I need to locate any information that anyone may be able to provide on the subject of problems(instrument related) encountered when
HPLC samples vials are overfilled.Thank you in advance for your help.

Sincerely,
Steven Baker, QC Chemistry
Instrument Specialist
Phone: 813-866-2253
E-Mail: sbaker@bausch.com

If there is no headspace, a vacuum can be created when the sample is withdrawn, and the syringe might not draw up enough sample. If it is a matter of a vial being 3/4 full instead of 1/2 full, I am not aware of any problems.

Depending on the type of autosampler, injections from a vial with no headspace can also give larger than expected peaks. This is due to the sample being forced into the needle/ loop by the pressure generated when the septum is punctured.

Question: To Michelle and Andy (or anyone who knows this).... If I am making a 10 mul injection and have a 100 mul loop, how much sample does the syringe draw for each injection? If it only draws 10 mul then your points may be valid.

Also on some autosamples, the sample vial is pressurized to force the sample into the sample loop. If there is not anough headspace to pressurize the vial correctly you will not fill the loop well enopugh and this can cause reproducibility errors.

Regards,
Mark

hi, this is also about vials but little different, i am looking for a vendor who can supply HPLC vials with snap on type cap, and with good quality septum so the cap resealability is good after needle punches, this can improve the reproducibility. right now the vial i use has plastic septum so it left a hole open after the first needle punch. so with time goes the solvent will evaporate. i know i can order from brand vendors like waters, agilent, etc. but i doubt their price may be higher, if i can find a generic vendor, the cost may be reduced. thanks for anyone who can share me the info.

To John B; what kind of injector are you using? Is it an autosampler or a manual injector. They work slightly differently.

We are using Agilent 1100 autosamplers.

The Agilent 1100 autosampler uses a piston metering pump that pulls a controlled volume (10 uL) into a loop (100 uL), then dispenses it onto the column. With a 100 uL loop, the injected volume can be anywhere from 1 to 100 uL, whatever is entered through the software. The loop does not get filled. To ensure that all of the desired volume is injected, the valve routes the solvent flow through the loop, which washes it out.

Anon,
We use national scientific vials with snap on caps. The septa do reseal and we have very good reproducability from these vials. We don't use enough to earn a price break from national, but we do use another local supplier. Look through vender catalogs like VWR or Krackeler scientific to see if they have something that will help your problem.

Hi John,

We have found the Agilent 1100 autosampler to be very accurate and precise but its design can also produce some specific problems.

If you overfill the vial a small amount of the contents can be forced onto the upper surface of the septum while the needle is being inserted. Then, when the needle is withdrawn this material is deposited onto the exterior of the needle. Since the 1100 does not use an exterior needle wash, this material can drain down the needle and rest in the depression of the needle seat and lead to carryover in subsequent injections.

We have found that to reduce carryover you need to use a septum with good wiping characteristics, we use PTFE faced silicone, these septa also reseal well.

The 1100 is also sensitive to cap diameter. The 1100 firmware tries to close the gripper fingers to a set position. If the cap diameter is too large this position cannot be reached and can lead to autosampler errors. We use Sun broker vials and caps.

I have to take issue with one point in Tom Mizukami's posting. At least when used with a ChemStation, the HP 1100 autosampler can easily do an external needle wash. I use this feature all the time. I have not attempted to do this using the "game boy" controller instead of the ChemStation.

Fill the vials 2/3 to 3/4 full, and cap with Agilent caps. You should have no problems. If samples are more viscous than typical, you can slow down the draw speed.