On the electronic structure of mesitylnickel complexes of α-diimines - Combining structural data, spectroscopy and calculations

Article abstract of DOI:10.1002/ejic.200300956

New organometallic nickel complexes of the type [(α-diiminediimine)- Ni(Mes)Br] and [(α-diimine)Ni(Mes)₂] (Mes = mesityl = 2,4,6-trimethylphenyl) were prepared and characterised spectroscopically in detail. A combination of spectroscopic techni

Full text of DOI:10.1002/ejic.200300956

FULL PAPER
On the Electronic Structure of Mesitylnickel Complexes of α-Diimines —
Combining Structural Data, Spectroscopy and Calculations
[b]
[b]
´ ˇ[c]
Axel Klein,*[a][‡] Martin P. Feth,* Helmut Bertagnolli, and Stanislav Zalis
Keywords: Density functional calculations / Electronic structure / EXAFS / Nickel / Raman spectroscopy
New organometallic nickel complexes of the type [(α-diimine)- ligands with the nickel centre. The low-lying electronic
Ni(Mes)Br] and [(α-diimine)Ni(Mes)2] (Mes
2,4,6-trimethylphenyl) were prepared and characterised
=
mesityl
=
transitions are assigned as mixed MLCT/LЈLCT or MLCT/
XLCT due to low-lying accepting π*-orbitals centred on the
spectroscopically in detail. A combination of spectroscopic diimine ligands and mixed metal/co-ligand MOs as donor
techniques (XRD, EXAFS, absorption, resonance Raman)
and quantum chemical (DFT) calculations reveals the inter-
levels.
( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
play of the diimine ligands and the mesityl or bromine co- Germany, 2004)
Introduction
investigation of their structures and electronic properties,
however, were scarce[15,18Ϫ20] until Yamamoto et al. recently
reported and discussed the optical spectra of some alkyl
and aryl complexes with 2,2Ј-bipyridine (bpy).[21] In this re-
port the spectra were mainly used to monitor elimination
processes. In order to fill that gap we initiated a detailed
and combined research project using various spectroscopic
techniques (absorption, emission and resonance Raman)
and quantum chemical calculations, with a focus on the op-
tical properties.
Organometallic nickel complexes with α-diimine ligands
like 2,2Ј-bipyridine (bpy), 1,10-phenanthroline (phen) or di-
azabutadienes (R-DAB) have gained an enormous interest
in the last decade. This is mainly due to their success as
effective catalysts in olefin polymerisation or olefin/CO co-
polymerisation.[1Ϫ9] It has been established, mainly by
Brookhart, that methylnickel complexes with carefully de-
signed diazabutadiene ligands exhibit higher activities than
the classical Ziegler catalysts combined with a greatly re-
One reason why such investigations have not been per-
duced sensibility towards poisoning by polar functions. De- formed so far is probably the high sensitivity of such alkyl-
tailed investigations on the activities of various derivatives nickel complexes towards oxygen or moisture.[19] They also
on polymerisation,[1a] theoretical calculations on the easily undergo thermal decomposition by reductive elimin-
mechanisms and energetics[2] and the use of these com- ation.[21,22] Aryl derivatives are much more inert towards
pounds for other catalytic purposes[10Ϫ13] have been pub- water or oxygen, although the threat of reductive elimin-
lished so far.
ation (forming biaryls) remains.[22] We have recently shown
In the 1960s these compounds were already known as that introducing the bulky mesityl co-ligand (Mes ϭ 2,4,6-
darkly coloured very reactive materials.[14Ϫ17] Detailed trimethylphenyl) results in very stable mesityl- and dimes-
itylnickel complexes[23] of 2,2Ј-bipyridine, and, with the
proper choice of ligands, compounds such as [(N∧N)Ni-
[a]
Institut für Anorganische Chemie, Universität Stuttgart,
(Mes)Br] (N∧N ϭ α-diimine) have allowed us to carry out
Pfaffenwaldring 55, 70569 Stuttgart, Germany
Fax: ϩ49-(0)711-685-4165
E-mail: aklein@iac.uni-stuttgart.de
a detailed study of the ligand-exchange behaviour of these
complexes using EXAFS and spectroscopic techniques like
NMR or absorption spectroscopy.[24]
[b]
Institut für Physikalische Chemie, Universität Stuttgart,
Pfaffenwaldring 55, 70569 Stuttgart, Germany
Fax: ϩ49-(0)711-685-4443
E-mail: m.feth@ipc.uni-stuttgart.de
In this paper we wish to report the synthesis and detailed
spectroscopic and structural (EXAFS ϩ XRD) characteris-
ation of a series of bromomesityl- and dimesitylnickel com-
plexes of various diimine ligands together with the results
of a detailed study of the electronic properties of selected
samples using a combination of spectroscopy (absorption,
resonance Raman) and quantum chemical calculations. The
latter were based on the molecular structures established by
single-crystal XRD measurements and EXAFS spec-
[c]
[‡]
´
J. Heyrovsky Institute of Physical Chemistry, Academy of
Sciences of the Czech Republic,
ˇ
Dolejskova 3, 18 000 Prague 8, Czech Republic
E-mail: zalis@jh-inst.cas.cz
Present address: Institut für Chemie, Karl-Franzens-
Universität Graz
Schubertstrasse 1, 8010 Graz, Austria
Fax: ϩ43-(0)316-380-9835
E-mail: axel.klein@uni-graz.at
Supporting information for this article is available on the
WWW under http://www.eurjic.org or from the author.
2784
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/ejic.200300956
Eur. J. Inorg. Chem. 2004, 2784Ϫ2796

Mesitylnickel Complexes of α-Diimines
FULL PAPER
Scheme 1. Preparation of the α-diimine complexes with numbering; tmphen ϭ 3,4,7,8-tetramethyl-1,10-phenanthroline, dmbpy ϭ 4,4Ј-
dimethyl-2,2Ј-bipyridine, bpy ϭ 2,2Ј-bipyridine, bpym ϭ 2,2Ј-bipyrimidine, bpz ϭ 2,2Ј-bipyrazine, bpm ϭ 4,4Ј-bipyrimidine, iPr-DAB ϭ
N,NЈ-diisopropyl-1,2-ethanediimine (N,NЈ-diisopropyl-1,4-diazabutadiene), terpy ϭ 2,2Ј,6ЈЈ-terpyridine
troscopy. The compounds under study are bromomesityl basic the diimine ligands are (e.g. bpz or bpm) the more the
complexes of the type [(N∧N)Ni(Mes)Br] (N∧N ϭ diimine equilibrium tends towards the starting materials. For iPr-
ligands, Mes ϭ mesityl ϭ 2,4,6-trimethylphenyl) with vari- DAB the procedure had to be varied to obtain reasonable
ous α-diimine ligands and, additionally, selected dimesityl yields (see Exp. Sect.). With the very weakly basic ligand
complexes [(N∧N)Ni(Mes)2] (see Scheme 1).
N,NЈ-di(2,6-xylyl)-1,2-ethanediimine (Xyl-DAB) no prod-
uct was formed upon reaction with [(PPh3)2Ni(Mes)Br] (1).
Once formed, and in the absence of PPh3 or other strong
ligands, the compounds are stable in solvents like CH2Cl2,
THF, toluene or acetone towards ligand-exchange reac-
tions.[24] The complex [(terpy)Ni(Mes)]Br (2h) was prepared
Results and Discussion
Synthesis and General Properties
The neutral complexes [(N∧N)Ni(Mes)Br] (N∧N ϭ di- following the same method. The 1H NMR spectroscopic
imine ligands) were synthesised from the precursor complex data reveal that all three pyridine units bind to nickel and
[(PPh3)2Ni(Mes)Br] (1) by ligand-exchange reactions in ace- that Br serves only as a counteranion to the cationic com-
tone or toluene solution [see Scheme 1 (A)] and were ana- plex. The dimesityl complexes [(N∧N)Ni(Mes)2] (3) were
lysed by 1H and 13C NMR spectroscopy and elemental prepared for selected samples by reacting the bromomesityl
analysis (see Table 1).
For rather basic diimine ligands like bpy or tmphen reac-
complexes with mesityllithium as shown in Scheme 1 (B).
The NMR spectroscopic data reflect the geometric
tion A proceeds very smoothly with high yields. The less properties of the compounds. The bromomesityl complexes
1
Table 1. H NMR spectroscopic data and details of the preparation of complexes [(N∧N)Ni(Mes)Br]
1H NMR
δ (ppm)[a]
Yield Formula
(%) (mol. mass)
Elem. anal.
found C, H, N (calcd.) (%)
2a 2,9
7.09, 9.28
2b 6,6Ј
9.20, 6.98
2c 6,6Ј
9.42, 7.20
2d 6,6Ј
9.63, 7.40
2e 3,3Ј
9.71, 9.52
2f 2,2Ј
5,6
8.25
3,3Ј
8.18, 8.16
3,3Ј
Me4,7
2.78
Me4,4Ј
2.51, 2.39
4,4Ј
8.18, 8.15
4,4Ј
9.21, 9.13
5,5Ј
9.12, 8.75 s
5,5Ј
Me3,8
2.65, 2.61 6.50
5,5Ј m-H
7.49, 7.14 6.43
5,5Ј m-H
7.70, 7.34 6.35
5,5Ј m-H
m-H
o-Me p-Me 92
3.09 2.75
o-Me p-Me 88
3.03 2.17
o-Me p-Me 95
3.05 2.14
o-Me p-Me 75
3.02 2.19
C25H27BrN2Ni 60.79 (60.77), 5.61 (5.51), 5.68 (5.67)
(494.11)
C21H23BrN2Ni 57.13 (57.06), 5.25 (5.24), 6.35 (6.34)
(442.04)
C
19H19BrN2Ni 55.33 (55.12), 4.65 (4.63), 6.72 (6.77)
(413.98)
17H17BrN4Ni 49.10 (49.09), 4.09 (4.12), 13.13 (13.47)
8.35, 8.31
C
7.92, 7.58 6.47
(415.97)
6,6Ј
9.80, 7.33
6,6Ј
m-H
o-CH3 p-CH3 78
C17H17BrN4Ni 48.92 (49.09), 4.02 (4.12), 13.21 (13.47)
(322.73)
C17H17BrN4Ni 49.02 (49.09), 4.13 (4.12), 13.41 (13.47)
6.50 s 2.91 s 2.20 s
m-H
o-CH3 p-CH3 75
10.04 s, 7.85 s 8.08 d, 8.02 d 9.36 d, 9.25 d
6.51 s 2.30 s 2.21 s
(322.73)
2g H imine
8.55 s, 8.43 s
2h 6,6ЈЈ;3,3ЈЈ
HCiPr
3.26 m
5,5ЈЈ
H3CiPr
1.03 d
3Ј,5Ј;4Ј
m-H
6.42
m-H
o-CH3 p-CH3 82
2.97 s 2.21 s
o-CH3 p-CH3 95
C17H27BrN2Ni 56.55 (57.07), 6.29 (6.59), 32.09 (31.60)
(398.03)
C24H26BrN3Ni 58.73 (59.26), 6.88 (1.18), 30.59 (29.17)
(495.10)
4,4ЈЈ
7.48 qd, 8.69 d 8.35 ddd
7.61
8.72, 8.52 6.72 s 2.96 s 2.29 s
[a]
Measured in [D6]acetone.
Eur. J. Inorg. Chem. 2004, 2784Ϫ2796
www.eurjic.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2785

´ ˇ
A. Klein, M. P. Feth, M. Bertagnolli, S. Zalis
FULL PAPER
give full sets of 1H and 13C resonances (see Table 1 and
The structure was solved in the orthorhombic Pbca space
Supporting Information) for the chelate ligands, confirming group (for details see Exp. Sect.). A solution in P21/n, which
the planar geometry and CS symmetry. For the dimesityl is the space group obtained for the bpy derivative 3c, gave
complexes C2 symmetry can be deduced from the NMR slightly worse R values. The crystal structure reveals that
spectra.
the complex forms dimers in the unit cell which are π-
stacked in a non-graphite-like fashion, with an interplanar
distance of 3.49(1) A. As displayed in Figure 2 the two cen-
˚
Structures
tral rings overlap but are not completely eclipsed (shift of
The structures of the investigated complexes are essential
for the discussion of electronic properties since the molecu-
lar entity must be unequivocal. Also, the optical spec-
troscopy might be affected by intermolecular interactions
like stacking and, finally, they serve as a basis for the quan-
tum chemical calculations. Selected samples were therefore
submitted to single-crystal XRD or powder EXAFS study.
˚
0.69 A). The interplanar distance is similar to that in the
recently studied isoleptic complex [(tmphen)Pt(Mes)2],[25d]
˚
but longer than the distance found in graphite (3.354 A).
˚
For the platinum complex a similar distance of 3.52(1) A
was found but the ligands stack in a staggered fashion and
are also not back-to-back, but offset.
XRD
The crystal structures of two examples of the series
[(N∧N)Ni(Mes)Br] with N∧N ϭ bpy (2c) or iPr-DAB
(2g)[24] and the structure of [(bpy)Ni(Mes)2] (3c)[23] have re-
cently been published. In this paper we report the crystal
structure of [(tmphen)Ni(Mes)2] (3a; Figure 1). The reason
for adding this structure to the series is because of the spe-
cial character of the tmphen ligand. In the series shown in
Scheme 1 tmphen is the most basic ligand and, more im-
portantly, it is a much more rigid ligand than bpy or the
bidiazines. This has a strong impact on the molecular struc-
ture since previous studies have revealed that the steric de-
mand of two cis oriented mesityl substituents in square-
planar complexes of nickel, palladium or platinum leads to
marked deviations not only in the coordination sphere but
also within the diimine ligands.[23Ϫ25] This rigidity also has
a strong impact on the optical properties, as we will see
later.
Figure 2. Stacking of two molecules of 3a in the unit cell
The molecular structure of 3a resembles in many respects
that of the bpy derivative 3c. The essential distances, for
example NiϪN or NiϪC, and the chelate bite angle
(around 82°) are very similar. The distortion of the square-
planar coordination plane is higher (15.1° vs. 12.4°), as ex-
pected. In both cases the distortion of the diimine ligand is
quite small (2.8° vs. 3.1°). The mesityl substituents are tilted
towards the coordination plane [C(21)ϪN(2)ϪNi(1)Ϫ
N(1)ϪC(31)] by 70.0(6)° and 69.0(6)°, respectively, as was
observed for other square-planar dimesityl complexes.[23,25]
In the related complex [(bpy)Ni(C6F5)2][22a] the C6F5 co-
ligands are oriented almost perpendicular to the coordi-
nation plane and both the NiϪN distances [1.938(3) and
˚
1.935(3) A) and the NiϪC distances [1.903(3) and 1.907(3)
˚
A] are markedly shorter. The NϪNiϪN bite angle (82.9°)
lies in the normal range whereas the CϪC bond between
˚
the two pyridyl units is noticeably elongated to 1.477(7) A.
The latter is a clear indication of reduced back-bonding to
the diimine ligand due to the fact that the fluoroaryl co-
ligands donate much less electron density than their mesityl
counterpart in the complexes 3a and 3c. This is also the
reason for the different NiϪC bond lengths in the two types
of arylnickel complexes. We thus conclude that the aryl co-
ligand strongly governs the geometry around the nickel
atom (and also the electron distribution).
Figure 1. Molecular structure of 3a (50% probability ellipsoids; H
˚
atoms omitted for clarity); selected bond lengths (A) and angles
(deg): NiϪN(1) 1.975(4), NiϪN(2) 1.960(4), NiϪC(11) 1.911(4),
NiϪC(21) 1.914(4), N(1)ϪC(1) 1.337(6), N(2)ϪC(10) 1.333(6),
N1ϪC(5) 1.345(13), N2ϪC(6) 1.385(12); N(1)ϪNiϪN(2)
82.55(16), C(11)ϪNiϪC(21) 90.47(18), N(1)ϪNiϪC(11) 94.01(17),
N(1)ϪNiϪC(21)
168.30(17),
N(2)ϪNiϪC(21)
94.79(17),
N(2)ϪNiϪC(11) 169.65(18)
2786
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2004, 2784Ϫ2796

Mesitylnickel Complexes of α-Diimines
FULL PAPER
˚
EXAFS Spectroscopy
bromine backscatterer at about 2.3 A and the third coordi-
˚
nation shell at about 2.8 A contains the further backbone-
In the Ni-K edge XANES spectra of all complexes the
typical two pre-edge peaks of square-planar NiII complexes
can be observed.[26] The pre-peak at about 8333 eV can be
assigned to a 1sǞ3d electron transition, while the second
peak, which occurs at about 8337 eV, is due to a 1sǞ4pz
transition with shakedown contributions.[26a]
The obtained structural parameters of the investigated
complexes determined by curve fitting analysis of the EX-
AFS spectra at the Ni-K and Br-K edge are summarised in
Table 2. [(dmbpy)Ni(Mes)Br] (2b), [(bpym)Ni(Mes)Br] (2d)
and [(bpz)Ni(Mes)Br] (2e) show very similar EXAFS func-
tions at the Ni-K and Br-K edge, leading to almost identical
structural parameters, which are in very good agreement
with values found from single crystal XRD and from the
EXAFS analysis of the complex [(bpy)Ni(Mes)Br] (2c).[24]
As an example, Figure 3 shows the k3-weighted EXAFS
spectra coupled with the magnitudes of the Fourier trans-
forms of 2e at the Ni-K (a and b) and at the Br-K edge (c
and d) The fitting of the Ni-K edge EXAFS spectra was
performed using a three-shell model in which the first coor-
carbon atoms of the mesityl and diimine ligand. Because of
the quite equal distances and backscatter behaviour of the
nitrogen and carbon backscatterers in the first coordination
shell, only one shell was fitted with nitrogen amplitude- and
phase-functions. The Br-K edge EXAFS spectra were fitted
with one coordination shell of nickel at about 2.3 A. In
complex 2a a slightly higher NiϪC/N distance (1.97 A) of
˚
˚
the first coordination shell was detected, although the
NiϪBr distance is of the same order of magnitude.
In Figure 4 (a) a comparison of the experimental k3-
weighted EXAFS functions of [(tmphen)Ni(Mes)2] (3a),
[(bpy)Ni(Mes)2] (3c) and [(bpz)Ni(Mes)2] (3e) at the Ni-K
edge is shown. The spectra of 3a, 2h, 3c and 3e were fitted
with a two-shell model: the first shell is due to the carbon
and nitrogen atoms of the imine and mesityl ligands coordi-
nating directly to the nickel centre atom and the second
shell consist of the backbone carbon atoms of the ligands.
The structural parameters of 3a, 2h and 3c determined by
EXAFS are very similar.
˚
dination shell at about 1.9 A consists of the two coordinat-
At first sight the complex [(bpz)Ni(Mes)2] (3e) seems to
ing nitrogen atoms of the diimine ligand and the carbon deviate from 3a, 2h and 3c. The peak of the first coordi-
atom of the mesityl group, the second shell contains the nation shell in the Fourier-transformed EXAFS spectrum
Table 2. Structural parameters of the solid complexes determined from the Ni-K and Br-K edge EXAFS spectrum
[(N∧N)Ni(Mes)Br]
[a]
Ϫ1
N∧N
r (A)
N
σ [A]
∆E0 (eV)
k-range (A
)
Fit-Index
37.3
˚
˚
˚
tmphen (2a)
NiϪC/N
NiϪBr
NiϪC
BrϪNi
NiϪC/N
NiϪBr
NiϪC
BrϪNi
NiϪC/N
NiϪBr
NiϪC
BrϪNi
NiϪC/N
NiϪBr
NiϪC
BrϪNi
NiϪC/N
NiϪBr
NiϪC
BrϪNi
NiϪC/N
NiϪC
1.97 Ϯ 0.02
2.30 Ϯ 0.02
2.82 Ϯ 0.03
2.30 Ϯ 0.02
1.93 Ϯ 0.02
2.29 Ϯ 0.02
2.79 Ϯ 0.03
2.29 Ϯ 0.02
1.92 Ϯ 0.02
2.30 Ϯ 0.02
2.81 Ϯ 0.03
2.30 Ϯ 0.02
1.91 Ϯ 0.02
2.29 Ϯ 0.02
2.82 Ϯ 0.03
2.29 Ϯ 0.02
1.92 Ϯ 0.02
2.29 Ϯ 0.02
2.80 Ϯ 0.03
2.29 Ϯ 0.02
1.89 Ϯ 0.02
2.79 Ϯ 0.03
2.6 Ϯ 0.3
0.8 Ϯ 0.1
1.9 Ϯ 1.4
0.9 Ϯ 0.1
3.1 Ϯ 0.3
0.9 Ϯ 0.2
3.8 Ϯ 1.1
1.0 Ϯ 0.1
3.1 Ϯ 0.3
1.0 Ϯ 0.2
5.6 Ϯ 1.7
0.9 Ϯ 0.1
2.8 Ϯ 0.3
1.2 Ϯ 0.2
2.8 Ϯ 0.8
0.9 Ϯ 0.1
3.3 Ϯ 0.3
0.8 Ϯ 0.1
4.6 Ϯ 1.4
1.0 Ϯ 0.1
3.6 Ϯ 0.4
4.0 Ϯ 1.2
0.108 Ϯ 0.009
0.071 Ϯ 0.011
0.089 Ϯ 0.027
0.063 Ϯ 0.006
0.093 Ϯ 0.006
0.062 Ϯ 0.008
0.099 Ϯ 0.019
0.064 Ϯ 0.007
0.081 Ϯ 0.008
0.062 Ϯ 0.011
0.110 Ϯ 0.030
0.060 Ϯ 0.006
0.056 Ϯ 0.006
0.054 Ϯ 0.008
0.063 Ϯ 0.019
0.051 Ϯ 0.007
0.093 Ϯ 0.009
0.058 Ϯ 0.008
0.090 Ϯ 0.027
0.064 Ϯ 0.006
0.075 Ϯ 0.008
0.060 Ϯ 0.018
22.3
3.90Ϫ13.90
13.8
25.6
4.00Ϫ13.10
3.70Ϫ13.00
41.0
27.6
dmbpy (2b)
18.5
28.0
4.40Ϫ13.40
3.60Ϫ14.90
47.8
21.8
[24]
bpy
(2c)
14.5
27.6
3.60Ϫ11.30
3.70Ϫ12.50
23.5
23.7
bpym (2d)
13.5
25.1
4.00Ϫ12.50
3.70Ϫ15.00
30.6
25.5
bpz (2e)
16.5
25.6
3.70Ϫ14.00
3.80Ϫ14.00
30.6
29.9
[(terpy)Ni(Mes)]Br (2h)
[(N∧N)Ni(Mes)2]
tmphen (3a)
NiϪC/N
NiϪC
NiϪC/N
NiϪC
NiϪC/N
NiϪC
1.91 Ϯ 0.02
2.81 Ϯ 0.03
1.94 Ϯ 0.02
2.85 Ϯ 0.03
2.01 Ϯ 0.02
2.90 Ϯ 0.03
3.5 Ϯ 0.4
3.6 Ϯ 1.1
3.7 Ϯ 0.4
5.3 Ϯ 1.6
4.4 Ϯ 0.4
4.3 Ϯ 1.6
0.083 Ϯ 0.008
0.095 Ϯ 0.029
0.087 Ϯ 0.009
0.111 Ϯ 0.033
0.104 Ϯ 0.009
0.107 Ϯ 0.033
29.5
28.9
21.8
3.80Ϫ11.90
3.80Ϫ11.50
3.40Ϫ11.00
36.7
32.7
36.2
bpy (3c)
bpz (3e)
[a]
AbsorberϪbackscatterer distance r, coordination number N, DebyeϪWaller factor σ, with their calculated deviations, shift of the
threshold energy ∆E0 and the fit-index R.
Eur. J. Inorg. Chem. 2004, 2784Ϫ2796
www.eurjic.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2787

´ ˇ
A. Klein, M. P. Feth, M. Bertagnolli, S. Zalis
FULL PAPER
Figure 3. Experimental (solid line) and calculated (dotted line) k3 χ(k) functions (a,c) and their Fourier transforms (b,d) of solid [(bpz)Ni-
(Mes)Br] (2e) at the Ni-K (a,b) and Br-K edge (c,d)
Figure 4. Comparison of the experimental k3 χ(k) functions (a) and their Fourier transforms (b) of solid [(tmphen)Ni(Mes)2] (3a) (solid
line), solid [(bpy)Ni(Mes)2] (3c) (dotted line) and solid [(bpz)Ni(Mes)2] (3e) (dashed line) (Ni-K edge)
of 3e shows a clear asymmetry, indicating longer and and this gives rise to the lengthening of the NiϪN bonds.
shorter bonded backscatterers in the local environment of The consequences of the basicity of the various ligands for
the Ni central atoms. However, a fit of two separate coordi- the preparation of the complexes has already been discussed.
nation shells did not lead to a satisfying result, thus only For the bromomesityl complexes this trend is reversed — the
one coordination shell for the direct environment of nickel longest average distances are found for the tmphen deriva-
˚
was fitted. The determined NiϪC/N distance of 2.01 A for tive. This can be explained in the following way. Assuming
this shell is significantly longer than in the other dimesityl that the NiϪC and NiϪN bond trans to Mes are essentially
complexes, such as in [(bpy)Ni(Mes)2] (3c) (NiϪC/N dis- the same as for the dimesityl complexes, and that the NiϪBr
˚
tance of 1.94 A). Comparing the bpy complex with the distances are more or less constant, the length of the NiϪN
tmphen derivative reveals slightly longer averaged distances bond trans to Br must be the reason for the observed trend.
for the former. In this series of ligands tmphen represents Br is a π-donor which is best accommodated by a good π-
the high-basicity end and bpy is therefore less basic. Bpz is acceptor diimine ligand and tmphen in that sense is the worst
a much better π-acceptor but therefore a much weaker base, in our series, whereas bpz is the best.
2788
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2004, 2784Ϫ2796

Mesitylnickel Complexes of α-Diimines
FULL PAPER
Optical Spectroscopy and Calculations
In agreement with recent work by Yamamoto[21] we can
assign the two long-wavelength bands to charge-transfer
transitions (CT) to the π* orbitals of the diimine ligands
and the UV bands to ligand-centred π-π* transitions. In
view of the observed solvatochromism of the extremely
long-wavelength bands an assignment of these bands to me-
tal-centred (LF) transitions is unlikely.
A closer look at the solvatochromism data (Table 4) re-
veals that the extent, as derived from the differences of the
observed maxima in pentane and DMF solution, increases
with the absolute energy, which parallels the series of in-
creasing basicity (or decreasing π-acceptor character;
Table 3). Generally the bromomesityl complexes exhibit
stronger solvatochromism than the dimesityl derivatives
and there is no marked difference between the dimethyl
complex [(bpy)Ni(Me)2] and its mesityl analogue. Extreme
values are found for 2a, underlining the special character of
the tmphen ligand mentioned above, and for 3e. All these
findings are in full agreement with the simple model of a
more or less strong polarisation M(δϪ)ǟL(δϩ) in the ground
state, which is transformed to a rather apolar MǟL excited
state by an MLCT excitation.[28] Strongly basic ligands fav-
our more pronounced polarisation, whereas the donating
Mes co-ligands prevent the negative polarisation of the me-
tal by the diimine ligand.
Absorption Spectroscopy and Calculations
Both dimesityl and bromomesityl nickel complexes are
brightly coloured, ranging from orange to green in the solid
as well as in solution. The absorption spectra (see Figure 5)
are dominated by two broad bands of medium intensity in
the visible region and an intense band in the UV region.
Additional weak, very-long-wavelength bands appear as
shoulders in the near IR (NIR) region which, in most cases,
are only clearly visible in apolar solvents (Table 3). The two
broad bands show a strong negative solvatochromism (see
for example Table 4Ϫ7) they are red-shifted for the dimes-
ityl complexes relative to their bromomesityl analogues, and
in apolar solvents they are partly structured. The UV band
is solvent invariant and appears to be characteristic of the
ligand.
Recently we have discussed the occurrence of mixed
MLCT/LЈLCT (ligand-to-ligand charge transfer) tran-
sitions (instead of pure MLCT transitions) due to contri-
butions of the co-ligands to the frontier orbitals in related
diarylplatinum complexes.[29] To probe for such contri-
butions from the mesityl or the bromine co-ligands we have
examined the dimesitylnickel complex, the dimethylnickel
complex and the bromo(mesityl)nickel complex of 2,2Ј-bi-
pyridine in a combined spectroscopic and quantum chemi-
cal study.
Figure 5. Absorption spectra of [(bpy)Ni(Me)2] (solid line), [(bpy)-
Ni(Mes)2] (dotted line) and [(bpy)Ni(Mes)Br] (dashed) in toluene
solution; the insert depicts the spectral deconvolution for [(bpy)-
Ni(Me)2]
First of all the molecular structures were calculated by
both ADF/BP and G03/B3LYP methods. Both methods re-
Table 3. Absorption maxima of the nickel complexes in toluene
[(N∧N)Ni(Mes)Br]
(N∧N)
λ (ε in 1000 Ϫ1·cmϪ1 [a]
)
λ1
λ2
λ3
λ4
λ5
tmphen (2a)
terpy (2h)
dmbpy (2b)
bpy (2c)
bpym (2d)
bpz (2e)
300
327, 343
341, 365 sh
344
342 (2.95)
352 sh
368 sh
361
430 sh, 452
427, 453
456 sh, 486 sh
448 sh, 483 sh
490 sh, 507 sh
402 sh, 475
427
485
487
507
516 (3.01)
528
564, 653 sh
559 sh, 590
632, 650 sh
527 sh
531 sh
555 sh
571 (1.71)
592 sh
790
729 sh
697 sh, 728 sh
300
304 (13.9)
293
329
Ϫ
iPr-DAB (2g)
bpm (2f)
Ϫ
356, 401 sh
538 sh, 610 sh
[(N∧N)NiMes2]
(N∧N)
tmphen (3a)
bpy (3c)
bpz (3e)
[(bpy)NiMe2]
297sh
307 (14.0)
333
355, 378 sh
377 (3.26)
408
505 sh, 543,
564
611 (3.11)
662
531 sh, 587 sh
462 sh, 628 sh
582 sh, 631 sh
710 sh, 785 sh
785 sh, 870 sh
294 (15.9)
416 (2.81)
678 (2.75)
[a]
Absorption maxima, λ, in nm; main maxima are in italics.
www.eurjic.org
Eur. J. Inorg. Chem. 2004, 2784Ϫ2796
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2789

´ ˇ
A. Klein, M. P. Feth, M. Bertagnolli, S. Zalis
FULL PAPER
Table 4. Long-wavelength absorption bands of selected nickel complexes in various solvents — solvatochromism[a]
[(N∧N)Ni(Mes)Br]
(N∧N)
λ (in nm)
THF (0.59)
ν˜ (in cmϪ1
)
CH2Cl2 (0.67)
pentane (E* ϭ 0)[b]
DMF (0.95)
∆ (cmϪ1 [c]
)
tmphen (2a)
bpy (2c)
518 (19300)
558 (17930)
594 (16850)
615 (16250)
494 (20240)
469 (21310)
500 (20000)
545 (18360)
572 (17490)
450 (22220)
423 (23630)
490 (20410)
547 (18270)
569 (17580)
460 (21760)
409 (24435)
464 (21530)
526 (19030)
547 (18270)
453 (22090)
5135
3600
2180
2020
1850
bpz (2e)
iPr-DAB (2g)
[(terpy)Ni(Mes)]ϩ
[(N∧N)Ni(Mes)2]
(N∧N)
tmphen (3a)
bpy (3c)
bpz (3e)
pentane
THF
CH2Cl2
DMF
∆ (cmϪ1 [c]
2440
2360
)
598 (16720)
645 (15500)
681 (14670)
546 (18330)
584 (17120)
642 (15580)
531 (18830)
576 (17350)
636 (15720)
522 (19160)
560 (17860)
629 (15890)
1220
toluene (0.30)
683 (14640)
608 (16450)
516 (19380)
THF
Ϫ
584 (17120)
500 (20000)
CH2Cl2
Ϫ
576 (17350)
490 (20410)
DMF
∆ (cmϪ1 [c]
)
1610
1445
2150
[(bpy)Ni(Me)2]
[(bpy)Ni(Mes)2]
[(bpy)Ni(Mes)Br]
615 (16250)
560 (17860)
464 (21530)
[a] Absorption maxima, λ, in nm; wavenumbers (in cmϪ1) in parentheses.[b] E* values are Manuta and Lees’ solvent parameters for charge-
transfer transitions.[27] [c] ∆ ϭ E(dmf) Ϫ E(pentane), or E(toluene) in cmϪ1. A complete table is provided with the Supporting Information.
Table 5. Selected calculated lowest TD DFT singlet excitation energies for [(bpy)Ni(Mes)2]
ADF/SAOP
Trans. energy in
eV (nm)
Experiment
Main character (%)
Osc.
strength
λ (nm) in
λ (nm) in
toluene (ε)[a]
DMF[a]
1A
1A
1A
1B
1A
1A
1B
1A
95 (58b Ǟ 59b)
1.57 (790)
2.18 (568)
2.34 (530)
2.54 (488)
2.65 (468)
2.89 (429)
2.97 (417)
3.08 (402)
0.004
0.026
0.005
0.004
0.006
0.004
0.012
0.130
789 (200)
745
554
535
493
429
388
362
306/297
38 (57b Ǟ 59b); 37 (56b Ǟ 59b); 10 (58b Ǟ 60b)
88 (58b Ǟ 60b); 3 (57b Ǟ 59b)
87 (58b Ǟ 63a); 8 (60a Ǟ 60b)
62 (57b Ǟ 60b); 31 (56b Ǟ 60b)
97 (59a Ǟ 63a);
70 (56b Ǟ 63a); 26 (57b Ǟ 63a)
53 (60a Ǟ 63a); 15 (57b Ǟ 60b);
10 (56b Ǟ 60b); 7 (57b Ǟ 59b);
611 (2610)
575 (1990)
548 (1420)
445 (1200)
410 (2240)
374 (3000)
307 (14000)
[a]
Experimental values from spectral deconvolution. Extinction coefficients, ε, in Ϫ1·cmϪ1
.
Table 6. Selected calculated lowest TD DFT singlet excitation energies for [(bpy)Ni(Me)2]
ADF/SAOP
Trans. energy in
eV (nm)
G03/B3LYP
Trans. energy in
eV (nm)
Experiment
Main character (%)
Osc.
strength
Osc.
strength
λ (nm) in
λ (nm) in
toluene (ε)[a]
DMF[a]
1B2
1A1
99 (5a2 Ǟ 6b1);
72 (7b1 Ǟ 8b1); 17 (7b1 Ǟ 9b1);
10 (5a2 Ǟ 6a2)
97 (5a2 Ǟ 9b1); 2 (7b1 Ǟ 6a2)
96 (7b1 Ǟ 6a2) 2 (5a2 Ǟ 9b1);
48 (5a2 Ǟ 6a2); 24 (7b1 Ǟ 8b1);
24 (7b1 Ǟ 9b1)
1.63 (760)
1.87 (663)
0.006
0.017
1.64 (756)
2.24 (553)
0.002
0.048
785 (380)
678 (4110)
760
618
1B2
1B2
1A1
2.02 (614)
2.56 (484)
2.78 (446)
0.006
0.026
0.125
2.75 (451)
3.08 (402)
3.34 (371)
0.008
0.029
0.119
631(3300)
461 (2830)
416 (4280)
581
434
392
[a]
Experimental values from spectral deconvolution. Extinction coefficients, ε, in Ϫ1·cmϪ1
.
produce the experimentally obtained bond lengths and for the dimesityl derivative. For the bromomesityl complex
angles well. The biggest difference between the two methods ADF/BP also gave a far more accurate NiϪBr distance
˚
was observed for the NiϪC bond, where ADF/BP gave a (2.324 A) compared with the experimental value of 2.300(1)
˚
˚
˚
˚
value of 1.930 A and G03/B3LYP yielded 1.906 A for the A (G03/BLYP gave 2.399 A). Comparing the experimental
dimethyl complex compared to the experimental value data of the dimethyl derivative [(bpy)Ni(Me)2] (monoclinic
[30]
Approximately the same was found P2/c, Z ϭ 4)[30] with all the so far structurally characterised
˚
of 1.923(4) A.
2790
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2004, 2784Ϫ2796

Mesitylnickel Complexes of α-Diimines
FULL PAPER
Table 7. Selected calculated lowest TD DFT singlet excitation energies for [(bpy)Ni(Mes)Br]
ADF/SAOP
G03/B3LYP
Experiment
Main character (%)
Trans. energy in
Osc.
strength
Trans. energy in
eV (nm)
Osc.
strength
λ (nm) in
λ (nm) in
eV (nm)
toluene (ε)[a]
DMF[a]
1A
1A
1A
1A
1A
48 (104a Ǟ 106a); 30 (103a Ǟ 106a)
89 (102a Ǟ 106a); 6 (104a Ǟ 106a)
83 (100a Ǟ 106a); 5 (104a Ǟ 107a)
52 (103a Ǟ 107a); 36 (104a Ǟ 107a)
43 (104a Ǟ 108a); 31 (103a Ǟ 108a);
20 (99a Ǟ 106a);
36 (100a Ǟ 107a); 36 (99a Ǟ 106a)
53 (100a Ǟ 107a); 15 (99a Ǟ 106a)
68 (105a Ǟ 109a); 10 (103a Ǟ 109a)
70 (100a Ǟ 108a); 10 (99a Ǟ 107a)
34 (103a Ǟ 109a); 30 (102a Ǟ 109a);
18 (99a Ǟ 107a); 8 (104a Ǟ 109a)
1.64 (756)
1.69 (734)
2.13 (582)
2.31 (537)
2.56 (484)
0.015
0.004
0.022
0.005
0.013
1.82 (682)
1.85 (669)
2.04 (607)
2.29 (540)
2.46 (502)
0.001
0.001
0.025
0.001
0.012
675 (80)
Ϫ
613 (270)
558 (1420)
539 (50)
555
522
496
473
512 (2710)
1A
1A
1A
1A
1A
2.71 (457)
2.82 (440)
2.89 (429)
3.14 (395)
3.44 (360)
0.010
0.025
0.010
0.038
0.030
2.60 (476)
2.72 (456)
Ϫ
3.01 (412)
Ϫ
0.040
0.004
Ϫ
0.009
Ϫ
477 (1830)
449 (960)
418 (900)
380 (1430)
340 (2740)
447
427
Ϫ
349
325
[a]
Experimental values from spectral deconvolution. Extinction coefficients, ε, in Ϫ1·cmϪ1
.
mesitylnickel complexes reveals markedly shorter NiϪC
bonds for the latter. This is in contrast to what would be
expected on steric grounds. The CϪNiϪC angles in the di-
mesityl derivatives are slightly enlarged to more than 90°
(86.6° for the dimethyl derivative). The NϪNiϪN chelate
bite and the NiϪN bonds, however, do not differ. As a
consequence, it seems that in [(bpy)Ni(Me)2] the inter-ring
˚
distance between the pyridine rings is elongated (1.485 A)
which normally means reduced back-bonding. ADF/BP
gives a value of 74.4° for the N1ϪNiϪC11ϪC12 tilt angle
of the mesityl co-ligands (74.9° by G03/B3LYP) and we feel
that from these structural data there is already evidence for
a notable co-ligand (Mes) contribution to the electronic
transitions. The Mes co-ligand therein could act either as
a π-donor or π-acceptor. Furthermore, from the EXAFS
measurements we have evidence of a strong π-donating in-
teraction of the bromine co-ligand with the nickel atom.
Both types of co-ligand contributions will be substantiated
by a view of the calculated frontier orbitals. The ground-
Figure 6. Qualitative MO schemes [(bpy)Ni(Me)2], [(bpy)Ni(Mes)2]
(3c) and [(bpy)Ni(Mes)Br] (2c) based on ADF/SAOP calculated
one-electron energies and percentage composition of selected fron-
tier molecular orbitals
state one-electron energies and percentage composition of tially either. In the case of [(bpy)Ni(Mes)Br] (2c) bromine p
selected highest occupied and lowest unoccupied molecular orbitals contribute to a large extent to the HOMOs whereas
orbitals have been calculated by DFT methods and the re- the contributions of the mesityl co-ligands are lower than
sults are shown qualitatively in Figure 6.
for the dimesityl derivative.
The labelling and relative position of ADF/SAOP calcu-
To gain insight to the electronic transitions, lowest TD
lated frontier molecular orbitals is depicted in Figure 6. The DFT singlet excitation energies have been calculated and
calculated composition is available in the Supporting Infor- compared to experimentally observed absorption maxima.
mation. The figure illustrates how the replacement of Mes The results are summarised in Tables 5Ϫ7. To this end the
by other co-ligands influences the composition of the fron- absorption spectra of the three compounds have been sub-
tier MOs. The character of the LUMO is only moderately mitted to a spectral deconvolution, as shown in Figure 5.
influenced by the variation of the co-ligand R.
However, the results of the deconvolutions were not unam-
More substantial changes are observed in the compo- biguous since both multiple electronic transitions and vi-
sition of the set of HOMOs which, in the case of [(bpy)- brational progression could contribute to the various max-
Ni(Mes)2] (3c), have large contributions of the Mes co-li- ima observed as shoulders in the spectra.
gand. This is different for [(bpy)Ni(Me)2], where the lower
All calculated transitions have mixed character. They can
lying occupied orbitals are formed by Me sp3 orbitals inter- be characterised as MLCT excitations for [(bpy)Ni(Me)2],
acting with central metal orbitals. Both ADF/BP and DFT/ whereas for [(bpy)Ni(Mes)2] (3c) and [(bpy)Ni(Mes)Br] (2c)
B3LYP give qualitatively the same composition for the the contribution from the Mes or Br co-ligands cannot be
HOMO and LUMO of [(bpy)Ni(Me)2], and the compo- neglected and the corresponding transitions should be as-
sition of the other frontier orbitals does not differ substan- signed as mixtures of MLCT/LЈLCT [ligand (Mes)-to-li-
Eur. J. Inorg. Chem. 2004, 2784Ϫ2796
www.eurjic.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2791

´ ˇ
A. Klein, M. P. Feth, M. Bertagnolli, S. Zalis
FULL PAPER
gand charge transfer] or MLCT/XLCT (halogen-to-ligand
charge transfer), respectively. Both co-ligands thus act as π-
donors. In the bromomesityl complex the donor character
of the mesityl group is lower than the dimesityl derivative.
The transition energies agree qualitatively quite well with
experimental ones for the dimesityl and dimethyl derivatives
(Table 5 and 6). Note that the solvatochromic effects are
enormous for most absorption bands. Generally, G03/
B3LYP predicts larger transition energies than ADF/SAOP,
as shown in Table 7. For 2c the ADF/SAOP calculations
predict very low-energy transitions around 1.64 eV
(750 nm), which cannot be found in the absorption spectra;
the TD-DFT method using ‘‘pure’’ functionals underesti-
mates the transitions energies also in the case of other hal-
ide complexes.[31] The transitions calculated at higher ener-
gies fit better to experimentally found ones; however the
intensity pattern is not reproduced well. G03 calculations
using hybrid B3LYP functionals give transitions with high
bromine contributions closer to the experimentally meas-
ured weak, low-lying absorption bands.
Figure 7. Resonance Raman spectra of the nickel complexes
[(bpy)Ni(Mes)2] (3c; λex ϭ 542 nm), [(bpy)Ni(Mes)Br] (2c; λex
ϭ
488 nm), [(bpz)Ni(Mes)Br] (2e; λex
ϭ
592 nm), [(iPr-
DAB)Ni(Mes)Br] (2g; λex ϭ 598 nm) and [(terpy)Ni(Mes)]Br (2h;
λ
ex ϭ 457.9 nm), obtained from KNO3 pellets upon irradiation into
the long-wavelength absorption bands; * denotes the 1051 cmϪ1
؊
Resonance Raman Spectroscopy
resonance of NO3
In many cases the character of the excited states of tran-
sition metal complexes has been studied by emission spec- respectively. For the other diimine ligands the assignment is
troscopy.[32] However, of the present nickel complexes only based on comparison with the bpy analogue.
2a, 2b, 2h and 3a show very weak emission in the solid or
At first sight (e.g. in Figure 7 for the two bpy complexes)
solution; the others do not emit either in the solid or in the rR spectra of the bromomesityl complexes of the aro-
solution, even at 110 K. Additionally, for the weakly emit- matic diimine ligands are much the same as those recorded
ting compounds the obtained results are ambiguous. We for the dimesityl complexes. For the iPr-DAB complex
therefore embarked on a combination of resonance Raman assignment of the observed resonances at 1542 cmϪ1
(rR) spectroscopy and quantum chemical calculations to ν(CϭN), or 1170, 1138 and 1020 cmϪ1 ν(CϪC) was facili-
probe the character of the excited states. The rR technique tated by comparison with recently examined organoplati-
allows us to draw direct conclusions about the character of num() and -platinum() complexes.[29b,36] A closer look at
the electronic transitions since it is based on the presump- the data reveals that in the dimesityl complexes some reso-
tion that only those Raman bands due to vibrations that nances are observed that do not appear for the bromomesityl
are influenced by the electronic transition that is excited are derivatives, for example for bpy: 1263, 1114 and 941 and
enhanced in intensity.[33] The rR spectra of the dimesityl 740 cmϪ1 (marked in italics in Table 8). Such resonances are
complexes and selected bromomesityl complexes dispersed characteristic of ν(CϭCring) or δring vibrations and could be
in KNO3 were recorded after irradiation of their lowest- assigned to the mesityl co-ligands. They appear only for the
energy absorption bands; examples are displayed in Fig- dimesityl complexes because the contribution of the mesityl
ure 7. Selected data are collected in Table 8 (see also Sup- co-ligands to the excited state is markedly higher (mixed
porting Information) in comparison to DFT calculated Ra- MLCT/LЈLCT) than for the bromomesityl complexes
man active vibrations for the model system [(bpy)Ni(Me)2] (mixed XLCT/MLCT) thus supporting the results from
and the resonances of free bpy.[34]
quantum chemical calculation (vide supra). The mentioned
Based on these calculations and data the resonances ob- resonances might also be interpreted as bpy-centred reso-
served for [(bpy)Ni(Mes)2] (3c) at 1608, 1561, 1485, 1322, nances and their occurrence explained by the differences in
1277, 1167, 1015, 655 and 370 cmϪ1 can be assigned to the symmetry between dimesityl (C2) and bromomesityl (Cs).
normal modes of the bpy ligand. They have been observed However, it is unlikely that the higher-symmetry molecules
previously in rR experiments with a number of bpy com- exhibit more resonances than the lower-symmetry ones.
plexes upon excitation into their metal-to-ligand(bpy)
Support for the partial XLCT character of the excited
charge-transfer bands.[35] The resonance at highest energy states in the bromomesityl complexes could not be obtained
is assigned to the ν(CϭN) vibration. Its energy decreases from rR spectra. The only relevant vibration, ν(Ni-Br),
along the series tmphen Ͼ bpy Ͼ bpz which is caused by which occurs around 160Ϫ170 cmϪ1, is quite weak and
the increasing π-accepting abilities from tmphen to bpz. therefore hard to detect in the rR experiment due to the
Resonances at 426, 343 and 241 cmϪ1 were assigned to the closeness to the excitation line. Furthermore, it is also pos-
metal-ligand modes ν(NiϪC), ν(NiϪN) and δ(NiϪN), sible that the NiϪBr σ-bond is not much distorted upon π-
2792
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2004, 2784Ϫ2796

Mesitylnickel Complexes of α-Diimines
FULL PAPER
Table 8. Selected resonance Raman data of [(bpy)Ni(Mes)2] (3c), [(bpy)Ni(Mes)Br] (2c) and [(terpy)Ni(Mes)]Br (2h) compared to calcu-
lated [(bpy)Ni(Mes)2] and the free ligand[a]
3c
2c
2h
Calculated[b]
bpy[c]
Assignment
ν˜ (cmϪ1
542 nm
)
ν˜ (cmϪ1
488 nm
)
ν˜ (cmϪ1
)
ν˜ (cmϪ1
)
ν˜ (cmϪ1
)
λex
457.9 nm
1
2
3
4
5
6
7
8
1608
1561
1485
Ϫ[c]
1322
1277
1263
1167
1114
1015
941
740
655
565
426
370
343
241
1603
1565
1490
Ϫ[c]
1325
1280
Ϫ[c]
1608
1577/1561
1503/1475
1408
1336
1299
Ϫ[c]
1590
1549
1463
1414
1296
1273
Ϫ
1152
Ϫ
1047
Ϫ
1579
1553
1448
1401
1270
1248
Ϫ
1210
1138
1090
Ϫ
ν(CϭCring) ϩ ν(CϭNring
ν(CϭCring) ϩ ν(CϭNring
ν(CϭCring)/δ(Η)
)
)
ν(CϭCring)/δ(Η)
ν(CϪCinterring
ν(CϭNring) ϩ ν(CϭCring
ν(CϭCring
)
)
)
1177
1172
δ(Η)/ν(CϭCring
δ(Η)/ν(CϭCring
)
)
9
Ϫ[c]
Ϫ[c]
10
11
12
13
14
15
16
17
18
1020
Ϫ[c]
Ϫ[c]
770
650
564
376
n.m.
n.m.
1023
Ϫ[c]
Ϫ[c]
676
654
δ(Η)/ring breathe
δ(ring)
δ(ring)
δ(ring bend)
ν(NiϪC)
δ(NϪNiϪN)out-of-plane
ν(NiϪN) ϩ δ(CϪNiϪC)
ν(NiϪN)
δ(NϪNiϪN)
Ϫ
Ϫ
643
542
421
383
358
224
651
618
Ϫ
398
Ϫ
557
n.m.
n.m.
n.m.
Ϫ
[a]
From resonance Raman experiments on KNO3 pellets at ambient temperature. Excitation wavelengths, λex, coincident with the long-
[b]
wavelength absorption maximum.
not measured; resonances of 3c marked in italics are assigned to mesityl vibrations (see text).
Calculated for [(bpy)Ni(Mes)2] by G03/B3LYP; scaling factor 0.9613.[37] [c] Not observed. n.m. ϭ
on Bruins Instruments Omega 10, HewlettϪPackard 8453 Diode
donation by the bromine atom and therefore not res-
onantly enhanced.
Array and Varian Cary 4E spectrophotometers. Resonance Raman
spectra of the complexes dispersed in KNO3 pellets were recorded
on a Dilor XY spectrometer equipped with a Wright Instruments
CCD detector, using a Spectra Physics 2040E Arϩ and Coherent
CR490 and CR590 dye lasers (with Coumarin 6 and Rhodamine
6G dyes) as excitation sources. Emission spectra were recorded on
a PerkinϪElmer LS5B spectrophotometer.
Conclusion
Quantum chemical calculations have revealed the inter-
play of the metal centre, the chelating diimine ligand and
the co-ligands methyl, mesityl or bromine to form the fron-
tier orbitals relevant for the optical transitions and excited
states. Strong experimental evidence for the π-accepting
mode of the chelate ligand and the π-donating modes of
Computational Details: Ground-state electronic structure calcu-
lations on complexes [(bpy)Ni(R)2] and [(bpy)Ni(Mes)Br] (2c) have
been done on the base of density-functional theory (DFT) methods
using the ADF2002.03[38] and Gaussian 03[39] program packages.
the mesityl and bromine co-ligands comes from a combi- The lowest excited states of the closed-shell complexes were calcu-
lated by the time-dependent DFT (TD DFT) method (both ADF
and G03 programs).
nation of spectroscopic methods. A structural study using
EXAFS spectroscopy shows some clear trends for the
metalϪligand distances that can be explained by varying σ-
donor and/or π-acceptor ability through the series of diim-
ine ligands and π-donating contributions from the bromine
co-ligand. Support for a specific electronic contribution
from the mesityl co-ligand comes from XRD results. The
absorption spectra agree qualitatively well with the calcu-
lated transitions. Resonance Raman spectroscopy strongly
Within the ADF program, Slater-type orbital (STO) basis sets of
triple-ζ quality with polarisation functions were employed; methyl
groups on the Mes co-ligands were described by basis sets of
double-ζ quality with polarisation functions. The inner shells were
represented by a frozen core approximation (1s for C, N, 1s-3p
for Br and 1s-3p for Ni were kept frozen). The following density
functionals were used within ADF: the local density approximation
(LDA) with VWN parametrisation of electron gas data or the func-
supports the π-accepting character of the diimine ligands. tional including Becke’s gradient correction[40] to the local-ex-
change expression in conjunction with Perdew’s gradient correc-
tion[41] to the LDA expression (ADF/BP). The scalar relativistic
(SR) zero-order regular approximation (ZORA) was used within
this study. Compositions and energies of molecular orbitals and
electronic transition energies and compositions were calculated by
the asymptotically correct SAOP functional,[42] which is suitable
also for higher-lying MOs and electronic transitions. Core electrons
were included in ADF/SAOP calculations.
The evidence for the co-ligand contributions is less strong
and should be further substantiated by rR studies on struc-
turally varied derivatives (H/D exchange, other aryl co-li-
gands, Cl/Br/I exchange). Such studies are already un-
derway.
Experimental Section
Instrumentation: 1H NMR spectra were recorded on Bruker AC250
Within Gaussian 03 6-31G* polarized double-ζ basis sets[43] for C,
N, H and Ni atoms; effective core pseudopotentials with corre-
or AM200 spectrometers. UV/Vis absorption spectra were recorded sponding optimised set of basis functions for Br atom[44] were used
Eur. J. Inorg. Chem. 2004, 2784Ϫ2796
www.eurjic.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2793

´ ˇ
A. Klein, M. P. Feth, M. Bertagnolli, S. Zalis
FULL PAPER
for geometry optimisation and frequency calculations. Correlation
in the case of the Br-K edge. The solid complexes were embedded
in a polyethylene matrix and pressed to pellets. The concentration
consistent polarized double-ζ basis sets[45] (cc-pVDZ) within TD-
DFT calculations. Hybrid Becke’s three parameter functional with of all samples was adjusted to yield an absorption jump, ∆µd, of
Lee, Yang and Parr correlation functional (B3LYP)[46] was used in
Gaussian 03 calculations (G03/B3LYP). Gaussian 03 was also used
for the calculations of the vibrations at G03/B3LYP optimised geo-
metries.
The calculations on [(bpy)Ni(Me)2] and [(bpy)Ni(Mes)2] (3c) were
performed in a C2v or C2 constrained symmetry, respectively. The
z axis is coincident with the C2 symmetry axis; the central atom
and bpy ligand lie in the y,z plane. All results discussed correspond
to optimised geometries. The calculations on [(bpy)Ni(Mes)Br]
were done without any symmetry constrains.
about 1.5.
Data evaluation started with background absorption removal from
the experimental absorption spectrum by subtraction of a Victo-
reen-type polynomial. Then, the background-subtracted spectrum
was convoluted with a series of increasingly broader Gauss func-
tions and the common intersection point of the convoluted spectra
was taken as energy E0.[48] To determine the smooth part of the
spectrum, corrected for pre-edge absorption, a piecewise poly-
nomial was used. It was adjusted in such a way that the low-R
components of the resulting Fourier transform were minimal. After
division of the background-subtracted spectrum by its smooth part,
the photon energy was converted into photoelectron wavenumbers
k. The resulting EXAFS function was weighted with k3. Data
analysis in k-space was performed according to the curved wave
multiple scattering formalism of the program EXCURV92 with
XALPHA phase and amplitude functions.[49] The mean-free path
of the scattered electrons was calculated from the imaginary part
of the potential (VPI was set to Ϫ4.00) and an overall energy shift
(∆E0) was assumed. The Amplitude Reduction Factor (AFAC) was
set to a value of 0.8 in the case of the Ni-K as well as the Br-K edge.
Crystal Structure Analysis for [(tmphen)Ni(Mes)2] (3a): Data collec-
tion was performed at T ϭ 173(2) K on a Siemens P4 dif-
˚
fractometer with Mo-Kα radiation (λ ϭ 0.71073 A). The structure
was solved by direct methods using the SHELXTL-PLUS pack-
age[47] and refinement was carried out with SHELXL-97 employing
full-matrix least-squares methods on F2 with F2o ϭ Ϫ3σ(F2o). For-
˚
mula (MW) C34H38N2Ni1 (533.37); cell dimensions (A): a ϭ
3
˚
16.297(3), b ϭ 16.578(3), c ϭ 20.732(4); V ؍ 5601.3(19) A ; Z ϭ
8; Abs. coefficient ϭ 0.718 mmϪ1; F(000) ϭ 2272; GooF at F2 ϭ
1.174; R values [I Ͼ 2σ(I)]: R1 ϭ 0.0801, wR2 ϭ 0.1679; R values
(all data): R1 ϭ 0.164, wR2 ϭ 0.2076; max./min. 0.609/Ϫ0.508
Materials and Procedures: The precursor complex [(PPh3)2Ni-
(Mes)Br] (1)
[23]
and the ligands bpm,[50] bpz,[51] and iPr-DAB[52]
e·AϪ3. Empirical absorption correction was performed using Ψ-
˚
were obtained following literature procedures. Mesityllithium was
prepared according to the method of Biali et al.[53] Other reagents
were commercially available and used without further purification.
All preparations and physical measurements were carried out in
dried solvents under an argon atmosphere, using Schlenk tech-
niques.
scans. All non-hydrogen atoms were treated anisotropically, hydro-
gen atoms were included by using appropriate riding models.
CCDC-226909 (for 3a) contains the supplementary crystallo-
graphic data for this paper. These data can be obtained free of
charge at www.ccdc.cam.ac.uk/conts/retrieving.html [or from
the Cambridge Crystallographic Data Centre, 12 Union Road,
Cambridge CB2 1EZ, UK; Fax: ϩ44-1223-336-033; E-mail:
deposit@ccdc.cam.ac.uk].
Preparation of [(N∧N)Ni(Mes)Br] (2) [N∧N ؍ tmphen (a), dmbpy
(b), bpy (c), bpym (d), bpz (e) bpm (f), and terpy (h)]: In a typical
reaction 2 g of [(PPh3)2Ni(Mes)Br] (1) (2.56 mmol) were suspended
in 250 mL of toluene, 1.1 equivalents of the ligand was added and
the solution stirred overnight at ambient temperature. The reaction
volume was then reduced to 20 mL and the resulting solids were
filtered off. They were washed with small amounts of toluene (2 ϫ
5 mL) and n-pentane (3 ϫ 20 mL) and dried. For bpz and bpm the
products were recrystallised twice from a mixture of CH2Cl2 and
n-heptane (1:8). From the filtrates only small amounts of clean
products could be isolated after evaporation to dryness and recrys-
tallisation from CH2Cl2/n-heptane (1:8). For analytical data see
Table 1 and Supporting Information.
EXAFS Measurements: The XANES and EXAFS measurements
were performed at the beamlines X1.1 (RÖMO II) and E4 at the
Hamburger Synchrotronstrahlungslabor des Deutschen Elek-
tronensynchrotrons (HASYLAB at DESY, Hamburg, Germany),
at beamline 2Ϫ3 at the Stanford Synchrotron Radiation Labora-
tory (SSRL) at SLAC (Stanford, USA) or at the newly designed
EXAFS-beamline KMC-2 at the Berliner Elektronenspeicherring-
Gesellschaft für Synchrotronstrahlung m.b.H (BESSY II, Berlin,
Germany).
For the measurements at the Ni-K edge (8332.8 eV) a Si(111)
double crystal monochromator was used at the SSRL and at
DESY. For the measurements at the Br-K edge (13474.0 eV) at be-
amline X1.1 a Si(311) double crystal monochromator was used. At
BESSY a SiGe(220) double graded-crystal (0.5% Ge/cm) mono-
chromator was used for the nickel and the bromine edge. The
synchrotron beam current was between 80 and 100 mA at HASY-
LAB (positron energy 4.45 GeV), between 100 and 250 mA at
BESSY (electron energy 1.7 GeV) and between 80 and 100 mA at
the SSRL (electron energy 1.7 GeV).
Preparation of Bromo-1,4-diisopropyl-1,4-diazabutadienemesityl-
nickel(II), [(iPr-DAB)Ni(Mes)Br] (2g): iPr-DAB (250 mg,
1.78 mmol) was added to a vigorously stirred suspension of
[(PPh3)2Ni(Mes)Br] (1; 250 mg, 0.32 mmol) in 500 mL of n-pen-
tane. After a few minutes the yellow suspension dissolved leaving
a violet voluminous precipitate that was finally filtered off. The
obtained violet solid was recrystallised from CH2Cl2/n-heptane
(1:8) and dried, yielding 112 mg of dark-violet microcrystalline ma-
terial. For nalytical data see Table 1 and Supporting Information.
All the experiments were carried out under ambient conditions at
25 °C. The tilt of the second monochromator crystal was set to 30%
harmonic rejection. Energy resolution was estimated to be about
Preparation of the Dimesitylnickel Complexes 3a, 3c, 3e: The dimes-
0.7Ϫ2.0 eV for the Ni-K edge and 4.0 eV for the Br-K edge. The itylnickel complexes [(N∧N)Ni(Mes)2] were prepared as described
spectra were collected in transmission mode with ion chambers. All
ion chambers were filled with nitrogen in the case of the measure-
ments at the Ni-K edge, and the second and third chamber with
argon in the case of the measurements at the Br-K edge. Energy
for the bpy complex[23] from the bromomesityl complexes and mes-
ityllithium.
3a: Yield 88%, dark violet powder. C34H38N2Ni1 (533.40): calcd. C
76.56, H 7.18, N 5.25; found C 76.59, H 7.23, N 5.27. 1H NMR
calibration was performed with nickel metal foil in the case of ([D6]DMSO): δ ϭ 8.26 (s, 2 H, H5,6), 7.64 (s, 2 H, H2,9), 6.45 (s,
measurements at the Ni-K edge and lead metal foil (Pb LIII-edge)
4 H, m-H), 2.67 (s, 6 H, H3C4,7), 2.64 (s, 12 H, o-CH3), 2.29 (s, 6
2794
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2004, 2784Ϫ2796

Mesitylnickel Complexes of α-Diimines
FULL PAPER
[11]
[12]
[13]
[14]
[15]
H, H3C3,8), 2.13 (s, 6 H, p-CH3) ppm. 13C NMR (CD2Cl2): δ ϭ
157.2, 150.5, 142.6, 127.3, 122.7 (all tmphen); 144.9 (Mes-1), 144.1
(Mes-2), 130.9 (Mes-4), 125.3 (Mes-3), 27.3 (o-CH3), 20.8 (p-CH3),
18.1 (H3C4,7), 14.9 (H3C3,8) ppm.
P. T. Matsanuga, J. C. Mavrapoulos, G. L. Hillhouse, Poly-
hedron 1995, 14, 175.
P. T. Matsanuga, C. R. Hess, G. L. Hillhouse, J. Am. Chem.
Soc. 1994, 116, 3665.
P. T. Matsanuga, G. L. Hillhouse, J. Am. Chem. Soc. 1993,
115, 2075.
P. Binger, M. J. Doyle, C. Krüger, Y.-H. Tsay, Z. Naturforsch.,
Teil B 1979, 34, 1289.
T. Yamamoto, A. Yamamoto, S. Ikeda, J. Am. Chem. Soc.
1971, 93, 3350.
3c: Yield 84%, dark violet powder. C28H30N2Ni1 (453.27): calcd. C
74.20, H 6.67, N 6.18; found C 74.35, H 6.65, N 6.22. 1H NMR
3
([D6]acetone): δ ϭ 8.38 (d, JH3,H4 ϭ 8.02 Hz, 2 H, H3,3Ј), 8.12
3
(dd,3JH4,H5 ϭ 6.90 Hz, 2 H, H4,4), 7.54 (d, JH5,H6 ϭ 4.84 Hz, 2
H, H6,6Ј), 7.44 (dd, 2 H, H5,5Ј), 6.37 (s, 4 H, m-H), 2.53 (s, 12 H,
o-CH3), 2.08 (s, 6 H, p-CH3) ppm. 13C NMR ([D6]acetone): δ ϭ
157.4 (bpy-2,2Ј), 149.9 (bpy-6,6Ј), 144.1 (Mes-2), 138.1 (bpy-4,4Ј),
136.6 (Mes-1), 130.9 (Mes-4), 127.6 (bpy-3,3Ј), 126.0 (Mes-3), 122.0
(bpy-5,5Ј), 27.3 (o-CH3), 20.8 (p-CH3) ppm.
[16]
[17]
W. Seidel, Z. Chem. 1985, 25, 411.
G. Wilke, G. Herrmann, Angew. Chem. 1966, 78, 591. G. Wilke,
G. Herrmann, Angew. Chem. Int. Ed. Engl. 1966, 5, 581.
T. R. Miller, I. G. Dance, J. Am. Chem. Soc. 1973, 95, 6970.
A. Arcas, P. Royo, Inorg. Chim. Acta 1978, 31, 97.
R. F. de Souza, L.C. Simon, M.C. Alves, J. Catal. 2003, 214,
165.
[18]
[19]
[20]
3e: Yield 78%, dark green microcrystalline powder. C26H28N4Ni1
(455.25): calcd. C 68.60, H 6.20, N 12.31; found C 68.72, H 6.22,
1
N 12.41. H NMR ([D6]acetone): δ ϭ 9.54 (s, 2 H, H3,3Ј), 8.75 (s,
[21]
T. Yamamoto, M. Abla, Y. Murakami, Bull. Chem. Soc. Jpn.
2002, 75, 1997.
2 H, H5,5Ј), 7.53 (s, 2 H, H6,6Ј), 6.49 (s, 4 H, m-H), 2.45 (s, 12 H,
o-CH3), 2.28 (s, 6 H, p-CH3) ppm.
[22] [22a]
T. Yamamoto, M. Abla, J. Organomet. Chem. 1997, 535,
209. [22b] T. Yamamoto, Synlett 2003, 4, 425. [22c] T. Yamamoto,
S. Wakabayashi, K. Osakada, J. Organomet. Chem. 1992, 428,
223.
Supporting Information: Fourteen tables with comparisons between
calculated and experimental structural data for 2c, 3c, and [(bpy)-
Ni(Me)2]; details of the crystal structure of 3a; 13C NMR spectro-
scopic data of selected complexes and absorption spectroscopic
data are also given. Furthermore, eight figures are provided show-
ing additional EXAFS spectra and the XANES spectra of 2e and
2d and a view of the unit cell of 3a from the crystal structure deter-
mination (see also the footnote on the first page of this article).
[23]
[24]
A. Klein, Z. Anorg. Allg. Chem. 2001, 627, 645.
M. P. Feth, A. Klein, H. Bertagnolli, Eur. J. Inorg. Chem.
2003, 839.
[25] [25a] A. Klein, E. J. L. McInnes, W. Kaim, J. Chem. Soc., Dalton
[25b]
Trans. 2002, 2371.
A. Klein, H.-D. Hausen, W. Kaim, J.
Organomet. Chem. 1992, 440, 207. [25c] A. Klein, M. Niemeyer,
Z. Anorg. Allg. Chem. 2000, 626, 1191.
Slageren, S. Zalis, Eur. J. Inorg. Chem. 2003, 1927.
[25d]
A. Klein, J. van
´
[26] [26a] G. J. Colpas, M.J. Maroney, C. Bagyinka, M. Kumar, W.S.
Willis, S.L. Suib, N. Baidya, P. K. Mascharak, Inorg. Chem.
1991, 30, 920. [26b] M. W. Renner, L. R. Furenlid, K. M. Barki-
gia, J. Fajer, J. Phys. IV France 1997, 7, 661. [26c] N. Kosugi, T.
Acknowledgments
We wish to thank HASYLAB at DESY (Hamburg, Germany), the
SSRL at SLAC (Stanford, USA) and BESSY (Berlin, Germany)
for providing synchrotron radiation. The SSRL is operated by the
U.S. Department of Energy, Office of Basic Energy Science. Minis-
try of Education of the Czech Republic and the COST Action D14
is gratefully acknowledged.
Yokoyama, K. Asakura, H. Kuroda, Chem. Phys. 1984, 91,
[26d]
249.
H. Nishida, T. Yokoyama, H. Kuroda, Chem. Phys.
1986, 104, 449.
[27]
[28]
D. M. Manuta, A. J. Lees, Inorg. Chem. 1983, 22, 3825.
W. Kaim, S. Ernst, S. Kohlmann, Chem. unserer Zeit 1987,
21, 50.
[29] [29a]
´ ˇ
A. Klein, J. van Slageren, S. Zalis, Eur. J. Inorg. Chem.
2003, 1917. [29b] A. Klein, J. van Slageren, S. Zalis, Inorg. Chem.
2002, 41, 5216.
´ ˇ
[1]
[1a] S. D. Ittel, L. K. Johnson, M. Brookhart, Chem. Rev. 2000,
[30]
[31]
[1b]
G. C. Tucci, R. H. Holm, J. Am. Chem. Soc. 1995, 117, 6489.
100, 1169.
C. M. Killian, L. K. Johnson, M. Brookhart,
Organometallics 1997, 16, 2005. [1c] S. Mecking, L. K. Johnson,
L. Wang, M. Brookhart, J. Am. Chem. Soc. 1998, 120, 888. [1d]
A. M. LaPointe, M. Brookhart, Organometallics 1998, 17,
1530.
ˇ
M. Turki, C. Daniel, S. Zalis, A. Vlcek, Jr, J. Van Slageren, D.
J. Stufkens, J. Am. Chem. Soc. 2001, 123, 11431.
[32] [32a] P. C. Ford, i n Inorganic and Organometallic Photochemistry
(Ed.: M. S. Wrighton), ACS, Washington, D.C., 1978, vol. 168,
73. [32b] A. B. P. Lever, Inorganic Electronic Spectroscopy, Elsev-
ier, Amsterdam, 1984.
[2]
A. Michalak, T. Ziegler, Organometallics 2001, 20, 1521.
[3] [3a]
S. Mecking, Angew. Chem. 2001, 113, 550; Angew. Chem.
[33]
[3b]
R. J. H. Clark, T. J. Dines, Angew. Chem. 1986, 98, 131. R.
Int. Ed. 2001, 40, 534.
2000, 6, 4623.
A. Held, S. Mecking, Chem. Eur. J.
J. H. Clark, T. J. Dines, Angew. Chem. Int. Ed. Engl. 1986,
[4]
[5]
[6]
[7]
[8]
[9]
25, 131.
G. J. P. Britovsek, V. C. Gibson, D. F. Wass, Angew. Chem.
1999, 111, 448; Angew. Chem. Int. Ed. 1999, 38, 428.
M. Schmid, R. Eberhardt, M. Klinga, M. Leskelä, B. Rieger,
Organometallics 2001, 20, 2321.
J. Heinicke, M. He, A. Dal, H.-F. Klein, O. Hetche, W. Keim,
U. Flörke, H.-J. Haupt, Eur. J. Inorg. Chem. 2000, 3, 431.
C. Wang, S. Friedrich, T. R. Younkin, R. T. Li, R. H. Grubbs,
D. A. Bansleben, M. W. Day, Organometallics 1998, 17, 3149.
M. Schmid, R. Eberhardt, J. Kukral, B. Rieger, Z. Natur-
forsch., Teil B 2002, 57, 1141.
[34] [34a]
J. S. Strukl, J. L. Walter, Spectrochim. Acta 1971, 27, 223.
J. S. Strukl, J. L. Walter, Spectrochim. Acta 1971, 27, 209.
[34b]
[35] [35a]
I. Srnova-Sloufova, B. Vlckova, T. L. Snoeck, D. J. Stuf-
[35b]
kens, P. Matejka, Inorg. Chem. 2000, 39, 3551.
R. Hage, J.
G. Haasnoot, D. J. Stufkens, T. L. Snoeck, J. G. Vos, J. Reedijk,
Inorg. Chem. 1989, 28, 1413. [35c] P. A. Mabrouk, M. S. Wrigh-
ton, Inorg. Chem. 1986, 25, 526. [35d] R. W. Balk, T. L. Snoeck,
D. J. Stufkens, A. Oskam, Inorg. Chem. 1980, 19, 3015.
J. van Slageren, D. J. Stufkens, S. Zalis, A. Klein, J. Chem.
Soc., Dalton Trans. 2002, 218. [36b] J. van Slageren, A. Klein, S.
Zalis, D. J. Stufkens, Coord. Chem. Rev. 2001, 219Ϫ221, 937.
M. W. Wong, Chem. Phys. Lett. 1996, 256, 391.
[36] [36a]
L. Johnson, L. Wang, S. McLain, A. Bennett, K. Dobbs, E.
Hauptman, A. Ionkin, S. Ittel, K. Kunitsky, W. Marshall, E.
McCord, C. Radzewich, A. Rinehart, K.J. Sweetman, Y. Wang,
Z. Yin, M. Brookhart, ACS Symp. Ser. 2003, 857, 131.
[37]
[38] [38a]
C. Fonseca Guerra, J. G. Snijders, G. Te Velde, E. J. Baer-
[10] [10a]
[38b]
J. L. Davis, B. A. Arndtsen, Organometallics 2000, 19,
ends, Theor. Chim. Acc. 1998, 99, 391.
S. J. A. van Gis-
[10b]
4657.
4400.
C. Geyer, S. Schindler, Organometallics 1998, 17,
bergen, J. G. Snijders, E. J. Baerends, Comput. Phys. Commun.
1999, 118, 119.
Eur. J. Inorg. Chem. 2004, 2784Ϫ2796
www.eurjic.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2795

´ ˇ
A. Klein, M. P. Feth, M. Bertagnolli, S. Zalis
FULL PAPER
[39]
[43b]
Gaussian 03, Revision B.2, M. J. Frisch, G. W. Trucks, H. B.
213.
V. A. Rassolov, J. A. Pople, M. A. Ratner, T. L.
Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, J. A.
Montgomery, Jr., T. Vreven, K. N. Kudin, J. C. Burant, J. M.
Millam, S. S. Iyengar, J. Tomasi, V. Barone, B. Mennucci, M.
Cossi, G. Scalmani, N. Rega, G. A. Petersson, H. Nakatsuji,
M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M.
Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene,
X. Li, J. E. Knox, H. P. Hratchian, J. B. Cross, C. Adamo, J.
Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Aus-
tin, R. Cammi, C. Pomelli, J. W. Ochterski, P. Y. Ayala, K.
Morokuma, G. A. Voth, P. Salvador, J. J. Dannenberg, V. G.
Zakrzewski, S. Dapprich, A. D. Daniels, M. C. Strain, O.
Farkas, D. K. Malick, A. D. Rabuck, K. Raghavachari, J. B.
Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul, S. Clifford, J.
Cioslowski, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz,
I. Komaromi, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-
Laham, C. Y. Peng, A. Nanayakkara, M. Challacombe, P. M.
W. Gill, B. Johnson, W. Chen, M. W. Wong, C. Gonzalez, J. A.
Pople, Gaussian, Inc., Pittsburgh PA, 2003.
Windus, J. Chem. Phys. 1998, 109, 1223.
A. Bergner, M. Dolg, W. Kuechle, H. Stoll, H. Preuss, Mol.
Phys. 1993, 80, 1431.
[44]
[45]
[46]
D. E. Woon, T. H. J. Dunning, J. Chem. Phys. 1993, 98, 1358.
P. J. Stephens, F. J. Devlin, C. F. Cabalowski, M. J. Frisch, J.
Phys. Chem. 1994, 98, 11623.
[47] [47a] G. M. Sheldrick, SHELXTL, Bruker Analytical X-ray Sys-
[47b]
tems, Madison, Wisconsin, USA, 1998.
SHELXL-97: A program for Crystal Structure Determination,
G. M. Sheldrick,
University of Göttingen, Göttingen, Germany, 1997.
[48] [48a]
T. S. Ertel, H. Bertagnolli, S. Hückmann, U. Kolb, D.
[48b]
Peter, Appl. Spectrosc. 1992, 46, 690.
M. Newville, P. Liv-
ins, Y. Yakoby, J. J. Rehr, E. A. Stern, Phys. Rev. B 1993, 47,
14126.
[49]
[50]
[51]
[52]
[53]
S. J. Gurman, N. Binsted, I. Ross, J. Phys. C 1986, 19, 1845.
F. Effenberger, Chem. Ber. 1965, 98, 2260.
J. J. Lafferty, F. H. Case, J. Org. Chem. 1967, 32, 1591.
J. M. Kliegman, R. K. Barnes, Tetrahedron 1970, 26, 2555.
S. E. Biali, Z. Rappoport, J. Am. Chem. Soc. 1985, 107, 1007.
Received December 19, 2003
[40]
A. D. Becke, Phys. Rev. A 1988, 38, 3098.
[41]
J. P. Perdew, Phys. Rev. A 1986, 33, 8822.
P. R. T. Schipper, O. V. Gritsenko, S. J. A. van Gisbergen, E. J.
[42]
Early View Article
Baerends, J. Chem. Phys. 2000, 112, 1344.
Published Online April 26, 2004
[43] [43a]
P. C. Hariharan, J. A. Pople, Theor. Chim. Acta 1973, 28,
2796
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2004, 2784Ϫ2796