Unveiling ignis fatuus: Microlightning between microbubbles

Abstract

Will-o’-the-wisps, ghostly blue flames seen at night in marshlands and long attributed to methane cool flames, have remained scientifically unexplained, a mystery caused by the lack of a known ignition mechanism. Here, we demonstrate that spontaneous electrical discharges, termed “microlightning”, can occur between rising methane-containing microbubbles in water. High-speed optical imaging reveals brief flashes between charged bubbles, arising from strong electric fields at curved gas–liquid interfaces. These discharges initiate nonthermal oxidation of methane, producing luminescence and measurable heat under ambient conditions. Our findings offer a scientific basis for ignis fatuus and reveal a general mechanism by which electrified interfaces can drive redox reactions in natural environments without the need for external ignition sources.

For centuries, faint blue flames known as ignis fatuus or will-o’-the-wisps have danced above marshes, cemeteries, and wetlands (1). Despite cultural ubiquity, their physical origin remains unresolved (2). Contemporary studies suggest these lights may result from “cool flames” of methane, where low-temperature oxidation leads to blue–violet chemiluminescence (3, 4). It is known that swamp gas contains mostly methane from decaying organic matter (5). However, the question of spontaneous ignition at ambient conditions persists: The activation energy for methane oxidation is >100 kcal/mol, far too high for passive initiation in nature (2).

Past explanations, invoking phosphine or static electricity, remain speculative (6). Our recent studies show that micrometer-sized water droplets can accumulate interfacial charge and discharge spontaneously, which is a phenomenon we termed microlightning (7–9). These microdischarges can generate reactive species under ambient conditions, demonstrating that strong electric fields at curved gas–water interfaces can mediate redox chemistry without applied voltage (10, 11).

We hypothesized that similar discharges might occur between methane bubbles in water. Bubbles possess curved gas–liquid interfaces, and recent experiments have shown that the electric field at the interface increases with increasing microdroplet curvature (12, 13). Moreover, the movement or coalescence or splitting apart can induce charge separation. If oppositely charged bubbles are near each other, the local electric field in the narrow interbubble gap may reach gas breakdown levels, producing a discharge that could ignite cool flame chemistry. Here, we examine a physically testable ignition route at the gas–liquid interface, where surface-localized microdischarges associated with rising and bursting bubbles can oxidize methane under ambient conditions. We discuss this as a plausible contributing mechanism to historical above-water luminescence.

Results and Discussion

We designed a transparent microbubble generator to introduce methane–air bubbles into water through a submerged nozzle (Fig. 1A). The rising microbubbles exhibited strong interfacial curvature and a high surface area-to-volume ratio, both factors known to enhance charge separation.

Fig. 1.

Microlightning between microbubbles. (A) Experimental setup for methane–air bubbling with optical access. Imaging is performed through a lateral 20x microscope lens positioned to view the bubble zone in the tank. Arrows indicate flow directions. (B) Schematic (Top) and high-speed frames (Bottom) showing two neighboring bubbles approaching and a localized flash captured at 0.05 ms. (Scale bar, 150 µm.)

Under dense bubbling conditions, brief, localized flashes were observed between adjacent bubbles. High-speed imaging (Movies S1 and S2) captured these submillisecond emissions, consistent with electrical discharges. A representative event is shown in Fig. 1B. These microlightning events closely resemble those observed previously between charged droplets, reinforcing a shared mechanism.

Photodiode detection confirmed light-emitting events even with air only (Fig. 2A), indicating that microlightning arises from interfacial charge interactions rather than gas composition. When methane was introduced, both the intensity and frequency of emission increased, and calibrated photon-counting reported absolute count rates for these events (Fig. 2B), consistent with activation of chemiluminescence at the interface. Furthermore, a fuel-swap with H₂/air under identical conditions also produced interbubble flashes and measurable visible luminescence.

Fig. 2.

Optical, thermal, and product measurements during bubbling. (A) Real-time photodiode current recorded during air and methane–air bubbling. (B) Calibrated photon-counting trace (counts/s) acquired under identical conditions. (C) Emission spectra collected under background, air, and methane–air bubbling. (D) Simultaneous temperature monitoring of the bulk water and the pump during bubbling. (E) Headspace mass-spectrometric signals (relative intensity) recorded before and after bubbling. (F) Time evolution of normalized mass-spectrometric signals for selected channels.

Optical emission spectra (Fig. 2C) supported this interpretation. Methane–air bubbling produced a broad emission enhancement centered around 330 to 370 nm—characteristic of electronically excited intermediates like electronically excited formaldehyde (CH₂O*) and hydroxyl radicals (•OH*), common in cool flames (14, 15). These broad features, along with sporadic spikes, point to nonthermal oxidation initiated at gas–liquid interfaces.

Thermal measurements (Fig. 2D) revealed a temperature increase in the bulk water that exceeded pump-generated heating, but only under methane–air bubbling. This excess heating implies exothermic chemical reactions, such as methane oxidation, initiated by microlightning.

To complement these optical and thermal signatures, headspace mass spectrometry tracked selected channels before and after bubbling (Fig. 2E) and over extended operation (Fig. 2F). Relative signals indicate a decrease at 16 amu (CH₄) and 32 amu (O₂) with a concomitant rise at 44 amu (CO₂), providing product-side support for partial oxidation under ambient conditions.

Conclusion

Microlightning between methane microbubbles offers a natural ignition mechanism for methane oxidation under ambient conditions. This discovery supports a long-suspected link between electrified interfaces and spontaneous cool flames, and it provides a physically grounded explanation for the occurrence of ignis fatuus. More broadly, our findings demonstrate that transient electrical discharges at gas–water interfaces can mediate redox chemistry, expanding the role of interfacial phenomena in natural and engineered systems.

Materials and Methods

Methane–air bubbles (with air-only and H₂/air controls) were introduced into water via a submerged nozzle. Submillisecond optical emissions at bubble–bubble contacts were recorded by high-speed video, a fast photodiode, and a calibrated photon-counting module; emission spectra were collected concurrently. Continuous logging of water and pump temperatures distinguished reaction-induced warming from pump heat loads. Quadrupole mass spectrometry monitored gases in the headspace before, during, and after bubbling, tracking changes at 16, 32, and 44 amu over time. Full device schematics, gas compositions/flows, optical alignments, detector calibrations, and data-processing protocols are provided in the SI Appendix.

Supplementary Material

Appendix 01 (PDF)

Movie S1.

A movie showing multiple spontaneous discharge events (microlightning) occurring between adjacent methane–air microbubbles under high–density bubbling conditions. The video was recorded using a high–speed camera (Photron T2410) at 24,000 frames per second (fps). It captures 0.5 seconds of real time and is played back at 3 fps, 8000 times slower than actual speed to highlight sub-millisecond flash events distributed across the field of view.

Movie S2.

A movie showing a close-up view of two methane–air microbubbles approaching and undergoing a microlightning event immediately prior to coalescence. The video was recorded using the same high–speed camera at 24,000 fps, capturing 0.2 seconds of real time and played back at 3 fps (8000× slower). A distinct flash can be observed in the narrow gap between the bubbles, corresponding to a transient interfacial discharge.

Data, Materials, and Software Availability

All study data are included in the article, the supporting information, and the publicly available link: https://github.com/YuXia19/Unveiling-ignis-fatuus-Microlightning-between-microbubbles (16).