In-situ measurement of hydrogen on airless planetary bodies using laser-induced breakdown spectroscopy

Remote-sensing observations on the surface of airless bodies, such as the Moon and asteroids, have confirmed the presence of hydrogen-bearing materials. However, their spatial distributions at small scales (mm–m) and depth profiles have great uncertainties. In-situ analyses of hydrogen-bearing materials with laser-induced breakdown spectroscopy (LIBS) have been proposed to resolve these problems, as the footprint of LIBS ablation is small (≲1 mm) and can penetrate into the subsurface by excavating the surface layer. Nevertheless, the measurement accuracy of hydrogen with LIBS on airless and hydrous planetary bodies has not been evaluated because it requires extensive calibration using hydrogen-rich geologic materials under a high-vacuum condition. In addition, whether hydrogen occurs as hydroxyl or ice has been difficult to ascertain via LIBS analysis because molecular information is typically lost in the ablation plasma. To resolve these problems, we conducted two experiments. First, compressed powders of rocks were measured by LIBS under vacuum (<3 × 10−2 Pa) to evaluate the calibration accuracies and detection limits in rocks and compacted soils on airless bodies. Several geostandards including basalts and feldspars were doped with various concentrations of hydroxyls (Mg(OH)2 and Ca(OH)2) to prepare hydrogen-rich samples up to 15 wt% in H2O-equivalent concentration (wt%H2O). Our results show that the hydrogen concentration can be accurately calibrated from the LIBS spectra by using multivariate models or matrix-matched calibration curves (i.e., calibration using samples with comparable bulk elemental abundances), facilitating the correction of significant matrix effects observed in the intensities of the 656 nm Hα line. We obtained a measurement accuracy of ±0.9 wt%H2O in the 0–12 wt%H2O range via matrix-matched calibration. This level of accuracy is sufficient for many planetary and resource exploration applications, such as designing hardware and operation for mining water on the Moon. We estimate the 2σ limit of detection (LOD) to be 0.4 wt%H2O based on the average of all samples, although better LODs were obtained for some individual matrix (e.g., 0.2 wt%H2O for basalt/feldspar–Ca(OH)2 mixtures). Such LOD shows that exploitable ice on the Moon can be detected with 2σ confidence by LIBS. Second, we demonstrate that the molecular structure of hydrogen can be distinguished by operating LIBS in tandem with heating lasers. In this method, the samples are heated prior to LIBS analysis using a continuous-wave laser with adjusted fluence and duration. Our results indicate that ice and hydroxyl can be distinguished because the Hα lines of ice-bearing samples decrease after the laser heating due to sublimation, but those of hydroxyl-bearing samples are retained. In addition, we report an enhancement of hydrogen emission from loose powders, suggesting that hydrogen in lunar soils may be measured with higher sensitivity. The results of this study show that LIBS is a versatile and powerful tool for accurate stand-off measurement of hydrogen-bearing materials on airless planetary bodies.