POLYELECTROLYTE MULTILAYER FILMS FOR ION SEPARATION AND WATER
PURIFICATION
By
Maneesha Adusumilli

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY
Chemistry
2010

ABSTRACT
POLYELECTROLYTE MULTILAYER FILMS FOR ION SEPARATION
AND WATER PURIFICATION
By
Maneesha Adusumilli
Polyelectrolyte multilayer (PEM) films formed by layer by layer (LbL) adsorption of
polycations and polyanions are attractive as the skins of separation membranes because of their
simple deposition, minimal thickness, and wide range of transport properties. Variation of the
polyelectrolyte type as well as deposition parameters such as supporting electrolyte
concentration, polyelectrolyte concentration, duration and temperature of adsorption allow
tailoring of films for specific separations. This dissertation examines the effect of ion-exchange
sites in PEM films on ion separations as well as removal of model organic pollutants by PEM
membranes.
The properties of poly(styrene sulfonate) (PSS)/ poly(diallyldimethylammonium chloride)
(PDADMAC) films vary dramatically with the number of polyelectrolyte layers deposited.
Attenuated total reflectance infrared spectroscopy shows that coatings with <7 PDADMAC/PSS
-

2-

bilayers do not absorb significant amounts of SCN or Ni(CN)4 , but films with >7 bilayers
exhibit an ion-exchange capacity of about 0.5 moles per L of film.

For Si-supported

PSS/PDADMAC films terminated with PSS, zeta (ζ) potentials change from negative to positive
as the number of adsorbed bilayers increases. These changes in film properties dramatically
-

2-

affect ion transport through (PSS/PDADMAC)nPSS-coated alumina membranes. The Cl /SO4

selectivities of these membranes are >30 with (PSS/PDADMAC)4PSS films but only 3 with

(PSS/PDADMAC)6PSS films. Trends in ζ potentials and selectivities with increasing numbers
of bilayers are consistent with the exponential growth mechanism where a polycation adsorbs
throughout the film to create large numbers of anion-exchange sites, and during polyanion
deposition some of the polycation diffuses to the surface of the film to complex with polyanions
from solution. Polycation that is not electrically compensated by the polyanion affords anion2-

-

exchange sites, and the presence of this fixed positive charge yields decreased Cl /SO4
selectivity.

Deliberate introduction of anion-exchange sites in PEMs can greatly enhance selectivities
in cation separations. Immersion of PSS/poly(allylamine hydrochloride) (PAH) membranes in
aqueous CuCl2 or FeCl3 solutions yields adsorbed Cu
exchange sites.

+

2+

Remarkably, the Na /Mg

2+

3+

or Fe

ions that serve as anion-

selectivity of (PSS/PAH)4 films on 50 kDa

polyethersulfone membranes increases from 17 to 109 after immersion in 5 mM FeCl3 for 10
min. Moreover, this modification is effective even after 18 h of cross-flow nanofiltration (NF).
Reflectance FTIR spectroscopy confirms the formation of anion-exchanges sites after Fe
2+

Cu

3+

or

adsorption. Due to its low pressure requirements, NF is a promising technique for removal

of endocrine disrupting and pharmaceutically active compounds from waste water. However, the
rejection of small organic compounds by commercial membranes may be too low for effective
water treatment. This work shows that coating of commercial NF-90 membrane with a bilayer of
poly(acrylic acid)/PAH increases acetaminophen rejection from 65% to 85%. (PSS/PAH)7 on
3

2

porous alumina shows a caffeine rejection of 93% along with a solution flux of 1.33 m /(m day).

To my husband, Upendra and my son, Rithvik.

iv

ACKNOWLEDGEMENTS

Though I am the “author” of this dissertation, there are many people who made this
possible and I will be indebted to them forever.
First and foremost, I thank my advisor, Dr. Bruening whose guidance and support over
the past six years kept me going in the right direction and made this dissertation possible. He
taught me how to write academic papers, made me a better researcher, and had confidence in me
when I doubted myself. Without his endless corrections and constant pushing me out of my
comfort zones my dissertation would not have been what it is today.
I would like to take this opportunity to thank my committee members, Dr. Blanchard, Dr.
Swain and Dr. Baker for being on my committee and for all the helpful discussions.
I spent more than five years in the lab and got to know many students in the group. I
sincerely thank them for providing a friendly lab environment and helping me overcome many
difficulties in research as well as in daily life.
Finally I come to the most personal source of gratitude. I thank my husband, Upendra for
his understanding and encouragement to complete this research. He was always there cheering
me up and stood by me through the good times and bad. I must also thank my son, Rithvik, who
had to endure my absence in order that I could conduct my research and write my dissertation. I
need to thank my little boy for being such a bundle of joy and laughter.

v

TABLE OF CONTENTS
LIST OF TABLES……………………………………………………………………………....ix
LIST OF FIGURES……………………………………………………………….…………....xii
LIST OF ABBREVATIONS……………………….…………………………………………xvii

Chapter 1: Introduction and Background.................................................................................. 1
1.1. Layer by Layer (LbL) Assembly of Polyelectrolyte Multilayers ........................................ 1
1.2. Mechanism of Polyelectrolyte Adsorption ........................................................................... 4
1.3. Factors that Affect Film Growth during LbL Polyelectrolyte Adsorption ......................... 6
1.3.1. Effect of Supporting Electrolyte .................................................................................... 6
1.3.1.1. Effect of Supporting Electrolyte Concentration .................................................... 6
1.3.1.2. Effect of Type of Supporting Electrolyte ............................................................... 9
1.3.2. Effect of Polyelectrolyte on Multilayer Adsorption ................................................... 12
1.3.2.1. Effect of Polyelectrolyte Type on Multilayer Growth ........................................ 12
1.3.2.2. Effect of Polyelectrolyte Molecular Weight ........................................................ 13
1.3.2.3. Effect of Polyelectrolyte Concentration on Film Thickness ............................... 16
1.3.2.4. Effect of Deposition pH (or Degree of Polyelectrolyte Ionization) on
the Growth of PEMs Containing Weak Polyelectrolytes .................................... 19
1.3.3. Effect of Polyelectrolyte Adsorption Time on the Growth of PEMs ........................ 21
1.3.4. Effect of Temperature on the Growth of PEMs.......................................................... 22
1.4. Swelling and Doping of Polyelectrolyte Multilayers ......................................................... 23
1.5. Applications of Polyelectrolyte Multilayer Films .............................................................. 25
1.5.1. Membrane-based Separations ....................................................................................... 25
1.5.1.1. Nanofiltration Membranes ..................................................................................... 25
1.5.1.2. Gas Separation Membranes ................................................................................... 29
1.5.1.3. Pervaporation Membranes ..................................................................................... 30
1.5.1.4. Membranes for Protein Separation and Purification ............................................ 31
1.5.2. Encapsulation of Drugs and Enzymes ......................................................................... 32
1.5.3. Catalytic Membrane Reactors...................................................................................... 33
1.5.4. Antibacterial Coatings .................................................................................................. 35
1.5.5. Environmental Applications ........................................................................................ 36
1.6. Dissertation Outline ............................................................................................................. 38
1.7. References............................................................................................................................. 41
Chapter 2: Methods and Techniques ........................................................................................ 52
2.1. Polyelectrolyte Multilayer Nomenclature and Deposition ................................................ 52
2.2. Determination of Film Thickness by Ellipsometry and Scanning Electron
Microscopy.............................................................................................................. 53
2.2.1. Ellipsometry .................................................................................................................. 53
2.2.2. Scanning Electron Microscopy (SEM) ....................................................................... 54
2.3. Molecular Spectroscopy ...................................................................................................... 54
2.3.1. Reflectance Fourier Transform Infrared (FTIR) Spectroscopy .................................. 54
vi

2.4.
2.5.
2.6.
2.7.
2.8.

2.3.2. Attenuated Total Reflectance Fourier Transform Infrared (ATR-FTIR)
Spectroscopy .................................................................................................................. 55
2.3.3. UV/Vis Spectroscopy .................................................................................................... 57
Ion Chromatography ............................................................................................................ 57
Atomic Absorption Spectroscopy and Inductively Coupled Plasma Optical
Emission Spectroscopy ................................................................................... 58
Streaming Potential Measurements ..................................................................................... 58
Nanofiltration Experiments ................................................................................................. 61
References............................................................................................................................. 64

Chapter 3: Variation of Ion-exchange Capacity, Zeta Potential, and Ion-transport
Selectivities with the Number of Layers in a Multilayer Polyelectrolyte
Film ........................................................................................................................... 65
3.1. Introduction............................................................................................................................... 65
3.2. Experimental Section ............................................................................................................... 67
3.3. Results and Discussion............................................................................................................. 70
3.3.1. Film Formation and Characterization........................................................................... 70
3.3.2. ATR-FTIR Studies of SCN Absorption in (PDADMAC/PSS)30PDADMAC
films ................................................................................................................................ 74
3.3.3. Competitive ion exchange ............................................................................................. 82
3.3.4. Ion binding as a function of the number of adsorbed PDADMAC/PSS
bilayers .......................................................................................................................... 86
3.3.5. Streaming Potential Measurements .............................................................................. 88
3.3.6. Nanofiltration Properties of PSS/PDADMAC films as a function of the
number of bilayers ......................................................................................................... 91
3.4. Conclusions ............................................................................................................................... 98
3.5. Appendix A ............................................................................................................................. 101
3.6. References ............................................................................................................................... 113
+

Chapter 4: Enhancing the Na / Mg

2+

Transport Selectivity of Polyelectrolyte
2+

3+

Multilayer Films Through Adsorption of Cu or Fe ..................................118
4.1. Introduction ............................................................................................................................. 118
4.2. Experimental Section ............................................................................................................. 119
4.3. Results and Discussion...........................................................................................................122
2+
3+
4.3.1. Adsorption of Cu and Fe in PEM Films ............................................................. 122
4.3.2. Effects of Metal Ion Adsorption on the Nanofiltration Properties of
(PSS/PAH)n Films on PES membranes ..................................................................... 127
4.3.2.1. Mixed Salt Solutions ........................................................................................... 127
3+
4.3.2.1.1. Effect of immersion time and Fe concentration on NF properties
of membranes treated with FeCl3. ............................................................... 131
4.3.2.2. Single Salt Experiments. ...................................................................................... 133
4.4. Conclusions ............................................................................................................................. 134
4.5. References ............................................................................................................................... 137
vii

Chapter 5: Polyelectrolyte Multilayer Films for Removal of Pharmaceutically
Active Compounds from Water...........................................................................139
5.1. Introduction ............................................................................................................................. 139
5.2. Experimental Section ............................................................................................................. 140
5.3. Results and Discussion...........................................................................................................142
5.3.1. Nanofiltration of ACT with NF-270, NF-90 and PSS/PAH membranes ................. 142
5.3.1.1. Effect of polyelectrolyte type and concentration on the rejection of
ACT ....................................................................................................................... 145
5.3.2. Nanofiltration of CFN with NF-270, NF-90 and PSS/PAH films ............................ 147
5.4. Conclusions ............................................................................................................................. 148
5.5. References ............................................................................................................................... 151
Chapter 6: Conclusions and Future Work ............................................................................. 154
6.1. References ............................................................................................................................... 158

viii

LIST OF TABLES

Table 1.1.

Ellipsometric thicknesses of (PDADMAC/PSS) 10 films adsorbed on a Si
wafer from solutions containing 1 mM polyelectrolyte (with respect to the
repeating unit) and a series of supporting electrolytes with an activity of
0.33. Hydration numbers, n, are also given for each ion. Adsorption
occurred for 5 min per layer with 3 min of rinsing after deposition of each
layer ......................................................................................................................... 10

Table 1.2.

Thickness changes per layer for PDADMAC/PSS films prepared using
different polyelectrolyte concentrations. All the films were made with no
added salt in deposition solutions .......................................................................... 18

Table 3.1.

Equilibrium ion concentrations in (PDADMAC/PSS)n films exposed to
solutions containing 2.5 mM NaSCN, or 2.5 mM K2Ni(CN)4, or 2.5 mM
of both salts ............................................................................................................. 87

Table 3.2.

Equilibrium ion concentrations in (PDADMAC/PSS)nPDADMAC films
exposed to solutions containing 2.5 mM NaSCN, or 2.5 mM K2Ni(CN)4,
or 2.5 mM of both salts .......................................................................................... 88

Table 3.3.

Rejections, solution fluxes, and selectivities from nanofiltration
experiments with (PSS/PDADMAC)nPSS-coated porous alumina
membranes and feed solutions containing 1000 ppm Cl

-

and 1000 ppm

2-

SO4 ....................................................................................................................... 93
Table 3.4.

Rejections, solution fluxes, and selectivities from nanofiltration
experiments with (PDADMAC/PSS)n-coated 100 kDa PES membranes
-

2-

and feed solutions containing 1000 ppm Cl and 1000 ppm SO4 ................... 94
Table 3.5.

Rejections, solution fluxes, and selectivities from nanofiltration
experiments with (PDADMAC/PSS)n-coated 50 kDa PES membranes and
-

2-

solutions containing 1000 ppm Cl and 1000 ppm SO4 ................................... 95
Table 3.6.

Rejections, solution fluxes, and selectivities from nanofiltration
experiments with (PSS/PDADMAC)nPSS-coated alumina membranes and
-

2-

feed solutions containing 1000 ppm SCN and 1000 ppm Ni(CN)4 ............... 95
ix

Table 3.7.

Rejections, solution fluxes, and selectivities from nanofiltration
experiments with (PSS/PDADMAC)4PSS-coated porous alumina
membranes and feed solutions containing different combinations of 1000
ppm anions .............................................................................................................. 96

Table 3.8.

Rejections, solution fluxes, and selectivities from nanofiltration
experiments with (PSS/PDADMAC)4PSS-coated porous alumina
membranes and solutions containing different combinations of 1000 ppm
anions. The experiments were performed sequentially (from top to
bottom) using the same two (PSS/PDADMAC)4PSS-coated alumina
membranes .............................................................................................................. 97

Table 3.9.

Rejections, solution fluxes, and selectivities from nanofiltration
experiments with (PSS/PDADMAC)4PSS-coated porous alumina
membranes and solutions containing different combinations of 1000 ppm
anions. The experiments were performed sequentially (from top to
bottom) using the same two (PSS/PDADMAC)4PSS-coated alumina
membranes .............................................................................................................. 98

Table A1.

Ellipsometric thicknesses and %swelling of PDADMAC/PSS films on
silicon wafer immersed in water .......................................................................... 105

Table 4.1.

Concentrations of Fe
and -SO3 in (PAH/PSS)4PAH films after
immersion of the film in an FeCl3 solution and rinsing ..................................... 123

Table 4.2.

Rejections, solution fluxes, and selectivities from nanofiltration
experiments with (PSS/PAH)n films deposited on 50 kDa PES membranes

3+

-

+

2+

and feed solutions containing 1000 ppm Na
and Mg . The
transmembrane pressure was 4.8 bar ...................................................................128
Table 4.3.

Rejections, solution fluxes, and selectivities from nanofiltration
experiments with (PSS/PAH)4 films deposited on 50 kDa PES membranes
-

2-

and feed solutions containing 1000 ppm Cl and SO4 ...................................130
Table 4.4.

Rejections, solution fluxes, and selectivities from nanofiltration
experiments with (PSS/PAH)4 films on 50 kDa PES membranes after
treatment of the membranes by immersion in FeCl3. The feed solutions
+
2+
contained 1000 ppm Na and Mg ....................................................................131
x

Table 4.5.

Rejections, solution fluxes, and selectivities from nanofiltration
experiments with (PSS/PAH)4 films on 50 kDa PES membranes with and
without treatment of the membranes by immersion in 5 mM FeCl3 for
+
2+
different times. The feed solutions contained 1000 ppm Na and Mg ........ 132

Table 4.6.

Rejections, solution fluxes, and selectivities from nanofiltration
experiments with NF-270 membranes before and after immersion in 0.5 M
+
2+
FeCl3 for 60 min. The feed solutions contained 1000 ppm Na and Mg ...133

Table 4.7.

Rejections and solution fluxes from single-salt nanofiltration experiments
3+
with (PSS/PAH)4 and Fe -modified (60 min) (PSS/PAH)4 films
deposited on 50 kDa PES membranes. The feed solutions contained 1000
+
2+
ppm Na or 1000 ppm Mg ............................................................................... 134

Table 5.1.

Physical and chemical properties of acetaminophen and caffeine .................... 142

Table 5.2.

Rejections and solution fluxes from nanofiltration experiments with NF270 and NF-90 membranes, and PSS/PAH films deposited on porous
alumina. The feed solutions contained 100 ppm ACT. In this case 0.02 M
PSS and PAH are used ......................................................................................... 144

Table 5.3.

Rejections and solution fluxes from nanofiltration experiments with NF270 and NF-90 membranes modified with polyelectrolyte coatings. The
feed solutions contained 100 ppm ACT .............................................................. 145

Table 5.4.

Rejections and solution fluxes from nanofiltration experiments with NF270 and NF-90 membranes modified with PSS/PAH films deposited from
solutions containing different polyelectrolyte concentrations. The feed
solutions contained 100 ppm of ACT ..................................................................146

Table 5.5.

Rejections and solution fluxes from NF experiments with PSS/PAH and
PAA/PAH films deposited on NF-90 membranes. The feed solutions
contained 100 ppm of ACT, and the transmembrane pressure was 4.8 bar ......146

Table 5.6.

Rejections and solution fluxes from nanofiltration experiments with NF270 and NF-90 membranes and PSS/PAH films deposited on porous
alumina. The feed solutions contained 100 ppm CFN ......................................148

xi

LIST OF FIGURES

Figure 1.1.

Schematic diagram of layer-by-layer adsorption of polyelectrolyte
multilayers ................................................................................................................. 2

Figure 1.2.

Structures of polyelectrolytes frequently used in building multilayers ................. 3

Figure 1.3.

Schematic drawing showing the buildup of a polyelectrolyte “bilayer”
during exponential growth. Step 1 includes diffusion of a polyelectrolyte
(in this case a polycation) into the entire film. During step 2, much of the
polycation diffuse out of the film and forms a complex with incoming
polyanion. H0 and H1 are the film thicknesses at the start and at the end of
step 1, and H2 is the film thickness at the end of step 2. h0 and h1 are the
increases in film thickness due to the deposition of polycation and
polyanion, respectively............................................................................................. 5

Figure 1.4.

Sketch of polyelectrolyte layers adsorbed in the presence and absence of
salt. The absence of salt results in thin layers of extended polyelectrolytes,
whereas at high ionic strength coiled polymers form thicker layers ..................... 7

Figure 1.5.

Thicknesses of LbL (PSS/PDADMAC)10 films deposited on silicon
wafers using a series of NaCl concentrations in the polyelectrolyte
deposition solutions .................................................................................................. 8

Figure 1.6.

Ellipsometric thicknesses of (PSS/PDADMAC)20 films deposited from
solutions containing 0.1 M of the sodium salt of the listed anions ..................... 12

Figure 1.7a.

Influence of the molecular weights of HA and PLL on the thickness of
(HA/PLL) n films grown using HA with Mw = 130000 and PLL with Mw
values of (▪) 20000, (○) 55000, and (▴ ) 360000 .................................................. 14

Figure 1.7b.

Influence of the molecular weights of HA and PLL on the thickness of
(HA/PLL) n films grown using PLL with Mw = 360000 and HA with Mw
values of (▪) 130000 and (○) 400000 ..................................................................... 15

Figure 1.8.

Thickness of (PSS/PDADMAC)5 films as a function of polyelectrolyte
concentration.
Deposition occurred on a silicon wafer from
polyelectrolyte solutions containing 1.0 M NaCl (aq) ........................................ 17

xii

Figure 1.9.

Structure of the polymeric dye poly(1-[ p-(3'-carboxy-4'hydroxyphenylazo) benzenesulfonamido]-1,2-ethandiyl) (PCBS) .................... 18

Figure 1.10.

Thickness as a function of number of polyelectrolyte (PE) layers for (a)
PAH/PSS and (b) PDADMAC/PAA multilayers deposited using different
deposition pH values for both polyelectrolytes .................................................... 20

Figure 1.11.

Effect of deposition temperature on the thickness of a
(PDADMAC/PSS)10 multilayer ............................................................................ 23

Figure 1.12.

Schematic diagram showing the binding of positively charged lysozyme to
a PEM film terminated with a polyanion. Charges are only shown for the
terminal layer .......................................................................................................... 32

Figure 1.13.

Schematic diagram of gold nanoparticle ( ) /polyelectrolyte ( ) films
immobilized in membranes for 4-nitrophenol reduction...................................... 35

Figure 2.1.

ATR-FTIR flow cell for the measurement of diffusion of ions in the
polymers. (The sample compartment was closed during measurements.)
The inset shows a schematic diagram of the ATR crystal in the flow cell ......... 56

Figure 2.2.

Schematic representation of electrical double layer and ζ potential.................... 60

Figure 2.3.

Electro Kinetic Analyzer flow cell setup for zeta potential measurements ........ 61

Figure 2.4.

Schematic drawing of the experimental NF apparatus ......................................... 62

Figure 3.1.

Structures of the polyelectrolytes used in this study ............................................ 68

Figure 3.2.

Ellipsometric thicknesses of (PDADMAC/PSS) n films on silicon wafers.
Polyelectrolytes were adsorbed from solutions containing 1 M NaCl ................ 71

Figure 3.3.

Cross-sectional SEM images of Si wafers coated with 10-bilayer (left) and
30-bilayer (right) PDADMAC/PSS films ............................................................. 72

Figure 3.4.

ATR-FTIR spectra of (PDADMAC/PSS)nPDADMAC films adsorbed on
a Ge ATR crystal. The numbers in the legend represent n, and the peaks
-1
-1
at 1008 cm and 1033 cm are due to the sulfonate groups of PSS. The
films were immersed in deionized water while the spectra were taken, and
the background spectrum was that of deionized water on a clean crystal .......... 73
xiii

-

-1

Figure 3.5.

Peak area of the -SO3 absorbance centered at 1033 cm in the ATRFTIR spectra of (PDADMAC/PSS)n multilayer films deposited on a Ge
crystal. Films were deposited from a solution containing 1 M NaCl. The
wide oscillations between the areas of peaks in PSS- and PDAMACterminated films are likely due to differences in the swelling of these
films. The more swollen PDADMAC-terminated films exhibit lower
absorbances. Films with an integer number of bilayers terminate with
PSS, while films with an additional 0.5 bilayer terminate with PDADMAC..... 74

Figure 3.6.

FTIR spectra of aqueous 0.5 M NaSCN on a bare Ge crystal (pink) and of
SCN absorbed in a (PDADMAC/PSS)30PDADMAC film on a Ge crystal
-

(blue). The SCN was absorbed for 6 h from a solution containing 1 mM
NaSCN, and the blue spectrum was taken while the film was still covered
with the NaSCN solution........................................................................................ 76
Figure 3.7.

ATR-FTIR spectra of a (PDADMAC/PSS)30PDADMAC film during
exposure to a 0.01 M NaSCN solution for varying lengths of time. The
absorbance in this region is due to the C=N stretch of SCN . The spectra
overlap after 180 min of exposure ......................................................................... 77

Figure 3.8.

Normalized area of the SCN peak (2058 cm ) in ATR-FTIR spectra of a

-

-1

(PDADMAC/PSS)30PDADMAC film exposed to different concentrations
of NaSCN for various periods of time. The film was deposited on a Ge
ATR crystal, and after measurements of spectra at a given NaSCN
concentration, the film was rinsed with 0.5 M NaCl to remove SCN prior
to changing the NaSCN concentration. The SCN peak area was
-

-1

normalized to the area of the –SO3 peak at 1033 cm ....................................... 79
-

Figure 3.9.

Equilibrium SCN concentrations in a (PDADMAC/PSS)30PDADMAC
film exposed to different concentrations of NaSCN. The inset shows an
expanded view at lower solution concentrations .................................................. 80

Figure 3.10.

ATR-FTIR spectra of absorbed SCN

-

-1

(peak at 2058 cm ) in a

(PDADMAC/PSS)30PDADMAC film after a 6 h exposure to an aqueous
solution of 50 mM (green) or 400 mM (red) NaSCN, and after rinsing of
these films with deionized water for 2 h (blue or pink). The pink spectrum
is difficult to see because it overlaps the blue one................................................ 82

xiv

Figure 3.11.

ATR-FTIR spectra of a (PDADMAC/PSS)30PDADMAC film during
exposure to a solution containing 0.5 mM NaSCN and 0.5 mM
-1
2K2Ni(CN)4. The peaks are due to SCN (2060 cm ) and Ni(CN)4
-1

-

2-

(2123 cm ). There was no change in SCN and Ni(CN)4 peaks after 60
min of exposure to the solution.............................................................................. 83
-

-1

2-

-1

Figure 3.12.

Normalized areas of the SCN (2060 cm ) and Ni(CN)4 (2123 cm )
peaks in ATR-FTIR spectra of a (PDADMAC/PSS)30PDADMAC film
during exposure to a solution containing 0.5 mM NaSCN and 0.5 mM
K2Ni(CN)4 .............................................................................................................. 84

Figure 3.13.

Equilibrium
SCN
and
Ni(CN)4
concentrations
in
a
(PDADMAC/PSS)30PDADMAC film exposed to solutions containing 0.5
mM K2Ni(CN)4 and varying concentrations of NaSCN. The total

-

2-

2-

absorption (green triangles) is the sum of the Ni(CN) 4 concentration in
-

the film multiplied by 2 and the SCN concentration in the film ........................ 85
Figure 3.14.

Zeta Potentials of Si wafers coated with (PDADMAC/PSS) n films.
Integer numbers of bilayers represent films terminated with PSS, while
films with an extra ½ bilayer are terminated with PDADMAC ......................... 89

Figure 3.15.

Zeta Potentials of 100 kDa PES membranes coated with
(PDADMAC/PSS)n films. Integer numbers of bilayers represent films
terminated with PSS, while films with an extra ½ bilayer are terminated
with PDADMAC .................................................................................................... 90

Figure A1.

Single-reflection ATR-FTIR experiment with a crystal exposed to a
NaSCN solution. The actual apparatus in this work employed 6
reflections .............................................................................................................. 102

Figure A2.

Schematic drawing of an optical system containing three layers (crystal,
swollen film, and solution) .................................................................................. 106

Figure A3.

Simulated values of absorbance at 2058 cm as a function of k using
equation 11. Refractive index and thickness of the film are taken as 1.345
and 260 nm, respectively. The water above the film had a refractive index
of 1.323.................................................................................................................. 107

-1

xv

Figure A4.

Peak Area versus the concentration of salt in a solution contacting the
ATR crystal for solutions containing NaSCN and K2Ni(CN)4. The inset
shows data for low concentrations .......................................................................108

Figure A5.

ATR-FTIR spectra of 0.1 M NaSCN in aqueous (purple) and 80% ethanol
in water (green) solutions containing 0.2 M PDADMAC. The ATR-FTIR
spectrum of 0.1 M NaSCN in water without PDADMAC is shown in blue ....109

Figure 4.1.

Structures of the polyelectrolytes used in this study .......................................... 120

Figure 4.2.

Reflectance FTIR spectra of (PAH/PSS) 4PAH-coated Au wafers before
and after a 1-h immersion in 0.5 M FeCl3 or 0.5 M CuCl2 or 0.5 M
NaSCN followed by rinsing with water, and the spectra of the CuCl 2 or
FeCl3-treated films, including a film immersed in 5 mM FeCl3, after a 1 h
immersion in 0.5 M NaSCN followed by rinsing. Some of the spectra are
offset for clarity ....................................................................................................124

Figure 4.3.

SEM images of 50 kDa PES membranes after (a) no treatment, (b)
adsorption of a (PAH/PSS)4 film and (c, d) after (PAH/PSS)4 adsorption
followed by immersion in 0.5 M FeCl3 and rinsing with water ........................ 126

Figure 5.1.

Structures of acetaminophen and caffeine .......................................................... 141

Figure 5.2.

UV/Vis spectrum of 20 ppm ACT in water. λ max=243 nm .............................. 143

Figure 5.3.

UV/Vis spectrum of 10 ppm CFN in water. λ max=272 nm .............................. 147

xvi

LIST OF ABBREVATIONS

1,2-BDS

1,2-Benzene Disulfonate

1,3-BDS

1,3-Benzene Disulfonate

1,4-BD

Hydroquinone

AAS

Atomic Absorption Spectroscopy

ACT

Acetaminophen

AFM

Atomic Force Microscopy

AH

Acridine Hydrochloride

ALG

Alginate

AP

Alkyl Phenol

ATR-FTIR

Attenuated Total Reflectance Fourier Transform Infrared
Spectroscopy

az6ac

1,4,7,10,13,16-hexacarboxymethyl 1,4,7,10,13,16-hexaazacyclooctadecane

Bs

Benzene sulfonate

CEES

Chloroethyl Ethyl Sulfide

CFN

Caffeine

CHI

Chitosan

CMR

Catalytic Membrane Reactor

DL-HFMR

Dual Layer Hallow Fiber Membrane Reactor

EDC

Endocrine Disrupting Compound

EE2

Ethinylestradiol

FESEM

Field Emission Scanning Electron Microscopy

FTIR

Fourier Transform Infrared Spectroscopy

HA

Hyaluronic Acid

ICP-OES

Inductively Coupled Plasma Optical Emission Spectrometry

LbL

Layer by Layer
xvii

MCT

Mercury-Cadmium-Telluride

MO

Methyl Orange

MPA

3-mercaptopropionic acid

Mw

Molecular Weight

MWCO

Molecular Weight Cut Off

NF

Nanofiltration

Np

Naphthalene

Ns

Naphthalene 2-sulfonate

P4VP

Poly(4-vinylpyridine)

PAA

Poly (acrylic acid)

PAH

Poly (allylamine hydrochloride)

PAN

Poly (acrylonitrile)

PCBS

Poly(1-[ p-(3'-carboxy-4'-hydroxyphenylazo) benzene
sulfonamide]-1,2-ethandiyl)

PDADMAC

Poly (diallyl dimethyl ammonium chloride)

PE

Polyelectrolyte

PEG

Poly (ethylene glycol)

PEI

Poly (ethylene imine)

PEM

Polyelectrolyte Multilayer

PES

Polyethersulfone

PET

Poly (ethylene-terephthalate)

PGA

Poly (L-glutamic acid)

Ph

Phenol

PhAC

Pharmaceutically Active Compound

PLL

Poly (L-lysine)

PMP

Poly (4-methyl-l-pentene)

PMMA

Poly (methyl methacrylate)

xviii

PRO

protamine sulfate

PS

Polystyrene

PSS

Poly (styrene sulfonate)

PVA

Poly (vinyl amine)

PVS

Poly (vinyl sulfate)

Py

Pyrene

QCM

Quartz Crystal Microbalance

SEM

Scanning Electron Microscopy

TCE

Trichloroethylene

Tp

triphenylene

RO

Reverse Osmosis

UF

Ultrafiltration

UV

Ultraviolet

UV/Vis

Ultraviolet/Visible Spectroscopy

XPS

X-ray Photoelectron Spectroscopy

ZVI NP

Zero valent iron nanoparticles

xix

Chapter 1
Introduction and Background

This dissertation describes the growth of polymer films through layer by layer adsorption
of polyelectrolytes, and the application of these thin films in anion separations, cation
separations, and the removal of pharmaceutically active compounds from water. The research
builds on a large body of prior work on adsorption of multilayer polyelectrolyte films and
enhancing the selectivity of anion and cation transport through these films. To put my research
in perspective, this introduction first discusses the layer by layer (LbL) method for forming
multilayer films. Subsequent sections of the chapter review studies of the incorporation of ions
and other species into multilayer polyelectrolyte films and previous efforts to use these films as
ion-separation membranes. Finally, the introduction gives a brief outline of the dissertation.

1.1.

Layer by Layer (LbL) Assembly of Polyelectrolyte Multilayers

Among the many methods for formation of thin films, e.g. spin and dip coating or
monolayer adsorption, layer-by-layer (LbL) deposition of complementary polymers has emerged
as an especially versatile technique for controlling film thickness and functionality. Figure 1.1
shows the most common form of the LbL method, alternating adsorption of polycations and
polyanions. Operationally, this method occurs through simple immersions of a selected substrate
in polycation and polyanion solutions, with rinsing to remove excess polymer after each
deposition step. Typical polyanions employed for deposition of these films include ionized
1

2

forms of poly(acrylic acid) (PAA), poly(styrenesulfonate) (PSS), poly(vinylsulfonic acid),

1

3

4,5

alginic acid,

hyaluronic acid (HA),

quaternary ammonium functionalities

6-8

10, 11

9

and pectic acid, whereas most polycations contain
12, 13

or protonated amines.

- -- - - -- -- - - -

+ + ++
+ + +
+ + +
++ + +++

Polyanion

+
+
+
+
+
+
+

Figure 1.1.

Water

i) Polyanion
ii) Rinse

-

-

-

Schematic

diagram

of

Water

Polycation

+

i) Polycation
ii) Rinse

+

1 bilayer
Formation

+

layer-by-layer

+
+
+

adsorption

of

polyelectrolyte

multilayers. (Adapted from Science, 1997, 277, 1232-1237). “For interpretation
of the references to color in this and all other figures, the reader is referred to the
electronic version of this dissertation”
(Figure 1.2 shows some of these polyelectrolytes). Nevertheless, LbL methods can also use a
much broader range of multiply charged species including proteins,
17-19

nanoparticles,

and exfoliated inorganic materials.

20, 21

2

16

viruses,

I should note that some LbL

methods also employ other interactions such as hydrogen bonding
30

14, 15

22-26

27-

or covalent linkages.

Figure 1.2.

Structures of polyelectrolytes frequently used in building multilayers.

The LbL strategy has a number of advantages over most coating techniques. First,
because a single adsorption step may deposit as little as a few angstroms (Å) of polymer, this
technique offers control over film thickness at the nanometer scale.

31

Second, conformal

adsorption occurs on substrates with a wide range of geometries, allowing coating of 3dimensional materials such as porous membranes and nanoparticles.

19, 31-37

Third, the wide

range of materials suitable for LbL adsorption and the ability to deposit species in a defined
3

order should afford a wide range of functional films.

17, 33, 34, 38-41

Post-deposition reactions

such as cross-linking and reduction of metal ions to form nanoparticles provide further ways to
modify film properties.
for catalysis,
coatings.

35, 48

42-47

Prior studies of post-deposition reactions yielded functional films

anti-reflective coatings,

49

optical shutters,

40, 50

and superhydrophobic

51

1.2.

Mechanism of Polyelectrolyte Adsorption

The key feature in the adsorption of polyelectrolyte multilayer (PEM) films is charge
overcompensation.

The initial layer adsorbs onto the substrate either by electrostatic or

hydrophilic attractions and creates a charged surface or reverses the substrate surface charge.
Adsorption of subsequent layers again overcompensates the charge on the surface to reverse the
substrate’s charge and allow adsorption of the next layer.

52-54

In many cases, the thickness of multilayer polyelectrolyte films increases linearly with
the number of adsorbed layers.

55-58

This suggests that the extent of charge overcompensation

does not vary greatly with the number of adsorbed layers, so the amount of polyelectrolyte
deposited in each step is approximately constant. However, for some polyelectrolyte systems
film thickness increases exponentially with the number of layers. Schaaf et al. suggested that the
exponential growth occurs when one of the polyelectrolytes diffuses “into” the entire film during
deposition.

59-61

Upon addition of the oppositely charged polyelectrolyte, the previously

adsorbed polyelectrolyte diffuses “out” of the whole film to form a very thick polyanionpolycation complex at the surface (see figure 1.3). Because one of the polyelectrolytes, usually
4

one with low charge density and high swelling in water, diffuses into the entire film, the
thickness of each layer increases with the number of layers.

62

h0

H1
H0
Substrate

Substrate

Step 1
Polycation
Polyanion

h1
H2

H2
Substrate

Substrate
Step 2

Figure 1.3.

Schematic drawing showing the buildup of a polyelectrolyte “bilayer” during
exponential growth. Step 1 includes diffusion of a polyelectrolyte (in this case
a polycation) into the entire film. During step 2, much of the polycation
diffuse out of the film and forms a complex with incoming polyanion. H0 and
H1 are the film thicknesses at the start and at the end of step 1, and H 2 is the
film thickness at the end of step 2. h0 and h1 are the increases in film
thickness due to the deposition of polycation and polyanion, respectively.
(Redrawn from J. Phys. Chem. B, 2009 113 (13), 4232-4241.)

5

1.3.

Factors that Affect Film Growth during LbL Polyelectrolyte
Adsorption

In addition to the polyelectrolyte selected for deposition, a number of adsorption
parameters such as supporting electrolyte concentration and composition,
of the polyelectrolytes,
and temperature

98-100

75-83

pH of the polyelectrolyte solutions,

84-91

63-74

molecular weight

adsorption time,

63, 92-97

also influence the amount of polyelectrolyte deposited in LbL methods.

Understanding the mechanisms of PEM formation and the role of process parameters in
determining thicknesses and interfacial properties of multilayer films is fundamental for their
future applications. Below I discuss the effects of some of these variables on LbL film growth.

1.3.1. Effect of Supporting Electrolyte
1.3.1.1. Effect of Supporting Electrolyte Concentration
A number of groups examined the importance of supporting electrolyte on the growth of
polyelectrolyte multilayer films.

63-67

In the absence of added salt, polyelectrolytes are highly

extended to maximize the distance between charged repeat units of the polymer. Under these
conditions, adsorbed layers are thin and overcompensate the surface charge only slightly (see
Figure 1.4).

101

For example, the thickness of a 10-bilayer PSS/poly(diallyl dimethyl ammonium

chloride) (PDADMAC) film prepared without addition of salt is about 60 Å.
average thickness per layer is only 3 Å.

6

63

Thus, the

1st layer

5th layer

no salt

with salt

Figure 1.4.

Sketch of polyelectrolyte layers adsorbed in the presence and absence of salt.
The absence of salt results in thin layers of extended polyelectrolytes,
whereas at high ionic strength, coiled polymers form thicker layers.
(Redrawn from Colloids Surf. Physicochem. Eng. Aspects 2002, 198-200, 293304.)

Many studies show a dramatic effect of supporting electrolyte concentration on the
thickness of PEMs.

63, 64, 68-74

In Figure 1.5, the thickness of a 10-bilayer PSS/PDADMAC

film shows an almost linear dependence on salt concentration between 10
in contrast to the early findings of Lvov et al.,

102

-2

and 2 M.

63

This is

who reported that the thickness of a layer pair

of PSS/poly(allylamine hydrochloride) (PAH) is proportional to the square of the ionic strength
in deposition solutions. However in both cases, thickness increases dramatically upon addition
of salt to deposition solutions. The reason for the differences in the above-mentioned systems
likely stems from the nature of the polyelectrolytes. (PDADMAC has a lower charge density
than PAH, and PDADMAC-containing films are prone to very high swelling in some cases.

7

53,

103-105

) The excess salt presumably screens charges on the polymer chains and allows them to

coil and form thicker layers (Figure 1.4). In addition, charge screening may require adsorption
of more polyelectrolyte to compensate the opposite charge in the previously adsorbed film.

Figure 1.5.

Thicknesses of LbL (PSS/PDADMAC)10 films deposited on silicon wafers
using a series of

NaCl concentrations in the polyelectrolyte deposition

solutions. (Reprinted with permission from Macromolecules 1999, 32(24),
8153-8160)
Using atomic force microscopy (AFM), McAloney et al. examined the morphology of
multilayer PDADMAC/PSS films deposited from solutions with salt concentrations ranging from
-4

10

to 1.0 M.

66

The roughness and thickness of (PDADMAC/PSS) 10 films deposited from

solutions with less than 0.1 M NaCl were approximately constant at 2 nm and 10 nm,

8

respectively, and the thickness increased linearly with the number of adsorbed bilayers. In
contrast, for PDADMAC/PSS films deposited from an ionic strength of 0.3 M or higher, the
growth rate rose as the polyelectrolyte attained a coil-like structure and surface roughness also
increased. The increase in surface roughness may contribute to the increase in film thickness.

66

Recently, Lee et al. studied the role of salt in the adsorption of DNA films assembled from
oligonucleotides composed of two homopolymeric diblocks (polyA nGn and polyTnCn).

A

combination of crystal microgravimetry and flow cytometry measurements showed that greater
film growth occurred as the salt concentration increased from 0.1 to 1 M.
salt concentrations may lead to delamination of the film.

106

However very high

107

1.3.1.2. Effect of Type of Supporting Electrolyte
In addition to the supporting electrolyte concentration, the identity of the supporting
electrolyte also affects film thickness. Dubas et al. determined the thicknesses of 10-bilayer
PDADMAC/PSS films adsorbed from solutions containing different salts. Table 1.1 shows
results for a cation series, with a constant anion (chloride). The trends in film thickness from
RbCl to LiCl suggest that the least hydrated cations

108

give rise to the thickest polyelectrolyte

films. The ion-exchange properties of protons and lithium are similar, which might explain the
comparable film thicknesses when adsorbing PEMs from HCl and LiCl solutions. CsCl and
RbCl have similar thicknesses even though the hydration number for Rb is lower than Cs.
Although in these studies the concentration of the ions also decreases slightly with the degree of
hydration to maintain a constant activity of the different salts, this change is not sufficient to
account for the large changes in film thickness with varying deposition solutions.

9

109

Table 1.1.

Ellipsometric thicknesses of (PDADMAC/PSS)10 films adsorbed on a Si
wafer from solutions containing 1 mM polyelectrolyte (with respect to the
repeating unit) and a series of supporting electrolytes with an activity of 0.33.
Hydration numbers, n, are also given for each ion. Adsorption occurred for
5 min per layer with 3 min of rinsing after deposition of each layer.
a

Salt

Concentration, M

ncation

HCl

0.43

2.7

LiCl

c

nanion

thickness, Å

b

2.0

705

0.44

5.2

2.0

820

NaCl

0.48

3.5

2.0

903

KCl

0.50

2.6

2.0

1200

RbCl

0.52

2.4

2.0

1466

CsCl

0.54

2.1

2.0

1420

a

Concentrations were selected based on activity coefficients to maintain a constant activity of
0.33.
b
+
Hydration number for H3O .
c

Ellipsometric thickness, ±3%. (Taken from Macromolecules 1999, 32(24), 8153-8160.)
Several other studies examine how the supporting salt anion in polyelectrolyte solutions

affects the growth of PSS/PAH films.

110-112

In these studies, the supporting electrolytes are all

sodium salts, but the nature of the anion varies along the Hofmeister series from cosmotropic to
−

−

−

−

−

−

−

− 112, 113

chaotropic anions (F , HCOO , BrO3 , Cl , ClO3 , Br , NO3 , ClO4 ).

For all the

investigated anions, the PSS/PAH film thickness increases linearly with the number of
deposition steps.

110

Cosmotropic anions bind strongly to water molecules and induce some

structure in the solution that results in “salting-out” of molecules such as proteins. Such anions
enhance the hydrogen bonding ability between water molecules to decrease the freedom of
motion. Because of their low electronegativity, high polarizability, and weak electrostatic fields,
10

chaotropic anions destabilize the hydrogen bonding between ion and solvent molecules to
increase the solubility of some molecules.

114

Figure 1.6 shows that for PSS/PDADMAC

multilayers, the chaotropic anions lead to thicker bilayers than cosmotropic anions.

112

The

chaotropic anions bind strongly to polycations, thus reducing the charge density on the
polyelectrolyte. This leads to the formation of coil structure thereby increasing the thickness of
the layer.

The film permeability to ions also changes when different salts are present during
adsorption. In electrochemical measurements, films consisting of up to 9.5 PSS/PAH bilayers
4−

are permeable to hexacyanoferrate(II) ions, Fe(CN) 6 . However, the peak intensities of anodic
3-/4−

and cathodic Fe(CN)6

currents depend on the sodium salt present during deposition, and
−

−

−

−

currents increase in the order F < Cl < NO3 < ClO4 . In contrast, PSS/PAH films prepared in
2+

the presence of NaF, NaCl, or NaNO3 are impermeable to Ru(NH3)6

after adsorption of the

third PAH layer, as both the anodic and cathodic current become zero. These results suggest the
presence of excess positive charges in the PSS/PAH multilayer film, which leads to a positive
Donnan potential. The stronger binding of PAH with anions as their chaotropic nature increases
likely results in a more loopy conformation that leads to less dense films and thus to a high
4−

mobility of the Fe(CN)6

ions through the films.

110

11

90

Br-

80

NO3-

Thickness (nm)

70

ClO3-

60

ClBrO3-

50
40

HCOOF-

30
20
10
0
2

4

6

8 10 12 14 16 18 20

Number of layers
Figure 1.6.

Ellipsometric thicknesses of (PSS/PDADMAC) 20 films deposited from
solutions containing 0.1 M of the sodium salt of the listed anions. (Reprinted
with permission from Langmuir 2004, 20(9), 3679-3683.)

1.3.2. Effect of Polyelectrolyte on Multilayer Adsorption
A number of polyelectrolyte properties (chemical structure, molecular weight,
concentration, and degree of ionization) affect the LbL growth of PEMs.

This subsection

examines each of these factors in detail.
1.3.2.1. Effect of Polyelectrolyte Type on Multilayer Growth
The mechanism of PEM growth, linear or exponential, depends on the constituent
polyelectrolytes as well as the ionic strength of the deposition solutions. PSS/PAH multilayers
provide an example of films whose thickness always increases linearly with the number of
12

adsorbed layers.

57, 58, 115

In contrast, polyanion/polycation systems that follow an exponential

growth mechanism include poly( L-glutamic acid) (PGA)/poly(L-lysine)
PGA/PAH,

116

PDADMAC/PSS (at high ionic strength)

66

and HA/chitosan (CHI).

steady-state incremental thickness increase per bilayer for PSS/PDADMAC
than the increment for PSS/PAH

117

(PLL),

63

59,

76, 81

61

The

is 7 times higher

under the same conditions, which demonstrates the

importance of polyelectrolyte type in determining ultimate film thickness. In addition to the
chemical structure of the polyelectrolytes, film thickness also varies with polyelectrolyte
concentration, molecular weight, and charge density, which often varies with pH. The following
sections discuss these factors.
1.3.2.2. Effect of Polyelectrolyte Molecular Weight
Several reports examined the influence of polyelectrolyte molecular weight (Mw) on the
thicknesses of PEMS.

75-81

In early studies Lösche et al. showed that the thicknesses of

PSS/PAH films are independent of PSS molecular weight.

117

Using in situ quartz crystal

microbalance (QCM) measurements, Richert et al. evaluated the effect of the Mw of CHI on the
growth of HA/CHI. The film thickness decreased by a factor of 1.75 when using CHI with a Mw
of 460000 rather than 110000.

81

(The HA Mw was 400000 in both cases.)

13

Figure 1.7a. Influence of the molecular weights of HA and PLL on the thickness of
(HA/PLL)n films grown using HA with Mw = 130000 and PLL with M w
values of (▪) 20000, (○) 55000, and (▴ ) 360000. (Reprinted with permission
from Langmuir 2007, 23(4), 1898-1904.)

14

Figure 1.7b. Influence of the molecular weights of HA and PLL on the thickness of
(HA/PLL)n films grown using PLL with M w = 360000 and HA with M w
values of (▪) 130000 and (○) 400000. (Reprinted with permission from
Langmuir 2007, 23(4), 1898-1904.)
Sun et al. found that growth rate of PAH/PAA films formed from lower Mw PAA
(15000) is greater than the growth rate for films adsorbed using higher Mw (90000) PAA.
Porcel et al. similarly examined the adsorption of HA/PLL

77

and PGA/PAH

83

82

systems by

varying the Mw of the polyelectrolytes. The films in both studies were constructed using a
spraying method (with rinsing after each layer) and were dried after deposition of every 2
bilayers to measure the ellipsometric thicknesses. In the case of the HA/PLL system, Figure
15

1.7a, 1.7b show that the thickness increases with the Mw of PLL but decreases with an increase
in the Mw of HA.

77

For the PGA/PAH system the thickness increments were almost

independent of the molecular weights of both polyelectrolytes, PGA, 44000 or 100000 and PAH,
15000 or 55000.

83

Based on all of these studies, it is difficult to predict how film thickness will

vary with polyelectrolyte molecular weight.
1.3.2.3. Effect of Polyelectrolyte Concentration on Film Thickness
Several groups showed that the amount of adsorbed polyelectrolyte in PEM films
increases with increasing polyelectrolyte concentration in the deposition solution.
119

23, 63, 95, 118,

Figure 1.8 shows that film thickness increases by as much as 30% on increasing

polyelectrolyte concentration from 2 to 50 mM when depositing a PSS/PDADMAC film on a Si
substrate by spin coating.

63

Note that this effect is considerably less than the effect of

supporting electrolyte concentration on film thickness.

Fleer et al. suggest that at high

polyelectrolyte concentrations, many polyelectrolyte chains interact with the interface at the
same time and each can adsorb on just a few binding sites, which results in relatively thick films.
Conversely, at lower polyelectrolyte concentrations, incoming polyelectrolytes will interact with
many binding sites at the surface to give thin films.

120

Baba et al. found that the thickness of the

PEM changes drastically by varying polyelectrolyte concentrations.

Table 1.2 shows the

thickness per bilayer for three different combinations of polyelectrolyte concentrations. These
thicknessess were calculated from adsorbed mass (obtained from QCM) using Sauerbrey
equation.

94

The difference in thicknesses for the films studied by Dubas et al. and Baba et al.

16

adsorbed from 10 mM PSS, PDADMAC might be due to their preparation method (spin coating
and dipping).

Figure 1.8.

63, 94

Thickness of (PSS/PDADMAC)5 films as a function of polyelectrolyte
concentration. Deposition occurred on a silicon wafer from polyelectrolyte
solutions containing 1.0 M NaCl (aq). (Reprinted with permission from
Macromolecules 1999, 32(24), 8153-8160.)

17

Table 1.2.

Thickness changes per layer for PDADMAC/PSS films prepared using
different polyelectrolyte concentrations. All the films were made with no
added salt in deposition solutions. (Taken from Colloids Surf. Physicochem.
Eng. Aspects 2000, 173(1-3), 39-49.)
Thickness per layer (nm)

PDADMAC (0.01 M)/PSS (0.01 M)

28.67

PDADMAC (0.01 M)/PSS (0.001 M)

3.09

PDADMAC (0.001 M)/PSS (0.01 M)

7.06

Garg et al. studied alternating adsorption of PAH and the polymeric dye poly(1-[ p-(3'carboxy-4'-hydroxyphenylazo)benzenesulfonamido]-1,2-ethandiyl) (PCBS, Figure 1.9) as a
function of polyelectrolyte concentration using a quartz crystal microbalance.

121

For films

formed at pH 7, the mass adsorbed increased by 50% when the PAH concentration increased
from 1 to 5 mM, but the thickness did not increase further on going from a polyelectrolyte
concentration of 5 to 10 mM repeat units.

This suggests a saturation limit for polymer

adsorption rather like an adsorption isotherm plateau.

Figure 1.9.

Structure of the polymeric dye poly(1-[ p-(3'-carboxy-4'-hydroxyphenylazo)
benzenesulfonamido]-1,2-ethandiyl) (PCBS).

18

1.3.2.4. Effect of Deposition pH (or Degree of Polyelectrolyte Ionization) on the Growth of
PEMs Containing Weak Polyelectrolytes
The pH of weak polyelectrolyte deposition solutions dramatically affects film thickness
along with permeability and morphology.

84-88

In weak polyelectrolytes, the ionization of

groups such as amines and carboxylic acids, and hence the polymer charge density, is a strong
function of pH. An increasing charge density on the adsorbing polymer will lead to thinner
adsorbed layers, whereas increasing charge density in the previously adsorbed polymer will
favor thicker adsorbed layers. For PAH/PAA films made with the pH of both polyelectrolytes at
2.5, about 70% of the carboxylates of adsorbed PAA are still in the carboxylic acid form. In
contrast, when the pH of both polyelectrolyte deposition solutions is 4.5, about 70% of the
functional acid groups in the film exist in the ionized carboxylate form. The thicknesses of
(PAH/PAA)30 films deposited at pH 2.5 and pH 4.5 are approximately 120 nm and 210 nm,
respectively.

86

Remarkably, for PAA/PAH films the thickness of an adsorbed polycation or

polyanion layer can vary from 5 to 80 Å with changes in deposition pH.
values may completely prevent film growth by favoring desorption.

19

90

89

However extreme pH

32

(a)
(a)

Thickness (nm)

28

pH = 4 = 4
pH

24

pH = 7 = 7
pH

pH
pH = 9 = 9

20

pH
pH = 11 = 11

16
12
8
4
0
0

1

2

3 4

5

6

7

8

9 10 11 12 13 14 15

No of PE layers
45

(b)
pH = 4

40

pH = 7

pH = 4

35

Thickness (nm)

pH = 9

pH = 7

30

pH =
pH = 9 11

25

pH = 11

20
15
10
5
0
0

1 2

3 4

5 6 7 8

9 10 11 12 13 14 15

No of PE layers
Figure 1.10. Thickness as a function of number of polyelectrolyte (PE) layers for (a)
PAH/PSS and (b) PDADMAC/PAA multilayers deposited using different
deposition pH values for both polyelectrolytes. (Redrawn from Colloids Surf.
Physicochem. Eng. Aspects 2008, 321(1-3), 258-261.)
20

Changes in charge density due to variations in pH depend on the specific polyelectrolyte
system. Figure 1.10a shows that for PAH/PSS the thickest films form under conditions where
the charge of the weak polyelectrolyte, PAH, is low (basic pH). In contrast, PDADMAC/PAA
films are thicker at low pH (Figure 1.10b), again where the charge density of the weak
polyelectrolyte is low. At low charge densities the weak polyelectrolytes will form more coiled
conformations due to lower electrostatic repulsion between repeat units. Additionally weaker
electrostatic repulsion between adsorbing polyelectrolyte molecules should contribute to the
formation of thicker films.

91

1.3.3. Effect of Polyelectrolyte Adsorption Time on the Growth of PEMs
The literature on polyelectrolyte multilayers contains a wide range of estimates (seconds
to hours) of the time required to form a layer.

63, 92-97

This wide range of times likely stems

from differences in the structure, molecular weights, and deposition pH values of the
polyelectrolytes used to build the multilayer films. Mass transport of polymer to the surface
probably limits most of the layer formation, and methods that enhance mass transport such as
spin coating can increase the initial rate of layer formation.

63, 122, 123

Addition of the final 20%

or so of the film likely occurs under a slower regime of permeation though surface polymer and
rearrangements. Typically, more than 95% of adsorption occurs during the first 1 min after
exposure to the polyelectrolyte solution in the case of PAH and PCBS in PAH/PCBS films

121

and in PDADMAC /PCBS films the first order rate constants of adsorption were 1.5 and 2 min
for PDADMAC and PCBS respectively.

124

21

1.3.4. Effect of Temperature on the Growth of PEMs
Increased deposition temperatures significantly enhance the thicknesses of PAH/PSS and
PDADMAC/PSS films, but this effect is stronger for PDADMAC/PSS than for PAH/PSS.

98, 99

At 55 °C, film thickness increases exponentially with the number of layers for PDADMAC/PSS
while it increases linearly for PAH/PSS. Figure 1.11 shows the thicknesses of manually dipped,
(PDADMAC/PSS)10 films deposited at temperatures ranging from 10 to 70
polyelectrolyte deposition solutions each contained 1 M NaCl.)

99

°C.

(The

The thickness increases in an

approximately linear fashion with respect to temperature throughout this range. Recent work by
Büscher et al. also shows that increasing the temperature of the deposition solution increases the
thickness of PAH/PSS deposited in the presence of KCl. Polyelectrolytes tend to precipitate at
higher temperatures, which results in the formation of rough, thick multilayers.

100

Secondary

interactions like hydrogen bonding, hydrophobicity and van der Waal’s forces, which are
temperature dependent, also affect the thickness of polyelectrolyte multilayer films.

22

125

Film Thickness (nm)

250
200
150

100

50

0
0

20

40

60

80

Temperature (ºC)
Figure 1.11. Effect of deposition temperature on the thickness of a (PDADMAC/PSS) 10
multilayer. (Reprinted with permission from Langmuir 2003, 19(22), 93119314.)

1.4.

Swelling and Doping of Polyelectrolyte Multilayers

In PEM formation, pairing of the charge on a polymer segment with the charge on an
oppositely charged polymer segment is termed intrinsic compensation. Conversely, electrical
compensation of charges on the polymer segments by counter ions from salt is called extrinsic
compensation.

Schlenoff termed the introduction of ion-exchange sites by extrinsic

compensation as “doping” the film.

126

The fractions of intrinsic and extrinsic charge

compensation have a dramatic effect on film swelling because intrinsically compensated charge
cross-links the film, whereas the formation of ion-exchange sites disrupts ionic cross-links and
23

likely renders the film more hydrophylic. Changes in ambient pH can also affect film swelling
through changes in charge compensation and introduction of new charged sites.
Dubas et al. reported that the volume swelling coefficients, Qswell (% swelling/salt
concentration in the swelling solution), for PAA/PDADMAC and PSS/PDADMAC are about
-1

400 M

-1

-1

and 20 M , respectively, whereas PSS/PAH (Qswell ~1 M ) films do not show any

change in thickness with increasing salt concentration.

105

Miller et al. used ellipsometry and

AFM to show that swelling of HA/CHI film is 4-fold higher than the swelling of PSS/PAH films
in water. Additionally, with PSS/PDADMAC films prepared from solutions containing 0.1 M
NaCl, the composition of the terminating layer does not make any difference in swelling.
However, when corresponding coatings are deposited from solutions containing 0.5 M NaCl,
films terminated with PDADMAC swell 4 times more than films terminated with PSS.

104

Carrière et al. also showed related effects for PSS/PAH films using neutron reflectometry
studies, where PAH terminated films swell 25 % less than PSS terminated films.

127

Using attenuated total internal reflectance Fourier transform infrared (ATR-FTIR)
spectroscopy, Jaber et al. found that the number of perchlorate ions that enter PDADMAC/PSS
after film formation is 2 times higher than the number that enter PAH/PSS.

128

Thus, the level of

intrinsic compensation is much greater in PAH/PSS than PDADMAC/PSS films. Lebedeva et
al. used much higher NaCl concentration to study swelling and found that at salt concentrations
greater than 3 M, softening of PAH/PSS, occurs probably due to breaking of some ionic
crosslinks.

129

24

1.5.

Applications of Polyelectrolyte Multilayer Films

Multilayer polyelectrolyte films are attractive in a wide range of potential applications
such as encapsulation of drugs
45, 134-140

37, 130, 131

antibacterial coatings,

membrane reactors.

33, 151, 152

and enzymes,

6, 141-146

fuel cells,

132, 133

147-149

membrane-based separations,

photovoltaic membranes,

150

3,

and

Below this introduction briefly explores some of these areas.

1.5.1. Membrane-based Separations
1.5.1.1. Nanofiltration Membranes
Nanofiltration (NF) is a pressure-driven separation technology similar to reverse osmosis
(RO), but the membrane has a pore size between 0.5 and 2 nm, and operating pressures are
between 5 and 40 bar.

153

Typically, these membranes allow passage of monovalent ions, but

they reject multivalent ions and low Mw organic compounds to a much greater degree. NF
applications include water softening, removal of organic pollutants, and filtration of acids and
bases, and because NF operates at lower pressures than RO, energy costs are less than for a
comparable RO treatment system.
Similar to RO membranes, NF membranes are asymmetric materials that contain a thin
selective layer on a highly permeable support. Several studies show that LbL adsorption of
oppositely charged polyelectrolytes on porous supports is a suitable method for forming NF
membranes for the separation or removal of ions,

154-156

sugars,

157, 158

and dyes.

159

In

particular, the high surface charge of many PEM films makes them attractive for separating

25

monovalent and divalent ions. The performance of a membrane is reported in terms of rejection,
R and selectivity, S which are defined by equations 1.1 and 1.2,


C perm
R  1 

C feed



  100%



C A, perm CB, feed

S

C A, feed CB, perm



------------ (1.1)

100  RA

------------ (1.2)

100  RB

where Cfeed and Cperm are the concentrations of solute A or B in the feed and the permeate,
respectively, and RA and RB are the rejections of solute A and B.
The solution flux, J, is calculated from the measured volume V of the permeate flowing across
the membrane of area A in the time period Δt as shown in equation 1.3

V
3
2
m /(m day)
At

J

------------ (1.3)
-

Harris et al. showed that a (PAH/PSS)5 film deposited on porous alumina exhibits a Cl
/SO4

2-

-

transport selectivity of 7 and a Cl /[Fe(CN6)]

3-

selectivity of 310 in diffusion dialysis.

160

Similarly, Krasemann et al. showed that a 60-bilayer film of PAH/PSS membrane has a transport
diffusion dialysis selectivity of 45 for NaCl over Na2SO4, but using such a large number of
layers is not practical for large-scale membrane fabrication.

161

Deposition of hybrid

PSS/PAH/PAA films on alumina supports and cross-linking of PAA/PAH at 115 °C for 2 h
enhances the ion-transport selectivity of layered polyelectrolyte membranes to give an average
-

diffusion dialysis Cl /SO4
-

a Cl /SO4

2-

2-

selectivity as high as 360.

45

In nanofiltration, Hong et al. achieved
3

2

selectivity of 26 and a solution flux of 2.7 m /(m day) at 4.8 bar with a
26

(PSS/PDADMAC)3PSS film on alumina. By terminating films with the polyanion PSS, and
depositing this layer from a solution containing 1.0 M NaCl to increase surface charge,
multivalent anions are electrostatically repelled from the membrane more than monovalent
anions.

139

Balachandra et al. examined templating of films with Cu

2+

to enhance the negative

charge in PAA/PAH films and increase anion-transport selectivity.
polyelectrolytes from solutions containing Cu

2+

Deposition of

ions and subsequent removal of the Cu

2+

from

the carboxylate groups to which they are bound ultimately leaves negatively charged COO
groups in the interior of the film.
-

a fourfold increase in Cl /SO4

2-

162

-

The presence of excess negative charge in the film leads to

selectivity.

Optimization of cation separations with PEM films, particularly sodium and magnesium
separations, is a less thoroughly explored area than anion separations. Krasemann et al. showed
that in diffusion dialysis, a (PAH/PSS) 5 membrane exhibits a selectivity of 31.3 for NaCl/MgCl2
permeation, whereas (PAH/PSS)60 membranes show a selectivity of 112.5. Interestingly, these
membranes terminated with PSS and still exhibit high NaCl/MgCl2 permeation selectivities.

161

Jin et al. studied rejections of chloride and sulfate salts of sodium and magnesium using 60bilayer poly(vinyl amine) (PVA)/poly(vinyl sulfate) (PVS) films on poly(acrylonitrile)
(PAN)/poly(ethylene terephthalate) (PET) supporting membranes.

The rejection of NaCl

increased with applied pressure but was independent of feed concentration. As pressure was
raised from 5 bar to 40 bar for 0.01 M Na2SO4, the rejection increased from 96.4% to 98.5%.
27

11

Of course, the large number of polyelectrolyte layers led to a very low film permeability.
+

2+

Ouyang et al. optimized the PSS/PAH PEM system to achieve a Na /Mg
3

2

high as 22 and a solution flux of 0.85 m /(m day) at 4.8 bar.
+

2+

Toutianoush et al. increased the Na /Mg
2+

45 to 55 by adsorbing Cu

NF selectivity as

163

selectivity of 60-bilayer PVS/PVA films from

ions in the film. They concluded that the copper ions in the

membrane form a complex with the amino groups of PVA, thereby increasing the crosslink
density in the membrane.

164

However, the Cu

2+

adsorption may also increase the amount of

positive charge in the membrane to enhance cation exclusion. Using 60-bilayer hexa-p-sulfonato+

2+

calix[6]arene/PVA and PSS/PVA films the same group achieved Na /Mg
and 37.2 in diffusion dialysis.

165

selectivities of 71.9

They also reported that membranes containing 60 bilayers of

PVA and 1,4,7,10,13,16-hexacarboxymethyl 1,4,7,10,13,16-hexaazacyclo-octadecane (az6ac)
exhibit a separation factor of 28 for NaCl/MgCl2 permeation in diffusion dialysis.

166

Jin et al. investigated the transport of phenol (Ph), hydroquinone (1,4-BD), naphthalene
(Np), pyrene (Py), triphenylene (Tp), alkali metal salts of benzene sulfonate (Bs), naphthalene 2sulfonate (Ns), methyl orange (MO), and isomeric benzene disulfonates (1,2-, 1,3BDS)compounds across PDADMAC/PSS and PAH/PSS multilayer membranes.

167

They

achieved separation factors of α(Ph/Py) ≈ 13 using (PDADMAC/PSS)60 and α(Ph/Np) ≈ 28,
α(Bs/Ns) ≈ 4 and

α(Bs/1,3-BDS) ≈ 65 using (PAH/PSS)60 membranes.

167

For neutral

compounds separation is based on size-selective transport and for charged compounds separation
is based on size and charge of the aromatic compounds. Size-based separation depends on the
28

effective pore size of the membrane, which in turn varies with the type of polyelectrolyte in the
film.
1.5.1.2. Gas Separation Membranes
Stroeve et al. alternately adsorbed PAH and PSS on porous polypropylene supports and
on highly permeable dimethyl silicone membranes to examine the gas separation properties of
PAH/PSS films.

168

The CO2/N2 selectivity on the polypropylene membrane was 23.8 for a

(PAH/PSS)20 film at 50

°C.

A (PAH/PSS)50-coated silicone membrane showed CO2/N2

selectivities that were 1.25 times the selectivity of the bare silicon membrane.
Leväsalmi et al.

169

deposited (PAH/PSS)200 on an oxidized poly(4-methyl-1-pentene)

(PMP) support. After film formation, the O2/N2 selectivity of the PMP increased 4-fold, the
H2/N2 selectivity increased 10-fold and the H2/O2 selectivity doubled. Van Ackern et al.

134

also

demonstrated that the CO2/N2 selectivity of (PAH/PSS)60 films on etched ion-track
polycarbonate supports was 2.42, which is due to greater solubility of CO2 than N2 in the
PAH/PSS film.

However, the selectivity for CO2 over N2 was only 1.5 with (poly(4-

vinylpyridine) (P4VP)/PSS)60 as the separation layer.

170

In another study, Blumstein et al.

prepared polyelectrolyte multilayers from poly(N-octadecyl-2-ethynylpyridinium bromide) and
PSS on Nafion and achieved an O2/N2 separation factor of 3.5.

171

Overall these studies

demonstrate that polyelectrolyte films are defect free and show some selectivity in gas
separations. However, in most cases the selective membranes contain large numbers of bilayers,
which is not attractive for large-scale membrane fabrication.
29

1.5.1.3. Pervaporation Membranes
A number of studies demonstrated effective pervaporative separations of liquid-liquid
mixtures using polyelectrolyte membranes.

136,137,170

Krasemann et al. compared ethanol-water

separations by polyelectrolyte multilayer membranes with different polycations, and separation
factors varied between 1.9 for (PDADMAC/PSS) 60 and 12.0 for poly(ethylene imine)
(PEI)/PSS)60.

170

Tieke et al. improved the separation efficiency for ethanol-water mixtures by

using membranes made of polyelectrolytes of high charge density. A maximum water-ethanol
separation factor of 700 occurred with (PVA/PVS) 60 on a PAN/PET support.

3

Meier-Haack et

al. used polyamide supports and achieved water-ethanol separation factors as high as 1400 after
adsorption of (PEI/PAA)6.

137

Very recently Zhang et al. reported composite membranes with a separation factor of 304
−2 −1

and a permeation flux of 0.512 kg m

h

for pervaporation of 95 wt% ethanol-water mixtures

at 75 °C. The membranes used in this study were prepared by depositing PEI on hydrolyzed
PAN ultrafiltration (UF) membranes under a DC electric field. These films gave fairly good flux
values and high separation factor as a very uniform and dense film was formed due to rapid
migration of positively charged PEI towards negatively charged PAN under the influence of the
electric field.

172

In other work, PDADMAC/PSS, PDADMAC/PAA and PEI/PAA membranes

deposited in an electric field separated isopropanol-water mixtures (90/10, w/w) very efficiently.
The separation factor and permeation flux of a (PEI/PAA) 4PEI membrane assembled in an
−2 −1

electric field were 1289 and 1.71 kg m

h

at 50 °C. The PEMs formed in an electric field

30

have a high selectivity and low permeability compared to normal PEM with the same number of
bilayers.

173

1.5.1.4. Membranes for Protein Separation and Purification
Membrane-based processes are also beginning to play critical roles in the separation and
purification of biotechnological products, and polyelectrolyte films in porous supports can serve
as novel membranes for protein capture.

Müller et al. reported that PEI/PAA multilayers

selectively bind one protein from a mixture of concanavalin A and lysozyme.

At pH 7.3

(PEI/PAA)3 preferentially adsorbed positively charged lysozyme and (PEI/PAA) 2PEI adsorbed
negatively charged concanavalin A.

174

PEM films formed in the pores of membrane terminated

with a polyanion have cation-exchange sites as shown in Figure 1.12. These cation-exchange
sites bind positively charged proteins like lysozyme, and the lysozyme-binding capacities of
(PSS/PEI)3PSS films increase with the ionic strength of capping layer to values as high as
16 mg/mL of membrane.

175

Several studies examined the interactions between proteins and polyelectrolyte
multilayers. In the case of human serum albumin and PAH/PSS multilayers, the protein adsorbs
on both PAH- and PSS-terminated films. However, on PSS-terminated multilayers a dense
protein monolayer forms, whereas on PAH-ending multilayers, adsorbed protein films have
thicknesses as high as four times the largest native protein dimension.

176

Salloum et al. studied

the adsorption of serum albumin, fibrinogen, and lysozyme on different PEM films. The
adsorption of the proteins depends on the surface charge, protein charge and thickness of the
PEM film.

177

31

Figure 1.12. Schematic diagram showing the binding of positively charged lysozyme to a
PEM film terminated with a polyanion. Charges are only shown for the
terminal layer. (Reprinted with permission from J. Membr. Sci. 2010, 354(12), 198-205.)

1.5.2. Encapsulation of Drugs and Enzymes
LbL technology is also attractive for encapsulation of proteins and other bioactive
compounds.

Encapsulation occurs by either directly coating the drug microcapsules with

polyelectrolyte multilayers, or by removing a template to form hollow polyelectrolyte
microcapsules and then loading the biomolecules into the capsules. Careful selection of the
encapsulating polyelectrolytes, including the incorporation of functional polyelectrolytes or
nanoparticles, allows tailoring of capsule properties for controlled and triggered release and
responsiveness to temperature, pH, light and magnetic fields. Micro-encapsulation by LbL
adsorption shows potential applications in biochemistry, pharmacy, controlled release, cosmetics
and catalysis.

178, 179

32

As examples, Caruso et al. encapsulated an enzyme (catalase) through LbL adsportion
onto enzyme crystal templates.

180

Encapsulated enzymes maintain a high activity because the

polyelectrolyte films restrict enzyme denaturation and degradation.

Multilayer films of
181

protamine sulfate (PRO) and PSS formed on ibuprofen-loaded CaCO3 microparticles
acridine hydrochloride (AH) deposited in CHI/alginate (ALG) nanocapsules

182

and

were used for

sustained drug release by Wang et al. The porous CaCO3 particles are useful for loading drugs
in an amorphous state, and the polyelectrolyte multilayer films coated on these particles help
control drug release.

181

The CHI/ALG multilayers were formed on polystyrene (PS)

nanoparticles which were later dissolved by tetrahydrofuran.

Positively charged AH was

deposited due to electrostatic interactions with negatively charged residual PS copolymer with
styrene sulfonate units in the capsule.

182

1.5.3. Catalytic Membrane Reactors
When applied in porous supports, LbL adsorption of polyelectrolytes and either enzymes
or precious metal nanoparticles provides a convenient method of forming catalytic membranes.
Such catalytic membrane reactors (CMRs) are attractive because they allow flow-through
reactions and avoid the need to separate the catalyst from reaction mixture.

183

Dotzauer et al.

showed that the adsorption of nanoparticle/polyelectrolyte films in either polymeric or alumina
membranes effectively immobilizes the gold nanoparticles without inhibiting access to catalytic
sites (Figure 1.13). The rate constants for nanoparticle-catalyzed 4-nitrophenol reduction in
solution and in membranes were essentially identical.

33

Alumina membranes modified with

nanoparticle/polyelectrolyte films catalyze the reduction of >99% of 0.4 mM 4-nitrophenol in
membrane residence times of 6.1 ms.

33

Weimann et al. encapsulated a biocatalyst, lipase B from Candida antarctica in a
PAH/PSS film on CaCO3 colloid particles. A slight increase (32%) in the enzyme catalytic
performance for esterification of caprylic acid and 1-octanol in n-hexane was observed because
immobilization results in better dispersion of the enzymes in organic solvents.

152

Smuleac et al.

fabricated enzyme-containing PEMs inside membrane pores to achieve high loading of
biomolecules with minimum loss of activity.

133

Specifically, adsorbed glucose oxidase and

alkaline phosphatase catalyzed the reactions shown below.

glucose  O2  gluconic acid  H2O2
p  nitrophenyl phosphate  p  nitrophenol  phosphate
Electrostatic immobilization of glucose oxidase in a PLL/PSS/PAH-modified membrane resulted
in a maximum loading of 699 µg of enzyme and 76% enzyme activity relative to enzyme activity
obtained in a homogeneous phase. Similarly, alkaline phosphatase retained 71% of its activity
after immobilization in a membrane. The advantage of using immobilization in membranes is
the convective flow, which eliminates diffusion limitations and transports the product away from
the catalytic site during operation.

133

34

Figure 1.13. Schematic diagram of gold nanoparticle (

) /polyelectrolyte (

) films

immobilized in membranes for 4-nitrophenol reduction. (Reprinted with
permission from Nano Lett. 2006, 6(10), 2268-2272.)
Increasing concern over the use of chemical warfare, combined with more frequent
potential exposure of soldiers and emergency care providers to toxic chemicals has heightened
the need for new protective measures. Reactive textile coatings that selectively degrade toxic
chemicals, including chemical warfare agents and environmental toxins such as NOx and SOx,
while still affording the wearer a high degree of water vapor permeability and thus greater
comfort, are attractive for protection against low to moderate exposures to toxic chemicals. A
recently developed PDADMAC/TiO2 LbL coating provides a reactive barrier with more than
99% efficiency against a saturated environment of the sulfur mustard simulant compound
chloroethyl ethyl sulfide (CEES) in the presence ultraviolet radiation.

184

1.5.4. Antibacterial Coatings
Klevens et al. estimated that approximately 64% of hospital-acquired infections
worldwide stem from attachment of viable bacteria to medical devices and implants,

185

and

these infections have an associated annual mortality of 100 000 persons in the US alone.

35

186

Among various methods for combatting the formation of bacterial films, PEMs provide
numerous opportunities for designing surfaces that either inhibit bacteria attachment and growth
or kill bacteria.

6, 143, 145, 187

Recently, Boulmedais et al.

144

constructed anti-adhesive multilayer films via the LbL

method with pegylated polypeptides. Films with one PLL/PGA-graft-poly(ethyl glycol) (PEG)
and three PLL/PGA-g-PEG bilayers reduced the adhesion of E. coli by 72% and 92%,
respectively, compared with bare substrates. Fu et al. alternately deposited CHI and heparin
onto aminolyzed PET films to construct anti-adhesive and antibacterial multilayer films.

146

The

multilayer films not only reduced the bacterial adhesion significantly, but also killed a part of the
bacteria that adhered to the surface.
More recently Malcher et al. designed a new antibacterial coating made of PLL/HA
multilayer films and liposome aggregates loaded with AgNO3.

142

+

Release of Ag ions should

kill bacteria through disruption of crucial metabolic proteins and enzymes. After 120 min of
contact with such AgNO3-containing coatings the number of E.Coli bacterial in a bacterial
suspension decreased by a factor of 10,000 relative to the initial bacterial population.
1.5.5. Environmental Applications
Industrial effluents constitute a major source of contaminated water with a wide range of
toxic pollutants, and membrane technologies such as reverse osmosis, ultrafiltration, and
nanofiltration may prove effective in treating such waters.

11, 188, 189

Polyelectrolyte

membranes could potentially be tailored for removal of specific contaminants at low operating
pressures. Catalytic polyelectrolyte films may also be effective in wastewater treatment. Huang
36

et al. assembled PAA/PAH multilayers on glass beads or carbon grids and immersed these
materials in FeSO4 solution for 30 min to allow ferrous cations to complex with available free
carboxylate groups on PAA. The substrates were then immersed in 10 mM NaBH4 solution to
0

reduce ferrous ions to Fe nanoparticles. The resulting zero valent iron nanoparticles (ZVI NPs)
can degrade trichloroethylene (TCE), a compound representative of a major class of chlorinated
organic contaminants.

190

Mass-normalized rate constants for pseudo-first-order degradation of
-1

-1

TCE are as high as 0.016 Lhr mg

for (PAA/PAH)15 assembled at pH 2.5. Xiao et al.

immobilized ZVI NPs on PAA/PDADMAC multilayers deposited on cellulose acetate
nanofibers. The decoloration efficiency of the ZVI NP-modified polymer nanofibers for acid
fuchsin, a model contaminant, was 86.8% in 40 min.

191

Priya et al. recently studied the use of

multilayer ultrathin composite films containing nanosized titanium dioxide particles and
polyelectrolytes (PAH or PSS) in photodegradation of rhodamine B under ultraviolet (UV)
irradiation.
film.

They achieved 100% degradation efficiency in 4 h with a PAH/(PSS/TiO 2)20

151

Compounds that disrupt the normal functionality of endocrine (endocrine-disrupting
compunds, EDCs) in humans or other animals are commonly present in municipal
wastewaters.

192

They include pharmaceuticals such as ethynylestradiol (EE2), plasticizers such

as bisphenol A and phthalates, surfactants such as the alkylphenols (APs), and organochlorine
pesticides. Health effects due to EDCs include a range of reproductive problems, changes in
hormone levels, early puberty, brain and behavior problems, impaired immune functions, and

37

various cancers.

193, 194

Pharmaceutically active compounds (PhAC) are also common

constituents of municipal waste water. Although the toxicological effects posed by residual
PhACs are still unknown, the presence of these compounds may induce unwanted effects in the
environment.

195

In aquatic systems, PhACs may affect lower animals with identical or similar

target organs, tissues or cells as humans.
and PhACs by NF membranes

198-200

196, 197

Several groups examined the removal of EDCs

some of which were modified by grafting,

201, 202

but no

literature describes the use of LbL films for PhAC or EDC removal. Hoshi et al. showed that Pt
electrodes coated with PAH/PVS or PAH/PSS films successfully exclude acetaminophen, a
PhAC,

203

So such films could be attractive as membranes for removal of such compounds.

1.6.

Dissertation Outline

This dissertation focuses on the characterization of polyelectrolyte multilayers and their
use in ion separations and water purification. Chapter 2 provides the experimental details for
film formation and characterization as well as the methods for examining separation efficiencies
with these membranes.

Chapter 3 examines why the separation performance of

PSS/PDADMAC membranes varies dramatically with the number of layers deposited on the
underlying porous substrate. ATR-FTIR spectroscopy shows that the ion exchange capacity in
the films changes with number of layers and reaches a maximum value, 0.5 M, after deposition
of 7 bilayers. Moreover, streaming potential measurements demonstrate that the surface charge
of PSS-terminated films changes from negative to positive as the number of layers increase.
These

changes

in

film

properties

dramatically

38

affect

ion

transport

through

(PSS/PDADMAC)nPSS-coated alumina membranes. Nanofiltration properties such as rejections
of sulfate initially increase with the number of adsorbed layers, reach a maximum and then
decrease with more layers.
Chapter 4 demonstrates a new method for enhancing the selectivity of PAH/PSS films for
cation separations. Immersing the films in aqueous FeCl3 generates fixed positive charges in the
+

2+

film to greatly increase Na /Mg

selectivity. FTIR spectroscopy shows the presence of

counterions in the multilayers, which presumably stems from the anion-exchange sites that
+

enhance selectivity in Na /Mg

2+

separations.

Chapter 5 examines the removal of acetaminophen and caffeine from water using NF with
commercial membranes and polyelectrolyte multilayer membranes. This work studies the effect
of the number of PEM bilayers, polyelectrolyte concentration and type of polyelectrolyte on the
rejections of acetaminophen. Commercial NF-90 membranes modified with only one bilayer of
PSS/PAA show 85% rejection of acetaminophen, whereas unmodified NF-90 membranes exhibit
a 65% acetaminophen rejection.

This work also examines the rejections of caffeine using

commercial NF membranes and PEM membranes. (PSS/PAH) 7 films have caffeine rejections of
94% and a solution flux that is 1.5 times higher than that of NF-90.
The final chapter of the thesis presents a summary of the research and its impact as well as
future directions for the work.

39

References

40

1.7.

References

1.

Fery, A.; Schöler, B.; Cassagneau, T.; Caruso, F., Langmuir 2001, 17(13), 3779-3783.

2.

Stanton, B. W.; Harris, J. J.; Miller, M. D.; Bruening, M. L., Langmuir 2003, 19(17),
7038-7042.

3.

Krasemann, L.; Toutianoush, A.; Tieke, B., J. Membr. Sci. 2001, 181(2), 221-228.

4.

Ye, S.; Wang, C.; Liu, X.; Tong, Z., J. Biomater. Sci. Polym. Ed. 2005, 16(7), 909-923.

5.

Wang, B.; Anzai, J.-i., Langmuir 2007, 23(13), 7378-7384.

6.

Chua, P.-H.; Neoh, K.-G.; Kang, E.-T.; Wang, W., Biomaterials 2008, 29(10), 14121421.

7.

Lu, H.; Hu, N., J. Phys. Chem. B 2006, 110(47), 23710-23718.

8.

Picart, C.; Lavalle, P.; Hubert, P.; Cuisinier, F. J. G.; Decher, G.; Schaaf, P.; Voegel, J.
C., Langmuir 2001, 17(23), 7414-7424.

9.

Kekkonen, J.; Lattu, H.; Stenius, P., J. Pulp Pap. Sci. 2002, 28(1), 6-12.

10.

Caruso, F.; Lichtenfeld, H.; Donath, E.; Mohwald, H., Macromolecules 1999, 32(7),
2317-2328.

11.

Jin, W.; Toutianoush, A.; Tieke, B., Langmuir 2003, 19(7), 2550-2553.

12.

Sukhorukov, G. B.; Möhwald, H.; Decher, G.; Lvov, Y. M., Thin Solid Films 1996, 284285, 220-223.

13.

Toutianoush, A.; Krasemann, L.; Tieke, B., Colloids Surf. Physicochem. Eng. Aspects
2002, 198-200, 881-889.

14.

Su, X.; Kim, B.-S.; Kim, S. R.; Hammond, P. T.; Irvine, D. J., ACS Nano 2009, 3(11),
3719-3729.

15.

Qu, X.; Lu, G.; Tsuchida, E.; Komatsu, T., Chemistry - A European Journal 2008,
14(33), 10303-10308.

16.

Liu, A.; Abbineni, G.; Mao, C., Adv. Mater. 2009, 21(9), 1001-1005.

17.

Kurt, P.; Banerjee, D.; Cohen, R. E.; Rubner, M. F., J. Mater. Chem. 2009, 19(47), 89208927.

41

18.

Jiao, Y.-H.; Li, Y.; Wang, S.; Zhang, K.; Jia, Y.-G.; Fu, Y., Langmuir 2010, 26(11),
8270-8273.

19.

Dotzauer, D. M.; Abusaloua, A.; Miachon, S.; Dalmon, J.-A.; Bruening, M. L., Appl.
Catal., B 2009, 91(1-2), 180-188.

20.

Bizeto, M. A.; Shiguihara, A. L.; Constantino, V. R. L., J. Mater. Chem. 2009, 19(17),
2512-2525.

21.

Izumi, C. M. S.; Constantino, V. R. L.; Temperini, M. L. A., J. Nanosci. Nanotechnol.
2008, 8(4), 1782-1789.

22.

Kim, B.-S.; Park, S. W.; Hammond, P. T., ACS Nano 2008, 2(2), 386-392.

23.

Wang, L.; Fu, Y.; Wang, Z.; Fan, Y.; Zhang, X., Langmuir 1999, 15(4), 1360-1363.

24.

Yang, S.; Zhang, Y.; Wang, L.; Hong, S.; Xu, J.; Chen, Y.; Li, C., Langmuir 2005, 22(1),
338-343.

25.

Zeng, G.; Gao, J.; Chen, S.; Chen, H.; Wang, Z.; Zhang, X., Langmuir 2007, 23(23),
11631-11636.

26.

Zhao, Y.; Bertrand, J.; Tong, X.; Zhao, Y., Langmuir 2009, 25(22), 13151-13157.

27.

Kohli, P.; Blanchard, G. J., Langmuir 2000, 16(22), 8518-8524.

28.

Major, J. S.; Blanchard, G. J., Chem. Mater. 2002, 14(6), 2574-2581.

29.

Seo, J.; Schattling, P.; Lang, T.; Jochum, F.; Nilles, K.; Theato, P.; Char, K., Langmuir
2009, 26(3), 1830-1836.

30.

El Haitami, A. E.; Thomann, J.-S.; Jierry, L.; Parat, A.; Voegel, J.-C.; Schaaf, P.; Senger,
B.; Boulmedais, F.; Frisch, B., Langmuir 2010, 26(14), 12351–12357.

31.

Quinn, J. F.; Johnston, A. P. R.; Such, G. K.; Zelikin, A. N.; Caruso, F., Chem. Soc. Rev.
2007, 36(5), 707-718.

32.

Zhang, G.; Ruan, Z.; Ji, S.; Liu, Z., Langmuir 2009, 26(7), 4782-4789.

33.

Dotzauer, D. M.; Dai, J.; Sun, L.; Bruening, M. L., Nano Lett. 2006, 6(10), 2268-2272.

34.

Bhattacharjee, S.; Bruening, M. L., Langmuir 2008, 24(6), 2916-2920.

35.

Kidambi, S.; Bruening, M. L., Chem. Mater. 2004, 17(2), 301-307.

36.

Kidambi, S.; Dai, J.; Li, J.; Bruening, M. L., J. Am. Chem. Soc. 2004, 126(9), 2658-2659.
42

37.

Johnston, A. P. R.; Cortez, C.; Angelatos, A. S.; Caruso, F., Curr. Opin. Colloid Interface
Sci. 2006, 11(4), 203-209.

38.

Nah, Y.-C.; Choi, W. S.; Kim, D.-Y., Sol. Energy Mater. Sol. Cells 2008, 92(12), 15471551.

39.

Tsai, W.-B.; Chen, R. P.-Y.; Wei, K.-L.; Chen, Y.-R.; Liao, T.-Y.; Liu, H.-L.; Lai, J.-Y.,
Acta Biomater. 2009, 5(9), 3467-3477.

40.

Stepp, J.; Schlenoff, J. B., J. Electrochem. Soc. 1997, 144(6), L155-L158.

41.

Benkirane-Jessel, N.; Lavalle, P.; Ball, V.; Ogier, J.; Senger, B.; Picart, C.; Schaaf, P.;
Voegel, J.-C.; Decher, G., Polyelectrolyte Multilayer Films–A General Approach to
(Bio)functional Coatings. Wiley-VCH: Weinheim: Germany, 2007; Vol. 2, p 1249-1305.

42.

Ott, P.; Gensel, J.; Roesler, S.; Trenkenschuh, K.; Andreeva, D.; Laschewsky, A.; Fery,
A., Chem. Mater. 2010, 22(11), 3323-3331.

43.

Veletanlic, E.; Goh, M. C., J. Phys. Chem. C 2009, 113(42), 18020-18026.

44.

Chen, H.; Dong, S., Talanta 2007, 71(4), 1752-1756.

45.

Stair, J. L.; Harris, J. J.; Bruening, M. L., Chem. Mater. 2001, 13(8), 2641-2648.

46.

Farhat, T. R.; Hammond, P. T., Chem. Mater. 2005, 18(1), 41-49.

47.

Dai, J.; Balachandra, A. M.; Lee, J. I.; Bruening, M. L., Macromolecules 2002, 35(8),
3164-3170.

48.

Ballarin, B.; Cassani, M. C.; Tonelli, D.; Boanini, E.; Albonetti, S.; Blosi, M.; Gazzano,
M., J. Phys. Chem. C 2010, 114(21), 9693-9701.

49.

Liu, X.; He, J., J. Phys. Chem. C 2008, 113(1), 148-152.

50.

Ball, V.; Bernsmann, F.; Werner, S.; Voegel, J.-C.; Piedra-Garza, L. F.; Kortz, U., Eur. J.
Inorg. Chem. 2009, 2009(34), 5115-5124.

51.

Nimittrakoolchai, O.-u.; Supothina, S., Macromolecular Symposia 2008, 264(1), 73-79.

52.

Schlenoff, J. B.; Dubas, S. T., Macromolecules 2001, 34(3), 592-598.

53.

Schlenoff, J. B.; Ly, H.; Li, M., J. Am. Chem. Soc. 1998, 120(30), 7626-7634.

54.

Joanny, J. F., Eur. Phys. J. B 1999, 9(1), 117-122.

55.

Ruths, J.; Essler, F.; Decher, G.; Riegler, H., Langmuir 2000, 16(23), 8871-8878.
43

56.

Ramsden, J. J.; Lvov, Y. M.; Decher, G., Thin Solid Films 1995, 254(1-2), 246-251.

57.

Caruso, F.; Niikura, K.; Furlong, D. N.; Okahata, Y., Langmuir 1997, 13(13), 3422-3426.

58.

Picart, C.; Ladam, G.; Senger, B.; Voegel, J.-C.; Schaaf, P.; Cuisinier, F. J. G.; Gergely1,
C., J. Chem. Phys. 2001, 115(2), 1086.

59.

Lavalle, P.; Picart, C.; Mutterer, J.; Gergely, C.; Reiss, H.; Voegel, J.-C.; Senger, B.;
Schaaf, P., J. Phys. Chem. B 2004, 108(2), 635-648.

60.

Hübsch, E.; Ball, V.; Senger, B.; Decher, G.; Voegel, J.-C.; Schaaf, P., Langmuir 2004,
20(5), 1980-1985.

61.

Lavalle, P.; Gergely, C.; Cuisinier, F. J. G.; Decher, G.; Schaaf, P.; Voegel, J. C.; Picart,
C., Macromolecules 2002, 35(11), 4458-4465.

62.

Hoda, N.; Larson, R. G., J. Phys. Chem. B 2009, 113(13), 4232-4241.

63.

Dubas, S. T.; Schlenoff, J. B., Macromolecules 1999, 32(24), 8153-8160.

64.

Guzman, E.; Ritacco, H.; Rubio, J. E. F.; Rubio, R. G.; Ortega, F., Soft Matter 2009,
5(10), 2130-2142.

65.

Lundström-Hämälä, L.; Johansson, E.; Wågberg, L., Starch - Stärke 2010, 62(2), 102114.

66.

McAloney, R. A.; Sinyor, M.; Dudnik, V.; Goh, M. C., Langmuir 2001, 17(21), 66556663.

67.

Monika, S., J. Phys.: Condens. Matter 2003, 15(49), R1781.

68.

Blomberg, E.; Poptoshev, E.; Caruso, F., Langmuir 2006, 22(9), 4153-4157.

69.

Boddohi, S.; Killingsworth, C. E.; Kipper, M. J., Biomacromolecules 2008, 9(7), 20212028.

70.

Lefaux, C. J.; Zimberlin, J. A.; Dobrynin, A. V.; Mather, P. T., J. Polym. Sci., Part B:
Polym. Phys. 2004, 42(19), 3654-3666.

71.

Podsiadlo, P.; Michel, M.; Lee, J.; Verploegen, E.; Wong Shi Kam, N.; Ball, V.; Lee, J.;
Qi, Y.; Hart, A. J.; Hammond, P. T.; Kotov, N. A., Nano Lett. 2008, 8(6), 1762-1770.

72.

Schoeler, B.; Kumaraswamy, G.; Caruso, F., Macromolecules 2002, 35(3), 889-897.

73.

Phuvanartnuruks, V.; McCarthy, T. J., Macromolecules 1998, 31(6), 1906-1914.

44

74.

Farhat, T. R.; Schlenoff, J. B., Langmuir 2001, 17(4), 1184-1192.

75.

Kolarik, L.; Furlong, D. N.; Joy, H.; Struijk, C.; Rowe, R., Langmuir 1999, 15(23), 82658275.

76.

Kujawa, P.; Moraille, P.; Sanchez, J.; Badia, A.; Winnik, F. M., J. Am. Chem. Soc. 2005,
127(25), 9224-9234.

77.

Porcel, C.; Lavalle, P.; Decher, G.; Senger, B.; Voegel, J. C.; Schaaf, P., Langmuir 2007,
23(4), 1898-1904.

78.

Radeva, T.; Milkova, V.; Petkanchin, I., J. Colloid Interface Sci. 2004, 279(2), 351-356.

79.

Sui, Z.; Salloum, D.; Schlenoff, J. B., Langmuir 2003, 19(6), 2491-2495.

80.

Wong, J. E. D ez-Pascual, A. M.; Richtering, W., Macromolecules 2008, 42(4), 12291238.

81.

Richert, L.; Lavalle, P.; Payan, E.; Shu, X. Z.; Prestwich, G. D.; Stoltz, J.-F.; Schaaf, P.;
Voegel, J.-C.; Picart, C., Langmuir 2004, 20(2), 448-458.

82.

Sun, B.; Jewell, C. M.; Fredin, N. J.; Lynn, D. M., Langmuir 2007, 23(16), 8452-8459.

83.

Porcel, C. H.; Izquierdo, A.; Ball, V.; Decher, G.; Voegel, J. C.; Schaaf, P., Langmuir
2005, 21(2), 800-802.

84.

Aulin, C.; Varga, I.; Claesson, P. M.; Wagberg, L.; Lindstrom, T., Langmuir 2008, 24(6),
2509-2518.

85.

Lulevich, V. V.; Vinogradova, O. I., Langmuir 2004, 20(7), 2874-2878.

86.

Yoo, D.; Shiratori, S. S.; Rubner, M. F., Macromolecules 1998, 31(13), 4309-4318.

87.

Schoeler, B.; Poptoshev, E.; Caruso, F., Macromolecules 2003, 36(14), 5258-5264.

88.

Moussallem, M. D.; Olenych, S. G.; Scott, S. L.; Keller, T. C. S.; Schlenoff, J. B.,
Biomacromolecules 2009, 10(11), 3062-3068.

89.

Shiratori, S. S.; Rubner, M. F., Macromolecules 2000, 33(11), 4213-4219.

90.

Adamczyk, Z. Zembala, M. Kolasińska, M. Warszyński, P., Colloids Surf., A 2007,
302(1-3), 455-460.

91.

Elzbieciak, M.; Kolasinska, M.; Warszynski, P., Colloids Surf. Physicochem. Eng.
Aspects 2008, 321(1-3), 258-261.

45

92.

Angelatos, A. S.; Wang, Y.; Caruso, F., Langmuir 2008, 24(8), 4224-4230.

93.

Ariga, K.; Lvov, Y.; Kunitake, T., J. Am. Chem. Soc. 1997, 119(9), 2224-2231.

94.

Baba, A.; Kaneko, F.; Advincula, R. C., Colloids Surf. Physicochem. Eng. Aspects 2000,
173(1-3), 39-49.

95.

Bertrand, P.; Jonas, A.; Laschewsky, A.; Legras, R., Macromol. Rapid Commun. 2000,
21(7), 319-348.

96.

Plech, A.; Salditt, T.; Münster, C.; Peisl, J., J. Colloid Interface Sci. 2000, 223(1), 74-82.

97.

Lvov, Y. M.; Rusling, J. F.; Thomsen, D. L.; Papadimitrakopoulos, F.; Kawakami, T.;
Kunitake, T., Chem. Commun. 1998, (11), 1229-1230.

98.

Salomäki, M.; Vinokurov, I. A.; Kankare, J., Langmuir 2005, 21(24), 11232-11240.

99.

Tan, H. L.; McMurdo, M. J.; Pan, G.; Van Patten, P. G., Langmuir 2003, 19(22), 93119314.

100.

Büscher, K.; Graf, K.; Ahrens, H.; Helm, C. A., Langmuir 2002, 18(9), 3585-3591.

101.

Schwarz, B.; Schönhoff, M., Colloids Surf. Physicochem. Eng. Aspects 2002, 198-200,
293-304.

102.

Lvov, Y. M.; Decher, G., Crystallogr. Rep 1994, 39, 628.

103.

Tieke, B.; Toutianoush, A.; Jin, W., Adv. Colloid Interface Sci. 2005, 116(1-3), 121-131.

104.

Miller, M. D.; Bruening, M. L., Chem. Mater. 2005, 17(21), 5375-5381.

105.

Dubas, S. T.; Schlenoff, J. B., Langmuir 2001, 17(25), 7725-7727.

106.

Lee, L.; Cavalieri, F.; Johnston, A. P. R.; Caruso, F., Langmuir 2009, 26(5), 3415-3422.

107.

Dubas, S. T.; Farhat, T. R.; Schlenoff, J. B., J. Am. Chem. Soc. 2001, 123(22), 53685369.

108.

Marcus, Y., In Liquid−Liquid Interfaces. Theory and Methods, CRC Press: Boca Raton,
FL, 1996.

109.

Helfferich, F., In Ion Exchange, McGraw-Hill: New York, 1962.

110.

El Haitami, A. E. Martel, D. Ball, V. Nguyen, H. C. Gonthier, E. Labb , P. Voegel,
J.-C.; Schaaf, P.; Senger, B.; Boulmedais, F., Langmuir 2009, 25(4), 2282-2289.

46

111.

Salomäki, M.; Laiho, T.; Kankare, J., Macromolecules 2004, 37(25), 9585-9590.

112.

Salomäki, M.; Tervasmäki, P.; Areva, S.; Kankare, J., Langmuir 2004, 20(9), 3679-3683.

113.

Mráček, A. Varhan ková, J. Lehocký, M. Gřundělová, L. Pokopcová, A. Velebný, V.,
Mol. Cells 2008, 13, 1025-1034.

114.

Leontidis, E., Curr. Opin. Colloid Interface Sci. 2002, 7(1-2), 81-91.

115.

Ladam, G.; Schaad, P.; Voegel, J. C.; Schaaf, P.; Decher, G.; Cuisinier, F., Langmuir
2000, 16(3), 1249-1255.

116.

Boulmedais, F.; Ball, V.; Schwinte, P.; Frisch, B.; Schaaf, P.; Voegel, J.-C., Langmuir
2002, 19(2), 440-445.

117.

Lösche, M.; Schmitt, J.; Decher, G.; Bouwman, W. G.; Kjaer, K., Macromolecules 1998,
31(25), 8893-8906.

118.

Voigt, U.; Jaeger, W.; Findenegg, G. H.; Klitzing, R. v., J. Phys. Chem. B 2003, 107(22),
5273-5280.

119.

Khopade, A. J.; Caruso, F., Langmuir 2002, 18(20), 7669-7676.

120.

Fleer, G. J.; Cohen Stuart, M. A.; Scheutjens, J. M. H. M.; Cosgrove, T.; Vincent, B.,
Polymers at Interfaces. Chapman & Hall: Cambridge, U.K.: 1993.

121.

Garg, A.; Heflin, J. R.; Gibson, H. W.; Davis, R. M., Langmuir 2008, 24(19), 1088710894.

122.

Cho, J.; Char, K.; Hong, J. D.; Lee, K. B., Adv. Mater. 2001, 13(14), 1076-1078.

123.

Lee, S.-S.; Hong, J.-D.; Kim, C. H.; Kim, K.; Koo, J. P.; Lee, K.-B., Macromolecules
2001, 34(16), 5358-5360.

124.

Lvov, Y.; Yamada, S.; Kunitake, T., Thin Solid Films 1997, 300(1-2), 107-112.

125.

Gopinadhan, M.; Ivanova, O.; Ahrens, H.; Günther, J.-U.; Steitz, R.; Helm, C. A., J.
Phys. Chem. B 2007, 111(29), 8426-8434.

126.

Farhat, T. R.; Schlenoff, J. B., J. Am. Chem. Soc. 2003, 125(15), 4627-4636.

127.

Carrière, D.; Krastev, R.; Schönhoff, M., Langmuir 2004, 20(26), 11465-11472.

128.

Jaber, J. A.; Schlenoff, J. B., Langmuir 2007, 23(2), 896-901.

47

129.

Lebedeva, O. V.; Kim, B.-S.; Vasilev, K.; Vinogradova, O. I., J. Colloid Interface Sci.
2005, 284(2), 455-462.

130.

Ai, H.; Jones, S.; Lvov, Y., Cell Biochem. Biophys. 2003, 39(1), 23-43.

131.

Liu, X.; Gao, C.; Shen, J.; Möhwald, H., Macromol. Biosci. 2005, 5(12), 1209-1219.

132.

Balabushevich, N. G.; Sukhorukov, G. B.; Larionova, N. I., Macromol. Rapid Commun.
2005, 26(14), 1168-1172.

133.

Smuleac, V.; Butterfield, D. A.; Bhattacharyya, D., Langmuir 2006, 22(24), 1011810124.

134.

van Ackern, F.; Krasemann, L.; Tieke, B., Thin Solid Films 1998, 327-329, 762-766.

135.

Sullivan, D. M.; Bruening, M. L., Chem. Mater. 2003, 15(1), 281-287.

136.

Lenk, W.; Meier-Haack, J., Desalination 2002, 148(1-3), 11-16.

137.

Meier-Haack, J.; Lenk, W.; Lehmann, D.; Lunkwitz, K., J. Membr. Sci. 2001, 184(2),
233-243.

138.

Rmaile, H. H.; Schlenoff, J. B., J. Am. Chem. Soc. 2003, 125(22), 6602-6603.

139.

Hong, S. U.; Malaisamy, R.; Bruening, M. L., J. Membr. Sci. 2006, 283(1-2), 366-372.

140.

Hollman, A. M.; Bhattacharyya, D., Langmuir 2004, 20(13), 5418-5424.

141.

Lichter, J. A.; Van Vliet, K. J.; Rubner, M. F., Macromolecules 2009, 42(22), 8573-8586.

142.

Malcher, M. Volodkin, D. Heurtault, B. a. Andr , P. Schaaf, P. M hwald, H. Voegel,
J.-C.; Sokolowski, A.; Ball, V.; Boulmedais, F.; Frisch, B., Langmuir 2008, 24(18),
10209-10215.

143.

Yuan, W.; Ji, J.; Fu, J.; Shen, J., J. Biomed. Mater. Res. B Appl. Biomater. 2008, 85B(2),
556-563.

144.

Boulmedais, F.; Frisch, B.; Etienne, O.; Lavalle, P.; Picart, C.; Ogier, J.; Voegel, J. C.;
Schaaf, P.; Egles, C., Biomaterials 2004, 25(11), 2003-2011.

145.

Dai, J.; Bruening, M. L., Nano Lett. 2002, 2(5), 497-501.

146.

Fu, J.; Ji, J.; Yuan, W.; Shen, J., Biomaterials 2005, 26(33), 6684-6692.

147.

Farhat, T. R.; Hammond, P. T., Adv. Funct. Mater. 2005, 15(6), 945-954.

48

148.

Jiang, S. P.; Liu, Z.; Tian, Z. Q., Adv. Mater. 2006, 18(8), 1068-1072.

149.

Argun, A. A.; Ashcraft, J. N.; Hammond, P. T., Adv. Mater. 2008, 20(8), 1539-1543.

150.

Tokuhisa, H.; Hammond, P. T., Adv. Funct. Mater. 2003, 13(11), 831-839.

151.

Priya, D. N.; Modak, J. M.; Raichur, A. M., ACS Appl. Mater. Interfaces 2009, 1(11),
2684-2693.

152.

Wiemann, L. O. Buthe, A. Klein, M. van den Wittenboer, A. D hne, L. AnsorgeSchumacher, M. B., Langmuir 2008, 25(1), 618-623.

153.

Timmer, J. M. K. Properties of nanofiltration membranes model development and
industrial application. Technical University of Eindhoven, Eindhoven, The Netherlands.,
2001.

154.

Hong, S. U.; Ouyang, L.; Bruening, M. L., J. Membr. Sci. 2009, 327(1-2), 2-5.

155.

Hong, S. U.; Bruening, M. L., J. Membr. Sci. 2006, 280(1-2), 1-5.

156.

Hong, S. U.; Malaisamy, R.; Bruening, M. L., Langmuir 2007, 23(4), 1716-1722.

157.

Miller, M. D.; Bruening, M. L., Langmuir 2004, 20(26), 11545-11551.

158.

Liu, X.; Bruening, M. L., Chem. Mater. 2004, 16(2), 351-357.

159.

Hong, S. U.; Miller, M. D.; Bruening, M. L., Ind. Eng. Chem. Res. 2006, 45(18), 62846288.

160.

Harris, J. J.; Stair, J. L.; Bruening, M. L., Chem. Mater. 2000, 12(7), 1941-1946.

161.

Krasemann, L.; Tieke, B., Langmuir 2000, 16(2), 287-290.

162.

Balachandra, A. M.; Dai, J.; Bruening, M. L., Macromolecules 2002, 35(8), 3171-3178.

163.

Ouyang, L.; Malaisamy, R.; Bruening, M. L., J. Membr. Sci. 2008, 310(1-2), 76-84.

164.

Toutianoush, A.; Tieke, B., Mater. Sci. Eng., C 2002, 22(2), 135-139.

165.

Toutianoush, A.; Jin, W.; Deligöz, H.; Tieke, B., Appl. Surf. Sci. 2005, 246(4), 437-443.

166.

Hoffmann, K.; Tieke, B., J. Membr. Sci. 2009, 341(1-2), 261-267.

167.

Jin, W.; Toutianoush, A.; Tieke, B., Appl. Surf. Sci. 2005, 246(4), 444-450.

49

168.

Stroeve, P.; Vasquez, V.; Coelho, M. A. N.; Rabolt, J. F., Thin Solid Films 1996, 284285, 708-712.

169.

Leväsalmi, J.-M.; McCarthy, T. J., Macromolecules 1997, 30(6), 1752-1757.

170.

Krasemann, L.; Tieke, B., Mater. Sci. Eng., C 1999, 8-9, 513-518.

171.

Zhou, P.; Samuelson, L.; Alva, K. S.; Chen, C.-C.; Blumstein, R. B.; Blumstein, A.,
Macromolecules 1997, 30(6), 1577-1581.

172.

Zhang, P.; Qian, J.; Yang, Y.; An, Q.; Liu, X.; Gui, Z., J. Membr. Sci. 2008, 320(1-2),
73-77.

173.

Zhang, G.; Gao, X.; Ji, S.; Liu, Z., J. Membr. Sci. 2008, 307(2), 151-155.

174.

Müller, M.; Kessler, B.; Houbenov, N.; Bohatá, K.; Pientka, Z.; Brynda, E.,
Biomacromolecules 2006, 7(4), 1285-1294.

175.

Liu, G.; Dotzauer, D. M.; Bruening, M. L., J. Membr. Sci. 2010, 354(1-2), 198-205.

176.

Ladam, G.; Gergely, C.; Senger, B.; Decher, G.; Voegel, J.-C.; Schaaf, P.; Cuisinier, F. J.
G., Biomacromolecules 2000, 1(4), 674-687.

177.

Salloum, D. S.; Schlenoff, J. B., Biomacromolecules 2004, 5(3), 1089-1096.

178.

Decher, G.; Schlenoff, J. B., Multilayer Thin Films: Sequential Assembly of
Nanocomposite Materials. Wiley-VCH: Weinheim: Germany, 2003; p 524.

179.

Liang, Z.; Wang, C.; Sun, Q.; Tong, Z., Prog. Chem. 2004, 16(4), 485-491.

180.

Caruso, F.; Trau, D.; Möhwald, H.; Renneberg, R., Langmuir 2000, 16(4), 1485-1488.

181.

Wang, C.; He, C.; Tong, Z.; Liu, X.; Ren, B.; Zeng, F., Int. J. Pharm. 2006, 308(1-2),
160-167.

182.

Wang, C.; Ye, S.; Sun, Q.; He, C.; Ye, W.; Liu, X.; Tong, Z., J. Exp. Nanosci. 2008, 3(2),
133 - 145.

183.

Mendes, A. M.; Magalhães, F. D.; Costa, C. A. V., Fluid Transport in Nanoporous
Materials. Springer: 2006.

184.

Krogman, K. C.; Zacharia, N. S.; Grillo, D. M.; Hammond, P. T., Chem. Mater. 2008,
20(5), 1924-1930.

185.

Costerton, J. W.; Stewart, P. S.; Greenberg, E. P., Science 1999, 284(5418), 1318-1322.

50

186.

Klevens, R. M.; Edwards, J. R.; Richards, C. L., Jr. ; Horan, T. C. G., R. P.; Pollock, D.
A.; Cardo, D. M., Public Health Rep. 2007, 122(2), 160– 6.

187.

Lichter, J. A.; Thompson, M. T.; Delgadillo, M.; Nishikawa, T.; Rubner, M. F.; Van
Vliet, K. J., Biomacromolecules 2008, 9(6), 1571-1578.

188.

Rachwal, A. J.; Khow, J.; Colbourne, J. S.; O'Donnell, J., Desalination 1994, 97(1-3),
427-436.

189.

Kochan, J.; Wintgens, T.; Wong, J. E.; Melin, T., Desalination and Water Treatment
2009, 9, 175-180.

190.

Huang, Q.; Shi, X.; Pinto, R. A.; Petersen, E. J.; Weber, W. J., Environ. Sci. Technol.
2008, 42(23), 8884-8889.

191.

Xiao, S.; Wu, S.; Shen, M.; Guo, R.; Huang, Q.; Wang, S.; Shi, X., ACS Appl. Mater.
Interfaces 2009, 1(12), 2848-2855.

192.

Benotti, M. J.; Trenholm, R. A.; Vanderford, B. J.; Holady, J. C.; Stanford, B. D.;
Snyder, S. A., Environ. Sci. Technol. 2008, 43(3), 597-603.

193.

http://e.hormone.tulane.edu/learning/human-effects.html (8/5/2010).

194.

J. Michael, Occup. Environ. Med. 2001, 58, 281-288.

195.

Sumpter, J. P., Drug Inf. J. 2007, 41(2), 143-147.

196.

Fent, K.; Weston, A. A.; Caminada, D., Aquat. Toxicol. 2006, 76, 122-159.

197.

Li, Z. H.; Randak, T., Vet. Med. (Praha). 2009, 52(7), 295-314.

198.

Comerton, A. M.; Andrews, R. C.; Bagley, D. M.; Hao, C., J. Membr. Sci. 2008, 313(12), 323-335.

199.

Braeken, L.; Van der Bruggen, B., Desalination 2009, 240(1-3), 127-131.

200.

Yoon, Y.; Westerhoff, P.; Snyder, S. A.; Wert, E. C.; Yoon, J., Desalination 2007, 202(13), 16-23.

201.

Ben-David, A.; Bernstein, R.; Oren, Y.; Belfer, S.; Dosoretz, C.; Freger, V., J. Membr.
Sci. 2010, 357(1-2), 152-159.

202.

Kim, J.-H.; Park, P.-K.; Lee, C.-H.; Kwon, H.-H., J. Membr. Sci. 2008, 321(2), 190-198.

203.

Hoshi, T.; Saiki, H.; Kuwazawa, S.; Tsuchiya, C.; Chen, Q.; Anzai, J.-i., Anal. Chem.
2001, 73(21), 5310-5315.
51

Chapter 2
Methods and Techniques
2.1.

Polyelectrolyte Multilayer Nomenclature and Deposition

This dissertation employs the (A/B) X nomenclature for polyelectrolyte multilayers, where
A is the initially deposited polyelectrolyte that is in direct contact with the substrate, B is the
oppositely charged polyelectrolyte, and X represents the number of bilayers. Integer values of X
indicate films that terminate with B, whereas n+0.5 values of X, where n is an integer, indicate
films that terminate with A. In some cases, we also use the terminology (A/B)XA to denote a
film containing X bilayers and an additional, terminating layer of A.
Deposition of PEM films on Si and Ge crystals or PES membranes begins by exposing
the top of the substrate to a solution containing 0.02 M polycation and then rinsing for 1 min
with deionized water from a squirt bottle (concentrations of polyelectrolytes are always given
with respect to the repeating unit). Subsequent adsorption of polyanions occurs similarly, and
repetition of the entire process yields the multilayer films. For Au-coated wafers, a monolayer of
3-mercaptopropionic acid (MPA) on the Au wafer serves as an adhesion layer between Au and a
polycation, which is the first deposited polyelectrolyte.

Adsorption of PEMs on alumina

membranes takes place similarly to deposition on Si, except the starting layer is PSS. The pH
and supporting salt concentrations in deposition solutions are mentioned in each chapter for
specific experiments.

52

2.2.

Determination of Film Thickness by Ellipsometry and Scanning
Electron Microscopy

2.2.1. Ellipsometry

1

Ellipsometry is a rapid, convenient method for determining the thicknesses of smooth
films whose thicknesses range from nanometers to a few microns. This technique measures a
change in the polarization of light as it reflects or transmits from an interface. The polarization
change is typically described by the ratio, ρ, of reflection coefficients for the transverse magnetic
and transverse electric modes. This ratio can be expressed as a function of both the amplitude
ratio, Ψ, and the phase difference, Δ, of the two reflection coefficients as shown in equation 2.1.

  tan   ei

------------ (2.1)

The measured values of Ψ and Δ depend on both the optical properties and thickness of a
film, so in principle ellipsometry allows determination of both film thickness and optical
constants. For thin films, the Cauchy model, equation 2.2, effectively describes the variation of
refractive index with wavelength, λ,

n    A 

B

2



C

------------ (2.2)

4

where the three terms A, B, and C are adjusted to fit the refractive index of the material.
Determination of optical constants is difficult for ultrathin films, however, because in this realm
the values of Ψ and Δ vary only weakly with refractive index. Because ellipsometry assumes a
smooth film and substrate, the validity of this technique diminishes for rough films.
Ellipsometry studies were performed with a J.A. Woollam model M-44 rotating analyzer
ellipsometer to determine the thicknesses of PEM films on Si wafers. The thicknesses of dry

53

films were determined using literature values of refractive indices for Si and SiO 2 and assuming
a PEM refractive index of 1.5. After deposition, films were dried with a stream of nitrogen, and
the dry thickness was determined at several points on each wafer.

2.2.2. Scanning Electron Microscopy (SEM)
For films with thicknesses >50 nm, cross-sectional SEM images provide a measure of
film thickness and roughness. Film-coated Si wafers were fractured immediately after exposure
to liquid nitrogen and subsequently sputter-coated (Pelco model SC-7 auto sputter coater) with
gold on the cross section of the wafer to minimize charging during SEM. PES membranes
coated with polyelectrolyte multilayers were cut with scissors, and the surface of the membrane
was sputter-coated with 5 nm of gold prior to imaging. SEM images were obtained with a
Hitachi S-4700 II field-emission scanning electron microscope.

2.3.

Molecular Spectroscopy

2.3.1. Reflectance Fourier Transform Infrared (FTIR) Spectroscopy
External reflection FTIR spectra were obtained with a Nicolet Magna 560 FTIR
spectrometer using a PIKE grazing angle (80 °) attachment. This instrument contains a MercuryCadmium-Telluride (MCT) detector and a KBr beam splitter and has a spectral range of 4000
-1

cm

-1

- 700 cm . The MCT detector is vital to achieve the sensitivity needed for reflectance
-1

spectroscopy. All spectra were recorded using 128 scans at 4 cm resolution, and a spectrum of
a gold wafer in air served as a background.

54

2.3.2. Attenuated Total Reflectance Fourier Transform Infrared (ATR-FTIR)
Spectroscopy
ATR-FTIR spectroscopy relies on absorption of the evanescent wave at the surface of the
ATR crystal. Figure 2.1 show an in situ ATR-FTIR setup where the crystal is covered with
solution. Total reflection, which results in the formation of an evanescent wave, occurs when
light propagates from an optically denser medium to a rarer medium and when the angle of
2

incidence is greater than critical angle, θc. Equation 2.3 defines the critical angle

 n2 

 n1 

------------ (2.3)

c  sin 1 

where n1 is the refractive index of the dense medium (ATR crystal) and n2 is the refractive index
of the rare medium (polymer film). Equation 2.4 gives the magnitude of the electric field, E, of
the evanescent wave as a function of the distance, z, from the surface where

E  E0 e

------------ (2.4)

z d p

E0 is the incident electric field amplitude, and d p is the penetration depth or the distance at which
the amplitude of the evanescent wave drops to 1/e times its amplitude at the crystal/film or
crystal/solution interface.

3, 4

For a particular wavelength, λ, equation 2.5 describes the

penetration depth,

dp 





2 n1 sin   n2
2

2

------------ (2.5)



2 1/2

where θ is the angle of incidence (45º in our experiments), n1 is the refractive index of the Ge
crystal (4.0) and n2 is the refractive index of the film.
55

5

Significant absorption in an ATR-FTIR

experiment occurs only within 2-3 penetration depths of the surface because at greater distances,
the electric field is <10% of its initial amplitude. Moreover, the light intensity and absorption are
proportional to the square of the electric field.

Solution

Polyelectrolyte Film

Solution

Ge Crystal

Peristaltic
Pump

Waste

Solution

Figure 2.1.

ATR Flow
Cell

ATR-FTIR flow cell for the measurement of diffusion of ions in the
polymers. (The sample compartment was closed during measurements.) The
inset shows a schematic diagram of the ATR crystal in the flow cell.

A Nicolet Magna 550 FTIR spectrophotometer was used to monitor the diffusion of ions
into polyelectrolyte multilayers. A flow cell with a volume of 550 µL, and a 70 mm x 10 mm x
6 mm 45º germanium crystal from Specac were used in these studies. All spectra were recorded
-1

using 32 scans at 4 cm

resolution, and a spectrum of a bare germanium crystal in water was

typically used as a background. The germanium crystal was cleaned by rinsing sequentially with
56

1 M CaCl2 and water and then wiping the crystal with a tissue soaked in methanol or
isopropanol.

6, 7

Cleaning was verified by the absence of sulfonate absorbances in ATR-FTIR

spectra taken with air as a background.

2.3.3. UV/Vis Spectroscopy

8

A Perkin Elmer Lambda 40 spectrophotometer was used to record UV/Vis spectra for
analysis of acetaminophen and caffeine. According to Beer’s law which is given by equation
2.6,

A   bC

------------ (2.6)

where A is the absorbance,  is the molar absorptivity coefficient or molar extinction coefficient
-1

-1

in Lmol cm , b is path length in cm and C is the concentration in molL

-1

or M. When the

extinction coefficient and the path length are constant, absorbance is directly proportional to the
concentration.

2.4.

Ion Chromatography

Concentrations of anions were determined by ion chromatography with conductivity
detection using a Dionex 600 Ion Chromatograph with an Ionpac AS16 column. Samples are
first passed through a shorter and less expensive guard column to protect the separator column
from potentially damaging sample material.

The eluent is 20 mM NaOH, the anions are

separated based on their tendency to bind to ion-exchange sites, and the concentrations are then
determined using conductivity measurements. For anion detection, an ASRS-Ultra anion selfregenerating suppressor (4 mm) is used to reduce the background conductivity of the eluent
57

because the concentration of anions is very low. Chromatograms are collected using a 1.0
mL/min flow rate, and the sample loop has a volume of 25 µL. Eluents and all sample solutions
are prepared with deionized water of resistivity 18.2 MΩ cm.

2.5.

Atomic Absorption Spectroscopy and Inductively Coupled Plasma
Optical Emission Spectroscopy

Atomic Absorption Spectroscopy allows the determination of trace concentrations of
elements, usually metals, through monitoring of light absorption due to specific electronic
transitions of atoms. The light source is a hallow cathode lamp that is element specific, and
atoms in the sample are vaporized in a high temperature source such as a flame or graphite
furnace. This work employed a Varian Spectra Atomic Absorption-200 Spectrometer to analyze
metals like Na and Mg.
A Varian 710-ES ICP Optical Emission Spectrometer was used for analysis of the iron in
polyelectrolyte multilayer films. In this case, plasma is used to excite the atoms and ionize a
sample, and the emission of light from specific elements is used along with a calibration curves
to determine their concentration.

2.6.

Streaming Potential Measurements

In a system with flow across a surface, convective transport of the excess charge in the
diffuse electrical double layer creates a streaming potential such that no net current flows
through the system. Based on the assumption that the currents due to convection of the diffuse
double layer and to migration stemming from the streaming current must be equal in magnitude
and opposite in direction, equation 2.7 provides an expression for the streaming potential, U.
58

9

 
U  o r P

------------ (2.7)

 Kb

In this equation, ΔP is the pressure drop across the capillary of length L, µ is the fluid viscosity,
Kb is liquid conductivity, εo is the permittivity of vacuum, εr is the dielectric constant of the
solvent, and ζ is the zeta potential, which is the potential at the interface of the Stern layer, where
ions are firmly bound to the surface, and the diffuse layer where ions are mobile (See Figure
2.2).
Streaming potential measurements were performed with a Brookhaven Instruments
Electro Kinetic Analyzer using a clamping cell (Anton Paar, Graz, Austria) as shown in figure
2.3. In this cell, the test sample is placed flush against a 10 mm by 20 mm piece of grooved
poly(methyl methacrylate) (PMMA) spacer. The clamping cell includes Ag/AgCl electrodes and
inlet/outlet tubing which introduce the test solutions (1 mM KCl and 1 mM MgCl2 in this work)
into the rectangular channels of the PMMA spacer.

9, 10

PES membranes with dimensions of 5

cm x 5 cm (cut from a flat sheet of 100 kDa membrane from Pall Corporation, Port Washington,
NY), served as substrates for streaming potential measurements.

These membranes were

initially soaked in 0.1 M NaOH for 1 hour and rinsed with deionized water prior to
polyelectrolyte adsorption.

59

+
+
+
+
+
+
+
+

-

-

+

-

+

-

-

-

-

+

-

+

+

Bulk Solution

Diffuse Layer

ψ
ψ0

+

-

+

Plane of Shear
Stern Plane

ψd
ζ
1/κ
Figure 2.2.

x

Schematic representation of electrical double layer and ζ potential.

11

The ζ potentials were measured with both a sheet of PMMA and the membranes. The
measurement with the PMMA provides the value of the spacer zeta potential. Even with the
membrane in place, the PMMA spacer still contributes to the streaming potential, so the ζ
potential of the sample is calculated from the ζ potential that is measured during the experiment,
ζaverage, using equation 2.8.

9

60

 sample  2 average   spacer

Figure 2.3.

------------ (2.8)

Electro Kinetic Analyzer flow cell setup for zeta potential measurements.

2.7.

Nanofiltration Experiments

Nanofiltration experiments were performed with a home built cross-flow apparatus
(Figure 2.4) that was pressurized with Ar at 4.8 bar, as described previously.

12

A centrifugal

pump circulated the analyte solution through the apparatus and across the membrane, which had
2

an exposed area of 1.5 cm . The exposed area was measured by nanofiltration with 0.1% Congo
red dye.

12

The flow rate across the membrane was set to 18 mL/min and controlled by a flow

meter located between the pump and membrane cells. This flow rate was selected to minimize
any concentration polarization in the nanofiltration cell. After 18 h of filtration, four permeate
samples were collected for time periods ranging from 5 to 30 min each, depending on the flux
61

through the membrane, and the feed was analyzed at the end of the experiment to calculate
rejections. The volume of the feed was 6- to 8-fold greater than the total volume of permeate, so
the feed concentration was relatively constant. The flux measurements reported are the average
of several steady-state solution flux measurements after the initial 18 h of filtration.

Flow meter

Prefilter

Water
bath

Ar
Pump

Membrane
cells

Feed
tank
Pressure
gauge
Figure 2.4.

Schematic drawing of the experimental NF apparatus.

62

12

References

63

2.8. References
1.

Fujiwara, H., Spectroscopic Ellipsometry: Principles and Applications. Wiley: 2007.

2.

Urban, M. W., Attenuated Total Reflectance Spectroscopy of PolymersTheory and
Practice. American Chemical Society: Washington, DC, 1996.

3.

Fieldson, G. T.; Barbari, T. A., AlChE J. 1995, 41(4), 795-804.

4.

Fieldson, G. T.; Barbari, T. A., Polym. Mater. Sci. Eng. 1994, 71, 150-151.

5.

Harrick, N. J., Internal Reflection Spectroscopy. John Wiley & Sons, Inc.: New York,
1967.

6.

Salloum, D. S.; Schlenoff, J. B., Biomacromolecules 2004, 5(3), 1089-1096.

7.

http://las.perkinelmer.com/content/TechnicalInfo/TCH_FTIRATR.pdf (8/4/2008).

8.

Skoog, D. A.; Holler, F. J.; Nieman, T. A., Principles of Instrumental Analysis. Saunders
College Publishing: Philadelphia, 1998.

9.

Walker, S. L.; Bhattacharjee, S.; Hoek, E. M. V.; Elimelech, M., Langmuir 2002, 18(6),
2193-2198.

10.

Ouyang, L.; Malaisamy, R.; Bruening, M. L., J. Membr. Sci. 2008, 310(1-2), 76-84.

11.

Bard, A. J.; Faulkner, L. R., Electrochemical Methods. 2nd ed.; Wiley: New York, 2001.

12.

Stanton, B. W.; Harris, J. J.; Miller, M. D.; Bruening, M. L., Langmuir 2003, 19(17),
7038-7042.

64

Chapter 3
Variation of Ion-exchange Capacity, Zeta Potential, and Iontransport Selectivities with the Number of Layers in a Multilayer
Polyelectrolyte Film
This chapter is published in Langmuir 2009, 25, (13), 7478-7485.

3.1. Introduction
Alternating adsorption of polycations and polyanions is a simple, versatile technique for
forming functional thin films.

1-5

The thicknesses of these films can be controlled by varying the

number of deposited layers, and a wide range of materials including synthetic polymers, proteins,
inorganic sheets, and nanoparticles can be incorporated into such coatings to tailor them for
specific applications.

6-13

In some potential uses of multilayer polyelectrolyte films, e.g.
14-22

membrane-based separations

and encapsulation of drugs

23-25

and enzymes,

26, 27

control of

transport through the film, which can be achieved by changing deposition conditions and
polyelectrolytes, is vital. Adsorption of different polyelectrolytes yields films with a wide range
of swelling and transport properties, and even the mechanism of film growth varies among
polyelectrolyte pairs.

28-36

Film permeability and the amount of polyelectrolyte deposited per adsorption step can
also vary dramatically with the number of layers in a multilayer polyelectrolyte film.
-

20, 37-40

-

Hong et al. recently found that the Cl /F nanofiltration selectivities of PSS/PDADMAC films on
porous alumina membranes reach a maximum value of 3.4 for (PSS/PDADMAC) 4PSS films and
65

then decline to essentially unity with (PSS/PDADMAC) 6PSS films.

35, 41

The research

described here focuses on better understanding the changes that occur in PSS/PDADMAC films
as the number of adsorbed layers increases.
Schlenoff et al. proposed a model for ion transport through polyelectrolyte multilayers in
which the ions hop between ion-exchange sites.

42

Thus, one of the film properties that this

manuscript investigates is the ion-exchange capacity of PSS/PDADMAC films as a function of
the number of deposited layers.

Although several techniques were employed to determine

whether counterions are present in multilayer polyelectrolyte films,

10, 43-47

attenuated total

reflectance Fourier-transform infrared (ATR-FTIR) spectroscopy is particularly attractive for
examining the absorption of counterions that have suitable infrared absorbances.

48, 49

By

focusing on specific IR absorption bands in the ion, ATR-FTIR spectroscopy can sensitively
distinguish between polymers and solutes, and this technique also reveals chemical interactions
between solutes and polymers through shifts in peak absorbances.

50

In this chapter, we use
-

ATR-FTIR spectroscopy to examine the absorption of both a monovalent ion, SCN , and a
2-

divalent ion, Ni(CN)4 , in PSS/PDADMAC films as a function of the number of bilayers.
These studies show that the number of anion-exchange sites in PSS/PDADMAC films increases
dramatically after deposition of 6 bilayers in the film.
We also investigate the zeta potentials of PSS/PDADMAC films as a function of the
number of deposited layers. Zeta potentials are indicative of the net film charge, which can
increase ion-transport selectivities by heavily excluding the partitioning of multivalent ions into
39

the film (Donnan exclusion ). A number of studies show that the sign of the zeta potentials of
66

polyelectrolyte films oscillates between positive and negative depending on whether the film
terminates with a polycation or a polyanion.

33, 51-53

In contrast, in the case of PSS/PDADMAC,

the situation is more complex and the trend in zeta potential changes as more layers are
deposited.
Finally, this chapter examines the separation of monovalent and divalent ions by
nanofiltration through PSS/PDADMAC films deposited on highly permeable supports. Trends
in monovalent ion/divalent ion transport selectivities as a function of the number of deposited
PSS/PDADMAC layers correlate well with corresponding trends in zeta potentials and ionexchange capacities, and remarkably high selectivities can only be achieved only with the
optimal number of bilayers.

3.2. Experimental Section
Materials.

Poly(sodium 4-styrenesulfonate) (PSS, Mw = 70 000 Da), poly(diallyl

dimethylammonium chloride) (PDADMAC, Mw = 200 000 – 350 000), KSCN, and NaSCN
were purchased from Sigma-Aldrich, and Figure 3.1 shows the structures of the polyelectrolytes.
Reagent grade KCl, NaCl, and Na2SO4 were obtained from Jade Scientific, and K2Ni(CN)4 was
purchased from Spectrum. Porous alumina supports (0.02-μm Whatman Anodisk filters) were
UV/O3 cleaned (Boekel UV-Clean Model 135500) with the feed side up for 15 min before film
deposition, and deionized water (Millipore purification system, 18.2 MΩcm) was used for
membrane rinsing and preparation of the polyelectrolyte solutions.

We utilized alumina

membranes with 20 nm-diameter surface pores to avoid significant deposition of polyelectrolytes
in membrane pores.

29, 54

67

Polyethersulfone (PES) membranes with molecular weight cutoffs of 50 and 100 kDa
were kindly provided by Pall Corporation (Port Washington, NY) and were initially soaked in
0.1 M NaOH for 1 hour, then soaked in DI water for 24 hours and rinsed with deionized water.
Silicon wafers (100 orientation) were obtained from Silicon Quest International (Santa Clara,
CA), and after 15 min of UV/Ozone cleaning, the ellipsometric thickness of the SiO 2 layer on
the surface of the wafer was 2 nm.

Figure 3.1.

Structures of the polyelectrolytes used in this study.

Film Deposition.
previously.

28, 29

Polyelectrolyte films were deposited using a procedure described

Briefly, deposition of PSS occurred by exposing the top of the substrate to 0.02

M PSS in 1.0 M NaCl for 5 min and rinsing for 1 min with deionized water. Deposition of
PDADMAC occurred similarly. The pH of PDADMAC and PSS deposition solutions was
adjusted to 7.4 with 0.1 M NaOH. After deposition of the desired number of polyelectrolyte
layers, membranes were stored in water until use.
Ellipsometry. Ellipsometry studies were performed with a J.A. Woollam model M-44 rotating
analyzer ellipsometer to determine the thicknesses of PDADMAC/PSS films on Si wafers. This
technique is described in more detail in chapter 2.

68

Scanning Electron Microscopy. Cross-sectional images of silicon wafers coated with
polyelectrolyte multilayers were taken using a Hitachi S-4700 II field-emission scanning electron
microscope (FESEM) as described in Chapter 2.
ATR-FTIR Studies. Attenuated total internal reflection FTIR spectroscopy was used to monitor
the diffusion of ions into polyelectrolyte multilayers (see Chapter 2). Multilayer formation was
accomplished by alternately passing the polyelectrolyte solutions and deionized water over the
ATR crystal in the flow cell for the times mentioned above.
Streaming Potential Studies.

Streaming potential measurements were performed with a

Brookhaven Instruments Electro Kinetic Analyzer using a clamping cell (see Chapter 2). A 1
mM KCl solution served as the electrolyte for these measurements.
Transport Studies. Nanofiltration experiments were performed as described in section 2.6. For
-

2-

Cl /SO4

separations, feed solutions contained 1000 ppm SO4

2-

-

and 1000 ppm Cl prepared by

dissolving Na2SO4 and NaCl in deionized water. Anion concentrations were determined using
ion chromatography (Dionex 600 Ion Chromatograph with an Ionpac AS16 column) with
conductivity detection, and all reported transport results are the averages of experiments with
two different membranes. The plus and minus values are the standard deviations of several
measurements on two different membranes.
-

In the case of SCN /Ni(CN)4

2-

-

and Cl /Ni(CN)4

2-

2-

experiments, Ni(CN)4

concentrations were determined by Ni atomic absorption spectroscopy (Varian Spectra Atomic
-

-

Absorption-200 Spectrophotometer). The SCN or Cl concentrations were determined with
conductivity measurements (Orion 115 Conductivity meter), where conductivities were
converted into concentrations using a calibration curve with a small correction for Ni(CN) 4
69

2-

2-

concentration. The correction, which was small because of the low permeability of Ni(CN) 4 ,
was based on a separate conductivity-concentration calibration curve for K2Ni(CN)4 and the Ni
-

2-

analysis. For SCN /SO4

-

experiments, the SCN and SO4

2-

concentrations were determined

using ion chromatography.

3.3. Results and Discussion
3.3.1. Film Formation and Characterization
Figure 3.2 shows the ellipsometric thicknesses of PDADMAC/PSS films on Si wafers as
a function of the number of PDADMAC/PSS bilayers. (The Si wafers contain a 2 nm layer of
SiO2 on their surface, and the films are adsorbed from solutions containing 1 M NaCl.) Similar
to literature data, the thicknesses of the PDADMAC/PSS multilayer films increase nonlinearly
with the number of bilayers,

37

and the thickness per bilayer is as high as 60 nm. Films with

more than 10 bilayers are visually inhomogeneous and rough, so it is difficult to determine their
thicknesses using ellipsometry.
The initial nonlinear and very rapid growth of PDADMAC/PSS films may occur by the
“exponential growth” mechanism suggested previously.

37, 40, 55-57

In this mechanism, one of

the polyelectrolytes, frequently the polycation, absorbs throughout the polyelectrolyte multilayer.
Upon deposition of the next layer, the previously deposited polyelectrolyte diffuses to the surface
and forms a thick layer of the polyanion/polycation complex.

70

350

Thickness(nm)

300
250
200
150
100
50
0
0

1

2

3

4

5

6

7

8

9

10

Number of Bilayers
Figure 3.2.

Ellipsometric thicknesses of (PDADMAC/PSS) n films on silicon wafers.
Polyelectrolytes were adsorbed from solutions containing 1 M NaCl.

Figure 3.3 shows SEM images of the cross sections of 10-bilayer and 30-bilayer
PDADMAC/PSS films on Si. The average thicknesses of these films are 400 nm and 1750 nm,
respectively. The ellipsometric thickness of the 10-bilayer film is 25% less than the SEM
thickness, which could be due to the film roughness or the fracturing of the film for SEM. ATRFTIR spectra such as those in Figure 3.4 provide a means to examine the layer by layer
adsorption of PDADMAC/PSS films on a Ge ATR crystal. Increases in the height of the
-1

sulfonate absorbances at 1008 and 1033 cm indicate the adsorption of PSS.

71

49, 58

Film

Film
1 µm

Figure 3.3.

5 µm

Cross-sectional SEM images of Si wafers coated with 10-bilayer (left) and 30bilayer (right) PDADMAC/PSS films.

The intensities of the sulfonate peaks initially increase rapidly as more layers are deposited and
then level off after deposition of about 12-15 bilayers (Figure 3.5). This plateau in absorbance
suggests that the thickness of the film is at least 2-3 times the penetration depth of the evanescent
-1

IR beam after deposition of 15 layers. The penetration depth at 1033 cm

for a film with a

refractive index of 1.44 is 0.62 μm. (Dry PDADMAC/PSS films have a refractive index around
1.55, and the refractive index of water is about 1.33, so a refractive index of 1.44 represents
approximately 100% swelling in water.) Thus, for films with 100% swelling, a thickness of 2-3
times the penetration depth is equivalent to a swollen film thickness of 1.2-1.9 μm. In agreement
with these IR studies, AFM data taken by Schlenoff et al. show that the wet thickness of
(PDAMAC/PSS)15 films prepared with solutions containing 1M NaCl is about 1.2 μm.

72

31, 49

0.14
1
3
5
7
9
11
13
15

Absorbance

0.12
0.10
0.08
0.06
0.04

0.02
0.00
1060

1040

1020

1000

980

Wavenumbers (cm-1)
Figure 3.4.

ATR-FTIR spectra of (PDADMAC/PSS) nPDADMAC films adsorbed on a
Ge ATR crystal. The numbers in the legend represent n, and the peaks at
-1

1008 cm

-1

and 1033 cm

are due to the sulfonate groups of PSS. The films

were immersed in deionized water while the spectra were taken, and the
background spectrum was that of deionized water on a clean crystal.
The oscillations in Figure 3.5 likely stem from different percent swelling values for PSSterminated and PDADMAC-terminated films.

As discussed in Appendix A the PDADMAC-

terminated films swell by 400% and have a refractive index of ~1.34 whereas films terminated
by PSS swell by 100% and have a refractive index of ~1.44. In more swollen films, the
sulfonate groups are on average farther from the ATR crystal surface, and this give rise to lower
IR peak intensities.

73

1.4
1.2

Peak Area

1.0

0.8
0.6
0.4
0.2
0.0
0

5

10

15

20

25

30

Number of bilayers (n)
-

Figure 3.5.

Peak area of the -SO3 absorbance centered at 1033 cm

-1

in the ATR-FTIR

spectra of (PDADMAC/PSS) n multilayer films deposited on a Ge crystal.
Films were deposited from a solution containing 1 M NaCl. The wide
oscillations between the areas of peaks in PSS- and PDAMAC-terminated
films are likely due to differences in the swelling of these films. The more
swollen PDADMAC-terminated films exhibit lower absorbances.

28

Films

with an integer number of bilayers terminate with PSS, while films with an
additional 0.5 bilayer terminate with PDADMAC.

-

3.3.2. ATR-FTIR Studies of SCN Absorption in (PDADMAC/PSS)30PDADMAC films
In an effort to better understand ion absorption and transport in PDADMAC/PSS
-

coatings, we utilized ATR-FTIR spectroscopy to examine partitioning of SCN and Ni(CN)4

2-

into PDADMAC/PSS films on Ge crystals. We studied these two anions because they have

74

different valences as well as distinct, strong IR stretches that appear in a region of the spectrum
-1

(2000 to 2100 cm ) where there is minimal interference from absorbances due to other
functional groups. In initial absorption studies, we employed (PDADMAC/PSS) 30PDADMAC
films that have a dry thickness (1750 nm) that is >5-times the penetration depth of the IR beam at
-1

2058 cm

(assuming a film refractive index of 1.55). Swollen films should be 2- to 5-fold

thicker than dry films and have a thickness >10 penetration depths. (The penetration depth at
-1

2058 cm

for a film with a refractive index of 1.34 is 0.31 μm. The refractive index is taken as

1.34 as these are PDADMAC terminated films and swell by 400%.) Thus, IR signals will only
be due to species in the film, not species in solution.
Figure 3.6 shows the ATR-FTIR spectrum of aqueous 0.5 M NaSCN on a bare Ge crystal
-

along with the spectrum of SCN absorbed into a (PDADMAC/PSS)30PDADMAC film from 1
-

mM NaSCN. Upon absorption in the film, the SCN peak absorbance that is due to a C=N
59

stretching vibration

-1

shifts from 2065 cm

(bare crystal exposed to NaSCN solution) to 2058

-1

cm , which may be due to a lower permittivity in the film than in water or to interaction of
-

SCN with the film.

60-63

-

Moreover, the area of the peak due to SCN absorbed from a 1 mM

solution is 17% greater than that of 0.5 M NaSCN on the bare crystal, indicating that the
-

concentration of SCN absorbed in the film is much higher than its 1 mM concentration in the
solution from which it is absorbed.

75

0.04
0.5 M NaSCN on bare Ge crystal
SCN adsorbed in (PDADMAC/PSS)30PDADMAC

Absorbance

0.03

0.02

0.01

0.00
2150

2100

2050

2000

1950

Wavenumbers (cm-1)
Figure 3.6.

FTIR spectra of aqueous 0.5 M NaSCN on a bare Ge crystal (pink) and of
-

SCN absorbed in a (PDADMAC/PSS)30PDADMAC film on a Ge crystal
-

(blue). The SCN was absorbed for 6 h from a solution containing 1 mM
NaSCN, and the blue spectrum was taken while the film was still covered
with the NaSCN solution.

76

0.040
720 min
360 min
180 min
60 min
30 min
10 min
5 min
2 min
1 min
5 sec
0 sec

0.035

Absorbance

0.030
0.025
0.020
0.015
0.010
0.005
0.000
2140

2090

2040

1990

Wavenumbers (cm-1)
Figure 3.7.

ATR-FTIR spectra of a (PDADMAC/PSS)30PDADMAC film during
exposure to a 0.01 M NaSCN solution for varying lengths of time. The
-

absorbance in this region is due to the C=N stretch of SCN . The spectra
overlap after 180 min of exposure.

77

-

-1

Figure 3.7 shows the increase in the SCN peak (2058 cm ) with time after exposure of a
(PDADMAC/PSS)30PDADMAC film to 0.01 M NaSCN, and Figure 3.8 shows how the area of
this peak evolves with time after exposure of a (PDADMAC/PSS) 30PDADMAC film to different
-

-

concentrations of SCN . The SCN peak areas in Figure 3.8 are normalized with respect to the
-1

sulfonate peak of PSS at 1033 cm

to minimize any variations in absolute absorbance due to
-

changes in swelling and refractive index upon SCN absorption.

42

As Figure 3.8 shows, the time

required to achieve equilibrium absorption decreases with an increasing concentration of SCN

-

in solution. Higher concentrations of anions in solution should increase the rates of both ionexchange and diffusion into the film to decrease the time required to saturate the anion-exchange
sites. Attempts to model the data in Figure 3.8 with equations for simple diffusion into the film
did not give a good fit to the data.

64

78

0.8
0.7

Peak Area

0.6
0.5
200mM
100mM
50mM
10mM
5mM
1mM
0.5mM
0.2mM
0.1mM

0.4
0.3
0.2
0.1
0.0
0

50

100

150

200

250

300

350

Time (min)
Figure 3.8.

-1

-

Normalized area of the SCN peak (2058 cm ) in ATR-FTIR spectra of a
(PDADMAC/PSS)30PDADMAC film exposed to different concentrations of
NaSCN for various periods of time. The film was deposited on a Ge ATR
crystal, and after measurements of spectra at a given NaSCN concentration,
-

the film was rinsed with 0.5 M NaCl to remove SCN prior to changing the
-

NaSCN concentration. The SCN peak area was normalized to the area of
-

-1

the –SO3 peak at 1033 cm .
At all solution NaSCN concentrations ≤50 mM, the equilibrium peak areas for absorbed
-

-

SCN are similar. To quantify the amount of SCN in the film, we obtained ATR-FTIR spectra
of a series of aqueous NaSCN solutions on a bare Ge crystal and used a calibration method that
takes into account differences in film and solution dielectric constants.

79

(The Appendix A

describes this method in detail.) Figure 3.9 shows that the equilibrium concentration of absorbed
-

SCN in the film is 0.5 M for all solution concentrations of NaSCN ≤50 mM.

1.2

[SCN-] in Film (M)

1.0

0.8

0.6

0.4
0

100

200

300

400

[SCN-] in Solution (mM)
Figure 3.9.

-

Equilibrium SCN concentrations in a (PDADMAC/PSS)30PDADMAC film
exposed to different concentrations of NaSCN. The inset shows an expanded
view at lower solution concentrations.

The data in Figure 3.9 suggest that there is a high concentration of ion-exchange sites in
-

-

the film and that at solution concentrations of SCN ≤50 mM, SCN absorption primarily results
-

-

from displacement of Cl by SCN at ion-exchange sites in the film as shown in reaction 3.1. In
80

+

this reaction, Pol represents anion-exchange sites that are presumably due to PDADMAC
quaternary amines that are not charge-compensated by a neighboring sulfonate of PSS.

Pol+ Cl- + Na +SCNPol Pol- + Na +SCN+

Pol+SCN-  Na + Cl-

------------ (3.1)
------------ (3.2)

Pol SCN-  Pol- Na +
+

-

There are two possible explanations for the increase in the concentration of SCN in the
+

-

film at solution concentrations of NaSCN >50 mM. One is that a significant number of Pol Pol
+

-

pairs dissociate to create anion-exchange, Pol , and cation-exchange, Pol , sites as shown in
reaction 3.2.

42, 65

Another explanation could be simple partitioning of NaSCN into the solvent-

swollen film, which should be more significant at higher NaSCN concentrations in solution.
-

Rinsing of films with deionized water does not remove the SCN that is absorbed from
solutions containing ≤50 mM NaSCN. However, a deionized-water rinse of a film exposed to
-

400 mM NaSCN decreases the amount of absorbed SCN to the levels seen after exposure of the
film to ≤50 mM NaSCN (Figure 3.10). Hence if new ion-exchange sites are created at high ionic
strength, they do not remain when the concentration of NaSCN in solution decreases.

81

0.07

0.05

after washing 50 mM
with water for 2hrs

0.04

after passing 400mM
NaSCN

0.03

Absorbance

0.06

after passing 50 mM
NaSCN

after washing 400 mM
with water for 2 hrs

0.02
0.01
0.00
2150

2100

2050

2000

1950

-1

Wavenumbers (cm )
Figure 3.10. ATR-FTIR spectra of absorbed SCN

-

-1

(peak at 2058 cm ) in a

(PDADMAC/PSS)30PDADMAC film after a 6 h exposure to an aqueous
solution of 50 mM (green) or 400 mM (red) NaSCN, and after rinsing of
these films with deionized water for 2 h (blue or pink). The pink spectrum is
difficult to see because it overlaps the blue one.

3.3.3. Competitive ion exchange
Competitive binding experiments, where a (PDADMAC/PSS) 30PDADMAC film is
exposed to a solution containing both 0.5 mM NaSCN and 0.5 mM K2Ni(CN)4, demonstrate
-

-1

-

initial rapid binding of SCN (peak at 2060 cm ) followed by displacement of SCN by
2-

Ni(CN)4

-1

(peak at 2123 cm ) (Figure 3.11). Figure 3.12 shows the normalized areas of the

82

-

SCN and Ni(CN)4

2-

peaks from the ATR-FTIR spectra in Figure 3.11. The data suggest that

-

2-

SCN diffuses into the film more quickly than does Ni(CN)4 , although binding of the doubly
charged Ni(CN)4

2-

is energetically favored.

0.030
60 min
20 min
18 min
16 min
15 min
12 min
10 min
7 min
5 min

0.025

Absorbance

0.020

0.015

0.010

0.005

0.000
2150

2100

2050

2000

Wavenumbers (cm-1)
Figure 3.11. ATR-FTIR spectra of a (PDADMAC/PSS)30PDADMAC film during
exposure to a solution containing 0.5 mM NaSCN and 0.5 mM K 2Ni(CN)4.
-

-1

The peaks are due to SCN (2060 cm ) and Ni(CN)4
-

2-

was no change in SCN and Ni(CN)4
solution.

83

2-

-1

(2123 cm ). There

peaks after 60 min of exposure to the

0.3
-

SCN
2Ni(CN)4

Peak Area

0.2

0.1

0.0
0

20

40

60

80

100

Time (min)
-

-1

2-

Figure 3.12. Normalized areas of the SCN (2060 cm ) and Ni(CN)4

-1

(2123 cm ) peaks

in ATR-FTIR spectra of a (PDADMAC/PSS)30PDADMAC film during
exposure to a solution containing 0.5 mM NaSCN and 0.5 mM K 2Ni(CN)4.
-

2-

Figure 3.13 presents the equilibrium concentrations of SCN and Ni(CN)4
-

2-

(PDADMAC/PSS)30PDADMAC film as a function of the SCN to Ni(CN)4

in a

ratio in the

solution. The concentration of NaSCN in solution was varied from 0.05 to 400 mM while the
concentration of Ni(CN)4

2-

was kept constant at 0.5 mM. The data show that at high solution
-

NaSCN concentrations, the SCN does displace Ni(CN)4

2-

from the film.

At solution

concentrations of NaSCN ≤50 mM, where the overall absorption is dominated by intrinsic ion

84

exchange (new adsorption sites are not created upon addition of salt and negligible free salt is
-

2-

present in the film), the exchange reaction between SCN and Ni(CN)4

should be represented

by reaction 3.3.

2 PolSCN  Ni(CN)42-

Pol2 Ni(CN)4  2 SCN-

------------ (3.3)

1.2

[Ion] in Film (M)

1.0
0.8
0.6

-

SCN

Ni(CN)4

0.4

2-

Total
0.2
0.0
0

50

100

150

200

250

300

350

400

[SCN -] in Solution (mM)
Figure 3.13. Equilibrium

SCN

-

and

2-

Ni(CN)4

concentrations

in

a

(PDADMAC/PSS)30PDADMAC film exposed to solutions containing 0.5 mM
K2Ni(CN)4 and varying concentrations of NaSCN.
2-

(green triangles) is the sum of the Ni(CN) 4
-

concentration in the film

multiplied by 2 and the SCN concentration in the film.

85

The total absorption

-

Based on the three data points in Figure 3.13 where the solution concentration of SCN is ≤50
-

mM and the concentration of SCN in the film is measurable, the equilibrium constant for
reaction 3 is around 4. Attainment of equal concentrations of the two anions in the film requires
-

2-

2-

a solution SCN /Ni(CN)4 ratio of about 50 when the solution concentration of Ni(CN) 4

is 0.5

mM.

3.3.4. Ion binding as a function of the number of adsorbed PDADMAC/PSS bilayers
Polyelectrolyte multilayer properties such as permeability and swelling can vary greatly
with the number of layers in a film,

66, 67

so we also examined anion absorption as a function of

the number of layers in PDADMAC/PSS coatings. Table 3.1 summarizes the results of these
studies. There is no detectable binding of ions in films with <6 bilayers, but the concentration of
absorbed ions in 7-bilayer films ranges from 0.04 to 0.07 M.

Film thickness, in contrast,

increases by only 30% upon the addition of the seventh bilayer. (Even though the film thickness
-

is less than the IR penetration depth, because the concentrations of SCN and Ni(CN)4
solution are low, they do not give rise to a detectable signal.)

2-

in

On going from 7 to 10

PDADMAC/PSS bilayers, the concentrations of absorbed ions in the film typically increase by a
factor of 3-4, but the total absorbed ion concentrations for films with 10 and 30 bilayers differ by
less than a factor of 2. Consistent with ion-exchange, in single-salt absorption, the concentration
-

2-

of SCN in the film is about double the concentration of Ni(CN) 4 .
-

absorption, the ratio of SCN to Ni(CN)4

2-

In competitive ion

in the film decreases as the number of deposited

layers increases. This may occur due to changes in ζ potentials (see below).

86

Table 3.1.

Equilibrium ion concentrations

a

in (PDADMAC/PSS)n films exposed to

solutions containing 2.5 mM NaSCN, or 2.5 mM K 2Ni(CN)4, or 2.5 mM of
both salts.

b

NaSCN in
solution
-

K2Ni(CN)4 in
solution
2-

Mixture of NaSCN and
K2Ni(CN)4 in solution
-

2-

number of
bilayers, n

[SCN ] in film
(M)

[Ni(CN)4 ] in
film (M)

[SCN ] in film
(M)

[Ni(CN)4 ] in
film (M)

4

0.00

0.00

0.00

0.00

5

0.00

0.00

0.00

0.00

6

0.00

0.00

0.00

0.01

7

0.07

0.04

0.04

0.04

10

0.31

0.15

0.02

0.15

30

0.46

0.27

0.00

0.28

a

The concentration of the ions in the films was calculated from ATR-IR spectra of films on a Ge
crystal as described in Appendix A.
b
Experiments with absorption of KSCN and NaSCN showed that the cation did not affect the
absorbed ion concentration.
For comparison, Table 3.2 shows the concentrations of ions in the film when PDADMAC
is the top layer. In this case, the surface is positively charged, so in thin films (4.5 bilayers, for
example), many of the observed ions may be adsorbed at the film surface. The concentrations of
ions in the film increase as the number of bilayers increases from 4.5 to 10.5, which shows that
not only are ions adsorbed on the surface but they also are absorbed at ion-exchange sites within
the polyelectrolyte multilayer. These ion-exchange sites evidently increase in concentration as
more layers are added to the film. The concentration of absorbed ions in the film with 30.5
bilayers is about 15% less than that in the 7.5-bilayer film, perhaps partly because the penetration
depth is reached and the extra ions adsorbed on the positively charged surface of the 30.5-bilayer
film are no longer observable. For 30 and 30.5-bilayer films, the concentrations of absorbed ions
87

are essentially independent of whether the top layer is PSS or PDADMAC (compare Tables 3.1
and 3.2).

Table 3.2.

Equilibrium ion concentrations in (PDADMAC/PSS)nPDADMAC films
exposed to solutions containing 2.5 mM NaSCN, or 2.5 mM K 2Ni(CN)4, or
2.5 mM of both salts.
NaSCN in
solution
-

K2Ni(CN)4 in
solution
2-

Mixture of NaSCN and
K2Ni(CN)4 in solution
-

2-

number of
a
bilayers

[Ni(CN)4 ] in
film (M)

[SCN ] in film
(M)

[Ni(CN)4 ] in
film (M)

4.5

0.40

0.21

0.01

0.19

5.5

0.42

0.24

0.03

0.21

6.5

0.50

0.27

0.03

0.22

7.5

0.55

0.29

0.05

0.24

10.5

0.60

0.32

0.01

0.31

30.5
a

[SCN ] in film
(M)

0.51

0.26

0.00

0.28

The extra 0.5 bilayer indicates that films are terminated with PDADMAC.

3.3.5. Streaming Potential Measurements
The results in Table 3.1 show that the concentration of ion-exchange sites increases
dramatically on going from 6 to 7 and 7 to 10 bilayers in (PDADMAC/PSS) n films where n is an
integer. Such increases should also be reflected in changes in the ζ potentials of the films, so we
measured streaming potentials on multilayer coatings deposited on both silicon wafers and PES
membranes.

88

60
50
40

Zeta Potential (mV)

30
20
10
0
-10
-20
-30

-40
-50
0

1

2

3

4

5

6

7

8

9

10 11 12 13 14 15 16

Number of bilayers (n)
Figure 3.14.

Zeta Potentials of Si wafers coated with (PDADMAC/PSS) n films. Integer
numbers of bilayers represent films terminated with PSS, while films with an
extra ½ bilayer are terminated with PDADMAC.

Figure 3.14 shows ζ potentials of PDADMAC/PSS films on silicon wafers as a function
of the number of bilayers. As seen in a number of previous studies,

68-70

the ζ potential shifts

between positive and negative values, respectively, as PDADMAC and PSS are added.
However, the magnitude of the ζ potential for PSS-terminated films begins to decrease after the
deposition of about 6 bilayers, and 16-bilayer films no longer have a negative ζ potential.

89

50
40

Zeta Potential

30
20
10
0
-10
-20
-30

0

1

2

3

4

5

6

7

8

9

10

Number of bilayers
Figure 3.15.

Zeta Potentials of 100 kDa PES membranes coated with (PDADMAC/PSS) n
films. Integer numbers of bilayers represent films terminated with PSS,
while films with an extra ½ bilayer are terminated with PDADMAC.

The ζ potentials of PDADMAC/PSS films on 100 kDa PES membranes (Figure 3.15)
exhibit a similar trend to films on silicon, but the ζ potentials of PSS-terminated films on PES
become positive after deposition of only 7 bilayers. Changes in ζ potentials may also affect the
selectivity of ion exchange. With 7-bilayer PDADMAC/PSS on Ge, the negative zeta potential
of the film likely provides some preferential electrostatic exclusion of Ni(CN) 4
2-

stronger binding of Ni(CN)4
-

2-

SCN and Ni(CN)4

2-

to offset the

to anion exchange sites in the film. Hence the amounts of bound

in 7-bilayer films are about the same (Table 3.1). As the zeta potential

becomes less negative when more bilayers are adsorbed, Ni(CN) 4
90

2-

becomes the anion

predominantly absorbed in the film.

This trend also agrees with the nanofiltration results

presented below.
The SEM images in Figure 3.3 confirm that film growth continues to occur even when
the surface potential does not become negative after deposition of PSS. This variation of zeta
potential is consistent with the exponential growth mechanism mentioned earlier, where large
amounts of PDADMAC enter the film during polycation adsorption and these polymers migrate
to the surface and complex with PSS during the PSS deposition step.

71, 72

Although the

PDADMAC surface charge is not fully compensated during PSS adsorption, more PDADMAC
diffuses into the film during the next adsorption step. Such a growth mechanism, while not
operative at low ionic strengths, seems to occur when PSS/PDADMAC films are deposited from
1 M NaCl.

73

3.3.6. Nanofiltration Properties of PSS/PDADMAC films as a function of the number of
bilayers
Changes in ζ potential and ion absorption should also greatly affect transport through
multilayer polyelectrolyte films, so we investigated the nanofiltration properties of
PSS/PDADMAC coatings on both porous alumina substrates and PES ultrafiltration membranes.
(All films terminate with PSS to enhance divalent anion rejection.) Table 3.3 shows results for
films on alumina, where the initially deposited layer in the polyelectrolyte film is PSS because
alumina is positively charged below pH 8.

74, 75

The results are reported in terms of rejection, R,

which is defined in equation 3.4, where Cperm and Cfeed are the solute concentrations in the
permeate and feed, respectively. The selectivity, S, for transport of solute A over solute B is

91

defined by equation 3.5, which can also be expressed in terms of rejections for solute A and
-

solute B as shown. In some cases, negative rejection of Cl occurs when the combination of a
+

2-

-

high flux of Na and a high SO4 rejection requires enhanced Cl transport to maintain electrical
-

neutrality in the permeate. In such cases the concentration of Cl in the permeate can actually
exceed that in the feed.


C perm 
  100%
R  1 

C feed 


C A, perm CB, feed 100  RA
S

C A, feed CB, perm 100  RB

------------ (3.4)

------------ (3.5)
2-

With both (PSS/PDADMAC)3PSS and (PSS/PDADMAC)4PSS films on alumina, SO4
-

rejection is about 97%, while Cl /SO4
previously.

35

2-

selectivity is above 30, similar to the results we saw

(Membranes containing (PSS/PDADMAC) 2PSS films show a selectivity of

around 1 because the film does not completely cover the substrate, data not shown.) After
adsorption of 4.5 bilayers, the addition of more bilayers results in a rapid decline in selectivity.
Film thickness still increases with the number of bilayers deposited, as shown by decreases in
2-

flux with adsorption of additional PDADMAC/PSS bilayers, but SO 4

-

2-

rejection and Cl /SO4

selectivity decline. This is consistent with both the drop in ζ potential after deposition of about 4
bilayers on PES supports and the rapid increase in anion adsorption seen after deposition of 6 to
7 PDADMAC/PSS bilayers on the Ge ATR crystal. The presence of excess positive charge in
2-

the membrane probably results in less electrostatic exclusion of the divalent SO4 . Faster
2-

hopping of SO4

between the highly concentrated ion exchange sites could provide a second
92

2-

mechanism for decreased SO4
with water.

rejection.

42

Films with more layers may also be more swollen

66

Table 3.3.

Rejections, solution fluxes, and selectivities from nanofiltration experiments
with (PSS/PDADMAC)nPSS-coated porous alumina membranes and feed
-

2- a

solutions containing 1000 ppm Cl and 1000 ppm SO4 .
number of
b
bilayers

ellipsometric solution flux,
c
3 2
thickness, nm m /m -day

-

2-

-

2-

Cl
rejection, %

SO4
rejection,%

Cl / SO4
selectivity

3.5

1.7±0.1

-16±2

96.9±0.2

37±2

4.5

36.6±0.4

1.15±0.06

-11±1

96.7±0.5

34±5

5.5

63±2

0.99±0.06

-5±3

93±1

15±3

6.5

100±4

0.83±0.04

3±2

86±2

6.8±0.6

7.5
a

18.0±0.1

143±7

0.65±0.03

2±1

68±6

3.2±0.7

Solutions were prepared with NaCl and Na2SO4.

b

The extra half bilayer indicates that coatings were terminated with PSS.

c

Thicknesses were measured with films deposited on Si wafers.
To better correlate ζ potentials and transport, we performed nanofiltration with the same

PES membranes used for streaming potential measurements.

(Commercial porous alumina

membranes are too small for the streaming potential apparatus.) Using 100 kDa PES supports
2-

(Table 3.4), flux decreases with the number of adsorbed PDADMAC/PSS bilayers, but SO 4
-

2-

rejection and Cl /SO4

selectivity peak after deposition of 6 bilayers. The polymeric substrates

are probably less charged than alumina, and thus it may take a few more layers to achieve
-

complete coverage and maximum selectivity. Previously we achieved a maximum Cl /SO4

2-

selectivity of 32 for 4.5-bilayer PSS/PDADMAC films on a 50 kDa PES membrane from
93

Millipore.

67

However, those films were deposited from a 0.5 M NaCl solution, rather than a 1

M solution, which could greatly affect selectivity. Variations in the support can also change
selectivity.

With a 50 kDa PES support from Pall Corporation, we achieved a maximum

selectivity of 13 using PDADMAC/PSS films deposited from 1 M NaCl as shown in Table 3.5.

Table 3.4.

Rejections, solution fluxes, and selectivities from nanofiltration experiments
with (PDADMAC/PSS)n-coated 100 kDa PES membranes and feed solutions
-

2- a

containing 1000 ppm Cl and 1000 ppm SO4 .
-

2-

-

2-

number of
bilayers,n

Cl
rejection, %

SO4
rejection,%

Cl / SO4
selectivity

3

7.5±1.7

-1.0±0.4

7±4

1.1±0.1

4

2.4±0.5

-6±2

59±2

2.6±0.2

5

0.95±0.05

-16.8±0.6

79±4

6±1

6

0.82±0.03

-13±3

85±3

8±1

7

0.73±0.02

-2±8

80±4

5±1

8
a

solution flux,
3 2
m /m -day

0.67±0.01

0.7±4

70±4

3.1±0.2

Solutions were prepared with NaCl and Na2SO4.
As a final correlation between transport and ion absorption, we examined nanofiltration
-

2-

of SCN /Ni(CN)4

mixtures using PSS/PDADMAC films prepared on alumina supports. The
-

2-

trend is similar to that for Cl /SO4

separations as shown in Table 3.6, with selectivity peaking

after deposition of 4.5-bilayers. However, the maximum selectivity is only 2.

94

Table 3.5.

Rejections, solution fluxes, and selectivities from nanofiltration experiments
with (PDADMAC/PSS)n-coated 50 kDa PES membranes and solutions
-

2- a

containing 1000 ppm Cl and 1000 ppm SO4 .
-

2-

-

2-

number of
bilayers,n

Cl
rejection, %

SO4
rejection,%

Cl / SO4
selectivity

4

2.17±0.09

-12±3

69±14

5±2

5

1.28±0.04

-1±6

87±4

8±3

6

0.97±0.02

-13±2

91±3

13±3

7

0.79±0.03

0.6±3

90±3

10±3

8
a

solution flux,
3 2
m /m -day

0.69±0.02

0.6±1

81±5

5±1

Solutions were prepared with NaCl and Na2SO4.

Table 3.6.

Rejections, solution fluxes, and selectivities from nanofiltration experiments
with (PSS/PDADMAC)nPSS-coated alumina membranes and feed solutions
-

2- a

containing 1000 ppm SCN and 1000 ppm Ni(CN)4 .
-

2-

-

number of
b
bilayers

SCN
rejection, %

Ni(CN)4
rejection,%

3.5

2.5±0.2

1±2

33±9

1.5±0.2

4.5

1.51±0.05

2.6±0.9

50±7

2.0±0.3

5.5

1.09±0.03

2±3

41±6

1.7±0.2

6.5

0.76±0.01

-3±4

32±3

1.5±0.1

7.5
a

solution flux,
3 2
m /m -day

0.66±0.02

-17±5

21±7

1.5±0.1

Solutions were prepared with KSCN and K2Ni(CN)4.

b

The extra half bilayer indicates that films terminate with PSS.

95

SCN / Ni(CN)4
Selectivity

2-

Table 3.7.

Rejections, solution fluxes, and selectivities from nanofiltration experiments
with (PSS/PDADMAC)4PSS-coated porous alumina membranes and feed
solutions containing different combinations of 1000 ppm anions .
solution flux,
3 2
m /m -day

-

-

SCN /SO4
-

a

96.7±0.5

34±5

2.6±0.9

50±7

2.0±0.3

-39±3

94.5±0.2

25±1

1.1±0.2

2- d

Cl / Ni(CN)4

-11±1

0.76±0.04

2- c

selectivity

1.51±0.05

2- b

SCN /Ni(CN)4

divalent ion
rejection,%

1.15±0.06

2- a

Cl /SO4

monovalent ion
rejection, %

3±3

45±2

1.8±0.1

Solution was prepared with NaCl and Na2SO4.

b

Solution was prepared with KSCN and K2Ni(CN)4.

c

Solution was prepared with NaSCN and Na2SO4.

d

Solution was prepared with KCl and K2Ni(CN)4.
-

2-

To understand why the SCN /Ni(CN)4
/SO4

2-

-

selectivity was so low compared to the Cl

selectivity, we performed nanofiltration of different combinations of ions with a series of

different membranes.
2-

Ni(CN)4

2-

is quite low compared to that of SO4 , even though both are divalent ions. The low
2-

rejection of Ni(CN)4
-

As Table 3.7 shows, these experiments reveal that the rejection of

2-

-

yields Cl / Ni(CN)4
-

2-

-

2-

and SCN /Ni(CN)4

selectivities of 2 or less, while

2-

both Cl /SO4 and SCN /SO4 selectivities are 25 or more.
-

2-

-

2-

-

2-

We also sequentially examined Cl /SO4 , SCN /Ni(CN)4 , and Cl /SO4
-

2-

in NF with the same membrane (Table 3.8). The initial Cl /SO4
-

SCN /Ni(CN)4

2-

selectivities

selectivity was 26, and the

selectivity was 3, similar to previous experiments. However, when the Cl

96

-

/SO4

2-

-

2-

selectivity was determined after the SCN /Ni(CN)4

decreased to 10, suggesting that Ni(CN)4

2-

nanofiltration, the selectivity

changes the film properties by making it more

2-

permeable to SO4 .

Table 3.8.

Rejections, solution fluxes, and selectivities from nanofiltration experiments
with (PSS/PDADMAC)4PSS-coated porous alumina membranes and
solutions containing different combinations of 1000 ppm anions.

The

experiments were performed sequentially (from top to bottom) using the
same two (PSS/PDADMAC)4PSS-coated alumina membranes.
solution flux,
3 2
m /m -day
-

2- a
2- b

SCN /Ni(CN)4
-

a

selectivity

-16.5±0.8

95.4±0.2

26±1

1.21±0.03

-26±1

52±9

2.7±0.6

-20.4±0.6

88.0±0.5

10.0±0.4

2- a

Cl /SO4

divalent ion
rejection,%

1.00±0.03

Cl /SO4
-

monovalent ion
rejection, %

1.07±0.03

Solution was prepared with NaCl and Na2SO4.

b

Solution was prepared with KSCN and K2Ni(CN)4.
-

2-

-

2-

Similar results were observed when sequentially examining SCN /SO4 , Cl /Ni(CN)4 ,
-

2-

and SCN /SO4

-

2-

separations. The first SCN /SO4

selectivity was 28, while the second was

only 12 as shown in Table 3.9. Although we do not understand the effect of Ni(CN) 4

2-

on the

film, it obviously alters film structure to decrease selectivity. Tieke and coworkers also showed
effects of ion absorption on selectivity.

76, 77

Overall, the trends in transport with the number of deposited layers are consistent with a
film that becomes less selective after deposition of 4 to 6 bilayers. This corresponds with the
97

point at which films rapidly increase in thickness and gain anion-exchange sites.

These

additional anion-exchange sites likely reduce selectivity by decreasing Donnan exclusion,
increasing hopping pathways for divalent ions, or increasing swelling.

Table 3.9.

Rejections, solution fluxes, and selectivities from nanofiltration experiments
with (PSS/PDADMAC)4PSS-coated porous alumina membranes and
solutions containing different combinations of 1000 ppm anions.

The

experiments were performed sequentially (from top to bottom) using the
same two (PSS/PDADMAC)4PSS-coated alumina membranes.
solution flux,
3 2
m /m -day
-

SCN /SO4
-

2- a

-

SCN /SO4
a

divalent ion selectivity
rejection,%

0.85±0.03

2- b

Cl / Ni(CN)4

monovalent ion
rejection, %
-41.0±0.6

94.9±0.1

27.7±0.8

1.01±0.03

0.7±1.8

44±6

1.8±0.2

-26.8±0.6

89.6±0.2

12.2±0.2

2- a

0.91±0.03

Solution was prepared with NaSCN and Na2SO4.

b

Solution was prepared with KCl and K2Ni(CN)4.

3.4. Conclusions
The properties of PDADMAC/PSS films vary dramatically with the number of layers in
the film. (PDADMAC/PSS)n films on a Ge ATR crystal do not absorb significant amounts of
-

2-

SCN or Ni(CN)4

when n is an integer less than 7, but films with ten bilayers have a

concentration of anion-exchange sites of about 0.5 M.
2-

Ni(CN)4

2-

Consistent with anion-exchange,
-

is the predominantly absorbed species from equimolar mixtures of SCN and
-

2-

Ni(CN)4 , even though SCN initially exchanges into the film more rapidly than Ni(CN) 4 .

98

Streaming potential measurements suggest that the high ion-exchange capacities of films with
more than 5-6 bilayers stem from excess positive charge due to more ammonium than sulfonate
groups in the film, even when PSS is the terminal layer. These data explain why transport
-

through PDADMAC/PSS films is such a strong function of the number of layers in the film. Cl
/SO4

2-

selectivity peaks after adsorption of 4-6 bilayers because a minimum number of layers are

required to fully cover the support, and deposition of additional layers results in a more
2-

positively charged film with decreased SO4

rejection. This knowledge is vital for designing

PSS/PDADMAC membranes.

99

Appendix

100

3.5. Appendix A
Calculation of Ion Concentrations in Polyelectrolyte Films
We plot ATR-IR spectra as absorbance, -log (I/Io), where I is the intensity of the light
-

reaching the detector when a SCN -containing film or solution is on the surface of the ATR
-

crystal and I o is the intensity of the light reaching the detector when no SCN is in the film or
-

-

solution. To convert the SCN ATR-IR absorbance to SCN concentration in a film, we use
ATR-IR spectra of aqueous NaSCN solutions for calibration.

However, because I/I o is a

complicated function of both film thickness and dielectric constant, the calibration requires
calculation of I/Io using Fresnel reflection coefficients.
-

For a single bounce on an ATR crystal covered with a SCN solution as shown in Figure
A1, the ratio of reflected intensity, I R, to incident intensity, Iin can be calculated using equation
A.1, where Ein is the amplitude of the incident electric field, ER is the amplitude of the reflected
field, r12 is the Fresnel reflection coefficient at the crystal (medium 1)/solution (medium 2)
interface, and r12* is the complex conjugate of this reflection coefficient.

I R / Iin 

2
ER
2
Ein

*
 r12 r12

------------ (A.1)

101

Inlet

Outlet
NaSCN Solution
Evanescent Wave
ATR Crystal

Incident Light

Figure A1.

Reflected Light

Single-reflection ATR-FTIR experiment with a crystal exposed to a NaSCN
solution.

78

The actual apparatus in this work employed 6 reflections.

The values for r12 depend on polarization and can be calculated using the Fresnel
equations shown in equation A.2, where n1 and n2 are the refractive indices of the crystal and the
solution, respectively, and θi and θt are the angles of incidence and refraction. (Snell’s law, n 1
sin θ1= n2 sin θt is used to calculate the value of cos θt.)

n cos i  n1 cos t
ll
r12  2
n2 cos i  n1 cos t

r12

n cos i  n2 cos t
 1
n1 cos i  n2 cos t

------------ (A.2)

Note that for light-absorbing solutions or films, n2 is a complex number (n2+ik2) as is cos θt.
For an unpolarized IR beam and N bounces in the ATR crystal, the total absorbance, Atot,
is given by equation A.3,
 NAll

Atotal   log[( Ill  I  ) / ( I o ll  I o )]   log[( I o ll10

102

 NA

 I o 10

) / ( I o ll  I o )] -------- (A.3)

where All and A┴ are the values of -log(I/Io) for a single bounce of radiation polarized parallel
and perpendicular to the plane of incidence, respectively

79, 80

. The I/Io values for a single
-

bounce are calculated by dividing IR/Iin for the solution or film containing SCN by the IR/Iin
-

value in the absence of SCN , which is 1. Thus, All and A┴ are simply log(IR/Iin) values for
-

SCN absorption in parallel and perpendicular polarizations, respectively. Equation A.3 can be
rewritten as equation A.4,

Atotal   log[ x10

 NAll

 (1  x)10

 NA

]

------------ (A.4)

where x represents the fraction of incident light that is polarized parallel to the plane of
incidence.

79, 80

The x value of 0.375 was calculated using a polarizer to determine the relative

intensities of the IR beam polarized perpendicular and parallel to the plane of incidence.
Calculation of All and A┴ and, hence, Atotal requires values for the angle of incidence, the
refractive index of the crystal, and the complex refractive index of the solution or film. Because
-

the solutions or films were relatively dilute with respect to SCN , the real part of the refractive
index was assumed to be that of the pure swollen film or pure solvent (see below for further
discussion of this point), while the value of k was used to fit the absorbance. The value of k is
-

directly proportional to SCN concentration, C, and determination of k at several solution
concentrations with no film on the crystal afforded the constant of proportionality Q shown in
equation A.5. Use of equation A.5 and the value of Q determined from spectra of standard
-

solutions on a bare crystal allows calculation of the SCN concentration in the film from the
value of k in the film.
103

C  k /Q

------------ (A.5)

For a film whose thickness is much greater than the penetration depth of the evanescent
-

wave, determination of SCN concentration is relatively straightforward. Equations A.1-A.4
allow iterative calculation of k from absorbance. The real part of the film refractive index
determined using in situ ellipsometry was taken as 1.442 for PSS-terminated films in water and
1.345 for PDADMAC-terminated films in water, and k2 was varied to fit the absorbance. These
-1

refractive indices of the water-swollen films at 2058 cm

(4859 nm) were estimated from the

Cauchy model used to fit the refractive index of swollen films on Si at the wavelengths of the
ellipsometer (414 – 736.1 nm). Refractive indices determined for swollen 4 and 4.5-bilayer
PDADMAC/PSS films were applied to all films terminated with the same polyelectrolyte. The
-1

refractive index of the aqueous solutions at around 2058 cm was 1.323.

81

In the case of films with <10 bilayers, film thickness is also an important parameter for
determining absorbed ion concentrations (see below). The thicknesses of 4- to 5.5-bilayer
PDADMAC/PSS films in water were determined using in situ ellipsometry as described in
reference 28. Table A1 summarizes the results for 4-, 4.5-, and 5.5-bilayer PDADMAC/PSS
films on Si wafers. The percent swelling of PSS-terminated films is around 150% while the
PDADMAC-terminated films swell around 400%.

104

Table A1.

Ellipsometric thicknesses and %swelling of PDADMAC/PSS films on silicon
wafer immersed in water.
a

number of bilayers

wet thickness (nm)

% swelling

4

37.7±0.7

95.0±1.9

152±5

4.5

59.7±1.1

291.8±8.2

389±18

5.5
a

dry thickness (nm)

86.9±1.8

465.2±30.3

435±32

The extra 0.5 bilayer indicates that films are terminated with PDADMAC.
Swollen thicknesses for films with more than 5.5 bilayers could not be obtained because

the films were rough and the ellipsometric model did not give good fits. For films with 5 to 10.5
bilayers, dry ellipsometric thicknesses (Figure 3.1) along with the swelling factors in Table A1
were used to obtain swollen thicknesses. Figure 3.5 suggests, however, that differences in
swelling of PSS and PDADMAC-terminated films decrease for films with more than about 8
bilayers, so the application of refractive indices determined for 4- or 4.5-bilayer films to films
with as many as 30 or 30.5 bilayers could cause errors in calculated concentration values for the
thicker films. However, the difference in calculated concentrations obtained using refractive
indices of 1.345 and 1.442 is <12%.

Even in the case of (PDADMAC/PSS)7PDADMAC films,

where both thickness and refractive index are input parameters, calculated values of absorbed ion
concentrations differ by only 5% when assuming 150% swelling and n=1.442 rather than 400%
swelling and n=1.345.

The effect of large differences in assumed swelling on calculated

concentrations is relatively small because even a 150% swollen 7.5-bilayer film would have a
thickness of 1.5 penetration depths, and the lower refractive index and higher thickness have
opposite effects on calculated concentrations.

105

Solution, n3+ik3
Film, n2+ik2
Crystal, n1

Incident Light

Figure A2.

Reflected Light

Schematic drawing of an optical system containing three layers (crystal,
swollen film, and solution).

Calculation of absorbed ion concentrations is more complicated for films with
thicknesses less than 2 or 3 penetration depths than for thicker films. In this case the optical
system consists of three layers as shown in Figure A2. The absorbance for a single bounce is
determined by the overall reflection coefficient, which is the term in brackets in equation A.6.

*

 r  r exp(2i )   r12  r23 exp(2i ) 
I R / Iin 
  12 23


Ei2 1  r r23 exp(2i )  1  r12 r23 exp(2i ) 
12



2
Er

  2

d



---------- (A.6)

n2 cos  2

In this equation, r12 is the Fresnel reflection coefficient for the crystal/film interface, r 23 is the
Fresnel reflection coefficient for the film/solution interface, d is the swollen film thickness, and λ
is the wavelength of interest. The values of cos θ 2 and cos θ3 needed for calculating r12 and r23
are determined using Snell’s law.

106

0.6

Absorbance

0.5

0.4
0.3
0.2

0.1
0.0
0

0.1

0.2

0.3

0.4

k
Figure A3.

Simulated values of absorbance at 2058 cm

-1

as a function of k using

equation A.6. Refractive index and thickness of the film are taken as 1.345
and 260 nm, respectively. The water above the film had a refractive index of
1.323.
-

Because the concentration of SCN in the film is much greater than the SCN

-

concentration in the solution, k3 is negligible, and n3 is simply the refractive index of water. (If
needed, the concentration-dependent k3 determined from solution studies with the ATR crystal
could be included.) The thickness and refractive index of the film can be estimated from in situ
ellipsometry of films on Si wafers so the only remaining unknown is k2, which can be iteratively
calculated from a value for the overall absorbance. Again, k 2 is converted to concentration using
equation A.5 with the Q value determined from standard aqueous solutions.
The above discussion uses the absorbance at a given wavelength to determine
concentration. However, ideally we would like to determine the concentration of a species in a
film based on the area of a peak, particularly when peak wavelengths show small shifts.
107

Peak Area (arbitrary units)

1.4
1.2
1.0
0.8
0.6
0.4
NaSCN
0.2

K2Ni(CN)4

0.0
0

100

200

300

400

500

600

Conc (mM)
Figure A4.

Peak Area versus the concentration of salt in a solution contacting the ATR
crystal for solutions containing NaSCN and K 2Ni(CN)4. The inset shows
data for low concentrations.

We can show by simulation that for low absorbances (k(λ)<0.38), at each wavelength the
absorbance is proportional to k(λ) (Figure A3). Thus, the overall peak area is proportional to the
k values at each wavelength, and these k values should vary linearly with concentration. Figure
A4 shows a plot of peak area versus the concentration of salt in a solution contacting the ATR
crystal for both NaSCN and K2Ni(CN)4. The plots are linear as expected. For each peak, we
can calculate an average absorbance, Aavg, such that peak area = AavgΔv, where Δv is the width
in wavenumbers over which the peak is integrated. From this Aavg, we can calculate a kavg
using the overall reflection coefficient and make a plot of k avg versus concentration. Over the
integration range, the value of λ does not make a significant difference (<3%) in the calculated
value of A for a given k and thickness. The effect of the dispersion in the refractive index of
108

-

-

SCN should be small because the concentrations of SCN in the film or solution are low enough
that the refractive index of the composite should not be greatly affected.

0.008

Absorbance

0.006

0.004

0.002

0.000
2150

2100

2050

2000

1950

Wavenumber (cm-1)
Figure A5.

ATR-FTIR spectra of 0.1 M NaSCN in aqueous (purple) and 80% ethanol in
water (green) solutions containing 0.2 M PDADMAC.

The ATR-FTIR

spectrum of 0.1 M NaSCN in water without PDADMAC is shown in blue.
To determine the concentration of the species in a film with a thickness much greater
than the penetration depth, we determine Aavg using the same width for the integration as in
solution spectra and calculate a kavg using the reflection coefficient equation A.1 and equation
A.4. Concentration can be determined using equation A.5 where Q is calculated from kavg
-

values for solutions containing different concentrations of SCN . The situation is slightly more

109

complicated with thinner films. We again determine Aavg and calculate the kavg value, but this
time the overall reflection coefficient (equation A.6) must be used to take into account the threephase (crystal-film-solution) interface. With a kavg for the film in hand, we can use equation A.5
to calculate the concentration in the film.
One major question that arises in this method is whether the value of k avg for a given
concentration is the same in water and in the film. Simple expressions such as the Maxwell
Garnett mixing rule would suggest that changes in k due to the changes in refractive index
between water (n=1.323) and the film (n=1.44, for example) would be less than 6%. However,
such expressions neglect specific interactions that may occur. To assess changes in k as a
function of both dielectric constant and interactions, we measured the spectra of 0.1 M NaSCN
in deionized water and in solutions containing 0.2 M PDADMAC in water, and 0.2 M
PDADMAC in 80% ethanol in water (v:v).

As shown in Figure A5, compared to the spectrum
-

of 0.1 M NaSCN in pure water, the peak position for SCN in solutions containing 0.2 M
-1

PDADMAC changes from 2065 cm

-1

to 2062 cm , and the peak area increases by 5% on going

from water to aqueous 0.2 M PDADMAC and by 8% on going from aqueous 0.2 M PDADMAC
to 0.2 M PDADMAC in 80% ethanol. From simulation, the absorbances for solutions with the
-1

same k values and refractive indices of 1.323 (pure water at 2058 cm ) and 1.355 (80% ethanol
in water, refractive index estimated from the difference in water and 80% ethanol refractive
indices at 589.3 nm) should differ by about 4%, which agrees reasonably well with the spectra
for NaSCN in 0.2 M PDADMAC in water and 80% ethanol. Overall these data suggest that film

110

-

concentrations of SCN calculated from ATR-FTIR spectra could differ by a maximum of 1020% from the true value.
We note that Jaber and Schlenoff utilized a powerful method for calculating the ratios of
different species in films with thicknesses greater than 2 or 3 penetration depths.

65

They simply

used the ratio of the areas of peaks in solution as a calibration and then determined the ratios of
species in the film from their peak areas. This method does not give an absolute concentration,
however, unless the concentration of one of the species in the film is known.

111

79

References

112

3.6. References
1.

Jaber, J. A.; Schlenoff, J. B., Curr. Opin. Colloid Interface Sci. 2006, 11(6), 324-329.

2.

Decher, G.; Schlenoff, J. B., Multilayer Thin Films: Sequential Assembly of
Nanocomposite Materials. Wiley-VCH: Weinheim: Germany, 2003; p 524.

3.

Decher, G.; Hong, J. D.; Schmitt, J., Thin Solid Films 1992, 210/211, 831.

4.

Decher, G., Science 1997, 277(5330), 1232-1237.

5.

Benkirane-Jessel, N.; Lavalle, P.; Ball, V.; Ogier, J.; Senger, B.; Picart, C.; Schaaf, P.;
Voegel, J.-C.; Decher, G., Polyelectrolyte Multilayer Films–A General Approach to
(Bio)functional Coatings. Wiley-VCH: Weinheim: Germany, 2007; Vol. 2, p 1249-1305.

6.

Cheung, J. H.; Fou, A. F.; Rubner, M. F., Thin Solid Films 1994, 244(1-2), 985-9.

7.

Decher, G.; Lehr, B.; Lowack, K.; Lvov, Y.; Schmitt, J., Biosens. Bioelectron. 1994, 9(910), 677-684.

8.

Fou, A. C.; Onitsuka, O.; Ferreira, M.; Rubner, M. F.; Hsieh, B. R., J. Appl. Phys. 1996,
79(10), 7501-7509.

9.

Hammond, P. T.; Whitesides, G. M., Macromolecules 1995, 28(22), 7569-7571.

10.

Laurent, D.; Schlenoff, J. B., Langmuir 1997, 13(6), 1552-1557.

11.

Lvov, Y. M.; Lu, Z.; Schenkman, J. B.; Zu, X.; Rusling, J. F., J. Am. Chem. Soc. 1998,
120(17), 4073-4080.

12.

Podsiadlo, P.; Michel, M.; Lee, J.; Verploegen, E.; Wong Shi Kam, N.; Ball, V.; Lee, J.;
Qi, Y.; Hart, A. J.; Hammond, P. T.; Kotov, N. A., Nano Lett. 2008, 8(6), 1762-1770.

13.

Dotzauer, D. M.; Dai, J.; Sun, L.; Bruening, M. L., Nano Lett. 2006, 6(10), 2268-2272.

14.

van Ackern, F.; Krasemann, L.; Tieke, B., Thin Solid Films 1998, 327-329, 762-766.

15.

Sullivan, D. M.; Bruening, M. L., Chem. Mater. 2003, 15(1), 281-287.

16.

Krasemann, L.; Toutianoush, A.; Tieke, B., J. Membr. Sci. 2001, 181(2), 221-228.

17.

Lenk, W.; Meier-Haack, J., Desalination 2002, 148(1-3), 11-16.

18.

Meier-Haack, J.; Lenk, W.; Lehmann, D.; Lunkwitz, K., J. Membr. Sci. 2001, 184(2),
233-243.
113

19.

Rmaile, H. H.; Schlenoff, J. B., J. Am. Chem. Soc. 2003, 125(22), 6602-6603.

20.

Hong, S. U.; Bruening, M. L., J. Membr. Sci. 2006, 280(1-2), 1-5.

21.

Hollman, A. M.; Bhattacharyya, D., Langmuir 2004, 20(13), 5418-5424.

22.

Stair, J. L.; Harris, J. J.; Bruening, M. L., Chem. Mater. 2001, 13(8), 2641-2648.

23.

Johnston, A. P. R.; Cortez, C.; Angelatos, A. S.; Caruso, F., Curr. Opin. Colloid Interface
Sci. 2006, 11(4), 203-209.

24.

Ai, H.; Jones, S.; Lvov, Y., Cell Biochem. Biophys. 2003, 39(1), 23-43.

25.

Liu, X.; Gao, C.; Shen, J.; Möhwald, H., Macromol. Biosci. 2005, 5(12), 1209-1219.

26.

Balabushevich, N. G.; Sukhorukov, G. B.; Larionova, N. I., Macromol. Rapid Commun.
2005, 26(14), 1168-1172.

27.

Smuleac, V.; Butterfield, D. A.; Bhattacharyya, D., Langmuir 2006, 22(24), 1011810124.

28.

Miller, M. D.; Bruening, M. L., Chem. Mater. 2005, 17(21), 5375-5381.

29.

Miller, M. D.; Bruening, M. L., Langmuir 2004, 20(26), 11545-11551.

30.

Klitzing, R. v., PCCP 2006, 8, 5012-5033.

31.

Dubas, S. T.; Schlenoff, J. B., Langmuir 2001, 17(25), 7725-7727.

32.

Wong, J. E.; Rehfeldt, F.; Hanni, P.; Tanaka, M.; Klitzing, R. v., Macromolecules 2004,
37(19), 7285-7289.

33.

Picart, C.; Lavalle, P.; Hubert, P.; Cuisinier, F. J. G.; Decher, G.; Schaaf, P.; Voegel, J.
C., Langmuir 2001, 17(23), 7414-7424.

34.

Caruso, F.; Niikura, K.; Furlong, D. N.; Okahata, Y., Langmuir 1997, 13(13), 3422-3426.

35.

Hong, S. U.; Malaisamy, R.; Bruening, M. L., J. Membr. Sci. 2006, 283(1-2), 366-372.

36.

Krasemann, L.; Tieke, B., Mater. Sci. Eng., C 1999, 8-9, 513-518.

37.

Schlenoff, J. B.; Dubas, S. T., Macromolecules 2001, 34(3), 592-598.

38.

Ouyang, L.; Malaisamy, R.; Bruening, M. L., J. Membr. Sci. 2008, 310(1-2), 76-84.

39.

Krasemann, L.; Tieke, B., Langmuir 2000, 16(2), 287-290.
114

40.

Lavalle, P.; Gergely, C.; Cuisinier, F. J. G.; Decher, G.; Schaaf, P.; Voegel, J. C.; Picart,
C., Macromolecules 2002, 35(11), 4458-4465.

41.

Hong, S. U.; Malaisamy, R.; Bruening, M. L., Langmuir 2007, 23(4), 1716-1722.

42.

Farhat, T. R.; Schlenoff, J. B., J. Am. Chem. Soc. 2003, 125(15), 4627-4636.

43.

Fery, A.; Scholer, B.; Cassagneau, T.; Caruso, F., Langmuir 2001, 17(13), 3779-3783.

44.

Schmitt, J.; Grünewald, T.; Decher, G.; Pershan, P. S.; Kjaer, K.; Lösche, M.,
Macromolecules 1993, 26(25), 7058-7063.

45.

Radeva, T.; Grozeva, M., J. Colloid Interface Sci. 2005, 287(2), 415-421.

46.

Tanchak, O. M.; Yager, K. G.; Fritzsche, H.; Harroun, T.; Katsaras, J.; Barrett, C. J., J.
Chem. Phys. 2008, 129(8), 084901-10.

47.

Salomäki, M.; Laiho, T.; Kankare, J., Macromolecules 2004, 37(25), 9585-9590.

48.

Ball, V.; Voegel, J.-C.; Schaaf, P., Langmuir 2005, 21(9), 4129-4137.

49.

Salloum, D. S.; Schlenoff, J. B., Biomacromolecules 2004, 5(3), 1089-1096.

50.

Jin, J. H.; Hong, S. U.; Won, J.; Kang, Y. S., Macromolecules 2000, 33(13), 4932-4935.

51.

Richert, L.; Lavalle, P.; Payan, E.; Shu, X. Z.; Prestwich, G. D.; Stoltz, J.-F.; Schaaf, P.;
Voegel, J.-C.; Picart, C., Langmuir 2004, 20(2), 448-458.

52.

Adamczyk, Z. Zembala, M. Kolasińska, M. Warszyński, P., Colloids Surf., A 2007,
302(1-3), 455-460.

53.

Schwarz, B.; Schönhoff, M., Langmuir 2002, 18(8), 2964-2966.

54.

Harris, J. J.; Stair, J. L.; Bruening, M. L., Chem. Mater. 2000, 12(7), 1941-1946.

55.

Köhler, K.; Shchukin, D. G.; Möhwald, H.; Sukhorukov, G. B., J. Phys. Chem. B 2005,
109(39), 18250-18259.

56.

Salomäki, M.; Vinokurov, I. A.; Kankare, J., Langmuir 2005, 21(24), 11232-11240.

57.

Tan, H. L.; McMurdo, M. J.; Pan, G.; Van Patten, P. G., Langmuir 2003, 19(22), 93119314.

58.

Müller, M.; Rieser, T.; Köthe, M.; Kessler, B.; Brissova, M.; Lunkwitz, K., Macromol.
Symp. 1999, 145(Polymer Sorption Phenomena), 149-159.

115

59.

Öhman, M.; Persson, D.; Leygraf, C., Prog. Org. Coat. 2006, 57(1), 78-88.

60.

Bron, M.; Holze, R., J. Electroanal. Chem. 1995, 385(1), 105-113.

61.

Firman, P.; Xu, M.; Eyring, E. M.; Petrucci, S., J. Phys. Chem. 1992, 96, 8631-8639.

62.

Xu, M.; Eyring, E. M.; Petrucci, S., Solid State Ionics 1996, 83(3-4), 293-300.

63.

Zhang, H.; Wang, J.; Xuan, X.; Zhuo, K., Electrochim. Acta 2006, 51(16), 3244-3248.

64.

Hong, S. U.; Barbari, T. A.; Sloan, J. M., J. Polym. Sci., Part B: Polym. Phys. 1997,
35(8), 1261-1267.

65.

Jaber, J. A.; Schlenoff, J. B., Langmuir 2007, 23(2), 896-901.

66.

Tanchak, O. M.; Barrett, C. J., Chem. Mater. 2004, 16(14), 2734-2739.

67.

Malaisamy, R.; Bruening, M. L., Langmuir 2005, 21(23), 10587-10592.

68.

Smith, R. N.; Reven, L.; Barrett, C. J., Macromolecules 2003, 36(6), 1876-1881.

69.

Wong, J.; Richtering, W., Surface Modification of Thermoresponsive Microgels via
Layer-by-Layer Assembly of Polyelectrolyte Multilayers. In Smart Colloidal Materials,
Springer-Verlag: Berlin, 2006; pp 45-51.

70.

Trotter, H.; Zaman, A. A.; Partch, R., J. Colloid Interface Sci. 2005, 286(1), 233-238.

71.

Lavalle, P.; Picart, C.; Mutterer, J.; Gergely, C.; Reiss, H.; Voegel, J.-C.; Senger, B.;
Schaaf, P., J. Phys. Chem. B 2004, 108(2), 635-648.

72.

Hübsch, E.; Ball, V.; Senger, B.; Decher, G.; Voegel, J.-C.; Schaaf, P., Langmuir 2004,
20(5), 1980-1985.

73.

McAloney, R. A.; Sinyor, M.; Dudnik, V.; Goh, M. C., Langmuir 2001, 17(21), 66556663.

74.

Fukuzaki, S.; Urano, H.; Nagata, K., J. Ferment. Bioeng. 1996, 81(2), 163-167.

75.

Mullet, M.; Fievet, P.; Reggiani, J. C.; Pagetti, J., J. Membr. Sci. 1997, 123(2), 255-265.

76.

Toutianoush, A.; Schnepf, J.; Hashani, A. E.; Tieke, B., Adv. Funct. Mater. 2005, 15(4),
700-708.

77.

Toutianoush, A.; Tieke, B., Mater. Sci. Eng., C 2002, 22(2), 135-139.

78.

http://mmrc.caltech.edu/FTIR/Harrick/GATR/ATR.pdf (11/13/2008).
116

79.

Freger, V.; Ben-David, A., Anal. Chem. 2005, 77, 6019-6025.

80.

Ohta, K.; Iwamoto, R., Appl. Spectrosc. 1985, 39, 418-425.

81.

Wieliczka, D. M.; Weng, S.; Querry, M. R., Appl. Opt. 1989, 28(9), 1714-1719.

117

Chapter 4
Enhancing the Na+/ Mg2+ Transport Selectivity of Polyelectrolyte
Multilayer Films Through Adsorption of Cu2+ or Fe3+
4.1. Introduction
Nanofiltration (NF) is an effective method for selectively removing divalent ions from
water streams in applications such as water softening.

1-7

Typical NF membranes are more

permeable than reverse osmosis membranes, and high permeability along with the low
monovalent ion rejections in NF allow water treatment at much lower pressures than in reverse
osmosis.

8

This research examines methods for increasing the selectivity of NF membranes

prepared by layer-by-layer (LbL) adsorption of multilayer polyelectrolyte films on polymeric
substrates. The polyelectrolyte films serve as a selective membrane “skin”, and the LbL method
is attractive because it allows fine control over the skin thickness through variation of the
number of deposited layers. Minimizing the thickness of the PEM is critical for achieving high
membrane permeabilities, but the skin should be thick enough to cover the substrate without
defects.
2+

Water softening by NF involves selective rejection of Ca
+

2+

and Mg
+

in the presence of
2+

monovalent cations such as Na , and several groups examined the Na /Mg

+

2+

and Na /Ca

selectivities of membranes containing a polyelectrolyte multilayer film as a selective skin.

9-12

(For NF and diffusion dialysis (DD), selectivity is simply the ratio of the fluxes of the ions of
interest.)

In particular, polyelectrolyte deposition at high ionic strength can increase ion-

transport selectivity, presumably through an increase in surface charge that leads to enhanced
118

exclusion of divalent ions.
rejections of charged ions.

13-15

16, 17

Fixed ionic charges present within the film also increase the
Toutianoush et al. studied the alternating permeation of NaCl

and BaCl2 through (PVA/PVS)60 and found that after 6 measurements for 180 min each, the
NaCl/BaCl2 selectivity changes from 8.6 to 17.5, and the permeation rate increases by about
10%.

17

+

2+

This work aims to further increase the Na /Mg

selectivities of multilayer

polyelectrolyte membranes through deliberate adsorption of transition metal cations throughout
the film.
Xiao et al. reported the formation of zero valent iron nanoparticles in PAA/PDADMAC
multilayers adsorbed on cellulose acetate nanofibers by immersing the PEM film in 0.18 M
2+ 18

ferrous chloride for 30 min and subsequently reducing the adsorbed Fe .
papers also demonstrated metal-ion adsorption in polyelectrolyte films.
describes a similar method to adsorb Fe
sites into these films.
+

Na /Mg

2+

3+

or Cu

2+

Several other

19, 20

This chapter

in membranes and introduce anion exchange

The adsorption of these ions yields remarkable improvements in

selectivities during NF through (PSS/PAH) n membranes. Even mM concentrations

of FeCl3 are sufficient to create ion-exchange sites and give high separation factors.

4.2. Experimental Section
Materials.

Poly(sodium 4-styrenesulfonate) (PSS, Mw = 70 000 Da), poly(allyl amine

hydrochloride) (PAH, Mw = 56 000 Da), CuCl2, FeCl3, and 3-mercapto propionic acid (MPA)

119

were purchased from Sigma-Aldrich, and reagent grade MgCl2 and NaCl were obtained from
Columbus Chemical Industries.

Figure 4.1 shows the structures of the polyelectrolytes.

Polyethersulfone (PES) membranes (50 kDa, Millipore catalog number PBQK02510) were
soaked in deionized water for 1 hour and rinsed with water prior to deposition of
polyelectrolytes.

Figure 4.1.

Structures of the polyelectrolytes used in this study.

Preparation of PEM Films and modification with Cu
followed a literature procedure.

21, 22

2+

3+

or Fe . Polyelectrolyte deposition

Briefly, adsorption of PSS occurred by exposing the top of

the substrate to 0.02 M PSS in 0.5 M NaCl for 2 min and rinsing for 1 min with deionized water.
(Concentrations of polyelectrolytes are always given with respect to the repeating unit). The
polycation layer was adsorbed during exposure of the substrate to 0.02 M PAH in 1.0 M NaCl
for 5 min prior to rinsing with water for 1 min. The pH of PSS and PAH deposition solutions
was adjusted to 2.3 with 0.1 M HCl. Polyanion and polycation adsorption steps were repeated to
give the desired number of polyelectrolyte layers, and the membranes were stored in water until

120

use. Unless otherwise noted, the top side of the PEM film was exposed to 0.5 M CuCl2 or FeCl3
for one hour and rinsed with water to modify the membranes with Cu

2+

3+

or Fe .

Characterization of doped PEM films. A monolayer of MPA on the Au wafer served as an
adhesion layer between Au and PAH.

23

After deposition, films were dried with nitrogen, and

the thicknesses were determined at several points on each piece of coated wafer. External
reflection Fourier-transform infrared (FTIR) spectra of films on Au-coated Si were obtained with
a Nicolet Magna 560 FTIR spectrometer using a PIKE grazing angle (80 °) attachment. A
UV/ozone-cleaned gold wafer was used to obtain the background spectrum.
To analyze the amount of iron in the films, the membranes were immersed in 1 M
sulfuric acid for 1 h (to make sure all the iron was extracted) and analyzed by inductively
coupled plasma optical emission spectrometry (ICP-OES) (Varian 710-ES ICP Optical Emission
Spectrometer). Top view images of PES membranes coated with polyelectrolyte multilayers
were taken using a Hitachi S-4700 II field-emission scanning electron microscope (FESEM).
The surface of the membrane was sputter-coated (Pelco model SC-7 auto sputter coater) with 5
nm of gold prior to imaging.
Transport Studies. Nanofiltration experiments were performed with a cross-flow apparatus that
was pressurized with Ar at 4.8 bar, as described previously.
solutions contained 1000 ppm Na

+

-

and 1000 ppm Mg
2-

MgCl2 in deionized water. For Cl /SO4

2+

13

+

2+

For Na /Mg

separations, feed

prepared by dissolving NaCl and
2-

separations, feed solutions contained 1000 ppm SO4

-

and 1000 ppm Cl prepared by dissolving Na2SO4 and NaCl in deionized water.

Cation

concentrations were determined using atomic absorption spectroscopy (Varian Spectra Atomic
121

Absorption-200 Spectrometer), and anion concentrations were determined using ion
chromatography (Dionex 600 Ion Chromatograph with an Ionpac AS16 column) with
conductivity detection. All the reported values are averages of several measurements on at least
two different membranes.

4.3. Results and Discussion
4.3.1. Adsorption of Cu

2+

and Fe

3+

This work aims to employ Cu

in PEM Films
2+

or Fe

3+

adsorption in multilayer polyelectrolyte films to

enhance the ion-transport selectivity of membranes capped with these films. Thus, the first step
in this research is to show that adsorption occurs and creates anion-exchange sites that may
increase selectivity. The most direct way to demonstrate ion adsorption is to dissolve the
adsorbed ions for subsequent analysis, and Table 4.1 shows the results of such studies. After
exposure of (PAH/PSS)4PAH films on gold wafers to 0.5 M or 0.005 M FeCl3 and subsequent
2

3+

rinsing with DI water, a 1.2 cm film contains approximately 0.002 µmol of Fe .
ellipsometric thickness of the (PAH/PSS)4PAH films prior to Fe

3+

The

adsorption is about 20 nm, so

3

assuming that the density of the film is 1 g/cm and that PSS accounts for 50% of the adsorbed
-

polyelectrolyte, the film should contain around 0.014 µmol of –SO3 groups. Thus, the ratio of
-

3+

–SO3 to Fe

in the film is 0.1-0.2. However, this number should be viewed with some caution

because the concentration of Fe

3+

eluted from the film is near the detection limit of ICP-OES.

.

122

3+

Table 4.1.

Concentrations of Fe

-

and -SO3 in (PAH/PSS)4PAH films after immersion

of the film in an FeCl3 solution and rinsing.
3+

Fe
(µmol)

Substrate
3+

(PAH/PSS)4.5 - 0.005 M Fe
(PAH/PSS)4.5 - 0.5 M Fe

3+

(Au wafer one sided)

(Au wafer one sided)

-

-SO3

(µmol)

-

a

3+

–SO3 /Fe

0.002

0.014

0.14

0.003

0.014

0.21

a

The –SO3- concentration was calculated from the film ellipsometric thickness using the
assumptions described in the text.

FTIR spectra can demonstrate the presence of anions in polyelectrolyte films and confirm
the formation of anion-exchange sites.
-

24, 25

-

Unfortunately, Cl ions do not absorb IR radiation,

-

-

so we exchanged SCN for Cl to facilitate counterion analysis. However, the SCN may adsorb
-

to the film to a greater extent than Cl because it exhibits significantly stronger ion-exchange
interactions.

24

123

4.5 PAH/PSS

4.5 PAH/PSS - FeCl3
4.5 PAH/PSS –FeCl3 (5 mM)- NaSCN
4.5 PAH/PSS - FeCl3 - NaSCN

0.01

4.5 PAH/PSS - CuCl2

Absorbance

4.5 PAH/PSS - CuCl2 - NaSCN

4.5 PAH/PSS
4.5 PAH/PSS - NaSCN
-

SCN

2150

1950

1750

1550

1350

1150

950

Wavenumbers (cm-1)
Figure 4.2.

Reflectance FTIR spectra of (PAH/PSS)4PAH-coated Au wafers before and
after a 1-h immersion in 0.5 M FeCl3 or 0.5 M CuCl2 or 0.5 M NaSCN
followed by rinsing with water, and the spectra of the CuCl2 or FeCl3treated films, including a film immersed in 5 mM FeCl3, after a 1 h
immersion in 0.5 M NaSCN followed by rinsing. Some of the spectra are
offset for clarity.

Figure 4.2 shows FTIR spectra of (PAH/PSS) 4PAH films before and after exposure to
3+

2+

aqueous FeCl3, CuCl2, or NaSCN solutions and after exposure of Fe - or Cu -treated films to
-1

a solution containing NaSCN. The dominant peaks in the spectrum between 1750 and 950 cm
stem from PSS and PAH,

26, 27

and there is no significant change in the IR spectrum of the

124

multilayer film after exposure to a solution of FeCl3 or CuCl2. After immersion in the NaSCN
3+

2+

solution and rinsing with water, both (PAH/PSS) 4PAH films and Fe - or Cu -treated
-1

-

(PAH/PSS)4PAH films show a new peak at 2058 cm due to SCN adsorption.

However, with
-

immersion in 0.5 M FeCl3 and subsequent exposure to NaSCN, the area of the SCN peak is
about 2.5-fold the area of the corresponding peak for an untreated film that was immersed in 0.5
M NaSCN, suggesting that Fe

3+

adsorption increases the number of ion-exchange sites in the
3+

-

film 2.5-fold. The relatively large SCN peak even without Fe

adsorption probably appears

-

because SCN adsorbs to the terminal polycationic layer of the film. We employed polycationterminated coatings both to avoid Fe
+

increase Na /Mg

2+

-

3+

adsorption at a polyanion-terminated surface and to
3+

selectivities. The FTIR spectra of 5 mM Fe -treated (PAH/PSS)4PAH after
-1

exposure to SCN also shows a peak at 2058 cm , but the peak area is only 1.5 times the area for
the untreated (PAH/PSS)4PAH after immersion in aqueous NaSCN. The corresponding SCN

-

peak area after treatment with 0.5 M CuCl2 and immersion in 0.5 M NaSCN is only 1.3-fold that
2+

for the untreated film, which suggests fewer ion-exchange sites in the Cu -treated films than in
3+

Fe -treated films.
Figure 4.3 shows SEM images of the top face of a bare PES membrane, and PES
membranes coated with (PSS/PAH)4 with and without subsequent treatment with 0.5 M FeCl3
and rinsing. On 4-bilayer PSS/PAH films treated with 0.5 M FeCl3 (Figure 4.3c and 4.3d), the

125

surface contains a number of particles, which may be Fe(OH)3 formed by reaction between
FeCl3 and water (the membranes are stored in water until use in SEM or nanofiltration studies).
No such particles are present on bare PES and PES coated with 4-bilayer PSS/PAH films. To
determine if the formation of Fe(OH)3 occurs during storage, a 4-bilayer PSS/PAH film
immersed in 0.5 M FeCl3 was dried immediately after rinsing and stored briefly in air. The SEM
image of this membrane shows no particles on the surface. Moreover, membranes treated with
0.005 M FeCl3 show no particles on the surface, even after storage in water.

200 µm

500 nm

500 nm

5 µm

Figure 4.3.

SEM images of 50 kDa PES membranes after (a) no treatment, (b)
adsorption of a (PAH/PSS)4 film and (c, d) after (PAH/PSS) 4 adsorption
followed by immersion in 0.5 M FeCl3 and rinsing with water.
126

4.3.2. Effects of Metal Ion Adsorption on the Nanofiltration Properties of (PSS/PAH)n
Films on PES membranes
4.3.2.1. Mixed Salt Solutions
Deposition of PSS/PAH coatings on PES 50 kDa ultrafiltration membranes yields
ultrathin membrane skins on a highly permeable support.

28

Most of our previous studies

examined deposition of polyelectrolyte films on porous alumina,

13, 27, 29-33

but polymeric

supports are less brittle and typically less expensive than ceramic supports. We begin the
polyelectrolyte deposition with a PSS layer that likely adsorbs to the membrane surface through
hydrophobic interactions, and all membranes terminate in PAH to provide a positively charged
+

surface that enhances Na /Mg

2+

selectivity. Upon deposition of a (PSS/PAH)4 film, the pure

water flux through the membrane at a transmembrane pressure drop of 4.8 bar decreases from
3

-2

7.7 m m

-1

day

3

-2

to 0.9 m m

-1

day , showing that the ultrathin film provides the bulk of the

hydraulic resistance in the membrane.
Table 4.2 presents data for nanofiltration of NaCl/MgCl2 solutions in terms of cationic
rejection, R, which is defined in equation 4.1, where Cperm and Cfeed are the solute
concentrations in permeate and feed, respectively. Equation 4.2 defines the selectivity, S, for
transport of solute A over solute B, which is also a function of the rejections for solute A and
solute B.


C perm 
  100%
R  1 

C feed 



------------ (4.1)

127

S

C A, perm CB, feed 100  R
A

C A, feed CB, perm 100  RB

Table 4.2.

------------ (4.2)

Rejections, solution fluxes, and selectivities from nanofiltration experiments
with (PSS/PAH)n films deposited on 50 kDa PES membranes and feed
solutions containing 1000 ppm Na

+

2+ a

and Mg .

The transmembrane

pressure was 4.8 bar.
+

2+

+

2+

Membrane

solution flux,
3 2
m /m -day

Na
rejection, %

Mg
rejection,%

Na / Mg
Selectivity

(PSS/PAH)4

0.51±0.03

-4.2±7.1

93.5±1.7

17.3±5.4

2+b

0.48±0.02

6.9±3.4

96.6±0.3

27.0±2.5

(PSS/PAH)4 - Fe

3+c

0.58±0.04

15.9±2.5

99.1±0.1

93.0±9.5

(PSS/PAH)5

0.42±0.03

-4.9±4.2

95.3±0.1

22.4±1.2

0.62±0.02

4.2±1.9

99.0±0.1

99.4±6.4

(PSS/PAH)4 - Cu

3+c

(PSS/PAH)5 - Fe
a

Solutions were prepared with NaCl and MgCl2.

b
c

The membrane was exposed to 0.5 M CuCl2 and rinsed with deionized water.

The membrane was exposed to 0.5 M FeCl3 and rinsed with deionized water.
For (PSS/PAH)4 films on PES, the Mg

2+

+

2+

rejection and Na /Mg

selectivity values of

94% and 17, respectively, agree reasonably with previous work in our group.
with adsorption of an additional PSS/PAH bilayer, whereas Mg

2+

14

Flux decreases
+

rejection and Na /Mg

selectivity increase slightly for (PSS/PAH) 5 relative to (PSS/PAH)4.

2+

Immersion of the

(PSS/PAH)4 membrane in a 0.5 M CuCl2 solution followed by rinsing with water increases
2+

Mg

+

2+

rejection to 97% and Na /Mg

selectivity to 27. More remarkably, similar treatment
128

with Fe

3+

leads to a 99% Mg

2+

+

rejection and a 5-fold increase in Na /Mg

2+

selectivity.

Moreover, after treatment with the FeCl3 solution, the fluxes through (PSS/PAH)4 and
(PSS/PAH)5 increases by 14 and 48%, respectively, even with the increased rejection.
+

2+

We presume that the increased Na /Mg

selectivity after Fe

3+

adsorption stems from an

increase in the number of ionic exchange sites in the film, which is consistent with the FTIR
spectra discussed above. Binding of Fe

3+

to the films via ionic interactions with sulfonate

groups likely gives rise to the high stability of this modification. Even after 18 h of cross-flow
filtration in a solution containing no Fe
3+

its selectivity. The Fe

3+

+

2+

and 1000 ppm Na and Mg , the membrane retains
+

ions may adsorb to the film through ion exchange with Na according

to equation 4.3, or they may actually disrupt ionic cross-links between sulfonate and ammonium
groups as shown in equation 4.4. Disruption of cross-links might be responsible for the increase
in flux after the FeCl3 treatment.



P-SO3 Na   Fe3+





 Fe3 P-SO  
3 




3 P-SO3 NH3 -P  Fe3+

2

 Na 



Fe3 P-SO3

------------ (4.3)

3  3 P-NH3
+

To provide evidence that the increase in Na /Mg

2+

------------ (4.4)

selectivity stems primarily from an
-

2-

increase in the number of anion-exchange sites in the film, we also examined the Cl /SO4

selectivity in nanofiltration of solutions containing NaCl and Na 2SO4. Table 4.3 shows that the
-

2-

polycation-terminated (PSS/PAH)4 films show minimal Cl /SO4

129

selectivity before and after

3+

treatment with Fe . If adsorption of Fe

3+

were inducing size-based discrimination among ions,

-

2-

we would expect to see an increase in Cl /SO4

selectivity for membranes treated with FeCl3
-

solutions, but this is not the case. In fact the Cl /SO4

2-

selectivity decreases slightly after

3+

adsorption of Fe , which is consistent with an increase in the positive charge in the film.
2-

Adsorption of SO4

might explain the smaller than expected difference in flux between the

3+

Fe -treated and untreated membranes. In fact the flux through these membranes is even higher
3

-2

-1

than the pure water flux of 0.9 m m day .

Table 4.3.

Rejections, solution fluxes, and selectivities from nanofiltration experiments
with (PSS/PAH)4 films deposited on 50 kDa PES membranes and feed
-

2- a

solutions containing 1000 ppm Cl and SO4 .
-

2-

-

2-

Membrane

solution flux,
3 2
m /m -day

Cl
rejection, %

SO4
rejection,%

Cl / SO4
Selectivity

(PSS/PAH)4

1.20±0.05

9.4±1.0

20.2±2.2

1.14±0.02

1.24±0.02

13.1±3.0

20.5±2.7

1.09±0.01

3+b

(PSS/PAH)4 - Fe
a

Nanofiltration solutions were prepared with NaCl and Na2SO4.

b

The membrane was exposed to 0.5 M FeCl3 and rinsed with deionized water.

130

3+

4.3.2.1.1. Effect of immersion time and Fe

concentration on NF properties of membranes

treated with FeCl3.
The NF properties of (PSS/PAH)x films exposed to FeCl3 solutions vary with both the
concentration of Fe

3+

in the solution and the exposure time. As Table 4.4 shows, flux through

(PSS/PAH)4-modified PES membranes is around 30% higher for films exposed to 0.5 M FeCl 3
than for films exposed to 5 mM or 50 mM FeCl3. Only with 0.5 M FeCl3 is the flux higher than
that through an unmodified membrane (see Table 4.4). This suggest that breaking of ionic crosslinks as described by equation 4.4 requires a high Fe

3+

concentration.

34

+

The Na and Mg

2+

rejections also increase with the concentration of FeCl3 in the modification solutions, but
differences in selectivity among the various modifications are not statistically significant because
rejection increases for both ions.

Table 4.4.

Rejections, solution fluxes, and selectivities from nanofiltration experiments
with (PSS/PAH)4 films on 50 kDa PES membranes after treatment of the
membranes by immersion in FeCl3. The feed solutions contained 1000 ppm
+

2+ a

Na and Mg .

+

2+

+

2+

FeCl3 concentration in
treatment solution

Na
rejection, %

Mg
rejection,%

Na / Mg
Selectivity

5 mM

0.43±0.02

-6.9±7.8

98.3±0.4

78.9±17.3

50 mM

0.45±0.03

-5.7±3.0

98.8±0.1

86.4±3.7

0.5 M
a

solution flux,
3 2
m /m -day

0.58±0.04

15.9±2.5

99.1±0.1

93.0±9.5

Nanofiltration solutions were prepared with NaCl and MgCl2.
131

Table 4.5.

Rejections, solution fluxes, and selectivities from nanofiltration experiments
with (PSS/PAH)4 films on 50 kDa PES membranes with and without
treatment of the membranes by immersion in 5 mM FeCl3 for different
+

2+ a

times. The feed solutions contained 1000 ppm Na and Mg .
+

2+

+

2+

Doping time
(min)

Na
rejection, %

Mg
rejection,%

Na / Mg
Selectivity

0

0.51±0.03

-4.2±7.1

93.5±1.7

17.3±5.4

5

0.50±0.03

-4.5±3.0

94.4±0.3

18.8±0.8

10

0.42±0.02

-5.3±1.8

99.0±0.1

109.4±10.1

30

0.43±0.02

-5.8±2.1

98.6±0.1

75.8±2.5

60
a

solution flux,
3 2
m /m -day

0.43±0.02

-6.9±7.8

98.3±0.4

78.9±17.3

Nanofiltration solutions were prepared with NaCl and MgCl2.
We investigated the effect of exposure time on membrane modification using the lowest

FeCl3 concentration, where the rate of membrane modification should be lowest. As Table 4.5
shows, full modification of the membrane requires about 10 min, and further exposure to 5 mM
FeCl3 may even lead to small decreases in selectivity and Mg
effect of adsorption of Fe

3+

2+

rejection. We also studied the

on separation properties using a commercial NF-270 membrane

(Table 4.6). The rejections and selectivity of the unmodified membrane are similar to those
reported in literature,
+

14, 35

2+

rejection and Na /Mg

but the FeCl3 treatment leads to only modest increases in Mg

selectivity.

132

2+

Table 4.6.

Rejections, solution fluxes, and selectivities from nanofiltration experiments
with NF-270 membranes before and after immersion in 0.5 M FeCl3 for 60
+

2+ a

min. The feed solutions contained 1000 ppm Na and Mg .
Membrane

2+

+

2+

3+

NF-270 - Fe

rejection, %

rejection, %

Na / Mg
Selectivity

0.92±0.04

NF-270

a

+

solution flux,
3 2
m /m -day

14.4±2.2

48.5±2.7

1.8±0.1

0.79±0.08

10.9±3.0

59.9±6.6

2.4±0.4

Na

Mg
b

b

Solutions were prepared with NaCl and MgCl2.

b

Solutions were analyzed using ICP-OES.

4.3.2.2. Single Salt Experiments
To better understand the higher selectivities and fluxes for the modified membranes, we
+

performed nanofiltration of solutions containing single salts (Table 4.7). Na

and Mg

2+

rejections and fluxes for (PSS/PAH)4 films are similar to results reported by Ouyang et al.

14

Solution flux through the (PSS/PAH)4 membrane is less for MgCl2 than for NaCl because of the
higher MgCl2 rejection and the higher overall concentration of ions in the MgCl2 solutions.
After treatment with 5 mM or 0.5 M FeCl3, sodium rejection increases from 40% to 50-55 %
whereas magnesium rejection increases from 95% to 99 %. Thus the salt passage decreases
more than 5-fold for MgCl2 and at most 25% for NaCl, which is consistent with exclusion of
2+

Mg

due to increased fixed positive charges in the film. The flux of the NaCl solution is higher
3+

for the Fe -doped films compared to the original PSS/PAH film, again suggesting the
3+

possibility of breaking the ionic cross-links in polymer chains at high concentrations of Fe .
133

3+

The flux of MgCl2 solutions does not change significantly after Fe
of Mg

2+

treatment, perhaps because

adsorption in films.

Table 4.7.

Rejections and solution fluxes from single-salt nanofiltration experiments
3+

with (PSS/PAH)4 and Fe -modified (60 min) (PSS/PAH)4 films deposited on
+

50 kDa PES membranes. The feed solutions contained 1000 ppm Na or
2+ a

1000 ppm Mg .

+

Na
Membrane

Mg

solution flux, rejection, %
3 2
m /m -day

2+

solution flux,
3 2
m /m -day

rejection,%

0.78±0.03

40.3±2.1

0.41±0.02

95.0±1.2

3+

1.02±0.02

55.0±1.7

0.44±0.02

99.1±0.1

3+

0.96±0.04

48.6±1.7

0.42±0.03

99.4±0.3

(PSS/PAH)4
(PSS/PAH)4 - 0.5 M Fe
(PSS/PAH)4 - 5 mM Fe
a

Solutions were prepared with NaCl or MgCl2.

4.4. Conclusions
Doping PEMs through exposure to solutions containing FeCl3 or CuCl2 introduces ionexchange sites into the film. Immersion of a (PSS/PAH)4 film on a 50 kDa PES membrane in as
+

2+

little as 5 mM FeCl3 yields a 5-fold increase in the Na /Mg

selectivity of the membrane with

a minimal decrease in permeability compared to untreated (PSS/PAH)4 films. Remarkably, the
membranes maintain their modification for at least 18 h of cross-flow nanofiltration with
3+

solutions that do not contain Fe . Increases in the density of anion-exchange sites in the

134

membrane give rise to exclusion of divalent Mg

2+

+

for an ion-exchange membrane. The high Na /Mg

2-

but not divalent SO4 , as would be expected
2+

selectivities and low operating pressures of

these membranes make them attractive for water softening applications.

135

References

136

4.5. References
1.

Anim-Mensah, A. R.; Krantz, W. B.; Govind, R., Eur. Polym. J. 2008, 44(7), 2244-2252.

2.

Bodzek, M.; Koter, S.; Wesolowska, K., Desalination 2002, 145(1-3), 321-327.

3.

Nanda, D.; Tung, K.-L.; Hsiung, C.-C.; Chuang, C.-J.; Ruaan, R.-C.; Chiang, Y.-C.;
Chen, C.-S.; Wu, T.-H., Desalination 2008, 234(1-3), 344-353.

4.

Yildiz, E.; Nuhoglu, A.; Keskinler, B.; Akay, G.; Farizoglu, B., Desalination 2003,
159(2), 139-152.

5.

Nanda, D.; Tung, K.-L.; Li, Y.-L.; Lin, N.-J.; Chuang, C.-J., J. Membr. Sci. 2010, 349(12), 411-420.

6.

Low, S. C.; Liping, C.; Hee, L. S., Desalination 2008, 221(1-3), 168-173.

7.

Rahimpour, A.; Jahanshahi, M.; Mortazavian, N.; Madaeni, S. S.; Mansourpanah, Y.,
Appl. Surf. Sci. 2010, 256(6), 1657-1663.

8.

Schäfer, A.; Fane, A. G.; Waite, T. D., Water Treatment. In Nanofiltration: Principles
and Applications 1st ed.; Elsevier Ltd: Oxford, UK, 2005.

9.

Krasemann, L.; Tieke, B., Langmuir 2000, 16(2), 287-290.

10.

Jin, W.; Toutianoush, A.; Tieke, B., Langmuir 2003, 19(7), 2550-2553.

11.

Toutianoush, A.; Jin, W.; Deligöz, H.; Tieke, B., Appl. Surf. Sci. 2005, 246(4), 437-443.

12.

Hoffmann, K.; Tieke, B., J. Membr. Sci. 2009, 341(1-2), 261-267.

13.

Stanton, B. W.; Harris, J. J.; Miller, M. D.; Bruening, M. L., Langmuir 2003, 19(17),
7038-7042.

14.

Ouyang, L.; Malaisamy, R.; Bruening, M. L., J. Membr. Sci. 2008, 310(1-2), 76-84.

15.

Farhat, T. R.; Schlenoff, J. B., Langmuir 2001, 17(4), 1184-1192.

16.

Balachandra, A. M.; Dai, J.; Bruening, M. L., Macromolecules 2002, 35(8), 3171-3178.

17.

Toutianoush, A.; Tieke, B., Mater. Sci. Eng., C 2002, 22(2), 135-139.

18.

Xiao, S.; Wu, S.; Shen, M.; Guo, R.; Huang, Q.; Wang, S.; Shi, X., ACS Appl. Mater.
Interfaces 2009, 1(12), 2848-2855.

137

19.

Wang, T. C.; Rubner, M. F.; Cohen, R. E., Langmuir 2002, 18(8), 3370-3375.

20.

Schuetz, P.; Caruso, F., Chem. Mater. 2004, 16(16), 3066-3073.

21.

Miller, M. D.; Bruening, M. L., Langmuir 2004, 20(26), 11545-11551.

22.

Miller, M. D.; Bruening, M. L., Chem. Mater. 2005, 17(21), 5375-5381.

23.

Caruso, F.; Niikura, K.; Furlong, D. N.; Okahata, Y., Langmuir 1997, 13(13), 3422-3426.

24.

Ball, V.; Voegel, J.-C.; Schaaf, P., Langmuir 2005, 21(9), 4129-4137.

25.

Farhat, T. R.; Schlenoff, J. B., J. Am. Chem. Soc. 2003, 125(15), 4627-4636.

26.

Caruso, F.; Furlong, D. N.; Ariga, K.; Ichinose, I.; Kunitake, T., Langmuir 1998, 14(16),
4559-4565.

27.

Stair, J. L.; Harris, J. J.; Bruening, M. L., Chem. Mater. 2001, 13(8), 2641-2648.

28.

Malaisamy, R.; Bruening, M. L., Langmuir 2005, 21(23), 10587-10592.

29.

Sullivan, D. M.; Bruening, M. L., J. Am. Chem. Soc. 2001, 123(47), 11805-11806.

30.

Sullivan, D. M.; Bruening, M. L., Chem. Mater. 2003, 15(1), 281-287.

31.

Sullivan, D. M.; Bruening, M. L., J. Membr. Sci. 2005, 248(1-2), 161-170.

32.

Dai, J.; Baker, G. L.; Bruening, M. L., Anal. Chem. 2005, 78(1), 135-140.

33.

Liu, X.; Bruening, M. L., Chem. Mater. 2004, 16(2), 351-357.

34.

Adusumilli, M.; Bruening, M. L., Langmuir 2009, 25(13), 7478-7485.

35.

Al-Zoubi, H.; Omar, W., Korean J. Chem. Eng. 2009, 26(3), 799-805.

138

Chapter 5
Polyelectrolyte Multilayer Films for Removal of Pharmaceutically
Active Compounds from Water
5.1. Introduction
Pharmaceutically active compounds (PhACs) and endocrine disrupting compounds
(EDCs) in environmental waters are an increasing problem for animals and aquatic life, although
the effect of these chemicals on humans is still debatable.
such as biological methods,
(NF)

16

8-10

chlorination,

11-13

1-7

Many water-treatment techniques

reverse osmosis

14, 15

and nanofiltration

can modify or isolate EDCs and PhACs in water. Among these treatment options, NF is

attractive because it removes a wide range of dissolved organic pollutants from water using
relatively low pressures.

17-21

Unfortunately, however, the small-molecule rejections of commercial NF membranes are
often too low for effective treatment of streams containing EDCs and PhACs. Kimura et al.
reported 44% rejection of caffeine by a commercial SC-3100 NF membrane (MWCO 200–
300 Da).

15

Yangali-Quintanilla et al. studied the rejections of acetaminophen (ACT) and

caffeine (CFN) using NF-90 membranes from DOW and found a steady state rejection of 71%
for ACT and 81% for CFN.

22

Yoon et al. performed dead-end, stirred-cell NF using a pressure

of 7.5 bar and obtained ACT and CFN rejections less than 40% for a NF membrane from ESNA,
Hydranautics, USA.

23

Zazouli et al. studied the influence of solution pH, ionic strength and

139

transmembrane pressure on ACT rejection. Rejections of ACT with SR3 membranes from
Koch, USA increased with the solution pH value because deprotonation of the phenolic group on
ACT created a negatively charged species that was more highly rejected by the negatively
charged membrane. However, the maximum ACT rejection was only 60%.

24

This research examines the rejections of ACT and CFN by NF membranes prepared
through layer-by-layer (LbL) adsorption of multilayer polyelectrolyte films on permeable
substrates. For comparison, we also study the ACT and CFN rejections of two commercial NF
membranes, NF-270 and NF-90. The polyelectrolyte films are attractive because they allow
variation of the thickness of the membrane skin through changing the number of deposited
layers, and control over the skin thickness permits optimization of the tradeoff between rejection
and membrane permeability. Moreover, at high pH, electrostatic repulsions between ACT and
terminal polyanion layers in polyelectrolyte membranes should lead to very high rejections. To
the best of our knowledge, this is the first work to use LbL films to remove PhACs from water.
However, Hoshi et al. showed that PAH/PSS films block the access of ACT to electrodes.

25

5.2. Experimental Section
Materials.

Poly(sodium 4-styrenesulfonate) (PSS, Mw = 70 000 Da), poly(allyl amine

hydrochloride) (PAH, Mw = 56 000 Da), poly(acrylic acid) (PAA, Mw = 90 000 Da), and reagent
grade ACT and CFN were purchased from Sigma-Aldrich. Figure 5.1 shows the structures of
the PhACs, and Table 5.1 provides a summary of their physical properties. Porous alumina
supports (0.02-μm Whatman Anodisk filters) were UV/O3 cleaned (Boekel UV-Clean Model
135500) with the feed side up for 15 min before film deposition, and deionized water (Millipore
140

purification system, 18.2 MΩcm) was used for membrane rinsing and preparation of the
polyelectrolyte solutions. Commercial NF-90 and NF-270 membranes were soaked in deionized
water for 1 hour. These membranes were then rinsed with water and used for NF experiments or
modified with polyelectrolyte multilayers.

Figure 5.1.

Structures of acetaminophen and caffeine.

Preparation of PEM Films. Polyelectrolyte adsorption followed a literature procedure.

26, 27

Briefly, deposition of PSS occurred by exposing the top of the substrate to 0.01 M PSS in 0.5 M
NaCl for 2 min and rinsing for 1 min with deionized water. (Concentrations of polyelectrolytes
are always given with respect to the repeating unit). The polycation layer was adsorbed during
exposure of the substrate to 0.01 M PAH in 0.5 M NaCl for 5 min and rinsing with water for 1
min.

The pH values of PSS and PAH deposition solutions were adjusted to 2.1 and 2.3,

respectively, using 0.1 M HCl. Polyanion and polycation adsorption steps were repeated to give
the desired number of polyelectrolyte layers, and the membranes were stored in water until use.
Transport Studies. Nanofiltration experiments were performed with a cross-flow apparatus that
was pressurized with Ar at 4.8 bar, as described previously.

141

28

Feed solutions contained 100 ppm

ACT or 100 ppm CFN, prepared by dissolving 0.1 g in 1 liter deionized water (Milli-Q, 18.2
MΩcm). Permeate and feed concentrations of ACT and CFN were determined using UV/Vis
spectroscopy (Perkin Elmer Lambda-40 Spectrometer) and a calibration curve. All the reported
values are average of several measurements on at least two different membranes.

Table 5.1.

Physical and chemical properties of acetaminophen and caffeine.

Compound

Molecular
Weight

Water Solubility
g/L

ACT

151

4.15

CFN

194

11.0

29
30

pKa at 20°C
9.4

29

10.4

30

Charge at
pH 6.4
Neutral
Neutral

5.3. Results and Discussion
5.3.1. Nanofiltration of ACT with NF-270, NF-90 and PSS/PAH membranes
We employed ACT as the primary compound for evaluating the PhAC rejection
properties of polyelectrolyte and commercial NF membranes. Figure 5.2 shows the UV/Vis
spectrum of aqueous 20 ppm acetaminophen, and similar spectra of permeate solution allow
rapid determination of ACT rejections from the absorbance at λmax, 243 nm. Experimental
results are expressed in terms of the rejection percentage, R, which is calculated by comparing
the permeate and feed concentration using equation 5.1,


C pi 
  100%
R  1 

C fi 



------------ (5.1)

where Cfi and Cpi are the concentrations of compound i in the feed and the permeate,
respectively. Solution flux is the volume flowing through the membrane per unit area and time.
142

3

Absorbance

2.5
2
1.5
1
0.5
0

200

225

250

275

300

325

Wavelength (nm)

Figure 5.2.

UV/Vis spectrum of 20 ppm ACT in water. λmax=243 nm.

Table 5.2 shows the rejections of ACT by different membranes. The NF-270 membrane
exhibits rejections <20% because of its relatively large pores.

In contrast, NF-90 and

(PSS/PAH)7 on porous alumina have similar rejections of around 65%, but the solution flux is
24 to 32% lower than that through NF-270.

PSS/PAH films terminated with PSS,

(PSS/PAH)7PSS, exhibit lower rejections than PAH-terminated films. The feed solution has a
pH of 6.4, which is lower than the pKa value of ACT, so this compound should be neutral during
filtration and unaffected by the charge on the membrane. The slight decrease in rejection on
going from (PSS/PAH)7 to (PSS/PAH)7PSS might stem from increased swelling for the PSSterminated film,

27, 31, 32

but this explanation is inconsistent with the trends in fluxes. We also

143

studied NF with (PSS/PAH)9 membranes and observed a 22% decrease in solution flux and no
increase in rejections relative to (PSS/PAH) 7 films (data not shown). Evidently, 7-bilayer films
are sufficient to achieve full coverage of the alumina support and addition of more layers only
decreases permeability.

Table 5.2.

Rejections and solution fluxes from nanofiltration experiments with NF-270
and NF-90 membranes, and PSS/PAH films deposited on porous alumina.
The feed solutions contained 100 ppm ACT.
solution flux,
3 2
m /m -day

NF-270
NF-90
(PSS/PAH) on alumina
7

(PSS/PAH) PSS on alumina
a

a

7

15.9±0.6
64.9±3.4
65.6±0.9

1.21±0.03

a

rejection, %

1.85±0.12
1.26±0.08
1.40±0.20

Membrane

54.4±3.3

Deposition occurred from solutions containing 0.02 M PSS and 0.02 M PAH.

In an effort to achieve higher rejections, we modified commercial membranes with
polyelectrolyte films (Table 5.3). Adsorption of (PSS/PAH) 3 on NF-270 increases rejection to
only 27%, perhaps because of the relatively large pores in the base membrane. However,
adsorption of a single PSS/PAH bilayer on NF-90 membranes increases rejection to 81% while
3

2

maintaining a solution flux 1.09 m /(m day). Even adsorption of a single PSS layer enhances
ACT rejection from 65 to 73%. On NF-90 membranes, rejection of ACT reaches a maximum of
81% with adsorption of 1 or 2 PSS/PAH layers and then decreases after deposition of a third
bilayer. Prior studies showed that the structure of PDADMAC/PSS

33, 34

films on alumina

substrates varies dramatically with the number of layers in the film, which leads to an optimal
144

number of bilayers for salt rejection and selectivity. Nevertheless, such behavior is unusual for
PSS/PAH films.

Table 5.3.

Rejections and solution fluxes from nanofiltration experiments with NF-270
and NF-90 membranes modified with polyelectrolyte coatings.

The feed

solutions contained 100 ppm ACT.
Membrane

solution flux,
3 2
m /m -day

rejection, %

PSS on NF-90

1.11±0.01

73.7±1.9

(PSS/PAH) on NF-90

1.09±0.03

80.7±1.0

(PSS/PAH) on NF-90

1.07±0.07

80.7±1.5

(PSS/PAH) on NF-90

1.02±0.01

73.8±2.4

(PSS/PAH) on NF-270

1.41±0.04

26.8±3.8

1
2
3

3

5.3.1.1. Effect of polyelectrolyte type and concentration on the rejection of ACT
The thickness of PEM films often increases with the concentration of polyelectrolytes in
deposition solutions. Instead of increasing the thickness through adsorption of additional layers,
we varied the polyelectrolyte concentrations in deposition solutions from 0.01 to 0.02 M. Such
concentration changes can lead to a 10 - 20% increase in PEM thickness.

35

However, on both

NF-270 and NF-90 membranes (Table 5.4), the use of the higher polyelectrolyte concentrations
led to decreases in flux and no significant change in rejection.

145

Table 5.4.

Rejections and solution fluxes from nanofiltration experiments with NF-270
and NF-90 membranes modified with PSS/PAH films deposited from
solutions containing different polyelectrolyte concentrations.

The feed

solutions contained 100 ppm of ACT.
Membrane

solution flux,
3 2
m /m -day

rejection, %

(PSS/PAH) (0.01 M deposition) on NF-90

1.07±0.07

80.7±1.5

(PSS/PAH) (0.02 M deposition) on NF-90

1.00±0.04

78.4±1.4

(PSS/PAH) (0.01 M deposition) on NF-270

1.41±0.04

26.1±3.8

(PSS/PAH) (0.02 M deposition) on NF-270

1.34±0.03

26.2±1.5

2
2

3
3

We also investigated whether varying the type of polyelectrolyte could increase the
rejection by modified membranes and found that deposition of (PAA/PAH) 1 on NF-90 leads to
an ACT rejection of 85% (Table 5.5). This small increase in rejection relative to modification
with (PSS/PAH) n may occur because the higher charge density in PAA than PSS leads to a more
cross-linked film.

Table 5.5.

27, 28

Rejections and solution fluxes from NF experiments with PSS/PAH and
PAA/PAH films deposited on NF-90 membranes.

The feed solutions

contained 100 ppm of ACT, and the transmembrane pressure was 4.8 bar.
3

2

Membrane

solution flux, m /m -day

rejection, %

(PSS/PAH) on NF-90

1.09±0.03

80.7±1.0

(PAA/PAH) on NF-90

1.08±0.01

84.8±1.3

1

1

146

5.3.2. Nanofiltration of CFN with NF-270, NF-90 and PSS/PAH films

1.2

Absorbance

1
0.8
0.6
0.4
0.2
0
210

235

260

285

310

Wavelength (nm)
Figure 5.3.

UV/Vis spectrum of 10 ppm CFN in water. λ max=272 nm.

Figure 5.3 shows the UV/Vis spectrum of CFN, and the clear λmax at 272 nm readily
allows determination of CFN concentrations using a calibration curve. Table 5.6 shows the
rejections of CFN using different membranes. NF-270 shows a relatively low 49% rejection of
CFN because of the large pores in this membrane. In contrast, NF-90 shows a CFN rejection of
78%, whereas (PSS/PAH)4 films show a similar rejection but a flux that is twice that through
NF-90. Upon addition of a terminating layer of PSS to create (PSS/PAH) 4PSS films, rejection
of CFN decreases to 52% and flux increases. Both of these changes are consistent with the
trends in ACT rejections in Table 5.2, and an increase in film swelling when terminating films
147

with PSS.

27, 32

Remarkably, (PSS/PAH)7 films exhibit a CFN rejection of 94% along with a

solution flux greater than that through NF90.

Thus, the polyelectrolyte membranes show

significantly higher rejections or fluxes than the commercial NF membranes.

The higher

rejections of CFN than ACT most likely stem from the slightly higher molecular weight (Table
5.1) or size of CFN.

Table 5.6.

Rejections and solution fluxes from nanofiltration experiments with NF-270
and NF-90 membranes and PSS/PAH films deposited on porous alumina.
The feed solutions contained 100 ppm CFN.

Membrane

solution flux,
3 2
m /m -day

rejection, %

NF-270

1.76±0.22

48.5±4.8

NF-90

0.94±0.02

77.7±4.6

(PSS/PAH) on alumina

2.14±0.02

79.2±1.4

(PSS/PAH) PSS on alumina

2.95±0.04

52±6

(PSS/PAH) on alumina

1.33±0.02

93.6±0.6

4

4

7

5.4. Conclusions
(PSS/PAH)7 membranes show ACT rejections similar to commercial NF-90 membranes,
but the CFN rejection by (PSS/PAH) 7 films is 94% whereas that for NF-90 membrane is only
78%. Thus, caffeine passage is 3.5-fold greater through the NF-90 membrane than through
(PSS/PAH)7. Moreover, the solution permeabilities of (PSS/PAH) 7 and NF-270 membranes are
almost the same. Modification of NF-90 membranes with 1 bilayer of PAA/PAH increases the

148

ACT rejection from 65% to 85% with only a 14% decrease in flux. PSS/PAH adsorption on NF90 also increases ACT rejection, but not to the same extent as addition of PAA/PAH. The high
rejections and permeabilities of PSS/PAH films on porous alumina and on NF-90 membranes
make polyelectrolyte adsorption an attractive option for creating nanofiltration membranes that
reject PhACs.

149

References

150

5.5. References
1.

Purdom, C. E.; Hardiman, P. A.; Bye, V. V. J.; Eno, N. C.; Tyler, C. R.; Sumpter, J. P.,
Chem. Ecol. 1994, 8(4), 275 - 285.

2.

Kubo, K.; Arai, O.; Omura, M.; Watanabe, R.; Ogata, R.; Aou, S., Neurosci. Res. 2003,
45(3), 345-356.

3.

Han, G. H.; Hur, H. G.; Kim, S. D., Environ. Toxicol. Chem. 2006, 25(1), 265-271.

4.

Schwab, B. W.; Hayes, E. P.; Fiori, J. M.; Mastrocco, F. J.; Roden, N. M.; Cragin, D.;
Meyerhoff, R. D.; D'Aco, V. J.; Anderson, P. D., Regul. Toxicol. Pharmacol. 2005,
42(3), 296-312.

5.

Safe, S., Toxicology 2004, 205(1-2), 3-10.

6.

Rudel, R. A.; Camann, D. E.; Spengler, J. D.; Korn, L. R.; Brody, J. G., Environ. Sci.
Technol. 2003, 37(20), 4543-4553.

7.

Tyler, C. R.; Jobling, S.; Sumpter, J. P., Crit. Rev. Toxicol. 1998, 28(4), 319-361.

8.

Balest, L.; Mascolo, G.; Di Iaconi, C.; Lopez, A., Water Sci. Technol. 2008, 58(4), 953956.

9.

Bevilaqua, J. V., Cammarota, M.C., Freire, D.M.G., Sant'Anna Jr., G.L., Brazilian
Journal of Chemical Engineering 2002, 19(2), 151-158.

10.

Yamada, K.; Tamura, T.; Azaki, Y.; Kashiwada, A.; Hata, Y.; Higashida, K.; Nakamura,
Y., J. Polym. Environ. 2009, 17(2), 95-102.

11.

Hu, J.; Cheng, S.; Aizawa, T.; Terao, Y.; Kunikane, S., Environ. Sci. Technol. 2003,
37(24), 5665-5670.

12.

Moriyama, K.; Matsufuji, H.; Chino, M.; Takeda, M., Chemosphere 2004, 55(6), 839847.

13.

Boyd, G. R.; Zhang, S.; Grimm, D. A., Water Res. 2005, 39(4), 668-676.

14.

Bódalo, A.; Gómez, J. L.; Gómez, M.; León, G.; Hidalgo, A. M.; Ruíz, M. A.,
Desalination 2008, 223, 323–329.

15.

Kimura, K.; Toshima, S.; Amy, G.; Watanabe, Y., J. Membr. Sci. 2004, 245(1-2), 71-78.

16.

Wintgens, T.; Gallenkemper, M.; Melin, T., Desalination 2002, 146(1-3), 387-391.

151

17.

Arsuaga, J. M.; López-Muñoz, M. J.; Aguado, J.; Sotto, A., Desalination 2008, 221(1-3),
253-258.

18.

Lee, S.; Lee, C.-H., Sep. Purif. Technol. 2007, 56(1), 1-8.

19.

Agenson, K. O.; Oh, J.-I.; Urase, T., J. Membr. Sci. 2003, 225(1-2), 91-103.

20.

Van der Bruggen, B.; Vandecasteele, C., Environ. Pollut. 2003, 122(3), 435-445.

21.

Kunst, B. Košutić, K., Removal of Emerging Contaminants in Water Treatment by
Nanofiltration and Reverse Osmosis. In Emerging Contaminants from Industrial and
Municipal Waste, Barceló, D.; Petrovic, M., Eds. Springer Berlin / Heidelberg: 2008;
Vol. 5S, pp 103-125.

22.

Yangali-Quintanilla, V.; Sadmani, A.; McConville, M.; Kennedy, M.; Amy, G., Water
Res. 2009, 43(9), 2349-2362.

23.

Yoon, Y.; Westerhoff, P.; Snyder, S. A.; Wert, E. C.; Yoon, J., Desalination 2007, 202(13), 16-23.

24.

Zazouli, M. A.; Susanto, H.; Nasseri, S.; Ulbricht, M., Water Res. 2009, 43(13), 32703280.

25.

Hoshi, T.; Saiki, H.; Kuwazawa, S.; Tsuchiya, C.; Chen, Q.; Anzai, J.-i., Anal. Chem.
2001, 73(21), 5310-5315.

26.

Miller, M. D.; Bruening, M. L., Langmuir 2004, 20(26), 11545-11551.

27.

Miller, M. D.; Bruening, M. L., Chem. Mater. 2005, 17(21), 5375-5381.

28.

Stanton, B. W.; Harris, J. J.; Miller, M. D.; Bruening, M. L., Langmuir 2003, 19(17),
7038-7042.

29.

http://www.drugbank.ca/drugs/DB00316 (07/30/2010).

30.

http://www.drugbank.ca/drugs/DB00201 (7/30/2010).

31.

Carrière, D.; Krastev, R.; Schönhoff, M., Langmuir 2004, 20(26), 11465-11472.

32.

Wong, J. E.; Rehfeldt, F.; Hanni, P.; Tanaka, M.; Klitzing, R. v., Macromolecules 2004,
37(19), 7285-7289.

33.

Hong, S. U.; Malaisamy, R.; Bruening, M. L., Langmuir 2007, 23(4), 1716-1722.

34.

Adusumilli, M.; Bruening, M. L., Langmuir 2009, 25(13), 7478-7485.

152

35.

Wang, Y.; Stedronsky, E.; Regen, S. L., J. Am. Chem. Soc. 2008, 130(49), 16510-16511.

153

Chapter 6
Conclusions and Future Work

Layer by layer adsorption of polyelectrolytes from aqueous solutions is an extremely
versatile method for fabricating thin films.

This technique is attractive for a range of

applications due to its fine control over film thickness, applicability to most substrates and
substrate geometries, and ability to incorporate a wide range of functional polyelectrolytes.
Chapter 1 reviews selected potential applications of PEMs and the factors that influence PEM
properties.
Chapter 3 demonstrates that the properties of PDADMAC/PSS coatings vary greatly with
the number of adsorbed layers in the film. Streaming potential measurements and ATR-FTIR
data suggest the presence of high anion-exchange capacities only in films with more than 7
bilayers, and the presence of these anion-exchange sites dramatically decreases the selectivity of
-

2-

Cl /SO4

separations. The understanding of film chemistry and structure provided by ATR-

FTIR spectroscopy and streaming potential measurements is vital for developing PEM
membranes to effectively separate anions or cations.
Chapter 4 describes a simple way to introduce fixed positive charges in PEM films by
exposing them to 5 mM FeCl3 for as little as 10 min. Treatment of (PSS/PAH)4 films (on 50
+

2+

kDa PES membranes) with FeCl3 yields a 5-fold increase in the Na /Mg

selectivity of the

membrane with a minimal decrease in permeability compared to untreated (PSS/PAH) 4 films.
Remarkably, the membranes maintain their modification for at least 18 h of nanofiltration with

154

3+

+

solutions that do not contain Fe . The high Na /Mg

2+

selectivities and low operating pressures

of these membranes make them attractive for water softening applications.
Chapter 5 examines the rejections of acetaminophen (ACT) and Caffeine (CFN) by PEM
films on porous substrates and by the commercial NF membranes, NF-270 and NF-90.
Modification of NF-90 with a single PAA/PAH bilayer yields an ACT rejection of 85%, whereas
the unmodified NF-90 membrane exhibits a similar flux and an ACT rejection of only 65%. The
best membrane in this study shows a CFN rejection of 94% and a solution flux of 1.33
3

2

m /(m day) at an operating pressure of only 4.8 bar. The ease of preparing these membranes
makes them very attractive for removal of pharmaceutically active compounds (PhACs) from
water.
This study examined the rejection properties of PEM films using PhACs, but the PEMs
may also be effective for removal of endocrine disrupting compounds (EDCs) from water.
Conventional NF can isolate EDCs like alkyl phenols (AP) from aqueous solutions, but the
removal of lower molecular weight APs is not always effective. López-Muñoz et al. found that
NF-270 rejected less than 10% of selected phenolic compounds, and NF-90 exhibited rejections
ranging from 10-50% depending on the compound studied.

1

HR98PP (Dow) membrane could reject up to 80% of phenol.

Bódalo et al. found that a DSS2

Jung et al. used NF membranes

to study the rejections of APs with alkyl groups ranging from C2 to C9 and found that a
poly(vinyl alcohol)/polyamide based membrane (NTR-729HF, Nitto Denko Co. Ltd, Japan)
rejected 4-n-octylphenol and 4-n-nonylphenol more than 95%, but the rejections of 4ethylphenol (4-EP) and 4-tert-butylphenol (4-BP) were lower than 82%.

3

Modification of these

membranes with polyelectrolyte multilayers may greatly enhance the rejections of APs.
155

Future work could also examine the modification of RO membranes with
polyelectrolytes. Due to the relatively low permeability of RO membranes, the polyelectrolyte
layer would not greatly decrease the flux through this system, and coverage of small defects by
the polyelectrolyte film could greatly enhance NaCl rejections. Treatment of the polyelectrolyte
layers with Fe

3+

may further enhance salt rejection. Finally, removal and regeneration of the

polyelectrolytes provides a potential way to combat membrane fouling.

4

Hopefully this work

only represents the beginning of efforts to improve membrane properties through polyelectrolyte
adsorption.

156

References

157

6.1. References

1.

López-Muñoz, M. J.; Sotto, A.; Arsuaga, J. M.; Van der Bruggen, B., Sep. Purif. Technol.
2009, 66(1), 194-201.

2.

Bódalo, A.; Gómez, E.; Hidalgo, A. M.; Gómez, M.; Murcia, M. D.; López, I.,
Desalination 2009, 245(1-3), 680-686.

3.

Jung, Y. J.; Kiso, Y.; Park, H. J.; Nishioka, K.; Min, K. S., Desalination 2007, 202(1-3),
278-285.

4.

Shan, W.; Bacchin, P.; Aimar, P.; Bruening, M. L.; Tarabara, V. V., J. Membr. Sci. 2010,
349(1-2), 268-278.

158