The use of computational fluid dynamics to predict the turbulent dissipation rate and droplet size in a stirred autoclave

Research output: Contribution to journalArticle

Abstract

The prediction of droplet sizes in emulsions is important for fields ranging from the chemical process industry to emergency planning in the event of an underwater oil release. Typically scale models have needed to be built and the results scaled up, but as computational resources have grown and turbulence models have matured it has become possible to use computational fluid dynamics (CFD) to simulate the behaviour of the fluid/s. While direct simulation of multiphase breakup at high Reynolds number is currently computationally impractical, this paper looks into the use of CFD along with a correlation function based on maximum turbulent kinetic energy dissipation rate to predict the Sauter mean diameter of droplets in a 1 in. baffle-and-vane type autoclave. The results show that using a RNG-k∊ turbulence model with a simplified 2D geometry gave droplet sizes within 26.2 μm of the Sauter mean diameter observed in experiments with no additional tuning of parameters. Correlating pipe and autoclave flows through the Reynolds number and the turbulent kinetic energy dissipation rate was also investigated. Using the traditional definitions of the Reynolds numbers the correlation is poor, the coefficient of determination of the linear fit to the log-log data is 0.64. The first modification replaced the diameter of the blade as characteristic length with the tip swept circumference which increased the coefficient of determination to 0.960. A further modification using data obtained from the turbulent fields of the simulation showed a significant improvement with the coefficient of determination increasing to 0.988.

Original languageEnglish
Pages (from-to)433-443
Number of pages11
JournalChemical Engineering Science
Volume196
DOIs
Publication statusPublished - 16 Mar 2019

Fingerprint

Coefficient of Determination
Autoclaves
Computational Fluid Dynamics
Droplet
Reynolds number
Dissipation
Computational fluid dynamics
Energy Dissipation
Turbulence Model
Turbulence models
Kinetic energy
Predict
Energy dissipation
Baffle
Process Industry
Emulsion
Circumference
Datalog
Chemical Processes
Breakup

Cite this

@article{059d605944304fb1bf8a0c4a92a05211,
title = "The use of computational fluid dynamics to predict the turbulent dissipation rate and droplet size in a stirred autoclave",
abstract = "The prediction of droplet sizes in emulsions is important for fields ranging from the chemical process industry to emergency planning in the event of an underwater oil release. Typically scale models have needed to be built and the results scaled up, but as computational resources have grown and turbulence models have matured it has become possible to use computational fluid dynamics (CFD) to simulate the behaviour of the fluid/s. While direct simulation of multiphase breakup at high Reynolds number is currently computationally impractical, this paper looks into the use of CFD along with a correlation function based on maximum turbulent kinetic energy dissipation rate to predict the Sauter mean diameter of droplets in a 1 in. baffle-and-vane type autoclave. The results show that using a RNG-k∊ turbulence model with a simplified 2D geometry gave droplet sizes within 26.2 μm of the Sauter mean diameter observed in experiments with no additional tuning of parameters. Correlating pipe and autoclave flows through the Reynolds number and the turbulent kinetic energy dissipation rate was also investigated. Using the traditional definitions of the Reynolds numbers the correlation is poor, the coefficient of determination of the linear fit to the log-log data is 0.64. The first modification replaced the diameter of the blade as characteristic length with the tip swept circumference which increased the coefficient of determination to 0.960. A further modification using data obtained from the turbulent fields of the simulation showed a significant improvement with the coefficient of determination increasing to 0.988.",
keywords = "Computational fluid dynamics, Droplet size, Scaling, Turbulence modeling",
author = "Booth, {Craig P.} and Leggoe, {Jeremy W.} and Aman, {Zachary M.}",
year = "2019",
month = "3",
day = "16",
doi = "10.1016/j.ces.2018.11.017",
language = "English",
volume = "196",
pages = "433--443",
journal = "Chemical Engineering Science",
issn = "0009-2509",
publisher = "Elsevier",

}

TY - JOUR

T1 - The use of computational fluid dynamics to predict the turbulent dissipation rate and droplet size in a stirred autoclave

AU - Booth, Craig P.

AU - Leggoe, Jeremy W.

AU - Aman, Zachary M.

PY - 2019/3/16

Y1 - 2019/3/16

N2 - The prediction of droplet sizes in emulsions is important for fields ranging from the chemical process industry to emergency planning in the event of an underwater oil release. Typically scale models have needed to be built and the results scaled up, but as computational resources have grown and turbulence models have matured it has become possible to use computational fluid dynamics (CFD) to simulate the behaviour of the fluid/s. While direct simulation of multiphase breakup at high Reynolds number is currently computationally impractical, this paper looks into the use of CFD along with a correlation function based on maximum turbulent kinetic energy dissipation rate to predict the Sauter mean diameter of droplets in a 1 in. baffle-and-vane type autoclave. The results show that using a RNG-k∊ turbulence model with a simplified 2D geometry gave droplet sizes within 26.2 μm of the Sauter mean diameter observed in experiments with no additional tuning of parameters. Correlating pipe and autoclave flows through the Reynolds number and the turbulent kinetic energy dissipation rate was also investigated. Using the traditional definitions of the Reynolds numbers the correlation is poor, the coefficient of determination of the linear fit to the log-log data is 0.64. The first modification replaced the diameter of the blade as characteristic length with the tip swept circumference which increased the coefficient of determination to 0.960. A further modification using data obtained from the turbulent fields of the simulation showed a significant improvement with the coefficient of determination increasing to 0.988.

AB - The prediction of droplet sizes in emulsions is important for fields ranging from the chemical process industry to emergency planning in the event of an underwater oil release. Typically scale models have needed to be built and the results scaled up, but as computational resources have grown and turbulence models have matured it has become possible to use computational fluid dynamics (CFD) to simulate the behaviour of the fluid/s. While direct simulation of multiphase breakup at high Reynolds number is currently computationally impractical, this paper looks into the use of CFD along with a correlation function based on maximum turbulent kinetic energy dissipation rate to predict the Sauter mean diameter of droplets in a 1 in. baffle-and-vane type autoclave. The results show that using a RNG-k∊ turbulence model with a simplified 2D geometry gave droplet sizes within 26.2 μm of the Sauter mean diameter observed in experiments with no additional tuning of parameters. Correlating pipe and autoclave flows through the Reynolds number and the turbulent kinetic energy dissipation rate was also investigated. Using the traditional definitions of the Reynolds numbers the correlation is poor, the coefficient of determination of the linear fit to the log-log data is 0.64. The first modification replaced the diameter of the blade as characteristic length with the tip swept circumference which increased the coefficient of determination to 0.960. A further modification using data obtained from the turbulent fields of the simulation showed a significant improvement with the coefficient of determination increasing to 0.988.

KW - Computational fluid dynamics

KW - Droplet size

KW - Scaling

KW - Turbulence modeling

UR - http://www.scopus.com/inward/record.url?scp=85056638113&partnerID=8YFLogxK

U2 - 10.1016/j.ces.2018.11.017

DO - 10.1016/j.ces.2018.11.017

M3 - Article

VL - 196

SP - 433

EP - 443

JO - Chemical Engineering Science

JF - Chemical Engineering Science

SN - 0009-2509

ER -