-
- Posts: 3
- Joined: Fri May 22, 2015 3:45 pm
Hello everybody.
I would get some help to solve a problem that is driving me crazy.
I am trying to establish a new methodology in my laboratory to analyze H2 rich biogas samples that we obtain from dark fermentation processes. I am using a 7820A GC with two six-port valves, and two capillary columns ( HP-PLOT Q and HP-Molesieve).
I have already created a method to analyze the biogas, but only qualitatively, I got this identifying the peak signals from CO2, CH4 and H2. The method is basically a current run with a temperature program, and right switching valve time to catch and separate both H2 and CH4 from others compounds in the biogas (CO2 and H2S), both H2 and CH4 keep for a period of time inside the HP-Molesieve capillary column, while CO2 and H2S are analyzed by TCD. After that, I turn the six-port valve on and let the analysis of both H2 and CH4 by TCD.
I am using three specialty gas calibration standards with different H2 concentrations (80%, 90% and 95%mol/mol) to create the calibration curve that relates peak Area vs %mol of H2. After the injection of the specialty gas calibration standards, I obtained a crazy relationship between H2 content (%mol) and Peak Area, I am getting a linear regression with a crazy slope and Y-intercept. To give you an idea, the linear regression splits the X-Axis around 76%mol H2. Therefore, quantifying smaller quantities of H2 is impossible (If I did that, I would need negative areas).
Initially, I thought that it behavior had some relationship with the high concentrated standard gas mixtures that I was using. Therefore, I got two more samples of H2 specialty gas calibration standards with 40% and 70%mol of H2 respectively, and repeated the analysis. Surprisingly, peak areas with these less concentrated standard gas mixtures were bigger than using high concentrated gas standard mixtures of H2.
I have heard that H2 peak signal is stronger than other compounds when using N2 as carrier gas, and that the gas matrix where H2 is diluted could generates really different peak signals (I read this in other forum), but I have never imagined that H2 quantification was so difficult.
I do not know what is happening here?, Could someone give me some advice about things that I could do or explain me why is so difficult to establish a right hydrogen calibration curve?.
I do not know what is the cause of my quantification problem trying to establish a relationship between Area signal and %mol of H2, sometimes, I think that it is related with my way of doing the injection (split ratio 1:1), or the configuration of my TCD detector (maybe changing TCD parameter could improve my situation), or if the problem is really the influence of the gas matrix where the H2 is diluted.
I would be very thankful for your attention, and help!.
Some of my chromatographic conditions are listed below:
Columns: HP-PLOT Q (30mX0.53mmX40μm) and HP-Molesieve (30mX0.53mmX50μm)
Injector temperature: 200⁰C.
Valve box temperature: 150⁰C.
Carrier gas: N2.
Carrier gas flow: 25 mL
Loop volume: 0.25 mL
Split ratio: 1:1
Oven temperature: 50 ⁰C for 10 min
TCD temperature: 250⁰C.
Reference gas flow: 48 mL/ml
Make up: 2 mL/min.
I would get some help to solve a problem that is driving me crazy.
I am trying to establish a new methodology in my laboratory to analyze H2 rich biogas samples that we obtain from dark fermentation processes. I am using a 7820A GC with two six-port valves, and two capillary columns ( HP-PLOT Q and HP-Molesieve).
I have already created a method to analyze the biogas, but only qualitatively, I got this identifying the peak signals from CO2, CH4 and H2. The method is basically a current run with a temperature program, and right switching valve time to catch and separate both H2 and CH4 from others compounds in the biogas (CO2 and H2S), both H2 and CH4 keep for a period of time inside the HP-Molesieve capillary column, while CO2 and H2S are analyzed by TCD. After that, I turn the six-port valve on and let the analysis of both H2 and CH4 by TCD.
I am using three specialty gas calibration standards with different H2 concentrations (80%, 90% and 95%mol/mol) to create the calibration curve that relates peak Area vs %mol of H2. After the injection of the specialty gas calibration standards, I obtained a crazy relationship between H2 content (%mol) and Peak Area, I am getting a linear regression with a crazy slope and Y-intercept. To give you an idea, the linear regression splits the X-Axis around 76%mol H2. Therefore, quantifying smaller quantities of H2 is impossible (If I did that, I would need negative areas).
Initially, I thought that it behavior had some relationship with the high concentrated standard gas mixtures that I was using. Therefore, I got two more samples of H2 specialty gas calibration standards with 40% and 70%mol of H2 respectively, and repeated the analysis. Surprisingly, peak areas with these less concentrated standard gas mixtures were bigger than using high concentrated gas standard mixtures of H2.
I have heard that H2 peak signal is stronger than other compounds when using N2 as carrier gas, and that the gas matrix where H2 is diluted could generates really different peak signals (I read this in other forum), but I have never imagined that H2 quantification was so difficult.
I do not know what is happening here?, Could someone give me some advice about things that I could do or explain me why is so difficult to establish a right hydrogen calibration curve?.
I do not know what is the cause of my quantification problem trying to establish a relationship between Area signal and %mol of H2, sometimes, I think that it is related with my way of doing the injection (split ratio 1:1), or the configuration of my TCD detector (maybe changing TCD parameter could improve my situation), or if the problem is really the influence of the gas matrix where the H2 is diluted.
I would be very thankful for your attention, and help!.
Some of my chromatographic conditions are listed below:
Columns: HP-PLOT Q (30mX0.53mmX40μm) and HP-Molesieve (30mX0.53mmX50μm)
Injector temperature: 200⁰C.
Valve box temperature: 150⁰C.
Carrier gas: N2.
Carrier gas flow: 25 mL
Loop volume: 0.25 mL
Split ratio: 1:1
Oven temperature: 50 ⁰C for 10 min
TCD temperature: 250⁰C.
Reference gas flow: 48 mL/ml
Make up: 2 mL/min.
