Unprecedented tetranuclear complexes with 20-electron NiII centers: The role of pressure and temperature on their solid-state and solution fragmentation

Article abstract of DOI:10.1002/anie.200701483

Complexes under pressure! The complexes [{NiCl₂(PN)} ₄] (PN = 2-diphenylphosphinomethyl-2-oxazoline or -2-thiazoline; see picture; Ni green, Cl pink, P orange, S yellow, N blue, C gray) have an unprecedented, centrosymmetric Ni₄Cl₈ core with 20-electron metal centers and undergo pressure-induced fragmentation into the mononuclear, square-planar 16-electron complexes [NiCl₂(PN)], which are in equilibrium with the respective tetramer in solution. (Figure Presented).

Full text of DOI:10.1002/anie.200701483

Communications
DOI: 10.1002/anie.200701483
Coordination Chemistry
Unprecedented Tetranuclear Complexes with 20-Electron
NiII Centers: The Role of Pressure and Temperature on
Their Solid-State and Solution Fragmentation**
Anthony Kermagoret, Roberto Pattacini, Patricia Chavez Vasquez,
Guillaume Rogez, Richard Welter, and Pierre Braunstein*
Dedicated to Professor D. Fenske on the
occasion of his 65th birthday
Angewandte
Chemie
6438
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 6438 –6441

Angewandte
Chemie
Ligands with phosphorus and nitrogen or oxygen donor atoms
allow fine-tuning of the structural properties and reactivity
(stoichiometric and catalytic) of their metal complexes.[1] In
particular, complexes bearing diversely substituted phosphi-
nooxazolines have been studied for their dynamic behavior
and catalytic activity in a number of reactions.[1c,2]
The unique reactivity of phosphinothiazolines[3] triggered
a comparison between the ligands 2-diphenylphosphino-
methyl-2-thiazoline (PNth) and 2-diphenylphosphinomethyl-
2-oxazoline (PNox). Their reaction in CH2Cl2 with NiCl2
afforded red solutions, which upon slow evaporation at
258C yielded red powders of [NiCl2(PNth)] (1) and [NiCl2-
(PNox)] (2), respectively. However, rapid precipitation
afforded green powders (see the Supporting Information),
and green single crystals of [{NiCl2(PNth)}4]·2CH2Cl2
(1a·2CH2Cl2) were grown from a solution of 1 (Figure 1).
In the centrosymmetric, tetranuclear structure of 1a, each
metal center is in a slightly distorted octahedral environment
defined by a chelating PNth ligand and four chlorine atoms:
Cl1 is terminally bound to Ni1, but Ni1 and Ni2 are
symmetrically bridged by Cl3 and Cl4, and Cl2 asymmetri-
cally bridges three metal centers with the unusual electron
count of 20. The Ni1···Ni2 distance of 3.0951(5) Š indicates
the absence of strong metal–metal interactions.
The Ni4Cl8 core of 1a is unprecedented; no complex with a
M4X8 (X = halogen) stoichiometry appears to have been
reported for Group 10 metals. Furthermore, its geometry is
new for any combination of transition metals with halogens,
and 1a represents the first complex in which a phosphine is
coordinated to a NiCl4 moiety. Only one example of a
phosphine bound to a NiX4 fragment has been reported (X =
Br).[5]
Figure 1. Molecular structure of 1a in 1a·2CH2Cl2. Hydrogen atoms
omitted for clarity. Selected bond distances [Š] and angles [8]:
Ni1···Ni2 3.0951(5), Ni1–Cl1 2.3357(7), Ni1–Cl2 2.6148(8), Ni1–Cl3
2.3870(8), Ni1–Cl4 2.4265(7), Ni2–Cl2 2.4534(8), Ni2–Cl2’ 2.4307(8),
Ni2–Cl3 2.4480(8), Ni2–Cl4 2.3972(8); Ni2-Cl2-Ni2’ 93.79(3), Ni1-Cl2-
Ni2’ 134.50(3), Ni1-Cl2-Ni2 75.20(2), Ni1-Cl3-Ni2 79.59(3), Ni1-Cl4-
Ni2 79.83(2). Symmetry operations generating equivalent atoms (’):
Àx, Ày, Àz.
The green complex [NiCl2(PNox)]4 (2a), analogous to 1a,
was obtained under similar conditions from PNox and
identified by X-ray powder diffraction (see the Supporting
Information). In the solid state, complexes 1a and 2a undergo
irreversible pressure-induced modification, with a color
change from green to deep red (at 20–35 kbar; Scheme 1).
[*] A. Kermagoret, Dr. R. Pattacini, P. Chavez Vasquez, Dr. P. Braunstein
Laboratoire de Chimie de Coordination
Institut de Chimie
UMR 7177 CNRS
UniversitØ Louis Pasteur
4 rue Blaise Pascal, 67070 StrasbourgCedex (France)
Fax: (+33)390-241-322
E-mail: braunstein@chimie.u-strasbg.fr
Homepage: http://www-chimie.u-strasbg.fr/~lcc/
Dr. G. Rogez
Groupe des MatØriaux Inorganiques
IPCMS
Scheme 1. Pressure-induced conversion of 1a and 2a into 1b and 2b,
respectively.
UMR 7504 CNRS
23, rue du Loess, B.P. 43
67034 StrasbourgCedex 2 (France)
Their high-pressure forms and the red powders obtained by
slow solvent evaporation were identified as the mononuclear
complexes [NiCl2(PNth)] (1b) and [NiCl2(PNox)] (2b), respec-
tively, by comparison of their FTIR spectra with those of the
corresponding bromides 3 and 4 (Figure 2 and Figure S-1 in
the Supporting Information).
Prof. R. Welter
DECOMET
Institut de Chimie
UMR 7177 CNRS
UniversitØ Louis Pasteur
4 rue Blaise Pascal, 67070 StrasbourgCedex (France)
The slightly distorted square-planar complexes 3 and 4
contain a PN chelate and two terminal bromides. The trans
[**] This work was supported by the Centre National de la Recherche
Scientifique and the Minist›re de l’Enseignement SupØrieur et de la
Recherche. We thank Dr. D. Mandon for assistance with the
variable-temperature UV/Vis measurements.
À
influence of the phosphorus donor makes the Ni1 Br2 bond
À
longer than the Ni1 Br1 bond. Both PN ligands show
À
À
analogous coordination geometries. The Ni N and Ni P
bond lengths in 3 are much shorter than those observed in 1a.
Supportinginformation for this article is available on the WWW
under http://www.angewandte.org or from the author.
Angew. Chem. Int. Ed. 2007, 46, 6438 –6441
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 6439

Communications
interaction. Below 20 K, the small increase of the cT product
up to 3.49 emuKmolÀ1 at 14 K and the final decrease below
14 K are probably due to a complicated interaction scheme
and to competing intramolecular interactions (Figure 3 and
Figure 2. ORTEP views of the molecular structures of 3 (left) and 4
(right). Ellipsoids enclose 50% of the electron density. Hydrogen
atoms omitted for clarity. Selected bond distances [Š] and angles [8]: 3:
Ni1–P1 2.147(1), Ni1–N1 1.910(3), Ni1–Br1 2.3095(6), Ni1–Br2
2.3665(6); N1-Ni1-P1 84.4(1), P1-Ni1-Br1 87.75(3), N1-Ni1-Br2
95.1(1), Br1-Ni1-Br2 92.72(2); 4: Ni1–P1 2.158(1), Ni1–N1 1.905(3),
Ni1–Br1 2.2983(7), Ni–Br2 2.3602(7), N1-Ni1-P1 85.0(1), P1-Ni1-Br1
87.73(4), N1-Ni1-Br2 94.0(1), Br1-Ni1-Br2 93.37(3).
The red solutions of 3 and 4 display magnetic moments of 1.3
and 1.0mB, respectively (Evans method[4]), which points to the
presence of a paramagnetic isomer in equilibrium with the
square-planar one (see below).
Relative to the FTIR spectrum of 2b (Figure S-2 in the
Supporting Information), that of 2a displays split absorptions,
which is consistent with two symmetry-independent ligands.
In contrast to 1a but similarly to 3 and 4, one independent
ligand is likely to be present in the unit cell of 1b.
Few fully characterized pressure-induced isomerizations
of nickel complexes have been reported: examples include
the conversion from tetrahedral to dinuclear, square-pyrami-
dal geometry,[6] and from mononuclear distorted tetrahedral
to square-planar geometry.[7] A square-planar to octahedral
isomerization was recently reported for a Pd thioether
species.[8] Our system represents a more dramatic situation,
involving: 1) the fragmentation of a tetranuclear complex
with 20-electron metal centers into four 16-electron square-
planar molecules and 2) an octahedral to square-planar
rearrangement, associated with an overall density increase
(estimated density from X-ray powder diffraction data for
2a·2CH2Cl2: 1.41 and for 2b, based on data for 4: 1.55 gcmÀ3,
see the Supporting Information).[9] The CH2Cl2 molecules
play a role in this phenomenon; their loss from the crystalline
powders leaves voids that lower the density of the solid by
approximately 10%, thus explaining the irreversibility of the
process.
&
*
Figure 3. cT ( ) and 1/c ( ) for 2a as a function of temperature. Solid
line: fit of the 1/c versus T curve to the Curie–Weiss law in the region
150–300 K.
the Supporting Information) as well as to zero-field splitting.
A paramagnetic impurity that obeys the Curie law (c = C/T)
would lead to a constant additive contribution in cT, which
could not explain the slight increase of the cT product
observed between 20 and 14 K. So far, no analytical fitting by
full diagonalization of the complete spin Hamiltonian was
successful over the whole temperature range. Further studies
are in progress to determine the magnitude and sign of the
different exchange constants and the characteristics of the
ground state.
The red solutions of 1a/b and 2a/b change to green at low
temperatures (Figure 4). A strong absorption band at 517 nm
dominates the spectrum recorded at 293 K. Two weak bands
centered at 615 and 700 nm were also detected. When the
temperature was decreased, the main absorption band
The magnetic properties of 2a in the solid state were
investigated in the range from 300 to 1.8 K with an applied
field of 5 kOe. The Curie constant C = 4.64 emuKmolÀ1,
determined from a fit of the 1/c versus T curve to the Curie–
Weiss law in the high-temperature region (150–300 K),
is as expected for four octahedral NiII ions (g = 2.15).[10]
The cT product decreases regularly upon cooling from
4.54 emuKmolÀ1 at 300 K to 3.47 emuKmolÀ1 at 20 K,
indicating at least one intramolecular antiferromagnetic
Figure 4. Visible spectra of 1a/b in MeCN/CH2Cl2 (5:2) in the range
293–193 K and of solid 1a at room temperature.
6440 www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 6438 –6441

Angewandte
Chemie
bridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_
request/cif.
disappeared and the two minor ones remained almost
unchanged. The pattern of the spectrum at 193 K is similar
to that of solid 1a, suggesting that 1a is present in solution and
favored at low temperature, being enthalpically stabilized
with respect to 1b. The equilibrium was fully restored at
293 K.
Received: April 5, 2007
Revised: June 7, 2007
Published online: July 16, 2007
The presence of two species in equilibrium in solution is
consistent with the paramagnetism detected by the Evans
method. The values of 2.8 and 2.3mB for 1 and 2, respectively,
in CD2Cl2 are not consistent with either square-planar or
octahedral coordination geometry. The paramagnetic NMR
data for 1–4 could thus result from an equilibrium between a
mononuclear diamagnetic complex and a polynuclear, prob-
ably tetranuclear, paramagnetic one. This equilibrium
explains why rapid precipitation of a solution of 1b or 2b
affords 1a or 2a, respectively, since precipitation shifts the
equilibrium towards the less soluble species. Equilibria
between square-planar and octahedral Ni complexes in
solution are not uncommon and may involve solvation[11]
and hemilability.[12] Other examples include tetrahedral to
square-planar isomerism.[13] However, the present equilibri-
um between mononuclear and polynuclear Ni complexes with
the same molecular formula appears unprecedented. The
tetranuclear form is more stabilized in the case of PNth than of
PNox, as observed by 1) the Evans method and 2) the
dependence of K on temperature.
Keywords: isomerization · magnetic properties · N,P ligands ·
nickel · polynuclear complexes
.
[1] See, for example, a) P. Braunstein, Chem. Rev. 2006, 106, 134 –
159; b) P. Braunstein, F. Naud, Angew. Chem. 2001, 113, 702 –
722; Angew. Chem. Int. Ed. 2001, 40, 680 – 699; c) G. Helmchen,
A. Pfaltz, Acc. Chem. Res. 2000, 33, 336 – 345; d) C. S. Slone,
D. A. Weinberger, C. A. Mirkin, Prog. Inorg. Chem. 1999, 48,
233 – 250.
[2] a) F. Speiser, P. Braunstein, L. Saussine, Acc. Chem. Res. 2005,
38, 784 – 793; b) P. Braunstein, G. Clerc, X. Morise, R. Welter, G.
Mantovani, Dalton Trans. 2003, 1601 – 1605; c) P. Braunstein, G.
Clerc, X. Morise, New J. Chem. 2003, 27, 68 – 72; d) P.
Braunstein, F. Naud, A. Dedieu, M.-M. Rohmer, A. DeCian,
S. J. Rettig, Organometallics 2001, 20, 2966 – 2981; e) P. Braun-
stein, F. Naud, S. J. Rettig, New J. Chem. 2001, 25, 32 – 39; f) P.
Braunstein, C. Graiff, F. Naud, A. Pfaltz, A. Tiripicchio, Inorg.
Chem. 2000, 39, 4468 – 4475; g) P. Braunstein, M. D. Fryzuk, M.
Le Dall, F. Naud, S. J. Rettig, F. Speiser, J. Chem. Soc. Dalton
Trans. 2000, 1067 – 1074; h) J. Sprinz, G. Helmchen, Tetrahedron
Lett. 1993, 34, 1769 – 1772.
[3] G. Margraf, R. Pattacini, A. Messaoudi, P. Braunstein, Chem.
Commun. 2006, 3098 – 3100.
[4] a) D. F. Evans, J. Chem. Soc. A 1959, 2003 – 2005; b) S. K. Sur, J.
Magn. Reson. 1989, 82, 169 – 173.
[5] P. B. Hitchcock, T. H. Lee, G. J. Leigh, Dalton Trans. 2003, 2276 –
2279.
The situations encountered in this work reveal many
unprecedented features and indicate that subtle differences
may result from the presence of oxygen or sulfur in PNox and
PNth, respectively.
Experimental Section
[6] G. J. Long, J. R. Ferraro, J. Chem. Soc. Chem. Commun. 1973,
719 – 720.
[7] J. R. Ferraro, K. Nakamoto, J. T. Wang, L. Lauer, J. Chem. Soc.
Chem. Commun. 1973, 266 – 267.
[8] D. R. Allan, A. J. Blake, D. Huang, T. J. Prior, M. Schröder,
Chem. Commun. 2006, 4081 – 4083.
Crystal data: 1a: C64H64Cl8N4Ni4P4S4·2CH2Cl2, T= 193 K, M =
¯
1829.60, triclinic P1, a = 11.248(1), b = 11.518(1), c = 14.843(2) Š,
a = 91.811(3), b = 94.041(3), g = 97.161(3)8, V= 1901.6(3) Š3, Z = 1,
1calcd = 1.598 gcmÀ3, m(MoKa) = 1.63 mmÀ1, F(000) = 932, 2qmax = 608,
R1 = 0.0485, wR2 = 0.1107, R(int) = 0.0369 parameters = 420, 17720
reflections measured, 7468 (I > 2s(I)). 3: C16H16NSPNiBr2, T=
[9] For
a recent review on high-pressure chemistry, see: W.
¯
Grochala, R. Hoffmann, J. Feng, N. W. Ashcroft, Angew.
Chem. 2007, 119, 3694 – 3717; Angew. Chem. Int. Ed. 2007, 46,
3620 – 3642.
193 K, M = 503.86, triclinic P1, a = 8.6595(3), b = 8.6983(2), c =
12.9457(4) Š, a = 91.476(1), b = 98.175(1), g = 111.337(1)8, V=
895.86(5) Š3, Z = 2, 1calcd = 1.868 gcmÀ3
,
m(MoKa) = 5.74 mmÀ1
,
[10] R. L. Carlin, Magneto-chemistry, Springer, Berlin, 1986.
[11] S. Mukhopadhyay, D. Mandal, D. Ghosh, I. Goldberg, M.
Chaudhury, Inorg. Chem. 2003, 42, 8439 – 8445.
[12] a) L. Sacconi, P. Nannelli, N. Nardi, U. Campigli, Inorg. Chem.
1965, 4, 943 – 949; b) L. Sacconi, N. Nardi, F. Zanobini, Inorg.
Chem. 1966, 5, 1872 – 1876.
[13] a) R. G. Hayter, F. S. Humiec, J. Am. Chem. Soc. 1962, 84, 2004 –
2005; b) R. G. Hayter, F. S. Humiec, Inorg. Chem. 1965, 4, 1701 –
1706.
F(000) = 496, 2qmax = 57.48, R1 = 0.0438, wR2 = 0.0965, R(int) =
0.0380 parameters = 199, 6954 reflections measured, 3318 (I >
2s(I)). 4: C16H16NOPNiBr2, T= 193 K, M = 503.86, triclinic P1, a =
¯
8.6330(3), b = 8.7360(3), c = 11.8890(4) Š, a = 104.2010(17), b =
96.7060(17), g = 96.3230(12)8, V= 854.28(5) Š3, Z = 2, 1calcd
=
1.896 gcmÀ3, m(MoKa) = 5.90 mmÀ1, F(000) = 480, 2qmax = 58.28, R1 =
0.0464, wR2 = 0.1194, R(int) = 0.0429 parameters = 199, 6589 reflec-
tions measured, 2768 (I > 2s(I)). CCDC-641491 (1a), 641489 (3), and
641490 (4) contain the supplementary crystallographic data for this
paper. These data can be obtained free of charge from The Cam-
Angew. Chem. Int. Ed. 2007, 46, 6438 –6441
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
6441