minerals

Article
CFD Simulation of Pipeline Transport Properties of
Mine Tailings Three-Phase Foam Slurry Backfill
Xin Chen ID
, Jian Zhou ID
, Qiusong Chen *, Xiuzhi Shi * and Yonggang Gou ID

School of Resources and Safety Engineering, Central South University, Changsha 410083, China;
chenxin_ck@csu.edu.cn (X.C.); csujzhou@hotmail.com (J.Z.); gouyonggang@csu.edu.cn (Y.G.)
* Correspondence: qiusong.chen@csu.edu.cn (Q.C.); shixiuzhi@263.net (X.S.)

Received: 12 July 2017; Accepted: 13 August 2017; Published: 17 August 2017

Abstract: A three-dimensional backfill pipeline transport model is developed using the computational
fluid dynamics (CFD) technique, which is applied to study the pipeline transport properties of
three-phase foam slurry backfill (TFSB). Based on rheological property tests and CFD simulations,
the foam phase, pressure, and velocity in the pipeline system are investigated using the CFD
mixture method for different bubble volume fractions and bubble diameters. The simulation results
indicate that TFSB can maintain a steady state during pipeline transport, experience a markedly
reduced pipeline transport resistance, and exhibit better liquidity than conventional cement slurry.
Furthermore, as the bubble volume fraction increases, the resistance of the pipeline decreases and
the fluidity improves. By contrast, the bubble diameter has little effect on the transport properties
of TFSB. The combined results of CFD simulations, slump tests, and strength tests indicate that,
when the bubble volume fraction is 15–20 vol %, TFSB can satisfy the necessary strength requirements
and exhibit self-flowing transport. The CFD technique provides an intuitive and accurate basis for
pipeline transport research and has the potential for wider application in studies of mine backfill.

Keywords: three-phase foam slurry backfill (TFSB); pipeline transportation properties; computational
fluid dynamics (CFD) simulation; bubble volume fraction

1. Introduction
Paste backfill technology represents an important direction for the development of mine backfill
techniques [1–5]. After many years of continuous exploration and practice, paste backfill technology
has been accepted and applied in numerous countries throughout the world by virtue of its
advantages in terms of environmental protection, emission reduction, safety, and other aspects [6–8].
Backfill pipeline systems have been established in many mines as transportation systems to transport
backfill material from the surface to the goaf [9]. As mining depth increased, these backfill pipelines
have become much longer and deeper than before, and it has become difficult to achieve self-flowing
transport of conventional cement slurry because of its poor fluidity. Although paste pumping
technology can solve this problem, the purchase and maintenance of the necessary backfill pump
equipment represents an enormous expenditure for a mine. Therefore, an increasing amount of
attention has been paid to improve the transport properties of paste slurry while satisfying the
transportation and strength requirements of the backfill [10].
Foam concrete is a new type of lightweight concrete material, which is composed of standard
foam and cement and possesses a large number of closed pores. With high fluidity and low
cement and aggregate usage, foam concrete is applied in sandwich structures, earth-retaining walls,
and running tracks or playgrounds [11]. Based on successful experiences with applying foam concrete,
Zhang et al. [12] and Chen et al. [13] have studied the application of foam slurry in mine backfill and
have concluded that the backfill can meet the requirements.

Minerals 2017, 7, 149; doi:10.3390/min7080149 www.mdpi.com/journal/minerals

Minerals 2017, 7, 149 2 of 19

For the continuous improvement of foam slurry backfill technology, research on the pipeline
transport properties of solid-liquid-gas three-phase foam slurry backfill (TFSB) is necessary. In the
past, research on pipeline transportation technology has been performed mostly based on analogy,
experience-based formulas, and laboratory experiments to determine the optimal slurry backfill
parameters [14,15]. However, such methods cannot be used to investigate the effects of foam factors
(bubble volume fraction, bubble diameter, and so on) on the motion characteristics, pipe pressure
loss, and changes in velocity of slurry backfill. With the development of computer technology,
computational fluid dynamics (CFD) [16,17], which is based on classical fluid mechanics and numerical
methods, has begun to be used to study the properties of slurry pipelines for backfill. Eesa and
Barigou [18] published a CFD analysis of viscous non-Newtonian flow under the influence of
a superimposed rotational vibration.
In this paper, the CFD technique is used to build a 3D pipeline transport model, which is applied
to simulate the transport properties of TFSB. These simulations are combined with laboratory tests to
study the effect of foam on the rheological properties of backfill slurry.

2. Materials and Methods

2.1. Foam Slurry Materials

2.1.1. Constituent Materials

TFSB refers to a backfill material that is synthesized from an admixture foam, cement, water,
and necessary additives, which are mixed according to certain proportions and hardened by physical
and chemical effects. In the case of a high slurry concentration and a low water-to-cement ratio,
it also maintains good fluidity [19]. Foam was produced by a high-performance soil foaming agent of
the SR-FL-001 [20] (dilution ratio 1:20 by weight). The basic properties of the standard foam are as
follows: a density of 40–60 kg/m3 , a foam diameter of 30–100 µm, and segregation-free foam when the
temperature is above 0 ◦ C. Cement was P.O 42.5R [21]. The water used in this research was tap water.
In this study, mine tailings from a lead–zinc mine in China were used as additives, whose particles are
less than 0.074 mm and about 80 wt %, and their mean grain size (d50 ) is 24.2 µm.

2.1.2. Manufacturing Process

The preparation of foam slurry is divided into three steps. Figure 1 summarizes the
manufacturing process.

1. Foam preparation: A special foaming agent is fully mixed with water according to a given
dilution ratio to generate foam. Then, the diluted liquid is loaded into a foaming gun along with
compressed air at a certain pressure to prepare a standard-density foam.
2. Cement slurry preparation: Cement, tailings, and water are mixed according to given proportions
to prepare the cement slurry.
3. TFSB preparation: The premade cement slurry and standard-density foam are mixed in a given
proportion to prepare the TFSB.
Minerals 2017, 7, 149 3 of 19
Minerals 2017, 7, 149 3 of 19

Figure 1. Manufacturing processes for foam slurry.

Figure 1. Manufacturing processes for foam slurry.

2.2. Test Methods

2.2. Test Methods
2.2.1. Rheological Test
2.2.1. Rheological Test
Rheological properties of TFSB were measured using a rheometer (R/S Plus) from Brookfield
Rheological properties of TFSB were measured using a rheometer (R/S Plus) from Brookfield
Engineering Laboratories, Inc (Middleboro, MA, USA). According to the proportion ratio, the slurry
Engineering Laboratories, Inc (Middleboro, MA, USA). According to the proportion ratio, the slurry
was mixed as uniformly as possible for 5 min and subsequently placed in the rheometer. The rotor
was mixed as uniformly as possible for 5 min and subsequently placed in the rheometer. The rotor
speed of rheometer increased gradually from zero and ran for 120 s. Record shear stress and plastic
speed of rheometer increased gradually from zero and ran for 120 s. Record shear stress and plastic
viscosity per second.
viscosity per second.
2.2.2.
2.2.2. Slump
Slump Tests
Tests
The slumptest
The slump testwas
wasoriginally
originally developed
developed to determine
to determine the the “workability”
“workability” or “consistency”
or “consistency” of
of fresh
fresh concrete, and the standard slump instrument is a tapered cylinder with a height
concrete, and the standard slump instrument is a tapered cylinder with a height of 30 cm, an upper of 30 cm, an
upper diameter
diameter of 10
of 10 cm, cm,
and and a bottom
a bottom diameter diameter of [22].
of 20 cm 20 cmThe
[22]. The was
slurry slurry was poured
poured into theinto the
slump
slump instrument and tamped. Then picked up the slump instrument quickly, and
instrument and tamped. Then picked up the slump instrument quickly, and the slurry flowed freely. the slurry
flowed
The freely.from
distance Thethe
distance
top of from the top
the slurry of top
to the the of
slurry to the instrument
the slump top of the was
slump instrument
measured afterwas
the
measured after the slurry shape is stable; the distance
slurry shape is stable; the distance is the slump value. is the slump value.

2.2.3. Strength Tests

The slurry
slurrywaswasmolded
moldedinin dimensions
dimensions of 70.7
of 70.7 mmmm × 70.7
× 70.7 mmmm × 70.7× 70.7
mm,mm, andspecimens
and the the specimens
were
were maintained by a TH10D laboratory humidifier (Texingda, Beijing, China),
maintained by a TH10D laboratory humidifier (Texingda, Beijing, China), which was kept working which was kept
working
to ensure to
thatensure that the temperature
the temperature was controlledwas within
controlled within
a range of 23–26 ◦
a range of 23–26
C and °Chumidity
that the and that wasthe
humidity was controlled within a range of 83%–87%. The compressive strength
controlled within a range of 83%–87%. The compressive strength was tested respectively after 7 and was tested
respectively after 7 and rigid
28 days by a WDW-2000 28 days by a WDW-2000
hydraulic pressure servorigid hydraulic
machine pressure
(Ruite, Guilin,servo machine (Ruite,
China).
Guilin, China).
2.3. CFD Simulation
2.3. CFD Simulation
As shown in Figure 2, a three-dimensional model was developed for a typical L-shaped backfill
pipeline for long-distance
As shown in Figure transport, which was used
2, a three-dimensional here towas
model study the transport
developed for aproperties of TFSB.
typical L-shaped
The pipe is made of seamless steel, and its detailed geometrical parameters are a
backfill pipeline for long-distance transport, which was used here to study the transport properties vertical height of
100 m, a The
horizontal ◦
of TFSB. pipe islength 500 m, a diameter
made of seamless steel, and ofits102 mm, an
detailed elbow angle
geometrical of 90 , and
parameters are an elbow
a vertical
radius of 1100
height m.m,The model can length
a horizontal be usedofto500analyze the velocity
m, a diameter andmm,
of 102 pressure distribution
an elbow angle ofof90°,
theand
slurry.
an
The
elbowmodel was
radius ofestablished and meshed
1 m. The model can be in Gambit,
used and the
to analyze thecalculation
velocity and accuracy wasdistribution
pressure confirmed by of the
mesh number.
slurry. The modelSix groups
was of different mesh
established and numbers
meshed are in tested
Gambit,on and
the model and compared
the calculation in terms
accuracy of
was
pressure,
confirmedvelocities,
by the meshand volume
number.fraction to ensure
Six groups a mesh-independent
of different mesh numbers solution. Theon
are tested results show that,
the model and
compared in terms of pressure, velocities, and volume fraction to ensure a mesh-independent
solution. The results show that, when the mesh number exceeds 2.08 × 106, the variations of
Minerals 2017, 7, 149 4 of 19
Minerals 2017, 7, 149 4 of 19

when the mesh

pressure, velocitynumber exceeds
and volume fraction 106so
2.08 ×are , the variations
little that theseofcan
pressure, velocity
be negligible and volume
generally. fraction
We found that
are × 6
the mesh number of 1.28 × 10 gives about a 1% deviation compared to the mesh number of 2.08gives
so little that these can be negligible
6 generally. We found that the mesh number of 1.28 10 × 106;
about a 1%
the mesh deviation
number compared
of 3.24 to about
× 106 gives the mesh
a 9%number
deviationof 2.08 × 106 ;tothe
compared themesh
meshnumber 2.08×× 10
numberofof3.24
6
106.
gives about the a 9%mesh
deviation compared 6
× 10 .for
Therefore, number of 2.08 to the
× 10 meshfound
6 was number of 2.08
to be sufficient Therefore, the mesh
the numerical number
simulation.
of 2.08 × 10 6 was found to be sufficient for the numerical simulation. As an index of the computational
As an index of the computational cost, the model takes approximately 2 h of computing time on a
cost, the model
workstation takes
with twoapproximately 2 h of computing
quad-core processors time on
(2.57 GHz) anda workstation
16GB of RAM with(Random-Access
two quad-core
processors
Memory). (2.57 GHz) and 16GB of RAM (Random-Access Memory).

Figure 2. Pipeline transport system for backfill in an underground mine.

Figure 2. Pipeline transport system for backfill in an underground mine.

2.3.1. Governing Conservation Equations

2.3.1. Governing Conservation Equations
TFSB is composed of a gaseous foam, solid aggregates and cement, and liquid water. In this
TFSB is composed of a gaseous foam, solid aggregates and cement, and liquid water. In this work,
work, to fully represent the effect of the foam on the transport properties of the backfill slurry,
to fully represent the effect of the foam on the transport properties of the backfill slurry, simplified
simplified CFD simulations of a two-phase flow are considered, in which the cement slurry is
CFD simulations of a two-phase flow are considered, in which the cement slurry is treated as the
treated as the first phase and the foam is treated as the second phase. The Fluent CFD tool was used
first phase and the foam is treated as the second phase. The Fluent CFD tool was used to perform
to perform the simulations, and the interfaces between the slurry and foam were tracked using the
the simulations, and the interfaces between the slurry and foam were tracked using the mixture
mixture method (MIX). When using MIX for interface tracking, it is customary to use a two-phase
method (MIX). When using MIX for interface tracking, it is customary to use a two-phase model, i.e.,
model, i.e., a model with a mixture pressure field, a mixture velocity field, mixture density, and a
a model with a mixture pressure field, a mixture velocity field, mixture density, and a second phase
second phase volume fraction [23]. In MIX, the finite volume discretized forms of the equations that
volume fraction [23]. In MIX, the finite volume discretized forms of the equations that govern the
govern the conservation of mass, momentum and volume fraction (Equations (1)–(3)) are solved
conservation of mass, momentum and volume fraction (Equations (1)–(3)) are solved over a collocated
over a collocated grid arrangement using a segregated solver approach (the PISO algorithm) for the
grid arrangement using a segregated solver approach (the PISO algorithm) for the pressure–velocity
pressure–velocity calculations. MIX allows phases to pass through each other; thus, the volume
calculations. MIX allows phases to pass through each other; thus, the volume fraction of each phase
fraction of each phase can take arbitrary values between 0 and 1. At the same time, the
can take arbitrary values between 0 and 1. At the same time, the characteristics of the mixed fluid in
characteristics of the mixed fluid in the control region are determined by each phase [23]. The
the control region are determined by each phase [23]. The formulas for the viscosity can be written as
formulas for the viscosity can be written as shown in Equation (4).
shown in Equation (4).  * 
ρ v = 0

 ∂
( )
∂
( m ) + ∇ ·
( ρm m ) + ∇ ⋅*mρ m vmm = 0
 ∂t ρ m
v ∂t =


 *
n
 m ∑ k =1 αk ρk v k /ρm

(1)

k
n n
 ρmv m== ∑ α ρα ρ v k ρ

(1)


k =1 k k k m
 k =1
n
∂  *
  
* * ρm =  α k ρk
  
* *T

ρm v m + ∇ · ρm v m v
m = −∇k =1
p + ∇ · µm ∇ v m + ∇ v m + ρm mg+
∂t
n
! (2)
*
  + ∇ · ∑ αk ρk v dr,k v dr,k  T
)
* *
∂
∂t
( ) ( F
ρ m v m + ∇ ⋅ ρ m v m v m = −∇) 
( 
k =1 p + ∇ ⋅  μ m ∇ v m + ∇ v m  + ρ m mg +
 
(2)
α p ρ p + ∇ ·  α p ρ p v m = −∇
∂
 
n *

 · α ρ v
 *

F + ∇ ⋅   α k ρ k v dr , k v dr , k  p p dr,p
(3)
∂t
 k =1
µ = µ 1−α +µ α

 (4)
m q p p p
Minerals 2017, 7, 149 5 of 19

where m is the mixture mass, ρm is the mixture density (kg/m3 ), ρk is the density of phase k (kg/m3 ),
* *
v m is the mixture average velocity (m/s), v k is the velocity of phase k (m/s), αk is the volume fraction
of phase k (%), n is the number of phases, ∇ p is the difference in pressure (Pa), µm is the viscosity of
*
the mixture (Pa·s), g is the acceleration of gravity (i.e., 9.8 m/s2 ), and F is the volume force (Pa).

2.3.2. Flow State

The flow state of the fluid can be determined based on the Reynolds number. The three-phase
flow of foam slurry is non-Newtonian, and the equation for the Reynolds number based on the plastic
viscosity can be written as follows [24]:
Re = ρvD/µ (5)

where Re is the Reynolds number, ρ is the density of the slurry (kg/m3 ), v is the velocity of the slurry
(m/s), and µ is the viscosity (Pa·s).
When the Reynolds number is less than 2300, the flow state of the fluid in the pipeline is laminar;
when the Reynolds number is more than 2300 but less than 4000, the flow is in a transition state
between laminar and turbulent flow; when the Reynolds number is greater than 4000, the flow
state is turbulent [25,26]. Generally, when the bubble volume fraction is 0–40 vol %, calculated by
Equations (2) and (3), their corresponding mixture viscosity is 0.4100–0.2780 Pa·s, and the mixture
density is 1750–1070 kg/m3 . Assuming the backfilling ability is 60 m3 /h, the velocity is 2.24 m/s.
Therefore, the Reynolds value is 975–879. The maximum Reynolds is 975 when the bubble volume
fraction is 0 vol % and less than 2300, the slurry flow state is laminar.

2.3.3. Boundary Conditions

The effects of heat exchange, vibration, stress waves, and compression are not considered in the
simulations [27]. The applicable boundary conditions are assumed as follows:

(1) At the walls: the wall is stationary and no-slip.

(2) At the pipe inlet, the velocity (v) function given below is used at the inlet face [28]:

v = k · Q/(900πD2 ) (6)

where Q is the slurry flow backfilling ability (m3 /h), D is the diameter of the backfill pipe (m),
the value of which is 0.102 m, k is a constant, and the value is 1.1. Thus, when the backfilling
ability is 60 m3 /h, the slurry velocity is 2.24 m/s.
(3) At the pipe outlet, the outflow function is used at the outlet face.
(4) In the calculation domain, a gravity field is applied, and the standard atmospheric pressure (1 bar)
is used as the reference to atmospheric pressure.

3. Results and Discussion

3.1. Foam Formation Characteristics

Foam is a porous membrane dispersion system formed of a mixture of liquid and gas, in which
the liquid is in a continuous phase and the gas is in a dispersed phase, and a standard foam flow is
the necessary condition and foundation for foam slurry preparation. During the foaming process,
surfactant molecules adhere to and orient on the gas–liquid interfaces to form a monolayer [29].
The foam formation process is illustrated in Figure 3. Figure 3a shows the formation of a bubble in the
water and the adsorption of a large number of surfactant molecules on its surface. The hydrophilic
groups are oriented toward the water, and the hydrophobic groups are oriented toward the air.
Figure 3b depicts the bubble rising in the water and the formation of two monolayer films, an inner
and an outer layer. Figure 3c shows the surfacing of the bubble, when it finally becomes a free bubble.
The surfactant molecules effectively overcome the surface tension of the bubble.
foam
foam formation
formation process
process is
is illustrated
illustrated in
in Figure
Figure 3.
3. Figure
Figure 3a
3a shows
shows the
the formation
formation of of aa bubble
bubble in
in the
the
water and the adsorption of a large number of surfactant molecules on its surface. The
water and the adsorption of a large number of surfactant molecules on its surface. The hydrophilic hydrophilic
groups
groups areare oriented
oriented toward
toward the
the water,
water, and
and the
the hydrophobic
hydrophobic groups
groups are
are oriented
oriented toward
toward thethe air.
air.
Figure
Figure 3b depicts the bubble rising in the water and the formation of two monolayer films, an inner
3b depicts the bubble rising in the water and the formation of two monolayer films, an inner
and
and an outer
an2017,
Minerals outer layer.
layer. Figure
7, 149 Figure 3c3c shows
shows the
the surfacing
surfacing of
of the
the bubble,
bubble, when
when it it finally
finally becomes
becomes aa6 of free
free
19
bubble. The surfactant molecules effectively overcome the surface tension of
bubble. The surfactant molecules effectively overcome the surface tension of the bubble.the bubble.

Figure 3. Foam formation process: (a)

(a) bubble formation, (b)
(b) adhesion and rise, and (c)
(c) foam
Figure 3.3.Foam
Foamformation
formation process:
process: bubble
(a) bubble formation,
formation, adhesion
(b) adhesion and rise,and
and rise, andformation.
(c) foam foam
formation.
formation.
As
As shown ininFigure 4, when the slurry is mixed with the foam in a mixing tank, under under
the action
As shown
shown in Figure
Figure 4,
4, when
when the
the slurry
slurry is
is mixed
mixed with
with the
the foam
foam in
in aa mixing
mixing tank,
tank, under the
the
of external
action forces, the solid particles and bubblesbubbles
collide with and adhere adhere
to each other at a certain
action of
of external
external forces,
forces, the
the solid
solid particles
particles and
and bubbles collide
collide with
with and
and adhere to to each
each other
other at
at aa
speed
certainand angle, ultimately forming TFSB. TFSB.
certain speed
speed and
and angle,
angle, ultimately
ultimately forming
forming TFSB.

Figure 4.
Figure 4.
4. Formation
Formation process
Formation process of
process of foam
foam slurry:
slurry: (a)
(a) mixture,
(a) mixture, (b)
mixture, (b) stirring
stirring and adhesion and
and adhesion (c) foam
and (c) foam
foam
slurry formation.
slurry formation.
formation.

The stability
The stability of
stability of the
of the foam
the foam
foam is is the
is the key
the key to
key to the
to the preparation
the preparation of
preparation of TFSB
of TFSB [30].
TFSB [30]. A
[30]. A greater
A greater viscosity
greater viscosity leads
viscosity leads to
leads to
to
The
better stability
better stability but
stabilitybut poorer
butpoorer
poorer foamability.
foamability. In addition,
In addition, the bubble
the bubble size is also an important factor
better foamability. In addition, the bubble size is size is also
also an an important
important factor
factor affecting
affecting
affecting the
the stability
stability of
of the
the foam;
foam; under
under normal
normal circumstances,
circumstances, when
when the
the bubble
bubble size
size is
is uniform,
uniform,
the stability of the foam; under normal circumstances, when the bubble size is uniform, the stability of
the stability
the bubbles of
stability is the
the bubbles
ofgood.bubbles is is good.
good.
the
3.2. Test
3.2. Test Results
Test Results Analyses
Results Analyses
Analyses

3.2.1. Rheological
3.2.1. Properties
Rheological Properties
TFSB is
TFSB is similar
similar to
to aa paste
paste slurry material
paste slurry material because
because of of its
its high
high mass
mass fraction,
fraction, low
low water
water content,
content,
and
and good
and good integrity.
integrity.InIn
goodintegrity. the
Inthe present
thepresent
present work,
work, thethe
work, flow
flow
the of
of slurry
of slurry
flow is
is modeled
is modeled
slurry as a as
modeled aa Bingham
Bingham
as fluid fluid
Bingham [31]. A
fluid [31]. A
fluid
[31]. A
fluid
that that
fluidbehavesbehaves as
as a solid
that behaves a solid
as a until until
solidauntil a
minimum minimum yield
yield stress
a minimum stress
yield τ τ is exceeded,
τ 00 is exceeded,
0 is exceeded,
stress and
and thatand that
exhibits exhibits a linear
a linear relation
that exhibits a linear
relation
between between
the shear the shear
stress and stress
the and
share the
rate of share rate
deformation, of deformation,
is referred to asis
a referred
Bingham
relation between the shear stress and the share rate of deformation, is referred to as a Bingham to as
plastic.a Bingham
The shear
plastic.
stress The
model shear
for a stress
Bingham model for
fluid isa Bingham
[32] fluid
plastic. The shear stress model for a Bingham fluid is [32] is [32]
·
τ = τ0 + µγ (7)
ττ == ττ 0 ++ μ
μ γγ (7)
(7)
0 ·
where τ is the shear stress (Pa), τ0 is the yield stress (Pa), µ is the plastic viscosity (Pa·s), and γ is the
share rate (1/s).
Figure 5a,b show the evolutions of the shear stress and the plastic viscosity with the share rate
for the cement slurry and the standard foam. Figure 5a shows that the relationship between the shear
stress and the shear rate is linearly increasing. The yield stresses of the cement slurry and the standard
foam are 68.23 and 15.54 Pa, respectively. Figure 5b shows that the relationship between the plastic
 shear stress and the shear rate is linearly increasing. The yield stresses of the cement slurry and the
standard foam are 68.23 and 15.54 Pa, respectively. Figure 5b shows that the relationship between
the plastic viscosity and the shear rate is exponentially decreasing. The plastic viscosity tends to
reach a stable value for shear rates higher than 60 r/s, and these stable values of the plastic viscosity
for the cement slurry and the standard foam are 0.41 and 0.08 Pa·s, respectively. In the present CFD
Minerals 2017, 7, 149 7 of 19
simulations, these values are used as the average values for the rheological properties of the
materials. However, in reality, slurry is not a homogeneous fluid. Therefore, in the simulations,
each phase
viscosity andisthetreated
shearas a homogeneous
rate is exponentiallybody, and all phases
decreasing. are mixed
The plastic in motion;
viscosity tends to the
reachrheological
a stable
properties
value of the
for shear cement
rates higherslurry
than and standard
60 r/s, foam
and these are as
stable shown
values of in
theTable 1. viscosity for the cement
plastic
slurry and the standard foam are 0.41 and 0.08 Pa·s, respectively. In the present CFD simulations,
these values are used Tableas
1. the
Theaverage
rheological properties
values for theof rheological
the cement slurry and standard
properties foam.
of the materials. However,
in reality, slurry is not aFluid
homogeneous
Type fluid. Therefore, in the simulations, each
Yield Stress τ0 (Pa) Plastic Viscosity μ (Pa·s) phase is treated as
a homogeneous body, Cement and allslurry
phases are mixed68.23in motion; the rheological
0.41 properties of the cement
slurry and standard foam are as
Standard foamshown in Table 1.
15.54 0.08

Figure5.5.Rheological
Figure Rheologicalproperties of the
properties cement
of the slurry
cement and standard
slurry foam:foam:
and standard (a) evolution of shearofstress
(a) evolution shear
with share
stress withrate andrate
share (b) and
evolution of plastic
(b) evolution ofviscosity with share
plastic viscosity rate.
with share rate.

Table 1. The rheological properties of the cement slurry and standard foam.

Fluid Type Yield Stress τ 0 (Pa) Plastic Viscosity µ (Pa·s)

Cement slurry 68.23 0.41
Standard foam 15.54 0.08

3.2.2. Slump
The method used to perform the laboratory experiments conducted in this study is depicted in
Figure 6a. Using this device, the slump behaviors of TFSB materials with different bubble volume
fractions were detected, and the results are shown in Figure 6b. As shown in Figure 6b, the slump
increases with the increasing of bubble volume fraction, which shows that the lubrication of bubbles is
Minerals 2017, 7, 149 8 of 19

3.2.2. Slump
The method used to perform the laboratory experiments conducted in this study is depicted in
Figure 6a. Using
Minerals 2017, 7, 149 this device, the slump behaviors of TFSB materials with different bubble volume 8 of 19
fractions were detected, and the results are shown in Figure 6b. As shown in Figure 6b, the slump
increases with the increasing of bubble volume fraction, which shows that the lubrication of
helpful to
bubbles is improve
helpful to theimprove
fluidity the
of slurry. Generally,
fluidity of slurry.the slump of the
Generally, paste is from
slump 15 to 25
of paste is cm.
fromWhen
15 tothe
25
slump is between 15 and 22 cm, the slurry is a small slump paste; in such a
cm. When the slump is between 15 and 22 cm, the slurry is a small slump paste; in such a case, thecase, the dehydration
equipment and
dehydration pumpingand
equipment equipment
pumpingisequipment
necessary. isEspecially
necessary.when the slump
Especially whenisthelessslump
than 20 cm,
is less
industrial applications are not easily made. Thus, the paste where the slump
than 20 cm, industrial applications are not easily made. Thus, the paste where the slump was was between 22 and
25 cm was
between 22widely
and 25 used
cm was andwidely
was called
used the
and“big
was slump
called paste”
the “big[33]. Figure
slump 6b shows
paste” that, when
[33]. Figure the
6b shows
bubble
that, whenvolume fraction
the bubble is 15, 20,
volume and 25
fraction vol20,
is 15, %, and
the 25
slump is 23.4,
vol %, 23.8, and
the slump 24.923.8,
is 23.4, cm, and
respectively.
24.9 cm,
The slurry has good fluidity and meets the requirements of the big slump
respectively. The slurry has good fluidity and meets the requirements of the big slump paste.paste.

Figure
Figure 6.
6. Slump
Slump tests:
tests: (a)
(a) method
method and
and (b)
(b) results.
results.

3.2.3.
3.2.3. Strength
Strength
The
The strength
strength ofof specimens
specimens was
was tested
tested with
with bubble
bubble volume
volume fractions
fractions from
from 00 to
to 40
40 vol
vol %,
%, and
and the
the
results after 7 and 28 days are shown in Figure 7. As shown in this figure, the strength
results after 7 and 28 days are shown in Figure 7. As shown in this figure, the strength decreases decreases
with
with the increase of
the increase of the
the bubble
bubble volume
volume fraction,
fraction, which
which shows
shows that
that the
the bubbles
bubbles areare of no advantage
of no advantage
in
in terms of improving the strength. The trend lines of strength with the bubble volume fraction are
terms of improving the strength. The trend lines of strength with the bubble volume fraction are
illustrated in Figure 7, and there is a linear relationship between the strength of 7 days
illustrated in Figure 7, and there is a linear relationship between the strength of 7 days and the bubble and the
bubble
volumevolume
fraction.fraction.
However, However,
for the for the strength
strength of 28there
of 28 days, days,isthere
a rapidis adecrease
rapid decrease
when the when the
bubble
volume fraction exceeds 20 vol %. The strength of backfill is an important factor to ensure the safety of
stopes [34].
 Minerals 2017, 7, 149 9 of 19

bubble volume
Minerals 2017, 7, 149fraction exceeds 20 vol %. The strength of backfill is an important factor to ensure
9 of 19
the safety of stopes [34].

Figure
Figure 7.
7. Results
Results of
of strength
strength tests:
tests: (a)
(a) method
method and
and (b)
(b) results.
results.

The foam can improve the fluidity of slurry, but the excessive foam is harmful to ensure the
The foam can improve the fluidity of slurry, but the excessive foam is harmful to ensure the
strength of backfill. Therefore, the bubble volume fraction from 15 to 20 vol % is a better choice.
strength of backfill. Therefore, the bubble volume fraction from 15 to 20 vol % is a better choice.
3.3. Effect of Bubble Volume Fraction
3.3. Effect of Bubble Volume Fraction
During the process of preparing TFSB, the quantity of the standard foam is measured based on
During the process of preparing TFSB, the quantity of the standard foam is measured based on
the bubble volume fraction, which indicates the volume fraction of the standard foam. In this work,
the bubble volume fraction, which indicates the volume fraction of the standard foam. In this work,
the influences of different bubble volume fractions and different average bubble diameters on the
the influences of different bubble volume fractions and different average bubble diameters on the
pipeline performance of foam slurry are simulated and studied. The basic properties of the cement
pipeline performance of foam slurry are simulated and studied. The basic properties of the cement
slurry and the standard foam are shown in Table 2.
slurry and the standard foam are shown in Table 2.
Table 2. The basic properties of the cement slurry and the standard foam.
Table 2. The basic properties of the cement slurry and the standard foam.
Yield Plastic
Cement-to-Sand Mass Fraction Density Bubble Volume Bubble
Fluid Type Mass Bubble Stress
Yield Viscosity μ
Plastic
Cement-to-Sand
Ratio (wt %) Density
(kg/m3) Bubble Volume
Fraction (vol %) Diameter
Fluid Type Fraction Diameter(μm)Stress
τ0τ(Pa)
0 Viscosity
(Pa·s)
Ratio (kg/m3 ) Fraction (vol %)
(wt %) (µm) (Pa) µ (Pa·s)
Cement slurry 1:5 70 1750 * * 68.23 0.41
Cement
Standard foamslurry * 1:5 70
* 1750
50 *
0–40 *
30–100 68.23
15.54 0.410.08
Standard foam * * 50 0–40 30–100 15.54 0.08
* Represent no corresponding parameters.
* Represent no corresponding parameters.

3.3.1. Foam Phase

3.3.1. Foam Phase
To study the effects of different amounts of foam addition on the slurry properties, nine
To study the effects of different amounts of foam addition on the slurry properties, nine simulated
simulated cases are considered in this work, with bubble volume fractions of 0, 5, 10, 15, 20, 25, 30,
cases are considered in this work, with bubble volume fractions of 0, 5, 10, 15, 20, 25, 30, 35, and 40 vol %.
35, and 40 vol %. Let us assume that the cement slurry and standard foam maintain a constant
Let us assume that the cement slurry and standard foam maintain a constant viscosity before mixing.
Minerals 2017, 7, 149 10 of 19
Minerals 2017, 7, 149 10 of 19

viscosity
The valuesbefore
of the mixing. The values
plastic viscosity of theviaplastic
obtained viscosity
theoretical obtained
calculation via theoretical
(Equation calculation
(4)) and simulation
(Equation
are shown (4)) and 3.simulation
in Table are shown
Analysis shows in simulated
that the Table 3. Analysis
viscosityshows
valuesthat the simulated
are close viscosity
to the theoretically
values are close to the theoretically calculated values, with a relative error between them
calculated values, with a relative error between them of no more than 0.4%. Thus, the mixture is of noevenly
more
than 0.4%. Thus, the mixture
mixed during pipeline transport. is evenly mixed during pipeline transport.

Table 3. The
Table 3. The value
value of
of plastic
plastic viscosity
viscosity by
by theoretical
theoretical calculation
calculation and
and simulation.
simulation.

Studied Bubble Volume Plastic Viscosity μ (Pa·s)

Bubble Volume Plastic Viscosity µ (Pa·s) Relative Error (%)
Case
Studied Case Fraction (vol %) Theoretical Calculation Simulation Relative Error (%)
1 Fraction
0 (vol %) Theoretical
0.4100 Calculation 0.4116Simulation 0.39
12 5 0 0.3935
0.4100 0.3902
0.4116 0.190.39
23 10 5 0.3770
0.3935 0.3690
0.3902 0.240.19
34 15 10 0.3770
0.3605 0.3690
0.3491 0.170.24
45 20 15 0.3605
0.3440 0.3491
0.3287 0.210.17
56 25 20 0.3440
0.3275 0.3287
0.3080 0.140.21
67 30 25 0.3275
0.3110 0.3080
0.2878 0.250.14
7 30 0.3110 0.2878 0.25
8 35 0.2945 0.2667 0.05
8 35 0.2945 0.2667 0.05
9 40 0.2780 0.2462 0.05
9 40 0.2780 0.2462 0.05

The foam phase in TFSB exists in the form of bubbles. The simulations show that the bubbles
The foam
maintain phase distribution
a uniform in TFSB exists in in thepipeline.
the form of In bubbles. The simulations
the vertical section of show that the
the pipe, the bubbles
TFSB is
maintain a uniform distribution in the pipeline. In the vertical section of
mainly affected by gravity and the wall viscosity. Because of the low air density, the bubbles the pipe, the TFSB is mainly
are
affected by gravity and the wall viscosity. Because of the low air density, the
easily pushed toward the wall of the pipe, which causes the bubble volume fraction to be small in bubbles are easily pushed
toward the wall
the center of the
of the pipe,At
pipe. which
the causes
elbow,the bubble
the foamvolume slurry fraction to be small
is influenced by in the center of
centrifugal the
force;
pipe. At the elbow,
consequently, the foam
the bubble slurryfraction
volume is influenced by centrifugal
distribution presentsforce; consequently,
an overall the bubble
decreasing trend fromvolume
the
fraction
inside to the outside of the elbow, and the bubble volume fraction distribution is not stableelbow,
distribution presents an overall decreasing trend from the inside to the outside of the with
and the bubble
variations in thevolume fraction
slurry flow. In distribution
the horizontal is not stable
section of with variations
the pipe, the TFSB in the slurry flow.
is mainly In the
affected by
horizontal section of the pipe, the TFSB is mainly affected by buoyancy. After
buoyancy. After long-distance transport, the slurry maintains a stable flow, and the bubble volume long-distance transport,
the slurrydistribution
fraction maintains ashowsstable aflow, and the trend
decreasing bubblefromvolume the fraction distribution
top to the bottom ofshows a decreasing
the pipe, but the
trend from is
difference thetoo
topsmall.
to the bottom
Hence, of thethebubble
pipe, but the difference
volume fraction isistoo small.constant,
almost Hence, the andbubble volume
the pipeline
fraction
transport is almost
of the constant,
foam slurry and the pipeline transport
is feasible. The bubble of the foam slurry
volume is feasible.
fraction The bubble
distribution for a volume
bubble
fraction distribution for a bubble volume fraction of 15 vol %, which
volume fraction of 15 vol %, which follows the identified bubble volume fraction distribution follows the identified bubble
law,
volume fraction distribution
is shown in Figure 8. law, is shown in Figure 8.

Figure 8. Cont.
Minerals 2017, 7, 149 11 of 19
Minerals 2017, 7, 149 11 of 19

Figure
Figure 8.8. Bubble
Bubblevolume
volumefraction
fraction (unit
(unit in vol
in vol %) distribution
%) distribution in theinpipe
thefor
pipe for a volume
a bubble bubble fraction
volume
fraction
of 15 volof
%:15(a)
vol %: (a)
inlet, (b)inlet,
elbow,(b)and
elbow, and (c) outlet.
(c) outlet.

3.3.2. Pressure
3.3.2. Pressure
To realize self-flowing transport of a backfill material through a pipeline, the gravitational
To realize self-flowing transport of a backfill material through a pipeline, the gravitational potential
potential energy must be greater than the resistance loss of the pipeline, which is defined as the
energy must be greater than the resistance loss of the pipeline, which is defined as the total pressure
total pressure difference between the inlet and outlet of the pipeline. The equations for the
difference between the inlet and outlet of the pipeline. The equations for the gravitational potential
gravitational potential energy and resistance loss are as follows [35]:
energy and resistance loss are as follows [35]:
p = ρ gh (8)
p = ρgh (8)
h f = p y = pin − pout (9)
h f = py = pin − pout (9)
where p is the gravitational potential energy (Pa), g is the acceleration of gravity (i.e., 9.82m/s2),
where p is the gravitational potential energy (Pa), g is the acceleration of gravity (i.e., 9.8 m/s ), h is
h is the height difference (m), hf is the resistance loss (Pa), py is the total pressure difference (Pa),
the height difference (m), h f is the resistance loss (Pa), py is the total pressure difference (Pa), pin is the
p
total
in
ispressure at the inlet (Pa), and (Pa),
the total pressure at the inlet pout isand pout ispressure
the total the totalatpressure
the outlet at (Pa).
the outlet (Pa).
Sixty points
points along
alongthe themiddle
middleline lineofofthethe pipe
pipe werewere chosen
chosen to investigate
to investigate the evolution
the evolution of theoftotal
the
total pressure.
pressure. As shownAs shown
in Figure in Figure 9, the
9, the total total pressure
pressure in the pipeline
in the pipeline decreasesdecreases fromtothe
from the inlet theinlet to
outlet,
the
and outlet, and
the total the total
pressure valuepressure
at a given value at aincreases
point given pointwith increases
an increase with an bubble
in the increase in thefraction.
volume bubble
volume fraction.
In simulations, theIninlet
simulations,
is supposed thetoinlet is supposed
be connected to theto atmosphere,
be connectedand to the atmosphere,pressure
an atmospheric and an
atmospheric
at the inlet is pressure
set as theat the inlet
reference is set as
pressure. the with
Thus, reference pressure.
the decrease of Thus,
pressure,with the isdecrease
there a negative of
pressure,atthere
pressure is a negative
the latter part andpressure
the outletatofthe thelatter part Table
pipeline. and the outlet
4 lists theof theand
inlet pipeline.
outlet Table 4 lists
pressures in the
inlet andfor
pipeline outlet pressures
different bubbleinvolumethe pipeline
fractions forand
different
shows bubble
that thevolume fractions
total resistance andand showsresistance
average that the
total
of theresistance and average
pipe decrease resistance
as the bubble of thefraction
volume pipe decrease
increases. as the
Whenbubblethe volume fraction fraction
bubble volume increases. is
When
greater thethanbubble
15 vol %, volume
the total fraction
reduction is greater
in resistancethan compared
15 vol %,with thethe total reduction
cement in resistance
slurry alone exceeds
compared
0.22 MPa, whichwith the cement slurry
corresponds alone exceeds
to a proportional 0.22 MPa,
reduction of morewhich
thancorresponds
12%. In addition, to a theproportional
reduction
reduction
in the average of more than 12%.
resistance exceeds In addition,
0.36 kPa/m, the also
reduction in the average
corresponding resistance exceeds
to a proportional reduction 0.36
of
kPa/m,
more than also12%.
corresponding
The results to thusa proportional
indicate thatreduction
the addition of more
of foam thancan12%.
reduceThethe
results thus indicate
resistance during
that
slurry thetransport
additionand of foam
improvecan reduce the resistance
the fluidity of the slurry.during slurry transport
Moreover, when theand improve
bubble volumethe fraction
fluidity
of the slurry.
exceeds 15 volMoreover,
%, the total when the bubble
resistance of thevolume
pipelinefraction exceeds
is less than 15 vol %, thepotential
the gravitational total resistance
energy of
the pipeline
1.72 is lessthe
MPa, allowing than the gravitational
realization potential
of self-flowing energyAsofshown
transport. 1.72 MPa, allowing
in Figure the is
10, there realization
a non-linear of
self-flowing transport.
relationship between theAs shown
total in Figure
resistance and the 10, bubble
there isvolume
a non-linear
fraction,relationship
according to between the total
the polynomial,
resistance
the critical and the volume
bubble bubble volume
fraction fraction,
is 12 vol %. according to the polynomial, the critical bubble volume
fraction is 12 vol %.
 2 5 0.28 -1.56 1.84 3.06 1.72 No
3 10 0.27 -1.51 1.78 2.96 1.72 No
4 15 0.25 -1.39 1.64 2.73 1.72 Yes
5 20 0.24 -1.33 1.57 2.60 1.72 Yes
6 25 0.22 -1.23 1.45 2.42 1.72 Yes
7 30 0.21 -1.14 1.35 2.25 1.72 Yes
Minerals
8 2017, 7, 149
35 0.19 -1.07 1.26 2.09 1.72 12 of 19
Yes
9 40 0.18 -1.04 1.22 2.02 1.72 Yes

Minerals 2017, 7, 149 12 of 19

Table 4. The resistances in the various studied cases.

Average Gravitational
Studied Bubble Volume Inlet Pressure Outlet Pressure Resistance Self-Flowing
Resistance Potential Energy
Case Fraction (vol %) (MPa) (MPa) (MPa) Transport
(kPa/m) (MPa)
1 0 0.0044 -1.86 1.86 3.09 1.72 No
2 5 0.28 -1.56 1.84 3.06 1.72 No
3 10 0.27 -1.51 1.78 2.96 1.72 No
4 15 0.25 -1.39 1.64 2.73 1.72 Yes
5 20 0.24 -1.33 1.57 2.60 1.72 Yes
6 25 0.22 -1.23 1.45 2.42 1.72 Yes
7 30 0.21 -1.14 1.35 2.25 1.72 Yes
8 35 0.19 -1.07 1.26 2.09 1.72 Yes
9 40 0.18 -1.04 1.22 2.02 1.72 Yes

Figure 9. Total pressure profiles for various bubble volume fractions.

Figure 9. Total pressure profiles for various bubble volume fractions.

Table 4. The resistances in the various studied cases.

Gravitational
Studied Bubble Volume Inlet Pressure Outlet Pressure Resistance Average Resistance Self-Flowing
Potential
Case Fraction (vol %) (MPa) (MPa) (MPa) (kPa/m) Transport
Energy (MPa)
1 0 0.0044 −1.86 1.86 3.09 1.72 No
2 5 0.28 −1.56 1.84 3.06 1.72 No
3 10 0.27 −1.51 1.78 2.96 1.72 No
4 15 0.25 −1.39 1.64 2.73 1.72 Yes
5 20 0.24 −1.33 1.57 2.60 1.72 Yes
6 25 0.22 −1.23 1.45 2.42 1.72 Yes
7 30 0.21 −1.14 1.35 2.25 1.72 Yes
8 35 0.19 −1.07 1.26 2.09 1.72 Yes
9 40 0.18 −1.04 1.22 2.02 1.72 Yes
Figure 9. Total pressure profiles for various bubble volume fractions.

Figure 10. Relationship between total resistance and bubble volume fraction.

Table 5 compares the pressures in the pipeline and the elbow for the various bubble volume
fractions. As shown in Table 5, the total and average resistances of the elbow also decrease as the
bubble volume fraction increases. Moreover, when the bubble volume fraction is greater than
15 vol %, the total resistance reduction at the elbow compared with the cement slurry alone exceeds
0.0012 MPa, corresponding to a proportional reduction of more than 14%, whereas the reduction in

Figure 10. Relationship between total resistance and bubble volume fraction.
Figure 10. Relationship between total resistance and bubble volume fraction.
Table 5 compares the pressures in the pipeline and the elbow for the various bubble volume
Table 5 compares
fractions. As shownthe pressures
in Table 5, the in theand
total pipeline
averageand the elbow
resistances forelbow
of the the various bubble
also decrease as volume
the
fractions.
bubbleAsvolume
shownfraction
in Table 5, the total
increases. and average
Moreover, resistances
when the of the elbow
bubble volume fractionalso decrease
is greater thanas the
bubble15volume
vol %, the total resistance
fraction reduction
increases. at the
Moreover, elbow
when thecompared with thefraction
bubble volume cement slurry alone
is greater exceeds
than 15 vol %,
0.0012
the total MPa, corresponding
resistance reduction attothea proportional reduction
elbow compared withofthe
more than 14%,
cement slurrywhereas the reduction
alone exceeds 0.0012inMPa,
Minerals 2017, 7, 149 13 of 19

corresponding to a proportional reduction of more than 14%, whereas the reduction in the average
resistance exceeds 0.72 kPa/m, corresponding to a proportional reduction of more than 13%. The results
indicate that, although the total resistance of the pipe elbow is less than 1% of that of the pipeline,
Minerals 2017, 7, 149 13 of 19
the average resistance of the elbow is 2 to 3 times that of the pipeline as a whole. This leads to a rapid
increase in the local
the average resistance,
resistance exceedsresulting in increased
0.72 kPa/m, resistance
corresponding at the elbow.
to a proportional reduction of more than
13%. The results indicate that, although the total resistance of the pipe elbow is less than 1% of that
Table
of the pipeline, 5. Comparison
the average between
resistance theelbow
of the pressures in the
is 2 to pipeline
3 times thatand the pipeline
of the elbow. as a whole.
This leads to a rapid increase in the local resistance, resulting in increased resistance at the elbow.
Elbow Pipe
Studied Bubble Volume
Table 5. Comparison between
Resistance the pressures
Average in the pipeline
Resistance and the elbow.
Average Resistance W (%)
Case Fraction (vol %)
(MPa) Resistance kPa/m) (MPa) (kPa/m)
Elbow Pipe
1Studied Bubble Volume
0 0.0086
Resistance 5.46
Average Resistance 1.86
Resistance 3.09
Average Resistance 43.41
W (%)
Case Fraction (vol %)
2 5 0.0083
(MPa) 5.30
kPa/m) 1.84
(MPa) 3.06
(kPa/m) 42.26
3 1 100 0.0077
0.0086 4.88
5.46 1.78
1.86 3.092.96 43.4139.34
4 2 155 0.0074
0.0083 4.74
5.30 1.64
1.84 3.062.73 42.2642.41
5 3 2010 0.0070
0.0077 4.46
4.88 1.57
1.78 2.962.60 39.3441.70
6 4 2515 0.0074
0.0066 4.74
4.19 1.64
1.45 2.732.42 42.4142.24
7 5 3020 0.0070
0.0062 4.46
3.94 1.57
1.35 2.602.25 41.7042.89
8 6 3525 0.0066
0.0056 4.19
3.59 1.45
1.26 2.422.09 42.2441.78
9 7 4030 0.0062
0.0051 3.94
3.24 1.35
1.22 2.252.02 42.8937.65
8 35 0.0056 3.59 1.26 2.09 41.78
9 W represents
40 the percentage
0.0051 of pipe average
3.24 resistance in1.22
elbow average resistance.
2.02 37.65
W represents the percentage of pipe average resistance in elbow average resistance.
3.3.3. Velocity
3.3.3. Velocity
Control of the transport velocity of the slurry is also an important factor in pipeline transport [36].
When theControl
velocity of is
thetoo
transport velocity of
low, settlement of the
the slurry is also anwill
solid particles important
occur. Byfactor in pipeline
contrast, whentransport
the velocity
[36].
is too When
high, the the velocity
stress inducedis too
bylow, settlement
the slurry of the
on the pipesolid
wallparticles will occur.
is enhanced, By frequency
and the contrast, when
withthe
which
velocity
the solid is too high,
particles impingethe stress
on theinduced by the
pipe wall slurry onwhich
increases, the pipe
canwall is enhanced,
easily andwall
to cause the the frequency
of the pipe to
wear with
andwhich
eventhe solid particles
deform. impingeof
The velocities onthe
theslurry
pipe wall
for increases, which can
various bubble easily fractions
volume to cause the
arewall
shown
of the pipe to wear and even deform. The velocities of the slurry for various bubble volume
in Figure 11. As shown in this figure, the maximum velocity among all considered cases is 4.68 m/s
fractions are shown in Figure 11. As shown in this figure, the maximum velocity among all
for a bubble volume fraction of 30 vol %, and the minimum velocity is 3.91 m/s for a bubble volume
considered cases is 4.68 m/s for a bubble volume fraction of 30 vol %, and the minimum velocity is
fraction of 35forvol
3.91 m/s %. These
a bubble velocities
volume doof
fraction not
35vary
vol %.monotonically with
These velocities do the
not bubble volume fraction
vary monotonically with but
satisfy the requirements for pipeline transport.
the bubble volume fraction but satisfy the requirements for pipeline transport.

Figure11.
Figure 11.Maximum
Maximum velocities
velocitiesininthe
thesimulations.
simulations.

The velocity distribution for a bubble volume fraction of 15 vol % is shown in Figure 12, which
The velocity
shows an obviousdistribution for ainbubble
flow core zone volume
the vertical fraction of
and horizontal 15 volof%theispipe.
sections shown in elbow,
In the Figure 12,
which shows the
however, an obvious
flow coreflow core
zone zone inbecomes
gradually the vertical and and
smaller horizontal
moves sections
from the of the pipe.
center of theInpipe
the elbow,
to
however, the flow
the outside core
of the zoneLet
elbow. gradually becomes
us consider smaller
the inlet sectionand
(y = moves
1 m), thefrom
elbowthesection
center(yof= the pipethe
x), and to the
outside
outletofsection
the elbow.
(x = 1 Let us consider
m) separately the inlet
to study the section
change in(y the
= 1velocity
m), the on elbow (y = x),atand
sectionsurface
the vertical the the
elbow
outlet in Case
section (x =4.1 As
m) shown in Figure
separately 13, the
to study theflow corein
change zone
the is shifted on
velocity toward the outside
the vertical of the
surface at the
elbow, and the shape of the flow core zone changes from an “O” shape to a “U” shape. Moreover,
Minerals 2017, 7, 149 14 of 19

Minerals 2017, 7, 149 14 of 19

elbow in Case 4. As shown in Figure 13, the flow core zone is shifted toward the outside of the elbow,
thereMinerals
is a high-velocity
2017, 7, 149 area concentrated at the outside of the elbow, and the shear stress on19 the
14 of
and the shape of the flow core zone changes from an “O” shape to a “U” shape. Moreover, there is
wall increases, which can easily cause wear.
a high-velocity area concentrated at the outside of the elbow, and the shear stress on the wall increases,
there is a high-velocity area concentrated at the outside of the elbow, and the shear stress on the
which
wallcan easily cause
increases, whichwear.
can easily cause wear.

Figure 12. Velocity distributions in the pipe for a bubble volume fraction of 15 vol %: (a) inlet,
Figure
Figure 12. Velocity distributions inin
thethe
pipe for afor
bubble volume fraction of 15 vol
of %:
15 (a)
volinlet, (b) elbow,
(b) 12. Velocity
elbow, and (c)distributions
outlet. pipe a bubble volume fraction %: (a) inlet,
and (c) outlet.
(b) elbow, and (c) outlet.

Figure 13. Cont.

Minerals 2017, 7, 149 15 of 19
Minerals 2017, 7, 149 15 of 19
Minerals 2017, 7, 149 15 of 19

Figure 13. Velocity contours at the elbow for a bubble volume fraction of 15 vol %: (a) the 3D total
Velocity contours
Figure 13.contours
pressure at the
(b) elbow forsection
a bubble(y volume fraction of 15 section
vol %: (a)
(y =the
x), 3D
andtotal
Figure 13. Velocityatcontours
the elbow,
at the the inlet
elbow for a bubble = 1 m), (c)
volume the elbow
fraction of 15 vol %: (a) the 3D (d)
total
pressure
the outletcontours
section at
(x =the
1 elbow,
m). (b) the inlet section (y = 1 m), (c) the elbow section (y = x), and (d) the
pressure contours at the elbow, (b) the inlet section (y = 1 m), (c) the elbow section (y = x), and (d)
outlet section (x = 1 m).
the outlet section (x = 1 m).
3.4. Effect of Bubble Diameter
3.4. Effect of Bubble Diameter
3.4. Effect of Bubble
The bubble Diameter
diameter is an important parameter affecting the performance of a foam [37]. When
The
the diameterbubble diameter
is too is an important
small, isthe parameter affecting the performance of a foam [37]. When the
The bubble diameter anbubbles
important cannot effectively
parameter adsorb
affecting solid particles.
the performance of By contrast,
a foam when
[37]. When
diameter
the is too small, the bubbles cannot effectively adsorb solid particles. By contrast, when the
the diameter
diameter is is too
too large,
small, athe defoaming
bubbles cannotphenomenon
effectivelycan easily
adsorboccur. Therefore,By
solid particles. it is necessary
contrast, to
when
diameter is
simulate thetooperformance
large, a defoaming of foamphenomenon
slurry can easily
pipeline occur. for
transport Therefore,
various it bubble
is necessary to simulate
diameters. The
the diameter is too large, a defoaming phenomenon can easily occur. Therefore, it is necessary to
the performance
diameter ofisfoam slurry pipeline transport forlevels
various bubble diameters. The diameter 30,of foam
simulate of thefoam 30–100
performance μm; thus, slurry
of foam five different
pipeline were
transport selected for the
for various simulations:
bubble diameters. 47.5,
The
is
65,30–100
82.5, andµm; thus,
100 μm. five different levels were selected for the simulations: 30, 47.5, 65, 82.5, and 100 µm.
diameter of foam is 30–100 μm; thus, five different levels were selected for the simulations: 30, 47.5,
The resistance
The resistance and velocity results are shown in Figure 14. As shown in this figure, the diameter
65, 82.5, and 100 μm.and velocity results are shown in Figure 14. As shown in this figure, the
has little
diameter influence on the resistance
has little influence on the inresistance
the pipe. in The theimpact
pipe. on
Thetheimpact
velocityonisthe
slightly larger,
velocity but the
is slightly
The resistance and velocity results are shown in Figure 14. As shown in this figure, the
velocity
larger, still
but thegenerally
velocity remains between
still generally 4 andbetween
remains 5 m/s, which is
4 andThe within
5 m/s, the acceptable
which range. Thus,
diameter has little influence on the resistance in the pipe. impact onisthe
within the
velocity acceptable
is slightly
the bubble
range. diameter has diameter
little effect onlittle
the transport properties of properties
TFSB.
larger, Thus,
but the thevelocity
bubble has
still generally effectbetween
remains on the transport
4 and 5 m/s, which of TFSB. the acceptable
is within
range. Thus, the bubble diameter has little effect on the transport properties of TFSB.

Figure 14. Cont.

Minerals 2017, 7, 149 16 of 19
Minerals 2017, 7, 149 16 of 19

Figure
Figure 14.
14. Resistance
Resistanceand
andvelocity
velocityresults
resultsfor
forvarious
variousbubble
bubblediameters:
diameters:(a)
(a)resistance
resistanceand
and(b)
(b)velocity.
velocity.

3.5.
3.5. Industrial
Industrial Applications
Applications
According
According to to the
the analysis
analysis of the CFD
of the CFD simulations,
simulations, the the rheological
rheological properties
properties of of TFSB
TFSB areare
predominantly affected by the bubble volume fraction. Using TFSB can effectively
predominantly affected by the bubble volume fraction. Using TFSB can effectively reduce the resistance reduce the
resistance
during slurryduring slurry
pipeline pipelineand
transport transport
the wearandonthe thewear
pipe,onandtheself-flowing
pipe, and self-flowing
transport can transport can
be realized
be realized when the bubble volume fraction
when the bubble volume fraction is higher than 15 vol %. is higher than 15 vol %.
Based
Based on the CFD
on the CFD simulations
simulations and and the
the laboratory
laboratory tests,
tests, the
the optimal
optimal material
material parameters
parameters can can
finally bedetermined
finally be determinedasasfollows:
follows: a cement-to-sand
a cement-to-sand ratio
ratio of 1:5,
of 1:5, a bubble
a bubble volume
volume fraction
fraction of 15–20of vol
15–20%,
vol %, aflow
a foam foamrate
flowof rate
8–12ofm8–12
3 /h, m 3/h, and a backfill ability of 50–60 m3/h. The number S01, N02, and
and a backfill ability of 50–60 m3 /h. The number S01, N02, and S05
S05
goafsgoafs
at theat0the 0 m in
m level level in were
mine mine chosen
were chosen as the industrial
as the industrial tests
tests area. area.
After theAfter theof
backfill backfill
the goafof and
the
goaf and the achievement
the achievement of condensation
of condensation stability, stability,
the rate of thepitrate
roofofsupport
pit roof was
support
largerwas larger
than 90%. than
The90%.
test
The
results in Table 6 show that the 7 and 28-day strengths greater than 1.2 and 2.3 MPa, respectively. MPa,
test results in Table 6 show that the 7 and 28-day strengths greater than 1.2 and 2.3 Thus,
respectively.
the TFSB showed Thus,a the
goodTFSB showed
backfill a good backfill effect.
effect.
A comparative analysis of the economic indicators for the conventional cement slurry and
TFSB shows that, although the Table use of TFSB
6. The incurs
results an additional
of industrial tests. cost for the foaming agent, it
simultaneously reduces the amount of cement consumed. As shown in Table 7, for a cement unit
price of 0.05 $/kg and a foaming agent unit price of 6 $/kg, the use of TFSBStrength saves $1.56
(MPa)per cubic
Goaf Number Density (kg/m3 ) Bubble Volume Fraction (vol %)
meter of the backfill. For mine “A”, which mines 1 million 800 thousand tons of
7 days stone annually
28 days and
has an annual backfill volume of 600 thousand cubic meters, under the assumption of a direct
S01 1496 16 1.21 2.33
savings in material
N02 cost of 1.56 $/m3, the mine would save
1402 19 220 thousand tons1.20 of cement per 2.17year and
would consequently
S05 reduce 1533
the annual backfill cost by 12936 thousand dollars, 1.38which represents
2.38 an
enormous economic benefit.
A comparative analysis of the economic indicators for the conventional cement slurry and TFSB
Table 6. The results of industrial tests.
shows that, although the use of TFSB incurs an additional cost for the foaming agent, it simultaneously
reduces the amount of cement consumed. As shown
Bubble Strengthunit
in Table 7, for a cement
Volume (MPa)
price of 0.05 $/kg
Goaf Number Density (kg/m3)
Fraction (vol %)
and a foaming agent unit price of 6 $/kg, the use of TFSB saves $1.56 per cubic meterdays
7 days 28 of the backfill.
For mine “A”,S01which mines 1 1496
million 800 thousand16tons of stone annually 1.21 and has an2.33
annual backfill
volume of 600N02thousand cubic1402
meters, under the 19assumption of a direct 1.20 savings in2.17material cost of
3 S05 1533 12 1.38 2.38
1.56 $/m , the mine would save 220 thousand tons of cement per year and would consequently reduce
the annual backfill cost by 936 thousand dollars, which represents an enormous economic benefit.
Table 7. The economic indicators for conventional backfill material and three-phase foam slurry
backfill (TFSB).

Project Specifications Conventional Cement Slurry TFSB

Cement consumption (kg/m3) 288 240
Foam agent consumption (kg/m3) 0 0.14
Unit cost ($/m3) 14.4 12.84
Cost savings ($/m3) 1.56
Minerals 2017, 7, 149 17 of 19

Table 7. The economic indicators for conventional backfill material and three-phase foam slurry
backfill (TFSB).

Project Specifications Conventional Cement Slurry TFSB

Cement consumption (kg/m3 ) 288 240
Foam agent consumption (kg/m3 ) 0 0.14
Unit cost ($/m3 ) 14.4 12.84
Cost savings ($/m3 ) 1.56

4. Conclusions
This paper develops a three-dimensional CFD transport model of a backfill pipeline, which is
used to study the pipeline transport properties of TFSB. Simulated results indicate that TFSB can
maintain a steady state during pipeline transport, experiences a markedly reduced pipeline transport
resistance and exhibits better liquidity than conventional cement slurry.
Furthermore, a multiphase flow model is established using the mixture method in FLUENT to
study the evolutions of the foam phase, pressure, and velocity in the pipeline system for different
bubble volume fractions and bubble diameters. The results show that, as the bubble volume fraction
increases, the resistance of the pipeline decreases and the fluidity improves. By contrast, the bubble
diameter has little effect on the transport properties of TFSB.
The combined results of the CFD simulations, slump tests, and strength tests indicate that,
when the bubble volume fraction is 15–20 vol %, TFSB can satisfy the necessary strength requirements,
good fluidity, and exhibit self-flowing transport. Furthermore, TFSB represents an enormous
economic benefit.
Finally, although the CFD technique provides an intuitive and accurate basis for pipeline transport
research with regard to TFSB, several factors were not addressed in the design of this study, including
temperature, free fall, and the solid particle phase. Therefore, a more comprehensive study of the
problem will be conducted in future work.

Acknowledgments: The authors would like to acknowledge the financial support from the National Key R & D
Program of China (Grant No. 2017YFC0602902), the Innovation-Driven Plan of Central South University of China
(Grant No. 2015CX005), and the Fundamental Research Funds for the Central Universities of Central South
University (No. 2016zztx096) for their financial support.
Author Contributions: X.C., Q.C. and X.S. conceived and designed the experiments; X.C. and Y.G. performed
the experiments; X.C. and J.Z. performed the simulations; X.C., J.Z. and Q.C. analyzed the data; X.C. and X.S.
contributed reagents/materials/analysis tools; X.C. wrote the paper.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Li, S.; Wang, X.M. Fly-ash-based magnetic coagulant for rapid sedimentation of electronegative slimes and
ultrafine tailings. Powder Technol. 2016, 303, 20–26. [CrossRef]
2. Liang, C.; Fall, M. Multiphysics modeling of arching effects in fill mass. Comput Geotech. 2017, 83, 114–131.
3. Wu, D.; Yang, B.G.; Liu, Y.C. Transportability and pressure drop of fresh cemented coal gangue-fly ash
backfill (CGFB) slurry in pipe loop. Powder Technol. 2015, 284, 218–224. [CrossRef]
4. Chen, Q.S.; Zhang, Q.L.; Wang, X.M.; Xiao, C.C.; Hu, Q. A hydraulic gradient model of paste-like crude
tailings backfill slurry transported by a pipeline system. Environ. Earth Sci. 2016, 75. [CrossRef]
5. Zhang, Q.L.; Chen, Q.S.; Wang, X.M. Cemented Backfilling Technology of Paste-Like Based on Aeolian Sand
and Tailings. Minerals 2016, 6, 132. [CrossRef]
6. Belem, T.; Fourie, A.B.; Fahey, M. Time-dependent failure criterion for cemented paste backfills. Richard
Jewell and Andy Fourie. In Proceedings of the 13th International Seminar on Paste and Thickened Tailings,
Perth: Australian Centre for Geomechanics, Australian, 3–6 May 2010; pp. 147–162.
Minerals 2017, 7, 149 18 of 19

7. Grabinsky, M.W. In situ monitoring for ground trothing paste backfill design. In Proceedings of the 13th
International Seminar on Paste and Thickened Tailings, Toronto, ON, Canada, 3–6 May 2010; pp. 85–98.
8. Zhou, J.; Li, X.B.; Mitri, H.S. Classification of rockburst in undergroud projects: Comparison of ten supervised
learning methods. J. Comput. Civ. Eng. 2016, 30, 04016003. [CrossRef]
9. Ju, F.; Zhang, J.X.; Zhang, Q. Vertical transportation system of solid material for backfilling coal mining
technology. Int. J. Min. Sci. Technol. 2012, 22, 41–45. [CrossRef]
10. Wu, D.; Sun, G.H.; Liu, Y.C. Modeling the thermo-hydro-chemical behavior of cemented coal gangue-fly ash
backfill. Constr. Build. Mater. 2016, 111, 658–662. [CrossRef]
11. Marradi, A.; Pinori, U.; Betti, G. The use of lightweight materials in road embankment construction. Proc. Soc.
Behav. Sci. 2012, 53, 1000–1009. [CrossRef]
12. Zhang, Y.P.; Su, X.P.; Tan, W.; Yang, S.Y.; Guo, W.J. The application value of foam mortar filling body.
Met. Mine 2013, 10, 40–42. (In Chinese)
13. Chen, Z.P.; Zhai, S.H.; Gao, X.; Dong, L. Research on the mechanical prosperities of foam mortar backfill and
its application. Met. Mine 2010, 8, 7–10. (In Chinese)
14. Zhang, W. Evaluation of effect of viscosity changes on bubble size in a mechanical flotation cell.
Trans. Nonferrous Met. Soc. China 2014, 24, 2964–2968. [CrossRef]
15. Yan, K.L.; Sun, C.Y.; Chen, J.; Chen, L.T.; Shen, D.J.; Liu, B.; Jia, M.L.; Niu, M.; Lv, Y.N.; Li, N.;
et al. Flow characteristics and rheological properties of natural gas hydrate slurry in the presence of
anti-agglomerant in a flow loop apparatus. Chem. Eng. Sci. 2014, 106, 99–108. [CrossRef]
16. Bakshi, A.; Altantzis, C.; Bates, R.B.; Ghoniem, A.F. Study of the effect of reactor scale on fluidization
hydrodynamics using fine-grid CFD simulations based on the two-fluid model. Powder Technol. 2016, 299,
185–198. [CrossRef]
17. Bisognin, P.C.; Fusco, J.M.; Soares, C. Heat transfer in fluidized beds with immersed surface: Effect of
geometric parameters of surface. Powder Technol. 2016, 297, 401–408. [CrossRef]
18. Eesa, M.; Barigou, M. CFD analysis of viscous non-Newtonian flow under the influence of a superimposed
rotational vibration. Comput. Fluids 2008, 37, 24–34. [CrossRef]
19. Shannag, M.J. Characteristics of lightweight concrete containing mineral admixtures. Constr. Build. Mater.
2011, 25, 658–662. [CrossRef]
20. Chen, Z.P. A New Type of High Efficient Foaming AGENT for Geotechnical Engineering. Chinese Patent ZL
200610033825.9, 27 February 2006.
21. GB/T18046-2008 (MOD, ASTM C989M). In Ground Granulated Blast Furnace Slag Used for Cement and Concrete;
General Administration of Quality Supervision, Inspection and Quarantine of PR China: Beijing, China, 2008.
22. Gao, J.; Fourie, A. Spread is better: An investigation of the mini-slump test. Miner. Eng. 2015, 71, 120–132.
[CrossRef]
23. B.F Inc. Fluent User’s Guide; Fluent Incorporated: Lebanon, NH, USA, 2010.
24. Hata, K.; Masuzaki, S. Critical heat fluxes of sub cooled water flow boiling in a short vertical tube at high
liquid Reynolds number. Nucl. Eng. Des. 2010, 240, 3145–3157. [CrossRef]
25. Chen, X.; Shi, X.Z.; Zhou, J.; Chen, Q.S.; Yang, C. Feasibility of Recycling Ultrafine Leaching Residue by
Backfill: Experimental and CFD Approaches. Minerals 2017, 7, 54. [CrossRef]
26. Zhao, G.H. Simulating and Experiment Research of Resistance Characteristics of coal-Water-Slurry.
Master’s Thesis, Southeast University, Nanjing, China, 2007. (In Chinese)
27. Kurnia, J.C.; Sasmito, A.P.; Mujumdar, A.S. CFD simulation of methane dispersion and innovative methane
management in underground mining faces. Appl. Math. Model. 2014, 38, 3467–3484. [CrossRef]
28. Yang, J.; Wang, X.M.; Zhang, Q.L.; Ke, Y.X. Pipeline transportation properties of high viscosity sulfur-content
filling slurry in three-phase flow. Chin. J. Nonferrous Met. 2015, 25, 1049–1055. (In Chinese)
29. Zhang, J.S.; Wang, C.Z.; Huang, L.X.; Qang, B.Q. Foam Concrete, 1st ed.; Harbin Institute of Technology Press:
Harbin, China, 2016.
30. Nambiar, E.K.K.; Ramamurthy, K. Sorption characteristics of foam concrete. Cem. Concr. Res. 2007, 37,
1341–1347. [CrossRef]
31. Bingham, E.C. Fluidity and Plasticity; McGraw-Hill: New York, NY, USA, 1922.
32. Gopala, V.R.; Nijeholt, J.A.L.À.; Bakker, P.; Haverkate, B. Development and validation of a CFD model
predicting the backfill process of a nuclear waste gallery. Nucl. Eng. Des. 2011, 241, 2508–2518. [CrossRef]
Minerals 2017, 7, 149 19 of 19

33. Slottee, J.S. Update on the application of paste thickeners for tailings disposal-mine paste backfill.
In Proceedings of the International Seminar on Paste and Thickened Tailings, Cape Town, South Africa,
31 March–2 April 2004.
34. Liu, G.S.; Li, L.; Yang, X.C.; Guo, L.J. A numerical analysis of the stress distribution in backfilled stopes
considering interfaces between backfill and rock walls. Int. J. Geomech. 2017, 17, 06016014-1–06016014-9.
[CrossRef]
35. Chen, X.; Shi, X.Z.; Zhou, J.; Qiu, X.Y.; Lin, X.F. CFD simulation of transportation properties of W type and
long distance pipeline for foam slurry backfilling. Chin. J. Nonferrous Met. 2016, 26, 1782–1793. (In Chinese)
36. Beltrán, J.P.; Ceci, P.; Miguez, P.; Casali, P. Construction of slurry pipelines. Proc. Eng. 2016, 138, 27–130.
[CrossRef]
37. Xiao, H.L. Study on the Performance of Foaming Agents Applied to the Foamed Concrete. Master’s Thesis,
Zhejiang University, Zhejiang, China, 2011. (In Chinese)

© 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).