Viral Templated Palladium Nanocatalysts

Article abstract of DOI:10.1002/cctc.201500381

Palladium (Pd) nanocatalysis plays key roles in many areas from environmental remediation, energy utilization to chemical synthesis. Viruses offer exciting opportunities and advantages as promising nanoparticle synthesis templates due to their controlled

Full text of DOI:10.1002/cctc.201500381

DOI: 10.1002/cctc.201500381
Minireviews
Viral Templated Palladium Nanocatalysts
Cuixian Yang and Hyunmin Yi*[a]
Palladium (Pd) nanocatalysis plays key roles in many areas
from environmental remediation, energy utilization to chemical
synthesis. Viruses offer exciting opportunities and advantages
as promising nanoparticle synthesis templates due to their
controlled dimensions and structures, and the ability to confer
precisely spaced functionalities by genetic modification to
permit improved and/or tunable metal nanoparticle formation.
We have utilized tubular tobacco mosaic virus (TMV) as biolog-
ically derived nanotemplates for controlled synthesis of catalyt-
ically active Pd nanoparticles. Here we describe key findings
and insights gained from our studies on the synthesis and
characterization, as well as on both aqueous and organic
phase catalytic reactions, dichromate reduction and the
Suzuki-coupling reaction. We hope that the methodologies
and results summarized here can spur further interdisciplinary
collaboration, and expect that more discoveries and improved
performances can be realized with future endeavors.
1. Introduction
Palladium (Pd) represents one of the most versatile and utilized
noble metal elements in a vast array of catalytic reactions and
applications,[1] such as organic coupling reactions,[2] hydroge-
nation of unsaturated olefins,[3] and alcohol oxidation.[4] Al-
though homogeneous, ligand-associated forms of Pd catalysts
are still widely used, heterogeneous nanocatalysts on high sur-
face area supports are often preferred, owing to a number of
advantages, including stability, simple catalyst recovery, and
reuse. Yet, harsh and unpredictable catalyst synthesis condi-
tions often involving high temperature calcination steps may
pose limitations in developing a priori design principles, and/
or achieving efficient generation and utilization of high activity
catalytic surfaces, particularly in the case of colloidal nanoparti-
cle catalysts prepared with capping agents.[5]
on the nature of their genomes, various viruses can be group-
ed into three distinct shape categories from materials perspec-
tive: filamentous, icosahedral (spherical), and rigid rod-
shaped.[6] Viruses in each category offer unique traits, and the
applications and strategies for utilization have evolved around
them.[7]
First developed by Smith,[8] random peptide libraries of the
filamentous bacterial virus M13 offers an exciting and intrigu-
ing capability to screen for affinity to virtually any material of
interest in an iterative screening process termed “biopan-
ning”.[9] Many research groups including the Belcher group
have exploited this potent property for nanomaterial synthesis,
and several types of functional inorganic nanomaterials have
been synthesized on M13 viral templates, in some cases with
promising properties for fuel cells,[10] lithium ion batteries[11]
and photocatalytic applications.[12] Particularly, Knecht and co-
workers utilized this biopanning technique to identify a peptide
sequence (TSNAVHPTLRHL) with affinity for Pd, and synthe-
sized small and uniform (1.9Æ0.3 nm) Pd nanoparticles using
short peptides for Stille coupling reaction.[13] A recent report
by Belcher and co-workers reported substantially improved
catalytic performance of lithium-oxygen batteries with the ad-
dition of low content Pd.[14]
Historically regarded as harmful to humans, viruses have
gained much attention in the past three decades as novel syn-
thesis templates for metal, metal oxide, and other ceramic
nanomaterials.[6] This attention is attributable to several inher-
ent advantages, including highly controlled dimensions and
structures in the nanoscale, diverse chemical functionalities
from the rich repertoire of amino acids comprising the coat
proteins, and the ability to impart additional functionalities
with precise spacing and control via genetic modification.
Except for animal viruses that pose biological hazard to
humans and mammals, the vast majority of the viruses utilized
for material synthesis are either bacterial or plant, and are
deemed noninfectious once functionalized as their infection
mechanisms and delicate self-assembly structures are disrupt-
ed at the molecular level. Although typically classified based
Viruses belonging to the icosahedral shape category have
also been extensively examined; for example, a series of pio-
neering works by Douglas and Young have exploited the pre-
cise inner capsid size and the pH-responsive swelling proper-
ties of plant viruses for the synthesis of various materials, in-
cluding photocatalytically active titania nanoparticles.[15]
Some of the most important pioneers in the viral nanotech-
nology field are Kern and his coworkers, who have long uti-
lized tobacco mosaic virus (TMV), the central biotemplate of
this article, for the synthesis of a wide range of metal and inor-
ganic nanomaterials including 3 nm diameter copper wires.[16]
Our studies on TMV-templated Pd nanocatalysts are much in-
spired by their works,[17] as well as those of Culver and
[a] Dr. C. Yang, Prof. H. Yi
Department of Chemical and Biological Engineering
Tufts University, 4 Colby St. Medford, MA 02155 (USA)
E-mail: hyunmin.yi@tufts.edu
This publication is part of a Special Issue on Palladium Catalysis. To view
the complete issue, visit:
http://onlinelibrary.wiley.com/doi/10.1002/cctc.v7.14/issuetoc
ChemCatChem 2015, 7, 2015 – 2024
2015
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Minireviews
Harris,[18,19] as further described below. Our objective here is to
offer a focused discussion on the key findings and encouraging
results on the preparation, characterization, catalytic activity
and stability of TMV-templated Pd nanocatalysts, not to pro-
vide a comprehensive overview or survey of the now rich viral
and bioinspired nanobiotechnology field. As such, many inspir-
ing works from pioneers such as Stubbs and Mann, as well as
studies on other important materials for catalytic applications,
are regretfully omitted in this work. Elegant review articles in
the general field,[20] as well as in-depth molecular and mecha-
nistic descriptions of bioinspired Pd and other catalysts can be
found elsewhere.[21] In the following Section 2, we describe our
earlier synthesis and characterization efforts via small angle X-
ray scattering (SAXS) and other techniques. In Section 3, we
summarize results on the catalytic activity and stability studies
on dichromate reduction reaction, with particular attention to
the new insights we gained on the reaction mechanisms, syn-
thesis parameters and the role and utility of TMV templates for
improved synthesis and fabrication. In Section 4, we offer the
lessons learned from our study on Suzuki coupling reaction,
followed by our recent reports on the integration of catalytical-
ly active Pd-TMV nanocomplexes into shape-controlled poly-
meric microparticle platforms via rapid microfluidic and robust
replica molding schemes in Section 5.
2. Controlled Synthesis and Characterization
of Pd Nanoparticles Formed on TMV Tem-
plates
TMV is a biologically derived nanotube that offers many
unique properties and advantages for catalytic Pd nanoparticle
synthesis. As the first discovered virus, a naturally occurring
wild-type TMV consists of 2130 identical coat proteins
(16.7 kDa) helically wrapped around a single-strand 6.4kb
mRNA genome, making up a rigid 300 nm length, 18 nm diam-
eter nanotube with a 4 nm diameter inner channel.[22] Such
highly defined structure, safety (i.e. non-infectious to mammali-
an cells), extraordinary stability (e.g. ꢀ908C, pH 2–10, various
high concentration organic solvents, etc.), and simple mass
production are only a few of the advantages that make TMV
a potent nanotemplate that can be readily functionalized to
generate high capacity functional materials. Importantly, Culver
and colleagues genetically displayed cysteines (one buried in
wtTMV) near outer surfaces of TMV coat proteins (TMV1cys
and TMV2cys), as shown in the Chimera model of TMV in Fig-
ure 1(a), and reported improved metal precursor adsorption
and Pd nanoparticle formation.[19,23,24] In many of our earlier
works, we utilized enhanced surface-adsorption of TMV1cys on
gold substrates via gold–thiol binding for high density surface
assembly of TMV, followed by Pd nanoparticle synthesis by
simple reduction of Pd precursor sodium tetrachloropalladate
(Na2PdCl4) with mild reducing agents such as sodium hypo-
phosphite (NaPH2O2) as shown in the schematic diagram of
Figure 1b.[25]
Figure 1. Controlled synthesis of Pd nanoparticles on surface-assembled
TMV templates. a) Chimera model of a fraction of TMV1cys showing geneti-
cally displayed cysteines in red. b) Schematic diagram showing surface as-
sembly followed by Pd nanoparticle formation. c) AFM image of surface-as-
sembled TMVs. d) Pd nanoparticles preferentially formed along TMV tem-
plates. e) Schematic diagram of GISAXS setup. f) Pd nanoparticle sizes vs.
sodium hypophosphite concentrations. Adapted from Ref. [25]. Copyright
(2010) American Chemical Society.
2.1. Grazing Incidence Small-angle X-ray Scattering (GISAXS)
for Examination of Controlled Pd Nanoparticle Synthesis
As shown in the atomic force microscopy (AFM) image of Fig-
ure 1c, simple exposure of clean gold surfaces to aqueous
TMV solution leads to high density and preferential surface-as-
sembly of TMV rising from the thiol groups genetically dis-
played near the outer edge of each coat protein.[25] Next, high
ChemCatChem 2015, 7, 2015 – 2024
www.chemcatchem.org
2016
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Minireviews
density of Pd nanoparticles along the TMV’s nanotubular struc-
tures are readily achieved in a consistent and controlled
manner under mild aqueous conditions, as shown in Fig-
ure 1d. To overcome AFM’s inherent size overestimation and
to examine the influence of synthesis parameters on the parti-
cle size, we performed a series of grazing incidence small-
angle X-ray scattering (GISAXS) experiments at the Advanced
Photon Source (beamline 12 ID-C) of Argonne National Labora-
tory.[25] In a typical GISAXS setup (Figure 1e), planar or surface-
assembled samples are irradiated with a monochromatic X-ray
beam at a grazing angle (e.g. ai =0.18 in most of our stud-
ies),[25] the X-ray scattered at small angle recorded on a 2D de-
tector as a function of the scattering angle (2q in the horizon-
tal direction and af in the vertical direction). A rich pool of in-
formation ranging from the nanostructure morphology to pre-
cise dimensions can be extracted over a large sample area,
giving statistically meaningful information (see Supporting In-
formation for a short tutorial and Guinier analysis in Manocchi
et al.).[26] Upon thorough examination of several reducing
agents, we observed that sodium hypophosphite yields con-
trolled synthesis of Pd nanoparticles in 4–16 nm diameter
range by simply changing the reducer concentration as shown
in Figure 1 f, where higher concentrations of reducer result in
smaller Pd nanoparticle formation. In short summary, this first
study established a simple and controlled Pd nanoparticle syn-
thesis Scheme on TMV templates in a surface-assembled
format, along with in-depth characterization methods on parti-
cle size and nanostructures via the GISAXS technique.
2.2. High Thermal Stability of Pd-TMV Nanocomplexes
Figure 2. Thermostability of surface-assembled Pd-TMV complexes. GISAXS
images at lower (leftmost column), critical transition/collapse (middle) and
complete degradation (rightmost) temperatures of a) TMV, b) 4 nm Pd-TMV
and c) 15 nm Pd-TMV nanocomplexes. d) Normalized scattering intensity
plot showing degradation at varying temperatures. Adapted from Ref. [26].
Copyright (2010) American Chemical Society.
GISAXS is indeed a potent technique that allows rapid in situ
examination of nanoscale structures (typically 1–60 nm feature
sizes) and their dynamic behavior in an accurate, statistically
meaningful and non-destructive manner unlike TEM. In the
next set of studies, we exploited these traits to examine ther-
mal stability of the Pd nanoparticle-TMV complexes in an
in situ GISAXS format.[25] Briefly, various samples with surface-
assembled TMVs with or without small (4 nm) to large (16 nm)
Pd nanoparticles were mounted on a heater, and GISAXS pat-
terns were recorded and analyzed by means of Guinier analysis
upon programmed heating at 58CminÀ1. As illustrated in the
representative scattering patterns in Figure 2a, randomly
aligned TMVs on surfaces with their monodisperse nanotubular
dimensions generate strong scattering in the vertical (af) direc-
tion, with the characteristic oscillations typical of monodisperse
nanotubular objects.[26] As heating progresses, the loss of such
scattering pattern at 1978C indicates degradation of TMV’s
nanotubular structure. In comparison, Pd-TMV complexes yield
strong isotropic scattering in the horizontal direction that is
largely maintained upon heating, until sudden disappearance
occurs at relatively well-defined temperatures, as partially illus-
trated in Figure 2b and c. Importantly, further line-cut plot,
normalized scattering intensity and Guinier analyses all re-
vealed that the Pd-TMV nanostructures have substantially
higher thermal stability than TMV or Pd nanoparticles grown
on the plain gold substrates. Specifically, Pd-TMV complexes
with 4 nm Pd particles showed stability in their overall nano-
structure and nanoparticle density up to 2438C, while the
complexes with 15 nm Pd particles were stable up to 2798C.[26]
Along with the simultaneous degradation of TMV and Pd nano-
particles observed through these studies, we hypothesized
that the TMV and the Pd nanoparticles grown on them provide
synergistic enhancement in the thermal stability. Although the
inherent lack of chemical and thermal stability of biotemplates
are of concern for catalytic applications, the high thermal sta-
bility of Pd-TMV nanocomplexes observed in this GISAXS study
showed potential for our approach.
2.3. Fundamental Role of TMV Templates in Pd Nanoparticle
Formation via in situ SAXS
Further exploiting the capability of the SAXS technique to ex-
amine dynamic behavior of nanoscale objects, we next carried
out a series of in situ Pd nanoparticle formation experiments
with TMV templates in a bulk solution format, as shown in the
schematic diagram of Figure 3a.[27]
ChemCatChem 2015, 7, 2015 – 2024
www.chemcatchem.org
2017
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Minireviews
stemming from the genetically displayed cysteine residues. In
short summary, this in situ SAXS study revealed fundamental
roles of TMV as a potent nanotemplate for enhanced precursor
adsorption and preferential growth of Pd nanoparticles.
3. TMV-templated Pd Nanocatalysts for Dichro-
mate Reduction
Hexavalent chromium (CrVI, dichromate) is a prevalent industri-
al pollutant in waste sites and drinking water sources, and
known to possess carcinogenic and mutagenic properties.[28]
Catalytic reduction of dichromate to nontoxic chromate (CrIII)
represents a promising alternative for environmental remedia-
tion and water pollution control. We have utilized this simple
surface-originated liquid phase electron transfer reaction to ex-
amine the catalytic activity of Pd nanoparticles grown on TMV
templates in various formats.
3.1. Viral-templated Pd Nanocatalysts in Surface-assembled
Format
Our first report on the catalytic activity of TMV-templated Pd
nanoparticles utilized the simple surface-assembled format as
shown in Figure 4a. Briefly, TMV was first self-assembled on
the gold-coated substrates by simply dipping in aqueous
TMV1cys solution. These TMV chips are then immersed into
aqueous Pd precursor solution with mild reducing agent to
Figure 3. In situ SAXS-based examination of Pd nanoparticle formation on
TMV templates. a) Schematic diagram of the in situ SAXS setup. b) Represen-
tative SAXS line-cut plot of Pd nanoparticle formation on TMV templates.
c) Representative TEM image of the Pd-TMV complex formed under the con-
dition in b). Adapted from Ref. [27]. Copyright (2011) American Chemical So-
ciety.
Briefly, separate aqueous solutions containing the Pd precur-
sor (Na2PdCl4) and the reducer (NaPH2O2) with TMV were
mixed in a quartz capillary, and 2D SAXS images over two
minute periods were recorded and analyzed. The representa-
tive line-cut plot of scattering intensity [I(q)] over scattering
vector q in Figure 3b illustrates several important findings.
First, TMV’s highly monodisperse nanotubular structure coated
with electron-dense Pd nanoparticles gives rise to characteris-
tic oscillating patterns as indicated with three small arrows.
The scattering intensity continues to rise as Pd particles grow,
while the overall oscillation patterns do not change over time,
indicating the retention of TMV’s monodisperse structures. Par-
ticularly, the local minima continue shifting to the low q range
over time, indicating enlargement of the tubular diameter of
the Pd-TMV complexes as Pd particles continue to grow (i.e. in-
verse correlation between scattering object’s size and the scat-
tering angle in Bragg’s law).[27] Finally, the increasing scatting
intensity in low q region (dashed circle) clearly indicates that
particles are formed on TMV templates and not in the bulk so-
lution. This important observation is in good agreement with
the findings of Lim et al.,[24] who showed much enhanced ad-
sorption (two-fold) of Pd precursors on TMV1cys over wtTMV
Figure 4. Catalytic activity and reusability of Pd nanoparticles formed on sur-
face-assembled TMV templates for dichromate reduction reaction. a) Sche-
matic diagram of Pd-TMV chip fabrication. b) Schematic diagram of a batch
reaction in a quartz cuvette with a Pd-TMV chip and in situ monitoring via
UV-vis. c) Photographs of the reaction mixtures before and after the dichro-
mate reduction reaction. d) Conversion and apparent pseudo-first order rate
constants over three repeated batch reactions with one Pd-TMV chip. Adapt-
ed from Ref. [28] with permission from Elsevier.
ChemCatChem 2015, 7, 2015 – 2024
www.chemcatchem.org
2018
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Minireviews
form 11–12 nm Pd nanoparticles along TMV templates. As
shown in the schematic diagram of Figure 4b and the photo-
graphs of Figure 4c, this surface-assembly scheme allowed the
catalysts in a chip-based format to be directly dipped in the re-
action solution containing dichromate and the electron donor
formic acid. The catalytic reactions were then monitored in real
time via simple in situ UV/vis spectrophotometry by measuring
absorbance at dichromate’s characteristic absorption peak at
l=350 nm.[28] By simply taking the logarithm of the conver-
sion, the apparent pseudo-first order rate constant of dichro-
mate reduction with Pd-TMV catalysts was readily determined
to be 0.2518 minÀ1 (Figure 4d). This chip-based format also en-
abled simple catalyst recovery (i.e. taking out the chips from
the reaction mixtures) and reuse; 88% of Pd-TMV catalyst’s re-
activity was retained upon three repeated batch reactions
without any regeneration treatment, whereas Pd on gold sur-
face without TMV templates showed rapid decline (52%) in
the reaction rate (Figure 4d). AFM and GISAXS measurements
showed that the overall average size of Pd nanoparticles does
not change significantly upon three recycled reactions, which
further indicated the stability of the Pd-TMV complexes under
the reaction conditions in our study (e.g., pH 3.0).[28]
X-ray photoelectron spectroscopy (XPS) also showed no
meaningful changes in the chemical state with no significant
decomposition of the Pd-TMV complexes or leaching of Pd
nanoparticles into the reaction solution. Although we didn’t
examine turnover frequency (TOF) or other quantitative param-
eters in this first work (further examined in Section 3.3),[29] this
work demonstrates the dual functionality and utility of TMV
templates in enhanced surface assembly and catalytically
active Pd nanoparticle formation, as well as the stability to
maintain Pd catalysts well-dispersed through multiple reaction
cycles.
Figure 5. Tunable and spatially controlled formation of Pd nanocatalysts on
surface-assembled TMV Templates. a) Schematic diagram of the TMV con-
centration-based surface loading control, and Pd loading density and rate
constant plot vs. TMV surface density. b) Schematic diagram of patterned
surface area-based control of Pd-TMV complexes, and a rate constant plot
vs. surface area. Adapted from Ref. [30] with permission from Elsevier.
3.2. Tunable and Programmable Surface Assembly of Pd-
TMV Nanocomplexes
ed linearly with the gold surface area. This linear relationship
between the rate constant and the area of gold patterns clear-
ly indicates the simple spatial control and patterned synthesis
of Pd nanocatalysts permitted by TMV’s selective surface as-
sembly and preferential Pd nanoparticle formation as enabling
bioengineered templates.
Owing to the genetically displayed cysteines, TMV1cys can be
utilized for tunable and preferential surface assembly. In the
next study, we exploited this property to develop two simple
routes in tuning the surface loading density and location of
TMV-templated Pd nanocatalysts on gold chips that are other-
wise highly challenging in traditional microfabrication tech-
niques as shown in Figure 5.[30] In the first approach (Fig-
ure 5a), the TMV surface coverage density was tuned by
simply varying the TMV concentration for assembly onto gold
substrates. Pd nanoparticles were selectively formed along
TMV templates only, and the resulting total Pd surface loading
density (XPS) as well as the reaction rate constant for dichro-
mate reduction showed linear relationship with TMV surface
density.
3.3. Catalytic Reaction Kinetics and Tunable Catalyst Synthe-
sis
Recently, we further examined the dichromate reduction mech-
anism and synthesis–structure–activity relationship of TMV-
templated Pd nanocatalysts in this surface-assembled format
as shown in Figure 6.[29] It was found that the reaction kinetics
of dichromate reduction follows a Langmuir–Hinshelwood
mechanism (Figure 6a), which involves competing surface ad-
sorption of dichromate and formic acid on the active surface
of Pd nanoparticles. The adsorption of dichromate onto the Pd
nanocatalyst surface was shown to be substantially stronger
(ꢀ300 times) than that of formic acid. By simply tuning the
concentrations of the Pd precursor and reducer during the syn-
In the second approach (Figure 5b), gold-patterned chips
with varying gold surface areas were fabricated by a standard
photolithographic patterning technique for the spatial control
of Pd-TMV catalysts. As TMVs were preferentially assembled on
gold, but not on silicon nitride (AFM), the catalytic activity
from patterned chips with different sizes of gold areas correlat-
ChemCatChem 2015, 7, 2015 – 2024
www.chemcatchem.org
2019
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Minireviews
capping agents to maintain the dispersed states. In addition,
we found size-dependent behavior with Pd-TMV chip catalysts,
that is, catalysts with larger Pd particle size and lower Pd load-
ing density led to higher catalytic activity per unit Pd surface
area (Figure 6c).
4. TMV-templated Pd Nanocatalysts for Suzuki
Coupling Reaction
Owing to the mild reaction conditions and ready availability of
precursors, the Pd-catalyzed Suzuki–Miyaura CÀC cross-cou-
pling reactions are extensively utilized to synthesize biaryl
compounds (Figure 7a).[31] We enlisted our surface-assembled
Pd-TMV complexes in the chip-based format (Figure 4a) to ex-
amine the catalytic activity and a number of important param-
eters in this Suzuki reaction.[32] Specifically, we carried out sev-
eral sets of small scale batch reaction experiments primarily
with phenylboronic acid and iodoanisole (Figure 7a) to deter-
mine optimal reaction conditions, reaction yields, selectivity,
catalyst stability, and recyclability, as well as reaction mecha-
nism in the ligand-free semi-heterogeneous format. We were
able to achieve complete conversion to the desired cross-cou-
pling product methoxy biphenyl under mid conditions in a mix-
ture of acetonitrile and water at 508C. Our simple chip-based
format enabled simple catalyst separation and reuse (Fig-
ure 7b), facile product recovery and reaction mechanism stud-
ies.
In particular, the solvent ratio played an important role in
the selectivity of the Suzuki reaction, where a higher water/
acetonitrile ratio significantly facilitated the cross-coupling
pathway, and suppressed the homo-coupling reaction (Fig-
ure 7c), presumably due to the enhanced solubility and acces-
sibility of base to activate the iodoanisole and boronate spe-
cies required for the transmetalation step in the reaction
mechanism.[33,34] The reactant content ratio study (Figure 7d)
showed that higher aromatic boronic acid content than iodo-
anisole in the reactant mixture enhanced the completion of
the cross-coupling reaction, while a certain amount of aromatic
boronic acid followed the homocoupling pathway.[34] The Pd
loading applied for each batch reaction is 0.15 mol% based on
the based on aryl halide, which is lower than in other hetero-
geneous catalyst systems; for example, 0.3 mol% for Pd/C,[35]
2.5 mol% for Pd/zeolite,[36] and 0.5 mol% for Pd/Chitosan.[37]
Importantly, a simple chip removal experiments along with
ICP-OES analysis enabled us to further explore the reaction
mechanism (i.e. active catalyst species) for the Suzuki reaction;
the reaction continued to proceed to completion even after
the Pd-TMV chip was removed unlike the dichromate reduc-
tion that originates entirely from the competitive adsorption of
dichromate and formic acid onto the Pd surfaces[29] (see also
Supporting Information in Yang et al.).[32] We found that Suzuki
reactions were catalyzed largely by trace amounts of Pd spe-
cies (1–90 ppb) that are leached into the reaction solution, and
not on the solid catalyst surface, consistent with findings in
other studies.[38] In sum, our simple surface-assembled chip-
based format served as the pre-catalyst pool for stable, readily
recyclable and efficient catalyst platforms for Suzuki coupling
Figure 6. Catalytic reaction kinetics and tunable catalyst synthesis. a) Lang-
muir-Hinshelwood mechanism of dichromate reduction, and an initial rate
plot vs. qAqB (q represents surface coverage of reactants). b) Effect of synthe-
sis conditions on the Pd particle size and surface loading density. c) 3D bar
chart of catalytic reactivity vs. catalyst structure Pd size and loading density).
Adapted from Ref. [30] with permission from Elsevier.
thesis procedure, Pd nanoparticle size, catalyst loading density
and catalytic activity of viral templated Pd catalysts were readi-
ly controlled, as shown in Figure 6b. Increase in Pd precursor
concentration at the same reducer concentration resulted in
higher Pd loading density, whereas higher reducer concentra-
tion at the same Pd precursor concentration led to formation
of smaller Pd nanoparticles. Under optimized synthesis condi-
tion, TMV-templated Pd nanocatalysts exhibited 68% higher
catalytic activity per unit Pd mass in comparison with commer-
cially available 5% Pd/C catalysts, suggesting the advantage of
TMV templates in the synthesis of Pd nanocatalysts with pris-
tine surfaces for higher catalytic activity due to the absence of
ChemCatChem 2015, 7, 2015 – 2024
www.chemcatchem.org
2020
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Minireviews
5. Integration of Catalytically Active Pd-TMV
Nanocomplexes with Polymeric Microparticles
5.1. Rapid Microfluidic Fabrication of Catalytically Active Hy-
drogel Microparticles
Microfluidic techniques have gained substantial attention for
fabrication of microscale polymeric materials owing to several
advantages, including rapid processes, uniform dimensions of
the resulting materials and the ability to confer multiple func-
tionalities in spatially discrete domains.[39] In our first attempt
to capture the catalytic activity of Pd-TMV nanocomplexes syn-
thesized in aqueous solutions (Figure 3)[27] in readily deploy-
able polymeric hydrogel formats, we enlisted a simple two-
phase flow Scheme to manufacture polyethylene glycol (PEG)-
based microparticles containing Pd-TMV complexes in collabo-
ration with Patrick Doyle’s group, as shown in Figure 8.[40] As
shown in the schematic diagram of Figure 8a and the micro-
graph of Figure 8b, aqueous microflows containing the poly-
merizable PEG diacrylate and Pd-TMV complexes are continu-
ously split into droplets and cross-linked to form microparticles
via photoinduced radical polymerization of the acrylates.
Owing to the laminar nature of the microfluidic flows, one can
readily introduce multiple functionalities into single microparti-
cles, in our study Janus particles containing one domain with
Pd-TMV complexes for catalytic activity and the other with iron
oxide nanoparticles for simple magnetic separation (Figure 8c–
e). The loading density and catalytic activity of the Pd-TMV
complexes (comparable to or higher than surface-assembled
ones) encapsulated in the polymeric microparticles were also
readily tunable, as shown in the rate constant plot of dichro-
mate reduction in Figure 8 f.
In sum, a simple microfluidic fabrication was shown to be ef-
fective in capturing the catalytically active Pd-TMV complexes
in hydrogel microparticle formats where the reactants were
permitted ready access to the catalytic sites.
5.2. Integration of TMV-templated Small Pd Nanocatalysts
with Polymeric Microparticles via Robust Replica Molding
Our TMV-templated Pd nanocatalysts consistently showed
comparable or higher specific catalytic activities to standard
Pd/C catalysts due to the pristine, capping agent-free nature
despite relatively larger particle sizes (4–11 nm diameter) rising
from the use of sodium hypophosphite reducer.[29] Smaller par-
ticle sizes, however, offer inherent advantages of larger surface
area per unit mass of precious Pd metal. Inspired by Lim
et al.’s study on Pd coating of TMV templates in the absence
of external reducer,[41] we recently examined spontaneous for-
mation of Pd nanoparticles under elevated temperature and
observed consistent formation of small (1–2 nm), uniform and
highly crystalline Pd nanoparticles along TMV1cys.[42] We then
enlisted a simple and robust replica molding technique to en-
capsulate the Pd-TMV complexes in a stable and shape-en-
coded PEG-based polymeric microparticle format (Figure 9).
In the synthesis step 1 of our integrated synthesis-fabrica-
tion approach (Figure 9a), TMV was simply mixed with the Pd
Figure 7. Viral-templated Pd nanocatalysts for Suzuki-coupling reaction.
a) Scheme of Suzuki cross- and homo-coupling reactions catalyzed by nano-
structured Pd-TMV complexes. b) Recycling test of Pd-TMV chip-catalyzed
Suzuki reaction. c) Effect of solvent CH3CN/H2O volume ratio on reaction se-
lectivity. d) Effect of reactant ratios on product yields. Adapted from Ref. [32]
with permission from The Royal Society of Chemistry.
reaction that further enabled studies on a number of impor-
tant reaction parameters.
ChemCatChem 2015, 7, 2015 – 2024
www.chemcatchem.org
2021
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Minireviews
Figure 8. Microfluidic fabrication of polymeric hydrogel microparticles con-
taining Pd-TMV complexes. a) Schematic diagram of a two-phase flow-focus-
ing microfluidic device. b) Photomicrograph of a microfluidic channel with
two PEGDA solution-based prepolymer fluids containing Pd-TMV and mag-
netic nanoparticles in each. Scale bars represent 100 mm. c) PEG hydrogel
microparticles containing Pd-TMV complexes. d) Janus microparticles con-
taining Pd-TMV and magnetic nanoparticles in each domain. e) Photograph
showing magnetic separation of the Janus particles in (d). f) Apparent rate
constant plot of microparticles in c) vs. Pd nanoparticle loading amount.
Adapted from Ref. [40] Copyright (2010) American Chemical Society.
Figure 9. Integration of catalytically active Pd-TMV nanostructures into poly-
meric hydrogel microparticles via replica molding. a) Synthesis of Pd-TMV
complexes and encapsulation into PEG microparticles by Replica Molding.
b,c) TEM images of small Pd nanoparticles along TMV templates. d–f) Photo-
micrographs of the fabricated Pd-TMV-PEG microparticles. g) Plot of Pd load-
ing densities in Pd-TMV-PEG microparticles vs. apparent reaction rate con-
stants. Adapted from Ref. [42]. Copyright (2013) American Chemical Society.
precursor (Na2PdCl4) solution at 508C for 30 min to form the
Pd nanoparticles well-dispersed along TMV surface (Figure 9b)
with near-complete conversion without any reducer. The Pd
nanoparticles formed on TMV templates are highly crystalline
(Figure 9c), showing (111) facet of the face centered cubic
(FCC) structure of metallic Pd crystals. Intriguingly, the as-pre-
pared Pd-TMV complexes formed randomly networked TMV ar-
chitectures. This was attributed to the screening of repulsive
forces among negatively charged amino acids that confer the
rigidity to TMV, and led to readily recoverable (i.e. simple pre-
cipitation by spin-down) and stable (i.e. readily re-dispersed)
forms. In the fabrication step 2, the Pd-TMV complexes were
mixed with PEG diacrylate (PEGDA) and photoinitiator to yield
the preparticle solution then placed on PDMS micromolds.
Cross-linked PEG-based hydrogel microparticles containing the
Pd-TMV complexes were formed upon photo-induced poly-
merization with a simple hand-held UV lamp. As shown in Fig-
ure 9d–f, the photomicrographs of the Pd-TMV-PEG microparti-
cles have uniform color, shape and highly consistent dimen-
sions without any apparent aggregation or deformation.
Particularly, the Pd density inside microparticles can be read-
ily tuned in a range of 0.035–0.56 mg Pd/mL microparticles
simply by changing the loading density of the Pd-TMV com-
plexes in the preparticle solution. The fabricated Pd-TMV-PEG
microparticles were applied for dichromate reduction, and
showed 6-fold higher catalytic activity per Pd mass than the
commercial Pd/C catalyst. The apparent reaction rate constant
ChemCatChem 2015, 7, 2015 – 2024
www.chemcatchem.org
2022
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Minireviews
is proportional to the Pd loading density in the kinetic con-
trolled regime indicating ready access of the reactants to cata-
lyst sites through polymer networks (25% PEG), then levels off
in the higher Pd-TMV loading density cases as slight internal
diffusion limitation was present (Figure 9(g)). In addition, recy-
cle and long term storage (2 weeks) experiments showed mini-
mal decrease in catalytic activity including the retention of the
microparticle dimensions, clearly indicating the utility and sta-
bility of our polymeric hydrogel microparticle-based encapsula-
tion strategy.[42]
vide disassembled coat protein-based peptides for controlled
nanoparticle synthesis under certain conditions. We have in
fact had a recent success on the synthesis of small (ꢀ2 nm)
and uniform silver nanoparticles with high pH Tollen’s reagent
that showed high catalytic activity and stability for p-nitrophe-
nol reduction reaction.[43] Similar approaches as well as genetic
modifications to introduce other amino acids or peptides
could be envisioned for further improvement in the controlled
synthesis of other catalytic materials; ample information and
precedents exist in the literature on peptide-based and related
approaches.[44] Controlled synthesis of alloys, sub-nanometer
clusters and core-shell catalysts for improved selectivity, as well
as multicomponent systems for multistep, complex reaction
systems are only a few of several active areas in a broader per-
spective; of an important final note are the continuing endeav-
ors of the Belcher group for a wide range of application areas
in this regard.[14,45]
The high catalytic activity, stability and recyclability of the
Pd-TMV PEG microparticles as catalysts indicate retention of
the Pd-TMV complexes within the polymeric networks due to
the relative dimensions of Pd nanoparticles (1–2 nm), net-
worked Pd-TMV complexes (up to micrometers), hydrogel
mesh size (ꢀ2–3 nm), and small molecule reactants and prod-
ucts (Š’s). On one hand, the integration of 1–2 nm Pd nanopar-
ticles with TMV nanostructures (300 nm length) allows Pd
nanoparticles to remain well-dispersed on networked TMV su-
perstructures and entrapped inside the hydrogel networks
under catalytic reaction conditions. On the other hand, the 2–
3 nm mesh size of hydrogel allows minimal mass transfer limi-
tation of small molecule reactants and products through mi-
croparticles with the 50 mm characteristic diffusion length (i.e.
100 mm microparticle size) examined in this study. Hence, TMV
biotemplates enable the synthesis, dispersion and retention of
small Pd nanoparticles within the PEG hydrogel network struc-
ture. Furthermore, the microparticle size and mesh size can be
readily controlled and manipulated for optimal payload of Pd
nanoparticles and catalytic performance.
Acknowledgements
We would like to extend our special thanks to the collaborators
that offered inspiration and significant contributions to the works
described here; Dr. Byeongdu Lee and Dr. Soenke Seifert at Ar-
gonne National Laboratory, Professor Patrick S. Doyle at MIT, Pro-
fessor Changsoo Lee at Chungnam National University and Dr.
Chang-hyung Choi at Harvard University.
Keywords: nanobiotechnology · nanocatalysts · palladium ·
Suzuki coupling reaction · tobacco mosaic virus
[1] I. Saldan, Y. Semenyuk, I. Marchuk, O. Reshetnyak, J. Mater. Sci. 2015, 50,
2337–2354.
6. Conclusions and Future Perspectives
[2] N. Miyaura, A. Suzuki, Chem. Rev. 1995, 95, 2457–2483.
[3] a) . Molnµr, A. Sµrkµny, M. Varga, J. Mol. Catal. A 2001, 173, 185–221;
b) M. Studer, H.-U. Blaser, C. Exner, Adv. Synth. Catal. 2003, 345, 45–65.
[4] E. Antolini, J. Power Sources 2007, 170, 1–12.
[5] a) D. Astruc, F. Lu, J. R. Aranzaes, Angew. Chem. Int. Ed. 2005, 44, 7852–
7872; Angew. Chem. 2005, 117, 8062–8083.
[6] T. Douglas, M. Young, Science 2006, 312, 873–875.
[7] J. M. Rego, H. Yi, Supramolecular Chemistry, Wiley, Hoboken, 2012.
[8] G. Smith, Science 1985, 228, 1315–1317.
[9] J. M. Galloway, S. S. Staniland, J. Mater. Chem. 2012, 22, 12423–12434.
[10] Y. Lee, J. Kim, D. S. Yun, Y. S. Nam, Y. Shao-Horn, A. M. Belcher, Energy En-
viron. Sci. 2012, 5, 8328–8334.
[11] Y. J. Lee, H. Yi, W. J. Kim, K. Kang, D. S. Yun, M. S. Strano, G. Ceder, A. M.
Belcher, Science 2009, 324, 1051–1055.
In this report, we described key procedures and findings on
the synthesis, characterization and fabrication of TMV-templat-
ed Pd nanocatalysts in various formats. The robust and precise
structures, along with the genetically displayed cysteines of
TMV templates enabled efficient synthesis of Pd nanocatalysts
in the wide size ranges under mild aqueous conditions as well
as assembly into both surface-assembled and hydrogel-embed-
ded formats. Through surface-originated dichromate reduction
and leaching-based Suzuki coupling reaction, the as-prepared
Pd nanocatalysts showed promising activity, stability and reus-
ability.
Despite these encouraging results, there remain several as-
pects for further development and improvement in our viral-
templated Pd nanocatalysis approach. The inherent lack of sta-
bility of biological templates under harsh, oxidizing, strong
acidic (pH<2) or basic (pH>10), and/or higher temperature
reaction conditions despite TMV’s unusually robust and stable
structures could be addressed in a number of ways, while our
chip-based Pd-TMV complexes showed excellent stability for
a range of Suzuki coupling reaction conditions.[32] Simple coat-
ing with mesoporous silica and/or hollow carbon rods is cur-
rently under investigation. Since the TMV’s overall structure is
primarily comprised of non-covalent self-assembly of the coat
proteins, TMV could also serve as a sacrificial template to pro-
[12] Y. S. Nam, A. P. Magyar, D. Lee, J.-W. Kim, D. S. Yun, H. Park, T. S. Pollom,
D. A. Weitz, A. M. Belcher, Nat. Nanotechnol. 2010, 5, 340–344.
[13] D. B. Pacardo, M. Sethi, S. E. Jones, R. R. Naik, M. R. Knecht, ACS Nano
2009, 3, 1288–1296.
[14] D. Oh, J. Qi, Y.-C. Lu, Y. Zhang, Y. Shao-Horn, A. M. Belcher, Nat.
Commun. 2013, 4, 2756.
[15] M. T. Klem, M. Young, T. Douglas, J. Mater. Chem. 2008, 18, 3821–3823.
[16] S. Balci, A. M. Bittner, K. Hahn, C. Scheu, M. Knez, A. Kadri, C. Wege, H.
Jeske, K. Kern, Electrochim. Acta 2006, 51, 6251–6257.
[17] M. Knez, M. Sumser, A. M. Bittner, C. Wege, H. Jeske, T. P. Martin, K. Kern,
Adv. Funct. Mater. 2004, 14, 116–124.
[18] a) X. Chen, K. Gerasopoulos, J. Guo, A. Brown, C. Wang, R. Ghodssi, J. N.
Culver, ACS Nano 2010, 4, 5366–5372; b) E. Dujardin, C. Peet, G.
Stubbs, J. N. Culver, S. Mann, Nano Lett. 2003, 3, 413–417.
[19] J.-S. Lim, S.-M. Kim, S.-Y. Lee, E. A. Stach, J. N. Culver, M. T. Harris, J.
Nanomater. 2010, 2010, 1–7.
ChemCatChem 2015, 7, 2015 – 2024
www.chemcatchem.org
2023
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Minireviews
[20] a) J. M. Alonso, M. Ł. Górzny, A. M. Bittner, Trends Biotechnol. 2013, 31,
530–538; b) A. M. Bittner, J. M. Alonso, M. L. Gorzny, C. Wege, Subcell.
Biochem. 2013, 68, 667–702; c) Z. Liu, J. Qiao, Z. Niu, Q. Wang, Chem.
Soc. Rev. 2012, 41, 6178–6194;d) G. Lomonossoff, D. Evans, in Plant
Viral Vectors, Vol. 375 (Eds.: K. Palmer, Y. Gleba), Springer, Berlin/Heidel-
berg, 2014, pp. 61–87.
[21] R. Coppage, M. R. Knecht, in Bio-Inspired Nanotechnology (Eds.: M. R.
Knecht, T. R. Walsh), Springer, New York, 2014, pp. 173–219.
[22] K. Namba, G. Stubbs, Science 1986, 231, 1401–1406.
[34] H. Lakmini, I. Ciofini, A. Jutand, C. Amatore, C. Adamo, J. Phys. Chem. A
2008, 112, 12896–12903.
[35] H. Sakurai, T. Tsukuda, T. Hirao, J. Org. Chem. 2002, 67, 2721–2722.
[36] L. Artok, H. Bulut, Tetrahedron Lett. 2004, 45, 3881–3884.
[37] S. S. Yi, D. H. Lee, E. Sin, Y. S. Lee, Tetrahedron Lett. 2007, 48, 6771–6775.
[38] a) N. T. S. Phan, M. Van Der Sluys, C. W. Jones, Adv. Synth. Catal. 2006,
348, 609–679; b) A. Gaikwad, A. Holuigue, M. Thathagar, J. Elshof, G.
Rothenberg, Chem. Eur. J. 2007, 13, 6908–6913; c) J. P. Simeone, J. R.
Sowa Jr, Tetrahedron 2007, 63, 12646–12654.
[23] a) H. Yi, S. Nisar, S. Lee, M. A. Powers, W. E. Bentley, G. F. Payne, R.
Ghodssi, G. W. Rubloff, M. T. Harris, J. N. Culver, Nano Lett. 2005, 5,
1931–1936; b) S. Lee, E. Royston, J. N. Culver, M. T. Harris, Nanotechnolo-
gy 2005, 16, S435–S441.
[24] J.-S. Lim, S.-M. Kim, S.-Y. Lee, E. A. Stach, J. N. Culver, M. T. Harris, J. Col-
loid Interface Sci. 2010, 342, 455–461.
[39] D. Dendukuri, P. S. Doyle, Adv.Mater. 2009, 21, 4071–4086.
[40] C. L. Lewis, Y. Lin, C. Yang, A. K. Manocchi, K. P. Yuet, P. S. Doyle, H. Yi,
Langmuir 2010, 26, 13436–13441.
[41] a) J.-S. Lim, S.-M. Kim, S.-Y. Lee, E. A. Stach, J. N. Culver, M. T. Harris, Nano
Lett. 2010, 10, 3863–3867; b) J.-S. Lim, S.-M. Kim, S.-Y. Lee, E. A. Stach,
J. N. Culver, M. T. Harris, J. Colloid Interface Sci. 2011, 356, 31–36.
[42] C. Yang, C.-H. Choi, C.-S. Lee, H. Yi, ACS Nano 2013, 7, 5032–5044.
[43] C. Yang, S. Jung, H. Yi, Biochem. Eng. J. 2014, 89, 10–20.
[44] M. Sarikaya, C. Tamerler, A. K. Y. Jen, K. Schulten, F. Baneyx, Nat. Mater.
2003, 2, 577–585.
[25] A. K. Manocchi, N. E. Horelik, B. Lee, H. Yi, Langmuir 2010, 26, 3670–
3677.
[26] A. K. Manocchi, S. Seifert, B. Lee, H. Yi, Langmuir 2010, 26, 7516–7522.
[27] A. K. Manocchi, S. Seifert, B. Lee, H. Yi, Langmuir 2011, 27, 7052–7058.
[28] C. Yang, A. K. Manocchi, B. Lee, H. Yi, Appl. Catal. B 2010, 93, 282–291.
[29] C. Yang, J. H. Meldon, B. Lee, H. Yi, Catal. Today 2014, 233, 108–116.
[30] C. Yang, H. M. Yi, Biochem. Eng. J. 2010, 52, 160–167.
[45] D. Oh, J. Qi, B. Han, G. Zhang, T. J. Carney, J. Ohmura, Y. Zhang, Y. Shao-
Horn, A. M. Belcher, Nano Lett. 2014, 14, 4837–4845.
[31] A. Suzuki, J. Organomet. Chem. 1999, 576, 147–168.
Received: April 5, 2015
[32] C. Yang, A. K. Manocchi, B. Lee, H. Yi, J. Mater. Chem. 2011, 21, 187–194.
[33] C. Rçhlich, A. S. Wirth, K. Kçhler, Chem. Eur. J. 2012, 18, 15485–15494.
Published online on June 23, 2015
ChemCatChem 2015, 7, 2015 – 2024
www.chemcatchem.org
2024
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim