Detecting trace-level gas components in gas sample

Abstract

The present invention relates to a method for detecting an active gas component, such as a trace level contained in a gas sample, by using a gas chromatograph in fluid communication with a detector sensitive to the gas component. The following steps are performed in sequence: (i) feeding a carrier gas doped with an amount of the active gas component to the gas chromatograph for a doping time; (ii) feeding a carrier gas not doped with the active gas component to the gas chromatograph for a purge time; and (iii) adding the gas sample to the carrier gas of step (ii) and feeding the resulting mixture to the gas chromatograph.

Description

[0001] This invention relates to a method for detecting trace amounts of reactive gas components contained in a gaseous sample, wherein the method uses a gas chromatograph in fluid communication with a detector sensitive to the reactive gas component to detect the reactive gas component. The invention also relates to a gas chromatographic system suitable for this method.

[0002] Many analyses in gas chromatography suffer from system loss due to the reaction or adsorption of reactive gas components—the component of interest—to or onto active sites in the gas chromatography system. Adsorption or reactivity of the component of interest can lead to deterioration in detection limits, repeatability, linearity, and equimolarity, especially at lower concentration levels. The adsorption level of the component of interest will depend on the component itself and the activity of the various surfaces that come into contact with it during transport through the gas chromatography system. A common practice in gas chromatography is to treat these surfaces with coatings such as Sulfinert to minimize such reactive surfaces. However, even after these treatments, reactive surfaces may still remain. The presence of such residual reactive surfaces is particularly problematic when measuring low concentrations of reactive components, such as ammonia or hydrogen sulfide, in hydrogen or propylene gas samples.

[0003] Active surfaces can be passivated by repeatedly injecting samples or calibration standards containing a certain amount of the component of interest. The problem with this method is that it takes time before actual sample analysis can be performed. Furthermore, it is sometimes uncertain whether all active surfaces on the active surface have been passivated and for how long. This method is also unsuitable when the concentration of the component of interest to be measured is extremely low.

[0004] US2012 / 0131987 describes a chromatographic method in which the active component is first measured in a first carrier gas, and then the sample is measured again with a different carrier gas after rinsing the system with a different carrier gas.

[0005] US5612489 describes a method for passivating active surfaces in a gas chromatography system by doping a carrier gas with a passivating component. This passivating component can be a component of interest. The problem with passivation performed with a component of interest is that it increases baseline noise and also causes problems when performing sample injection via valve switching. When measuring extremely low concentrations of the component of interest in the sample, the peak of the component of interest is found to be unquantifiable. This method is also difficult to implement when performing sample injection using valve switching, i.e., so-called gas sampling valve injection (GSV-injection).

[0006] The purpose of this invention is to provide a method for detecting trace amounts of reactive gas components contained in gas samples, which avoids the disadvantages of existing methods.

[0007] This is achieved through the following method. A method for detecting trace amounts of reactive gas components contained in a gas sample, wherein the reactive gas component is detected by using a gas chromatograph in fluid communication with a detector sensitive to the reactive gas component, wherein the following steps are performed sequentially: (i) A carrier gas doped with a certain amount of active gas components is fed to a gas chromatograph for a certain doping time; (ii) Feeding a carrier gas, undoped of reactive gas components, to the gas chromatograph for a certain rinsing time; and (iii) Add the gas sample to the carrier gas in step (ii) and feed the resulting mixture to a gas chromatograph.

[0008] The applicant discovered that this method achieves an improved signal-to-noise ratio, enabling the measurement of low levels of reactive gas components in sample gases. Furthermore, the method exhibits good linearity. Without wishing to be constrained by the following theoretical limitations, the applicant believes that in step (i), a carrier gas doped with a certain amount of reactive gas components is first fed into the gas chromatograph, where the active sites in the system are chemically bonded. Additionally, a certain amount of unbonded reactive gas components will be present in the gas chromatograph system, of which the gas chromatograph is part. By performing step (ii), the unbonded reactive gas components are flushed from the gas chromatograph system, while the chemically bonded reactive gas components at the active sites are retained. By subsequently performing step (iii), the reactive gas components of the sample will not interact with the active surfaces due to the presence of chemically bonded reactive gas components at the active sites. In this specification, the sequence of steps (i) to (ii) is referred to as intermittent carrier doping. As explained, the components of interest present in the sample gas will reach the detector largely unaffected by adsorption losses, thus producing normal positive peaks. This peak will have the best possible signal-to-noise ratio (S / N ratio) because by performing step (ii) in time before step (iii) to remove excess dopants from the system, the baseline noise will not be increased.

[0009] Step (i) should be performed for a suitable period of time sufficient for the reactive gas component to chemically bond with all or almost all active sites in the gas chromatography system. This minimum time may depend on the reactive gas component and the gas chromatography system itself. For most applications using a gas chromatography system, step (i) is performed continuously, except when performing steps (ii) and (iii). This allows for the measurement of multiple samples in a sequential manner.

[0010] Step (ii) should be performed for a suitable period of time, sufficient to flush out unbonded reactive gas components from the gas chromatography system, while chemically bonded reactive gas components at the active sites remain. The carrier gas is the same as in step (i), except that it does not contain the reactive gas components. The optimal time for performing step (ii) may depend on the active component and the chromatographic system and can be readily determined. Preferably, the active component in step (iii) is eluted to the detector after the carrier gas in step (i) incorporating a certain amount of the reactive gas component. Since the component of interest is the same as the dopant, the active gas component in step (iii) will not be able to catch up with the dopant component in step (i) during elution by the gas chromatography system. This means that step (ii) is preferably performed for a very short time. If other components in the sample besides the active component that elute faster by the gas chromatography system will be detected, theoretically these other components or all components may catch up with the dopant component in step (i). In this case, it may be more preferable to perform step (ii) for a longer period of time. For most applications, especially those described in this application, the flushing time is preferably between 10 seconds and 300 seconds. Prolonged rinsing time may cause chemically bonded active gas components to leave the active sites, which is undesirable for the reasons previously described.

[0011] Step (iii) is suitable to be performed immediately after step (ii). This can be achieved by using a switching valve, which is standard practice in gas chromatography systems. In step (iii), the sample gas is added to the carrier gas. This is typically performed by injecting a small amount of sample gas into the carrier gas.

[0012] After performing step (iii), step (i) can be performed, making the gas chromatography system available for the next measurement. The timing of step (i) after step (iii) depends on the active component, the further composition of the gas sample, and the chromatographic system. Ideally, step (i) is performed when all peaks of interest are detected by the detector. For most applications, especially those described in this application, step (i) is preferably performed within 300 seconds after step (iii). By performing step (i), active sites from which the active component has been removed in previous steps (ii) and (iii) can be chemically bonded again to the active component.

[0013] Between steps (iii) and (i), a fresh carrier gas, undoped with the active component, can be introduced into the gas chromatograph. When an analysis is to be performed on a new sample, steps (i) through (iii) are performed. This avoids unnecessary consumption of the active component.

[0014] The reactive gas component can be any component in the sample gas that needs to be detected. Examples of reactive gas components include ammonia, hydrogen sulfide, hydrogen cyanide, carbon monoxide, carbon dioxide, oxygen, hydrogen, fluorine, chlorine, sulfur dioxide, formaldehyde, formic acid, hydrogen chloride, and water, as well as mixtures thereof.

[0015] The sample gas preferably consists of more than 50% by volume of hydrogen, carbon monoxide, carbon dioxide, methane, ethane, propane, butane, air, nitrogen, helium, ethylene, propylene, or mixtures thereof.

[0016] Preferably, the sample gas consists of more than 80% by volume hydrogen, more preferably more than 95% by volume, and even more preferably more than 99% by volume hydrogen, and the reactive gas component is any one or more of ammonia, hydrogen sulfide, formic acid, sulfur dioxide, formaldehyde, oxygen, carbon monoxide, carbon dioxide, hydrogen chloride, and water. More preferably, the sample gas consists only of hydrogen and trace impurities comprising one or more reactive gas components selected from the following: ammonia, hydrogen sulfide, formic acid, sulfur dioxide, formaldehyde, oxygen, carbon monoxide, carbon dioxide, hydrogen chloride, and water. “Trace” means that the total amount of these impurities does not exceed 100 ppm.

[0017] Preferably, the sample gas consists of more than 80% by volume of propylene, and the active gas component is any one or more of hydrogen sulfide, sulfur dioxide, and thiols. More preferably, the sample gas consists only of propylene and trace impurities comprising one or more active gas components selected from the following active gas components: hydrogen sulfide, sulfur dioxide, and thiols. “Trace” means that the total amount of these impurities does not exceed 100 ppm.

[0018] The trace amounts of active gas components in the gas sample are preferably less than 100 ppm, more preferably less than 10 ppm, even more preferably less than 5 ppm, and most preferably less than 1 ppm.

[0019] The carrier gas can be any inert gas. Preferably, the carrier gas can be a component consisting of hydrogen, helium, argon, and nitrogen.

[0020] The content of the active component in the carrier gas used in step (i) can be within the range of the active component content in the sample in step (iii). However, higher contents can also be used, such as up to 100 times the content of the active component in the gas sample. Due to the rinsing step (ii), excess unbonded active component in the gas chromatography system is removed before the actual measurement in step (iii). Incorporating the active component into the carrier gas can be performed by known methods, such as, for example, dynamic dilution, permeation, or calibration leak. Dynamic dilution doping is performed by continuously mixing the carrier gas with a gas stream containing pure active component or a mixture of active component and carrier gas. Permeation doping is performed by continuously diffusing the active component from a separate reservoir through a permeable material into the carrier gas. Calibration leak doping is performed by continuously injecting a certain amount of active component into the carrier gas through a controlled leak, typically consisting of an orifice or capillary. Alternatively, doping can be performed by using a mixture of carrier gas and active component dispensed from an industrial gas cylinder.

[0021] The carrier gas in step (i) may contain one or more reactive gas components. When the sample contains more than one analyte, the carrier gas in step (i) may contain more than one reactive gas component. Preferably, the carrier gas in step (i) contains one reactive gas component, and more preferably, it contains the most reactive reactive gas component. For example, when the sample contains a mixture of sulfur compounds as listed above, the reactive component in the carrier gas in step (i) is preferably hydrogen sulfide.

[0022] The pressure of the gas sample in steps (i) through (iii), and optionally the carrier gas pressure, can be ambient pressure or higher. Higher pressures of the gas sample and optionally the carrier gas, between 10 kPa and 500 kPa, may be beneficial for further improving the S / N ratio.

[0023] The method is particularly useful for applications that use gas sampling valve injection (GSV injection), but it may also be useful for GC applications that use liquid sampling valve (LSV) or automated liquid sampling (ALS) injection.

[0024] The above method is preferably performed in a gas chromatography system as described below. This invention also relates to such a system. The gas chromatography system includes a gas chromatography column having an upstream end and a downstream end, a detector fluidly connected to the downstream end of the gas chromatography column, and a first switching valve fluidly connected to a second switching valve, each switching valve having a different switching valve position. The first switching valve has an inlet for a carrier gas, an inlet for a dopant carrier gas, and an outlet for either the carrier gas or the dopant carrier gas, depending on the position of the first switching valve. The outlet for the carrier gas or the dopant carrier gas is fluidly connected to the inlet of the second switching valve. The second switching valve is further provided with an inlet for a gas sample, an outlet leading to an exhaust conduit, and a second switching valve outlet fluidly connected to the upstream end of the gas chromatography column. The first and second switching valves have switching valve positions (a) to (b), in which switching valve positions exist... (a) Fluid communication exists between the inlet for doping carrier gas at the outlet of the second switching valve and the upstream end of the gas chromatography column. (b) Fluid communication between the inlet for carrier gas via the outlet of the second switching valve and the upstream end of the gas chromatography column, and (c) Fluid communication exists between the inlet for carrier gas, the inlet for gas sample, and the upstream end of the gas chromatography column via the outlet of the second switching valve. In the switching valve position (a), step (i) of the method of the present invention can be performed. In the switching valve position (b), step (ii) of the method of the present invention can be performed. In the switching valve position (c), step (iii) of the method of the present invention can be performed.

[0025] When the method uses permeation doping, it is preferred that the first switching valve has an outlet for a carrier gas that is fluidly connected via a flow path containing a furnace to an inlet for the doping carrier gas, the furnace containing a permeation tube containing an active component, and wherein the inlet for the doping carrier gas is fluidly connected to the outlet for the doping carrier gas.

[0026] When using a sample pressure higher than ambient pressure, it is preferable to provide an exhaust conduit with a back pressure regulator. By setting a gas sample pressure higher than ambient pressure on the back pressure regulator, more sample is introduced into the system during injection.

[0027] The chromatographic column and detector of the gas chromatography system suitable for the process of this invention can be any known chromatographic column and detector.

[0028] The gas chromatography system consists of the following three parts. Figures 1 to 3 exemplified.

[0029] Figure 1A schematic diagram of a gas chromatography system (1) according to the invention, which allows gas sampling valve injection (GSV injection), is shown. The gas chromatography system is provided with a gas chromatography column (2) having an upstream end (3) and a downstream end (4); and a detector (5) fluidly connected to the downstream end (4) of the gas chromatography column (2). A first switching valve (6) fluidly connected to a second switching valve (7) is shown. Figures 1 to 3 The diagram shows switching valves (6) and (7) with different switching valve positions. The first switching valve (6) has an inlet (8) for carrier gas, an inlet (9) for doping carrier gas, and an outlet (10) for either carrier gas or doping carrier gas, depending on the position of the first switching valve (6). The inlet (9) for doping carrier gas is fluidly connected to a region (11) where the carrier gas contacts a permeation tube containing an active gas component. The carrier gas itself flows from a carrier gas source (12) to the region (11) via the first switching valve (6). The carrier gas source (12) can be two gas reservoirs that are used intermittently as both a carrier gas source and a doping carrier gas source.

[0030] The outlet (10) for carrier gas or for doping carrier gas is fluidly connected to the inlet (13) of the second switching valve (7). The second switching valve (7) is further provided with an inlet (14) for gas samples, an outlet (15) leading to an exhaust conduit (16), and a second switching valve outlet (17) fluidly connected to the upstream end (3) of the gas chromatography column (2). A sample loop (18) with an upstream end (19) and a downstream end (20) exists.

[0031] exist Figure 1 In this method, the first switching valve (6) and the second switching valve (7) have switching valve positions for performing step (i) of the method of the present invention. The switching valve positions allow fluid communication between the carrier gas source (12), the region (11), the inlet (9) for doping the carrier gas, and the upstream end (3) of the gas chromatography column (2) via the outlet (17) of the second switching valve. The upstream end (19) of the sample loop (18) is connected to the inlet (14) for the gas sample, and the downstream end (20) of the sample loop (18) is connected to the exhaust duct (16) provided with the back pressure regulator (22). Therefore, in step (i), the sample loop (18) is filled with the gas sample.

[0032] exist Figure 2 In this method, the first switching valve (6) and the second switching valve (7) have switching valve positions for performing step (ii) of the method of the present invention. The switching valve positions allow fluid communication to be formed between the carrier gas source (12), the inlet (8) for the carrier gas, and the upstream end (3) of the gas chromatography column (2) via the outlet (17) of the second switching valve. The doped carrier gas obtained during step (ii) is discharged via the exhaust port (23). The sample loop (18) is as follows: Figure 1The connection is made in such a way that at the end of step (ii), the sample loop (18) is filled with the gas sample.

[0033] exist Figure 3 In the method, the first switching valve (6) and the second switching valve (7) have switching valve positions for performing step (iii) of the method of the present invention. The switching valve positions enable fluid communication between the inlet (8) for the carrier gas, the inlet (13), the outlet (17) of the second valve, the sample loop (18), and the upstream end (3) of the gas chromatography column (2) via the outlet (17). The inlet (14) for the gas sample is connected to the exhaust conduit (16).

[0034] like Figure 1 As shown, the gas chromatography system (1) according to the present invention can also be repeated, wherein different detectors (5) can be used to analyze different active components in each parallel system. The sample can then be fed into the second switch valve (7) of one system and then into the second switch valve (7) of the next system. For example, in one system, trace amounts of ammonia in hydrogen can be measured using a nitrogen chemiluminescence detector and trace amounts of sulfur in hydrogen can be measured using a sulfur chemiluminescence detector.

[0035] The present invention will be illustrated by the following examples.

[0036] Comparative Experiment A Nitrogen samples containing 10 ppm ammonia and 1 ppm ammonia were analyzed by gas chromatography under undoped conditions to reduce active surface area. A Restek 60 m × 0.53 mm × 7 μm MXT-1 column was used, and a nitrogen chemiluminescence detector (NCD) was employed. Figure 4a shows the chromatogram of the 10 ppm sample, and Figure 4b shows the chromatogram of the 1 ppm sample, where "A" represents the ammonia peak. The chromatograms show that at the 10 ppm concentration level, the system activity already exhibits significant tailing on the ammonia peak. At the 1 ppm level, the ammonia peak completely disappears due to adsorption within the system.

[0037] Comparative Experiment B The same sample from Experiment A was analyzed using the same gas chromatography system, column, and detector, but with 10 ppm ammonia incorporated into the helium carrier gas (doped carrier gas). This carrier doping method is described in US Patent US5612489. Figure 5a shows the chromatogram of the 10 ppm sample, and Figure 5b shows the chromatogram of the 1 ppm sample, where "A" represents the ammonia peak.

[0038] The chromatograms show that excessive ammonia elution to the detector from the chromatographic system results in a high baseline. The 10 ppm ammonia chromatogram shows a normal peak shape, but the peak appears smaller than expected, which can be explained by the baseline drop as the injected sample plug reaches the detector. This can be explained by the fact that the sample plug used in the GSV injection does not contain ammonia background / dopants, meaning that this background deficiency needs to be compensated for before the ammonia present in the sample can generate a peak. This causes the peak to be smaller. Therefore, even at the 10 ppm level, doping will reduce the response to some extent. This is more evident in the analysis of a 1 ppm ammonia sample, where the chromatogram shows that the ammonia peak is no longer quantifiable due to the baseline drop as the injected sample plug reaches the detector.

[0039] Besides the problems associated with GSV injection, introducing doping also increases detector noise, leading to a decrease in the S / N ratio. This makes analyzing ammonia at this concentration level and below (e.g., 0.1 ppm) more problematic. Ideally, carrier doping would be sufficient to occupy all active sites in the system, resulting in minimal baseline increase. In this way, the baseline caused by GSV injection would be minimized while improving the S / N ratio. In practice, this is difficult to achieve and maintain.

[0040] Example 1 Experiment B was repeated, except that the doped carrier gas was replaced with an ammonia-free carrier gas before sample injection. After sample injection, the ammonia-free carrier gas was replaced with the doped carrier gas. Figure 6a shows the chromatogram of the 10 ppm sample, and Figure 6b shows the chromatogram of the 1 ppm sample, where "A" represents the ammonia peak. This example is... Figures 1 to 3 The gas chromatography system shown is used, and the pressure of the gas sample in the gas sample loop is the ambient pressure.

[0041] The chromatogram shows that the signal baseline drops just before the peak of interest (i.e., the ammonia peak) elutes toward the detector. This direct baseline drop is thought to be due to the removal of unbonded ammonia from the system (i.e., through elution). Since bonded ammonia elutes from the system more slowly, it is assumed that most active sites will still be bonded to ammonia from the dopant carrier gas as the sample plug passes through these sites. Therefore, this bonded ammonia prevents adsorption of ammonia injected into the sample plug. The intermittent carrier doping in this example allows ammonia in the gas sample to reach the detector and is largely unaffected by adsorption losses, resulting in a normal positive peak on the detector. This peak will have an optimal S / N ratio because excess dopant has been removed from the system in time and therefore does not increase baseline noise.

[0042] Example 2 Example 1 was repeated, with the gas sample pressure in the gas sample loop at 150 kPa. Figure 7a shows the chromatogram obtained from a 10 ppm sample, and Figure 7b shows the chromatogram obtained from a 1 ppm sample, where "A" represents the ammonia peak. A comparison of the chromatograms in Figures 6a and 6b with the corresponding chromatograms in Figures 7a and 7b shows that an improved S / N ratio was observed when the method was performed at higher pressures.

[0043] Example 3 For the nitrogen sample containing 3 ppm molar ammonia, Example 2 was repeated. Figure 8a shows the resulting chromatogram, where "A" represents the ammonia peak. Figure 8b shows the gas chromatogram of the same sample without doping. It can be seen that the S / N ratio of the ammonia peak in the chromatogram of Figure 8a is much higher than that in the chromatogram of Figure 8b.

[0044] Example 4 Example 1 was repeated under the condition of 100 ppb ammonia concentration in pure hydrogen. Step (i) of the method of the present invention was performed for 0.08 minutes, step (ii) for 2.99 minutes, and step (iii) for 2.90 minutes. The measurements were repeated 3 times (n=3). The results are presented in Table 1.

[0045] Example 5 Example 4 was repeated for a pure hydrogen sample containing 1.00 ppb H₂S, 1.00 ppb COS, 1.00 ppb methyl sulfide (MeSH), 0.94 ppb ethyl sulfide (EtSH), and 1.00 ppb dimethyl sulfide (DMS). The method of the present invention... step (i) Step (ii) was performed using a carrier with H2S as the active component for 0.01 minutes, (iii) for 2.99 minutes, and (iii) for 2.90 minutes. A Restek 60 m × 0.53 mm × 7 μm MXT-1 column was used, and a sulfur chemiluminescence detector (SCD) was employed. The results are presented in Table 1.

[0046] Table 1

[0047] In Table 1, RSD% is the "relative standard deviation" of the results. LOQ is the "limit of quantitation". The LOQ is calculated as follows: LOQ = (10 × signal / noise) × concentration. LOD is the "limit of detection". The LOD is calculated as follows: LOD = (3 × signal / noise) × concentration.

Claims

1. A method for detecting trace amounts of an active gaseous component contained in a gaseous sample, the method detecting the active gaseous component by using a gas chromatograph in fluid communication with a detector sensitive to the active component, wherein the following steps are performed sequentially: (i) A carrier gas doped with a certain amount of the active gas component is fed to the gas chromatograph for a certain doping time; (ii) The carrier gas, which is not doped with the active gas component, is fed to the gas chromatograph for a certain rinsing time; as well as (iii) The gas sample is added to the carrier gas of step (ii), and the resulting mixture is fed to the gas chromatograph.

2. The method according to claim 1, wherein the rinsing time is between 10 seconds and 300 seconds.

3. The method according to any one of claims 1 to 2, wherein step (i) is performed within 600 seconds after step (iii).

4. The method according to any one of claims 1 to 3, wherein the active gas component is ammonia, hydrogen sulfide, hydrogen cyanide, carbon monoxide, carbon dioxide, oxygen, hydrogen, fluorine, chlorine, sulfur dioxide, formaldehyde, formic acid, hydrogen chloride, and water, and mixtures thereof.

5. The method according to any one of claims 1 to 4, wherein the sample gas consists of more than 50% by volume of hydrogen, carbon monoxide, carbon dioxide, methane, ethane, propane, butane, air, nitrogen, helium, ethylene, propylene, or mixtures thereof.

6. The method of claim 5, wherein the sample gas comprises more than 80% by volume hydrogen or propylene, and the active gas component is selected from at least one or more of the following active gas components: ammonia, hydrogen sulfide, formic acid, sulfur dioxide, formaldehyde, oxygen, carbon monoxide, carbon dioxide, hydrogen chloride, and water.

7. The method of claim 6, wherein the sample gas consists of more than 99% by volume hydrogen.

8. The method according to any one of claims 1 to 7, wherein the trace amount of the gas component in the gas sample is less than 100 ppm.

9. The method according to any one of claims 1 to 8, wherein the pressure of the gas sample in step (iii) is between 10 kPa and 500 kPa.

10. The method according to any one of claims 1 to 9, wherein a gas sampling valve is used for injection.

11. A gas chromatography system, the gas chromatography system comprising a gas chromatography column having an upstream end and a downstream end, a detector fluidly connected to the downstream end of the gas chromatography column, and a first switching valve fluidly connected to a second switching valve, each switching valve having a different switching position. The first switching valve has an inlet for carrier gas, an inlet for doped carrier gas, and an outlet for either the carrier gas or the doped carrier gas, depending on the position of the first switching valve. The outlet fluid for the carrier gas or the doped carrier gas is connected to the inlet for the second switching valve. The second switching valve is further provided with an inlet for a gas sample, a sample loop, an outlet leading to an exhaust conduit, and a second switching valve outlet fluidly connected to the upstream end of the gas chromatography column. The first switching valve and the second switching valve have switching valve positions (i) to (iii), wherein in the switching valve positions thereexist (i) Fluid communication exists between the inlet for doping carrier gas at the outlet of the second switching valve and the upstream end of the gas chromatography column. (ii) fluid communication between the inlet for carrier gas at the outlet of the second switching valve and the upstream end of the gas chromatographic column, and (iii) Fluid communication between the inlet for carrier gas at the outlet of the second switching valve, the sample loop, and the upstream end of the gas chromatography column.

12. The gas chromatography system of claim 11, wherein the first switching valve has an outlet for a carrier gas, the outlet being fluidly connected via a flow path comprising a furnace to an inlet for doping a carrier gas, the furnace comprising a permeation tube containing an active component, and wherein the inlet for doping a carrier gas is fluidly connected to the outlet for doping a carrier gas.

13. The gas chromatography system according to any one of claims 11 to 12, wherein the exhaust duct is provided with a back pressure regulator.

Images