Improving the rates of Pd-catalyzed reactions by exciting the surface plasmons of AuPd bimetallic nanotriangles

Article abstract of DOI:10.1039/c7ra07264c

Gold nanoparticles exhibit unique optical properties due to surface plasmon oscillations when they interact with light. By utilizing their optical properties, the rates of many chemical reactions have been improved in the presence of visible light. The properties of plasmonic nanoparticles are highly tunable based on the size and shape of the nanoparticle. Here, we have used anisotropic AuPd bimetallic nanotriangles to improve the rates of Pd-catalyzed reactions in the presence of visible light. We synthesized AuPd core-shell bimetallic nanotriangles and performed Suzuki cross-coupling and hydrogenation reactions in light and dark conditions. Upon illuminating AuPd nanotriangles with an array of green LEDs (power ～ 500 mW), enhanced catalytic activity of palladium was observed. In order to understand the relative contributions of individual plasmonic effects, such as plasmonic hot electron transfer and plasmonic heating effects, the reaction temperatures were monitored, and careful control experiments were run at different temperatures. Our results indicated that the enhancement in the rate of these Pd-catalyzed reactions is primarily due to the plasmonic heating effect.

Full text of DOI:10.1039/c7ra07264c

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Improving the rates of Pd-catalyzed reactions by
exciting the surface plasmons of AuPd bimetallic
nanotriangles
Cite this: RSC Adv., 2017, 7, 40218
a
b
b
a
Mahesh K. Gangishetty, Adriana M. Fontes, Marcos Malta, Timothy L. Kelly*
a
and Robert W. J. Scott
*
Gold nanoparticles exhibit unique optical properties due to surface plasmon oscillations when they interact
with light. By utilizing their optical properties, the rates of many chemical reactions have been improved in
the presence of visible light. The properties of plasmonic nanoparticles are highly tunable based on the size
and shape of the nanoparticle. Here, we have used anisotropic AuPd bimetallic nanotriangles to improve the
rates of Pd-catalyzed reactions in the presence of visible light. We synthesized AuPd core–shell bimetallic
nanotriangles and performed Suzuki cross-coupling and hydrogenation reactions in light and dark
conditions. Upon illuminating AuPd nanotriangles with an array of green LEDs (power ꢀ 500 mW),
enhanced catalytic activity of palladium was observed. In order to understand the relative contributions
of individual plasmonic effects, such as plasmonic hot electron transfer and plasmonic heating effects,
the reaction temperatures were monitored, and careful control experiments were run at different
temperatures. Our results indicated that the enhancement in the rate of these Pd-catalyzed reactions is
primarily due to the plasmonic heating effect.
Received 30th June 2017
Accepted 9th August 2017
DOI: 10.1039/c7ra07264c
rsc.li/rsc-advances
including C–C bond forming reactions. Also, heterogeneous Pd
catalysts can be easily recovered from the reaction mixture and
Introduction
Plasmonic nanoparticles have been extensively studied as light reused for several reaction cycles. Recently, by using plasmonic
harvesting materials in various applications such as photovol- nanoparticles, the rates of Pd-catalyzed reactions such as cross-
22–24
1
–4
5
6
16,17
18–21
taics,
biotechnology, catalysis, and surface-enhanced coupling,
hydrogenation,
and oxidation reactions
7
Raman spectroscopy. In particular, employing their light have been improved, and it has been found that bimetallic
trapping properties to drive chemical transformations is an nanoparticles show superior activity compared to the pure
emerging eld. Upon interacting with certain frequencies of metallic nanoparticles.
6
25,26
The most studied bimetallic cata-
light, metal nanoparticles show unique optical properties due lysts in plasmon-enhanced Pd-catalyzed reactions are AuPd
to the collective oscillation of their conduction band electrons. bimetallic nanoparticles. Here, the Au/Pd ratio has been shown
This phenomenon is called localized surface plasmon reso- to be critical for both preserving the Au plasmonic properties as
26
nance. This phenomenon has also been observed in some metal well as the catalytic activity of Pd. In addition to this, the power
8
9
oxides such as WOx and Ce:In O . The optical characteristics of the incoming light also plays a key role in the plasmonic
2
3
17,26
of plasmonic nanoparticles are tunable based on the size and enhancement.
the shape of the nanoparticle. The energy accumulated in
10
It has been demonstrated that two plasmonic effects
these oscillating surface plasmons can be used in several ways contribute to the enhancement of Pd-catalytic activity in AuPd
to accelerate the rates of chemical reactions. This energy can be bimetallic nanoparticle systems: plasmonic local heating, and
6
,11
16,17,26
either directly used to drive a chemical reaction, or it can be plasmonic hot electron transfer.
transferred to an existing catalyst such as a semiconducting surface plasmons, they can decay non-radiatively via generation
metal oxide (e.g., TiO , CeO
) or a metal (e.g., Cu, Pd, and Ni) of hot electrons in the metal nanoparticles. The energy of these
Among these catalysts, enhancing the activity hot electrons is either redistributed among the ground state
of Pd nanoparticles using solar energy is attractive since Pd electrons through electron–electron scattering or dissipated in
Aer the excitation of
2
2
4,12–15
nanoparticle.
4,27,28
nanoparticles are known to catalyze many organic reactions, the form of heat by electron–phonon scattering.
The heat
generated by the metal nanoparticles can increase the temper-
ꢁ
6
ature at their surface up to 500 C. In a colloidal solution of
metal nanoparticles, this local heat dissipates into the solution
and increases the overall temperature of the bulk solution. This
plasmonic photothermal effect was found to accelerate the rates
a
Department of Chemistry, University of Saskatchewan, 110 Science Place, Saskatoon,
SK, S7N 5C9, Canada. E-mail: robert.scott@usask.ca; tim.kelly@usask.ca
b
Instituto de Qu ´ı mica, Universidade Federal da Bahia, Campus Ondina, Salvador, BA,
Brazil
40218 | RSC Adv., 2017, 7, 40218–40226
This journal is © The Royal Society of Chemistry 2017

View Article Online
Paper
RSC Advances
2
9
30
of chemical reactions when Au/ZnO, Au/Fe
tallic AuPd nanoparticles
2
O
3
,
and bime- Aldrich. Potassium tetrachloropalladate (99%), cetyl-
17,21,31,32
were employed as catalysts. In trimethylammonium chloride (CTAC, 96%) and all the boronic
addition to the plasmonic heating effect, some reports on AuPd acids used for Suzuki reactions were purchased from Alfa Aesar.
catalyzed reactions also observed a contribution from plas- Potassium iodide and sodium hydroxide were purchased from
monic hot electron transfer from Au to Pd in improving the Fisher Scientic. All chemicals were used directly without any
26,33
reaction rates.
It has been demonstrated that in AuPd further purication, and aqueous solutions were prepared
nanoparticles, the Fermi energy levels of Au and Pd are close to using Milli-Q water (18.2 MU cm).
34
each other (Pd is 5.0 eV, and Au is 4.7 eV). Upon exciting the
surface plasmons in Au, the hot electrons can easily be trans- Characterization
ferred from Au to Pd. As a result of this transfer, the number of
Transmission electron microscopy (TEM) images were collected
active catalytic sites on Pd increases, and this improves the
using a HT7700 TEM operating at 100 kV. TEM samples were
prepared by drop casting nanoparticle solutions onto a carbon-
coated 300 mesh Cu TEM grid (Electron Microscopy Sciences).
UV-vis spectra were measured using a Varian Cary 50 Bio UV-vis
16
overall rate of the reaction. Such hot electron transfer from Au
to Pd was described by recording a single particle photo-
luminescence (PL) spectra from an Au–Pd nanorod and an Au
35
nanorod. The authors observed a decrease in the PL intensity
for the AuPd nanorod compared to the Au nanorod, which was
attributed to hot electron transfer from Au and Pd. Further, by
1
spectrophotometer with an optical path length of 1 cm. H NMR
spectra were collected by using a Bruker Avance 500 MHz
spectrometer.
using a TiO spacer between Au to Pd in an AuPd nanoparticle,
2
Wang et al. controlled the electron transfer from Au to Pd. This
Methods
resulted in a decrease in the conversion of a Suzuki reaction
17
Synthesis of Au nanotriangles. Au nanotriangles were
compared to AuPd nanoparticles without any spacer. However,
the difference in conversion was very low, and it also depended
on the power of the incoming light. Another study, with the help
of DFT calculations, reported that hot electrons generated in
AuPd nanoparticles are directly transferred to the adsorbed
40
synthesized by following a literature procedure. Briey,
aqueous CTAC (16.0 mL, 0.10 M) was dissolved in 20.0 mL Milli-
Q water in a 250 mL Erlenmeyer ask. To this solution, aqueous
4
solutions of KI (0.75 mL, 0.1 M), HAuCl (0.80 mL, 25.4 mM) and
16
NaOH (0.20 mL, 0.10 M) were added sequentially. At each
addition, the solution was stirred manually by shaking the ask.
Aqueous ascorbic acid (1.0 mL, 0.064 M) was then added, and
then immediately, an aqueous NaOH (0.20 mL 0.10 M) solution
was rapidly injected into the reaction mixture, while simulta-
neously shaking the ask. Finally, the entire mixture was kept
undisturbed for 15 min until the solution became dark blue in
color. This dark blue solution was centrifuged at 5500 rpm
molecules and drive the chemical transformation.
All these plasmonic effects are highly dependent on the
power of the incoming light as well as on the morphology of the
metal nanoparticle. Various morphologies of Au and Ag nano-
1
7,35,36
32
21
particles such as nanorods,
and spherical nanoparticles
nanowheels, nanocages
1
6,20,26,37
have been used to improve
the catalytic activity of palladium. Here, we employed plas-
monic Au nanotriangles to harvest light in order to photo-
physically improve the rates of Pd-catalyzed reactions. Due to
the presence of sharp features, the nanotriangles are expected
(4194 g) for 30 min, and then the supernatant was decanted.
The Au nanotriangles were then redispersed in water.
38,39
Synthesis of AuPd nanotriangles. A concentrated Au nano-
triangle solution (15 mL, optical density ¼ 4.5 at 610 nm) was
dispersed in aqueous CTAC (30 mL, 0.010 M) in an Erlenmeyer
to show strong plasmonic effects compared to other shapes.
Also, due to the plate-like morphology of a nanotriangle, the
number of active sites (i.e., surface area) available for catalytic
reactions to take place is higher compared to a spherical
nanoparticle. By using AuPd bimetallic nanotriangles, we per-
formed Suzuki cross-coupling and hydrogenation reactions in
both dark and light conditions. The reactions performed in the
light showed higher conversions compared to the ones carried
out in the dark. The mechanism of this plasmonic enhance-
ment was investigated by performing several control experi-
ments. Our results indicate that the enhancements in the rates
of both Suzuki and hydrogenation reactions upon illumination
are mainly due to plasmonic heating effects. Further stability
studies show that the sharp-featured triangles are unstable, and
aer two cycles of Suzuki reactions their catalytic activity is
severely compromised.

2 4
ask. To this mixture, an aqueous K PdCl (2.0 mL, 0.010 M)
solution was added while stirring the solution, which was
maintained at 40 C using a hot water bath. Aer 10 min of
stirring, aqueous ascorbic acid (2.0 mL, 0.040 M) was added, the
ꢁ
ꢁ

ask was covered and the stirring was continued for 8 h at 40 C
by covering the ask with paralm. Finally, this solution was
centrifuged at 5500 rpm (4194 g) for 30 min, and the AuPd
bimetallic nanotriangles were collected by decanting the
supernatant. The AuPd nanotriangles were then dispersed in
water.
Catalytic reactions
Suzuki cross-coupling reactions. In a model cross-coupling
reaction, 0.5 mmol of p-iodobenzoic acid and 0.6 mmol of
phenylboronic acid were dissolved in a mixture of 8.0 mL of
Milli-Q water and 2 mL of ethanol. To this mixture, 2.0 mmol of
K CO dissolved in 2.0 mL of water was added. The solution was
Experimental section
Materials
2
3
Hydrogen tetrachloroaurate(III) (99%), 4-iodobenzoic acid, and mixed well to dissolve all of the reactants. Next, 2.0 mL of the
-methyl-3-buten-2-ol (MBE, 98%) were purchased from Sigma- as-synthesized AuPd nanotriangle solution (optical density ¼
2
This journal is © The Royal Society of Chemistry 2017
RSC Adv., 2017, 7, 40218–40226 | 40219

View Article Online
RSC Advances
Paper
42
4
.5 at 532 nm) was added. The nal optical density of the surface alloying between Au and Pd. The morphologies of the
nanotriangles in the resulting solution was ꢀ0.75 at 532 nm. Au and AuPd nanotriangles were conrmed by TEM analysis
This reaction was performed both with and without light illu- (Fig. 1b–d). Fig. 1b shows the presence of a large population of
mination using a home-made LED source (Luxeon), which Au nanotriangles; the average edge length of these nano-
consisted of 8 green LEDs which have a maximum emission triangles was found to be 43 ꢂ 4 nm. However, the sample also
wavelength of 530 nm and a emission range between 450 and contains ca. 10% of quasi-spherical Au nanoparticles along with
6
00 nm (power ꢀ 500 mW). At each hour, 200 mL aliquots of the the nanotriangles. Presence of multiple shapes broadens the
reaction mixture were collected, the reaction was quenched with absorption spectrum; which leads to a good match with the
00 mL of concentrated H SO , and the products extracted with incoming green LED spectrum. Fig. 1c shows a TEM image of
2
2
4
dichloromethane. The solvent was evaporated in a stream of air, the AuPd nanotriangles, and a higher resolution TEM image is
and the leover solids were dissolved in CDCl and character- shown in Fig. 1d. The Z-contrast difference in the TEM images
3
1
ized by H NMR. The rest of the cross-coupling reactions were of AuPd (Fig. 1c and d) indicates the spatial distribution of Au
performed in a similar way except for the number of moles of and Pd within the nanotriangle; the lighter regions are the Pd
K
2
CO
functional group, we used 2.0 mmol of K
CO was used otherwise. Also, the concentrated H
3
in the reaction mixture. For reactants containing an acid shell, and the darker region corresponds to the Au core. The Pd
CO ; 1.0 mmol of shells of the AuPd nanotriangles are polycrystalline, and thus,
SO was the resulting AuPd nanotriangles appear textured with unevenly
2
3
K
2
3
2
4
added only when the reactants contained carboxylic groups, distributed Au and Pd sites on their surface. A similar
otherwise the reaction mixture was directly extracted into morphology was observed in AuPd nanowheels due to the Pd
32
dichloromethane.
alloying with Au. The morphology likely occurs due to the
Hydrogenation of 2-methyl-3-buten-2-ol. In a 1 L round deposition of Pd at multiple sites on each particle, which could
bottom ask, 3.0 mL of AuPd nanotriangle solution (optical be due to defects in the CTAC coating on the surface of the Au
43
density 0.75 at 532 nm) was added to 6 mL Milli-Q water. To this nanotriangles. For catalytic purposes, the polycrystalline
mixture, 150 mL of methanol was added and then the ask was nature of the Pd shell may be favorable as the available surface
sealed tight using a septum. The ask was then purged with area of the Pd catalyst would be substantially higher than
hydrogen gas for 20 minutes; the ask was then lled with H
gas until the ask reached an initial differential pressure of
.370 atm. The pressure in the ask was measured by a differ-
ential pressure manometer (Extech Instrument, Model
2
a perfect epitaxial core–shell structure.
0
Plasmon enhanced Suzuki-coupling reactions
#407910), and for all hydrogenation reactions the same initial Aer conrming the morphology of the AuPd nanotriangles,
pressure was maintained. The AuPd nanotriangle solution was they were used to catalyze Suzuki cross-coupling reactions. The
then stirred under hydrogen for 30 minutes. Then 2-methyl-3- reactions were performed both with and without light illumi-
buten-2-ol (0.10 mL) was injected into the ask. Similar to the nation. For the dark reactions, the reaction asks were covered
cross-coupling reactions, hydrogenation reactions were per- with aluminum foil; for the light reactions, the asks were
1
formed in both light and dark conditions. For the H NMR illuminated with a green LED (power ꢀ 500 mW, l
¼ 530 nm)
max
analysis, an aliquot of 100 mL of the reaction mixture was light source (Fig. 2a). The green light source was chosen to
collected every 20 minutes and was directly extracted in 500 mL match the wavelength (534 nm) of the AuPd plasmon band
CDCl
3
.
(Fig. 1a). At this wavelength, the surface plasmons of both the
triangular and spherical AuPd nanoparticles will be excited. The
light source setup is shown in Fig. 2a; it contains an array of 8
green LEDs mounted on each heat sink, and four of these heat
Results and discussion
AuPd bimetallic nanotriangles
sinks are assembled in such a way that a 25 mL ask can be
ꢁ
Au nanotriangles were synthesized via a literature procedure, placed in the center and receive 360 illumination.
which is a competitive process involving both reduction/
To investigate the plasmonic effect of AuPd nanotriangles on
oxidative etching of Au in the presence of CTAC as a stabi- the Suzuki cross-coupling reaction, we chose to study a model
40
lizing agent. Here, ascorbic acid was used as the reducing reaction between p-iodobenzoic acid and phenylboronic acid, as
agent and potassium iodide as an oxidative etchant in the shown in Scheme 1. This reaction was performed in both dark
presence of air. To synthesize the AuPd bimetallic nano- and light illumination, and the reaction kinetics were moni-
1
triangles, K
2
PdCl
4
was reduced onto the as-synthesized Au tored by H NMR. The percent conversions were calculated from
1
nanotriangles using ascorbic acid as the reducing agent. These the H NMR spectra, and are plotted against time (Fig. 2b). The
Au and AuPd nanotriangles were then characterized by UV-vis cross-coupling reaction performed in the light showed signi-
spectroscopy and TEM; the results are shown in Fig. 1. In the cantly higher conversions compared to the dark reactions. In
UV-vis spectrum, the Au nanotriangles show a broad plasmon both cases, only hetero-cross-coupling products were seen; i.e.
peak at 605 nm with a shoulder at 534 nm. The plasmon peak at no homocoupling products were observed. For example, aer
41
6
05 nm is attributed to the in-plane dipole mode, while the 1 h of reaction time, the light reaction had proceeded to ꢀ40%
shoulder at 534 nm is likely due to the presence of spherical Au conversion, whereas the dark reaction had proceeded to only
2,40
nanoparticles. Upon coating with Pd, a ca. 60 nm blue shi in ꢀ23% conversion. With an increase in time, the difference in
the plasmon peak was observed. This might be due to some conversion between the light and dark reactions was more
40220 | RSC Adv., 2017, 7, 40218–40226
This journal is © The Royal Society of Chemistry 2017

View Article Online
Paper
RSC Advances
Fig. 1 (a) Normalized UV-vis spectra of Au and AuPd bimetallic nanotriangles; the spectra are normalized to the maximum of the plasmon band.
TEM images of (b) as-synthesized Au nanotriangles, (c) AuPd bimetallic nanotriangles and (d) magnified image of AuPd nanotriangles.
1
Fig. 2 (a) The LED light source with an array of green LEDs mounted on heat sinks, (b) percent conversions calculated from H NMR are plotted
against time; the reaction is between p-iodobenzoic acid and phenylboronic acid. The green triangles indicate the control dark reaction per-
ꢁ
formed in a hot water bath and the inverted blue triangles indicate the light reaction performed in a 25 C water bath. The solid lines are a best-fit
using a first-order kinetic model.
This journal is © The Royal Society of Chemistry 2017
RSC Adv., 2017, 7, 40218–40226 | 40221

View Article Online
RSC Advances
Paper
45,46
highly dependent on the power of the excitation light source.
The low power LEDs used here are likely insufficient to produce
a signicant population of hot electrons; hence, the contribu-
Scheme 1 The model Suzuki reaction using AuPd nanotriangles as tion of these hot electrons in improving the overall rate of the
catalysts.
reaction is low. Nonetheless, the dark control reactions under-
performed the light reactions even when both were run at the
same temperature. It is therefore possible that there is a minor
pronounced. Aer 5 h of the reaction time, the light reaction contribution from plasmonic hot electron transfer to the overall
showed ca. 35% higher conversion compared to the dark. At the enhancement.
same time, the temperature of both the reaction mixtures was
We extended our studies to examine the effect of different aryl
measured. We noticed a rise in the nal temperature of the light halides and different boronic acids. The results for various
ꢁ
reaction from 25 C (the initial temperature of the reaction substrates are shown in Table 1. Selectivity for the heterocoupling
ꢁ

ask) to 37 C, while the dark reaction showed no change in product was 100% in all cases (though, for entry 2 and 3 both
temperature. To determine if this increase in temperature was reactions would have the same product). In all cases, the reac-
due to the presence of AuPd nanotriangles or due to light- tions performed in the light showed an average of ꢀ20% higher
induced heating of the solvent, we illuminated water with and conversion compared to those carried out in the dark. A large
without AuPd nanotriangles. Aer illuminating both the asks difference in the conversions between the light and dark reac-
ꢁ
ꢁ
for 1 h, an increase in temperature from 25 C to 37 C for the tions was observed when aromatic iodides were used as

ask containing AuPd nanotriangles was observed, whereas the substrates. This is particularly seen for entry 4, where a nitro-
ask containing only water showed only a 2 C (light-induced substituted aryl iodide (p-iodonitrobenzene) was reacted with
ꢁ

heating) rise in temperature. This clearly conrms that the phenylboronic acid. This difference was slightly higher even
AuPd nanotriangles are the cause of the temperature increase; when compared to the other reactions where other aryl iodides
this is due to the plasmonic local heating generated by non- were reacted with the same boronic acid (entries 1 and 2). The
31
radiative plasmon decay (electron–phonon scattering).
light reaction in entry 4 showed complete conversion, and the
To determine how signicant a role this plasmonic heating dark reaction showed a decreased conversion compared to
effect has on the overall enhancement of the reaction rate, we entries 1 and 2. The presence of an electron-withdrawing group
performed a control reaction in the dark using a hot water bath to such as –NO
maintain the temperature of the reaction mixture at 37 C (the oxidative addition (typically the rate determining step) in Suzuki
on an aryl halide generally improves the rate of
2
ꢁ
same temperature reached by the light reaction ask). The reac- reactions. It is possible that the further acceleration of the reac-
1
tion progress was monitored hourly by H NMR, and the tion in the light could be coming from plasmonic hot electron
conversion plotted in Fig. 2b. The light reaction still showed transfer. However, the exact reason why the nitro-group shows
ꢁ
higher conversions compared to the dark reaction run at 37 C, a positive effect on the light reaction and a negative effect on the
although the nal temperatures of both the reaction mixtures dark reaction is unclear. On the other hand, the difference in
ꢁ
were same (37 C). However, the difference in conversion is only conversions was lower when the same halide (p-iodoni-
9
% aer 5 h of the reaction time; this is much lower compared to trobenzene) was reacted with methyl-substituted phenylboronic
the previous case (35%). From these results, it is clear that the acid (entry 7); the difference in conversion between light and dark
rate of reaction is accelerated by the excitation of the AuPd was very low. Here, due to the presence of an electron-donating
surface plasmon modes. The effect of plasmonic excitation is also methyl group on the boronic acid, both the light and dark reac-
similar to that of conventional heating in improving the overall tions showed almost complete conversions. Interestingly, in the
rate. To further support these conclusions, another Suzuki reac- case of entry 5, where 2-thiopheneboronic acid was used, the
tion was performed in which the temperature of the light reaction reaction showed almost double the conversion in the presence of
ꢁ

ask was xed at 25 C using a water bath. Aer 5 h, this reaction light as opposed to the dark. Generally, sulfur-containing
showed signicantly higher conversion (ca. 20%) than the dark (thiophene-based) substrates have been shown to poison the Pd
ꢁ
47
reaction. However, the nal temperature increased to 29 C (even catalyst and cause low yields in cross-coupling reactions. They
ꢁ
though the reaction ask was kept in a 25 C water bath). This oen need either high catalyst loadings or elevated temperatures
slight rise in temperature may be due to the direct photothermal to achieve good yields. In such reactions, the use of plasmonic
heating by both solvent and the water bath, or simply an insuf- nanoparticles might be a promising way of improving the overall
ciently large thermal mass in the cooling bath. A number of conversions, while keeping the catalyst loading to a minimum.
other papers have outlined a similar role of plasmonic heating for Finally, among all of the reactions, the aryl bromide substrate
21,29,30,44,45
enhancing the rate of a variety of catalytic reactions.
(entry 3) showed the lowest overall conversion, regardless of
However, in addition to the plasmonic heating effect, many whether or not the reaction was illuminated. This can be attrib-
reports have observed that plasmonic hot electron transfer uted to the poor leaving ability of Br as compared to I.
16,17
contributes to the rate of a cross-coupling reaction.
In many
of these reports, a high power laser source was used to excite the
Au nanoparticle surface plasmons, whereas, in the current
Catalyst stability and recyclability
study, we used a relatively low power (ꢀ500 mW) and unfocused The main advantage of employing nanoparticles in catalysis is
LED light source. The generation of plasmonic hot electrons is that the catalyst can easily be separated from the reaction
40222 | RSC Adv., 2017, 7, 40218–40226
This journal is © The Royal Society of Chemistry 2017

View Article Online
Paper
Table 1 Percent conversions measured by H NMR for different Suzuki reactions performed in both the dark and the light
RSC Advances
1
a
Entry
1
Aryl halide
Boronic acid
Time
3 h
Conversion (light)
92%
Conversion (dark)
73%
2
3
4
3 h
3 h
3 h
85%
34%
100%
71%
30%
63%
5
6
2 h
57%
25%
2 h
2 h
100%
91%
89%
7
100%
a
ꢁ
The substrate to catalyst loading ratio for all reactions was 2.5 : 0.002. The initial reaction temperature was 25 C, and the solvent was a 80 : 20
mixture of water and ethanol.
mixture and re-used for several reaction cycles. To test the
Using the same reactants that were used in the model Suzuki
recyclability of the AuPd nanotriangles, we rst characterized reaction between p-iodobenzoic acid and phenylboronic acid,
them aer a Suzuki coupling reaction between p-iodobenzoic we studied the catalyst recyclability over three reaction cycles in
acid and phenylboronic acid. Fig. 3a shows the UV-vis spectra of the presence of light. Aer each cycle, the AuPd nanotriangles
the AuPd nanotriangles before and aer the Suzuki reaction. were puried by centrifugation and then reused by redispersing
The spectra were recorded for the AuPd nanotriangles used in them in the same amount of the reaction mixture. Aer 3 h of
1
the model cross-coupling reactions shown in Fig. 2b. Aer reaction, the percent conversion was calculated from the
H
completion of the reaction, the AuPd nanotriangles were NMR spectrum, and the data is plotted against cycle number
collected by centrifuging the reaction mixture, and then redis- (Fig. 3b). In cycle 1, aer 3 h of reaction, ca. 70% conversion was
persed in a volume of water equivalent to the initial volume of observed, and the nal temperature of the reaction ask was
ꢁ
ꢁ
the reaction mixture. A slight decrease in the intensity of the 37 C. In cycle 2, the nal temperature decreased slightly to
plasmon peak at 532 nm was observed aer the reaction 35 C, and the percent conversion was reduced to ca. 50%. The
(Fig. 3a). This suggests a slight decrease in the population of decrease in conversion may be due to the decrease in the
AuPd nanotriangles aer the reaction. However, an increase in number of available active sites, since the nanoparticles are
rd
the optical density was observed in the wavelength range from aggregated aer the reaction (Fig. 3c). In the 3 cycle, the
2
00–300 nm aer one reaction cycle. This might be due to the reaction yielded only 1% conversion and a nal temperature of
ꢁ
Rayleigh scattering from large aggregates in the reaction 31 C; this could be due to almost complete aggregation of the
mixture. From TEM analysis (Fig. 3c), aggregation of some of nanotriangles as the color of the solution substantially became
the nanotriangles was evidenced aer one reaction cycle. As lighter aer redispersing the nanotriangles for the 3 reaction
nd
a result of the aggregation, the nanotriangles tended to settle at cycle. However, stability and hence catalytic recyclability of
the bottom of the reaction ask upon keeping the ask undis- these nanoparticles may be further improved by either inte-
48,49
turbed for several days. We are uncertain as to why aggregation grating them in porous metal oxide matrix
might be occurring in this system, but we believe it is due to the them with thin shells.
loss of their surface stabilizer (CTAC) during the Suzuki
or by coating
50
coupling reaction. A control experiment showed that the
aggregation is not light-induced, and no such aggregation was
seen for the hydrogenation system discussed below.
Plasmon enhanced hydrogenation reaction
To determine if the plasmonic properties of AuPd nanotriangles
can be used to drive other Pd-catalyzed reactions, we studied the
This journal is © The Royal Society of Chemistry 2017
RSC Adv., 2017, 7, 40218–40226 | 40223

View Article Online
RSC Advances
Paper
Fig. 3 (a) UV-vis absorption spectra of AuPd nanotriangles before and after cross-coupling reactions performed between p-iodobenzoic acid
and phenylboronic acid. (b) Recyclability of AuPd nanotriangles and final temperature of the same coupling reaction performed under light
illumination, and (c) TEM image showing aggregation of the AuPd nanotriangles, recorded after one cycle of the light reaction.
hydrogenation of 2-methyl-3-buten-2-ol (MBE) using the nano- control dark reaction (Fig. 2b). The fact that similar conversions
ꢁ
triangles as the catalyst. The reactions were performed in both are observed in both the reactions (light and dark at 30 C)
the presence and absence of light illumination. We used the clearly suggests that the enhancement is due to plasmonic
same LED source (Fig. 2a) to irradiate the reaction mixture in heating effects alone. If other plasmonic effects contributed
light reaction. Here, we used a larger (1 L) round-bottom ask to signicantly to the enhancement, the light reaction would show
maintain an excess of H
The AuPd nanotriangles were added to the ask, and an excess
of H gas was then supplied. Before injecting the substrate
MBE) into the ask, the ask was stirred to homogenize the
2
gas for the hydrogenation reaction. higher conversions compared to the high-temperature dark
2
(
solution. At 20 min intervals, the reaction was sampled and
1
analyzed by H NMR, and the percent conversions are plotted
against time (Fig. 4). At each time interval, the light reaction
showed signicantly higher conversions compared to the dark
reaction. For example, aer 20 min, the light reaction had
proceeded to 42% conversion, whereas the dark reaction
showed only 18% conversion. The difference became more
pronounced as the reaction proceeded. Similar to the plasmon
enhanced Suzuki reactions, an increase in the temperature of
ꢁ
ꢁ
the light reaction (from 23 C to 30 C) was observed. However,
the magnitude of the temperature increase was not as signi-
cant as in the Suzuki reactions. This is most likely due to the
slightly lower AuPd catalyst loading, which was nearly 4 times
lower compared to the Suzuki reaction. Again, to understand
the role of plasmonic heating in the enhancement of the
hydrogenation reaction, a control dark reaction was performed
ꢁ
Fig. 4 Percent conversion of hydrogenation reactions performed on
-methyl-3-buten-2-ol using AuPd nanotriangles in the presence and
by heating the ask to 30 C in a conventional hot water bath
2
(i.e., the same temperature reached by the light reaction). Aer
absence of light. The black squares show the reaction performed at
120 min of reaction time, the control dark reaction showed
ꢁ
23
C under ambient conditions, the red circles show the reaction
almost same conversion as the light reaction (Fig. 4). This is in performed under illumination that achieved a final temperature of
ꢁ
contrast to the Suzuki reactions, where the light reaction 30 C, and the green triangles show the dark reaction performed at
ꢁ
3
0 C using hot water bath. The solid lines are a best-fit using a first-
showed slightly higher conversions (ꢀ9%) compared to the
order kinetic model.
40224 | RSC Adv., 2017, 7, 40218–40226
This journal is © The Royal Society of Chemistry 2017

View Article Online
Paper
RSC Advances
21,51
reaction. Similar results have been reported in the literature;
Research Council of Canada (NSERC) and the University of
Long et al., observed the contribution of only plasmonic heating Saskatchewan are acknowledged for nancial support.
51
effect in the hydrogenation of styrene. Since it has been shown
that Au nanoparticles alone can catalyze the hydrogenation
References
46,52
reaction (although it requires high temperatures),
we also
attempted to perform the same hydrogenation reaction using
pure Au nanotriangles as the catalyst. No product formation was
observed in either the light or dark conditions. This conrmed
that the Pd is the active site on the AuPd catalyst for the
hydrogenation reactions. Another possible mechanism to drive
the hydrogenation reactions is by plasmonic hot electron
transfer; however, to achieve the dissociation of H2 by hot
electrons generated on Au, a high power incoming light source
1 M. K. Gangishetty, K. E. Lee, R. W. J. Scott and T. L. Kelly, ACS
Appl. Mater. Interfaces, 2013, 5, 11044–11051.
2 M. K. Gangishetty, R. W. J. Scott and T. L. Kelly, Langmuir,
2014, 30, 14352–14359.
3 H. A. Atwater and A. Polman, Nat. Mater., 2010, 9, 205–213.
4 C. Clavero, Nat. Photonics, 2014, 8, 95–103.
5 J. N. Anker, W. P. Hall, O. Lyandres, N. C. Shah, J. Zhao and
R. P. Van Duyne, Nat. Mater., 2008, 7, 442–453.
6 S. Linic, U. Aslam, C. Boerigter and M. Morabito, Nat. Mater.,
2015, 14, 567–576.
7 L. Guerrini and D. Graham, Chem. Soc. Rev., 2012, 41, 7085–
7107.
8 Z. Lou, Q. Gu, Y. Liao, S. Yu and C. Xue, Appl. Catal., B, 2016,
184, 258–263.
9 E. L. Runnerstrom, A. Bergerud, A. Agrawal, R. W. Johns,
C. J. Dahlman, A. Singh, S. M. Selbach and D. J. Milliron,
Nano Lett., 2016, 16, 3390–3398.
ꢃ2
46
(on order of W cm ) is required. The LED light source used in
the current study was likely not intense enough to produce
sufficiently hot electrons, and no activity of Au was seen for this
hydrogenation reaction. This result agrees with earlier cross-
coupling and hydrogenation results that showed that photo-
thermal heating is mostly responsible for the catalytic
enhancement. Again, we note that this is a consequence of the
low power LED source used in this study.
1
1
1
1
1
1
1
1
0 K. L. Kelly, E. Coronado, L. L. Zhao and G. C. Schatz, J. Phys.
Chem. C, 2002, 107, 668–677.
1 M. J. Kale, T. Avanesian and P. Christopher, ACS Catal., 2014,
Conclusions
In conclusion, we have synthesized triangular AuPd nano-
particles and employed them to improve the rates of Suzuki
coupling and hydrogenation reactions. Upon exciting the
surface plasmons in AuPd nanotriangles using green LEDs
4
, 116–128.
2 P. Wang, B. Huang, Y. Dai and M.-H. Whangbo, Phys. Chem.
Chem. Phys., 2012, 14, 9813–9825.
3 R. Jiang, B. Li, C. Fang and J. Wang, Adv. Mater., 2014, 26,
(whose wavelength was near the maximum of the plasmon
5274–5309.
band), an improvement in the rates of Suzuki and hydrogena-
tion reactions was observed. Our results revealed that the
enhancement in the rate of reaction is mainly due to the
contribution of plasmonic heating effects, with at most a minor
contribution from plasmonic hot electron transfer. We have
investigated the recyclability of these catalysts; we found that
the catalyst survived for 2 cycles of Suzuki reactions. TEM
showed aggregation of the AuPd nanotriangles causing
a reduction in the catalytic activity. This study clearly demon-
strates that a plasmonic enhancement in the rates of Pd-
catalyzed reactions can be observed even using a low power
and unfocused LED light source, and that, under these low
powers, there is only a plasmonic photothermal effect contrib-
uting to the rate enhancement.
4 J. Li, S. K. Cushing, J. Bright, F. Meng, T. R. Senty, P. Zheng,
A. D. Bristow and N. Wu, ACS Catal., 2013, 3, 47–51.
5 L. Liu, S. Ouyang and J. Ye, Angew. Chem., Int. Ed., 2013, 125,
6821–6825.
6 Q. Xiao, S. Sarina, A. Bo, J. Jia, H. Liu, D. P. Arnold, Y. Huang,
H. Wu and H. Zhu, ACS Catal., 2014, 4, 1725–1734.
7 F. Wang, C. Li, H. Chen, R. Jiang, L.-D. Sun, Q. Li, J. Wang,
J. C. Yu and C.-H. Yan, J. Am. Chem. Soc., 2013, 135, 5588–
5601.
1
8 Q. Xiao, S. Sarina, E. R. Waclawik, J. Jia, J. Chang,
J. D. Riches, H. Wu, Z. Zheng and H. Zhu, ACS Catal.,
2016, 6, 1744–1753.
1
9 L.-B. Zhao, M. Zhang, Y.-F. Huang, C. T. Williams, D.-Y. Wu,
B. Ren and Z.-Q. Tian, J. Phys. Chem. Lett., 2014, 5, 1259–
1
266.
20 Z. Liu, Y. Huang, Q. Xiao and H. Zhu, Green Chem., 2016, 18,
17–825.
Conflicts of interest
8
There are no conicts to declare.
21 X. Zhao, R. Long, D. Liu, B. Luo and Y. Xiong, J. Mater. Chem.
A, 2015, 3, 9390–9394.
2
2 J. Cui, Y. Li, L. Liu, L. Chen, J. Xu, J. Ma, G. Fang, E. Zhu,
H. Wu, L. Zhao, L. Wang and Y. Huang, Nano Lett., 2015,
15, 6295–6301.
Acknowledgements
T. L. K. is a Canada Research Chair in Photovoltaics. This
research was undertaken, in part, thanks to funding from the 23 Q. Xiao, Z. Liu, A. Bo, S. Zavahir, S. Sarina, S. Bottle,
Canada Research Chairs program. A. M. F. thanks the Coordi-
nation for the Improvement of Higher Level Personnel-CAPES
J. D. Riches and H. Zhu, J. Am. Chem. Soc., 2015, 137,
1956–1966.
for
a
scholarship (doctoral sandwich grant, PDSE 24 T. Jiang, C. Jia, L. Zhang, S. He, Y. Sang, H. Li, Y. Li, X. Xu and
H. Liu, Nanoscale, 2015, 7, 209–217.
99999.009761/2014-04). The Natural Sciences and Engineering
This journal is © The Royal Society of Chemistry 2017
RSC Adv., 2017, 7, 40218–40226 | 40225

View Article Online
RSC Advances
Paper
2
2
2
2
2
5 S. Linic, P. Christopher and D. B. Ingram, Nat. Mater., 2011, 39 K. L. Kelly, E. Coronado, L. L. Zhao and G. C. Schatz, J. Phys.
0, 911–921. Chem. C, 2003, 107, 668–677.
6 S. Sarina, H. Zhu, E. Jaatinen, Q. Xiao, H. Liu, J. Jia, C. Chen 40 L. Chen, F. Ji, Y. Xu, L. He, Y. Mi, F. Bao, B. Sun, X. Zhang and
and J. Zhao, J. Am. Chem. Soc., 2013, 135, 5793–5801. Q. Zhang, Nano Lett., 2014, 14, 7201–7206.
7 M. L. Brongersma, N. J. Halas and P. Nordlander, Nat. 41 J. E. Millstone, S. Park, K. L. Shuford, L. Qin, G. C. Schatz and
Nanotechnol., 2015, 10, 25–34. C. A. Mirkin, J. Am. Chem. Soc., 2005, 127, 5312–5313.
8 W. R. Erwin, H. F. Zarick, E. M. Talbert and R. Bardhan, 42 Y. Ding, F. Fan, Z. Tian and Z. L. Wang, J. Am. Chem. Soc.,
Energy Environ. Sci., 2016, 9, 1577–1601. 2010, 132, 12480–12486.
9 C. Wang, O. Ranasingha, S. Natesakhawat, P. R. Ohodnicki, 43 M. Jaschke, H. J. Butt, H. E. Gaub and S. Manne, Langmuir,
1
M. Andio, J. P. Lewis and C. Matranga, Nanoscale, 2013, 5,
968–6974.
1997, 13, 1381–1384.
44 T. Bora, D. Zoep and J. Dutta, Sci. Rep., 2016, 6, 26913.
6
3
3
3
3
3
3
3
3
3
0 G. K. Larsen, W. Farr and S. E. Hunyadi Murph, J. Phys. 45 P. Dombi, A. H ¨o rl, P. R ´a cz, I. M ´a rton, A. Tr u¨ gler, J. R. Krenn
Chem. C, 2016, 120, 15162–15172. and U. Hohenester, Nano Lett., 2013, 13, 674–678.
1 J. R. Adleman, D. A. Boyd, D. G. Goodwin and D. Psaltis, 46 S. Mukherjee, F. Libisch, N. Large, O. Neumann, L. V. Brown,
Nano Lett., 2009, 9, 4417–4423.
2 X. Huang, Y. Li, Y. Chen, H. Zhou, X. Duan and Y. Huang,
Angew. Chem., Int. Ed., 2013, 52, 6063–6067.
3 Q. Xiao, S. Sarina, E. Jaatinen, J. Jia, D. P. Arnold, H. Liu and
H. Zhu, Green Chem., 2014, 16, 4272–4285.
4 P. Verma, Y. Kuwahara, K. Mori and H. Yamashita, J. Mater.
Chem. A, 2016, 4, 10142–10150.
J. Cheng, J. B. Lassiter, E. A. Carter, P. Nordlander and
N. J. Halas, Nano Lett., 2013, 13, 240–247.
47 Y. Li, X. M. Hong, D. M. Collard and M. A. El-Sayed, Org. Lett.,
2000, 2, 2385–2388.
48 D. Han, Z. Bao, H. Xing, Y. Yang, Q. Ren and Z. Zhang,
Nanoscale, 2017, 9, 6026–6032.
49 A. Samanta, T. Rajesh and R. Nandini Devi, J. Mater. Chem. A,
2014, 2, 4398–4405.
5 Z. Zheng, T. Tachikawa and T. Majima, J. Am. Chem. Soc.,
2
015, 137, 948–957.
50 Y. Mu, H. Zhang, W. Zheng and X. Cui, New J. Chem., 2017,
41, 786–792.
51 R. Long, Z. Rao, K. Mao, Y. Li, C. Zhang, Q. Liu, C. Wang,
Z.-Y. Li, X. Wu and Y. Xiong, Angew. Chem., Int. Ed., 2015,
54, 2425–2430.
6 M. Wen, S. Takakura, K. Fuku, K. Mori and H. Yamashita,
Catal. Today, 2015, 242, 381–385.
7 S. Zhang, C. Chang, Z. Huang, Y. Ma, W. Gao, J. Li and Y. Qu,
ACS Catal., 2015, 5, 6481–6488.
8 E. Hao and G. C. Schatz, J. Chem. Phys., 2004, 120, 357–366. 52 T. Mitsudome and K. Kaneda, Green Chem., 2013, 15, 2636–
654.
2
40226 | RSC Adv., 2017, 7, 40218–40226
This journal is © The Royal Society of Chemistry 2017