English

• The global greenhouse effect has led to the recognition of hydrogen as an excellent alternative energy source in the 21^st century, especially for vehicles [1-3]. In a hydrogen vehicle, the fuel cell system requires high purity and quality hydrogen, as even small amounts of contaminants can have irreversible effects on the performance and lifespan of the fuel cells [4-9]. To mitigate the detrimental effects of pollutants, several international standards have been established including ISO 14687:2019 [10], EN 17124:2018 [11], and SAE J2719:2020 [12], which sets the amount fractions of 14 contaminants and their respective maximum thresholds. Sensitive analytical methods have been developed for measuring ISO 14687:2019 impurities in fuel cell graded hydrogen. In recent years, Heleen Meuzelaar used laser-based spectroscopic methods for HCHO, HCOOH, HCl, and HF analysis in hydrogen fuel to promote traceable hydrogen purity analysis. These methods have sufficient high sensitivity and low uncertainty [13]. A new system by combining online cryogenic trapping of trace sulfur compounds and subsequent GC-SCD was developed by Pan's group for sulfur compound detection in hydrogen. This method obtained a much lower LOD than the minimum concentration level of total SCs by ISO 14687:2019, which shows great promise for practical applications [14]. Furthermore, Sang Woo Kim's group developed a hydrogen impurity analyzer (HIA) based on mobile gas chromatography with a thermal conductivity detector for the quality assurance of hydrogen fuel, where O₂, Ar, N_2, and CH₄ were selected as target impurities [15]. Cavity-enhanced Raman Spectroscopy technology was used for the detection of gaseous impurities in hydrogen for fuel cells. At a total pressure of 0.1 and 2.5 MPa, the LOD can reach sub-ppm- and ppb-levels. This method has excellent application prospects in gaseous impurity analysis in hydrogen [16]. In addition to the above research, other analytical methods have also been introduced [17-23]. From the reported research methods, it can be seen that currently, most analysis methods focus on the analysis of inorganic components and sulfides, with relatively little research on other active impurity components in hydrogen.

  As one of the key impurities, NH₃ can readily react with the proton in the proton exchange membrane to generate NH₄⁺. This reaction reduces the proton transfer capacity of the electrolyte membrane, leading to a significant degradation in fuel cell performance [24-27]. NH₃ has a high reaction active and low threshold (100 nmol/mol), which immaterially improves the difficulty of accurate quantitative analysis. To date, a gas chromatography-mass spectrometry (GC–MS) method was employed to analyze the concentration of NH₃ in nitrogen. The method utilized single ion monitoring (m/z = 17) with a LOD of 0.5 µmol/mol. It is important to note that this LOD is higher than the defined threshold for NH₃ [28]. ASTM D7550:2009 describes an ion chromatography method for measuring NH₃ in hydrogen with sensitivities in the low nmol/mol. This method must involve dissolving NH₃ in a solution and analyzing it using ion chromatography. However, it should be noted that this process is complex and can be influenced by environmental conditions [29]. One ASTM D7941/D7941M:2023 method using cavity ring-down spectroscopy (CRDS) was proposed for the determination of contaminants in fuel-cell-grade hydrogen, which proposes that the LOD of NH₃ is 0.86 nmol/mol [30]. A Fourier transform infrared (FTIR) spectroscopy was developed in ASTM D7653:2018 for measuring several impurities in hydrogen. It provides an LOD of 20 nmol/mol for NH₃ analysis [31]. Although the LOD of CRDS and FTIR methods meet the requirements, they have not yet been validated. The GB/T 37244:2018 [32] recommends the use of the ion-selective electrode method [33], which has some disadvantages including the need to prepare multiple standard solutions, susceptibility to interference from manual operation, and limited applicability to air and industrial waste gas. To sum up, there is still a lack of effective methods for monitoring NH₃ in hydrogen.

  FTIR, as a mature precise spectroscopy method, has remarkable advantages of simple operation, fast analysis speed, high accuracy, low detection limit, etc. It was widely used in the qualitative and quantitative analysis of various micro and trace gases [34-37]. Here a novel method based on an online dilution module (ODM-FTIR) was developed for qualitative and quantitative analysis of NH₃ in hydrogen. In contrast to traditional FTIR, an online dilution module was integrated with FTIR. Additionally, a multi-point calibration for the FTIR instrument can be easily achieved by using a single bottle of standard gas at any time. The ODM-FTIR can be operated fully automatic automatically with remote operation. The schematic diagram is shown in Fig. 1. The performance of the analytical system was verified using NH₃ in hydrogen as an example. NH₃ standard gas was diluted to different amount fractions using a dynamic dilution module and then detected using FTIR. This ODM-FTIR analysis platform has a good linear relationship, low detection limit, and good accuracy in an amount fraction range, which can meet the requirements of accurate analysis of trace NH₃ in hydrogen with good long-term stability. This work provides an idea for the analysis of other impurities in hydrogen.

  Figure 1

  

  Figure 1.  Schematic diagram based on ODM-FTIR for NH₃.

  DownLoad: Full-Size Img PowerPoint

  In this work, NH₃ in hydrogen is prepared at our laboratory in NIMTT (National Institute of Measurement and Testing Technology of China, Chengdu, China) according to ISO 6142–1:2015 [38]. An aluminum gas cylinder (internal water volume: 8 L) is purchased from Shenyang Zhongfu Kejin Pressure Vessels, Co., Ltd. and is further modified in NIMTT by a proprietary technique to passivate the internal surface. This gas cylinder is equipped with a stainless diaphragm valve with minimum dead volume. High purity H₂ was purchased from Jiangsu Wujiang Messer Industrial Gas Co., Ltd. Pure NH₃ comes from Zhonghao Guangming Research & Design Institute of Chemical Industry Co., Ltd.

  The gas cylinder is evacuated in advance using a turbo molecular pump. In terms of the preparation of NH₃ samples in hydrogen, the “master” standards of NH₃ with the concentration of 500 µmol/mol are prepared in the gas cylinder. NH₃ samples with amount fraction of 10 µmol/mol are further prepared by using the H₂ dilution “Master” standard, detailed principles and procedures were referenced in the paper published by NIMTT in 2020 [39].

  The fourier transform infrared spectrometer was used for measuring the sampling signal (Thermo Fisher Scientific Inc., ANTARIS IGS Analyzer, USA). The gas cell had a 10-meter path long and was operated at 40 ℃ to minimize the inhomogeneity of the gas in the cell. The mirrors within the gas cell were coated with gold to limit the deposition of carbon fines introduced in the H₂ sample. Background spectra (found by scanning with the gas cell filled by H₂) were subtracted from the sample spectrum and their ported value was processed from the average value of the 128 scans, with NH₃ discriminated at 1047.15-1045.14 ⁻¹.

  Dynamic dilution module is used for dilution of standard gas to provide different amounts of fraction trace NH₃, which plays a significant role in calibrating instruments. A two-stage dilution system is designed with dilutions ranging from 30 times to 10,000 times, the design principle is shown in Fig. 1. The flow rates of NH₃ reference standard and H₂ were controlled by mass-flow controllers (MFC1, MFC2, and MFC3) respectively. According to the dilution ratio, a certain flow of NH₃ standard gas (MFC1) and H₂ (MFC2) are mixed and diluted at the coil (1), the obtained diluted gas reaches high-precision pressure sensor, which can accurately monitor the pressure of the mix gas. Under the control of the secondary dilution ratio, the excess diluted mix gas is vented. Moreover, the flow rate of the remaining gas was further adjusted by a critical flow orifice and mixed with H₂ (MFC3) in coil (2) to obtain the target amount of fraction gas. All gas lines are SilcoTek® passivated with 1/16 inch external diameter stainless steel tubing in order to reduce physics or chemistry adsorption.

  For online analysis, the dynamic dilution module was integrated with the FTIR and the NH₃ reference standard was connected directly to the inlet of the dynamic dilution module, so that gases with different amount fractions constantly flowed through the gas cell respectively.

  The gas cell should be continuously purged with H₂ until it is cleaned. During this process, a background spectrum should be collected using H₂ gas. A vacuum pump is used to extract the H₂ from the gas cell. When the pressure of the gas cell is slightly higher than atmospheric pressure, open the outlet valve of the gas cell to make the gas keep flowing state. Subsequently, the instrument reaches stable state, the target gas is measured, and measurement is repeated 7 times for each amount fraction point.

  To evaluate the FTIR spectrometer on real hydrogen gas, the ODM-FTIR analysis platform was installed on-site in a natural gas hydrogen production plant for nearly a month. The hydrogen keeps purging the sampling line during online analysis. When undertaking measurements, sample gas went through the gas cell with 128 scans taken over 210 s. The characteristics of analytical methods were investigated to make sure this method meets the requirements of the ISO 21087:2019 [40].

  The quantitative method of NH₃ is established by the TQ analysis software of FTIR. In accordance with the characteristic absorption peak of NH₃, select an appropriate wavenumber range to integrate the area of the characteristic absorption peak, and establish the standard curve by using the integrated area of the absorption peak and amount fraction of NH₃.

  The LOD was given by (Eq. 1) [40]:

  (1)

  where x_i is the FTIR reading for sample i; x_mean is the average value; m is the number of replicates used during validation of the analytical method. n is the number of replicates in the routine use.

  The limit of quantification (LOQ) was given by (Eq. 2) [40]:

  (2)

  For determine the quantitative wavenumber range more accurately, the characteristic absorption peak of NH₃ was initially determined by measuring a standard substance of NH₃ (Fig. 2A). This standard substance was selected for peak area integration. Given the necessity for the selected absorption peak to exhibit both high signal value and stability without interference from other components, the wavenumber range of 1047.15–1045.14 cm⁻¹ was chosen for area integration in this study (Fig. 2B).

  Figure 2

  

  Figure 2.  (A, B) The characteristic absorption peak of NH₃.

  DownLoad: Full-Size Img PowerPoint

  The parameters such as injection pipeline, resolution and temperature are the key parameters that affect the performance of the analysis system, which are optimized in this work. NH₃ is a highly active and strongly adsorptive alkaline gas. However, its low content (nmol/mol) can easily get adsorbed on the sampling pipe, leading to uncontrollable measurement results. In order to obtain the best suitable pipeline with smallest adsorption, different pipelines with PFA pipe, Peek pipe, and stainless steel passivation pipe with the same length were investigated using NH₃ standard gas (800 nmol/mol). The measurement results indicate that the response values of the PFA pipe are slightly higher compared to the Peek pipe and stainless steel passivation pipe (Fig. 3). Therefore, the PFA pipe was chosen for the subsequent experiment.

  Figure 3

  

  Figure 3.  The effect of injection line with the same length.

  DownLoad: Full-Size Img PowerPoint

  Resolution has a great influence on the spectral signal and precision. In order to obtain smaller spectral noise and finer spectra, spectral signals with resolution from 0.5 cm⁻¹ to 4 cm⁻¹ were investigated using NH₃ standard gas (1.00 µmol/mol). With the increase of the resolution, the fine structure of the spectrum gradually disappears, and the spectral signal also shows a significant decreasing trend (Fig. 4). To ensure the analytical spectrum can accurately and reliably reflect the fine structure and response signal of the object, 0.5 cm⁻¹ was selected for the follow-up experiments.

  Figure 4

  

  Figure 4.  The influence of resolution on the fine structure of spectrum (A) and response signal (B).

  DownLoad: Full-Size Img PowerPoint

  Temperature is another key factor affecting the response value. In the actual analysis process, in addition to considering the signal of the component to be measured, we should also fully consider the condensation of the component to be measured and the possible adverse effects of environmental temperature changes. The effect of temperature on the response value of NH₃ (10.18 µmol/mol) is analyzed at four different temperatures: 40 ℃, 50 ℃, 60 ℃, and 70 ℃. Fig. 5A shows the corresponding absorption spectra at different temperatures. It can be clearly seen that the absolute peak area of the response spectral decreases with increasing temperature, which is consistent with the conclusion derived from the ideal state equation of gas (Fig. 5B). To obtain a lower LOD, 40 ℃ is selected for the follow-up experiments.

  Figure 5

  

  Figure 5.  The influence of gas cell temperature on the response signal spectrum (A) and the response signal comparison of different temperature in terms of the peak area (B).

  DownLoad: Full-Size Img PowerPoint

  Under optimal conditions, the analytical performance of this method is demonstrated using NH₃ as a reference. Figs. 6A and B illustrates the calibration curve and spectral response obtained for contents ranging from 50 nmol/mol to 1000 nmol/mol, respectively. The regression equation is y = 0.0000074x + 0.000024 (where x is the content of NH₃). The linear coefficient was better than 0.999, the LOD of NH₃ was calculated to be 2 nmol/mol and the LOQ was 3.3 nmol/mol.

  Figure 6

  

  Figure 6.  The analytical performance of this method for NH₃. (A) The linear relationship with various contents of NH₃ ranging from 50 nmol/mol to 1000 nmol/mol. (B) Spectral response of NH₃.

  DownLoad: Full-Size Img PowerPoint

  The accuracy and repeatability were further assessed through analysis of reference material at 210 nmol/mol, 510 nmol/mol, and 790 nmol/mol. Measurement was performed over three days (seven replicate analysis each day for each content). Statistical analysis was performed to determine the accuracy and repeatability of the measurement, as shown in Table 1. Test results indicate that this method has satisfactory accuracy and reproducibility.

  Table 1

  

  Table 1.  Investigation of analytical accuracy and precision.

  DownLoad: CSV

  The long-term stability of this method was investigated by ODM-FTIR. Three content points in the linear range, high, medium, and low, were selected and analyzed for day 33 under the same analytical conditions. As shown in Table 2, the relative standard deviation (RSD) of the analysis results ranged from 1.2% to 2.7%, indicating that the ODM-FTIR method holds great long-term stability for sample detection application.

  Table 2

  

  Table 2.  Investigation of long-term stability.

  DownLoad: CSV

  Table 3 summarizes the analytical figures of merit characterizing this method and compares its performance to other methods [41-43]. The LOD of NH₃ obtained by this method is comparable to those obtained by other methods. In addition, the LOD might be further decreased by simply increasing the length of the gas cell. The ISO 21087:2019 addresses the validation of analytical methodology for the analysis of hydrogen in the laboratory. Here, detailed standardized procedures for analyzing NH₃ in hydrogen samples are given and the limit of detection and limit of quantification, working range, trueness, and precision are fit for purpose criteria in ISO 21087:2019.

  Table 3

  

  Table 3.  Analytical figures of merit compared with other methods for NH₃ analysis in hydrogen.

  DownLoad: CSV

  The demonstrated results revealed the great potential of the developed system in determining NH₃ in H₂ with high sensitivity, precision and accuracy. To verify the practical utility of this analysis system, it was applied to the determination of trace of NH₃ in real H₂ samples.

  "Low-carbon energy, low-carbon transportation" is the concept of hosting the Chengdu 2021 FISU World University Games 80 hydrogen buses and 100 food refrigerated logistics vehicles were deployed. Chengdu Hydrogen Energy Co., Ltd. is responsible for providing hydrogen energy during the games, and they produce hydrogen gas from natural gas using the steam methane reforming (SMR) process. It is important to monitor the quality of hydrogen as the lifetime and performance of hydrogen fuel cells are dependent on it.

  To ensure that the quality of the hydrogen products meets the requirement of PEMFCV, the developed ODM-FTIR system was applied to monitor the NH₃ content in Chengdu Hydrogen Energy Co., Ltd. for 21 days. The ODM-FTIR system can also analyze methane, carbon dioxide, carbon monoxide, formaldehyde, formic acid, and hydrogen chloride in the hydrogen gas simultaneously. An online hydrogen quality monitoring data software was developed, which can display the measured values of the ODM-FTIR analysis system in real-time (Fig. S1 in Supporting information).

  During the 21 days, the correctness of the threshold concentration of NH₃ was verified in the pre-operation, mid-operation, and post-operation respectively, to ensure the reliability of online monitoring data. Verification results show that the accuracy is less than 20%, which meets the requirements for the determination of NH₃ in hydrogen (Table 4). The data is collected every hour and the distribution of online measurement data of hydrogen fuel samples is shown in Fig. 7A. Measurement results show that the content of NH₃ in hydrogen samples is lower than the ISO 14687:2019 threshold, indicating that the NH₃ impurity content in hydrogen sample meets the requirements of PEMFCV. The measurement results of the ODM-FTIR method are consistent with the conclusions obtained from the on-site CRDS method, which is validated by NH₃ standard gas, further indicating that this method can be used for accurate monitoring of ammonia content in hydrogen samples (Fig. 7B).

  Table 4

  

  Table 4.  Investigation of analytical accuracy in HRS samples determined by ODM-FTIR online analysis.

  DownLoad: CSV

  Figure 7

  

  Figure 7.  (A, B) Distribution map of online measurement data of hydrogen fuel samples in pipelines of hydrogen production plant.

  DownLoad: Full-Size Img PowerPoint

  In this work, a sensitive and accurate method for determining NH₃ content in hydrogen using the ODM-FTIR system is developed. The method demonstrates a linear correlation for NH₃ in the content range of (50–1000) nmol/mol, with a correlation coefficient greater than 0.999. The LOD was as low as 2 nmol/mol and the LOQ was 3.3 nmol/mol. This method offers high precision, easy operation, high speed, good stability, and high selectivity, making it an ideal choice for determining NH₃ content in hydrogen samples. Additionally, it provides insights for onsite quantitatively analyzing other impurity components in hydrogen.

  Declaration of competing interest

  The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

  CRediT authorship contribution statement

  Wenqing Deng: Data curation, Writing – original draft. Fanfeng Deng: Formal analysis. Ting Zhang: Investigation. Junjie Lin: Software. Liang Zhao: Resources. Gang Li: Resources. Yi Pan: Conceptualization, Project administration, Supervision, Validation, Writing – review & editing, Methodology. Jiebin Yang: Project administration, Resources, Validation.

  Acknowledgment

  We greatly appreciate the financial support by Sichuan Science and Technology, China (No. 2023YFG0070).

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2024.110085.

  
  
• 1. [1]

     J.H. Wee, Renew. Sust. Energ. Rev. 11 (2007) 1720–1738. doi: 10.1016/j.rser.2006.01.005

  2. [2]

     K.Y. Shen, S. Park, Y.B. Kim, Int. J. Hydrogen. Energy 45 (2020) 16773–16786. doi: 10.1016/j.ijhydene.2020.04.147

  3. [3]

     X. Cheng, Z. Shi, N. Glass, et al., J. Power. Source 165 (2007) 739–756. doi: 10.1016/j.jpowsour.2006.12.012

  4. [4]

     B. Shabani, M. Hafttananian, S. Khamani, A. Ramiar, A.A. Ranjbar, J. Power. Source 427 (2019) 21–48. doi: 10.1016/j.jpowsour.2019.03.097

  5. [5]

     W.M. Yan, H.S. Chu, Y.L. Liu, F. Chen, J.H. Jang, Int. J. Hydrogen. Energy 36 (2011) 5435–5441. doi: 10.1016/j.ijhydene.2011.01.158

  6. [6]

     K. Narusawa, M. Hayashida, Y. Kamiya, et al., JSAE Rev. 24 (2003) 41–46. doi: 10.1016/S0389-4304(02)00239-4

  7. [7]

     X.Y. Zhang, H.M. Galindo, H.F. Garces, et al., J. Electrochem. Soc. 157 (2010) B409–B414. doi: 10.1149/1.3284646

  8. [8]

     H. Li, H.J. Wang, W.M. Qian, et al., J. Power Sources 196 (2011) 6249–6255. doi: 10.1016/j.jpowsour.2011.04.018

  9. [9]

     X.Z. Yuan, H. Li, Y. Yi, et al., Int. J. Hydrogen. Energy 37 (2012) 12464–12473. doi: 10.1016/j.ijhydene.2012.05.125

  10. [10]

      International Standard., Hydrogen fuel quality-product specification, ISO 14687, 2019.

  11. [11]

      International Standard., Hydrogen fuel-product specification and quality assurance-Proton exchange membrane (PEM) fuel cell applications for road vehicles, EN 17124, 2018.

  12. [12]

      International Standard., Hydrogen Fuel Quality for Fuel Cell Vehicles, SAE J2719, 2020.

  13. [13]

      H. Meuzelaar, J. Liu, S. Persijn, J.V. Wijk, A.M.H.V.D. Veen, Int. J. Hydrogen. Energy 45 (2020) 34024–34036. doi: 10.1016/j.ijhydene.2020.09.046

  14. [14]

      Y. Pan, F.F. Deng, Z. Fang, et al., Chin. Chem. Lett. 32 (2021) 3440–3445. doi: 10.1016/j.cclet.2021.05.067

  15. [15]

      S.W. Kim, B.A. Trisna, M. Yin, et al., Int. J. Hydrogen. Energy 48 (2023) 13012–13023. doi: 10.1016/j.ijhydene.2022.12.233

  16. [16]

      P.Y. Wang, W.G. Chen, J.X. Wang, et al., Anal. Chem. 95 (2023) 6894–6904. doi: 10.1021/acs.analchem.3c00066

  17. [17]

      Z. Noda, K. Hirata, A. Hayashi, et al., Int. J. Hydrogen. Energy 42 (2017) 3281–3293. doi: 10.1016/j.ijhydene.2016.12.066

  18. [18]

      R. Mukundan, E.L. Brosha, C.J. Romero, D. Poppe, T. Rockward, J. Electrochem. Soc. 167 (2020) 147507. doi: 10.1149/1945-7111/abc43a

  19. [19]

      K. Arrhenius, O. Büker, A. Fischer, S. Persijn, N.D. Moore, Meas. Sci. Technol. 31 (2020) 075010. doi: 10.1088/1361-6501/ab7cf3

  20. [20]

      C. Beurey, B. Gozlan, M. Carré, et al., Front. Energy Res. 8 (2020) 615149.

  21. [21]

      B. Urasinska-Wojcik, J.W. Gardner, IEEE Sens. Lett. 2 (2017) 1–4.

  22. [22]

      D.Z. Zhang, S.J. Yu, X.W. Wang, et al., J. Hazard. Mater. 423 (2022) 127160. doi: 10.1016/j.jhazmat.2021.127160

  23. [23]

      Z.W. Qiu, Y.T. Xue, J.R. Li, et al., Chin. Chem. Lett. 32 (2021) 2807–2811. doi: 10.1016/j.cclet.2021.02.029

  24. [24]

      N. Rajalakshmi, T.T. Jayanth, K.S. Dhathathreyan, Fuel Cells 3 (2003) 177–180. doi: 10.1002/fuce.200330107

  25. [25]

      F.A. Uribe, S. Gottesfeld, T.A. Zawodzinski, J. Electrochem. Soc. 149 (2002) A293–A296. doi: 10.1149/1.1447221

  26. [26]

      F.A. Uribe, T. Zawodzinski, S. Gottesfeld, ECS. PVS. 27 (1998) 229–237. doi: 10.1149/199827.0229pv

  27. [27]

      R. Halseid, P.J.S. Vie, R. Tunold, J. Power Sources 154 (2006) 343–350. doi: 10.1016/j.jpowsour.2005.10.011

  28. [28]

      A.S. Brown, G.M. Vargha, M.L. Downey, et al., NPL Report AS 64, National Physical Laboratory, Teddington, UK, 2011. Available at https://eprintspublications.npl.co.uk/5212/.

  29. [29]

      International Standard., Standard Test Method for Determination of Ammonium, Alkali and Alkaline Earth Metals in Hydrogen and Other Cell Feed Gases by Ion Chromatography, ASTM D7550, 2009.

  30. [30]

      International Standard., Standard Test Method for Hydrogen Purity Analysis Using a Continuous Wave Cavity Ring-Down Spectroscopy Analyzer, ASTM D7941/D7941M, 2023.

  31. [31]

      International Standard., Standard Test Method for Determination of Trace Gaseous Contaminants In Hydrogen Fuel by Fourier Transform Infrared (FTIR) Spectroscopy, ASTM D7653, 2018.

  32. [32]

      National standard., Fuel specification for proton exchange membrane fuel cell vehicles—Hydrogen, GB/T 37244, 2018.

  33. [33]

      National standard., Air Quality—Determination of Ammonia—Ion Selective Electrode Method, GB/T 14669, 1993.

  34. [34]

      A.A. Stec, P. Fardell, P. Blomqvist, et al., Fire. Saf. J. 46 (2011) 225–233. doi: 10.1016/j.firesaf.2011.02.004

  35. [35]

      F. Hase, M. Frey, T. Blumenstock, et al., Atmos. Meas. Tech. 8 (2015) 3059–3068. doi: 10.5194/amt-8-3059-2015

  36. [36]

      J. Mohn, M.J. Zeeman, R.A. Werner, W. Eugster, L. Emmenegger, Isot. Environ. Health Stud. 44 (2008) 241–251. doi: 10.1080/10256010802309731

  37. [37]

      S. Yamanouchi, K. Strong, O. Colebatch, et al., Environ. Res. Commun. 3 (2021) 051002. doi: 10.1088/2515-7620/abfa65

  38. [38]

      International Standard;, Gas Analysis-Preparation of Calibration Gas Mixtures–Part 1: Gravimetric Method For Class I Mixtures, ISO 6142-1, 2015.

  39. [39]

      Y. Pan, Y.J. Zhang, Z.A. Li, et al., Microchem. J. 156 (2020) 104833. doi: 10.1016/j.microc.2020.104833

  40. [40]

      International Standard, Gas Analysis-Analytical Methods For Hydrogen Fuel-Proton Exchange Membrane (PEM) Fuel Cell Applications For Road Vehicles, ISO 21087, 2019.

  41. [41]

      K. Arrhenius, A. Anna, H. Yaghooby, et al., Analysis of Hydrogen Quality According to Standard ISO/DIS 14687-2 Pre-Study, Energiforsk, Stockholm, Sweden, 2015, p. 177. Report 2015 Available at https://energiforskmedia.blob.core.windows.net/media/18581/analysis-of-hydrogen-quality-energiforskrapport-2015-177.pdf.

  42. [42]

      T. Bacquart, K. Arrhenius, S. Persijn, et al., J. Power Sources 444 (2019) 227170. doi: 10.1016/j.jpowsour.2019.227170

  43. [43]

      L. Dong, J. Wright, B. Peters, et al., Appl. Phys B 107 (2012) 459–467.

• 

  Figure 1  Schematic diagram based on ODM-FTIR for NH₃.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  (A, B) The characteristic absorption peak of NH₃.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  The effect of injection line with the same length.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  The influence of resolution on the fine structure of spectrum (A) and response signal (B).

  下载: 全尺寸图片 幻灯片
  

  Figure 5  The influence of gas cell temperature on the response signal spectrum (A) and the response signal comparison of different temperature in terms of the peak area (B).

  下载: 全尺寸图片 幻灯片
  

  Figure 6  The analytical performance of this method for NH₃. (A) The linear relationship with various contents of NH₃ ranging from 50 nmol/mol to 1000 nmol/mol. (B) Spectral response of NH₃.

  下载: 全尺寸图片 幻灯片
  

  Figure 7  (A, B) Distribution map of online measurement data of hydrogen fuel samples in pipelines of hydrogen production plant.

  下载: 全尺寸图片 幻灯片

  Table 1.  Investigation of analytical accuracy and precision.

  下载: 导出CSV

  Table 2.  Investigation of long-term stability.

  下载: 导出CSV

  Table 3.  Analytical figures of merit compared with other methods for NH₃ analysis in hydrogen.

  下载: 导出CSV

  Table 4.  Investigation of analytical accuracy in HRS samples determined by ODM-FTIR online analysis.

  下载: 导出CSV
•