Column test solutions

[CSV]

Hello!
At our lab we have so far not used any standard procedure to test the perfomance of our columns, which has led to a "piling up" of columns with doubtful performance. In an effort to clean up in our column storage we are thinking of implementing a routine testing of all columns (upon arrival and during usage). Most of our columns are reversed phase (C18, CN etc).
I'm considering a test solution containing uracil, toluene and naphtalene. Does anyone have any ideas to share about concentrations, injection volumes, performance/acceptance criteria or something else concerning this issue?

Take care!

[maris]

I don't see this approach as helpful. Any standard mixture/procedure which sufficiently differs from the actual analysis condition will not refer the column performance.

[tom jupille]

Actually, Maris, I would agree with your statement as far as column chemistry goes, but a standard test will pick up things like a head space void or partially plugged frit.

[ConstantinS]

Rather strange choice of adsorbates...

Testing a column is made for following purposes (I'll try to formulate):

1. To obtain maximum N for a column.
2. Having the column of known N - to test the connection with chromatograph.

For these purposes toluene is good enough.

Then,

3. To check the ability of absorbent not to broaden peaks of bases. Here you can use pyridine, 22'-bipyridile and so on.
4. To check the ability of absorbent not to broaden peaks of acids. Here you can use phenol, or any organic acid (in thes casr you have to acidify eluent).
5. To check methylene selectivity of reversed phase. Use benzene in addition to toluene, for example.
6. To check the ability of a phase to retain very polar substances and to operate under 100% water containing eluens. Use organic acids (pair oxalic/tartaric), nucleotides and so on.
7. To characterize in general the selectivity of all your colums in order to compare them. Creativity is wellcome here.
For example, for all types of polymeric reversed phases I developed such mixture: acetone, pyridine, p-chlorophenol, benzaldehyde, toluene, cumene.
On most polym. phases the chromotogram looks like a good fence with equal intervars But it's not so on other reversed phases.

So, think about the aim of testing, and then you will know what to do.

[Patrik Appelblad]

There is a standard reference material (SRM) available for your particluar purpose. Have a look at the NIST SRM 870. I believe that is what you are looking for.

I quote the first seven sentences of the SRM 870 paper (a pdf file, readily available on the web)..

"SRM 870 is a mixture of five organic compounds in methanol intended for use in characterizing general aspects of liquid chromatographic (LC) column performance, including efficiency, void volume, methylene selectivity, retentiveness, and activity toward chelators and organic bases. Other possible uses include (1) column classification to aid column selection during method development, (2) as a control material for monitoring LC column performance over time, and (3) in quality control for column manufacturing. SRM 870 consists of a mixture of the following five organic compounds in methanol: uracil, toluene, ethylbenzene, quinizarin, and amitriptyline (see Figure 1 for structures). The concentrations and relative detection responses of the components are listed in Table 1."


/Patrik

[ravi]

What leads you believe that they are doubtful? If they are doubtful, why keep them at all? If you have a doubtful column and its tests okay with your sample mix, would you still use it for development with a very short timeline?

Answering those questions will probably be quicker than testing all those columns.

[Provetech]

Our lab is using a mixture of uracyl, toluene and naphtalene for RP phase testing. We want to view every month: plates number, retnetion factor, void volume, symmetry and resolution evolution. This afford us to estimate aging of the stationnary phase and determine if the performances are good enough to perform quality control of our synthetysed products. It's important to tell we're not a "pure analysis" lab but a R&D lab wich make a control by HPLC of the molecules we send to customer, so maybe you need a more "quality" procedure in agreement with GLP or ISO....