Why hydrogen is hard to sense

Summary

Most cheap gas sensors are infrared absorption sensors. CO2, CH4, and CO have a charge asymmetry that oscillates as they vibrate, so they absorb IR strongly at signature wavelengths and an NDIR detector reads them out for a few dollars. H2 is two identical atoms sharing electrons symmetrically: it has no permanent dipole moment, vibration and rotation do not create one, and ordinary IR absorption is forbidden. There is no NDIR-style hydrogen sensor and there cannot be one.

Every practical H2 sensor therefore uses a workaround: heat flow, speed of sound, electrical resistance of a palladium film, catalytic combustion, electrochemical oxidation, or the residual (very weak) quadrupole and Raman couplings. Each workaround trades away cost, selectivity, temperature/pressure robustness, or ppm-level sensitivity. That one fact of molecular symmetry explains why soil-gas surveys run on electrochemical cells rather than lasers, why no drop-in downhole H2 sensor exists, and why every technology in this wiki is a compromise.

Why the optical routes fail

No dipole. H2 is homonuclear: the D∞h symmetry forces the electric dipole moment to zero at every internuclear distance, so there is no dipole-allowed rovibrational spectrum [1]. Dipole-allowed IR absorbers like CO, HCl, and H2O have line strengths of typically 10⁻²¹–10⁻¹⁹ cm/molecule; H2 has none. This is why H2 is absent from an ordinary FTIR spectrum and why NDIR is structurally inapplicable.

Quadrupole lines, 5–7 orders weaker. The electric-quadrupole term survives, with selection rule $\Delta J = 0, \pm2$ (O, Q, S branches) [1]. Herzberg first observed this spectrum in 1949 [2]. The strongest H2 quadrupole lines, near 4000 cm⁻¹ (≈2.5 µm), have intensities of order 10⁻²⁶ cm/molecule, the values in HITRAN since HITRAN2012 [1]; weaker branches and overtones fall toward 10⁻²⁷. Every laser-absorption H2 sensor therefore needs meters of pathlength, a high-finesse cavity, or both (see IR absorption and proxies). Laboratory instruments do get there: comb-referenced CRDS has measured (2–0) band lines near 1.2 µm to sub-100 kHz accuracy [3], resolved the collision-induced (H2)2 dimer at room temperature [4], and reached noise-equivalent absorption of ~4×10⁻¹⁰ cm⁻¹ Hz^(−1/2) on the 3–0 band [5]. These are metrology tools, not field sensors.

Collision-induced absorption is a density effect, not a sensing route. During a collision, two molecules transiently induce a dipole in each other, giving a continuum absorption that scales roughly with density squared ($\alpha \approx \alpha_1\rho_a\rho_p + \alpha_2\rho_a\rho_p^2$) [6][7]. It is negligible at atmospheric density and dominant only in planetary and stellar atmospheres.

Downhole laser absorption

At borehole pressures (tens to low hundreds of MPa), pressure broadening and incipient CIA continuum must be accounted for if laser absorption is ever used downhole. Nobody has industrialized CIA itself for trace H2.

Electronic transitions sit in the vacuum UV, where air is opaque. The dipole-allowed Lyman (B¹Σᵤ⁺–X¹Σg⁺) and Werner (C¹Πᵤ–X¹Σg⁺) bands span roughly 90,000–119,000 cm⁻¹ (11.2–13.6 eV; ~90–115 nm strong, weak tail toward ~160–180 nm) [8][9], and atomic hydrogen has Lyman-α at 121.567 nm [10]. O2 absorbs essentially completely across this range: the Schumann–Runge continuum covers ~130–175 nm and structured bands ~175–200 nm [11][12], so a VUV photon survives centimeters to meters of open air. (Solar Lyman-α sits in a local minimum of the O2 cross-section and penetrates to ~70–75 km altitude [13]; an aeronomy detail, not a sensing opportunity.) Even in vacuum, window materials cap the technique: LiF cuts off near 103–105 nm and is hygroscopic, MgF2 near 115 nm [14][15]. Two-photon laboratory schemes exist for plasma diagnostics: TALIF on atomic H (2×205.1 nm excitation, Balmer-α fluorescence at 656.3 nm) [16] and two-photon excitation of H2 to the E,F¹Σg⁺ state near 201–202 nm [17][18]. Neither is field-deployable. See UV/EUV & fluorescence.

The reason there is no cheap IR hydrogen sensor: H₂’s only infrared coupling is the electric-quadrupole mechanism, five to seven orders of magnitude weaker than the dipole-allowed bands behind NDIR CO₂ and CH₄ sensors.

The physical handles that remain

Every practical H2 sensor exploits one of these properties, in most cases not an optical spectrum at all:

Physical property	H2 value	Comparison	Sensor family it enables
Polarizability	≈5.41 atomic units ≈ 0.80 ų [19][20]	~1/3 of N2’s	Raman scattering: weak, but an independent selection rule
Thermal conductivity	≈0.18 W/(m·K) near room temp. [24]	~7× air (~0.026 W/(m·K)); highest of any gas except He	Thermal-conductivity (katharometer/MEMS) sensing
Speed of sound	≈1315 m/s @ 25 °C [24]	~3.8× air (343 m/s), via $c=\sqrt{\gamma RT/M}$ and H2’s low molar mass	Acoustic/ultrasonic gas-ID sensing
Molecular diffusivity in air	≈0.76×10⁻⁴ m²/s [21]	among the highest of any gas	Shapes leak-detection geometry and soil-gas transport modeling
Catalytic dissociative chemisorption on Pd/Pt	H2 → 2H(ads)	—	Heat of combustion (pellistors); work-function shift
Absorption into Pd lattice → PdHx	up to ~3% linear lattice expansion; resistivity increase up to ~80%	—	Palladium resistive, capacitive, and Pd-coated fiber sensors
Electrochemical oxidation at a Pt electrode	H2 → 2H⁺ + 2e⁻	—	Amperometric electrochemical cells
Residual quadrupole IR / Raman activity	~10⁻²⁶ cm/molecule (quadrupole IR)	5–7 orders weaker than dipole absorbers	Laser absorption (TDLAS/WMS/CRDS); Raman spectroscopy

Sensor families this enables

This table is the map the wiki is organized around. No family covers the whole problem; each trades sensitivity, selectivity, robustness, or field-readiness against the others.

Family	What it senses	Headline verdict	Page
Catalytic combustion (pellistor / catalytic bead)	Heat of combustion on a heated Pt/Pd bead [25]	Cheap and fast, but needs ambient O2 and cannot distinguish H2 from other combustibles	Catalytic pellistors
Thermal conductivity (TCD / MEMS-TC)	H2’s unusually high thermal conductivity [22][23]	Immune to poisoning, very low drift, but chemically nonspecific and not ppm-capable	Thermal conductivity & acoustic sensors
Chemiresistive / metal-oxide (MOX)	Surface conductance change on heated SnO2-type films [26][27]	Sub-ppm capable and cheap, but humidity- and cross-gas-sensitive	MOX chemiresistors
Electrochemical (amperometric)	Faradaic current from H2 oxidation at a Pt electrode [28]	Today’s soil-gas workhorse: ppm-level, compact, low-power; limited (1–3 yr) cell life and CO cross-sensitivity	Electrochemical cells
Palladium hydride, resistive/capacitive	Volume/resistivity change on PdHx formation [29][30]	Mature and non-consumable per vendor claims; H2S-vulnerable and temperature-limited	Palladium resistive sensors
Palladium work-function (MIS/MIM/Schottky/HEMT)	Interfacial dipole shifting a threshold voltage or barrier height	O2-independent, best-matched family for high-temperature downhole work; H2S poisoning is the likely failure mode	Work-function devices
Pd-coated optical fiber (FBG/SPR)	PdHx transduced optically instead of electrically [31][32][33]	The only technology with zero downhole electronics; demonstrated only at 1–4 vol%, room T/P	Fiber-optic sensors
Raman spectroscopy	H2’s anisotropic polarizability, independent of dipole/quadrupole rules [34]	Multi-species readout, no sample consumption; intrinsically weak signal	Raman spectroscopy
Enhanced Raman / plasmonic-field techniques	Same polarizability handle, boosted by a plasmonic or waveguide-enhanced local field	Frontier route to closing Raman’s sensitivity gap for trace H2	Enhanced Raman
Laser absorption (TDLAS / WMS / CRDS)	The weak quadrupole lines directly [35][3][4][5]	Selective and non-consuming; needs meters of path or a high-finesse cavity, capital-instrument grade	IR absorption and proxies
UV/EUV & fluorescence (TALIF, VUV absorption)	Lyman/Werner electronic bands, atomic Lyman-α	Vacuum-chamber only; O2 destroys the signal in open air	UV/EUV & fluorescence
Plasmonic metamaterials	Engineered resonant structures (often Pd-functionalized) coupling light to PdHx	Research frontier, not field-proven	Plasmonic metamaterials

Acoustic/speed-of-sound gas-ID sensing appears in the handle table but is not carried forward as an exploration-grade family: helium and any other low-molar-mass gas confounds the reading. See Thermal conductivity & acoustic sensors.

Performance envelope

Two anchor numbers

The tropospheric H2 background is 531 ± 6 ppb (≈0.53 ppm), from a 50-site global network over 1991–1996, ~2-year atmospheric lifetime [36], corroborated regionally at 0.53–0.59 ppm [37]. The lower explosive limit is 4 vol% (flammable to 75 vol%, minimum ignition energy ~0.02 mJ, burning velocity ~265 cm/s) [38][39]. Every sensor family works somewhere on the eleven decades between these.

The sensitivity landscape: every technology family in this wiki, placed by detection limit and response time (dots: geometric mean of the published range; bars: the range). More on the charts page.

Sensor family	Typical demonstrated range	Best-case LOD	Response time (t90)	Field readiness	Refs
Catalytic bead (pellistor)	0–100% LEL (0–4 vol% H2 typical span)	Not ppm-capable; ~100s ppm–%LEL	<10 s classic; MEMS <12 s, some <2 s	Mature, ubiquitous safety instrument	[40][41][25]
Thermal conductivity (MEMS-TC)	0–16 vol% to 0–100% depending on model	Coarse; not sub-100 ppm capable	<100 ms (best-in-class MEMS)	Mature; safety/LEL role, not trace exploration	[22][23][42]
Electrochemical (amperometric)	0–1,000 ppmv (soil-gas survey instruments); 0–2,000 ppm (OEM cells)	~1 ppmv precision; <0.7 ppm RMS noise	<45 s, 0→400 ppm	Mature; de facto soil-gas field standard	[28][43][44][45][46]
Metal-oxide semiconductor (MOX/SnO2)	sub-ppm to % (claimed)	Sub-ppm demonstrated in lab	Seconds, degraded by humidity	Lab-proven; not field-proven for quantitative exploration	[26][27][47]
Pd/Pd-alloy resistive (solid-state)	ppm to %	Manufacturer claims no cross-sensitivity to other gases	Seconds (design-dependent)	Commercial (industrial/process instrument class)	[29][30][48][49][50]
Pd-coated fiber-optic (FBG/SPR)	~1–4 vol% demonstrated	Tens of pm wavelength shift per vol% H2	Seconds (diffusion-limited by film thickness)	Research/early-commercial; only zero-downhole-electronics candidate	[31][32][33]
Raman gas analyzer	Percent-level process streams (commercial); ppm-level in specialized lab setups	Weak signal limits ppm-level field use today	Seconds (spectral acquisition)	Commercial for process streams; not yet trace-background capable	[51][52][53]
TDLAS / WMS on quadrupole line (2121.8 nm)	Demonstrated 0.02 vol% precision, 1 m path, 1 s integration	Sub-0.1 vol%-class demonstrated	Seconds	Lab/process demonstration; not field-hardened	[35]
CRDS on quadrupole lines	Research-grade; not a field product	NEA ~4×10⁻¹⁰ cm⁻¹ Hz^(−1/2) demonstrated	Seconds per spectrum	Laboratory only (metrology tool)	[3][4][5]

Every LOD, response-time, and field-readiness entry above is a vendor or paper claim; see Reading the specs for how to separate lab best-case numbers from field-realistic ones.

Cross-sensitivity & environment

Cross-cutting summary; each family’s page has the detail.

Factor	Who it hits and how	Ref
Temperature	Electrochemical cells rated roughly −30…+50 °C [28]; TC/MEMS tolerate up to 105 °C with no chemistry to disrupt [22]; commercial Pd:Ni lines rated only to ~60 °C in process configurations [48], below wellbore and many summer surface temperatures	[22][28][48]
Humidity	First-order confound for MOX (baseline and kinetics both shift; “anti-humidity” coatings remain an active research topic) [26][27][47]. Electrochemical cells are rated across wide RH bands; buried field probes regulate internal humidity to protect electronics [45]. TC sensors are largely immune	[26][27][45][47]
Pressure	500–1400 mbar has only minor effect on a widely used electrochemical field analyzer, but depressurization below ~1 atm causes measurable overestimation of H2 [46]: relevant at altitude, during barometric pumping, and at wellheads without pressure conditioning	[46]
O2 availability	Pellistors stop working or under-read in O2-depleted or inert atmospheres [40]; two-electrode electrochemical cells also rely on O2 at the counter electrode	[40]
CO	Principal residual interferent for several electrochemical H2 cells [45]; CO-selective inlet membranes can nearly eliminate it short-term (unverified) [54]. Seep redox environments can co-produce trace CO	[45][54]
CH4	Low cross-sensitivity for electrochemical cells [44]. Pellistors respond to any combustible and cannot distinguish H2 from CH4 without a dual-bead scheme	[44]
H2S	Poisons the two most H2-selective chemical mechanisms, catalytic and Pd-based: sulfide layers block the sites H2 dissociation needs, measured directly on single Pd nanoparticles [55][56]. For pellistors, sulfur (with silicones, lead, halogenated solvents) is an irreversible poison, distinct from recoverable inhibitors [41]	[41][55][56]
CO2	Weak direct interferent, but elevated soil CO2 is itself a seep/microbial co-indicator, and CO2 channels are commonly bundled in the same field instrument	[44]

H2S deserves emphasis: natural hydrogen accumulations are frequently co-evolved with H2S, so prefer electrochemical or TC sensors as the primary quantitative instrument, use catalytic/Pd sensors as a secondary check, and budget Pd/pellistor replacement or thermal regeneration where H2S is expected [55].

Failure modes to design around

1. Catalytic-sensor failure in O2-poor or H2S-rich boreholes;
2. electrochemical-cell drift and finite (1–3 year) life requiring recalibration;
3. irreversible Pd/pellistor poisoning by sulfur/silicone/lead with no in-field recovery;
4. MOX humidity artifacts at unattended stations;
5. pressure-transient (depressurization) artifacts during sample extraction or barometric pumping [46].

Use for geologic H₂

Field verdict

No single technology spans the eleven decades from atmospheric background (0.53 ppm) to the 4 vol% LEL, and none is field-proven at borehole temperature and pressure. Soil-gas surveys are well served by electrochemical multi-gas analyzers. Continuous monitoring runs on the same electrochemical technology, trading multi-year lifetime for ppm resolution. Downhole logging is unsolved: borehole temperature, the zero-downhole-electronics preference, and each chemical sensor’s failure mode rule out nearly every family. Pd-coated fiber is the one physically plausible downhole path, unvalidated above room T/P or beyond ~4 vol%.

Soil-gas surveys. The best-served use case. Portable electrochemical multi-gas analyzers report H2 at 0–1,000 ppmv with ~1 ppmv precision alongside O2, CO2, CH4, and optionally CO/H2S; dual-range instruments extend to 0–10,000 ppmv and 0.1–100 vol% for rich seeps [44]. Purpose-built probes bury the cell at ~1 m to damp diurnal temperature swings, with CO as the main residual cross-sensitivity [45]. The GA5000 metrology assessment [46] is worth reading before any campaign: generally reliable across field T/RH/P, with the depressurization bias noted above. Standard sampling methodology (~1 m depth, ~2 cm drainage radius, up to 500 mL/min pumped) brings representativeness and cross-contamination concerns that are procedural, not sensor, limitations [57].

Downhole / borehole logging. Unsolved. Borehole temperatures reach 150–175 °C onshore and ~200 °C offshore while downhole-rated Si/SiC electronics stop at ~175 °C commercially [58][59], and Hestia prefers no downhole electronics at all. On top of that: pellistors need O2 a reducing borehole may lack, electrochemical electrolytes are neither temperature-rated nor long-lived, and Pd hydride sensing loses sensitivity with temperature and is H2S-vulnerable. Pd-coated fiber is the exception in principle (passive sensing point, surface interrogator, same fiber infrastructure as DTS completions [31][32][33]), but demonstrated performance is 1–4 vol% at room T/P [31]. The pragmatic answer today is mud-gas / wireline fluid sampling with surface analysis [57].

Continuous monitoring stations. Power budget, calibration-free lifetime, and drift immunity dominate over LOD. TC/MEMS offers sub-mW power and multi-year calibration-free life [22][23][42] but cannot resolve ppm anomalies; it fits wellhead safety, not baseline science. Electrochemical stations are the only demonstrated ppm-resolution unattended option, at the cost of multi-year service life and desiccant/recalibration management [45]. A ruggedized Raman or TDLAS/CRDS station would remove the poisoning failure modes but exists nowhere as a field product; that gap is a candidate R&D investment.

See Geologic context for the exploration-side framing.

Vendors & products

Landscape summary; detailed vendor tables live on each technology page.

Technology	Representative vendors / products	Field readiness	Page for full table
Catalytic bead / pellistor	SGX Sensortech, Winsen, Draeger, Blackline Safety	Mature, ubiquitous	Catalytic pellistors
MEMS thermal conductivity	Infineon XENSIV TCI, Posifa PGS1000/PGS4100, SGX TC-1326	Mature; safety/LEL role	Thermal conductivity & acoustic sensors
Chemiresistive / MOX	SnO2-based OEM chemiresistors; sensor-array research prototypes	Lab-proven; not field-proven for quantitative exploration	MOX chemiresistors
Electrochemical	Alphasense H2-AF/H2-B4; Geotech GA2000/GA5000/BIOGAS5000; Ion Science/Geotech Variotec 460; MONHyTOR (buried probe)	Mature; de facto field standard	Electrochemical cells
Pd/Pd-alloy resistive (solid-state)	H2scan HY-OPTIMA / HY-ALERTA lines	Commercial (industrial/process instrument)	Palladium resistive sensors
Pd/Pt work-function (MIS/MIM/Schottky/HEMT)	AppliedSensor/ams HLS-440 family; H2scan hybrid MIS+resistor lines; SenSiC SiC-FET	Mature (MIS) to research-stage (HEMT/nanostructured)	Work-function devices
Pd-coated optical fiber	Academic groups (FBG/SPR designs); no dominant commercial vendor identified	Research/early-commercial	Fiber-optic sensors
Raman gas analyzer	Endress+Hauser Raman Rxn5; NETL (DOE) Raman Gas Analyzer prototype	Commercial for process streams	Raman spectroscopy
Laser absorption (TDLAS/WMS)	Research-prototype instruments (2121.8 nm quadrupole line)	Lab/process demonstration	IR absorption and proxies

OEM element manufacturers do not publish list prices; analyzers and spectroscopic systems are complete instruments at higher price points. No verified list prices were located for any SKU, so all cost-class statements in this wiki are relative.

Open questions

1. No field-validated downhole H2 sensor exists in the surveyed literature. Before committing downhole R&D capital, look for any Pd-FBG demonstration above 100 °C / tens of MPa, and ask whether an oilfield DTS/DAS vendor could piggyback a Pd-functionalized channel on an existing deployment.
2. Quantify H2S poisoning kinetics for the specific sensors Hestia will buy, via vendor support or an independent test lab; basin-realistic ppm-level poisoning curves are not in the literature reviewed.
3. Reconcile the two diffusivity figures (0.61 vs. 0.76–0.80 cm²/s) against the primary Marrero & Mason (1972) compilation; diffusivity feeds soil-gas transport models.
4. Evaluate whether a Raman or TDLAS/CRDS system could be ruggedized into a continuous exploration-baseline station before assuming it is purely a research problem.
5. Get an authoritative price quote sheet; no verified list prices were found for any SKU.
6. Build a decision-tree / selection-matrix artifact mapping Hestia’s three deployment contexts against the performance table and cross-sensitivity flags.

Related pages

• Reading the specs: how to read the datasheet numbers behind the performance-envelope table above
• Raman spectroscopy
• Enhanced Raman
• IR absorption and proxies
• UV/EUV & fluorescence
• Fiber-optic sensors
• MOX chemiresistors
• Work-function devices
• Palladium resistive sensors
• Thermal conductivity & acoustic sensors
• Catalytic pellistors
• Electrochemical cells
• Plasmonic metamaterials
• Geologic context

References

Full reference list (59 entries)

1. Somogyi, W., Yurchenko, S.N., Yachmenev, A., “Calculation of electric quadrupole linestrengths for diatomic molecules: Application to the H2, CO, HF and O2 molecules,” J. Chem. Phys. 155, 214303 (2021). DOI: 10.1063/5.0063256. https://pubs.aip.org/aip/jcp/article/155/21/214303
2. Herzberg, G., “Quadrupole Rotation-Vibration Spectrum of the Hydrogen Molecule,” Nature 163, 170 (1949). https://ui.adsabs.harvard.edu/abs/1949Natur.163..170H
3. “The high-accuracy spectroscopy of H2 rovibrational transitions in the (2-0) band near 1.2 μm,” Phys. Chem. Chem. Phys. (2023). https://pubs.rsc.org/en/content/articlelanding/2023/cp/d3cp01136d
4. Fleurbaey, H., Kassi, S., Campargue, A., “Room temperature detection of the (H2)2 dimer,” Phys. Chem. Chem. Phys. 26, 21974–21981 (2024). DOI: 10.1039/D4CP02605E. https://pubs.rsc.org/en/content/articlelanding/2024/cp/d4cp02605e
5. NIST, “The 3-0 Quadrupole Band in H2 by Cavity Ring-Down Spectroscopy.” https://www.nist.gov/publications/3-0-quadrupole-band-h2-cavity-ring-down-spectroscopy
6. “High-temperature (1000–7000 K) collision-induced absorption of H2 pairs computed from first principles, with application to cool and dense stellar atmospheres.” https://www.sciencedirect.com/science/article/abs/pii/S0022407300000236
7. “Pressure-induced infrared absorption of hydrogen and hydrogen–foreign gas mixtures.” https://cdnsciencepub.com/doi/10.1139/p58-009
8. Wikipedia, “Lyman–Werner photons.” https://en.wikipedia.org/wiki/Lyman%E2%80%93Werner_photons
9. “Highly accurate transition frequencies in the H2 Lyman and Werner absorption bands.” https://www.academia.edu/19365646/Highly_accurate_transition_frequencies_in_the_H_2_Lyman_and_Werner_absorption_bands
10. Wikipedia, “Lyman-alpha.” https://en.wikipedia.org/wiki/Lyman-alpha
11. ScienceDirect Topics, “Schumann-Runge Band.” https://www.sciencedirect.com/topics/earth-and-planetary-sciences/schumann-runge-band
12. Blake, A.J., “An atmospheric absorption model for the Schumann-Runge bands of oxygen,” J. Geophys. Res. Space Phys. (1979). https://agupubs.onlinelibrary.wiley.com/doi/abs/10.1029/JA084iA07p03272
13. “The H Lyman-Alpha actinic flux in the middle atmosphere.” https://www.researchgate.net/publication/241275320_The_H_Lyman-Alpha_actinic_flux_in_the_middle_atmosphere
14. “Passivation Strategies for Far-Ultraviolet Al Mirrors Using Plasma-Based AlF3 Processing,” Chem. Mater. (2025). https://pmc.ncbi.nlm.nih.gov/articles/PMC12461835/
15. NASA NTRS, “Enhanced Aluminum reflecting and solar-blind filter coatings for the [FUV].” https://ntrs.nasa.gov/api/citations/20170007449/downloads/20170007449.pdf
16. Niemi, K., Schulz-von der Gathen, V., Döbele, H.F., “Absolute calibration of atomic density measurements by laser-induced fluorescence spectroscopy with two-photon excitation,” J. Phys. D: Appl. Phys. 34, 2330 (2001). https://pureadmin.qub.ac.uk/ws/files/551190/niemi01JPD34_2330.pdf
17. “Laser-induced fluorescence of hydrogen via the E,F1Σg+→B1Σu+ transition: Rotational-state-dependent collisional quenching.” https://www.sciencedirect.com/science/article/abs/pii/S0009261411001503
18. “Quantitative determination of H2, HD, and D2 internal-state distributions by (2+1) resonance-enhanced multiphoton ionization.” NASA Technical Reports Server. https://ntrs.nasa.gov/citations/19910057997
19. “Polarizability of molecular hydrogen and gas metrology,” Phys. Rev. A 111, 012801 (2025). https://link.aps.org/doi/10.1103/PhysRevA.111.012801
20. “On the Polarizability of the Hydrogen Molecule.” https://www.researchgate.net/publication/230982280_On_the_Polarizability_of_the_Hydrogen_Molecule
21. h2tools.org, “Hydrogen Diffusion in Air” (H2 Tools FAQ). https://h2tools.org/faq/hydrogen-diffusion
22. Infineon Technologies, “XENSIV™ TCI H2 sensor: Hydrogen sensor based on thermal conductivity principle.” https://www.infineon.com/cms/en/product/sensor/gas-sensors/h2-sensors/tci/
23. Posifa Technologies, “Hydrogen Sensor for fuel cell and safety applications - PGS1000.” https://posifatech.com/pgs1000/
24. Engineering ToolBox, “Hydrogen - Thermophysical Properties” (thermal conductivity, sound velocity). https://www.engineeringtoolbox.com/hydrogen-d_1419.html
25. Winsen Sensor, “Catalytic Bead Hydrogen Sensors: Principles, Features, and Applications.” https://www.winsen-sensor.com/knowledge/catalytic-bead-hydrogen-sensors.html
26. Frontiers, “Review: Influences of Semiconductor Metal Oxide Properties on Gas Sensing Characteristics.” https://www.frontiersin.org/journals/sensors/articles/10.3389/fsens.2021.657931/full
27. “Review of noble metal and metal-oxide-semiconductor based chemiresistive hydrogen sensors” (2024). https://www.sciencedirect.com/science/article/abs/pii/S0924424724004345
28. Alphasense, H2-AF hydrogen sensor. https://www.alphasense.com/products/view-by-target-gas/hydrogen-h2
29. “High-Performance Nanostructured Palladium-Based Hydrogen Sensors—Current Limitations and Strategies for Their Mitigation,” ACS Sensors (2020/2021). https://pubs.acs.org/doi/10.1021/acssensors.0c02019
30. “Recent Advances in Palladium Nanoparticles-Based Hydrogen Sensors for Leak Detection.” https://pmc.ncbi.nlm.nih.gov/articles/PMC6832907/
31. “Tilted Fiber Bragg Grating Sensor Using Chemical Plating of a Palladium Membrane for the Detection of Hydrogen Leakage,” Sensors 18(12), 4478 (MDPI, 2018). https://www.mdpi.com/1424-8220/18/12/4478
32. “Palladium-based optical fiber Bragg grating hydrogen sensors: A comprehensive review,” Opt. Laser Technol. (2024). https://www.sciencedirect.com/science/article/abs/pii/S0030399224003086
33. “Optical Fiber Grating Hydrogen Sensors: A Review,” Sensors 17(3), 577 (MDPI, 2017). https://www.mdpi.com/1424-8220/17/3/577
34. Li, H. et al., “Raman vibrational spectral characteristics and quantitative analysis of H2 up to 400°C and 40 MPa,” J. Raman Spectrosc. 49 (2018). https://analyticalsciencejournals.onlinelibrary.wiley.com/doi/10.1002/jrs.5420
35. “Hydrogen Sensor Based on Tunable Diode Laser Absorption Spectroscopy,” Sensors 19(23), 5313 (MDPI, 2019). https://www.mdpi.com/1424-8220/19/23/5313
36. Novelli, P.C., Lang, P.M., Masarie, K.A., Hurst, D.F., Myers, R., Elkins, J.W., “Molecular hydrogen in the troposphere: Global distribution and budget,” J. Geophys. Res. Atmos. 104(D23), 30427–30444 (1999). DOI: 10.1029/1999JD900788. https://agupubs.onlinelibrary.wiley.com/doi/abs/10.1029/1999JD900788
37. Barnes, D.H. et al., “Hydrogen in the atmosphere: Observations above a forest canopy in a polluted environment,” J. Geophys. Res. Atmos. (2003). https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2001JD001199
38. Praxair, “Hydrogen Compressed Gas” Safety Data Sheet (LEL 4%, UEL 75%). https://dept.harpercollege.edu/chemistry/msds/Hydrogen%20Compressed%20Gas%20Praxair.pdf
39. Industrial Monitor Direct, “Hydrogen Gas LEL Limits and Exposure Safety for Industrial Facilities” (comparative ignition-energy/burning-velocity figures). https://industrialmonitordirect.com/blogs/knowledgebase/hydrogen-gas-lel-limits-and-exposure-safety-for-industrial-facilities
40. Winsen Sensor knowledge base on catalytic-bead sensors (see ref. 25); SGX Sensortech, “Pellistor Application Note 6 — Pellistor Poisoning.” https://www.sgxsensortech.com/content/uploads/2014/08/AN6-%E2%80%93-Pellistor-Poisoning.pdf
41. SGX Sensortech, “Pellistor Application Note 6 — Pellistor Poisoning” (see ref. 40).
42. SGX Sensortech, “TC-1326 MEMS - Thermal Conductivity Sensor Elements.” https://www.sgxsensortech.com/products-services/industrial-safety/tc-1326-mems-thermal-conductivity-sensor-elements/
43. Winsen Sensor, “Electrochemical Hydrogen Sensors: Principles, Applications, and Advantages.” https://www.winsen-sensor.com/knowledge/electrochemical-hydrogen-sensors.html
44. Mainson, M., Heath, C., Pejcic, B., Frery, E., “Sensing Hydrogen Seeps in the Subsurface for Natural Hydrogen Exploration,” Applied Sciences 12(13), 6383 (MDPI, 2022). DOI: 10.3390/app12136383. https://www.mdpi.com/2076-3417/12/13/6383
45. “In-soil hydrogen concentration monitoring using MONHyTOR: application for natural hydrogen exploration,” Adv. Geosci. (ADGEO) 67, 57 (2025). https://adgeo.copernicus.org/articles/67/57/2025/
46. “Assessing the Portable Gas Analyzer GA5000 for Accurate H2 Monitoring in Geological Exploration” (2024/2025). https://hal.science/hal-04861335
47. “A High-Precision Hydrogen Sensor Array Based on Pt-Modified SnO2 for Suppressing Humidity and Oxygen Interference,” Chemosensors 13(8), 294 (MDPI). DOI: 10.3390/chemosensors13080294
48. H2scan, “HY-OPTIMA 700B Series Process Hydrogen Analyzer.” https://h2scan.com/product/hy-optima-700b-series-process-hydrogen-analyzer/
49. H2scan, “Solid-State Area Hydrogen Monitor HY-ALERTA 5021.” https://h2scan.com/product/hyalerta-hydrogen-monitor/
50. “Effects of poisoning gases on and restoration of PdCuSi metallic glass in a capacitive MEMS hydrogen sensor.” https://www.sciencedirect.com/science/article/abs/pii/S0360319919341527
51. Endress+Hauser, “How to measure hydrogen in natural gas with Raman spectroscopy.” https://endressprocessautomation.com/how-to-measure-hydrogen-in-natural-gas-with-raman-spectroscopy/
52. NETL (U.S. DOE), “Raman Gas Analyzer” fact sheet. https://netl.doe.gov/sites/default/files/rdfactsheet/Fact%20Sheet%20R-D106,%20June%202022,%20Raman%20Gas%20Analyzer.pdf
53. “Demystifying Raman analyzers: A comprehensive guide,” Swagelok / PI Process Instrumentation. https://www.piprocessinstrumentation.com/instrumentation/article/53098867/swagelok-demystifying-raman-analyzers-a-comprehensive-guide
54. “Electrochemical Sensor for Hydrogen Leakage Detection at Room Temperature.” https://pmc.ncbi.nlm.nih.gov/articles/PMC11723446/ (does not support the claim)
55. “Nanoscale Analysis of Sulfur Poisoning Effects on Hydrogen Sorption in Single Pd Nanoparticles” (2025). https://pmc.ncbi.nlm.nih.gov/articles/PMC12574217/
56. “Hydrogen Sensors Based on Pd-Based Materials: A Review,” Sensors 25(11), 3402 (MDPI). https://www.mdpi.com/1424-8220/25/11/3402
57. “Natural H2 exploration: tools and workflows to characterize a play,” Sci. Technol. Energy Transition (STET) (2023). https://www.stet-review.org/articles/stet/full_html/2023/01/stet20230116/stet20230116.html
58. “Sensing Electronics for Downhole Drilling in Extreme Environments.” https://labpartnering.org/technologies/3f72efda-2142-450a-aa56-547d82b5ed4b
59. “Downhole Electronic Components: Achieving Performance Reliability,” JPT (SPE). https://jpt.spe.org/downhole-electronic-components-achieving-performance-reliability