Reactions of {TiO(salen)}n [salen = N,N′-bis(salicylidene) ethylenediamine] in aromatic aldehydes and ketones

Article abstract of DOI:10.1016/j.poly.2011.11.013

Polymeric {TiO(salen)}_n prepared from [Ti(OCHMe ₂)₂(acac)₂] and salen-H₂ in MeOH and with an apparent chain structure, is mostly insoluble but dissolves when strongly heated in aromatic aldehydes or ketones, depositing the insoluble form again on cooling. p-Tolualdehyde performed differently, giving a red solution from which insoluble red crystals were deposited. X-ray crystallography showed the complex was [{Ti(salen)(O₂CC₆H₄Me-4)-O- Ti(salen)(O₂CC₆H₄Me-4)]. Solvent (1) (solvent = p-tolualdehyde or p-toluic acid) containing a Ti-O-Ti bridge. Reaction of {TiO(salen)}_n with p-tert-butylbenzoic acid in the presence of excess p-tolualdehyde gave [{Ti(salen)(O₂CC₆H ₄CMe₃-4)-O-Ti(salen)(O₂CC₆H ₄CMe₃-4) ]. 4-CH₃C₆H₄CHO (2) for which an X-ray crystal structure determination indicated similar connectivity around the metal as for (1). {TiO(salen)}_n heated strongly in isobutyrophenone (PhCOCHMe₂) also gave red crystals which X-ray analysis showed structural similarity to (1) and (2) but incorporating a PhCO₂-ligand. {TiO(salen)}_n heated in acetophenone gave a product which X-ray analysis showed was a dioxo-bridged dimer (4) in which C-C bond formation had occurred on one side of each salen ligand.

Full text of DOI:10.1016/j.poly.2011.11.013

Polyhedron 33 (2012) 97–106
Contents lists available at SciVerse ScienceDirect
Polyhedron
journal homepage: www.elsevier.com/locate/poly
Reactions of {TiO(salen)}n [salen = N,N0-bis(salicylidene)ethylenediamine]
in aromatic aldehydes and ketones
b
a
Alastair J. Nielson a, , Shane G. Telfer , Joyce M. Waters
⇑
a Chemistry, Institute of Natural Sciences, Massey University at Albany, Private Bag 102904, North Shore Mail Centre, Auckland, New Zealand
b MacDiarmid Institute for Advanced Materials and Nanotechnology, Institute of Fundamental Sciences, Massey University, Palmerston North, New Zealand
a r t i c l e i n f o
a b s t r a c t
Article history:
Polymeric {TiO(salen)}n prepared from [Ti(OCHMe2)2(acac)2] and salen-H2 in MeOH and with an apparent
chain structure, is mostly insoluble but dissolves when strongly heated in aromatic aldehydes or ketones,
depositing the insoluble form again on cooling. p-Tolualdehyde performed differently, giving a red
solution from which insoluble red crystals were deposited. X-ray crystallography showed the complex
was [{Ti(salen)(O2CC6H4Me-4)-O-Ti(salen)(O2CC6H4Me-4)]. Solvent (1) (solvent = p-tolualdehyde or
p-toluic acid) containing a Ti–O–Ti bridge. Reaction of {TiO(salen)}n with p-tert-butylbenzoic acid in
the presence of excess p-tolualdehyde gave [{Ti(salen)(O2CC6H4CMe3-4)-O-Ti(salen)(O2CC6H4CMe3-4)].
4-CH3C6H4CHO (2) for which an X-ray crystal structure determination indicated similar connectivity
around the metal as for (1). {TiO(salen)}n heated strongly in isobutyrophenone (PhCOCHMe2) also gave
red crystals which X-ray analysis showed structural similarity to (1) and (2) but incorporating a
PhCO2-ligand. {TiO(salen)}n heated in acetophenone gave a product which X-ray analysis showed was
a dioxo-bridged dimer (4) in which C–C bond formation had occurred on one side of each salen ligand.
Ó 2011 Elsevier Ltd. All rights reserved.
Received 9 September 2011
Accepted 9 November 2011
Available online 28 November 2011
Keywords:
Oxotitanium(iv) Schiff base complexes
Reactions in aldehydes and ketones
C–C bond formation
X-ray structures
1. Introduction
an oxotitanium Schiff base complex. An unusual C–C bond
formation occurring at the Schiff base C@N bond carbon is also
Transition metal oxides are important in many heterogeneous
catalytic reactions [1]. In homogeneous catalysis, the Ti–O–Ti
bridging system, which shares a relationship with heterogeneous
systems, has been found to provide a useful framework in com-
plexes that are utilised for some catalytic organic transformations
[2]. Formation of a Ti–O–Ti bridge can be achieved by adding Ag2O
to a complex containing a Ti–Cl bond [2e] but a more convenient
way is to add water to a preformed titanium complex. For example,
in attempting to prepare bridging dioxo complexes of titanium, we
have found that the reaction of the complexes [Ti(OCHMe2)2L]
[L = diamine bis(phenolate)] in ethanol with various amounts of
water led to the complexes [Ti(OEt)L-O-Ti(OEt)L] and no amount
of forcing could cause the remaining ethoxo ligand to hydrolyse
and form a bridging dioxo system [3]. In other work it has been
shown that the Ti–O–Ti system can be generated by exposing iodo
or mono-alkylated Schiff base complexes to atmospheric moisture
[4,5]. [TiCl2Cp-O-TiCl2Cp] can also be formed in this way from
[TiCl3Cp] [6]. However it is known that the use of water to prepare
oxotitanium complexes is in general hard to control [7]. We report
here the results of work (see Scheme 1) which indicates that the
Ti–O–Ti function can be formed when an aromatic acid reacts with
revealed.
2. Experimental
2.1. General
All preparations and manipulations were carried out without
the exclusion of air and moisture. Salen-H2 (N,N0-bis(salicylid-
ene)ethylenediamine) was prepared by refluxing salicylaldehyde
and ethylenediamine in ethanol. {TiO(salen)}n was prepared from
salen-H2 and [Ti(OCHMe2)2(acac)2] in methanol according to a
published procedure [8]. Aldehydes and ketones were distilled in
air prior to use. IR spectra were recorded on a Perkin-Elmer Para-
gon 1000 FT-IR spectrometer (bs = broad strong, m = medium,
s = strong, vw = very weak, w = weak). C, H and N analyses were
determined by Dr. A. Cunninghame and associates, University of
Otago, New Zealand.
2.2. Reaction of {TiO(salen)}n in p-tolualdehyde
{TiO(salen)}n (0.24 g, 0.73 mmole) was suspended in p-tolualde-
hyde (5 ml) and the solution was heated to the boiling point of the
aldehyde until all the complex dissolved. The yellow solution was
allowed to stand whereupon the solution slowly turned red over a
⇑
Corresponding author.
E-mail address: a.j.nielson@massey.ac.nz (A.J. Nielson).
0277-5387/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.poly.2011.11.013

98
A.J. Nielson et al. / Polyhedron 33 (2012) 97–106
Scheme 1. (i) p-Tolualdehyde/air/heat to dissolve, (ii) p-tert-butylbenzoic acid in p-tolualdehyde/air/heat to dissolve, (iii) isobutyrophenone/air/heat to dissolve, (iv)
acetophenone/heat to dissolve.
period of days. Further standing for several weeks led to the
deposition of a small amount of deep red crystals. Found: C,
64.65; H, 5.02; N, 5.28%. An X-ray crystal structure determination
showed that the crystals chosen contain p-tolualdehyde and p-to-
luic acid as solvent of crystallisation in almost equal amounts.
2.4. Reaction of {TiO(salen)}n with isopropylbenzophenone
{TiO(salen)}n (0.366 g, 11.10 mmole) was added to isopropyl-
benzophenone (10.8 g) and the mixture was heated to the boiling
point of the ketone whereupon the complex dissolved and a yellow
solution was formed. On further vigorous heating for 25 min a pale
red solution had formed which was allowed to stand for several
days at room temperature giving a small quantity of orange-red
crystals, one of which was examined by X-ray diffraction.
C56H50N4O10.5Ti2 requires C, 64.56; H, 4.84; N, 5.38%. IR spectrum:
1710m, 1700m, 1633s, 1613s, 1569m, 1555m, 1507w, 1471m,
1450s, 1412m, 1377m, 1345s, 1319m, 1288s, 1255m, 1208m,
1172m, 1152m, 1138m, 1128m, 1111w, 1097m, 1052m1, 1035m,
1023w, 993vw, 966vw, 953vw, 910s, 849m, 808s, 755s, 729bs,
693w, 649s, 613m, 602m, 556m, 512w, 460s cmꢀ1. A single crystal
was examined by X-ray diffraction.
2.5. Reaction of {TiO(salen)}n with acetophenone
{TiO(salen)}n (0.324 g, 0.98 mmole) was added to acetophenone
(5 ml) and the mixture was heated to the boiling point of the ke-
tone until the complex dissolved. The yellow solution was allowed
to stand at room temperature for several days giving a small quan-
tity of yellow crystals, one of which was examined by X-ray
diffraction.
2.3. Reaction of {TiO(salen)}n with p-tert-butylbenzoic acid in p-
tolualdehyde
{TiO(salen)}n (0.372 g, 11.28 mmole) was added to p-tert-butyl-
benzoic acid (0.20 g, 11.22 mmole) and p-tolualdehyde (15 ml)
was added. The mixture was heated to the boiling point of the
aldehyde whereupon the complex dissolved and a red solution
was formed immediately. Further heating gave a dark red solution
which was allowed to stand at room temperature for several days
giving an orange-red microcrystalline solid. Found: C, 66.02; H,
5.28; N, 5.05%. C62H64N4O10Ti2 requires C, 66.49; H,5.67; N,
5.00%. IR spectrum: 1702m, 1688m, 1634s, 1614s, 1598s, 1555m,
1451s, 1410m, 1390m, 1376m, 1344s, 1320m, 1287s, 1256m,
1208m, 1187m, 1173m, 1154m, 1130, 1097, 1054m, 1035w,
1018w, 995w, 968w, 954w, 910s, 856m, 808s, 782m, 754s,
732bs, 709m, 649s, 613m, 582w, 478w, 459s, 410m cmꢀ1. A single
crystal was examined by X-ray diffraction.
2.6. X-ray crystallography
Crystallographic and refinement data are given in Table 1. Data
collected with graphite monchromated Mo K
a X-radiation, were
measured on a Siemens SMART CCD area detector diffractometer
for complexes (1) and (3) and on a Rigaku Spider diffractometer,
using a Rigaku VariMax-HF confocal optical system with mono-
chromated Cu K
a radiation for complexes (2), and (4). Both sys-
tems used the omega scan method. The data were corrected for
Lorentz and polarisation effects and for absorption by multi-scan
methods [9]. The structures were solved by direct methods [10]

A.J. Nielson et al. / Polyhedron 33 (2012) 97–106
99
Table 1
Crystal data for complexes (1)–(4).
(1)
(2)
(3)
(4)
Empirical formula
Mr
T (K)
C
56H50N4O10.5Ti2
C61H60N4O10Ti2
1104.93
150(2)
C56H50N4O10Ti2
1034.80
90(2)
C56H54N4O10Ti2
1038.83
293(2)
1042.80
90(2)
k (Å)
0.71073
0.21 ꢁ 0.18 ꢁ 0.16
orange, fragment
triclinic
1.54178
0.15 ꢁ 0.10 ꢁ 0.04
red, plate
triclinic
0.71073
0.24 ꢁ 0.08 ꢁ 0.04
red, needle
triclinic
1.54178
Crystal size (mm)
Colour, crystal shape
Crystal system
Space group
a (Å)
0.05 ꢁ 0.05 ꢁ 0.01
yellow, plate
triclinic
ꢀ
P1
P1
P1
P1
9.5987(3)
10.7128(3)
13.2935(4)
70.895(2)
73.259(2)
82.900(2)
1236.27(6)
1
9.717(1)
10.768(2)
13.642(2)
106.697(7)
95.149(7)
97.742(7)
1342.3(4)
1
10.058(5)
10.565(5)
12.912(5)
70.285(5)
68.977(5)
82.303(5)
1205.5(10)
1
12.551(3)
12.957(3)
15.742(3)
97.93(3)
94.95(3)
101.33(3)
2469.2(9)
2
b (Å)
c (Å)
a
(°)
b (°)
c
(°)
V (Å3)
Z
Dcalc (mg mmꢀ3
)
1.411
0.391
546
1.367
3.059
578
1.425
0.398
538
1.397
3.289
1084
Absorption coefficient (mmꢀ1
F(000)
)
h range for data collection (°)
Reflections collected
2.01–28.33
27820
6.7–74.7
17400
1.78–27.63
9909
6.69–72.26
45270
Unique reflections (Rint
Reflections with [I > 2r(I)]
)
11146 (0.0490)
7406
7215 (0.0743)
3450
6312 (0.0787)
3040
9300 (0.0551)
6090
Refined parameters
651
645
530
659
Absorption correction Tmax, Tmin
Goodness of fit (all data)
0.9401, 0.9224
1.012
1.00, 0.74
1.071
0.9843, 0.9105
1.017
1.00, 0.73
1.120
Final R indices [I > 2
r
(I)] (R1, wR2)
0.0497, 0.1095
0.0941, 0.1327
0.884
0.0901, 0.2096
0.1704, 0.2762
0.894
0.0751, 0.1605
0.1799, 0.2162
0.880
0.0819, 0.2038
0.1199, 0.2464
0.992
R indices (all data) (R1, wR2)
|E⁄E ꢀ 1|
and refined by full-matrix least-squares methods on F2 [11].
Hydrogen atoms were placed in calculated positions and were re-
fined with a riding model (except for the two H atoms of the water
molecule in compound 4, sites for which were located on a differ-
ence map). Diagrams were prepared by ORTEP [12a] and Fig. 5 by
SHELX [12b]. Selected bond distances and angles are listed in
Tables 2–5.
indicate whether P1 or P1 was to be preferred. Attempts at direct
method solutions for both 2 and 3 assuming the centrosymmetric
ꢀ
ꢀ
space group, P1, produced molecules which were not credible.
However with non-centrosymmetric P1 chemically sensible solu-
tions were obtained. For compound 1 structure solution in both
P1 and P1 was possible and resulted in almost identical results
for the complex moiety. However the presence of a half-weighted
solvent molecule lying across the centre of symmetry in P1 (with
the centre at the mid-point of one of the bonds of the phenyl ring)
was more difficult to unscramble than a fully weighted equivalent
ꢀ
ꢀ
The quality of the crystals for complexes 1–3 was poor and for 2
and 3 the results are only presented to show connectivity and to
compare with complex 1. All four compounds crystallised in tri-
clinic space groups with a value for |E⁄E ꢀ 1| of 0.992 for 4 clearly
ꢀ
solvent in space group P1. Moreover space group P1 required that
ꢀ
indicating the centrosymmetric space group P1. However for the
compounds1, 2, and 3 the intensity statistics did not unequivocally
Table 3
Comparison of selected bond lengths (Å) and angles (°) for (2) and (3).
Table 2
(2)
(3)
Selected bond lengths (Å) and angles (°) for complex (1).
Ti(1)–O(1)
Ti(2)–O(1)
1.84(2)
1.79(2)
1.99(1)
1.86(1)
1.99(1)
2.10(2)
2.15(1)
1.22(2)
1.28(2)
1.36(2)
1.26(2)
Ti(1)–O(1)
Ti(2)–O(1)
1.86(2)
1.77(2)
1.87(1)
1.87(1)
2.02(1)
2.13(2)
2.16(2)
1.28(2)
1.22(3)
1.29(2)
1.23(2)
Ti(1)–O(1)
1.778(8)
1.888(6)
1.876(6)
2.005(5)
2.130(7)
2.129(8)
1.331(9)
1.295(10)
1.303(9)
1.227(9)
Ti(2)–O(1)
1.860(8)
1.851(6)
1.885(6)
2.002(6)
2.143(6)
2.161(7)
1.246(9)
1.280(10)
1.285(10)
1.238(9)
Ti(1)–O(11)
Ti(1)–O(12)
Ti(1)–O(13)
Ti(1)–N(11)
Ti(1)–N(12)
N(11)–C(7)
N(12)–C(10)
C(17)–O(13)
C(17)–O(14)
Ti(2)–O(21)
Ti(2)–O(22)
Ti(2)–O(23)
Ti(2)–N(21)
Ti(2)–N(22)
N(21)–C(37)
N(22)–C(40)
C(47)–O(23)
C(47)–O(24)
Ti(1)–O(11)
Ti(1)–O(12)
Ti(1)–O(13)
Ti(1)–N(11)
Ti(1)–N(12)
N(11)–C(10)
N(12)–C(7)
C(17)–O(13)
C(17)–O(14)
Ti(1)–O(12)
Ti(1)–O(11)
Ti(1)–O(14)
Ti(1)–N(11)
Ti(1)–N(12)
N(11)–C(7)
N(12)–C(10)
C(17)–O(14)
C(17)–O(15)
O(1)–Ti(1)–O(11)
O(1)–Ti(1)–O(12)
O(1)–Ti(1)–O(13)
O(1)–Ti(1)–N(11)
O(1)–Ti(1)–N(12)
Ti(1)–O(11)–C(1)
Ti(1)–O(12)–C(16)
Ti(1)–O(13)–C(17)
O(13)–C(17)–C(18)
O(13)–C(17)–O(14)
C(18)–C(17)–O(14)
Ti(1)–O(1)–Ti(2)
98.5(3)
95.3(3)
169.7(3)
88.0(3)
83.4(3)
136.3(6)
129.0(6)
144.1(5)
113.1(7)
126.3(7)
120.5(7)
179.5(5)
O(1)–Ti(2)–O(21)
O(1)–Ti(2)–O(22)
O(1)–Ti(2)–O(23)
O(1)–Ti(2)–N(21)
O(1)–Ti(2)–N(22)
Ti(2)–O(21)–C(31)
Ti(2)–O(22)–C(46)
Ti(2)–O(23)–C(47)
O(23)–C(47)–C(48)
O(23)–C(47)–O(24)
C(48)–C(47)–O(24)
97.2(3)
94.9(3)
168.4(3)
89.3(3)
84.6(3)
133.0(5)
132.1(5)
148.5(5)
117.9(7)
123.8(8)
118.4(8)
O(1)–Ti(1)–O(11)
O(1)–Ti(1)–O(12)
O(1)–Ti(1)–O(13)
O(1)–Ti(1)–N(11)
O(1)–Ti(1)–N(12)
Ti(1)–O(11)–C(1)
Ti(1)–O(12)–C(16)
Ti(1)–O(13)–C(17)
O(13)–C(17)–C(18)
O(13)–C(17)–O(14)
C(18)–C(17)–O(14)
Ti(1)–O(1)–Ti(2)
94.4(6)
95.0(6)
170.6(6)
87.3(6)
86.5(6)
127(1)
133(1)
146(1)
112(1)
123(2)
126(1)
177(1)
O(1)–Ti(1)–O(12)
O(1)–Ti(1)–O(11)
O(1)–Ti(1)–O(14)
O(1)–Ti(1)–N(11)
O(1)–Ti(1)–N(12)
Ti(1)–O(12)–C(16)
Ti(1)–O(11)–C(1)
Ti(1)–O(14)–C(17)
O(13)–C(17)–C(18)
O(13)–C(17)–O(15)
C(18)–C(17)–O(15)
Ti(1)–O(1)–Ti(2)
93.7(7)
98.0(7)
170.5(8)
88.9(7)
86.1(8)
135(1)
137(1)
149(1)
117(2)
127(2)
117(2)
178(1)

100
A.J. Nielson et al. / Polyhedron 33 (2012) 97–106
Table 4
Selected bond lengths (Å) and angles (°) for complex (4).
Ti(1)–O(11)
Ti(1)–O(21)
Ti(1)–O(12)
Ti(1)–O(13)
Ti(1)–N(11)
Ti(1)–N(12)
C(7)–N(11)
C(22)–N(12)
C(7)–C(8)
1.789(3)
1.914(3)
1.913(3)
1.929(4)
2.263(5)
2.255(4)
1.492(6)
1.275(7)
1.530(7)
1.517(8)
1.493(7)
1.222(7)
C(8)–C(9)
C(9)–C(10)
C(9)–O(14)
Structure 1.
Ti(1)–O(11)–Ti(2)
Ti(1)–O(21)–Ti(2)
O(11)–Ti(1)–O(21)
N(11)–Ti(1)–O(12)
N(11)–Ti(1)–O(13)
N(11)–Ti(1)–O(11)
N(11)–Ti(1)–O(21)
N(11)–Ti(1)–N(12)
O(12)–Ti(1)–O(11)
O(12)–Ti(1)–O(21)
O(12)–Ti(1)–O(13)
O(12)–Ti(1)–N(12)
O(13)–Ti(1)–O(11)
O(13)–Ti(1)–O(21)
N(12)–Ti(1)–O(11)
97.2(2)
98.8(2)
82.5(1)
83.6(2)
157.3(2)
100.3(2)
80.4(2)
74.4(2)
98.2(2)
163.9(2)
95.5(2)
87.2(2)
102.3(2)
100.1(2)
172.0(2)
N(12)–Ti(1)–O(21)
90.7(2)
Ti(1)–O(12)–C(1)
Ti(1)–O(13)–C(16)
Ti(1)–N(11)–C(7)
Ti(1)–N(11)–C(24)
Ti(1)–N(12)–C(22)
Ti(1)–N(12)–C(23)
N(11)–C(24)–C(23)
N(11)–C(7)–C(8)
N(12)–C(23)–C(24)
C(7)–C(8)–C(9)
132.9(4)
137.6(4)
120.1(4)
108.4(3)
126.5(4)
115.4(3)
109.3(5)
108.7(4)
108.6(4)
113.5(5)
117.1(5)
121.2(5)
121.6(5)
113.2(4)
C(8)–C(9)–C(10)
C(8)–C(9)–O(14)
C(10)–C(9)–O(14)
C(7)–N(11)–C(24)
Structure 2.
with the potential to break the chain structure do not appear to re-
act. For example, when {TiO(salen)}n is heated with 4-tert-butyl-
pyridine in an attempt to form the monomeric complex
[TiO(salen)(pyCMe3-4)] (Structure 2), the solid does not dissolve
even when it is heated for long periods in organic solvents or in
the neat ligand. This reflects the known chemistry of titanium
oxo complexes where monomeric species containing a terminal
oxo ligand are very rare and the chemistry of this ligand is invari-
ably bridging [8].
We have noticed that when {TiO(salen)}n is heated in aromatic
aldehydes or ketones the insoluble solid dissolves to give a yellow
solution but on cooling and allowing the solution to stand, the ori-
ginal insoluble non-crystalline material is deposited. However in
some cases when the solutions were allowed to stand for long peri-
ods (i.e. days to several weeks), colour changes took place and a
small amount of crystalline material was deposited. Invariably,
only a few crystals were formed which sometimes precluded nor-
mal characterisation procedures so that only X-ray diffraction
studies could be employed.
Table 5
Bond lengths (Å) and angles (°) for the water
molecule hydrogen bonding network and the
weak intramolecular hydrogen bond in (4).
O(60)ꢂꢂꢂO(11)
O(60)ꢂꢂꢂO(1300)
O(60)ꢂꢂꢂN(2100)
O(23)ꢂꢂꢂN(11)
H(60)ꢂꢂꢂO(11)
H(61)ꢂꢂꢂO(1300)
H(21A)ꢂꢂꢂO(60)
H11AꢂꢂꢂO(23)
O(60)–H(60)ꢂꢂꢂO(11)
O(60)–H(61)ꢂꢂꢂO(1300)
O(60)ꢂꢂꢂH(21A)–N(2100)
H(60)–O(60)–H(61)
N(11)–H(11A)ꢂꢂꢂO(23)
2.761(5)
2.895(6)
2.953(6)
3.048(5)
1.82
2.18
2.06
2.21
175(5)
157(6)
167.0
101(5)
152.8
Heating insoluble {TiO(salen)}n at the boiling point of purified
p-tolualdehyde led to dissolution of the compound, producing a
yellow solution which was no different in colour from that of the
original solid material. However on standing for several days the
solution darkened to a deep red and after several weeks the solu-
tion had deposited a small amount of red crystals. The IR spectrum
of this material showed that the 746 cmꢀ1 band due to the
Ti@OꢂꢂꢂTi@O system in the starting material was no longer present
but there was a new absorption at 729 cmꢀ1 which other studies
indicate arises from an asymmetric Ti–O–Ti stretch [14]. In addi-
tion there were C@O absorptions in the vicinity of 1700 cmꢀ1 indi-
cating aldehyde was present and there was a new band in at
1613 cmꢀ1 which the starting material did not have, suggesting an-
other type of carbonyl group was present. The complex was highly
insoluble in organic solvents which precluded NMR studies but the
structure was obtained by X-ray analysis.
centrosymmetry be forced on the complex moiety. This constraint
was removed if space group P1 were chosen. Some disorder was
also found in the solvent where a mix of p-tolualdehyde and
p-toluic acid appeared to be present. The solvent appeared merely
to have a space filling role since no close approaches to the com-
plex moiety leading to potential H-bonds were observed.
3. Results and discussion
The complex corresponding to the formulation TiO(salen), pre-
pared by the reaction of salen-H2 [salen-H2 = N,N0-bis(salicylid-
ene)ethylenediamine] with [Ti(OCHMe2)2(acac)2] in methanol
solution in the present work, is a highly insoluble material which
has a broad IR absorption at 746 cmꢀ1. This absorbtion is in a sim-
ilar position to that found for the Ti@O stretch in the complex
formed from salen-H2 and [TiO(acac)2] in MeOH for which an
X-ray crystal structure determination has shown a polymeric chain
structure is present [13] (Structure 1). The position of the Ti@O
stretch for the product in the present preparation thus suggests a
similar chain. The product does not dissolve in the usual organic
solvents even when the solutions are strongly heated and ligands
The structure of (1) (Fig. 1, bond lengths and angles in Table 2)
shows that two Ti atoms of the molecule are separated by a single
oxygen atom resulting in a Ti–O–Ti bridging system. At each end of
the bridge and lying trans to the internal oxygen are oxygen atoms
of a p-toluato ligand. The rest of the coordination sphere around
each of the titanium atoms is made up of the oxygen and nitrogen

A.J. Nielson et al. / Polyhedron 33 (2012) 97–106
101
Fig. 1. (a) Molecular structure of complex (1), (b) stepped structure of the salen ligands in (1).
atoms of the two salen ligands. The disposition of these ligands is
such that one salen ligand points inwards and one points outwards
and the two lie in a stepped arrangement (Fig. 1b). A similar
arrangement of these ligands has been observed in the complex
[Ti(salen-1)(O2CCH3)-O-Ti(salen-1)(O2CCH3)] {salen-1 = bis[3,5-di
(tert-butyl)salicyliden]cyclohexane-1,2-diamine} which was pre-
pared from the dioxo bridged complex [{TiO(salen-1)}2] and acetic
anhydride and appears to be the only other complex of this type
that is known [15]. Included in the unit cell of the present complex
is a disordered molecule of solvent which refines best as a mixture
of p-tolualdehyde and p-toluic acid. The IR spectrum of the bulk
sample clearly shows that the aldehyde is present but the presence
of p-toluic acid in the bulk sample of the crystals could not be
determined by IR spectroscopy since absorptions for this molecule
are masked by those of the complex.
The Ti–O–Ti bridging system is slightly asymmetric [Ti(1)–O(1)
and Ti(2)–O(1) bond lengths [1.778(8) and 1.860(8) Å, respec-
tively] similar to that found in [Ti(OEt)L-O-Ti(OEt)L] [L-H2 = N-(2-
pyridylmethyl)-N,N-bis(20-hydroxy-30,50-di-tert-butylbenzyl)amine]
[3] in which a cis-disposition of the Ti–O and Ti–OEt ligands occurs
due to the end-capping ligand L. The Ti–O–Ti bond angle is essen-
tially linear [179.5(5)°] which is also observed for the [Ti(OEt)L-O-
Ti(OEt)L] complex. Ti–O bond lengths and bond angles for a similar
complex containing this type of oxygen bridge are 1.8218(4) Å and
180.0° in [Ti(OiPr)L0-O-Ti(OiPr)L0] [L = amine bis(carboxylato)] [16].
In organometallic examples this angle can vary. Whereas 180° is ob-
served in [Cp TiCl2-O-TiCl2Cp] [17] and [Ti(PhCH2)3-O-Ti(PhCH2)3]
[18] and the angle is essentially linear (173.8°) in [Cp2TiCl-O-TiC-
lCp2] [19], the solid-state structure of [Cp⁄TiCl2-O-TiCl2Cp⁄] shows
three independent molecules with bent Ti–O–Ti bond angles of
159.1°, 154.4°, and 157.7° [19]. In [Cp⁄TiMe2-O-TiMe2Cp⁄] two inde-
pendent molecules show angles of 180° and153.4° [20]. The linear
ligands of this kind normally occur at lower wavelength than
observed here [14a,c]. The Ti–O bond lengths for the ligand in
(1), Ti(1)–O(13) and Ti(2)–O(23), are 2.005(6) and 2.002(6) Å
which are slightly longer than the Ti–O bonds associated with
the salicylaldimine ligands [for example the Ti(1)–O(11) and
Ti(1)–O(12) bond lengths are 1.888(6) and 1.876(6) Å, respec-
tively]. The C@O bond lengths and the bond angles associated with
this functional group for both p-toluato ligands are normal [repre-
sentative are the C(17)–O(14) bond length at 1.227(9) Å; the
O(13)–C(17)–C(18), O(13)–C(17)–O(14) and C(18)–C(17)–O(14)
bond angles at 113.1(7)°, 126.3(7)° and 120.5(7)°, respectively].
The two salen ligands in (1) are orientated in the structure so
that the open sides of the ligands which are made up by the oxygen
atoms are diametrically opposed so that the benzene rings of one
ligand point outwards and inwards in the other. The two ligands
adopt a stepped configuration across the oxo bridge in which one
step points downwards and the other one upwards (Fig. 1b). For
the salen ligand at Ti(1), the step downwards is associated with a
slight opening up of the O(1)–Ti(1)–O(12) angle [95.3(3)°] and a
slight decrease in the O(1)–Ti(1)–N(12) angle [83.4(3)°] before
the benzene ring becomes tilted downwards and on the other side
of the ligand the step upwards is also associated with an opening
up of the O(1)–Ti(1)–O(11) angle [98.5(3)°] and hardly any de-
crease in the O(1)–Ti(1)–N(11) angle [88.0(3)°] before the benzene
ring tilts downwards. This situation is reversed at Ti(2).
For the p-toluato ligand at Ti(1), the O(1)–Ti(1)–O(13) angle is
almost linear [169.7(3)°] and the ligand rotation is such that the
O atom of the C@O bond points backwards and lies neatly above
C(8) which is the carbon atom making the downwards pucker of
the ethylene diamine chelate ring. This carbon lies adjacent to
the upwards step of the salen ligand. As expected, the C@O bond
and benzene ring are co-planar. At Ti(2) a similar situation occurs
except that the C@O oxygen atom which points forwards now lies
neatly below C(38) which is the carbon atom making the upwards
pucker of the ethylene diamine chelate ring. The carbon in this case
is adjacent to the downwards step of the salen ligand.
Ti–O–Ti angle of 180.0° has been suggested to involve some p-dona-
tion from oxygen p orbitals to both metals [21]. The structure of (1)
shows a terminal p-toluato ligand which is unusual in Ti-acetato or
benzoato ligands where bridging is commonly observed [22].
The band observed in the IR spectrum at 1613 cmꢀ1 for (1) is
apparently the asymmetric C@O stretch. Bridging and bidentate
The isolation and characterisation of the bridging oxo complex
(1) gives rise to questions about how the complex was formed.

102
A.J. Nielson et al. / Polyhedron 33 (2012) 97–106
Fig. 2. Molecular structure of complex (2).
In particular, how the Ti–O–Ti bridge was formed in the purified p-
tolualdehyde solution and also how the corresponding acid was
found both as a deprotonated ligand and a protonated acid as a
non-coordinated solvent molecule. The apparent slow change of
colour from the yellow of the soluble form of {TiO(salen)}n to the
red associated with the Ti–O–Ti complex suggested a slow air as-
sisted oxidation might be taking place involving the transforma-
tion of p-tolualdehyde to p-toluic acid. To test this prediction the
initial {TiO(salen)}n compound was reacted with one equivalent
of p-tert-butylbenzoic acid in the presence of excess p-tolualde-
hyde and on heating the solution a red colour rapidly developed
and on standing, orange-red microcrystals were formed.
Structure 3.
The complex was characterised as [Ti(O2CC6H4CMe3-4)(salen)-
O-Ti(O2CC6H4CMe3-4)(salen)]. 4-CH3C6H4CHO (2). The analytical
data was well represented by this formulation which involves
one molecule of p-tolualdehyde solvate. The IR spectrum indicated
a Ti–O–Ti bridge from the broad absorption at 732 cmꢀ1 and there
was a strong C@O absorptions at 1614 cmꢀ1 similar to that found
for complex (1) in which the p-toluato ligand was terminal. In
addition there were peaks in the 1700 cmꢀ1 region typical of the
C@O group of free p-tolualdehyde. An X-ray crystal structure for
(2) was obtained but poor quality crystals precluded a precise
determination. However the structure is presented to indicate
the connectivity (Fig. 2) which shows a similar arrangement of
the salen and acid group ligands about the metal as found for com-
plex (1) and there is a similar formation of a Ti–O–Ti bridge and
one unit of incorporated solvent (not shown in Fig. 2). A selection
of representative bond lengths and angles for the molecule is pre-
sented in Table 3 where it can be seen that within the standard
deviations, the structural features about the two Ti centres are
no different from those of complex (1).
The formation of (2) using one equivalent of the aromatic acid
suggests that in the preparation of complex (1), free acid must have
been generated by air oxidation of the aldehyde. Whether the dis-
solved TiO(salen) compound assisted in this process is unknown
but it is noted that aldehydes can be oxidised to the corresponding
acids when only air and moisture is present [23] and metal cata-
lysts can assist in this regard [23,24]. The mechanism of the forma-
tion of the Ti–O–Ti bridge is not clear but can be postulated as
arising from protonation of the nucleophilic oxo ligand to give
OH and benzoato ligands (Eq. (1)) which could be regarded as an
addition of the acid O–H moiety across the Ti@O double bond.
However the disposition of the 4-substituted benzoate ligand and
the Ti–O–Ti bridge in both (1) and (2) is trans which would involve
an intra-molecular rearrangement over the salen ligand or inter-
molecular rearrangement from an initial cis-disposition of these
groups if an addition across the Ti@O bond had occurred. Hydroxo
ligands are not particularly common in Ti chemistry apart from bis-
cyclopentadienyl chemistry and it can be envisioned to lose H2O
when two such ligand on different metals come together (Eq. (2)
and see Structure 3) [25].
Ti@O þ ArCOOH ! TiðOHÞðO2CArÞ
ð1Þ
ð2Þ
2TiðOHÞðO2CArÞ ! TiðO2CArÞ-O-TiðO2CArÞ þ H2O
ðStructure 3Þ
Related to this type of reaction is the conversion of [Ti(Me)X(salen)]
(X = Cl, Br) type complexes to [TiX(salen)-O-TiX(salen)] when the
former reacts with atmospheric moisture [4].
In general, when {TiO(salen)}n is heated in aromatic ketones
and the complex dissolves, yellow solutions similar to the colour
of the yellow complex are formed and on standing the insoluble
starting compound is returned. However when isobutyrophenone
(PhCOCHMe2) was used and the solution boiled for ca. 25 min
the solution slowly turned red similar to the aromatic aldehyde
reaction described above and on standing at room temperature
for several weeks a small quantity of red crystals was formed.
Insufficient material was produced for normal characterisation
purposes but the complex was identified from an X-ray crystal
structure determination. Poor quality crystals and solvent disorder
precluded a precise determination but the structural connectivity
is presented here (Fig. 3) since the formation of the complex is of
interest. The complex shows similar structural features to complex
(1) with a Ti–O–Ti bridge and salen and benzoato ligands but with
isobutyrophenone as the unit of solvent (not shown in Fig. 3).
A list of representative bond lengths and angles for the complex

A.J. Nielson et al. / Polyhedron 33 (2012) 97–106
103
Fig. 3. Molecular structure of complex (3).
Fig. 4. Molecular structure of complex (4).
(3) is shown in Table 3 where a comparison can be made with
complex (2).
Further evidence for the formation of an enol in a neat solution
of a ketone was obtained when acetophenone (PhCOMe) was em-
ployed. When {TiO(salen)}n was heated in this ketone until the
complex dissolved and then the yellow solution was allowed to
stand for several weeks, a small amount of a highly crystalline
material was formed rather than the more typical amorphous solid
of {TiO(salen)}n. Insufficient material was produced for normal
characterisation but the complex was identified from an X-ray
crystal structure determination.
The complex is a dimer consisting of an asymmetric dioxo
bridge (Fig. 4) instead of the Ti–O–Ti bridging system found for
complexes (1)–(3). The remainder of the coordination sphere about
each metal centre is made up of a modified salen ligand. On one
side of each ligand there is a normal salen type imine bond
(C@N) with the typical coordination to the metal of the oxygen
and nitrogen atoms attached to the aromatic ring. On the other
The reaction leading to (3) is of interest since it would appear from
the previous reactions that benzoic acid must have been formed in the
isobutyrophenone solution under the rather harsh conditions em-
ployed (heating at the boiling point of 217° for 25 min in air). The
isobutyrophenone employed in the reaction was distilled prior to use
and although the commercial preparation procedure is unknown, it
is likely to involve a Friedel-Crafts acylation reaction and not involve
benzoic acid. As to how benzoic acid could be produced in the reaction
is unknown but formation of the enol form of isobutyrophenone fol-
lowed by oxidation could give rise to the acid (Eqs. (3) and (4)).
PhCOCHMe2 ! PhCðOHÞ@CMe2
ð3Þ
ð4Þ
½Oꢃ
PhCðOHÞ@CMe2 ! PhCO2H þ Me2C@O

104
A.J. Nielson et al. / Polyhedron 33 (2012) 97–106
are reacted in DMF [13]. In comparison, the chain structure present
in [{TiO(salen)}2].0.33 MeOH which forms when salen-H2 and
[{TiO(acac)2}2] are reacted in MeOH, are shorter [1.705(2)–
1.706(2) Å] and longer [2.095(2)–2.134(2) Å], respectively [13].
Although there are not a great number of dioxo bridged Schiff base
structures reported for titanium it would appear that there are
variances. For example, in a ‘double stranded’ Schiff base complex
in which two Schiff base ligands are joined across the bridge, the
bond distances are more similar [1.837(3) and 1.852(2) Å] [30].
In [{Ti(salen-2)O}2] (salen-2 = bis[5-(tert-butyl)salicylidene]cyclo-
hexane-1,2-diamine) in which no such bridge exists, the shorter
Ti–O bond lengths are 1.822(2) and 1.843(2) Å and the longer ones
are 1.859(2) and 1.875(2) Å [31]. In a Ti dioxo bridged complex
similar to the present one containing a salen ligand and a salan
type ligand where a PhCH2 group is attached to the carbon of the
C–N bond, the shorter Ti–O bond lengths are 1.778(5) and
1.795(5) Å and the longer ones are 1.923(6) and 1.925(6) Å [28b].
The structure of (4) also includes a water molecule which lies to
the side of bridging oxygen O(11). [O(11)–O(60) separation
2.761(5) Å.] H atoms were identified in the difference map and re-
fined. One H atom of the water molecule [H(60)] (see Fig. 5) makes
a hydrogen bond with O(11) at a distance of 1.82 Å and the other H
atom [H(61)] makes a hydrogen bond with an oxygen of the salen
ligand on the adjacent molecule in the unit cell O(1300) at a distance
of 2.07 Å. In addition, the NH hydrogen on the adjacent molecule
[H(21A)] acts as a weak hydrogen bond donor to the oxygen atom
of the water molecule, the HꢂꢂꢂO separation being 2.06 Å. Within
the unit cell the two molecules of the complex are linked by the
two water molecules making up a network binding the entities to-
gether. Table 5 contains the pertinent bond lengths and angles for
this network. In addition, there is a very weak intramolecular
hydrogen bond made between the salan ligand NH hydrogen asso-
ciated with Ti(1), [H(11A)] and the salen ligand oxygen bound to
Ti(2), [O(23)].
Structure 4.
side there is a salan type bond (C–N) where an addition across the
original C@N system has occurred [C(7) and C(37)]. This modifica-
tion involves the formation of a carbon–carbon bond by attach-
ment of a PhCOCH2 fragment to the carbon and a hydrogen atom
to the nitrogen. Each titanium has a similar coordination geometry.
However at the point of inclusion of the new carbon–carbon bonds,
different stereochemistry is observed. This centre has an (R) config-
uration for the ligand attached to Ti(1) and an (S) configuration for
the ligand at Ti(2). The addition is thus stereospecific for the ligand
around each metal but is racemic with respect to the whole mole-
cule. Enantioselective additions are now a common feature of or-
ganic C@N bond chemistry [26] but in general salen ligands are
unreactive and are commonly chosen as they are stable under a
wide range of conditions [27]. However there are some cases
where additions to the C@N system of coordinated salicylaldimine
ligands occurs [28]. In other types of complexes which contain a
coordinated imine ligand, addition reactions readily occur [29].
Bond lengths and angles for the complex (4) are contained in
Table 4. As the two Ti centres have essentially the same coordina-
tion aspects, only one side of the molecule [Ti(1)] is described. For
the asymmetric dioxo bridge the Ti(1)–O(11) bond length
[1.789(3) Å] is shorter than the Ti(1)–O(21) bond length
[1.914(3) Å]. These bond lengths are similar to those found for
the dioxo bridge in [{TiO(salen)}2].2DMF (Structure 4) [1.804(2)
and 1.907(2) Å] which forms when salen-H2 and [{TiO(acac)2}2]
In complexes (1)–(3), the salen ligands are essentially planar
with trans orientated oxygens but in (4) where the addition across
one of the C@N bonds has occurred, a ‘butterfly’ arrangement is
Fig. 5. Hydrogen bonding network for the water molecule and the weak intramolecular hydrogen bond in complex (4).

A.J. Nielson et al. / Polyhedron 33 (2012) 97–106
105
present where the oxygens are now cis. The salen section of the li-
gand lies below the oxo bridge system and the salan section lies to
one side of it. For the salen section, the Ti(1)–O(13) bond length
[1.929(4) Å] which lies trans to the sp3 nitrogen atom of the salan
ligand section, is slightly longer those observed in complex (1) [Ti–
O bond lengths 1.888(6) and 1.876(6) Å] where both of the oxygen
atoms lie trans to sp2 nitrogens of the C@N linkages. In addition,
the Ti(1)–N(12) bond (C@N nitrogen) in (4) [bond length
2.255(4) Å] is also longer than those observed in complex (1) [Ti–
N bond lengths 2.130(7), 2.129(8) Å]. However the C@N bond
lengths in both complexes are very similar [C(22)–N(12) bond
length in (4), 1.275(7) Å, equivalent bond lengths in (1), 1.331(9)
and 1.295(10) Å].
For the salan section of the ligand in (4), the Ti(1)–O(12) bond in
which the oxygen atom lies trans to the bridging oxygen with the
longer Ti–O bond, has a length [1.913(3) Å] similar to that observed
for the salen section [1.929(4) Å]. There is also no difference in the
Ti–N bond lengths even though one involves an sp3 nitrogen
[Ti(1)–N(11) bond length 2.263(5) Å] which lies trans to the salen
oxygen and the other an sp2 nitrogen [Ti(1)–N(12) bond length
2.255(4) Å] which lies trans to the short Ti–O bond oxygen of the
oxo bridge. The C–C bonds associated with the organic framework
attached at the carbon which was previously the imine carbon
[C(7)] are all normal [C(7)–C(8), C(8)–C(9) and C(9)–C(10) bond
lengths 1.530(7), 1.517(8) and 1.493 Å] as is the C@O bond
[1.222(7) Å].
with the soluble form to produce red oxo-bridged complexes of the
type [Ti(O2CAr)(salen)-O-Ti(O2CAr)(salen)]. We have found that on
standing, solutions of a variety of aromatic and aliphatic aldehydes
in which {TiO(salen)}n has become solubilised, slowly form red
solutions but at present only complexes (1)–(3) have been charac-
terised due to the formation of crystalline solids. The X-ray crystal
structures of these complexes show that the unoxidised aldehyde
can be taken up as solvent of crystallisation in the lattice and the
oxidised aldehyde is taken up in the form of a carboxylato ligand
on the metal. Heating {TiO(salen)}n in ketones also causes it to dis-
solve and in most cases the polymeric {TiO(salen)}n precipitates
out again on standing. However when dissolved in PhCOCHMe2
and allowed to stand [Ti(O2CPh)(salen)-O-Ti(O2CPh)(salen)] was
produced which suggests that the ketone alkyl C–C bond was being
cleaved but this feature cannot be verified at present. Heating
{TiO(salen)}n in PhCOMe led to the addition of a PhCOCH2-group
to the carbon of one of the two C@N bonds available in the salen
ligand apparently by the formation of the enol form of the ketone.
In this case the polymeric form of {TiO(salen)} converts to a dimer
with an asymmetric dioxo bridge.
Acknowledgements
We thank Mr. Graham Freeman for running the IR spectra. We
are grateful to Ms. Tania Groutso of the University of Auckland for
the X-ray data collections for complexes (1) and (3).
The Ti–O–Ti bond angles which make up the asymmetric bridge
in (4) are essentially equivalent at 97.2(2)° and 98.8(2)°, the O(11)–
Ti(1)–O(21) bond angle is 82.5(1)° and the four atoms [Ti(1), O(11),
Ti(2) and O(21)] lie in essentially the same plane and form a trap-
ezoidal structure. Above this set of atoms lies the salen section of
the ligand. Although the oxygen bridge Ti(1)–O(11) bond is shorter
than the Ti(1)–O(21) bond [bond lengths 1.789(3) and 1.914(3) Å,
respectively] the salen ligand oxygen pushes away from both to
the same degree [O(13)–Ti(1)–O(11) and O(13)–Ti–O(21) bond an-
gles 102.3(2)° and 100.1(2)°, respectively]. However this is not the
case for the ethylenediamine ring nitrogen atom which lies below
the oxo bridge system. N(11) pushes back much further from O(11)
[N(11)–Ti(1)–O(11) bond angle 100.3(3)°] [Ti–O(11) involves the
shorter Ti–O bond of the dioxo bridge] than it does for O(21)
[N(11)–Ti(1)–O(21) bond angle 80.4(3)°] which involves the short-
er Ti–O bond of the bridge. This would appear to be due to the
hydrogen attached to N(11) which points more directly towards
O(11) and is therefore closer to it than is O(21).
For the rest of the salan section of the ligand the Ti(1)–O(12)–
C(1) bond angle [132.9(4)°] narrows a small amount in comparison
with the equivalent salen section angle [Ti(1)–O(13)–C(16) bond
angle 137.6(3)°]. About the sp3 nitrogen atom, N(11), the Ti(1)–
N(11)–C(24) bond angle which is contained in the ethylene dia-
mine ring, [108.4(4)°], is normal for this type of nitrogen. The
C(7)–N(11)–C(24) bond angle widens somewhat [113.2(4)°] but
the Ti(1)–N(11)–C(7) bond angle which is contained in the salan
ring, widens out considerably [120.1(4)°]. The bond angles for
the attached PhCOCH2-group at C(7) are all normal [C(8)–C(9)–
C(10), C(8)–C(9)–O(14) and C(10)–C(9)–O(14) bond angles
117.1(5)°, 121.2(5)° and 121.6(5)°, respectively].
Appendix A. Supplementary data
CCDC 842648, 842650, 842089 and 842649 contain the supple-
mentary crystallographic data for complexes (1), (2), (3) and (4),
respectively. These data can be obtained free of charge via http://
www.ccdc.cam.ac.uk/conts/retrieving.html, or from the Cambridge
Crystallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ,
UK; fax: (+44) 1223-336-033; or e-mail: deposit@ccdc.cam.ac.uk.
References
[1] (a) C.N.R. Rao, B. Raveau, Transition Metal Oxides, VCH, Weinheim, 1995;
(b) J.M. Thomas, W.J. Thomas, Principles and Practice of Heterogeneous
Catalysis, VCH, Weinheim, 1997;
(c) G. Ertl, H. Knözinger, J. Weitkamp, Handbook of Heterogeneous Catalysis,
VCH, Weinheim, 1997.
[2] (a) T. Hashimoto, M. Omote, K. Maruoka, Org. Biomol. Chem. 6 (2008) 2263;
(b) T. Hashimoto, M. Omote, Y. Hato, T. Kano, K. Maruoka, Chem. Asian J. 3
(2008) 407;
(c) T. Hashimoto, M. Omote, T. Kano, K. Maruoka, Org. Lett. 9 (2007) 4805;
(d) T. Kano, T. Hashimoto, K. Maruoka, J. Am. Chem. Soc. 127 (2005) 11926;
(e) M. Hanawa, D. Uraguchi, S. Konishi, T. Hashimoto, K. Maruoka, Chem. Eur. J.
9 (2003) 4405;
(f) M. Hanawa, T. Hashimoto, K. Maruoka, J. Am. Chem. Soc. 125 (2003) 1708.
[3] A.J. Nielson, J.M. Waters, Polyhedron 29 (2010) 1715.
[4] S.J. Coles, M.B. Hursthouse, D.G. Kelly, A.J. Toner, N.M. Walker, J. Chem. Soc.,
Dalton Trans. (1998) 3489.
[5] S.J. Coles, M.B. Hursthouse, D.G. Kelly, A.J. Toner, N.M. Walker, J. Organomet.
Chem. 580 (1999) 304.
[6] (a) P. Corradini, G. Allegra, J. Am. Chem. Soc. 81 (1959) 5510;
(b) R.D. Gorsich, J. Am. Chem. Soc. 82 (1960) 4211.
[7] L.M. Babcock, W.G. Klemperer, Inorg. Chem. 28 (1989) 2003.
[8] N.F. Choudhary, P.B. Hitchcock, G.J. Leigh, Inorg. Chim. Acta 306 (2000) 24.
[9] (a) R.H. Blessing, Acta Crystallogr., Sect. A 51 (1995) 33;
(b) T. Higashi, ABSCOR, Rigaku Corporation, Tokyo, Japan, 1995.
[10] G.M. Sheldrick, Acta Crystallogr., Sect. A 46 (1990) 467.
[11] G.M. Sheldrick, Program for the Refinement of Crystal Structures, University of
Göttingen, Germany, 1997.
4. Conclusion
[12] (a) L.J. Farrugia, ORTEP-3 for Windows, Version 1.06, University of Glasgow,
Scotland, UK 1997.;
This work shows that {TiO(salen)}n which has a polymeric chain
structure, will react in neat solutions of aromatic aldehydes and
ketones. When heated in the aldehydes, the polymer dissolves,
possibly forming a monomeric species, the structure of which re-
mains unknown at this stage. On standing in air, it is apparent that
oxidation of the aldehyde takes place and the resulting acid reacts
(b) G.M. Sheldrick, SHELXL 93, Institut Für Anorganische Chemie, Universität
Göttingen, 1993.
[13] M. Tsuchimoto, Bull. Chem. Soc. Jpn. 74 (2001) 2101.
[14] (a) K. Nakamoto, Infrared and Raman Spectra of Inorganic and Coordination
Compounds, Part B/V, Wiley, New York, 1997. p. 58, 172;
(b) P. Piszczek, M. Richert, A. Wojtczak, Polyhedron 27 (2008) 602;

106
A.J. Nielson et al. / Polyhedron 33 (2012) 97–106
(c) P. Piszczek, M. Richert, A. Grodzicki, A. Wojtczak, Polyhedron 24 (2005)
663.
(b) H. Barrow, D.A. Brown, N.W. Alcock, H.J. Clase, M.G.H. Wallbridge, J. Chem.
Soc., Dalton Trans. (1994) 195.
[15] Y.N. Belokon, P. Cartac, A.V. Gutnova, V. Maleeva, M.A. Moskalenkoa, L.V.
Yashkinaa, N.S. Ikonnikova, N.V. Voskoboevb, V.N. Khrustaleva, M. North, Helv.
Chim. Acta 85 (2002) 3301.
[23] J.R. McNesby, C.A. Heller, Chem. Rev. 54 (1954) 325.
[24] D.R. Larkin, J. Org. Chem. 55 (1990) 1563.
[25] H.W. Roesky, Solid State Sci. 3 (2001) 777.
[16] M. Shavit, E.Y. Tshuva, Eur. J. Inorg. Chem. (2008) 1467.
[17] U. Thewalt, D. Schomburg, J. Organomet. Chem. 127 (1977) 169.
[18] H. Stoeckli-Evans, Helv. Chim. Acta 57 (1974) 684.
[19] Y. Le Page, J.D. McCown, B.K. Hunter, R.D. Heyding, J. Organomet. Chem. 193
(1980) 201.
[20] S.P. Varkey, M. Schormann, T. Pape, H.W. Roesky, M. Noltemeyer, R. Herbst-
Irmer, H.-G. Schmidt, Inorg. Chem. 40 (2001) 2427.
[21] (a) F. Caruso, M. Rossi, C. Opazo, C. Pettinari, Bioinorg. Chem. Appl. 3 (2005)
317;
[26] S. Kobayashi, Y. Mori, J.S. Fossey, M.M. Salter, Chem. Rev. 111 (2011) 2626.
[27] D.A. Atwood, M.J. Harvey, Chem. Rev. 101 (2001) 37.
[28] (a) F. Franceschi, E. Solari, C. Floriani, M. Rosi, A. Chiesi-Villa, C. Rizzoli, Chem.
Eur. J. 5 (1999) 708;
(b) P.R. Woodman, N.W. Alcock, I.J. Munslow, C.J. Sanders, P. Scott, J. Chem.
Soc., Dalton Trans. (2000) 3340;
(c) Y. Wang, S. Parkin, D. Atwood, Inorg. Chem. 41 (2002) 558;
(d) D. Qin, F. Han, Y. Yao, Y. Zhang, Q. Shen, Dalton Trans. (2009) 5535.
[29] See for example: G.C. Martin, J.M. Boncella, E.J. Wucherer, Organometallics 10
(1991) 2804.
(b) F. Bottomley, S.-K. Goh, Polyhedron 15 (1996) 3045;
(c) F. Caruso, C. Pettinari, F. Marchetti, P. Natanti, C. Phillips, J. Tanski, M. Rossi,
Inorg. Chem. 46 (2007) 7553.
[30] M.L.N. Rao, H. Houjou, K. Hiritani, Chem. Commun. (2002) 420.
[31] Amber Davis, Colin A. Kilner, Terence P. Kee, Inorg. Chim. Acta 357 (2004)
3493.
[22] (a) See for example: T. Frot, S. Cochet, G. Laurent, C. Sassoye, M. Popall, C.
Sanchez, L. Rozes, Eur. J. Inorg. Chem. (2010) 5650. and refs. therein;