Mechanistic insight into the Z-E isomerization catalysis of azobenzenes mediated by bare and core-shell gold nanoparticles

Article abstract of DOI:10.1039/c4cy01442a

We explored the catalytic effect of 15 nm diameter gold nanoparticles (AuNPs) upon the thermal Z-E isomerization reaction of azobenzene and nine 4 and 4-4′ substituted azobenzenes (ABs). The kinetics follows a first order rate in ranges of [ABs] = 5 to 50

Full text of DOI:10.1039/c4cy01442a

Catalysis
Science &
Technology
View Article Online
View Journal
PAPER
Mechanistic insight into the Z–E isomerization
catalysis of azobenzenes mediated by bare and
core–shell gold nanoparticles†
ab
Cite this: DOI: 10.1039/c4cy01442a
Sabrina Simoncelliac and Pedro F. Aramendía
*
We explored the catalytic effect of 15 nm diameter gold nanoparticles (AuNPs) upon the thermal Z–E
isomerization reaction of azobenzene and nine 4 and 4-4′ substituted azobenzenes (ABs). The kinetics
follows a first order rate in ranges of [ABs] = 5 to 50 μM and [AuNPs] = 50 pM to 1 nM. A kinetic analysis of
this compartmentalized system renders the thermal Z–E isomerization rate constant associated with each
AuNP. Enhancements of 10- to 106-fold were measured for this rate constant in comparison to the same
free ABs in solution. Experiments with selective Au facet coverage, as well as the kinetics studied in
gold–silica core–shell nanoparticles (AuNP@SiO2) of different thicknesses, demonstrate the surface nature
of the catalysis and allow one to evaluate the diffusion coefficient of azobenzene in the silica layer.
Received 4th November 2014,
Accepted 5th January 2015
DOI: 10.1039/c4cy01442a
www.rsc.org/catalysis
distance.7 In this approach, Whitesell et al.8 reported a
~100-fold increase in the isomerization quantum yield of
Introduction
The interaction of metallic nanoparticles (NPs) with azobenz-
enes has received considerable attention during the past
decades in view of the changes that the photochromic trans-
formation confers to nanoparticle properties and photo-
controllable layers on metal surfaces.1 The adsorption of
switchable molecules onto NPs can modify the optical proper-
ties of the latter either by local variation of the dielectric con-
stant of the encompassing medium, due to the changes in
the electronic properties of the isomers,2,3 or by controlling
the proximity or state of aggregation of the particles upon
photoisomerization.4–6 The isomerization efficiency of photo-
chromic probes located in the vicinity of a NP has also
attracted great interest. Metallic NPs can deactivate the
electronic state of organic molecules placed near their surface
through energy transfer, thus modulating the switching capa-
bility of the probe. This is a function of the photochromic–NP
azobenzenes located at ~1.25 nm from the surface of a
2.5 nm diameter gold nanoparticle (AuNP) with respect to the
ones located at ~0.54 nm. An appealing but less studied
aspect of the interaction is the analysis of the isomerization
kinetic rates of azobenzenes placed near metallic NPs. In
2008, Shin et al. investigated the kinetic rates of E–Z photo-
conversion and Z–E thermal isomerization of an azobenzene
(AB)–alkanethiol self-assembled layer around a 2 nm diameter
AuNP.9 They reported an increase between two and three
times, respectively, for the isomerization rate constants com-
pared to that of the free dye in solution. On the other hand,
the group of Tamada et al. obtained identical reaction kinetic
rates for a 5.2 nm diameter AuNP capped with 4-hexyl-4′-
IJ12-IJdodecyldithio)dodecyloxy)azobenzene and free AB mole-
cules.10 Prior to the work of Yoon et al. in 2011, nobody had
outlined a detailed picture of the interactions involved in the
isomerization of azobenzenes near the surface of metallic
NPs.4 In this report, by using surface enhanced Raman
spectroscopy measurements, a 15-fold acceleration in the
thermal Z–E isomerization kinetic rate of azobenzenes linked
to AuNP aggregates was attributed to the weakening of the
N–N double bond of the Z-AB isomer near the metallic
surface. Further, Scaiano et al. were the first ones to report
the AuNP catalysis of the thermal Z–E isomerization of
unbounded 4 and 4-4′-substituted ABs in suspensions of
‘pseudo-naked’ AuNPs.11 In this case, the proposed mecha-
nism involves the formation of an azobenzene radical cation
intermediate through electron transfer from the probe to the
gold surface. In this work, average rates were measured in
this AuNP suspension in water. In this contribution, we have
a Centro de Investigaciones en Bionanociencias “Elizabeth Jares-Erijman”
(CIBION-CONICET), Godoy Cruz 2390, 1425 Buenos Aires, Argentina.
E-mail: pedro.aramendia@cibion.conicet.gov.ar
b Departamento de Química Inorgánica, Analítica y Química Física, Facultad de
Ciencias Exactas y Naturales, Universidad de Buenos Aires, Pabellón 2, Ciudad
Universitaria, 1428 Buenos Aires, Argentina
c Instituto de Química Física de Materiales, Ambiente y Energía (INQUIMAE-CONICET),
Pabellón 2, Ciudad Universitaria, 1428 Buenos Aires, Argentina
† Electronic supplementary information (ESI) available: Gold nanoparticle
characterization, kinetic traces for NO2-DAB, DABCYL and NO2-HAB thermal
Z–E isomerization, pH control experiments, global kinetic rate influence on the
relative ABs–AuNPs concentration, kinetic traces for the Z–E isomerization of AB
in suspensions of different thicknesses of AuNPs@SiO2, determination of the
AuNP associated first order thermal isomerization rate constant, and derivation
of the diffusion equation. See DOI: 10.1039/c4cy01442a
This journal is © The Royal Society of Chemistry 2015
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
thus examined the kinetics of the thermal Z–E isomerization
for a family of 4 or 4-4′ substituted azobenzenes in AuNP sus-
pensions with the aim of answering the following questions:
What is the magnitude of the catalytic effect of the AuNPs?
Can we obtain the Z–E thermal isomerization rate of azobenz-
enes over the AuNPs? What is the role of the metallic surface?
Is it necessary for the photochromic compound to be in direct
contact with the metallic surface or does the NP influence on
the catalysis extend in volume due to the high polarizability
of the NP plasmonic interaction? What is the influence of the
substituents in the isomerization catalysis? What is the role
of the pH of the medium?
Fig. 2 (A) UV-Vis temporal evolution of a 15 μM AB aqueous solution
with 300 pM AuNP suspension (diameter = 15 1 nm) measured after
UV flash photolysis. The inset shows the enlarged spectral region of
the n–π* band. The band with a maximum at around 520 nm repre-
sents the AuNP extinction. (B) Growth of the E isomer concentration
measured at 320 nm (black triangles). The red line is the mono-
exponential fit to the experimental data. The dashed black line near
the abscissa axis represents the temporal evolution of the E isomer in
the absence of AuNPs in this time interval.
Results
Fig. 1 shows transmission electron microscopy (TEM) micro-
graphs and histograms of the size distribution of the AuNP
and the silica coated AuNP (AuNP@SiO2) used in this work.
AuNPs are highly monodisperse (with a diameter of 15 1 nm,
standard deviation). On the other hand, the obtained
AuNPs@SiO2 exhibit >99% effectiveness in the coating process
but a thickness standard deviation in the order of 20%.
Catalyzed Z–E thermal isomerization by AuNPs was
observed for AB (Fig. 2) and the following nine 4 or 4-4′
substituted azobenzenes: DM-AB, DC-AB, DEO-AB, MO-AB,
NO2-AB, NO2-MOAB, NO2-DAB, DABCYL and NO2-HAB (Fig. 3
and S3†). Fig. 2, 3, and S3† show the temporal spectral evolu-
tion of azobenzenes in a AuNP suspension. The spectral
information of the AuNP/azobenzene systems corresponds to
the characteristic spectral changes of the Z → E isomerization
reaction for the free dye in solution, for example, for AB,
increase of the π–π* band (λmax ~ 320 nm) and decrease of
the n–π* band (λmax ~ 450 nm) (Fig. 2-A), and they are invari-
ant in the presence of AuNPs.12,13
that the Z–E isomerization is rate determining compared to
other dynamic events. Specifically, the addition of 0.3 to
0.4 nM AuNPs of 15 nm diameter to 10 μM azobenzene solu-
tions resulted in accelerated Z–E isomerization. The high
activity of AuNP catalysis gives rise to a 104-fold increase in
the global first order isomerization rate constant, kobs, of AB
and MO-AB in suspensions of AuNPs in water. On the other
hand, for the systems of AuNPs and symmetrically substituted
azobenzenes (DM-AB, DC-AB, and DEO-AB) in mixtures of
acetonitrile (ACN) : water (5 : 1) the isomerization reaction is
accelerated ~100 times with respect to the free dye in solution.
For NO2-AB and NO2-MOAB (also in 5 : 1 ACN : water) the
The small scattering by the AuNP did not affect the mea-
surements since the kinetic data were derived from differen-
tial absorption data (see Fig. 2 and 3).
Fig. 2-B and S3† as well as the insets in Fig. 3 display the
time evolution of the build-up of the E isomer in AuNP sus-
pensions. The kinetics follows a first order rate with very
good approximation in all cases and in ranges of [ABs] = 5 to
50 μM and [AuNPs] = 50 pM to 1 nM. This points to the fact
Fig. 3 UV-Vis temporal evolution of a 300 pM or 400 pM AuNP
suspension (diameter = 15 1 nm) and 10 μM azobenzene (A) DM-AB,
(B) DC-AB, (C) DEO-AB, (D) MO-AB, (E) NO2-AB and (F) NO2-MOAB
solutions after UV flash photolysis. The inset corresponds to the growth
of the E isomer concentration measured at (A) 330 nm, (B) 330 nm,
(C) 360 nm, (D) 345 nm, (E) 330 nm, and (F) 370 nm. The red lines
represent the monoexponential fit to the experimental data. The
dashed black line corresponds to the temporal evolution of the E
isomer in the absence of AuNPs in the same time interval.
Fig. 1 TEM micrographs (upper panel) and size histograms (lower
panel) of (A) AuNPs of 15
increasing SiO2 shell thicknesses of (B) 2.7 0.6 nm, (C) 4 1 nm, (D)
1 nm, (E) 13 1 nm and (F) 15 2 nm. Scale bars: 20 nm in all
1 nm diameter and AuNP@SiO2 with
7
cases. Histograms were constructed with sample set size n > 50.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Catalysis Science & Technology
Paper
catalytic acceleration of the Z–E isomerization is much lower,
with factors of 30 and 2, respectively. Regarding NO2-DAB and
DABCYL a 103 times increase is observed for the thermal
isomerization rate, while a 50-fold increase is detected for
NO2-HAB.
The addition of AuNPs to the reaction medium provokes
a decrease in the pH. Further, it is well known that acidic
pH highly increases the Z–E thermal isomerization rate
in azobenzenes.14–17 Therefore, control experiments were
performed for azobenzene solutions at the same pH of the
AuNP suspensions but in the absence of AuNPs. The isomeri-
zation kinetics for AB, DM-AB, DC-AB, DEO-AB, MO-AB,
NO2-AB and NO2-MOAB, in such control experiments, were
identical to the ones observed for the corresponding azobenz-
enes in solution in the absence of acid and AuNPs (see
Fig. S4†). On the other hand, for NO2-DAB, DABCYL and
NO2-HAB the isomerization rate constants for samples with
the same pH are identical, regardless of the presence or
absence of AuNPs (see Fig. S5†). Thus, a potential AuNP cata-
lytic action upon the thermal Z–E isomerization reaction for
these three last mentioned azobenzenes is masked by the
changes in acidity that AuNP addition causes to the solution.
Fig. 4-A and B illustrate the dependence of kobs with
[AuNP] and [AB] concentrations, respectively.
Fig. 5 Inverse of the thermal Z–E isomerization reaction rate constant
as a function of the ijZ-AB]T/ijAuNP] for (A) aqueous and (B–C) ACN : water
(5 : 1) solutions of different azobenzene/AuNP systems. The red line
represents the best linear fit (see eqn (2)).
The trends can be interpreted considering a reaction scen-
ery where Z isomers decay to the stable E form either in the
continuous solution phase or under the influence of the dis-
persed AuNPs that accelerate the reaction rate. Thus, the
increase of kobs with AuNP concentration is related to the
increase in the relative amount of Z molecules influenced by
AuNPs. On the other hand, the increase of AB in the medium
decreases the molar fraction of AB molecules associated with
AuNPs, which in turn is manifested as a decreased reaction
rate constant. From these experiments it stands out that the
magnitude of the observed Z → E first-order kinetic rate con-
stant only depends on the relative AB–AuNP concentration
(see Fig. 5 and S7†).
common approach is to use target-blocking agents to deter-
mine the changes of reactivity.19,20 Polyvinylpyrrolidone (PVP)
selectively binds to gold and silver {100} facets.21 On the
other hand, sodium citrate binds stronger to {111} compared
to {100} facets. Therefore, we decided to functionalize the cit-
rate stabilized 15 nm AuNPs (bare AuNPs) with PVP, poison-
ing the catalytic activity of the {100} facets of the AuNPs.
Fig. 6 shows kobs as a function of the relative concentration
of Z-AB and AuNPs for experiments with (empty circles) and
without (black triangles) PVP functionalization.
The catalytic surface of a metal is heterogeneous by
nature.18 One of the main objectives in catalysis is to deter-
mine the number of catalytically different sites and the con-
tribution of each of them in global reaction. In this regard, a
Fig. 4 Dependence of the Z–E thermal isomerization first order rate
constant for an aqueous solution (A) on the AuNP concentration
(50 μM AB) and (B) on the AB concentration (300 pM AuNPs). The
diameter of the AuNPs is 15 1 nm.
Fig. 6 Inverse of the thermal Z–E isomerization reaction rate constant
as a function of the ijZ-AB]T/ijAuNP] for aqueous solution of azobenzenes
and citrate stabilized (black triangles) or PVP-functionalized AuNPs
(empty circles). The red line represents the best linear fit (see eqn (2)).
This journal is © The Royal Society of Chemistry 2015
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
Interestingly, PVP-functionalized AuNP samples are much
less efficient for the catalyzed thermal Z–E isomerization
reaction than the bare-AuNPs. This can be explained as a
decrease in the number of AB binding sites per AuNP in the
presence of PVP to the more reactive {100} facets.
Time dependence evolution of the catalytic growth of E-AB
by PVP-covered (blue circles) and silica-coated (green trian-
gles and red squares) AuNPs is presented in Fig. 7.
AuNPs@SiO2 synthesis requires transferring AuNPs from
water to ethanol solutions using PVP as the coupling agent.
Thus, bare AuNPs were functionalized with PVP and dis-
persed in ethanolic media for strict data comparison. Adsorp-
tion effects over SiO2 were discarded in a control experiment
that gave equal isomerization rate constants for free AB in
solution and for AB in suspensions of silica NPs (without a
Au core) (dashed line, Fig. 7). The Z–E isomerization rate
constant for AB in suspensions of PVP functionalized
AuNPs is ~35-fold higher than the corresponding for suspen-
Fig. 8 Growth of the E-AB isomer concentration measured at 320 nm
for ethanolic solutions of 8 μM AB in suspension of AuNPs@SiO2: (A)
2.7 0.6 nm ([AuNP] = 0.8 nM) and (B) 15 2 nm ([AuNP] = 1.0 nM)
silica shell thicknesses immediately after (black triangles) or after 12 h
(grey circles) of NP addition. The blue lines are the monoexponential
fits to the data. The insets correspond to the logarithmic representa-
tion of the absorbance growth. The data recorded immediately after
NP addition were temporally delayed for a better comparison. The blue
and red lines correspond to the best linear fits.
sions of AuNPs@SiO2 with 4
1 nm thickness. Further,
the sigmoidal temporal dependence for the E-AB growth
observed for core–shell NPs with a silica shell thickness of
15 2 nm (red squares, Fig. 7) points to a diffusion controlled
process. Specifically, immediately after NP addition, there is an
initial time delay that allows Z-AB molecules to diffuse suffi-
ciently close to the Au core to exhibit Z–E catalyzed reaction. The
exponential increase of the absorbance of the E isomer is delayed
as the reaction proceeds to complete conversion to this isomer.
Evaluation of the diffusion regime observed in the differ-
ent silica shell thicknesses of AuNPs@SiO2 was performed by
comparing the temporal build-up of the E concentration
isomer immediately after NP addition (Fig. 8, black triangles)
and after leaving the AB/AuNPs@SiO2 solution to incubate in
the dark overnight (Fig. 8, grey circles). The thinner silica
shell AuNPs@SiO2 tested, 2.7 0.6 nm, displays almost the
same temporal dependence for the E isomer growth, inde-
pendent of the mixing time in the dark before photolysis
(Fig. 8-A). Therefore, AB reaches a stationary diffusion regime
in the silica layer during the Z isomer decay for this thick-
ness. On the other hand, for NPs with the thickest silica
shells, faster Z–E conversion is detected after allowing
azobenzene molecules to diffuse inside the silica layer over-
night (Fig. 8-B and S8†). This indicates a non-stationary diffu-
sion regime for AB molecules in silica shells thicker than
2.7 0.6 nm immediately after NP addition. Insets of Fig. 8
show the log data representation for the exponential growth
phase of the kinetic traces. Parallel curves point to the sta-
tionary state diffusion regime for AB in the silica layer. Dis-
tinctively, for 2.7 0.6 nm silica shell thickness, parallelism
of both plots is observed, while the thicker the silica layer, the
higher is the ratio of slopes between the incubated and the
as-prepared samples. This ratio is 2 for 4 1 nm thickness
and 3 for 7 1 nm, 13 1 nm, and 15 2 nm silica layers.
Discussion
The kinetics and mechanism of the catalyzed thermal back
isomerization of azobenzenes in AuNP suspensions are ana-
lyzed by studying the dependence of kobs as a function of the
relative AB–AuNP concentration (see Fig. S7† and 5).
To interpret the physicochemical meaning of the observed
first order rate constant, we propose a compartmentalized
reaction scenery (see Scheme 1) where the Z and E isomers
are distributed in two phases, the dispersed phase (formed
by the sum of the volume of influence of all the AuNPs) and
the continuous (solvent) phase. Assuming rapid equilibration
Fig. 7 Growth of the E-AB isomer concentration measured at 320 nm
for ethanolic solutions of 8 μM AB in suspension of the following:
2 nM, 15
1 nM AuNPs@SiO2 with 4
squares) silica shell thickness; and 2 nM 71
1 nm diameter, PVP functionalized AuNPs (blue circles);
1 nm (green triangles); 15
2 nm (red
8 diameter silica nano-
particles (black dashed line) immediately after the addition of NPs and
UV flash photolysis. The inset corresponds to a time range expansion
for the reaction catalyzed by PVP covered gold nanoparticles.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Catalysis Science & Technology
Paper
azobenzenes. On the other hand, the constant average
number of Z molecules associated with each single AuNP
evidenced in the kinetic experiment can be considered as the
maximum number of available adsorption sites in the surface
of each AuNP, nmax. This latter parameter can be estimated
using different assumptions. In this work, it was approxi-
mated as 103 binding sites for a 15 nm diameter AuNP and
considered independent of the derived structure of the
azobenzene (see S7† for detailed calculation).
Scheme
1 Mechanism for the microheterogeneous dark Z–E
isomerization catalyzed by AuNPs.
The acceleration of the reaction rate constant measured
on Au {100} facets is in line with the reported reduction of
the activation energy by a factor of four for the thermal Z–E
isomerization of adsorbed tetra-tert-butyl-azobenzene on Au
{111} compared to the free molecule in the liquid phase.22 In
another work, the authors find a change in the lifetime of
3-IJ4-IJ4-hexyl-phenylazo)-phenoxy)-propane-1-thiol from more
than 35 h in dichloromethane solution to ca. 60 s in a self-
assembled mixed monolayer with n-butanethiol on Au {111}.23
Table 1 reports the values of kAuNP calculated from the
best linear fit of the experimental data of Fig. 5. The
enhancement factor in kAuNP for AB and MO-AB in aqueous
suspensions of AuNPs is in the order of 105. On the other
hand, DM-AB, DC-AB and DEO-AB show a 103-fold acceleration
in suspensions of ACN: water (5 : 1). Further, NO2 substituted
azobenzenes only show a 102 or 101 AuNP catalytic factor in
the same solvent mixture.
for the reactant exchange steps between compartments com-
pared to the reactive steps (i.e. k+[Z-AB]S, k− ≫ kS, and kAuNP),
and neglecting reactant exchange by direct AuNP interac-
tions, the first order reaction rate constant can be expressed
as a weighted average of the first order isomerization rate
constants in solution, kS, and in the vicinity of AuNPs, kAuNP
(uncatalyzed and catalyzed reaction pathways, respectively).
The weighting factors are the mole fractions of Z-AB mole-
cules associated with each phase. Further, expressing the
concentration of the Z isomer in the disperse phase as a
function of the average number of Z molecules associated
with each single AuNP, <n>, the following expression for kobs
can be obtained (see S6†),
n
AuNP


 k
kobs = kS 
 kS
AuNP
(1)


Z-AB


T
The acceleration of the reaction rate induced by AuNPs is
related to the weakening of the azo double bond character on
the surface of the gold nanoparticle.4 Azobenzenes carrying
an electronegative group present intrinsically a weaker –N–N–
double bond character and therefore the expected AuNP cata-
lytic effect should be poorer. However, if we only consider
the electronegativity factor of the azobenzene substituents,
we cannot explain the acceleration observed for the MO-AB
Z–E thermal isomerization relative to the symmetrically
substituted azobenzenes such as DM-AB or DC-AB. To give an
explanation for this fact, we should take into account that
ACN is reversibly adsorbed on the gold surface with a bind-
ing energy of 46 kJ mol−1 onto the Au(100) facets,24 thus com-
peting with the azobenzene molecules for the adsorption
sites (Ead ~ 160 kJ mol−1 for Au(111)).25 Therefore, decrease in
the global isomerization reaction rate constant for systems
measured in ACN : water (5 : 1) with respect to the ones mea-
sured in water can be attributed to a decrease in the value of
where ijZ-AB]T is the total concentration of the Z-AB isomer
obtained immediately after flash photolysis. This parameter
is evaluated from the difference absorption spectrum.
For all the studied azobenzenes and for all the combina-
tions of [AuNP]/[AB] tested it holds that kobs ≫ kS, accordingly
kAuNP ≫ kS, and therefore, eqn (1) can be simplified to eqn (2),
n
AuNP


kAuNP
kobs
=
(2)
Z-AB


T
Fig. 5 displays the inverse of kobs as a function of
ijZ-AB]T/ijAuNP] for different azobenzene–AuNP systems. As
expected, the increase of kobs as a function of ijAuNP]/ijZ-AB]T
is related to the higher proportion of Z-AB molecules associ-
ated with the disperse phase. The linear relationship observed
in Fig. 5 suggests that <n> is a constant value in the concen-
tration range tested.
The experiments with PVP blocking the {100} facets give
the hint of the predominant surface character of the catalysis
and point to these planes as the more catalytically active
ones. The experiments performed with AuNPs@SiO2 suspen-
sions allowed further elucidation of the spatial catalytic
extension of the AuNPs. Particularly, it was possible to deter-
mine whether the isomerization catalysis was active at nano-
meter distance from the AuNP surface or if it only takes
place upon direct contact with the metallic surface. The
distinct catalytic behavior for even the thinner silica shell
AuNPs@SiO2 with respect to the bare NPs as well as the
competitive experiments with PVP enlightens the surface
catalytic nature of the AuNPs on the Z–E isomerization in
Table 1 Parameters of the best linear fit of the inverse of eqn (2) for the
experimental data of Fig. 5
a
Azobenzene
nmax kNP (s−1
)
kNP (s−1
)
kS (s−1
)
kNP/kS
AB
1 × 103
4 × 101
2 × 101
6 × 101
3 × 103
3 × 101
5 × 101
1 × 100−2
1 × 10−6
7 × 10−6
6 × 10−6
3 × 10−5
6 × 10−6
1 × 10−4
1 × 10−3
1 × 106
6 × 103
4 × 103
2 × 103
5 × 105
2 × 102
4 × 101
DM-AB
DC-AB
DEO-AB
MO-AB
NO2-AB
NO2-MOAB
4 × 10
2 × 10−2
6 × 10−2
3 × 100
3 × 10−2
5 × 10−2
a
Obtained considering in all cases nmax = 103 (see S7†).
This journal is © The Royal Society of Chemistry 2015
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
Table 2 Diffusion rate constant, kdiff, and diffusion coefficient, DAB‐SiO
,
masks the AuNP catalytic effect. Selective blocking experi-
ments with PVP suggested that the {100} crystalline facets are
particularly active for the catalysis. Experiments performed
with core–shell AuNPs with different silica spacers around
the gold surface reveal the exclusive surface nature of the
AuNP mediated catalysis. Further, it was possible to deter-
mine the diffusion coefficient of AB in silica layers as
2
for AB molecules in silica as a function of silica shell thickness, hSiO . kdiff
2
was determined with the best monoexponential fit of the data of Fig. 7
and S8† (grey circles). The DAB‐SiO was calculated using eqn (3)
2
hSiO (nm)
kdiff (10−3 s−1
)
DAB‐SiO (10−11 cm2 s−1
)
2
2
2.7 0.6 nm
0.68
0.82
1.6
1.8
1.9
1.1 0.5
1.6 0.7
4
7
1 nm
1 nm
DAB‐SiO ~ 10−10–10−11 cm2 s−1.
5
11
8
2
4
3
2
13 1 nm
15 2 nm
Experimental
Chemicals
nmax. Further, the NO2 groups can also be reversibly adsorbed
on the gold surface (Ead ~ 58.6 kJ mol−1 for Au(111))26 and
therefore compete with the azo moiety for the surface bind-
ing sites, without the same effect in the reactive rate.
All reagents given as follows were used as received unless
specified otherwise: HAuCl4Ĵ3H2O (≥99.9%, Aldrich),
trisodium citrate dihydrate (≥99%, Aldrich), 2-hydroxy-4′-IJ2-
hydroxyethoxy)-2-methylpropiophenone (Irgacure 2959) (98%,
Aldrich), tetraethyl orthosilicate (≥98.0%, Fluka) (TEOS),
IJ3-aminopropyl)triethoxysilane (≥98%, Sigma Aldrich),
polyvinylpyrrolidone (PVP10) (average molecular weight,
10 000 g mol−1, Sigma Aldrich), acetonitrile (HPLC grade,
Sintorgan), azobenzene (98%, Sigma Aldrich) (AB), 4-methoxy-
azobenzene (≥99.0%, Aldrich) (MO-AB), 4-nitroazobenzene
Finally, we can use the kinetic curves in AuNPs@SiO2 to
estimate the diffusion coefficient of AB in silica, DAB‐SiO , by
2
solving Fick's equation for steady state diffusion in a spheri-
cal geometry (see S8†). Eqn (3) holds for the DAB‐SiO as a
2
function of the diffusion kinetic rate constant, kdiff, the
AuNPs radius, rNP, the thickness of the silica shell, hSiO , and
2
[AuNPs].
(90%, Aldrich) (NO2-AB), 4,4′-dimethylazobenzene (AldrichCPR
(DM-AB), 4,4′-dichloroazobenzene (AldrichCPR
(DC-AB),
4,4′-diethoxyazobenzene (97%, ALFA AESAR) (DEO-AB),
4-nitro-4′-methoxyazobenzene (NO2-MOAB), 4-nitro-4′-
)
h
kdiff
SiO2
.
DAB-SiO
=
(3)
)
2
4 AuNP  N rNP(rNP + hSiO
)


A
2
Table 2 summarizes kdiff and DAB‐SiO for the different
2
dimethylaminoazobenzene (≥98.0%, TCI) (NO2-DAB), 4-IJ4′-
nitrophenylazo)phenol (≥97.0%, TCI) (NO2-HAB), and
4-ij4-IJdimethylamino)phenylazo]benzoic acid N-succinimidyl
ester (Sigma Aldrich) (DABCYL). All of the following reagents
were of analytical grade: hydrochloric acid, ammonia, ethanol,
and sodium hydroxide.
silica shell thicknesses. The small diffusion coefficient for AB
in silica IJ~10−10–10−11 cm2 s−1) in comparison to free diffu-
sion of AB in water (~10−6 cm2 s−1)27 is related to the tortuous
pathway for diffusing molecules inside the disordered and
microporous silica shell (estimated porous size of ~15 Å).28
Further, the hydrophilic silica pores (active hydroxyl groups)
may promote the adsorption of molecules in the silica pores
(chemical traps) precluding the arrival of Z-AB molecules to
Nanoparticle synthesis
the gold surface. As regards the higher DAB‐SiO values for
2
Au nanoparticles with an average radius of 15
1 nm
thicker silica shells, this can be understood as a more compact
silica matrix for the proximal layers compared to the more
expanded outer part.
(see Fig. 1) were prepared following the Turkevich protocol.29
‘Semi-naked’ gold nanoparticles were photochemically syn-
thesized as described elsewhere30 obtaining nanoparticles
with a mean radius of 10 2 nm (see Fig. S1†). Au–SiO2 core–
shell nanoparticles (AuNP@SiO2) of 2.7 0.6, 4 1, 7 1,
Conclusions
Z–E thermal isomerization reaction of 4 or 4-4′ substituted
azobenzenes is highly accelerated in sub-nanomolar 15 1 nm
diameter AuNP suspensions. Microheterogeneous kinetic anal-
ysis was found useful to describe the combined process of the
thermal Z isomer decay in the continuous solution phase and
under the catalytic influence of the dispersed AuNPs. With this
approach it was possible to determine the first order isomeriza-
tion rate constants in the vicinity of AuNPs. The magnitude of
the acceleration is determined not only by the nature of the
substituents but also by the solvent and surfactants present in
the reaction medium. The maximum acceleration factor
detected, of 106 times, was for AB in aqueous solution and in
the absence of external surfactants. The competitive adsorp-
tion of acetonitrile, polyvinylpyrrolidone (PVP) and NO2
groups in the azo moiety binding sites of the gold surface
13
1, and 15
2 nm silica shell thicknesses (see
Fig. 1-B to F) were synthesized following the van Blaaderen
procedure.31 Briefly, PVP10 was used as a coupling agent to
transfer the colloids from water to a 4.2% v/v ammonia–
ethanol medium. Silica deposition over the gold core was
performed by quickly adding TEOS under vigorous stirring.
The solution was kept under stirring at room temperature
for 12 h. The silica-coated nanoparticles were purified by 3 cycles
of redispersion in ethanol and centrifugation at 12 000 RCF.
The final concentration of the nanoparticle solution was
determined by gold quantification of the nanoparticle sus-
pension with a graphite furnace atomic absorption spectro-
meter. Silica nanoparticles (mean radius = 71 8 nm, see
Fig. S2†) employed as control samples were prepared following
a standard Stöber protocol.32
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Catalysis Science & Technology
Paper
Spectroscopy
6 D. S. Sidhaye, S. Kashyap, M. Sastry, S. Hotha and
B. L. V. Prasad, Langmuir, 2005, 21, 7979–7984.
7 E. Dulkeith, A. C. Morteani, T. Niedereichholz, T. A. Klar,
J. Feldmann, S. A. Levi, F. C. J. M. van Veggel,
D. N. Reinhoudt, M. Möller and D. I. Gittins, Phys. Rev. Lett.,
2002, 89, 203002.
8 J. Zhang, J. K. Whitesell and M. A. Fox, Chem. Mater.,
2001, 13, 2323–2331.
9 K. Shin and E. J. Shin, Bull. Korean Chem. Soc., 2008, 29,
1259–1262.
UV-VIS spectra were measured on a Hewlett Packard 8452A,
on a Shimadzu UV3101PC (Shimadzu Corporation, Kyoto,
Japan), or on a Cary 50 UV-Vis (Agilent Technologies, Santa
Clara, USA). Laser flash photolysis experiments were carried
out using a Surelite-II OPO Plus (a pump with a Nd-YAG at
355 nm) (Continuum, Santa Clara, USA) as the excitation
source. Data were recorded with a LFP 111 laser-flash photo-
lysis system (Luzchem Inc., Ottawa, Canada).
System characterization
10 A. Manna, P.-L. Chen, H. Akiyama, T.-X. Wei, K. Tamada and
W. Knoll, Chem. Mater., 2003, 15, 20–28.
Nanoparticle mean radius and silica shell thickness were
measured by either transmission electron microscopy (TEM)
using a Philips EM 301 apparatus or scanning electron
microscopy (SEM) using a JEOL JSM-7500F.
11 G. L. Hallett-Tapley, C. D'Alfonso, N. L. Pacioni,
C. D. McTiernan, M. González-Béjar, O. Lanzalunga, E. I. Alarcon
and J. C. Scaiano, Chem. Commun., 2013, 49, 10073–10075.
12 N. Nishimura, T. Sueyoshi, H. Yamanaka, E. Imai,
S. Yamamoto and S. Hasegawa, Bull. Chem. Soc. Jpn.,
1976, 49, 1381–1387.
Sample preparation
Typical experiments were performed in 1 cm × 1 cm quartz
cuvettes combining 50 to 500 μL of AuNPs or AuNP@SiO2
with water or ethanol or mixtures of acetonitrile–water until a
final volume of 3 mL was reached. Kinetic experiments were
performed after gradual addition of 5 to 45 μL of 5 mM or
1 mM ACN solutions of the corresponding AB to the previous
mentioned NP suspension. All azobenzenes/AuNP suspen-
sions except for AB and MO-AB were prepared in 5 : 1 ACN :
water mixtures. E–Z photochemical conversion was performed
with either a commercial photographic xenon flash lamp or
with the corresponding excitation wavelength of the OPO
laser. For the pH dependence experiments solutions of HCl or
NaOH were also used.
13 P. P. Birnbaum, J. H. Linford and D. W. G. Style, Trans.
Faraday Soc., 1953, 49, 735–744.
14 G. S. Hartley, J. Chem. Soc., 1938, 633–642.
15 N. J. Dunn, W. H. Humphries, A. R. Offenbacher, T. L. King
and J. A. Gray, J. Phys. Chem. A, 2009, 113, 13144–13151.
16 A. M. Sanchez and R. H. Rossi, J. Org. Chem., 1995, 60,
2974–2976.
17 S. Ciccone and J. Halpern, Can. J. Chem., 1959, 37, 1903–1910.
18 H. S. Taylor, Proc. R. Soc. London, Ser. A, 1925, 108, 105–111.
19 S. M. Oxford, J. D. Henao, J. H. Yang, M. C. Kung and
H. H. Kung, Appl. Catal., A, 2008, 339, 180–186.
20 M. M. Nigra, I. Arslan and A. Katz, J. Catal., 2012, 295,
115–121.
Acknowledgements
21 Y. Sun, B. Mayers, T. Herricks and Y. Xia, Nano Lett.,
2003, 3, 955–960.
PFA is a research staff and SS is a research fellow from
CONICET. The work was performed under support from
CONICET (PIP 11220100100397) and UBA (grant number:
20020100100234). SS acknowledges ELAP (Emerging Leaders
in the Americas Program) for the DFAIT fellowship that
supported her visit to Canada. We thank Prof. Juan C. Scaiano
(University of Ottawa) for helpful suggestions and for providing
the facility used to perform the laser flash photolysis experi-
ments and to synthesize the ‘semi-naked’ gold nanoparticles.
22 S. Hagen, P. Kate, M. V. Peters, S. Hecht, M. Wolf and
P. Tegeder, Appl. Phys. A: Mater. Sci. Process., 2008, 93,
253–260.
23 U. Jung, O. Filinova, S. Kuhn, D. Zargarani, C. Bornholdt,
R. Herges and O. Magnussen, Langmuir, 2010, 26,
13913–13923.
24 T. Solomun, K. Christmann and H. Baumgaertel, J. Phys.
Chem., 1989, 93, 7199–7208.
25 E. R. McNellis, J. Meyer and K. Reuter, Phys. Rev. B: Condens.
Matter Mater. Phys., 2009, 80, 205414.
Notes and references
26 M. E. Bartram and B. E. Koel, Surf. Sci., 1989, 213, 137–156.
27 W. Dozier, J. Drake and J. Klafter, Phys. Rev. Lett., 1986, 56,
197–200.
1 R. Klajn, J. F. Stoddart and B. A. Grzybowski, Chem. Soc.
Rev., 2010, 2039, 2203–2237.
2 Y. B. Zheng, Y.-W. Yang, L. Jensen, L. Fang, B. K. Juluri,
A. H. Flood, P. S. Weiss, J. F. Stoddart and T. J. Huang, Nano
Lett., 2009, 9, 819–825.
3 S. J. van der Molen, J. Liao, T. Kudernac, J. S. Agustsson,
L. Bernard, M. Calame, B. J. van Wees, B. L. Feringa and
C. Schönenberger, Nano Lett., 2008, 9, 76–80.
4 J. H. Yoon and S. Yoon, Phys. Chem. Chem. Phys., 2011, 13,
12900–12905.
28 A. van Blaaderen and A. Vrij, J. Colloid Interface Sci.,
1993, 156, 1–18.
29 J. Turkevich, P. C. Stevenson and J. Hillier, Discuss. Faraday
Soc., 1951, 11, 55–75.
30 K. L. McGilvray, J. Granger, M. Correia, J. T. Banks and
J. C. Scaiano, Phys. Chem. Chem. Phys., 2011, 13, 11914–11918.
31 C. Graf, D. L. J. Vossen, A. Imhof and A. van Blaaderen,
Langmuir, 2003, 19, 6693–6700.
5 R. Klajn, K. J. M. Bishop and B. A. Grzybowski, Proc. Natl.
Acad. Sci. U. S. A., 2007, 104, 10305–10309.
32 W. Stöber, A. Fink and E. Bohn, J. Colloid Interface Sci.,
1968, 26, 62–69.
This journal is © The Royal Society of Chemistry 2015
Catal. Sci. Technol.