Need some GC clarity

[rbjenkins]

So I've been going back an forth between reads and forums bout the age old questions of Constant Pressure vs Constant Flow in the last few days. If any other pros out there could help it would be greatly appreciated. I'm not new to GC, but there is too much conflicting information and not enough specific info on what I'm looking for. Sorry for the long post...

I currently work with Waters Empower 2, Agilent 6890's and 7890's at a new company I've moved to but have had years of exp at my previous work place, a contract lab, and yet at this company it seems as though I'm missing something or need to start correct procedures.

At the contract lab we used Waters Empower 3 with the same Agilents and had all but two of the GCs set up with -S/SL EPC rear inlet -PP EPC front inlet -TCD rear detector -FID front detector so that we could maximize capabilities and versatility with each instrument, if this info helps.

I was taught through the previous company:
Capillary column: Always try to use a S/SL EPC injector, unless you have to use an On-Column injector, and run with Constant Flow. Set dimensions of capillary column appropriately.
-- Reason: allows the EPC to control the flow through the small diameter much more effectively than with pressure due to the much lower values in pressure making it difficult to change and maintain during an oven temp ramp.

Packed column: Always try to use a PP EPC injector and run with Constant Pressure. Set dimensions of packed column to length and max diameter (750um) and measured flow at the detector with an external calibrated flow meter.
-- Reason: allows the EPC to control the pressure through the much larger diameter more effectively than with flow due to the over exceeding values in mL/min from the flow controller and read outs on the column. Also because it only increases the pressure at the head end of the system and not throughout. (ex: common method at new company uses 2m x 4mm packed column set at ~60psi to give 80mL/min but the readout says 2000mL/min on the column readout).

I had no trouble with columns and method until I got to this company and I was constantly being corrected. Most of their methods run packed columns at very high flow rates (methods made in the 80's and 90's) and they run their systems without column dimensions and under constant flow. We never ran without dimensions because we were told that the EPC and only compensate for what it's programmed for. It works but I tested the flow with an external flow meter and it's off anywhere from 2% - 10% depending on the flow demand and temperature (higher flow more % off). I don't know if this is a fluke because of the poor preventative maintenance that I've seen or if it's actually incorrect method execution.

Any feedback would be appreciated.

Rob

[GasMan]

With packed columns it is recommended to use constant flow and NOT to put any dimensions in for the columns. If you put in dimensions, an Agilent GC will assume that you are working with an open tubular column, and all flows will be CALCULATED. No dimensions are interpreted as 'I am a packed column', and when you choose constant flow, you will be using the built in flow sensor of the pneumatic controller to actually control the flow. This make for easier set up, you just enter the flow that you want, and you do not have to worry about pressure programing if you are temperature programing and wish to keep a constant flow in your column.

Gasman

[Stunt]

GasMan nailed it. Don't input dimensions unless you want your flows calculated for you. I really hope you're not clipping that packed column...

[Peter Apps]

Although a huge amount of attention is paid to flow rate, GC is actually remarkably robust to departures form the so-called optima. Which is just as well, because the gas is flowing at the optimum velocity at only one point along the column. Although you can measure a 10 % departure from the set flow I very much doubt that you could measure its influence on resolution.

It is, of course, very important that the flow be repeatable from run to run so that retention times are repeatable.

There have been lots of discussions about carrier flow - try a search of the archives looking for posts by lmb.

Peter

[rbjenkins]

Thank you guys for the replies! Looks like I'll be going through some thorough cleaning procedures after finding a liner and septum in a S/SPL inlet that had not been changed in 5 months So I assume it's probably just a fluke of the system. Time to start some over due PMs.

Thanks again,

Rob