Chromatography Forum

.

Chromatography of a set of chlorophenolls and internal standard.
Injector 225, oven start 150 deg_hold 0.5 min; ramp 10 degr per min up to 250 degr; Detector FID 325 deg. Column HP- 5 capillary, 8 psig Helium, split 1:10. First peak is internal std, others are analytes. 0.8 ug loaded on column w/ 1uL injection volume.

Can anyone help tell me :

1) why do the peak tail off/sway off to the right side and what can be done to fix this if possible?

2) some of the artifact peaks appeared near the baseline in close proximity to the analyte peaks--so I tried the best manual integration in order to avoid capturing them. Was this appropriate?

3) is the resolution sufficient between the peaks?

You need to improve your peak shapes, the tailing is bad. You need to provide more information, maybe your column and analyte polarity are a poor match (like polar analytes on non-polar column). Is this a capillary column or packed, seems like an old system with integrator. And are you using Internal Standard because you're doing manual injections?

Hi!

I see a lot of tailing.

Is this a standard method?

If yes, that can due to contamination (liner, column). If not, try a higher split.

Your analytes are pretty polar, and you are using a very non-polar column. I would suggest a more polar column, probably a wax type column. Be sure not to exceed the upper temperature limits on a wax column. Most are limited to 260C max during temperature programming, although the SGE SolGel wax column has an upper temperature limit of 300C.

1. The conjecture about the apparent mis-match between analyte polarity and phase type were well-observed. The HP-5 column (upon which the cited data were borne) was the one I available in the lab---and my usual approach to "method development" is to use what I have available and optimize up from there. So, it appears that I probably need to purch a column that has at least 30% to 50% wax ---but short of justifying a high dollar purchase order to upper mgm't there must be a way to improve the chromatography with the materials I have at-hand?
2. A higher split ratio would just lower the peak height and not really affect the tailing (which was computed at 1.4 tailing factor) no?
3. My solvent blank and matrix blank, which were injected at the end of multiple injections of the test STD, were both clean of peaks. Thankfully issues with analyte clinging/carryover were not observed.

You should really have at least three columns, one low polarity, one mid polarity, and one high polarilty if you have to deal with a variety of sample types. A columns is usually around $400, not that large an expense. How many hours have you spent so far on trying to improve this chromatography? I would be willing to bet that the total resources spent on method development so far are more than the proper column would cost.

See below a chromatogram using SLB-5. It is technically the same stationary phase, but shows no peak tailing for many polar compounds. I am sure they have a lot of proprietery column treatment procedure to make the chromatogram look like the way it looks.

http://www.sigmaaldrich.com/analytical- ... -2009.html

I would try astroguilles suggestion of using a higher split. You will be surprised that peak height will not proportionally come down with increase in split ratio. Besides, you are way above the noise levels.

What is the dimensions of your column? If you are using a 0.53mm id column (and 30 m long), your average linear flow velocity at start is about 63 cm/s which is too high. If you are using 0.32 or 0.25 id, the hold up should be above 2 mins.

Cheers,
Suresh.

A couple of things come to mind:

1) Check flow rates including carrier gas flow, makeup gas flow, Hydrogen flow and air flow rate.
2) Initial oven temperature seems too high, try lower it to about 50 C. What's the solvent?

Jumpshooter,

I am going to take a different tack. Put your compounds in a different solvent, like hexane. Even though a -5 has good wetability, it is a bit much to ask it to eat methanol. Granted you are not trying to re-focus (more on that in a second) but I still think you are overloading the early part of the phase and band broadening as a result. If you have to extract in Meoh, go with a 10:1 dilution in hexane and go splitless....

I would also suggest you try a much cooler starting temp. to re-focus. With a -5 and hexane I would try starting at 45C perhaps.

Chlorophenols should be do-able on a 5 since the environmental bobs do it all the time with 8270 type compounds. Therefore, I don't suggest column first, I suggest conditions. Besides, both of these options are cheap to try.

Best regards.

1. Yes, you would certainly win that bet--as I have spent numerous work hours getting the peaks resolved (initially they were all scrunched together) so the $$ spent on a DB-17 would perhaps have yielded the desired outcome. Alas, we are limited here--but good suggestion on eventually getting three "types" of phases.

What can I do with what I have available now?

2. OK, as suggested I will try setting a 1:20 and 1:50 split on re-setting the method for this on Injector and post up results.

Jumpshooter,

What liner are you using for this analysis?

Best regards.

Three phases of GC columns most used in method development?

5% phenyl methyl silicone low polarity

100% Polyethylene glycol PEG high polarity

3% cyanopropyl 3% phenyl methyl silicone intermediate polarity


OV-1701, OV-225, OV-17, OV-2330 are also very useful given the appropriate analysis.

But there is a lot more science than just phase choice in performing Gas Chromatography.

best wishes,

Rodney George
bored consultant

I think your chromatography for this analysis will probably always do some tailing, but there are a few things you can try that might help. What is the length, internal diameter, and phase thickness for the column you are currently using? From the initial temperature, head pressure, and solvent retention time it looks like you are probably using either a short column or a megabore column.

1. You probably will get better peak shape if you inject less onto the column, as noted above the tailing and height do not always decrease in proportion. Try increasing the split ratio to about 25:1, this will do two things that could help, not only is less material being injection onto the column the higher flow rate through the injector should produce a narrower injection band.

2. You may be able to improve peak shape by starting at a lower initial temperature, and by also increasing the column flow rate.

3. Go back to the basics, pull the column out of the injector and detector and make sure you have good cuts on both ends. A badly cut column can have a significant effect on peak shape of polar analytes.

This might be a good opportunity to start working towards getting in a few other columns, you can present this as a opportunity to save money by reducing method development costs in the future. You have a very good example here of how spending a little money up front would have been a net saving. Present in a positive manner and maybe you can get three columns for future use.

Hi Jumpshooter

Chlorophenols are difficult customers.

This is almost certainly adsorbtive activity, which could be in the inlet, the column or both.

In either case increasing the split ratio will definitely make things worse.

If the activity is in the inlet you will see an improvement in peak shape if you lower the column programme starting temperature, or increase the inlet temperature by a substantial margin. If this makes no difference the activity is in the column. If the activity is in the inlet you need to look at reducing any glass wool to the very minimum, and deactivating the wool after it is inserted in the liner, use deactivated liners, and you might need to change them once a day.

Brand new 5% phenyl columns give sharp chlorophenol peaks, but the performance can degrade quite rapidly. Wax columns are more robust in this respect. How much work has this column had already ? If you do not already have them you need oxygen and water scrubbers in the carrier gas lines to protect the column. Any damage to the phase might be concentrated in the first few metres - chop a metre off and see if it makes any difference.

Peter

http://i44.tinypic.com/1zogex5.jpg[/IMG][/img]

I have considered all of your suggestions and implemented the following:
1) adjusted initial temp to 50C, 2) slowed the oven ramp to 5C per minute to effect a better resolved elution of peaks 1 to 3, and 3) adjusted the split ratio to 20:1.
The result is the presented chromatogram. Upon 5 consecutive injections of the sample ("STD W") I observed reduced variances and improved accuracy:
peak # %RSD
1 1.52
2 0.95
3 2.01
4 2.21

Notwithstanding, that there is evidently some mild taing (USP tail = 1.32)at this point I am satisfied with the chromatography. It was quite an undertaking to get all four analytes well-resolved and consistently chromatographed as well as a learning experience thanks to all of you forum writers who lent actionable commentary.