Extremely exact pipetting, what volume is reasonable?

[dresdentl]

For calibration samples I have to pipette standard and internal standard. I'm pipetting 2,5 µl and 10 µl. Would it be better to do this precedure with higher volumes, e.g. 25 and 100 µl?

According to the data sheet a 10 µl pipette schould be nearly the same (a few % failure)

[Peter Apps]

For those volumes I would use a microsyringe, not a pipette, and I would check weigh and correct for actual weight added. It's not foolproof (see Evaluation of the source of bias caused by losses of solvent vapour
during sample preparation. Peter J. Apps • Marcelle´ Archer, Accred Qual Assur (2010) 15:171–180 for examples of what can go wrong) but you will get better precision and accuracy than relying on volumetrics alone.

Peter

[gstaepels]

Just a week ago, I attended a pipetting workshop. More than 100 people were in attendance, and the general feeling was: "Pipetting is not all that simple!". First of all I would suggest that you calibrate the pipette with purified water. Make 10 measurements and calculate the standard deviation. Is your pipette a variabel one, you should check the pipette at 100%, 50% and 10% of its nominal volume. The regression should be linear. Pipetting organic (volitile) material is more difficult. Consult the user manual of your pipette or visit your suppliers web site for the correct procedure using the pipette.
Even now, when I am aware of the possible errors, I still tend to weigh the standards and analytes.
I hope this helps.

[gstaepels]

One thing more: the temperature of the water with which you calibrate is extremely important. It is suggested to condition the water at the lab temperature for at least 2 - 3 hours, and that the density of water at the measured temperature into account to calculate the volume.

[HW Mueller]

It was hinted on, the pipetting technique is just as important if not even more so. Everybody, including doctoral students, who started in my lab had to complete a session with pipettes on the balance/scale. Some complained about this "waste of time", until they had played a bit with the weighing. This also helped in having everybody use the same pipetting technique which became necessary as we had large differences of results on glucose analyzers, etc. (any apparati which had sample directly pipetted into them) from one operator to the other. We standardized such things as pipetting against the wall of the container or free into air, or into solution. Also, I replaced all "air cushion" pipets with positive displacement types wherever possible.

[Consumer Products Guy]

Our experience is that most laboratory errors are caused by improper volumetric techniques; of course, even the most accurate and expensive instrumentation ultimately always relies upon good volumetric techniques. We purchased a few commercially-available video tapes, didn't like them, so made our own volumetric techniques video.

Maybe there's something better available these days.

[lmh]

Absolutely agree with HW Mueller about getting new lab members to weigh water until we are convinced we are all getting the same result. When I worked as research assistant in a group with fairly high staff throughput I had the same problem, that many, particularly experienced post-docs, regarded this as a patronising waste of time and refused to cooperate.

My remedy was to take their fully-calibrated and top-notch pipettes the day before they arrived, and hide them in a dirty drawer full of old junk. On the day they arrived, I would grub around in the drawer and "find" the pipettes, spinning some story such as "these have been around a bit and might be out of calibration/their last user was a bit clumsy. You'll have to check they're pipetting correctly, and I'll get them calibrated if they aren't..."

No one ever realised that I was primarily checking the operator rather than the pipette.

[mbicking]

Agree with all of the above. To measure such small volumes accurately, the gravimetric approach would be best (taking suitable precautions).

Years ago I developed a one-day seminar on "Analytical Lab Techniques" which included a session on volumetric glassware of all kinds (and a few other boring but important topics like Significant Figures and Rounding, Calculations, etc.). Although it has never been popular, I think it is very important, as others here have noted, mostly because few lab people these days have ever had a Quantitative Analysis course. In fact, it is on the short course list at Eastern Analytical Symposium later this month. If anyone has any material that they would be willing to share, or real world data, I would be happy to consider using it, and acknowledging the source. Alas, I cannot offer any monetary compensation, but there might be some adult beverages available should we meet.