Process intensification with selected membrane processes

Accepted Manuscript Title: Process intensification with selected membrane processes Author: Kamalesh K. Sirkar Anthony G. Fane Rong Wang Ranil Wickramasinghe PII: DOI: Reference:

S0255-2701(14)00214-1 http://dx.doi.org/doi:10.1016/j.cep.2014.10.018 CEP 6494

To appear in:

Chemical Engineering and Processing

Received date: Revised date: Accepted date:

25-5-2014 4-10-2014 29-10-2014

Please cite this article as: Kamalesh K.Sirkar, Anthony G.Fane, Rong Wang, Ranil Wickramasinghe, Process intensification with selected membrane processes, Chemical Engineering and Processing http://dx.doi.org/10.1016/j.cep.2014.10.018 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Process Intensification with Selected Membrane Processes Kamalesh K. Sirkar*1, Anthony G. Fane2,4 , Rong Wang2,5, Ranil Wickramasinghe3,6 1

Chemical, Biological and Pharmaceutical Engineering New Jersey Institute of Technology Newark, NJ 07102, USA 2 School of Civil and Environmental Engineering Singapore Membrane Technology Center Nanyang Environment and Water Research Institute Nanyang Technological University Singapore 639798, Singapore 3 Ralph E. Martin Department of Chemical Engineering University of Arkansas Fayetteville, AR 72701-1201, USA

RI

PT

*Corresponding author. Tel. 1-973-596-8447; fax: 1-973-642-4854; [email protected] Email addresses of other authors: [email protected]; [email protected]; 6 [email protected]

U

Membrane-based phase contacting achieves high levels of process intensification (PI). Combining two separations on two sides, membranes achieve process intensification. Combining reaction and separation, membrane bioreactors deliver high levels of PI. Membrane distillation, forward osmosis and pressure retarded osmosis achieve high PI.

N

1. 2. 3. 4.

SC

Highlights

M

A

Abstract Membrane devices, membrane processes and membrane-based conventional chemical engineering processes have achieved extraordinary levels of process intensification (PI).

D

Generally membrane-based devices require smaller equipment to achieve a given device

TE

production rate. Further they can eliminate dispersion-based operation and achieve extraordinary

EP

selectivity. Exploiting the compartmentalization of two regions on two sides of the membrane,

CC

membrane devices can combine two processes carried out on two sides of the membrane in one membrane device. Selected membrane processes and their applications are briefly reviewed here

A

in terms of the level of PI achieved. The membrane processes selected are generally recently commercialized or being commercialized or have great potential for commercialization: Membrane bioreactor; Membrane gas-liquid contacting; Membrane solvent extraction; Forward osmosis; Pressure-retarded osmosis; Membrane distillation; Membrane distillation bioreactor.

Examples of potential process intensification by membrane processes are briefly illustrated for processing of lignocellulose to biofuels in a biorefinery and for produced water treatment. Keywords: Membrane processes; Membrane contactors; Membrane distillation; Membrane bioreactor; Process intensification 1. Introduction Process intensification is of increasing interest in chemical process and allied industries.

PT

Several definitions of process intensification exist in the literature. One of the earliest uses of the

RI

phrase “Process Intensification” appeared in an article by Ramshaw (1983) which described a rotating packed bed for gas-liquid contacting (HIGEE) for which US patents were issued 1981

SC

and 1982. Ramshaw (1983) identified process intensification (PI) as a strategy for reducing the

U

plant size for a given production goal; correspondingly PI also stands for reducing equipment

N

size for a given performance level in terms of the mass transfer rate or the reaction rate per unit

A

equipment volume. Stankiewicz and Moulijn (2000) defined PI to be any chemical engineering

M

development that leads to a “substantially smaller, cleaner and more energy-efficient

D

technology”. However they specified that PI should concern only engineering methods and

TE

equipment and not a new chemical route or a new catalyst. Therefore goals like atom economy and green chemistry are not part of PI. Sustainability may still be included as long as a cleaner

EP

and more energy-efficient technology has been achieved since those two are integral parts of

CC

sustainability.

If one considers the expositions of PI provided by Stankiewicz and Moulijn (2000) or that by

A

Charpentier (2007), it becomes clear that their focus is on intensifying traditional processes and devices employed in chemical engineering especially reaction engineering. Membranes appear in their lists only through membrane reactors, membrane contactors, membrane absorption,

membrane distillation and membrane crystallizer. In their papers, there is no mention of membrane processes such as reverse osmosis for which there does not appear to be any analog in classical chemical engineering processes and devices. Yet reverse osmosis (RO), for example RO desalination, should qualify as an example of extraordinary process intensification compared to the classical methods of thermal desalination since there is no new chemical route or a new catalyst.

PT

In reverse osmosis, the notion of classical filtration has merely been taken to its logical

RI

conclusion in terms of an increasingly smaller particle size being separated ultimately ending with molecular level filtration. In fact one of the terms originally proposed for what goes under

SC

the title “reverse osmosis” was “hyperfiltration” (Spiegler and Kedem, 1966) since in actual

U

reverse osmosis one cannot reverse the salt leakage that inevitably takes place during osmosis.

N

Therefore the conventional membrane processes of nanofiltration, ultrafiltration and

A

microfiltration are also logical candidates for PI along with hyperfiltration.

M

In the PI framework by Stankiewicz and Moulijn (2000), any development in the process

D

equipment or method that contributes to dramatic improvements in manufacturing and

TE

processing qualifies as process intensification. It must ultimately result in a cheaper and more sustainable technology. There are two specific types of developments. The first type of

EP

development leads to dramatic improvements in processing by eliminating or bypassing

CC

limitations of conventional chemical engineering equipment. A second type of development consists of a combination of two or more processes or functions in one device or process

A

resulting in PI. It is expected that such developments will lead to smaller/compact devices for the same production goal.

Illustration of the first type of development leading to dramatic improvements in processing by eliminating or bypassing limitations of conventional equipment is provided by the processes of membrane absorption, membrane extraction and membrane stripping. The devices for such processes are broadly identified as membrane contactors. In conventional devices based on phase dispersion, there are limits of phase flow rate ratios beyond which there would be flooding. By having immobilized gas-liquid interface (Esato and Eisemann, 1975; Tsuji et al., 1981) or liquid-

PT

liquid interface (Kiani et al., 1984) at the pore mouths of porous hydrophobic membranes, the

RI

limitation on the flow rate ratio of the two phases is removed as long as appropriate pressure conditions are maintained and breakthrough pressure difference between the phases is avoided.

SC

There are a number of other basic advantages in such membrane contactors: A much enhanced

U

surface area per unit equipment volume leading to highly compact devices when hollow fiber

N

membranes are used; modular equipment for easy scale up or scale down; nondispersive

A

operation eliminating foaming and weeping in gas-liquid systems, prevention of emulsification

M

in liquid-liquid systems; elimination of the need for coalescence in liquid-liquid contacting

D

processes; no need for density difference in liquid-liquid systems.

TE

The earliest illustrations of the second type of development are provided by membrane distillation (MD) (Gore, 1982; Schneider and van Gassel, 1984; Schofield et al., 1987) and

EP

membrane crystallization (MC) (Wu and Drioli, 1989). The process of membrane distillation,

CC

especially direct contact membrane distillation (DCMD), combines evaporation of a volatile species on one surface of a porous non-wetted hydrophobic membrane with its condensation in

A

the distillate stream on the other surface of the membrane; the primary application has been removal of water from a hot brine into a cold distillate stream. In air gap membrane distillation (AGMD), the two functions take place in two distinct parts of one separation device: evaporation

takes place on one surface of the porous hydrophobic membrane as the water vapor emerging from the other side of the non-wetted membrane encounters a contiguous cold surface and condenses. In membrane crystallization, the process of membrane distillation is utilized to concentrate the solution initiating crystallization. An even earlier illustration of the second type of development is found in a supported liquid membrane (SLM) or an immobilized liquid membrane (ILM) or later in a contained liquid

PT

membrane (CLM). In a SLM, the pores of a porous/microporous membrane are filled with a

RI

liquid which is held by capillary forces. Feed liquid solution meant to undergo solvent extraction contacts one surface of the pore liquid on one side of the membrane where solvent extraction

SC

takes place. The feed liquid and the pore liquid (solvent) must be immiscible. On the other side

U

of the membrane, back extraction of the extracted solute takes place from the pore solvent into a

N

back extraction solvent with which the pore liquid must be immiscible. The phrase ILM is used

A

more often when a gas mixture is separated through the liquid membrane immobilized in the

M

pores of the support membrane. Contained liquid membrane achieves separation in a similar

D

fashion except the liquid membrane is now placed between two separate membranes or two

TE

separate sets of porous hollow fibers. Membrane reactors illustrate also a combination of two or more processes/functions in one

EP

device. The functions include among others combining multiple reactions on two sides of the

CC

membrane, combining reaction with separation or separation with reaction, immobilizing or segregating a catalyst along with separation and reaction. Distributed introduction of reactants

A

through the membrane, nondispersive operation, heat exchange, functioning as an electrode etc. are additional functions implemented via membranes in a membrane reactor. The large number of functions that can be carried out in a membrane reactor for chemical, biochemical,

biopharmaceutical and petrochemical manufacturing has been summarized in Sirkar et al. (1999). An early review of membrane bioreactors is available in Cheryan and Mehaia (1986). Although there are quite a few large-scale applications of membrane bioreactors in amino acid production and pharmaceutical synthesis, the largest volume application is the membrane bioreactor (MBR) for municipal water treatment. In a MBR, the suspended growth activated sludge system utilizes a microporous membrane for ultrafiltration which carries out solid/liquid

PT

separation and eliminates the secondary clarifiers.

RI

In the next section we focus on selected individual membrane processes and membranebased processes and provide illustrations of the process intensification achieved. When we refer

SC

to membrane-based processes ideally we refer to conventional processes in chemical engineering

U

such as absorption, desorption, stripping, extraction, back extraction, distillation, crystallization

N

and chemical reaction implemented using a membrane. On the other hand, any type of filtration

A

process using a membrane is identified as a membrane process. We do not dwell much on this

M

distinction in the next section as individual processes are considered. References to appropriate

D

literature as well as important reviews will be added for each process. In the last section two

TE

potential examples for PI by membrane processes are provided: The process route to biofuels from lignocellulose in a biorefinery; produced water from oil exploration. Both illustrate how

EP

membrane processes can introduce considerable PI in challenging separation applications.

CC

2. Selected Membrane Processes for Process Intensification We first identify the membrane processes and devices of interest here. Our primary

A

criteria for process selection are as follows. These are new membrane processes or membranebased processes and are either commercialized or being commercialized. Some processes have been commercialized over the last decade and a half: Membrane absorbers, membrane

bioreactors, membrane extractors, membrane strippers. The processes of membrane distillation (MD), forward osmosis (FO) and pressure-retarded osmosis (PRO) are being commercialized. 2.1 Membrane Bioreactor (MBR) The membrane bioreactor for wastewater treatment is a classic example of a hybrid membrane process and PI. The conventional activated sludge process (CASP) involves an aerobic suspended growth bioreactor followed by a settling tank to provide a settled treated

PT

wastewater of low BOD. The settled sludge is returned to the bioreactor and a small fraction is

RI

discharged, as ‘waste sludge’, to maintain a constant level of mixed liquor suspended solids (MLSS), or biomass, in the reactor. In the CASP the concentration of MLSS is a compromise as

SC

higher values improve the biological removal of BOD but also lead to less efficient settling and

N

U

residual SS in the treated water. A value of MLSS of about 3 - 5 g/L is typical.

A

The advent of ultrafiltration membranes in the late 1960s provided the opportunity to replace

M

the settling tank with a more efficient solid/liquid separation. Initially the approach was to

D

consider MLSS values > 20 g/L as settling was no longer an issue and high MLSS values would

TE

enhance BOD removal and reduce the amount of ‘waste sludge’. However the disadvantages of very high MLSS were soon appreciated as less efficient oxygen input, high viscosity for

EP

pumping and potentially more membrane fouling. A compromise for current generation MBRs is

CC

MLSS values of 10 to 15 g/L. The early MBRs were also designed to operate at relatively high fluxes (> 50 L/m2hr) due to membrane costs. This required high crossflow velocities through the

A

side-stream membrane modules and typical energy demands were around 10 kWh/m3 permeate. As a result the MBR was a niche application until the early 1990s. At that time the concept of the

submerged membrane bioreactor was introduced by Yamamoto and co-workers (1989). This approach was rapidly adopted and further developed, with the following features; (i) the membranes, either hollow fibers or flat sheets, are submerged directly in the submerged growth bioreactor and permeate is removed by suction; (ii) operational fluxes are now reduced to about 15 to 25 L/m2hr, due to the lower unit costs of membrane modules;

PT

(iii) fouling is controlled by coarse bubble aeration.

RI

These parameters have been optimized so that energy demand is now down to about < 0.4 kWh/m3 (Tao, 2014). This value approaches that for the conventional activated sludge process.

SC

Since the early 1990s the MBR has experienced exponential growth for both municipal and

U

industrial wastewaters and there are probably more than 3000 MBRs worldwide with sizes upto

N

> 100 MLD, and a market of > $500m pa. The submerged membrane configuration is depicted in

A

Figure 1 (a) and the modern version of the side stream membrane MBR is in Figure 1 (b). An

M

important feature of the modern side stream system is that the feed to the module is two-phase

D

flow, with a combined mixed liquor and injected air stream. This has allowed operation at a

TE

somewhat higher flux than submerged membranes with reasonable energy demand. The side stream MBRs tend to use capillary (large bore) fibers operating inside-out, and the submerged

EP

MBRs operate outside-in with hollow fibers or with vertical flat sheets. The municipal MBRs

CC

tend to use submerged hollow fibers and the industrial MBRs use both submerged hollow fibers and flat sheets and side stream systems, although there exceptions to these observations. For

A

more information on submerged membranes see Fane (2008). The wastewater MBR satisfies most of the PI criteria. Table 1 compares CASP and the MBR

data in terms of loading, footprint and performance. The MBR is more compact and delivers

PT RI

EP

TE

D

M

A

N

U

SC

Figure 1(a). MBR with submerged membrane module.

CC

Figure 1 (b). MBR with a side-stream membrane module.

A

significantly better performance. Although the energy demand for MBRs is more than CASP it could be equal or better if the CASP effluent was combined with MF/UF to achieve similar TSS removals. MBRs can be operated at higher sludge retention times (SRTs) than CASPs, and the sludge waste decreases as SRT increases. On the issue of overall costs of processing it is

2.2 Membrane Gas-Liquid Contacting (Absorption, Stripping, Degassing) Blood oxygenation provided probably the earliest example of nondispersive gas-liquid contacting via two types of gas-filled porous membranes: Porous hydrophobic flat Goretex® membrane (Esato and Eiseman, 1975) and porous hydrophobic polypropylene hollow fibers (Tsuji et al., 1981). The first illustration of its application to chemical industry involved removal of CO2 from a gas stream flowing on one side of porous hydrophobic hollow fibers of

PT

polypropylene (PP) by absorption and reaction with an aqueous solution of caustic flowing on

RI

the other side (Qi and Cussler, 1985a, 1985b). An additional application using the same hollow fiber membrane involved supplying O2 to and removing CO2 from the fermentation medium in a

SC

continuous tubular fermenter for alcohol production (Frank and Sirkar, 1985, 1986); these are the

U

earliest examples of putting porous hollow fiber membranes submerged inside a bioreactor.

A

also demonstrated (Semmens et al., 1989).

N

Stripping of volatile organic compounds (VOCs) from water through similar membranes was

M

The most important advantages of such processes are: (1) Independent variation of the flow

D

rates of the gas phase and the liquid phase on two sides of the membrane without any loading or

TE

flooding (Qi and Cussler, 1985b; Frank and Sirkar, 1985) as long as the excess pressure of the liquid phase over that of the gas phase does not exceed a critical value so that the membrane

EP

pores remain gas-filled (Sirkar, 1992); (2) High surface area per unit equipment volume of as

CC

much as 50 cm2/cm3 (Qi and Cussler, 1985b); (3) No foaming or weeping; (4) High values of Kl a (10-30 times larger than packed bed values for CO2 absorption in water, (Karoor and Sirkar,

A

1993)) or low values of HTU (7-20 cm for SO2 absorption in water, (Karoor and Sirkar, 1993)); (5) The membrane contactor may be as much as 10 times plus smaller and lighter than a nonmembrane contactor having similar performance (for ammonia absorption in water at 7000

lb/min (Chem. Eng., 2001)); (6) Modular devices allowing easy scale up or scale down; (7) Elimination of foaming and weeping. Reviews of such membrane contactors are available in Sirkar (1992), Reed et al. (1995), Gableman and Hwang (1999), Sengupta and Pittman (2009) and Criscuoli and Drioli (2009). Numerous applications of membrane contactor-based gas-liquid contacting can be found in industry. A very large installation involves deoxygenation application in ultrapure water

PT

production where 48 hollow fiber modules are being used; each module (14x28 PVC Contactor;

RI

flange to flange length of 45 inch) has a vessel diameter of 14 inch with a membrane surface area of 220 m2 (Sirkar, 2008) (Figure 2); on the gas side a high vacuum is used sometimes with a bit

SC

of nitrogen (Sengupta et al., 1998). Most industrial applications employ polypropylene hollow

A

CC

EP

TE

D

M

A

N

U

fibers and nonwetting solutions usually aqueous in nature.

Figure 2. Commercial water degassing system with 48 Liqui-Cel® modules each having a surface area of 220 m2 (Courtesy of Membrana, Charlotte, NC).

In applications of acid-gas removal an extensive amount of research is being conducted involving a variety of membranes and membrane materials (Zhang and Wang, 2013) at various pressures and temperatures of operation (Jie et al., 2014) using different classes of absorbents (Kumar et al., 2002; Kosaraju et al., 2005; Albo et al., 2010; Chau et al., 2014). The key issue there is to avoid pore wetting by the absorbent liquid and still have high transfer rates (Wang et al., 2005). Thus the ideal membranes used for membrane contactor-based acid-gas removal

PT

should be highly hydrophobic, and possess a relatively small maximum pore size and narrow

RI

pore size distribution based on the Laplace-Young equation (Franken et al., 1987; Zhang and Wang, 2014). In addition, the membrane should have good chemical and thermal resistances, as

SC

the absorbent liquids with high CO2 loading capacity are normally highly corrosive with low

U

compatibility to commonly used polymeric materials (Wang et al., 2004). In addition, developing

N

novel non-volatile absorption liquids that possess high CO2 loading capacity and high

A

compatibility to polymeric membranes used in the contactor is also critical to facilitate practical

M

application of membrane contactor technology (Zhang and Wang, 2014).

D

Optimizations of fluid dynamics and membrane module design are also important, as the

TE

mass transfer coefficients in the boundary layers close to both of membrane surfaces affect the overall performance of gas transfer based on the resistance-in-series model. Several types of

EP

membrane module configurations have been reported including longitudinal flow module, cross-

CC

flow module, and coiled module (Wickramasinghe et al., 1992). However, most attention has been on the membrane absorption process and device; much less attention has been paid to

A

membrane stripping. 2.3 Membrane Solvent Extraction

The earliest example of potential process intensification in chemical industry was illustrated in membrane solvent extraction by Kiani et al. (1984). A porous hydrophobic flat membrane was used to carry out nondispersive solvent extraction of acetic acid from its aqueous solution which was flowing on one side of the membrane into the solvent wetting the pores of the membrane and flowing on the other side of the membrane. The aqueous solution pressure was either equal to or higher than that of the organic solvent phase; the aqueous-organic phase interface was

PT

immobilized at the pore mouth on the aqueous side of the membrane. No dispersion of one phase

RI

as drops in the other phase was needed; there was no need either for a density difference between the two phases. Therefore the problems of dispersion and coalescence were dispensed with. To

SC

ensure such conditions, the excess aqueous phase pressure should not exceed that of the organic

U

phase by a certain amount. The phase pressure conditions are reversed when a porous

N

hydrophilic membrane is used with the aqueous phase in membrane pores (Prasad and Sirkar,

A

1987).

M

When Kiani et al. (1984) calculated the volumetric mass transfer coefficients for acetic acid

D

extraction achievable using different values of the porous hydrophobic hollow fiber based

TE

surface area per unit extractor volume, the values turned out to be 3-14 times that achieved in a perforated plate column. Such potentials were experimentally demonstrated by Prasad and Sirkar

EP

(1990) using hydrophobic hollow fiber extractor modules where the height of transfer unit

CC

(HTU) values as low as 3 cm was achieved. The earliest review of membrane solvent extraction is available in Prasad and Sirkar (1992). Reid et al. (1995) provide a perspective of membrane

A

solvent extraction as well. Gableman and Hwang (1999) provide a secondary review. There have been numerous studies on reactive extraction of heavy metals and organic acids amongst others. Schlosser et al. (2005) provide an overview of organic acid extractions. Ortiz and Irabien (2009)

provide an overview of modeling of membrane-assisted solvent extraction for metallic pollutants. There are a few hollow fiber-based commercial membrane solvent extraction units around the world; some of them have been listed/illustrated in Sirkar (2008). Figure 3 illustrates a small one (Klassen and Jansen, 2001) employing 4 inch diameter PP hollow fiber-based modules. Most utilize porous hydrophobic PP hollow fibers; one utilizes hydrophilic polyacrylonitrile hollow

PT

fibers for diltiazem production. Hydrophobic hollow fibers possessing solvent resistance higher

RI

than that of PP are desirable for many applications. For applications involving back extraction from an organic solvent, hydrophilic hollow fibers containing the aqueous phase in the pores are

CC

EP

TE

D

M

A

N

U

SC

needed.

A

Figure 3. Recovery of aromatic compounds from polluted water coming from a reactor by membrane solvent extraction; Plant capacity: 15 m3/hr (KoSa Netherlands BV, Netherlands). (Reprinted with permission from Klassen and Jensen (2001). Coypyright 2001 American Institute of Chemical Engineers). (Four inch diameter vertically placed membrane modules are at the far end). 2.4 Forward Osmosis

Different from pressure-driven RO processes, forward osmosis (FO) is an osmotically driven natural process for water transfer through a semi-permeable membrane under an osmotic pressure gradient, which arises from a feed water of low concentration and a draw solution of high concentration, across the membrane. In recent years, FO has received intensive studies for a range of potential applications, including seawater/brackish water desalination, wastewater treatment, food and pharmaceutical processing as well as energy production (McCutcheon and

PT

Wang, 2013). Because of the unique features of FO, most of the FO processes qualify as process

RI

intensification which is elaborated as follows.

The idea of using FO for seawater/brackish water desalination was proposed in the 1970s

SC

(Kessler and Moody, 1976). When commercial HTI FO membranes became available during the

U

1990s, there was renewed interest in FO for water production in the membrane community (Cath

N

et al., 2006). The major advantages of FO for seawater desalination are the potentially lower

A

electrical energy consumption (waste heat can be used) and fouling tendency of the FO process

M

as compared with the RO process. However, the key challenges have been the lack of an optimal

D

FO membrane that can mitigate concentration polarization to achieve improved water flux, and

TE

ideal draw solutions that can extract the water from the diluted draw solution and be regenerated with less energy requirement. Efforts have been made to develop novel FO membranes (Wang et

EP

al., 2010; Setiawan et al., 2012), and to identify and invent good draw solutes such as

CC

thermolytic ammonium bicarbonate (NH4HCO3) that can be regenerated by heating (McGinnis, 2002), water-soluble salts or particles that can be recovered by integrating with other processes

A

like membrane distillation (MD) or nanofiltration (NF) (Zhao et al., 2012). However, many potential applications can use available draws, such as seawater or brine. The FO process then provides low energy process intensification and there are several examples of this use of FO. The

FO-RO integrated system is attractive, where the FO unit serves as an osmotic dilution process using impaired water to reduce the concentration of seawater/RO brine that can go through the RO unit for desalination to achieve a higher water recovery as compared to a single stage RO process (Cath and Childress, 2011). A recent analysis (Sim et al, 2013), described below (see Pressure Retarded Osmosis (PRO)) illustrates the benefits of a hybrid process combining FO+RO+PRO for lower energy water production.

PT

A similar concept to osmotic dilution is osmotic concentration, which extends the FO

RI

process as an effective pre-treatment process where the diluted draw solution does not require regeneration. Using seawater or RO brine as draw solution, FO can be applied in dewatering or

SC

concentrating a variety of wastewater resources such as landfill leachates and activated sludge

U

from anaerobic digesters, which will be subsequently further treated in other processes

N

(Holloway et al, 2007). Thus the footprint of the next-stage process can be reduced significantly

A

and the potential transportation cost can be decreased considerably. Another interesting and

M

smart FO process involving osmotic dilution and concentration is to use fertilizers as the draw

D

solute and the diluted fertilizers are then directly applied for fertigation as products (Phuntsho et

TE

al, 2012).

Osmotic concentration can also be used for the enrichment of aqueous pharmaceutical

EP

products such as protein and lysozyme, and the dewatering of liquid foods such as fruit juices to

CC

improve their quality and reduce storage and transportation costs (Petrotos and Lazarides, 2001). These products are normally heat sensitive and have large molecular sizes. The removal of water

A

from the aqueous stream using an osmotic process offers many advantages over conventional chemical or thermal concentration approaches, which include high purity of products due to the RO or NF-like dense selective layer of FO membrane (Shi, et al., 2011; Setiawan et al., 2013),

and unchanged original physical properties of products because of the elimination of chemicals and heating as well as the inherent “natural” feature of the FO process. An important merit of the osmotic dewatering/concentration process is that it is energyfriendly. Since the concentrates of the FO process are the target products, there is a great flexibility for choosing an appropriate draw solute, and no separation is required for the diluted draw solution. It can be seen that osmotic concentration illustrates well the concept of process

PT

intensification in terms of reduced footprint, elimination of conventional chemical engineering

RI

equipment and less energy consumption. However, optimized FO membranes and suitable draw solutes are needed to realize these processes for practical applications.

SC

2.5 Pressure Retarded Osmosis

U

It is well understood that a significant amount of energy is released when two streams with

N

different salinities, such as river water and seawater, are mixed. To harvest the salinity gradient

A

energy, one type of osmotic process, pressure retarded osmosis (PRO), is considered as one of

M

the promising techniques (Logan and Elimelech, 2012).

D

The PRO is an intermediate osmotic process between the FO and RO. Similar to the RO,

TE

an external hydraulic pressure is applied on the high salinity draw solution side, but the hydraulic pressure is lower than the osmotic pressure gradient across the membrane. As a result, the net

EP

water flow is in the direction from low salinity feed water toward the pressurized high salinity

CC

draw solution, similar to FO. The increased volume of pressurized draw solution can be utilized

A

to do useful work such as driving a hydroturbine for power generation (Loeb, 1976). The pioneering experimental work on PRO was reported by Loeb and Norman in 1975

(Loeb and Norman, 1975) to demonstrate the feasibility of the PRO process for power generation.

However, the observed water flux and power density were much lower than

expected due to the lack of desirable membranes that can mitigate the severe concentration polarization that occurred in the thick and dense support layer of the RO membrane. The resurgence of interest in PRO came with the advancement of membrane fabrication technology in recent years (Achilli et al., 2009; Chou et al., 2012). The world's first osmotic power plant using river water and seawater with a capacity of 4 kW was inaugurated by Statkraft, an European company specializing in renewable energy, in November 2009 in Tofte, Norway

PT

(Helfe, et al., 2014). In 2012, Japan started the ‘‘Mega-ton water project’’ employing seawater

RI

reverse osmosis (SWRO) concentrate and treated wastewater for osmotic power recovery (Ku rihara and Hanakawa, 2013).

SC

The hybrid process of SWRO, wastewater treatment (WWT) and the PRO is of particular

U

interest, which represents an excellent example of process intensification. Using the SWRO brine

N

as the draw solute and treated wastewater as the feed, the PRO generated osmotic power can be

A

used for the SWRO process. The integration of SWRO-WWT-PRO contributes to reducing the

M

total energy consumption in SWRO plants and mitigating the environmental impact caused by

D

concentrated brine disposal (Kim, et al., 2012). It was reported that for an appropriately designed

TE

hybrid process, the potential capital cost savings may range from 8.7%–20% compared to conventional designs of seawater desalination plants (Sim et al., 2013). Strategic co-location of

EP

desalination and PRO systems can make use of the synergies available in the water-energy nexus

CC

for more sustainable desalination and energy production. 2.6 Membrane Distillation

A

Although membrane distillation (MD) may be used to remove selectively any volatile

component in solution, it is being primarily studied for desalination. In desalination by MD, pure water vapor is recovered from hot brine passing at ~35-950C on one side of a porous

hydrophobic membrane with gas-filled pores. Polymeric membranes are used in general. Conditions on the other side of the porous membrane define four types of MD (Drioli et al., 2005; Khayet, 2008; Sirkar, 1992): (a) Direct contact MD (DCMD) where cold distillate flowing on the other side condenses locally the water vapor coming through the pores; (b) air gap MD (AGMD) where the evaporated water vapor diffuses through an air gap on the other side and condenses on a cold surface at a distance; (c) vacuum MD (VMD) where water vapor coming

PT

through the pores to the other side of the membrane is withdrawn by a vacuum pump to a

RI

separate condenser running at a low pressure; (d) sweep gas MD (SGMD) where a noncondensable gas (e.g., air) is passed on the other side to remove the water vapor which is

SC

recovered in a separate and large condenser outside the MD device. Most developments have

U

focused on DCMD; much less has taken place on AGMD, VMD and SGMD in that order.

N

The advantages of DCMD compared to conventional thermal distillation processes are:

A

compact device, lower footprint, reduced scaling potential, lower capital cost and modular

M

structure. Devices are based on flat membranes including spiral-wound modules (Phattaranawik

D

et al., 2003) or hollow fibers (Li and Sirkar, 2004; Yang et al., 2011). Seawater has been

TE

concentrated in a DCMD pilot plant up to ~ 19% salt using rectangular cross-flow based polymeric hollow fiber membrane modules producing distilled water at a rate of 0.62 gpm (2.35

EP

LPM) with hot brine entering at 90-930C and distilled water coming in at 20-270C (Song et al.,

CC

2008). The average water flux at this level of production varied between 15 to 33 kg/ (m2-h) (8.8–19.4 gfd). A high value of heat transfer coefficient on the brine side is essential to reduce

A

temperature polarization and achieve the highest possible water vapor pressure based driving force leading to a high flux. The maximum membrane surface area used in the pilot plant was 6.6 m2 for ten modules with each module having a membrane surface area of 0.66 m2. An additional

multi-module configuration indicated a potential production capacity of 0.95 gpm (3.6 LPM). Each rectangular cross-flow module used in the pilot plant provided ~200-400 m2/m3 membrane surface area per unit volume depending on whether the ID or OD is used. Novel DCMD modules already developed at NJIT have around 1000 m2/m3. Therefore such modules are quite compact compared to conventional spiral-wound RO modules. DCMD-based plants will have much smaller footprint than conventional thermal distillation-based plants.

PT

Scaling of metallic heat transfer surfaces is a major problem in conventional thermal

RI

distillation-based desalination plants. Scaling has been essentially avoided in the polymeric hollow fiber membrane module designs employed in laboratory experiments (Li and Sirkar,

SC

2004) and pilot plant studies mentioned above. Laboratory studies using saturation indices (SI)

U

values of 1.1-1.9 for gypsum and 10-64 for CaCO3 indicated that the hollow fibers used and the

N

module design employed were essentially immune to scaling during experiments lasting for as

A

long as 360 min ( He et al., 2008, 2009). Precipitates were flowing all around without the water

M

flux being affected. This was achieved by the following features: a porous superhydrophobic

D

plasma polymerized fluorosilicone coating on porous hydrophobic polypropylene hollow fibers,

TE

rectangular cross-flow of the hot brine around the hollow fibers and potential oscillations of the hollow fibers. Pore wetting should be avoided. Polymeric materials being investigated include

EP

polytetrafluoroethylene (PTFE), polyvinylidene fluoride (PVDF) and polypropylene (PP).

CC

Another metric for thermal distillation methods is the gained output ratio (GOR) -- kg of water evaporated per kg of steam used for heating. Recycling of heat transferred to the distillate

A

is essential to achieving high GOR. In desalination by conventional distillation, GOR can go up to 10-15+. Most published DCMD studies have not paid attention to this aspect. Lee et al. (2011) employed a countercurrent cascade of small hollow fiber modules and experimentally achieved a

GOR value close to 6. Their modules did not have thermal insulation; the loss of heat to the ambient reduced the GOR achieved. Their model verified experimentally with other quantities such as water vapor flux predicted a GOR value of 10+. The proceedings of the 2011 Ravello Conference on DCMD provide an introduction to a number of studies of larger scale MD operations of various types (Drioli, 2011). One of the shortcomings of DCMD is that a supply of distilled water is needed at the

PT

beginning. The AGMD process generates distilled water directly from the brine by condensation

RI

on a cold surface. Singh and Sirkar (2012) provide a short introduction to AGMD and illustrate two-hollow-fiber-set based AGMD modules achieving a very high water vapor flux level in

SC

AGMD.

U

2.7 Membrane Distillation Bioreactor

N

The Membrane Distillation Bioreactor (MDBR) combines two PI concepts, the MBR

A

(section 2.1) and MD (section 2.6). As described earlier the conventional MBR uses a

M

microporous UF, or MF, membrane to replace the settler in the conventional activated sludge

D

process (CASP). One limitation of this is the inability of the UF membrane to retain slowly

TE

degradable organics below about 5 to 10kDa in MWt. In order to biodegrade these organics it would be beneficial to increase their organic residence time (ORT) to much greater than the

EP

hydraulic residence time (HRT). This becomes feasible if ‘tighter’ membranes are used in the

CC

MBR, for example using a nanofiltration (NF) membrane (Choi et al. 2002). However the NF membrane requires higher pressures or delivers very low fluxes under suction in the submerged

A

MBR. Using a MD membrane in the MDBR overcomes this (Fane et al. 2012). The MDBR employs a MD membrane to retain the mixed liquor and provide the treated wastewater. The MDBR membranes can be either submerged in the bioreactor or located in a side stream; the

submerged membrane MDBR is shown schematically in Figure 4. A temperature of about 50 to 600C is required to drive water across the membrane and consequently thermophilic biomass are required. In addition the MD membrane retains nonvolatile salts that accumulate in the reactor so the biomass also need to be salt tolerant. The MDBR is a ‘high retention’ MBR, wherein low molecular weight molecules do not leave the reactor through the membrane, as they would

TE

D

M

A

N

U

SC

RI

PT

through MF or UF membranes in the conventional MBR.

Figure 4. Membrane distillation bioreactor.

The MDBR can treat wastewater to a high quality, suitable for reuse. With adequate

EP

bubbling and occasional cleaning, fluxes of the order of 10 litre/m2-hr can be sustained. An

CC

attractive feature of the MDBR is that its major energy use can be provided by low grade waste heat. As a PI concept it can be compared with CASP+UF+RO, or MBR+RO, for water

A

reclamation. It can deliver water very close to the RO permeate quality. In a recent analysis (Goh et al. 2014) it showed a 30% lower electrical energy demand than a combined MBR and RO

process, and would be expected to have a lower cost structure due to less equipment and smaller footprint. The MDBR is a developmental concept and not yet widely used. However it has been demonstrated in the petrochemical industry (Khiang et al. 2010) providing a highly treated water, and driven by waste process heat. Other target industries include food and pharmaceutical wastewaters.

Recently the MDBR has been operated under anaerobic conditions with the

PT

generation of biogas.

RI

3. Some Process Examples 3.1 Lignocellulose to Biofuels

SC

Biofuels derived from lignocellulosic biomass are one of the leading renewable energy

U

candidates to replace fossil-based transportation fuels (Huber et al., 2006; Himmel et al., 2007;

N

Farrell et al., 2006). Commercialization of 2nd generation drop-in biofuels from lignocellulosic

A

biomass such as agricultural wastes, crop residues, forestry wastes, municipal wastes and aquatic

M

biomass will require the development of efficient separation and purification operations.

D

Membrane based separation processes are attractive as they could lead to significant process

TE

intensification and hence reduced operating costs (Drioli et al., 2012). As an example consider Figure 5 which is a schematic representation of a lignocellulose to

EP

biofuel biorefinery. Of particular interest are the pretreatment, conditioning and enzymatic

CC

hydrolysis steps which typically run in batch mode. Commonly pretreatment involves the use of dilute sulfuric acid typically 0.1-0.2 wt% to hydrolyze the hemicellulose present to its

A

monomeric sugars (mainly xylose) and enhances the enzymatic digestibility of the cellulose (Parajo et al., 1988; Grzenia et al., 2010).

After dilute acid pretreatment, the pH of the

hydrolysate is between 1 and 2. Further, compounds that inhibit subsequent bioconversion of the

solubilized sugars to the desired biofuel are also produced (Grzenia et al., 2008). Consequently hydrolysate conditioning is essential prior to fermentation. In this step the pH of the hydrolysate is adjusted and these inhibitory compounds are removed. Saccharification typically refers to the enzymatic hydrolysis of cellulose to glucose using a cocktail of cellulase enzymes (Huang et al.,

M

A

N

U

SC

RI

PT

2008).

D

Figure 5. Schematic representation of a lignocellulose to biofuel biorefinery.

TE

Development of a continuous pretreatment and enzymatic hydrolysis step is highly desirable. Figure 6 gives the variation of glucose concentration and the rate of glucose production with

EP

time for batch and continuous hydrolysis. The rate of glucose production is limited by product

CC

inhibition (Himmel et al., 2007; Ables et al., 2013; Cartensen et al., 2012). On the other hand, in a continuous hydrolysis process where glucose is continuously removed using membrane

A

filtration (see Figure 5), the rate of glucose production is up to an order of magnitude greater. Further enzyme usage efficiency is greater.

As can be seen from Figure 6, the glucose concentration in the product stream from a continuous process could be up to 5 times lower than in a batch process. In reality due to the decreasing rate of glucose production with increasing sugar concentration, the glucose concentration is only around 50 – 60 g L-1 (Cazetta et al., 2007; Brethauer and Wyman, 2010) for batch enzymatic hydrolysis. However in the case of bioethanol, concentrating the sugar stream to 200 g L-1 often maximizes the rate of ethanol production. Thus an integrated membrane process for sugar

PT

removal (microfiltration or ultrafiltration) and concentration (nanofiltration) prior to

RI

fermentation could lead to significant reduction in equipment size and processing time. Development of a catalytic membrane that catalyzes the hydrolysis of hemicellulose and

SC

cellulose to its monomeric sugars and allows passage of the product sugar through the membrane

U

could further enable combining pretreatment, hydrolysate conditioning and saccharification into

N

one step (Qian et al., 2013).

A

There are numerous other membrane separation processes such as pervaporation that

M

could be used for product recovery and purification. In fact inclusion of membrane processes

D

could enable the development of cost effective manufacturing processes for biofuels. However it

TE

is important to realize that replacement of one or more unit operations by a single membrane based unit operation will require integration of the membrane process into the overall

A

CC

EP

manufacturing process.

PT RI SC U N

3.2 Produced Water Treatment

M

A

Figure 6. Variation of glucose concentration and glucose production rate as a function of time for batch and continuous processing.

D

The direct contact membrane distillation (DCMD) process was employed to recover purified

TE

water from hot produced water that was de-oiled by a dissolved air flotation process (WEMCO).

EP

In a cross-flow membrane module having porous fluorosilicone coated polypropylene hollow fibers, Singh et al. (2013) have demonstrated that essentially pure water can be obtained in the

CC

distillate from post-WEMCO produced water. In a conventional produced water treatment

A

process there are 6-7 process steps needed to treat post-WEMCO produced water in order to obtain purified water from a reverse osmosis unit at the very end (Webb et al., 2009). This illustrates the extraordinarily high levels of PI achievable by membrane processes. 4. Concluding Remarks

Membrane processes and membrane-based contacting and reaction processes have demonstrated considerable process intensification. This has been achieved by one or more of the following: Equipment volume reductions sometimes by an order of magnitude for the same production level; achieving multistep processes in one device; nondispersive processing in phase contacting-based processes; enhancement of operational flexibility by avoiding flooding, loading, weeping, foaming; reduction in energy required for processing. Introduction of such

PT

processes in the process flow sheets for biorefinery, chemical plants, desalination and waste

RI

water treatment among others could introduce extraordinary levels of process intensification. References

A

CC

EP

TE

D

M

A

N

U

SC

Ables, C., F. Carstensen, M. Wessling, Membrane processes in biorefinery applications, J. Membrane Sci., 444, 285-317 (2013). Achilli, A., T.Y. Cath, A.E. Childress, Power generation with pressure retarded osmosis: An experimental and theoretical investigation, J.Membrane Sci., 343, 42-52 (2009). Albo, J., P. Luis, A. Irabien, Carbon dioxide capture from flue gases using a cross-flow membrane contactor and the ionic liquid 1-Ethyl-3-methylimidazolium Ethylsulfate, Ind. Eng. Chem. Res. 49, 11045-11051 (2010). Brethauer, S., C. E. Wyman, Review; continuous hydrolysis and fermentation for cellulosic ethanol production, Bioresource technology, 101, 4862-4874 (2010). Carstensen, F., A. Apel, M. Wessling, In situ product recovery: submerged membrane vs. external loop membranes, J. Membrane Sci., 394-395, 1-36 (2012). Cath, T.Y., A.E. Childress, M. Elimelech, Forward osmosis: principles, applications, and recent developments, J. Membrane Sci., 281, 70–87 (2006). Cath, T. Y. and A. E. Childress, Systems and Methods for Purification of Liquids, US Patent, 8083942 B2, 2011. Cazetta, M.L., M. A. P. C. Celligoi, J. B. Buzato, I. S. Scarmino, Fermentation of molasses by Zymomonas mobilis: effects of temperature and sugar concentrate on ethanol production, Bioresource Technology, 98, 2824-2828 (2007). Charpentier, J. C., Modern chemical engineering in the framework of globalization, sustainability and technical innovation, I&EC Res., 46, 3465-3485 (2007). Chau, J., G. Obuskovic, X. Jie, K. K. Sirkar, Pressure swing membrane absorption process for shifted syngas separation: modeling vs. experiments for pure ionic liquid. J. Membrane Sci., 453, 61-70 (2014). Chemical Engineering, A new approach to absorbing ammonia, October, 2001, pp. 95. Cheryan, M.; M. A. Mehaia, Membrane Bioreactors, in: W.C. McGregor, (Ed.), Membrane Separations in Biotechnology; Bioprocess Technology; Marcel Dekker, New York, Vol. 1, 1986, pp. 255-301.

A

CC

EP

TE

D

M

A

N

U

SC

RI

PT

Chou, S., R. Wang, L. Shi, Q, She, C.-Y. Tang and A.G. Fane, Thin-film composite hollow fiber membranes for pressure retarded osmosis (PRO) process with high power density, J. Membrane Sci., 389, 25-33 (2012). Choi J.H, S. Dockko, K. Fukushi, K. Yamamoto, A novel application of a submerged nanofiltration membrane bioreactor (NF MBR) for wastewater treatment. Desalination,146, 413420 (2002). Cornel, P., M. Krause, Membrane bioreactors for wastewater treatment, Chapter 9 in Li, N., A.G. Fane, W.S. Ho., T. Matsuura(Eds.), Advanced Membrane Technology and Applications, J. Wiley & Sons, Hoboken, New Jersey (2008), pp.217-238. Criscuoli, A., E. Drioli, Membrane Contactors for Gaseous Stream Treatments, in A.K. Pabby, S.S.H. Rizvi, A.M. Sastre (Eds.), Handbook of Membrane Separations: Chemical, Pharmaceutical, Food and Biotechnological Applications, CRC Press, Boca Raton, FL. 2009, pp.1041-1055. Drioli, E., Proceedings of the International Workshop on Membrane Distillation and Related Technologies, ITM—CNR, Ravello, Italy, October 11, 2011; www.itm.dnr.it/Ravello2011.html. Drioli, E., E. Curcio, G. di Profio, State of the art and recent progresses in membrane contactors, Chemical Engineering Research and Design, 83 (3), 223–233 (2005). Drioli, E., A. Brunetti, G. Di Profio, G. Barbieri, Process intensification strategies and membrane engineering, Green Chem., 14, 1561-1572 (2012). Esato, K., B. Eiseman, Experimental evaluation of Gore-Tex membrane oxygenator, J. Thorac. Cardiovascular Surgery, 69(5), 690-697(1975). Fane, A.G., Submerged membranes, Chapter 10 in Li, N., A. G. Fane, W. S. Ho., T. Matsuura ( Eds.), Advanced Membrane Technology and Applications, J. Wiley & Sons, Hoboken, New Jersey (2008), pp. 239-270. Fane A.G., J. Phattaranawik, F.S. Wong, Contaminated inflow treatment with Membrane Distillation Bioreactor. US Patent 8,318,017B2 (PCT June 2006) (2012). Farrell, A.E., R. J. Plevin, B. T. Turner, A. D. Jones, M. O'Hare, M. and D. M. Kammen, Ethanol can contribute to energy and environmental goals, Science, 311, 506-508 (2006). Frank, G.T., K. K. Sirkar, Alcohol production by yeast fermentation and membrane extraction, Biotech. Bioeng. Symp. Ser., 15, 621-631(1985). Frank, G.T., K. K. Sirkar, An integrated bioreactor-separator: in-situ recovery of fermentation products by a novel dispersion-free solvent extraction technique, Biotech. Bioeng. Symp. Ser., 17, 303-316 (1986). Franken, ACM, J.A.M. Nolten, M.H.V. Mulder, D. Bargeman, C.A. Smolders, Wetting criteria for the applicability of membrane distillation, J. Membrane Sci., 33, 315-328 (1987). Gabelman, A., S. Hwang, Hollow fiber membrane contactors, J. Membrane Sci., 159, 61-106 (1999). Goh S., J. Zhang, Y. Liu, A.G. Fane, Membrane distillation bioreactor (MDBR) – a lower GreenHouse-Gas (GHG) option for industrial wastewater reclamation, submitted to Chemosphere (Wastewater-energy nexus issue) (2014). Gore, D.W., Gore-Tex membrane distillation, In Proc. 10th Annual Convention of Water Supply Improvement Association, July 25-29, 1982, Honolulu, Hawaii. Grzenia, D.L., D. J. Schell, S. R. Wickramasinghe, Membrane extraction for removal of acetic acid from biomass hydrolysates, J. Membrane Sci., 322, 189-195 (2008). Grzenia, D.L., D. J. Schell, S. R. Wickramasinghe, Detoxification of biomass hydrolysates by reactive membrane extraction, J. Membrane Sci., 348, 6-12 (2010).

A

CC

EP

TE

D

M

A

N

U

SC

RI

PT

Helfer F., C. Lemckert, Y. G. Anissimov, Osmotic power with Pressure Retarded Osmosis: Theory, performance and trends – A review, J. Membrane Sci., 453, 337 -358 (2014). Holloway, R. W., A. E. Childress, K. E. Dennett and T. Y. Cath, Forward osmosis for concentration of anaerobic digester concentrate, Water Res., 41, 4005–4014 (2007). Himmel, M.E., S. Y. Ding, D. K. Johnson, W. S. Adney, M. R. Nimlos, J. W. Brady and T. D. Foust, Biomass recalcitrance: Engineering plants and enzymes for biofuels production, Science, 315, 804-807 (2007). Huang, H.-J., S. Ramaswamy, U. W. Tschirner, B. V. Ramero, A review of separation technologies in current and future biorefineries, Separation and Purification Technology, 63, 121 (2008). Huber, G.W., S. Iborra, A. Corma, Synthesis of transportation fuels from biomass: chemistry, catalysts and engineering, Chem. Rev., 106, 4044-4098 (2006). Jie, X., J. Chau, G. Obsukovic, K. K. Sirkar, Enhanced pressure swing membrane absorption process for CO2 removal from shifted syngas with dendrimer-ionic liquid mixtures as absorbent, I&EC Res., 53, 3306-3325 (2014). Judd, S., The MBR Book, Second Edition, Elsevier, Amsterdam (2011). Karoor, S., K.K. Sirkar, Gas absorption studies in microporous hollow fiber membrane modules, I&EC Res., 32, 674-684 (1993). Kessler, J.O., C.D. Moody, Drinking water from sea water by forward osmosis, Desalination,18, 297–306 (1976). Khaing T-H, J. Li, Y. Li, N. Wai, F.S. Wong, Feasibility study on petrochemical wastewater treatment and reuse using a novel submerged membrane distillation bioreactor. Sep. Purif. Technol.,74, 138-143 (2010). Khayet, M., Membrane Distillation, in N.N. Li, A.G. Fane, W.S.W. Ho, T. Matsuura (Eds.), “Advanced Membrane Technology and Applications”, John Wiley, Hoboken, NJ, Chap 6, 297 (2008). Kiani, A., R. R. Bhave, K. K. Sirkar, Solvent extraction with immobilized interfaces in a microporous hydrophobic membrane, J. Membrane Sci., 20, 125-145 (1984). Kim, J., J. Lee, J.H. Kim, Overview of pressure-retarded osmosis (PRO) process and hybrid application to sea water reverse osmosis process, Desalination and Water Treatment, 43, 193-200 (2012). Klaassen, R., A.E. Jansen, The membrane contactor: Environmental applications and possibilities. Environ. Prog. 20, 37-43 (2001). Kosaraju, P., A. S. Kovvali, A. Korikov, K. K. Sirkar, Hollow fiber membrane contactor based CO2 absorption-stripping using novel solvents and membranes, I&EC Res., 44, 1250-1258 (2005). Kraume, M., U. Bracklow, M. Vocks, A. Drews, Nutrients removal in MBRs for municipal wastewater treatment, Water Sci. & Technol.,51(6-7), 391-402 (2005). Kumar, P. S., J.A. Hogendoorn, P.H.M. Feron, G.F. Versteeg, New absorption liquids for the removal of CO2 from dilute gas streams using membrane contactors, Chem. Eng. Sci., 57, 16391651 (2002). Kurihara, M., M. Hanakawa, Mega-ton Water System: Japanese national research and development project on seawater desalination and wastewater reclamation, Desalination, 308, 131–137 (2013).

A

CC

EP

TE

D

M

A

N

U

SC

RI

PT

Lee, H., F. He, L. Song, J. Gilron, K.K. Sirkar, Desalination with a cascade of crossflow hollow fiber membrane distillation devices integrated with a hollow fiber heat exchanger, AIChE J., 57 (7), 1780–1795 (2011). Li, B., K.K. Sirkar, Novel membrane and device for direct contact membrane distillation-based desalination process, I&EC Res., 43, 5300–5309 (2004). Loeb, S., Production of energy from concentrated brines by pressure-retarded osmosis : I. Preliminary technical and economic correlations, J. Membrane Sci., 1, 49-63 (1976). Loeb,S., R.S. Norman, Osmotic Power Plants, Science, 189, 654-655 (1975). Logan, B. E., M. Elimelech, Membrane-based processes for sustainable power generation using water, Nature, 488 313-319 (2012). McCutcheon, J.R., R. Wang, Osmotic processes for a sustainable 21st century — Guest editorial, Desalination, 312 (2013) 1. McGinnis, R.L., Osmotic desalination process, US Patent 6,391,205 B1 (2002). Ortiz, I., J. A. Irabien, Membrane-Assisted Solvent Extraction for the Recovery of Metallic Pollutants: Process Modeling and Optimization, in A.K. Pabby, S.S.H. Rizvi, A.M. Sastre (Eds.), Handbook of Membrane Separations: Chemical, Pharmaceutical Food and Biotechnological Applications, CRC Press, Boca Raton, FL. 2009, pp.1023-1039. Parajo, J. C., H. Dominguez, Biotechnological production of xylitol. Part 3. Operation in culture media made form lignocellulose hydrolysates, Bioresource Technology, 66(1), 25-40 (1988). Petrotos, K.B., H.N. Lazarides, Osmotic concentration of liquid foods, J. of Food Engineering, 49, 201–206 (2001). Phattaranawik, J., R. Jiraratananon, A. G. Fane, Heat transport and membrane distillation coefficients in direct contact membrane distillation, J. Membrane Sci., 212 (1-2), 177-193 (2003). Phuntsho, S., H.K. Shon, S.K. Hong, S.Y. Lee, S. Vigneswaran, J. Kandasamy, Fertiliser drawn forward osmosis desalination: The concept, performance and limitations for fertigation. Reviews in Environmental Science and Biotechnology, 11, 147-168 (2012). Prasad, R., K. K. Sirkar, Solvent extraction with microporous hydrophilic and composite membranes, AIChE J., 33, 1057-1066 (1987). Prasad, R., K.K. Sirkar, Hollow fiber solvent extraction: Performances and design, J. Membrane Sci., 50, 153-175 (1990). Prasad, R., K.K. Sirkar, Membrane-based Solvent Extraction, in: W.S.W. Ho and K.K. Sirkar (Eds.), Membrane Handbook, Van Nostrand Reinhold, New York (1992); Reprinted, Kluwer Academic, Boston, 2001, pp. 727-763. Qi, Z., E.L. Cussler, Microporous hollow fibers for gas absorption I. Mass transfer in the liquid, J. Membrane Sci., 23, 321-332 (1985a). Qi, Z., E.L. Cussler, Microporous hollow fibers for gas absorption II. Mass transfer across the membrane, J. Membrane Sci., 23, 333-345 (1985b). Qian, X., J. Lei. S. R. Wickramasinghe, Novel polymeric solid acid catalysts for cellulose hydrolysis, RSC Advances, 3, 24280-24287 (2013). Qin, J., K. A. Kekre, G. H. Tao, H. Seah, New option of MBR-RO process for production of NEWater from domestic sewage, J. Membrane Sci., 272 (1-2), 70-77 (2006). Ramshaw, C., HIGEE distillation—an example of process intensification, The Chem. Eng., 389 (2), 13-14 (1983).

A

CC

EP

TE

D

M

A

N

U

SC

RI

PT

Reed, B. W., M. I. Semmens, E.L. Cussler, Membrane Contactors, in: R. D. Noble, S.A. Stern (Eds.), Membrane Separations: Principles and Applications, Elsevier: Amsterdam, The Netherlands, 1995; pp. 467-498. Schneider, K., T. S. Van Gassel, Membrandestillation, Chem. Ing. Tech., 56, 514-521 (1984). Schlosser, S., R. Kerte´sz, J. Marta´k, J., Recovery and separation of organic acids by membrane-based solvent extraction and pertraction: An overview with a case study on recovery of MPCA. Sep. Purif. Technol., 41, 237-266 (2005). Schofield, R.W., A.G. Fane, C.J. D. Fell, Heat and mass transfer in membrane distillation, J. Membrane Sci., 33, 299-313 (1987). Semmens, M., R. Qin, A. Zander, Using a microporous hollow fiber membrane to separate VOCs from water, Journal AWWA, April,162-167 (1989). Sengupta, A., P. A. Petersen, B.D. Miller, J. Schneider, C.W. Fulk, Jr., Large-Scale application of membrane contactors for gas transfer from or to ultrapure water. Sep. Purif. Technol., 14, 189200 (1998). Sengupta, A., R. A. Pittman, Application of Membrane Contactors as Mass Transfer Devices, in A.K. Pabby, S.S.H. Rizvi, A.M. Sastre (Eds.), Handbook of Membrane Separations: Chemical, Pharmaceutical Food and Biotechnological Applications, CRC Press, Boca Raton, FL. 2009, pp.7-24. Setiawan L., R. Wang, K. Li, A. G. Fane, Fabrication and characterization of forward osmosis hollow fiber membranes with antifouling NF-like selective layer, J. Membrane Sci., 394-395, 8088 (2012). Setiawan, L., R. Wang, S. Tan, L. Shi, A. G. Fane, Fabrication of poly(amide-imide) polyethersulfone dual layer hollow fiber membranes applied in forward osmosis by combined polyelectrolyte cross-linking and depositions, Desalination, 312, 99–106 (2013). Shi, L., S. Chou, R. Wang, W.X. Fang, C.Y. Tang, A.G. Fane, Effect of substrate structure on the performance of thin-film composite forward osmosis hollow fiber membranes, J. Membrane Sci., 382, 116-123 (2011). Sim, V. S.T., Q.H. She, T.H. Chong, C.Y. Y. Tang, A. G. Fane, W. B. Krantz, Strategic CoLocation in a Hybrid Process Involving Desalination and Pressure Retarded Osmosis (PRO) Membranes, 3, 98-125 (2013) Singh, D., P. Prakash, K. K. Sirkar, Deoiled produced water treatment using direct-contact membrane distillation, Ind. Eng. Chem. Res., 52, 13439−13448 (2013). Sirkar, K. K., Membranes, Phase interfaces and separations: Novel techniques and membranes -An overview, I&EC Res., Ind. Eng. Chem. Res., 47, 5250–5266 (2008). Sirkar, K. K., Other New Membrane Processes, in: W.S.W. Ho and K.K. Sirkar (Eds.), Membrane Handbook, Van Nostrand Reinhold, New York (1992); Reprinted, Kluwer Academic, Boston, 2001, pp. 885-912. Sirkar, K. K., P.V. Shanbhag, A.S. Kovvali, Membrane in a reactor: A functional perspective, I&EC Res., 38(10), 3715-3737 (1999). Song, L., Z. Ma, X. Liao, P.B. Kosaraju, J.R. Irish, K.K. Sirkar, Pilot plant studies of novel membranes and devices for direct contact membrane distillation-based desalination, J. Membrane Sci., 323, 257–270 (2008). Spiegler, K. S., O. Kedem, Transport coefficients and salt rejection in uncharged hyperfiltration membranes, Desalination, 1, 311-326 (1966). Stankiewicz, A.I., J. A. Moulijn, Process intensification: Transforming chemical engineering, CEP, January, 22-34 (2000).

A

CC

EP

TE

D

M

A

N

U

SC

RI

PT

Tao, G.H, Public Utilities Board Singapore, personal communication (2014). Tsuji, T., K. Suma, K. Tanishita, H. Fukazawa, M. Kanno, H. Hasegawa, A. Takahashi, Development and clinical evaluation of hollow fiber membrane oxygenator, Trans. Am. Soc. Artif. Intern. Organs, 27, 280-284 (1981). Wang, R., D.F. Li, C. Zhou, M. Liu, D.T. Liang. Impact of DEA solutions with and without CO2 loading on porous polypropylene membranes intended for use as contactors, J. Membrane Sci., 229, 147-157(2004). Wang, R., L. Shi, C.Y. Tang, S. Chou, C. Qiu, A.G. Fane, Characterization of novel forward osmosis hollow fiber membranes, J. Membrane Sci., 355,158–167(2010). Wang, R., H.Y. Zhang, P.H.M. Feron, D.T. Liang, Influence of membrane wetting on CO2 capture in microporous hollow fiber membrane contactors. Separation and Purification Technology 46, 33-40 (2005). Webb, C.; L. N. S. P. Nagghappan, G. Smart, J. Hoblitzell, R. Franks, Desalination of oilfieldproduced water at the San Ardo water reclamation facility, CA, SPE Western Regional Meeting, San Jose, CA, Mar 24−26, 2009; Society of Petroleum Engineers: Richardson, TX, 2009; SPE 121520-PP. Wickramasinghe, S.R., M.J. Semmens, E.L. Cussler, Mass transfer in various hollow fiber geometries, J. of Membrane Sci., 69, 235-250 (1992). Wu, Y., E. Drioli, The behavior of membrane distillation of concentrated aqueous solution, Water Treatment, 4, 399-415 (1989). Xie, J., J. Chau, G. Obsukovic, K.K. Sirkar, Enhanced pressure swing membrane absorption process for CO2 removal from shifted syngas with dendrimer−ionic liquid mixtures as absorbent, I&EC Res., 53, 3306-3325 (2014). Yamamoto, K., M. Hiasa, T. Mahood, T. Matsuo, Direct solid-liquid separation using hollow fiber membrane in an activated sludge aeration tank, Water Sci. & Technol.,21, 43-54 (1989). Yang, X. R Wang, L Shi, AG Fane, Performance improvement of PVDF hollow fiber-based membrane distillation process, J. Membrane Sci., 369 (1–2), 437–447 (2011). Zhang, Y., R. Wang, Gas-liquid membrane contactor for acid gas removal: Recent advances and future challenges, COCHE, 2(2), 255-262 (2013). Zhang, Y., R. Wang, Novel method for incorporating hydrophobic silica nanoparticles on polyetherimide hollow fiber membranes for CO2 absorption in gas-liquid membrane contactor, J. Membrane Sci., 452 (2014) 379-389. Zhao, S.F., L. Zou, C.Y. Tang, et al., Recent developments in forward osmosis: Opportunities and challenges, J. Membrane Sci., 396 1-21 (2012).

Table 1. Comparison of CASP and MBR Parameter

CASP

MBR

Loading rate (kgBOD/m3d)(a)

0.25

0.3-2.5

BOD removals (%)(a)

85-95

98-99

~60

> 99

99

>99

89

97-99

Footprint (relative)

100

30 to 50

Energy (kWh/m3)

<0.5

0.4-0.7

Solid waste (relative)

100

< 100

(a)

TSS removals (%)

Amm removals (%)

(a)

(a)

P removals (%)

(a) From Table 3 (Kraume et al., 2005)

PT

difficult to make a clear comparison with CASP due to the effect of footprint on land costs and

RI

whether the treated water is assigned a value. In applications where the interest is in reuse, the

SC

higher quality effluent from the MBR will have cost and technical advantages. For example in water reclamation to indirect potable standards the options are CASP+UF+RO or MBR+RO. It

U

has been established that the MBR provides a better feed to the RO than the CASP+UF option

N

(Qin, 2006). In many cases the compact nature of the MBR is a key criterion for selection, for

A

example in basements of large buildings, on board ship etc. More details on MBRs can be found

A

CC

EP

TE

D

M

in Cornel and Krause (2008) and Judd (2011).