Question of equivalence of liquid and gas calibration in GC

Discussions about GC and other "gas phase" separation techniques.

11 posts Page 1 of 1
Hello,
Is it possible that liquid and gas calibration of a species yield the same area and residence time, using GC?

I have a Shimadzu GC 2014 with capillary column in my lab. The main objective is to use the GC to quantify concentration of hydrocarbon products of the thermal decomposition of jet fuels.

In the past, I have calibrated the GC with C1 to C4 species, using calibration gas standards (mainly purchased from Airgas or Air Liquide). My next step is to calibrate bigger hydrocarbon species (C5 and up). However, these species are rarely available in gas form. Thus, I need to calibrate using liquid standards.

My first attempt was to use a 1000-ppm mixture of benzene in methanol because I have a gas standard bottle for benzene. Here is how I made the solution:
The mole fraction of benzene is
x = 1000 moles of benzene/ 1,000,000 total moles

I converted moles to masses by using the corresponding molar mass
x = 78,110 g of benzene/(32,007,960 g of methanol + 78,110 g of benzene)

I reduced the ratio to
x = 0.122 g of benzene/(50 g of methanol + 0.122 g of benzene)

Then, I made the solution accordingly.

I used split injection method. However, the liquid sample gave an area that was 3 times higher than the gas sample did. And the residence time shifted earlier than the gas sample.

The gas standard bottle is of 1000 ppm benzene in balance of nitrogen.

Could anyone please help explain why I cannot obtain similar results for both gas and liquid samples?

Many thanks,

Khanh
How much (volume) liquid did you inject, and how much gas standard ? What was the split ratio and was it the same for both ? What column are you using ?

Peter
Peter Apps
Is the gas standard 1000 mg/1L N2 or is it 1000mg/1Kg N2 or 1000uMole/Mole N2?

If it is in Liters, then 1000ppm(mg/L) in N2 and 1000ppm(mg/L) in Methanol should give similar response if equal amounts are injected such as injecting 1ul of each because you would be injecting 1000ng/ul or 0.001mg/ul.

To me it is much easier to work with standards given in even amounts of weight per volume when working with chromatography. If you need molar amounts just do the conversion on the final numbers after the analysis is complete. Usually when something is listed as ppm concentration it involves mass and volume instead of moles.
The past is there to guide us into the future, not to dwell in.
Peter Apps wrote:
How much (volume) liquid did you inject, and how much gas standard ? What was the split ratio and was it the same for both ? What column are you using ?

Peter

1. Liquid injection:
I injected 1 micro-litter of the solution.

2. Gas injection:
I used a continuous gas flow. I connected the gas bottle to a Sierra flow controller, and set the flow rate to be 30 cubic centimeters per minute (SCCM). I should point out one thing that I set the inlet and outlet pressure at 0.5 atm for the gas injection to simulate the condition of my experiment.

The split ratio I used for both cases was: 10.

The column I am using is SH-Rtx-1 (crossbond 100% dimethyl polysiloxane), which has 60 m length, 0.32 mm ID, and 1 micro-meter df.
Perfect gas law may be working against you here, depending on the temperatures of things between your regulator/flow meter and column head...
Thanks,
DR
Image
khdang178 wrote:
Peter Apps wrote:
How much (volume) liquid did you inject, and how much gas standard ? What was the split ratio and was it the same for both ? What column are you using ?

Peter

1. Liquid injection:
I injected 1 micro-litter of the solution.

2. Gas injection:
I used a continuous gas flow. I connected the gas bottle to a Sierra flow controller, and set the flow rate to be 30 cubic centimeters per minute (SCCM). I should point out one thing that I set the inlet and outlet pressure at 0.5 atm for the gas injection to simulate the condition of my experiment.

The split ratio I used for both cases was: 10.

The column I am using is SH-Rtx-1 (crossbond 100% dimethyl polysiloxane), which has 60 m length, 0.32 mm ID, and 1 micro-meter df.


You cannot do chromatography with a continuous flow of sample into the column, you must have a valve and loop system or some kind of syringe to inject a sample of gas onto the column. What is the volume (and pressure) of gas that is injected onto the column ?

Peter
Peter Apps
Peter Apps wrote:
khdang178 wrote:
Peter Apps wrote:
How much (volume) liquid did you inject, and how much gas standard ? What was the split ratio and was it the same for both ? What column are you using ?

Peter

1. Liquid injection:
I injected 1 micro-litter of the solution.

2. Gas injection:
I used a continuous gas flow. I connected the gas bottle to a Sierra flow controller, and set the flow rate to be 30 cubic centimeters per minute (SCCM). I should point out one thing that I set the inlet and outlet pressure at 0.5 atm for the gas injection to simulate the condition of my experiment.

The split ratio I used for both cases was: 10.

The column I am using is SH-Rtx-1 (crossbond 100% dimethyl polysiloxane), which has 60 m length, 0.32 mm ID, and 1 micro-meter df.


You cannot do chromatography with a continuous flow of sample into the column, you must have a valve and loop system or some kind of syringe to inject a sample of gas onto the column. What is the volume (and pressure) of gas that is injected onto the column ?

Peter

Peter,

There is a valve within the GC, which takes in 1 mL of gas sample for analysis.

Sorry that I did not include the piece of information in the previous post.

Khanh
Now at last, with the nformation that I need, I can answer your question; I would not expect a 1ul liquid injection and a 1ml gas injection with the same concentration to ever give the same peak area, even without the non-linear effects of the splitter. Some simple arithmetic which I am not going to do on your behalf will tell you why.

Peter
Peter Apps
Peter Apps wrote:
Now at last, with the nformation that I need, I can answer your question; I would not expect a 1ul liquid injection and a 1ml gas injection with the same concentration to ever give the same peak area, even without the non-linear effects of the splitter. Some simple arithmetic which I am not going to do on your behalf will tell you why.

Peter


I did some calculation, and saw that the number of moles of benzene in the liquid solution is very different from that in the gas sample.

Thus, I was just wondering if there would be any possible method to correlate the two? Since I am really interested in calibrating the GC with n-dodecane (n-C12H26), which is extremely difficult to have in gas phase.

Best,

Khanh
You have already "correlated" then by calculating how much benzene is in each.

I think that you are making this much more difficult than it really is. Say you want to calculate the mass per volume of a liquid solution that will give the same peak area as 1 ml of a gas with a given mole/mole content. Calculate the number of moles of analyte in 1 ml of gas. If you want to inject 1ul it has to have the same number of moles in it as in the 1 ml of gas. Convert from moles to mass and you have the mass per volume concentration of the liquid you need to inject.

You still have the purely mechanical problem of liquid split injections not transferring sample to column quantitatively. You best chance is to do splitless injections.

Peter
Peter Apps
khdang178 wrote:
Peter Apps wrote:
Now at last, with the nformation that I need, I can answer your question; I would not expect a 1ul liquid injection and a 1ml gas injection with the same concentration to ever give the same peak area, even without the non-linear effects of the splitter. Some simple arithmetic which I am not going to do on your behalf will tell you why.

Peter


I did some calculation, and saw that the number of moles of benzene in the liquid solution is very different from that in the gas sample.

Thus, I was just wondering if there would be any possible method to correlate the two? Since I am really interested in calibrating the GC with n-dodecane (n-C12H26), which is extremely difficult to have in gas phase.

Best,

Khanh



Dear Khanh,

This is interesting because I am facing similar problem now but rather the other way around. I am analyzing benzene gases, but I have liquid standard, which I know the concentration in ug/ml. To be honest, I'm still not sure how am I going to make the standard curve with liquid benzene to quantify the gas benzene. Do share if you have any idea about it. And, can you show me how did you do the calculation? The concentration of same volume of liquid benzene and gases benzene would not be the same right?

Best,

Nik
11 posts Page 1 of 1

Who is online

In total there are 8 users online :: 1 registered, 0 hidden and 7 guests (based on users active over the past 5 minutes)
Most users ever online was 188 on Wed Nov 02, 2016 2:58 am

Users browsing this forum: Bing [Bot] and 7 guests