Download the full version of the ebook at ebookname.

com

Fluid Dynamics and Transport of Droplets and

Sprays 2ed. Edition William A. Sirignano

https://ebookname.com/product/fluid-dynamics-and-transport-
of-droplets-and-sprays-2ed-edition-william-a-sirignano/

OR CLICK BUTTON

DOWNLOAD EBOOK

Download more ebook instantly today at https://ebookname.com

Instant digital products (PDF, ePub, MOBI) available
Download now and explore formats that suit you...

Computational Fluid Dynamics Principles and Applications

Jiri Blazek

https://ebookname.com/product/computational-fluid-dynamics-principles-
and-applications-jiri-blazek/

ebookname.com

Dynamics of Markets 2ed. Edition Joseph L. Mccauley

https://ebookname.com/product/dynamics-of-markets-2ed-edition-joseph-
l-mccauley/

ebookname.com

The Dynamics of Socio Economic Development An Introduction

2ed. Edition Szirmai A.

https://ebookname.com/product/the-dynamics-of-socio-economic-
development-an-introduction-2ed-edition-szirmai-a/

ebookname.com

Professional Cocoa Application Security 1st Edition Graham

J. Lee

https://ebookname.com/product/professional-cocoa-application-
security-1st-edition-graham-j-lee/

ebookname.com
An invitation to cognitive science 2nd Edition Lila R.
Gleitman

https://ebookname.com/product/an-invitation-to-cognitive-science-2nd-
edition-lila-r-gleitman/

ebookname.com

Working With Dreams Initiation into the Soul s Speaking

About Itself 1st Edition Wolfgang Giegerich

https://ebookname.com/product/working-with-dreams-initiation-into-the-
soul-s-speaking-about-itself-1st-edition-wolfgang-giegerich/

ebookname.com

The Method of the Divine Government Physical and Moral 2nd

Edition James Mccosh

https://ebookname.com/product/the-method-of-the-divine-government-
physical-and-moral-2nd-edition-james-mccosh/

ebookname.com

Handbook of Affective Sciences 1st Edition Richard J.

Davidson

https://ebookname.com/product/handbook-of-affective-sciences-1st-
edition-richard-j-davidson/

ebookname.com

The Importance of Being Earnest Webster s German Thesaurus

Edition Oscar Wilde

https://ebookname.com/product/the-importance-of-being-earnest-webster-
s-german-thesaurus-edition-oscar-wilde/

ebookname.com
Pharmaceutical Industry Innovation and Developments
Innovation and Developments 1st Edition David A. Mancuso

https://ebookname.com/product/pharmaceutical-industry-innovation-and-
developments-innovation-and-developments-1st-edition-david-a-mancuso/

ebookname.com
This page intentionally left blank
FLUID DYNAMICS AND TRANSPORT OF DROPLETS
AND SPRAYS

Second Edition

This book serves as both a graduate text and a reference for engineers
and scientists exploring the theoretical and computational aspects of
the fluid dynamics and transport of sprays and droplets. Attention is
given to the behavior of individual droplets, including the effects of
forced convection due to relative droplet–gas motion, Stefan convec-
tion due to the vaporization or condensation of the liquid, multicom-
ponent liquids (and slurries), and internal circulation of the liquid. This
second edition contains more information on droplet–droplet interac-
tions, the use of the mass-flux potential, conserved scalar variables,
spatial averaging and the formulation of the multicontinua equations,
the confluence of spatial averaging for sprays and filtering for tur-
bulence, direct numerical simulations and large-eddy simulations for
turbulent sprays, and high-pressure vaporization processes. Two new
chapters introduce liquid-film vaporization as an alternative to sprays
for miniature applications and a review of liquid-stream distortion and
breakup theory, which is relevant to spray formation.

William A. Sirignano is the Henry Samueli Professor of Mechanical

and Aerospace Engineering and former Dean of the School of Engi-
neering at the University of California, Irvine. Before that, he was the
George Tallman Ladd Professor and Department Head at Carnegie-
Mellon University and a professor at Princeton University. His major
research and teaching interests include spray combustion, turbulent
combustion and ignition, aerospace propulsion, fluid dynamics, and
applied mathematics. Dr. Sirignano has written more than 450 research
papers, book articles, and reports and has given more than 300 confer-
ence presentations and research seminars. He has been a formal con-
sultant to 30 industrial organizations and federal laboratories.
Sirignano is a member of the National Academy of Engineering
and a Fellow of the AIAA, ASME, AAAS, APS, and SIAM. He is the
recipient of the Pendray Aerospace Literature Award, Propellants and
Combustion Award, Energy Systems Award, Wyld Propulsion Award,
and Sustained Service Award from the AIAA; the ASME Freeman
Scholar Fluids Engineering Award; The Combustion Institute Alfred
C. Egerton Gold Medal; and the Institute for the Dynamics of Explo-
sions and Reactive Systems Oppenheim Award.
Fluid Dynamics and Transport of Droplets
and Sprays

SECOND EDITION

William A. Sirignano
University of California, Irvine
CAMBRIDGE UNIVERSITY PRESS
Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore,
São Paulo, Delhi, Dubai, Tokyo

Cambridge University Press

The Edinburgh Building, Cambridge CB2 8RU, UK

Published in the United States of America by Cambridge University Press, New York

www.cambridge.org
Information on this title: www.cambridge.org/9780521884891
© William A. Sirignano 2010

This publication is in copyright. Subject to statutory exception and to the

provision of relevant collective licensing agreements, no reproduction of any part
may take place without the written permission of Cambridge University Press.
First published in print format 2010

ISBN-13 978-0-511-67564-5 eBook (NetLibrary)

ISBN-13 978-0-521-88489-1 Hardback

Cambridge University Press has no responsibility for the persistence or accuracy

of urls for external or third-party internet websites referred to in this publication,
and does not guarantee that any content on such websites is, or will remain,
accurate or appropriate.
To my wife and children, Lynn Sirignano, Justin Sirignano,
Monica Sirignano, Jacqueline Vindler.
Contents

Preface xi
Nomenclature xiv

1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1 Overview 1
1.2 Droplet-Size Determination 4

2 Isolated Spherically Symmetric Droplet Vaporization and Heating . . . . 8

2.1 Theory of Spherically Symmetric Droplet Vaporization and Heating 11
2.1.1 Gas-Phase Analysis 12
2.1.2 Liquid-Phase Analysis 19
2.1.3 Chemical Reaction 24
2.2 Radiative Heating of Droplets 26

3 Convective Droplet Vaporization, Heating, and Acceleration . . . . . . . 30

3.1 Convective Droplet Vaporization 31
3.1.1 Evaluation of Reynolds Number Magnitude 33
3.1.2 Physical Description 35
3.1.3 Approximate Analyses for Gas-Phase Boundary Layer 40
3.1.4 Approximate Analyses for Liquid-Phase Flows 47
3.1.5 Droplet Drag Coefficients 56
3.1.6 Results from Approximate Analyses 57
3.1.7 Exact Analyses for Gas-Phase and Liquid-Phase Flows 64
3.1.8 Free Convection 71
3.2 Low Reynolds Number Behavior 73
3.3 Droplet Vaporization in an Oscillating Gas 76
3.4 Individual Droplet Behavior in an Unsteady Flow 79

4 Multicomponent-Liquid Droplets . . . . . . . . . . . . . . . . . . . . . . . . . 90
4.1 Spherically Symmetric Diffusion 93
4.1.1 Continuous-Thermodynamics Models 97

vii
viii Contents

4.2 Liquid-Phase Mass Diffusion with Convective Transport 98

4.2.1 Approximate Analyses 98
4.2.2 Exact Analyses 106
4.3 Metal-Slurry Droplet Vaporization and Combustion 107
4.3.1 Burning of a Fuel Droplet Containing a Single Metal Particle 108
4.3.2 Liquid Vaporization from Fine-Metal-Slurry Droplets 116
4.3.3 Metal-Particle Combustion with Oxide Condensation 129
4.4 Emulsified-Fuel-Droplet Vaporization and Burning 130

5 Droplet Behavior under Near-Critical, Transcritical, and

Supercritical Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
5.1 High-Pressure Droplet Behavior in a Quiescent Environment 136
5.2 Convective Effects and Secondary Atomization 143
5.3 Molecular-Dynamics Simulation of Transcritical Droplet
Vaporization 147

6 Droplet Arrays and Groups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150

6.1 Heating and Vaporization of Droplet Arrays 153
6.2 Group Vaporization and Combustion 165
6.3 Generalized Theory for Droplet-Array Vaporization and Burning 168
6.3.1 Basic Formulation 168
6.3.2 Analysis of Vaporization Without Combustion 170
6.3.3 Combustion Analysis 173
6.3.4 Array Combustion with Nonunitary Lewis Number 179
6.3.5 Array Vaporization with Multicomponent Liquids 189
6.4 Droplet Collisions 192
6.4.1 Droplet–Droplet Collisions 193
6.4.2 Droplet–Wall Collisions 196

7 Spray Equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199

7.1 Averaging Process for Two-Continua Formulations 200
7.1.1 Averaging of Dependent Variables 204
7.1.2 Averaging of Derivatives 207
7.1.3 Averaged Gas-Phase Equations 210
7.1.4 Averaged Vorticity and Entropy 214
7.1.5 Averaged Liquid-Phase Partial Differential Equations 216
7.1.6 Averaged Liquid-Phase Lagrangian Equations 218
7.1.7 The Microstructure 220
7.2 Two-Continua and Multicontinua Formulations 223
7.2.1 Continuity Equations 223
7.2.2 Momentum Conservation 226
7.2.3 Energy Conservation 228
7.2.4 Hyperbolic Character of Liquid-Phase Equations 230
7.2.5 Subgrid Models for Heat, Mass, and Momentum Exchange 232
7.3 Discrete-Particle Formulation 233
7.4 Probabilistic Formulation 234
Contents ix

8 Computational Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237

8.1 Efficient Algorithms for Droplet Computations 237
8.2 Numerical Schemes and Optimization for Spray Computations 245
8.2.1 Two-Phase Laminar Axisymmetric Jet Flow 246
8.2.2 Axisymmetric Unsteady Sprays 255
8.2.3 Solution for Pressure 269
8.3 Point-Source Approximation in Spray Calculations 269

9 Spray Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285

9.1 Spherically Symmetric Spray Phenomena 287
9.2 Counterflow Spray Flows 289
9.3 One-Dimensional Planar Spray Ignition and Flame Propagation 296
9.4 Vaporization and Combustion of Droplet Streams 301
9.5 Flame Propagation Through Metal-Slurry Sprays 305
9.6 Liquid-Fueled Combustion Instability 308
9.7 Spray Behavior in Near-Critical and Supercritical Domains 310
9.8 Influence of Supercritical Droplet Behavior on Combustion
Instability 311

10 Spray Interactions with Turbulence and Vortical Structures . . . . . . . . 314

10.1 Vortex–Spray Interactions 318
10.2 Time-Averaged Turbulence Models 321
10.3 Direct Numerical Simulation 324
10.4 Large-Eddy Simulations 329
10.4.1 Proper Two-Way Coupling for LES Closure 332
10.4.2 Gas-Phase Equations 333
10.4.3 Liquid-Phase Equations 335
10.4.4 Vortex–Droplet Interactions 336

11 Film Vaporization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 340

11.1 Introduction 340
11.2 Miniature Film-Combustor Concept 342
11.3 Analysis of Liquid-Film Combustor 347
11.3.1 Assumptions and Governing Equations 348
11.3.2 Liquid-Phase Thermal Analysis 349
11.3.3 Fluid-Dynamics Analysis 350
11.3.4 Scalar Analysis 351
11.3.5 Results 354
11.4 Concluding Remarks 360

12 Stability of Liquid Streams . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 361

12.1 Introduction 361
12.2 Formulation of Governing Equations 364
12.3 Round Jet Analyses 366
12.3.1 Temporal Stability Analysis 367
12.3.2 Surface Energy 368
x Contents

12.3.3 Spatial Stability Analysis 370

12.3.4 Nonlinear Effects 371
12.3.5 Viscous Effects 376
12.3.6 Cavitation 376
12.4 Planar Sheet Analyses 381
12.4.1 Linear Theory 381
12.4.2 Fan Sheets 385
12.4.3 Nonlinear Theory 385
12.5 Annular Free Films 396
12.5.1 Linear Theory 397
12.5.2 Nonlinear Theory 399
12.5.3 Effect of Swirl 401
12.6 “Conical” Free Films 402
12.7 Concluding Remarks 406

Appendix A The Field Equations . . . . . . . . . . . . . . . . . . . . . . . . . . . 409

Appendix B Conserved Scalars . . . . . . . . . . . . . . . . . . . . . . . . . . . . 415

Appendix C Droplet-Model Summary . . . . . . . . . . . . . . . . . . . . . . . . 422

Bibliography 427
Index 460
Preface

The fluid dynamics and transport of sprays comprise an exciting field of broad
importance. There are many interesting applications of spray theory related to
energy and power, propulsion, heat exchange, and materials processing. Spray phe-
nomena also have natural occurrences. Spray and droplet behaviors have a strong
impact on vital economic and military issues. Examples include the diesel engine and
gas-turbine engine for automotive, power-generation, and aerospace applications.
Manufacturing technologies including droplet-based net form processing, coating,
and painting are important applications. Applications involving medication, pesti-
cides and insecticides, and other consumer uses add to the impressive list of im-
portant industries that use spray and droplet technologies. These industries involve
annual production certainly measured in tens of billions of dollars and possibly
higher. Many applications are still under development. The potentials for improved
performance, improved market shares, reduced costs, and new products and appli-
cations are immense. Continuing effort is needed to optimize the designs of spray
and droplet applications and to develop strategies and technologies for active con-
trol of sprays in order to achieve the huge potential.
In the first edition of this book and in this second edition, I have attempted to
provide some scientific foundation for movement toward the goals of optimal design
and effective application of active controls. The book, however, does not focus on
design and controls. Rather, I discuss the fluid mechanics and transport phenomena
that govern the behavior of sprays and droplets in the many important applications.
Various theoretical and computational aspects of the fluid dynamics and transport of
sprays and droplets are reviewed in detail. I undertook this writing because no pre-
vious treatise exists that broadly addresses theoretical and computational issues re-
lated to both spray and droplet behavior. There are other books that address either
sprays on a global scale or individual isolated droplets on the fine scale. However,
no other book has attempted a true integration of these two critically related topics.
My research interests have focused on the theoretical and computational aspects of
the spray problem. Therefore, this book emphasizes those aspects. Major but not
total attention is given to the works of my research team because we have many re-
search publications and review papers on this subject. On the basis of these research

xi
xii Preface

studies over the years, a decent comprehensive portrayal of the field is achievable.
I have given some emphasis to liquid-fuel droplets and to combustion applications
because my experience is centered in that domain and, more important, because the
high temperatures and rapid vaporization make the dynamics of the phenomenon
much more interesting and general. Rapidly vaporizing sprays have a richness of the
scientific phenomena and several, often disparate, time scales. The discussions are
often also relevant to other important applications including materials processing,
heat exchange, and coatings. Because the field of droplet and spray studies is still
developing in terms of both science and technology, a critical review is undertaken
here.
This book was developed largely on the basis of my lecture notes generated dur-
ing several offerings of a graduate course. This treatise can serve both as a graduate-
level text and as a reference book for scientists and engineers.
All of the material from the first edition is retained here, although much of
it has been reorganized into different chapters. Attention is given to the behavior
of individual droplets, including the effects of forced convection that are due to
relative droplet–gas motion, Stefan convection that is due to the vaporization or
condensation of the liquid, multicomponent liquids (and slurries), and internal cir-
culation of the liquid. Flow-field details in the gas boundary layer and wake and
in the liquid-droplet interior are examined. Also, the determinations of droplet lift
and drag coefficients and Nusselt and Sherwood numbers and their relationships
with Reynolds number, transfer number, Prandtl and Schmidt numbers, and spac-
ing between neighboring droplets are extensively discussed. Results from droplet
analyses are presented in a manner that makes them useful as subgrid models in
spray computations. Several examples of spray computations for which these mod-
els are used are presented. The two-phase flow equations governing spray behavior
are presented in various forms and thoroughly discussed. Attention is given to issues
of computational accuracy and efficiency. Various configurations for spray flows are
studied. Droplet interactions with vortical and turbulent fields are analyzed. Droplet
behavior under near-critical and supercritical conditions is discussed.
In addition to updating and reorganizing the material from the first edition, new
content has been added. This second edition is more than 50% longer than the first
edition. More information has been added on the topics of droplet–droplet interac-
tions, the use of the mass-flux potential, conserved scalar variables, spatial averaging
and the formulation of the multicontinua equations, the confluence of spatial aver-
aging for sprays and filtering for turbulence, direct numerical simulations and large-
eddy simulations for turbulent sprays, and high-pressure vaporization processes. A
new chapter has been included on liquid-film vaporization as an alternative to sprays
for miniature applications. Another new chapter has a review of theory on liquid-
stream distortion and breakup, which is very relevant to spray formation.
My interactions over more than three decades with 13 postdoctoral associates
and 21 graduate students on the subject of sprays have been very productive, stimu-
lating, and instructive. These junior (at the time) collaborators are well represented
in the references. They are Boris Abramzon, Suresh K. Aggarwal, Nasser Ash-
griz, Rakesh Bhatia, Jinsheng Cai, C.H. “Jeff” Chiang, Gaetano Continillo, Sadegh
Dabiri, Jean-Pierre Delplanque, Amar Duvvur, Eva Gutheil, Howard Homan,
Preface xiii

Randall Imaoka, Inchul Kim, Pedro Lara-Urbaneja, C.K. “Ed” Law, D.N. Lee,
Steven Lerner, Jun Li, Mansour Masoudi, Constantine Megaridis, Carsten Mehring,
Kamyar Molavi, Gopal Patnaik, Satya Prakash, M.S. Raju, Roger Rangel, David
Schiller, Bartendu Seth, Simone Stanchi, Douglas Talley, Albert Tong, Guang Wu,
and Jinxiang Xi. Exciting interactions with senior collaborators are also recognized:
H.A. Dwyer, D. Dunn-Rankin, S.E. Elghobashi, G.J. Fix, D.D. Joseph, F. Liu,
V.G. McDonell, B.R. Sanders, E. Suuberg, and S.C. Yao are identified here. Several
federal funding agencies and industrial organizations have been supportive of my
spray research; special recognition for continuing support goes to Julian Tishkoff of
AFOSR; David Mann, Kevin McNesby, and Ralph Anthenien of ARO; and Gabriel
Roy of ONR. Here I am acknowledging only those who worked with me or sup-
ported me specifically in the area of sprays. There are many others to thank for
associations on other scientific problem areas. Also, I am thankful for the opportu-
nities for intellectual exchanges and friendship with many individuals from around
the world who have contributed to the disciplines of fluid dynamics, transport, and
combustion and/or to technologies for energy, power, and propulsion.
 Nomenclature

a droplet acceleration; constant of curvature in stagnation-point

flow; half of undisturbed sheet thickness
a, b constants in Eq. (1.3); variable parameters in Section 3.1,
Eq. (3.48); also constant parameters in equation of state (5.3)
A constant in Section 3.1; area
à liquid-vortex strength
bT , bM corrections to transfer number
B Spalding transfer number
BH energy transfer number
BM mass transfer number
Bo = ρl a R2 /σ Bond number
cl liquid specific heat
cp specific heat at constant pressure
C D, C L drag and lift coefficients
CF friction coefficient
d droplet diameter; distance of vortex from droplet path
D mass diffusivity; droplet center-to-center spacing; drag force;
orifice diameter
e thermal energy; correction in Section 8.3 for application of am-
bient conditions
e unit vector
E activation energy
Ei externally imposed electric field
f Blasius function; droplet distribution function (probability
density function)
fH defined by Eq. (3.72e)
fi fugacity of species i
F drag force per unit mass on droplet; friction force
F Di aerodynamic forces per unit volume on droplets
F, G functions defined in Section 3.1, Eqs. (3.44)

xiv
Nomenclature xv

Fr Froude number
G response factor for combustion instability; also Chiu group
combustion number
g acceleration due to gravity
g 1 , g2 functions defined in Section 3.1 by Eq. (3.44)
Gr = g R3 /ν 2 Grashof number
h enthalpy; heat transfer coefficient; scale factors; film thickness
In modified Bessel function of the first kind
k nondimensional constant in Section 3.1; also turbulence wave
number
k, l, m, n wave numbers
kE generalized Einstein coefficient
K constant in vaporization-rate law; constants in model equa-
tions; strain rate; cavitation number
K(t, t − τ ) kernel in history integral, Section 3.4
lc characteristic length for liquid-stream pinch-off
L latent heat of vaporization; differential operator; orifice length
Le Lewis number
m droplet mass
ṁ droplet mass vaporization rate
mf mass of fluid displaced by droplet or particle
mp mass of particle or droplet
Ṁ mass source term (rate per unit volume); vaporization rate per
unit volume for spray
MA1 , MA2 acceleration numbers defined by Eqs. (3.65) and (3.68)
n droplet number density; direction normal to interface; fuel
mass flux
ni number of moles of species i
n unit normal vector
N number of droplets; number of species in multicomponent
mixture; ratio of droplet heating time to droplet lifetime
N number of droplets in array
Nu Nusselt number
Oh Ohnesorge number
p pressure
Pe Peclet number
Pr Prandtl number
qα electric charge on droplet
q̇ heat flux
Q energy per unit mass of fuel
r spherical radial coordinate
r̃ cylindrical radial coordinate
r
nondimensional annular radius perturbation
rf flame radius
R droplet radius, radius of curvature of liquid sheet
R2 droplet-radii ratio
xvi Nomenclature

R gas constant
Re = RU/ν droplet Reynolds number (In discussions in which a Reynolds
number based on diameter or some special properties is used,
it is specified in the text.)
s nondimensional radius; transformed variable defined by Eq.
(9.14b)
S superscalar; weighted area in numerical interpolation schemes;
distribution function for radiative heat transfer; jet or droplet-
surface area
Sc Schmidt number
Sh Sherwood number
St Strouhal number
t time
tc characteristic time for liquid-stream pinch-off
T temperature
u, v velocity
uo mean jet velocity
U free-stream velocity; particle or droplet velocity; Bernoulli jet
velocity
v velocity in the argument of distribution function; specific vol-
ume
v velocity vector
V volume in eight-dimensional phase space; liquid volume; char-
acteristic velocity
ẇ chemical-reaction rate, fractional change per unit time
W molecular weight
We = ρl R(U) /σ Weber number
2

x spatial coordinate; stream coordinate in planar case

X mole fraction; position
y normal coordinate in boundary layer; transverse coordinate in
planar case
y liquid sheet centerline position
ỹ liquid sheet thickness
Y nondimensional sheet centerline position
Yn mass fraction of species n
z axial or downstream spatial coordinate in cylindrical coordi-
nates; spanwise coordinate
Z nondimensional temperature

Greek Symbols
α thermal diffusivity; wave phase; growth rate
β Shvab–Zel’dovich variables
γ reciprocal of Lewis number; fraction of radiation absorbed at surface and in
interior; ratio of specific heats
circulation
Nomenclature xvii

δ distance ratio in Section 6.3; thickness of oxide layer; fraction of radiation

absorbed at droplet surface; Dirac delta function; differential quantity
δ̃ upstream distance for application of boundary condition
δ99 the point, measured from the stagnation point, where 99% of the mass frac-
tion variation has occurred
 change in a quantity
U relative droplet velocity
ε radiative emissivity
εn mass-flux fraction for species n
ε0 electrical permittivity of a vacuum
η Blasius coordinate; nondimensional radius
θ void volume fraction; porosity; azimuthal angle
κ reciprocal of characteristic length of temperature variation
λ thermal conductivity; eigenvalues
µ dynamic viscosity; chemical potential
ν stoichiometric coefficient (mass of fuel per mass of oxygen); kinematic
viscosity
ρ density
ρr ratio of particle (or droplet) density to gas density
ρ̄ bulk density
σ surface-tension coefficient; Stefan–Boltzmann constant, nondimensional
vortex-core size
τ nondimensional time; viscous stress
τH droplet-heating time
τL droplet lifetime
τ∗ droplet-heating time with uniform temperature
φ velocity potential; mass-flux potential; normalized stream function; entropy
variable defined by Eq. (7.18)
φm metal volume fraction
φr ratio of acceleration numbers defined by Eq. (3.68)
ϕ generic variable
 fractional volume of fluid
χ ratio of effective thermal diffusivity to thermal diffusivity; ratio of thermal
diffusivities for slurry droplets
ψ liquid volume fraction; stream function
ω vorticity; frequency

Subscripts
0 initial condition
∞ condition at infinity
Al aluminum property
b boiling point Al (aluminum property); bubble
c critical conditions
e edge of boundary layer
eff effective value
xviii Nomenclature

F fuel vapor
g gas phase
H related to thermal transfer
i index for vectorial component; index for initial value
j index for vectorial component
k integer index
L lift
l liquid phase
M related to mass transfer
max maximum
m index for species; index for metal
mix mixture
n, p integers for numerical mesh points
N nitrogen
ox oxide
O oxygen
p particle
P product
r, θ, z components in cylindrical coordinates
s droplet-surface condition
T related to thermal transfer
wb wet bulb
+, − upper or lower liquid–gas interface

Superscripts
k index for droplet group
. rate or time derivative


derivative; perturbation quantity
 1 Introduction

1.1 Overview
A spray is one type of two-phase flow. It involves a liquid as the dispersed or dis-
crete phase in the form of droplets or ligaments and a gas as the continuous phase. A
dusty flow is very similar to a spray except that the discrete phase is solid rather than
liquid. Bubbly flow is the opposite kind of two-phase flow wherein the gas forms the
discrete phase and the liquid is the continuous phase. Generally, the liquid density
is considerably larger than the gas density; so bubble motion involves lower kine-
matic inertia, higher drag force (for a given size and relative velocity), and different
behavior under gravity force than does droplet motion.
Important and intellectually challenging fluid-dynamic and -transport phenom-
ena can occur in many different ways with sprays. On the scale of an individual
droplet size in a spray, boundary layers and wakes develop because of relative
motion between the droplet center and the ambient gas. Other complicated and
coupled fluid-dynamic factors are abundant: shear-driven internal circulation of the
liquid in the droplet, Stefan flow that is due to vaporization or condensation, flow
modifications that are due to closely neighboring droplets in the spray, hydro-
dynamic interfacial instabilities leading to droplet-shape distortion and perhaps
droplet shattering, and droplet interactions with vortical structures in the gas flow
(e.g., turbulence).
On a much larger and coarser scale, we have the complexities of the integrated
exchanges of mass, momentum, and energy of many droplets in some subvolume of
interest with the gas flow in the same subvolume. The problem is further compli-
cated by the strong coupling of the phenomena on the different scales; one cannot
describe the mass, momentum, and energy exchanges on a large scale without de-
tailed knowledge of the fine-scale phenomena. Note that, in some practical appli-
cations, these scales can differ by several orders of magnitude so that a challenging
subgrid modelling problem results.
Detailed consideration will be given to applications in which the mass vaporiza-
tion rate is so large that the physical behavior is modified. This is the most complex
situation, and therefore its coverage leads to the most general formulation of the

1
2 Introduction

theory. In particular, as the vaporization rate increases, the coupling between the
two phases becomes stronger and, as the droplet lifetime becomes as short as some
of the other characteristic times, the transient or dynamic character of the problem
emerges in a dominant manner.
The fast vaporization rate is especially prominent in situations in which the am-
bient gas is at very high temperatures (of the order of 1000 K or higher). Combustion
with liquid fuels is the most notable example here. The spray combustion regime is
a most interesting limiting case of the more general field of thermal and dynamic be-
havior of sprays. In the high-temperature domain, rapid vaporization causes droplet
lifetimes to be as short as the time for a droplet to heat throughout its interior. It
can be shorter than the time for liquid-phase mass diffusion to result in the mixing of
various components in a multicomponent liquid. The combustion limit is inherently
transient from the perspective of the droplet, richer in terms of scientific issues, and
more challenging analytically and numerically than low-temperature spray prob-
lems. Vaporization might still be longer than other combustion processes such as
mixing or chemical reaction; therefore it could be the rate-controlling process for
energy conversion.
The spray problem can be complicated by the presence of spatial temperature
and concentration gradients and internal circulation in the liquid. Interaction among
droplets is another complication to be treated.
There is a great disparity in the magnitudes of the scales. Liquid-phase mass dif-
fusion is slower than liquid-phase heat diffusion, which, in turn, is much slower than
the diffusion of vorticity in the liquid. Transport in the gas is faster than transport
in the liquid. Droplet diameters are typically of the order of a few tens of microm-
eters (µm) to a few hundreds of micrometers in diameter. Resolution of internal
droplet gradients can imply resolution on the scale of micrometers or even on a sub-
micrometer scale. Combustor or flow chamber dimensions can be 5 to 6 orders of
magnitude greater than the required minimum resolution. Clearly, subgrid droplet-
vaporization models are required for making progress on this problem.
Experiments have been successful primarily in resolving the global character-
istics of sprays. The submillimeter scales associated with the spray problem have
made detailed experimental measurements very difficult. If an attempt is made to in-
crease droplet size, similarity is lost; the droplet Reynolds number can be kept con-
stant by decreasing velocity but the Grashof number grows, implying that buoyancy
becomes relatively more important. Also, the Weber number increases as droplet
size increases; surface tension becomes relatively less important, and the droplet is
more likely to acquire a nonspherical shape. Modern nonintrusive laser diagnostics
have made resolution possible on a scale of less than 100 µm so that, in recent years,
more experimental information has been appearing. Nevertheless, theory and com-
putation have led experiments in terms of resolving the fluid-dynamical characteris-
tics of spray flows.
Classical texts on droplets, including burning-fuel droplets, tend to consider an
isolated spherical droplet vaporizing in a stagnant environment. In the simplified
representation, the liquid has one chemical component, ambient-gas conditions are
subcritical, and vaporization occurs in a quasi-steady fashion. The classical result
is that the square of the droplet radius or diameter decreases linearly with time
1.1 Overview 3

because heat diffusion and mass diffusion in the surrounding gas film are the rate-
controlling (slowest) processes; this behavior is described as the d2 law. These im-
portant phenomena are discussed in later chapters. Although most researchers are
now addressing these relevant and interesting factors that cause major deviations
from classical behavior, there are still some researchers who persist in the study of
the classical configuration. Here we will relax these simplifications, one at a time, to
gain a more accurate and more relevant understanding. Convective effects that are
due to droplet motion or natural convection and subsequent internal liquid circu-
lation are thoroughly studied. Transient heating (or cooling) and vaporization (or
condensation) that are due to changing ambient conditions, unsteady liquid-phase
diffusion, or unsteady gas-phase diffusion are analyzed. Multicomponent-liquid (in-
cluding emulsions and slurries as well as blended liquids) droplet vaporization is
studied. Near-critical and supercritical ambient conditions (and their effects on dif-
fusion processes, phase change, solubility, and liquid-surface stripping that is due
to shear) will be discussed. Interactions of droplets with other droplets and with
turbulent or vortical structures are analyzed. Distortion of the spherical shape and
secondary atomization of the droplets are also discussed. The effects of radiative
heating of the liquid and of exothermic chemical reaction in the gas film are also
studied.
Current texts do not explain in a unified fashion the various approaches to cal-
culation of the behaviors of the many droplets present in a spray. Efficient and ac-
curate methods for predicting the trajectories, temperatures, and vaporization rates
of a large number of droplets in a spray are discussed here. Sprays in both laminar
and turbulent environments are discussed.
Some comments about primary atomization and droplet-size determination are
given in Section 1.2. In Chapters 2 and 3, we discuss the vaporization of individual
droplets and study the phenomenon on the scale of the droplet diameter. Chap-
ter 2 considers the case in which there is no relative motion between the droplet
and the distant gas, and Chapter 3 covers the situation with a relative velocity. The
theoretical models and correlations of computational results for individual droplets
can be used to describe exchanges of mass, momentum, and energy between the
phases in a spray flow. The vaporization of multicomponent droplets, including
slurry droplets, is discussed in Chapter 4. Droplet behavior under near-critical or
transcritical thermodynamic conditions is considered in Chapter 5. Secondary at-
omization and molecular-dynamic methods are also discussed there. Interactions
among droplets and their effects on the modification of the theory are discussed
in Chapter 6. The spray with its many droplets is examined first in Chapter 7. The
spray equations are examined from several aspects; in particular, two-continua, mul-
ticontinua, discrete-particle, and probabilistic formulations are given. The choice of
Eulerian or Lagrangian representation of the liquid-phase equations within these
formulations is discussed, including important computational issues and the rela-
tionship between the Lagrangian method and the method of characteristics. Some
specific computational issues are discussed in Chapter 8. Some of the theories and
information in this book have already had an impact on computational codes; mod-
ification of the codes to address more recent advances should not be difficult. One
shortcoming, of course, is the limited experimental verification, as just discussed.
4 Introduction

Applications of the spray theory to special laminar-flow configurations are discussed

in Chapter 9. Turbulence–spray interactions are surveyed in Chapter 10. Vorticity–
droplet interactions and turbulence–droplet interactions have not yet been fully in-
tegrated into a comprehensive spray theory. These interaction studies are still active
research domains, and, so far, little application to engineering practice has occurred.
In Chapter 11, we discuss exchanges of mass, momentum, and energy between the
phases for a type of configuration other than droplets and sprays, i.e., liquid films.
In Chapter 12, current research is discussed for the distortion and disintegration of
liquid streams in processes leading to the formation of sprays. Important material on
the underlying governing field equations, use of conserved scalars, and a summary
of droplet models are given in the Appendices.

1.2 Droplet-Size Determination

The droplet size is an important factor in its behavior. Droplet shape is another fac-
tor with profound implications. Surface tension will tend to minimize the droplet
surface area, given its volume, resulting in a spherical shape for sufficiently small
droplets. The size of a spherical droplet will be represented most commonly by its
diameter d or radius R. In most sprays, droplets of many different sizes will exist.
Vaporization, condensation, droplet coalescence, and droplet shattering will cause
a temporal variation in droplet sizes. For a spray, a distribution function of the in-
stantaneous diameter f (d) is typically used to describe a spray. This function gives
the number of droplets possessing a certain diameter. Often an average droplet di-
ameter dm n is taken to represent a spray. In particular,

∞
f (d)dm dd
dm n =
0∞ n
. (1.1)
0 f (d)d dd

In practice, f (d) will not be a continuous function. However, for a spray with many
droplets (millions can be common), the function is well approximated as a contin-
uously varying function. One example of an average droplet is the Sauter mean
diameter d32 , which is proportional to the ratio of the total liquid volume in a spray
to the total droplet-surface area in a spray.
The aerodynamic forces on a droplet will depend on its size in a functional man-
ner different from the dependence of droplet mass on the size. As a result, smaller
droplets undergo more rapid acceleration or deceleration than larger droplets.
Heating times and vaporization times will be shorter for smaller droplets. Accuracy
in the initial droplet-size distribution is mandatory, therefore, if we wish to predict
droplet behavior. Unfortunately, we must currently rely mostly on empirical meth-
ods to represent droplet distribution; it cannot be predicted from a first-principles
approach for most liquid-injection systems.
Liquid streams injected into a gaseous environment tend to be unstable under
a wide range of conditions. An important parameter is the Weber number,

ρU 2 L
We = , (1.2)
σ
1.2 Droplet-Size Determination 5

where ρ is the gas density, U is the relative gas–liquid velocity, L is the character-
istic dimension of the stream, and σ is the surface-tension coefficient. The Weber
number (We) is the ratio of the aerodynamic force related to dynamic pressure to
the force of surface tension. Depending on the stream shape, oscillation of the
stream and breakup occur above some critical value of the Weber number. These
interface oscillations can occur at any wavelength, but some wavelengths will have
larger rates of amplitude growth. Below the critical value of the Weber number,
the surface-tension forces are large enough to overcome the aerodynamic force that
tends to distort the stream. So here, the basic shape of the stream is maintained with-
out disintegration. At higher Weber numbers, the aerodynamic force dominates,
leading to distortion and disintegration. This process is called atomization.
Disintegration or atomization typically results in liquid ligaments or droplets
with a characteristic dimension that is smaller than the original length scale asso-
ciated with the stream. Disintegration will continue in a cascade fashion until the
decreased length scale brings the Weber number for the resulting droplets below
the critical value for the droplets. Other parameters will affect the critical value
of the Weber number; they include the ratio of liquid density to gas density and a
nondimensional representation of viscosity (e.g., Reynolds number).
Practical atomization systems use a variety of mechanisms to achieve the criti-
cal Weber numbers that are necessary. Jet atomizers use a sufficiently large pressure
drop across an orifice to obtain the necessary liquid velocity. Air-assist and air-blast
atomizers force air flow as well as liquid flow. The critical Weber number depends
on the relative air–liquid velocity here. Some atomizers use swirl vanes for the liq-
uid or air to create a tangential component of velocity; this can increase the relative
velocity. Rotary atomizers involve spinning cups or disks upon which the liquid is
flowed; the centrifugal effect creates the relative velocity. Sometimes other means
are used for atomization, including acoustic or ultrasonic oscillations, electrostatic
forces, and the injection of a bubbly liquid. An excellent review of practical atom-
ization systems is given by Lefebvre (1989).
There are three general approaches to the prediction of the droplet sizes that
result from atomization of a liquid stream. The most widely used approach involves
the use of empirical correlations. Another approach requires the solution of the
Navier–Stokes equations or of their inviscid limiting form, the Euler equations, to
predict disintegration of the liquid stream. Often the linearized form of the equa-
tions is taken. The third approach assumes that, in addition to conservation of mass
momentum and energy, the droplet-size distribution function satisfies a maximum-
entropy principle.
In the first approach, it is common practice to fit experimental data to a number-
distribution function for the droplet radius or diameter. With the current level of the
theory, this is the most commonly used approach. The Rosin–Rammler distribution
equation governs the volume of liquid contained in all droplets below a given diam-
eter d. In particular, the fractional volume of liquid (d) is described as

d  

3

0 f (d ) dd db

∞ ≡ (d) = 1 − exp − , (1.3)
3 a
0 f (d) dd
Random documents with unrelated
content Scribd suggests to you:
it was most probably in a contio that the supreme pontiff scourged to
death a man who had wronged a Vestal.[794] The people gathered in
the same kind of meeting to witness an oath,[795] a judicial process,
[796] or the levy of a fine.[797] But it was preëminently the listening

assembly, hence the definition offered by Gellius,[798] “To hold a

contio is to address the people, without calling on them for a vote.” It
applied not only to the isolated meeting summoned to hear edicts,
reports, communications of every kind, including arguments and
appeals for or against a given policy,[799] but also to the preparatory
stage of the voting assembly whether addresses were delivered or
not. Occasionally in early times there was speaking on the merits of
candidates at the opening of an electoral assembly, and the voting
was sometimes interrupted for the purpose.[800] Before the age of
Cicero this rare proceeding had disappeared. In his day the canvass
for candidates had been made before the holding of the preliminary
contio, which accordingly was brief and formal. Because much time
was required for the voting of the centuries,[801] speaking on the day
of their assembly had to be minimized. For this reason the contio for
advising the adoption of a resolution by the comitia centuriata was
held on an earlier day. Such was the meeting summoned in the
Campus Martius by the consul P. Lentulus for the purpose of urging
the people to vote in the ensuing centuriate assembly for the recall of
Cicero.[802] In judicial proceedings before any of the assemblies the
testimonies of witnesses and the pleadings occupied the greater part
of the time, and for this reason judicial assemblies were frequently
termed simply contiones. They will be described in a later chapter.
[803]

Informal contiones could be called at any time while the sun was
up,[804] with or without[805] an interval between the summons and
the meeting, and in any place[806] at the pleasure of the person who
convoked them. In public assemblies of every kind the people
remained standing throughout the session.[807] The magistrate who
was about to summon an auspicated contio repaired to the templum
which he intended to occupy during the meeting.[808] After taking the
auspices there, and finding the omens favorable, he ordered the
crier to call the citizens.[809] His directions to this assistant were
prefaced by a solemn wish that the gathering of the citizens might be
well, fortunate, auspicious, and advantageous to the Roman people,
the commonwealth, himself, his colleague or colleagues, and his
magistracy.[810] After first issuing the summons from the templum
the crier repeated it while making the circuit of the walls.[811]
Meantime the presiding officer invited the senators, his colleague or
colleagues, and the various other magistrates to assist him with their
presence and advice.[812] The invitation was extended even to
opposing tribunes;[813] and on the other hand a presiding tribune
was especially anxious to secure the presence and favorable
influence of patrician magistrates and of the leading men of the
state.[814] When the president saw his friends about him and the
people gathered, he called the contio to order,[815] and proceeded to
open the meeting with a prayer.[816] In the case of a resolution to be
brought before the assembly the magistrate was accustomed first to
submit it to the senate, which considered the bill and perhaps
suggested alterations;[817] but sometimes measures were brought
into the comitia without this senatorial deliberation.[818] In the contio
the presiding officer had power to exclude or to limit discussion of his
proposals. Ordinarily he found it advantageous to instruct the people
regarding the subject on which they were to vote; and it was for this
purpose that one or more contiones were held previous to the
comitia. The right to address the people belonged primarily to the
presiding magistrate. Although the king enjoyed the superior right,
the notion that in the regal period no private persons spoke in the
assembly seems to be unwarranted.[819] The custom of the republic
prescribed that the president should grant the privilege not only to
his colleagues[820] but also to all the higher magistrates.[821] In some
cases the invitation was extended to senators[822] and to other
distinguished private persons.[823] The tribunes sometimes gave the
privilege to freedmen,[824] to foreign kings,[825] and to ambassadors.
[826] The early republic did not allow women to be present in political
meetings;[827] but in time this severity began to relax. Livy[828]
represents the elder Cato as saying that his generation permitted
women to take part in affairs of state and to interfere in contiones
and comitia—evidently an exaggeration, as the context proves that
the women referred to did not actually come into the assembly, and
the speaker intimates that custom disapproved of their doing so.
From the time of the Gracchi they occasionally spoke in public. Dio
Cassius[829] states that Tiberius Gracchus brought his mother and
children into a contio to join their entreaties with his; and according
to Valerius Maximus[830] a tribune of the plebs required Sempronia,
sister of the Gracchi, to come forward in a similar meeting and give
her opinion on the subject under consideration. In the year 43 some
ladies attended a contio to protest against being taxed by the
triumvirs. Hortensia spoke for the complainants.[831] It was an
accepted custom that no tribune should intercede against a measure
till an opportunity had been afforded private persons to speak for or
against it.[832] When after the victory of Pydna a tribune of the plebs
had introduced a motion to grant a triumph to Aemilius Paulus, and
the debate had been thrown open to the assembly, all for a time
remained silent, for no doubt was entertained as to its passing; but
finally Servius Galba, who as a military tribune had served under
Paulus and was his enemy, came forward and obstructed the
measure by a long harangue.[833] Although the president could, and
perhaps often did, throw the debate open to the citizens in this way,
he was not compelled to do so. The tribunician assembly was more
deliberative than any other—a circumstance which accounts for its
designation as a concilium.[834] Those invited to speak, if citizens,
had to be of good standing and not under disqualification through a
special law or usage. The rex sacrorum was prohibited not only from
holding any other office but also from addressing an assembly.[835]
The spendthrift[836] and the man condemned for extortion[837] were
likewise forbidden. When the right was granted as a special
distinction, the receiver was probably placed thereby on a footing of
equal dignity with the magistrates.[838]
 The president could also compel a citizen to speak. The holder of
the imperium had a right to summon any man into a public meeting,
and order him to answer any question put to him.[839] Tribunes of the
plebs, however, who lacked the power of summoning, exercised this
coercive function against citizens and even consuls, not through a
direct right but by a usurpation, probably based on their power to
arrest and imprison.[840]
The president extended permission by asking a man to give his
opinion on the subject under discussion, and it was not in good order
even for a magistrate to address the assembly unless invited, though
he had ground for resentment if he was passed over in favor of
private persons.[841] When Caesar as consul, 59, brought his
agrarian bill before the comitia without the consent of the senate and
in spite of its silent disapproval,[842] he first asked his colleague
whether he had any objection to the proposal. Bibulus offered none
but declared that he would allow no innovation during his consulship.
Thereupon Caesar begged him for support, and requested the
people to join in the entreaty, saying, “You will have the law on the
sole condition that he is willing.” Then Bibulus, answering in a loud
voice, “You will not have this law the present year, even if all of you
want it,” left the assembly. Slighting the other magistrates, Caesar
invited Pompey and Crassus to address the meeting, though they
were but senators and therefore, as contrasted with magistrates,
merely private citizens. After Pompey had spoken at length,
commending the details of the law, Caesar asked if he would support
it against opponents, at the same time requesting the people to beg
of him this favor. They did so, doubtless by acclamation; and
Pompey, greatly flattered because the consul and the people
besought help of him, a private citizen, promised to stand by the law.
[843]

A magistrate spoke from the platform, a private person from a

lower position, presumably from one of the steps; for the chairman to
bring a private speaker upon the stage was a cause of offence to his
colleagues.[844] When the president granted an opportunity to speak,
he had a right to fix the amount of time to be used. In the debate on
the law for assigning provinces to Caesar and Pompey, 55, the
presiding tribune granted one hour to Favorinus and two hours to
Cato, both opponents of the measure.[845] The speaker could use
the time in whatever way he pleased; a few persons by concert
might waste the whole day in trivialities, as is sometimes done in the
senate of the United States of America, so as to prevent voting on
the subject for that date. In the case above mentioned Favorinus,
doubtless for lack of real argument, exhausted his hour in
lamentation over the shortness of the time allowed him,[846] and
Cato spent his two hours on irrelevant or minor matters, merely that
he might be silenced by the president while still appearing to have
something to say. He persisted in speaking accordingly till an officer
dragged him from the rostra and ejected him from the Forum. Even
then he returned several times to interrupt the proceedings with his
shouting.[847] If the speaker approved the measure, he might close
with the words, “This law of yours and your purpose and sentiments I
praise and most heartily approve”;[848] or more formally, “In my
opinion this bill as presented ought to be passed, and may it prove
well, auspicious, and fortunate both to yourselves and to the
republic”;[849] or if opposed to the proposition, he might conclude
with this form of disapproval, “It is my judgment that this law should
by no means be repealed.”[850]
Sometimes the magistrates were invited in the order of their rank
and afterward private citizens; in other cases, especially in
tribunician contiones, private persons were called first that they
might speak with perfect freedom, uninfluenced by the opinion of
their magistrates.[851] As the president had absolute control, he
could alter the usage to suit his own interest, and could certainly
reserve to himself the advantage of speaking last.[852] It often
happened that there was not enough time in one day for the
discussion of a question. In that case the magistrate adjourned the
meeting to a specified date.[853]
After the deliberation, or after the formality of opening the contio
which was merely preliminary to the comitia, the president ordered
the assembly to form into voting groups—curiae, centuries, or tribes.
He could say, for instance, “I order you to take your proper places in
the comitia centuriata,”[854] or more generally, “If you think fit,
quirites, move apart (into your voting groups).”[855] At the same time
he ordered the departure of all who lacked the qualification for
voting.[856] The lictors of the magistrates with imperium and the
beadle (viator) of the tribune attended to clearing away the
unqualified.[857]

Schulze, C. F., Volksversammlungen der Römer, 141 ff., 243 ff.; Rubino, J.,
Röm. Verfassung und Geschichte, 240-53; Lange, L., Röm. Altertümer, ii. 715-23,
and see indices s. v.; Mommsen, Th., Röm. Staatsrecht, i. 191-209; iii. 370-8, and
see index s. v.; Willems, P., Droit public Rom. 158 f.; Herzog, E., Röm.
Staatsverfassung, i. 632-6, 1057 f., and see index s. v.; Karlowa, O., Röm.
Rechtsgeschichte, i. 48 f., 379-81; Madvig, J. N., Verf. u. Verw. d. röm. Staates, i.
219; Soltau, W., Altröm. Volksversammlungen, 37 ff.; Humbert, G., Contio, in
Daremberg et Saglio, Dict. i. 1484 f.; Liebenam, W., Contio, in Pauly-Wissowa,
Real-Encycl. iv. 1149-53; Ruggiero, Diz. ep. ii. 1185, s. Contio; Lodge, G., Lex.
Plaut. i. 307, s. Contio; Forcellini, Tot. Lat. Lex. ii. 349 f., s. Concio; Dupond, A.,
Constitution et magistratures Rom. 60-3; Ihne, History of Rome, iv. 40-2.
 CHAPTER VIII
THE CALATA COMITIA

In seeking for the origin and primitive character of the Roman

assembly we are enabled by comparative study to reach a stage of
growth far anterior to the beginnings of Roman tradition. In its
earliest known form the European popular assembly had the
following characteristics, provisionally enumerated here, but
established in the next chapter: (1) the people who attended were
the mass of freemen of a tribe, especially the warriors; (2) they stood
or sat promiscuously, without reference to sub-tribal groups; (3)
measures were proposed by none but chiefs or nobles, generally
after previous discussion in council, the common members wholly
lacking initiative; (4) the speakers were as a rule, though not
exclusively, chieftains; (5) the vote was by acclamation, the clash of
weapons, or some similar demonstration; as a correlate of (3) and
(4) may be added, (6) sovereignty, so far as the idea existed, resided
not in the assembly, which of itself could take no action, but in the
king and chieftains, who made use of the assembly (a) for the
publication of news or of projects, (b) for securing by their eloquence
the coöperation of the tribe in a plan already formed in council.
However far developed beyond this crude institution the comitia
curiata or the comitia centuriata of the republican period may have
been, traces of all the characteristics above mentioned may be found
in the historical Roman assembly[858]—a fact which justifies the
comparative method of approach to the subject.
We need not hesitate to begin with the unorganized contio as the
earliest form of Roman assembly, to which we may attach the other
features of the European gathering named above. The first problem
is to determine under what influence and for what purpose the
gathering of the people came to be organized in curiae. The notion
that the object was primarily for voting is groundless. The Athenians
had the germ of a tribal assembly in the division of the people by
phylae on the occasion of ostracophory[859] and of the passing of
other privilegia (νόμοι ἐπ’ ἀνδρί). The organization was not in this
case for the purpose of using the tribes as voting units, but merely
for bringing order and solemnity to the proceeding. Apparently the
assembly of Alamanni was arrayed in military form for ratifying
emancipations,[860] though in the process the military companies did
not vote as units. In like manner, but for a wider range of functions,
we find at Rome the meeting of the people in curiae, less frequently
in centuries, merely for listening, for witnessing, or for receiving
purification. The circumstances that the business of such assemblies
was largely religious, and of such a character that it must have
originated in the earliest Roman times, and that in the greater
number of cases these gatherings were under sacerdotal presidency
suggest that the sacerdotes, particularly the pontiffs, introduced the
curiate organization from the army to make their religious meetings
more orderly and dignified.[861]
All assemblies which met under pontifical presidency for religious
purposes were called calata,[862] evidently from “calare,” a verb
which must originally have been in common use in the sense of “to
call,” but which in historical time was restricted to the technical
language of the sacerdotes.[863] In the latter connection it designates
the peculiar method of summoning used by the pontiffs.[864]
Probably, at least in earlier time, their calatores acted as curiate
lictors in convoking the calata comitia curiata,[865] over which they
presided. In all meetings of the kind in the regal period the people
were grouped in curiae; under the republic the centuriata comitia
calata were also used for certain purposes.[866] The usual meeting-
place of the calata comitia curiata was in front of the curia Calabra
on the Capitoline Hill.[867] With reference to their object, they may be
classed as non-voting and voting; the former were purely religious,
the latter were for the settlement of questions which were in part
civil.[868] First to be noted of the non-voting assemblies were those in
which the people gathered in comitia under the presidency of the
king,[869] in the republic under the rex sacrorum, to hear the
proclamation of the fasti. On the calends of each month a pontifex
minor, as clerk of the college,[870] announced to them on what day,
whether the fifth or seventh, the nones would come.[871] On the
nones the king again summoned the people to hear the calendar of
the month,[872] read probably by the same pontifex minor. This
custom fell into disuse with the publication of the calendar in the
Forum, beginning in 304.[873]
Equally passive were those comitia calata which under the
presidency of the supreme pontiff witnessed the inauguration of the
three flamines maiores,[874] probably of the king in the regal period,
and certainly of the rex sacrorum under the republic.[875] As warlike
Mars had his shrines outside the pomerium,[876] his chief temple
being in the Campus Martius,[877] it is a probable conclusion that his
flamen was inaugurated there—in the regal period in some form of
military assembly, under the republic in the comitia centuriata.[878]
The inaugural ceremonies were performed by an augur;[879] in the
case of the sacerdotes it was the supreme pontiff who requested this
service of him,[880] whereas the king could doubtless command an
augur without the coöperation of the pontiff. A closely related
function was the appointment of Vestals by lot, under the conduct of
the supreme pontiff in a public assembly, probably the calata comitia.
[881] The destatio sacrorum and the abjuration of social rank, other
acts which these comitia merely witnessed, will be considered in
connection with the transitio ad plebem and the adrogatio.[882] The
ceremonies attended to by the rex sacrorum on March 24 and again
on May 24 may have been in comitia calata, though this is doubtful.
[883]

Assemblies of the people were organized in curiae by the pontiffs

for the religious purposes mentioned above, while political
measures, so far as submitted to the people, continued for a time,
we may suppose, to be decided by din in contiones. But when a
desire for a more precise vote began to be felt, the curiate
organization naturally offered itself as most convenient for the
purpose. The contention that in primitive Rome, as among other
early peoples,[884] the assembly expressed its feeling or opinion by
noisy demonstration finds strong support in the most probable
derivation of suffragium, “vote,” which connects it with frangere,
fragor, “a breaking,” “crash,” “din,” “applause,”[885] the prefix sub-
expressing the dependence of the action upon the proposal of the
speaker, as in the military succlamare, succlamatio.[886] We may
well believe that even after the organization of the assembly as
comitia—that is, in curiate, centuriate, or tribal divisions[887]—the
voting within the component groups continued for a time to be by din,
as is suggested by the phrase sex suffragia, applied to the six oldest
groups of knights in the comitia centuriata.[888] Voting by heads in
large gatherings is in fact a slow, cumbersome process, the product
of a well-developed political life. In all probability it originated in the
centuriate assembly—in which the military array facilitated the taking
of individual opinion[889]—and afterward extended to the other
comitia. This line of reasoning suggests that when in the regal period
a desire began to be felt for a more precise vote, and the curiate
organization readily offered itself for the purpose, the expedient was
adopted of taking the vote of each curia in order by din and then of
deciding the question at issue by a majority of the thirty curial votes.
[890] There can be little doubt that this step also was first taken by
the pontiffs.
The testamentary calata comitia met twice a year, probably on
fixed days.[891] It has been a disputed question whether the oldest
form of testament here referred to required a vote of the people.
Rubino[892] strongly upheld the negative on the ground (1) of
analogy with the procedure in inaugurations, (2) of analogy with
other forms of testament, none of which required a vote, (3) of the
word testamentum itself, which refers to witnessing, (4) of the
conviction that the patricians would not leave to the popular
assembly the making of private law, (5) on the authority of
Theophilus,[893] who mentions the people’s witnessing of the
testament, (6) on the statement of Gellius[894] that wills of the kind
were made “in populi contione.” Against this reasoning may be urged
(1) the analogy from the adrogatio, (2) the analogy from the
testamentary adoption, to both of which cases the simple testament
was similar, and both of which required a vote of the people,[895] (3)
the consideration that the act of witnessing in the assembly did not
necessarily exclude a vote, (4) the statement of Gaius[896] that
calata comitia were convoked “for making”—not for witnessing—
testaments, (5) the circumstance that the contio was often a
preliminary stage of the voting assembly[897] in addition to the fact
that pontifical language applies the term to comitia in general.[898]
These arguments offset all the points offered by Rubino, unless it be
the fourth, which is a purely subjective consideration. Arguments (1),
(2), and (4) are especially effective for establishing the fact of a vote
in the case under consideration. But the problem can be most
satisfactorily solved (6) by comparative investigation. In the
constitution of the early Indo-European family the estate belonged
jointly to all the male members, and for that reason could not be
given away by the pater.[899] The primitive Germans accordingly
made no wills, but left their property to their children, or in failure of
children to the near kin.[900] In Attica the right to bequeath was
instituted by a law of Solon, which allowed it to those only who had
no legitimate sons;[901] in Sparta the right was introduced by
Epitadeus, perhaps early in the fourth century b.c.[902] Testaments
were unknown in Gortyn at the time when the Twelve Tables of this
city were published,[903] and similar conditions existed in other states
of Greece.[904] The rule holds, too, for ancient India.[905] The Slavic
householder could not alienate his land without the consent of the
community.[906] As there is no reason to assume a more advanced
condition for primitive Rome, we may conclude that, as indicated
above, the calata comitia not only witnessed but ratified testaments.
[907]

Mommsen has attempted to fix these days as March 24 and May

24,[908] on which the rex sacrificulus performed comitial ceremonies
not clearly described by the sources.[909] He admits, however, that
the testamentary comitia met under the pontifex maximus rather than
under the rex sacrorum[910]—a fact directly opposed to his
contention. We should be surprised also to find the testamentary
days so close together.[911] But the most effective argument against
his view is that this function performed by the rex sacrorum could not
have been the holding of comitia, for the time during which it
continued was nefas.[912] The ancient authorities state that “the
sacrificial king, after performing sacred rites, comes into, or makes a
sacrifice in (venit or litat), the comitium,”[913] but they do not mention
an assembly; hence we may infer that in the fasti for these days
reference is to some other function than the holding of comitia. The
form of testament above described fell early into disuse,[914] so that
the conditions and ceremonies attending it became a subject of
study for antiquarians.
Adoptions ordinarily came before the praetor. The legal object was
the perpetuation of the family and its religion. The law granted the
privilege accordingly to those only who had no children and who
were incapable of having children. It required further that the act
should not imperil the continuance of the family from whom the
adopted came.[915] Adrogatio was the adoption of a person who was
his own master and who accordingly consented to pass under the
paternal power of another. The word signifies that the act to which it
applies required a vote of the people.[916] It was not undertaken
rashly or without careful consideration.[917] The persons concerned
were required first to present the case to the college of pontiffs, who
took into account “what reason any one has for adopting children,
what considerations of family or dignity are involved, what principles
of religion are concerned.”[918] The age of the man who wished to
arrogate was considered—whether in this respect he was capable of
having children of his own, and care was taken that the property of
the arrogated person should not be insidiously coveted.[919] The
adrogator was asked whether he wished the candidate for adoption
to be his real son, and the candidate was asked whether he would
allow himself to be placed in this condition;[920] and the testimonies
were confirmed by an oath formulated by Q. Mucius Scaevola.[921]
If the pontiffs gave their consent, the case came before the comitia
curiata under the presidency of the chief of the college,[922] who put
the question in the following form: “Do you wish and order that L.
Valerius be the son of L. Titus by the same legal rights as if born of
the father and mother of that family, and that the latter have the
power of life and death over the former as a father over a son? This
order I request of you, Romans, to grant, just as I have pronounced
the words.”[923] The curiae decided by vote.[924] At the same
meeting the arrogated son was required to declare that he forsook
the religion of the family or gens of his birth—detestatio
sacrorum[925]—and by a similar declaration the adrogator received
him into the sacra of the new family.[926] This form of adoption could
not apply to youths before they had put on the manly gown, or to
wards or women; for children and women had no part in an
assembly, and guardians were not allowed under any circumstances
to place their wards in the power of another.[927]
A modification of adrogatio is testamentary adoption, of which the
only well-known case is that of Octavius, the heir of the dictator
Caesar. Octavius came before a praetor with witnesses and formally
accepted the inheritance;[928] afterward he was declared adopted by
a vote of the curiae.[929] As this case is nearly akin to the adrogatio,
there can be no doubt that the vote was taken in the calata comitia
under pontifical presidency.[930]
Distinct from the adrogatio, though analogous to it, was the direct
passing of individuals and of gentes from the patrician to the
plebeian rank—transitio ad plebem. The motive was a desire to
qualify for the tribunate of the plebs,[931] or more generally to widen
the range of one’s eligibility to office.[932] The history of the republic
affords several instances of the transition of individuals;[933] and two
plebeian gentes, the Octavia[934] and the Minucia,[935] boasted of
having passed over from the patricians. Even if these boasts rest
upon genealogical falsifications,[936] the Romans thought such an
act legally possible; and they formulated a process applicable to
every case whether of individuals or of gentes. It was through some
other ceremony than the adrogatio, for the latter could not apply to
groups of persons. Clodius was following the more general
procedure here referred to when in the year 60 he tried to make
himself a plebeian without recourse to adrogatio. First he abdicated
his nobility by an oath, probably taken in the comitia calata;[937] then
coming before an assembly of the plebs, he held himself ready to
receive plebeian rights through a resolution introduced by the tribune
Herennius.[938] The process allowed the retention of the name,
sacra, and all other privileges not dependent on the patriciate.[939]
But Metellus, the consul, objected that a curiate law was needed to
make the act valid, and the senate evidently agreed with him.[940]
Metellus may have had in mind the transition through the adrogatio,
which required a curiate law, or more probably he was thinking of a
vote of the curiae in addition to the other formalities which Clodius
was passing through.[941] The complete process accordingly would
have been the abjuration of the patriciate, confirmed by a curiate law,
and the reception of plebeian rights through a plebi scitum. Clodius
was not so foolish as to suppose that a process of transitio invented
by himself would prove acceptable to the senate and magistrates,
and must therefore have followed as closely as possible the formula
which he believed to be legal. But when Metellus raised the
objection, and when the tribunes persisted in interceding against the
plebi scitum,[942] he yielded for the present, and in the following year
had himself arrogated by a plebeian named Fonteius, from whom he
was forthwith emancipated.[943] This procedure, too, allowed him to
retain the gentile name of his birth,[944] his imagines and sacra,[945]
and consequently his inheritance. The oath taken in the calata
comitia accordingly was not the detestatio sacrorum usual in
arrogations, but a form of declaration which reserved these
privileges, with the understanding that in this case the arrogatio was
not for the customary object but to enable him to change his rank.
[946]

Analogous to the transitio ad plebem is the elevation of a plebeian

to the patrician rank. The Romans believed that eminent plebeians,
including foreigners of distinction newly admitted to citizenship, were
sometimes granted the patriciate not only through the regal period
but also in the opening years of the republic. For the republican age
they represented the bestowal as a double act, a resolution of the
people followed by coöptation into the senate.[947] In stating that the
first consuls chose the best men from the commons, made them
patricians, and with them filled the senate to the number of three
hundred, Dionysius[948] apparently has in mind the consuls’ function
of recruiting the senate before the Ovinian legislation,[949] together
with their initiative in granting the patriciate. The Roman view that the
bestowal required a vote of the people is further proved by the
procedure of Julius Caesar and of Octavianus in creating new
patricians; for in this function they doubtless followed tradition as
nearly as possible. In 45 a plebi scitum,[950] proposed by L. Cassius
Longinus and supported by a senatus consultum,[951] empowered
Caesar to recruit the patrician rank. Octavianus proceeded in a
similar manner except that a consular law,[952] approved also by a
senatus consultum,[953] was passed for the purpose. As the object
was religious, we may suppose that the qualifications of the
candidates were previously examined by the pontifical college. On
the analogy of the transitio ad plebem it may be assumed further that
the candidate abjured his plebeian rank in the calata comitia, which
then confirmed his declaration by vote.[954]
But whether the Romans were right in supposing patricians to
have been created in the early republic has been doubted.
Mommsen[955] takes the ground that when the curiae ceased to be
exclusively patrician, elevation to the rank became impossible, and
that therefore no cases of the kind occurred after the fall of the kings.
But in such a matter it is absurd to speak of impossibilities;
everything was possible which the governing power approved, and
the argument falls when its basis, the purely patrician state, has
been removed.[956] The cessation was in fact due to the growing
exclusiveness of the patricians, who as they came to supplant the
king in the government, learned to value their privileged position so
highly they were unwilling longer to share it with others. Just when
the closing of their rank was effected has not been ascertained, but
there is no good reason for rejecting the Roman view that for a time
after the fall of the kings plebeians continued to be admitted: in
reality the indications are strong for a relatively late closing.[957]
 We may next inquire how patricians were created in the time of the
kings. As the history of the regal period is in general a reconstruction
with material drawn from later time, so in this particular case ancient
writers sometimes date back to the age of the kings the usage of the
republic. Dionysius[958] accordingly states that “the Romans by vote
transferred Servius Tullius from the plebeian to the patrician order,
just as they had previously transferred Tarquin the Elder and still
earlier Numa Pompilius.” But the Romans preferred to reconstruct
the process on an entirely different principle. Regarding the kings as
the founders of all the fundamental institutions, the patricians looked
upon their superior rank as a gift of these monarchs. The patriciate
depended upon senatorial membership, which was at the disposal of
the kings.[959] This view is well adapted to explain the creation of the
senate; but for the period after its establishment Livy[960] adds to the
adlectio of the king a coöptatio by the patres (senators). Livy’s
account of the usage here given is reasonable; the king indicated his
preference as to the choice of advisers, but a powerful council, such
as the senate must have been, at least in the later regal period,
would have the final decision on the question of admitting a new
member. The conclusion is that toward the end of the monarchy, if
not from the beginning, plebeians were admitted to the senate, and
through it to the patriciate, by the coöperation of the king and the
senate, the people having nothing to do with the matter.[961] But after
the overthrow of the monarchy the vote of the people was
substituted for the will of the king, coöptation by the senate
continuing as before.[962]
The patriciate was acquired not only through bestowal by the
state, but also through the adoption of a plebeian into a patrician
family. Several cases of the kind have been ascertained.[963] The act
took place before the praetor[964] and did not concern the comitia.
Probably a preliminary examination by the pontiffs was necessary to
adoptions as well as to arrogations.[965]

Rubino, J., Röm. Verfassung, 241-53; Mommsen, Röm. Forschungen, i. 123-7,

397-409; Röm. Chronologie, 241 ff.; Röm. Staatsrecht, ii. 33-41; iii. 38-40; Lange,
L., Röm. Altertümer, i. 131-4, 177 f., 356 f., 362, 398-401, 459, 795; ii. 518, see
also indices s. Adrogatio, Calatores, Detestatio sacrorum; Transitio ad plebem, in
Kleine Schriften, ii. 1-90; Madvig, J. N., Verf. u. Verw. d. röm. Staates, i. 222-6;
Herzog, E., Röm. Staatsverfassung, i. 108-11, 1062-4, 1075; Mispoulet, J. B.,
Institutions politiques des Romains, i. 202 f.; Willems, P., Droit public Rom. 53 f.;
Drumann-Gröbe, Gesch. Roms, ii. 187 ff.; Wissowa, G., Religion und Kultus der
Römer, 440 f.; Hallays, A., Comices à Rome, 16-9; Mercklein, D. L., Coöptation
der Römer, 11-44 (of the gentes and of the senate); Helbig, W., in Comptes rendus
de l’acad. des inscr. et belles-lettres, xxi (1893). 350-3; Büdinger, M., Cicero und
die Patriciat, in Denkschr. d. Kaiserl. Akad. d. Wiss. Phil.-hist. Cl. xxxi (1881). 211-
73; Der Patriciat und das Fehderecht in den letzten Jahrzehnten der röm. Rep.,
ibid. xxxvi (1888). 81-125; Baudry, F., Adrogatio, in Daremberg et Saglio, Dict. i. 83
f.; Saglio, E., Calator, ibid. i. 814; Humbert, G., ibid. i. 1375 f.; Detestatio sacrorum,
ibid. ii. 113; Leonhard, Adrogatio, in Pauly-Wissowa, Real-Encycl. i. 419-21;
Samter, Calatores, ibid. iii. 1335 f.; Kübler, Calata comitia, ibid. iii. 1330-4;
Ruggiero, E., Diz. ep. ii. 1185; Smith, Dict. i. 26 f.; Nettleship, Contrib. to Lat.
Lexicog. 400.
 CHAPTER IX.
THE COMITIA CURIATA

The primitive European assembly, of which the Roman is a variety,

may be reconstructed in broad outline by a comparison of the forms
and functions of the institution as found among the earliest Italians,
Greeks, Celts, Germans, Slavs, and kindred peoples, among whom
it differed in detail while possessing the same general features. The
usual tendency of development was toward the abridgment of
popular powers to the advantage of the nobles or of the king;[966] but
in some instances may be discovered a growth in the opposite
direction.
Generally the assembly did not have fixed times of meeting but
convened only when called by the king or chiefs. This is known to be
true of the Homeric Greeks,[967] of the Slavs,[968] and of the
Romans,[969] and may be regarded as the more primitive condition.
In addition to extraordinary sessions the German assembly acquired
the right to meet regularly twice a month at fixed times[970]—a right
which gave the people a valuable political advantage. In like manner
the Lacedaemonians met once a month;[971] the Athenians probably
once a prytany (tenth of a year) after Cleisthenes, and certainly four
times a prytany after Pericles.[972] The Celtic assemblies convened
annually or triennially at fixed seasons.[973] Among all these
peoples, however, subjects for consideration were presented by
none but the king or chief, the assembly itself being wholly without
initiative. Such subjects were as a rule previously discussed in a
council of chiefs or nobles.[974] The person who summoned the
assembly naturally made the first speech, which explained the
purpose of the meeting and the character of the subject to be
considered. If it was an enterprise in which he desired the support or
coöperation of the community, he attempted to rouse for it the
enthusiasm of his hearers.[975] The discussion might then be
continued by the chiefs or any others distinguished for age, military
prowess, or eloquence.[976] Among the Germans, who possessed
more than the average degree of liberty, any one spoke who could
gain a hearing; in the Homeric assembly a commoner who dared lift
up his voice against king or noble was liable to severe chastisement
as a disorderly person;[977] and conditions at Rome, as well as in
Etruria,[978] seem to have been equally unfavorable to the ordinary
freeman.
A considerable variety of business came before the assembly. It
might be summoned to hear the announcement of news of interest to
the community,[979] the reading of the calendar for the month,[980]
the declaration of a policy or opinion by a king or chief,[981] or for
witnessing acts affecting the interests of the community.[982] More
important were judicial cases,[983] questions of war and peace,[984]
and elections.[985]
The problem as to the relative power of the king and council on the
one hand and of the assembly on the other is difficult. It was a
disadvantage to the people, over and above their lack of initiative, to
have no means of precisely expressing their will. The Greeks
signified their approval by acclamation,[986] the Germans by clashing
their weapons,[987] and the Celts by both;[988] either demonstration
aimed to express, not the will of the majority,[989] but the intensity of
conviction on the part of the assembly as a whole. It lacked as well
the means of legally enforcing its will.[990] The Achaeans in
assembly approved the petition of Chryses, a suppliant priest;
nevertheless King Agamemnon rejected it.[991] After the people had
divided the spoils of war, Agamemnon seized the prize they had
given another.[992] The Trojans were ready to surrender Helen for
the sake of peace; but Priam, to gratify his son, refused, and the war
went on.[993] In his relations with individuals the king often acted
unjustly and tyrannically. Even in affairs which concerned the entire
community he might take large liberty. Without consulting the
assembly he could count on the support of the people in a war of
defence. Treaties of peace, which were often guest-friendships and
intermarriages between royal families,[994] did not come before the
people for ratification as a right, but only in cases in which their
pledge seemed necessary for the prevention of private warfare. The
right of the magistrate to conclude peace with or without discussion
in the council or senate was recognized by the states of Italy as late
as the Second Samnite war.[995] The king might even declare an
offensive war on his own responsibility, if without consulting the
people he could feel sure of their support.[996] Enterprises requiring
their coöperation he usually submitted to them to win their approval,
as he had no means of coercing the entire community. His
independence of the assembly increased with the growth of heredity.
The idea of sovereignty, strictly speaking, was unknown to primitive
times; yet so far as people thought of political power, they assigned it
to the king and council.[997] Nevertheless the fact of the assembly’s
existence and the need of eloquence for persuading it prove it to
have been a real force. The suppression of the German assembly or
the prohibition of carrying arms to the meeting was looked upon as
intolerable tyranny.[998] For the disturbance of an Irish assembly the
penalty was death.[999] Public opinion was a check on royalty,[1000]
and in extreme cases the people rebelled and killed their king.[1001]
The strengthening of the kingship naturally tended to weaken the
assembly. The Lacedaemonian kings had a right to make war on
whatever state they pleased, and any citizen who obstructed this
power was accursed;[1002] if, too, in anything the people gave a
wrong decision, the kings and council could set it right.[1003] Under
the Frankish monarchy the general assembly seems to have entirely
disappeared in the sixth century a.d., to be revived in the latter part
of the seventh,[1004] in a form which took little account of the
commons.[1005] In the other Germanic tribes which entered the
Empire the effect of the migration was to strengthen the king and to
weaken in a corresponding degree the power of the people.[1006] In
Russia Tartar domination, converting the legitimate princes into
tyrants, effected the downfall of the assemblies.[1007] The building up