Chromatography Forum

Thanks rod, I'm trying to consider all the possible answers.

Can you describe me the cleaning procedure you told me before, so I can ask the vendor about it.

Thanks, for everyone.

During the manufacturing process of a porous polymer the product is taken from the reactor and is 'washed' by a proprietary process to remove residual monomer and catalysts, etc from the interior pores of the beads.

If this is done piecemeal, in an inadequate manner, the performance of the polymer beads is poor and excessive shrinkage occurs when the beads are heated, giving a column of poor chromatographic performance.

This 'washing' should be done (there are many ways to do it) thoroughly to minimize these beads issues. Supelco is familiar with the process I gave them. I don't know if other vendors can do this or not. I no longer work for Supelco and don't wish to give an impression I am critical of other vendors or that I presently work for Supelco. (trying to be diplomatic and noncommittal, here).

While Restek offers wire bundle terminations (not my favorite) Supelco does offer screens and frit terminations. I like frits the best, myself.

best wishes,

Rod

As Rod says the (micro)packed route is cheaper. The method we use is in-house but really you can just look up the J&W alumina PLOT example chroms.

The vendor has offered me Stainless Steel Packed Column, 2mm i.d., 1/8in o.d., 6 ft or 12 ft. Which is more appropriate for my application? Should I ask for a smaller i.d.? or different length? material?

I have been looking some chromatograms and it seems that this column (porapack N 80/100) can be used to measure CO2 (in other conditions and using a TCD). May be in a future, I can use the column for that too.

You are doing trace analysis. 2mm ID is too big. 1 mm ID is much preferred. You should have the smaller ID so your head pressure will be higher. This will better compress your sample volume (0.5-2.0mL) at the head of the column, giving you a better sharper peak (lower detection limits). Don't forget your frit terminations (reduction of intra-column void volume) to improve your column performance, also.

If you can't get 100/120 mesh take 80/100 mesh. Same reasons as above.

I would want a minimum of 6ft but I would prefer a 10ft (3m) column for general use.

Yes, this will be a wonderful column for CO2 analysis also. It is nicely spaced equally between methane and ethylene. (Q has it much closer to methane)

Get what you need, not what they want to sell you.

Rod

Obvious, thank rod.

I asked for the 80/100 because you suggested it before, should I ask for the 100/120?

Johnny Rod

I don't wish an argument, but I do see concerns with recommending an alumina capillary column for this analysis.

So please, tell me. How long will the alumina capillary be inert to absorption of olefins and alkynes when wet samples will be injected upon them? How much ethylene will be lost?

For decades alumina columns have been avoided in process analyzers because an ACCURATE measurement of trace amounts were needed, not just getting a peak through a column, ignoring the losses and the inaccuracy. Even gas standards in a cylinder need replacement, they don't remain stable forever.

If the deactivation of the alumina is complete (is this possible?) will it be durable? , and for how long? If you know the answers to these questions, I and many others would enjoy your sharing this information. Things do change and I would hate to be the last to know (seriously).

It is easy to create beautiful chromatograms. I have done it myself. But no tool is perfect and no tool lasts forever. But there is often a better choice, or a best choice.

Trace ethylene analysis has been performed with porous polymer columns for decades and as far as I know, no capillary column is superior. But I can be taught.

best wishes,

Rod

It isn't a critical issue. The only significant gain will be a higher head pressure, and you may not need this. You can do just fine with 80/100. And given the same head pressure you WILL get a faster flow (faster analysis) with the 80/100.

Just INSIST on a 1mm ID column and clean packing and frits or screens.

Condition your column slowly from ambient to 140C and hold overnight before you start running samples. Use a 20cc/min flow and insert isothermal holds of 60 min into your temperature program at 80X, 100C, and 120C.

You should then have an excellent column.

best wishes,

Rod

I have found this chromatogram that I show below,



It seems that with a short column (1m) and a 2.3mm ID column the sample can be resolved well. What are the advantages for my application of have a longer and with a smaller ID column? I know that longer column mean better resolution but is it necessary? Longer columns also mean longer elution times.

Rod, I know you have a lot of experience and knowledge, maybe I am forgetting something, or I don't know. I don't want what I said sounds like I think that you gave me bad advice or something like that. I just want to discuss so I can take the best decision possible and increase my knowledge.

Note the lower temperature and the relative broad peak widths.

You can certainly possibly get by with a 2m column (maybe even a 1m column) but you will not be able to compress your initial sample plug due to the lower head pressure involved. You want to measure accurately low trace amounts (less than 1 ppm) not merely detect such amounts.

You could also use a 1 ft. 1/4" 4mm ID column to separate 125 microliters of sample at percent levels with a 10 psi head pressure.

A 3m 1mm ID column will require 60-80 psi to squeeze a 1000 microliters of sample into the same peak plug length at the head of the column. (8X pressure, remember the pressure-temperature-volume equation?)

Tell me, how will this effect your LOD at the tail of the column?

The best way for you to learn is to try different lengths, flows, and temperatures and find out what works best for you. If you got the money, I got the time.

Good luck,

Rod

The vendor just told me he doesn't have packed column of 1 mm ID...

It isn't so bad the 2mm ID?

I'm asking to other suppliers but, here there isn't a lot of options...

Where are you located?

You have not tried Supelco or Restek who are in the US but ship worldwide.

What I have described will not work or will not work as well as a 2mmID. I do not recommend a 2mm ID. Your headspressure will be much less with a 2mm ID.

Good luck,

Rod

Sorry for the delayed answer, I'm still waiting for some vendors answers....

I'm from Uruguay (third world country). I only can purchase items from local suppliers (I don't make the rules in my workplace), so buy directly from Supelco or Restek is not an option.

I have a doubt, do packed columns allow split/splitless injection? or only direct injection? which is better for my application?

I am sorry to hear you are limited in your options. That makes your work difficult to do well.

Split injection is possible with a micropacked column (1mm ID) if you have the right graphite ferrules and nuts. The technique that allows you to get the most sample onto a column while not overloading your column is best. If you want to measure trace amount accurately and as low as possible.

Normally you have different injector parts. Perhaps you should describe your GC injector and detector hardware. Are you injecting by a gas sampling valve? Are you using a syringe with a split/splitless injector?

If you only have a capillary injector and syringe then use a 1-2mL splitless injection with a alumina capillary column and try to get the most sample on column you can. When your peak broadens rather than gets taller then stop increasing your sample on column.

best wishes,

Rod

The suppliers are not offering me 1mm ID columns. Are you sure there are packed columns of 1mm id available? I didn't find them in the catalogs.