Hello,

    Yes, even with the naked eye you would see a void if there is one. The lack of a void is a good thing.

    Have you tried running your method on a brand new column - one that has not been stored for any period of time? We can take care of this for you.

    Thanks,

    --Doug