Chromatography Forum

Hello, I'm a student using a GC for my experimental studies. I am not an expert and my procedure that I am following is based on what I understood from reading literature, lab experiment guides and brochures I was able to find. I would appreciate it, greatly, if I could get some feedback and help. Thank you very much.

The GC available to me is an Agilent GC 6890N equipped with FID and capillar column HP-5 30m x 0.320mm x 0.25 micron. The software is Chemstation Rev.A.10.02 (1757). I am not very knowledgeable about Chemstation and the help option is not very helpful to me because it does not explain what is the purpose of each option. Is there anything that explains it?

My experimental study involves Benzene, Toluene and Xylene at ppm level (mg/L) in Methanol as a solvent. Currently my solution is not a mixture of Benzene, Toluene and Xylene in methanol BUT only individual component mixed in methanol (I will be making mixture of the 3 components in later stages of my study). I am trying to quantify how much Benzene, Toluene and Xylene are present in my solutions after adding activated carbon to them. The solutions are supposed to be tested at different times (for example 12hrs, 24hrs, etc).

I am using Ethyl Acetate (EA) in methanol as my internal standard at 500ppm (mg/L). I also prepared solutions of benzene (B) in methanol at 25ppm, 50ppm-400ppm (at 50ppm increments) to get a calibration curve. To get my calibration curve, I take 0.5mL of my EA solution and 0.5mL of my B solution and transfer to an autosampler vial to be injected in the GC. I then record the area and height of the EA and B peaks, and calculate the ratio of Area B/Area EA and Height B/Height EA and plot each against the concentration of B.

I was able to do only some injections and from 0.2micro liter injections I got information but the peak areas are small and the plot R squared value I got was 0.9903 based on area while it was 0.9963 based on height. I don't know how low the concentration of my components will go after adding activated carbon when I start my experiment, so I want to make the peak area or height as large as possible to get good results. How can I improve my peaks?

Benzene (ppm).... Area Benzene......Area Ethyl Acetate......Area B/Area EA.....Height Benzene....Height Ethyl Actetate.....Height B/Height EA
400.....................39.66035.............19.17882..................2.067924408.........36.65300...............19.66506................1.863864133
300......................37.32846..............26.9337..................1.385938805.........36.49706...............27.77709................1.313926693
200......................17.53905.............16.51722..................1.061864527.........15.39798..............16.39149.................0.939388671
100......................9.09984..............19.85968..................0.458206779.........7.77229.................19.36886.................0.401277618
25........................3.01499..............23.78933..................0.126737071.........2.76540.................24.41411.................0.113270564

The following are the GC settings:
Inlet
Heater: 250C
Pressure: 9.91psi
Total flow: 51.5mL/min
Split ratio: 10.0:1
Split Flow: 44.4mL/min

He flow:
P: 9.91psi
Flow: 2.2mL/min
Avg Velocity: 35cm/sec

Oven
Initial at 40C hold for 2min, Ramps up to 100C at 30C/min holding for 1 min. Post run is 100C for 0.5min (Total run time is 5.50min).

Detector
Heater:250C
H2: 40mL/min
Air:450mL/min
Makeup flow He: 45mL/min

I am sorry for the long post. I will be very thankful for any feedback.
Thank you.

For quantification you need to be working on peak area. For sensitivy though it's all about peak height. It isn't used in the calculation of amounts but generally the limit of detection is a peak three times the baseline noise amplitude (i.e. signal:noise ratio of 3), and 10x for limit of quantitation. Chemstation can caclulate this for you using the Performance reports (you need to define noise ranges in Data Analysis under the Report menu). Narrower peaks are taller peaks, though on a 5.5min run you would expect your peaks to be pretty sharp? To make them narrower, can you speed up the chromatography? Either warmer oven temps and/or higher carrier flow. Flow is the dominant factor at low retention times, temperature at longer times (broadly). Your 35cm/s is pretty good, if you have good resolution (separation) then you could increase this out of the optimum (20-40cm/s) and gain speed at the expense of resolution you don't need. Also are you running in constant flow mode? You should, if not then the linear speed reduces with oven temperature so loss of reesolution, peak broadening, and longer run times, all at once.

The two very simple things to try before you do anything else are to increase the injection volume to 1 ul (which will also improve the repeatability and linearity), and if the peaks are still not big enough, to operate the inlet in splitless mode. Doing both of these together will increase your peak areas by about 50 times.

Methanol is a tricky solvent for flash vaporization (in a hot inlet such as you are using) you might need to optimize injection conditions. We can look at that later.

What type of liner do you have in the GC inlet ?

Peter

Thank you both for the reply. I will try my best to answer.

The narrowest peak and with the largest height was for methanol since it was the solvent, it appears at around 1.620min. The ethyl acetate peak appears at around 2.247min and is smaller than the methanol peak but is very clearly apparent on the chromatogram. The issue I have is with the analytes, benzene is very small and appears at around 2.560min (I still haven't tried toluene and xylene, yet). Initially, I was not able to get any area for benzene since it was too small but I was just trying around on chemstation until I was lucky to get a peak area (I don't know what I did or what I clicked on, and I am actually concerned that the area I got is not correct but the area chemstation calculated is at the same retention time for benzene, so it should be ok right? ) I will be very honest here and tell you that most of the options I read on chemstation don't make a lot of sense to me.

I would not say that the peaks are narrow, except for methanol. I have to zoom in a lot to see the benzene peak which looks to short and pretty broad to me.

Johnny Rod: The option for the column is constant pressure so I believe it is constant flow mode.
What is the significance of the height, limit of detection and limit of quantitation, and also what do you mean by resolution. Can you please explain further?

Peter Apps: Thank you for your suggestions. I will try 1micro L injections and splitless mode. Can you please elaborate what is the difference between split and splitless mode and what is the significance? Unfortunately I don't know what liner is present, how can I check this?

Thank you.

In split mode a large proportion of the carrier gas that flows into the inlet is vented, and only a small portion goes to the column. Since the sample is mixed with the carrier gas, this means that only a small portion of the sample goes to the column and the detector, and the peak is correspondingly small. In splitless injection "all" the sample laden carrier gas goes to the column and the detector, and the peaks are larger. If you select splitless mode for the GC method you need to set a splitless time of 30s. The solvent peak will get much wider, so you might have to use a slower column temperature program to sepatta ethe benzene form the solvent.

To find out what inlet liner you have, you need to remove it from the inlet - is there anyone at your lab who can show you how to do this ?

Peter

I'm afraid not, the lab techs are on leave currently. I wouldn't want to mess around with the equipment.

So just to get things clear, if I use splitless mode I will have to change my ramp to a much slower ramp? What about the inlet temp, heater temp?

30 C per minute is already an extremely fast ramp rate, and way above the optimum for the column length that you have. Iin order to get the system working and the settings where you need them to be I suggest that you reduce it to 10 C/min (which is still quite fast).

Before you try splitless injections, try just increasing the injection volume - this will increase your peaks by 5 times. Only if this is not enough do you need to go to splitless injections.

If you do want to use splitless, than the only thing that you change (besides the column ramp rate) is the split / splitless setting on the inlet, and set a splitless time (which might be called purge on or off in Chemstation, it's a long time since I used it) of 30 s.

Don't do anyting to the inlet until there is someone around who can help you. Check with them about septum changes as well.

Peter

Hello, I'm a student using a GC for my experimental studies. I am not an expert and my procedure that I am following is based on what I understood from reading literature, lab experiment guides and brochures I was able to find. I would appreciate it, greatly, if I could get some feedback and help. Thank you very much.
.....
.

Perhaps I can help you in the following way. Would you like send your files with measured chromatograms (usually these files are the following names: FID1A.ch or FID2B.ch) on e-mail unichrom@unichrom.com

Regards,
Siarhei

Peter Apps: I will be in the lab tomorrow morning. I will try 1micro liter injections and this time I will inject the previous solutions again and the other solutions I prepared earlier (25-400ppm, in 50ppm increments). I will also inject each solution several times and take the average of the peak areas(based on an article I remember reading which did 5 injections).

I will continue with the current GC temperature and flow settings before altering them, if I did not get any progress with higher injection volume.

I have been thinking about my internal standard, ethyl acetate which is 500ppm. I have been adding the same quantity of 0.5mL of Ethyl acetate such that it is a fixed quantity to 0.5mL of my benzene solutions. But the area of Ethyl acetate is not the same in each injection which I did earlier. Is this normal?

CharapitsaS: Thank you very much!! I will be back tomorrow in the lab and will be able to copy the files and send them to you once I get back home. There is not internet access in the lab, unfortunately.

Peter Apps: I will be in the lab tomorrow morning. I will try 1micro liter injections and this time I will inject the previous solutions again and the other solutions I prepared earlier (25-400ppm, in 50ppm increments). I will also inject each solution several times and take the average of the peak areas(based on an article I remember reading which did 5 injections).

I will continue with the current GC temperature and flow settings before altering them, if I did not get any progress with higher injection volume.

I have been thinking about my internal standard, ethyl acetate which is 500ppm. I have been adding the same quantity of 0.5mL of Ethyl acetate such that it is a fixed quantity to 0.5mL of my benzene solutions. But the area of Ethyl acetate is not the same in each injection which I did earlier. Is this normal? It is normal, but not good. In other words it is a common problem.

.

Is there anyone at the lab who knows how to run the GC properly and can advise and train you on how to set up the instrument ? An hour of competent instruction at the bench will provide much more useful knowledge than days of internet posting.

The FID detector that you are using should give you a peak that can be reliably integrated with high signal:noise from 1 ng of benzene - calculate the quantity of benzene in the solution that you inject, and divide it by the split ratio to get how much is going to the column. If the lowest concentration of your calibration series gives less than 0.1 ng on column you are going to struggle no matter what you do with the instrument.

It is helpful to us, and will make your calculations easier if you express concentrations as proper SI units (i.e. mass/volume, volume/volume or mass/mass rather than "ppm", which can have several different meanings.

Peter

Is there anyone at the lab who knows how to run the GC properly and can advise and train you on how to set up the instrument ? An hour of competent instruction at the bench will provide much more useful knowledge than days of internet posting.

Peter is right about this. At the very least, all GC operators should know how to safely replace a liner and septum. Someone should show you how to do this. If you feel courageous and bold there should be some Agilent Maintenance videos on that computer from the installation for changing the liner and septum. It's simple, but watching someone do it really helps. The Agilent videos are usually pretty good (although the quality is sometimes poor as most were made in the 90s. I think you could search YouTube for some of them too.

I'm back. I couldn't go to the lab on Sunday, but I managed today.

Thank you for your concern. I agree with you both 100% but the lab techs are on leave right now and I can't find anyone else who knows how to use a GC. I don't want to sound rude, but I sat several times for hours with the lab techs trying to get peak areas but they did not know how. Although I am very thankful to them for teaching me how to run the GC and also for permission to use it. They were the only ones willing to spend their time trying to figure it out.

Someone should show you how to do this. If you feel courageous and bold there should be some Agilent Maintenance videos on that computer from the installation for changing the liner and septum. It's simple, but watching someone do it really helps. The Agilent videos are usually pretty good (although the quality is sometimes poor as most were made in the 90s. I think you could search YouTube for some of them too.

I wouldn't mind changing things myself and I think this will provide me with a better learning of GC's but I can't because it is not my machine and I was just given permission to use the GC for injections.

I have been thinking about my internal standard, ethyl acetate which is 500ppm. I have been adding the same quantity of 0.5mL of Ethyl acetate such that it is a fixed quantity to 0.5mL of my benzene solutions. But the area of Ethyl acetate is not the same in each injection which I did earlier. Is this normal? It is normal, but not good. In other words it is a common problem.

So how can I solve this issue of having different peak area for my internal standard?

The FID detector that you are using should give you a peak that can be reliably integrated with high signal:noise from 1 ng of benzene - calculate the quantity of benzene in the solution that you inject, and divide it by the split ratio to get how much is going to the column. If the lowest concentration of your calibration series gives less than 0.1 ng on column you are going to struggle no matter what you do with the instrument.
It is helpful to us, and will make your calculations easier if you express concentrations as proper SI units (i.e. mass/volume, volume/volume or mass/mass rather than "ppm", which can have several different meanings.

I apologize for the confusion, for ppm I am referring to mg of benzene per liter of methanol. I don't understand about the peak integration and noise, can you please explain?


I am trying my best to get my experiment going with good results. Today something strange has happened and it is very frustrating after having what looked like good results when I first posted. I did 3 injections of 1uL for each of my benzene solutions 25ppm-400ppm (in 50 ppm increments) (ppm = mg/L)

The parameters in the method which I used earlier changed, I am not sure how but I think it's my mistake. I was lucky that I caught the change before doing many injections and having to repeat them. But some of the parameters I could not return back to the original settings.

The settings which I could not change back was total flow is 26.7mL/min and split flow 22.1mL/min, they were before total flow 51.5mL/min and split flow 44.4mL/min. All the other parameters were the same as the beginning. I am not sure if this is the reason but my peak areas today were much smaller than initially! And now after I compiled the peak areas on excel, I am seeing that my area for ethyl acetate is only between 0.5-3.2 and for benzene it is between 0.25-4.6. In the first injections, the peak areas was much higher. I don't know what went wrong!! The R squared value for my plot of ratio of peak area benzene/ethyl acetate versus concentration Benzene is 0.98.

Edit: I have been trying to find out where I made a mistake to get such results and I have a question. Is there a difference between the results in the report and the ones output when using the auto-integrate option in data analysis?

Perhaps I can help you in the following way. Would you like send your files with measured chromatograms (usually these files are the following names: FID1A.ch or FID2B.ch) on e-mail unichrom@unichrom.com

Regards,
Siarhei

CharapitsaS I copied all the output files and there are several. I don't want to flood your email with all of them, which one is the best for me to send you?

I'm back. I couldn't go to the lab on Sunday, but I managed today.

Thank you for your concern. I agree with you both 100% but the lab techs are on leave right now and I can't find anyone else who knows how to use a GC. I don't want to sound rude, but I sat several times for hours with the lab techs trying to get peak areas but they did not know how I don't want to be rude either, but if the lab techs who know how to run the machine could not get decent peak areas, and they recommended programme rates of 30C/min and 0.2 ul injections, and didn't try splitless, then I begin to suspect that there might be some issues of instrument maintenance and set up that need to be addressed - for example when was the last time that the septum and liner were changed and is the column installed properly in the inlet and detector ? Is there a maintenance log for the instrument that you can check ?. Although I am very thankful to them for teaching me how to run the GC and also for permission to use it. They were the only ones willing to spend their time trying to figure it out.

Someone should show you how to do this. If you feel courageous and bold there should be some Agilent Maintenance videos on that computer from the installation for changing the liner and septum. It's simple, but watching someone do it really helps. The Agilent videos are usually pretty good (although the quality is sometimes poor as most were made in the 90s. I think you could search YouTube for some of them too.

I wouldn't mind changing things myself and I think this will provide me with a better learning of GC's but I can't because it is not my machine and I was just given permission to use the GC for injections.

I have been thinking about my internal standard, ethyl acetate which is 500ppm. I have been adding the same quantity of 0.5mL of Ethyl acetate such that it is a fixed quantity to 0.5mL of my benzene solutions. But the area of Ethyl acetate is not the same in each injection which I did earlier. Is this normal? It is normal, but not good. In other words it is a common problem.

So how can I solve this issue of having different peak area for my internal standard? Making the leap of faith that the instrument is properly set up, then poor repeatability with methanol as a solvent can usually be improved by changing inlet and injection conditions

The FID detector that you are using should give you a peak that can be reliably integrated with high signal:noise from 1 ng of benzene - calculate the quantity of benzene in the solution that you inject, and divide it by the split ratio to get how much is going to the column. If the lowest concentration of your calibration series gives less than 0.1 ng on column you are going to struggle no matter what you do with the instrument.
It is helpful to us, and will make your calculations easier if you express concentrations as proper SI units (i.e. mass/volume, volume/volume or mass/mass rather than "ppm", which can have several different meanings.

I apologize for the confusion, for ppm I am referring to mg of benzene per liter of methanol. I don't understand about the peak integration and noise, can you please explain?"Integration" is the term used for measuring the area of a peak (which should be related to quantity of analyte). Noise is the random fluctuations in the baseline signal. The higher the peak is in relation to the noise (the signal:noise ratio) the more repeatable is the integration. Signal:noise ratios nned to be above ten for quantitative work.


I am trying my best to get my experiment going with good results. Today something strange has happened and it is very frustrating after having what looked like good results when I first posted. I did 3 injections of 1uL for each of my benzene solutions 25ppm-400ppm (in 50 ppm increments) (ppm = mg/L)

The parameters in the method which I used earlier changed, I am not sure how but I think it's my mistake. I was lucky that I caught the change before doing many injections and having to repeat them. But some of the parameters I could not return back to the original settings.

The settings which I could not change back was total flow is 26.7mL/min and split flow 22.1mL/min, they were before total flow 51.5mL/min and split flow 44.4mL/min. All the other parameters were the same as the beginning. I am not sure if this is the reason but my peak areas today were much smaller than initially!By decreasing the total flow you have decreased the split ratio, which should have made the peaks bigger, not smaller, so something else must have changed as well. What were the injection volumes ? And now after I compiled the peak areas on excel, I am seeing that my area for ethyl acetate is only between 0.5-3.2 and for benzene it is between 0.25-4.6. In the first injections, the peak areas was much higher. I don't know what went wrong!!Don't worry about it. Go back to the Chemstation method editor and just put all the setting to what they should be, and then see what the peak sizes look like with some new injections The R squared value for my plot of ratio of peak area benzene/ethyl acetate versus concentration Benzene is 0.98 This is actually pretty good, all things considered. How good do you need it to be to run your charcoal experiments ? Are you expecting the aromatics concentrations to increase or decrease when you add charcoal ? - although it will act as an adsorbent I would not be at all surprised to find that it has high levels of extractable aromatics.

Edit: I have been trying to find out where I made a mistake to get such results and I have a question. Is there a difference between the results in the report and the ones output when using the auto-integrate option in data analysis?The area that is reported for a peak will depend on the integration parameters that the software uses, especially if the signl:noise is too low, so it is possible that you have two sets of different areas for the same peaks

Perhaps I can help you in the following way. Would you like send your files with measured chromatograms (usually these files are the following names: FID1A.ch or FID2B.ch) on e-mail unichrom@unichrom.com

Regards,
Siarhei

CharapitsaS I copied all the output files and there are several. I don't want to flood your email with all of them, which one is the best for me to send you?

The harsh reality is, manufacturer's claims and TV shows about forensics notwithstanding, that you cannot just walk up to a GC and start injecting samples into it unless it has already been set up and validated for the analysis that you want to do. Kudos to you for being willing to try, but as you are finding out the hard way that there is a certain level of technical knowledge that you need. What you have achieved already may be good enough for what you need to do. What levels of accuracy and precision do you need for the charcoal experiments ?

Peter

I'm back. I couldn't go to the lab on Sunday, but I managed today.

Perhaps I can help you in the following way. Would you like send your files with measured chromatograms (usually these files are the following names: FID1A.ch or FID2B.ch) on e-mail unichrom@unichrom.com

Regards,
Siarhei

CharapitsaS I copied all the output files and there are several. I don't want to flood your email with all of them, which one is the best for me to send you?

I am waiting for files with extension *.ch
It will be enough.

Siarhei

Hello again. I would like to thank you all for taking the time to read my posts and respond with very useful information. Thank you.

I tried to re-analyze my injections I did yesterday which was for 1uL injections and re-integrate using auto-integrate option. But the peak areas were still low and there was very small differences.

I returned back to 0.2uL injections (3 injections of each solution) using the same settings (except for total flow and split flow which I could not return back to the original settings) and the peak areas were much higher and are almost similar to the ones I posted in the beginning. The plot fitting gave a better R squared value 0.9931.
Well, I think there is some progress in this case. It could be that the GC can't handle high injection volume with the settings used.

I noticed that there has been a shift in the retention time for Benzene. Initially it used to be around 2.560 minutes and now it is around 2.570 minutes. Is this normal because of the difference in the total flow and split flow?

I am waiting for files with extension *.ch
It will be enough.
Siarhei

CharapitsaS I sent you the .ch files, please have a look and let me know about your comments.

I don't want to be rude either, but if the lab techs who know how to run the machine could not get decent peak areas, and they recommended programme rates of 30C/min and 0.2 ul injections, and didn't try splitless, then I begin to suspect that there might be some issues of instrument maintenance and set up that need to be addressed - for example when was the last time that the septum and liner were changed and is the column installed properly in the inlet and detector ? Is there a maintenance log for the instrument that you can check ?.

Peter I think that there is definitely an issue with the GC. To my knowledge I don't think it is maintained consistently and I could not find a log book for the GC in the lab. The GC is used for the main purpose of analyzing student samples they prepare for their organic chemistry lab course. The lab techs have saved methods for each lab the students work on during the semester. They don't do any other analysis besides just checking if they got the correct final product expected.

The FID detector that you are using should give you a peak that can be reliably integrated with high signal:noise from 1 ng of benzene - calculate the quantity of benzene in the solution that you inject, and divide it by the split ratio to get how much is going to the column. If the lowest concentration of your calibration series gives less than 0.1 ng on column you are going to struggle no matter what you do with the instrument.
It is helpful to us, and will make your calculations easier if you express concentrations as proper SI units (i.e. mass/volume, volume/volume or mass/mass rather than "ppm", which can have several different meanings.

Shouldn't the amount of benzene in the injections be the same as in my solution? So if my solution is 50ppm = 50mg/L then I have 50mg in my injection, correct? So it is much higher than 0.1ng.

I apologize for the confusion, for ppm I am referring to mg of benzene per liter of methanol. I don't understand about the peak integration and noise, can you please explain?"Integration" is the term used for measuring the area of a peak (which should be related to quantity of analyte). Noise is the random fluctuations in the baseline signal. The higher the peak is in relation to the noise (the signal:noise ratio) the more repeatable is the integration. Signal:noise ratios nned to be above ten for quantitative work.

The chromatogram peak settles at 3.2 according to the screen on the GC but the peak areas of ethyl acetate and benzene vary from one injection to the next. Is there a specific option on chemstation to get the signal:noise ratio?

I did a quick divide of my peak areas for ethyl acetate and benzene with 3.2 on excel and it's not above 10 for all the injections except 3. Does this remove the validity of using GC for my experiment?!

By decreasing the total flow you have decreased the split ratio, which should have made the peaks bigger, not smaller, so something else must have changed as well. What were the injection volumes ?

The total flow decreased but I was able to specify the same split ratio 10.0:1. I couldn't and still can't figure out how to return back to the original settings for total flow and split flow. I checked several other options and they were not changed.
Just so there is no confusion the earlier settings had a total flow of 51.5mL/min, split flow 44.4mL/min and split ratio of 10.0:1 and after the settings were changed the total flow is now 26.7mL/min, split flow 22.1mL/min and split ratio is the same at 10.0:1.

The injection volumes I did yesterday was 1uL which gave the smaller areas but today I did 0.2uL injections again and got higher peak areas.

And now after I compiled the peak areas on excel, I am seeing that my area for ethyl acetate is only between 0.5-3.2 and for benzene it is between 0.25-4.6. In the first injections, the peak areas was much higher. I don't know what went wrong!!Don't worry about it. Go back to the Chemstation method editor and just put all the setting to what they should be, and then see what the peak sizes look like with some new injections The R squared value for my plot of ratio of peak area benzene/ethyl acetate versus concentration Benzene is 0.98 This is actually pretty good, all things considered. How good do you need it to be to run your charcoal experiments ? Are you expecting the aromatics concentrations to increase or decrease when you add charcoal ? - although it will act as an adsorbent I would not be at all surprised to find that it has high levels of extractable aromatics.

I want to have good accuracy and precision for my experiments. Actually the main purpose of using activated carbon is to test how much of benzene, xylene and toluene it can adsorb. The main purpose of my experiment is to find the most suitable activated carbon for the removal of BTX. I expect the concentrations to decrease but how low I don't know yet which is making me nervous.

Edit: I have been trying to find out where I made a mistake to get such results and I have a question. Is there a difference between the results in the report and the ones output when using the auto-integrate option in data analysis?The area that is reported for a peak will depend on the integration parameters that the software uses, especially if the signl:noise is too low, so it is possible that you have two sets of different areas for the same peaks

The results output and the ones generated from auto-integrate option have very slight differences. But I will, for now, continue using the auto-integrate option to report my peak areas.

The harsh reality is, manufacturer's claims and TV shows about forensics notwithstanding, that you cannot just walk up to a GC and start injecting samples into it unless it has already been set up and validated for the analysis that you want to do. Kudos to you for being willing to try, but as you are finding out the hard way that there is a certain level of technical knowledge that you need. What you have achieved already may be good enough for what you need to do. What levels of accuracy and precision do you need for the charcoal experiments ?

Of course not, I never expected direct results but to be frank it is taking some time and effort. Seeing as no one is well experienced in operating GCs here to help guide me I think I will have to accept my slow progress.

I want to have good accuracy and precision for my experiments. I am still in the beginning but would like to present my results with confidence. I was able to obtain 3 different types of carbon from manufacturers and I want to test which of them is the best for adsorbing benzene, toluene and xylene. So it is very important that my GC method will be able to detect the analytes after adsorption.

So do you think I can start my experiment, at least for benzene? My plan is to run solutions of 400ppm and 200ppm of benzene with 3 different types of activated carbon for 12, 24, 48hrs and test the amount of benzene adsorbed using the GC.

A practical question: What would be the best way to filter out the carbon from solution? Is there a better way than using filter paper and centrifuging?
I mixed some carbon with water in a test tube today, just to check, but after filtering it out using a filter paper and then centrifuging not all of the carbon settled in the bottom of the test tube, some got stuck to the sides.

Also, I noticed that the pressure on the helium cylinder attached to the GC is dropping. It was at around 1600psi and it's dropped to 1400psi. Am I consuming too much Helium? The other cylinders of H2 and Compressed Air did not drop pressure. I have been running the GC for roughly 5-6 hours for the past few times.