Alkylated PAH analysis

[Sean]

I have a mixture of PAH in an alkane solution that I am having trouble separating with ordinary methods. Most of the PAH in this solution have a high degree of alkylation making them extremely difficult to resolve (there are a very large number of PAH, and they tend to coelute).

My understanding is that a normal phase column will separate PAH by ring number with little sensitivity towards the alkyl groups. So my plan is to use a normal phase column to fractionate the sample by ring number, then use a reverse phase column to analyze each fraction.

I plan to buy a cyanopropyl column for this purpose, but I am wondering if there is a stationary phase better suited for this purpose. Is there a column that is more sensitive to ring number and less to alkyl groups?

Thanks in advance.

[Uwe Neue]

It is not normal phase per se that tends to separate the PAH according to ring number. You need an amino column to do that.

[Sean]

It is not normal phase per se that tends to separate the PAH according to ring number. You need an amino column to do that.

Thanks for the clarification. I was aware that I needed a specific stationary phase, but I didn't make it clear in the post. I believe my choices are either a cyano or amino column.

I had heard that an amino column would be ideal, but I am also told that compounds with carbonyl groups (which are present in my sample) will bind permanantly to the stationary phase. Is this true?

[Uwe Neue]

I am not aware that cyano columns do this group separation as well. I have some doubts though. Nitro-phases do this though via clear pi-pi interactions.

You are right that amino phases are prone to react with aldehydes and ketones. Since you have such things in your sample, the amino column is not a good idea.

I would go for the nitro-phase, but you may have to dig until you find a supplier. Check with Macherey-Nagel in Germany!

[HW Mueller]

How do phenyl phases or Hypercarb behave with PAH?

[Uwe Neue]

Phenyl columns give more retention for pi acceptors, but I don't think that this will help here.