Soluble monometallic salen complexes derived from O-functionalised diamines as metalloligands for the synthesis of heterobimetallic complexes

Article abstract of DOI:10.1039/b920706f

O-Functionalised salen ligands were employed as bridging ligands in the synthesis of homo- and heterometallic salen complexes with early and late transition metals (H₂salen: N,N′-bis(salicylidene) ethylenediamine, systematic name: 2,2′-{ethane-1,2- diylbis(nitrilomethylidine)}diphenol). A new type of O-functionalised salen ligand was synthesised, which contains alkyl groups to enhance the solubility in organic solvents as well as carboxyl groups to allow introduction of an early transition metal. Two new salen ligands derived from O-functionalised diamines were synthesised from 3,5-di-tert-butylsalicylaldehyde (bsal) and 3,4-diaminobenzoic acid (4cpn) or (R)-2,3-diaminopropionic acid (cen). By using the aryl diamines, a conjugated backbone is obtained, and the alkyl diamines can be used to introduce a chiral centre. The salen ligand derived from 3,4-diaminobenzoic acid was accessible only via a zinc(ii)-mediated template reaction. Monometallic salen complexes could be obtained by template synthesis with nickel(ii) and copper(ii). The analogous chromium(iii), manganese(iii) and molybdenum(iv) salen complexes were synthesised directly from the salen ligands. The crystal structures of the molybdenum(iv) salen complex and a decomposition product thereof gave insight into the stability of this compound. Starting from (R)-2,3-diaminopropionic acid the corresponding nickel(ii), chromium(iii), manganese(iii) and molybdenum(iv) salen complexes were obtained. Reactions of the conjugated nickel(ii) salen complex with metallocene derivatives resulted in the formation of soluble di- and trinuclear heterobimetallic complexes, depending on the stoichiometry used. The compounds were characterised by NMR, IR and EPR spectroscopy, mass spectrometry and, for selected complexes, by X-ray crystallography. For selected mono- and bimetallic salen complexes the catalytic activity in the epoxidation of styrene was tested under different reaction conditions and with different oxidising agents. The highest values (up to 24%) for the conversion of styrene to styrene oxide were obtained with manganese(iii) salen complexes.

Full text of DOI:10.1039/b920706f

View Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
Soluble monometallic salen complexes derived from O-functionalised diamines
as metalloligands for the synthesis of heterobimetallic complexes†
Michael Schley, Sebastian Fritzsche, Peter Lo¨nnecke and Evamarie Hey-Hawkins*
Received 5th October 2009, Accepted 9th February 2010
First published as an Advance Article on the web 12th March 2010
DOI: 10.1039/b920706f
O-Functionalised salen ligands were employed as bridging ligands in the synthesis of homo- and
heterometallic salen complexes with early and late transition metals (H2salen: N,N¢-bis(salicylidene)-
ethylenediamine, systematic name: 2,2¢-{ethane-1,2-diylbis(nitrilomethylidine)}diphenol). A new
type of O-functionalised salen ligand was synthesised, which contains alkyl groups to enhance the
solubility in organic solvents as well as carboxyl groups to allow introduction of an early transition
metal. Two new salen ligands derived from O-functionalised diamines were synthesised from
3,5-di-tert-butylsalicylaldehyde (bsal) and 3,4-diaminobenzoic acid (4cpn) or (R)-2,3-diaminopropionic
acid (cen). By using the aryl diamines, a conjugated backbone is obtained, and the alkyl diamines can
be used to introduce a chiral centre. The salen ligand derived from 3,4-diaminobenzoic acid was
accessible only via a zinc(II)-mediated template reaction. Monometallic salen complexes could be
obtained by template synthesis with nickel(II) and copper(II). The analogous chromium(III),
manganese(III) and molybdenum(IV) salen complexes were synthesised directly from the salen ligands.
The crystal structures of the molybdenum(IV) salen complex and a decomposition product thereof gave
insight into the stability of this compound. Starting from (R)-2,3-diaminopropionic acid the
corresponding nickel(II), chromium(III), manganese(III) and molybdenum(IV) salen complexes were
obtained. Reactions of the conjugated nickel(II) salen complex with metallocene derivatives resulted in
the formation of soluble di- and trinuclear heterobimetallic complexes, depending on the stoichiometry
used. The compounds were characterised by NMR, IR and EPR spectroscopy, mass spectrometry and,
for selected complexes, by X-ray crystallography. For selected mono- and bimetallic salen complexes the
catalytic activity in the epoxidation of styrene was tested under different reaction conditions and with
different oxidising agents. The highest values (up to 24%) for the conversion of styrene to styrene oxide
were obtained with manganese(III) salen complexes.
plex in the catalytic epoxidation of chromene.6 This is in contrast
to the large number of existing studies on alkoxy-substituted salen
complexes and early/late heterobimetallic compounds thereof,7
with special focus on their magnetic behaviour.8
We recently reported the synthesis of soluble monometallic
salen complexes derived from O-functionalised salicylaldehydes9
and hydroxy-functionalised diamines10 as metalloligands for the
synthesis of heterobimetallic complexes and immobilisation of
some of these complexes on organo-modified mesoporous silica
gel as well as their catalytic activity in the epoxidation of olefins.11
Now we report the synthesis and characterisation of two
new salen ligands derived from carboxy-functionalised diamines,
Na(H2bsalcen) and H3bsal4cpn,12 and their mono- and hetero-
bimetallic complexes.
1
Introduction
Monometallic salen complexes have been studied extensively after
Kochi et al.1 reported their highly chemoselective catalytic activity
in epoxidation reactions. Further extensive studies were carried out
by Jacobsen et al.2 and Katsuki et al.,3 who reported independently
the excellent performance of chiral manganese(III) complexes with
a wide variety of salen ligands having various chiralities and bulky
alkyl or aryl groups in the enantioselective catalytic oxidation of
unfunctionalised olefins.4
Despite extensive work on tuning of asymmetric induction
of salen ligands, hardly any attention has been paid to O-
functionalised salen complexes, e.g., with hydroxyl or carboxyl
substituents, or their activity in catalytic epoxidation.5 There is
only one report on employing a carboxy-functionalised salen com-
2
Results and discussion
Institut fu¨r Anorganische Chemie der Universita¨t Leipzig, Johannisallee 29,
D-04103, Leipzig, Germany. E-mail: hey@rz.uni-leipzig.de; Fax: (+49)-
341-9739319; Tel: (+49)-341-9736151
† Electronic supplementary information (ESI) available: Cyclic voltammo-
gram of the complexes [Na{Cu(bsalcen)}] (1b) and [Tp*2La{Cu(bsalcen)}]
(4b) in dichloromethane; and cyclic voltammograms of the com-
plexes [Ni(Hbsal4cpn)] (2a) and [Tp*2La{Ni(bsal4cpn)}] (5a) in
dichloromethane and thf. CCDC reference numbers 736166–736176. For
ESI and crystallographic data in CIF or other electronic format see DOI:
10.1039/b920706f
2.1 Salen ligands derived from carboxy-functionalised alkyl and
aryl diamines
For comparison of the properties of salen complexes having a
flexible alkyl diimine backbone with those having a conjugated
aryl diimine backbone, different ligands were synthesised. For all
ligands, 3,5-di-tert-butyl-salicylaldehyde (bsal) was used, which
4090 | Dalton Trans., 2010, 39, 4090–4106
This journal is
The Royal Society of Chemistry 2010
©

View Online
results in enhanced solubility of the ligands in organic solvents
due to its tert-butyl groups.
The synthesis of carboxy-functionalised salen ligands derived
from alkyl diamines (see Scheme 1) was carried out as described
for related ligands.13
The salen ligand H3bsal4cpn could finally be obtained in a
template synthesis using the weakly coordinated zinc(II) cation,
which prefers tetrahedral coordination and thus forms a kineti-
cally labile Zn salen complex from which the metal cation is readily
eliminated, e.g., in methanol. It is also known that zinc(II) com-
pounds can activate starting materials by coordination.17 Using
this approach and heating to reflux a mixture of two equivalents of
salicylaldehyde, one equivalent of 3,4-diaminobenzoic acid (4cpn)
and two equivalents [ZnCl2(thf)2] in thf for 1 h resulted in a dark
green solution. Removal of thf and addition of methanol gave
the salen ligand H3bsal4cpn as a nearly insoluble yellow solid
(Scheme 3). In the FAB mass spectrum the molecular ion peak is
Scheme 1 Synthesis of the salen ligand Na(H2bsalcen).
observed atm/z = 585.4, and in the IR spectrum nO–H (3425cm-1)
-1
=
and nC O (1690 cm ) bands indicate the presence of a COOH
This procedure could be easily employed for the synthesis
of Na(H2bsalcen) by addition of a stoichiometric amount of
NaOH to enantiopure (R)-2,3-diaminopropionic acid mono-
hydrochloride (cen). However, racemisation took place at some
stage (as shown by X-ray crystallography on the hetero-
bimetallic complexes [Tp*2La{Ni(bsalcen)}]·n-hexane (4a) and
[Tp*2La{Cu(bsalcen)}]·n-hexane (4b); see Section 4). Racemisa-
tion of a-amino acids on formation of Schiff bases was already
described by Grigg et al.14
group. H3bsal4cpn is soluble in thf and dmso, but only slightly
soluble in methanol and ethanol.
A
salen ligand derived from the carboxy-functionalised
Scheme 3 Synthesis of the salen ligand H3bsal4cpn with [ZnCl2(thf)2].
aryl diamine 3,4-diaminobenzoic acid was mentioned in the
literature,15 and only the in situ synthesis of N,N¢-bis(salicylidene)-
3,4-diaminobenzoic acid from salicylaldehyde and 3,4-diamino-
benzoic acid, which was used without further characterisation
in the synthesis of the corresponding nickel(II) complex (charac-
terised by elemental analysis only), was reported.16
Attempts to obtain the ligand H3bsal4cpn with catalytic
amounts (10 mol%) of [ZnCl2(thf)2] or ZnCl2·4H2O were not
successful.
2.2 Monometallic (d-block) and heterometallic (d-block/s-block)
salen complexes derived from carboxy-functionalised alkyl- and
aryl diamines
Therefore, we tried to synthesise the carboxy-functionalised
ligand H3bsal4cpn employing the general procedure, i.e., treating
two equivalents of salicylaldehyde with one equivalent of diamine
in methanol. The reaction mixture was heated to reflux for
2 h and then stirred for 18 h at room temperature. However,
the yield of the isolated yellow solid was only 23%. The 1H
NMR spectrum of the solid in methanol showed the formation
The complexes [M{Na(bsalcen)}] (M = Ni, 1a; Cu, 1b) were
prepared from the salen ligand Na(H2bsalcen) and the appropriate
transition metal chloride in methanol. NaOH or NaOCH3 was
used as deprotonating agent. These complexes were characterised
by 1H NMR (1a), 13C NMR (1a), EPR (1b) and IR spectroscopy,
elemental analysis and mass spectrometry.
The synthesis of the corresponding Mn, Cr and Mo complexes
[M{Na(bsalcen)}] (M = MnCl, 1c; CrCl, 1d; MoCl2, 1e) was
performed in thf or toluene using Na(H2bsalcen) and the water-
free metal chlorides in the presence of two equivalents of Et3N
(Scheme 4).
1
of the “half-salen” ligand, whereas the H NMR spectrum of
the supernatant solution showed only unchanged salicylaldehyde
(see Scheme 2). An explanation for the incomplete reaction is
formation of zwitterions in which only the unprotonated amino
group will react with the aldehyde to form the half-salen ligand.
Scheme 4 Synthesis of monometallic complexes [M{Na(bsalcen)}] (M =
Ni, L = H2O, m = 2, n = 6, base = 2 NaOH, 1a; M = Cu, L = H2O, m = 2,
n = 0, base = 2 NaOCH3, 1b; M = MnCl, L = thf, m = 1, n = 2, base = 2
Et3N, air, 1c; M = CrCl, L = thf, m = 2, n = 3, base = 2 Et3N, 1d and M =
MoCl2, L = CH3CN, m = 2, n = 2, base = 2 Et3N, 1e).
Scheme
aminobenzoic acid.
2
Reaction of 3,5-di-tert-butylsalicylaldehyde with 3,4-di-
Attempts to prepare the salen ligand H3bsal4cpn in boiling
toluene with increased reaction time, use of dry thf and five
equivalents of Na2SO4 as drying agent, or addition of two
equivalents of NaOH in methanol, only resulted in formation of
the half-salen ligand and unreacted salicylaldehyde (Scheme 2).
The solubility of the complexes in toluene facilitates separation
from insoluble ammonium chloride. However, complete separa-
tion was difficult, as was apparent from the elemental analysis and
ESI MS of the chromium(III) and molybdenum(IV) complexes, in
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 4090–4106 | 4091
©

View Online
Table 1 Selected IR vibrations (vib.) of Na(H2bsalcen) and complexes derived therefrom
-1
vib. M-Cl/cm-1
further vib. in CsI/cm-1
=
Compound
vib. C N/cm
Na(H2bsalcen)
1629 (s)
1611 (s)
1610 (s)
1618 (s)
—
391 (w)
397 (m)
392 (w), 319 (w)
646 (w), 539 (m), 488 (w), 454 (w), 278 (w)
644 (w), 563 (m), 491 (w), 473 (w), 270 (w)
639 (w), 547 (m), 510 (w), 494 (w), 279 (w)
644 (w), 560 (m), 483 (w), 465 (w), 278 (w)
[MnCl{Na(bsalcen)}] (1c)
[CrCl{Na(bsalcen)}] (1d)
[MoCl2{Na(bsalcen)}] (1e)
which Et3N adducts were detected. The complexes, 1c, 1d, and
1e were characterised by IR spectroscopy, mass spectrometry and
elemental analysis, and their magnetic moments were determined.
The IR spectra of Na(H2bsalcen) and salen complexes derived
=
therefrom have characteristic C N vibrations and, where appro-
priate, M–Cl vibrations (Table 1).
-1
=
In the complexes the C N vibration is shifted by ca. 20 cm
Scheme 6 Synthesis of monometallic complexes [M(Xbsal4cpn)] (M =
MnCl, X = [NHEt3], L = thf, m = 1, n = 2, oxidation by air, 2c; M =
CrCl, X = H, L = thf, m = 2, n = 3, 2d and M = MoCl2, X = H, L =
CH3CN, m = 2, n = 2, 2f)
to lower wavenumbers compared to Na(H2bsalcen), which is
=
attributable to the lower C N bond order due to coordination
to the metal atom. The M–Cl vibrations are expected between 300
and 450 cm-1.18 Only one M–Cl valence vibration is expected and
observed for the square-pyramidal chromium(III) 1d (397 cm-1)
and manganese(III) complexes 1c (391 cm-1), whereas the slightly
distorted octahedral molybdenum(IV) complex 1e with two trans
chloro ligands exhibits two absorptions, for the symmetric and,
at higher wavenumbers, the antisymmetric vibration, at 392 and
319 cm-1 which are in the range observed for [MoCl2(salen)]
(309 cm-1),19 and dimeric MoCl5 (418 to 345 cm-1).20 The Cr–Cl
vibration for chromium(III) complex 1d is in the range observed
for CrOCl3 (404 cm-1).21
Presumably, this complex forms an oligomeric structure in the
solid state.
All monometallic complexes were characterised by IR spec-
troscopy and mass spectrometry. The nickel complex was addi-
tionally characterised by 1H NMR and 13C NMR spectroscopy.
2.3 Heterobimetallic (d-block/d-block, d-block/f-block) salen
complexes derived from carboxy-functionalised alkyl or aryl
diamines
The synthesis of metal salen complexes with the salen lig-
and containing a carboxy-functionalised aryl diimine backbone,
H3bsal4cpn, was achieved via two routes. Firstly, monometallic
complexes [Ni(Hbsal4cpn)] (2a) and [Cu(Hbsal4cpn)] (2b) were
obtained by template synthesis starting from 3,4-diaminobenzoic
acid, 3,5-di-tert-butylsalicylaldehyde and hydrated nickel(II) chlo-
ride or copper(II) chloride (Scheme 5).
Starting from the monometallic complexes or the heterometallic
sodium salts described in Section 2.2., heterobimetallic complexes
with early transition metals and lanthanoids were synthesised. It
was necessary to find co-ligands which were able to stabilise the
coordination sphere of the second metal and which allowed the
preparation of discrete dinuclear or trinuclear complexes. In view
of the catalytic oxidation reactions the co-ligands should also be
stable against oxidation. These requirements were achieved by use
of cyclopentadienyl (Cp) and hydridotrispyrazolyl (Tp) ligands
and their derivatives.
2.3.1 Titanium- and zirconium-containing heterobimetallic
salen complexes. Two equivalents of the nickel(II) complex 2a
reacted with one equivalent of the early transition metal met-
allocene derivatives [CpR2M¢(CH3)2] (M¢ = Ti, Zr; CpR = Cp
(C5H5) or Cp◦ (C5EtMe4)) via methane formation to give the
symmetrical heterotrinuclear complexes [Cp2Zr{Ni(bsal4cpn)}2]
Scheme 5 Synthesis of monometallic complexes [NHEt3][M(bsal4cpn)]
(with M = Ni, n = 6, 2a; M = Cu, n = 0, 2b).
(3a) and [Cp◦ Ti{Ni(bsal4cpn)}2] (3b) (Scheme 7).
2
The heterobimetallic complexes were characterised by 1H NMR
spectroscopy, IR spectroscopy and elemental analysis. The 1 : 1
reaction of [CpR2M¢(CH3)2] with the nickel(II) complex 2a led,
however, to an inseparable mixture of trinuclear and binuclear
complexes and unconverted starting material.
Secondly, monometallic complexes [MnCl(Hbsal4cpn)]
(2c), [CrCl(Hbsal4cpn)] (2d), [Fe(bsal4cpn)] (2e) and
[MoCl2(Hbsal4cpn)] (2f) could be obtained starting from
the salen ligand H3bsal4cpn and the appropriate metal chloride
complex (Mn, Cr, Mo) (Scheme 6) or Fe(OAc)2. In order to avoid
formation of NHEt3 salts (as was observed for the Mn complex),
no amine (Fe, Cr) or nBu3N (Mo, the ammonium salt of which is
very soluble in thf, while the Mo complex is not) was employed
for the synthesis of the chromium(III) complex 2d (obtained in
98% yield) and molybdenum(IV) complex 2f (65% yield).
2.3.2 Lanthanoid-containing heterobimetallic salen complexes.
Heterobimetallic Ni/Ln and Cu/Ln complexes could easily be
obtained by salt elimination. Thus, the monometallic nickel(II)
(1a, 2a) and copper(II) complexes (1b, 2b) were treated with
a stoichiometric amount of NaOH followed by reaction with
[LnTp*2(OTf)] to give the heterobimetallic lanthanoid complexes
The iron(III) complex 2e was isolated as a zwitterion (positive
charge at the Fe atom and negative charge at the carboxyl group).
4092 | Dalton Trans., 2010, 39, 4090–4106
This journal is
The Royal Society of Chemistry 2010
©

View Online
=
molecular structure, Section 2.4.4), in which one of the two C N
bonds was reduced to a C–N single bond. The synthesis of this
complex could not be reproduced nor could the reducing agent be
unambiguously determined. Addition of hydride and enolate to
the imine bond of an aluminium(III) salophen complex was also
observed by Portalone, Rissanen, et al.24 We assume that Tp*,
which contains a hydridic B–H group, could be responsible for the
reduction, because reducing agents like NaBH4 are well known for
the reduction of Schiff bases.25
2.3.3. 1H
NMR
investigation
of
the
complexes
[Na{Ni(bsal4cpn)}] (2a¢) and [Tp*2Ln{Ni(bsal4cpn)}] (Ln =
La, 5a; Ln = Nd, 6a). In order to investigate the stability
of the lanthanoid-containing heterobimetallic complexes in
solution, the mono- and heterobimetallic complexes were
Scheme 7 Synthesis of heterobimetallic salen complexes containing early
transition metals [CpR2M¢{Ni(bsal4cpn)}2] (with M¢ = Zr and CpR
=
1
analysed by H NMR spectroscopy in different solvents. First,
C5H5, 3a and with M¢ = Ti and CpR = C5EtMe4, 3b).
the spectra were measured in the less polar solvents C6D6 and
CD2Cl2. Solutions of the nickel complex 2a¢ showed very broad
signals. This broadening is ascribed to the associative dynamic
behaviour of carboxyl groups and was not diminished by cooling
the solutions. Accordingly, sharp signals were observed for the
monometallic complex in [D8]thf. On the other hand, coordination
of the carboxyl group to the lanthanoid cation also eliminates
this associative dynamic behaviour, resulting in sharp proton
4a, 4b with an alkyl backbone as well as the aryl derivatives 5a, 5b
(La) and 6a, 6b (Nd) in addition to NaOTf (Scheme 8).
This procedure could not be applied in the synthesis of
complexes containing Fe and Cr which form stable betaines.
Thus, for the preparation of heterobimetallic complexes with Fe
and Cr the monometallic betaine complexes [M(bsal4cpn)] were
converted to the carboxylic acids with a solution of HCl/diethyl
ether and then treated with [LaTp*2(pz*)]. Formation of the
heterobimetallic complexes is accompanied by the generation of
pz*H, which then coordinates as a neutral axial ligand in the
complex [Tp*2La{CrCl(pz*H)(bsal4cpn)}] (5c), as indicated by
an N–H absorption in the IR spectrum at 3293 cm-1 and the
elemental analysis. On the other hand, the IR spectrum and
elemental analysis of the complex [Tp*2La{FeCl(bsal4cpn)}] (5d)
indicated no coordination of pz*H. Thus, the Fe atom presumably
has a square-pyramidal coordination sphere, as was observed for
similar complexes.22,23
1
resonances. In Table 2 selected H NMR signals of aromatic H
atoms of the nickel complex 2a¢ and the heterometallic La- or
Nd-containing complexes 5a and 6a are summarised.
The differences between analogous signals of the complexes
[Na{Ni(bsal4cpn)}] (2a) and [Tp*2La{Ni(bsal4cpn)}] (5a) are
1
rather small. Due to the paramagnetism of NdIII, the H NMR
signals of [Tp*2Nd{Ni(bsal4cpn)}] (6a) are shifted strongly down-
field. The highest downfield shift is observed for H atoms 10 and
12, which are located in the position ortho to the COO group. This
observation confirms stable coordination of the carboxyl group to
the lanthanoid cation.
In the synthesis of the heterobimetallic Cr/La com-
plex [Tp*2La{CrCl(pz*H)(bsal4cpn)}] (5c) a few crystals of
[Tp*2La(Cr{pz*H}2{b(H)sal4cpn})] (5f) were also obtained (see
2.3.4. Molecular structure of [(Tp*2La{Fe(bsal4cpn)})2(l-O)]
(5e). It is well known that the formation of dimeric oxo-bridged
Scheme 8 Synthesis of lanthanoid-containing heterobimetallic salen complexes 4a, 4b, 5a, 5b, 5c and 5d.
The Royal Society of Chemistry 2010 Dalton Trans., 2010, 39, 4090–4106 | 4093
This journal is
©

View Online
Table 2 Selected 1H NMR chemical shifts of the complexes
[X{Ni(bsal4cpn)}] (X = Na, 2a¢; Tp*2La, 5a; Tp*2Nd, 6a, in [D8]thf)a
range of comparable complexes.26 The Fe ◊ ◊ ◊ La distances within
the two FeLa complex fragments average 959 pm. The Fe1 ◊ ◊ ◊ Fe2
distance is 345.5(4) pm.
2.4 Molecular structures of mono- and heterobimetallic salen
complexes derived from carboxy-functionalised aryl and alkyl
diamines
2.4.1. Molecular structure of the monometallic com-
plexes [Ni(Hbsal4cpn)] (2a), [CrCl(CH3CN)(Hbsal4cpn)] (2d¢),
[CrCl(pz*H)(Hbsal4cpn)] (2d¢¢) and [MoCl2(Hbsal4cpn)] (2f). In
the solid state, all complexes form dimers by intermolecular hy-
drogen bonding between the carboxyl groups (Fig. 2–4), a feature
which is usually observed for carboxylic acids. The intermolecular
O3 ◊ ◊ ◊ O4 distances are about 260 pm and comparable to those of
similar compounds.27
1
2
3
CH
chem. shift d 2a¢
chem. shift d 5a
chem. shift d 6a
7,7¢
9
10
12
8.69, 8.34
7.87
7.61
8.68, 8.36
7.83
7.64
11.00, 12.66
12.50
20.56
8.63
8.57
21.61
a The sodium salt of 2a (= 2a¢) was employed.
iron complexes is responsible for the continuous deactivation
during the catalytic oxidation.1 The introduction of a bulky
substituent in the diimine backbone was assumed to avoid the
formationof theoxo-bridged dimer and thus maintain the catalytic
activity of the Fe complex 2e. Oyaizu et al. have already shown
that additional phenyl groups in the diimine backbone reduce
the formation of oxo-bridged dimers.23 However, addition of
one equivalent of NaOH to [Fe(bsal4cpn)] (2e) followed by two
equivalents of [LaTp*2(OTf)] gave [(Tp*2La{Fe(bsal4cpn)})2(m-
O)] (5e).
The oxo-bridged dimer crystallises with 3.5 toluene molecules.
The Fe atoms have square-pyramidal coordination spheres and
are located about 55 pm and 53 pm above the N2O2 plane (Fig. 1).
The Fe1–O9–Fe2 angle of 155.9(2)◦ is rather small compared to
the analogous angle of 176.5(2)◦ in the oxo-bridged dimer which
contains a sterically less hindered 1,2-phenylenediimine backbone.
The Fe(1, 2)–O(1, 2, 5, 6) and Fe(1, 2)–O9 bond lengths are in the
Fig. 2 ORTEP of the monomeric unit of [Ni(Hbsal4cpn)] (2a). In the
solid state, 2a forms hydrogen-bonded dimers. H atoms (other than H1O4)
are omitted for clarity. The size of the ellipsoids relates to a probability of
50%.
While chromium(III) and molybdenum(IV) prefer an octahedral
coordination sphere, nickel(II) is coordinated in a square-planar
fashion. Remarkably, the organic ligand is not planar, despite
its conjugated backbone. Salen complexes with a conjugated
ligand cannot compensate ring strain by formation of l or d
conformations of the five-membered rings as in non-conjugated
salen complexes. Comparison of the MN2O2 planes (see
Fig. 2–4) reveals almost no strain in the nickel and chromium
complex and high strain in the Mo complex. The dihedral angle
between the salicylidene rings and the MoN2O2 plane in the Mo
complex is approximately 30◦ and thus about six times larger than
in the nickel and chromium complexes (< 5◦) and more than twice
as large as in a non-conjugated hexacoordinate manganese(III)
salen complex.28 The distortion in the molybdenum complex 2f is
also obvious from the small O–Mo–N bond angles of 159.0(2) and
159.1(2)◦ as compared to 172 to 176◦ in the nickel and chromium
complexes. This clearly indicates that the distortion is not due to
steric interaction of adjacent tert-butyl groups but due to the size
of the central atom. The size of the molybdenum atom causes a
drastic expansion of the N2O2 coordination sphere, which is also
Fig. 1 ORTEP of the molecular structure of [(Tp*2La{Fe(bsal4cpn)})2-
(m-O)] (5e). H atoms are omitted for clarity. The size of the ellipsoids relates
to a probability of 50% (metal atoms) or 5% (other atoms).
4094 | Dalton Trans., 2010, 39, 4090–4106
This journal is
The Royal Society of Chemistry 2010
©

View Online
Fig. 4 ORTEP of [MoCl2(Hbsal4cpn)] (2f). The molecules form dimers
via intermolecular COOH hydrogen bonds (acetonitrile molecules are
omitted for clarity; ellipsoid probability: 50%).
Fig.
3
ORTEP of the molecular structures of [CrCl(CH3CN)-
(Hbsal4cpn)]·6CH3CN (2d¢) and [CrCl(pz*H)(Hbsal4cpn)]·1,4-dioxane
(2d¢¢). The 1,4-dioxane molecule, the CH3CN molecules and the H atoms
(excepted O3H and N4H) are omitted for clarity. The size of the ellipsoids
relates to a probability of 50%.
reflected in the O1 ◊ ◊ ◊ O2 distance of about 320 pm, as opposed to
only about 250 and 280 pm in the nickel and chromium complexes,
respectively.
Fig. 5 ORTEP of the dimeric structure of the decomposition product
2f¢ of [MoCl2(Hbsal4cpn)] (2f). The hydrogen atoms and the C6D6 and
[nBu3NH]Cl molecules are omitted for clarity. The size of the ellipsoids
relates to a probability of 40%.
2.4.2. Molecular structure of a decomposition product 2f¢ of
[MoCl2(Hbsal4cpn)]. During the NMR spectroscopic measure-
ments on the diamagnetic molybdenum(IV) salen complex 2f in
moist C6D6 only broad signals were observed and, furthermore,
red needles formed in the NMR tube, which were shown to be
a decomposition product by X-ray crystallography. The asym-
metric unit contains two independent molecules which form a
dimer via (presumably) oxo bridges, four C6D6 molecules and one
[nBu3NH]Cl (Fig. 5). Although refinement was only possible to
an R1 value of 10.1%, a closer look at the obtained product is
valuable, as it was formed in an unusual manner and offers new
insights into the stability of such salen complexes.
The molybdenum atom is coordinated in a distorted octahedral
fashion by one chloro ligand, one oxygen and one nitrogen atom
of a rearranged salen ligand and three oxygen atoms, two of which
bridge two molybdenum centres. The intramolecular Mo–O bond
lengths of the oxygen atoms O7, O8, O9 and O10 are 164.7(8) and
170.3(8) pm which is typical for terminal Mo–O double bonds.29
The Mo–O bond lengths of the bridging oxygen atoms Mo1–O8
and Mo2–O9 are only slightly larger than Mo1–O7 and Mo2–O10,
by 3.5 or 5.6 pm, whereas the intermolecular Mo–O distances are
much longer (253.8 and 255.3 pm). Molybdenum was, therefore,
apparently oxidised from the original molybdenum(IV) salen
complex 2f to the formal oxidation state +6.
Besides the change in the oxidation state of molybdenum, the
second benzylidene ring is missing and a new heterocycle was
formed. Apparently, the coordinated salen ligand was hydrolysed,
3,5-di-tert-butylsalicylaldehyde was eliminated and rearrange-
ment of the ligand resulted in the formation of an imidazole ring,
according to the proposed mechanism shown in Scheme 9.
=
The assignment of the position of the C N double bond in
the imidazole ring is not directly possible, as the distances C37–
N4 (135(1) pm) and C37–N3 (137(1) pm) are nearly identical,
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 4090–4106 | 4095
©

View Online
Scheme 9 Proposed mechanism for the formation of the decomposition product 2f¢ of [MoCl2(Hbsal4cpn)] (2f).
but it can be derived from a similar molecular structure from
Behrens et al.30 for formation of an imidazole ring starting
from Schiff bases. Behrens et al. and Do¨ring et al.31 describe
oxidative formation of an imidazole ring using catalytic amounts
of copper(II).
Two further arguments support the proposed mechanism:
1) During the reaction of [MoCl2(Hbsal4cpn)] (2f) with one
equivalent of water in toluene in air a red solid was formed. Efforts
to crystallise the solid by different crystallisation techniques
resulted, however, only in oils.
2) In the 1H NMR spectra of the supernatant solution of
the obtained decomposition product as well as in the solution
obtained in 1), the proposed hydrolysis product 3,5-di-tert-
butylsalicylaldehyde could be identified.
Other possible “destruction” mechanisms could be anticipated,
whereby both imine groups are cleaved and one aldehyde is
expelled from the coordination sphere of the Mo complex. Re-
organisation of coordinated aldehyde and diamine on the Mo com-
plex could also lead to the imidazole ring and then to the observed
complex. However, this should result in different isomers, whereby
after imidazole formation the carboxyl group is either in para or
in meta position to the N atom coordinating at Mo, which was not
observed in the crystal structure of 2f¢. Further experiments using
different ratios of aldehyde and diamine (with or without oxygen
atmosphere) did not produce the decomposition product.
The formation of the decomposition product 2f¢ illustrates
the reactivity of the hydrogen atom at the imine carbon atom.
Attempts to use 2-hydroxy-4-methoxybenzophenone instead of
salicylaldehyde in the synthesis of the salen ligand (to replace the
hydrogen atom at the imine carbon atom with an aryl substituent)
resulted, however, in inseparable product mixtures which indicates
that ketones have a distinctly lower reactivity.
2.4.3. Molecular structures of the heterometallic com-
plexes [Tp*2La{Ni(bsal4cpn)}] (5a), [Tp*2La{Cu(bsal4cpn)}] (5b),
[Tp*2La{Ni(bsalcen)}] (4a), and [Tp*2La{Cu(bsalcen)}] (4b).
The molecular structures of the heterobimetallic complexes 5a
and 5b are shown in Fig. 6.
The Ni and Cu atoms are coordinated in a square-planar fashion
by the N2O2 coordination sphere of the bsal4cpn ligand. The
M–N and M–O (M = Ni, Cu) bond lengths are in the expected
Fig. 6 ORTEP of the molecular structures of [Tp*2La{Ni(bsal4cpn)}]·
1,4-dioxane·0.5n-hexane (5a) and [Tp*2La{Cu(bsal4cpn)}]·n-hexane (5b).
The n-hexane molecules and the H atoms are omitted for clarity. The size
of the ellipsoids relates to a probability of 50%.
range for salen complexes.32 The Ni–N and Ni–O bond lengths
are nearly identical to those in the monometallic Ni complex 2a.
The distances between the two transition metal atoms within the
4096 | Dalton Trans., 2010, 39, 4090–4106
This journal is
The Royal Society of Chemistry 2010
©

View Online
heterobimetallic complexes are about 933 pm (d(Ni ◊ ◊ ◊ La)) and
937 pm (d(Cu ◊ ◊ ◊ La)). The La–O3 and La–O4 bond lengths of
the complexes differ between 251.8(2) and 248.7(3) pm (5b) and
248.4(7) and 255.7(7) pm (5a). This is due to the steric interaction
between the bulky Tp* ligands and the metallosalen ligand.
The complex [LaTp*2(O2NO)],33 for instance, has two nearly
identical La–O bond lengths of 258.4(1) pm and 258.6(1) pm.
The La–N bond lengths for the 5a and 5b are in the range of
258.5(9) to 277.5(8) and 258.9(3) to 275.6(4) pm, respectively,
and are thus slightly longer than the La–N bonds of the starting
material [LaTp*2(OTf)(CH3CN)] (254.0(5) pm to 270.8(5) pm).33
The reason is the stronger electron-donating effect of the carboxyl
group in the heterobimetallic complexes.
The M–N and M–O (M = Ni, Cu) bond lengths are in the
expected range for salen complexes.32 The difference between
the La–O3 (4b: 263.7(2) pm, 4a: 263.7(3) pm) and La–O4 (4b:
253.4(2) pm, 4a: 250.8(3) pm) bond lengths is larger than in the
heterobimetallic complexes 5a and 5b. This can be explained by the
shorter metal-metal distances of about 593.1(8) pm (d(Ni ◊ ◊ ◊ La))
and 607.4(7) pm (d(Cu ◊ ◊ ◊ La)) and the resulting increased steric
repulsion between the bulky Tp* ligands and the metallosalen
ligand. For steric reasons, the COO ligand would be expected to
occupy an equatorial position, but the COO groups are located
in the sterically unfavourable axial position in both complexes.
Because of this conformation the distance between the atoms
C35H and O1 is rather short, e.g., C35 ◊ ◊ ◊ O1 325.7(4) pm (4b) and
352.48(7) pm (4a), respectively, and the C35–H bond is directed
towards O1, indicating a weak hydrogen bond.34
The complexes 4a and 4b crystallise as racemic mixtures of the
(S, d) and (R, l) enantiomers with one n-hexane molecule in the
asymmetric unit. Only the (S, d) enantiomer of nickel(II) complex
4a and the (R, l) enantiomer of copper(II) complex 4b are depicted
in Fig. 7.
2.4.4. Molecular
structure
of
[Tp*2La(Cr{pz*H}2-
atom in the
{b(H)sal4cpn})] (5f). The
chromium
heterobimetallic complex 5f is octahedrally coordinated
(Fig. 8). Two pz*H molecules occupy the axial position and form
Cr–N(3,5) bonds with an average bond length of 211.4 pm.
This bond length is comparable to that of the monometallic
chromium(III) complex 2d¢¢ (212.7(2) pm). The 3,5-dimethyl-
pyrazole rings are fixed by N6H ◊ ◊ ◊ O2 and N4H ◊ ◊ ◊ O4 hydrogen
bonds, similar to the monometallic complex. The N6 ◊ ◊ ◊ O2 and
N4 ◊ ◊ ◊ O4 distances are 259(1) pm and 264(1) pm, respectively.
Fig. 8 ORTEP of the molecular structure of [Tp*2La(Cr{b(H)sal4cpn})]
(5f). The n-hexane molecules, the H atoms and the t-butyl groups are
omitted for clarity. The size of the ellipsoids relates to a probability of
50%.
=
One of the C N imine bonds was reduced to a C–N single bond,
as revealed by the N2–C30–C16 bond angle of 112.1(7)◦ and the
C30–N2 bond length of 145.7(7) pm. Accordingly, the Cr–N bond
lengths in the equatorial plane differ by about 10 pm (d(Cr–N2) =
192.9(4) pm and d(Cr–N1) = 202.2(4) pm).
2.5 Cyclovoltammetric experiments
Fig. 7 ORTEP of the molecular structures of [Tp*2La{Ni(bsalcen)}]·
n-hexane (4a) and [Tp*2La{Cu(bsalcen)}]·n-hexane (4b). The n-hexane
molecules and the H atoms are omitted for clarity. The size of the ellipsoids
relates to a probability of 50%.
In accordance with our previous work9 we investigated the
electrochemical properties of the mono- and heterobimetallic salen
complexes. Besides many cyclovoltammetric measurements of
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 4090–4106 | 4097
©

View Online
heterobimetallic complexes,35 wherein the two metals are bridged
by donor atoms, there exist only a few examples of cyclovoltam-
metric investigations of heterobimetallic complexes which contain
two transition metals located at least two bonds apart.36,37 Weaker
interactions between the transition metal atoms are expected in the
latter, which represent the type of complexes in our studies, but the
influence of the early transition metal on the redox potential of the
late transition metal coordinated by the N2O2 coordination sphere
is an important aspect with regard to the application of these
complexes, e.g., in catalytic oxidation reactions. For this reason the
behaviour of the complexes was studied, especially under oxidative
conditions.
A representative selection of mono- and heterobimetallic com-
plexes was chosen for the cyclovoltammetric experiments. For the
ligand H3bsal4cpn, the monometallic complex [Ni(Hbsal4cpn)]
(2a) and the heterobimetallic complex [Tp*2La{Ni(bsal4cpn)}]
(5a) were chosen. In case of the ligand Na(H2bsalcen), the
complexes [Na{Cu(bsalcen)}] (1b) and [Tp*2La{Cu(bsalcen)}]
(4b) were studied (see Fig. SI 1, ESI†).
Table 3 Results of catalytic oxidation experiments with monometallic
and heterobimetallic salen complexes
Reaction
Time/h
Conversion
[%]
Yield styrene
oxide [%]
Complex
Oxidant
2c
mcpba/nmo
tbhp
tbhp
tbhp
tbhp
tbhp
tbhp
tbhp
phIO
phIO
phIO
phIO
0.75 (15)
18
21
18
18
21
18
18
18
13.5 (24.4)
11.5
0
11.2
33.6
0.8
13.5 (24.4)
2.0
0
2.4
2.0
0.3
1.0
5.6
9.2
2ab
2bb
1a
1b
5b
4a
4b
2d
5c
2.4
19.6
23.2
6.4
3.7 (11.5)
5.2
18
1 (18)
18
3.3
3.0 (9.0)
5.0
2ea
5d
Conditions: CH2Cl2, 20 ◦C, complex : oxidant : styrene, 0.04 : 2 : 1, ncomplex
=
3 ¥ 10-5 mol. Oxidants: mcpba/nmo = meta-chloroperoxybenzoic acid/N-
methylmorpholine N-oxide, tbhp = tert-butylhydroperoxide, phIO =
iodosobenzene,a ncomplex = 6 ¥ 10-5 mol, b Na-salt employed.
oxTidhaetiocnopapt eEr(pII=) c0o.6m6pVlex(v1s.bSCfirEst). uTnwdoerqguoaessireavnersiribreleverresdibolxe
signals appear at E1/2 = 0.90 V and E1/2 = 1.15 V (vs. SCE),
which were assigned to CuII/CuIII and CuIIIL/CuIIIL+· (L = salen
ligand) by comparison with similar complexes.38 The analogous
quasireversible redox signals in the heterobimetallic complex 4b
appear at E1/2 = 0.85 V and E1/2 = 1.15 V (vs. SCE). The
irreversible signal has vanished in the heterobimetallic complex,
which indicates the oxidation of the carboxy group in the
monometallic complex 1b but not in the heterobimetallic complex
4b. It is well known that carboxylates undergo Kolbe electrolysis,
which includes decarboxylation. Most likely the coordination
of the carboxyl group to the [LaTp*2]+ fragment prevents this
reaction. A comparison of redox potentials of 1b and 4b is not
reasonable because of the Kolbe electrolysis of the monometallic
complex.
The cyclovoltammogram of the mononuclear nickel(II) complex
2a in dichloromethane (see Fig. SI 2, ESI†) shows two quasire-
versible signals at E1/2 = 1.10 V and E1/2 = 1.39 V (vs. SCE)
which were assigned to NiII/NiIII and NiIIIL/NiIIIL+· (L = salen
ligand), respectively. The analogous signals of the heterobimetallic
complex 5a appear at E1/2 = 1.09 V and E1/2 =1.44 V (vs. SCE)
(see Fig. SI 2, ESI†). The difference between equivalent redox
potentials of the mono- and heterobimetallic complexes is rather
small and only significant in the case of E1/2(NiIIIL/NiIIIL+·).
The redox potential E1/2(NiIIIL/NiIIIL+·) of the heterobimetallic
complex 5a is 50 mV higher than that of the monometallic
equivalent 2a, which indicates a small electron-withdrawing effect
caused by the early transition metal. To confirm this electron-
withdrawing effect, another cyclic voltammetric measurement in
the negative potential range was performed. As dichloromethane
is unsuitable in the negative potential range, thf was used as solvent
(see Fig. SI 3, ESI†). The E1/2(NiI/NiII) values of the mono-
and heterobimetallic complex were determined to be -1.66 and
-1.55 V (vs. SCE), respectively. The difference of 110 mV proves
an electron-withdrawing effect of the early transition metal. The
fact that this electron-withdrawing effect has a bigger influence
on E1/2(NiI/NiII) than on E1/2(NiII/NiIII) and E1/2(NiIIIL/NiIIIL+·)
was already observed in electrochemical investigations of Ni/Ru
heterobimetallic salen complexes.37
2.6 Catalytic investigations
Selected monometallic and heterobimetallic complexes were
tested in the homogeneous catalytic epoxidation of styrene.
meta-Chloroperoxybenzoic acid/N-methylmorpholine N-oxide
(mcpba/nmo), iodosobenzene (phIO) and tert-butyl hydroperox-
ide (tbhp) were used as oxidants, depending on the transition metal
involved (Table 3).
The sodium salts of the complexes 2a and 2d were tested because
of their better solubility in CH2Cl2.
Catalytic epoxidation reactions of olefins with Cu- and Ni-
containing salen complexes were studied previously by Decinti
et al., Kochi et al. and Burrows et al.,39 who observed a low activity
of these complexes. Similarly, we achieved moderate conversions
with the mono- and heterobimetallic Ni- and Cu-containing salen
complexes. The maximum conversion of 33.6% was observed for
the monometallic complex 1b but with a very low styrene oxide
yield of 2%.
Surprisingly the Mn, Cr and Fe complexes showed low catalytic
activity in the oxidation of styrene. The maximum conver-
sion of 24% after 15 h was obtained with the manganese(III)
complex 2c.
Extended reaction times with mcpba/nmo and phIO as oxidant
did not result in a proportional increase of the conversion,
while the opposite was observed for tbhp. This indicates that
the complexes likely decompose in the presence of the oxidants
mcpba/nmo and phIO whereas they are stable in the presence of
tbhp. This is also indicated by brightening of the reaction solutions
or suspensions after addition of mcpba/nmo and phIO within the
stated reaction time.
Table 3 reveals that the heterobimetallic complexes always
have a lower conversion rate than their monometallic analogues.
However, these lower conversion rates were accompanied by
higher selectivity with respect to styrene oxide. We assume that
steric hindrance of the early transition metal complex fragment
influences the reactivity by hampering approach to the catalytically
active centre (transition metal).
4098 | Dalton Trans., 2010, 39, 4090–4106
This journal is
The Royal Society of Chemistry 2010
©

View Online
[MoCl4(CH3CN)2],42,43 [Cp2Zr(CH3)2] and [Cp◦ Ti(CH3)2],42,44
3
Conclusions
2
[ZnCl2(thf)2],45 [LaTp*2(OTf)] and [NdTp*2(OTf)]46 were syn-
thesised according to literature procedures. [LaTp*2(pz*)] was
already obtained as a by-product by Carpentier et al.47 A new
and improved synthesis was developed for this compound.
Several monometallic salen complexes derived from O-fun-
ctionalised diamines were prepared and used as metalloligands in
the synthesis of heterobimetallic complexes. Monometallic salen
complexes could be obtained via template synthesis from nickel(II)
and copper(II). The analogous chromium(III), manganese(III) and
molybdenum(IV) salen complexes were synthesised directly from
the salen ligands. Reactions of the conjugated nickel(II) salen
complex 2a with metallocene derivatives gave soluble di- and trin-
uclear heterobimetallic complexes depending on the stoichiometry
used. Cyclovoltammetric experiments on 2a, 5a, 1b and 4b showed
the electron-withdrawing influence of the Tp*2La fragment. The
catalytic activity in the epoxidation of styrene was low for all tested
complexes. The highest values (up to 24%) for the conversion of
styrene to styrene oxide were obtained with manganese(III) salen
complexes.
Improved synthesis of [LaTp*2(pz*)]
A solution of Na(pz*) in thf (2.17 mL of a 0.809 M solution,
1.76 mmol) was added via syringe at room temperature to a
solution of [LaTp*2(OTf)] (1.55 g, 1.76 mmol) in thf (30 mL). The
solvent was removed in vacuo after stirring for 14 h. The retained
solid was suspended in toluene (10 mL) and the suspension filtered.
Subsequently, the filtrate was reduced in volume until a white
solid precipitated. This solid was redissolved by warming the
◦
suspension. Storage at -35 C overnight gave a white crystalline
solid, which was isolated by cold filtration and dried in vacuo. The
yield can be increased by layering the filtrate with n-hexane (twice
the volume) and storing the solution for 15 h at room temperature
and then for about 15 h at -35 ◦C. The resulting white precipitate
was separated by filtration and dried in vacuo. The two crops of
crystalline solid were combined.
4
Experimental
All manipulations were carried out with standard high-vacuum
and dry-nitrogen techniques. NMR spectra: Avance DRX 400
(Bruker), standards: 1H NMR (400 MHz): TMS, 13C NMR (100.6
MHz): TMS. The IR spectra were recorded as KBr pellets on a
Perkin-Elmer Spectrum 2000 FT-IR spectrometer in the range
400-4000 cm-1. The FAB mass spectra were obtained on an
MAT-8030 VG ZAB-HSQ (Finnigan). The ESI mass spectra
were recorded on an FT-ICR-MS device type Apex II (Bruker-
Daltonics). For air- and water-sensitive samples atmospheric pres-
sure photo-ionisation (APPI) was used as ionisation technique; the
samples were dissolved in toluene–methanol and evaporated in
vacuo. EPR spectra were recorded at 9.4 GHz (X band) using
an ESP 300E spectrometer (Bruker). Magnetic moments were
determined with the magnetic balances MK1 and MK2 (Johnson-
Matthey) and corrected for the diamagnetic increments. Elemental
analysis was performed on a VARIO EL (Heraeus). Melting points
were determined in sealed capillaries and are uncorrected.
Cyclic voltammetry was carried out under N2 atmosphere in
dry, O2-free dichloromethane with a three-electrode cell at room
temperature (20 2 ◦C). A platinum working electrode (diameter
2 mm, Metrohm), a platinum-plate counter-electrode (Meinsberg)
and a Ag/AgCl reference electrode (filled with LiCl-saturated
ethanol solution, Metrohm) in combination with a salt bridge,
which was freshly filled with the supporting electrolyte before all
measurements were made. A 0.1 M solution of [nBu4N]BF4 in dry,
O2-free dichloromethane was used as supporting electrolyte. The
analyte concentration was 3 mM and the scan rate 0.1 V s-1. The
cyclovoltammograms were recorded with a PAR 273A (Princeton
Applied Research, now Ametek) potentiostat and analysed with
“PowerCV”. The system Fc/Fc+ was used as external standard.
All redox potentials are referenced to SCE and were recalculated
based on the system Fc/Fc+.40
1
Yield of [LaTp*2(pz*)]: 0.84 g (58%). H NMR ([D6]benzene):
dH 6.04 (1H, s, Hpz*), 5.62 (6H, s, HTp*), 2.23 (18H, s, HTp*), 1.97
(6H, s, Hpz*), 1.79 (18H, s, HTp*). IR (KBr): 3119 (w), 2927 (m),
2551 (w), 2525 (w), 1545 (s), 1517 (m), 1489 (m), 1441 (s), 1417
(s), 1380 (m), 1362 (s), 1342 (s), 1205 (s), 1076 (s), 1068 (s), 1031
(s), 981 (m), 832 (m), 809 (s), 778 (s), 697 (m), 650 (s), 460 (w), 430
(w) cm-1.
Synthesis of Na(H2bsalcen)
A solution of 3,5-di-tert-butylsalicylaldehyde (1.67 g, 7.12 mmol)
in methanol (25 mL) was added via dropping funnel over
20 min at room temperature to a stirred suspension of (R)-2,3-
diaminopropionic acid mono-hydrochloride (0.50 g, 3.56 mmol)
and sodium hydroxide (0.28 g, 7.12 mmol) in methanol (15 mL).
The resulting yellow reaction mixture was heated to reflux for
20 min. The supernatant was filtered off from the bright yellow
solid (impurities) via a filter cannula. From the yellow solution
the solvent was evaporated in vacuo giving Na(H2bsalcen). Yield:
1.97 g (98%), mp 192–195 ◦C. Found: C, 70.04; H, 7.99; N, 4.88%.
C33H47N2O4Na requires C, 70.94; H, 8.48; N, 5.01%. IR (KBr):
3406 (w), 2958 (s), 1695 (sh), 1629 (s), 1527 (m), 1470 (s), 1441 (s),
1361 (m), 1251 (s), 1173 (m), 1113 (m), 1025 (m), 877 (m), 828 (m),
803 (m), 646 (w), 529 (w) cm-1. IR (CsI): 646 (w), 539 (m), 488
(w), 454 (w), 278 (w) cm-1.1H NMR ([D6]dmso): dH 14.47 (1H, s,
C(2)OH), 14.07 (1H, s, C(2¢)OH), 8.52 (1H, s, C(7)H=N), 8.41
(1H, s, C(7¢)H=N), 7.25 (1H, d, 4JHH = 2.4 Hz, C(6)H), 7.50 (1H,
4
4
d, JHH = 2.0 Hz, C(6¢)H), 7.19 (1H, d, JHH = 2.4 Hz, C(4)H),
7.15 (1H, d, 4JHH = 2.0 Hz, C(4¢)H), 4.16 (1H, dd, 2JHH = 12.2 Hz,
3
3
The reagents employed in the synthesis of Schiff base
ligands and their complexes were used as received. 3,4-
Diaminobenzoic acid and (R)-2,3-diaminopropionic acid mono-
hydrochloride were purchased from Merck, NiCl2·6H2O,
CuCl2 and nBu3N from Fluka and Fe(OAc)2 from Aldrich.
3,5-Di-tert-butylsalicylaldehyde41 was synthesised by follow-
ing literature procedures. [MnCl2(thf)2], [CrCl3(thf)3] and
3JHH = 3.6 Hz, C(8)H2), 4.00 (1H, dd, JHH = 8.4 Hz, JHH
=
2
3
3.6 Hz, C(9)H), 3.89 (1H, dd, JHH = 12.2 Hz, JHH = 8.4 Hz,
C(8)H2), 1.36 (9H, s, C(3)C(CH3)3), 1.34 (9H, s, C(3¢)C(CH3)3),
1.23 (18H, s, C(5)C(CH3)3 + C(5¢)C(CH3)3). 13C NMR ([D6]dmso):
1
=
dC 172.8 (s, C(9)COONa), 169.0 (d, JCH = 125 Hz, C(7)H N),
1
=
167.7 (d, JCH = 126 Hz, C(7¢)H N), 160.5 (s, C(2)OH), 159.4
(s, C(2¢)OH), 140.6 (s, C(1)), 140.0 (s, C(1¢)), 137.2 (s, C(3)),
Dalton Trans., 2010, 39, 4090–4106 | 4099
This journal is
The Royal Society of Chemistry 2010
©

View Online
1
137.0 (s, C(3¢)), 127.3 (d, JCH = 163 Hz, C(4)H + C(4¢)H +
3.45 (1H, d, 3JHH = 12.0 Hz, C(8)H2), 1.32 (9H, s, C(3)C(CH3)3),
1.31 (9H, s, C(3¢)C(CH3)3), 1.23 (9H, s, C(5)C(CH3)3), 1.21
C(6)H + C(6¢)H), 119.6 (s, C(5) + C(5¢)), 73.9 (d, 1JCH = 132 Hz,
C(9)H), 63.3 (t, 1JCH = 141 Hz, C(8)H2), 36.0 (s, C(3)C(CH3)3 +
1
(9H, s, C(5¢)C(CH3)3). 13C{ H} NMR ([D6]dmso): dC 164.3 (s,
=
=
C(3¢)C(CH3)3), 35.2 (s, C(5)C(CH3)3 + C(5¢)C(CH3)3), 32.7 (q,
C(10)OONa), 163.0 (s, C(7)H N + C(7¢)H N), 161.7 (s, C(2)O),
161.5 (s, C(2¢)O), 138.9 (s, C(1)), 138.7 (s, C(1¢)), 135.1 (s, C(3)),
135.0 (s, C(3¢)), 127.6 (s, C(4)H), 127.4 (s, C(4¢)H), 127.1 (s,
C(6)H), 126.8 (s, C(6¢)H), 120.5 (s, C(5)), 120.4 (s, C(5¢)), 69.3 (s,
C(9)H), 62.8 (s, C(8)H2), 35.8 (s, C(3)C(CH3)3 + C(3¢)C(CH3)3),
34.0 (s, C(5)C(CH3)3 + C(5¢)C(CH3)3), 31.9 (s, C(3)C(CH3)3 +
C(3¢)C(CH3)3), 30.1 (s, C(5)C(CH3)3 + C(5¢)C(CH3)3). FAB MS:
m/z 637.2 ([M + Na]+, 100.0%), 615.3 ([M + H]+, 24.1%), 593.3
([M+–Na + H]+, 19.8%).
1JCH = 130 Hz, C(3)C(CH3)3 + C(3¢)C(CH3)3), 30.7 (q, JCH
=
1
124 Hz, C(5)C(CH3)3 + C(5¢)C(CH3)3). FAB MS: m/z 625.3 ([M +
3 Na - 2 H]+, 18.6%), 603.3 ([M + 2 Na - H]+, 71.0%), 581.3 ([M +
Na]+, 100.0%).
Synthesis of H3bsal4cpn
A solution of 3,5-di-tert-butylsalicylaldehyde (4.68 g, 0.02 mol)
in thf (40 mL) was added via dropping funnel over 20 min to
a suspension of 3,4-diaminobenzoic acid (1.52 g, 0.01 mol) in thf
(50 mL). Subsequently a solution of [ZnCl2(thf)2] (2.80 g, 0.01 mol)
in thf (25 mL) was added over 10 min. The resulting yellow-green
reaction mixture was heated to reflux for 45 min and the solvent
was evaporated in vacuo. The resulting solid was suspended in
methanol (25 mL) and stirred for 15 min. After filtering off the
supernatant via a filter cannula the remaining yellow solid was
washed two times with methanol (25 mL) and dried in vacuo.
Yield of H3bsal4cpn: 5.27 g (90%), mp 231–233 ◦C. IR (KBr):
3425 (w), 2988 (s), 1690 (m), 1616 (s), 1575 (s), 1438 (m), 1362
(m), 1252 (m), 1172 (s), 1132 (w), 1026 (w), 771 (m) cm-1.
1H NMR ([D6]dmso): dH 13.58 (1H, s, C(2)OH), 13.50 (1H, s,
C(2¢)OH), 9.05 (1H, s, C(7)H=N), 8.99 (1H, s, C(7¢)H=N), 8.03
Synthesis of [Cu{Na(bsalcen)}] (1b)
A solution of NaOCH3 (0.097 g, 1.79 mmol) in methanol
(10 mL) was quickly added at room temperature to a solution
of Na(H2bsalcen) (0.50 g, 0.9 mmol) in methanol (20 mL). After
stirring the yellow solution for 10 min CuCl2 (0.12 g, 0.9 mmol)
dissolved in methanol (10 mL) was added at room temperature.
The dark-green suspension was heated to reflux for 30 min.
Subsequently, the solvent was removed in vacuo and the retained
green solid was suspended in thf (15 mL). After filtration the
solvent of the filtrate was removed in vacuo and the resulting green
solid was washed with diethyl ether (5 mL) and dried in vacuo.
Yield of 1b: 0.32 g (58%), mp 262 ◦C. Found: C, 63.08; H, 7.13;
N, 4.63%. C33H45N2O4NaCu requires C, 63.90; H, 7.31; N, 4.51%.
IR (KBr): 2957 (s), 1614 (sh), 1596 (s), 1536 (s), 1464 (sh), 1430 (s),
1410 (s), 1387 (m), 1362 (m), 1321 (m), 1273 (m), 1256 (m), 1236
3
(1H, s, C(12)H), 7.95 (1H, d, JHH = 8.0 Hz, C(9)H), 7.58 (1H,
3
d, JHH = 8.0 Hz, C(10)H), 7.57 (1H, s, C(6)H), 7.50 (1H, s,
C(6¢)H), 7.43 (1H, s, C(4)H), 7.41 (1H, s, C(4¢)H), 1.38 (18H, s,
(w), 1200 (w), 1172 (m), 1025 (w), 842 (w), 538 (w) cm-1. ESI MS
C(3)C(CH3)3 + C(3¢)C(CH3)3), 1.30 (18H, s, C(5)C(CH3)3
+
neg. (methanol): m/z 596.3 ([M–Na]-, 100%). EPR (thf): a0
=
Cu
C(5¢)C(CH3)3). (The COOH proton could not be identified -
87.9 ¥ 10-4 cm-1, g0 = 2.102 (at 293 K); gꢀ = 2.205, g^ = 2.051,
regardless the usage of [D6]dmso.) 13C{ H} NMR ([D8]thf):
1
Aꢀ = 203.7 ¥ 10-4 cm-1, A^ = 33.7 ¥ 10-4 cm-1 (at 130 K); a0
,
Cu
Cu
Cu
=
=
dC 167.6 (C(7)H N), 167.0 (C(11)COOH), 166.8 (C(7¢)H N),
159.8 (C(2)OH), 159.6 (C(2¢)OH), 147.6 (C(8)), 143.6 (C(13)),
141.2 (C(1)), 141.1 (C(1¢)), 137.7 (C(3)), 137.5 (C(3¢)), 130.8
(C(11)COOH), 129.6 (C(10)H), 129.1 (C(12)H), 128.8 (C(9)H),
128.5 (C(4)H), 128.4 (C(4¢)H), 121.7 (C(6)H), 120.7 (C(6¢)H),
119.8 (C(5)), 119.7 (C(5¢)), 33.4 (C(3)C(CH3)3 + C(3¢)C(CH3)3),
AiCu
=
1.0 ¥ 10-4 cm-1, gi = 0.001.
Synthesis of [MnCl{Na(bsalcen)}] (1c)
Solid [MnCl2(thf)2] (0.21 g, 0.79 mmol) was added at room
temperature over 15 min to a solution of Na(H2bsalcen) (0.46 g,
0.79 mmol) in thf (25 mL) and Et3N (0.22 mL, 0.16 g, 1.6 mmol).
The resulting light brown suspension was heated to reflux for 1 h.
Afterwards, air was bubbled through the dark brown solution
through an inlet for 1 h and then the solution was heated to reflux
again for 2 h. The solvent was evaporated from the resulting dark
brown solution and the remaining dark brown solid was suspended
in toluene (10 mL). After stirring at room temperature the formed
ammonium chloride was filtered off via a filter cannula and a dark
brown solid was obtained after removing the solvent in vacuo.
Yield of 1c: 0.46 g (90%), mp 220–223 ◦C. Found: C, 61.73; H,
7.43; N, 4.67%. C33H45N2O4NaMnCl requires C, 61.25; H, 7.01; N,
4.33%. IR (KBr): 3047 (sh), 2960 (s), 1611 (s), 1536 (s), 1463 (m),
1435 (s), 1390 (s), 1307 (m), 1254 (s), 1176 (s), 1098 (m), 1026 (m),
842 (m), 751 (m), 563 (m) cm-1. IR (CsI): 644 (w), 563 (m), 491 (w),
473 (w), 391 (m), 270 (w) cm-1. FAB MS: m/z 589.1 ([M–Cl–Na +
H]+, 62.9%), 544.1 ([M–Cl–COO–Na]+, 89.4%).
32.5 (C(5)C(CH3)3 + C(5¢)C(CH3)3), 28.9 (C(3)C(CH3)3
+
C(3¢)C(CH3)3), 27.1 (C(5)C(CH3)3 + C(5¢)C(CH3)3). FAB MS:
m/z 585.4 ([M + H]+, 58.9%), 376.2 ([M–C15H22O2 + H2O + H]+,
51.5%), 219.3 ([M–C15H22O2–C7H8N2O2–CH3 + 2 H2O]+, 100.0%).
Synthesis of [Ni{Na(bsalcen)}] (1a)
A solution of NaOH (0.04 g, 1.08 mmol) in methanol (10 mL) and
NiCl2·6H2O (0.13 g, 0.54 mmol) dissolved in methanol (10 mL)
was added via dropping funnel at room temperature over 15 min
to a solution of Na(H2bsalcen) (0.30 g, 0.54 mmol) in methanol
(10 mL). After heating the resulting red-brown suspension to reflux
for 1 h the supernatant solution was filtered off via a filter cannula
from the bright red solid. The solvent was evaporated from the
dark red filtrate and the remaining solid was dried in vacuo.
Yield of 1a: 0.28 g (84%), red solid, mp 284–289 ◦C. Found:
C, 63.80; H, 7.09; N, 4.13; O, 10.80%. C33H45N2O4NaNi requires
1
C, 64.40; H, 7.37; N, 4.55; O, 10.40%. H NMR ([D6]dmso): dH
Synthesis of [CrCl{Na(bsalcen)}] (1d)
7.75 (1H, s, C(7)H=N), 7.60 (1H, s, C(7¢)H=N), 7.15 (2H, s,
C(6)H + C(6¢)H), 7.03 (2H, s, C(4)H + C(4¢)H), 3.66 (1H, d,
[CrCl3(thf)3] (0.44 g, 1.16 mmol) was suspended in toluene
(10 mL), added to a solution of Na(H2bsalcen) (0.65 g, 1.16 mmol)
3JHH = 12.0 Hz, C(8)H2), 3.54 (1H, t, JHH = 12.0 Hz, C(9)H),
3
4100 | Dalton Trans., 2010, 39, 4090–4106
This journal is
The Royal Society of Chemistry 2010
©

View Online
in toluene (20 mL) and Et3N (0.32 mL, 0.24 g, 2.32 mmol) and
stirred at room temperature for 1 h. After heating the suspension
to reflux for 30 min the reaction mixture was reduced to approx.
20 mL and kept afterwards for 4 h at -1◦8 ◦C. The precipitated
ammonium chloride was filtered off at 0 C via a filter cannula.
The solvent was evaporated from the dark brown filtrate and the
remaining dark brown solid was dried in vacuo.
1.9 Hz, C(4¢)H), 1.40 (18H, s, C(3)C(CH3)3 + C(3¢)C(CH3)3),
1
1.29 (18H, s, C(5)C(CH3)3 + C(5¢)C(CH3)3). 13C{ H} NMR
=
([D8]thf): dC 167.1 (C(7)H N), 164.4 (C(11)COOH), 163.7
(C(7¢)H N), 158.2 (C(2)O), 157.9 (C(2¢)O), 146.4 (C(8)), 142.9
=
(C(13)), 139.9 (C(1)), 139.1 (C(1¢)), 136.8 (C(3)), 136.6 (C(3¢)),
130.7 (C(11)COOH), 130.3 (C(10)H), 129.4 (C(12)H), 128.7
(C(9)H), 128.3 (C(4)H), 128.0 (C(4¢)H), 120.3 (C(6)H + C(6¢)H),
117.3 (C(5)), 116.2 (C(5¢)), 35.9 (C(3)C(CH3)3 + C(3¢)C(CH3)3),
Yield of 1d: 0.73 g (98%), mp 338–340 ◦C. Found: C, 61.40; H,
8.20; N, 4.24%. C33H45N2O4NaCrCl C, 61.53; H, 7.04; N, 4.35%.
IR (KBr): 3047 (sh), 2958 (s), 2711 (w), 2486 (w), 1610 (s), 1532
(s), 1459 (s), 1411 (s), 1387 (s), 1323 (m), 1257 (s), 1170 (s), 1134
(s), 1026 (m), 814 (m), 746 (m), 545 (m) cm-1. IR (CsI): 639 (w),
547 (m), 510 (w), 494 (w), 397 (m), 279 (w) cm-1. FAB MS: m/z
586.2 ([M–Cl]+, 100.0%).
34.2 (C(5)C(CH3)3 + C(5¢)C(CH3)3), 31.6 (C(3)C(CH3)3
+
C(3¢)C(CH3)3), 30.0 (C(5)C(CH3)3 + C(5¢)C(CH3)3). ESI MS
(CH3OH/isopropanol): m/z 639.274 ([M - H]-, 100.0%).
Yield of 2b: 1.42 g (44%), red solid, mp 325◦C (decomp.). Found:
C, 68.30; H, 7.49; N, 4.63%. C37H46N2O4Cu requires C, 68.76; H,
7.17; N, 4.33%. IR (KBr): 3428 (m), 2959 (sh), 1690 (m), 1618 (s),
1597 (s), 1578 (s), 1523 (s), 1427 (s), 1358 (s), 1256 (s), 1198 (s),
1175 (s), 1129 (m), 1025 (w), 792 (m), 766 (w), 533 (m) cm-1. ESI
MS (isopropanol): m/z 644.262 ([M - H]-, 100.0%). EPR (thf):
Synthesis of [MoCl2{Na(bsalcen)}] (1e)
a0 = 85.2 ¥ 10-4 cm-1, g0 = 2.106 (at 293 K); gꢀ = 2.217, g^ =
Cu
A suspension of [MoCl4(CH3CN)2] (0.20 g, 0.63 mmol) in toluene
(10 mL) was added at room temperature over 10 min to a
solution of Na(H2bsalcen) (0.35 g, 0.63 mmol) in toluene (15 mL)
and Et3N (0.18 mL, 0.13 g, 1.26 mmol). The resulting black
brown suspension was heated to reflux for 90 min. The formed
ammonium chloride was filtered off via a filter cannula and a
black solid was obtained after removing the solvent in vacuo.
Black crystals suitable for X-ray analysis could be obtained from
an acetonitrile solution of 1e at room temperature.
2.050, Aꢀ = 202.7 ¥ 10-4 cm-1, A^ = 30.8 ¥ 10-4 cm-1 (at 130
Cu
Cu
K); a0Cu, AiCu
=
1.0 ¥ 10-4 cm-1, gi = 0.001.
Synthesis of [NHEt3][MnCl(bsal4cpn)] (2c)
Solid [MnCl2(thf)2] (0.16 g, 0.6 mmol) was added at room
temperature over 5 min to a solution of H3bsal4cpn (0.35 g,
0.6 mmol) in thf (35 mL). The resulting light brown suspension was
stirred for 1 h at room temperature and heated to reflux afterwards
for 20 min. Et3N (0.17 mL, 0.12 g, 1.2 mmol) was added via pipette
to the reaction mixture, and then air was bubbled through the dark
brown solution via an inlet for 45 min. The resulting dark brown
solution was reduced to dryness and the remaining dark brown
solid was suspended in toluene (10 mL). After stirring for 2 h at
room temperature the formed light brown solid was left to settle
over night. The supernatant solution was filtered off via a filter
cannula and a dark brown solid was retained after removing the
solvent in vacuo.
Yield of 1e: 0.33 g (72%), mp 290–295 ◦C. Found: C, 54.50; H,
7.27; N, 4.36%. C33H45N2O4NaMoCl2 requires C, 54.78; H, 6.27;
N, 3.87%. IR (KBr): 3017 (sh), 2961 (s), 1618 (m), 1603 (m), 1537
(m), 1463 (m), 1362 (m), 1261 (s), 1246 (s), 1098 (m), 1025 (m), 805
(s), 560 (m) cm-1. IR (CsI): 644 (w), 560 (m), 491 (w), 483 (w), 465
(w), 392 (m), 319 (w), 278 (w) cm-1. ESI MS (toluene–CH3CN):
m/z 749.37 ([M–2 Cl + O + Et3NH]+, 62.9%). Magnetic moment:
2.51 BM (calc. 2.83 BM (spin only)).
Template synthesis of [Ni(Hbsal4cpn)] (2a) and [Cu(Hbsal4cpn)]
(2b)
Yield of 2c: 0.25 g (63%), mp 242–245 ◦C. Found: C, 68.90; H,
7.14%. C37H46N2O4MnCl requires C, 66.02; H, 6.89%. IR (KBr):
3430 (m), 2960 (m), 1670 (sh), 1610 (s), 1577 (s), 1463 (s), 1389 (s),
1361 (s), 1313 (s), 1255 (s), 1198 (s), 1180 (s), 1097 (s), 1027 (s),
807 (s), 781 (s), 545 (m) cm-1. ESI MS (thf/CH3CN): m/z 774.380
([M + NHEt3]+, 100.0%), 637.280 ([M–Cl]+, 25.9%).
General method: A solution of 3,4-diaminobenzoic acid (0.76 g,
5 mmol) in methanol (70 mL) and the corresponding hydrated
metal chloride (5 mmol) dissolved in methanol (40 mL) was added
via dropping funnel at room temperature over 10 min to a solution
of 3,5-di-tert-butylsalicylaldehyde (2.34 g, 10 mmol) in methanol
(70 mL). After refluxing the resulting brown suspension for 1 h
and reducing the volume to 30 mL, the reaction mixture was
left overnight and a red solid precipitated, which was washed
three times with methanol (10 mL) and dried in vacuo. Red crystals
suitable for X-ray analysis were obtained from 1,4-dioxane.
Yield of 2a: 1.82 g (57%), red solid, mp 380 ◦C. Found: C,
64.60; H, 7.44; N, 3.69%. C37H46N2O4Ni + 2 H2O + CH3OH
requires C, 64.33; H, 7.67; N, 3.95%. IR (KBr): 3422 (m), 3054
(m), 2959 (s), 1688 (m), 1618 (s), 1579 (s), 1526 (s), 1463 (m),
1426 (s), 1386 (s), 1359 (s), 1260 (m), 1200 (s), 1180 (s), 1130
(m), 1026 (m), 788 (m), 541 (m) cm-1. 1H NMR ([D6]dmso):
Synthesis of [CrCl(Hbsal4cpn)] (2d)
A solution of [CrCl3(thf)3] (0.19 g, 0.5 mmol) in toluene (25 mL)
was added via dropping funnel at room temperature within 15 min
to a solution of H3bsal4cpn (0.29 g, 0.5 mmol) in toluene (20 mL).
After heating the suspension to reflux for 15 min and reducing
the resulting dark red mixture to dryness, the solid was washed
two times with diethyl ether (10 mL) and dried in vacuo. Red
crystals suitable for X-ray analysis were obtained either from
acetonitrile ([CrCl(CH3CN)(Hbsal4cpn)], 2d¢) or on addition of
pz*H in 1,4-dioxane/n-hexane ([CrCl(pz*H)(Hbsal4cpn)], 2d¢¢).
Yield of 2d: 0.33 g (98%), dark red solid, mp 350 ◦C. Found: C,
65.20; H, 7.83; N, 4.23%. C37H46N2O4CrCl requires C, 66.31; H,
6.92; N, 4.18%. IR (KBr): 3385 (w), 3051 (sh), 2959 (s), 1715 (sh),
1612 (s), 1576 (s), 1526 (s), 1461 (sh), 1362 (s), 1328 (sh), 1259 (s),
1198 (s), 1172 (s), 1099 (s), 1023 (s), 872 (m), 804 (s), 730 (m) cm-1.
dH 13.24 (1H, s, C(11)COOH), 8.98 (1H, s, C(7)H=N), 8.91
3
(1H, s, C(7¢)H=N), 8.72 (1H, s, C(12)H), 8.27 (1H, d, JHH
=
3
8.8 Hz, C(9)H), 7.84 (1H, d, JHH = 8.8 Hz, C(10)H), 7.62
4
4
(1H, d, JHH = 2.4 Hz, C(6)H), 7.48 (1H, d, JHH = 1.9 Hz,
C(6¢)H), 7.37 (1H, d, 4JHH = 2.4 Hz, C(4)H), 7.37 (1H, d, 4JHH
=
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 4090–4106 | 4101
©

View Online
FAB MS: m/z 669.2 ([M]+, 10.0%), 634.3 ([M–Cl]+, 100.0), 619.3
([M–Cl–CH3]+, 69.0%), 585.3 ([M–Cr–Cl–3 H]+, 38.0%). Magnetic
moment: 3.60 BM (calc. 3.87 BM (spin only)).
3JHH = 8.8 Hz, C(9)H), 8.03 (2H, s, C(7¢)H=N), 7.69 (2H, s,
C(12)H), 7.65 (2H, s, C(6)H), 7.60 (2H, s, C(6¢)H), 7.08 (2H, d,
3JHH = 8.8 Hz, C(10)H), 7.02 (2H, s, C(4)H), 6.84 (2H, s, C(4¢)H),
6.20 (10H, s, (C5H5)2), 1.73 (36H, s, C(3)C(CH3)3 + C(3¢)C(CH3)3),
1.41 (18H, s, C(5)C(CH3)3), 1.29 (18H, s, C(5¢)C(CH3)3). FAB MS:
m/z 1498.5 ([M]+, 0.7%), 640.3 ([M–Ni(Hbsal4cpn)–Zr(C5H5)2 +
H]+, 100.0%).
Synthesis of [Fe(bsal4cpn)] (2e)
Solid Fe(OAc)2 (0.12 g, 0.68 mmol) was added at room tem-
perature to a solution of H3bsal4cpn (0.4 g, 0.68 mmol) in
thf/methanol (25 mL/25 mL). The resulting dark brown sus-
pension was heated to reflux for 45 min in air. After cooling
the solution to room temperature, the solvent was removed in
vacuo. The retained solid was suspended in methanol (10 mL) and
then filtered. The dark brown solid was washed with diethyl ether
(10 mL) and dried for 15 h in vacuo at 10◦0 ◦C.
Synthesis of [Cp◦ Ti{Ni(bsal4cpn)}2] (3b)
2
In a round-bottom Schlenk flask 2a (0.48 g, 0.74 mmol) was dried
in vacuo for 30 min and then suspended in toluene (40 mL). A
solution of [Cp◦ Ti(CH3)2] (0.14 g, 0.37 mmol) in toluene (50 mL)
2
was added via dropping funnel at room temperature over 15 min.
After addition, slight gas evolution was observed. The reaction
mixture was stirred at room temperature for 14 h. A small amount
of precipitate was filtered off via a filter cannula. After removing
the solvent in vacuo from the red filtrate, a dark red solid was
obtained.
Yield of 2e: 0.37 g (85%), mp > 300 C. Found: C, 69.27; H,
7.26%. C37H45N2O4Fe requires C, 69.70; H, 7.11%. IR (KBr): 2958
(s), 1663 (m), 1609 (s), 1579 (s), 1551 (m), 1531 (s), 1463 (m), 1426
(s), 1388 (m), 1377 (m), 1361 (s), 1312 (m), 1272 (m), 1253 (s), 1238
(m), 1198 (m), 1176 (m), 1132 (m), 1099 (w), 870 (w), 842 (m), 822
(w), 783 (w), 774 (w), 749 (w), 543 (m) cm-1. ESI MS (methanol–
acetonitrile/thf): m/z 1293.6 ([2M + 2H + OH]+, 100.0%), 1275.6
([2M + H]+, 32.1%), 710.3 ([M + H + thf]+, 47.1%). Magnetic
moment: 5.99 BM (calc. 5.90 BM (d5 high spin, spin only)).
Yield of 3b: 0.46 g (76%), mp 400 ◦C (decomp.). Found: C,
69.29; H, 8.15; N, 3.25%. C96H124N4O8Ni2Ti requires C, 70.86;
H, 7.68; N, 3.44%. IR (KBr): 3349 (m), 3255 (m), 2960 (s),
2869 (m), 1610 (sh), 1580br, 1424br, 1383br, 1261 (m), 1198
(m), 1179 (m), 1024br, 786 (m), 729 (w), 666 (m), 622 (m), 495
(s) cm-1. 1H NMR ([D6]dmso): dH 7.62 (4H, s, C(7)H=N +
C(7¢)H=N), 7.44 (8H, br, C(6)H, C(6¢)H, C(4)H + C(4¢)H), 6.87
Synthesis of [MoCl2(Hbsal4cpn)] (2f)
nBu3N (1.81 mL, 1.41 g, 7.6 mmol) was added via pipette to a
solution of H3bsal4cpn (2.20 g, 3.8 mmol) in thf (30 mL). Then
solid [MoCl4(CH3CN)2] (1.21 g, 3.8 mmol) was added at room
temperature over 15 min to the orange solution. After refluxing
the suspension for 50 min and reducing the resulting black-brown
solution to half its volume, the supernatant dark brown solution
was filtered off via a filter cannula. The remaining black solid
was washed two times with thf (10 mL) and dried in vacuo. Black
crystals suitable for X-ray analysis were obtained from acetonitrile
at room temperature.
3
3
(2H, d, JHH = 8.0 Hz, C(10)H), 6.76 (2H, d, JHH = 8.0 Hz,
C(9)H), 6.42 (2H, s, C(12)H), 3.65 (4H, s, (C5(CH2CH3)(CH3)4)2),
1.68 (6H, s, (C5(CH2CH3)(CH3)4)2), 1.66 (6H, s, (C5(CH2CH3)-
(CH3)4)2), 1.59 (6H, s, (C5(CH2CH3)(CH3)4)2), 1.36 (6H, s,
(C5(CH2CH3)(CH3)4)2), 1.31 (36H, s, C(3)C(CH3)3 + C(3¢)C-
(CH3)3), 1.19 (36H, s, C(5)C(CH3)3 + C(5¢)C(CH3)3), 1.04 (6H, s,
(C5(CH2CH3)(CH3)4)2).
Synthesis of [Tp*2La{Ni(bsalcen)}] (4a)
Yield of 2f: 1.85 g (65%), mp 222 ◦C. Found: C, 59.70; H,
7.14%. C37H46N2O4MoCl2 requires C, 59.28; H, 6.18%. IR (KBr):
3415 (w), 3043 (sh), 2961 (s), 1715 (m), 1602 (m), 1573 (s), 1530
(s), 1463 (m), 1363 (m), 1261 (s), 1184 (m), 1099 (s), 1027 (m), 803
(s), 770 (s), 705 (m) cm-1. FAB MS: m/z 750.1 ([M]+, 90.0%), 715.2
([M–Cl]+, 76.0%), 696.2 ([M–Cl–OH–2 H]+, 100.0%), 585.3 ([M–
Mo–2 Cl–2 H]+, 60.0). Magnetic moment: 2.85 BM (calc. 2.83 BM
(spin only)).
A solution of [LaTp*2(OTf)] (0.20 g, 0.23 mmol) in toluene
(10 mL) was added over 5 min at 0 ◦C to a suspension of 1a (0.14 g,
0.23 mmol) in toluene (10 mL). The suspension was stirred for 2 h
at 0 ◦C and then filtered. The solvent was removed in vacuo and the
retained brown solid was washed with a small amount of diethyl
ether. The resulting compound was dried in vacuo. Red crystals
suitable for X-ray analysis were obtained from n-hexane.
Yield of 4a: 0.18 g (60%), mp 250–255 ◦C. Found: C, 57.46; H,
6.58; N, 14.53%. C63H89B2N14O4LaNi requires C, 57.08; H, 6.77;
N, 14.79%. IR (KBr): 2958 (s), 2555 (w), 2513 (w), 1615 (s), 1573
(s), 1543 (s), 1437 (s), 1416 (s), 1360 (s), 1325 (s), 1259 (s), 1203 (s),
1176 (s), 1099 (s), 1070 (s), 1035 (s), 982 (m), 869 (m), 807 (s), 784
(s), 697 (m), 637 (m) cm-1. 1H NMR ([D6]benzene): dH 7.60 (1H, d,
Synthesis of [Cp2Zr{Ni(bsal4cpn)}2] (3a)
In a round-bottom Schlenk flask 2a (0.7 g, 1.1 mmol) was dried
in vacuo for 30 min and afterwards suspended in toluene (25 mL). A
solution of [Cp2Zr(CH3)2] (0.137 g, 0.55 mmol) in toluene (50 mL)
was added via a Schlenk flask with a cone inlet at room temperature
over 10 min. After the addition, slight gas evolution was observed.
The reaction mixture was heated to reflux for 30 min and then
stirred at room temperature for 10 h. A small amount of precipitate
was filtered off via a filter cannula. After removing the solvent
in vacuo from the red filtrate a dark red solid was obtained.
Yield of 3a: 0.73 g (88%), mp 360 ◦C (decomp.). IR (KBr): 3047
(sh), 2957 (m), 1720 (w), 1646 (w), 1577 (s), 1524 (s), 1424 (s), 1358
(m), 1259 (s), 1179 (s), 1100 (s), 1022 (m), 810 (s), 730 (m) cm-1.
1H NMR ([D8]toluene): dH 8.62 (2H, s, C(7)H=N), 8.17 (2H, d,
4JHH = 2.0 Hz, Harom.), 7.52 (1H, d, 4JHH = 2 Hz, Harom.), 7.16 (1H, s,
4
HC N), 6.69 (1H, d, JHH = 2.0 Hz, Harom.), 6.62 (1H, d, 4JHH
=
=
=
2.0 Hz, Harom.), 6.53 (1H, s, HC N), 5.72 (6H, s, HTp*), 4.76 (2H,
br, BHTp*), 3.06 (1H, d, 2JHH = 14 Hz, H2C.), 2.82 (1H, d, 3JHH
=
6 Hz, HCCOO), 2.70 (1H, dd, 2JHH = 14 Hz,3JHH = 6 Hz, H2C),
2.11 (18H, s, HTp*), 2.10 (18H, s, HTp*), 1.83 (9H, s, HtBu), 1.78
(9H, s, HtBu), 1.41 (9H, s, HtBu), 1.25 (9H, s, HtBu). 13C{ H} NMR
1
([D8]benzene): dC 181.1 (COO), 165.6 (C=N), 162.1 (C=N), 150.4
(CTp*), 145.7 (CTp*), 163.6, 163.2, 140.7, 140.6, 135.4, 135.2, 129.0,
128.6, 126.8, 126.5, 120.7, 120.5 (Carom), 106.8 (CTp*), 71.1 (CH2),
4102 | Dalton Trans., 2010, 39, 4090–4106
This journal is
The Royal Society of Chemistry 2010
©

View Online
59.3 (CCOO), 36.4 (CtBu), 36.3 (CtBu), 34.1 (CtBu), 33.9 (CtBu), 31.8
(CtBu), 31.6 (CtBu), 30.29 (CtBu), 30.27 (CtBu), 13.8 (CTp*), 13.2 (CTp*).
114.1, 106.6 (CTp*), 2 ¥ 36.4 (CtBu), 2 ¥ 34.1 (CtBu), 2 ¥ 31.5 (CtBu),
2 ¥ 30.2 (CtBu), 13.5 (CTp*), 13.4 (CTp*).
Synthesis of [Tp*2La{Cu(bsal4cpn)}] (5b)
Synthesis of [Tp*2La{Cu(bsalcen)}] (4b)
The synthesis of 5b was carried out according to the synthesis of
5a. Brown crystals suitable for X-ray analysis were obtained from
n-hexane.
A solution of [LaTp*2(OTf)] (0.20 g, 0.22 mmol) in thf (5 mL) was
added at room temperature to a solution of 1b (0.14 g, 0.22 mmol)
in thf (15 mL). The resulting suspension was stirred for 6 h at room
temperature. Afterwards, the solvent was removed in vacuo, the
retained solid suspended in toluene (10 mL) und the resulting
suspension filtered. The volume of the filtrate was reduced to
1–2 mL in vacuo. Subsequently, the green solution was covered
with a layer of n-hexane (8 mL). Brown crystals suitable for X-ray
analysis formed overnight.
◦
Yield of 5b: 0.28 g (65%), brown solid, mp 230 C. Found: C,
57.91; H, 6.15; N, 14.29%. C67H89B2N14O4LaCu requires C, 58.37;
H, 6.51; N, 14.22%. IR (KBr): 2957 (s), 2556 (w), 2523 (w), 1612
(sh), 1604 (s), 1579 (s), 1645 (s), 1523 (s), 1463 (s), 1427 (s), 1385
(s), 1360 (s), 1259 (s), 1198 (s), 1174 (s), 1131 (m), 1068 (m), 1031
(m), 981 (w), 829 (m), 809 (s), 791 (m), 779 (m), 695 (w), 649 (m),
534 (w), 459 (w) cm-1. Magnetic moment: 1.89 BM (calc. 1.73 BM
Yield of 4b: 0.14 g (50%), mp 235 ◦C. Found: C, 57.69; H, 6.49;
N, 14.82%. C63H89B2N14O4LaCu requires C, 57.87; H, 6.74; N,
14.74%. IR (KBr): 2958 (s), 2553 (w), 2523 (w), 1615 (s), 1568 (m),
1442 (s), 1528 (s), 1439 (s), 1415 (s), 1389 (s), 1361 (s), 1325 (m),
1271 (m), 1256 (m), 1233 (w), 1202 (s), 1168 (s), 1070 (m), 1036
(m), 981 (m), 864 (w), 832 (w), 807 (m), 789 (m), 778 (m), 746
(spin only)). EPR (thf): a0 = 88.2 ¥ 10-4 cm-1, g0 = 2.104 (at 293
Cu
K); gꢀ = 2.214, g^ = 2.049, Aꢀ = 201.4 ¥ 10-4 cm-1, A^ = 35.7 ¥
Cu
Cu
10-4 cm-1 (at 130 K); (a0Cu, AiCu
=
1.0 ¥ 10-4 cm-1, gi = 0.001).
Synthesis of [Tp*2Nd{Ni(bsal4cpn)}] (6a)
(w), 729 (w), 696 (w), 646 (m), 534 (w), 463 (w) cm-1. Magnetic
The synthesis of 6a was carried out according to the synthesis of
5a.
Cu
moment: 1.68 BM (calc. 1.73 BM (spin only)). EPR (thf): a0
=
87.9 ¥ 10-4 cm-1, g0 = 2.101 (at 293 K); gꢀ = 2.203, g^ = 2.051,
Yield of 6a: 0.36 g (73%), red solid, mp 290 ◦C. Found: C, 58.35;
H, 6.50; N, 14.35%. C67H89B2N14O4NdNi requires C, 58.35; H,
6.50; N, 14.28%. IR (KBr): 2957 (s), 2557 (w), 2524 (w), 1616 (m),
1605 (m), 1579 (s), 1545 (s), 1525 (s), 1417 (s), 1360 (s), 1263 (m),
1200 (s), 1179 (s), 1068 (m), 1033 (s), 831 (m), 809 (m), 788 (m), 778
(m), 648 (w) cm-1. 1H NMR ([D8]thf): dH 21.61 (1H, Harom.), 20.56
Aꢀ = 205.7 ¥ 10-4 cm-1, A^ = 32.9 ¥ 10-4 cm-1 (at 130 K); a0
,
Cu
Cu
Cu
AiCu
=
1.0 ¥ 10-4 cm-1, gi = 0.001.
Synthesis of [Tp*2La{Ni(bsal4cpn)}] (5a)
=
(1H, Harom.), 12.50 (1H, Harom.), 12.66 (1H, HC N), 11.00 (1H,
A solution of NaOH (14 mg, 0.36 mmol) in methanol (1.5 mL)
was added at room temperature to a suspension of 2a (0.23 g,
0.36 mmol) in methanol (10 mL). After stirring for 1 h the solvent
was removed in vacuo. The retained red solid was suspended in
toluene/thf (15 mL/5 mL) and the suspension filtered. The solvent
of the filtrate was removed in vacuo. The resulting red solid was
dried, weighed and then suspended in toluene (10 mL). A solution
of [LaTp*2(OTf)] (0.33 g, 0.37 mmol) in toluene (10 mL) was
added. The suspension was stirred for 3 h at room temperature and
then filtered. The solvent was removed in vacuo and the retained
red solid washed with n-hexane (5 mL) and dried in vacuo. Red
crystals suitable for X-ray analysis were obtained by diffusion of
n-hexane vapour into a solution of the product in 1,4-dioxane.
Yield of 5a: 0.26 g (54%), mp 240 ◦C. Found: C, 58.78; H,
6.36; N, 14.67%. C67H89B2N14O4LaNi requires C, 58.58; H, 6.53;
N, 14.28%. IR (KBr): 2957 (s), 2553 (w), 2523 (w), 1616 (m), 1605
(m), 1579 (s), 1544 (s), 1524 (s), 1417 (s), 1402 (s), 1384 (s), 1361
(s), 1263 (m), 1200 (s), 1180 (s), 1131 (m), 1099 (w), 1069 (m), 1033
(m), 981 (w), 846 (w), 830 (m), 809 (m), 787 (s), 778 (s), 697 (w),
649 (m), 541 (w), 465 (w), 437 (w) cm-1. 1H NMR ([D6]benzene):
=
HC N), 8.90 (1H, Harom), 8.50 (1H, Harom), 8.00 (1H, Harom), 7.89
(1H, Harom), 2.04 (9H, HtBu), 2.01 (9H, HtBu), 1.85 (9H, HtBu), 1.68
1
(9H, HtBu). 13C{ H} NMR ([D8]thf): dC 193.3, 167.3, 166.5, 160.2
(C=N), 158.9 (C=N), 151.9, 150.8, 144.2, 143.9, 142.0, 141.5,
138.1, 137.8, 131.8, 131.4, 129.1, 129.7, 122.9, 122.6, 37.2 (CtBu),
35.2 (CtBu), 32.1 (CtBu), 30.9 (CtBu), 15.0 (CTp*). CHTp* not observed.
Magnetic moment: 4.02 BM (calc. 3.62 BM).
Synthesis of [Tp*2Nd{Cu(bsal4cpn)}] (6b)
The synthesis of 6b was carried out according to the synthesis of
5a.
Yield of 6b: 0.58 g (66%), brown solid, mp 237 C. Found: C,
◦
57.49; H, 6.87; N, 13.64%. C67H89B2N14O4NdCu requires C, 58.15;
H, 6.48; N, 14.17%. IR (KBr): 2958 (s), 2554 (w), 2524 (w), 1615
(sh), 1604 (s), 1580 (s), 1545 (s), 1523 (s), 1427 (s), 1385 (s), 1361
(s), 1259 (m), 1199 (s), 1174 (s), 1131 (m), 1099 (w), 1068 (m), 1032
(m), 982 (w), 830 (m), 809 (m), 792 (m), 779 (m), 648 (m) cm-1.
Cu
Magnetic moment: 4.01 BM (calc. 3.92 BM). EPR (thf): a0
=
88.2 ¥ 10-4 cm-1, g0 = 2.105 (at 293 K); gꢀ = 2.214, g^ = 2.049,
3
Cu
Cu
Cu
Aꢀ = 203.1 ¥ 10-4 cm-1, A^ = 35.7 ¥ 10-4 cm-1 (at 130 K); (a0
,
dH 8.36 (1H, s, Harom.), 7.81 (1H, d, JHH = 8.8 Hz, Harom.), 7.65
(1H, d, 4JHH = 2.4 Hz, Harom.), 7.62 (1H, s, HC N), 7.58 (1H, d,
=
AiCu
=
1.0 ¥ 10-4 cm-1, gi = 0.001).
4
4
=
JHH = 2.4 Hz, Harom.), 7.47 (1H, s, HC N), 6.94 (1H, d, JHH
=
3
2.0 Hz, Harom.), 6.66 (1H, d, JHH = 8.8 Hz, Harom.), 6.39 (1H, d,
4JHH = 2.0 Hz, Harom.), 5.64 (6H, s, HTp*), 4.89 (2H, br, BHTp*),
2.16 (18H, s, HTp*), 2.06 (18H, s, HTp*), 1.77 (9H, s, HtBu), 1.75
Synthesis of [Tp*2La{CrCl(pz*H)(bsal4cpn)}] (5c)
Solid [LaTp*2(pz*)] (0.26 g, 0.31 mmol) was added at room
temperature to a solution of 2d (0.22 g, 0.31 mmol) in thf (10 mL).
The resulting red solution was stirred for 18 h at room temperature.
Subsequently the solvent was removed in vacuo and the retained
solid suspended in n-hexane (10 mL). The suspension was stirred
(9H, s, HtBu), 1.38 (9H, s, HtBu), 1.28 (9H, s, HtBu). 13C{ H} NMR
1
([D8]benzene): dC 180.4 (COO), 154.8 (C=N), 154.7 (C=N), 150.1
(CTp*), 145.4 (CTp*), 166.2, 165.5, 145.8, 142.9, 141.5, 140.8, 136.9,
136.4, 133.9, 130.9, 130.6, 127.6, 127.1, 127.9, 120.2, 120.1, 115.8,
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 4090–4106 | 4103
©

View Online
4104 | Dalton Trans., 2010, 39, 4090–4106
This journal is
The Royal Society of Chemistry 2010
©

View Online
for 10 min and then filtered. The collected red solid was dried in
vacuo.
on a Siemens CCD diffractometer (SMART)48 in w scan mode.
Data reduction with SAINT,49 including the program SADABS50
for empirical absorption correction. The data for 2a·1,4-dioxane,
5a·1,4-dioxane·0.5n-hexane and 5b·n-hexane were collected on a
Stoe IPDS imaging plate diffractometer, f scan mode, numerical
absorption correction with XRED,51 and the data for 4b·n-hexane
and 2d¢¢·1,4-dioxane on an Xcalibur-S diffractometer (Oxford
Diffraction), w and f scan mode. Data reduction with CrysAlis
Pro,52 including the program SCALE3 ABSPACK53 for empirical
absorption correction. Used radiation: Mo-Ka (l = 71.073 pm).
The structures were solved by direct methods and the refinement of
all non-hydrogen atoms was performed with SHELX97.54 Details
are summarised in Table 4. Structure figures were generated with
DIAMOND-3.55
Yield of 5c: 0.21 g (45%), mp > 300 ◦C. Found: C, 55.89; H, 6.85;
N, 14.92%. C67H89B2N14O4LaCrCl·C5H8N2 requires C, 57.70; H,
6.53; N, 14.96%. IR (KBr): 3404 (w), 3293 (w), 2958 (s), 2553 (w),
2524 (w), 1610 (s), 1579 (s), 1543 (s), 1526 (s), 1417 (s), 1379 (s),
1362 (s), 1274 (m), 1256 (m), 1197 (s), 1172 (s), 1134 (m), 1097 (m),
1070 (m), 1035 (s), 982 (m), 872 (w), 829 (m), 809 (m), 783 (s), 748
(w), 697 (w), 650 (m), 540 (m), 464 (w), 431 (w) cm-1. Magnetic
moment: 4.11 BM (calc. 3.92 BM (spin only)).
Synthesis of [Tp*2La{FeCl(bsal4cpn)}] (5d)
HCl/diethyl ether (0.15 mL of 2 M solution, 0.30 mmol) was
added at room temperature to a stirred solution of 2e (0.19 g,
0.30 mmol) in thf (10 mL). The solution was stirred for 10 min.
Subsequently the solvent was removed in vacuo. The retained solid
was dissolved in thf (15 mL) and was then treated with a solution
of [LaTp*2(pz*)] (0.24 g, 0.3 mmol) in thf (6 mL). The solution
was stirred for 14 h at room temperature and then the solvent
was removed in vacuo. The resulting brown solid was dissolved in
hot n-hexane (10 mL). A microcrystalline brown solid precipitated
overnight and was dried in vacuo.
Acknowledgements
We gratefully acknowledge financial support from the DFG
(Deutsche Forschungsgemeinschaft, Graduate College 378
“Mechanisms and Applications of Non-Conventional Oxidation
Reactions” and the International Postgraduate Programme (IPP)
at Universita¨t Leipzig. We thank Prof. Kirmse and Dr Jan Griebel
for the EPR measurements.
◦
Yield of 5d: 0.21 g (50%), mp > 300 C. Found: C, 57.18; H,
6.82; N, 14.01%. C67H89B2N14O4LaFeCl requires C, 57.21; H, 6.38;
N, 14.01%. IR (KBr): 2958 (s), 2553 (w), 2521 (w), 1608 (s), 1579
(s), 1543 (s), 1530 (s), 1424 (s), 1377 (s), 1361 (s), 1274 (m), 1255
(s), 1199 (s), 1177 (s), 1134 (m), 1119 (w), 1099 (w), 1068 (m), 1032
(s), 981 (m), 870 (m), 830 (s), 810 (s), 780 (s), 748 (m), 730 (w),
697 (m), 665 (w), 649 (m), 542 (m), 492 (w), 461 (w), 426w cm-1.
Magnetic moment: 5.07 BM (calc. 5.92 BM (spin only)).
Notes and references
1 K. Srinivasan, P. Michaud and J. K. Kochi, J. Am. Chem. Soc., 1986,
108, 2309.
2 W. Zhang, J. L. Loebach, R. W. Scott and E. N. Jacobsen, J. Am. Chem.
Soc., 1990, 112, 2801.
3 R. Irie, K. Noda, Y. Ito, M. Matsumoto and T. Katsuki, Tetrahedron
Lett., 1990, 31, 7345.
4 (a) C. T. Dalton, K. M. Ryan, V. M. Wall, C. Bousquet and D. G.
Gilheany, Top. Catal., 1998, 5, 75; (b) L. Canali and D. C. Sherrington,
Chem. Soc. Rev., 1999, 28, 85; (c) E. N. Jacobsen, M. H. Wu in
E. N. Jacobsen, A. Pfaltz and H. Yamamoto (Hrsg.), Comprehensive
Asymmetric Catalysis, Springer, Berlin, 1999, 649; (d) T. Katsuki, Adv.
Synth. Catal., 2002, 344, 131.
5 (a) A. R. Silva, C. Freire and B. de Castro, New J. Chem., 2004, 28, 253;
(b) P.G. Cozzi, Chem. Soc. Rev., 2004, 33, 410.
6 Y. N. Ito and T. Katsuki, Tetrahedron Lett., 1998, 39, 4325.
7 (a) K. Dey, A. K. Biswas and A. K. Sinha Roy, Indian J. Chem., 1981,
20A, 848; (b) A. Shafir, D. Fiedler and J. Arnold, J. Chem. Soc., Dalton
Trans., 2002, 555.
8 (a) O. Kahn, P. Tola and H. Coudanne, Inorg. Chim. Acta, 1978, 31,
L405; (b) Y. Journaux, O. Kahn and H. Coudanne, Angew. Chem., Int.
Ed. Engl., 1982, 21, 624; (c) W.-F. Yeung, P.-H. Lau, T.-C. Lau, H.-Y.
Wei, H.-L. Sun, S. Gao, Z.-D. Chen and W.-T. Wong, Inorg. Chem.,
2005, 44, 6579.
9 S. Fritzsche, P. Lo¨nnecke, T. Ho¨cher and E. Hey-Hawkins, Z. Anorg.
Allg. Chem., 2006, 632, 2256.
10 S. Schley, P. Lo¨nnecke and E. Hey-Hawkins, J. Organomet. Chem.,
2009, 694, 2480.
11 (a) T. Luts, W. Suprun, D. Hofmann, O. Klepel and H. Papp, J. Mol.
Catal. A: Chem., 2007, 261, 16; (b) T. Luts, R. Frank, W. Suprun, S.
Fritzsche, E. Hey-Hawkins and H. Papp, J. Mol. Catal. A: Chem., 2007,
273, 250.
Synthesis of [(Tp*2La{Fe(bsal4cpn)})2(l-O)] (5e)
NaOH in methanol (1 mL of a 1.02 M solution, 1.07 mmol)
was added at room temperature to a suspension of 2e (0.68 g,
1.07 mmol) in methanol (10 mL). During the addition the solid
dissolved and the colour changed from green to dark brown. The
solution was stirred for 1 h and separated from a small amount
of solid by filtration. The solvent of the filtrate was removed in
vacuo. Subsequently, the orange-brown solid was dissolved in
toluene (20 mL) and treated with a solution of [LaTp*2(OTf)]
(0.94 g, 1.07 mmol) in toluene (20 mL). The brown suspension
was stirred for 16 h at room temperature and then filtered. The
solvent of the filtrate was removed in vacuo, the retained orange
solid was washed with n-hexane and dried in vacuo. Crystals were
obtained by dissolving the solid in toluene (10 mL) and vapour-
phase diffusion of n-hexane into this solution after three weeks.
◦
Yield of 5e: 0.59 g (41%), mp > 300 C. Found: C, 59.40; H,
6.92; N, 13.99%. C134H178B4N28O9La2Fe2·C7H8 requires C, 59.42;
H, 6.58; N, 13.76%. IR (KBr): 2958 (s), 2556 (w), 2524 (w), 1610
(s), 1600 (sh), 1579 (s), 1544 (s), 1527 (s), 1423 (s), 1387 (s), 1361
(s), 1258 (m), 1198 (s), 1175 (s), 1069 (m), 1034 (m), 868 (m), 829
(m), 809 (m), 780 (m), 649 (m), 537 (m) cm-1. Magnetic moment:
3.01 BM (calc. 8.37 BM (spin only)).
12 (a) Nomenclature: 3,5-di-tert-butyl-salicylaldehyde (bsal) and (R)-2,3-
diaminopropionic acid (cen) give the salen ligand H3bsalcen; 3,5-di-
tert-butyl-salicylaldehyde (bsal) and 3,4-diaminobenzoic acid (4cpn)
give the salen ligand H3bsal4cpn.
13 Y.N. Ito and T. Katsuki, Tetrahedron Lett., 1998, 39, 4325.
14 (a) R. Grigg and H. Q. N. Gunaratne, Tetrahedron Lett., 1983, 24, 4457;
(b) K. Amornraksa, R. Grigg, H. Q. Nimal Gunaratne, J. Kemp and V.
Sridharan, J. Chem. Soc., Perkin Trans. 1, 1987, 2285.
15 U. Casellato, P. Guerriero, S. Tamburini, P.A. Vigato and C. Benelli,
Inorg. Chim. Acta, 1993, 207, 39.
X-ray crystallography
The data for 2d¢·6CH3CN, 2f·4CH3CN, the decomp. product
2f¢, 4a·n-hexane, 5e·3.5toluene and 5f·2n-hexane were collected
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 4090–4106 | 4105
©

View Online
16 S.S. Mandal, U. Varshney and S. Bhattacharya, Bioconjugate Chem.,
1997, 8, 798.
34 (a) G. A. Jeffrey, An Introduction to Hydrogen Bonding, Oxford
University Press, Oxford, 1997; (b) G. R. Desiraju and T. Steiner, The
Weak Hydrogen Bond, Oxford University Press, Oxford, 1999.
35 N. Wheatley and P. Kalck, Chem. Rev., 1999, 99, 3379.
36 (a) M. J. Begley, P. Mountford, P. J. Stewart, D. Swallow and S. Wan,
J. Chem. Soc., Dalton Trans., 1996, 1323; (b) M. Leschke, H. Lang and
R. Holze, J. Solid State Electrochem., 2003, 7, 518.
37 N. Onozawa-Komatsuzaki, R. Katoh, Y. Himeda, H. Sugihara, H.
Arakawa and K. Kasuga, Bull. Chem. Soc. Jpn., 2003, 76, 977.
38 R. Klement, F. Stock, H. Elias, H. Paulus, P. Pelika´n, M. Valko and M.
Maru´r, Polyhedron, 1999, 18, 3617.
17 (a) G. Reddelien, Chem. Ber., 1910, 43, 2476; (b) E.C. Escudero-Ada´n,
J. Benet-Buchholz and A.W. Kleij, Dalton Trans., 2008, 734; (c) A.M.
Castilla, S. Curreli, E.C. Escudero-Ada´n, M.M. Belmonte, J. Benet-
Buchholz and A.W. Kleij, Org. Lett., 2009, 11, 5218.
18 J. Weidlein, U. Mu¨ller and K. Dehnicke, Schwingungsfrequen-
zen II - Nebengruppenelemente, Georg Thieme Verlag, Stuttgart,
1986.
19 A. Van Den Bergen, K.S. Murray and B.O. West, Aust. J. Chem., 1972,
25, 705.
20 U. Kynast, Dissertation, Universita¨t Marburg, 1984.
21 W. Levason, J.S. Ogden and A.J. Rest, J. Chem. Soc., Dalton Trans.,
1980, 419.
22 (a) M. Gerloch and F. Mabbs, J. Chem. Soc., 1967, 1598; (b) M. Lubben,
A. Meetsma, F. v. Bolhuis and B. L. Feringa, Inorg. Chim. Acta, 1994,
215, 123; (c) H. C. Shyu, H.-H. Wei, G.-H. Lee and Y. Wang, J. Chem.
Soc., Dalton Trans., 2000, 915.
23 K. Oyaizu and E. Tsuchida, Inorg. Chim. Acta, 2003, 355, 414.
24 M. Cametti, A. D. Cort, M. Colapietro, G. Portalone, L. Russo and K.
Rissanen, Inorg. Chem., 2007, 46, 9057.
39 (a) S. Zolezzi, E. Spodine and A. Decinti, Polyhedron, 2003, 22, 1653;
(b) D. J. Koola and J. K. Kochi, Inorg. Chem., 1987, 26, 908; (c) H.
Yoon and C. J. Burrows, J. Am. Chem. Soc., 1988, 110, 4087.
40 P. Zanello, A. Cinquantini, S. Mangani, G. Opromolla, L. Pardi, C.
Janiak and M. D. Rausch, J. Organomet. Chem., 1994, 471, 171.
41 J.F. Larrow and E.N. Jacobsen, J. Org. Chem., 1994, 59, 1939.
42 B. Heyn, B. Hipler, G. Kreisel, H. Schreer and D. Walther, Anorganische
Synthesechemie, Springer, Berlin, 1986.
43 J.R. Dilworth, R.L. Richards, G.J.-J. Chen and J.W. McDonald, Inorg.
Synth., 1990, 28, 33.
25 (a) J. H. Billman and A. C. Diesing, J. Org. Chem., 1957, 22, 1068; (b) I.
Correia, J. C. Pessoa, M. T. Duarte, M. F. M. Da Piedade, T. Jackush,
T. Kiss, M. M. C. A. Castro, C. F. G. C. Geraldes and F. Avecilla,
Eur. J. Inorg. Chem., 2005, 732.
26 W.-J. Ruan, G.-H. Hu, S.-J. Wang, J.-H. Tian, Q.-L. Wang and Z.-A.
Zhu, Chin. J. Chem., 2005, 23, 709.
27 R. Atencio, R. Gaitan, S. Pekerar and J.D. Medina, Acta Crystallogr.,
Sect. C: Cryst. Struct. Commun., 1997, 53, 1068.
28 P.J. Pospisil, D.H. Carsten and E.N. Jacobsen, Chem.–Eur. J., 1996, 2,
974.
44 E. Samuel and M.D. Rausch, J. Am. Chem. Soc., 1973, 95, 6263.
45 A. Dashti, K. Niediek, B. Werner and B. Neumu¨ller, Z. Anorg. Allg.
Chem., 1997, 623, 394.
46 S.-Y. Liu, G. H. Maunder, A. Sella, M Stevenson and D.A. Tocher,
Inorg. Chem., 1996, 35, 76.
47 F. A. Kunrath, O. L. Casagrande Jr., L. Toupet and J.-F. Carpentier,
Polyhedron, 2004, 23, 2437.
48 SMART: Area-Detector Software Package. (1993) Siemens Industrial
Automation, Inc., Madison, WI.
49 SAINT: Area-Detector Integration Software. Version 6.01 (1999)
Siemens Industrial Automation, Inc., Madison, WI.
50 SADABS: G.M. Sheldrick, SADABS, Program for Scaling and Cor-
rection of Area-detector Data, Go¨ttingen, 1997.
51 X-RED Version 1.22 (2001), STOE Data Reduction Program, STOE
& Cie GmbH Darmstadt.
29 M.V. Caparelli, B. Piggott, S.D. Thorpe and R.N. Sheppard, Inorg.
Chim. Acta, 1985, 98, L53.
30 K. Brychcy, K. Dra¨ger, K.-J. Jens, M. Tilset and U. Behrens, Chem.
Ber., 1994, 127, 465.
31 M.E. Bluhm, M. Ciesielski, H. Go¨rls, O. Walter and M. Do¨ring, Inorg.
Chem., 2003, 42, 8878.
52 CrysAlis Pro: Data collection and data reduction software package,
32 A search of copper- and nickel-containing salen complexes with an
aromatic conjugated diimine backbone in the Cambridge Structural
Database (CCD) resulted in 27 Ni salen complexes and 34 Cu salen
complexes. The range of Ni–O, Cu–O, Ni–N and Cu–N bond lengths of
these complexes were 182.1–190.0 pm, 180.5–193.9 pm, 181.3-189.3 pm
und 183.3-199.5 pm, respectively.
Oxford Diffraction Ltd.
53 SCALE3 ABSPACK: Empirical absorption correction using spherical
harmonics.
54 SHELX97 [Includes SHELXS97, SHELXL97]: G.M. Sheldrick (1997).
SHELX97. Programs for Crystal Structure Analysis (Release 97-2),
University of Go¨ttingen, Germany.
33 R. J. H. Clark, S. Y. Liu, G. H. Maunder, A. Sella and M. R. J. Elsegood,
J. Chem. Soc., Dalton Trans., 1997, 2241.
55 DIAMOND 3: K. Brandenburg, Crystal Impact GbR, Bonn,
Germany.
4106 | Dalton Trans., 2010, 39, 4090–4106
This journal is
The Royal Society of Chemistry 2010
©