Chromatography Forum

First of all thanks for a great forum. Even thou this is my first post I have been reading here a lot.

I’m a student of chemistry, and right now I’m working on a project about changes of the fat composition of the body of people developing diabetes 2. My assignment is to develop an easy to use method for quantification of the levels of mono-, di-, and triaglycerols in both fat and muscle samples.

The equipment I’m working on is a HP 6890 GC with a HP 5973 MSD system for detection. The software is MSD ChemStation D.03.00.611. The column in the GC system is a Agilent HP-5ms, 30m, 0,25mm, 0,25um. I have my injection temp at 300C, 1:5 split, oven starting at 100, 1 min hold, 20C/s to 325 and hold for 30 min. The solvent is hepthane, and flow is 40 cm/s.

My problem is, that when I inject a sample of 4 acylglycerides, monoolein(C18:1), 1,2 and 1,3-Diolein, Triolein, derivatized by BSTFA to trimethylsilyl ethers, I only see one peak, monoolein. I have read a few articles in witch the have no problem eluting even the triaglyceroles at an oven temp of only 325 C, but I can not replicate their results. I have been thinking of buying a new column, possibly one of the new Phenomnex inferno columns capable of handling 430 C, as the one currently in use is, possibly, from 2001. Du you think this would solve my problem?

Best regards
Rasmus Forsberg

We typically use a METAL capillary for the derivatized mono- and diglycerides, and the triglycerides, can see glyceryl tristearate no problem. These columns are designed for temperatures up to 400C, have a thin cross-linked dimethylpolysiloxane film, and are either 10m or 15m, all design elements for large molecules such as these. See Phenomenex, Quadrex, others for these.

What is the phase of the column again?

Tricacyl glycerides are a problem due to several factors: 1. labile (they can breakdown. 2. the volatility of the compounds is very high so 350C may not be enough. 3. Need to look at the column they use (probably higher temps used.

AOCS (American Oil Chemisty Society) method Cd 11b-91 covers mono- and di-glycerides. Tri-glycerides are very hard due to high boiling points.

What is the phase of the column again?

HP-5 (HP-5ms) 350C not enough, use thin cross-linked dimethylpolysiloxane film coated on metal capillary like I stated above.

Tri-glycerides are very hard due to high boiling points.

Sounds like I've had better success with triglycerides than Phycal.

Unsaturated lipids are quite unstable, so first ensure that your standards are still true-to-label, as they often polymerise. Human lipids can be a mixture of complex and unsaturated fatty acid molecules, so if you can't get the stds to work, you're going to have more problems with samples.

Second, W.W.Christie has some useful tips on general techniques.
http://lipidlibrary.aocs.org/GC_lipid/gc_lip.html

Third, visit a library and obtain some of the papers by others who have researched human acylglycerides using GC. You may find it easier to fractionate samples into the neutral MAGs, DAGs, and TAGs, and then derivatise and analyse each type.

We typically use a METAL capillary for the derivatized mono- and diglycerides, and the triglycerides, can see glyceryl tristearate no problem. These columns are designed for temperatures up to 400C, have a thin cross-linked dimethylpolysiloxane film, and are either 10m or 15m, all design elements for large molecules such as these. See Phenomenex, Quadrex, others for these.

What advantages are ther to using a metal column instead of the Inferno column here. http://www.phenomenex.com/cms400min/zebronzb5ht.aspx

What is the phase of the column again?

5 % -Phenyl 95 %-Dimethylpolysiloxane

Unsaturated lipids are quite unstable, so first ensure that your standards are still true-to-label, as they often polymerise.

The standart is a mixture of the four lipids. Just recived from Sigma-Aldrich last week, so brand new.


I know "The Lipid Libary", very usefull site.

You may find it easier to fractionate samples into the neutral MAGs, DAGs, and TAGs, and then derivatise and analyse each type.

A lot papers seperate the lipid samples in MAG, DAG and TAG before GC analysis, but I would like to develop a method that dosen't require these time consuming steps.

Apart form its being the default column that comes with the instrument is thetre any reason to use the column that you have ? Thinner films, wider bore and shorter will get the compounds out sooner without having to go to temperatures that may compromise the samples.

Peter

The HP-5 phase seemes to be the most widely used phase for this type of analysis, but som uses a 1% Phenyl phase. Then again few uses 50% Phenyl or various wax-types. Wax types are mostly used if people wants to separate the various unsaturated subtypes.

I realise that the dimentions of my column, 30m, 0,25mm, 0,25um, is probably not optimal, but they don't seem to be very different from what is used in a lot of the artikels i read, so i expected to see at least the DAG's, with a reasnoble retention time. But then again often they have some parameters witch are better for this type of analysis, e.g. wider bore, thiner film or oven temps up til 360C.


This column being the only one this lab's got is a pretty good reason too.


What do you think of my thourghts on bying a new column? It would probably be the Inferno I linked to earlier with the dimentions 15m, 0,32mm, 0,18 um.

Should i maybe use a even thinner phase to avoid higher oven temps?

Hi Rasmus

On a system in poor enough condition it would be possible to elute all of the components as one, undifferentiated lump. If the column really has been in use since 2001 it is almost certainly past its useful lifetime, and you are going to need a new one. If the main aim is to elute heavy molecules, and resolution is not a major constraint then a short, thin film column is the direction to go in. Since higher final temepratures will get the job done quicker, then one of the specialist high temprature columns seems like a sensible choice. How about inlet maintanance ?, when was the liner last changed ? Source clean on the MS ?.

One minor operational point - are you running with a constant gas flow rate ? If not the carrier flow will slow dramatically as the column temperature rises, and retention times will be correspondingly longer.

Peter

Hi Peter

It would be nice to be able to separate the DAG and MAG after chain lenght and maybe saturation, so I don't want the column to be so short that this is impossible.


The GC and MS was serviced last week by a technician from Agilent, but this resultet in a leaking injection port I had to fix. As I am the only chemist in the lab, and just recently got transfered here, the GC system suffers from lack of maintance, as nobody dares touch "the magic machine". I don't mind taking the machinery apart, but everyone else seem to, so I get a lot of experience in this.

I run constant flow rate, 40cm/s. In the service the septun, liner, god seal and filament was replaced.

Rasmus

You might wish to review columns on the web that are used for BIODIESEL analysis which elute your analytes in the 20 min time range but use 0.53mm ID and 0.16µ films of 5% phenyl phase in a metal column.

Having done research with these columns I could recommend both Varian and Restek web sites for you to review their published chromatograms.

I also believe they use on-column injection which is why I would suggest you use the megabore ID or use a deactivated union to connect a megabore precolumn to a 0.32mm ID analytical column.

best wishes,

Rodney George
consultant

Hi Rasmus

A 530um column requires volume flows that are probably too high for the MS vacuum pumps, so you would need to have a flow restrictor at one end of the column. Varian did some work on the benefits in terms of speed and resolution of running a megabore column under reduced pressure. Jack Cochran et al published a paper on plumbing restrictors into inlets - I don't recall the citation I'm afraid, but it might be available through the Restek web site. Restek make an inlet liner that can do either on-column or direct injections to 530um columns, I've never tried one.

Peter

What is the phase of the column again?

HP-5 (HP-5ms) 350C not enough, use thin cross-linked dimethylpolysiloxane film coated on metal capillary like I stated above.

Tri-glycerides are very hard due to high boiling points.

Sounds like I've had better success with triglycerides than Phycal.

We have higher boiling Triglycerides in the samples I work with, so 350C may not be high enough and I really don't want to remove our workhorse FAME analysis column to go higher in temperature (max 250C for this column).

Peter is right about the MS pumps. I forgot you were using MS for detection as I was focused on the retention and delivery of the triglycerides onto the column.

Getting triglycerides volatile enough to enter the column with a split injection is hard enough without loss due to decomposition, and then not discriminate is (almost) impossible. That is why on-column is used so commonly in analyses like biodiesel.

Good luck.

Rodney George

Just a little update on my project.

I have had a little vacation, so hence my lack of updates.


Phenomnex have very good support here in Denmark, so I spoke to them about my problem and my interests in a new column. I ended up ordering a Zebron ZB-35HT 30m x 0.25mm x 0.1um Capillary GC Column. I would have liked a 0,32 mm column, but it was not available. Hopefully the phase, 35 % phenyl 65 % Dimethylpolysiloxane, will give me good separation of the fat subtypes. The column is capable of handling oven temps up to 400 C. If I have good enough separation but need shorter retention times I can separate the column in a 10 m and a 20 m piece.

I will probably receive the column next week, and have a few samples through it in a day or two.