Three-step growth of vapor-deposited ice under mesospheric temperature and water vapor conditions (2507.18477v1)

Abstract: Polar mesospheric clouds provide clues to physicochemical processes in the mesosphere and lower thermosphere. However, the heterogeneous nucleation and growth processes of water ice under polar mesospheric conditions are poorly understood, especially at the nanoscale. This study used reflection high energy electron diffraction and infrared reflection absorption spectroscopy to analyze the structure of vapor-deposited ice at polar mesospheric temperature (120 K) under vapor pressure conditions. The ice appeared to grow in three steps during vapor deposition, being amorphous water for the first 15 nm, then cubic ice up to 50 nm, and finally hexagonal ice subsequently. This three step growth suggests that the three observed phases can coexist in polar mesospheric clouds, depending on the thickness of water ice. The finding of the three-step growth also shows that the Ostwald rule of stages can hold for vapor deposited ice at low temperature.

• The paper demonstrates that vapor-deposited water ice grows in three sequential phases—amorphous, cubic, and hexagonal—in line with the Ostwald rule of stages.
• The study employs in situ RHEED and IRRAS to precisely track thickness-dependent phase transitions at 120 K under controlled water vapor conditions.
• The findings have significant implications for understanding mesospheric cloud microphysics and improving atmospheric remote sensing models.

Three-Step Growth of Vapor-Deposited Ice under Mesospheric Conditions: Structural Evolution and Implications

Introduction

This paper investigates the nanoscale structural evolution of vapor-deposited water ice under conditions analogous to the Earth's polar mesosphere, specifically at 120 K and water vapor pressures of $10^{-6}$ Pa. The work addresses a critical gap in understanding the phase behavior and growth mechanisms of ice in polar mesospheric clouds (PMCs), which are composed of nanometer-scale ice particles and play a significant role in mesospheric chemistry and climate diagnostics. The research employs in situ reflection high-energy electron diffraction (RHEED) and infrared reflection-absorption spectroscopy (IRRAS) to resolve the phase transitions and coexistence of amorphous and crystalline ice forms during deposition, with a focus on the thickness dependence of these transitions.

Experimental Approach

A custom ultrahigh-vacuum system was developed, integrating a closed-cycle helium cryostat for substrate cooling, a controlled water vapor doser, and simultaneous RHEED and IRRAS capabilities. The substrate was a mirror-polished Al disk, and deposition was performed with H$_2$O containing 3.5 mol% HDO to facilitate OD-stretch IR measurements. The system allowed precise control and measurement of ice thickness, deposition rate, and phase identification at the nanoscale.

Key experimental parameters:

• Substrate temperature: 120 K
• Water vapor pressure: $2.2 \times 10^{-6}$ Pa (local, near substrate)
• Growth rate: $0.15 \pm 0.01$ nm/min
• Ice thickness range: 5.6–63 nm (30–420 min deposition)

Results: Three-Step Growth Sequence

Amorphous Water Formation (≤15 nm)

For initial depositions up to $\sim$15 nm, IRRAS spectra were well reproduced by models for amorphous solid water (ASW), and RHEED patterns showed only diffuse halos, confirming the absence of long-range order. The sticking probability of water molecules at 120 K is near unity, and the low surface diffusion coefficient ($D_s$) at this temperature and flux regime prevents rearrangement into crystalline sites, favoring rapid hydrogen bond formation and amorphous growth.

Cubic Ice Nucleation and Growth (15–50 nm)

Between 13–17 nm, RHEED revealed the emergence of Debye-Scherrer rings corresponding to cubic ice (Ic) reflections, while IRRAS still indicated a predominantly amorphous bulk. This suggests surface-limited nucleation of nanocrystalline Ic domains atop the ASW layer. The transition is attributed to increased $D_s$ on laterally smoother ASW surfaces at higher thickness, as supported by STM/AFM studies showing reduced surface roughness with increasing ASW thickness. The critical thickness for Ic nucleation aligns with theoretical predictions for size-dependent phase stability in water clusters and films.

Hexagonal Ice Emergence (≥45 nm)

For ice thicknesses exceeding 45 nm, RHEED patterns showed the appearance of hexagonal (Ih) reflections, indicating the onset of hexagonal stacking sequences. IRRAS spectra diverged from ASW models and developed features characteristic of polycrystalline ice I. The transition from Ic to Ih is consistent with mechanisms such as epitaxial 2D nucleation on Ic(111) planes (identical to Ih(0001)) and double spiral growth mediated by screw dislocations, as observed in prior STM/AFM studies. The transformation is kinetically hindered in the bulk at 120 K, so the observed Ih is formed at the surface during ongoing deposition.

Phase Coexistence and Kinetics

Analysis of the decoupled OD-stretch band in IRRAS for thickest films (63 nm) revealed a persistent $\sim$30% amorphous fraction, indicating that once ASW forms, it does not readily crystallize at 120 K on experimental timescales ($\sim$420 min). Similarly, Ic-to-Ih transformation in the bulk is negligible at this temperature. Thus, the final structure is a mixture of ASW, Ic, and Ih, with phase fractions determined by deposition history and thickness.

Theoretical and Practical Implications

Ostwald Rule of Stages

The observed sequential formation—ASW $\rightarrow$ Ic $\rightarrow$ Ih—demonstrates the applicability of the Ostwald rule of stages under far-from-equilibrium vapor deposition at low temperature. The system does not directly form the thermodynamically stable Ih, but instead passes through metastable intermediates (ASW, Ic) as dictated by kinetic accessibility and relative free energies. The high supersaturation ratio ($\sim 8.8 \times 10^3$) further supports the non-equilibrium nature of the process.

Mesospheric Cloud Microphysics

The results imply that PMCs in the Earth's mesosphere can contain a heterogeneous mixture of amorphous, cubic, and hexagonal ice, with phase composition dependent on particle size (thickness). This has direct consequences for the interpretation of remote sensing data, as the physical properties (density, thermal conductivity, vapor pressure, optical response) of these phases differ significantly. The findings also constrain models of ice nucleation and growth in the upper atmosphere, supporting the dominance of heterogeneous nucleation on meteoric smoke particles and the importance of nanoscale phase behavior.

Broader Relevance to Nucleation Theory

The paper provides a rare experimental validation of the Ostwald rule in a vapor-solid system at cryogenic temperature and high supersaturation, challenging the classical single-step nucleation paradigm. The thickness-dependent phase evolution highlights the need to consider finite-size and surface effects in models of crystal growth, especially for nanomaterials and atmospheric aerosols.

Future Directions

Further systematic studies across a broader range of temperatures (100–145 K) and vapor pressures ($10^{-7}$–$10^{-5}$ Pa) are needed to map the full phase diagram of vapor-deposited ice under mesospheric conditions. High-resolution surface microscopy (STM/AFM) during deposition could elucidate the microscopic mechanisms of Ic-to-Ih transformation. The impact of substrate material, impurity content (e.g., meteoric smoke analogs), and deposition flux on phase selection and kinetics warrants investigation. These insights are also relevant for astrophysical ices and cryogenic materials science.

Conclusion

This work establishes that vapor-deposited water ice under mesospheric conditions exhibits a three-step growth sequence: initial formation of amorphous water, followed by surface nucleation and growth of cubic ice, and eventual emergence of hexagonal ice at greater thickness. The coexistence of these phases at the nanoscale is governed by kinetic constraints and surface diffusion, not solely by temperature or vapor pressure. The findings have significant implications for mesospheric cloud microphysics, atmospheric modeling, and the general understanding of nonclassical nucleation pathways in solid-state systems.