Synthesis and characterization of mono- And bi-metallic Mn(III) complexes containing salen type ligands

Article abstract of DOI:10.1039/a908304i

A mononuclear complex 1, of Mn(in) with a SchifF-base ligand L [L = N,N′-ethylenebis(3-formyl-5-methylsalicylaldimine)] has been synthesized and structurally characterized. A series of binuclear complexes, MnML′Cl_xsolv, [M = Zn(II), Cu(n), Ni(n

Full text of DOI:10.1039/a908304i

View Article Online / Journal Homepage / Table of Contents for this issue
Synthesis and characterization of mono- and bi-metallic Mn(III)
complexes containing salen type ligands
Debasis Das and Cheu Pyeng Cheng*
Department of Chemistry, National Tsing Hua University, Hsin-Chu 300, Taiwan.
E-mail: cpcheng@mx.nthu.edu.tw
Received 18th October 1999, Accepted 1st February 2000
Published on the Web 8th March 2000
A mononuclear complex 1, of Mn() with a Schiff-base ligand L [L = N,NЈ-ethylenebis(3-formyl-5-
methylsalicylaldimine)] has been synthesized and structurally characterized. A series of binuclear complexes,
MnMLЈClxsolvy [M = Zn(), Cu(), Ni(), Co(), Fe(), and Mn(); x = 3 or 4 depending on the oxidation state
of M; LЈ is the aniline condensation product of L; y = 1, 2 or 3 and solv = H2O or CH3OH] have been prepared
by the reaction of 1, aniline and M(/) chloride with a 1:3:1.2 molar ratio in methanolic media. These bimetallic
complexes were characterized by elemental analyses, UV-Vis and IR spectroscopies, magnetic susceptibilities and
thermogravimetric analyses. The results of magnetic susceptibility measurements at 300 K indicate that all the metal
ions are in their high spin states. The catalytic properties of these complexes for epoxidation of alkenes have been
investigated using PhIO as a terminal oxidant. The binuclear Mn() complexes are less efficient epoxidation catalysts
than the mononuclear Mn() complex, 1. The reduced activities of the bimetallic complexes may be due to both the
low rates of formation of a Mn()᎐O intermediate and the transfer of oxygen from this intermediate to the alkene.
᎐
Furthermore, the bimetallic complexes also transformed into a catalytically inactive species in the presence of PhIO
as indicated by UV-Vis spectroscopic studies.
salicylaldimine) and in 1981 Gagne et al.25 synthesized Ni()
and Cu() complexes of a similar ligand system. Those com-
Introduction
Mn() complexes of salen type ligands have attracted much
plexes were characterized by elemental analyses, magnetic
interest in the last few decades because of their unique catalytic
moment, IR and UV-Vis spectroscopic studies. The authors
activity, especially as epoxidation catalysts, in the presence
presumed that the metal ions bind preferentially to the N2O2
of a terminal oxidant like iodosylbenzene.1–19 Numerous salen
site. To date no mononuclear metal complex of ligand L has
ligands have been prepared to investigate steric, electronic and
yet been characterized by X-ray crystallography. Therefore, we
overall structural effects on the catalytic activity of the Mn()
prepared Mn()LCl, 1 and characterized it structurally by the
center. N,NЈ-Ethylenebis(3-formyl-5-methylsalicylaldimine) (L)
single crystal X-ray diffraction method (vide infra).
is an extensively studied salen type ligand. Surprisingly, to the
best of our knowledge, its Mn() complex has not yet been
Bimetallic complexes containing Mn() could not be suc-
cessfully prepared from Mn()LCl. This difficulty was over-
reported. Moreover, no mononuclear complex of the ligand
come by condensing the formyl groups with aniline to create
has been structurally characterized by X-ray crystallography in
order to delineate the donor atoms of the ligand. This ligand
may be used to synthesize binuclear complexes with a second
metal ion in close proximity to the Mn() center and thereby
another coordination site for the second metal in Mn()LЈCl
(Scheme 1).25 The six bimetallic complexes MMnLЈClxsolvy (for
solv = CH3OH: M = Zn(), x = 3, y = 2, 2; for solv = H2O:
M = Cu(), x = y = 3, 3; M = Ni(), x = y = 3, 4; M = Co(),
provide a means of studying the effect of a second metal on the
x = 3, y = 1, 5; M = Fe(), x = 4, y = 1, 6; M = Mn(), x = 4,
catalytic activity of Mn(). Further understanding of metal–
y = 1, 7) were prepared by the reactions of Mn()LЈCl with
metal interactions is very important because the interactions
are critical in some bimetallic metalloproteins. Some illustrative
various metal chlorides. We tried but failed to prepare Pd()-
Mn()LЈClxsolvy by the same procedure. This failure may be
examples of bimetallic metalloproteins involved in the reaction
ascribed to the size of the second N2O2 site which is too small to
and transport of dioxygen in biological systems20–23 are given
below. In cytochrome c oxidase, the binding and initial reduc-
tion of dioxygen occurs on a site of coupled copper and iron
centers. The transport of oxygen requires a bicopper center in
hemocyanin and a biferrous center in hemerythrin. Here, we
report the syntheses of bimetallic Mn() complexes based on
the monomanganese complex of L. The catalytic activities of
these complexes for the epoxidation of alkenes were investi-
gated to probe the influence of the second metal on the catalytic
efficiency of the Mn() center.
accommodate the second row transition metal ion.
The mononuclear Mn() complex 1 exhibits characteristic
IR bands at 1660, 1625 and 1546 cmϪ1 assigned to C᎐O, C᎐N
᎐
᎐
and skeletal vibrations.24,25 The 1660 cmϪ1 band is absent in the
spectra of all the binuclear complexes whereas the other two
bands remain, suggesting the complete condensation of the
formyl groups of L with aniline. The effective magnetic
moments at 300 K and the expected spin-only values are col-
lected in Table 1. The spin-only values in Table 1 were calcu-
¹

²
lated using the equations µMn = 2[SMn(SMn ϩ 1)] for complex 1
¹
2
2

²
and µMn-M = (µMn ϩ µM) for binuclear complexes. The effective
magnetic moment of complex 1 agrees well with that predicted
for a high-spin d4 configuration Mn() ion in the complex. The
effective magnetic moments of the binuclear complexes are
in good agreement with their corresponding spin-only value
assuming no magnetic interaction between the metal ions.
Results and discussion
Synthesis and characterization
In 1972, Okawa snd Kida24 first reported Ni() and Cu()
complexes of
L
(L = N,NЈ-ethylenebis(3-formyl-5-methyl-
DOI: 10.1039/a908304i
J. Chem. Soc., Dalton Trans., 2000, 1081–1086
This journal is © The Royal Society of Chemistry 2000
1081

View Article Online
Table 1 Effective magnetic moments of the complexes 1–7 at 300 K
Table 2 Selected bond distances (Å) and bond angles (Њ) for complex
1ؒ0.39H2O, Mn()LClؒ0.39H2O [L = N,NЈ-ethylenebis(3-formyl-5-
methylsalicylaldiimine)]
Complex
µeff/µB
µspin only a /µB
M: Sb
Mn(1)–O(1)
Mn(1)–N(1)
Mn(1)–Cl(1)
O(2)–C(16)
N(2)–C(10)
N(2)–N(9)
1.878(3)
1.987(3)
2.3691(14)
1.318(5)
1.291(5)
1.472(5)
1.437(6)
1.437(6)
Mn(1)–O(2)
Mn(1)–N(2)
O(1)–C(1)
N(1)–C(7)
N(1)–C(8)
C(8)–C(9)
C(1)–C(6)
C(11)–C(16)
1.878(3)
1.969(3)
1.341(5)
1.284(5)
1.481(5)
1.513(7)
1.395(6)
1.411(6)
1
2
3
4
5
6
7
4.79
4.78
5.09
5.61
6.16
7.48
6.78
4.90
4.90
5.20
5.66
6.24
7.68
6.93
—
Zn(): 0
Cu(): 1/2
Ni(): 1
Co(): 3/2
Fe(): 5/2
Mn(): 2
C(6)–C(7)
C(10)–C(11)
a Calculated based on the assumption that Mn() is in a high spin state
in all the complexes. b The metal ion and the spin state of the second
O(1)–Mn(1)–N(1)
O(1)–Mn(1)–O(2)
O(1)–Mn(1)–Cl(1)
N(1)–Mn(1)–Cl(1)
O(1)–Mn(1)–N(2)
89.65(13)
91.47(13)
99.81(11)
95.96(11)
160.70(15)
O(2)–Mn(1)–N(2)
N(1)–Mn(1)–N(2)
O(2)–Mn(1)–Cl(1)
N(2)–Mn(1)–Cl(1)
O(2)–Mn(1)–N(1)
91.26(14)
81.92(14)
101.84(11)
98.32(11)
161.70(15)
metal in complexes 2–7 used to calculate µspin only
.
Fig. 1 ORTEP35 drawing of complex 1ؒ0.39H2O with the atom
numbering scheme. Thermal ellipsoids are scaled to enclose 50%
electron density. Hydrogens have an arbitrary radius of 0.1 Å.
formyl groups in the salen ligand.1–3,26,27 The four atoms in the
basal plane deviate slightly from planarity. The average devi-
ation from the best plane through these four atoms is 0.0042 Å,
with O(1) and N(2) sitting above the plane while O(2) and
N(1) sit below it. Mn() lies 0.3035 Å above the basal plan. A
comparison among analogous penta coordinated Schiff-base
complexes of Mn(), [Mn(salen)Cl],26 [Mn(acen)Cl] [H2acen =
N,NЈ-ethylenebis(acetylacetone imine)]27 and [Mn(tetramethyl-
salen)Cl]3 shows that these complexes all have similar structures
but in [Mn(salen)Cl] the Mn atom is only 0.19 Å above the
N2O2 plane, whereas in 1, [Mn(acen)Cl] and in [Mn(tetramethyl-
salen)Cl] complexes the corresponding distances are 0.3035,
0.344 and 0.305 Å, respectively. This distinction is well corre-
lated with their respective Mn–Cl bond lengths. The Mn–Cl
distances are 2.3691(14), 2.381(1) and 2.391(2) Å in 1, (acen)
and (tetramethylsalen) complexes, respectively, whereas this
distance in the (salen) complex is noticeably longer, 2.461(1) Å.
Scheme 1
Therefore, these data indicate that both Mn() and the second
metal ion are in their high-spin states. Complexes 1, 2 and 7 are
EPR silent and thereby exclude the presence of Mn() impur-
ities in these complexes. The coordinating solvents have been
verified by the appropriate weight losses in their thermogravi-
metric (TG) analyses. These weight losses and their correspond-
ing temperatures are indicated in the Experimental section.
Structure of complex 1
The structure of this complex is shown in Fig. 1; selected bond
distances and angles are given in Table 2. Mn() in the complex
is penta coordinated and the geometry can be best described as
square pyramidal (Fig. 1). It also contains 0.39 mol of water of
crystallization which is different from that found for the sample
described in the Experimental section. This results from differ-
ent crystallization conditions. The two phenolic O atoms and
two imine N atoms of the Schiff base occupy the basal plane
of the square pyramid. This indicates clearly that the formyl
groups in this ligand do not coordinate to Mn(). The struc-
ture of this N2O2 coordination site also supports the proposed
structures of CuL and NiL.24,25 The Mn–O distances of
1.878(3) Å (for both) and Mn–N distances of 1.969(3) and
1.987(3) Å (Table 2) are comparable to those previously
reported for Mn()(salen) derivatives in which there are no
Epoxidation of alkenes
The catalytic activities of the monomanganese complex 1 and
the bimetallic complexes 2–7 were compared for the epoxid-
ation of alkenes by a terminal oxidant. The epoxidations were
carried out in CH3CN using PhIO as the terminal oxidant at
25 ЊC. The amounts of PhI and epoxide formed on epoxidation
of (E)-stilbene catalyzed by complexes 1–7 as a function of
time are shown in Fig. 2 and 3. Within 5 min, 0.160 mmol of
PhI and 0.117 mmol of trans-stilbene oxide were formed on
epoxidation of (E)-stilbene by complex 1, whereas, those
amounts were 0.120 to 0.020 mmol and 0.082 to 0.0025 mmol,
respectively, when the binuclear complexes 2–7 were used as
catalysts. In all cases, the initial rates of formation of PhI and
1082
J. Chem. Soc., Dalton Trans., 2000, 1081–1086

View Article Online
Table 3 Yields of epoxides from the reactions of alkenes with PhIO catalyzed by complexes 1–7a
Yield of epoxides with different catalystsb (%)
Olefins
1
2
3
4
5
6
7
(E)-PhCH᎐CH–Ph
86
63
22
12
38
21
25
11
5
63.5
15
8
37
21
73
21
10
47
39
11
6
45
19
9
82
31
17
51
43
14
9
᎐
(Z)-PhCH᎐CH–Ph
48 (cis)
25 (trans)
63
53
28
᎐
PhCH᎐CH2
12
28
᎐
c
(E)-PhCH᎐CH–CH3
19
᎐
c
c
c
(Z)-PhCH᎐CH–CH3
8
6
᎐
c
c
c
15 (trans)
a 0.300 mmol of alkene, 1.00 × 10Ϫ2 mmol of catalyst and 0.300 mmol of PhIO in 5 mL acetonitrile at 25 ЊC. b Yields are based on iodobenzene
formed, determined by GC analysis. The products for (E)-alkenes are trans-epoxides only. c Yield is negligibly low.
of the nature of alkene. The reactivities of the alkenes follow
the order: (E)-stilbene > (Z)-stilbene > styrene > (E)-β-methyl-
styrene > (Z)-β-methylstyrene. It is noteworthy that trans-
alkenes produce trans-epoxides, whereas cis-disubstituted
alkenes produce both cis- and trans-epoxides on epoxidation.
The experimental results indicated that the presence of a
second metal in complexes 2–7 reduces the catalytic activity of
the Mn() centers. In order to further understand the inter-
action of the catalyst with PhIO, we studied the reactions of the
catalysts 1–7 with PhIO (0.010 mmol catalyst and 0.300 mmol
PhIO in 5 mL CH3CN) in the absence of any substrate. The
Mn() complexes react with PhIO to produce iodoxybenzene
PhIO2.28 Complex 1 produced 0.130 mmol of PhI and the
binuclear complexes yielded PhI in the range of 0.060 to 0.090
mmol after 1 h of reaction. These data, which are consistent
with the slow rates of formation of PhI for complexes 2–7
shown in Fig. 2, reveal that the low efficiency of the binuclear
Fig. 2 PhI formation as a function of time in the epoxidation of
(E)-stilbene with PhIO catalyzed by complexes 1–7 (0.300 mmol of
alkene, 1.00 × 10Ϫ2 mmol of catalyst and 0.300 mmol of PhIO in
acetonitrile (5 mL) at 25 ЊC).
complexes for alkene epoxidation compared to that of the
corresponding mononuclear Mn() is not due to the decom-
position of PhIO by any other path. If a Mn()᎐O intermedi-
᎐
ate, which was first proposed by Kochi et al.1 and later
supported by the electrospray tandem mass spectroscopic
data,29 is formed in both the decomposition of PhIO and the
epoxidation reaction, then this intermediate is more slowly
formed for the binuclear complexes 2–7 than for 1. Since the
yields of epoxide relative to PhI are low for the complexes 2–7
catalyzed reactions, the effectiveness of the transfer of an oxy-
gen atom from the Mn()᎐O intermediate to the alkene is also
᎐
reduced for these complexes. One reasonable explanation is that
the presence of the second metal may destabilize the Mn()᎐O
᎐
intermediate because of a repulsive charge interaction between
Mn() and the second metal as well as a possible charge-
induced bonding change. However, since the order of effective-
ness of catalysts 2–7 does not correlate with the charge of the
second metal ion, there must be other factors which influence
the activity of the bimetallic complexes. The affinity of the
second metal for alkenes is not well recognized. So the coordin-
ation of alkenes to the second metal may not be a significant
factor. The presence of the second metal may also affect the
coordination environments of the reactive Mn() center or
Fig. 3 Epoxide formation as a function of time for the epoxidation of
(E)-stilbene with PhIO catalyzed by complexes 1–7 (0.300 mmol of
alkene, 1.00 × 10Ϫ2 mmol of catalyst and 0.300 mmol of PhIO in
acetonitrile (5 mL) at 25 ЊC).
Mn()᎐O intermediate. Unfortunately, up to now, we failed to
᎐
obtain single crystals of any binuclear complex. On the other
hand, crystal structures of Mn() complexes of salen deriv-
atives indicate that salen derivatives are pliable in forming the
manganese complexes.30 Therefore, a change in the coordin-
ation environments for Mn() induced by the second metal
may play an important role in the catalytic activity of com-
plexes 1–7.
To examine the nature of the complexes 1–7 in the catalytic
process, we studied the UV-Vis spectral changes for complexes
1–7 in the presence of PhIO (1.00 × 10Ϫ3 mmol complex in
acetonitrile (5 mL) was treated with 0.300 mmol PhIO). The
results for complexes 1 and 5 are shown in Fig. 4A and 4B,
respectively. Complex 5 was chosen as a representative example
epoxide were rapid, except for the reaction catalyzed by com-
plex 3. After 30 min the rates of formation of PhI and epoxide
formation slowed down in every case. From Fig. 2 and 3, the
effectiveness of the catalysts in producing trans-stilbene oxide
followed the order 1 > 7 > 5 > 4 ≈ 2 > 6 > 3. The percentages
of epoxide (with respect to PhI) formed after a one hour
reaction period by complexes 1–7 are collected in Table 3. We
also studied the epoxidation of (Z)-stilbene, styrene, (E)-β-
methylstyrene and (Z)-β-methylstyrene under the same condi-
tions. From the results given in Table 3, it is clear that the order
of catalytic activity of the metal complexes is independent
J. Chem. Soc., Dalton Trans., 2000, 1081–1086
1083

View Article Online
oxidation product was detected for the binuclear catalysts. The
poor efficiency of complex 1 is proposed to be related to its
penta coordinate geometry as reported earlier.3,31 We also per-
formed the oxidation of cyclohexane using complexes 1–7 as
catalyst. No oxidation product was detected even with complex
1. This indicates that the Mn()᎐O species generated from
᎐
complexes 1–7 are not as reactive as that which was obtained
from Kochi’s 5,5Ј-dinitrosalen Mn() complex to activate the
C–H bonds of alkanes.1
Conclusion
A series of bimetallic Mn() complexes MMnLЈClxsolvy (for
solv = CH3OH: M = Zn(), x = 3, y = 2, 2; for solv = H2O:
M = Cu(), x = y = 3, 3; M = Ni(), x = y = 3, 4; M = Co(),
x = 3, y = 1, 5; M = Fe(), x = 4, y = 1, 6; M = Mn(), x = 4,
y = 1, 7), were synthesized from the monomanganese complex
Mn()LCl, 1. The X-ray structure of 1 indicates that the
coordination site is N2O2, confirming the earlier assumptions of
the coordination site of L. The presence of the second metal ion
does not alter the high spin state of the Mn() ions in the
bimetallic complexes 2–7. In the epoxidation of alkenes, includ-
ing (E)-stilbene, (Z)-stilbene, styrene, (E)-β-methylstyrene and
(Z)-β-methylstyrene, by PhIO, 1–7 yielded essentially the same
products, i.e., trans-alkene to trans-epoxide and cis-alkene to a
mixture of trans- and cis-epoxides. From the rate of formation
of PhI from PhIO and epoxide from (E)-stilbene, it is obvious
that the second metal ion in 2–7 slows down the formation of
the active intermediate Mn()᎐O and reduces the efficiency of
᎐
the transfer of an oxygen atom to the alkene. The influence of
the second metal follows the order Cu() > Fe() > Zn() ≈
Ni() > Co() > Mn(). Furthermore, in the presence of
PhIO, the bimetallic complexes also transform into inactive
species in contrast to the monometallic complex 1, which is
stable and reduced PhIO to PhI effectively. Therefore, the sec-
ond metal exhibits a negative effect on the catalytic activity of
Mn() in catalyzing the epoxidation of alkenes by PhIO.
Fig. 4 UV-Vis spectral pattern of complex 1 (A) and 5 (B): (a)
2.00 × 10Ϫ3 M solution of complex in acetonitrile at room temper-
ature; (b) one hour after addition of PhIO (0.300 mmol) (sample
diluted by ≈2-fold) and (c) after one day of the reaction (sample diluted
by ≈2-fold).
of the binuclear complexes as they all exhibited similar changes
in their UV-Vis spectra. Fig. 4 shows spectra before addition of
PhIO (a); after 1 h of reaction (b) and after one day of reaction
(c). PhIO was completely reacted after one day in a solution
of complex 1 while a substantial amount of PhIO remained
unreacted in the solutions of complexes 2–7. Complex 1 and the
binuclear complexes exhibited similar spectra after 1 h reaction
(spectra (b); Fig. 4). After one day, complex 1 produced
spectrum (c) which is identical to its original one (spectrum (a),
Fig. 4A) except for the contribution from PhI, whereas,
binuclear complexes yielded spectrum (c) (Fig. 4B) which was
somewhat different from their original spectra (the spectrum
was measured after removing the unreacted PhIO from the
reaction mixture).
The species 1c and 5c which were responsible for spectra (c)
in Fig. 4A and 4B, respectively and were obtained after one day
reaction of 1 and 5 with PhIO, were further treated with fresh
PhIO (0.300 mmol). GC analyses were performed on the reac-
tion mixtures after 1 h of reaction. 5c was observed to be
inactive to produce PhI, whereas, 1c was found to be as reactive
as 1 and this reactivity was noticed to be retained beyond ten
cycles. Furthermore 1c is as effective as 1 in catalyzing the
epoxidation reaction while 5c is not an active catalyst. From
these observations, we may infer that initially complex 1 as well
as the binuclear complexes react with PhIO in a similar fashion
but at reduced rates for the binuclear complexes. However, the
binuclear complexes lose their reactivity gradually due to their
conversion to some inactive species, whereas, complex 1 retains
its reactivity for a long time and consequently exhibits a higher
efficiency.
Experimental
Physical measurements
Elemental analyses (C, H, and N) of the complexes were carried
out on a Heraeus CHN rapid analyzer. The metal contents of
the complexes were determined by a Perkin-Elmer ICP mass
spectrometer (model Optima 3000 DV). NMR spectra were
recorded on a Bruker AC300 spectrometer, IR spectra (as KBr
discs) were recorded on a Jasco FT-IR (model 300E) spectro-
photometer and electronic spectra (in dry methanol) were
recorded on a Perkin-Elmer Lambda 5 UV-Vis spectro-
photometer. Magnetic susceptibilities were measured on a
Quantum Design MPMS SQUID magnetometer (applied field
5 T) at 300 K. Diamagnetic corrections were applied using
Pascal’s constants.32 GC analysis was performed on a Shimadzu
Gas Chromatograph 8A instrument equipped with a flame
ionization detector. Thermogravimetric analysis was carried
out with a Seiko 300 Thermal Analyzer under a nitrogen
atmosphere.
Synthesis
2,6-Diformyl-4-methylphenol25
and
N,NЈ-ethylenebis(3-
formyl-5-methylsalicylaldimine)24 (L) were prepared according
to the literature methods. Alkenes were purchased from Aldrich
and used without further purification. The epoxides were
obtained either from Aldrich or prepared by the epoxidation
of the corresponding alkene with m-chloroperbenzoic acid
(Aldrich) in methylene chloride.33 Other chemicals and solvents
were reagent grade, and were used without further purification.
Solvents were dried according to standard procedures and
distilled before use.
In addition to the oxidation of alkenes of styrene derivatives,
we also studied the catalytic oxidation of simple alkenes like
cyclohexene with complexes 1–7. Very low percentage yields
of cyclohexene epoxide (15%) and cyclohexanol (5%) were
observed when complex 1 was used as a catalyst, whereas, no
1084
J. Chem. Soc., Dalton Trans., 2000, 1081–1086

View Article Online
MnLClؒ2H2O (1ؒ2H2O). To a heated suspension of L (7.04 g,
0.02 mol) in ethanol (50 mL), an aqueous solution (20 mL) of
MnCl2ؒ4H2O (3.96 g, 0.02 mol) was added dropwise. A brown
color developed immediately upon dissolution of the ligand.
The resulting solution was allowed to stir open to the atmos-
phere for 12 h. A brown complex precipitated and was isolated
by filtration, washed with water and dried in vacuum at 50 ЊC
for 12 h. Yield 8.5 g (89% with respect to L). The complex thus
obtained was recrystallized from acetonitrile (Found: C, 50.36;
H, 5.05; N, 5.96; Mn, 11.38. C20H18N2O4MnClؒ2H2O requires
C, 50.37; H, 4.62; N, 5.88; Mn, 11.53%). IR(KBr)/cmϪ1: 1660
analysis: 0.19 mg weight loss (2.63% of 7.21 mg complex;
expected weight loss: 2.44%) at 118 ЊC.
MnFeLЈCl4ؒH2O (6). Yield: 68% (Found: C, 49.74; H, 3.95;
N, 7.21; Mn, 7.03; Fe, 7.50) C32H28N4O2Cl4MnFeؒH2O requires
C, 49.82; H, 3.89; N, 7.27; Mn, 7.13; Fe, 7.62%). IR(KBr)/cmϪ1
:
1638 (C᎐N) and 1548 (skeletal vibration). UV-Vis(MeOH)/nm:
᎐
499 (sh, ε = 7500), 359 (ε = 15000). TG analysis: 0.21 mg weight
loss (2.74% of 7.29 mg complex; expected weight loss: 2.33%)
at 122 ЊC.
(C᎐O), 1625 (C᎐N) and 1546 (skeletal vibration). UV-Vis-
Mn2LЈCl4ؒH2O (7). In this case the reflux was carried out
in open atmosphere. Yield: 76% (Found: C, 49.78; H, 3.73; N,
7.22; Mn, 14.02. C32H28N4O2Cl4Mn2ؒH2O requires C, 49.88;
᎐
᎐
(MeOH)/nm: 421 (sh, ε = 9500), 361 (ε = 12000). TG analysis:
0.356 mg weight loss (7.56% of 4.71 mg complex; expected
weight loss: 7.56%) at 123 ЊC.
H, 3.89; N, 7.27; Mn, 14.27%). IR(KBr)/cmϪ1: 1629 (C᎐N) and
᎐
1550 (skeletal vibration). UV-Vis(MeOH)/nm: 428 (sh,
ε = 11000), 372 (ε = 16000). TG analysis: 0.15 mg weight loss
(2.41% of 7.48 mg sample; expected weight loss: 2.33%) at
115 ЊC.
MnZnLЈCl3ؒCH3OH (2). Aniline (0.41 mL; 4.50 mmol)
in dry methanol (10 mL) was added dropwise to a methanolic
(15 mL) solution of MnLClؒ2H2O (0.715 g, 1.50 mmol) with
constant stirring at room temperature. After 1 h solid ZnCl2
(0.245 g, 1.80 mmol) was added to that solution and the
stirring was continued for an additional 1 h. The mixture was
then cooled in an ice bath for 40 min. The yellowish brown
product was isolated by filtration, washed with dry methanol
rapidly to remove the unreacted complex 1ؒ2H2O which is highly
soluble in methanol and dried under vacuum. Yield: 0.87 g
(76%). Recrystallization was accomplished by dissolving the
complex in a minimum amount (5 mL) of dry methanol fol-
lowed by dropwise addition of dry ether (10 mL) (Found: C,
52.42; H, 4.36; N, 7.30; Mn, 7.19; Zn, 8.83. C32H28N4O2Cl3-
MnZnؒCH3OH requires C, 52.18; H, 4.22; N, 7.38; Mn, 7.24;
Iodosylbenzene. This was prepared by hydrolysis of the corre-
sponding diacetate with aqueous sodium hydroxide as reported
in the literature.34 Freshly prepared PhIO was used in every
epoxidation experiment.
Epoxidation of alkenes catalyzed by complexes 1–7
All epoxidation reactions were carried out by the same pro-
cedure. In a typical experiment, 0.300 mmol of alkene and
1.00 × 10Ϫ2 mmol of catalyst were treated with acetonitrile
(5 mL) under a nitrogen atmosphere. Chlorobenzene (0.300
mmol) was added as an internal standard for GC analysis. A
GC analysis was performed by withdrawing aliquots in small
portions from the reaction mixture to establish the initial condi-
tions. Afterwards 0.066 g (0.300 mmol) of PhIO was added
to the solution under a nitrogen flow. The reaction mixture was
then sampled periodically by withdrawing aliquots in small
portions for GC analysis. Quantification of the product was
done by the internal standard method.
Zn, 8.62%). IR (KBr)/cmϪ1: 1637 (C᎐N) and 1542 (skeletal
᎐
vibration). UV-Vis(MeOH)/nm: 428 (sh, ε = 9000), 366 (ε =
14000). TG analysis: 0.20 mg weight loss (4.27% of 4.68 mg
sample; expected weight loss: 4.22%) at 112 ЊC.
MnCuLЈCl3ؒ3H2O (3). To a methanolic solution (15 mL) of
MnLClؒ2H2O (0.715 g, 1.50 mmol), aniline (0.41 mL, 4.50
mmol) in methanol (10 mL) was added dropwise. The resulting
solution was refluxed for 2 h under a nitrogen atmosphere and
solid CuCl2ؒ2H2O (0.308 g, 1.80 mmol) was added to it. The
mixture was further refluxed for 2 h. Solvent was removed
partially on a rotary evaporator and the product was isolated
by filtration, washed with methanol and dried under vacuum.
Yield: 0.91 g (78%). This crude product was recrystallized twice
from methanol solution by addition of ether (Found: C, 49.14;
H, 4.33; N, 7.16; Mn, 6.83; Cu, 8.05. C32H28N4O2Cl3MnCuؒ
3H2O requires C, 49.29; H, 4.36; N, 7.18; Mn, 7.05; Cu, 8.16%).
The product of epoxidation was identified by 1H NMR spec-
troscopy. Typically, the sample was prepared by the following
procedure.1 Iodosylbenzene (0.066 g, 0.300 mmol) was added to
a solution containing 0.054 g (0.300 mmol) of (E)- or (Z)-
stilbene and 1.00 × 10Ϫ2 mmol of catalyst in dry acetonitrile
(5 mL) under a nitrogen atmosphere. The suspension was
stirred vigorously for 1 h. Then the solvent was removed under
reduced pressure. Anhydrous ether (5 mL) was added to the
brown residue and the ether solution was carefully filtered
through a cannula. Then, ether was removed under reduced
pressure (alternatively, the ether solution was chromatographed
on a silica gel using hexane and ether as an eluent). The residue
IR(KBr)/cmϪ1: 1637 (C᎐N) and 1542 (skeletal vibration).
᎐
UV-Vis(MeOH)/nm: 440 (sh, ε = 6000), 371 (ε = 18000). TG
analysis: 0.31 mg weight loss (7.06% of 4.39 mg sample;
expected weight loss: 6.93%) at 130 ЊC.
The remaining binuclear complexes were prepared by the
same procedure as that for MnCuLЈCl3ؒ3H2O (3). The
analytical data are given below.
1
was dissolved in 0.4 mL CDCl3 and a H NMR spectrum was
measured. The singlet peaks at δ 4.35 and 3.85 are charac-
teristic of the proton resonances of (Z)- and (E)-stilbene oxide
respectively.
X-Ray crystallography
MnNiLЈCl3ؒ3H2O (4). Yield: 79% (Found: C, 49.72; H, 4.43;
N, 7.22; Mn, 6.86; Ni, 7.50. C32H28N4O2Cl3MnNiؒ3H2O
requires C, 49.60; H, 4.39; N, 7.23; Mn, 7.09; Ni, 7.58%).
Single crystals of complex 1ؒ0.39H2O suitable for X-ray diffrac-
tion were grown by slow diffusion of anhydrous ether into an
acetonitrile–acetone (1:1 v/v) mixed solution of 1 at room tem-
perature. Crystal data are summarized in Table 4. Intensity data
were collected at 293 K on a Siemens Smart CCD diffract-
ometer using monochromated Mo-Kα radiation (λ = 0.71073
Å) from a graphite single crystal. A total of 14554 reflections
were collected with 5553 independent reflections (Rint = 4.5%).
The intensities were corrected for Lorentz-polarization
effects and absorption using the SHELXTL-PC36 package
(Tmin, max = 0.8296, 0.9630). The structure was solved by direct
methods. All non-hydrogen atoms were refined anisotropi-
cally by full-matrix least-squares. All hydrogen atoms were
IR(KBr)/cmϪ1: 1631 (C᎐N) and 1543 (skeletal vibration).
᎐
UV-Vis(MeOH)/nm: 390 (sh, ε = 1400), 368 (ε = 19000). TG
analysis: 0.36 mg weight loss (7.07% of 5.09 mg sample;
expected weight loss: 6.97%) at 127 ЊC.
MnCoLЈCl3ؒH2O (5). Yield: 73% (Found: C, 51.92; H,
4.11; N, 7.62; Mn, 7.36; Co, 7.81. C32H28N4O2Cl3MnCoؒH2O
requires C, 52.00; H, 4.06; N, 7.58; Mn, 7.44; Co, 7.98%).
IR(KBr)/cmϪ1: 1627 (C᎐N) and 1554 (skeletal vibration).
᎐
UV-Vis(MeOH)/nm: 395 (sh, ε = 11000), 370 (ε = 20000). TG
J. Chem. Soc., Dalton Trans., 2000, 1081–1086
1085

View Article Online
Table 4 Crystallographic data for complex 1ؒ0.39H2O, Mn()LClؒ
0.3H2O [L = N,NЈ-ethylenebis(3-formyl-5-methylsalicylaldiimine)]
10 H. Fu, G. C. Look, W. Zhang, E. N. Jacobsen and C.-H. Wong,
J. Org. Chem., 1991, 56, 6497.
11 N. Hosoya, R. Irie and T. Katsuki, Synlett., 1992, 261.
12 A. Hatayama, N. Hosoya, R. Irie, Y. Ito and T. Katsuki, Synlett.,
1992, 407.
Empirical formula
Formula weight
C20H18.78ClMnN2O4.39
447.87
13 H. Sasaki, R. Irie and T. Katsuki, Synlett., 1993, 300.
14 N. Hosoya, A. Hatayama, K. Yanai, H. Fuji, R. Irie and T. Katsuki,
Synlett., 1993, 641.
15 J. F. Larrow, E. N. Jacobsen, Y. Gao, Y. Hong, X. Nie and C. M.
Zepp, J. Org. Chem., 1994, 59, 1993.
16 T. Hamada, R. Irie and T. Katsuki, Synlett., 1994, 479.
17 T. Hamada, T. Fukuda, H. Imanishi and T. Katsuki, Tetrahedron,
1996, 52, 515.
18 T. Linker, Angew. Chem., Int. Ed. Engl., 1997, 36, 2060.
19 R. Irie, T. Hashihayata, T. Katsuki, M. Akita and Y. Moro-oka,
Chem. Lett., 1998, 1041.
20 B. G. Malmströn, Annu. Rev. Biochem., 1982, 51, 21.
21 M. Wikströn and G. T. Babcock, Nature (London), 1980, 348, 16.
22 S. Otsuka and T. Yamanaka (Editors), Metalloproteins, Chemical
Properties and Biological Effects, Kodanska, Tokyo, 1988, ch. 5.
23 I. Bertini, H. B. Gray, S. J. Lippard and J. S. Valentine, Bioinorganic
Chemistry, University Science Books, Mill Valley, CA, 1994,
ch. 4 and 5.
24 H. Okawa and S. Kida, Bull. Chem. Soc. Jpn., 1972, 45, 1759.
25 R. R. Gagne, C. L. Spiro, T. J. Smith, C. A. Hamann, W. R. Thies
and A. K. Shiemeke, J. Am. Chem. Soc., 1981, 103, 4073.
26 V. L. Pecoraro and W. M. Butler, Acta Crystallogr., Sect. C, 1986,
42, 1151.
27 L. J. Boucher and V. W. Day, Inorg. Chem., 1977, 16, 1360.
28 M. J. Nappa and C. A. Tolman, Inorg. Chem., 1985, 24, 4711.
29 D. Feichitinger and D. A. Plattner, Angew. Chem., Int. Ed. Engl.,
1997, 35, 1718.
30 R. Irie, T. Hashihayata, T. Katuki, M. Akita and Y. Moro-oka,
Chem. Lett., 1998, 1041.
Space group
a/Å
C2/c
26.0049(2)
b/Å
c/Å
14.6410(9)
13.6091(9)
β/Њ
117.987(2)
4575.4(5)
V/Å3
Z
8
Dc/Mg mϪ3
1.300
0.0721
µcal/mmϪ1
θ/Њ
1.65 to 28.32
Final Ra indices [I > 2σ(I)]
Final R indices [all reflections]
R(F) = 0.0622, Rw(F2) = 0.2032
R(F) = 0.1201, Rw(F2) = 0.2342
2
a R(F) = Σ|Fo Ϫ Fc|/Σ|Fc|; Rw(F2) = [Σw(Fo2 Ϫ Fc2)2/Σw(Fo )2]1/2
.
placed in ideal positions according to the riding model with
isotropic U values in structure factor calculations. All calcu-
lations were performed on a Micro VAX III computer with the
Siemens SHELXTL PLUS (VMS)37 program.
CCDC reference number 186/1835.
See http://www.rsc.org/suppdata/dt/a9/a908304i/ for crystal-
lographic files in .cif format.
Acknowledgements
We thank the National Science Council of Taiwan, ROC, for
financial support (Grant No. NSC88-2811-M-007-0018).
31 P. O Norrby, C. Linde and B. Akermark, J. Am. Chem. Soc., 1995,
117, 11035.
32 A. Earnshaw, Introduction to Magnetochemistry, Academic Press,
New York, 1968.
References
33 L. F. Fieser and M. Fieser, Reagents for Organic Synthesis, Wiley,
New York, 1963, vol. 1, 1378.
1 K. Srinivasn, P. Michaud and J. K. Kochi, J. Am. Chem. Soc., 1986,
108, 2309.
34 H. Saltzman and J. G. Sharefkin, Org. Synth., 1973, Coll. Vol. V, 658.
35 C. K. Johnson, ORTEP, Report ORNL-5138, Oak Ridge National
Laboratory, Oak Ridge, TN, 1976.
36 G. M. Sheldrick, SHELXTL-PC, release 4.4, Siemens Analytical
X-Ray Instruments, Madison, WI, 1992.
37 G. M. Sheldrick, SHELXTL PLUS, Program package for structure
solution and refinement, version 4.2, Siemens Analytical X-Ray
Instruments, Madison, WI, 1990.
2 W. Zhang, J. L. Loebach, S. R. Wilson and E. N. Jacobsen, J. Am.
Chem. Soc., 1990, 112, 2801.
3 D. R. Oki and D. J. Hodgsen, Inorg. Chim. Acta, 1990, 170, 65.
4 C. Bowers and P. K. Dutta, J. Catal., 1990, 122, 271.
5 W. Zhang and E. N. Jacobsen, J. Org. Chem., 1991, 56, 2296.
6 N. H. Lee and E. N. Jacobsen, Tetrahedron Lett., 1991, 32, 6533.
7 R. Irie, Y. Ito and T. Katsuki, Synlett., 1991, 265.
8 R. Irie, K. Noda, Y. Ito, N. Matsumoto and T. Katsuki, Tetrahedron:
Asymmetry, 1991, 2, 481.
Paper a908304i
9 E. N. Jacobsen, W. Zhang and M. L. Guler, J. Am. Chem. Soc., 1991,
113, 6703.
1086
J. Chem. Soc., Dalton Trans., 2000, 1081–1086