close

Вход

Забыли?

вход по аккаунту

?

Theenergy challenge of direct contact membrane distillation in low temperature concentration.

код для вставкиСкачать
ASIA-PACIFIC JOURNAL OF CHEMICAL ENGINEERING
Asia-Pac. J. Chem. Eng. 2007; 2: 400?406
Published online 10 August 2007 in Wiley InterScience
(www.interscience.wiley.com) DOI:10.1002/apj.072
Research Article
The energy challenge of direct contact membrane
distillation in low temperature concentration
V. A. Bui,1 * M. H. Nguyen1 and J. Muller2
1
2
Centre for Plant and Food Science, University of Western Sydney, Locked Bag 1797, Penrith South DC NSW 1797, Australia
Siemens Water Technologies, South Windsor, Australia
Received 24 October 2006; Revised 5 February 2007; Accepted 5 February 2007
ABSTRACT: Direct contact membrane distillation (DCMD) was operated at low temperatures from 25 to 40 ? C to suit
the purpose of thermally concentrating sensitive liquid foods, especially fruit juices to high solid content concentrate
with most of the quality attributes preserved.
A lab scale DCMD unit has been set up at the Centre for Plant and Food Science, University of Western Sydney.
Hollow fibre modules (HFM) using five types of fibres of polyvinylidene fluoride (PVDF) and Halar material, with
mass transfer areas ranging from 281 to 573 cm2 were employed. Experiments for concentration of glucose solutions
from 30 to 60% (w/w) were carried out. Results indicated that not only the operating conditions were important, but
also the membrane properties. It was found that Halar fibres were performing 2?3 times better than PVDF fibres in
term of removing water from the feed, and 3?4 times better in term of energy saving. Results also showed that an
increase of the feed inlet temperature from 25 to 40 ? C improved the mass flux up to 6 times and energy efficiency
(EE) up to 2.5 times depending on the feed concentration.
With flux up to 2.88 kg m?2 h?1 for PVDF and 5.83 kg m?2 h?1 for Halar fibres when concentrating 30% glucose
solution at 40 ? C, DCMD appeared to be an attractive concentration technique, when product quality is the priority.
However, with EE from as low as 2.1?14.9%, PVDF fibres employed in the study seemed not to be very suitable
for DCMD liquid food concentration under low temperature condition. DCMD in Halar fibres with EE up to 45.6%
still encounters the challenge of energy and could only be cost competitive to osmotic distillation and evaporative
concentration when cheaper energy sources or heat recovery measures are employed. ? 2007 Curtin University of
Technology and John Wiley & Sons, Ltd.
KEYWORDS: direct contact membrane distillation; DCMD; energy efficiency; hollow fibre
INTRODUCTION
Concentration is one important unit operation in the
liquid food industry, especially in fruit juice processing,
since it removes a significant amount of water, hence
a major reduction in packaging, transport and storage
cost with much greater stability of the concentrates;
and determines the quality of the final products. The
techniques for concentration and their advantages and
disadvantages are discussed elsewhere (Ramteke et al .,
1993; van Niestelrooij, 1998; Jiao et al ., 2004).
Osmotic distillation (OD) and direct contact membrane distillation (DCMD) have recently emerged as
alternatives to other concentration techniques when
high final concentration and quality are required. The
*Correspondence to: V. A. Bui, Centre for Plant and Food Science,
University of Western Sydney, Locked Bag 1797, Penrith South DC
NSW 1797, Australia.
E-mail: v.bui@uws.edu.au
? 2007 Curtin University of Technology and John Wiley & Sons, Ltd.
schematic principle of the two processes is shown in
Fig. 1.
In these processes, water molecules at the feedmembrane interface vaporise, then diffuse through the
membrane, condense at the membrane-stripping solution interface, and eventually are being swept away by
the stripping solution (Schofield et al ., 1987; Johnson
et al ., 1989; Hogan et al ., 1998).
The two processes differ in the method of creating
the driving force ? the water vapour pressure difference
across the membrane surfaces. While OD uses hygroscopic brine as a stripping solution, DCMD relies on the
temperature difference across the membrane to create
the process driving force.
From the energy point of view, OD has the advantage that the process is isothermal, thus minimal or no
heat transfer occurs. The process has been successfully
demonstrated in pilot plant operation for many fruit
juices such as apple and grape juice with final concentration up to 65?70? Brix (Hogan et al ., 1998; Bui
and Nguyen, 2005), and orange and passion fruit juices
Asia-Pacific Journal of Chemical Engineering
DIRECT CONTACT MEMBRANE DISTILLATION
?
Hydrophobic
membrane
Feed
solution
with high
water
vapour
pressure
Stripping
solution
with low
water
vapour
pressure
Water
vapour
transfer
through the
membrane
Figure 1.
DCMD.
Schematic principle of OD and
(Shaw et al ., 2001; Vaillant et al ., 2001). However, OD
has been encountering the problem of poor consumers?
perception, because of the use of brine solution being a
chemical, and is facing the potential problem of corrosion by the brine and the cost of its reconcentration.
DCMD does not use chemicals in the concentration
process, but uses cold water instead, and therefore overcomes the disadvantages of OD. However, since mass
driving force is due to the temperature difference across
the membrane, the DCMD process is disadvantageous
in losing heat during operation. DCMD has conventionally been applied in waste water treatment and desalination, operating at temperatures from 60 to 90 ? C (Calabro et al ., 1991; Lawson and Lloyd, 1997; Alklaibi
and Lior, 2004). In these areas of application, large P
can be created at relatively small T as illustrated in
Fig. 2. Therefore, operating DCMD at feed temperature
above 60 C, as in the case of waste water treatment
and desalination, will have the benefit of high mass
flux at relatively small heat loss because of conduction
across the membrane. Moreover, the small value of T
offers the options for heat recovery (Fane et al ., 1987;
Schneider et al ., 1988; Kurokawa and Sawa, 1996) and
utilizing waste heat (Drioli et al ., 1999) from other processes to reduce the cost of energy.
For application in concentrating thermally sensitive
liquid foods such as fruit juices, DCMD is required
to operate at low temperature of 25?40 ? C. Within
such a range of temperature, a competitive mass flux
could be achieved only if a much larger T across the
membrane is maintained. In other words, application of
DCMD in liquid food concentration may be challenged
by the problem of bigger heat loss. In addition, from
engineering point of view, large T between the
two streams incapacitates the option for internal heat
recovery and the availability of ?waste heat? within a
liquid food processing plant is questionable. Thus, in
this case a major concern about the heat inefficiency of
DCMD has been raised.
Since heat and mass transfer occur simultaneously
in DCMD, temperature and concentration polarisation
also occur in this process as demonstrated in Fig. 3.
These phenomena occur in a similar way as in OD
(Bui et al ., 2005) except that the bulk temperatures are
different. These temperature and concentration (on feed
side only) profiles, the determining factors for the heat
and mass driving force and therefore the heat efficiency
of DCMD, are affected by the hydrodynamic conditions
of the process such as operating condition and system
configuration. However, little is known about this effect
particularly when DCMD is operated at low temperature
range.
Therefore, a study was set out to determine the effect
of operating condition, as well as the membrane itself
on heat efficiency of DCMD process operating at low
temperature as a step forward for its application in the
food industry.
Water vapour pressure (kPa)
105
5癈
90
MATERIALS AND METHODS
6癈
75
Materials
7癈
60
10癈
45
DCMD experiments were carried out with the following
materials:
16癈
30
53癈
15
0
0
10
20
30
40
50
60
70
80
90
100
Temperature (癈)
The relationship between temperature and
water vapour pressure by Antoine equation (adapted from
Fernandez-Pineda et al., 2002). This figure is available in
colour online at www.apjChemEng.com.
Figure 2.
? 2007 Curtin University of Technology and John Wiley & Sons, Ltd.
? Feed: aqueous glucose solutions 30?60% total
solid (TS) prepared from glucodin powder with
99.9% purity (glucose) (Glucodin Energy Powder,
Australia).
? Permeate: distilled water kept unchanged at 10 ? C
through out the experiments. This temperature was
chosen on the basis that decreasing the permeate temperature below 10 ? C results in insignificant improvement of P across the membrane (Fig. 2) while it
Asia-Pac. J. Chem. Eng. 2007; 2: 400?406
DOI: 10.1002/apj
401
402
V. A. BUI, M. H. NGUYEN AND J. MULLER
Asia-Pacific Journal of Chemical Engineering
Cold water
Feed
Hollow fibre
10
ro
12
ri
Tf
1
Tf,m
Tp,m
Tp
Qp
9
Qm
Qf
8
Cf,m
5
Cf
Lumen side
8
11
Shell side
4
7
J
6
2
3
Figure 3. Temperature and concentration profiles at the
membrane interfaces of hollow fibres and heat fluxes in
DCMD (subscripts f for feed, p for cold water permeate and
m for membrane).
linearly proportionates the heat loss as well as the
potential problem of ice forming when temperature
approaches 0 ? C.
Setting up the lab scale DCMD unit
Five types of hollow fibres, obtained from Siemens
Water Technologies, Australia, with properties shown
in Table 1 were potted in modules, that is, hollow
fibre modules (HFM) with effective length 350 mm and
packing density 50%. The fibre bundles with 0.25 mm
thick mesh support were housed in a 1/2 brass pipe
which was insulated by a 15-mm thick insulation tube.
The DCMD unit, shown in Fig. 4, was equipped with
two diaphragm pumps with flow rate adjustable from 0
to 6 l min?1 through the flow regulating valves 11, a
hot water bath and a cold water bath with temperature
controlled within �2 ? C accuracy. Temperatures of
the two streams were recorded immediately at the inlets
Figure 4. Lab scale DCMD unit using hollow fibre module.
1 ? HFM; 2,3 ? feed and permeate bottles; 4,5 ? diaphragm
pumps; 6,7 ? hot and cool water baths; 8 ? flow meters;
9 ? pressure gauges; 10 ? thermometers; 11 ? flow regulating valves; 12?50 ml pipette for flux recording.
and outlets of the HFM. Flow rate of the glucose
feed was measured by oval gear flowmeter GM2RSP2RD (GPI-USA) with accuracy � of the reading,
while one of the cold water permeate side by gravity
flowmeter LMR1-10 006. The operating pressures of
the two streams were also observed throughout the
experiments so as not to exceed the critical limit of the
penetration pressure Pbp indicated in Table 1 to ensure
the integrity of the process.
Experiments were conducted under full factorial
design in regard to the operating conditions applied to
the feed at the shell side. The cold water permeate
stream was maintained at fixed inlet temperature of
Tpi = 10 ? C and velocity ?p = 1.2 m s?1 , and the two
streams were pumped counter currently through the
HFMs in all experiments. The feed was pumped through
the shell side to ease the problem of high pressure due
to high viscosity and velocity.
The factors under investigation in the study were as
follows:
? Feed concentration: 30, 40, 50 and 60%
? Feed inlet temperature: 25, 30, 35 and 40 ? C
Table 1. Properties of hollow fibres and HFMs.
Code
Material
di (祄)
do (祄)
? ? 祄
dp (祄)
Pbp (kPa)
? (%)
Ai (cm2 )
PV37
375
625
125
310
74
573.1
PV65
PVDF
650
1000
170
0.2
380
64
385.9
PV80
HL31
800
1300
250
310
650
170
HL50
Halar
500
800
150
0.1
350
60
281.5
400
75
436.3
450
85
467.3
(PVDF, polyvinylidene fluoride; Halar, copolymer of ethylene and chlorotrifluoroethylene).
? 2007 Curtin University of Technology and John Wiley & Sons, Ltd.
Asia-Pac. J. Chem. Eng. 2007; 2: 400?406
DOI: 10.1002/apj
Asia-Pacific Journal of Chemical Engineering
DIRECT CONTACT MEMBRANE DISTILLATION
? Feed velocity: 0.4, 0.6 and 0.8 m s?1 , controlled
via the flow rate of the feed stream (flow rate =
velocity � void area in the shell side). This range of
velocity was chosen because of the importance of
feed velocity in improving the process performance
(Bui and Nguyen, 2006) and the limit of the process
allowable pressure (Pbp ).
? Five HFMs with codes as listed in Table 1. The
effective length of the HFMs was fixed at 350 mm.
It should be emphasized that the concentration process by DCMD in this study was conducted at low
temperature aiming at its potential application in the
food industry.
Flux and energy efficiency determination
The mass flux (J ) is defined as the amount of water
transferred across membrane of one unit area in one
unit of time. It is usually expressed in (kg m?2 h?1 ),
(l m?2 h?1 ), (kg m?2 s?1 ), or (l m?2 s?1 ). In this
study, the DCMD flux (J ) was measured by the
indirect method used by Bui et al . (2004). However, the
readings in this study were measured by the increase
of water volume in the permeate bottle instead of
the decrease in the feed volume. This method of flux
recording has the advantage of avoiding the effect of
temperature on the accuracy since the permeate side was
kept constantly at 10 ? C throughout the experiments.
DCMD process is characterised by the simultaneous
heat and mass transfer with the associated concentration
and temperature polarisations as depicted in Fig. 3.
The total heat transferred from the feed to the permeate stream can be described by Eqn (1). It indicates the
energy balance involving heat transfer through the thermal boundary layers and the membrane wall in DCMD.
QT = Qf = Qm = Qp
(1)
The amount of heat transferred through the membrane
wall Qm in DCMD can be determined by Eqn (2). The
first part in this equation is due to conduction Qcond ,
and the second part is due to vapour diffusion Qdiff as
a result of mass transfer.
Qm = Ai � hm � (Tfm ? Tpm ) + Ai � J � Hv
(2)
Of these two parts, Qdiff is the useful energy, while
Qcond is considered as a loss of energy in this process.
Therefore, the energy efficiency (EE) of a DCMD
system is defined as given in Eqn (3).
EE =
Qdiff
Qdiff
=
QT
Qdiff + Qcond
(3)
While the useful energy Qdiff can be estimated as
defined in the second part of Eqn (2), the total heat QT
is due to temperature change from the inlet to the outlet
of the permeate stream as given in Eqn (4).
(4)
QT = ?p .?p .Al . Tpo ? Tpi
So, by controlling the flow rates and the inlet temperatures of the two streams, recording the temperature
changes between the inlets and outlets of HFM, and
observing the increase of the permeate volume from
the reading on the pipette 12 as shown in Fig. 4, one
can determine the mass flux and the heat efficiency of
the DCMD process.
RESULTS AND DISCUSSIONS
Lab scale experiments were carried out on all the five
HFMs specified in Table 1 under the conditions as
described in the experimental design. Mass flux was
measured and EE was calculated in accordance with
the above equations. Results indicating the performance
of the fibres at the extreme experimental conditions as
well as the average values over the range of operating
conditions are listed in Table 2.
It can be noticed that Halar fibres were performing
2?3 times betterthan polyvinylidene fluoride (PVDF)
fibres in terms of removing water from the feed,
and 3?4 times better in terms of EE even though
Halar fibres have smaller pore size. It again confirms
Table 2. DCMD performance of HFMs.
Module
PV37
PV65
Mass flux J (kg m?2 h?1 )
Max
2.88
2.30
Min
0.11
0.10
Average
1.16
0.94
Energy efficiency (EE ) (%)
Max
14.9
13.8
Min
2.1
2.9
Average
9.0
9.7
? 2007 Curtin University of Technology and John Wiley & Sons, Ltd.
PV80
HL31
HL50
1.66
0.16
0.73
5.13
0.30
1.91
5.83
0.39
2.17
10.5
3.4
7.3
34.1
7.8
21.7
45.6
13.0
29.8
Asia-Pac. J. Chem. Eng. 2007; 2: 400?406
DOI: 10.1002/apj
403
V. A. BUI, M. H. NGUYEN AND J. MULLER
Effect of feed velocity
Feed velocity is a much more important operating factor
for DCMD rather than for OD as the effect is due to both
reducing the polarisations at the membrane surfaces
and maintaining the temperature difference across the
membrane (Bui and Nguyen, 2006). This role is even
more important when the process is operated at low
temperature since large T has to be maintained as
discussed in the introduction section.
The effect of feed velocity is illustrated in Fig. 5. It
was found that feed velocity steadily improved the heat
efficiency of DCMD process. This effect was getting
stronger at higher feed concentration, though insignificant for all HFMs according to ANOVA analysis at
? = 0.015. Therefore, it should not be a serious concern for the level of feed velocity in regard to the EE,
but rather the mass flux of the process.
PV80
Energy efficiency - %'
the inapplicability of Knudsen and Poiseuille transport
models of gaseous water vapour through the membrane
when air is stagnantly present in the pores since mass
flux is proportional to the pore radius according to these
models. Instead, Fickian diffusion should be considered
as the governing law for this case as it has been proven
in an OD study of Bui et al . at similar temperature range
(Bui et al ., 2005). It is similar to the cases of OD studies
in which membrane pore size did not show any significant effect on flux (Mansouri and Fane, 1999; Narayan
et al ., 2002; Nagaraj et al ., 2006). The higher flux and
EE of Halar fibres may be due to the higher porosity,
lower thermal conductivity of the polymer material and
smaller tortuosity than those of PVDF fibres.
The measured EE up to 14.9% for PVDF fibres
appeared to be too low when compared to 1/1.32 or
75.8% of single effect evaporative concentration with
aroma recovery (Ramteke et al ., 1993) and presumably
of OD since reconcentration of the brine in this process
relies on evaporation. DCMD in Halar fibre modules
with EE up to 45.6% is still facing the challenge of
energy demand and therefore could only be cost competitive to OD and evaporative concentration in term
of energy demand when cheaper energy sources or heat
recovery measures are employed. It is worth noting that
the obtained flux for PVDF fibres fell well within the
data in literature while Halar fibres were first tested by
DCMD in this study. Details of the effect of operating
conditions are discussed in the following sections.
Asia-Pacific Journal of Chemical Engineering
45
40
35
30
25
20
15
10
5
0
0.2
Feed concentration plays an important role in determining the solution viscosity and water activity, thus
affecting the flow and water vapour pressure of the
? 2007 Curtin University of Technology and John Wiley & Sons, Ltd.
0.4
HL50
PV37
0.6
HL31
1
0.8
Effect of feed velocity ?f on DCMD
energy efficiency (for Cf = 30%, values are means at
different Tfi ). This figure is available in colour online at
www.apjChemEng.com.
Figure 5.
feed stream. Therefore, it was expected that feed
concentration had a negative effect on the performance
and EE of the DCMD process.
As shown in Fig. 6, feed concentration statistically
imposed an insignificant effect on EE for low flux
PVDF modules (? = 0.025) when feed concentration
changed from 30 to 50%, but significant for higher feed
concentration from 50 to 60%.
For Halar fibres where high mass flux occurred,
an increase in feed concentration significantly reduced
the EE of the process. This phenomenon indicates the
increase of heat resistance was much less than the
increase of mass resistance on the feed side because of
the increased feed concentration. It can be explained by
the severe effect of concentration polarisation in DCMD
when high flux occurs. Therefore, applying DCMD to
achieve high solid content will experience not only
reduced mass flux, but also lower heat efficiency.
Effect of feed inlet temperature
Feed inlet temperature generally improves DCMD process by increasing the driving force because of increased
PV80
PV65
HL50
PV37
HL31
40
35
30
25
20
15
10
5
0
25
Effect of feed concentration
PV65
Feed velocity - m.s-1
Energy efficiency - %'
404
30
35
40
45
50
55
60
65
Feed concentration - %
Figure 6.
Effect of feed concentration Cf on
DCMD energy efficiency (for ?f = 0.8 m s?1 and Tfi =
35 ? C). This figure is available in colour online at
www.apjChemEng.com.
Asia-Pac. J. Chem. Eng. 2007; 2: 400?406
DOI: 10.1002/apj
Asia-Pacific Journal of Chemical Engineering
DIRECT CONTACT MEMBRANE DISTILLATION
temperature gradient across the membrane, as well as
by reducing the viscosity of the feed stream.
ANOVA analysis of the data, shown in Fig. 7, indicated significant effect of feed temperature on the heat
efficiency of the process. The effect was more profound
in the case of high mass flux with Halar fibres. An
increase of feed temperature from 25 to 40 ? C improved
the EE of Halar fibres by 10?12%, while those figures
for PVDF fibres ranged 3.5?6%. Therefore, feed temperature plays an important role not only in the operation of DCMD, but also in increasing the EE of the
process.
Effect of membrane thickness
The membrane, a barrier between the feed and the
permeate in DCMD, is desirable to be thick enough to
maintain the T across the membrane (which create the
process?s mass driving force) and to reduce the heat loss
due to conduction through the membrane as well, and
as thin as possible to pose less mass resistance to allow
high flux. Therefore, it is important to find an optimal
PV80
PV65
PV37
HL50
HL31
Energy efficiency - %'
35
30
25
20
CONCLUSIONS
15
10
5
0
20
25
30
35
40
Feed inlet temperature - 癈
45
Figure 7. Effect of feed inlet temperature Tfi on
DCMD energy efficiency (for Cf = 50% and ?f =
0.8 ms?1 ). This figure is available in colour online at
www.apjChemEng.com.
0.4 m/s
Energy efficiency - %'
thickness of membrane (of known porosity, tortuosity
and material) for DCMD operation.
Figure 8 shows the effect of PVDF fibre thickness
on the heat efficiency of DCMD for glucose solution of
30% concentration at different temperatures and feed
velocities. Clearly, the fibre thickness had a significant
role in determining the heat efficiency of the DCMD
process. The highest EE achieved fell between 150
and 170 祄 depending on the operating feed velocity.
It was found that the peak migrated towards thinner
membrane when the feed velocity was increased. It
should be interpreted that thinner fibres are desirable
for DCMD operation at high feed velocity, while thicker
fibres are more suitable for DCMD operation at low feed
velocity. This finding is important for DCMD operation
since thin membrane should be considered for use in
concentration of low solid content solutions where high
feed velocity is allowable due to low viscosity, and thick
membrane for high solid content solutions. However,
the conclusion made here may be biased since the
porosity of the three PVDF fibres studied were not of
the same porosity which effectively affects the thermal
conductivity of the fibres, and subsequently the EE of
the process.
For Halar fibres, the optimal thickness might have
fallen below 150 祄 (Table 2). The conclusion could
not be drawn here since only two types of Halar fibres
were available.
0.6 m/s
0.8 m/s
12.0
11.0
10.0
9.0
8.0
7.0
120
140
160
180
200
220
Membrane thickness - mm
240
260
Effect of membrane thickness on DCMD
energy efficiency (for Cf = 30%; values are means at
different temperatures and feed velocities). This figure is
available in colour online at www.apjChemEng.com.
Figure 8.
? 2007 Curtin University of Technology and John Wiley & Sons, Ltd.
DCMD was successfully carried out to remove water
from 60% (w/w) glucose solution, showing its ability
to achieve high solid content concentrate at low temperature, hence applicable in concentration of thermally
sensitive liquid foods when high solid content and high
quality are the priority.
It was found that feed temperature significantly
improved DCMD performance in all the cases in terms
of mass flux and heat efficiency. Feed concentration,
however, reduced the EE significantly in the cases of
high flux such as those with Halar fibres. The effect was
more profound at higher range of feed concentration.
The study also found that there was an optimal
thickness of membrane in DCMD application. Thick
membrane is more suitable for low feed velocity DCMD
operation, while thin membrane is desirable in high feed
velocity operation.
With flux up to 2.88 kg m?2 h?1 for PVDF and
5.83 kg m?2 h?1 for Halar fibres when concentrating
30% glucose solution at 40 ? C, DCMD appears to be
an attractive alternative concentration technique, when
product quality is the priority. PVDF fibres, employed
in the study, appeared to result in too low EE when
compared to that of single effect evaporative concentration, and therefore, are not very suitable for liquid
Asia-Pac. J. Chem. Eng. 2007; 2: 400?406
DOI: 10.1002/apj
405
406
V. A. BUI, M. H. NGUYEN AND J. MULLER
food concentration under low temperature conditions.
DCMD with Halar fibres with EE up to 45.6% still
encounters the challenge of energy demand, and could
have to overcome the challenge of energy and be cost
competitive to OD and evaporative concentration when
cheaper energy sources or heat recovery measures are
employed.
NOMENCLATURE
The dimensions of all the nomenclatures in this paper
are in SI units unless specified.
Internal mass transfer area of HFMs
Ai
Cross section of permeate flow
Al
External diameter of fibres
do
Internal diameter of fibres
di
Nominal pore size
dp
h
Heat transfer coefficient
EE Energy efficiency, %
J
Mass flux
Pbp Bubble point or liquid penetration pressure
T
Temperature
Q
Heat flux
HV Latent heat of vaporisation of water
GREEK SYMBOLS
?
?
?
?
Thickness of the fibre wall
Velocity of the fluid streams
Fluid density
Porosity of membrane
SUBSCRIPTS
f, p
m
i, o
For feed and permeate stream, respectively
For membrane and membrane surface
For inlet and outlet, respectively
Acknowledgements
The authors would like to acknowledge the University
of Western Sydney and the Australian government for
providing scholarship and the facilities, and Siemens
Water Technologies Australia for providing with the
hollow fibres.
Asia-Pacific Journal of Chemical Engineering
Bui AV, Nguyen HM. Scaling up of osmotic distillation from
laboratory to pilot plant for concentration of fruit juices.
Int. J. Food Eng. 2005; 1: Article 5; 20 pp. Available at:
http://www.bepress.com/ijfe/vol1/iss2/art5.
Bui AV, Nguyen MH. The role of operating conditions in
osmotic distillation and direct contact membrane distillation ? a
comparative study. Int. J. Food Eng. 2006; 2: Article 1; 14 pp.
Available at: http://www.bepress.com/ijfe/vol2/iss5/art1.
Bui VA, Nguyen MH, Muller J. A laboratory study on glucose
concentration by osmotic distillation in hollow fibre module. J.
Food Eng. 2004; 63(2): 237?245.
Bui VA, Nguyen MH, Muller J. Characterisation of the polarisations
in osmotic distillation of glucose solutions in hollow fibre module.
J. Food Eng. 2005; 68(3): 391?402.
Calabro V, Drioli E, Matera F. Membrane distillation in the textile
wastewater treatment. Desalination 1991; 83(1?3): 209?224.
Drioli E, Lagana F, Criscuoli A, Barbieri G. Integrated membrane
operations in desalination processes. The International Workshop
on Desalination Technologies for Small and Medium Size
Plants With Limited Environmental Impact, Desalination 1999;
122(2?3): 141?145.
Fane AG, Schofield RW, Fell CJD. The efficient use of energy in
membrane distillation. Desalination 1987; 64: 231?243.
Fernandez-Pineda C, Izquierdo-Gil MA, Garcia-Payo MC. Gas
permeation and direct contact membrane distillation experiments
and their analysis using different models. J. Membr. Sci. 2002;
198(1): 33?49.
Hogan PA, Canning RP, Peterson PA, Johnson RA, Michaels AS. A
new option: osmotic distillation. Chem. Eng. Prog. 1998; 94(7):
49?61.
Jiao B, Cassano A, Drioli E. Recent advances on membrane
processes for the concentration of fruit juices: a review. J. Food
Eng. 2004; 63(3): 303?324.
Johnson RA, Valks RH, Lefebvre MS. Osmotic distillation ? a low
temperature concentration process. Aust. J. Biotechnol. 1989; 3(3):
206?207 & 217.
Kurokawa H, Sawa T. Heat recovery characteristics of membrane
distillation. Heat Trans. Jpn. Res. 1996; 25(3): 135?150.
Lawson KW, Lloyd DR. Membrane distillation. J. Membr. Sci.
1997; 124(1): 1?25.
Mansouri J, Fane AG. Osmotic distillation of oily feeds. J. Membr.
Sci. 1999; 153(1): 103?120.
Nagaraj N, Patil G, Babu BR, Hebbar UH, Raghavarao KSMS,
Nene S. Mass transfer in osmotic membrane distillation. J.
Membr. Sci. 2006; 268(1): 48?56.
Narayan AV, Nagaraj N, Hebbar UH, Chakkaravarthi A, Raghavarao KSMS, Nene S. Acoustic field-assisted osmotic membrane
distillation. Desalination 2002; 147(1?3): 149?156.
Ramteke RS, Singh NI, Rekha MN, Eipeson WE. Methods for
concentration of fruit juices: a critical evaluation. J. Food Sci.
Technol. 1993; 30(6): 391?402.
Schneider K, Holz W, Wollbeck R, Ripperger S. Membranes and
modules for transmembrane distillation. J. Membr. Sci. 1988;
39(1): 25?42.
Schofield RW, Fane AG, Fell CJD. Heat and mass transfer in
membrane distillation. J. Membr. Sci. 1987; 33(3): 299?313.
Shaw PE, Lebrun M, Dornier M, Ducamp MN, Courel M,
Reynes M. Evaluation of concentrated orange and passionfruit
juices prepared by osmotic evaporation. Lebensm.-Wiss. Technol.
2001; 34(2): 60?65.
Vaillant F, Jeanton E, Dornier M, O?Brien GM, Reynes M,
Decloux M. Concentration of passion fruit juice on an industrial
pilot scale using osmotic evaporation. J. Food Eng. 2001; 47(3):
195?202.
van Niestelrooij M. Freeze concentration and its application in the
fruit juice industry. Fruit Process. 1998; 11(8): 460?462.
REFERENCES
Alklaibi AM, Lior N. Membrane-distillation desalination: status and
potential. Desalination 2004; 171(2): 111?131.
? 2007 Curtin University of Technology and John Wiley & Sons, Ltd.
Asia-Pac. J. Chem. Eng. 2007; 2: 400?406
DOI: 10.1002/apj
Документ
Категория
Без категории
Просмотров
4
Размер файла
142 Кб
Теги
theenergy, challenge, concentrations, low, temperature, direct, contact, membranes, distillation
1/--страниц
Пожаловаться на содержимое документа