Polynuclear gold complexes with bridging selenido ligands. Theoretical studies of gold-gold interactions

Article abstract of DOI:10.1021/om000119g

The compounds [Se(AuPPh₃)₂] and [Se(Au₂dppf)] are good precursors for the synthesis of highly aurated selenium-centered derivatives. Thus the reaction of [Se(AuPPh₃)₂] with various molar ratios of [Au(OTf)PPh₃] affords the homoleptic species [Se(AuPPh₃)_n](OTf)_n-2 (n = 3-6). Similarly, the treatment of [Se(Au₂dppf)] with various stoichiometric ratios of [Au₂(OTf)₂(μ-dppf)] gives the polynuclear derivatives [(Au₂dppf){Se(Au₂dppf)}₂](OTf)₂, [Se(Au₂dppf)₂](OTf)₂, and [Se(Au₂dppf)₃](OTf₄. The compounds [Au{Se(AuPPh₃)₂}₂]ClO₄ and [Au{Se(Au₂dppf)}₂]ClO₄ have been obtained by reaction of the precursors with [Au(tht)₂]-ClO₄ in a molar ratio of 2:1. Some of these derivatives have been characterized by X-ray diffraction studies and show unusual geometries, associated with gold-gold interactions. The latter have been studied theoretically using [Se(AuPH₃)_n]^(n-2)+ (n = 2-6) models at the HF and MP2 levels with quasi-relativistic pseudopotentials. There is a good agreement between experimental and theoretical geometries.

Full text of DOI:10.1021/om000119g

Organometallics 2000, 19, 4985-4994
4985
P olyn u clea r Gold Com p lexes w ith Br id gin g Selen id o
Liga n d s. Th eor etica l Stu d ies of Gold -Gold In ter a ction s
Silvia Canales,† Olga Crespo,† M. Concepcio´n Gimeno,† Peter G. J ones,‡
Antonio Laguna,*,† and Fernando Mendizabal§
Departamento de Quı´mica Inorga´nica, Instituto de Ciencia de Materiales de Arago´n,
Universidad de Zaragoza-CSIC, E-50009 Zaragoza, Spain, and Institut fu¨r Anorganische
und Analytische Chemie der Technischen Universita¨t, Postfach 3329,
D-38023 Braunschweig, Germany, and Facultad de Ciencias, Universidad de Chile,
Casilla 653, Santiago de Chile, Chile
Received February 8, 2000
The compounds [Se(AuPPh3)2] and [Se(Au2dppf)] are good precursors for the synthesis of
highly aurated selenium-centered derivatives. Thus the reaction of [Se(AuPPh3)2] with
various molar ratios of [Au(OTf)PPh3] affords the homoleptic species [Se(AuPPh3)n](OTf)n-2
(n ) 3-6). Similarly, the treatment of [Se(Au2dppf)] with various stoichiometric ratios of
[Au2(OTf)2(µ-dppf)] gives the polynuclear derivatives [(Au2dppf){Se(Au2dppf)}2](OTf)2, [Se-
(Au2dppf)2](OTf)2, and [Se(Au2dppf)3](OTf)4. The compounds [Au{Se(AuPPh3)2}2]ClO4 and
[Au{Se(Au2dppf)}2]ClO4 have been obtained by reaction of the precursors with [Au(tht)2]-
ClO4 in a molar ratio of 2:1. Some of these derivatives have been characterized by X-ray
diffraction studies and show unusual geometries, associated with gold-gold interactions.
The latter have been studied theoretically using [Se(AuPH3)n](n-2)+ (n ) 2-6) models at the
HF and MP2 levels with quasi-relativistic pseudopotentials. There is a good agreement
between experimental and theoretical geometries.
In tr od u ction
metallic chemistry.1 P. Pyykko¨ has summarized the
available theoretical and experimental evidence of intra-
and intermolecular interactions between heavy metal
atoms, such as Au(I).3 In general, when heavy metals
are involved in such interactions, the closed-shell at-
tractions are of metallophilic or van der Waals type, as
has been found for d8, d10 coinage-metals, and d10s2
centers. Experimental evidence for the aurophilic Au-
(I)- - -Au(I) attraction is available from crystallography,
NMR and Raman spectroscopies, and optical spectro-
scopic measurements of the interaction strength.4 It is
manifested in molecular conformations with relatively
close Au- - -Au interactions of ca. 3 Å, whose strength
has been estimated at ca. 33 kJ /mol.5 These attractions
are weaker than most covalent or ionic bonds, but
stronger than other van der Waals bonds and compa-
rable in strength to typical hydrogen bonds. The phe-
nomenon cannot be satisfactorily explained by classical
theories such as hybridization.3
In the past few years a great deal of attention has
been paid to the chemistry of polynuclear gold deriva-
tives with a central heteroatom, not only from the
experimental and structural but also from the theoreti-
cal standpoint.1 Interesting hypercoordinated species of
the type [C(AuPR3)5]+, [C(AuPR3)6]2+, [N(AuPR3)5]2+
,
[P(AuPR3)5]2+, and [P(AuPR3)6]3+ have been described,2
apart from other complexes with more conventional
stoichiometry, and all have in common the presence of
gold-gold interactions of ca. 3 Å. Usually, the chemistry
of the first-row elements of the p-block is known to
follow classical rules of bonding, and only in cases of
extreme electron deficiency does the traditional electron
count need to be reconsidered to account for special
types of molecular or solid state structures. Many of
these heteroatom-centered complexes are electron-
deficient, and the gold-gold interactions provide a
significant contribution to their stability.
From a theoretical point of view, the attractions close
to equilibrium distances are caused by dispersive in-
The attraction between these closed-shell systems has
recently captured attention in inorganic and organo-
† Universidad de Zaragoza-CSIC.
(3) Pyykko¨, P. Chem. Rev. 1997, 97, 597.
‡ Institut fu¨r Anorganische und Analytische Chemie der Technis-
chen Universita¨t Braunschweig.
(4) (a) J ones, P. G. Gold Bull. 1981, 14, 102. (b) Schmidbaur, H.
Gold Bull. 1990, 23, 11. (c) Gade, L. H. Angew. Chem., Int. Ed. Engl.
1997, 36, 1171. (d) Mansour, M. A.; Connick, W. B.; Lachicotte, R. J .;
Gysling, H. J .; Eisenberg, R. J . J . Am. Chem. Soc. 1998, 120, 1329. (e)
Canales, F.; Gimeno, M. C.; Laguna, A.; J ones, P. G. J . Am. Chem.
Soc., 1996, 118, 4839.
(5) (a) Dziwok, K.; Lachmann, J .; Wilkinson, D. L.; Mu¨ller, G.;
Schmidbaur, H. Chem. Ber. 1990, 123, 423. (b) Schmidbaur, H.; Graf,
W.; Mu¨ller, G. Angew. Chem., Int. Ed. Engl. 1988, 27, 417. (c)
Narayanaswamy, R.; Young, M. A.; Parkhurst, E.; Ouellette, M.; Kerr,
M. E.; Ho, D. M.; Elder, R. C.; Bruce, A. E.; Bruce, M. R. M. Inorg.
Chem. 1993, 32, 2506. (d) Harwell, M. E.; Mortimer, M. D.; Knobler,
C. B.; Anet, F. A. L.; Hawthorne, M. F. J . Am. Chem. Soc. 1996, 118,
2679.
§ Universidad de Chile.
(1) (a) Schmidbaur, H. Chem. Soc. Rev. 1995, 391. (b) Laguna, A.
In Gold, Progress in Chemistry, Biochemistry and Technology, Inter-
national; Schmidbaur, H., Ed.; Wiley: New York, 1999. (c) Schmidbaur,
H. Pure Appl. Chem. 1993, 65, 691.
(2) (a) Scherbaum, F.; Grohmann, A.; Mu¨ller, G.; Schmidbaur, H.
Angew. Chem., Int. Ed. Engl. 1989, 28, 463. (b) Scherbaum, F.;
Grohmann, A.; Huber, B.; Kru¨ger, C.; Schmidbaur, H. Angew. Chem.,
Int. Ed. Engl. 1988, 27, 1544. (c) Grohmann, A.; Riede, J .; Schmidbaur,
H. Nature 1990, 345, 140. (d) Schmidbaur, H.; Weidenhiller, G.;
Steigelmann, O. Angew. Chem., Int. Ed. Engl. 1991, 30, 433. (e) Zeller,
E.; Schmidbaur, H. J . Chem. Soc., Chem. Commun. 1993, 69.
10.1021/om000119g CCC: $19.00 © 2000 American Chemical Society
Publication on Web 10/28/2000

4986 Organometallics, Vol. 19, No. 24, 2000
Canales et al.
Sch em e 1a
a
(i) Na2CO3, MeOH, (ii) [Au(OTf)PPh3], (iii) 2 [Au(OTf)PPh3], (iv) 1/2 [Au(tht)2]ClO4, (v) 3 [Au(OTf)PPh3], (vi) 4 [Au(OTf)PPh3].
teractions together with relativistic effects.6 The attrac-
tive contributions are dominated by pair-excitations
from the heavy metal, such as gold 5d, strengthened
by relativistic effects. In all the cases where there are
inter- or intramolecular metal-metal attractions, the
correlation effects are essential; if they are omitted, at
the Hartree-Fock (HF) level, repulsive interactions are
obtained.6 Hence, it is necessary to use at least second-
order Møller-Plesset level (MP2) methods for the
dispersive forces, which are included in the electronic
correlation effects.7
Although the chemistry of carbon-, nitrogen-, phos-
phorus-, or arsenic-centered species has developed
rapidly, the corresponding chemistry of chalcogen de-
rivatives is still growing. We have previously reported
the synthesis and structural characterization of sulfur-
centered complexes,4e,8 and here we describe the syn-
thesis of polynuclear gold complexes with selenium as
the central heteroatom. We have also studied theoreti-
cally the gold-gold interactions in these complexes. The
synthesis of the compound [Se(AuPPh3)4](OTf)2 with a
quadruply bridging selenido ligand has been the subject
of a preliminary communication.9
Resu lts a n d Discu ssion
We have previously reported that the sulfur deriva-
tives [S(AuPR3)2]10 or [S(Au2dppf)] (dppf ) 1,1′-bis-
(diphenylphosphino)ferrocene)4e are excellent starting
materials for the synthesis of highly aurated homoleptic
sulfur-centered derivatives. Our aim was to obtain
similar complexes with selenium as central element.
The complex [Se(AuPPh3)2] 1 was prepared in high yield
by reacting [AuCl(PPh3)] with selenourea followed by
basic hydrolysis.11 We have studied further auration
around the selenium center by reaction of 1 with various
stoichiometric ratios of [Au(OTf)PPh3] to give the tri-
nuclear [Se(AuPPh3)3]OTf 2, tetranuclear [Se(Au-
PPh3)4](OTf)2 3, pentanuclear [Se(AuPPh3)5](OTf)3 4, or
hexanuclear [Se(AuPPh3)6](OTf)4 5 species (see Scheme
1).
The products are air-stable colorless solids, insoluble
in nonpolar solvents, and readily soluble in dichlo-
romethane or chloroform. The conductivities of the
complexes measured in acetone are in agreement with
the proposed formulas. Although there is not many data
for conductivities of highly charged compounds, the tri-
and tetracationic species show values similar to those
found in the sulfur-centered complexes.8b
(6) (a) Pyykko¨, P.; Zhao, Y.-F. Angew. Chem. 1991, 103, 622. (b)
Pyykko¨, P.; Li, L.; Runenberg, N. Chem. Phys. Lett. 1994, 218, 133. (c)
Pyykko¨, P.; Mendizabal, F. Inorg. Chem. 1998, 37, 3018. (d) Runenberg,
N.; Schu¨tz, M.; Werner, H.-J . J . Chem. Phys. 1999, 110, 7210. (e)
Pyykko¨, P.; Schneider, W.; Bauer, A.; Bayler, A.; Schmidbaur, H. Chem.
Commun. 1997, 1111.
(7) Møller, C.; Plesset, M. S. J . Chem. Phys. 1934, 46, 618.
(8) (a) Canales, F.; Gimeno, M. C.; J ones, P. G.; Laguna, A. Angew.
Chem., Int. Ed. Engl. 1994, 33, 769. (b) Canales, F.; Gimeno, M. C.;
Laguna, A.; Villacampa, M. D. Inorg. Chim. Acta 1996, 244, 95. (c)
Canales, F.; Gimeno, M. C.; Laguna, A.; J ones, P. G. Organometallics
1996, 15, 3412. (d) Calhorda, M. J .; Canales, F.; Gimeno, M. C.;
J ime´nez, J .; J ones, P. G.; Laguna, A.; Veiros, L. F. Organometallics
1997, 16, 3837. (e) Canales, F.; Canales, S.; Crespo, O.; Gimeno, M.
C.; J ones, P. G.; Laguna, A. Organometallics 1998, 17, 1617.
(9) Canales, S.; Crespo, O.; Gimeno, M. C.; J ones, P. G.; Laguna A.
J . Chem. Soc., Chem. Commun. 1999, 679.
The 31P(1H) NMR spectra for complexes 1-5 show a
sharp single resonance, indicating the equivalence of all
the phosphine groups. The variable-temperature experi-
+
ments confirm that all the AuPR3 fragments are
equivalent, maybe because of a rapid exchange in
solution even at low temperature. An upfield displace-
ment is observed when a new gold fragment coordinates
to the selenium center. We have previously observed
this upfield shift in the sulfur derivatives.8b
The structure of the [Se(AuPPh3)3]+ unit is known and
is analogous to that in the oxonium and sulfonium salts
(10) Lensch, C.; J ones, P. G.; Sheldrick, G. M. Z. Naturforsch., Teil
B 1982, 37b, 944.
(11) J ones, P. G.; Tho¨ne, C. Chem. Ber. 1991, 124, 2725.

Polynumclear Gold Complexes
Organometallics, Vol. 19, No. 24, 2000 4987
Sch em e 2a
a(i) Na2CO3, MeOH, (ii) [Au(OTf)PPh3] or [Au(OTf)PPh2Me], (iii) 2 [Au(OTf)PPh3] or 2 [Au(OTf)PPh2Me], (iv) 1/2 [Au(tht)2]ClO4,
(v) 2 [Au2(OTf)2(µ-dppf)], (vi) [Au2(OTf)2(µ-dppf)], (vii) 1/2 [Au2(OTf)2(µ-dppf)].
and consists of trigonal pyramids with basal gold atoms
and selenium at the apex.10 The synthesis and struc-
tural characterization of the tetragold species has been
previously communicated by us;9 the structure shows a
square pyramidal geometry with the selenium atom in
the apical position and in the base the four gold atoms
forming short gold-gold interactions. We believe that
the structures of the complexes with five and six gold
atoms could be trigonal pyramidal and octahedral,
respectively, with close gold-gold distances. The high
charge of the cations has however prevented the growth
of single crystals for an unambiguous determination.
The reaction of 1 with [Au(tht)2]ClO4 in a molar ratio
F igu r e 1. Structure of complex 7 in the crystal with the
atom-numbering scheme. Displacement parameter ellip-
soids represent 50% probability surfaces. The H atoms are
omitted for clarity.
of 2:1 leads to the pentanuclear complex [Au{Se-
(AuPPh3)2}2] 6. Complex 6 is an air-stable white solid
that behaves as a 1:1 electrolyte in acetone solutions.
The 31P(1H) NMR spectrum presents one singlet for all
phosphorus atoms even at low temperature. In the
positive-ion liquid secondary mass spectrum (LSIMS)
the cation molecular peak appears at m/z ) 2193 (15%)
with coincident experimental and isotopic distribution.
Treatment of [Au2Cl2(µ-dppf)] with selenourea fol-
lowed by basic hydrolysis affords the complex [Se(Au2-
dppf)] 7, in which the two gold atoms are bridged by
the 1,1′-bis(diphenylphosphino)ferrocene ligand and the
selenium atom. Complex 7 is an orange air-stable solid
that is nonconducting in acetone solutions.
The NMR data are consistent with the formulation,
and only one signal is observed in the 31P(1H) NMR
spectrum, because of the equivalence of both phosphorus
atoms. In the 1H NMR spectrum, apart from the
multiplets for the phenyl protons, two broad multiplets
appear for the R and â protons of the cyclopentadienyl
rings. The low-temperature spectrum presents four
signals for the cyclopentadienyl protons, which is a
typical behavior of 1,1′-disubstituted ferrocene deriva-
tives; the protons are strictly not equivalent but are
rendered so by fluxionality processes. Up to eight signals
Ta ble 1. Selected Bon d Len gth s [Å] a n d An gles
[d eg] for Com p ou n d 7
Au(1)-P(2)
Au(1)-Se
Au(1)-Au(2)
2.263(3)
2.4218(11)
2.9353(8)
Au(2)-P(1)
Au(2)-Se
2.254(3)
2.4055(11)
P(2)-Au(1)-Se
175.47(8) C(8)-P(1)-Au(2)
109.1(3)
P(2)-Au(1)-Au(2) 124.64(7) C(21)-P(1)-Au(2) 112.9(3)
Se-Au(1)-Au(2)
P(1)-Au(2)-Se
52.30(3) C(11)-P(1)-Au(2) 119.1(3)
170.67(7) C(3)-P(2)-Au(1) 112.9(3)
P(1)-Au(2)-Au(1) 117.93(7) C(31)-P(2)-Au(1) 113.2(3)
Se-Au(2)-Au(1)
Au(2)-Se-Au(1)
52.80(3) C(41)-P(2)-Au(1) 117.2(3)
74.90(3)
are observed in the low-temperature spectra of these
derivatives. The LSIMS shows the molecular peak at
m/z ) 1028 (87%).
The crystal structure of complex 7 has been deter-
mined by X-ray diffraction studies, and the molecule is
shown in Figure 1. Selected bond lengths and angles
are collected in Table 1. The selenido ligand is bonded
to the gold atoms of an “Au2(µ-dppf)” unit with a gold-
gold contact of 2.9353(8) Å. This value is shorter than

4988 Organometallics, Vol. 19, No. 24, 2000
Canales et al.
that found for [Se(AuPPh3)2] (3.051(1) Å).11 An extreme
distortion of ideal geometry at the selenium center [Au-
Se-Au ) 74.90(3)°] is observed, even more so than for
[Se(AuPPh3)2] [Au-Se-Au ) 79.1(1)°].11
The presence in 7 of a bridging ligand, with less steric
demand than two triphenylphosphine ligands, probably
allows a shorter gold-gold contact and a more pro-
nounced narrowing of the angle at selenium. The same
effects are observed in [S(Au2dppf)]4e (shorter gold-gold
contact and more narrowing) related to [S(AuPPh3)2].10
A comparison of 7 with [S(Au2dppf)] reveals a larger
angle Au-E-Au (E ) S, Se) for the sulfido ligand. This
may be associated with the presence of more diffuse 4p
orbitals in the selenium derivative or a greater electron
repulsion of the lone pairs. The distortion of linear
geometry in the gold atoms is more marked for Au(2)
[P(1)-Au(2)-Se ) 170.67(7)°] than for Au(1) [P(2)-Au-
(1)-Se ) 175.47(8)°]. The distances Au-Se (Au(1)-Se
) 2.4218(11), Au(2)-Se ) 2.4055(11) Å) are dissimilar
but longer than those found for [Se(AuPPh3)2] (2.394-
(1), 2.398(1) Å).11 The distances Au-P are also margin-
ally dissimilar (Au(1)-P(2) ) 2.263(3), Au(2)-P(1) )
2.254(3) Å); the shorter compares well with those in [Se-
(AuPPh3)2] (2.254(2), 2.255(2) Å).11
F igu r e 2. Structure of the cation of complex 12b. Dis-
placement parameter ellipsoids represent 50% probability
surfaces. Carbon atoms are spheres of arbitrary radius. The
H atoms are omitted for clarity.
Complex 7 can serve as a building block for preparing
polynuclear selenium-centered derivatives (Scheme 2).
We have studied the reactions of 7 with several gold(I)
compounds. Thus the treatment of 7 with 1 or 2 equiv
of [Au(OTf)PR3] affords the complexes [Se(Au2dppf)-
(AuPR3)]OTf (PR3 ) PPh3 8, PPh2Me 9) or [Se(Au2dppf)-
(AuPR3)2](OTf)2 (PR3 ) PPh3 10, PPh2Me 11). They
behave as 1:1 or 1:2 electrolytes in acetone solution,
respectively.
spectrum is very complicated as a consequence of the
inequivalence of all protons and phosphorus nucleus.
In the H NMR spectrum several signals appear in the
region of the Cp protons, 16 in total, some of which are
overlapped. The 31P NMR spectrum presents four
resonances for the four different phosphorus environ-
ments. In the positive-ion liquid secondary mass spec-
trum the cation molecular peak appears at m/z ) 2251
(100%).
1
1
The H NMR spectra show multiplets for the phenyl
All attempts to crystallize complex 12a failed, and
then we added to a solution of complex 12a another
anion in order to get single crystals. We have deter-
mined the crystal structure of this complex with the
anion NO3- (Figure 2), 12b; a selection of bond lengths
and angles are collected in Table 2. The structure
displays two [Au{Se(Au2dppf)}2]+ moieties joined through
gold-gold interactions and related by crystallographic
2-fold symmetry. There are several gold-gold contacts
in the range 2.908(4)-3.182(3) Å; the shortest is be-
tween the gold atoms bridging the [Se(Au2dppf)] units,
Au(5)-Au(5)# 2.908(4) Å (# -x+1, y, -z+1/2). The
selenium atoms act as triply bridging ligands with
dissimilar distances to the gold atoms; for the gold
atoms in the [Se(Au2dppf)] units the distances are of
the same order, 2.426(4)-2.462(4) Å, and are shorter
for the gold atoms bridging these units, 2.396(4) and
2.398(4) Å; this is presumably attributable to the greater
trans influence of P donors. The latter is similar to the
shorter distance found in complex 7 (with a two-
coordinate selenium donor atom) and the former to those
in [Se(AuPPh3)3]PF6. The angles around the selenium
atoms are very irregular; those in the [Se(Au2dppf)]
units are narrower, 74.33(11)° and 75.21(11)°. All the
gold atoms display a slightly distorted linear geometry
(ignoring the gold-gold contacts).
protons and two multiplets for the cyclopentadienyl
protons; compounds 9 and 11 also present a doublet for
the methyl group of the phosphine, the integral ratio
agreeing with the proposed formulas. The 31P(1H)
spectra show two signals because of the two different
phosphorus environments. In the positive-ion liquid
secondary mass spectra the cation molecular peaks, [M
- OTf]+, appear for complex 8 at m/z ) 1486 (100%)
and for compound 9 at m/z ) 1425 (100%); for complex
10 the peak [Se(Au2dppf)(AuPPh3)2]+ appears at m/z )
1945, although with very low intensity, and is absent
in the spectrum of complex 11.
Although these complexes have not been character-
ized by X-ray diffraction, we assume that compounds 8
and 9 have a structural framework similar to that found
in the complex [Se(AuPPh3)3]+, with short contacts
between the gold(I) centers, whereas 10 and 11 could
have a structure similar to [Se(AuPPh3)4]2+
.
In the mass spectra of these selenium-centered de-
rivatives the fragment corresponding to the cation [Au-
{Se(Au2dppf)}2]+ always appears with high intensity.
This indicates the stability of this compound, and we
have synthesized it by treatment of 2 equiv of [Se(Au2-
dppf)] with 1 equiv of [Au(tht)2]ClO4. The reaction leads
to the compound [Au{Se(Au2dppf)}2]ClO4 12a as an
orange solid that behaves as a 1:1 electrolyte in acetone
solution.
We have also synthesized a hexanuclear derivative
that contains two SeAu3 cores connected through a
bridging dppf ligand. The reaction of complex 7 with a
freshly prepared solution of [Au2(OTf)2(µ-dppf)] in dichlo-
romethane (molar ratio 2:1) affords [(Au2dppf){Se(Au2-
The NMR spectra at room temperature show broad
bands for the cyclopentadienyl protons and a broad
singlet for the phosphorus atoms; the low-temperature

Polynumclear Gold Complexes
Organometallics, Vol. 19, No. 24, 2000 4989
Ta ble 2. Selected Bon d Len gth s [Å] a n d An gles [d eg] for Com p ou n d 12ba
Au(1)-P(1)
Au(1)-Se(1)
Au(1)-Au(1)#1
Au(1)-Au(2)
Au(1)-Au(5)
Au(2)-P(2)
Au(2)-Se(1)
Au(3)-P(3)
Au(3)-Se(2)
2.291(10)
2.465(4)
2.920(3)
2.985(2)
3.041(3)
2.264(10)
2.426(4)
2.278(10)
2.462(4)
Au(3)-Au(4)
Au(3)-Au(3)#1
Au(3)-Au(5)
Au(4)-P(4)
Au(4)-Se(2)
Au(5)-Se(2)#1
Au(5)-Se(1)
Au(5)-Au(5)#1
2.960(2)
3.015(3)
3.182(3)
2.255(10)
2.437(4)
2.396(4)
2.398(4)
2.908(4)
P(1)-Au(1)-Se(1)
P(1)-Au(1)-Au(1)#1
Se(1)-Au(1)-Au(1)#1
P(1)-Au(1)-Au(2)
Se(1)-Au(1)-Au(2)
Au(1)#1-Au(1)-Au(2)
P(1)-Au(1)-Au(5)
Se(1)-Au(1)-Au(5)
Au(1)#1-Au(1)-Au(5)
Au(2)-Au(1)-Au(5)
P(2)-Au(2)-Se(1)
P(2)-Au(2)-Au(1)
Se(1)-Au(2)-Au(1)
P(3)-Au(3)-Se(2)
P(3)-Au(3)-Au(4)
Se(2)-Au(3)-Au(4)
P(3)-Au(3)-Au(3)#1
Se(2)-Au(3)-Au(3)#1
Au(4)-Au(3)-Au(3)#1
P(3)-Au(3)-Au(5)
Se(2)-Au(3)-Au(5)
172.1(3)
99.1(2)
Au(4)-Au(3)-Au(5)
Au(3)#1-Au(3)-Au(5)
P(4)-Au(4)-Se(2)
85.37(6)
70.38(4)
174.4(3)
88.67(10)
120.3(2)
51.80(9)
139.91(6)
130.5(3)
50.33(10)
89.88(4)
70.85(6)
175.6(2)
122.6(2)
52.99(9)
163.4(3)
119.5(2)
52.45(9)
108.3(2)
83.71(10)
130.92(6)
106.7(3)
87.81(11)
P(4)-Au(4)-Au(3)
123.7(3)
Se(2)-Au(4)-Au(3)
Se(2)#1-Au(5)-Se(1)
Se(2)#1-Au(5)-Au(5)#1
Se(1)-Au(5)-Au(5)#1
Se(2)#1-Au(5)-Au(1)
Se(1)-Au(5)-Au(1)
Au(5)#1-Au(5)-Au(1)
Se(2)#1-Au(5)-Au(3)
Se(1)-Au(5)-Au(3)
Au(5)#1-Au(5)-Au(3)
Au(1)-Au(5)-Au(3)
Au(5)-Se(1)-Au(2)
Au(5)-Se(1)-Au(1)
Au(2)-Se(1)-Au(1)
Au(5)#1-Se(2)-Au(4)
Au(5)#1-Se(2)-Au(3)
Au(4)-Se(2)-Au(3)
53.22(10)
175.44(15)
95.73(11)
88.21(11)
129.78(12)
52.30(10)
90.12(4)
81.17(11)
97.95(10)
71.71(5)
146.41(7)
92.76(13)
77.38(11)
75.21(11)
111.04(17)
94.71(13)
74.33(11)
a
Symmetry transformation used to generate equivalent atoms: #1 -x+1, y, -z+1/2.
dppf)}2](OTf)2 13. Compound 13 is an orange solid that
behaves as a 2:1 electrolyte in an acetone solution.
The room-temperature NMR spectra show broad
bands that sharpen when the experiment is carried out
1
at low temperature. The H NMR spectrum shows five
close multiplets for the Cp protons in a ratio of 4:4:8:
4:4, although some of the resonances may be overlapped.
The 31P(1H) NMR spectrum presents three different
phosphorus environments (in each SeAu3P3 unit). The
inequivalence of these phosphorus and hydrogen atoms
could be a consequence of different gold-gold interac-
tions, as mentioned for complex 12a . In the positive-
ion LSIMS spectrum the dication molecular peak ap-
pears at m/z ) 3154 (61%).
The structure of complex 13 has been confirmed by
an X-ray diffraction study, and the cation is shown in
Figure 3. Selected bond lengths and angles are given
in Table 3. Compound 13 crystallizes with four dichlo-
romethane molecules. The structure of the cation con-
sists of an Se2Au6 core that shows short intramolecular
Au-Au interactions. These distances vary from 2.9028-
(5) to 3.4325(6) Å. The shortest contacts, Au(5)-Au(6)
) 2.9028(5) Å, Au(3)-Au(6) ) 2.9779(6) Å, and Au(3)-
Au(4) ) 3.0088(5) Å, involve gold atoms of the [Au2-
dppf]2+ fragment that are bonded to the same selenium
atom. The selenium atoms lie 1.566 and 1.623 Å out of
the planes formed by their gold neighbors. The Au-Se-
Au angles are very different: at Se(1) two of the angles
are ca. 86°, whereas Au(4)-Se(1)-Au(3) is 75.86(3)°; at
Se(2) the values range from 72.43(3)° for Au(5)-Se(2)-
Au(6) to 89.02(3)° for Au(5)-Se(2)-Au(2), in accordance
with the longest contact value found for Au(5)-Au(2).
The Au-P and Au-Se distances, which range from
2.259(3) to 2.280(3) Å and 2.4240(10) to 2.4704(10) Å,
compare well with those found in [Se(AuPPh3)3]PF6.10
The gold atoms display distorted linear geometries with
F igu r e 3. Perspective view of the cation of complex 13.
Displacement parameter ellipsoids represent 50% prob-
ability surfaces. The H atoms are omitted for clarity.
deviations from linearity [Se-Au-P from 174.68(7)° to
172.17(8)°] similar to those found in [Se(AuPPh3)3]PF6
[Se-Au-P from 176.9(1)° to 172.7(1)°], except for Au-
(2) [Se(2)-Au(2)-P(2) ) 164.40(7)°], which shows con-
siderable distortion.
Finally, we have studied the reaction of 7 with [Au2-
(OTf)2(µ-dppf)] in 1:2 and 1:3 molar ratios to give the
polynuclear derivatives [Se(Au2dppf)2](OTf)2 14 and [Se-
(Au2dppf)3](OTf)4 15. They are orange solids that show
a conductivity value of a 1:2 electrolyte (14), and for 15
the value is similar to that obtained for compound 5.
Complexes 14 and 15 show two broad resonances for

4990 Organometallics, Vol. 19, No. 24, 2000
Canales et al.
the cyclopentadienyl protons in the 1H NMR spectra and
one singlet for all the phosphorus atoms in the 31P(1H)
NMR spectra. The low-temperature NMR spectra for
complex 14 has been measured in CDCl3 at -55 °C and
in CD2Cl2 at -85 °C, and the results are similar. In the
31P(1H) NMR spectrum the singlet splits into several
resonances, from which we have identified those belong-
ing to complex 13. We have observed that in this type
of complex fast equilibrium processes are also present,
and we accordingly believe that compound 14 could be
in equilibrium with complex 13 and [Au2(OTf)2(µ-dppf)]
(see eq 1). For complex 15 the 31P(1H) NMR spectrum
shows four signals; these can be attributed to different
phosphorus environments and also to the presence of
the fac and mer isomers.
Ta ble 3. Selected Bon d Len gth s [Å] a n d An gles
[d eg] for Com p ou n d 13
Au(1)-P(1)
Au(1)-Se(1)
Au(1)-Au(2)
Au(1)-Au(4)
Au(1)-Au(3)
Au(2)-P(2)
Au(2)-Se(2)
Au(2)-Au(6)
Au(2)-Au(5)
Au(3)-P(3)
2.277(2)
Au(6)-Se(2)
Au(3)-Se(1)
Au(3)-Au(6)
Au(3)-Au(4)
Au(4)-P(4)
Au(4)-Se(1)
Au(5)-P(5)
Au(5)-Se(2)
Au(5)-Au(6)
Au(6)-P(6)
2.4658(9)
2.4704(10)
2.9779(6)
3.0088(5)
2.259(3)
2.4240(10)
2.262(3)
2.4474(10)
2.9028(5)
2.267(2)
2.4420(9)
3.1514(6)
3.3392(5)
3.3775(6)
2.269(3)
2.4487(10)
3.1512(5)
3.4325(6)
2.280(3)
P(1)-Au(1)-Se(1)
P(1)-Au(1)-Au(2)
Se(1)-Au(1)-Au(2)
P(1)-Au(1)-Au(4)
Se(1)-Au(1)-Au(4)
172.53(7)
104.70(6)
80.86(2)
128.63(6)
46.44(2)
P(5)-Au(5)-Au(6)
Se(2)-Au(5)-Au(6)
P(5)-Au(5)-Au(2)
Se(2)-Au(5)-Au(2)
123.48(6)
54.08(2)
141.42(7)
45.50(2)
Au(6)-Au(5)-Au(2) 58.951(12)
Au(2)-Au(1)-Au(4) 126.546(14) P(6)-Au(6)-Se(2)
172.29(6)
118.91(6)
53.49(2)
103.49(6)
83.57(3)
49.88(2)
P(1)-Au(1)-Au(3)
Se(1)-Au(1)-Au(3)
Au(2)-Au(1)-Au(3) 86.748(13) P(6)-Au(6)-Au(3)
Au(4)-Au(1)-Au(3) 53.222(11) Se(2)-Au(6)-Au(3)
P(2)-Au(2)-Se(2)
P(2)-Au(2)-Au(6)
Se(2)-Au(2)-Au(6)
P(2)-Au(2)-Au(1)
Se(2)-Au(2)-Au(1)
Au(6)-Au(2)-Au(1) 88.754(13) Au(1)-Se(1)-Au(3)
P(2)-Au(2)-Au(5)
Se(2)-Au(2)-Au(5)
Au(6)-Au(2)-Au(5) 52.110(11) Au(2)-Se(2)-Au(6)
Au(1)-Au(2)-Au(5) 132.605(16) C(131)-P(5)-Au(5) 116.7(3)
137.29(7)
46.92(2)
P(6)-Au(6)-Au(5)
Se(2)-Au(6)-Au(5)
164.40(7)
136.88(6)
50.36(2)
102.73(6)
90.47(3)
Se(2)-Au(6)-Au(2)
Au(5)-Au(6)-Au(2) 68.939(13)
Au(3)-Au(6)-Au(2) 94.074(14)
Au(4)-Se(1)-Au(1)
Au(4)-Se(1)-Au(3)
86.66(3)
75.86(3)
86.87(3)
89.02(3)
72.43(3)
79.76(3)
123.80(6)
45.47(2)
Au(5)-Se(2)-Au(2)
Au(5)-Se(2)-Au(6)
P(3)-Au(3)-Se(1)
P(3)-Au(3)-Au(6)
Se(1)-Au(3)-Au(6)
P(3)-Au(3)-Au(4)
Se(1)-Au(3)-Au(4)
172.19(7)
115.31(6)
72.49(2)
120.91(6)
51.37(2)
C(111)-P(5)-Au(5) 112.6(3)
C(121)-P(5)-Au(5) 108.7(4)
C(51)-P(3)-Au(3)
C(81)-P(3)-Au(3)
C(71)-P(3)-Au(3)
110.0(3)
117.0(3)
112.3(3)
111.1(3)
107.1(3)
120.9(3)
111.5(3)
117.1(3)
111.5(3)
107.9(3)
113.1(3)
Au(6)-Au(3)-Au(4) 122.876(16) C(6)-P(2)-Au(2)
P(3)-Au(3)-Au(1)
Se(1)-Au(3)-Au(1)
Au(6)-Au(3)-Au(1) 87.589(14) C(1)-P(1)-Au(1)
Au(4)-Au(3)-Au(1) 62.736(12) C(21)-P(1)-Au(1)
P(4)-Au(4)-Se(1)
P(4)-Au(4)-Au(3)
Se(1)-Au(4)-Au(3)
P(4)-Au(4)-Au(1)
Se(1)-Au(4)-Au(1)
131.69(6)
46.21(2)
C(31)-P(2)-Au(2)
C(41)-P(2)-Au(2)
In the positive-ion liquid secondary mass spectrum
(LSIMS+) of complex 14 the fragment corresponding to
the molecular peak less a triflate anion appears, [Se-
(Au2dppf)2OTf]+, at m/z ) 1975 (15%). For complex 15
the molecular peak does not appear probably due to the
high charge, and only fragmentation peaks are ob-
served.
174.68(7)
122.35(6)
52.77(2)
130.55(6)
46.89(2)
C(11)-P(1)-Au(1)
C(56)-P(4)-Au(4)
C(91)-P(4)-Au(4)
C(101)-P(4)-Au(4) 115.1(3)
C(116)-P(6)-Au(6) 109.6(3)
Au(3)-Au(4)-Au(1) 64.042(12) C(141)-P(6)-Au(6) 113.2(3)
P(5)-Au(5)-Se(2) 172.17(8) C(151)-P(6)-Au(6) 111.8(3)
the selenium-selenium distances are longer than gold-
gold distances when considering two “SeAu3” cores.
The crystal structure of compound 14 has been
studied by X-ray diffraction, but the results are made
very imprecise by the presence of channels containing
large amounts of solvent together with one triflate
anion; no single chemical species could be identified in
these channels. In view of this, we do not present full
data. The cation is shown in Figure 4. Complex 14
crystallizes in the monoclinic space group P21/n with
unit cell dimensions a ) 17.3845, b ) 15.2513, c )
33.1138 Å, â ) 97.478°. The structure is similar to that
found for [Se(AuPPh3)4](OTf)2.9 The gold atoms in the
SeAu4 core exhibit gold-gold interactions that range
from 2.889 to 2.936 Å (2.8959(8)-2.9605(8) Å in [Se-
(AuPPh3)4](OTf)2. The Se-Au distances (from 2.492 to
2.511 Å) also compare well with those obtained for [Se-
(AuPPh3)4](OTf)2 (2.4654(13)-2.54347(14) Å). The cat-
ions are paired across symmetry centers to form loose
dimers (Figure 5). The shortest intercation distances
(Au1‚‚‚Au4a ) 3.298 Å) are similar in length to the Se‚
‚‚Au3′ contacts found for the dimers in [Se(AuPPh3)4]-
(OTf)2. This is the main difference between the two
structures, and 14 thus resembles more closely the
triply bridging selenido compounds [Se(AuPPh3)3]OTf
and [Se(Au2dppf){Se(Au2dppf)}2](OTf)2 (13), in which
Th eor etica l Ca lcu la tion s
The compounds described in the Experimental Section
present short Au(I)-Au(I) contacts. To understand their
origin, we have carried out ab initio calculations at MP2
and HF levels with quasi-relativistic pseudopotentials
(Table 4). The computational details are described in
the molecular orbital calculations section. The total
energies and optimized geometries are shown in Tables
5 and 6, respectively. Geometries of the [Se(AuPH3)n](n-2)+
(n ) 2-6) models are illustrated in Figure 5. The results
indicate that the various structural parameters change
substantially when they are compared at the HF and
MP2 levels; the gold-gold distances are significantly
shortened and Au-Se-Au angles narrowed for the
latter method and are comparable in magnitude with
known experimental values (see Table 6). This gives
some impression of the contribution of electron correla-
tion to the intramolecular contacts for these clusters.
Since it has been suggested that aurophilic attraction
is primarily a correlation effect, the HF calculations
provide a convenient way of turning off the attraction.6

Polynumclear Gold Complexes
Organometallics, Vol. 19, No. 24, 2000 4991
Ta ble 4. Ba sis Sets a n d P seu d op oten tia ls (P P s)
Used in th e P r esen t Wor k
atom PP
basis
remarks
H -
(4s1p)/[2s1p]
Rp ) 0.8021
Rd ) 0.34
Rd ) 0.25
Rf ) 0.20
P Bergner 5-VE
Se Bergner 6-VE
Au Andrae 19-VE
(4s4p1d)/[2s2p1d]
(4s4p1d)/[2s2p1d]
(8s6p5d1f)/[6s5p3d1f]
Ta ble 5. Tota l En er gies of th e System s (in a u )
system
Hartree-Fock
MP2
Se(AuPH3)2+C2v
Se(AuPH3)3 C3v
Se(AuPH3)42+ Td
-295.346744
-438.300352
-581.133855
-581.133855
-581.104087
-723.117459
-723.111230
-867.632215
-296.185816
-439.484695
-582.618990
-582.621580
-582.642334
-724.916119
-724.913063
-868.524890
2+
Se(AuPH3)4 C3v
2+
Se(AuPH3)4
Se(AuPH3)5
C
C
4v
3+
3v
Se(AuPH3)53+ Cs
4+
Se(AuPH3)6
D
3d
The relevant experimental structural data are in-
cluded in Table 6. We have taken the average crystal-
lographic angles and distances for these systems.9-11 It
should be mentioned that at the experimental level the
-PPh3 groups are replaced by -PH3 and the ferrocene
bridged ligands by -H.
F igu r e 4. Structure of the cation of complex 14 in the
crystal.
Se(Au P H3)2. This system has also been studied
theoretically by Pyykko¨ et al.,13 who found a gold-gold
contact when their MP2 calculations included electronic
correlation and quasi-relativistic effects. The difference
between the previous calculations and ours (Table 6) is
that we have fully optimized the structure. However,
both calculations generate an Au-Se-Au angle and
Au-Au distance very similar at the MP2 level, 75.4°,
74.4°, 292 pm, and 295 pm, respectively. At the HF level
the angles and distances are appreciably larger.
Se(Au P H3)3+ . In this cluster too, when the electronic
correlation at the MP2 level is taken into account, the
values of the distances and angles approach the experi-
mental data. However, the calculated Au- -Au distances
and Au-Se-Au angles are smaller than experimental
values, because the MP2 level exaggerates the auro-
philic attractions, as for the previous neutral A-frame.12
On the other hand, the wider bond angles in the
experimental system may be caused by steric effects
between bulky phenyl groups. There exist for some
similar systems (e.g., X(AuPH3)3+ , X ) O, S),8d calcula-
tions based on density functional theory (DFT) with
electronic correlation and relativistic effects. However,
they are not as consistent with the experimental data,
being instead comparable to Hartree-Fock calculations.
This is because the density functional calculations are
unable to describe van der Waals interactions among
metallic centers; such interactions are not related in any
simple way to the electron density.3
F igu r e 5. Geometries of the [Se(AuPH3)n](n-2)+ (n ) 2-6)
models.
There exists theoretical evidence indicating that the
MP2 method may exaggerate the aurophilic attraction.2
A more exact method would be CCSD (coupled cluster
single and double), but it is very expensive for a study
of this magnitude of the systems in which we are
interested.
For the systems with n ) 4 and 5, various geometrical
arrangements of the (AuPH3) fragments are possible
around the central Se atom. It is a well-known principle
in coordination chemistry that systems of the type MLn
prefer a tetrahedral geometry for n ) 4, whereas
trigonal bipyramids are typical for n ) 5. Systems
analogous to the ours, such as [X(AuL)n]m+ (X ) C, N,
O, S, P; n ) 3-6), have previously been studied by
Pyykko¨ et al.,12 establishing that when attractive forces
are present between the -AuPH3 fragments, deviations
from the ideal structure must be expected. Thus, the
tetrahedral symmetry was lowered to C4v, resulting in
four Au-Au distances shorter than in the tetrahedral
case. In the pentacoordinated case, a C3v arrangement
of the ligands is preferred over the Cs structure.
Se(Au P H3)42+. C4v, C3v, and Td relative MP2 energies
are 0, +54, +61 kJ /mol. This order rationalizes for the
first time the observed C4v structure,9 and shows that
the geometry adopted by the cluster is directly related
to the van der Waals interactions established between
the gold centers (MP2) and the number of these, with a
maximum for the structure with C4v symmetry. If we
compare the relative energies at the HF level, we find
the opposite trend (see Table 5).
Se(Au P H3)53+. For this type of cluster a trigonal
bipyramidal C3v (distorted D3h) symmetry is preferred
(12) Pyykko¨, P.; Tamm, T. Organometallics 1998, 17, 4842.
(13) Li, J .; Pyykko¨, P. Chem. Phys. Lett. 1992, 197, 586.

4992 Organometallics, Vol. 19, No. 24, 2000
Canales et al.
Ta ble 6. Ma in Geom etr ic P a r a m eter s of th e Selen iu m Cen ter ed System s (in p m a n d d egr ees).
Se-Au
bond
Au- -Au
distance
Au-P
bond
Au-Se-Au
system
method
angle
[Se(AuPPh3)2]
[Se(AuPH3)2] C2v
expt11
HF
MP2
expt10
HF
MP2
expt9
HF
MP2
HF
239.4
245.7
241.9
244.7
250.4
246.5
249.5
254.7
243.8
254.6
247.6
258.9
247.6
261a
305.1
363.7
292.4
301.8
385.0
297.5
292.5
415.8
398.1
415.7a
350a-429b
333.8
288.8
451a
225.4
241.5
231.5
227.4
241.5
232.5
227
79.1
95.49
74.36
80.1
100.5
74.2
[Se(AuPPh3)3]PF6
[Se(AuPH3)3]+ C3v
[Se(AuPPh3)4](CF3SO3)2
71.52
109.5
109.5
109.42a
90.4a-125b
80.24
71.36
120d
[Se(AuPH3)4]2+ Td
241.3
232.3
241.3
234.9
241.9
234.9
240a
2+
[Se(AuPH3)4]
[Se(AuPH3)4]
[Se(AuPH3)5]
C
C
C
3v
4v
3v
MP2
HF
MP2
HF
2+
3+
267c
373b
240c
90b
MP2
HF
246a
423d
230a
120d
248c
348b
232c
90b
[Se(AuPH3)5]3+ Cs
262c
453f
240a
116.4f
82.3b
118.3d
80.2b
90.0
266e
363e
240c
MP2
243c
433d
230a
247e
352b
232c
4+
[Se(AuPH3)6]
D
HF
268
354
240
3d
MP2
248
302
232
90.0
a
b
d
Equatorial. Axial to equatorial. c Axial. Equatorial to equatorial. e Nonaxial. f Axial to nonaxial.
Ma ter ia ls. The starting materials [Se(AuPPh3)2],11 [Au2Cl2-
over the Cs (C4v for the Se-Au core) by 8 kJ /mol (at the
MP2 level). For this system it has not been possible to
obtain the X-ray structure, but the geometry is analo-
gous with that for the carbon derivatives.12 Our calcula-
tions serve as a prediction of the geometry of the
selenium compounds.
Se(Au P H3)64+. The structural framework is calcu-
lated to display octahedral geometry (distorted D3h
symmetry). Again it has not been possible to determine
its X-ray structure. Due to the effect of the electronic
correlation at the MP2 level, the Au-Se and Au-Au
distances are shorter than at the HF level (see Table
6).
(µ-dppf)],14 and [Au(tht)2]ClO4 were prepared by published
15
procedures. [Au(OTf)(PPh3)] was prepared from [AuCl(PPh3)]16
by reaction with AgOTf in dichloromethane; [Au2(OTf)2(µ-
dppf)] from [Au2Cl2(µ-dppf)] and 2 equiv of AgOTf in dichlo-
romethane.
Sa fety Note. Caution! Perchlorate salts of metal complexes
with organic ligands are potentially explosive. Only small
amounts of material should be prepared, and these should be
handled with great caution.
Syn th esis of [Se(Au P P h 3)n ](OTf)n -2 (n ) 3 (2), 4 (3), 5
(4), 6 (5)). To a solution of [Se(AuPPh3)2] (0.100 g, 0.1 mmol)
in 20 mL of dichloromethane was added the corresponding
amount of [Au(OTf)PPh3] (0.061 g, 0.1 mmol, 2; 0.122 g, 0.2
mmol, 3; 0.183, 0.3 mmol, 4; 0.244 g, 0.4 mmol, 5) and the
mixture stirred for 15 min. Concentration of the solution to
ca. 5 mL and addition of hexane (10 mL) gave complex 2, 3, 4,
or 5 as a white solid. Complex 2: yield 85%; ΛM 102 Ω-1 cm2
Con clu sion s
mol-1. Anal. Found: C, 41.23; H, 2.76; S, 2.14. Calcd for C55H45
Au3F3O3P3SSe: C, 41.12; H, 2.80; S, 1.99. NMR data. 31P(1H),
δ: 34.7 (s) ppm. Complex 3: yield 81%; ΛM 165 Ω-1 cm2 mol-1
Anal. Found: C, 39.65; H, 2.71; S, 2.42. Calcd for C74H60
-
Several gold compounds with a central selenium atom
have been prepared and characterized. They show
unusual geometries with short gold-gold interactions.
Theoretical calculations at the MP2 level with quasi-
relativistic pseudopotentials are in agreement with
experimental geometries. We attribute the change in the
ideal geometry of the compounds to the presence of
strong Au- - -Au interactions that contribute to the
stabilization of the different compounds.
.
-
Au4F6O6P4S2Se: C, 40.12; H, 2.71; S, 2.84. 31P(1H), δ: 33.5 (s)
ppm. Complex 4: yield 76%; ΛM 293 Ω-1 cm2 mol-1. Anal.
Found: C, 39.02; H, 2.55; S, 3.91. Calcd for C93H75Au5F9O9P5S3-
Se: C, 39.56; H, 2.66; S, 3.40. 31P(1H), δ: 32.6 (s) ppm. Complex
5: yield 71%; ΛM 350 Ω-1 cm2 mol-1. Anal. Found: C, 39.55;
H, 2.38; S, 2.57. Calcd for C112H90Au6F12O12P6S4Se: C, 39.19;
H, 2.62; S, 3.23. 31P(1H), δ: 31.7 (s) ppm.
Syn th esis of [Au {Se(Au P P h 3)2}2]ClO4, 6. To a solution
of [Se(AuPPh3)2] (0.200 g, 0.2 mmol) in 20 mL of dichlo-
romethane was added [Au(tht)2]ClO4 (0.047 g, 0.1 mmol), and
the mixture was stirred for 30 min. Evaporation of the solvent
to ca. 5 mL and addition of diethyl ether (10 mL) gave a white
solid of complex 6. Yield: 57%; ΛM 95 Ω-1 cm2 mol-1. Anal.
Found: C, 38.11; H, 2.61. Calcd for C72H60Au5 ClO4P4Se2: C,
37.72; H, 2.62. 31P(1H), δ: 35.3 (s) ppm.
Exp er im en ta l Section
In str u m en ta tion . Infrared spectra were recorded in the
range 4000-200 cm-1 on a Perkin-Elmer 883 spectrophotom-
eter using Nujol mulls between polyethylene sheets. Conduc-
tivities were measured in ca. 5 × 10-4 mol dm-3 solutions with
a Philips 9509 conductimeter. C, H, and S analyses were
carried out with a Perkin-Elmer 2400 microanalyzer. Mass
spectra were recorded on a VG Autospec, with the liquid
secondary-ion mass spectra (LSIMS) technique, using ni-
trobenzyl alcohol as matrix. NMR spectra were recorded on a
Varian Unity 300 spectrometer and a Bruker ARX 300
spectrometer in CDCl3 (otherwise stated). Chemical shifts are
cited relative to SiMe4 (1H, external) and 85% H3PO4 (31P,
external).
Syn th esis of [Se(Au 2d p p f)], 7. To a solution of [Au2Cl2-
(µ-dppf)] (1.019 g, 1 mmol) in 20 mL of acetone was added a
(14) Gimeno, M. C.; Laguna, A.; Sarroca, C.; J ones, P. G. Inorg.
Chem. 1993, 32, 5926.
(15) Uso´n, R.; Laguna, A.; Laguna, M.; J ime´nez, J .; Go´mez M. P.;
Sainz, A.; J ones, P. G. J . Chem. Soc., Dalton Trans. 1990, 3457.
(16) Uso´n, R.; Laguna, A. Inorg. Synth. 1982, 21, 71.

Polynumclear Gold Complexes
Organometallics, Vol. 19, No. 24, 2000 4993
Ta ble 7. Deta ils of Da ta Collection a n d Str u ctu r e Refin em en t for Com p lexes 7, 12b, a n d 13
7
12b‚8CH2Cl2
13‚3.5CH2Cl2
chemical formula
cryst habit
cryst size/mm
cryst syst
space group
a/Å
C34H28Au2FeP2Se
yellow prism
0.32 × 0.18 × 0.15
monoclinic
P21/c
10.942(2)
15.816(2)
18.066(3)
104.030(14)
3033.4(8)
4
C144H128Au10Cl16Fe4N2O6P8Se4
C107.5H91Au6Cl7Fe6O6S2Se2
yellow plate
0.25 × 0.16 × 0.08
monoclinic
P21/n
20.5931(18)
19.5047/16)
28.984(2)
102.814(3)
11351.7(17)
4
yellow tablet
0.28 × 0.17 × 0.10
monoclinic
C2/c
34.142(7)
22.064(5)
24.354(5)
118.695(9)
16093(6)
4
b/Å
c/Å
â/deg
U/Å3
Z
Dc/g cm-3
M
2.249
1027.25
1920
2.190
5306.35
9904
2.105
3598.16
6796
F(000)
T/°C
-100
-130
-130
2θmax/deg
µ(Μo KR)/cm-1
transmission
no. of reflns measd
no. of unique reflns
Rint
50
50
52
115
0.890-0.775
5385
5319
0.078
107
91
0.801-0.350
46 530
14 132
0.212
0.802-0.389
115 873
23 217
0.108
Ra (F>4σ(F))
wRb (F2, all reflns)
no. of reflns used
no. of params
no. of restraints
Sc
0.0382
0.0793
5319
362
316
0.865
1.877
0.1115
0.3442
14 132
306
13
0.975
0.0438
0.1049
23 217
1224
263
0.949
2.224
max. ∆F/e Å-3
4.877
R(F)) ∑||Fo| - |Fc ||/∑|Fo |. wR(F2) ) [∑{w(Fo - Fc2)2}/∑{w(Fo2)2}]0.5; w-1 ) σ2(Fo2) + (aP)2 + bP, where P ) [Fo + 2Fc2]/3 and a
a
b
2
2
and b are constants adjusted by the program. S ) [∑{w(Fo - Fc2)2}/(n - p)]0.5, where n is the number of data and p the number of
parameters.
c
2
solution of SeC(NH2)2 (0.123 g, 1 mmol) in acetone (20 mL)
and the mixture stirred for 1 h. The white precipitate formed
was filtered off, then dissolved in 10 mL of methanol, and a
solution of Na2CO3 in water was added. The mixture was
stirred for 4 h and then washed with CH2Cl2. The organic
phase was separated and dried with MgSO4. Evaporation of
the solvent and addition of diethyl ether gave an orange solid
of complex 7. Yield: 45%; ΛM 0.6 Ω-1 cm2 mol-1. Anal. Found:
C, 39.5; H, 2.5. Calcd for C34H28Au2FeP2Se: C, 39.7; H, 2.7.
S, 2.93. Calcd for C62H54Au4F6FeO6P4S2Se: C, 35.11; H, 2.54;
S, 3.02. 31P(1H), δ: 28.5 (s, 2P, dppf), 20.7 (s, 2P, PPh2Me) ppm.
1H, δ: 4.11 (m, 4H, C5H4), 4.57 (m, 4H, C5H4), 7.2-7.8 (m, 40H,
Ph).
Syn th esis of [Au {Se(Au 2d p p f)}2]ClO4, 12a . To a solution
of complex 7 (0.204 g, 0.2 mmol) in 20 mL of dichloromethane
was added [Au(tht)2]ClO4 (0.047 g, 0.1 mmol) and the mixture
stirred for 30 min. Evaporation of the solvent to ca. 5 mL and
addition of ca. 10 mL of diethyl ether gave a yellow solid of
complex 12a : Yield 93%; ΛM 110 Ω-1 cm2 mol-1. Anal. Found:
C, 34.27; H, 2.32. Calcd for C68H56Au5ClFe2O4P4Se2: C, 34.72;
1
31P(1H), δ: 31.4 (s) ppm; H, δ: 3.89 (m, 4H, C5H4), 5.27 (m,
4H, C5H4), 7.4-7.6 (m, 20H, Ph).
1
Syn th esis of [Se(Au 2d p p f)(Au P R 3)]OTf (P R3 ) P P h 3 8,
P P h 2Me 9). To a solution of complex 7 (0.102 g, 0.1 mmol) in
20 mL of dichloromethane was added [Au(OTf)PPh3] (0.061 g,
0.1 mmol) or [Au(OTf)PPh2Me] (0.054 g, 0.1 mmol) and the
mixture stirred for 1 h. Evaporation of the solvent to ca. 5 mL
and addition of hexane afforded orange solids of complexes 8
and 9, respectively. Complex 8: Yield 80%; ΛM 98 Ω-1 cm2
H, 2.38. 31P(1H), rt, δ: 28.2 (br, 4P, dppf) ppm. H, rt, δ: 4.4
(m, br, 8H, C5H4), 4.7 (m, br, 8H, C5H4), 7.2-7.6 (m, 40H, Ph).
31P(1H), -55 °C, δ: 21.5 (s, 1P, dppf), 23.5 (s, 1P, dppf), 26.0
1
(s, 1P, dppf), 27.3 (s, 1P, dppf) ppm. H, -55 °C., δ: 3.34 (m,
1H, C5H4), 3.43 (m, 1H, C5H4), 3.58 (m, 1H, C5H4), 3.69 (m, H,
C5H4), 3.79 (m, 1H, C5H4), 4.05 (m, 1H, C5H4), 4.20 (m, 1H,
C5H4), 4.26 (m, 1H, C5H4), 4.33 (m, 1H, C5H4), 4.40 (m, 1H,
C5H4), 4.71 (m, 1H, C5H4), 4.80 (m, 1H, C5H4), 5.01 (m, 1H,
C5H4), 5.48 (m, 1H, C5H4), 5.58 (m, 1H, C5H4), 6.8-8.2 (m, 60H,
Ph).
mol-1. Anal. Found: C, 38.89; H, 2.55; S, 2.10. Calcd for C53H43
-
Au3 F3FeO3P3SSe: C, 38.82; H, 2.80; S, 1.95. 31P(1H), δ: 29.3
(s, 2P, dppf), 35.9 (s, 1P, PPh3) ppm; 1H, δ: 4.08 (m, 4H, C5H4),
4.45 (m, 4H, C5H4), 7.3-7.5 (m, 35H, Ph). Complex 9: Yield
83%; ΛM 102 Ω-1 cm2 mol-1. Anal. Found: C, 36.16; H, 2.62;
S, 1.96. Calcd for C48H41Au3F3FeO3P3SSe: C, 36.62; H, 2.61;
S, 2.03. 31P(1H), δ: 28.7 (s, 2P, dppf), 21.2 (s, 1P, PPh2Me) ppm.
1H, δ: 4.06 (m, 4H, C5H4), 4.50 (m, 4H, C5H4), 7.2-7.6 (m, 30H,
Ph).
Syn th esis of [(Au 2d p p f){Se(Au 2d p p f)}2](OTf)2, 13. To a
solution of complex 7 (0.204 g, 0.2 mmol) in 20 mL of
dichloromethane was added [Au2(OTf)2(µ-dppf)] (0.124 g, 0.1
mmol) and the mixture stirred for 30 min. Evaporation of the
solvent to ca. 5 mL and addition of ca. 10 mL of hexane gave
an orange solid of complex 13: Yield 66%; ΛM 170 Ω-1 cm2
mol-1. Anal. Found: C, 37.46; H, 2.38; S, 1.71. Calcd for
Syn th esis of [Se(Au 2d p p f)(Au P R3)2](OTf)2 (P R3 ) P P h 3
10, P P h 2Me 11). To a solution of complex 7 (0.102 g, 0.1 mmol)
in 20 mL of dichloromethane was added [Au(OTf)PPh3] (0.122
g, 0.2 mmol) or [Au(OTf)PPh2Me] (0.108 g, 0.2 mmol) and the
mixture stirred for 1 h. Evaporation of the solvent to ca. 5 mL
and addition of hexane afforded orange solids of complexes 10
and 11, respectively. Complex 10: Yield 55%; ΛM 150 Ω-1 cm2
C
104H84Au6F6Fe3O6P6S2Se2: C, 37.09; H, 2.54; S, 1.93. 31P(1H),
rt, δ: 28 (br, 6P, dppf) ppm. 1H, rt, δ: 4.43 (m, br, 16H, C5H4),
7.2-7.8 (m, 60H, Ph). 31P(1H), -55 °C, δ: 22.8 (s, 2P, dppf),
26.9 (s, 2P, dppf), 29.0 (s, 2P, dppf) ppm. H, -55 °C, δ: 4.06
1
(m, 2H, C5H4), 4.15 (m, 2H, C5H4), 4.40 (m, 4H, C5H4), 4.51
(m, 8H, C5H4), 4.65 (m, 4H, C5H4), 6.8-8.2 (m, 60H, Ph).
Syn th esis of [Se(Au 2d p p f)2](OTf)2, 14. To a solution of
complex 7 (0.102 g, 0.1 mmol) in 20 mL of dichloromethane
was added [Au2(OTf)2(µ-dppf)] (0.124 g, 0.1 mmol) and the
mixture stirred for 30 min. Then the solvent was evaporated
to ca. 5 mL, and addition of hexane gave an orange solid of
mol-1. Anal. Found: C, 38.89; H, 2.40; S, 3.07. Calcd for C72H58
-
Au4 F6FeO6P4S2Se: C, 38.58; H, 2.58; S, 2.85. 31P(1H), δ: 28.0
(s, 2P, dppf), 31.1 (s, 2P, PPh3) ppm. 1H, δ: 4.18 (m, 4H, C5H4),
4.60 (m, 4H, C5H4), 7.3-7.6 (m, 50H, Ph). Complex 11: Yield
76%; ΛM 156 Ω-1 cm2 mol-1. Anal. Found: C, 34.76; H, 2.64;

4994 Organometallics, Vol. 19, No. 24, 2000
Canales et al.
complex 14: Yield 75%; ΛM 160 Ω-1 cm2 mol-1. Anal. Found:
C, 36.66; H, 2.40; S, 2.32. Calcd for C70H56Au4F6Fe2O6P4S2Se:
C, 36.96; H, 2.46; S, 2.80. 31P(1H), rt, δ: 28.3 (br, 4P, dppf)
ppm. 1H, rt, δ: 4.3 (m, br, 8H, C5H4), 4.5 (m, br, 8H, C5H4),
7.4-7.8 (m, 40H, Ph). 31P(1H), -55 °C, δ: 22.8 (s, 2P, dppf,
13), 26.9 (s, 2P, dppf, 13), 29.0 (s, 2P, dppf, 13), 27.8 (s, 4P,
regions of severely disordered dichloromethane which were
arbitrarily refined as chlorine sites of varying occupation. The
high residuals, associated with this problem, indicate that the
structure should be interpreted with caution, but we are
confident that its main features are correct. The agreement
factors are necessarily poor. The largest residual electron
density in this complex is near the gold atoms. Complex 13
crystallizes with four molecules of dichloromethane, of which
one was arbitrarily assigned occupancy of 0.5 to obtain
reasonable U values.
Molecu la r Or bita l Ca lcu la tion s. The Gaussian 98 pack-
age18 was used. The basis sets and pseudopotentials (PP) used
in the production runs are given in Table 4. The 19-valence
electron (VE) quasi-relativistic (QR) pseudopotential of An-
drae19 was employed for gold. We have employed one f-type
polarization function for Au. The f orbitals are necessary for
the weak intermolecular interactions, as was demonstrated
previously for different metals.5 The atoms C, O, P, and Se
were also treated by Stuttgart pseudopotentials,20 including
only the valence electron for each atom. For these atoms,
double-ú basis sets were used, augmented by d-type polariza-
tion functions. For the H atom, a double-ú plus one p-type
polarization function was used (see Table 1).21
1
dppf, 14), 30.0 (s, 2P, dppf) ppm. H, -55 °C, δ: 3.6-5.0 (m,
br, C5H4), 6.6-8.2 (m, Ph).
Syn th esis of [Se(Au 2d p p f)3](OTf)4, 15. To a solution of
complex 7 (0.102 g, 0.1 mmol) in 20 mL of dichloromethane
was added [Au2(OTf)2(µ-dppf)] (0.249 g, 0.2 mmol) and the
mixture stirred for 30 min. Then the solvent was evaporated
to ca. 5 mL, and addition of hexane gave an orange solid of
complex 15: Yield 73%; ΛM 340 Ω-1 cm2 mol-1. Anal. Found:
C, 36.03; H, 2.46; S, 3.07. Calcd for C106H84Au6F12Fe3O12P6S4-
Se: C, 36.15; H, 2.38; S, 3.63. 31P(1H), rt, δ: 27.6 (s, 6P, dppf)
ppm. 1H, rt, δ: 4.47 (m, br, 24H, C5H4), 7.3-7.6 (m, 60H, Ph).
31P(1H), -55 °C, δ: 26.3 (s, dppf), 27.2 (s, dppf), 27.8 (s, dppf),
31.3 (s, dppf) ppm. 1H, -55 °C, δ: 4.16 (m, 8H, C5H4), 4.56
(m, 8H, C5H4), 4.67 (m, 4H, C5H4), 5.13 (m, 4H, C5H4), 6.6-
8.0 (m, 60H, Ph).
Cr ysta l Str u ctu r e Deter m in a tion s. The crystals were
mounted in inert oil on glass fibers and transferred to the cold
gas stream of a Siemens P4 (7) or Bruker SMART diffracto-
meter. Data were collected using monochromated Mo KR
radiation (λ ) 0.71073 Å) in ω-scan mode. An absorption
correction was applied on the basis of ψ-scans for 7, whereas
for 12b-14 the program SADABS, based on multiple scans,
was used. The structures were solved by direct methods and
refined on F2 using the program SHELXL-97.17 All non-
hydrogen atoms were refined anisotropically except for solvent.
Hydrogen atoms were included using a riding model. Further
crystal data are given in Table 7. Special details of refine-
ment: Compound 12b : The structure contains extensive
We have fully optimized the structures for the models Se-
(AuPH3)2 (C2v), Se(AuPH3)3+ (C3v), Se(AuPH3)42+ (Td, C3v, C4v),
3+
4+
Se(AuPH3)5 (Cs), and Se(AuPH3)6 (D3d), at Hartree-Fock
(HF) and second-order Møller-Plesset perturbation theory
(MP2) levels (see Tables 5 and Table 6). In the experimental
structures, the ligand on the gold atoms is triphenylphosphine,
-P(C6H5)3. This ligand implies a large computational effort;
this was replaced by the phosphine group -PH3. If the rotation
of the terminal -PH3 ligands would have broken the sym-
metry, their dihedral angle was fixed.
Ack n ow led gm en t. We thank the Direccio´n General
de Investigacio´n Cient´ıfica y Te´cnica (No. PB97-1010-
C02-01), the Caja de Ahorros de la Inmaculada (CB16/
99), and the Fonds der Chemischen Industrie for
financial support. This work was partially supported by
Project CSIC-Universidad de Chile and Fondecyt No.
1990038. The authors thank Professor Bruce Cassels
for the access to the Silicon Graphics Station Octane
and the Gaussian 98 package.
(17) Sheldrick, G. M. SHELXL-97, a program for crystal structure
refinement; University of Go¨ttingen, Germany, 1997.
(18) Frisch, M. J .; Trucks, G. W.; Schlegel, H. B.; Gill, P. M. W.;
J ohnson, B. G.; Robb, M. A.; Cheeseman, J . R.; Keith, K. T.; Petersson,
G. A.; Montgomery, J . A.; Raghavachari, K.; Al-Laham, M. A.;
Zakrzewski, V. G.; Ortiz, J . V.; Foresman, J . B.; Cioslowski, J .;
Stefanov, B. B.; Nanayakkara, A.; Challacombe, M.; Peng, C. Y.; Ayala,
P. Y.; Chen, W.; Wong, M. W.; Andres, J . L.; Replogle, E. S.; Gomperts,
R.; Martin, R. L.; Fox, D. J .; Binkley, J . S.; Defrees, D. J .; Baker, J .;.
Stewart, J . P.; Head-Gordon, M.; Gonzalez, C.; Pople, J . A. Gaussian
98; Gaussian, Inc.: Pittsburgh, PA, 1998.
(19) Andrae, D.; Ha¨usserman, M.; Dolg, H.; Stoll, H.; Preuss, H.
Theor. Chim. Acta 1990, 77, 123.
(20) Bergner, A.; Dolg, M.; Ku¨chle, W.; Stoll, H.; Preuss, H. Mol.
Phys. 1993, 80, 1431.
(21) Dunning, T. H.; Hay, P. J . In Modern Theoretical Chemistry,
Vol. 3 Schaefer, H. F., III, Ed.; Plenum Press: New York, 1997; pp
1-28.
Su p p or tin g In for m a tion Ava ila ble: Three X-ray crys-
tallographic files, in CIF format, for complexes 7, 12b, and
13. This material is available free of charge via the Internet
at http://pubs.acs.org.
OM000119G