Structural, spectroscopic and magnetic properties of M[R 2P(E)NP(E)R′2]2 complexes, M = Co, Mn, e = S, Se and R, R′ = Ph or iPr. Covalency of M-S bonds from experimental data and theoretical calculations

Article abstract of DOI:10.1039/b517938f

The S/Se-containing bidentate ligands LH of the type R₂P(E) NHP(E)R′₂, E = S, Se and R, R′ = Ph or ⁱPr have been employed to synthesize ML₂ (M = Mn, Co) complexes which contain the biologically important MS₄ core. Theoretical calculations on the LH and L^- forms of the ligands probe the geometric and electronic changes induced by the deprotonation of the LH form, which are correlated with structural data from X-ray crystallography. These results reflect the flexibility of the ligands, which enables them to be rather versatile with respect to the formation of ML₂ complexes with varied geometries and MEPNPE metallacycle conformations. A series of old and new ML₂ complexes have been synthesized and their structural, spectroscopic and magnetic properties characterized in detail. The nephelauxetic ratio β of the CoL₂ complexes provides evidence of covalent interactions, whereas the EPR properties of the MnL₂ complexes are interpreted on the basis of predominant ionic interactions, between the metal center and the ligands, respectively. Additional evidence for the existence of covalent interactions in the CoL₂ complexes (R = Ph, ⁱPr, or mixed Ph/ⁱPr), is offered by comparisons between their ³¹P NMR. The aforementioned notations are supported by extensive theoretical calculations on the ML₂ (E = S, R = Me) modelled structures, which probe the covalent and ionic character of the M-S bonds when M = Co or Mn. Wider implications of the findings of the present study on the M-S covalency and its importance in the active sites of various metalloenzymes are also discussed. The Royal Society of Chemistry 2006.

Full text of DOI:10.1039/b517938f

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
Structural, spectroscopic and magnetic properties of M[R2P(E)NP(E)Rꢀ2]2
complexes, M = Co, Mn, E = S, Se and R, Rꢀ = Ph or iPr. Covalency of
M–S bonds from experimental data and theoretical calculations†‡
Dimitrios Maganas,a Sarah S. Staniland,b Alexios Grigoropoulos,a Fraser White,b Simon Parsons,b
Neil Robertson,*b Panayotis Kyritsis*a and Georgios Pneumatikakisa
Received 19th December 2005, Accepted 28th February 2006
First published as an Advance Article on the web 13th March 2006
DOI: 10.1039/b517938f
The S/Se-containing bidentate ligands LH of the type R2P(E)NHP(E)Rꢀ2, E = S, Se and R, Rꢀ = Ph or
iPr have been employed to synthesize ML2 (M = Mn, Co) complexes which contain the biologically
important MS4 core. Theoretical calculations on the LH and L− forms of the ligands probe the
geometric and electronic changes induced by the deprotonation of the LH form, which are correlated
with structural data from X-ray crystallography. These results reflect the flexibility of the ligands, which
enables them to be rather versatile with respect to the formation of ML2 complexes with varied
geometries and MEPNPE metallacycle conformations. A series of old and new ML2 complexes have
been synthesized and their structural, spectroscopic and magnetic properties characterized in detail.
The nephelauxetic ratio b of the CoL2 complexes provides evidence of covalent interactions, whereas the
EPR properties of the MnL2 complexes are interpreted on the basis of predominant ionic interactions,
between the metal center and the ligands, respectively. Additional evidence for the existence of covalent
interactions in the CoL2 complexes (R = Ph, iPr, or mixed Ph/iPr), is offered by comparisons between
their 31P NMR. The aforementioned notations are supported by extensive theoretical calculations
on the ML2 (E = S, R = Me) modelled structures, which probe the covalent and ionic character of
the M–S bonds when M = Co or Mn. Wider implications of the findings of the present study on
the M–S covalency and its importance in the active sites of various metalloenzymes are also discussed.
centers have been obtained by the reconstitution of the FeS(Cys)4
protein rubredoxin by Co(II),6 or the coordination of Co(II) to the
Introduction
Cys-rich Zn- or Cd-metallothioneins.7 The coordination of Co(II)
to S-containing ligands has also been achieved in small peptides
or “maquettes”, which retain critical structural and spectroscopic
properties of analogous Co(II) biological sites.8
On the other hand, Mn(II) is not coordinated to S-containing
amino acids in biological sites, while it prefers O- and N-
containing biological ligands.9 It should be stressed, however, that
Mn is extensively coordinated to sulfides, as for example, in the
organometallic complexes reviewed recently.10
Extensive crystallographic studies during the last few years have
revealed the existence of M–S bonds in the active site of many
metalloenzymes, involving either sulfide ions (S2−), Cys-thiolate,
Met-thioether, or dithiolenic moieties, most prominently with the
transition metals M = Fe, Ni, Cu, Zn, Mo, W.1 In this respect,
beyond the above metals, one should also consider Co, since
it has recently been shown that the enzyme ATP sulfurylase is
a metalloenzyme containing Co(II) and Zn(II) in its active site,
with the metal ion bound possibly to three S(Cys) and one
N(His) residues.2 In addition, the active site of the enzyme nitrile
reductase contains either Fe(III) or Co(III), coordinated to post-
translationally oxidized Cys ligands,3 that has been modelled by a
wealth of small complexes.4 Moreover, due to the lack of spectro-
scopic and magnetic properties of Zn(II)-containing enzyme active
sites, Zn-enzymes, like alcohol dehydrogenase, or “Zn-fingers”, are
frequently reconstituted by Co(II), with the formation of S(Cys)-
rich coordination spheres around Co(II).5 Analogous Co(II)S4
Our research groups are interested in the synthesis of complexes
of certain first-series transition elements, with S/Se-containing
bidentate ligands LH of the type R2P(E)NHP(E)Rꢀ2, E = S or
Se. These ligands are considered to be inorganic analogues of
b-diketonates, since deprotonation is always accommodated by
an extensive delocalization of p electronic density, a phenomenon
verified by X-ray crystallography and spectroscopic data.11,12
A
wealth of ML2 complexes have been synthesized and thoroughly
characterized in the past, with the reported examples involving
numerous main-group, as well as transition elements.11 The interest
in metal complexes containing ligands of this type stems mainly
from the structural flexibility of the deprotonated forms (L−),
since (i) proper selection of the R, Rꢀ groups leads to symmetrical
(R = Rꢀ) or asymmetrical coordination spheres (R = Rꢀ), and
(ii) MEPNPE metallacycles adopt either pseudo-chair or pseudo-
boat conformations, depending on the metal and the peripheral
R groups.12b
aInorganic Chemistry Laboratory, Department of Chemistry, National and
Kapodistrian University of Athens, Panepistimiopolis, GR-157 71 Athens,
Greece. E-mail: kyritsis@chem.uoa.gr; Fax: +30 210 7274782; Tel: +30
210 7274268
bSchool of Chemistry, University of Edinburgh, King’s Buildings, West Mains
Road, Edinburgh, UK. E-mail: neil.robertson@ed.ac.uk; Fax: +44 131
6504743; Tel: +44 131 6504755
† Fig. 1 in colour; Tables S1–S2; Fig. S1–S6.
‡ In memoriam to Professor Andreas T. Tsatsas.
This journal is
The Royal Society of Chemistry 2006
Dalton Trans., 2006, 2301–2315 | 2301
©

View Article Online
The structural flexibility of the ligands is explored in the present
work by theoretical calculations, which probe the structural and
electronic changes induced by deprotonation of the ligands, as well
as by their coordination to a metal ion.
EPR spectroscopy. EPR measurements were carried out at
77 K on an X-band Bruker ER200D-SCR spectrometer, connected
to a Datalink 486DX PC running EPR Acquisition System,
version 2.42 software.
The second important feature of these ligands is their large
X-Ray crystallography. X-Ray diffraction intensities were
collected with Mo-Ka radiation on a Bruker Smart APEX
CCD diffractometer equipped with an Oxford Cryosystems low-
temperature device operating at 150 K. Absorption corrections
were carried out with the multiscan procedure SADABS.16 The
structure was solved by direct methods (SHELXS97)17 and
refined by full-matrix least squares against |F|2 using all data
(SHELXL97).18 H-atoms were placed in calculated positions
and allowed to ride on their parent atoms. All non H-atoms
were modelled with anisotropic displacement parameters. The
crystallographic data of our studies are shown in the ESI†.
CCDC reference numbers 293307–293310.
˚
E · · · E “bite” (ca. 4 A), which induces less strain in the metal–
ligand ring and, therefore, allows them to accommodate the
requirements of various metal-coordination geometries.11,12b Thus,
there is a range of observed geometries, some of which are charac-
teristically distorted from ideal ones and confer specific electronic
properties to the MS4 core, that might resemble those of the active
site of certain metalloenzymes. Such a case, involving the study
of distorted tetrahedral or square planar13 Ni[R2P(S)NP(S)Rꢀ2]2
i
(R, Rꢀ = Ph or Pr) complexes, is currently under experimental
and theoretical investigation, (P. Kyritsis, N. Robertson et al.,
unpublished), in order to address critical questions on the elec-
tronic properties of NiS4-containing metalloenzyme active sites,
like hydrogenase and CO dehydrogenase.14
For crystallographic data in CIF or other electronic format see
DOI: 10.1039/b517938f
In addition to the theoretical study of the ligands, the syn-
thesis and characterization of the following new ML2 com-
plexes Co[iPr2P(S)NP(S)Ph2]2 (3), Mn[iPr2P(S)NP(S)iPr2]2 (5),
Mn[iPr2P(S)NP(S)Ph2]2 (6) and Mn[Ph2P(Se)NP(Se)Ph2]2 (7) is
described. The structures of these complexes have been determined
by X-ray crystallography. Moreover, the previously reported
complexes Co[Ph2P(S)NP(S)Ph2]2 (1), Co[iPr2P(S)NP(S)iPr2]2 (2),
Mn[Ph2P(S)NP(S)Ph2]2 (4), and Zn[Ph2P(S)NP(S)Ph2]2 (8) have
also been prepared. The whole set of complexes have been
studied by spectroscopic methods (1H and 31P NMR, IR, UV-VIS,
EPR), and their magnetic properties investigated by magnetisation
measurements between 350 and 2 K. All complexes studied
have been crystallized in a tetrahedral or pseudo-tetrahedral
geometry. The tetrahedral nature of the complexes examined in
this study is of interest from a magnetic point of view, since
all complexes are typically high-spin and paramagnetic. The
structural, spectroscopic and magnetic properties of these MS4-
containing complexes (M = Mn, Co), combined with the results
of appropriate theoretical calculations, are discussed with respect
to the relative covalency of their M–S bonds. This feature is
often discussed in the case of Cu–S, Fe–S and Ni–S containing
proteins,15 but has not yet been extensively studied in the case of
Co–S containing enzymes or “maquettes”.
Theoretical calculations. Ground-state electronic structure
calculations have been performed using Density Functional
Theory (DFT) methods, by employing the Gaussian 98 software
package.19 The hybrid (HF/DFT) Density Functional B3LYP
was applied, which consists of the non local hybrid exchange
functional, as defined by Becke’s three parameters equation and
the non local Lee–Yang–Parr correlation functional.20 For the Co
and Mn metal centers, the SDD21 effective core potential was
applied, whereas for all non-metal centers, the triple-f6–311G*
basis set with p-type polarization function was used.22 The elec-
tronic ground-state was left to full geometry optimization without
any symmetry constraint, starting either from the crystallographic
structures or from built models when the crystallographic co-
ordinates were not available. The density matrix in geometry
optimization energy calculations was converged to a tight root
mean square threshold of 10−8 au. The optimum structure, located
as a saddle point of the potential energy, was verified by the absence
of imaginary frequencies. The derived wave functions were found
free of internal instabilities. The estimation of the Atomic Orbitals
(AOs) contribution in Molecular Orbitals (MOs) of the complexes,
as well as the Overlap Population Analysis (OPA), were performed
with the AOMix software.23 The analysis of the MO compositions
in terms of fragment molecular orbitals, the construction of orbital
interaction diagrams, as well as the charge decomposition analysis
was performed using AOMix-CDA.23a The chemical bonding in
the complexes, at the B3LYP optimized ground state geometries,
was also studied by Natural Bond Orbital (NBO)24 analysis of the
appropriate density matrix, by B3LYP single point calculations.
Quantitative analysis of the NBOs, as well as the corresponding
Natural Atomic Orbitals (NAOs) and the Natural Localized
Molecular Orbitals (NLMOs) was also applied, in order to esti-
mate bond covalencies. Determination of the Natural Population
Analysis NLMO/NPA bond orders was computed using the
NAOs of the NLMOs, and the sign of these values is obtained
from the overlap integrals of Natural Hybrid Orbitals (NHOs),
located on the covalently bonded atoms.25 All NBO calculations
were performed with the NBO 5.0 software package, whereas
the NBOView 1.0 software was used for the visualization of the
NBOs and their corresponding NLMOs.26 The visualization of the
Kohn–Sham MOs was performed by the Molekel 3.4 software.27
Experimental
Materials and instrumentation
UV-VIS, IR and NMR spectroscopy. IR spectra were run in
the range 4000–200 cm−1 on a Perkin-Elmer 883 IR spectropho-
1
tometer, as KBR discs. H and 31P spectra were recorded in a
Varian Unity Plus 300 MHz instrument. The 1H and 31P chemical
shifts are relative to SiMe4 and 85% H3PO4, respectively. Electronic
spectra were recorded on a Cary 300 Varian spectrophotometer
and a Perkin-Elmer Lamda 9 spectrophotometer.
Magnetic susceptibility studies. Magnetic susceptibility mea-
surements were carried out between 2 K and 300 K, using a
Quantum design MPMS2 SQUID magnetometer with MPMS
MultiVu Application software to process the data. The magnetic
field employed was 0.1 T.
2302 | Dalton Trans., 2006, 2301–2315
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
Synthesis
Mn[Ph2P(Se)NP(Se)Ph2]2 (7). In a Schlenck-type flask were
added 20 mL CH3OH and 0.21 g (0.5 mmol) K[Ph2P(Se)-
NP(Se)Ph2]. The mixture was stirred at room temperature for
30 min, and subsequently 0.058 g (0.25 mmol) MnCl2·6H2O were
added. The mixture was stirred for 2 h and during that time
a solid pale pink product was formed, which was collected by
filtration. The product (yield 85%) was crystallized by slow mixing
of a CH2Cl2 solution of the complex and hexane, affording pink
cuboidal crystals, which were suitable for X-ray crystallography.
IR (cm−1) m(Ph2P–Se) 536 s, m(PNP) 1205br.
All manipulations were carried out under a pure Ar atmosphere,
unless otherwise mentioned. All chemical reagents used were
purchased from Aldrich. The LH ligands Ph2P(S)NHP(S)Ph2
(PhLH),28 iPr2P(S)NHP(S)iPr2 (iPrLH),29 iPr2P(S)NHP(S)Ph2
(
iPr,PhLH),30 Ph2P(Se)NHP(Se)Ph2,31 their corresponding KL salts,
as well as complexes 1,32 2,32 4,33 and 8,34 were prepared according
to published procedures.
Co[iPr2P(S)NP(S)Ph2]2 (3). In a Schlenck-type flask were
added 20 mL CH3OH and 0.21 g (0.5 mmol) K[iPr2P(S)NP(S)Ph2].
The mixture was stirred at room temperature for 30 min, and
subsequently 0.059 g (0.25 mmol) CoCl2·6H2O were added. The
mixture was stirred for 2 h and during that time a solid blue product
was formed, which was collected by filtration. The product (yield
68%) was crystallized by slow mixing of an acetone solution of the
complex and methanol (1 : 3), affording blue needle-like crystals,
which were suitable for X-ray crystallography. IR (cm−1) m(Ph2P–S)
573 s, m(iPr2P–S) 527 s, m(PNP) 1231br; dH(CDCl3) 1.49 [4H, s, CH],
4.75 [8H, br, m-C6H5], 6.03 [4H, d, 5J(PH) = 58.18 Hz, p-C6H5],
Results and discussion
Theoretical calculations on the ligands LH and L−
In the past, DFT calculations were performed on models of LH
forms of the ligands, in which the peripheral R groups were
replaced by H atoms.35 However, to the best of our knoweledge,
up to now there has been no detailed theoretical study on the
delocalization of p electronic density36 and its effect on the
structural and electronic characteristics of this family of ligands.
The existence of a small degree of p electronic delocalization
even in the LH forms, has already been noticed, demonstrated by
the P–N bond lengths that are between those of single and double
bonds, as well as the very small degree of pyramidalisation at the
N atom.12,31
3
3
7.77 [12H, d, J(PH) = 123.49 Hz, CH3], 8.41 [8H, d, J(PH) =
141.25 Hz, o-C6H5], 9.45 [12H, s(br), CH3]; dP(CDCl3) − 87.8
[s(br), Ph2P], −188.4 [s(br), iPr2P]
Mn[iPr2P(S)NP(S)iPr2]2 (5). In a Schlenck-type flask were
added 20 mL CH3OH, 0.15 g (0.5 mmol) iPr2P(S)NHP(S)iPr2 and
0.056 g (0.5 mmol) potassium butoxide (C4H9OK), in order to
deprotonate in situ the ligand. The mixture was stirred at room
temperature for 30 min, and subsequently 0.049 g (0.25 mmol)
MnCl2·6H2O were added. The mixture was stirred for 2 h and
during that time a solid pale pink product was formed, which was
collected by filtration. The product (yield 86%) was crystallized
by slow mixing of a CH2Cl2 solution of the complex and hexane,
affording pink cuboidal crystals, which were suitable for X-ray
crystallography. IR (cm−1) m(PS) 549 m, m(PNP) 1223br.
In order to investigate the increase of the delocalization of
p electronic density upon deprotonation of LH, as well as its
influence on bond lengths and angles, DFT calculations were
performed on the LH and L− forms of the following four ligands
(Fig. 1): Me2P(S)NHP(S)Me2
(
MeLH),37 PhLH,38 iPrLH,29 and
iPr,PhLH,30 starting from the published crystallographic coordinates.
Taking into consideration the existence of the ligand ion PhL− in
two different geometries, depending on the counter cation,12a,39
the L− forms were produced from the corresponding LH forms
by substracting the N-proton and then leaving the structure to
fully optimize its geometry, with tight criteria of convergence. The
aim was to investigate the effects of the additional N lone pair
(LP)40 that is created upon deprotonation, on the structural and
electronic properties of the ligands.
Mn[iPr2P(S)NP(S)Ph2]2 (6). In a Schlenck-type flask were
added 20 mL CH3OH and 0.21 g (0.5 mmol) K[iPr2P(S)NP(S)Ph2].
The mixture was stirred at room temperature for 30 min and
subsequently 0.058 g (0.25 mmol) MnCl2·6H2O were added. The
mixture was stirred for 2 h and during that time a solid pale pink
product was formed, which was collected by filtration. The product
(yield 85%) was crystallized by slow mixing of a CH2Cl2 solution
of the complex and hexane, affording pink cuboidal crystals, which
were suitable for X-ray crystallography. IR (cm−1) m(Ph2P–S) 576 s,
m(iPr2P–S) 528 s, m(PNP) 1233br.
All ligands were optimized in C1 symmetry without any
constraint criteria. As can be seen from the data in Table 1, the
optimized structures of the LH ligands are in excellent agreement
with the corresponding X-ray structures, since the differences are
in the accepted limits for a B3LYP level DFT calculation. Upon
deprotonation of the LH forms, there are changes in the following
three structural features:
Table 1 Selected structural properties from the crystal structures (exp) and the optimized geometries (calc) of the LH and L− forms of the ligands
MeLH
iPrLH
iPr,PhLH
PhLH
R = Ph/Me,
MeL−
calc
iPrL−
calc
iPr,PhL−
calc
PhL−
calc
i
Rꢀ = Pr
exp
calc
exp
calc
exp
calc
exp
calc
˚
av. P(R)–S/A
1.950
1.678
1.966
1.718
2.017
1.610
1.936
1.948
1.674
1.703
1.075
129.5
114.8
1.962
1.968
1.721
1.737
1.013
130.4
115.9
2.013
2.016
1.602
1.621
1.919
1.670
1.971
1.718
2.011
1.587
ꢀ
˚
av. P(R )–S/A
1.947
1.973
2.023
1.615
˚
av. P(R)–N/A
ꢀ
˚
av. P(R )–N/A
1.688
0.975
130.7
114.6
1.738
1.019
134.5
116.1
˚
av. N–H/A
0.894
133.2
110.9
1.015
134.8
112.1
1.015
136.8
114.9
av. P–N–P/◦
av. P–S–N/◦
137.1
118.5
141.0
119.9
141.5
120.7
131.7
115.7
159.3
118.6
This journal is
The Royal Society of Chemistry 2006
Dalton Trans., 2006, 2301–2315 | 2303
©

View Article Online
Fig. 1 Ground state optimized structures of the LH and L− forms of the ligands, as well as of complexes 11 and 12.
(a) P–S and P–N bond lengths. In the optimized structures
of all ligand ions (L−) mentioned above and presented in Fig. 1, the
P–S bonds are lengthened, whereas the P–N bonds are shortened,
compared to the corresponding LH forms (Table 1), a feature
strongly correlated to p delocalization. Therefore, it is confirmed
that deprotonation of the ligands is the prime factor that enhances
p delocalization along the EPNPE fragment.
further. These observations will be further discussed below with
respect to pyramidalisation effects.
In the majority of the ML2 complexes, the P–N–P angle varies
between 132◦ and 142◦, depending on the R groups attached to
P atoms.11,12b Therefore, it is mainly deprotonation, accompanied
by an increase of p delocalization, that should be considered as
the main reason for the enlargement of P–N–P angles upon the
formation of the ML2 complexes. This proposal is also supported
by theoretical calculations on the MMeL2 complexes (M = Co,
Mn), described in the following paragraphs.
(b) Enlargement of the P–N–P angles. Electronic delocal-
ization has also been considered to affect the magnitude of
the P–N–P angles. In the crystal structure of PhL− with bis
(triphenylphosphine) iminium as the counter cation, a linear
P–N–P fragment has been observed.12a Theoretical calculations,
presented below, demonstrate the opening of the P–N–P angle as
a result of deprotonation, in the absence of any counter cation.
In the case of iPrL− and iPr,PhL−, the P–N–P angle is increased
from 134.5◦ and 130.4◦ in the LH form, to 141.0◦ and 141.5◦
in the L− form, respectively, whereas in MeL− the P–N–P angle
remains almost unchanged, compared to the corresponding LH
form. These values are typical of the P–N–P angles observed in
previously reported pseudo-tetrahedral ML2 complexes contain-
ing these ligands,29,32,37 as well as the complexes 3 and 6. In the
case of PhL−, the P–N–P angle (159.3◦) approaches linearity even
(c) Angles between the P–S–N planes. Another characteristic
structural feature of the LH forms is the relative positions of the P–
S–N planes. This has already been described for a series of LH lig-
ands in terms of anti or syn conformations, in which the S–P · · · P–
S torsion angles are 150–180◦ and <90◦, respectively.11c The factor
that decides which structure is adopted by a given LH ligand, is
the nature of the peripheral R groups. It is of interest to note
that according to published crystallographic data, all ligands con-
taining the SPNPS fragment adopt the anti conformation, which
seems to be energetically favourable, except for those possessing
iPr groups, which adopt the syn conformation.11c This trend is also
observed in our theoretical calculations, for both the protonated
and deprotonated forms of the ligands. It should be stressed that
2304 | Dalton Trans., 2006, 2301–2315
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
i
upon deprotonation, the ligands that possess Pr groups retain
(b) Origin of p delocalization and chemical bonding. NBO
analysis of the spin-restricted B3LYP density matrix with respect
to 6–311G* basis set, verifies that delocalization of p electronic
density stems from distribution of the S and N LPs among
both P atom p-orbitals.40 The increase of p electronic density
delocalization upon deprotonation is a direct consequence of the
additional N LP. Depictions of these considerations are provided
by the visualization of the proper NBOs and the corresponding
NLMOs of N and S LPs. In the case of LH forms, both N and
S LP NLMOs demonstrate p delocalization towards the P atoms,
which is manifested as a distortion of the associated N and S LP
NBOs (Fig. 2i,ii). This is also the case in the L− forms. However,
the existence of the additional N LP upon deprotonation enhances
p delocalization, as is mirrored in the contour plot of the second
N LP NLMO (Fig. 2iii).
their syn conformation, whereas all others adopt the anti one. This
shows that in the absence of a counter cation, the values of the
S–P · · · P–S torsion angles are mainly affected by the steric effects
of the R groups. Therefore, flexibility of the system when R = Me
(small size),37 Ph (planar),38 or even Bu (freedom of rotation),41
results in large torsion angles, which is not the case when the
ligand possesses the bulky and not easily rotated iPr groups.29,30
Trends in p delocalization among the LH and L− forms
(a) Pyramidalisation effects in the LH and L− forms. The low
degree of pyramidalisation of the N atom has been considered as
evidence that p electron delocalization already exists in the LH
forms.12,31 The crystal structures of iPr,PhLH30 and PhLH38 show that
the N atom lies between 0.18 A and 0.22 A out of the plane of
its substituents (sum of bond angles close to 360◦), whereas in the
case of MeLH,37 both P atoms, N and H are coplanar (sum of bond
angles 360◦ exactly), indicating close to sp2 and sp2 hybridization
of N, respectively. It should be noted that in the crystal structure
of iPrLH,29 no pyramidalisation of N is observed, since both P
atoms, N and H lie in the same plane. However, our theoretical
˚
˚
˚
analysis reveals that the N atom lies 0.21 A out of the plane of its
substituents. The planarity observed in the crystal structure may
be due to packing interactions and hydrogen bonding.
By calculating the NHO compositions of N atom character
in the P–N bonds for the LH forms, we predict an sp1.7–1.9
hybridization for N, thus explaining the near planarity of the three
bonds (both P–N and N–H). Deprotonation results in a further
decrease of the NHO composition of N character in the P–N
bonds, giving sp1.1–1.5 hybridization.
Fig. 2 Selected NBOs (top) and the corresponding NLMOs (bottom) for
LH (i,ii), and L− (i, ii, iii) forms of the ligands. All contour plots were
generated with NBOView 1.0 using a value of 0.03, and step size of 0.03.26
The results presented in Table 2 reflect the opening of the P–
N–P angle and its tendency to become linear, as shown by the
NHO compositions of N character in P–N bonds. Therefore, in
the case of MeLH, the small difference in the hybridization of N
between the LH and L− forms, corresponds to a negligible change
in the P–N–P angle upon deprotonation. On the other hand, the
N hybridization in PhLH is strongly reduced from sp1.7 to sp1.1 upon
deprotonation, inducing the largest opening of the P–N–P angle.
The determination of NLMO/NPA bond orders also reflects the
expected differences regarding P–S and P–N bonds, since the for-
mer are notably decreased and the latter are considerably increased
upon deprotonation. For example, the NLMO/NPA bond orders
calculated for the P–N bonds increase from approximately 0.5 in
PhLH to 0.626 in PhL−, suggesting greater double bond character.
On the contrary, the P–S bond orders diminish from approximately
1.21 to 1.104, indicating smaller double bond character after
deprotonation (Table 3). These results are in excellent agreement
i
Ligands bearing Pr peripheral groups lie in the middle between
these two extreme cases.
Table 2 NHO composition of NBOs containing N character in the P–N
bonds for the optimised structures of the LH and L− forms
Table 3 NLMO/NPA bond orders for the optimised structures of the
LH and L− forms of the ligands
Ligand
MeLH
MeL−
NBO
Occupancy
N character (%) N hybridization
NLMO/NPA bond orders
P3–N5
P4–N5
P3–N5
P4–N5
P3–N5
P4–N5
P3–N5
P4–N5
1.87795
1.88883
1.98276
1.98634
1.98298
1.98297
1.98263
1.98235
1.98240
1.98303
1.98342
1.98313
1.98261
1.98253
1.98551
1.98551
78.3
78.6
70.7
71.4
74.1
74.2
72.3
72.2
73.2
74.4
71.2
72.4
73.7
73.6
72.5
72.5
sp1.7
sp1.8
sp1.7
sp1.5
sp1.8
sp1.9
sp1.5
sp1.5
sp1.9
sp1.8
sp1.5
sp1.4
sp1.7
sp1.7
sp1.1
sp1.1
Bond
MeLH
MeL−
iPrLH
iPrL−
P3–N5
P4–N5
P3–S1
P4–S2
0.509
0.502
1.263
1.253
0.678
0.643
1.107
1.107
0.496
0.494
1.213
1.204
0.631
0.630
1.074
1.078
iPrLH
iPrL−
iPr,PhLH P3–N5
P4–N5
NLMO/NPA bond orders
iPr,PhL−
PhLH
PhL−
P3–N5
P4–N5
P3–N5
P4–N5
P3–N5
P4–N5
Bond
iPr,PhLH
iPr,PhL−
PhLH
PhL−
P3–N5
P4–N5
P3–S1
P4–S2
0.521
0.485
1.223
1.223
0.663
0.613
1.103
1.109
0.498
0.507
1.216
1.211
0.626
0.626
1.104
1.104
This journal is
The Royal Society of Chemistry 2006
Dalton Trans., 2006, 2301–2315 | 2305
©

View Article Online
1
Table 4 Selected 1H and 31P–{ H} NMR data at 298 K in CDCl3
1H chemical shifts/ppm
31P chemical shifts/ppm
Ph groups
iPr groups
CH3
Compound
Ph2P
iPr2P
2J (31P–31P)
o-C6H5
m-C6H5
p-C6H5
CH3
C–H
1
−32.8
8.85 (16), br
8.41 (8), d
6.58 (16), br
4.75 (8), br
5.97 (8), br
6.03 (4), d
2
−331.9
10.62 (24), br 7.30 (24), br
9.45 (12), br 7.77 (12), d
1.45 (8), s
1.49 (4), s
3
−87.8 (1)
−188.4 (1)
8
37.6
929
64.4
67.8
1030
34.3
57.7
17.8 Hz
30.8 Hz
PhLH47
iPrLH29
iPr,PhLH30
91.3
100.0
51.5
Table 5 Selected IR data (cm−1) for ligands and complexes 1–7
Compound
v(P–N–P)
v(PhP–E)a
v(iPrP–S)
PhLH34
935–925–920
933–904–878
935–907–880
937–926–918
1210 (br)
1218 (br)
1231 (br)
1217 (br)
1223 (br)
1233 (br)
1205 (br)
648 (s)
—
624 (s)
546 (s)
563 (s)
—
573 (s)
564 (s)
—
576 (s)
536 (s)
—
646 (m)
613 (s)
iPrLH29
iPr,PhLH30
[Ph2P(Se)]2NH31
132
232
3
—
560 (s)
527 (s)
—
549 (m)
528 (s)
—
4
5
6
7
Fig. 3 Molecular structure and atom numbering of complex 3 (ORTEP
diagram, 50% thermal ellipsoids). Atoms with labels ending in “_2” are at
equivalent positions (−x,y,1/2 −z).
a E = S or Se.
with the 31P NMR and IR spectroscopic data (Tables 4 and 5).
Greater double bond character of the P–N bonds is reflected as an
increase of m(PNP) in the IR spectra of the deprotonated ligands.
Moreover, the electronic density on P atoms is increased, since
˚
˚
other, ranging between 2.3061(4)A and 2.3356(4) A, possibly due
to different R groups (Ph or iPr) attached to the P atoms. However,
this is not the case for the analogous Co[Ph2P(S)NP(S)Me2]2
complex, in which the Co–S bond lengths are all equivalent to
each other.42
P N is less polarized than P S, hence the 31P chemical shifts
move to lower frequencies after deprotonation.
=
=
The geometrical and electronic changes upon deprotonation of
LH described above, explain the flexibility of this family of ligands
that enables them to be rather versatile in structural and metal-
coordinating terms.11,12 Moreover, the DFT calculations presented
so far, prove that the additional N LP and the necessity of the extra
charge distribution that is generated upon deprotonation, is the
dominant factor that imposes the structural changes observed by
X-ray crystallography in these ligands.
As expected, the P◦–N–P angles are increased compared to
iPr,PhLH, from 129.5(3) in the latter, to 139.21(8)◦ in 3, whereas
in the case of the CoPh,MeL2, P–N–P angles are barely affected,
compared to Ph,MeLH.42 This observation provides experimental
evidence concerning the effect of the iPr groups on the enlargement
of the P–N–P angle. The same phenomenon has also been
observed in the case of the iPrLH ligand and the pseudo-tetrahedral
complexes it forms with Mn(II) (5), as well as Co(II),32 Cd(II)
and Zn(II) ions.29 In these complexes, the P–N–P angles vary
Description of structures of ML2 complexes
◦
between 137◦ and 143 . Presence of the Pr bulky groups seems
i
CoiPr,PhL2 (3). To the best of our knowledge, complex 3 is
the second Co(II) complex with an asymmetric LH ligand, the
first being the complex Co[Ph2P(S)NP(S)Me2]2.42 Selected bond
lengths and angles of complex 3 are listed in Table S1† and the
ORTEP representation of its crystal structure is presented in Fig. 3.
The complex has crystallographically-imposed twofold sym-
metry and the CoS4 core exists in a distorted tetrahedral form,
as depicted from the magnitude of S–Co–S angles, which range
between 95.84(2)◦ and 114.225(13)◦, the S · · · S non bonding
to be responsible for the opening of the P–N–P angle which does
not take place when the ligand possesses other R groups such
as Me37 or OPh.43 Even in complexes 132 and 433 bearing PhL−,
the P–N–P angle does not change, compared to PhLH. Therefore,
although deprotonation enhances the P–N–P angle enlargement,
as discussed above, steric effects imposed by the peripheral groups
seem to regulate the magnitude of the P–N–P angle in the ML2
complexes.
˚
The P–S and P–N bond lengths in complex 3 are 2.0194(5) A
˚
˚
˚
˚
˚
distances ranging between 3.467 A and 3.898 A, as well as the
dihedral angles between the CoS2 pairs being 87.2◦, 81.3◦ and
83.9◦. Moreover, the Co–S bond lengths are not equivalent to each
and 2.0361(5) A, and 1.5826(12) A and 1.5915(12) A, respectively,
having, as expected, intermediate magnitudes between single and
double bonds, compared to iPr,PhLH.30
2306 | Dalton Trans., 2006, 2301–2315
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
The CoSPNPS ring in complex 3 adopts a pseudo-boat con-
formation, but in contrast to all other known Co(II) complexes
with this type of ligands, Co and N are the “bow” and the “stern”
of the boat instead of P and S (Fig. S1†). This should be the
case for all tetrahedral complexes involving Ph,iPrLH, as it was also
observed for the corresponding Mn(II) complex 6. The Co–S–P
angles are of the same order as those in complex 6, at 98.553(18)◦
and 101.835(18)◦. These angles are considered to be in the expected
range for M–S–P angles in a six-membered pseudo-boat ring,
with M (M = Co, Mn) and N atoms at the apices. This leads to an
axial–equatorial interaction between Ph and iPr groups attached
to different P atoms on the same chelate ring (Fig. S1†). However,
the axial and equatorial positions are also distorted due to the very
specific boat ring geometry. Complexes 3 and CoPh,MeL2 provide a
rare example of different MSPNPS metallacycle conformations
upon variation of the peripheral R groups. CoPh,MeL2 adopts a
chair conformation and consequently the interactions between the
corresponding peripheral groups in the same chelate ring are axial–
axial, implying relaxation of the chelate ring due to the presence
of Me groups42 (Fig. S2†).
single and double, due to the delocalization of p electronic density.
The P–N–P angles in complex 5 are increased to 138.20(11)◦ and
139.87(12)◦ compared to iPrLH (131.58◦).
The MnSPNPS ring adopts a pseudo-boat conformation, with
the P and S atoms at the apices, being the “bow” and the “stern”
of the boat respectively, in agreement with other metal complexes
of similar ligands12b (Fig. S3).† The Mn–S–P angles, ranging
between 104.75(3)◦ and 105.60(3)◦, are the expected ones for a
boat conformation with P and S at the apices. In this conformation,
the interactions between the iPr groups on different P atoms of the
same chelate ring are axial–equatorial, thus affording a more stable
conformation compared to a pseudo-chair one.33
MniPr,PhL2 (6). Complex 6 was crystallized in space group P1,
with one molecule per unit cell. Space group P1 is rather unusual
for crystal structures of non-enantiopure materials. Selected bond
lengths and angles of complex 6 are listed in Table S1† and the
ORTEP representation of its molecular structure is presented in
Fig. 5.
MniPrL2 (5). Selected bond lengths and angles of complex 5
are listed in Table S1†. The representation of the ORTEP plot of
its molecular structure is presented in Fig. 4.
Fig. 4 Molecular structure and atom numbering of complex 5 (ORTEP
diagram, 50% thermal ellipsoids).
Fig. 5 Molecular structure and atom numbering of complex 6 (ORTEP
diagram, 50% thermal ellipsoids).
The Mn–S bond lengths are typical of Mn–S unstrained bonds.
It should be noted that, as in the case of complex 3, the Mn–S
bond lengths are not equivalent. The compound can be described
as a spiro-bicyclic structure with Mn being the spiro-atom. The
MnS4 core adopts a slightly distorted tetrahedral geometry, as it
can be reflected by the magnitude of the S–Mn–S angles, which
lie between 107.35(2)◦ and 111.53(2)◦. The endocyclic bite angles
S(1)–Mn–S(2) and S(3)–Mn–S(4) are 109.59(2)◦ and 110.82(2)◦,
respectively, with the former being nearly the angle of an ideal
tetrahedral geometry. Planes defined by the endocyclic S–Mn–S
angles are nearly perpendicular to one another (angle between
planes is 88.10◦). Therefore, the molecule is very close to having
an ideal tetrahedral geometry. It is only the slight deviation from
the angle of an ideal tetrahedron, as well as the asymmetric bond
lengths, that break this symmetry.
The Mn–S bond lengths are again typical of unstrained Mn–S
˚
˚
bonds, ranging between 2.4348(10) A and 2.4578(10) A, although
they are crystallographically non-equivalent to each other. The
MnS4 core adopts a distorted tetrahedral geometry, as can be
deduced from the S–Mn–S angles that range between 96.91(3)◦
and 117.01(4)◦, as well as the value (78.99(3)◦) of the angle between
S(1)–Mn–S(2) and S(3)–Mn–S(4) planes. The P–N–P angles are
increased, compared to iPr,PhLH,30 from 129.5(3)◦ in the latter, to
141.2(2)◦ and 140.8(2)◦ in complex 6.
It is of interest to note that the apices of the pseudo-boat
conformation of the metallacycle are occupied by Mn and N (Fig.
S1),† instead of the expected S and P atoms. This fact is rather
unusual for this type of complex in tetrahedral geometry, as it has
up to now been observed only in complex 3 and in FeMeL2.44 It
˚
˚
The average P–S (2.028 A) and P–N (1.592 A) bond lengths are
enlarged and shortened, respectively, compared to the free ligand
i
should be pointed out that the interactions between the Pr and
iPrLH.29 In the latter, the P–S bond lengths are 1.941(1) A and
Ph groups on different P atoms in the same chelate ring are again
axial–equatorial, although the axial and the equatorial positions
are now distorted due to a different boat conformation (Fig. S1).†
˚
˚
˚
˚
1.949(1) A, whereas the P–N ones are 1.682(3) A and 1.684(2) A.
Therefore, the P–S and P–N bonds in complex 5 are between
This journal is
The Royal Society of Chemistry 2006
Dalton Trans., 2006, 2301–2315 | 2307
©

View Article Online
Mn[Ph2P(Se)NP(Se)Ph2)]2 (7). Selected bond lengths and
angles of complex 7 are listed in Table S1† and the ORTEP
representation of its molecular structure is presented in Fig. 6.
Complex 7 was synthesized in order to assess the structural and
spectroscopic changes induced by replacing S by Se in complex 4.
observed. However, these values are consistent with those observed
for complex 1,34 and show the L− forms to be strong-field ligands.
The nephelauxetic parameter (b = Bꢀ/Bfree ion) is calculated as 0.68
(since B for the free Co(II) ion is 971 cm−1), indicating a significant
degree of covalency for such Co(II)S4 systems.8a,45,46
NMR spectroscopy
To the best of our knowledge, the 1H NMR spectrum of complexes
1 and 2 have not been reported up to now. The 1H NMR spectrum
of 1 at 298 K reveals broad signals for each set of aromatic protons,
i.e. d = 8.85 ppm (ortho), 6.58 ppm (meta) and 5.97 ppm (para).
Similar behaviour has been reported for tetrahedral paramagnetic
Ni(II) complexes bearing PhL−.47 On the other hand, this is not
the case for the diamagnetic complex ZnPhL2 (8), where the ortho
and para aromatic proton resonances are not separated. In the
case of complex 2, the presence of the paramagnetic Co(II) center
(S = 3/2) strongly affects, as expected, the chemical shifts of the
iPr peripheral groups. The methyl groups become magnetically
inequivalent and extremely deshielded, giving rise to two broad
signals at 10.62 ppm and 7.30 ppm, respectively, while the C–
H proton is observed at 1.45 ppm. Complex 3 shows a very
Fig. 6 Molecular structure and atom numbering of complex 7 (ORTEP
diagram, 50% thermal ellipsoids).
1
complicated H NMR spectrum, illustrating again the effect of
unpaired electrons of Co(II) on the chemical shifts.
˚
The Mn–Se bond lengths vary between 2.5344(11) A and
˚
2.5691(11) A, but they are slightly longer compared to typical
Proton-decoupled 31P NMR spectra of complexes 1–3 were
recorded at 298 K in CDCl3. Table 4 incorporates both previously
published32 and new data from this work. The dramatic reduction
in the 31P chemical shifts compared to the corresponding deproto-
nated ligands can be attributed to the paramagnetic Co(II) center.
On the other hand, the large difference in chemical shifts between
1 and 2 (−32.8 ppm and −331.9 ppm, respectively) cannot solely
be explained by the electron-donating effect of the iPr groups. It is
worth noticing that a single peak at −150.0 ppm was reported for
complex 2,32 whereas a peak at −331.9 ppm was observed in this
work. The latter was double-checked by monitoring the spectrum
also at 500 MHz (Bruker), and therefore, we believe that the value
quoted in ref. 32 must be erroneous.
Mn–S bonds in the complexes 4–6, due to the larger radius of the
Se atom compared to the S atom. This spiro-bicyclic Mn(II)Se4
compound adopts a slightly distorted tetrahedral conformation,
with the Se–Mn–Se angles varying between 102.90(4)◦ and
113.75(4)◦. The Se · · · Se non bonding distances range between
˚
˚
4.276 A and 4.264 A, and the dihedral angles between MnSe2
pairs are 87.41◦, 87.01◦, and 84.37◦. As described extensively
above, the P–N–P angles are slightly increased compared to the
corresponding LH form, being 132.2(3), 136.3(3) in complex 7
and 132.3(2)◦ in LH.31 The MnSePNPSe ring adopts the usual
pseudo-boat conformation with the P and Se atoms at the apices.
As is expected, due to this chelate ring geometry, the interaction
between the Ph groups on different P atoms on the same chelate
ring is axial–equatorial (Fig. S3).†
Furthermore, this immensely up-field shifted new resonance
for complex 2, discloses the fact that complex 3, bearing the
asymmetrical iPr,PhL− ligand, exhibits intermediate chemical shifts
compared to complexes 1 and 2. The 31P NMR spectrum of
complex 3 reveals two signals for the non-equivalent P atoms,
without any well resolved hyperfine couplings, at −87.8 ppm for
PPh2 and at −188.4 ppm for PiPr2. In other words, PPh2 is shielded,
while PiPr2 is deshielded compared to 1 and 2 which contain the
symmetrical ligands PhL− and iPrL−, respectively. We propose that
this trend is explained via an electron density flow from the PiPr2
Electronic spectroscopy
An ideal tetrahedral d7 ion is expected to show transitions from
the ground 4A2 state to the excited 4T2 (m1), 4T1(F) (m2) and
4T1(P) (m3) states.45 The electronic spectrum of 3 (0.5 mM in
CH2Cl2) showed ligand-field bands at 16 181 cm−1, 14 144 cm−1,
13 123 cm−1 (shoulder) and 6630 cm−1, consistent with Co(II) d–d
transitions that are typical of a tetrahedral Co(II)S4 coordination
environment. The band at 16 181 cm−1 is believed to arise from a
formally forbidden transition to a doublet state,45a the transitions
at 14 144 and 13 123 cm−1 to m3, and the broad unresolved band
centred at 6630 cm−1 to m2. The m1 transition is expected to be in the
IR region (ca. 3000 cm−1) and thus it is not observed. Splittings
within the observed bands may arise due to distortion from ideal
symmetry or spin–orbit coupling.46b A further band was observed
at 32 154 cm−1 with a shoulder at 27 100 cm−1, assigned as S →
Co(II) charge transfer. From the ligand-field bands, the values of
Dt = 3980 cm−1 and Bꢀ = 660 cm−1 can be derived using the Tanabe–
Sugano diagram. There is some uncertainty in this determination
due to the difficulty in assigning the centre of the broad bands
i
fragment carrying the electron donating Pr groups, to the PPh2
fragment bearing the electron withdrawing Ph groups, via the
central metal atom.
This trend has not been previously considered, although it seems
to also appear, to a less significant extent, in the thoroughly
studied protonated forms of the ligands (Table 4), due to their
delocalized p electronic density. However, the presence of a central
metal ion with unpaired electrons is clearly promoting, if not
altering, this electron density flow, since, apart from the Co(II)
complexes discussed here, similar behaviour is documented in the
corresponding series of tetrahedral paramagnetic complexes with
Ni(II) (S = 1), as will be reported elsewhere. The participation
2308 | Dalton Trans., 2006, 2301–2315
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
of paramagnetic metal ions in this electronic channel is further
supported by the comparison of the 31P NMR spectra of the
respective series of Zn(II) complexes 8 (this work), ZniPrL2 (9)29
and ZniPr,PhL2 (10),30 in which the same trend is observed, albeit
with minor differences in 31P chemical shifts that resemble those of
the free ligands (Table 4). The diamagnetic nature of Zn(II) seems
to allow electron density flow only among the chelate rings.
The above results can be considered as evidence of extensive
delocalization of electronic density from the Co(II) center onto the
ligands in complexes 1–3, due to the presence of highly covalent
Co(II)–S bonds.
similarity of the EPR parameters in solution would suggest that
the complexes do not differ greatly in geometry when in solution.
Therefore, for the series of the MnL2 complexes, it seems likely that
the deviations in their geometry in the solid state reflect packing
effects rather than electronic properties of the ligands.
The hyperfine coupling constant A1 for [MnCl4]2− is 79.3 G and
the value of A1 has previously been taken as evidence of the level of
covalency in the Mn–ligand bonding.48c,d The significantly larger
values of A for 4–7 in comparison with [MnCl4]2− (Table 6) are
consistent with location of the spin density on 4–7 mainly on the
metal centre and not the ligands and thus a low component of
covalent bonding in the metal–ligand interaction. Interestingly, 7
appears to be the least covalent of the complexes with a distinctly
larger value of A than the sulfur-donor complexes 4–6.
IR spectroscopy
The IR spectrum of complexes synthesized in this work exhibit
the expected and well documented trends, regarding the P–N and
P–S bond orders.29,31,39 The P–N bonds are strengthened, while
the P–S(Se) bonds are weakened, as can be deduced from the
comparison of the m(PNP) and m(PS) or m(PSe) between the metal
complexes and the free ligands (Table 5). Higher energies of the
m(PNP), correspond, not only to stronger P–N bonds, but to larger
values of PNP angles, as well (Table S1).†
CoL2 complexes. The CoL2 complexes 1, 2 and 3 studied in
this work are EPR silent in the X-band at temperatures as low
as 77 K. We attempted to record spectra at room temperature
(in solution and as powders) and also frozen at liquid nitrogen
temperature, however, these were all silent. EPR silent Co(II)
species have been observed before, and the effect was attributed to
rapid spin relaxation through spin–lattice interactions.49
Magnetic susceptibility studies
EPR spectroscopy
Magnetisation versus temperature was recorded for complexes
1–7 at temperatures ranging from 2 K to 340 K, at 0.1 T and
the molar magnetic susceptibilities, corrected for a diamagnetic
contribution, were plotted against temperature (Fig. 8). Data for
1–7 could all be fit in the high-temperature region to the Curie–
Weiss model with Curie constants of C = 1.89 (1), C = 2.17 (2), C =
1.85 (3), C = 4.42 (4), C = 4.25 (5), C = 4.56 (6) and C = 4.73 (7)
MnL2 complexes. Room temperature EPR studies of com-
plexes 4, 5, 6 and 7 showed a characteristic six-line pattern arising
from coupling to the I = 5/2 Mn nucleus. Frozen glass EPR
studies of 4–7 show a signal of 16 lines comprising 6 main peaks
with 5 sets of doublets due to spin forbidden transitions (Fig. 7,
Table 6). This pattern is characteristic of Mn(II) in a tetrahedral site
with an isotropic g parameter, consistent with the nearly perfect
tetrahedral geometry observed in the solid state for 4–7.48 The
Fig. 8 A plot of the square of the effective magnetic moment of 3 against
temperature showing a sharp drop at low temperature due to zero-field
splitting of the ground state. The solid line is a fit to the model in ref. 51
and yields a zero-field splitting parameter of 24 K.
Fig. 7 Frozen glass EPR spectrum of complex 5 at 9.48 GHz and 77 K.
2−
Table 6 EPR parameters for complexes 4–7 compared to MnCl4
Complex
g
A1/G (main line splitting, average)
A2/G (Doublet splitting, average)
4
2.06
2.07
2.01
2.01
2.00
89.5
93.7
90.9
97
22.1
23.4
25.2
31
5
6
7
[MnCl4]2−48c
79.3
—
This journal is
The Royal Society of Chemistry 2006
Dalton Trans., 2006, 2301–2315 | 2309
©

View Article Online
emu K mol−1. Complexes 1–7 show a negligible Weiss constant,
indicating no significant intermolecular interactions. This is to
be expected as the X-ray crystal structures show only limited
independent paramagnetism (TIP) is estimated according to
TIP = 8Nb2/D using an averaged value of D).
i
short contacts between the peripheral groups (Ph or Pr), which
Covalency of the M(II)–S bonds from experimental data
do not mediate spin interactions efficiently. Thus, the peripheral
groups effectively shield the metal centers from intermolecular
interactions, since the M · · · M distances between molecules are
MnL2 complexes. The Mn(II)–S bonds in the MnL2 complexes
studied in this work, appear to have a relatively low covalent
character, since spin density is maily located on the central metal
ion Mn(II), as is deduced from the EPR data. The Mn–Se bonds
in complex 7 seem to be slightly less covalent compared to the
Mn–S bonds. This might be due to the smaller electronegativity of
Se compared to S, or to less well matched orbital energies for the
Mn–Se compared to the Mn–S bonding interaction.
˚
large (of the order of 10 A). Therefore, complexes 1–7 behave like
nearly perfect paramagnetic systems.
High spin Co(II) and Mn(II) have 3 and 5 unpaired electrons,
respectively, and no first order spin–orbit contribution, giving
a theoretical spin-only Curie constant of 1.875, and 4.377 emu
K mol−1 respectively, which is in close agreement with the
experimental results. For the Mn(II) complexes, the g-values are
within errors consistent with those determined by EPR (see above).
Literature examples of tetrahedral Co(II) complexes typically
have higher Curie constants, in the region of 2.6, due to mixing in
of low-lying excited states.50 In the case of 2, this may account for
the higher value of C as this complex is the closest to a tetrahedral
geometry, whereas 1 and 3 show larger distortions that will split
the three-fold degeneracy of the first excited state and so mixing-in
of an orbital component to the magnetic moment will be much
reduced. This leads to values closer to the spin-only value.
The data for complexes 1 and 3 were fit over the entire
temperature range using an expression that accounts for the
observed zero-field splitting of the ground state which results in
a drop in effective magnetic moment at low temperature.51 From
these fits, values of the zero-field splitting parameter of J/kB =
20 K and 24 K were obtained for 1 and 3, respectively. These values
are largely comparable to literature data but are slightly larger than
typical values51 possibly arising from the strength of the L− ligand
field. A comparable fit for complex 2 was not possible, due to
more complex magnetic behaviour at lower temperature, and no
further analysis was attempted in the absence of low-temperature
structural data.
We sought to fit the data for complexes 1 and 3 using eqn (1),
which is appropriate for the powder susceptibility of an axially
distorted Co(II) four coordinate system and this should apply
exactly to our complexes.52 Using a range of reasonable values
for the parameters e.g. (in cm−1) D1 = 3300, D2 = 3700, d (zero-
field splitting) = 10, k = − 160, we could successfully reproduce
the data for the complexes. However, we were not able to obtain
an optimised fit, since there are too many parameters to be
determined from the relatively simple curve. Thus, the data are
consistent with this model but we cannot use the data to extract
the resulting parameters. To fit the parameters additional data
would be required, such as oriented single crystal magnetic data,
however, the crystals available were not sufficiently large to enable
this type of experiment.
CoL2 complexes. The nephelauxetic ratio b has been used in
the past as an indicator of covalency for Co(II)–S containing
complexes, with values ranging between 0.6 and 0.7.46 The value
of 0.68 that was obtained for complex 3 in this work, is well within
these limits and proves that the Co(II)–S bonds in this complex
have considerable covalent character.
The value of the spin–orbit coupling parameter k is considered
as another indicator of covalency. Values of k within 60 to 90%
of the free ion value (−178 cm−1) have been obtained for a series
of tetrahedral [Co(II)X4]2− complexes, in which the b values range
between 0.65 to 0.80.53
A third indicator of covalency, is the value of the orbital
reduction parameter k, which for Co(II)S4 complexes was deter-
mined to be smaller compared to other Co(II) complexes.51 More
specifically, for Co(II)S4 chromophores, values of k are 0.7,51 for
a CoN4 chromophore 0.8,54 for Co(II)O4 0.9,55 and for [CoCl4]2−
0.92.56 These data were considered as indicative of more extensive
covalency of Co(II)–S bonds compared to the other cases.
Theoretical calculations on the ML2 complexes (M = Mn, Co)
The complexes CoMeL2 (11) and MnMeL2 (12) were chosen as
models, in order to investigate the electronic properties of the ML2
(M = Co, Mn) complexes. Complex 11 has already been synthe-
sized and structurally characterized by X-ray crystallography, but
the crystallographic coordinates are not available in the Crystal-
lographic Database. By comparing the published crystallographic
coordinates37 to the previously crystallized complex 1,32 we found
similarities that led us to build a theoretical model of 11 from the
crystal structure of 1. The same procedure was applied for the
as yet unknown complex 12, by using the previously published
structure of complex 4.33 Complexes 11 and 12 are expected to be
paramagnetic with 3 and 5 unpaired electrons, respectively.
For complexes 11 and 12, the Density Functional wavefunctions
(i.e. the Kohn Sham orbitals) were calculated in two ways: first,
a spin-restricted calculation (ROB3LYP, ꢂS2ꢃ = 3.75 and ꢂS2ꢃ =
8.75) was carried out as the simplest description of an open shell
system.57 In this case, the total spin density is reflected in either
the HOMO or the LUMO wave functions (Singly Occupied MOs,
SOMOs), since all other orbitals are either doubly occupied or
doubly unoccupied. However, since the spin polarization effect is
important, spin-unrestricted calculations (UB3LYP, ꢂS2ꢃ ≈3.75,
and ꢂS2ꢃ ≈8.75) were also performed.57 Single point calculations
by using the TZVP basis set for all atoms, on the ground state
calculated geometries, were then performed for the visualization
of the frontier MOs and the NBO analysis. In order to describe
v = (Nb2/3kT)[5y2 + 10x2 + (2d/kT)(y2 − x2)] + TIP
x = 1 − (4kk/D2)
(1)
y = 1 − (4kk/D1)
Eqn (1): N = Avogadro’s number, b = the Bohr magneton,
k = Boltzman constant, d = zero-field splitting of the ground 4B1
4
4
state, D1 = splitting between ground B1 state and B2 state and
D2 = splitting between ground 4B1 state and 4E state. Temperature-
2310 | Dalton Trans., 2006, 2301–2315
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
Table 7 Selected NLMO/NPA bond orders for MeL, 11 and 12 in TZVP basis set
[(SPMe2)2]N−
Bond
11
12
Bond Order
Bond
Bond Order
Bond
Bond Order
av P–N
av P–S
—
0.67
1.09
—
av. P–N
av. P–S
av. Co–S
0.71
0.97
0.22
av.P–N
av.P–S
av. Mn–S
0.70
0.96
0.18
the bonding between the metal and the ligands, the ML+ and L−
molecular fragments were chosen for the charge decomposition
analysis, as well as the MO composition of the whole molecules in
terms of molecular fragments.58
both bond lengths (Table 8) and bond orders suggest that the
covalent contribution in the M–S bonds in CoMeL2 is higher
compared to MnMeL2.
The chemical bonding in transition metal complexes is usually
described in terms of covalent and ionic interactions between the
metal centers and the ligands:
Properties of the M–S bonds
The comparison between the NBOs, in complexes 11 and 12, de-
scribing free and coordinated MeL− ligands indicates that P–S and
P–N bonds are only slightly further weakened and strengthened,
respectively, due to complexation. The P–N–P angles in 11 and
12 are very similar to those of MeL−, again implying that in the
absence of steric interactions between R peripheral groups, P–N–
P angles in tetrahedral ML2 complexes are mainly affected by the
deprotonation and not by the complexation of the ligands (Tables 7
and 8).
In ML2 complexes, MeL− lies at higher energy than the previously
calculated free MeL−, by 6.75 Kcal mol−1 (M = Co) and 7.83 Kcal
mol−1 (M = Mn), with respect to the TZVP basis set. In other
words, it seems that ligands of this family become energetically
poised to coordinate to metal ions.
The Co–S bond in 11 is shorter than the Mn–S bond in 12.
As it has been reported previously,58 shorter bond lengths usually
correspond to higher bond orders, although variations have been
observed. Various types of bond order evaluations (NLMO/NPA,
Meyers, Wiberg, Lowdin) support this hypothesis (Table 7). The
Co–S bond order is greater than the Mn–S one (Table 7); therefore,
(i) Ionic character of the M–S bonds
It has been shown58 that for an M–S(thiolate) bond, the ionic
character is mirrored in the absolute values of the atomic charges
of the M and S atoms, as higher values correspond to higher ionic
character. The M–S bonds in 12 presents a higher degree of ionicity
compared to 11, as can be derived from the NPA population
analysis (Table 9).
(ii) Covalent character of the M–S Bonds
The covalent interaction between two molecular fragments is
usually described in terms of electron donation. Molecular orbitals
HOMO-0 (Spr*), −1 (Spp*), −2 (Spp), −3 (Spr), −5(Lpr) of the
MeL− molecular fragment (Fig. 9) participate in M–S bonds. It
should be pointed out that electron density is centered on S
atoms in the first four MOs, while in HOMO-5 electron density is
distributed among SPNPS rings.
Table 8 Selected structural properties from the crystal structures (exp)
and the optimized geometries (calc) of MMeL2 complexes
11
1
12
4
Fig. 9 Frontier MOs of the MeL− molecular fragment.
Bond
av. M–S (A)
Calc
Exp
Exp
Calc
Exp
˚
2.376
2.065
1.613
109.4
102.6
117.2
128.7
2.314
2.021
1.586
109.8
102.4
116.9
128.9
2.329
2.011
1.588
113.9
100.3
118.3
129.5
2.481
2.064
1.610
108.1
102.5
118.1
131.6
2.443
2.013
1.588
111.8
99.8
The metal centers in 11 and 12 typically contain M2+ ions.
However, calculated NPA partial charges are lower than +2
(Table 9), suggesting ligand to metal electron donation and thus
revealing the covalent character of the M–S bonds. Calculation
of the Mulliken and Natural Magnetic spin densities of the metal
ions and the sulfur atoms, also reveals that part of the electron
˚
av. P–S (A)
˚
av. P–N (A)
av. S–M–S endo (◦)
av. M–S–P (◦)
av. S–P–N (◦)
av. P–N–P (◦)
118.7
133.2
Table 9 NPA charges and MPA/NPA spin densities for 11 and 12
11
12
Atom
NPA charges
MPA/NPA spin densities
Atom
NPA charges
MPA/NPA spin densities
Co
S1
S2
S3
S4
+1.08
−0.70
−0.70
−0.70
−0.70
2.521/2.501
0.105/0.113
0.105/0.113
0.106/0.113
0.106/0.113
Mn
S5
S7
S8
S9
+1.26
−0.74
−0.74
−0.74
−0.74
4.769/4.438
0.036/0.110
0.037/0.110
0.037/0.110
0.037/0.110
This journal is
The Royal Society of Chemistry 2006
Dalton Trans., 2006, 2301–2315 | 2311
©

View Article Online
Table 10 Total contributions of unoccupied fragment molecular orbitals
to the occupied orbitals of complexes 11 and 12
(Fig. 10), since the a MOs are occupied in both Mn(II) and Co(II),
and thus contribute equally to the covalent character of the M–
S bonds in both cases. The a spin LUMO of the ML+ fragment
(64–65% in M 4s character) is mixed with the HOMO-2 (Spp)
and the HOMO-3(Spr) of the L− fragment (Fig. S4).† This results
in a small and relatively similar covalent interaction between the
two fragments for both 11 and 12. This is also reflected in the
contribution of the a-spin ML+ unoccupied fragment molecular
orbitals to the occupied orbitals of the complexes as a whole
(Table 10).
11
12
Donation (L → ML+−) and Polarization of ML+
% (a-spin UFO of ML+)
% (b-spin UFO of ML+)
21.0
142.9
23.1
36.9
Donation (ML+ → L) and Polarization of L
In the case of 12, the b LUMOs 0, 1, 2, 3, 4 of the MnL+
fragment derive from the 3d metal orbitals (Fig. S5),† as it is
supported by MPA. Since the Zeff of Mn2+ is large, the difference in
energy between the thiolate-based HOMOs of L− and metal-based
LUMOs of MnL+ is also large. As a consequence, a weak MO
interaction is observed, producing a quite high HOMO-LUMO
gap (4.6 eV). Contribution of the 3d orbitals in the occupied
MOs of 12 corresponds to values smaller than 4%, revealing
negligible interaction with the ligand MOs, and thus a minor
covalent contribution.58
The covalent character in the Mn–S bond emanates from mixing
of the LUMO (30% in Mn 4s character) of the MnL+ fragment
with the HOMOs of the L− fragment (Fig S5†).58 This additional
interaction causes a further ligand to metal charge donation
and the contribution of the b-spin MnL+ unoccupied fragment
molecular orbitals to the occupied MOs of complex 12 increases
to 36.9% (Table 10).
% (a-spin UFO of L)
6.1
6.6
0.19 e
0.37 e
5.8
4.2
0.14 e
0.30 e
% (b-spin UFO of L)
a-spin L → ML+ charge donation
b-spin L → ML+ charge donation
density is transferred from the central metal atom to the sulfur
atoms (Table 9).
The covalent character of the M–S bond can be depicted by
the interaction of the ML+ and L− fragment orbitals. Ligand to
metal electron donation can be considered as the mixing of the
unoccupied MOs of the ML+ fragment with the occupied MOs
of the L− fragment. The Charge Decomposition Analysis reveals
that the metal–ligand covalent interraction is dominant, involving
charge donation from the L− fragment to the ML+ one (Table 10).
On the contrary, back donation is negligible.58
The discussion on the ineraction diagrams of the molecular
fragments will be focused on the b frontier MOs of the fragments
Fig. 10 b-spin energetic levels of the frontier MOs of complexes 11 and 12.
2312 | Dalton Trans., 2006, 2301–2315 This journal is The Royal Society of Chemistry 2006
©

View Article Online
On the contrary, in the case of 11, metal orbitals dz2 and dyz
that were unoccupied in 12, become occupied are stabilized in
the HOMO manifold. The sulfur character in the metal based
LUMOs increases to ∼19% while 3d metal character in the ligand
based HOMOs Spp and Spr increases to 8% and 3%, respectively
(Fig. 10). Thus fragment mixing is enhanced and the energy of the
LUMO manifold is lowered, resulting in a smaller HOMO-LUMO
gap (3.6 eV).
It is noteworthy that HOMO-12 originates from 3dyz of Co(II)
and Spr and Lpr of the ligand (Fig. 10, S6†), showing clearly the
r covalent character of the Co–S bond. It should be remembered
that in the case of 12, the b spin LUMOs (Mn 3d character)
do not participate in any interactions, thus elucidating the lower
covalent character of the Mn–S bond compared to the Co–S one,
as well as the smaller ligand to metal charge transfer for complex
12 (Table 10).
Additional information on the estimation of the M–S covalency
can be obtained by determining the number of the metal–ligand
shared electrons (Table 11).25 The covalency in the M–S bond has
r character, as can be seen by the corresponding NBOs in Fig. 11.
In agreement with the experimental data described above, the
covalency in the Co–S bond estimated by theoretical calculations
is found to be larger compared to the Mn–S bond. The extent of
covalency in the Mn–S and Co–S bonds, arising from p ligand
to metal electron donations is reflected in the picture of the S LP
NBOs and the corresponding NLMOs for the CoL2 and MnL2
complexes, respectively (Fig. 11).
Fig. 11 (A) Co–S and Mn–S bonding interactions within the NBO basis,
showing the r covalent character of the M–S bonds. (B) NBOs (up) and
the corresponding NLMOs (bottom), showing the extence of covalency in
the M–S bonds due to p ligand to metal donation. All contour plots were
generated with NBOView 1.0 using a value of 0.03, and step size of 0.03.26
of plastocyanin, ascorbate oxidase and nitrite reductase, as well
as the di-copper and S(Cys)-bridged site of cytochrome oxidase.
In the latter case, the extensive delocalization of electronic density
is thought to result in lower reorganization energy, facilitating
electron-transfer by this site. Moreover, the covalent nature of
Fe–S bonds is manifested in the electronic and redox properties
of Fe/S-containing rubredoxins and ferredoxins.15,59 Numerous
other metalloenzymes have been shown to contain covalent M–
S bonds that control the electronic properties of their active site
needed for specific catalytic functions. Representative examples
include the Fe-containing enzymes nitrile hydratase,25 and super-
oxide reductase,60 as well as the Ni superoxide dismutase.61
On the other hand, although covalency has been observed in
various Co(II)–S complexes, to the best of our knoweledge, there
has been no XAS or any other study of the covalency of Co–S in the
active site of Co–S containing enzymes or “maquettes”. The work
presented here supports previous studies on Co(II)–S containing
complexes, showing that the Co(II)–S bonds are typically covalent,
in a large variety of coordination spheres. It would be expected that
similar effects operate in the active sites of the Co(II) enzymes,
and it would be interesting to study covalency aspects of Co(II)–S
bonds in the active sites of these enzymes.
Covalency of M–S bonds in biological systems
Compared to the above estimation of covalency in M–S containing
sites by experiment and theory, a more direct experimental
method, that of S K-edge X-ray absorption, has recently been
developed by Solomon and colleagues, and applied to a series of
inorganic and bioinorganic systems.15,59 According to this analysis,
the intensity of the pre-edge peaks in the XAS spectrum are
direct indicators of M–S covalency. Rather strong covalency is
established in the case of Cu(II)–S and Ni(II)–S complexes. On the
other hand, Co(II)–S is considered as a typically covalent bond,
whereas the Mn(II)–S bond is shown to lack any considerable
covalent character.59a Preliminary results by S K-edge XAS studies
on the covalency of Mn(II), Co(II), and Ni(II) complexes with
the ligands used in the present study, are in agreement with the
above observations, (P. Kyritsis, E. I. Solomon et al., unpublished).
XAS S K-edge studies have also been carried out on many met-
alloenzymes containing M–S bonds.15,59 Functional consequences
of the covalency of M–S bonds have been proposed for the Cu(II)–
S(Cys) bond as the exit point in the electron transfer mechanism
Table 11 Calculated M–L bond covalencies in terms of shared electrons for complexes 11 and 12. Bond covalencies were calculated with respect to the
6–311 G* basis set
11
12
Bond
Bond covalency
Total covalency
1.624 e−
Bond
Bond covalency
Total covalency
1.463 e−
Co–S1
Co–S2
Co–S3
Co–S4
0.401 e−
0.394 e−
0.408 e−
0.421 e−
Mn–S5
Mn–S7
Mn–S8
Mn–S9
0.376 e−
0.358 e−
0.374 e−
0.355 e−
This journal is
The Royal Society of Chemistry 2006
Dalton Trans., 2006, 2301–2315 | 2313
©

View Article Online
The situation with respect to Mn(II)–S bonds in biology is
completely different. Mn is found in many enzymes, the most char-
acteristic being the O2 evolving complex (OEC) in Photosystem
II.9 However, a large number of enzymes would implicate Mn, for
its sensing and transportation.62 To the best of our knoweledge,
there has been no well defined biological system in which Mn
is coordinated to S-containing ligands. It is characteristic that
attempts to coordinate Mn to S(Met) in a molecular variant of the
Mn transport regulator of Bacillus subtilis were not successful.63
The lack of Mn–S containing biological sites can be explained
in terms of unfavorable hard Mn(II)–soft (S) interactions and the
pronounced preference of Mn for O- and N-containing ligands. It
is of interest to note, that the bond between Mn(II) and the softer
(compared to S) Se in complex 7, is the least covalent among the
series 4–7.
The lack of preference of Mn(II) for S-containing ligands, was
utilized by nature since the OEC of Photosystem II employs Mn–
O chemistry, unimpeded by sulfide reactions, to carry out the H2O
splitting reaction in PSII.9 The Mn(II)S4 core of complex 4 was
once considered to resemble in some aspects the Fe(III)S4 active
site in rubredoxins, since it has the same electronic configuration
(high-spin 3d5).64 However, there are certainly critical differences
between the two sites. For example, the high covalency of Fe(III)–S
bonds in rubredoxin is considered one of the factors that would
control the reduction potential of the Fe centre, and hence its
electron transfer reactivity.59b
to the National Center for Scientific Research Demokritos, and
the Laboratory of Physical Chemistry, University of Athens, for
using their computing facilities. P. K. and G. P. would like to
thank the Special Account of the University of Athens (grants
KA 70/4/7575 & KA 70/4/5735), as well as the Empirikion
Foundation, for financial support. A. G. would like to thank the
Greek State Scholarship Foundation for a postgraduate studies
grant.
References
1 See reviews in Bioinorganic Enzymology, ed. Richard H. Holm and
Edward I. Solomon, Chem. Rev., 1996, 96, 2237.
2 O. Y. Gavel, S. A. Bursakov, J. J. Calvete, G. N. George, J. J. G. Moura
and I. Moura, Biochemistry, 1998, 37, 16225.
3 A. Miyanaga, S. Fushinobu, K. Ito and T. Wakagi, Biochem. Biophys.
Res. Commun., 2001, 288, 1169.
4 J. A. Kovacs, Chem. Rev., 2004, 104, 825.
5 (a) L. Banci, A. Bencini, C. Benelli, D. Gatteschi and C. Zanchini,
Struct. Bonding, 1982, 52, 37; (b) M. T. Werth, S.-F. Tang, G. Formicka,
M. Zeppezauer and M. K. Johnson, Inorg. Chem., 1995, 34, 218;
(c) A. B. Ghering, J. E. Shokes, R. A. Scott, J. G. Omichinski and
H. A. Godwin, Biochemistry, 2004, 43, 8346, and references therein.
6 S. W. May and J.-Y. Kuo, Biochemistry, 1978, 17, 3333.
7 M. Good and M. Vasˇa´k, Biochemistry, 1986, 25, 3328.
8 (a) J. R. Anglin and A. Davison, Inorg. Chem., 1975, 14, 234; (b) A.
Lombardi, D. Marasco, O. Maglio, L. Di Costanzo and V. Pavone,
Proc. Natl. Acad. Sci. USA, 2000, 97, 11922; (c) A. L. Nivorozhkin,
B. M. Segal, K. B. Musgrave, S. A. Kates, B. Hedman, K. O. Hodgson
and R. H. Holm, Inorg. Chem., 2000, 39, 2306; (d) A. K. Petros, S. E.
Shaner, D. L. Tierney and B. R. Gibney, Inorg. Chem., 2004, 43, 4793;
(e) V. Nanda, M. M. Rosenblatt, A. Osyczka, Z. Getahun, P. L. Dutton,
J. G. Saven and W. F. DeGrado, J. Am. Chem. Soc., 2005, 127, 5804.
9 The Biological Chemistry of the Elements, J. J. R. F. da Silva and R. J. P.
Williams, Oxford University Press, 2nd edn, 2001, ch. 14.
10 R. D. Adams, O.-S. Kwon and S. Miao, Acc. Chem. Res., 2005, 38, 183.
11 (a) T. Q. Ly and J. D. Woollins, Coord. Chem. Rev., 1998, 176, 451;
(b) I. Haiduc, J. Organomet. Chem., 2001, 623, 29; (c) C. Silvestru
and J. E. Drake, Coord. Chem. Rev., 2001, 223, 117; (d) J. Parr, M. B.
Smith and M. R. J. Elsegood, J. Organomet. Chem., 2002, 664, 85; (e) I.
Haiduc, Dichalcogenoimidodiphosphinato Ligands in Comprehensive
Coordination Chemistry II: From Biology to Nanotechnology, ed.
J. A. McCleverty and T. J. Meyer, in Fundamentals, ed. A. B. P. Lever,
2004, Elsevier, Amsterdam, vol. 1, pp. 323–347; (f) P. Moore, W.
Errington and S. P. Sangha, Helv. Chim. Acta, 2005, 88, 782, and
references therein.
Conclusions and perspectives
In this work, a detailed theoretical study of the electronic
properties of the R2P(S)NHP(S)Rꢀ2 ligands (LH) is presented. The
structural changes that take place upon deprotonation of LH, as
well as the subsequent enhancement of p delocalization among the
SPNPS fragments, are elucidated by well known computational
methods.
In addition, new ML2 complexes ofCo(II) and Mn(II) containing
these ligands are synthesized in order to complete the whole
series M[R2P(S)NP(S)Rꢀ2]2, R, Rꢀ = Ph and/or iPr. Combination
of spectroscopic data and theoretical calculations on the ML2
(M = Mn, Co) complexes, reveal a higher ionic character of the
Mn–S bonds compared to Co–S. The CoL2 complexes are shown
to contain Co–S bonds of considerable covalent character. This
aspect is rather general, irrespective of the specific nature of their
coordination sphere, since similar behaviour has been observed
in previously described Co(II)S4 containing complexes. Therefore,
it is expected that it will also manifest itself in the active sites of
Co(II)–S containing enzymes or “maquettes”.
12 (a) I. Haiduc, R. Cea-Olivares, S. Hernandez-Ortega and C. Silvestru,
Polyhedron, 1995, 14, 2041; (b) J. D. Woollins, J. Chem. Soc., Dalton
Trans., 1996, 2893.
13 E. Simo´n-Manso, M. Valderrama and D. Boys, Inorg. Chem., 2001, 40,
3647.
14 (a) A. Volbeda and J. C. Fontecilla-Camps, Coord. Chem. Rev., 2005,
249, 1609; (b) D. J. Evans, Coord. Chem. Rev., 2005, 249, 1582.
15 E. I. Solomon, B. Hedman, K. Hodgson, A. Dey and R. K. Szilagyi,
Coord. Chem. Rev., 2005, 249, 129.
16 G. M. Sheldrick. SADABS version 2004/1. University of Go¨ttingen,
Germany.
The relative covalencies of the M–S bonds (M = Mn, Co)
in complexes 1–7 are currently being addressed by more direct
experimental techniques, such as XAS and EPR/ENDOR spec-
troscopy. Our common efforts are also focused on the study of the
magnetic and spectroscopic properties of the corresponding NiL2
complexes. (P. Kyritsis, N. Robertson et al., in preparation).
17 G. M. Sheldrick. SHELXS97. University of Go¨ttingen, Germany.
18 G. M. Sheldrick. SHELXL97. University of Go¨ttingen, Germany.
19 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A.
Robb, J. R. Cheeseman, V. G. Zakrzewski, J. A. Montgomery, Jr, R. E.
Stratmann, J. C. Burant, S. Dapprich, J. M. Millam, A. D. Daniels,
K. N. Kudin, M. C. Strain, O. Farkas, J. Tomasi, V. Barone, M. Cossi, R.
Cammi, B. Mennucci, C. Pomelli, C. Adamo, S. Clifford, J. Ochterski,
G. A. Petersson, P. Y. Ayala, Q. Cui, K. Morokuma, D. K. Malick,
A. D. Rabuck, K. Raghavachari, J. B. Foresman, J. Cioslowski, J. V.
Ortiz, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi,
R. Gomperts, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y.
Peng, A. Nanayakkara, C. Gonzalez, M. Challacombe, P. M. W. Gill,
B. G. Johnson, W. Chen, M. W. Wong, J. L. Andres, M. Head-Gordon,
E. S. Replogle and J. A. Pople, Gaussian 98, revision A.9; Gaussian,
Inc.: Pittsburgh, PA, 1998.
Acknowledgements
We would like to thank Dr S.I. Gorelski for his very helpful com-
ments, as well as Dr C. Makedonas and Mr D. Liakos for providing
valuable assistance in the computational studies. We are grateful
2314 | Dalton Trans., 2006, 2301–2315
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
20 (a) A. D. Becke, J. Chem. Phys., 1993, 98, 5648; (b) P. J. Stephens, F. J.
Devlin, C. F. Chabalowski and M. J. Frich, J. Phys. Chem., 1994, 98,
11623.
43 M. Nouaman, Z. Zak, E. Herrmann and O. Navratil, Z. Anorg. Allg.
Chem., 1993, 619, 1147.
44 M. R. Churchill and J. Wormald, Inorg. Chem., 1971, 10, 1778.
45 (a) F. A. Cotton and M. Goodgame, J. Am. Chem. Soc., 1961, 83,
1777; (b) R. H. Holm and F. A. Cotton, J. Chem. Phys., 1960, 32, 1168;
(c) R. H. Holm and F. A. Cotton, J. Chem. Phys., 1959, 31, 788.
46 (a) A. Davison and D. L. Reger, Inorg. Chem., 1971, 10, 1967; (b) R. W.
Lane, J. Ibers, R. B. Frankel, G. C. Papaefthymiou and R. H. Holm,
J. Am. Chem. Soc., 1977, 99, 84; (c) D. Coucouvanis, D. G. Holah and
F. J. Hollander, Inorg. Chem., 1975, 14, 2657; (d) M. Vasˇa´k, J. H. R.
Ka¨gi, B. Holmquist and B. L. Vallee, Biochemistry, 1981, 20, 6659.
47 R. Ro¨sler, C. Silvestru, G. Espinosa-Pe´rez, I. Haiduc and R. Cea-
Olivares, Inorg. Chim. Acta, 1996, 241, 47.
48 (a) B. T. Allen and D. W. Nebert, J. Chem. Phys., 1964, 41,
1983; (b) F.-D. Tsay, S. L. Manatt and S. I. Chan, Chem. Phys. Lett.,
1972, 17, 223; (c) F.-D. Tsay and L. Helmholtz, J. Chem. Phys., 1969,
50, 2642; (d) J. S. Van Wieringen, Discuss. Faraday Soc., 1955, 19, 118.
49 (a) L. Benisvy, E. Bill, A. J. Blake, D. Collison, E. S. Davies, C. D.
Garner, C. I. Guindy, E. J. L. McInnes, G. McArdle, J. McMaster, C.
Wilson and J. Wolowska, Dalton Trans., 2004, 3647; (b) R. Stahl-Brada
and W. Low, Phys. Rev., 1959, 113, 775; (c) K. Fukui, N. Kojima, H.
Ohya-Nishiguchi and N. Hirota, Inorg. Chem., 1992, 31, 1338.
50 R. L. Carlin, Magnetochemistry, Springer Verlag, 1986.
51 K. Fukui, H. Ohya-Nishigughi and N. Hirota, Bull. Chem. Soc. Jpn.,
1991, 64, 1205.
21 M. Dolg, U. Wedig, H. Stoll and H. Preuss, J. Chem. Phys., 1987, 86,
866.
22 (a) A. D. McLean and G. S. Chandler, J. Chem. Phys., 1980, 72, 5639;
(b) R. Krishnan, J. S. Binkley, R. Seeyer and J. A. Pople, J. Chem. Phys.,
1980, 72, 650.
23 (a) S. I. Gorelsky, AOMix program, http://www.sg-chem.net/; (b) S. I.
Gorelsky and A. B. P. Lever, J. Organomet. Chem., 2001, 635,
187.
24 (a) A. E. Reed, L. A. Curtiss and F. Weinhold, Chem. Rev., 1988, 899;
(b) F. Weinhold, Natural Bond Orbital Methods in Encyclopedia of
Computational Chemistry, ed. P. v. R. Schleyer, N. L. Allinger, T. Clark,
J. Gasteiger, P. A. Kollman, H. F. Schaefer, III and P. R. Schreiner, John
Wiley & Sons, Chichester, UK, 1998, vol. 3, pp. 1792–1811.
25 C. H. Chang, A. J. Boone, R. J. Bartlett and N. G. J. Richards, Inorg.
Chem., 2004, 43, 458.
26 (a) NBO 5.0, E. D. Glendening, J. K. Badenhoop, A. E. Reed,
J. E. Carpenter, J. A. Bohmann, C. M. Morales and F. Weinhold,
Theoretical Chemistry Institute, University of Wisconsin, Madison,
WI,2001; (b) M. Wendt and F. Weinhold, NBOView 1.0, Theoretical
Chemistry Institute, University of Wisconsin, Madison, WI, 2001.
27 (a) MOLEKEL 4.3, P. Flu¨kiger, H. P. Lu¨thi, S. Portmann and J. Weber,
Swiss Center for Scientific Computing, Manno (Switzerland), 2000–
2002; (b) S. Portmann and H. P. Lu¨thi, MOLEKEL: An Interactive
Molecular Graphics Tool, Chimia, 2000, 54, 766.
52 F. E. Mabbs and D. J. Machin, Magnetism and Transition Metal
Complexes, Chapman and Hall Ltd., London, 1973.
28 F. T. Wang, J. Najdzionek, K. L. Neneker, H. Wasserman and D. M.
Braitsch, Synth. React. Inorg. Met.-Org. Chem., 1978, 8, 119.
29 D. Cupertino, R. Keyte, A. M. Z. Slawin, D. J. Williams and J. D.
Woollins, Inorg. Chem., 1996, 35, 2695.
30 D. J. Birdsall, A. M. Z. Slawin and J. D. Woollins, Polyhedron, 2001,
20, 125.
53 F. A. Cotton, D. M. L. Goodgame and M. Goodgame, J. Am. Chem.
Soc., 1961, 83, 4690.
54 L. Banci, C. Benelli, D. Gatteschi and F. Mani, Inorg. Chem., 1982, 21,
1133.
55 K. Kojima, J. Matsuda, N. Kojima, T. Ban and I. Tsujikawa, Bull.
Chem. Soc. Jpn., 1987, 60, 3213.
31 P. Bhattacharyya, A. M. Z. Slawin, D. J. Williams and J. D. Woollins,
J. Chem. Soc., Dalton Trans., 1995, 1607.
56 W. DeW. Horrocs, Jr. and D. A. Barlone, J. Am. Chem. Soc., 1976, 98,
6512.
57 S. DeBeer George, M. Metz, R. K. Szilagyi, H. Wang, S. P. Cramer,
Y. Lu, W. B. Tolman, B. Hedman, K. O. Hodgson and E. I. Solomon,
J. Am. Chem. Soc., 2001, 123, 5757.
58 S. I. Gorelsky, L. Basumallick, J. Vura-Weis, R. Sarangi, K. O.
Hodgson, B. Hedman, K. Fujisawa and E. I. Solomon, Inorg. Chem.,
2005, 44, 4947.
59 (a) K. R. Williams, B. Hedman, K. O. Hodgson and E. I. Solomon,
Inorg. Chim. Acta, 1997, 263, 315; (b) K. Rose, S. E. Shadle, M. K.
Eidsness, D. M. Kurtz, Jr., R. A. Scott, B. Hedman, K. O. Hodgson
and E. I. Solomon, J. Am. Chem. Soc., 1998, 120, 10743; (c) T. Glaser, B.
Hedman, K. O. Hodgson and E. I. Solomon, Acc. Chem. Res., 2000, 33,
859; (d) R. K. Szilagyi, B. S. Lim, T. Glaser, R. H. Holm, B. Hedman,
K. O. Hodgson and E. I. Solomon, J. Am. Chem. Soc., 2003, 125, 9158.
60 A. D. Fielder, H. L. Halfen, J. A. Halfen and T. C. Brunold, J. Am.
Chem. Soc., 2005, 127, 1675.
61 A. D. Fielder, P. A. Bryngelson, M. J. Maroney and T. C. Brunold,
J. Am. Chem. Soc., 2005, 127, 5449.
62 D. G. Kehres and M. E. Maguire, FEMS Microbiol. Rev., 2003, 27,
263.
32 L. M. Gilby and B. Piggott, Polyhedron, 1999, 18, 1077.
33 O. Siiman and H. B. Gray, Inorg. Chem., 1974, 13, 1185.
34 A. Davison and E. S. Switkes, Inorg. Chem., 1971, 10, 837.
35 M. C. Aragoni, M. Arca, A. Garau, F. Isaia, V. Lippolis, G. L. Abbati
and C. Fabretti, Z. Anorg. Allg. Chem., 2000, 626, 1454.
36 See reviews in Delocalization-pi and sigma, Chem. Rev., 2005, 105,
3433.
37 C. Silvestru, R. Ro¨sler, I. Haiduc, R. Cea-Olivares and G. Espinosa-
Pe´rez, Inorg. Chem., 1995, 34, 3352.
38 (a) S. Husebye and K. Maartmann-Moe, Acta Chem. Scand. Ser. A,
1983, 37, 439; (b) J. F. Nixon and P. B. Hitchcock, Inorg. Chim. Acta,
1985, 96, 77.
39 A. M. Z. Slawin, J. Ward, D. J. Williams and J. D. Woollins, J. Chem.
Soc., Chem. Commun., 1994, 421.
40 The NBO program authors,26 define a LP as a doubly occupied
NBO that is not involved in bonding, except from the case of (a
maximum possible) spin orbital occupancy of 1.0, in which case the
NBO algorithm analyzes a and b spin density separately.
41 D. C. Cupertino, R. W. Keyte, A. M. Z. Slawin and J. D. Woollins,
Polyhedron, 1999, 18, 707.
63 A. Glasfeld, E. Guedon, J. D. Helmann and R. G. Brennan, Nat. Struct.
42 C. Silvestru, R. Ro¨sler, J. E. Drake, J. Yang, G. Espinoza-Pe´rez and I.
Haiduc, J. Chem. Soc., Dalton Trans., 1998, 73.
Biol., 2003, 10, 652.
64 O. Siiman, M. Wrighton and H. B. Gray, J. Coord. Chem., 1972, 2, 159.
This journal is
The Royal Society of Chemistry 2006
Dalton Trans., 2006, 2301–2315 | 2315
©