Metal-metal interactions in dinuclear d8 metal cyanide complexes supported by phosphine ligands. Spectroscopic properties and ab initio calculations of [M2(μ-diphosphine)2(CN)4] and trans-[M(phosphine)2(CN)2] (M = Pt, Ni)

Article abstract of DOI:10.1021/ic020077j

Structural, spectroscopic properties on the dinuclear [M₂(dcpm)₂(CN)₄] (M = Pt, 1a; Ni, 2a, dcpm = bis(dicyclohexylphosphino)methane) and [M₂(dmpm)₂(CN)₄] (M = Pt, 1b; Ni, 2b, dmpm = bis(dimethylphosphino)-methane) and the mononuclear trans-[M(PCy₃)₂(CN)₂] (M = Pt, 3; Ni, 4, PCy₃ = tricyclohexylphosphine) and theoretical investigations on the corresponding model compounds are described. X-ray structural analyses reveal Pt···Pt and Ni···Ni distances of 3.0565(4)/3.189(1) A? and 2.957(1)/3.209(8) A? for 1a/1b and 2a/2b, respectively. The UV-vis absorption bands at 337 nm (ε 2.41 × 10⁴ dm³ mol^-1 cm^-1) for 1a and 328 nm (ε 2.43 × 10⁴ dm³ mol^-1 cm^-1) for 1b in CH₂Cl₂ are assigned to ¹(5d_σ* → 6p_σ) electronic transitions originating from Pt(II)-Pt(II) interactions. Resonance Raman spectroscopy of 1a, in which all the Raman intensity appears in the Pt-Pt stretch fundamental (93 cm^-1) and overtone bands, verifies this metal-metal interaction. Complexes 1a and 1b exhibit photoluminescence in the solid state and solution. For the dinuclear nickel(II) complexes 2a and 2b, neither spectroscopic data nor theoretical calculation suggests the presence of Ni(II)-Ni(II) interactions. The intense absorption bands at λ > 320 nm in the UV-vis spectra of 2a and 2b are tentatively assigned to d → d transitions.

Full text of DOI:10.1021/ic020077j

Inorg. Chem. 2002, 41, 3866−3875
8
Metal−Metal Interactions in Dinuclear d Metal Cyanide Complexes
Supported by Phosphine Ligands. Spectroscopic Properties and ab
Initio Calculations of [M (µ-diphosphine) (CN) ] and
2
2
4
trans-[M(phosphine) (CN) ] (M) Pt, Ni)
2
2
†
Bao-Hui Xia, Chi-Ming Che,* David Lee Phillips, King-Hung Leung, and Kung-Kai Cheung
Department of Chemistry and the HKU-CAS Joint Laboratory on New Materials,
The UniVersity of Hong Kong, Pokfulam Road, Hong Kong SAR, P.R. China
Received January 25, 2002
Structural, spectroscopic properties on the dinuclear [M
bis(dicyclohexylphosphino)methane) and [M (dmpm) (CN) ] (M) Pt, 1b; Ni, 2b, dmpm ) bis(dimethylphosphino)-
methane) and the mononuclear trans-[M(PCy (CN) ] (M ) Pt, 3; Ni, 4, PCy ) tricyclohexylphosphine) and
2 2 4
(dcpm) (CN) ] (M ) Pt, 1a; Ni, 2a, dcpm )
2
2
4
3
)
2
2
3
theoretical investigations on the corresponding model compounds are described. X-ray structural analyses reveal
Pt‚‚‚Pt and Ni‚‚‚Ni distances of 3.0565(4)/3.189(1) Å and 2.957(1)/3.209(8) Å for 1a/1b and 2a/2b, respectively.
4
3
-1
-1
4
3
The UV−vis absorption bands at 337 nm (ꢀ 2.41 × 10 dm mol cm ) for 1a and 328 nm (ꢀ 2.43 × 10 dm
mol- cm ) for 1b in CH
1
-1
1
2
2 σ σ
Cl are assigned to (5d * f 6p ) electronic transitions originating from Pt(II)−Pt(II)
interactions. Resonance Raman spectroscopy of 1a, in which all the Raman intensity appears in the Pt−Pt stretch
-
1
fundamental (93 cm ) and overtone bands, verifies this metal−metal interaction. Complexes 1a and 1b exhibit
photoluminescence in the solid state and solution. For the dinuclear nickel(II) complexes 2a and 2b, neither
spectroscopic data nor theoretical calculation suggests the presence of Ni(II)−Ni(II) interactions. The intense absorption
bands at λ > 320 nm in the UV−vis spectra of 2a and 2b are tentatively assigned to d f d transitions.
Introduction
1,3[(nd
in the nd
σ
σ
*)(n + 1)p ] excited states produced upon excitation
6
6b,c,e
σ σ
* f (n + 1)p transition. Experimental studies
Since the initial report on the fascinating photophysical
and photochemical properties of the dinuclear d -d complex
8
8
8
8
(
2) d -d : (a) Stiegman, A. E.; Rice, S. F.; Gray, H. B.; Miskowski, V.
M. Inorg. Chem. 1987, 26, 1112-1116. (b) Che, C.-M.; Yam, V. W.-
W.; Wong, W.-T.; Lai, T.-F. Inorg. Chem. 1989, 28, 2908-2910. (c)
Yip, H.-K.; Lai, T.-F.; Che, C.-M. J. Chem. Soc., Dalton Trans. 1991,
1639-1641. (d) Bailey, J. A.; Miskowski, V. M.; Gray, H. B. Inorg.
Chem. 1993, 32, 369-370. (e) Yip, H.-K.; Cheng, L.-K.; Cheung,
K.-K.; Che, C.-M. J. Chem. Soc., Dalton Trans. 1993, 2933-2938.
4
- 1
[Pt
2 2 5 2 4
(P O H ) ] , there has been continued interest in
luminescent dinuclear metal complexes, particularly those
8
10
2
8
8
with d and d metal ions. To explain the d -d metal
interaction, Gray and co-workers first presented a qualitative
molecular orbital model (see Figure S1 in the Supporting
(
f) Navarro, J. A. R.; Romero, M. A.; Salas, J. M.; Quir o´ s, M.;
Information), in which the higher lying ndz2 and (n + 1)p
3
Bahraoui, J. E.; Molina, J. Inorg. Chem. 1996, 35, 7829-7835. (g)
Lai, S.-W.; Chan, M. C.-W.; Cheung, T.-C.; Peng, S.-M.; Che, C.-M.
z
orbitals combine to yield the nd
σ
/d
σ
* and (n + 1)p
σ σ
/p * MOs
10
10
Inorg. Chem. 1999, 38, 4046-4055. d -d : (h) King, C.; Wang,
J.-C.; Khan, M. N. I.; Fackler, J. P., Jr. Inorg. Chem. 1989, 28, 2145-
(MO ) molecular orbital). Many spectroscopic studies on
2
149. (i) Fu, W.-F.; Chan, K.-C.; Miskowski, V. M.; Che, C.-M.
dinuclear Rh(I), Ir(I), and Pt(II) complexes have revealed
weak metal-metal bonding interactions in the ground state.
This results from mixing of MOs with same symmetry
between the nd
scopic and photophysical studies on dinuclear d systems
have revealed that there is a net metal-metal single bond in
Angew. Chem. 1999, 111, 2953-2955; Angew. Chem., Int. Ed. 1999,
4
3
8, 2783-2785. (j) Che, C.-M.; Mao, Z.; Miskowski, V. M.; Tse,
M.-C.; Chan, C.-K.; Cheung, K.-K.; Phillips, D. L.; Leung, K.-H.
Angew. Chem., Int. Ed. 2000, 39, 4084-4088.
5
σ
and (n + 1)p
σ
levels. Extensive spectro-
(
3) Mann, K. R.; Gordon, J. G., II; Gray, H. B. J. Am. Chem. Soc. 1975,
97, 3553-3555.
8
(
4) (a) Dallinger, R. F.; Miskowski, V. M.; Gray, H. B.; Woodruff, W.
H. J. Am. Chem. Soc. 1981, 103, 1595-1596. (b) Rice, S. F.; Gray,
H. B. J. Am. Chem. Soc. 1981, 103, 1593-1595. (c) Rice, S. F.; Milder,
S. J.; Gray, H. B. Coord. Chem. Rew. 1982, 43, 349-354. (d) Rice,
S. F.; Miskowski, V. M.; Gray, H. B. Inorg. Chem. 1988, 27, 4704-
4708.
*
To whom correspondence should be addressed. E-mail: cmche@hku.hk.
Correspondence on resonance Raman spectroscopy.
†
(1) Roundhill, D. M.; Gray, H. B.; Che, C.-M. Acc. Chem. Res. 1989, 22,
55-61.
(5) Smith, D. C.; Gray, H. B. Coord. Chem. ReV. 1990, 100, 169-181.
3866 Inorganic Chemistry, Vol. 41, No. 15, 2002
10.1021/ic020077j CCC: $22.00 © 2002 American Chemical Society
Published on Web 06/26/2002

8
Dinuclear d Metal Cyanide Complexes
and theoretical considerations have shown that the simplified
MO diagram (see Supporting Information) is inadequate and
cannot accurately describe the excited states. A more
satisfactory description was given by a valence bond treat-
ment in which the monomer excitations are involved in
regarding the presence of Ni(II)-Ni(II) interaction. Cotton
and co-workers had calculated the electronic structure of
11
2 4 2 4
Ni (HNCHNH) as a model for Ni (form) . Their treatment
indicated that there is no net metal-metal bonding interaction
at the “zeroth order” level calculation, despite the rather short
Ni‚‚‚Ni distances. Herein, we describe the structural and
spectroscopic properties of dinuclear and mononuclear plati-
num(II) and nickel(II) cyanide complexes supported by the
electron-rich phosphine ligands dmpm (bis(dimethylphos-
phino)methane), dcpm (bis(dicyclohexylphosphino)methane),
6b
excitonic dipole-dipole coupling.
Some theoretical studies on d -d metal-metal interac-
8
8
tions have previously been undertaken. Alvarez and co-
7
workers performed ab initio calculations on the nature of
8
weak M‚‚‚M contacts between d square-planar ML
4
mol-
ecules and predicted the effect of adding axial groups (along
the M-M orientation) on the strength of the M-M interac-
tion. These workers also probed the effect of pyramidality
on d -ML complexes and proposed a correlation between
4
M‚‚‚M distances and the degree of pyramidalization at the
metal atom.
and PCy
ab initio calculations on the models [M
(CN) ] and trans-[M(PMe (CN) ] (M ) Pt or Ni). We
conclude that there is metal-metal interaction in the di-
nuclear [Pt (µ-diphosphine) (CN) ] complexes. For the
[Ni (µ-phosphine) (CN) ] congeners, both experimental data
3
(tricyclohexylphosphine). We have also performed
2
(µ-H PCH PH
2
2
2 2
) -
4
3
)
2
2
8
8
2
2
4
2
2
4
Metal-metal interactions in oligomeric square-planar
Pt(II) complexes have been extensively studied including
and theoretical calculations provide no evidence for Ni-Ni
interaction in the ground state.
2-
9
those of the [Pt(CN)
complex [Pt (dppm)
4
]
chain compounds. The dinuclear
2
b
(CN)
4
]
(1c, dppm ) bis(diphen-
Experimental Section
2
2
ylphosphino)methane) was previously synthesized by Che
and co-workers to probe the interaction between Pt-
3
General. The phosphine ligands dcpm, dmpm, and PCy were
purchased from Strem. All the chemicals and solvents (AR grade)
for synthesis were used as received. The solvents for photophysical
2 2
(phosphine) (CN) moieties. Compared to the numerous
spectroscopic works on Pt(II) complexes, less attention has
been paid to dinuclear Ni(II) systems and most of the studies
have been confined to structural reports. These studies have
revealed short intramolecular Ni‚‚‚Ni distances for some
polynuclear Ni(II) complexes. For examples, the Ni‚‚‚Ni
2g
studies were purified as previously described. The resonance
12
Raman apparatus and methods have previously been described.
[Pt (dcpm) (CN) ] (1a) and [Pt (dmpm) (CN) ] (1b) were synthe-
2
2
4
2
2
4
sized by a modified literature method.13 The starting materials
Pt(L)Cl and Ni(L)Cl (L ) dcpm or dmpm) were prepared
following procedures similar to that for Pt(dppm)Cl
Ni(dppm)Cl2,14 respectively. For the preparation of trans-
[M(PCy (CN) ] (M ) Pt, 3; Ni, 4), the starting materials
Pt(PCy Cl and Ni(PCy Cl were prepared by reacting PCy with
Pt(COD)Cl and NiCl ‚6H
2
2
1
3
2
and
distances in [Ni
2
(form)
4
] (form ) N,N′-di-p-tolylform-
10a
amidinato), [Ni
X ) Cl , CN , N
5
(µ -tpda)
5
4
X
2
] (H
2
10b
-tpda ) tripyridyldiamine,
-
-
-
-
3
)
2
2
3
, or NCS ), and Ni
7
7 4 2
(µ -teptra) Cl
-teptra ) tetrapyridyltriamine)1 are 2.485(2), ca.
.30-2.40, and 2.215-2.383(2) Å, respectively. These
relatively short intermetal distances have stimulated interest
0c
3
)
2
2
3
)
2
2
3
(H
3
2
1
2
2
2
O, respectively, and were isolated. H
and 31P NMR spectra were recorded on a Bruker DPX-500
multinuclear FT-NMR spectrometer. Chemical shifts (δ, ppm) were
1
reported relative to tetramethylsilane ( H NMR) and 85% H
PO
3 4
(
6) (a) Che, C.-M.; Butler, L. G.; Gray, H. B.; Crooks, R. M.; Woodruff,
W. H. J. Am. Chem. Soc. 1983, 105, 5492-5494. (b) Smith, D. C.;
Miskowski, V. M.; Mason, W. R.; Gray, H. B. J. Am. Chem. Soc.
3
1
( P NMR). Infrared spectra were obtained on a Bio-Rad FTS-165
spectrometer. UV-vis spectra were obtained on a Hewlett-Packard
1990, 112, 3759-3767. (c) Marshall, J. L.; Hopkins, M. D.; Mis-
8
453 diode array spectrophotometer. Positive ion FAB mass spectra
were recorded on a Finnigan MAT95 mass spectrometer.
Pt (dcpm) (CN) ], 1a. A methanolic solution of NaCN (0.03
g, 0.6 mmol) was dropwise added to a methanolic (40 mL)
suspension of Pt(dcpm)Cl (0.20 g, 0.3 mmol). After 3 h a colorless
solution was obtained. Removal of the solvent gave a white solid.
After extraction with CH Cl , the filtrate was reduced to 5 mL.
kowski, V. M.; Gray, H. B. Inorg. Chem. 1992, 31, 5034-5040. (d)
Miskowski, V. M.; Rice, S. F.; Gray, H. B.; Milder, S. J. J. Phys.
Chem. 1993, 97, 4277-4283. (e) Miskowski, V. M.; Rice, S. F.; Gray,
H. B.; Dallinger, R. F.; Milder, S. J.; Hill, M. G.; Exstrom, C. L.;
Mann, K. R. Inorg. Chem. 1994, 33, 2799-2807. (f) Dallinger, R. F.;
Carlson, M. J.; Miskowski, V. M.; Gray, H. B. Inorg. Chem. 1998,
[
2
2
4
2
3
7, 5011-5013. (g) Leung, K. H.; Phillips, D. L.; Che, C.-M.;
Miskowski, V. M. J. Raman Spectrosc. 1999, 30, 987-993.
2
2
(
(
(
7) Novoa, J. J.; Aull o´ n, G.; Alemany, P.; Alvarez, S. J. Am. Chem. Soc.
1
995, 117, 7169-7171.
8) Aull o´ n, G.; Alemany, P.; Alvarez, S. Inorg. Chem. 1996, 35, 5061-
067.
9) (a) Miller, J. S.; Epstein, A. J. Prog. Inorg. Chem. 1976, 20, 1-151.
b) Miller, J. S. Extended Linear Chain Compounds; Plenum: New
Diethyl ether was added to give a white precipitate. This was filtered
and washed with MeOH and diethyl ether. Crystals suitable for
X-ray crystal analysis were obtained by slow diffusion of diethyl
5
1
ether into dichloromethane solution. Yield: 0.20 g, 85%. H NMR
(
York, 1982; vols. 1-3. (c) Schindler, J. W.; Fukuda, R. C.; Adamson,
A. W. J. Am. Chem. Soc. 1982, 104, 3596-3600. (d) Gliemann, G.;
Yersin, H. Struct. Bonding 1985, 62, 87-153. (e) Lechner, A.;
Gliemann, G. J. Am. Chem. Soc. 1989, 111, 7469-7475. (f) Mis-
kowski, V. M.; Houlding, V. H. Inorg. Chem. 1991, 30, 4446-4452.
(500 MHz, CDCl
): δ 3.03 (t, 4 H; CH
), 1.25-2.93 (m, 88 H;
): δ 13.75 (t, J(PtP) )
3
2
31
1
1
6
C H11). P{ H} NMR (500 MHz, CDCl
3
-
1
2
313.6 Hz). IR (KBr, cm ): ν 2125 (CtN). MS (+FAB): m/z
+
+
1310 [M] , 1284 [M - CN] . Anal. Calcd for C54
92 4 4 2
H N P Pt : C,
(
g) Houlding, V. H.; Miskowski, V. M. Coord. Chem. ReV. 1991, 111,
49.44; H, 7.07; N, 4.27. Found: C, 49.01; H, 7.42; N, 4.00.
1
45-152. (h) Miskowski, V. M.; Houlding, V. H.; Che, C.-M.; Wang,
Y. Inorg. Chem. 1993, 32, 2518-2524. (i) Bailey, J. A.; Hill, M. G.;
Marsh, R. E.; Miskowski, V. M.; Schaefer, W. P.; Gray, H. B. Inorg.
Chem. 1995, 34, 4591-4599.
(11) Cotton, F. A.; Matusz, M.; Poli, R.; Feng, X. J. J. Am. Chem. Soc.
1988, 110, 1144-1154.
(
10) (a) Cotton, F. A.; Matusz, M.; Poli, R. Inorg. Chem. 1987, 26, 1472-
(12) Zheng, X.; Phillips, D. L. J. Chem. Phys. 1998, 108, 5772-5783.
(13) Hassan, F. S. M.; Markham, D. P.; Pringle, P. G.; Shaw, B. L. J.
Chem. Soc., Dalton Trans. 1985, 279-283.
1474. (b) Wang, C.-C.; Lo, W.-C.; Chou, C.-C.; Lee, G.-H.; Chen,
J.-M.; Peng, S.-M. Inorg. Chem. 1998, 37, 4059-4065. (c) Lai, S.-
Y.; Lin, T.-W.; Chen, Y.-H.; Wang, C.-C.; Lee, G.-H.; Yang, M.-H.;
Leung, M.-K.; Peng, S.-M. J. Am. Chem. Soc. 1999, 121, 250-251.
(14) Van Hecke, G. R.; Horrocks, W. D., Jr. Inorg. Chem. 1966, 5, 1968-
1974.
Inorganic Chemistry, Vol. 41, No. 15, 2002 3867

Xia et al.
Table 1. Crystallographic Data and Structure Refinement for Complexes
1a‚2.5 H2O
1b‚H2O
2a‚2.5 H2O
2b
3
4
formula
fw
cryst syst
space group
a (Å)
C54H92N4P4Pt2
1310.56
monoclinic
P2/a (No. 13)
18.097(3)
14.759(2)
23.701(3)
90.0
99.87(2)
90.0
6263(1)
2740
C14H28N4P4Pt2
766.06
C54H92N4P4Ni2
1036.50
monoclinic
P2/a (No. 13)
17.999(3)
14.513(3)
23.895(1)
90.0
99.298(7)
90.0
6160(1)
2340
C14H28N4P4Ni2
493.70
monoclinic
P21/n
9.160(1)
11.472(1)
10.238(1)
90.0
90.427(2)
90.0
C38H66N2P2Pt
807.99
C38H66N2P2Ni
671.60
triclinic
P 1h (No. 2)
8.757(2)
9.149(2)
9.149(2)
79.58(3)
66.63(3)
66.63(3)
617.4(2)
366
triclinic
P 1h (No. 2)
9.927(2)
10.274(2)
11.050(2)
65.23(2)
69.80(2)
89.72(2)
947.4(4)
416
triclinic
P 1h (No. 2)
9.972(2)
10.289(2)
10.908(2)
66.00(2)
68.91(2)
89.64(2)
941.0(4)
366
b (Å)
c (Å)
R (deg)
â (deg)
γ (deg)
3
V (Å )
1075.8(2)
512
20.52
F(000)
µ (cm-1)
46.05
1.445
115.85
2.110
7.55
1.168
38.00
1.416
6.28
1.185
Dc (g cm-3)
1.524
2
θmax (deg)
51.1
11976
8753
597
0.042, 0.066
1.76
+1.18, -1.41
51.12
50.0
11334
5947
600
0.055, 0.085
2.19
+0.92, -0.55
55.06
53.2
3638
3638
196
0.032, 0.046
1.69
+0.69, -2.02
53.2
3435
2880
196
0.045, 0.063
1.68
+0.40, -0.53
reflns unique
no. of reflns with I >3σ(I)
no. of params
R, Rw
2117
2117
2473
a
2473a
119
0.05, 0.14
1.18
160
0.028, 0.087
0.73
GOF
residual F (e Å-3)
+1.394, -4.456
+0.461, -0.475
a
Reflections with I >2σ(I).
[
Pt
2
(dmpm)
2
(CN)
4
], 1b. NaCN (0.077 g, 1.6 mmol) in methanol
(0.27
CH
8.27.
trans-[Ni(PCy
purified by the same method as that for 3. Crystals were obtained
by slow diffusion of diethyl ether into CH Cl solution. Yield: 87%.
): δ 1.2-2.6 (m). P{ H} NMR (500
3
OH: C, 55.75; N, 3.34; H, 8.40. Found: C, 55.96; N, 2.91; H,
(
20 mL) was dropwise added to a suspension of Pt(dmpm)Cl
g, 0.67 mmol) in methanol (20 mL). After 24 h, removal of solvent
from the mixture gave a white solid. The solid was washed with
water and diethyl ether and then dried. Extraction with CHCl
removal of solvent on a rotary evaporator gave a white powder.
Crystals suitable for X-ray structure analysis were obtained by
diffusion of diethyl ether into methanol solution. Yield: 0.14 g,
4%. H NMR (500 MHz, CDCl
m, 24 H; CH
2
3 2 2
) (CN) ], 4. This complex was prepared and
3
and
2
2
1
31
1
H NMR (500 MHz, CDCl
MHz, CDCl ): δ 38.37 (s). IR (KBr, cm ): ν 2109 (CtN). FAB
MS: m/z: 672 [M] ; 645 [M - CN + 1] . Anal. Calcd for
38 66 2 2
C H N P Ni: C, 67.96; N, 4.17; H, 9.91. Found: C, 68.48; N,
3
-
1
3
+
+
1
5
(
1
3
): δ 3.23 (m, 4 H; CH
2
); 2.07
31
1
3
). P{ H} NMR (500 MHz, CDCl
3
): δ -18.28 (t,
4.19; H, 9.96.
-1
J(PtP) ) 2282.90 Hz). IR (KBr, cm ): ν 2139 (CtN). MS
Photophysical Measurements. Steady-state emission and ex-
citation spectra at room temperature and 77 K were measured on
a Spex 1681 Flurolog-2 model F111AI fluorescence spectropho-
tometer equipped with a Hamamatsu R928 PMT detector. The
spectra at 77 K in the solid state and in MeOH/EtOH (4:1 v/v)
were recorded for samples in quartz tubes immersed in a quartz-
walled optical Dewar flask filled with liquid nitrogen. For measure-
ments of solution emission spectra at room temperature, the
solutions were subjected to four freeze-pump-thaw cycles prior
to the measurements. Emission quantum yields were determined
+
+
(
C
+FAB): m/z 767 [M + 1], 740 [M - CN] . Anal. Calcd for
Pt ‚CHCl : C, 20.34; N, 6.32; H, 3.30. Found: C, 20.04;
N, 6.53; H, 3.34.
Ni (dcpm) (CN)
purified in the same manner as that for 1a. Crystals were obtained
14
H
28
N P
4 4
2
3
[
2
2
4
], 2a. This compound was prepared and
1
by evaporation of a CH
500 MHz, CDCl
2
Cl
2
/hexane solution. Yield: 86%. H NMR
), 1.25-2.52 (m, 88 H;
(
C
3
): δ 2.61 (t, 4 H; CH
2
31
1
6
H
11). P{ H} NMR (500 MHz, CDCl
1
3
): δ 32.87 (s). IR (KBr,
-
+
cm ): ν 2110 (CtN). MS (+FAB): m/z 1038 [M ], 1011 [M -
+
15
CN + 1] . Anal. Calcd for C54
92 4 4 2 2
H N P Ni ‚H O: C, 61.38; N, 5.30;
by the method of Demas and Crosby with quinine sulfate in 1.0
H, 8.97. Found: C, 61.65; N, 4.95; H, 9.40.
Ni (dmpm) (CN) ], 2b. This complex was prepared and purified
r
N sulfuric acid as the standard (Φ ) 0.546).
X-ray Crystallography. Crystal data and details of collection
[
2
2
4
similarly to1b. Crystals were obtained from slow evaporation of a
methanol solution of the complex. Yield: 70%. H NMR (500 MHz,
and refinement are summarized in Table 1.
1
Graphite-monochromatized Mo KR radiation (λ ) 0.71073 Å)
was used, and all the non-hydrogen atoms were refined anistropi-
cally, except for the water molecules of 1a‚2.5H O, 1b‚H O, and
3
1
1
CDCl
(
3
): δ 2.82 (p, 4 H; CH
2
), 1.81 (t, 24 H; CH
3
). P{ H} NMR
-
1
500 MHz, CDCl3:): δ 2.89 (s). IR (KBr, cm ): ν 2119 (CtN).
MS(+FAB): m/z 493 [M ], 466 [M - CN + 1] . Anal. Calcd for
2
2
+
+
2a‚2.5H O. All the hydrogen atoms were placed at calculation
2
C
1
14
H
28
N
4
P
4
Ni
1.36; H, 5.77.
trans-[Pt(PCy
and NaCN (0.018 g, 0.4 mmol) in MeOH (20 mL) were stirred for
h to give a white suspension. Removal of solvent produced a
white solid. The crude solid was extracted with CH Cl , and the
CH Cl extract was concentrated to 5 mL. Addition of diethyl ether
gave a white solid, which was washed with MeOH and diethyl
ether. Crystals were obtained by slow evaporation of a CH Cl
hexane solution. Yield: 0.11 g, 87%. H NMR (500 MHz,
2
: C, 34.06; N, 11.35; H, 5.72. Found: C, 33.73; N,
positions without refinement.
For crystals 1a‚2.5H O (0.15 × 0.15 × 0.05 mm), 1b‚H O (0.30
2
2
3
)
2
(CN)
2
], 3. Pt(PCy
3
)
2
Cl
2
(0.130 g, 0.16 mmol)
× 0.2 × 0.1), 2a‚2.5H O (0.30 × 0.25 × 0.05 mm), 3 (0.30 ×
2
0.15 × 0.10 mm), and 4 (0.20 × 0.15 × 0.05 mm), the crystal
4
data were collected at 301 K on a MAR diffractometer with a 300
mm image plate detector. The images were interpreted and
intensities integrated using the program DENZO.16 For 1a, 2a, 3,
and 4, their structures were solved by Patterson methods and
2
2
2
2
2
2
/
1
(15) Demas, J. N.; Crosby, G. A. J. Phys. Chem. 1971, 75, 991-1024.
(16) DENZO: In The HKL ManualsA description of programs DENZO,
XDISPLAYF, and SCALEPACK, written by Gewirth, D., with the
cooperation of the program authors Otwinowski, Z., and Minor, W.,
1995, Yale University, New Haven, CT.
31
1
CDCl
3
): δ 1.19-2.75 (m). P{ H} NMR (500 MHz, CDCl
3
): δ
1
-1
28.56 (t, J(PtP) ) 2193.0 Hz). IR (KBr, cm ): ν 2124 (CtN).
+
MS(+FAB): m/z 807 [M + 1]. Anal. Calcd for C38
66 2 2
H N P Pt‚
3868 Inorganic Chemistry, Vol. 41, No. 15, 2002

8
Dinuclear d Metal Cyanide Complexes
expanded by Fourier methods (PATTY 17) and refined by full-matrix
least-squares using the software package TeXsan18 on a Silicon
Graphics Indy computer. The crystal data of 2b (0.20 × 0.12 ×
Table 2. Selected Bond Distances (Å) and Angles (deg)
1
2.340(2)
2.329(2)
1.16(1)
a‚2.5H2O
Pt(1)-P(1)
Pt(2)-P(3)
N(1)-C(1)
Pt(1)‚‚‚Pt(2)
Pt(1)-C(1)
Pt(2)-C(4)
N(3)-C(3)
2.003(8)
1.985(9)
1.10(1)
0.10 mm) were collected at 294 K on a Bruker SMART CCD
diffractometer. The structures of 1b and 2b were determined by
3.0565(4)
2
direct methods and refined by full-matrix least-squares on F
employing the SHELXL97 program19 on a PC.
Pt(1)‚‚‚Pt(2)-P(3)
Pt(2)‚‚‚Pt(1)-P(1)
P(1)-Pt(1)-P(2)
P(1)-Pt(1)-C(1)
C(1)-Pt(1)-C(2)
88.84(5)
91.57(5)
177.25(7)
88.8(2)
Pt(2)‚‚‚Pt(1)-C(1)
Pt(1)‚‚‚Pt(2)-C(3)
P(3)-Pt(2)-P(4)
P(3)-Pt(2)-C(4)
C(3)-Pt(2)-C(4)
87.6(2)
96.5(3)
177.34(8)
94.7(3)
Computational Details. Previous theoretical work on a dinuclear
gold complex20 showed that the ab initio MP2 (second-order
Møller-Plesset) calculations gave a good representation of the
ground-state electronic structure. Here, we also performed ab initio
theoretical calculation at the MP2 level on the model molecules.
All the calculations were performed using the Gaussian 94 program
package21 on a Silicon Graphics Origin2000 machine. For the sake
of simplicity, the ancillary cyclohexyl groups of the phosphine
ligands were substituted by hydrogen and methyl for the dinuclear
177.4(3)
175.8(4)
1
2.313(5)
2.02(3)
1.17(4)
3.189(1)
b‚H2Oa,b
Pt(1)-P(1)
Pt(1)-C(1)
N(1)-C(1)
Pt(1)‚‚‚Pt(1)*
Pt(1)-P(2)
Pt(1)-C(2)
N(2)-C(2)
2.310(5)
1.98(2)
1.20(4)
Pt(1)*‚‚‚Pt(1)-P(1)
C(1)-Pt(1)-P(1)
N(1)-C(1)-Pt(1)
C(2)-Pt(1)-C(1)
89.8(1)
87.8(8)
169(3)
Pt(1)*‚‚‚Pt(1)-P(2)
C(2)-Pt(1)-P(2)
N(2)-C(2)-Pt(1)
P(1)-Pt(1)-P(2)
88.7(1)
87.5(7)
178(3)
and mononuclear models, respectively. [Ni
was set to have a C symmetry just like the case of the crystalline
form of 2b; [Pt (µ-H PCH PH (CN) ] was set to have a C
symmetry (the crystal lattice of 1a displays pseudo-C symmetry
2 2 2 2 2 4
(µ-H PCH PH ) (CN) ]
i
175.6(7)
178.3(1)
2
2
2
)
2 2
4
2
2
2
2.241(2)
1.867(8)
1.147(9)
2.957(1)
a‚2.5H2O
for the repeating unit). The two metal centers were placed on the
Z axis, and the four phosphorus atoms were put in the YZ plane
and the four CN groups in the XZ plane. The molecule trans-
Ni(1)-P(1)
Ni(1)-C(1)
N(1)-C(1)
Ni(1)‚‚‚Ni(2)
Ni(2)-P(3)
Ni(2)-C(3)
N(3)-C(3)
2.231(2)
1.860(9)
1.151(9)
3 2 2
[M(PMe ) (CN) ] (M ) Pt or Ni) was put in the XY plane with the
phosphorus atoms on the X axis and CN on the Y axis; the molecule
is of C2h symmetry. We employed the quasi-relativistic pseudopo-
tentials for Pt, Ni, and P by Hay and Wadt, with 19, 19, and 5
Ni(2)‚‚‚Ni(1)-P(1)
Ni(1)‚‚‚Ni(2)-P(3)
P(1)-Ni(1)-P(2)
P(1)-Ni(1)-C(1)
C(1)-Ni(1)-C(2)
92.39(5)
89.50(5)
174.39(7)
89.9(2)
Ni(2)‚‚‚Ni(1)-C(1)
Ni(1)‚‚‚Ni(2)-C(3)
P(3)-Ni(2)-P(4)
P(3)-Ni(2)-C(3)
C(3)-Ni(2)-C(4)
97.2(2)
86.8(2)
178.58(8)
85.5(2)
22
valence electrons, respectively; the double-ú LANL2DZ basis sets
177.8(3)
173.9(3)
associated with the pseudopotentials were adopted. For [M
2
(µ-
2
bb
H
2
PCH PH (CN) ] (M ) Pt or Ni), the LANL2DZ basis sets were
2
2
)
2
4
Ni(1)-P(1)
Ni(1)-C(1)
P(1)-C(3)
N(1)-C(1)
Ni(1)‚‚‚Ni(2)
2.1945(4)
1.870(1)
1.828(1)
1.143(2)
3.209(8)
Ni(1)-P(2)
Ni(1)-C(2)
P(2)-C(3A)
N(2)-C(2)
2.2023(4)
1.860(1)
1.836(1)
1.146(2)
adopted for C, N, and H atoms; while for the monomer systems
the standard 6-31g* basis sets were adopted for C, N, and H atoms,
and a d-polarization function was added to the P atoms (R )
d
2
3
0
.334 ). The basis sets were taken as Pt(8s6p3d)/[3s3p2d], Ni-
8s6p3d)/[3s3p2d], P(3s3p)/[2s2p], C(10s5p)/[3s2p], N(10s5p)/
3s2p], and H(4s)/[2s] for [M (µ-H PCH PH (CN) ] (M ) Pt or
(CN) ] (M ) Pt or Ni), we adopted
(
[
Ni(1A)‚‚‚Ni(1)-P(1)
C(1)-Ni(1)-P(1)
C(1)-Ni(1)-P(2)
N(1)-C(1)-Ni(1)
C(2)-Ni(1)-C(1)
86.1
Ni(1A)‚‚‚Ni(1)-P(2)
92.3
2
2
2
2
)
2
4
90.50(4) C(2)-Ni(1)-P(1)
91.29(4) C(2)-Ni(1)-P(2)
178.40(1) N(2)-C(2)-Ni(1)
170.47(6) P(1)-Ni(1)-P(2)
89.05(4)
89.29(4)
173.4(1)
178.09(1)
Ni); for trans-[M(PMe
)
3 2
2
P(3s3p1d)/[2s2p1d], C(10s4p1d)/[3s2p1d], N(10s4p1d)/[3s2p1d],
and H(4s)/[2s]. Thus, the calculations employed 190 basis func-
a
tions and 132 electrons for [M
Ni) and 224 basis functions and 108 electrons for trans-[M(PMe
CN) ] (M ) Pt, Ni). The molecular structures were partially
optimized.
(µ-H PCH PH
2 2 2 2
)
2
(CN)
4
] (M ) Pt,
This molecule is disordered against the Pt-Pt orientation, i.e., rotated
along the bond ca. 51°, only one set of possible structural data is described
here. b This molecule has an inversion center.
3 2
) -
(
2
Results and Discussion
(
17) PATTY: Beurskens, P. T.; Admiraal, G.; Beurskens, G.; Bosman, W.
P.; Garcia-Granda, S.; Gould, R. O.; Smits, J. M. M.; Smykalla, C.
The DIRDIF Program System, Technical Report of the Crystallography
Laboratory; University of Nijmegen: Nijmegen, The Netherlands,
Syntheses. The diphosphine ligands dcpm and dmpm have
been selected for the syntheses of dinuclear Ni(II) and
Pt(II) complexes. Unlike the widely used dppm ligand, the
dcpm and dmpm ligands are optically transparent in the UV
region. Therefore, it is feasible to probe for metal-metal
bonded electronic transitions at shorter wavelengths than for
1992.
(18) TeXsan: Crystal Structure Analysis Package; Molecular Structure
Corporation: The Woodlands, TX, 1985 and 1992.
(
19) SHELX97: Sheldrick, G. M. Programs for Crystal Structure
Analysis, Release 97-2; University of G o¨ ttingen: G o¨ ttingen, Germany,
dppm complexes. We prepared [Pt
[Pt (dmpm) (CN) ] (1b) by slight modification of Shaw’s
method. [Ni (dcpm) (CN) ] (2a) and [Ni (dmpm) (CN)
2 2 4
(dcpm) (CN) ] (1a) and
1997.
(
(
20) Zhang, H.-X.; Che, C.-M. Chem. Eur. J. 2001, 7, 4887-4893.
21) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Gill, P. M. W.; Johnson,
B. G.; Robb, M. A.; Cheeseman, J. R.; Keith, T.; Petersson, G. A.;
Montgomery, J. A.; Raghavachari, K.; Al-Laham, M. A.; Zakrzewski,
V. G.; Ortiz, J. V.; Foresman, J. B.; Cioslowski, J.; Stefanov, B. B.;
Nanayakkara, A.; Challacombe, M.; Peng, C. Y.; Ayala, P. Y.; Chen,
W.; Wong, M. W.; Andres, J. L.; Replogle, E. S.; Gomperts, R.;
Martin, R. L.; Fox, D. J.; Binkley, J. S.; Defrees, D. J.; Baker, J.;
Stewart, J. P.; Head-Gordon, M.; Gonzalez, C.; Pople, J. A. Gaussian
2
2
4
13
2
2
4
2
2
4
]
(2b) were similarly synthesized, which are rare examples of
face-to-face dinuclear Ni(II) complexes with bridging diphos-
phine ligands. To aid spectral assignment, we have also
prepared and studied the spectroscopic properties of the
monomeric congeners trans-[M(PCy
Ni, 4).
3 2 2
) (CN) ] (M ) Pt, 3;
94, Revision C.3; Gaussian, Inc.: Pittsburgh, PA, 1995.
(22) (a) Wadt, W. R.; Hay, P. J. J. Chem. Phys. 1985, 82, 284-298. (b)
Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 299-310.
X-ray Crystal Structures. Selected bond distances and
angles of dinuclear compounds are listed in Table 2. Because
(23) Check, C. E.; Faust, T. O.; Bailey, J. M.; Wright, B. J.; Gilbert, T.
M.; Sunderlin, L. S. J. Phys. Chem. A 2001, 105, 8111-8116.
Inorganic Chemistry, Vol. 41, No. 15, 2002 3869

Xia et al.
disfavors face-to-face arrangement of neighboring MP (CN)
2
2
moieties. All the P-M-P angles fall between 174.39(7)°
and 178.58(8)°. The C-M-C angles of 173.9(3)-177.8-
(3)° in 1a, 1b, and 2a are larger than that in 2b (170.47-
(
6)°). Indeed, the NiP (CN) units in 2b are significantly
2
2
distorted from square-planar geometry. The average Pt-C
distances of 2.008/2.00 Å and Pt-P distances of 2.334/2.312
Å in 1a/1b respectively are comparable to those values found
in 1c (Pt-C ) 2.003 Å, Pt-P ) 2.322 Å) and 3 (see above).
In 2a and 2b, the average Ni-C distances (1.873 and 1.865
Å, respectively) and the average Ni-P distances (2.236 and
2
.1984 Å, respectively) are comparable to the average Ni-C
2
4a
distances of 1.868 Å in [CdNi(CN)
4
(C
in cis-[Ni(tdppcyme)Cl
-1,3,5-(COOCH -C
The intriguing structural feature is the intramolecular
6 8 2 3
H N )(NH )] and
2
4b
Ni-P distances of 2.185 Å
tdppcyme ) cis,cis-1,3,5-(PPh
2
]
(
)
2 3
)
3 3
6 5
H ).
Figure 1. ORTEP drawing of complex 2a with thermal ellipsoids on the
0% probability level (hydrogen atoms are not shown) and the numbering
scheme.
M‚‚‚M distance in each structure. The Pt‚‚‚Pt distance in 1a
and 1b is 3.0565(4) and 3.189(1) Å, respectively, which are
relatively shorter than the value of 3.301(1) Å in 1c2b but
4
4- 25
2 2 5 2 4
slightly longer than that of 2.925(1) Å in [Pt (P O H ) ] .
9
a
Miller noted that for monomeric square-planar Pt(II)
complexes stacked to form a continuous chain of metal
atoms, the Pt‚‚‚Pt distances lie in the range 3.09-3.50 Å.
Hence, the Pt‚‚‚Pt distances in 1a and 1b suggest the presence
of weak Pt-Pt interactions.
The Ni‚‚‚Ni distances in 2a and 2b are 2.957(1) and 3.209-
(
8) Å, respectively. They are comparable to the Ni‚‚‚Ni
2
6a
separations of 3.207(6) Å in [Ni(dmgBF
2
)]
2
(dmgBF
2
)
26b
2-
2 3
difluoroborondimethylglyoximato) and 3.298 Å in [Ni L ]
(L ) biphenyl-2,2′-dithiolate). However, they are signifi-
cantly longer than the Ni‚‚‚Ni distances in some polynuclear
Ni(II) complexes, including those mentioned in the Introduc-
Figure 2. ORTEP drawing of complex 2b with thermal ellipsoids on the
3 4 2
tion as well as in the linear chain [Ni (bipyam-H) Cl
] unit2
7a
5
0% probability level (hydrogen atoms are not shown) and the numbering
scheme.
(2.443(1) and 2.431(1) Å, bipyam-H ) bis(2-pyridyl)-amide)
2
7b
and in [Ni
-tpda ) tripyridyldiamine).
It is noteworthy that, irrespective of the metal atom, the
5
(µ
5
-tpda)
4
Cl
2
]
(2.385(2) and 2.305(1) Å,
1
2
a/2a, 1b/2b are isostructural, only the ORTEP drawings of
a and 2b are shown (see Figures 1 and 2, respectively).
H
2
The other crystal structural data and ORTEP drawings are
given in the Supporting Information.
The structures of 3 and 4 are similar. Each metal atom
exhibits square-planar trans-[MP
coordination. The M-P and M-C distances in 3/4 are
M‚‚‚M distances of the dmpm-bridged complexes are longer
than those of the dcpm-bridged analogues, i.e., 1b > 1a and
2
8
2
b > 2a. According to previous work, the cone angles of
2 2
(CN) ] (P, phosphine ligand)
the diphosphine ligands are in the order dcpm > dppm >
dmpm. The dcpm ligand is more bulky than dmpm, which
results in twisting of the two MP (CN) units away from an
2 2
eclipsed configuration, and hence closer proximity for the
two metal atoms. The Pt‚‚‚Pt distance for 1c (with dppm
ligand) is the longest among the three complexes 1a, 1b,
2.336(1)/2.2426(6) Å and 2.021(5)/1.857(3) Å, respectively.
Complexes 1a, 1b, 2a, and 2b are structurally related to
[Pt (dppm) (CN) ] (1c) and [Pd (dppm) (CN) ], while 1b
2 2 4 2 2 4
and 2b contain a crystallographically imposed center of
symmetry. In these four complexes, each metal atom exhibits
square-planar geometry with two trans-cyano groups and two
trans-P atoms from bridging diphosphine ligands. The two
2b
2c
(
24) (a) H o¨ kelek, T.; U¨ lk u¨ , D. Acta Crystallogr., Sect. C 1988, 44, 832-
8
34. (b) Heins, M.; St o¨ ssel, P.; Mayer, H. A.; Fawzi, R.; Steimann,
MP
2
(CN)
2
units in all of the dimers are in normal face-to-
(CN) units in 1a
M. J. Organomet. Chem. 1999, 587, 258-266.
(25) Dos Remedios Pinto, M. A. F.; Sadler, P. J.; Neidle, S.; Sanderson,
face orientation. However, the two MP
and 2a are not eclipsed but skewed to each other with
dihedral angles P-M-M-P/C-M-M-C of ca. 23.1/28.1
2
2
M. R.; Subbiah, A.; Kuroda, R. J. Chem. Soc., Chem. Commun. 1980,
13-15.
(26) (a) Stephens, F. S.; Vagg, R. S. Acta Crystallogr., Sect. B 1977, 33,
3
1
159-3164. (b) Erkizia, E.; Conry, R. R. Inorg. Chem. 2000, 39,
674-1679.
(1a) and 21.99°/27.6° (2a), respectively, which are different
from that in 1b, 2b, and other [M (dppm) (CN) ] (M ) Pt
2
2
4
(
27) (a) Aduldecha, S.; Hathaway, B. J. Chem. Soc., Dalton Trans. 1991,
993-998. (b) Shieh, S.-J.; Chou, C.-C.; Lee, G.-H.; Wang, C.-C.; Peng,
S.-M. Angew. Chem. 1997, 109, 57-59; Angew. Chem., Int. Ed. Engl.
or Pd) analogues. Such a structural difference between the
dmpm-, dppm-, and dcpm-bridged metal complexes is
attributed to the steric bulkiness of the dcpm ligand which
1997, 36, 56-59.
(28) Tolman, C. A. Chem. ReV. 1977, 77, 313-348.
3870 Inorganic Chemistry, Vol. 41, No. 15, 2002

8
Dinuclear d Metal Cyanide Complexes
Table 3. Spectroscopic and Photophysical Properties of 1a-4
λabs/nm (ꢀ /dm mol cm-1
3
-1
)
λem/nm (τ/µs)a
complex
medium (T/K)
c
[
[
[
Pt2(dcpm)2(CN)4], 1a
CH2Cl2b
glass (77)
solid (298)
solid (77)
CH2Cl2
263 (8900), 337 (24100), 388 (300)
386, 448 (0.1)
d
c
380, 448 (2.6)
c
388, 450 (0.2)
c
388, 450 (3.0)
Pt2(dmpm)2(CN)4], 1b
Ni2(dcpm)2(CN)4], 2a
235 (6000), 256 (6000), 328 (24300), 375 (50)
nonemissive
438 (3.6)
444 (1.0)
d
glass (77)
solid (298)
solid (77)
CH2Cl2
MeOH
MeOH
CH2Cl2
CH2Cl2
442 (2.9)
251 (55400), 306 (5300), 361 (3700), 391 (3900)
252 (53000), 306 (6400), 358 (3900), 395 (4900)
240 (66200), 267 (6600), 298 (5300), 342 (4700), 380 (3600)
231 (6300), 251 (6600), 260 (6200), 273 (2900), 289 (1400)
264 (49600), 294 (8500), 312 (2900), 344 (700)
nonemissive
nonemissive
nonemissive
nonemissive
nonemissive
[Ni2(dmpm)2(CN)4], 2b
[Pt(PCy3)2(CN)2], 3
[Ni(PCy3)2(CN)2], 4
a
For emission spectra measurements excitation wavelength: 338 nm for 1a, 330 nm for 1b. For luminescence lifetime measurements: excitation at 355
nm. b The emission quantum yield is 1.68 × 10
-3
.
c
The lifetime for the higher-energy emissions cannot be measured due to the limitation (τ < 10 ns) of
d
the instrument. Solvent: EtOH/MeOH (1:4 v/v).
As emphasized by Mason,29,30c the group-theoretical clas-
sification of the excited states is largely independent of the
various assumptions about the nature of the terminal orbital.
2-n
The close similarity for the spectra of [Pt(CN)
n
(PR
3 4-n
) ]
(
3
n ) 0, 2), all of which show a series of UV bands between
3
3
-1
-1
00 and 220 nm with ꢀ 1-6 × 10 dm mol cm , suggests
(Pt) contribution. This is similar to the well-
(M) contribution in analogous
a dominant 6p
established dominant (n + 1)p
z
z
transitions of valence-isoelectronic Rh(I) and Ir(I) complexes.6b
However, the very complicated spectrum of the low-
symmetry compound 3 does not contribute to our under-
2
9
standing. We note that older literature assumes that the
π*(PR ) levels derive from 3d(P) orbitals. Theoretical
discussions including calculations performed by us (vide
infra) indicate that 3d(P) orbitals play no significant role in
bonding or in the composition of the low-energy states.
However, the σ*(P-R) orbitals lie at relatively low energy
Figure 3. UV-vis absorption spectra of 1a (- - -), 1b (‚‚‚), and 3 (s) in
CH2Cl2 at room temperature.
3
3
1
and 1c. It thus appears that both electronic and steric effects
of the bridging diphosphine auxiliaries affect the M‚‚‚M
distances.
3
1d
and can play a role as π*-acceptor orbitals.
Absorption Spectra and Photophysical Properties. (a)
Platinum(II) Complexes. Spectroscopic and photophysical
data are set out in Table 3. The electronic absorption spectra
of 1a, 1b, and 3 are shown in Figure 3. In dichloromethane,
The electronic spectra of dinuclear 1a and 1b are similar
to that of 3 for λ < 300 nm, but differ dramatically by
3
-1
exhibiting an intense band at 337 nm (ꢀ 24100 dm mol
-
1
cm ) for 1a and 328 nm (ꢀ 24300) for 1b. The bands also
show a weak, long-wavelength shoulder (1a 388 nm (ꢀ 300
the monomeric trans-[Pt(PCy
bands at 231 (ꢀ 6300 dm mol cm ), 251 (6600), 260
3
)
2
-1
(CN)
2
-1
] 3 shows absorption
3
3
-1
-1
dm mol cm ); 1b 375 nm (ꢀ 50)) that can be assigned to
spin-forbidden analogues of the intense band. These absorp-
tion bands are comparable to the spin-allowed and spin-
(6200), 273 (2900), and 289 nm (1400), which are very
similar to the spectral data reported for trans-[Pt(P(n-Bu ) -
3 2
29
(
CN)
2
] by Mason et al. These bands are clearly electrically
forbidden d
σ
* f p
σ
transitions of the [Pt
2
(dppm)
2
(CN)
4
2
]
dipole-allowed because their extinction coefficients are
b
3
3
-1
-1
analogue, which occurs at 324 and 390 nm, respectively.
greater than 10 dm mol cm (thus excluding assignments
to d-d transitions, which are Laporte forbidden). The spec-
troscopic assignments for complexes of this type, as well as
The inferred singlet-triplet excited-state splittings for 1a and
-
1
1
b are ∼4000 cm , which are comparable to the corre-
-
1
4- 1
2
-
sponding value of 4259 cm in [Pt (P ] .
2
2 5 2 4
O H )
of their homoleptic congeners [Pt(CN)
4
]
and [Pt(phos-
2+
The compounds also exhibit intense luminescence (Figure
4). Complex 1b only emits in rigid media (glassy solutions
or crystalline solid) with lifetimes of ∼1.0-4.0 µs for the
single ∼440 nm emission, indicating phosphorescence from
the triplet excited state. Complex 1a differs by emitting
weakly in fluid solution, and also exhibiting two emissions.
The 450 nm emission with lifetime in the microsecond region
phine)
ture.
4
9,30
] , have been extensively discussed in the litera-
There is general agreement that the transitions relate
2
to states of the type 1 {[5d(Pt)][6p
,3
-
z
(Pt),π*(L)]} (L ) CN ,
PR ), but the relative importance of the ligand contributions
3
to the terminal orbital is not well established in most cases.
(
29) Solar, J. M.; Ozkan, M. A.; Isci, H.; Mason, W. R. Inorg. Chem. 1984,
3, 758-764.
30) (a) Mason, W. R., III; Gray, H. B. J. Am. Chem. Soc. 1968, 90, 5721-
729. (b) Piepho, S. B.; Schatz, P. N.; McCaffery, A. J. J. Am. Chem.
2
(
(31) (a) Xiao, S.-X.; Trogler, W. C.; Ellis, D. E.; Berkovitch-Yellin, Z. J.
Am. Chem. Soc. 1983, 105, 7033-7037. (b) Marynick, D. S. J. Am.
Chem. Soc. 1984, 106, 4064-4065. (c) Pacchioni, G.; Bagus, P. S.
Inorg. Chem. 1992, 31, 4391-4398. (d) Gilheany, D. G. Chem. ReV.
1994, 94, 1339-1374.
5
Soc. 1969, 91, 5994-6001. (c) Isci, H.; Mason, W. R. Inorg. Chem.
1
1
975, 14, 905-912. (d) Cowman, C. D.; Gray, H. B. Inorg. Chem.
976, 15, 2823-2824.
Inorganic Chemistry, Vol. 41, No. 15, 2002 3871

Xia et al.
Figure 5. UV-vis absorption spectra of 2a (- - -) and 2b (‚‚‚) in
CH3OH and 4 (s) in CH2Cl2 at room temperature.
2
-
compounds9d and other dinuclear Pt(II)
chain [Pt(CN)]
4
2
d
complexes.
b) Ni(II) Complexes. The electronic absorption spectra
(
of 2a, 2b, and 4 are shown in Figure 5. No luminescence
was detected from these compounds.
The spectrum of 4 is very similar to that reported for trans-
2
9
[Ni(PEt
3 2 2
) (CN) ]. Like 3 and related Pt(II) complexes, 4
displays a complicated spectrum consisting of a series of
3
3
-1
-1
absorption bands with ꢀ values around 10 dm mol cm
at λ e320 nm. There is a clear difference, however, in that
4
3
-1
-1
also displays a very intense (ꢀ 49600 dm mol cm )
band at 264 nm. For the dinuclear 2a and 2b derivatives,
-
1
Figure 4. (a) The excitation (- - -) and emission spectra (‚‚‚‚‚‚) in the
glass state (MeOH/EtOH 4:1 v/v) at 77 K and the emission spectrum in
CH2Cl2 (s) at room temperature of 1a. (b) The excitation (s) and emission
this band is blue-shifted by 1900 and 3800 cm , respec-
tively. There is certainly no correlation with Ni‚‚‚Ni dis-
tances. However, there is a correlation with Ni-P distances,
whose average values are 2.243 Å (4) > 2.236 Å (2a) >
(- - -) spectra in the glass state (MeOH/EtOH 4:1 v/v) at 77 K and emission
spectrum in the solid state at 77 K (-‚-‚) and room temperature (‚‚‚) of 1b.
2
.198 Å (2b). Thus, the transition energy decreases as the
is attributable to phosphorescence, while the shorter wave-
length band at ∼386 nm with lifetime <10 ns is assigned to
fluorescence from the singlet excited state.
Ni-P distance increases.
Comparisons to the spectra of homoleptic complexes are
2-
informative. The spectrum of the complex [Ni(CN)
not exhibit an analogous band; it is, in fact, very similar to
the spectrum of [Pt(CN) ] . The spectra of Ni(II) phosphine
4
complexes do display comparable bands; for example,
4
]
does
These photophysical properties are diagnostic for dinuclear
d8-d8 [d
1,3
1,3,4,6
The [ndz2(M) f
1,3
σ
* f p
σ
] excited states.
2-
(n + 1)p
z
(M)] transitions are specifically perturbed by
ground- and excited-state interactions; as a result, the
2+
[Ni(bis(diethylphosphino)ethane)
2
]
has a band at 252 nm
1
,3
] components of [ndz2(M) f (n +
1,3
[
nd
σ
* f (n + 1)p
σ
3
-1
-1 29
(ꢀ 34500 dm mol cm ). Clearly, this band is associated
6
1
)p
z
(M)] are strongly stabilized, whereas effects upon other
with the Ni-P unit.
6
b
nd(M) f ((n + 1)p(M), π*(L)) transitions are minor.
The absence of a similar transition for Pt(II) complexes
suggests that it is strongly blue-shifted for the third transition
series. Such behavior is incompatible for d(M) f p(M) or
MLCT (metal-to-ligand charge-transfer) transitions, but is
consistent with an LMCT (ligand-to-metal charge-transfer)
6
Unlike for dinuclear Rh(I) and Ir(I) isocyanide complexes,
there is no clear correlation of the dinuclear Pt(II) ∼330 nm
1
2
band to a [nd
z
f (n + 1)p
z
] transition for monomer 3.
2
Indeed, it is possible that nd
z
z
f (n + 1)p character may be
30a
spread over several transitions for 3, owing to mixing of
assignment. The obvious assignment is P(σ) f Ni(d 2 2).
x -y
p
z
(Pt) with π*(L) orbitals. The resonance Raman results
Such an LMCT assignment does not appear to have been
previously considered for this band, but is reasonable for
the first transition series metal ion considering the strongly
reducing nature of phosphines. In any case, the UV bands
are clearly unrelated to the nature of the Ni-Ni interactions.
The spectra of dinuclear 2a and 2b show a number of
moderately intense bands (ꢀ ∼ 4000 dm mol cm ) in
the 300-450 nm region that are not evident in the monomer
spectrum. However, assignment of these bands to Ni-Ni
1
presented below suggest pure [nd
σ σ
* f (n + 1)p ] character
for the ∼330 nm band of the dinuclear complexes, perhaps
because energetic resolution of the low-energy transitions
results in reduced metal-ligand mixing.
The energetic ordering of the low-energy Pt(II)-Pt(II)
3
-1
-1
bands tracks the Pt‚‚‚Pt distance; a shorter distance by 0.1325
1
Å (Table 2) results in a [nd
σ
σ
* f (n + 1)p ] transition energy
-
1
that is blue-shifted by ∼800 cm , consistent with a weaker
Pt-Pt interaction. Correlations of this type have been clearly
demonstrated for the absorption/emission spectra of linear-
transitions is confounded both by their weakness and by their
number (there is only one (d
1
σ σ
* f p ) transition).
3872 Inorganic Chemistry, Vol. 41, No. 15, 2002

8
Dinuclear d Metal Cyanide Complexes
In fact, low-spin square-planar Ni(II) complexes do display
1
-3
3
-1
-1
very weak transitions (ꢀ ∼ 10
dm mol cm ) in the
visible region with d-d transitions terminating in the empty
x -y orbital. Such transitions are strongly blue-shifted for
2
2
d
3
0a
Pt(II).
]2
4
-
The most thoroughly examined case is the [Ni(CN)
complex, which has been examined by solution absorption
30a,b
and MCD spectroscopy
and single-crystal absorption
Bands have been located at 310 (ꢀ 700
30d,32
spectroscopy.
3
-1
-1
dm mol cm ), 330 (ꢀ 500), 370, 440 (ꢀ ∼ 2), and 500
nm. The third and fifth of these were only evident in the
single-crystal spectra.32 Detailed assignments are not avail-
able for these bands, and singlet-triplet assignments may
be applied for some,3 particularly the very weak 440 and
0a
500 nm bands.
Nevertheless, the important point is that d-d transitions
are known to occur in the visible to near UV region for
2-
[
Ni(CN)
cyanide, d-d transitions should actually be red-shifted for
trans-[Ni(PR (CN) ] complexes. We therefore assign the
00-450 nm bands of 2a and 2b to d-d transitions.
The reason for the weakness of d-d transitions in
4
] . Since phosphines are weaker-field ligands than
3
)
2
2
3
Figure 6. Resonance Raman spectra of 1a in CH2Cl2 solution obtained
with 319.9 and 341.5 nm excitation wavelengths. The spectra have been
intensity corrected, and the Rayleigh line, glass bands (+ mark), and solvent
bands (* mark) have been subtracted.
monomers is that they are Laporte forbidden. This is rigorous
for 4, as inversion site symmetry imposes C-Ni-C and
P-Ni-P bond angles of 180°. In contrast, 2a and 2b show
marked pyramidalization at the Ni atoms, with C-Ni-C and
P-Ni-P angles of 175.9° and 176.5°, respectively, for 2a,
and 170.5° and 178.1°, respectively, for 2b. This distortion
may allow the single-centered d-d transitions to mix with
higher-energy dipole-allowed transitions, thus promoting their
intensity.
We note that our explanation for the lowest-energy bands
of the dinuclear Ni(II) complexes does not invoke Ni-Ni
interaction; instead, we attribute them to distortions from
planarity at the Ni atoms imposed by the steric constraints
of the bridging ligands. Indeed, we have found no evidence
in the electronic spectra for significant Ni-Ni interaction.
Resonance Raman Studies. To further probe the metal-
metal interactions of the dinuclear metal complexes, reso-
nance Raman spectra of 1a and 2a have been recorded upon
excitation into their lowest-energy dipole-allowed absorption
bands. Figure 6 shows the resonance Raman spectrum of 1a
indicates substantial MLCT character for the transition, which
is due to the low energy of empty diimine π* orbitals.2
When π* ligand orbitals lie at substantially higher energies,
as is true for the ligands in the present study according to
d,9g,i
3
3
electron-transmission spectroscopy, the terminal orbital of
the lowest-energy electronic transitions becomes virtually
pure metal (n + 1)p in nature, and the characteristic pattern
2
of almost exclusive resonance enhancement of ν(M ) and
overtones emerges. The validity of this argument is supported
by transient-IR measurements on a dinuclear Rh(I) isocyanide
34
complex, which showed the ν(CtN) value of the triplet
-1
excited state to be only 16 cm lower than that of the ground
state. Excited-state distortions involving ligands are accord-
ingly predicted to be very small, hence resonance enhance-
ment of the corresponding vibration modes should also be
very small.
The smaller ν(Pt
value of 118 cm in [Pt
-1
2
) (93 cm ) for 1a compared with the
obtained in CH
2
Cl
2
solution excited at 319.9 and 341.5 nm,
-1
4- 6a
2
(P
2
O
5
2
H )
4
]
indicates weaker
respectively. The Raman intensity appears only in a single
low-frequency mode plus overtones. The resonance-enhanced
metal-metal interaction, consistent with the larger Pt‚‚‚Pt
distances in the former complex. The assigned ν(Pt
2
) of 93
-
1
Raman band appears at ∆ν ) 93 cm and is assigned to
the Pt-Pt stretch. This implies that the excited-state distortion
is localized at the Pt-Pt bond, in agreement with the
-1
-1
cm can also be compared to the ν(Au
2
) of 88 cm for
2+ 35a
-1
[
Au
2
(dcpm)
2
]
and ν(Ag
2
) of 80 cm
2
for [Ag -
2+ 35b
(dcpm)
2
] . Clearly the metal-metal bonding interaction
σ σ
assignment of the metal-centered 5d * f 6p transition to
in 1a must be stronger than that in [Au(I)]
Using a previously described method of calculation, the
2
and [Ag(I)]
2
.
the 337 nm absorption band for 1a (see above) and previous
work.4 Moreover, this result contrasts sharply with reso-
nance Raman studies of dinuclear Pt(II) complexes contain-
6g
,6
(33) (a) Jordan, K. D.; Burrow, P. D. Acc. Chem. Res. 1978, 11, 341-
348. (b) Burrow, P. D.; Ashe, A. J., III; Bellville, D. J.; Jordan, K. D.
J. Am. Chem. Soc. 1982, 104, 425-429.
(34) Doorn, S. K.; Gordon, K. C.; Dyer, R. B.; Woodruff, W. H. Inorg.
Chem. 1992, 31, 2284-2285.
2d
ing R-diimine ligands, which display resonance enhance-
ment of R-diimine ligand modes upon excitation into the
low-energy “metal-metal” transition band. This latter result
(
35) (a) Leung, K. H.; Phillips, D. L.; Tse, M.-C.; Che, C.-M.; Miskowski,
V. M. J. Am. Chem. Soc. 1999, 121, 4799-4803. (b) Che, C.-M.;
Tse, M.-C.; Chan, M. C. W.; Cheung, K.-K.; Phillips, D. L.; Leung,
K. H. J. Am. Chem. Soc. 2000, 122, 2464-2468.
(
32) Ballhausen, C. J.; Bjerrum, N.; Dingle, R.; Eriks, K.; Hare, C. R. Inorg.
Chem. 1965, 4, 514-518.
Inorganic Chemistry, Vol. 41, No. 15, 2002 3873

Xia et al.
Table 4. Resonance Raman Bands of 1a in CH2Cl2 Solution
compounds, the basis sets are relatively small and no
d-polarization functions were added for second- and third-
row atoms. However, as described below, the optimized
geometry parameters for both the mononuclear and dinuclear
models are in good agreement with the corresponding bond
distances and angles determined by X-ray crystal analysis.
intensityb
Raman shift (cm-
1 a
Raman bands
)
319.9 nm
341.5 nm
0.09
63
100
-
2νPt-Pt
νPt-Pt
-186
-93
93
-
64
100
νPt-Pt
absolute Raman cross section
of νPt-Pt (Å /molecule)
Furthermore, the d(P) orbital parentage in the trans-[M(PMe
(CN) ] frontier MO compositions is small. Previous work
on the dinuclear [Au complex has indi-
3 2
) -
2
-
9
-9
-9
-9
expt
calcd
calcd
8.6 × 10
3.58 × 10
7.85 × 10
22
9.25 × 10
1.34 × 10
1.25 × 10
18.4
2
c
-9
-9
2+
20
2 2 2 2 2
(µ-H PCH PH ) ]
d
cated that the 3d(P) orbitals have little effect on the ground
and excited states. Other theoretical calculations31 on Pt-P
or Pd-P systems have also revealed that the d-polarization
functions for P atoms act as polarization functions and the d
orbitals are not involved in metal-to-ligand charge transfer.
2
3
νPt-Pt
186
280
νPt-Pt
a
Estimated uncertainties are about 4 cm-1 for the wavenumbers.
b
Relative intensities are based on integrated areas of Raman bands.
Estimated uncertainties are about 5% for intensities greater than 100, 10%
for intensities between 50 and 100, and 20% for intensities below 50.
31d
The unimportant nature of 3d(P) orbitals in bonding is
c
Calculated using the parameters of Table S2A (Supporting Information)
now understood to be very general; the low-energy
σ*(P-C) or (P-H) orbitals and not the 3d(P) orbitals are
thought to act as π-acceptor orbitals according to current
theories. We believe that the LANL2DZ basis sets with no
d-polarization function added to P atoms are adequate, and
they were used in the dinuclear calculations in order to reduce
computational time and resources.
with eq 1 and 2 in ref 6g and the exponential decay damping function for
solvent dephasing. Calculated using the parameters of Table S2B (Sup-
porting Information) with eq 1 and 2 in ref 6g and the overdamped Brownian
oscillator damping function for solvent dephasing.
d
excited-state Pt-Pt stretching vibration is estimated to be
-
1
1
45 cm , which is in agreement with the electronic
assignment of a transition to an excited state with increased
metal-metal bonding. Using the parameters in Table 4 and
The optimized molecular structure parameters are sum-
marized in Table 5. The calculated M‚‚‚M distances of
3.1339 Å for Pt‚‚‚Pt and 3.2447 Å for Ni‚‚‚Ni are very
similar to the experimental M‚‚‚M distances of 3.189(1) and
3.209(8) Å found in complexes 1b and 2b, respectively.
However, they deviate from the M‚‚‚M distances of
3.0565(4) and 2.957(1) Å in 1a and 2a, respectively. The
6g
a related calculation method, the Pt‚‚‚Pt distances change
by 0.11 Å for 1a in the excited state relative to the ground-
state value. The variation for the Pt-Pt bond length for 1a
4
- 6g
2 2 5 2 4
is smaller than that of 0.225 Å for [Pt (P O H ) ] .
The resonance Raman spectrum of 2a in methanol with
416.0 nm excitation has also been recorded (see Supporting
Information, Figure S6). The Raman intensity is very weak.
The spectrum shows only two weak low-frequency peaks at
optimized [M
to the experimental data for [M
2
(µ-H
2
PCH
2
PH
2
)
2
(CN)
4
] structures are similar
(CN) ] complexes,
2
(dmpm)
2
4
-
1
-1
-
68 and 75 cm , and no overtones appear. The -68 cm
and the shorter M‚‚‚M distances in 1a and 2a compared to
1b and 2b are attributed to the steric effect of the dcpm
ligand, as argued in a previous section. The other optimized
structural parameters for the dinuclear and mononuclear
models are consistent with the X-ray structural data. This
indicates that the calculations give reasonable representations
of the ground states.
-1
feature is the quasi-anti-Stokes analogue of the 75 cm band.
These weak Raman peaks may be attributed to the Ni-Ni
stretching vibration, although no overtone appears and the
assignment is only tentative.
Theoretical Calculations. In this work, we have per-
formed ab initio MP2 calculations on the molecules [M
2
(µ-
H
2
PCH PH (CN) ] and trans-[M(PMe (CN) ] (M ) Pt,
2
2
)
2
4
3
)
2
2
The partial molecular orbital compositions for [Pt
PCH PH (CN) ] (see Table S5) and those for other
model compounds are given in Supporting Information. The
MO diagram of the Pt d-orbitals in [Pt (µ-H PCH PH
(CN) ] is depicted in Figure 7. From Table S5 we can see
that all the 5d(Pt) orbitals are involved in the filled MOs
2
(µ-
Ni). In the geometry optimization we have adopted different
basis sets for the dinuclear and mononuclear model mol-
ecules, since their molecular sizes are different. The basis
sets for the mononuclear compounds include d-polarization
functions for second- and third-row atoms. For the dinuclear
H
2
2
2
)
2
4
2
2
2
2 2
) -
4
Table 5. The Optimized Structural Data of the Model Complexes [M2(µ-H2PCH2PH2)2(CN)4] and trans-[M(PMe3)2(CN)2] (M ) Pt, Ni)a
Pt dimer [1b]
Ni dimer [2b]
Pt monomer [3]
Ni monomer [4]
Bond Length (Å)
3.2447 [3.209(8)]
2.1798 [2.1984(4)]
1.8413 [1.865(1)]
1.2226 [1.145(2)]
1.9099 [1.832(1)]
M‚‚‚M
M-P
M-C
C-N
P-C
3.1339 [3.189(1)]
2.3606 [2.311(5)]
2.0224 [2.00(2)]
1.2233 [1.185(4)]
1.8980 [1.88(2)]
2.3436 [2.336(1)]
2.0066 [2.021(5)]
1.1826 [1.105(7)]
1.8818 [1.849(5)]
2.2322 [2.2426(6)]
1.897 [1.857(3)]
1.2104 [1.148(4)]
1.8768 [1.852(2)]
Bond Angle (deg)
91.76 [89.2]
93.178 [92.15]
173.64 [170.47(6)]
92.44 [90.03(4)]
176.49 [178.09(1)]
P-M-M
C-M-M
C-M-C
C-M-P
P-M-P
90.06 [89.25(1)]
93.265 [91.975(8)]
173.47 [175.6(7)]
89.99 [87.7(7)]
180.0 [180.0]
90.0 [90.0(1)]
180.0 [180.0]
180.0 [180.0]
90.0 [90.0(8)]
180.0 [180.0]
179.88 [178.3(1)]
a
The values within square brackets are the corresponding bond distances/bond angles of complexes 1b, 2b, 3, and 4 determined by X-ray crystal analysis.
3874 Inorganic Chemistry, Vol. 41, No. 15, 2002

8
Dinuclear d Metal Cyanide Complexes
is more dz2 contribution in the σ-bonding MO (28a) than in
2
the σ*-antibonding one (33a). The difference in d
z
composi-
tion leads to σ-bonding interaction along the Pt-Pt axis.
Furthermore, in the filled MOs there is small 6s6p parentage
that would indirectly contribute to the metal-metal interac-
tion.
Compared to [Pt
2
(µ-H
2
PCH
PH
2
PH
2
)
2
(CN)
4
], the MO com-
positions of [Ni (µ-H
2
2
PCH
2
2
)
2
(CN)
4
] are not explicit for
Ni-Ni bonding. However, comparing the MO compositions
of the dinuclear and mononuclear model compounds, we can
highlight one clear trend: the occupied MOs of dinuclear
compounds are significantly less perturbed from those of the
monomer for Ni than for Pt. The 3d(Ni) orbitals appear to
-
1
be split by a few 1000 cm (0.2-0.9 ev) in the dimer,
whereas the 5d(Pt) orbitals undergo much larger splitting in
the dimer (0.7-2.7 ev), particularly in the case of the
2
predominately d
z
parentage levels (2.66 eV, see Figure 7
and Tables S5 and S6 in the Supporting Information). This
11
is consistent with Cotton’s calculation indicating negligible
Ni-Ni interaction at a noticeably shorter intermetal distance
(2.485 Å).
Acknowledgment. This work was supported by the
University of Hong Kong and the Hong Kong Research
Grants Council of the Hong Kong SAR (P.R. China) [HKU
Figure 7. The calculated MO diagram of the partial Pt d-orbitals in [Pt2(µ-
H2PCH2PH2)2(CN)4].
7298/99P]. We thank Prof. Zhong-Yuan Zhou and Dr. Nian-
Yong Zhu for some X-ray structural analyses and Dr. V. M.
Miskowski for helpful discussions on spectroscopic assign-
ments and comments on the manuscript.
and contribute to the metal-metal and/or metal-ligand
interactions. To check the metal-metal interaction, we pay
attention to those MOs that have large metal parentage
(
8
2
>60%). The 22b and 24b MOs which have 81.4% dyz and
7.0% dxz character, respectively, are the π orbitals, and the
3b and 28b MOs with 73.1% dyz and 78.1% dxz character,
Supporting Information Available: X-ray crystallographic file
for all compounds in CIF format; selected bond distances and
angles for compounds 3 and 4; parameters for resonance Raman
simulations for compound 1a; calculated MO energies and orbital
compositions for dimer and monomer model complexes; ORTEP
diagrams for compounds 1a, 1b, 3, and 4; the resonance Raman
spectrum of compound 2a in methanol and related input and
output gross orbital populations files for calculation. This ma-
terial is available free of charge via the Internet at http://pubs.
acs.org.
respectively, are the π* orbitals. These four MOs are
responsible for the π type Pt-Pt interaction. However, they
are fully occupied and there is no net π bonding. The δ
bonding of Pt-Pt is not significant due to the small parentage
of the dxy and dx -y in the δ type MOs. The important feature
would be the σ type metal-metal MOs. The rather large
2
2
2
2
d
d
z
-d
z
overlap is found in 28a and 33a MOs, which have
z2 compositions of 78.1% and 66.0%, respectively. There
IC020077J
Inorganic Chemistry, Vol. 41, No. 15, 2002 3875