Chromatography Forum

Dear All,

I would like to request your thoughts and opinions about the differences and relative advantages of On-Column and PTV injection for the analyses of triglycerides.

We know OC requires retention gap and that PTV may degrade some compounds. Any other opinions?

Also, we would like to hear about the GC instrumental requirements for HT-GC work such as, temperature controls, pressure or constant flow controls, septa, injection liners, etc.

Thanks,

josebenjamin

You can use whatever works for the lipids you are interested in.

For "highly saturated" lipids ( eg coconut oil ), almost any system will work - providing it is inert.

For "polyunsaturated oils" (eg Omega-3 fish oils), the chances of consistently getting them intact through a long column at high temperature are fairly remote, regardless of injection system type and inertness.

Most other triglycerides are somewhere between those extremes. I've favoured injection small quantities ( split injection ) onto a relatively short ( 5 metre ), narrow bore ( 0.1mm ID ), thin film ( 0.1um ) non-polar column, but the literature abounds with alternatives.

Bruce Hamilton

Another thing to consider is the cleanness of your samples. For dirty samples PTV has the advantage that it is easier (?) to change an inlet liner than to change a retention gap. PTV is also easier to implement with an ordinary autoinjector.

Peter

Though not having worked for some years with lipids there is still the curiosity:
josebenjamin, why would a PTV have increased tendency to degrade some compounds?
Bruce, is it known why/how polyunsaturated TG degrade? Are temps too high here? In work of the early 1980´s I didn´t see any thermal degradation of FA at up to ~280°.

Hello Hans

There does not seem to be any reason why a PTV would degrade compounds any more than an on-column injection, given similar temperatures. Nonetheless, in practise people tend to ramp the temperature of a PTV ten to 100 times faster than a column temperature programme - hence the analytes suffer more thermal degradation. With slower ramps in the inlet you get less degradation, and the peaks are still sharp - you only need very rapid heating if the inlet contains adsorptive material.

Regards, Peter

Hm, if the ramping rate is responsible than conventional injection, into a constant temp injector, should be degrading the stongest? I always thought that the faster you get the evaporation done the less time for degradation. Could it be that residence time plays a cruical role here? (I did some pioneering work on this in the early 80´s, if I remember correctly, with ramping times of 0.2 sec or so..... The whole thing was shelved as the university, etc. didn´t help me to get a patent right, etc., etc., etc. At that time German universities still considered it a sacrilege to do something commercial or even work with industry as I tried at the time).

Dear All,

Thanks for your comments. It seems like there is no clear concensus about what works better for lipids. On Column is regarded as the "safest" method to inject, but prone to contamination and requires a retention gap for good peak shape. PTV is more likely to degrade, and the extent of degration could be a function of residence time and temperature, and catalysis on the surfaces. Thermo has a PTV with specially developed silicosteel surfaces that claims to be more inert. PTV is also regarded as easier to use and to clean.

Any other ideas?

josebenjamin

Degradation is almost certainly a function of temperature and time. Given a splitless transfer to the column, analytes can be in the inlet for tens of seconds. With heating rates of 10K/s (or higher on some PTVs) temperatures can be up to 300 - 400 degC while analyte is still in the inlet. The guideline I use is that the PTV should be programmed to reach the column programme maximum within the splitless time. Since splitless times with PTVs tend to be longer than with conventional hot inlets the heating rates are usually 1 or 2K/s.

Hans, did you publish any of you rinlet work ? If you did, could you e-mail (papps@csir.co.za) me the references ? Thank you.

Peter

Though not having worked for some years with lipids there is still the curiosity:
Bruce, is it known why/how polyunsaturated TG degrade? Are temps too high here? In work of the early 1980´s I didn´t see any thermal degradation of FA at up to ~280°.

It's about a decade since I worked with lipids, mostly on an HP5792a with split and SGE on-column injectors and FID.

The degradation of my samples appeared to be a mixture of thermal cracking and polymerisation. Any trace of oxygen or inlet activity will greatly accelerate decomposition, and quickly kill partially polar columns. An inert system would separate lipids well for weeks.

There usually is an increase in baseline signal at final temperatures when PUFA TAGs degrade, indicating a gradual elution of decomposition products.

The elution of a full profile ( FFA, MAGs, DAGs, Sterols, TAGs ) usually takes about 20 mins, and the unsaturated TAGs react in the injector and column and don't elute, even the smaller TAGs ( C46 - C54 ). The injector temparature was around 300C and the column went from 200 to 340C. Marine PUFA lipids ( C20:5, C22:6 ) tend to degrade in the injector.

Saturated TAGs are thermally stable and were used for frying food, however similar-sized polyunsaturated TAGs produce peaks that are much smaller, even in the absence of Oxygen.

Bruce Hamilton

Josebenjamin, from Peter´s and Bruce´s statements, + from my experience I would conclude that if you do it right (regarding temp., "residence" time, catalysis and amts injected) one should be best off with a PTV.
It should be incomparibly better than any other method if you do it this way:

Hook up, in series, a two columns (Dean´s) system to the PTV. Inject the sample on reverse flow of the first column so that sovents are swept out of the PTV. When that is done go to a forward flow (very high rate) through the PTV, into the first column and ramp up the temp. to a max of 300° in less than 1sec. After the right time the flow is normalized and one can take a heart cut, etc.

Peter, there are no refs on this, but a fair amt of material which was send to the Fraunhofer Gesellschaft for the purpose of getting support in obtaining a patent (after the university claimed that they had no money). Fraunhofer G. deemed it patentable , but since the Gerstel KAS and Siemens PTV had appeared, they said that they could probably not get enough roylties on the patent. The end. There is not enough energy and resource to continue alone (my GC is dead from overuse and old age).

A bit more history: As far as I can tell it was me who pried Gerstel (Gerstel sr in person) and a Siemens engineer to go in this direction. It seems, though, that I might have activated connections to Schomburg (at the Max Planck in Mülheim, where I had a stipend before working with GC in Giessen), as both Gerstel´s and Siemens´devices reflected his ideas more than mine. Siemens superiors never really supported the engineers that developed their PTV (no advertising) so it was quickly pulled from the market. Soon after this the Siemens analytical GC disappeared as well.
Strangely: the system closest to mine later appeared as the ATAS OPTIC (now PAS?), even though I had no connection there. None could be moved to modify their stuff to work with a Dean´s system as eplained above.

Thermo have published a comparison paper of their PTV and Cold On Column for use on fats and waxes
Fats

Thanks for pasting the link to the PTV vs on-column comparison. It is an example of a situation where rapid heating of the PTV inlet is appropriate because the starting temperature of the column programme is much higher than the initial inlet temperature. Once the inlet has caught up with the column temperature the inlet only has to be about 10 K hotter than the column at any given time for peaks to come out sharp (as long as the inlet is clean). With these temperatures any compound that survives its trip down the column would not be degraded in the inlet.

A further thought - so that automation is easier most "on-column" injectors need a piece of 530 um column to inject into. This is connected to the analytical column by a press-fit connector. How is this different chromatographically from a 1 mm i.d. inlet liner with a press-fit taper connection to the column, and how is that different from a 2 or 3 mm i.d liner with a press fit taper connection to the column ?

The history of GC inlet development is a case study for a sociologist of science. I wonder how many of the younger generation in GC think that the commercial hardware available today delivers the best performance that has ever been achieved ?

Peter

How is this different chromatographically from a 1 mm i.d. inlet liner with a press-fit taper connection to the column, and how is that different from a 2 or 3 mm i.d liner with a press fit taper connection to the column ?

The Thermo Cold On Column is the original Grob designed device with secondary cooling. In this method the injection actually occurs inside the oven inside the pre-column. The oven temperature should be above the boiling point of any solvent and air is directed down the column to cool the top most part of the column. As the flooded volume reaches the point where the cooling effect of the air is countered by the heat of the oven evaporation of the solvent occurs. Vapour forcing back up the column is recondensed and your compounds of interest should be focused in the pre column. This is the gentlest method of introducing the sample as the compounds of interest are never exposed to a raised temperature to force them into the system (as in PTV and temperature tracked on-column devices). The HOT device used in the paper details an extended version of secondary cooling allowing oven temperatures up to 100C above solvent boiling point to be used.

Assuming that the temperatures etc are the same in the different systems, and back diffusion in wider tubes excepted, does the diameter of the tube at the point of injection really make any difference ?

Peter