UV/EUV electronic transitions & fluorescence — the physics veto

If you shine a UV light on a hydrogen leak, you will not see it glow, and there is no version of “hydrogen UV camera” that a few more years of engineering will produce. Hydrogen’s own light signature — the specific colors of light it absorbs and re-emits — sits in a part of the spectrum called the vacuum ultraviolet, roughly 90–120 nanometers, far short of anything a normal lens, window, or even ordinary air will pass.

Air itself is the problem: at these wavelengths, the oxygen and water vapor already present in every cubic centimeter of atmosphere absorb the light within a fraction of a millimeter, so the light never reaches a detector unless the whole optical path is evacuated or purged with inert gas, the way it is done at particle-accelerator light sources and inside combustion-diagnostics laser labs. That is a laboratory instrument, not a soil-gas probe, a downhole logging tool, or a weatherproof monitoring station.

This page documents that limit precisely — the underlying physics, the handful of real laboratory techniques that do exploit these transitions, and why none of them can be miniaturized into field hardware — so that a plausible-sounding “UV fluorescence hydrogen sensor” idea can be checked against hard numbers rather than re-invented from a combustion-diagnostics paper.

Summary

Hydrogen’s electrons only absorb and emit strongly-allowed UV light at wavelengths shorter than about 115–120 nanometers (the “Lyman” and “Werner” molecular bands) or, for lone hydrogen atoms produced by breaking H₂ apart, at the single atomic Lyman-alpha line at 121.6 nm — both deep in the vacuum ultraviolet (VUV), a band so named because ordinary air absorbs it so completely that it only propagates in a vacuum.

Every technique that tries to exploit this — passive VUV absorption spectroscopy, atomic-hydrogen resonance fluorescence, two-photon laser-induced fluorescence (TALIF), multiphoton ionization (REMPI), or laser-induced breakdown spectroscopy (LIBS) — either requires a hard vacuum or actively purged optical path, a pulsed laser system on an aligned optical table, or both. None of that survives being carried across a field site, lowered down a borehole, or left running unattended for weeks.

The one real-world adjacent technology, handheld LIBS, does produce a visible hydrogen emission line in the field, but it measures total elemental hydrogen from whatever the laser vaporizes — mostly soil moisture and organics — and cannot distinguish that from a genuine H₂ gas anomaly. This is a physical limit, not a research gap, and effort belongs in the vibrational/rovibrational, thermal, electrochemical, and catalytic sensor families covered elsewhere in this wiki.

At a glance — published performance envelope

Detection limit

n/a

Range

n/a

Response t₉₀

n/a

Recovery

n/a

Selectivity

none

Drift

high

Operating T

n/a

Humidity

n/a

Power

500–5,000 W

Unit cost

$12k–$1000k

TRL

2

Main interferents: H₂O, O₂, O3: Hartley band peak 1.15e-17 cm^2 near 255 nm; practical absorption footprint 220-280 nm, N2: transparent above ~110 nm but strong electronic · poisons: LiF window solarization: cumulative VUV-dose-induced color-center formation permanently degrades transmission, LiF hygroscopicity: ambient humidity degrades LiF optical surfaces, Hydrocarbon. Ranges span the published devices/variants covered below — per-claim citations are in the body; data: src/data/modalities/uv-euv-and-fluorescence.yaml.

Mechanism & physics

The native electronic spectrum of H₂ and H

Molecular hydrogen’s lowest electronic transitions from the ground state $X\,^1\Sigma_g^+$ are the Lyman band system ($B\,^1\Sigma_u^+ \leftarrow X\,^1\Sigma_g^+$, band origins below ~112 nm, extending to ~90 nm) and the Werner band system ($C\,^1\Pi_u \leftarrow X\,^1\Sigma_g^+$, ~90–115 nm, interleaved with the Lyman bands) [1][2]. Both are dipole-allowed from the totally symmetric ground state and together carry essentially all of cold H₂’s oscillator strength below ~115 nm.

The rovibronic structure is dense — hundreds of individual $(v',J')\leftarrow(v'',J'')$ lines — each tabulated ab initio (Einstein $A$ coefficients, oscillator strengths) by Abgrall, Roueff, and collaborators [3][4][5]. Photon energy follows the usual relation

$h\nu = E_{\text{upper}}(v',J') - E_{\text{lower}}(v'',J'')$

with $h\nu$ in the 10.8–13.8 eV range for the whole Lyman/Werner manifold (90–115 nm, via $E(\text{eV}) = 1239.84/\lambda(\text{nm})$).

These are bound–bound transitions, and the process governing what happens after absorption is the Solomon process: resonant absorption in a Lyman or Werner line, followed by spontaneous fluorescent decay that lands ~10–15% of the time in the vibrational continuum of the ground electronic state (i.e., photodissociation) and ~85–90% of the time back into a bound rovibrational level of $X$ [1].

This branching ratio is one reason H₂ cannot simply be “excited and read out” cleanly at a fixed, discriminable wavelength: a meaningful fraction of absorbed VUV photons dissociate the molecule instead of fluorescing coherently back.

A second, compounding effect is self-shielding: the Lyman/Werner lines are so numerous and strong that a modest H₂ column density becomes optically thick in its own resonance lines. The interstellar-medium literature, where this was first quantified, reports detectability of diffuse H₂ down to column densities of a few $10^{12}\,\text{cm}^{-2}$, with the strongest lines saturating around $10^{14}$–$10^{15}\,\text{cm}^{-2}$ for a Doppler parameter $b\sim3\,\text{km/s}$ [6] (unverified), following the classic Draine & Bertoldi (1996) parameterization [7].

Translated to a field number: at ambient air density ($N\approx2.5\times10^{19}\,\text{cm}^{-3}$), a soil-gas anomaly of just 1% H₂ along a 1 mm path already presents a column density of order $2.5\times10^{16}\,\text{cm}^{-2}$ — one to two orders of magnitude above saturation. In plain terms: even if VUV light reached a real soil-gas plume, the gas would already be optically thick in its own lines, making quantitative retrieval from an unresolved band hopeless without extremely short path lengths and isolated, weak lines. That is a laboratory spectroscopy exercise, not a field sensor design.

Atomic hydrogen: Lyman-alpha and the Balmer series

Atomic H — which only exists once something has broken the H–H bond (dissociation energy 4.48 eV) via a plasma, flame, electron impact, or UV photon — has its own line spectrum. The transition of greatest relevance is Lyman-alpha, $2p\rightarrow1s$ at 121.567 nm (10.2 eV), the most intense VUV line in the solar spectrum [8].

This is a commonly confused point in casual “hydrogen and UV” discussions: Lyman-alpha is an atomic transition, invisible from intact H₂ gas. Any field scheme that invokes “Lyman-alpha detection of hydrogen” implicitly proposes a dissociation step first, which is a far more elaborate instrument than a passive gas sensor.

The visible-range Balmer series — H-alpha at 656.279 nm ($n{=}3\rightarrow2$), H-beta at 486 nm — is the practically accessible fluorescence signature used in laser diagnostics and LIBS (see Variants, below). H-alpha’s spontaneous-emission (Einstein $A$) coefficient of $4.41\times10^{8}\,\text{s}^{-1}$ makes it the strongest visible hydrogen line and the natural channel for two-photon excitation schemes [9].

Why air is opaque in this window

Even setting H₂ itself aside, the transmission medium — ambient air — is essentially opaque across this spectral region:

• O₂ Schumann–Runge continuum (~130–175 nm): a strong, structureless continuum with peak cross sections “reaching greater than $10^{-17}$ cm²” [10][11].
• O₂ Schumann–Runge bands (~175–204 nm): a rotationally/vibrationally structured system bridging the continuum to near-UV [12][13].
• O₂ Herzberg continuum (200–242 nm): much weaker ($10^{-23}$–$10^{-24}$ cm²) but still dominates stratospheric ozone photochemistry there [13].
• O₃ Hartley band (peak ~255 nm): cross section $1.15\times10^{-17}\,\text{cm}^2$ (consensus/HITRAN value at 253.65 nm), practically relevant to 220–280 nm rather than the 90–175 nm window of primary interest, but still a factor for any UV source/detector operated near its edge [14][15].
• N₂: essentially transparent from ~100 nm out to its own ionization continuum, becoming a strong absorber only below the $B\,^2\Sigma_u^+$ threshold at 66.123 nm — N₂ is not the dominant absorber in the 90–175 nm band; O₂ and H₂O are [16][17].
• H₂O vapor: cross sections of $10^{-17}$–$10^{-18}\,\text{cm}^2$ across 120–180 nm, continuum-dominated above 145 nm [18][19] — and, as discussed under cross-sensitivity below, this is the field-relevant absorber given real humidity swings.

Worked attenuation-length example. Using the O₂ partial density at sea level ($n_{O_2}\approx5.2\times10^{18}\,\text{cm}^{-3}$) and the peak Schumann–Runge continuum cross section of $\gtrsim10^{-17}\,\text{cm}^2$ [10][11]:

$L_{1/e} = \frac{1}{n_{O_2}\,\sigma} \approx \frac{1}{(5.2\times10^{18})(10^{-17})} \approx 0.02\ \text{cm} \approx 190\ \mu\text{m}$

At the peak of the Schumann–Runge continuum, ordinary air attenuates VUV light $1/e$ in roughly the thickness of two sheets of paper. Even at the single known local minimum in the O₂ cross section — a narrow “window” almost exactly at the Lyman-alpha wavelength, absorption coefficient ~0.25–0.30 cm$^{-1}$ ($\sigma\sim1\times10^{-20}\,\text{cm}^2$ per O₂ molecule) [20] — the attenuation length only improves to:

$L_{1/e} \approx \frac{1}{(5.2\times10^{18})(1\times10^{-20})} \approx 19\ \text{cm}$

That 19 cm best case is why solar Lyman-alpha penetrates only to the mesosphere (~70–75 km) rather than the ground [20]. No realistic field optical path — soil probe tube, downhole sonde, open-path standoff, tens of centimeters to meters — works without hard vacuum or inert purge over its entire length.

Ionization energetics: a second, independent wall

Setting fluorescence aside, one might instead ask about VUV/EUV photoionization, as in a photoionization detector (PID). H₂’s first ionization potential is 15.43 eV (= 4.48 eV dissociation + 13.6 eV H ionization − 2.65 eV H₂⁺ binding energy) [21], corresponding to ~80.3 nm — deep VUV/soft-EUV, shortward of both the LiF (105 nm) and MgF₂ (115 nm) cutoffs discussed below, requiring windowless or all-reflective vacuum optics.

This is harder than the ubiquitous 10.6 eV krypton-lamp PIDs used for VOC field screening: their lamp photon energy (10.6 eV, 117 nm) sits below H₂’s ionization threshold, so commercial 10.6 eV (and even the less common 11.7 eV) PID lamps cannot ionize H₂ at all [22]. That is a hard energetic gap of more than 4 eV, not a calibration issue. It is worth stating explicitly, because PIDs are the most common handheld “UV gas sensor” a field team already owns.

Variants & subtypes

This limit covers a family of related but distinct techniques, each failing for a related but slightly different reason.

Variant	Mechanism & typical setting	Why it can’t be a field H₂ sensor
Passive VUV absorption spectroscopy (Lyman/Werner, 90–115 nm)	Broadband/tunable VUV source through a sample cell, absorption read directly — as astronomers measure interstellar H₂ column density. Setting: synchrotron VUV beamlines, e.g. DESIRS at SOLEIL (5–40 eV, resolving power >100,000) [23][24]	Requires a windowless or LiF/MgF₂-windowed, evacuated or inert-purged optical path end to end
Atomic-H resonance absorption/fluorescence at Lyman-alpha (121.567 nm)	A Lyman-alpha source (H/D discharge lamp or frequency-multiplied laser) probes pre-existing atomic H. Setting: astrophysics, fusion/plasma diagnostics, photodissociation dynamics [8][25]	Only sees atomic H, which does not exist in an intact H₂ soil-gas seep without a prior dissociation step
TALIF (two-photon absorption laser-induced fluorescence of atomic H)	Two ~205 nm photons drive $1s\rightarrow3s/3d$; decay emits the 656.3 nm Balmer-alpha photon in 15–20 ns, read with a gated PMT/ICCD. Setting: combustion and plasma diagnostics — the workhorse technique for atomic H [26][27]	Needs a flame/plasma/dissociation source plus a pulsed deep-UV laser table; fs variants exist to suppress photolytic interference [26][28][29]
H₂ two-photon E,F–X excitation (~202 nm, or 193.26 nm ArF)	Two-photon excitation of intact H₂ to $E,F\,^1\Sigma_g^+$, fluorescence read on the $E,F\rightarrow B$ system; a landmark ArF-193 nm variant imaged H₂ fuel and hot O₂ simultaneously in a turbulent diffusion flame [30][31][32][33][34]	Demonstrated only in flames/imaging; not a calibrated absolute-concentration instrument
REMPI (resonance-enhanced multiphoton ionization)	Resonant multiphoton excitation, then ionization detected as ion/electron current rather than light; demonstrated for atomic H (2-photon 1S–2S) and H₂ directly via Rydberg intermediates, described as “ultrasensitive” [35][37]	Requires high/ultra-high vacuum for clean ion collection — stricter than fluorescence-based TALIF
VUV-LIF in purged/vacuum cells	Catch-all: any scheme exciting H₂/H with a genuinely VUV (<180 nm) photon and collecting fluorescence, laser/synchrotron/lamp-based	Requires hard vacuum or active inert-gas purge (dry N₂ to ~115 nm, Ar/He shorter) along the entire optical path [38][39]
LIBS (laser-induced breakdown spectroscopy)	A pulsed laser (typically Nd:YAG, ~1064 nm) focused onto a sample creates a ~10,000–20,000 K microplasma that atomizes and excites everything in the focal volume; H-alpha (656.3 nm) is prominent and early-appearing as it cools [40][41]. The one genuinely field-deployable technique here	Measures total elemental hydrogen (moisture + organics + any H₂), destructively and pointwise; Hα/N or Hα/O ratios are in fact used deliberately as a humidity proxy (linearity >99.9%) — the opposite of H₂-selective sensing [42][43][44]

Performance envelope

Laboratory capability, not field specification

None of these techniques exists as a packaged H₂ field sensor, so every figure below is a demonstrated laboratory capability in a controlled cell, flame, or vacuum system — not a calibrated field-instrument specification. That absence is itself the finding, not a gap in the literature search.

Parameter	Lyman/Werner absorption (lab/synchrotron)	Atomic-H TALIF (205 nm → 656 nm)	H₂ two-photon E,F–X (~202 nm)	REMPI (H, 2+1 @ 1S–2S)	Field LIBS (Hα 656.3 nm, total-H)
Wavelength(s) needed	90–115 nm excitation [1][2]	205 nm ×2 pump, 656.3 nm probe [26][27]	~201.6–202.3 nm ×2, or 193.26 nm ×2 [30][32]	~243 nm ×2 (1S–2S two-photon) [35]	1064 nm ablation; 190–950 nm spectrometer readout [45]
Detection quantity	H₂ column density/concentration	Atomic H number density	Molecular H₂ number density (relative, imaging)	Atomic H (or D, T), highly state-selective	Total elemental H — H₂O + organics + any H₂
Demonstrated sensitivity	Column densities down to a few $10^{12}\,\text{cm}^{-2}$ via strong resonance lines [6] (unverified)	Quantifies flame/plasma H densities ~$10^{13}$–$10^{16}\,\text{cm}^{-3}$ (context-dependent, not a fixed LOD) [26][46]	Relative concentration imaging in flames; not an absolute-ppm instrument [32][34]	High — single-quantum-state selectivity, described as “ultrasensitive” in vacuum [37]	Elemental LOD is matrix-dependent; handheld units report H among detectable elements H→U [45]
Environment required	Vacuum or He/Ar purge; synchrotron/windowless source [23][24]	Combustion/plasma chamber or purged/vacuum cell; pulsed laser table [26][28]	Flame/discharge cell, optical access; pulsed tunable UV laser [32][34]	High/ultra-high vacuum, ion optics, MS or biased collector [35][37]	Ambient air OK — the one open-air-capable technique — but destructive, point sampling [40][45]
Response/recovery	N/A — single-shot spectral scan, not continuous	Single shot (ns–fs); rep-rate limited (kHz-class fs systems) [28][29]	Single-shot planar image possible [32]	Per-shot with ns dye laser [35]	~1 s per shot/read cycle in handheld units [45]
Selectivity vs. interferents	High (line-resolved), but H₂ must already be isolated in a cell	High for atomic H; “H₂ selectivity” is indirect, via dissociation	Molecule-specific, only inside a prepared cell/flame	Extremely high (isotope- and state-selective)	None — total H regardless of chemical form [42][43]
T/P operating range	Room temp typical (90 K/295 K O₂ data exists) [10]	Flame temps (1500–2500+ K) or plasma; not ambient soil-gas T/P	Flame temperatures	Cold beams to room temp	Ambient field T; optics rated roughly −10 to 45 °C (vendor-dependent)
Power/footprint	Synchrotron, or optical-table laser + vacuum pumps, kW-class facility power	Optical table: pulsed dye/OPO or fs laser, hundreds of W	Optical table, excimer or dye laser	Optical table + vacuum system + ion detection electronics	Handheld, internal battery, ~10s of W, self-contained [45]
Cost class	Synchrotron beamtime (allocated, not purchasable) or $500k+ lab VUV spectrometer + source [24][47]	$100k–$500k+ (fs laser, optics, ICCD/PMT, vacuum/purge cell)	$100k–$300k+ (tunable excimer/dye laser + optics)	$200k–$1M+ (vacuum chamber, ion optics, laser)	$12,000–$55,000 for a handheld/portable unit [45][48][49]
TRL for field H₂ gas sensing specifically	2–3 (lab technique, no field H₂-sensor product)	2–3 (combustion/plasma diagnostic, no field product)	2 (demonstrated only in flames/imaging)	2 (fundamental-physics/vacuum diagnostic)	8–9 as a LIBS elemental analyzer product, but 1–2 as a dedicated H₂ gas sensor — it isn’t one

Optics and instrumentation constraints

Below ~105–115 nm essentially all conventional refractive optics are unusable.

• LiF is the shortest-cutoff common VUV window material (~105 nm, ~40% transmission at 121 nm) but is hygroscopic and solarizes (VUV-induced color-center darkening) under cumulative dose [50][51][52]; MgF₂ cuts on around 115–120 nm and is more chemically robust, also used as a protective mirror overcoat extending usable reflectance to ~115 nm [53][54][55][56][57][58].
• Below ~40 nm even normal-incidence mirrors lose efficiency, forcing grazing-incidence designs — standard synchrotron/lithography engineering, foreign to portable gas-sensor practice.
• Standard silica optical fiber is opaque below ~180–200 nm, so there is no fiber-based way to pipe VUV light anywhere, including down a borehole.
• On the detection side, solar-blind AlGaN photodiodes (cutoff tunable 200–280 nm, dark current below $10^{-6}\,\text{A/cm}^2$) [59][60][61][62][63] and Cs-Te photocathode PMTs (~160–320 nm response) [64] are genuinely solar-blind, but true sub-160 nm VUV photocathodes require vacuum operation and are components inside spectrometers, not standalone gas sensors [65].
• Sources run from deuterium arc lamps (115–400 nm) [66][67] and ArF/F₂ excimer lasers (193/157 nm) [68][69][70][71] to tabletop high-harmonic generation (10–60 nm, femtosecond-driven) [72][73][74] and laser-produced tin-plasma EUV sources (13.5 nm, lithography-scale) [75][76][77] — every one an optical-table-to-building-scale instrument, not a component that fits in a sensor housing.

Figures of merit

The comparison axes defined on the Figures of merit page — LOD, sensitivity, selectivity, response time, drift — describe a device, and this page has established that no device exists here for them to describe (see Performance envelope, above). What decides feasibility instead is a short list of physical constants that fix the outcome before any instrument design begins; see Concentration units for how the column-density and number-density units below relate to the ppm/vol% conventions used elsewhere on this wiki.

Figures-of-merit table for this modality (6 rows)

Figure of merit	Symbol / value	What it constrains	Value here	How to read it
Transition energy / wavelength	$h\nu = 1239.84/\lambda(\text{nm})$	Which photon source and optics would even need to reach the transition	H₂ Lyman ($B\leftarrow X$) & Werner ($C\leftarrow X$): 90–115 nm, 10.8–13.8 eV [1][2]. Atomic H Lyman-α ($2p\to1s$): 121.567 nm, 10.2 eV [8]. TALIF: 205 nm ×2-photon pump → 656.3 nm H$_\alpha$ probe [26][27]	All three sit shortward of, or right at the edge of, every conventional optical material’s transmission window (see Optics cutoff row) — a property of hydrogen’s electronic structure, not of any instrument’s design
Oscillator strength / transition probability (Einstein $A$)	tabulated per line, not a single constant	How strongly a transition can be driven, and how cleanly it decays back out	H₂ Lyman/Werner manifold: ab initio Einstein $A$ coefficients and oscillator strengths tabulated line-by-line [3][4][5]; Solomon-process branching ~85–90% back to bound $X$, ~10–15% photodissociative [1]. Atomic H$_\alpha$: $A=4.41\times10^{8}\,\text{s}^{-1}$ [9]	Even successful excitation doesn’t return a clean, single-wavelength readout — a large fraction of absorbed photons dissociate the molecule instead of fluorescing coherently back
Air attenuation length, $L_{1/e}=1/(n\sigma)$	derived from O₂ Schumann–Runge cross section	Whether the signal can travel any distance through ordinary air at all	Worst case (Schumann–Runge continuum peak, ~145 nm): $L_{1/e}\approx0.02\ \text{cm}$ (≈190 μm) [10][11]. Best case anywhere in the 90–200 nm band (narrow O₂ transmission window, coincident with Lyman-α): $L_{1/e}\approx19\ \text{cm}$ [20]	The figure of merit that actually forecloses field use before selectivity or LOD ever enter the comparison — 19 cm best-case is shorter than a single soil-gas probe tube
Optics cutoff wavelength	material transmission edge	Whether any refractive window or lens can be used at all	LiF: ~105 nm (~40% transmission at 121 nm; hygroscopic, solarizes under cumulative VUV dose) [50][51][52]. MgF₂: ~115–120 nm (more chemically robust; also used as a mirror overcoat) [53][54][55]	Below ~105 nm no common window material transmits at all; below ~40 nm even mirrors need grazing-incidence designs — these are fixed material limits, not design choices
Photoionization energetics	H₂ ionization potential	Whether a simpler photoionization route (rather than fluorescence) exists	H₂ IP = 15.43 eV (~80.3 nm) [21]. Common 10.6 eV (and 11.7 eV) PID lamps fall 4+ eV short and cannot ionize H₂ at all [22]	Rules out the one “UV gas sensor” a field team is most likely to already own
Lab LOD, TALIF / VUV-LIF (the variants that do work, in vacuum/purge)	context-dependent, not a fixed spec	What the one demonstrably-working measurement family can see, and what it costs to see it	H₂ column density via Lyman/Werner absorption: down to a few $10^{12}\,\text{cm}^{-2}$ (unverified) [6]. Atomic-H TALIF: $10^{13}$–$10^{16}\,\text{cm}^{-3}$ flame/plasma densities, context-dependent, not a fixed LOD [26][46]	Requires hard vacuum or He/Ar/N₂ purge along the entire optical path, plus a pulsed UV/VUV laser table [23][24][26][28][38][39] — every number in this row comes with that requirement

What dominates

None of the six rows above is an engineering spec that better electronics, cleverer signal processing, or more R&D funding could move — each is a fixed constant of hydrogen’s electronic structure or of Earth’s atmospheric composition. The attenuation-length row is the one that decides the outcome: at 190 μm to 19 cm best-to-worst case, it rules out any field optical path — soil probe, borehole, open path — before selectivity, LOD, or cost class become relevant.

Wherever a real laboratory number does exist (TALIF, VUV-LIF, synchrotron absorption spectroscopy), it comes attached to “requires vacuum or inert purge, plus a pulsed laser table”; that requirement, not the number next to it, is what determines field deployability, and it is the same for every technique on this page: lab-only.

Environmental covariance & cross-sensitivity

Because no packaged field sensor exists on this principle, “cross-sensitivity” here means why every field-relevant gas and condition makes the underlying physics worse, not a calibration table for an existing instrument.

• Humidity/water vapor: H₂O is one of the strongest VUV absorbers in exactly the band of interest, $10^{-17}$–$10^{-18}\,\text{cm}^2$ across 120–188 nm [18][19] — comparable to or exceeding the O₂ Schumann–Runge continuum. This compounds the problem, because geologic H₂ seeps are frequently associated with wet or water-saturated ground (serpentinization and iron-oxide reduction reactions that generate H₂ typically occur in water-bearing rock), so the worst-case optical environment co-occurs with the target signal.
• CO₂: transparent ($<10^{-22}\,\text{cm}^2$) above ~200 nm at room temperature, but a significant VUV absorber at shorter wavelengths and elevated temperature — relevant only for a hypothetical high-T downhole deployment [78][79].
• CH₄: solid-phase (ice-analog) VUV absorption cross section $5.7^{+0.5}_{-1.1}\times10^{-18}\,\text{cm}^2$ averaged over 120–160 nm [79]; gas-phase values differ but confirm CH₄ absorbs in exactly H₂’s own Lyman/Werner window. Since CH₄ commonly co-occurs with geologic H₂, it would spectrally interfere rather than merely dilute.
• H₂S: a major absorption band spans 40–160 nm, secondary at 160–250 nm [79] — again overlapping the H₂ Lyman/Werner window and the O₂ Schumann–Runge continuum.
• He: inert and non-absorbing in most of this range, with a resonance line at 58.4 nm (24.6 eV ionization energy) [80] used as a source line in photoionization spectroscopy. He is a common co-tracer in geologic H₂ seeps, but that is irrelevant to this failure mode — it matters only as a potential TALIF/REMPI calibration gas, not a field interferent.
• O₂ itself: simultaneously not required by any of these mechanisms (they are not combustion-based) and the dominant interferent absorber across nearly the whole range. O₂ is not a “poison” here in the catalytic-sensor sense; the atmospheric medium itself is the dominant absorber.
• Pressure: cross sections are per-molecule and pressure-independent, but optical depth scales with number density — downhole (elevated P) makes the attenuation problem proportionally worse.
• Temperature: the Schumann–Runge continuum has a weak, measured temperature dependence (90 K and 295 K data exist) [10], second-order next to baseline opacity; downhole heat mainly threatens source/detector survivability, not transparency.
• Optics degradation (a field-reliability failure mode, not an analyte effect): LiF windows are hygroscopic and solarize under cumulative VUV dose [51]; MgF₂ is more robust but demands extreme surface cleanliness, since hydrocarbon contamination films absorb strongly in the VUV. Both are irrelevant to ordinary gas sensors but first-order problems for any VUV path exposed to real dust, moisture, and soil-gas organics.

Use for geologic H₂

Field verdict

Not viable at any point on the soil-gas, downhole, or continuous-monitoring spectrum, and not because the technology is immature. Every UV/EUV/fluorescence technique that can see H₂’s or H’s native electronic transitions requires a hard vacuum or actively purged inert-gas optical path, a pulsed laser system on an aligned optical table, or both. That is incompatible with a backpack-carried soil probe, a downhole sonde, or an unattended weatherproof station.

The one field-real adjacent technology, handheld LIBS, does emit a visible hydrogen line, but measures total elemental hydrogen (dominated by soil moisture), not a gas-phase H₂ anomaly. Use this section’s reasoning with anyone proposing a “UV fluorescence H₂ sensor”: the answer is no, backed by the atmospheric-opacity and vacuum-requirement numbers above, not a TRL argument that more funding would resolve.

Soil-gas surveys: not viable. A soil-gas probe needs to be rugged, weatherproof, low-power, and capable of continuous sampling or a walking traverse. Every technique above requires a hard vacuum or purge, a pulsed UV/VUV laser on an optical table, or both — none of it survives a backpack, let alone weeks in a weatherproof enclosure.

Even the least exotic candidate, LIBS, measures total elemental hydrogen with no way to attribute the signal to a genuine H₂ gas-phase anomaly against the ever-present, often larger, soil-moisture background — precisely the worst-case correlation flagged above (wet ground and H₂ seeps co-occurring).

Downhole/borehole logging: not viable, for compounding reasons. Downhole tools must tolerate high temperature and pressure and strongly prefer passive, fiber-optic readout with no downhole electronics. Every technique here needs a pulsed laser and fast electronic gating at the measurement point, or a delicate vacuum-sealed cell with fragile, hygroscopic/solarizing LiF or MgF₂ windows rated for benign lab environments — nothing here is rated for wellbore shock, vibration, or elevated T/P.

Nor is there a fiber-based transport solution: standard silica fiber is opaque below ~180–200 nm, so “send VUV light down a fiber” has no practical answer (contrast with Fiber-optic sensors, which operate at wavelengths where silica is transparent).

Continuous monitoring stations: not viable for the same reasons, plus an added maintenance burden — unattended pulsed-laser operation and an actively maintained vacuum/purge system is a capital load utterly disproportionate to the ppm-to-percent range relevant to H₂ exploration (atmospheric background ~0.53 ppm, anomalies from ~1 ppm to percent level). A single misaligned mirror or solarized LiF window silently zeroes the entire signal chain, with no simple field diagnostic to flag it, which is the opposite of the fail-safe behavior wanted in a remote station.

Field-portable LIBS: these units are real, ruggedized, handheld instruments that do observe an Hα line in real time. A colleague will eventually ask whether we can just point the LIBS gun at the ground.

We can, but it will report the hydrogen content of whatever surface film the laser ablates, dominated by moisture and organics, which is not the same quantity as soil-gas H₂ concentration and cannot be deconvolved into one without a separate calibrated moisture correction the instrument doesn’t provide. LIBS is best treated as a humidity/bulk-elemental tool that happens to also emit at 656.3 nm, not a hydrogen-exploration sensor.

Summary: this is a consequence of where H₂’s electronic structure and terrestrial atmospheric chemistry both happen to sit on the wavelength axis, not a maturity gap R&D funding will close. Budget and attention belong with the vibrational/rovibrational (IR absorption and proxies), thermal (Thermal conductivity & acoustic), electrochemical (Electrochemical cells), and catalytic (Catalytic pellistors) sensor families covered elsewhere in this wiki — see also Why H₂ is hard for how this fits the broader pattern.

Vendors & products

No vendor sells a “UV fluorescence hydrogen gas sensor” for field use; the physics rules it out. The landscape instead consists of component suppliers for the lab techniques above, plus the one adjacent, commercially real technology (handheld LIBS) a team might actually procure.

Category	Vendor / product	Key specs	Price class
VUV windows (LiF)	ISP Optics/Edmund Optics, Crystran, Del Mar Photonics, DM Photonics — “VUV grade” LiF	Transmission from 105 nm; ~40% at 121 nm; solarizes with cumulative VUV dose [50][51][52]	$100s–$1,000s per window
VUV/UV windows (MgF₂)	Firebird Optics, Crystran, Thorlabs, ANCORP	Transmission to ~115–120 nm; >50% at 120 nm; UHV/CF/KF viewport options [53][54][55]	$100s–$1,000s
Deuterium VUV lamps	McPherson Model 632/634; Hamamatsu L1835, L7293/L13301, S2D2	115 nm–visible (MgF₂ window), 30 W, ~2,000 h life; photoionization-source variants available [66][67][81][82]	Lab component pricing (contact vendor)
VUV monochromator/spectrometer	McPherson Model 234/302	200 mm focal length, f/4.5, 1200 g/mm aberration-corrected grating, MCP/CCD options [83][84]	Research-instrument pricing (unverified — contact vendor)
Synchrotron VUV beamline (facility, not a product)	SOLEIL DESIRS	5–40 eV, resolving power to ~250,000 [23][24]	Beamtime allocation (peer-reviewed proposal only)
ArF excimer laser	Coherent LEAP UV series; Gigaphoton GT40A (45 W); BeamQ Laser 193/248/308/351 nm	193 nm; mJ-class pulses; kHz rep rates in lithography-grade units [69][85][86]	~$195,000 for a compact research system (unverified); lithography-grade higher
Solar-blind photodetectors	AlGaN p-i-n photodiodes/APDs (no dominant off-the-shelf SKU identified)	Cutoff tunable 200–280 nm; dark current <$10^{-6}\,\text{A/cm}^2$ [59][60][61]	Component pricing, varies widely
Solar-blind PMT	Hamamatsu R821 (Cs-Te photocathode)	Spectral response ~160–320 nm [64]	Component pricing (datasheet, price on request)
Handheld LIBS (Hα-capable, total-H)	SciAps Z-901/Z-902C+/Z-903	190–950 nm range (Z-903); PULSAR laser 5–6 mJ/pulse, up to 50 Hz; detects H among other elements [45][48]	$4,755–$15,090 (Z-200C+/Z-902C+), by configuration (price unverified)
Handheld/portable LIBS (general)	Malvern Panalytical, Thermo Fisher Niton Apollo, Torontech, Wisens	General elemental/alloy ID and soil screening	~$40,000–$55,000 for specialized units (unverified) [87][88][49]

Note

Photoionization detector (PID) manufacturers — Ion Science, RAE Systems/Honeywell, Dräger — are deliberately not listed above as H₂ sensors. As established in Mechanism & physics, standard 10.6 eV PID lamps cannot ionize H₂ (IP 15.43 eV) at all, so despite being the most common handheld “UV gas sensor” in the field-safety world, PIDs are not a candidate H₂ technology [22]. This is mentioned to close off that line of inquiry, not to recommend the product category.

Open questions

1. Self-shielding coefficients: the fitted constants in the Draine & Bertoldi (1996) self-shielding formula were not independently re-verified from the primary paper here; confirm before citing exact coefficients elsewhere in the wiki [7].
2. Wavelength-resolved O₂ cross-section table: replace the single-point ”$\gtrsim10^{-17}\,\text{cm}^2$ near 145 nm” figure with a full digitized Schumann–Runge curve for the wiki’s reference-data appendix.
3. Commercial solar-blind AlGaN photodiode SKU: strong device-physics literature exists but no single dominant off-the-shelf part number was found — worth pinning down for a component-level BOM if a UV-flame/arc detection need arises elsewhere (not H₂ sensing per se).
4. The MDPI “Sensing Hydrogen Seeps” review (HTTP 403 on fetch during research) is likely the most on-topic soil-gas H₂ sensing survey and worth cross-referencing against this page’s verdict.
5. Fiber VUV transmission cutoff: get an actual attenuation-vs-wavelength curve for silica fiber below ~180–200 nm if the downhole-transport question ever needs a harder number than the qualitative claim here.
6. “Why does IR work but UV doesn’t?”: colleagues will reasonably ask why IR absorption and proxies isn’t subject to the same veto. Short answer: H₂ is homonuclear and has no IR-active fundamental vibrational mode either, so NDIR/TDLAS approaches for H₂ rely on proxies or weak quadrupole/collision-induced absorption — adjacent but out of scope here, and treated on that page.
7. LiF transmission at 121 nm: vendor figures range from ~40% to “high transparency from 105 nm”; upgrade to a peer-reviewed optical-constants reference (e.g., Palik’s Handbook) if available in-house.

Related pages

• Why H₂ is hard
• Raman spectroscopy
• Enhanced Raman
• IR absorption and proxies
• Fiber-optic sensors
• MOX sensors
• Thermal conductivity & acoustic
• Catalytic pellistors
• Electrochemical cells
• Geologic context

References

Full reference list (88 entries)

1. IOPscience, “Rovibrationally Resolved Direct Photodissociation Through the Lyman and Werner Transitions of H2 for FUV/X-ray-Irradiated Environments,” ApJ 746, 78 (2012). https://iopscience.iop.org/article/10.1088/0004-637X/746/1/78

2. ResearchGate search-result summary of H2 Lyman/Werner band literature (band origins below 112 nm; B¹Σu⁺−X¹Σg⁺ and C¹Πu−X¹Σg⁺ assignments), incl. “The Lyman and Werner bands of H2.” https://www.researchgate.net/publication/237360185_The_Lyman_and_Werner_bands_of_H2

3. Abgrall, Roueff, Launay, Roncin, Subtil, “Wavelengths, oscillator strengths and transition probabilities of the H2 molecule for the Lyman and Werner systems,” A&AS 79, 313 (1989). https://ui.adsabs.harvard.edu/abs/1989A&AS…79..313A/abstract

4. “Highly accurate transition frequencies in the H2 Werner and Lyman absorption bands,” (ResearchGate PDF). https://www.researchgate.net/publication/235791099_Highly_accurate_transition_frequencies_in_the_H2_Werner_and_Lyman_absorption_bands

5. “Highly accurate transition frequencies in the H2 Lyman and Werner absorption bands” (Academia.edu PDF). https://www.academia.edu/19365646/Highly_accurate_transition_frequencies_in_the_H_2_Lyman_and_Werner_absorption_bands

6. Search-result synthesis on H2 self-shielding sensitivity (column densities from a few 10¹² cm⁻² up to saturation at 10¹⁴–10¹⁵ cm⁻²), drawn from interstellar H2/H I absorption-line literature incl. IOPscience “Interstellar H I and H2 in the Magellanic Clouds,” ApJ 745, 173 (2012). https://iopscience.iop.org/article/10.1088/0004-637X/745/2/173 (does not support the claim; the cited Welty et al. paper reports H2 column densities of 0.3–90×10¹⁹ cm⁻² in the Magellanic Clouds, four to five orders of magnitude above the range attributed to it here).

7. Draine, B.T. & Bertoldi, F., “Structure of Stationary Photodissociation Fronts,” ApJ 468, 269 (1996); arXiv:astro-ph/9603032. https://arxiv.org/abs/astro-ph/9603032; see also citing/using discussion in Wolcott-Green, Haiman & Bryan, “Photodissociation of H2 in Protogalaxies,” MNRAS 418, 838 (2011), arXiv:1106.3523. https://arxiv.org/abs/1106.3523 (Note: the specific numeric fitting-formula coefficients commonly attributed to Draine & Bertoldi 1996 were not independently re-verified from the primary text in this session and are therefore not reproduced here; the qualitative self-shielding behavior is asserted with confidence, but the specific column-density thresholds attributed to [6] are unverified — see the note on reference [6].)

8. Wikipedia, “Lyman-alpha,” (121.567 nm, 10.2 eV, strong atmospheric absorption, requires vacuum spectroscopy or satellite observation). https://en.wikipedia.org/wiki/Lyman-alpha

9. Search-result summary citing Balmer series Hα Einstein A coefficient (4.41×10⁸ s⁻¹) and 656.3 nm wavelength, per Wikipedia “Balmer series.” https://en.wikipedia.org/wiki/Balmer_series

10. “Absolute absorption cross-section measurements of Schumann–Runge continuum of O2 at 90 and 295 K,” J. Mol. Spectrosc. (2005); ADS record 2005JMoSp.229..238Y. https://ui.adsabs.harvard.edu/abs/2005JMoSp.229..238Y; ScienceDirect abstract: https://www.sciencedirect.com/science/article/abs/pii/S0022285204002772

11. Gies et al., “The Schumann-Runge continuum of oxygen at wavelengths greater than 175 nm,” JGR 87, A10, 8307 (1982) — cross section context “>10⁻¹⁷ cm², comparable to the ionization continuum.” https://ui.adsabs.harvard.edu/abs/1982JGR…87.8307G/abstract

12. ScienceDirect Topics, “Schumann-Runge Band” overview (175–204 nm structured band system). https://www.sciencedirect.com/topics/earth-and-planetary-sciences/schumann-runge-band

13. Search-result synthesis on O2 Herzberg continuum (~200–242 nm, weak but stratospherically important) and Schumann-Runge bands (175–204 nm), general VUV atmospheric absorption context.

14. Ozone Hartley band peak cross section values (1.1476×10⁻¹⁷ cm² Hearn 1961; 1.145×10⁻¹⁷ cm² HITRAN 2016 at 253.65 nm; 1.1329×10⁻¹⁷ cm² consensus): IOPscience, “Recommendation of a consensus value of the ozone absorption cross-section at 253.65 nm,” Metrologia (2019). https://iopscience.iop.org/article/10.1088/1681-7575/ab0bdd; AMT, “Accurate measurements of ozone absorption cross-sections in the Hartley band” (2015). https://amt.copernicus.org/articles/8/1245/2015/

15. Search-result synthesis: ozone practical absorption footprint 220–280 nm in ambient air; general UV/ozone spectroscopy sources.

16. Search-result synthesis (VUV Wikipedia-sourced): “VUV spectrum is in the wavelength range of 10–200 nm … strongly absorbed by oxygen in the air … can only propagate in a vacuum”; nitrogen purge extends usable range to ~115 nm.

17. Search-result synthesis on N2 VUV absorption structure: transparent down to ~100–110 nm, ionization continuum dominates shortward of B²Σu⁺ threshold at 66.123 nm, per ResearchGate figure “VUV absorption cross sections of N2 and CH4.” https://www.researchgate.net/figure/VUV-absorption-cross-sections-of-N2-blue-and-CH4-green-and-the-VUV-solar-spectrum-at_fig1_307613202

18. “Absorption cross section measurements of water vapor in the wavelength region 120 to 188 nm,” Chem. Phys. 211, 387 (1996); ADS record. https://ui.adsabs.harvard.edu/abs/1996CP…211..387Y/abstract

19. Search-result synthesis on H2O VUV cross section magnitude (10⁻¹⁷–10⁻¹⁸ cm², 1200–1800 Å range) and continuum/structure transition near 145 nm.

20. Ogawa, “Absorption coefficients of O2 at the Lyman-alpha line and its vicinity,” JGR 73, 21, 6759 (1968). https://agupubs.onlinelibrary.wiley.com/doi/abs/10.1029/JA073i021p06759; search-result synthesis on the O2 transmission-window minimum near 1216 Å and mesospheric Lyman-alpha penetration to 70–75 km via cross sections ≤10⁻²¹–10⁻²⁰ cm².

21. Search-result synthesis (PhysicsForums-sourced derivation) of H2 first ionization energy = 15.43 eV via dissociation (4.48 eV) + H ionization (13.6 eV) − H2⁺ binding energy (2.65 eV). https://www.physicsforums.com/threads/ionization-energy-of-the-dihydrogen.594160/

22. TSI Inc., “Photo-Ionization Detection (PID) Technology,” Application Note TSI-147 (10.6 eV krypton lamp principle; major/minor air components have IP >12.0 eV so don’t interfere with VOC PID measurement). https://tsi.com/getmedia/c2814d77-a757-4269-9682-ff6fab17ed2b/TSI-147_Photo_Ionization_Detection_Technology-A4?ext=.pdf; Wikipedia, “Photoionization detector.” https://en.wikipedia.org/wiki/Photoionization_detector

23. Nahon et al., “DESIRS: a state-of-the-art VUV beamline featuring high resolution and variable polarization for spectroscopy and dichroism at SOLEIL,” J. Synchrotron Rad. (2012). PubMed record: https://pubmed.ncbi.nlm.nih.gov/22713882/

24. “The high-resolution absorption spectroscopy branch on the VUV beamline DESIRS at SOLEIL,” J. Synchrotron Rad. (2016). https://journals.iucr.org/s/issues/2016/04/00/ve5054/

25. ScienceDirect, “Photodissociation dynamics of water at Lyman alpha (121.6 nm).” https://www.sciencedirect.com/science/article/abs/pii/S000926140700379X

26. Search-result synthesis on two-photon 205 nm TALIF of atomic H with 656 nm Balmer-alpha fluorescence detection, incl. “Strategy of interference-free atomic hydrogen detection in flames using femtosecond multi-photon laser-induced fluorescence,” Int. J. Hydrogen Energy (2016). https://www.sciencedirect.com/science/article/abs/pii/S0360319916318183

27. “Two-Photon 205-nm Excitation of Atomic Hydrogen in Flames: Stimulated Emission, Time-Resolved Fluorescence, and Doppler-Free Excitation,” in Twenty-Fifth Symposium (International) on Combustion. https://www.sciencedirect.com/science/article/abs/pii/B978012251930750039X

28. “Femtosecond two-photon LIF imaging of atomic hydrogen in high-pressure methane-air flames,” Proc. Combust. Inst. (2022). https://www.sciencedirect.com/science/article/abs/pii/S1540748922004217

29. “Femtosecond two-photon laser-induced fluorescence imaging of atomic hydrogen in a laminar methane–air flame assisted by nanosecond repetitively pulsed discharges,” Plasma Sources Sci. Technol. (2020). https://iopscience.iop.org/article/10.1088/1361-6595/ab9234

30. Search-result synthesis on H2 two-photon 201–202 nm E,F–X excitation spectra (peaks at 201.64, 201.75, 201.98, 202.31 nm).

31. “Laser-induced fluorescence of hydrogen via the E,F¹Σg⁺→B¹Σu⁺ transition: Rotational-state-dependent collisional quenching,” Chem. Phys. Lett. (2011). https://www.sciencedirect.com/science/article/abs/pii/S0009261411001503

32. “Two-dimensional imaging of molecular hydrogen in H2–air diffusion flames using two-photon laser-induced fluorescence,” Opt. Lett. 16, 9, 660 (1991). https://opg.optica.org/ol/abstract.cfm?uri=ol-16-9-660

33. NASA Technical Reports Server record of the same paper. https://ntrs.nasa.gov/citations/19910049746

34. PubMed record of the same paper. https://pubmed.ncbi.nlm.nih.gov/19774030/

35. Wikipedia, “Resonance-enhanced multiphoton ionization.” https://en.wikipedia.org/wiki/Resonance-enhanced_multiphoton_ionization

36. “Determining Absolute Neutrino Mass using Quantum Technologies,” arXiv:2412.06338 (does not support the claim; this paper concerns cyclotron radiation emission spectroscopy and Rydberg-atom quantum magnetometry for neutrino-mass measurement and does not discuss REMPI). https://arxiv.org/pdf/2412.06338

37. “Ultrasensitive Detection of Hydrogen Molecules by (2+1) REMPI,” eScholarship. https://escholarship.org/uc/item/2zk9g8mx

38. Search-result synthesis on VUV purge requirements (dry N2 to ~115 nm limited by O2, Ar/He shorter).

39. Springer, “Vacuum-ultraviolet photodetectors,” PhotoniX (2020) — general VUV instrumentation context. https://link.springer.com/article/10.1186/s43074-020-00022-w

40. “Hydrogen Balmer α line behavior in Laser-Induced Breakdown Spectroscopy depth scans of Au, Cu, Mn, Pb targets in air,” Spectrochim. Acta B (2010). https://www.sciencedirect.com/science/article/abs/pii/S0584854710001825

41. “Roughness effects on the hydrogen signal in laser-induced breakdown spectroscopy,” Spectrochim. Acta B (2017). https://www.sciencedirect.com/science/article/abs/pii/S0584854717301854

42. “Laser-induced breakdown spectroscopy for the measurement of air humidity using spectral line intensity ratio,” RSC Advances (2025). https://pubs.rsc.org/en/content/articlelanding/2025/ra/d5ra05015d; PMC full text: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12308682/

43. “Correlation between laser-induced breakdown spectroscopy signal and moisture content,” Spectrochim. Acta B (2012). https://www.sciencedirect.com/science/article/abs/pii/S0584854712001942

44. Search-result synthesis on Hα(656.3 nm)/N or O line-ratio humidity correlation (linearity >99.9%).

45. SciAps, “Z-Series LIBS Analyzers” product page (Z-903: 190–950 nm range, detects H–U; PULSAR laser 5–6 mJ/pulse, up to 50 Hz, 1 ns pulses; ~4 lb with battery). https://www.sciaps.com/products/libs/z-series; Malvern Panalytical, “SciAps Z-903 Handheld LIBS Analyzer.” https://www.malvernpanalytical.com/en/products/product-range/z-series/z-903

46. “Quantitative measurement of atomic hydrogen in low-pressure methane flames using two-photon LIF calibrated by krypton,” Combustion and Flame (2020). https://www.sciencedirect.com/science/article/abs/pii/S0010218020305575

47. General search-result context on VUV spectrometer/synchrotron facility cost scale (no single citable public price; qualitative characterization only).

48. “SciAps Z200c+ LIBS Analyzer,” Golden Analyzer Supply product listing (price $4,755 as listed on the cited page at time of verification; comparable Z-902C+/Z-200C+ configurations are listed elsewhere at $13,900–$15,090 — price unresolved). https://www.goldenanalyzersupply.com/product/sciaps-z200c-libs/

49. DirectIndustry eMagazine, “Affordable Handheld Device Allows Analysis of Lithium in Soils and Minerals” (context on handheld LIBS/portable analyzer pricing, €40,000–55,000 range for specialized units). https://emag.directindustry.com/2022/11/09/affordable-handheld-device-allows-analysis-of-lithium-in-soils-and-minerals/

50. SOT (Singapore Optics Technology), “Lithium Fluoride (LiF) Optical Crystal | UV to IR Windows” (transparency 105 nm–6 µm). https://sot.com.sg/optical-crystal-lithium-fluoride-lif/

51. Del Mar Photonics, “LiF windows VUV grade” (transmission at Lyman-alpha, solarization/color-center degradation). http://www.delmarphotonics.com/LiF-Lithium_Fluoride-VUV-grade/LiF%20windows%20VUV%20grade.htm

52. Alkor, “Lithium Fluoride LiF windows and lenses” (~40% transmission at 121 nm). https://alkor.net/LiF.html

53. Firebird Optics, “Magnesium Fluoride (MgF2) Viewports for Vacuum Systems.” https://www.firebirdoptics.com/vacuum-viewports/magnesium-fluoride-mgf2-viewports

54. Crystran, “Magnesium Fluoride MgF2 - Optical Materials.” https://www.crystran.com/optical-materials/magnesium-fluoride-mgf2/

55. ANCORP, “Magnesium Fluoride (MgF2) Viewports.” https://ancorp.com/product-category/viewports/magnesium-fluoride-mgf2-viewports/

56. Search-result synthesis on Al+MgF2 VUV mirror coatings (~250 Å MgF2 overcoat, peak reflectivity near 120 nm, usable reflectance 115–200 nm >80%).

57. ResearchGate figure, “Measured reflectance of an Al+MgF2 mirror from 310 Å to 1600 Å.” https://www.researchgate.net/figure/Measured-reflectance-of-an-Al-MgF2-mirror-from-310-A-to-1600-A-The-MgF2-thickness-is_fig4_23578223

58. “Vacuum ultraviolet high-reflectance aluminum mirrors on copper substrate for application in noble liquid time projection chambers,” (2021). https://www.sciencedirect.com/science/article/pii/S0042207X2100751X

59. “The road ahead for ultrawide bandgap solar-blind UV photodetectors,” J. Appl. Phys. 131, 150901 (2022). https://pubs.aip.org/aip/jap/article/131/15/150901/2836676/

60. “Progress on AlGaN-based solar-blind ultraviolet photodetectors and focal plane arrays,” Light: Sci. Appl. 10, 94 (2021). https://www.nature.com/articles/s41377-021-00527-4

61. “High-performance solar-blind AlGaN photodetectors,” Bilkent University repository PDF. https://repository.bilkent.edu.tr/bitstream/handle/11693/27381/High_performance_solar_blind_AlGaN_photodetectors.pdf

62. “Enhanced performance of AlGaN solar-blind ultraviolet avalanche photodiodes through electric field optimization,” (2024). https://www.sciencedirect.com/science/article/abs/pii/S2773012324002978

63. Search-result synthesis on AlGaN photodiode dark current (3 fA), cutoff (~266 nm), and responsivity (~0.051 A/W) figures from multiple AlGaN p-i-n photodiode papers.

64. Hamamatsu, “Photomultiplier Tube R821” datasheet TPMH1210E (Cs-Te photocathode, ~160–320 nm spectral response). https://www.hamamatsu.com/content/dam/hamamatsu-photonics/sites/documents/99_SALES_LIBRARY/etd/R821_TPMH1210E.pdf

65. Search-result synthesis on CsI/VUV photocathode microchannel-plate detector use in VUV spectrometers.

66. Hamamatsu, “Deuterium Lamps (D2 Lamps)” technical datasheet TLS1017E. https://www.hamamatsu.com/content/dam/hamamatsu-photonics/sites/documents/99_SALES_LIBRARY/etd/D2lamps_TLS1017E.pdf

67. McPherson, “MgF2 Window Deuterium” light source, Model 632/634 datasheet (115 nm–visible, 30 W, 2000 h lifetime, no flowing gas/coolant/pump required). https://mcphersoninc.com/lightsources/model632.html

68. Search-result synthesis on ArF excimer laser lithography specs (15 mJ pulse energy, 18 ns pulses, 1 kHz, natural bandwidth ~500 pm, line-narrowed to ~0.5 pm).

69. SPIE, “ArF excimer laser for 193-nm lithography.” https://www.spiedigitallibrary.org/conference-proceedings-of-spie/3051/0000/ArF-excimer-laser-for-193-nm-lithography/10.1117/12.276007.short

70. Search-result synthesis on F2 (157 nm) excimer laser doublet lines (157.63/157.52 nm) and line-narrowing to <0.2 pm.

71. USPTO patent document, “Molecular fluorine laser,” (F2 laser 157 nm technical background). https://image-ppubs.uspto.gov/dirsearch-public/print/downloadPdf/6577665

72. MDPI, “Development of a Table-Top High-Power, High-Stability, High-Harmonic-Generation Extreme-Ultraviolet Laser Source,” Photonics 12, 942 (2025). https://www.mdpi.com/2304-6732/12/9/942

73. “High photon flux table-top coherent extreme-ultraviolet source,” Nature Photonics (2014). https://www.nature.com/articles/nphoton.2014.214

74. KMLabs, “High Harmonic Generation | HHG | XUUS” product/technology page. https://www.kmlabs.com/high-harmonic-generation

75. Fujimoto et al., “Development of Laser-Produced Tin Plasma-Based EUV Light Source Technology for HVM EUV Lithography,” Physics Research International 2012, 249495. https://onlinelibrary.wiley.com/doi/10.1155/2012/249495

76. “Evolution analysis of EUV radiation from laser-produced tin plasmas based on a radiation hydrodynamics model,” Sci. Rep. 7, 45212 (2017). https://www.nature.com/articles/srep45212

77. “Production of 13.5 nm light with 5% conversion efficiency from 2 µm laser-driven tin microdroplet plasma,” Appl. Phys. Lett. 123, 234101 (2023). https://pubs.aip.org/aip/apl/article/123/23/234101/2925750/

78. “VUV-absorption cross section of carbon dioxide from 150 to 800 K and applications to warm exoplanetary atmospheres,” A&A 609, A34 (2018); arXiv:1709.08415 (corrected — arXiv:1502.07620 is a different, earlier Venot et al. preprint, “VUV-absorption cross section of CO2 at high temperatures and impact on exoplanet atmospheres,” not the cited A&A 2018 paper). https://www.aanda.org/articles/aa/full_html/2018/01/aa31295-17/aa31295-17.html

79. “Vacuum-UV spectroscopy of interstellar ice analogs. II. Absorption cross-sections of nonpolar ice molecules,” A&A (2014); arXiv:1405.7802 (CH4 cross section 5.7×10⁻¹⁸ cm², 120–160 nm); search-result synthesis on H2S VUV absorption bands (40–160 nm major, 160–250 nm minor). https://arxiv.org/pdf/1405.7802

80. “Ionization Quantum Yields and Absorption Coefficients of Selected Compounds at 58.4 and 73.6–74.4 nm,” PMC. https://pmc.ncbi.nlm.nih.gov/articles/PMC6716037/ ; NIST, “Atomic Data for Helium (He)” (IP 24.587387 eV). https://physics.nist.gov/PhysRefData/Handbook/Tables/heliumtable1.htm

81. Hamamatsu, “Deuterium Lamps for Photoionization,” L7293/L13301 datasheet TLS1019E. https://www.hamamatsu.com/content/dam/hamamatsu-photonics/sites/documents/99_SALES_LIBRARY/etd/L7293_L13301_TLS1019E.pdf

82. Lab-Club, “L1835 high-intensity VUV deuterium lamp, water-cooled, 115-400 nm.” https://www.lab-club.com/L1835

83. McPherson, “VUV and Deep Ultraviolet Spectrometer,” Model 234/302. https://mcphersoninc.com/spectrometers/vuvuvvis/model234302.html

84. AZoOptics, “Vacuum Ultraviolet Monochromator - Model 234/302 from McPherson.” https://www.azooptics.com/optics-equipment-details.aspx?EquipID=137

85. BeamQ Laser, “Excimer lasers 193 nm, 248 nm, 308 nm, and 351 nm” product listing (~$195,000). https://beamq.com/excimer-lasers-193-nm248-nm308-nm-and-351-nm-p-1923.html

86. Gigaphoton, “193nm High Power Laser for Drilling with High Aspect Ratio into Wide Bandgap Materials” (GT40A, 45 W). https://www.gigaphoton.com/wp-content/uploads/2018/03/193nm-High-Power-Laser-for-Drilling-with-High-Aspect-Ratio-into-Wide-Bandgap-Materials.pdf

87. Malvern Panalytical, “Handheld LIBS analyzers” product category page. https://www.malvernpanalytical.com/en/products/category/handheld-libs-analyzers

88. Thermo Fisher Scientific, “Niton Apollo Handheld LIBS Analyzer” product page. https://www.thermofisher.com/us/en/home/industrial/spectroscopy-elemental-isotope-analysis/portable-analysis-material-id/industrial-elemental-radiation-solutions/niton-apollo-handheld-libs-analyzer.html