Synthesis and X-ray structure study on new Au(I) polymer architectures based on multi-sulfur tentacles

Article abstract of DOI:10.1016/j.jorganchem.2004.10.061

Multi-thiolate ligands are used as a scaffold to construct a series of supramolecules, which cover the following entries; [(1,3-S₂-C ₆H₄){AuP(C₆H₄-3-CF₃) ₃}₂]_n (1), [(1,4-S₂-C ₆H₄){AuP(C₆H₄-3-CF₃) ₃}₂]_n (2), (1,4-S₂-C ₆H₄){AuP(C₆H₅)₂(2- pyridine)}₂ (3), [(1,3,5-S₃-C₆H ₃){AuP(C₆H₅)₂(2-pyridine)} ₃]_n (4), and [(1,3,5-S₃-C₆H ₃){AuP(C₆H₄-3-CF₃)₃} ₃]_n (5). The molecular and crystal structures of these new derivatives have been elucidated by single crystal X-ray diffraction. Aurophilic interactions have been demonstrated for 1, 2, 4, and 5 to produce new supramolecular architectures. Nano-channels are formed by aurophilic and π-π interactions for 1, in which benzene molecules are trapped. An 8 (eight)-shaped loop is formed in solid state for 2. Infinite zigzag chains are constructed for 4 and 5.

Full text of DOI:10.1016/j.jorganchem.2004.10.061

Journal of Organometallic Chemistry 690 (2005) 1332–1339
www.elsevier.com/locate/jorganchem
Synthesis and X-ray structure study on new Au(I)
polymer architectures based on multi-sulfur tentacles
a
a
a,
a
Keiko Nunokawa , Kazuya Okazaki , Satoru Onaka *, Mitsuhiro Ito ,
a
Tetsuya Sunahara , Tomoji Ozeki , Hiroyuki Imai , Katsuya Inoue
b
c
c
a
Department of Environmental Technology, Graduate School of Engineering, Nagoya Institute of Technology, Gokiso-cho, Showa-ku,
Nagoya 466-8555, Japan
b
Department of Chemistry and Materials Science, Tokyo Institute of Technology, O-okayama, Meguro-ku, Tokyo 152-8551, Japan
c
Institute for Molecular Science, Myodaiji, Okazaki 444-8585, Japan
Received 13 August 2004; accepted 4 October 2004
Available online 7 February 2005
Abstract
Multi-thiolate ligands are used as a scaffold to construct a series of supramolecules, which cover the following entries; [(1,3-S2–
C6H4){AuP(C6H4–3-CF3)3}2]n (1), [(1,4-S2–C6H4){AuP(C6H4–3-CF3)3}2]n (2), (1,4-S2–C6H4){AuP(C6H5)2(2-pyridine)}2 (3), [(1,3,5-
S3–C6H3){AuP(C6H5)2(2-pyridine)}3]n (4), and [(1,3,5-S3–C6H3){AuP(C6H4–3-CF3)3}3]n (5). The molecular and crystal structures of
these new derivatives have been elucidated by single crystal X-ray diffraction. Aurophilic interactions have been demonstrated for 1,
2, 4, and 5 to produce new supramolecular architectures. Nano-channels are formed by aurophilic and p–p interactions for 1, in
which benzene molecules are trapped. An 8 (eight)-shaped loop is formed in solid state for 2. Infinite zigzag chains are constructed
for 4 and 5.
ꢀ 2004 Elsevier B.V. All rights reserved.
Keywords: Aurophilicity; Multi-S ligands; Supramolecules; Dendrimers; Helical structures
1. Introduction
interesting supramolecules by the use of 1,3,4-thiadiaz-
ole-2,5-dithiolate, which is similar to the above ligands.
We have also been interested in using thiolate ligands to
control aurophilicity [4] and led to explore a pre-dendri-
mer chemistry based on multi-sulfur tentacles as a
logical extension of our research programs. 1,3- and 1,4-
Dimercaptobenzene and 1,3,5-trimercaptobenzene are
our choice in addition to the above trithiocyanurates,
on which a unique Au12 cluster has been constructed re-
cently [4d]. In the present study, these multi-thiolate li-
gands were reacted with two kinds of gold phosphine
chlorides where the phosphine represents P(C6H4–3-
CF3)3 and P(C6H5)2(2-pyridine) other than PPh3. These
ligands were chosen because of the following two rea-
sons: the selection of P(C6H4–3-CF3)3 was rationalized
in the previous paper in order to reinforce the aurophi-
licity [4b]. P(C6H5)2(2-pyridine) was expected to exhibit
The design and the synthesis of supramolecular struc-
tures in the solid state have attracted keen interest not
only because of topological beauty [1] but also of recent
applications to nano-technologies [2]. The use of auro-
philic interaction has led to many fruitful outcomes
for constructing gold-based extended structures, which
should serve as such prospective nano-materials [3]. Thi-
olate ligands are especially attractive scaffold to build-up
such nano-molecules [4–7]. In this vein, trithiocyanuric
acid, its salts, and its derivatives are intriguing and many
fantastic supramolecules have been reported recently [8–
11]. Schmidbaur and co-workers [12] have synthesized
*
Corresponding author. Tel.: +81527355160; fax: +81527355160.
E-mail address: onaka.satoru@nitech.ac.jp (S. Onaka).
0022-328X/$ - see front matter ꢀ 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2004.10.061

K. Nunokawa et al. / Journal of Organometallic Chemistry 690 (2005) 1332–1339
1333
coordinating ability of resulting Au–S derivatives to
other transition metal(s) through a pyridine N atom.
This paper describes the synthesis and X-ray structure
analysis of new Au(I) supramolecules, [(1,3-S2–C6H4)-
{AuP(C6H4–3-CF3)3}2]n (1), [(1,4-S2–C6H4){AuP(C6H4–
3-CF3)3}2]n (2), (1,4-S2–C6H4){AuP(C6H5)2(2-pyridine)}2
(3), [(1,3,5-S3–C6H3){AuP(C6H5)2(2-pyridine)}3]n (4),
and [(1,3,5-S3–C6H3){AuP(C6H4–3-CF3)3}3]n (5). Infi-
nite polymers have been obtained for 1, 2, 4, and 5.
Fig. 2. An ORTEP drawing of (1,4-S2–C6H4){AuP(C6H4–3-CF3)3}2
(2). F atoms are omitted for clarity.
2. Results and discussion
good yield; two P–Au–S moieties are bent to the same
rotational direction as in 2 (Fig. 3). The 1,3,5-S3C6H3
scaffold affords 4 and 5. The molecular structure of 4
is shown in Fig. 4. Three P–Au–S groups are bent to
the same rotational direction as in 2 and 3, thus forming
a propeller-like structure. Fig. 5 shows the molecular
structure of 5, which is similar to that of 4 and a known
trimercapto-triazine–Au–PPh3 analogue [10]. No intra-
molecular Au–Au interaction is observed for all of these
derivatives when two independently refined moieties are
treated as a molecule for 2. However, intermolecular
2.1. X-ray molecular structural analysis
The pyridyl N atoms for 3 and 4 have been deter-
mined mainly by respective Fourier peak heights. An
additional reasoning for the determination of N atom
is based on the absence of an aromatic hydrogen peak
in the Fourier map around the assumed atom.
Fig. 1 shows an ORTEP drawing of the molecular
structure of 1. An interesting structural feature of 1 is
that the two P–Au–S moieties are bent to come close
intramolecularly in spite of the sterical crowding due
to bulky PR03 groups. A plausible explanation is that
two phenyl groups (C21–C26 and C51–C56), each of
which comes from different PR03 ligands, are bent to
form a preferable intramolecular p–p interaction; the de-
tail of this stack is discussed below. When 1,4-dithiol
was used as a scaffold, 2 was obtained in a good yield
and two P–Au–S moieties are bent to the same rota-
tional direction in contrast to 1 (Fig. 2). When this dith-
iol and PPh2(2-pyridine) were used, 3 was obtained in a
Fig. 3. The molecular structure of (1,4-S2–C6H4){AuP(C6H5)2(2-
pyridine)}2 (3).
Fig. 1. An ORTEP drawing of (1,3-S2–C6H4){AuP(C6H4–3-CF3)3}2
(1). F atoms are omitted for clarity.
Fig. 4. The molecular structure of (1,3,5-S3–C6H3){AuP(C6H5)2(2-
pyridine)}3 (4).

1334
K. Nunokawa et al. / Journal of Organometallic Chemistry 690 (2005) 1332–1339
Fig. 5. An ORTEP drawing of (1,3,5-S3–C6H3){AuP(C6H4–3-CF3)3}3 (5). F atoms are omitted for clarity.
aurophilic interaction has been induced for 1, 2, 4, and
˚
5. The Au–Au distance is 3.001 (1) A for 1 and
gand is used, only S3C6 group of the Au3P3S3C6 core
forms a plane and the deviations from the best plane
are especially large for P atoms. Otherwise, the Au–S
and Au–P lengths for 5 are quite close to those of 4.
It is interesting to compare bonding parameters of
1–5 with those of similar Au(I)-PPh3 derivatives con-
structed on 1,3-S2–C6H4 and trimercapto-triazine scaf-
folds [7a,10]. Although the Au–S and the Au–P
lengths for 1 are slightly longer than those of the PPh3
analogues, [7a] these bond lengths in 2–5 are little influ-
enced by replacing the phosphine ligand from PPh3 to
PPh2(2-py) and P(C6H4–3-CF3)3 for the present series
of compounds.
˚
˚
2.992 A (Au1–Au3) and 2.971 A (Au2–Au4) for 2, while
˚
˚
the distance is 3.059 (1) A for 4 and 3.0909(5) A for 5,
respectively. The Au–Au distance in 1 is significantly
˚
shorter than that of the PPh3 analogue (3.0834(8) A)
[7a] and the Au–Au distances in 2 are shorter than that
of 1. The previous paper demonstrated that the intro-
duction of a bulky CF3 substituent at a meta-position
of each phenyl group in P(C6H5)3 induces aurophilicity
even for simple AuClPR3-type compounds [4b]. Perhaps
the same mechanism is effective on reinforcing the auro-
philicity in 1 as in the case of our previous study. The
structural details of the resulting supramolecules are de-
scribed in the crystal packing section. The Au–S bond
2.2. Crystal structures
˚
lengths are about 2.31 A for the present series of com-
˚
pounds and the Au–P distances are about 2.26 A, both
Fig. 6 shows the crystal packing of 1 viewed along the
c-axis, which clearly exhibits a cage structure made from
of which are close to those of known PPh3–thiolate
Au(I) derivatives [4]. P–Au–S bond angles for these
new compounds are almost linear (Table 2). Au–Au–S
angles are acute (about 80ꢁ), while Au–Au–P angles
are obtuse (about 105ꢁ) for 1, 4, and 5 where aurophilic
interactions are operative. However, these angles come
close to 90ꢁ in 2. A glance of the molecular structure
of 4 suggested that the Au3S3P3C6 core forms a plane.
However, the deviation from the best plane defined by
˚
Au–Au interactions (r(Au–Au) = 3.0021(7) A) and p–p
stacks. The shortest C–C distance among the carbon
atoms of the 1,3-S2–benzene ring (C1–C6) and those
of the benzene ring (C31–C36) in AuP(C6H4–3-CF3)3
˚
is 3.436 A for C4 and C33. In addition, distances among
˚
C2–C6 and C31–C36 are below 4.0 A and these two
planes are almost parallel. These findings support p–p
interaction between these benzene rings. It might be
anticipated from a glance of Fig. 6 that p–p stacks are
also operative between two C31–C36 planes. However,
distances among relevant carbon atoms are beyond the
range of the normal p–p interaction. In this vein, the
cage formation is incomplete. However, Fig. 7 indicates
˚
the Au3S3P3C6 core is significant for P1 (1.162 A) and
˚
S1 (1.289 A), which are bonded to Au1. Au1 is involved
in the intermolecular Au–Au interaction. The aurophilic
interaction should induce such a molecular deformation.
In 5, where the sterically more demanding phosphine li-

K. Nunokawa et al. / Journal of Organometallic Chemistry 690 (2005) 1332–1339
1335
Table 1
Crystal data
Compound
1
2
3
4
C59H49Au3Cl4N3P3S3
5
Empirical formula
Formula weight
Crystal system
Space group
C54H28Au2F18P2S2
1516.6
Monoclinic
P21/n
C55H36Au2Cl2F18P2S2
3018.3
C40H32Au2N2P2S2
1145.6
C75H45Au3F27P3S3
2238.1
1721.8.0
Monoclinic
P21
Monoclinic
Pc
Monoclinic
P21/c
Monoclinic
P21/n
˚
a (A)
24.84(1)
16.268(9)
13.728(8)
97.40(4)
4
17.426(3)
16.001(3)
18.393(3)
99.580(1)
2
18.710(1)
14.6611(8)
15.5821(9)
113.287(1)
4
8.858(2)
13.744(3)
23.371(5)
91.763(4)
2
16.708(3)
16.450(3)
29.009(5)
102.211(3)
4
˚
b (A)
˚
c (A)
b (ꢁ)
Z
dcalcd (g cmꢀ3
)
1.82
1.95
1.94
2.01
1.90
Crystal dimensions (mm3)
0.7 · 0.4 · 0.3
5501(5)
55.5
0.35 · 0.15 · 0.15
5057(2)
60.9
0.31 · 0.19 · 0.07
3926.0(4)
78.2
0.20 · 0.10 · 0.10
2844(1)
81.4
0.25 · 0.15 · 0.1
7792.6(6)
58.8
3
˚
V (A )
l(Mo Ka) (cmꢀ1
Diffractometer
2hmax (ꢁ)
)
MAC MXC3
45.0
298
SMART APEX
55.0
100
SMART APEX
55.0
100
SMART APEX
55.0
100
SMART APEX
55.0
293
Temperature (K)
Unique reflections
Reflections with |Fo| > 3r(|Fo|)
No. of parameters refined
7179
5110
15,890
13,172
8999
7937
10,648
10,320
676
11,217
8408
668
0.06
1324
0.039
0.054
461
0.06
1066
0.036
0.083
R
0.05
0.12
Rw2
0.17
0.17
P
P
P
Mo Ka radiation (k = 0.71073 A); R = iFo| ꢀ |Fc/i/|Fo|; Rw = { (|Fo| ꢀ |Fc|)2/ w(Fo)2|1/2, where w ¼ 1=r2ðF oÞ.
2
˚
that each cage contains two benzene molecules with side
to side and the C6 axes of trapped benzene molecules are
almost perpendicular to the cage or the channel. This
finding supports the view that the void behaves like a
nano-tube; the shortest edge-to-edge distance (S–S) is
inspection of the crystal-packing does not indicate sig-
nificant p–p stacks among aromatic rings, including pyr-
idine in 4. Each void of crystals for 2 and 4 contains
solvent CH2Cl2 molecule.
It is amazing that the solid structures of 2 and 3 are
dramatically changed by only slightly modifying the
aryl phosphine ligand as was described above. Neither
infinite chain nor helix is formed for 3. Although the
substituent effect of the CF3 group at a meta-position
on the aurophilicity was demonstrated in the previous
paper, [4b] it was supposed from recent developments
in crystal engineering [14,15] that the hydrogen bond
may assist the formation of the helix in 2. Therefore,
a source of hydrogen bond was sought among F atoms
and aromatic hydrogens in the molecular packing dia-
gram in 2. Two hydrogen atoms (H2 and H5) in the
core 1,4-S2–C6H4 pedestal are involved in hydrogen
bonds with one of the F atoms in a CF3 substituent
of the (3-CF3–C6H4)3P ligand; r(H–F) = 2.608 and
˚
˚
10.0 A, while the longest distance (P–P) is 17.3 A. Thus,
formed nano-channels are fused together to yield a dis-
torted honeycomb structure. Quite interestingly, the
PPh3 analogue of 1 does not provide such a channel-like
structure according to Laguna and co-workers [7a]. The
thermogravimetric analysis has shown that 1 loses its
weight slowly above 120 ꢁC, sharply above 139 ꢁC, and
color changes from pale yellow to yellow. The resulting
solid is not soluble in common laboratory solvents,
which suggests that the channel structure is destroyed
upon the loss of benzene molecules.
The (1,4-S2–C6H4){AuP(C6H4–3-CF3)3}2 unit is self-
assembled by aurophilicity at two opposite sites of the
1,4-S2–C6H4 scaffold to afford an 8 (eight)-shaped helix
(Fig. 8(a) and (b)) for 2; the main helical axis is parallel
to the c-axis. Aurophilicity plays a crucial role in con-
structing the helix like a hydrogen bond for protein heli-
ces and/or DNA. The details of the helical structure are
described below. 31P NMR shows a singlet at the highest
field among these five derivatives (d 31.9). The helical
structure should be responsible for this high-field shift.
The crystal-packing diagrams for 4 and 5 demon-
strate the formation of quasi one-dimensional chains
composed mainly of aurophilic interaction, the skele-
tons of which look like the teeth of a saw (Figs. 9 and
10). Fig. 9 also demonstrates that three P–Au–S moieties
in a unit are almost parallel to the corresponding three
P–Au–S moieties of neighboring molecules. Detailed
˚
2.649 A. Fig. 11 shows a part of the hydrogen bonds
among these atoms. These hydrogen bonds may assist
the formation of an 8-shaped loop. However, it is
apparent that the aurophilicity plays the crucial role
for the present ‘‘double helix’’ formation. This work af-
fords another example of the essential contribution of
aurophilicity to construct an ‘‘artificial DNA’’ other
than hydrogen bonds. Recently, Puddephatt et al.
[16] have reported quite interesting helical loops in
self-assembled polymers in which aurophilicity plays a
crucial role. Lippert et al. [17] have also reported inter-
esting molecular architectures composed of Ag ions
and 2,20-bipyrazine and classified them into an infinite
helix, an infinite chain, and an infinite loop. Our

1336
K. Nunokawa et al. / Journal of Organometallic Chemistry 690 (2005) 1332–1339
Table 2
˚
Selected bond lengths (A) and angles (ꢁ)
Compounds
Au–Au
1
2
3
4
5
3.001(1)
2.9714(8)
2.9923(8)
2.339(3)
3.055(7)
3.0909(5)
Au–S
Au–P
S–C
2.330(5)
2.320(4)
2.309(2)
2.306(2)
2.318(3)
2.325(6)
2.293(3)
2.308(2)
2.312(2)
2.297 (2)
2.315(3)
2.303(3)
2.227(3)
2.297(3)
2.260(3)
2.270(3)
1.77(1)
1.77(1)
1.76(1)
1.79(1)
1.81(1)
1.82(1)
1.83(1)
1.80(1)
1.83(1)
1.82(1)
2.271(5)
2.274(4)
2.262(2)
2.264(2)
2.269(3)
2.268(3)
2.266(3)
2.259(2)
2.265(2)
2.246(3)
1.76(2)
1.76(2)
1.772(8)
1.774(8)
1.77(2)
1.79(3)
1.77(2)
1.769(8)
1.781(7)
1.781(7)
P–C
1.81(2)
1.83(2)
1.83(2)
1.819(7)
1.830(8)
1.818(7)
1.823(7)
1.822(8)
1.825(7)
1.85(2)
1.84(2)
1.83(2)
1.84(2)
1.81(2)
1.86(2)
1.826(9)
1.821(9)
1.817(8)
1.814(8)
1.814(9)
1.814(9)
1.814(8)
1.81(1)
1.82(1)
1.82(1)
1.80(1)
1.83(1)
1.82(1)
1.82(1)
1.81(1)
P–Au–S
Au–S–C
Au–P–C
172.6(2)
174.1(2)
176.3(1)
176.30(6)
177.90(6)
172.5(2)
172.7(2)
178.8(2)
173.33(4)
174.52(7)
173.11(9)
174.0(1)
176.6(1)
106.5(4)
101.1(6)
105.0(6)
105.8(3)
107.4(3)
103.5(7)
108.6(7)
105.7(7)
101.7(3)
103.6(3)
105.0(3)
106.2(4)
107.4(4)
113.8(4)
115.6(4)
113.8(4)
112.4(4)
112.1(4)
115.3(4)
112.0(4)
111.7(4)
114.7(4)
112.3(4)
115.1(4)
112.7(3)
85.77(7)
83.08(8)
114.0(10)
105.3(6)
116.8(5)
111.8(5)
111.0(6)
118.1(6)
114.1(2)
112.7(3)
112.2(2)
111.4(2)
113.4(2)
115.8(2)
110.7(7)
113.6(7)
112.6(6)
119.1(7)
112.8(8)
108.1(8)
118.2(9)
113.2(7)
112.3(9)
113.3(3)
110.5(3)
119.2(3)
115.7(3)
114.9(3)
108.9(3)
113.8(3)
111.1(3)
114.0(3)
Au–Au–S
Au–Au–P
81.4(1)
81.3(1)
76.4(1)
76.2(1)
85.68(5)
81.06(5)
80.48(8)
97.21(8)
101.34(8)
102.92(8)
105.4(1)
100.7(1)
106.7(1)
111.1(1)
89.38(6)
97.28(6)
‘‘double helix’’ in 2 belongs to an infinite loop. Finally,
we would like to point out the possible contribution of
the present type of manifestation of aurophilicity to the
gold-sensing of DNA which has been highlighted
recently [13].
Synthesis of gold dendrimers has recently been tar-
geted for preparing gold nano-particles, [3] establishing
drug-delivery systems, and/or generating new intelligent
nano-materials [18]. Symmetric S3 pedestals are
quite promising to construct dendrimers by the use of

K. Nunokawa et al. / Journal of Organometallic Chemistry 690 (2005) 1332–1339
1337
Fig. 8. (a) A crystal packing diagram of 2 which exhibits an infinite
chain structure (side view). (b) A crystal packing diagram of 2 which
exhibits an 8-shaped ‘‘double helix’’ (top view).
Fig. 6. A crystal packing diagram of 1 along the c-axis. Only
important molecular parts are shown for clarity. The dotted Æ line
(ꢁ ꢁ ꢁ) indicates Au–Au interaction and hatched rings are concerned with
p–p interactions.
Fig. 9. A crystal packing diagram of 4 which exhibits an infinite quasi
one-dimensional chain, the skeleton of which is similar to saw-toothed.
constructing such oligomers. Therefore, another selec-
tion of PPh2(py) ligands in which N is located at a
3- or 4-position has been attempted and efforts for con-
structing a ‘‘larger dendrimer’’ are continuing in this
laboratory. Our recent paper has proved that this kind
of strategy is quite promising for constructing hetero-
nuclear supramolecules by the use of Au(S–4py)PR3
as a building block [19].
3. Experimental
Fig. 7. Benzene-filled channels in 1. Channels are almost along the c-
axis direction.
3.1. General comments
Reactions were carried out under an argon atmosphere
by standard Schlenk techniques. The reaction vessel was
covered with a piece of black cloth. The PPh2(2-pyridyl)
ligand was purchased from Aldrich. P(C6H4–3-CF3)3
was synthesized as reported previously [4b]. 1,3-Benzene-
dithiol, 1,4-benzenedithiol, and 1,3,5-benzenetrithiol
were purchased from Nihon Fine-chemicals Co. Ltd.
31P{1H} NMR spectra were recorded on a Varian
XL-200 spectrometer at 80.984 MHz. 31P{1H} NMR
divergent procedures. In this vein, 4 and 5 are quite
interesting as a kind of ‘‘dendrimer’’. Several attempts
have been made to generate a higher ‘‘dendrimer’’ by
connecting the pyridine-N in each PPh2(2-py) ligand
for 3 and 4 with other Au(I) and/or heterometal com-
ponents. Successful outcomes have not been obtained
because N-positions in this ligand are not suitable for

1338
K. Nunokawa et al. / Journal of Organometallic Chemistry 690 (2005) 1332–1339
Fig. 10. A crystal packing diagram of 5 which exhibits an infinite quasi one-dimensional chain.
3.3. (1,4-S2–C6H4){AuP(C6H4–3-CF3)3}2 Æ C6H6 Æ
CH2Cl2 (2)
160 mg (0.23 mmol) of AuClP(3-CF3–C6H4)3 was dis-
solved in 2 ml of CH2Cl2. To this was added dichlorom-
ethane solution (1 ml) of 1,4-benzenedithiol (17 mg, 0.12
mmol) and the mixture was stirred at room temperature
overnight. Solvents were distilled off at reduced pressure
and the resulting yellow solid was recrystallized from
CH2Cl2/benzene to afford bright yellow crystals of 2.
Yield 120 mg, 72%. 31P{1H} NMR (CDCl3); d 31.9.
Anal. Calc. for C48H28Au2F18P2S2 Æ CH2Cl2 Æ C6H6: C,
40.53; H, 2.22%. Found: C, 40.91 ; H, 2.06%.
Fig. 11. One of the hydrogen bonds between an F atom of the CF3 and
an aromatic hydrogen for 2.
3.4. (1,4-S2–C6H4){AuP(C6H5)2(2-pyridine)}2 (3)
200 mg (0.4 mmol) of AuClP(C6H4)2(2-pyridine) was
dissolved in 30 ml of CH2Cl2. To this was added metha-
nol solution (12 ml) of 1,4-benzenedithiol (28 mg, 0.2
mmol) and Na2CO3 (430 mg, 4 mmol), and the mixture
was stirred at room temperature for 2 h. Solvents were
distilled off at reduced pressure and the resulting pale yel-
low solid was extracted with each 5 ml of CH2Cl2 two
times. Then, the solvent was distilled at reduced pressure.
The residue was recrystallized from CH2Cl2/hexane to
afford pale yellow crystals of 3. Yield 155 mg, 73%.
31P{1H} NMR (CDCl3); d 38.1. Anal. Calc. for
C40H32Au2N2P2S2 (vacuum-dried sample): C, 45.29; H,
3.04; N 2.64%. Found: C, 45.29; H, 3.09; N 2.67%.
chemical shifts are referenced to external 85%
H3PO4.
3.2. (1,3-S2–C6H4){AuP(C6H4–3-CF3)3}2 (1)
700 mg (1 mmol) of AuClP(C6H4–3-CF3)3 was dis-
solved in mixed solvents of CH2Cl2 (20 ml) and acetone
(30 ml). To this was added methanol solution (40 ml) of
1,3-benzenedithiol (71 mg, 0.5 mmol) and KOH
(68 mg), and the mixture was stirred at room temperature
overnight. Solvents were distilled off at reduced pressure
and the resulting pale yellow solid was washed with small
amounts of methanol, hexane, and benzene successively.
The residue was extracted with each 10 ml of CH2Cl2
two times and the solvent was partially distilled at reduced
pressure. Then, small amount of benzene was added and
the mixture was stored in a refrigerator to afford pale yel-
low crystals of 1. Yield 520 mg in the form of single crys-
tals, 71%. Low solubility of this molecule in ordinary
organic solvents made it impossible to obtain a 31P
NMR spectrum. Anal. Calc. for C48H28Au2F18P2-
S2 Æ C6H6: C, 41.98; H, 2.22%. Found: C, 42.11; H, 2.50%.
3.5. (1,4-S3–C6H3){AuP(C6H5)2(2-pyridine)}3 Æ
2CH2Cl2 (4)
300 mg (0.6 mmol) of AuClP(C6H5)2(2-pyridine) was
dissolved in 30 ml of CH2Cl2. To this was added a meth-
anol solution (20 ml) of 1,3,5-benzenetrithiol (35 mg, 0.2
mmol) and Na2CO3 (430 mg, 4 mmol), and the mixture
was stirred at room temperature for 2 h. Solvents were
distilled off at reduced pressure and the resulting pale

K. Nunokawa et al. / Journal of Organometallic Chemistry 690 (2005) 1332–1339
1339
yellow solid was extracted with each 5 ml of CH2Cl2 two
References
times. Then, the solvent was distilled at reduced pres-
sure. The residue was recrystallized from CH2Cl2/hexane
to afford pale yellow crystals of 4. Yield 240 mg, 77%.
31P{1H} NMR (CDCl3); d 38.3. Anal. Calc. for
C57H45Au3N3P3S3 Æ 2CH2Cl2: C, 41.15; H, 2.87; N,
2.44%. Found: C, 41.39 ; H, 2.93; N, 2.38%.
[1] B.J. Holliday, C.A. Mirkin, Angew. Chem. Int. Ed. 40 (2001)
2022, and references therein.
[2] A. Muller, H. Reuter, S. Dillinger, Angew. Chem. Int. Ed. 34
¨
(1995) 2328;
R.W. Saalfrank, B. Demleitner, in: J.-P. Sauvage (Ed.), Transition
Metals in Supramolecular Chemistry, Perspectives in Supramo-
lecular Chemistry, vol. 5, Wiley–VCH, Weinheim, 1999, p. 1, and
references therein.
3.6. (1,3,5-S3–C6H3){AuP(C6H4–3-CF3)3}3 Æ C6H6 (5)
[3] D. Shmid (Ed.), Clusters and Colloids, VCH, Weinheim, 1994;
H. Schmidbaur (Ed.), Gold, Wiley, Chichester, 1999;
C. Joachim, S. Roth (Eds.), Atomic and Molecular Wires,
Kluwer, Dordrecht, 1997;
210 mg (0.3 mmol) of AuClP(C6H4–3-CF3)3 was dis-
solved in 15 ml of CH2Cl2. This solution was dropped into
a methanol solution (5 ml) of 1,3,5-benzenetrithiol (18 mg,
0.1 mmol) and KOH (17 mg), and the mixture was stirred
at room temperature for 1 h. The solution was subjected to
filtration and the solvent was distilled off at reduced pres-
sure. The resulting solid was extracted with benzene–
CH2Cl2 to give yellow solids 5. Yield 160 mg, 75%. Single
crystals were grown from CH2Cl2/hexane. 31P{1H} NMR
(CDCl3); d 35.3. Anal. Calc. for C75H45Au3F27P3S3; C,
40.23; H, 2.03%. Found: C, 39.94 ; H, 1.87%.
G.D. Mendenhall, A. Greenberg, J.F. Liebma, Molecules – from
Molecules to Materials, Chapman & Hall, London, 1995, and
references therein.
[4] (a) T. Yoshida, S. Onaka, M. Shiotsuka, Inorg. Chim. Acta 342
(2003) 319;
(b) K. Nunokawa, S. Onaka, T. Tatematsu, M. Ito, J. Sakai,
Inorg. Chim. Acta 322 (2001) 56;
(c) S. Onaka, Y. Katsukawa, M. Shiotsuka, O. Kanegawa, M.
Yamashita, Inorg. Chim. Acta 312 (2001) 100;
(d) K. Nunokawa, T. Sunahara, S. Onaka, K. Okazaki, H. Imai,
K. Inoue, T. Ozeki, Chem. Lett. 33 (10) (2004) 1300.
[5] M. Nakamoto, H. Koijman, M. Paul, W. Hiller, H. Schmidbaur,
Z. anorg. Allg. Chem. 619 (1993) 1341.
3.7. X-ray crystallography
`
[6] R.M. Davila, A. Elduque, T. Grant, R.J. Staples, J.P. Facjler Jr.,
Inorg. Chem. 32 (1993) 1749.
Selected crystals of 1 and 5 were glued to the top of a
fine glass rod for room temperature measurements and se-
lected crystals of other compounds 2–4 were attached on a
nylon loop for 100 K measurements. The reflection data
for 1 were collected by use of a MAC MXC3 diffractom-
eter with graphite-monochromated Mo Ka radiation
[7] (a) M.C. Gimeno, P.G. jones, A. Laguna, M. Laguna, R.
Terroba, Inorg. Chem. 33 (1994) 3932;
(b) I. del Rio, R. Terroba, E. Cerrada, M.B. Hursthouse, M.
Laguna, M.E. Light, A. Ruiz, Eur. J. Inorg. Chem. (2001) 2013,
and references therein.
[8] B.-C. Tzeng, C.-M. Che, S.-M. Peng, J. Chem. Soc. Chem.
Commun. (1997) 1771.
˚
(k = 0.71073 A) and those for 2–5 were collected on a Bru-
ker SMART-APEX CCD diffractometer with graphite-
[9] K. Henke, D.A. Atwood, Inorg. Chem. 37 (1998) 224;
K. Henke, A.R. Hutchinson, M.K. Krepps, S. Parkin, D.A.
Atwood, Inorg. Chem. 40 (2001) 4443.
˚
monochromated Mo Ka radiation (k = 0.71073 A) at
[10] W.J. Hunks, M.C. Jennings, R.J. Puddephatt, Inorg. Chem. 38
(1999) 5930.
100 K. The structures of these crystals were solved by di-
rect method using Sir-97 in a WinGx program package
and refined with anisotropic thermal parameters by full-
matrix least-squares programs on SHELXS-97 in a WinGx
program package [20]. The refinements were made on F2
data and the final R and wR2 values are given in Table
2. Tables for atomic coordinates, thermal parameters,
and bond lengths and angles are available as supplemen-
tary materials. Selected bond lengths and angles are given
in Table 1. The supplementary crystallographic data are
contained in CCDC 227471 (1), 227473 (2), 227472 (3),
227469 (4), and 237195 (5).
[11] M. Hong, Y. Zhao, W. Su, R. Cao, M. Fujita, Z. Zhou, A.S.C.
Chan, Angew. Chem. Int. Ed. 39 (2000) 2468.
[12] J.D.E.T. Wilton-Ely, A. Schier, N.W. Mitzel, H. Schmidbaur,
Inorg. Chem. 40 (2001) 6266.
[13] C.A. Mirkin, R.L. Letsinger, R.C. Mucic, J.J. Storhoff, Nature
382 (1996) 607;
R. Elghanian, J.C. Storhoff, R.C. Mucic, R.L. Letsinger, C.A.
Mirkin, Science 277 (1997) 1078.
[14] G.A. Jeffrey, An Introduction to Hydrogen Bonding, Oxford
University Press, Oxford, 1997, and references therein.
[15] G.R. Desiraju, T. Steiner, The Weak Hydrogen Bond, Oxford
Science, Oxford, 1999, and references therein.
[16] R.J. Puddephatt, J. Chem. Soc. Chem. Commun. (1998) 1055;
R.J. Puddephatt, Coord. Chem. Rev. 216–217 (2001) 313;
Z. Qin, M.C. Jennings, R.J. Puddephatt, Chem. Eur. J. 8 (2002)
735.
Acknowledgments
[17] R.-D. Schnebeck, E. Freisinger, B. Lippert, Eur. J. Inorg. Chem.
(2000) 1193.
This research was funded by Grants-in Aid for Scien-
tific research on Priority Area (No. 11136218, No.
12023221 ‘‘Metal-assembled Complexes’’) and by
Grants-in Aid for Scientific research (No. 14540514
(S.O.) and 15550047 (T.O.)) from the Ministry of Edu-
cation, Science, Sports and Culture, Japan. Thanks are
also due to CREST, JST (T.O.).
[18] P. Lange, A. Schier, H. Schmidbaur, Inorg. Chem. 35 (1996)
637.
[19] K. Nunokawa, S. Onaka, Y. Mizuno, K. Okazaki, T. Sunahara,
M. Ito, M. Yaguchi, H. Imai, K. Inoue, T. Ozeki, H. Chiba, T.
Yoshida, J. Organomet. Chem. 690 (2004) 48.
[20] G.M. Sheldrick, SHELX-97, Program for Analysis of Crystal
Structures, University of Go¨tiingen, Germany, 1997.