Hydrogen purity determination is a critical quality control activity for fuel cell vehicles, semiconductor manufacturing, chemical synthesis, metallurgy, and power generation, where even trace impurities can degrade performance or damage sensitive equipment. Accredited third-party laboratories typically quantify hydrogen purity by measuring individual impurities at ppm and ppb levels using gas chromatography, spectroscopy, and dedicated trace analyzers, in accordance with standards such as ISO 14687, ISO 21087, ASTM D7653, and the GB/T 3634 series. This article explains the main sample types, testing items, analytical methods, calculation principles, and typical applications of hydrogen purity determination.
Technical Overview
Hydrogen purity refers to the mole fraction of hydrogen (H2) in a gas mixture and is typically reported as a percentage or using grade notation such as 3N (99.9%), 5N (99.999%), and 6N (99.9999%). Common commercial classifications include industrial hydrogen (approximately 99.0–99.5%), pure hydrogen (99.99%), high-purity hydrogen (99.999%), ultra-high-purity hydrogen (99.9999%), and fuel cell grade hydrogen, which must reach at least 99.97% as specified in ISO 14687 for hydrogen fuel used in proton exchange membrane (PEM) fuel cell vehicles.
Because hydrogen is the overwhelming major component of the sample, directly measuring the hydrogen content with high accuracy is analytically difficult. Instead, hydrogen purity is almost always determined indirectly: each impurity is quantified individually at ppm (µmol/mol) or even ppb (nmol/mol) concentrations, the impurity values are summed, and purity is calculated by difference as 100% minus the total impurity content. This approach demands analytical techniques with very low detection limits and rigorous calibration.
A complete determination workflow consists of representative sampling at the production source or delivery point, calibration with certified reference gas mixtures traceable to national metrology institutes, instrumental analysis of each impurity group, purity calculation with a documented measurement uncertainty budget, and comparison against the applicable specification limits. ISO 21087 provides the overall methodological framework for impurity analysis of hydrogen fuel, while ISO 14687 defines the quality thresholds that the analysis must verify.
Testing Samples
Industrial Hydrogen
Industrial hydrogen is produced by steam methane reforming, water electrolysis, chlor-alkali by-product recovery, methanol reforming, or coke oven gas purification, and typically exhibits purity between 98.5% and 99.9%. Characteristic impurities include oxygen, nitrogen, carbon monoxide, carbon dioxide, methane, light hydrocarbons, and moisture. Testing against the requirements of GB/T 3634.1 confirms whether the product meets the industrial quality tier.
High-Purity, Ultra-High-Purity, and Electronic Grade Hydrogen
Grades of 5N and 6N are required for semiconductor fabrication, LED and photovoltaic manufacturing, optical fiber drawing, and quartz melting, where traces of oxygen or moisture can oxidize wafers or disturb deposition processes. In addition to the common permanent gases, these grades carry stringent limits on oxygen, nitrogen, moisture, total hydrocarbons, carbon monoxide, and carbon dioxide, typically verified against GB/T 3634.2 and GB/T 16942 for electronic applications.
Fuel Cell Grade Hydrogen
Hydrogen intended for PEM fuel cell vehicles must comply with ISO 14687 Grade D, which specifies a minimum purity of 99.97% and individual limits for fourteen impurity parameters, including carbon monoxide at 0.2 µmol/mol, total sulfur at 0.004 µmol/mol, total hydrocarbons at 2 µmol/mol, oxygen at 5 µmol/mol, and water at 5 µmol/mol. These extremely low thresholds require specialized ultra-trace analytical techniques.
Sampling Containers and Sample Handling
Valid results begin with valid samples. Hydrogen is normally collected into passivated 316L stainless steel cylinders with inert interior surfaces, connected through low-dead-volume fittings and particle filters. Cylinders must be conditioned by repeated pressurization and evacuation with the sample gas to remove adsorbed moisture and residual air. Elastomer seals that could outgas hydrocarbons are avoided, cylinder pressure and sample temperature are recorded, and sampling practices follow the guidance of ISO 19880-1 for hydrogen fueling infrastructure.
Testing Items
A full hydrogen purity determination characterizes the following parameters, with the exact list dictated by the applicable purity grade and specification:
- Hydrogen purity (mole fraction, determined by difference)
- Oxygen content
- Nitrogen and argon content
- Helium content
- Carbon monoxide, carbon dioxide, and methane content
- Total hydrocarbons expressed on a methane-carbon basis
- Moisture content and dew point temperature
- Total sulfur compounds (hydrogen sulfide, carbonyl sulfide, mercaptans)
- Ammonia content
- Formaldehyde and formic acid content
- Total halogenated compounds
- Particulates and oil mist in compressed hydrogen streams
Testing Methods
Gas Chromatography with Thermal Conductivity Detection (GC-TCD)
GC-TCD is the workhorse method for permanent gases such as oxygen, nitrogen, methane, carbon monoxide, argon, and helium. Molecular sieve 5A columns separate oxygen, nitrogen, methane, and carbon monoxide, while porous polymer columns handle carbon dioxide. The thermal conductivity detector is universal, stable, and non-destructive, with practical detection limits of roughly 1–10 µmol/mol, making it ideal for industrial and high-purity grades. Calibration is performed with certified reference gas mixtures, and special column configurations resolve the oxygen–argon co-elution problem.
Gas Chromatography with Flame Ionization Detection (GC-FID)
The flame ionization detector provides high sensitivity for hydrocarbon impurities and is the reference approach for total hydrocarbons reported on a methane-carbon equivalent basis. When equipped with a methanizer, a nickel catalyst reactor that converts carbon monoxide and carbon dioxide into methane, the FID achieves sub-ppm detection of these critical fuel cell poisons. Dedicated total hydrocarbon analyzers based on FID are also deployed for continuous on-line quality monitoring at production facilities.
GC with Pulsed Discharge Helium Ionization Detection (GC-PDHID)
The pulsed discharge helium ionization detector is a universal detector capable of measuring light permanent gases and hydrocarbons at single-digit ppb levels in a single chromatographic run. It forms the basis of ASTM D7653, the standard test method for trace gaseous impurities in high-purity hydrogen fuel, and is widely used by accredited third-party laboratories for fuel cell grade verification because one instrument configuration can cover a large portion of the ISO 14687 impurity list.
Gas Chromatography–Mass Spectrometry (GC-MS)
GC-MS combines chromatographic separation with mass-selective detection and is employed for screening unknown organic contaminants, speciating halogenated compounds, and investigating contamination incidents. It is particularly valuable when a hydrogen stream shows an unexplained total hydrocarbon result and molecular identification is required before corrective action can be defined.
Trace Oxygen Analysis
Oxygen is measured with electrochemical (galvanic fuel-cell type) sensors, paramagnetic analyzers, or trace oxygen monitors, following methods such as GB/T 6285 for the electrochemical determination of trace oxygen in gases. Modern instruments reach sub-ppm detection limits, and interferences from the hydrogen matrix are corrected through appropriate cell design and calibration strategy.
Moisture and Dew Point Determination
Moisture is one of the most demanding parameters because adsorption and desorption on surfaces distort results. Chilled-mirror hygrometry provides the highest accuracy and is the fundamental reference technique, while electrolytic phosphorus pentoxide cells (GB/T 5832.1) and capacitive aluminum oxide or polymer sensors (GB/T 5832.2 dew point method) support routine and field measurements. Measured dew point temperatures are converted to ppm(v), and passivated, heated transfer lines together with long stabilization times are essential for reliable ultra-trace moisture data.
Spectroscopic Methods (FTIR, CRDS, TDLAS)
Fourier transform infrared spectroscopy (FTIR) enables simultaneous multi-component quantification of ammonia, formaldehyde, formic acid, carbon monoxide, and carbon dioxide directly in the hydrogen matrix. Cavity ring-down spectroscopy (CRDS) and tunable diode laser absorption spectroscopy (TDLAS) offer ppb-level sensitivity for trace moisture and other targeted species, with fast response and no consumables, making them well suited for on-line monitoring at hydrogen fueling stations and electrolyzer outlets.
Impinger Sampling with Ion Chromatography
For polar species that gas chromatography handles poorly, such as ammonia, hydrogen halides, formate, and sulfate, the sample is bubbled through a dilute absorbing solution in impinger trains and the resulting solution is analyzed by ion chromatography with suppressed conductivity detection. This technique achieves detection limits at the low nmol/mol to pmol/mol level and is a standard approach for several ISO 14687 impurity parameters.
Sulfur Compound Analysis
Total sulfur must be controlled to 0.004 µmol/mol for fuel cell hydrogen, which drives the use of gas chromatography with flame photometric detection (FPD) or sulfur chemiluminescence detection (SCD), often combined with cryogenic or adsorptive preconcentration. Because sulfur species are extremely surface-active, inert-coated sample pathways and passivated cylinders are mandatory to avoid loss during transport and analysis.
Purity Calculation and Data Evaluation
After all impurities are quantified, hydrogen purity is calculated as 100% minus the arithmetic sum of the measured impurity mole fractions, including water vapor. Each result is reported with its measurement uncertainty, and the overall impurity budget is compared with the applicable grade limit as required by GB/T 3634.1, GB/T 3634.2, or ISO 14687. ISO 21087 defines the validation requirements that analytical procedures must satisfy, including calibration traceability, detection limit verification, and participation in inter-laboratory comparisons, ensuring that the purity value issued by an accredited third-party laboratory is technically defensible.
Applications
Hydrogen purity determination supports quality assurance, regulatory compliance, and process safety across a wide range of industries:
- Fuel cell electric vehicles: verification of hydrogen fuel quality at production plants, tube trailers, and refueling stations against ISO 14687 and the fuel quality provisions of ISO 19880-1.
- Semiconductors and electronics: incoming inspection of 5N–6N hydrogen for epitaxy, diffusion, annealing, and solder-reflow protective atmospheres.
- Chemical industry: quality control of hydrogen feeds for ammonia synthesis, methanol production, and catalytic hydrogenation, where sulfur and carbon monoxide poison catalysts.
- Metallurgy and materials: certification of hydrogen used for bright annealing, reduction of metal oxides, and powder metallurgy sintering atmospheres.
- Power generation: monitoring of hydrogen coolant purity in large turbine generators, where contamination reduces cooling efficiency and raises explosion risk.
- Aerospace and research: characterization of hydrogen for propulsion testing, laboratory studies, and the development of electrolyzers and fuel cells under IEC 62282 framework applications.
- Gas manufacturing and distribution: batch release testing, cylinder certification, and investigation of contamination complaints throughout the supply chain.
FAQ
What purity is required for hydrogen used in fuel cell vehicles?
ISO 14687 Grade D requires a minimum hydrogen purity of 99.97% and limits the total of all non-hydrogen gases to 100 µmol/mol, with individual caps such as 0.2 µmol/mol for carbon monoxide and 0.004 µmol/mol for total sulfur, because even trace catalyst poisons degrade PEM fuel cell performance irreversibly.
Why is purity determined from impurity measurements instead of measuring hydrogen directly?
A direct 100%-scale hydrogen measurement cannot reliably distinguish 99.999% from 99.9999%, since the analytical uncertainty of a full-scale determination is far larger than the impurity levels involved. Measuring each impurity at its own ppm or ppb level and calculating purity by difference provides dramatically better resolving power at high purity grades.
Which method is most commonly used for hydrogen purity testing?
Gas chromatography is the backbone of virtually every testing scheme. A multi-detector GC platform combining TCD, FID with methanizer, and PDHID covers most of the impurity list, supplemented by moisture analyzers, FTIR or laser spectrometers, ion chromatography, and sulfur-specific detectors for the remaining parameters.
What precautions are necessary when sampling hydrogen for purity testing?
Samples must be collected into passivated stainless steel cylinders that have been repeatedly purged and pressurized with sample gas, using metal-diaphragm valves and low-dead-volume connections. Air ingress, elastomer outgassing, moisture adsorption, and sulfur adsorption onto active surfaces are the main sources of false results, so cylinder conditioning and leak checking are mandatory.
What is the difference between high-purity and ultra-high-purity hydrogen?
High-purity hydrogen (5N) contains no more than 10 µmol/mol of total impurities, while ultra-high-purity hydrogen (6N) is limited to 1 µmol/mol. The 6N grade imposes proportionally tighter limits on oxygen, nitrogen, moisture, and hydrocarbons and is specified mainly by the semiconductor and optoelectronics industries.
What detection limits can modern methods achieve?
GC-PDHID and laser-based spectroscopy such as CRDS routinely reach low ppb (nmol/mol) detection limits, which is sufficient to verify even the strictest ISO 14687 thresholds, including the 0.004 µmol/mol total sulfur requirement, provided that sampling, calibration, and quality assurance follow the validated procedures described in ISO 21087.