Mass, momentum, and energy transfer in supersonic aerosol deposition processes

Chenxi Li, Narendra Singh, Austin Andrews, Bernard A. Olson, Thomas E Schwartzentruber, Chris Hogan

Research output: Contribution to journalArticle

Abstract

Aerosol deposition (AD) can lead to the formation of dense coatings via deposition of particles from a gas flow; in AD the aerosol is passed through a converging-diverging nozzle, facilitating inertial particle impaction on a desired substrate at supersonic particle velocities. Unique from thermal spray methods, AD can be applied near room temperature and unique from cold spray, in AD the aerosol is typically at atmospheric pressure upstream of the nozzle. Though AD has been successfully demonstrated previously, a number of aspects related to particle motion in AD systems remain poorly understood. In this work, we simulated compressible flow field profiles and particle trajectories for typical AD working conditions for a slit type converging-diverging nozzle with a planar substrate. In examining the fluid flow profile, we show that the velocity and pressure profiles, as well as the shock structure are sensitive to the upstream and downstream operating pressures of the nozzle. These ultimately affect particle impaction speed. Importantly, in AD, the particle drag regime is dynamic; both particle Knudsen numbers and Mach numbers can vary by orders of magnitude. To aid particle trajectory simulations, we trained a neural network to predict the drag force on the particles based on existing experimental data, theoretical limits, and new direct simulation Monte Carlo (DMSC) results. The neural network based drag law, which depends upon both Mach and Knudsen numbers, shows better agreement with the DSMC simulation data than pre-existing correlations. With it, particle trajectory simulation results reveal that for a given particle density, there exists an optimal particle diameter to maximize particle impaction speed. We also find that in AD particles undergo size dependent inertial focusing, i.e. there is a particular particle diameter where the particle deposition linewidth is minimized. Particles smaller than this diameter are underfocused, and particles larger than this are overfocused, and hence have larger deposition linewidths in both cases. Using trajectory simulations, we additionally developed a framework that can be used to evaluate the position-dependent mass, momentum and kinetic energy fluxes to the deposition substrate for any aerosol size distribution function upstream of the nozzle. It is shown for typical aerosol concentrations achievable in the laboratory, the kinetic energy flux can approach a magnitude normally observed in convective heat transfer with phase change, hence translational kinetic energy to thermal energy transfer in AD is likely a key contributor to the formation of dense coatings.

LanguageEnglish (US)
Pages1161-1171
Number of pages11
JournalInternational Journal of Heat and Mass Transfer
Volume129
DOIs
StatePublished - Feb 1 2019

Fingerprint

Momentum transfer
Aerosols
Energy transfer
mass transfer
momentum transfer
aerosols
Mass transfer
energy transfer
Particles (particulate matter)
nozzles
Nozzles
particle trajectories
kinetic energy
Trajectories
Kinetic energy
upstream
drag
Drag
Knudsen flow
Mach number

Keywords

  • Aerosol deposition
  • Converging-diverging nozzle
  • Direct simulation Monte Carlo
  • Inertial focusing
  • Inertial impaction

Cite this

Mass, momentum, and energy transfer in supersonic aerosol deposition processes. / Li, Chenxi; Singh, Narendra; Andrews, Austin; Olson, Bernard A.; Schwartzentruber, Thomas E; Hogan, Chris.

In: International Journal of Heat and Mass Transfer, Vol. 129, 01.02.2019, p. 1161-1171.

Research output: Contribution to journalArticle

@article{eca6e6a3d7d64f5ea2e59790bbf51af4,
title = "Mass, momentum, and energy transfer in supersonic aerosol deposition processes",
abstract = "Aerosol deposition (AD) can lead to the formation of dense coatings via deposition of particles from a gas flow; in AD the aerosol is passed through a converging-diverging nozzle, facilitating inertial particle impaction on a desired substrate at supersonic particle velocities. Unique from thermal spray methods, AD can be applied near room temperature and unique from cold spray, in AD the aerosol is typically at atmospheric pressure upstream of the nozzle. Though AD has been successfully demonstrated previously, a number of aspects related to particle motion in AD systems remain poorly understood. In this work, we simulated compressible flow field profiles and particle trajectories for typical AD working conditions for a slit type converging-diverging nozzle with a planar substrate. In examining the fluid flow profile, we show that the velocity and pressure profiles, as well as the shock structure are sensitive to the upstream and downstream operating pressures of the nozzle. These ultimately affect particle impaction speed. Importantly, in AD, the particle drag regime is dynamic; both particle Knudsen numbers and Mach numbers can vary by orders of magnitude. To aid particle trajectory simulations, we trained a neural network to predict the drag force on the particles based on existing experimental data, theoretical limits, and new direct simulation Monte Carlo (DMSC) results. The neural network based drag law, which depends upon both Mach and Knudsen numbers, shows better agreement with the DSMC simulation data than pre-existing correlations. With it, particle trajectory simulation results reveal that for a given particle density, there exists an optimal particle diameter to maximize particle impaction speed. We also find that in AD particles undergo size dependent inertial focusing, i.e. there is a particular particle diameter where the particle deposition linewidth is minimized. Particles smaller than this diameter are underfocused, and particles larger than this are overfocused, and hence have larger deposition linewidths in both cases. Using trajectory simulations, we additionally developed a framework that can be used to evaluate the position-dependent mass, momentum and kinetic energy fluxes to the deposition substrate for any aerosol size distribution function upstream of the nozzle. It is shown for typical aerosol concentrations achievable in the laboratory, the kinetic energy flux can approach a magnitude normally observed in convective heat transfer with phase change, hence translational kinetic energy to thermal energy transfer in AD is likely a key contributor to the formation of dense coatings.",
keywords = "Aerosol deposition, Converging-diverging nozzle, Direct simulation Monte Carlo, Inertial focusing, Inertial impaction",
author = "Chenxi Li and Narendra Singh and Austin Andrews and Olson, {Bernard A.} and Schwartzentruber, {Thomas E} and Chris Hogan",
year = "2019",
month = "2",
day = "1",
doi = "10.1016/j.ijheatmasstransfer.2018.10.028",
language = "English (US)",
volume = "129",
pages = "1161--1171",
journal = "International Journal of Heat and Mass Transfer",
issn = "0017-9310",
publisher = "Elsevier Limited",

}

TY - JOUR

T1 - Mass, momentum, and energy transfer in supersonic aerosol deposition processes

AU - Li, Chenxi

AU - Singh, Narendra

AU - Andrews, Austin

AU - Olson, Bernard A.

AU - Schwartzentruber, Thomas E

AU - Hogan, Chris

PY - 2019/2/1

Y1 - 2019/2/1

N2 - Aerosol deposition (AD) can lead to the formation of dense coatings via deposition of particles from a gas flow; in AD the aerosol is passed through a converging-diverging nozzle, facilitating inertial particle impaction on a desired substrate at supersonic particle velocities. Unique from thermal spray methods, AD can be applied near room temperature and unique from cold spray, in AD the aerosol is typically at atmospheric pressure upstream of the nozzle. Though AD has been successfully demonstrated previously, a number of aspects related to particle motion in AD systems remain poorly understood. In this work, we simulated compressible flow field profiles and particle trajectories for typical AD working conditions for a slit type converging-diverging nozzle with a planar substrate. In examining the fluid flow profile, we show that the velocity and pressure profiles, as well as the shock structure are sensitive to the upstream and downstream operating pressures of the nozzle. These ultimately affect particle impaction speed. Importantly, in AD, the particle drag regime is dynamic; both particle Knudsen numbers and Mach numbers can vary by orders of magnitude. To aid particle trajectory simulations, we trained a neural network to predict the drag force on the particles based on existing experimental data, theoretical limits, and new direct simulation Monte Carlo (DMSC) results. The neural network based drag law, which depends upon both Mach and Knudsen numbers, shows better agreement with the DSMC simulation data than pre-existing correlations. With it, particle trajectory simulation results reveal that for a given particle density, there exists an optimal particle diameter to maximize particle impaction speed. We also find that in AD particles undergo size dependent inertial focusing, i.e. there is a particular particle diameter where the particle deposition linewidth is minimized. Particles smaller than this diameter are underfocused, and particles larger than this are overfocused, and hence have larger deposition linewidths in both cases. Using trajectory simulations, we additionally developed a framework that can be used to evaluate the position-dependent mass, momentum and kinetic energy fluxes to the deposition substrate for any aerosol size distribution function upstream of the nozzle. It is shown for typical aerosol concentrations achievable in the laboratory, the kinetic energy flux can approach a magnitude normally observed in convective heat transfer with phase change, hence translational kinetic energy to thermal energy transfer in AD is likely a key contributor to the formation of dense coatings.

AB - Aerosol deposition (AD) can lead to the formation of dense coatings via deposition of particles from a gas flow; in AD the aerosol is passed through a converging-diverging nozzle, facilitating inertial particle impaction on a desired substrate at supersonic particle velocities. Unique from thermal spray methods, AD can be applied near room temperature and unique from cold spray, in AD the aerosol is typically at atmospheric pressure upstream of the nozzle. Though AD has been successfully demonstrated previously, a number of aspects related to particle motion in AD systems remain poorly understood. In this work, we simulated compressible flow field profiles and particle trajectories for typical AD working conditions for a slit type converging-diverging nozzle with a planar substrate. In examining the fluid flow profile, we show that the velocity and pressure profiles, as well as the shock structure are sensitive to the upstream and downstream operating pressures of the nozzle. These ultimately affect particle impaction speed. Importantly, in AD, the particle drag regime is dynamic; both particle Knudsen numbers and Mach numbers can vary by orders of magnitude. To aid particle trajectory simulations, we trained a neural network to predict the drag force on the particles based on existing experimental data, theoretical limits, and new direct simulation Monte Carlo (DMSC) results. The neural network based drag law, which depends upon both Mach and Knudsen numbers, shows better agreement with the DSMC simulation data than pre-existing correlations. With it, particle trajectory simulation results reveal that for a given particle density, there exists an optimal particle diameter to maximize particle impaction speed. We also find that in AD particles undergo size dependent inertial focusing, i.e. there is a particular particle diameter where the particle deposition linewidth is minimized. Particles smaller than this diameter are underfocused, and particles larger than this are overfocused, and hence have larger deposition linewidths in both cases. Using trajectory simulations, we additionally developed a framework that can be used to evaluate the position-dependent mass, momentum and kinetic energy fluxes to the deposition substrate for any aerosol size distribution function upstream of the nozzle. It is shown for typical aerosol concentrations achievable in the laboratory, the kinetic energy flux can approach a magnitude normally observed in convective heat transfer with phase change, hence translational kinetic energy to thermal energy transfer in AD is likely a key contributor to the formation of dense coatings.

KW - Aerosol deposition

KW - Converging-diverging nozzle

KW - Direct simulation Monte Carlo

KW - Inertial focusing

KW - Inertial impaction

UR - http://www.scopus.com/inward/record.url?scp=85054882979&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85054882979&partnerID=8YFLogxK

U2 - 10.1016/j.ijheatmasstransfer.2018.10.028

DO - 10.1016/j.ijheatmasstransfer.2018.10.028

M3 - Article

VL - 129

SP - 1161

EP - 1171

JO - International Journal of Heat and Mass Transfer

T2 - International Journal of Heat and Mass Transfer

JF - International Journal of Heat and Mass Transfer

SN - 0017-9310

ER -