DESIGN, FABRICATION, AND UTILIZATION OF SILICA INVERSE OPAL
STRUCTURES FOR FLOW-THROUGH CATALYST SUPPORTS
By
Douglas B. Gornowich

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
Chemistry - Doctor of Philosophy
2013

ABSTRACT
DESIGN, FABRICATION, AND UTILIZATION OF SILICA INVERSE OPAL
STRUCTURES FOR FLOW-THROUGH CATALYST SUPPORTS
By
Douglas B. Gornowich
Enzymes are used in a wide range of industries for various different chemical processes.
Optimizing the performance of enzymes remains an area of high interest in many research labs.
An enzyme is often immobilized on or within a support structure, which allows for the
biocatalyst to be relatively easy to recover post-reaction. Immobilization can also increase the
structural stability of the enzyme, which is beneficial from a cost standpoint because pure
enzymes can be expensive. By choosing the appropriate support and immobilization chemistry,
it is possible to maximize the efficiency of a biocatalyst. The purpose of this work was to
design, fabricate and utilize inverse opal structures as a support for the immobilization of
enzymes.
We have developed a flow-through silica inverse opal structure that was used for the
immobilization of biocatalysts. The inverse opal structures were created using polystyrene
nanospheres as a template, sol-gel chemistry to deposit silica in the interstitial spaces between
the nanospheres, and solvent dissolution to remove the template. Scanning electron microscopy
and dynamic light scattering were used to characterize the nanospheres and structures. The silica
inverse opal structure has a relatively high surface area, and a surface that is amenable to a wide
range of surface modification reactions. Two enzymes were chosen to evaluate our catalyst
support structure; glucose oxidase and alkaline phosphatase. Absorbance and fluorescence
measurements were used for the enzyme assays.

Our results show an enhancement in reactivity that is associated with enzyme
immobilization and nano-confinement, and also underscore limitations inherent to this approach.
Three different reaction formats were examined: solution phase, immobilized enzymes on planar
supports, and enzymes immobilized on the flow-through inverse opal structure. Glucose oxidase
exhibited an increase in reactivity when comparing planar vs. solution phase and inverse opal vs.
planar structural formats. This finding indicated an enhancement due to the immobilization
process and due to the nanoconfinement of the enzyme within the inverse opal structure. In
contrast, alkaline phosphatase exhibited a reduced activity when comparing solution phase vs.
enzyme immobilized on planar and inverse opal structures. This finding illustrated the
importance of identifying immobilization chemistry that maintains the enzyme in an active form
and binds the enzyme in a way that leaves the reactive site accessible. An enhancement was
observed for the inverse opal structure vs. the planar support, indicating that there remains the
positive effect associated with nano-confinement of the enzyme.
This project proved to be enlightening by showing the enhancements in activity for
glucose oxidase, and also by showing that there are limitations that need to be addressed in the
alkaline phosphatase results. There will be continued work to further characterize and optimize
the flow-through inverse opal structures. In addition, it may be useful to examine the use of
other materials for the inverse opal support itself. The results of this work are promising for the
utilization of inverse opal structures to immobilize and optimize the performance of enzymes.

ACKNOWLEDGEMENTS

I would like to acknowledge a few people that have helped me grow as a scientist and a
person over the course of my graduate career. First and foremost, I would like to thank my
graduate research advisor, Dr. Gary J. Blanchard. His door was always open, which allowed for
an open line of communication throughout my time spent at MSU. I appreciate the thoughtful
insight that he provided as well as the constructive criticism that helped to mold me into the
scientist that I am today. I would also like to thank my guidance committee for all of their
assistance along the way; Dr. Greg Swain, Dr. Merlin Bruening and the late Dr. Greg Baker. In
addition, I would like to thank Dr. John McCracken for stepping in as a committee member for
my defense.
I would also like to thank all of the past and current Blanchard group members that were
there to lend a helping hand. I would like to specifically thank the following past members; Dr.
Monique Lapinski, Dr. Kelly Greenough, Dr. Janelle Newman, Dr. Alexis Blevins, Dr. Benjamin
Oberts, Dr. Monika J. Dominska, Dr. Heather A. Pillman, and Dr. Margaretta Koster. I would
also like to thank the following members of the group that have helped me throughout my time at
MSU; Christine Hay, Iwan Setiawan, Chen Qiu, Fredy Pratama, Stephan Baumler, and Matthew
Oliver.
I would also like to express gratitude to a list of groups and facilities that allowed me to
utilize their instrumentation for my graduate work. I would like to thank the folks over at the
Michigan State University Center for Advanced Microscopy facility for allowing me to use the
scanning electron microscope. I would also like to thank Dr. Xuefei Huang and his research
group for allowing me access to their dynamic light scattering instrument. Finally, I would like

iv

to thank Mike Rich for granting me access to the instruments at the Composite Materials and
Structure Center at MSU.
Last, but certainly not least, I would like to thank all of my family and friends that
provided me with encouragement and support throughout my journey in graduate school. I
would especially like to thank my wife, Sheila, for being understanding and positive when things
were tough. I simply could not have done it without her support.

v

TABLE OF CONTENTS

LIST OF TABLES ....................................................................................................................... viii
LIST OF FIGURES ....................................................................................................................... ix
LIST OF SCHEMES.................................................................................................................... xiii
CHAPTER 1: Introduction to Enzyme Immobilization and Inverse Opal Structures .................... 1
Introduction................................................................................................................................. 1
Enzyme Immobilization ............................................................................................................... 1
Inverse Opals .............................................................................................................................. 6
Model Enzymes ......................................................................................................................... 13
Literature Cited......................................................................................................................... 17
CHAPTER 2: Fabrication of Inverse Opal Structures .................................................................. 23
Introduction............................................................................................................................... 23
Polystyrene Sphere Synthesis.................................................................................................... 25
Polystyrene Sphere Deposition Techniques .............................................................................. 39
Sol-gel Methods for Silica Deposition ...................................................................................... 42
Polystyrene Sphere Removal..................................................................................................... 45
Flow-through Nanoporous Solid Structure Fabrication .......................................................... 49
Conclusions ............................................................................................................................... 53
Literature Cited......................................................................................................................... 54
CHAPTER 3: Enhancement of GOx Activity by Confinement in an Inverse Opal Structure ..... 58
Introduction............................................................................................................................... 58
Experimental ............................................................................................................................. 60
Results and Discussion ............................................................................................................. 63
Conclusions ............................................................................................................................... 76
Literature Cited......................................................................................................................... 78
CHAPTER 4: Evaluating the Catalytic Efficiency of Alkaline Phosphatase Confined in an
Inverse Opal Structure .................................................................................................................. 82
Introduction............................................................................................................................... 82
Experimental ............................................................................................................................. 84
Results and Discussion ............................................................................................................. 86
Conclusions ............................................................................................................................. 114
Literature Cited....................................................................................................................... 115
CHAPTER 5: Conclusions ......................................................................................................... 118
Summary ................................................................................................................................. 118
Future Directions .................................................................................................................... 120
vi

Overall Conclusions................................................................................................................ 122
Literature Cited....................................................................................................................... 123

vii

LIST OF TABLES

Table 3.1 Comparison of GOx reactivity for selected catalytic systems. For all measurements
reported in this Table, the glucose concentration was 100 mM. .................................................. 71
Table 4.1. Comparison of different substrates for various reaction formats. (Uncertainties are
±1σ) ............................................................................................................................................. 106
Table 4.2. Relative turnover rates for the substrates and ALP in different formats. The turnover
rates are normalized to a solution phase rate of 28.8/s for 4-MBP. ............................................ 112

viii

LIST OF FIGURES

Figure 1.1: Illustration of three different enzyme immobilization techniques; binding to a
prefabricated support (A), entrapment (B), and enzyme aggregation/crystals (C). Note, this is
just to illustrate the different methods and is not drawn to scale. For interpretation of the
reference to color in this and all other figures, the reader is referred to the electronic version of
this dissertation. .............................................................................................................................. 4
Figure 1.2: Illustration of the steps involved in the fabrication of an inverse opal structure. ....... 7
Figure 1.3: Enzyme immobilization using (3-aminopropyl)triethoxysilane (APTES)
and glutaraldehyde to cross-link amine groups. ........................................................................... 11
Figure 1.4: Plot of 2√(Dt)/d vs. d to indicate the enhanced probability of substrate-enzyme
interactions (t = 1 sec). The dashed line highlights the d used for the experimental conditions in
this work........................................................................................................................................ 12
Figure 2.1: Schematic for the fabrication of an inverse opal on a planar substrate. .................... 24
Figure 2.2: SEM image of polystyrene spheres with a bimodal size distribution. ...................... 29
Figure 2.3: SEM image of polystyrene spheres with an average diameter of 900 nm synthesized
using methanol as the solvent. ...................................................................................................... 31
Figure 2.4: SEM image of polystyrene spheres with an average diameter of 1.1 μm synthesized
using ethanol as the solvent. ......................................................................................................... 32
Figure 2.5: SEM image of polystyrene spheroidal structures with an average diameter of 2.2 μm
synthesized using ethanol as the solvent....................................................................................... 33
Figure 2.6: Plot of polystyrene sphere size vs. steric stabilizer concentration (top left) for free
radical dispersion polymerization reactions with ethanol as the solvent. SEM images (top right
and bottom) of polystyrene spheres at each PVP concentration. .................................................. 34
Figure 2.7: SEM image of polystyrene spheres with an average diameter of 180 nm synthesized
using an emulsifier polymerization reaction method. ................................................................... 35
Figure 2.8: SEM image of polystyrene spheres with a diameter of ca. 1 μm deposited on a glass
microscope slide that were synthesized using a free radical dispersion polymerization method. 37
Figure 2.9: Dynamic light scattering data showing the particle size distribution of the
polystyrene spheres with an average diameter of 1.15 μm. .......................................................... 38

ix

Figure 2.10: SEM image of polystyrene spheres with an average diameter of 1 μm deposited on
a glass microscope slide using solvent evaporation...................................................................... 40
Figure 2.11: SEM image of polystyrene spheres with an average diameter of 1 μm deposited on
a glass microscope slide using a spin-coater................................................................................. 41
Figure 2.12: SEM image of polystyrene spheres with an average diameter of 1 μm deposited on
a glass microscope slide using the nanodipper at a rate of 2 μm/sec. ........................................... 43
Figure 2.13: SEM images of silica inverse opal structures on a glass microscope slide. The
spheres and sol-gel chemistry were performed in subsequent steps by adding solvent evaporation
and adding the precursor solution on drop wise. .......................................................................... 46
Figure 2.14: SEM image of a silica inverse opal structure on a glass microscope slide. The
spheres and sol-gel chemistry were performed in subsequent steps using the nanodipper at 2
μm/sec. The spheres were deposited using 3 consecutive withdrawals from a suspension,
sintered at 65°C for 1 hour, and then withdrawn from the sol-gel precursor solution 1 time.
Toluene was used to remove the spheres. ..................................................................................... 47
Figure 2.15: SEM image of an inverse opal that utilized heat to remove the polystyrene scaffold.
This image was taken at low magnification to show the major cracking caused by the burning
procedure....................................................................................................................................... 48
Figure 2.16: SEM image of an inverse opal that was synthesized on a gold TEM grid (1500
mesh). ............................................................................................................................................ 50
Figure 2.17: SEM images of an inverse opal structures that were synthesized on PAS. The top
image represents an inverse opal formed using subsequent steps for the sphere and sol-gel
depositions, and the bottom micrograph represents an inverse opal formed using the co-assembly
technique. ...................................................................................................................................... 52
Figure 3.1: (a) Dynamic light scattering data showing the particle size distribution of the
polystyrene spheres used to template the inverse opal structures used as enzyme supports. (b)
SEM micrograph of the polystyrene spheres. ............................................................................... 65
Figure 3.2: Schematic of the NS/PAS assembly. Top, SEM micrographs of inverse opal, PAS,
and interfacial region, and bottom, SEM micrograph of an on-edge view of the NS/PAS that was
fractured. ....................................................................................................................................... 66
Figure 3.3: TGA data for NS/PAS. The data were normalized to zero mass at 640°C. ............. 68
Figure 3.4: Investigation of the effect of Fe
with Fe

2+/3+

2+/3+

on GOx activity. Solution phase measurements

present during reaction (■) and in the absence of Fe

2+/3+

(∆). ............................... 72

Figure 3.5: Absorbance values at 340 nm for the NS/PAS at different flowrates. ...................... 74
x

Figure 4.1: ALP immobilization chemistry used for planar and NS/PAS support experiments. 88
Figure 4.2: Reactions for ALP with 1-NP (top), 2-NP (middle), and 4-MBP (bottom). ........... 90
Figure 4.3: Fluorescence excitation and emission spectra for 4-MBP (dotted) and 4-MB (solid),
λ ex = 365 nm and λ em = 448 nm for each. ................................................................................... 91
Figure 4.4: Fluorescence excitation and emission spectra for 1-NP (dotted) and 1-naphthol
(solid), λ ex = 332 nm and λ em = 460 nm for each. ....................................................................... 92
Figure 4.5: Fluorescence excitation and emission spectra for 2-NP (dotted) and 2-naphthol
(solid), λ ex = 352 nm and λ em = 408 nm for each. ....................................................................... 93
Figure 4.6: Absorbance spectra for 4-MB/4-MBP (top), 1-naphthol/1-NP (middle), and 2naphthol/2-NP (bottom). ............................................................................................................... 94
Figure 4.7: (a) Time-dependent fluorescence from the reaction of 4-MBP with ALP at different
-7
4-MBP initial concentrations, as indicated in the legend. [ALP] = 2.08x10 M. (b)
Concentration-dependence of long-time fluorescence intensity from the same reactions. .......... 95
Figure 4.8: (a) Time-dependent fluorescence from the reaction of 1-NP with ALP at different 1-7
NP initial concentrations, as indicated in the legend. [ALP] = 5.21x10 M. (b) Concentrationdependence of long-time fluorescence intensity from the same reactions. .................................. 96
Figure 4.9: (a) Time-dependent fluorescence from the reaction of 2-NP with ALP at different 2-7
NP initial concentrations, as indicated in the legend. [ALP] = 5.21x10 M. (b) Concentrationdependence of long-time fluorescence intensity from the same reactions. .................................. 97
Figure 4.10: (a) Initial rate data for the reaction of 4-MBP with ALP at different 4-MBP initial
-7
concentrations, as indicated in the legend. [ALP] = 2.08x10 M. (b) Concentration-dependence
of the initial rate data for the same reactions. ............................................................................. 100
Figure 4.11: (a) Initial rate data for the reaction of 1-NP with ALP at different 1-NP initial
-7
concentrations, as indicated in the legend. [ALP] = 5.21x10 M. (b) Concentration-dependence
of the initial rate data for the same reactions. ............................................................................. 101
Figure 4.12: (a) Initial rate data for the reaction of 2-NP with ALP at different 2-NP initial
-7
concentrations, as indicated in the legend. [ALP] = 5.21x10 M. (b) Concentration-dependence
of the initial rate data for the same reactions. ............................................................................. 102
Figure 4.13: Overview of the construction of the NS/PAS supports used for ALP
immobilization (top) and a cross-section SEM image of the NS/PAS interface (bottom). ........ 104

xi

Figure 4.14: Calibration data for the reaction products used in this work. Slopes are the
sensitivity factors for our system, given in counts/M and are best fits of data points contained in
the linear range for each reaction product................................................................................... 109
Figure 4.15: Time-dependent fluorescence data for ALP immobilized on planar silica for each of
-6
the three different substrates. The concentration of the substrate was 9.17x10 M for all cases.
..................................................................................................................................................... 110

xii

LIST OF SCHEMES

Scheme 1.1: GOx reaction, oxidation of β-D-glucose to form D-gluconolactone, which is then
spontaneously hydrolyzed to D-gluconic acid. The reduced enzyme reacts with oxygen to
regenerate GOx and produces hydrogen peroxide. ....................................................................... 15
Scheme 1.2: Non-specific dephosphorylation of a phosphate monoester by reacting with alkaline
phosphatase. .................................................................................................................................. 16
Scheme 2.1: Reaction schematic for a free radical dispersion polymerization reaction for
polystyrene sphere synthesis. PVP was present in the reaction mixture as a stabilizer............... 27
Scheme 2.2: Overall reaction schematic for the sol-gel process used to fill in the interstitial
space with a TEOS precursor solution. Hydrolysis (top), water condensation (middle), and
alcohol condensation (bottom) reactions. ..................................................................................... 44

xiii

CHAPTER 1: Introduction to Enzyme Immobilization and Inverse Opal Structures
Introduction
1,2

Enzyme immobilization is a key area of interest in many research labs.

Enzymes are

used in a wide variety of industries, including dairy (cheese), beverage (fermentation), biofuels
(fermentation), and food safety (diagnostic test kits). Consequently, optimizing the performance
of the various enzymes used in these processes is important. Maximizing the performance of an
enzyme is a balance of maximizing reactivity and while maintaining long term stability of the
enzyme. The two main areas of interest in this field are examining the different techniques used
to immobilize the enzyme and the design of the material/support used to immobilize the enzyme.
Designing a support that would increase the performance of a given enzyme would be a valuable
contribution to science, and this dissertation reports on an investigation utilizing inverse opal
structures as a scaffold to immobilize enzymes. This Chapter will discuss enzyme
immobilization, give an overview of inverse opal structures, sometimes referred to as
nanoporous solids (NS), and introduce the enzyme/NS combinations that were used in this work.
Chapter 2 will focus on the fabrication of inverse opal structures on planar and porous supports.
Chapters 3 and 4 provide examples of the utility and limitations of NS structures used as flowthrough heterogeneous catalytic reactors for two different enzymes. Chapter 5 will include
overall conclusions and future directions for this project.
Enzyme Immobilization
As mentioned above, the immobilization of enzymes often results in enhanced
performance relative to that in the solution phase for a given protein. There are, however, a few
areas of concern regarding the immobilization process. The standard by which enzymes are

1

evaluated in an industrial setting, to maximize the cost efficiency of a process, is a ratio
quantitating the kg of product per kg of enzyme. Ideally, enzymes should not cost more than a
few percent of the total production costs.

2

Immobilizing enzymes provides several advantages

over solution phase enzymatic reactions including: facile separation of the enzyme and
reactants/products, reduction of protein contamination in the product, ease of enzyme recovery
and reuse, increased structural stability of the biocatalyst, and enzyme denaturation that may be
caused by products formed during the reaction can be minimized when using a flow-through
system. Another advantage to immobilizing the enzyme is the possibility of arranging sequences
of enzymes in specific order(s) to perform cascaded chemical reactions.

3,4

The reactivity of the

enzyme is often affected when it is tethered to a support structure. Depending on the manner in
which the enzyme is immobilized, it is possible to enhance the reactivity of the biocatalyst
relative to that in the solution phase.

5-7

There are, however, some potential disadvantages to the immobilization process. The
most prominent issue to contend with is ensuring that the immobilization method does not make
the active site of the protein inaccessible to substrate molecules. If the enzyme is attached in a
manner where the active site is not readily accessible, its reactivity will be decreased, sometimes
markedly, relative to that in the solution phase. Another issue that may arise for certain
biocatalysts is steric crowding of the proteins. This can be a problem if the steric effects alter the
conformation of the enzyme in a way that changes the conformation of or access to the binding
pocket. Using various surface chemistry immobilization techniques and inserting chemical
spacers between enzymes can often minimize the aforementioned limitations encountered when
8

biocatalysts are immobilized.

2

There are three main techniques used for enzyme immobilization: binding to a
prefabricated support structure, encapsulation or entrapment within a material, and aggregation
of enzymes due to covalent cross-linking. The first two approaches can involve biocatalysts that
are either physisorbed or bonded covalently to the support matrix. The main distinction between
the two techniques is that the enzyme is attached to the support in sequential steps in the first
case. For the entrapment method, the enzyme and support are formed simultaneously, resulting
in an enzyme/support matrix. The cross-linking methodology is completely different than the
other two techniques because the enzyme itself is used as the support structure. Fig. 1.1 provides
a simple illustration of the different methods, and the following three paragraphs will discuss
each system in more detail.
Enzymes bound to a pre-fabricated carrier exhibit certain advantages when it comes to
immobilization. The biocatalysts can be either physisorbed or covalently attached to the surface,
which ensures that the proteins are accessible to substrate molecules. However, the surface
attachment used is an important factor in determining the conformation of the protein when it is
attached to the surface, and this will be discussed later in this chapter. There have been several
different prefabricated materials reported in the literature used for enzyme immobilization.
9

10

These supports include acrylic resins, hydrogels

and polymers.

11,12

A major focus has also

been centered on utilizing selected inorganic supports as enzyme carriers, including nanoporous
5-7,13,14

and mesoporous silica,
20

macroporous titania.

15,16

silica nanospheres,

zeolites,

17,18

19

alumina,

and

Many of these support structures provide an active surface for the

attachment of enzymes, but a major drawback for many of them are expense and difficulty in
fabrication.

3

A

B

C

Figure 1.1: Illustration of three different enzyme immobilization techniques; binding to a
prefabricated support (A), entrapment (B), and enzyme aggregation/crystals (C). Note, this is
just to illustrate the different methods and is not drawn to scale. For interpretation of the
reference to color in this and all other figures, the reader is referred to the electronic version of
this dissertation.

4

21-27

The second method for immobilizing enzymes is entrapment of the biocatalyst.

For

this technique, enzymes and the entrapping matrix material are typically co-assembled in a onepot reaction. The primary advantage of this technique is that high enzyme loadings are
achievable. The physical interactions between the enzyme and the material are relatively weak
and, as a result, enzyme leakage can be a problem. The way many researchers get around this is
to include reactive functional groups within the matrix to facilitate covalent attachment between
the biocatalyst and the surrounding material. Entrapment is typically performed using organic
21,22

polymers,

23

ionic liquids

24-27

or sol-gel based metal oxides.

The main limitation in this

work is that due to the nature of the encapsulating process, a large amount of biocatalysts can be
buried within the matrix. If the active sites are rendered inaccessible by virtue of enzyme
entrapment within the matrix, much lower activity may be observed. Entrapment and
prefabricated supports are often referred to the enzyme “carrier” matrix.
The third class of enzyme immobilization techniques is cross-linked enzyme aggregates
or crystals (carrier-free) where the enzyme itself is the carrier.

28-31

For this method, bioactive

proteins are cross-linked together to form insoluble aggregates or crystals that can be physically
28-31

separated from solution.

This approach dates back to 1964 when it was discovered that
32,33

surface NH2 groups could be cross-linked using glutaraldehyde.

Since then, there have

been several attempts at improving crystallization techniques, including control over the pore
34

size in the crystals.

Other efforts have been made to simplify the separation of the crystals by
35

attaching them to magnetic particles.

The main drawback to this approach is the

crystallization process itself, because of the requirement of high purity enzymes and

5

consequently high cost. Another inherent disadvantage of this approach is a result of the
physical issue of a large amount of enzyme residing within the crystal structure and thus being
inaccessible. Such enzyme is essentially used as filler material and is not catalytically active.
A great deal of research is currently ongoing and focused on addressing the limitations of
each of the three different approaches.

1,2

Carrier-bound techniques on pre-fabricated supports

have been the gold standard over the past decades, and continue to be at the forefront of
biocatalysis systems. Recent advances in encapsulation and aggregation techniques have shown
promise. The next section of this Chapter will detail a novel approach that uses dielectric inverse
opal materials as support structures for enzyme immobilization. There are several benefits that
accrue from the use of this structural motif, including confinement of the enzymes in a nanoscale
volume and the chemical functionality of the matrix itself for use in tethering enzymes.
Inverse opals
5,36-44

Inverse opal materials have been studied extensively over the last decade.

An

inverse opal structure is fabricated using a process that is illustrated in Fig. 1.2. In general, the
procedure of how such nanoporous solids are fabricated is described by the following steps: 1)
nanospheres are deposited in an organized assembly on a support, 2) the interstitial space
between the spheres is filled with a different material, and 3) the nanospheres are removed,
leaving an inverse “image” of the nanosphere assembly which is characterized by openings
(nanosphere touch points) that allow for facile flow through the structure. It is also possible to
combine the first two steps and co-deposit the nanospheres in a suspension of the filler material.
The properties of these materials will be discussed in this section, including the wide variety of
options for composition of the matrix, the moderately high surface area, and control over the

6

A

Cleaned bare support

B

nanosphere
deposition

matrix growth
D

C

nanosphere
removal

Figure 1.2: Illustration of the steps involved in the fabrication of an inverse opal structure.

7

pore size. The utility of inverse opals for use as supports for enzyme immobilization will be
illustrated in Chapters 3 and 4. There are, of course, limitations to the use of such structures, and
they are considered in this work as well.
There are many different possibilities for the nanosphere and filling matrix materials,
allowing for the use of these structures in a variety of different applications. Typically, either
organic polymer or inorganic oxide spheres are used for the formation of the nanosphere
scaffold. Polystyrene and silica are the nanospheres used most widely.

5,39,41-45

There have

been several different materials used to fill the void spaces between nanospheres. For dielectric
36,46

materials, silica,

39,40,46-48

titania,

46,48

zirconia,

46,48

alumina,

41-43,50-52

reported. Metals have also been used widely, including Pt,
50,53,54

Ni,

50

and Cu.

49

and composites
41-43,51

Pd,

Au,

have been

37,52

Inorganic oxide inverse opals are typically formed with the use of sol-gel

chemistry, and metals are electrodeposited into the interstitial space. The ability to choose the
matrix material makes inverse opal structures amenable for a variety of different uses.
Another characteristic property of inverse opals is that they have a relatively high
surface-area-to-volume (SA/V) ratio. The surface area can be calculated based on the
dimensions and geometric arrangement of the nanospheres comprising the scaffold structure.
Voids are designated as the spaces vacated by the nanospheres that had been used as the scaffold,
and the openings left by the nanosphere contact points are termed pores. The pore size is
determined by the size of the spheres used for the scaffold, a relationship that implies very
sensitive control over the characteristic openings within the matrix. Kuhn and coworkers have
shown that the diameter of the pores are ca. 20% of the diameter of the nanospheres.

8

37

For a

hexagonal close-packed structure, each sphere on the first layer would have 9 contact points with
neighboring spheres and one pore on the bottom (with the support), and each subsequent layer
would have 12 contact points. The surface area (SA) of each void on the first layer can be
calculated by SAvoid = (SAnanosphere – 10Apore ) and SAvoid = (SAnanosphere – 12Apore ) for
2

2

subsequent layers. SA nanosphere is 4πr and the Apore is estimated to be 0.2πr for the circular
pores. For illustrative purposes, assume that an inverse opal structure is fabricated using
nanospheres with a diameter of 500 nm. Consequently, the pores would have a diameter of 100
3

4

nm. For 1cm of material, there would be 2x10 voids along each side of the material, resulting
8

2

in 4x10 voids per layer. Using the equations above, the surface area would be 1.57 cm for the
2

4

first layer and 1.26 cm for each subsequent layer. With 2x10 layers, the resulting total surface
4

2

3

area is 2.5x10 cm per geometric cm of material. It is worth noting that the surface area for
each layer is independent of nanosphere size, however using smaller spheres will allow for more
layers to be deposited in a given volume.
The relatively high surface area and wide range of matrix materials make inverse opals an
auspicious carrier for enzyme immobilization. For all of the work presented in this dissertation,
silica was used as the matrix material and it was deposited using sol-gel chemistry. The details
of the NS structure fabrication are outlined in Chapter 2 of this dissertation. The focus in
Chapter 1 is on the rationale for the design of the inverse opals used. The moderately high SA/V
ratio affords a relatively high loading density of biocatalysts to be achieved in a limited volume
despite the limitation of a single monolayer of coverage within the matrix (vide infra). In
addition, silica is amenable to a wide range of surface modification chemistry for enzyme

9

immobilization. Most of the enzyme attachment strategies can be performed under relatively
mild reaction conditions, minimizing enzyme denaturation during immobilization. Enzyme
immobilization by glutaraldehyde (Fig. 1.3) cross-linking of primary amines on the support to
primary amines on the surface of the enzyme molecules is a well-established and relatively
55-60

efficient method.

Having the enzyme immobilized on a support structure is desirable

because the enzyme can be stored in buffer and can have prolonged stability.

58

Another appealing characteristic of NS structures is that they are sufficiently porous to
5

allow for flow-through designs to be developed. Confining proteins in small spaces has been
61

shown to increase the stability of the molecule.

When substrate-containing solutions are

flowed through the NS, it is possible that enhanced reactivity relative to that in the solution phase
can be observed by increasing the number of substrate-enzyme interactions based on geometric
confinement within the inverse opal structure. A plot of characteristic diffusion length of a
substrate molecule relative to the diameter of the nanoscale void space of the inverse opal (Fig.
1.4), 2√(Dt)/d vs. d, illustrates the increased relative probability of reactive substrate-enzyme
interactions for the NS systems, where D is the diffusion coefficient for the substrate molecule
2

(cm /sec), t is time (sec), and d is the diameter of the nanospheres used. For the flow rates used
in these experiments, residence times for the substrate molecules within the NS are on the order
of 5-10 seconds. From this model it is clear that the smaller the void volume, the higher the
probability of a substrate-enzyme reaction event occurring. Another important factor to consider
in a flow-through structural motif is that the products of the reaction are swept away from the
enzyme. This can be advantageous for certain enzymes such as glucose oxidase, where one of

10

NH2 NH2 NH2 NH2
OH

OH

OH

OH

APTES

SiO2

SiO2

NH2

Glutaraldehyde
O

O

O Si O
O

NH2
+
N

N

N

N

N

N

N

O

N

N

N

O

N

O

N

SiO2

SiO2

Figure 1.3: Enzyme immobilization using (3-aminopropyl)triethoxysilane (APTES) and
glutaraldehyde to cross-link amine groups

11

O

2(Dt)1/2/d

1000

100

10

0

2000

4000

6000

8000

10000

d (nm)

Figure 1.4: Plot of 2√(Dt)/d vs. d to indicate the enhanced probability of substrate-enzyme
interactions (t = 1 sec). The dashed line highlights the d used for the experimental conditions in
this work.

12

the products (H2 O 2 ) of the reaction can denature the native enzyme over prolonged exposure
62-64

times.

With all of the previously mentioned potential benefits in mind, there are a few
limitations to the use of NS structures as support structures for the attachment of enzymes. The
biggest challenge is to create the NS in a physically robust format that will allow a reactant
stream to flow through the matrix. Inverse opal structures are very fragile by nature and a
support is required to help maintain the integrity of the structure. Chapter 2 addresses this issue
in more detail. Another limitation is associated with the fragile structure and small pore size that
limits the flow-rates that can be used. An issue specific to dielectric inverse opals is the
quantitation of the enzyme loading in the NS matrix. Spectroscopic techniques are difficult to
65

employ due to the inverse opals acting as photonic crystals.

Due to the alternating dielectric

constant within the matrix, photonic band gaps are seen at locations in the UV and visible
spectrum, depending on the dimensions of the inverse opal structure. Thus performing
quantitative spectroscopic measurements is a challenge in such media. Gravimetric methods can
provide some quantitative data, but the measurements can be time consuming and challenging, in
both execution and interpretation.
Model Enzymes
Glucose oxidase (GOx) and alkaline phosphatase (ALP) were chosen as model enzymes
for our inverse opal flow-through catalytic nano-reactors. These two enzymes were not chosen
with any specific industrial application in mind, but rather as starting points to illustrate the
potential and limitations of this approach to the immobilization of biocatalytic systems. These
enzymes are readily available for purchase and are relatively inexpensive. GOx (Aspergillus

13

niger) and ALP (bovine intestinal mucosa) are similar in size (160 kDa), yet they are different
with respect to substrate specificity. GOx is an enzyme that exhibits a high specificity for β-Dglucose and the reaction is shown in Scheme 1.1. In contrast, ALP dephosphorylates phosphate
monoesters into alcohols with very little substrate specificity (Scheme 1.2). It was our hope that
these two enzymes would be sufficient to demonstrate our flow-through system and provide
some insight into how this structural motif for flow-through supported catalytic reactions could
be improved. Three different reaction formats were examined; free enzymes in solution, enzyme
immobilized on planar solid supports, and enzyme attached to nanoporous solid structures.
These results are presented in Chapter 3 for GOx and Chapter 4 for ALP and, taken together,
they demonstrate some of the advantages and limitations associated with this approach to
enzyme immobilization.

14

H2O2
+
+ GOxH2

+ GOx

O2

Scheme 1.1: GOx reaction, oxidation of β-D-glucose to form D-gluconolactone, which is then
spontaneously hydrolyzed to D-gluconic acid. The reduced enzyme reacts with oxygen to
regenerate GOx and produces hydrogen peroxide.

15

O
R

O

P
P
OH

O
ALP
OH

R

H2O

OH

+

HO

P
P
OH

OH

Scheme 1.2: Non-specific dephosphorylation of a phosphate monoester by reacting with alkaline
phosphatase.

16

Literature Cited

17

Literature Cited
1. Hwang, E. T.; Gu, M. B. Engineering in Life Sciences 2013, 13, 49.
2. Sheldon, R. A. Advanced Synthesis & Catalysis 2007, 349, 1289.
3. Bruggink, A.; Schoevaart, R.; Kieboom, T. Organic Process Research & Development 2003,
7, 622.
4. Veum, L.; Hanefeld, U. Chemical Communications 2006, 0, 825.
5. Gornowich, D. B.; Blanchard, G. J. The Journal of Physical Chemistry C 2012, 116, 12165.
6. Lei, C.; Soares, T. A.; Shin, Y.; Liu, J.; Ackerman, E. J. Nanotechnology 2008, 19, 1.
7. Lei, C.; Shin, Y.; Liu, J.; Ackerman, E. J. Journal of the American Chemical Society 2002,
124, 11242.
8. Manta, C.; Ferraz, N.; Betancor, L.; Antunes, G.; Batista-Viera, F.; Carlsson, J.; Caldwell, K.
Enzyme and Microbial Technology 2003, 33, 890.
9. Katchalski-Katzir, E.; Kraemer, D. M. Journal of Molecular Catalysis B: Enzymatic 2000, 10,
157.
10. Temiño, D. M.-R. D.; Hartmeier, W.; Ansorge-Schumacher, M. B. Enzyme and Microbial
Technology 2005, 36, 3.
11. Ivanov, A. E.; Edink, E.; Kumar, A.; Galaev, I. Y.; Arendsen, A. F.; Bruggink, A.;
Mattiasson, B. Biotechnology Progress 2003, 19, 1167.
12. de Oliveira, P. C.; Alves, G. M.; de Castro, H. F. Biochemical Engineering Journal 2000, 5,
63.
13. Moelans, D.; Cool, P.; Baeyens, J.; Vansant, E. F. Catalysis Communications 2005, 6, 307.
14. Borole, A.; Dai, S.; Cheng, C.; Rodriguez, M., Jr.; Davison, B. Applied Biochemistry and
Biotechnology 2004, 113, 273.

18

15. Kim, M. I.; Ham, H. O.; Oh, S.-D.; Park, H. G.; Chang, H. N.; Choi, S.-H. Journal of
Molecular Catalysis B: Enzymatic 2006, 39, 62.
16. Wang, Y.; Caruso, F. Chemical Communications 2004, 1528.
17. Díaz, J. F.; Balkus Jr, K. J. Journal of Molecular Catalysis B: Enzymatic 1996, 2, 115.
18. Yan, A.-X.; Li, X.-W.; Ye, Y.-H. Applied Biochemistry and Biotechnology 2002, 101, 113.
19. Reshmi, R.; Sanjay, G.; Sugunan, S. Catalysis Communications 2006, 7, 460.
20. Zhu, Y.; Cao, H.; Tang, L.; Yang, X.; Li, C. Electrochimica Acta 2009, 54, 2823.
21. Besic, S.; Minteer, S. In Enzyme Stabilization and Immobilization; Minteer, S. D., Ed.;
Humana Press: 2011; Vol. 679, p 113.
22. Kobayashi, J.; Mori, Y.; Kobayashi, S. Chemical Communications 2006, 0, 4227.
23. Nakashima, K.; Kamiya, N.; Koda, D.; Maruyama, T.; Goto, M. Organic & Biomolecular
Chemistry 2009, 7, 2353.
24. Braun, S.; Rappoport, S.; Zusman, R.; Avnir, D.; Ottolenghi, M. Materials Letters 1990, 10,
1.
25. Avnir, D. Accounts of Chemical Research 1995, 28, 328.
26. Avnir, D.; Braun, S.; Lev, O.; Ottolenghi, M. Chemistry of Materials 1994, 6, 1605.
27. Gill, I. Chemistry of Materials 2001, 13, 3404.
28. Cao, L.; van Rantwijk, F.; Sheldon, R. A. Organic Letters 2000, 2, 1361.
29. Roy, J. J.; Abraham, T. E. Journal of Molecular Catalysis B: Enzymatic 2006, 38, 31.
30. St. Clair, N. L.; Navia, M. A. Journal of the American Chemical Society 1992, 114, 7314.

19

31. Zelinski, T.; Waldmann, H. Angewandte Chemie International Edition in English 1997, 36,
722.
32. Quiocho, F. A.; Richards, F. M. Proceedings of the National Academy of Sciences of the
United States of America 1964, 52, 833.
33. Quiocho, F. A.; Richards, F. M. Biochemistry 1966, 5, 4062.
34. Talekar, S.; Shah, V.; Patil, S.; Nimbalkar, M. Catalysis Science & Technology 2012, 2,
1575.
35. Tudorache, M.; Nae, A.; Coman, S.; Parvulescu, V. I. RSC Advances 2013, 3, 4052.
36. Hatton, B.; Mishchenko, L.; Davis, S.; Sandhage, K. H.; Aizenberg, J. Proceedings of the
National Academy of Sciences 2010, 107, 10354.
37. Szamocki, R.; Reculusa, S.; Ravaine, S.; Bartlett, P. N.; Kuhn, A.; Hempelmann, R.
Angewandte Chemie International Edition 2006, 45, 1317.
38. Abramova, V.; Sinitskii, A. Superlattices and Microstructures 2009, 45, 624.
39. Cao, Y.; Wang, Y.; Zhu, Y.; Chen, H.; Li, Z.; Ding, J.; Chi, Y. Superlattices and
Microstructures 2006, 40, 155.
40. Yu, H. M.; Yim, J.-H.; Choi, K. Y.; Lim, J. S. The Journal of Supercritical Fluids 2012, 67,
71.
41. Dimos, M. M.; Blanchard, G. J. The Journal of Physical Chemistry C 2010, 114, 6019.
42. Dimos, M. M.; Blanchard, G. J. Journal of Electroanalytical Chemistry 2011, 654, 13.
43. Dimos, M. M.; Blanchard, G. J. The Journal of Physical Chemistry C 2011, 115, 11247.
44. Gornowich, D. B.; Blanchard, G. J. The Journal of Physical Chemistry C 2013, in
preparation.

20

45. Mu, W.; Hwang, D.-K.; Chang, R. P. H.; Ketterson, J. B. Journal of Raman Spectroscopy
2011, 42, 941.
46. Holland, B. T.; Blanford, C. F.; Do, T.; Stein, A. Chemistry of Materials 1999, 11, 795.
47. Turner, M. E.; Trentler, T. J.; Colvin, V. L. Advanced Materials 2001, 13, 180.
48. Holland, B. T.; Blanford, C. F.; Stein, A. Science 1998, 281, 538.
49. Zhou, Q.; Dong, P.; Cheng, B. Journal of Crystal Growth 2006, 292, 320.
50. Jiang, P.; Cizeron, J.; Bertone, J. F.; Colvin, V. L. Journal of the American Chemical Society
1999, 121, 7957.
51. Bartlett, P. N.; Birkin, P. R.; Ghanem, M. A. Chemical Communications 2000, 1671.
52. Bartlett, P. N.; Baumberg, J. J.; Birkin, P. R.; Ghanem, M. A.; Netti, M. C. Chemistry of
Materials 2002, 14, 2199.
53. Sumida, T.; Wada, Y.; Kitamura, T.; Yanagida, S. Langmuir 2002, 18, 3886.
54. Huang, Y.-J.; Lai, C.-H.; Wu, P.-W.; Chen, L.-Y. Materials Letters 2009, 63, 2393.
55. Robinson, P. J.; Dunnill, P.; Lilly, M. D. Biochimica et Biophysica Acta (BBA) Enzymology 1971, 242, 659.
56. Van Aken, B.; Ledent, P.; Naveau, H.; Agathos, S. Biotechnology Letters 2000, 22, 641.
57. Betancor, L.; López-Gallego, F.; Alonso-Morales, N.; Dellamora, G.; Mateo, C.; FernandezLafuente, R.; Guisan, J. M. 2006; Vol. 22, p 57.
58. López-Gallego, F.; Betancor, L.; Mateo, C.; Hidalgo, A.; Alonso-Morales, N.; DellamoraOrtiz, G.; Guisán, J. M.; Fernández-Lafuente, R. Journal of Biotechnology 2005, 119, 70.
59. Surinenaite, B.; Bendikiene, V.; Juodka, B. Prog. Biotechnol. 1998, 15, 577.

21

60. Subramanian, A.; Kennel, S. J.; Oden, P. I.; Jacobson, K. B.; Woodward, J.; Doktycz, M. J.
Enzyme and Microbial Technology 1999, 24, 26.
61. Zhou, H.-X.; Dill, K. A. Biochemistry 2001, 40, 11289.
62. Kleppe, K. Biochemistry 1966, 5, 139.
63. Krishnaswamy, S.; Kittrell, J. R. Biotechnology and Bioengineering 1978, 20, 821.
64. Malikkides, C. O.; Weiland, R. H. Biotechnology and Bioengineering 1982, 24, 2419.
65. Yablonovitch, E. Physical Review Letters 1987, 58, 2059.

22

CHAPTER 2: Fabrication of Inverse Opal Structures
Introduction
Inverse opal structures have gained the interest of several research groups over the past
1-20

decade.

There are many different procedures for preparing the NS. There are a vast range of

materials that can be used as templates and as the matrix itself. A schematic for the construction
of a NS is illustrated in Fig. 2.1. The basic process involved in the fabrication of this type of
materials starts with layers of polymer or silica spheres deposited on a support. The interstitial
space between the spheres is then backfilled with a different material, and finally the sphere
template is removed either by dissolution or thermal decomposition. There are different methods
associated with the deposition of the spheres. The size of the spheres used can be changed
according to the desired application, offering a high degree of control over the pore size of the
system. The material that is deposited in between the spheres can also be chosen based on the
application that is intended for the resulting NS.
3-5

The Blanchard group has experience using metal inverse opal structures.

In that work,

silica nanosphere assemblies were used as the scaffold to build the NS and metals were
electrodeposited in the interstitial space. The silica spheres were then removed by dissolution
with hydrofluoric acid (HF). Platinum and palladium were both used to create NS structures and
the methods used in that work provided highly ordered NS structures. One of the main
limitations of that work was the construction of the NS on a planar, non-permeable support.
Consequently, any chemical reactions that take place on the surface of the nanoporous metal are
strictly limited by diffusion of reactive species in the NS, and flow is not possible.
Using this knowledge for inverse opal preparation, the work described in this research
required constructing inverse opal structures using dielectric materials. To take this one step

23

A

Cleaned bare support

B

nanosphere
deposition

matrix growth
D

C

nanosphere
removal

Figure 2.1: Schematic for the fabrication of an inverse opal on a planar substrate.

24

further and to eliminate the limitations associated with diffusion as the only means of transport,
the focus of this work also shifted to creating a NS material that would afford the ability to flow
a stream of reactant containing solution through the structure. The first section of this Chapter
focuses on forming NS structures on planar supports, and concludes with utilizing a combination
of techniques to form flow-through inverse opal structures. Polystyrene spheres (PS) were used
in the construction of scaffold structures for the NS and silica was deposited in the interstitial
space between the spheres by means of sol-gel chemistry. Once formed, the resulting matrix was
amenable to a wide range of surface modifications using several different chemical reactions.
Polystyrene Sphere Synthesis
The first step in the process of fabricating dielectric inverse opals is to synthesize
polymer spheres. For this work, polystyrene was chosen for two reasons. The first is that there
are multiple synthetic procedures available for the preparation of spheres in a wide range of sizes
21-32

that yield a monodisperse size distribution.

Nanosphere monodispersity is important for the

growth of well-organized inverse opal structures. In addition, polystyrene spheres are relatively
simple to remove using solvent dissolution or heating methods. It is important to note that all
polystyrene spheres used in this work were non-cross-linked polymers. Cross-linked spheres
were synthesized, but they are not as soluble in solvents such as toluene or ethyl acetate.
Removing the cross-linked spheres from the matrix would be too difficult for solvent dissolution
and, as a result only non-cross-linked polymers were used.
21-32

There are a number of options for synthesizing polymer spheres.

Two common
21-32

approaches that are utilized are dispersion and emulsion polymerization reactions.

These

syntheses are one-pot reactions and both yield polymer spheres in the size range of 0.1-15 μm in

25

diameter depending on the reaction conditions.

26

Both approaches were utilized in this work,

and the characterization of the resulting spheres will be described in detail in the following
sections.
Dispersion polymerization in an alcoholic medium has been studied extensively in recent
21,24-33

decades.

This one pot reaction requires a stabilizer, initiator, monomer, and solvent.

The most common stabilizers used are (hydroxypropy1)cellulose, poly(acry1ic acid), or poly(N30,31

vinylpyrrolidone) (PVP).

The stabilizer enables the formation of latex spheres, however
28,34

the understanding of the details involved in this process are qualitative.

For this work, PVP

(MW = 55,000) was used for all dispersion reactions. The initiator triggers the polymerization of
the monomers in solution. The most common initiators break down upon heating, forming free
radical species in solution. Benzoyl peroxide and azobisisobutyronitrile (AIBN) are two
24,26,27

common free radical sources for dispersion polymerization,

and AIBN was used for all

reactions reported in this work. Polystyrene and poly(methyl methacrylate) (PMMA) are two
common polymers used to synthesize spheres. Polystyrene was used for the work contained in
this dissertation. An important component of dispersion polymerization is the reaction medium,
and in most cases alcohols are used as the solvent. Methanol and ethanol were both utilized in
this research. The overall reaction scheme for the dispersion polymerization reaction used for
this work is illustrated in Scheme 2.1. The stabilizer identity and concentration, initiator
concentration, monomer concentration, and reaction medium all play important roles in
31

determining the size and polydispersity of the resulting polymer spheres.

26

N

H3C
N

N

C

C

N

N
CH3
heat
C
C
R-OH
CH3

C
CH3

CH3

CH3
H2C

N

H 3C

CH

+

CH3
C
H2C

C

N

CH

R-OH

CH3

H3C

C
H 2C

C
H3C C

N2

+

2 H3C C

CH3
C

H 3C

N
H2C

CH

C
H2C

CH

+

C
CH

N
CH2 CH n

R-OH

Scheme 2.1: Reaction schematic for a free radical dispersion polymerization reaction for
polystyrene sphere synthesis. PVP was present in the reaction mixture as a stabilizer.

27

For a typical reaction, the initiator (AIBN) was added to a round-bottom flask containing
ethanol. The reaction flask was then heated to reflux at a temperature of 80°C for approximately
30 minutes. Then the stabilizer (PVP) and monomer (polystyrene) were added to the reaction
flask. The reaction was allowed to proceed with stirring at 80°C under reflux for 24 hours, and
then allowed to cool to room temperature. The spheres were washed three times with ethanol via
centrifugation at 3900 rpm. The ethanol was allowed to evaporate, and the spheres were
suspended in aqueous solutions at various w/w concentrations for deposition onto planar support
surfaces.
The first key component to synthesizing monodisperse polymer spheres is to use purified
chemicals. Styrene monomer is available commercially, but it contains a stabilizer to prevent
polymerization while in storage. Consequently, this stabilizer must be removed prior to use.
There are two approaches to removing the stabilizer; washing with a strong base or distillation of
the monomer. It was found that distilling the monomer was the most effective way, because this
method purified the reactant just prior to use and resulted in comparatively monodisperse
spheres. If care was not taken to remove the stabilizer from the styrene, a bimodal distribution of
spheres was the typical result, as shown in Fig. 2.2. Spheres with a wide size distribution are not
ideal to use as a scaffold for NS construction because they do not assemble in a highly ordered
manner.
As noted previously, all of the constituents in the dispersion polymerization reaction play
a role in determining the size of the polymer spheres. The first component that was varied was
the alcohol solvent used for the reaction medium. Methanol, ethanol, and isopropanol were all
used as reaction media for dispersion polymerization reactions. A solution containing 45 mL of
alcohol, 5 mL of water, 5 mL of styrene monomer, 50 mg of AIBN, and 0.6 g of PVP was used

28

10 μm
Figure 2.2: SEM image of polystyrene spheres with a bimodal size distribution.

29

for the each reaction following the procedure previously described. Nitrogen was bubbled into
the mixture throughout the reaction. The PS sphere sizes were ca. 900 nm, 1.1 μm, and 2.2 μm
for methanol, ethanol, and isopropanol, respectively. These data suggest a correlation between
alcohol aliphatic chain length and nanosphere diameter. The methanol (Fig. 2.3) and ethanol
(Fig. 2.4) reactions yield monodisperse spheres, while the isopropanol (Fig. 2.5) produces a
bimodal distribution with some irregularly shaped polymer structures.
The amount of stabilizer was also varied for a series of reactions with ethanol as the
solvent. Reactions were performed using 0.2%, 0.8% and 1.2% (w/v) PVP and the diameters of
the spheres were 2.3, 1.5 , and 1.1 μm, respectively. The amount of ethanol, initiator (AIBN),
styrene, and water were all held constant at 45 mL, 50 mg, 5 mL, and 5 mL, respectively. The
results reveal an inverse relationship between polymer sphere size and PVP concentration (Fig.
2.6). Higher steric stabilizer concentrations yield smaller spheres, consistent with literature
31

reports.

While the other constituents of the dispersion reaction solution could be varied, the

spheres synthesized with the previously mentioned methods were suitable for the purposes of
this work.
Another type of polymerization reaction was investigated briefly in an effort to obtain
smaller spheres, on the order of a few hundred nanometers in diameter. Emulsion
polymerization was utilized for this work, and has been previously characterized in the
22,23,35

literature.

In this reaction format, potassium persulfate is used as the initiator for the

polymerization. A solution of 50 mL of methanol, 20 mL of water, 4 mL of styrene, and 100 mg
of potassium persulfate was used for this reaction. The mixture was heated under reflux at 70°C
in an oil bath for 24 hours under nitrogen. The SEM image of PS spheres with an average
diameter of 180 nm synthesized using this technique is presented in Fig. 2.7. Using the two

30

2 μm
Figure 2.3: SEM image of polystyrene spheres with an average diameter of 900 nm synthesized
using methanol as the solvent.

31

2 μm
Figure 2.4: SEM image of polystyrene spheres with an average diameter of 1.1 μm synthesized
using ethanol as the solvent.

32

5 μm
Figure 2.5: SEM image of polystyrene spheroidal structures with an average diameter of 2.2 μm
synthesized using ethanol as the solvent.

33

2.5

0.2% PVP

2.0

2.0
size (µm)

Sphere Diameter (µm)

2.4

1.5

1.6

2 μm

1.21.0
0.2
0.2 0.40.4 0.60.6 0.80.8 1.01.0 1.2 1.2
% PVP (w/v)
% PVP

1.2% PVP

0.8% PVP

5 μm

2 μm

Figure 2.6: Plot of polystyrene sphere size vs. steric stabilizer concentration (top left) for free
radical dispersion polymerization reactions with ethanol as the solvent. SEM images (top right
and bottom) of polystyrene spheres at each PVP concentration.

34

500 nm
Figure 2.7: SEM image of polystyrene spheres with an average diameter of 180 nm synthesized
using an emulsifier polymerization reaction method.

35

synthetic techniques described previously, a wide size range of PS spheres (180 nm – 2 μm
diameter) were synthesized and used throughout the NS fabrication experiments.
The polymer spheres were characterized using two different techniques, scanning
electron microscopy (SEM) and dynamic light scattering (DLS). For a highly ordered NS to be
formed, the spheres for the scaffold must have a monodisperse size distribution. Fig. 2.8 is a
SEM image of polystyrene spheres with a diameter of ca. 1 μm that were deposited on a glass
microscope slide. All samples were coated with gold using an Emscope SC500 sputter coater
with a deposition current of 20 mA for 3 minutes in preparation to be analyzed using SEM. At
20 mA, the gold deposition rate results in a thickness of 7 nm/minute.36 All micrographs were
obtained using a JEOL 6400V microscope with a LaB 6 emitter at an accelerating voltage of 12
kV.
The size of the PS spheres was measured using a Malvern Zetasizer Nano ZS. Fig. 2.9
contains dynamic light scattering (DLS) data for polystyrene spheres, with an average diameter
of 1.15 μm and the distribution of particle sizes. This instrument is equipped with backscatter
detection technology which measures scattered light at an angle of 173° relative to the axis of
light propagation. This detection method provides several advantages over traditional DLS
techniques, where scattering is measured from light that passes through the sample. With
backscatter detection geometry, the (scattered) light does not have to pass through the sample,
allowing for strongly scattering samples to be analyzed. Backscatter detection geometry reduces
the effect known as multiple scattering, where light scattered by one particle is subsequently
scattered by other particles. Particulate contamination issues are also reduced by collecting the
data in backscatter configuration because particulate contaminants are often relatively large
compared to the size of the nanospheres. Larger particles scatter in the forward direction mostly,

36

20 μm
Figure 2.8: SEM image of polystyrene spheres with a diameter of ca. 1 μm deposited on a glass
microscope slide that were synthesized using a free radical dispersion polymerization method.

37

18
16
14
Intensity (%)

12
10
8
6
4
2
0
-2

100

1000

10000

Size (d, nm)

Figure 2.9: Dynamic light scattering data showing the particle size distribution of the
polystyrene spheres with an average diameter of 1.15 μm.

38

so collecting in backscatter mode greatly reduces signals from contaminants that may be present
37

in the sample.

For all of this work, the polystyrene spheres were suspended in a 1% wt./wt.

aqueous solution. Prior to each DLS analysis, the aqueous suspension of spheres was sonicated
for 30 minutes to break apart any aggregates that may have formed during storage.
Polystyrene Sphere Deposition Techniques
Once the PS spheres have been synthesized and characterized, the next step in the process
of building inverse opal structures is to deposit the spheres in an organized manner onto a
support. Three different techniques were used for this purpose. The first and most elementary
method used was solvent evaporation. A microscope slide was cut so that it would fit on a SEM
stub and cleaned using piranha solution (3:1 sulfuric acid:hydrogen peroxide). An aqueous or
alcoholic suspension of PS spheres was then added to the slide drop wise and the solvent was
allowed to evaporate at room temperature. Fig 2.10 is a representative SEM image of PS spheres
deposited using this technique. This technique was able to produce opal structures, but there is
limited control over the spreading and evaporation of the solution on the support. Consequently,
it was difficult to reproduce the same thickness from sample to sample. A second deposition
technique was spin-coating. Glass microscope slides were placed on a spin-coater (Headway
Research Inc.), and sphere suspensions were placed onto the surface while spinning at velocities
ranging from 1000-3000 rpm. Fig 2.11 is a SEM image of PS spheres that were deposited onto a
slide using a spin-coater. High quality packing of the spheres was not observed for the range of
parameters used for this work, so this method was not used for inverse opal preparation.
3-5

The third method used for sphere deposition was vertical dipping.

For this technique,

a SDI Nanodip Coater (nanodipper) was used to slowly withdraw a planar substrate from an

39

20 μm
Figure 2.10: SEM image of polystyrene spheres with an average diameter of 1 μm deposited on
a glass microscope slide using solvent evaporation.

40

20 μm
Figure 2.11: SEM image of polystyrene spheres with an average diameter of 1 μm deposited on
a glass microscope slide using a spin-coater.

41

aqueous sphere suspension. Capillary forces at the meniscus cause a multi-layer assembly of
spheres to be deposited on the substrate as it is withdrawn from the solution. Withdrawal speeds
of 1 μm/sec up to 25 μm/sec were used for this work. A rate of 2 μm/sec yielded the best
packing for size of PS spheres used in this work, and was used for most of the routine work that
utilized the nanodipper. Fig. 2.12 is a SEM micrograph of PS spheres that were deposited on a
clean glass microscope slide at a rate of 2 μm/sec. Examining the three methods shows that spincoating yields a highly disordered arrangement of PS spheres that is not useful to use for building
NS materials. The solvent evaporation and vertical dipping techniques both yield suitable
scaffolds for NS construction. It was also found that sintering the spheres at 60-70°C for ca. 1
hour fuses the scaffold together to form a more robust structure. Vertical dipping results in an
assembly with the highest proportion of hexagonally closed packed regions and is the method of
choice for planar substrates. Vertical dipping also affords control over the number of layers of
spheres deposited on the support. Multiple dipping cycles can be performed, adding layers of
spheres with each cycle. Highly ordered structures with multiple layers formed by vertical
38

dipping is consistent with literature reports.

Sol-gel Methods for Silica Deposition
Sol-gel chemistry was used to fill the interstitial space between the PS spheres. Silica
was chosen as the dielectric material because of its well understood sol-gel reactions and because
it is amenable to a wide range of surface modification reactions. Scheme 2.2 is an overall
schematic for the sol-gel reaction using tetraethyl orthosilicate (TEOS) as the precursor.

39

TEOS undergoes hydrolysis in the presence of water, and subsequent water and alcohol
condensation reactions form siloxane bonds. As the precursor gels, an interconnected silica
network is formed around the PS sphere scaffold. This sol-gel reaction is catalyzed by both

42

20 μm
Figure 2.12: SEM image of polystyrene spheres with an average diameter of 1 μm deposited on
a glass microscope slide using the nanodipper at a rate of 2 μm/sec.

43

O

O

Si

O

O

O
O

Si

+ H2O

O

OH + O

Si

O

OH

O

O
O

O

Si
O

hydrolysis
reesterfication

OH + O

Si
O

O

H2O condensation
hydrolysis

H2O condensation
hydrolysis

O

O

Si

O

O
Si

O

+

OH

O

O
Si

O O

O
O

Si

O
O O

OH

O

+

H2O

O
Si

O

+

Scheme 2.2: Overall reaction schematic for the sol-gel process used to fill in the interstitial
space with a TEOS precursor solution. Hydrolysis (top), water condensation (middle), and
alcohol condensation (bottom) reactions.

44

OH

acids and bases through different mechanisms. An acid catalyzed system was utilized for all of
the work described in this dissertation. The precursor solution was prepared by combining 72
mL of water and 8 mL of ethanol, adjusting to pH 2 with concentrated hydrochloric acid, adding
4 mL of TEOS, and allowing the solution to stir for one hour. Two methods utilized for this
process were by adding the sol-gel precursor solution drop-wise onto the PS scaffold, and
vertical dipping using the nanodipper. Fig 2.13 contains SEM images for the evaporation
method and Fig. 2.14 is a representative image of a NS fabricated using the nanodipper at a rate
of 2 μm/sec. The PS sphere assemblies had been sintered for ca. 1 hour at 65°C before being
submerged into the sol-gel solution. Both techniques worked well for utilizing silica sol-gel
chemistry to fill in the interstitial regions between PS spheres.
Polystyrene Sphere Removal
There are two options available to remove the polymer spheres from within the matrix
after it has solidified; heating to high temperatures to remove the spheres by thermal degradation
or dissolving the spheres using a suitable solvent. Both approaches were studied, and each
technique has advantages and disadvantages. For the heating technique, a temperature program
was created using a high temperature furnace. The PS sphere/silica composite structure was
placed in the furnace, the temperature was ramped from room temperature to 500°C at a rate of
5°C/min, the temperature then held constant at 500°C for 2 hours, and allowed to cool to room
temperature. For the solvent dissolution technique, the composite structure was soaked in
toluene for a period of 24-48 hours at room temperature. The heating method is much less time
consuming and it guarantees that all of the polystyrene is removed from the matrix. However,
this process typically resulted in significant cracking (see Fig. 2.15) of the NS structure and
shrinkage of the silica network. Heating also results in a completely dehydrated silica network,

45

20 μm

Figure 2.13: SEM images of silica inverse opal structures on a glass microscope slide. The
spheres and sol-gel chemistry were performed in subsequent steps by adding solvent evaporation
and adding the precursor solution on drop wise.

46

500 nm
Figure 2.14: SEM image of a silica inverse opal structure on a glass microscope slide. The
spheres and sol-gel chemistry were performed in subsequent steps using the nanodipper at 2
μm/sec. The spheres were deposited using 3 consecutive withdrawals from a suspension,
sintered at 65°C for 1 hour, and then withdrawn from the sol-gel precursor solution 1 time.
Toluene was used to remove the spheres.

47

1 mm
Figure 2.15: SEM image of an inverse opal that utilized heat to remove the polystyrene scaffold.
This image was taken at low magnification to show the major cracking caused by the burning
procedure.

48

rendering it much less reactive than the structure resulting from sphere dissolution with toluene.
The toluene method was less destructive, however only spheres that are accessible to the solvent
are removed from the matrix. Any polymer residue buried within the matrix will remain in the
NS.
Flow-through Nanoporous Solid Structure Fabrication
Using the knowledge and experience gained in the process of constructing NS on planar
substrates, we designed a NS structure appropriate for flow-through applications. Using a
combination of the techniques previously described in this chapter, two different supports were
used in an attempt to create a flow-through system. The first support matrix used was
transmission electron microscopy (TEM) grids. The other support used for this work was porous
alumina supports (PAS). The TEM grids and PAS are both commercially available materials
that had three specific qualities important to this work. These are that they are relatively inert to
the reactions we perform in their presence, that they are tolerant of a wide range of solvents, and
they are able to withstand high temperatures. Both supports will be described in this section.
The first attempts at a flow-through system were made using 1500-2000 mesh TEM
grids. A grid was mounted on a clean gold substrate, and the same dipping techniques that were
described previously were used to deposit the spheres and fill in the void space with silica.
Moderate success was made with the TEM grids using this method. The grid spaces were filled
in with the NS matrix (Fig. 2.16), however a large portion of them remained partially filled or
completely empty. This would provide an unrestricted path for a reactant containing solution to
bypass the nanoporous solid structure, which was not desirable. In addition, the TEM grids are
extremely delicate and require great care to be taken when handling them. As a result, it was
concluded that TEM grids were not going to work for flow-through applications.

49

5 μm
Figure 2.16: SEM image of an inverse opal that was synthesized on a gold TEM grid (1500
mesh).

50

At this point, we turned to PAS for the support matrix for the NS materials. The PAS are
available commercially in a wide range of diameters and pore sizes. For all of the work
presented in this dissertation, PAS with a diameter of 22 mm, thickness of 60 μm, and pore size
of 200 nm (Whatman) were used. Due to the thinness and relatively fragile nature of the PAS,
using the nanodipper was not an option for deposition of the spheres. As a result, the PAS were
placed in an Advantec glass microanalysis filter holder with a porous glass frit as a structural
support. Two different methods were examined for the NS fabrication on the PAS. The first
involved performing steps analogous to the planar support. A 0.5 mL aliquot of aqueous 1%
(wt./wt.) PS sphere suspension was deposited on the PAS and allowed to dry for 24 hours.
Subsequently a 0.5 mL aliquot of sol-gel precursor solution was added drop-wise onto the PS
scaffold. The sol-gel solution consisted of TEOS, ethanol, water and concentrated hydrochloric
acid in a 4:6:3:1 mass ratio, respectively. A vacuum was applied using a water aspirator, and the
PS/silica matrix was allowed to dry at room temperature for 24 hours. The second method
involved using a co-assembly technique where PS spheres were suspended in the sol-gel
precursor solution, and both the PS spheres and silica network were deposited simultaneously.
For this method, 0.5 mL of an aqueous 1% (w/w) PS sphere solution was combined with 0.5 mL
of a sol-gel precursor solution containing TEOS, ethanol, water and concentrated hydrochloric
acid in a 4:6:3:1 mass ratio, respectively. The PS sphere/sol-gel mixture was then deposited on
the PAS. A vacuum was applied using a water aspirator, and the sample was allowed to dry for
24 hours. The PS/silica matrix was then removed from the filter holder and soaked in toluene for
24 hours to remove the PS scaffold for structures prepared by both methods. Fig. 2.17 contains
SEM images of structures prepared by both techniques to be used for flow-through applications.
It was found that the method utilizing sequential steps often resulted in “capping” of the silica

51

20 μm

20 μm
Figure 2.17: SEM images of an inverse opal structures that were synthesized on PAS. The top
image represents an inverse opal formed using subsequent steps for the sphere and sol-gel
depositions, and the bottom micrograph represents an inverse opal formed using the co-assembly
technique.

52

network. The NS was present underneath the layer of silica, but the pores were blocked by a
layer of silica. The co-assembly technique resulted in a NS with the pores fully exposed at the
surface. As a result, the co-assembly technique was used for all of the NS/PAS matrices in the
2

flow-through experiments throughout this work. A co-assembly technique has been reported to
8

result in crack-free inverse opal structures on planar substrates in the literature, however our
results have shown some areas of cracking remain.
Conclusions
This chapter was intended to provide an overview of the procedures used in the
fabrication of silica inverse opal structures. Silica NS structures were constructed on a variety of
supports, including glass microscope slides, TEM grids, and PAS. The knowledge gained
through the process of fabricating NS on planar substrates was translated and modified to create
NS structures on a porous support system. The goal of creating flow-through nanoporous solids
was achieved using a PAS and a modified co-assembly technique. With this foundation, the
decoration of these structures with catalytic species will form the foundation for the studies
presented in this dissertation.

53

Literature Cited

54

Literature Cited
1. Nishijima, Y.; Ueno, K.; Juodkazis, S.; Mizeikis, V.; Misawa, H.; Tanimura, T.; Maeda, K.
Optics Express 2007, 15, 12979.
2. Gornowich, D. B.; Blanchard, G. J. The Journal of Physical Chemistry C 2012, 116, 12165.
3. Dimos, M. M.; Blanchard, G. J. The Journal of Physical Chemistry C 2010, 114, 6019.
4. Dimos, M. M.; Blanchard, G. J. Journal of Electroanalytical Chemistry 2011, 654, 13.
5. Dimos, M. M.; Blanchard, G. J. The Journal of Physical Chemistry C 2011, 115, 11247.
6. Szamocki, R.; Reculusa, S.; Ravaine, S.; Bartlett, P. N.; Kuhn, A.; Hempelmann, R.
Angewandte Chemie International Edition 2006, 45, 1317.
7. Szamocki, R.; Velichko, A.; Mücklich, F.; Reculusa, S.; Ravaine, S.; Neugebauer, S.;
Schuhmann, W.; Hempelmann, R.; Kuhn, A. Electrochemistry Communications 2007, 9, 2121.
8. Hatton, B.; Mishchenko, L.; Davis, S.; Sandhage, K. H.; Aizenberg, J. Proceedings of the
National Academy of Sciences 2010, 107, 10354.
9. Kim, J. K.; Moon, J. H.; Lee, T.-W.; Park, J. H. Chemical Communications 2012, 48, 11939.
10. Abramova, V.; Sinitskii, A. Superlattices and Microstructures 2009, 45, 624.
11. Cao, Y.; Wang, Y.; Zhu, Y.; Chen, H.; Li, Z.; Ding, J.; Chi, Y. Superlattices and
Microstructures 2006, 40, 155.
12. Holland, B. T.; Blanford, C. F.; Do, T.; Stein, A. Chemistry of Materials 1999, 11, 795.
13. Holland, B. T.; Blanford, C. F.; Stein, A. Science 1998, 281, 538.
14. Holland, B. T.; Blanford, C. F.; Stein, A. Science 1998, 281, 538.
15. Turner, M. E.; Trentler, T. J.; Colvin, V. L. Advanced Materials 2001, 13, 180.

55

16. Yang, P.; Deng, T.; Zhao, D.; Feng, P.; Pine, D.; Chmelka, B. F.; Whitesides, G. M.; Stucky,
G. D. Science 1998, 282, 2244.
17. Yu, H. M.; Yim, J.-H.; Choi, K. Y.; Lim, J. S. The Journal of Supercritical Fluids 2012, 67,
71.
18. Ma, J.; Parajuli, B. R.; Ghossoub, M. G.; Mihi, A.; Sadhu, J.; Braun, P. V.; Sinha, S. Nano
Letters 2013, 13, 618.
19. Mathew, S. S.; Ma, S.; Kretzschmar, I. Journal of Materials Research 2013, 28, 369.
20. Wu, M.; Zheng, A.; Deng, F.; Su, B.-L. Applied Catalysis B: Environmental 2013, 138–139,
219.
21. Goodwin, J. W.; Hearn, J.; Ho, C. C.; Ottewill, R. H. Colloid & Polymer Science 1974,
252, 464.
22. Ottewill, R.; Shaw, J. Colloid & Polymer Science 1967, 215, 161.
23. Zhang, J.; Chen, Z.; Wang, Z.; Zhang, W.; Ming, N. Materials Letters 2003, 57, 4466.
24. Bamnolker, H.; Margel, S. Journal of Polymer Science Part A: Polymer Chemistry 1996, 34,
1857.
25. Jiang, S.; David Sudol, E.; Dimonie, V. L.; El-Aasser, M. S. Journal of Applied Polymer
Science 2008, 107, 2453.
26. Kawaguchi, S.; Ito, K.; Okubo, M., Ed.; Springer Berlin / Heidelberg: 2005; Vol. 175, p 299.
27. Lu, Y. Y.; El-Aasser, M. S.; Vanderhoff, J. W. Journal of Polymer Science Part B: Polymer
Physics 1988, 26, 1187.
28. Paine, A. J. Journal of Colloid and Interface Science 1990, 138, 157.
29. Paine, A. J. Macromolecules 1990, 23, 3109.

56

30. Paine, A. J. Journal of Polymer Science Part A: Polymer Chemistry 1990, 28, 2485.
31. Paine, A. J.; Luymes, W.; McNulty, J. Macromolecules 1990, 23, 3104.
32. Tseng, C. M.; Lu, Y. Y.; El-Aasser, M. S.; Vanderhoff, J. W. Journal of Polymer Science
Part A: Polymer Chemistry 1986, 24, 2995.
33. Im, S. H.; Khalil, G. E.; Callis, J.; Ahn, B. H.; Gouterman, M.; Xia, Y. Talanta 2005, 67,
492.
34. Barrett, K. E. J. Dispersion polymerization in organic media, edited by K. E. J. Barrett. With
a foreword by C. H. Bamford; Wiley: London, New York, 1974.
35. Liu, Z.; Jin, Z.; Li, W.; Qiu, J. Materials Research Bulletin 2006, 41, 119.
36. Flegler, S. L.; Heckman, J. W.; Klomparens, K. L. Scanning and Transmission Electron
Microscopy: An Introduction; 1st ed.; Oxford University Press: New York, New York, 1993.
37. In Zetasizer Nano User Manual; Malvern Instruments Ltd.: Worcestershire, 2007; Vol.
MAN0317, p 14.1.
38. Gu, Z.-Z.; Fujishima, A.; Sato, O. Chemistry of Materials 2002, 14, 760.
39. Brinker, J. C., Scherer, G. W. Sol-Gel Science: The Physics and Chemistry of Sol-Gel
Processing; Academic Press, Inc.: San Diego, CA, 1990.

57

CHAPTER 3: Enhancement of GOx Activity by Confinement in an Inverse Opal Structure
Introduction
The immobilization of enzymes on support structures has proven to be an attractive
research area for both fundamental and pragmatic reasons. Among the motivations for using
immobilized enzymes are the minimization of post-reaction cleanup and the ability to recover
and reuse the enzyme.

From a more fundamental perspective, it has been shown that

immobilization can either enhance or diminish the reactivity of an enzyme, depending on the
1,2

specific system and manner of immobilization,
3

can be improved by immobilization.

and that the long-term stability of the enzyme

The two methods used most widely are binding the

enzyme to a pre-fabricated matrix and entrapment of the enzyme within a polymer. A variety of
4

5-7

support materials have been demonstrated, including synthetic resins, mesoporous silica,
8,9

silica nanospheres,

10,11

and zeolites.

Each of these supports offers advantages and

disadvantages. The support-related issues of most importance to this work are (a) the ability to
bind enzymes to the support surface covalently under mild reaction conditions, (b) high porosity
to enable efficient reactant stream flow, and (c) relatively high surface area. Inverse opal
structures satisfy all of these requirements and possess several potential structurally-based
advantages.

This material format is seeing increased utilization for the immobilization of

12-15

enzymes,

with promising results having been reported recently that illustrate the range of
16,17

utility of inverse opals for the immobilization of enzymes.

58

We are interested in exploring inverse opal structures as catalyst supports. We have
examined the efficiency of the catalytic conversion of glucose to gluconic acid and hydrogen
peroxide with glucose oxidase (GOx) immobilized on a silica inverse opal structure. The ability
to perform this reaction in a flowing mode is important because hydrogen peroxide denatures
18,19

GOx, thereby terminating the reaction.

In a flow-through system, the product(s) are

separated from the enzyme on a continuous basis, minimizing the adverse effect(s) of prolonged
exposure of the enzyme to H2 O 2 . We have chosen to use the glucose/GOx system because it is a
well characterized reaction and the selectivity of GOx for glucose precludes the operation of
more than one chemical reaction. Inverse opals are attractive for such studies because of the
control one can exert over the internal void space and pore dimensions, and the ability to create
20-25

these structures using a variety of oxide materials.

Inverse opals, sometimes referred to as

nanoporous solids (NS), used in this work were constructed in our labs using assemblies of
polystyrene spheres as a scaffold. Silica was deposited in the void spaces by a sol-gel growth
method, and the polymer spheres were subsequently removed by dissolution. The resulting
matrix is an interconnected structure that has a high surface area-to-volume (S/V) ratio (ca. 1.22
5

2

3

x 10 cm /cm for an inverse opal constructed using 1 μm diameter spheres) and is amenable to
a wide range of surface modifications that can be used for enzyme attachment. For inverse opals
the pore diameter between void regions is taken to be 20% of the diameter of the spheres used in
the formation of the structure, affording synthetic control over flow properties.

26

Nanoporous solids offer the potential to enhance the catalytic activity of GOx due the
high S/V and physical dimensions of the structure of the solid matrix. The high S/V ratio
provides a large support area for the deposition of GOx in a structural format that produces a

59

high local density of catalyst. The nanoporous structure is conducive to frequent substrateenzyme interactions, thus enhancing the turnover rate of the enzyme relative to solution phase
enzymes or the enzyme bound to a planar surface. Our data show that the turn-over rate of GOx
bound to inverse opal structures is enhanced relative to GOx bound to a planar silica surface or
free in solution, and we understand these findings in the context of geometric confinement,
surface binding of the enzyme and the kinetics of the GOx/glucose reaction.
Experimental
Nanoporous solid fabrication: Polystyrene spheres (PS) with an average diameter of
27

1.15 μm were synthesized using an adaptation of a free radical polymerization method.

Styrene monomer (99%, Aldrich) was distilled at 80°C to remove the stabilizer. The purified
styrene was then added to a reaction flask with ethanol, 2,2’-azobisisobutyronitrile (AIBN, 98%,
Aldrich), and polyvinylpyrrolidone (PVP, M w ~ 55,000, Aldrich). The reaction solution was
refluxed at 80oC for 24 hours.

The resulting polystyrene spheres were then washed by

centrifuging 3 times with ethanol rinses. The solvent was then allowed to evaporate, and a 2%
(w/w) suspension of the polystyrene spheres was prepared in water (Milli-Q). The sphere
suspension was sonicated for 30 minutes prior to each use to disperse any polystyrene aggregates
that may have formed.
The method used for the sol-gel inverse opal formation procedure (vide infra) was
28

adapted from the literature.

The sol consisted of tetraethylorthosilicate (TEOS, 99%, Aldrich),

ethanol (anhydrous), water (Milli-Q) and concentrated hydrochloric acid (12 M, CCI) in a
4:6:3:1 mass ratio, respectively (total volume 10 mL). The sol was allowed to stir for 1 hour at
room temperature prior to use.

60

To form the inverse opal structure, 0.5 mL of the sol was combined with 0.5 mL of the
2% (w/w) nanosphere solution. The resulting mixture was deposited on a porous alumina
support (PAS, d = 22 mm, thickness = 60 μm, Whatman) with a 200 nm nominal pore diameter.
The PAS was placed in an Advantec glass microanalysis filter holder with a porous glass frit as a
structural support. Using a water aspirator, a vacuum was applied under the PS/sol/PAS and the
resulting nanosphere assembly was left to dry at room temperature for a period of 24 hours. The
nanosphere/sol structure mounted on the PAS was then removed from the filtration apparatus
and placed in toluene (≥99.5%, Mallinckrodt) to remove the polystyrene nanospheres from the
sol matrix. It was possible to flow toluene through the matrix after 24 hours of soak time.
Nanoporous solid surface modification: Glucose oxidase derived from Aspergillus niger
(lyophilized powder, Aldrich) was immobilized on the surface of the nanoporous solid using a
29

method adapted from the literature.

The NS/PAS assembly was first immersed in a solution

containing 10% (v/v) 3-aminopropyltriethoxysilane (APTES, 99%, Aldrich) in toluene for a
period of 30 minutes at 35°C. A cross-linking moiety was then added to the surface by placing
30

the matrix in a 1% (v/v) aqueous glutaraldehyde solution.

The NS/PAS assembly was then

incubated in a 1 mg/mL GOx solution containing 50 mM sodium acetate buffer (pH 4) for 24
hours at 4°C. The resulting GOx-modified NS/PAS was stored in sodium acetate buffer at 4°C
when not in use.
Flow-through experiment: An in-house-made Teflon® flow cell was used to hold the
NS/PAS in place for the flow-through experiments. β-D-Glucose in sodium acetate buffer
solutions in a series of concentrations was flowed through the assembly at flow-rates ranging
from 25-50 μL/min. The eluent was collected in 1 mL fractions and analyzed for H 2 O 2 .

61

31,32

GOx assay: Fenton’s assay

was used to determine the concentration of hydrogen

peroxide formed as a result of GOx catalytic oxidation of β-D-glucose to gluconolactone, which
spontaneously hydrolyzes to form D-gluconic acid and H2 O 2 . Fenton’s reagent (FR) is 1 mM
FeSO 4 in sodium acetate buffer (pH 4). We estimate from the data reported by Woodward et al.
32

ε 340 = 3,000 L/mol-cm for Fenton’s reagent.

Hydrogen peroxide oxidizes Fe

2+

to Fe

3+

resulting in an increase in a near UV absorption band which is monitored at 340 nm. A 3 mL
aliquot of FR was added to a quartz cuvette along with 20 μL of the sample to perform the assay.
Absorbance measurements were made using a Cary 300 UV-visible double beam absorption
spectrometer.
Free GOx in solution analysis: A 1.5 mL aliquot of a 1 mg/mL GOx in sodium acetate
buffer solution was combined with 1.5 mL of FR in a cuvette. A 20 μL aliquot of an aqueous
solution containing glucose was injected into the cuvette and the solution was stirred
continuously throughout the measurement.

Absorbance spectra were obtained at 2 minute

intervals.
GOx immobilized on planar support: A 5 mm x 5 mm silicon wafer with a ca. 15 Å SiO x
layer was cleaned by sonication in water, acetone, and ethanol for 15 minutes each and dried in
an oven.

29

The clean support wafer was then treated in the same manner as previously described

for the immobilization of GOx on the NS/PAS. The GOx containing wafer was then placed in a
cuvette along with 3 mL of FR. A 20 μL aliquot of an aqueous solution containing controlled
amounts of glucose was injected into the cuvette and absorbance spectra were obtained in 5
minute intervals.

62

Scanning Electron Microscopy (SEM): All samples were coated with gold using an
Emscope SC500 sputter coater with a deposition current of 20 mA for 3 minutes. All images
were obtained using a JEOL 6400V microscope with a LaB 6 emitter at an accelerating voltage of
12 kV.
Dynamic Light Scattering (DLS): The size of the polystyrene spheres was measured
using a Malvern Zetasizer Nano ZS. The spheres were suspended in an aqueous solution for
analysis.
Thermogravimetric Analysis (TGA): The TGA analysis was performed using a Perkin
Elmer TGA-7 thermogravimetric analyzer and TAC 7 thermal analysis controller.

The

temperature program involved three steps; heat from room temperature to 120°C at 10°C/min,
hold at 120°C for 30 minutes, and heat from 120°C to 800°C at 10°C/min.
Results and Discussion
As noted above, the use of inverse opal (nanoporous solid, NS) structures as catalyst
supports is attractive for several reasons. We have found experimentally that this structural
motif gives rise to enhanced catalytic performance in the electrocatalytic oxidation of methanol
33-35

with Pt as the matrix material.

In that work we found evidence for a geometrically-based

performance enhancement and indications that the formation of the Pt inverse opal structure gave
rise to a different distribution of crystalline faces than the deposition of a planar Pt surface. We
are also interested in catalytic processes in systems where the support material does not function
as the catalyst. For such systems, where the physical support is chemically inert, it is possible to
decorate the support surface with a catalyst such as GOx. We focus in this work on our results
for GOx bound to a silica inverse opal structure. We compare the reaction turnover rate for the
GOx/glucose reaction for the inverse opal-bound enzyme to that for solution phase reaction and

63

for the enzyme bound to a planar silica substrate. Through these comparisons we can evaluate
the relative contributions of different system properties. Our data demonstrate enhancements in
the catalytic behavior of GOx for both steric and geometric reasons.
The construction of an inverse opal structure requires the use of spheres with a relatively
narrow size distribution for constructing the scaffold. We have examined the size distribution of
our spheres using dynamic light scattering (DLS) and SEM measurements. Results from these
measurements are consistent, showing an average diameter of 1.15 μm with a relative
polydispersity of 2.25% (Fig. 3.1). These spheres were used for all inverse opal preparation
described in this work.
The construction of an inverse opal structure on a porous alumina substrate is an essential
step in our ability to evaluate the enzymatic activity of GOx. We show in Fig. 3.2 a schematic of
the NS/PAS assembly, with SEM micrographs of the inverse opal, the PAS and the interfacial
region for a PAS where the two structures have separated. We show in the bottom panel a sideview of a NS/PAS assembly where physical separation between the two materials is beginning to
occur. The NS assembly is ca. 27 layers thick. It is instructive given this information to
estimate the surface area of the inverse opal structure. The surface area can be approximated
geometrically. For a hexagonal close-packed arrangement, the surface area of each void is given
by Avoid = Ananosphere – 10Apore for the first layer and Avoid = Ananosphere – 12Apore for each
subsequent layer. The reason for the difference between the first and subsequent layers has to do
with the points of contact a given nanosphere makes. For the first layer there are three contact
points with the second layer, six neighboring contact points in the layer and one contact point
with the support for a total of 10. For the second and subsequent layers, there are three contact
points with each adjacent layer of nanospheres and six contact points with in-plane neighbors for

64

18

a

16
14

Intensity (%)

12
10
8
6
4
2
0
-2

100

1000

10000

Size (d, nm)

b

10 μm

Figure 3.1: (a) Dynamic light scattering data showing the particle size distribution of the
polystyrene spheres used to template the inverse opal structures used as enzyme supports. (b)
SEM micrograph of the polystyrene spheres.

65

Nanoporous Solid Adlayer

Adlayer/Support
Porous Alumina Support

20 μm

Figure 3.2: Schematic of the NS/PAS assembly. Top, SEM micrographs of inverse opal, PAS,
and interfacial region, and bottom, SEM micrograph of an on-edge view of the NS/PAS that was
fractured.

66

2

a total of 12. For nanospheres of radius r, we approximate the area to be π(0.2r) for each
2

circular pore. For spheres with a diameter of 1.15 μm and a PAS deposition area of 1.33 cm the
2

2

surface areas are 3.73 cm and 3.65 cm for the first and subsequent layers, respectively. For a
2

27 layer structure (Fig. 3.2), we estimate the total surface area to be 98.6 cm . With this
information in hand, it is important to evaluate the amount of GOx present on/in the NS/PAS
assembly.
36

The hydrodynamic radius of GOx has been reported
2

-13

shape, the “footprint” of GOx would be A = πr = 5.81x10
14

coverage of 1.70 x 10

as 43 Å. Assuming a spherical
2

cm /GOx, or a monolayer surface

GOx for our nanoporous solid structures. Evaluation of the amount of

GOx poses a substantial experimental challenge. Because silica is not conductive, we cannot use
electrochemical means to evaluate the amount of GOx present. Spectroscopic measurements
also pose a challenge owing to the diffractive nature of the inverse opal (photonic crystal).
Another issue that is not amenable to either spectroscopic or electrochemical characterization is
the amount of GOx that is bound or adsorbed to the PAS. For these reasons, we have used
gravimetric analysis to evaluate surface loading in our systems.
For the TGA measurements on NS/PAS assemblies, three different samples were
analyzed; NS/PAS with GOx immobilized (sample), NS/PAS with no GOx present (blank), and
the PAS with GOx (control). The data shown in Fig. 3.3 were normalized first to zero mass at
640°C, then the NS/PAS with and without GOx were matched at 520°C to avoid the physically
unrealistic result of negative mass. The step-wise decrease we see for the upper two scans at
120°C is the result of dehydration; the sample is held at this temperature for 30 minutes prior to

67

5

normalilzed mass (mg)

4
3
NS/PAS + GOx

2
NS/PAS

1
PAS + GOx

0

0

100

200

300

400

500

600

700

temperature (oC)

Figure 3.3: TGA data for NS/PAS. The data were normalized to zero mass at 640°C.

68

scanning to higher temperatures. Subtracting the blank and control mass loss from the mass loss
associated with the NS/PAS with GOx immobilized resulted in an experimental enzyme loading
of ca. 35 µg, or 1.32 x 10

14

GOx molecules. This value indicates that ca. 0.77 monolayer of

GOx is present on the surface of the NS. It is significant that the TGA data show some GOx is
immobilized within the PAS. We performed control experiments to determine the activity of the
GOx contained in the PAS and found that it was inactive. This finding indicates that enzyme
activity of GOx is associated with enzyme that is bound to the NS matrix and not to the PAS. It
is has been shown that enzymes immobilized on alumina can exhibit varying levels of
reactivity.

37-39

Our TGA data indicate a significant amount of GOx trapped within the PAS, but

the enzyme assay indicates an absence of enzymatic activity. The GOx in the PAS must either
be denatured by the alumina or the active sites of the native enzyme must not be accessible to
substrate (glucose). The important point of this finding is that the enzymatic activity we observe
is associated with GOx bound to the surface of the inverse opal, and not associated with enzyme
on/in the PAS.
With the surface-loading of GOx established for our NS/PAS assemblies, we consider
next the reactivity of the bound GOx.

Solutions containing a range of β-D-glucose

concentrations in sodium acetate buffer were flowed at selected rates through the NS/PAS using
a syringe pump. For all of these measurements, a pH 4 buffer was used to avoid the formation of
iron oxide. We recognize that this is not the optimum pH for GOx enzymatic activity, but the
enzyme exhibits sufficient activity at pH 4 to allow the measurements to be made.

32,40

Our

results provide two key findings. The first is that the enzyme activity is dependent on the
reactant flow-rate. Second, GOx immobilized on the NS exhibits a higher turnover rate than

69

solution phase GOx. There are a number of possible explanations for these findings, but it is
clear that the inverse opal matrix plays a significant role in mediating the enzymatic activity of
GOx. We consider below the factors that give rise to this observed enhancement.
It is instructive to consider the turnover rate of GOx under several experimental
arrangements. Data comparing the reactivity of GOx in aqueous solution to its reactivity when
bound to planar and nanoporous solid surfaces are shown in Table 3.1. For a 100 mM glucose
solution we measure a turnover rate of 2.2 per second determined by Fenton’s assay for GOx
(present in solution at 1 mg/mL, pH 4). For GOx bound to a planar silica surface we observe a
rate of 19 turnovers per second for a 100 mM glucose solution, an enhancement of a factor of 8.6
due to surface binding by the reaction we have used. This finding is consistent with other
literature reports that show an increase in enzyme reactivity for bound enzymes relative to free
1,2

enzymes in solution.

It is important to evaluate the comparability of data from these two experimental
4

configurations. For the solution phase measurements, the ratio of glucose/GOx = 1.6x10 and
4

for GOx bound to planar silica, the ratio glucose/GOx = 1.9x10 . For these two systems there is
a large and functionally equal stoichiometric excess of glucose given the rate constant for this
41

reaction,

rendering the turnover rate results comparable. It is also useful to consider the

possible role of iron in solution. To evaluate the possibility that the presence of Fe

2+/3+

affects

the activity of GOx, assays were performed both with and without iron present during the
reaction. The data in Figure 3.4 show that Fe

2+/3+

does not affect the activity of GOx to within

the experimental uncertainty. The efficiency of the GOx reaction for a bound enzyme relative to

70

Table 3.1 Comparison of GOx reactivity for selected catalytic systems. For all measurements
reported in this Table, the glucose concentration was 100 mM.

Sample format

Flowrate
(μL/min)

TO Enhancement
GOx

TO/sec.

relative to solution
phase reactivity

b

NS/PAS

25

0.77 ML

NS/PAS

37.5

0.77 ML

NS/PAS

50

Planar solid

97

44.1

0.77 ML

28

12.7

--

1 ML

19

8.6

--

0.5 mg/mL

2.2

PAS with GOx

b

87.3

Solution phase

a

192

50

0.241 mg

undetectable

b
b

a

1
undetectable

solution phase turnover rate extrapolated to initial rate at t = 0
0.77 ML = 0.77 monolayer. Based on TGA data and surface area calculation.

71

0.30
0.25

A340

0.20
0.15
0.10
0.05
0.00

0

2

4

6

8

10

time (min)

Figure 3.4: Investigation of the effect of Fe
with Fe

2+/3+

2+/3+

on GOx activity. Solution phase measurements

present during reaction (■) and in the absence of Fe

72

2+/3+

(∆).

the solution phase enzyme cannot be explained in the context of access to the reactive site of
GOx. Rather, this enhancement can be understood in the context of the surface binding of the
enzyme locking it into a more reactive distribution of conformers than exists in the solution
phase. Another possible explanation is that there is a higher concentration of oxygen at the
surface where the reaction occurs for the immobilized enzyme. The concentration of oxygen
plays an important role in the kinetics of the reaction; however no oxygen concentration studies
were performed for these experiments.
Examination of the data for the NS/PAS assembly shows that for this flow-based system
the turnover rate (mean values) depends on the flow rate of the glucose solution (Fig. 3.5),
although the uncertainties indicate that they are not statistically different.

For all of the

experiments on this structural format, the turnover rate is substantially higher than that seen for
either the solution phase or planar solid-supported GOx. The flow rate dependence demonstrates
that the kinetics of the glucose/GOx reaction are modest, consistent with known data on this
reaction, k ~ 1.7x10

-4 -1

s

41

in solution.

What is more significant, however, is that the turnover

rate for GOx bound to the NS/PAS assembly is higher than it is for GOx bound to the planar
silica surface using the same binding chemistry. There is thus, in addition to the enhancement
associated with binding GOx to a silica surface, an enhancement in turnover rate that derives
from the inverse opal structure. We understand this effect based on the inverse opal structure
functioning as an assembly of nanoreactors.

34

The effect is based on the action of diffusion

within the nanoporous solid structure. As a reactant molecule passes through the nanoporous
solid structure, it will execute diffusive motion. If the distance between the walls of the nanoconfining volume in the nanoporous solid is less than the characteristic diffusion length,
2(Dt)

1/2

, then the reactant will experience multiple interactions with the catalyst-coated walls of

73

0.25
0.20

A340

0.15
0.10
0.05
0.00
20

25

30

35

40

45

50

flowrate (µL/min)

Figure 3.5: Absorbance values at 340 nm for the NS/PAS at different flowrates.

74

55

the nanoporous solid. We assert that the probability of a reaction event occurring is proportional
to the number of reactant-enzyme interactions.

The smaller the confining volume of the

nanoporous solid, the higher the reaction probability. For our experimental conditions, flow
rates of 25, 37.5 and 50 µL/minute correspond to 10.2, 6.8 and 5.1 second residence times,
respectively, in the nanoporous solid structure. D for glucose in aqueous solution is 6.75x10
2

cm /s

42

and d for our assemblies is 1.15x10

-4

cm. The quantity 2(Dt)

1/2

-6

/d provides a gauge of

the geometric enhancement for the NS/PAS assembly compared to the planar surface. We
calculate these enhancements to be 144 for 25 µL/min flow rate, 118 for 37.5 µL/min flow rate
and 100 for 50 µL/min flow rate. We observe experimentally these enhancements to be 10, 5.1
and 1.5 for the three flow rates.

The difference between experimental and calculated

enhancements lies partly in the admitted over-simplification of the geometric estimate we have
used and partly in the kinetic limitations to the turnover rates we report for the flow-through
NS/PAS system.
The flow rates we have used in this work were selected for two reasons. The first is that
the glucose/GOx reaction is kinetically limited, and fast flow rates lead to comparatively low
reaction efficiency. The second reason is that we are limited to slower flow rates due to the
fragile NS and PAS structures. We have attempted flow rates up to 1 mL/min, with the typical
result being fractures in the NS/PAS matrix. Such mechanical breakage allows leakage through
the NS/PAS without exposure to bound enzyme, producing results with negligible apparent
turnover rates. It is also important to note that the overall efficiency of the conversion of glucose
to hydrogen peroxide is ca. 6% for the 25 μL/min flow rate. This finding indicates that glucose
can pass through the matrix unreacted, likely due to the reaction rate and the large stoichiometric
excess of glucose used. Either the unreacted glucose makes it through the NS matrix without

75

contacting an enzyme or there are defect regions within the NS matrix that provide the reactant
with an open path to the PAS. Using a NS matrix fabricated from smaller PS spheres may give
some insight into which of these possibilities is responsible for the observed reaction efficiency.
The enhancements we report for GOx immobilized on silica inverse opal structures are
higher than the values reported for mesoporous silica matrices.

6,7

The enhanced reactivity we

observe for GOx immobilized on the planar silica support is similar to the values reported in that
work. We believe the increased turnover rate observed in that work is due to the fact that the
previous work was performed under diffusion-limited conditions. Having the ability to flow the
reactant through the supported catalyst structure increases the frequency of enzyme-substrate
interactions, resulting in a higher turnover rate. In addition to flow, the GOx molecules in our
NS may be immobilized in a way that results in a higher number of available active sites due to
the surface binding chemistry used.
Conclusions
The NS/PAS flow-through catalyst support we have demonstrated is a potentially useful
matrix for enzyme immobilization that is capable of producing enhanced turnover rates. We
report a GOx turnover enhancement of 87.3 for a 25 μL/min flow rate relative to solution phase
reactivity and an enhancement of 8.6 for the planar solid relative to solution phase.

The

enhancement associated with the nanoporous solid matrix is due to both structural stabilization
of the enzyme and geometric advantages associated with the inverse opal structure. 1.15 μm
Diameter polystyrene spheres were chosen as inverse opal templates to allow facile flow through
the structure. We are in the process of investigating the effect(s) of templating sphere size as
well as the matrix material and enzyme binding chemistry to elucidate the parameter-dependence
and optimum configuration, and to understand the potential of this structural motif for even

76

greater catalytic turnover rate enhancement. An important consideration is the improvement of
the structural integrity of the NS/PAS assembly to allow higher reactant flow rates. This issue
will be critical for systems where the catalyst possesses relatively fast reaction kinetics.

77

Literature Cited

78

Literature Cited
1. Minton, A. P. Journal of Biological Chemistry 2001, 276, 10577.
2. Zhou, H.-X.; Dill, K. A. Biochemistry 2001, 40, 11289.
3. Gupta, M. N. Thermostability of enzymes; Springer-Verlag: Secaucus, New Jersey, 1993.
4. Katchalski-Katzir, E.; Kraemer, D. M. Journal of Molecular Catalysis B: Enzymatic 2000, 10,
157.
5. Borole, A.; Dai, S.; Cheng, C.; Rodriguez, M.; Davison, B. Applied Biochemistry and
Biotechnology 2004, 113, 273.
6. Lei, C.; Soares, T. A.; Shin, Y.; Liu, J.; Ackerman, E. J. Nanotechnology 2008, 19, 1.
7. Lei, C.; Shin, Y.; Liu, J.; Ackerman, E. J. Journal of the American Chemical Society 2002,
124, 11242.
8. Kim, M. I.; Ham, H. O.; Oh, S.-D.; Park, H. G.; Chang, H. N.; Choi, S.-H. Journal of
Molecular Catalysis B: Enzymatic 2006, 39, 62.
9. Wang, Y.; Caruso, F. Chemical Communications 2004, 1528.
10. Díaz, J. F.; Balkus, K. J. Journal of Molecular Catalysis B: Enzymatic 1996, 2, 115.
11. Yan, A.-X.; Li, X.-W.; Ye, Y.-H. Applied Biochemistry and Biotechnology 2002, 101, 113.
12. Zhu, Y.; Cao, H.; Tang, L.; Yang, X.; Li, C. Electrochimica Acta 2009, 54, 2823.
13. Walcarius, A.; Kuhn, A. TrAC Trends in Analytical Chemistry 2008, 27, 593.
14. Szamocki, R.; Velichko, A.; Mücklich, F.; Reculusa, S.; Ravaine, S.; Neugebauer, S.;
Schuhmann, W.; Hempelmann, R.; Kuhn, A. Electrochemistry Communications 2007, 9, 2121.

79

15. Bon Saint Côme, Y.; Lalo, H.; Wang, Z.; Etienne, M.; Gajdzik, J.; Kohring, G.-W.;
Walcarius, A.; Hempelmann, R.; Kuhn, A. Langmuir 2011, 27, 12737.
16. Wang, Z.; Etienne, M.; Kohring, G.-W.; Bon-Saint-Côme, Y.; Kuhn, A.; Walcarius, A.
Electrochimica Acta 2011, 56, 9032.
17. Qu, F.; Nasraoui, R.; Etienne, M.; Côme, Y. B. S.; Kuhn, A.; Lenz, J.; Gajdzik, J.;
Hempelmann, R.; Walcarius, A. Electrochemistry Communications 2011, 13, 138.
18. Kleppe, K. Biochemistry 1966, 5, 139.
19. Krishnaswamy, S.; Kittrell, J. R. Biotechnology and Bioengineering 1978, 20, 821.
20. Holland, B. T.; Blanford, C. F.; Stein, A. Science 1998, 281, 538.
21. Holland, B. T.; Blanford, C. F.; Do, T.; Stein, A. Chemistry of Materials 1999, 11, 795.
22. Yan, H.; Blanford, C. F.; Holland, B. T.; Parent, M.; Smyrl, W. H.; Stein, A. Advanced
Materials 1999, 11, 1003.
23. Turner, M. E.; Trentler, T. J.; Colvin, V. L. Advanced Materials 2001, 13, 180.
24. Wijnhoven, J. E. G. J.; Vos, W. L. Science 1998, 281, 802.
25. Yang, P.; Deng, T.; Zhao, D.; Feng, P.; Pine, D.; Chmelka, B. F.; Whitesides, G. M.; Stucky,
G. D. Science 1998, 282, 2244.
26. Szamocki, R.; Reculusa, S.; Ravaine, S.; Bartlett, P. N.; Kuhn, A.; Hempelmann, R.
Angewandte Chemie International Edition 2006, 45, 1317.
27. Im, S. H.; Khalil, G. E.; Callis, J.; Ahn, B. H.; Gouterman, M.; Xia, Y. Talanta 2005, 67,
492.
28. Hatton, B.; Mishchenko, L.; Davis, S.; Sandhage, K. H.; Aizenberg, J. Proceedings of the
National Academy of Sciences 2010, 107, 10354.

80

29. Subramanian, A.; Kennel, S. J.; Oden, P. I.; Jacobson, K. B.; Woodward, J.; Doktycz, M. J.
Enzyme and Microbial Technology 1999, 24, 26.
30. Lopez-Gallego, F.; Betancor, L.; Mateo, C.; Hidalgo, A.; Alonso-Morales, N.; DellamoraOrtiz, G.; Guisan, J. M.; Fernandez-Lafuente, R. Journal of Biotechnology 2005, 119, 70.
31. Woodward, J.; Lennon, K. W.; Zanin, G.; Wagner, M.; Scott, M. A. Biotechnology Letters
1985, 7, 197.
32. Woodward, J.; Wagner, M.; Lennon, K. W.; Zanin, G.; Scott, M. A. Enzyme and Microbial
Technology 1985, 7, 449.
33. Dimos, M. M.; Blanchard, G. J. The Journal of Physical Chemistry C 2010, 114, 6019.
34. Dimos, M. M.; Blanchard, G. J. Journal of Electroanalytical Chemistry 2011, 654, 13.
35. Dimos, M. M.; Blanchard, G. J. The Journal of Physical Chemistry C 2011, 115, 11247.
36. Nakamura, S.; Hayashi, S.; Koga, K. Biochimica et Biophysica Acta (BBA) - Enzymology
1976, 445, 294.
37. Bosley, J.; Peilow, A. Journal of the American Oil Chemists' Society 1997, 74, 107.
38. de Oliveira, P. C.; Alves, G. M.; de Castro, H. F. Biochemical Engineering Journal 2000, 5,
63.
39. Tzanov, T.; Costa, S. A.; Gübitz, G. M.; Cavaco-Paulo, A. Journal of Biotechnology 2002,
93, 87.
40. Bright, H. J.; Appleby, M. J. Biol. Chem. 1969, 244, 3625.
41. Odebunmi, E. O.; Owalude, S. O. Journal of Applied Science and Environmental
Management 2007, 11, 95.
42. Gladden, J. K.; Dole, M. Journal of the American Chemical Society 1953, 75, 3900.

81

CHAPTER 4: Evaluating the Catalytic Efficiency of Alkaline Phosphatase Confined in an
Inverse Opal Structure
Introduction
Enzyme immobilization on support structures is an area of significant research activity
because of its potential for improvements in chemical processing and catalyst recovery.
Stabilization, regeneration, and the ability to easily separate the enzyme from the reaction
products are three of the most commonly targeted areas for improvement in the field. A variety
of support structures have been investigated, ranging from organic polymers to inorganic
supramolecular structures such as zeolites,

1,2

3-5

mesoporous silica,

and silica nanospheres.

6,7

It

has also been reported that enzymes can exhibit increased stability when immobilized in a
8-11

confined space.

We have reported previously on the immobilization and enhanced reactivity
12

of glucose oxidase confined in a silica inverse opal structure.

Other groups have also reported

on the immobilization of enzymes using inverse opal structures,

11,13-15

matrix format may be a useful material for biocatalysis applications.

16,17

suggesting that this

There are a number of

potential issues associated with the immobilization of enzymes in this structural motif, and it the
purpose of this work to explore some of these issues.
Inverse opal structures, sometimes referred to as nanoporous solids (NS), have
characteristics that make them an attractive choice for enzyme immobilization when used in a
flow-through system. Nanoporous solids have a relatively high surface area-to-volume ratio (ca.
5

2

3

12

1.22 x 10 cm /cm for a NS constructed using 1 μm diameter spheres)

affording a useful

loading density for enzymes. While other materials exist with higher surface area-to-volume
ratios, the combination of structural regularity, relative ease of formation and a usefully high

82

achievable loading density make inverse opal NS structures attractive for our purposes. The
regularity of the nanoscale confinement afforded by NS materials allows for predictable
12

enhancements in catalytic activity relative to catalysts bound to a planar surface.

The

geometric confinement provided by NS materials leads to a higher frequency of interactions
between the substrate and the enzyme, thus increasing the probability of a reactive event. The
manner in which the NS materials are constructed allows for significant control over the pore
15

size within these materials.

Finally, the ability to utilize these materials in a flow-through

mode reduces the exposure of the enzyme to reaction product(s), which have been shown in
18,19

some cases to denature the enzyme.

We reported previously on the immobilization of glucose oxidase (GOx) on a silica NS
12

matrix.

In that work we found that the reactivity of NS-immobilized GOx was higher than

either solution phase GOx or GOx immobilized on a planar silica surface. In this work we focus
on the reactivity of alkaline phosphatase (ALP) immobilized on a silica inverse opal structure
and compare its reactivity when immobilized in this manner to its solution phase and planar
surface-bound reactivity. ALP derived from bovine intestinal mucosa is a dimeric zinc metallo20

enzyme containing two identical subunits.

We chose ALP because it non-specifically

dephosphorylates a wide range of phosphate monoesters, resulting in an alcohol and phosphate.
A non-specific enzyme affords the ability to choose a variety of substrate molecules, which
allows for the interrogation the active site(s) of ALP to see if any information can be gained
regarding substrate specificity.

Our experimental data address both the issue of substrate-

dependent reactivity for this enzyme and the effect of immobilizing ALP. Our findings suggest
that the immobilization of ALP reduces its activity relative to that of the solution phase enzyme,

83

in contrast to our findings for GOx, even though the enhancement associated with
immobilization in a NS material relative to a planar surface remains the same.

Taken

collectively, these data demonstrate the important structural and enzyme-specific issues
associated with this approach to enzyme immobilization.
Experimental
Nanoporous solid fabrication: Polystyrene spheres with an average diameter of 1.15 μm
were synthesized using an adaptation of a free radical polymerization method. Styrene monomer
(99%, Aldrich) was distilled at 80°C to remove the stabilizer. The purified styrene was then
added to a reaction flask with ethanol, 2,2’-azobisisobutyronitrile (AIBN, 98%, Aldrich), and
polyvinylpyrrolidone (PVP, M w ~ 55,000, Aldrich). The reaction solution was refluxed at 80oC
for 24 hours. The resulting polystyrene spheres were then washed by centrifuging 3 times with
ethanol rinses. The solvent was then allowed to evaporate, and a 2% (w/w) suspension of the
polystyrene spheres was prepared in water (Milli-Q). The sphere suspension was sonicated for
30 minutes prior to each use to disperse any polystyrene aggregates that may have formed.
The same method used for the sol-gel inverse opal formation from previous work

12

was

used for this work. The sol consisted of tetraethylorthosilicate (TEOS, 99%, Aldrich), ethanol
(anhydrous), water (Milli-Q) and concentrated hydrochloric acid (12 M, CCI) in a 4:6:3:1 mass
ratio, respectively (total volume 10 mL). The sol was allowed to stir for 1 hour at room
temperature prior to use.
To form the NS structure, 0.5 mL of the sol was combined with 0.5 mL of the 2% (w/w)
nanosphere solution. The resulting mixture was deposited on a porous alumina support (PAS, d
= 22 mm, thickness = 60 μm, Whatman) with a 200 nm nominal pore diameter. The PAS was
placed in an Advantec glass microanalysis filter holder with a porous glass frit as a structural

84

support. Using a water aspirator, a vacuum was applied under the NS/PAS and the resulting
nanosphere assembly was left to dry at room temperature for a period of 24 hours.

The

nanosphere/sol structure mounted on the PAS was then removed from the filtration apparatus
and placed in toluene (≥99.5%, Mallinckrodt) to remove the polystyrene nanospheres from the
sol matrix. It was possible to flow toluene through the matrix after 24 hours of soak time.
ALP materials: All of the following chemicals were purchased from Sigma Aldrich and
used without further purification; 1-naphthol (≥99%), 2-naphthol (99%), 4-methylumbelliferone
(4-MB, ≥98%), 1-naphthyl phosphate (1-NP, disodium salt), 2-naphthyl phosphate (2-NP,
disodium salt, 97%), 4-methylumbelliferyl phosphate (4-MBP, disodium salt, premium grade),
alkaline phosphatase (ALP, derived from bovine intestinal mucosa, lyophilized powder, MW
~160 kDa), and diethanolamine (DEA, reagent grade, ≥98%).
ALP immobilization: A 5 mm × 5 mm silicon wafer with a ca. 15 Å SiO x layer was
cleaned by immersing the chip in a piranha solution (3:1 concentrated sulfuric acid:30%
hydrogen peroxide) for 30 minutes. The PAS was used without further modification. ALP was
immobilized on the surface of a planar silicon support, PAS (control), and NS/PAS using a
21-23

combination of methods adapted from the literature.

The various supports were first

immersed in a solution containing 10% (v/v) 3-aminopropyltriethoxysilane (APTES, 99%,
Aldrich) in toluene for a period of 30 minutes at 35°C. A linking moiety was then added to the
surface by placing the matrix in a 1% (v/v) aqueous glutaraldehyde solution. The supports were
then incubated in a 1 mg/mL ALP solution containing 1.0 M DEA buffer (pH adjusted to 9.8
with 6 M HCl) and 0.50 mM magnesium chloride for 24 hours at 4°C. The resulting ALPmodified supports were stored in DEA buffer at 4°C when not in use. The DEA buffer was
prepared fresh on a weekly basis and stored in the fridge.

85

Flow-through ALP on PAS/NS format: An in-house-made Teflon® flow cell was used to
hold the NS/PAS in place for the flow-through experiments. Each PAS and PAS/NS were
flushed with DEA buffer for 30 minutes to remove any nonspecifically bonded ALP from within
the matrix. After the substrates were rinsed, separate 1x10-4 M 4-MBP, 1-NP, and 2-NP in DEA
buffer solutions were flowed through the PAS and PAS/NS at 50 μL/min.

Samples were

collected in 2 mL fractions and analyzed using fluorescence measurements.
Fluorescence assays for ALP activity: All fluorescence measurements were conducted
using a Jobin-Yvon Fluorolog III spectrometer. The spectral resolution for the fluorescence
experiments was 3 nm for all measurements. All fluorescence measurements were performed at
room temperature. All measurements were made with DEA buffer as the solvent and at a total
volume of 3 mL in all cases. For the 1-naphthol assay, λ ex = 332 nm, for 2-naphthol, λ ex = 352
nm and for 4-methylumbelliferone, λ ex = 365 nm.
ALP immobilized on planar support analysis: A Si chip with ALP immobilized on the
surface was placed in the bottom of a cuvette. Buffer was added and the various substrates were
introduced (3 mL total volume) and allowed to react without stirring. The fluorescence intensity
was monitored in 2 minute intervals at the corresponding λ max for each of the substrates.
Results and Discussion
The ability to bind enzymes to a solid support is attractive for a variety of reasons, as
discussed above, but there are also a number of potential problems with this approach. Among
the issues that are challenging to evaluate and control are the details of how the enzyme is bound
to the solid support structure. There are a number of well-established methods for binding
biomolecules to (polar) surfaces, all of which rely on the identity of the functional groups that

86

are present at the enzyme surface. The details of the binding can vary significantly depending on
which amino acid functionality on the enzyme is used for this purpose, both in terms of possible
changes in the tertiary structure of the enzyme upon binding and on the accessibility of the
reactive binding pocket. It is thus possible to either enhance or diminish the reactivity of a
surface bound enzyme depending on the identity of the enzyme and the manner in which it is
bound to the support surface. In a previous report we have investigated the surface binding of
glucose oxidase (GOx) to a silica inverse opal structure. We found that the surface bound
enzyme exhibited higher reactivity than the solution phase enzyme, and that a nanoscale
confinement effect was operative (vide infra) that served to enhance the activity for inverse opalsupported enzyme relative to the same system supported on a planar silica surface. In this work
our focus is two-fold. We have chosen to use the enzyme alkaline phosphatase (ALP) because it
is a promiscuous catalyst, reacting with a variety of phosphate monoesters, allowing the
interrogation of the binding pocket by challenging the enzyme with a selection of different
phosphates. In addition, because ALP is a structurally different enzyme than GOx, we anticipate
being able to evaluate the generality, or not, of the surface-enhancement exhibited by the binding
of GOx to our inverse opal structures. In the discussion that follows we briefly consider the
rationale for the selection of the substrates used in this work and then consider the results for
each.

Comparison of solution phase results for ALP dephosphorylation of the substrates

provides some information on the nature of the reaction. Subsequent comparison of the solution
phase results to those for ALP bound to both inverse opal and silica surfaces allows for direct
evaluation of the effect of surface immobilization for this enzyme when the well-established
glutaraldehyde surface binding reaction is used (Fig. 4.1).

87

NH2 NH2 NH2 NH2
OH

OH

OH

OH
APTES

SiO2

SiO2

Glutaraldehyde
O

NH2

O

O Si O
O

ALP
NH2

+

ALP ALP ALP ALP
N

N

N

N

N

N

N

O

N

N

N

O

N

O

N

SiO2

SiO2

Figure 4.1: ALP immobilization chemistry used for planar and NS/PAS support experiments.

88

O

We have chosen three phosphate probes for these studies.

These are 4-

methylumbelliferyl phosphate (4-MBP), 1-nahpthyl phosphate (1-NP) and 2-naphthyl phosphate
(2-NP). These substrates were chosen because they are characterized by very low fluorescence
quantum yields and upon dephosphorylation the resulting alcohol products (Fig. 4.2) emit light
efficiently. Thus detection of the reaction process is facile and sensitive. The excitation and
emission spectra are shown in Figs. 4.3-4.5 for 4-MBP/4-methylumbelliferone, 1-NP/1-naphthol,
and 2-NP/2-naphthol, respectively. From these data it is clear that the substrate molecules do not
exhibit measureable fluorescence under the enzyme reaction conditions. Absorbance spectra for
each R-PO 4 /R-OH pair are shown in Fig. 4.6. The absorbance data indicate that each product
was selectively excited, with minimal interference from the substrate molecules. In addition to
these substrates providing for facile detection, the difference in chemical structure may provide
some insight into the structure-dependence of the enzymatic reaction and it is this point we
address next in the context of the reaction kinetics for solution phase reactions.
Solution phase ALP reactivity:

We have determined the substrate concentration-

dependence of the ALP enzymatic reaction rate with 4-MBP, 1-NP and 2-NP.

The

concentration-dependent time-evolution of the reactions and the concentration-dependence of the
long-time fluorescence intensities are shown in Figs. 4.7-4.9. These data demonstrate that the
reactions proceed rapidly in all cases, essentially reaching completion within two minutes after
initiation. It is important to note that for 1-NP and 2-NP, the fluorescence intensities achieved at
the completion of the reaction are not linear in substrate concentration-dependence. This finding
is indicative of either nonlinearity in the concentration-dependence of 1- and 2-naphthol
fluorescence or that the reaction proceeds to completion at lower substrate concentrations but
-4

does not achieve completion for substrate concentrations of ca. 10 M.

89

O
O

P

-

- O

OH

O

-

O

O
P

O

+

O
-

O

O
P

O

+
O

OH

ALP

H2O

O

O

Figure 4.2: Reactions for ALP with 1-NP (top), 2-NP (middle), and 4-MBP (bottom).

90

HPO4

+
HO

O

2-

+

ALP

H2O

2-

HPO4

+

-

O

-

ALP

H2O

+

HPO4

2-

normalized fluorescence

1.0

0.8

0.6

0.4

0.2

0.0
250

300

350

400

450

500

550

600

wavelength, nm

Figure 4.3: Fluorescence excitation and emission spectra for 4-MBP (dotted) and 4-MB (solid),
λex = 365 nm and λem = 448 nm for each.

91

normalized fluorescence

1.0

0.8

0.6

0.4

0.2

0.0
250

300

350

400

450

500

550

600

wavelength, nm

Figure 4.4: Fluorescence excitation and emission spectra for 1-NP (dotted) and 1-naphthol
(solid), λex = 332 nm and λem = 460 nm for each.

92

normalized fluorescence

1.0

0.8

0.6

0.4

0.2

0.0
250

300

350

400

450

500

550

600

wavelength, nm

Figure 4.5: Fluorescence excitation and emission spectra for 2-NP (dotted) and 2-naphthol
(solid), λex = 352 nm and λem = 408 nm for each.

93

Normalized Absorbance
Normalized Absorbance

Normalized Absorbance

1.0

4-MB
4-MBP

0.8
0.6
0.4
0.2
0.0

1.0
0.8

1-naphthol
1-NP

0.6
0.4
0.2
0.0

1.0
0.8

2-naphthol
2-NP

0.6
0.4
0.2
0.0

250 300 350 400 450 500 550 600
wavelength, nm

Figure 4.6: Absorbance spectra for 4-MB/4-MBP (top), 1-naphthol/1-NP (middle), and 2naphthol/2-NP (bottom).

94

100

a

80
9.17x10-8 M
9.17x10-7 M

6
counts (x10 )

60
40

9.17x10-6 M

20
0

0

50

100

100

150 200
time (sec)

250

300

b

80

counts (x106)

60
40
20
0
2

4
6
concentration (M, x10-5)

8

10

Figure 4.7: (a) Time-dependent fluorescence from the reaction of 4-MBP with ALP at different
-7
4-MBP initial concentrations, as indicated in the legend. [ALP] = 2.08x10 M. (b)
Concentration-dependence of long-time fluorescence intensity from the same reactions.

95

10

a

counts (x106)

8

9.17x10-8 M
9.17x10-7 M
9.17x10-6 M
9.17x10-5 M

6
4
2
0
0

50

100

150
200
time (sec)

250

10

300

b

counts (x106)

8
6
4
2
0
0

2

4

6

8

10

concentration (M, x10-5)

Figure 4.8: (a) Time-dependent fluorescence from the reaction of 1-NP with ALP at different 1-7
NP initial concentrations, as indicated in the legend. [ALP] = 5.21x10 M. (b) Concentrationdependence of long-time fluorescence intensity from the same reactions.

96

a

50

9.17x10-8 M
9.17x10-7 M
9.17x10-6 M
9.17x10-5 M

counts (x106)

40
30
20
10
0
0

50

100

150

time (sec)

200

250

300

b

50

counts (x106)

40
30
20
10
0
0

2

4

6

concentration (M, x10-5)

8

10

Figure 4.9: (a) Time-dependent fluorescence from the reaction of 2-NP with ALP at different 2-7
NP initial concentrations, as indicated in the legend. [ALP] = 5.21x10 M. (b) Concentrationdependence of long-time fluorescence intensity from the same reactions.

97

The functional form of the kinetic data is sufficiently complex that it is difficult to assign
a detailed model for the reaction based on these results. One effective means of evaluating the
kinetic information presented in Figs. 4.7-4.9 is to measure the initial rates of reaction. We
present these data in Figs. 4.10-4.12. These data show clearly that the kinetics of reaction for 4MBP with ALP are different than for 1-NP and 2-NP with ALP. For 4-MBP there appears to be
a much less pronounced concentration dependent initial rate than is seen for either NP. We are
aware that the concentration range over which we report the data for 4-MBP is smaller than it is
for either 1-NP or 2-NP and this is for reasons of the fluorescence quantum yield of the reaction
produces. With that caveat, at the low range of the concentrations we have studied, it appears
that the initial rate for 4-MBP reaction with ALP is close to concentration-independent (Fig.
4.10).

For 1-NP (Fig. 4.11) and 2-NP (Fig. 4.12) there is a much more pronounced

concentration-dependence to the initial rates and for these two compounds the initial rates are
similar for a given concentration. This finding suggests that the location of the phosphate
functionality on the naphthyl phosphates is of secondary importance relative to the size of the
substrate. We view this finding as somewhat surprising in light of the influence of the naphthyl
substitution on the absorbance spectra (Fig. 4.6). The difference in spectral response for these
two substrates is consistent with a different electron density distribution in their π and π* states,
yet the dephosphorylation reaction does not seem to be affected to a great extent. It is perhaps
because the reaction involves the cleavage of σ-bond(s) and the π-electrons are not involved
directly in the reaction.
It is also instructive to consider that the initial rates of reaction for 1-NP and 2-NP appear
to exhibit an onset of saturation at higher substrate concentrations.
-4

concentration range of the substrate is less than 10

98

Given that the upper

M and that there is no spectroscopic

evidence for aggregate formation, it appears that the presence of sufficient reaction product (1and 2- naphthol) could serve to occupy the reactive site of ALP and thus diminish the achievable
rate of reaction for this system.
It is clear from our solution phase data that the reaction of these substrates with ALP is
facile and that the detection methodology we use is sufficiently sensitive for the purpose. It is
also important to note that, based on the initial rate data (Figs. 4.10-4.12) that the kinetics of
reaction for 4-MBP are different from those of the naphthyl phosphates. We believe that the
reason for this difference lies in the structural differences of the substrates, although the extent to
which structural issues influence the reaction kinetics is limited, as seen in the similarity of the
initial rate data for 1-NP and 2-NP. With this information in hand, we examine next these same
reactions where ALP is bound to inverse opal and planar silica support structures.
Surface-bound enzyme reactivity: One of the central issues of this work is the catalytic
efficiency of surface-bound ALP relative to solution phase ALP. As noted above, the binding of
an enzyme to a surface is likely to alter the reactivity of the enzyme but it is not clear, a priori,
whether the reactivity will be increased or decreased upon attachment to a surface. In our
previous work (Chapter 3) we observed enhanced catalytic activity for surface bound GOx
relative to solution phase GOx, and there was a further reactivity dependence on the structure of
the support matrix. In this work, as is described below, we find that the reactivity of ALP is
reduced when it is bound to a surface, but the dependence of enzyme reactivity on the structure
of the support matrix is similar to what was observed for GOx. These data, taken collectively,
serve to provide a useful limit on the range of applicability of surface-binding of enzymes using
glutaraldehyde linking chemistry.

99

a

normalized counts

0.40
0.35
0.30
0.25

9.17x10-8 M
9.17x10-7 M
9.17x10-6 M

0.20
0.15
0.10

0

2

4

rate (∆counts/sec)

0.04

6

8

m = 0.023 ± 0.001
m = 0.022 ± 0.006
m = 0.028 ± 0.002

10 12 14 16 18 20

time (sec)

b

0.03
0.02
0.01
0.00

0

2

4
6
concentration (M, x 10-5)

8

10

Figure 4.10: (a) Initial rate data for the reaction of 4-MBP with ALP at different 4-MBP initial
-7
concentrations, as indicated in the legend. [ALP] = 2.08x10 M. (b) Concentration-dependence
of the initial rate data for the same reactions.

100

normalized counts

0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1

a
9.17x10-8 M
9.17x10-7 M

m = 0.003 ± 0.004
m = 0.013 ± 0.001

9.17x10-6 M
9.17x10-5 M

0

2

4

6

m = 0.018 ± 0.001
m = 0.032 ± 0.001

8 10 12 14 16 18 20
time (sec)

initial rate (∆counts/sec)

0.04

b

0.03
0.02
0.01
0.00

0.0

2.0

4.0

6.0

concentration (M, x10-5)

8.0

10.0

Figure 4.11: (a) Initial rate data for the reaction of 1-NP with ALP at different 1-NP initial
-7
concentrations, as indicated in the legend. [ALP] = 5.21x10 M. (b) Concentration-dependence
of the initial rate data for the same reactions.

101

a

normalized counts

0.6
9.17x10-8 M
9.17x10-7 M

0.5
0.4

m = 0.017 ± 0.002
m = 0.027 ± 0.001

9.17x10-6 M
9.17x10-5 M

0.3

m = 0.033 ± 0.0005
m = 0.036 ± 0.003

0.2
0

2

4

6

8 10 12 14 16 18 20
time (sec)

b

rate (∆counts/sec)

0.04
0.03
0.02
0.01
0.00

0

2
4
6
concentration (M, x10-5)

8

10

Figure 4.12: (a) Initial rate data for the reaction of 2-NP with ALP at different 2-NP initial
-7
concentrations, as indicated in the legend. [ALP] = 5.21x10 M. (b) Concentration-dependence
of the initial rate data for the same reactions.

102

A schematic of the nanoporous solid/porous alumina support (NS/PAS) structure used
here is shown in Fig. 4.13. The details of constructing this interface have been presented in
Chapter 2 of this dissertation. ALP was bound to the NS using a glutaraldehyde linking reaction
(Fig. 4.1) and a key issue for such systems is characterizing the loading density achieved with
this process.

We have chosen to use silica as a support material because of its surface

functionality and consequent ease of surface-binding chemistry. Another key aspect in our
choice of silica as a support matrix is its comparative inertness relative to a variety of metals
(e.g. Au, Pt, Pd) and some oxides (e.g. Al 2 O 3 ), which have been shown to denature other
enzymes.

12

A critical limitation to the use of silica, however, lies in the ability to quantitate the

amount of enzyme present on the surface.

Because of the insulating nature of silica,

electrochemical characterization is not possible and the inverse opal structure precludes
quantitative spectroscopic measurements because of their diffractive properties at optical
wavelengths. We have used thermogravimetric analysis previously (TGA) to quantitate the
amount of GOx present on silica inverse opal structures, and we determined that the surface
chemistry we used produced a coverage of ca. 0.77 monolayer. For ALP bound to silica inverse
opals, however, we have not been able to acquire reliable TGA data. We believe that the reason
for this limitation lies with the buffer system used, which contains diethanolamine. The DEA in
the buffer produces residue upon heating, resulting in a mass loss well in excess of that
achievable even with full monolayer coverage of ALP. Based on the surface binding chemistry
we use, one monolayer of enzyme is the maximum coverage achievable and the geometric
properties of the inverse opal structure allow for accurate calculation of surface area. For these
reasons, we have estimated the surface coverage of ALP to be the same as that which we
achieved for GOx. The number of ALP molecules on the surface was calculated using a radius

103

Nanoporous solid adlayer

Adlayer/support
Porous alumina support

20 μm

Figure 4.13: Overview of the construction of the NS/PAS supports used for ALP
immobilization (top) and a cross-section SEM image of the NS/PAS interface (bottom).

104

24

of 7.7 nm which has been reported in the literature.

We recognize that this situation is far from

ideal but, based on the observed reactivities for these systems, any reasonable deviation from the
estimated surface coverage will not have a significant effect on the larger lessons learned from
this work. In other words, even if our estimate of surface coverage is substantially too high, it
will not alter the fact that the reactivity of bound ALP is much less than that of solution phase
ALP.
In contrast to the solution phase measurements, studies of ALP bound to a silica
nanoporous solid were performed under flowing conditions at a flow rate of 50 µL/min. We
have performed flow-based experiments on both NS/PAS and PAS structures, where the latter
porous alumina structure was reacted with ALP and the same glutaraldehyde linking chemistry
used for binding of the enzyme to the nanoporous solid silica support. Our motivation for
performing this latter study was to determine whether or not the enzyme activity was influenced
by exposure to the alumina surface. As noted above, GOx was found to denature when exposed
to the porous alumina support surface. We present these data in tabular form (Table 4.1).
The data presented in Table 4.1 provide a direct comparison of the reactivity of ALP
toward the three substrates in four different structural formats, three with the enzyme surfacebound and the fourth in the solution phase. Because of the different structural formats, there are
a number of specific points that needed to be addressed in order to make these different bodies of
data comparable. For the solution phase measurements, as discussed above, the turnover rates
were determined from the initial rate data, the detection sensitivity for each substrate (number of
fluorescence counts per unit solution concentration) and the concentration of the enzyme. For
13

our instrument, the sensitivity for 4-methylumbelliferone is 1.44x10
4.91x10

11

11

counts/M and for 2-naphthol 1.46x10

105

counts/M, for 1-naphthol

counts/M (Table 4.1, Fig. 4.14). These

Table 4.1. Comparison of different substrates for various reaction formats. (Uncertainties are
±1σ)
Substrate
a

[substrate]
(M)
-6

Solution Phase

4-MBP
1-NP

9.17x10

2-NP

9.17x10
[substrate]
(M)

-6

1.06x10

-6

-6

9.35x10

TO/sec

-7

28.8±8.9

-7

2.04±0.03

-7

17.9±3.3

2.08x10
5.21x10
5.21x10

Initial
slope
(cts/s)

Calib.
(cts/M)

-5

6349

(1.44±0.08)
13
x10

4.41x10

-5

248

(4.91±0.28)
11
x10

5.05x10

-5

179

(1.46±0.13)
11
x10

1.23x10

9.17x10

1-NP

9.17x10

2-NP

Planar

6.00x10

-6

4-MBP

9.17x10

Substrate

[ALP]
(M)
-6

9.17x10

Substrate

[substrate]
(M)
-5

4-MBP
PAS

Rate
(M/s)

6.52x10

1-NP

2.10x10

2-NP

5.26x10

-5
-5

Rate
c
(mol/s)
-11

5.43x10

-11

1.75x10

-11

4.39x10

106

Rate
(M/s)

[ALP]

-10
-10

ALP
(mol)

-9

b
-7

(8.16±4.62)
-4
x10

-7

(9.35±5.46)
-4
x10

-7

(2.27±0.56)
-3
x10

5.40x10
5.40x10

5.40x10

TO/sec
-9

2.82x10

-9

2.82x10

-9

2.82x10

TO/sec

(1.93±0.45)x10
(6.02±1.33)x10
(1.55±0.16)x10

-2
-3
-2

Table 4.1 (cont’d)
Substrate

[substrate]
(M)
-5

a

3.94x10

1-NP

5.37x10

2-NP

NS/PAS

4-MBP

7.04x10

-5
-5

Rate
(mol/s)b
-11

3.29x10

-11

4.47x10

-11

5.86x10

ALP
(mol)
3.04x10
3.04x10
3.04x10

TO/sec
-9
-9
-9

(1.08±0.32)x10
(1.47±0.03)x10
(1.93±0.06)x10

-2
-2
-2

N=2 for 4-MBP solution phase, N=3 for all other measurements.

b

Determined by the number of moles of ALP bound to the silica surface and the volume of the
system determined by the diffusion length (depth) and the geometric area of the surface.
c

Rate calculated using the calibration factors given for the planar surface data and the flow rate
of 50 µL/min.

107

sensitivity factors could, in principle, be calculated from the extinction coefficient and
fluorescence quantum yields of the chromophores at the wavelengths of excitation and emission,
and the throughput and gain factors for the fluorescence spectrometer, but none of these
quantities are known reliably. Rather than use estimates, we determined these calibration factors
experimentally (Fig. 4.14).

With these data we determine the turnover rates for the

substrate/ALP reactions, and show these data in Table 4.1.
Determining the turnover rate for the enzyme bound to the planar surface is accomplished
by measuring the count rate for production of the product (Fig. 4.15) and converting that count
rate to a concentration with the sensitivity factors given in Table 4.1 (vide infra). Determining
the amount of enzyme present and the relevant volume of the reactive system involves two
assumptions. Because the concentration of surface-bound ALP is not directly measurable, we
estimate the surface loading to be 0.77 monolayer, as discussed above based on the surface
chemistry used to attach the enzyme. The central issue in this determination is the relevant
volume in which the reaction occurs. Because we operate under non-stirred conditions for these
measurements, diffusion is, at best, the limiting factor in delivery of substrate to the enzyme. We
thus assume the relevant distance from the enzyme-coated surface is the diffusion length, l =
2(Dt)

1/2

. Taking D = 10

-5

2

-3

cm /s, for t = 1 s, l = 6.32x10

cm. Under these assumptions we

estimate the turn-over rates for the substrates used here at a planar ALP-covered silica surface.
We note that, of all the experimental formats, the determinations for ALP bound to a planar silica
surface is associated with the largest number of approximations and is thus characterized by the
greatest uncertainty.
Quantitation of the turnover rates for the three substrates at the NS/PAS and PAS
structures involves the measurement of product under flow-through conditions at a flowrate of

108

16
14
12

counts (x107)

10
m = (1.46±0.13)x1011 counts/M
m = (4.91±0.28)x1011 counts/M

1-naphthol

8

2-naphthol

6

m = (1.44±0.08)x1013 counts/M

4-MB

4
2
0
-2

0

2

4

6

8

10

concentration (M, x10-5)

Figure 4.14: Calibration data for the reaction products used in this work. Slopes are the
sensitivity factors for our system, given in counts/M and are best fits of data points contained in
the linear range for each reaction product.

109

1.8

4-MBP
1-NP
2-NP

1.6

Normalized Fluorescence

1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
0

2

4

6

8 10 12 14 16 18 20 22 24 26 28 30 32

time, minutes

Figure 4.15: Time-dependent fluorescence data for ALP immobilized on planar silica for each of
-6
the three different substrates. The concentration of the substrate was 9.17x10 M for all cases.

110

50 µL/min in all cases. We determine the turnover rates for these experimental configurations
based on the flowrate and the amount of product formed, assuming a surface coverage of 0.77
monolayer (vide infra). Because we use the same surface coverage for all of our surface-bound
ALP data interpretation, the absolute turnover rates may be consistently biased by a small
amount but comparison between the surface-bound experimental formats is valid.
Our turnover rate data are presented in Table 4.1 and are summarized in normalized form
in Table 4.2. A number of important points emerge from these data. The first is that when
comparing the reactivity of the different substrates for solution phase measurements we find that
there is a noticeable structure-dependence on turnover rate. The turnover rate for 4-MBP and 2NP differ by a factor of two and both are approximately an order of magnitude larger than that
for 1-NP. Given the locations of the phosphate groups on these substrates, it is tempting to
speculate that the shape of the binding pockets are such that 2-NP and the phosphate-containing
ring of 4-MBP fit the ALP binding pockets better than 1-NP does. This result is interesting in
light of the kinetic initial rate data (Figs. 4.10-4.12) which give similar concentrationdependencies for the two naphthyl derivatives and a distinctly different concentrationdependence for 4-MBP. The TO rate data thus are related to steric factors while the kinetic data
are reflective of mechanistic differences between different classes of substrates. The fact that the
4-MBP and 1-NP TO rate results are so similar suggests that the rate-limiting factor, at least in
the concentration range we have examined, is determined to some extent by delivery of the
substrate to the enzyme. For the concentration of substrate and enzyme used, we estimate for 4MBP and ALP that there should be on the order of 500 collisions between the two compounds
per second, and the measured TO rate is ca. 5% of the collision rate. For 1-NP and 2-NP we
estimate there to be ca. 700 collisions with ALP per second, and the TO rates for these two

111

Table 4.2. Relative turnover rates for the substrates and ALP in different formats. The turnover
rates are normalized to a solution phase rate of 28.8/s for 4-MBP.
Solution

PAS

4-MBP

1

6.7x10

1-NP

0.069

2.2x10

2-NP

0.63

5.4x10

112

NS/PAS
-4
-4
-4

-4

3.8x10

-4

5.1x10

-4

6.7x10

Planar
-5

2.8x10

-5

3.3x10

-5

7.9x10

substrates is ca. 0.3% and 2.6% of the collision rate, respectively. The issue that we cannot
address is the probability of a reactive interaction taking place between the substrate and ALP
based on geometric factors alone.
Comparison of the TO rates for surface-bound ALP to solution phase ALP for the
substrates we have studied shows that the binding of the enzyme to a silica surface using
glutaraldehyde binding chemistry results in a substantial loss of reactivity. This loss of reactivity
could be associated either with the binding of the enzyme in such a way that access to the
reactive site is blocked or with the enzyme being locked into a conformation that is less reactive
than the solution phase conformation. If the latter explanation were operative, we would expect
there to be a substantial difference in the reactivity of ALP when bound to silica and alumina and
we find this not to be the case. The findings for ALP stand in contrast to those for GOx in this
regard.

12

Further investigations using different enzyme binding chemistry may be able to

provide more insight into this issue.
Despite the fact that the surface-binding of ALP reduces its reactivity relative to what it is
in the solution phase, the issue of reactivity enhancement associated with nanoconfinement can
still be evaluated for this system. As was discussed in Chapter 1, flowing substrate solution
through a structure with features that have a characteristic length scale shorter than the diffusion
length of the substrate gives rise to more opportunity for interaction between substrate and
enzyme than if the enzyme were attached to a planar surface. The enhanced frequency of
interactions leads to higher reactivity and thus the apparent enhancement. Comparing our data
for the nanoporous structure(s) to that for the planar silica surface shows the enhancement effect
expected on geometric grounds. This finding, while not surprising, provides further proof of the
geometric enhancement in catalytic activity observed for nanoporous structures. We anticipate

113

that the enhancement could be increased with a decrease in the diameter of the characteristic
void space in our nanoporous structures, but this enhancement would occur at the expense of the
flow rate achievable and thus the throughput of the system. The optimum balance between
reactivity enhancement and throughput depends on the kinetics of the enzyme used and is
expected to be different for each enzyme-substrate pair. We have not attempted to identify this
optimum operating point for either ALP or GOx and, in light of our findings for surface binding
of ALP, finding the optimum for that system would serve no practical purpose.
Conclusions
We have reported on the reactivity of the enzyme alkaline phosphatase (ALP) with
substrates 1-naphthyl phosphate, 2-naphthyl phosphate and 4-methylumbelliferyl phosphate, in
solution and with the enzyme bound to planar and nanoporous silica structures. Our data
demonstrate that both the mechanism and the rate of reactivity depend on the chemical identity
of the substrate. Comparison of the solution phase data to that for surface bound enzyme
demonstrates a substantial loss of reactivity upon surface attachment. There are two possible
causes for this loss of reactivity; blocking of the enzyme reactive site by virtue of the way in
which it is bound to the support structure, and loss of reactivity due to locking the enzyme into a
less reactive conformation upon binding. The fact that the TO rate for ALP is the same for
alumina and silica surfaces suggests that the former explanation is operative. Despite the loss of
reactivity upon surface binding, we observe the geometrically-derived enzyme reactivity
enhancement associated with nanoscale confinement. Our data, taken collectively, serve to
underscore the importance of enzyme surface binding chemistry and geometric factors in
achieving enhanced reactivity.

114

Literature Cited

115

Literature Cited
1. Díaz, J. F.; Balkus Jr, K. J. Journal of Molecular Catalysis B: Enzymatic 1996, 2, 115.
2. Yan, A.-X.; Li, X.-W.; Ye, Y.-H. Applied Biochemistry and Biotechnology 2002, 101, 113.
3. Borole, A.; Dai, S.; Cheng, C.; Rodriguez, M.; Davison, B. Applied Biochemistry and
Biotechnology 2004, 113, 273.
4. Lei, C.; Soares, T. A.; Shin, Y.; Liu, J.; Ackerman, E. J. Nanotechnology 2008, 19, 1.
5. Lei, C.; Shin, Y.; Liu, J.; Ackerman, E. J. Journal of the American Chemical Society 2002,
124, 11242.
6. Kim, M. I.; Ham, H. O.; Oh, S.-D.; Park, H. G.; Chang, H. N.; Choi, S.-H. Journal of
Molecular Catalysis B: Enzymatic 2006, 39, 62.
7. Wang, Y.; Caruso, F. Chemical Communications 2004, 1528.
8. Kim, J.; Jia, H.; Lee, C.-w.; Chung, S.-w.; Kwak, J. H.; Shin, Y.; Dohnalkova, A.; Kim, B.G.; Wang, P.; Grate, J. W. Enzyme and Microbial Technology 2006, 39, 474.
9. Minton, A. P. Journal of Biological Chemistry 2001, 276, 10577.
10. Zhou, H.-X.; Dill, K. A. Biochemistry 2001, 40, 11289.
11. Zhu, Y.; Cao, H.; Tang, L.; Yang, X.; Li, C. Electrochimica Acta 2009, 54, 2823.
12. Gornowich, D. B.; Blanchard, G. J. The Journal of Physical Chemistry C 2012, 116, 12165.
13. Walcarius, A.; Kuhn, A. TrAC Trends in Analytical Chemistry 2008, 27, 593.
14. Szamocki, R.; Velichko, A.; Mücklich, F.; Reculusa, S.; Ravaine, S.; Neugebauer, S.;
Schuhmann, W.; Hempelmann, R.; Kuhn, A. Electrochemistry Communications 2007, 9, 2121.

116

15. Bon Saint Côme, Y.; Lalo, H.; Wang, Z.; Etienne, M.; Gajdzik, J.; Kohring, G.-W.;
Walcarius, A.; Hempelmann, R.; Kuhn, A. Langmuir 2011, 27, 12737.
16. Wang, Z.; Etienne, M.; Kohring, G.-W.; Bon-Saint-Côme, Y.; Kuhn, A.; Walcarius, A.
Electrochimica Acta 2011, 56, 9032.
17. Qu, F.; Nasraoui, R.; Etienne, M.; Côme, Y. B. S.; Kuhn, A.; Lenz, J.; Gajdzik, J.;
Hempelmann, R.; Walcarius, A. Electrochemistry Communications 2011, 13, 138.
18. Kleppe, K. Biochemistry 1966, 5, 139.
19. Krishnaswamy, S.; Kittrell, J. R. Biotechnology and Bioengineering 1978, 20, 821.
20. Fosset, M.; Chappelet-Tordo, D.; Lazdunski, M. Biochemistry 1974, 13, 1783.
21. Subramanian, A.; Kennel, S. J.; Oden, P. I.; Jacobson, K. B.; Woodward, J.; Doktycz, M. J.
Enzyme and Microbial Technology 1999, 24, 26.
22. Ehlert, N.; Müller, P. P.; Stieve, M.; Behrens, P. Microporous and Mesoporous Materials
2010, 131, 51.
23. Surinenaite, B.; Bendikiene, V.; Juodka, B. Prog. Biotechnol. 1998, 15, 577.
24. Ey, P. L.; Ferber, E. Biochimica et Biophysica Acta (BBA) - Enzymology 1977, 480, 403.

117

CHAPTER 5: Conclusions
Summary
Enzymes play an important role in many processes that span a multitude of different
industries including food, beverage, drug discovery, chemical manufacturing, and biofuels. In
2010, the global market for industrial enzymes was estimated to be worth $3.3 billion.1
Consequently, a great deal of research is aimed at optimizing various parameters for maximum
2,3

enzyme efficiency (kg product/kg enzyme).

Immobilization of enzymes continues to be an

area of focus because immobilizing the enzyme has several advantages, including increased
4,5

enzyme stability,

the relative ease with which the enzyme can be recovered and reused, and in
6-8

some cases enhancement in reactivity.

The goal of this project was to evaluate the utility of

the inverse opal structure function as a flow-through catalyst support. To this end, an inverse
opal structural format was designed and utilized as a support for immobilized enzymes that could
be used in a flow-through mode. Chapter 2 outlined the procedure involved in creating a flowthrough inverse opal matrix, and Chapters 3 and 4 provided experimental results on two different
enzymes that were chosen to characterize this approach to catalyst immobilization and reaction.
These results provide a useful initial point in evaluating the utility of inverse opal structures as
enzyme supports, but there are limitations to this approach based on the enzyme used and the
manner in which it is bound to the inverse opal structure.
Inverse opal structures have been created using a wide range of different materials, most
8-27

commonly metals and inorganic oxides.

The moderately high surface area provided by these

materials allows for a correspondingly high enzyme loading density.

Most inverse opal

structures are created on planar support structures, which don’t allow for a facile flow-through

118

experimental configuration. For biocatalysis applications using a non-flow through format, the
delivery of substrate to the enzyme is limited by diffusion, thereby not taking advantage of the
full surface area provided by the multi-layer NS structure. The flow-through NS system that was
described in Chapter 2 overcomes this limitation and allows for a reactant stream to be flowed
through the matrix. Consequently, the number of substrate-enzyme interactions is increased,
8

which can lead to an enhancement in enzyme reactivity. However, the chemistry used to attach
the enzyme to the support structure must leave the enzyme in an active form with the binding
pocket readily accessible, or the activity may be diminished.

23

Glucose oxidase (GOx) and alkaline phosphatase (ALP) were chosen as model enzymes
to investigate whether or not an enhancement in activity could be observed for our flow-through
inverse opal structure. The two enzymes are similar in size; however GOx has only one specific
target substrate (β-D-glucose), while ALP dephosphorylates a wide range of phosphate
monoesters.

The results for GOx showed an enhancement in reactivity for the enzyme

immobilized on a planar support relative to its reactivity in solution. This finding indicated an
enhancement in reactivity due to the enzyme immobilization.

The data also revealed an

enhancement in reactivity for GOx immobilized on the NS structure relative to the reactivity of
the enzyme on the planar support.

These data demonstrated an enhancement due to the
8

nanoconfinement of the enzyme within the NS matrix.

The concentration of oxygen at the

support surface may have also contributed to the enhancement in turnover rate, however oxygen
studies were not performed. Results for ALP, however, showed different behavior. The solution
phase reactivity of ALP was higher than when it was used in immobilized reaction formats.
While there was still an enhancement observed for the enzyme within the NS structure relative to

119

its reactivity when bound to a planar structure, the data illustrated the limitations involved with
the immobilization process.

23

There are a couple of limitations that are important to point out. The methods used for
the immobilization process can determine the reactivity of the enzyme based on which
functionalities are used in the covalent linkage between the enzyme and the support. As a result,
each enzyme behaves differently for a given immobilization technique, making it challenging to
determine the optimum immobilization chemistry for a given enzyme/support system. Another
important issue is the quantitation of enzyme loading on the silica inverse opals. Thermogravimetric measurements were used for this work; however the ALP loadings were not able to
be determined due to large mass losses associated with the buffer system. The incorporation of
metals either within the dielectric matrix or the fabrication of a flow-through inverse opal using a
conductive material will prove beneficial for the quantitation of the enzyme loading.

A

conductive NS matrix would allow electrochemical methods to be used for surface coverage
evaluation.
Future Directions
While this work has illustrated the utility of NS structures as supports for enzyme
immobilization, there is still work to be done. With an enhancement observed for an enzyme
supported on an inverse opal structure, it is important to optimize the flow-through NS system to
maximize the enzyme efficiency. The diffusion length vs. void size model indicates that as the
diameter decreases, the number of substrate-enzyme interactions per unit time will increase. A
relationship between pore size and flow rate would be useful to evaluate so that an enzyme could
have an optimized set of parameters to maximize the reactivity. The optimized inverse opal
structural parameters differ for each enzyme/substrate system and is also expected to depend on

120

the enzyme-attachment chemistry used.

Thus it will also be useful to compare several

immobilization reactions that will bind the enzyme to the support using different functionalities
present on the outer “surface” of the enzyme. As mentioned previously, each enzyme may
exhibit different levels of reactivity based on how the enzyme is immobilized.
28

immobilization techniques have been investigated in the literature for planar supports,

Other
and it

would be useful to compare these immobilization methodologies for the flow-through NS
system.
As noted above, a present limitation is the quantitation of the enzyme loading. The
gravimetric analysis is not ideal for this measurement, because the mass losses are low and such
measurements are characterized by a high background.

23

Consequently, exploration of

alternative inverse opal matrix materials with an emphasis on conductive materials would be an
important direction to move in. The use of electrochemistry to provide information on the
surface coverage of the enzyme could also allow for additional characterization of how the
enzyme is bound to the support, depending on the details of the enzyme electrochemistry.
Devising flow-through metal NS structures or incorporating metal nanowires into the inverse
opal structure could be useful avenues of investigation. Enzyme stability is another issue that
requires evaluation. As discussed in Chapter 1, immobilization of the enzyme can increase its
stability, and this issue would have to be characterized more thoroughly if this structural motif is
to be considered for practical applications. In the work presented in this dissertation, long term
stability was not addressed as most experiments were completed within a one- to two-week timeframe. In the future, the long term stability of enzymes immobilized within the NS structures
must be addressed to calculate a measure of kg of product formed per kg of enzyme.

121

Overall Conclusions
In conclusion, creating a flow-through NS proved to be challenging. Due to the fragile
nature of the material, careful consideration had to be taken when fabricating and handling these
structures. While there are still some areas that need to be investigated further, this work has
illustrated the potential and limitations for immobilizing enzymes on dielectric inverse opal
structures. Once the various parameters are optimized, flow-through NS structures could prove
to be a useful format for biocatalysis applications, but their expansion to industrial-scale
operations would be problematic. It is most likely that this approach to enzyme immobilization
and reactivity will find its greatest use in research laboratory or other small volume enzymatic
reaction applications.

122

Literature Cited

123

Literature Cited
1. http://www.bccresearch.com/report/enzymes-industrial-applications-bio030f.html.
2. Sheldon, R. A. Advanced Synthesis & Catalysis 2007, 349, 1289.
3. Hwang, E. T.; Gu, M. B. Engineering in Life Sciences 2013, 13, 49.
4. López-Gallego, F.; Betancor, L.; Mateo, C.; Hidalgo, A.; Alonso-Morales, N.; DellamoraOrtiz, G.; Guisán, J. M.; Fernández-Lafuente, R. Journal of Biotechnology 2005, 119, 70.
5. Zhou, H.-X.; Dill, K. A. Biochemistry 2001, 40, 11289.
6. Lei, C.; Shin, Y.; Liu, J.; Ackerman, E. J. Journal of the American Chemical Society 2002,
124, 11242.
7. Lei, C.; Soares, T. A.; Shin, Y.; Liu, J.; Ackerman, E. J. Nanotechnology 2008, 19, 1.
8. Gornowich, D. B.; Blanchard, G. J. The Journal of Physical Chemistry C 2012, 116, 12165.
9. Abramova, V.; Sinitskii, A. Superlattices and Microstructures 2009, 45, 624.
10. Cao, Y.; Wang, Y.; Zhu, Y.; Chen, H.; Li, Z.; Ding, J.; Chi, Y. Superlattices and
Microstructures 2006, 40, 155.
11. Gladden, J. K.; Dole, M. Journal of the American Chemical Society 1953, 75, 3900.
12. Holland, B. T.; Blanford, C. F.; Do, T.; Stein, A. Chemistry of Materials 1999, 11, 795.
13. Holland, B. T.; Blanford, C. F.; Stein, A. Science 1998, 281, 538.
14. Holland, B. T.; Blanford, C. F.; Stein, A. Science 1998, 281, 538.
15. Turner, M. E.; Trentler, T. J.; Colvin, V. L. Advanced Materials 2001, 13, 180.

124

16. Yang, P.; Deng, T.; Zhao, D.; Feng, P.; Pine, D.; Chmelka, B. F.; Whitesides, G. M.; Stucky,
G. D. Science 1998, 282, 2244.
17. Yu, H. M.; Yim, J.-H.; Choi, K. Y.; Lim, J. S. The Journal of Supercritical Fluids 2012, 67,
71.
18. Ma, J.; Parajuli, B. R.; Ghossoub, M. G.; Mihi, A.; Sadhu, J.; Braun, P. V.; Sinha, S. Nano
Letters 2013, 13, 618.
19. Mu, W.; Hwang, D.-K.; Chang, R. P. H.; Ketterson, J. B. Journal of Raman Spectroscopy
2011, 42, 941.
20. Dimos, M. M.; Blanchard, G. J. The Journal of Physical Chemistry C 2010, 114, 6019.
21. Dimos, M. M.; Blanchard, G. J. Journal of Electroanalytical Chemistry 2011, 654, 13.
22. Dimos, M. M.; Blanchard, G. J. The Journal of Physical Chemistry C 2011, 115, 11247.
23. Gornowich, D. B.; Blanchard, G. J. The Journal of Physical Chemistry C 2013, in
preparation.
24. Qu, F.; Nasraoui, R.; Etienne, M.; Côme, Y. B. S.; Kuhn, A.; Lenz, J.; Gajdzik, J.;
Hempelmann, R.; Walcarius, A. Electrochemistry Communications 2011, 13, 138.
25. Szamocki, R.; Reculusa, S.; Ravaine, S.; Bartlett, P. N.; Kuhn, A.; Hempelmann, R.
Angewandte Chemie International Edition 2006, 45, 1317.
26. Szamocki, R.; Velichko, A.; Mücklich, F.; Reculusa, S.; Ravaine, S.; Neugebauer, S.;
Schuhmann, W.; Hempelmann, R.; Kuhn, A. Electrochemistry Communications 2007, 9, 2121.
27. Walcarius, A.; Kuhn, A. TrAC Trends in Analytical Chemistry 2008, 27, 593.
28. Subramanian, A.; Kennel, S. J.; Oden, P. I.; Jacobson, K. B.; Woodward, J.; Doktycz, M. J.
Enzyme and Microbial Technology 1999, 24, 26.

125