Retention time shifting GCxGC with deans switch

[emi_r]

Hello to all GC specialists,
I work now with the GCxGC, had previously worked with LC. I try to read but unfortunately not get any further We are struggling with shifting RT (on the first column) of all analytes. The chromatogram looks every time as shifted by 0.x min. After several runs during the day, the RT stabilize somewhat but the next day they are different again and then stabilize at a different time.
This is an Agilent 7890B with a deans switch after the first column to the MS or to the purged splitter (which then leads either to the MS or to the FID). Depending on the method we can then do the heartcuts or do a dual measurement with FID and MS in parallel.

I would be very grateful for all hints!

The method is as follows:
GC
GC Summary
Run Time 90 min
Post Run Time 20 min
Oven
Temperature
Setpoint On
(Initial) 60 °C
Hold Time 1 min
Post Run 270 °C
Program
#1 Rate 2.5 °C/min
#1 Value 240 °C
#1 Hold Time 17 min

Equilibration Time 10 min
Max Temperature 325 °C
Maximum Temperature Override Disabled
Slow Fan Disabled
ALS
Front Injector
Syringe Size 10 μL
Injection Volume 1 μL
Dwell Time (PreInj) 0 min
Dwell Time (PostInj) 0 min
Solvent Wash Draw Speed 300 μL/min
Solvent Wash Dispense Speed 3000 μL/min
Sample Wash Draw Speed 300 μL/min
Sample Wash Dispense Speed 3000 μL/min
Injection Dispense Speed 6000 μL/min
Viscosity Delay 1 sec
Sample Depth Disabled
Injection Type Standard
L1 Airgap 0.1 μL
Front SS Inlet He
Mode Split
Heater On 290 °C
Pressure On 297 kPa
Total Flow On 50.636 mL/min
Septum Purge Flow On 3 mL/min
Gas Saver On 20 After 3 min mL/min
Split Ratio 20 :1
Split Flow 45.367 mL/min
Thermal Aux 2 (User Configurable)
Temperature
Setpoint On
(Initial) 250 °C
Post Run 0 °C

Column
Column #1
Pressure
Setpoint On
(Initial) 297 kPa
Post Run 157 kPa
0 °C—325 °C (325 °C): 60 m x 320 μm x 0.25 μm
Column lock Unlocked
In Front SS Inlet He
Out Aux PCM C
(Initial) 60 °C
Pressure 297 kPa
Flow 2.2684 mL/min
Average Velocity 14.408 cm/sec
Holdup Time 6.9407 min
Column #2
Pressure
Setpoint On
(Initial) 240 kPa
Post Run 0 kPa

HP Wax
0 °C—325 °C (325 °C): 25 m x 250 μm x 0.25 μm
Column lock Unlocked
In Aux PCM C He
Out MSD
(Initial) 60 °C
Pressure 240 kPa
Flow 5.5945 mL/min
Average Velocity 94.89 cm/sec
Holdup Time 0.43911 min

Column Outlet Pressure 0 kPa

Front Detector FID
Makeup N2
Signal is modified by Column Compensation Curve #2
Heater On 290 °C
H2 Flow On 45 mL/min
Air Flow On 300 mL/min
Makeup Flow On 5 mL/min
Carrier Gas Flow Correction Does not affect Makeup or Fuel Flow
Flame On
Electrometer On
Valve 1
Name Deansswitch
Other On

Valve 2
Name Stopflow MSD
Other Off
PCM C
PCM C He
Pressure
Setpoint On
(Initial) 63 kPa
Post Run 0 kPa
***Excluded from Affecting GC's Readiness State***
Aux PCM C He
***Excluded from Affecting GC's Readiness State***
Aux PCM C He Supplies Column 2
Valve Box
Heater On 220 °C

[HPLC chemist]

What is your molecular target? What is your sample matrix? How much is your retention time shifting (more than 2.0 %RSD)?

Can you use a thicker film on the column?

[emi_r]

What is your molecular target? What is your sample matrix? How much is your retention time shifting (more than 2.0 %RSD)?

Can you use a thicker film on the column?

Hello!

We analyze essential oils, the sample is diluted 1:10 with ethanol. The RT shifting is about 0.3% RSD, but it's critical for the clean heartcuts. Do you suspect that it is the column?

[HPLC chemist]

Yes. Try a longer column with a thicker film. Do you derivatize the EO before a direct injection (FAME)? Underivatized oils can foul the front of the GC column, which means you will have to cut off the 1st few cm.

[HPLC chemist]

In order to prevent column fouling many EO's are analyzed with a 'headspace (HS)' sampler.

[emi_r]

No, they are not derivatized. Only diluted. Before I remove the column for the first time in my life , maybe I will measure a standard with 2-3 analytes for a few days to see if the RT shifting are still there? Could I maybe bake out the colum at the end of the run?

[HPLC chemist]

Yes, to the bake out. It will work up to a point since this fouling is caused by many non-volatile compounds like underivatized fatty acids (a major component). This youtube video explains the technique of trimming a GC column. You will find a LOT more information on youtube. Even visit the website at restek.com (a major column manufacturer).

https://www.youtube.com/watch?v=0Mne5U2 ... e=youtu.be

[GOM]

First, thank you for providing plenty of useful information

HPLC chemist and rb6banjo are ones who appear to have experience in Dean switching and heart cutting and may have comments about

the RT shifting is about 0.3% RSD, but it's critical for the clean heartcuts.

May I make the following observation

Column #1

0 °C—325 °C (325 °C): 60 m x 320 μm x 0.25 μm

Average Velocity 14.408 cm/sec

HP Wax
0 °C—325 °C (325 °C): 25 m x 250 μm x 0.25 μm

In Aux PCM C He
Out MSD

Average Velocity 94.89 cm/sec

Both of these linear velocities for your columns appear to be outside the normal optimum for separation efficiency for Helium. Perhaps for heart cutting this is not such an issue?

What is your essential oil? I am trying to recover my essential oil database after a few laptop failures

[Peter Apps]

You have a pressure and temperature that are set to zero post-run. Try leaving them at the same setting as during the run. You specify only one column stationary phase - if you are doing heart cutting (or comprehensive GC) you need columns with different stationary phases.

Peter

[emi_r]

First, thank you for providing plenty of useful information

Both of these linear velocities for your columns appear to be outside the normal optimum for separation efficiency for Helium. Perhaps for heart cutting this is not such an issue?

What is your essential oil? I am trying to recover my essential oil database after a few laptop failures

This is a peppermint oil.

The linear velocities and actually all parameters were set by the technician during installation. Do you think I should enter other values? Or maybe they were so adapted because of all the switsches and splitters?
I'm very glad that there are so many experts here I would be grateful for any hint regarding the method and the settings.

Yes, to the bake out. It will work up to a point since this fouling is caused by many non-volatile compounds like underivatized fatty acids (a major component). This youtube video explains the technique of trimming a GC column. You will find a LOT more information on youtube. Even visit the website at restek.com (a major column manufacturer).

https://www.youtube.com/watch?v=0Mne5U2 ... e=youtu.be

Thanks! I'll bake out the column first. Up to what temperature can I go if the maximum for the column is 325 degrees Celsius.

You have a pressure and temperature that are set to zero post-run. Try leaving them at the same setting as during the run. You specify only one column stationary phase - if you are doing heart cutting (or comprehensive GC) you need columns with different stationary phases.

Peter

Do you mean I should set the conditions from the end of the run as a post run? Also leave the high temperature like that? Will that be fine for the column and the system?
The first column is the HP-5: 60m x 320μm x 0.25μm and the second, where the heartcuts are led, is the InnoWAX: 25m x 250μm x 0.25μm.


I talked to my colleague, and he said that the standards (without matrix) have also shifted. He tested it before I came.
Is it true that if the column were contaminated, would only some analytes be displaced? For me, it looks like the whole chromatogram has been moved to 0.x min. From day to day in different direction and sometimes even weaker.

Is there a Agilnet measurement file where you can read all the pressures taken during the measurement? So that I can compare whether the conditions in the device are stable?

Excuse my English

[Peter Apps]

Usually in GC the temperatures that are programmed are set to return to their initial values after a run (for example if your column is programmed from 40C to 200C it would return to 40C,) and temperatures that are not programmed stay they same pre- during and post-run. Similarly for pressures. By having post-run settings of zero you make it more difficult for the system to stabilise.

Is the shift in retention time exactly the same for all the peaks on the chomatogram ? If it is then it points to problem being with the timing of the injection, or something switching at the beginning of the run - anything to do with gas flow control or temperature will have different effects at different places on the chromatogram.

Peter

[emi_r]

Is the shift in retention time exactly the same for all the peaks on the chomatogram ? If it is then it points to problem being with the timing of the injection, or something switching at the beginning of the run - anything to do with gas flow control or temperature will have different effects at different places on the chromatogram.

Yes, that's what makes me wonder. The retention time shift is the same at every point of the chromatogram. This can already be seen in the solvent peak and at the end of the chromatogram, the last peak is shifted in the same way as the first and the middle one. I have already read that the temperature does not affect all analytes equally. But did not know that it is the same with the pressure .... The chromatograms are really identical except for the shift. You could cut out all the chromatograms, move them a bit and put them on top of each other - they would be identical.

How can I find out if the start signal is set correctly?

[Peter Apps]

A shift of a whole chromatogram is actually really unusual, and can only be related to something happening at slightly different times at the beginning of the run. The most likely culprit is that column 2 post run pressure is set to zero post run - when the run starts it has to jump to 240 KPa, which it will never be able to do as repeatably as you want. This is putting a variable back-pressure on column 1 just as the injection is being made, which could well shift the whole chromatogram. How repeatable are your peak areas ?, and does peak area correlate with retention shift ?

Peter

[emi_r]

A shift of a whole chromatogram is actually really unusual, and can only be related to something happening at slightly different times at the beginning of the run. The most likely culprit is that column 2 post run pressure is set to zero post run - when the run starts it has to jump to 240 KPa, which it will never be able to do as repeatably as you want. This is putting a variable back-pressure on column 1 just as the injection is being made, which could well shift the whole chromatogram. How repeatable are your peak areas ?, and does peak area correlate with retention shift ?

Peter

Thank you Peter!
Do you think I can change the pressure to 240 kPa at the end of the run (post run)? Will the pressure be set to 0 after the post run? Or are the settings loaded directly from the next run?
I have now figured out the function of how to record the pressures in each run. I will record them from now on and compare them.
I'm working in constant pressure mode, so all pressures need to be stable during the run, is that right?

[emi_r]

How repeatable are your peak areas ?, and does peak area correlate with retention shift ?

The peak areas are quite repeatable. I insert a picture.