Luminescent gold(i) macrocycles with diphosphine and 4,4'-bipyridyl ligands

Article abstract of DOI:10.1039/b005251p

The complexes [{-Ph₂P(CH₂)_aPPh2AuNC5H4C5H4NAu-y2l:+ [CF3C

Full text of DOI:10.1039/b005251p

View Article Online / Journal Homepage / Table of Contents for this issue
Luminescent gold(I) macrocycles with diphosphine and
4,4Ј-bipyridyl ligands
Marie-Claude Brandys, Michael C. Jennings and Richard J. Puddephatt*
Department of Chemistry, University of Western Ontario, London, N6A 5B7, Canada
Received 30th June 2000, Accepted 10th October 2000
First published as an Advance Article on the web 1st December 2000
Ϫ
The complexes [{–Ph2P(CH2)nPPh2AuNC5H4C5H4NAu–}x]2xϩ [CF3CO2
]
2x (n = 1–6) were prepared as colourless,
air-stable solids by reaction of silver trifluoroacetate with the corresponding precursor complex [(CH2)n(Ph2PAuCl)2],
and subsequent treatment of the products [(CH2)n(Ph2PAuO2CCF3)2] with 4,4Ј-bipyridyl. The complexes are
suggested to exist in solution as an equilibrium mixture of linear oligomers and cyclic complexes. When n = 1, 3 or
5 the cationic complexes were shown to exist as 26-, 30- and 34-membered macrocyclic rings respectively; only when
n = 1 are there significant intramolecular Au ؒ ؒ ؒ Au contacts of 3.106(1) and 3.084(1) Å. Some of the complexes are
strongly emissive at room temperature in solution and in the solid state.
polymers was possible in some cases. The conformations of the
Introduction
polymers were determined by the nature of the diphosphine
There is growing interest in the synthesis of gold() complexes
ligands as indicated in Scheme 1.9
with macrocyclic, oligomeric or polymeric structures.1 The con-
This article describes an investigation of new gold()-
containing oligomers or macrocycles using 4,4Ј-bipyridyl as the
rigid-rod, bidentate N-donor ligand. The aim was to search for
struction of unusual molecular architectures is aided by the
tendency of gold() to form linear, 2-coordinate complexes that
may then undergo additional weaker gold–gold attractions.
Experimental and theoretical studies indicate that these
similarities and differences compared to the chemistry of the
trans-bis(pyridyl)ethylene ligands used previously (Scheme 1).9
aurophilic interactions with distances from 2.5 to 4 Å may give
The photoluminescence behaviour of the products was also
studied and molecular modeling was performed in an attempt
to rationalize the conformations adopted.
bonding forces of ca. 30 kJ molϪ1, comparable to hydrogen
bonding forces. As with hydrogen bonding in organic chemistry,
the aurophilic attractions can lead to unique molecular con-
formations, stereochemistry, crystal packing and physical
Results
properties in gold chemistry.1–8 If the gold() centres are bonded
to bidentate ligands, polymers or rings may be formed, and the
Au ؒ ؒ ؒ Au interactions can determine the favoured structure.9
For example, in cationic complexes containing diphosphine
and trans-bis(pyridyl)ethylene ligands, the occurrence of a
crossover from ring to polymer was demonstrated for the
complexes illustrated in Scheme 1. In solution, the lability of
the N-donor ligand allowed easy equilibration between rings
and oligomers and the orderly build-up of crystalline rings or
The complexes [(CH2)n(Ph2PAuCl)2] 2–7 (n = 1–6), were pre-
pared as colourless, air-stable products by treating two equiv-
alents of [AuCl(SMe2)], 1, with the corresponding diphosphine
ligand in acetone at room temperature.10–15 The 31P and 1H
NMR data can be found in Table 1. Addition of silver tri-
fluoroacetate to a suspension of 2–7 in tetrahydrofuran results
in formation of silver chloride and the corresponding soluble
bis(trifluoroacetate) complexes, as described in Scheme 2. These
Scheme 1 X = CF3CO2, P = PPh2.
DOI: 10.1039/b005251p
J. Chem. Soc., Dalton Trans., 2000, 4601–4606
This journal is © The Royal Society of Chemistry 2000
4601

View Article Online
Table 1 Comparison of 31P and 1H NMR data for complexes 2–13
Complex
δ(31P)
δ(CH2P)
δ(CH2CH2P)
2, 8
24.25, 25.59
32.12, 26.75
27.02, 21.70
29.92, 24.70
29.52, 24.29
29.98, 24.22
3.60, 4.40
2.62, 2.80
2.80, 2.94
2.45, 2.50
2.42, 2.49
2.37, 2.48
—
—
3, 9
4, 10
5, 11
6, 12
7, 13
1.91, 1.95
1.80, 1.83
1.67, 1.66
1.56, 1.55
Fig. 1 A view of the structure of complex 8. Two distant trifluoro-
acetate ions are not shown.
each type of methylene group for the diphosphine ligands.
A compilation of the NMR data can be found in Table 1.
The data are consistent either with formation of symmetrical
ring complexes in each case or with very rapid equilibration
between ring and chain structures, probably by easy reversible
displacement of bipyridine nitrogen donors from gold() by
trifluoroacetate.9
Molecular structures of the macrocycles [{–Ph2P(CH2)nPPh2-
AuNC5H4C5H4NAu–}2]4؉[CF3CO2؊]4, n ؍ 1, 8; n ؍ 3, 10; n ؍ 5,
12
Many attempts were made to grow crystals of the complexes
8–13. Finally crystals were grown for 8, 10 and 12 (n = 1, 3
and 5 respectively) from concentrated solutions of dichloro-
methane layered with pentane, but no single crystals for 9,
11 and 13 (n = 2, 4 and 6 respectively) suitable for single crystal
X-ray analysis could be grown. The structures of 8, 10 and 12
allow comparison of structures with different lengths of spacer
groups between the two phosphorus donors of the diphosphine
ligands.
Fig. 1 illustrates the molecular structure of the cationic
macrocycle [{–Ph2PCH2PPh2AuNC5H4C5H4NAu–}2]4ϩ in com-
plex 8, while Tables 2 and 3 contain selected bond lengths and
angles. The complex adopts a macrocyclic structure and the
geometry about each gold() centre is approximately linear, with
angles N(11)–Au(1)–P(1), N(21)–Au(2)–P(2), N(31)–Au(3)–
P(3) and N(41)–Au(4)–P(4) of 175.4(3), 176.7(3)Њ, 177.4(4) and
175.1(3)Њ respectively. Fig. 1 features a 26-membered ring with
four gold() atoms. The transannular gold–gold distances are
Au(1)–Au(2) 3.1061(8) and Au(3)–Au(4) 3.0841(8) Å, indic-
ating a significant aurophilic interaction, and the similar
spacing between roughly parallel pyridyl groups allows a
π- stacking attraction also.1–8
Scheme 2 P = PPh2. Reagents: (i) AgO2CCF3, ϪAgCl; (ii) 4,4Ј-
bipyridine.
complexes were not isolated, but their solutions were filtered
to remove silver chloride then treated with 4,4Ј-bipyridine to
give the cationic complexes [{Ph2P(CH2)nPPh2AuNC5H4C5H4-
NAu}x]2xϩ[CF3CO2
]
8–13 (n = 1–6) by displacement of the
Ϫ
2x
weakly coordinated trifluoroacetate ligands. These complexes
are colourless, air-stable solids but they are light sensitive and
were stored in the dark. They are slightly soluble in chloroform
and dichloromethane, and were characterized in solution by
NMR spectroscopy. The isolated complexes gave satisfactory
analytical data, and study by thermogravimetric analysis
indicated that each decomposed over a temperature range
from 170 to 245 ЊC to leave a residue of metallic gold.
The key question for the new complexes is whether they
exist as rings or chains or as an equilibrium mixture of both
(Scheme 3). The room temperature 31P NMR spectra of
8–13 contain only a singlet, indicating effective equivalence
of all phosphorus atoms. In addition, single resonances
were observed in the 1H NMR spectra for each of the bipyridyl
protons ortho and meta with respect to nitrogen, and for
Scheme 3 P = PPh2.
4602
J. Chem. Soc., Dalton Trans., 2000, 4601–4606

View Article Online
Fig. 2 A view of the structure of the cation 10. Two distant trifluoro-
acetate ions are not shown.
Table 2 Selected bond lengths [Å] and angles [Њ] for complex 8
Au(1)–N(11)
Au(1)–P(1)
Au(1)–Au(2)
Au(2)–N(21)
Au(2)–P(2)
2.098(9)
2.230(3)
3.1061(8)
2.075(10)
2.236(3)
Au(3)–N(31)
Au(3)–P(3)
Au(3)–Au(4)
Au(4)–N(41)
Au(4)–P(4)
2.044(10)
2.241(3)
3.0841(8)
2.107(10)
2.232(3)
Fig. 3 (a) A view of the structure of the cation 12. The disordered tri-
fluoroacetate ions are not shown and only the major component of the
disordered Au(4) and P(4) atoms is shown. Bond distances (Å) : Au(1)–
P(1) 2.245(9); Au(2)–P(2) 2.249(8); Au(3)–P(3) 2.22(1); Au(4)–P(4)
2.18(3). (b) A space filling model showing the partial encapsulation
of one trifluoroacetate anion, weakly bonded to Au(3) with
d(Au3 ؒ ؒ ؒ O86) = 3.16(4) Å.
N(11)–Au(1)–P(1)
N(21)–Au(2)–P(2)
175.4(3)
176.7(3)
N(31)–Au(3)–P(3)
N(41)–Au(4)–P(4)
177.4(4)
175.1(3)
Table 3 Selected bond lengths [Å] and angles [Њ] for complex 10
Au(1)–N(11)
Au(1)–P(1)
2.084(6)
2.240(2)
Au(2)–N(21)
Au(2)–P(2)
2.081(6)
2.240(2)
N(11)–Au(1)–P(1)
178.2(2)
N(21)–Au(2)–P(2)
178.3(2)
A view of the structure of the cation [{–Ph2P(CH2)3PPh2-
AuNC5H4C5H4NAu–}2]4ϩ in complex 10 is shown in Fig. 2, and
selected bond lengths and angles are listed in Table 3. The
geometry about the gold() center is linear with angles N(11)–
Au(1)–P(1) and N(21)–Au(2)–P(2) of 178.2(2) and 178.3(2)Њ
respectively and the complex adopts a macrocyclic structure,
containing a 30-membered ring with four gold() atoms. There
is a crystallographic centre of symmetry at the ring centre. The
intramolecular gold–gold distance, Au(1) ؒ ؒ ؒ Au(2) 5.279(1) Å,
lies well outside the range for a bonding interaction.1–9 For each
ring there are four trifluoroacetate counter ions and two of
these are partly enclosed in the macrocyclic ring, as shown in
Fig. 2, this being aided by twisting of the bridging bipy ligand
to create a suitable cavity. The twist of the 4,4Ј-bipyridyl ligand
in 10 (Fig. 2) is clearly greater than in 8 (Fig. 1). The shortest
Au ؒ ؒ ؒ O contacts in 10 are Au(1) ؒ ؒ ؒ O(55) 3.05(1) and
Au(2) ؒ ؒ ؒ O(55) 3.37(1) Å; this is probably best considered a
weak ionic attraction between the cation and anion.
Fig. 4 (a) A view of the structure of the complex 14. (b) A part of the
one-dimensional chain structure formed by intermolecular Au ؒ ؒ ؒ Au
bonding.
A view of the molecular structure of the cation [–{Ph2P-
(CH2)5PPh2AuNC5H4C5H4NAu–}2]4ϩ in complex 12 is shown
in Fig. 3(a), as determined by a partial structure determination.
The crystals were of poor quality but the data sufficient to
determine the structure of the cation. The intramolecular
gold–gold distances, the longest being Au(1)ؒ ؒ ؒAu(3) 8.05(1) Å,
are clearly non-bonding, as expected with the long (CH2)5
bridges between phosphorus donors. There is disorder in the
position of Au(4) and so there are two values of the distance
Au(2) ؒ ؒ ؒ Au(4) of 6.82(2) and 7.24(2) Å, both of which are
significantly shorter than the Au(1) ؒ ؒ ؒ Au(3) distance. There is
also resolved disorder in the position of P(4) but, though the
lighter atoms in this region [especially C(10) and the pyridyl
ring containing N(41)] must also be disordered, this was not
resolved and so bond distances are not discussed. The complex
contains an unsymmetrical 34-membered ring with four gold()
atoms. There is a large cavity within the ring, but only one
trifluoroacetate anion is partly enclosed (Fig. 3b), and it is
possible that the inefficient solid state packing leads to the
observed poor crystal quality. The increase in the ring cavity
size with increasing number of methylene spacer groups in the
diphosphine ligand is clear by comparison of Figs. 1–3.
In one attempt to grow crystals of complex 11, crystals of the
parent trifluoroacetate complex 14 were obtained. The struc-
ture is shown in Fig. 4, and bond parameters are listed in Table
4. The molecules exist in the anti conformation, such that there
is a center of symmetry at the midpoint of the C(9)–C(9Ј) bond
(Fig. 4a). If this conformation is also preferred in the bipyridine
complex it would of course lead to a polymeric rather than
cyclic structure. The complex forms one-dimensional chains in
the solid state through weak intermolecular Au ؒ ؒ ؒ Au bonding
(Fig. 4b) with d(Au ؒ ؒ ؒ Au) = 3.584(1) Å. The structure is
similar to that of the iodide derivative [IAuPPh2CH2CH2-
J. Chem. Soc., Dalton Trans., 2000, 4601–4606
4603

View Article Online
Table 4 Selected bond lengths [Å] and angles [Њ] for complex 14
Au–O(7)
Au–P
C(4)–C(5)
2.081(9)
2.214(3)
1.51(2)
C(5)–O(7)
C(5)–O(6)
1.18(1)
1.19(2)
O(7)–Au–P
O(7)–C(5)–O(6)
171.0(3)
128(1)
C(5)–O(7)–Au
123.8(9)
Table 5 Energies and transannular Au ؒ ؒ ؒ Au distances in complexes
8–13, as predicted by molecular mechanics calculationsa
Complex
E/kcal molϪ1
d(Au ؒ ؒ ؒ Au)/Å
8
9
10
11
12
13
Ϫ18
3.4
5.6
5.7
5.9
6.1
6.6
26
Ϫ10
25
Ϫ1
25
a The PAuN bond was constrained to be linear. No anions were present
in the MM calculations.
Fig. 5 The room temperature emission spectra of complex 9: (a) in
dichloromethane solution with λex = 254 nm and (b) in a solid KBr disk
with λex = 300 nm.
CH2CH2PPh2AuI], but this has a shorter intermolecular con-
tact d(Au ؒ ؒ ؒ Au) = 3.148(1) Å, probably due to the presence of
the softer iodide ligand that enhances the aurophilic attraction.8
Table 6 Emission spectral data (nm) of the complexes in dichloro-
methane solution or in the solid state as KBr disks
Solution
excitation
Solution
emission
Solid
Solid
Molecular modeling
Complex
excitation emission
Molecular mechanics calculations were carried out for the
cations in complexes 8–13, in the macrocyclic ring form, and
data are listed in Table 5. The trend in calculated energies is that
the ring structures are more favored for the diphosphine ligands
(CH2)n(PPh2)2, when n = 1, 3 or 5 compared to when n = 2, 4 or
6. This is presumably because the syn conformation of the
diphosphine required for macrocycle formation is more favor-
able with n odd, since the CH2 groups can then have the
staggered conformation with respect to each other. It is sug-
gested that there is a higher relative concentration of non-cyclic
oligomers, with the diphosphines in the anti conformation as
established in complex 14, in solution in the cases with n = 2, 4
and 6 and that this may be at least a partial reason for our
inability to crystallize these compounds, despite many attempts.
Table 5 also gives the mean Au ؒ ؒ ؒ Au separation for the
complexes. It overestimates the value when n = 1, probably
because there are no parameters for the transannular Au ؒ ؒ ؒ Au
attraction incorporated in the model. It underestimates the
value of d(Au ؒ ؒ ؒ Au) when n = 5, but the energy differences
with varying Au ؒ ؒ ؒ Au separation are small in this case, and it
is likely that anion inclusion (Fig. 3b) leads to the higher value
of d(Au ؒ ؒ ؒ Au) observed in the solid state structure of 12.
8
9
10
11
12
13
265, 350
270, 330
265, 340
265, 325
265, 330
265, 330
390a
390d
410c
385d
395g
395g
355
430b,c
305
417, 440, 468, 500e
420b,f
288, 378b
300
416, 440, 470, 500e
420b,e
265, 360b
279, 360b
440b,e
Approximate excitation energies are given for maximum emission inten-
sity; the emission energies are not significantly affected by changing λex.
a λex = 267 nm. b The emission was very weak and broad in these
cases. λex = 340 nm. λex = 254 nm. λex = 300 nm. λex = 288 nm.
g λex = 330 nm.
c
d
e
f
that the solution state emission can be assigned to a gold-
centred 5d–6p excited state. Since the 6p(Au) and π*(bipy)
orbitals can mix, a different combination may be involved for
the solution and solid state emissions.
Complexes 8–13 in dichloromethane solution each give a
single emission band, with the peak maximum ranging from
390 to 410 nm, as illustrated in Fig. 5 for complex 9. For all
complexes there was a red shift on going from solution to the
solid state (Table 6). This is a common effect in the emission
spectra of gold() complexes,17 and can be attributed to the
presence of intermolecular Au ؒ ؒ ؒ Au interactions or associ-
ation through π–stacking effects.
Photoluminescence studies
The emission spectra of complexes 8–13 at room temperature
were studied both in the solid state as KBr disks and in solution
in dichloromethane and data are listed in Table 6. In the solid
state 8, 10, 12 and 13 gave only a very broad, weak emission in
the region 420–440 nm. However, 9 and 11 give strong emission
with well defined structure as illustrated in Fig. 5 for 9. The
approximate energy difference between neighboring bands for
both 9 and 11 was 1330 cmϪ1. This splitting is similar to that
observed for ligand based π–π* transitions for bipyridine
derivatives, but it should be noted that 4,4Ј-bipyridine is not
emissive at room temperature and its protonated form displays
weak emission so a pure π–π* transition is not expected to lead
to room temperature emission.16 This factor indicates that the
emission is strongly affected by the presence of the gold()
centres and may indicate that the excitation has considerable
d(Au)–π*(bipy) character. The featureless solution emission,
combined with the observation of a significant red shift from
solution to solid state emission1,17 (Table 6, Fig. 5), suggests
Discussion
Each complex 8, 10 and 12 contains the expected linear, two-
coordinate gold() centres, with each gold atom bound to a
nitrogen atom and a phosphorus atom of the bridging 4,4Ј-
bipyridyl (bipy) and diphosphine ligand respectively. It is inter-
esting to compare the macrocyclic structures observed in this
work with the structures found for the corresponding com-
plexes with the bridging ligand trans-1,2-bis(4-pyridyl)ethylene
(bipyen) studied previously.9a A summary is given in Table 7.
With the ligand bipyen the structures were rationalized in
terms of the effects of Au ؒ ؒ ؒ Au bonding. Thus, for the
diphosphines Ph2P(CH2)nPPh2 with n = 1 or 2, the ring struc-
tures naturally contain interannular Au ؒ ؒ ؒ Au distances short
enough to give a significant aurophilic attraction and this gave a
thermodynamic preference for ring formation (Table 7). When
n > 2 this effect was not present and polymeric structures, in
4604
J. Chem. Soc., Dalton Trans., 2000, 4601–4606

View Article Online
Table 7 Summary of the structures of complexes [{Ph2P(CH2)n-
PPh2AuNNAu}x]2xϩ, as a function of the number of methylene groups n
and nature of the bridging ligand NN = bipy (bipyridine) or bipyen
repeated until the vapors produced were no longer brown. The
final solution (30 mL) was cooled to room temperature and
methanol (ca. 200 mL) added. With minimum light exposure,
SMe2 (ca. 10 mL, excess) was added dropwise, with continuous
stirring, until no red colour appeared. The white precipitate of
the product was recovered by filtration, washed with methanol,
diethyl ether and pentane and dried under vacuum. Yield: 13.8
g (93%). NMR in acetone-d6: δ(1H) 2.86 [s, 6H, CH3].
x, ya
n
bipy
bipyen
d(Au ؒ ؒ ؒ Au) b/Å
1
2
3
4
5
6
Ring, 2, 26
?
Ring, 2, 30
?
Ring, 2, 34
?
Ring, 2, 28
Ring, 2, 30
Polymer, ∞
Polymer, ∞
Polymer, ∞
Polymer, ∞
3.1
3.6
5.3
[CH2(Ph2PAuCl)2], 2. A solution of dppm (0.083 g, 0.215
mmol) in acetone (15 mL) was added to a suspension of
[AuCl(SMe)2] (0.127 g, 0.430 mmol) in acetone (25 mL). The
mixture was stirred for 2 h, to give a white precipitate, which
was collected by filtration, washed with acetone, diethyl ether
and pentane, and dried under vacuum. Yield: 0.16 g (89%).
Similarly prepared were [(CH2)2(Ph2PAuCl)2], 3, yield 76%;
[(CH2)3(Ph2PAuCl)2], 4, yield 84%; [(CH2)4(Ph2PAuCl)2], 5,
yield 85%; [(CH2)5(Ph2PAuCl)2], 6, yield 92%; [(CH2)6(Ph2-
PAuCl)2], 7, yield 90%. The complexes were characterized by
comparison of the NMR data with literature values.10–15
7.6
a y = Ring size when x = 2; ? indicates not known. b Approximate
average transannular Au ؒ ؒ ؒ Au separation in the ring structures only.
which the diphosphine could have either the syn (n = 3 or 5) or
anti (n = 4 or 6) conformation, were observed.9 For example,
when n = 3, a sinusoidal polymer was formed with closest gold–
gold contact of 5.98(1) Å, which is much too long to allow any
bonding to occur. In contrast, complex 10 has the ring structure
even though the transannular Au ؒ ؒ ؒ Au distance of 5.279(1) Å
is clearly non-bonding. In both complexes the diphosphine
adopts the syn conformation. Similarly, when n = 5, the com-
plex with bipyen is a sinusoidal polymer but with bipy the solid
state structure is the macrocyclic 12, although with syn con-
formation of the diphosphine in both cases. Why are different
structures obtained? First, it should be emphasized that there is
likely to be a fast equilibrium between structural forms in solu-
tion and so the form that crystallizes may not be the only, or
even the major, complex present. There will be a tendency for
the least soluble form present to crystallize, and so any discus-
sion based on electronic/steric effects may not be valid. The
main differences between bipy and bipyen are the distance
between the nitrogen donors (longer for bipyen) and the ten-
dency to twist from planarity (greater for bipy), but it is not
obvious that either effect would greatly influence the relative
stabilities of cyclic versus chain structures. The twisting of the
bipy ligand might affect π-stacking in the macrocycles, but this
is not evident in the structures of 10 and 12.
[{–Ph2PCH2PPh2AuNC5H4C5H4NAu–}2]4؉ [CF3CO2؊]4, 8.
Silver trifluoroacetate (0.057 g, 0.257 mmol) was added to a
suspension of [(CH2)(Ph2PAuCl)2] (0.108 g, 0.128 mmol) in
THF (15 mL). The mixture was stirred for 45 min, then filtered
through Celite into a flask containing a solution of 4,4Ј-bipyr-
idyl (0.0106 g, 0.0679 mmol) in THF (25 mL). After 1 h, the
white solid product was collected by filtration, washed with
THF, acetone, diethyl ether and pentane and dried under
vacuum. Yield: 0.11 g (72%). Calc. for C39H30Au2F6N2O4P2: C,
40.34; H, 2.61; N, 2.41. Found: C, 39.03; H, 2.56; N, 2.21%.
NMR in CDCl3: δ(1H) 9.04 [m, 4H, o-H py]; 7.47–7.36 [m,
28H, m-H py ϩ Ph]; 4.40 [m, 2H, CH2P]; δ(31P) 25.59 (s). TGA:
Au yield 35.2% (found), 33.94% (calc.); decomposition onset
245 ЊC.
The following complexes were prepared similarly.
[{–Ph2P(CH2)2PPh2AuNC5H4C5H4NAu–}x]2x؉ [CF3CO2؊]2x,
9. From [(CH2)2(Ph2PAuCl)2] (0.327 g, 0.379 mmol). Yield : 0.35
g (80%). Calc. for C20H16AuF3NO2P: C, 40.90; H, 2.75; N, 2.39.
Found: C, 40.53; H, 2.65; N, 2.44%. NMR in CDCl3: δ(1H) 8.81
(m, 4H, o-H py); 7.71–7.51 (m, 24H, m-H py ϩ Ph); 2.80 (m,
4H,CH2P); δ(31P) 26.75 (s). TGA: Au yield 34.0% (found),
33.5% (calc.); decomposition onset 245 ЊC.
The fact that the macrocyclic complexes 8, 10 and 12 could
be crystallized and are not luminescent, whereas 9 and 11 are
emissive but do not crystallize well, might suggest that different
structures are adopted and that 9 and 11 may have polymeric
structures in the solid state. This explanation could not be
confirmed crystallographically, but the molecular modeling
calculations do lend support to the hypothesis. In addition,
known polymeric gold() complexes are much more weakly
emissive than molecular analogues.1
[{–Ph2P(CH2)3PPh2AuNC5H4C5H4NAu–}2]4؉ 4[CF3CO2؊],
10. From [(CH2)3(Ph2PAuCl)2] (0.1052 g, 0.1199 mmol). Yield:
0.11 g (76%). Calc. for C41H34Au2F6N2O4P2: C, 41.42; H, 2.89;
N, 2.36. Found: C, 41.43; H, 2.85; N, 2.07%. NMR in CDCl3:
δ(1H) 8.80 [m, 4H, o-H py]. TGA: Au yield 33.2% (found),
33.1% (calc.); decomposition onset 231 ЊC.
[{–Ph2P(CH2)4PPh2AuNC5H4C5H4NAu–}x]2x؉ [CF3CO2؊]2x,
11. From [(CH2)4(Ph2PAuCl)2] (0.109 g, 0.123 mmol). Yield:
0.10 g (64%). Calc. for C21H18AuF3NO2P: C, 41.94; H, 3.02; N,
2.33. Found: C, 40.52; H, 3.09; N, 2.32%. NMR in CDCl3:
δ(1H) 8.77 (m, 4H, o-H py). TGA: Au yield 33.1% (found),
32.8% (calc.); decomposition onset 212 ЊC.
Experimental
NMR spectra were recorded using a Gemini 300 MHz
spectrometer. 1H and 31P chemical shifts are reported relative to
TMS and 85% H3PO4 respectively. Thermogravimetric analyses
were performed using a Perkin-Elmer Series 7 TGA instrument
at a heating rate of 20 ЊC minϪ1 on samples of 10 to 40 mg under
nitrogen. Molecular modeling experiments were performed
using the CAChe Scientific software package.18 Emission
spectra were recorded by using a Fluorolog-3 (ISA Jobin
Yvon-Spex) spectrofluorimeter. Typically, a 1 nm slit width was
used for solid samples and a 3 nm slit width for solution
samples in quartz cuvettes.
[{–Ph2P(CH2)5PPh2AuNC5H4C5H4NAu–}2]4؉ 4[CF3CO2؊],
12. From [(CH2)5(Ph2PAuCl)2] (0.164 g, 0.181 mmol). Yield:
0.05 g (64%). Calc. for C43H38Au2F6N2O4P2: C, 42.43; H, 3.15;
N, 2.30. Found: C, 41.57; H, 3.20; N, 2.30%. NMR in CDCl3:
δ(1H) 8.80 [m, 4H, o-H py]. TGA: Au yield 35.0% (found),
32.4% (calc.); decomposition onset 171 ЊC.
Preparations
[{–Ph2P(CH2)6PPh2AuNC5H4C5H4NAu–}x]2x؉ [CF3CO2؊]2x,
13. From [(CH2)6(Ph2PAuCl)2] (0.171 g, 0.186 mmol). Yield:
0.05 g (60%). Calc. for C22H20AuF3NO2P: C, 42.92; H, 3.28; N,
2.28. Found: C, 42.19; H, 3.28; N, 2.24%. NMR in CDCl3:
δ(1H) 8.82 [m, 4H, o-H py]. TGA: Au yield 32.9% (found),
32.0% (calc.); decomposition onset 202 ЊC.
[AuCl(SMe2)], 1. This procedure is modified from the liter-
ature method.10 Au (9.88 g, 50.2 mmol) was dissolved in boiling
aqua regia (160 mL). The volume was reduced to ca. 30 mL
by boiling, then HCl (ca. 100 mL, conc.) was added, and the
volume again reduced to ca. 30 mL. This procedure was
J. Chem. Soc., Dalton Trans., 2000, 4601–4606
4605

View Article Online
Table 8 Crystal data and structure refinements for complexes 8, 10, 12 and 14
8
10
12a
14
Formula
M
C79H62Au4Cl2F12N4O8P4
2405.97
C44H41Au2ClF6N2O4P2
1267.11
C84H76Au4F9N4O6P4
2320.23
C32H28Au2F6O4P2
1046.42
T/K
Crystal system
Space group
150
150
Monoclinic
P21/n
200
200
monoclinic
C2/c
Orthorhombic
Pna2(1)
Orthorhombic
Iba2
a/Å
b/Å
c/Å
26.5731(7)
18.7220(3)
16.8531(4)
—
8384.4(3)
4
13.2510(2)
19.1399(5)
18.1822(4)
105.789(2)
4437.43(17)
4
17.3326(12)
23.8685(13)
48.0435(18)
—
19875.7(19)
8
21.4364(7)
13.2988(5)
13.2773(6)
117.081(2)
3370.1(2)
4
β/Њ
V/Å3
Z
µ/mmϪ1
7.199
6.806
6.013
8.86
Reflections collected/independent
R1, wR2 [I > 2σ(I)]
34614, 20837
0.057, 0.130
43213, 9761
0.045, 0.125
24887, 12866
0.110, 0.235
24580, 3864
0.064, 0.146
a The data were of low quality and only a partial structure was obtained.
Crystal structure determinations
were refined with anisotropic thermal parameters. Crystal data
are given in Table 8.
Data were collected using a Nonius Kappa-CCD diffract-
ometer using COLLECT (Nonius, 1998) software, with
λ = 0.71073 Å. The unit cell parameters were calculated and
refined from the full data set. Crystal cell refinement and data
reduction were carried out using the Nonius DENZO package.
The data were scaled using SCALEPACK (Nonius, 1998)
and no other absorption corrections were applied. Friedel
pairs were kept separate. The SHELXTL 5.1 program
package19 was used to solve the structure by direct methods,
followed by refinement using successive difference Fourier
syntheses.
CCDC reference number 186/2237.
Acknowledgements
We thank the NSERC (Canada) for financial support to R. J. P.
and for a scholarship to M.-C. B.
References
1 D. M. P. Mingos and R. P. F. Kanter, J. Organomet. Chem., 1990,
384, 405; G. Jia, R. J. Puddephatt, J. D. Scott and J. J. Vittal,
Organometallics., 1993, 12, 3565; S.-J. Shieh, X. Hong, S.-M. Peng
and C.-M Che, J. Chem. Soc., Dalton Trans., 1994, 3067; P. M. Van
Calcar, M. M. Olmstead and A. L. Balch, J. Chem. Soc., Chem.
Commun., 1995, 1773; M. J. Irwin, G. Jia, N. C. Payne and R. J.
Puddephatt, Organometallics, 1996, 15, 51; M. J. Irwin, L. M. Muir,
K. W. Muir and R. J. Puddephatt, Chem. Commun., 1997, 219.
2 S. S. Pathameni and G. R. Desiraju, J. Chem. Soc., Dalton Trans.,
1993, 319.
3 J. J. Guy, P. G. Jones, M. J. Mays and G. M. Sheldrick, J. Chem. Soc.,
Dalton Trans., 1977, 8.
4 P. G. Jones, G. M. Sheldrick and E. Hadicke, Acta Crystallogr., Sect.
B, 1980, 36, 2777.
5 K. P. Hall and D. M. P. Mingos, Inorg. Chem., 1984, 22, 237.
6 Y. Jiang, S. Alvarez and R. Hoffmann, Inorg. Chem., 1985, 24, 749.
7 N. Rosch, A. Gorling, D. E. Ellis and H. Schmidbaur, Angew.
Chem., Int. Ed. Engl., 1989, 28, 1357; H. Schmidbaur, W. Graf and
G. Müller, Angew. Chem., Int. Ed. Engl., 1988, 27, 417.
8 P. M. Van Calcar, M. M. Olmstead and A. L. Balch, Inorg. Chem.,
1997, 36, 5231.
9 (a) M. J. Irwin, J. J. Vittal, G. P. A. Yap and R. J. Puddephatt, J. Am.
Chem. Soc., 1996, 118, 13101; (b) R. J. Puddephatt, Chem.
Commun., 1998, 1055; (c) M. J. Irwin, L. M. Rendina, J. J. Vittal and
R. J. Puddephatt, Chem. Commun., 1996, 1281.
10 A. Tamaki and J. K. Kochi, J. Organomet. Chem., 1974, 64, 411.
11 I. J. B. Lin, J. M. Hwang, D.-F. Feng, M. C. Cheng and Y. Wang,
Inorg. Chem., 1994, 33, 3467.
12 H. Schmidbaur, A. Wohlleben, F. Wagner, O. Orama and G.
Huttner, Chem. Ber., 1997, 110, 1748.
Crystals
of
[Au4(µ-Ph2PCH2PPh2)2(µ-NC5H4C5H4N)2]-
[CF3CO2]4ؒCH2Cl2 were grown by slow diffusion of pentane
into a methylene chloride solution. A white wedge shaped
crystal was mounted on a glass fiber. The cation was refined
anisotropically except for the phenyl rings which were also
constrained to be regular hexagons (AFIX 66). Of the four
anions, two refined well anisotropically but the other two were
disordered (modeled as 60/40 and 50/50 abundance), and the
geometries were constrained to be the same. The CH2Cl2
molecule of solvation was disordered also (50/50 abundance)
and the C–Cl (1.65 Å) and the Cl ؒ ؒ ؒ Cl (2.74 Å) distances were
fixed and the atoms kept isotropic. The crystal was twinned and
the BASF parameter refined to a value of 0.500.
Crystals of [(µ-Ph2P(CH2)3PPh2)Au(µ-NC5H4C5H4N)Au]2-
[CO2CF3]4ؒCH2Cl2ؒC5H12 were grown similarly. A pale yellow
cube was mounted on a glass fiber. All of the non-hydrogen
atoms in the cation were refined with anisotropic thermal
parameters. One of the crystallographically distinct trifluoro-
acetate anions was well behaved but the other was disordered
and modeled as two units in a 60/40 mix. The 60% model was
refined anisotropically and the 40% model isotropically, with
fixed geometry. The molecules of solvation were refined at half
occupancy.
Crystals of [Au4(µ-Ph2P(CH2)5PPh2)2(µ-NC5H4C5H4N)2]-
[CF3CO2]4 were grown similarly. A colorless needle was cut and
the resulting block mounted on a glass fiber. The crystal
was of poor quality and several constraints were needed, such
that the structure was only of sufficient quality to determine the
overall geometry. The pyridyl groups were constrained to be
flat, phenyl groups to be hexagons (AFIX 66); the P(CH2)5P
group to have P–C 1.85, C–C 1.54 Å and the geometry of the
trifluoroacetate anions was fixed. The atoms Au4, P4 and C10
were disordered over two positions and refined isotropically,
while Au1, Au2, Au3, P1, P2, P3, and C1–C9 were anisotropic.
The crystal was twinned and the BASF factor refined to a value
of 0.294.
13 P. A. Bates and J. M. Waters, Inorg. Chim. Acta, 1985, 98, 125.
14 D. F. Chodosh, D. S. Eggleston, G. R. Girard and D. T. Hill, Inorg.
Chim. Acta, 1985, 108, 221.
15 M. K. Cooper, K. Hendrick, M. L. McPartlin, E. Mitchell and
A. Scott, Inorg. Chim. Acta, 1984, 84, L9.
16 A. Sarkar and S. Chakravorti, J. Lumin., 1995, 65, 163; D. Li,
X. Hong, C.-M. Che, W.-C. Lo and S.-M. Peng, J. Chem. Soc.,
Dalton Trans., 1993, 2929.
17 L. Hao, R. J. Lachicotte, H. J. Gysling and R. Eisenberg, Inorg.
Chem., 1999, 38, 4616; V. W. W. Yam and K. K. W. Lo, Chem. Soc.
Rev., 1999, 28, 323; J. M. Forward, D. Bohmann, J. P. Fackler, Jr.
and R. J. Staples, Inorg. Chem., 1995, 34, 6330; W. B. Jones, J. Yuan,
R. Narayanaswamy, M. A. Young, R. C. Elder, A. E. Bruce and
M. R. M. Bruce, Inorg. Chem., 1995, 34, 1996.
18 CAChe Scientific 3.7, Molecular Mechanics, Conjugated Gradients,
Oxford Molecular Group, Oxford, 1994.
19 G. M. Sheldrick, SHELXTL, version 5.1, Siemens Analytical X-Ray
instruments, Madison, WI, 1998.
Crystals of [(CF3CO2)Au(µ-Ph2P(CH2)4PPh2)Au(O2CCF3)]
were grown from dichloromethane–pentane. A tiny, colorless
needle was mounted on a glass fibre. All non-hydrogen atoms
4606
J. Chem. Soc., Dalton Trans., 2000, 4601–4606