Accepted Manuscript

Chemistry and fabrication of polymeric nanofiltration membranes: A review

Mou Paul, Steven D. Jons

PII: S0032-3861(16)30651-6
DOI: 10.1016/j.polymer.2016.07.085
Reference: JPOL 18891

To appear in: Polymer

Received Date: 20 December 2015

Revised Date: 24 June 2016
Accepted Date: 29 July 2016

Please cite this article as: Paul M, Jons SD, Chemistry and fabrication of polymeric nanofiltration
membranes: A review, Polymer (2016), doi: 10.1016/j.polymer.2016.07.085.

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.
 ACCEPTED MANUSCRIPT

Chemistry and fabrication of polymeric nanofiltration membranes: A

review

Abstract

PT
The review article describes research on different methods for manufacturing polymeric NF membranes.
The primary focus is on NF composite membranes formed by interfacial polymerization. Within that

RI
space, topics covered in detail include the composite structure, monomeric and polymeric reactants
employed to endow specific characteristics, and additives used to influence the reaction. Other
approaches covered that have been commercially important are manufacture of NF membranes by

SC
phase inversion and different post-treatments (e.g. coatings, grafting) that may be applied to a more
porous support. Emerging technologies also discussed include layer-by-layer coatings, incorporation of
aquaporins, and use of glassy polymers with high internal porosity. Common themes include research

U
directed to improved performance (flux, rejection, selectivity, reduced fouling) and modifications that
allow for success in more challenging feeds (pH extremes, high temperature, chlorine, solvents).
AN
Emphasis on Interfacially Polymerized Membranes
Relative Number of Citations

M
Post-treatments
Additives
Reactive monomers & polymers
D

Composite structure
TE
C EP
AC
 ACCEPTED MANUSCRIPT

Contents
Chemistry and fabrication of polymeric nanofiltration membranes: A review ........................................... 1
Abstract ......................................................................................................................................................... 1
1.0 Introduction ........................................................................................................................................... 4
2.0 Methods of NF membrane formation.................................................................................................... 5

PT
2.1 Phase inversion .................................................................................................................................. 6
2.1.1 Membrane formation ................................................................................................................. 6

RI
2.1.2 Improved Stability ....................................................................................................................... 8
2.1.3 Improved performance ............................................................................................................. 11

SC
2.2 Post-treatment of a porous support ................................................................................................ 12
2.2.1 Polymer coatings ....................................................................................................................... 13
2.2.2 Surface grafting ......................................................................................................................... 15

U
2.2.3 Membrane charge ..................................................................................................................... 16
AN
2.3 Layer-by-layer (LBL) polyelectrolyte membranes ............................................................................ 16
2.4 Aquaporin-based membranes ......................................................................................................... 18
M
2.5 Glassy polymers with high free volume ........................................................................................... 20
3.0 Interfacial polymerization .................................................................................................................... 20
3.1 Historical perspective on the interfacial polycondensation ............................................................ 21
D

3.2 Composite membrane ..................................................................................................................... 22

TE

3.2.1 Web for strength ....................................................................................................................... 23

3.2.2 Porous support layer ................................................................................................................. 23
EP

3.2.3 Intermediate layer .................................................................................................................... 27

3.2.4 Surface coatings ........................................................................................................................ 28
3.3 Monomers ........................................................................................................................................ 29
C

3.3.1 Monomers with charged groups ............................................................................................... 29

AC

3.3.2 Monomers providing hydrophilic character ............................................................................. 31

3.3.3 Monomer structure................................................................................................................... 32
3.3.4 Monomers for chlorine resistance ............................................................................................ 35
3.3.5 Monomers for extending pH range........................................................................................... 38
3.3.6 Monomers for solvent stability ................................................................................................. 40
3.4 Polymeric reagents........................................................................................................................... 40
3.5 Additives........................................................................................................................................... 45
 ACCEPTED MANUSCRIPT

3.5.1 Surfactants & phase transfer catalysts ..................................................................................... 45

3.5.2 Co-solvents/solvents ................................................................................................................. 48
3.5.3 Organic molecules ..................................................................................................................... 49
3.5.4 Inorganic Salts ........................................................................................................................... 50
3.5.5 Nanoparticles ............................................................................................................................ 50

PT
3.6 Post Treatments ............................................................................................................................... 52
4.0 Conclusions and Recommendations .................................................................................................... 54

RI
5.0 References............................................................................................................................................ 59

U SC
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT

1.0 Introduction
Microfiltration (MF), ultrafiltration (UF), nanofiltration (NF), and reverse osmosis (RO) are all pressure-
driven separation processes that fall along a continuum of performance needs. MF is used for removing
suspended particles greater than about 0.1 micron. UF commonly excludes dissolved molecules greater
than 5,000 molecular weight. RO membranes have high rejection of almost all salts and neutral species
greater than 100 Dalton MW. And, NF is somewhere between UF and RO.

PT
The term nanofiltration (NF) was originally coined at Filmtec Corporation to characterize loose RO
membranes with pores diameters greater than about 1 nm in size. A more critical definition of an NF

RI
membrane may be formulated based on a number of approximate characteristics.

Pore diameters less than 2 nm.

SC
Passage of an appreciable amount of monovalent ions.
Substantially higher rejection of divalent ions than monovalents.
Molecular weight cutoff (MWCO) for neutral species is in the 150-2000 Daltons range.
Rejection of neutrals and positive ions relates principally to their size and shape.

U
This description is similar to the list of NF features provided by Bjarne Nicholaisen, as reiterated in the
AN
2005 book, Nanofiltration: Principles and Applications.[1] It is somewhat more inclusive, as that
description specified more than 30% passage of monovalents, virtually total rejection of divalents, and a
molecular weight cutoff range of 150-300 Daltons. The NF literature is inconsistent on these points, and
acceptance of a common set of defining characteristics for nanofiltration remains elusive. In any case, it
M
is clear that one cant define a membrane in terms of its performance without also establishing the
conditions of operation. (For instance, just decreasing the pressure by a factor of two might commonly
double the passage of divalent ions.) Hence, using some definitions, there is a reasonable argument
D

that whether a membrane is NF or RO could depend on the process in which it is being used. Because it
is even more difficult to determine if a membrane is NF by a description of its fabrication, the authors
TE

have been fairly inclusive in this review. Nonetheless, it important to recognize particular ways in which
NF is perceived as advantaged.

As compared to RO, the success of NF often results from its selective separation of one solute over
EP

another. The Mry-sur-Oise plant has been producing drinking water from the Oise river since 1999,
with objectives to remove atrazine and pass a substantial fraction of calcium and bicarbonate ions.[2] In
the dairy industry, NF is used to demineralize whey products by passing undesirable monovalent salts
C

(Na, K, Cl) and simultaneously concentrating constituents such as lactose, protein, and calcium.[3] The
waste streams from textile dying can comprise both dyes and salt, both of which would be removed by
AC

RO, but NF can be used to keep the valuable salts while concentrating dyes into a more manageable
volume.[4] NF is also used in separation of different neutral species, e.g. the simple monosaccharides
xylose (150 MW) and glucose (180 MW).[5]

NF membranes are also frequently advantaged by their ability to avoid an osmotic pressure limitation.
To produce water for downhole injection on offshore oil production platforms, NF has been used to
treat seawater and prevent barium sulfate scale. High recoveries are possible at reasonable pressures
because the membrane passes most of the sodium and chloride ions while retaining sulfate.[6] NF is
also used to remove sulfate in the chloralkali process, where concentrations of NaCl in the feed may
exceed 200 g/L and would correspond to unreasonably high osmotic pressure for RO.[7] As zero liquid
 ACCEPTED MANUSCRIPT

discharge (ZLD) is becoming more often required, NF will play a larger role in treating already
concentrated RO reject streams, reducing the volume of water needing more costly final stage
operations (e.g. evaporation).

In still other cases, nanofiltration membranes are more akin to a "loose RO" and their principal
advantage is a relatively high permeability while retaining sufficient rejection for the specific application.
FilmTec NF90 was originally introduced to pass about 10% of monovalents, while retaining high

PT
removal of other contaminants (e.g. sulfite, iron, pesticides, herbicides, TOC). However, without
concern for NaCl passage, many customers principally appreciate its relatively high water permeability
(~13L/m2hr/bar) and sufficient rejection for solutes in their applications. Based on both its performance
and chemistry, one of the authors (SDJ) has often argued that it is better to classify this membrane as

RI
loose RO, but NF90 maintains a firm position in the NF literature.

There are many review articles relevant to NF membranes. Cadotte and Petersen have described the

SC
basics of interfacial polymerization and variations in chemistry used to form historical nanofiltration
membranes.[8, 9] More recently, a comprehensive text on nanofiltration was published in 2005 by
Schfer, Fane, and Waite that covers membrane fabrication in two of the chapters.[1] Hilal et al. have
published two articles that broadly discuss the topic of nanofiltration membranes, including various

U
methods of fabrication and modification.[10, 11] The topic of nanofiltration with organic solvents is
covered in depth by Marchetti et al.[12] Vandezande et al. have written a detailed overview of the
AN
types of solvent resistant NF membranes[13] and Hermans et al. provided a more focused discussion of
the fabrication of thin film composite membranes for solvent resistant nanofiltration[14]. Van der
Bruggen et al. discuss several key challenges facing nanofiltration applications that continue to motivate
research.[15] There have been several other reviews covering thin film composite membranes in
M
general, especially for reverse osmosis.[16-19] Recent reviews describe efforts to incorporate
nanomaterials into membranes, and this may sometimes be applied to NF membrane fabrciation.[20,
21] Other articles focus on layer-by-layer membranes[22] and polyelectrolyte membranes for
D

nanofiltration[23, 24].
TE

With a bias in selection towards literature from the last decade, this paper focuses on reviewing several
approaches used to fabricate polymeric NF membranes. Special attention is given to the class of NF
membranes made by interfacial polymerization. Within that space, an effort is particularly made to
describe constituents of the reaction. These include monomers, polymeric reagents, and various
EP

additives. It is often the case that membrane research is performed to address two basic performance
properties: flux and salt passage. However, an attempt is also made to call out approaches directed to
other issues (e.g. fouling, selectivity, stability to different environments, chlorine resistance, etc.)
C

2.0 Methods of NF membrane formation

AC

Polymeric nanofiltration membranes may be fabricated by several approaches. Common approaches

include interfacial polymerization, phase inversion, and post-treatment of a more porous support
(surface coatings, grafting, etc). Some of the emerging technologies include layer-by-layer coatings,
incorporation of aquaporins, and use of glassy polymers with high internal porosity. Due to space
constraints, this review still omits many options (esp. carbon nanotubes, graphene, graphene oxide) that
may become very significant, but appear further from commercialization. (The interested reader may
initially refer to several computational studies [25-28] and recently published reviews of progress with
these materials.[29, 30]) Also, by focusing on polymeric membranes, this selective summary also avoids
 ACCEPTED MANUSCRIPT

discussion of commercially available ceramic membranes which can have advantage in stability to very
extreme operating conditions. However, boundaries had to be drawn, and this more distinct art space is
worthy of its own review.

An interfacial polymerization is the most common approach to making the barrier layer for commercial
nanofiltration membranes, including those sold by Dow. To provide context for discussion of other
methods, a short description is provided here. However, variants on the process will be covered in

PT
depth in a later section.

In the interfacial polymerization process, a polycondensation reaction between two monomers is carried
out at the boundary of immiscible solutions (an aqueous phase and organic phase). In practice, the

RI
aqueous phase monomer solution, which often comprises piperazine (PIP) or m-phenylenediamine
(MPD), is first imbibed in a porous support and then excess solution is removed from the surface.
Following this, the organic phase monomer solution, which usually comprises trimesoyl chloride (TMC),

SC
is made to contact the porous support. A barrier layer is rapidly formed at the interface. While use of
PIP monomer in newer literature almost always indicates a nanofiltration membrane, MPD is more
commonly (though not always) used in reactions forming tighter reverse osmosis membranes.

U
The interfacial polymerization process has two important advantages over other methods. 1) An
extremely thin barrier layer is formed at the interface of the two immiscible solutions. Dows NF 270
AN
membrane has a polyamide barrier layer of about 20 nm thickness. The NF90 membrane, made with
MPD chemistry, has a barrier layer of about 200 nm. 2) The process is reliable because it is self-limiting.
The reaction continues at a high rate wherever reactive monomers can contact. However, upon forming
the barrier film at the oil/water interface, the transport of monomers is hindered and the reaction
M
slows. As a consequence of this process, a thin, uniform, and integral barrier layer can be made that has
both high permeability and good retention of solutes. At the same time, the thin layer can provide
challenges for operation in some environments.
D

2.1 Phase inversion

TE

A phase inversion process has been used for forming both commercial and experimental NF
membranes. (In fact, the phase-inversion process is particularly significant for NF, since the barrier layer
EP

of almost all interfacially polymerized composite membranes is itself formed on a phase inverted
support layer.) However, this section focuses on use of phase inversion to form the NF discriminating
layer. An advantage of this approach is that it is less complicated and requires fewer steps than
interfacial polymerization. In addition to flat sheet geometry, it allows facile formation of hollow fiber
C

membranes. And, as will be evident, a wide range of polymers can be employed.

AC

2.1.1 Membrane formation

In the phase inversion process, a thermodynamically stable polymer solution is caused to separate into a
polymer-rich and polymer-lean phase, eventually resulting in a bicontinuous porous solid. This liquid-
liquid de-mixing process might be brought about in several ways.

1) Immersion precipitation takes place when a polymer solution (comprising at least a polymer and
a solvent) is immersed in a coagulation bath containing a non-solvent (for the polymer), and
 ACCEPTED MANUSCRIPT

wherein the solvent and nonsolvent are miscible. The solvent surrounding the polymer moves
into the non-solvent and causes phase separation.
2) Evaporation-induced phase separation is caused when increased concentration of the polymer is
brought about by evaporation of a solvent for the polymer. Precipitation or de-
mixing/precipitation results.
3) Vapor-induced phase separation results when the polymer solution is exposed to a nonsolvent
in the atmosphere (typically water). Absorption of this non-solvent causes de-

PT
mixing/precipitation.
4) A thermally induced phase separation is made to occur by changing (generally decreasing) the
temperature, so that the polymer becomes less soluble in the solvent. The solvent is

RI
subsequently removed from the resulting polymer matrix by extraction, evaporation, or freeze
drying.

NF phase separated membranes are usually formed with an immersion precipitation. However, a

SC
combination of the above processes may be used. Commonly, the polymer solution may include a
volatile solvent or co-solvent (e.g. THF), so that an initial evaporation creates a higher concentration of
polymer at the surface. The polymer may be immersed thereafter in the nonsolvent (usually water), and
the pre-concentration of polymer at the surface causes a thicker barrier layer to be formed that is more

U
likely to have an integrally skinned top surface (with pores of sufficiently small size to operate in the
NF regime). If pre-concentration by evaporation is not used, membranes with this tight surface can also
AN
be made using a coagulation bath containing a weak nonsolvent. Polymers used in phase inversion may
include charged groups, and it is typical in NF that charge (Donnan) interactions dominate rejection of
ionic species, so that a much larger MWCO is observed for sterically rejected neutral species compared
M
to ions. Also, as will be described later, various post-treatments can subsequently be used to convert
even membranes formed with much larger pores on the top surface into NF membranes.

An advantage of immersion precipitation is that the membrane structure can be very asymmetric. A
D

dense skin layer provides selectivity, and a much more open bulk structure provides mechanical support
without greatly increasing flow resistance. The thickness of the dense skin for NF phase inverted
TE

membranes is not always measurable or even well defined and thickness obviously varies between
membranes. Nonetheless, the dense discriminating layer at the top surface is typically thicker than
would usually be obtained by interfacial polymerization for NF membranes (10-200 nm). At the same
time, the discriminating layer of phase inverted membranes is usually thinner than most free-standing
EP

coatings (without defects) can easily be applied. Below this discriminating layer, pore size increases
rapidly as one moves into the bulk structures. Because of this large asymmetry, the majority of
resistance to flow for immersion precipitation membranes is sometimes ascribed to the region near the
C

thin discriminating layer. However, closer analysis has often shown otherwise [31], and opportunities
may exist to further optimize and reduce contributions to flow resistance from within the bulk or lower
AC

surface.

An optimized phase inversion will typically extend beyond the simple ternary system of polymer,
solvent, and non-solvent. In commercial formulations, any of these components may actually be a
combination of constituents that produce the desired membrane properties. A plethora of additives
(e.g. salts, alcohols, surfactants, particulates, low molecular weight organic molecules, high and low MW
polymers) to the dope or non-solvent can further pronouncedly affect the resulting membrane. Figure 1
illustrates the asymmetric top surfaces of two membranes made by immersion precipitation and the
resulting morphology changes when 3% of sulfonated poly (arylene ether sulfone) SPAES just replaces
some of the PES in a 23% casting solution (18% PES / 5% PEG / NMP). In this case, Hwang et. al.[32]
ascribed differences in morphology (increased size and spacing of finger-like structure and thicker skin
 ACCEPTED MANUSCRIPT

layer) principally to viscosity increases in the dope and reduced diffusional exchange rate between
solvent and non-solvent below the surface. The sulfonated polymer also increased NaCl rejection from
about 0% to 14% due to both changes in membrane charge and surface pore size.

Membrane formation depends on more than the chemicals used. Humidity in the air, temperatures of
the dope and quench, the time before quench, and the time duration of quench are all also key
variables. The thermodynamics and kinetics that define the structures formed by a phase inversion

PT
process have been the subject of many book chapters and papers, and it continues to be the subject of
current research.[33-36] Because this process is much better addressed elsewhere,[35-39] and because
it is not required for comprehension of subsequent discussions, it will remain an area beyond the scope
of this review. Instead, the limited discussion of phase inversion membranes will be focused on some of

RI
the key approaches being researched to address market challenges.

U SC
AN
M
D

Figure 1. Top surfaces of asymmetric PES membranes formed by phase inversion, with 0% and 3%
substitution of SPAES into dope.[32]
TE

2.1.2 Improved Stability

EP

An important thrust of the open literature during recent years has been aimed at further improving the
ability of NF membranes to operate in harsh environments. Common NF interfacially polymerized
C

membranes are generally limited to aqueous applications with less than 0.1 ppm chlorine in the feed,
leaving membranes susceptible to both biofouling and accidental chlorination. By contrast, several
AC

membranes formed by phase inversion often easily surpass this limit though typically with reduced
flux and/or rejection. For instance, one commercial cellulose acetate NF membrane is specified as
tolerant to 1 ppm continuous chlorine, but it has a more restrictive pH operating range (pH 2-7.5) and
water permeability (~0.12 gfd/psi) that is less than half that of most NF polyamide membranes. A phase
inverted poly(ether sulfone) hollow fiber module specifies up to 200 ppm continuous chlorine tolerance,
but this commercial membrane has higher passage (1000 Dalton MWCO) than other NF membranes.
Most interfacially polymerized NF modules are also limited by other operating ranges: feed pH > 1-3,
feed pH < 9-11, and a maximum operating temperature commonly between 40o and 50oC. There are
several spaces where acid or base conditions can limit the lifetime of these NF elements. In some food
and beverage applications, daily high pH (and also low pH) membrane cleanings are used to remove
 ACCEPTED MANUSCRIPT

organics and prevent bacteria growth. Metal plating operations can use membranes to separate metals
(Cu, Fe, Al) from strong acids (sulfphuric, phosphoric, hydrochloric) in the pickling baths. In the mining
industry, NF membranes can recover sulfuric acid used in leaching metal ores. Cleaning operations in
several industries can include recovery and re-use of sodium hydroxide in cleaning solutions. Other
polymeric barrier layers based on phase inversion or coatings are easily stable at greater pH and
temperature extremes than common interfacially polymerized membranes, but these frequently lack
sufficient performance (water permeability or rejection). For instance, a common NF application in the

PT
dairy industry is concentration of lactose (MW 324 Daltons), and interfacially polymerized piperazine-
based membranes are used despite a shorter lifetime because of their separation characteristics. One
need for new membrane development is to achieve improved stability, while maintaining equivalent

RI
performance (and costs) to typical interfacially polymerized NF membranes. Alternatively, new
membrane types provide opportunities to attain stable operations at conditions well beyond those
ranges compatible with the dominant interfacially polymerized membranes.

SC
Specifically with respect to phase inversion membranes, research in the literature appears to be focused
on domains not addressable with common polyamide NF membranes. This sometimes involves testing
membranes in very strong acid and caustic solutions in combination with high temperature.[40, 41]
With respect to organic solvents, current limitations and potential applications have been discussed in

U
several papers.[12-14, 42, 43] Marchetti et. al. [12] provides a recent detailed review of applications,
noting that these membranes are currently most commonly used in the areas of pharmaceutical
AN
polishing and catalyst recovery. For applications involving pH, high temperature, and solvents, it should
be emphasized that the membrane is one of many components in a separation module that need to be
compatible with the operating conditions.
M
Attempting to create more stable NF membranes by phase inversion, different groups have explored
several polymer systems with various degrees of success (depending on the solvent system, conditions,
and test criteria). Examples of polymers used in this research include polyarylene sulfide sulfone
D

(PASS)[44], polyphenylsulfone PPSU[45] [46], polyaniline (PANI)[47-49], polysulfone (PS) with low
molecular weight additives[50], variations of polyetheretherketone (PEEK)[40, 51-56], variations of
TE

polyimide (PI)[57-65], variations of polybenzimidazole (PBI)[66-68], polypyrrole post reacted on

polysulfone (PS)[69], and blended polyimide (PI) with polyphenylsulfone (SPUS)[70]. A few of the
approaches to improve stability are noted below.
EP

PEEK is a polymer with high Tg and excellent chemical resistance but its insolubility in common solvents
makes it particularly difficult to form into an NF membrane. By optimizing fabrication conditions,
including an extensive solvent exchange and drying process after membrane formation, da Silva Burgal
et al. was able to make NF membranes from PEEK polymer in a mix of methanesulfonic acid and
C

sulphuric acid.[54] They found that PEEK membranes and mildly post-sulfonated PEEK membranes were
both much more stable in solvents (THF, EtOH, Acetone, and esp. DMF) than PEEK membranes that had
AC

been post-sulfonated to a substantially greater degree.[55] Based on low mass loss over months, the
PEEK appeared to have good stability to strong acids and bases. In a separate paper,[41] these
membranes were run at a high temperature (80oC) and high concentration of base (0.9 M triethylamine)
in DMF, successfully demonstrating use of the membrane in a continuous catalytic process with
improved productivity and purity. Taking a different direction, Hendrix et al. created two new polymers,
TBPEEK [53] and BPAPEEK [40]. In each case, groups were attached to the PEEK backbone that made the
polymers sufficiently soluble for casting. Modifications of PEEK to enable making phase inversion
membranes have been done before,[52] but the two new polymers were expected to maintain a more
similar character to the high performing PEEK. Both polymers were cast from dope mixtures containing
NMP and THF, allowing loose NF membranes to be formed by adjusting the volatile THF. TBPEEK
 ACCEPTED MANUSCRIPT

membranes were run with hexane, MeOH, acetonitrile, and IPA, but soaking tests showed they werent
stable in toluene or acetone. Membranes from both TBPEEK and BPAPEEK polymers demonstrated
mechanical stability under extreme pH conditions (30% KOH, 150oC), but a dramatic decrease in dye
retention proved loss of integrity in the discriminating layer. Still, authors suggested that the polymers
might be sufficient for ultrafiltration or as a support layer for interfacially polymerized membranes. In a
different paper, Hendrix et al. also examined crosslinking PEEK-based polymers.[56] They used
monomers diphenolic acid and difluorobenzophenone to create a new polymer (VAPEEK) with both

PT
sufficient solubility for casting and having functional groups containing carboxylic acid. The VAPEEK
polymer could then be cast from a solution of THF and an activating agent. The activated carboxylic
acids reacted with multifunctional amines in the coagulation bath to create amide crosslinks. The

RI
membranes had good rejection of dye in IPA and acetone, but were not operated with other solvents.

Crosslinking the phase inverted membranes to form covalent bonds is a continuing area for research
that pushes stability further for NF.[12] Taking polyimides as an example, these membranes are

SC
conveniently cast in polar aprotic solvents, but then the presence of those species (NMP, DMF, DMAc
DMSO) in feed waters poses a problem for stable operation. Various methods of crosslinking after
casting can allow the advantageous morphology of a phase inversion to be retained (e.g. against
swelling or dissolution) under more challenging conditions.[71] Crosslinking of polyimides has been

U
performed by UV initiation of a monomer after phase inversion[72], by reaction of diols with
incorporated carboxylic acid groups,[73] or by reaction with diamines or polymeric amines to form
AN
intermolecular amides.[58, 74] In particular, See-Toh et al. first found that great increases in membrane
stability resulted from post-crosslinking polyimides with aliphatic diamines (e.g. 1,2-ethylenediamine,
1,6-hexanediamine).[58] As an alternative to post-treatment by soaking membranes, crosslinkers may
M
be included within a dilute casting solution[59] or applied in the aqueous coagulation bath.[59, 60]
Several studies have further explored the stability of crosslinked polyimides with different solvents and
evaluated recipes and casting conditions to improve their performance.[61-65] Soroko et al. has
examined the roles of different polyimides, polyimide molecular weight and structure (block vs.
D

random), concentration of the 1,4-dioxane co-solvent, and evaporation time before casting.[63-65]
Membranes were evaluated in tests using a homologous series of styrene oligomers that were dissolved
TE

in either toluene (for non-crosslinked membranes) or DMF (for crosslinked membranes).

Valtcheva et al. cast membranes of polybenzimidazole (PBI) in DMAc and crosslinked these with either
,-dibromo-p-xylene (DBX) or 1,4-dibromobutane (DBB).[67] Both did well under acid and base
EP

conditions, but the aromatic crosslinker (DBX) performed better than the aliphatic (DBB) in DMF.
Comparisons of weight loss were made to a commercial polyimide membrane to show greater stability
in several solvents and at pH extremes. In their most recent study, conditions of DBX crosslinking (e.g.
C

time, temperature, concentration, solvent) were examined in a DOE, and one conclusion was that
carrying out the reaction beyond 24 hours resulted in negligible change in MWCO and a detrimental
AC

change in permeability.[68]

Loh et al. made integrally skinned polyaniline (PANI) membranes by phase inversion on a polypropylene
support, and these membranes were crosslinked with either gluteraldehyde or ,-dichloro-p-xylene.
Membranes were found to be stable in MeOH, THF, DMF, acetone, and ethyl acetate. [47] Sairam et al.
made similar PANI membranes by thermal crosslinking.[48] A polyester non-woven was used instead
because the crosslinking conditions (180oC, 1hr) were problematic for the polypropylene. However, the
instability of the polyester backing is also a problem for polar aprotic solvents (e.g. DMF), so this would
limit use. Spiral wound elements were made of each type of membrane.[49] Tests of the
gluteraldehyde crosslinked PANI element were run for two days at 65oC with oligostyrenes in DMF to
verify stability.
 ACCEPTED MANUSCRIPT

2.1.3 Improved performance

Commercial membranes are addressing well many needs in water purification, but there is always a
desire for higher flux, greater selectivity, and less fouling. For instance, the textile industry has a great
need for membranes that better separate dyes from salts with minimal energy, and fouling is often
found to be a problem.[75, 76] Modifying the membrane charge has been a common approach. In

PT
some fortunate cases, inclusion of charged species in the dope of a phase inversion membrane is found
to have favorable effects on each of flux, rejection, and fouling.[77]

The following paragraph mentions several recent articles in which NF membranes were created by

RI
phase inversion with charged materials in the dope. Zhang et al. used a new variant of poly(arylene
ether sulfone), having pendant tertiary amine groups, to create a positively charged membrane with a
one-step phase inversion process.[78] Membrane tests showed good ability to separate NaCl (about

SC
36% retention) from positively charged dye molecules (> 98% retention) having 408, 479, and 653
Dalton molecular weights. The membrane also exhibited reasonable tolerance to chlorine (1000 ppm
NaOH, 20 days). Bolong et al. cast a phase inversion membrane from a blended dope comprising
negatively charged macromolecules, polyethersulfone, and NMP.[79] Rejection of bisphenol A (BPA) by

U
the membrane was increased at all conditions tested, but including macromolecule in the dope greatly
decreased flux when operating in wastewater (as opposed to DI water). Yu et al. and Xing et al.
AN
prepared positively[80] and negatively[81] charged NF membranes by blending modified SiO2
nanoparticles into the PES dope. The SiO2 nanoparticles were surrounded by branched networks of
either positively or negatively charged grafted chains, and inclusion of either of these additives was able
to increase passage of salts while improving rejection of neutral dyes. Li et al. made membranes from
M
blended dopes with polysulfone (PSf) and sulfonated poly(ether ether ketone) (SPEEK), and then tested
these in IPA.[82] The sulfonated polymer created a more open structure with greater flux, while
increasing the concentration of THF in the dope created a tighter surface. In the alcohol environment,
D

the charged polymer (SPEEK) still contributed to improved rejection of negative dye molecules. Hwang
has examined blending of polyethersulfone (PES) with sulfonated poly(arylene ether sulfone) (SPAES)
TE

and the subsequent impact of thermal treatment.[32] An emphasis was on morphology of the resulting
membrane. Among other things, they noted that thermal treatments reduce surface pore size, but with
a concurrent and more rapid loss in permeability.
EP

To inhibit fouling and enhance rejection, another common approach is to modify the surface of an
existing phase inverted NF membrane. A classic approach to prevent fouling has been to add neutral
hydrophilic groups onto the membrane surface. Seman et al. used UV-initiated graft polymerization to
incorporate one of two hydrophilic monomers (either anionic acrylic acid or neutral N-vinylpyrrolidone)
C

on a commercial polyethersulfone NF membrane.[83] In fouling tests with humic acid, it was found that
the vinylpyrrolidone reduced fouling more than acrylic acid, and that the impact of both was more at pH
AC

7 than pH 3. Zhou et al. caused a layer of sulfonate polyetheretherketone (SPEEK) to adsorb on the
surface of a positively charged PEI membrane, reducing the isoelectric point and increasing rejection of
negatively charged challenge species.[84] Ba et al. show that this combination of positive and negative
constituents results in a more neutral composite membrane that is resistant to fouling.[85] They also
found that PVA adsorption was less effective than polyacrylic acid in both reducing the positive charge
and preventing fouling from negatively charged materials like humic acid and sodium alginate.[86]
Buonomenna et al. modified a poly(ether ether ketone) (PEEKWC) NF membranes using a low
temperature plasma treatment to introduce amino groups.[87] While this is only one of several papers
that could illustrate the impact of charge interactions on ion rejection, it was interesting that one of the
 ACCEPTED MANUSCRIPT

modified membranes was able to pass 100% of methyl orange (327 Dalton MW monovalent anion) and
retain 100% of methylene blue (320 MW monovalent cation).

The examples above help to illustrate the variety of approaches that can be used to modify NF
membrane surfaces. Methods for modifying phase inverted membranes are likely to fall in one of the
following buckets.

PT
Species may be grafted to the surface. UV-initiated graft polymerization was the approach
taken by Seman et al. to incorporate the hydrophilic groups that decreased fouling onto their
polyethersulfone membrane. However, initiation of grafting may also take place through a
redox system (e.g. K2S2O8/Na2S2O5), by plasma initiation, or by gamma rays, and reaction with

RI
the surface usually has an impact on membrane performance.
Species may be adsorbed to the surface. Ba et al. used the adsorption of polymers to their
advantage, preventing fouling by other species. Surfactants, polymers, and foulants will all

SC
associate (to some extent reversibly) with the membrane surface. Whether through charge or
steric effects, all have the possibility to change membrane performance (flux and rejection) as
well.
Plasma may be used to treat the surface. Buonomenna et al. used NH3 gas in the presence of a

U
plasma to introduce amino groups onto the PEEKWC membrane. In the presence of O2, plasmas
can also often be used to increase carboxyl content at the surface.
AN
Functionality may be incorporated by conventional chemically reactions. This was not amongst
the examples of the previous paragraph. However, classic organic chemistry reactions are
illustrated with the VAPEEK polymer, in that it included activated carboxylic acids so the polymer
M
could be crosslinked by reaction with amines. Also, concentrated sulfuric acid is often used to
incorporate a sulfone group on polysulfone, and this usually increases both flux and rejection (of
salts).
D

In a particularly interesting set of papers, Maruf et al. introduce a new method to modify phase inverted
membranes that falls outside of the above categorical approaches. They imprinted the surface of a
TE

microfiltration membrane with a submicron oscillating pattern.[88, 89] While it is commonly stated in
the literature that surface roughness is a contributor to fouling, their experiments showed that oriented
micropatterned surfaces were able reduce particulate fouling. While not yet demonstrated for NF, this
physical approach will be an interesting area for further exploration.
EP

In terms of assessing new NF membranes, the wide range of intended use results in many different
criteria that could characterize Improved performance. Even for just aqueous separations, a plethora
of different tests, test conditions, and challenge species are used. In the future, a community
C

agreement on some standardized testing could help to provide the uniformity required to better assess
and compare membranes.
AC

2.2 Post-treatment of a porous support

Nanofiltration membranes may also be made through post-treatment of a microfiltration or

ultratfiltration support layer. While a variety of support materials are possible (e.g. nonwovens,
electrospun filaments, stretched polyolefins), a common support structure is a phase inverted
membrane. A phase inverted membrane can provide the advantages of optimal pore size (to avoid
excessive span distances), evenly distributed pores across the surface (to reduce the need for lateral
flow through the barrier layer), an asymmetric structure (to reduce resistance from pore flow), and a
 ACCEPTED MANUSCRIPT

smooth surface (to enable application of thinner subsequent layers). Further, many of the same
approaches described previously to modify phase-inverted NF membranes can also be used to convert
UF or MF membranes into NF membranes.

Several approaches (e.g. dip coating, Mayer rod, doctor blade, spin coating, spraying[90]), have been
implemented to coat a polymeric layer on the support. Dynamic coatings (during filtration) are also
used, and these may preferentially deposit a coating towards looser regions. To obtain a very thin

PT
coating without defects, Peng et. al used multiple coating, draining, and drying steps (with crosslinking)
of a dilute solution.[91] In some cases, the coating is chemically reacted with the surface, whereas it
may alternatively only be attached by adsorption or by electrostatic interactions. To maximize the
active surface area and maintain high permeability, it is typically desirable that the NF discriminating

RI
layer is principally created on the surfaces of porous supports. However, whether for reasons of
durability or enhanced separation properties, modifications or polymer formation within pores of the
support can be desirable,[74, 92-94]

SC
2.2.1 Polymer coatings

U
One of the advantages of post-treating supports is that a wide variety of polymers can be selected for
application, depending on ones needs.
AN
Poly(dimethyl siloxane) (PDMS) is a relatively permeable rubber with favorable chemical and mechanical
properties. It has formed the discriminating layer of a few commercial NF membranes targeted for
separations involving solvents and pH extremes.[95] Because of its low polarity, it is typically used in
M
apolar solvents (where it does swell). It can be made sufficiently stable in some organic solvents by
crosslinking. Stafie et al. examined the impact of crosslinking degree on swelling and permeation
properties, finding that crosslinking impacted hexane permeation but not measured retention.[96]
D

Lieven et al. noted that swelling causes PDMS membranes to become less rejecting after exposure to
non-polar solvents, and they found that incorporation of fillers (esp. zeolites) into the PDMS layer
TE

decreases this effect.[97] To allow for thinner PDMS coatings (while still covering the fillers), Vanherck
et al. found it advantageous to include a hollow sphere filler with shells comprising nano-sized zeolite
crystals.[98] In recent years, less activity appears to be directed towards formulating membranes with
this polymer, but several recent studies have attempted to better characterize/predict performance of
EP

PDMS membranes, including the impact of pressure on swelling and compaction.[99-102]

Polyvinyl alcohol formed the basis of early reverse osmosis membranes, and it has recently been
attracting attention for nanofiltration. Several groups have coated PVA onto PS or PES UF supports and
C

studied performance of NF membranes after crosslinked with different molecules (maleic acid, malic
acid, succinic acid, gluteraldehyde, suberic acid). [103-105] This includes the previously mentioned
AC

efforts to attain very thin coating through application of multiple layers.[91] Another recent thrust has
been applying both PVA and a second polymer to the support. To more substantially increase flux and
provide increased charge for rejection, Liu et al. mixed PVA with poly (sodium-p-styrene sulfonate),
applied the blend to a PS support, and then crosslinked the coating with gluteraldehyde to form a
negatively charged NF membrane.[106] Song et al. used a similar approach with a SPEEK/PVA blend to
stabilize SPEEK polymer applied to PS support.[107] Zheng et al. applied a blend of PVA and
polyquaternium-10 to a polypropylene hollow fiber support and crosslinked with gluteraldehyde to
result in a positively charged NF membrane.[108] Zhang et al. coated both the PVA and thermal
crosslinker (3-mercaptoropyltriethoxysilane) onto a polysulfone support and then further oxidized SH
groups to introduce sulfonic acid groups into the membrane.[109] The resulting membrane
 ACCEPTED MANUSCRIPT

demonstrated good stability after strong acids and bases. To provide a membrane with good chlorine
tolerance and improved performance (flux and rejection) compared to PVA membranes, Bano et al.
coated a blend of PVA and sodium alginate onto a porous polysulfone support, and then crosslinked the
coating with a two-step process (using CaCl2 and glutaraldehyde).[110]

Chitin itself is a hydrophobic homopolymer easily produced from the shells of crabs and shrimps.
Various hydrophilic derivatives of chitin (esp. chitosan) have been coated onto UF supports and

PT
crosslinked. Polymers include chitosan[111, 112], sulfated chitosan[113, 114],
carboxymethylchitosan[115-117], chitin xanthate[118], glycolchitin[119], and two chitosans modified
with mesogenic compounds[112]. Chitin xanthate was crosslinked with hydrogen peroxide.[118]
Several other studies mentioned crosslinking the polymers with one or more of epichlorohydrin,

RI
gluteraldehyde, and hexamethylene. Most (not all) of these studies identified a MWCO, and it was
generally between 540-700 Da, corresponding to a relatively loose NF. A typical observation was
moderately high (85-95%) rejection of Na2SO4 and substantially less rejection of MgSO4 during single salt

SC
tests. (In these single salt tests, Mg2+ reduces the effective negative charge on the membrane.) In a
series of publications, Huang et al. also described making several quaternized copolymers of chitosan
(e.g graft copolymers of trimethyl ammonium chloride onto chitosan) that were applied to PAN UF
supports and crosslinked by several means.[120] The MWCO was relatively large (e.g. 920 Da) and the

U
single salt tests showed a rejection order (CaCl2> MgCl2> MgSO4 > NaCl > KCl > Na2SO4 > K2SO4). For
both chlorides and sulfates, the order of cations is consistent with expected rejection by the positive
AN
membrane. All of the above-mentioned studies were done with UF supports of PAN or PS, but chitosan
membranes have also recently been made on polypropylene[121] and poly(1,4-phenylene ether ether
sulfone)[122].
M
Proteins excreted by marine mussels have inspired implementation of dopamine (3,4-
dihydroxyphenethylamine) in a variety of applications. A dilute solution can self-polymerize under mild
conditions to result in a water-resistant, thin and strong, hydrophilic polydopamine (PDA) coating. It has
D

been used to modify membrane surfaces for MF, UF, and RO. There are also several examples of
coating substrates with polydopamine to obtain NF membranes. Li et al. obtained an NF membrane by
TE

just dip coating a PSf UF support in a solution of self-polymerized dopamine, rinsing, and drying.[123]
Zhang et al. formed a membrane by soaking a PES UF membrane in dopamine and allowing the
dopamine to oxidize to polydopamine and adhere.[124] The PDA-coated membrane was then
contacted with PEI to make a positive NF membrane. Li et al. soaked a PS UF support in a dopamine
EP

solution. Subsequently the support was heated and immersed in a hyperbranched PEI. This, in turn, was
subjected to several different crosslinkers (including gluteraldehyde) to form an NF membrane.[125] Lv
et al. co-deposited dopamine and PEI on a PAN UF membrane, and then the coated membrane was
C

further crosslinked with gluteraldehyde to make another positively charged NF membrane.[126]

Perry et al. [127] describes creating membranes with high pH stability by crosslinking polyethyleneimine
AC

(PEI) on a porous support modified to also enhance stability. In some examples, they modified
polyacrylonitrile and PVDF supports by heat treating with a solution of PEI. It was also described that
PES supports could be sulfonated before heat treating with PEI. In preferred embodiments, the PEI is
further crosslinked by cyanuric halide, making an NF membrane more difficult to remove. Perry et al.
[94] also described fabrication of a stable membrane by reacting a polymer (e.g. PEI) and a reactant,
such that a pressure difference across the membrane caused membrane to form within the supports
pores. Both patents were assigned to Bio Pure Technology, and a paper by Freger and Perry [128]
described the BPT membranes as a highly crosslinked melamine-polyamide. They used a number of
techniques (TEM, AFM ,and ATR-FTIR) to characterize their degradation in extreme conditions (e.g.
exposure to 20% sulfuric acid at 90oC).
 ACCEPTED MANUSCRIPT

Coatings of sulfonated polysulfone or sulfonated polyethersulfone have been applied to a porous

support to create negatively charged membranes with good resistance to acids and chlorine.[129, 130]
In the 1980s, a commercial sulfonated polyethersulfone membrane was made on a polysulfone support
and the composite was touted as stable for a month in either 10,000 ppm chlorine or pH 0.5-13 at
80oC.[131] However, the MWCO of that particular membrane is on the high side of NF, and so not
suitable for many applications.

PT
Beyond the above common classes, there have been a wide variety of polymeric coatings with different
characteristics that have been applied to the more porous sublayer. To improve fouling, both

RI
amphiphilic copolymers[132] and a (crosslinked) terpolymer having high zwitterionic content have also
been used.[133] A chelating polymer containing negatively charged groups was adsorbed onto PEI to
improved rejection of ions and adsorb heavy metals.[134]

SC
2.2.2 Surface grafting

Graft polymerization allows new functional groups to be covalently attached to a substrate. Methods

U
used to generate free radicals for polymer grafting include ultraviolet (UV) radiation, low-temperature
plasma, electron beam, -ray irradiation, and redox reactions. While each have been used in making NF
AN
membranes, the most commonly mentioned approach in recent papers appears to be UV
photoinitiation.

With respect to UV photoinitiation, it has been found that polysulfone, polyethersulfone, and Lenzing
M
P84 (a copolyimide that also contains a substituted benzophenone group in its repeating unit) are all
polymers that generate free radicals upon irradiation with UV light and so dont require photoinitiators.
Hence, these are among the most common substrate polymers mentioned for UV-induced
D

photopolymerization. Bernstein et al. created a very negative polyelectrolyte hydrogel suitable for NF
on a polyethersulfone UF support membrane though UV-initiated graft polymerization of vinyl sulfonic
TE

acid and a crosslinking monomer.[135] Homayoonfal et al. varied UV irradiation times, acrylic acid
monomer concentrations, and inclusion of PEGs with different molecular weights in the polysulfone
casting solution (to increase pore size), so that an NF membrane could be created with higher
permeability and reasonable rejection.[136] Behnke et al. introduced side groups containing double
EP

bonds to the P84 polyimide and then cast the modified polymer upon a polyethylene UF support before
crosslink by UV-irradiation.[137] A cardo polyetherketone was found to be more UV-photosensitive
than polysulfones, and so Qiu has cast phase inverted supports from it and then simultaneously grafted
both sodium allyl sulfonate and acrylic acid monomers, so as to improve rejection of ions.[138] Several
C

other references have described photografting different monomers to create either positive or negative
charges on the membranes.[139-142] Some of the above studies include concurrent or subsequent
AC

crosslinking steps.[135-137, 139-141] In another case, Li et al also attached monomers onto PS by UV-
induced graft polymerization, but then used a subsequent crosslinking step to quaternize them.[143]

Papers also described plasma initiation for grafting. Wang et al. created a negatively charged NF
membrane by grafting 2-acrylamido-2-methylpropanesulfonic acid onto a polysulfone UF support that
was, itself, pre-treated by plasma.[144] Chen et al. created a very hydrophobic NF membrane for
separating de-waxing solvents (e.g. toluene, MEK) from de-waxed oil, using a low temperature plasma
to graft styrene onto a PAN UF membrane.[145]
 ACCEPTED MANUSCRIPT

Likely for reasons of cost, availability, and scalablity, fewer researchers form NF membranes using
grafting through electron beams. In combination with UV, electron beams were recently used for
grafting 2-acrylamido-2-methylpropanesulfonic acid onto polysulfone to create a high charge
density.[146]

2.2.3 Membrane charge

PT
Most current commercial NF membranes have a negative charge at neutral pH, and this is often
appropriate to provide desired rejection and limit fouling. A variety of other experimental NF
membranes in recent literature have been created by coating different sulfonated polymers on a porous

RI
support.[93, 147-151] Large negative charge may also be induced by grafting, as described earlier.[135,
138, 139, 142]

SC
There has been a growing interest in positively charged NF membrane. A positively charged membrane
may enhance separation of specific cations for softening applications, removal of heavy metals, or
separation of amino acids below their isoelectric point. By reducing the rejection of large negative
molecules (e.g. dyes), it can provide a degree of selectivity between two otherwise fully rejected

U
species. Adsorption and fouling may also be avoided during operation with cationic surfactants or dyes.
One effective method to create positive membranes is to functionalize a surface with ammonium
AN
groups. The quaternized chitosans created by Huang created positively charged NF membranes on PAN
support.[152, 153] Other quaternized polymers have similarly been applied to UF supports and
crosslinked,[154] or created by simultaneous crosslinking and quaternization of polymers on the
support.[155, 156] In a different approach, polyethyleneimine (PEI) has been reacted or bound to the
M
surface. Branched PEI has been used in several studies to modify a P84 copolyimide asymmetric
membrane, creating positively charged NF membranes due to free amine groups at the membrane
surface.[74, 157] Similarly, Feng et al. coated a hydrolyzed PAN support (having carboxyl groups) with a
D

PEI layer and then crosslinked the layer with different reactive modifiers (e.g. gluteraldehyde,
epichlorohydrin, toluene diisocyanate).[158] Reaction of PEI onto an intermediate layer of dopamine
TE

has resulted in a positive zeta potential at neutral pH.[124] Several other studies created positively
charged membranes through UV-initiated graft polymerization of monomers onto polysulfone UF
support membranes.[141, 142, 159]
EP

2.3 Layer-by-layer (LBL) polyelectrolyte membranes

The layer-by-layer (LBL) approach to fabricating membranes uses alternating applications of cationic and
C

anionic polyelectrolytes. Decher first reported using the LBL method to make films from
polyelectrolytes [160], and research into using these layered structures for NF membranes followed
AC

shortly.[161-163] The space has recently been discussed in several review articles.[22-24] This LBL
fabrication process provides a number of advantages. It provides the potential to accurately control
membrane thickness on the nanometer-scale, and (like interfacial polymerization) there is a degree of
self-healing from the fabrication process. It can be both studied and implemented with a minimal
investment in equipment. These multilayer membranes can be formed from low-cost, water soluble,
commercially available polymers. These membranes are usually formed on flat sheet supports, but LBL
membranes have also been constructed on both the inner and outer surfaces of hollow fibers[164].
Fabrication is relatively simple to control, but the great variety of possible variations in implementation
also provides versatility. In particular, properties of the membrane can be readily controlled through
polymer compositions, concentrations, numbers of applied layers, and application conditions/methods.
 ACCEPTED MANUSCRIPT

A wide range of polyelectrolyte pairs can be implemented. Examples of cationic materials used in recent
studies include poly(ethyleneimine) (PEI), chitosan (CHI), poly(diallyldimethylammonium chloride)
(PDADMAC), and poly(allylamine hydrochloride) (PAH). Polymers with negative charge include sodium
alginate (ALG), polyacrylic acid (PAA), poly(vinylsulfate) (PVS), sodium carboxymethyl cellulose (CMCNa),
and poly(styrenesulfonate) (PSS). While most LBL membranes are created with just a single pair of
polymers, Shi et al. had found advantage in utilizing more than one pair of monomers, in that it provided
greater flexibility to tune selectivity for separations of glucose, maltose, and oligosaccharides.[165]

PT
Deng et al. mentioned that incorporating copolymers in polyelectrolyte provides greater opportunity to
control the membranes chemistry and structure, and that study used a polyanion comprising both
strongly ionized (for stability) and weakly ionized (for properties tunable with pH) segments.[166] Week

RI
polyacids (e.g. polyacrylic acid) were observed to provided more possibilities for control of morphology
and performance than strong polyacids, as charge density was more susceptible to changes in pH and
salt concentration.[167]

SC
Film thickness and permeability depend markedly on the number of layer pairs applied, although not
necessarily in a simple relationship. Some membranes are made with only a few layers,[168-170]
whereas others incorporate more than 50 bilayer pairs.[161, 171, 172] The number of sequential
layered pairs is a parameter easily studied and it is commonly noted that an increase in the number of

U
layers results in greater thickness and corresponds to a decrease in permeability but not always. In
particular, with the LBL(CHI/ALG) system, Lajimi et al. observed that at about 15 layer pairs, there was a
AN
maximum in flux, charge density, and hydrophilicity.[173] This effect was attributed to a transition from
a loose stratified structure to a tightened interpenetrating granular structure. Ahmadiannamini et al.
examined the LBL (PDDA/SPEEK) system, studying the impacts of both the number of successive bilayers
M
(5, 10 ,15, 20) and the NaCl concentration (0 M, 0.1 M, 0.5 M) during deposition.[174] They found that
flux continuously decreased as the number of bilayers increased. However, they also observed
increased selectivity (Cl-/SO42-) between 5-10 bilayers and reduced selectivity between 15-20 bilayers.
The former effect was attributed to a reduced number of defects available to pass the larger SO42- ions,
D

whereas the latter observation was related to a change in swelling that reduced the importance of ionic
size in exclusion.
TE

The outermost layer can be particularly significant for performance of the LBL membrane. Zeta
potential has been seen to alternate between positive and negative in subsequent layers.[175]
Terminating the LBL structure with a polycation results in a positive surface charge, whereas terminating
EP

with a polyanion results in a negative surface charge. Also, the last surface coating can impact
hydrophilicity, which may be relevant to fouling.[169, 175] Permeability of the LBL membrane has been
seen to vary (even-odd) with added layers and ascribed to differences in swelling as the outer surface
C

changes.[176] While not always the case, Miller et al. further found that the terminal polymer could
profoundly impacted swelling of the LBL membrane, presumably because certain terminal capping
AC

polymers were able to penetrate into the entire film and disrupt crosslinking.[177]

To better understand different even-odd swelling effects, de Grooth et al. studied interactions of LBL
(PDADAMAC/PSS) membranes with their adjacent support (SPES UF membrane, 75 kDa MWCO) and
observed two limiting cases.[176] 1) For thin films where the multilayer membrane was coated on the
inside of the pores, transport of both water and ions was governed by convective flow through the small
pores. Ion rejection was predominantly based on size, and swelling of the membrane resulted in a
decreased permeability. 2) For thick films on the top surface, transport was governed by solution
diffusion through these layers. The rejection of ions was impacted by charge in the multilayer
membrane, both on the surface and in the bulk. Swelling of the membrane increased water
permeability.
 ACCEPTED MANUSCRIPT

Conditions at which the polyelectrolyte is applied are also important. Krasemann et al. found that
greater membrane thickness was obtained by using conditions for layer application with either
increased salt concentration or increased pH.[161] It was asserted that higher NaCl concentration
reduced electrostatics interactions between polymer chains and resulted in a membranes incorporating
coils (instead of a flat surface) with more charged chain segments (that decrease permeability but
improve rejection of ions). By contrast, increased pH during polyelectrolyte application caused greater
thickness through incorporating more non-ionized chain segments that swell. Consistent with the first

PT
part, Ouyang et al. made a series of positively charged LBL membranes and found that higher surface
charge, increased Mg2+ and Ca2+ rejection, and greater monovalent/divalent selectivity could be
obtained by applying the final deposition layer in the presence of supporting electrolyte having high

RI
ionic strength.[178] Rajabzadeh et al. found that rejection was principally controlled by the charge
density of the capping layer, in that they observed similar performance when base layers were produced
with a low concentration of supporting electrolyte.[168]

SC
In a typical approach, individual polyelectrolytes are applied and then excess and weekly associated
chains are rinsed from the surface to make a thin layer. Attractions between opposing charges promote
even surface coverage, and electron repulsion from ions of the same charge can discourage a buildup of
excess polyelectrolyte molecules. Each application may be performed by just contacting the surface

U
with a polyelectrolyte or by flowing the polyelectrolyte across the surface. Xu et al. showed that
improved rejection was possible by a dynamic process in which solutions were applied during
AN
pressurized filtration.[166] This selectively directs polymer deposition to regions of greater flux,
resulting in fewer layers being required to produce a uniform membrane. Su et al. reported that a
combination of static and dynamic methods was most favored, where low molecular weight polymers
M
were applied without permeation and higher molecular weight polymers were applied at elevated
pressure to induce flux.[169]

LBL layer membranes can be sensitive to operating conditions. Membranes may swell or even dissociate
D

in high ionic strength, at low and high pH, or in the presence of chlorine. For this reason, several LBL
membranes have been further crosslinked.[164, 170, 179] To improve stability, Saeki et al. has
TE

examined crosslinking polyelectrolytes through both amide bond formation and siloxane bond
formation.[180] While performance of membranes based on amide bonds was degraded in chlorine,
membranes based on polyelectrolytes modified with reactive silane groups showed high stability in both
high ionic strengths and with chlorine. In a different approach, Shan et al. suggests the instability of
EP

some LBL membranes may be used to advantage.[181] As an approach to mitigate fouling, a pH

responsive LBL membrane may be applied to a UF support, fouled during operation, and then the LBL
film may be removed from the support (along with foulants) layer before it is re-applied. For instance,
C

Shan et al. demonstrated that LBL (PSS/PAH) membranes on a UF support may be fouled by SiO2 and
then regenerated at high pH. In that case, a backflushing step was found to greatly improve the
AC

efficiency of regeneration. (Backflushing has also recently been used as part of the stability test for LBL
(PDADMAC/PSS) membranes.[182] This allowed optimization of ionic strength during fabrication, so as
to get stronger attractive forces to form between the polyelectrolyte layers.)

2.4 Aquaporin-based membranes

Aquaporins (AQP) are proteins that form pores in membranes of biological cells. They are capable of
selectively transporting water molecules high at rates, and Kumar et al. showed that membranes
manufactured to incorporate these species could have permeabilities of about 600 L/m2/hr/bar, which is
 ACCEPTED MANUSCRIPT

nearly 50 times greater than even NF270 membrane.[183] Several recent articles discuss the potential
for aquaporins to dramatically lower the energy for desalination, and describe the challenges of
incorporating these water channel proteins into membranes.[184-188] Duong et al. made nanofiltration
membranes by embedding AQPs in vesicles of tri-block polymer coated onto gold-coated porous
alumina substrates.[189] Zhong et al. used a phase inverted cellulose acetate membrane as support and
formed an NF membrane by applying AQPs in a tri-block copolymer to the support and UV
polymerizing.[190] Several groups have since immobilized intact liposomes containing aquaporins

PT
(proteoliposomes) on a support within a layer that prevents extraneous salt passage.[191-194] For
instance, Sun et al. embedded vesicles containing AQPs within a LBL NF membrane[193] and used
magnetic nanoparticles within proteoliposomes to increased the concentration of AQPs on the

RI
membranes surface.[194] Zhao et al. demonstrated that reverse osmosis membranes could be made
more permeable by incorporating liposomes containing AQPs within the aqueous phase of an interfacial
polymerization (MPD+TMC).[195] Figure 2 illustrates (a) the interfacial polymerization process and (b) a
proposed conceptual model for increased permeability of the barrier layer by traversing incorporated

SC
proteoliposomes. Li et al. has also used a similar approach to make hollow fiber membranes with water
and salt permeabilities similar to very high flow (Dow XLE) commercial membranes.[196]

U
AN
M
D
TE

Figure 2. Illustrations convey (a) incorporation of aquaporins by interfacial polymerization and (b) a
conceptualization of the resulting barrier layer[195]
C EP
AC
 ACCEPTED MANUSCRIPT

2.5 Glassy polymers with high free volume

Polymers having high free volume due to the rigid nature of their macromolecular chains are potential
candidates for solvent resistant nanofiltration. Polymers of Intrinsic Microporosity (PIM) are a class of
polymers having a rigid molecular backbone and sites that force contortions, so that inefficient packing
results.[197] An archetypal example, PIM-1, is shown in Figure 3. It has a backbone consisting of fused

PT
rings separated by spiro-centre structures that cause it to erratically twist and turn, resulting in high
internal porosity and interconnected void regions having pores less than 2 nm. PIM are only soluble in a
few solvents (esp. THF, chloroform), and they are expected to have good stability in many other solvents

RI
without need for crosslinking. Fritsch et al. has formed composite nanofiltration membranes by dip-
coating PIM polymers on PAN support.[198] Their experiments further examined the impact of
crosslinkers and of coating the PIM on a gutter layer (Teflon or PDMS) established between the PIM
polymers and support. As compared to a commercial solvent resistant NF membrane, a PIM membrane

SC
without crosslinking had both lower MWCO (based on filtration of a polystyrene mix) and more than 30
times higher permeability of n-heptanes. Measurements in toluene also showed better performance
(rejection and permeance) for a crosslinked PIM compared to a commercial solvent resistant

U
membrane. Volkov et al. formed NF membranes by casting poly[1-(trimethylsilyl)-1-propyne] (PTMSP),
another glassy polymer with a free volume fraction up to 25%, on PAN support, and measured moderate
AN
permeabilities to methanol, ethanol, and acetone.[199] As compared to rubbery polymers like PDMS,
both PTMSP and PIM-1 have shown much higher permeability of ethanol.[200]
M
D
TE

Figure 3. Structure of PIM-1

EP

3.0 Interfacial polymerization

C
AC

Interfacial polymerization is the process used to make most commercial nanofiltraiton membranes.[201]
However, it is also the approach used to make membrane in the vast majority of reverse osmosis
elements. Because of the similarity in these membranes, it is frequently difficult to distinguish advances
in NF from the even larger body of work directed to reverse osmosis. Consequently, the following
section will be guided by the intent to highlight those aspects that seem most relevant to understanding
nanofiltration membranes.
 ACCEPTED MANUSCRIPT

3.1 Historical perspective on the interfacial polycondensation

Several previously mentioned reviews describe the history of interfacially polymerized membranes.[8,
19, 202] Its roots are typically traced back through various seminal discoveries including: 1) the phase
inversion process that enabled the creation of asymmetric cellulose acetate membranes by Loeb and
Sourirajan, 2) the development of composite membranes having a discriminating layer built on a more

PT
porous substrate, and 3) different approaches (e.g. float casting, meniscus coating) initially used to
create thin film layers. Significant attention may also be given to early phase inversion membranes (and
supports) from materials other than cellulose acetate (esp. polysulfone). However, as this review article

RI
emphasizes chemistry, a slightly different path is highlighted.

Wallace Carothers, the inventor of nylon, had shown in 1929 that polycondensation reactions at high
pressure and temperature (for example, melt polymerization well above 200oC) were able to produce

SC
high molecular weight polyamide.[203] In 1959, Wittbecker and Morgan [204, 205] produced a high
molecular weight copolymer by employing a much faster Schotten-Baumann reaction of an acid chloride
with a compound containing an active hydrogen atom (-OH, -NH, -SH). Their reaction took place at the
interface of two liquid phases, and they coined the name interfacial polycondenstation. Several

U
subsequent studies by Morgan et al. [206, 207] demonstrated significant points of differentiation for the
interfacial polycondensation. In a typical step growth polymerization, more than 99.9% purity of
AN
monomers and a strict 1:1 stoichiometric balance of the functional groups of the reactants are needed
to ensure the production of high molecular weight polymers. By contrast, the interfacial
polycondensation can tolerate significant levels of impurity in the monomers and a strict stoichiometric
M
balance is not required. Instead, functional group equivalence is maintained at the interface even
though the bulk concentrations are different. Also, Carothers [208] initially made low molecular weight
polyamides from diacid chlorides, and heating at elevated temperature was necessary to increase the
molecular weight. The interfacial polycondensation method could produce a high molecular weight
D

copolymer at low temperatures in a very short time. Perhaps most important for membranes, because
the produced polymer is not soluble in either phase, a very thin film could be readily created.
TE

Following the discovery of interfacial polycondensation, there were a few early attempts to make
membranes by reacting monomers through this process. Lonsdale et al. [209] attempted to react
ethylenediamine with succinyl chloride, but the resulting membrane had no rejection of salts. Scala et
EP

al. [210] reacted various aliphatic diamines and aliphatic diacid chlorides, but this generally resulted in
low flows. By contrast, Cadottes NS-1 membrane, later re-named NS-100, was the first interfacially
polymerized RO membrane that combined high flux with sufficient salt rejection to desalinate water at
C

reasonable pressure. This polyurea membrane was made by reaction of toluene diisocyanate (TDC) with
polyethyleneimine (PEI) on a polysulfone support [211-213]. The first polyamide based RO membrane
AC

was subsequently made when TDC was replaced by isophthaloyl chloride (IPC).[214] Wrasidlo [215] and
Riley et.al.[216, 217] made analogous membranes that became commercially important by reacting
polyepiamine with TDC and IPC. Other polymeric amines studied (alone or in conjunction with a
monomeric amine) included polyvinyl amines [218], poly(diallyl amines)[219], and polyvinylimidazolines
[220-224]. For a short time following the success of NF-100, it was commonly thought that polymeric
amines were required to achieve good membrane performance. However, building on research by
Credali et al. [225] using monomeric piperazine-based molecules with various carboxylic acids, Cadotte
[226] reacted piperazine and isophthaloyl chloride to achieve a membrane from these two monomeric
reactants that demonstrated high salt rejection (98%) at seawater feed conditions. He further improved
the trade-off between flux and rejection when a trifunctional acid chloride, TMC, was reacted with
 ACCEPTED MANUSCRIPT

piperazine to produce NS-300 membrane [226, 227]. To this date, this piperazine-TMC chemistry is the
basis of most commercial TFC nanofiltration membranes, although membrane properties may differ
substantially due to other constituents and process conditions. Examples of commercial piperazine-
based membranes include Dow's SR90, NF270, and NF200, Torays UTC60, TriSeps TS40, XN45, and
DSIs/Osmonics Desal-5 (which is similar to GEs DK and DL membranes). Shortly after working on the
piperazine-TMC membrane, Cadotte also discovered that a membrane with unprecedented high flux
and high rejection was produced by reacting an aromatic diamine, m-phenylene-diamine (MPD), with

PT
TMC. This reaction, referred to as FT-30 chemistry, is now used to produce almost all commercial
reverse osmosis membranes [228].

Subsequent sections of the paper describe research advancing interfacially polymerized NF membranes,

RI
with an emphasis on areas of chemistry. Broad topics include the composite membrane, new
monomers, polymeric reactants, additives, and post-treatments. The focus will be on studies of NF
membranes, but examples from RO will also be included.

SC
3.2 Composite membrane

U
The structure of an interfacially polymerized composite membrane comprises at least a porous support
layer and a thin film discriminating layer. Flat sheet commercial membranes also include a web layer for
AN
added strength. Other additional polymeric layers contacting the discriminating thin film (below or
above it) may further be present to enhance adhesion, molecular separations, or fouling resistance.
Figure 4 illustrates components of a typical polyamide membrane, including the cross section of a 34 nm
polyamide discriminating layer obtained from Dows NF200 membrane. The different layers of the thin
M
film composite (TFC) are created individually and are almost always of different chemical composition.
A composite structure allows for optimization of the separate parts for their respective purposes. The
thin film discriminating layer will be described in detail later. The following section describes these other
D

important components.
TE
C EP
AC

Figure 4. Components of a typical interfacially polymerized NF composite membrane

 ACCEPTED MANUSCRIPT

3.2.1 Web for strength

Commercial composite membranes are usually constructed on a sheet of non-woven, fibrous, paper
web onto which the porous support layer is applied. Suitable non-woven polyester fibrous sheets are
manufactured by several companies (e.g. Ahlstrom, Awa Paper Mfg., Freudenberg, Hirose Paper, Kavon
Filter Products, Neenah Technical Materials). Tightly woven fabrics such as sail cloth are also used, but

PT
are not typically for commercial NF membranes. The primary purpose of this web is to provide tensile
strength to prevent the composite membrane from deforming in operation into channels within the
permeate spacer of a spiral wound element. The web is often designed for use in reverse osmosis
applications that often employ more than 1000 psi (~70 bar) applied pressure. Most NF applications are

RI
not similarly challenging with respect to pressure, and so an opportunity exists in many cases to
optimize this layer of the NF composite for much lower pressure requirements.

SC
The web can influence performance of the composite membrane, although not typically to the same
extent as the support or discriminating layer. In certain aggressive cases, a polyester web may be
weakened by high temperatures and extremes of pH during operation or frequent cleanings. For this
reason, polypropylene has been occasionally used as an alternate web materials for use in some

U
applications.[229] Excessive penetration of the porous support layer into the web during its fabrication
can cause flow resistance. However, this needs to be balanced against the fact that sufficient adhesion
AN
between these two layers is required to prevent delamination (assuming accidental backpressure) of the
support layer from the web in operation.[230-232] A dense web with substantially spaced apart paths
for fluid permeation can require lateral flow to take place through the support, also increasing flow
resistance.
M
Most NF membranes are sold in a spiral wound configuration and use a web-backed flat sheet.
However, other form factors would not use a web at all. For instance, the thin film discriminating layer
D

is often formed on either the internal[233, 234] or external[235, 236] surfaces of hollow fibers.
Microcapillary sheets[237, 238] and multi-channel fibers[239] provide similar options. Nanofiltration
TE

membranes have been made on dual layer hollow fibers.[235] Multilayer extrusion may be used to
create support layers for both hollow fiber[240] and flat sheet[241], and one theoretical advantage is
that even greater asymmetry (and increased permeance) can result from a further tailoring of individual
layers. Several patent applications include integration of the composite membrane and permeate spacer
EP

for use in spiral elements.[242, 243]

3.2.2 Porous support layer

C

A porous support layer is applied to one surface of the web. During the interfacial polymerization, the
AC

porous support serves as a reservoir for the water soluble reactant. Upon bringing the support layer
into contact with the oil phase reactant, polymerization happens at or near the water/oil interface. The
extent of reaction that takes place within the growing polymer phase, as compared to within the water
or (especially) oil phase, has not been determined. Although the bulk of polymer results above the
support layers surface, some polymer may also be found within the pores. Preferably, a part of the
polymeric discriminating is at least physically attached to the porous support layer, if not chemically
bound.

Most commonly, the porous support layer is an asymmetric ultrafiltration membrane formed by
diffusion induced phase inversion. Studies with reverse osmosis membranes (MPD+TMC) have observed
 ACCEPTED MANUSCRIPT

that changes to the support pore size impacts the location, thickness, and permeability of the resulting
polyamide barrier layer[244, 245] Recent studies by Lu et al.[246] found that the solvent (NMP or DMF)
used in casting polysulfone affected resulting polyamide structures. Increased hydrophilicity of the
support was also seen by Ghosh et al. to decrease water permeability for the resulting interfacially
polymerized RO membrane, and this was explained based on both differences in initial MPD diffusion
rates and the orientation of menisci within liquid-filled pores after excess water had been removed with
an air knife.[245] Li et al. formed polyamide (PPD+TMC) membranes on different cellulose acetate

PT
propionate support layers and observed that large pores resulted in a rougher polyamide,[247], as also
seen by Ghosh. [248, 249]Several studies have since examined changes to the support layer for
piperazine-based NF recipes. Misdan et al. made piperazine-based nanofiltration membranes on

RI
supports having different pore sizes, formed from different concentrations of polysulfone in NMP.[250]
It was found that polyamide thickness decreased with increasing pore size and that flux was highest in
their examples at about 15% polysulfone, but that rejection was not substantially changed between 12-
20% polysulfone. In another study, Misdan et al. compared piperazine-based membranes formed on

SC
polysulfone, polyethersulfone, and polyetherimide, and they also found that the support impacted the
resulting polyamide, but that hydrophilicity was more important than pore size/structure.[251] Zhu et
al. also examined the influence of support properties on the subsequent polyamide (PIP+TMC) layer

U
when the amount of PANI in the PES/PANI casting dope was varied from 0-0.2%.[252] Elaborating on
several of the concepts described by Misdan[251], Li[247] and Ghosh[245], they attempted to explain
AN
changes in polyamide structure (crosslinking and thickness) and performance (flux and rejection) with
differences in the support, this time observing better correlations with pore size instead of
hydrophilicity. Jimenez-Solomon et al.[253] made NF membranes by polymerizing MPD and TMC on
both crosslinked polyimide and poly(ether ether ketone) supports. While the resulting polyamide
M
structure was clearly different by SEM, authors attributed the relative differences in flux while using THF
and toluene solvents to differences in the hydrophilicity of the two supports. Ramon et al.[248, 249]
modeled the influence of the support pore size and distribution on localized flux and fouling,
D

determining that the support features will be more important for highly permeable barrier layers (e.g.
nanofiltration). While it is obvious from all these studies that properties of the support clearly affect the
interfacial polymerization process, it is also reasonable that reactions using different co-monomers,
TE

additives, and polymerization conditions (e.g. temperatures, nipping or air knife methods, and
polymerization time) have different reaction rates, result in different structures, and have different
limiting trends.
EP

It has been stated that low surface roughness of the support is crucial to avoid defects in the interfacial
polymerization of NF membranes.[253] On the other hand, supports with more than 100 nm periodic
ridges have been used to make RO membranes, and the patterned surface was seen to reduce scaling
C

propensity of the polyamide.[254] (This support is akin to the UF membrane that was mentioned earlier
as having submicron periodic imprinting and that resulted in a decrease in particulate fouling.[88, 89])
AC

Aside from its impact on the interfacial polymerization, the support layer also has an effect on
performance during membrane operation. Because NF membranes typically have greater permeability
than RO membranes, one might expect that the contribution of the support to flow resistance would be
generally greater for NF membranes. In the case of NF270, one of Dows high flow NF membranes, the
uncompacted porous support layer has more than 20 times higher water permeability than the
composite. Compaction of the support under operating pressure is still observed to reduce permeability
of the composite, but common parameters (e.g. % solids, temperatures, additives) of the phase
inversion can be optimized to create a support layer less susceptible to compaction.[255-257]
Incorporation of nanoparticles within a support has also been seen to reduce compactions.[258, 259]
 ACCEPTED MANUSCRIPT

Especially for high flux membranes (such as NF), the porous support may result in inhomogeneous flow
patterns through the interfacially polymerized membrane. Because of this, numerical calculations
indicate that fouling propensity may be favorably impacted by a support with small pores and high
surface porosity.[248, 249]

Several studies have modified the support used for interfacial polymerization. An increase in
hydrophilicity of the support could assist in wetting at low pressures, although that rarely appears to be

PT
a problem in the field. In a preferred situation, it might allow the manufacturer to omit humectants for
drying. It might also open up a range of new supports. Polypropylene would be among the cheapest of
polymer materials. By ozonation and grafting with acrylamide, Pan et al. was able to make a
polypropylene support sufficiently hydrophilic to wet an amine solution and demonstrate interfacial

RI
polymerization.[260] However, despite using a recipe typically resulting in RO membrane, 60% rejection
of Na2SO4 resulted. Yu et al. used plasma polymerization to incorporate amino groups onto PTFE
surfaces, and these groups were then used to link the substrate to a subsequently polymerized

SC
polyamide (EDA+TMC).[261] Song et al. mixed SPSf in different ratios with PS, and then used three
different amines (MPD, PPD, PIP) to form interfacially polymerized membranes with TMC.[262] An
interesting observation was that an interpenetrating layer between the support and the polyamide was
seen when SPSf was used. There was also a noted decrease in delamination bubbles, suggesting

U
better adhesion between the two layers.

AN
The solvent resistance of typical interfacially polymerized thin film composite membranes is often
determined by the support layer, as the polyamide barrier layer is highly crosslinked. The support layer
is most commonly made with polysulfone or polyethersulfone. These support layers are stable to many
solvents, but they are not stable in polar aprotic solvents (e.g. DMF, NMP, DMAC, TEP, DMSO). By
M
contrast, the polyamide discriminating layer for interfacially polymerized membranes is thin, defect free,
and stable in these solvents. Hence, a recent thrust has been toward creating an improved and
compatible support layer for these applications.
D

As described previously for phase inverted membranes, crosslinking can improve the stability of a
TE

support layer made by phase inversion. Crosslinked polyimides have been used several times as a
support for interfacially polymerized solvent resistant membranes.[263] To simplify the manufacturing
process, Hermans et al. added two diamines (hexanediamine and MPD) to the coagulation bath.[264]
Ideally, a single diamine would suffice, but they found that the former was needed for crosslinking the
EP

polyimide support and the latter was required for the interfacial polymerization. Jimenez-Solomon et al.
formed interfacially polymerized membranes (MPD+TMC) on both PEEK supports and crosslinked
polyimide supports.[253] They found that PEEK was the optimal support for a non-polar solvent
(toluene), but the polyimide was better for use with a polar aprotic solvent (THF). Aburabie immersed a
C

dope solution (polythiosemicarbazide/DMSO) in water and crosslinked it with ,-dibromo-p-xylene

before forming polyamide (TMC+diaminopiperazine) on its surface.[265] Prez-Manrquez et al.
AC

crosslinked a PAN support with hydrazine and made an interfacially polymerized nanofiltration
membrane from diamino piperazine (DAP) and TMC. The resulting membrane had excellent stability in
DMF.[266]

High temperature operation of NF membranes would have advantage in several applications, especially
those where it could avoid a need to cool the feed water. Some factors associated with spiral wound
modules, particularly the permeate spacer and sometimes the permeate tube, can be more limiting to
high temperature operation than the composite membrane itself. However, there are several examples
that show increased temperature stability of piperazine-based membranes by using different support
layers. Poly (phthalazione ether nitrile ketone) has been examined by Hu et al. as a novel support for
 ACCEPTED MANUSCRIPT

high temperature polyamide (piperazine + TMC) membranes.[267] Going to 80oC and even boiling, flux
increased at similar pressures (as expected) and rejection of Na2SO4 stayed approximately stable. Han et
al. showed that an interfacially polymerized (piperazine+TMC) composite membrane formed on a phase
inverted support of copoly(phthalazinone biphenyl ether sulfone) (PPBES) had improve lifetimes at 85oC,
as compared to a similar thin film composite membrane on polysulfone.[268] Membranes of
poly(phthalazinone ether sulfone ketone) (PPESK) have also shown better thermal stability than
polysulfone and these have been used as a porous support layer for the piperazine reaction.[269] Wu et

PT
al. created a (PIP+TMC) composite with a UF support of poly(phthalazinone ether amide) polymer, a
Tg>300oC material.[270] Filtration of dyes was performed at 80oC, but no comparison was made to the
performance of membranes made on polysulfone. Liang et al. has implemented an annealing step on

RI
the polysulfone support, and this was found to favorably impact the pore structure of its bulk layer.[271]
This aspect was partially credited for an improved stability of the piperazine-based nanofiltration
membrane at high temperature. In summary, there are a number of options to use alternative supports
to increase the stability of the composite membrane to high temperature, but changes would need to

SC
be concurrent with other modifications to the spiral wound element construction.

While not especially common, polyamide membrane are sometimes used in conjunction with moderate
concentrations of alcohols.[272, 273] However, citing long-term instability that was reported for

U
exposure of polysulfone phase inverted [ultrafiltration] membranes to 70% ethanol,[274] Kosaraju et al.
used a polypropylene support for the interfacial polymerization of poly(ethyleneimine) (PEI) and
AN
isophthaloyl dichloride (IPD).[275] They found the resulting membranes were stable (4 weeks) to
methanol and ethanol. Oh et al. modified a porous PAN support with hydroxide to create carboxylic
acids, and an interfacially polymerized (piperazine +TMC) membrane was formed thereon.[276] They
M
attributed improved stability after soaking in strong (50/50) water/alcohol solutions to ionic bonds
formed between the modified support and the polyamide discriminating layer. Peng et al. activated a
hydrolyzed PAN (HPAN) support so that carboxyl groups would react with amine, attempting to increase
the density of polyamide and prevent defects.[277] An interfacial polymerization (PIP+TMC) was
D

performed and the composite operated satisfactorily in 30% ethanol.

TE

Electrospinning has recently been discussed as an alternative to phase inverted porous supports for
interfacially polymerized membranes.[278-280] When internal polarization is known to be a problem,
such as in forward osmosis, greater openness of the support layer could likely provide an advantage.
Bui et al.[281] electrospun fibers from solution (polysulfone/polyethersulfone/DMF/NMP) onto a
EP

polyester non-woven fabric and, after interfacial polymerization (MPD+TMC), observed good
performance in forward osmosis tests. Several references show these supports working for interfacially
polymerized nanofiltration membranes as well. Yoon et al. [282] used electrospun PAN fibers as the
C

intermediate layer in a piperazine-based interfacial polymerization, and these composites were found to
have higher fluxes than membranes similarly made on a commercial phase-inverted PAN support. Using
AC

electrospun polyethersulfone fibers applied to a PET polyester web, Yung et al.[283] found that the ratio
of solvents (NMP and DMF) could be used to optimize adhesion between these layers. Kaur et al. [284]
spun PAN fibers onto a nonwoven polyester sheet, pressed the composite under heat to decrease the
effective pore size and improve adhesion to the polyester, and created an interfacially polymerized
(piperazine + TMC) membrane. SEMs in Figure 5 show electrospun surfaces with and without hot-
pressing (87oC, 0.28MPa), and the corresponding interfacially polymerized membranes formed on each.
The surface of both was rough compared to conventional membranes on standard PS supports, and the
shape of fibers could be easily seen through the polyamide surface. Hot-pressing before polymerization
resulted in improved adhesion of the spun layer and better stability of the polyamide membrane at
higher (>130 psi) operating pressures. The same group also electrospun a top layer of fibers made from
 ACCEPTED MANUSCRIPT

different concentrations of PAN and found that smaller fiber sizes resulted in smaller pores on the
support, and this also increased MgSO4 rejection of the subsequent piperazine-based thin film
composite.[285] It is interesting that images within each of these papers demonstrate how
electrospinning results in a layering of fibers with an indeterminate surface pore size but one much
larger than the 5-50 nm reported elsewhere for the pore size of phase inverted UF supports often used
for commercial interfacial polymerization.[19] One of the authors has made interfacially polymerized
membrane on pores larger than 1 micron, but the thin polyamide must span a larger distance and

PT
reliability can be a problem in that case.

RI
U SC
AN
M
D
TE
C EP
AC

Figure 5. SEMs of electrospun support (a) with and (b) without hot-pressing, and the corresponding (c,
d) interfacially polymerizied piperazine membranes formed thereon. [284]

3.2.3 Intermediate layer

 ACCEPTED MANUSCRIPT

Between the support and interfacially polymerized barrier layer, the composite may comprise an
intermediate layer as well. For instance, it was reported that Desal 5 piperazine-based membrane
contained a sulfonated polymer between the microporous polysulfone support and the thin
polypiperazineamide layer.[9] However, as it is often difficult to detect this type of modification in the
final product, membrane manufacturers may choose to keep aspects such as this as a trade secret.
Nonetheless, there are several examples in the open literature suggesting how an intermediate layer
may have advantage. Li et al. applied an intermediate layer of dopamine between the PES support and

PT
piperazine membrane.[286] This enable better wetting of the PS support and enhanced interfacial
strength between the support and active layer. Zhang et al. found that forming the support layer in the
presence of poly(ethylene oxide) allowed it to segregate into the polyamide and even favorably impact

RI
fouling of the NF composite.[287] When Pluronic F127 was present in the substrate dope, the
composite showed less flux decline during operation with BSA and there was improved flux recovery
after water washing. Song et al. observed an interpenetrating region between the support and
discriminating layer caused by inclusion of SPSf in the support.[262] They attributed increased adhesion

SC
of the two layers to the material betwixt. One might hypothesize that several of the previously
described reactive and non-reactive post treatments for phase inverted membranes might also be
advantageously applied to support layers used for thin film composite NF membranes as well.

U
3.2.4 Surface coatings

AN
A composite membrane may also include a coating on top of the interfacially polymerized layer to
enhance rejection, reduce fouling, or just protect the membrane during element manufacturing. With
respect to fouling, two common goals include reducing surface roughness and making the membrane
M
more hydrophilic. Towards this end, neutral hydrophilic polymers are often applied. Either positively or
negatively charged coatings can be appropriately used against specific foulants. Coatings may be just
adsorbed on the surface, crosslinked with itself, or covalently attached to the interfacially polymerized
D

polymer. Patents [288-291] describe different adsorbed or chemically bound coatings that may be
applied to RO and NF membranes. Other applications mention removable coatings for purposes of
TE

protecting the membrane in fabrication, storing the membrane in a stable environment, or to enable
better removal of foulants or scale.[292, 293] For many further examples, the reader is referred to
citations within these applications and patents. Some information on coatings used in commercial
membranes and their impact on performance is suggested from an analysis of 17 commercial
EP

membranes by FTIR and XPS.[294] A companion paper studying performance and physical
characteristics noted that the effect of PVA coating on semi-aromatic piperazine membranes was small
compared to the impact measured for aromatic polyamides.[295]
C

Specific to nanofiltration, Draevi et al. [296] studied the impact of PVA coating on NF270. PVA caused
the rejection of hydrophobic solutes to increase, the rejection of hydrophilic solutes to decreased, and
AC

the passage of low interacting solutes (e.g. NaCl) was unchanged. Akbari [297] immersed interfacially
polymerized (PIP+TMC) membranes in different concentrations of chitosan solutions for different times,
allowing in situ reaction of amines and hydroxyl groups of chitosan. The modified membranes had
greater hydrophilicity, reduced roughness, and fouled less in the presence of CTAB cationic surfactant.
Zhou et al.[298] started with a commercial NF membrane (PIP+TMC) and coated different concentration
of poly(ethyleneimine) on the surface. The adsorbed layer caused a slight increase in water
permeability and minimal change in neutral PEG or sucrose passage. However, the membrane acquired
a positive charge and rejection of divalent cations and methylene blue increased. Additional examples
of coatings specific to NF will be described in later discussions of post treatments.
 ACCEPTED MANUSCRIPT

3.3 Monomers

Although a reaction including PIP and TMC is most common for NF membranes, several alternatives
have been examined. For amine monomers, differences in the reactivity and solubility are both

PT
impactful to the resulting polyamide. The tertiary structure and flux of the polyamide is often seen to
be particularly affected by the rate at which an amine crosses the interface into the oil phase. By
contrast, TMC is a trifunctional acid chloride and is typically the monomer responsible for the
polyamides crosslinked network structure. Also, unreacted acid chloride groups of TMC hydrolyze into

RI
carboxylic acid groups during the membrane fabrication process, and these carboxylic acid groups
impact membrane performance by increasing hydrophilicity and by rejecting salts through Donnan
exclusion.

SC
Several studies with alternative monomers have been aimed generally at improving conventional
performance properties (flux, rejection, fouling). These studies are grouped roughly into those using
monomers to impact the barrier layer through charge, hydrophilicity, or structure. Subsequent

U
groupings are focused on use of monomers having impact on chlorine resistance or pH/solvent stability.
These groups are described below.

3.3.1 Monomers with charged groups

AN
M
Increased charge introduced through the monomers can have an impact on flux, rejection of ionics, and
fouling. One effect of increased charge is through greater Donnan interactions, so that rejection by the
membrane of similarly charged molecules will often be improved. Increased charge content can also
D

result in swelling of the polymer, which may favorably impact flux but can negatively impact rejection
(especially of neutral species). In some cases, resulting membrane charge can also modify fouling
TE

characteristics both favorably and otherwise.

Carboxylic acid containing amine monomers such as 3,5-diaminobenzoic acid, an aromatic amine
monomer with a carboxylic acid group, have been used in the past to improve flux such as in asymmetric
EP

membranes[299] or polyester based membranes[300]. More recently Ahmad et al. [301-304] used 3,5-
diaminobenzoic acid (BA) in addition to piperazine to make a more hydrophilic NF membrane. In a first
study,[301] flux, NaCl passage, and Na2SO4 passage all increased with BA, but passage of Na2SO4
maintained a satisfactory level. In subsequent reports, [302-304] flux was seen to increase with BA
C

content, but sulfate passage changed minimally at most conditions. The pressure dependence of the
effects of BA was measured and ascribed to differences in polyamide compaction due to the different
AC

molecular confirmations (planar vs. chair).[303]

Sulfonic acid groups containing monomers have also been a popular choice to increase hydrophilicity of
the membranes. TFC RO membranes have been made with a blend of MPD and m-phenylenediamine-5-
sulfonic acid[305] or with 2,2-benzidinedisulfonic acid[306], or with sulfonated cardo poly (arylene
ether sulfone) bearing pendant amino groups[307], or with 2,5-diaminobenzenesulfonic acid to make
novel composite mosaic membranes [308]. In all cases flux of the membrane improved with the
inclusion of sulfonic acid containing monomer. Liu et al.[309] synthesized two novel sulfonated
aromatic diamine monomers, 2,5-bis(4-amino-2-trifluoromethyl-phenoxy)benzenesulfonic acid (6FAPBS)
and [1,1'-Biphenyl]-3,3'-disulfonic acid, 4,4'-bis[4-amino-2-(trifluoromethyl)phenoxy]- (6FBABDS), and
 ACCEPTED MANUSCRIPT

then used them as comonomers with piperazine and TMC to produce NF membranes for treatment of
dye solutions. The flux of the membranes increased with increasing sulfonated monomer concentration;
6FBABDS showing higher flux than 6FAPBS, but rejection of rhodamine and methyl orange dye also
decreased at the same time. Akbari et al. [310] also prepared NF membranes for treatment of textile
wastewater by reacting different concentrations of 2,5-diaminobenzenesulfonic acid (2,5-DABSA) and
piperazine with TMC. At an optimal composition of 50% DABSA in the amine mixture, the resulting
membrane showed substantially greater flux, improved rejection of direct yellow 12 dye, and similar

PT
Na2SO4 passage compared to a PIP-TMC control membrane. The sulfonic acid groups on DABSA
monomer improved membrane hydrophilicity (reduced contact angle) and less fouling from a cationic
molecule was even observed. Jin et al. [311] added taurine, a hydrophilic compound in the aqueous

RI
phase with piperazine to improve antifouling properties. Taurine (2-aminoethane-1-sulfonic acid), has
both a hydrophilic charged sulfonic acid group and a reactive amine group. Water contact angle and zeta
potential measurements showed that surface hydrophilicity and surface negative charge increased with
increasing taurine concentration. MgSO4 passage increased with taurine concentration, and an optimum

SC
ratio of piperazine to taurine had to be maintained in order to obtain stable rejection and flux.
Membranes made with 0.2% taurine showed better antifouling resistance to BSA protein than the
control membrane.

U
Li et al. [312] made positively charged membranes by interfacial polymerization using TMC and the
monomer 1,4-Bis(3-aminopropyl)piperazine (DAPP). The isoelectric point was about pH 9.5, and
AN
passage of single salts varied with the cation: (e.g. LiCl > NaCl > MgSO4 > MgCl2). In mixed feeds formed
of LiCl and MgCl2 salts, negative rejection of Li+ ion was observed.
M
Jewrajka et al.[313] reported the use of a novel acid chloride, pyridine tricarboxylic acid chloride with or
without TMC to react with MPD for making membranes with antimicrobial property due to the presence
of the quaternary ammonium group generated via the quaternization of the pyridine nitrogen by an
alkyl halide. An et al. [314] reported an NF membrane containing zwitter ion for improved fouling
D

resistance wherein the amine phase contained zwitter ionic species derived from tertiary amines such as
2,6-diamino pyridine and sultone monomers such as 1,3- propane sultone.
TE

Amine-reactive monomers are also used to modify charge. Li et al. [315] formed NF membranes by
reacting piperazine (PIP) with 3,3,5,5-biphenyl tetraacyl chloride (mm-BTEC). With four acid chlorides
on the molecule, there is the possibility for both increased crosslinking and an increase in residual
EP

carboxylic acids (Figure 6). The increased flux and more negative zeta potential, as compared to
PIP+TMC membranes, demonstrated greater charge density. Rejection of Na2SO4 was improved also,
especially at low pH. (Other monomers with multiple acid chlorides will be discussed in more detail
C

later.) While not specifically aimed at NF, a number of amine-reactive small molecules having charged
groups have recently been mentioned in the patent literature and used to obtain either increased flux or
AC

greater rejection of NaCl for RO. Examples demonstrate charges on the membrane may be introduced
through carboxylic groups[316-320], phosphonylchlorides[321], and anhydrides.[318, 322]
 ACCEPTED MANUSCRIPT

PT
RI
Figure 6. Possible chemical structures of the mm-BTEC/PIP polymer chains. (A) linear polymer chain

SC
with two pendant COOH groups; (B) linear polymer chain with a pedant COOH group; (C) totally
cross-linked polymer chain.[315]

U
3.3.2 Monomers providing hydrophilic character
AN
Ethyloxy groups are well known hydrophilic groups. Based on this understanding, Jin et al. [323]
synthesized an ethyloxy based monomer, 2, 2'-oxybis-ethylamine (2,2'-OEL) and fabricated thin-film
M
composite nanofiltration membranes via interfacial polymerization between an amine mixture of
piperazine and 2,2-OEL and TMC. In contrast to the long chain PEG derivative, the OEL monomer has a
very short chain and thus can swell less, imparting more strength for the polyamide skin layer. They
D

reported an increase in flux and resistance to protein fouling with increasing 2,2-OEL monomer
concentration.
TE

The hydroxyl group is another well-known hydrophilic group, and it can also react interfacially with acid
chlorides to produce polyester based TFC membranes. In some cases, these polyester based
membranes have significantly higher flux than the corresponding polyamide based membranes and may
EP

be attractive where high salt rejections is not necessary. Also due to the lack of the amide bond, these
membranes can be more chlorine stable. Kwak et al.[324] studied various bisphenols substituted with
methyl and halogen groups and found that methyl group substitution increased flux and decreased
rejection, whereas the opposite trend was observed for halogen substitution. Kim et al. [300] reported
C

forming a high rejection (~97% NaCl rejection) polyesteramide membrane by reacting 4,4-
dihydroxybiphenyl and MPD with TMC. Jayarani et al. [325, 326] also studied various hydroxyl based
AC

monomers such as m-aminophenol (m-AP), bisphenol-A (Bis-A), and hydroquinone (HQ) for use as
comonomers with MPD to synthesize polyesteramine TFC membranes. Hydroquinone/MPD membranes
had similar NaCl rejection (95-98%) to control MPD-TMC membrane. The chlorine resistance of the
membranes was also improved. Razdan et al.[327] developed novel polyesteramide NF membranes for
potential application in diafiltration applications (dyes, antibiotics) where a high rejection of species with
MWCO > 400 Daltons is necessary, but a high passage of small molecules and salts is desired. They
found that sterically hindered diols, when added in conjunction with MPD, prevented close packing of
the polymer layers and made a more open structure than the control MPD-TMC membrane. They
altered the trade-off between water permeability and monovalent salt passage by selecting bulky group
diols: phenolphthalein and tetrabromo bisphenol-A. (Low flow and decreased salt passage resulted
 ACCEPTED MANUSCRIPT

from hydrophobic bromo groups on tetrabromo bisphenol-A.) More recently, Abu Seman et al. [328-
330] synthesized polyester based NF membranes by interfacially polymerizing bisphenol A based
monomers with trimesoyl chloride to improve antifouling property.

Tang et al. [331, 332] reported a polyester based NF membrane synthesized via interfacial
polymerization between triethanolamine (TEOA) and TMC. The effect of polymerization time, monomer
concentration, and pH of the aqueous solution etc. were investigated. The membranes in general

PT
showed lower MgSO4 rejection compared to commercial NF membranes. However, these membranes
may have potential application in acidic feed as the tertiary amine group can change to a quaternary
ammonium group at low pH and produce higher flux. In other reports [333, 334] from this group, -
cyclodextrin (-CD)/polyester thin film nanofiltration composite membranes were prepared via

RI
interfacial polymerization of TEOA with TMC in presence of -CD in the aqueous phase (Figure 7). A
study on the effects of the concentration and pendant group of -CD on membrane performance
determined an optimized concentration of 1.8% for -CD at which the membrane flux was almost

SC
double that of the base polyester membrane. Incorporation of sulfonic acid groups through sulfated -
CD monomer improved membrane rejection. The membranes also showed promising antifouling
properties.

U
AN
M
D
TE

Figure 7. (a) Schematic of the preparation procedure of -CD/polyester nanofiltration composite

membrane; (b) Interfacial polymerization between TMC and TEOA in the presence of -CD.[333]

Colquhoun et al. [335] reported novel hydrophilic NF membranes made with poly-ylids. These
EP

membranes are synthesized via interfacial polymerization between 1,1-diamino-4,4-bipyridinium di-

iodide and aromatic di- or tri-acyl chlorides. These materials are isoelectronic with polyesters but
contain polar ylid linkages [N+N-CO] instead of ester bonds [C-O-CO] and thus are hydrophilic. Poly-ylid
C

membranes made on Udel support showed a MWCO of 300 Daltons and moderately high salt rejection.

3.3.3 Monomer structure

AC

Specific monomers can also impact properties of the membrane through changes other than charge or
hydrophilicity. To explore different effects of monomer structure, a variety of comparative studies have
been performed using both alternate amine and amine-reactive monomers (e.g. acid chlorides)

In both academic and industrial environments, several studies have considered the functionality and
structure of different amines. In his first project report on NS-100, Cadotte provided performance data
for different aliphatic monomeric amines such as 1,6-hexanediamine, 1,3-propanediamine, 1,2-
ethanediamine and hydrazine [211]. A comparative study was subsequently done by Petersen and
 ACCEPTED MANUSCRIPT

coworkers [336] on 1,2-ethanediamine, 1,4-cyclohexanediamine , 1,3-bis (aminomethyl) cyclohexane,

and piperazine. All three membranes made from amines other than piperazine showed lower flux, but
higher NaCl rejection than that of made from piperazine. It was hypothesized that the interchain
hydrogen bonding by the amidic hydrogen of the amide groups prevented membrane swelling and
hence the number of amidic hydrogen dictates membrane flux. To their point, 1,2-ethanediamine with
the highest number of amidic hydrogen showed the lowest flux.

PT
Verissimo et al. [337] investigated the effect of the diamine structure on surface charge, structure and
performance, forming interfacially polymerized NF membranes by reacting TMC with piperazine (PIP),
N,N'-diaminopiperazine (DAP), 1,4-bis(3-aminopropyl)-piperazine (DAPP) or N-(2-aminoethyl)-piperazine
(EAP). They generally correlated the flux of the membranes with the hydrophobic/hydrophilic character

RI
of the monomers (characterized by the octanol-water partition coefficient), with the DAP-TMC
membrane showing the highest flux. However PIP-TMC membrane did not follow the trend, and they
ascribed this to a rougher surface structure created with piperazine (and its comparatively greater

SC
surface area). Surface charge experiments revealed that PIP and DAP based membranes were negatively
charged over the pH range examined, whereas DAPP and EAP membranes were amphoteric with
isoelectric points near pH 6.5 and pH 6, respectively.

U
Saha and Joshi [338] examined performance resulting from different combinations of amine and acid
chloride monomers. Amines included MPD, 3,5-diaminobenzoic acid, PIP and 2-aminoethyl piperazine
AN
(alone or in combination) and the acid chlorides included TMC or isophthaloyl chloride (alone or in
combination). Membranes prepared form 2-aminoethyl piperazine showed very low flux, as was also
reported by Verissimo et al. When considering amine mixture containing MPD and piperazine, it was
expectedly observed that rejection increased when the fraction of MPD increased in a MPD/PIP mixture.
M
However, they also observed that greatest flux was obtained with a 50/50 MPD/PIP composition.
Unreported studies at Dow have similarly observed a maximum in water permeability with addition of
MPD to PIP.
D

Li et al. [339] conducted a comparative study with aliphatic straight chain amines (diethylenetriamine,
TE

triethylenetetramine and tetraethylenepentamaine) and compared results to those obtained with the
piperazine molecule. Similar to Verissimos study, they found that the solubility of the amine monomer
in the organic phase controlled the structure and the performance of the membrane; the more soluble
amine favoring a membrane with greater permeation. According to their study, the amine with the
EP

longest chain length (tetraethylenepentamaine) showed the highest solubility in the organic phase and
also the highest flux and most hydrophilicity (due to a greater number of amino groups) among the
three straight chain amines. However, piperazine, the monomer with the highest solubility in the
organic phase, was still the best monomer in terms of producing the highest flux membrane.
C

McCray et al. prepared NF membranes with different aliphatic amines: p-xylylenedimaine[340] and
AC

tetrakis (N-methyl-amino-methyl) methane[341]. These were reacted (separately or in a mixture) of

isophthaloyl chloride and TMC to produce membranes with varying range of NaCl rejection and flux.
The membranes also showed good chlorine resistance. Tomaschke [342] reported synthesis of
bipiperidine based NF membranes in presence of monomeric amine salts. The membranes showed
higher flux than the piperazine based membranes and good MgSO4 rejection. Chen et al. [343] used 1,4-
diaminocyclohexane (DCH) as the amine monomer and sodium N-cyclohexylsulfamate (SCHS) to
prepare NF membranes with improved flux and Na2SO4 rejection. Tomaschke also prepared membranes
with performance range from RO to NF with benzoyl piperazine based monomers (such as 3,5-
diaminobenzoyl piperazine) which showed higher flux , but lower NaCl rejection than mPD based
membranes.[344]
 ACCEPTED MANUSCRIPT

Chen et al. [345] reacted a macrocyclic polyamine (cyclen) with TMC to fabricate NF membrane. In
comparison to flux values reported for several other aqueous monomers, authors found high flux and
Na2SO4 rejection which they attributed to the multiple amino groups on the macrocycle. They also
incorporated silica nanoparticles in the aqueous phase containing cylcen and reported obtaining even
higher flux.

In the drug industry many drugs are chiral and they are present in different isomeric forms which can

PT
differ in their properties. Hence there is a desire to separate optically pure isomers of chiral drugs to
maintain the desired effects. Singh et al.[346-348] reported a novel enantioselective TFC NF membrane
where the chiral selective layer was formed through interfacial polymerization between L-lysine,
piperazine and TMC and different monomer concentration ratios were investigated. Membranes made

RI
with equal ratio of piperazine and L-lysine showed the highest enantioselectivity and permeated D-
enantiomer of lysine monohydrochloride preferentially whereas membranes made with piperazine only
did not show any enantioseparation.

SC
The impact on performance of acid chloride structure has also been considered in several membrane
studies. Qian et al. reported [349] making microporous NF membranes made on PAN support through
interfacial polymerization of piperazine and acid chlorides containing tetrahedral carbon or silicon cores:

U
(Tetrakis(4-carboxyphenyl)methane (TPMC) or tetrakis(4-carboxyphenyl)silane (TPSC) molecules). Due
to the introduction of these rigid molecules into the polyamide network, the membrane had higher free
AN
volume and microporosity than that of the counterpart NF membranes made from piperazine and TMC.
As a result, these membranes showed much higher flux than TMC-PIP membranes.

Li et al. [350] have extensively studied biphenyl acid chlorides with varying functionality for both RO and
M
NF applications. RO membranes made from the interfacial polymerization of MPD with tri- and tetra-
functional biphenyl acid chlorides: 3,4,5-biphenyl triacyl chloride (BTRC) and 3,3,5,5-biphenyl tetraacyl
chloride (BTEC). BTRC and BTEC both showed higher rejection of NaCl (but lower flux) than controls
D

(MPD+TMC ), presumably due to the higher crosslinking and rigid structures of the biphenyls. Further
studies [351] were carried out with three isomeric tetrafunctional biphenyl acid chlorides 3,3',5,5'-
TE

biphenyl tetraacyl chloride (mm-BTEC), 2,2',4,4'-biphenyl tetraacyl chloride (om-BTEC), and 2,2',5,5'-
biphenyl tetraacyl chloride (op-BTEC) as shown in Figure 8. Membranes prepared from mm-BTEC
showed highest rejection as mm-BTEC was more reactive and could form a better crosslinked network
structure than the sterically hindered and less reactive 2,2-substituted acyl chlorides om-BTEC and op-
EP

BTEC.
C
AC

Figure 8. Structures of three isomeric tetrafunctional acid chlorides[351]

These studies were extended for NF membranes [315] by preparing membranes via interfacial
polymerization of mm-BTEC with piperazine in water. In this case, an effect opposite that with RO
membrane (MPD) was observed i.e. membranes made from mm-BTEC/PIP showed higher flux and lower
 ACCEPTED MANUSCRIPT

rejection than control membranes made form TMC/PIP. The observed differences may principally stem
from the lower partition coefficient for piperazine in the oil phase, so that a slower rate of piperazine
diffusing into the oil phase limits extent of reaction and a more charged and swollen membrane results.
This is consistent with zeta potential and XPS studies that showed higher concentrations of COOH
groups for membranes made with mm-BTEC, as compared to TMC based membranes, suggesting fewer
acid chloride groups participated in the interfacial polymerization reaction.

PT
In 2011, Wang et al. [352] reported making a positively charged NF membrane from the same monomer
mm-BTEC by changing the organic solvent from cyclohexane to toluene. They showed that the solubility
and diffusivity of piperazine in toluene were higher than that in cyclohexane, allowing piperazine
monomer to diffuse more into the organic layer and to participate in greater extent in the interfacial

RI
polymerization reaction. This resulted in a positively charged membrane as evident from the higher
number of amine groups obtained by zeta potential and XPS experiments. In 2013, Wang et al. reported
[353] the use of a hexafunctional 2,2',4,4',6,6'-biphenyl hexaacyl chloride (BHAC) monomer with

SC
piperazine on PAN support to make a positively charged NF membrane. The rigid and contorted
structure of BHAC led to higher free volume membranes with higher flux than the membranes made
with BTEC monomers, but without compromising salt rejection.

U
3.3.4 Monomers for chlorine resistance

AN
A recognized weaknesses of polyamide based membranes is the lack of chlorine stability. Using a
variety of model compounds and membranes, substantial work by Glater and Zachariah[354-357],
Kawaguchi and Tamura[358], Soice [359] and others have shown that reversible chlorination occurs at
M
the amidic hydrogen of the amide bonds which then progresses to irreversible ring chlorination via
Ortons rearrangement or sometimes direct ring chlorination is also possible depending on the pH of the
solution and the concentration of chlorine. However, several different mechanisms have also been
D

proposed for the membrane performance loss due to chlorine exposure. Glater [355] suggested that
membrane flux decreases when chlorine substitution tightens the membrane and flux increases due to
TE

the cleavage of the amide bonds. Soice [360] used pendent drop experiments to examine mechanical
properties of MPD-based membrane and suggested that the major cause is not bond cleavage, rather
the physical separation of the polyamide skin layer from the polysulfone support. Kwon [361] examined
an MPD membrane by FT-IR and ascribed performance changes to loss of interchain hydrogen bonding
EP

as a result of the chlorination of amidic hydrogen. Mechanical analysis based on induced wrinkling
recently demonstrated embrittlement of MPD membrane, but enhanced chain flexibility on piperazine-
based NF270.[362] Hence, significant research has been done that influences monomer selection for
preparation of a chlorine resistant membrane. Some of the areas heavily explored are the following:
C

monomers containing secondary or tertiary amines so that the amide bond does not have any amidic
hydrogen, aliphatic monomers to avoid aromatic ring chlorination, and electronic withdrawing group
AC

substitution on aromatic amines or acid chlorides to prevent aromatic ring chlorination.

NF membranes are primarily based on piperazine, which is a heterocyclic aliphatic secondary diamine.
Membranes from piperazine would show better chlorine resistance than RO membranes formed from
its aromatic counterpart, MPD. An early study by Parrini [363] found that a piperazinamide-based
membrane actually had reduced impact from chlorine resistance than a commercial cellulose acetate
membrane. Studies [364] on other aliphatic amines such as diethylenetriamine and ethylenediamines
also demonstrated superior chlorine resistance compared to MPD-based FT-30 membranes.
Kawaguchi and Tamura [358] examined different diamines and found that tertiary polyamides made
from piperazine and isophthaloyl chloride did not show any oxidation by chlorine whereas reversible N-
 ACCEPTED MANUSCRIPT

chlorination at the amide bond occurred with aliphatic primary diamines and irreversible ring
chlorination occurred with aromatic primary diamines.

Konagaya and Watanabe [365] compared chlorine resistance of polyamides made from isophthaloyl
chloride with aliphatic, cycloaliphatic and aromatic diamines. Similar to the findings by Kawaguchi and
Tamura, they also concluded that polyamides based on piperazine and isophthaloyl chloride had the
highest chlorine resistance of those examined. They found that aliphatic or cycloaliphatic diamines with

PT
secondary amine groups or a short methylene chain between end groups were desirable for chlorine
resistance, and the chlorine resistance of aromatic diamines could be improved when the ortho position
to the amine group is substituted by a Cl or CH3 group. In another paper [366] they reported that
asymmetric membranes prepared from a copolymer of mixed diamines, 4,4-diaminodiphenylsulfone

RI
and piperazine, with isophthaloyl chloride showed stable performance after exposure to 50 ppm of
chlorine for 100 hours at pH 5, whereas Nomex type aromatic polyamide rejection started to drop
within 10 hours under similar conditions. It was concluded that the electron withdrawing group on the

SC
aromatic diamine helped to increase chlorine resistance of the copolyamide.

Shintani et al. [367] studied 17 kinds of diamines and 2 kinds of acid chlorides to understand the effect
of monomer structure on the chlorine resistance of resulting polyamides. Observing changes in

U
molecular weight of polyamides following exposure to chlorine, they also noted that aromatic
polyamides were less chlorine stable than those from aliphatic and cycloaliphatic amines. They
AN
determined that polyamides from diamines with amino groups at the ortho position showed better
chlorine resistance than that of polyamides made from monomers with amino groups at m- or p-
positions. This was ascribed to the increased steric resistance for chlorine attack at the amide bond at
the ortho position. A membrane prepared from N,N-dimethyl-m-phenylenediamine (N,N-DMMPD)
M
and TMC showed very high chlorine resistance compared to commercial membrane control based on
MPD. However, the experimental membrane showed lower salt rejection than the conventional
polyamide RO membranes. In a subsequent study [368] they synthesized an NF membrane by reacting
D

N-phenylethylenediamine (N-PED) with TMC. This particular polyamide had properties suitable for NF
application and sufficient chlorine stability, though the chlorine resistance was lower than that of
TE

membranes based on N,N-DMMPD. Laboratory and field test data confirmed higher chlorine resistance
of the experimental membranes compared to a commercial RO membrane (MPD+TMC/IPC).

Buch et al.[369] prepared an NF membrane by reacting TMC with one of the diamines studied by
EP

Shintani, the cycloaliphatic diamine 1,3-cyclohexanebis (methylamine). They concluded that N-

chlorination of the polyamide skin layer is significant to adversely affect membrane performance, even
in the absence of ring chlorination. Yu et al. [370] combined the effect of a methyl group substituent on
aromatic ring and a cycloaliphatic acid chloride to prepare RO membranes from m-phenylenediamine-4-
C

methyl (MMPD) and cyclohexane-1,3,5-tricarbonyl chloride (HTC). A resulting RO membrane

demonstrated more stable operation after soaking in pH 8 chlorine solutions, but also had greater initial
AC

salt passage than the control (MPD+TMC) membrane.

La et al.[371] prepared a fluoropolyamide based chlorine resistant RO membrane by reacting TMC with
a hexafluoroalcohol containing the diamine, 2,2'-bis(1-hydroxyl-1-trifluoromethyl-2,2,2-trifluoroethyl)-
4,4'-methylenedianiline (BHTTM). The hexafluoroalcohol group, by virtue of both its electron
withdrawing capacity and bulkiness, can offer chlorine resistance to the resulting polyamide. Hu et al.
[372, 373] also reported the use of this monomer to prepare a novel silica-fluoropolyamide based
chlorine resistant NF membrane. They demonstrated that flux could be increased by incorporating a
small amount of silica in the otherwise low-flux fluoropolyamide layer, and that rejection of the
membrane was not negatively impacted at sufficiently low silica concentrations. They also found both
 ACCEPTED MANUSCRIPT

improved flux and salt passage for this membrane could be obtained by soaking in 5000 ppm-hours
chlorine, noting that a similar treatment had previously been shown [374] to similarly improve RO
performance.

Hong et al.[375] interfacially reacted a tetrafunctional monomer 1,2,4,5-benzene tetracarbonyl chloride

(BTC) with MPD to produce a poly(amic acid) based membrane. The acid groups were then thermally
imidized in presence of triethylamine (TEA) or N,N,N, M- tetramethyl hexane diamine (TMHD) catalysts

PT
to produce a chlorine resistant polyimide RO membrane. Zhang et al. [376] utilized the same
tetrafunctional BTC monomer and instead of thermally imidizing the carboxylic acid groups, ionically or
covalently bonded the acid groups with ethylene diamine (EDA) to produce a chlorine resistant
polyamide based NF membrane. The membranes were made on a polyetherimide support by reacting

RI
BTC in hexane with MPD in the aqueous phase. Next the membranes were either post treated with an
aqueous solution of EDA to ionically bond with carboxylic acids or post treated with 1-ethyl-3-(3-
dimethylaminopropyl) carbodiimide hydrochloride (EDC)/N-hydroxysuccinimide (NHS) to activate the

SC
acid groups followed by a post treatment with aqueous EDA to covalently bond the amine groups with
COOH groups. Membranes made with covalently bonded EDA particularly showed both chlorine
resistance and high rejection of low molecular weight organics, suggesting their potential applicability
for intended pharmaceutical applications.

U
Motivated by the chlorine resistance attribute of the asymmetric sulfonated biphenol based polyarylene
AN
ether sulfone RO membranes[377-380], Xie et al.[381] synthesized a sulfonic acid group containing
diamine monomer, disulfonated bis[4-(3-aminophenoxy)phenyl]sulfone (S-BAPS) and reacted it with
TMC for preparing thin film composite sulfonated polysulfone based RO membranes. However, the
membrane showed much lower NaCl rejection compared to conventional RO membranes. It also
M
demonstrated lower chlorine resistance than a control MPD-TMC membrane because even though, the
membrane was based on sulfonated polyarylene ether sulfone, it still contained the amide group, which
is susceptible for chlorine attack and also the membrane was less crosslinked than a control MPD-TMC
D

membrane. Kim et al.[382] used same kind of sulfonated diamine based monomer synthesized via direct
step polymerization and reported good NaCl rejection and higher chlorine resistance than the control
TE

MPD-TMC membrane.

Zhang et al. [383] formed an NF membrane by reaction of TMC with Bis-2, 6-N, N-(2-hydroxyethyl)
diaminotoluene to form a crosslinked polyamide. Hydroxyl groups on this amine monomer can further
EP

react with TMC to create esters that contribute to crosslinking. Esters, being electron withdrawing
groups, can also provide a barrier to aromatic ring chlorination. A methyl group at the ortho position
also increases chlorine resistance. The membrane was subjected to 10,000 ppm-hours of NaOCl (up to
50 hours using 200 ppm NaOCl, pH not specified) and was described as having high chlorine resistance.
C

However, performance properties changed substantially over this treatment range.

AC

In addition to the monomer route, substantial work has been done on coatings to improve chlorine
resistance. As shown in Figure 9a, Wei et al.[384] coated the membranes with 3-monomethylol-5,5-
dimethylhydantoin (MDMH) after the initial interfacial polymerization to react with residual acid
chlorides before the hydrolysis step. The grafted N-H group from the MDMH unit acted as a sacrificial
pendant group under chlorine exposure (Figure 9b). The coated membrane showed little change in flux
and salt rejection after chlorination with 1002000 ppm-hours chlorine at pH 4, but demonstrated lower
salt rejection overall compared to conventional PA membranes, as the membrane had fewer COOH acid
groups due to the reaction of acid chloride and MDMH.

a)
 ACCEPTED MANUSCRIPT

PT
RI
b)

U SC
AN
Figure 9. a) Reactive coating of MDMH on membrane surface via reaction with residual acid chlorides;
M
b) Reversible chlorination reaction at the pendant N-H group of MDMH[384]
D

They also grafted [385] 3-allyl-5,5-dimethylhydantoin (ADMH) by free radical graft copolymerization
using 2,2-azobis(isobutyramidine) dihydrochloride as an initiator on a commercial RO membrane. Here
TE

also the hypothesis was that the N-H groups in the coated layer will act as sacrificial groups under
chlorine exposure. The grafted membrane showed improved chlorine resistance, but suffered from flux
loss due to the addition of the coating layer. Similarly Liu et al. [386] coated commercial RO membranes
EP

by hydrophilic copolymers poly(N-isopropylacrylamide-co-acrylamide) (P(NIPAM-co-Am)) to use the

coating layer as a sacrificial layer. They intended to increase the resistance to hydrolysis and
chlorination by the intermolecular hydrogen bonding of the barrier layer and also to avoid flux loss due
to an additional coating layer by choosing a hydrophilic coating. Shin et al.[387] coated commercial
C

polyamide membranes with silanes via a sol-gel method so that the silane can act as a protective layer
for chlorine exposure. Wu et al. [388]improved chlorine resistance of an NF membrane (PEI+TMC) by
AC

forming thereon one or more subsequent interfacially polymerized layers of crosslinked PIP+TMC layer.
Cheng et al.[389] developed an NF membrane with dual resistance to chlorine and fouling for
application in antibiotics by interfacially polymerizing an amine functionalized PEG with TMC.

3.3.5 Monomers for extending pH range

There are many applications that could benefit from membranes with better pH stability. Examples
include separation of metals from acidic process streams, sulfuric acid concentration, treatment of
effluent streams from pulp and paper and from the textile industry, and daily cleanings in sanitary
 ACCEPTED MANUSCRIPT

applications (e.g. dairy). Early parts of this review have described efforts to address these needs using
phase inverted membranes or coated supports. However, other research has also been directed to
improving performance of interfacially polymerized membranes, especially in selecting monomers more
amenable for stable operation with pH extremes.

Polysulfonamides exhibit more pH stability than polyamides due to greater stability of the sulfonyl bond
towards oxygen compared to that of the carbonyl bond [390]. However it is often more difficult to

PT
synthesize polysulfonamides due to the weaker reactivity of sulfonyl chloride compared to acid
chloride[391]. Naphthalene based sulfonyl chlorides such as naphthalene-1,3,6-trisulfonylchloride
(NTSC) have been a favorite choice for researchers [392-396]. NTSC was used either as the sole
monomer or in conjunction with TMC in the organic phase for interfacial polymerization with piperazine.

RI
Membranes made from NTSC often showed minimal change in performance after prolonged exposure
to low pH (e.g. soaking in 8 w/v% sulfuric acid for 30 days [393]) compared to the same exposure for a
TMC based membrane.

SC
Lee et al. [397] prepared a membrane with improved pH stability by interfacially polymerizing
polyethyleneimine with cyanuric acid chloride. The membranes maintained salt rejection when exposed
at high pH, whereas comparative membranes made with polyethyleneimine and TMC suffered loss in

U
salt rejection with high pH exposure. Xue et al. [398] also reported the use of cyanuric chloride and/or
phosphonitrilic chloride trimer as an oil phase monomer for synthesizing an NF membrane.
AN
Several papers have compared the tolerance of different commercial interfacially polymerized NF
membranes at pH extremes.[392, 399-401] Dow has tighter cleaning guidelines for piperazine-based
membranes than those membranes dominated by MPD (including NF90). It has often been observed
M
that piperazine-based membranes were less stable to pH extremes than those membranes based on
MPD.[401, 402] However, the exact role of the monomer mix would be an apt area for fundamental
study, as various factors have yet to be de-convoluted. For example, performance degradation at pH
D

extremes might be impacted by various factors (e.g. modification of hydrogen bonds, swelling,
hydrolysis, or even delamination of the barrier layer from the support), and monomers will impact only
TE

some issues. It was also interesting that Platt [400] found that the relative stability between NF
membranes depended even on the type of acid, whether nitric or sulfuric.

For some users, pH stability is less about issues of chemical degradation than just having a consistent
EP

performance over time and/or across a range of feed pH values. For instance, Mnttri [403] noted that
when a membrane has dissociable groups, it may likely swell during cleanings. While they recognize the
effect is typically reversible, they find performance can take longer than several hours or days thereafter
to re-equilibrate. (It may be useful to note that a slow return to equilibrium can also be observed with
C

NF270 in switching from a calcium to a sodium-based feed, presumably due to the replacement of
calcium ions complexed with carboxylic acids.) Other researchers have also considered reversibility of
AC

performance following cleaning.[404]

Operating pH also impacts the equilibrated membrane performance of interfacially polymerized NF

membranes due to disassociated charged groups and swelling.[405, 406]). For instance, NF270 has an
isoelectric point at about pH 3, and a corresponding minimum in rejection for several charged solutes
has been seen [406-408] Some papers mention that their new membrane would be potentially
advantageous at low pH.[409-411]
 ACCEPTED MANUSCRIPT

3.3.6 Monomers for solvent stability

Solvent stable membranes is a growing area of importance in nanofiltration. Fabrication of various

membranes directed to this need has been described in previous sections and in recent review
articles.[12-14, 43] However, many commercial membranes intended for use with solvents suffer from
relatively low flow. Also, with interfacially polymerized membranes, it is often the case that stability is
principally limited by conventional supports (e.g. PS, PES) and not by the barrier layer. Hence, there is

PT
opportunity for improved interfacially polymerized thin film composite membranes, tuning each
component to achieve the desired properties.

Several papers have been published that explore various combinations of supports and interfacially

RI
polymerized NF barrier layers. Kim et al. [412] conducted a thorough study by preparing membranes
from a mixture with varying ratios of MPD and piperazine, reacted with TMC at different concentrations
on PAN support. The membranes were stable in alcohols, ketones and hexane but showed no flux for

SC
the nonpolar solvent hexane, poor flux for EtOH and IPA and reasonable flux for MeOH. Livingston et al.
[413, 414] reported the synthesis of NF membranes on a stabilized support (polyimide (P84) crosslinked
with hexadiamine) by reacting imbibed amines (PIP, MPD, or hexane diamine) with TMC. It was found
that flux could be increased by first imbibing PEG400 into the support before reaction, potentially

U
impacting polymerization by enhancing amine uptake, modifying diffusion of amine within pores during
the polymerization, or preventing formation of membrane within the pores. Alternatively, it was found
AN
even more effective to post-treat the membrane after interfacial polymerization with an activating
solvent (DMF or DMSO). They speculated that these solvents swell the polyamide and can wash away
the low molecular weight fragments of the polyamide which decrease flux but dont substantially
M
contribute to rejection. With this post-treatment, membranes were formed that exceeded performance
of commercial solvent resistant membranes (higher permeability and lower MWCO for polystyrene
oligomers). Jimenez-Solomon et al. [253] explored improving the permeance to non-polar solvents (e.g.
hexane, heptane) by changing both the support and barrier layer. Using both crosslinked polyimide and
D

PEEK supports, NF membranes were prepared (MPD+TMC) and some were post-reacted with
pentafluoropropylamine to create a more hydrophobic surface. A related study [263] made
TE

hydrophobic NF membranes using pre- and post-treatments described previously and also examined a
mixture of TMC and a monoacyl chloride containing fluorine. They found that flux could be increased by
incorporating hydrophobic groups like F or Si in the membrane, either by grafting onto membrane
surface or through incorporating hydrophobic monomers during interfacial polymerization. Zhang et
EP

al.[415] reacted polyethyleneimine with a mixture of hydroxyl terminated trifluoride PDMS-TMC on a

crosslinked PAN support to prepare a hybrid hydrophilic-hydrophobic NF membrane. Hermans et al.
[264]utilized a crosslinked polyimide support for making solvent resistant NF membrane and simplified
C

the process by adding diamines in the coagulation bath. Minhas et al. [416] employed a polypropylene
support (Celgard 2400) for synthesis of solvent resistant NF membranes by interfacially polymerizing
AC

ethylene diamine and terephthaloyl chloride. Other groups have also utilized polypropylene support for
preparing solvent resistant membranes by interfacially polymerizing polyetherimide and isophthaloyl
chloride on top of it [275, 417]. Perez-Manriquez et al.[266] reacted N,N-diaminopiperazine with TMC
on a chemically crosslinked PAN support to produce a solvent stable NF membrane.

3.4 Polymeric reagents

In addition to monomeric reagents, interfacial polymerization can alternatively employ a polymer as one
of the two reactants. It was previously mentioned that the ground-breaking NS-100 membrane was
 ACCEPTED MANUSCRIPT

formed by reaction of polyethyleneimine (PEI) and toluene-di-isocyanate (TDC).[211, 213, 418] Other
early studies also involved exploration of different polymeric amines (e.g. polyepiamines [215-217],
polyvinyl amines [218], poly(diallyl amines)[219] and polyvinylimidazolines [220-224]). However, most
current commercial RO and NF membranes are now based on small monomers like PIP and MPD.
Nonetheless, particularly in the last five years, it seems that there has been resurgence in the
exploration of the role that polymeric reagents can play in interfacial polymerized NF membranes.

PT
PEI is a cationic polyelectrolyte containing primary and secondary amine groups which can be used in
place of the piperazine monomer. In contrast to a negatively charged NF membrane (e.g. PIP+TMC), the
resulting PEI-TMC membranes are positively charged at neutral pH. This difference enables membranes
to selectively repel divalent cations on the basis of both size and charge. Hence, for low pressure water

RI
softening, positively charged NF membranes may enable a less crosslinked network structure with
higher water flux to efficiently be used. Positively charged membranes can also have applications in the
separation of positively charged drugs or heavy metal ion removal from waste water.

SC
Fang et al. [419] developed a PEI+TMC based hollow fiber NF membrane for such a targeted low
pressure water softening application. They found that a medium molecular weight (50,000-100,000
g/mole) PEI with sodium dodecyl sulfate surfactant in the aqueous phase and a high pH are needed to

U
produce a membrane with optimized performance. Wei et al. [420, 421] also reported the use of PEI for
making a positively charged hollow fiber NF membrane. They started with a 70,000 g/mole PEI and
AN
optimized the reaction conditions by studying the effects of PEI and TMC concentration, amine wetting
time, reaction time, and the effect of an acid scavenger, Na3PO4. From the study, they concluded that
no acid scavenger is needed for PEI-TMC reaction since PEI itself can offer excess amine groups which
can absorb the acid by product.
M
Sun et al. [422, 423] investigated PEI as the aqueous reactant for making an interfacially polymerized NF
membrane on dual-layer hollow fiber membranes. In an earlier report [424] they had studied the effect
D

of the molecular weight of PEI when used as a crosslinker for the surface modification of a single layer
polyamide-imide based hollow fiber membrane. However, the crosslinked UF membrane could not
TE

offer high flux or solute rejection as generally observed for an interfacially polymerized NF membrane.
Hence they tried to implement PEI for an interfacially polymerized membrane by reacting it with
isophthaloyl chloride. The support itself offered a great advantage over the normal PES, PP or PAN
supports as the carbonyl groups of the imide rings of the polyamide-imide support can react with the
EP

amine groups of PEI and thus can form a stabilized layer of PEI. As expected, flux and salt passage
decreased with increasing molecular weight of PEI, and an optimum concentration range was observed
for a fixed molecular weight PEI. Interestingly, these membranes have a positively charged barrier layer
and a negatively charged support layer, and authors describe this dual layer as suitable for rejecting
C

both negatively and positively charged dye molecules.

AC

As PEI is a macromolecule, it is not as highly reactive as the monomeric piperazine. In general, the
membranes formed from PEI have a loose structure with large pores and a thick dense layer. Wu et
al.[425] made a layer-by-layer TFC membrane from PEI and TMC to have a better control on the
reactivity of PEI and thus the ability to make a controlled membrane network structure. PEI and TMC
can be applied on the porous support by first coating with PEI and then TMC or vice-versa. They found
out that the deposition sequence of PEI and TMC is important to control membrane morphology and
performance. Membranes made with PEI-TMC deposition order were more permeable than membranes
made with TMC-PEI deposition order and the former also had a more evenly distributed valley-ridge
morphology compared to the latter. Increasing the number of deposition cycles or reactant
concentrations improved salt rejection, but lowered the flux of the membranes.
 ACCEPTED MANUSCRIPT

Wu et al. [426] made an NF membrane by reacting TMC with a mixture of PEI and monomeric amine PIP
in the aqueous phase. While PEI membranes might typically have less crosslinked structure, the
addition of the more highly reactive monomeric amine enabled control of the crosslinking density and
enhanced membrane stability. They also studied the effect of the order of interfacial polymerization on
membrane performance by making 2-ply membranes containing two layers with PIP-TMC and PEI-TMC.
Figure 10 shows the process of making a 2-ply membrane where in first cycle PEI reacts with TMC and
then in second cycle the residual acid chlorides from TMC react with PIP. They observed that for a single

PT
ply membrane, addition of a small amount of PIP in the PEI increases membrane flux without adversely
affecting salt rejection. The performance of the 2-ply membranes depended on the PIP/PEI ratio. By
optimizing this ratio, 2- ply membranes with better flux and rejection than singly ply membranes made

RI
with the blend of PIP and PEI could be obtained. Among the 2-ply membranes, again an effect of the
PIP/PEI ratio was observed, showing higher flux for (PIP/TMC-PEI/TMC) membrane than (PEI/TMC
PIP/TMC) membrane at a low PIP/PEI ratio and the opposite at a higher PIP/PEI ratio.

U SC
AN
M
D

Figure 10. Process of making a 2-ply PEI/TMC-PIP/TMC by 2 cycles of interfacial polymerization with
PEI and PIP respectively [426]
TE

Chiang et al. [427] reported a comparison study between ethylene diamine (EDA), diethylenetriamine
EP

(DETA), and hyperbranched PEI. Membranes were formed by reaction with TMC and also compared to a
fourth membrane made by reaction of hyperbranched PEI and TPC. They measured zeta potential,
thickness, and rejection of PEGs and different salts (at pH 3-6). Among observations, they found that
the PEI+TPC and EDA+TMC membranes had the same MWCO, but the PEI membrane had higher flux
C

and greater salt rejection. They attributed this to flexible pendant charged amine groups that drifted
inside of pores and hindered ion transport but had less effect on water molecules.
AC

Polyvinyl amines contain vinyl amine and acryl amide units which can take part in the interfacial
polymerization reaction with acid chlorides. Yu et al.[411] and Liu et al. [428] fabricated NF membranes
by interfacially polymerizing polyvinylamines or sericines with TMC or isophthaloyl chloride or 5-
isocyanate-isophrhaloyl chloride or 5-chloroformyloxyisophthaloyl chloride. Due to high water
permeability and positive charge at low pH, Yu had suggested that these membranes may be particularly
suitable for treating acidic feeds and antifouling applications.

Polyethylene glycols or PEGs and their derivatives are very popular hydrophilic polymeric reagents due
to their excellent antifouling properties. There are several reports [429-432] where PEG and its
 ACCEPTED MANUSCRIPT

derivatives have been used as coatings on polyamide membrane surface after the polymerization is
completed. In a few studies mentioned below, PEG-derivatives have also been used as a monomer for
interfacial polymerization.

Gol et al. [433] reported in-situ PEGylation of NF membranes by interfacial polymerization of amine
mixtures containing piperazine and amine terminated PEGs with TMC. Figure 11 shows the comparative
study between piperazine terminated PEG, MPD terminated PEG and alkyl amine terminated PEG.

PT
Membranes synthesized from piperazine terminated PEG showed almost similar membrane
performance as the conventional NF membranes, but with improved antifouling resistance. Membranes
made from the MPD terminated PEGs showed better fouling resistance but their flux and salt rejection
were worse than the conventional membranes. Authors attributed the different impacts on flux and

RI
rejection to the reactivity differences between MPD-terminated PEG and piperazine as demonstrated by
the trend of global nucleophilicity of the amines towards TMC in Figure 11. This is a novel route to
incorporate antifouling macromolecules in the polymer network during the interfacial polymerization

SC
process.

U
AN
M
D
TE

Figure 11. PEGylated NF membranes made with different amine terminated PEGs and the
nucleophilicity of the amines toward TMC [433]
EP

Cheng et al.[434, 435] reported making a hydrophilic NF membrane by reacting TMC with amine
terminated PEG molecules. Various polymerization conditions, such as monomer concentration and
reaction times, were investigated. These membranes had higher flux and are hypothesized to have
antifouling character. Sfora et al.[436] reported the synthesis of hydrophilic NF membranes by
C

interfacial polymerization between TMC and different monomeric and oligomeric amines located on or
inside a poly(ethylene oxide-b-amide) (PEBAX) layer. The PEBAX layer was added on top of a PVDF
AC

support. Ethylene glycol based amines such as poly(propylene glycol-b-ethyleneglycol-b-propylene

glycol) bis(2-amine propyl ether) (PPGEG) , jeffamine, and MPD were investigated. The ethylene glycol
based amines particularly demonstrated higher flux with a relatively low molecular weight cut-off. Kang
et al. [437] coated a MPD-TMC membrane with aminopolyethylene glycol monomethylether (MPEG-
NH2) monomer before the hydrolysis step so that the amine groups could react with unreacted acid
chloride groups and the PEG layer is thus chemically bound onto the membrane surface. The membrane
showed good antifouling properties.

Other hydrophilic polymeric reagents used in preparation of interfacially polymerized membranes

include tannic acid, sericin, polydopamines etc. Zhang et al. [438] reported a polyester based NF
 ACCEPTED MANUSCRIPT

membrane synthesized from the interfacial polymerization between a natural polymer, tannic acid, and
TMC. Tannic acid is a polyphenol and can be synthesized from glucose and gallic acid. The membrane
showed very good antifouling properties because of the presence of hydroxyl groups. It also
demonstrated better chlorine stability than polyamide based NF membranes. Zhou et al. [439] studied
another natural hydrophilic polymer, sericin, which contains hydroxyl, carboxyl, and amino groups for
making NF membranes by interfacially polymerizing with TMC, and the membranes showed favorable
antifouling properties.

PT
It has been previously mentioned how coatings comprising dopamine have been used to form
hydrophilic membranes on UF supports.[123-126] Zhao et al. [440] has also used dopamine as an
aqueous phase monomer to react with TMC to make an interfacially polymerized NF membrane. In this

RI
study, some portion of the dopamine was self-polymerized as a layer on the polyethersulfone UF layer;
interfacial polymerization with TMC occurred with the amine and hydroxyl groups of the rest of the
dopamine monomer and also with the catechol groups of the polydopamine. Dopamine concentration,

SC
interfacial polymerization time, and reaction temperature had significant effect on membrane
performance, and the membranes showed promising alcohol stability due to the adhesion property of
polydopamine. They also claimed better chlorine tolerance for dopamine-TMC membrane than PIP-TMC
membranes due to the ester bonds formed with the hydroxyl groups of dopamine with TMC. However,

U
there was still a high flux increase and fairly moderate dye rejection drop (12%) with chlorine exposure.

AN
Dendrimers have an ordered hyper-branched structure with a large number of terminal functional
groups and thus have drawn attention as a highly reactive macromolecule in many applications such as
drug delivery, catalysis and also in membrane synthesis. Zwijnenburg et al. [441, 442] reported the use
of a poly(propylene imine) (PPI) dendrimer for synthesizing NF membrane by reacting with TMC.
M
However, the synthesis of PPI requires high pressure and suffers from high material cost and low yield. Li
et al. [410, 443] used a common dendrimer polyamidoamine (PAMAM) shown in Figure 12, which is
relatively easier to synthesize, but similar in structure and property with PPI for making NF membranes
D

on a poly (ether ether ketone-cardo) (PEK-C) support. The resulting PAMAM-TMC membrane was a
positively charged membrane with potential application for treating acidic feeds. Jin et al. [444-447] also
TE

has reported the synthesis of PAMAM-TMC and PAMAM-TMC- SiO2 NF membranes on polysulfone
support.
C EP
AC

Figure 12. Structure of PAMAM G1.0 dendrimer [410]

While not as ordered as dendrimers, hyperbranched polymers (HBPs) have a more facile synthesis
compared to the multistep synthesis need for dendrimers. Like dendrimers they have hyper branched
structure and a large number of functional groups. Wei et al. [448, 449] used a hydroxyl functional
aliphatic hyperbranched polyester (HPE) in interfacially polymerizing with TMC or terephthaloyl
chloride. A study of HPE-TMC morphology revealed that membrane surface roughness increased with
increasing HPE concentration. Membrane flux decreased and salt rejection increased with increasing
 ACCEPTED MANUSCRIPT

TMC concentration due to formation of a more crosslinked network. A later study [450] by this group
investigated the impacts of HPE molecule structure and molecular weight, and found that membrane
flux and salt rejection increased with increasing molecular weight of HPEs.

3.5 Additives

PT
To this point in the paper, discussion of the interfacial polymerization reaction has focused on reactive
entities. However, a wide variety of other species may be added to the primary reaction, and these can
be essential to making an optimum membrane and controlling important properties. The different
components may be added to either the aqueous or organic phase (or both).

RI
3.5.1 Surfactants & phase transfer catalysts

SC
Surfactants and phase transfer catalysts are commonly used additives for interfacial polymerization. A
primarily and unique role that surfactants play in the aqueous phase is to help wet the UF support
material of a TFC membrane, which can help in uniform amine impregnation. Surfactants can also

U
influence reactions by modifying surface tension and by increasing the solubility of reactants. Polymeric
surfactants may further influence the reaction by occupying space near the interface and by
AN
contributing to adhesion of the interfacially polymerized membrane to the support (at least in initial
stages). In contrast to surfactants that are defined by structure, phase transfer catalysts (PTC) are
defined by their function: to increase the polymerization rate by facilitating the transport of reactants
from one phase to another. For instance, common PTCs for anionic molecules are quaternary
M
ammonium salts. Added to the aqueous phase, they can complex and increase amine solubility in the
organic phase, thus enhancing the polymerization rate. Many times the same reagent can act as both
surfactant and phase transfer catalyst. For example, onium compounds with long alkyl chains such as
D

cetyl trimethylammonium bromide can act as both surfactant and PTC.

TE

Mansourpanah et al. [451] compared the effect of a cationic (cetyl trimethylammonium bromide),
anionic (sodium dodecyl sulfate), and non-ionic (triton X-100) surfactant in the organic phase on the
resulting NF membrane (PIP+TMC) performance. As compared to membranes made with the non-ionic
surfactant, membranes made with cationic and anionic surfactant showed higher flux (SDS showing
EP

higher flux than CTAB) and better rejection. The CTAB and Triton-X-100, owing to their positive charge
and hydrogen bonding capacity respectively, can absorb PIP and thus can promote its diffusion into the
organic phase to result in a thicker, denser film. In contrast, SDS, being a negatively charged surfactant,
does not adsorb PIP and can lead to a thinner skin layer with higher flux. Moreover, it was suggested
C

that SDS can lead to a higher free volume polymer due to its bigger polar head with higher space
hindrance and smaller hydrophobic tail compared to CTAB, resulting in higher flux. In a later publication
AC

[452], the same group reported use of the same surfactants in the same concentration ranges in the
aqueous phase. The trends were different and ascribed to better solubility of these water soluble
surfactants in the aqueous phase compared to the organic phase. For example, SDS showed higher flux
at concentrations >0.1%, but poor rejection due to the defects present in the film. SDS being a water
soluble surfactant might have formed micelles at higher concentration which did not occur due to its
poor solubility in the organic phase. The neutral surfactant Triton-X-100 showed the best rejection with
low flux. CTAB did not show any change in performance. Saha and Joshi [338] also reported the effect of
SDS concentration on membrane performance.
 ACCEPTED MANUSCRIPT

A variety of polymeric surfactants have been used in making interfacially polymerized membranes. One
of the early polymeric surfactants used for piperazine chemistry was PVP[453]. Recently Yeon Su Kim et
al.[454] reported adding different PVP based surfactants in the aqueous phase with piperazine to
increase flux. Chen et al.[455] added PEG in the aqueous amine solution to improve PS support wetting.
Kim et al. [456] described making a membrane with piperazine monomer and a non-ionic surfactant
polyoxyethylene p-tert octylphenyl ether having concentration of 0.01 to 1% in the aqueous phase to
improve membrane flux. Cationic polymeric surfactants such as sulfonium salts containing

PT
polymethacrylates have also been used in early studies [457].

Tertiary aliphatic amines, quaternary ammonium, phosphonium, arsonium, sulfonium compounds,

crown ethers all are examples of phase transfer reagents used for the synthesis of polymers such as

RI
polycarbonates, polyesters, polysulfonates etc. Tertiary amines or ammonium salts have been studied
extensively as phase transfer catalysts or flux enhancing additives for interfacially polymerized thin film
composite membranes. Tomaschke et al. have used tetramethyl ammonium hydroxide [458] and

SC
triethylamine camphorsulfonic acid salt [459, 460] in the aqueous phase to improve flux. Jegal et
al.[461] studied three PTCs based on benzyl ammonium salts. They reported that triethylbenzyl
ammonium bromide improved membrane flux, most probably by increasing the surface area for
polymerization. More recently Xiang et al.[462, 463] compared the effect of three symmetrical

U
(tetraethylammonium chloride (TEAC), tetrabutylammonium bromide (TBAB), camphorsulfonic acid
triethylamine salt (CAS-TEA)) and three asymmetrical (benzyltrimethylammonium chloride (BTMAC),
AN
benzyltrimethylammonium chloride (BTEAC), dodecyltrimethylammonium chloride (DTMAC))
ammonium salts on NF membrane structure and performance. Figure 13 shows how alkyl ammonium
salts facilitate interfacial polymerization by forming a complex with piperazine monomer in the aqueous
M
phase and thus increasing the diffusion of PIP into the organic phase because of the improved solubility
of the complex in that phase. The study found out that the thickness (or the transport of PIP monomers
into the organic phase) of the membranes increased in the order of CSA-TEA< TEAC< DTMAC <
BTMAC~BTEAC < TBAB which is proportional to the size of the catalysts. The paper describes a specific
D

cluster structure formed from TBAB and uses it to explain why transport efficiency likely increases in this
manner. They also report that water absorbing capacity of the ammonium salt is a critical factor, as
TE

excess water can hydrolyze TMC in the organic phase. Hence a weak membrane with lowest salt
rejection and highest water flux was obtained with CSA-TEA salt and weak adhesion between support
and film was seen for TEAC also. The high molecular weight ammonium salts did not show any such flux-
rejection trade-off as these larger ammonium salt complexes left larger free volumes in the membrane
EP

after they were washed away.

C
AC
 ACCEPTED MANUSCRIPT

PT
Figure 13. Role of ammonium salts as phase transfer reagents; QX presents an ammonium salt,
consisting of a highly lipophilic cation (Q+) and halide anion (X); M+Y represents the monomer in

RI
aqueous solution; and R+X stands for monomers in organic solution [463]

Yung et al. [283] used a novel additive, ionic liquid for fabricating NF membranes. Ionic liquids or liquid

SC
electrolytes are salts in the liquid state. They are environmentally friendly solvents and have been
reported in other interfacial polymerization processes [464, 465]. Figure 14 shows the structures of two
kinds of ionic liquids, 1-octyl-3-methylimidazolium chloride (OMIC) and 1-butyl-3-methyl-imidazolium
chloride (BMIC) which were added in the aqueous phase during the polymerization process. OMIC

U
behaved as a surfactant owing to its long hydrophobic tail. Hence with increasing concentration, OMIC
molecules can aggregate or form micelles at the interface leading to high porosity and increased flux,
AN
but with lower salt rejection. In contrast, due to a short hydrophobic tail, BMIC mainly behaved as a
phase transfer catalyst by complexing with piperazine and facilitating its transport into the organic
phase. Hence membranes made with BMIC were more crosslinked and had lower flux and higher
rejection.
M
D
TE
EP

Figure 14. Structures of two ionic liquids- BMIC and OMIC [283]
C

Mickols[466] and later Kim [467] have reported the use of phosphate based additives such as tributyl
phosphate and triphenyl phosphate(TBP) in organic phase to improve membrane flux. It was suggested
AC

that phosphate groups can associate or complex with the acid chloride group of the TMC and this
complex formation generates a concentration polarization of TMC in the organic phase which can affect
interfacial polymerization. It has more recently been seen that TBP increases the solubility of water
(and a hydrolyzed monomer) within the hydrocarbon solvent. [468] It can selectively catalyze the
hydrolysis of molecules comprising acid chlorides prior to reaction, which is useful in preparing new
molecules and increasing the resulting number of carboxylic acids within the membrane.
 ACCEPTED MANUSCRIPT

3.5.2 Co-solvents/solvents

For TFC membranes, co-solvents have been used as additives to improve membrane performance,
mainly the flux.[469] Among different co-solvent additives, NMP, DMF[470, 471], dimethyl sulfoxide
(DMSO)[472, 473], alcohols such as IPA [461, 474], and acetone[475, 476] have been reported to
enhance flux of thin film composite RO membranes. Co-solvents can increase the miscibility of the

PT
aqueous and the organic phase and a modified solubility facilitates diffusion of the amine from aqueous
to the organic phase. As a result, the surface roughness and the surface area of the membranes
increased, resulting in higher permeability of the membranes. Akbari et al. [477] also applied this
concept to NF membranes and showed that by adding DMSO in the aqueous phase, the membrane flux

RI
increased by almost 46% whereas salt rejection remained unaltered. Morphology analysis revealed
increased surface roughness for the NF membrane with the addition of DMSO in the aqueous phase.
Jegal et al.[461] demonstrated that the addition of a small amount of alcohols such as i-propanol and n-

SC
propanol in the aqueous phase increases the miscibility of the water phase with hexane and can make a
more diffused interface. The membrane formed at this diffused interface was rougher and showed
increase in flux. Liu et al. [478] reported the fabrication of a highflux thin film composite hollow fiber

U
NF membrane by adding cyclic ethers in the organic phase during the interfacial polymerization. They
hypothesized that the alicyclic solvent which had similar structure to PIP would improve its solubility and
diffusivity in the organic phase. The study showed that by increasing the polarity from dioxane, to
AN
oxolane, to trioxane, the interfacial polymerization reaction zone becomes wider due to the increased
miscibility of the two phases, leading to a drop in the local concentration of the functional groups. This
results in decreasing barrier layer thickness with increasing co-solvent polarity as shown in Figure 15.
M
Additionally the surface roughness increased with increasing co-solvent polarity as the interface stability
reduced due to higher miscibility of the two phases upon addition of the co-solvents. The low
concentration of reactants at the interface can also result in a loosely crosslinked network structure.
Hence the water permeability of the membranes improved when the polarity of the cosolvents
D

increased, but at the same time the MWCO increased also. This challenge was overcome by selecting
dioxane as the desired additive. It has the lowest polarity and thus does not widen the interfacial
TE

polymerization reaction zone, resulting in a dense barrier layer, but at the same time makes a thin
membrane enough to produce higher permeability than the control membrane.
C EP
AC

Figure 15. Thickness and surface roughness of TFC hollow fiber membranes with different co-solvents:
(a, f) no co-solvent; (b, g) dioxane (1 wt%); (c, h) oxolane (1 wt%) and (d, i) trioxane (1 wt%) [478]
 ACCEPTED MANUSCRIPT

More recently, Duan et al. [479] added hexamethyl phosphoramide to the aqueous phase and obtained
a membrane with almost 73% increase in flux compared to the control with minimal loss in salt
rejection.

Kim et al. [412] examined NF membranes prepared with two organic solvents, isoparaffin or hexane, as
a function of drying time following interfacial polymerizations. They found that isoparaffin based
membranes showed higher rejection of PEG200. However, relatively long reaction times (up to 30

PT
hours) were practiced, and they attributed differences to the lower boiling point of isoparaffin that
enabled curing reactions to proceed longer. Hu et al. [267] studied the effect of monomer
concentration, reaction time, and organic solvent type for solvent stable NF membranes which were
made on solvent stable poly(phthalazione ether nitrile ketone) (PPENK) support. Among the different

RI
solvents studied (toluene, xylene, hexane and cyclohexane), cylcohexane was the best for showing
optimum flux and rejection. As expected, a denser skin was obtained with increased reaction time and
there was an optimum range of PIP and TMC concentrations selected for flux and Na2SO4 rejection. In

SC
the RO space, Ghosh et al.[480] conducted a thorough study on four different organic solvents which
can provide a wide range of MPD solubility and diffusivity in those solvents. They demonstrated that
membrane flux and salt rejection could be improved if MPD diffusion is increased, but its solubility in
the organic phase is decreased.

U
3.5.3 Organic molecules
AN
Rana et al.[481] had studied the effect of blending a new type of hydrophilic surface modifying
macromolecules (LSMMs) synthesized by end-capping urethane prepolymer with poly(ethylene glycol)s,
M
in polyethersulfone for making UF membranes. The group extended this study by incorporating these
LSMMS in the polyamide layer by either adding the pre-synthesized macromolecules to the organic
phase or by forming the macromolecules in-situ during interfacial polymerization [482]. Membranes
D

made with in-situ formed macromolecules showed better NaCl rejection. In a further extension, these
membranes were tested with different foulants and they demonstrated good antifouling properties.
TE

Membrane made with both silver nanoparticles and LSMMs showed good antimicrobial properties[483].

NF membranes are generally hydrophilic and sometimes not suitable for non-aqueous solvent
separation. Silicones are well known for solvent resistance and high solvent flux. Kim et al. [484]
EP

developed a silicone coated NF membrane for potential non aqueous application. They incorporated
polydimethylsiloxane in the membrane by adding it in the organic layer alongside TMC when
interfacially polymerizing with a mixture of piperazine and MPD in the aqueous phase. Optimization
study pointed out that low concentration of monomers and high concentration of PDMS are needed for
C

high solvent flux. The membranes showed high flux of hexane like hydrophobic solvents. Zhou et al.
[485]also used a 63000 g/mole PDMS in the organic phase to produce a membrane with almost double
AC

the flux of an unmodified membrane with MWCO increasing from 200 to 340 Da, but with a stable
Na2SO4 rejection. An et al. [486] reported the addition of PVA in the aqueous phase during interfacial
polymerization between piperazine and TMC to enhance flux. Surface roughness decreased with
increasing PVA content in the polyamide which helped to impart antifouling properties.

In contrast to RO, NF membrane preparation usually requires an acid scavenger to enhance the
polymerization rate. A variety of different acid scavengers/acceptors may be used to enhance the rates
of reaction. These may be simple inorganic molecules (e.g. sodium hydroxide, sodium tertiary
phosphate). Alternatively, several small organic molecules such as N,N-dimethylpiperazine, , tetrabutyl
ammonium hydroxide, triethylamine also can be used.
 ACCEPTED MANUSCRIPT

3.5.4 Inorganic Salts

A few reports describe the effect of inorganic salts as additives for polyamide NF membrane
preparation. Tang et al.[487] studied the influence of LiBr on NF membrane properties when it was
added in the aqueous phase with amines triethanolamine (TEOA) or N-methyl-diethanolamine
(MDEOA). They hypothesized that LiBr can break the hydrogen bonds generated by the OH groups of

PT
TEOA molecules and thus can increase the number of the available OH groups and also the reactivity of
the OH groups as demonstrated in Figure 16, thus enhancing the crosslinking reaction between TEOA
and TMC. However, LiBr can also complex with the carbonyl group of TMC prompting the hydrolysis of
TMC, thus making a loosely crosslinked network structure. These two competing effects of LiBr resulted

RI
in low salt rejection at LiBr concentration less than 1% when TMC complexation is dominant and high
salt rejection at concentrations higher than 1% when TEOA-LiBr interaction is dominant. The opposite
effect was observed in case of MDEOA monomer as the MDEOA-LiBr interaction was strong in this case.

U SC
AN
M

Figure 16. Interaction between Li+ ion and the hydroxyl oxygen atom of TEOA increasing free OH
D

groups by breaking internal H-bonds and improving the reactivity of OH groups [487]

Fan et al. [488]reported adding CaCl2 in the aqueous phase which improved water permeability for NF
TE

membranes synthesized interfacially with tetraethylenepentamine and TMC. They hypothesized that Ca
ion complexed with the carbonyl group of TMC and was encapsulated in the polyamide active layer. This
resulted in a comparatively loose polyamide membrane leading to increased flux. Additionally, the
EP

membranes also showed improved chlorine resistance due to the complexation of the carbonyl groups
of amide with Ca ions.

3.5.5 Nanoparticles
C

Great attention has recently been paid to a new class of additives in the area of interfacially polymerized
AC

membranes. Addition of nanoparticles to the barrier layer purports to offer various attractive
properties for membranes such as high permeability, increased mechanical strength, better selectivity,
antimicrobial properties etc. The term TFN or thin film nanocomposite membrane was coined by Hoek
and his coworkers in a patent application and journal article.[489, 490] The application focused on
incorporating zeolite nanoparticles into the interfacial polymerization, but it described many other
options that have since been the focus of several research efforts. Among various types of particles now
added to interfacial polymerizations, some of the more common include zeolites, silica nanoparticles,
TiO2, and carbon nanotubes. A subsequent patent has also described advantages of incorporating
nanoparticles within the support layer.[258] Several newer review articles [491, 492] provide more
comprehensive discussions of use of nanoparticles for RO/NF.
 ACCEPTED MANUSCRIPT

Zeolite nanoparticles have been extensively studied for TFN RO membranes [257, 259, 490, 493-496],
and these particles have recently been included in an NF membrane. Namvar-Mahboub [497, 498]
prepared solvent resistant TFN NF membranes by incorporating amine functionalized UZM-5 zeolite
nanoparticles in the polyamide layer on top of a silica modified polyetherimide support.

Silica nanoparticles have been studied broadly because of their small size, large surface area, narrow
particle size distribution, hydrophilic property, and thermal resistance[499] and have been used in a

PT
mesoporous form[500] or as non-porous silica nanoparticles[501]. Li et al. [502] introduced a
mesoporous silica SBA-15 as an additive for making nanocomposite NF membranes. They described SBA-
15 as having a very high specific surface area, ordered structure, and controllable pore sizes, which
make it an attractive candidate for increasing membrane hydrophilicity. They reported an increase in

RI
flux from 32 to 46 LMH by using SBA-15 and improvement of antifouling properties, though the MgSO4
salt rejection dropped from 98% to 85%. Tang et al. [503, 504] also reported amino modified
mesoporous silica for making nanocomposite NF membrane. This same group[505] has also studied NF

SC
membranes made with silica nanospheres (235 nm). The said nanocomposite membranes showed an
increase in MgSO4 rejection from 87% to 94% with a moderate increase in flux. Jin et al. [444-446] chose
a novel polyamide amine (PAMAM) dendrimer-based NF membrane for studying the effect of SiO2
nanoparticles. This PAMAM has a hyperbranched structure with inner cavity properties. They

U
hypothesized that these cavities could be used as a carrier for silica nanoparticles. The use of SiO2
improved the thermal resistance, water permeability and antifouling resistance of these PAMAM based
AN
NF membranes. Kebria et al. [506]also used SiO2 nanoparticles to increase the hydrophilicity of a
polyethyleneimine (PEI) based NF membrane. Alternatively, it has also been seen that poorly dispersed
SiO2 nanoparticles can agglomerate and hinder membrane formation or its performance. Hu et al. [507]
M
used a 20% silica sol to achieve better dispersion. Ma et al.[508] modified silica nanoparticles with CTAB
surfactant to enhance dispersion and made silica/polyamide-imide nanocomposite membranes.

Lee et al.[509] incorporated TiO2 nanoparticles in the organic phase for producing TFN NF membranes
D

and reported an increase in membrane flux compared to the control. The optimum membrane
performance was obtained at a TiO2 concentration of 5%, above which the membrane falls apart due to
TE

the interference of TiO2 nanoparticles with interfacial polymerization. Rajaeian et al. [510] modified
TiO2 nanoparticles with an aminosilane coupling agent called N-[3-(Trimethoxysilyl) propyl]
ethylenediamine (AAPTS) and then added them in the amine solution to minimize the agglomeration
between particles and also to enhance the uniform distribution of the particles in the polyamide layer
EP

which can be facilitated by the organic monolayer coating. Optimum membrane performance was
obtained at low concentration (0.005%) of modified TiO2 nanoparticles and the thermal resistance of the
membranes was also improved due to TiO2 incorporation. Bai et al.[511] improved the dispersion of TiO2
C

nanoparticles in the aqueous phase by using a dispersant agent 2-hydroxypropyl trimethyl ammonium
chloride chitosan (HACC) for making a polyethyleneimine based positively charged NF membrane. The
AC

flux and dye rejection of the membranes increased, while salt rejection decreased with increasing TiO2
content in the membrane. Peyravi et al.[512] prepared solvent resistant TFN NF membranes by
incorporating aminated or chlorinated TiO2 nanoparticles in aqueous and organic phase during the
interfacial polymerization between ethylenediamine and isophthaloyl chloride on a PEI coated polyimide
support. The membranes showed good stability in organic solvents and high methanol flux for solvent
separation.

Sorribas et al. prepared solvent resistant NF membranes by incorporating metal-organic

framework(MOF) nanoparticles [513] in the organic layer during interfacial polymerization. Methanol
and THF flux increased for MOF modified membranes compared to the control. An and Ji et al. [514-516]
reported the preparation of NF membranes by adding carboxy or sulfo betaine type colloid
 ACCEPTED MANUSCRIPT

nanoparticles in the aqueous phase. Improved hydrophilicity and fouling resistance were reported.
Kong et al. [517] described the preparation of TFN NF membranes by incorporating Ti and Si based metal
alkoxides in the organic phase. They hypothesized that metal alkoxides will have superior solubility in
the organic phase compared to the inorganic nanoparticles. The optimized membrane showed almost
two fold increase in flux compared to the control with minimal loss in rejection. Kim and Deng [518]
incorporated hydrophilized ordered mesoporous carbons (H-OMCs) in the polyamide layer made from
MPD and TMC. The membranes had improved permeability and surface hydrophilicity and less

PT
adsorption of BSA protein.

RI
3.6 Post Treatments

SC
In many instances membrane are post treated to improve existing properties or to impart new
attributes. This section describes some of the different post treatments applied to interfacially
polymerized membranes. Polymeric surface coatings are one class of post treatments that has been
mentioned previously in this review.[288-298] While discussed earlier in terms of degradation, post

U
treatment with chlorine may also be used to modify performance of NF properties.[388, 519] Several of
the other post treatments mentioned below are similar to those described previously for making
modifying UF supports to make NF membranes. AN
One approach is to react the residual amine or acid chloride functional groups on the membrane surface
after completion of the interfacial polymerization. Mi et al. [520] put a reactive coating on NF
M
membranes (PIP+TMC) by immersing the interfacially formed membrane in an aqueous solution of N-
aminoethyl piperazine propane sulfonate (AEPPS), so that the amine groups can react with the
unreacted, residual acid chloride groups on the membrane surface. The resulting membrane almost
D

doubled in flux without suffering any significant loss in Na2SO4 rejection and also showed an
improvement in the antifouling properties. Chiang et al. [521] synthesized NF membranes by
TE

interfacially polymerizing diethylenetriamine and TMC on a PAN support and then post treated the
membranes with iodopropionic acid through N-alkylation reaction of the secondary amide groups. The
membrane was further treated with iodomethane so that the tertiary amines were converted to
quaternary amines. Thus a zwitter ionic membrane surface was formed. These membranes showed
EP

improved resistance to humic acid fouling.

Polymers have also been attached to the membrane to change its character. Li [522] grafted a
fluorinated polyamine to the carboxylic acid groups on the surface of an NF membrane (PIP+TMC) by
C

immersion at elevated temperatures for 3 hrs. Fluorination of the surface had a small impact on flux
and reduced rejection of salts (due to fewer negative charges on the surface). Dye rejection was still
AC

high, and the relative flux declines during operation were reduced for BSA and humic acid. Belfer [523]
performed redox grafting of HEMA, either onto commercial membranes (NF70, NF90, NF200 and NF270)
or by in situ treatment of spiral wound elements (NF70, NF90, NF270). Modification caused a modest
decrease in fouling. Non-charged hydrophilic monomers were seen to cause increased rejection. Jae-
Hyuk Kim[524] et al. modified a commercial NF membrane (MPD+TMC) by free-radical graft
polymerization of methacrylic acid and grafted chains were subsequently crosslinked with ethylene
diamine. As this crosslinking reduced negative charge, the membrane was further reacted with
succininc acid to increase negative charge again. Zeta potential, contact angle and performance in the
presence of pollutants were tracked at each stage. Mansourpanah and Habili [525]grafted acrylic acid
to an NF membrane (PIP+TMC). They studied how the morphology and performance of membranes
 ACCEPTED MANUSCRIPT

differed when UV radiation was applied either during or after the interfacial polymerization and for
different times (30, 60, 120 seconds). Application of UV radiation during the interfacial polymerization
was found to be effective and preferred in terms of increased flux (compared to no UV-polymerization),
increased rejection of Na2SO4, and decreased fouling propensity. Himstedt [526] used UV initiated
polymerization to grow acrylic acid nanobrushes from the surface of NF270 membrane. The glucose flux
and rejection was more pH dependent over pH 3-7. The configuration of grafted polymers switched
with pH between expanded and collapsed conformations, and it was hypothesized that this feature

PT
might be useful for removing foulants.

Eun-Sik Kim et al. [527] treated commercial nanofiltration membranes with a low-pressure NH3 plasma.
For NF270 and NF90 membranes, there was no apparent change in salt rejection with the duration of

RI
plasma treatment, but permeability, hydrophilicity and negative charge on the surface all increased.
Treatment caused BSA to be less adsorbed and relative flux decline was less in tests with aldrich humic
acid.

SC
Navarro et al.[528] treated a piperazine-based membrane DS5-DL for 14 or 35 days with 0.5 M HF, 8 M
H3PO4, or both. Both treatments increased flux, but the H3PO4 caused more flux, greater passage of
cationic impurities, and larger pore radii (as determined by glucose and sucrose). Results were

U
interpreted in terms of partial hydrolysis of the polyamide.

AN
In some cases, a membrane is post treated after formation by soaking it in aqueous solution containing
non-reactive flux enhancer additives. As previously noted, Jimenez-Solomon et al. [414] soaked
interfacially polymerized solvent resistant NF membranes in an activating solvent of DMF to swell the
membrane and rinse out small molecular fragments. Kuehne et al.[529] reported flux increment by 30-
M
70% with no significant loss in rejection by soaking the membranes in water containing glycerol and SDS
or organic salts such as salt of camphorsulfonic acid and triethylamine and then drying in oven. Chen et
al.[530] post-treated NF membranes after interfacial polymerization by immersing them in aqueous
D

solutions containing hydroxyl based modifiers such as diethylene glycol, butylene glycol etc. to improve
the flux.
TE

Treating membranes with surfactants has long been known to modify membrane performance. (See
Wilbert [531] and references therein.) Nath et al.[532] recently contacted an NF polyamide membrane
with different concentrations of cationic CTAB, anionic SDS, and non-ionic X-100 surfactants. AFM
EP

suggested a decrease in the height of surface features due to surfactant. A decrease in flux was
generally observed as surfactant concentrations increased, but increased rejection of yellow 160 dye
was noted for membranes treated with cationic and anionic surfactants.
C

NF membranes may be derived from reverse osmosis membranes by post treatment. Cadotte et.al.
[533] found that controlled contact of an interfacially polymerized RO membrane with a strong mineral
AC

acid, such as phosphoric acid at elevated temperature, resulted in an NF membrane. Subsequent

treatment with a rejection enhancing agent (e.g. tannic acid, polymeric dispersion, water soluble
polymer) could then selectively plug excessive defects created in the barrier layer. Wheeler[534]
describes how an RO membrane exposed to low levels of an oxidizable ion (Mn+2, Sn+2, Fe+2) that forms
membrane-ion complexes can be subsequently opened up by exposure to acidic permanganate. Upon
dissolving the resulting manganese dioxide crystals, an NF membrane may be obtained. Mickols[535]
treated RO membranes with triethanolamine to swell the polyamide, making a tight NF. Low level
chlorination of reverse osmosis membranes will often initially decrease salt passage, but more excessive
exposure will increase both water and salt permeability, making an NF membrane. Recent sustainability
efforts have attempted to use NaOCl to convert end-of-life spiral RO elements into low pressure UF or
 ACCEPTED MANUSCRIPT

NF elements, principally for use in brackish water pretreatment or tertiary treatment of

wastewater.[536, 537]

4.0 Conclusions and Recommendations

PT
In this review article, we have tried to summarize different methods for making polymeric NF
membranes and identify significant research activity in each area. Special attention was given to three
approaches that are used to make commercial membranes: interfacial polymerization, phase inversion,
and different post-treatments (e.g. coatings, grafting) that may be applied to a more porous support.

RI
Emerging technologies also discussed include layer-by-layer coatings, incorporation of aquaporins, and
use of glassy polymers with high internal porosity. However, the primary focus of this review was
interfacially polymerized NF membranes. Within that confine, topics covered in particular detail include

SC
the composite structure, monomeric and polymeric reactants employed to endow specific
characteristics, additives used to influence the reaction, and post treatments.

The growth of nanofiltration is such that more than about two thirds of all publications concerning NF

U
membranes were written in the last decade. Even in the smaller focus space of interfacially polymerized
NF membranes, this review is able to mention only a small fraction of relevant art. Nonetheless, the act
AN
of considering many references has suggested a few points to mention.

A strong focus of recent activity has been on membrane performance and stability in challenging
environments (e.g. pH extremes, high temperature, chlorine, solvents). Especially as new interfacially
M
polymerized membranes are explored for operations at those conditions, it would be beneficial to
better understand not only their performance but also their means of failure. Taking pH an example,
performance can be influenced by breakage of hydrogen bonds and modification of the tertiary
D

structure, reversible and irreversible swelling, irreversible hydrolysis, and even delamination of the
barrier layer from the support. In some of these cases, thin membranes that have generally favorable
TE

permeability will be more impacted. It is important to separate factors to understand the extent to
which chemistry and physical modifications will be necessary and effective for interfacially polymerized
membranes. At the same time, other membrane types have been examined that already do have
greater tolerance to pH, temperature, and solvent exposure. For some applications, these robust
EP

membranes have sufficient properties (flux, rejection, selectivity, cost) for commercial success, but
replacements of conventional polyamides in other applications will require improvements to base
membrane performance.
C

The introduction noted that some NF membranes are valued for their ability to produce sufficient flux
and rejection at low energy, whereas others are prized for properties relating to selective separation of
AC

different species. Many papers are using PEGs, styrene oligomers, several neutral molecules, or dyes of
different molecular weights to characterize performance of NF membranes. However, many more limit
characterization to a few salts and frequently identify the best membrane based on just high flux and
good rejection of some specific salt. Neither approach is wrong, but a unique and appreciated property
of nanofiltration is selectivity. While not relevant to every situation, we would encourage researches to
challenge new membranes such that both ionic selectivities are measured and a MWCO for neutral
molecules is determined.

Ideally the art would converge on a few sets of generally accepted standard challenge tests that could
enable determination of fundamental parameters and facilitate comparisons. In gas separations,
 ACCEPTED MANUSCRIPT

Robeson plots (log selectivity ij vs log permeability Pi) have identified the upper bounds of current
technology. For reverse osmosis membranes, chemistry changes resulting in increased water
permeability usually (though not always) also result in greater permeabilities for salts and low molecular
weight neutral species. It is common in RO to compare membranes using a plot of A values (water
permeability) vs. BNaCl values (NaCl permeability). Similar approaches are meritorious to compare the
many commercial and experimental NF membranes that have been produced, but published data
typically lacks uniformity. We suggest that the field would benefit from the NF community defining a

PT
few recognized testing protocols that would be appropriate to most membranes used in aqueous
separations. Beyond specific feed compositions (e.g. 2000 ppm MgSO4, 2000 ppm NaCl, simulated
seawater, specific PEG mix), defining test conditions (including test cell, flux, and recovery) would allow

RI
membranes of widely varying permeabilities to be compared with calculable impacts of polarization.
While any research group might implement these test protocols, establishing testing centers that would
be available to perform these defined tests for others would enable some researchers to concentrate
efforts on membrane development. A more ambitious plan could also include tests for stability and

SC
performance with different solutes.

Recent literature demonstrates many new membranes and manufacturing techniques, often showing
that a particular approach has resulted in improved performance compared to available commercial

U
membranes. A more uniform set of test procedures would clarify the relative differences and
advantages. However, there are many other factors that enter into decisions to commercialize. Since
AN
cost is frequently as important a driver for customers as improved water permeability, solute rejection,
or selectivity, among the important factors considered is the ease of manufacture. Processes not easily
made continuous and high speed are disadvantaged but not excluded. However, better discussion and
M
cooperation between academicians and manufacturers will help to identify the challenges to be the
met, and better the rate at which new technologies might be adopted.
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT

Table 1. Name, chemical structure and reference number of selected monomers mentioned in the
review article

Monomer Name & Reference no. Chemical Structure

Monomers from section 3.3.1 that can provide charged groups

3,5-diaminobenzoic acid

PT
[299,300,301-304,338]

RI
2,5-bis(4-amino-2-trifluoromethyl-
phenoxy)benzenesulfonic acid [309]

SC
[1,1'-Biphenyl]-3,3'-disulfonic acid, 4,4'-
bis[4-amino-2-(trifluoromethyl)phenoxy]-

U
[309]

2,5-diaminobenzenesulfonic acid (2,5-

DABSA) [310]
AN
Taurine (2-aminoethane-1-sulfonic acid)
M
[311]
1,4-Bis(3-aminopropyl)piperazine (DAPP)
[312]
D

2,4,6-pyridine tricarboxylic acid chloride

TE

[313]

Monomers from section 3.3.2 providing hydrophilic character

EP

2, 2'-oxybis-ethylamine (2,2'-OEL) [323]

4,4-dihydroxybiphenyl
C

[300]
AC

m-aminophenol (m-AP)
[325,326],
bisphenol-A (Bis-A) [325,326,329,330],
and hydroquinone (HQ)[325,326]

Phenolphthalein,
tetrabromo bisphenol-A
[327]
 ACCEPTED MANUSCRIPT

triethanolamine (TEOA)
[331,332]
I-
NH
N+ 2
1,1-diamino-4,4-bipyridinium di-iodide
[335]
N+

PT
H2N
I-

Monomers from section 3.3.3 with varying functionality and structure

1,6-hexanediamine, 1,3-propanediamine,

RI
1,2-ethanediamine and hydrazine
[211]
1,4-cyclohexanediamine

SC
[336,343,369] ,
1,3-bis (aminomethyl) cyclohexane[336]
N,N'-diaminopiperazine (DAP) [266, 337],

U
1,4-bis(3-aminopropyl)-piperazine (DAPP)
[337]and N-(2-aminoethyl)-piperazine
(EAP) [337,338]
diethylenetriamine, triethylenetetramine
and tetraethylenepentamaine
AN
[339]
M
p-xylylenedimaine and tetrakis (N-
methyl-amino-methyl) methane
[340]
D

3,5-diaminobenzoyl piperazine
TE

[344]

Cyclen
EP

[345]

L-lysine
[346-348]
C

Tetrakis(4-carboxyphenyl)methane
AC

(TPMC) and tetrakis(4-

carboxyphenyl)silane (TPSC)
[349]

3,3',5,5'-biphenyl tetraacyl chloride (mm-

BTEC), 2,2',4,4'-biphenyl tetraacyl
chloride (om-BTEC), and 2,2',5,5'-biphenyl
tetraacyl chloride (op-BTEC)
[350,351]
 ACCEPTED MANUSCRIPT

2,2',4,4',6,6'-biphenyl hexaacyl chloride

(BHAC) [353]

Monomers from section 3.3.4 for chlorine resistance

PT
4,4-diaminodiphenylsulfone [366]

RI
N,N-dimethyl-m-phenylenediamine
(N,N-DMMPD)
[367]

SC
N-phenylethylenediamine (N-PED)
[368]
m-phenylenediamine-4-methyl (MMPD)
[370]

U
cyclohexane-1,3,5-tricarbonyl chloride
[370] AN
2,2'-bis(1-hydroxyl-1-trifluoromethyl-
M
2,2,2-trifluoroethyl)-4,4'-
methylenedianiline (BHTTM) [371]
D

1,2,4,5-benzene tetracarbonyl chloride

(BTC) [375]
TE

disulfonated bis[4-(3-
aminophenoxy)phenyl]sulfone (S-
BAPS)[381]
EP

Bis-2, 6-N, N-(2-hydroxyethyl)

diaminotoluene [383]
Monomers from section 3.3.5 for extending pH range
C

naphthalene-1,3,6-trisulfonylchloride
(NTSC)[392-396]
AC

Cl

cyanuric acid chloride[397,398] N N

Cl N Cl
 ACCEPTED MANUSCRIPT

5.0 References

1. Schfer, A.I., A.G. Fane, and T.D. Waite, Nanofiltration : principles and applications. 2005,
Oxford: Elsevier.

PT
2. Ventresque, C., et al., An outstanding feat of modern technology: the Mery-sur-Oise
nanofiltration Treatment plant (340,000 m3/d). Desalination, 2000. 131(13): p. 1-16.
3. Rice, G., et al., Membrane-Based Dairy Separation: A Comparison of Nanofiltration and

RI
Electrodialysis. Developments in Chemical Engineering and Mineral Processing, 2005. 13(1-2): p.
43-54.
4. Allgre, C., et al., Treatment and reuse of reactive dyeing effluents. Journal of Membrane
Science, 2006. 269(1-2): p. 15-34.

SC
5. Sjman, E., et al., Separation of xylose from glucose by nanofiltration from concentrated
monosaccharide solutions. Journal of Membrane Science, 2007. 292(12): p. 106-115.
6. Davis, R., I. Lomax, and M. Plummer, Membranes solve North Sea waterflood sulfate problems.

U
Oil and Gas Journal, 1996. 94(48): p. 59-64.
7. Twardowski, Z., US5587083, Nanofiltration of concentrated aqueous salt solutions. 1996.
8.

9.
AN
Petersen, R.J.C., J.E., Thin film composite reverse osmosis membranes, in Handbook of Industrial
Membrane Technology,, M.E. Porter, Editor. 1990, Noyes Publications: Park Ridge, NJ.
Petersen, R.J., Composite reverse osmosis and nanofiltration membranes. Journal of Membrane
Science, 1993. 83(1): p. 81-150.
M
10. Hilal, N., et al., A comprehensive review of nanofiltration membranes:Treatment, pretreatment,
modelling, and atomic force microscopy. Desalination, 2004. 170(3): p. 281-308.
11. Mohammad, A.W., et al., Nanofiltration membranes review: Recent advances and future
D

prospects. Desalination, 2015. 356: p. 226-254.

12. Marchetti, P., et al., Molecular separation with organic solvent nanofiltration: A critical review.
Chem. Rev. (Washington, DC, U. S.), 2014. 114(21): p. 10735-10806.
TE

13. Vandezande, P., L.E. Gevers, and I.F. Vankelecom, Solvent resistant nanofiltration: separating on
a molecular level. Chem Soc Rev, 2008. 37(2): p. 365-405.
14. Hermans, S., et al., Recent developments in thin film (nano)composite membranes for solvent
EP

resistant nanofiltration. Current Opinion in Chemical Engineering, 2015. 8: p. 45-54.

15. Van der Bruggen, B., M. Mnttri, and M. Nystrm, Drawbacks of applying nanofiltration and
how to avoid them: A review. Separation and Purification Technology, 2008. 63(2): p. 251-263.
16. Li, D. and H. Wang, Recent developments in reverse osmosis desalination membranes. Journal of
C

Materials Chemistry, 2010. 20(22): p. 4551.

17. Lau, W.J., et al., A recent progress in thin film composite membrane: A review. Desalination,
AC

2012. 287: p. 190-199.

18. Ismail, A.F., et al., Thin film composite membrane Recent development and future potential.
Desalination, 2015. 356: p. 140-148.
19. Tomaschke, J.E., Membrane preparation, interfacial composite membranes, in Encyclopedia of
Separation Science, I.D. Wilson, Editor. 2000, Academic Press, Oxford, . p. 3319-3331.
20. Lau, W.J., et al., A review on polyamide thin film nanocomposite (TFN) membranes: History,
applications, challenges and approaches. Water Research, 2015. 80: p. 306-324.
21. Yin, J. and B. Deng, Polymer-matrix nanocomposite membranes for water treatment. Journal of
Membrane Science, 2015. 479: p. 256-275.
 ACCEPTED MANUSCRIPT

22. Joseph, N., R. Hoogenboom, and P. Ahmadiannamini, Layer-by-layer preperation of

polyelectrolyte multilayer membranes for separation. Polymer Chemistry, 2013. 5: p. 1817-1831.
23. Zhao, Q., et al., Polyelectrolyte complex membranes for pervaporation, nanofiltration and fuel
cell applications. Journal of Membrane Science, 2011. 379(12): p. 19-45.
24. Ng, L.Y., A.W. Mohammad, and C.Y. Ng, A review on nanofiltration membrane fabrication and
modification using polyelectrolytes: Effective ways to develop membrane selective barriers and
rejection capability. Advances in Colloid and Interface Science, 2013. 197198: p. 85-107.

PT
25. Mller, E.A., Purification of water through nanoporous carbon membranes: a molecular
simulation viewpoint. Current Opinion in Chemical Engineering, 2013. 2(2): p. 223-228.
26. Cohen-Tanugi, D. and J.C. Grossman, Nanoporous graphene as a reverse osmosis membrane:

RI
Recent insights from theory and simulation. Desalination, 2015. 366: p. 59-70.
27. Cohen-Tanugi, D. and J.C. Grossman, Water desalination across nanoporous graphene. Nano
Lett, 2012. 12(7): p. 3602-3608.
28. Rikhtehgaran, S. and A. Lohrasebi, Water desalination by a designed nanofilter of graphene-

SC
charged carbon nanotube: A molecular dynamics study. Desalination, 2015. 365: p. 176-181.
29. Li, Z., et al., Superstructured Assembly of Nanocarbons: Fullerenes, Nanotubes, and Graphene.
Chem Rev, 2015. 115(15): p. 7046-7117.

U
30. Hegab, H.M. and L. Zou, Graphene oxide-assisted membranes: Fabrication and potential
applications in desalination and water purification. Journal of Membrane Science, 2015. 484: p.

31.
95-106.
AN
Valadez-Blanco, R. and A.G. Livingston, Solute molecular transport through polyimide
asymmetric organic solvent nanofiltration (OSN) membranes and the effect of membrane-
formation parameters on mass transfer. Journal of Membrane Science, 2009. 326(2): p. 332-342.
M
32. Hwang, H.Y., et al., PES/SPAES blend membranes for nanofiltration: The effects of sulfonic acid
groups and thermal treatment. Desalination, 2012. 289: p. 72-80.
33. Guillen, G.R., et al., Direct microscopic observation of membrane formation by nonsolvent
D

induced phase separation. Journal of Membrane Science, 2013. 431: p. 212-220.

34. Ferreira, J.C.C., et al., Variation of the physicochemical and morphological characteristics of
solvent casted poly(vinylidene fluoride) along its binary phase diagram with dimethylformamide.
TE

Journal of Non-Crystalline Solids, 2015. 412: p. 16-23.

35. Holda, A.V., Ivo, Understanding and guiding the phase inversion process for synthesis of solvent
resistant nanofiltration membranes. Journal of Applied Polymer Science 2015.
EP

36. Sadrzadeh, M. and S. Bhattacharjee, Rational design of phase inversion membranes by tailoring
thermodynamics and kinetics of casting solution using polymer additives. Journal of Membrane
Science, 2013. 441: p. 31-44.
37. Guillen, G.R., et al., Preparation and Characterization of Membranes Formed by Nonsolvent
C

Induced Phase Separation: A Review. Industrial & Engineering Chemistry Research, 2011. 50(7):
p. 3798-3817.
AC

38. Mulder, M., Basic Principles of Membrane Technology. 1991, Netherlands: Kluwer Academic
Publishers.
39. Matsuura, T., Synthetic Membranes and Membrane Separation Processes. 1994, London: CRC
Press.
40. Hendrix, K., G. Koeckelberghs, and I.F.J. Vankelecom, Study of phase inversion parameters for
PEEK-based nanofiltration membranes. Journal of Membrane Science, 2014. 452: p. 241-252.
41. Peeva, L., et al., Experimental strategies for increasing the catalyst turnover number in a
continuous Heck coupling reaction. Journal of Catalysis, 2013. 306: p. 190-201.
 ACCEPTED MANUSCRIPT

42. Darvishmanesh, S., et al., Performance of Solvent-Pretreated Polyimide Nanofiltration

Membranes for Separation of Dissolved Dyes from Toluene. Ind. Eng. Chem. Res., 2010. 49: p.
9330-9338.
43. Cheng, X.Q., et al., Recent Advances in Polymeric Solvent-Resistant Nanofiltration Membranes.
Advances in Polymer Technology, 2014. 33(S1): p. n/a-n/a.
44. Liu, L., et al., Preparation and characterization of asymmetric polyarylene sulfide sulfone (PASS)
solvent-resistant nanofiltration membranes. Materials Letters, 2014. 132: p. 11-14.

PT
45. Darvishmanesh, S., et al., Novel polyphenylsulfone membrane for potential use in solvent
nanofiltration. Journal of Membrane Science, 2011. 379(12): p. 60-68.
46. Sani, N.A., W.J. Lau, and A.F. Ismail, Performance of Polyphenylsulfone-based Solvent Resistant

RI
Nanofiltration Membranes in Removing Dyes from Methanol Solution. Teknologi Jurnal, 2014: p.
29-32.
47. Loh, X.X., et al., Crosslinked integrally skinned asymmetric polyaniline membranes for use in
organic solvents. Journal of Membrane Science, 2009. 326(2): p. 635-642.

SC
48. Sairam, M., et al., Nanoporous asymmetric polyaniline films for filtration of organic solvents.
Journal of Membrane Science, 2009. 330(12): p. 166-174.
49. Sairam, M., et al., Spiral-wound polyaniline membrane modules for organic solvent nanofiltration

U
(OSN). Journal of Membrane Science, 2010. 349(12): p. 123-129.
50. Hoda, A.K. and I.F.J. Vankelecom, Integrally skinned PSf-based SRNF-membranes prepared via

51.
Science, 2014. 450: p. 499-511.
AN
phase inversionPart B: Influence of low molecular weight additives. Journal of Membrane

Buonomenna, M.G., et al., Asymmetric PEEKWC membranes for treatment of organic solvent
solutions. Journal of Membrane Science, 2011. 368(12): p. 144-149.
M
52. Buonomenna, M.G., et al., Preparation, characterization and use of PEEKWC nanofiltration
membranes for removal of Azur B dye from aqueous media. Reactive and Functional Polymers,
2009. 69(4): p. 259-263.
D

53. Hendrix, K., et al., Synthesis of modified poly(ether ether ketone) polymer for the preparation of
ultrafiltration and nanofiltration membranes via phase inversion. Journal of Membrane Science,
2013. 447: p. 96-106.
TE

54. da Silva Burgal, J., et al., Controlling molecular weight cut-off of PEEK nanofiltration membranes
using a drying method. Journal of Membrane Science, 2015. 493: p. 524-538.
55. da Silva Burgal, J., et al., Organic solvent resistant poly(ether-ether-ketone) nanofiltration
EP

membranes. Journal of Membrane Science, 2015. 479: p. 105-116.

56. Hendrix, K., et al., Crosslinking of modified poly(ether ether ketone) membranes for use in
solvent resistant nanofiltration. Journal of Membrane Science, 2013. 447: p. 212-221.
57. Vandezande, P., et al., High throughput study of phase inversion parameters for polyimide-based
C

SRNF membranes. Journal of Membrane Science, 2009. 330(12): p. 307-318.

58. See Toh, Y.H., F.W. Lim, and A.G. Livingston, Polymeric membranes for nanofiltration in polar
AC

aprotic solvents. Journal of Membrane Science, 2007. 301(12): p. 3-10.

59. Hendrix, K., K. Vanherck, and I.F.J. Vankelecom, Optimization of solvent resistant nanofiltration
membranes prepared by the in-situ diamine crosslinking method. Journal of Membrane Science,
2012. 421422: p. 15-24.
60. Vanherck, K., et al., A simplified diamine crosslinking method for PI nanofiltration membranes.
Journal of Membrane Science, 2010. 353(12): p. 135-143.
61. Vanherck, K., et al., Cross-linked polyimide membranes for solvent resistant nanofiltration in
aprotic solvents. Journal of Membrane Science, 2008. 320(12): p. 468-476.
 ACCEPTED MANUSCRIPT

62. See-Toh, Y.H., M. Silva, and A. Livingston, Controlling molecular weight cut-off curves for highly
solvent stable organic solvent nanofiltration (OSN) membranes. Journal of Membrane Science,
2008. 324(12): p. 220-232.
63. Soroko, I., M.P. Lopes, and A. Livingston, The effect of membrane formation parameters on
performance of polyimide membranes for organic solvent nanofiltration (OSN): Part A. Effect of
polymer/solvent/non-solvent system choice. Journal of Membrane Science, 2011. 381(12): p.
152-162.

PT
64. Soroko, I., et al., The effect of membrane formation parameters on performance of polyimide
membranes for organic solvent nanofiltration (OSN). Part B: Analysis of evaporation step and the
role of a co-solvent. Journal of Membrane Science, 2011. 381(12): p. 163-171.

RI
65. Soroko, I., M. Sairam, and A.G. Livingston, The effect of membrane formation parameters on
performance of polyimide membranes for organic solvent nanofiltration (OSN). Part C. Effect of
polyimide characteristics. Journal of Membrane Science, 2011. 381(12): p. 172-182.
66. Chen, D., et al., Solvent resistant nanofiltration membrane based on polybenzimidazole.

SC
Separation and Purification Technology, 2015. 142: p. 299-306.
67. Valtcheva, I.B., et al., Beyond polyimide: Crosslinked polybenzimidazole membranes for organic
solvent nanofiltration (OSN) in harsh environments. Journal of Membrane Science, 2014. 457: p.

U
62-72.
68. Valtcheva, I.B., P. Marchetti, and A.G. Livingston, Crosslinked polybenzimidazole membranes for

69.
AN
organic solvent nanofiltration (OSN): Analysis of crosslinking reaction mechanism and effects of
reaction parameters. Journal of Membrane Science, 2015. 493: p. 568-579.
Li, X., P. Vandezande, and I.F.J. Vankelecom, Polypyrrole modified solvent resistant nanofiltration
membranes. Journal of Membrane Science, 2008. 320(12): p. 143-150.
M
70. Jansen, J.C., et al., Influence of the blend composition on the properties and separation
performance of novel solvent resistant polyphenylsulfone/polyimide nanofiltration membranes.
Journal of Membrane Science, 2013. 447: p. 107-118.
D

71. Vanherck, K., G. Koeckelberghs, and I.F.J. Vankelecom, Crosslinking polyimides for membrane
applications: A review. Progress in Polymer Science, 2013. 38(6): p. 874-896.
72. Behnke, S. and M. Ulbricht, Two New Preparations for Organophilic Nanofiltration Membranes
TE

Based on Photocrosslinked Polyimide. Procedia Engineering, 2012. 44: p. 244-246.

73. Staudt-Bickel, C. and W. J. Koros, Improvement of CO2/CH4 separation characteristics of
polyimides by chemical crosslinking. Journal of Membrane Science, 1999. 155(1): p. 145-154.
EP

74. Ba, C., J. Langer, and J. Economy, Chemical modification of P84 copolyimide membranes by
polyethylenimine for nanofiltration. Journal of Membrane Science, 2009. 327(12): p. 49-58.
75. Dasgupta, J., et al., Remediation of textile effluents by membrane based treatment techniques: A
state of the art review. Journal of Environmental Management, 2015. 147: p. 55-72.
C

76. Chidambaram, T., Y. Oren, and M. Noel, Fouling of nanofiltration membranes by dyes during
brine recovery from textile dye bath wastewater. Chemical Engineering Journal, 2015. 262: p.
AC

156-168.
77. Bowen, W.R., T.A. Doneva, and H.-B. Yin, Separation of humic acid from a model surface water
with PSU/SPEEK blend UF/NF membranes. Journal of Membrane Science, 2002. 206(12): p.
417-429.
78. Zhang, Q., et al., Positively charged nanofiltration membrane based on cardo poly(arylene ether
sulfone) with pendant tertiary amine groups. Journal of Membrane Science, 2011. 375(12): p.
191-197.
79. Bolong, N., et al., Negatively charged polyethersulfone hollow fiber nanofiltration membrane for
the removal of bisphenol A from wastewater. Separation and Purification Technology, 2010.
73(2): p. 92-99.
 ACCEPTED MANUSCRIPT

80. Yu, L., et al., High flux, positively charged loose nanofiltration membrane by blending with poly
(ionic liquid) brushes grafted silica spheres. Journal of Hazardous Materials, 2015. 287: p. 373-
383.
81. Xing, L., et al., A negatively charged loose nanofiltration membrane by blending with poly
(sodium 4-styrene sulfonate) grafted SiO2 via SI-ATRP for dye purification. Separation and
Purification Technology, 2015. 146: p. 50-59.
82. Li, X., S. De Feyter, and I.F.J. Vankelecom, Poly(sulfone)/sulfonated poly(ether ether ketone)

PT
blend membranes: Morphology study and application in the filtration of alcohol based feeds.
Journal of Membrane Science, 2008. 324(12): p. 67-75.
83. Seman, M.N.A., M. Khayet, and N. Hilal, Comparison of two different UV-grafted nanofiltration

RI
membranes prepared for reduction of humic acid fouling using acrylic acid and N-
vinylpyrrolidone. Desalination, 2012. 287: p. 19-29.
84. Zhao, S., et al., Enhancing the performance of polyethylenimine modified nanofiltration
membrane by coating a layer of sulfonated poly(ether ether ketone) for removing sulfamerazine.

SC
Journal of Membrane Science, 2015. 492: p. 620-629.
85. Ba, C. and J. Economy, Preparation and characterization of a neutrally charged antifouling
nanofiltration membrane by coating a layer of sulfonated poly(ether ether ketone) on a

U
positively charged nanofiltration membrane. Journal of Membrane Science, 2010. 362(12): p.
192-201.
86.
AN
Ba, C., D.A. Ladner, and J. Economy, Using polyelectrolyte coatings to improve fouling resistance
of a positively charged nanofiltration membrane. Journal of Membrane Science, 2010. 347(12):
p. 250-259.
87. Buonomenna, M.G., et al., Polymeric membranes modified via plasma for nanofiltration of
M
aqueous solution containing organic compounds. Microporous and Mesoporous Materials, 2009.
120(12): p. 147-153.
88. Maruf, S.H., et al., Critical flux of surface-patterned ultrafiltration membranes during cross-flow
D

filtration of colloidal particles. Journal of Membrane Science, 2014. 471: p. 65-71.

89. Maruf, S.H., et al., Use of nanoimprinted surface patterns to mitigate colloidal deposition on
ultrafiltration membranes. Journal of Membrane Science, 2013. 428: p. 598-607.
TE

90. Boricha, A.G. and Z.V.P. Murthy, Preparation, characterization and performance of nanofiltration
membranes for the treatment of electroplating industry effluent. Separation and Purification
Technology, 2009. 65(3): p. 282-289.
EP

91. Peng, F., et al., Transport, structural, and interfacial properties of poly(vinyl alcohol)polysulfone
composite nanofiltration membranes. Journal of Membrane Science, 2010. 353(12): p. 169-
176.
92. Childs, R.F., et al., Nanofiltration using pore-filled membranes: effect of polyelectrolyte
C

composition on performance. Separation and Purification Technology, 2001. 2223: p. 507-517.

93. He, T., et al., Preparation and characterization of nanofiltration membranes by coating
AC

polyethersulfone hollow fibers with sulfonated poly(ether ether ketone) (SPEEK). Journal of
Membrane Science, 2008. 307(1): p. 62-72.
94. Perry, M. and T.I.L. Petach, EP1064073B1, Selective membrane and process for its preparation.
2006.
95. Linder, C., et al., US5265734, Silicon-derived solvent stable membranes. 1993.
96. Stafie, N., D.F. Stamatialis, and M. Wessling, Effect of PDMS cross-linking degree on the
permeation performance of PAN/PDMS composite nanofiltration membranes. Separation and
Purification Technology, 2005. 45(3): p. 220-231.
 ACCEPTED MANUSCRIPT

97. Gevers, L.E.M., I.F.J. Vankelecom, and P.A. Jacobs, Solvent-resistant nanofiltration with filled
polydimethylsiloxane (PDMS) membranes. Journal of Membrane Science, 2006. 278(12): p.
199-204.
98. Vanherck, K., et al., Hollow filler based mixed matrix membranes. Chem Commun (Camb), 2010.
46(14): p. 2492-4.
99. Cocchi, G., M.G. De Angelis, and F. Doghieri, Solubility and diffusivity of liquids for food and
pharmaceutical applications in crosslinked polydimethylsiloxane (PDMS) films: II. Experimental

PT
data on mixtures. Journal of Membrane Science, 2015. 492: p. 612-619.
100. Leitner, L., C. HarscoatSchiavo, and C. Vallires, Experimental contribution to the understanding
of transport through polydimethylsiloxanenanofiltration membranes: Influence of swelling,

RI
compaction and solvent on permeation properties. Polymer Testing, 2014. 33: p. 88-96.
101. Zeidler, S., U. Ktzel, and P. Kreis, Systematic investigation on the influence of solutes on the
separation behavior of a PDMS membrane in organic solvent nanofiltration. Journal of
Membrane Science, 2013. 429: p. 295-303.

SC
102. Ben Soltane, H., D. Roizard, and E. Favre, Effect of pressure on the swelling and fluxes of dense
PDMS membranes in nanofiltration: An experimental study. Journal of Membrane Science, 2013.
435: p. 110-119.

U
103. Gohil, J.M. and P. Ray, Polyvinyl alcohol as the barrier layer in thin film composite nanofiltration
membranes: Preparation, characterization, and performance evaluation. Journal of Colloid and

104.
AN
Interface Science, 2009. 338(1): p. 121-127.
Jahanshahi, M., A. Rahimpour, and M. Peyravi, Developing thin film composite poly(piperazine-
amide) and poly(vinyl-alcohol) nanofiltration membranes. Desalination, 2010. 257(1-3): p. 129-
136.
M
105. Peng, F., Z. Jiang, and E.M.V. Hoek, Tuning the molecular structure, separation performance and
interfacial properties of poly(vinyl alcohol)polysulfone interfacial composite membranes.
Journal of Membrane Science, 2011. 368(12): p. 26-33.
D

106. Liu, M., et al., Enhancing the permselectivity of thin-film composite poly(vinyl alcohol) (PVA)
nanofiltration membrane by incorporating poly(sodium-p-styrene-sulfonate) (PSSNa). Journal of
Membrane Science, 2014. 463: p. 173-182.
TE

107. Song, J., et al., Stabilization of composite hollow fiber nanofiltration membranes with a
sulfonated poly(ether ether ketone) coating. Desalination, 2015. 355: p. 83-90.
108. Zheng, Y., et al., Positively charged thin-film composite hollow fiber nanofiltration membrane for
EP

the removal of cationic dyes through submerged filtration. Desalination, 2013. 328: p. 42-50.
109. Zhang, Y., et al., Preparation and properties of novel pH-stable TFC membrane based on organic
inorganic hybrid composite materials for nanofiltration. Journal of Membrane Science, 2015.
476: p. 500-507.
C

110. Bano, S., et al., Chlorine resistant binary complexed NaAlg/PVA composite membrane for
nanofiltration. Separation and Purification Technology, 2014. 137: p. 21-27.
AC

111. Musale, D.A. and A. Kumar, Effects of surface crosslinking on sieving characteristics of
chitosan/poly(acrylonitrile) composite nanofiltration membranes. Separation and Purification
Technology, 2000. 21(12): p. 27-37.
112. Mu, T., et al., Preparation and characterization of novel chitosan composite nanofiltration
membrane containing mesogenic units. Desalination, 2012. 298: p. 67-74.
113. Miao, J., G.-h. Chen, and C.-j. Gao, A novel kind of amphoteric composite nanofiltration
membrane prepared from sulfated chitosan (SCS). Desalination, 2005. 181(13): p. 173-183.
114. Miao, J., et al., Amphoteric composite membranes for nanofiltration prepared from sulfated
chitosan crosslinked with hexamethylene diisocyanate. Chemical Engineering Journal, 2013. 234:
p. 132-139.
 ACCEPTED MANUSCRIPT

115. Miao, J., et al., Preparation and characterization of N,O-carboxymethyl chitosan/Polysulfone

composite nanofiltration membrane crosslinked with epichlorohydrin. Desalination, 2008. 233(1
3): p. 147-156.
116. Miao, J., et al., Preparation and characterization of N,O-carboxymethyl chitosan
(NOCC)/polysulfone (PS) composite nanofiltration membranes. Journal of Membrane Science,
2006. 280(12): p. 478-484.
117. Zhou, C., et al., Fabrication and characterization of novel composite nanofiltration membranes

PT
based on zwitterionic O-carboxymethyl chitosan. Desalination, 2013. 317: p. 67-76.
118. Dong, T.-T., G.-H. Chen, and C.-J. Gao, Preparation of chitin xanthate/polyacrylonitrile NF
composite membrane with cross-linking agent hydrogen peroxide and its characterization.

RI
Journal of Membrane Science, 2007. 304(12): p. 33-39.
119. Sun, H., et al., A novel composite nanofiltration (NF) membrane prepared from
glycolchitin/poly(acrylonitrile) (PAN) by epichlorohydrin cross-linking. Journal of Membrane
Science, 2007. 297(12): p. 51-58.

SC
120. Huang, R., et al., Preparation and characterization of composite NF membrane from a graft
copolymer of trimethylallyl ammonium chloride onto chitosan by toluene diisocyanate cross-
linking. Desalination, 2009. 239(13): p. 38-45.

U
121. Padaki, M., et al., New polypropylene supported chitosan NF-membrane for desalination
application. Desalination, 2011. 280(13): p. 419-423.
122.

123.
AN
Shenvi, S.S., et al., Preparation and characterization of PPEES/chitosan composite nanofiltration
membrane. Desalination, 2013. 315: p. 135-141.
Li, X.-l.Z., Li-ping; Jiang, Jin-hong; Yi, Zhuan; Zhu, Bao-ku; Xu, You-yi, Hydrophilic nanofiltration
membranes with self-polymerized and strongly-adhered polydopamine as separating layer.
M
Chinese Journal of Polymer Science, 2012. 30(2): p. 152-163.
124. Zhang, R., et al., A novel positively charged composite nanofiltration membrane prepared by bio-
inspired adhesion of polydopamine and surface grafting of poly(ethylene imine). Journal of
D

Membrane Science, 2014. 470: p. 9-17.

125. Li, M., et al., Bioinspired fabrication of composite nanofiltration membrane based on the
formation of DA/PEI layer followed by cross-linking. Journal of Membrane Science, 2014. 459: p.
TE

62-71.
126. Lv, Y., et al., Nanofiltration membranes via co-deposition of polydopamine/polyethylenimine
followed by cross-linking. Journal of Membrane Science, 2015. 476: p. 50-58.
EP

127. Perry, M., et al., WO2010082194, Solvent and acid stable membranes, methods of manufacture
thereof and methods of use thereof inter alia for separating metal ions from liquid process
streams. 2010.
128. Freger, V., et al., Characterization of novel acid-stable NF membranes before and after exposure
C

to acid using ATR-FTIR, TEM and AFM. Journal of Membrane Science, 2005. 256(12): p. 134-
142.
AC

129. Ikeda, K., S. Yamamoto, and H. Ito, US4818387, Sulfonated polysulfone composite
semipermeable membranes and process for producing the same. 1989.
130. Tomaschke, J.E., A.J. Testa, and J.G. Vouros, US4990252, Stable membranes from sulfonated
polyarylethers. 1991.
131. Ikeda, K., et al., New composite charged reverse osmosis membrane. Desalination, 1988. 68(2
3): p. 109-119.
132. Asatekin, A., et al., Antifouling nanofiltration membranes for membrane bioreactors from self-
assembling graft copolymers. Journal of Membrane Science, 2006. 285(12): p. 81-89.
 ACCEPTED MANUSCRIPT

133. Ji, Y.-L., et al., Novel composite nanofiltration membranes containing zwitterions with high
permeate flux and improved anti-fouling performance. Journal of Membrane Science, 2012.
390391: p. 243-253.
134. Gao, J., et al., Chelating polymer modified P84 nanofiltration (NF) hollow fiber membranes for
high efficient heavy metal removal. Water Research, 2014. 63: p. 252-261.
135. Bernstein, R., E. Antn, and M. Ulbricht, Tuning the nanofiltration performance of thin film
strong polyelectrolyte hydrogel composite membranes by photo-grafting conditions. Journal of

PT
Membrane Science, 2013. 427: p. 129-138.
136. Homayoonfal, M., A. Akbari, and M.R. Mehrnia, Preparation of polysulfone nanofiltration
membranes by UV-assisted grafting polymerization for water softening. Desalination, 2010.

RI
263(13): p. 217-225.
137. Behnke, S. and M. Ulbricht, Thin-film composite membranes for organophilic nanofiltration
based on photo-cross-linkable polyimide. Reactive and Functional Polymers, 2015. 86: p. 233-
242.

SC
138. Qiu, C., Q.T. Nguyen, and Z. Ping, Surface modification of cardo polyetherketone ultrafiltration
membrane by photo-grafted copolymers to obtain nanofiltration membranes. Journal of
Membrane Science, 2007. 295(12): p. 88-94.

U
139. Akbari, A., et al., Application of nanofiltration hollow fibre membranes, developed by
photografting, to treatment of anionic dye solutions. Journal of Membrane Science, 2007.

140.
297(12): p. 243-252.
AN
Zhong, P.S., et al., Positively charged nanofiltration (NF) membranes via UV grafting on
sulfonated polyphenylenesulfone (sPPSU) for effective removal of textile dyes from wastewater.
Journal of Membrane Science, 2012. 417418: p. 52-60.
M
141. Liu, F., et al., Positively charged loose nanofiltration membrane grafted by diallyl dimethyl
ammonium chloride (DADMAC) via UV for salt and dye removal. Reactive and Functional
Polymers, 2015. 86: p. 191-198.
D

142. Akbari, A., et al., New UV-photografted nanofiltration membranes for the treatment of colored
textile dye effluents. Journal of Membrane Science, 2006. 286(12): p. 342-350.
143. Li, X.-L., et al., A novel positively charged nanofiltration membrane prepared from N,N-
TE

dimethylaminoethyl methacrylate by quaternization cross-linking. Journal of Membrane Science,

2011. 374(12): p. 33-42.
144. Wang, X.-l., et al., Preparation and characterization of negatively charged hollow fiber
EP

nanofiltration membrane by plasma-induced graft polymerization. Desalination, 2012. 286: p.

138-144.
145. Chen, J., et al., Nanofiltration membrane prepared from polyacrylonitrile ultrafiltration
membrane by low-temperature plasma: 5. Grafting of styrene in vapor phase and its application.
C

Surface and Coatings Technology, 2007. 201(15): p. 6789-6792.

146. Xu, H.-m., J.-f. Wei, and X.-l. Wang, Nanofiltration hollow fiber membranes with high charge
AC

density prepared by simultaneous electron beam radiation-induced graft polymerization for

removal of Cr(VI). Desalination, 2014. 346: p. 122-130.
147. Guan, S., et al., Effect of additives on the performance and morphology of sulfonated copoly
(phthalazinone biphenyl ether sulfone) composite nanofiltration membranes. Applied Surface
Science, 2014. 295: p. 130-136.
148. Shao, L.-L., et al., Preparation and characterization of sulfated carboxymethyl cellulose
nanofiltration membranes with improved water permeability. Desalination, 2014. 338: p. 74-83.
149. Song, J., et al., Composite hollow fiber nanofiltration membranes for recovery of glyphosate from
saline wastewater. Water Research, 2013. 47(6): p. 2065-2074.
 ACCEPTED MANUSCRIPT

150. Zhang, S., X. Jian, and Y. Dai, Preparation of sulfonated poly(phthalazinone ether sulfone ketone)
composite nanofiltration membrane. Journal of Membrane Science, 2005. 246(2): p. 121-126.
151. Dalwani, M., et al., Sulfonated poly(ether ether ketone) based composite membranes for
nanofiltration of acidic and alkaline media. Journal of Membrane Science, 2011. 381(12): p. 81-
89.
152. Huang, R., et al., A novel composite nanofiltration (NF) membrane prepared from graft
copolymer of trimethylallyl ammonium chloride onto chitosan (GCTACC)/poly(acrylonitrile) (PAN)

PT
by epichlorohydrin cross-linking. Carbohydrate Research, 2006. 341(17): p. 2777-2784.
153. Huang, R., et al., Studies on nanofiltration membrane formed by diisocyanate cross-linking of
quaternized chitosan on poly(acrylonitrile) (PAN) support. Journal of Membrane Science, 2006.

RI
286(12): p. 237-244.
154. Ji, Y., et al., Preparation of novel positively charged copolymer membranes for nanofiltration.
Journal of Membrane Science, 2011. 376(12): p. 254-265.
155. Cui, Y., et al., Positively-charged nanofiltration membrane formed by quaternization and cross-

SC
linking of blend PVC/P(DMA-co-MMA) precursors. Journal of Membrane Science, 2015. 492: p.
187-196.
156. Tongwen, X. and Y. Weihua, A novel positively charged composite membranes for nanofiltration

U
prepared from poly(2,6-dimethyl-1,4-phenylene oxide) by in situ amines crosslinking. Journal of
Membrane Science, 2003. 215(12): p. 25-32.
157.

158.
AN
Gao, J., et al., Polyethyleneimine (PEI) cross-linked P84 nanofiltration (NF) hollow fiber
membranes for Pb2+ removal. Journal of Membrane Science, 2014. 452: p. 300-310.
Feng, C., et al., Studies on a novel nanofiltration membrane prepared by cross-linking of
polyethyleneimine on polyacrylonitrile substrate. Journal of Membrane Science, 2014. 451: p.
M
103-110.
159. Deng, H., et al., High flux positively charged nanofiltration membranes prepared by UV-initiated
graft polymerization of methacrylatoethyl trimethyl ammonium chloride (DMC) onto polysulfone
D

membranes. Journal of Membrane Science, 2011. 366(12): p. 363-372.

160. Decher, G., Fuzzy nanoassemblies - toward layered polymeric multicomposites. Science, 1997.
277: p. 1232-1327.
TE

161. Krasemann, L. and B. Tieke, Selective Ion Transport across Self-Assembled Alternating
Multilayers of Cationic and Anionic Polyelectrolytes. Langmuir, 2000. 16(2): p. 287-290.
162. Harris, J.J., J.L. Stair, and M.L. Bruening, Layered Polyelectrolyte Films as Selective, Ultrathin
EP

Barriers for Anion Transport. Chemistry of Materials, 2000. 12(7): p. 1941-1946.

163. Stanton, B.W., et al., Ultrathin, Multilayered Polyelectrolyte Films as Nanofiltration Membranes.
Langmuir, 2003. 19(17): p. 7038-7042.
164. Liu, C., L. Shi, and R. Wang, Crosslinked layer-by-layer polyelectrolyte nanofiltration hollow fiber
C

membrane for low-pressure water softening with the presence of SO42 in feed water. Journal of
Membrane Science, 2015. 486: p. 169-176.
AC

165. Shi, J., et al., Composite polyelectrolyte multilayer membranes for oligosaccharides nanofiltration
separation. Carbohydrate Polymers, 2013. 94(1): p. 106-113.
166. Deng, H.-Y., et al., Polyelectrolyte membranes prepared by dynamic self-assembly of poly (4-
styrenesulfonic acid-co-maleic acid) sodium salt (PSSMA) for nanofiltration (I). Journal of
Membrane Science, 2008. 323(1): p. 125-133.
167. Ahmadiannamini, P., et al., Multilayered polyelectrolyte complex based solvent resistant
nanofiltration membranes prepared from weak polyacids. Journal of Membrane Science, 2012.
394395: p. 98-106.
168. Rajabzadeh, S., et al., Preparation of low-pressure water softening hollow fiber membranes by
polyelectrolyte deposition with two bilayers. Desalination, 2014. 344: p. 64-70.
 ACCEPTED MANUSCRIPT

169. Su, B., et al., Preparation and performance of dynamic layer-by-layer PDADMAC/PSS
nanofiltration membrane. Journal of Membrane Science, 2012. 423424: p. 324-331.
170. Chen, Q., et al., High-flux composite hollow fiber nanofiltration membranes fabricated through
layer-by-layer deposition of oppositely charged crosslinked polyelectrolytes for dye removal.
Journal of Membrane Science, 2015. 492: p. 312-321.
171. El-Hashani, A., A. Toutianoush, and B. Tieke, Use of layer-by-layer assembled ultrathin
membranes of dicopper-[18]azacrown-N6 complex and polyvinylsulfate for water desalination

PT
under nanofiltration conditions. Journal of Membrane Science, 2008. 318(12): p. 65-70.
172. Jin, W., A. Toutianoush, and B. Tieke, Use of Polyelectrolyte Layer-by-Layer Assemblies as
Nanofiltration and Reverse Osmosis Membranes. Langmuir, 2003. 19(7): p. 2550-2553.

RI
173. Lajimi, R.H., et al., Change of the performance properties of nanofiltration cellulose acetate
membranes by surface adsorption of polyelectrolyte multilayers. Desalination, 2004. 163(13): p.
193-202.
174. Ahmadiannamini, P., et al., Multilayered PEC nanofiltration membranes based on SPEEK/PDDA

SC
for anion separation. Journal of Membrane Science, 2010. 360(12): p. 250-258.
175. Malaisamy, R., A. Talla-Nwafo, and K.L. Jones, Polyelectrolyte modification of nanofiltration
membrane for selective removal of monovalent anions. Separation and Purification Technology,

U
2011. 77(3): p. 367-374.
176. de Grooth, J., et al., The role of ionic strength and oddeven effects on the properties of

177.
p. 311-319.
AN
polyelectrolyte multilayer nanofiltration membranes. Journal of Membrane Science, 2015. 475:

Miller, M.D. and M.L. Bruening, Correlation of the Swelling and Permeability of Polyelectrolyte
Multilayer Films. Chemistry of Materials, 2005. 17(21): p. 5375-5381.
M
178. Ouyang, L., R. Malaisamy, and M.L. Bruening, Multilayer polyelectrolyte films as nanofiltration
membranes for separating monovalent and divalent cations. Journal of Membrane Science,
2008. 310(12): p. 76-84.
D

179. Shi, D., et al., Separation performance of polyimide nanofiltration membranes for concentrating
spiramycin extract. Desalination, 2006. 191(13): p. 309-317.
180. Saeki, D., et al., Stabilization of layer-by-layer assembled nanofiltration membranes by
TE

crosslinking via amide bond formation and siloxane bond formation. Journal of Membrane
Science, 2013. 447: p. 128-133.
181. Shan, W., et al., Polyelectrolyte multilayer films as backflushable nanofiltration membranes with
EP

tunable hydrophilicity and surface charge. Journal of Membrane Science, 2010. 349(12): p. 268-
278.
182. Ng, L.Y., et al., Development of nanofiltration membrane with high salt selectivity and
performance stability using polyelectrolyte multilayers. Desalination, 2014. 351: p. 19-26.
C

183. Kumar, M., et al., Highly permeable polymeric membranes based on the incorporation of the
functional water channel protein Aquaporin Z. Proc Natl Acad Sci USA, 2007. 104(52): p. 20719-
AC

24.
184. Tang, C., et al., Biomimetic aquaporin membranes coming of age. Desalination, 2015. 368: p. 89-
105.
185. Grzelakowski, M., et al., A framework for accurate evaluation of the promise of aquaporin based
biomimetic membranes. Journal of Membrane Science, 2015. 479: p. 223-231.
186. Shen, Y.-x., et al., Biomimetic membranes: A review. Journal of Membrane Science, 2014. 454: p.
359-381.
187. Tang, C.Y., et al., Desalination by biomimetic aquaporin membranes: Review of status and
prospects. Desalination, 2013. 308: p. 34-40.
 ACCEPTED MANUSCRIPT

188. Zhao, J., et al., Biomimetic and bioinspired membranes: Preparation and application. Progress in
Polymer Science, 2014. 39(9): p. 1668-1720.
189. Duong, P.H.H., et al., Planar biomimetic aquaporin-incorporated triblock copolymer membranes
on porous alumina supports for nanofiltration. Journal of Membrane Science, 2012. 409410: p.
34-43.
190. Zhong, P.S., et al., Aquaporin-embedded biomimetic membranes for nanofiltration. Journal of
Membrane Science, 2012. 407408: p. 27-33.

PT
191. Li, X., et al., Preparation of high performance nanofiltration (NF) membranes incorporated with
aquaporin Z. Journal of Membrane Science, 2014. 450: p. 181-188.
192. Sun, G., et al., Stabilization and immobilization of aquaporin reconstituted lipid vesicles for water

RI
purification. Colloids and Surfaces B: Biointerfaces, 2013. 102: p. 466-471.
193. Sun, G., et al., A layer-by-layer self-assembly approach to developing an aquaporin-embedded
mixed matrix membrane. RSC Adv., 2013. 3(2): p. 473-481.
194. Sun, G., et al., Highly permeable aquaporin-embedded biomimetic membranes featuring a

SC
magnetic-aided approach. RSC Advances, 2013. 3(24): p. 9178.
195. Zhao, Y., et al., Synthesis of robust and high-performance aquaporin-based biomimetic
membranes by interfacial polymerization-membrane preparation and RO performance

U
characterization. Journal of Membrane Science, 2012. 423424: p. 422-428.
196. Li, X., et al., Nature gives the best solution for desalination: Aquaporin-based hollow fiber

197.
68-77.
AN
composite membrane with superior performance. Journal of Membrane Science, 2015. 494: p.

Budd, P.M. and N.B. McKeown, Highly permeable polymers for gas separation membranes.
Polymer Chemistry, 2010. 1(1): p. 63.
M
198. Fritsch, D., et al., High performance organic solvent nanofiltration membranes: Development and
thorough testing of thin film composite membranes made of polymers of intrinsic microporosity
(PIMs). Journal of Membrane Science, 2012. 401402: p. 222-231.
D

199. Volkov, A.V., et al., High permeable PTMSP/PAN composite membranes for solvent
nanofiltration. Journal of Membrane Science, 2009. 333(12): p. 88-93.
200. Tsarkov, S., et al., Solvent nanofiltration through high permeability glassy polymers: Effect of
TE

polymer and solute nature. Journal of Membrane Science, 2012. 423424: p. 65-72.
201. Gagliardi, M., Global markets and technologies for nanofiltration. 2014, BCC Research:
Wellesley, MA.
EP

202. Schfer, A.F., A.; and Waite, T., Nanofiltration: Principles and Applications. 2004: Elsevier
Science.
203. Mark, H. and G.S. Whitby, eds. Collected Papers of Wallace Hume Carothers on High Polymeric
Substances. 1940, Wiley-Interscience, New York.
C

204. Wittbecker, E.L. and P.W. Morgan, Interfacial polycondensation. I. J. Polym. Sci., 1959. 40: p.
289-97.
AC

205. Wittbecker, E.L., Reflections on "Interfacial polycondensation. I," by Emerson L. Wittbecker and
Paul W. Morgan, J. Polym. Sci., XL, 289 (1959). J. Polym. Sci., Part A: Polym. Chem., 1996. 34(4):
p. 515-16.
206. Morgan, P.W. and S.L. Kwolek, Interfacial polycondensation. II. Fundamentals of polymer
formation at liquid interfaces. Journal of Polymer Science, 1959. 40(137): p. 299-327.
207. Morgan, P.W., Interfacial Polymerization, in Encyclopedia Of Polymer Science and Technology.
2011, John Wiley & Sons, Inc.
208. Carothers, W.H., US2130523, Linear polyamides and their production. 1938.
209. Lonsdale, H.K., S. United, and A. Gulf General, Research on improved reverse osmosis
membranes. United States. Office of Saline Water. Research and development progress
 ACCEPTED MANUSCRIPT

report,no. 577. 1970, Washington, D.C.: U.S. Dept. of the Interior, Office of Saline Water. vii, 143
p.
210. Scala, L., D.F. Ciliberti, and D. Berg, US3744642, Interface condensation desalination membranes.
1973, Westinghouse Electri: US.
211. Cadotte, J.E. and L.T. Rozelle, In-situ-formed condensation polymers for reverse osmosis
membranes, NTIS Report No. PB-229337. 1972.
212. Cadotte, J.E., US Patent, US4039440A, Reverse osmosis membrane. 1977, . p. 6

PT
213. Rozelle, L.T., et al., eds. Nonpolysaccharide membranes for reverse osmosis: NS-100 membranes.
Reverse Osmosis and Synthetic Membranes, ed. S. Sourirajan. 1977, National Resource Council
Canada: Ottawa. 249-61.

RI
214. Cadotte, J.E., et al., Continued evaluation of in situ-formed condensation polymers for reverse
osmosis membranes, NTIS Report No. PB-253193. 1976, Midwest Res. Inst. p. 89.
215. Wrasidlo, W.J., US4005012, Semipermeable membranes and the method for the preparation
thereof. 1977.

SC
216. Riley, R.L., et al., Spiral-wound poly(ether/amide) thin-film composite membrane systems.
Desalination, 1976. 19(1-2-3): p. 113-26.
217. Riley, P.L., et al., Spiral-wound thin-film composite membrane systems for brackish and seawater

U
desalination by reverse osmosis. Desalination, 1977. 23(13): p. 331-355.
218. Naaktgeboren, A.J., et al., EP174045A1, Semipermeable composite membrane. 1986. p. 15.
219.

220.
AN
Kawaguchi, T., et al., US4242208A, Semipermeable composite membrane and process for
preparation thereof. 1979, Teijin Ltd., Japan . p. 114.
Hurndall, M.J., E.P. Jacobs, and R.D. Sanderson, Chemical composition of thin-film composite
reverse osmosis membranes made from poly(2-vinylimidazoline). J. Membr. Sci., 1993. 78(3): p.
M
283-98.
221. Hurndall, M.J., E.P. Jacobs, and R.D. Sanderson, New composite reverse osmosis membranes
made from poly(2-vinylimidazoline). J. Appl. Polym. Sci., 1992. 46(3): p. 523-9.
D

222. Hurndall, M.J., E.P. Jacobs, and R.D. Sanderson, The performance of novel reverse osmosis
membranes made from poly-2-vinylimidazoline. I. Desalination, 1992. 86(2): p. 135-54.
223. Hurndall, M.J., et al., Modified poly-2-vinylimidazoline reverse osmosis membranes. II.
TE

Desalination, 1992. 89(2): p. 203-22.

224. Hurndall, M.J., et al., Poly(2-vinylimidazoline) composite reverse osmosis membranes.
Desalination, 1993. 90(1-3): p. 41-54.
EP

225. Credali, L., A. Chiolle, and P. Parrini, New polymer materials for reverse osmosis membranes.
Desalination, 1974. 14(2): p. 137-50.
226. Cadotte, J.E., M.J. Steuck, and R.J. Petersen, Research on 'in situ'-formed condensation polymer
for reverse osmosis membranes, NTIS.gov PB-288 387/4. 1978, Midwest Res. Inst. p. 50.
C

227. Cadotte, J.E., US4259183A, Reverse osmosis membrane. 1981.

228. Cadotte, J.E., US4277344, Interfacially synthesized reverse osmosis membrane and its use in
AC

removing solute from solute-containing water. 1980.

229. Cadotte, J.E., US4895661, Alkali resistant hyperfiltration membrane. 1990.
230. Shinjou, I. and R. Shoji, US4795559A, Semipermeable membrane support. 1989.
231. Vilakati, G.D., et al., Relating thin film composite membrane performance to support membrane
morphology fabricated using lignin additive. Journal of Membrane Science, 2014. 469: p. 216-
224.
232. Knappe, P.H., et al., US6755970, Back-flushable spiral wound filter and methods of making and
using same. 2004.
233. Verssimo, S., K.V. Peinemann, and J. Bordado, New composite hollow fiber membrane for
nanofiltration. Desalination, 2005. 184(13): p. 1-11.
 ACCEPTED MANUSCRIPT

234. Yang, F.J., et al., Synthesis of polypiperazine-amide thin-film membrane on PPESK hollow fiber UF
membrane. Chinese Chemical Letters, 2007. 18(8): p. 966-968.
235. Liu, T.-Y., et al., Fabrication of a high-flux thin film composite hollow fiber nanofiltration
membrane for wastewater treatment. Journal of Membrane Science, 2015. 478: p. 25-36.
236. Liu, J.-Q., et al., An improved process to prepare high separation performance PA/PVDF hollow
fiber composite nanofiltration membranes. Separation and Purification Technology, 2007. 58(1):
p. 53-60.

PT
237. Koenhen, D.M., US6787216, Method for manufacturing multiple channel membranes, multiple
channel membranes and the use thereof in separation methods. 2004.
238. Billovits, G.F., et al., US8911625, Spiral wound module including membrane sheet with capillary

RI
channels. 2014.
239. Spruck, M., et al., Preparation and characterization of composite multichannel capillary
membranes on the way to nanofiltration. Desalination, 2013. 314: p. 28-33.
240. Sun, S.-P., S.-Y. Chan, and T.-S. Chung, A slowfast phase separation (SFPS) process to fabricate

SC
dual-layer hollow fiber substrates for thin-film composite (TFC) organic solvent nanofiltration
(OSN) membranes. Chemical Engineering Science, 2015. 129: p. 232-242.
241. Shuji, f.H., k Shimura; Kiyohiko, Takaya; Masakazu, Koiwa; Mashiro, Kimura, JP 2014144441A,

U
Composite semipermeable membrane. 2014, Toray Ind. Inc.: Japn.
242. Tang, M.W., US20120085697, Cast-on-tricot asymmetric and composite separation membranes.

243.
2012.
AN
Hirozawa, H., et al., US20140151286, Separation membrane, separation membrane element,
and method for producing separation membrane. 2014.
244. Singh, P.S., et al., Probing the structural variations of thin film composite RO membranes
M
obtained by coating polyamide over polysulfone membranes of different pore dimensions.
Journal of Membrane Science, 2006. 278(12): p. 19-25.
245. Ghosh, A.K. and E.M.V. Hoek, Impacts of support membrane structure and chemistry on
D

polyamidepolysulfone interfacial composite membranes. Journal of Membrane Science, 2009.

336(12): p. 140-148.
246. Lu, X., et al., Influence of active layer and support layer surface structures on organic fouling
TE

propensity of thin-film composite forward osmosis membranes. Environ Sci Technol, 2015. 49(3):
p. 1436-44.
247. Li, X., et al., Thin-Film Composite Membranes and Formation Mechanism of Thin-Film Layers on
EP

Hydrophilic Cellulose Acetate Propionate Substrates for Forward Osmosis Processes. Industrial &
Engineering Chemistry Research, 2012. 51(30): p. 10039-10050.
248. Ramon, G.Z., M.C.Y. Wong, and E.M.V. Hoek, Transport through composite membrane, part 1: Is
there an optimal support membrane? Journal of Membrane Science, 2012. 415416: p. 298-305.
C

249. Ramon, G.Z. and E.M.V. Hoek, Transport through composite membranes, part 2: Impacts of
roughness on permeability and fouling. Journal of Membrane Science, 2013. 425426: p. 141-
AC

148.
250. Misdan, N., et al., Formation of thin film composite nanofiltration membrane: Effect of
polysulfone substrate characteristics. Desalination, 2013. 329: p. 9-18.
251. Misdan, N., et al., Study on the thin film composite poly(piperazine-amide) nanofiltration
membrane: Impacts of physicochemical properties of substrate on interfacial polymerization
formation. Desalination, 2014. 344: p. 198-205.
252. Zhu, S., et al., Improved performance of polyamide thin-film composite nanofiltration membrane
by using polyetersulfone/polyaniline membrane as the substrate. Journal of Membrane Science,
2015. 493: p. 263-274.
 ACCEPTED MANUSCRIPT

253. Jimenez-Solomon, M.F., et al., Beneath the surface: Influence of supports on thin film composite
membranes by interfacial polymerization for organic solvent nanofiltration. Journal of
Membrane Science, 2013. 448: p. 102-113.
254. Maruf, S.H., et al., Fabrication and characterization of a surface-patterned thin film composite
membrane. Journal of Membrane Science, 2014. 452: p. 11-19.
255. Persson, K.M., V. Gekas, and G. Trgrdh, Study of membrane compaction and its influence on
ultrafiltration water permeability. Journal of Membrane Science, 1995. 100(2): p. 155-162.

PT
256. Chakrabarty, B., A.K. Ghoshal, and M.K. Purkait, Preparation, characterization and performance
studies of polysulfone membranes using PVP as an additive. Journal of Membrane Science, 2008.
315(12): p. 36-47.

RI
257. Pendergast, M.T.M., et al., Using nanocomposite materials technology to understand and
control reverse osmosis membrane compaction. Desalination, 2010. 261(3): p. 255-263.
258. Hoek, E.M.V., A.K. Ghosh, and J.M. Nygaard, US8029857, Micro-and nanocomposite support
structures for reverse osmosis thin film membranes. 2011.

SC
259. Pendergast, M.M., A.K. Ghosh, and E.M.V. Hoek, Separation performance and interfacial
properties of nanocomposite reverse osmosis membranes. Desalination, 2013. 308: p. 180-185.
260. Pan, K., P. Fang, and B. Cao, Novel composite membranes prepared by interfacial polymerization

U
on polypropylene fiber supports pretreated by ozone-induced polymerization. Desalination, 2012.
294: p. 36-43.
261.
AN
Yu, C.-H., et al., PTFE/polyamide thin-film composite membranes using PTFE films modified with
ethylene diamine polymer and interfacial polymerization: Preparation and pervaporation
application. Journal of Colloid and Interface Science, 2009. 336(1): p. 260-267.
262. Song, Y., F. Liu, and B. Sun, Preparation, characterization, and application of thin film composite
M
nanofiltration membranes. Journal of Applied Polymer Science, 2005. 95(5): p. 1251-1261.
263. Jimenez Solomon, M.F., Y. Bhole, and A.G. Livingston, High flux hydrophobic membranes for
organic solvent nanofiltration (OSN)Interfacial polymerization, surface modification and
D

solvent activation. Journal of Membrane Science, 2013. 434: p. 193-203.

264. Hermans, S., et al., Efficient synthesis of interfacially polymerized membranes for solvent
resistant nanofiltration. Journal of Membrane Science, 2015. 476: p. 356-363.
TE

265. Aburabie, J., et al., Thin-film composite crosslinked polythiosemicarbazide membranes for
organic solvent nanofiltration (OSN). Reactive and Functional Polymers, 2015. 86: p. 225-232.
266. Prez-Manrquez, L., et al., Cross-linked PAN-based thin-film composite membranes for non-
EP

aqueous nanofiltration. Reactive and Functional Polymers, 2015. 86: p. 243-247.

267. Hu, L., et al., Preparation and performance of novel thermally stable polyamide/PPENK
composite nanofiltration membranes. Applied Surface Science, 2012. 258(22): p. 9047-9053.
268. Han, R., et al., Preparation and characterization of thermally stable poly(piperazine
C

amide)/PPBES composite nanofiltration membrane. Journal of Membrane Science, 2011. 370(1

2): p. 91-96.
AC

269. Yang, F., et al., Preparation and characterization of polypiperazine amide/PPESK hollow fiber
composite nanofiltration membrane. Journal of Membrane Science, 2007. 301(12): p. 85-92.
270. Wu, C., et al., Preparation, characterization and application of a novel thermal stable composite
nanofiltration membrane. Journal of Membrane Science, 2009. 326(2): p. 429-434.
271. Liang, S., et al., Annealing of supporting layer to develop nanofiltration membrane with high
thermal stability and ion selectivity. Journal of Membrane Science, 2015. 476: p. 475-482.
272. Heffernan, R., et al., Disinfection of a polyamide nanofiltration membrane using ethanol. Journal
of Membrane Science, 2013. 448: p. 170-179.
 ACCEPTED MANUSCRIPT

273. Geens, J., et al., Polymeric nanofiltration of binary wateralcohol mixtures: Influence of feed
composition and membrane properties on permeability and rejection. Journal of Membrane
Science, 2005. 255(12): p. 255-264.
274. Shukla, R. and M. Cheryan, Performance of ultrafiltration membranes in ethanolwater
solutions: effect of membrane conditioning. Journal of Membrane Science, 2002. 198(1): p. 75-
85.
275. Kosaraju, P.B. and K.K. Sirkar, Interfacially polymerized thin film composite membranes on

PT
microporous polypropylene supports for solvent-resistant nanofiltration. Journal of Membrane
Science, 2008. 321(2): p. 155-161.
276. Oh, N.-W.J., Jegal, Lee, Kew-Ho, Preparation and Characterization of Nanofiltration Composite

RI
Membranes Using Polyacrylonitrile (PAN). II. Preparation and Characterization of Polyamide
Composite Membranes. J. Appl. Polym. Sci. , 2001. 80 p. 2729-2736.
277. Peng, J., et al., Polyamide nanofiltration membrane with high separation performance prepared
by EDC/NHS mediated interfacial polymerization. Journal of Membrane Science, 2013. 427: p.

SC
92-100.
278. Ahmed, F.E., B.S. Lalia, and R. Hashaikeh, A review on electrospinning for membrane fabrication:
Challenges and applications. Desalination, 2015. 356: p. 15-30.

U
279. Subramanian, S. and R. Seeram, New directions in nanofiltration applications Are nanofibers
the right materials as membranes in desalination? Desalination, 2013. 308: p. 198-208.
280.

281.
AN
Wang, X., et al., Nanofiltration membranes based on thin-film nanofibrous composites. Journal
of Membrane Science, 2014. 469: p. 188-197.
Bui, N.-N., et al., Electrospun nanofiber supported thin film composite membranes for engineered
osmosis. Journal of Membrane Science, 2011. 385386: p. 10-19.
M
282. Yoon, K., B.S. Hsiao, and B. Chu, High flux nanofiltration membranes based on interfacially
polymerized polyamide barrier layer on polyacrylonitrile nanofibrous scaffolds. Journal of
Membrane Science, 2009. 326(2): p. 484-492.
D

283. Yung, L., et al., Fabrication of thin-film nanofibrous composite membranes by interfacial
polymerization using ionic liquids as additives. J. Membr. Sci., 2010. 365(1-2): p. 52-58.
284. Kaur, S., et al., Hot pressing of electrospun membrane composite and its influence on separation
TE

performance on thin film composite nanofiltration membrane. Desalination, 2011. 279(1-3): p.

201-209.
285. Kaur, S., et al., Influence of electrospun fiber size on the separation efficiency of thin film
EP

nanofiltration composite membrane. Journal of Membrane Science, 2012. 392393: p. 101-111.

286. Li, Y., et al., Preparation of thin film composite nanofiltration membrane with improved
structural stability through the mediation of polydopamine. Journal of Membrane Science, 2015.
476: p. 10-19.
C

287. Zhang, Y., et al., Appearance of poly(ethylene oxide) segments in the polyamide layer for
antifouling nanofiltration membranes. Journal of Membrane Science, 2011. 382(12): p. 300-
AC

307.
288. Hachisuka, H. and K. Ikeda, US6177011, Composite reverse osmosis membrane having a
separation layer with polyvinyl alcohol coating and method of reverse osmotic treatment of
water using the same. 2001.
289. Mickols, W.E., US6280853, Composite membrane with polyalkylene oxide modified polyamide
surface. 2001.
290. Koo, J., et al., US6913694, Selective membrane having a high fouling resistance. 2005.
291. Niu, Q.J., W.E. Mickols, and C. Zhang, US7905361, Modified polyamide membrane. 2011.
292. Haynes, T.N. and I.A.R. Marsh, WO2014158660A1, Composite polyamide membrane including
dissolvable polymer coating. 2014.
 ACCEPTED MANUSCRIPT

293. Hiep, T.H. and M.J. Fabig, WO2015139073A1, Membranes comprising sacrificial coating. 2015.
294. Tang, C.Y., Y.-N. Kwon, and J.O. Leckie, Effect of membrane chemistry and coating layer on
physiochemical properties of thin film composite polyamide RO and NF membranes: I. FTIR and
XPS characterization of polyamide and coating layer chemistry. Desalination, 2009. 242(13): p.
149-167.
295. Tang, C.Y., Y.-N. Kwon, and J.O. Leckie, Effect of membrane chemistry and coating layer on
physiochemical properties of thin film composite polyamide RO and NF membranes: II.

PT
Membrane physiochemical properties and their dependence on polyamide and coating layers.
Desalination, 2009. 242(13): p. 168-182.
296. Draevi, E., et al., Coating layer effect on performance of thin film nanofiltration membrane in

RI
removal of organic solutes. Separation and Purification Technology, 2013. 118: p. 530-539.
297. Akbari, A., Z. Derikvandi, and S.M. Mojallali Rostami, Influence of chitosan coating on the
separation performance, morphology and anti-fouling properties of the polyamide nanofiltration
membranes. Journal of Industrial and Engineering Chemistry, 2015. 28: p. 268-276.

SC
298. Zhou, Y., et al., Surface modification of polypiperazine-amide membrane by self-assembled
method for dye wastewater treatment. Chinese Journal of Chemical Engineering, 2015. 23(6): p.
912-918.

U
299. Gupta, K.C., Synthesis and evaluation of aromatic polyamide membranes for desalination in
reverse-osmosis technique. J. Appl. Polym. Sci., 1997. 66(4): p. 643-653.
300.
AN
Kim, J.-j., C.-k. Kim, and S.-y. Kwak, Composite reverse osmosis membrane having active layer of
aromatic polyester or copolymer of aromatic polyester and aromatic polyamide. 1997, Korea
Institute of Science and Technology, S. Korea . p. 6 pp.
301. Ahmad, A.L., B.S. Ooi, and J.P. Choudhury, Preparation and characterization of co-polyamide thin
M
film composite membrane from piperazine and 3,5-diaminobenzoic acid. Desalination, 2003.
158(13): p. 101-108.
302. Ahmad, A.L., et al., Development of a highly hydrophilic nanofiltration membrane for
D

desalination and water treatment. Desalination, 2004. 168: p. 215-221.

303. Ahmad, A.L., B.S. Ooi, and J.P. Choudhury, Effect of Hydrophilization Additive and Reaction Time
on Separation Properties of Polyamide Nanofiltration Membrane. Sep. Sci. Technol., 2004. 39(8):
TE

p. 1815-1831.
304. Ahmad, A.L., et al., Composite Nanofiltration Polyamide Membrane: A Study on the Diamine
Ratio and Its Performance Evaluation. Ind. Eng. Chem. Res., 2004. 43(25): p. 8074-8082.
EP

305. Yong, Z., et al., Polyamide thin film composite membrane prepared from m-phenylenediamine
and m-phenylenediamine-5-sulfonic acid. Journal of membrane science, 2006. 270(1): p. 162-
168.
306. Baroa, G.N.B., J. Lim, and B. Jung, High performance thin film composite polyamide reverse
C

osmosis membrane prepared via m-phenylenediamine and 2,2-benzidinedisulfonic acid.

Desalination, 2012. 291: p. 69-77.
AC

307. Chen, G., et al., Novel thin-film composite membranes with improved water flux from sulfonated
cardo poly (arylene ether sulfone) bearing pendant amino groups. Journal of Membrane Science,
2008. 310(1): p. 102-109.
308. Liu, J., et al., A novel method for fabricating composite mosaic membrane with unique NF
selectivity. China Particuol., 2006. 4(2): p. 98-102.
309. Liu, Y., et al., Novel sulfonated thin-film composite nanofiltration membranes with improved
water flux for treatment of dye solutions. Journal of Membrane Science, 2012. 394395: p. 218-
229.
310. Akbari, A., et al., Novel sulfonated polyamide thin-film composite nanofiltration membranes with
improved water flux and anti-fouling properties. Desalination, 2016. 377: p. 11-22.
 ACCEPTED MANUSCRIPT

311. Jin, J., et al., Taurine as an additive for improving the fouling resistance of nanofiltration
composite membranes. J. Appl. Polym. Sci., 2015. 132(11): p. 41620/1-41620/7.
312. Li, X., et al., Preparation and characterization of positively charged polyamide composite
nanofiltration hollow fiber membrane for lithium and magnesium separation. Desalination,
2015. 369: p. 26-36.
313. Jewrajka, S.K., et al., Use of 2,4,6-pyridinetricarboxylic acid chloride as a novel co-monomer for
the preparation of thin film composite polyamide membrane with improved bacterial resistance.

PT
Journal of Membrane Science, 2013. 439: p. 87-95.
314. An, Q., et al., CN102294178, Polyamide nanofiltration membrane containing amphoteric ion and
its preparation method. 2011.

RI
315. Li, L., S. Zhang, and X. Zhang, Preparation and characterization of poly(piperazineamide)
composite nanofiltration membrane by interfacial polymerization of 3,3,5,5-biphenyl tetraacyl
chloride and piperazine. Journal of Membrane Science, 2009. 335(12): p. 133-139.
316. Paul, M., et al., US20130287944, Composite Polyamide Membrane. 2013.

SC
317. Roy, A., et al., US8968828, Composite polyamide membrane. 2015.
318. Jons, S.D., et al., US20130287946, Composite Polyamide Membrane. 2013.
319. Kurth, C.J., et al., US8603340B2, Hybrid nanoparticle TFC membranes, E.S.C.A.U.S. NanoH2O Inc,

U
et al., Editors. 2013: US.
320. Rosenberg, S., et al., US9073015, Composite polyamide membrane. 2015.
321.

322.
AN
Zhang, C., et al., US20140170314, Composite polyamide membrane derived from monomer
including amine-reactive and phosphorous-containing functional groups. 2014.
Murakami Mutsuo, S.J.P., K.J.P. Tateishi Yasushi, and S.J.P. Fusaoka Yoshinari, Composite
semipermeable membrane, processfor producing the same, and method of purifying water with
M
the same, T.J.P. Toray Industries Inc, Editor. 2002: US.
323. Jin, J.-b., et al., Preparation of thin-film composite nanofiltration membranes with improved
antifouling property and flux using 2,2'-oxybis-ethylamine. Desalination, 2015. 355: p. 141-146.
D

324. Kwak, S.-Y., et al., Correlations of chemical structure, atomic force microscopy (AFM)
morphology, and reverse osmosis (RO) characteristics in aromatic polyester high-flux RO
membranes. J. Membr. Sci., 1997. 132(2): p. 183-191.
TE

325. Jayarani, M.M. and S.S. Kulkarni, Thin-film composite poly(esteramide)-based membranes.
Desalination, 2000. 130(1): p. 17-30.
326. Jayarani, M.M., et al., Synthesis of model diamide, diester and esteramide adducts and studies
EP

on their chlorine tolerance. Desalination, 2000. 130(1): p. 1-16.

327. Razdan, U. and S.S. Kulkarni, Nanofiltration thin-film-composite polyesteramide membranes
based on bulky diols. Desalination, 2004. 161(1): p. 25-32.
328. Abu Seman, M.N., M. Khayet, and N. Hilal. Development of antifouling properties and
C

performance of nanofiltration membranes by interfacial polymerization and photografting

techniques. 2012. CRC Press.
AC

329. Abu Seman, M.N., M. Khayet, and N. Hilal, Development of antifouling properties and
performance of nanofiltration membranes modified by interfacial polymerization. Desalination,
2011. 273(1): p. 36-47.
330. Abu Seman, M.N., M. Khayet, and N. Hilal, Nanofiltration thin-film composite polyester
polyethersulfone-based membranes prepared by interfacial polymerization. J. Membr. Sci., 2010.
348(1-2): p. 109-116.
331. Tang, B., Z. Huo, and P. Wu, Study on a novel polyester composite nanofiltration membrane by
interfacial polymerization of triethanolamine (TEOA) and trimesoyl chloride (TMC). J. Membr.
Sci., 2008. 320(1+2): p. 198-205.
 ACCEPTED MANUSCRIPT

332. Tang, B. and P. Wu, Fabrication of a polyester composite nanofiltration membrane. 2008, Fudan
University, Peop. Rep. China . p. 7pp.
333. Wu, H., B. Tang, and P. Wu, Preparation and characterization of anti-fouling -
cyclodextrin/polyester thin film nanofiltration composite membrane. J. Membr. Sci., 2013. 428:
p. 301-308.
334. Wu, H., B. Tang, and P. Wu, Antifouling cyclodextrin-polymer composite nanofiltration
membrane and its preparation method. 2012, Fudan University, Peop. Rep. China . p. 6pp.

PT
335. Colquhoun, H.M., et al., Hydrophilic poly-ylids derived from 4,4-bipyridyl: synthesis, structure
and membrane-forming characteristics. Polymer, 1995. 36(2): p. 443-446.
336. Petersen, R.J., P. K. Eriksson, and J. E. Cadotte. , Highly permeable reverse osmosis membranes.

RI
Frontiers of Macromolecular Science, 1989: p. 511.
337. Verissimo, S., K.V. Peinemann, and J. Bordado, Influence of the diamine structure on the
nanofiltration performance, surface morphology and surface charge of the composite polyamide
membranes. J. Membr. Sci., 2006. 279((1-2)): p. 266-275.

SC
338. Saha, N.K. and S.V. Joshi, Performance evaluation of thin film composite polyamide
nanofiltration membrane with variation in monomer type. J. Membr. Sci., 2009. 342(1-2): p. 60-
69.

U
339. Li, Y., et al., Separation performance of thin-film composite nanofiltration membrane through
interfacial polymerization using different amine monomers. Desalination, 2014. 333(1): p. 59-65.
340.
341.
342.
AN
McCray, S.B., US4853122, p-xylylenediamide/diimide composite RO membranes. 1989.
McCray, S.B., US4876009, Tetrakis-amido high flux membranes. 1989.
Tomaschke, J.E., US6464873, Interfacial polymerization of bipiperidine-polyamide membranes
for reverse osmosis, nanofiltration, and seawater desalination. 2000, Hydranautics, USA .
M
343. Chen, G.-E., et al., Fabrication and characterization of a novel nanofiltration membrane by the
interfacial polymerization of 1,4-diaminocyclohexane (DCH) and trimesoyl chloride (TMC). RSC
Adv., 2015. 5(51): p. 40742-40752.
D

344. Tomaschke, J.E., Amine monomers and their use in preparing interfacially synthesized polyamide
membranes for reverse osmosis and nanofiltration. 1999, Hydranautics, USA . p. 11 pp.
345. Chen, G.-E., et al., Preparation and characterization of a composite nanofiltration membrane
TE

from cyclen and trimesoyl chloride prepared by interfacial polymerization. J. Appl. Polym. Sci.,
2015. 132(33).
346. Singh, K., et al., Optical Resolution of Racemic Mixtures of Amino Acids through Nanofiltration
EP

Membrane Process. Sep. Sci. Technol. (Philadelphia, PA, U. S.), 2014. 49(17): p. 2630-2641.
347. Singh, K., et al., Optical Resolution of -Amino Acids by Reverse Osmosis using Enantioselective
Polymer Membrane Containing Chiral Metal-Schiff Base Complex. Sep. Sci. Technol., 2010.
45(10): p. 1374-1384.
C

348. Singh, K., et al., Resolution of racemic mixture of -amino acid derivative through composite
membrane. J. Membr. Sci., 2011. 378(1-2): p. 531-540.
AC

349. Qian, H., J. Zheng, and S. Zhang, Preparation of microporous polyamide networks for carbon
dioxide capture and nanofiltration. Polymer, 2013. 54(2): p. 557-564.
350. Li, L., et al., Polyamide thin film composite membranes prepared from 3,4,5-biphenyl triacyl
chloride, 3,3,5,5-biphenyl tetraacyl chloride and m-phenylenediamine. Journal of Membrane
Science, 2007. 289(12): p. 258-267.
351. Li, L., et al., Polyamide thin film composite membranes prepared from isomeric biphenyl tetraacyl
chloride and m-phenylenediamine. J. Membr. Sci., 2008. 315(1+2): p. 20-27.
352. Wang, H., Q. Zhang, and S. Zhang, Positively charged nanofiltration membrane formed by
interfacial polymerization of 3,3',5,5'-biphenyl tetraacyl chloride and piperazine on a
poly(acrylonitrile) (PAN) support. J. Membr. Sci., 2011. 378(1-2): p. 243-249.
 ACCEPTED MANUSCRIPT

353. Wang, T., et al., A novel highly permeable positively charged nanofiltration membrane based on
a nanoporous hyper-crosslinked polyamide barrier layer. J. Membr. Sci., 2013. 448: p. 180-189.
354. Glater, J. and M.R. Zachariah, A mechanistic study of halogen interaction with polyamide
reverse-osmosis membranes. ACS Symposium Series, 1985. 281(Reverse Osmosis Ultrafiltr.): p.
345-58.
355. Glater, J., S.-k. Hong, and M. Elimelech, The search for a chlorine-resistant reverse osmosis
membrane. Desalination, 1994. 95(3): p. 325-45.

PT
356. Glater, J., et al., Halogen interactions with typical reverse osmosis membranes. Proc. Water
Reuse Symp., 1982(2): p. 1399-409.
357. Glater, J., et al., Reverse osmosis membrane sensitivity to ozone and halogen disinfectants.

RI
Desalination, 1983. 48(1): p. 1-16.
358. Kawaguchi, T. and H. Tamura, Chlorine-resistant membrane for reverse osmosis. I. Correlation
between chemical structures and chlorine resistance of polyamides. Journal of Applied Polymer
Science, 1984. 29(11): p. 3359-3367.

SC
359. Soice, N.P., et al., Oxidative degradation of polyamide reverse osmosis membranes: Studies of
molecular model compounds and selected membranes. Journal of Applied Polymer Science,
2003. 90(5): p. 1173-1184.

U
360. Soice, N.P., et al., Studies of oxidative degradation in polyamide RO membrane barrier layers
using pendant drop mechanical analysis. Journal of Membrane Science, 2004. 243(1-2): p. 345-

361.
355.
AN
Kwon, Y.-N. and J.O. Leckie, Hypochlorite degradation of crosslinked polyamide membranes.
Journal of Membrane Science, 2006. 282(1+2): p. 456-464.
362. Lee, J.-H., et al., Correlating chlorine-induced changes in mechanical properties to performance
M
in polyamide-based thin film composite membranes. Journal of Membrane Science, 2013. 433: p.
72-79.
363. Parrini, P., Polypiperazinamides: new polymers useful for membrane processes. Desalination,
D

1983. 48(1): p. 67-78.

364. Larson, R.E., J.E. Cadotte, and R.J. Petersen, The FT-30 seawater reverse osmosis membrane--
element test results. Desalination, 1981. 38: p. 473-483.
TE

365. Konagaya, S. and O. Watanabe, Influence of chemical structure of isophthaloyl dichloride and
aliphatic, cycloaliphatic, and aromatic diamine compound polyamides on their chlorine
resistance. J. Appl. Polym. Sci., 2000. 76(2): p. 201-207.
EP

366. Konagaya, S., H. Kuzumoto, and O. Watanabe, New reverse osmosis membrane materials with
higher resistance to chlorine. J. Appl. Polym. Sci., 2000. 75(11): p. 1357-1364.
367. Shintani, T., H. Matsuyama, and N. Kurata, Development of a chlorine-resistant polyamide
reverse osmosis membrane. Desalination, 2007. 207(13): p. 340-348.
C

368. Shintani, T., et al., Development of a chlorine-resistant polyamide nanofiltration membrane and
its field-test results. J. Appl. Polym. Sci., 2007. 106(6): p. 4174-4179.
AC

369. Buch, P.R., D.J. Mohan, and A.V.R. Reddy, Preparation, characterization and chlorine stability of
aromatic-cycloaliphatic polyamide thin film composite membranes. J. Membr. Sci., 2008. 309(1-
2): p. 36-44.
370. Yu, S., et al., Aromatic-cycloaliphatic polyamide thin-film composite membrane with improved
chlorine resistance prepared from m-phenylenediamine-4-methyl and cyclohexane-1,3,5-
tricarbonyl chloride. Journal of Membrane Science, 2009. 344(12): p. 155-164.
371. La, Y.-H., et al., Novel thin film composite membrane containing ionizable hydrophobes: pH-
dependent reverse osmosis behavior and improved chlorine resistance. Journal of Materials
Chemistry, 2010. 20(22): p. 4615-4620.
 ACCEPTED MANUSCRIPT

372. Hu, D., Z.-L. Xu, and Y.-M. Wei, A high performance silica-fluoropolyamide nanofiltration
membrane prepared by interfacial polymerization. Sep. Purif. Technol., 2013. 110: p. 31-38.
373. Hu, D., et al., Novel Composite Nanofiltration Membrane Prepared by Interfacial Polymerization
of 2,2'-Bis(1-Hydroxyl-1-trifluoromethyl-2,2,2-trifluoroethyl)-4,4'-methylenedianiline and
Trimesoyl Chloride. Sep. Sci. Technol., 2013. 48(4): p. 554-563.
374. Jons, S.D., et al., US5876602, "Treatment of composite polyamide membranes to improve
performance". 1999.

PT
375. Hong, S., et al., Interfacially synthesized chlorine-resistant polyimide thin film composite (TFC)
reverse osmosis (RO) membranes. Desalination, 2013. 309: p. 18-26.
376. Zhang, J., et al., Novel diamine-modified composite nanofiltration membranes with chlorine

RI
resistance using monomers of 1,2,4,5-benzene tetracarbonyl chloride and m-phenylenediamine.
J. Mater. Chem. A, 2015. 3(16): p. 8816-8824.
377. Park, H.B., et al., Highly Chlorine-Tolerant Polymers for Desalination. Angewandte Chemie, 2008.
120(32): p. 6108-6113.

SC
378. Paul, M., et al., Synthesis and crosslinking of partially disulfonated poly(arylene ether sulfone)
random copolymers as candidates for chlorine resistant reverse osmosis membranes. Polymer,
2008. 49(9): p. 2243-2252.

U
379. Wang, F., et al., Direct polymerization of sulfonated poly(arylene ether sulfone) random
(statistical) copolymers: candidates for new proton exchange membranes. Journal of Membrane

380.
Science, 2002. 197(12): p. 231-242.
AN
Hickner, M.A., et al., Alternative polymer systems for proton exchange membranes (PEMs).
Chemical reviews, 2004. 104(10): p. 4587-4612.
381. Xie, W., et al., Polyamide interfacial composite membranes prepared from m-phenylene diamine,
M
trimesoyl chloride and a new disulfonated diamine. Journal of Membrane Science, 2012. 403
404: p. 152-161.
382. Kim, S.G., et al., Novel thin-film composite membrane for seawater desalination with sulfonated
D

poly(arylene ether sulfone) containing amino groups. Desalination and Water Treatment, 2012.
43(1-3): p. 230-237.
383. Zhang, Z., et al., Preparation of polyamide membranes with improved chlorine resistance by bis-
TE

2,6-N,N-(2-hydroxyethyl) diaminotoluene and trimesoyl chloride. Desalination, 2013. 331: p. 16-

25.
384. Wei, X., et al., A novel method of surface modification on thin-film-composite reverse osmosis
EP

membrane by grafting hydantoin derivative. Journal of Membrane Science, 2010. 346(1): p. 152-
162.
385. Wei, X., et al., Surface modification of commercial aromatic polyamide reverse osmosis
membranes by graft polymerization of 3-allyl-5,5-dimethylhydantoin. Journal of Membrane
C

Science, 2010. 351(12): p. 222-233.

386. Liu, M., et al., Thin-film composite polyamide reverse osmosis membranes with improved acid
AC

stability and chlorine resistance by coating N-isopropylacrylamide-co-acrylamide copolymers.

Desalination, 2011. 270(13): p. 248-257.
387. Shin, D.H., N. Kim, and Y.T. Lee, Modification to the polyamide TFC RO membranes for
improvement of chlorine-resistance. Journal of Membrane Science, 2011. 376(12): p. 302-311.
388. Wu, D., et al., Effects of chlorine exposure on nanofiltration performance of polyamide
membranes. Journal of Membrane Science, 2015. 487: p. 256-270.
389. Cheng, X.Q., et al., Nanofiltration membrane achieving dual resistance to fouling and chlorine for
green separation of antibiotics. Journal of Membrane Science, 2015. 493: p. 156-166.
390. Soulier, J.-P., et al., Synthesis and properties of some polyamides and polysulfonamides. Journal
of Applied Polymer Science, 1974. 18(8): p. 2435-2447.
 ACCEPTED MANUSCRIPT

391. Arnold, F.E., S. Cantor, and C.S. Marvel, Aromatic polysulfonamides. Journal of Polymer Science
Part A-1: Polymer Chemistry, 1967. 5(3): p. 553-561.
392. Liu, M., et al., Acid stable thin-film composite membrane for nanofiltration prepared from
naphthalene-1,3,6-trisulfonyl chloride (NTSC) and piperazine (PIP). J. Membr. Sci., 2012. 415-416:
p. 122-131.
393. Yu, S., et al., Thin-film composite nanofiltration membranes with improved acid stability
prepared from naphthalene-1,3,6-trisulfonylchloride (NTSC) and trimesoyl chloride (TMC).

PT
Desalination, 2013. 315: p. 164-172.
394. Zhao, J., et al., Method for preparation of acid-resistant polysulfonamide nanofiltration
composite membrane from heterocyclic polyamine and 1,3,6-naphthalenetrisulfonyl chloride by

RI
interfacial polymerization. 2011, Hangzhou Fangran Filtration Membrane Technology Co., Ltd.,
Peop. Rep. China . p. 16pp.
395. Trushinski, B.J., et al., Polysulfonamide thin-film composite reverse osmosis membranes. Journal
of Membrane Science, 1998. 143(12): p. 181-188.

SC
396. Kurth, C.J., et al., US7138058, Acid stable membranes for nanofiltration. 2006.
397. Lee, K.P., et al., pH stable thin film composite polyamine nanofiltration membranes by interfacial
polymerisation. J. Membr. Sci., 2015. 478: p. 75-84.

U
398. Xue, L., et al., CN103933881, Method for preparing composite nanofiltration membrane with
interfacial polymerization. 2014, Ningbo Institute of Materials Technology & Engineering,

399.
AN
Chinese Academy of Sciences, Peop. Rep. China .
Tanninen, J., et al., Long-term acid resistance and selectivity of NF membranes in very acidic
conditions. Journal of Membrane Science, 2004. 240(12): p. 11-18.
400. Platt, S., et al., Stability of NF membranes under extreme acidic conditions. Journal of Membrane
M
Science, 2004. 239(1): p. 91-103.
401. Simon, A., et al., Effects of caustic cleaning on pore size of nanofiltration membranes and their
rejection of trace organic chemicals. Journal of Membrane Science, 2013. 447: p. 153-162.
D

402. Marsh, A.R., et al., US7156997, Package assembly for piperazine-based membranes. 2007.
403. Mnttri, M., A. Pihlajamki, and M. Nystrm, Effect of pH on hydrophilicity and charge and their
effect on the filtration efficiency of NF membranes at different pH. Journal of Membrane
TE

Science, 2006. 280(12): p. 311-320.

404. Simon, A., W.E. Price, and L.D. Nghiem, Changes in surface properties and separation efficiency
of a nanofiltration membrane after repeated fouling and chemical cleaning cycles. Separation
EP

and Purification Technology, 2013. 113: p. 42-50.

405. Meihong, L., et al., Study on the thin-film composite nanofiltration membrane for the removal of
sulfate from concentrated salt aqueous: Preparation and performance. Journal of Membrane
Science, 2008. 310(12): p. 289-295.
C

406. Dalwani, M., et al., Effect of pH on the performance of polyamide/polyacrylonitrile based thin
film composite membranes. Journal of Membrane Science, 2011. 372(12): p. 228-238.
AC

407. Simon, A., W.E. Price, and L.D. Nghiem, Impact of chemical cleaning on the nanofiltration of
pharmaceutically active compounds (PhACs): The role of cleaning temperature. Journal of the
Taiwan Institute of Chemical Engineers, 2013. 44(5): p. 713-723.
408. Mullett, M., R. Fornarelli, and D. Ralph, Nanofiltration of Mine Water: Impact of Feed pH and
Membrane Charge on Resource Recovery and Water Discharge. Membranes (Basel), 2014. 4(2):
p. 163-80.
409. Tang, B., Z. Huo, and P. Wu, Study on a novel polyester composite nanofiltration membrane by
interfacial polymerization of triethanolamine (TEOA) and trimesoyl chloride (TMC): I.
Preparation, characterization and nanofiltration properties test of membrane. Journal of
Membrane Science, 2008. 320(12): p. 198-205.
 ACCEPTED MANUSCRIPT

410. Li, L., et al., A novel nanofiltration membrane prepared with PAMAM and TMC by in situ
interfacial polymerization on PEK-C ultrafiltration membrane. J. Membr. Sci., 2006. 269(1-2): p.
84-93.
411. Yu, S., et al., Study on polyamide thin-film composite nanofiltration membrane by interfacial
polymerization of polyvinylamine (PVAm) and isophthaloyl chloride (IPC). Journal of Membrane
Science, 2011. 379(12): p. 164-173.
412. Kim, I.-C., J. Jegal, and K.-H. Lee, Effect of aqueous and organic solutions on the performance of

PT
polyamide thin-film-composite nanofiltration membranes. Journal of Polymer Science Part B:
Polymer Physics, 2002. 40(19): p. 2151-2163.
413. Livingston, A.G., Y.S. Bhole, and M.F. Jimenez Solomon, WO2012010889A1, Solvent-resistant

RI
polyamide nanofiltration membranes. 2012, Imperial Innovations Limited, UK .
414. Jimenez Solomon, M.F., Y. Bhole, and A.G. Livingston, High flux membranes for organic solvent
nanofiltration (OSN)Interfacial polymerization with solvent activation. Journal of Membrane
Science, 2012. 423424: p. 371-382.

SC
415. Zhang, H., et al., Cross-linked polyacrylonitrile/polyethyleneiminepolydimethylsiloxane
composite membrane for solvent resistant nanofiltration. Chemical Engineering Science, 2014.
106: p. 157-166.

U
416. Minhas, F.T., et al., Solvent resistant thin film composite nanofiltration membrane:
Characterization and permeation study. Appl. Surf. Sci., 2013. 282: p. 887-897.
417.

418.
AN
Sirkar, K.K., A.P. Korikov, and P.B. Kosaraju, US20080197070A1, Composite membranes formed
by interfacial polymerization on polyolefins. 2008, New Jersey Institute of Technology, USA.
Cadotte, J.E., US4039440A, Reverse osmosis membrane. 1977.
419. Fang, W., L. Shi, and R. Wang, Interfacially polymerized composite nanofiltration hollow fiber
M
membranes for low-pressure water softening. J. Membr. Sci., 2013. 430: p. 129-139.
420. Wei, X., et al., Application of Positively Charged Composite Hollow-Fiber Nanofiltration
Membranes for Dye Purification. Ind. Eng. Chem. Res., 2014. 53(36): p. 14036-14045.
D

421. Wei, X., et al., Characterization of a positively charged composite nanofiltration hollow fiber
membrane prepared by a simplified process. Desalination, 2014. 350: p. 44-52.
422. Sun, S.P., et al., Novel thin-film composite nanofiltration hollow fiber membranes with double
TE

repulsion for effective removal of emerging organic matters from water. J. Membr. Sci., 2012.
401-402: p. 152-162.
423. Sun, S. and T. Chung, WO2013039456A1, Thin film composite nanofiltration hollow fiber
EP

membranes. 2013, National University of Singapore, Singapore .

424. Sun, S.P., T.A. Hatton, and T.-S. Chung, Hyperbranched Polyethyleneimine Induced Cross-Linking
of Polyamide-imide Nanofiltration Hollow Fiber Membranes for Effective Removal of
Ciprofloxacin. Environ. Sci. Technol., 2011. 45(9): p. 4003-4009.
C

425. Wu, D., et al., Thin film composite nanofiltration membranes assembled layer-by-layer via
interfacial polymerization from polyethylenimine and trimesoyl chloride. J. Membr. Sci., 2014.
AC

472: p. 141-153.
426. Wu, D., et al., Thin film composite nanofiltration membranes fabricated from polymeric amine
polyethylenimine imbedded with monomeric amine piperazine for enhanced salt separations.
React. Funct. Polym., 2015. 86: p. 168-183.
427. Chiang, Y.-C., et al., Nanofiltration membranes synthesized from hyperbranched
polyethyleneimine. J. Membr. Sci., 2009. 326(1): p. 19-26.
428. Liu, M., et al., Thin-film composite membrane formed by interfacial polymerization of
polyvinylamine (PVAm) and trimesoyl chloride (TMC) for nanofiltration. Desalination, 2012. 288:
p. 98-107.
 ACCEPTED MANUSCRIPT

429. Jo, S. and K. Park, Surface modification using silanated poly(ethylene glycol)s. Biomaterials, 2000.
21(6): p. 605-616.
430. Van Wagner, E.M., et al., Surface modification of commercial polyamide desalination
membranes using poly(ethylene glycol) diglycidyl ether to enhance membrane fouling resistance.
Journal of Membrane Science, 2011. 367(12): p. 273-287.
431. McCloskey, B.D., et al., A bioinspired fouling-resistant surface modification for water purification
membranes. Journal of Membrane Science, 2012. 413414: p. 82-90.

PT
432. Kang, G.-d. and Y.-m. Cao, Development of antifouling reverse osmosis membranes for water
treatment: A review. Water Research, 2012. 46(3): p. 584-600.
433. Gol, R.M., et al., Effect of amine spacer of PEG on the properties, performance and antifouling

RI
behavior of poly(piperazineamide) thin film composite nanofiltration membranes prepared by in
situ PEGylation approach. J. Membr. Sci., 2014. 472: p. 154-166.
434. Cheng, X.Q., L. Shao, and C.H. Lau, High flux polyethylene glycol based nanofiltration membranes
for water environmental remediation. J. Membr. Sci., 2015. 476: p. 95-104.

SC
435. Shao, L. and X. Cheng, CN104117296A, "Polyamide composite nanofiltration membrane with
main chain containing ether-oxygen structure, and preparation method thereof". 2014, Harbin
Institute of Technology, Peop. Rep. China .

U
436. Sfora, M.L., S.P. Nunes, and K.V. Peinemann, Composite nanofiltration membranes prepared by
in situ polycondensation of amines in a poly(ethylene oxide-b-amide) layer. Journal of Membrane

437.
Science, 1997. 135(2): p. 179-186.
AN
Kang, G., et al., A novel method of surface modification on thin-film composite reverse osmosis
membrane by grafting poly(ethylene glycol). Polymer, 2007. 48(5): p. 1165-1170.
438. Zhang, Y., et al., Composite nanofiltration membranes prepared by interfacial polymerization
M
with natural material tannic acid and trimesoyl chloride. J. Membr. Sci., 2013. 429: p. 235-242.
439. Zhou, C., et al., Thin-film composite membranes formed by interfacial polymerization with
natural material sericin and trimesoyl chloride for nanofiltration. J. Membr. Sci., 2014. 471: p.
D

381-391.
440. Zhao, J., et al., Dopamine composite nanofiltration membranes prepared by self-polymerization
and interfacial polymerization. J. Membr. Sci., 2014. 465: p. 41-48.
TE

441. Zwijnenburg, A., J.W. De Rijk, and C.J. Rekers, US5833854A, Semipermeable composite
membrane and its preparation. 1998, Stork Friesland B.V., Neth. .
442. Zwijnenburg, A., C.J.N. Rekers, and J.W. De Rijk, US5833854, Semipermeable composite
EP

membrane, its preparation, and use for reverse osmosis or nanofiltration. 1997, Stork Friesland
B.V., Neth. .
443. Li, L., et al., CN1636623A, Nanofiltration membrane and its preparation process. 2005, Peop.
Rep. China .
C

444. Jin, L., et al., Preparation and characterization of a novel PA-SiO2 nanofiltration membrane for
raw water treatment. Desalination, 2012. 298: p. 34-41.
AC

445. Jin, L.M., et al., Synthesis of a novel composite nanofiltration membrane incorporated SiO2
nanoparticles for oily wastewater desalination. Polymer, 2012. 53(23): p. 5295-5303.
446. Jin, L.-m., et al., Preparation and characterization of a novel organic-inorganic nanofiltration
membranes prepared by interfacial polymerization. Adv. Mater. Res., 2012. 374-377: p. 1059-
1063.
447. Jin, L.-m., et al., Composite nanofiltration membranes synthesized from PAMAM and TMC by
interfacial polymerization. J. Harbin Inst. Technol. (Engl. Ed.), 2012. 19(1): p. 116-120.
448. Wei, X.-Z., et al., New type of nanofiltration membrane based on crosslinked hyperbranched
polymers. J. Membr. Sci., 2008. 323(2): p. 278-287.
 ACCEPTED MANUSCRIPT

449. Wei, X.-Z., J. Yang, and G.-L. Zhang, Preparation and characterization of nanofiltration
membranes synthesized by hyperbranched polyester and terephthaloyl chloride (TPC). Polym.
Compos., 2012. 20(3): p. 261-270.
450. Wei, X., et al., Structure influence of hyperbranched polyester on structure and properties of
synthesized nanofiltration membranes. J. Membr. Sci., 2013. 440: p. 67-76.
451. Mansourpanah, Y., S.S. Madaeni, and A. Rahimpour, Fabrication and development of interfacial
polymerized thin-film composite nanofiltration membrane using different surfactants in organic

PT
phase; study of morphology and performance. J. Membr. Sci., 2009. 343(1-2): p. 219-228.
452. Mansourpanah, Y., et al., Using different surfactants for changing the properties of
poly(piperazineamide) TFC nanofiltration membranes. Desalination, 2011. 271(1-3): p. 169-177.

RI
453. Kamiyama, Y., N. Yoshioka, and K. Nakagome, US4619767, Semipermeable composite
membrane. 1982, Nitto Electric Industrial Co., Ltd., Japan .
454. Kim, Y.S., et al., KR2005074166A, Method for producing polyamide based nanofiltration
composite membrane having high flow rate compared with existing method by interfacial

SC
polymerization. 2005, Saehan Industries Incorporation, S. Korea .
455. Chen, S.-H., et al., Preparation and separation properties of polyamide nanofiltration membrane.
J. Appl. Polym. Sci., 2002. 83(5): p. 1112-1118.

U
456. Kim, B.J. and Y.S. Kim, Nanofiltration composite membrane and manufacturing method therefor.
2003, Hy Sung Corporation, S. Korea . p. No pp. given.
457.

458.
Dow Chemical Co., USA .
AN
Fibiger, R.F., et al., US4769148A, Novel polyamide reverse-osmosis composite membranes. 1988,

Tomaschke, J.E., US4872984A, Interfacially synthesized reverse osmosis membrane containing

an amine salt and processes for preparing the same. 1989.
M
459. Tomaschke, J.E., US5246587A, Interfacially synthesized reverse osmosis polyurethane composite
membranes and their manufacture. 1993.
460. Tomaschke, J.E. and I.E. Ary, Reverse-osmosis membranes, their manufacture by interfacial
D

polymerization and use. 1993, Hydranautics, USA . p. 23 pp.

461. Jegal, J., S.G. Min, and K.-H. Lee, Factors affecting the interfacial polymerization of polyamide
active layers for the formation of polyamide composite membranes. J. Appl. Polym. Sci., 2002.
TE

86(11): p. 2781-2787.
462. Xiang, J., et al., Effect of amine salt surfactants on the performance of thin film composite
poly(piperazine-amide) nanofiltration membranes. Desalination, 2013. 315: p. 156-163.
EP

463. Xiang, J., et al., Effect of ammonium salts on the properties of poly(piperazineamide) thin film
composite nanofiltration membrane. J. Membr. Sci., 2014. 465: p. 34-40.
464. Sethi, A.R., et al., Ionic liquids as solvents for organic synthesis. NATO Sci. Ser., II, 2003. 92(Green
Industrial Applications of Ionic Liquids): p. 457-464.
C

465. Welton, T., Ionic liquids in Green Chemistry. Green Chem., 2011. 13(2): p. 225.
466. Mickols, W.E., US6878278, Composite membrane and method for making the same. 2005.
AC

467. Kim, I.-C., et al., Preparation of high flux thin film composite polyamide membrane: The effect of
alkyl phosphate additives during interfacial polymerization [Erratum to document cited in
CA158:565853]. Desalination, 2013. 326: p. 46.
468. ROY, A., et al., US9051227, In-situ method for preparing hydrolyzed acyl halide compound. 2015.
469. Hirose, M., et al., US5614099, Highly permeable composite reverse osmosis membrane, method
of producing the same, and method of using the same. 1997.
470. Chau, M.M., US4950404, High flux semipermeable membranes. 1990.
471. Chau, M.M., W.G. Light, and A.X. Swamikannu, US5271843A, Chlorine-tolerant, thin-film
composite membrane. 1993.
 ACCEPTED MANUSCRIPT

472. Kim, S.H., S.-Y. Kwak, and T. Suzuki, Positron Annihilation Spectroscopic Evidence to Demonstrate
the Flux-Enhancement Mechanism in Morphology-Controlled Thin-Film-Composite (TFC)
Membrane. Environmental Science & Technology, 2005. 39(6): p. 1764-1770.
473. Kwak, S.-Y., S.G. Jung, and S.H. Kim, Structure-Motion-Performance Relationship of Flux-
Enhanced Reverse Osmosis (RO) Membranes Composed of Aromatic Polyamide Thin Films.
Environmental Science & Technology, 2001. 35(21): p. 4334-4340.
474. Hirose, M. and K. Ikeda, US5576057, Method of producing high permeable composite reverse

PT
osmosis membrane. 1996.
475. Kong, C., et al., Co-solvent-mediated synthesis of thin polyamide membranes. Journal of
Membrane Science, 2011. 384(12): p. 10-16.

RI
476. Kong, C., et al., Controlled synthesis of high performance polyamide membrane with thin dense
layer for water desalination. Journal of Membrane Science, 2010. 362(12): p. 76-80.
477. Akbari, A. and S.M.M. Rostami, Development of permeability properties of polyamide thin film
composite nanofiltration membrane by using the dimethyl sulfoxide additive. J. Water Reuse

SC
Desalin., 2014. 4(3): p. 174-181.
478. Liu, T.-Y., et al., Fabrication of a high-flux thin film composite hollow fiber nanofiltration
membrane for wastewater treatment. J. Membr. Sci., 2015. 478: p. 25-36.

U
479. Duan, M., et al., Influence of hexamethyl phosphoramide on polyamide composite reverse
osmosis membrane performance. Separation and Purification Technology, 2010. 75(2): p. 145-

480.
155.
AN
Ghosh, A.K., et al., Impacts of reaction and curing conditions on polyamide composite reverse
osmosis membrane properties. Journal of Membrane Science, 2008. 311(12): p. 34-45.
481. Rana, D., T. Matsuura, and R.M. Narbaitz, Novel hydrophilic surface modifying macromolecules
M
for polymeric membranes: Polyurethane ends capped by hydroxy group. Journal of Membrane
Science, 2006. 282(12): p. 205-216.
482. Tarboush, B.J.A., et al., Preparation of thin-film-composite polyamide membranes for
D

desalination using novel hydrophilic surface modifying macromolecules. Journal of Membrane

Science, 2008. 325(1): p. 166-175.
483. Rana, D., et al., Development of antifouling thin-film-composite membranes for seawater
TE

desalination. Journal of Membrane Science, 2011. 367(1): p. 110-118.

484. Kim, I.-C. and K.-H. Lee, Preparation of interfacially synthesized and silicone-coated composite
polyamide nanofiltration membranes with high performance. Ind. Eng. Chem. Res., 2002. 41(22):
EP

p. 5523-5528.
485. Zhou, Y., et al., A novel method to improve flux of nanofiltration composite membrane prepared
by interfacial polymerization. Desalin. Water Treat., 2011. 34(1-3): p. 37-43.
486. An, Q., et al., Influence of polyvinyl alcohol on the surface morphology, separation and anti-
C

fouling performance of the composite polyamide nanofiltration membranes. J. Membr. Sci.,

2011. 367(1-2): p. 158-165.
AC

487. Tang, B., C. Zou, and P. Wu, Study on a novel polyester composite nanofiltration membrane by
interfacial polymerization. II. The role of lithium bromide in the performance and formation of
composite membrane. J. Membr. Sci., 2010. 365(1-2): p. 276-285.
488. Fan, X., et al., Improved performance of composite nanofiltration membranes by adding calcium
chloride in aqueous phase during interfacial polymerization process. J. Membr. Sci., 2014. 452: p.
90-96.
489. Hoek, E., Y. Yan, and B.H. Jeong, WO2006098872A2, Nanocomposite membranes and methods
of making and using same. 2006.
490. Jeong, B.-H., et al., Interfacial polymerization of thin film nanocomposites: A new concept for
reverse osmosis membranes. J. Membr. Sci., 2007. 294(1+2): p. 1-7.
 ACCEPTED MANUSCRIPT

491. Humplik, T., et al., Nanostructured materials for water desalination. Nanotechnology, 2011.
22(29): p. 292001.
492. Lau, W.J., et al., A review on polyamide thin film nanocomposite (TFN) membranes: History,
applications, challenges and approaches. Water Res., 2015. 80: p. 306-324.
493. Lind, M.L., et al., Influence of Zeolite Crystal Size on Zeolite-Polyamide Thin Film Nanocomposite
Membranes. Langmuir, 2009. 25(17): p. 10139-10145.
494. Nguyen, T.-V., et al., Relating fouling behavior and cake layer formation of alginic acid to the

PT
physiochemical properties of thin film composite and nanocomposite seawater RO membranes.
Desalination, 2014. 338: p. 1-9.
495. Fathizadeh, M., A. Aroujalian, and A. Raisi, Effect of added NaX nano-zeolite into polyamide as a

RI
top thin layer of membrane on water flux and salt rejection in a reverse osmosis process. J.
Membr. Sci., 2011. 375(1-2): p. 88-95.
496. Lind, M.L., et al., Effect of mobile cation on zeolite-polyamide thin film nanocomposite
membranes. Journal of Materials Research, 2009. 24(05): p. 1624-1631.

SC
497. Namvar-Mahboub, M. and M. Pakizeh, Optimization of preparation conditions of polyamide thin
film composite membrane for organic solvent nanofiltration. Korean J. Chem. Eng., 2014. 31(2):
p. 327-337.

U
498. Namvar-Mahboub, M., M. Pakizeh, and S. Davari, Preparation and characterization of UZM-
5/polyamide thin film nanocomposite membrane for dewaxing solvent recovery. J. Membr. Sci.,

499.
2014. 459: p. 22-32.
AN
Namvar-Mahboub, M. and M. Pakizeh, Development of a novel thin film composite membrane
by interfacial polymerization on polyetherimide/modified SiO2 support for organic solvent
nanofiltration. Sep. Purif. Technol., 2013. 119: p. 35-45.
M
500. Yin, J., et al., Fabrication of a novel thin-film nanocomposite (TFN) membrane containing MCM-
41 silica nanoparticles (NPs) for water purification. J. Membr. Sci., 2012. 423-424: p. 238-246.
501. Jadav, G.L. and P.S. Singh, Synthesis of novel silica-polyamide nanocomposite membrane with
D

enhanced properties. J. Membr. Sci., 2009. 328(1+2): p. 257-267.

502. Li, Q., et al., Effects of ordered mesoporous silica on the performances of composite
nanofiltration membrane. Desalination, 2013. 327: p. 24-31.
TE

503. Tang, B., H. Wu, and P. Wu, CN102794116A, Method for preparation of mesoporous silica
sphere-polymer nanocomposite nanofiltration membrane with stable structure and good
permeability, hydrophilicity and anti-fouling property. 2012.
EP

504. Wu, H., B. Tang, and P. Wu, Optimizing polyamide thin film composite membrane covalently
bonded with modified mesoporous silica nanoparticles. Journal of Membrane Science, 2013.
428: p. 341-348.
505. Li, Q., et al., Influence of silica nanospheres on the separation performance of thin film composite
C

poly(piperazine-amide) nanofiltration membranes. Appl. Surf. Sci., 2015. 324: p. 757-764.

506. Kebria, M.R.S., M. Jahanshahi, and A. Rahimpour, SiO2 modified polyethyleneimine-based
AC

nanofiltration membranes for dye removal from aqueous and organic solutions. Desalination,
2015. 367: p. 255-264.
507. Hu, D., Z.-L. Xu, and C. Chen, Polypiperazine-amide nanofiltration membrane containing silica
nanoparticles prepared by interfacial polymerization. Desalination, 2012. 301: p. 75-81.
508. Ma, X., et al., Preparation and Characterization of Silica/Polyamide-imide Nanocomposite Thin
Films. Nanoscale Res. Lett., 2010. 5(11): p. 1846-1851.
509. Lee, H.S., et al., Polyamide thin-film nanofiltration membranes containing TiO2 nanoparticles.
Desalination, 2008. 219(13): p. 48-56.
 ACCEPTED MANUSCRIPT

510. Rajaeian, B., et al., Fabrication and characterization of polyamide thin film nanocomposite (TFN)
nanofiltration membrane impregnated with TiO2 nanoparticles. Desalination, 2013. 313: p. 176-
188.
511. Bai, X., et al., Study on modification of positively charged composite nanofiltration membrane by
TiO2 nanoparticles. Desalination, 2013. 313: p. 57-65.
512. Peyravi, M., et al., Novel thin film nanocomposite membranes incorporated with functionalized
TiO2 nanoparticles for organic solvent nanofiltration. Chemical Engineering Journal, 2014. 241:

PT
p. 155-166.
513. Sorribas, S., et al., High Flux Thin Film Nanocomposite Membranes Based on Metal-Organic
Frameworks for Organic Solvent Nanofiltration. J. Am. Chem. Soc., 2013. 135(40): p. 15201-

RI
15208.
514. An, Q., et al., CN103285748A, Method for preparing carboxybetaine type colloid nanoparticle
filled polyamide nanofiltration membrane. 2013.
515. Ji, Y., Q. An, and C. Gao, CN103285752A, Polyamide nanofiltration membrane containing

SC
sulfobetaine-type colloid nanoparticle and preparation method thereof. 2013.
516. Ji, Y., et al., CN104028126A, Preparation of sulfonic acid-type amphoteric polyelectrolyte
nanoparticle hybrid polyamide nanofiltration membranes. 2014.

U
517. Kong, C., et al., Enhanced performance of inorganic-polyamide nanocomposite membranes
prepared by metal-alkoxide-assisted interfacial polymerization. J. Membr. Sci., 2011. 366(1-2): p.

518.
382-388.
AN
Kim, E.-S. and B. Deng, Fabrication of polyamide thin-film nano-composite (PA-TFN) membrane
with hydrophilized ordered mesoporous carbon (H-OMC) for water purifications. Journal of
Membrane Science, 2011. 375(12): p. 46-54.
M
519. Do, V.T., et al., Effects of hypochlorous acid exposure on the rejection of salt, polyethylene
glycols, boron and arsenic(V) by nanofiltration and reverse osmosis membranes. Water
Research, 2012. 46(16): p. 5217-5223.
D

520. Mi, Y.-F., et al., A novel route for surface zwitterionic functionalization of polyamide
nanofiltration membranes with improved performance. J. Membr. Sci., 2015. 490: p. 311-320.
521. Chiang, Y.-C., et al., A facile zwitterionization in the interfacial modification of low bio-fouling
TE

nanofiltration membranes. J. Membr. Sci., 2012. 389: p. 76-82.

522. Li, Y., et al., Surface fluorination of polyamide nanofiltration membrane for enhanced antifouling
property. Journal of Membrane Science, 2014. 455: p. 15-23.
EP

523. Belfer, S., et al., Modification of NF membrane properties by in situ redox initiated graft
polymerization with hydrophilic monomers. Journal of Membrane Science, 2004. 239(1): p. 55-
64.
524. Kim, J.-H., et al., Surface modification of nanofiltration membranes to improve the removal of
C

organic micro-pollutants (EDCs and PhACs) in drinking water treatment: Graft polymerization
and cross-linking followed by functional group substitution. Journal of Membrane Science, 2008.
AC

321(2): p. 190-198.
525. Mansourpanah, Y. and E. Momeni Habili, Preparation and modification of thin film PA
membranes with improved antifouling property using acrylic acid and UV irradiation. Journal of
Membrane Science, 2013. 430: p. 158-166.
526. Himstedt, H.H., K.M. Marshall, and S.R. Wickramasinghe, pH-responsive nanofiltration
membranes by surface modification. Journal of Membrane Science, 2011. 366(12): p. 373-381.
527. Kim, E.-S., Q. Yu, and B. Deng, Plasma surface modification of nanofiltration (NF) thin-film
composite (TFC) membranes to improve anti organic fouling. Applied Surface Science, 2011.
257(23): p. 9863-9871.
 ACCEPTED MANUSCRIPT

528. Navarro, R., et al., Effect of an acidic treatment on the chemical and charge properties of a
nanofiltration membrane. Journal of Membrane Science, 2008. 307(1): p. 136-148.
529. Kuehne, M.A., et al., Flux enhancement in TFC RO membranes. Environmental Progress, 2001.
20(1): p. 23-26.
530. Chen, Y., et al., CN104437106A, One kind of method for preparation of nanofiltration membrane
by interfacial polymerization. 2015, Beijing Origin Water Membrane Technology Co., Ltd., Peop.
Rep. China .

PT
531. Chapman Wilbert, M., J. Pellegrino, and A. Zydney, Bench-scale testing of surfactant-modified
reverse osmosis/nanofiltration membranes. Desalination, 1998. 115(1): p. 15-32.
532. Nath, K., T.M. Patel, and H.K. Dave, Performance characteristics of surfactant treated

RI
commercial polyamide membrane in the nanofiltration of model solution of reactive yellow 160.
Journal of Water Process Engineering, 2015.
533. Cadotte, J.E. and D.R. Walker, US4812270, Novel water softening membranes. 1989.
534. Wheeler, J.W., US5262054, Process for opening reverse osmosis membranes. 1993.

SC
535. Mickols, W.E., US5755964, Method of treating polyamide membranes to increase flux. 1998.
536. Garca-Pacheco, R., et al., Transformation of end-of-life RO membranes into NF and UF
membranes: Evaluation of membrane performance. Journal of Membrane Science, 2015. 495: p.

U
305-315.
537. Lawler, W., et al., Production and characterisation of UF membranes by chemical conversion of
AN
used RO membranes. Journal of Membrane Science, 2013. 447: p. 203-211.
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT

Highlights

Review (537 references) focusing on chemistry of polymeric nanofiltration membranes.

Emphasis on interfacially polymerized NF composite membranes.

PT
Discusses composite structure, reactants, additives, and post treatments.

Identifies need for improved uniformity of membrane testing within NF field.

RI
U SC
AN
M
D
TE
C EP
AC