P1: ARS/mbg P2: ARS/vks QC: ARS

November 16, 1998 19:16 Annual Reviews AR075-16

Annu. Rev. Fluid Mech. 1999. 31:567–603

Copyright °c 1999 by Annual Reviews. All rights reserved

DIRECT NUMERICAL SIMULATION

OF FREE-SURFACE AND
INTERFACIAL FLOW
Annu. Rev. Fluid Mech. 1999.31:567-603. Downloaded from www.annualreviews.org
by University of Massachusetts - Amherst on 09/24/12. For personal use only.

Ruben Scardovelli1 and Stéphane Zaleski2

1DIENCA, Lab. di Montecuccolino, Via dei Colli, 16, 40136 Bologna, Italy; and
2Modélisation en Mécanique, UMR CNRS 7607, Université Pierre et Marie
Curie - Paris 6, Paris, France; e-mail: raus@mail.ing.unibo.it; zaleski@lmm.jussieu.fr

KEY WORDS: interfaces, free surfaces, surface tension, multiphase flow

1. INTRODUCTION
The numerical simulation of flows with interfaces and free-surface flows is a
vast topic, with applications to domains as varied as environment, geophysics,
engineering, and fundamental physics.
In engineering, as well as in other disciplines, the study of liquid-gas inter-
faces is important in combustion problems with liquid and gas reagents. The
formation of droplet clouds or sprays that subsequently burn in combustion
chambers originates in interfacial instabilities, such as the Kelvin-Helmholtz
instability. What can numerical simulations do to improve our understanding of
these phenomena? The limitations of numerical techniques make it impossible
to consider more than a few droplets or bubbles. They also force us to stay at
low Reynolds or Weber numbers, which prevent us from finding a direct solu-
tion to the breakup problem. However, these methods are potentially important.
First, the continuous improvement of computational power (or, what amounts
to the same, the drop in megaflop price) continuously extends the range of af-
fordable problems. Second, and more importantly, the phenomena we consider
often happen on scales of space and time where experimental visualization is
difficult or impossible. In such cases, numerical simulation may be a useful
prod to the intuition of the physicist, the engineer, or the mathematician.
A typical example of interfacial flow is the collision between two liquid
droplets. Finding the flow involves the study not only of hydrodynamic fields
in the air and water phases but also of the air-water interface. This latter part
567
0066-4189/99/0115-0567$08.00
P1: ARS/mbg P2: ARS/vks QC: ARS
November 16, 1998 19:16 Annual Reviews AR075-16

568 SCARDOVELLI & ZALESKI

is more difficult than the former because the interface is subject to a number of
relevant physical phenomena, at scales much smaller than the typical sizes of
the droplets. For instance, a major difficulty is caused by the change in interface
topology that occurs when the colliding droplets coalesce. This complicates
the physics and sharpens the requirements that a numerical method must satisfy
in order to resolve the motion in a satisfactory way.
These difficulties are dealt with in a variety of ways, depending on the physi-
cal modeling and the numerical methods. First comes a choice of modeling the
interface as either a thin or a thick region of space. In engineering problems,
such as combustion studies, the interface may be considered a discontinuity
Annu. Rev. Fluid Mech. 1999.31:567-603. Downloaded from www.annualreviews.org

of density and pressure as well as of other physical variables. The location

by University of Massachusetts - Amherst on 09/24/12. For personal use only.

of this discontinuity in three-dimensional space will be assumed to be a two-

dimensional, twice-differentiable manifold (to simplify, a smooth surface S )
that evolves according to predetermined rules.
The second choice is to decide which forces and thermal effects to model.
The main thrust of this review is as follows. In the bulk of the fluids, we
restrict ourselves to viscosity, constant in each phase, and perhaps gravity. On
interfaces, we assume a constant surface tension. Our physical assumptions
thus lead us to a version of the Navier-Stokes equation, with variable density
and viscosity and capillary tension on S. This should be contrasted with both
a more elementary approach, in which only perfect fluids are considered, and
many more complex approaches, where more realistic theories of interfaces
or other transport effects (involving temperature, passive scalars, and surface-
active substances) are considered.
Once this framework is set, we attack the subject matter of this review:
simulation methods for the Navier-Stokes equation with interfaces. On one
hand the numerical methods are analogous to otherwise well-known methods
in the bulk of the phases, such as finite elements, finite volumes, and finite
differences. On the other hand, there are specific problems due to the pre-
sence of the interface: location of the discontinuity and computation of surface
stresses.
From this point of view the various methods for interface simulation can be
divided into two great classes, depending on the nature, fixed or moving, of the
grid used in the bulk of the phases. In fixed-grid methods, there is a predefined
grid that does not move with the interface. The interface has to somehow cut
across this fixed grid. As shown in Figures 1 and 2, this fixed grid may be either
structured or unstructured. In moving-grid methods, the interface is a boundary
between two subdomains of the grid. The interface then identifies, at some order
of approximation, with element boundaries. Again, the grid may be structured
and even near-orthogonal or more general. This simplifies the analysis near
the interface. However, when the interface undergoes large deformations, the
grid has to be remeshed. Other complications may occur, especially when the
P1: ARS/mbg P2: ARS/vks QC: ARS
November 16, 1998 19:16 Annual Reviews AR075-16

DIRECT SIMULATION OF INTERFACIAL FLOW 569

Annu. Rev. Fluid Mech. 1999.31:567-603. Downloaded from www.annualreviews.org
by University of Massachusetts - Amherst on 09/24/12. For personal use only.

Figure 1 An interface cutting across a fixed, structured grid.

topology changes. In addition to the moving-grid and fixed-grid methods, there

are special cases, such as particle methods, in which grids are not needed.
In this article we emphasize fixed-grid methods. These are more interesting
because of their relatively simple description and greater ease of program-
ming. Easier extension to three dimensions of space is one of the principal

Figure 2 An interface cutting across a fixed, unstructured grid.

P1: ARS/mbg P2: ARS/vks QC: ARS
November 16, 1998 19:16 Annual Reviews AR075-16

570 SCARDOVELLI & ZALESKI

advantages expected from this simplified formulation. Moreover, much of

this paper is concerned with so-called volume-of-fluid (VOF) methods. Many
commercially available codes use this method to represent interfaces: SOLA-
VOF (Nichols et al 1980), NASA-VOF2D (Torrey et al 1985), NASA-VOF3D
(Torrey et al 1987), RIPPLE (Kothe & Mjolsness 1992), and FLOW3D (Hirt
& Nichols 1988).
This review remains at an introductory level. The reader may read with profit
other reviews and books (i.e. Floryan & Rasmussen 1989, Shyy et al 1996,
Sethian 1996, Rothman & Zaleski 1997). The review by McHyman (1984) is
particularly clear, pleasant to read, and refreshing, as well as informative to the
Annu. Rev. Fluid Mech. 1999.31:567-603. Downloaded from www.annualreviews.org

novice practitioner.
by University of Massachusetts - Amherst on 09/24/12. For personal use only.

A great deal of useful information may be found on a variety of World Wide

Web sites. It is difficult to provide a pointer to the rapidly evolving list of
these sites. However, useful information and links may be found on our site
(http://www.lmm.jussieu.fr/∼zaleski/zaleski.html).
This review reflects our personal point of view: Some aspects of the subject
have been emphasized; others have been omitted. For instance, interface simu-
lation using boundary integral methods is not described in detail; only relevant
references are given below. Similarly, lattice-gas and Boltzmann lattice-gas
methods are described in other reviews and books (i.e. Rothman & Zaleski
1994, 1997; Benzi et al 1992). The same holds for level-set methods (Sussman
et al 1994, Sethian 1996, Sussman & Smereka 1997).

2. PHYSICAL MODELING OF INTERFACES

The study of interfaces starts when equilibrium thermodynamics takes into
account spatial inhomogeneities. Consider two pure phases such as water and
oil. Oil and water do not mix: The equilibrium state of the mixture in a flask
consists of two layers of liquid separated by a thin region called the interface.

2.1 Interface Description

There are many ways in which the geometrical interface S may be defined. The
simplest one is to define a height function h. We then get the equation z =
h(x, y, t). This describes the interface satisfactorily whenever h is single-
valued. Alternatively, one may define the phase-characteristic function χ with
χ = 1 in phase 1 and χ = 0 in phase 2. For instance, when h is single-valued,
we may take χ(x, y, z, t) = H [z − h(x, y, t)], where H is the Heaviside step
function. Then phase 1 fills the z > h region and phase 2 the z < h region.
A short discussion of several equations for interface motion may be found in
Whitham (1974). Here we simply remark that evolution of the interface S is
defined by its normal velocity V on each point of S. We also note n, the unit
normal to the interface.
P1: ARS/mbg P2: ARS/vks QC: ARS
November 16, 1998 19:16 Annual Reviews AR075-16

DIRECT SIMULATION OF INTERFACIAL FLOW 571

The normal velocity is related to the fluid velocity field u in various ways.
In the simplest case, without phase change, one assumes continuity of fluid
velocity
u1 = u2 . (1)
(We note in general xi , the limiting values of a variable x when the interface
is approached from phase i.) This may be written in jump notation, i.e. the
notation [x] = x2 − x1 :
[u] S = 0. (2)
Annu. Rev. Fluid Mech. 1999.31:567-603. Downloaded from www.annualreviews.org

The interface velocity is then the normal velocity, the same on both sides of the
by University of Massachusetts - Amherst on 09/24/12. For personal use only.

interface:
V = u1 · n = u2 · n. (3)
In the case of evaporation, or more generally phase change, there may be a mass
flow q from phase 1 to phase 2. This mass flow is connected to the velocities on
both sides of the interface in the following way. Consider the frame of reference
where the interface is at rest. The normal velocities in that frame of reference
are u 0 = u · n − V . The mass flow is the amount of mass that goes from phase 1
to phase 2, and it must be the same on both sides of the interface by conservation
of mass (the Rankine-Hugoniot condition):
ρ1 u 01 = ρ2 u 02 = q. (4)
Then, in the general frame of reference,
ρ1 (u1 · n − V ) = ρ2 (u2 · n − V ) = q. (5)
Thus, if there is no phase change, i.e. q = 0, one recovers Equation 3, which
now appears clearly to be a consequence of mass conservation. The continuity
of tangential velocities [ut = u − (u · n)n], on the other hand, is a physical
assumption akin to the assumption that the slip velocity on a solid wall vanishes.
Then interface motion may be described in weak form by
∂t χ + V n · ∇χ = 0, (6)
which in the no-phase-change case amounts to
∂t χ + u · ∇χ = 0. (7)
(Notice that ∇χ = nδ S .) Here we use a weak formulation of the partial dif-
ferential equation (PDE), since derivatives of the discontinuous function χ
are singular. In this weak formulation, the PDE is interpreted by the space
integrals of Equation 7. In other words, the function cannot be differentiated,
but integrals of Equation 7 are well-defined: They express volume evolution
and correspond to interface motion with velocity V.
P1: ARS/mbg P2: ARS/vks QC: ARS
November 16, 1998 19:16 Annual Reviews AR075-16

572 SCARDOVELLI & ZALESKI

2.2 Navier-Stokes Equation: Whole-Domain Formulation

We now consider the incompressible flow of two immiscible liquids. The whole
flow fills a domain Ä. This domain may be decomposed into any number of
subdomains filled with the individual phases. However, in this paragraph we
express the Navier-Stokes equation over the whole domain Ä. Incompressibility
implies the divergence free condition

∇ · u = 0. (8)

The momentum balance is modeled by the Navier-Stokes equation,

Annu. Rev. Fluid Mech. 1999.31:567-603. Downloaded from www.annualreviews.org

ρ(∂t u + u · ∇u) = −∇ p + 2∇ · µD + 2σ κδ S n,
by University of Massachusetts - Amherst on 09/24/12. For personal use only.

(9)

where ρ is the density and µ the dynamical viscosity. These are two phys-
ical properties of the bulk phases that may have discontinuities across phase
boundaries. The rate-of-strain tensor is D:
µ ¶
1 ∂u j ∂u i
Di j = + . (10)
2 ∂ xi ∂x j
In these equations, the term 2σ κδ S n represents capillary forces, with a surface
tension σ and n is again the unit normal to the interface. The mean curvature
is κ = (1/r1 + 1/r2 )/2. It may also be expressed as
1
κ = − ∇ S · n, (11)
2
where ∇ S is the gradient operator restricted to the surface S.

2.3 Whole-Domain Conservation Law Form

We may also rewrite the Navier-Stokes equation as a set of conservation laws.
Conservation of momentum is then obvious. In particular, capillary effects may
be represented by a tensor T. This tensor is tangent to the interface and is given
by

T = −σ (1 − n ⊗ n)δ S , (12)

where 1 is the kronecker symbol tensor δi j .

One may show that the capillary force may be written (Lafaurie et al 1994)

2σ κnδ S = −∇ · T, (13)

with the equivalence of Equation 9 in conservative form

∂t (ρu) = −∇ · ( p1 + ρu ⊗ u + T − 2µD). (14)

P1: ARS/mbg P2: ARS/vks QC: ARS
November 16, 1998 19:16 Annual Reviews AR075-16

DIRECT SIMULATION OF INTERFACIAL FLOW 573

2.4 Navier-Stokes Equation: Jump-Condition Form

Another useful formulation of the momentum-balance equation decomposes
the problem into any number of bulk-phase domains within which the usual,
nonsingular Navier-Stokes equation holds; for the interface between domains,
some quantities are required to be continuous, whereas others are required to
have specific jumps.
The jumps result from a balance between the singular terms in the whole-
domain formulation. Balancing the most singular terms in Equations 9 or 14
leads to
Annu. Rev. Fluid Mech. 1999.31:567-603. Downloaded from www.annualreviews.org

[− p1 + 2µD] S · n = 2σ κn. (15)

by University of Massachusetts - Amherst on 09/24/12. For personal use only.

This condition may be split into a normal and tangential stress condition
[− p + 2µn · D · n] S = 2σ κ, (16)

[µt(k) · D · n] S = 0, (17)

where the vectors tk may be any set of d − 1 independent tangent (to S) vectors,
and d is the dimension of space.
Away from the interface, Equation 9 takes the usual form
∂t u + u · ∇u = −∇( p/ρ) + ν∇ 2 u, (18)
with ν = µ/ρ. All the above formulations are equivalent.
2.5 Problems with Reconnecting Interfaces
When two interfaces reconnect and change topology, as in Figures 3 and 4, the
above macroscopic model becomes even more singular. Indeed, immediately
after reconnection the surface is no longer a smooth surface but has cusp-like
singularities.
The curvature κ is then ill-defined and arbitrarily large in the instants imme-
diately following reconnection. Another difficulty arises in the case represented
in Figure 4. In that case it is reasonable to assume that the velocity field is hy-
perbolic near the reconnection point. The interfaces then approach each other
exponentially, and their separation d(t) varies as d0 exp(−λt), where λ is the
largest eigenvalue of the rate-of-strain tensor D of the local velocity field. It is
obvious that the separation d(t) rapidly reaches scales much smaller than the
scales on which microscopic molecular forces act to attract or repel interfaces
(de Gennes 1985). The exact nature of these forces depends on the fluids and
on the presence of surfactants. As a rule of thumb, they must be considered
prevalent on scales of 10–40 nm.
This difficulty is less important, from the numerical point of view at least,
in the case of the filamentary reconnection (Figure 3). Indeed, in that case the
P1: ARS/mbg P2: ARS/vks QC: ARS
November 16, 1998 19:16 Annual Reviews AR075-16

574 SCARDOVELLI & ZALESKI

Annu. Rev. Fluid Mech. 1999.31:567-603. Downloaded from www.annualreviews.org
by University of Massachusetts - Amherst on 09/24/12. For personal use only.

Figure 3 Interface reconnection through filament breakup. This kind of reconnection is accel-
erated by capillary tension. The Navier-Stokes equation predicts it will happen in finite time.

macroscopic impact of microscopic physics may be limited. In other words,

the macroscopic interface motion may be relatively less dependent on interface
physics. This is because universal macroscopic solutions that lead to a singu-
larity in finite time may be found (Eggers 1997). The nature of this universal
singularity does not depend on the large-scale flow around the singularity or
on microscopic properties (Eggers 1993, Stone 1994). One may then hope to
solve the problem on a sufficiently fine scale to capture the singularity, just as
one captures a shock in computational aerodynamics.
2.6 Free-Surface Flow
Free-surface flow is a limiting case of flow with interfaces, in which the treat-
ment of one of the phases is simplified. For instance, for some cases of air-water
flow, we may consider (a) the pressure p in the air to depend only on time and
not on space [through, say, some function pair (t)] and (b) the viscous stresses
in the air to be negligible. The whole-domain forms, Equations 9 or 14, are
no longer available and Form 18 must be used. The jump conditions become
boundary conditions on the border of the liquid domain:

(− p + 2µn · D · n)| S = pair + 2σ κ (19)

P1: ARS/mbg P2: ARS/vks QC: ARS
November 16, 1998 19:16 Annual Reviews AR075-16

DIRECT SIMULATION OF INTERFACIAL FLOW 575

Annu. Rev. Fluid Mech. 1999.31:567-603. Downloaded from www.annualreviews.org
by University of Massachusetts - Amherst on 09/24/12. For personal use only.

Figure 4 Interface reconnection through sheet breakup. This reconnection may occur in a hyper-
bolic region of the flow, in two-dimensional as well as in three-dimensional flow. The computation
of the final phase of reconnection requires the estimation of microscopic forces.

and
t(k) · D · n| S = 0. (20)
It may be noted that the last condition, which results from the nullity of tangen-
tial stresses, is purely kinematic: It does not involve material properties at all.
For this reason it has interesting consequences for the value of the vorticity on
a free surface (Batchelor 1967, Longuet-Higgins 1998).

3. SPECIAL CASES: PARTICULATE

AND GRID-FREE METHODS
Arguably, the most important special case is inviscid flow. The Navier-Stokes
equation degenerates into the Euler equation. A boundary integral formulation
P1: ARS/mbg P2: ARS/vks QC: ARS
November 16, 1998 19:16 Annual Reviews AR075-16

576 SCARDOVELLI & ZALESKI

then exists, in which the interface is both a free boundary between phases and
a vortex sheet (Kane 1994). The boundary integral may at first view be solved
more simply than the full Navier-Stokes or Euler equations. In a cubic domain
of size L and with a spatial grid size h, one needs N = L/ h points along
each dimension. The discretization of the full two-dimensional problem would
require N 2 grid points. If the interface is also of length |S| ' O(L), then the
boundary integral may be discretized with O(N ) points, which is much less than
what is needed for the full Euler equation. However, if there are many droplets
or bubbles, or if the interface is very convoluted, then |S| À L. However, for
many problems the boundary-integral method offers a substantial savings of
Annu. Rev. Fluid Mech. 1999.31:567-603. Downloaded from www.annualreviews.org

computations and memory requirements.

by University of Massachusetts - Amherst on 09/24/12. For personal use only.

The formulation of the Euler equation as a boundary-integral problem is

somewhat tricky, especially when there is a density jump. The equations of
vortex sheet evolution were derived in part by Zalosh (1976), with corrections
indicated by Rottman & Olfe (1977). Many other formulations and numerical
methods have been proposed (Baker et al 1982, 1984; Og̃uz & Prosperetti 1990).
Various applications to bubble and droplet flow may be found in the literature
(Lundgren & Mansour 1988, 1991; Og̃uz & Prosperetti 1993; Boulton-Stone
& Blake 1993).
Creeping flow is another interesting special case. There is also a boundary
integral formulation of the problem (Stone 1994, Tsai & Miksis 1994), which
is predominantly used for creeping flows with interfaces. When the Reynolds
number vanishes, lattice-gas (Rothman & Zaleski 1994, 1997) and Boltzmann
lattice-gas (Benzi et al 1992) methods are also a possibility. It was argued
(Ladd 1994) that Boltzmann lattice-gas methods (and perhaps simple finite
differences) may be more efficient than boundary-integral methods when the
number of particles is large. Recent applications of lattice-gas methods to creep-
ing multiphase flows may be found (Olson & Rothman 1997).
The lattice-gas and Boltzmann lattice-gas methods, despite their refreshing
originality, are difficult to generalize beyond their domain of validity. Whenever
either density or viscosity jump across the interface, the jump conditions of
Equations 1 and 17 are not satisfied anymore.
One way to eliminate the problems posed by fixed or moving grids is to elim-
inate the grid, in part or completely. Partial elimination of the grid is achieved
in the famous particle in cell (PIC) method (Harlow 1964). This method has
evolved in a number of ways. Some versions retain coupling between particles
and a mesh and others do not.
The domain is then traversed by a number of particles, with or without direct
physical meaning. The smoothed particle hydrodynamics method (Monaghan
1992) is a general framework for solving differential equations using particles
and smoothing kernels that define the intensity of interaction between particles,
P1: ARS/mbg P2: ARS/vks QC: ARS
November 16, 1998 19:16 Annual Reviews AR075-16

DIRECT SIMULATION OF INTERFACIAL FLOW 577

depending on their mutual distance. The interaction between particles may be

viewed either as a device to solve the differential equation or as an actual force
between particles. When particles of different kinds (oil and water, to use a
standard example) repel each other, phases may separate and an interface with
surface tension may be established. By simply letting particles attract each other,
Monaghan and coworkers were able to create a free surface (Monaghan et al
1994, Monaghan 1994), thus creating a realistic simulation of wave breaking.
Phase separation may also be attained (Okuzono 1997).
An interesting review of developments up to 1989 may be found in Floryan
& Rasmussen (1989). The statement in that reference, that “[t]he SPH method
Annu. Rev. Fluid Mech. 1999.31:567-603. Downloaded from www.annualreviews.org

is relatively recent; it has been developed in the context of astrophysical hydro-

by University of Massachusetts - Amherst on 09/24/12. For personal use only.

dynamics ... and seems to be hardly known in other areas of fluid dynamics” is,
however, still quite accurate.

4. MOVING-GRID AND ADAPTIVE-GRID METHODS

Moving grid methods have been particularly successful for the study of the
motion of small amplitude waves and weakly deformed bubbles. Curvilinear
grids have been used for some time now to follow the motion of a rising bubble
(Ryskin & Leal 1984a,b). The latter was a free-surface calculation and thus
the fluid dynamics of the gas was not treated. A full calculation, with gas
effects included, was performed by Dandy & Leal (1989). More recently, simi-
lar methods were used by Magnaudet et al (1995) and Cuenot et al (1997) with
a quasiconformal mapping technique developed by Duraiswami & Prosperetti
(1992). An example of an orthogonal grid is given in Figure 5.
There are many methods that use nonorthogonal grids. When the grid points
move simply in a Lagrangian way, the grid may deform considerably (see
Figure 6). It is more sensible in most cases to move the point in a mixed manner
between the Lagrangian motion and the fixed Eulerian point of view (Hirt et al
1974). Frequent regridding or rezoning may, however, be necessary in both
cases, with in some cases the additional complication of having to add and/or
remove a few grid points. This is performed, for instance, in the free Lagrangian
method (Fritts et al 1985, Fyfe et al 1988).
Recently, interesting applications of deforming finite elements to large inter-
face deformation were made (Fukai et al 1993, 1995). Representations of the
deforming grids may be found in Waldvogel et al (1996).

5. FIXED-GRID METHODS
In fixed-grid methods the grid used to solve the Navier-Stokes equation is
entirely or quasi-entirely fixed. The grid may be quasi-entirely fixed when, for
P1: ARS/mbg P2: ARS/vks QC: ARS
November 16, 1998 19:16 Annual Reviews AR075-16

578 SCARDOVELLI & ZALESKI

Annu. Rev. Fluid Mech. 1999.31:567-603. Downloaded from www.annualreviews.org
by University of Massachusetts - Amherst on 09/24/12. For personal use only.

Figure 5 An example of an orthogonal grid, used to compute oscillations of a bubble in a liquid.

The bubble is axisymmetric. The bubble axis lies at the bottom of the figure. Only a fraction of
the grid is shown. [Figure courtesy of J Magnaudet and coworkers.]

example, it is adaptive just for those cells cut by the interface, as in Figure 7
(Glimm et al 1986, 1987).
5.1 Marker Methods
In marker methods, tracers or marker particles are used in the algorithm to
locate the phases. Interfacial or surface-marker methods use marker particles
only on the interfaces. Volume-marker methods have marker particles in the
whole domain.
For two-phase flow, surface markers are more accurate than volume markers
because they track exactly the location of the interface. However, when there
are more than two phases, it may become difficult to handle the complexity of
triple lines (lines in three-dimensional space where three phases meet) and other
effects associated with the presence of several phases. Volume markers afford
a simple way of dealing with the problem. Volume markers become distorted
as time goes by, just as does the Lagrangian grid of Figure 6.
P1: ARS/mbg P2: ARS/vks QC: ARS
November 16, 1998 19:16 Annual Reviews AR075-16

DIRECT SIMULATION OF INTERFACIAL FLOW 579

Annu. Rev. Fluid Mech. 1999.31:567-603. Downloaded from www.annualreviews.org
by University of Massachusetts - Amherst on 09/24/12. For personal use only.

Figure 6 In a grid that follows the fluid motion, a considerable distortion may appear even if
the interface has undergone relatively mild deformation. [Redrawn after Figure 5 of McHyman
(1984).]

An illustration of the surface-marker method, in two particular cases, is

shown in Figures 8 and 9. One advantage of the use of surface markers is that
is allows formation of very thin liquid bridges that do no break, as shown in
Figure 8. However, this is a real gain only in some cases. The situation may be
made clear by considering the spiraling wave of Figure 9. If both phases have

Figure 7 An example of local adaptation on an otherwise fixed grid. (Left) The undeformed finite
element triangular grid; (Right) the thick line representing the interface is made up of sides of the
new triangles.
P1: ARS/mbg P2: ARS/vks QC: ARS
November 16, 1998 19:16 Annual Reviews AR075-16

580 SCARDOVELLI & ZALESKI

Annu. Rev. Fluid Mech. 1999.31:567-603. Downloaded from www.annualreviews.org
by University of Massachusetts - Amherst on 09/24/12. For personal use only.

Figure 8 An example of the use of surface markers on an underlying square grid. The use of
surface markers allows maintenance of the thin layer shown in the figure. In other types of fixed-grid
methods, this layer may break.

Figure 9 Another example of the use of surface markers: A vortical structure entrains the interface
in a spiraling motion. Surface markers allow the capture of details of interface motion on scales
much smaller than the grid spacing. However, as explained in the text, some small-scale detail of
the hydrodynamics may be lost when scales much smaller than the grid are created.
P1: ARS/mbg P2: ARS/vks QC: ARS
November 16, 1998 19:16 Annual Reviews AR075-16

DIRECT SIMULATION OF INTERFACIAL FLOW 581

the same viscosity and density and there is no surface tension, the interface is
“transparent” to the fluid: The fluid, in a way, does not know that there is an
interface. The phases are only distinguished by their “color” as an ideal passive
scalar with zero diffusivity. In that case it makes sense to track details smaller
than the grid.
On the other hand, if there is, for instance, a difference in density, there
will be scales in the velocity and pressure fields much smaller than the grid,
and these scales will not be well resolved using the uniform square grid of
Figure 9.
Surface-marker methods have been used extensively by the Glimm group
Annu. Rev. Fluid Mech. 1999.31:567-603. Downloaded from www.annualreviews.org

(Glimm et al 1986, 1987) and by the Tryggvason group (Tryggvason & Unverdi
by University of Massachusetts - Amherst on 09/24/12. For personal use only.

1990; Unverdi & Tryggvason 1992a,b). Marker methods have other advantages,
and an important one is the high degree of accuracy that may be achieved by
representing the interface through high-order interpolation polynomials. This
accuracy may improve the accuracy of surface tension calculations (Popinet &
Zaleski 1998a,b).
5.2 Volume of Fluid Methods
As stated in the introduction, we consider in some detail VOF methods relative
to two-dimensional incompressible flows, without mass transfer. Extension to
compressible fluids can be found (Norman & Winkler 1986, Miller & Puckett
1996, Puckett & Saltzman 1992, Saurel & Abgrall 1998). Fluxes are defined
on cell faces of a square mesh with constant grid spacing h = 1x = 1y. The
generalization to three-dimensional domains and to nonuniform rectangular
and parallelepipedic grids can be done with some extra conceptual effort, but
it would unnecessarily complicate this presentation (Popinet et al 1997).
A discrete analog of the characteristic function χ used in Equation 7 is the
scalar field Ci j , known as volume fraction or color function (it actually should
be area fraction in two dimensions, but the standard appellation is volume
fraction). The color function Ci j represents the portion of the area of the cell
(i, j) filled with phase 1:
ZZ
Ci j h ≈
2
χ (x, y) d x d y.
(i, j)

We have 0 < C < 1 in cells cut by the interface S and C = 0 or 1 away from it.
An example of a color function corresponding to a circle arc is shown in
Figure 10. In an incompressible flow, mass conservation is equivalent to con-
servation of volume and hence of the characteristic function χ . There is a vast
literature of numerical methods for conservation laws. However, an explicit
account needs to be taken of the special nature of the problem, which is entirely
concentrated on the interface S. Moreover, particular care has to be given to the
P1: ARS/mbg P2: ARS/vks QC: ARS
November 16, 1998 19:16 Annual Reviews AR075-16

582 SCARDOVELLI & ZALESKI

Annu. Rev. Fluid Mech. 1999.31:567-603. Downloaded from www.annualreviews.org
by University of Massachusetts - Amherst on 09/24/12. For personal use only.

Figure 10 The exact volume-of-fluid color function for a smooth circular arc over a square grid.

constraint 0 < C < 1, because numerical errors in the estimation of the fluxes
can lead to values of C outside the physical range of validity.
VOF methods have been known for several decades and have gone through a
continuous process of improvement. Their use and effectiveness are widespread,
for several reasons:

1. They preserve mass in a natural way, as a direct consequence of the devel-

opment of an advection algorithm based on a discrete representation of the
conservation law (7).

2. No special provision is necessary to perform reconnection or breakup of the

interface and in this sense the change of topology is implicit in the algorithm.

3. They can be relatively simply extended from two-dimensional to three-

dimensional domains.

4. The scheme is local in the sense that only the C values of the neighboring
cells are needed to update the C value in the cell (i, j). For this reason, it
is relatively simple to implement these algorithms in parallel, in particular
within the framework of domain decomposition techniques.

In general, a VOF algorithm solves the problem of updating the volume

fraction field C given the fixed grid, the velocity field u, and the field C at the
previous step. In two dimensions, the interface is considered to be a continuous,
piecewise smooth line; the problem of its reconstruction is that of finding an
approximation to the section of the interface in each cut cell, by knowing only
P1: ARS/mbg P2: ARS/vks QC: ARS
November 16, 1998 19:16 Annual Reviews AR075-16

DIRECT SIMULATION OF INTERFACIAL FLOW 583

the volume fraction C in that cell and in the neighboring ones. The simplest
types of VOF-methods are the simple line interface calculation (SLIC) of Noh
& Woodward (1976) or the SOLA-VOF algorithm of Hirt & Nichols (1981).
These are first order, O(h), in the accuracy of the reconstruction of the interface.
Typically, the reconstructed interface is made up of a sequence of segments
aligned with the grid, as shown in Figure 11a. The reconstruction is relatively
crude, and its advection, even with simple velocity fields such as translations or
solid body rotations, generates a large amount of flotsam. A few improvements
have been made by several authors (Chorin 1980, Lafaurie et al 1994), but these
features still remain.
Annu. Rev. Fluid Mech. 1999.31:567-603. Downloaded from www.annualreviews.org

More accurate VOF techniques attempt to fit the interface through piecewise
by University of Massachusetts - Amherst on 09/24/12. For personal use only.

linear segments. This was already implemented in some of the earliest VOF
methods (DeBar 1974). The more accurate techniques are now known as the
piecewise linear interface construction (PLIC) method (Ashgriz & Poo 1991,
Li 1995, Parker & Youngs 1992, Rider & Kothe 1995, Pilliod & Puckett 1997,
Rudman 1997, 1998, Rider & Kothe 1998, Gueyffier et al 1998).
To understand the reconstruction problem we have represented in each cell
of Figure 12a possible distribution of the color function on the square grid.
In Figure 12b, it is harder to determine the location of the interface, but a
robust reconstruction method should be able to handle this case too without any
ambiguity.
One of the critical simplifying features of VOF/PLIC algorithms is that one
does not attempt to reconstruct the interface as a chain of joined segments (a
continuous chain of segments) but rather as a discontinuous chain with however
asymptotically small discontinuities.
Whenever the curvature is small (i.e. the radius of curvature is large with
respect to the grid size) the method will be accurate. However, it is robust
in the sense that it does not have a catastrophic behavior when the curvature
increases or even becomes infinite, for instance at reconnection. Specifically,

Figure 11 Two attempts at interface reconstruction. (a) First-order or simple line interface cal-
culation; (b) second-order or piecewise linear interface construction.
P1: ARS/mbg P2: ARS/vks QC: ARS
November 16, 1998 19:16 Annual Reviews AR075-16

584 SCARDOVELLI & ZALESKI

Annu. Rev. Fluid Mech. 1999.31:567-603. Downloaded from www.annualreviews.org

Figure 12 Representing the color function as shaded squares of size proportional to the fractional
by University of Massachusetts - Amherst on 09/24/12. For personal use only.

volumes may help to explain the reconstruction problem. (a) A well-behaved distribution of the
color function, corresponding to a smooth circular arc. (b) A more confused distribution.

when the curvature is small, the jumps in interface position at discontinuities

are of order ||[x]|| ' O(κh 2 ). When the curvature is large, we lose all details
at scales smaller than the grid size h.
A VOF/PLIC algorithm is divided into two parts: a reconstruction step and
a propagation step. (This splitting also occurs in the numerical resolution of
conservation laws.) We now describe these two components.
5.2.1 RECONSTRUCTION The key part of the reconstruction step is the deter-
mination of the orientation of the segment. This is equivalent to the determi-
nation of the unit normal vector n to the segment. Then, the normal vector n
and the volume fraction C uniquely determine the straight line.
A typical reconstruction is shown in Figure 11b. Several algorithms have
been developed for the calculation of the normal vector (Puckett & Saltzman
1992, Parker & Youngs 1992). Here, we briefly review two of them. In the first
one (Li 1995), a normal (nonunit) vector m is estimated by a finite-difference
formula:
mh = ∇ h C. (21)
This vector m is cell-centered and it is approximated by first evaluating the
cell-corner values of m, for example at position (i + 1/2, j + 1/2):
1
m x,i+1/2, j+1/2 = (Ci+1, j − Ci, j + Ci+1, j+1 − Ci, j+1 ), (22)
2h
1
m y,i+1/2, j+1/2 = (Ci, j+1 − Ci, j + Ci+1, j+1 − Ci+1, j ). (23)
2h
The required cell-centered value is obtained by averaging:
1
mi j = (mi+1/2, j−1/2 + mi−1/2, j−1/2 + mi+1/2, j+1/2 + mi−1/2, j+1/2 ). (24)
4
P1: ARS/mbg P2: ARS/vks QC: ARS
November 16, 1998 19:16 Annual Reviews AR075-16

DIRECT SIMULATION OF INTERFACIAL FLOW 585

This estimation of the normal is sometimes called the Parker and Youngs ap-
proach (PY). In two dimensions, the discrete gradient ∇ h C is constructed from
the volume fraction values of a 3 × 3 block of cells centered at (i, j).
An alternate approach is a least-square method (Puckett 1991). The same
block of cells is considered, but now the interface is approximated by a straight
line in the whole block, with the constraint that the volume fraction of the
central cell is always the true value C. By changing the slope of the line, one
can minimize the discrete error between the true value C and the value C̃ given
by the linear approximation. For example, the discrete error E in L 2 norm at
cell (i, j) is given by the expression
Annu. Rev. Fluid Mech. 1999.31:567-603. Downloaded from www.annualreviews.org

à 1 ! 12
by University of Massachusetts - Amherst on 09/24/12. For personal use only.

X
E(m) = (C̃ i+k, j+l (m) − Ci+k, j+l )2
. (25)
k,l=−1

More recently, the ELVIRA approach was proposed (Pilliod & Puckett 1997).
In ELVIRA the slope is chosen among six “candidates” given by the backward,
central, and forward differences of the column sums of the volume fractions
done along the x and the y directions, respectively. The candidate that yields
the lowest value of the error in Equation 25 wins.
The ELVIRA approach reconstructs exactly (with machine precision) a
straight line interface. This method is thus truly second order, O(h 2 ), unlike PY,
which is intermediate between first and second order (Pilliod & Puckett 1997).
Numerical experiments show, however, that at low-to-medium resolution it does
not always give better results than PY, but as the resolution increases, ELVIRA
gives the best results, as should be expected. This typically occurs when there
are more than 20 grid points along the bubble diameter.
In the second part of the reconstruction, a linear interface that divides the
computational cell into two parts containing the proper area of each fluid must
be found. In general, the “forward” problem of finding the area within a square
on each side of a given linear interface is more straightforward than the “inverse”
problem of obtaining the equation for the linear interface, given the fraction
of area contained on each side and the normal direction. Both are needed in
the reconstruction and propagation steps of PLIC. We achieve this by deriving
an explicit expression that relates the “cut” area to a parameter α, which fully
defines the straight line (Gueyffier et al 1998).
In two dimensions, the problem can be stated as follows. Given a square cell
of side h in the (x, y) plane and a straight line (such as EH in Figure 13) with
normal vector m, find the area of the region below the line that also lies within
the square cell. This corresponds to the area ABFGD in Figure 13. To obtain an
expression for this area, let us suppose that the components m x and m y of the
normal m are both positive (this can always be arranged by a simple coordinate
P1: ARS/mbg P2: ARS/vks QC: ARS
November 16, 1998 19:16 Annual Reviews AR075-16

586 SCARDOVELLI & ZALESKI

Annu. Rev. Fluid Mech. 1999.31:567-603. Downloaded from www.annualreviews.org
by University of Massachusetts - Amherst on 09/24/12. For personal use only.

Figure 13 Geometrical basis for Expression 27.

transformation); in case one of the components vanishes, the calculation of the

area becomes trivial. The most general equation for a straight line in the (x, y)
plane with normal m is
m x x + m y y = α, (26)
where α is a parameter. The area of the region contained below this line within
the square cell (i, j) (ABCD of Figure 13) is given by
· µ ¶
α2 α − mx h 2
Ci j = 1 − H (α − m x h)
2m x m y α
µ ¶ ¸
α − myh 2
− H (α − m y h) . (27)
α
The prefactor α 2 /2m x m y on the right-hand side of this equation is simply the
area of the triangle AEH. In case points E and H lie within the original square,
this is the desired area. When α > m x h, point E is to the right of point B and
we must subtract the area of the small triangle BEF to obtain the proper area.
Since triangle BEF is geometrically similar to triangle AEH, the ratio of their
areas is equal to the square of the ratio of the sides BE to AE, given by
µ ¶
Area of BEF α − mx h 2
= .
Area of AEH α
P1: ARS/mbg P2: ARS/vks QC: ARS
November 16, 1998 19:16 Annual Reviews AR075-16

DIRECT SIMULATION OF INTERFACIAL FLOW 587

This corresponds to the second term within the square brackets on the right-
hand side of Formula 27, which also contains the Heaviside step function
H (α − m x h), since the area of the triangle BEF is only subtracted if E is
to the right of B. Similarly, the third term within the square brackets in the
formula subtracts the area of the triangle DGH when α > m y h and point H
lies above point D. The single Formula 27 thus provides the area of the region
below the straight line (Equation 26), which lies in the original square of side
h for all possible cases. The area is a continuous, one-to-one, monotonically
increasing function of α. It ranges from zero, when α = 0, to h 2 , when α
reaches its maximum value of (m x + m y )h. There are two critical values of α,
Annu. Rev. Fluid Mech. 1999.31:567-603. Downloaded from www.annualreviews.org

corresponding to the zeros of the arguments of the Heaviside step functions in

by University of Massachusetts - Amherst on 09/24/12. For personal use only.

Formula 27, at which the function changes form. This occurs when the straight
line (Equation 26) passes through the corners B and D of the square, i.e. when
α = m x h or α = m y h.
In practice, not only does one need the “forward” Formula 27 between the
cut area and the parameter α, but the method also requires the “inverse” problem
of determining the α that corresponds to a given cut area and normal direction
in a computational cell. There are a number of ways to achieve this. One can
simply use a standard root-finding approach to find the particular value of α at
which the cut area has the desired value, for instance, an iterative method may be
found in Rider & Kothe (1998). Another option is as follows: Corresponding to
each critical value of α for which the interface passes through one of the corners
of the square, there exists a critical value of the cut area. In between any two
critical values, the function on the right hand side of Formula 27 is a known
polynomial in α whose roots can be evaluated analytically. Thus, to resolve the
inverse problem, we first identify which two critical values bound it on either
side and then obtain the root of the correct polynomial in α in that range.

5.2.2 PROPAGATION The second step of the VOF algorithm is propagation.

Once the interface has been reconstructed, its motion by the underlying flow
field must be modeled by a suitable advection algorithm. Here, we describe
in some detail the fractional step or operator split method, which updates the
volume fraction C by advecting the interface along one spatial direction at a
time. Intermediate C values are calculated during this process, and the final
C field is obtained only after advection of the interface along all coordinate
directions.
One oft-used way to calculate fluxes along the x direction is shown in
Figure 14, where all the fluid to the right of the dashed line will cross the right
boundary during time τ . In our calculations (Gueyffier et al 1998) we use a
Lagrangian approach; that is, we compute directly the motion of the interface
segments (Figure 15). We have found this scheme to be more robust. Because
P1: ARS/mbg P2: ARS/vks QC: ARS
November 16, 1998 19:16 Annual Reviews AR075-16

588 SCARDOVELLI & ZALESKI

Annu. Rev. Fluid Mech. 1999.31:567-603. Downloaded from www.annualreviews.org
by University of Massachusetts - Amherst on 09/24/12. For personal use only.

Figure 14 A simple scheme for the split computation of the fluxes. As in the upwind differencing
scheme, the whole block of fractional volume in a band of width uτ is transferred from the upwind
cell to the downwind cell. No account, however, is taken of the change of shape of the interface
during the time step.

in practice the time-stepping is performed separately in each spatial direction,

we only describe the advection of the interface along one spatial coordinate,
say x.
For each cell, three contributions are calculated: the area fluxes φ − and φ +
entering the cell (i, j), respectively, from cells (i − 1, j) and (i + 1, j) and the

Figure 15 A Lagrangian propagation step. The advection of the reconstructed segment is com-
puted, defining a new segment that straddles two cells. From that new segment, fluxes are calculated.
P1: ARS/mbg P2: ARS/vks QC: ARS
November 16, 1998 19:16 Annual Reviews AR075-16

DIRECT SIMULATION OF INTERFACIAL FLOW 589

area φ 0 of the fluid contained at the beginning of the step in the control cell
that remains there. The updated volume fraction in each cell after the fractional
step along the x direction is then given by
£ ¤
Ci,(x)j = φi,−j + φi,0 j + φi,+j . (28)

Then, the overall fractional-step procedure requires two reconstructions of the

interface and an advection step along each one of the two coordinate directions.
The Lagrangian advection method allows us to take into account the stretching
or compression of the interface during each single fractional step. The proce-
Annu. Rev. Fluid Mech. 1999.31:567-603. Downloaded from www.annualreviews.org

dure can be made second-order accurate by alternating the advection directions

by University of Massachusetts - Amherst on 09/24/12. For personal use only.

at each time step.

Unsplit algorithms are geometrically more complex because they also need
to take into account fluxes along the transverse direction, for example from cell
(i, j) to cell (i + 1, j + 1). First-order and second-order accurate algorithms
have been developed by several authors (Collela 1990, Bell et al 1990, LeVeque
1996, Puckett et al 1997, Rider & Kothe 1998). The results they present indi-
cate that for a given order of accuracy of the advection model, the unsplit
algorithm shows better resolution, especially near regions of high variation of
the derivatives, such as corners. Moreover it is less prone to asymmetries in the
numerical solution.
We conclude with a few remarks about possible developments on interface
reconstruction. Several schemes can be devised to reduce the discontinuities at
the cell faces present in the PLIC representation. We may calculate an aver-
age curvature at cell (i, j) using the PLIC reconstruction in a 3 × 3 block of
cells centered at cell (i, j) with a least-square procedure. We then calculate two
new intersections of the fitted circle with the faces of the cell and add an extra
internal point to approximate the given curvature and to conserve area. Pre-
liminary results (Manservisi et al 1998) indicate that this reconstruction of the
interface is consistently better than in any other scheme we considered. Several
other groups (DB Kothe, private communication) are also considering third-
order reconstruction. Propagation of the interface with this approach is under
development.

6. IMPLEMENTATION OF SURFACE TENSION

The surface tension term in the Navier-Stokes equation creates the most obvious
difficulties, since it is a singular term. In several implementations of the method,
these difficulties are manifest in both numerical instabilities and/or noise, and
in poor accuracy of capillary effects.
On adaptive grids such as those of Figures 6 and 7, surface tension may,
however, be easily represented as a jump condition for the pressure shown in
P1: ARS/mbg P2: ARS/vks QC: ARS
November 16, 1998 19:16 Annual Reviews AR075-16

590 SCARDOVELLI & ZALESKI

Equation 16 from one element to the next, or as a boundary condition for the
pressure shown in Equation 19.
On fixed grids, there are two options:
1. Smoothing the capillary force: Surface tension is implemented in a simple,
albeit approximate way, by distributing it over neighboring grid points.
2. Accurate finite volume balance of surface tension forces as realized in
Popinet & Zaleski 1998a: This option has been used only for two-dimension-
al calculations because of the complexity of the geometric considerations
involved.
Annu. Rev. Fluid Mech. 1999.31:567-603. Downloaded from www.annualreviews.org
by University of Massachusetts - Amherst on 09/24/12. For personal use only.

In the following section we only describe the first option.

6.1 Smoothing the Discontinuity of Capillary Forces
Several articles introduce what amounts in effect to distributing the surface
tension over grid points neighboring the interface. Consider the phase-charac-
teristic function χ. It may be smoothed by convolution with a kernel
Z
χ̃ (x) = χ (x0 )H (x − x0 ; ²) dx0 , (29)
V

where H (x; ²) is an integration kernel verifying H → δ S , when ² → 0. The

computation is based on the following remark: For a given value of c let Sc be
the surface with χ̃(x) = c, κ̃(x) the curvature of Sc at x, and m = ∇ χ̃. Then
from Equation 11
µ ¶
m
κ̃(x) = −∇ · . (30)
2||m||
When ² → 0, κ̃ → κ and one finds the “real” curvature. After Brackbill et al
(1992), this method is sometimes called the continuous surface force (CSF)
method. To implement the CSF method it is not necessary to perform a great
deal of smoothing. We, as well as several authors with whom we had private
communications, have directly implemented finite differences of χ without
smoothing or with very narrow kernels without any major problem. A discussion
of the merits of better kernels may be found for instance in Aleinov & Puckett
(1995). We (Lafaurie et al 1994) defined the finite filter
1 1
[F(C)]i j = Ci j + [Ci, j−1 + Ci, j+1 + Ci−1, j + Ci+1, j ]. (31)
2 8
The action of this filter may be repeated m f times to yield an additional degree
of smoothing C̃ = F (m f ) (C). The gradient m is then computed by finite dif-
ferences, using for instance Expressions 22 and 23. We have found that one or
two iterations of this filter were optimal.
P1: ARS/mbg P2: ARS/vks QC: ARS
November 16, 1998 19:16 Annual Reviews AR075-16

DIRECT SIMULATION OF INTERFACIAL FLOW 591

It is possible to connect the use of a filtered surface tension with the idea of a
thickened interface (Anderson et al 1998). A thick interface occurs naturally in
the Van der Waals-Cahn-Hilliard theory of interfaces. In this theory, matter is
considered at an intermediate scale between the classical continuum mechanics
scale at which interfaces are sharp and the molecular scale where matter is
discontinuous. At this scale, continuous fields of density and velocity still exist
and interfaces have a finite thickness. An example of a dynamical equation
with thick interfaces for a one-species, liquid-vapor flow as well as numerical
simulations of that model may be found in Nadiga & Zaleski (1996). For two
incompressible phases, the Van der Waals-Cahn-Hilliard framework was used
Annu. Rev. Fluid Mech. 1999.31:567-603. Downloaded from www.annualreviews.org

by Jacqmin (1996). An application to moving contact lines may be found in

by University of Massachusetts - Amherst on 09/24/12. For personal use only.

Seppecher (1996).
6.2 Momentum Conservation
The smoothing method makes it possible to find an approximation of the sin-
gular tensor defined in Equation 12,
T = σ (I − n ⊗ n)||∇ C̃||, (32)
where the unit normal n is
∇ χ̃
n= . (33)
||∇ χ̃||
The momentum-conserving formulation is a finite difference approximation
of Equation 32. In practice, a 4- or 6-point approximation was used for the
computation of the normals.
As in the method used by Brackbill et al (1992), this approximation con-
verges in theory toward the true normal as H → δ S . When the kernel H² is
narrowed one should keep m f large in order to have h ¿ ². However, numer-
ical experiences have shown that the best results are obtained when m f = 1
or 2. Larger values of m f increase the spurious currents.
6.3 Other Methods for Surface Tension
The direct measurement of normal vectors and their derivatives is possible when
the interface is tracked by markers or when it is an element boundary.
We have tested a method of surface markers together with the MAC finite
difference method (Popinet & Zaleski 1998a). The surface stresses were again
computed and integrated over the boundary of a finite volume for the x and
y momentum. When carefully implemented, this method makes it possible to
considerably reduce the spurious currents.
6.4 Capillary Waves
The simulation of capillary waves is an important test for the interface simula-
tion methods in general. It has been performed in detail by Fyfe et al (1988) for
P1: ARS/mbg P2: ARS/vks QC: ARS
November 16, 1998 19:16 Annual Reviews AR075-16

592 SCARDOVELLI & ZALESKI

their free Lagrangian method. Their computations were carefully analyzed for
possible errors coming from finite domain size, nonlinear effects, and viscous
effects. High levels of accuracy were obtained for waves on planar interfaces,
but the accuracy was less for capillary oscillations of cylinders (two-dimensional
flows around circular patches). For the conservative method described above,
tests were presented by Gueyffier et al (1998). For comparisons with the level-
set method see Sussman et al (1994) and Sussman & Smereka (1997). For all
these fixed-grid methods the accuracy is of the order of 1%. Marker meth-
ods with smoothed surface tension also give comparable results (Nobari et al
1996). However, somewhat more accuracy, of the order of 0.5%, may be ob-
Annu. Rev. Fluid Mech. 1999.31:567-603. Downloaded from www.annualreviews.org

tained by an accurate finite-volume balance (Popinet & Zaleski 1998a).

by University of Massachusetts - Amherst on 09/24/12. For personal use only.

6.5 Spurious Currents

A major difficulty with many methods of interface calculations is the existence
of the so-called spurious or parasite currents. These currents are vortices ap-
pearing in numerical simulations in the neighborhood of interfaces despite the
absence of any external forcing. They are observed with many surface tension
simulation methods, including the CSF method, the conservative method of
Lafaurie et al (1994), and the lattice-Boltzmann method (Gunstensen 1992), in
which they were first discovered. They result in a limitation of the range of
parameters that may be accessed by the model.
It is difficult to give a systematic expression for the amplitude of spurious
currents because they often fluctuate in time, but direct measurements (Lafaurie
et al 1994) yield approximately
u p ' 0.01 σ/µ. (34)
These currents were also measured around droplets in the connected marker
method of Tryggvason and coworkers. There spurious currents appear to be of
the order
u p ∼ 10−5 σ/µ
(Tryggvason, unpublished lecture notes).
6.6 Problems with Large Surface Tension
The simulation methods for flows with interfaces are limited in the range of
dimensionless numbers that may be reached. The situation is analogous to that
of direct numerical simulations of homogeneous flows, where there is an upper
limit for the Reynolds number that grows like a power of the number of grid
points.
For interface methods, calculations often become difficult when the surface
tension becomes large and the density jump is also large. This may happen for
bubbles or droplets of small radius R, when surface tension is large compared
P1: ARS/mbg P2: ARS/vks QC: ARS
November 16, 1998 19:16 Annual Reviews AR075-16

DIRECT SIMULATION OF INTERFACIAL FLOW 593

with viscosity. This means that either the capillary number Ca = µU/σ is small
or that the Laplace number La = σρ R/µ2 is large. Numerical experiments
show that at small Ca, but moderate La, no catastrophic instability occurs,
and that the only troublesome effects are the spurious currents. However, at
large La, parasite currents may grow and become large enough to destroy the
interface. This may be easily explained by the connection between La and
Reynolds numbers of spurious currents given by Equation 34,

La = 100 Re p , (35)
Annu. Rev. Fluid Mech. 1999.31:567-603. Downloaded from www.annualreviews.org

where Re p = u p R/ν. In practice, computations become difficult when La

by University of Massachusetts - Amherst on 09/24/12. For personal use only.

∼106 . This is approximately the value for a 1-cm droplet or bubble. Thus, this
droplet size represents the borderline for current-day simulations.
For larger droplets and Reynolds numbers, one needs to compare surface
tension and inertia. At small Weber numbers, We = U 2 ρ R/σ , capillary ten-
sion dominates inertial phenomena. Since this happens for smaller R, it may
explain why several articles and private communications report difficulties with
smaller particles.

6.7 Problems with Small Surface Tension

At large Weber numbers, problems also occur because the computed objects
(droplets, bubbles, filaments) become very small. In this situation, the breakup
phenomenon tends to create smaller structures until the Weber number becomes
small enough, at around We = 10. If and when these structures reach sizes
smaller than the mesh, a loss of accuracy occurs.
First-order methods, such as the SOLA-VOF method (Hirt & Nichols 1981),
are numerically unstable in the absence of surface tension. The interface is
progressively destroyed by the generation of flotsam, and surface tension is the
only way to keep the interface well defined. On the other hand, PLIC methods
(Li 1995) and other more accurate methods such as front tracking avoid this
numerical instability. The small structures created at large Weber numbers are
often well resolved and have physical meaning.
Moreover, at low Reynolds numbers, small surface tension implies large
Ca. This is equally catastrophic, since viscous flow may create cusps in the
interface.

7. TESTS AND APPLICATIONS

7.1 A Reconnection Test: Sessile Droplets
Sessile droplets are created by letting the liquid flow very slowly out of a crank.
A pendant drop forms and remains attached to the crank only through a thin
liquid bridge, which eventually pinches in several places.
P1: ARS/mbg P2: ARS/vks QC: ARS
November 16, 1998 19:16 Annual Reviews AR075-16

594 SCARDOVELLI & ZALESKI

This problem was studied in an inviscid framework by Schulkes (1994) and

Og̃uz & Prosperetti (1993) using a boundary-integral method. Eggers & Dupont
(1994) obtained an approximation of the full Navier-Stokes equation, which is
in part a lubrication approximation valid in the limit of slowly varying thickness
R(z). It is a lubrication approximation only in part because it includes a surface-
tension term that is exact, and thus the Eggers-Dupont equation is also valid for
static shapes with arbitrary slopes of R(z), such as droplets. Both approaches
(Schulkes 1994, Eggers & Dupont 1994) were found to be in very good agree-
ment with the experiments of Peregrine (Peregrine et al 1990). More recently,
full three-dimensional and two-dimensional axisymmetric viscous simulations
Annu. Rev. Fluid Mech. 1999.31:567-603. Downloaded from www.annualreviews.org

have been carried out for this physical situation using the VOF method by
by University of Massachusetts - Amherst on 09/24/12. For personal use only.

Gueyffier et al (1998) and the results show an excellent agreement with the
pictures of Peregrine et al (1990).

7.2 Sedimentation of Droplet Arrays in Creeping Flow

The asymptotic fall velocity of a droplet is given by the classic Hadamard-
Rybczynski expression (Batchelor 1967)
2 1+K
UHR = (ρout − ρdrop )a 2 g, (36)
3µout 2 + 3K
where µout and ρout are the dynamical viscosity and density of the outer fluid and
K = µout /µdrop is the ratio between the dynamic viscosities of the two fluids.
Sangani (1987) found solutions to the Stokes equation for periodic arrays of
droplets. The ratio between the sedimentation velocity of the array of drops U
and Hadamard-Rybczynski velocity for a single drop UHR only depends on the
viscosity ratio K and on the volume fraction between drops and the outer fluid c.
These theoretical results were compared with three-dimensional calculations
with our VOF/PLIC method. The results are shown in Figure 16. The error
increases with K. This is due to the manner in which viscosity is averaged in
cut cells. This creates O(h) errors in the viscous stresses. This problem was
addressed by Coward et al (1997).

7.3 Breakup of Liquid Jets and Kelvin-Helmholtz

Instability
The phenomenon of droplet peeling on the surface of liquid jets has been abun-
dantly studied (Hoyt & Taylor 1977, Reitz & Bracco 1982). These jets are
often produced in engine nozzles. The mechanism for the breakup of the jet is
complex and difficult to observe because of the very small space- and timescales
involved. Numerical simulation is therefore well placed to deal with this prob-
lem. Breakup of liquid jet has been studied using VOF methods at relatively low
We by Richards et al (1993, 1994). For fluids of interest in combustion studies
P1: ARS/mbg P2: ARS/vks QC: ARS
November 16, 1998 19:16 Annual Reviews AR075-16

DIRECT SIMULATION OF INTERFACIAL FLOW 595

Annu. Rev. Fluid Mech. 1999.31:567-603. Downloaded from www.annualreviews.org
by University of Massachusetts - Amherst on 09/24/12. For personal use only.

Figure 16 Ratio between sedimentation velocity of an infinite array of drops and Stokes velocity
for a single drop.

(such as liquid-oxygen/hydrogen or hydrocarbon/air mixtures), the principal

difficulties occur from the very large values of the capillary number Ca. This
number is about 0.25 in a typical water jet in air at 10 m/s water velocity, whereas
it can reach the value 10 in hydrocarbons or cryogenic fluids. Figure 17 shows an
example of such a peeling obtained for Ca = 0.02, whereas Figure 18 shows an
example for Ca = 1. The simulations give different results. In the second case,
structures of very small size are formed because of the systematic stretching of
a thin filament of liquid. The precision of the calculation is limited by the forma-
tion of these very small structures. Our three-dimensional simulations analyze
the response to spanwise perturbations of the above two-dimensional base-flow
(Zaleski et al 1996). The pinching mechanism by which liquid droplets de-
tach from the edges of sheets in two-dimensional flow is not necessarily the
realistic three-dimensional regime. We have been able so far to distinguish two
scenarios by which three-dimensional flow sets in. For small three-dimensional
perturbations, the simulation remains two-dimensional until the sheet breaks
and a cylinder pinches off. Subsequently, the capillary instability of jets gives
three-dimensional structure to the flow. In this case, the two-dimensional sim-
ulations are useful for predicting the evolution of the flow and the droplet size.
P1: ARS/mbg P2: ARS/vks QC: ARS
November 16, 1998 19:16 Annual Reviews AR075-16

596 SCARDOVELLI & ZALESKI

Annu. Rev. Fluid Mech. 1999.31:567-603. Downloaded from www.annualreviews.org
by University of Massachusetts - Amherst on 09/24/12. For personal use only.

Figure 17 Droplet formation in a 2D sheared layer. The parameters are ρ L /ρG = 10, ReG =
7500, and W eG = 150. The liquid phase is the grey area. The lines are iso-vorticity levels.

In contrast, for higher three-dimensional perturbations, a different scenario is

observed, as shown in Figure 19 (this simulation took about 48 h to complete
on an IBM RS6000/370). The rim-like edge concentrates in a protruding finger
that will be subject to the capillary instability. However, the unstable cylinder
is now streamwise. The formation of streamwise filaments or fingers is ob-
served in real experiments on shear layers (Raynal et al 1995) and it is also
reminiscent of the crown formation in droplet impact experiments (Yarin &
Weiss 1995).

7.4 Other Applications

Although the surface-wave problem has been studied for a long time using
boundary-integral methods for inviscid flow, it is only recently that several
P1: ARS/mbg P2: ARS/vks QC: ARS
November 16, 1998 19:16 Annual Reviews AR075-16

DIRECT SIMULATION OF INTERFACIAL FLOW 597

Annu. Rev. Fluid Mech. 1999.31:567-603. Downloaded from www.annualreviews.org
by University of Massachusetts - Amherst on 09/24/12. For personal use only.

Figure 18 Droplet formation at higher capillary and Weber numbers. The intermediate steps are
not shown. The level lines of the vortices are omitted in b to avoid overcrowding the plot.
P1: ARS/mbg P2: ARS/vks QC: ARS
November 16, 1998 19:16 Annual Reviews AR075-16

598 SCARDOVELLI & ZALESKI

Annu. Rev. Fluid Mech. 1999.31:567-603. Downloaded from www.annualreviews.org
by University of Massachusetts - Amherst on 09/24/12. For personal use only.

Figure 19 Breakup of the interface through a secondary instability of the sheet. ρ L /ρG =
10, Re = 1000, and W e = 300.

authors have attempted to solve the full Navier-Stokes equations. The dissipa-
tion of surface waves was studied by Yang & Tryggvason (1998). Wave breaking
and the mixing of the fluid from the plunging jet was studied by Abadie & Calt-
agirone (1998). Breaking waves with plunging jets and several splash-ups were
simulated by Chen et al (1998).
Applications of surface marker methods to biology are a promising area of
research. One example in this direction is Agresar et al (1998). Boiling flow is
also a new application of surface-tracking methods (Juric & Tryggvason 1998).
Finally, bubbly flow with a large number of bubbles has been intensively
studied both in two and three dimensions by several groups (see Esmaeeli &
Tryggvason 1996, 1998a,b; Tomiyama 1998).
P1: ARS/mbg P2: ARS/vks QC: ARS
November 16, 1998 19:16 Annual Reviews AR075-16

DIRECT SIMULATION OF INTERFACIAL FLOW 599

Of great importance for modeling multiphase flow is the study of single

bubbles in various configurations. Three-dimensional numerical simulation is
required for the study of asymmetric flow such as bubbles in shear flows (Ervin
& Tryggvason 1997).

8. CONCLUSION
In our opinion, the main progress witnessed over the past 10 years is the advent
of many three-dimensional calculations. Two-dimensional calculations have
also matured, producing either highly accurate results or calculations over im-
Annu. Rev. Fluid Mech. 1999.31:567-603. Downloaded from www.annualreviews.org

pressively large grids.

by University of Massachusetts - Amherst on 09/24/12. For personal use only.

The one new method that appeared is level sets. It is seducing by its sim-
plicity but has not yet produced the wide range of results, especially in three-
dimensions, achieved by older methods. The VOF method has been consider-
ably improved by the systematic use in the scientific literature of the higher
order piecewise linear interface calculation method.
Important calculations have not yet been done. For instance, there is no
three-dimensional calculation of a falling drop or rising bubble in the difficult
air/water conditions and in the nontrivial regimes where axial symmetry does
not hold. Many complex flows, such as atomizing jets, splashing droplets, or
breaking waves, have been studied only in a preliminary way. Finally, for some
flows, quantitative results are difficult to get, in particular, agreement between
simulations and linear theory for shear flow instabilities.
For engineering applications, new methods need to be developed that could
adapt to complex geometries, such as combustors and nozzles, to cite just two
of a whole world of examples.
It is our opinion that these difficulties should soon be resolved. This would
allow these methods to connect directly with scientific problems and engineer-
ing applications in multiphase flow. From there the simulation of interfaces
could move on to incorporate more complex physics, such as models of inter-
molecular forces, complex surface rheology, and the transport and effects of
tensio-active molecules.

Visit the Annual Reviews home page at

http://www.AnnualReviews.org

Literature Cited

Abadie S, Caltagirone J. 1998. Breaking waves. and adhesion of circulating cells. J. Comput.
Paris: C. R. Acad. Sci. In press Phys. 143:346–80
Agresar G, Linderman J, Tryggvason G, Pow- Aleinov I, Puckett EG. 1995. Computing sur-
ell K. 1998. An adaptive, Cartesian Front- face tension with high-order kernels. In Proc.
tracking method for the motion, deformation 6th Int. Symp. Comput. Fluid Dyn. ed. K
P1: ARS/mbg P2: ARS/vks QC: ARS
November 16, 1998 19:16 Annual Reviews AR075-16

600 SCARDOVELLI & ZALESKI

Oshima, pp. 6–13. Lake Tahoe, CA. May axisymmetric free-surface flow. Phys. Rev.
be obtained from the author’s web page at Lett. 71:3458–61
http://math.math.ucdavis.edu/∼egp/ Eggers J. 1997. Nonlinear dynamics and break-
Anderson DM, McFadden GB, Wheeler AA. up of free-surface flows. Rev. Mod. Phys.
1998. Diffuse-interface methods in fluid me- 69:865–929
chanics. Annu. Rev. Fluid Mech. 30 Eggers J, Dupont TD. 1994. Drop forma-
Ashgriz N, Poo JY. 1991. FLAIR: Flux line- tion in a one-dimensional approximation of
segment model for advection and interface the Navier-Stokes equation. J. Fluid Mech.
reconstruction. J. Comput. Phys. 92:449–68 262:205–21
Baker GR, Meiron DI, Orszag SA. 1982. Gener- Ervin E, Tryggvason G. 1997. The rise of bub-
alized vortex methods for free surface flows bles in a vertical shear flow. ASME J. Fluid
problems. J. Fluid Mech. 123:477 Eng. 119:443–49
Baker GR, Meiron DI, Orszag SA. 1984. Esmaeeli A, Tryggvason G. 1996. An inverse
Boundary integral methods for axisymmetric energy cascade in two-dimensional, low Rey-
Annu. Rev. Fluid Mech. 1999.31:567-603. Downloaded from www.annualreviews.org

and three dimensional Rayleigh-Taylor insta- nolds number bubbly flows. J. Fluid Mech.
bility problems. Physica D 12:19–31 314:315–30
by University of Massachusetts - Amherst on 09/24/12. For personal use only.

Batchelor GK. 1967. An Introduction to Fluid Esmaeeli A, Tryggvason G. 1998a. Direct nu-
Dynamics. Cambridge: Cambridge Univ. merical simulations of bubbly flows. Part I.
Press Low Reynolds number arrays. Submitted for
Bell J, Dawson C, Shubin G. 1990. An un- publication
split, higher order Godunov method for scalar Esmaeeli A, Tryggvason G. 1998b. Direct nu-
conservation laws in multiple dimensions. J. merical simulations of bubbly flows. Part II.
Comput. Phys. 74:171–200 Moderate Reynolds number arrays. Submit-
Benzi R, Succi S, Vergassola M. 1992. The lat- ted for publication
tice Boltzmann equation: theory and appli- Floryan JM, Rasmussen H. 1989. Numeri-
cations. Phys. Rep. 222:145–97 cal methods for viscous flow with moving
Boulton-Stone JM, Blake JR. 1993. Gas bub- boundaries. Appl. Mech. Rev. 42:323–40
bles bursting at a free surface. J. Fluid Mech. Fritts MJ. Cowley W, Trease HE, eds. 1985. The
254:437–66 Free Lagrange Method. Lect. Notes Phys.
Brackbill J, Kothe DB, Zemach C. 1992. A con- Vol. 238. New York: Springer-Verlag
tinuum method for modeling surface tension. Fukai J, Shiiba Y, Yamamoto T, Miyatake O,
J. Comput. Phys. 100:335–54 Poulikakos D, et al. 1995. Wetting effects on
Chen G, Kharif C, Li J, Zaleski S. 1998. the spreading of a liquid droplets colliding
Two-dimensional Navier-Stokes simulation with a flat surface: experiment and model-
of breaking waves. Submitted for publication ing. Phys. Fluids 7:236–47
Chorin AJ. 1980. Flame advection and propa- Fukai J, Zhao Z, Poulikakos D, Megaridis CM,
gation algorithms. J. Comput. Phys. 35:1–11 Miyatake O. 1993. Modeling of the deforma-
Collela P. 1990. Multidimensional upwind tion of a liquid droplet impinging on a flat
methods for hyperbolic conservation laws. J. surface. Phys. Fluids A 5:2588–99
Comput. Phys. 87:171–200 Fyfe DE, Oran ES, Fritts MJ. 1988. Surface ten-
Coward AV, Renardy YY, Renardy M, Richards sion and viscosity with Lagrangian hydro-
JR. 1997. Temporal evolution of periodic dis- dynamics on a triangular mesh. J. Comput.
turbances in two-layer Couette flow. J. Com- Phys. 76:349–84
put. Phys. 132:346–61 Glimm J, Bryan OM, Menikoff R, Sharp D.
Cuenot B, Magnaudet J, Spennato B. 1997. 1986. Front tracking applied to Rayleigh-
The effects of slightly soluble surfactants on Taylor instability. SIAM J. Sci. Stat. Comput.
the flow around a spherical bubble. J. Fluid 7:230–51
Mech. 339:25–53 Glimm J, McBryan O, Menikoff R, Sharp DH.
Dandy DS, Leal LG. 1989. Buoyancy-driven 1987. Front tracking applied to Rayleigh-
motion of a deformable drop through a qui- Taylor instability. SIAM J. Sci. Stat. Comput.
escent liquid at intermediate Reynolds num- 7:230–51
bers. J. Fluid Mech. 208:161–92 Gueyffier D, Nadim A, Li J, Scardovelli R, Za-
DeBar R. 1974. Fundamentals of the KRAKEN leski S. 1998. Volume of fluid interface track-
code. Tech. rep. UCIR-760. Lawrence Liver- ing with smoothed surface stress methods for
more Nat. Lab. three-dimensional flows. Submitted for pub-
de Gennes PG. 1985. Wetting: statics and dy- lication
namics. Rev. Mod. Phys. 57:827–63 Gunstensen AK. 1992. Lattice-Boltzmann stud-
Duraiswami R, Prosperetti A. 1992. Orthogo- ies of multiphase flow through porous media.
nal mapping in two dimensions. J. Comput. PhD thesis. MIT, Cambridge, MA
Phys. 98:254–68 Harlow F. 1964. The Particle-in-Cell Comput-
Eggers J. 1993. Universal pinching of 3D ing Method for Fluid Dynamics, In Methods
P1: ARS/mbg P2: ARS/vks QC: ARS
November 16, 1998 19:16 Annual Reviews AR075-16

DIRECT SIMULATION OF INTERFACIAL FLOW 601

of Computational Physics, Vol. 3, pp. 313– dynamics. Annu. Rev. Astron. Astrophys. 30:
43. New York: Academic 543–74
Hirt CW, Amsden AA, Cook JL. 1974. An arbi- Monaghan J. 1994. Simulating free surface
trary Lagrangian Eulerian computing method flows with SPH. J. Comput. Phys. 110:399–
for all speeds. J. Comput. Phys. 14:227–53 406
Hirt CW, Nichols BD. 1981. Volume of fluid Monaghan J, Bicknell P, Humble R. 1994. Vol-
(VOF) method for the dynamics of free boun- canoes, tsunamis and the demise of the Mi-
daries. J. Comput. Phys. 39:201–25 noans. Physica D 77:217–28
Hirt CW, Nichols BD. 1988. Flow-3D users Nadiga BT, Zaleski S. 1996. Investigations of
manual. Tech. Rep. Flow Sciences, Inc. a two-phase fluid model. Eur. J. Mech. B 15:
Hoyt JW, Taylor J. 1977. Waves on water jets. 885–96
J. Fluids Mech. 83:119–27 Nichols BD, Hirt CW, Hotchkiss RS. 1980.
Jacqmin D. 1996. An energy approach to the SOLA-VOF: A solution algorithm for tran-
continuum surface tension method. AIAA sient fluid flow with multiple free boundaries.
Annu. Rev. Fluid Mech. 1999.31:567-603. Downloaded from www.annualreviews.org

Pap. 96-0858 Tech. Rep. LA-8355, Los Alamos Nat. Lab.

Juric D, Tryggvason G. 1998. Computations Nobari M, Jan Y-J, Tryggvason G. 1996. Head-
by University of Massachusetts - Amherst on 09/24/12. For personal use only.

of boiling flows. Int. J. Multiphase Flow. In on collision of drops—a numerical investiga-

press tion. Phys. Fluids 8:29–42
Kane JH. 1994. Boundary-Element Analysis. Noh W, Woodward P. 1976. SLIC (simple line
Eaglewood Cliffs, NJ: Prentice Hall interface calculation). In Proc. 5th Int. Conf.
Kothe DB, Mjolsness RC. 1992. RIPPLE: A Fluid Dyn. Vol. 59, Lect. Notes Phys. ed. A
new model for incompressible flows with free van de Vooren, P Zandbergen, pp. 330–40.
surfaces. AIAA J. 30:2694–700 Berlin: Springer-Verlag
Ladd AJC. 1994. Numerical simulations of par- Norman ML, Winkler K-H. 1986. 2-D Eulerian
ticulate suspensions via a discretized Boltz- hydrodynamics with fluid interfaces, self-
mann equation. Part 1. Theoretical founda- gravity and rotation. In Astrophysical Radi-
tion. J. Fluid Mech. 271:285–310 ation Hydrodynamics, ed. K-H Winkler, ML
Lafaurie B, Nardone C, Scardovelli R, Za- Norman, pp. 187–221. New York: Reidel
leski S, Zanetti G. 1994. Modelling, merging Og̃uz HN, Prosperetti A. 1990. Bubble entrain-
and fragmentation in multiphase flows with ment by the impact of drops on liquid sur-
SURFER. J. Comput. Phys. 113:134–47 faces. J. Fluid Mech. 203:143–79
LeVeque R. 1996. An unsplit, higher order Go- Og̃uz HN, Prosperetti A. 1993. Dynamics of
dunov method for scalar conservation laws in bubble growth and detachment from a nee-
multiple dimensions. SIAM J. Numer. Anal. dle. J. Fluid Mech. 257:111–45
33:627–65 Okuzono T. 1997. Smoothed-particle method
Li J. 1995. Calcul d’interface affine par mor- for phase separation in polymer mixtures.
ceaux (piecewise linear interface calcula- Phys. Rev. E 56:4416–27
tion). CR Acad. Sci. Paris, Sér. IIb, 320:391– Olson JF, Rothman DH. 1977. Two-fluid flow in
96 sedimentary rock: simulation, transport, and
Longuet-Higgins MS. 1998. Vorticity and cur- complexity. J. Fluid Mech. 341:343–70
vature at a free surface. J. Fluid Mech. 356: Parker BJ, Youngs DL. 1992. Two and three
149–53 dimensional Eulerian simulation and fluid
Lundgren TS, Mansour NN. 1988. Oscillations flow with material interfaces. Tech. Rep. UK
of drops in zero gravity with weak viscous Atomic Weapons Establ.
effects. J. Fluid Mech. 194:479–510 Peregrine DH, Shoker G, Simon A. 1990. The
Lundgren TS, Mansour NN. 1991. Vortex ring bifurcation of liquid bridges. J. Fluid Mech.
bubbles. J. Fluid Mech. 224:177–96 212:25–39
Magnaudet J, Rivero M, Fabre J. 1995. Acceler- Pilliod JE Jr, Puckett EG. 1997. Second-
ated flows around a rigid sphere or a spherical order accurate volume-of-fluid algorithms
bubble. Part I. Steady straining flow. J. Fluid for tracking material interfaces. Tech. Rep.
Mech. 284:97–136 Lawrence Berkeley Nat. Lab., No. LBNL-
Manservisi S, Popinet S, Scardovelli R, Zaleski 40744
S. 1998. Two front-tracking algorithms for Popinet S, Scardovelli R, Zaleski S. 1997. Code
2D interface problems. Proc. ICMF98, Lyon, de suivi de volume ciam en maillage irreg-
8–12 June ulier. Intern. LMM, Rep., Lab. Modélisation
McHyman J. 1984. Numerical methods for Mécanique, UMR CNRS 7607, Univ. Pierre
tracking interfaces. Physica D 12:396–407 Marie Curie
Miller GH, Puckett EG. 1996. A high-order Popinet S, Zaleski S. 1998a. A front tracking
Godunov method for multiple condensed algorithm for the accurate representation of
phases. J. Comput. Phys. 128:134–64 surface tension. Int. J. Numer. Methods Flu-
Monaghan J. 1992. Smoothed particle hydro- ids. In press
P1: ARS/mbg P2: ARS/vks QC: ARS
November 16, 1998 19:16 Annual Reviews AR075-16

602 SCARDOVELLI & ZALESKI

Popinet S, Zaleski S. 1998b. Simulation of axi- mechanics. Part 2: Buoyancy-driven motion

symmetric free-surface viscous flow around of a gas bubble through a quiescent liquid. J.
a non-spherical bubble in the sonolumines- Fluid Mech. 148:19–36
cence regime. Proc. ICMF98, Lyon, 8–12 Sangani AS. 1987. Sedimentation in ordered
June emulsions of drops at low Reynolds numbers.
Puckett EG. 1991. A volume-of-fluid interface J. Appl. Math. Phys. (ZAMP) 38:542–56
tracking algorithm with applications to com- Saurel R, Abgrall R. 1998. A simple method for
puting shock wave refraction. In Proc. 4th Int. compressible multifluid flows. SIAM J. Sci.
Symp. Comput. Fluid Dyn. Davis, CA, ed. H Comput. In press
Dwyer, pp. 933–38 Schulkes RMS. 1994. The evolution and bi-
Puckett EG, Almgren AS, Bell JB, Marcus furcation of a pendant drop. J. Fluid Mech.
DL, Rider WJ. 1997. A high-order projection 278:83–100
method for tracking fluid interfaces in vari- Seppecher P. 1996. Moving contact lines in
able density incompressible flows. J. Comp. the Cahn-Hilliard theory. Int. J. Engin. Sci.
Annu. Rev. Fluid Mech. 1999.31:567-603. Downloaded from www.annualreviews.org

Phys. 130:269–82 34:977–92

Puckett EG, Saltzman JS. 1992. A 3D adaptive Sethian JA. 1996. Level Set Methods. Cam-
by University of Massachusetts - Amherst on 09/24/12. For personal use only.

mesh refinement algorithm for interfacial gas bridge Univ. Press

dynamics. Physica D 60:84–93 Shyy W, Udaykumar H, Rao MM, Smith RW.
Raynal L, Villermaux E, Hopfinger EJ. 1995. 1996. Computational Fluid Dynamics With
Couche de Mélange eau/air à grands rapports Moving Boundaries. London: Taylor & Fran-
de quantité de mouvement. Rapport Final, cis
Contrat SEP no 910023 Stone HA. 1994. Dynamics of drop deforma-
Reitz D, Bracco F. 1982. Mechanism of atomi- tion and breakup in viscous fluids. Annu. Rev.
sation of a liquid jet. Phys. Fluids 25:1730–42 Fluid Mech. 26:65–102
Richards JR, Beris AN, Lenhof AM. 1993. Sussman M, Smereka P. 1997. Axisymmet-
Steady laminar flow of liquid-liquid jets at ric free boundary problems. J. Fluid Mech.
high Reynolds numbers. Phys. Fluids 5: 341:269–94
1703–17 Sussman M, Smereka P, Osher S. 1994. A
Richards JR, Lenhof AM, Beris AN. 1994. Dy- level set approach for computing solutions to
namic breakup of liquid-liquid jets. Phys. imcompressible two-phase flow. J. Comput.
Fluids 6:2640–55 Phys. 114:146–59
Rider WJ, Kothe DB. 1995. Stretching and tear- Tomiyama A. 1998. Struggle with computa-
ing interface tracking methods. AIAA Pap. tional bubble dynamics. In Third Int. Conf.
95-1717 on Multiphase Flow, ICMF ’98, Lyon, June
Rider WJ, Kothe DB. 1998. Reconstructing vol- 8–12, 1998. Proc. on CD-ROM
ume tracking. J. Comput. Phys. 141:112–52 Torrey MD, Cloutman LD, Mjolsness RC, Hirt
Rothman DH, Zaleski S. 1994. Lattice-gas mod- CW. 1985. NASA-VOF2D: a computer pro-
els of phase separation: interfaces, phase gram for incompressible flow with free sur-
transitions, and multiphase flow. Rev. Mod. faces. Tech. Rep. LA-101612-MS, Los Alamos
Phys. 66:1417–79 Nat. Lab.
Rothman DH, Zaleski S. 1997. Lattice-Gas Cel- Torrey MD, Mjolsness RC, Stein LR. 1987.
lular Automata. Cambridge, UK: Cambridge NASA-VOF3D: a three-dimensional com-
Univ. Press puter program for incompressible flow with
Rottman JW, Olfe DB. 1977. Comment on dis- free surfaces. Tech. Rep. LA-11009-MS, Los
cretized simulations of vortex sheet evolution Alamos Nat. Lab.
with buoyancy and surface tension effects. Tryggvason G, Unverdi SO. 1990. Computa-
AIAA J. 15:1214–16 tions of three-dimensional Rayleigh-Taylor
Rudman M. 1997. Volume-tracking methods for instability. Phys. Fluids A 2:656–59
interfacial flow calculations. Int. J. Numer. Tsai TM, Miksis MJ. 1994. Dynamics of a drop
Meth. Fluids 24:671–91 in a constricted capillary tube. J. Fluid. Mech.
Rudman M. 1998. A volume-tracking method 274:197–217
for incompressible multifluid flows with large Unverdi SO, Tryggvason G. 1992a. Computa-
density variations. Int. J. Numer. Meth. Flu- tion of multifluid flow. Physica D 60:70–83
ids. In press Unverdi SO, Tryggvason G. 1992b. A front-
Ryskin G, Leal LG. 1984a. Numerical solution tracking method for viscous, incompressible,
of free-boundary problems in fluid mechan- multi-fluid flows. J. Comput. Phys. 100:25–
ics. Part 2: Buoyancy-driven motion of a gas 37
bubble through a quiescent liquid. J. Fluid Waldvogel J, Poulikakos D, Wallace DB. Ma-
Mech. 148:1–18 rusak R. 1996. Transport phenomena in pico-
Ryskin G, Leal LG. 1984b. Numerical so- liter size solder droplet deposition. ASME J.
lution of free-boundary problems in fluid Heat Transfer 118:148–56
P1: ARS/mbg P2: ARS/vks QC: ARS
November 16, 1998 19:16 Annual Reviews AR075-16

DIRECT SIMULATION OF INTERFACIAL FLOW 603

Whitham G. 1974. Linear and Non-Linear Zaleski S, Li J, Scardovelli R, Zanetti G. 1996.

Waves. New York: Wiley-Interscience Direct simulation of multiphase flows with
Yang Y, Tryggvason G. 1998. Dissipation of density variations. In Colloq. IUTAM “Vari-
energy by finite amplitude surface waves. able Density Low Speed Turbulent Flows.”
Comput. Fluids. In press Kluwer
Yarin AL, Weiss DA. 1995. Impact of drops on Zalosh RG. 1976. Discretized simulations of
solid surfaces: self-similar capillary waves, vortex sheet evolution with buoyancy and
and splashing as a new type of kinematic dis- surface tension effects. AIAA J. 14:1517–
continuity. J. Fluid Mech. 283:141–73 23
Annu. Rev. Fluid Mech. 1999.31:567-603. Downloaded from www.annualreviews.org
by University of Massachusetts - Amherst on 09/24/12. For personal use only.
 Annual Review of Fluid Mechanics
Volume 31, 1999

CONTENTS
Linear and Nonlinear Models of Aniosotropic Turbulence, Claude
1
Cambon, Julian F. Scott
Transport by Coherent Barotropic Vortices, Antonello Provenzale 55

Nuclear Magnetic Resonance as a Tool to Study Flow, Eiichi Fukushima 95

Computational Fluid Dynamics of Whole-Body Aircraft, Ramesh

125
Agarwal

Liquid and Vapor Flow in Superheated Rock, Andrew W. Woods 171

Annu. Rev. Fluid Mech. 1999.31:567-603. Downloaded from www.annualreviews.org

The Fluid Mechanics of Natural Ventilation, P. F. Linden 201

by University of Massachusetts - Amherst on 09/24/12. For personal use only.

Flow Control with Noncircular Jets, E. J. Gutmark, F. F. Grinstein 239

Magnetohydrodynamics in Materials Processing, P. A. Davidson 273

Nonlinear Gravity and Capillary-Gravity Waves, Frédéric Dias,

301
Christian Kharif
Fluid Coating on a Fiber, David Quéré 347
Preconditioning Techniques in Fluid Dynamics, E. Turkel 385

A New View of Nonlinear Water Waves: The Hilbert Spectrum, Norden

417
E. Huang, Zheng Shen, Steven R. Long

Planetary-Entry Gas Dynamics, Peter A. Gnoffo 459

VORTEX PARADIGM FOR ACCELERATED INHOMOGENEOUS

FLOWS: Visiometrics for the Rayleigh-Taylor and Richtmyer-Meshkov 495
Environments, Norman J. Zabusky

Collapse, Symmetry Breaking, and Hysteresis in Swirling Flows,

537
Vladimir Shtern, Fazle Hussain

Direct Numerical Simulation of Free-Surface and Interfacial Flow, Ruben

567
Scardovelli, Stéphane Zaleski