processes

Article
A Mathematical Modeling of the Reverse Osmosis
Concentration Process of a Glucose Solution
Chenghan Chen * and Han Qin
Biomass Energy Technology Research Center, Biogas Institute of Ministry of Agriculture, Chengdu 610041,
China; qinhan@caas.cn
* Correspondence: chenchenghan@caas.cn




Received: 21 March 2019; Accepted: 27 April 2019; Published: 8 May 2019


Abstract: A mathematical modeling of glucose–water separation through a reverse osmosis (RO)

membrane was developed to research the membrane’s performance during the mass transfer process.
The model was developed by coupling the concentration–polarization (CP) model, which uses
one-dimensional flow assumption, with the irreversible thermodynamic Spiegler–Kedem model.
A nonlinear parameter estimation technique was used to determine the model parameters Lp (hydraulic
permeability constant), σ (reflection coefficient), and Bs (solute transport coefficient). Experimental
data were obtained from the treatment of a pre-treated glucose solution using a laboratory-scale
RO system, and studies on the validation of the model using experimental results are presented.
The calculated results are consistent with the experimental data. The proposed model describes the
RO membrane concentration process and deduces the expression of k (mass transfer coefficient in the
CP layer). The verification shows that the expression of k well-describes the reverse osmosis mass
transfer of a glucose solution.

Keywords: reverse osmosis; mass transfer process; concentration polarization; mathematical model

1. Introduction
In the last few decades, in China, with the development of industry and the increase in pollution,
water shortage has been getting worse. Wastewater recycling can effectively alleviate this situation
and contribute to the sustainability of development. Considering the relatively efficient and simple
process, membrane technology is a promising method for water treatment [1–4]. Additionally, it is
reported that reverse osmosis (RO) technology is a relatively appropriate technique for manipulating
sugar concentration [5]. In recent years, RO has gained widespread interest and has become a widely
adopted modern tool for industrial applications and research laboratories. This has resulted in a great
demand for RO membranes, making it necessary to reduce the cost of the RO system. According to the
literature reported [6], the development of a proper mathematical model that adequately describes the
performance of the RO process is crucial for the optimum design of an RO membrane which can make
the RO system more efficient, thereby reducing the overall cost.
A number of mechanistic and mathematical models have been proposed to describe the
mass transfer and hydrodynamic permeability in RO. In the RO separation processes, the mass
transfer is mainly governed by the inside-membrane mass transfer and the outside-membrane
mass transfer. The inside-membrane mass transfer is explained by the solution–diffusion model,
proposed by Lonsdale et al. [7,8] and the preferential sorption–capillary flow model, proposed by
Sourirajan et al. [8,9]. The main subject of investigation in the outside-membrane mass transfer is
the mass transfer process near the membrane, where concentration–polarization (CP) significantly
affects the mass transfer process. Hence, the study of outside-membrane mass transfer focuses on the
modeling of CP; it is necessary to model the CP phenomenon to predict the RO separation process.

Processes 2019, 7, 271; doi:10.3390/pr7050271 www.mdpi.com/journal/processes

Processes 2019, 7, 271 2 of 12

Various attempts were made to understand and describe the CP phenomenon by a theoretical
analysis, and the main models were based on film theory, such as the Kimura–Sourirajan model [10–14]
and the Spiegler–Kedem model [15–19], which are both related to the mass transfer coefficient. In other
literature, a model based on the solution–diffusion mass transport theory, CP, and pressure-dependent
dynamic membrane resistance was developed to predict the performance of RO for mixed salt and dye
solutions [20,21]. Pastagia et al. [22] developed an unsteady-state mass transfer model for a binary
solute system composed of reactive black dye and a reactive red dye. Moreover, Jamal [23] developed
a model without CP for the prediction of feed concentration, permeate concentration, rejection, and
flux as a function of the operating time. Overall, these models proved to be useful and applicable in
the RO separation process.
The objective of this work was to research the mass transfer during the RO concentration process.
To fulfill this requirement, a mathematical model based on glucose–water separation through an RO
membrane was developed. Considering the actual structure of the RO membrane, the irreversible
thermodynamic Spiegler–Kedem model is appropriate for describing the inside-membrane mass
transfer, while a CP model based on the one-dimensional flow assumption was selected to explain
the outside-membrane mass transfer. The mathematical model proposed in this study couples the
CP model with the irreversible thermodynamic Spiegler–Kedem model. Experimental data (glucose
concentration experiment) were obtained from the treatment of a pretreated glucose solution using a
laboratory-scale RO system, and studies on the validation of the model using experimental results are
presented The calculated results are consistent with the experimental data. Also, the expression of k
is deduced.

2. Theory
In the RO separation process, the mass transfer occurs mainly in the feed boundary layer (CP
layer) and inside the membrane. The general process of mass transfer is shown schematically in
Figure 1. In order to build a mass transfer model, a one-dimensional flow is assumed to be valid for
the transport of solvent and solute through the membrane. In addition, the CP layer is assumed to be
fully developed. On this basis, the mass transfer equation is expressed as

dc
Jw · Cp = Jw · C − D (1)
dx
where Jw is the solvent (pure water) flux, Cp is the permeate solution concentration, C is the solute
concentration in the CP layer, and D is the diffusion coefficient. The Equation (1) is transformed into
the following equation
dc
D = Jw · C − Jw · Cp (2)
dx
According to Figure 1, x ranges from 0 to δcp , and C ranges from Cf to Cδ,1 . Integrating Equation
(2) gives the following equation
Cδ,1 − Cp Jw ·δcp Jw
= e( D ) = e( k ) (3)
C f − Cp
where Cδ,1 is the solute concentration at the membrane surface (feed side, as shown in Figure 1), Cf is
the feed solution concentration, δcp is the CP layer thickness, and k is the mass transfer coefficient in
the CP layer and is defined as
D
k= (4)
δcp
According to Figure 1, the solute flux Js is expressed by the following equation
 
Js = Bs (Cδ,1 − Cδ,2 ) = Bs Cδ,1 − Cp = Jw · Cp (5)

where Bs is the solute transport coefficient.

where Jw is the solvent (pure water) flux, Cp is the permeate solution concentration, C is the solute
concentration in the CP layer, and D is the diffusion coefficient. The Equation (1) is transformed into
the following equation

dc
Processes 2019, 7, 271 D = Jw ⋅C − Jw ⋅Cp (2)12
3 of
dx

Figure 1. Schematic diagram of mass transfer in the reverse osmosis (RO) membrane separation process.
CP: concentration–polarization.

For the reverse osmosis process, the total flux of the solvent and solute (Jv ), which represents the
volumetric flux on the permeate side of the RO membrane and can reflect the concentration capacity of
the RO membrane, is expressed by the following equation

Jv = Jw + Js ≈ Jw (6)

According to the principle of mass conservation, we get the following equations

Q f C f = Qb Cb + Qp Cp (7)

Q f = Qb + Qp (8)

where Qf , Qb, , and Qp are the feed solution flow, retentate flow, and permeate flow, respectively.
Cb is the retentate solution concentration, and the relationship between Qp and Jw is shown in the
following equation
Qp
Jv = (9)
S
where S is the effective RO membrane area.
The recovery (y), which represents the water production capacity, is defined as the fraction of
the feed flow which passes through the membrane. The higher the recovery, the stronger the water
production capacity of the RO system, and the stronger the concentration capacity of the RO system.
The value of y is calculated by the following equation

Qp
y= × 100% (10)
Qf

To describe the mass transfer process of the RO membrane, we examined the parameters
of the glucose solution and the microstructure of the RO membrane (PA2-4040, HYDECANME,
HYDRANAUTICS, Oceanside, CA, USA). The effective thickness (calculated by deducting the thickness
of the support layer) of the RO membrane was 0.2 µm. In contrast, the diameter of the glucose molecule
(<1 × 10−3 µm) and the diameter of the water molecule (≈4 × 10−4 µm) are much smaller than the
RO membrane thickness, and the membrane pore of the RO membrane was generally less than 1 nm
(roughly the same as the diameter of the glucose molecule). Therefore, we believe that the solution
interacted with the RO membrane during the membrane separation of the glucose solution, and the
solution-diffusion model was discarded. In this work, the irreversible thermodynamic Spiegler–Kedem
Processes 2019, 7, 271 4 of 12

model [24–27] was assumed to be appropriate for explaining the separation performance of solute
through the membrane. Therefore, Jv is expressed by the following equation

Jv = Lp (∆p − σ∆π) (11)

where Lp is the hydraulic permeability constant, 4p is the transmembrane pressure, 4π is the difference
in the osmotic pressure across the membrane, and σ is the reflection coefficient. The reflection coefficient
represents the solute separation capability of a membrane, where σ = 0 means no separation, and σ = 1
means complete separation (100% separation). For permeable membranes, 0 < σ < 1. 4π is calculated
using the following equation
 
∆π = RT (Cδ,1 − Cδ,2 ) = RT Cδ,1 − Cp (12)

where R is the gas law constant, T is the temperature, and Cδ,2 is the solute concentration at the
membrane surface (permeate side, as shown in Figure 1).
The observed membrane rejection fraction Ro is given by

C f − Cp
Ro = (13)
Cf

and the real rejection fraction Rr is given by

Cδ,1 − Cδ,2 Cδ,1 − Cp

Rr = = (14)
Cδ,1 Cδ,1

According to the Spiegler–Kedem equation, the real rejection fraction Rr is expressed as

σ(1 − F)
Rr = (15)
1 − σF
F is defined as
F = exp[−Jv (1 − σ)/Bs ] (16)

where Bs is the solute transport coefficient.

Rewriting Equation (15), we get

Bs σ(1 − Rr )
Jv = ln (17)
1−σ σ − Rr

Substituting Equations (13) and (14) in Equation (3), we get

1 − Ro Jv 1 − Rr
   
ln = + ln (18)
Ro k Rr

Rewriting Equation (18), we get

Rr (1 − Ro )
k = Jv /ln (19)
Ro (1 − Rr )

Substituting Equations (3) and (12) in Equation (11), we get

   Jv 
Jv = Lp (∆p − σ∆π) = Lp ∆p − σRT C f − Cp e( k ) (20)
   
 Jv 

J v = Lp ( Δp − σΔπ ) = Lp  Δp − σ RT ( C f − C p ) e k   (20)
 
Processes 2019, 7, 271 5 of 12
Substituting Equation (19) in Equation (20), we get

 R (1 − Ro ) 
J v = Lp  Δp − σ RT ( C f − C p ) r
Substituting Equation (19) in Equation (20), we get
 (21)
"   RrR (1 − R#r )
(1 − Ro ) 
o
Jv = Lp ∆p − σRT C f − Cp (21)
Ro (1 − Rr )
3. Experimental
3. Experimental
3.1. Membrane and Module
3.1. Membrane and Module
A polyamide composite membrane (PA2-4040, HYDECANME, USA) was chosen as the reverse
A polyamide composite membrane (PA2-4040, HYDECANME, USA) was chosen as the reverse
osmosis experiment membrane material in the present work. The membrane characteristics are
osmosis experiment membrane material in the present work. The membrane characteristics are shown
shown in Table 1.
in Table 1.
Table
Table 1. Main
Main characteristics
characteristics of
of the
the used
used membrane.

Type
Type PA2-4040
PA2-4040
Membrane properties
Membrane properties Composition
Composition polyamide
polyamide
Permeable
Permeable capacity
capacity (average,
(average, m)3·d−1)
m3 ·d−1 7.2 7.2
Effective
Effective membrane
membrane 2
area area
(m ) (m2) 7.9 7.9
Recovery
Recoveryraterate
(single, %) %)
(single, 15 15
Usage conditions
Usage conditions Maximum
Maximum pressure
pressure (Mpa)(Mpa) 4.14 4.14
Temperature (◦ C)(°C)
Temperature 5–45 5–45
Maximum flowflow −1 )
(m3 ·h(m 3.6 3.6
Maximum 3·h−1)

The
The experimental
experimental device
device (Figure
(Figure 2) mainly
mainly consisted
consisted of
of aa cooling
cooling and
and heating
heating circulation
circulation tank
tank to
to
change
change the
the temperature
temperature of the
the model
model solution,
solution, aa pump
pump for
for feeding
feeding the
the glucose
glucose solution,
solution, aa valve
valve (check
(check
valve) to
to control
control the
the solution
solution flux,
flux, the
the reverse
reverse osmosis
osmosis membrane
membrane module
module (including
(including pressure
pressure gauges
gauges
and
and aa flow
flow meter), and a device for sampling the solution.

Figure
Figure 2.
2. Schematic
Schematic illustration
illustration of
of the
the reverse
reverse osmosis
osmosis system.
system.

3.2. Preparation
3.2. Preparation of
of Model
Model Solutions
Solutions and
and Concentration
Concentration Experiment
Experiment
In this
In this study,
study,glucose anhydrous
glucose anhydrous(AR) waswas
(AR) purchased from from
purchased the Kelong Chemical
the Kelong ReagentReagent
Chemical Factory,
Chengdu, China. China.
Model Model
solutions (24.8–166.8 −3
mol·mmol·m
) were prepared using ultrapure water, by
Factory, Chengdu, solutions (24.8–166.8 −3) were prepared using ultrapure water,
adding suitable amounts of the analytes (calculated).
by adding suitable amounts of the analytes (calculated).
The glucose model solutions were selected to carry out reverse osmosis concentration experiments
using the reverse osmosis system (Figure 2). Model solutions were transferred into the membrane
module by a feed pump. This was followed by the outflow of concentrate and permeate from the
reverse osmosis membrane module and the reflow into the circulation tank. During this procedure, the
concentrate was sampled by a sampling device and subsequently analyzed to evaluate the separation
 The glucose model solutions were selected to carry out reverse osmosis concentration
experiments using the reverse osmosis system (Figure 2). Model solutions were transferred into the
membrane module by a feed pump. This was followed by the outflow of concentrate and permeate
from the reverse osmosis membrane module and the reflow into the circulation tank. During this
Processes 2019, 7, 271 6 of 12
procedure, the concentrate was sampled by a sampling device and subsequently analyzed to evaluate
the separation effect of the reverse osmosis system. The concentrations of the glucose solutions
(amount
effect of theofreverse
sample for each
osmosis test:The
system. 20concentrations
μL) were quantified using
of the glucose a high-performance
solutions liquid
(amount of sample for
chromatography
each test: 20 µL) weresystem (HPLC, Agilent
quantified LC1200, Agilent, Santa
using a high-performance liquidClara, CA, USA) equipped
chromatography with a
system (HPLC,
differential
Agilent refraction
LC1200, detector
Agilent, (RID) and
Santa Clara, CA, an Aminex
USA) HPX-87H
equipped with acolumn (Bio-Rad
differential Co., detector
refraction Hercules,(RID)
CA,
USA). Experimental data were measured three times to obtain an average
and an Aminex HPX-87H column (Bio-Rad Co., Hercules, CA, USA). Experimental data were measuredvalue under each
experimental
three times to condition.
obtain an average value under each experimental condition.

4. Results and Discussion

4.1. Influence of Experimental Parameters of the Reverse Osmosis System

System
The pump
pump used
used was
was aa vertical
vertical multistage
multistage centrifugal
centrifugal pump
pump (CDLF1,
(CDLF1, 50HZ,
50HZ, CNP, Hangzhou,
CNP,Hangzhou,
China),
China), and in the experimental design, one of the valves was installed at the outlet membrane
and in the experimental design, one of the valves was installed at the outlet of the RO of the RO
(valve 3, Figure
membrane 2). The
(valve feed flow
3, Figure 2). and
Thetransmembrane
feed flow andpressure in the system
transmembrane were changed
pressure by adjusting
in the system were
the valveby
changed 3. adjusting
The relationship
the valvebetween
3. The the feed solution
relationship and the
between transmembrane
feed solutionpressure is shown in
and transmembrane
Figure
pressure3. is
According
shown intoFigure
Figure3.3, According
the feed flow rate decreased
to Figure when
3, the feed the rate
flow transmembrane pressure
decreased when the
increased, and the
transmembrane two parameters
pressure increased,wereand roughly
the two linear.
parameters were roughly linear.

250

200
Qf (m · s ，×10 )
-6

150
-1
3

100

50

0
300 350 400 450 500 550 600 650
3
△P (Pa，×10 )

Figure 3. Relation between the feed solution flow and the transmembrane pressure.
pressure.

Figure
Figure 4 reflects
reflects the effect
effect of
of transmembrane
transmembrane pressurepressure on on the RO process. According to Figure 4,
within the range of experimental parameters (310 kPa < △p <617
within the range of experimental parameters (310 kPa < 4p < 617kPa;
kPa; 3636 mol·m−3<<CCf f<< 140
mol·m −3
140 mol·m
−3
mol·m−3 ),
when
when otherother experimental
experimental conditions
conditions are constant, JJvv,, y,
are constant, y, RRoo,, and Cb increase
increase with
with an increase in
transmembrane
transmembrane pressure. and △p
pressure. Jvv and 4p are approximately
approximately linear,
linear, which
which is is consistent
consistent with the irreversible
thermodynamic
thermodynamic Spiegler–Kedem
Spiegler–Kedemmodel modelselected
selectedininthis
thispaper
paper(Equation
(Equation(11)).(11)).As
Asthe △p
the4p increased,
increased, y
increased
y increased slowly,
slowly,followed
followedbybyananincrease
increaseiningrowth
growthrate,
rate,and
andthe theCCb bmaintained
maintained approximately
approximately the the
same trend. According to Equation (10), the recovery formula is a ratio. Combined
same trend. According to Equation (10), the recovery formula is a ratio. Combined with Figure 3 and with Figure 3 and Jv
trend
Jv trendover 4p.△p.
over In the formula
In the formula of y,ofthey, numerator
the numeratorincreases linearly
increases with with
linearly △p (consistent
4p (consistent with thewithtrend
the
of Jv , Equation
trend (9)), and
of Jv, Equation (9)),the
anddenominator
the denominator decreases approximately
decreases approximately linearly 4p. △p.
withwith
linearly Therefore, the
Therefore,
overall
the overallratio ratio
should increase,
should and the
increase, ratethe
and of rate
the value
of theincrease should beshould
value increase gradual,bewhich is consistent
gradual, which is
with the trend
consistent y in
withofthe Figure
trend of y4.inAccording to Figure 4b,
Figure 4. According to with
Figure the4b,increase
with the 4p, Ro isof
of increase △p, Roimproved,
slightly is slightly
although
improved,the trend isthe
although nottrend
obvious.
is notAccording to Figure 3,
obvious. According toQ f decreased
Figure with increases
3, Qf decreased of 4p. This
with increases of
△p. This means that a unit volume of feed solution can acquire more pressure to facilitate the RO
means that a unit volume of feed solution can acquire more pressure to facilitate the RO process. The
macroscopic
process. The performance of this processofwas
macroscopic performance thisan increase
process was inanRoincrease
and Jv . in Ro and Jv.
 Processes
Processes 2019,
Processes 2019,
7, 7,
2019, 271x7,FOR
x FOR PEER
PEER REVIEW
REVIEW 77of
7 of of1212
12

1.00 82
1.00 82
5.5
5.5
15 80
15 80
5.0 0.98
5.0 0.98 R
Jv Ro o
Jv 78
4.5 y C 78
Cb b
Jv (m · m · s ，×10 )

4.5
-7

y
Jv (m · m · s ，×10 )
-7

0.96
10 0.96 76

Cb (mol· m )
-3
4.0 10 76

Cb (mol· m )
y (%)

-3
4.0

y (%)
-1

Ro
-1

74

Ro
-2

3.5 0.94 74
-2

3.5 0.94
3

5 72
3

3.0 5 72
3.0
0.92
0.92
2.5 (b) 70
2.5 (a) (b) 70
(a)
2.0 0 0.90 68
2.0 0 0.90 68
300 350 400 450 500 550 600 650 300 350 400 450 500 550 600 650
300 350 400 450 500 3 550 600 650 300 350 400 450 500 3 550 600 650
△P (Pa，×10
3 ) △P (Pa，×10
3 )
△P (Pa，×10 ) △P (Pa，×10 )

Figure
Figure 4. Influence of of transmembrane pressure onon
RORO process.(a)(a) Solvent flux and recovery vary with
Figure 4.4.Influence
Influence transmembrane
transmembranepressure pressure on process.
RO process. Solvent flux
(a) Solvent and recovery
flux vary
and recovery with
vary
transmembrane pressure, feed solution concentrate (C f) = 70.17 mol·m −3, pH = 6, T = 306 K; (b) Solute
transmembrane
with transmembrane pressure, feed solution
pressure, concentrate
feed solution (Cf) = 70.17
concentrate = 70.17
(Cf )mol·m −3, pH = 6,−3
mol·m T ,=pH 306=K;6,(b) = 306 K;
T Solute
rejectionandandretentate
retentatesolution
solutionconcentration
concentrationvary
varywith
withtransmembrane
transmembranepressure,
pressure,feed
feedsolution
solution
(b)rejection
Solute rejection and retentate solution concentration vary with transmembrane pressure, feed
concentrate
concentrate (C(C f) = 70.17 mol·m , pH = 6, T = 306 K.
−3
f) = 70.17 mol·m−3, pH = 6,−3
solution concentrate (Cf ) = 70.17 mol·m T, =pH 306=K.6, T = 306 K.
Figure
Figure 5 shows
5 shows thethe effect
effect of Conf on the RO process. It can be seen in Figure 5 that an increase in Cf
Figure 5 shows the effect ofofCCf fon the RO
the RO process.
process. ItItcan
canbebeseen
seenininFigure
Figure5 5that anan
that increase in in
increase Cf Cf
led to an increase in the C b and decreases in Jv and y. The trend of Ro was not significant (slightly lower
ledledto to
anan increase
increase ininthetheCCband
b
anddecreases
decreases in in JJvv and
and y.y.The
Thetrend
trendofofRRo was not significant (slightly lower
o was not significant (slightly lower
trend).
trend). In In
thethe RO RO process,
process, the the
CP CP phenomenon
phenomenon occurs
occurs near
near thethe boundary
boundary layer
layer ofof the
the membrane.
membrane.
trend). In the RO process, the CP phenomenon occurs near the boundary layer of the membrane. When
When Q f is constant, the increase in Cf leads to an enhanced CP phenomenon, which can result in a
Qf When Qf is constant,
isdecrease
constant, the increasethe inincrease
C f leadsin to
Cf an
leads to an enhanced
enhanced CP CP phenomenon,
phenomenon, which which
can can
result in result
a in a in
decrease
decrease inin Jv, Jfollowed
v, followed by a decrease in y. On the other hand, the enhancement of the CP phenomenon
by a decrease in y. On the other hand, the enhancement of the CP phenomenon
Jv , followed
also leads by a decrease in y. Ontrend
the otherRhand, the enhancement of the CP phenomenon also leads
also leads toto a slightly
a slightly lower
lower trend forfor Ro. oThe. The trend
trend ofofCbCincreasing
b increasing with Cf was predictable and
with Cf was predictable and
slightly lower trend for Ro . The trend of Cb increasing with Cf was predictable and imaginable.
to a imaginable.
imaginable.
1.00 150
1.00 150
5.0
5.0
2.5
2.5 125
0.98
4.5 0.98 125
4.5
Jv (m · m · s ，×10 )

4.0
-7

0.96 100
Jv (m · m · s ，×10 )

4.0
-7

2.0 100

Cb (mol· m )
0.96

-3
2.0

Cb (mol· m )
-3
y (%)

3.5
-1

(a) (b)
y (%)

Ro

3.5
-1

(a) (b) 75
Ro

0.94
-2

0.94 75
-2

3.0 Jv
3

3.0 Jv 1.5
3

1.5
y 50
y 0.92 R 50
2.5 0.92 Ro o
2.5 C
Cb b
25
2.0 1.0 0.90 25
2.0 20 40 60 80 100 120 140 1.0 0.90 20 40 60 80 100 120 140
20 40 60 80 100 -3 120 140 20 40 60 80 100 -3 120 140
Cf (mol·-3 m ) Cf (mol·-3 m )
Cf (mol· m ) Cf (mol· m )

Figure
Figure
Figure 5. Influence
5. 5.Influence
Influence of
thethe
ofofthe feed feed
feed solution
solution
solution concentration
concentration
concentration onon
on RO
RORO process.
process.
process. (a)
(a)(a) Solvent
Solvent
Solvent fluxflux
andand
and recovery
recovery
recovery
vary with feed solution concentration, feed solution flow (Q f=
) = 1.56 × 10 −4mm 3 −1
3·s ,−1
pH = =
6, T = = K;
306 (b)(b)
−4
vary with feed solution concentration, feed solution flow (Q ) 1.56 ×
vary with feed solution concentration, feed solution flow (Qf)f = 1.56 × 10 m ·s , pH = 6, T = 306 K;
−4 3 −1 ·s , pH 6, T 306 K;
Solute
(b)Solute
Solute rejection
rejection and
and retentate
retentate solution
solution concentration
concentration vary
vary with
with feed
rejection and retentate solution concentration vary with feed solution concentration, feed solution
solution concentration,
concentration, feed
feed
f ) f)==1.56 = 6,=T6,= T = K.
solution
solution flow flow
(Q(Q (Qf)1.56
= 1.56
× 10 −4−4m
× 10 3m ·s,−1
33·s −1, pH = 6, T = 306 K.
, pH 306 K.
solution flow × 10 −4 m ·s−1 pH 306

4.2. Model
4.2.
4.2.
Validation
Model
Model Validation
Validation
Some Some
Some mathematical
mathematical
mathematical formulas
formulas
formulas for the
forfor
the mass
the mass
mass transfer
transfer
transfer inin
in the
thetheRO
RORO process
process
process were
were
were given
given in
inin
the the
the previous
previous
previous
section
section (Section
(Section 2). 2).
In In
thisthis section,
section, model
model validation
validation was
was carried
carried out
out to
to verify
verify that
section (Section 2). In this section, model validation was carried out to verify that the calculated values the calculated
calculated values
values
of ofof
thesethese mathematical
mathematical formulas
formulas werewere consistent
consistent with
with the
the experimental
experimental
these mathematical formulas were consistent with the experimental data. In order to verify the data.
data. In
In order
order to
to verify
verify the
the
models,
models,
models, wewe we first
firstfirst determined
determined
determined somesome some
parameter parameter values
valuesvalues
parameter in the in in
model. the model. The
The non-linear
the model. non-linear
parameter
The non-linear parameter
estimation
parameter
estimation
technique
estimation techniqueofofLevenberg–Marquardt,
of technique
Levenberg–Marquardt, Levenberg–Marquardt,
combined with combined withthe
the Gauss–Newton
combined with theGauss–Newton
Gauss–Newton algorithm
algorithm [17,20,28–31],
algorithm
was [17,20,28–31],
used in this
[17,20,28–31], waswas
study used
used inin
to this
estimate
this study
study toto
the estimate
parameters,
estimate thethe parameters, and
and the calculated
parameters, and thethe calculated ofvalues
values values
calculated Bs andofofBsBs and
σand
were
0.85 σ were
σ were −8
× 10 0.850.85 −1× 10
m·s× 10and−8
−8 m·s−1 and 0.9981, respectively. We also calculated the Lp and Cδ,1; the results are
m·s0.9981,
−1 and respectively. We alsoWe
0.9981, respectively. calculated the Lp and
also calculated the C ; theCresults
Lpδ,1and are shown
δ,1; the results are in
Tableshown
shown inin
2. From Table
Table
Table 2. 2.
2 From
From
andTable Table
Equation 2 and
2 and Equation
Equation
(14), (14),
(14),
we further wewe further
further
calculated calculated
calculated
the the
the real
real rejection real rejection
rejection
fraction R fraction
fraction
. As RrR. r.
shown
r
in AsAs shown
shown
Table in Table
2, Lpinremained
Table 2, 2,
Lp Lp remained
remained
basically basically
thebasically
same. the
the same.
same.
Processes 2019, 7, x FOR PEER REVIEW 8 of 12

Processes 2019, 7, 271 8 of 12

Table 2. Solute concentration at the membrane surface (Cf = 70.17 mol·m−3, pH = 6, T = 306 K, R = 8.314
J·mol−1·K−1).
Table 2. Solute concentration at the membrane surface (Cf = 70.17 mol·m−3 , pH = 6, T = 306 K,
Qf
R = 8.314 J·mol−1 ·K−1 ).
△p Cp Cδ,1 Lp
(m3·s−1, ×10−6) (Pa, ×103) (mol·m−3) (mol·m−3) (m·s−1·Pa−1, ×10−12)
Qf
248.8889 4p
311 Cp
3.9142 Cδ,1
70.57104 Lp
1.3923
(m 3 ·s−1 , ×10−6 )
218.6111 (Pa, ×103 )
346.5 (mol·m −3
3.8976 )
−3
70.72304 )
(mol·m (m·s−1 ·Pa −1 , ×10−12 )
1.3611
175.5556
248.8889 421.5
311 3.3972
3.9142 71.06723
70.57104 1.3423
1.3923
218.6111
152.2222 346.5
488.5 3.8976
3.2748 70.72304
71.49108 1.3611
1.3238
175.5556 421.5 3.3972 71.06723 1.3423
123.3333 505.5 3.2471 71.97543 1.3048
152.2222 488.5 3.2748 71.49108 1.3238
75.5556
123.3333 530
505.5 3.0858
3.2471 73.44194
71.97543 1.3889
1.3048
59.7222
75.5556 553
530 2.9635
3.0858 74.33926
73.44194 1.3709
1.3889
48.3333
59.7222 578
553 2.9635
2.9635 75.48901
74.33926 1.3538
1.3709
48.3333
36.1111 578
598.5 2.9635
2.8078 75.48901
77.80078 1.3538
1.353
36.1111 598.5 2.8078 77.80078 1.353
29.7222 616.5 2.7411 79.9817 1.3529
29.7222 616.5 2.7411 79.9817 1.3529

In the RO experiment, the feed solution flow Qf and the permeate flow Qp were obtained
In the to
according ROtheexperiment,
flow meter.theThen,
feed solution flow
the solvent Qf and
(pure the permeate
water) flux Jv wasflow Qp were
obtained obtained according
(experimental value)
to the flow
based meter. Then,
on Equation the solvent
(9). Combined (pure
with water) (21)
Equation Jv was
flux and obtained
Table (experimental
2, we calculated the Jvalue) based
v (theoretical
on Equation (9). Combined with Equation (21) and Table 2, we calculated the J
value). The comparison of the two Jv values is shown in Figure 6. Meanwhile, we calculated the v (theoretical value).
The comparison
relative error between eachJvgroup
of the two valuesofistwo
shown in Figure
Jv values, and6.theMeanwhile,
maximumwe calculated
value the relative
of the relative errorerror
was
12.81%. each group of two Jv values, and the maximum value of the relative error was 12.81%.
between

6
experimental value
theoretical value
5
Jv (m · m · s ，×10 )
-7

4
-1
-2

3
3

0 50 100 150 200 250

3 -1 -6
Qf (m · s ，×10 )

Figure 6. Comparison of the experimental and theoretical solvent flux values.

values.

4.3. Estimation of
4.3. Estimation of the
the Mass
Mass Transfer
Transfer Coefficient,
Coefficient, kk
The
The mass
mass transfer
transfer coefficient
coefficient (k)
(k) represents
represents the
the transfer
transfer rate
rate of
of the
the solute
solute in
in the
the CP
CP layer. The kk
layer. The
value reflect the strength of the concentration polarization, and with a known k
value can reflect the strength of the concentration polarization, and with a known k value, the salt
can value, the salt
diffusion coefficient and boundary layer thickness can be determined using a dimensionless
diffusion coefficient and boundary layer thickness can be determined using a dimensionless analysis. analysis.
The
The RO
ROmembrane
membrane (PA2-4040, HYDECANME,
(PA2-4040, HYDECANME, USA)USA)
used in this in
used workthiswas a spiral
work waswound membrane.
a spiral wound
The mass transfer
membrane. coefficient
The mass (k) coefficient
transfer was calculated according
(k) was to theaccording
calculated followingto empirical formula
the following [12]
empirical
formula [12] D
k = αReβ Scγ (22)
γ hD
β
d
k = α Re Sc (22)
where α, β, and γ are constants that vary for different dphysical
h situations, which are determined
experimentally, and Re is the Reynolds number. Experiments indicate that the flow rate of the fluid,
Where α, β, and γ are constants that vary for different physical situations, which are determined
the fluid channel size, and the physical properties of the fluid (density and viscosity) affect the type of
experimentally, and Re is the Reynolds number. Experiments indicate that the flow rate of the fluid,
flow. These main influencing factors can be combined into a dimensionless parameter (Re), the value
Processes 2019, 7, 271 9 of 12

of which can reflect the type of flow. Sc is the Schmidt number, and dh is the hydraulic diameter of the
channel. Re and Sc are defined as
ρudh µ
Re = , Sc = (23)
µ ρD
where u, ρ, and µ are the velocity, the density, and the viscosity of the fluid, respectively.
For a spiral wound membrane, dh is approximated by the following formula

dh ≈ 2h (24)

where h is the rectangular feed channel thickness (h = 0.7112 mm, HYDECANME).

According to Equation (22) and the experimental work of this article, When Qf changes during
the RO process (since the Cf does not change, Sc and D/dh are constant) only Re changes. Therefore,
Equation (22) is rewritten by the following equation

k = α1 Reβ (25)

where α1 represents a combination constant.

Combining Table 2 with Equation (19), we can calculate the value of k and Re under different feed
solution flow conditions (as shown in Table 3), then we performed a power function fitting according
to Table 3 and Equation (25) and β = 0.9243. Equation (22) is rewritten by the following equation

D
k = αRe0.9243 Scγ (26)
dh

Table 3. Value of the mass transfer coefficient (k) and Reynolds number (Re) under different Qf
(Cf = 70.17 mol·m−3 , pH = 6, T = 306 K, R = 8.314 J·mol−1 ·K−1 ).

Qf (m3 ·s−1 , ×10−6 ) k (m·s−1 , ×10−6 ) dh (m, ×10−4 ) Re Sc

248.8889 37.6 1.4224 151.2 482
218.6111 33.61 1.4224 132.8 482
175.5556 27.15 1.4224 106.6 482
152.2222 21.33 1.4224 92.5 482
123.3333 16.29 1.4224 74.9 482
75.5556 9.49 1.4224 45.9 482
59.7222 7.86 1.4224 36.3 482
48.3333 6.49 1.4224 29.4 482
36.1111 4.77 1.4224 21.9 482
29.7222 3.88 1.4224 18.1 482

Also, when Cf changed during the RO process, ρ, µ, and D of the solution were slightly changed
as a result of the change in the concentration of the solution, leading to a slight change in Re. Since
the change in Re was slight, we ignored it (Re ≈ 95) in this paper. Equation (22) is rewritten by the
following equation
k = α2 Scγ (27)

where α2 represents a variable value.

We calculated the value of k and Sc under different feed solution concentration conditions, as
shown in Table 4, and the value of D was estimated from the literature [30]). According to Table 4,
using the same method as above, the value of γ was obtained (γ = 0.3495). Equation (22) was rewritten
by the following equation
D
k = αRe0.9243 Sc0.3495 (28)
dh
Processes 2019, 7, x FOR PEER REVIEW 10 of 12

D
k = α Re0.9243 Sc 0.3495 (28)
Processes 2019, 7, 271
dh 10 of 12

Table 4. Value of k and Schmidt number (Sc) under different Cf (Qf = 155.56×10−6 m3·s−1, pH = 6, T = 306
Table 4. Value of k and Schmidt number (Sc) under different Cf (Qf = 155.56 × 10−6 m3 ·s−1 , pH = 6,
K, R = 8.314 J·mol−1·K−1).
T = 306 K, R = 8.314 J·mol−1 ·K−1 ).
Cf (mol·m−3) −3 k (m·s−1, ×10−6)−1 Sc D/ dh (m·s−1−1
, ×10−6)
Cf (mol·m ) k (m·s , ×10−6 ) Sc D/dh (m·s , ×10−6 )
36.1567 22.2 484 1.1710
36.1567 22.2 484 1.1710
62.4444 21.95 482 1.1703
62.4444 21.95 482 1.1703
87.5533
87.5533 21.77 21.77 480 480 1.1696
1.1696
111.506
111.506 21.6 21.6 479 479 1.1689
1.1689
137.149
137.149 21.4 21.4 478 478 1.1682
1.1682

Finally, we calculated
Finally, we calculatedthe
thevalue
valueofofααaccording
accordingtotoTable
Table4 4and
andmodified
modified the
the value
value of of α according
α according to
to Table
Table 3, α3,=α0.031,
= 0.031,
andand obtained
obtained thethe final
final expression
expression of the
of the massmass transfer
transfer coefficient
coefficient
0.9243 0.3495 DD
kk ==0.031 Re0.9243
0.031Re Sc0.3495
Sc (29)
(29)
ddhh

In this
this paper,
paper,according
accordingtotoEquation
Equation(29),
(29),the Reynolds
the Reynoldsnumber
number was thethe
was main factor
main affecting
factor the
affecting
mass transfer
the mass coefficient.
transfer Combining
coefficient. Table
Combining 3 and
Table 3 andEquation
Equation(29),
(29),we
weplotted
plottedthe
the experimental
experimental and
theoretical values of k, as shown in Figure 7. We
We also
also calculated
calculated the
the relative
relative error between each group
of the two k values; the maximum value of the relative error error was
was 12.2%.
12.2%.

30

25
experimental value
theoretical value
20
k (m· s ，×10 )
-6

15
-1

10

0
0 20 40 60 80 100
Re
Figure 7. Comparison of the k values.

5. Conclusions
5. Conclusions
A mathematical
A mathematicalmodeling
modeling of glucose–water separation
of glucose–water through
separation an RO membrane
through was developed
an RO membrane was
in this article. The model coupled the CP model, which used a one-dimensional
developed in this article. The model coupled the CP model, which used a one-dimensional flow assumption,
flow
with the irreversible
assumption, with thethermodynamic Spiegler–Kedem
irreversible thermodynamic model. The model.
Spiegler–Kedem non-linear
The parameter
non-linear estimation
parameter
technique was used to determine the model parameters, such
estimation technique was used to determine the model parameters, as Lp , σ, and B
such sas Lp, σ, andon
. Research Bsthe validation
. Research on
of the model with experimental results was presented, and experimental data were
the validation of the model with experimental results was presented, and experimental data were obtained from the
treatmentfrom
obtained of a pre-treated
the treatmentglucose
of a solution using
pre-treated a laboratory-scale
glucose solution using RO asystem. The comparison
laboratory-scale of the
RO system.
The comparison of the results showed that the calculated values were consistent withto the
results showed that the calculated values were consistent with the experimental data. According the
calculated results,
experimental data.the maximum
According torelative error between
the calculated results,the
thetwo values was
maximum 12.81%.
relative errorThe expression
between of k
the two
was also deduced. The model well-describes the membrane performance during
values was 12.81%. The expression of k was also deduced. The model well-describes the membrane the mass transfer
process, and the verification showed that the expression of k well-describes the reverse osmosis mass
transfer of glucose solution.

Author Contributions: The concentrations of the glucose solutions were quantified by co-author H.Q. using
HPLC, and the rest was completed independently by C.C.
Processes 2019, 7, 271 11 of 12

Funding: This research was funded by Central Public-interest Scientific Institution Basal Research Fund [No.
12018206030202210_04]. And the APC was funded by Central Public-interest Scientific Institution Basal Research
Fund [12016206030201701].
Acknowledgments: The authors are grateful to the Central Public-interest Scientific Institution Basal Research
Fund [No. 12018206030202210_04] and Central Public-interest Scientific Institution Basal Research Fund
[12016206030201701].
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Atab, M.S.; Smallbone, A.J.; Roskilly, A.P. A hybrid reverse osmosis/adsorption desalination plant for
irrigation and drinking water. Desalination 2018, 444, 44–52. [CrossRef]
2. Chung, T.S.; Zhang, S.; Wang, K.Y.; Su, J.; Ling, M.M. Forward osmosis processes: Yesterday, today and
tomorrow. Desalination 2012, 287, 78–81. [CrossRef]
3. Ray, S.S.; Chen, S.S.; Sangeetha, D.; Chang, H.M.; Thanh, C.N.D.; Le, Q.H.; Ku, H.M. Developments in
forward osmosis and membrane distillation for desalination of waters. Environ. Chem. Lett. 2018, 16,
1247–1265. [CrossRef]
4. Ray, S.S.; Chen, S.S.; Nguyen, N.C.; Nguyen, H.T.; Dan, N.P.; Thanh, B.X. Exploration of polyelectrolyte
incorporated with Triton-X 114 surfactant based osmotic agent for forward osmosis desalination. J. Environ.
Manag. 2018, 209, 346–353. [CrossRef] [PubMed]
5. Zhou, F.; Wang, C.; Wei, J. Separation of acetic acid from monosaccharides by NF and RO membranes:
Performance comparison. J. Membr. Sci. 2013, 429, 243–251. [CrossRef]
6. Sundaramoorthy, S.; Srinivasan, G.; Murthy, D.V.R. An analytical model for spiral wound reverse osmosis
membrane modules: Part I—Model development and parameter estimation. Desalination 2011, 280, 403–411.
[CrossRef]
7. Lonsdale, H.K.; Merten, U.; Riley, R.L. Transport properties of cellulose acetate osmotic membranes. J. Appl.
Polym. Sci. 1965, 9, 1341–1362. [CrossRef]
8. Merdaw, A.A.; Sharif, A.O.; Derwish, G.A.W. Water permeability in polymeric membranes, Part II. Desalination
2010, 257, 184–194. [CrossRef]
9. Soltanieh, M.; GILL’, W.N. Review of reverse osmosis membranes and transport models. Chem. Eng. Commun.
1981, 12, 279–363. [CrossRef]
10. Goosen, M.F.A.; Sablani, S.; Cin, M.D.; Wilf, M. Effect of cyclic changes in temperature and pressure on
permeation properties of composite polyamide seawater reverse osmosis membranes. Sep. Sci. Technol. 2010,
46, 14–26. [CrossRef]
11. Ghernaout, D. Reverse osmosis process membranes modeling—A historical overview. J. Civ. Constr. Environ.
Eng. Civ. 2017, 2, 112–122.
12. Qiu, T.Y.; Davies, P.A. Concentration polarization model of spiral-wound membrane modules with application
to batch-mode RO desalination of brackish water. Desalination 2015, 368, 36–47. [CrossRef]
13. Kim, D.Y.; Lee, M.H.; Gu, B.; Kim, J.H.; Lee, S.; Yang, D.R. Modeling of solute transport in multi-component
solution for reverse osmosis membranes. Desalin. Water Treat. 2010, 15, 20–28. [CrossRef]
14. Khanarmuei, M.; Ahmadisedigh, H.; Ebrahimi, I.; Gosselin, L.; Mokhtari, H. Comparative design of plug
and recirculation RO systems; thermoeconomic: Case study. Energy 2017, 121, 205–219. [CrossRef]
15. Spiegler, K.S.; Kedem, O. Thermodynamics of hyperfiltration (reverse osmosis): Criteria for efficient
membranes. Desalination 1966, 1, 311–326. [CrossRef]
16. Jain, S.; Gupta, S.K. Analysis of modified surface force pore flow model with concentration polarization
and comparison with Spiegler–Kedem model in reverse osmosis systems. J. Membr. Sci. 2004, 232, 45–62.
[CrossRef]
17. Murthy, Z.V.P.; Chaudhari, L.B. Separation of binary heavy metals from aqueous solutions by nanofiltration
and characterization of the membrane using Spiegler–Kedem model. Chem. Eng. J. 2009, 150, 181–187.
[CrossRef]
18. Attarde, D.; Jain, M.; Gupta, S.K. Modeling of a forward osmosis and a pressure-retarded osmosis spiral
wound module using the Spiegler-Kedem model and experimental validation. Sep. Purif. Technol. 2016, 164,
182–197. [CrossRef]
Processes 2019, 7, 271 12 of 12

19. Hidalgo, A.M.; León, G.; Gómez, M.; Murcia, M.D.; Gómez, E.; Gómez, J.L. Application of the
Spiegler–Kedem–Kachalsky model to the removal of 4-chlorophenol by different nanofiltration membranes.
Desalination 2013, 315, 70–75. [CrossRef]
20. Ahmad, A.L.; Chong, M.F.; Bhatia, S. Mathematical modeling of multiple solutes system for reverse osmosis
process in palm oil mill effluent (POME) treatment. Chem. Eng. J. 2007, 132, 183–193. [CrossRef]
21. Al-Bastaki, N. Removal of methyl orange dye and Na2 SO4 salt from synthetic waste water using reverse
osmosis. Chem. Eng. Process. Process Intensif. 2004, 43, 1561–1567. [CrossRef]
22. Pastagia, K.M.; Chakraborty, S.; DasGupta, S.; Basu, J.K.; De, S. Prediction of permeate flux and concentration
of two-component dye mixture in batch nanofiltration. J. Membr. Sci. 2003, 218, 195–210. [CrossRef]
23. Jamal, K.; Khan, M.A.; Kamil, M. Mathematical modeling of reverse osmosis systems. Desalination 2004, 160,
29–42. [CrossRef]
24. Ślȩzak, A. Irreversible thermodynamic model equations of the transport across a horizontally mounted
membrane. Biophys. Chem. 1989, 34, 91–102. [CrossRef]
25. Van Gauwbergen, D.; Baeyens, J. Modelling reverse osmosis by irreversible thermodynamics. Sep. Purif.
Technol. 1998, 13, 117–128. [CrossRef]
26. Sablani, S.S.; Goosen, M.F.A.; Al-Belushi, R.; Wilf, M. Concentration polarization in ultrafiltration and reverse
osmosis: A critical review. Desalination 2001, 141, 269–289. [CrossRef]
27. Gupta, V.K.; Hwang, S.T.; Krantz, W.B.; Greenberg, A.R. Characterization of nanofiltration and reverse
osmosis membrane performance for aqueous salt solutions using irreversible thermodynamics. Desalination
2007, 208, 1–18. [CrossRef]
28. Marquardt, D.W. An algorithm for least-squares estimation of nonlinear parameters. J. Soc. Ind. Appl. Math.
1963, 11, 431–441. [CrossRef]
29. Mendes, P.; Kell, D. Non-linear optimization of biochemical pathways: Applications to metabolic engineering
and parameter estimation. Bioinformatics 1998, 14, 869–883. [CrossRef]
30. Dong, J.; Lu, K.; Xue, J.; Dai, S.; Zhai, R.; Pan, W. Accelerated nonrigid image registration using improved
Levenberg–Marquardt method. Inf. Sci. 2018, 423, 66–79. [CrossRef]
31. Zhao, C.; Ma, P.; Zhu, C.; Chen, M. Measurement and correlation on diffusion coefficients of aqueous glucose
solutions. J. Chem. Ind. Eng. 2005, 56, 1–5.

© 2019 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/).