Rh(I) complexes bearing N, N and N,P ligands anchored on glassy carbon electrodes: Toward recyclable hydroamination catalysts

Article abstract of DOI:10.1021/ja405783g

A series of N,N-donor ligands (bis(pyrazol-1-yl)methane (bpm), bis(N-methylimidazol-2-yl)methane (bim), 1-(phenylmethyl)-4-(1H-pyrazol-1-yl methyl)-1H-1,2,3-triazole (PyT)), and one N,P-donor ligand precursor (1-(3,5-dimethylpyrazol-1-yl)(2-bromoethane) (dmPyBr)) were synthesized and functionalized with aniline. Diazotization of the aniline into an aryl diazonium, using nitrous acid in aqueous conditions, was performed in situ such that the ligands could be reductively adsorbed onto glassy carbon electrode surfaces. The N,N-donor ligands (bpm, bim, PyT) were immobilized in a single step, while several steps were required to immobilize the N,P-donor ligand (dmPyP) to prevent oxidation of the phosphine group. The complexation of the anchored ligands with the metal complex precursor ([Rh(CO)₂(μ-Cl)] ₂) led to the formation of anchored Rh(I) complexes with each of the ligands (bpm, bim, PyT, dmPyP). X-ray photoelectron spectroscopy (XPS) confirmed the formation of the anchored ligands as well as the anchored complexes. The surface coverage of functionalized electrodes was estimated by means of cyclic voltammetry, and the nature of the coverage was close to being a monolayer for each immobilized complex. The anchored Rh(I) complexes were active as catalysts for the intramolecular hydroamination of 4-pentyn-1-amine to form 2-methyl-1-pyrroline.

Full text of DOI:10.1021/ja405783g

Article
pubs.acs.org/JACS
Rh(I) Complexes Bearing N,N and N,P Ligands Anchored on Glassy
Carbon Electrodes: Toward Recyclable Hydroamination Catalysts
Andrey A. Tregubov, Khuong Q. Vuong, Erwann Luais, J. Justin Gooding, and Barbara A. Messerle*
School of Chemistry, The University of New South Wales, Sydney 2052, Australia
S
* Supporting Information
ABSTRACT: A series of N,N-donor ligands (bis(pyrazol-1-yl)-
methane (bpm), bis(N-methylimidazol-2-yl)methane (bim), 1-
(phenylmethyl)-4-(1H-pyrazol-1-yl methyl)-1H-1,2,3-triazole
(PyT)), and one N,P-donor ligand precursor (1-(3,5-dimethylpyr-
azol-1-yl)(2-bromoethane) (dmPyBr)) were synthesized and func-
tionalized with aniline. Diazotization of the aniline into an aryl
diazonium, using nitrous acid in aqueous conditions, was performed
in situ such that the ligands could be reductively adsorbed onto glassy
carbon electrode surfaces. The N,N-donor ligands (bpm, bim, PyT)
were immobilized in a single step, while several steps were required to
immobilize the N,P-donor ligand (dmPyP) to prevent oxidation of
the phosphine group. The complexation of the anchored ligands with
the metal complex precursor ([Rh(CO)2(μ-Cl)]2) led to the
formation of anchored Rh(I) complexes with each of the ligands (bpm, bim, PyT, dmPyP). X-ray photoelectron spectroscopy
(XPS) confirmed the formation of the anchored ligands as well as the anchored complexes. The surface coverage of
functionalized electrodes was estimated by means of cyclic voltammetry, and the nature of the coverage was close to being a
monolayer for each immobilized complex. The anchored Rh(I) complexes were active as catalysts for the intramolecular
hydroamination of 4-pentyn-1-amine to form 2-methyl-1-pyrroline.
adsorption of aryl diazonium salts has been widely used for
surface functionalization, in particular for biosensor fabrica-
tion.18−20 Electron donation from the carbon electrode to the
aryl diazonium salt results in the formation of a phenyl radical
which is then covalently attached to the glassy carbon (GC)
surface (Scheme 1).
INTRODUCTION
■
Homogeneous catalysts are readily characterized and optimized
leading to highly effective catalysis, while heterogeneous
catalysts have the advantage that they are more readily
recovered and reused. The immobilization of homogeneous
catalysts on solid supports leads to catalysts that take advantage
of the best of both homogeneous and heterogeneous catalysis.
Ideally, homogeneous catalysts should be immobilized on
robust surfaces with robust linkers to give the most effective
systems. A wide range of supports have been used for catalyst
immobilization, including zeolites,1,2 polymers,3−5 carbon-
based materials,6,7 and silica.8−13 Of these, silica is the most
common support for the immobilization of metal complexes,
and although the (Si−O−Si) linkage used for immobilization is
quite robust, it can undergo hydrolysis under basic con-
ditions.14 Several important catalyzed reactions, including the
Heck reaction,15 utilize basic conditions,16,17 limiting the
breadth of application of silica supported catalysts. Carbon-
based materials are useful as catalyst supports as they are
typically very stable under harsh conditions including low or
high pH and high temperatures. The robust attachment of
homogeneous catalysts to inert carbon surfaces is an important
challenge, and the use of the stable covalent C−C bond as a
linkage between a carbon surface and catalyst ligand is an
important approach to catalyst immobilization.
Scheme 1. Attachment of Molecules to Carbon-Based
Surfaces via Reduction of Aryldiazonium Salts
Experimental and theoretical studies21,22 confirm that the
covalent C−C linkage formed between the surface and aryl
layer in this way is very strong, and it has been demonstrated
that it is highly resistant toward leaching even under harsh
conditions.18,19,23 It is therefore rather surprising that there are
only limited examples of this approach for immobilizing
The attachment of a variety of molecules to stable carbon-
Received: June 9, 2013
based surfaces through C−C bonds via the reductive
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ja405783g | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
catalysts on carbon surfaces. Stark and Reiser immobilized
TEMPO (2,2,6,6-tetramethylpiperidine-1-oxyl)24 and Cu(II)−
azabis(oxazoline) complexes25 on graphene-coated cobalt
nanoparticles. The immobilized TEMPO catalyst is a highly
active catalyst for the oxidation of primary and secondary
alcohols and does not lose activity over several catalytic runs.
More recently, Geneste et al.26 reported a Cu(II) complex
electrochemically immobilized on graphite felt and tested its
catecholase activity in continuous flow catalysis. In each of
these examples, the catalysis was performed under thermal
conditions without any electron transfer from or to the
supports. The attachment of molecules to carbon electrodes
leading to the formation of effective electrocatalysts using the
aryl diazonium method has also been demonstrated. For
example, De Lacey et al. attached Desulfovibrio gigas hydro-
genase to an aniline layer on a glassy carbon electrode,27 and
the covalently attached enzyme was active for dihydrogen
oxidation via direct electron transfer to the electrode and was
significantly more stable than the physisorbed enzyme.
The catalyzed hydroamination reaction is a significant
reaction that involves the addition of an N−H bond across
an unsaturated C−C bond and is a highly efficient approach to
the synthesis of many valuable nitrogen containing compounds
(amines, imines, and enamines).28 We have shown previously
that Rh(I) and Ir(I) complexes bearing a range of N,N-donor
and N,P-donor ligands are highly effective catalysts for
intramolecular hydroamination of alkynamines29−32 and alken-
amines.33,34 Here we report the immobilization of a selection of
these catalysts on glassy carbon surfaces. Bidentate pyrazolyl-
and imidazolyl-based ligands including bis(pyrazol-1-yl)-
methane (bpm) and bis(N-methylimidazol-2-yl)methane
(bim) and mixed pyrazol-1-yl-1,2,3-triazol-4-ylmethane (PyT)
and 1-(3,5-dimethylpyrazol-1-yl)-2-diphenylphosphinoethane
(dmPyP) ligands were immobilized on glassy carbon electrodes
and the corresponding supported Rh(I) complexes (Figure 1)
synthesized. The application of these immobilized complexes as
catalysts for the intramolecular hydroamination reaction is
demonstrated.
Figure 2. Aniline-functionalized ligands (1−3) and pro-ligand (4).
dmPyBr-NH2·HCl (4), respectively, in acidic media following reported
methods.35 A conventional three-electrode system was set up for the
modification of the GC plate (Goodfellow, product code VC000502)
using a platinum wire as the counter electrode, an Ag/AgCl 1.0 M KCl
electrode as the reference electrode, and finally the GC plate (1 × 1 ×
0.1 cm) as the working electrode. All potentials were reported versus
the Ag/AgCl reference electrode. The glassy carbon electrodes were
polished successively with 1.0, 0.3, and 0.05 μm alumina (Al2O3)
slurries and Milli-Q water on microcloth pads. The carbon plates were
thoroughly rinsed with Milli-Q after polishing and then soaked for 30
min in dichloromethane. Before modification, the electrodes were
dried under a nitrogen stream.
Aqueous sodium nitrite (100 mM, 100 μL) was added to a solution
of the aniline-functionalized ligands (1, 2, or 3) and pro-ligand (4) (1
mM), and aqueous hydrochloric acid (0.5 M, 10 mL) at 0 °C was
added to generate the aryl diazonium salt in the electrochemical cell in
situ. The mixture was degassed by bubbling with nitrogen and allowed
to react for about 10 min at 0 °C. The electrochemical reduction of the
in situ generated diazonium salt was carried out by scanning in a
potential range between 0.6 and −1.0 V versus Ag/AgCl for two cycles
at the scan rate of 100 mV × s−1 leading to the formation of the
immobilized ligands GC-bpm, GC-bim, GC-PyT, and GC-PyBr. After
this grafting step, the GC plates were rinsed with copious amounts of
Milli-Q water, acetonitrile, and ethanol, then dried under a stream of
nitrogen.
To prepare the immobilized phosphine−pyrazolyl ligand dmPyP on
GC, a stepwise approach was taken. A GC electrode functionalized
with the pro-ligand (4) was treated with 1 mL of diphenylphosphide
solution (0.1 M, generated by reacting 0.1 g Ph2PH (0.5 mmol) in 5
mL of tetrahydrofuran (THF) with 0.35 mL of n-butyllithium (1.6 M
in hexanes) at 0 °C overnight. The solvent was decanted, and the
electrode was washed with deoxygenated THF (3 × 2 mL),
deoxygenated water (3 × 2 mL), and then with deoxygenated
methanol (3 × 2 mL) to afford the immobilized ligand, GC-dmPyP.
Complexation of Immobilized Ligands with Rh(I). The
modified electrodes bearing the immobilized ligands (GC-bpm, GC-
bim, GC-PyT, and GC-dmPyP) were placed in a THF solution of
[Rh(CO)2(μ-Cl)]2 (0.5 mM, 2 mL) under an atmosphere of argon for
2 h. After decantation of the solvent, the electrodes were consecutively
rinsed three times with THF (5 mL) and methanol (5 mL) and then
dried under a stream of nitrogen.
Figure 1. Bidentate Rh(I) complexes with the N,N donor ligands
(bpm, bim, and PyT) and N,P donor ligand (dmPyP) immobilized on
a glassy carbon support.
Characterization of the Modified Surfaces. Surface analysis of
the modified GC plates was performed using XPS with an EscaLab 250
Xi (Thermo Scientific) spectrometer with a monochromated Al Kα
source. The measurements were recorded at a pressure of below 10−8
mbar in the analysis chamber and a takeoff angle normal to the sample
surface. The pass energy and the step size for the survey scan was 100
and 1.0 eV, respectively, and 20 and 0.1 eV, respectively, when
monitoring narrow scans. The analysis was focused on the center of
the modified area to avoid contamination from the edges of the
electrodes. Spectral analysis was performed using Avantage 4.73, the
background spectra were considered as Shirley type and curve fitting
was carried out using a mixture of Gaussian−Lorentzian functions.
GENERAL EXPERIMENTAL SECTION
■
Synthesis of Ligands. Three different aniline-functionalized
ligands bpm-NH2, bim-NH2, and PyT-NH2 (Figure 2 (1−3)) and
one ligand precursor dmPyBr-NH2·HCl were synthesized (Figure 2
(4)). The procedures for the synthesis of these compounds are
provided in the Supporting Information.
Glassy Carbon (GC) Plate Modification. The modifications of
GC plate surfaces with the N,N-donor ligands (bpm, bim and PyT)
and pro-ligand dmPyBr were achieved using the in situ generated aryl
diazonium salts of bpm-NH2 (1), bim-NH2 (2), PyT-NH2 (3), and
B
dx.doi.org/10.1021/ja405783g | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Attempts to confirm the presence of Rh-bound CO groups on the
surface using surface IR were not successful. This is likely to be due to
the fact that the complex layer on the surface is very thin (CV showed
the layer to be a monolayer) and that carbon materials strongly absorb
light in the UV−vis−IR regions masking IR signals. Carbon surfaces
also lack the ability to support electromagnetic field enhancement in
surface-enhanced Raman spectroscopy.36
Scheme 2. Synthesis of Anchored Rh(I) Complexes Using
Ligands Immobilized According to Scheme 1: (a) N,N-
Donor Ligands; (b) N,P Donor Ligand
Evaluation of Rh Content on Electrode Surface. The
evaluation of Rh content on the electrode surface was accomplished
using cyclic voltammetry (potential range from −0.5 and 1.5 V versus
Fc/Fc+) in acetonitrile using n-Bu4NBF4 (1 M) as the supporting
electrolyte. The cyclic voltammetry was acquired using an Autolab
PGStat12 Potentiostat. The glassy carbon rodlike electrodes (surface
area approximately 0.07 cm2/each) functionalized with Rh complexes
according to methodology described above were used as working
electrodes. Platinum wires were used both as the counter electrode and
the reference electrode. All potentials were reported versus Fc/Fc+
couple. The surface coverage (Γ) was evaluated using formula Γ = Q/
nFA, where Q is the charge measured during the oxidation of a surface-
bound Rh complex, F is the Faraday constant, n is the number of
electrons employed in the electrochemical reaction (for the oxidation
Rh(I) to Rh(II) n = 1), and A is the surface area of the working
electrode.37 The ratio of Rh coverage to surface area obtained for
rodlike electrodes was then used to calculate the Rh content for the
glassy carbon plates (surface area of one side is 1 cm2, the full surface
area is 2.4 cm2 in order to calculate the TON of catalyzed reaction
described below).
Catalyzed Hydroamination of 4-Pentyn-1-amine Using
Immobilized Complexes. A functionalized electrode was placed in
1.0 mL of anhydrous THF in an amber vial purged with nitrogen. 4-
Pentyn-1-amine (0.0200 0.0005 g, 0.024 mmol) was added, the vial
was sealed with a Teflon-lined screw-on lid, and the reaction was
heated at 60 °C for 72 h under an inert atmosphere. To optimize the
flow of the substrate past the catalyst on the electrode surface, the vial
was spun (300 rpm) for the entire reaction time (for a picture of the
setup, see Figure SI11 in the Supporting Information). The reaction
was then cooled to ambient temperature, and the conversion of
reductive adsorption of their aryl diazonium salts, which were
generated in situ in aqueous conditions using electrochemical
assistance as illustrated in Scheme 1. The cyclic voltammogram
of the diazonium salt generated from bpm-NH2 is shown in
Figure 3. The first scan contains the broad irreversible peak
1
substrate to product was determined using H NMR spectroscopy.
The turnover numbers (TON) was calculated using the following
formula:
[moles of product]
TON =
[moles of Rh]
where [moles of product] is the amount of 2-methyl-1-pyrroline
measured from 1H NMR spectra, and [moles of Rh] is the amount of
Rh on the electrode surface estimated by cyclic voltammetry.
RESULTS AND DISCUSSION
Figure 3. Cyclic voltammagram of diazonium salt bpm-N2+ generated
from bpm-NH2.
■
Synthesis of Aryl Amino-Functionalized Ligands. The
immobilization of metal complexes on glassy carbon surfaces
via C−C bond described here required the initial functionaliza-
tion of the N,N-donor ligands and the pro-ligand of the N,P-
donor ligands with aniline moieties (Figure 2, (1−4)). The
position of the aniline linker varied depending on the symmetry
and structure of a ligand. In the case of the bis N,N ligands
(bpm and bim) the linker was introduced at the CH2 bridge
between the azole rings (Figure 2 (1 and 2)). The ligand
pyrazol-1-yl-1,2,3-triazol-4-ylmethane (PyT) and pro-ligand 1-
(3,5-dimethyl-pyrazol-1-yl)-2-bromoethane (dmPyBr) were
functionalized with aniline at the N1 position of the 1,2,3-
triazolyl and the C4 position of the pyrazolyl ring, respectively
(Figure 2, (3 and 4)).
+
attributed to the reduction of the aryl diazonium salt bpm-N2 .
The second scan does not have any peaks showing that the
electrode surface is fully functionalized with the aryl layer which
blocks the electron transfer between the electrode surface and
the diazonium salt and confirms the complete surface coverage.
The cyclic voltammograms of diazonium salts generated from
bim-NH2, PyT-NH2, and PyBr-NH2·HCl can be found in the
Supporting Information (Figures SI1−SI3, Supporting In-
formation).
The anchored metal complexes GC-[Rh(bpm)(CO)2]+, GC-
[Rh(bim)(CO)2]+, and GC-[Rh(PyT)(CO)2]+ were synthe-
sized by the addition of a solution of [Rh(CO)2(μ-Cl)]2 in
THF to the immobilized ligands, in an analogous fashion to the
synthesis of the parent homogeneous complexes [Rh(N,N)-
(CO)2]+. The formation of the Rh(I) dicarbonyl complexes
was confirmed using XPS (see below); however, the precise
Immobilization of Functionalized Ligands and Prep-
aration of Immobilized Catalysts. The synthesis of the
anchored Rh(I) complexes of the N,N ligands (bpm, bim, PyT)
was accomplished in two steps (Scheme 2(a)). The function-
alized N,N ligands were grafted on the GC electrodes by the
C
dx.doi.org/10.1021/ja405783g | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
a
Table 1. XPS Data of Anchored Ligands and Pro-ligand dmPyBr
binding energy
N 1s
atomic ratio
N/Br
sample
Na
Nb
Nc
Br 3d
P 2p
Nb/Nc
N/P
GC-bpm
n/a
401.2
401.5
401.0
401.4
401.5
399.7
399.5
399.8
399.9
400.0
n/a
n/a
n/a
71.4
n/a
n/a
1.1
1.0
1.3
1.0
1.0
n/a
n/a
n/a
1.7
n/a
n/a
n/a
n/a
1.9
GC-bim
n/a
n/a
GC-PyT
401.9
n/a
n/a
GC-dmPyBr
GC-dmPyP
n/a
n/a
133.0
n/a
a
Binding energies are in eV.
Figure 4. XP spectra: (a) unmodified plate, (b) plate functionalized with bpm, (c) plate (b) after treatment with [Rh(CO)2(μ-Cl)]2.
nature of the counter-anions expected for the positively charged
immobilized complexes could not be established. The surface
shows the presence of little or no counterion by XPS, indicating
that the surface possesses little or no charge. We have also
shown previously with XPS of mixed layers containing charged
moieties that if the charge is completely neutralized there are
no counterions present.38 The unclear nature and location of
the counteranion after immobilization of a cationic complex has
also been reported by Laws et al.39 Laws et al. immobilized the
complex [CoCp(η5-C5H4N2)][PF6]2 on glassy carbon electro-
des, and XPS showed only a small amount of fluorine and none
of the phosphorus peaks expected for the PF6 counteranion.
The synthesis of the anchored Rh(I) complex of the N,P-
donor ligand dmPyP (4) was achieved following a modification
of the immobilization procedure used for the analogous N,N-
donor ligands (1−3) (Scheme 2(a)). The susceptibility of
phosphines to oxidation in the presence of nitrous acid,40 which
is necessary for diazotization of aniline moiety, makes the direct
immobilization of the functionalized N,P ligand undesirable,
and the addition of the phosphine arm of the ligand was
performed after the immobilization of the pro-ligand (4)
dmPyBr. The aniline hydrochloride-functionalized proligand
(4) dmPyBr was anchored on the glassy carbon electrode via
electrochemical reduction of its aryl diazonium salt (sample
GC-dmPyBr) and subsequent treatment of the electrode with a
solution of LiPPh2 led to the formation of the desired anchored
ligand GC-dmPyP (Scheme 2(b)). In contrast to the N,N
ligands mentioned above, the complexation between the
homogeneous counterpart dmPyP ligand and [Rh(CO)2(μ-
Cl)]2 results in the formation of a neutral Rh (I) complex
[Rh(dmPyP)(CO)Cl], with the CO cis and Cl trans relative to
the phosphine donor.41 The analogous immobilized complex
GC-[Rh(dmPyP)(CO)Cl] was formed on reaction of the
immobilized dmPyP ligand (GC-dmPyP) with a solution of
[Rh(CO)2(μ-Cl)]2 (Scheme 2(b)), as confirmed by XPS
(below).
X-ray Photoelectron Spectroscopy. X-ray photoelectron
spectroscopy (XPS) was used to characterize both the anchored
ligands and complexes. XPS data (binding energies and atomic
ratios) for the anchored ligands are summarized in Table 1. The
XP spectra shown in Figure 4 are for the immobilized ligand
GC-bpm and the corresponding immobilized Rh complex,
while the analogous spectra for the samples GC-bim, GC-
dmPyBr, and GC-dmPyP are included in the Supporting
Information. The XP survey spectra of the GC-bpm, GC-bim,
GC-dmPyBr, GC-dmPyP plates contain N 1s peaks at ca. 400
eV, which confirms the success of the immobilization of ligands.
The N 1s high resolution spectra of samples GC-bpm, GC-
bim, GC-dmPyBr, and GC-dmPyP display two peaks at ca. 401
and 400 eV (Table 1, Nb and Nc, respectively). The peak with
D
dx.doi.org/10.1021/ja405783g | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
a
Table 2. XPS Data Obtained for Anchored Complexes (N in the N/Rh Ratio Is the Sum of All N Species Present)
binding energy
atomic ratio
N 1s
Nb
sample
Na
Nc
Br 3d
P 2p
Rh 3d
Nb/Nc
N/Rh
P/Rh
Cl/Rh
GC-[Rh(bpm)(CO)2]
GC-[Rh(bim)(CO)2]
GC-[Rh(PyT)(CO)2]
GC-[Rh(dmPyP)(CO)Cl]
n/a
401.4
401.8
401.2
401.3
400.4
400.5
400.2
400.5
n/a
n/a
309.9
310.0
310.3
309.7
1.2
0.9
3
4.1
3.8
5.3
1.8
n/a
n/a
n/a
1.1
n/a
n/a
n/a
0.8
n/a
n/a
n/a
402.1
n/a
n/a
n/a
132.5
198.6
1.2
a
Binding energies are in eV.
lower binding energy was assigned to the pyridine like nitrogen
atom in each case (for GC-bpm, GC-dmPyBr, GC-dmPyP - Nb
is N2 of pyrazole; for GC-bim Nb is N3 of imidazole), whereas
the one at higher binding energy was assigned to the pyrrole-
like nitrogen of each ligand (for GC-bpm, GC-dmPyBr, GC-
dmPyP; Nc is N1 of pyrazole; for GC-bim Nc is N1 of
imidazole).42−45 The atomic percentages ratio for Nb and Nc is
very close to the expected ratio of 1:1.
The N 1s region of XP spectrum of the sample GC-PyT was
fitted with three peaks assigned to N1 of 1,2,3 triazole (401.9
eV), N2 of pyrazole and N1 of pyrazole (401.0 eV), and N2/
N3 of 1,2,3 triazole (399.8 eV).46,47
GC-bpm, GC-dmPyBr, GC-dmPyP; Nb is N2 of pyrazole; for
GC-bim Nb is N3 of imidazole). The sample GC-[Rh(PyT)-
(CO)2] has three different nitrogen emissions assigned to N1
of 1,2,3 triazole (402.1 eV), N3 of 1,2,3-triazole and N1,2 of
pyrazole (401.2 eV), and N2 of 1,2,3-triazole (400.2 eV). The
ratio of the atomic percentages of the last two nitrogen
environments (Nb/Nc in the Table 2) is close to the expected
ratio and correlates well with the assignment. The atomic
percentage ratios N:Rh for samples GC-[Rh(bpm)(CO)2] and
GC-[Rh(bim)(CO)2] (4.1 and 3.8, respectively) confirm the
stoichiometry of anchored Rh(I) complexes with bis-azole
ligands. In case of the sample GC-[Rh(PyT)(CO)2] the atomic
ratio N:Rh is close to 5, which satisfies the expected value for
the Rh(I) complex bearing mixed pyrazolyl-triazolyl ligand. The
difference of the binding energies of the N 1s level of Nc
between the anchored ligands (GC-bpm, GC-bim, GC-PyT,
and GC-dmPyP) and those treated with [Rh(CO2)(μ-Cl)]2
(GC-[Rh(bpm)(CO)2], GC-[Rh(bim)(CO)2], GC-[Rh(PyT)-
(CO)2], and GC-[Rh(dmPyP)(CO)Cl]) is ca. 0.5 eV. Figure 5
The XP survey spectrum of the sample GC-dmPyBr (Figure
SI6(b) in the Supporting Information) comprises the XPS
bromine signal at 71.4 eV, and the atomic content of bromine
emission correlated well with that expected for nitrogen (N:Br
= 2:1). The survey spectrum of the sample GC-dmPyP (Figure
SI6(c), Supporting Information), which was obtained by the
treatment of GC-dmPyBr with LiPPh2, contained a phosphorus
peak at 133.0 eV and no bromine signal, as expected. This
result indicates the full substitution of bromide with diphenyl
phosphide. The observed ratio of N:P of 1.9:1, was close to the
expected ratio of 2:1. The narrow peak for the phosphorus
signal indicates the presence of only one type of phosphorus
species, with a binding energy specific for phosphines.48 The
presence of the O 1s peak in all spectra of all of the surface
bounds ligands can be explained by contamination of the
surface due to physisorption of oxygen molecules and/or native
oxidized carbon species of the GC plates containing functional
groups such as phenols or carboxylic acids. The success of the
complexation of the anchored ligands with the Rh(I) precursor,
[Rh(CO)2(μ-Cl)]2, was also confirmed using XPS. The binding
energies for Rh 3d5/2, N 1s and P 2p and atomic ratios are given
in the Table 2. The survey spectra of samples GC-[Rh(bpm)-
(CO)2] (Figure 4(c)), GC-[Rh(bim)(CO)2] (Figure SI4(c) in
the Supporting Information), GC-[Rh(PyT)(CO)2] (Figure
SI5(c), Supporting Information), and GC-[Rh(dmPyP)(CO)-
Cl] (Figure SI 7(b), Supporting Information) contain the
emission of rhodium at ca. 310 eV (Rh 3d5/2). These values of
binding energies for Rh 3d indicate the presence of rhodium in
+1 oxidation state and its coordination to N and P donor
atoms.44,49
Figure 5. Overlay of the N 1s high-resolution scans of the XP spectra
of the samples GC-bpm and GC-[Rh(bpm)(CO)2].
illustrates the difference of the binding energies of N 1s level
between GC-bpm and GC-[Rh(bpm(CO)2]. This difference
indicates the binding between Rh and N as donor atoms of the
ligands
Surface Coverage. The cyclic voltammogram (CV) of the
anchored complex GC-[Rh(bpm)(CO)2] is shown in Figure 6.
The irreversible oxidation peak at ca. 0.4 V was assigned to the
oxidation of anchored Rh(I) to Rh(II). The second quasi-
reversible peak at ca. 0.8 V may be due to the couple Rh(III)/
Rh(II). The reversibility of the couple Rh(II)/Rh(I) and
Rh(III)/Rh(II) could be ligand-dependent.50 In contrast, the
cyclic voltammogram of the immobilized ligand bpm does not
contain any peaks.
The assignment of the N 1s peaks for the anchored
complexes GC-[Rh(bpm)(CO)2], GC-[Rh(bim)(CO)2], and
GC-[Rh(dmPyP)(CO)(Cl)] is similar to the assignment for
the anchored uncomplexed ligands (GC-bpm, GC-bim, and
GC-dmPyP, respectively). The peak at higher binding energy
was assigned to the pyrrole-like nitrogen of each ligand (for
GC-bpm, GC-dmPyBr, GC-dmPyP; Nc is N1 of pyrazole; for
GC-bim Nc is N1 of imidazole) whereas the one at lower
binding energies was assigned to pyridine-like nitrogen (for
The experimental value of Γ obtained from the integration of
the oxidation peak is 6.5 × 10−10 mol/cm2. The theoretical
value of Γ can be estimated based on the footprint of anchored
organometallic complex (see the Supporting Information). The
E
dx.doi.org/10.1021/ja405783g | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Scheme 3. Catalyzed Intramolecular Hydroamination of 4-
Pentyn-1-amine
product in the 1H NMR spectrum. For the immobilized
complexes of the N,N-donor ligands the conversion of 4-
pentyn-1-amine to 2-methyl-1-pyrroline was ca. 50%, and for
the immobilized complex with the N,P-donor ligand the
conversion was significantly reduced (ca. 15%). This mirrored
the observations for the analogous homogeneous complexes:
the homogeneous complexes based on N,N-donor ligands are
much faster than those with N,P-donor ligands.30 The fact that
full conversions of 4-pentyn-1-amine to 2-methyl-1-pyrroline
were not achieved with our immobilized catalysts suggests that
the catalyst may deactivate during the catalysis process, or that
the mass transport effects reduce the interaction of the
substrate in solution with the catalyst bound to the surface. If
the lack of full conversion is due to catalyst deactivation, the
exact nature of the deactivation pathway is not known at this
stage, and could be due to product inhibition or clustering of
metal centers. Hashmi et al. recently reported extremely
reactive mononuclear Au complexes with NHC bound ligands
as catalysts for the tandem hydroalkoxylation reaction and
made the observation that bulky ligands led to a more stable
catalyst, attributing this to the increased in site separation
between metal centers which led to lesser extent of catalyst
deactivation by clustering of metal centers.51 Any catalyst
deactivation is unlikely to result from the breaking of the very
strong C−C bond between the ligand and the surface, as has
been reported previously for species attached to carbon surfaces
using this approach.19,22,23,52 In this work we have shown (see
Catalyst Stability) that the ligand remains bound to the surface
during the course of a catalytic run.
Figure 6. CV using glassy carbon electrodes of GC-bpm and GC-
[Rh(bpm)(CO)2] as working electrodes.
molecular surface of the cationic fragment of complex
[Rh(bpm)(CO)2]BPh4 was roughly estimated to be ca. 37
Å2. Therefore the theoretical value of Γ for such molecule is
estimated to be ca. 4.5 × 10−10 mol/cm2. The experimental
surface coverage obtained from CV is compatible with the
presence of a monolayer. Similar surface coverages were
obtained for the other complexes under investigation (see
Table 3). The experimental values of Γ were used to estimate
the amount of Rh on the glassy carbon electrode with an
apparent surface area of ca. 2.4 cm2 (full available surface area)
in order to evaluate the turnover number (TON) in the
catalyzed hydroamination of 4-pentyn-1-amine.
Intramolecular Hydroamination of 4-Pentyn-1-amine
to 2-Methyl-1-pyrroline Catalyzed by Immobilized
Complexes. The catalytic activity of the immobilized
complexes GC-[Rh(N,N)(CO)2] and GC-[Rh(N,P)(CO)Cl]
was investigated using intramolecular hydroamination of 4-
pentyn-1-amine as a test reaction (Scheme 3). The cyclization
using the immobilized complexes was carried out in sample
tubes thoroughly sealed with Teflon tape, with 20 mg of
substrate in 1 mL of THF at 60 °C. To achieve effective
distribution of the homogeneous substrate in the reaction
media the sample tubes were spun for the entire reaction time
using a mechanical stirrer (see the Supporting Information).
This set up minimizes any mechanical damage of the GC
supports on stirring. The results of the catalyzed hydro-
amination using the immobilized complexes, as well as the
efficiency of their homogeneous counterparts, are given in
Table 4. The homogeneous catalysis given in the Table 4 was
carried out in NMR tubes equipped with Youngs Teflon valve,
using the comparable quantity (40 mg) of the substrate used
for the catalysis using immobilized complexes, with 1.5 mol %
of the corresponding homogeneous Rh complex in deuterated
THF solvent at 60 °C.
The TONs for the hydroamination of 4-pentyn-1-amine
catalyzed by the immobilized complexes were calculated using
1
the number of moles of product determined from H NMR
spectra divided by the number of moles of Rh on the electrode
surface estimated using cyclic voltammetry. The calculated
TONs for the immobilized complexes are significantly higher
(3.7 × 104−9.6 × 104) than those of their homogeneous
counterparts (Table 4). The two highest turn over frequencies
(TOFs) reported in the literature for this particular trans-
formation are attributed to [Me2Si(η5-Me4C5)(tBuN)U-
32
(NMe2)2]53 and [Ir(Me2PyP)(COD)]BPh4 at 1210 and
3100 h−1, respectively, with corresponding TONs of 333 and
71. In the vast majority of homogeneous metal catalyzed
hydroamination reactions between 0.5 and 5.0 mol % of catalyst
have been used which allows a maximum TON of between 20
and 200 and limits the comparison with the low concentration
of the surface immobilized catalysts under investigation here.
All immobilized Rh(I) complexes of the N,N-donor ligands
were active catalysts for the hydroamination of 4-pentyn-1-
amine, and 2-methyl-1-pyrroline was the only product
observed. The control experiment carried out using a blank
(unfunctionalized) electrode did not show any traces of the
Table 3. Experimental and Theoretical Surface Coverage for the GC Supports Modified with Organometallic Complexes (See
the Supporting Information for a Calculation Example)
sample
theoretical Γ, mol/cm2
experimental Γ, mol/cm2
amt of Rh on 2.4 cm2 glassy carbon electrode, mol
GC-[Rh(bpm)(CO)2]
GC-[Rh(bim)(CO)2]
GC-[Rh(PyT)(CO)2]
GC-[Rh(dmPyP)(CO)Cl]
4.5 × 10−10
4.4 × 10−10
3.8 × 10−10
3.2 × 10−10
5.6 × 10−10
4.8 × 10−10
5.1.0 × 10−10
4.0 × 10−10
1.34 × 10−9
1.15 × 10−9
1.22 × 10−9
0.96 × 10−9
F
dx.doi.org/10.1021/ja405783g | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Table 4. Comparison of the Efficiency of the Catalyzed Cyclization of 4-Pentyn-1-amine to 2-Methyl-1-pyrroline (Scheme 3)
a
Using the Immobilized Catalysts GC-[Rh(N,N)(CO)2] and GC-[Rh(dmPyP)(CO)Cl] and Their Homogeneous Counterparts
conversion S→P, % (h)
TON
b
c
b
c
catalyst
complex immobilized on GC electrode
homogeneous complex complex immobilized on GC electrode
homogeneous complex
[Rh(bpm)(CO)2]
[Rh(bim)(CO)2]
[Rh(PyT)(CO)2]
[Rh(dmPyP)(CO)Cl]
50 (72)
46 (72)
49 (72)
15 (72)
90 (12)57
98 (14)57
70 (5)34
89000
96000
96000
37000
6057
6557
4634
4832
d
72 (45),32
,
Conditions:
a
b
Conditions:[Rh(N,N)(CO)2]BPh4 for N,N ligands, [Rh(PyP)(CO)Cl] for N,P ligand. Conditions: 0.24 mmol of the substrate, ca. 5.4 × 10−4 mol
c
d
% of Rh. Conditions: 0.50 mmol of the substrate, 1.5 mol % of Rh. The analogous complex [Rh(PyP)(CO)Cl] was used.
Figure 7. Catalyst recycling: XP spectra of GC-[Rh(bpm)(CO)2] before and after catalysis: (a) N 1s; )b) Rh 3d.
The hydroamination of 4-pentyn-1-amine catalyzed by
heterogeneous catalysts has not previously been reported.
However, the catalyzed cyclization of 5-hexyn-1-amine using
supported catalysts has been reported. The silica supported Pd
complex silica/cis-[PdMe(Cl)(PMe3)2] can promote the
cyclization of 5-hexyn-1-amine with up to 85% conversion
after 20 h (0.88 mmol of substrate, 1.2 mol % of Pd) and a
surface may be achieved using carbon materials with higher
available surface areas, and this is currently under investigation
in our laboratory.
Catalyst Stability. The long-term stability of the
immobilized complex GC-[Rh(bpm)(CO)2] under catalytic
conditions was assessed by employing a sample of this complex
in three consecutive catalytic runs, and determining the amount
of cyclized product formed after each run. The surface
composition of the immobilized catalyst was also determined
using XPS after the first and third catalytic runs. After the
completion of the first and third catalytic runs, the Rh 3d peaks
have the same binding energies, which indicates that the Rh
oxidation state does not change during catalysis. Following the
first catalysis run, the intensity of the Rh 3d peak in the XP
spectrum decreased by 24% and conversion to product
decreased by 30% (from 23% to 16% conversion), while the
intensity of the N 1s peak in the XP spectrum only decreased
by 5% (Figure 7). Following the third run, the intensity of the
Rh 3d peak decreased by a further 9% and the level of
conversion to product decreased by a further 56% to 7%, while
the intensity of the N 1s peak did not change significantly
(<1%). With a change in N/Rh ratio from 4:1 prior to catalysis
to 5.3:1 after the first run, it is clear that the Rh is leaching from
the surface during catalysis, although the insignificant changes
on the XPS data for N 1s shows that the ligand itself remains
bound to the surface and completely intact. These observations
are consistent with previously reported work showing that the
pyrazolyl ligands are relatively weak metal donors and can
decoordinate from Rh in solution.57,58 Rh leaching due to N−
Rh bond cleavage has also been observed previously from the
([Rh(COD)NH2(CH2)2NH(CH2)2Si(OMe)3]BF4) complex
calculated TON of 72.54 Muller et al. also studied the
̈
cyclization of 5-hexyn-1-amine using late transition metal
(Zn, Rh, Cu) ion-exchanged BEA zeolites.55 The conversions of
5-hexyn-1-amine to cyclized product achieved using these
catalysts were 85−100% after 8 h (0.53 mmol % of substrate,
1.0 mol % of metal), and the TON values (80−100) for these
catalysts were comparable to those achieved using the silica
supported Pd complex in the previous example. In both cases
the activity of the heterogeneous supported catalyst was higher
than that of a homogeneous counterpart, however, the TONs
were relatively low.
The amounts of Rh used in the cyclization of 4-pentyn-1-
amine assisted with immobilized Rh complexes in the present
work were relatively small (approximately 1.3 × 10−9 mol, or
5.4 × 10−4 mol %) in comparison with Rh loading used for
hydroamination catalyzed with homogeneous Rh complexes.
This leads to incomplete conversion of the full 20 mg of
substrate, but with extremely high TONs. These high levels of
catalyst efficiency for the immobilized complexes highlight the
potential of the methodology used here for catalyst
immobilization. The relatively high TON reported here are
consistent with high turnover numbers reported for the surface-
bound enzyme glucose oxidase relative to those of the glucose
oxidase catalysis in solution.56 A higher Rh loading on the
G
dx.doi.org/10.1021/ja405783g | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
which had been immobilized on carbon nanotubes via Si−O−C
linkage.59 It should be possible to reduce the metal leaching
observed for the catalysts under investigation here by using
more strongly binding metal donors in the ligands, such as
−-heterocyclic carbenes.
Research Postgraduate Scholarship). Financial support from
UNSW is gratefully acknowledged. This research was
supported under Australian Research Council’s Discovery
Projects funding scheme (project numbers DP1094564 and
DP130101838).
CONCLUSIONS
REFERENCES
■
■
Using the reductive adsorption of aryl diazonium salts we
immobilized a series of N,N-donor ligands and an N,P−donor
ligand on glassy carbon electrodes. The N,N-donor ligands
were immobilized electrochemically in a single step, while the
immobilization of the N,P-donor ligand required several steps.
The complexation between immobilized ligands and [Rh-
(CO)2(μ-Cl)]2 leads to anchored Rh(I) complexes of these
ligands. The success of ligand immobilization as a monolayer
on the GC surface and the stoichiometry of the anchored
complexes were confirmed by XPS and also by cyclic
voltammetry. Anchored Rh(I) complexes of the N,N-donor
ligands were active catalysts for the intramolecular hydro-
amination of 4-pentyn-1-amine, forming 2-methyl-1-pyrroline,
with extremely high TON (up to 9.6 × 104). The immobilized
catalysts with the N,N-donor ligand could be recycled, although
there was evidence of Rh leaching from the surface. The
catalyzed reaction is limited by the relatively small surface area
of glassy carbon, and the use of materials with high specific
surface area (e.g., porous carbon, carbon nanotubes, graphene)
may resolve this issue, and the synthesis of immobilized Rh(I)
complexes by means of spontaneous adsorption of aryl
diazonium salts on these materials is currently being conducted
in our laboratory.
(1) Ayala, V.; Corma, A.; Iglesias, M.; Rincon, J. A.; Sanchez, F. J.
Catal. 2004, 224, 170.
(2) Dominguez, I.; Fornes, V.; Sabater, M. J. J. Catal. 2004, 228, 92.
(3) Takeda, K.; Oohara, T.; Shimada, N.; Nambu, H.; Hashimoto, S.
Chem.Eur. J. 2011, 17, 13992.
(4) Kollhofer, A.; Plenio, H. Chem.Eur. J. 2003, 9, 1416.
(5) Benaglia, M.; Puglisi, A.; Cozzi, F. Chem. Rev. 2003, 103, 3401.
(6) Roman-Martinez, M. C.; Diaz-Aunon, J. A.; De Lecea, C. S. M.;
Alper, H. J. Mol. Catal. A: Chem. 2004, 213, 177.
(7) Lemus-Yegres, L. J.; Such-Basanez, I.; Roman-Martinez, M. C.;
Salinas-Martinez de Lecea, C. Appl. Catal., A 2007, 331, 26.
(8) Hamasaka, G.; Kawamorita, S.; Ochida, A.; Akiyama, R.; Hara, K.;
Fukuoka, A.; Asakura, K.; Chun, W. J.; Ohmiya, H.; Sawamura, M.
Organometallics 2008, 27, 6495.
(9) Zhou, W.; Li, Y. M.; He, D. H. Appl. Catal., A 2010, 377, 114.
(10) Sahoo, S.; Kumar, P.; Lefebvre, F.; Halligudi, S. B. J. Mol. Catal.
A: Chem. 2007, 273, 102.
(11) Hu, R. H.; Zha, L. F.; Cai, M. Z. Catal. Commun. 2010, 11, 563.
(12) Bourque, S. C.; Maltais, F.; Xiao, W. J.; Tardif, O.; Alper, H.;
Arya, P.; Manzer, L. E. J. Am. Chem. Soc. 1999, 121, 3035.
(13) Miao, Q. J.; Fang, Z. P.; Cai, G. P. Catal. Commun. 2003, 4, 637.
(14) Glasspoole, B. W.; Webb, J. D.; Crudden, C. M. J. Catal. 2009,
265, 148.
(15) Heck, R. F.; Nolley, J. P., Jr. J. Org. Chem. 1972, 37, 2320.
(16) Miyaura, N.; Suzuki, A. Chem. Rev. (Washington, D.C.) 1995, 95,
2457.
(17) Sonogashira, K. J. Organomet. Chem. 2002, 653, 46.
(18) Pinson, J.; Podvorica, F. Chem. Soc. Rev. 2005, 34, 429.
(19) Gooding, J. J. Electroanalysis 2008, 20, 573.
(20) McCreery, R. L. Chem. Rev. 2008, 108, 2646.
(21) Jiang, D. E.; Sumpter, B. G.; Dai, S. J. Phys. Chem. B 2006, 110,
23628.
This is the first time that Rh(I) complexes have been
anchored onto carbon surfaces via robust C−C bonds. The
surface attachment significantly increases the catalytic activity of
the Rh(I) complexes, leading to TONs that are orders of
magnitude greater than those obtained for the homogeneous
counterparts. This work could have significant implications in
approaches to the anchoring of homogeneous catalysts on
surfaces and lead to improvements in industrial applications of
immobilized homogeneous catalysts.
(22) Toupin, M.; Belanger, D. J. Phys. Chem. C 2007, 111, 5394.
(23) Creager, S. E.; Liu, B.; Mei, H.; DesMarteau, D. Langmuir 2006,
22, 10747.
(24) Schatz, A.; Grass, R. N.; Stark, W. J.; Reiser, O. Chem.Eur. J.
ASSOCIATED CONTENT
* Supporting Information
Synthesis of aniline-functionalized N,N ligands and N,P pro-
ligand. CV of diazonium salts generated from bim-NH2, PyT-
NH2, and dmPyBr-NH2·HCl. XPS of GC-bim, GC-PyT, GC-
dmPyBr, GC-dmPyP, and their respective Rh(I) complexes.
This material is available free of charge via the Internet at
http://pubs.acs.org.
2008, 14, 8262.
■
(25) Schatz, A.; Grass, R. N.; Kainz, Q.; Stark, W. J.; Reiser, O. Chem.
Mater. 2010, 22, 305.
(26) Marion, R.; Muthusamy, G.; Geneste, F. J. Catal. 2012, 286,
S
266.
(27) Ruediger, O.; Abad, J. M.; Hatchikian, E. C.; Fernandez, V. M.;
De Lacey, A. L. J. Am. Chem. Soc. 2005, 127, 16008.
(28) Muller, T. E.; Hultzsch, K. C.; Yus, M.; Foubelo, F.; Tada, M.
Chem. Rev. 2008, 108, 3795.
(29) Dabb, S. L.; Ho, J. H. H.; Hodgson, R.; Messerle, B. A.; Wagler,
J. Dalton Trans. 2009, 634.
(30) Burling, S.; Field, L. D.; Messerle, B. A.; Rumble, S. L.
Organometallics 2007, 26, 4335.
(31) Field, L. D.; Messerle, B. A.; Vuong, K. Q.; Turner, P. Dalton
Trans. 2009, 3599.
AUTHOR INFORMATION
Corresponding Author
b.messerle@unsw.edu.au
■
Notes
The authors declare no competing financial interest.
(32) Field, L. D.; Messerle, B. A.; Vuong, K. Q.; Turner, P.; Failes, T.
Organometallics 2007, 26, 2058.
(33) Nguyen, T. O.; Man, B. Y. W.; Hodgson, R.; Messerle, B. A.
Aust. J. Chem. 2011, 64, 741.
(34) Hua, C.; Vuong, K. Q.; Bhadbhade, M.; Messerle, B. A.
Organometallics 2012, 31, 1790.
(35) Liu, G. Z.; Chockalingham, M.; Khor, S. M.; Gui, A. L.;
Gooding, J. J. Electroanalysis 2010, 22, 918.
ACKNOWLEDGMENTS
■
We are grateful to Dr. Gavin Edwards for help with developing
the methodology for the ligand functionalization and to Dr.
Nadim Darwish for discussions of the electrochemistry
experiments. We thank Professor Mike George and his group
at the University of Nottingham for their help. A.T. thanks the
Australian Government for the award of an IPRS (International
(36) Anariba, F.; Viswanathan, U.; Bocian, D. F.; McCreery, R. L.
Anal. Chem. 2006, 78, 3104.
H
dx.doi.org/10.1021/ja405783g | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
(37) Bard, A. J.; Faulkner, L. R. Electrochemical Methods:
Fundamentals and Applications; Wiley: New York, 1980.
(38) Gui, A. L.; Yau, H. M.; Thomas, D. S.; Chockalingam, M.;
Harper, J. B.; Gooding, J. J. Langmuir 2013, 29, 4772.
(39) Laws, D. R.; Sheats, J.; Rheingold, A. L.; Geiger, W. E. Langmuir
2010, 26, 15010.
(40) Bakac, A.; Schouten, M.; Johnson, A.; Song, W. J.; Pestovsky,
O.; Szajna-Fuller, E. Inorg. Chem. 2009, 48, 6979.
(41) Burling, S.; Field, L. D.; Messerle, B. A.; Khuong, Q. V.; Turner,
P. Dalton Trans. 2003, 4181.
(42) Katrib, A.; Elrayyes, N. R.; Alkharafi, F. M. J. Electron Spectrosc.
Relat. Phenom. 1983, 31, 317.
(43) Shen, C.; Haryono, M.; Grohmann, A.; Buck, M.; Weidner, T.;
Ballav, N.; Zharnikov, M. Langmuir 2008, 24, 12883.
(44) Strzelec, K.; Wasikowska, K.; Cypryk, M.; Pospiech, P. J. Appl.
Polym. Sci. 2012, 124, 3538.
(45) Vasnin, S. V.; Cetrullo, J.; Geanangel, R. A.; Bernal, I. Inorg.
Chem. 1990, 29, 885.
(46) Ciampi, S.; Boecking, T.; Kilian, K. A.; Harper, J. B.; Gooding, J.
J. Langmuir 2008, 24, 5888.
(47) James, M.; Ciampi, S.; Darwish, T. A.; Hanley, T. L.; Sylvester,
S. O.; Gooding, J. J. Langmuir 2011, 27, 10753.
(48) Cai, M. Z.; Sha, J. C.; Xu, Q. H. J. Mol. Catal. A: Chem. 2007,
268, 82.
(49) Evans, J.; Hayden, B. E.; Newton, M. A. Surf. Sci. 2000, 462, 169.
(50) Feller, M.; Ben-Ari, E.; Gupta, T.; Shimon, L. J. W.; Leitus, G.;
Diskin-Posner, Y.; Weiner, L.; Milstein, D. Inorg. Chem. 2007, 46,
10479.
(51) Blanco, J. M. C.; Boehling, C. R. N.; Serrano-Becerra, J. M.;
Hashmi, A. S. K. Angew. Chem., Int. Ed. 2013, 52, 7963.
(52) Allongue, P.; Delamar, M.; Desbat, B.; Fagebaume, O.; Hitmi,
R.; Pinson, J.; Saveant, J. M. J. Am. Chem. Soc. 1997, 119, 201.
(53) Stubbert, B. D.; Stern, C. L.; Marks, T. J. Organometallics 2003,
22, 4836.
(54) Richmond, M. K.; Scott, S. L.; Alper, H. J. Am. Chem. Soc. 2001,
123, 10521.
(55) Penzien, J.; Haessner, C.; Jentys, A.; Kohler, K.; Muller, T. E.;
Lercher, J. A. J. Catal. 2004, 221, 302.
(56) Gooding, J. J.; Erokhin, P.; Hibbert, D. B. Biosens. Bioelectron.
2000, 15, 229.
(57) Burling, S.; Field, L. D.; Messerle, B. A.; Turner, P.
Organometallics 2004, 23, 1714.
(58) Ho, J. H. H.; Wagler, J.; Willis, A. C.; Messerle, B. A. Dalton
Trans. 2011, 40, 11031.
(59) Lemus-Yegres, L. J.; Perez-Cadenas, M.; Roman-Martinez, M.
C.; Salinas-Martinez de Lecea, C. Microporous Mesoporous Mater. 2011,
139, 164.
I
dx.doi.org/10.1021/ja405783g | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX