For a lab doing only small number of runs a day (say, no more than 10 unknown samples, usually fewer), does it make sense to get an autoinjector vs. manual? Is it required for GLP/GMP compliant systems? Does it produce more consistent data?

Always better an Automatic Injector if you are working on a GLP/GMP system.

(10 Unknown samples x 3 injections each sample solution) + 2 blank system + 5 Suitability test + ( Nš of External STD x 3 each level ) = ......

I am not sure about the GLP/GMP requirements, but I think an autosampler is definitely worth the money. I have worked both with and without them. With an autosampler, someone doesn't have to hover around the instrument, and that person is free to work on other things. Also, you may only do 10 injections per day now, but if that number increases, you'll want to be able to use the LC around the clock. I have never noticed a difference in reproducibility between the two--they both use a sample loop. The advantages of manual injection are that you get more flexibility as far as injection volume, and there seems to be less carryover.

To Anon,

An autoinjector in NOT required for GLP/GMP testing but to go along with thesecond part of your question the RSD from repeat ionjections is always lower using an autoinjector so it is generally easier to meet reproducibility requirements when using an ALS. And to mirror Michelle, it is certainly worth the cost to free up an analyst to do other things ("analysts are damn expensive autosamplers" as one of my former bosses used to say).

Regards,
Mark

Definitely get the autosampler; better precision, etc., no coffee breaks, sick time, etc.

Michele,

I have to disagree with you that you get more flexibility with a manual injector regarding injection volume. With a manual injector you have a fixed loop of x microliters. In order to achieve reproducibility (reasonable RSD's), you need to overfill the loop, preferably flushing though with more than one loop volume. This is the proper way to use a manual injector. If you start injecting amounts of less that than the loop volume, you reproducibility (RSD) will become unacceptable for any quantitative work, regardless of whether it is GLP/GMP. Just good science. Using an autosampler, you still have a fixed loop or flow path, but you can set the volume from one microliter to the maximum volume that the autoinjector can handle. Since this is electronically and mechanically set, it will be far more reproducible that you ever can be in trying to do variable volumes in a manual injector.

You should flush a manual injector with at least 3x the volume. This gives extremely reproducible results. It is more trouble to change out a manual autosampler loop, which is required to change the injection volume than to simply enter a number into software (another knock against manual injectio), but you can change it to a wide range of volumes. I have loaded as much as 1.5 mL of sample onto the head of a column using a manual injector (1.5 mL loop) and selectively eluted, in effect doing online (or on-column) SPE and concentrating the analytes. From what I understand, this is difficult to do with an autosampler.