Aurophilic complexes as gold atom sources in organic media

Article abstract of DOI:10.1039/b111679g

The decomposition, either thermal or under H₂, of [O(Au¹PR₃)₃](CF₃SO₃) (R = Ph 1, R = Me 2) in organic solvents has been studied by ³¹P{¹H} NMR, UV-vis spectroscopy an

Full text of DOI:10.1039/b111679g

Aurophilic complexes as gold atom sources in organic media†
Manuel Bardají,*a Pawel Uznanski,b Catherine Amiens,*c Bruno Chaudretc and Antonio Lagunaa
a Departamento de Química Inorgánica, Instituto de Ciencia de Materiales de Aragón, Universidad de
Zaragoza-CSIC, E-50009 Zaragoza, Spain
b Centre for Molecular and Macromolecular Studies, PAS, Sienkiewicza 112, 90-363 Lodz, Poland
c
Laboratoire de Chimie de Coordination du CNRS, 205, route de Narbonne, 31077 Toulouse Cédex 04,
France. E-mail: amiens@lcc-toulouse.fr
Received (in Cambridge, UK) 2nd January 2002, Accepted 5th February 2002
First published as an Advance Article on the web 19th February 2002
The decomposition, either thermal or under H2, of
[O(AuIPR3)3](CF3SO3) (R = Ph 1, R = Me 2) in organic
solvents has been studied by 31P{1H} NMR, UV–vis spectros-
copy and TEM; during the reaction, the phosphine acts as an
efficient oxygen trap and gold nanoparticles are produced
which may be stabilized by PVP in acetonitrile (mean
diameter 4.5 nm) or oleylamine in toluene (mean diameter 9
nm).
contain finely divided gold metal. Most of this powder sticks to
the walls of the reactor and could not be isolated for further
characterisation. Hence, to better characterise the system, the
reaction was reproduced in the presence of 100 mg of poly(N-
vinylpyrrolidone) (PVP, 40 000 g mol21). The reaction was
monitored by UV–vis spectroscopy. Once the reflux started, a
brown colloidal solution formed, probably containing small
clusters, which rapidly transformed into a typical ruby-coloured
colloidal solution: l = 525 nm for colloid 1 (obtained from
complex 1) and 515 nm for colloid 2 (obtained from complex 2).
The observation of the typical plasmon band implies the
presence of gold nanoparticles with sizes in the range 4–30
nm.19 The nanoparticles could be stored for a long time by
evaporating to dryness the solution to give a red/violet residue,
which could then be redissolved in chlorinated solvents or
acetonitrile. Transmission electron microscopy (TEM) images
of colloid 1 display nanoparticles of 4.5 0.5 nm mean diameter
and narrow size distribution. In the case of colloid 2,
nanoparticles of similar size (4.3 0.5 nm mean diameter) and
also narrow size distribution are observed (Fig. 1). As PVP is
not a stabiliser of choice for gold,20 these results suggest that the
phosphine contributes to the stabilisation of the nanoparticles.
However, since a powder is obtained in the absence of polymer,
the interactions between the phosphine ligand and the gold
surface is probably weak.
In recent years intensive work has been focused on the synthesis
and properties of nanoparticles. These properties were shown to
differ from those of the corresponding bulk materials and to be
highly dependent on the surface state.1 Many applications are
expected in catalysis,2 gas sensors technology,3,4 high density
recording media5 and microelectronics,3 etc., which require
well characterized nanoparticles assemblies, with narrow size
distributions and well controlled surface states. Furthermore,
the properties of these objects may be modulated through the
formation of bimetallic species.6 In this respect, nanoparticles
including gold atoms could be of particular interest for their
optical and catalytic properties.7 This requires the use of a gold
precursor, which would decompose in conditions close to those
of the other metals in the absence of contaminants such as water.
Besides, strongly coordinating ligands such as thiols should be
avoided since they alter the surface of the particles which is thus
distorted8 or passivated, with modified magnetic properties.9
Halide ions also form strong bonds with gold or silver surfaces.
For instance, chloride ions are used to anchor nanoparticles on
gold surfaces10 and their presence is essential for the stabilisa-
Since our precursors contain phosphines, 31P NMR can be
used to follow the decomposition reaction. The 31P{1H} NMR
tion
of
gold–phosphine
clusters11
such
as
[Au39(PPh3)14Cl6]Cl2 and [Au55(PPh3)12Cl6]13 or even of
nanoparticles.14 There is therefore a need for alternatives to
gold chloro derivatives as nanoparticle precursors15,16 and for
conditions allowing the use of labile and non-contaminating
ligands.
12
We report here our investigations on the gold( ) complexes
I
[O(AuIPR3)3](CF3SO3) (R = Ph 1, R = Me 2)17 as potential
gold atom precursors in organic media. These complexes are
easily prepared, they display short gold–gold distances and are
available with two phosphines of different electronic and steric
properties. These complexes have been widely used to carry out
auronation reactions and as alternative precursors for the
synthesis of small gold clusters (4–8 gold atoms).18 In the
context of this work, the phosphine would act both as a reducing
agent and as an oxygen trap. The counter-anion is non-
coordinating and should not alter the surface properties of the
nanoparticles.
Thermal decompositions of these salts were carried out in
refluxing acetonitrile. Typically, 5 mg of the oxonium salt were
dissolved in 20 mL degassed acetonitrile and the solution was
refluxed under argon for 2 h. At this temperature phosphine or
phosphine oxide ligands alone were not able to stabilize gold
nanoparticles, and a brown powder was formed, thought to
† Electronic supplementary information (ESI) available: UV–vis spectra,
TEM images and size histograms of colloids 1, 3 and 4. See http://
www.rsc.org/suppdata/cc/b1/b111679g/
Fig. 1 TEM images and size histograms of colloids 2 (a) and 5 (b).
598
CHEM. COMMUN., 2002, 598–599
This journal is © The Royal Society of Chemistry 2002

spectrum of colloid 1 showed both the presence of free
phosphine oxide and a broad resonance at 41.2 ppm. When
running the experiment at 255 °C, this broad peak splits into a
sharp signal assigned to [Au(PPh3)2]+ and several small signals
assigned tentatively to phosphino-clusters present in the
solution. It is well-known that the presence of a small amount of
extra phosphine can induce fast exchanges in phosphino–gold
derivatives.21 As determined through integration of the NMR
signals, one third of the initial phosphine was transformed into
phosphine oxide. [Au(PPh3)2](CF3SO3) has been heated for
more than 9 h in refluxing acetonitrile without any change in its
NMR spectrum. In the case of colloid 2 the corresponding
[Au(PMe3)2](CF3SO3) complex and phosphine oxide were the
only products observed. Therefore the 31P NMR experiments
demonstrate that the thermal decomposition of the oxonium
salts is an intramolecular redox process which produces gold(0),
and [Au(PR3)2](CF3SO3). Furthermore, these experiments
show that the phosphines effectively behave as oxygen traps.
This contrasts with the observation that these oxonium salts
and narrow size distribution. In toluene, the use of a weaker
stabiliser, oleylamine, also leads to gold nanoparticles of narrow
size distribution centered around 9 nm. We now intend to
develop these precursors with the aim of synthesizing bimetallic
Au/M nanoparticles and study their physical properties.
This work was supported by the Accion Integrada HF1999-
0122, the Programme Picasso No 00688XH, the Dirección
General de Investigación Científica y Técnica (Project PB97-
1010-C02-01) and the European Associated Laboratory
(financed by the CNRS and the Polish Committee for Scientific
Research, project PBZ-KBN 7 013/T08/39). We thank V.
Collière for technical assistance in TEM experiments at the
TEMSCAN service of the Université Paul-Sabatier.
Notes and references
1 (a) G. Schmid Clusters and Colloids: From Theory to Applications, ed.
G. Schmid, VCH, Weinheim, 1994; (b) L. J. de Jongh Physics and
Chemistry of Metal Cluster Compounds, Kluwer, Dordrecht, 1994.
2 (a) L. N. Lewis, Chem. Rev., 1993, 93, 2693; (b) J. S. Bradley, in ref.
1(a), ch. 6, pp. 459–544; (c) H. Bönnemann, G. Braun, W. Brijoux, R.
Brinkmann, A. Schulze Tilling, K. Seevogel and K. Siepen, J.
Organomet. Chem., 1996, 520, 143.
3 Nanotechnology, Molecularly Designed Materials, ed. G.-M. Chow and
K. E. Gonsalves, ACS Symposium Series 622, Science and Engineer-
ing, Inc., 20–24 August 1995, American Chemical Society, Wash-
ington, DC, 1996.
4 C. Nayral, E. Viala, V. Collière, P. Frau, F. Senocq, A. Maisonnat and
B. Chaudret, Appl. Surf. Sci., 2000, 164, 219.
5 S. Sun, C. B. Murray, D. Weller, L. Folks and A. Moser, Science, 2000,
287, 1989, and references therein.
6 N. Toshima and T. Yonezawa, New J. Chem., 1998, 22, 1179.
7 (a) S. Link, Z. L. Wang and M. A. El-Sayed, J. Phys. Chem. B, 1999,
103, 3529; (b) Y. Mizukoshi, T. Fujimoto, Y. Nagata, R. Oshima and Y.
J. Maeda, Phys. Chem. B, 2000, 104, 6028; (c) A. M. Molenbroek, J. K.
Norskov and B. S. Clausen, J. Phys. Chem. B, 2001, 105, 5450; (d) A.
Henglein, J. Phys. Chem. B, 2000, 104, 2201.
8 F. Dassenoy, K. Philippot, T. Ould Ely, C. Amiens, P. Lecante, E.
Snoeck, A. Mosset, M.-J. Casanove and B. Chaudret, New J. Chem.,
1998, 22, 703.
9 P. W. Selwood Chemisorption and Magnetization, Academic Press,
New York, 1975.
10 N. Prokopuk and D. Shriver, Chem. Mater., 1998, 10, 10.
11 P. J. Dyson and D. M. P. Mingosin Gold: Progress in Chemistry,
Biochemistry and Technology, ed. H. Schmidbaur, Wiley, New York,
1999, p. 512.
12 B. K. Teo, X. Shi and H. Zhang, J. Am. Chem. Soc., 1992, 114, 2743.
13 G. Schmid, R. Pfeil, R. Boese, F. Bandermann, S. Meyer, G. H. M. Calis
and J. W. A. Van der Velden, Chem. Ber., 1981, 114, 3634.
14 J. Fink, C. J. Kiely, D. Bethell and D. Schiffrin, Chem. Mater., 1998, 10,
922.
15 (a) M. Faraday, Philos. Trans. R. Soc. London, 1857, 147, 145; (b) J.
Turkevich, P. C. Stevenson and J. Hillier, Discuss. Faraday Soc., 1951,
11, 55; (c) M. Brust, M. Walker, D. Bethell, D. J. Schiffrin and R.
Whyman, J. Chem. Soc., Chem. Commun., 1994, 801.
16 S. Gómez, K. Philippot, V. Collière, B. Chaudret, F. Senocq and P.
Lecante, Chem. Commun., 2000, 1945.
17 (a) M. I. Bruce, B. K. Nicholson and O. B. Shawkataly, Inorg. Synth.,
1989, 26, 326; (b) K. Angermaier and H. Schmidbaur, Inorg. Chem.,
1994, 33, 2069.
18 (a) Y. Yang and P. R. Sharp, J. Am. Chem. Soc., 1994, 116, 6983; (b) E.
Zeller, H. Beruda and H. Schmidbaur, Inorg. Chem., 1993, 32, 3203; (c)
V. Ramamoorthy, Z. Wu, Y. Yi and P. R. Sharp, J. Am. Chem. Soc.,
1992, 114, 1526.
19 C. D. Keating, M. D. Musick, M. H. Keefe and M. J. Natan, J. Chem.
Educ., 1999, 76, 949.
20 (a) T. Teranishi, I. Kiyokawa and M. Miyake, Adv. Mater., 1998, 10,
596; (b) D. De Caro, V. Agelou, A. Duteil, B. Chaudret, R. Mazel, C.
Roucau and J. S. Bradley, New J. Chem., 1995, 19, 1265.
21 G. Schmid, N. Klein, L. Korste, U. Kreibig and D. Schönauer,
Polyhedron, 1988, 7, 605.
give three phosphino–gold( ) units in auronation reactions.
I
We have carried out experiments under a reducing atmos-
phere of H2, in order to lower the decomposition temperature
and to get cleaner reactions. The decomposition occurred at 40
°C under 3 bar H2 and even at room temperature under 5 bar H2.
31P{1H} NMR spectra of the colloidal solution formed at room
temperature from complex 1, denoted colloid 3, show only
phosphine oxide and a broad signal at 34 ppm. No traces of
[Au(PPh3)2](CF3SO3) were present even at 255 °C. Therefore,
we assume that the full reduction of the oxonium complex
affords free phosphine oxide and gold nanoparticles. Removal
of phosphine oxide was attempted by washing with diethyl
ether, however traces of the phophine oxide remained trapped
within the polymer framework. In the case of complex 2, traces
of [Au(PMe3)2](CF3SO3) were observed on the 31P{1H} NMR
spectra of the corresponding colloidal solution, i.e. colloid 4.
This can be related to the fact that the thermal decomposition of
complex 2 is easier and thus this path could not be totally
suppressed, even under 5 bar of dihydrogen. Surprisingly,
though the decomposition under dihydrogen is easier, the
growth of the gold nanoparticles is less controlled. TEM images
of colloids 3 and 4, evidence bimodal size distributions which
correspond to particles of 2.8 and 6.6 nm mean diameter for
colloid 3, and 3 and 6.2 nm for colloid 4. The nanoparticles are
however stable in solution.
We have further investigated this route in a non-polar organic
medium. As PVP is not readily soluble in such media, we turned
to amine ligands as stabilizing agents. These ligands have been
recently used by our group to stabilise ruthenium22 or nickel
nanoparticles.23 They have been shown to be fluxional22 and to
induce the formation of rod-shape nanoparticles.23 Hence, the
decomposition of complex 1 (3 mg) has been studied in toluene
(3 mL), in the presence of oleylamine (100 mL). Full
decomposition of the gold precursor was achieved after
refluxing the solution for 20 mn: UV–vis spectra are consistent
with the formation of gold nanoparticles, lmax = 528 nm
(colloid 5). TEM images display spherical gold nanoparticles of
narrow size distribution centered around 9 0.5 nm. Inter-
estingly, these nanoparticles tend to self-assemble on the
microscope grid as can be observed in Fig. 1; additionally, the
colloidal solution remained unchanged for several days. This is
different from previous observations for THF solutions of gold
nanoparticles stabilised by primary alkyl amines such as RNH2,
R = C8H17, C12H25 or C16H33.16 Most probably, the bending of
the oleyl chain insures better protection against coalescence.
In conclusion, we have shown that phosphino–gold oxonium
salts can act as gold atom sources in organic media under mild
conditions. In acetonitrile, in the presence of poly(N-vi-
nylpyrrolidone), UV–vis spectroscopy and TEM measurements
evidence the formation of gold nanoparticles of 4.5 nm diameter
22 C. Pan, K. Pelzer, K. Philippot, B. Chaudret, F. Dassenoy, P. Lecante
and M.-J. Casanove, J. Am. Chem. Soc., 2001, 123, 7584.
23 N. Cordente, M. Respaud, F. Senocq, M.-J. Casanove, C. Amiens and B.
Chaudret, Nano Lett., 2001, 1, 565.
CHEM. COMMUN., 2002, 598–599
599