Chromatography Forum

If I run my alkane calibration mix several times in a row, I find that the retention times of all the peaks will be shifted plus or minus two seconds from run to run. The shift is consistent across the entire run (i.e. every peak is shifted the same amount), so I am wondering what could cause this? I am using a 5890 Series II plus with a 5972 MS. Two seconds doesn't create any big problems for my identifications, but just the same I'd like to get to the bottom of this. The column is a 30m x .25 x .25 HP5MS using hydrogen as a carrier gas. I've run maybe 100 samples on this column since new.

What are your temperature and flow/pressure programmes for inlet and column, split or splitless injections or something more fancy like PTV ? What are the column dimensions.

Peter

What are your temperature and flow/pressure programmes for inlet and column, split or splitless injections or something more fancy like PTV ? What are the column dimensions.

Peter

Column dimensions have already been given.

Additionally what is injection volume, solvent and inlet liner volume ?

Oven temp is 55C-246C 3C/min
Inlet is 220C
Transfer line 280C
Flow is .85ml/min starting at .5psi (approx) carrier is Hydrogen
Liner is plain 4mm liner with glass wool, (4x78mm)
Split 1:120
Injection volume is .5ul
Solvent is Hexane

It is likely to be a pressure/flow control problem. 0.5 psi is too low for the flow module to control properly. The reason the pressure is so low is the very low volume flow rate, which is also making you analyses take more than twice as long as they need to, and is therefore cancelling out the reason to use hydrogen as carrier whichis faster analyses wit the same resolution. Increase the volume flow rate to 2 ml/min, or set the linear flow rate to 50 cm/s and retention times will be shorter and stable.

Peter

Interesting

Feeding in your conditions into the Agilent flow calculator

30m x 250um x 0.25um

Oven initial 55°C
Carrier Hydrogen
Inlet pressure 0.5psi
Outlet 0 psi ( assumed) using a 5972 MSD

would indicate that you are already within a good optimum linear velocity at 54cm/sec for hydrogen with the ms 5972 outlet and that you are using the correct calculated inlet pressure of approximately 0.5 psi at your starting oven temp of 55°C

"The shift is consistent across the entire run (i.e. every peak is shifted the same amount)"

Although you appear to be using the correct flow/linear velocity/pressure parameters, I would agree with Peter's comment that with your 5890

" It is likely to be a pressure/flow control problem. 0.5 psi is too low for the flow module to control properly."

That would indeed be consistent with and be a likely explanation for your observed slight variations of +/- 2 seconds in retention time

I appreciate that this is just an academic exercise for you

Perhaps try increasing your inlet pressure to 5psi with a resulting linear velocity of 70 cm/sec. Still within acceptable limits for efficiency with hydrogen as a carrier gas on the van Deemter curve at your initial oven temp. By 246°C (?) you will be at an even better 51cm/sec.

Let us know the result? It may help to eliminate some causes.
I have assumed that you are not doing pressure/flow programming.

As always, please check my calculations

Interesting

Feeding in your conditions into the Agilent flow calculator

30m x 250um x 0.25um

Oven initial 55°C
Carrier Hydrogen
Inlet pressure 0.5psi
Outlet 0 psi ( assumed) using a 5972 MSD

would indicate that you are already within a good optimum linear velocity at 54cm/sec for hydrogen with the ms 5972 outlet and that you are using the correct calculated inlet pressure of approximately 0.5 psi at your starting oven temp of 55°C

"The shift is consistent across the entire run (i.e. every peak is shifted the same amount)"

Although you appear to be using the correct flow/linear velocity/pressure parameters, I would agree with Peter's comment that with your 5890

" It is likely to be a pressure/flow control problem. 0.5 psi is too low for the flow module to control properly."

That would indeed be consistent with and be a likely explanation for your observed slight variations of +/- 2 seconds in retention time

I appreciate that this is just an academic exercise for you

Perhaps try increasing your inlet pressure to 5psi with a resulting linear velocity of 70 cm/sec. Still within acceptable limits for efficiency with hydrogen as a carrier gas on the van Deemter curve at your initial oven temp. By 246°C (?) you will be at an even better 51cm/sec.

Let us know the result? It may help to eliminate some causes.
I have assumed that you are not doing pressure/flow programming.

As always, please check my calculations

The last part will depend if using constant pressure or constant flow mode. If using constant flow mode the pressure will increase to keep the flow constant with increasing oven temperature.

I would agree that it is probably the low pressure that is causing the shifts, but it is difficult to use higher pressures since the pumping capacity of the 5972 is not very high. Even 2ml/min is going to result in fairly high source pressures with hydrogen carrier, though the oil diffusion pump on the 5972 will handle hydrogen more efficiently than a similar sized turbo pump will.

With mass spec, you don't gain much reduction in run times with hydrogen since you are limited on total flow through the mass spec, unless you are using something like the 7000QQQ which has a much larger pumping system. Hydrogen carrier is to save money versus using helium while having mostly the same run times.

A 5972 has a flow limit of 2ml/min for helium, and it pumps it out more efficiently than it does hydrogen. Not sure if the 5890 can operate with pulsed injection or not, been too long since I used one to remember, but this would be a good situation to use it if possible.

Is it possible to use a restrictor with the EPC so I can increase the pressure at the same flow? The manual mentions restrictors for auxiliary EPC.

Also, I am using the 5890 in constant flow mode.

James was correct in pointing out my error of not taking into account that 1.7ml/min of hydrogen may well exceed the pump capacity.

How much time does your method wait at the starting temp. before injecting the next sample? It's possible the oven temp is not stabilized when the next injection starts.

Is it possible to use a restrictor with the EPC so I can increase the pressure at the same flow? The manual mentions restrictors for auxiliary EPC.

Also, I am using the 5890 in constant flow mode.

Restek did some work on putting restrictors at the head of the column to allow the whole column to be under vacuum. It was a couple of years ago, but it might still be on their web page.

I wait 1min at the start of each run.

I wait 1min at the start of each run.

Is that one minute initial oven temperature or oven equilibration time? There are separate settings for those. Most default equilibration time to 0.5minutes but you might want to extend it to 1.0 minutes.

Another thing to look at is the flaps on the back of the oven. If those are not closing completely then it could be causing some fluctuations as the oven heats, but if the shift is the same amount of seconds for each peak that is probably not what is happening.