Crystallographic and computational studies of luminescent, binuclear gold(I) complexes, AuI2(Ph2P(CH 2)nPPh2)2I2 (n = 3-6)

Article abstract of DOI:10.1021/ic301954n

Four crystalline dimers of the type, Au^I₂(μ-PnP) ₂I₂, where PnP is PPh₂(CH₂) _nPPh₂ with n = 3, 4, 5, and 6 have been prepared and characterized by single-crystal X-ray

Full text of DOI:10.1021/ic301954n

Article
pubs.acs.org/IC
Crystallographic and Computational Studies of Luminescent,
Binuclear Gold(I) Complexes, AuI2(Ph2P(CH2)nPPh2)2I2 (n = 3−6)
Sang Ho Lim,† Jennifer C. Schmitt,‡ Jason Shearer,*,‡ Jianhua Jia,† Marilyn M. Olmstead,†
James C. Fettinger,† and Alan L. Balch*,†
†Department of Chemistry, University of California, Davis, California 95616, United States
‡Department of Chemistry, University of Nevada, Reno, Nevada 89557, United States
S
* Supporting Information
ABSTRACT: Four crystalline dimers of the type, AuI2(μ-
PnP)2I2, where PnP is PPh2(CH2)nPPh2 with n = 3, 4, 5, and 6
have been prepared and characterized by single-crystal X-ray
diffraction and by 31P NMR and infrared spectroscopy. AuI2(μ-
P3P)2I2 and AuI2(μ-P6P)2I2 are centrosymmetric dimers with
the planar AuIP2I units oriented in antiparallel fashion.
Remarkably, noncentrosymmetric AuI2(μ-P5P)2I2 has its
planar AuIP2I units oriented in parallel manner. AuI2(μ-
P4P)2(μ-I)2 is unique, since it contains four-coordinate gold
centers that are bridged by both iodide and diphosphine
ligands. All four compounds are luminescent as solids at room
temperature. B3LYP, B2PLYP, and spectroscopically oriented configuration interaction (SORCI) calculations have been
conducted to give insight into the electronic and geometric structures of the ground and first excited triplet states of the three
trigonal-planar complexes. The emission energies for the trigonal planar complexes are more strongly correlated with changes in
the Au−I bond length rather than changes in the P−Au−P angle.
we have reported the solid state interconversions of four
different crystalline compounds containing the Au2(dppe)2I2
unit (dppe is bis-(diphenylphosphino)ethane) that have been
observed and characterized through X-ray diffraction and
emission spectroscopy.16
INTRODUCTION
■
The luminescence arising from gold(I) complexes has received
considerable attention.1−4 While monomeric, two-coordinate
gold(I) complexes are generally nonemissive, they can become
emissive when aurophilic interactions between gold(I) centers
are allowed to occur. Such aurophilic interactions are attractive
connections between gold centers that are promoted by a
combination of relativistic and correlation effects.5−8 Although
many two-coordinate gold(I) complexes are nonluminescent in
solution, they can become luminescent in the solid state
because of self-association through aurophilic interactions.9−11
Mononuclear three-coordinate gold(I) complexes with a
planar geometry are frequently luminescent both in the solid
state and in solution.12−14 In general, aurophilic interactions are
not needed for three-coordinate gold(I) complexes to be
emissive.
Recently, the remarkably flexible molecule, Au2(μ-dppe)2Br2,
was shown to be a trigonal-planar complex in which aurophilic
interactions influenced the luminescence.15 Dimeric Au2(μ-
dppe)2Br2 can accommodate Au···Au separations that range
from 3.8479(3) to 3.0995(10) Å. These different Au···Au
separations occur in an array of crystalline solvates of the dimer.
Variations in the Au···Au separations along with alterations in
the Au−Br distances and in the P−Au−P angles result in
differences in the luminescence properties of crystals of Au2(μ-
dppe)2Br2. Crystals with Au···Au separations less than 3.5 Å
show green luminescence, while those with Au···Au separations
greater than 3.5 Å display orange luminescence. Additionally,
For monomeric, trigonal planar complexes that lack
aurophilic interactions, the excitation process is generally
2
2
attributed to a transition from the filled, in-plane dx −y ,dxy
orbitals on gold into a suitable empty orbital, which may be the
gold pz orbital or a ligand-based orbital. Several experimental
and computational studies have examined the structural
changes that are involved during excitation of three-coordinate
gold(I) complexes. For complexes of the types [Au(PR3)3]+
and Au(PR3)2X (X = halide) excitation is proposed to cause
extensive distortion from a trigonal planar ground state to a
distorted T-shape or beyond for the lowest triplet excited
state.17,18 Photocrystallographic studies of Au(PR3)2Cl·CHCl3
and Au(PR3)2Cl showed bond shortening in the excited
state.19,20 However, a similar study of a second polymorph of
Au(PR3)2Cl revealed little change between the structures of the
crystals before and after photoexcitation.21
Here we report on the preparation, structure, and
luminescence of a series of trigonal planar gold(I) complexes
̈
that were designed under the naive notion that chelated
monomers shown in Scheme 1 could be prepared from the
Received: September 6, 2012
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ic301954n | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Scheme 1. Trigonal Planar AuI Compounds
complexes consist of a single resonance and indicate that all
phosphorus atoms in the complexes are equivalent.
Structure of AuI2(μ-P3P)2I2. Crystal data for all complexes
considered here are given in Table 1. Selected bond distances
and angles are presented in Tables 2 and 3. For comparison,
structural data for the monomeric AuI(PPh3)2I is included in
the tables.22 A drawing of the structure of AuI2(μ-P3P)2I2 is
shown in Figure 1. The dimeric AuI2(μ-P3P)2I2 molecule packs
about a crystallographic center of symmetry so that one-half of
the dimer resides in the asymmetric unit. The gold center is
three-coordinate with bonds to two phosphorus atoms of two
different bridging ligands and to the terminal iodide ligand. The
AuIP2I unit is planar. The sum of the two P−Au−I angles and
the P−Au−P angle is 359.8°, but these individual bond angles
themselves differ considerably. The centrosymmetric nature of
the complex positions the two polar AuIP2I units in an
antiparallel alignment. The Au···Au separation is rather long,
5.0671(6) Å, and there is no significant aurophilic interaction
within the molecule.
Figure 2 shows a drawing of just the 12-membered ring of
atoms at the center of AuI2(μ-P3P)2I2. This view emphasizes
the antiparallel orientation of the two AuIP2I planes. Figure 2
also shows a comparable view of the other three compounds
reported here.
diphosphine ligands, Ph2P(CH2)nPPh2, and that alteration of
the length of the polymethylene chain would alter the P−Au−P
angles in these complexes in a sensible fashion. As shown
below, such chelated monomers were not obtained. Rather
bridged, binuclear complexes were obtained.
Structure of AuI2(μ-P4P)2(μ-I)2. A drawing of a molecule
of AuI2(μ-P4P)2(μ-I)2 is shown in Figure 3. One half of the
molecule is present in the asymmetric unit. The rest of the
molecule is generated by inversion through the crystallographic
center of symmetry. Within the series of four molecules
considered here, AuI2(μ-P4P)2(μ-I)2 is the only one with four-
coordinate gold ions. The gold ion is in a highly distorted
tetrahedral environment with bonds to two phosphorus atoms
and two bridging iodide ligands. The two Au−I distances
(3.0580(3), 3.0967(4) Å) are nearly equal. These are the
longest Au−I bonds seen in this series of molecules. Likewise,
the two Au−P distances (2.3092(7), 2.3248(6) Å) are nearly
RESULTS AND DISCUSSION
■
Synthesis and Crystal Growth. The reaction of a
suspension of AuII with the appropriate PnP ligand (n = 3−
6) produces colorless solids that can be crystallized from
dichloromethane/diethyl ether or chloroform/diethyl ether to
yield colorless crystals of the dimeric products. Infrared and 31
P
NMR spectra of these compounds are given in the
Experimental Section. The 31P NMR spectra for all four
Table 1. Crystal Data and Structure Refinement for AuI2(μ-PnP)2I2
AuI2(μ-P4P)2(μ-I)2
AuI2(μ-P5P)2I2·3CHCl3
AuI2(μ-P6P)2I2
I
Au2 (μ-P3P)2I2
formula
formula weight
T, K
C54H52Au2I2P4
1472.58
C56H56Au2I2P4
1500.62
90(2)
C58H60Au2I2P4·3(CHCl3)
1885.77
90(2)
C60H64Au2 I2P4
1556.73
90(2)
90(2)
color and habit
crystal system
space group
a, Å
colorless block
triclinic
colorless block
monoclinic
P21/c
colorless block
monoclinic
P21
colorless block
triclinic
P1
̅
P1
̅
10.527(5)
10.853(5)
12.979(5)
91.340(5)
113.602(5)
111.965(5)
1233.7(9)
1
10.124(2)
18.692(2)
15.574(2)
90
10.821(2)
20.111(4)
15.845(3)
90
9.984(2)
11.648(2)
12.673(2)
101.425(5)
93.982(5)
100.789(5)
1410.3(4)
1
b, Å
c, Å
α, deg
β, deg
119.930(5)
90
94.334(1)
90
γ, deg
V, Å3
2554.1(7)
2
3438.4(11)
2
Z
d
calcd, g cm−3
1.982
1.951
1.821
1.833
μ, mm−1
unique data
restraints
params.
7.356
7.108
5.639
6.440
5673
7857
20323
6463
0
0
1
0
145
289
701
307
a
R1
0.0356
0.0758
0.0189
0.0396
0.0221
0.0559
0.0140
0.0332
b
wR2
a
b
2
2
For data with I > 2σI, R1 = ∑||Fo| − |Fc||/∑|Fo|. For all data. wR2 = [∑w(Fo − Fc2)2/∑w(Fo )2]1/2
.
B
dx.doi.org/10.1021/ic301954n | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 2. Selected Experimental and Computed Bond Distances (Å) for AuI2(μ-PnP)2I2 and AuI(PPh2R)I
bond lengths, Å
AuI2(μ-P3P)2I2
AuI2(μ-P4P)2I2
AuI2(μ-P5P)2I2·3CHCl3
AuI2(μ-P6P)2I2
AuI(PPh2R)2I
a
d
Au1−I1
2.9310(14)
3.0580(3)
3.0006(6)
{3.007}
[3.079]
2.8052(4)
{2.973}v
[2.953]
2.754(1) Ph
{2.970} Me
[2.906] Me
b
{3.051}
c
[3.115]
Au2−I2
Au1−P1
Au1−P2
Au2−P3
Au2−P4
3.0967(4)
2.3092(7)
2.3248(6)
3.0831(7)
{3.016}
[3.098]
d
2.3048(18)
{2.372}
2.3089(9)
{2.378}
[2.354]
2.3347(6)
{2.402}
[2.385]
2.333(2) Ph
{2.409} Me
[2.662] Me
[2.383]
2.3217(17)
{2.393}
2.3017(9)
{2.378}
[2.365]
2.2997(6)
{2.375}
[2.384]
{2.412} Me
[2.665] Me
[2.390]
2.3093(8)
{2.379}
[2.386]
2.3115(8)
{2.379}
[2.341]
Au···Au
5.0671(6)
{5.151}
[5.247]
4.1952(2)
8.0206(3)
{8.089}
[8.242]
5.3095(2)
{5.595}
[7.011]
a
b
c
d
Experimental. {} Computed for the Ground State. [] Computed for the First Excited Triplet. Data from Bowmaker, G. A.; Dyason, J. C.; Healy,
P. C.; Engelhardt, M.; Pakawatchai, C.; White, A. H. J. Chem. Soc., Dalton Trans. 1987, 1089.
Table 3. Selected Experimental and Computed Bond Angles (deg) for AuI2(μ-PnP)2I2 and AuI(PPh2R)2I
bond angles (deg)
AuI2(μ-P3P)2I2
AuI2(μ-P4P)2(μ-I)2
AuI2(μ-P5P)2I2·3CHCl3
AuI2(μ-P6P)2I2
AuI(PPh2R)2I
a
d
P1−Au1−P2
152.88(5)
143.31(2)
164.63(3)
{164.0}
[175.3]
172.32(3)
{164.9}
[170.6]
94.19(2)
{96.7}
133.042(19)
{136.1}
132.13(7) Ph
{147.3} Me
[179.8] Me
b
{155.0}
c
[163.3]
[158.3]
P3−Au2−P4
P1−Au1−I1
d
109.14(4)
{108.9}
[103.3]
96.218(15)
102.311(19)
{115.7}
113.93(5) Ph
{107.1} Me
[90.0] Me
[90.2]
[106.8]
P1−Au1−I1A
P2−Au1−I1
105.098(15)
115.841(16)
97.77(4)
{96.0}
101.01(2)
{98.2}
123.086(15)
{102.1}
{105.4} Me
[90.5] Me
[93.4]
[88.7]
[94.5]
P2−Au1−I1A
P3−Au2−I2
92.369(19)
93.45(2)
{98.5}
[95.3]
P4−Au2−I2
93.36(2)
{97.2}
[94.0]
I1−Au1−I1A
94.061(10)
a
b
c
d
Experimental. {} Computed for the Ground State. [] Computed for the First Excited Triplet. Data from Bowmaker, G. A.; Dyason, J. C.; Healy,
P. C.; Engelhardt, M.; Pakawatchai, C.; White, A. H. J. Chem. Soc., Dalton Trans. 1987, 1089.
equivalent and are similar to those of the other molecules
reported here. As best seen in Figure 2, the iodide ligands lie
well outside of the planes of the two P−Au−P units.
geometry. For Au1, the sum of the two P−Au−I angles and the
P−Au−P angle is 359.83°, while for Au2 that sum is 359.13°.
Notice that the P−Au−P angles (164.63(3), 172.32(3)°) are
quite wide. Remarkably, the two polar AuIP2I units are aligned
in parallel fashion as can be seen in Figures 2 and 4.
Structure of AuI2(μ-P5P)2I2. The structure of AuI2(μ-
P5P)2I2 is shown in Figure 4. This is the only non-
centrosymmetric molecule in the series reported here. It has
no crystallographically imposed symmetry, and there are some
significant differences in the bond lengths and angles involving
the two gold centers. Each gold ion has planar, three-coordinate
Structure of AuI2(μ-P6P)2I2. Figure 5 shows a drawing of
the AuI2(μ-P6P)2I2 molecule. The molecule resides at a
crystallographic center of symmetry with one-half of the
molecule in the asymmetric unit. The structure of AuI2(μ-
C
dx.doi.org/10.1021/ic301954n | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 1. Drawing of the structure of AuI2(μ-P3P)2I2. Thermal
ellipsoids are shown at the 50% probability level. Hydrogen atoms are
not shown for clarity.
P6P)2I2 is similar to that of AuI2(μ-P3P)2I2 with the two planar
AuIP2I units arranged in antiparallel fashion as is apparent in
the perspective used in Figure 2. The sum of the two P−Au−I
angles and the P−Au−P angle is 358.44°.
Luminescence. Each of the compounds is luminescent at
room temperature. Figure 6 shows the emission and excitation
spectra for the planar, three-coordinate complexes, AuI2(μ-
P3P)2I2, AuI2(μ-P5P)2I2, and AuI2(μ-P6P)2I2. The excitation
profiles for these three complexes are similar, but the emission
maxima differ. The absorption maxima increase as the length of
the methylene chain (and hence n) increases, but there is no
apparent correlation of the emission energies with the P−Au−P
angles in the ground state of these molecules. Four-coordinate
AuI2(μ-P4P)2(μ-I)2 is also luminescent with an excitation
maximum at 353 nm and emission maximum at 440 nm.
Optimized Geometric Structures from Computational
Studies. To better understand the connection between the
structure of these complexes and their corresponding
luminescence properties, electronic structure calculations of
these complexes were performed. To obtain input structures for
these compounds, geometry optimizations (GOs) were
performed using the widely employed B3LYP hybrid density
functional on the compounds AuI2(μ-P3P)2I2 (n3), AuI2(μ-
P5P)2I2 (n5), and AuI2(μ-P6P)2I2 (n6). To determine the
influence of the tether on the AuIP2I unit, we also performed a
GO at the B3LYP level on the unconstrained compound
AuI(PPh2Me)2I. These calculations utilized a large def2-TZVPP
basis set23 for all atoms except gold, which was treated with a
VTZ-PP basis set on the valence electrons24 and the Stuttgart/
Bonn pseudopotential (ECO60MDF) on all core electrons.25
The resulting singlet state GO structures of n3 and n6 are in
good agreement with the experimental data (Tables 2 and 3).
We note that there is a small but consistent overestimation of
the Au-ligand bond lengths in the B3LYP structures vs the
experimentally determined bond lengths; however, this is a
well-known artifact of the B3LYP functional.26
Figure 2. Comparison of the ring conformations of AuI2(μ-PnP)2I2;
(A) n = 3, (B) n = 4, (C) n = 5, (d) n = 6. For reference, the dashed
lines connect the gold centers.
For n5 the computations produced the unusual parallel
alignment of the two AuP2I units seen in Figures 2 and 4, but
the large differences in the two gold(I) centers observed
experimentally could not be reproduced; we find that both
gold(I) centers in the GO structure of n5 are similar to one
another. Although some asymmetry is observed between the
D
dx.doi.org/10.1021/ic301954n | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 3. Drawing of the structure of AuI2(μ-P4P)2(μ-I)2. Thermal
ellipsoids are shown at the 50% probability level. Hydrogen atoms are
not shown for clarity.
Figure 5. Drawing of the structure of AuI2(μ-P6P)2I2. Thermal
ellipsoids are shown at the 50% probability level. Hydrogen atoms are
not shown for clarity.
Figure 6. Emission spectra for crystals of the planar, three-coordinate
complexes AuI2(μ-PnP)2I2 (n = 3, 5, 6) at room temperature.
tether, which is undoubtedly influencing the geometry about
the gold(I) center.
Figure 4. Drawing of the structure of AuI2(μ-P5P)2I2·3CHCl3.
Thermal ellipsoids are shown at the 50% probability level. Hydrogen
atoms are not shown for clarity.
Unlike the P−Au−P bond angles, the Au−I bond lengths
display a systematic trend across the n3−n6 series. As the
polymethylene tether increases there is a systematic decrease in
the Au−I bond length; the bond-length decreases from 3.051 Å
(n3) to 2.973 Å (n6). At 2.970 Å, the unconstrained compound
AuI(PPh2Me)2I demonstrates the shortest Au−I bond length of
the compounds examined computationally. The elongation of
the Au−I bond length as the polymethylene tether is shortened
can be readily rationalized. There are significant steric
interactions between the I atom and the methylene H atoms.
As the tether is elongated the phenyl groups are more capable
of “splaying open” allowing the methylene groups to open up
away from the Au−Au central axis. Thus, the Au−I bond length
can contract.
In addition to the ground state singlet structure (S0), we also
performed GOs of the first triplet excited state (T1) of n3, n5,
n6, and AuI(PPh2Me)2I. Selected metric parameters are shown
in Tables 2 and 3. The untethered complex AuI(PPh2Me)2I
displays an excited state geometry with all of the hallmarks of
those reported for other three coordinate gold(I) complexes in
their first excited triplet state.16,17 The overall structure is best
described as T-shaped; the P−Au−I bond angle compresses to
two gold(I) centers, the difference between the two gold(I)
geometries is minor (Tables 2 and 3). The two Au-centers in
the GO structure of n5 are most consistent with the less open
P−Au−P bond angle of AuI2(μ-P5P)2I2 (labeled Au2) derived
crystallographically. Even at the MP2 level of theory the
asymmetry observed for the two gold(I)-centers in AuI2(μ-
P5P)2I2 cannot be reproduced; we observe that the two P−
Au−P bond angles remain at ∼165°.
There are two key metric parameters of note in the
discussion below: the P−Au−P bond angles and the Au−I
bond lengths (Tables 2 and 3). The calculated P−Au−P bond
angles increase from 136.1° (n6) to 155.0° (n3) to 164.1° (n5),
as is observed crystallographically. Compound n6 possesses
metric parameters that are the closest to AuI(PPh2Me)2I, but it
still demonstrates significant distortions from the untethered
compound. Thus, all of the tethered complexes demonstrate
significant asymmetry from the unconstrained compound
AuI(PPh2Me)2I. These differences in metric parameters
between the complexes are likely a result of the polymethylene
E
dx.doi.org/10.1021/ic301954n | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
∼90° and the P−Au−P opens to 179.8°. The Au−P bond
length increases by ∼0.3 Å to 2.66 Å, while the Au−I bond
length undergoes a contraction to 2.91 Å. In contrast, the T1
surface for n3, n5, and n6 are not T-shaped. Furthermore, the
same nonsystematic changes in the P−Au−P bond angle
observed on the S0 surface are reflected on the T1 surface. The
P−Au−P bond angles become progressively more open from
n6 (158.3°) to n3 (163.3°) to n5 (170.6/175.3°). Thus, the
polymethylene tether is imposing significant geometric
constraints on the T1 surface as well as the S0 surface.
character (Figure 7b). Despite the fact that the lowest energy
emissions are all T1 → S0 processes originating from similar
states, we calculate different emission energies for all four
compounds investigated. There is a systematic red-shift in
emission energies as the polymethylene tether is elongated.
Compound n3 has a calculated emission energy of 425 nm, n5
has an emission energy of 476 nm, while n6 has its emission
maximum at 505 nm. Untethered AuI(PPh2Me)2I displayed the
most red-shifted emission energy of all of the compounds
investigated (529 nm).
Calculated Excitation and Emission Energies. Using the
GO structures from above we calculated the energies of the S1
← S0 and T1 → S0 transition energies. To achieve a high
degree of accuracy for these excitation and emission energies
we employed Neese’s spectroscopically oriented configuration
interaction (SORCI) methodology.27 Unlike time-dependent
density functional theory (TD-DFT) methods, the SORCI
method is a true multiconfigurational, ab initio computational
technique, and thus yields far more accurate excited state
energies.28 For these SORCI calculations, a modest CAS(4,4)
reference space was utilized, which proved adequate to
reproduce the experimentally observed excitation and emission
energies (Table 4). The S1 ← S0 excitation energies for n3, n5,
Origin of the Differential Emission Energies. Of the
different structural parameters, the two parameters that vary the
most from the S0 to the T1 structures are the Au−I bond
length and the P−Au−P bond angle. The potential energy
surfaces (PESs) of AuI(PPh2Me)2I were, therefore, constructed
for the lowest energy singlet and triplet surfaces to examine the
influence of the P−Au−P bond angle distortion and the Au−I
bond length. Geometries for the PESs were constructed at the
B3LYP level of theory using AuI(PPh2Me)2I. To deconvolve
the influence of the P−Au−P bond angle on the Au−I bond
length and the Au−I bond length on the P−Au−P bond angle
constrained GOs were performed. The PESs examining the P−
Au−P bond distortion (from 105 to 180° in 2.5° increments)
utilized a constrained Au−I bond length of 3.000 Å, while the
PESs examining the Au−I bond length (from 2.800 to 3.150 Å
in 0.025 Å increments) utilized a P−Au−P bond angle of 160°.
Final single-point energies for each GO structure along the S0
and T1 surfaces were calculated using the B2PLYP double-
hybrid functional.29 This functional is similar to the B3LYP
hybrid-functional, except it also mixes in a second-order
Møller−Plesset perturbation theory (MP2) correction (along
with the Hartee−Fock (HF) correction) to the DFT derived
energies yielding highly accurate energies.29,30
Table 4. Computed and Experimental Excitation and
Emission Maxima for Trigonal Planar Gold(I) Complexes
compd
n3
n5
n6
AuIIP2
390
S1 ← S0 (nm) excitation computed
experimental
387
370 sh
425
388
370
486
465
391
380
505
499
T1 → S0 (nm) emission computed
experimental
529
420
n6, and AuI(PPh2Me)2I are all similar to one another, and
similar to those experimentally observed. In all cases we see the
lowest energy excitation corresponds to a wavelength of ∼390
nm. Thus, the calculations strongly suggest that the nature of
the initial excitation process is best described as an excitation
from the first to second singlet surfaces. In all of these cases the
excitation process is best described as a metal-to-ligand charge-
transfer (MLCT) process originating from a state composed of
mostly Au(5dxy)/I(5px) antibonding character to a state that is
composed of exclusively phosphine ligand character (Figure
7a).
The PES for the S0 state results in a rather shallow energy
profile as the P−Au−P (Figure 8a). There is a minimum on the
PES for AuI(PPh2Me)2I (Au−I bond length 3.000 Å) at 138.3°,
which increase to 9.9 kJ mol−1 at 180°. There is another
maximum observed at 105°, which at 9.4 kJ mol−1 is not as
positive in energy as the point at 180°. This can be contrasted
with the T1 surface, where the range of 132−136° represents a
maximum on the T1 surface. In contrast there is a minimum of
the T1 surface at 175.3°, which is 19.3 kJ mol−1 lower in energy
than the maximum at 132.2°. This is consistent with the nature
of the electronic structures of the two states. Along the T1
surface the highest occupied molecular orbital (HOMO) is
made of a P(σ)-Au(6s) antibonding interaction. As the bond
distorts from 180° to 132.2° more Au(6s) character becomes
mixed into the HOMO, and the Au−P bonds become more
The lowest energy T1 → S0 emission process corresponds to
roughly the inverse of the excitation process; it corresponds to
a decay from a state corresponding to phosphine-ligand
character to a state composed of Au(5dxy)/I(5px) antibonding
Figure 7. Isosurface transition difference plots derived from the SORCI calculations (TZVPP/VTZ-PP-ECP) for the (A) S1← S0 and (B) T1→ S0
transitions for AuIIP2. The red areas correspond to the regions of the molecule that are gaining electron density while the ice-blue areas correspond
to regions of the molecule that are losing electron density.
F
dx.doi.org/10.1021/ic301954n | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 8. (A) B2PLYP (TZVPP/VTZ-PP-ECP) PES constructed for the P−A−P bond angle distortion maintaining a constant Au−I bond length.
The vertical lines represent the transitions from the T1 to S0 surfaces based on the predicted excited state P−Au−P bond angles from the B3LYP
geometry optimized structures. Energies are relative to the AuIP2I T1 → S0 transition. (B) B2PLYP (TZVPP/VTZ-PP-ECP) PES constructed for
Au−I bond length distortion with a fixed P−Au−P bond angle.
covalent. Thus, the antibonding HOMO of the T1 state
increases in energy as the P−Au−P bond angle contracts, and
the overall energy of the compound increases. In contrast, the
P(σ)-Au(6s) covalency is not strongly affected along the S0
surface as the bond distorts, hence the rather small influence on
the energy of the S0 state as the P−Au−P bond angle is
distorted.
angle and Au−I bond length. Along the P−Au−P PES we
predict a small change in the T1 → S0 energy as the bond P−
Au−P bond angle is closed, with a ΔE = 753.6 cm−1 going from
the least energetic emission energy (AuI(PPh2Me)2I) to the
largest emission energy (n6). For the Au−I bond length PES
we see a more dramatic change in emission energy as a function
of Au−I bond length with a ΔE = 1,659 cm−1 going from
AuI(PPh2Me)2I to n3.
This can be contrasted with the changes in the energy of the
S0 and T1 surfaces as the Au−I bond length is altered (Figure
8b). For the S0 state there is a large change in energy as the
Au−I bond length changes. There is a minimum along the S0
PES for AuI(PPh2Me)2I at 2.982 Å (P−Au−P bond angle
160°). Contraction of the Au−I bond length to 2.800 Å leads to
a 37.6 kJ mol−1 increase in energy, while elongation to 3.150 Å
leads to only a 7.7 kJ mol−1 increase in energy. In contrast with
the S0 surface, the T1 surface shows little influence in the
change of energy as the Au−I bond length is altered. A
contraction of the Au−I bond length from the equilibrium
bond length of 2.911 Å on the T1 surface to 2.800 Å shows an
increase in energy of only 3.8 kJ mol−1, while elongation of the
bond to 3.150 Å leads to an increase of 11.81 kJ mol−1. Once
again, we can relate this to the electronic structure of the S0 vs
T1 states. The HOMO of the S0 state is composed of a
Au(5dx2‑y2)/I(5pz) antibonding interaction. As one compresses
the Au−I bond, the Au−I bond covalency increases and the
energy of the HOMO increases. Hence, the energy of the state
increases. In contrast, contraction of the Au−I bond length
leaves the T1 surface virtually unaltered; as we have relieved
antibonding character in the Au−I bond, contraction of the
Au−I bond is not as energetically unfavorable on the T1 PES.
Using the constructed PESs we can determine the emission
energies of AuI(PPh2Me)2I as a function of P−Au−P bond
Combining the energies of the two PESs allows us to
understand the origins in the emission energies. Although the
observed ΔE from the PESs are underestimated by at least one-
half, the trend obtained in the SORCI calculations and the
experimental data is observed. From the PES results we derive a
systematic change in emission energies as the polymethylene
tether length is increased. Compound n3 would be predicted to
have the highest emission energy while AuI(PPh2Me)2I will
have the lowest emission energy. This compares well with both
the SORCI calculations for all four compounds and the
experimental data for n3, n5, and n6. Thus, it is reasonable to
propose that the differences in emission energy observed
experimentally are largely a result of changes in the Au−I bond
length with minor contributions from the P−Au−P bond angle
distortions.
CONCLUSIONS
■
The results reported here show that the diphosphine ligands
prefer to form bridged, binuclear complexes rather than
chelated monomers. The dimers with an odd number of
methylene groups in the PnP ligand have these methylene
groups arranged in a regular zigzag fashion. However, when an
even number of methylene groups is present, these groups
assume a less regular, kinked arrangement. These features are
G
dx.doi.org/10.1021/ic301954n | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
bis(diphenylphosphino)hexane: yield, 200 mg (83.2%). Colorless
blocks suitable for X-ray structure determination and emission
spectroscopy were obtained by slow diffusion of diethyl ether into a
dichloromethane solution of the product.
seen in the related, bridged, binuclear complexes that have been
observed for a number of planar, four-coordinate metal
complexes formed from PnP-type ligands.31−34 As the
computational and experimental studies show, the tether
between the phosphorus atoms imposes interesting constraints
on bond distances and angles within the gold coordination
sphere.
Infrared spectrum: 3044w, 2913w, 1585w, 1478m,1428s, 1401w,
1102s, 1024w, 976w, 947m, 854w, 773w, 739m, 687s, 528m, 504s,
473s, 447m.
31P NMR spectrum: singlet at 35.6 ppm.
Unlike the situation with Au2(μ-dppe)2Br2 and Au2(μ-
dppe)2I2, where the crystallization of a variety of different
solvates produced significant changes in the aurophilic
interactions and luminescence because of the flexibility within
the molecule,15,16 we did not encounter such variations in the
complexes reported here. Only AuI2(μ-P5P)2I2·3CHCl3 crystal-
lized as a solvate. The others did not need incorporation of
solvent molecules to crystallize. In none of the compounds
reported here did the gold centers approach one another close
enough to allow aurophilic interactions.
The four complexes reported here are luminescent as solids
at room temperature. For the three trigonal planar complexes,
the emission energy is more strongly correlated with changes in
the Au−I bond length rather than changes in the P−Au−P
angle. This observation was rationalized in terms of bonding
interactions involved in these gold complexes; contraction of
the Au−I bond results in a significant destabilization of the S0
surface because of an increase in the energy of the antibonding
Au(5dx2‑y2)/I(5pz) orbital. Such a large destabilization of the
filled Au(5dx2‑y2)/I(5pz) antibonding orbital is not produced
upon P−Au−P bond angle distortions.
X-ray Crystallography and Data Collection. The crystals were
removed from the glass tubes in which they were grown together with
a small amount of mother liquor and immediately coated with a
hydrocarbon oil on a microscope slide. A suitable crystal of each
compound was mounted on a glass fiber with silicone grease and
placed in the cold stream of a Bruker SMART 1000 CCD with
graphite monochromated Mo Kα radiation at 90(2) K.
The structures were solved by direct methods and refined using all
data (based on F2) using the software SHELXTL 5.1. A semiempirical
method utilizing equivalents was employed to correct for absorptions.
Hydrogen atoms were added geometrically and refined with a riding
model.36
Physical Measurements. Infrared spectra were recorded on a
Bruker ALPHA FT-IR spectrometer. Fluorescence excitation and
emission spectra were recorded on a Perkin-Elmer LS50B
luminescence spectrophotometer.
Computational Details. GOs on ground and triplet excited states
were performed using Turbomole v. 6.337 with the crystallographic
data as input structures. In addition, the singlet and triplet structures of
AuI(PPh2Me)2I were initially calculated within Turbomole. Absorption
and emission spectra were simulated using ORCA v. 2.9.0.38 The PES
was constructed for the molecule AuI(PPh2Me)2I using ORCA v.
2.8.20. All calculations utilized an all electron doubly polarized basis
set of triple-ζ quality on all atoms except gold, which utilized a
Stuttgart/Bonn pseudopotential and corresponding 19-valence elec-
tron triple-basis set. GOs were performed at the B3LYP level of
theory,39 which in ORCA made use of the RIJCOX approximation.40
Single point energies were calculated using the B2PLY double-hybrid
functional of Grimme and co-workers.41 Absorption and emission
spectra were simulated using the spectroscopically oriented config-
uration interaction (SORCI) approach,42 using a CAS(4,4) reference
space, a selection threshold of 1 × 10−6 Eh, a prediagonalization
threshold of 10−4, and a natural orbital threshold of 10−5.
EXPERIMENTAL SECTION
■
Materials. A previously reported procedure was used for the
preparation of (tht)AuCl (tht = tetrahydrothiophene).35 The
phosphine ligands were purchased from Alfa Inorganics.
AuI2(μ-P3P)2I2. A 100 mg (0.309 mmol) portion of AuI was
suspended in 30 mL of dichloromethane. 1,3-Bis(diphenylphosphino)-
propane (250 mg, 0.606 mmol) was added to this suspension. After
stirring for 2 h, all the solids dissolved. The solution was filtered, and
then the solvent was removed in a vacuum. The white solid was
collected and washed with diethyl ether: yield, 160 mg (70.3%).
Colorless prisms suitable for the X-ray structure determination and
emission spectroscopy were obtained by slow diffusion of diethyl ether
into a dichloromethane solution of the product.
ASSOCIATED CONTENT
■
S
* Supporting Information
X-ray crystallographic files in CIF format for AuI2(μ-P3P)2I2,
AuI2(μ-P4P)2(μ-I)2, AuI2(μ-P5P)2I2·3(CHCl3), and AuI2(μ-
P6P)2I2. This material is available free of charge via the
Internet at http://pubs.acs.org.
Infrared spectrum: 3050w, 2923w, 1481m, 1434s, 1305w, 1097s,
1028w, 995w, 949m, 841w, 741m, 688s, 528m, 510s, 478s, 456m.
31P NMR spectrum: singlet at 35.1 ppm.
AuI2(μ-P4P)2(μ-I)2. This complex was prepared using the method
outlined for Au2(μ-P3P)2I2 from 100 mg (0.309 mmol) of AuI and
260 mg (0.610 mmol) of 1,4-bis(diphenylphosphino)butane: yield,
190 mg (81.8%). Colorless blocks suitable for the X-ray structure
determination and emission spectroscopy were obtained by slow
diffusion of diethyl ether into a dichloromethane solution of the
product.
AUTHOR INFORMATION
Corresponding Author
*E-mail: shearer@unr.edu (J.S.), albalch@ucdavis.edu (A.L.B.).
Notes
■
The authors declare no competing financial interest.
Infrared spectrum: 3049w, 2922w, 1479m, 1435s, 1401w, 1099s,
1028w, 996w, 893w, 742m, 693s, 517m, 481m, 454m.
ACKNOWLEDGMENTS
■
31P NMR spectrum: singlet at 35.6 ppm.
We thank the Petroleum Research Fund (Grant 37056-AC)
and the U.S. National Science Foundation (Grants CHE-
1011760, CHE-0716843, and CHE-0844234) for support and
Ms. K. England for experimental asistance.
AuI2(μ-P5P)2I2·3(CHCl3). This complex was prepared using the
method outlined for Au2(μ-P3P)2I2 from 265 mg (0.602 mmol) of 1,5-
bis(diphenylphosphino)pentane: yield, 200 mg (84.6%). Colorless
blocks suitable for the X-ray structure determination and emission
spectroscopy were obtained by slow diffusion of diethyl ether into a
chloroform solution of the product.
REFERENCES
■
(1) Balch, A. L. Struct. Bonding (Berlin) 2007, 123, 1.
(2) He, X.; Yam, V. W.-W. Coord. Chem. Rev. 2011, 255, 2111.
(3) Yam, V. W. W.; Chung-Chin Cheng, E. C. C. Chem. Soc. Rev.
2008, 37, 1806.
Infrared spectrum: 3049w, 2919w, 1480m,1428s, 1406w, 1102s,
1013w, 993w, 947m, 840w, 741m, 685s, 527m, 503s, 476s, 451m.
31P NMR spectrum: singlet at 35.7 ppm.
AuI2(μ-P6P)2I2. This complex was prepared using the method
outlined for Au2(μ-P3P)2I2 from 275 mg (0.605 mmol) of 1,5-
(4) Tiekink, E. R.T.; Kang, J.-G. Coord. Chem. Rev. 2009, 253, 1627.
H
dx.doi.org/10.1021/ic301954n | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(5) Pyykko, P. Chem. Rev. 1997, 97, 597.
(40) Neese, F.; Wennmohs, F.; Hansen, A.; Becker, U. Chem. Phys.
2009, 356, 98.
(41) Schwabe, T.; Grimme, S. Acc. Chem. Res. 2008, 41, 569.
̈
(6) Pyykko, P. Angew. Chem., Int. Ed. 2004, 43, 4412.
̈
(7) Pyykko, P. Chem. Soc. Rev. 2008, 37, 1967.
̈
(8) Schmidbaur, H.; Schier, A. Chem. Soc. Rev. 2008, 37, 1931.
(9) White-Morris, R. L.; Olmstead, M. M.; Jiang, F.; Tinti, D. S.;
Balch, A. L. J. Am. Chem. Soc. 2002, 124, 2327.
(42) Neese, F. J. Chem. Phys. 2003, 119, 9428.
(10) White-Morris, R. L.; Olmstead, M. M.; Balch, A. L. J. Am. Chem.
Soc. 2003, 125, 1033.
(11) Rios, D.; Pham, D. M.; Fettinger, J. C.; Olmstead, M. M.; Balch,
A. L. Inorg. Chem. 2008, 47, 3442.
(12) Ziolo, R.; Lipton, S.; Z. Dori, Z. Chem. Commun. 1970, 1124.
(13) McCleskey, T. M.; Gray, H. B. Inorg. Chem. 1992, 31, 1773.
(14) King, C.; Khan, M. N. I.; Staples, R. J.; Fackler, J. P., Jr. Inorg.
Chem. 1992, 31, 3236.
(15) Lim, S. H.; Olmstead, M. M.; Balch, A. L. J. Am. Chem. Soc.
2011, 133, 10229.
(16) Lim, S. H.; Olmstead, M. M.; Balch, A. L. Chem. Sci. 2013, 4,
311−318.
(17) Barakat, K. A.; Cundari, T. R.; Omary, M. A. J. Am. Chem. Soc.
2003, 125, 14228.
(18) Sinha, P.; Wilson, A. K.; Omary, M. A. J. Am. Chem. Soc. 2005,
127, 12488.
(19) Hoshino, M.; Uekusa, H.; Ohashi, Y. Bull. Chem. Soc. Jpn. 2006,
79, 1362.
(20) Hoshino, M.; Uekusa, H.; Sonoda, S.; Takuhiro Otsuka, T.;
Kaizu, Y. Dalton Trans. 2009, 3085.
(21) Hoshino, M.; Uekusa, H.; Ishii, S.; Otsuka, T.; Kaizu, Y.; Ozawa,
Y.; Toriumi, K. Inorg. Chem. 2010, 49, 7257.
(22) Bowmaker, G. A.; Dyason, J. C.; Healy, P. C.; Engelhardt, M.;
Pakawatchai, C.; White, A. H. J. Chem. Soc., Dalton Trans. 1987, 1089.
(23) (a) Weigend, F.; Ahlrichs, R. Phys. Chem. Chem. Phys. 2005, 7,
3297. (b) Schaefer, A.; Horn, H.; Ahlrichs, R. J. Chem. Phys. 1992, 97,
2571. (c) Schaefer, A.; Huber, C.; Ahlrichs, R. J. Chem. Phys. 1994,
100, 5829. (d) Eichkorn, K.; Weigend, F.; Treutler, O.; Ahlrichs, R.
Theor. Chem. Acc. 1997, 97, 119. (e) Weigend, F.; Furche, F.; Ahlrichs,
R. J. Chem. Phys. 2003, 119, 12753. (f) Peterson, K. A.; Figgen, D.;
Goll, E.; Stoll, H.; Dolg, M. J. Chem. Phys. 2003, 119, 11113.
(24) Peterson, K. A.; Puzzarini, C. Theor. Chem. Acc. 2005, 114, 283.
(25) Figgen, D.; Rauhut, G.; Dolg, M.; Stoll, H. Chem. Phys. 2005,
311, 227.
(26) Neese, F. Coord. Chem. Rev. 2009, 253, 526.
(27) Neese, F. J. Chem. Phys. 2003, 119, 9428.
(28) Neese, F.; Petrenko, T.; Ganyushin, D.; Olbrich, G. Coord.
Chem. Rev. 2007, 251, 288.
(29) Schwabe, T.; Grimme, S. Acc. Chem. Res. 2008, 41, 569.
(30) Neese, F.; Hansen, A.; Wennohs, F.; Grimme, S. Acc. Chem. Res.
2009, 42, 641.
(31) Balch, A. L.; Tulyathan., B. Inorg.Chem. 1977, 16, 2840.
(32) Wang, H.-H.; Pignolet, L. H.; Reedy, P. E., Jr.; Olmstead, M. M.;
Balch, A. L. Inorg. Chem. 1987, 26, 377.
(33) Balch, A. L.; Davis, B. J.; Olmstead, M. M. Inorg. Chem. 1990,
29, 3066.
(34) Samar, D.; Fortin, J.-F.; Fortin, D.; Decken, A.; Harvey, P. D. J.
Inorg. Organomet. Polym 2006, 15, 411.
(35) Uson, R.; Laguna, A.; Laguna, M. Inorg. Synth 1989, 26, 85.
(36) Sheldrick, G. M. Acta Crystallogr., Sect. A 2008, 64, 112.
(37) Turbomole, version 6.2; University of Karlsruhe and
Forschungszentrum Karlsruhe GmbH: Karlsruhe, Germany, 2011.
(38) (a) Neese, F. ORCA, An ab Initio, Density Functional, and
Semiempirical Program Package, version 2.8.20; Institute for Physical
and Theoretical Chemistry, Universitat Bonn: Bonn, Germany, 2010.
̈
(b) Neese, F. ORCA, An ab Initio, Density Functional, and Semiempirical
Program Package, version 2.9; Max-Planck-Institute for Bioninorganic
Chemistry: Mulheim a. d. Ruhr, Germany, 2011.
̈
(39) (a) Vosko, S. J.; Wilk, M.; Nusair, M. Can. J. Phys. 1980, 58,
1200. (b) Becke, A. J. Chem. Phys. 1986, 84, 4524. (c) Becke, A. J.
Chem. Phys. 1988, 88, 1053. (d) Becke, A. Phys. Rev. A 1988, 38, 3098.
(e) Perdew, J. P. Phys. Rev. B 1986, 34, 7406. (f) Perdew, J. P. Phys.
Rev. B 1986, 33, 8822.
I
dx.doi.org/10.1021/ic301954n | Inorg. Chem. XXXX, XXX, XXX−XXX