Fiber-optic & chemochromic hydrogen sensors

Fiber-optic hydrogen sensors do not use electricity at the measurement point at all. A thin coating of palladium metal (or a palladium/platinum-catalyzed tungsten-oxide film) is applied to an optical fiber, and when hydrogen touches that coating it either swells the metal — because hydrogen atoms squeeze into gaps in palladium’s crystal lattice — or changes the coating’s color and opacity. Either effect nudges the light traveling through the fiber: it shifts the wavelength that bounces back, bends an interference pattern, or dims the light that gets through, and that optical change is read out at the surface, potentially kilometers away, by an instrument that never has to send power downhole.

A closely related, even simpler family — chemochromic tapes and paints — uses the same palladium/tungsten-oxide chemistry but is read by eye or camera instead of by an interrogator: a hydrogen leak turns a patch of tape from gray to black.

The appeal for geologic hydrogen work is structural: no downhole batteries or electronics, no ignition risk, and no vulnerability to the electrical noise endemic to a working wellsite. Today’s fiber-optic hydrogen sensors are mostly laboratory demonstrations rather than field-hardened products, they are compromised by the same hydrogen sulfide that so often accompanies natural hydrogen, and, in a problem unique to this technology, hydrogen itself can damage the ordinary glass fiber used to carry the signal, independent of whatever coating is meant to detect it.

Mechanism & physics

Fiber-optic hydrogen sensing covers a family of optical read-out schemes — grating wavelength, interferometric phase, reflected/transmitted intensity — driven by a common chemical front end: a thin film of palladium (Pd), a Pd alloy (Pd-Ag, Pd-Ni, Pd-Au, Pd-Hf, Pd-Pt), or a catalyzed tungsten oxide (WO₃) that undergoes a large, reversible, concentration-dependent change in mechanical strain, refractive index, or optical absorption. The fiber converts that chemical-film state into a wavelength shift, phase shift, or intensity change readable remotely with no electricity at the sensing point [1][2].

The field traces to Butler’s 1984 Pd-coated Mach–Zehnder fiber interferometer [1], extended mechanistically by Butler & Ginley in 1988 [2], and carried into gratings by Sutapun, Tabib-Azar & Kazemi’s 1999 demonstration of two Pd-coated FBGs multiplexed on one fiber [9], the basis for the distributed hydrogen-FBG concepts discussed below.

The Pd–H chemistry that drives almost everything

Hydrogen molecules dissociate catalytically on the Pd surface into atomic hydrogen, which dissolves interstitially into the face-centered-cubic Pd lattice. At low hydrogen chemical potential this forms a dilute solid solution, the α-phase (H/Pd atomic ratio x ≲ 0.03). Sieverts’ law describes the equilibrium solubility of a diatomic gas that dissociates on the metal surface:

$c_H = K_S \sqrt{p_{H_2}}$

— dissolved hydrogen concentration is proportional to the square root of H₂ partial pressure, with $K_S$ the temperature-dependent Sieverts constant. This √p dependence is why Pd-film optical sensors are characteristically nonlinear in H₂ concentration, and several fiber Fabry–Pérot studies fit a $\sqrt{c_{H_2}}$ response law [5]. Diffusive membrane flux is likewise Sieverts-law-limited when the exponent n ≈ 0.5, the regime for thick Pd membranes/foils/tubes rather than thin evaporated films [6].

As hydrogen chemical potential rises further, Pd undergoes a first-order phase transition into a concentrated β-hydride phase (x ≳ 0.6), with an α+β coexistence region between. X-ray diffraction shows the Pd unit cell expanding by roughly 4.3% (linear, 30 °C) on full hydriding [3], corresponding to a volumetric expansion widely cited as ~10% for α→β [4].

This lattice dilation — mechanical, not electronic — is the direct origin of every strain-based fiber H₂ sensor (FBG, tilted FBG, most MZI and FP designs), and it also drives two consequences that dominate long-term engineering: hysteresis (absorption/desorption branches of the isotherm don’t coincide, propagating into hysteretic wavelength-vs-concentration curves) and mechanical fatigue — the repeated ~10% volumetric swing drives dislocation glide and grain-boundary motion until the film blisters, cracks, or delaminates [4].

Diffusion and the response-time number

Hydrogen transport within a Pd film is Fickian at the sub-second scale, with a measured diffusion coefficient for α-phase Pd thin films of D ≈ 1.3 × 10⁻⁷ cm²/s at room temperature [7]. The natural time constant $\tau_{diff} \sim L^2/D$ predicts sub-millisecond diffusion times for a 100 nm sputtered film, yet measured sensor response times run seconds to many minutes; response is almost never diffusion-limited in real devices. It is instead governed by:

• dissociative-chemisorption kinetics at the Pd surface (poisoned/slowed by O₂, H₂O, CO, H₂S — see below)
• nucleation/growth kinetics of the β-hydride phase
• mechanical relaxation of the strain into the glass

Thick Pd tubes show response times of >200 minutes at 23 °C but <2 minutes at 95 °C for 4% H₂ [8]; this reflects Arrhenius-accelerated surface kinetics layered on the diffusion scaling.

FBG strain transduction

A fiber Bragg grating reflects light at the Bragg wavelength $\lambda_B = 2\, n_{eff}\, \Lambda$, where $n_{eff}$ is the fiber core effective index and $\Lambda$ the grating period. Both are perturbed by axial strain $\varepsilon$ and temperature change $\Delta T$:

$\frac{\Delta \lambda_B}{\lambda_B} = (1 - p_e)\,\varepsilon + (\alpha_\Lambda + \alpha_n)\,\Delta T$

with $p_e$ the effective photo-elastic coefficient of silica (~0.22), $\alpha_\Lambda$ the fiber’s thermal expansion coefficient, and $\alpha_n$ its thermo-optic coefficient. In a Pd-coated FBG, hydride-induced lattice dilation strains the underlying fiber via the coating bond, giving $\varepsilon = \varepsilon(c_{H_2})$ — the grating converts a chemical concentration into strain, and strain into wavelength shift.

Because the same equation carries an explicit, uncorrelated $\Delta T$ term, every bare Pd-FBG is intrinsically temperature-cross-sensitive. Reported raw sensitivities span ~8–90 pm per % H₂ across Pd/Pd-alloy compositions and geometries, with etched/side-polished/tapered designs reaching 0.3–0.5 pm per ppm in dissolved-gas (transformer oil) applications by concentrating the evanescent field [8][9][10][11][12][13].

Temperature cross-talk and compensation

Because a single Pd-FBG cannot distinguish a hydriding-strain shift from an ambient-temperature one, three compensation architectures recur:

• a dual/reference grating (an uncoated FBG co-located with the sensing FBG measures $\Delta T$ alone, subtracted algebraically)
• differential thermal-expansion substrates (gratings bonded to two different-CTE materials so a common $\Delta T$ gives a predictable, subtractable differential shift)
• self-compensating composite coatings (e.g., Pd/Au tilted-FBG stacks engineered so the net thermo-mechanical response cancels to first order) [14]

Technica’s T910 twin-FBG hydrogen sensor markets itself as “self-temperature compensated” using a customized twin-FBG transducer rather than a single grating [15][16]. Despite these schemes, no quantitative, full T/P/RH-compensated field accuracy spec for any Pd-FBG product was found; this is flagged as an open question below.

Interferometric transduction (Fabry–Pérot, Mach–Zehnder)

A fiber-tip Fabry–Pérot (FP) cavity forms between two reflecting interfaces separated by cavity length $L$ (e.g., a cleaved fiber end and a suspended or bonded Pd membrane); resonance occurs when $2nL = m\lambda$. Hydride-induced expansion or bending changes $L$ directly, so the best FP designs report the highest sensitivities in this survey: up to 30.3 pm/ppm over 0–100 ppm H₂ with a fiber-tip graphene–Au–Pd cantilever [17], and 3.6 pm/ppm with a ~3.3 ppm calculated LOD with a suspended 30 nm Pd membrane [5].

A Mach–Zehnder interferometer (MZI) splits light into a Pd-coated sensing arm and an unperturbed reference arm (or, in-fiber, couples core and cladding modes across a tapered waist); the phase difference $\Delta\phi = \frac{2\pi}{\lambda}\,\Delta(nL)$ changes as the coating’s effective index responds to hydrogen. Butler’s original 1984 sensor was of this lineage: a 10 μm Pd film on the signal arm of an MZI, sensitive over roughly 0.002–2000 Pa H₂ partial pressure [1][2].

Modern in-fiber MZI variants using polymer-infiltrated, Pt-loaded WO₃ coatings report the largest raw wavelength-shift-per-concentration numbers in this review — up to ~1949 nm per volume-% H₂ over 0–0.8% — though this reflects a highly nonlinear, narrow-range response rather than a ppm-level LOD claim [18].

Evanescent-field transduction

Side-polished, D-shaped, and tapered fibers remove part of the cladding so the evanescent tail of the guided mode directly samples an external Pd (or Pd/WO₃) film. Hydriding changes the film’s complex refractive index $\tilde n = n + ik$, shifting the effective index seen by the mode and hence the local Bragg condition (evanescent-FBG hybrids) or the propagation loss (evanescent-absorption designs) — the same lattice-strain/complex-index physics above, applied where field–matter overlap is maximized, not a distinct mechanism.

Gasochromic WO₃: a different, non-strain mechanism

WO₃ gasochromics do not rely on Pd lattice strain. H₂ dissociates on a Pt or Pd catalyst dispersed on/in the WO₃ film (“hydrogen spillover”), and the atomic hydrogen intercalates into the WO₃ lattice as a small polaron, forming a hydrogen tungsten bronze:

$WO_3 + xH^+ + xe^- \rightarrow H_xWO_3$

Intervalence charge transfer between W⁶⁺ and the newly formed W⁵⁺ sites produces a strong, broad optical absorption from the visible into the near-IR, turning the film from pale to blue-black — read as decreased transmittance or increased extinction coefficient. Unlike Pd-strain sensors, this is intrinsically reversible on O₂/air exposure, which re-oxidizes W⁵⁺ back to W⁶⁺ [19][20][21].

Because the mechanism is chemical rather than mechanical, gasochromic WO₃ avoids Pd’s blistering/delamination fatigue, but suffers a distinct aging pathway: water photolysis at the Pt/WO₃ interface produces water that poisons the catalyst over repeated cycles, degrading stability unless mitigated (e.g., periodic UV irradiation to decompose accumulated water) [22].

Passive chemochromic indicators

NASA/UCF-lineage chemochromic tapes and paints use hydrogen-sensitive pigments — tungsten and molybdenum oxo-compounds and heteropoly acids (WO₃, MoO₃, silico-tungstic, phospho-tungstic, phospho-molybdic acids) — combined with nanoscale Pd/Pt activator particles in a gas-permeable binder (silicone, paint, ink, tape) [23][24]. The chemistry is the same spillover/intercalation as gasochromic WO₃ above, but read by eye or camera, and engineered along a reversible ↔ irreversible spectrum: reversible formulations re-oxidize in air (repeatable use), while irreversible formulations stay dark permanently once triggered, acting as a “witness mark” for post-incident inspection [25][26].

At a glance — published performance envelope

Detection limit

3–5,000 ppm

Range

3–50,000 ppm

Response t₉₀

0.34–12,000 s

Recovery

0.5–12,000 s

Selectivity

high

Drift

medium

Operating T

−54 … 232 °C

Humidity

0–98 % RH

Power

0 mW – 15 W

Unit cost

$130–$50k

TRL

4

Main interferents: O₂, relative humidity, temperature, total pressure · poisons: H₂S, water, SO₂. Ranges span the published devices/variants covered below — per-claim citations are in the body; data: src/data/modalities/fiber-optic-sensors.yaml.

Variants & subtypes

#	Family	Transduction	Distinguishing physics
1	Pd-coated fiber Bragg grating (FBG)	Wavelength shift	Pd/Pd-alloy film hydrides, lattice dilates (~10% Δvolume in β-phase [4]), strains the underlying grating; $\Delta\lambda_B$ scales with hydride-induced strain [8][9].
2	Tilted FBG (TFBG) / long-period grating (LPG)	Wavelength shift + cladding-mode coupling	Cladding-mode resonances are far more evanescently exposed than core modes, giving stronger overlap with a thin Pd or Pt-SiO₂ coating; some designs achieve passive temperature self-compensation via composite coatings [14][27][28].
3	Pd micromirror (Butler lineage)	Reflected intensity	Thin Pd film evaporated on a cleaved multimode fiber end acts as a mirror whose reflectivity/absorption changes as the film hydrides; the simplest fiber H₂ optrode [1][29].
4	Evanescent-field: tapered fiber	Intensity / index change	Fiber pulled to a micron-scale waist so the guided mode’s evanescent field extends into a Pd or Pd-nanocomposite coating over an extended length, raising sensitivity per unit H₂ concentration at the cost of fiber fragility [30][31][32].
5	Evanescent-field: side-polished / D-shaped fiber	Wavelength shift (if combined with FBG) or intensity	Cladding mechanically polished near the core to expose an evanescent “window,” frequently combined with an FBG beneath for both strong evanescent overlap and narrowband read-out — the highest dissolved-gas (transformer oil) sensitivities in this survey come from this class (up to 0.477 pm per μL/L) [33][34][35].
6	Evanescent-field: Pd/WO₃-supported	Intensity	Pd nanoparticles catalyze H₂ dissociation and spill hydrogen onto an adjacent/matrix WO₃, combining Pd’s kinetics with WO₃’s optical contrast; response/recovery reported as fast as ~10 s [36].
7	Interferometric: fiber Fabry–Pérot (FP)	Phase / fringe shift	Pd film or suspended Pd/graphene membrane forms one mirror of a miniature FP cavity; cavity-length or membrane-curvature change with hydriding gives the highest reported ppm-level sensitivities in this survey [5][17][37][38].
8	Interferometric: fiber Mach–Zehnder (MZI)	Phase / spectral shift	Pd- or Pt/WO₃-coated sensing arm (or in-fiber tapered/core-cladding-mode-coupling structure) accumulates a differential optical path length relative to a reference arm/mode [1][18][39][40][41].
9	Gasochromic WO₃ (thin-film, waveguide, or fiber-coupled)	Optical density / transmittance	Pt/Pd-catalyzed hydrogen spillover into WO₃ forms a tungsten bronze with strong visible/NIR absorption; reversible on air exposure [19][20][21][42].
10	Passive chemochromic indicators (tape, paint, pigment)	Human/camera-read color change	Same spillover/intercalation chemistry as (9), formulated into a gas-permeable, no-power, no-instrumentation consumable; reversible or irreversible variants exist [23][25][43][44].
11	Distributed/multiplexed FBG arrays (WDM/TDM)	Many wavelength shifts on one fiber	Multiple discrete Pd-FBGs (each at a unique Bragg wavelength) or a dense ultra-weak FBG (UFBG) array are multiplexed along a single fiber and interrogated by wavelength or time gating, in principle giving a spatial hydrogen profile along a wellbore or pipeline [9][45][46].

Performance envelope

All rows are demonstrated laboratory (or, where noted, transformer-industry field) results, not necessarily achieved simultaneously by a single device. Values are as reported in the cited source; ranges reflect the spread across designs within a subtype.

Sensor type	Sensitivity	LOD	Response time	Recovery time	Conditions / notes	Ref
Pd-FBG (pure Pd film, various thickness)	8.4–81.8 pm/% H₂	~500 ppm–1% (device-dependent)	tens of s to >200 min	comparable to response	0.1–3% H₂ range; Arrhenius-accelerated (200 min @23 °C → <2 min @95 °C for 4% H₂ on Pd tube)	[8]
Pd-alloy FBG (Pd-Ni, Pd-Ag, Pd-Hf)	10–47 pm/% H₂ (up to 700 pm at 35 °C for one Pd/Ag design)	not separately stated	seconds–minutes	minutes	Alloying reduces hysteresis/deactivation vs. pure Pd	[8][47]
Etched/thin-cladding Pd-FBG	>20 pm/%(v/v), 0–1% H₂	sub-%	not stated	not stated	Applied Optics 2015 etched-grating design	[10]
Pt-loaded WO₃-coated FBG	536 pm shift at 10,000 ppm (25 °C)	50 ppm (optimized)	tens of s	tens of s	Combines WO₃ optical contrast with FBG read-out	[12]
Side-polished/D-shaped FBG (Pd/Ag), dissolved-gas	0.477 pm/(μL/L) — ~11.4× a plain FBG	ppm-level dissolved H₂	~30 s (4% H₂) to ~2 min (1% H₂)	not separately stated	Power-transformer oil application; validated against conventional DGA	[33][34][35]
Tapered fiber + Pd nanoparticles	up to 91% response at 1000 ppm	<0.5 ppm (RT, one design) to ~40 ppm (another)	18.8 s (1000 ppm)	not stated	Sensitivity and stability trade off against taper fragility	[31][32]
Pd-supported WO₃ evanescent fiber	up to 250% response @0.7% H₂	not separately stated	~10 s	~10 s	Fast Pd-catalyzed WO₃ evanescent design	[36]
Fiber Fabry–Pérot, bulk/foil Pd	1.44 nm/%	500 ppm (resolution)	185 s	comparable	Compact FP design, FBG+nanofilm variant	[37][48]
Fiber-tip FP, graphene–Au–Pd cantilever	30.3 pm/ppm (0–100 ppm)	sub-ppm implied by slope	fast (cantilever resonance shift)	fast	Highest ppm-sensitivity found in this review	[17]
Fiber-tip FP, suspended 30 nm Pd membrane	3.6 pm/ppm (@500 ppm), 1.5 pm/ppm (@0.5%)	~3.3 ppm (calculated from noise floor)	~11 min (t₉₀, 0.5% H₂)	~comparable	Best directly-stated ppm LOD for a Pd-only FP design	[5]
Fiber Mach–Zehnder, pure Pd film	sensitivity ~2× a comparison taper design	not stated	not stated	not stated	0–3.6% H₂ range	[39][40]
In-fiber MZI, Pt-WO₃/polymer	up to ~1949 nm/vol% (0–0.8% H₂)	not separately stated as ppm LOD	not stated	not stated	Very high raw sensitivity, narrow/nonlinear range	[18]
WO₃ gasochromic thin film (Pd overlayer, optimized anneal)	Sensor response 6.3 @25 ppm, 10.2 @1000 ppm	25 ppm (best demonstrated)	8.5 min (25 ppm) → 3.5 min (1000 ppm)	5 min (25 ppm) → 2.5 min (1000 ppm)	25 ppm cited as 40–400× lower than typical (0.1–1%) prior WO₃ gasochromic LODs	[20]
WO₃-PdPt-Pt fiber-coupled nanocomposite	not separately stated as pm/ppm	3 ppm	0.34–0.44 s at 4,000–10,000 ppm	fast; hysteresis grows above ~4000 ppm	Uses standard photodetector, not spectrometer, for lower system cost	[49]
WO₃-Pd₂Pt-Pt nanocomposite fiber sensor	—	—	tens of s (5–12.5 ppm range across generations)	—	Predecessor designs to the above	[50][51]
Chemochromic tape (DetecTape)	visual: gray→black	vendor states detection at concentrated leaks; quantitative ppm threshold not published	”changes color within minutes”	non-reversible (single-use witness)	−54 °C to +232 °C operating range; $129–185 per roll (1”/2”, 15 ft)	[43][52]
Chemochromic paint/pigment (NASA/Element One-class formulations)	visual color change	as low as 0.04% (400 ppm) = 1/100 LFL cited by one formulation	not separately quantified (general “rapid” claims)	reversible or irreversible depending on formulation	Multiple format options: paint, tape, caulk, molded parts, textiles	[44][25][26]
Twin-FBG catalytic-combustion optrode (Technica T910)	not published	not published	”within seconds” (vendor claim)	not published	Multiplexable/daisy-chainable; EMI/lightning immune; self-temperature-compensated (vendor claims)	[15]

DOE benchmark context

For orientation, the U.S. DOE’s long-standing fuel-cell hydrogen-sensor targets call for roughly ≤1 s response time and detection limits below ~1000 ppm (0.1%) (program targets cited across multiple secondary sources — treat as indicative rather than a directly verified primary spec).

Judged against that bar, the best fiber-optic point sensors (graphene/Pd FP designs, WO₃-PdPt-Pt fiber-coupled devices) now clear the ppm sensitivity bar but very few clear the 1-second response bar simultaneously — there is a persistent sensitivity/speed trade-off across this whole technology family. For cross-technology orientation more broadly, Hübert et al.’s widely cited review of hydrogen-sensor operating principles remains the standard benchmark against which any fiber-optic figure of merit should be read [79].

Cost class. Chemochromic tape is a low-cost, disposable consumable (~$130–185 per 15 ft roll, no instrumentation) [43].

Point Pd-FBG or Pd-FP sensor heads themselves are inexpensive optical components once fabricated, but a usable system requires an optical interrogator (broadband source + spectrometer or tunable laser + photodetector); industrial multi-channel FBG interrogators are marketed as field-proven units but public list pricing was not found in this survey (unverified) — qualitatively, this places fiber-optic point systems in a moderate-to-high capital-cost class dominated by the interrogator, not the sensing head, which is precisely the argument for the multiplexing/distributed-sensing value proposition (see row 11 above).

Figures of merit

The universal figures of merit — sensitivity, LOD, selectivity, response time, dynamic range, drift, power, operating envelope — are defined in detail on the Figures of merit page; the table below specializes each one to fiber-optic/chemochromic transduction, where the raw “signal” is a wavelength, phase, or intensity shift rather than a current or resistance, plus the modality-specific FoMs (Bragg-shift sensitivity, temperature cross-sensitivity, multiplexing capacity, intrinsic safety) that don’t exist for any electrical sensor. Concentrations below follow the ppm/% conventions on Concentration units.

Figures-of-merit table for this modality (11 rows)

Figure of merit	Symbol / formula	What it means here	Typical value for this technology	How to read it
Sensitivity (Bragg-wavelength shift)	$\Delta\lambda_B/\Delta c_{H_2}$ (pm per % H₂), from $\lambda_B=2n_{eff}\Lambda$	How far the reflected Bragg line moves per unit of hydride-induced strain — the wavelength-domain specialization of the universal sensitivity slope $S$	8.4–81.8 pm/% H₂ for pure-Pd-film FBGs; 10–47 pm/% H₂ for Pd-alloy FBGs, with one optimized Pd/Ag design reaching ~700 pm at 35 °C; interferometric (FP) designs report far higher per-ppm slopes (up to 30.3 pm/ppm for a graphene–Au–Pd cantilever) but only over narrow, low-concentration ranges [8][17][47]	Compare only within one transduction family and coating thickness — pm/% (FBG) is not directly comparable to pm/ppm (FP) or nm/vol% (MZI); this unit mismatch is the sensitivity-units problem the definitions page warns against
Temperature cross-sensitivity & compensation	$(\alpha_\Lambda+\alpha_n)\,\Delta T$ term in $\Delta\lambda_B/\lambda_B$	A bare Pd-FBG cannot distinguish a hydride-strain shift from an ambient-temperature shift, and temperature can multiplicatively rescale the H₂ sensitivity itself, not just add an offset	No single quantitative pm/°C spec for a Pd-FBG was found in this survey (unverified); documented magnitude of the problem: room-temperature H₂ sensitivity ~2.5× cryogenic sensitivity, and one Pd/Ag design’s sensitivity jumped >10× (47 pm → 700 pm) between 25 °C and 35 °C [8]	Treat a single-grating spec as valid only at its calibration temperature; look for a dual/reference grating, a differential-CTE substrate, or a self-compensating composite coating (e.g., Technica’s “self-temperature-compensated” T910 twin-FBG) before trusting a reading across a temperature-varying wellbore [14][15][16]
Limit of detection (LOD)	$k\,\sigma_{blank}/S$ (see definitions page)	Smallest H₂ concentration this transduction scheme can distinguish from noise	~500 ppm–1% for basic Pd-FBGs; single-digit ppm for the best designs (3.3 ppm calculated, suspended-Pd-membrane FP [5]; 3 ppm demonstrated, WO₃-PdPt-Pt fiber-coupled evanescent design [49]); 25–50 ppm for optimized gasochromic-WO₃ and Pt-WO₃-FBG designs [12][20]	None of these ppm-level numbers states its averaging time or noise convention (3σ vs. SNR = 1) — per the definitions page’s LOD caveat, treat a bare ppm figure as unverified until integration time is known, especially since response times for the same designs span seconds to >200 minutes
Selectivity / cross-sensitivity	$K_{ij}=S_j/S_i$	Pd’s native chemical selectivity for H₂ over the CH₄/CO₂/CO/He that commonly co-evolve with geologic hydrogen, and its one serious weakness	~2–3 orders-of-magnitude rejection of CO/CO₂/CH₄ (10.1% response to H₂ vs. 0.1% CO, 0.05% CO₂, ≈0.03% CH₄ in one comparison) [55]; He rejection good in non-fiber Pd devices (unverified for fiber specifically); H₂S is a documented catalytic poison, with response and response time both heavily degraded after 61 days of sulfide exposure in one study [56]	The CO/CH₄/CO₂ rejection is genuinely strong and favors Pd/fiber routes for geologic seep-gas mixtures, but selectivity is only as good as the worst interferent (per the definitions page), and in sour geologic H₂ systems that interferent is H₂S, which this technology has not been shown to tolerate over multi-month field deployment
Response / recovery time	$\tau_{diff}\sim L^2/D$, $D\approx1.3\times10^{-7}\ \text{cm}^2/\text{s}$ [7]	Naive Fickian diffusion through a thin Pd film predicts sub-millisecond response; real devices run seconds to many minutes because response is actually governed by surface chemisorption kinetics (poisoned/slowed by O₂, H₂O, CO, H₂S), not by diffusion	Seconds to >200 minutes across the Pd-FBG family, strongly Arrhenius-accelerated with temperature (>200 min at 23 °C → <2 min at 95 °C for 4% H₂ on a Pd tube) [8]; fastest designs (WO₃-PdPt-Pt fiber-coupled, Pd/WO₃ evanescent) report sub-second to ~10 s response [49][36]	Because response is kinetics-, not diffusion-, limited, a thinner film or hotter operating point genuinely speeds response, but at the cost of accelerated hydride-cycling fatigue (see drift/stability row) and lower sensitivity at high T
Dynamic range	$N_{\text{decades}}=\log_{10}(C_{\max}/C_{\min})$	Concentration span over which a given device produces usable output	Wide variation by design: Butler’s original Pd/MZI spanned roughly 0.002–2000 Pa H₂ partial pressure (~6 decades) [1][2]; most FBG designs are validated only over ~0.1–3% H₂ (~1.5 decades) [8]; the highest-sensitivity FP cantilever design is characterized only over 0–100 ppm [17]	A wide range from one design doesn’t transfer to another family — check that the quoted range is the same one over which sensitivity/LOD were measured, since the Sieverts’-law $\sqrt{p_{H_2}}$ nonlinearity means a single calibration slope rarely holds across a device’s full range
Drift / long-term stability	Cyclic “deactivation” (%), hysteresis width	How much the calibration slope and baseline degrade with repeated hydride cycling (mechanical fatigue) and contaminant exposure (chemical poisoning)	Pure-Pd films show ~7% deactivation after cycling vs. <1% for Pd-Au alloys [58]; compliant/flexible substrates avoid buckling/blistering through 20 cycles where rigid substrates fail [13][4]; no quantitative field drift-rate (%/yr) or full T/P/RH-compensated accuracy spec for any Pd-FBG product was found (unverified)	Alloying and compliant-substrate engineering measurably reduce the ~10% volumetric-swing fatigue that is otherwise structural to any Pd-strain sensor — but with no published field drift-rate number, budget for periodic recalibration rather than trusting a “maintenance-free” claim
Power	$P_{\text{head}} + P_{\text{interrogator}}$	The sensing element itself draws no electrical power; all power is spent at the surface instrument reading it out	0 W at the sensing head; interrogator power not separately quantified in this survey, but broadly a 0–15 W class instrument (broadband source + spectrometer, or tunable laser + photodetector)	Fiber optics does not shrink the LOD–response time–power triangle; it relocates the power cost off the downhole sensing point and onto a serviceable surface instrument
Operating envelope	$C_{\text{reported}}=C_{\text{raw}}\cdot f_T(T)\cdot f_{RH}(RH)\cdot f_P(P)$	Distinguishes where the fiber merely survives from where its H₂ reading is actually trustworthy	Individual lab studies span roughly −50 °C to 95 °C [8]; chemochromic tape’s operating range is the widest in this family at −54 °C to +232 °C [43], but its output is qualitative, not a calibrated ppm number; no full T/P/RH-compensated accuracy envelope has been published for any Pd-fiber product	An “operating” range is not a “compensated” one (per the definitions page) — nothing stops a Pd-FBG from producing a number outside its calibrated envelope, and the >10× sensitivity swing over just a 10 °C window (row 2) implies the true compensated envelope is narrower than the operating one
Multiplexing capacity	Gratings per fiber (WDM/TDM channel count)	How many discrete Pd-FBG sensing points one interrogator can read along a single fiber run — the feature that could turn a point sensor into a depth-resolved wellbore profile	2 Pd-coated FBGs multiplexed on one fiber in the field-founding 1999 demonstration [9]; WDM/TDM/ultra-weak-FBG-array infrastructure mature in structural-health-monitoring contexts supports far larger counts [45][46][66], but no field-demonstrated multiplexed/distributed FBG hydrogen profile in an actual borehole or pipeline was found (unverified beyond the 2-grating lab demo)	The multiplexing math is proven for strain/temperature/acoustic sensing generally — the open question is whether channel count and spacing hold up once every sensing grating also needs its own temperature-reference grating, roughly doubling the practical channel budget
Intrinsic-safety advantage	n/a (structural/qualitative)	An all-optical, unpowered sensing element cannot itself supply the electrical or thermal energy needed to ignite an H₂ (LEL 4 vol%) atmosphere, and is immune to wellsite EMI	Qualitative/binary: zero electronics at the measurement point vs. a powered transducer required by every electrochemical, catalytic-bead, or MOX alternative; Technica markets its T910 explicitly around EMI/lightning immunity [15]	This is the one figure of merit fiber optics wins outright; it is the primary reason this technology stays in the downhole toolkit despite the maturity gap on the rows above

What dominates. For a geologic-H₂ downhole job, the gating figures of merit are temperature cross-sensitivity (row 2) and long-term stability under H₂S exposure and hydride-cycling fatigue (row 7), not sensitivity or LOD; both sensitivity and LOD are already good enough on paper, with single-digit-ppm designs demonstrated in the lab [5][49]. The trade-off is the classic sensitivity–selectivity–stability triangle: the same exposed, reactive Pd surface sites that give strong native selectivity over CH₄/CO₂/CO are the sites most vulnerable to H₂S poisoning and to the ~10% volumetric hydride swing that blisters the film, and no published design has demonstrated both a compensated temperature response and multi-month H₂S tolerance in one packaged product. Fiber optics wins the power/intrinsic-safety corner of the LOD–response time–power triangle outright: zero downhole electricity is a structural advantage, which is why it remains in the toolkit despite the maturity gap on the other vertices.

Environmental covariance & cross-sensitivity

Temperature. Every Pd-FBG has an intrinsic, uncompensated temperature cross-term in $\Delta\lambda_B$, and hydriding sensitivity itself is temperature-dependent — one study found room-temperature sensitivity roughly 2.5× that at cryogenic temperature, recovering only 57% of it at −50 °C via localized IR-laser heating [8]. Another Pd/Ag micro-machined FBG design showed sensitivity jumping from 47 pm to 700 pm (>10×) going from 25 °C to 35 °C toward 4% H₂ [8]. Temperature can multiplicatively rescale the hydrogen sensitivity itself, a harder compensation problem than simple offset subtraction.

Relative humidity. Water vapor competes for the Pd surface sites needed for dissociative H₂ chemisorption. In one humidity study (source not independently pinned down in this survey), response magnitude was relatively insensitive to humidity, but response time lengthened by roughly 5× at 60 °C and 2.5× at 80 °C as RH rose — the effect shrinks at higher operating temperature because less water stays adsorbed. The qualitative rule across Pd-alloy thin-film studies: response magnitude falls with rising temperature, humidity, and O₂; response time falls with rising temperature and H₂ concentration. For WO₃ gasochromics, accumulated water from the H₂ + ½O₂ → H₂O by-product is itself a slow catalyst-poisoning mechanism (water photolysis, see Mechanism above) [22].

Oxygen. O₂ competes directly with H₂ for Pd surface sites. The same Pd sensor in H₂/N₂ vs. H₂/air showed response time rising from 3 s to 33 s and response magnitude dropping from 1.15% to 0.48% at 0.1% H₂/50 °C, attributed to competitive H₂/O₂ adsorption [54]. This is relevant wherever local O₂ fugacity varies with redox conditions, as downhole and in soil gas.

CO, CO₂, CH₄ — good native selectivity, not perfect. Pd is chemically far more reactive toward H₂ than toward these co-evolved gases: one comparison reported sensor response of 10.1% to H₂ vs. 0.1% to CO, 0.05% to CO₂, and ≈0.03% to CH₄ at comparable exposure — two-to-three orders of magnitude rejection [55]. A separate nanowire-array study (uncited here) found H₂ response ~90× larger than CH₄ response at 50 ppm. This is genuinely strong selectivity relative to most electrochemical/MOX competitors, and a major reason Pd/fiber routes suit CH₄/CO₂/He-bearing geologic H₂ seeps.

H₂S — the main weakness. Sulfur species are documented Pd catalytic poisons that block the sites needed for H₂ dissociation; in one poisoning study (uncited here), Pd sensor response and response time were both heavily degraded after 61 days of sulfide contamination, and single-Pd-nanoparticle studies confirm sulfur suppresses hydrogen sorption at the nanoscale [56]. Because geologic H₂ systems very commonly co-evolve H₂S, any Pd-based fiber sensor in a sour downhole or soil-gas environment should be assumed to drift or degrade over weeks-to-months unless a sacrificial filter, periodic thermal regeneration, or a sulfur-tolerant Pd alloy is used. “Repeatable thermal refreshing” reverses poisoning in resistive (non-fiber) Pd nanowire sensors and could in principle adapt to a heated downhole Pd-fiber optrode, but was not found demonstrated in a fiber-optic H₂ sensor [57].

Helium. He does not adsorb/dissociate on Pd; one resistive Pd device showed only a minor signal pit from He exposure — good rejection, but from a non-fiber device, offered here (unverified for fiber-optic Pd sensors specifically).

Pressure. No fiber-optic-specific pressure-compensation dataset was found. By extension from Sieverts’ law above, total system pressure at fixed H₂ mole fraction changes the H₂ partial pressure and therefore equilibrium hydride loading — a downhole Pd-fiber sensor calibrated at surface pressure will read differently downhole at the same H₂ mole fraction, independent of any mechanical pressure effect on the fiber. This is a first-principles inference, not a measured downhole calibration curve, and should be validated before deployment.

Aging, delamination, and blistering. The α↔β transition’s ~10% volumetric swing [4] causes cumulative mechanical damage each cycle; stress stabilizes somewhat after several cycles via grain-boundary motion, but films can still fail over the sensor’s service life [4]. Demonstrated mitigations: compliant/flexible substrates (avoided buckling/blistering through 20 cycles vs. rigid substrates) [13][4]; alloying (Pd-Au shows <1% “deactivation” after cycling vs. 7% for pure Pd) [58]; and polyimide adhesion layers between a thick Pd coating and the bare fiber, reducing blistering from poor Pd–silica adhesion [8].

A downhole-specific, non-obvious failure mode: hydrogen darkening of the fiber itself. Independent of any deliberate sensing coating, free molecular H₂ diffusing into ordinary silica-core telecom fiber reacts with the glass to form Si–OH groups, producing broadband IR loss known as hydrogen darkening [59][60]. At 1550 nm this can add up to ~1 dB/km (vs. ~0.2 dB/km baseline for standard single-mode fiber), cutting a distributed sensing system’s usable range/SNR roughly 5-fold [59]. This is a serious issue for any downhole DFOS system — DTS, DAS, or a wellbore FBG chain — in a naturally H₂-rich or H₂-storage well, entirely separate from whether the fiber is measuring hydrogen. Mitigations include hydrogen-resistant glass chemistries, hermetic (e.g., carbon) coatings that slow H₂ ingress, and dual-wavelength/dual-fiber self-calibration [59][60][61][62]. Field teams should account for this: an uncompensated distributed fiber run down a natural-hydrogen well can lose range and accuracy over its service life from the same gas it is meant to characterize.

Use for geologic H₂

Field verdict

Not competitive for portable soil-gas surveys today: no commercial or demonstrated fiber-optic soil-gas instrument exists, and electrochemical multi-gas analyzers are the established tool there. But for downhole/wellbore continuous monitoring, fiber-optic hydrogen sensing is the most structurally credible technology covered here, because it is the only route that puts zero electronics at depth. It is not yet a qualified product: ppm-level sensitivity, T/P compensation, and H₂S-tolerant long-term stability have each been demonstrated separately, but never together in one field-hardened device, and hydrogen darkening of the carrier fiber itself must be engineered around regardless of sensing chemistry chosen.

Soil-gas surveys (near-surface, ppm-to-% range, portable). Soil-gas H₂ surveys for natural hydrogen exploration overwhelmingly use portable electrochemical multi-gas analyzers (e.g., the GA5000/BIOGAS5000 class of instrument), not fiber optics [63][64][65]. No commercial or demonstrated fiber-optic soil-gas H₂ instrument was found. Fiber-optic point sensors could in principle offer better long-term drift and no consumable electrochemical cell, but interrogator hardware, cost, and the lack of a rugged, batteryless field-portable product currently rule fiber optics out of this niche. Chemochromic tape/paint has a narrow role as a zero-power qualitative screening aid (below), but nothing quantitative and portable exists.

Downhole/wellbore — the strongest application. This is where fiber-optic H₂ sensing has a real role in the toolkit, for structural reasons:

• No downhole electronics required. A passive Pd-FBG, Pd-mirror, or Pd/WO₃-coated fiber segment needs no power, battery, or electronics package at depth — only glass fiber and a coating. This is the main structural advantage over any electrochemical, catalytic-bead, or semiconductor (MOX) H₂ sensor, all of which need a powered transducer at the measurement point (compare MOX chemiresistors and electrochemical cells).
• Intrinsic safety and EMI immunity. An all-optical, unpowered element cannot itself ignite an H₂ (LEL 4 vol%) atmosphere, and it is immune to the electrical noise endemic to a working wellsite (pumps, ESPs, wireline trucks, cathodic protection) — Technica markets its FBG sensor around exactly this EMI/lightning immunity [15].
• Kilometer reach and multiplexability on one fiber. Standard telecom fiber supports km-scale runs with acceptable loss, and WDM/TDM techniques already mature for structural-health monitoring [45][46][66] are directly transferable to a chain of Pd-FBGs along a wellbore, in principle giving a depth-resolved hydrogen profile from one wellhead interrogator. However, no field-demonstrated multiplexed/distributed FBG hydrogen profiling in an actual borehole, well, or pipeline was found. The closest analogs are NETL’s subsurface wellbore-integrity program, which targets hydrogen/natural-gas storage wells but reports pH/corrosion/strain/temperature sensing rather than a validated H₂ profile [67][68][69], and the 2001 IOS/Boeing/NASA Stennis multi-point fiber-optic leak-detection demonstration on a Delta IV static-fire test — real, but a launch-pad application, not a borehole [70][71].
• The hydrogen-darkening caveat above is a first-order downhole risk that must be engineered around regardless of the sensing chemistry chosen.
• Sensitivity is still maturing relative to the exploration target. Anomalies of interest start around ~1 ppm above the ~0.53 ppm atmospheric background. The best fiber-tip FP designs now claim few-ppm LODs (3.3 ppm [5]; 30.3 pm/ppm slope [17]), and the best gasochromic fiber-coupled design claims 3 ppm [49] — encouraging, but laboratory demonstrations, not field-hardened, T/P-compensated, H₂S-tolerant products. Fiber optics is the most credible route for continuous, no-electronics downhole hydrogen monitoring, but ppm-level field sensitivity, T/P compensation, and long-term H₂S-exposed stability are not yet demonstrated together in one qualified product.

Continuous monitoring stations (surface, mains-powered). Here the “no downhole electronics” advantage disappears, so fiber optics must compete on EMI immunity near HV infrastructure, multiplexing economy, and non-consumable operation. Gasochromic WO₃ fiber-coupled sensors with sub-second response at ppm-to-% levels [49] look attractive for a fixed monitoring-station role if catalyst-poisoning/water-photolysis aging is actively managed (e.g., periodic UV regeneration) [22].

Chemochromic tape/paint as a field aid, not a survey instrument. DetecTape and NASA/Element-One-lineage pigments are genuinely useful as a zero-power, no-instrumentation qualitative screening layer: wrap fittings, flanges, and piping, then visually or by drone-camera scan for color change during site walks. They are not quantitative, mostly threshold-triggered around the 0.04%–percent level rather than the ppm exploration-anomaly range [44], and — for non-reversible formulations — single-use “witness” marks rather than continuous monitors. Their role is safety/leak-check tape on surface hardware, not subsurface exploration.

Vendors & products

Vendor	Product / line	Technology	Key specs found	Price class	Source
Technica Optical Components, LLC (Camarillo, CA)	T910 Hydrogen Detection Sensor	Catalytic-combustion-type twin custom FBG transducer, single-mode fiber, encapsulated hydrogen-responsive material	Detection “within seconds”; daisy-chainable/multiplexable on one fiber; EMI/lightning immune; self-temperature-compensated; stainless-steel packaging	Not published (unverified)	[15][16]
Intelligent Optical Systems, Inc. (IOS) (Torrance, CA)	Multi-point fiber-optic hydrogen leak-detection system (custom, program-specific)	Optrodes with temperature-sensitive/chemochromic-type indicators on standard telecom fiber + low-cost light source	Field-demonstrated May 2001 with Boeing on a Delta IV orbital rocket static-fire test at NASA Stennis; separately developed a hydrogen sensor for Ford’s fuel-cell vehicle program (2000)	Program/contract-based (a $600k NASA follow-on contract reported in 2001); not a retail product	[70][71]
Element One, Inc. (Boulder, CO)	DetecTape® H2 (visual hydrogen leak tape)	Chemochromic silicone tape (NASA/UCF-lineage pigment chemistry)	Gray→black on exposure, “changes color within minutes”; operating range −54 °C to +232 °C; 1″ or 2″ width × 15 ft rolls; non-hermetic, gas-permeable so sniffers still work; $129–185 per roll; tested/validated by NREL (2016)	Low (consumable)	[43][52][53]
Element One, Inc.	Chemochromic paints, inks, decals, pigments (“Visual Hydrogen Detection” line)	Same pigment chemistry as DetecTape, in paint/ink/molded-part/textile/composite/film formats	Detection threshold as low as 0.04% (1/100 LFL) cited for one formulation; reversible or non-reversible options	Low–moderate depending on format	[44][72]
NASA Kennedy Space Center / Florida Solar Energy Center (UCF)	Chemochromic detector IP (US 8,293,178; US 8,920,730; US 8,945,473; reversible-variant US 8,623,662)	Hydrogen-sensitive pigment (WO₃, MoO₃, silico-/phospho-tungstic and -molybdic acids) + nano-metal activator, gas-permeable binder	Patent family underlying multiple commercial licensees, historically including tape products in the DetecTape lineage	N/A (IP; licensed)	[23][24][25][26][73][74][75][76][77]
FemtoFiberTec GmbH (Berlin)	Custom femtosecond-inscribed Type-II FBGs (Bragg components, not a packaged H₂ sensor)	fs-laser-inscribed FBGs, ~50% reflectivity, 1550 nm, in SMF-28 fiber	Component-level supplier for researchers building custom Pd-coated FBG hydrogen sensors, not a finished hydrogen-sensor product	Component-level	seen as a supplier across multiple academic Pd-FBG papers (unverified as a standalone source)
Multiple academic/national-lab groups (no packaged commercial product found)	WO₃-PdPt-Pt and WO₃-Pd₂Pt-Pt fiber-coupled nanocomposite sensors	Gasochromic WO₃ + dual Pd/Pt catalyst on fiber-coupled optrode, read by standard photodetector	3 ppm LOD; sub-second response at 4,000–10,000 ppm	Prototype / lab-scale	[49][50][51]

Notable absence. No vendor was found offering a commercially packaged, field-hardened downhole (HPHT) Pd-fiber or gasochromic-fiber hydrogen sensor product with a public datasheet quoting LOD, T/P range, and lifetime — the closest is the ongoing NETL-led subsurface wellbore-integrity sensor program, which is a government-lab R&D effort (with industry partners and multiple patents/publications) rather than a shipping commercial product [67][68][69].

Open questions

1. No field-validated distributed FBG hydrogen profile in an actual borehole or pipeline was found. The WDM/TDM multiplexing infrastructure is mature for strain/temperature/acoustic sensing [45][46][66]; worth contacting NETL’s wellbore-integrity program [67][68][69] and/or Technica [15] to ask whether a multi-point T910-class chain has ever run in a real well.
2. Quantify downhole pressure effects on Pd-FBG/Pd-FP calibration. The Sieverts’-law argument above is a first-principles inference, not a measured calibration curve — needs a high-pressure test-chamber study before trusting a surface-calibrated Pd-fiber sensor at depth.
3. H₂S tolerance budget. No fiber-optic-specific H₂S time-to-failure curve was found (only general Pd literature [56] and an uncited 61-day resistive-sensor data point). Given likely sour target formations, this needs dedicated bench testing before selecting a Pd-fiber product.
4. Get an actual datasheet from Technica for the T910 (LOD, T/P range, lifetime, accuracy) — the public page lacks these numbers [15]; a direct vendor RFI is warranted before procurement.
5. Confirm whether IOS still sells any hydrogen-sensing product. Strong 2000–2001 references, but no current catalog SKU, and the company’s “About Us” page omits hydrogen sensing today [78] — may have been discontinued or spun out.
6. Hydrogen-darkening mitigation choice for any downhole fiber run (sensing or otherwise) — hydrogen-resistant fiber chemistry, hermetic coating, or dual-wavelength self-calibration [59][60][61][62] — affects system design regardless of which H₂-sensing chemistry, if any, rides on the same fiber.
7. No soil-gas-portable fiber-optic H₂ instrument exists today — worth periodically re-scanning the market in case a ruggedized interrogator + Pd-FP probe emerges to challenge the electrochemical GA5000-class instruments used for surface surveys [63][64][65].
8. Reconcile the wide range of reported Pd-FBG “pm/%” sensitivities (roughly 8–90 pm/%, plus higher-sensitivity outliers) into one normalized figure of merit (e.g., pm per %·nm-of-Pd-thickness) for apples-to-apples comparison.
9. Cost modeling. No public interrogator pricing was found; get vendor quotes for a representative multi-channel FBG interrogator sized to a plausible number of downhole points before writing a procurement case.
10. Chemochromic tape reversibility spec for DetecTape was not confirmed (vendor literature is ambiguous) — confirm directly with Element One before specifying it as a repeatable monitor vs. single-use witness tape.

Related pages

• Why H₂ sensing is hard
• Raman spectroscopy
• Enhanced Raman techniques
• IR absorption and proxies
• UV/EUV and fluorescence
• MOX chemiresistors
• Work-function devices
• Palladium-resistive sensors
• Thermal conductivity & acoustic sensors
• Catalytic pellistors
• Electrochemical cells
• Plasmonic metamaterials
• Geologic context for H₂ exploration

References

Full reference list (79 entries)

1. Butler, M.A. “Optical fiber hydrogen sensor.” Applied Physics Letters 45, 1007 (1984). https://pubs.aip.org/aip/apl/article-abstract/45/10/1007/50257/Optical-fiber-hydrogen-sensor?redirectedFrom=fulltext
2. Butler, M.A. & Ginley, D.S. “Hydrogen sensing with palladium-coated optical fibers.” Journal of Applied Physics 64, 3706 (1988). https://pubs.aip.org/aip/jap/article-abstract/64/7/3706/15775/Hydrogen-sensing-with-palladium-coated-optical?redirectedFrom=PDF
3. “Change in the crystalline structure during the phase transition of the palladium–hydrogen system.” Physical Chemistry Chemical Physics (2015). https://pubs.rsc.org/en/content/articlelanding/2015/cp/c5cp02725j
4. “Dislocations, texture and stress development in hydrogen-cycled Pd thin films: An in-situ X-ray diffraction study.” International Journal of Hydrogen Energy (2022). https://www.sciencedirect.com/science/article/pii/S0360319922004773
5. “Fabry–Pérot Cavities with Suspended Palladium Membranes on Optical Fibers for Highly Sensitive Hydrogen Sensing.” Molecules (2023). https://pmc.ncbi.nlm.nih.gov/articles/PMC10574199/
6. “Study of n value and α/β palladium hydride phase transition within the ultra-thin palladium composite membrane.” Journal of Membrane Science (2008). https://www.sciencedirect.com/science/article/abs/pii/S0376738808006224
7. “Hydrogen diffusion and solubility in palladium thin films.” University of Kentucky. https://scholars.uky.edu/en/publications/hydrogen-diffusion-and-solubility-in-palladium-thin-films/
8. “Optical Fiber Grating Hydrogen Sensors: A Review.” Sensors 17(3), 577 (2017). https://pmc.ncbi.nlm.nih.gov/articles/PMC5375863/
9. Sutapun, B., Tabib-Azar, M. & Kazemi, A. “Pd-coated elastooptic fiber optic Bragg grating sensors for multiplexed hydrogen sensing.” Sensors and Actuators B: Chemical (1999). https://www.sciencedirect.com/science/article/abs/pii/S0925400599002403
10. “Fiber optic hydrogen sensor based on an etched Bragg grating coated with palladium.” Applied Optics 54(35), 10342 (2015). https://opg.optica.org/ao/abstract.cfm?uri=ao-54-35-10342
11. “Greatly etched fiber Bragg grating hydrogen sensor with Pd/Ni composite film as sensing material.” Sensors and Actuators B: Chemical (2013). https://www.sciencedirect.com/science/article/abs/pii/S0925400512006983
12. “Performance of fiber Bragg grating hydrogen sensor coated with Pt-loaded WO3 coating.” Sensors and Actuators B: Chemical (2014). https://www.sciencedirect.com/science/article/abs/pii/S0925400513010198
13. “Enhanced sensitivity of fiber Bragg grating hydrogen sensor using flexible substrate.” Sensors and Actuators B: Chemical (2014). https://www.sciencedirect.com/science/article/abs/pii/S0925400514002172
14. “Pd/Au nanofilms based tilted fiber Bragg grating hydrogen sensor.” Optics Communications (2022). https://www.sciencedirect.com/science/article/abs/pii/S0030401821006738
15. “T910 Hydrogen Detection Sensor.” Technica Optical Components, LLC. https://technicasa.com/t910-hydrogen-detection-sensor/
16. “Technica Optical Components, LLC” company profile. RP Photonics. https://www.rp-photonics.com/bg/profiles/technica_optical_components.html
17. “Fiber-tip Fabry–Pérot interferometer with a graphene–Au–Pd cantilever for trace hydrogen sensing.” Lab on a Chip (2023). https://pubs.rsc.org/en/content/articlelanding/2023/lc/d2lc01149b
18. “Highly sensitive hydrogen sensor based on an in-fiber Mach-Zehnder interferometer with polymer infiltration and Pt-loaded WO3 coating.” Optics Express 29(3), 4147 (2021). https://opg.optica.org/oe/fulltext.cfm?uri=oe-29-3-4147&id=446897
19. “Pt and Pd as catalyst deposited by hydrogen reduction of metal salts on WO3 films for gasochromic application.” Solar Energy Materials and Solar Cells (2013). https://www.sciencedirect.com/science/article/abs/pii/S0169433213003358
20. “WO3 Thin-Film Optical Gas Sensors Based on Gasochromic Effect towards Low Hydrogen Concentrations.” Materials 16(10), 3831 (2023). https://pmc.ncbi.nlm.nih.gov/articles/PMC10223437/
21. “Improving the efficiency of gasochromic response by using a dual sample system with WO3 thin films with Pd and Pt catalysts.” Scientific Reports (2025). https://www.nature.com/articles/s41598-025-31815-3
22. “Water photolysis effect on the long-term stability of a fiber optic hydrogen sensor with Pt/WO3.” Scientific Reports (2016). https://www.nature.com/articles/srep39160
23. Roberson, L., Captain, J., Williams, M., Smith, T., Tate, L.C. “Chemochromic detector for sensing gas leakage and process for producing the same.” US Patent 8,293,178 B2 (2012). https://image-ppubs.uspto.gov/dirsearch-public/print/downloadPdf/8293178
24. “Gas permeable chemochromic compositions for hydrogen sensing.” US Patent Application 20070224081A1. https://patents.google.com/patent/US20070224081A1/en
25. “Visual hydrogen detector with variable reversibility.” NASA Technical Reports Server, 20120014374. https://ntrs.nasa.gov/citations/20120014374
26. Muradov, N. et al. “Methods of forming visual hydrogen detector with variable reversibility.” US Patent 8,623,662 B1. https://patents.google.com/patent/US8623662?oq=nazim+muradov+reversible
27. “Tilted Fiber Bragg Grating Sensor Using Chemical Plating of a Palladium Membrane for the Detection of Hydrogen Leakage.” Sensors (2018). https://pmc.ncbi.nlm.nih.gov/articles/PMC6308443/
28. “Intrinsically Safe Optical Fiber Hydrogen Sensor Using Pt-SiO2 Coated Long-Period Fiber Grating.” Nanomaterials 16(2), 95 (2026). https://www.mdpi.com/2079-4991/16/2/95
29. “Review of the Status and Prospects of Fiber Optic Hydrogen Sensing Technology.” Photonics 11(9), 473 (2024). https://www.mdpi.com/2227-9040/11/9/473
30. “Room temperature operated hydrogen sensor using palladium coated on tapered optical fiber.” Sensors and Actuators B: Chemical (2022). https://www.sciencedirect.com/science/article/abs/pii/S0921510722004809
31. “Tapered Optical Fiber Functionalized with Palladium Nanoparticles by Drop Casting and Laser Radiation for H2 and Volatile Organic Compounds Sensing Purposes.” Sensors (2017). https://pmc.ncbi.nlm.nih.gov/articles/PMC5620954/
32. “Palladium/Graphene Oxide Nanocomposite for Hydrogen Gas Sensing Applications Based on Tapered Optical Fiber.” Sensors (2022). https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9697936/
33. “Highly Sensitive Dissolved Hydrogen Sensor Based on Side-Polished Fiber Bragg Grating.” IEEE (2015). https://ieeexplore.ieee.org/document/7101266/
34. “Pd/Ag coated fiber Bragg grating sensor for hydrogen monitoring in power transformers.” PubMed (2015). https://pubmed.ncbi.nlm.nih.gov/25933889/
35. “Research on High Sensitive D-Shaped FBG Hydrogen Sensors in Power Transformer Oil.” Sensors (2016). https://pmc.ncbi.nlm.nih.gov/articles/PMC5087429/
36. “Fiber-optic evanescent-wave hydrogen gas sensor using palladium-supported tungsten oxide.” Sensors and Actuators B: Chemical. https://www.researchgate.net/publication/222034627_Fiber-optic_evanescent-wave_hydrogen_gas_sensor_using_palladium-supported_tungsten_oxide
37. “Fiber optic hydrogen sensor based on a Fabry–Perot interferometer with a fiber Bragg grating and a nanofilm.” Lab on a Chip (2021). https://pubs.rsc.org/en/content/articlelanding/2021/lc/d1lc00012h
38. “High-sensitivity and fast-response fiber-tip Fabry–Pérot hydrogen sensor with suspended palladium-decorated graphene.” Nanoscale (2019). https://pubs.rsc.org/en/content/articlelanding/2019/nr/c9nr04274a
39. “Femtosecond laser fabricated micro Mach-Zehnder interferometer with Pd film as sensing materials for hydrogen sensing.” Optics Letters 37(11), 1940 (2012). https://opg.optica.org/ol/abstract.cfm?uri=ol-37-11-1940
40. “An optical fiber hydrogen concentration sensor based on Mach-Zehnder interferometer coated with a film of palladium.” IEEE (2018). https://ieeexplore.ieee.org/document/8595983/
41. “In-line Mach-Zehnder Interferometer and FBG with Pd film for simultaneous hydrogen and temperature detection.” Sensors and Actuators B: Chemical (2015). https://www.sciencedirect.com/science/article/abs/pii/S0925400514006819
42. “Optical hydrogen sensing with high-Q guided-mode resonance of Al2O3/WO3/Pd nanostructure.” Scientific Reports (2023). https://www.nature.com/articles/s41598-023-28204-z
43. “Passive Leak Detection Using Commercial Hydrogen Colorimetric Indicator.” NREL/OSTI Technical Report (2016). https://www.osti.gov/biblio/1326889
44. “Visual Hydrogen Detection.” Element One. https://elem1.com/visual-hydrogen-detection/
45. “High-density fiber Bragg grating array enabling fully distributed sensing with sub-millimeter spatial resolution.” Optics Letters 51(1), 117 (2026). https://opg.optica.org/ol/abstract.cfm?uri=ol-51-1-117
46. “Ultra-Weak Fiber Bragg Grating Sensing Network Coated with Sensitive Material for Multi-Parameter Measurements.” Sensors (2017). https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5539773/
47. “Optical Fiber Hydrogen Sensor Based on π-Phase-Shifted Grating and Sputtered Pd/Hf Composite Film.” Photonic Sensors (2025). https://link.springer.com/article/10.1007/s13320-025-0750-8
48. “A compact hydrogen sensor based on the fiber-optic Fabry-Perot interferometer.” Optics and Lasers in Engineering (2020). https://www.sciencedirect.com/science/article/abs/pii/S0030399219319711
49. “Early Detection of Hydrogen Leakage Using Fiber Optic Hydrogen Sensor Based on WO3-PdPt-Pt Nanocomposite Films.” Sensors (2025). https://pmc.ncbi.nlm.nih.gov/articles/PMC12157849/
50. “Fiber Optical Hydrogen Sensor Based on WO3-Pd2Pt-Pt Nanocomposite Films.” Nanomaterials 11(1), 128 (2021). https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7826938/
51. “Improved performance of fiber-optic hydrogen sensor of porous Pt/WO3 based on ZIF-8.” International Journal of Hydrogen Energy (2023). https://www.sciencedirect.com/science/article/abs/pii/S0360319923029336
52. “DetecTape® H2 Visual Hydrogen Leak Detector Tape.” Element One / DetecTape product page. https://detectape.com/product/detectape/
53. DetecTape homepage. https://detectape.com/
54. “Recent Advances in Palladium Nanoparticles-Based Hydrogen Sensors for Leak Detection.” Sensors (2019). https://pmc.ncbi.nlm.nih.gov/articles/PMC6832907/
55. “Hydrogen Sensors Based on Pd-Based Materials: A Review.” Sensors 25(11), 3402 (2025). https://www.mdpi.com/1424-8220/25/11/3402
56. “Nanoscale Analysis of Sulfur Poisoning Effects on Hydrogen Sorption in Single Pd Nanoparticles.” (2025). https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12574217/
57. “Long-term reliable wireless H2 gas sensor via repeatable thermal refreshing of palladium nanowire.” Nature Communications (2024). https://www.nature.com/articles/s41467-024-53080-0
58. “Nanostructured Pd–Au based fiber optic sensors for probing hydrogen concentrations in gas mixtures.” International Journal of Hydrogen Energy (2013). https://www.sciencedirect.com/science/article/abs/pii/S0360319913000669
59. “Hydrogen Darkening.” distributedsensing.net. http://www.distributedsensing.net/hydrogen-darkening
60. “FIBER OPTICS: Overcoming Hydrogen Darkening.” Journal of Petroleum Technology / SPE. https://jpt.spe.org/fiber-optics-overcoming-hydrogen-darkening
61. “Evaluation of in-situ optical loss of polyimide-coated optical fiber under hydrothermal environments.” Geothermics (2023). https://www.sciencedirect.com/science/article/abs/pii/S1068520023003632
62. “Optical fibers present opportunities and challenges for geophysical applications.” Offshore Magazine. https://www.offshore-mag.com/geosciences/article/16760286/optical-fibers-present-opportunities-and-challenges-for-geophysical-applications
63. “Exploring natural hydrogen hotspots: a review and soil-gas survey design for identifying seepage.” Geoenergy (2023). https://www.lyellcollection.org/doi/full/10.1144/geoenergy2023-014
64. “Natural H2 Transfer in Soil: Insights from Soil Gas Measurements at Varying Depths.” Geosciences 14(11), 296 (2024). https://www.mdpi.com/2076-3263/14/11/296
65. “Natural H2 exploration: tools and workflows to characterize a play.” Science and Technology for Energy Transition (STET) (2023). https://www.stet-review.org/articles/stet/full_html/2023/01/stet20230116/stet20230116.html
66. “FBG Arrays for Quasi-Distributed Sensing: A Review.” Photonic Sensors (2021). https://link.springer.com/article/10.1007/s13320-021-0615-8
67. “NETL and Partners Develops Novel Embedded Sensors for Subsurface Wellbore Integrity.” National Energy Technology Laboratory. https://netl.doe.gov/node/12775
68. “Selective H2 Sensing Through Use of Palladium and Platinum-based Nanoparticle Functional Sensor Layers Integrated with Engineered Filter Layers.” NETL. https://netl.doe.gov/node/9222
69. “NETL Patents Fiber Optic Sensor Technology for Hydrogen Leak Detection.” NETL. https://netl.doe.gov/node/12415
70. “NASA Awards Intelligent Optical Systems Inc. $600,000 Follow-On Contract to Develop Fiber Optic Hydrogen Leak Detection System.” SpaceNews (2001). https://spacenews.com/nasa-awards-intelligent-optical-systems-inc-600000-follow-on-contract-to-develop-fiber-optic-hydrogen-leak-detection-system/
71. “Intelligent Optical to Develop Hydrogen Sensor for Ford’s Fuel Cell.” Photonics.com (2000). https://www.photonics.com/Article.aspx?AID=7202
72. “EERE Success Story - Element One Reduces Cost of Hydrogen Leak Detection Systems.” U.S. Department of Energy. https://www.energy.gov/cmei/success-stories/articles/eere-success-story-element-one-reduces-cost-hydrogen-leak-detection
73. “Color Changing Hydrogen Sensors.” NASA Technical Reports Server, 20150023595. https://ntrs.nasa.gov/citations/20150023595
74. “Color Changing Material for Hydrogen Leak Detection.” NASA Technical Reports Server, 20140012641. https://ntrs.nasa.gov/citations/20140012641
75. “Hydrogen Detection Tape Saves Time and Lives.” NASA Spinoff 2016. https://spinoff.nasa.gov/Spinoff2016/ps_5.html
76. “Innovative Hydrogen Leak Detection Tape Earns Prestigious Awards.” NASA. https://www.nasa.gov/technology/innovative-hydrogen-leak-detection-tape-earns-prestigious-awards/
77. “Chemochromic detector for sensing gas leakage and process for producing the same.” US Patent 8,920,730 B2. https://patents.google.com/patent/US8920730B2/en
78. “About Us.” Intelligent Optical Systems, Inc. https://intopsys.com/company/about-us/
79. Hübert, T., Boon-Brett, L., Black, G., Banach, U. “Hydrogen sensors – A review.” Sensors and Actuators B: Chemical 157, 329–352 (2011). https://www.sciencedirect.com/science/article/abs/pii/S0925400511003674