NMP and tailing

[cawarhur]

I've read that some types of chemicals simply always tail (ex. amines). I'm currently looking at samples with NMP (n-methylpyrolidinone) and the NMP peak tails pretty much no matter what. NMP is an amide. Do amides tail? Is there any credence to the "some things always tail" statement?

Thanks!

[HbJ]

I'm routinely screening for NMP and I don't experience much tailing with it, if at all.

There are some possibilities why those high-boiling solvents tail (excerpt):
- Active sites on column.
- Too thick film (>0,5 ym).
- Inlet problems.

My setup for NMP is usally a DB5(ms) with 30mx0.25x0.5.

[prabhakarpawar]

My method for NMP,
*DB-624,30MX0.53mmID,3.0umfilm thickness.
*Autosampler used for NMP content.NMP peak sharp.

[cawarhur]

Is it possible that NMP's dipole causes it to tail on an HP-5?

I just can't reason why the NMP peak would be the only peak to tail out of all the peaks in our sample...and it's mid-chromatogram, too, so it doesn't seem like injection technique is suspect.

@Both people who look at NMP: what kind of liners do you use?

Straight, taper, double taper?
Glass wool or no?
Liner ID?
Are you running in split or splitless mode?

[chromatographer1]

NMP will tail if they are in solution with or flow by a Lewis Acid. The equilibration process which is forming the base causes the tailing. The tailing you see is either caused by a physical structure through which the sample plug is passing or is caused by a chemical process from the components in the sample.

I hope you find the cause and are able to fix the problem.

best wishes,

Rod

[cawarhur]

If we find ourselves in that scenario, is there a way around it, do we accept the ugly peakshape, or is our data bunk?

[HbJ]

You might fill us in about the details of your methods (temperature, column, liner, injection parameters, brand of GC etc.).

As I've been doing many thousand solvent analyses on several GCs and have very rarely seen tailing I'll give you some of my general advices to try:

1.) Get your chromatography right! Obey insertion distances for your liner/injector and detector(!) by the mm (check your instrument manual). After you've slided the nut and ferrule over your column cut around 3 cm of the column to prevent absorption from small ferrule debris that fell into the column. Also check with a new liner, new septum (Thermo Red for the win!) and a sufficient injector temperature.

2.) Use pre-columns and change them frequently. We have no idea about your matrix but it's possible it contains non-volatile matter. Such contaminants on the column can absorb NMP easily and distort the peak because of the solvent's high boiling point and wide solution properties (plus gank isn't specified as a column coating).

3.) Run test mixes. I'm checking the performance of my columns with Grob's mix as well as acetic compounds (acetic acid e.g.) standards as well as base (DMEA, EDIPA) standards (bases and acids separated of course). I'd recommend to try to a basic mixture like DMEA, EDIPA and TEA (check max temperature of your column first. TEA can be hard to get rid of with low-temperature tolerant columns).

4.) Check your matrix. Rod's hint towards Lewis acids is right to the head (I was more thinking about Lewis acidic sites on the column). This can only ruled out if you tell us more about your matrix.

5.) Use modern, inert columns. Some years ago I had the dubious "pleasure" to test a DB-5 that came from an unopened box from ten years ago. Inertness for bases was a total fail (tailing all over, distorted peaks, small response) as well as massive column bleeding. Go for the brand of your choice

[chromatographer1]

You change the pH of the solution to a more basic condition if the problem is chemical-acid/base in nature, or neutralize any acidic surface in the sample flow path.

You fix the flow path if a physical cause is the root of the problem. Particles in the column, poor placement of the tip of the column in the injector, poor choice of injector design, etc.

There is a reason why it is called RESEARCH, but I don't know if I can remember why.

Just remember, fifty years from now you won't care. (a bit of philosophy never hurts)

Even with great care I was never able to get a perfectly symmetrical peak for pyridine with headspace. I could eliminate most of the tailing, but some tailing was always present. I showed adequate recovery even at 1ng per vial, so life was good. (that is 1ppm of a 1mg sample, dissolved into 25 microliters of water, or 40mg/mL of sample solution)

best wishes,

Rod

[cawarhur]

HbJ:

GC: Agilent 6850 with autosampler
Column: HP-5 30 x 250 x 0.25 (less than 1 y/o!)
Liner: Agilent 4mm ID dual chamber, deactivated glass wool, split liner (changed often)
Inj. Parameters: 10:1 split, T=250C
Septum: Agilent non-stick, long-life septa

(Brand loyalty?)

Also in our samples: DCM, xylenes, 2,4,6-trimethylaniline, and bis-mesityl formamidine.

I'll see if I can't get a chromatogram on here in the next hour or so.

@chromatographer1: I don't necessarily want to think about 50 years from now, but I'm guessing in 50 hours from now I won't care either, so let's stick with that

You guys are great by the way. I appreciate your help!

[chromatographer1]

Hmmmmm

glass wool

Indeed.

Tailing..........

hmmmmmm

best wishes,

Rod

[cawarhur]

So, throw me a bone here, I'm a newbie at this (I wasn't born with knowledge of gas chromatography!)...I should be using a liner with no wool even if the wool & liner are deactivated. Perhaps some sort of baffled liner?

[HbJ]

Liners... choices, choices :p

From my experience Agilent's stock liners are not even close to top-of-breed in terms of deactivation, especially with their glass wool.

To elaborate a bit: NMP tends to stick to everything it touches. It seems to especially like silanol-groups on glass (as in liner glass or glass wool).

After having tried almost all kinds and brands of liners, I'd go for SGE (really good overall) or the Restek Sky (order some with wool, some without). My absolute favorite is their Cyclo liner 23310.1

[chromatographer1]

cawarhur

I hope you never insert glass wool into a liner yourself. (oh, the agony of defeat) You have much to learn about glass wool and active sites, grasshopper.

HBJ has given you excellent advice.

I don't need to add anything to it, but only a caution that trying different brands is a good thing.

best wishes,

Rod

[krickos]

5.) Use modern, inert columns. Some years ago I had the dubious "pleasure" to test a DB-5 that came from an unopened box from ten years ago. Inertness for bases was a total fail (tailing all over, distorted peaks, small response) as well as massive column bleeding. Go for the brand of your choice

Well even less than 1 year old I have the similar experiance as HbJ when it comes to thin filmed HP-5, it sucks for like Deca-BDE as well. Simple test really if you like a DB-5 around or a Restek 5 amine column to verify.

Also done residual solvent testing with DB-624 in the past on this one, no peakshape issues either.

[cawarhur]

Another question along those lines then, if the glass wool/liner is a problem, couldn't the column also be problem? Should I consider purchasing a new column (same type, just not used and abused)? Would a thicker film help?