See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/225589907

Drag coefficients of variously shaped solid particles

Article in Theoretical Foundations of Chemical Engineering · June 2011

DOI: 10.1134/S0040579511020084

CITATIONS
READS
72
3,165

1 author:

Gudret Kelbaliyev
Azerbaijan National Academy of Sciences
45 PUBLICATIONS 488 CITATIONS

All content following this page was uploaded by Gudret Kelbaliyev on 04 December 2015.

The user has requested enhancement of the downloaded file.

ISSN 0040-5795, Theoretical Foundations of Chemical Engineering, 2011, Vol. 45, No. 3, pp. 248–266. © Pleiades Publishing, Ltd., 2011.
Original Russian Text © G.I. Kelbaliyev, 2011, published in Teoreticheskie Osnovy Khimicheskoi Tekhnologii, 2011, Vol. 45, No. 3, pp. 264–283.

Drag Coefficients of Variously Shaped Solid Particles,

Drops, and Bubbles
G. I. Kelbaliyev
Institute of Chemical Problems, Academy of Sciences of Azerbaijan, H. Javid ave. 29, Baku, AZ1143
Azerbaijan e-mail: Kelbaliev@yahoo.com
Received June 9, 2009; in final form, February 24, 2010

Abstract—This article, which is largely a review, deals with the drag force and drag coefficient for rigid
spher- ical and deformable particles in ordinary and non-Newtonian fluids. The most important theoretical
formu- las for small Reynolds numbers of Re 1 and semiempirical formulas for the drag coefficient in a
wide Re range up to 106 are presented. The deformation of drops and bubbles and its effect on the drag
coefficient are considered.
DOI: 10.1134/S0040579511020084

INTRODUCTION there are mass transfer processes in which the particle

One of the significant parameters governing the size varies with time in a wide range
migration of dispersed particles in a flow is the drag (agglomeration, coalescence, unsteady-state particle
force arising from the interaction between the particle growth or dimi- nution). As the drop grows in size,
surface and the flow past the particle. This force plays the flow past it begins to differ from the flow past a
the key role in mass, heat, and momentum transfer. solid particle because, along with the above
The drag coefficient of a spherical particle in the clas- physical phenomena, other effects come into play.
sical sense is given by the formula These are drop shape and surface pulsations (because
FS of the mobility of the interface and the uniform
C = , pressure distribution on it), internal liquid
( )
D
ρU 2
2 sM circulation, shape variations, fragmen- tation, etc.
C ∞ Because of this, various expressions have been
where U ∞ is the flow velocity far from the particle. suggested in the literature for calculating the drag
The drag FS depends on the shape of the surface past coefficient for solid particles, drops, and bubbles both
which the fluid moves, on the hydrodynamic for small Re values and for Re varying in a fairly wide
characteristics of the flow, on the size and properties range [1–13]. For gas bubbles and liquid drops, the
of the particles, and on the properties of the determination of the drag coefficient at large Re
medium. Numerous experimental studies aimed at values is complicated by shape distortion depending
determining the drag coefficient resulted in the on the Weber number We, Morton number Mo,
construction of the so- called standard Rayleigh Acrivos number Ac, and Bond number Bo. Table 1
curve for a single solid sphere moving with a constant presents the definitions of, and the relationships
velocity in a quiescent isother- mal liquid. For between, these numbers. In the general case, the drag
complex-shaped bodies, theoretical experienced by a liquid drop moving in a medium can
be expressed as
determination of the drag coefficient involves serious 2 + 3γ [1 + ζ(γ, Ac, Re)],
F =1 FS
difficulties and, in most cases, this quantity is esti- SK
31+γ
mated using experimental data and empirical correla-
tions. This is particularly true for the drag where FSK is the drag force acting on the drop. In view
coefficient of deformable particles (drops and of We = Ac Re2, we obtain ζ (γ, Ac, Re) =
bubbles) at large Re values with the corresponding
specific features of the flow (boundary layer ζ (γ, We, Re), although, as will be demonstrated
breakaway, hydrodynamic wake, drag crisis, etc.) and below, this equality is not valid in the general case.
for nonspherical single particles. In chemical The variation of the character of the CD (Re)
technology there are many processes involving solid dependence over a wide Re range is due to the hydro-
particles whose size changes in the flow from zero to dynamic changes in the flow past the particle. Assum-
some limit (crystallization, condensa- tion, ing that the drag coefficient in isothermal flow
swelling, granulation) and many processes in which depends on the flow velocity and viscosity and on the
particles can disappear entirely as a result of particle shape, introducing some equivalent
physicochemical transformations (dissolution, subli- diameter d, and using dimensional and similarity
mation, evaporation). Along with these processes, methods, we

248
 DRAG COEFFICIENTS OF VARIOUSLY SHAPED SOLID 2
Table 1. Similarity parameters and their interrelation
Number Definition Relation
gd3 Δρ Ar = BoAc–1
1 Archimedes Ar =
νc2 ρc Ar = MoAc − 3
gη4c Δρ
2 Morton Mo = 3
Mo = We3 Re−4 Fr−2 = BoWe2 Re−4
ρcσ ρc d d

2
3 Weber ρcU d We = Re 4/3 Mo1/3Fr −1/3 = Ac Re 2 We = Re 2 (MoBo) −1/2
We = σ d d d

gd 2
4 Bond or E@otv@os Bo = Δρ
ρ Bo = ArAc = WeFr−1

ρcνc2
5 Acrivos Ac = Ac = We Re−d2
σd
6 Froude 2 Δρ
Fr = U Fr = WeBo−1
dg ρc

obtain C = AU n d m ν k . With the dimensions of these

parameters extensive bibliography. Note that the Stokes solution
D taken into account, n = –k and m = –k.
In view of the arbitrariness of the coefficient A, the (2) is not the exact solution of the flow-past-particle
expression for the drag coefficient appears as problem with inertial terms. Calculations demon-
strated that, for distances larger than the particle
size,
the neglected terms carry a significant correction.
A
C = A Re−k = f (Re). (1) Ossen suggested linearizing the Navier–Stokes equa-
D
Re tion by replacing the convective term (U ∇)U with
As was demonstrated by theoretical calculations (U∞∇)U ∞,, where U ∞ is the velocity of unperturbed
and experimental studies, the drag coefficient for
spherical particles (solid particles, liquid drops, gas flow at infinity [4, 5, 14]. This leads to the following
bubbles) at Re 1 (Stokes flow) obeys the following expression:
equation [1, 8, 12]: 3
ζ ( Re ) = Re + ..... (4)
24 16
C = . (2)
D
Re An analysis carried out by Proudman and Pearson
For somewhat larger Re values, Eq. (1) can be rewrit- [4, 5] by the matched asymptotic expansions
ten as method led to the relationship
24
C = (1 + (3) ζ(Re) =
3
Re +
9
Re2 ln (5)
ζ(Re)), Re
.
D
Re 16 160 2
where ζ (Re) is, in most cases, an empirical function Quantitatively, expansion (5) is inappropriate because
of Re derived from experimental data. For drops the ln Re term makes the result dependent on the
and method of determining the Reynolds number. A
bubbles and large Reynolds numbers, Eq. (1) is repre- three-term
sented as CD A expansion was obtained by Chester and Breach [7]:
=
Re
f (Re, Mo, We, Ac).
The purpose of this work is to analyze the state
3
ζ (Re) = Re +
9
Re2 ln
Re
(+ A + )
16 160 2
of the art in the determination of the drag coefficients
+ O ⎛⎜ Re ln Re⎞⎟ ,
2
of rigid and deformable particles in ordinary and (6)
non- 4 2
Newtonian fluids. OF A SOLID SPHERICAL PARTICLE
Theoretical and experimental studies of the drag
DRAG COEFFICIENT coefficient for a solid spherical particle have been car-

THEORETICAL FOUNDATIONS OF CHEMICAL ENGINEERING Vol. No. 20

 2 KELBALIY
ried out by many researchers [1–6, 12, 13]. The ⎝ ⎠
com- plete list of the relevant works would make 5 323
A = γ1 + ln 2 − ≈ 0.835,
up an 3 360
where γ1 is the Euler constant, whose value is
0.577. Note that there is still no substantiated
method for improving asymptotic series with
logarithmic terms. Voloshchuk and Sedunov [14]
reported that the three- term expansion (6)
effectively refines Eq. (4) only at

THEORETICAL FOUNDATIONS OF CHEMICAL ENGINEERING Vol. 45 No. 3

 DRAG COEFFICIENTS OF VARIOUSLY SHAPED SOLID 2
Re ≤ 0.44, and augmented Eq. (6) to make it better up a monotonically increasing function satisfying the fol-
to Re ≤ 5. Thus, the solutions of the hydrodynamic lowing conditions [14]:
problem of flow past a spherical particle demonstrate
that the Stokes equation for the drag force is only the d (Re) 2
first term of the series in powers of Re. ζ ≤ 3, ζ (Re) ~ Re3 .
d Re 16
The above approximate theoretical solutions are
appropriate for very small Reynolds numbers of Levich [8] estimates the region of flow breakaway
from the surface of drops and bubbles as
Re 1, i.e., for laminar flow. For large Re values, the
1
theoretical solutions of the hydrodynamic problem are θ≈ , (7)
complicated by the appearance of convective terms in Re
the Navier–Stokes equation and by various that is, the flow almost does not break away from the
hydrody- namic phenomena (boundary layer bubble. Michaelides [6] reported the following empir-
breakaway, turbu- lence, wake formation, etc.). ical correlation for the wake of the flow past a solid
Because of this, numer- ous formulas fitted to the spherical particle:
experimental Rayleigh curve have been reported for
different Re ranges.
Wake formation is associated with the specific flow
θ≈ − 0.483
⎡ Re ⎤
pattern around the spherical body and with the 180 42.5
ln
( )
20 ⎥⎦
.
⎢⎣
appearance of convective forces. At Re ≈ 10, the Using experimental data of other authors [6, 9], we
iner- tial terms begin to distort the symmetric flow ◦
obtained, for the θ ≤ 180 range, the following simpler
pattern and the laminar boundary layer begins to break
away from the rear part of the sphere surface [4]; at formula with a correlation factor of r 2 = 0.995:
Re ≈ 20,
according to experimental data of Taneda [9], an
annular vortex forms behind the sphere. Van Dyke θ = 276
1/7
. (8)
[15] analyzed the liquid velocity field and discovered Re
that,
even at Re ≥ 8, there is a wake as an annular vortex It follows from (7) and (8) that the flow breakaway
behind the sphere. At Re ≈ 100, the flow at the rear angle for a drop or bubble is much smaller than for a
part of the sphere becomes unsteady and the system of solid particle. Owing to the mobility of the surface of
vortices spreads downstream over a distance approxi- the drop and bubble, the flow breakaway from the sur-
mately equal to the sphere diameter [4]. At Re ≈ 500, face occurs in the very small region in which flow tur-
the flow becomes irregular and a system of vortices bulization at large Re values takes place. Accordingly,
separates from the rear part of the sphere with a fre- the flow breakaway point for the drop or bubble, as
quency increasing with Re to from a wake. As Re distinct from the solid particle, is situated so low that
increases, annular vortices form and separate from the the turbulent wake covers only an insignificant part of
surface constantly, thus causing periodic changes in the surface. Thus, the total drag on a solid particle can
the flow velocity field and in the drug. At fairly be considered as the drag force acting on the surface
large Re values, part of the laminar boundary layer upstream of the flow breakaway point and having a
on the sphere surface before the breakaway line vis- cous character plus the drag force acting in the
becomes unsteady and turbulent. This turbulization break- away region. As was noted by Levich [8], the
of the boundary layer appreciably shifts the drag force is dissipative in character and can be
expressed in terms of energy dissipation, and it is
breakaway line downstream, the turbulent wake equal to the Stokes force at small Re values. This flow
behind the sphere pattern past the spherical body poses serious
narrows, and a drag crisis occurs at Re ≈ 3 × 10. This limitations on the description of the Rayleigh curve in
causes a dramatic decrease in the drag down to a wide Re range. An analysis of the existing formulas
CD ≈ 0.1. for calculating the drag coefficient for solid spherical
Note that, although the drag curve is differentiable particles demon- strated that, for Re ≤ 500, most of
and continuous, its derivatives are not monotonic the equations
functions in wide Re ranges. In particular, at Re reduce to the following forms:
values corresponding to the drag crisis, the derivatives
show a 24
C = 1 + A Ren ,
discontinuity of the first kind: as the discontinuity is
approached from the left Re → +3 × 105, C ≈ 0.51;
D
Re ( (9)

THEORETICAL FOUNDATIONS OF CHEMICAL ENGINEERING Vol. No. 20

 2 D
KELBALIY
where A = 0.17 and n = 2/3—Klyachko–Mazin for-
as it is approached from the right Re → −3 × 105, mula [10, 11]; A = 0.189 and n = 0.632—Prup-
CD ≈ 0.1. In drag coefficient calculations, the data pacher–Steinberger formula [16]; A = 0.15 and n =
pertaining to this region are smoothed to some extent. 0.687—Schiller–Neumann formula [1, 17, 18]; A =
For very small Re values, an analysis of calculated 0.1538 and n = 2/3—Kelbaliev–Ceylan formula [19];
and experimental data leads to the conclusion that ζ A = 0.1935, n = 0.6305, and 20 ≤ Re ≤ 260 —Clift–
(Re) is

THEORETICAL FOUNDATIONS OF CHEMICAL ENGINEERING Vol. 45 No. 3

 DRAG COEFFICIENTS OF VARIOUSLY SHAPED SOLID 2
Gauvin formula [20]; A = 3/16 and n = 2/3—Stuck- Table 2. Determining the particle drag coefficient in differ-
enbruck formula [6, 11]; ent Re ranges
24 n
C = 1 + A Rem Re A n Re A n
(10)
, ( <0.1 24 1.0 10–100 16.8 0.60
D
Re 0.1–1.0 26.9 0.95 100–800 6.1 0.38
where A = 0.0665, m = 2/3, n = 3/2, and 0.5 ≤ Red ≤ 1.0–10 26.5 0.80 800–1000 5.8 0.37
2 × 103 [3]; A = 0.27, m = 1, and n = 0.43 [3, 21]. At
Re ≤ 500, the Klyachko–Mazin and Schiller–Neu-
mann equations coincide within acceptably low accu-
racy: For wide Re ranges, the formulas for the drag coeffi-
24 cient of a spherical particle are more complicated and
CD = +
(11) cumbersome (Table 3). These formulas are valid for
ζ(Re), comparatively large Re values and allow the drug
Re coef-
where ζ (Re) = 6.48 Re−0.573 + + 0.36, 20 ≤ Re ≤ ficient to be calculated with a relative error of below
500
5–10%. Note that Eq. (17) was obtained by the
[3, 10], ζ (Re) = 0.03276 ×
× Re
−0.18−0.05 lg Re
, New- ton method [18], which enables one to
combine sepa-
6 rate piecewise approximations into a single formula
0.01 ≤ Re ≤ 20 [20], ζ (Re) = + 0.4,
1 + Re for a wide Re range. The Clift–Gauvin equation
1 < Re < 105 [6]; (15)
A (12) describes the entire drag curve for 0.01 ≤ Re ≤ 106 as
a piecewise approximation for particular Re ranges.
Equations (13), (14), and (17) presume the observance
n
,CD =
Re of the Newton law in the 104 ≤ Re ≤ 105, range, which
where the coefficients A and n take the values listed in means that the friction force is proportional to the
Table 2 [10]. This table can be supplemented with the square of flow velocity (CD ≈ 0.44). This leads to a sig-
following data: for Re = 04–7 × 104, A = 0.16 and n = nificant deviation of the calculated data from the
–0.102. A number of approximations for the drag actual experimental curve in this Re range. Equation
coefficients as Eq. (12) are presented in [1–6; 11, (21) fits the drag curve only in the drag crisis region.
19, 22–31]. In order to widen the Re range, the above As follows from the data presented in Table 3, use of
for- mulas were brought into the following forms: series or sums of functions is the only way to solve
the prob-
24 lem of approximating the drag curve for a spherical
CD = 1 + a Rem
Re + ψ(Re), (13)
( particle in a wide Re range. By using the drag coeffi-
cient in the form of Eq. (3), it is possible in most cases
24 to reduce the problem of fitting the drag curve to
C = (1 + k Re)n + ψ 1 (Re), (14) experimental data for spherical particles to
D
Re
−1 appropri- ately choosing a ζ(Re). function. However,
(
where (Re) = 0.42 1 + 42300 Re−1 ) , a = 0.15, it should be remembered that the highest
computational effi-
m = 0.687, Re < 2 × 105 [1, 6]; ciency is attained with formulas having the minimum
−
1
possible number of terms and parameters to be
(Re) = 0.42 (1 + 1.902 × 104 Re−1.18 ) , deter- mined.
A satisfactory fit to the standard drag curve in the
a = 0.241, m = 0.687, Re < 1.5 × 105 [11] 0.1 ≤ Re ≤ 106 range is provided by the following
series [31]:
;
(Re) = 0.407 (1 + 8710 C = ∞
(
A Rem(n) exp −b Rek(n) (23)
−1 )
Re−1 ) ,
∑
5
a = 0.15, m = 0.684, Re < 2 × 10 [22]; D n n
n=1

THEORETICAL FOUNDATIONS OF CHEMICAL ENGINEERING Vol. No. 20

2 − KELBALIY
1
(Re) = 0.413 (1 + 16300 Re−1.09 ) , Here, A, b, m, and k are the coefficients of the series
depending on the term number. This equation, con-
a = 0.173, m = 0.657, Re < 2 × 105 [25]; sisting of similar functions, is very flexible and can
−
1 describe the smallest changes in the drag over the
(Re) = 0.4251 (1 + 6880.9 Re−1 ) , widest Re range. In addition to the formulas listed
in Table 3, there are other correlations for the
a = 0.1806, m = 0.6459, Re < 2 × 104 [28]; spherical particle drag coefficient in particular Re
ranges [1, 6, 11, 30–34]. Here, we suggest one more
(Re) = 0.47 ⎡⎣1 − exp ( −0.04 Re0.38⎦)⎤ , equation for calculating the drag coefficient of a
solid spherical
k = 0.27, n = 0.43, Re < 2 × 105 [27]. particle over the 0.01 ≤ Re ≤ 106 range:

THEORETICAL FOUNDATIONS OF CHEMICAL ENGINEERING Vol. 45 No. 3

 DRAG COEFFICIENTS OF VARIOUSLY SHAPED SOLID 2
СD taking into account the compressibility of the liquid
104 was reported for the sphere drag coefficient:

103 CD = Re24 ×
⎡ ⎛
2 1 + 0.15Re0.687 ⎡ 0. 427 3 ⎤⎞⎤
10 ⎢ ( )⎜1 + exp ⎢ − 4.63 − 0.88 ⎥ ⎟⎥
×⎢ 1+
M⎡ ⎝
3.82 + 1.28 exp − ( )
⎣ M 1. 25 Re ⎤ ⎦ ⎠ , (26)
⎥
⎥
10 ⎣ Re M ⎥⎦
2 ⎦
1 ⎢⎣ Re < 3 × 10 .
5

1
3 At Re < 500, this equation coincides with Schiller and
Neumann’s formula. Experimental studies demon-
0.1 strated that an increase in M causes an increase in
0.01 0.1 10 102 103 104 105 106 Re the Re value at which the turbulization of the
boundary layer occurs.
1
Fig. 1. Drag coefficient of a solid spherical particle:
24
(1) CD ≈ 0.44;,(2) CD = ; (3) Schiller–Neumann equa-
Re DRAG COEFFICIENT
tion. FOR NONSPHERICAL PARTICLES
The spherical shape of a particle is an idealization,
2 and real particles deviate from the strict spherical
24 ⎡ ⎤
shape, appearing as an oval, ellipse, disc, cylinder,
CD = ⎢1 + Re3 ζ(Re)⎥ + or an irregularly shaped body. Irregularly shaped
Re ⎣ ⎦ parti-
cles are characterized by a sphericity factor, which is
the ratio of the surface area of the sphere having the
( (
+ 0.22 1 − exp −1.2 × 10 −24 Re 4
(24) same volume as the irregularly shaped particle (s) to
s
)) , Re2
ζ(Re) = + the actual surface area of the particle (sd): ϕ = . For
5 + 6.5Re2 sd
Re cubic particles, ϕ = 0.806; for cylindrical particles,
+ 2 10
. ϕ = 0.69; for a disc, −ϕ = 0.32. Numerous expres-
sions and experimental data have been reported for the
3 × 104 + 36 Re3 + 2.8 × 10−13 Re 3 drag coefficient of irregularly shaped particles [28,
For 0.01 ≤ Re < 800, Eq. (24) reduces to 36–41] (Table 4). Formula (31) in Table 4 is
applicable to a round disc with thickness H and
diameter d. the
L
drag coefficient of a cylindrical body with 5 < < 50
⎛ 8
⎞ d
CD ≈ 24 ⎜ 1+ Re 3
⎟ (25) and Re < 1 is given by the following expression[6]:
2⎟ 4
Re ⎜⎜ 5 + 6.5Re ⎟ C = . (32)
⎝ ⎠ D ⎡
Re ln
⎣⎢ d
L
()
− 0.1197⎤
⎥⎦
which is a modification of Eq. (9). Figure 1 plots The expression for the drag coefficient of a cylindrical
the body in lateral flow can be derived from experimental
Rayleigh curve and the data calculated via Eq. (24). Ud d
data [32] in the following form (Re = , is the
This equation provides a satisfactory fit with
cyl-
accept- able accuracy of 8–10% and a correlation νc
factor of inder diameter):
r 2 ≈ 0.9431. The error of approximation arises from CD
the low accuracy of the data in the drag crisis region.
Note that a considerable effect of the compressibil- 10 ⎛ Re1.875+ 0.368 × 10−3 Re2.55 ⎞
THEORETICAL FOUNDATIONS OF CHEMICAL ENGINEERING Vol. No. 20
 2 KELBALIY=
ity of the liquid is observed at high velocities of the ⎜1 + ⎟ (33)
Re0.778 60 + 6.8 Re1.15+ 0.4 × 10−14 Re3.95
flow past the particle, when the drag crisis takes place. ⎝
This effect is taken into account in terms of the
Mach
( )
+ 0.36 1 − exp(−1.2 × 10−24 Re4) .
U
number M = , where c is the speed of sound in the Figure 2 presents a comparison between the data cal-
c culated via Eq. (33) and experimental drag coefficient
given medium [35]. At M 1, the liquid can be data for a cylindrical particle for 0.1 ≤ Re ≤ 106. Equa-
con- tion (33) has the following particular solutions:
sidered to be incompressible. The following equation

THEORETICAL FOUNDATIONS OF CHEMICAL ENGINEERING Vol. 45 No. 3

 DRAG COEFFICIENTS OF VARIOUSLY SHAPED SOLID 2
Table 3. Drag coefficients of spherical particles over a wide Re range
@Но-
En- мер Refer-
try Expression фор- ence
мулы
⎧24 , Re < 0.01 ⎫
⎪
⎪ Re ⎪
⎪
⎪
⎪ Re
( 2 4) (
1 + 0.1315 Re 0.82 − 0.05 l o g Re , )
0.01 < Re ≤ 20 ⎪
⎪
⎪ 24
⎪ Re
( )(
1 + 0.1935 Re 0.6305 , ) 20 < Re ≤ 260
⎪
⎪
⎪ 2 ⎪
1.6425−1.1242 l o g Re + 0.1558 l o g Re
1 ⎪10 , 260 < Re ≤ 1500 ⎪ (15) [1]
⎪ −2.4571 + 2.5558 l o g Re − 0.9295 l o g Re + 0.1049 l o g Re 2 3
4⎪
C D = ⎨10 , 1500 < Re ≤ 1.2 × 10 ⎬
⎪ −1.9181 + 0.637 l o g Re − 0.0636 l o g Re 2
4 4 ⎪
⎪10 , 1.2 × 10 < Re ≤ 4.4 × 10 ⎪
⎪ −4.339 + 1.5809 l o g Re − 0.1546 l o g Re 2
4 5⎪

⎪ 10 , 4.4 × 10 < Re ≤ 3.38 × 10 ⎪

⎪ 29.78 − 5.3 l o g Re, 5 5 ⎪ 3.38 × 10 < Re ≤ 4 × 10
⎪ ⎪
⎪0.19 l o g Re − 0.49, 4 × 105 < Re ≤ 106 ⎪
⎪ 4 ⎪
⎪ 8 × 10 6 ⎪
0.19 − , Re > 10
⎩
⎪ ⎡ Re 1/10 10 10 ⎭
⎪
⎤ 10
⎛ 21 ⎞ ⎛ 4 ⎞
CD = ⎢
⎣(ϕ1 + ϕ2 )
1
−1
+ϕ
−1 + ϕ4 ⎥ , ϕ1 =
24
Re ( ) +
⎜Re
⎝
0.67
⎠
⎟ + 0.33
⎜Re
⎝ ⎠
10
⎟ + 0.4 ,
3 ⎦ 10
2 ⎛1 .57 × 108 ⎞ (16) [21]
1 1 6

ϕ2 = , ϕ3 = ⎜ ⎟ , ϕ4 = ,Re < 10
⎡ −10
⎤
1.625
× ⎡ 2.63 −10 − 10 ⎤
Re −17
(×10 )
−10
( )
0.11
0.148Re + 0.5 ⎝ ⎠ Re + 0.2
⎢⎣ ⎥⎦ ⎢
⎣ ⎦⎥
11 1/30 4/5
24 ⎡ ⎤
3 CD =
Re
⎢1 + 18.5 Re3.6+ Re
⎣
2 ( ) ⎥
⎦
+
4 Re
9330 + Re4/5
, 0.1 ≤ Re ≤ 5 × 104 (17) [19]

24(Re) 0.124
4 C = 10m , m (Re) = 0.261Re0.369− 0.105 Re0.431− , Re < 3 × 105 (18) [24]
D
Re 1 + l o g2 Re
2

5
C
D
=
0. 2841

Re
2
1+ (
9.04

Re
)
λ (Re) , λ (Re) = 0.96208 Re 2 + 2.73646 × 10 −5 Re 3 − 3.93861 × 10 −10 Re 4
(19) [26]
+ 2.47686 × 10−15 Re5 − 7.15934 × 10−21 Re6 + 7.43723 × 10−27 Re7, Re < 3 × 106
CD = 24 ⎡1 + 0.545 Re + 0.1Re b (1 − 0.03 Re)⎤ ,
1/2

6 Re ⎢ 1 + a Re ⎥ (20) [29]
⎣ ⎦
a = 0.09 + 0.077 exp(−0.4 Re), b = 0.4 + 0.77 exp (−0.04 Re), Re < 3 × 105
⎧28.12 − 5.3 l o g Re, 1.7 × 105 ≤ Re ≤ 2 × 105,⎫
C
⎪
= 0.1 log Re − 0.46, 2 × 105 < Re ≤ 5 × 105,
⎪
7 D ⎨ ⎬ (21) [11]
⎪ 4 −1 5 ⎪
⎩0.19 − 4 × 10 Re , Re > 5 × 10 ⎭

THEORETICAL FOUNDATIONS OF CHEMICAL ENGINEERING Vol. No. 20

2 KELBALIY
CD =
24
Re
( ) 3
A + B Re + C Re , A = 1, B = , m = 1, C = 0, Re < 1,
m n

16
8 A = 1, B = 0.1935, m = 0.6305, C = 0, 1 < Re < 285, (22) [30]
A = 1, B = 0.015, m = 1, C = 0.2283, n = 0.424, 285 < Re < 2000,
0.44
A = 0, B = , m = 1, C = 0, 2000 < Re < 3.5 × 105
24

THEORETICAL FOUNDATIONS OF CHEMICAL ENGINEERING Vol. 45 No. 3

 DRAG COEFFICIENTS OF VARIOUSLY SHAPED SOLID 2
Table 4. Drag coefficients of irregularly particles

@Но-
En- мер Refer-
try Expression фор- ence
мулы

30 67.289
1 CD = + , 0.2 ≤ ϕ ≤ 1.0, 0.1 ≤ Red ≤ 20 (27) [35, 36]
Re exp(5.03ϕ)
b3 Re
CD =
24
Re
(
1 + b1 Reb2 ) +b + Re
(
, 1b = exp 2.3288 − 6.4581ϕ + 2.488ϕ2 , b2 = 0.0964 + 0.5565ϕ, )
4
2 b3 = exp(4.905 − 13.8944ϕ + +18.4222ϕ2 − 10.2599ϕ3), (28) [28, 38]
b
4
(
= exp 1.4681 + 12.258ϕ − 20.7322ϕ2 + 15.8855ϕ3 . )
24 73.69 Re exp(−5.748ϕ)
C = [1 + 8.171exp(−4.0655ϕ)]Re0.0964 +0.5565ϕ+ ,
3 D
Re Re + 5.378 exp(6.2122ϕ) (29) [28]
0.7 ≤ ϕ < 1, 1 < Re ≤ 20
24
⎡1 + 0.118(K K Re)
0.6567
⎤+ 0.4305
C = ,
Re ⎣ ⎦
D 1 2
K1 K 3305
2 1+
4 K1K2 Re (30) [39]
1 2 −0.5741
K = + , logK 2 = 1.8148 (− l o g ϕ)
1
3 3 ϕ
64 Re Ude 64 ⎡ Re 2 Re2 Re ⎤

CD =
π Re
(1 + 2π ) , Re ≤ 0.01, Re =
ν
c
, CD =
πRe ⎣
⎢1 +
2π
+
5π
2
ln ( 2 ⎦)⎥ , Re < 1

5 CD =
64
π Re
(
1 + 10m , ) m = −0.883 + 0.906 l og Re− 0.025l og2 Re, (31) [1, 6]
1/3
0.01 < Re < 1.5, d e = 6d 2H( )
64
π Re
, CD =
1 + 0.138 Re0.792 , ( ) 1.5 ≤ Re < 133,
@H − толщина диска, d − диаметр диска

C ≈ 10 , Re ≤ 1 [32], DRAG COEFFICIENT

D
0.778 OF DROPS AND BUBBLES
10 ⎛ Re The motion of a liquid drop or gas bubble differs
3
1.875
⎞Re
CD ≈ 0.778 ⎜
1+ 1.156 ⎟
, Re ≤ 6.10 , (34) from the motion of a solid spherical particle of the
Re ⎝ 60 + 6.8 Re ⎠
same volume and mass. The liquid flowing past a drop
10 initiates, through friction, the circulation of the inter-
CD≈
0.778
(1 + 0.1076 Re ), 0.778
0.1 ≤ Re ≤ 6.103. nal liquid of the drop. This effect depends on the
R η
dimensionless parameter dγ,, which is the ratio of
These relationships for a cylindrical body can be ηc
approximated as the dynamic viscosity of the internal liquid to that
of the external liquid and characterizes the degree
CD ≈ 1.0, 2
6.10 ≤ Re ≤ 6.103 of mobility of the drop surface. In addition, as the
exter- nal liquid flows past the drop, forces arising
from pres- sure nonuniformity act on the drop
CD ≈ 1.1–1.4, 8.103 < Re < 2.105. surface to distort the spherical shape of the drop.
These forces are
As follows from Fig. 3, the drag crisis associated turbuliza-
with the shift of the boundary layer breakaway zone to tion takes place at Re ≥ 3 × 105.
the downflow part of the cylinder because of
THEORETICAL FOUNDATIONS OF CHEMICAL ENGINEERING Vol. No. 20
 2 KELBALIY
opposed by surface tension, which tends to
conserve the spherical shape. The presence of
adsorbed matter on the drop or bubble surface can
exert a significant effect on the surface forces, on
the drag coefficient, and on the drop (bubble)
settling or floating velocity.

THEORETICAL FOUNDATIONS OF CHEMICAL ENGINEERING Vol. 45 No. 3

 DRAG COEFFICIENTS OF VARIOUSLY SHAPED SOLID 2
Taking into account these specific features of flow
past liquid drops and using the matched asymptotic СD
expansions method in the determination of the drop 100
drag coefficient, Taylor and Acrivos carried out a the-
oretical analysis and arrived at the following
formula
for Red 1: 10

8
C DK =Re
⎡2 +
3γ Re 2 + 3γ 1 ⎛2 + 3γ⎞ 2 Re 2 ⎤
Re (35) 1
1
× ⎢ +
⎢ 1 + γ 16 1 + γ 40 ⎜ 1 + γ ⎟
⎝ ⎠ (2) ln
2⎥
. 2

+ ⎣

⎦
Chester and Breach [7] and Ockendon and Evans [43]
modified the formula for relatively large Reynolds 0.1
10–1 100 101 102 103 104 105 106 Re
numbers of Re < 6. For large Re values, numerical
Fig. 2. Drag coefficient of a cylinder in a lateral flow:
methods were suggested for solving the Navier– −0.78
Stokes equation (Re < 6) in view of the appearance of (1) CD ≈ 1.1; and (2) CD = 10Re .
convec- tive terms [44, 45]. There have been
theoretical and experimental studies, including
numerical solutions, on the drag coefficient, with ζ (Re) → 2,for Re > 0.4 and ζ(Re) → 3, for
deformation, and settling
(floating) velocity of drops and bubbles for moderate Re ≤ 0.01.. Figure 3 compares the drag coefficients of
and large Re values [1, 13, 46–53]. Table 5 presents bubbles calculated via formula (39) to experimental
some formulas for calculating the drag coefficients of data [54].
drops and bubbles. Unfortunately, most of the formu- Experimental data for the minimum in the drag
las for bubbles are valid only for small and moderate 1
Re and We values (Re < 50, We < 1). The exceptions
curve suggest the relationship Re Mo6 ≈ 7 [55], and,
are formulas (39), (45), and (46). Equations (39) and
(40) for the drag coefficient of gas bubbles was if Re Mo1/6 > 7, the drop will undergo deformation. In
derived from experimental data of Raymond and another work [52], this condition is expressed as
Rozant [54] by the Newton method. Equation (39) We Re0.85 > 165. It was reported [29] that the drop will
is valid for
Re ≤ 100, Mo ≤ 7 ; for Re ≤ 1.385, it reduces not be deformed if Mo < 1.2 × 10–7 Bo8.15 for Bo < 5 and
to
Eq. (40), which is independent of Mo. As was demon- if Mo < 0.2 × 10–7 Bo2.83 for Bo > 5. Based on the data
strated by numerous experimental data on the of Harmathy [56], Ceylan et al. [31] suggested the for-
behav- ior of drops and bubbles, their drag mula
coefficient and degree of distortion depend on Re, C
Mo, and We. In
particular, the Morton number depends on the prop- DK
≈ 6[1 − exp(−0.126Bo)]. (50)
erties of the medium and drops and takes the CD
following
values for different liquids at normal temperature: Here, CD is the drag coefficient of the solid particle.
mineral oil, Mo = 1.45 × 10−2; water + 42% glycerol, For Bo < 20, the following formula was obtained
Mo = 1.75 × 10−−48; water + 13% ethanol, using experimental data reported by Harmathy [56]:
Mo = 1.17 × 10 ; methanol, Mo = 0.89 × 10−10; dis- CDK
tilled water, Mo = 3.1 × 10–11. The second term in ≈ 1.275 Bo.
CD
Eq. (39) characterizes the dependence of the drag Based on formula (50), the steady-state value of the
coefficient of a bubble on its deformation, Re, and drag coefficient for drops and bubbles for large Re
Mo. As follows from Eq. (40), at Re < 0.4 spherical 4 8
drops and bubbles behave as solid particles and their
( )
drag coefficient is given by formula (2), although for
Re CD ≈ and Bo can be found as CD∞ ≈ 6CD = ..
If 9 3

THEORETICAL FOUNDATIONS OF CHEMICAL ENGINEERING Vol. No. 20

2 KELBALIY
0.4 ≤ Re ≤ 1.385 the drag coefficients for a solid spherical particle and
– CD = 16 . Thus, the following for a gas bubble are known, the drag coefficient for a
Re liquid drop can be determined as follows [3, 11]:
formula can be suggested for the 0.01 ≤ Re < 0.4
range [19]:
CDK = γCD + CDG ,
1+γ
⎡ where CDK, CD, and CDG are the drag coefficients of the
8 8 ⎤ 1 ⎥ (49)
CDK = ζ(Re) = ⎢1
+ (
1 − 0.5 1 + 250 Re )
5
⎥ drop, solid particle, and gas bubble, respectively. This
formula is precise at Re 1 and, as was demonstrated
Re Re ⎣⎢ ,
⎦

THEORETICAL FOUNDATIONS OF CHEMICAL ENGINEERING Vol. 45 No. 3

 DRAG COEFFICIENTS OF VARIOUSLY SHAPED SOLID 2
Table 5. Drop and bubble drag coefficients

@Но-
En- мер Refer-
try Expression фор- ence
мулы
24 ⎛ 2 3 + γ ⎞
1 C DK = ⎜ ⎟ , Re 1 (36) [8]
Re 1 + γ
⎝ ⎠
⎧ 1.68 , 0.1 ≤ Re ≤ 0.5⎫
1/4
⎪Re ⎪
2 CDK =⎨ ⎬ (37) [1]
⎪ 14.9 , 1 ≤ Re ≤ 10 ⎪
⎪⎩Re 0.78 ⎪⎭
CDG = 48 ⎛⎜1 + 12Mo 1/3⎞ ⎟ + 0.9 Mo3/2
1/3

3 1/3 3/2 (38) [46]

Re⎝1 + 36Mo
1/55
⎠ (
1.4 1 + 30Mo ) + Bo
⎧16ζ (Re) ⎡ Re 12
⎤ 8
4/3
Re Mo
1/3 ⎫
(39)

4 ⎪
⎪
0
⎪ Re ⎢⎣1 + 1.385 ⎦⎥ + 3
24 (( )
1 + Mo1/3
)
−1/3
+ Re 4/3 Mo1/3
, Re < 100 ⎪
⎪
⎪ [19]
CDG =⎨ ⎬
⎪16ζ0 (Re) ⎛ Re 12 ⎞ 1/55 1 ⎪
⎪ Re ⎝
⎜1 + ( ) ⎠⎟
1.385
, Re < 10, ζ 0 (Re) = 1 +
5 −2 ⎪
(40)

⎩⎪ (
1 − 0.5 1 + 250 Re ) ⎪⎭
1 ⎡ ⎛ 24 4 ⎞ 14.9 ⎤
5 C = γ + + , 0.1 ≤ Re < 10 (41) [47]
⎜ 1/3 ⎟
1+ γ⎢ 0.78 ⎥
DK
⎣ ⎝ Re Re ⎠ Re ⎦
48 ⎡ H (χ)⎤
6 C = G (χ) + , Re < 20 (42) [48]
DK ⎢ 1 1/2 ⎥
Re ⎣ Re ⎦
8 3γ + 2 ⎡ 3λ + 2 ⎤ 3γ + 2
7 C = 1 + 0.05 Re − 0.01 Re ln Re, 1 < Re < 5, 0<γ<∞ (43) [51]
DK
Re 1 + γ ⎢⎣ 1+λ ⎥⎦ 1+γ

8 C
DK
( 2
= 1.87 783γ + 2142γ + 1080 Re −0.74 [(60 + 20γ) ( 4 + 3γ)] ) −1
, 2 < Re < 50 (44) [11]
⎧⎡ ⎛ 24 4 ⎞ 14.9 ⎤ 3γ + 2 ⎫ 2
−1

9 C DK = ⎨⎢γ ⎜ + 1/3 ⎟ + 0.78 ⎥ + 40 ( ) + 15γ + 10⎬ ⎡⎣(1 + γ ) 5 + Re ⎤⎦ , ( ) (45) [51]

Re £ 400
⎩⎣ ⎝Re Re ⎠ Re ⎦ Re ⎭
⎧16 , Re < 1.5 ⎫
⎪
⎪Re ⎪
⎪
14.9
, 1.5 < Re ≤ 80
10 ⎪ Re0.78 ⎪ (46) [52]
C DK = ⎪
⎨49.9 2.21 ⎪
⎬
( )
−8
1− + 1.17 × 10 Re 2.615
, 80 < Re ≤ 1
⎪ 1/2
530⎪
⎪ Re Re ⎪
⎪⎩2.61, Re > 1530 ⎪⎭
26.5 ⎡(1.3 + γ) − 0.5⎤
2
11 (47) [53]
CDK = 0.78 ⎢ ⎥ , Re ≤ 5, γ ≤ 1.4
Re ⎣(1.3+ γ) (2 + γ)⎦

{ }
−1
16 ⎡8 1 −0.5 ⎤
12 (48) [47]

THEORETICAL FOUNDATIONS OF CHEMICAL ENGINEERING Vol. No. 20

2 KELBALIY
CDK =
Re
1+
⎢
⎣Re
+
2
(1 + 3.31Re )⎥⎦ , Re < 50

THEORETICAL FOUNDATIONS OF CHEMICAL ENGINEERING Vol. 45 No. 3

 DRAG COEFFICIENTS OF VARIOUSLY SHAPED SOLID 2
by calculations, is accurate to within 5% at
Re ≤ 100.The correlations between the drag coeffi- СD
cients of the solid spherical particle, drop, and bubble + абвгде
for 5 < Re < 1000 are presented below [6]: 100
2−γ
C (Re, γ) = C 4γ 3
D
(Re, 0) + C (Re, 2), 10
D
2 6D + γ

0≤γ<2 1
4 γ−2 1 3
CD (Re, γ) = (Re, 2) + C (Re, ∞),
γ+2 D 2
CD
2+γ
2 ≤ γ ≤ ∞,
0.1
where CD (Re, 0), CD (Re, 2), and CD (Re, ∞) are 1 10 100 1000 Re
the
drag coefficients for a bubble, a drop with γ = 2, and a Fig. 3. Bubble drag coefficient as a function of Re for Mo
solid particle, respectively. The drag coefficient is sig-
= (a) 7.0, (b) 1.4, (c) 0.023, (d) 9× 10–4, and (e) 1.1 × 10–
nificantly affected by variations of the shape of 5
drops and bubbles as a result of their deformation. .
It was found that the behaviors of drops and
bubbles are qualitatively the same and differ A still simpler formula was reported by Raymond
markedly from the behavior of solid particles. The and Rozant [54]:
shape of a drop or bub- ble is not fixed. It forms as
the drop or bubble moves and is determined by the We
current balance of the pres- sure force acting on the . χ≈ 1 −
surface of the deformable par- ticle from the side of 9
the medium, which tends to com- press the particle in Taylor and Acrivos [42], assuming that the radius of
the motion direction, and the sur- face tension, which the deformed drop is r = R [1 + ζ(cos θ)], theoretically
opposes this compression. The pressure force is obtained the following expansion by asymptotic
proportional to the velocity head, matching of the external and internal expansions for
ρU 2 the function ζ(cos θ) :
FD ~ c ∞ ,, and the surface tension is proportional to
2
FD ζ ( cos θ) = Ac Re2 P (cos θ)
the capillary pressure, = 2dσ . If We ~ We F 1 at 2 3
2

Fσ ~ λV
3 11 + 10γ
r σ − λV Ac Re P3 (cos θ) + ...,
Re 1, the drops and bubbles will be strictly spherical. 70 1+γ
When FD ≥ Fσ, the drop surface loses stability and the
drop undergoes deformation, taking on the shape of an
oblate ellipsoid of revolution and, with a further
λ =
V
4 (1 +
1
(8081 +2057 γ + 40103 γ +43 γ )
2 3

(51)
increase in Re and We, various configurations, includ- 3 2
γ)
ing that of a thread. Here, the deformation factor is γ Δρ 1
– 2,
defined as the ratio of the minor semiaxis of the 12 ρC (1 + γ)
ellipse a0
(a ) to the major semiaxis (b ,): . Mypa [57] sug-
χ= 3 + 2γ
0 0
b0 FSK = 1 FS
gested the following estimator for the deformation fac- 3 1+γ
tor at large Re values: ⎡ 3 + 2γ ⎛3 + 2γ⎞ ⎤
1 1 2 2
× ⎢1 + Re + ⎜ ⎟ Re ln Re + ... ⎥,
1/2 ⎡ We = χ4χ−−4/3 χ2( )
χ3 −+ 1χ − 2 2
−3
⎣⎢ 8 1 + γ 40 ⎝ 1 + γ ⎠ ⎥⎦
× ⎤χ2arcsec ( ) χ −1 . where P (cos θ) and P (cos θ) are second- and third-
⎣⎢
( ⎥⎦ ( ) 2 3
)
order Legendre polynomials. For gas bubbles in a liq-

THEORETICAL FOUNDATIONS OF CHEMICAL ENGINEERING Vol. No. 20

 2 KELBALIY 5
Wellek et al. [58] reported the empirical formula uid medium, γ → 0 and λV → . Thus, drop defor-
b 48
Y = 0 = 1 + 0.09 mation exerts no effect on the drag up to the terms of
a0 .
We
0.95 the order of Re ln Re, inclusive. Unfortunately, lack of
Taylor and Acrivos [42] suggested the following the necessary physical data concerning the drag on
for- mula for estimating the deformation of drops: drops at medium Re values encumbers the in-depth
analysis of this expansion.
χ = 1 − 9 We. According to Sherman [59], the variation of the
16 shape of a drop at small deformations is described
by

THEORETICAL FOUNDATIONS OF CHEMICAL ENGINEERING Vol. 45 No. 3

 DRAG COEFFICIENTS OF VARIOUSLY SHAPED SOLID 2
π
where β = κS. Experimental data of Raymond and
1.0 24
χ −4
+ а b c d e Rozant [54] suggest that β = 1.475 × 10 Mo
−1/5
and
+ + + 1
0.8 + 1
χS = Mo . Figure 4 compares the experimental and
++ + 8
+ 2
+
0.6 + calculated gas bubble deformation factors for different
+
+ + Mo values. As Mo decreases, the deformation factor
falls sharply and, accordingly, the shape of the bubble
0.4 (drop) changes significantly.
A somewhat different approximation for the defor-
mation factor can be derived from Eq. (51) by
retain-
0.2 ing only its first term and assuming that We = Ac Re2 d
(
and P (cos θ) = 0.5 3 cos2 θ − 1 :
2
)
λ
0 2 4 6 8 10 12 d,
mm
( )
r = R (1 − ζ ( cos θ)) = R ⎡1 − V We 3 cos2 θ − 1 ⎤.
(54)
Fig. 4. Bubble deformation factor as a function of the ⎣⎢ 2 ⎥⎦
equivalent size for Mo = (a) 7.0, (b) 1.4, (c) 0.023,
(d) × 10–4, and (e) 1.1 × 10–5. For θ = 0 and r = a0, Eq. (54) reduces to
r = a0 = R (1 − λV We).
the following relationship (n the notation used in π λ
this work): 2 (
For θ = , we have r = b = R 1 + V We and the

⎜2 ( K )
0 2
1
⎟1 deformation factor will appear as
1+χ Δρ 16γ + 16 χ= 0= V
. (55)
1 − χ Gd g 2 ⎛ ρc ⎞ 4 19γ + 16 a 1 − λ We
= Mo , b0 1 + λV We
⎝ ⎠
2
1
⎛ σ ⎞2
where K = is the capillary constant and G is the Using experimental data of Raymond and Rozant
⎝⎜ d ⎠⎟ [54], λV in the first approximation can be represented
shear modulus. as follows [19, 60]:
In terms of elasticity theory, drop deformation is
described by the following equation [19, 60]: 1 − 3 We = 1 − 3
1 Δχ
= −κSΔd, (52)
λ V = 12
( 1 25 Re ) (12 1 25
There has been a comparison between the observed
υχ − χS and calculated (Eq. (55)) bubble deformation
factors for various We and Re values. It follows from
these data
where υ is the volume of a drop, κS is the elasticity that the rate of decrease of the deformation factor of a
coefficient, and χS is the steady-state value of the drop increases with increasing We and Re. As Mo
deformation factor. By passing to the Δd → 0, limit, increases, the deformation factor increases progres-
we obtain sively more rapidly even at small drop sizes.
Similar experimental data on the drag coefficient
Δχ ∂χ and defor- mation of air bubbles in various liquid
lim = = −κS (χ − χS ) υ.
media were pre- sented by Haberman and Morton
Δd→0 Δd ∂d
3 [61].
πd The deformation of drops and bubbles exerts a sig-
For spherical drops, assuming that ϑ = ,, we can
d nificant effect on their drag coefficient. Under the
6 assumption that the volume of the liquid in a drop is
bring the above equation to
shape-independent,
4 the volume of an ellipsoidal
a0 drop
3 is υ = πa2b . Taking into account that and that
dχ χ=
= − πd κ ( χ − χ ) 0 0
b0
3 4
dd 6 S S
(53) the volume of a spherical drop is υ = πR3, we obtain

THEORETICAL FOUNDATIONS OF CHEMICAL ENGINEERING Vol. No. 20

2 KELBALIY
χ ( d → 0) = 1. 3
1
The solution of Eq. (52) will then appear as follows ⎛⎜ χυ ⎟
⎞3
[19, 60]: 1
2
4 a0 =⎜⎜ π⎟ ⎟ = Rχ and b0 = 4 πa
3 υ 2 = Rχ −3. The drag
4
(
0
χ = χS + (1 − χS ) exp −βd ⎝3 ⎠ 3
),

THEORETICAL FOUNDATIONS OF CHEMICAL ENGINEERING Vol. 45 No. 3

 DRAG COEFFICIENTS OF VARIOUSLY SHAPED SOLID 2
coefficient of an ellipsoidal drop in a lateral flow for Y
small and moderate Re values will then appear as
7 аbc
R
CDE ≈ CDC 1
2
= χ CDC,
3
(56) 6
a0b0
5
This expression interrelates the drag coefficients of an
ellipsoidal particle ((CDE )) and a spherical particle 4
(CDC ). The variation of the shape of drops and bubbles
and their deformation at a constant volume is associ- 3
ated with the variation of their surface area, 2
8 πa b
S = 1 +0 χ0 . (57) 1
Equation (57) describes the dependence of the surface 0 0.05 0.10 0.15 0.20 0.25 0.30 d, mm
area of a deformed drop or bubble on the deformation Fig. 5. Drop deformation factor for γ = (a) 1.0, (b) 3.0, and
factor. It follows from this equation that, with increas- (c) 5.0.
ing deformation (decreasing deformation factor),
the surface area of the drop increases. When there
is no the liquid shows pseudoplastic properties; at n > 1, the
liquid is dilatant. This fact imposes special constraints
deformation, χ = 1, a = b = R, and S = 4πR2 ; that is, on the hydrodynamic description of non-Newtonian
0
the drop is spherical and its surface area is equal to the fluids and modifies the description of the drag coeffi-
surface area of the sphere. The deformation of cient in these liquids.
drops depends mainly on the capillary number Ca
and vis- cosity ratio γ. There has been an Theoretical and experimental hydrodynamic stud-
experimental study of drop deformation at γ= 1.0, ies and drag coefficient data for solid particles, drops,
3.0, and 5.0 and various Re values [62], with the and bubbles in non-Newtonian fluids are presented in
deformation factor defined as the ratio of the length [3, 51, 67–75]. The following formula was obtained
of the deformed drop (L) to the thickness of the by asymptotic methods for the drag coefficient of a
drop: Y = L/a. The following drop deformation bub- ble in a non-Newtonian fluid (n < 1) for small
equation was derived from experimental data in that Rey- nolds numbers of Re < 1 [3, 76]:
study:
1
Y = ,
s s
(
Y + (1 − Y ) exp −β d
0
)
2
(58) 16
CD = X (n),
1 1 2n Re (59)
n−3
where Ys is the steady-state value: Ys = γ2, and 13 + 4n − 8n 2
7 X (n) = 2n − 13 (2n + 1) (n + 2),
1
−
2
β = 1.08 × 10−3 Re2 γ 4. Figure 5 presents a compari- where X (n) is the exponent n dependent parameter
son0 between the calculated and experimental drag
coefficients of a drop [62] for different γ values. characterizing the rheological properties of the
The deformation of drops and bubbles is associated flow. Calculations using formula (59) demonstrate
with changes in their shape and is accompanied by that, for pseudoplastic and dilatant liquids, the drag
their breakup into smaller particles [60, 63]. The coefficient is, respectively, larger and smaller than the
changes in the particle size have a significant effect on drag coeffi- cient of a bubble in non-Newtonian
the drag coefficient. At the same time, the flow [3]. In addi- tion to Eq. (59), there have been
minimum and maximum sizes of drops and bubbles experimental X data and X (n) dependences reported
depend on the drag coefficient. The effects of by Chabra [51]:
external factors (physical phenomena on the surface)
n−3
on the drag coef- ficient of drops and bubbles are
considered in other X (n) = 2n3 2 [1 − 3.83(n − 1)], 0.7 ≤ n ≤ 1;
works [64–66]. n−3 2
2 13 + 4n − 8n
X (n ) = (33γ2 ) 2 ,
(2n + 1) (n + 2)
γ > 10;
DRAG COEFFICIENT OF PARTICLES n+3 2 (60)
IN A NON-NEWTONIAN FLUID X n − 2 ⎡2 (2n + 1) (2 − n)⎤
Non-Newtonian fluids differ from ordinary liquids
THEORETICAL FOUNDATIONS OF CHEMICAL ENGINEERING Vol. No. 20
2 KELBALIY
( ) =3 ⎢⎣ 2 ;
in that their viscosity n
changes with an increasing shear ⎦⎥ n−3 2
rate. The extent to which a liquid is non-Newtonian is 1 + 7n − 5n
characterized by the shear rate exponent n. At n < 1, X (n) = 2n3 , n < 1.
2 n (n + 2)

THEORETICAL FOUNDATIONS OF CHEMICAL ENGINEERING Vol. 45 No. 3

 DRAG COEFFICIENTS OF VARIOUSLY SHAPED SOLID 2
X(n) a number of formulas characterizing the deformation
of bubbles in a non-Newtonian fluid:
1.8
b
Y = 0 = 0.0628δ0.46, 20 ≤ δ ≤ 100,
K
a0
1.2 YK = 1.4, δ ≤ 4,
Y = 6.17δ−1.07, 4 ≤ δ ≤ 20,
δ = ReK Mo0.078, Mo = Wen + 2Fr 2 − 3n Re−4,
t t
0.6 0.64 ≤ n ≤ 0.9
Table 6 presents a variety of formulas for calculat-
ing the drag coefficient of solid spherical particles,
drops, and bubbles in non-Newtonian fluids. Formu-
0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 n las (62), (65), and (67) in Table 6 refer to the drag
coefficient of a solid spherical particle; formulas (63),
Fig. 6. X (n) as a function of n (64) , and (66), to irregularly shaped particles with dif-
ferent shape factors; formula (68), to gas bubbles.
According to Chabra [51], the drag coefficient of a
bubble in a non-Newtonian fluid in the 5 ≤ Re ≤ 25
K(n) range is given by the formula
50
K (n)
45 CDG = .
Re t
45 Here, K (n) is an n-dependent coefficient. Based on
45
experimental data of Chabra [51], we suggest the fol-
lowing formula:
25
exp(2.53n3 )
K (n) = 30.6 ⎛ 6 ⎞ .
25
⎜ n + 1⎟
7

⎝ ⎠
0.2 0.4 0.6 0.8 1.0 n As is clear from Fig. 7, the coefficient K (n) varies
sig- nificantly in the 0 ≤ n ≤ 1 range and, passing
through a
Fig. 7. K as a function of n. minimum at n = 1, tends to K (n) ≈ 48,, which
charac- terizes the drag coefficient of a bubble in a
Newtonian fluid suggested by Levich [8] for small Re
The second formula in Eq. (60) characterizes the values.
behavior of drops in a non-Newtonian fluid. These By analogy with Eq. (24), we suggest another
theoretical expressions do no not provide a general equa- tion for the drag coefficient of solid particles in
solution for the problem. a non- Newtonian fluid: 2
24X (n, Re ) ⎡ ⎤
A satisfactory approximation for X (n) over the CD = t
⎢ 1 + Re3 ξ( Ret )⎥
t
0.1 ≤ n ≤ 1.8 range for solid particles in a non-Newto- Ret ⎣ ⎦
nian fluid is provided by the following relationship
based on experimental data of Chabra and Richardson ( (
+ 0.3 1 − exp −0.2 × 10 −19 Re 4 t , ))
[69]:
⎡ 3
5⋅6
n −1 1
Re2 ⎤
⎥
⎢
3 X (n, Ret ) = 1 + 4n ⎢n1)+3 (1 − t1 3⎥
,
7 ⎡ 3⎛ ⎞⎤ ⎢ ⎥
Re2 1 + (1 − n)3 (70)
(5n + 9) exp ⎢−n2 ⎜1 +2n
2
⎟⎥(.n) =
X (61) ⎣ t ⎦
450 ⎢ ⎜ 10 ⎟⎥ 0.8 < n ≤ 1, 0.1 ≤ ≤ 105,
Re
t
⎣ ⎝ ⎠⎦
ξ(Ret ) = Re2
t 2 +
THEORETICAL FOUNDATIONS OF CHEMICAL ENGINEERING Vol. No. 20
 2 KELBALIY
It follows from (61) that, at X (n) ≈ 1. Figure 6 illus- 5 + 8Ret
trates the fit between the calculated and
experimental Ret
+ 5 2 10
.
6 − −13
X (n) data. Chabra and Richardson [69] also reported 6.8 × 10 Re 4+ 32 Re3t+ 952.8 ×10 Re 3

THEORETICAL FOUNDATIONS OF CHEMICAL ENGINEERING Vol. 45 No. 3

 DRAG COEFFICIENTS OF VARIOUSLY SHAPED SOLID 2
Table 6. Drag coefficients of particles in non-Newtonian fluids

@Номер
En-
Expression фор- Refer-
try ence
мулы

24 0. 4305k2 −1/2
⎣⎡1 + 0.1118 ( k Re ) ⎦⎤ +
0.654
C = , Re = k Re, k −1 = 0.33 + 0.67ϕ ,
D 2 t t 1 1
1 Re t 1 + 3305 (62) [47]
k2 Ret
0.574 −3 5
l o g k2 = 1.815 (− l o g ϕ) , 10 ≤ Red ≤ 10 , 0.1 < ϕ < 1
24 0.413
2 C
D
=
Re t
(
1 + 0.173 Re0.657
t ) + 1 + 16300 Re −1.09
, Re t< 1600, 0.16 ≤ n ≤ 1 (63) [64]
t

3 C
D
=
24
Re t
(
1 + 0.24 Ret 0.5 ) + 1 +1 +0.0000238
370 Re
Re
, −1
t
Re t < 2.5 × 104 (64) [67]
t

24 73. 69 Re t exp ( −5. 075ϕ)

⎣⎡1 + 8.1723 exp ( −4.066ϕ) t Re ⎦⎤ +
0.0964 + 0.557ϕ
4 C = (65) [38]
D
Re t Re t + 5.378 exp(6.212ϕ)

⎧35.2 (2n) ⎡ 20.9 (2n)⎤ ⎫

⎪ + n ⎢1 − ⎥, 0.2 2n ( ) ≤ Re ≤ 24 2n
t
( ) 0.38 ≤ n ≤ 1⎪
1.03 1.11
5 ⎪ Ret ⎢⎣ Re ⎥⎦ ⎪ (66) [68]
CD = ⎪
⎨37 2n ( ) t
⎬
⎪
⎪ 1.1 + 0.25 + 0.36n, 24 2n ( ) ≤ Re t
≤ 100 2n ( ) ⎪
⎩ Ret ⎭
32.5
6 CD = (1 + 2.5 Ret ), 0.2
Re t < 150, 0.77 ≤ n < 1, 0.35 ≤ ϕ < 0.7 (67) [69]
Re t
16 0.431
7 CD =
Re t
(
1 + 0.173 Ret0.657 ) + 1 + 16300 Re −1.09
, Re t < 1000, 0.16 ≤ n < 1 (68) [72]
t

24 n2 − n + 3
CD = X *(n), X *(n) = X (n) = 32n − 3
, Re t < 10−5
Re 33n
t
2
1− n
8 X *(n) = X (n) + (
lg 103 Ret , ) 10−5 ≤ Ret < 10−3 (69) [70]
3n + 1
4n4
X *(n) = X (n) + , 10−3 ≤ Re < 103, 0.56 ≤ n ≤ 0.89
n−3 t
24 Ret 3

For Ret ≤ 1000, this equation reduces to EFFECTS OF VARIOUS PHENOMENA

2 ON THE DRAG COEFFICIENT OF PARTICLES
24 X ( n, Re t ) ⎛ 3
−5
⎞
Apart from the above parameters, the following
C ≈ ⎜1 + 0.125Re + 2.51 × 10 Re
2

⎟.
t t
D Ret ⎝ factors produce a significant effect on the friction
⎠
forces and the drag coefficient of particles: flow turbu-
Figure 8 illustrates the fit between the drag coefficient smaller than in Newtonian fluids.
data calculated via Eq. (70) and experimental data
[72–75]. It follows from Fig. 8 and calculations using
Eq. (70) that, as n decreases, the drag coefficient of
solid particles in viscoplastic liquids become

THEORETICAL FOUNDATIONS OF CHEMICAL ENGINEERING Vol. No. 20

 2 KELBALIY
lence, roughness of the particle surface, bulk
porosity of the particles, and mass transfer
processes on the drop and bubble surface
(dissolution of gas bubbles in the liquid,
evaporation of drops, etc.). These factors cause
unsteady changes in the particle size and many
other features.

THEORETICAL FOUNDATIONS OF CHEMICAL ENGINEERING Vol. 45 No. 3

 DRAG COEFFICIENTS OF VARIOUSLY SHAPED SOLID 2
СD particle. This can markedly change the friction forces
103 and drag coefficient. Particles freely moving in the
абв
+г д flow can spin under the action of the Magnus force,
change their orientation, and execute complicated
102 nonlinear motion. Depending on its size, the
particle can be entrained by a pulsating turbulent flow
to some extent, and the Reynolds number in this
10 case is esti-
U − Vd d
mated as Re = . The smaller the particle, the
νc
1 larger the extent to which it is entrained by the
turbu- lent pulses; that is, the extent of entrainment
of parti- cles tends to unity as the particle size
increases. For drops and bubbles, an increase in Re
raises the proba-
0.1
0.1 1 10 102 103 104 105 Re bility of their ellipsoidal distortion (We ~ Re 2 and

)
their breakup frequency [60, 63]. Polyanin et al. [11]
Fig. 8. The drag coefficient of a solid particle in a non- suggested formulas for calculating the drag coefficient
Newtonian fluid as a function of Re according to the
data of different authors: (a) n = 0.84–0.86 [64], (b) n = of ellipsoidal particles in lateral and longitudinal
0.75– 0.90 [69], (c) n = 0.75–0.92 [69], (d) n = 0.56– flows. Tropea et al. [79] reported experimental data on
0.75 [70], the effect of surface roughness on the drag crisis. As
and (e) n = 0.73 [71]. The dashed line represents the drag the roughness of the particle surface increases, the
coefficient of the particle in a Newtonian fluid.
laminar boundary layer is turbulized progressively
more rapidly and the drag crisis occurs at smaller Re
There has been a qualitative analysis of the effect values. Various expressions taking into account
of the turbulence of the external hydrodynamic field surface roughness have been suggested for estimating
on the drag on a solid spherical particle in the case of the drag and turbulent dif- fusion coefficients of
the particle size being of the same order as the particles, the particle settling veloc- ity, and heat
internal scale of turbulence or larger [4, 18, 77–79]. transfer [77, 78, 80, 81].
At low tur- bulence intensity in the external flow and
small Re val- ues, turbulence slightly raises the drag The drag force on a permeable sphere is equivalent
by enhancing the dissipation in the wake. When the to the drag force of an impermeable sphere with a
particle is in tur- −1
bulent ⎛ 2k ⎞ [4], where k is
[18]. Itflow, the drag
was noted [79]crisis
that,occurs
as the at smaller Re
turbulence values
intensity smaller diameter of d = d × × 2⎟
+
increases from 0.5 to 2.5%, the critical Re value k ⎝ d⎠
decreases from2.7 × 105 to 1.25 × 105.. At a turbulence the permeability coefficient. Expressions for the drag
intensity of 30% and above, the drag crisis can occur coefficient of porous cylindrical particles were pre-
at much smaller Re values. The effect of turbulence on sented by Tropea et al. [79].
the drag coefficient of solid particles is estimated as A marked effect on the drag on drops is exerted by
FTP mass transfer processes, including evaporation from
C ( Re ) = . the drop surface, which reduces the drag coefficient:
D
πd −0.84
2
ρd CDM = 27 Re , Re < 1. The following formula was
( U )2 suggested for the drag coefficient of a spherical drop
8 with an evaporating surface [65, 66]:
With isotropic turbulence taken into account [78], the
drag coefficient of solid particles appears as
3
24 1 + 0.545Re + 0.1Re0.5 (1 − 0.03Re)
⎡ −4 ⎛ ⎞⎤ CDK = ,
CDT = CD ⎢1 + 8.76 × 10 λ d Re 1 + A ReBP
,
⎝ ⎜0 ⎠ ⎥⎟⎦⎥ 10 < Re < 200, 1 < ReP < 20, (71)
where λ0 is the Kolmogorov scale of turbulence and flow.
CDT is the drag coefficient of the particle in turbulent

THEORETICAL FOUNDATIONS OF CHEMICAL ENGINEERING Vol. No. 20

 2 KELBALIY
A = 0.09 + 0.077 exp(−0.4 Re), B = 0.4 + 0.77 exp(−0.04 Re),
Note the difference between a fixed particle in a VPd
flowing medium and a particle moving in the flow. where ReP = is the Reynolds number defined in
For the fixed particle, the flow pattern around it may νc
depend on the degree of turbulence and high-intensity terms of the evaporation rate and VP is the rate of evap-
turbulence can alter the structure of the boundary oration from the drop surface. With the Stefan flow
layer and break down the turbulent wake behind the taken into account, the drag coefficient of an evapo-

THEORETICAL FOUNDATIONS OF CHEMICAL ENGINEERING Vol. 45 No. 3

 DRAG COEFFICIENTS OF VARIOUSLY SHAPED SOLID 2
rating spherical drop for small Red values appears as
follows [51, 70]: hypothetical case of Re → 0, the drag coefficient
tends to infinity: CD → ∞; that is, we have a singular
prob-
lem in which the particle size is zero and the drag on
CD = Re(124
+ B ), Y − Y∞ the particle is infinitely high. In order to eliminate this
B = s
, m 1−Y
m s singularity, it is likely necessary to set CD = 0 at
where Ys, andY∞ are the mole fractions of vapor on the Re → 0. It is likely that the drag curve will pass
drop surface and far from the drop. Ackovic [82] through a maximum at some Re → Rem value. Obvi-
reported calculation of the drag coefficient for ously, the boundary of the singularity determined
unsteady-state bubble growth in a binary mixture. the lower limit of the validity of the Stokes law.
Other factors in the friction force and drag If the particle size is small compared to the free
coeffi- cient are the wall effect and interaction with
other par- ticles due to the superposition of path of the molecules of the liquid (λ,), implying that
λ
hydrodynamic fields. The wall effect is taken into the Knudsen number is Kn = 1,then molecular
account by introducing
the coefficient f d a
= F (ϑ, Re), ϑ = (D is the slip will take place, which will reduce the drag coeffi-
cient (Kn ~ M/Re). In this case, it is possible to apply
channel
D
diameter) [1, 5, 31, 51]. In the particular case of Re >
100, the coefficient accounting for the wall effect is the Cunningham correction factor [1, 4, 10], with
3 which the drag coefficient can be expressed in the
form of the following semiempirical relationship:
( )
written as f = 1 − ϑ2 2 . [1]. The drag coefficient
of
drops and bubbles depends strongly on the presence of
)
−1
adsorbed surfactants on the surface. The presence U
CD ≈ 7.09Kn [ξ (1 − Kn ) + 2RT ,
of traces of adsorbable admixtures in the solution

strengthens the interface, making the drops to

Kn] (
make the right choice among the empirical ξ(Re)
behave as solid particles. Chia-Shun Yih [64] functions, guided by their experience and intu-
presented a number formulas taking into account
the effect of the concentration of adsorbed ition. The ξ(Re) function should be continuously dif-
surfactants on the drag coefficient of drops and ferentiable over the entire Re range, and it is desirable
bubbles at small Re values. that it be fairly simple and contain the smallest possi-
ble number of unknown coefficients.
Note that experimental drag coefficient data for
CONCLUSIONS solid particles over the 2 × 105 ≤ Re ≤ 107 range vary
The above expressions for the drag coefficient of widely [1, 4, 6, 32, 64, 65]. A similar situation takes
particles are empirical and, for very small Re
values, semiempirical or theoretical. Aside from the place at very small Reynolds numbers of Re < 10 −3. It
above formulas for the drag coefficient of solid follows from the formulas presented here that, in the
particles, drops, and bubbles in Newtonian and non-
Newtonian fluids, there are many relationships valid
in various ranges of Re, We, Mo, and Bo [1, 6, 25,
26, 38, 47, 51,
64, 65].
It is natural that serious problems are
encountered at large Re values, which arise from the
changes in the flow pattern around the particle. This
leads to unpre- dictable variation of the derivative
of the drag coeffi- cient with respect to Re. An
analysis of numerous expressions and
approximations for the drag coeffi- cient of solid
particles, drops, and bubbles demon- strates that,
for obtaining a good result, a researcher should

THEORETICAL FOUNDATIONS OF CHEMICAL ENGINEERING Vol. No. 20

 2 KELBALIY
where ξ is the slip factor, (( U
)) is the ratio of the
flow velocity to the molecular motion velocity, T is
2RT
temperature, and R is the gas constant.
Since there are a wide variety of formulas for
calcu- lating the drag coefficient of solid particles,
drops, and bubbles, it is essential to find the best
formula. How- ever, here it is difficult to give a clue
to researchers because the right choice will be
determined by the fol- lowing factors:
(1) Flow pattern of the dispersion. When the
parti- cle size in the flow varies continuously
because of par- ticle coagulation or breakup and
some other physical processes (evaporation,
dissolution, condensation, etc.), it is appropriate to
choose formulas valid in a wide Re range. In
particular, this is true for the turbu- lent flow of
liquid–gas and gas–liquid dispersions in which the
drop or bubble size varies from 1 µm to 200 µm
and above, which covers a wide Re range and, if the
particles undergo deformation, wide ranges of Mo
and We. Obviously, simple formulas should be
used in CD calculations for a narrow Re range. In
par-
ticular case of solid spherical particles and Re ≤
500, the best choice ensuring an acceptable degree
of accu- racy is relationship (9), the Klyacko–
Mazin equation,
the Schiller–Neumann equation, or relationship
(25). For a narrow Re range, an acceptable
choice is Eq. (12).
(2) Re, Mo, and We ranges and the accuracy of
the formula for calculating the drag coefficient. The
rela- tive error of most of the above formulas in a
given Re range is <10%. However, although the
formulas are simple and do not exceed the
acceptable error in their validity range, they lead to
appreciable errors even when the Re value is slightly
outside this range.

THEORETICAL FOUNDATIONS OF CHEMICAL ENGINEERING Vol. 45 No. 3

 DRAG COEFFICIENTS OF VARIOUSLY SHAPED SOLID 2
(3) Particle shape and size. Note that serious diffi- SUBSCRIPTS AND SUPERSCRIPTS
culties are encountered in the determination of the
drag coefficient of deformable drops and bubbles at c—continuous medium;
Re 1, We > 1, and Mo 1 in both Newtonian and non- d—dispersed phase.
Newtonian fluids. Under these conditions, drops and
bubbles can take a shape that cannot be analyzed
experimentally or described theoretically. REFERENCES
1. Clift, R., Grace, J.R., and Weber, M.E., Bubbles,
Drops, and Particles, New York: Academic, 1978.
NOTATION 2. Bird, R., Stewart, W.E., and Lightfoot, E., Transport
Phenomena, New York: Wiley, 1960.
CD—drag coefficient of a solid particle;
3. Brounshtein, B.I. and Shchegolev, V.V.,
CDK —drag coefficient of a drop; Gidrodinamika, masso- i teploperenos v kolonnykh
CDG —drag coefficient of a bubble; apparatakh (Fluid Dynamics and Mass and Heat
Transfer in Columns), Leningrad: Khimiya, 1988.
c —speed of sound, m/s;
4. Soo, S.L., Fluid Dynamics of Multiphase Systems, Lon-
d —diameter of a spherical or cylindrical particle, don: Blaisdell, 1970.
m; 5. Proudman, I. and Pearson, J.R., Expansion at Small
FS —friction force, N; Reynolds Number for the Flow past a Sphere and
Circu- lar Cylinder, J. Fluid Mech., 1957, vol. 2, p.
f —wall effect; 237.
G —shear modulus, s/m; 6. Michaelides, E.E., Particles, Bubbles and Drops: Their
L—length, m; Motion, Heat and Mass Transfer, Singapore: World
n—shear rate exponent for non-Newtonian fluids; Sci- ence, 2006.
7. Chester, W. and Breach, D.R., On the Flow past a
R—radius, m; Sphere at Low Reynolds Numbers, J. Fluid Mech.,
r2—correlation factor; 1968, vol. 37, no. 4, p. 751.
sM —midsection area of the particle, m2; 8. Levich, V.G., Fiziko-khimicheskaya gidrodinamika
S—surface area of a sphere, m2; (Physicochemical Fluid Dynamics), Moscow: Fizmat-
giz, 1962.
Sd—surface area of an irregularly shaped particle, 9. Taneda, S., Studies on Wake Vortices. III:
m2; Experimental Investigation of the Wake behind a
U—flow velocity, m/s; Sphere at Low Rey- nolds Numbers, Rep. Res. Inst.
Appl. Mech. Kyushu Univ., 1956, vol. 4, p. 99.
Vd —particle velocity, m/s;
10. Mednikov, E.P., Turbulentnyi perenos i osazhdenie
Y—drop deformation factor; aero- zolei (Turbulent Transfer and Aerosol
γ—ratio of the viscosity of the particle to the Deposition), Moscow: Nauka, 1980.
vis- cosity of the medium; 11. Polyanin, A.D., Kutepov, A.M., Vyazmin, A.V., and
λ0—Kolmogorov scale of turbulence; Kazenin, D.A., Hydrodynamics, Mass and Heat
Transfer in Chemical Engineering, London: Taylor and
λ—free path length of the molecules of the liquid; Francis, 2002.
νc—kinematic viscosity of the medium, m2/с/s; 12. Happel, J. and Brenner, H., Low Reynolds Number
ρd —density of the particle, kg/m3; Hydrodynamics with Special Applications to Particulate
Media, Englewood Cliffs, N.J.: Prentice-Hall, 1965.
ρc—density of the medium, kg/m3; 13. Zapryanov, Z., Tabakova, S., Dynamics of Bubbles,
σ—surface tension, N/m; Drops and Rigid Particles, Dortrecht: Kluwer, 1999.
χ—deformation factor of drops and bubbles; 14. Voloshchuk, V.M. and Sedunov, Yu.S., Protsessy
ς —non-Newtonian viscosity coefficient; Ac koagu- lyatsii v dispesnykh sistemakh (Coagulation
—Acrivos number; Processes in Disperse Systems), Leningrad:
Gidrometizdat, 1975.
BO—Bond number; 15. Van Dyke, M., Perturbation Methods in Fluid Mechanics,
Ca—capillary number; Stanford, Calif.: Parabolic, 1975.
Fr—Froude number; 16. Pruppacher, H.R. and Steinberger, E.H., An Experi-
Kn—Knudsen number; mental Determination of the Drag on a Sphere at
Mo—Morton number; Low Reynolds Numbers, J. Appl. Phys., 1968, vol. 38, p.
4129.
M—Mach number;
U − Vp d 17. Hay, K.J., Liu, Z.C., and Hanratty, T.J., Relation of
Re = —Reynolds number for a particle; Deposition to Drop Size When the Rate Law Is
Nonlin- ear, Int. J. Multiphase Flow, 1986, vol. 22, p.
νc 829.
2−n
18. Bagchi, P. and Balachandar, B., Effect of Turbulence
on the Drag and Lift of a Particle, Phys. Fluids,
THEORETICAL FOUNDATIONS OF CHEMICAL ENGINEERING Vol. No. 20
 2 2003, KELBALIY
ρU d
Ret = c —Reynolds number for non- vol. 15, no. 11, p. 3496.
New- 19. Kelbaliyev, G. and Ceylan, K., Development of New
ς
tonian fluids. Empirical Equations for Estimation of Drag Coefficient,

THEORETICAL FOUNDATIONS OF CHEMICAL ENGINEERING Vol. 45 No. 3

 DRAG COEFFICIENTS OF VARIOUSLY SHAPED SOLID 2
Shape Deformation and Rising Velocity of Gas (Engl. Transl.), vol. 37, no. 6, p. 606].
Bubbles or Liquid Drops, Chem. Eng. Commun., 2007, 35. Milne-Thompson, L., Theoretical Hydrodynamics, Lon-
vol. 194, p. 1623. don: Chapman and Hall, 1960.
20. Majumder, A.K. and Barnwal, J.P., A Computational
Method to Predict Particle Free Terminal Settling
Velocity, Indian Mining Eng. J., 2004, vol. 85, p. 17.
21. Almedeij, J., Drag Coefficient of Flow around a
Sphere: Matching Asymptotically the Wide Trend,
Powder Tech- nol., 2008, vol. 186, no. 3, p. 218.
22. Brown, P.P. and Lawler, D.F., Sphere Drag and
Settling Velocity Revisited, J. Environ. Eng., 2003,
vol. 129, p. 222.
23. Peria, E., Anta, J., Puertas, J., and Teijero, T.,
Estima- tion of Drag Coefficient and Setting
Velocity of the Cockle Cenastoderma Edule Using
Particle Image Velocimetry, J. Coastal Res., 2008, vol.
24, p. 150.
24. Flemmer, R.L.C. and Banks, C.L., On the Drag
Coeffi- cient of a Sphere, Powder Technol., 1986,
vol. 48, p. 217.
25. Turton, R. and Levenspiel, O.A., A Short Note on
the Drag Correlation for Spheres, Powder Technol.,
1986, vol. 47, no. 1, p. 83.
26. Concha, F. and Barrientos, A., Settling Velocities of
Particulate Systems. 3. Power-Series Expansions for
the Drag Coefficient of a Sphere and Production of the
Set- tling Velocity, Int. J. Miner. Process., 1982, vol. 9,
no. 2, p. 167.
27. Nian-Sheng, Cheng., Comparison of Formulas for
Drag Coefficient and Settling Velocity of Spherical
Particles, Powder Technol., 2009, vol. 189, no. 3, p.
395.
28. Gabito, J. and Tsouris, C., Drag Coefficient and
Settling Velocity for Particles of Cylindrical Shape,
Powder Tech- nol., 2008, vol. 183, no. 2, p. 314.
29. Kurose, R. and Makino, H., Effect of Outflow from
the Surface of a Sphere on Drag, Shear Lift and
Scalar Dif- fusion, Phys. Fluids, 2003, vol. 15, no. 3, p.
2338.
30. Balduga, J., Henczka, M., Shekunov, B.Y., Fluid
Dynamics, Mass Transfer and Particle Formation, in
Supercritical Fluid Technology for Drug Product
Develop- ment, York, P., Kampella, U.B., and
Shekunov, B.Y., Eds., New York: Marcel Dekker,
2004, p. 91.
31. Ceylan, K., Altunbas, A., and Kelbaliyev, G., A New
Model for Estimation of Drag Force in the Flow of
New- tonian Fluids around Rigid of Deformable
Particles, Powder Technol., 2001, vol. 119, p. 250.
32. Kutateladze, S.S., Teploperedacha i
gidrodinamicheskoe soprotivlenie (Heat Transfer and
Drag), Moscow: Ener- goatomizdat, 1990.
33. Aerov, M.E. and Todes, O.M., Gidravlicheskie i
teplovye osnovy raboty apparatov so statsionarnym i
kipyashchim zernistym dloem (Hydraulic and Thermal
Principles of Operation of Fixed- and Fluidized-Bed
Apparatuses), Leningrad: Khimiya, 1968.
34. Kondrat’ev, A.S. and Naumova, E.A., Calculation of
the Velocity of Free Settling of Solid Particles in a
New- tonian fluid, Teor. Osn. Khim. Tekhnol., 2003,
vol. 37, no. 6, p. 646 [Theor. Found. Chem. Eng.

THEORETICAL FOUNDATIONS OF CHEMICAL ENGINEERING Vol. No. 20

 2 Statie, E., Salcudean, M., Gartshore, I., and Bibeau,
36.
KELBALIY
Sci., 2000, vol. 55, p. 943.
E., A Computational Study of Particles Separation 55. Sis, H., Kelbaliyev, G., and Chander, S., Kinetics of
in Hydrocyclones, J. Pulp Paper Sci., 2002, vol. 38, Drop Breakage in Stirred Vessels under Turbulent Con-
p. 84. ditions, J. Dispersion Sci. Technol., 2005, vol. 26, p.
37. Chien, S.F., Settling Velocity of Irregularly Shaped 566.
Par- ticles, Proc. 69th Annual Technical Conf. and
Exhibition, New Orleans, 1994.
38. Haider, A. and Levenspiel, O., Drag Coefficient
and Terminal Velocity of Spherical and
Nonspherical Parti- cles, Powder Technol., 1989, vol.
58, p. 63.
39. Ganser, G.A., A Rational Approach to Drag
Prediction of Spherical and Nonspherical Particles,
Powder Tech- nol., 1993, vol. 77, p. 143.
40. White, F.M., Viscous Flow, New York: McGraw-
Hill, 2006, 3rd ed.
41. Yin, C., Rosedae, L., Kaer, S.K., and Sorenson, H.,
Modeling the Motion of Cylindrical Particles in
Non- uniform Flow, Chem. Eng. Sci., 2003, vol. 58, p.
3489.
42. Taylor, T. and Acrivos, A., On the Deformation
and Drag of a Falling Drop at Low Reynolds
Number, J. Fluid Mech., 1964, vol. 18, p. 466.
43. Ockendon, J.R. and Evans, G.A., The Drag on a
Sphere in Low Reynolds Number Flow, J. Aerosol
Sci., 1972, vol. 3, no. 4, p. 237.
44. LeClair, B.P., Hamielek, A.E., and Pruppacher,
H.R., A Numerical Study of the Drag on a Sphere at
Low Rey- nolds Numbers, J. Atmos. Sci., 1970, vol.
27, p. 308.
45. Bhaga, D. and Weber, M.E., Bubbles in Viscous
Liquids: Shape, Wakes and Velocities, J. Fluid
Mech., 1981, vol. 105, p. 61.
46. Bozzano, G. and Dente, M., Shape and Velocity of
Sin- gle Bubbles Motion: A Novel Approach,
Comput. Chem. Eng., 2001, vol. 25, p. 571.
47. Grace, J.R., Hydrodynamics of Liquid Drops in
Immis- cible Liquids, Handbook of Fluids in
Motion, Chere- misinoff, N.P. and Gupta, E., Eds.,
Ann Arbor, Mich.: Ann Arbor Science, 1983 p. 273.
48. Helenbrook, B.T. and Edwards, C.F., Quasi-Steady
Deformation and Drag of Contaminated Liquid
Drops, Int. J. Multiphase Flow, 2002, vol. 28, p.
1631.
49. Maxworty, T., Grann, C., Kurten, M., Durst, F.,
Exper- iments on the Rise of Air Bubbles in Clean
Viscous Liq- uids, J. Fluid Mech., 1996, vol. 321, p.
421.
50. Sajjadi, S., Zerfa, M., and Brooks, B.M., Dynamic
Behaviors of Drops in Oil/Water Dispersion, Chem.
Eng. Sci., 2002, vol. 57, p. 663.
51. Chabra, R.P., Bubbles, Drops, and Particles in Non-
New- tonian Fluids, Boca Raton, Fla.: CRC, 2007.
52. Rodi, W. and Fueyo, N., Engineering Turbulence
Mod- eling and Experiments, Proc. 5th Int. Symp. on
Engineer- ing Turbulence Modelling and
Measurements, Mallorca, Spain, 2002.
53. Fenn, Z.G. and Michaelides, E.E., Heat and Mass
Transfer Coefficient of Viscous Spheres, Int. J.
Heat Mass Transfer, 2001, vol. 44, no. 23, p. 4445.
54. Raymond, F. and Rozant, J.M., A Numerical and
Experimental Study of the Terminal Velocity and
Shape of Bubbles in Viscous Liquids, Chem. Eng.

THEORETICAL FOUNDATIONS OF CHEMICAL ENGINEERING Vol. 45 No. 3

 2 KELBALIY
56. Harmathy, T.Z., Velocity of Large Drops and Bubbles
in Media of Infinite or Restricted Extent, AIChE J., 70. Ceylan, K., Herdem, S., and Abbasov, T., A
1960, vol. 6, no. 2, p. 281. Theoretical Model for Estimation of Drag Force in the
Flow of Non- Newtonian Fluids around Solid spherical
57. Moore, D.W., The Boundary Layer on Spherical Gas Particles, Pow- der Technol., 1999, vol. 103, p. 286.
Bubbles, J. Fluid Mech., 1965, vol. 16, p. 161.
71. Shul’man, Z.P. and Berkovskii, B.M., Pogranichnyi
58. Wellek, R.M., Angrawal, A.K., and Skelland, A.H., sloi nen’yutonovskikh zhidkostei (Boundary Layer in
Shape of Liquid Drops Moving in Liquid Media, Non- Newtonian fluids), Minsk: Nauka i Tekhnika,
AIChE J., 1966, vol. 12, p. 854. 1966.
59. Sherman, Ph., Emulsion Science, London: Academic,
1968. 72. Dewsbury, K., Karamanev, D.G., and Margaritis,
S.A., Hydrodynamic Characteristics of Free Rise of
60. Sarimeseli, A. and Kelbaliyev, G., Modeling of the Light Solid Particles and Gas Bubbles in Non-
Break-Up of Deformable Particles in Turbulent Flow, Newtonian flu- ids, Chem. Eng. Sci., 1999, vol. 54, p.
Chem. Eng. Sci., 2004, vol. 59, p. 1233. 4825.
61. Haberman, W.L. and Morton, R.K., An 73. Kelessids, V.C., An Explicit Equation for the Terminal
Experimental Investigation of the Drag and Shape of Velocity of Solid Spheres Falling in Pseudoplastic Liq-
Air Bubbles Ris- ing in Various Liquids, D.W. Taylor uids, Chem. Eng. Sci., 2004, vol. 59, p. 4437.
Model Basin Report, Department of the Navy,
Washington, DC, 1953. vol. 802. 74. Miura, H., Takahachi, T., and Ichikawa, K., Bed
62. Hovenkamp, B., Investigation of Porous Media Flow Expansion in Liquid–Solid Two-Phase Fluidized
with Regard to the Emulsion Science, Doctoral Beds with Newtonian and Non Newtonian Fluids
(Eng.) Dissertation, Swiss Federal Inst. of Technology, over the Wide Range of Reynolds Numbers, Powder
Zurich, 2002. Technol., 2001, vol. 117, p. 239.
63. Kelbaliev, G.I. and Ibragimov, Z.I., Coalescence and 75. Pinelli, D. and Magelli, F., Solids Falling and
Fragmentation of Drops in an Isotropic Turbulent Distribu- tion in Slurry Reactors with Dilute
Flow, Teor. Osn. Khim. Tekhnol., 2009, vol. 43, no. Pseudoplastic Sus- pension, Ind. Eng. Chem. Res.,
3, p. 16 [Theor. Found. Chem. Eng. (Engl. Transl.), vol. 2001, vol. 40, p. 4456.
43, no. 3, p. 314]. 76. Hirose, T. and Moo-Yung, M., Bubble Drag and
64. Chia-Shun, Yih., Advances in Applied Mechanics, New Mass Transfer in Non-Newtonian Fluids: Creeping
York: Academic, 1972. Flow with Power-Law Fluids, Can. J. Chem. Eng.,
65. Crowe, C.T., Multiphase Flow Handbook, Boca Raton, 1969, vol. 47, no. 3, p. 265.
Fla.: CRC, Taylor and Francis Group, 2006. 77. Loitsyanskii, L.G., Mechanics of Liquids and Gases,
66. Clift, K.A. and Lever, D.A., Isothermal Flow past a Oxford: Pergamon, 1966.
Blowing Sphere, Int. J. Numer. Methods Fluids, 1985, 78. Bricato, A., Ciofalo, M., Grisafi, F., and Micale, G.,
vol. 5, p. 709. Numerical Prediction of Flow Fields in Baffled
67. Karamanev, D.G., Equation for Calculation of the Stirred Vessels, Chem. Eng. Sci., 1998, vol. 53, no. 21,
Ter- minal Velocity and Drag Coefficient of Solid p. 3653.
Spheres and Gas Bubbles, Chem. Eng. Commun., 1996, 79. Tropea, C., Yarin, A.L., and Foss, J.F., Springer
vol. 147, p. 73. Hand- book of Experimental Fluid Mechanics, New
68. Dewsbury, K.H., Karamanev, D.G., and Margaritis, A., York: Springer, 2007.
Rising Solid Hydrodynamics at High Reynolds 80. Ceylan, K. and Kelbaliyev, G., The Roughness Effect
Num- bers in Non-Newtonian Fluids, Chem. Eng. on Friction and Heat Transfer in the Fully Developed
Sci., 2002, vol. 87, p. 120. Tur- bulent Flow, Appl. Therm. Eng., 2003, vol. 23, p.
69. Chabra, R.P. and Richardson, J.F., Non-Newtonian 557.
Flow in the Process Industries: Fundamentals and 81. Altunbas, A., Kelbaliyev, G., and Ceylan, K., Eddy
Engi- neering, Oxford: Butterworth–Heinemann, 1999. Dif- fusivity of Particles in Turbulent Flow in Rough
Chan- nels, J. Aerosol Sci., 2002, vol. 33, p. 1075.
82. Ackovic, R., Drag of a Growing Bubble at Rectilinear
Accelerated Ascension in Purer Liquids and Binary
Solutions, Theor. Appl. Mech., 2003, vol. 30, no. 3,
p. 177.

THEORETICAL FOUNDATIONS OF CHEMICAL ENGINEERING Vol. 45 No. 3 2011

 DRAG COEFFICIENTS OF VARIOUSLY SHAPED SOLID 2

View publication stats

THEORETICAL FOUNDATIONS OF CHEMICAL ENGINEERING Vol. No. 20