.: 1 . .11.)... 5.. 5
f? . . . , I
q...

30......-

.1.

Wu u.¢~...- ¢ .5.-.

.
1.
A
u
.
1;

 

3%. at 133% an? , ...,.w_.§.s...=.§ .1..

 

 

This is to certify that the

dissertation entitled

Applications of Multilayer Polyelectrolyte
Films in Corrosion Inhibition, Ion
Separation, and Catalysis

presented by

Jinhua Dai

has been accepted towards fulﬁllment
of the requirements for

Ph.D. degreein Chemistry

M

Major professor U

 

 

Date [Me/oz

MS U i: an Afﬁrmative Action/Equal Opportunity Institution 0-12771

 

 

 

LIBRARY

MiChigan State
University

 

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

FEB 1 2 2005

 

Whig
“a 1 4 2339

 

 

163969

 

 

 

 

 

 

 

 

 

 

 

6/01 c:/ClRC/DateDue.p65-p. 15

APPLICATIONS OF MULTILAYER POLYELECTROLYTE FILMS IN
CORROSION INHIBITION, ION SEPARATION, AND CATALYSIS

By

J inhua Dai

A DISSATATION
Submitted to
Michigan State University
In partial fulﬁllment of the requirements
For the degree of

DOCTOR OF PHILOSOPHY
Department of Chemistry

2002

ABSTRACT

APPLICATIONS OF MULTILAYER POLYELECTROLYTE FILMS IN CORROSION
INHIBITION, ION SEPARATION, AND CATALYSIS

By

Jinhua Dai

Alternating adsorption of polycations and polyanions provides a very convenient
means of depositing ultrathin ﬁlms with controlled thicknesses and compositions. This
dissertation investigates the application of these ﬁlms in areas ranging from surface
protection to encapsulation of nanoparticle catalysts. In corrosion reduction, for example,
deposition of multilayered poly(acrylic acid)/poly(allylamine hydrochloride) (PAA/PAH)
ﬁlms on Al yields ultrathin coatings that protect Al from Cl'-induced corrosion. Although
the resistance of these ﬁlms is minimal, they form a passivating layer on the surface
oxide and increase oxide resistance by 2-3 orders of magnitude. Heating of PAA/PAH
coatings produces cross-linked polyamides that have ﬁlm resistances of about 5 M0 cmz,
even when the ﬁlm is only lO-nm thick. Although the resistance of these ﬁlms is high,
decreases in corrosion current due to cross-linked ﬁlms are still result primarily from
passivation of the surface oxide, as oxide resistance is larger than ﬁlm resistance.

Partial Fischer esteriﬁcation of PAA allows tailoring of the hydrophobicity and charge
density in multilayered PAH/derivatized PAA (d-PAA) ﬁlms. The hydrophobicity of
these ﬁlms depends on both the extent of esteriﬁcation and the nature of the derivatizing
alcohol. Even though PAH/d-PAA ﬁlms are composed of polyelectrolytes, the presence
of hydrophobic ester groups can result in advancing water contact angles as high as 101°.

Cross-linking of hydrophobic PAH/d-PAA ﬁlms via heat-induced amidation stabilizes

coatings over a wide pH range and allows base-promoted hydrolysis of the ester groups
of PAH/d-PAA. After hydrolysis, ﬁlms are extremely hydrophilic and selectively
permeable to Ru(NH3)(,3+ over F e(CI\D(,3 ° due to the high density of newly formed -COO'
groups. To introduce ﬁxed-charge density in PAA/PAH ﬁlms, we also partially
derivatized PAA using 2-nitrobenzyl bromide. Post-deposition UV irradiation of ﬁlms
formed using these photolabile polymers cleaves the 2-nitrobenzyl ester groups to yield
ﬁxed-charge sites. Such ﬁxed charges can enhance Cl'/SO42' selectivities up to 17-fold
relative to underivatized membranes. Transport simulations suggest that both Donnan
exclusion and selective diffusion contribute to selectivity.

Alternating adsorption of polyethyleneimine (FED-metal ion complexes and
polyanions also results in the formation of multilayered polyelectrolyte ﬁlms, and post-
deposition reduction of the metal ions in these ﬁlms yields composite coatings containing
metal nanoparticles. UV/visible spectroscopy and transmission electron microscopy
conﬁrm the formation of well-dispersed nanoparticles with sizes (2-30 run) that depend
on the concentration of metal ions initially in the ﬁlm. Electrodes coated with PEI-Ag
(0)/PAA or PEI-Pt(0)/PAA can catalyze electrochemical reduction of methylene bromide
or 02. Interestingly, PEI-Ag(0)/PAA ﬁlms are effective as antimicrobial coatings. PEI-
Pd(O)/PAA-coated alumina powders catalyze the hydro genations of structurally different
unsaturated alcohols (allyl alcohol, 1-penten-3-ol, 3-methyl-1-penten-3-ol, crotyl alcohol,
and 3-methyl-buten-1-ol) at very different rates. This allows selective hydrogenation of
one molecule in a mixture. Additionally, the polyelectrolyte-encapsulated nanoparticles
exhibit signiﬁcantly lower rates of unwanted substrate isomerizations than do commercial

Pd catalysts.

To my wife, Dan Xu

iv

ACKNOWLEDGEMENTS

First, I would like to thank my advisor, Professor Merlin Bruening. I was so lucky
to be accepted into your group, and have really enjoyed the four years’ scientiﬁc
training under your guidance. You showed me to how be insightﬁil (‘sit down and
think about why is this’), to pay more attentions to details, and to be precise. You
taught me how to write a paper and how to give a presentation. I owe every single
piece of achievement in this dissertation to you. Thank you so very much.

Next, I want to thank my wife, Dan Xu, for your constant support in the years. You
took over most of the housework, which allow me focusing on my study. I could not
ﬁnish my Ph.D without you.

Professor Baker, I thank you very much for your help with my problems that I met
in organic/polymeric synthesis. I was never scared of asking you questions because
always greeted me with a smile.

I also would like to thank Dr. Anton Jensen for the happy cooperation in a project
described in Chapter 3, and Dr. J ian Cui for his help with the anti-microbial testing.

Last, but not least, the Bruening group, past and present. Dan, a mechanist and a
computer scientist in addition to a chemist, thank you so much for taking care of our
computers and many set-ups in the lab. Yinda, I beneﬁted a lot from the stimulating
discussions with you. Anagi, Jeremy, Skanth, Matt, Brian, Wenxi, Kangping,

Xiaoyun, Millind, Sandra, Keith, and Bo: it was so nice being with you.

This work was partially supported by the
Department of Energy Ofﬁce of Basic

Energy Sciences, the American

Chemical Society Petroleum Research

Fund, and the Center for Fundamental

Materials Research at Michigan State

University. v

TABLE OF CONTENTS

List of Tables

List of Figures

List of Abbreviations

Chapter 1: Introduction and Background

1.1 Ultrathin Organic Films
1.2 Layer-by-Layer Adsorption of Oppositely Charged Polyelectrolytes
1.3 Fundamentals of Polyelectrolyte Films
1.4 Potential Applications of Polyelectrolyte Films
1.5 Motivation and Research Goals
1.5.1 Polyelectrolyte Films as Protecting Coatings
1.5.2 Controlling the Permeability of Layered Polyelectrolyte Films
1.5.3 Controlling Ion-Transport through Layered Polyelectrolyte Membranes
1.5.4 Catalytic Nanoparticle/Polyelectrolyte ﬁlms
1.6 References

Chapter 2: Ultrathin Polyelectrolyte Films as Corrosion Inhibition Coatings

2.1 Introduction
2.2 Experimental
2.2.1 Materials
2.2.2 PAA/PAH Film Synthesis
2.2.3 Film Characterization
2.2.4 Electrochemical Studies
2.3 Results and Discussion
2.3.1 Synthesis of PAA/PAH Films on Al
2.3.2 Electrochemical Impedance Spectroscopy
2.2.3 Protection of A1 by PAA/PAH Films
2.4 Conclusions
2.5 References

vi

ix

xiii

\O\OOOO\Uiv—-

13
15
18

27

27
29
29
29
3O
3O
31
31
35
36
41
42

Chapter 3: Controlling the Permeability of Multilayered Polyelectrolyte Films

through Derivatization, Cross-Linking, and Hydrolysis

3.1 Introduction
3.2 Experimental
3.2.1 Materials
3.2.2 Small-Scale Esteriﬁcations of PAA
3.2.3 Large-Scale Ethyl-, 1-Propyl-, and l-Butyl-PAA Syntheses
3.2.4 Large-Scale Benzyl-PAA Synthesis
3.2.5 Large-Scale (2-Methyl-1-butyl)-PAA Synthesis
3.2.6 Film Formation
3.2.7 Film Characterization
3.3 Results and Discussion
3.3.1 Film Formation, Cross-Linking, and Hydrolysis
3.3.2 Contact Angles
3.3.3 Electrochemical Studies
3.3.4 Ion Detection Using Hydrolyzed Films
3.4 Conclusions
3.5 References

Chapter 4: Controlling [on Transport through Multilayer Polyelectrolyte

Membranes by Derivatization with Photolabile Functional Groups

4.1 Introduction
4.2 Experimental
4.2.1 Materials
4.2.2 Synthesis of 2-nitrobenzyl-derivatized PAA
4.2.3 Synthesis of 2-nitrobenzyloxycarbonyl—derivatized PAH
4.2.4 Film Synthesis and Characterization
4.2.5 Photolysis
4.2.6 Transport Studies
4.3 Results and Discussion
4.3.1 Derivatization of PAA and PAH
4.3.2 Film Formation and Photochemistry
4.3.3 Ion-transport Studies
4.3.4 Modeling of Ion Transport
4.4 Conclusions
4.5 Appendix
4.6 References and Notes

vii

46

46
46
46
47
47
48
48
49
50
51
51
55
57
61
63
64

66

66
7O
7O
71
71
73
74
74
75
75
75
79
81
85
86
88

Chapter 5: Metal Nanoparticles/Polyelectrolyte Multilayer ﬁlms:
Synthesis and Catalytic Applications

5.1 Introduction
5.2 Experimental
5.3 Results and Discussion
5.3.1 Synthesis of Metal Nanoparticles
5.3.2 Electrocatalysis
5.3.3 Inhibition of Bacteria Growth
5.4 Conclusions
5.5 References and Notes

Chapter 6: Selective Hydrogenation Using Nanoparticle—Containing
Polyelectrolyte Films

6.1 Introduction
6.2 Experimental
6.2.1 Materials
6.2.2 Synthesis of Polyelectrolyte Films Containing Pd Nanoparticles
6.2.3 Hydrogenation
6.3 Results and Discussion
6.3.1 Synthesis of Polyelectrolyte-Encapsulated Pd Nanoparticles
6.3.2 Hydrogenation
6.3.3 Competitive Hydrogenation
6.3.4 Suppression of Isomerization
6.4 Conclusions
6.5 References
6.6 Appendix

Chapter 7: Conclusions and Future Work

viii

91

91
94
95
95
101
103
105
106

109

109
111
111
113
114
115
115
116
120
123
124
125
127

130

Table 2.1

Table 2.2

Table 3.1

Table 4.1

Table 4.2

Table 6.1

LIST OF TABLES

Film thicknesses and equivalent circuit parameters for Al electrodes
coated with cross-linked 9-bilayer PAA/PAH ﬁlms.

Equivalent circuit parameters for Al electrodes coated with unheated
PAA monolayers and 9-bilayer PAA/PAH ﬁlms that were deposited
at pH 3.5 and pH 4.5.

Thicknesses and advancing water contact angles for 6-bilayer
PAH/d-PAA ﬁlms.

Fluxes and selectivities through membranes prepared from PAA,
PAH, NBPAA, and NBPAH deposited on porous alumina.

Estimated charge densities, diffusion coefﬁcients,
Donnan selectivities, and diffusional selectivities
for derivatized PAA/PAH membranes.

Rates of hydrogenation for structurally related unsaturated
alcohols using commercial 5%—Pd-on-alumina or
PEI-Pd(0)/PAA-on-alumina as catalysts.

ix

39

39

56

80

83

118

Figure 1.1

Figure 1.2
Figure 1.3

Figure 1.4

Figure 1.5

Figure 1.6

Figure 1.7

Figure 2.1

Figure 2.2

Figure 2.3

Figure 2.4

Figure 2.5

Figure 2.6

Figure 3.1

Figure 3.2

LIST OF FIGURES

(A) A Langmuir-Blodgett trough for deposition of monolayers on
solid substrates; (B) Transfer of a LB ﬁlm from the water-air
interface to a solid, hydrophilic substrate.

A schematic diagram of the formation of a self-assembled monolayer.

Formation of an ultrathin ﬁlm by alternating polyelectrolyte adsorption.

Schematic diagram of heat-induced cross-linking of PAA/PAH ﬁlms.
Cross-linking converts polyelectrolyte ﬁlms into passivating coatings.

A composite separation membrane with a selective skin layer and a
porous support.

Schematic drawing of the introduction of net, ﬁxed charges into a
polyelectrolyte membrane via photolysis

Reduction of metal ions in a polyelectrolyte ﬁlm to form a uniform
dispersion of metal nanoparticles.

Synthesis and cross-linking of layered PAA/PAH ﬁlms.

External reﬂection FTIR spectra of 1, 3, 5, 7, and 9-bi1ayer
PAA/PAH on aluminum.

External reﬂection F TIR spectra of 9-bilayer PAA/PAH ﬁlms
before and after heating at 215 °C for 2h under N2 protection.

Equivalent circuits used to simulate impedance data.

Impedance plots for ‘bare’ Al, and Al coated with 9 bilayers of
PAA/PAH, 9 bilayers of cross-linked PAA/PAH, or a monolayer
of unheated PAA.

Tafel plots of ‘bare’ Al and Al coated with 9 bilayers of
cross-linked PAA/PAH (deposited at pH 4.5).

Synthesis, cross-linking, and hydrolysis of PAH/d-PAA ﬁlms.

External reﬂection FTIR spectra of a PAH/PAA ﬁlm and several different

PAH/d-PAA ﬁlms on Au.

10

13

15

17

28

33

34

36

37

40

45

Figure 3.3

Figure 3.4

Figure 3.5

Figure 3.6

Figure 4.1

Figure 4.2

Figure 4.3

Figure 4.4

Figure 4.5

Figure 5.1

Figure 5.2

Figure 5.3

External reﬂection FTIR spectra of unheated, heated, and
hydrolyzed PAH/Pr-PAA ﬁlms (six bilayers) on Au.

Cyclic voltammograms of Ru(NH3)6Cl3 and K3Fe(CN)(,
in 0.5 M Na2S04 at Au electrodes coated with different
PAH/d-PAA ﬁlms.

Cyclic voltammograms of Ru(NH3)6Cl3 in 0.5 M NaZSO4
at Au electrodes coated with cross-linked, hydrolyzed
PAH/d-PAA ﬁlms.

Cyclic voltammograms of Ru(NH3)6Cl3 in 0.1M NaCl containing
different concentration of Ca2+ at Au electrodes coated with 6
bilayers of cross-linked, hydrolyzed PAH/Pr-PAA ﬁlms.

Schematic drawing of the formation of multilayer polyelectrolyte
membrane and subsequent introduction of net, ﬁxed charges via
photolysis.

1H NMR spectra of PAA and PAH that were 50% derivatized
with 2—nitrobenzyl and 2-nitrobenzyloxycarbonyl groups,
respectively

Reﬂection FTIR spectra of a 10.5-bi1ayer 50%NBPAA/PAH

ﬁlm on an Al-coated wafer before (a) and after (b) UV irradiation.
Spectrum (0) is that of the irradiated ﬁlm after a 30 min immersion
in pH 9.2 buffer solution and rinsing with ethanol.

Reﬂection FTIR spectra of a 10.5-bilayer PAA/50%NBPAH ﬁlm
on an Al-coated wafer before (a) and after (b) UV irradiation.

Schematic diagram of a model for ion transport through a multilayer
polyelectrolyte membrane.

UV-visible spectra (A) and TEMs (B,C) of PEI-protected silver
colloids reduced with NaBH.;. The absorbance decreased 50% in
the ﬁrst 10 hours, and average particle size increased from 8 nm

at 1h (B) to 21 nm at 10 h (C).

UV-visible spectra of NaBH4-reduced PEI-Ag(0)/PAA ﬁlms and
TEM image of a NaBH4-reduced PEI-Ag(0)/PAA ﬁlm. ). Silver
particle size is 4.0 :1: 0.6 nm,

TEM image of a 5.5-bilayer PEI-Pt(0)/PAA (NaBH4 reduced) ﬁlm.

xi

54

58

60

62

69

72

77

78

82

96

98

99

Figure 5.4

Figure 5.5

Figure 5.6

Figure 5.7

Figure 6.1

Figure 6.2

Figure 6.3

Figure 6.4

Figure 6.5

TEM images of thermally reduced (heating for 2 h at 150 °C
under N2) 5.5-bilayer PEI-Ag(0)/PAA ﬁlms prepared using different
concentrations of Ag+ during PEI adsorption.

Reﬂection FTIR spectra of a 5.5-bilayer PEI-AgVPAA ﬁlm on A1
before and aﬁer heating at 150 °C for 2h under N2.

(A) Cyclic voltammograms of CHzBrz (25 mM) at bare,
PEI/PAA-coated, and PEI-Ag(0)/PAA-coated Au electrodes,
and at a bulk Ag electrode. (B) Cyclic voltammograms at bare,
PEI/PAA-coated, and PEI-Pt(0)/PAA-coated Au electrodes in
Oz-saturated 1.0 M H2804.

(left) E. C oli. solutions photographed after 12 h of growth

at 37 °C in LB broth. The test tubes were coated with different ﬁlms
as noted on the picture. (right) Visible spectra of the solutions in

the four test tubes.

TEM image of 5 bilayers of PEI-Pd(0)/PAA on a carbon-coated
copper grid.

Percent of substrates hydrogenated vs reaction time for allyl alcohol,
1-penten—3-ol, and 3-methyl-1-penten-3-ol on a commercial
5%-Pd-on alumina catalyst and on a 7-bi1ayer PEI-Pd(0)/PAA-coated
alumina catalyst.

Gas chromatograms of reaction mixtures from competitive
hydrogenations of allyl alcohol and l-penten-l-ol (left), and allyl
alcohol and 3-methy1-1-penten-3-ol (right).

Gas chromatograms from competitive hydrogenations of allyl
alcohol, crotyl alcohol, and 3-methyl-l-buten-l-ol on
5%-Pd-on-alumina and on 7-bilayer PEI-Pd(0)/PAA-on-alumina.

Gas chromatograms of the reaction mixture from hydrogenation

of 1-penten-3-ol on 5%-Pd-on-alumina and on 7-bilayer
PEI-Pd(0)/PAA-on-alumina catalysts. Substrate isomerization was
greatly suppressed by the polyelectrolye ﬁlms.

xii

100

100

102

104

106

119

121

122

123

Ben

Cox

CV
DBU
DMSO
d-PAA
Et
FTIR

1001'

LIST OF ABBREVIATIONS

ion transport selectivity

benzyl

capacitance of ﬁlm

capacitance of oxide

cyclic voltammograms (or voltammetry)
1,8-diazabicyclo[5 .4.0]undec-7-ene
dimethyl sulfoxide

derivatized PAA

ethyl

Fourier-transform infrared
corrosion current

ﬂux

Langrnuir—Blodgett

2-methyl-1 -butyl
3-mercaptopropionic acid
multilayer polyelectrolyte ﬁlms
monolayer self-assembly
2-nitrobenzy1
2-nitrobenzyloxycarbonyl
tetraethylammonium chloride
poly(acrylic acid)
poly(allylamine hydrochloride)
polyethyleneimine
polyelectrolyte-encapsulated Pd nanoparticles
polyelectrolyte ﬁlm

l-propyl

poly(sodium 4-styrenesulfonate)
poly(4-vinylpyridine)

resistance of ﬁlm

resistance of oxide

transmission electron microscopy
thermogravimetric analysis
turnover ﬁ'equency

xiii

Chapter 1

Introduction and Background

1.1 Ultrathin Organic Films

Thin (pm) to ultrathin (nm) ﬁlms are important for applications ranging from
corrosion prevention to drug delivery.1 For example, protective layers in microelectronic
devices must be very thin due to space limitations} 3 and anti-corrosion coatings on
heat-exchangers should be thin enough that they do not greatly diminish heat transfer.4
In membrane separations, for a given material with a speciﬁc selectivity, the separation
layer should be as thin as possible in order to achieve high ﬂuxes. Additionally,
pharmaceutical tablets often contain very thin coatings that protect drugs from oxygen,
moisture, and light (the three key causes of drug degradation), and simultaneously allow
rapid release of drug compounds to the body.5

Polymers are probably the most ubiquitous coating materials because they are
relatively robust and inexpensive. The wide variety of available polymers also affords a
range of material properties, and methods for deposition of these macromolecules allow
coating of substrates ranging from automobiles to pharmaceuticals. Films are oﬁen
prepared by dissolving the appropriate polymer in organic solvent(s), applying a coating
of the material, and allowing the solvent(s) to evaporate. The biggest drawback to this
strategy is the unavoidable release of volatile organic solvents.6 Furthermore, commonly
used methods for ﬁlm formation such as brushing, rolling, spraying, dip coating, and

curtain coating are not suitable for forming ultrathin ﬁlms with controlled thicknessesf’

In ideal cases, spin coating may yield ultrathin ﬁlms on smooth, ﬂat surfaces, but this
technique does not allow for preparation of ﬁlms on a large scale.

The Langmuir-Blodgett (LB) technique is a powerful method for forming ultrathin
organic ﬁlms of amphiphilic molecules with control over ﬁlm thickness on the nm scale.
Langmuir carried out the ﬁrst systematic study of monolayers of amphiphilic molecules
at the water-air interface as early as in 1917,7 and Blodgett developed a method to
transfer the Langmuir ﬁlms onto solid substrates.8 In a typical experiment, a drop of a
dilute solution (~1%) of an amphiphilic molecule in a volatile solvent (e.g., CHC13) is
spread on the water-air interface of a trough (Fig. 1.1A).9 The amphiphiles orient
themselves so that their hydrophilic head groups are dispersed at the water surface while
their hydrophobic tails protrude to the air. Moving a barrier in the trough changes surface
area and controls the ﬁlm density. Films are transferred onto a surface by withdrawal of
a substrate from the water (for hydrophilic surfaces) or insertion of the substrate into the
water (for hydrophobic surfaces) (Fig. 1.1B). Multilayers can be obtained through
retraction and immersion cycles. The LB method provides chemists with the capability
of constructing ordered, ultrathin molecular assemblies, but the technique demands an
experienced operator and careful control of experimental environment and purity of
chemicals. For example, some groups recommend that the LB trough be installed on a
shock absorber in a laminar-ﬂow hood to minimize vibration and to ensure a dust-free
atmosphere.9 The system has little tolerance to organic contamination, and thus
necessitates meticulous cleaning procedures and high purity reagents. Additionally, the

need for amphiphilic constituents limits ﬁlm composition, while the weak van der Waals

forces holding ﬁlms together result in a fragile system.'0 Thus in spite of its appeal, the

LB technique is not likely to yield practical coatings.

a Hydrophobic tail

1
h Hydrophilic head

  

 

 

 

 

 

 

 

 

_@ ©—
:3 ®’
63’
\® GD/
1. ea

 

 

Feasts? 615111111.

I

 

Figure 1.1: (A) A Langmuir-Blodgett trough for deposition of monolayers on
solid substrates. a, bath; b, a moving barrier that allows control of the pressure
applied to the monolayer; c, a motor that moves the barrier; d, a processor for
controlling pressure; e, a balance that measures the surface pressure; f, a motor
with a gearbox that lowers and raises the substrate; g, a solid substrate. (B)
Transfer of a LB ﬁlm from the water-air interface to a solid, hydrophilic

substrate. Figures are adapted from reference 8.

Monolayer self-assembly (MSA) provides another method to form ultrathin
ﬁlms.9’ ”'13 This technique is more convenient than the LB procedure and also produces
more robust ﬁlms. Amphiphilic molecules with thiol, carboxylate, or silane groups
spontaneously form assemblies that are covalently or ionically connected to substrates
such as Au, A1203/A1, and SiOz/Si (Fig. 1.2).9 Although the procedure for ﬁlm formation
is convenient, simply involving immersion of a substrate in a dilute solution of the
amphiphile, the ﬁlm is usually limited to one layer. Substrate requirements also restrict

the applications of these ﬁlms.

(«V4

m w 4}. 5111.115?
#22113sz

Figure 1.2: A schematic diagram of the formation of a self-

 

assembled monolayer.

1.2 Layer-by-Layer Adsorption of Oppositely Charged Polyelectrolytes
Formation of ultrathin ﬁlms by alternating adsorption of polycations and polyanions
overcomes many of the limitations imposed by the LB and MSA methods. Figure 1.3

schematically portrays the principle of this technique.'4 A charged (negatively charged in

 

 

c e: 9
++

c . . © , . e
1. lmmersuon In + 1. lmmersron m e

c positively charged ° + + negatively charged

c polyelectrolyte ° polyelectrolyte e

o —> o e

c, 2. Rinse e 2. Rinse e

6) @
Repetition of
e @

Figure 1.3: Formation of an ultrathin ﬁlm by alternating polyelectrolyte adsorption.

the ﬁgure) substrate is ﬁrst immersed into a solution containing an oppositely charged
polyelectrolyte for a few minutes. The polymer adsorbs onto the substrate due to
electrostatic interactions, and the sign of the surface charge reverses. After rinsing with
water to remove any physisorbed polymer, the substrate is then immersed into a second
solution of polyelectrolyte to produce another layer of polymer. This procedure may be
repeated as many times as desired to produce multilayer ﬁlms.

Film formation by alternating polyelectrolyte deposition has the following assets. 1.

9

Film formation is extremely easy and convenient, involving only a simple “dip and rinse’

procedure. 2. Film thicknesses are tunable on the nanometer scale simply by controlling

the number of layers”’ 15 or deposition conditions such as the pH or ionic strength of the
deposition solutions.”'18 3. Any multiply charged species (polymeric, inorganic, or
biological) can be selected to construct ﬁlms (vide infra). This asset contributes largely
to the versatility of the method. 4. Films can form on a wide variety of surfaces, as the
only required characteristic of the substrate is that it must be charged. This requirement
is easily met because many solid substrates are naturally charged. Importantly, ﬁlms can
be deposited on irregular substrates such as small particles19 or test tubes.20 5. Film

synthesis is environmentally friendly because in most cases the only solvent used is water.

1.3 Fundamentals of Polyelectrolyte Films

Sequential adsorption of oppositely charged colloids was ﬁrst reported in a seminal
paper by Iler.21 Decher and coworkers “rediscovered” this concept of layer-by-layer
growth of ﬁlms and extended it to preparation of mutilayers of polymeric
polyelectrolytes.” 22‘ 23 Although ﬁlm formation is based on electrostatic attractions
between positively and negatively charged species, the true driving force is presumably
due to an entropy increase rather than an enthalpy change.24 Due to the requirement of
electrical neutrality, the number of ionic bonds in the system is the same before and after
adsorption, and thus adsorption results in a negligible enthalpy change. Nevertheless,
complexation of polyions to a charged surface releases many monovalent and divalent
counterions that were previously associated with the polyelectrolytes and the surface,
thus increasing the entropy of the system.

Although multilayer polyelectrolyte ﬁlms are formed by sequential adsorption of

polycations and polyanions, they generally do not have a highly stratiﬁed structure.

Neighboring layers are highly intertwined in most of cases because of loops and tails in

14, 25. 26

the polyions. These tails and loops may be in part responsible for the charge
overcompensation at the ﬁlm surface,27 and the entanglement of adjacent layers also
facilitates complete charge compensation within multilayer ﬁlms. Several studies
detected no counterions in the bulk of polyelectrolyte ﬁlms, suggesting that polycations
are completely charge-compensated by the neighboring polyanions, and vice versa.“‘27
The thicknesses of multilayer polyelectrolyte ﬁlms are highly dependent on adsorption
conditions. For a given substrate and polyelectrolyte system, variables that affect ﬁlm
thickness include the pH and ionic strength of deposition solutions, deposition time,
rinsing and drying methods, and solvent composition.28 The effects of pH and ionic
strength are more pronounced than other factors. For polycarboxylates and polyamines,
pH affects ﬁlm thickness because it controls the charge density in the polyelectrolytes by
altering the degree of protonation/deprotonation.16 In most cases, a high charge density
results in relatively thin ﬁlms because chains are in an extended conﬁguration parallel to
the substrate surface. Moreover, with high charge densities, charge compensation of the
surface can be accomplished with less polymers. Addition of supporting electrolyte to
the deposition solution also allows adjustment of ﬁlm thickness over a wide range.
Though there are exceptions, high concentrations of salt generally increases thickness and

density.15’ 25’ ”'32 The salt screens neighboring charges on the polyelectrolytes and allows

formation of loops and tails.

1.4 Potential Applications of Polyelectrolyte Films

Since the early studies of alternating polyelectrolyte adsorption by Decher and

14, 22. 23

coworkers, the simplicity and versatility of this method have stimulated

investigations of possible applications of multilayer polyelectrolyte ﬁhns. As previously
mentioned, virtually any multiply charged species may be used to construct ﬁlms with

this technique, and even when considering only functional polymers, the scope of ﬁlms

. . , 4 . . . 5, 6 . ‘ ’ 7.4
that can be formed 18 vast. Conducting,33 3 light-emitting,3 3 redox-actrve,25 3‘ 3 0and

reactive polymers4M3 have been used for ﬁlm formation. Speciﬁcally modiﬁed

36, 40, 44

polyelectrolytes including polymeric metal-ion complexes, polymers bearing dyes

41.45-48 51-54
0

for non-linear optics, and labeled polymers (e.g., biotin-,49‘ 50 dye-, r
ﬂuorescent marker-labeled”58 systems) were also prepared for possible applications as
conducting ﬁlms, optical devices, modiﬁed electrodes, and bio/chemical sensors.

In addition to synthetic polymers, natural polyelectrolytes have been used for layer-
by-layer electrostatic adsorption. Biomacromolecules employed include nucleic

49' 6"“ and polysaccharides,(’5 as well as certain charged

acids,”‘ 60 proteins,
supramolecular biological assemblies like viruses"6 and membrane fragments.°7'69 These
investigations were primarily motivated by a desire to achieve biological properties such
as enzymatic activity and biocompatibility in a controlled synthetic system.

The inclusion of charged nanoparticles (predominantly inorganic) in polyelectrolyte
ﬁlms further expands the spectrum of useful materials that can be prepared by layer-by-
layer electrostatic adsorption. These charged nanoparticles (or platelets) can be
considered to be rigid polyelectrolytes, and most of the particles used thus far were

70-73

deposited as negatively charged colloids. Stable colloidal dispersions of silica, metal

l. 21 . 74-78 76, 79-82 55

. . . i-
oxrdes, polyoxometalates, semiconductor nanopartrclesf’ 87 fullerenes,

88-95 97-100

metal colloids, metallo-supramolecular complexes, charged latex

‘9' 21'96’ 97 or microcrystallitesg8 have been successfully employed to prepare

spheres,
coatings. The resulting ﬁlms are nanoparticle-polymer hybrids, which sometimes
combine the unique properties (e. g., electrical, optical, magnetic, catalytic) of
nanoparticles with the mechanical properties of polymers (e.g., ﬂexibility, strength).
These studies have laid the foundation for possible preparation of miniaturized

photoelectronic or magnetic recording devices and efﬁcient catalysts using alternating

polyelectrolyte deposition.

1.5 Motivation and Research Goals
1.5.1 Polyelectrolyte Films as Protecting Coatings

For applications of thin ﬁlms as protective coatings, membranes, and controlled-
release coatings, ﬁlm permeability is among the ﬁrst factors that must be considered. In
the case of anti-corrosion coatings, ﬁlms should be highly impermeable so that the
corrosive species cannot access the substrate. As mentioned earlier, in some cases
protective coatings should also be as thin as possible so as not to mask substrate
properties. Although alternating polyelectrolyte deposition allows formation of ultrathin
ﬁlms, these coatings are usually rather permeable due to their hydrophilic nature and the
fact that they swell in aqueous solution.'7 Harris and coworkers reported a technique for
cross-linking of polyelectrolyte ﬁlms to reduce their permeability."9 In this procedure,
heating of poly(acrylic acid) (PAA)/poly(allylamine hydrochloride) (PAH) ﬁlms on gold

allows formation of amide bonds from the —C00' groups of PAA and the —NH4+ groups

of PAH (Fig. 1.4). These cross-linked, ultrathin ﬁlms have a resistance of ~105 Q—cmz.
Chapter 2 of this thesis describes deposition of such ﬁlms on aluminum and examination
of their ability to inhibit corrosion. Electrochemical impedance studies and polarization
curves show that cross-linked PAA/PAH ﬁlms with thicknesses of only 10 nm decrease
the rate of Al corrosion by 2-3 orders of magnitude.100 I note that after the publication of
this work, Farhat and Schlenoff reported that an ultrathin polyelectrolyte multilayer ﬁlm

can suppress the corrosion of stainless steel under anodic conditions in salt solutions.”"

 

 

R 000’ + (Kc 1? H m
.O/C +H3 NI-Ia -d O— 3

- 0s
§-§C 091.] NH; 5 'o/C g—H NH;
E 3‘ + cross-link E +
is): C90 NHa — E if}: 3..“ H3
“3 0% o ,
(kc cqo H; _ c H
‘d H. __ 0’ — 3

 

Figure 1.4: Schematic diagram of heat-induced cross-linking of
PAA/PAH ﬁlms. Cross-linking converts polyelectrolyte ﬁlms into

passivating coatings.

10

1.5.2 Controlling the Permeability of Layered Polyelectrolyte Films

The minimal thickness of polyelectrolyte ﬁlms could also be exploited to prepare
high-ﬂux membranes, but this will require formation of ﬁlms that are selectively
permeable. This thesis is particularly concerned with ion-transport through
polyelectrolyte ﬁlms. A few groups performed fundamental studies in this area. Using
voltammetry at rotating disk electrodes coated with poly(sodium 4-styrenesulfonate)
(PSS)/poly(diallylmethylammonium chloride) ﬁlms, Schlenoff and coworkers'02
investigated the transport of electroactive ions through polyelectrolyte multilayers. They
found that limiting currents are strongly attenuated by ﬁlms with thicknesses on the order
of a few hundred A, and that the rate of diffusion of ions through these ﬁlms decreases
with increasing ion charge. Ion transport also depends on the sign of the charge at the
ﬁlm surface, presumably because of Donnan exclusion. Mdhwald and von Klitzing
examined the transport of neutral quenchers in polyelectrolyte ﬁlms using time-
dependent ﬂuorescence.” 103 They observed that the quenchers diffuse more slowly in
the bulk of the ﬁlm than in the outer layers, indicating that the outer part of the ﬁlm is
more loosely packed, probably due to less entanglement. They also found that
preparation of ﬁlms without added salt gives a dramatic increase in permeability.
Electrochemical and in situ ellipsometric investigations of layered polyelectrolyte ﬁlms
by our group showed that the permeability of PSS/PAH and PAA/PAH ﬁlms to
Fe(CN)63' and Ru(NH3)(,3+ depends on the solution pH, the number of bilayers in the ﬁlm,
the concentration of salt, and the nature of constituent polymers.17 More recently,
Sukhorukov104 and Donath105 investigated the release of ﬂuorescein and ibufrofen from

layer-by-layer polyelectrolyte mutilayer capsules, and revealed that the diffusion of the

11

core molecules is dependent on several parameters such as the crystal size, the
polyelectrolyte capsule thickness, and the solubility of the core material in the bulk
solutions. Among the results of the above studies, two are particularly important to this
thesis. The ﬁrst is that ion transport varies with the composition of polyelectrolyte ﬁlms.
The second is that Donnan inclusion/exclusion plays an important role in ion transport.
In this work, we further probed the effects of ﬁlm constituents and charge density on
ion transport through polyelectrolyte ﬁlms. Speciﬁcally, we partially derivatized PAA
with different alkyl groups through Fischer esteriﬁcation reactions (Scheme 1.1).106
These partially derivatized PAAs (d-PAAs) still contain —COO' groups that are essential
for formation of PAH/d-PAA ﬁlms. In contrast to underivatized PAA/PAH, the d-
PAA/PAH ﬁlms are highly impermeable to F e(CN)(,3 ' and Ru(NH3)63+ due to the
hydrophobicity of substituent alkyl groups. Interestingly, after cross-linking followed by
base-catalyzed hydrolysis of ester groups, Ru(NH3)63+ ﬂux through the ﬁlms is lO-fold
higher than that of Fe(CN)63'. This occurs due to the removal of hydrophobic alkyl
groups and the introduction of negative —COO' groups during hydrolysis. Chapter 3 of

this thesis presents electrochemical studies that demonstrate control of the permeability

of multilayered polyelectrolyte ﬁlms through derivatization, cross-linking, and hydrolysis.

H2804 W—X

 

COOH A j]: d-PAA
R
PAA NaOH
ii
Scheme 1.1 W_x
X
00'

C00“

12

1.5.3 Controlling Ion-Transport through Layered Polyelectrolyte Membranes
Building on the work described in Chapter 3, Chapter 4 of this thesis reports studies
on controlling ion transport through multilayer polyelectrolyte membranes by
derivatization with photolabile functional groups. Membrane separations are attractive
because of their operational simplicity and low energy cost, and polymer membranes are
used extensively due to their low cost and relatively easy manufacture. Simultaneously
achieving high ﬂux and selectivity, however, is an ongoing challenging in most
membrane applications.107 This problem stems from the fact that selectivity is generally
inversely proportional to permeability over a wide range of membrane materials. '07’ '08
Synthesis of ever-thinner membranes provides a means of increasing ﬂux without
decreasing selectivity, but deposition of membranes with a selective, defect-free layer
that is less than 50 nm thick is very difﬁcult. Adsorption of ultrathin polyelectrolyte
ﬁlms on porous supports may provide a way to overcome this challenge (Fig. 1.5). The

porous support supplies mechanical strength, while the ultrathin polyelectrolyte ‘skin’,

which is responsible for selectivity, should allow high ﬂux.

‘“ Polyelectrolyte ﬁlm

‘- Porous support

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 1.5: A composite separation membrane with a

selective skin layer and a porous support.

13

Previous studies indicate, however, that simple polyelectrolyte ﬁlms do not show

1 9 . l
0 Stair ‘0 and coworkers compared

particularly high ion-transport selectivity. Harris,
transport of Cl', S042} and Fe(CN)(,3' through PSS/PAH and PAA/PAH ﬁlms coated on
porous alumina supports. They obtained a selectivity of 4-9 for Cl'/SO42' and a
selectivity of 310-1500 for Cl'/Fe(CN)63' using 5-bilayer ﬁlms. Once complete pore
coverage occurs, the presence of additional polyelectrolyte bilayers decreases anion ﬂux,
but has little impact on selectivity. Krasemann and Tieke did similar research and
observed a selectivity of 45 for Cl‘/SO42' and 113 for NaVMgz‘L.32 The differences in CI‘
/SO42' selectivity between the two studies are likely due to different supports and
deposition conditions. Krasemann and Tieke used 60 bilayers of PSS/PAH coated on a
porous polymer support. Both studies suggest that selectivity likely results from Donnan
exclusion of multiply charged ions.

Assuming that Donan exclusion does play an important role in ion transport through
polyelectrolyte ﬁlms, the introduction of net charges into polyelectrolyte ﬁlms should
greatly enhance ion-transport selectivity. Although polyelectrolyte ﬁlms contain a high
concentration of charges, polycations and polyanions electrically compensate each other
and there is little net charge in these ﬁlmszs’ 3 I To introduce net charge, we partially
derivatized PAA with photolabile functional groups, and then formed derivatized
PAA/PAH ﬁlms on porous alumina supports. Photolysis of speciﬁc functional
nitrobenzyl ester groups liberates —COO' groups in the ﬁlm that must be charge-
compensated by ions in solution (Fig. 1.6).1 '1 Photolysis is preferable to base-catalyzed

hydrolysis because it is mild and does not corrode the alumina support. Transport studies

show that the introduction of net charges into polyelectrolyte ﬁlms increase C1'/SO42’

14

 

 

Figure 1.6: Schematic drawing of the introduction of net,

ﬁxed charge into a polyelectrolyte membrane via photolysis.

selectivity by as much as 17-fold, and the selectivity is tunable by controlling the degree
of derivatization. We also simulated ion transport through this system using a simple
model. The simulations indicate that Donnan exclusion is not entirely responsible for
ion-transport selectivity. Diffusivity differences between various ions are also important,
and in some cases dominant, in determining selectivity.
1.5.4 Catalytic Nanoparticle/Polyelectrolyte films

Chapter 5 of this thesis focuses on developing new methods for preparing
polyelectrolyte ﬁlms containing catalytic nanoparticles. This study expands the
applications of polyelectrolyte ﬁlms to include preparation of catalysts for electrodes,

membranes, and synthesis of ﬁne chemicals.I 12’ ”3 Interestingly, although a number of

15

studies described development of nanoparticle/polyelectrolyte ﬁlms, we are unaware of
studies of catalysis with these systems.

Nanoparticles are attractive catalysts because of their high surface area and unique
size-dependent properties." “4 However, practical application of these materials will
require their immobilization on supports so that the catalysts can be recovered and
recycled. Because of their easy manipulation and relatively cheap price, polymeric
catalyst supports have been explored recently by several groups.1 ”“23 Examples include
immobilization of Pd nanoparticles on poly(vinylpyridine) nanospheres,115 in the interior

116,117
1,

of a block copolymer she] or in the inside of dendrimers.l '8'120 The polymers

stabilize nanoparticles, regulate the access of reactant to the particles, and expedite

119.121.122.124
catalyst recovery.

In applications, the polymer-capped nanoparticles are
generally dispersed in solutions in a batch reactor. In spite of the improvements in
encapsulation, the complete recovery of catalysts is still challenging. Uniformly
dispersed catalytic nanoparticle/polymer composite membranes should have advantages
in this area,121 but as discussed earlier, making ultrathin polymer ﬁlms is challenging.
In Chapter 5, I present a new method of forming ultrathin ﬁlms that contain catalytic
metal nanoparticles (Fig. 1.7).20 This method circumvents the cumbersome procedures
for preparation of colloids that are commonly adapted in constructing inorganic
nanoparticle/polyelectrolyte composites.“ 89’ 93’ 95' 122425 I also demonstrate that metal

nanoparticle—containing polyelectrolyte ﬁlms are active as electrocatalysts and

antimicrobial ﬁlms.

16

Q
1 . .50).

19‘

 

69 = metal ion

Figure 1.7: Reduction of metal ions in a polyelectrolyte

ﬁlm to form a uniform dispersion of metal nanoparticles

The ﬁnal work in this thesis aims at utilizing catalytic nanoparticle/polyelectrolyte
ﬁlms to afford selective hydrogenations. Such selective catalysis may allow the
performance of organic reactions such that puriﬁcation is not needed after each step.

This work, described in Chapter 6, represents the culmination of this thesis as it combines

selective transport through polyelectrolyte ﬁlms with nanoparticle catalysis.

1.6 References

1

10

ll

12

13

14

Fendler, J. H. Nanoparticles and Nanostructured Films: Preparation,
Characterization and Applications; Wiley-VCH: Weinhein, 1998.

Bellucci, F.; Nicodemo, L.; Monetta, T.; Kloppers, M. J.; Latanision, R. M.
Corrosion Sci. 1992, 33, 1203-1226.

Negi, Y. S.; Goyal, R. K. Mol. Crys. Liq. Crys. Sci. T ech., C: Mol. Mater. 2001, 14,
103-120.

Melo, L. F.; Bott, T. R. Exp. Therm. Fluid Sci. 1997, 14, 375-391.

Johnson, J. L. In Coatings Technology Handbook; Satas, D., Tracton, A. A., Eds;
Marcel Dekker: New York, 2001, pp 863-866.

Wicks, Z. W.; Jones, F. N.; Pappas, S. P. Organic Coatings: Science and
Technology, 2nd ed.; John Wiley & Sons: New Yorker, 1999.

Langmuir, I. J. Am. Chem. Soc. 1917, 39, 1848-1852.
Blodgett, K. B. J. Am. Chem. Soc. 1935, 57, 1007-1010.

Ulman, A. An Introduction to Ultrathin Organic Films: from Langmuir-Blodgett to
Self-Assembly; Academic Press: San Diego, 1991.

Embs, F.; Funhoff, D.; Laschewsky, A.; Licht, U.; Ohst, H.; Prass, W.; Ringsdorf,
H.; Wegner, G.; Wehrmann, R. Adv. Mater. 1991, 3, 25-30.

Allara, D. L.; Nuzzo, R. G. Langmuir 1985, 1, 45-52.
Nuzzo, R. G.; Fusco, F. A; Allara, D. L. J. Am. Chem. Soc. 1987, 109, 2358-2368.
Nuzzo, R. G.; Dubois, L. H. Chem. Rev. 1996, 96, 1533-1554.

Decher, G. Science 1997, 277, 1232-1237.

18

15

16

l7

l8

19

20

21

22

23

24

25

26

27

28

29

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

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

Harris, J. J .; Bruening, M. L. Langmuir 2000, 16, 2006-2013.

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

Caruso, F. Adv. Mater. 2001, 13, 11-22.

Dai, J .; Bruening, M. L. Nano Letters 2002, 2, 497-501.

Iler, R. K. J. Coll. Int. Sci 1966, 21, 569-594.

Decher, G.; Hong, J. D. Macromol. C hem, Macromol. Symp. 1991, 46, 321-327.

Decher, G.; Hong, J. D. Ber. Bunsenges. Phys. Chem. 1991, 95, 1430-1434.

Arys, X.; Jonas, A. M.; Laschewsky, A.; Legras, R. In Supramolecular Polymers;
Ciferri, A., Ed.; Marcel Dekker: New York, 2000, pp 505-564.

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

Joly, S.; Kane, R.; Radzilowski, L.; Wang, T.; Wu, A.; Cohen, R. E.; Thomas, E. L.;
Rubner, M. F. Langmuir 2000, 16, 1354-1359.

Lowack, K.; Helm, C. A. Macromolecules 1998, 31 , 823-833.

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

Lvov, Y.; Decher, G.; Mohwald, H. Langmuir 1993, 9, 481-486.

19

30

31

32

33

34

35

36

37

38

39

40

41

42

43

Clark, S. L. M., Martha F .; Hammond, Paula T. Macromolecules 1997, 30, 7237-
7244.

Laurent, D.; Schlenoff, J. B. Langmuir 1997, 13, 1552-1557.

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

Lukkari, J.; Salomaeki, M.; Viinikanoja, A.; Aeaeritalo, T.; Paukkunen, J .;
Kocharova, N.; Kankare, J. J. Am. Chem. Soc. 2001, 123, 6083-6091.

Fou, A. C.; Rubner, M. F. Macromolecules 1995, 28, 71 15-7120.

Liu, Y. Q.; Ma, H.; Jen, A. K. Y. Chem. Commun. 1998, 2747-2748.

Wu, A; Yoo, D.; Lee, J.-K.; Rubner, M. F. J. Am. Chem. Soc. 1999, 121, 4883-
4891.

Hodak, J .; Etchenique, R.; Calvo, E. J. Langmuir 1997, 13, 2708-2716.

Kaschak, D. M.; Lean, J. T.; Waraksa, C. C.; Saupe, G. B.; Usami, H.; Mallouk, T.
E. J. Am. Chem. Soc. 1999, 121, 3435-3445.

Meier, L. P.; Heule, M.; Caseri, W. R.; Shelden, R. A.; Suter, U. W.
Macromolecules 1996, 29, 718-723.

Keller, S. W.; Johnson, S. A.; Brigham, E. S.; Yonemoto, E. H.; Mallouk, T. E. J.
Am. Chem. Soc. 1995, 117, 12879-12880.

Koetse, M.; Laschewsky, A.; Mayer, B.; Rolland, 0.; Wischerhoff, E.
Macromolecules 1998, 31 , 9316-9327.

Laschewsky, A.; Mayer, B.; Wischerhoff, E.; Arys, X.; Jonas, A.; Kauranen, M.;
Persoons, A. Angew. Chem. Int. Ed. Engl. 1997, 36, 2788-2791.

Sun, J .; Wu, T.; Sun, Y.; Wang, Z.; Zhang, X.; Shen, J .; Cao, W. Chem. Commun.
1998, 1853-1854.

20

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

Lee, J.-K.; Yoo, D.; Rubner, M. F. Chem. Mater. 1997, 9, 1710-1712.

Lindsay, G. A.; Roberts, M. J.; Chaﬁn, A. P.; Hollins, R. A.; Merwin, L. H.;
Stenger—Smith, J. D.; Yee, R. Y.; Zarras, P.; Wynne, K. J. Chem. Mater. 1999, 11,
924-929.

Roberts, M. J .; Lindsay, G. A.; Herman, W. N.; Wynne, K. J. J. Am. Chem. Soc.
1998, 120, 11202-11203.

Laschewsky, A.; Wischerhoff, E. Macromolecules 1997, 30, 8304-8309.

Balasubramanian, S.; Wang, X.; Wang, H. C.; Ke Yang; Kumar, J .; Tripathy, S. K.
Chem. Mater. 1998, 10, 1554-1560.

Hong, J. D.; Lowack, K.; Schmitt, J .; Decher, G. Prog. Colloid Polym. Sci. 1993,
93, 98-102.

Cassier, T.; Lowack, K.; Decher, G. Supramol. Sci. 1998, 5, 309-315.

Fischer, P.; Laschewsky, A. Macromolecules 2000, 33, 1100-1102.

Ferreira, M.; Rubner, M. F. Macromolecules 1995, 28, 7107-7114.

Delcorte, A.; Bertrand, P.; Wischeroff, E.; Laschewsky, A. Langmuir 1997, 13,
5125-5136.

Dante, S.; Advincula, R.; Frank, C. W.; Stroeve, P. Langmuir 1999, 15, 193-201.

Liu, Y.; Wang, Y.; Lu, H.; Claus, R. O. J. Phys. Chem. B 1999, 103, 2035-2036.

Kerimo, J .; Adams, D. M.; Barbara, P. F.; Kaschak, D. M.; Mallouk, T. E. J. Phys.
Chem. B 1998, 102, 9451-9460.

Kaschak, D. M.; Mallouk, T. E. J. Am. Chem. Soc. 1996, 118, 4222-4223.

Klitzing, R. v.; Mohwald, H. Macromolecules 1996, 29, 6901-6906.

21

59

60

61

62

63

64

65

66

67

68

69

7O

71

72

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

Montrel, M. M.; Sukhorukov, G. B.; Shabarchina, L. I.; Apolonnik, N. V.;
Sukhorukov, B. I. Mater. Sci. Eng. C 1998, C5, 275-279.

Ladam, G.; Schaaf, P.; Cuisinier, F. J. G.; Decher, G.; Voegel, J .-C. Langmuir 2001,
1 7, 878-882.

Caruso, F .; Mohwald, H. J. Am. Chem. Soc. 1999, 121, 6039-6046.

Caruso, F.; Niikura, K.; Furlong, D. N.; Okahata, Y. Langmuir 1997, I3, 3427-
3433.

Eremenko, A; Kurochkin, 1.; Chemov, S.; Barmin, A.; Yaroslavov, A.; Moskvitina,
T. Thin Solid Films 1995, 260, 212-216.

Lvov, Y.; Onda, M.; Ariga, K.; Kunitake, T. J. of Biomater. Sci, Polym. Ed. 1998,
9, 345-355.

Lvov, Y.; Haas, H.; Decher, G.; Mohwald, H. Langmuir 1994, 10, 4232—4236.

He, J .-A.; Samuelson, L.; Li, L.; Kumar, J .; Tripathy, S. K. Adv. Mater. 1999, 11,
435-446.

He, J.-A.; Samuelson, L.; Li, L.; Kumar, J .; Tripathy, S. K. J. Phys. Chem. B 1998,
102, 7067-7072.

He, J .-A.; Samuelson, L.; Li, L.; Kumar, J .; Tripathy, S. K. Langmuir 1998, 14,
1674-1679.

Caruso, F.; Caruso, R. A.; M'o‘hwald, H. Science 1998, 282, 1111-1114.

Caruso, F.; Lichtenfeld, H.; Giersig, M.; Mdhwald, H. J. Am. Chem. Soc. 1998, 120,
8523-8524.

Caruso, F.; Mdhwald, H. Langmuir 1999, 15, 8276-8281.

22

73

74

75

76

77

78

79

80

81

82

83

84

85

86

87

Lvov, Y.; Ariga, K.; Onda, M.; Ichinose, I.; Kunitake, T. Langmuir 1997, 13, 6195-
6203.

Cassagneau, T.; Fendler, J. H.; Mallouk, T. E. Langmuir 2000, 16, 241-246.

Aliev, F. G.; Correa-Duarte, M. A.; Mamedov, A.; Ostrander, J. W.; Giersig, M.;
Liz—Marzan, L. M.; Kotov, N. A. Adv. Mater. 1999, 11, 1006-1010.

Keller, S. W.; Kim, H.-N.; Mallouk, T. E. J. Am. Chem. Soc. 1994, 116, 8817-8818.

Rosidian, A.; Liu, Y.; Claus, R. 0. Adv. Mater. 1998, 10, 1087-1091.

Chen, Z.-h.; Yang, Y.-a.; Qiu, J.-b.; Yao, J .-n. Langmuir 2000, 16, 722-725.

Caruso, F.; Kurth, D. G.; Volkmer, D.; Koop, M. J .; Miiller, A. Langmuir 1998, 14,
3462-3465.

Ichinose, 1.; Tagawa, H.; Mizuki, 8.; Yuri Lvov, a.; Kunitake, T. Langmuir 1998,
14, 187-192.

Moriguchi, 1.; F endler, J. H. Langmuir 1998, 10, 2205-2211.

Cheng, L.; Niu, L.; Gong, J.; Dong, S. Chem. Mater. 1999, 11, 1465-1475.

Cassagneau, T.; Mallouk, T. E.; Fendler, J. H. .1. Am. Chem. Soc. 1998, 120, 7848-
7859.

Kotov, N. A.; Dékany, I.; Fendler, J. H. J. Phys. Chem. 1995, 99, 13065-13069.

Sun, Y.; Hao, B; Zhang, X.; Yang, 8.; Shen, J .; Chi, L.; Fuchs, H. Langmuir 1997,
13, 5168-5174.

Gao, M.; Richter, B.; Kirstein, S.; Mdhwald, H. J. Phys. Chem. B 1998, 102, 4096-
4103.

Zhang, X.; Shen, J. Adv. Mater. 1999, 11, 1139-1143.

23

88

89

90

91

92

93

94

95

96

97

98

99

100

101

102

Feldheim, D. L.; Grabar, K. C.; Natan, M. J .; Mallouk, T. E. J. Am. Chem. Soc.
1996, 118, 7640-7641.

Schrof, W.; Rozouvan, S.; Van Keuren, E.; Horn, D.; Schmitt, J.; Decher, G. Adv.
Mater. 1998, 10, 338-341.

Cassagneau, T.; Fendler, J. H. J. Phys. Chem. B 1999, 103, 1789-1793.

Schmitt, J .; Decher, G.; Dressick, W. J .; Brandow, S. L.; Geer, R. E.; Shashidhar,
R.; Calvert, J. M. Adv. Mater. 1997, 9, 61-65.

Liu, Y.; Claus, R. o. J. Appl. Phys. 1999, 85, 419-424.

Liu, Y.; Wang, Y.; Claus, R. 0. Chem. Phys. Lett. 1998, 298, 315-319.

Yonezawa, T.; Onoue, S.-y.; Kunitake, T. Adv. Mater. 1998, 10, 414-416.

He, J.-A.; Valluzzi, R.; Yang, K.; Dolukhanyan, T.; Sung, C.; Kumar, J .; Tripathy,
S. K. Chem. Mater. 1999, 11, 3268-3274.

Caruso, F.; Susha, A. S.; Giersig, M.; Mdhwald, H. Adv. Mater. 1999, 11, 950-953.

Serizawa, T.; Takeshita, H.; Akashi, M. Langmuir 1998, 14, 4088-4094.

He, J .-A.; Yang, K.; Kumar, J .; Tripathy, S. K. J. Phys. Chem. B 1999, 103, 11050-
11056.

Harris, J. J.; DeRose, P. M.; Bruening, M. L. J. Am. Chem. Soc. 1999, 121, 1978-
1979.

Dai, J.; Sullivan, D. M.; Bruening, M. L. Ind. Eng. Chem. Res. 2000, 39, 3528-
3535.

Farhat, T. R.; Schlenoff, J. B. Electrochem. Solid-State Lett. 2002, 5, Bl3-B15.

Farhat, T. R.; Schlenoff, J. B. Langmuir 2001, 1 7, 1184-1192.

24

103 Klitzing, R. v.; Mohwald, H. Thin Solid Films 1996, 352—356.

104 Antipov, A. A.; Sukhorukov, G. B.; Donath, E.; Mdhwald, H. J. Phys. Chem. B
2001, 105, 2281-2284.

105 Qiu, X.; Leporatti, S.; Donath, E.; Mohwald, H. Langmuir 2001, 1 7, 5375-5380.

106 Dai, J .; Jensen, A. W.; Mohanty, D. K.; Emdt, J .; Bruening, M. L. Langmuir 2001,
1 7, 931-937.

107 Liu, C.; Martin, C. R. Nature (London) 1991, 352, 50-52.

108 Liu, C.; Espenscheid, M. W.; Chen, W.-J.; Martin, C. R. J. Am. Chem. Soc. 1990,
112, 2458-2459.

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

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

111 Dai, J .; Balachandra, A. M.; Lee, J. 1.; Bruening, M. L. Macromolecules 2002, 35,
3164-3170.

112 Bengtson, G.; Scheel, H.; Theis, J .; Fritsch, D. Chem. Eng. J. 2002, 85, 303-311.

113 Vankelecom, I. F. J .; Jacobs, P. A. Catalysis Today 2000, 56, 147-157.

114 Rao, C. N. R.; Kulkarni, G. U.; Thomas, P. J .; Edwards, P. P. Chem. Eur. J. 2002, 8,
28-35.

115 Pathak, S.; Greci, M. T.; Kwong, R. C.; Mercado, K.; Surya Prakash, G. K.; Olah, G.
A.; Thompson, M. E. Chem. Mater. 2000, 12, 1985-1989.

116 Bronstein, L. M.; Chemyshov, D. M.; Volkov, I. O.; Ezernitskaya, M. G.; Valetsky,
P. M.; Matveeva, V. G.; Sulman, E. M. J. Catalysis 2000, 196, 302-314.

117 Underhill, R. S.; Liu, G. Chem. Mater. 2000, 12, 3633-3641.

25

118

119

120

121

122

123

124

125

Rahim, E. H.; Kamounah, F. S.; Frederiksen, J .; Christensen, J. B. Nano Lett. 2001,
1, 499-501.

Yeung, L. K.; Crooks, R. M. Nano Lett. 2001, 1, 14-17.

Niu, Y.; Yeung, L. K.; Crooks, R. M. J. Am. Chem. Soc. 2001, 123, 6840-6846.

Liu, C.; Xu, Y.; Liao, S.; Yu, D. Appl. Catal., A 1998, 172, 23-29.

Lvov, Y.; Munge, B.; Giraldo, 0.; Ichinose, I.; Suib, S. L.; Rusling, J. F. Langmuir
2000, 16, 8850-8857.

Mamedov, A. A.; Belov, A.; Giersig, M.; Mamedova, N. N.; Kotov, N. A. J. Am.
Chem. Soc. 2001, 123, 7738-7739, and references therein.

Ostrander, J. W.; Mamedov, A. A.; Kotov, N. A. J. Am. Chem. Soc. 2001, 123,
1101-1110.

Liu, Y.; Wang, A.; Claus, R. 0. Appl. Phys. Lett. 1997, 71, 2265-2267.

26

Chapter 2

Ultrathin Polyelectrolyte Films as Corrosion Inhibition Coatings

2.1 Introduction

In spite of the fact that surface passivation generally increases with coating thickness,
there are applications where ultrathin coatings would be advantageous. In heat-
exchangers, for example, coatings should be thin enough that they do not drastically
reduce heat-transfer coefﬁcients.I Thin coatings are also important for protecting
electronic materialsz'4 and microelectrode arrays.5

To provide satisfactory protection, a ﬁlm must exhibit high chemical resistance in a
corrosive environment and good adhesion to the underlying structure. Langrnuir-
Blodgett ﬁlms are not attractive in either aspect and thus are rarely used as anti-corrosion
coatings.6’7 Self-assembled monolayers are chemically adsorbed onto substrates, and full
coverage on very uniform structures may occur. Although they can supply good short-
term protection in some cases, they frequently have defects, and lack thermal and
mechanical stability.5'8'IO

The strong electrostatic attachment of polyelectrolytes to surfaces provides good
adhesion to the substrate, and layered polyelectrolyte ﬁlms have many advantages as
described in the introduction of this thesis (section 1.2). Unfortunately, layered
polyelectrolyte ﬁlms are relatively permeable because their ionic composition allows

them to absorb water.l "'3 Because of this, we investigated conversion of poly(acrylic

acid)/(poly(allylamine hydrochloride) (PAA/PAH) ﬁlms into cross-linked polyamides

27

 

 

 

 

 

 

04C coo-
‘O/C
M 04C coo-
Al COO’ Al 'O/C
—————> Osc -coo-
(1) "O’C
04C coo-
_ _ '0/C
E I 1,, (2)
NH;
V
' 0s coo-
C +
'o/ +H,N NH3
04C coo- +
.O/ +H3N NH3
A1
Osc coo- NH .
.O/ +H3N 3
04C coo: +
. 'o/ +H3N “3

lheat

_ ﬁg NH3,

‘ El} NH;
Al f SC? NH;
11. NH:

Figure 2.1: Synthesis and cross-linking of layered PAA/PAH

0240242

0

 

 

cm

ﬁlms. Repetition of steps 1 and 2 produces multilayers.

28

through heating (Fig. 2.1). These cross-linked, nylon-like ﬁlms are highly passivating.l4
Here we report on the possibility of using these ﬁlms to protect Al from corrosion.
Electrochemical studies show that such ﬁlms protect the native oxide layer of Al, and
even a 10 nm-thick, nylon-like ﬁlm reduces corrosion currents by 2-3 orders of
magnitude. For comparison, we also investigated corrosion inhibition by non-cross-
linked PAA/PAH ﬁlms and PAA monolayers. The ultrathin cross-linked ﬁlms discussed
here should have a stability advantage over non-cross-linked and monolayer ﬁlms, but
they do have the disadvantage of needing a heat-treatment. The heat treatment (215 °C),
however, is compatible with most metals, and such treatments are common in the

deposition of practical expoxy coatings.IS

2.2 Experimental

2.2.] Materials. Poly(acrylic acid) (Mw~90,000, 25 wt % solution in water) and
poly(allylamine hydrochloride) (Mw~70,000) were obtained from Aldrich. Solutions of
0.01 M PAA, 0.01 M PAH and 0.5 M NaCl were prepared in 18.2 MQcm Millipore
water, and pH values were adjusted with 0.1 M NaOH or HCl. (For polymers,
concentration is given with respect to the repeating unit). Al substrates were made by
sputtering 200 nm of Al on Si(100) wafers. This substrate has a native oxide layer with
an ellipsometric thickness of about 40 A after 15 minutes of cleaning in a UV/ozone
cleaner. In this chapter, when we refer to Al or “bare” A], we mean A1 with its native
oxide.

2.2.2 PAA/PAH film synthesis. Al substrates were cleaned for 15 minutes in a

UV/ozone cleaner (Boekel Industries, model 135500) and subsequently immersed in a

29

0.01 M PAA solution for 5 minutes to form the ﬁrst PAA layer on Al. The wafers were
then rinsed with H2O for 1 minute, immersed in a 0.01 M PAH solution for 5 min, and
rinsed with water for 1 min. This cycle was repeated until the desired number of bilayers
was deposited, and the ﬁlms were then blown dry with N2. Cross-linking was performed
by heating at 215 °C under N2 ﬂow for 2 hours. The temperature was measured with a
mercury thermometer that touched the bottom of the round-bottom ﬂask next to the
substrate. Heating at lower temperatures results in less cross-linking and higher
permeability. M

2.2.3 Film Characterization. External reﬂection FTIR spectra of ﬁlms were obtained
with a Nicolet Magna-IR 560 spectrometer using a Pike grazing angle (80°) attachment.
The whole instrument was enclosed in a glove box that was purged with the N2 from the
sample compartment. This decreases unwanted peaks due to water vapor. Film
thicknesses were measured with a M-44TM rotating analyzer spectroscopic ellipsometer (J.
A. Woollam), and the ﬁlms were assumed to have a reﬂective index of 1.5. For thickness
calculations, the Al/A12O3 system was assumed to be one homogeneous substrate, and
values of refractive index, n, and absorption coefﬁcient, k, for this system were
determined on each substrate prior to ﬁlm deposition. Modeling Al/Al2O3 with two
layers while assuming literature values for n and k of A1 and A1203 allowed
determination of the oxide thickness. This two-layer substrate model gives the same
thicknesses for adsorbed polyelectrolyte ﬁlms (within experimental error) as the
assumption of a homogeneous substrate.

2.2.4 Electrochemical studies. Impedance data and polarization curves were obtained

with a CH Instruments electrochemical analyzer (Model 604). Measurements were made

30

using a Ag/AgCl (3M KCl) reference electrode and a platinum wire counter electrode.
The working electrode was “bare” or coated Al contained within an O-ring holder that
exposed 2.2 cm2 of the sample. We intentionally used a rather corrosive electrolyte
solution (0.5 M NaCl, adjusted to pH 3.0 with HCl) for electrochmical measurements to
differentiate between the impedances of “bare” and ﬁlm-coated Al. At neutral pH, the
impedance of “bare” Al is as high as 107 Q-cm2 after a 4 h immersion in 0.5 M NaCl due
to its native oxide. Impedance data acquisition started aﬁer four hours of immersion in
NaCl solutions to achieve stable values. For “bare” Al, the impedance data were
acquired at the open circuit potential (ca. —1 .20 V vs Ag/AgCl) using a sinusoidal voltage
of 5 mV. Other impedance measurements were made at a DC potential of —0.75 V vs.
Ag/AgCl, because coated Al has a much more positive open circuit potential (-0.48 to -
0.75 V) than does “bare” Al. Applying potentials higher than —0.7 V (pitting potential of
Al4 ) causes rapid corrosion in some cases. The frequency range for impedance
measurements was 105 Hz to 10 mHz, and ﬁtting of impedance data to equivalent circuits
was performed using LEVM 7.0 software written by J. Ross MacDonald (available from
Solatron).l6 Polarization curves (scan rate of 1 mV/sec) were acquired 10 minutes after
ﬁnishing impedance measurements (after a total of 5 hours of immersion). All

electrochemical experiments were performed on at least three different electrodes.

2.3 Results and Discussion
2.3.1 Synthesis of PAA/PAH Films on Al. Synthesis of PAA/PAH ﬁlms on Al occurs
in a controlled layer-by-layer fashion (Fig. 2.1). We begin the synthesis by immersing an

Al-coated wafer into a solution of PAA to form carboxylate linkages to the oxidized

31

surface.'4’15 Subsequent alternating immersions in PAH and PAA solutions produce a
multilayer ﬁlm. External reﬂection F TIR spectra (Figure 2.2) demonstrate the layer-by-
layer grth of these coatings as absorbances increase regularly with the number of
deposited bilayers. The dominant peaks in these spectra are due to the -C00'
asymmetric (1575 cm") and symmetric ( 1400 cm") stretches, and the -COOH carbonyl
stretch (1710 cm'l).

By varying the pH of deposition solutions, one can vary the thickness and composition
of PAA/PAH ﬁlms.16 The FTIR spectra in Figure 2.3 show that the ratio of -COOH to
-C00' groups increases with a decrease in deposition pH. When depositing at pH 5.5,
the ratio of the absorbances (peak heights) due to the acid carbonyl and the -C00'
asymmetric stretch is 0.1 (Fig. 2.33). At pH 3.5 this ratio increases to 0.5 (Fig. 2.3b).
This change is explained by the variation of protonation/deprotonation of PAA with pH.
Rubner and coworkers noted that bilayer thicknesses in PAA/PAH ﬁlms vary
dramatically with the pH at which ﬁlms are deposited.” This is certainly true in the
present case as 9-bilayer ﬁlms deposited at pH 5.5 are 30% as thick as 9-bilayer ﬁlms

deposited at pH 4.5 (See Table 2.1 on page 39).

32

 

 

 

 

9
0.01
7
O
O
C
3
3 5
In
.0
<
3
1
1700 1600 1500 1400

Wavenumbers (cm‘l)

Figure 2.2: External reﬂection FTIR spectra of 1, 3, 5, 7, and 9 bilayers of
PAA/PAH on aluminum. Films were synthesized at pH 5.5.

33

 

 

 

0.02
f
o ‘ ‘ e
0
s A
.D .
3 d
m
2
C
a

 

1800 1700 1600 1500 1400 1300
Wavenumbers (cm")

Figure 2.3: External reﬂection FTIR spectra of 9-bilayer PAA/PAH ﬁlms
before (a, b, c) and after (I, e, f) heating at 215 °C for 2h under N2 protection.
Deposition pH values were as follows: pH 5.5, a and d; pH 3.5, b and e; pH
4.5, c and f. The spectra of ﬁlms deposited at pH 5.5 (a and d) were multiplied

by 2 for ﬁgure clarity.

In a procedure similar to a method for forming nylon, heating PAA/PAH ﬁlms at 215
°C under N2 efﬁciently converts carboxylate/ammonium salts into amide groups.l7 Upon
heating, absorbance peaks due to amide groups (1675 and 1530cm") appear in external
reﬂection FTIR spectra, and peaks due to -C00' groups vanish (Fig. 2.3). The spectra
suggest nearly complete conversion of carboxylates to amides. Amide formation occurs
for ﬁlms deposited at all of the pH values, but the number of remaining -COOH groups
does vary with deposition pH. As expected, deposition at lower pH results in more
residual -COOH groups (Fig. 2.3).

2.3.2 Electrochemical Impedance Spectroscopy. Electrochemical impedance
spectroscopy is one of the most powerful methods for studying corrosion because it often
provides mechanistic information about corrosion processes.”‘26 In this technique, one
determines the impedance of an electrochemical cell over a wide range of frequencies
and then uses an equivalent circuit to describe impedance as a function of frequency.
Various equivalent circuit models were proposed to simulate the impedance behavior of

- 10.23-25.27-29
coated alumrnum,

and most of these are similar to the circuits shown in Figure
2.4. For “bare” Al, the insulating oxide ﬁlm forms a leaky capacitor resulting in circuit A.
The presence of an organic coating that is highly blocking complicates the circuit by the
addition of another leaky capacitor as shown in circuit B. As a parallel RC element
results in a semicircle in a plot of real versus imaginary components of impedance, the
introduction of an organic coating often results in a plot with two partial

semicircles.'7"8’23 Note that the magnitude of the diameter of each semicircle is

proportional to the resistance in the parallel RC circuit.3O Thus, larger semicircles in

35

impedance plots represent better corrosion protection. Scully and Hensley suggest that

coating resistances need to be at least 107 Q-cm2 to provide effective protection of Al.31

 

 

 

 

 

 

 

c C'
l L
g °l" 1 1
l 1 Cox
———1vwv—~ ’WW—— --
Rs ——'wvv——— Rs
ROX R
f
ROX
A B

Figure 2.4: Equivalent circuits used to simulate impedance data. Circuit A was
used to ﬁt “bare” Al and Al coated with non-cross-linked PAA/PAH ﬁlms. Circuit B
was used to ﬁt Al coated with cross-linked PAA/PAH ﬁlms. The physical meaning
of the different symbols is RS, solution resistance, Rex, oxide resistance, Rf, ﬁhn

resistance, Cf, ﬁlm capacitance, and Cox, oxide capacitance.

2.3.3 Protection of Al by PAA/PAH Films. Impedance data show that PAA/PAH ﬁlms
protect the underlying aluminum oxide layer. Figure 2.5 shows impedance plots for
"bare" Al (inset) and Al coated with 9-bilayers of cross-linked (triangles) and unheated
(squares) PAA/PAH ﬁlms. Figure 2.5 also shows the impedance plot for Al coated with
a monolayer of PAA (diamonds). All of the coated Al electrodes have values of R0,‘ that

are 2-3 orders of magnitude higher than for "bare" Al (Tables 2.1 and Table 2.2).

36

 

 

-2” (Mn-cmz)

 

 

 

 

 

 

 

012 3 4 5 6
zwmnemn

Figure 2.5: Impedance plots for ‘bare’ Al (circles, inset) and Al coated with 9-
bilayer PAA/PAH (squares), 9-bilayer cross-linked PAA/PAH (triangles), or a
monolayer of unheated PAA (diamonds). Films were deposited at pH 4.5. Lines
represent ﬁts to the data using the equivalent circuits in ﬁgure 2.4. All
impedance data were measured in 0.5 M NaCl at pH 3.0 after a 4-h immersion

time.

37

Tafel plots in Figure 2.6 also conﬁrm the protective value of PAA/PAH coatings. The
corrosion currents for "bare" A1 are 2-3 orders of magnitude higher than those for Al
coated with cross-linked PAA/PAH ﬁlms. This is in excellent agreement with the 2-3
order of magnitude increase in oxide resistance determined by impedance. For unheated
PAA/PAH ﬁlms and PAA monolayers, it is difﬁcult to obtain complete Tafel plots like
those in Figure 2.6 due to their very positive open circuit potential (ca. —0.48 V). A
couple of plots that we did obtain show corrosion currents within the same order of
magnitude as those for Al coated with cross-linked ﬁlms.

The impedance data explain why ultrathin ﬁlms reduce the corrosion current for Al.
Binding of a monolayer of PAA to the surface passivates the oxide layer and, in turn,
protects Al (Table 2.2). The ﬁlm resistance of a single PAA layer or unheated PAA/PAH
ﬁlms should be small, but the ﬁlm apparently inhibits dissolution of the oxide. We
surmise that a protective layer of Al carboxylates is forming on the surface similar to

32’” and self-assembled monolayersg’34

protective layers formed by corrosion inhibitors
The layer inhibits Cl‘ adsorption, thus protecting the oxide.35 This layer will not be as
ordered as protective ﬁlms of self-assembled monolayers, but it does passivate the oxide
ﬁlm. Multilayer PAA/PAH ﬁlms likely provide more complete coverage than do PAA
monolayers, resulting in higher Rox values (Table 2.2).

Although passivation of the surface oxide likely dominates corrosion currents,

impedance plots show deﬁnite differences between cross-linked and unheated PAA/PAH

ﬁlms. As seen in Figure 2.5, the impedance plot for Al coated with cross-linked ﬁlms

38

Table 2.1: Film thicknesses and equivalent circuit parameters“ for A1 electrodes coated

with cross-linked 9-bilayer PAA/PAH ﬁlms.

 

Coated at Coated at Coated at

 

Bare Al pH 3.5 pH 4.5 pH 5.5“
Thickness before - 330 : 20 410 : 30 130 : 15
heating (A)b
Thickness after _ 260 i 20 320 i 20 100 i 10
heating (A)b
R. (MQ-cmz) - 4 : 0.8 3 : 0.6 7 : 1
R,,. (MQ—cmz) 0.07 : 0.02 80 : 60 60 : 44 70 :50
Cr(1.lF/cm2) - 0.55 : 0.05 0.75 : 0.10 2.4 : 0.6
c.x (uF/cm’) 7.5 : 1.0 2.5 :01 2.6 : 0.2 0.8 : 0.4
1,... (nA/cmz) 800 : 150 3.5 : 1 3.5 : 1 3.5 : 1

" Equivalent circuit elements are deﬁned in Figure 2.4 and were determined from
impedance spectroscopy.

b Thicknesses were determined using ellipsometry.

C Impedance plots for ﬁlms deposited at pH 5.5 do not show two apparent semicircles as
do plots for ﬁlms deposited at pH 3.5 and 4.5. However, phase angle-frequency plots

suggest two time constants in all cases, so data were ﬁt with circuit B in ﬁgure 2.4.

Table 2.2: Equivalent circuit parameters for Al electrodes coated with unheated PAA
monolayers and 9-bilayer PAA/PAH ﬁlms that were deposited at pH 3.5 and pH 4.5.

 

 

Monolayer 9 bilayer PAA/PAH
pH 3.5 pH 4.5 pH 3.5 pH 4.5
R0,(MQ-cmz) 11 :2 10:2 40: 10 30: 10
C... (”F/cmz) 4.2 : 0.1 4.3 : 0.1 4.4 : 0.3 4.5 : 0.3

39

 

"bare" Al

coated Al

Log (I, Alcmz)
60
o

'10 ‘ .

 

 

 

-11 T I I I
-0.6 -O.8 -1.0 -1.2 -1.4 -1.6

Potential (V vs AgIAgCI)

Figure 2.6: Tafel plots of bare Al and Al coated with 9-bilayer of cross-linked
PAA/PAH (deposited at pH 4.5). The data were taken after a 5-h immersion time
in 0.5 M NaCl adjusted to a pH of 3.0.

40

contains two semicircles, while that for unheated ﬁlms contains only one. Plots of phase-
angle versus frequency also show that there is an additional time constant for cross-linked
ﬁlms. The ﬁlm resistance, Rf, for unheated ﬁlms is negligible,‘4 while values of Rf for
cross-linked ﬁlms are around 5 MQ-cm2 (Table 2.1). Even though this value is
substantial, ﬁlm resistance will not signiﬁcantly decrease corrosion currents because R0"
is the dominant resistance in these systems. As mentioned earlier, Rf is generally > 107
Q-cm2 in ﬁlms used for practical corrosion protection.3 ' One has to consider, however,
that the Rf values in this study result from ﬁlms as thin as 100 A.

Both Rf and R0,. show little variation with deposition pH (Table 2.1). As gauged by
the value of Rf per unit thickness, cross-linked ﬁlms deposited at pH 5.5 are superior to
ﬁlms deposited at pH 3.5 and 4.5, but the differences in Rf/thickness are less than an
order of magnitude. Thus the presence of excess -COOH groups does not appreciably

increase the permeability of cross-linked ﬁlms in acidic conditions.

2.4 Conclusions

PAA/PAH ﬁlms can be synthesized on A1 by simple "dip and rinse" layer-by-layer
methods, and heating of these ﬁlms results in amidation of PAA/PAH. Cross-linking by
amidation yields ﬁlms with resistances of 5 MQ—cmz, whereas ﬁlm resistance before
cross-linking is negligible. Interestingly, even unheated PAA monolayers provide
substantial short-terrn protection for Al by passivating the surface oxide. Although the
resistance of cross-linked PAA/PAH ﬁlms is probably not sufﬁcient for practical

corrosion protection, considering the thickness of these ﬁlms, 5 MQ-cm2 is encouraging.

41

2.5 References

10

ll

12

13

14

15

16

17

Melo, L. F .; Bott, T. R. Exp. Therm. Fluid Sci. 1997, 14, 375-391.

Kirchner, C.; George, M.; Stein, 8.; Parak, W. J .; Gaub, H. E.; Seitz, M. Adv.
F unct. Mater. 2002, 12, 266-276.

Negi, Y. S.; Goya], R. K. M01. Crys. Liq. Crys. Sci. T ech., C: Mol. Mater. 2001, 14,
103-120.

Bellucci, F .; Nicodemo, L.; Monetta, T.; Kloppers, M. J.; Latanision, R. M.
Corrosion Sci. 1992, 33, 1203-1226.

Schmitt, G.; FaBbender, F.; Liith, H.; Schoning, M. J .; Schultze, J.-W.; BuB, G.
Mater. Corr. 2000, 51 , 20-25.

Jaiswal, A.; Singh, R. A.; Dubey, R. S. Corrosion 2001, 5 7, 307-312.

Ulman, A. An Introduction to Ultrathin Organic Films: from Langmuir-Blodgett to
Self-Assembly; Academic Press: San Diego, 1991.

Tsuji, N.; Nozawa, K.; Armaki, K. Corros. Sci. 2000, 42, 1523-1538.
Zamborini, F. P.; Crooks, R. M. Langmuir 1998, 14, 3279-3286.
French, M.; Creager, S. E. Langmuir 1998, 14, 2129-2133.

Harris, J. J .; Bruening, M. L. Langmuir 2000, 16, 2006-2013.

Losche, M.; Schmitt, J .; Decher, G.; Bouwman, W. G.; Kjaer, K. Macromolecules
1998, 31, 8893-8906.

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

Harris, J. J.; DeRose, P. M.; Bruening, M. L. J. Am. Chem. Soc. 1999, 121, 1978-
1979.

Grandle, J. A.; Taylor, S. R. Corrosion 1994, 50, 792-803.
Macdonald, J. R. Solid State Ionics 1992, 58, 97-107.

Deﬂorian, F.; F edrizzi, L.; Bonora, P. L. Progress in Organic Coatings 1993, 23,
73-88.

42

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

Deﬂorian, F .; Miskovic-Stankovic, V. B.; Bonora, P. L.; F edrizzi, L. Corrosion
1994, 50, 438-446.

Kendig, M.; Scully, J. Corrosion 1990, 46, 22-29.

Mansfeld, F.; Chen, C.; Lee, C. C.; Xiao, H. Corros. Sci. 1996, 38, 497-513.
Bonora, P. L.; Deﬂorian, F.; Fedrizzi, L. Electrochim. Acta 1996, 41 , 1073-1082.
Grifﬁn, A. J. J .; Brotzen, F. R. J. Electrochem. Soc. 1994, 141, 3473-3479.

Su, P.-C.; Devereux, O. F. Corrosion 1998, 54, 419-427.

Mertens, S. F.; Xhoffer, C.; De Cooman, B. C.; Temmerrnan, E. Corrosion 1997,
53, 381-388.

Scholl, H.; Jimenez, M. M. D. Corros. Sci. 1992, 33, 1967-1978.
Lin, S.; Shih, H.; Mansfeld, F. Corrosion Science 1992, 33, 1331-1349.

Bessone, J .; Mayer, C.; Jiittner, K.; Lorentz, W. J. Electrochim. Acta. 1983, 28,
171-175.

Bellucci, F.; Kloppers, M.; Latanision, R. M. J. Electrochem Soc. 1991, 138, 40-48.
Mansfeld, F.; Kendig, M. W.; Tsai, S. Corrosion 1982, 38, 478-485.

Bard, A. J .; Faulkner, L. R. Electrochemical Methods: Fundamentals and
Applications; 2nd ed.; John Wiley & Sons: New York, 2001.

Scully, J. R.; Hensley, S. T. Corrosion 1994, 50, 705-716.
Heeg, B.; Klenerman, D. Langmuir 2000, 16, 1783-1792.

Brusic, V.; Frisch, M. A.; Eldridge, B. N.; Novak, F. P.; Kaufman, F. B.; Rush, B.
M.; Frankel, G. S. J. Electrochem. Soc. 1991, 138, 2253-2259.

Ogawa, H.; Chihera, T.; Taya, K. J. Am. Chem. Soc. 1985, 107, 1365-1369.

EI-Awady, A. A.; Abd-EI-Nabey, B. A.; Aziz, S. G. J. Chem. Soc. Faraday Trans.
1993, 89, 795-802.

43

Chapter 3

Controlling the Permeability of Multilayered Polyelectrolyte Films through
Derivatization, Cross-Linking, and Hydrolysis
3.1 Introduction
The convenience and versatility of layer—by-layer polyelectrolyte deposition make
polyelectrolyte ﬁlms (PF 5) attractive for possible applications in sensors,"3 separation

. 4 -
13 l and corrosron

membranes,4'“ capillary electrophoresis,” light-emitting diodes,
protection.IS Several of these applications require control over ﬁlm permeability.
Development of corrosion-resistant coatings, for example, requires PFs that are
impermeable to ions, while separation membranes composed of PFs should be selectively
permeable. Synthesis of impermeable PFs will likely require development of
hydrophobic16 or highly cross-linked ﬁlms.“ ‘8 In contrast, PFs are attractive for ion-
separation membranes in part because their high charge density may result in Donnan
exclusion of multiply charged ions.4’ 5
This chapter describes the preparation (see Fig. 3.1) and use of partially esteriﬁed
poly(acrylic acid) (d-PAA) to control both hydrophobicity and charge density in
poly(allylamine hydrochloride) (PAH)/d-PAA ﬁlms. The hydrophobicity of PAH/d-PAA
ﬁlms results in coatings that passivate surfaces. In contrast, ester hydrolysis affords
highly charged ﬁlms that are selectively permeable to ions. We employ only partial
esteriﬁcation of PAA because unreacted -COO’ groups are necessary for formation of
PFs. Esteriﬁcation of 50-70% of the -COOH groups of PAA still results in very
hydrophobic PAH/d-PAA ﬁlms (advancing water contact angles of up to 101 °) that

greatly restrict access to an underlying electrode.

44

”coo Moore,
4300- pAH (new?
coo— W -COO’*H3
~coo- cow 3

"coom, ROOC
Hg’OOC

~COO‘ +H3 ROOC
‘ Hg-ooc

lCOO +H3 ROOC

 

 

 

 

 

 

 

 

 

 

 

SCOO— +H3 H3R%%%
H
1% Q COR
.811 ROOC CO"
H
£H ROOC 00R
00'
Q H H Q OOR
"C
" H
4 h drol ze
_ O H _ ( ) y y
~63 OO
8H ‘OOC 00'
H
H '00C 00.
£14 H 00:
L H 00

Figure 3.1: Synthesis, cross-linking (heating), and hydrolysis of PAH/d-PAA ﬁlms.

Repetition of steps 1 and 2 produces multilayer ﬁlms.

45

To increase charge density in a PAH/d-PAA coating, we ﬁrst heat the ﬁhn to form
amide cross-links from carboxylate-ammonium ion pairs.18 Subsequently, basic
hydrolysis of ester groups results in a high density of -C00' groups in the ﬁlm. Unlike
the -COO' groups present during deposition of PAH/d-PAA, carboxylates formed by
hydrolysis are not charge-compensated by a neighboring ammonium group in PAH.
Several previous studies show that charge compensation in many PFs is almost
completely intrinsic (polycations are compensated by polyanions, and vice versa).19'2'
The charges on carboxylate groups formed by hydrolysis, however, must be charge-
compensated by cations from the solution (extrinsic compensation).20 Binding of
divalent cations to extrinsically compensated -COO' sites in hydrolyzed PAH/d-PAA

ﬁlms dramatically alters ﬁlm permeability and provides a possible means of sensing low

. . + . . +
concentratrons of ions such as Ca2 , even 1n a 5 order of magmtude excess of Na .

3.2 Experimental Section

3.2.1 Materials. PAA (Aldrich, Mw = 750 000), PAH (Aldrich, MW = 70 000), 3-
mercaptopropionic acid (MPA) (Aldrich), K3Fe(CN)6 (Mallinckrodt), Ru(NH3)6Cl3 (Alfa
Aesar), AlCl3-6H2O (Spectrum), CaCl2-2H2O (Baker), NaCl (Baker), Na2SO4 (Baker),
ethanol (Pharmco, 200 proof), l-butanol (Fisher, Certiﬁed ACS), 2-methyl-1-butanol
(Fisher, Certiﬁed ACS), pentane (Aldrich, HPLC grade), diethyl ether (Fisher,
anhydrous), and sulfuric acid (Fisher, Certiﬁed ACS) were used as received. Propanol
and benzyl alcohol were distilled over magnesium and iodine. All aqueous solutions

were prepared with 18 MQcm Milli-Q water

46

3.2.2 Small-Scale Esterifications of PAA. Fischer esteriﬁcation reactions were
performed and described by Dr. Anton Jensen and his associates at Central Michigan
University. Small-scale reactions were initially run to test the reactivity of the alcohols.
This involved adding 0.5-1.0 g of PAA, 10-20 mL of alcohol, and 3 drops of
concentrated sulfuric acid to a round-bottom ﬂask ﬁtted with a condenser. The mixtures
were then heated with stirring to reﬂux (ethanol, 1-propanol, and l-butanol) or about 100
°C (benzyl alcohol). IR spectra were taken at regular intervals by placing a drop from the
solution onto a salt plate and drying. At these same intervals, 2 mL aliquots were
removed from the reaction for 1H NMR analysis. These aliquots were added to distilled
water, pentane, or diethyl ether to precipitate the polymer, which was then dried in a
vacuum oven for about an hour. About 50 mg of dried precipitate was then dissolved in
do-DMSO for analysis. The extent of derivatization was calculated from the 1H NMR
spectra. Once small-scale reactions were analyzed by IR and 1H NMR, a correlation
could be made between relative intensities of ester and acid carbonyl peaks in the IR
spectra and extent of derivatization. Therrnogravimetric analysis (TGA) of the polymers
was conducted at a heating rate of 10 °C/min under nitrogen using a DuPont 2900 TGA.
These data were collected by Dr. Dilip Mohanty of Central Michigan University.

3.2.3 Large-Scale Ethyl-, 1-Propyl—, and l-Butyl-PAA Synthesis. For the ethyl, 1-
propyl, and l-butyl ester derivatives, large-scale reactions were run by adding 10 g of
PAA, about 130 mL of alcohol, and 10 drops of sulfuric acid to a round-bottom ﬂask
ﬁtted with a reﬂux condenser. The solution was then heated with stirring to reﬂux. The
reactions were followed via IR by comparing the relative intensities of ester and sour

carbonyl stretches with the small-scale-reaction IR spectra. Thus, reactions were allowed

47

to proceed until the desired percent derivatization, as estimated by IR, was reached. The
products were then precipitated and analyzed by 1H NMR as with the small scale
reactions. For the ethyl derivative, aﬁer 3 h of reﬂux PAA was 67% ethyl esteriﬁed
(82% yield). For the l-propyl derivative, after 1 h and 50 min of reﬂux PAA was 61% l-
propyl esteriﬁed (66% yield). For the 1-butyl derivative, after 40 min of reﬂux PAA was
69% 1-butyl esteriﬁed (71% yield). The 19% ethyl ester polymer that was used in the
ﬁlm studies was the ﬁrst aliquot removed from a small-scale reaction aﬁer 30 min of
heating.

3.2.4 Large-Scale Benzyl-PAA Synthesis. For the benzyl ester derivative, a large scale
reaction was run by adding 3.0 g of PAA, about 30 mL of alcohol, and 3—4 drops of
sulfuric acid to a round-bottom ﬂask ﬁtted with a reﬂux condenser. The clear, gelatinous
slurry was then heated to 100 °C. The product was precipitated as a white solid, soxhlet
extracted with diethyl ether, and dried under high vacuum with minimal heating before
analysis by 1H NMR. After 2 h and 10 min of reﬂux, PAA was 52% benzyl esteriﬁed
(quantitative yield).

3.2.5 Large-Scale (2-Methyl-1-butyl)-PAA Synthesis. Only a large-scale reaction was
run for the 2-methyl-1-butyl ester derivative. This was done by adding 10 g of PAA, 220
mL of 2-methyl-1-butanol, and 6 drops of sulfuric acid into a round-bottom ﬂask ﬁtted
with a reﬂux condenser and a mechanical stirrer. The solution was heated to 105 °C.

The mixture was heated initially for 3 h, stopped for 11 h, and recommenced for an
additional 4.5 h, all with continuous stirring. At the end of the heating, the polymer was
precipitated in pentane, dried under vacuum for 20 h at 62 °C, and analyzed by 1H NMR.

The product was found to be 30% derivatized (93% yield).

48

3.2.6 Film Formation. To synthesize PAH/d-PAA ﬁlms, we ﬁrst cleaned a gold-coated
Si slide (2.4 X 1.2 cm piece of a Si( 100) wafer that was sputter-coated with 20 nm of Cr
followed by 200 nm of Au) for 15 min in a Boekel UV/ozone cleaner. Next, we formed a
monolayer of MPA on the gold by immersing the slide in 2 mM MPA in ethanol for 45
min, rinsing with ethanol, and drying with N2. (MPA was used because it can be cross-
linked to PAH in a later step. Otherwise, 3-mercapto-l-propanesulfonic acid might be a
better choice because it is not pH-sensitive.22) Deposition of a layer of PAH occurred by
immersion (5 min) of the MPA-coated slide into an aqueous 0.02 M PAH, 0.2 M NaCl
solution that was adjusted to pH 5.5 with NaOH. (Salt was added to increase the
thickness of PAH layers.” 20‘ 23‘ 24 The molarity of the polymer is given with respect to
the repeating unit.) The slide was then removed from solution, rinsed with water for 1
min, dried with N2, and immersed in a solution containing PAA or d-PAA for 5 min.
Rinsing again with water and drying with N2 completed the synthesis of the ﬁrst bilayer
of PAH/PAA. Subsequent bilayers were prepared with the same PAH/PAA deposition
steps.

Derivatized PAA molecules were not soluble in water, so deposition solutions for
these polymers were prepared in mixtures of water and methanol or water and THF.”’ 25
Solutions containing PAA derivatized with ethyl, l-propyl, or 2-methyl—1-buty1 alcohol
(Et-, Pr-, and mBu-PAA) were prepared by dissolving 200 mg of polymer in 90 mL of
methanol, adding 10 mL of water, and adjusting the apparent pH (measured with a pH
meter) to 7.0 with a few drops of 0.1 M NaOH. To prepare solutions for deposition of
PAA derivatized with butyl alcohol (Bu-PAA), 200 mg of polymer was dissolved (with

sonication) in 90 mL of methanol, and then 60 pL of N-methylmorpholine (added to

49

deprotonate the -COOH groups of Bu-PAA) and 10 mL of water were added to this
solution. (Bu-PAA precipitates in solutions of methanol and water in the absence of N-
methylmorpholine.) Solutions for deposition of PAA derivatized with benzyl alcohol
(Ben-PAA) were prepared by dissolving 200 mg of Ben-PAA in 90 mL of THF and
adding 60 uL of N—methylmorpholine and 10 mL of water. (Ben-PAA does not dissolve
in methanol). For deposition of underivatized PAA, we used 0.01 M aqueous solutions
of PAA that were adjusted to pH 5.0 with NaOH.

Cross-linking of ﬁlms was performed by heating at 180 °C under N2 for 2 h. Heating
at higher temperatures causes more loss of ester groups, while lower temperatures result
in less cross-linking.18 Hydrolysis of cross-linked ﬁlms was effected by immersion of the
sample in a solution of 2.0 M NaOH in 90%methanol+10%water for 4 h at 60 °C
followed by rinsing with 1 mM NaOH and drying with N2. We rinsed with 1 mM NaOH
rather than pure water to avoid protonation of -C00' groups.

3.2.7 Film Characterization. External reﬂection FTIR spectra were obtained with a
Nicolet Magna-560 FTIR spectrometer using a Pike grazing angle attachment (80° angle
of incidence). The spectrometer employs a MCT detector. Film thickness was measured
with a rotating analyzer ellipsometer (J. A. Woollam model M-44), assuming a ﬁlm
refractive index of 1.5. Contact-angle measurements were performed with a FTA200
(First Ten Angstroms) contact angle analyzer. We used the following procedure in
measuring the advancing contact angle of water. A 2 11L drop of water was formed on
the tip of a stainless steel syringe needle and placed on a ﬁlm by raising the substrate
until contact was made between the drop and sample. Subsequently water was added to

the drop at a rate of 0.2 uL/s for about 7 s to produce the advancing contact angle. Then

50

an image of the drop was taken and the left and the right contact angles were determined
and averaged. The needle was in the drop when the contact angle was measured. All
samples were dried in a ﬂask under ﬂowing N2 for 2 h before contact angle measurement.
Cyclic voltammograms were obtained with a CH Instruments electrochemical analyzer
(model 604). Measurements were made using a Ag/AgCl (3 M KCl) reference electrode
and a platinum wire counter electrode. The working electrode was one of the Au-coated

Si slides contained within a plastic/O-ring holder that exposed 0.1 cm2 of the sample.

3.3 Results and Discussion

3.3.1 Film Formation, Cross-Linking, and Hydrolysis. Formation of PAH/d-PAA
ﬁlms occurs by alternating immersion of MPA-coated Au slides in solutions containing
PAH or d-PAA as shown schematically in Figure 3.1. External reﬂectance FTIR spectra
(Fig. 3.2) conﬁrm the formation of PAH/d-PAA ﬁlms and show signiﬁcant differences
among the different PAH/d-PAA coatings. Peaks due to the -C00' asymmetric (1575
cm'l) and symmetric (1400 cm'l) stretches and the ester C=O (1735 cm") and C-0 (1170
and 1265 cm']) stretches appear in the spectra of all PAH/d-PAA ﬁlms. (The ester
carbonyl peaks probably also contain a small contribution from the carbonyl stretch of
underivatized, un-ionized -COOH groups.) The ratio of absorbances of the ester carbonyl
peaks to those of -C00' peaks is not the same in each of the spectra, and this ratio is
affected both by the extent of derivatization and by the nature of derivatized groups.

Higher percentages of derivatization lead to higher ratios of ester carbonyl peak heights

51

 

  
  

ester C=O
ester C-O
0.02

A PAH/Bu(69)-PAA

PAH/Pr(61 )-PAA

   

 

-COO' hi

 
   

 

 

 

Absorbance

 

3

ﬂ PAH/Et(67)-PAA J

PAH/Et(19)-PAA
PAH/Ben(52)-PAA

AH/mBu(30)-PAA

PAH/PAA

??

E

 

 

 

I I I I I I

3200 2900 2600 2300 2000 1700 1400 1100

Wavenumber (cm")

Figure 3.2: External reﬂection F TIR spectra of a PAH/PAA ﬁlm and several
different PAH/d-PAA ﬁlms on Au. The fraction of -COOH groups in d-PAA

that were esteriﬁed is given in parentheses. The ﬁlms contained six bilayers.

52

to -C00' peak heights as shown by the spectra of PAH/Et(67)-PAA and PAH/Et(19)-
PAA in Figure 3.2. PAH/Bu-PAA ﬁlms have unusually small -COO' peaks. The high
hydrophobicity of these ﬁlms may result in higher pK,l values for -COOH groups and
hence smaller -COO' peaks. An additional reason for the low -COO’ peaks in spectra of
PAH/Bu-PAA ﬁlms may be the fact that due to the low solubility of Bu-PAA, PAH/Bu-
PAA ﬁlms were prepared under different conditions than most other ﬁlms (see
Experimental Section). However, preparation of PAH/Pr-PAA ﬁlms under the conditions
used for PAH/Bu-PAA ﬁlms did not alter the IR spectrum of PAH/Pr—PAA.

Heating PAH/d-PAA ﬁlms at 180 °C under N2 cross-links coatings by converting
carboxylate-ammonium salts into amides (Fig. 3.1, step 3). The reﬂectance FTIR spectra
in Figure 3.3 give evidence for this transformation. Upon heating, peaks due to amide
groups appear (1675 and 1530 cm'l), and peaks due to -C00' groups vanish. The spectra
suggest nearly complete conversion of carboxylates to amides. Aﬁer heating, the height
of the peak due to the ester carbonyl stretch drops 15-25% for all PAH/d-PAA ﬁlms
except PAH/Bu-PAA, which shows no signiﬁcant change. We speculate that drops in
carbonyl absorbance result from partial conversion of ester groups to amides. The lack of
conversion of esters to amides in PAH/Bu-PAA ﬁlms may be due to higher thermal
stability of the butyl ester (TGAs for Et-, Pr-, and Ben-PAA show a ~10% weight loss at
150-180 °C. Weight loss for Bu-PAA occurs at higher temperatures).

Reﬂectance FTIR spectra also demonstrate the success of base-promoted hydrolysis of
esters in cross-linked PAH/d-PAA ﬁlms (Fig. 3.3). After exposure of ﬁlms to 2 M NaOH
in 90% MeOH for 4 h (60 °C),26 ester carbonyl peaks decrease by at least 65%, and -

COO' stretches again appear in IR spectra. It is not possible to estimate the percentage of

53

 

 

ester C=O
0.01

 

ester C-O

 

 

ydrolyzed

 

3000 2900 2800

 

 

 

g unheated i‘midel')
3 ide(ll
a: 0.01
O
m
n
< A J
heated
hydrolyzed

 

 

 

I I I I I I

3200 2900 2600 2300 2000 1700 1400 1100

Wavenumber (cm")

Figure 3.3: External reﬂection FTIR spectra of unheated, heated, and hydrolyzed
PAH/Pr-PAA ﬁlms (six bilayers) on Au. The inset shows the IR spectra
(2800 cm'1 to 3000 cm") of heated and hydrolyzed PAH/Bu-PAA ﬁlms.

ester groups that were hydrolyzed using the carbonyl region of the spectrum, because the
ester-carbonyl peak and the acid-carbonyl peak overlap. However, the nearly complete
disappearance of methyl (2965 and 2880 cm") and C-0 stretches (1170 and 1265 cm")
of ester groups (Fig. 3.3) suggests that hydrolysis is almost quantitative. The
disappearance of stretches due to methyl groups is especially clear in the spectra of
hydrolyzed PAH/Bu-PAA ﬁlms (inset, Fig. 3.3). The presence of the strong amide peak
at 1675 cm'1 also shows that amide groups are stable under hydrolysis conditions.

The ellipsometric thicknesses of six-bilayer ﬁlms range from 200 to 300 A (Table 3.1).
Film thickness results chieﬂy from the d-PAA deposition step, as the thickness increase
after deposition of each PAH layer is only about 10 A. Heating at 180 °C decreases ﬁlm
thickness by about 15%. This decrease may be due to loss of absorbed water27 as well as
dehydration of carboxylate-ammonium salts. Hydrolysis, on the other hand, does not
decrease thickness, suggesting that cross-linked ﬁlms do not collapse after ester cleavage.
Hydrolyzed ﬁlms may also absorb solvent.

3.3.2 Contact Angles. The presence of ester groups in PAH/d-PAA coatings dramatically
alters ﬁlm hydrophobicity. For unheated ﬁlms, the advancing water contact angles
(Table 3.1) increased from 42° for underivatized PAH/PAA to as high as 101° for
PAH/Bu-PAA. Contact angles appear to increase with the amount of derivatization as
well as the length of alkyl groups. After heating, however, contact angles on PAH/d-
PAA ﬁhns dropped in most cases, probably because polymers are less ﬂexible after
cross-linking and hydrophobic groups cannot collect at the air interface. Upon removal

of hydrophobic groups through ester cleavage, PAH/d-PAA ﬁlms become very

55

hydrophilic. This dramatic change provides further evidence for successful hydrolysis of

cross-linked ﬁlms.

Table 3.1: Thicknesses and advancing water contact angles for 6-bilayer
PAH/d-PAA ﬁlms.

d-P A A“ before
heatrng
PAA 247 : 10

Et (19%) 195 : 9
Et (67%) 210 : 8
Pr (61%) 230: 12
Bu (69%) 295 : 13
Ben (52%) 255 : 12

mBu (30%) 263 : 10

thickness (A)
after after
heating hydrolysis
217 : 8
171 : 9 176 : 6
178 : 8 188 : 6
194 : 11 204 : 7
269 : 10 281 : 11
229 i 11 225 : 6
227 i 7 239 : 12

before
heating

42:2
80:2
85:4
89:2
101:4
81:2

88:4

contact angle (degf

after
heating

68:4
65:5
70:2
72:3
91:2
96:3

82:3

after
hydrolysis”

<10
<10
<10
<10
<10

<10

" The number in parentheses represents the fraction of -COOH groups that were

derivatized as determined by NMR.

b Water is rapidly adsorbed by hydrolyzed ﬁlms.
" Contact angles of PAH/Et(l9%)-PAA and PAH/mBu(30%)-PAA are the average of

values at four different spots on one ﬁlm. The other values are the average of eight

different spots on two ﬁlms.

56

3.3.3 Electrochemical Studies. The presence of ester groups in PAH/d-PAA ﬁlms
greatly decreases their permeability to electrochemically active probe molecules. Peak
currents in cyclic voltammograms (CVs) of Ru(NH3)(,3+ at all PAH/d-PAA-coated
electrodes are signiﬁcantly lower than those at PAH/PAA-coated electrodes (Fig. 3.4,
top). For ﬁlms containing l-propyl, l-butyl, 2-methy-1-butyl, and benzyl esters, peak
currents decreased by about 2 orders of magnitude compared to underivatized PAA/PAH
films. (Peak currents at PAH/d-PAA electrodes were calculated after subtracting currents
measured at the same electrode in the absence of the redox couple (see inset of Fig. 3.4)).
Similar trends occur in CVs of Fe(CN)(,3’ (Fig. 3.4, bottom), showing that hydrophobic
ﬁlms are passivating to both cations and anions. The ethyl-derivatized ﬁlms do not
passivate electrodes as well as other PAH/d-PAA ﬁlms, probably due to the smaller size
and lower hydrophobicity of ethyl groups. However, from the CVs at PAH/Et(19)-PAA-
and PAH/Et(67)-PAA-coated electrodes (Fig. 3.4, top), one ﬁnds that a higher percentage
of derivatization does result in more electrode passivation. Cross-linking of ﬁhns does
not dramatically affect peak currents in most cyclic voltammograms (it signiﬁcantly
increases passivation due to PAH/PAA and PAH/Et-PAA ﬁlms because these coatings
were initially highly permeable), but it does stabilize ﬁlms and allow us to perform base-
promoted hydrolysis of ester groups.

Hydrolysis results in greatly increased cathodic currents in CVs of Ru(NH3)(,3+ (Fig.
3.5), showing that the previously passivating PAH/d-PAA ﬁlms become permeable to
Ru(NH3)(,-3+. Several pieces of evidence suggest that in the case of propyl-derivatized
PAA (and to some extent other cases also) much of the cathodic current is due to

adsorbed Ru(NH3)(,3+. First, the cathodic peak is shifted to more negative potentials.

57

20

 

15‘

 

        

-_l_.__._—-————-_-:";'

Pr, Bu, mBu, Ben

 

 

 

 

 

 

 

 

Current (11A)

 

 

 

 

 

 

 

 

0.5 0.4 0.3 0.2 0.1 0.0 -0.1
Potential (V vs AglAgCl)

Figure 3.4: Cyclic voltammograms of 1 mM Ru(NH3)6Cl3 (top) and 1 mM
K3Fe(CN)6 (bottom) in 0.5 M Na2SO4 at Au electrodes coated with different
PAH/d-PAA ﬁlms (six bilayers). The inset shows enlarged CVs obtained in
the presence (solid line) and absence (dotted line) of Ru(NH;1)63+ at a Au
electrode coated with a six bilayer PAH/Bu-PAA ﬁlm. The scan rate in all

CVs was 0.1 V/s, and electrode areas were 0.1 cm2

58

This likely occurs because the presence of negatively charged -COO' groups stabilizes
the oxidized form of the couple.28 When 0.1 M tetraethylammonium chloride (NEtaCl)
was used as supporting electrolyte rather than 0.5 M Na2804, the cathodic peak shifted
another 200 mV negative and the cathodic peak current increased by ca. 40%. This
occurs because ML; cations are large enough that they do not compete with Ru(NH3)(,3+
for adsorption sites. Second, the anodic peak is very small in several cases. The absence
of the anodic peak suggests that Ru(NH3)(,2+ produced during the cathodic scan is
displaced from the ﬁlm by the more highly charged Ru(NI-13)(,3 +. Third, the cathodic peak
currents at electrodes coated with hydrolyzed ethyl-, propyl-, and butyl-derivatized ﬁlms
were higher than that at bare gold because of a high concentration of Ru(NH3)63+ in the
ﬁlms. Thus the CVs suggest that there is a high density of -C00’ groups in hydrolyzed
ﬁlms and that these groups are capable of binding ions. Cathodic peak currents at
PAH/Et(67)-PAA-coated electrodes are 2-5 times larger than those at PAH/Et(19)-PAA-
coated electrodes. This suggests that the amount of adsorption can be varied by
controlling the extent of derivatization prior to hydrolysis. These hydrolyzed PAH/d-
PAA ﬁhns behave somewhat like Naﬁon or Naﬁon—impregnated membranes where
cations residing in the solution phase exchange with cations electrostatically bound
to -SO3' sites in the membrane.”30
Peak currents in CVs of Fe(CN)63' at electrodes coated with hydrolyzed PAH/d-PAA
ﬁlms are about 10-fold lower than those for Ru(NH3)(,3+ at corresponding electrodes
(Fig.3.5, inset). The high density of negatively charged -COO' groups likely hinders
access of Fe(CN)(,3' to the electrode surface. The negative charges also prevent the

adsorption of Fe(CN)(,3'.

59

30

 

 

20‘

 

  
  

 

 

.3
0

Current (11A)

0

    

 

: 01°: - PAH/Et(19)-PAA
. ‘ . PAH/Et(67)-PAA

. PAH/Pr-PAA

- PAH/Bu-PAA

bare Au

 

 

 

I I I

 

I

0.0 -0.1 -0.2 -0.3 -0.4 -0.5
Potential (V vs AgIAgCI)

Figure 3.5: Cyclic voltammograms of 1 mM Ru(NH3)6C13 in 0.5 M Na2S04
at Au electrodes coated with cross-linked, hydrolyzed PAI-I/d-PAA ﬁlms (six
bilayers). The inset compares CVs of 1 mM Ru(NH)6Cl3 and 1 mM
K3Fe(CN)6 at 3 Au electrode coated with cross-linked, hydrolyzed PAH/Pr-

PAA. The scan rate in all CVs was 0.1 V/s, and electrode areas were 0.1 cm2

60

3.3.4 Ion Detection Using Hydrolyzed Films. CVs of Ru(NH3)(,3+ at electrodes coated
with hydrolyzed PAH/d-PAA ﬁhns are affected by low concentrations of other divalent
and trivalent metal cations in solution. These effects might allow PAH/d-PAA electrodes
to function as sensors for these ions. Figure 3.6 shows CVs of Ru(NH3)63+ that were
obtained at an electrode coated with PAH/Pr-PAA that was cross-linked and
subsequently hydrolyzed. The cathodic peak current and position are affected by
concentrations of Ca2+ as low as 10'5 M, even though the measurement is being made in
the presence of 0.1 M NaCl. With further addition of Ca2+, the peak current continues to
drop and peak positions become even more positive. This likely occurs because Ca2+
displaces some of the adsorbed Ru(NH3)(,3+ and blocks access to the underlying gold
surface. Peak positions shift to more positive potentials as the presence of Ca2+ makes
the enviromnent around Ru(NH3)¢=,3+ less negatively charged. After the ﬁhn is rinsed with
1 mM HCl for 5 min (to remove Ca2+ from the ﬁlm) and then with 1 mM NaOH for 5
min (to deprotonate -COOH groups), the reduction current of Ru(NH3)(,3+ in the absence
of Ca2+ returns to within 5% of its original value. The electrode responds to Mg2+ in a
similar way, and selectivities need to be ﬁirther investigated. We also added AlCl3 to a 1
mM Ru(NH3)(,3+, 0.1 M NaCl solution and obtained similar results. However, the ﬁhn
does not show a higher sensitivity to Al“ than to Ca“. One might expect the trivalent

cation to bind more strongly to -C00' groups, but at pH 6, most aluminum ions exist as

A1(0H)2*.3‘

61

 

N
O
I

_x
O
1

Current (11A)

 

 

 

 

 

_ -O.2 -0.3 -O.4 -0.5 -0.6
Potential (V vs AgIAgCI)

Figure 3.6: Cyclic voltammograms of 1 mM Ru(NH3)6Cl3 in 0.1 M
NaCl containing different concentration of Ca2+ at a Au electrode
coated with six bilayers of cross-linked, hydrolyzed PAH/Pr-PAA.

The scan rate was 0.1 V/s, and the electrode area was 0.1 cm2

62

3.4 Conclusions

Variation of the ester functionalities in PAH/d-PAA ﬁlms along with subsequent
cross-linking and hydrolysis afford control over coating hydrophobicity and charge
density. Control of these properties will be important in future applications of layered
polyelectrolytes such as ion—separation membranes, sensors, and protective coatings.
Esteriﬁcation with hydrophobic groups can result in advancing water contact angles at
least as high as 101°. Cross-linking through amidation strengthens PAH/d-PAA ﬁlms
and allows subsequent modiﬁcations. Hydrolysis of ester groups in cross-linked ﬁhns
yields a high density of -C00’ groups that renders these coatings more permeable to
positively charged Ru(N H 3)(,3 + than to Fe(CN)(,3'. Binding of Ca2+ to hydrolyzed
PAH/Pr-PAA occurs at concentrations as low as 10'5 M and results in decreases in peak
currents in CVs of Ru(NH3)63+. Such effects may prove useful in ion-sensing
applications. Taken together, the results in this chapter show the potential of PAH/d-

PAA deposition for producing coatings with properties tailored for speciﬁc applications.

63

3.5 References

10

ll

12

l3

14

15

16

17

Caruso, F .; Rodda, E.; Furlong, D. N.; Niikura, K.; Okahata, Y. Anal. Chem. 1997, 69,
2043-2049.

Caruso, F .; Niikura, .K.; Furlong, D. N.; Okahata, Y. Langmuir 1997, 13, 3427-3433.
Shipway, A. N.; Lahav, M.; Blonder, R.; Willner, 1. Chem. Mater. 1999, 11, 13-15.
Krasemann, L.; Tieke, B. Langmuir 2000, 16, 287-290.

Harris, J. J.; Stair, J. L.; Bruening, M. L. Chem. Mater. 2000, 12, 1941-1946.
Ackem, F. v.; Krasemann, L.; Tieke, B. Thin Solid Films 1998, 32 7-329, 762-766.

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

Kotov, N. A.; Magonov, S.; Tropsha, E. Chem. Mater. 1998, 10, 886-895.
Leviisalmi, J .-M.; McCarthy, T. J. Macromolecules 1997, 30, 1752-1757.
Lukas, J .; Richau, K.; Schwarz, H.-H.; Paul, D. J. Membr. Sci. 1997, 131, 39-47.
Stair, J. L.; Harris, J. J.; Bruening, M. L. Chem. Mater. 2001, 13, 2641-2648.
Graul, T. W.; Schlenoff, J. B. Anal. Chem. 1999, 71, 4007-4013.
Liu, Y. Q.; Ma, H.; Jen, A. K. Y. Chem. Commun. 1998, 2747-2748.

Wu, A.; Yoo, D.; Lee, J.-K.; Rubner, M. F. J. Am. Chem. Soc. 1999, 121, 4883-
4891.

Dai, J .; Sullivan, D. M.; Bruening, M. L. Ind. Eng. Chem. Res. 2000, 39, 3528-
3535.

Cochin, D.; Laschewsky, A. Macromol. Chem. Phys. 1999, 200, 609-615.

Chen, J.; Huang, L.; Ying, L.; Luo, G.; Zhao, X.; Cao, W. Langmuir 1999, 15,
7208-7212.

64

18

19

20

21

22

23

24

25

26

27

28

29

30

31

Harris, J. J.; DeRose, P. M.; Bruening, M. L. J. Am. Chem. Soc. 1999, 121, 1978-
1979.

Dubas, S. T.; Schlenoff, J. B. Macromolecules 1999, 32, 8153-8160.
Schlenoff, J. B.; Ly, H.; Li, M. J. Am. Chem. Soc. 1998, 120, 7626-7634.
Lowack, K.; Helm, C. A. Macromolecules 1998, 31 , 823-833.

Lvov, Y. M.; Lu, 2.; Schenkman, J. B.; Zu, X.; Rusling, J. F. J. Am. Chem. Soc.
1998, 120, 4073-4080.

Lvov, Y.; Decher, G.; Mdhwald, H. Langmuir 1993, 9, 481-486.
Laurent, D.; Schlenoff, J. B. Langmuir 1997, 13, 1552-1557.

Cochin, D.; Passmann, M.; Wilbert, G.; Zentel, R.; Wischerhoff, E.; Laschewsky,
A. Macromolecules 1997, 30, 4775-4779.

Bryand, W. M. D.; Smith, D. M. J. Am. Chem. Soc. 1936, 58, 1014-1017.

Ldsche, M.; Schmitt, J .; Decher, G.; Bouwman, W. G.; Kjaer, K. Macromolecules
1998, 31, 8893-8906.

Derrnody, D. L.; Peez, R. F.; Bergbreiter, D. E.; Crooks, R. M. Langmuir 1999, 15,
885-890.

Liu, C.; Martin, C. R. J. Electrochem. Soc. 1990, 13 7, 510-515.

Yeager, H. L. In Perﬂuorinated Ionomer Membranes; Yeager, H. L., Eisenberg, A.,
Eds; ACS Symposium Series No. 180; American Chemical Society: Washington,
DC, 1982, pp 41-63.

Porterﬁeld, W. W. Inorganic Chemistry, a United Approach; Addison-Wesley:
Reading, MA, 1984, page 229.

65

Chapter 4

Controlling Ion Transport through Multilayer Polyelectrolyte Membranes by
Derivatization with Photolabile Functional Groups

4.1 Introduction

Membrane-based separations are attractive because of their operational simplicity and
low energy costs, but limitations in membrane ﬂux and/or selectivity restrict many
applications of this technology.1 The most common method for increasing ﬂux is to
decrease membrane thickness, and thus commercial membranes are generally prepared
with a thin selective skin on a porous support that provides mechanical strengthz’3
Methods for forming thin membrane skins include phase inversion}5 interfacial
polymerization,6 plasma grafting,7’8 casting,9 and even deposition of ﬁlms from the
air/water interface. 10 In spite of the success of these methods, synthesizing ultrathin,
selective membranes is an ongoing challenge.

Multilayer polyelectrolyte ﬁlms (MPFs) are potentially attractive materials for
synthetic membrane skins because of their convenient synthesis and versatility.

Synthesis of these ﬁlms, ﬁrst demonstrated by Decher and coworkers,”‘12

simply
involves sequential adsorption of polycations and polyanions on an initially charged
surface (see chapter 1 of this thesis). The layer-by-layer deposition procedure affords
control over ﬁlm thickness on the nanometer scale, and the minimal total thickness of
these ﬁlms should allow high ﬂux through polyelectrolyte multilayers deposited on
highly permeable supports. Moreover, selectivity of polyelectrolyte membranes may be

tailored through selection of constituent polyelectrolytes as well as post-deposition cross-

linking of ﬁlms.13 Separation applications of polyelectrolyte membranes recently

66

explored by several research groups include gas separations,'4"7 selective pervaporation
from water/organic solvent mixtures,'8"° and ion separations.'3’2°‘21

This thesis focuses on understanding and controlling ion ﬂux through multilayer
polyelectrolyte membranes (MPMs) via control of the composition of these ﬁlms. In
previous studies of ion transport through MPMs,13’20’21 ﬂuxes of monovalent ions were
higher than those of divalent ions. Krasemann and Tieke suggested that Donnan
exclusion resulting from ﬁxed charge in the membrane is responsible for this
selectivity.20 If Donnan exclusion is the dominant factor in effecting ion-transport
selectivity in MPMs, then introduction of more net ﬁxed charge into these membranes
should lead to higher monovalent/divalent ion-transport selectivities. The purpose of this
work is to explore the relationship between ion-transport selectivity and net ﬁxed-charge
density in MPMs. In a similar work from our group, Balanchandra showed that
templating with Cu2+ also provides a convenient method for introducing net ﬁxed charge
and can yield Cl'/SOaz' selectivities as high as 600.22

Although MPFs form due to electrostatic interactions, in the bulk of the ﬁlm, charges
on polycations are electrically compensated by charges on polyanions and vise versa.23’24
Schlenoff demonstrated that introduction of net ﬁxed charge into MPFs requires post-
deposition modiﬁcation,23 and several studies showed that MPFs can be modiﬁed after
deposition to alter properties such as conductivity”26 or stability.“28 We chose to use
partially derivatized poly(acrylic acid) (PAA) and poly(allylamine hydrochloride) (PAH)
chains as model constituents for controlling charge in MPFs. To introduce net charge

into ﬁlms, we partially derivatized PAA and PAH with photolabile 2-nitrobenzyl (NE)

and 2-nitrobenzyloxycarbonyl (N BOC) groups (scheme 4.1). The underivatized —C00'

67

groups in PAA and —NH3+ groups in PAH still allow for ﬁlm formation through
electrostatic adsorption, while UV irradiation of derivatized PAA/PAH ﬁlms cleaves
ester or carbamate bonds and liberates free acid or amine groups. These groups can then
protonate/deprotonate to yield net negative (or positive) ﬁxed charges in membranes (Fig.
4.1). The newly formed charges are not electrically compensated by adjacent polyions
but by oppositely charged ions from solution.23 With the introduction of net charge into
the membrane, Donnan exclusion should be more pronounced, and monovalent/divalent

ion transport selectivity should increase.

 

 

CH28I
N02 x 1-X
n DBU (9:0 COOH CHO
; O hv 11 NO
COOH In DMSO, 3h at 45 °C CH2 —> +
N02 COOH
N02 0 WX
QCHzO—C—Cl (EH2 1c”?
M NH ”Hz CHO
can : c':=o M M No
——-—>
' 2 base O 9H2 + "' COz
NHZHCI I NH2
CH2
N02
Scheme 4.1

68

Porous alumina support

8880888888

1 immersion
( l in NBPAA

 

nnse,
(2) Immersion
in PAH

 

 

Repetition of step (1) and (2)
Produces multilayers

 

 

 

1. 111112. base

 

Figure 4.1: Schematic drawing of the formation of a multilayer

polyelectrolyte membrane and subsequent introduction of net, ﬁxed

charge via photolysis. R represents the 2-nitrobenzyl (NB) photolabile

functional group. Interweaving of layers is not shown for ﬁgure clarity.

69

In principle, the simplest way to control net ﬁxed charge in PAA/PAH membranes is
to vary the pH of deposition solutions. Rubner showed that deposition of PAA/PAH
ﬁlms at low pH and subsequent immersion in higher pH solutions create ion-exchange
sites in the ﬁlm.29 However, when we attempted depositing PAA/PAH membranes at pH
2.5 and running transport experiments at pH 5-6, we did not observe any increase in
selectivity relative to PAA/PAH membranes prepared at pH 5.0. The use of deposition
pH to control ﬁxed charge could result in morphological changes in ﬁlm structure that
may decrease selectivity.”3 '

In contrast, introduction of ﬁxed charge using photocleavable groups increases the
selectivity of PAA/PAH ﬁlms by an order of magnitude. The use of derivatization with
photocleavable groups also allows knowledge of and control over the net, ﬁxed charge in
the membrane. Thus, by changing the extent of derivatization, one can control charge
density, and hence, tailor ion-transport selectivity. With an estimate of charge density
obtained from the extent of derivatization, we utilized a simple model to simulate ion

transport through these membranes. The model suggests that both Donnan exclusion and

selective diffusion play important roles in effecting selective transport.

4.2 Experimental Section

4.2.1 Materials. PAA (Mv = 450,000), PAH (MW = 70,000), 2-nitrobenzyl bromide
(99%), 2-nitrobenzyl alcohol (99%), phosgene solution (~20% in toluene), 1,8-
diazabicyclo[5.4.0]undec-7-ene (DBU, 98%), and dimethyl sulfoxide (DMSO, reagent)
were purchased from Aldrich and used as received. Nitrobenzylchloroforrnate was

synthesized according to a literature procedure.32 Deionized water (18 MQ-cm, Milli-Q)

70

was used in preparation of aqueous solutions. Membrane supports (Whatman AnodiscTM
membrane ﬁlters with a nominal surface pore diameter of 0.02 pm) were purchased from
Thomas Scientiﬁc.

4.2.2 Synthesis of 2-nitrobenzyl-derivatized PAA (NBPAA). PAA (720 mg, 10 mmol
with respect to the repeating unit) was dissolved in 20 mL of DMSO (sonicate for 1h, stir
for 2 h at 45 °C to get complete dissolution), and DBU (1.52 g, 10 mmol, in 5 mL DMSO)
was added dropwise to this solution while stirring. 1.51 g (7 mmol) 2-nitrobenzyl
bromide was then added. The mixture was stirred at 45 °C for 3 h after which 1 mL of
acetic acid was added to neutralize DBU, and the solution was stirred for an additional 15
min. Derivatized polymer was then precipitated upon dropwise addition of the mixture
into 1500 mL of pH 3 water. The precipitate was collected by ﬁltration, washed
thoroughly in water while stirring, and again collected by ﬁltration. The polymer was
dried under vacuum at 60 °C for 48 h, and 1.1 g of polymer was obtained (78% yield).

H1 NMR (d-DMSO as solvent) showed that ~5 0% of the —COOH groups in PAA were
derivatized by NB groups (Fig. 4.2). The 23% and 65% NB-derivatized PAA were
synthesized in a similar way (56% and 84% yield), except that the quantity of 2-
nitrobenzyl bromide added was 0.86 g (4 mmol) and 1.73 g (8 mmol) for 23% and 65%
derivatization, respectively. 23% NBPAA was precipitated into and washed with ethyl
acetate-saturated pH 3 water.

4.2.3 Synthesis of 2-nitrobenzyloxycarbonyl-derivatized PAH (NBPAH). PAH (1.55
g, 17 mmol with respect to the repeating unit) was dissolved in 100 mL of water, and the
solution was placed in an ice-bath. After adjusting the pH of the solution to 3.0 with HCl,

1.84 g (8.5 mmol) of 2-nitrobenzylchloroformate in 5 mL dry THF was added

71

 

 

 

 

 

 

Figure 4.2: lH NMR spectra of PAA (top, d-DMSO) and PAH (bottom, d-
DMSO+trace of D20) that were 50% derivatized with 2-nitrobenzyl and 2-
nitrobenzyloxycarbonyl groups, respectively. The percentage derivatization
was calculated from the ratio of the integrated signals due to the benzylic
protons and the methylene protons on the polymer backbone. The chemical

structures on the right indicate the proton assignments.

72

dropwise to the solution over a 2 h period while maintaining the solution pH between 3
and 4 through addition of 1 M NaOH. The reaction was then allowed to proceed at room
temperature for another 2 h at pH 4. The suspension was collected by centrifugation.
The polymer was redissolved in DMSO, precipitated upon addition to a 20% ethyl
acetate/80% 2-propanol (v/v) mixture, and dried under vacuum at 50 °C for 72 h to yield
1.77 g of product (71% yield). 1H NMR showed that ~50% of the amine groups in PAH
were derivatized (Fig. 4.2).

4.2.4 Film synthesis and characterization. Before synthesizing MPMs on porous
alumina supports, we prepared ﬁlms on Al-coated wafers (200 nm of Al sputtered on
Si(100)) to determine the optimum conditions for deposition. The Al wafer was
UV/ozone cleaned for 15 min before immersion in a solution containing a polyanion
(PAA or NBPAA). After a 10 min immersion, the wafer was rinsed with water for 1 min,
immersed in a solution containing a polycation (PAH or NBPAH) for 10 min, and rinsed
with water for 1 min to form the ﬁrst bilayer. This procedure was repeated until the
desired number of bilayers was deposited. (Drying with N2 occurred only after
deposition of the entire ﬁlm.) The solutions used to form ﬁlms were: PAA (0.01 M, pH
5.0), PAH (0.02 M, pH 5.0), NBPAA (200 mg in 20 mL DMSO + 80 mL MeOH + 80 mg
KOH), NBPAH (0.1 mM, pH 5.0). Concentrations of polymers are given with respect to
the repeating unit, and pH values were adjusted with 0.1 M HCl or 0.1 M NaOH. Film
thickness was measured with a rotating analyzer ellipsometer (J. A. Woollam model M-
44), assuming a ﬁlm refractive index of 1.5. Reﬂectance FTIR spectra were obtained

with a Nicolet Magma-560 FTIR spectrometer using a Pike grazing angle attachment (80°

73

angle of incidence). The spectrometer employs a MCT detector. Films were also
deposited on porous alumina using the above procedure.
4.2.5 Photolysis. Polyelectrolyte ﬁlms and membranes were irradiated in air with a
medium-pressure mercury arc lamp (Hanovia, 450 W) without a ﬁlter at a distance of 30
cm from the source. The irradiation time was 60 min for all samples. Samples were
dried with N2 purging for 2 h before photolysis.
4.2.6 Transport studies. The diffusion dialysis apparatus consists of two cells separated
by the membrane, which is contained in the middle of a 2.5 cm-long neck. The source
cell was ﬁlled with 90 mL of a 0.1 M solution of the desired salt (NaCl, Na2S04, or
M gCl2), while the receiving cell initially contained 90 mL of deionized water. The
measured pH values of the unbuffered 0.1 M salt solutions were 5.3 (NaCl), 5.6 (Na2S04),
and 5.0 (MgCl2). The area of the membrane exposed to the solution was 2 cm2. The
concentration of salt in the receiving cell was monitored at 5 min intervals for 45 min
using a conductivity meter (Orion model 115). Solutions in both cells were vigorously
stirred to minimize concentration polarization at the membrane surface. Membranes
were soaked in water for at least 12 h before use, and three different membranes were
prepared and tested for each type of ﬁlm.

The conductivity in the receiving cell was transformed to concentration using a
calibration curve. The ﬂux (J) and selectivity (a) of ion transport were then calculated

according to equations 1 and 2

AC V
=._.._ 1
J At A ()

J
6112:: (2)

74

where [Z—f is the concentration change with time calculated from the slope of a plot of

concentration versus time, V is the volume of solution in the receiving cell, A is the

exposed area of the membrane, and the subscripts 1 and 2 represent two different ions.

4.3 Results and Discussion

4.3.1 Derivatization of PAA and PAH. The 2-nitrobenzyl group and its analogues have
been intensively employed as photoremovable protecting groups in a variety of synthetic
strategies.”‘34 These photolabile groups were also, though less frequently, used to
modify polymers.”37 Derivatization of PAA with 2-nitrobenzyl bromide occurs under
mild conditions using a method developed by Nishikubo and coworkers for modiﬁcation
of poly(methacrylic acid).36 Derivatization of PAH, however, is more challenging
because both PAH and deprotonated poly(allyl amine) are not very soluble in most
organic solvents. To overcome this problem, we reacted PAH with 2-
nitrobenzylchloroformate in aqueous solution while controlling pH. This procedure is
similar to the common practice in peptide synthesis of protecting amines with NBOC
groups using an aqueous reaction.”38 Amines react with 2-nitrobenzylchloroformate at a
much faster rate than does water, and thus hydrolysis of 2-nitrobenzylchloroformate is
not a problem. In fact at high pH, this reaction occurs so rapidly that polymer
immediately begins to precipitate upon addition of nitrobenzylchloroforrnate. To slow
down the reaction and achieve uniform derivatization without precipitation, we ran the
reaction at a pH of about 4 so most of the amine groups would be protonated.39

4.3.2 Film formation and photochemistry. We prepared polyelectrolyte solutions in

water in all cases where the polyelectrolyte was sufﬁciently soluble. Using a deposition

75

pH of 5, we were able to obtain homogeneous ﬁlms with reasonable thicknesses (Table

29.40

4.1 on page 80 ). NBPAH was only slightly soluble in water, but concentrations of
10’4 M were sufﬁcient for ﬁlm formation.41 Derivatized PAA, however, did not dissolve
in water, so we employed a DMSO/methanol cosolvent for these polymers and added a
stoichiometric amount of KOH to deprotonate —CO0H groups.41413 To assess whether
changing deposition solvent affects ﬁlm properties, we also prepared underivatized
PAA/PAH ﬁlms using PAA dissolved in DMSO/methanol containing KOH. These ﬁlms
have basically the same IR spectra and ion-transport selectivities as ﬁlms prepared only
in water, but their thickness is ~30% less than for the entirely water-based deposition.
These experiments indicate that the enhanced selectivity (vide infra) of NBPAA/PAH
membranes is not due to deposition from an organic solvent.

Figure 4.3 shows the reﬂectance F TIR spectra of a 105-bilayer 50%NBPAA/PAH
ﬁlm before and after UV irradiation. (The addition of an extra half bilayer results in
NBPAA being the top layer in these ﬁlms.) The strong absorbance at 1740 cm'1 is
primarily due to the ester carbonyl groups of derivatized PAA, but underivatized,
protonated —CO0H groups likely also contribute to this peak. After 60 min of UV
irradiation, peaks due to —N02 groups (1530 cm'I and 1345 cm'1 ) completely
disappeared, showing that photolysis was complete. The peak at 1720 cm'1 after
irradiation is due largely to ~CO0H groups produced by photolysis (Scheme 1) and thus
decreases signiﬁcantly (~60%) after deprotonation in pH 9.2 buffer. We also see an
increased —C00' absorbance at 1575 cm‘l aﬁer deprotonation. Similar results were

obtained for PAA/50%NBPAH ﬁlms (Fig.4.4). The carbamate carbonyl absorbance at

1716 cm’1 and peaks due to —N02 groups vanished after 60 min of UV irradiation,

76

 

0.01

 

 

 

/ before irradiation

 

 

Absorbance

 

/after irradiation

 
 
  

 

irradiated then depronated

I I I I I I

1800 1700 1600 1500 1400 1300 1200 1100

    

 

 

 

;—

Wavenumbers (cm'l)

Figure 4.3: Reﬂection F TIR spectra of a 105-bilayer 50%NBPAA/PAH
ﬁlm on an Al-coated wafer before (a) and after (b) UV irradiation.
Spectrum (c) is that of the irradiated ﬁlm after a 30 min immersion in pH

9.2 buffer solution and rinsing with ethanol.

77

 

0.01

Before irradiation

/

 

Absorbance

    

\ After irradiation

f

1800 1700 1600 1500 1400 1300 1200

 

 

 

Wavenumbers (cm'l)
Figure 4.4: Reﬂection FTIR spectra of a 105-bilayer PAA/50%NBPAH

ﬁlm on an Al-coated wafer before (a) and after (b) UV irradiation.

78

indicating complete photolysis. The mechanism of photocleavage of NB or NBOC

37,38.44,45 . . . . .
and thus rt 1s efﬁcrent 1n SOIld

groups proceeds via an intramolecular pathway,
ﬁlms as well as in solution.

4.3.3 Ion-transport studies. Table 4.1 summarizes ﬂuxes and selectivities for several
salts in diffusion dialysis experiments with PAA/PAH and derivatized PAA/PAH
membranes. The table shows three interesting trends for 105-bilayer membranes: (1)
photolyzed NBPAA/PAH membranes have 10- to 20-fold higher Cl'/S042' selectivities
than corresponding underivatized PAA/PAH membranes; (2) Cl'/S042' selectivity for
photolyzed 105-bilayer NBPAA/PAH membranes increases with the extent of
derivatization; and (3) photolyzed PAA/NBPAH membranes show increased cation
selectivities compared to PAA/PAH membranes (3-4 times) and minimal anion
selectivities. NaCl ﬂuxes are about the same for all coated membranes. Thus high
selectivity can be achieved without a signiﬁcant decrease in ﬂux.

As Figure 4.1 schematically shows, the difference between underivatized PAA/PAH
membranes and UV-irradiated NBPAA/PAH membranes is that the latter contain net,
ﬁxed negative charge in the bulk of the ﬁlm. Photolysis may also increase net charge at
the surface of these ﬁlms. Higher percentages of PAA derivatization lead to more
negative ﬁxed charge in membranes, and hence higher anion-transport selectivity.
Similarly, the introduction of net positive charges using photolyzed PAA/NBPAH
increases Donnan exclusion of cations, and hence monovalent/divalent cation transport
selectivity. However, this selectivity enhancement is not as signiﬁcant as that displayed

for anions by NBPAA/PAH membranes. A possible reason for the lower enhancement

may be incomplete conversion of carbamates to amines during photocleavage. Although

79

Table 4.1. Fluxes and selectivities through membranes prepared from PAA, PAH,

NBPAA, and NBPAH deposited on porous alumina.

 

Type of membrane” Thickness Flux (X109 mole/cm2 s) selectivityi
(131)“ NaCl Na2SO4 MgCl2 cr/soi' Na+/Mg2+
Bare porous alumina 43: 2 32:1 32:1 13:01 1.3:0.1
PAA/PAH (10) 420:16 22:1 3.6:O.3 19:03 60:08 12:1
50%NBPAA/PAH (10) 260:20 18:1 0.15:0.01 2.0:0.l 120:10 9.1:0.5
PAA/50%NBPAH (10) 400:35 14:2 15:1 0.35 :0.1 10:01 43:7
50%NBPAA/50%NBPAH (10) 320:20 22:2 18:03 20:02 12:1 1 1:1
PAA/PAH (10.5) 450:30 20:1 22:02 21:02 9.3:0.6 10:1
23%NBPAA/PAH (10.5) 330:15 18:2 0.19:0.02 2.1:0.2 100:4 8.7:0.1
50%NBPAA/PAH (10.5) 270:15 16:2 0.11:0.02 2.0:0.l 150:10 8.1:1.1
65%NBPAA/PAH (10.5) 275:15 17:2 0.10:0.01 2.1:0.2 170:20 8.0:0.7
PAA/50%NBPAH (10.5) 410:35 15:2 13:1 04:01 12:01 38:8

" The number in parentheses represents the number of bilayers deposited.

1’ The percentage values represent the degree of derivatization of the particular polymer

as calculated from NMR spectra.

‘ Post-UV irradiation thickness of ﬁlms prepared on Al-coated wafers. Thicknesses

before irradiation were 10-25% higher.

‘1 Selectivities were calculated as the average of selectivity values from different

membranes and not simply as a ratio of average ﬂux values.

80

removal of the nitrobenzyloxycarbonyl group is quantitative, side reactions may result in
lower yields of amines. Patchornik and coworkers reported a 25% yield of amines when
photocleaving nitrobenzyloxycarbonyl-protected amino acids.38This may also explain
why the Cl'/SO42' selectivity of 10-bilayer 50%NBPAA/50%NBPAH membranes is
twice as large as that of 10-bilayer PAA/PAH membranes. If derivatized PAH introduces
less charge than does derivatized PAA, 50%NBPAA/50%NBPAH membranes would
have a net negative charge and hence an increased Cl'/SOaz' selectivity relative to
PAA/PAH.

Because Donnan exclusion can result from charges at both the surface and the interior
of a ﬁlm, changing the outer layer of a membrane from a polyanion to a polycation
should affect selectivity.'3’21‘23’41 Table 4.1 shows some effect of changing the sign of
surface charge, but the change in selectivity is not very large. Capping a membrane with
a PAA or NBPAA top layer (10.5 bilayer membranes) increases anion selectivity by 20-
50%, while capping with PAH (or NBPAH) (10 bilayer membranes) may slightly
increase cation selectivity. In previous studies, however, Cl'/8042' selectivity increased
by as much as two orders of magnitude when the outer layer of the ﬁlm was a polyanion
rather than a polycation.l3 The reason for the small effect in the present case is probably
that ﬁlms were prepared in the absence of supporting electrolyte. Preparation under these
conditions reduces the net charge on the membrane surface,23'46’47 and hence the bulk of
the membrane will play a much larger role in determining selectivity.

4.3.4 Modeling of Ion Transport. To estimate the contributions of Donnan exclusion
and selective diffusion to ion-transport selectivity, we utilized a simple model for ion

- - . - 4 ,4 0, 0
transport based on prev1ous Simulations of ion-exchange membranes 8 9 and MPMs.2 5

81

The model includes Donnan equilibria at the feed/membrane interface and at the
alumina/polyelectrolyte ﬁlm boundary (F ig.4.5) as well as Nemst-Planck transport within
the membrane. Figure 4.5 portrays the concentration proﬁle in the membrane. We did
not include a charged surface layer because reversing the charge in the top layer of the
membrane had only a small effect on selectivity. To calculate diffusion coefﬁcients in

the membrane, we performed the procedure given in the appendix.

porous support ' __Cs
. e

Concentration ——9
Source phase

Receiving phase

 

Donnan equilibia

 

Figure 4.5: Schematic diagram of a model for ion transport through a multilayer
polyelectrolyte membrane. The line represents the concentration proﬁle of the

excluded ion. Symbols are deﬁned in the text.

82

 

cal

NE

111

C01

111

1111

1151

la

The transport model requires knowledge of the charge density, C1, in MPMs. For our
calculation, we assumed that all ﬁxed charges resulted from photolysis of NBPAA or
NBPAH. Pure PAA contains ~16 M -CO0H groups. Assuming an equal volume
fraction for PAA and PAH components, and considering 50% derivatization, the
concentration of net charge in 50%NBPAA/PAH ﬁlms would be approximately 4M.
With this estimation, we calculated diffusion coefﬁcients in different membranes along
with values for diffusion, at), and electrostatic (Donnan), org, selectivities. The results are

listed in Table 4.2.

Table 4.2: Estimated charge densities, diffusion coefﬁcients, Donnan selectivities, and

diffusional selectivities for Derivatized PAA/PAH membranes.

Membrane 23%NBPAA/PAH 50%NBPAA/PAH 65%NBPAA/PAH PAA/5 0%NBPAHb

z.c. (M) -1.84 —4.0 -4.6 1.0
D or 8042' Ct 8042' 0' so? Na+ Mg2+
(x1010 cmZ/s)
135 5 204 12 255 15 65 3.5
011)" 27 17 17 19
aE “ 3.7 8.8 10 2.3

" See equation 8 in the Appendix for the deﬁnition of 019, 015, and AC.
b To obtain ﬁxed charge density in this ﬁlm, the initial estimate of amine concentration
was divided by 4 to reﬂect a low yield of amine groups from photocleavage

(reference 38).

83

Calculations with this simple model suggest that ion-transport selectivity is only
partially due to Donnan exclusion, and that diffusive selectivity is a larger factor than
Donnan exclusion. However, the contribution of Donnan exclusion to selectivity does
increase with increasing charge density in the membrane as would be expected.
Diffusive selectivity is consistent with the fact that the calculated diffusion coefﬁcients
(ranging from 5.2x10‘IO to 2.5x10'8 cmZ/s) in MPFs are very small.

The slow difﬁision and the difference in diffusion coefﬁcients could result from steric
hindrance and/or electrostatic interactions. Farhat and Schlenoff suggest that transport
through polyelectrolyte multilayers occurs by hopping of ions between transient ion-
exchange sites.47 Because more highly charged ions require more ion-exchange sites,
their movement through MPMs may be slower than that of monovalent ions. Yeager
reported that the diffusion coefﬁcient of Cl' could be 14 times larger than that of 8042' in
perﬂuorinated carboxylate membranes.5 ' Even for similarly charged ions such as Na+
and Cs+, diffusion coefﬁcients can differ by a factor of four in NaﬁonTM membranes.52
By modeling Donnan dialysis data for ion transport through cation-exchange membranes,
Miyoshi obtained effective difﬁision coefﬁcients of 6.33x 10'8 and 1.34x10'8 cmZ/s, for
Na+ and M g”, respectively.53 Thus the low diffusion coefﬁcients and the diffusional
selectivities resulting from these simple simulations are within reasonable limits.

The limitations of the ion-transport model must be kept in mind, however. The
simulation does not consider activity coefﬁcients and possible differences in solubilities
of different ions in MPMs.54’55 This could account for the fact that calculated diffusivity
selectivities in derivatized membranes are about 2-fold larger than total selectivities for

pure PAA/PAH membranes. Additionally, charge densities are only approximate.

84

However, even with these limitations, the modeling results and data for transport through
underivatized membranes suggest that Donnan exclusion is only partially responsible for
selectivities. Future work on determining partition coefﬁcients would be helpful for a

better understanding of transport, but such studies will be challenging given the minimal

thickness of MPFs.

4.4 Conclusions

Introduction of net, ﬁxed charge into MPMs through derivatization and photolysis is
straightforward and quantitatively controllable. The presence of ion-exchange sites in the
membrane signiﬁcantly increases ion-transport selectivities, and this increase correlates
with the charge density introduced into the membrane. Simulations suggest that the
enhancement of selectivity is due to both Donnan exclusion and diffusivity differences
among ions. In addition, the methods for derivatizing poly(acrylic acid) and poly(allyl

amine) may provide a general way to modify the properties of MPFs.

85

4.5 Appendix

To calculate diffusion coefﬁcients in MPMs, we applied the following procedure. The
ion concentration on the left side of the porous support/polyelectrolyte membrane
interface, Cf, was calculated using equation 3

J .
cf 2 ‘Iroatcd X C‘S (3)

bare
where med and Jim. are ﬂuxes through coated and bare porous alumina supports,
respectively, and C is is the concentration of ion in the source-phase solution (0.1M).
This equation assumes that ion concentrations in the receiving phase are negligible and
that ﬂux through the alumina is proportional to a linear concentration gradient. The
concentration of excluded ion just inside the left end of the coated polyelectrolyte ﬁlm
(Figure 4.5), C "'L , was calculated assuming Donnan equilibrium. For a salt AXBy with
counterion charge ZA and excluded ion charge 23, C I,“ can be expressed by equation 4

71

ZBCBHL + Zr Cr Z
ZtCﬁ

 

CE" = C4 (4)

where 2,. and C, are the sign and concentration of the ﬁxed charge in the membrane.
Next, we employed the Nemst-Planck equation, (5), to calculate the excluded ion
concentration, C ,7” , at the right side of the polyelectrolyte ﬁlm.

. F
J,=—Dlﬂ—ﬂgﬂ (5)
(it RT dx

In this equation, D is the diffusion constant, ¢ is the electrical potential, x is the distance
in membrane, and F, R, and T are the Faraday constant, the gas constant, and the

temperature, respectively. Subscript i indicates the ith ion. Because the concentration of

86

the excluded ion in the membrane is low, the migration term in the Nernst-Planck
equation is negligible compared to the diffusion term for this ion. Assuming constant
ﬂux, equation 5 for the excluded ion can then be simpliﬁed to equation 6

Ch")? - (Am J (6)

Jr; = “Dh[ d

where d is the thickness of the polyelectrolyte ﬁlm. Rearrangement of equation 6 allows

calculation of C 3'” once C g'L is calculated. Finally, using equation 7, the concentration

of the excluded ion in the source phase, C 3 , could be calculated.

Zn

2 Cm” +2 C "21
_ B 8 S): I] (7)
ZACA

 

C; = C,’,"”[

We performed the entire calculation iteratively, using the experimental ﬂux and varying
the diffusion coefﬁcient of the excluded ion until the calculated excluded ion
concentration in the source phase matched that used experimentally (0.1 M). Selectivity
for species 1 (e. g., Cl') over species 2 (e. g., S042) is given by equation 8

J3 DB AC;
alt-*3 = I: 'x |=aoxae (8)
" 2 J32 DB2 AC2:

 

where AC;l = cg” — cg'L (Figure 4.5), and a0 and a5 denote selectivities arising from

selective diffusion of ions and electrostatic exclusion, respectively.

87

4.6 References and Notes
1 Liu, C.; Martin, C. R. Nature London 1991, 352, 50-52.

2 Kesting, R. E.; Fritzche, A. K. Polymeric Gas Separation Membranes; John Wiley &
Sons: New York, 1993.

3 Brandt, D. C.; Leitner, G. F.; Leitner, W. E. In Reverse Osmosis Membrane
Technology, Water, Chemistry, and Industrial Applications; Amjad, Z., Ed.; Van
Nostrand Reinhold: New York, 1993, pp 1-36.

4 Niwa, M.; Kawakami, H.; Nagaoka, S.; Kanamori, T.; Shinbo, T. J. Membr. Sci. 2000,
171, 253-261.

5 Shieh, J .-J .; Chung, T.-S. Ind. Eng. Chem. Res. 1999, 38, 2650-2658.

6 Parthasarathy, A.; Brumlik, C. J .; Martin, C. R.; Collins, G. E. J. Membr. Sci. 1994,
94, 249-254.

7 Chen, H.; Belfort, G. J. Appl. Polym. Sci. 1999, 72, 1699-1711.

8 Kilduff, J. E.; Mattaraj, S.; Pieracci, J. P.; Belfort, G. Desalination 2000, 132, 133-
142.

9 Linder, C.; Kedem, 0. J. Membr. Sci. 2001, 181, 39-56.

10 Zhang, L.-H.; Hendel, R. A.; Cozzi, P. G.; Regen, S. L. J. Am. Chem. Soc. 1999,
121, 1621-1622.

11 Decher, G.; Hong, J. D. Macromol. Chem, Macromol. Symp. 1991, 46, 321-327.
12 Decher, G.; Hong, J. D. Ber. Bunsenges. Phys. Chem. 1991, 95, 1430-1434.

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

14 Ackem, F. v.; Krasemann, L.; Tieke, B. Thin Solid Films 1998, 32 7-329, 762-766.

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

16 Kotov, N. A.; Magonov, S.; Tropsha, E. Chem. Mater. 1998, 10, 886-895.
17 Levasalmi, J .-M.; McCarthy, T. J. Macromolecules 1997, 30, 1752-1757.

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

88

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

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

Krasemann, L.; Tieke, B. Langmuir 2000, 16, 287-290.
Harris, J. J .; Stair, J. L.; Bruening, M. L. Chem. Mater. 2000, 12, 1941-1946.

Balanchandra, A. M.; Dai, J.; Bruening, M. L. Macromolecules 2001, 35, 3171-
3178.

Schlenoff, J. B.; Ly, H.; Li, M. J. Am. Chem. Soc. 1998, 120, 7626-7634.
Laurent, D.; Schlenoff, J. B. Langmuir 1997, 13, 1552-1557.
Kotov, N. A.; Dekany, I.; F endler, J. H. Adv. Mater. 1996, 8, 637-641.

Joly, S.; Kane, R.; Radzilowski, L.; Wang, T.; Wu, A.; Cohen, R. E.; Thomas, E. L.;
Rubner, M. F. Langmuir 2000, 16, 1354-1359.

Chen, J .; Huang, L.; Ying, L.; Luo, G.; Zhao, X.; Cao, W. Langmuir 1999, 15,
7208-7212.

Harris, J. J .; DeRose, P. M.; Bruening, M. L. J. Am. Chem. Soc. 1999, 121, 1978-
1979.

Yoo, D.; Shiratori, S. S.; Rubner, M. F. Macromolecules 1998, 31 , 4309-4318.
Fery, A.; Scholer, B.; Cassagneau, T.; Caruso, F. Langmuir 2001, 17, 3779-3783.

Mendelsohn, J. D.; Barrett, C. J .; Chan, V. V.; Pal, A. J .; Mayes, A. M.; Rubner, M.
F. Langmuir 2000, 16, 5017-5023.

Dyer, R. G.; Tumbull, K. D. J. Org. Chem. 1999, 64, 7988-7995.

Corrie, J. E. T.; Trentham, D. R. In Bioorganic Photochemistry; Morrison, H., Ed.;
John Wiley: New York, 1993; Vol. 2, pp 243-305.

Pillai, V. N. R. In Organic Photochemistry; Padwa, A., Ed.; Marcel Dekker: New
York, 1987; Vol. 9, pp 225-323.

Matuszczak, S.; Cameron, J. F.; Fréchet, J. M. J.; Wilson, C. G. J. Mater. Chem.
1991, 1, 1045-1050.

Nishikubo, T.; IIzawa, T.; Takahashi, A.; Shimokawa, T. J. Polym. Sci. A: Polym.
Chem. 1990, 28, 105-117.

89

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

Beecher, J. E.; Cameron, J. F.; Fréchet, J. M. J. J. Mater. Chem. 1992, 2, 811-816.
Patchornik, A.; Amit, B.; Woodward, R. B. J. Am. Chem. Soc. 1970, 92, 6333-6335.
Yoshikawa, Y.; Matsuoka, H.; Ise, N. Br. Polym. J. 1986, 18, 242-246.
Shiratori, S. S.; Rubner, M. F. Macromolecules 2000, 33, 4213-4219.
Dubas, S. T.; Schlenoff, J. B. Macromolecules 1999, 32, 8153-8160.

Dai, J .; Jensen, A. W.; Mohanty, D. K.; Emdt, J.; Bruening, M. L. Langmuir 2001,
17, 931-937.

Cochin, D.; Passmann, M.; Wilbert, G.; Zentel, R.; Wischerhoff, E.; Laschewsky,
A. Macromolecules 1997, 30, 4775-4779.

Gravel, D.; Giasson, R.; Blanchet, D.; Yip, R. W.; Sharma, D. K. Can. J. Chem.
1991, 69, 1193-1200.

DeMayo, P. Adv. Org. Chem. 1960, 2, 367-425.

Schlenoff, J. B.; Dubas, S. T. Macromolecules 2001, 34, 592-598.
Farhat, T. R.; Schlenoff, J. B. Langmuir 2001, 17, 1184-1192.
Meyer, K. H.; Sievers, J .-F. Helv. Chim. Acta 1936, 19, 649-664.

Teorell, T. In Progress in Biophysics and Biophysical Chemistry; Buter, J. A. V.,
Randall, J. T., Eds.; Academic Press: New York, 1953; Vol. 3, pp 305-369.

Lebedev, K.; Ramirez, P.; Mafé, S.; Pellicer, J. Langmuir 2000, 16, 9941-9943.
Herrera, A.; Yeager, H. L. J. Electrochem. Soc. 1987, 134, 2446-2451.

Rollet, A.-L.; Simonin, J.-P.; Turq, P. Phys. Chem. Chem. Phys. 2000, 2, 1029-
1034.

Miyoshi, H. J. Membr. Sci. 1998, 141, 101-110.
Paula, S.; Volkov, A. A.; Deamer, D. W. Biophys. J. 1998, 74, 319-329.

Makino, K.; Oshima, H.; Kondo, T. Colloid Polym. Sci. 1987, 265, 911-915.

90

Chapter 5

Metal Nanoparticles/Polyelectrolyte Multilayer films:
Synthesis and Catalytic Applications
5.1 Introduction
Due to their small size, metal nanoparticles often have properties that are different
from those of bulk metals."2 These novel properties may ﬁnd applications in areas such
as photoelectronicsf‘4 catalysis,5’6 magnetism,7’8 and sensing?"0 In many cases, the
particles will be most useful in the form of thin ﬁlms, but preparation of tunable,
homogeneous ﬁlms that contain nanoparticles is an on-going challenge. Several
technologies have been exploited to fabricate such ﬁlms. Mechanical mixing of metal
particles with polymers is simple and straightforward but often yields heterogeneous
ﬁlms.ll Spin-coating or smearing of polymer-metal ion mixtures followed by reduction
may result in homogenous coatings, but deposition of ultrathin ﬁlms can be difﬁcult, and
coverage of substrates with unusual geometries is not possible. 12'”
Layer-by-layer deposition of nanoparticles allows ﬁne control over ﬁhn thickness and

formation of ﬁlms on a wide variety of surfaces. Interactions used to form such ﬁlms
include ligand-metal ion-ligand bridges15 and covalent bondingm’l7 The most convenient
and versatile technique for layer-by-layer deposition, however, is probably the alternating
adsorption of oppositely charged polyelectrolytes.18 Control of ﬁlm thickness on the
nanometer scale results simply from controlling deposition conditions such as the number
of layers, pH, and the supporting electrolyte concentration present during adsorption.'9’20

This method has been used extensively to construct composite ﬁlms containing inorganic

colloid (e. g., sulﬁdes, selenides, tellurides, and oxides of Cd, Pb, Ti, Zn, Si, Fe, and

91

Mn).2 "22 ”‘24 A few groups synthesized colloid/polyelectrolyte ﬁlms by utilizing

- 25 - 21.27
dendrrmer-, crtrate-, ’

or polyelectrolyte-stabilized28 gold nanoparticles. The
common and crucial feature in all of these studies is to prepare properly charged,
stabilized metal nanoclusters before ﬁlm construction. Alternating adsorption of colloids
and polymeric electrolytes then results in ﬁlm formation.

This chapter of the thesis reports the synthesis of nanoparticle-containing ﬁlms
through formation of a polyelectrolyte-metal ion complex, layer-by-layer adsorption of
this complex and a polyanion, and ﬁnally post-deposition reduction of the metal ions
(Scheme 5.1). The diameters of the nanoparticles depend on the concentration of metal
ions in the ﬁlm. This method has several assets. First, because the metal ions are well
distributed along the polymer chains, nanoparticles are dispersed throughout the ﬁlm.
Second, the surrounding polymer limits particle aggregation and thus yields a small
particle size. Finally, the process circumvents the need to develop methods to synthesize
and isolate uniform, stabilized colloids in solution.

Rubner and coworkers reported the synthesis of poly(acrylic acid)-
Ag(0)/poly(allylamine hydrochloride) (PAH) ﬁlms by a related method.29 They
immersed a preformed poly(acrylic acid) (PAA)/PAH ﬁlm into a solution containing Ag+
to trap the metal ion in the ﬁlm by ion exchange. Reduction of the Ag+ with H2 resulted
in Ag nanoparticles. This ion-exchange method was also used to construct
semiconductor nanocomposites by ﬁrst adsorbing a metal ion into a polyelectrolyte ﬁlm

and subsequently treating the ﬁlm with H2S.30'3'

92

 

 

e 49*
+N—CH2CH2R /CH2CH2_3_
112 x N y

(13
CHZCHgml—i3

alternating deposition
and rinse

 

 

JReduction by ©= metal '00

heating or NaBH4

 

.= nanoparticle

Scheme 5.1

93

 

By comparison, the method described herein involves formation of the metal complex
before ﬁhn deposition. This affords a convenient method for controlling the amount of
metal ion in the ﬁlm, which in turn allows control over particle size after reduction.
Transmission electron microscopy (TEM) shows that particle size can be tuned from 4 to
30 nm, and initial studies of these particle-containing ﬁlms show them to be active as

both catalysts and antimicrobial coatings.

5.2 Experimental Section

Materials. Polyethyleneimne (PEI) (M., = 25,000), poly(acrylic acid) (25 wt% in water,
M., = 90,000), tetraamineplatinum(II) chloride hydrate, and sodium perchlorate hydrate
were purchased from Aldrich. Silver nitrate and methylene bromide were purchased
from E. Merck and J. T. Baker, respectively. Films were prepared on quartz slides, Au-
coated Si (200 nm sputtered Au on 20 nm Cr on Si(100)), and copper-coated carbon TEM
grids. Prior to deposition, Au-coated Si and TEM grids were cleaned with UV/ozone for
15 min, while quartz slides were cleaned in hot piranha solution (H2804zH202, 3:1) for 30
min and copiously rinsed with water. Caution: Pirhana solution reacts violently with
organic compounds and should be used with care. Waste piranha should be stored in
loosely capped bottles. To deposit one PEI-AgI/PAA bilayer, substrates were immersed
in PEI-Ag+ solution (0.1, 0.5, or 3.0 mM AgN03, 1 mg/mL PEI; pH adjusted to 7.0 with
0.1 M HN03) for 5 min, rinsed with water for 1 min, immersed in a 3 mg/mL PAA
solution (pH adjusted to 5.0 with 0.1 M NaOH) for 5 min, and rinsed again with water for
1 min. Bilayers containing other metal ions were prepared in the same fashion.

Mutilayers were obtained by repetitive deposition of PEI-Ag+ and PAA. Reduction of

94

ions was effected by heating at 150 °C for 2 h under the protection of N2, or by
immersing the ﬁlm in a freshly prepared 1 mM NaBH4 solution for 30 min.

UV-visible spectra were recorded using a Perkin-Elmer Lamda 40 spectrometer. TEM
was performed on a J EOL 100CX microscope operating at 100 kV, and cyclic
voltammograms were obtained with a CH Instruments electrochemical analyzer (model
604). Measurements were made using a Ag/AgCl (3 M KCl) reference electrode and a
platinum wire counter electrode. Working electrodes were gold-coated silicon wafers
placed in an 0-ring cell that exposed 0.1 cm2 of area. In the antimicrobial test, one seed
colony of E. Coli (INFalphaF, Invitrogen) was inserted into the test tube and allowed to

grow overnight at 37 °C in Luria-Bertani broth.

5.3 Results and discussion

5.3.1 Synthesis of metal nanoparticles. Branched polyethyleneimine (PEI) contains
primary, secondary, and tertiary amino groups and thus is an attractive polymer for
metal-ion complexation.32 Initially, we tried to prepare PEI-stabilized Ag colloids in
solution and subsequently form ﬁlms by alternating adsorption of PAA and the PEI-
stabilized colloids. A brownish solution formed immediately after adding 1 mL of 0.1 M
NaBH4 to a 20 mL mixture of PEI (1 mg/mL) and 0.5 mM AgN03. The brown color
gradually faded to yellow over a 10 h period, and this clear yellow color did not change
for 3 months. No precipitation was observed. UV-visible spectra and TEM conﬁrmed
the initial formation and growth of silver clusters (Fig. 5.1). Particle size increased from

a value of 8 nm (1 hour after addition of NaBHa) to 21 nm nine hours later.

95

Absorbance

A _ -. . .
1h .
2‘ .‘I' '_
:-‘_V‘.I" I.
100’hm' -.
. “ ' ' - ' ' :"C
11 10h “H -.'°'- ”at."
' I "2...": a H.
.3 a ‘ .r. ‘ .6 I
0 v ‘ ... .
0 1 1 I 1 1 1 A . ‘ " 4 .4 9 o. 1
'. '5 .- . -' '.
250 300 350 400 450 500 550 600 , ' '- ‘.'-: ‘, ,'
Wavelength (nm) .Myz. _ . ‘ .

 

 

 

 

 

 

 

 

 

 

 

Figure 5.1: UV-visible spectra (A) and TEMs (B,C) of PEI-protected silver
colloids reduced with NaBH4. The absorbance decreased 50% in the ﬁrst 10
hours, and average particle size increased from 8 nm at 1h (B) to 21 nm at 10 h

(c). [Ag*]=0.5 mM, [PEI]=1 mg/mL.

96

The surface plasmon absorbance decreased ~50% with a red shift of 8 nm over the same
time period. These results indicate that PEI stabilizes silver nanoparticles temporarily
and prevents precipitation over longer time periods. However, PEI does not inhibit
gradual particle aggregation. Mayer and coworkers investigated various polyelectrolytes
as colloid stabilizers in solution and found that many simple polyelectrolytes are not
satisfactory.33

Interestingly, in multilayered polyelectrolyte ﬁlms prepared using solutions of PEI-
Ag(0) colloids and PAA, surface plasmon absorbance was very weak even after
deposition of 10 bilayers. This suggests that the silver nanoparticles did not transfer into
the ﬁlm along with the PEI. During ﬁlm formation, adsorbed PEI may release colloids to
maximize electrostatic interactions. In contrast, NaBHa reduction of silver ions within
PEI-AgI/PAA ﬁlms (Scheme 5.1) easily yields stable nanoparticles. The surface
plasmon absorbance that results from these Ag(0) particles increases with the number of
layers in the ﬁlm as would be expected (Fig. 5.2). Additionally, this absorbance dropped
less than 10% after aging a ﬁlm for 16 days, suggesting that when silver ions are reduced
within PEI/PAA ﬁlms, movement of the resulting nanoparticles is restricted, and hence
particle aggregation is very slow. The TEM image in Figure 5.1 shows that the particles
produced by NaBHa reduction are relatively monodisperse (diameters of 4.0 :1: 0.6 nm).

Similar particles can be formed by including Pt(II) in PEI/PAA ﬁlms (Fig. 5.3).

97

 

Absorbance

 

 

 

 

250 300 350 400 450 500 550 600
Wavelength (nm)

Figure 5.2: (Top) UV-visible spectra of NaBH4-reduced PEI-
Ag(0)/PAA ﬁlms. Numbers on the plot refer to the number of
bilayers in the ﬁlm. (Bottom) TEM image of a NaBHa-reduced PEI-
Ag(0)/PAA ﬁlm (5.5 bilayers, the additional half bilayer results in
PEI-Ag(0) being the top layer in the ﬁlm). Silver particle size is 4.0 :t
0.6 nm, and 0.5 mM AgN03 was used in ﬁlm deposition.

98

Figure 5.3: TEM image of a 5.5-bilayer PEI-
Pt(0)/PAA (NaBH4 reduced) ﬁlm. Platinum
nanoparticle size: 10 i 4 nm. 0.5 mM
Pt(NH3)4Cl2 was used in the ﬁlm deposition,
and other conditions were the same as the

synthesis of PEI-Ag(0)/PAA ﬁlm.

 

Heat-induced reduction also creates silver nanoparticles in PEI-AgI/PAA ﬁlms. For
ﬁlms synthesized under the same conditions (0.5 mM Ag+), the diameters of thermally
reduced particles are ~2.5 times as large as those of chemically reduced particles
(compare TEMS in Fig. 5.2 and Fig. 5.4B). This is probably because heating (150 °C)
promotes aggregation of Ag atoms. The TEM images in Figure 5.4 also show that
particle size depends on the Ag+/PEI ratio in deposition solutions. The average particle
sizes when using 0.1 mM, 0.5 mM, and 3.0 mM Ag+ during deposition were 6, 10, and 26
nm, respectively. Control of particle size should provide a convenient means for tuning
ﬁlm properties.

Reduction of Ag" by heating should also result in cross-linking of PEI-Ag+/PAA
systems via heat-induced amide formation from carboxylate-ammonium complexes.34
The appearance of an amide peak in the external reﬂection FTIR spectra of heated PEI-
AgI/PAA ﬁlms (Fig. 5.5) conﬁrms that some cross-linking occurs. Cross-linking should
strengthen the ﬁlm, and control of the cross-linking density may allow tailoring of ﬁlm

permeability. Such control could ﬁnd application in selective catalysis.

99

 

Figure 5.4: TEM images of thermally reduced (heating for 2 h at 150 °C under

N2) 5.5—bilayer PEI-Ag(0)/PAA ﬁlms prepared using different concentrations of
Ag during PEI adsorption. (A) [Ag+] = 0.1 mM, particle size = 6 : 1 nm; (B)
[Ag+] = 0.5 mM, particle size = 10 : 2 mm; (C) [Ag+] = 3.0 mM, particle size = 26

: 5nm.

 

0.25 -

0.20 ~

1
0‘1 5 Amide

Absorbance

0.10 .

0.05 1
B

 

 

 

0 .00 I I I I
3500 3000 2500 2000 1 500 1 000

Wavenumbers (cm'l)

Figure 5.5: Reﬂectance FTIR spectra of a 5.5-bilayer PEI-AgI/PAA
ﬁlm on A] before (A) and after (B) heating at 150 °C for 2h under N2.
Appearance of the amide peak indicates partial cross-linking.

100

5.3.2 Electrocatalysis. Nanoparticles are especially attractive for catalysis because of
their high surface area to mass ratio. Because bulk Ag electrodes catalyze

. . - 5
electrochemical reduction of some organrc halogens} ’36

we performed preliminary
catalysis experiments using Ag nanoparticles to reduce methylene bromide. In this
reaction, Ag likely binds with Br' and a carbene intermediate to facilitate breaking of the
C-Br bond. The main products of the reduction are ethylene and methane.36 In cyclic
voltammograms using Au electrodes coated with PEI-Ag(0) /PAA ﬁlms (5.5 bilayers,
NaBH4 reduced), the methylene bromide reduction peak occurred at 0.77 V (vs Ag/AgCl),
while at bare or PEI/PAA-coated Au electrodes, no reduction peak was observed in this
region (Fig. 5.6A). Thus the nanoparticles are catalytic. The reduction peak potential at
the PEI-Ag(0)/PAA-coated Au electrode was, however, 0.12 V negative of that at a bare
Ag electrode. This could be due to electrical resistance in the ﬁhn, hindered access of
reagent to the nanoparticles, or a lower intrinsic catalytic activity of the nanoparticles.
Although the composite nanoparticle ﬁlm is less effective per apparent surface area than
the silver wire, one has to consider that composite ﬁlms can contain a much lower mass
of precious metal than bulk electrodes.

PEI-Pt(0)/PAA ﬁlms also exhibit electrocatalytic properties. Figure 563 shows
cyclic voltammograms at bare, PEI/PAA-, and PEI-Pt(0)/PAA-coated gold electrodes in
oxygen-saturated 1.0 M H2S04. Oxygen reduction begins at ~0.15 V (vs Ag/AgCl) for
the electrode coated with PEI-Pt(0)/PAA, while no reduction currents appear for bare
gold and PEI-PAA-coated gold electrodes from 0.6 to —0.1 V.37 The studies with both
Ag and Pt nanoparticles show that at least a portion of the nanoparticles are in electrical

contact

101

 

 

 

 

 

an,
A .°
I1mA/cm2
I’
7-
71"
E ‘/ o //"
'8
0
o 1 1 1 r 1 1
E 04 06 0.8 10 12 14
g
3
U

 

 

 

 

 

0.6 0.4 0.2 0.0
Potential (V vs AgIAgCI)

Figure 5.6: (A) Cyclic voltammograms of CH2Br2 (25 mM) at bare (ﬁlled circles),
PEI/PAA-coated (dashed line), and PEI-Ag(0)/PAA-coated (solid line) Au
electrodes, and at a bulk Ag electrode (open circles). The supporting electrolyte was
1.0 M NaClOa, the scan rate was 100 mV/s, and ﬁlms contained 5.5 bilayers of
polyelectrolyte. (B) Cyclic voltammograms at bare (ﬁlled circles), PEI/PAA-coated
(dashed line), and PEI-Pt(0)/PAA-coated (solid line) Au electrodes in 02-saturated

1.0 M H2S04. Films contained 9.5 bilayers of polyelectrolyte, and the scan rate was
50 mV/s.

102

with the gold substrate, and these particles can be accessed by analyte molecules in
solution.

5.3.3 Inhibition of bacteria growth. For centuries silver metal and silver salts have
been used to treat burn wounds.”39 Although these compounds are effective anti-
microbial agents, their use likely results in unwanted adsorption of silver ions in
epidermis cells and sweat glands.38 To reduce the likelihood of silver-ion adsorption,
membranes containing metallic silver, e. g., silver-coated nylon,39 are being investigated
as anti-microbial agents. Because polyelectrolyte ﬁlms can form on various substrates
including natural and synthetic polymers,40 PEI-Ag(0) ﬁlms could potentially be used to
construct antibacterial membranes for wound treatment.

To investigate the inhibition of bacterial growth by PEI-Ag(0)/PAA ﬁlms, we
compared the growth of E. Coli in bare glass test tubes and test tubes coated with PEI-
Ag(0)/PAA (Fig. 5.7). E. Colt-seeded solutions in both bare and 5.5-bilayer PEI/PAA-
coated test tubes became turbid after 12 h of incubation at 37 °C. Under the same
conditions, tubes coated with 5.5-bilayers of PEI-Ag+/PAA or PEI-Ag(0)/PAA (NaBHa
reduced) remained clear, showing that the latter two ﬁlms inhibited the growth of bacteria.
Optical density measurements quantitatively show the differences in turbidity as the
absorbance at 600 nm is proportional to the number of bacteria.41 For the blank tube and
the tube coated with a PEI/PAA ﬁlm, absorbances at 600 nm were 0.66 and 0.56,
respectively (Fig. 5.7), while absorbance values for PEI-AgI/PAA- and PEI-Ag(0)/PAA-
coated tubes were nearly zero (0.010 and 0.012, respectively). Signiﬁcantly, the silver

nanoparticle-included ﬁlms have the same antibacterial effect as ﬁlms containing silver

103

ions. The nanoparticle ﬁlms may be more desirable because they should minimize the

amount of Ag+ absorbed in the body.

 

 

 

 

 

 

1
1.0
g \
_ - m
g X\
140.5-
.0
1 2 3 4 <
3,4
0 I l
400 500 600 700 800
Wavelength (nm)
Blank PEI/PAA- PEI-Ag‘lPAA- PEI-Ag°/PAA-
tube\ /c3ated tube coated tube coated tube
Turbid Clear

Figure 5.7: (left) E. Coli. solutions photographed after 12 h of growth at 37 °C in
LB broth. The test tubes were coated with different ﬁlms as noted on the picture.
(right) Visible spectra of the solutions in the four test tubes. Solutions were

diluted 3-fold before acquiring the spectra.

104

5.4 Conclusions. The ability of PEI to form complexes with Ag+ and Pt(II) (and
presumably a wide variety of metal ions) allows the formation of nanoparticles by
reduction of metal ions in PEI/PAA ﬁlms. The size of the nanoparticles can be varied
easily by changing the concentration of metal ions present during PEI/PAA deposition.
Silver and Pt nanoparticles in PEI/PAA ﬁlms are electrocatalytically active, showing that
the nanoparticles are accessible to analyte molecules and electrically connected with the
electrode. Additionally, PEI-Ag(0)/PAA ﬁhns are effective in inhibiting bacterial growth.
The use of PEI-metal ion complexes and alternating polyelectrolyte deposition followed
by reduction appears to be a general method for forming nanoparticles, and should also

be applicable to forming bimetallic particles.

105

5.5 References and notes:

1 Fendler, J. H. Nanoparticles and Nanostructured Films: Preparation,
Characterization and Applications; Wiley-VCH: Weinhein, 1998.

2 Schmid, G. In Natoscale materials in chemistry; Klabunde, K. J ., Ed.; Wiley-
Interscience: New York, 2001, pp 15-59.

3 Sudeep, P. K.; Ipe, B. L; Thomas, K. 0.; George, M. V.; Barazzouk, S.; Hotchandani,
S.; Kamat, P. V. Nano Lett. 2002, 2, 29-35.

4 Merschdorf, M.; Pfeiffer, W.; Thon, A.; Voll, S.; Gerber, G. Appl. Phys. A 2000, 71 ,
547-552.

5 Zhong, C.-J.; Maye, M. M. Adv. Mater. 2001, 13, 1507-1511.
6 Gates, B. C. Chem. Rev. 1995, 95, 511-522.

7 Li, X. G.; Takahashi, S.; Watanabe, K.; Kikuchi, Y.; Koishi, M. Nano Lett. 2001, 1,
475-480.

8 Sun, X.; Gutierrez, A.; Yacaman, M. J .; Dong, X.; J in, S. Mater. Sci. Eng., A 2000,
286, 157-160.

9 Kim, Y.; Johnson, R. C.; Hupp, J. T. Nano Lett. 2001, 1, 165-167.

10 Ruiz, A.; Arbiol, J .; Cirera, A.; Cornet, A.; Morante, J. R. Mater. Sci. Eng. C 2002,
19, 105-109.

1 1 Ghosh, K.; Maiti, S. N. J. Appl. Polym. Sci. 1996, 60, 323-331.
12 Liu, C.; Xu, Y.; Liao, 8.; Yu, D. Appl. Catal., A 1998, 172, 23-29.

13 Fritzsche, W.; Porwol, H.; Wiegand, A.; Bommann, S.; Kohler, J. M. Nanostruct.
Mater. 1998, 10, 89-97.

14 Zhang, Z.; Zhang, L.; Wang, S.; Chen, W.; Lei, Y. Polymer 2001, 42, 8315-8318.

15 Zamborini, F. P.; Hicks, J. F.; Murray, R. W. J. Am. Chem. Soc. 2000, 122, 4514-
4515.

16 Sarathy, K. V.; Thomas, P. J .; Kulkarni, G. U.; Rao, C. N. R. J. Phys. Chem. B
1999, 103, 399-401.

17 Musick, M. D.; Keating, C. D.; Keefe, M. H.; Natan, M. J. Chem. Mater. 1997, 9,
1499-1501.

106

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

Decher, G. Science 1997, 277, 1232-1237.
Dubas, S. T.; Schlenoff, J. B. Macromolecules 1999, 32, 8153-8160.
Shiratori, S. S.; Rubner, M. F. Macromolecules 2000, 33, 4213-4219.

Mamedov, A. A.; Belov, A.; Giersig, M.; Mamedova, N. N.; Kotov, N. A. J. Am.
Chem. Soc. 2001, 123, 7738-7739.

Lvov, Y.; Munge, B.; Giraldo, 0.; Ichinose, 1.; Suib, S. L.; Rusling, J. F. Langmuir
2000, 16, 8850-8857.

Liu, Y.; Wang, A.; Claus, R. 0. Appl. Phys. Lett. 1997, 71 , 2265-2267.

Ostrander, J. W.; Mamedov, A. A.; Kotov, N. A. J. Am. Chem. Soc. 2001, 123,
1101-1110.

He, J.-A.; Valluzzi, R.; Yang, K.; Dolukhanyan, T.; Sung, C.; Kumar, J .; Tripathy,
S. K.; Samuelson, L.; Balogh, L.; Tomalia, D. A. Chem. Mater. 1999, 11, 3268-
3274.

Schrof, W.; Rozouvan, S.; Van Keuren, E.; Horn, D.; Schmitt, J.; Decher, G. Adv.
Mater. 1998, 10, 338-341.

Feldheim, D. L.; Grabar, K. C.; Natan, M. J .; Mallouk, T. E. J. Am. Chem. Soc.
1996, 118, 7640-7641.

Liu, Y.; Wang, Y.; Claus, R. 0. Chem. Phys. Lett. 1998, 298, 315-319.

Joly, 8.; Kane, R.; Radzilowski, L.; Wang, T.; Wu, A.; Cohen, R. E.; Thomas, E. L.;
Rubner, M. F. Langmuir 2000, 16, 1354—1359.

Xiong, H.; Cheng, M.; Zhou, Z.; Zhang, X.; Shen, J. Adv. Mater. 1998, 10, 529-532.
Dante, S.; Hou, Z.; Risbud, S.; Stroeve, P. Langmuir 1999, 15, 2176-2182.
Belﬁore, L. A.; Indra, E. M. J. Ploym. Sci. B 2000, 38, 552-561.

Mayer, A. B. R.; Hausner, S. H.; Mark, J. E. Polym. J. 2000, 32, 15-22.

Harris, J. J .; DeRose, P. M.; Bruening, M. L. J. Am. Chem. Soc. 1999, 121, 1978-
1978.

Guerrini, M.; Mussini, P.; Rondinini, 8.; Toni, G.; Vismara, E. Chem. Commun.
1998, 1575-1576.

107

36

37

38

39

40

41

F edurco, M.; Satoretti, C. J.; Augustynski, J. Langmuir 2001, 1 7, 2380-2387.

As with Ag, the Pt nanoparticles in ﬁlm are less catalytically active than a bare Pt
electrode. At a freshly cleaned (with concentrated HN03) Pt wire, reduction of
oxygen begins at around 0.4 V (vs Ag/AgCl).

Klasen, H. J. Burns 2000, 26, 117-138.

Balogh, L.; Swanson, D. R.; Tomalia, D. A.; Hagnauer, G. L.; McManus, A. T.
Nano Lett. 2001, I, 18-21.

Phuvanartnuruks, V.; McCarthy, T. J. Macromolecules 1998, 31 , 1705 - 2018.

Brown, T. A. Essential Molecular Biology: a Practical Approach; Oxford
University Press: New York, 1991; Vol. 1, pp22-26.

108

Chapter 6

Selective Hydrogenation Using Nanoparticle—Containing Polyelectrolyte Films

6.1. Introduction

Metal nanoparticles are attractive for catalysis because their large surface area to
volume ratio results in a high catalytic efﬁciency per amount of precious metal
empolyed."2 Additionally, reaction rates at a particular metal can vary with nanoparticle
size, allowing optimization of catalytic properties.3 Unfortunately, however, aggregation
of naked nanoparticles often prohibits control over particle size.4 To overcome this
problem, catalytic nanoparticles are either immobilized on solid supports (e.g., carbon,
metal oxides, and zeolites) or stabilized by capping the colloids with ligands ranging
from small organic molecules to large polymerss'7 Recently, dendrimers,8 block
copolymer nanospheres,9 and cross-linked lyotropic liquid crystals'0 were employed to
encapsulate metal nanoparticles.

In addition to protecting nanoparticles from aggregation, encapsulating materials and
ligand stabilizers can sometimes impart selectivity to catalysis. For example,
hydrogenation of 2-hexyne catalyzed by 2-n-butylphenanthrolinc-protected palladium
particles yields exclusively cis-2-hexene,” and hydrogenation of allyl alcohol is 1 1-fold
faster than hydrogenation of 3-methyl-l-penten-3-ol when using dendrimer-encapsulated
palladium particles as catalysts.12 Triblock copolymers that encapsulate Pd nanospheres
also function as nanoﬁlters that allow preferential hydrogenation of methyl methacrylate

in the presence of ethylene glycol dimethacrylate.9 Thus, variation of the matrix around

109

catalytic nanoparticles should allow tuning of catalyst properties while helping to
stabilize particles and decrease leaching of catalyst.

Multilayer polyelectrolyte ﬁlms should provide an especially convenient and versatile
matrix for encapsulating nanoparticles. Several studies show that adsorption of
multilayer polyelectrolytes and inorganic colloids provides a controlled method for
depositing nanoparticle-containing ﬁlms.l3 '17 Such methods are convenient because ﬁlm
deposition occurs simply by alternating immersion of a charged substrate in solutions

18.19

containing polyelectrolytes and colloids. However, stabilized colloids must be

prepared prior to ﬁlm deposition. To overcome this limitation, Rubnerzo‘21 and

.2
Stroeve22 3

utilized ion-exchange to insert metal ions into multilayer polyelectrolyte
ﬁlms after deposition. Subsequent reactions of the metal ions yielded nanoparticle-
containing ﬁlms. We recently reported a variation of this method.24 The procedure
involves alternating adsorption of polymeric metal-ion complexes and polyanions
followed by reduction of the metal ions. Control over the concentration of metal ions
present in the polycation deposition solution allows control over particle size, and cyclic
voltammetry shows that nanoparticles in these ﬁlms are accessible to analytes and
electrocatalytically active.

In other studies, we showed that control over the composition of multilayer
polyelectrolyte membranes can yield remarkably ion-transport (Cl'/S042'=1000) or gas
transport (C02/CH4=68) selectivities.25‘26 Preliminary studies of the transport of neutral
organic molecules through multilayer polyelectrolyte ﬁlms indicate that even in simple

poly(acrylic acid) (PAA)/poly(allylamine hydrochloride) (PAH) ﬁlms, methanol

diffusion is ten times faster than isopropanol difﬁrsion. We therefore surmised that

110

polyelectrolyte ﬁlms might function as highly selective ﬁlters for catalysis by
encapsulated nanoparticles.

This chapter reports our initial investigations of selective hydrogenation by
polyelectrolyte-encapsulated Pd nanoparticles (PEPst). To increase ﬁlm surface area
and enhance reaction rate, we deposited PEPst on lSO-pm-diameter alumina particles
(Scheme 6.1). The alumina particles are small enough that they can be suspended in
solution by vigorous stirring, but large enough so that simple ﬁltration allows catalyst
recovery. To demonstrate selectivity, we investigated the hydrogenation of ﬁve different
water-soluble unsaturated alcohols (Scheme 6.2). Although these alcohols have very
similar reactivity on ‘naked’ Pd catalysts, reaction rates at PEPst differ by up to a
factor of 31. Selectivity is enhanced by both capping of particle-containing ﬁlms with a
more selective polycation/polyanion pair and heat-induced cross-linking. We expect that
even higher selectivities can be obtained by carefully controlling the composition and

chemistry of multilayer polyelectrolyte ﬁlms.

6.2 Experimental

6.2.1 Materials. Polyethyleneimine (PEI) (M.,. = 25,000), poly(acrylic acid) (25 wt% in
water, M.,. = 90,000), palladium (5 wt% on alumina powder), a-alumina (100 mesh), allyl
alcohol (99%), 1-penten-3-ol (99%), 3-methyl-1-penten-3-ol (99%), 3-methyl-2-buten-l-
01 (99%), and crotyl alcohol (97%, mixture of isomers) were purchased from Aldrich,
while potassium tetrachloropalladate(II) (99.99%) was obtained from Alfa Aesar. All

reagents were used as received except crotyl alcohol, which was vacuum distilled before

111

 

+ 1311(1)

1,8;CH20H2MVCHZCH21;
l
CH20H21‘IH3

alternating deposition
and rinse

 

 

 

 

O=Pd(0) nanoparticle

Scheme 1. Synthesis of Pd-containing poly(ethyleneimne)/poly(acrylic acid)

112

WOH /\£OH At}. \:\—OH >=-\_
ﬂ OH
(2)

OH
(1) (3) (4) (5)

Scheme 2. Substrates for catalytic hydrogenations.

use. Prior to synthesis of ﬁlms, alumina was washed 4 times (suspension in water
followed by centri fugation and decantation). All solutions were prepared with 18 MO
cm Milli-Q water.

6.2.2 Synthesis of polyelectrolyte ﬁlms containing Pd nanoparticles. Synthesis
followed the general layer-by-layer adsorption method shown in scheme 6.1. First, 15 g
of a—alumina was added to a 100 mL PEI-Pd(II) solution (1mg/mL PEI, 2.0 mM K2PdC14,
pH adjusted to 9.0 with 0.1 M HCl), which was stirred vigorously for 10 min.
Subsequently, the alumina was precipitated by centriﬁrgation for 2 min. The supernatant
was saved for the next polycation deposition, and the precipitated alumina was washed
with 3 cycles of stirring in fresh water and centrifugation. To deposit a polyanion layer,
100 mL of 20.0 mM PAA (pH adjusted to 4.0 with 0.1 M NaOH, molarity of PAA given
with respect to the repeating unit) was added to the PEI-Pd(II)-coated alumina, and the
particles were precipitated and washed as described above to yield 1 adsorbed bilayer of
PEI-Pd(II)/PAA. This procedure was repeated until 7 bilayers of PEI-Pd(II)/PAA were
deposited. Reduction of the Pd(II) in the ﬁlms by exposure to 100 mL of fresh 1.0 mM
NaBHa for 30 min (with stirring) provided PEI-Pd(0)/PAA-coated alumina. Particles

were washed three times after exposure to NaBHa.

113

For the convenience of characterization, we also synthesized PEPdN ﬁlms on Al-
coated slides (200 nm of Al sputtered on Si(100)) using a similar dip-and-rinse procedure.
Reﬂection F TIR (N icolet Magna-560) and ellipsometry (J. A. Woolam M-44) were used
to monitor the ﬁlm formation. Films (5 bilayers) were also deposited on copper-coated
carbon grids for TEM measurements. TEM was performed on a J EOL 100CX
microscope operating at 100kV. The Al-coated slides and TEM grids were UV/ozone
cleaned for 15 min before deposition.

6.2.3 Hydrogenation. Catalytic hydrogenations were run in a 100mL 3-neck, pear-
shaped ﬂask. H2 was bubbled through a ﬁit at the bottom of the solution at a rate of 5
mL/min (atmospheric pressure), and the solution was vigorously stirred throughout the
reaction. In most instances, the hydrogenation solution initially contained 1.0 mmol of
substrate (or total 1 mmol of substrates with eqimolar of each component in competitive
hydrogenation) and either 10 mg of commercial 5%-Pd-on—alumina or 250 mg of alumina
coated with PEPst in 50 mL water. For 3-methyl-2-buten-1-ol hydrogenation with the
commercial catalyst, 2 mmol of substrate and 5 mg catalyst were used so that low
conversion of the alcohol could be effectively studied. Solutions of catalyst and solvent
(H20) were bubbled with H2 for 15 min before adding the substrates, all of which were
liquids. Gas chromatography (GC) (Shimadzu GC-17A equipped with a RTx-BACI
column) was used to monitor the reactions. The sensitivity of the ﬂame-ionization
detector was assumed to be the same for products and reactants because they contain the
same number of carbon atoms. For reactions with more than one product, GC-MS and

1H NMR (Fig. 6.3, appendix) were used to identify the products.

114

To calculate turnover frequencies, the amount of Pd in the catalyst must be known.
For both the commercial and synthesized catalysts, the percentage of palladium in the
material was determined by atomic emission spectrometry. The standard solutions (0.1 to
0.5 mM) were prepared by dissolving K2PdC14 in 0.1 M of HNO3, and sample solutions
were prepared by stirring 250 mg of catalyst in 2 mL of aqua regia for 15 min. The
solution was diluted to 12.5 mL and centriﬁrged (the (rt-alumina support does not dissolve
in aqua regia), and the supernatant was analyzed. To see if the dissolved polymer or a
small amount of dissolved alumina interfered with the Pd measurement, we also prepared
blank solutions by dissolving 7-bilayer PEI/PAA ﬁlms (without Pd), and did not observe
emission at 633 nm. The determined Pd quantities in 250 mg of 7-bilayer PEI(0)/PAA-
on-alumina and 10 mg of 5% Pd-on-alumina were 2.14x10'6 mole and 5.29x10'6 mole,
respectively. The value for the commercial catalyst corresponds to about 6% by wt,

showing that the analyses are reasonable.

6.3 Results and Discussion

6.3.] Synthesis of PEPst. We previously reported the synthesis of Ag and Pt
nanoparticles by reduction of metal ions in multilayered polyelectrolyte ﬁlms.24
Catalysis of many organic reactions, however, requires Pd nanoparticles. The TEM
image (Fig. 1) conﬁrms that Pd nanoparticles can also be produced by reduction of Pd(II)
ions in PEI/PAA ﬁlms. The particles have diameters of 1-3 mm and are well distributed
throughout the ﬁlm. Ellipsometry shows that 7-bilayer PEI-Pd(0)/PAA ﬁlms on Al-

coated wafers have thicknesses of 350:30 A.

115

 

50 nm

Figure 6.1: TEM image of 5 bilayers of PEI-Pd(0)/PAA on

a carbon-coated copper grid.

6.3.2 Hydrogenation. To show that encapsulation of Pd nanoparticles in polyelectrolyte
ﬁlms can result in selective catalysis, we hydrogenated a series of unsaturated alcohols
(Scheme 6.2) using PEPdN ﬁlms on alumina as well as a commercial Pd-on-alumina
catalyst. Pseudo zero-order kinetics were observed for all hydrogenations at low
conversions (up to 80% as shown in Figure 6.2 as well as Figures 6.1 and 6.2 in the
Appendix.) Turnover frequencies (TOFs) were calculated from the slopes in these plots
(measured in the linear range).

Table 1 summarizes the TOFs for hydrogenation of the two series of molecules. The
ﬁrst series consists of allyl alcohol (1), l-penten-3-ol (2), and 3-methyl-l-penten-3-ol (3),
and members in this group differ in the substituents at the (rt-carbon of the double bond.
Group 2 consists of allyl alcohol. crotyl alcohol (4), and 3-methyl-buten-1-ol (5), and
these molecules differ in the number of methyl groups attached to the terminal carbon of

the double bond.

116

0n ‘naked’ 5%-Pd-on-alumina catalyst, the reaction rates for substrates of group 1 are
very close (1/221/3=0.87), showing that the bulky alkyl groups at the (rt-carbon do not
have a signiﬁcant effect on reactivity. The data for group 2 are a little more complicated.
The TOF of 4 is somewhat lower than that of 1, probably because of steric hindrance due
to the extra methyl group.27 For similar reasons, the TOP of 5 should be even lower, but
a higher (Table 6.1) TOF was observed. However, plots of conversion versus time for 5
show two linear regions (Fig. 6.2,appendix). The linear region at low conversions (up to
~20%) has a 3-fold greater slope than the region at higher conversions. Table 6.1 lists
TOFs calculated based on both regions. No matter which TOF we choose for 5, however,
rate differences among the molecules in group 2 are still small (1/4=1.43, 1/5=0.52 or
1.63, Table 6.1).

In contrast, PEPst afford signiﬁcantly different hydrogenation rates for the various
molecules. The TOP of 1 on PEI-Pd(0)/PAA-on-alumina is only about 30% lower than
that on the commercial Pd-on-alumina catalyst, showing that the catalytic efﬁciency of
PEPst is not severely decreased by the presence of the polyelectrolyte ﬁlm. However,
the ratios of hydrogenation rates for 1/2 and 1/3 were 2.6 and 5.7, respectively, for the
PEI-Pd(0)/PAA catalyst. This difference in reaction rates likely reﬂects differences in
the sizes of the substrates.

This effect of substituents on reaction rate is more pronounced for substrates in group
2 (1, 4, 5). The hydrogenation rates of 4 and 5 were 6 and 31 times lower, respectively
than that of 1. The methyl groups attached to the double-bonded carbon likely provide
greater steric hindrance to adsorption on polyelectrolyte-protected catalytic sites than do

substituents at the a-carbon.

117

Table 6.]. Rates of hydrogenation for structurally related unsaturated alcohols using

commercial 5%-Pd-on-alumina or PEI-Pd(0)/PAA on alumina as catalysts.“

 

TOF (mol hydrogenation per mol Pd per h)

 

 

 

Substrates 5% Pd 7-bilayer Capped Cross-linked
PEI'Pd/PAA PEI-Pd/PAA” PEI-Pd/PAA‘
1 W0” 1300:150 1030 752 364
OH
2 / 1500:120 400:30 140 79

 

3 /\tOH 15001100 182 56 35

4 KOH + 910:50 184

\
5 _ 2500:15() 33
/—\—0H (~800)“

 

 

 

 

 

 

 

 

 

" Hydrogenations on commercial catalyst were repeated 3 times. On synthesized
catalysts, we chose 2 as a representative example and repeated hydrogenation 3 times.
Other hydrogenations were run only once.

b 7 bilayers of PEI-Pd(0)/PAA capped with 1 bilayer of PAH/PAA.

‘ Cross-linking was carried out by heating PEI-Pd(0)/PAA-coated (7 bilayers) alumina at
150 °C for 2h under a N2 atmosphere.

“ Calculated based on reaction rate at high (above 20%) conversion.

118

100

 

 

80-

60*

Hydrogenation (%)
A
O

 

 

 

 

 

 

 

0 2 4 6 8 0 20 40 60 80
Reaction time (min)

Figure 6.2: Percent of substrates hydrogenated vs reaction time for allyl alcohol
(inverted triangles), l-penten-3-ol (triangles), and 3-methyl-1-penten-3-ol (circles) on
a commercial 5%-Pd-on-alumina catalyst (left) and on a 7-bilayer PEI-Pd(0)/PAA-
coated alumina catalyst. Percent hydrogenation was obtained from the integration of

product and reactant peaks in a GC spectrum.

Hydrogenation rates for PEPst can be tuned by depositing a capping bilayer of
polyelectrolyte on the ﬁlm. For example, capping 7 bilayers of PEI-Pd(0)/PAA-on-
alumina with 1 bilayer of PAH/PAA (no additional Pd) decreased the reactivity of
substrates 1, 2, and 3 to different degrees, and yielded reactivity ratios of 1/2 = 5.4 and
1/3 = 13. This represents an approximately 2-fold increase over similar ratios for 7-
bilayer PEI-Pd(0)/PAA ﬁlms.

Cross-linking of polyelectrolyte ﬁlms via heat-induced amide formation from
carboxylate and ammonium groups also increases the selectivity among hydrogenation
rates. Selectivities (1/2 = 5 and 1/3 =10) obtained by heating the catalysts at 150 °C for

2h under N2 are similar to those achieved by depositing a capping layer. As might be

119

expected, the increases in selectivity for both capped and cross-linked ﬁlm are
accompanied by a decrease in rate. This likely results from slower diffusion of substrates
to nanoparticles.
6.3.3 Competitive hydrogenation. In practical applications of selective catalysts, such
as minimizing the number of puriﬁcation steps involved in organic reactions, one
substrate must be hydrogenated in the presence of several impurities.28 The two
examples in Figure 6.3 demonstrate the selectivity of PEPst in competitive
hydrogenation. In an equimolar mixture of 1 and 2, use of the commercial 5%-Pd-on-
alumina catalyst results in 94 % conversion of 2 when 1 is 99% converted to either the
hydrogenated product or isomer (Figure 6.3, left). However, when using the 7-PEI-
Pd/PAA-on-alumina catalyst, the conversion of 2 decreased to 57% when l was 99%
converted. Conversion of 2 dropped to only 40% with the addition of a PAH/PAA
capping layer on the polyelectrolyte ﬁlm. Selectivity improvements are more dramatic
for the hydrogenation of a mixture of 1 and 3 (Figure 6.3, right), where the conversion of
3 is 75% for 5%-Pd-on-alumina, 29% for 7-bilayer PEI-Pd/PAA-on-alumina, and 17%
for PAI—UPAA-capped 7-bilayer PEI-Pd/PAA-on-alumina when 1 has been 99%
converted. These selectivities appear to be lower than the ratios of TOFs shown in Table
6.1, and may represent the fact that rate of hydrogenation of 1 will decrease at high
conversions.

Relative to commercial catalyst, PEPst also exhibited enhanced selectivity in
competitive hydrogenation of a mixture of the three components in group 2 (Figure 6.4).

Using 5%-Pd-on-alumina, when 1 was 99% converted to products, the conversions of 4

120

and 5 were 50% and 20%, respectively. For 7-bilayer PEI-Pd(0)/PAA-on-alumina

catalyst, conversions of 4 and 5 decreased to 20% and 5%.

0H

WOH
126,1 1:
© ' C?
@511 0...... JUL

8 e
@ 1
11L JUL/Lani 10;

3 4 1 2 3 4
Retention time (min)

/\/

 

 

1 L
@E

 

 

 

 

 

 

 

 

 

 

Figure 6.3: Gas chromatograms of reaction mixtures from competitive
hydrogenations of allyl alcohol and l-penten-l-ol (left), and allyl alcohol and 3-
methyl-l-penten-3-ol (right). Catalysts used for reactions were (a) 5% Pd-on-
alumina, (b) 7-bilayer PEI-Pd(0)/PAH-on-alumina, and (c) 1-bilayer PAH/PAA
capped 7-bilayer PEI-Pd(0)/PAH-on-alumina. The circled s, p, and i represent

substrate, hydrogenation product, and isomerization product, respectively.

121

 

 

 

 

 

 

 

/\/OH \-=\_OH >=\_OH
/=\_OH
.: A 1L -
® 0
(8)
@@
m GIL it
o 1 5 '3 4 £5 13

Retention time (min)

Figure 6.4: Gas chromatograms from competitive hydrogenations of allyl
alcohol (left), crotyl alcohol (middle), and 3-methyl-1-buten-l-ol (right) on
5%-Pd/alumina (bottom) and on 7-PEI-Pd/PAA/alumina (top). Polyelectrolyte
ﬁlm-immobilized Pd catalysts signiﬁcantly improved hydrogenation
selectivity. The circled s, p, and i represent substrate, hydrogenation product,

and isomerization product.

122

6.3.4 Suppression of isomerization. Substrate isomerization is a common but unwanted
process in hydrogenation, and in some cases, this side reaction is even dominant over
hydrogenation.29 Minimization of isomerization is thus desirable to improve the yield of
hydrogenation reactions. PEPst signiﬁcantly inhibited substrate isomerization, as
shown in Figure 6.5. When the substrate is 99% converted on the 5%-Pd commercial
catalyst, the peaks due to the isomerization of allyl alcohol (to propanal) and l-penten-3-
01 (to 3-pentanone) are signiﬁcant compared to product peaks, 31% and 25%,
respectively. In contrast, when using the 7-PEI-Pd/PAA-on-alumina catalyst, the isomer
peaks decrease to <10% of the product peak. The fraction of isomer decreases to less

than 5% when using PAH/PAA-capped 7-bilayer PEI-Pd(0)/PAA-on-alumina.

 

 

 

 

“'1
__IU ___.AJLA__

1.5 2.0 2.5 3.0 1.5 2.0 2.5 3.0

Retention time (min)

 

 

Figure 6.5: Gas chromatograms of the reaction mixture from hydrogenation of 1-
penten-3-ol on 5%-Pd-on-alumina (left) and on 7-bilayer PEI-Pd(0)/PAA-on-alumina
(right) catalysts. Substrate isomerization was greatly suppressed by the
polyelectrolye ﬁlms. The small unassigned peak in the left chromatograrn is

probably due to an isomerization intermediate.

123

6.4 Conclusions

Formation of a polyethyleneimine-Pd(II) complex followed by layer-by-layer
polycation/polyanion deposition and post-deposition reduction is a convenient method to
synthesize immobilized Pd catalysts. The polyelectrolyte matrix both stabilizes the
particles and introduces selectivity. Moreover, the surrounding polyelectrolyte ﬁlm
signiﬁcantly decreases the amount of unwanted isomerization. Speciﬁc modiﬁcations of
the polyelectrolyte ﬁlm such as cross-linking and capping with PAA/PAH enhance
catalytic selectivity. Further exploitation of the versatility of polyelectrolyte ﬁlms should

increase selectivtivity in hydrogenation as well as other reactions.

124

 

 

6.5 References

1

10

ll

12

13

14

15

l6

l7

l8

Fendler, J. H. Nanoparticles and Nanostructured Films: Preparation,
Characterization and Applications; Wiley-VCH: Weinhein, 1998.

Schmid, G. In Nanoscale Materials in Chemistry; Klabunde, K. J ., Ed.; Wiley-
Interscience: New York, 2001, pp 15-59.

Li, Y.; Boone, E.; El-Sayed, M. A. Langmuir 2002, 18, 4921 -4925.
Zhao, M.; Sun, L.; Crooks, R. M. J. Am. Chem. Soc. 1998, 120, 4877 -4878.
Bdnnemann, H.; Richards, R. M. Eur. J. Inorg. Chem. 2001, 2455-2480.

Rao, C. N. R.; Kulkarni, G. U.; Thomas, P. J .; Edwards, P. P. Chem. Eur. J. 2002, 8,
28-35.

Li, Y.; El-Sayed, M. A. J. Phys. Chem. B 2001, 105, 8938-8943.

Crooks, R. M.; Zhao, M.; Sun, L.; Chechik, V.; Yeung, L. K. Accounts Chem. Res.
2001, 34,181-190.

Underhill, R. S.; Liu, G. Chem. Mater. 2000, 12, 3633-3641.
Ding, J. H.; Liu, D. L. Chem. Mater. 2000, 12, 22-24.

Schmid, G.; Maihack, V.; Lanterrnarm, F. J. Chem. Soc. Dalton Trans. Inorg. Chem.
1996, 589-595.

Niu, Y.; Yeung, L. K.; Crooks, R. M. J. Am. Chem. Soc. 2001, 123, 6840-6846.

Mamedov, A. A.; Belov, A.; Giersig, M.; Mamedova, N. N.; Kotov, N. A. J. Am.
Chem. Soc. 2001, 123, 7738-7739, and references therein.

Lvov, Y.; Munge, B.; Giraldo, 0.; Ichinose, 1.; Suib, S. L.; Rusling, J. F. Langmuir
2000, 16, 8850-8857.

Liu, Y.; Wang, A.; Claus, R. 0. Appl. Phys. Lett. 1997, 71, 2265-2267.

Feldheim, D. L.; Grabar, K. C.; Natan, M. J.; Mallouk, T. E. J. Am. Chem. Soc.
1996, 118, 7640-7641.

Schrof, W.; Rozouvan, S.; Van Keuren, E.; Horn, D.; Schmitt, J.; Decher, 0. Adv.
Mater. 1998, 10, 338-341.

Iler, R. K. J. Coll. Int. Sci 1966, 21 , 569-594.

125

 

19

20

21

22

23

24

25

26

27

28

29

Decher, G. Science 1997, 277, 1232-1237.
Wang, T. C.; Rubner, M. F .; Cohen, R. E. Langmuir 2002, 18, 3370-3375.

Joly, 8.; Kane, R.; Radzilowski, L.; Wang, T.; Wu, A.; Cohen, R. E.; Thomas, E. L.;
Rubner, M. F. Langmuir 2000, 16, 1354-1359.

Fojas, A. M.; Murphy, E.; Stroeve, P. Ind. Eng. Chem. Res. 2002, 41, 2662-2667.
Dante, S.; Advincula, R.; Frank, C. W.; Stroeve, P. Langmuir 1999, 15, 193-201.
Dai, J .; Bruening, M. L. Nano Letters 2002, 2, 497-501.

Sullivan, D. M.; Bruening, M. L. J. Am. Chem. Soc. 2001, 123, 11805-11806.
Sullivan, D. M.; Bruening, M. L. Chem. Mater. 2002, in press.

March, J. Advanced Organic Chemistry: reactions, mechanics, and structure;
Wiley: New York, 1992.

Cerveny, L.; Ruzicka, V. Catal. Rev. Sci. Eng. 1982, 24, 503-566.
Zharrnagambetova, A. K.; Ergozhin, E. E.; Sheludyakov, Y. L.;

Mukhamedzhanova, S. G.; Kurmanbayeva, I. A.; Selenova, B. A.; Utkelov, B. A. .1.
Mol. Catal. A 2001, 1 7 7, 165-170.

126

 

Hydrogenation (%)

100

6.6 Appendix

 

 

801

60"

40‘

 

20‘

 

 

 

 

 

 

0 I I I r I I I
0 1 00 200 300 400 0 100 200 300 400 500

Reaction time (min)

Figure 6.1, appendix. Hydrogenation vs reaction time of substrate 1 (inverted
triangle), 2 (triangle), and 3 (circle) on 1 bilayer PAH/PAA-capped 7-PEI-
Pd/PAA-on-alumina catalyst (left) and on cross-linked 7-PEI-Pd/PAA-on-alumina
catalyst (right). Hydrogenation (%) was obtained from the integration of
hydrogenated product peak of GC spectrum.

127

 

Hydrogenation (%)

 

 

100

80-

604

401

20‘

(b

I
l

{\<

 

 

 

 

 

 

I r

012345678010203040506070

Reaction time (min)

Figure 6.2, appendix. Hydrogenation vs reaction time of substrate 1
(inverted triangle), 4 (square), and 5 (diamond) on 5%-Pd-on-alumina
catalyst (left) and on 7-PEI-Pd/PAA-on-alumina catalyst (right).
Hydrogenation (%) was obtained from the integration of hydrogenated
product peak of GC spectrum

128

 

 

a b c
CH3-CH2-CH2-OH

 

 

 

 

 

 

 

a
c
d e /o
CH3-CH2-C\’
H l o
f
' Ii t e d k.
MW -KN ‘W \JJ
I I I ‘ I I 'I" I ' I I "'Irj"l
10 9 8 7 6 5 4 3 2 1 0
PW“
0H 1
a b |
CH3-CH2-q-CH5-CH3 1
d e
CHrCHi‘ﬁ: ~CH2-CH3
a) e ., a,
\t m A 1 ﬁJ \__.___

 

Figure 6.3, appendix. 1H NMR spectra of the hydrogenation products of allyl
alcohol (top, 7-PEI-Pd/PAA-on-alumina) and 1—penten—3-ol (bottom, 5%—Pd—

on-alumina).

129

Chapter 7

Conclusions and future work

This thesis demonstrates the versatility of multilayer polyelectrolyte ﬁlms in
applications that require ultrathin coatings. Chapter 2 showed that control over the
composition of PAA/PAH ﬁlms through heat-induced amide formation yields highly
passivating ﬁlms. In chapters 3 and 4, we showed that manipulation of polyelectrolyte
composition through derivatization affords control over coating hydrophobicity and
charge density. This allowed enhancement of ion-transport selectivity and modeling of
transport through PAA/PAH ﬁlms and their derivatives. Chapters 5 and 6 explored the
use of multilayer polyelectrolyte ﬁlms for encapsulating catalytic nanoparticles. In
addition to stabilizing nanoparticle size, the polyelectrolyte ﬁlms function as a
discriminating matrix that allows selective hydrogenation.

Although these studies clearly demonstrate the potential of multilayer polyelectrolyte
ﬁlms, real applications of these materials will require signiﬁcant developments in
deposition procedures as well as a better understanding of these systems. Coverage of
porous supports on a large scale and studies of ﬁlm durability must be performed to know
if polyelectrolyte ﬁlms can truly serve as practical separation membranes.

As work with nanoparticles encapsulated in polyelectrolyte ﬁlms is just beginning,
much research is needed to understand these systems. First of all, the mechanisms behind
selective catalysis are not clear. Is selectivity reﬂective of hindered diffusion to the
nanoparticles or different rates of reaction at the nanoparticle surface? Further studies of

hydrogenation as a function of substrate concentration will help to clarify this issue.

130

Additionally, we have only begun to explore the reactions in which Pd nanoparticles
may serve as catalysts. Selective hydrogenation within a molecule may be particularly
interesting. For example, in the hydrogenation of dehydrolinalool, linalool is often the
desired product, because it is an important building block in organic reactions (Scheme
7.1), However, dehydrolinalool is easily over-hydrogenated to dihydrolinalool. Bronstein
and coworkers reported that Pd colloids surrounded by poly(4-vinylpyridine) (PVP) in
polystyrene-block—PVP micelles provide high selectivity for hydrogenation to linalool.1
We expect that Pd nanoparticles in polyelectrolyte ﬁlms will behave similarly and
provide the additional convenience and versatility associated with polyelectrolyte ﬁlms.

This is just one example of the reactions that could be investigated using encapsulated

nanoparticles.
H H H
DHL LN DiHL
dehydrolinalooI linalool dihydrolinalool
Scheme 7.1

This dissertation clearly shows the potential of multilayer polyelectrolyte ﬁlms in
several areas. I am optimistic that with further work, the potential of these ﬁlms will be

truly exploited for practical applications.

1 Bronstein, L. M.; Chemyshov, D. M.; Volkov, I. 0.; Ezernitskaya, M. (3.;
Valetsky, P. M.; Matveeva, V. G.; Sulman, E. M. J. Catalysis 2000, 196, 302-314.

131