Identity of the active site in gold nanoparticle-catalyzed sonogashira coupling of phenylacetylene and iodobenzene

Article abstract of DOI:10.1021/ja1063179

XPS, TEM, and reaction studies were used to examine the catalytic behavior of gold species deposited on lanthana toward the cross-coupling of phenylacetylene and iodobenzene. Atomically dispersed Au^I and Au ^III were catalytically ine

Full text of DOI:10.1021/ja1063179

Published on Web 08/17/2010
Identity of the Active Site in Gold Nanoparticle-Catalyzed Sonogashira
Coupling of Phenylacetylene and Iodobenzene
Simon K. Beaumont, Georgios Kyriakou, and Richard M. Lambert*
Department of Chemistry, Cambridge UniVersity, Lensfield Road, Cambridge, CB2 1EW, United Kingdom
Received July 16, 2010; E-mail: rml1@cam.ac.uk
Au-containing Schiff base complexes that were used as reference
homogeneous catalysts. It was found that only the Au complex
Abstract: XPS, TEM, and reaction studies were used to examine
the catalytic behavior of gold species deposited on lanthana
I
was an effective homogeneous catalyst, leading to the view that
toward the cross-coupling of phenylacetylene and iodobenzene.
I
the active agent in the Au/ceria catalyst was Au . (In passing we
I
III
Atomically dispersed Au and Au were catalytically inert, whereas
note that the homogeneous catalysis reference data reported by
Gonzalez-Arellano et al. have been brought into question because
traces of soluble Pd species may actually have been the active
0
metallic Au nanoparticles were both active and very selective.
Thus it is metallic gold and not ionic gold that provides the
0
catalytically active sites. Au nanoparticles supported on silica,
8
agent. ) Clearly, this stands in contrast with our recent proposal,
γ-alumina, and BaO were active but relatively unselective;
however, as with lanthana, ceria-supported Au0 nanoparticles
showed high selectivity. This strong promoting effect of the
lanthanide oxide supports on Sonogashira selectivity cannot be
accounted for in terms of acid/base, redox, or SMSI effects; it
may be tentatively ascribed to metal f support hydrogen spillover.
5
as described above. However, our Au/silica catalysts were
substantially less selective than the gold-ceria nanomaterial used
7
by Gonzalez-Arellano et al. Accordingly, we set out to examine
the apparent contradictions regarding the identity of the active site
in gold-catalyzed heterogeneous Sonogashira coupling and to
investigate the related issue of whether or not rare earth oxide
supports do induce higher selectivity. We are led to the conclusion
0
that in all the cases reported here metallic Au nanoparticles are
I
Metal-catalyzed coupling reactions that result in formation of
new C-C bonds are strategically important in organic synthesis,
and metal nanoparticles suspended in solution are often used to
indeed the active agent whereas immobilized Au is not. However,
consistent with the results of Gonzalez-Arellano et al., it is found
that rare earth oxide supports strongly promote selectivity toward
cross-coupling compared to neutral, acidic, and basic support
materials.
1-4
provide the catalytic agent.
Sonogashira coupling is an important
example of such chemistry, a prototypical case being the cross-
coupling of phenylacetylene with iodobenzene (Scheme 1). When
catalyzed by silica-supported Au nanoparticles, detailed evaluation
of reaction data and quantitative analysis of the solid and solution
phases by XPS and ICP-MS, respectively, led to the conclusion
that the reaction was overwhelmingly a heterogeneous process with
An unambiguous test as to whether gold species in higher
oxidation states and immobilized on metal oxides are active
Sonogashira catalysts is made possible by the recent discovery by
Goguet et al. who showed that it is possible to prepare a gold-on-
I
lanthana material in which the metal is present only in the Au and
5
III
9
Au(0) metal clusters providing the active sites. This view was also
Au oxidation states. On the basis of XANES, XPS, TEM, and
ISS measurements, these authors concluded that the ionic gold was
present as monatomic entities as opposed to multiatom clusters and,
most importantly for present purposes, no Au(0) was detectable.
We therefore synthesized this lanthana-based material following a
modification of the procedure described by Goguet et al., hereafter
designated Catalyst A, and evaluated it for Sonogashira coupling
of iodobenzene and phenylacetylene (Scheme 1). For purposes of
comparison, using a seeded growth procedure we synthesized
Catalyst B, which consisted of ∼20 nm Au particles with a narrow
size dispersion supported on lanthana.
consistent with the pronounced metal particle size effects that were
5
observed and with the results of recent single crystal observations
6
of the same reaction carried out on an extended Au(111) surface.
Scheme 1. Coupling Reactions of Iodobenzene and
Phenylacetylene To Yield Both the Desired Sonogashira
Cross-Coupling Product Diphenylacetylene (DPA) and the Two
Homocoupling Side Products Diphenyldiacetylene and Biphenyl
Figure 1 shows the respective morphologies from which it is
apparent that whereas the ∼20 nm Au particles in Catalyst B are
clearly visible, images of Catalyst A were indistinguishable from
9
those of the oxide support alone, in good agreement Goguet et al.
2 3
who concluded that Au/La O prepared in this way contained only
atomically dispersed gold species. Figure 2 shows the corresponding
I
Au 4f XP spectra. It is clear that Catalyst A contained Au as a
III
Recently Gonzalez-Arellano et al. reported on a new type of gold-
ceria nanomaterial that was highly selective as a heterogeneous
catalyst for this same reaction, which they attributed to the ability
of nanocrystalline ceria to stabilize a small amount of the gold as
majority species and Au as a minority species (Au 4f7/2 peaks
1
0
centered at 85.6 and 88.1 eV respectively ) with no detectable
Au . On the other hand, Catalyst B contained only metallic Au
(Au 4f7/2 peak centered at 83.8 eV ) and no detectable ionic gold.
Note that the observed binding energy shifts between Catalysts A
and B (1.8 and 4.3 eV) are substantially greater than the 0.6-1.0
0
0
1
0
I
0
III
Au although Au and Au were the predominant components of
7
the solid catalyst. Their conclusion was based on a comparison of
1
1,12
the catalytic behavior of their Au/ceria material with that of three
eV values typical of metal particle quantum size effects
and
1
2246 9 J. AM. CHEM. SOC. 2010, 132, 12246–12248
10.1021/ja1063179  2010 American Chemical Society

C O M M U N I C A T I O N S
can therefore be reliably ascribed to changes in Au oxidation state.
Moreover, the two components in the Au 4f XP spectra of Catalyst
A are separated by 2.5 eV, in excellent agreement with reference
spectra of Au(I) and Au(III) molecular compounds, which are also
comparison to be made between Catalysts A, B, and C the
corresponding turnover numbers (TONs) are shown, calculated on
a per surface gold atom basis as explained in the Supporting
Information. It can be seen (column 4) that Catalyst A, which
contained only ionic gold, is at least 100 times less active, and
possibly much less active, toward Sonogashira coupling than the
1
0
separated by 2.5 eV.
0
Au nanoparticles present in Catalyst B, which contained no ionic
gold. These results amount to very strong evidence that (i) lanthana-
I
III
stabilized Au species (and indeed Au species) are not the active
site whereas (ii) oxide-supported metallic Au nanoparticles are
I
responsible for the observed catalysis. It thus appears that Au
immobilized on lanthana does not behave in an analogous way to
I
Au complexes in solution, if the latter are indeed effective for
7
Sonogashira coupling. Notice also the much higher selectivity of
2 3
La O supported Au nanoparticles (Catalyst B) compared to the
silica-supported case (Catalyst C). Interestingly, this effect of a rare
earth oxide support in promoting Sonogashira selectivity is
consistent with the observations of Gonzalez-Arellano et al. who
Figure 1. Representative TEM images of La
a) Catalyst A containing finely dispersed/mononuclear gold, invisible to
TEM, immobilized on a La support; (b) Catalyst B containing 20 nm
gold particles (dark features) deposited on a La support.
2 3
O supported gold catalysts:
7
(
reported 89% selectivity for their Au/nanocrystalline ceria catalyst.
2 3
O
2
We confirmed this selectivity-promoting effect of CeO by preparing
O
2 3
13
and testing Catalyst D using Au impregnated onto ceria, which
is a low surface area material. Although the resulting conversion
was substantially lower than that achieved by Gonzalez-Arellano
et al. with their nanocrystalline ceria (15% after 160 h reaction),
this likely reflects the much lower active metal area in our case
2
2
(
the ceria surface area was <1 m /g compared to 20 m /g for the
2
La ; see also TEM data in the Supporting Information). The
O
3
important point is that we too obtained >85% cross-coupling
selectivity, in accord with their findings.
2 3 2
It therefore seems plausible that La O and CeO act in a similar
way to enhance selectivity in Au-catalyzed Sonogashira coupling
by a mechanism that is not accessible when silica is used as the
support. Related to this, it is significant that while Au supported
on both lanthana and ceria catalyzed the homocoupling of phenyl-
acteylene to diphenyldiacetylene (DPDA Scheme 1) silica-supported
Au did not. Note that both DPDA formation and Sonogashira cross-
coupling formally require removal of the acidic hydrogen atom on
phenylacetylene. Therefore, in order to examine whether this
behavior correlates with the basicity of the support, we synthesized
and tested catalysts (see Supporting Information) consisting of gold
supported on a strongly basic support (BaO) and an acidic support
Figure 2. Au 4f XP spectra for Catalysts A (top) and B (bottom) showing
the different oxidation states of gold present and corresponding to the
morphologies shown in the illustration (left).
2 3
Table 1. Results of Catalytic Testing of the Two Au/La O
Catalysts A and B (160 h at 145 °C, 0.5 mmol of IB and PA; 30
mg of catalyst; 0.3 mmol of base) and Previously Reported
5
Results for Au/SiO
2
Catalyst C for Comparison
(γ-Al
2 3
O ): both gave much lower selectivities, comparable to Au/
SiO , and as with silica, neither produced DPDA. Thus no
2
b
TON Molecules
correlation exists between Sonogashira selectivity (and DPDA
Selectivity to
a
DPA
Final Yield
of DPA/mmol
of reactant per
surface Au atom
formation) and the basicity of the support materials, which falls in
c
14,15
Catalyst A
Dispersed Au
Undetectable
82%
Undetectable
0.40
,10
the order BaO > CeO
2 2
, La O
3
> TiO
2
, SiO
2
2 3
> γ-Al O .
Another possible cause for the exceptional behavior of ceria and
lanthana relative to the other supports could be the intervention of
(
0.5 wt %)/La
Catalyst B
0 nm Au
10 wt %)/La
2 3
O
275
158
redox chemistry, relatively facile in the case of CeO
2
, much less
2
(
1
6
O
2 3
2 3
so in the case of La O .
Post-reaction XPS of the ceria and
c
Catalyst C
38%
0.16
lanthana-supported catalysts showed no evidence of oxide reduction
in either case (see Supporting Information). Given the results of
2
0 nm Au
(
10 wt %)/SiO
2
1
7
3+
Behm and co-workers who found that Ce species in partially
reduced Au/ceria catalysts survived ex situ transfer from reactor to
spectrometer, we may tentatively conclude that effects due to
oxygen vacancies were not responsible for the observed catalytic
a
Selectivity is defined as 100% × concentration DPA/concentration
b
of (DPA + BP + DPDA). Estimated in number of IB molecules
consumed per surface goc ld atom calculated as described in the
Supporting Information. Quoted upper limit based on detection
sensitivity. Actual TON likely to be much lower.
behavior of Au/CeO
can exhibit strong metal support interaction (SMSI), the absence
of a high temperature reduction step during the preparation of
our catalysts would appear to rule out any such effects as an
explanation for Sonogashira selectivity enhancement.
2 2 3
and Au/La O . Although Au/ceria catalysts
1
8
1
9
Table 1 summarizes the results obtained with Catalysts A and B
after 160 h of reaction carried out at 145 °C in DMF solvent; also
shown and for comparison purposes are data reported previously
for Catalyst C, a Au/SiO
method as that for the Au/La
2
material prepared using exactly the same
Catalyst B. To enable a direct
Finally, however, we note that both Sonogashira cross-coupling
and phenylacetylene homocoupling may formally be considered to
2 3
O
J. AM. CHEM. SOC. 9 VOL. 132, NO. 35, 2010 12247

C O M M U N I C A T I O N S
produce hydrogen as product (on a metal surface HI is thermody-
(3) de Souza, R. O. M. A.; Bittar, M. S.; Mendes, L. V. P.; da Silva, C. M. F.;
6
da Silva, V. T.; Antunes, O. A. C. Synlett 2008, 12, 1777–1780.
namically disfavored relative to 1/2 H
2
+ 1/2 I
2
). Consistent with
(
4) Kanuru, V. K.; Humphrey, S. M.; Kyffin, J. M. W.; Jefferson, D. A.; Burton,
this, the characteristic orange-brown color of the iodine coproduct
was observed in every case. Thus,
J. W.; Armbr u¨ ster, M.; Lambert, R. M. Dalton Trans. 2009, 7602–7605.
(5) Kyriakou, G.; Beaumont, S. K.; Humphrey, S. M.; Antonetti, C.; Lambert,
R. M. ChemCatChem, DOI: 10.1002/cctc.201000154.
PhCt CH + PhI f PhCt CPh + H(a) + I(a)
(6) Kanuru, V. K.; Kyriakou, G.; Beaumont, S. K.; Papageorgiou, A. C.;
Watson, D. J.; Lambert, R. M. J. Am. Chem. Soc. 2010, 132, 8081–8086.
2
PhCt CH f PhCt CsCt CPh + 2 H(a)
(7) Gonzalez-Arellano, C.; Abad, A.; Corma, A.; Garc ´ı a, H.; Iglesias, M.;
S a´ nchez, F. Angew. Chem., Int. Ed. 2007, 46, 1536–1538.
Additionally, it is known that both ceria- and lanthana-supported
catalysts exhibit metal f support hydrogen spillover effects even
at room temperature.
(8) Lauterbach, T.; Livendahl, M.; Rosell o´ n, A.; Espinet, P.; Echavarren, A. M.
Org. Lett. 2010, 12, 3006–3009.
2
0-22
(9) Goguet, A.; Ace, M.; Saih, Y.; Sa, J.; Kavanagh, J.; Hardacre, C. Chem.
One may therefore speculate that
Commun. 2009, 4889–4891.
scavenging of hydrogen from the gold by the support acts to
promote the above two reactions, but not the homocoupling of
iodobenzene to produce biphenyl, in accord with observation.
In summary, the performance of Au/lanthana catalysts, prepared
so as to contain either Au nanoparticles or Au and Au , indicates
that it is metallic gold that provides the catalytically active sites.
(10) Casaletto, M. P.; Longo, A.; Martorana, A.; Prestianni, A.; Venezia, A. M.
Surf. Interface Anal. 2006, 38, 215–218.
(
11) In the case of the ceria-supported catalysts, it was found that deposition of
2
0 nm Au particles resulted in nanoparticle agglomeration due to the low
surface area of this support. Instead, these catalysts were prepared by an
impregnation method. In order to check that the preparation method did
not significantly affect Sonogashira selectivity, control measurements with
0
I
III
Au/SiO
2
and Au/La
2
O
3
prepared by impregnation were performed. These
and silica-
materials exhibited essentially the same selectivity as the La
2
O
3
2 2 3
When supported on CeO or La O the Au nanoparticles exhibit
supported monodisperse ∼20 nm Au nanoparticles reported in Table 1 but
a reduction in activity resulting from the lower gold surface area of the
impregnated catalysts.
strongly enhanced selectivity toward Sonogashira coupling. This
behavior cannot be accounted for in terms of redox, acid/base, or
SMSI effects; it may instead be tentatively ascribed to metal f
support hydrogen spillover.
(
12) Santra, A. K.; Goodman, D. W. J. Phys: Condens. Matter 2003, 15 (2),
R31.
(13) Nosova, L. V.; Stenin, M. V.; Nogin, Y. N.; Ryndin, Y. A. Appl. Surf. Sci.
992, 55, 43–48.
(14) Mitchel, M. D.; Vannice, M. A. Ind. Eng. Chem. Fundam. 1984, 23, 88–
6.
1
Acknowledgment. S.K.B. acknowledges financial support from
Cambridge University, Trinity Hall, Cambridge, the UK Society
of the Chemical Industry, and the International Precious Metals
Institute. G.K. acknowledges financial support by the UK Engineer-
ing and Physical Sciences Research Council.
9
(
(
15) Ryndin, Y. A.; Hicks, R. F.; Bell, A. T. J. Catal. 1981, 70, 287–297.
16) Pettigrew, D. J.; Trimm, D. L.; Cant, N. W. Catal. Lett. 1994, 28, 313–
319.
(
17) Karpenko, A.; Leppelt, R.; Cai, J.; Plzak, V.; Chuvilin, A.; Kaiser, U.;
Behm, R. J. J. Catal. 2007, 250, 139–150.
(
18) Hermann, J. J.; Ramaroson, E.; Tempere, J. F.; Guilleux, M. F. Appl. Catal.
Supporting Information Available: Experimental details of syn-
thesis, catalytic testing, and characterization techniques; additional
catalytic data; Ce and La 3d XP spectra; representative TEM of Au/
CeO Catalyst D. This material is available free of charge via the
2
Internet at http://pubs.acs.org.
1
989, 53, 117–134.
(19) Tauster, S. J. In Strong Metal-Support Interactions; Baker, R. T. K., Tauster,
S. J., Dumesic, J. A., Eds.; ACS Symposium Series 298; American
Chemical Society: Washington, DC, 1986.
(
(
(
20) Bensalem, A.; Bozon-Verduraz, F.; Perrichon, V. J. Chem. Soc., Faraday
Trans. 1995, 91, 2185–2189.
21) Bernal, S.; Calvino, J. J.; Cifredo, G. A.; Rodriguez-Izquierdo, J. M.;
Perrichon, V.; Laachir, A. J. Catal. 1992, 137, 1–11.
References
22) Bensalem, A.; Bozon-Verduraz, F. React. Kinet. Catal. Lett. 1997, 60,
(1) Li, P.; Wanga, L.; Lia, H. Tetrahedron 2005, 61, 8633–8640.
(2) Thathagar, M. B.; Kooyman, P. J.; Boerleidera, R.; Jansen, E.; Elsevier,
C. J.; Rothenberg, G. AdV. Synth. Catal. 2005, 347, 1965–1968.
71–77.
JA1063179
1
2248 J. AM. CHEM. SOC. 9 VOL. 132, NO. 35, 2010