A gateway for ion transport on gas bubbles pinned onto solids

The study addresses whether and how ions are exchanged between the electric double layers (EDLs) at gas bubble surfaces and the solid surfaces on which they are pinned. Gas bubbles in water bear a negatively charged EDL due to hydroxide adsorption and countercations, which can influence catalysis by shielding electrodes, blocking reaction sites, or causing overpotentials. The central hypothesis is that an ion transport gateway forms where the EDLs of bubbles and (semi)conducting solids overlap, enabling exchange of counterions. To probe this directly at single-bubble scale, the authors employ ultramicroelectrode-based electrochemical impedance spectroscopy (EIS) on oxygen bubbles pinned to hematite (α-Fe2O3) and gold, with implications for water-splitting catalysis and geochemical systems.

Prior work has characterized bubble populations and their electrochemical effects using AFM and scanning electrochemical microscopy, revealing spatial distributions and resistive effects of bubbles on catalytic surfaces. The charging of gas–water interfaces is well established, with microbubble zeta potentials, EDL models, and hydroxide adsorption on hydrophobic interfaces reported. Studies have shown bubbles can increase IR drop, induce overpotential and impact current distribution at gas-evolving electrodes. However, previous studies largely examined ensembles rather than the direct electrochemical response of single bubbles, leaving unresolved the mechanisms of ion transport within overlapping EDLs at the bubble–solid contact.

Oxygen bubbles were generated electrochemically by water oxidation at +1.0 V (2-electrode configuration) on hematite (α-Fe2O3) or gold working electrodes with a Pt counter electrode in air-saturated aqueous electrolytes: 1.0 mM LiCl, 1.0 mM NaCl, 1.0 mM KCl, and 0.1 mM HCl + 1.0 mM NaCl. Electrodes were cleaned by sequential sonication in propanol, ethanol, and ultrapure water. Bubbles (diameters ~144–460 µm) formed after ~40 s and remained pinned for hours. Single-bubble EIS was performed using AC-SECM with a 25 µm-diameter Pt ultramicroelectrode (UME) as working electrode, Pt counter, and Ag wire pseudo-reference. The UME was positioned over a single bubble and lowered to partially compress it, ensuring electrochemical contact among UME, bubble, and substrate (hematite, gold, or PTFE). Spectra were recorded at open circuit potential with an AC amplitude of ±50 mV over 51 frequencies from 1 to 10^5 Hz; low ionic strength minimized bulk solution contributions. In select experiments, the same bubble was laterally moved 3–5 mm from hematite or gold onto adjacent PTFE to compare responses; additional measurements were conducted while applying external potentials of −1.0 V or +1.0 V through the supporting hematite or gold electrode. Bubble diameters decreased on average by 3.2 ± 1.9% after each EIS acquisition, attributed to gas loss under AC. Impedance data were modeled in Matlab using an equivalent circuit consisting of the UME capacitance (active at low frequency) in series with a bubble element modeled as a parallel combination of polarization resistance (Rbubble) and a constant phase element representing double-layer capacitance (Cbubble), with non-ideality factor φ co-optimized (0.5–1.0). Solution impedance was neglected as sufficiently small. Model selection favored parsimony to avoid intercorrelated parameters observed in more complex circuits.

• EIS of single oxygen bubbles pinned on hematite exhibited a distinct semi-circle in the ~30–3000 Hz range, assigned to bubble surface processes, and no response above ~3000 Hz, indicating no space charge or bulk electron transport in hematite under these conditions. Low-frequency response (<~30 Hz) arose from UME capacitance. An equivalent circuit with parallel Rbubble and Cbubble captured the bubble response.
• Across 29 bubbles and repeated measurements up to ~80 min, Rbubble spanned ~0.1–1 MΩ and Cbubble was ~5–10 pF s−1 (CPE parameter), with values independent of bubble diameter and largely insensitive to electrolyte identity and conductivity (LiCl, NaCl, KCl, and 0.1 mM HCl + NaCl).
• The constant phase element non-ideality factor was near unity (φ ≈ 0.9–1.0) on both hematite and gold, indicating a predominantly homogeneous bubble surface response; variability likely reflects differences in contact area at the bubble–substrate interface.
• When the same bubble was moved from hematite (or gold) onto PTFE, the double-layer capacitance remained similar, but polarization resistance increased markedly, evidencing inhibited ion transport between compositionally similar PTFE and bubble interfaces compared to hematite/gold interfaces.
• Applying external potentials through the supporting solid modulated bubble impedance: on hematite, −1 V reduced Rbubble by about 20% relative to open circuit, whereas +1 V increased Rbubble beyond values on PTFE. Similar trends were observed on gold. Negative potentials attract cations from the bubble toward the solid, enhancing ion exchange; positive potentials inhibit cation transfer to the solid and cannot drive anions toward the negatively charged bubble, increasing resistance.

The data directly support the existence of a gateway for ion transport at the interface of pinned gas bubbles and (semi)conducting solids, arising from overlapping EDLs of differing (electro)chemical potentials. The semicircular impedance feature and its modeling as parallel Rbubble–Cbubble reflect ion mobility within the bubble EDL and interfacial exchange with the solid’s EDL. The insensitivity to electrolyte type and bubble size, the near-ideal CPE behavior, and the sensitivity of Rbubble to substrate identity (hematite/gold vs PTFE) and to applied potential collectively indicate that the EDL composition and potential of the supporting solid govern the ease of counterion exchange across the overlap region. Negative applied potentials facilitate cation transfer from bubble to solid, reducing resistivity, while positive potentials suppress this transfer, increasing resistivity. These findings elucidate mechanisms by which bubbles can electrically shield catalysts, alter local potential distributions, and contribute to reaction overpotentials, with implications for electrolysis, flotation, microfluidics, porous media geochemistry, and fuel cell operation.

The study reveals and characterizes a gateway for ion transport between oxygen bubble surfaces and supporting (semi)conducting solids due to overlapping EDLs. Single-bubble EIS with a UME demonstrates that this gateway manifests as a constant double-layer capacitance with a variable polarization resistance that depends on substrate identity and applied potential. The gateway is active at open circuit and can be tuned: negative potentials enhance, while positive potentials inhibit, cation exchange between bubble and solid. Variability in bubble impedance likely stems from differences in contact area at the bubble–solid interface. These insights are directly relevant to photoelectrochemical water splitting on hematite and extend to broader systems where bubbles interact with (semi)conducting surfaces and porous materials. Future work could map contact-area effects, expand to other gases and substrates, quantify ion-specific effects, and integrate operando measurements during gas-evolving reactions.

• The impedance analysis relies on a simplified equivalent circuit; while more complex models improved fits, they were rejected due to parameter intercorrelation, potentially limiting mechanistic resolution.
• Solution impedance was neglected as small under low ionic strength; behavior at higher conductivities was not examined in detail.
• The study focused on specific materials (hematite, gold, PTFE), low electrolyte concentrations, and oxygen bubbles of ~144–460 µm; generalization to other systems, sizes, and gas types remains to be validated.
• Bubble compression by the UME and slight bubble size reduction during repeated measurements may influence interfacial contact area and measured impedance.