No abstract

Experimental measurements of bulk diffusion in ice and surface diffusion on ice were performed using laser resonant desorption (LRD) techniques. Bulk diffusion in ice was examined using ice sandwich structures and continuous source experiments together with LRD depth-profiling analysis. Surface diffusion was monitored using prepare-refill-probe LRD experiments. New experimental results were obtained for the bulk diffusion of NH 3 and CH 3 OH. These species probably exist as hydrates in the ice. The LRD measurements for CH 3 OH hydrate diffusion, combined with previous results, provide evidence for a vacancy-mediated diffusion mechanism. The diffusion rates for NH 3 hydrates are much larger than diffusion rates for H 2 O self-diffusion in ice and are attributed to the disruption of the ice lattice. LRD prepare-refill-probe experiments revealed that surface diffusion was not measurable for almost all of the species examined on ice. Only butane displayed a measurable surface mobility that was attributed to its unique size and chemical nature. These new measurements of bulk diffusion in ice and surface diffusion on ice should be useful in developing our understanding of kinetic processes in and on ice that are relevant to heterogeneous atmospheric chemistry and ice core analysis.

A new infrared laser resonant desorption (LRD) technique has been developed that permits depth-profiling and diffusion measurements in ice. This LRD technique utilizes an Er:YAG rotary Q-switched laser with an output wavelength of lambda = 2.94 microm and a pulse duration of approximately 100 ns. The Er:YAG laser light resonantly excites O-H stretching vibrations in the H2O molecules that form the ice. This laser resonant heating induces H2O desorption at the ice surface. Control experiments were conducted on pure and isotopically mixed laminated ice films to determine the optimum experimental parameters for the LRD depth-profiling and diffusion measurements. Depending on laser energy, the measured desorption depth was either less than, comparable to, or larger than the optical penetration depth of approximately 0.8 microm at lambda = 2.94 microm. LRD studies were used to analyze H2 18O/H2 16O stacked multilayers and laminate sandwich structures. These measurements revealed that the LRD technique can depth-profile into ice films with submicrometer spatial resolution and high sensitivity. Two types of experiments employing LRD depth-profiling were demonstrated to monitor diffusion in ice. HCl hydrate diffusion in ice was measured versus time after depositing ice/HCl/ice sandwich structures. Na diffusion into ice was studied after adsorbing Na using a continuous Na source for a given exposure time at the diffusion temperature.

The kinetics of HDO surface and bulk diffusion on ultrathin (25–192 BL; 90–700 Å) single-crystal H216O ice multilayers were studied using a combination of laser-induced thermal desorption (LITD) probing and isothermal desorption depth-profiling. The single-crystal hexagonal ice multilayers were grown epitaxially on a single-crystal Ru(001) metal substrate with the basal (001) facet of ice parallel to the Ru(001) surface. HDO surface diffusion on the single-crystal ice multilayer was not observed within the resolution of the LITD experiment at T=140 K. These LITD surface diffusion experiments yielded an upper limit to the HDO surface diffusion coefficient of Ds⩽1×10−9 cm2/s at T=140 K. The bulk diffusion coefficients were measured along the c axis of the hexagonal ice crystal which is perpendicular to the (001) plane. HDO was observed to diffuse readily into the underlying H216O ice multilayer. The measured HDO bulk diffusion coefficients ranged from D=2.2(±0.3)×10−16 cm2/s to D=3.9(±0.4)×10−14 cm2/s over the temperature range from 153 to 170 K. The HDO bulk diffusion coefficients were measured for H216O thicknesses of 25–192 BL (1 BL=1.06×1015 molecules/cm2) and initial HDO adlayer thicknesses of 2–9 BL. The HDO bulk diffusion was independent of H216O film thickness and initial HDO coverage. Arrhenius analysis of the temperature-dependent bulk diffusion coefficients yielded a diffusion activation energy of EA=17.0±1.0 kcal/mol and a diffusion preexponential of Do=4.2(±0.8)×108 cm2/s. Compared with extrapolations from macroscopic diffusion kinetics obtained earlier at temperatures close to the melting point, these bulk diffusion coefficients are larger and may reflect the perturbation of the ultrathin ice films induced by the nearby interfaces. The differences between these HDO diffusion kinetics and recently measured kinetics for H218O indicate that H/D exchange and molecular transport make comparable contributions to the HDO diffusion coefficient.

The diffusion of HDO into ultrathin single-crystal H2 16O ice multilayers was investigated using a novel combination of laser-induced thermal desorption (LITD) probing and isothermal desorption depth-profiling. The single-crystal hexagonal ice multilayers were grown epitaxially on a Ru(001) metal substrate, and the diffusion coefficients were measured perpendicular to the basal (0001) facet. The measured HDO diffusion coefficients ranged from D = (2.2 ± 0.3) × 10-16 to D = (3.9 ± 0.4) × 10-14 cm2/s over the temperature range 153−170 K. Arrhenius analysis of the temperature-dependent diffusion coefficients yielded a diffusion activation energy of E A = 17.0 ± 1.0 kcal/mol and a preexponential factor of D 0 = (4.2 ± 0.8) × 108 cm2/s. The similarity of the diffusion coefficients for HDO and H2 18O indicates that H/D exchange does not contribute significantly to HDO diffusion in ice. The agreement between the diffusion kinetics for HDO and H2 18O argues that the HDO diffusion occurs via a molecular transport mechanism. The large diffusion preexponentials for both HDO and H2 18O diffusion into the ultrathin ice multilayers also suggest that bulk transport properties in ice may be perturbed by close proximity to the ice surface.