This	document	is	the	Accepted	Manuscript	version	of	a	Published	Work	that	appeared	in	final	form	in	ACS	Catalysis,	
copyright	©	American	Chemical	Society	aAer	peer	review	and	technical	ediCng	by	the	publisher.	To	access	the	final	
edited	and	published	work	see	hEps://pubs.acs.org/doi/abs/10.1021/acscatal.6b03441	
Research Article
pubs.acs.org/acscatalysis

Gold−Ligand-Catalyzed Selective Hydrogenation of Alkynes into cisAlkenes via H2 Heterolytic Activation by Frustrated Lewis Pairs
Jhonatan L. Fiorio,† Núria López,‡ and Liane M. Rossi*,†
†

Downloaded via UNIV ROVIRA I VIRGILI on April 12, 2019 at 13:15:51 (UTC).
See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

Departamento de Química Fundamental, Instituto de Química, Universidade de São Paulo, Av. Prof. Lineu Prestes, 748, 05508-000
São Paulo, SP, Brazil
‡
Institute of Chemical Research of Catalonia, ICIQ, The Barcelona Institute of Science and Technology, Av. Països Catalans 16,
43007 Tarragona, Spain
S
* Supporting Information

ABSTRACT: The selective hydrogenation of alkynes to alkenes is an important
synthetic process in the chemical industry. It is commonly accomplished using
palladium catalysts that contain surface modifiers, such as lead and silver. Here we
report that the adsorption of nitrogen-containing bases on gold nanoparticles results
in a frustrated Lewis pair interface that activates H2 heterolytically, allowing an
unexpectedly high hydrogenation activity. The so-formed tight-ion pair can be
selectively transferred to an alkyne, leading to a cis isomer; this behavior is controlled
by electrostatic interactions. Activity correlates with H2 dissociation energy, which
depends on the basicity of the ligand and its reorganization on activation of hydrogen.
High surface occupation and strong Au atom−ligand interactions might affect the
accessibility and stability of the active site, making the activity prediction a
multiparameter function. The promotional effect found for nitrogen-containing bases
with two heteroatoms was mechanistically described as a strategy to boost gold
activity.
KEYWORDS: gold, hydrogenation, alkyne, piperazine, frustrated Lewis pair

■

INTRODUCTION

comparison to the enormous number of studies with Au NP
catalysts for oxidations, there are fewer studies in hydrogenations and gold was often considered only a modest
hydrogenation catalyst.11−16 Ultrafine gold particles supported
on alumina were regarded as a unique metal for the selective
hydrogenation of acetylene into ethylene, and gold received
attention as a selective catalyst for this gas-phase reaction.17,18
More recently, a new interest emerged in the selective
hydrogenation of carbonyl or nitro groups.19−22 The main
virtue was gold’s unique chemoselectivity, which is controlled
by the thermodynamics of the adsorption of the organic
reactants and products.18,23 In general, the low activity
associated with gold is a consequence of poor H2 dissociation
on gold,24,25 though nanostructuring can improve activity (H2
dissociation) at structural defects on nanoparticles or nanoporous materials.18
The use of molecular modifiers or auxiliary ligands is a wellknown strategy to enhance selectivity in heterogeneous
catalysis, mainly as a result of ensemble control,26−28 and
similarly capping ligands (e.g., nitrogen-containing bases) were
found to influence the catalytic properties of metal nanoparticles.29−35 Quinoline is the most frequently used modifier,
either by itself or with lead, for the selective hydrogenation of

Catalytic hydrogenation is a key process technology in the
petrochemical industry and has evolved into an important
transformation for the manufacture of pharmaceuticals and fine
chemicals, replacing stoichiometric chemical reduction methods
that generate large amounts of toxic waste. Catalytic hydrogenations are usually accomplished by the activation of
molecular hydrogen via homolytic H2 dissociation on metal
surfaces, such as Rh, Pt, Pd, and Ni.1,2 Selective hydrogenations
with discrimination between reducible functional groups, such
as carbon−carbon triple and double bonds, are very
challenging. Palladium remains the catalyst of choice for
many hydrogenations, but the selective conversion of alkynes
into cis-alkenes requires the use of a palladium catalyst partially
poisoned with a metal such as lead (Lindlar catalysts).3
Auxiliary ligands containing N, S, and O groups have been
combined with Pd and a secondary metal (Pb, V) to improve
selectivity. The presence of toxic lead and catalyst deactivation
are driving forces for the search of lead-free Lindlar catalyst
replacements for alkyne to cis-alkene hydrogenations. More
active and selective palladium catalysts have appeared, but there
is still room for the development of catalyst systems based on
alternative designs.4−6
Gold nanoparticles (Au NPs) have attracted a significant
amount of interest as catalysts, due to their exceptional
selectivity and surprisingly high activity, which is not replicated
by other metals, for reactions such as CO oxidation.7−10 In
© 2017 American Chemical Society

Received: December 4, 2016
Revised: February 22, 2017
Published: March 13, 2017
2973

DOI: 10.1021/acscatal.6b03441
ACS Catal. 2017, 7, 2973−2980

ACS Catalysis

Research Article
36,37

with deionized water. The recovered material was dispersed in
10 mL of water, and an aqueous NaBH4 solution (10 mL, 0.2
mol L−1) was added dropwise with vigorous stirring. After
further stirring for 20 min, the solid was recovered by filtration,
thoroughly washed with water, and dried under vacuum. Both
Au/SiO2 and Au/Fe3O4@SiO2 catalysts contain 3.4 wt % of
gold as determined by FAAS.
Preparation of Au/Support (TiO2, CeO2, and Al2O3 ). Au
NPs supported on TiO2, CeO2, and Al2O3 were prepared by
the DP method, as reported in the literature.45,46 Typically, 1.0
g of oxide support was added to 100 mL of an aqueous solution
of HAuCl4 (2.1 × 10−3 mol L−1) and urea (0.42 mol L−1). The
mixture was heated to 80 °C with vigorous stirring for 4 h,
centrifuged, washed, and dispersed in 10 mL of water. Then, an
aqueous NaBH4 solution (10 mL, 0.2 mol L−1) was added
dropwise with vigorous stirring. After further stirring for 20
min, the solid was recovered by filtration, thoroughly washed
with water, and dried under vacuum. The Au loadings were as
follows: Au/TiO2, 3.4 wt %; Au/CeO2, 3.5 wt %, Au/Al2O3, 2
wt %.
General Procedure for Catalytic Hydrogenation of
Alkynes. A typical procedure for the semihydrogenation of
alkynes is as follows: alkyne (1 mmol), N-containing base (1
mmol), Au/SiO2 (1 mol %), and 2 mL of solvent were placed
in a modified Fischer−Porter 100 mL glass reactor. The reactor
was purged five times with H2, leaving the vessel at 6 bar. The
resulting mixture was vigorously stirred using a Teflon-coated
magnetic stirrer, and the temperature was maintained with an
oil bath on a hot stirring plate connected to a digital controller
(ETS-D5 IKA). After the desired time, the catalyst was
removed by centrifugation and the products were analyzed by
GC with an internal standard to determine the conversion of
alkyne and the selectivity for alkene. For kinetic studies, the
reaction was performed using 4 times the amount of each
component given above and aliquots were collected and
analyzed by GC at different reaction time intervals. To
determine the isolated yield of the obtained products, after
the reaction was complete, the crude reaction mixture was
transferred to a extraction funnel and treated with HCl (20 mL,
10% v/v). The aqueous phase was extracted three times with 5
mL of dichloromethane. The organic layers were dried with
anhydrous MgSO4, and the solvent was removed under reduced
pressure, to give the corresponding desired alkene. 1H NMR
confirmed the purity of the isolated products. Details on other
catalyst tests (catalyst recycling, scale-up, isotope H2−D2
exchange, isotopic effect) are given in the Supporting
Information.
Computational Details. Density functional theory calculations were performed on slabs representing the gold surface
with the VASP code.47,48 The RPBE was the functional of
choice,49 and due to the large size of the molecules dispersion
terms were introduced through our modified version of the D2
formulation50 for the metal. The inner electrons were replaced
by PAW,51 and the valence electrons were expanded in plane
waves with a kinetic cutoff energy of 450 eV. The Au(111)
surface was represented by a p(4 × 4) supercell and four metal
layers (>12 Å vacuum) for the adsorption and reaction. The kpoint sampling was 3 × 3 × 1 in the Monkhorst−Pack
scheme.52 Transition states were located through the climbing
image nudged elastic band (ci-NEB) method;53 the nature of
the structures was assessed through the vibrational analysis with
a 0.02 Å step.

alkynes performed on Pd Lindlar and Pt catalysts.
Nitrogen-containing bases have also been explored as
promoters for hydrogenation reactions by gold via H2
activation (quinoline38 and pyridine39) and using other
hydrogen sources, such as PhMe2SiH40 and HCOOH,41 in
combination with reaction modifiers. The use of H2 as the
hydrogen source is highly desired, as it avoids the production of
residues, but the channels for activation of H2 on gold are not
yet fully understood.
Herein, we report the unique promotion effect of nitrogencontaining bases on the catalytic activity of gold nanoparticles.
The starting inactive gold nanoparticles supported on silica
become highly active for the selective hydrogenation of alkyne
into cis-alkenes. The hydrogenation proceeded smoothly and
was fully selective using H2 as the hydrogen source under
relatively mild conditions (80 °C, 6 bar of H2). According to
density functional theory calculations, molecular hydrogen
dissociation occurs at the ligand−gold interface, generating a
tight-ion pair that can be selectively transferred to the adsorbed
alkyne in a cis configuration controlled by electrostatic
interactions.

■

EXPERIMENTAL SECTION
Materials and Methods. All reagents used for the support
and catalyst preparation were of analytical grade, purchased
from Sigma-Aldrich, and were used as received.
Tetrachloroauric(III) acid was purchased as a 30 wt % aqueous
solution in dilute HCl (Sigma-Aldrich). Alkynes, alkenes,
amines, and standards for GC analysis were purchased from
Sigma-Aldrich at the highest grade available and used as
received. Liquid amines used as additives were purified by
standard methods just before use. Ethanol (99.5%) and DMF
(99%) were purchased from Synth (Brazil) and used as
received. Toluene was purchased from Synth (Brazil) and dried
before use. The glass reactor was thoroughly cleaned with aqua
regia (HCl/HNO3 3/1 v/v), rinsed with copious amounts of
pure water, and then dried in an oven prior to use. Gold
content in the catalyst was measured by FAAS analysis, on a
Shimadzu AA-6300 spectrophotometer using an Au hollow
cathode lamp (Photron). Metal leaching into the supernatant
solution was measured by ICP-AES, performed on a Spectro
Arcos ICP AES instrument. UV−vis spectra were recorded on a
Shimadzu UV-1700 instrument. TEM analyses were performed
with a JEOL 2100 instrument. X-ray diffraction (XRD) of the
samples was recorded using a Rigaku miniflex diffractometer
with Cu Kα radiation (λ = 1.54 Å) at a 2θ range from 20 to 90°
with a 0.02° step size and measuring time of 5 s per step. GC
analyses were carried out with a Shimadzu GC-2010 instrument
equipped with a RTx-Wax column (30 m × 0.25 mm × 0.25
mm) and an FID detector. Method: Ti = 40 °C, Tf = 200 °C,
25 min, T FID, and SPLIT = 200 °C. Internal standard:
biphenyl. The conversion and selectivity were determined from
the total amount of detected products and reactant.42
Preparation of Au/SiO2 and Au/Fe3 O4 @SiO2. Silica
nanospheres (SiO2) and silica-coated magnetite nanoparticles
(Fe3O4@SiO2) were obtained by means of a reverse microemulsion process and functionalized with (3-aminopropyl)triethoxysilane, as reported elsewhere.43 Gold nanoparticles
were prepared by the impregnation−reduction method, as
reported in the literature.44 Typically, 1.0 g of the aminomodified support was added to 20 mL of HAuCl4 aqueous
solution (1.3 × 10−2 mol L−1) with continuous stirring. After 30
min, the mixture was filtered and the solid thoroughly washed
2974

DOI: 10.1021/acscatal.6b03441
ACS Catal. 2017, 7, 2973−2980

ACS Catalysis

Research Article

■

RESULTS AND DISCUSSION
Screening of a series of N-containing bases (B1−B19) (see
Table S1 in the Supporting Information) with a broad range of
pKa values and chemical structures (primary, secondary, and
tertiary amines and also heteroaromatic and cycloaliphatic
amines) for the hydrogenation of phenylacetylene 1a over Au
NPs supported on silica (Au/SiO2) with an average diameter of
2.3 ± 0.6 nm (see Figure S1 in the Supporting Information),
provided the results shown in Figure 1.

obtained activity is not related to any leaching of the
catalytically active metal (no remaining activity in the
supernatant), which suggests that gold surfaces containing
adsorbed auxiliary ligands catalyze the hydrogenation reaction.
In order to further examine the stability of the Au NP catalyst
upon recycles, a magnetic version of the silica support was
synthesized to avoid any mass loss during the recycles. The
magnetic catalyst (Au/Fe3O4@SiO2) was easily recovered by
magnetic separation, and the recovered solid remains active for
five successive reactions using fresh portions of 1a, piperazine
(B10), and solvent (Figure S2 in the Supporting Information).
ICP AES analysis of the liquid products isolated after magnetic
separation showed negligible amounts of gold (detection limit:
0.10 ppm). These results corroborate the filtration test,
suggesting that the catalytically active metal remains at the
support and does not leach into the solution.
The Au/SiO2 catalyst was further tested in the hydrogenation
of 1a using different solvents and temperatures. The “green”
and protic solvent ethanol gave the best catalytic results (full
conversion), while aprotic solvents such as toluene and DMF
were slightly less efficient (Table S2 in the Supporting
Information). Decreasing the reaction temperature to 80 °C
did not affect the conversion and selectivity, while further
reduction of the temperature to 60 °C caused a decrease in
conversion to 53% (Table S2). On the basis of the above
experiments, we chose ethanol as solvent, piperazine (B10) as
the auxiliary ligand, and a temperature of 80 °C as the
optimized conditions for the semihydrogenation reaction. We
further examined Au NPs immobilized on different supports,
such as Au/Al2O3, Au/TiO2, and Au/CeO2, under the
aforementioned conditions (Table S2). TiO2 provided the
same activity and selectivity as for SiO2, but the CeO2 support
was detrimental to activity while the Al2O3 support was
detrimental to selectivity. Next, we turned our attention to
analyze the effect of adding different amounts of piperazine
(B10) (Figure 2). A very low conversion of 1a (13%) was
obtained using 1a:ligand:Au 100:1:1 (Figure 2a), but full

Figure 1. Screening of nitrogen-containing bases for phenylacetylene
(1a) hydrogenation with Au/SiO2 catalyst. Reaction conditions: 1
mmol of 1a, 0.01 mmol of Au (50 mg of Au/SiO2), 1 mmol of amine,
2 mL of ethanol, 100 °C, 6 bar of H2, 24 h.

Au/SiO2 itself performed poorly (conversion <1%), but it is
highly selective and provides exclusively the alkene 1b. The Au/
SiO2 catalyst becomes active in the presence of some of the Ncontaining bases studied, but there was no correlation with the
amine pKa values. In general, nitrogen-containing bases with
two heteroatoms (pyrazine (B1), 2,2-bipyridine (B3), morpholine (B7), 1,4-diazabicyclo[2.2.2]octane (DABCO; B8),
piperazine (B10), and imidazole (B17)) exhibited a higher
promotion effect in the hydrogenation of 1a by gold than
species with one nitrogen atom. Morpholine (B7) was able to
give full conversion of 1a, but a competing hydroamination
process39 occurred and the ketone product was formed in the
crude products (selectivity for acetophenone 1d: 60%).
Piperazine (B10) provided full conversion of 1a and 100%
selectivity to the alkene 1b, giving the best catalytic result
among the amines tested. It is worth mentioning that a blank
experiment with B10 and without the catalyst or with the
support without Au NPs showed no conversion of 1a (Table S2
in the Supporting Information). Another experiment carried
out in the presence of catalyst but in the absence of H2
(reaction performed under a nitrogen atmosphere) showed no
conversion of 1a (Table S2). A filtration test revealed that the

Figure 2. Time course of hydrogenation of phenylacetylene catalyzed
by Au/SiO2 with increasing amounts of piperazine: (a) 0.04 mmol; (b)
0.4 mmol; (c) 2 mmol; (d) 4 mmol. Reaction conditions: 4 mmol of
phenylacetylene, 0.04 mmol of Au, 0.04−4 mmol of amine, 8 mL of
ethanol, 80 °C, 6 bar of H2.
2975

DOI: 10.1021/acscatal.6b03441
ACS Catal. 2017, 7, 2973−2980

ACS Catalysis

Research Article
a

Table 1. Scope of Semihydrogenation of Alkynes to Alkenes Using Au/SiO2 and Piperazine B10

a

Reaction conditions unless specified otherwise: 1 mmol of alkyne, 0.01 mmol of Au, and 1 mmol of piperazine (B10), in 2 mL of ethanol at 80 °C
and 6 bar of H2. bDetermined by GC using internal standard techniques (>99% denotes that no other product was detected); values in parentheses
refer to isolated yields. cZ/E ratio was determined by 1H NMR spectroscopy. d1% of alkane detected by GC. e0.02 mmol of Au and toluene as
solvent (2 mL). f0.02 mmol of Au. gToluene as solvent (2 mL).

Due to the high selectivity of the Au/SiO2:piperazine (1:100
Au:B10) catalytic system for the conversion of phenylacetylene
(1a) into styrene (1b) (Table 1, entry 1), it was tested in a
competition hydrogenation experiment starting from a mixture
of phenylacetylene (1a) and styrene (1b) (Table 1, entries 2
and 3). These experiments revealed that the reaction selectivity
is maintained even in large excess of the alkene 1b and are
relevant to evaluate the application of our gold catalyst system
for the purification of styrene, which is an important monomer
and usually contains small quantities of acetylenic compounds,
which poison olefin polymerization catalysts and must be
removed via selective hydrogenation before polymerization.54
Under competitive conditions, Au/SiO2 catalyst converted
preferentially the alkyne 1a to alkene 1b and less than 1% of
ethylbenzene 1c was formed.
Finally, the general applicability and limitation of the
methodology was evaluated through the hydrogenation of

conversion was achieved with any other concentration of B10
studied (Figure 2b−d). The hydrogenation of styrene (1b) to
ethylbenzene (1c) was completely suppressed when a large
excess of B10 was used (1a:ligand:Au 100:100:1). It is worth
noting that the reaction rate is enhanced with an increase in the
concentration of the piperazine (reaction rates 0.87, 9.0, 15.7,
and 15.4 mmol gcat−1 h−1, for 1, 10, 50, and 100 equiv of B10,
respectively), suggesting that not only does piperazine play a
very important role in the activation of Au NPs as
hydrogenation catalysts (activity enhancement effect) but it
also works as a selectivity enhancer to avoid the subsequent
hydrogenation of the alkene into alkane. Particularly, in large
excess of piperazine (1a:ligand:Au 100:100:1), which would
correspond to about 240 ligands per surface Au atom
(estimated surface atoms 42%), the alkene desorbs without
being converted into the alkane.
2976

DOI: 10.1021/acscatal.6b03441
ACS Catal. 2017, 7, 2973−2980

ACS Catalysis

Research Article

alkynes under the optimized conditions (100:1:100 alkyne:Au:B10). A variety of terminal and internal alkynes were
readily hydrogenated to the desired alkene and cis-alkene with
moderate to excellent yield with, importantly, hardly any overreduction to alkane (Table 1). Moreover, both electrondeficient substituents, such as esters (Table 1, entry 25), and
electron-rich groups, for example amino and methoxy (Table 1,
entries 15 and 16), were tolerated well. Clearly, the most
challenging substrates are alkynes with an alkene moiety (Table
1, entries 6, 11, and 19). We were pleased that our Au NP
catalyst system was able to reduce only the alkyne unit without
any detectable concurrent reduction of the alkene moieties in
both the parent and product molecules. The results depicted in
Table 1 confirmed that a broad range of sensitive and reducible
functional groups, including halide, ether, and ester substituents
(Table 1, entries 12−16 and 25), were tolerated in the alkyne
hydrogenation process. The prominent isoprene, which is an
essential building block in the polymer industry, was obtained
via hydrogenation of 2-methylbut-1-en-3-yne (9a) (Table 1,
entry 11) in 84% isolated yield. (Z)-Alkenes were mostly
formed from internal alkynes (Table 1, entries 20−25).
Notably, the semihydrogenation of nonpolar internal alkynes
such as diphenylacetylene (18a) (Table 1, entry 20) and 1phenyl-1-propyne (19a) (Table 1, entry 21) proceed quite
smoothly under mild conditions. The only functional groups
that do not tolerate the reaction conditions are aldehydes,
ketones, and carboxylic acids (Table S3 in the Supporting
Information), which are known to couple with amines. The
catalyst system was also applicable for scaled-up conditions; 1a
(50 mmol, 5.1 g) and 18a (50 mmol, 8.9 g) were successfully
converted into the alkenes 1b (4.6 g, 88%) and 19b (7.1 g,
79%), respectively. Considering full conversion, a turnover
number (TON) of 5000 was reached, with an excellent
productivity of approximately 200 mol mol−1 h−1. These TON
and productivity values (not optimized) are remarkably higher
than those reported for other heterogeneous Au catalyst
systems for semihydrogenation of alkynes employing molecular
hydrogen or any other hydrogen source.39,55−58
On the basis of the aforementioned facts and bearing in mind
that primary and secondary amines are used as ligands in
homogeneous catalysis to promote heterolytic cleavage of H2 to
give metal hydride species,38,34 we suggest that the key role of
piperazine (B10) is to facilitate the crucial heterolytic H2
activation at the Au NPs. Namely, the nitrogen atom of B10
can serve as a basic ligand to promote the heterolytic H2
cleavage, providing a favorable situation for hydrogen activation
under very mild conditions. The reaction network and the
configurations are presented in Figure 3. Piperazine (B10) is
adsorbed on the gold surface through one of the N centers; due
to the configuration in a boat site the second N is higher on the
surface. The adsorption energy is sufficiently low (−0.27 eV) to
ensure that no leaching of gold occurs.59 Upon adsorption, the
H2 molecule can split between the ligand, forming a quaternary
N center and a second H that goes to the surface. Such
heterolytic splitting has been proposed for phosphine
ligands,60−63 nitrogen ligands on metals,64 and single-atom
catalysts,3,65 while the active participation of the reactant in the
activation of H2 was also found for Ag catalysts in hydrogenation.66 The metal−ligand system can be understood as a
tight ion pair induced by the frustrated Lewis pair (FLP)
formed by the ligand and the surface.67,68 H2 dissociation
occurs on the Au(111) surface through an energy barrier of
1.45 eV; the reaction is endothermic by 0.69 eV. Instead, if the

Figure 3. Reaction energy profiles with the corresponding
configurations for the hydrogenation of 1-phenyl-1-propyne (19a).
The inset corresponds to the charge distribution with respect to the
neutral fragments after the first hydrogenation step. The separation
between the yellow and blue colors indicates the electrostatic
interactions. Yellow spheres represent Au, blue N, black C, and
white H.

surface is decorated with piperazine (B10) the dissociation
shows a barrier of 0.32 eV (therefore close to 0 if zero-point
vibrational energies, ZPVE, are taken into account) and the
final state is thermoneutral in comparison to the reactants (H2
and the FLP). Then the alkyne can be adsorbed on the
Au(111). Alkynes are weakly adsorbed on Au, but due to the
van der Waals contributions the adsorption energy is −0.28 eV
for 1-phenyl-1-propyne (19a). From the coadsorbed state, the
hydride can be transferred to the alkyne. The reaction is slightly
endothermic by 0.28 eV, the barrier being 0.74 eV from the
coadsorbed state. Upon the first hydrogenation the moiety is
negatively charged and interacts by electrostatic forces with the
protonated piperazine (B10); this is shown in the inset of
Figure 3. From this pair, proton transfer occurs, with a barrier
of 0.13 eV, and the total process is exothermic by −2.39 eV.
The alkene requires 0.53 eV for desorption. The particular
structures for the transition states are shown in Figure 4. Note

Figure 4. Transition state geometries for the hydrogenation of 1phenyl-1-propyne (19a). Yellow spheres represent Au, blue N, black
C, and white H.

that even if H2 is barrierless according to the potential energy
surface the introduction of entropies in the gas phase molecules
implies that there is an entropic barrier on going from the gas
to the adsorbed phase.69 Finally, an alternative path considering
first the transfer of the H+ and then the hydride was calculated;
however, the intermediate is more than 1.5 eV higher in energy
than that of the H−, H+ transfer and thus was discarded.
Calculations also highlight the limited basicity window for
the N-containing bases able to provide the aforementioned
promotion effect for hydrogenation on gold surfaces. Piperazine
(B10) is unique because it is not strongly bound to the surface,
and thus cannot extract the metal, for instance in comparison to
2977

DOI: 10.1021/acscatal.6b03441
ACS Catal. 2017, 7, 2973−2980

ACS Catalysis

Research Article

Finally, to gain a better understanding of the H2 activation
process, we conducted H2−D2 exchange experiments (Au
catalyst, toluene, H2:D2 1:1, 6 bar, 80 °C, 24 h). The H2−D2
exchange, attested by the formation of HD in the gas phase
(detected by mass spectrometry m/z 3), was much higher over
the Au/B10 systems than that with pure Au catalyst. These
results suggest that H2 and D2 molecules dissociate and
recombine to produce HD, and this process is significantly
enhanced by piperazine. When H2 was switched for D2 in the
hydrogenation reaction under similar conditions, it was
noteworthy that a primary isotope effect (kH/kD ≈ 2.7, Figure
S4 in the Supporting Information) was observed, suggesting
that hydrogen is involved in the rate-determining step.70 To
obtain quantitative information about H2 dissociation, different
amounts of 1a were hydrogenated under limited H2 conditions
(H2 was preadsorbed on Au/SiO2 catalyst and the B10/Au/
SiO2 catalyst system and the reaction was conducted under an
inert atmosphere; see details in the Supporting Information and
Table S5). Within the limits of experimental error, when 1a was
increased from 0.01 to 0.05 mmol, a similar amount of 1b
(∼0.008 mmol) was obtained, allowing us to estimate that H2
was the limiting reagent (∼0.008 mmol of H2 activated by
B10/Au catalyst). For pure Au catalyst, a low yield of 1b
(∼0.0001−0.00045 mmol) was obtained, confirming the very
unfavorable hydrogen dissociation by Au in the absence of B10.
On the basis of the experimental and theoretical results
described above, we propose the mechanism shown in Figure 6
for the hydrogenation of alkynes by Au NPs, which is possible
in the presence of adsorbed piperazine due to the heterolytic
activation of H2 via frustrated Lewis pairs.

pyridine (B6). The binding energies for the linear complexes
from the atomic reference are exothermic for pyridine (B6)
(−0.1 eV) and very endothermic for piperazine (B10) (>5 eV).
On comparison of the N-containing bases (Figure 5), the

Figure 5. Reaction rates (experimental) as a function of the computed
activation energies for heterolytic H2 dissociation at the N ligand−
Au(111) interface (without zero-point correction). The fitting shows
that the activity is completely dominated by the H2 activation. The red
points stand for the molecules containing ligands with single N
donors. As shown here they do not belong to the same family: for
quinoline the adsorption is too strong, blocking the reactants on the
surface, while for piperidine the ligand cannot compete with the
reactant for adsorption and thus cannot effectively activate H2.

activity (reaction rates were obtained from the kinetic curves
given in Figure S3 in the Supporting Information) is found to
correlate with the energy barrier for H2 dissociation at the
ligand/Au(111) interface for piperazine, DABCO, pyrazine,
imidazole, ethyleneamine, and diethylenetriamine (Table S4 in
the Supporting Information). The barrier is the lowest for B10,
thus indicating the unique catalytic activity of the piperazine−
gold interface. The barrier for the activation is formed by
different terms. One stems from the formation of a proton−
hydride pair, this is a constant going from H2 to a H+···H− pair
separated by an average distance of 2 Å. The second appears as
a consequence of the intrinsic basicity of the N-containing
bases. A third term appears since at the transition state some of
the N-containing bases need to rearrange in order to be an
active part in the transition state; this is the case for quinoline
(B5), which presents a much larger H2 splitting barrier than
would correspond to its thermodynamic (ΔE) value. The
fourth term depends on the adsorption of H− to the metal,
which is in this case invariant. Notice that B5 due to its large
size can also block too many adsorption sites. Finally, the
structures with two basic N heteroatoms seem to be more
active for hydrogenation purposes. There are two reasons for
this: the first is that two molecules can be dissociated for each
ligand, thus being doubly active. This is clear with imidazole
(B17), as only one of the N atoms seems capable of trapping
H2 (note that diethylenetriamine (B12) converts 23% while
imidazole (B17) about 7% even if the barriers are comparable).
The second reason is that having only one N heteroatom does
not ensure adsorption to the surface for systems, such as
piperidine (B15), which contains only one N. Piperidine (B15)
thus easily leaves the surface, as in the transition state structure
there is no element binding it to the surface, making the
configuration unstable.

Figure 6. Proposed mechanism for gold-catalyzed catalytic hydrogenation of alkynes into cis-alkenes via a frustrated Lewis pair
approach.

■

CONCLUSIONS
In summary, we have described that high conversion of a range
of alkynes with excellent selectivity for cis-alkenes was achieved
by combining supported Au NPs, N-containing bases, and H2
as hydrogen source. We emphasize the role of the nitrogencontaining ligand as a means for substantial enhancement of
2978

DOI: 10.1021/acscatal.6b03441
ACS Catal. 2017, 7, 2973−2980

ACS Catalysis

Research Article

(8) Ciriminna, R.; Falletta, E.; Della Pina, C.; Teles, J. H.; Pagliaro,
M. Angew. Chem., Int. Ed. 2016, 55, 14210−14217.
(9) Wittstock, A.; Bä
umer, M. Acc. Chem. Res. 2014, 47, 731−739.
(10) Stratakis, M.; Garcia, H. Chem. Rev. 2012, 112, 4469−4506.
(11) Bond, G. C.; Sermon, P. A. J. Chem. Soc., Chem. Commun. 1973,
444b−445b.
(12) Bond, G. Gold Bull. 2008, 41, 235−241.
(13) Cárdenas-Lizana, F.; Keane, M. A. J. Mater. Sci. 2013, 48, 543−
564.
(14) Wang, X.; Cárdenas-Lizana, F.; Keane, M. A. ACS Sustainable
Chem. Eng. 2014, 2, 2781−2789.
(15) García-Mota, M.; Cabello, N.; Maseras, F.; Echavarren, A. M.;
Pérez-Ramírez, J.; Lopez, N. ChemPhysChem 2008, 9, 1624−1629.
(16) Zhang, Y.; Cui, X.; Shi, F.; Deng, Y. Chem. Rev. 2012, 112,
2467−2505.
(17) Jia, J.; Haraki, K.; Kondo, J. N.; Domen, K.; Tamaru, K. J. Phys.
Chem. B 2000, 104, 11153−11156.
(18) Segura, Y.; López, N.; Pérez-Ramírez, J. J. Catal. 2007, 247,
383−386.
(19) Grirrane, A.; Corma, A.; García, H. Science 2008, 322, 1661−
1664.
(20) Corma, A.; Serna, P.; García, H. J. Am. Chem. Soc. 2007, 129,
6358−6359.
(21) Claus, P.; Brü A.; Mohr, C.; Hofmeister, H. J. Am. Chem.
cker,
Soc. 2000, 122, 11430−11439.
(22) Corma, A.; Serna, P. Science 2006, 313, 332−334.
(23) Vilé, G.; Albani, D.; Almora-Barrios, N.; López, N.; PérezRamírez, J. ChemCatChem 2016, 8, 21−33.
(24) Hammer, B.; Norskov, J. K. Nature 1995, 376, 238−240.
(25) Fujitani, T.; Nakamura, I.; Akita, T.; Okumura, M.; Haruta, M.
Angew. Chem., Int. Ed. 2009, 48, 9515−9518.
(26) Mallat, T.; Baiker, A. Appl. Catal., A 2000, 200, 3−22.
(27) Schoenbaum, C. A.; Schwartz, D. K.; Medlin, J. W. Acc. Chem.
Res. 2014, 47, 1438−1445.
(28) Giesbrecht, P. K.; Herbert, D. E. ACS Energy Lett. 2017, 2, 549−
555.
(29) Marshall, S. T.; O’Brien, M.; Oetter, B.; Corpuz, A.; Richards, R.
M.; Schwartz, D. K.; Medlin, J. W. Nat. Mater. 2010, 9, 853−858.
(30) Chen, G.; Xu, C.; Huang, X.; Ye, J.; Gu, L.; Li, G.; Tang, Z.; Wu,
B.; Yang, H.; Zhao, Z.; Zhou, Z.; Fu, G.; Zheng, N. Nat. Mater. 2016,
15, 564−569.
(31) Kwon, S. G.; Krylova, G.; Sumer, A.; Schwartz, M. M.; Bunel, E.
E.; Marshall, C. L.; Chattopadhyay, S.; Lee, B.; Jellinek, J.; Shevchenko,
E. V. Nano Lett. 2012, 12, 5382−5388.
(32) Moreno, M.; Ibañez, F. J.; Jasinski, J. B.; Zamborini, F. P. J. Am.
Chem. Soc. 2011, 133, 4389−4397.
(33) Wu, B.; Huang, H.; Yang, J.; Zheng, N.; Fu, G. Angew. Chem.,
Int. Ed. 2012, 51, 3440−3443.
(34) Schrader, I.; Warneke, J.; Backenkö J.; Kunz, S. J. Am. Chem.
hler,
Soc. 2015, 137, 905−912.
(35) Kahsar, K. R.; Schwartz, D. K.; Medlin, J. W. J. Am. Chem. Soc.
2014, 136, 520−526.
(36) Chinchilla, R.; Nájera, C. Chem. Rev. 2014, 114, 1783−1826.
(37) García-Mota, M.; Gómez-Díaz, J.; Novell-Leruth, G.; VargasFuentes, C.; Bellarosa, L.; Bridier, B.; Pérez-Ramírez, J.; López, N.
Theor. Chem. Acc. 2011, 128, 663−673.
(38) Ren, D.; He, L.; Yu, L.; Ding, R. S.; Liu, Y. M.; Cao, Y.; He, H.
Y.; Fan, K. N. J. Am. Chem. Soc. 2012, 134, 17592−17598.
(39) Li, G.; Jin, R. J. Am. Chem. Soc. 2014, 136, 11347−11354.
(40) Yan, M.; Jin, T.; Ishikawa, Y.; Minato, T.; Fujita, T.; Chen, L.-Y.;
Bao, M.; Asao, N.; Chen, M.-W.; Yamamoto, Y. J. Am. Chem. Soc.
2012, 134, 17536−17542.
(41) Li, S.-S.; Tao, L.; Wang, F.-Z.-R.; Liu, Y.-M.; Cao, Y. Adv. Synth.
Catal. 2016, 358, 1410−1416.
(42) Scanlon, J. T.; Willis, D. E. J. Chromatogr. Sci. 1985, 23, 333−
340.
(43) Jacinto, M. J.; Kiyohara, P. K.; Masunaga, S. H.; Jardim, R. F.;
Rossi, L. M. Appl. Catal., A 2008, 338, 52−57.

catalytic activity and selectivity, since the auxiliary ligands play a
key role by opening a new channel for the heterolytic
dissociation of molecular hydrogen on the frustrated Lewis
pair. This ability to switch on the catalytic activity of Au NPs
provides a promising new method for controlling a reactive
system with ligands. The control is such that stereochemistry of
the addition is also achieved due to electrostatic forces. A
delicate balance among (i) the basicity of the ligand, (ii) the
reorganization energy of the ligand to activate H2, (iii) the
possibility of site blocking by large ligands, and (iv) the metal
leaching induced by some ligands is responsible for the
activity−selectivity−stability controls in the catalyst system. We
believe this remarkable promotional effect by N-containing
ligands may have implications for gold catalysis while
stimulating more applications in the field of selective
hydrogenations not envisaged before. The versatile concept
presented can be transferable to other metal nanoparticle
catalysts.

■

ASSOCIATED CONTENT

S
* Supporting Information

These materials are available free of charge via the Internet at
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acscatal.6b03441.
Additional experimental results and details, characterization of catalysts, procedures, and characterization data,
including NMR spectra (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*E-mail for L.M.R.: lrossi@iq.usp.br.
ORCID

Jhonatan L. Fiorio: 0000-0001-7435-7430
Núria López: 0000-0001-9150-5941
Liane M. Rossi: 0000-0001-7679-0852
Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS
The authors are grateful to the INCT-Catalise and the Brazilian
government agencies FAPESP (Grant 2016/167387), CNPq,
and CAPES for financial support. N.L. thanks the BSC-RES for
providing generous computational resources.

■

REFERENCES

(1) Kyriakou, G.; Boucher, M. B.; Jewell, A. D.; Lewis, E. A.; Lawton,
T. J.; Baber, A. E.; Tierney, H. L.; Flytzani-Stephanopoulos, M.; Sykes,
E. C. H. Science 2012, 335, 1209−1212.
(2) Vilé, G.; Albani, D.; Almora-Barrios, N.; López, N.; PérezRamírez, J. ChemCatChem 2016, 8, 21−33.
(3) Vilé, G.; Albani, D.; Nachtegaal, M.; Chen, Z.; Dontsova, D.;
Antonietti, M.; López, N.; Pérez-Ramírez, J. Angew. Chem., Int. Ed.
2015, 54, 11265−11269.
(4) Jagadeesh, R. V.; Surkus, A.-E.; Junge, H.; Pohl, M.-M.; Radnik, J.;
Rabeah, J.; Huan, H.; Schunemann, V.; Bruckner, A.; Beller, M. Science
2013, 342, 1073−1076.
(5) Westerhaus, F. A.; Jagadeesh, R. V.; Wienhö G.; Pohl, M.-M.;
fer,
Radnik, J.; Surkus, A.-E.; Rabeah, J.; Junge, K.; Junge, H.; Nielsen, M.;
Brü
ckner, A.; Beller, M. Nat. Chem. 2013, 5, 537−543.
(6) Chen, F.; Kreyenschulte, C.; Radnik, J.; Lund, H.; Surkus, A.;
Junge, K.; Beller, M. ACS Catal. 2017, 7, 1526−1532.
(7) Hashmi, A. S. K.; Hutchings, G. J. Angew. Chem., Int. Ed. 2006,
45, 7896−7936.
2979

DOI: 10.1021/acscatal.6b03441
ACS Catal. 2017, 7, 2973−2980

ACS Catalysis

Research Article

(44) Liu, X.; Wang, A.; Yang, X.; Zhang, T.; Mou, C. Y.; Su, D. S.; Li,
J. Chem. Mater. 2009, 21, 410−418.
(45) Dekkers, M. A. P.; Lippits, M. J.; Nieuwenhuys, B. E. Catal. Lett.
1998, 56, 195−197.
(46) Zanella, R.; Giorgio, S.; Henry, C. R.; Louis, C. J. Phys. Chem. B
2002, 106, 7634−7642.
(47) Kresse, G.; Hafner, J. Phys. Rev. B: Condens. Matter Mater. Phys.
1993, 47, 558−561.
(48) Kresse, G.; Furthmü J. Comput. Mater. Sci. 1996, 6, 15−50.
ller,
(49) Hammer, B.; Hansen, L.; Nørskov, J. Phys. Rev. B: Condens.
Matter Mater. Phys. 1999, 59, 7413−7421.
(50) Almora-Barrios, N.; Carchini, G.; Błoński, P.; López, N. J. Chem.
Theory Comput. 2014, 10, 5002−5009.
(51) Kresse, G. Phys. Rev. B: Condens. Matter Mater. Phys. 1999, 59,
1758−1775.
(52) Monkhorst, H. J.; Pack, J. D. Phys. Rev. B 1976, 13, 5188−5192.
(53) Henkelman, G.; Uberuaga, B. P.; Jónsson, H. J. Chem. Phys.
2000, 113, 9901.
(54) Shao, L.; Huang, X.; Teschner, D.; Zhang, W. ACS Catal. 2014,
4, 2369−2373.
(55) Mitsudome, T.; Yamamoto, M.; Maeno, Z.; Mizugaki, T.;
Jitsukawa, K.; Kaneda, K. J. Am. Chem. Soc. 2015, 137, 13452−13455.
(56) Vasilikogiannaki, E.; Titilas, I.; Vassilikogiannakis, G.; Stratakis,
M. Chem. Commun. 2015, 51, 2384−2387.
(57) Li, S.-S.; Liu, X.; Liu, Y.-M.; He, H.-Y.; Fan, K.-N.; Cao, Y.
Chem. Commun. 2014, 50, 5626−5628.
(58) Liang, S.; Hammond, G. B.; Xu, B. Chem. Commun. 2016, 52,
6013−6016.
(59) Jover, J.; García-Ratés, M.; López, N. ACS Catal. 2016, 6, 4135−
4143.
(60) Cano, I.; Huertos, M. a; Chapman, A. M.; Buntkowsky, G.;
Gutmann, T.; Groszewicz, P. B.; van Leeuwen, P. W. N. M. J. Am.
Chem. Soc. 2015, 137, 7718−7727.
(61) Liu, C.; Abroshan, H.; Yan, C.; Li, G.; Haruta, M. ACS Catal.
2016, 6, 92−99.
(62) Cano, I.; Chapman, A. M.; Urakawa, A.; van Leeuwen, P. W. N.
M. J. Am. Chem. Soc. 2014, 136, 2520−2528.
(63) López-Serrano, J.; Duckett, S. B.; Lledós, A. J. Am. Chem. Soc.
2006, 128, 9596−9597.
(64) Chen, Z.; Pronkin, S.; Fellinger, T.-P.; Kailasam, K.; Vilé, G.;
Albani, D.; Krumeich, F.; Leary, R.; Barnard, J.; Thomas, J. M.; PérezRamírez, J.; Antonietti, M.; Dontsova, D. ACS Nano 2016, 10, 3166−
3175.
(65) Liu, P.; Zhao, Y.; Qin, R.; Mo, S.; Chen, G.; Gu, L.; Chevrier, D.
M.; Zhang, P.; Guo, Q.; Zang, D.; Wu, B.; Fu, G.; Zheng, N. Science
2016, 352, 797−800.
(66) Vilé, G.; Baudouin, D.; Remediakis, I. N.; Copéret, C.; López,
N.; Pérez-Ramírez, J. ChemCatChem 2013, 5, 3750−3759.
(67) Welch, G. C.; Juan, R. R. S.; Masuda, J. D.; Stephan, D. W.
Science 2006, 314, 1124−1126.
(68) Chernichenko, K.; Madarász, A.; Pápai, I.; Nieger, M.; Leskelä
,
M.; Repo, T. Nat. Chem. 2013, 5, 718−723.
(69) Chorkendorff, I.; Niemantsverdriet, J. W. Concepts of Modern
Catalysis and Kinetics; Wiley-VCH: Weinheim, FRG, 2003.s.
(70) Carey, F. A.; Sundberg, R. J. Advanced Organic Chemistry Part A:
Structure and Mechanisms; Springer: Boston, MA, USA, 2007.

2980

DOI: 10.1021/acscatal.6b03441
ACS Catal. 2017, 7, 2973−2980