Surface Fields Drive Trihydrogen Formation on Hydrated Silica

The trihydrogen cation, $\mathrm{H_3^+}$, starts ion–molecule reaction chains that build complexity in interstellar environments, yet its canonical route, $\mathrm{H_2^+ + H_2 \rightarrow H_3^+ + H}$, had not been established on an inorganic surface under radiation-driven ionization. Sahel-Schackis and colleagues test that chemistry on hydrated silica nanoparticles exposed to intense X-ray free-electron-laser pulses. Their central result is not simply an H3+ mass peak: the relative hydrogen-ion products track the electric field created as each particle charges, across particle types and aggregation states.

Core Contribution

The paper makes a case for self-induced surface electric field as the controlling variable for water-derived $\mathrm{H^+}$, $\mathrm{H_2^+}$, and $\mathrm{H_3^+}$ yields. X-ray ionization ejects electrons from silica, leaving a positively charged particle whose surface field reaches the V/nm range. At that interface, the authors argue, the field promotes hole transfer from silica into adsorbed water, creating the charged and neutral hydrogen-bearing fragments needed for trihydrogen formation.

That attribution is stronger than a simple intensity dependence because the experiment separates particle size, composition, and monomer/dimer aggregation. The reported overlap of monomer and dimer time-of-flight spectra at matched effective fields supports the narrower interpretation: neighboring particles matter through their contribution to the local field rather than through a distinct aggregation-specific reaction pathway. Figure 1 lays out this chain from X-ray charging through surface water fragmentation and ion detection.

Technical Approach

The experiment combines three single-particle measurements at the European XFEL: ion velocity-map imaging for transverse momentum and time-of-flight mass spectra, electron time-of-flight spectroscopy for emitted electrons, and coherent diffractive imaging to characterize individual particles and identify monomers or dimers. The ion VMI uses a Timepix3 timestamping camera and records ion flight times up to 3.54462 $\mu\mathrm{s}$. Coherent diffraction patterns are classified with a convolutional neural network trained using simulated dimer augmentation.

The authors use silica particles spanning several sizes and compositions, including 100 nm, 300 nm, and 500 nm silica, plus 150 nm Au@SiO2 core-shell particles. Their supplementary characterization gives measured diameters of $126 \pm 7$ nm, $312 \pm 14$ nm, and $504 \pm 32$ nm for the silica populations. Particle position in the Gaussian X-ray focus changes the deposited energy, while diffraction-derived sizing and electron counts are used to estimate per-event fields.

For the dimer test, a 300 nm SiO2 monomer producing about 4,000 detected ions is estimated to emit about 54,000 electrons and reach 3.2 V/nm. A dimer with about 8,000 detected ions is estimated at about 97,000 emitted electrons, or about 48,500 per particle, corresponding to 2.9 V/nm per particle; the neighbor’s field is then added by linear superposition. This treatment resolves contact pairs but assumes independent point-charge sources and neglects mutual shadowing, dielectric screening, and altered geometry in fused or necked aggregates.

The mechanistic account is backed by density-functional partial-density-of-states calculations and nonadiabatic quantum molecular dynamics. The calculations show water-derived O 2p states moving into the intrinsic surface-gap region as applied field increases, increasing the opportunity for hole localization in the water layer. Figure 4 presents that field-dependent band tilting.

Results and Analysis

The experimental intensity scans place H3+ formation below the reported warm-dense-matter transition at a deposited energy density of 1 MJ/kg. In contrast, the Si+ feature changes slope at that transition, indicating a different charging and expansion regime. This separation matters: the paper does not need warm dense matter to explain the trihydrogen signal, and it narrows the claimed chemistry to the earlier surface-chemistry regime.

Binning spectra by inferred field shows $\mathrm{H^+}$ and $\mathrm{H_3^+}$ yields rising with field while $\mathrm{H_2^+}$ declines. At matched 2–4 V/nm effective field, monomer and dimer spectra overlap, including at particle contact; supplementary spectra show the same correspondence over 1–3 and 3–5 V/nm selections. The evidence therefore supports field as a useful organizing parameter, though it does not directly measure every elementary step of $\mathrm{H_2^+ + H_2 \rightarrow H_3^+ + H}$.

The simulations provide a timescale consistent with that interpretation: after hole creation in silica, average hole population transfers to the adsorbed water layer within about 60 fs at 5 V/nm. One trajectory shows a field-driven $\mathrm{H_2O^+}$ fragment leaving the surface between 90 and 110 fs, while $\mathrm{H_3O^+}$ remains in the hydrogen-bond network and desorbs more slowly. Together, the measurements and simulations make a credible interface-charge-transfer mechanism, but the astrophysical implication remains an analogy: the experiment uses extreme 1.88 keV pulsed irradiation rather than a direct reproduction of an interstellar grain environment.

Caveats in Practice

The study measures product-ion yields under a specific XFEL excitation geometry, rather than an absolute H3+ formation rate on realistic cosmic dust. Field estimates also depend on electron-count inference and the point-charge superposition approximation for dimers. The simulations model a finite hydrated silica interface under applied fields, so they illuminate a plausible microscopic path without sampling the full diversity of surface defects, water coverage, and radiation histories expected outside the laboratory.