GC column - sign of end of lifetime

[Cassandre]

Hi everybody !

I'm a new GC user and I am wondering what the signs for changing the column are.
Is it a high noise ? a sudden poor resolution ? peak tailing ?
all of them ?

when do you decide to change your column ?
do you check this parameters regularly ?

thank you very much for your help.

Cassandra

[dblux_]

Hi everybody !

I'm a new GC user and I am wondering what the signs for changing the column are.
Is it a high noise ? a sudden poor resolution ? peak tailing ?
all of them ?

Cassandra

Shortened retention times (and decreasing during runs), excessive column bleeding, easy column overloading - typical for overheated DBWax column.

Wojtek

[Peter Apps]

You need to change your column when it can no longer do its job, in other words when the quality of the results is no longer fit for purpose. BEFORE you change your column check that the inlet is in good condition, since this is a far more common cause of the problems that you list. Cutting a meter or so off the inlet end of the column, rather than replacing the whole column, will often improve performance.

Peter

[Omaejel]

I have some columns that have been running almost seven days a week for at least 4 years. We hardly ever change one here unless it breaks or we lose compounds on the front end. An old column runs great. Hardly any bleed for your GRO ranges so they have great RSD's. I hate changing out columns. It is a last resort for me when I am working on something.

[thohry]

GC columns normally last very long. One of my colleagues uses an HP-1 for essential oil analysis and she replaced the column after 7 years of use almost everyday.
You can see the signs of : bleeding, peak broadening or skewing etc ... that the column can not separate well before replace with a new column.

One of my experience is to cut off 1 or 2 meters at the inlet end of the bad column and it may help.

[kvaak]

To check that your GC column is working properly, you can have same control sample e.g. for fatty acid analysis some rapeseed or olive oil etc. If the ratio of peak areas is getting worst, then you should do something for your system. The first thing is to cut about meter or half in the inlet side, if that doesn't work, the maybe change the column. In addition, the samples can affect to the condition of column. For example I have noticed that silylated compounds damage columns more than for example methylated or acetylated.

[AZBiodiesel]

i just recently had to replace a column after about 2 and a half months of use. It was an agilent DB5-ht. I have been reading a lot about the life of columns hand how you should be getting a certain number of injections out of it, on the order of thousands. However I saw a loss of resolution after 400 samples, probably less than that. I look at supelco bulletin 853b for any problems with the column, i think its a pretty thorough guide.
Anyway i start to get peak tailing, peak broadening and an lowering of the areas. I just keep track of the areas of my internal standards and if they start getting low then i can tell the column is getting worn.

I use MSTFA to derivitize our biodiesel/glycerides, which is a silyating reagent, if what kvaak says is right this could be one reason why my column doesn't last as long, especially if we run samples with high glycerol/glycerides in them. Good to know!

[Peter Apps]

Assuming that you have clean carrier gas and that you are not exceeding its maximum temperature, the lifetime of a column depends on the samples that get put through it. With clean, easy samples like essential oils a column can survive thousands of runs, with mucky, reactive raw extracts containing heavy compounds you might be lucky to get into double figures.

Peter

[AICMM]

AZbiodiesel,

You are in a pretty tight spot. The temperatures you have to run at and the stuff you throw at the column, you will be lucky if you get 400 to 600 runs out of a column. On the other hand, I have worked on 1980's instruments with the original column set (packed's.) Just depends.

One other quick comment, always keep in mind column rinsing and column reversal. Both are good steps to take if you really think your column is the cause of your current problems.

Best regards.

[Ron]

Thousands of injections and good chromatography just doesn't happen when you are running biodiesel or similar analyses. In that case hundreds of injections is the norm. 400 injections while still maintaining good chromatography is actually pretty good.

[thohry]

If you use on-column technique, the column will deteriorate very soon (compared to S/SP inlet)