Skip to content
H₂ Sensing Compendium

Catalytic pellistors

Summary

Section titled “Summary”

A pellistor is a self-heated catalytic bead that burns a combustible gas flamelessly at its surface and reads the resulting heat as a resistance change, reported as percent of the Lower Explosive Limit (%LEL) rather than parts-per-million. It is cheap, rugged, certified for hazardous areas, and has been the oil-and-gas and industrial-safety default combustible-gas sensor for roughly a century, but it answers one question only: whether a space is approaching an explosive atmosphere. That threshold sits hundreds of times above the ppm-level hydrogen anomalies geologic exploration targets.

For Hestia, a pellistor belongs at the wellhead or gas-gathering skid as a safety backstop, not in the soil-gas survey kit or on a downhole tool string.

At a glance — published performance envelope

Detection limit
40–2,000 ppm
Range
0–40,000 ppm
Response t90
3–30 s
Recovery
5–60 s
Selectivity
none
Drift
high
Operating T
−40 … 75 °C
Humidity
0–95 % RH
Power
5–400 mW
Unit cost
$20–$150
TRL
9

Main interferents: O₂, CH₄, CO, propane butane and higher alkanes · poisons: silicones siloxanes HMDS by far most common and severe, hydrogen sulfide high concentration or prolonged exposure, lead compounds tetraethyl lead. Ranges span the published devices/variants covered below — per-claim citations are in the body; data: src/data/modalities/catalytic-pellistors.yaml.

How it works

Section titled “How it works”

A catalytic-bead sensor, or pellistor (“pelletised resistor”), is a calorimeter with a resistive readout: two beads on fine platinum-wire coils, an active/detector bead whose ceramic (alumina) matrix carries a noble-metal catalyst (Pt, Pd, or Pt/Pd alloys), and a compensator bead of matched thermal mass, chemically treated so it cannot catalyze combustion [1][2]. The coils double as heaters, holding both beads near 400–550 °C, and as resistance thermometers reading their own temperature [2][3]. The design descends from a 1923 mining-methane sensor patent [5] and reached the oilfield by the late 1920s, becoming the standard mud-logging “hot-wire” technique from 1939 [6].

Response. Combustible gas at the detector bead oxidizes catalytically without a flame; the heat of reaction raises that bead’s temperature and resistance relative to the compensator. The two beads sit in opposite arms of a Wheatstone bridge, and the out-of-balance voltage is the signal [1][2][4]. Because platinum resistance rises linearly with temperature (R(T)=R0[1+α(T−T0)]R(T)=R_0\left[1+\alpha(T-T_0)\right], the same relation used in Pt RTD thermometry), the temperature differential shows up as a bridge output

Vout≈Vs4⋅ΔRdetRV_{out} \approx \frac{V_s}{4}\cdot\frac{\Delta R_{det}}{R}

for small imbalances, where VsV_s is the bridge supply voltage [4]. This is why datasheets specify performance in mV per volume-% target gas (12 mV/% methane for the SGX MP-7217, >25 mV/% for the Honeywell CAT25 [7][8]) rather than in resistance units.

Catalyst role. Combustible gas/air mixtures do not ignite spontaneously at bead operating temperature: the uncatalyzed activation energy EaE_a is too high to overcome thermally at ~500 °C [4][5]. The noble-metal catalyst provides a lower-energy reaction path, so combustion runs to completion at an electrically sustainable temperature. Response rises with bead voltage, plateaus, then falls as heat loss dominates; methane’s unusually high activation energy is why pellistors are tuned to its plateau at standard operating voltage, while hydrogen and other gases plateau lower, one reason cross-gas response ratios are voltage-dependent [4].

Diffusion-limited response. A hot, unpoisoned catalyst consumes essentially all fuel that reaches it, so the rate-limiting step is diffusion of fuel to the bead, not surface chemistry. Firth, Jones & Jones (1973) modeled pellistor output at a fuel’s lower explosive limit as

V(LEL)=K D12 ΔHox [LEL]V_{(LEL)} = K \, D_{12}\, \Delta H_{ox}\, [\text{LEL}]

where KK is a construction constant, D12D_{12} the fuel’s binary diffusion coefficient in air, ΔHox\Delta H_{ox} the heat of oxidation, and [LEL][\text{LEL}] the fuel’s lower explosive limit in vol % [4][9]. D12D_{12} can be estimated from Chapman-Enskog kinetic theory [4], which lets SGX and others publish theoretical “k-factor” response tables relative to methane for gases with no directly measured data, though real responses (heavy/aromatic vapors, after a flame arrestor) can deviate from the prediction [3]. Larger molecules diffuse more slowly and respond lower than methane; hydrogen, the smallest and fastest-diffusing common fuel, is essentially always the highest-responding gas on a pellistor [4][8].

Four effects perturb this ideal picture in real hardware [4]:

  • a depletion zone of combustion products around the bead;
  • can geometry: a closed can with a small gas-access hole restricts diffusion and lowers output relative to methane;
  • the mandatory flame arrestor/sinter, a stainless-steel mesh required to stop combustion propagating outward, which slows heavier molecules more than methane, depressing their relative response [4][7];
  • porous catalyst structures used for poison resistance trade sensitivity for a larger reservoir of active sites, so porous-bead sensors run less sensitive than solid-surface beads [4].

The sinter is both a safety requirement (flameproof enclosure) and a performance cost, slowing response and depressing heavy-hydrocarbon sensitivity.

Device classes

Section titled “Device classes”

Every catalytic-combustion variant below shares the energy-balance physics above; they differ in thermal mass, catalyst exposure, power, and poison protection.

SubtypeDistinguishing physicsRepresentative product
Classic wire-coil / bead pellistorPt coil wound into a catalyst-loaded ceramic (alumina) bead; the original and still-dominant form since the mid-20th centuryHoneywell/City Technology CAT25 [8]; Winsen MC105 [10]
Poison-resistant (“ZD”/“PR”) bead variantsSame physics, but modified catalyst loading, porous protective structure, and/or a different noble-metal formulation specifically engineered to resist silicone and H₂S deactivationSGX VQ548ZD/VQ549ZD (“Poison Resistance: Silicones & H₂S”) [3]; Dräger CatEx “PR” (“new pellistor composition…highly resistant to poisoning”) [11]
MEMS micro-pellistorBead-and-coil replaced by a pair of micromachined silicon diaphragms with planar Pt heater meanders, catalyzed vs. inert coatings: same Wheatstone-bridge combustion physics at drastically reduced thermal mass and footprint (Ø14 mm can housing a chip)SGX MP-7217 / VQ548MP [7][12]
Pulsed/duty-cycled MEMS operationThe same MEMS sensor in a different operating mode: the low thermal mass of the MEMS structure (<30 ms to reach 400–500 °C) lets the heater run for a fraction of a duty cycle (e.g., 4–10%) instead of continuously, cutting average power by up to ~25× while increasing net sensitivity, because the transient signal measured before gas-diffusion/catalysis reach equilibrium is larger than the continuous-mode steady-state signalSGX MP-7217 cycled-mode application note [12]
Thermal-conductivity “pellistor” cousinSame physical bead-pair package and Wheatstone-bridge readout, but the compensator/detector pair is matched and uncatalyzed; signal derives from the difference in thermal conductivity between the sample gas and air rather than from combustion. Needs no oxygen and can read up to 100 vol %, but cannot discriminate a combustible gas from any other high-thermal-conductivity gas (e.g., He) and cannot resolve ppm-level concentrations; see thermal-conductivity & acoustic sensorsSGX VQ546M/VQ546MR (methane, %v/v) [3]
Thermoelectric (Seebeck-readout) catalytic sensorInstead of a resistance-differential Wheatstone bridge requiring continuous excitation of both elements, a single element carries a catalyzed zone and a bare zone; the differential reaction-heat directly generates a Seebeck-effect thermopile voltage, in principle removing the bridge-excitation electronics from the signal path (though heating to light-off temperature is still typically required)Origin traced to McAleer (1985); continuing development noted in the gas-sensor review literature [5]
Historical “hot-wire” bare-filament total-combustibles detectorA single uncoated (or lightly catalyzed) platinum filament in a Wheatstone bridge, without the bead/ceramic encapsulation: the pellistor’s direct ancestor, still recognizable in mud-logging “total gas” detectorsMud-logging gas-trap analyzers, in continuous use since the late 1930s [6]

Performance envelope

Section titled “Performance envelope”

Every commercial catalytic sensor is specified against methane as the reference gas and %LEL as the reporting unit, reflecting the technology’s safety-alarm design history, not trace-gas quantification.

ParameterTypical valueSource/model
Resolution / practical LOD~1 %LEL ≈ 400 ppm H₂ (display/alarm resolution); some fixed systems quote 1 %LEL / 1 vol % as the specified resolutionDräger CatEx125PR [11]
Minimum sensitivity12 mV/% CH₄ (MEMS, at 3.0 V) to >25 mV/% CH₄ (classic bead)SGX MP-7217 [7]; Honeywell CAT25 [8]
Measurement range0–100 %LEL standard; some heads to 5 vol % CH₄ (MEMS) or 100 vol % CH₄ (TC-mode variant)SGX MP-7217, VQ546M [7][3]
Response time (T90)3 s device-level minimum (MEMS chip w/ optimized delivery) to ~20 s (housed sensor head, methane, standard test manifold)SGX MP-7217 <12 s [7]; VQ500 <20 s [3]; CAT25 <10 s (methane) [8]; Winsen MC105 ≤10 s (mfr. spec) [10]
Recovery time (T90)≤30 s typicalWinsen MC105 [10]
SelectivityNone: responds to essentially all combustible gases and vapours; relative gas-to-gas response (“k-factor”) is catalyst- and construction-dependent, not a fixed physical constant (see below)[4][3][8][13]
Operating temperature−40 °C to +50/+55 °C (classic bead heads); −40 °C to +60 °C operating / +75 °C certified (MEMS)CAT25 [8]; VQ500 [3]; MP-7217 [7]
Operating pressure~1 atm ±10%; 70–130 kPa certified range (MEMS)CAT25 [8]; MP-7217 [7]
Humidity range0–90–95 %RH non-condensing (varies by model; VQ500 spec’d 0–80%)CAT25 [8]; MP-7217 [7]; VQ500 [3]
Long-term zero drift<1.0 mV/month (MEMS, 90-day test) to <5 %LEL/month (classic bead, worst-case 12-month-averaged spec); <±5 %LEL CH₄/yr (CAT25)MP-7217 [7]; VQ500 [3]; CAT25 [8]
Long-term sensitivity/span drift<0.6 mV/month (MEMS) to <5 %Full-Scale/month (classic bead worst case); <±2 %LEL CH₄/yr (CAT25); ±2 mV/yr (MC105)[7][3][8][10]
Expected operating life>2 years (CAT25, warrantied); up to 5 years claimed (Winsen MC105)[8][10]
Power consumption39–47 mA @ ~3 V ≈ 120–140 mW continuous (MEMS); 135–230 mW (classic bead heads, model-dependent); 230 mW max (CAT25); ~375 mW (MC105, 150 mA @ 2.5 V); down to ~5 mW average in a 4% duty-cycle pulsed mode[7][3][8][10][12]
Warm-up / T90-to-operating-temperature<30 ms (MEMS heater reaches 400–500 °C)SGX MP-7217 cycled-mode note [12]
Cost class (bare sensor element)~US$20–35 component price observed (VQ548ZD-S ≈ US$30.83, LCSC listing) [14]; full fixed-point instrument systems commonly cited in the US$500–3,000 range in market commentary (unverified, not independently confirmed against a manufacturer price list) [15]
TRL9: continuously commercial and standardized (IEC 60079, EN 60079, UL 913) since at least the 1970s, with roots to a 1923 mining sensor patent [5]

Gas-to-gas response factors (“k-factors”)

Section titled “Gas-to-gas response factors (“k-factors”)”

Two independently sourced tables show the same shape (smaller, lighter, more hydrogen-rich molecules respond higher) and real device-to-device variability in exactly how much higher hydrogen reads relative to methane.

Honeywell CAT25, relative response (methane = 100), experimentally measured at 50 %LEL CH₄ [8]:

GasRelative sensitivityGasRelative sensitivity
Methane100Methanol84
Hydrogen107Ethanol64
Ethane82Propan-2-ol49
Propane63Acetone50
Butane51Butan-2-one (MEK)48
Pentane50Diethyl ether40
Hexane46Ethyl acetate46
Heptane44Toluene44
Octane38Xylene31
Ethylene81Acetylene47

RAE Systems LEL-1 and LEL-2 pellistor cartridges, relative response and derived correction factor (CF) [13] (figures unverified against a primary RAE Systems source):

GasLEL-1 rel. sensitivity (CF)LEL-2 rel. sensitivity (CF)
Methane100 (1.0)100 (1.0)
Hydrogen43 (2.3)83 (1.2)
Propane63 (1.6)63 (1.6)
n-Pentane45 (2.2)50 (2.0)

The two-fold spread in hydrogen’s relative response between these pellistor cartridge families (43% vs. 83–107% of the methane signal at the same nominal %LEL) means a %LEL reading calibrated on methane cannot be converted to a hydrogen ppm value with a universal factor. The conversion is catalyst-specific, must come from that product’s own datasheet or be verified empirically, and even then carries real uncertainty. City Technology itself disclaims the table as “for guidance only… an instrument should be calibrated using the gas under investigation” [8].

Firth, Jones & Jones’s theoretical alkane series, computed from diffusion coefficients and heats of oxidation and normalized to methane = 1.00, shows the same fall-off with molecular size [4]:

MoleculeLEL (vol %)Relative response vs. CH₄
Methane4.41.00
Ethane2.50.68
Propane1.70.51
n-Butane1.40.52
n-Pentane1.40.51
n-Hexane1.00.40
n-Heptane1.10.42
n-Octane0.80.32
n-Nonane0.70.28

This fall-off with molecular size underlies every manufacturer’s k-factor guidance table, and is why a soil-gas or downhole matrix that co-evolves H₂ with CH₄ and heavier associated hydrocarbons produces an ambiguous, superposed pellistor signal (see Cross-sensitivity & environment, below).

Figures of merit

Section titled “Figures of merit”

The rows below apply the universal definitions on the Figures of merit page to catalytic pellistors specifically. Because this technology’s native reporting unit is %LEL rather than ppm, converting between the two (see Concentration units; 1 %LEL(H₂) = 400 ppm H₂) matters when reading any pellistor spec.

Figures-of-merit table for this modality (12 rows)
Figure of meritSymbol / formulaWhat it means hereTypical value for this technologyHow to read it
Sensitivity (bridge output)Vout≈Vs4⋅ΔRdetRV_{out} \approx \frac{V_s}{4}\cdot\frac{\Delta R_{det}}{R}, reported as mV per vol-% gasThe bridge millivolt output per unit of target gas: the physically meaningful quantity, not resistance12 mV/% CH₄ (MEMS, SGX MP-7217 @ 3.0 V) to >25 mV/% CH₄ (classic bead, Honeywell CAT25) [7][8]Hydrogen reads higher than methane on the same bridge by a device-specific factor (see response factor row); a 1:1 ratio should not be assumed
%LEL resolution / practical LODDisplay/alarm resolution floor, not an IUPAC 3σ3\sigma statisticThe smallest reliably actionable reading, set by the %LEL display grid rather than by blank-noise statistics~1 %LEL ≈ 400 ppm H₂ (Dräger CatEx125PR) [11]This is ~100–1,000× coarser than an electrochemical cell’s ppm-level LOD; no vendor publishes an averaging-time-dependent (Allan-deviation) floor because integration time isn’t a lever this diffusion-limited technology uses
Selectivity / cross-sensitivityKij=Sj/SiK_{ij}=S_j/S_i; overall selectivity =1/max⁡j(Kij)=1/\max_j(K_{ij})How well H₂ can be told apart from CH₄, CO, and other combustibles reaching the beadEffectively non-selective: responds to essentially all combustible gases at 28–107% of the methane signal depending on species and product [4][3][8][13]Selectivity sits near the floor of the universal metric by design; a %LEL reading alone can never confirm the gas is H₂ rather than co-evolved CH₄/CO
Response factor vs. CH₄ (“k-factor”)Response factor =Sgas/SCH4= S_{gas}/S_{CH_4} at fixed %LELConverts a methane-calibrated %LEL reading toward an actual H₂ concentrationH₂: 107% of CH₄ (Honeywell CAT25) [8]; 43% (RAE LEL-1) and 83% (RAE LEL-2) of CH₄ (unverified against a primary RAE source) [13]No universal H₂/CH₄ factor exists: City Technology explicitly disclaims its own table as “for guidance only” and requires calibration on the actual target gas [8]
Response time (t90t_{90})t90≈2.303 τt_{90}\approx2.303\,\tau for near-exponential kineticsTime to reach 90% of a step-change reading~3 s (MEMS chip, optimized delivery) to ~20 s (housed head, CH₄, standard test manifold) [7][3][8][10]Fast in absolute terms, but not useful on its own here: the %LEL resolution ceiling above already excludes ppm-level exploration use regardless of speed
Recovery time (t90t_{90})Diffusion/desorption-governed, family-specificTime to fall back to baseline after gas clears≤30 s typical [10]Comparable to response time, unlike the large response/recovery asymmetry seen in MOX chemiresistors, because diffusion through the sinter, not surface chemisorption, governs both directions
Dynamic rangeNdecades=log⁡10(Cmax⁡/Cmin⁡)N_{\text{decades}}=\log_{10}(C_{\max}/C_{\min})Full span of usable output0–100 %LEL standard; some heads to 5 vol % CH₄ (MEMS) or 100 vol % CH₄ (TC-mode variant) [7][3]In ppm-H₂ terms this ~2-decade range (400–40,000 ppm) sits entirely above the diagnostically interesting 1–1,000+ ppm geologic-H₂ band; dynamic range and useful range don’t overlap for this job
Drift / stability (zero & span)R0(t)=R0(0)+dZtR_0(t)=R_0(0)+d_Z t; S(t)=S(0)(1+dSt)S(t)=S(0)(1+d_S t)Long-term creep in baseline and in bridge sensitivityZero drift <1.0 mV/month (MEMS, 90-day test) to <5 %LEL/month worst-case (classic bead); <5 %LEL CH₄/yr (CAT25). Span drift <0.6 mV/month (MEMS) to <5 %FS/month worst-case; <2 %LEL CH₄/yr (CAT25) [7][3][8]Offset in safety service by a bump-test/recalibration culture (up to 4×/yr); uncorrected drift folds directly into %LEL alarm-threshold accuracy
O₂ dependenceNo formula: a hard thresholdCatalytic combustion requires molecular O₂ as the oxidantMinimum ~10–12 vol % O₂ for reliable, complete combustion [16][17][18]Below threshold the sensor can read falsely low or zero with no fault flag, indistinguishable from “no gas present”; this is not graceful degradation, and it is a consequential entry in this table for anoxic/downhole settings
Poison resistance (siloxane / H₂S)Baseline/sensitivity shift under ISO 26142 exposure, product-dependentPermanent (poisoning) vs. temporary (inhibition) loss of catalytic activity from named contaminantsStandard bead (“CAT-1”): SO₂ 500 ppm dropped baseline ~84%, H₂ sensitivity ~45% (fails ISO 26142’s 20% threshold); H₂S 50/40 ppm baseline shift fully reversible; short ISO HMDS test (10 ppm) showed no significant loss, but CAT25’s own longer 100-min 2 ppm HMDS test showed decay to ~68–70% of original signal. A poison-resistant analog (“CAT-2”) passed the SO₂ and H₂S tests with no/minimal effect [8][19]Poison resistance is a product-specific engineering feature (ZD/PR catalyst variants), not a universal trait of the technology, and the short standardized test likely understates long-duration siloxane risk, per the study’s own authors
Power consumptionHeater-dominated; duty-cycled vs. continuousElectrical draw to sustain the ~400–550 °C bead120–140 mW continuous (MEMS) to 230 mW max (classic bead, CAT25); down to ~5 mW average in a 4% duty-cycle pulsed mode [7][3][8][10][12]Determines run time for a fixed safety-monitoring station; irrelevant to exploration fitness, since the LOD/O₂/temperature limits already exclude that use case
Operating envelopeCreported=Craw⋅fT(T)⋅fRH(RH)⋅fP(P)C_{\text{reported}}=C_{\text{raw}}\cdot f_T(T)\cdot f_{RH}(RH)\cdot f_P(P)Compensated range where the stated accuracy actually holds, nested inside the wider operating range−40 °C to +50/+55 °C (classic bead) or +60 °C operating/+75 °C certified (MEMS); ~1 atm ±10%, 70–130 kPa certified (MEMS); 0–90–95 %RH [3][7][8]No datasheet surveyed extends this envelope anywhere near deep-borehole temperature or pressure, and the entire calibration implicitly assumes near-atmospheric, O₂-rich air

Cross-sensitivity & environment

Section titled “Cross-sensitivity & environment”
FactorEffect on a pellistorDesign responseRef
OxygenCatalytic combustion needs molecular O₂; below a minimum ~10–12 vol % O₂, combustion is incomplete and response falls, with no fault flag distinguishing “no gas” from “not enough O₂ to burn it”Treat any anoxic or O₂-depleted stream as a blind spot this technology cannot monitor[16][17][18]
TemperatureMatched-bead design keeps zero-offset temperature sensitivity small (SGX VQ500 ZD/W: ≤0.2% CH₄-equivalent variation, −10 to +40 °C vs. the 20 °C reading) [3], but ambient sits underneath the bead’s own ~400–550 °C operating point; as ambient nears the rated ceiling (+50 to +75 °C) the detector/compensator temperature differential narrowsNo catalytic sensor surveyed is rated anywhere near deep-borehole ambient temperatures[3]
HumidityCity Technology’s CAT25 datasheet shows zero output rising with RH, reaching a ~4 %LEL CH₄-equivalent offset at 100% RH (comparable in magnitude to a real low-level gas signal), driven by water vapor’s distinct thermal conductivity and asymmetric detector/compensator response; the manufacturer flags this as guidance only, not a specification [8]Cross-sensitivity worsens with catalytic-element surface area, so is worse for high-surface-area/porous poison-resistant beads[8]
CO₂ / other inert gasesAny gas with thermal conductivity different from air perturbs bead temperature independent of combustion; the compensator cancels this imperfectly, especially at the percent-level CO₂ concentrations common in soil gasSGX’s own MEMS datasheet flags CO₂-driven thermal-conductivity variation as a compensated-but-imperfect effect[7][3]
PressureGeneral trade-literature guidance holds that temperature, humidity, and pressure have no significant accuracy impact within the rated operating band (SGX MP-7217: 70–130 kPa, roughly sea level ±30%)The claim is scoped to near-atmospheric conditions; no vendor datasheet surveyed specifies performance at downhole hydrostatic/lithostatic pressure[7]
CH₄Read by the same catalytic mechanism as H₂, at ~50–100% of the H₂-normalized signal depending on product; indistinguishable from H₂ on a pellistor aloneThis is what non-selectivity means in practicek-factor tables above
COItself combustible; adds to (not subtracts from) the total %LEL signal, again indistinguishable from H₂ or CH₄——
HeChemically inert but has a thermal conductivity even higher than H₂; mainly perturbs baseline/compensation similarly to CO₂, a smaller effect here than in the thermal-conductivity-mode sibling device, where He is directly confounded with combustible gas——
H₂SActs as an inhibitor at low/moderate concentration (temporary, reversible in clean air) and, per SGX’s own precautions, as a permanent poison at high levels [7]. Palmisano et al. found 50/40 ppm H₂S produced a fully reversible baseline shift on one commercial sensor (“CAT-1”) and no measurable effect on a poison-resistant model (“CAT-2”) [19]“Poison resistance” is a product-specific engineering feature, not a universal trait of the technology[7][19]

Quantified poisoning data (Palmisano et al. 2015, JRC–NREL collaborative study, ISO 26142 protocol) [19]:

ContaminantConcentrationEffect on catalytic sensor “CAT-1”Effect on poison-resistant “CAT-2”
H₂S50/40 ppmBaseline drop, fully reversibleNo measurable effect
SO₂500 ppmBaseline dropped ~84% (X0=−0.84X_0=-0.84); H₂ sensitivity fell ~45% (fails ISO 26142’s 20% pass/fail threshold); only partial recovery after exposureSensitivity “slightly decreased” but baseline unaffected: passes
HMDS (siloxane)10 ppm, ISO 26142 standard testNo significant degradation within the short ISO test exposure; minor sensitivity increase observedNo significant degradation
NO₂20 ppmBaseline decreased slightly, sensitivity change small; largely recoveredNot tested on this model

The authors caution that this short ISO 26142 siloxane test “may not properly reflect sensor robustness to siloxanes” in real, longer-duration field exposure [19], a real gap given the safety-industry field reports that silicone/HMDS is “by far the most common and virulent poison” in practice, causing full, irreversible loss of response with the sensor still reading a deceptive 0% [7][20]. Consistent with that caution, CAT25’s own poison-resistance chart (2.5 vol % CH₄ + 2 ppm HMDS) shows output falling from 100% to roughly 68–70% of original over a 100-minute continuous exposure, a substantially more aggressive trajectory than the short ISO test would suggest [8][19].

Failure modes: silent zero-oxygen failure (no fault flag, reads as “no gas”); irreversible poisoning by silicones/siloxanes, high H₂S, lead, and phosphorus compounds (catalyst surface passivation) [7][4]; reversible inhibition by lower-level H₂S/SO₂ and halogenated solvents (Freons, trichloroethylene, methyl chloride) [7]; sensitivity/zero drift requiring scheduled recalibration; catastrophic overload, where exposure above 100 %LEL can permanently damage sensitivity/zero calibration via carbon-soot deposition or physical destruction of the catalytic surface [1][3].

Use for geologic H₂

Section titled “Use for geologic H₂”

Soil-gas surveys. Not suitable. Hestia’s exploration signal runs from the ~0.53 ppm atmospheric background through anomalies typically reported at 1–1,000+ ppm in the published natural-hydrogen soil-gas literature [21][22]. A pellistor’s practical resolution floor, ~1 %LEL ≈ 400 ppm H₂, is blind to essentially the entire diagnostically interesting range, and would only respond to an exceptionally strong, already-hazardous seep. This is why the geologic-H₂ soil-gas literature surveyed here uses electrochemical analyzers (e.g., the GA5000/BIOGAS5000 platform) instead, for ppm-level survey work: see electrochemical cells [21]. A pellistor’s own humidity zero-drift and complete non-selectivity to co-evolved CH₄/CO₂/H₂S would further corrupt any push toward trace detection even before the fundamental LOD ceiling is considered.

Downhole / borehole logging. Categorically excluded, for three independent, compounding reasons:

  1. Oxygen dependence. Catalytic combustion requires >~10–12 vol % O₂ [16][17][18]. Deep, reducing/anoxic subsurface and formation gas is frequently O₂-depleted, exactly the regime this technology fails in, and fails silently.
  2. Active, powered electronics with an internal ignition source. A pellistor is a continuously (or duty-cycled) self-heated element at 400–550 °C, not a passive transducer, against the “minimize powered/active elements” preference for HPHT tool strings.
  3. Temperature ceiling. No datasheet surveyed rates ambient operation above +55 to +75 °C [7][3][8], an order of magnitude below deep-borehole ambient. The sensor depends on a controlled differential between a hot catalyzed bead and a matched compensator, and that differential collapses as ambient nears the bead’s own operating point.

Continuous fixed monitoring stations. The correct use case, as a safety instrument rather than an exploration sensor. At a wellhead, gas-gathering skid, laboratory, or enclosed workspace where the operating question is whether the space is approaching an explosive atmosphere, pellistors are the mature, inexpensive, certified (ATEX/IECEx/UL), field-proven default, with an established bump-test/calibration culture already familiar to EHS departments [8][11][17] (see Standards & regulations for the certification landscape). Deploy pellistor-based fixed detectors as a parallel safety backstop at facilities handling or venting H₂ at concentration, never as a substitute for the ppm-capable electrochemical/optical instrumentation used for the actual exploration signal.

Historical nuance. Catalytic “hot-wire” total-combustible-gas detectors have monitored oil-and-gas mud-logging gas traps continuously since 1939, building on 1920s Wheatstone-bridge “hot wire” concepts [6]. That measures percent-level total combustibles degassing from mud at the surface, not ppm-level H₂ in soil gas or downhole, and gas chromatography substantially displaced it in mud logging by the 1970s [6]. The mud-logging history is the closest real precedent for the coarse, non-selective, percent-level role pellistors remain suited to today, not evidence they fit modern geologic-H₂ exploration duty. See why H₂ is hard.

Vendors & products

Section titled “Vendors & products”
ManufacturerProduct lineModel/SKUKey specs (as documented)Price classRef
SGX Sensortech (Amphenol)VQ500 series (classic bead, 4-series flameproof head)VQ548ZD, VQ549ZD (poison-resistant), VQ547TS (LEL ammonia), VQ546M/MR (thermal-conductivity, %v/v)T90 < 20 s; 135–230 mW; −40 to +55 °C; ATEX/IECEx/CSA/ULBare element ≈ US$30 (VQ548ZD-S, LCSC) [14][3]
SGX Sensortech (Amphenol)MEMS PellistorMP-7217 (= VQ548MP)T90 < 12 s; 110–140 mW continuous, <5–10 mW pulsed/duty-cycled; −40/+60 °C operating, +75 °C certified; Ø14 mm, 1 gComponent-level, comparable to VQ500 class[7][12]
Honeywell / City TechnologyCiTipeL CAT seriesCAT25 (P/N 2111B2125), CAT16T90 < 10 s (CH₄); 230 mW max; life > 2 yr; span drift < ±2 %LEL CH₄/yr; poison resistance “some” (HMDS, H₂S)Industrial component pricing (not independently found)[8]
Figaro EngineeringTGS catalytic seriesTGS6812-D00Detects H₂, CH₄, LPG; adsorbent sensor cap reduces cross-sensitivity to organic vapours; “long life,” linear outputComponent-level[23]
Winsen ElectronicsMC seriesMC101, MC105, MC107, MC109, MC119MC105: 1% CH₄ → 20–50 mV, T90 ≤10 s, recovery ≤30 s, 5-yr claimed life, 150 mA @ 2.5 V (≈375 mW), drift ±2 mV/yrLow-cost component tier[10][24][25]
DrägerCatalytic Ex sensing heads (DrägerSensor CatEx)CatEx 125 PR (Poison Resistant), CatEx SR (shock-resistant)0–100 %LEL / 0–100 vol % CH₄; resolution 1 %LEL / 1 vol %; certified hydrocarbon range methane→nonane; “new pellistor composition” explicitly engineered against H₂S and silicone poisoningMining/industrial premium tier[11]
MSA Safety (incl. legacy General Monitors)XCell catalytic bead sensorsXCell cat-bead element (used in Ultima X-series and S4000CH fixed detectors)Integrated fixed-point detector platform; marketed on “sensitivity, reliability and longevity”Industrial fixed-system pricing(unverified)
Contrast: not a pellistorNevadaNano MPS (molecular property spectrometer, MEMS-based)—Marketed explicitly against catalytic bead limitations: no O₂ dependence claimed, no calibration required, 15+ yr life, 1.3–20 mWIncluded here only to frame the pellistor’s chief competitor/successor technology[16]

Only bare-die/element pricing (VQ548ZD-S ≈ US$30.83, LCSC listing [14]) is independently verified against a live distributor. Full fixed-point instrument system pricing, informally cited at US$500–3,000, is market commentary, not a confirmed quote, pending direct vendor confirmation [15].

Open questions

Section titled “Open questions”
  1. Full-instrument pricing is unverified. Only bare-element pricing (~US$30, VQ548ZD-S [14]) is confirmed against a live distributor listing. Fixed-point system pricing (informally US$500–3,000 [15]) needs a direct vendor quote before it goes into any procurement-facing page.
  2. Do current “poison-resistant” catalyst formulations change the hydrogen-specific poisoning profile? Palmisano et al. [19] tested specific, unnamed commercial models circa 2014–2015; Dräger markets a “new pellistor composition” [11] and SGX markets ZD-variant catalysts [3] as improvements, but no independent third-party test data on these current formulations against H₂S/HMDS has been located.
  3. Does pulsed/duty-cycled MEMS operation change long-term drift or poisoning kinetics relative to continuous operation [12]? Not addressed in any source found; plausible in either direction.
  4. No source addresses pellistor performance in a simultaneously high-pressure and elevated-background-temperature environment, representative of a near-surface geothermal/fumarolic H₂ seep as opposed to a cold soil-gas survey or a deep anoxic borehole.
  5. Real-world (non-ISO-26142) HMDS/siloxane long-exposure poisoning kinetics remain an open question even in the primary literature; Palmisano et al. say so explicitly [19]. Treat any Hestia field site near silicone vapor sources (sealants, some drilling fluid additives, personal-care products) as a poisoning risk pending better data.
  6. No vendor datasheet reviewed assumes an O₂-depleted, CO₂/H₂S/CH₄-rich soil-gas matrix; every quoted spec implicitly assumes ambient (20.9%) O₂ air. Independently monitor local O₂ before considering a pellistor near, e.g., a confined-space cellar with possible CO₂ ponding.
  7. Response-factor tables disagree across vendors by roughly 2× for hydrogen specifically (43–107% of methane response across the CAT25 and RAE LEL-1/LEL-2 examples above). Treat k-factors as product-specific lookup values, never a universal physical constant.
  8. The Firth/Jones/Jones (1973) Combustion and Flame volume number needs verification. The primary source consulted here cites Volume 20, Issue 3, pp. 303–311, while one independent search-engine synthesis suggested Volume 21; unresolved against the journal record [9].
  9. Long-term drift mechanism is inferred, not directly confirmed. General heterogeneous-catalysis literature on Pt/Pd sintering, particle migration/coalescence, Ostwald ripening, and single-atom decomposition [26] plausibly underlies the drift documented in every pellistor datasheet surveyed, but no source located here studies these mechanisms in situ on a pellistor bead specifically.
Section titled “Related pages”

References

Section titled “References”
Full reference list (26 entries)
  1. Blackline Safety, “Pellistor (Catalytic Bead) Sensor Technical Notes.” Live URL now redirects to the help-center homepage (link dead as of 2026-07); archived copy: https://web.archive.org/web/20260115205602/https://support.blacklinesafety.com/articles/pellistor-catalytic-bead-sensors
  2. Wikipedia, “Catalytic bead sensor.” https://en.wikipedia.org/wiki/Catalytic_bead_sensor
  3. SGX Sensortech, “VQ500 Series Datasheet,” DS-0223 Issue 2, 10-May-2016. https://www.sgxsensortech.com/content/uploads/2016/06/DS-0223-VQ500-Datasheet-V2.pdf
  4. Clairair Ltd, “A Discussion on Pellistor Gas Sensor Responses.” https://www.clairair.co.uk/wp-content/uploads/2014/05/Pellistor_article.pdf (mirror: https://www.chromatographytoday.com/article/gas-detection/8/clairair-ltd/a-discussion-on-pellistor-gas-sensor-responses/467/download)
  5. Yunusa, Z., Hamidon, M.N., Kaiser, A., Awang, Z., “Gas Sensors: A Review,” Sensors & Transducers, Vol. 168, Issue 4, April 2014, pp. 61–75. https://www.sensorsportal.com/HTML/DIGEST/april_2014/Vol_168/P_1957.pdf
  6. AAPG Wiki, “Mudlogging: gas extraction and monitoring.” https://wiki.aapg.org/Mudlogging:_gas_extraction_and_monitoring
  7. SGX Sensortech, “MP-7217 Datasheet,” DS-0165 Issue 5, 03-May-2016. https://www.sgxsensortech.com/content/uploads/2014/07/DS-0165-MP-7217-Datasheet-V5.pdf
  8. City Technology / Honeywell, “CAT 25 CiTipeL® Combustible Gas Sensor — Product Data Sheet,” Doc. Ref. cat25.indd, ECN I 4643, Issue 4, 14 Nov 2016. https://prod-edam.honeywell.com/content/dam/honeywell-edam/sps/siot/ru-ru/products/sensors/gas-sensors/other-combustibles/documents/sps-ast-citytech-cat-25-datasheet.pdf
  9. Firth, J.G., Jones, A., Jones, T.A., “The principles of the detection of flammable atmospheres by catalytic devices,” Combustion and Flame, Vol. 20, Issue 3, pp. 303–311 (1973), as cited in Clairair [4]; journal record: https://www.sciencedirect.com/science/article/abs/pii/0010218073900217
  10. Winsen Electronics, “MC105 Catalytic Flammable Gas Sensor User’s Manual.” https://www.winsen-sensor.com/d/files/mc105-manual-v1_6b.pdf
  11. Dräger, “Catalytic Ex Sensors” product page. https://www.draeger.com/en_seeur/Products/Catalytic-Ex-Sensors
  12. SGX Sensortech, “Pulsed Power Operation of the MPEL MEMS Pellistor,” Application Note MiCS-MP-7217, 2047 rev 0. https://www.sgxsensortech.com/content/uploads/2014/08/Pulsed-Power-Operation-of-the-MPEL-MEMS-Pellistor.pdf
  13. Environmental Science & Technology Ltd, “LEL Correction Factor Guide” (RAE Systems LEL-1/LEL-2 sensor data) — unverified against a primary RAE Systems source; RAE’s own Technical Note TN-156 shows different hydrogen correction factors under different product-family names, and no “LEL-1”/“LEL-2” cartridge naming was located. https://www.environst.com/lel-correction-factors
  14. LCSC Electronics, “VQ548ZD” product/price listing. https://www.lcsc.com/product-detail/C3659340.html
  15. Mordor Intelligence, “Gas Detectors Market – Companies & Report” (market-size/price commentary; does not support the specific US$500–3,000 fixed-point-system figure attributed to it — that report cites $500–1,500 for portable units and >US$1M for refinery-scale fixed installs). https://www.mordorintelligence.com/industry-reports/gas-detectors-market
  16. NevadaNano, “Catalytic Bead Gas Sensor Limitations.” https://nevadanano.com/catalytic-bead-gas-sensor-limitations/
  17. Industrial Scientific, “LEL Sensors Compared: Catalytic Bead, IR, and MPS™.” https://www.indsci.com/en/blog/catalytic-bead-ir-lel-sensors-which-to-choose
  18. JMS Consultants, “The Requirement of Oxygen for Catalytic Sensors.” https://resources.jmsconsultants.co.uk/news/oxygen-catalytic
  19. Palmisano, V., Weidner, E., Boon-Brett, L., Bonato, C., Harskamp, F., Moretto, P., Post, M.B., Burgess, R., Rivkin, C., Buttner, W.J., “Selectivity and resistance to poisons of commercial hydrogen sensors,” International Journal of Hydrogen Energy, 40 (2015) 11740–11747. DOI: 10.1016/j.ijhydene.2015.02.120
  20. JMS Consultants, “Catalytic Gas Sensor Poisoning: A Silent Threat to Safety.” https://resources.jmsconsultants.co.uk/news/catalytic-sensor-poisoning
  21. Langhi, L., Strand, J., “Exploring natural hydrogen hotspots: a review and soil-gas survey design for identifying seepage,” Geoenergy (2023). DOI: 10.1144/geoenergy2023-014. https://www.lyellcollection.org/doi/full/10.1144/geoenergy2023-014
  22. “Natural hydrogen soil gas emissions near Harvey, Perth Basin: a comparative study of survey methods,” Science and Technology for Energy Transition (STET) (2025). https://www.stet-review.org/articles/stet/full_html/2025/01/stet20250031/stet20250031.html
  23. Figaro Engineering Inc., “TGS6812-D00” product page. https://www.figarosensor.com/product/entry/tgs6812.html
  24. Winsen Electronics, “Catalytic Sensors: Principles, Applications, and Performance.” https://www.winsen-sensor.com/knowledge/catalytic-sensors.html
  25. Winsen Electronics, “Catalytic Bead Hydrogen Sensors: Principles, Features, and Applications.” https://www.winsen-sensor.com/knowledge/catalytic-bead-hydrogen-sensors.html
  26. “Colloidally Engineered Pd and Pt Catalysts Distinguish Surface- and Vapor-Mediated Deactivation Mechanisms,” ACS Catalysis (general Pt/Pd sintering mechanism background, not pellistor-specific). https://pubs.acs.org/doi/10.1021/acscatal.2c04683