Oxido Pincer Ligands - Exploring the Coordination Chemistry of Bis(hydroxymethyl)pyridine Ligands for the Late Transition Metals

Article abstract of DOI:10.1002/ejic.200900023

ination of the 2,6-bis(hydroxymethyl)pyridine-based oxido pincer ligands RR'pydimH₂ [R = RydimH₂); R = R' = Me (pydipH ₂); R = 2-tolyl, R' = Me (pydotH₂)] towards late transition metals Coⁿ,Niⁿ

Full text of DOI:10.1002/ejic.200900023

FULL PAPER
DOI: 10.1002/ejic.200900023
Oxido Pincer Ligands – Exploring the Coordination Chemistry of
Bis(hydroxymethyl)pyridine Ligands for the Late Transition Metals
Axel Klein,*[a] Said Elmas,[a] and Katharina Butsch[a]
Keywords: N,O ligands / Ligand design / Transition metals / Binding modes / UV/Vis spectroscopy
Coordination of the 2,6-bis(hydroxymethyl)pyridine-based
oxido pincer ligands RRЈpydimH2 [R = RЈ = H (pydimH2); R
= RЈ = Me (pydipH2); R = 2-tolyl, RЈ = Me (pydotH2)] towards
late transition metals CoII, NiII, CuII, ZnII, PdII and PtII allows
the formation of molecular species (complexes), which exhi-
bit three main structural motifs in the solid state. The two
main species are pentacoordinate [(RRЈpydimH2)MCl2] and
hexacoordinate [(RRЈpydimH2)2M]X2, both of which are
stable in solution and can be interconverted by changing the
solvent polarity. The disproportionation equilibrium [(RRЈpy-
dimH2)MCl2] h [(RRЈpydimH2)2M]2+ + [MCl4]2– was studied
by optical spectroscopy. The chiral ligand pydotH2 allows the
formation of chiral complexes. In the square-planar com-
plexes [(pydimH2)2MCl2] (M = PdII or PtII) the oxido donor
functions of the ligands do not take part in the coordination.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2009)
Firstly, the possibility of substituting the protons on the
methylene C atom allows interesting electronic and steric
Introduction
Tridentate ligands of the so-called pincer type combine variations to be introduced into ligands RRЈpydimH2
the possibility of preparing highly stable transition metal (Scheme 1) and their complexes. The most interesting varia-
complexes with defined stereochemistry with almost unlimi- tion is surely the introduction of one alkyl, aryl or alkynyl
ted possibilities for ligand derivatisation and fine-tuning of group, which leads to two stereogenic centres (R = H, RЈ
the properties of these complexes. This has allowed the ra- = alkyl, aryl, alkynyl). Alternatively, chiral disubstituted py-
tional design of such complexes for applications in catalysis dimH2 derivatives have been created by using menthyl or
or molecular sensing.[1,2] Initially coined for E C E donor other chiral groups. These have been used in several studies
∧ ∧
combinations with E = N, S, P or O and comprising a cen- with early transition metal ions such as TiIV,[4] ZrIV,[5,6]
tral M–C bond,[1] the term “pincer ligand” has since been HfIV,[6] VV[7] or MoVI.[7,8] Another interesting aspect of li-
extended to any combination of donor atoms from the per- gands of the RRЈpydimH2 type is the possibility of stepwise
iodic table.[2] The vast majority of pincer complexes are, deprotonation upon coordination, which goes along with
however, restricted to combinations of C, N, P or S donors, increasing hardness of the metal ion. Examples for com-
and the use of O-donor functions is far less developed. The plexes of early transition metals containing exclusively the
main reason for this choice is the expected (and found) high fully deprotonated ligand dianion RRЈpydim2– include
stability of complexes of the so called “soft” donors C, N, [Ti(RRЈpydim)2], [Ti(RRЈpydim)X2] (R = RЈ = Ph, H; X =
P and S in combination with the rather “soft” late transi- Cl or OiPr),[9] [Cp*Ti(L)(pydim)], [Cp*Ti(LЈ)(pydim)]+
tion metals (groups 8–11, 4d or 5d elements), in agreement (Cp* = pentamethylcyclopentadienide; L = Me, OH; LЈ =
with the completely empirical but highly intuitive HSAB H2O, MeCN),[10] Na[VO2(pydim)],[11] [Cp*Ta(Me)(pyd-
concept.[3] This concept suggests that the binding of these im)(X)] (X = OTf, H2O),[10] [MoO2(pydim)][12] or the above-
metals to O donor atoms is likely to be weak, therefore it mentioned chiral complexes.
is not surprising to find a relatively small number of com-
plexes of late transition metals with oxido pincer ligands.
Oxido pincer ligands based on 2,6-bis(hydroxymethyl)-
∧ ∧
pyridine (pydimH2, Scheme 1), which has an O N O do-
nor set, provide a number of interesting features for coordi-
nation chemistry and the application of their complexes.
[a] Universität zu Köln, Institut für Anorganische Chemie,
Scheme 1. O,N,O oxido pincer ligands used in this study. 2,6-Bis-
Greinstrasse 6, 50939 Köln, Germany
(hydroxymethyl)pyridine (pydimH2), 2,6-bis(1-hydroxy-1-methyl-
ethyl)pyridine (pydipH2) and 2,6-bis(1-hydroxy-1-o-tolylethyl)pyr-
idine (pydotH2).
E-mail: axel.klein@uni-koeln.de
Supporting information for this article is available on the
WWW under http://www.eurjic.org or from the author.
Eur. J. Inorg. Chem. 2009, 2271–2281
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2271

A. Klein, S. Elmas, K. Butsch
FULL PAPER
In contrast to this, late transition metals usually coordi-
nate to RRЈpydimH2 in its protonated form, as in [M-
(pydimH2)2]2+ (M = Co, Ni, Cu, Zn)[13–15] [Cu(O2CCH2-
CH3)2(pydimH2)],[16] [Cu(nif)2(pydimH2)] {nif = niflumate,
2-{[3-(trifluoromethyl)phenyl]amino}nicotinate},[17] or [Cu-
(pydic)(pydimH2)].[18] Exceptions to this rule have been
found for the CuII complexes [Cu2(η2,µ-pydimH)2(η3-
pydimH2)2]2+[14b,19] and [Cu(pydimH2)(pydimH)]+.[14b]
An intermediate case arises in the polynuclear manga-
nese complexes (and molecular magnets) [Mn9(O2CCH2-
CH3)12(pydim)(pydimH)2(L)2] {LH2 = [6-(hydroxymethyl)-
pyridin-2-yl](6-hydroxypyridin-2-ylmethoxy)methanol}[20]
or [Mn4(O2CCH3)2(pydimH)6](ClO4)2,[21] where the mono-
deprotonated ligand pydimH– is found. This agrees nicely
with our general rule of deprotonation and increasing hard-
ness. Again, however, there is an exception: the fully pro-
tonated ligand pydimH2 is found in [MnCl(η3-
pydimH2)(η2-pydimH2)]+,[22] even though the complex also
contains MnII, as in the examples with the partly deproton-
ated ligand.
Scheme 2. Synthesis of the ligands pydipH2 and pydotH2.
Both ligands were obtained in good to excellent yields
1
and fully characterised by elemental analysis and H and
13C NMR spectroscopy (see Experimental Section).
Crystal and Molecular Structure of pydotH2
Suitable crystals of the ligand pydotH2 were obtained
from a concentrated solution in ethanol. The crystal struc-
At the far end of metal softness, the metal ions PdII and
PtII have been found to coordinate only through the pyr-
idine N-atom in complexes [MCl2(pydimH2)2] (M = Pd[23]
or Pt[24]); coordination of the hard O-donor functions was
not observed.
In this paper we report on a study focussing on the coor-
dination chemistry of three simple oxido pincer ligands
based on 2,6-bis(hydroxymethyl)pyridine (Scheme 1)
towards late transition metals CoII, NiII, CuII, ZnII, PdII
and PtII, which have an increasing “soft” character along
the series. Within this series of ligands, the steric bulk at the
methyloxy C atom and the basicity of this atom increases
from pydimH2 to pydipH2 and pydotH2 (Scheme 1), and
pairs of enantiomers are expected for complexes of the
pydotH2 ligand.
¯
ture was solved in the triclinic space group P1. Close H···O
contacts between hydroxy groups of two neighbouring
molecules are observed in the crystal structure, although the
H···O distances of 2.075(7) and 2.260(6) Å [O···O distance:
2.879(4) Å] qualify them as medium to weak H bonds.[27]
The H atoms on O1 and O2 were found during refinement,
whereas other H atoms were added by using appropriate
riding models, Figure 1 shows the molecular structure.
The ligand pydipH2 has been synthesised previously by
lithiation of 2,6-dibromopyridine and addition of acetone
in situ.[25] The method was first mentioned by Osborn and
Kress (using menthone instead of acetone)[5,7] and has been
developed recently by Schmalz et al. to a broadly applicable
method, coined the “in situ quench (ISQ)” method.[25] The
ligand pydipH2 was alternatively recently prepared from di-
methyl pyridine-2,6-dicarboxylate and MeLi.[26]
Figure 1. View of the molecular structure of pydotH2 (thermal el-
lipsoids represent 30% probability). The H atoms on O1 and O2
are displayed, whereas others have been omitted for clarity. The
dotted line represents the H bond of 2.016(3) Å [angle O(1)–H(1)···
N(1) 123.0(2)°; N···O 2.557(5) Å].
Results and Discussion
One of the two hydroxy units (O1) in the free ligand py-
dotH2 is almost co-planar with the pyridine core [tilt angle
OCCN 2.57(2)°]. This is probably caused by a short O–
H···N contact of 2.016(3) Å [angle N···H–O 123.0(2)°]
Synthesis of the Ligands
Whereas the ligand pydimH2 is commercially available, which can be qualified as a medium to strong H bond.[27]
pydipH2 was synthesised from dimethyl pyridine-2,6-dicar- The other hydroxy group is tilted away from the pyridine
boxylate and MeLi by modifying a literature procedure plane with a tilt angle of 37.59(2)°. Consequently, the O–
(Scheme 2).[26] The ligand pydotH2 was prepared by using H···N hydrogen bond is longer [2.426(3) Å; angle 103.0(2)°;
the ISQ method starting from 2,6-diacetylpyridine and tol- N···O 2.729(5) Å] and thus much weaker. The methyl sub-
ylmagnesium bromide.
stituents of the two tolyl groups are both oriented towards
2272
www.eurjic.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2009, 2271–2281

Bis(hydroxymethyl)pyridine Ligands for the Late Transition Metals
the core of the pincer ligand and, most importantly, in the Section for details). Importantly, during the refinement pro-
crystal structure we find only the meso configuration of the cess we were able to find all H atoms of the ligands’ OH
molecule, with C8 in the (R) configuration and C15 in the functions. Other H atoms were usually refined by using ap-
(S) configuration.
propriate riding models. For the discussion of the crystal
and molecular structures we will divide the structures into
the three groups A, B and C defined in Scheme 3.
Synthesis of the Metal Complexes
The first type (A) contains the structures of the neutral
complexes [(RRЈpydimH2)MCl2] (Figure 2). Most of them
crystallised as solvates (EtOH or H2O), and two poly-
Ethanol turned out to be the best solvent to dissolve
both the organic ligand and the metal chlorides. Complexes
of the type [(RRЈpydimH2)MCl2] (Scheme 3, Route A; M
= Co, Ni, Cu, Zn) were synthesised by mixing solutions of
the ligands with 3d metal salts in a 1:1 ratio (see Experi-
mental Section for details). Alternatively, complexes of the
composition [(RRЈpydimH2)2M][MCl4] were obtained by
the same procedure for some of the ligands with M = Co
and Zn (Route B1). Similarly, the application of a ligand/
metal ratio of 2:1 allowed the formation of complexes
[(RRЈpydimH2)2M]Cl2 (Route B2).
¯
morphs (P21/c or P1) of the compound [(pydimH2)ZnCl2]
were crystallised and characterised (see Table 3 for details).
The latter suggests that the space groups are probably not
of particular importance for the structure discussion. There
are frequent but weak H bonding interactions between the
solvent molecules and the complexes or between complex
molecules. The shortest contacts are observed between O–
H protons and chlorido ligands in [(pydotH2)CoCl2]·EtOH,
with an H···Cl distance of 2.054(8) Å [O···Cl 3.094(5) Å]
and a Cl···H–O angle of 170.9(4)°, which classifies this in-
teraction as a medium to strong H bond.[27] All the com-
plexes exhibit a fivefold coordination sphere around the
central metal atom with quite similar M–N and M–O dis-
tances between 1.93 and 2.07 Å for the M–N bonds and
2.00 and 2.22 Å for the M–O bonds (see Table 1). The pyr-
idine ring and the oxido donor function of the RRЈpydim
ligands are almost coplanar and the NOMN coordination
plane is perpendicular to the ClMCl plane.
A marked difference can be observed in the orientation
of the chlorido co-ligands. Thus, the two chlorido ligands
complete a trigonal-bipyramidal arrangement around the
central metal atom for Co, Ni and Zn, whereas Cu displays
a square-pyramidal coordination sphere (see Figure 2).
Both polyhedra are distorted, with the Cl–M–Cl angle for
the trigonal bipyramids varying from 113° to 118°. For the
N–M–Cl angle, a smaller (102–117°) and a larger angle
(128–145°) add up to a sum of about 245°. The square-
pyramidal structure of the CuII derivative is characterised
by a small Cl–M–Cl angle of 101.24(3)° and a large N–M–
Cl angle of 165.16(5)°. The difference between Co, Ni and
Zn on the one hand and the CuII derivative on the other
can easily be explained by the d9 configuration of CuII, with
the expected Jahn–Teller distortion leading to a preferred 4
(square planar) + 1 coordination. Indeed, the axial chlorido
ligand exhibits a markedly elongated distance [2.696(8) Å]
compared to its equatorial counterpart [2.208(8) Å]. Inter-
estingly, the O–M–O angle seems to be rather flexible, rang-
ing from 144° to 160°. As expected, the highest value is
found for the square-pyramidal CuII system. Furthermore,
the chiral pydotH2 ligand in the molecular structure of
[(pydotH2)CoCl2]·EtOH (Figure 2) displays a meso configu-
ration.
Scheme 3. Formation of the metal complexes.
The reaction of K2[MCl4] (M = Pd or Pt) with pydimH2
gave complexes of the composition [(pydimH2)2MCl2]
(Route C); the Pt complex has been described before.[24] The
other ligands did not react under the chosen conditions
(stirring at ambient temperature), therefore the alternative
complex precursors [(dmso)2MCl2] were used, but again the
reactions with pydipH2 or pydotH2 failed. The products ob-
tained from the reaction of these precursors with pydimH2
were the novel complexes trans-[(pydimH2)2PdCl2] and cis-
and trans-[(pydimH2)(dmso)PtCl2] (obtained as a mixture).
The oxidation state of the metal atom remained unchanged,
and the RRЈpydimH2 ligand remained protonated in all
these reactions.
The second type of structure (B) is found for those of
the composition [(RRЈpydimH2)2M][X]n, where X can be
various anions like Cl, as in [(pydipH2)2Cu]Cl2·H2O (see
Figure 3), or trifluoroacetic acid (TFA), as in [(pydimH2)2-
Crystal and Molecular Structures of the Complexes in the
Solid State
The crystal structures discussed in this section were Co][TFA]2. Alternatively, tetrachloridometallates [MCl4]2–
¯
solved in various space groups ranging from triclinic P1 to can act as counterion, as in [(pydipH2)2Co][CoCl4] (see
monoclinic C2/c (see Tables 3, 4 and 5 in the Experimental Table 4 for details of the crystal structures). Various hydro-
Eur. J. Inorg. Chem. 2009, 2271–2281
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
2273

A. Klein, S. Elmas, K. Butsch
FULL PAPER
Figure 2. View of the molecular structures of [(pydimH2)CuCl2]·H2O (left) and [(pydotH2)CoCl2]·EtOH (right) (thermal ellipsoids repre-
sent 30% probability). The H atoms on the ligands’ oxido functions are displayed, whereas other H atoms and the ethanol molecule have
been omitted for clarity.
Table 1. Selected structural data for complexes of the type [(RRЈpydimH2)MCl2].
[(pydotH2)ZnCl2]·
EtOH
[(pydimH2)ZnCl2] [(pydimH2)ZnCl2] [(pydotH2)CoCl2]·
[(pydotH2)NiCl2]·
EtOH
[(pydimH2)CuCl2]·
H2O
¯
in P21/c
in P1
EtOH
M–N
M–O
2.070(3)
2.165(3)
2.153(3)
2.257(1)
2.218(1)
143.9(2)
74.3(1)
2.042(3)
2.222(2)
2.153(2)
2.277(1)
2.233(1)
151.7(1)
75.7(1)
2.037(3)
2.142(4)
2.162(4)
2.285(1)
2.230(1)
144.6(1)
74.9(1)
2.038(2)
2.143(2)
2.132(2)
2.283(1)
2.247(1)
146.28(8)
75.43(9)
75.56(9)
117.24(4)
105.83(7)
136.89(7)
90.5(1)
1.962(6)
2.074(6)
2.063(6)
2.294(2)
2.242(2)
150.5(2)
77.8(3)
1.929(1)
2.001(2)
1.997(2)
2.6960(9)
2.2076(7)
159.16(7)
79.24(7)
80.36(6)
101.24(3)
93.58(5)
165.16(5)
93.0(1)
M–Cl
O–M–O
O–M–N
74.4(1)
76.1(1)
75.2(1)
77.5(2)
Cl–M–Cl
Cl–M–N
118.32(6)
106.8(1)
134.8(1)
90.6(1)
113.58(4)
117.77(9)
128.58(9)
90.6(1)
110.42(5)
105.9(1)
143.7(1)
88.5(1)
113.32(9)
101.6(2)
145.1(2)
89.5(1)
NOOM/
ClClM
gen-bonding interactions are also observed here. The short-
est are found between the carboxylate anions and the H–O
function in [(pydimH2)2Co][TFA]2, where the O···H dis-
tance of 1.674(5) Å [O···O 2.577(6) Å] and the O–H···O an-
gle of 172.9(5)° classify them as medium to strong H
bonds.[27] The metal centres in the dicationic complexes are
coordinated by two RRЈpydim ligands in an approximately
octahedral environment. Table 4 lists selected bond lengths
and angles.
At first sight, the bond lengths and angles around the
metal atoms are quite similar in all the structures, and the
two pyridine rings are oriented almost perpendicular to
each other. However, a closer look reveals differences be-
tween the structures of the Co and Zn derivatives and that
of the Cu complex. For example, the four oxido donor
atoms are not arranged in a square-planar fashion due to
the geometrical restriction enforced by the pincer ligand.
Figure 3. View of the molecular structure of [(pydipH2)2Cu]-
Cl2·H2O (thermal ellipsoids represent 30% probability). The H
atoms on the ligands’ oxido functions are displayed, whereas other
H atoms and the water molecule have been omitted for clarity.
The deviation of the square plane towards a tetrahedron Shortest H···Cl contact: Cl(1)···H(2) [2.156(7) Å]; Cl(1)···O(2)
3.055(5) Å; angle O–H···Cl 170.9(5)°.
can best be described by the sum of the angles around the
metal atom, which is about 305° for the Co and Zn deriva-
tives and thus far from the required 360° for a perfect plane.
The corresponding value for the Cu derivative is 313°
(Table 4). The most striking finding are the bond lengths in complexes also reveals that the M–N distances are shorter
the Cu complex [(pydipH2)2Cu]Cl2·H2O. The two axial N– for the CuII derivative than for the Co or Zn analogues and
M bonds are markedly shorter than the equatorial M–O that the four (equatorial) M–O bonds in the copper com-
bonds, which is the opposite of what would have been ex- plex are slightly longer than those of the Co or Zn deriva-
pected from a normal Jahn–Teller distortion. Our series of tives. We therefore assume a tetragonal compressive effect
2274
www.eurjic.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2009, 2271–2281

Bis(hydroxymethyl)pyridine Ligands for the Late Transition Metals
for this complex. This is in line with the Jahn–Teller theo-
rem, which includes both tetragonal compressions or exten-
sions.[28,29]
Weber et al. have recently investigated the crystal struc-
tures of complexes with pydimH2 and Co, Ni and Cu.[14]
The complexes [(pydimH2)2Co]2+ with various anions such
–
2–
–
as NO3 , SO4 or p-TolOSO3 exhibit very similar struc-
tural parameters when compared with our pydimH2 and
pydipH2 derivatives.[14a]
Interestingly,
a
complex
[(pydimH2)(pydimH)CoIII]Cl2 was obtained, in which de-
protonation of the ligand occurs concomitantly with oxi-
dation of the central metal atom (CoII to CoIII). Similarly,
a CuII complex containing a partially deprotonated ligand –
[(pydimH2)(pydimH)Cu]ClO4 – was reported.[14b] In both
cases the corresponding M–O bonds are about 10% shorter
than the corresponding bonds with protonated O atoms
(M–OH). Four further copper complexes [(pydimH2)2Cu]2+
have been crystallised with structural parameters very sim-
ilar to our structure with the pydipH2 ligand. The axial N–
Cu bonds are also rather short in this complex, with values
ranging from 1.935(4) Å to 2.102(3) Å, clearly indicating a
Jahn–Teller-like compression.[14b] A number of nickel com-
plexes [(pydimH2)2Ni]2+ with various counterions (Cl–,
Figure 4. View of the molecular structure of trans-
[(pydimH2)(dmso)PtCl2] (thermal ellipsoids represent 30% prob-
ability). The H atoms on the ligand’s oxido functions are displayed,
whereas other H atoms have been omitted for clarity.
ular structures. The coordination geometry around the
metal atom is square-planar, as expected, and the bond
lengths around the metal atom in the two stereoisomeric Pt
complexes show clear trends in terms of the trans influence
of ligands. For example, the Pt–N bond is slightly shorter
in the cis-configured Pt derivative, which is in line with
dmso being a stronger ligand than Cl. Upon comparing the
Pt–S bond lengths, we can see that this bond is shorter in
the cis derivative than in the trans derivative, which means
that Cl is a weaker ligand than pydimH2. In summary, the
ligands can be sorted into a series with decreasing trans
influence: dmso Ͼ pydimH2 Ͼ Cl. The influence of the pyd-
imH2 ligand is very similar to the well-known ligand 1,4-
pyrazine (pyz) in the dinuclear complexes trans-[(µ-
pyz){(R2SO)PtCl2}2] (R = nBu, nPr or cBu), which show
Pt–S bond lengths of about 2.222(2) Å.[31] From the series
of ligand strengths, we assume that the cis isomer of
[(pydimH2)(dmso)PtCl2] is thermodynamically favoured,
and the trans isomer is presumably the kinetic product. The
latter finding leads to the assumption of an associative li-
gand-exchange reaction through a five-coordinate transi-
tion state (or intermediate).[30,32] This, and its superior solu-
bility, enabled us to isolate it. This is important since the
two complexes are interesting candidates as building blocks
for di- or oligonuclear (heterometallic) complexes, as the
dmso ligand can easily be replaced by other ligands. Future
work will be devoted to this subject.
NO3 , SO42–, p-TolOSO3 ) have also been reported. These
complexes also show shortened axial M–N distances of
about 1.97 Å, whereas the equatorial M–O(H) bonds lie in
the same range as those of Ni and Cu derivatives (about
2.1 Å).[14a] Unfortunately, we were not able to include the
crystal structure of a nickel derivative in our series, but we
are quite confident that its structural parameters would be
quite similar.
A third type of structure is represented by the complex
trans-[(pydimH2)2PdCl2]·2H2O (data collected in Table 5),
which is isostructural with the already reported PtII deriva-
tive.[24] The two pydimH2 ligands bind to the metal atom
through their pyridine donor function, and the OH group
is bent away. There are medium-strong hydrogen-bonding
interactions between the OH group and the co-crystallised
water molecules, with an O···H distance of 2.225(5) Å
[O···O 2.780(4) Å] and an O–H···O angle of 161.1(6)°. Due
to the steric demand of the pydimH2 ligands the complexes
are trans-configured, although the cis configuration would
be preferred on electronic grounds.[30] This finding strongly
supports the proposal that the failure to form complexes
with the bulky ligands pydipH2 and pydotH2 is due to re-
pulsion of the ancillary RЈ groups of the bulkier RRЈpyd-
imH2 ligands.
–
–
The two compounds cis- and trans-[(pydimH2)(dmso)-
PtCl2] formed from the reaction of the complex precursor
[(dmso)2PtCl2] with 1 equiv. of pydimH2 were also crystal-
lised and structurally characterised (see, for example Fig-
ure 4; crystallographic data listed in Table 5).
Molecular Structures of the Complexes in Solution
NMR Spectroscopy
Unfortunately, no NMR spectra could be obtained for
The crystal structure of the cis derivative shows a short most of the complexes due to their paramagnetism (CoII,
O–H···O contact of 1.943(5) Å [angle: 149.5(4)°; O···O NiII, CuII). A largely broadened H NMR spectrum is ob-
1
2.682(4) Å] which can be classified as an H bond of medium served for [(pydotH2)2NiCl2] (see the Supporting Infor-
strength. The analogous interaction in the trans derivative is mation), and we therefore assume a high-spin configuration
much weaker [2.108(6) Å; 146.8(4)°; 2.829(4) Å], although for this complex. The ZnII complexes generally gave good
neither is likely to have any significant impact on the molec- NMR spectra in [D6]acetone solution, and selected reso-
Eur. J. Inorg. Chem. 2009, 2271–2281
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
2275

A. Klein, S. Elmas, K. Butsch
FULL PAPER
Table 2. Selected NMR spectroscopic data for the oxido pincer ligands and their ZnII complexes.[a]
Compound
1H
3,5-Py
13C
4-Py
OH
4-Py
2,6-Py
160.3
–
165.8
162.8
C(OH)
pydimH2
7.77
7.98
7.72
8.19
7.62
7.63
7.89
7.89
7.34
7.46
7.32
7.74
7.02
7.07
6.97
6.96
4.43
136.8
64.3
[b]
[c]
[c]
[c]
[(pydimH2)ZnCl2]
pydipH2
–
–
–
4.16
6.94
5.44
5.39
7.67
7.54
137.6
72.3
72.8
[(pydipH2)2Zn]2+
141.4
R,S
pydotH2
pydotH2
137.6[d]
137.6[d]
141.4[d]
141.4[d]
164.4[d]
164.4[d]
162.8[d]
162.8[d]
76.3[d]
76.3[d]
76.7[d]
76.7[d]
R,R
[(pydotH2R,S)ZnCl2]
[(pydotH2R,R)ZnCl2]
[a] Chemical shifts (δ) in ppm, as measured in [D6]acetone. [b] Not observed due to broadening. [c] Not observed due to low solubility.
[d] Isomers could not be assigned separately.
nances are summarised together with those of the free li-
gands in Table 2 (see the Experimental Section for complete plexes [(RRЈpydimH2)2M]2+ (M = Co, Ni, Cu. Zn) in
data). MeCN solution remain unchanged for several days, thus
The spectra of most of the octahedrally configured com-
It is evident from Table 2 that coordination of the metal indicating that these species are configurationally stable in
1
ions leads to a general shift to lower field for the H and this rather polar solvent. Interestingly, [(pydimH2)2Ni]Cl2 is
13C resonances. The spectra were found to be unchanged not stable in MeCN solution as the solution immediately
even after one week in the corresponding solvent, thereby turns from light green to intense yellow. The nature of the
proving the stable coordination of the ligands. Several sets decomposition product is not yet clear, although it is likely
of signals were observed for the chiral ligand pydotH2 and that the oxido pincer ligand is replaced by MeCN mole-
its Zn complex, and we were able to assign these to the cules. The rather low energy of the ligand-field bands of the
(R,S) or (R,R) enantiomers (see Table 2).
pydipH2 derivative suggests rather weak bonding to NiII.
The NMR spectra of both crude and recrystallised [(pyd- We assume that for the slightly weaker pydimH2 ligand
imH2)2PtCl2] point to the exclusive trans configuration of MeCN readily replaces the pincer ligand to form species
this complex,[24] whereas a 1:1 mixture of cis/trans isomers such as [(MeCN)6Ni]2+ [33]
.
was observed for [(pydimH2)(dmso)PtCl2]. The 3JH
When compounds of the type [(RRЈpydimH2)2Co]-
(dmso),Pt
coupling is slightly larger for the cis complex (24.1 Hz), [CoCl4] are dissolved in the apolar solvent CH2Cl2 the com-
than for the trans derivative (20.9 Hz), which suggests that plexes undergo “coordination disproportionation”[34] ac-
the pydimH2 ligand possesses a slightly larger trans influ- cording to Equation (1).
ence than the chlorido ligand. We can also conclude that
the bulkiness of the ancillary R and RЈ groups of the
[(RRЈpydimH2)2M][MCl4] i 2 [(RRЈpydimH2)MCl2]
(1)
RRЈpydimH2 ligands has a strong impact on the observed
stereochemistry and very probably inhibits complex forma-
tion for the more bulky ligands pydipH2 and pydotH2. A
trans configuration is assumed for the palladium complex
This fully reversible process is shown for the system
[(pydotH2)2Co][CoCl4] i 2 [(pydotH2)CoCl2] in Figure 5.
The visible part of the spectrum of the first form (in pure
MeCN solution) is dominated by the absorption bands of
[CoCl4]2–, and the absorptions of the dicationic complex
[(pydotH2)2Co]2+ are obscured by this relatively strong ab-
1
[(pydimH2)2PdCl2] from the very high similarity of the H
shifts with those of the platinum analogue.
Optical Spectroscopy of the Complexes in Solution
Since NMR spectroscopy could not provide evidence for sorption. The spectrum of the neutral species [(pydotH2)-
stable coordination in solution for the majority of our com- CoCl2] (in pure CH2Cl2 solution) consists of a relatively
plexes, absorption spectra were recorded for selected exam- weak structured band with
a maximum at 628 nm
ples. Bands in the UV spectrum at around 270 nm (λ1) were (15920 cm–1) and shoulders at 587 nm (17030 cm–1) and
observed for all compounds. These bands are typical for 400 nm (25000 cm–1). These absorptions are attributed to
4
4
4
2,6-disubstituted pyridine ligands (π-π* transitions), and the AЈ2Ǟ4AЈ2, AЈ2Ǟ4EЈ, and AЈ2Ǟ4EЈЈ transitions in a
discrimination between coordinated and free ligands on this d7 high-spin system.[29,35]
basis is not feasible. The ZnII complexes do not exhibit any
The pentacoordinate complexes [(RRЈpydimH2)CuCl2]
further absorption bands, whereas the Co, Ni, Co, Pd, and with pydimH2 and pydipH2 undergo disproportionation in
Pt complexes show additional bands arising from the metal MeCN solution to yield the octahedral species [(RRЈpydim)2-
atoms. For the 3d elements Co, Ni, and Cu these are typical Cu]2+ and [CuCl4]2–. However, this disproportionation oc-
ligand-field (d-d) transitions, whereas charge-transfer tran- curs far slower than for the corresponding CoII systems.
sitions are assumed for the observed long-wavelength ab- The complex ion [CuCl4]2– is clearly characterised by the
sorption bands for Pd and Pt. A table containing all spec- intense charge-transfer band (LMCT) at 462 nm
troscopic data and assignments is provided in the Support- (21645 cm–1).[36] A high-energy shoulder visible at around
ing Information.
389 nm points to partial solvolysis in MeCN solution to
2276
www.eurjic.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2009, 2271–2281

Bis(hydroxymethyl)pyridine Ligands for the Late Transition Metals
the oxido pincer ligands and explain the failure of the bulk-
ier ligands pydipH2 or pydotH2 to coordinate to the two
platinum metals.
Future work will focus on the development of further
oxido pincer ligands of the type RRЈpydim with a strong
emphasis on chiral derivatives. The use of deprotonated li-
gands or the deprotonation reaction “on site” will also be
studied. An interesting aspect in this respect is the coupling
of the protonation/deprotonation with changes in the redox
state. Generally, our expectation that the pyridine function
will reliably bind to the metal atom and that the oxido func-
tions are much more weakly bound is largely fulfilled for
the series of complexes presented herein. The reactivity pat-
terns of these complexes that are expected to arise from
these weak interactions, for example C–H or C–X acti-
vation, will also be studied in the future in the light of cata-
lytic applications.
Figure 5. Absorption spectra of the equilibrium [(pydotH2)2Co]-
[CoCl4] i 2 [(pydotH2)CoCl2] in mixtures of MeCN and CH2Cl2.
The traces represent mixtures from 10:0 (pure MeCN), 9:1, 8:2 to
1:9 and 0:10 (pure CH2Cl2).
give [CuCl3(MeCN)]–.[36,37] The octahedral species
[(RRЈpydim)2Cu]2+ shows long-wavelength bands at
around 900 nm (11100 cm–1) accompanied by long-wave-
length shoulders. These bands can clearly be assigned to the
expected Jahn–Teller-distorted ligand-field transition. The
nickel complex [(pydotH2)NiCl2] exhibits an absorption
spectrum typical for a pentacoordinate high-spin d8 sys-
tem[29,35] and is completely stable in all solvents.
Experimental Section
General: Dimethyl pyridine-2,6-dicarboxylate, 2,6-diacetylpyridine,
2,6-bis(hydroxymethyl)pyridine, and MeLi were purchased from
Aldrich. All reactions were carried out under inert-gas conditions
by using Schlenk techniques. Solvents (CH2Cl2, thf, toluene, Et2O,
and MeCN) were dried by using an MBRAUN MB SPS-800 sol-
vent purification system.
The spectra of the square-planar PtII and PdII complexes
Instrumentation: NMR spectra were recorded with a Bruker Avance
1
n
[(pydimH2)MCl2] look very similar to the spectrum of II 300 MHz spectrometer by using a triple-resonance H, BB in-
verse probe head. Unambiguous assignment of the 1H and 13C res-
trans-[(py)2PtCl2], which confirms that the complex entity
onances was achieved from 1H TOCSY, 1H COSY, gradient-se-
is conserved in MeCN solution. Again, no spectral changes
1
lected H, 13C HSQC, and HMBC spectra. All 2D NMR experi-
were observed after several days.
ments were performed by using standard pulse sequences from the
Bruker pulse program library. Chemical shifts are quoted relative
to TMS. UV/Vis/NIR absorption spectra were measured with a
Varian Cary50 Scan or Shimadzu UV-3600 photospectrometer. El-
Conclusion and Outlook
emental analyses were performed by using a HEKAtech CHNS
EuroEA 3000 Analyzer.
Our study of the coordination chemistry of the three 2,6-
bis(hydroxymethyl)pyridine-based oxido pincer ligands
RRЈpydimH2 [R = RЈ = H (pydimH2); R = RЈ = Me
(pydipH2); R = 2-tolyl, RЈ = Me (pydotH2)] with late tran-
sition metals CoII, NiII, CuII, ZnII, PdII, and PtII has re-
vealed a wealth of structural motifs, including square-
planar (Pt, Pd), trigonal-bipyramidal (Co, Ni, Zn), square-
pyramidal (Cu), and octahedral (Co, Ni, Cu, Zn) coordina-
tion. The oxido functions coordinate to the 3d elements
with no deprotonation of the OH group but fail to bind to
the (very soft) palladium and platinum atoms. The 3d met-
als form two main complex species, which interconvert
upon changing the solvent polarity. Polar solvents promote
the formation of the dicationic species [(RRЈpydimH2)2M]
2+, whereas apolar solvents favour the neutral complexes
[(RRЈpydimH2)MCl2]. Their identity and the equilibrium
Synthesis of the Oxido Pincer Ligands
2,6-Bis(1-hydroxy-1-methylethyl)pyridine (pydipH2): This ligand
was synthesised according to a modified literature procedure.[26]
Thus, MeLi (54 mL of a 1.6  solution, 86 mmol) was added drop-
wise to dimethyl pyridine-2,6-dicarboxylate (4 g, 20.05 mmol) in thf
(100 mL) at –78 °C. After stirring at room temperature for 2 h, the
mixture was quenched with an aqueous saturated NH4Cl solution
(100 mL) and extracted with CH2Cl2 (5ϫ30 mL). The dark oil ob-
tained upon evaporation of the CH2Cl2 extracts was purified by
chromatography on silica gel by using cyclohexane/ethyl acetate
(5:1) to give an orange oil. This oil was finally crystallised from an
acetone/pentane mixture (1:1) at 0 °C to give colourless crystals.
1
Yield: 2.74 g (70%). H NMR (300 MHz, CDCl3): δ = 7.72 (t, J =
7.79 Hz, 1 H, 4-Py), 7.32 (d, J = 7.79 Hz, 2 H, 3,5-Py), 4.16 (m, 2
H, OH), 1.57 (s, 12 H, CH3) ppm. 13C NMR (75 MHz, [D6]ace-
tone): δ = 165.8 (2 C, 2,6-Py), 137.6 (1 C, 4-Py), 116.5 (2 C, 3,5-
Py), 72.3 [2 C, C(OH)], 30.1 (4 C, CH3) ppm. C11H17NO2 (195.26):
calcd. C 67.66, H 8.78, N 7.17; found C 67.51, H 8.80, N 7.16.
1
between the two forms have been established by H NMR
spectroscopy for the diamagnetic ZnII complexes and by
UV/Vis/NIR spectroscopy for the paramagnetic CoII, NiII,
and CuII derivatives. PdII and PtII form square-planar com-
plexes of the type trans-[(pydimH2)2MCl2]. The complexes
[(pydimH2)(dmso)MCl2] form cis and trans isomers (ratio
2,6-Bis(1-hydroxy-1-o-tolylethyl)pyridine (pydotH2): A freshly pre-
pared solution of o-tolylmagnesium bromide (20.4 mmol) in Et2O
(20 mL) was added dropwise to 2,6-diacetylpyridine (1632 mg,
ca. 1:1). These findings emphasize the relative bulkiness of 10 mmol) in Et2O (50 mL) at 0 °C. The reaction mixture was stirred
Eur. J. Inorg. Chem. 2009, 2271–2281
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
2277

A. Klein, S. Elmas, K. Butsch
FULL PAPER
overnight and then quenched with an aqueous saturated NH4Cl
solution (50 mL). After extraction with CH2Cl2 (3ϫ30 mL) and
evaporation of the organic solvents, the product was recrystallized
as a colourless solid from ethanol/water (1:2). Yield: 1.56 g (45%).
1H NMR (300 MHz, [D6]acetone): (R,R/S,S) isomer: δ = 7.69 (dd,
J1 = 7.30, J2 = 1.90 Hz, 2 H, 6-Tol), 7.62 (t, J = 7.76 Hz, 1 H, 4-
Py), 7.13–7.25 (m, 4 H, 4,5-Tol), 7.07 (m, 2 H, 3-Tol), 7.02 (m, 2
H, 3,5-Py), 5.39 (s, 2 H, OH), 1.95 (s, 6 H, CH3Tol), 1.88 (s, 6 H,
CH3) ppm; (R,S/S,R) isomer: δ = 7.69 (dd, J1 = 7.30, J2 = 1.90 Hz,
2 H, 6-Tol), 7.63 (t, J = 7.74 Hz, 1 H, 4-Py), 7.13–7.25 (m, 4 H,
4,5-Tol), 7.07 (d, J = 7.21 Hz, 2 H, 3-Tol), 7.02 (d, J = 7.74 Hz, 2
H, 3,5-Py), 5.44 (s, 2 H, OH), 1.95 (s, 6 H, CH3Tol), 1.86 (s, 6 H,
CH3) ppm. 13C NMR (75 MHz, [D6]acetone): δ = 164.4 (2 C, 2,6-
Py), 144.0 (2 C, 1-Tol), 137.6 (1 C, 4-Py), 131.9 (2 C, 5-Tol), 127.4
(2 C, 4-Tol), 126.5 (2 C, 2-Tol), 125.2 (2 C, 3-Tol), 125.1 (2 C, 6-
Tol), 118.1 (2 C, 3,5-Py), 76.3 [2 C, C(OH)], 30.9 (2 C, CH3), 20.6
(2 C, CH3Tol) ppm (isomers not assigned). C23H25NO2 (347.5):
calcd. C 79.51, H 7.25, N 4.03; found C 77.36, H 7.47, N 4.06.
139.6 (1 C, 4-Py), 120.8 (2 C, 3,5-Py), 64.9 [2 C, C(OH)] ppm.
C14H18Cl2N2O4Pd (455.63): calcd. C 36.90, H 3.98, N 6.15; found
C 36.91, H 3.98, N 6.13.
[(pydimH2)2PtCl2]: pydimH2 (140 mg, 1.0 mmol) was dissolved in
acetone (10 mL) and added dropwise to an aqueous solution (2 mL
H2O) of K2[PtCl4] (206 mg, 0.5 mmol). After stirring of the reac-
tion mixture at room temperature for 2 h, all volatiles were distilled
off with a rotary evaporator at 65 °C. The residue was extracted
three times with an acetone/pentane mixture (3:1) (30 mL) to give
a light yellow product upon evaporation of the remaining solvent.
1
Yield: 405 mg (77%). H NMR (300 MHz, [D6]acetone): δ = 8.17
(t, J = 7.50 Hz, 2 H, 4-Py), 7.67 (d, J = 7.50 Hz, 4 H, 3,5-Py), 5.85
(br. s, 8H CH2), 4.85 (br. s, 4 H, OH) ppm. 13C NMR (75 MHz,
[D6]acetone): δ = 164.1 (4 C, 2,6-Py), 139.6 (2 C, 4-Py), 120.8 (4 C,
3,5-Py), 63.8 (4 C, CH2) ppm. No satellites due to coupling to 195Pt
were observed.
[(pydipH2)2Co][CoCl4]: Synthesized from CoCl2·6H2O and pydipH2
and isolated as blue crystals. Yield: 377 mg (58%).
C22H34Cl4Co2N2O4 (650.19): calcd. C 40.64, H 5.27, N 4.31; found
C 40.85, H 5.35, N 4.32.
General Procedure for Complex Synthesis: In a typical reaction, the
ligand (1 mmol) was dissolved in ethanol (5 mL) and subsequently
added dropwise to a solution of the corresponding MII salt
(1 mmol) in ethanol (10–15 mL). The mixture was stirred at room
temperature overnight, then the solvent was removed under re-
duced pressure and the residue recrystallized three times from ace-
tone and pentane to afford the coloured products.
[(pydipH2)2Ni]Cl2: Synthesized from anhydrous NiCl2 and pydipH2
(2 equiv.) and isolated as a turquoise solid. Yield: 359 mg (69%).
C22H34Cl2N2NiO4 (520.12): calcd. C 50.80, H 6.59, N 5.39; found
C 50.78, H 6.61, N 5.35.
[(pydipH2)CuCl2]: Synthesized from CuCl2·2H2O and pydipH2 and
isolated as a blue powder. Yield: 310 mg (94%). C11H17Cl2CuNO2
(329.71): calcd. C 40.07, H 5.20, N 4.25; found C 40.62, H 5.40, N
4.13.
[(pydimH2)2Co][CoCl4]: Synthesized from CoCl2·6H2O and
pydimH2 in a 1:1 ratio and isolated as blue crystals. Yield: 269 mg
(50%). C14H18Cl4Co2N2O4 (537.98): calcd. C 31.26, H 3.37, N
5.21; found C 31.27, H 3.36, N 5.18.
[(pydipH2)2Cu]Cl2: Synthesized from CuCl2·2H2O and pydipH2
(2 equiv.) and isolated as a turquoise powder. Yield: 521 mg (96%).
C22H36Cl2CuN2O5 (542.98): calcd. C 47.66, H 6.68, N 5.16; found
C 47.05, H 6.55, N 5.02.
[(pydimH2)2Co][TFA]2: CoCl2·6H2O (238 mg, 1.00 mmol) and
pydimH2 (290 mg, 2.08 mmol) were stirred in trifluoroacetic acid
(10 mL) overnight. The solvent was evaporated and the resulting
oily purple crude product stored at 0 °C. After 3 weeks, purple
crystals were collected and washed three times with cold ethanol.
Yield: 201 mg (36%). C18H18CoF6N2O8 (563.27): calcd. C 38.38,
H 3.22, N 4.97; found C 39.01, H 3.12, N 4.85.
[(pydipH2)2Zn][ZnCl4]: Synthesized from ZnCl2·2H2O and pydipH2
and isolated as colourless to yellowish fine needles. Yield: 557 mg
(84%). 1H NMR (300 MHz, [D6]acetone): δ = 8.19 (t, J = 7.95 Hz,
1 H, 4-Py), 7.74 (d, J = 7.95 Hz, 2 H, 3,5-Py), 6.94 (s, 1 H, OH),
1.78 (s, 12 H, CH3) ppm. 13C NMR (75 MHz, [D6]acetone): δ =
162.8 (2 C, 1-Py), 141.4 (1 C, 4-Py), 119.5 (2 C, 3,5-Py), 72.8 [2 C,
C(OH)], 29.9 (2 C, CH3) ppm. C22H34Cl4N2O4Zn2 (663.11): calcd.
C 39.85, H 5.17, N 4.22; found C 38.45, H 5.09, N 4.11.
[(pydimH2)NiCl2]: Synthesized from anhydrous NiCl2 and pydimH2
and isolated as a bright green powder. Yield: 234 mg (87%).
C7H9Cl2NNiO2 (268.75): calcd. C 31.28, H 3.38, N 5.21; found C
30.35, H 3.26, N 5.12.
[(pydimH2)CuCl2]: Synthesized from CuCl2·3H2O and pydimH2
and isolated as turquoise crystals. Yield: 175 mg (64%).
C7H9Cl2CuNO2 (273.60): calcd. C 30.73, H 3.32, N 5.12; found C
30.76, H 3.30, N 5.13.
[(pydotH2)CoCl2]·EtOH: Synthesized from CoCl2·6H2O and
pydotH2 and isolated as blue crystals. Yield: 261 mg (50%).
C25H31Cl2CoNO3 (523.36): calcd. C 57.37, H 5.97, N 2.68; found
C 56.59, H 5.89, N 2.58.
[(pydimH2)ZnCl2]: Synthesized from ZnCl2·2H2O and pydimH2
[(pydotH2)NiCl2]·EtOH: Addition of pydotH2 to a solution of an-
hydrous NiCl2 immediately led to a pink solution. The usual
1
and isolated as a colourless solid. Yield: 203 mg (74%). H NMR
(300 MHz, [D6]acetone + some drops of water): δ = 7.98 (t, J =
7.80 Hz, 1 H, 4-Py), 7.46 (d, J = 7.80 Hz, 2 H, 3,5-Py), 4.87 (s, 4
H, CH2OH) ppm; signals for OH were not found. No 13C NMR
spectrum could be obtained due to low solubility in common or-
ganic solvents. C7H9Cl2NO2Zn (275.45): calcd. C 30.52, H 3.29, N
5.09; found C 30.09, H 3.25, N 4.89.
workup afforded
a
pink powder. Yield: 355 mg (68%).
C25H31Cl2NNiO3 (523.12): calcd. C 57.40, H 5.97, N 2.68; found
C 57.45, H 6.23, N 2.64.
[(pydotH2)CuCl2]·EtOH: Synthesized from CuCl2·2H2O and
pydotH2 and isolated as a green solid. Yield: 177 mg (55%).
C25H31Cl2CuNO3 (527.98): calcd. C 56.87, H 5.92, N 2.65; found
C 56.16, H 5.84, N 2.67.
[(pydimH2)2PdCl2]: pydimH2 (140 mg, 1.0 mmol) was dissolved in
acetone (10 mL) and added dropwise to an aqueous solution (2 mL
H2O) of K2[PdCl4] (326 mg, 1.0 mmol). The mixture was stirred at
room temperature for 1 h, then all volatiles were removed. The
light-yellow product was extracted from the residue with CH2Cl2.
[(pydotH2)ZnCl2]·EtOH: Synthesized from ZnCl2·2H2O and
pydotH2 and isolated as a colourless solid. Yield: 493 mg (93%).
1H NMR (300 MHz, [D6]acetone): (R,S/S,R) isomer (77%): δ =
7.89 (t, J = 7.86 Hz, 1 H, 4-Py), 7.81 (m, 2 H, 6-Tol), 7.67 (s, 2 H,
OH), 7.32 (m, 2 H, 5-Tol), 7.21 (m, 2 H, 4-Tol), 7.18 (m, 2 H, 3-
1
Yield: 160 mg (36%). H NMR (300 MHz, [D6]acetone): δ = 8.03
(t, J = 7.78 Hz, 1 H, 4-Py), 7.70 (d, J = 7.78 Hz, 2 H, 3,5-Py), 6.05
(d, J = 5.51 Hz, 4 H, CH2OH), 5.20 (t, J = 5.51 Hz, 2 H, OH) Tol), 6.97 (d, J = 7.86 Hz, 2 H, 3,5-Py), 2.21 (s, 6 H, CH3), 2.06 (s,
ppm. 13C NMR (75 MHz, [D6]acetone): δ = 163.5 (2 C, 2,6-Py), 6 H, CH3Tol) ppm; (S,S/R,R) isomer (23%): δ = 7.89 (t, J =
2278
www.eurjic.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2009, 2271–2281

Bis(hydroxymethyl)pyridine Ligands for the Late Transition Metals
7.86 Hz, 1 H, 4-Py), 7.81 (m, 2 H, 6-Tol), 7.54 (s, 2 H, OH), 7.32 2 C, C(OH)], 42.3 [t, J = 57.5 Hz, 2 C, CH3 (dmso)] ppm. 1H NMR
(m, 2 H, 5-Tol), 7.21 (m, 2 H, 4-Tol), 7.18 (m, 2 H, 3-Tol), 6.96 (d,
(300 MHz, [D6]acetone): trans product: δ = 8.11 (t, J = 7.76 Hz, 1
J = 7.86 Hz, 2 H, 3,5-Py), 2.06 (s, 6 H, CH3Tol), 2.00 (s, 6 H, CH3) H, 4-Py), 7.77 (d, J = 7.76 Hz, 2 H, 3,5-Py), 5.55 (d, J = 5.49 Hz,
ppm. 13C NMR (75 MHz, [D6]acetone): δ = 162.8 (2 C, 2,6-Py), 4 H, CH2), 5.04 (t, J = 5.96 Hz, 2 H, OH), 3.45 [t, JPt,H = 20.9 Hz,
141.4 (1 C, 4-Py), 138.2 (2 C, 1-Tol), 132.4 (2 C, 5-Tol), 128.8 (2 6 H, CH3 (dmso)] ppm. 13C NMR (75 MHz, [D6]acetone): trans
C, 4-Tol), 126.9 (2 C, 2-Tol), 125.5 (2 C, 6-Tol), 125.5 (2 C, 3-Tol), product: δ = 163.0 (2 C, 2,6-Py), 140.3 (1 C, 4-Py), 122.1 (t, JPt,C
120.2 (2 C, 3,5-Py), 76.7 [2 C, C(OH)], 30.1 (2 C, CH3), 21.0 (2 C,
= 34.7 Hz, 2 C, 3,5-Py), 65.2 (t, JPt,C = 38.6 Hz, 2 C, CH2), 43.9 [t,
CH3Tol) ppm. C25H31Cl2NO3Zn (529.81): calcd. C 56.67, H 5.90, JPt,C = 61.1 Hz, 2 C, CH3 (dmso)] ppm. C9H15Cl2NO3PtS (483.27):
N 2.64; found C 56.65, H 5.80, N 2.59.
calcd. C 22.37, H 3.13, N 2.90, S 6.64; found C 22.45, H 3.15, N
2.91, S 6.80.
Alternative Synthesis of [(pydimH2)2PdCl2]: pydimH2 (140 mg,
1.0 mmol) was dissolved in CH2Cl2 (10 mL) and added dropwise
to a solution of [(dmso)2PdCl2] (333 mg, 1.0 mmol) in CH2Cl2
(25 mL). A yellow precipitate formed after 15 min, and the reaction
was finished after 20 min. The solvent was then evaporated to
around half of the original volume and the precipitate washed sev-
eral times with small portions of CH2Cl2 and pentane to yield a
light yellow solid. Yield: 217 mg (48%). 1H, 13C NMR spectro-
scopic data and elemental analysis were identical to those obtained
for the material of the original synthesis method (see above).
pydipH2·HCl: The synthesis of pydipH2·HCl was originally in-
tended as a further purification of the pydipH2 ligand. Thus, HCl
gas (generated from concentrated sulfuric acid and sodium chlo-
ride) was slowly added to an Et2O solution of the ligand pydipH2
in an open flask equipped with a stirring bar. The reaction was
complete when the formation of a white precipitate had finished.
The white solid was washed several times with small portions of
1
Et2O and dried. H NMR (300 MHz, [D6]acetone): δ = 7.95 (d, J
= 7.77 Hz, 1 H, 5-Py), 7.86 (t, J = 7.77 Hz, 1 H, 4-Py), 7.59 (d, J
= 7.77 Hz, 1 H, 3-Py), 4.70 (s, 1 H, OH), 4.20 (s, 1 H, OHЈ), 1.58
(s, 6 H, CH3), 1.56 (s, 6 H, CH3Ј) ppm. C11H18ClNO2 (231.72):
calcd. C 57.02, H 7.83, N 6.04; found C 57.14, H 7.89, N 6.03.
cis/trans-[(pydimH2)(dmso)PtCl2]: pydimH2 (210 mg, 1.5 mmol)
was dissolved in CH2Cl2 (20 mL) and added dropwise to a solution
of [(dmso)2PdCl2] (500 mg, 1.5 mmol) in CH2Cl2 (50 mL). The
mixture was refluxed at 50 °C for about 55 h. The white precipitate
formed was filtered off and washed three times with small portions
of warm CH2Cl2. The yellow filtrate was concentrated to around
half of the original volume, and a yellow solid was precipitated by
adding pentane. Alternatively, the yellow material can be extracted
with Et2O, as the white solid is not soluble in that solvent. Yield:
360 mg (50% white = cis product), 205 mg (28% yellow = trans
[pydipH3]2[CoCl4]: Synthesized from CoCl2·6H2O (119 mg,
0.5 mmol) and pydipH2·HCl (2 equiv., 196.5 mg, 1.00 mmol) and
isolated as light blue crystals. Yield: 152 mg (51%).
C22H36Cl4CoN2O4 (593.28): calcd. C 44.54, H 6.12, N 4.72; found
C 44.14, H 5.99, N 4.73.
X-ray Crystallographic Studies: Crystal structure determinations
were performed at 293(2) K by using graphite-monochromated
1
product). H NMR (300 MHz, [D6]acetone): cis product: δ = 8.10
(t, J = 7.81 Hz, 1 H, 4-Py), 7.74 (d, J = 7.81 Hz, 2 H, 3,5-Py), 5.53 Mo-Kα radiation (λ = 0.71073 Å) with a IPDS II (STOE and Cie.)
(d, J = 5.51 Hz, 4 H, CH2), 5.11 (t, J = 5.85 Hz, 2 H, OH), 3.52 diffractometer. The structures were solved by direct methods for
[t, JPt,H = 24.14 Hz, 6 H, CH3 (dmso)] ppm. 13C NMR (75 MHz, the free ligand pydotH2 or by using the Patterson method for all
[D6]acetone): cis product: δ = 162.9 (2 C, 2,6-Py), 140.0 (1 C, 4-
Py), 122.0 (t, JPt,C = 24.5 Hz, 2 C, 3,5-Py), 63.8 [t, JPt,C = 28.0 Hz,
complexes (SHELXS-97)[38] and refined by full-matrix least-
squares techniques against F2 (SHELXL-97).[39] Numerical absorp-
Table 3. Crystal structure and refinement data of complexes [(RRЈpydimH2)MCl2].
Compound
[(pydimH2)ZnCl2][a]
[(pydimH2)CuCl2]·
[(pydotH2)CoCl2]·
[(pydotH2)NiCl2]·
[(pydotH2)ZnCl2]·
H2O
EtOH
EtOH
EtOH
Empirical formula
Formula mass
Crystal system
Space group
a [Å]
b [Å]
c [Å]
α [°]
β [°]
C7H9Cl2NO2Zn
275.45
C7H11Cl2CuNO3
291.62
C25H31Cl2CoNO3
523.36
C25H31Cl2NiNO3
523.12
C25H31Cl2NO3Zn
529.81
monoclinic
P21/c (No. 14)
10.5505(10)
10.5893(10)
10.7490(9)
90
119.391(4)
90
1046.34(17), 1
552
triclinic
triclinic
triclinic
triclinic
¯
¯
¯
¯
P1 (No. 2)
P1 (No. 2)
P1 (No. 2)
P1 (No. 2)
7.3053(15)
8.0944(16)
9.8959(19)
99.439(16)
109.502(15)
96.265(16)
535.61(18), 2
294
10.8545(16)
10.9814(18)
11.595(2)
87.35(2)
71.492(18)
81.867(18)
1297.4(4), 2
546
10.795(2)
10.898(2)
11.476(3)
87.36(3)
72.09(3)
82.10(2)
10.8962(16)
10.9913(16)
11.5914(19)
87.058(19)
71.406(18)
81.799(17)
1302.3(3), 2
552
γ [°]
Volume [Å3], Z
F(000)
1272.4 (5), 2
548
Reflections collected
Data/restraints/parameters
Goof on F2
Final R indices
[I Ͼ 2σ(I)]
R indices
14678
2915/0/121
1.334
R1 = 0.049,
wR2 = 0.135
R1 = 0.057,
wR2 = 0.142
0.793/–0.701
8721
2948/0/144
1.300
R1 = 0.032,
wR2 = 0.097
R1 = 0.036,
wR2 = 0.094
0.590/–0.481
5843
5843/0/304
0.535
R1 = 0.033,
wR2 = 0.057
R1 = 0.121,
wR2 = 0.057
0.272/–0.245
15575
5751/2/304
0.569
R1 = 0.054,
wR2 = 0.077
R1 = 0.252,
wR2 = 0.120
0.434/–0.357
15629
5809/0/299
0.814
R1 = 0.055,
wR2 = 0.111
R1 = 0.125,
wR2 = 0.126
0.842/–0.598
(all data)
Largest difference peak/hole
[eÅ–3]
CCDC-
701771
701773
701780
701781
701782
¯
[a] Another crystalline fraction of the compound was solved in triclinic P1, the results of the crystal structure refinement were of
comparable quality, the data has been deposited with The Cambridge Crystallographic Data Centre (CCDC-701772).
Eur. J. Inorg. Chem. 2009, 2271–2281
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
2279

A. Klein, S. Elmas, K. Butsch
Table 4. Crystal and refinement data and selected bond lengths [Å] and angles [°] for complexes of the type [(RRЈpydimH2)2M][X]2.
FULL PAPER
Compound
[(pydimH2)2Co]-
[TFA]2
[(pydimH2)2Co][CoCl4]·
5/4H2O
[(pydipH2)2Co]-
[CoCl4]
[(pydipH2)2Cu]-
Cl2·H2O
[(pydipH2)2Zn]-
[ZnCl4]
Empirical formula
Formula mass
Crystal system
Space group
a [Å]
b [Å]
c [Å]
α [°]
β [°]
C18H18CoF6N2O8
563.27
C14H18Cl4Co2N2O4·5/4 H2O C22H34Cl4Co2N2O4
C22H36Cl2CuN2O5
542.98
C22H34Cl4Zn2N2O4
663.11
560.48
650.19
triclinic
monoclinic
P21/c (No. 14)
11.0149(16)
11.5966(15)
20.456(4)
90
monoclinic
C2/c (No. 15)
17.474(4)
10.537(2)
17.204(3)
90
monoclinic
P2/c (No. 13)
9.1291(18)
10.151(2)
28.321(6)
90
96.88(3)
90
2605.5(9), 4
1140
monoclinic
P2/c (No. 13)
17.455(3)
10.5405(12)
17.231(3)
90
113.553(19)
90
2906.1(8), 1
680
¯
P1 (No. 2)
7.4407(10)
12.4969(18)
13.401(2)
70.737(17)
87.707(18)
75.065(16)
1135.2(3), 2
570
121.381(15)
90
113.55(3)
90
γ [°]
Volume [Å3], Z
F(000)
2230.7(6), 1
1134
2903.9(10), 4
1336
Reflections collected
13636
5434
12614
21099
3425
Data/restraints/parameters 5062/0/380
1162/10/296
0.538
3463/0/168
0.873
5842/0/316
0.883
3425/0/162
0.803
Goof on F2
0.911
Final R indices
[I Ͼ 2σ(I)]
Rint = 0.049
Rint = 0.256
Rint = 0.103
Rint = 0.048
Rint = 0.276
R indices (all data)
R1 = 0.039,
wR2 = 0.086
R1 = 0.069,
wR2 = 0.094
0.420/–0.418
R1 = 0.0772,
wR2 = 0.1646
R1 = 0.2784,
wR2 = 0.2031
0.298/–0.370
R1 = 0.0443,
wR2 = 0.0983
R1 = 0.0973,
wR2 = 0.1061
0.477/–0.642
R1 = 0.036,
wR2 = 0.074
R1 = 0.070,
wR2 = 0.083
0.329/–0.426
R1 = 0.192,
wR2 = 0.285
R1 = 0.100,
wR2 = 0.271
1.648/–0.403
Largest difference peak/
hole [eÅ–3]
CCDC-
701775
701774
701776
701777
701778
M–N
M–O
O–M–O
O–M–N
2.045(2), 2.032(2)
2.153(2), 2.119(2)
152.4(1), 146.9(1)
76.1(1), 76.9(1)
75.0(1), 76.9(1)
165.5(1)
2.030(7), 2.261(3)
2.115(7), 2.069(6)
150.4(3), 153.6(3)
75.9(3), 77.7(3)
76.7(3), 76.7(3)
172.3(3)
2.047(3)
2.121(3), 2.093(4)
152.4(1)
1.957(2), 1.958(2)
2.173(2), 2.156(2)
156.8(1), 156.4(1)
78.05(8), 78.36(8)
78.44(8), 78.43(8)
178.22(8)
2.038(10)
2.121(10), 2.126(9)
153.1(3)
76.6(1), 75.9(1)
76.5(4), 76.6(4)
N–M–N
NOOM/NOOM
174.5(1)
93.2(1)
174.6(5)
94.0(1)
91.6(1)
91.6(1)
91.2(1)
Table 5. Crystal and refinement data and selected bond lengths [Å] and angles [°] for complexes [(RRЈpydimH2)2PdCl2] and cis- and
trans-[(RRЈpydimH2)(dmso)MCl2].
Compound
trans-[(pydimH2)2PdCl2]·2H2O cis-[(pydimH2)(dmso)PtCl2]
trans-[(pydimH2)(dmso)PtCl2]
Empirical formula
Formula mass
Crystal system
Space group
a [Å]
b [Å]
c [Å]
C28H40Cl4N4O12Pd2
979.3 gmol–1
monoclinic
C2/m (No. 12)
7.0674(10)
15.685(3)
8.3102(11)
90
93.466(16)
90°
919.5(3), 1
496
C9H15Cl2NO3PtS
966.5 gmol–1
triclinic
C9H15Cl2NO3PtS
1933.1 gmol–1
monoclinic
P21/c (No. 14)
12.5764(16)
8.7029(9)
13.9796(15)
90
112.527(13)
90
1413.3(3), 1
912
¯
P1 (No. 2)
7.9393(11)
9.1867(13)
9.6526(14)
92.402(17)
95.430(17)
100.288(17)
688.35(17), 1
456
α [°]
β [°]
γ [°]
Volume [Å3], Z
F(000)
Reflections collected
Data/restraints/parameters
Goof on F2
Final R indices
[I Ͼ 2σ(I)]
4166
1098/0/68
1.117
R1 = 0.0318,
wR2 = 0.0897
R1 = 0.0319,
wR2 = 0.0897
0.430/–0.690
701783
7970
3073/0/159
0.993
R1 = 0.034,
wR2 = 0.076
R1 = 0.043,
wR2 = 0.078
1.462/–1.424
701784
13331
3249/0/159
0.779
R1 = 0.040,
wR2 = 0.062
R1 = 0.097,
wR2 = 0.071
1.336/–1.970
701785
R indices (all data)
Largest difference peak/hole [eÅ–3]
CCDC-
M–N
2.046(3)
2.051(5)
2.065(8)
M–Cl
M–S
2.306(1)
–
2.302(2), 2.295(2)
2.200(2)
2.298(2), 2.295(2)
2.224(3)
Cl–M–Cl
Cl–M–N
N–M–N/S
NNClClM/py
180.00
90.00(2)
180.00
99.0(1)
90.10(7)
88.0(2), 178.0(2)
91.9(2)
175.6(1)
87.3(2), 89.0(2)
177.7(2)
91.8(1)
94.6(1)
2280
www.eurjic.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2009, 2271–2281

Bis(hydroxymethyl)pyridine Ligands for the Late Transition Metals
[16] M. Melnik, M. Koman, L. Macaskova, T. Glowiak, R. Grob-
elny, J. Mrozinsky, J. Coord. Chem. 1998, 43, 159–167.
[17] M. Melnik, C. E. Holloway, J. Coord. Chem. 1999, 49, 69–73.
[18] M. Koman, M. Melnik, J. Moncol, Inorg. Chem. Commun.
2000, 3, 262–266.
tion corrections (X-RED V1.22; Stoe & Cie, 2001) were performed
after optimising the crystal shapes by using X-SHAPE V1.06
(Stoe & Cie, 1999).[40] The non-hydrogen atoms were refined with
anisotropic displacement parameters without any constraints. The
H atoms of the oxo pincer ligands’ OH groups were all found dur-
ing the refinement process, other H atoms were included by using
appropriate riding models. For further details see Tables 3, 4 and
5. CCDC-701771, -701772, -701773, -701774, -701775, -701776,
[19] A. K. Gupta, J. Kim, Acta Crystallogr., Sect. C 2003, 59, m262-
m264.
[20] C. Boskovic, W. Wernsdorfer, K. Folting, J. C. Huffman, D. N.
Hendrickson, G. Christou, Inorg. Chem. 2002, 41, 5107–5118.
-701777, -701778, -701779, -701780, -701781, -701782, -701783, [21] a) J. Yoo, E. K. Brechin, A. Yamaguchi, M. Nakano, J. C.
Huffman, A. L. Maniero, L.-C. Brunel, K. Awaga, H. Ishim-
oto, G. Christou, D. N. Hendrickson, Inorg. Chem. 2000, 39,
3615–3623; b) E. K. Brechin, J. Yoo, M. Nakano, J. C. Huff-
man, D. N. Hendrickson, G. Christou, Chem. Commun. 1999,
783–784.
-701784, -701785 contain the supplementary crystallographic data
for this paper. These data can be obtained free of charge from The
Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/
data_request/cif.
[22] S. Onaka, L. Hong, M. Ito, T. Sunahara, H. Imai, K. Inoue,
J. Coord. Chem. 2005, 58, 1523–1530.
[23] T. Kawato, H. Koyama, H. Kanatomi, Y. Muramoto, Inorg.
Chim. Acta 1991, 183, 107–112.
[24] F. D. Rochon, A. L. Beauchamp, C. Bensimon, Can. J. Chem.
1996, 74, 2121–2130.
Supporting Information (see footnote on the first page of this arti-
cle): Data from absorption spectroscopy, including selected absorp-
tion spectra, the paramagnetically broadened 1H NMR spectrum
of [(pydotH2)NiCl2], as well as figures for all 15 crystal and molecu-
lar structures.
[25] a) S. Goto, J. Velder, S. El Sheikh, Y. Sakamoto, M. Mitani, S.
Elmas, A. Adler, A. Becker, J.-M. Neudörfl, J. Lex, H.-G.
Schmalz, Synlett 2008, 9, 1361–1365; b) S. El Sheikh, H.-G.
Schmalz, Curr. Opin. Drug Discovery Dev. 2004, 7, 882–894.
[26] E. Gómez, R. Flores, G. Huerta, C. Alvarez-Toledano, R. A.
Toscano, V. Santes, N. Nava, P. Sharma, J. Organomet. Chem.
2003, 672, 115–122.
Acknowledgments
Dr. Ingo Pantenburg and Ingrid Müller from the Institute for Inor-
ganic Chemistry are thanked for the X-ray measurements. We are
also grateful to Johnson Matthey for a loan of K2PtCl4 and
K2PdCl4.
[27] T. Steiner, Angew. Chem. 2002, 114, 50–80; Angew. Chem. Int.
Ed. 2002, 41, 48–76.
[28] J. E. Huheey, E. A. Keiter, R. L. Keiter, Inorganic Chemistry:
Principles of Structure and Reactivity, 4. ed., Harper Collins
College Publishers, New York, 1993.
[29] C. Furlani, Coord. Chem. Rev. 1968, 3, 141–167.
[30] A. Klein, T. Schurr, A. Knödler, D. Gudat, K.-W. Klinkh-
ammer, V. K. Jain, S. Zálisˇ, W. Kaim, Organometallics 2005,
24, 4125–4131.
[1] J. T. Singleton, Tetrahedron 2003, 59, 1837–1857.
[2] M. Albrecht, G. van Koten, Angew. Chem. 2001, 113, 3866–
3898; Angew. Chem. Int. Ed. 2001, 40, 3750–3781.
[3] a) J. A. Davies, F. R. Hartley, Chem. Rev. 1981, 81, 79–90; b)
R. G. Pearson, Inorg. Chem. 1988, 27, 734–740; c) R. G. Pear-
son, J. Am. Chem. Soc. 1963, 85, 3533–3539.
[4] J. M. Hawkins, K. B. Sharpless, Tetrahedron Lett. 1987, 28,
[31] J. R. L. Priqueler, F. D. Rochon, Inorg. Chim. Acta 2004, 357,
2825–2828.
2167–2175.
[5] R. M. Gauvin, J. A. Osborn, J. Kress, Organometallics 2000,
19, 2944–2946.
[32] a) M. R. Plutino, M. L. Scolaro, A. Albinati, R. Romeo, J. Am.
Chem. Soc. 2004, 126, 6470–6484; b) S. Lanza, D. Minniti, P.
Moore, J. Sachinidis, R. Romeo, Inorg. Chem. 1984, 23, 4428–
4433.
[6] P. E. Fanwick, L. M. Kobriger, A. K. McMullen, I. P.
Rothwell, J. Am. Chem. Soc. 1986, 108, 8095–8097.
[7] S. Bellemin-Laponnaz, K. S. Coleman, P. Dierkes, J.-P. Mas-
son, J. A. Osborn, Eur. J. Inorg. Chem. 2000, 1645–1649.
[8] J. M. Hawkins, J. C. Dewan, K. B. Sharpless, Inorg. Chem.
1986, 25, 1501–1503.
[9] K. V. Zaitsev, M. V. Bermeshev, S. S. Karlov, Y. F. Oprunenko,
A. V. Churakov, J. A. K. Howard, G. S. Zaitseva, Inorg. Chim.
Acta 2007, 360, 2507–2512.
[10] R. Fandos, B. Gallego, A. Otero, A. Rodriguez, M. J. Ruiz, P.
Terreros, C. Pastor, Organometallics 2007, 26, 2896–2903.
[11] R. J. Fites, A. T. Yeager, T. L. Sarvela, W. A. Howard, G. Zhu,
K. Pang, Inorg. Chim. Acta 2006, 359, 248–256.
[12] J. M. Berg, R. H. Holm, Inorg. Chem. 1983, 22, 1768–1771.
[13] a) V. T. Yilmaz, S. Guney, O. Andac, W. T. A. Harrison, J. Co-
ord. Chem. 2003, 56, 21–32; b) O. Andac, S. Guney, Y. Topcu,
V. T. Yilmaz, W. T. A. Harrison, Acta Crystallogr., Sect. C
2002, 58, m17–m20.
[33] M. P. Feth, A. Klein, H. Bertagnolli, Eur. J. Inorg. Chem. 2003,
839–852.
[34] IUPAC Gold Book, International Union of Pure and Applied
Chemistry, 2005–2008, http://goldbook.iupac.org/index.html.
[35] a) M. Ciamporlini, N. Nardi, Inorg. Chem. 1967, 6, 445–449;
b) M. Ciamporlini, N. Nardi, Inorg. Chem. 1966, 5, 41–44.
[36] S. E. Manahan, R. T. Iwamoto, Inorg. Chem. 1965, 4, 1409–
1413.
[37] H. Kaczmarek, J. Photochem. Photobiol. A: Chem. 1996, 95,
61–65.
[38] G. M. Sheldrick, SHELXS-97: A program for Crystal Structure
Resolution, University of Göttingen, Germany, 1997.
[39] G. M. Sheldrick, SHELXL-97: A Program for the Refinement
of Crystal Structures, University of Göttingen, Germany, 1997.
[40] STOE X-RED, Data Reduction Program, Version 1.22/Win-
dows, STOE & Cie, Darmstadt, 2001; STOE X-SHAPE, Crys-
tal Optimisation for Numerical Absorption Correction, Ver-
sion 1.06/Windows, STOE & Cie, Darmstadt, 1999.
Received: January 9, 2009
[14] a) S. Winter, W. Seichter, E. Weber, J. Coord. Chem. 2004, 57,
997–1014; b) S. Winter, W. Seichter, E. Weber, Z. Anorg. Allg.
Chem. 2004, 630, 434–442.
[15] M. Koman, M. Melnik, Polyhedron 1997, 16, 2721–2726.
Published Online: April 16, 2009
Eur. J. Inorg. Chem. 2009, 2271–2281
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
2281