Analysis of isothermal and coolingratedependent immersion freezing by a unifying stochastic ice nucleation model
Immersion freezing is an important ice nucleation pathway involved in the formation of cirrus and mixedphase clouds. Laboratory immersion freezing experiments are necessary to determine the range in temperature, T, and relative humidity, RH, at which ice nucleation occurs and to quantify the associated nucleation kinetics. Typically, isothermal (applying a constant temperature) and coolingratedependent immersion freezing experiments are conducted. In these experiments it is usually assumed that the droplets containing ice nucleating particles (INPs) all have the same INP surface area (ISA); however, the validity of this assumption or the impact it may have on analysis and interpretation of the experimental data is rarely questioned. Descriptions of ice active sites and variability of contact angles have been successfully formulated to describe ice nucleation experimental data in previous research; however, we consider the ability of a stochastic freezing model founded on classical nucleation theory to reproduce previous results and to explain experimental uncertainties and data scatter. A stochastic immersion freezing model based on first principles of statistics is presented, which accounts for variable ISA per droplet and uses parameters including the total number of droplets, N _{tot}, and the heterogeneous ice nucleation rate coefficient, J _{het}( T). This model is applied to address ifmore »
 Authors:

^{[1]};
^{[1]}
 Stony Brook Univ., NY (United States). School of Marine and Atmospheric Sciences
 Publication Date:
 Grant/Contract Number:
 SC0008613
 Type:
 Accepted Manuscript
 Journal Name:
 Atmospheric Chemistry and Physics (Online)
 Additional Journal Information:
 Journal Name: Atmospheric Chemistry and Physics (Online); Journal Volume: 16; Journal Issue: 4; Journal ID: ISSN 16807324
 Publisher:
 European Geosciences Union
 Research Org:
 Stony Brook Univ., NY (United States). School of Marine and Atmospheric Sciences
 Sponsoring Org:
 USDOE Office of Science (SC), Biological and Environmental Research (BER) (SC23)
 Country of Publication:
 United States
 Language:
 English
 Subject:
 54 ENVIRONMENTAL SCIENCES
 OSTI Identifier:
 1254489
Alpert, Peter A., and Knopf, Daniel A.. Analysis of isothermal and coolingratedependent immersion freezing by a unifying stochastic ice nucleation model. United States: N. p.,
Web. doi:10.5194/acp1620832016.
Alpert, Peter A., & Knopf, Daniel A.. Analysis of isothermal and coolingratedependent immersion freezing by a unifying stochastic ice nucleation model. United States. doi:10.5194/acp1620832016.
Alpert, Peter A., and Knopf, Daniel A.. 2016.
"Analysis of isothermal and coolingratedependent immersion freezing by a unifying stochastic ice nucleation model". United States.
doi:10.5194/acp1620832016. https://www.osti.gov/servlets/purl/1254489.
@article{osti_1254489,
title = {Analysis of isothermal and coolingratedependent immersion freezing by a unifying stochastic ice nucleation model},
author = {Alpert, Peter A. and Knopf, Daniel A.},
abstractNote = {Immersion freezing is an important ice nucleation pathway involved in the formation of cirrus and mixedphase clouds. Laboratory immersion freezing experiments are necessary to determine the range in temperature, T, and relative humidity, RH, at which ice nucleation occurs and to quantify the associated nucleation kinetics. Typically, isothermal (applying a constant temperature) and coolingratedependent immersion freezing experiments are conducted. In these experiments it is usually assumed that the droplets containing ice nucleating particles (INPs) all have the same INP surface area (ISA); however, the validity of this assumption or the impact it may have on analysis and interpretation of the experimental data is rarely questioned. Descriptions of ice active sites and variability of contact angles have been successfully formulated to describe ice nucleation experimental data in previous research; however, we consider the ability of a stochastic freezing model founded on classical nucleation theory to reproduce previous results and to explain experimental uncertainties and data scatter. A stochastic immersion freezing model based on first principles of statistics is presented, which accounts for variable ISA per droplet and uses parameters including the total number of droplets, Ntot, and the heterogeneous ice nucleation rate coefficient, Jhet(T). This model is applied to address if (i) a time and ISAdependent stochastic immersion freezing process can explain laboratory immersion freezing data for different experimental methods and (ii) the assumption that all droplets contain identical ISA is a valid conjecture with subsequent consequences for analysis and interpretation of immersion freezing. The simple stochastic model can reproduce the observed time and surface area dependence in immersion freezing experiments for a variety of methods such as: droplets on a coldstage exposed to air or surrounded by an oil matrix, wind and acoustically levitated droplets, droplets in a continuousflow diffusion chamber (CFDC), the Leipzig aerosol cloud interaction simulator (LACIS), and the aerosol interaction and dynamics in the atmosphere (AIDA) cloud chamber. Observed timedependent isothermal frozen fractions exhibiting nonexponential behavior can be readily explained by this model considering varying ISA. An apparent coolingrate dependence of Jhet is explained by assuming identical ISA in each droplet. When accounting for ISA variability, the coolingrate dependence of ice nucleation kinetics vanishes as expected from classical nucleation theory. Finally, the model simulations allow for a quantitative experimental uncertainty analysis for parameters Ntot, T, RH, and the ISA variability. We discuss the implications of our results for experimental analysis and interpretation of the immersion freezing process.},
doi = {10.5194/acp1620832016},
journal = {Atmospheric Chemistry and Physics (Online)},
number = 4,
volume = 16,
place = {United States},
year = {2016},
month = {2}
}