Halogenated oxo- and peroxotitanium porphyrinates as sensitizers for the photooxygenation of olefinic compounds

Article abstract of DOI:10.1016/S1381-1169(98)00213-1

A series of porphines, oxotitanium(IV) and peroxotitanium(IV) porphyrinates were tested as photosensitizers of the singlet oxygen ene reaction with cyclohexene and cis-cyclooctene. The relative order of activity is: H₂(P) ≥ O=Ti(P) ≥ Ti(O₂)P. Best stability is obtained for porphines or porphyrinate complexes with fluorine substituents. Oxotitanium(IV) porphyrinates are active in the decomposition of allylic hydroperoxides. The (photo)catalytic epoxidation occurs less readily than the thermal reaction at 82°C. Attempts of photocatalytic hydrogen peroxide activations lead to fast destruction of O=Ti(P).

Full text of DOI:10.1016/S1381-1169(98)00213-1

Ž
.
Journal of Molecular Catalysis A: Chemical 140 1999 13–24
Halogenated oxo- and peroxotitanium porphyrinates as
sensitizers for the photooxygenation of olefinic compounds
a,)
Peter E. Esser b,1, Birgit Drießen-Holscher a,2, Wilhelm Keim
¨
a
Institut fur Technische Chemie und Petrolchemie, RWTH Aachen, Worringerweg 1, 52074 Aachen, Germany
¨
b
Huls Infracor, Verfahrenstechnik, 45764 Marl, Germany
¨
Received 18 February 1998; accepted 8 June 1998
Abstract
Ž
.
Ž
.
A series of porphines, oxotitanium IV and peroxotitanium IV porphyrinates were tested as photosensitizers of the
Ž . Ž .
singlet oxygen ene reaction with cyclohexene and cis-cyclooctene. The relative order of activity is: H2 P )O5Ti P G
Ti O2 P. Best stability is obtained for porphines or porphyrinate complexes with fluorine substituents. Oxotitanium IV
Ž
.
Ž
.
Ž
.
porphyrinates are active in the decomposition of allylic hydroperoxides. The photo catalytic epoxidation occurs less readily
than the thermal reaction at 828C. Attempts of photocatalytic hydrogen peroxide activations lead to fast destruction of
Ž .
O5Ti P . q 1999 Elsevier Science B.V. All rights reserved.
Ž
.
Ž
.
Keywords: Cycloolefins; Photooxygenation; Oxotitanium IV porphyrinates; Peroxotitanium IV porphyrinates; Photosensitizer; Singlet
oxygen; Ene reaction
1. Introduction
for new selective activation pathways using
molecular oxygen or hydrogen peroxide is of
w
x
The catalytic oxidation of hydrocarbons de-
rived from petrochemical feedstocks in the liq-
uid or gas phase belongs to the most important
methods in the chemical industry. The search
current interest 1–3 .
Photooxygenation reactions usually proceed
very selectively at ambient temperatures using
only light, oxygen and a sensitizer. Singlet oxy-
gen is often the active species and a broad
variety of compounds can be obtained starting
Ž
.
from hetero aromatics, alkenes or alka-1,3-di-
Abbreviations: OEP: octaethylporphinato; TMP: meso-tetra-
w
x
enes 4–6 .
Ž
.
Ž
kis 2,4,6-trimethylphenyl porphinato; TTP: meso-tetrakis 4-
Longer reaction times cause oxidative degra-
dation of the used sensitizer dyes, such as rose
bengale, methylene blue or H2TPP. Quast et al.
have shown the bleaching of meso-tetraphenyl-
porphines to be dependent of the substituents of
the macrocycle as well as the nature and con-
centration of the alkene applied. Especially dur-
ing cyclohexene photooxygenation, complete
.
methylphenyl porphinato; TPP: meso-tetraphenylporphinato; TP-
Ž
.
PCl4: meso-tetrakis 4-chlorophenyl porphinato; TPPCl8: meso-te-
trakis 2,6-dichlorophenyl porphinato; TPPF8: meso-tetrakis 2,6-
difluorophenyl porphinato; TPPF20 : meso-tetrakis 2,3,4,5,6-penta-
Ž
.
Ž
.
Ž
.
Ž .
fluorophenyl porphinato; P : porphinato
)
Corresponding author. Fax: q49-241-8888-177; E-mail:
keim@itc.itc.rwth-aachen.de
1
Ž .
Telefax: int.q 0 2365-495589; E-mail:
p.esser@huels-infracor.de
2 E-mail: Drihoel@itc.itc.rwth-aachen.de
1381-1169r99r$ - see front matter q 1999 Elsevier Science B.V. All rights reserved.
Ž
.
PII: S1381-1169 98 00213-1

(
)
14
P.E. Esser et al.rJournal of Molecular Catalysis A: Chemical 140 1999 13–24
Ž
.
destruction of H2TPP and H2TPPF20 is ob-
served. Under the same conditions, more than
30% of the initial H2TPPCl8 amount remained
dried with calciumhydride Aldrich , distilled
under argon and stored in the dark. cis-
Cyclooctene was used with a purity of 95.5%.
Cyclohexen-3-ol and cycloocten-3-ol were pre-
pared by photooxygenation of cyclohexene or
cis-cyclooctene and subsequent reduction with
triphenyl phosphine or by reduction of cyclo-
w x
7 . According to DiMagno et al., H2TPPF20 is
bleached only slightly during photooxygenation
of allylic alcohols. Cyclohexene as substrate
causes comparable decomposition of H2TPPF20
w x
Ž .
and H2TPP 8 .
hexen-3-one Fluka with lithium aluminiumhy-
Ž
.
The physical separation of the sensitizer from
the reaction products during the irradiation pro-
cedure using perfluorinated biphasic solvent
dride Merck . 2,3-Epoxycyclohexanol was syn-
thesized from cyclohexen-3-ol and m-chloroper-
Ž
.
benzoic acid Merck, 55% . cis-Cyclooctene-
oxide, trans-cyclooctane-1,2-diol and BF3 P
Et2O were purchased from Fluka.
w
x
systems 9–11 leads to an enhanced chemical
w
x
stability of the sensitizer 8,12 .
Ž
.
When transition metal porphyrinates with iron
13–20 and manganese 21–23 , osmium 24 ,
Titanium tetrachloride Fluka was distilled
from copper chloride under argon before use.
Hydrogen peroxide 30%, Merck , t-butyl-
w
x
w
x
w x
w
x
w
x
Ž .
niobium 25,26 or molybdenum 27,28 as cen-
tral atom are used as photocatalysts, they can
cause oxygenation via a radical mechanism.
Some oxovanadium and oxotitanium porphyri-
nates are reported to be singlet oxygen sensitiz-
Ž
.
.
hydroperoxide 3 molrl in i-octane, Fluka ,
Ž
2,3,5,6-tetrachlorobenzoquinone PCA Fluka ,
2,3-dichloro-5,6-dicyanobenzoquinone DDQ
Fluka , titanocene dichloride Aldrich and tita-
nium acetylacetonate Merck were used as pur-
Ž
.
Ž
.
w
x
Ž
.
ers 29,30 .
To our knowledge, no systematic investiga-
chased.
tion concerning the influence of solvents or
The NMR spectra were recorded with a
Bruker-AC300 spectrometer and the UV–Vis
spectra with the Perkin-Elmer Lambda 14 spec-
trometer using toluene or dichloromethane as
solvent. The IR spectra were measured with a
Nicolet 510rP spectrometer in a KBr matrix.
The Finnigan MAT 95 served for the detection
of the SIMS spectra.
Ž
.
halogen substituents on the macrocyclus re-
garding the stability of oxotitanium porphyri-
nates and the selectivity during the photooxy-
genation of alkenes has been published.
We reported recently the syntheses of various
new oxo- and peroxotitanium porphyrinates with
perfluorinated or perchlorinated phenyl groups
Ž .
w
x
and the structure of Ti O TPPF20 31,32 . This
paper shows the photooxygenation of cyclohex-
ene and cis-cyclooctene with a series of tita-
nium porphyrinate complexes as sensitizers. The
influence of substituents and solvents on the
porphyrinate stability and the selectivity of oxy-
genation will be shown in comparison to metal-
free porphines.
2.2. Synthesis of porphyrinates
The free porphines H2TPP, H2TMP, H2TTP,
Ž
.
H2TPPCl4 and H2TPP OMe
were prepared
12
according to the conditions of the Rothemund
w
x
synthesis 33–38 . H2OEP was provided by
Midcentury. The porphine H2TPPF20 was syn-
thesized from 2,3,4,5,6-pentafluorobenzalde-
hyde and pyrrole in the presence of boron triflu-
oride–diethyl ether and subsequent oxidation
with p-chloroanile or 2,3-dichloro-5,6-di-
cyanobenzoquinone under equilibrium condi-
2. Experimental
2.1. General
w x
All solvents were dried, purified and stored
under inert argon atmosphere. Cyclohexene was
tions 39 . By a similar procedure, H2TPPF8
w
x
w x
40 and H2TPPCI8 41 can be obtained.

(
)
P.E. Esser et al.rJournal of Molecular Catalysis A: Chemical 140 1999 13–24
15
The reaction of the porphines with titanium
tetrachloride in dry toluene under argon atmo-
sphere and subsequent hydrolysis with dilute
mineral acid yields oxotitanium porphyrinates
under argon. After 20 or 60 h, the reaction
mixture was cooled and analyzed by GC and
TLC.
w
x
31,32,42 . After chromatography on silica, the
oxotitanium porphyrinates were reacted with an
excess of aqueous hydrogen peroxide in
dichloromethane. The light sensitive peroxotita-
3. Results and discussion
3.1. Syntheses of titanium porphyrinates
Ž
.
nium IV porphyrinates were obtained after
w
x
Ž
.
chromatography 31,32,43 .
The syntheses of the oxotitanium IV por-
phyrinates starts from the free porphines and
titanium tetrachloride in toluene according to
Fig. 1. Derivatisation with aqueous hydrogen
peroxide of the pure compound dissolved in
dichloromethane yields the light-sensitive per-
2.3. Photochemical procedures
All photoreactions were carried out in a 45-ml
Pyrexe photoreactor 3 cm in diameter
equipped with a magnetic stirrer, a gas supply
and a heat exchanger in the reaction mixture.
The with water cooled light source was placed
outside of the reactor at a distance of 5 cm.
Ž
.
Ž
.
w
x
oxotitanium IV porphyrinates 31,32,42,43 .
Good yields, ranging from 80% to 92% in both
reaction steps, can be reached by this method.
Table 1 and Fig. 2 respectively show the abbre-
viations and the structures of the synthesized
complexes used as catalysts for the activation of
oxygen, t-butylhydroperoxide or hydrogen per-
oxide.
Ž
Different high-pressure mercury lamps TQ 150,
TQ 150 Z1 or TQ 150 Z2, Hereus Noble-light
.
were used. When hydrogen peroxide or t-
butylhydroperoxide were used as oxidant, the
oxygen stream was replaced by an argon sup-
ply. Excess hydrogen peroxide and water were
removed after the irradiation by adding small
amounts of manganese dioxide and molecular
sieve as a drying agent.
3.2. Photooxygenation reactions with dioxygen
The catalytic activity of the synthesized tita-
Ž
.
nium IV porphyrinates was evaluated during
the photooxygenation of cyclohexene 10a and
cis-cyclooctene 10b with dioxygen as oxidant.
The observed products are shown in the Fig. 3.
Ž .
The mixture of the olefin, the solvent s , the
catalyst and n-decane as internal standard was
stirred for half an hour in the dark, while oxy-
gen was bubbled through the mixture with a
constant stream of 10 lrh. Before and during
the irradiation, samples from the reactor were
withdrawn and analysed by gas chromatography
and thin layer chromatography. At the end of
the irradiation, the reaction mixture was reduced
by the addition of triphenyl phosphine or lithium
aluminium hydride in anhydrous diethylether
and was also gas chromatographically analysed.
3.2.1. Influence of the titanyl group and the
substituents at the porphine core
The photooxygenation of cyclohexene and
cis-cyclooctene using different tetraarylpor-
2.4. Thermal procedures
In a Schlenk tube, the weighted amounts of
the olefin, the catalyst, n-decane and t-butyl-
hydroperoxide were heated in 4 ml of toluene
Ž
.
Ž
.
Ž .
Fig. 1. Syntheses of per oxotitanium IV porphyrinates: H2 P -
free porphyrine.

(
)
16
P.E. Esser et al.rJournal of Molecular Catalysis A: Chemical 140 1999 13–24
Table 1
Ž
.
Titanium IV porphyrinates and abbreviations
Ž .
M: Ti O2
R1
R2
M: Ti5O
Ž
.
.
.
.
.
.
.
.
Ž .
Ti O2 TPP 1c
Ž .
Ti O2 TPPCl8 2c
Ž .
Ti O2 TMP 3c
Ž .
Ti O2 TPPF20 4c
H
H
H
H
H
H
H
Et
H
Phenyl 1a
Ti O TPP 1b
Ž
2,6-Dichlorophenyl 2a
2,4,6-Trimethylphenyl 3a
2,3,4,5,6-Pentafluorophenyl 4a
4-Methylphenyl 5a
4-Chlorophenyl 6a
2,6-Difluorophenyl 7a
H 8a
Ti O TPPCl8 2b
Ž
Ti O TMP 3b
Ž
Ti O TPPF20 4b
Ž
Ti O TTP 5b
Ž
Ž .
Ti O2 TPPCl4 6c
Ž .
Ti O2 TPPF8 7c
Ti O TPPCl4 6b
Ž
Ti O TPPF8 7b
Ž
Ž
.
Ti O OEP 8b
. Ž
Ti O2 OEP 8c
Ž
Ž
. .
Ž
Ž
.
.
H2T 3,4,5- OMe PP 9a
Ti O T 3,4,5- OMe 3P P 9b
3
phines as photosensitizers results in the selec-
tetraarylporphines. Small amounts of the corre-
Ž
.
Ž .
tive formation )94% of the corresponding
sponding peroxotitanium IV –tetraarylporphyri-
Ž
.
Ž
.
allylic hydroperoxides 11a and 11b Table 2 .
Nearly no by-product formation occurs. Cyclo-
hexene and cis-cyclooctene show similar reac-
tivities. With the partial fluorinated porphine
H2TPPF8 7a, the best results with respect to
stability and hydroperoxide yield are obtained.
The porphines H2TPP 1a, H2TMP 3a and
H2TPPCl8 2a, which were reported to be supe-
nates are formed during the irradiation TLC .
The allylic hydroperoxides 11a and 11b are the
main products with selectivities from 38 to 77%.
Major side products are the allylic alcohol
12ar12b, the a,ß-unsaturated ketone 13ar13b
and the epoxide 14ar14b. In a few cases, the
1,2-epoxycyclohexan-3-ol 15a could be detected
in small amounts.
w x
rior sensitizers 7 , exhibit lower, but still good
The best results of stability and activity can
Ž
.
yields. H2TPP 1a is not stable in the presence of
cylohexenyl-hydroperoxide 11a. After 6 h of
irradiation, nearly total porphine bleaching was
observed. The low hydroperoxide yields ob-
tained with the porphine H2TPPF20 4a are sur-
prising because the quantum yield should in-
crease with the fluorine and especially with the
be obtained with oxotitanium IV tetraarylpor-
phyrinates disubstituted in the 2,6 position of
the aryl group. Fluorine and chlorine sub-
stituents are more suitable than methyl groups.
Ž .
Especially, Ti O TPPF20 4b catalyses the de-
composition of the hydroperoxides and exhibits
further enhanced epoxidation activity. The in-
troduction of the titanyl group enhances the
catalytic performance of H2TPPF20 4a signifi-
cantly. The para- or unsubstituted oxotita-
w
x
chlorine content of the porphine core 44 . Ac-
cording to product yield and stability, the fol-
lowing order of sensitizer quality was found:
7a))3a;2a))4a;1a.
Ž
.
nium IV tetraarylporphyrinates are completely
destroyed during the irradiation. The following
order of sensitizer quality was found: 7b;4b)
2b)3b)1b)5b)6b)8b;9b.
Ž
.
Ž
Oxotitanium IV porphyrinates Tables 3 and
.
4 exhibit lower yields than the corresponding
Ž
.
Ž
.
Fig. 2. Structure of per oxotitanium IV porphyrinates.
Fig. 3. Product spectra of the photooxygenation of cyclic olefins.

(
)
P.E. Esser et al.rJournal of Molecular Catalysis A: Chemical 140 1999 13–24
17
Table 2
Photooxygenation of cyclohexene 10a and cis-cyclooctene 10b with different tetraarylporphyrines
a Porphine was destroyed.
TQ 150 Z2, 40 ml CH2Cl2.
3.2.2. Influence of the catalyst concentration
The amount of the hydroperoxide decomposi-
tion during the irradiation depends on the aryl
substituents and the concentration of the used
oxotitanium porphyrinates. The influence of the
concentration on the reaction was investigated
Ž .
Ž
.
in the case of Ti O TPP 1b as catalyst Fig. 4 .
With increasing catalyst concentration, signifi-
Table 3
Photooxygenation of cyclohexene 10a with different oxotitanium porphyrinates
a 0.020 mmol.
0.037 mmol Catalyst, TQ 150 Z2, 6.1 h, 40 ml CH2Cl2.

(
)
18
P.E. Esser et al.rJournal of Molecular Catalysis A: Chemical 140 1999 13–24
Table 4
Photooxygenation of cis-cyclooctene 10b with different oxotitanium porphyrinates
0.036 mmol Catalyst, TQ 150 Z2, 40 ml CH2Cl2.
cant amounts of the allylic alcohol 12a could be
detected. In the dark, no further decomposition
occurred at room temperature. This is in agree-
oxygen yields mainly cyclohexenone. The
Ž
turnover frequency is less than 0.4 molr mol=
.
h . No formation of hydroperoxides was re-
w
x
ment with published results 45 . The overall
yields increased only slightly with the amount
of catalyst and varied between 15 and 17% after
an irradiation time of 4 h. These results indicate
a photochemically induced hydroperoxide de-
composition.
ported.
During the course of this investigation, the
formation of peroxotitanium porphyrinates could
be observed when cyclohexene and cis-
cyclooctene were photooxygenated with oxotita-
Ž
.
nium IV porphyrinates as catalysts. The perox-
Ž
.
Ž
.
otitanium IV porphyrinates Ti O2 TPPCl8 2c,
( )
Ž
.
Ž
.
Ž
.
3.2.3. Peroxotitanium IV porphyrinates as cat-
Ti O2 TMP 3c, Ti O2 TPPF20 4c and Ti O2 -
TPPF8 7c were therefore tested as catalysts as
well. The catalytic performance is, in general,
lower in comparison to the corresponding oxoti-
tanium porphyrinates. The selectivity to the hy-
droperoxide ranges from 80 to 85%. During the
irradiation, most of the peroxo complex is con-
alysts
The peroxotitanium porphyrinates are known
w
x
to be photosensitive 31,32 . They release sin-
Ž
.
w
x
glet dioxygen during irradiation Fig. 5 46,47 .
The irradiation of cyclohexene in benzene or
Ž
.
ethyl acetate in the presence of Ti O2 TPP and
Ž
.
.
Fig. 4. Photooxygenation of 19.3 mmol 10a with different concentrations of 1b 4 h, TQ 150 Z2, 40 ml CH2Cl2

(
)
P.E. Esser et al.rJournal of Molecular Catalysis A: Chemical 140 1999 13–24
19
coordination at the oxotitanium moiety could be
Ž
.
the reason Fig. 9 . In ethanolic solution, for
Ž .
example, an equilibium between Ti O TPP 1b
and EtOH Ti O TPP exists, whereas no coordi-
nation occurs in chloroform or toluene 48 .
Mono- or disubstituted amines show only a
weak coordination and triethylamine or DMSO
Ž .
Fig. 5. Photodeoxygenation of peroxotitanium porphyrinates P s
Ž
.
Ž .
Ž
.
TPP, T m-T P.
w
x
w x
verted to the corresponding oxo complex. The
stabilities of the peroxo catalysts are nearly as
Ž
.
high as the oxo complexes Fig. 6 . In the case
of cis-cyclooctene 10b as substrate, only small
differences in yield and selectivity were mea-
molybdenum is well-documented 21–23,50 .
Ž .
Because the catalysts Ti O TMP 3b and
Ž
.
sured Table 5 .
Ti O TPPF20 4b are more sensitive to a change
in the reaction medium, a detailed investigation
of the influence of additives on their activity
and selectivity was undertaken. The results are
summarized in Fig. 8. Addition of imidazole
increases the product yield and the selectivity of
hydroperoxide formation. The irradiation of
3.2.4. Variation of the solÕents and influence of
additiÕes
It became obvious during the course of these
investigations that the catalyst activities differ
with the solvent used. Best results could be
obtained in benzene or toluene, whereas in
dichloromethane, the yield was reduced by al-
most 50%. A faster catalyst decomposition oc-
curred in dichloromethane. The solvent, there-
fore, significantly influences the course of the
reaction.
Ž .
Ti O TMP 3b with imidazole causes an addi-
tional absorption in the UV–Vis spectrum at
Ž
.
433 nm log es5.1 . As in the case of aceto-
nitrile, axial coordination may be the reason for
Ž
.
this Fig. 9 .
DABCO quenches the reaction effectively.
The addition of titanium complexes reduces the
product formation and the hydroperoxide forma-
Fig. 7 shows some results using pure
dichloromethane and acetonitrile as cosolvent
Ž
.
50% . Although both solvents show similar
singlet oxygen lifetimes, the turnover numbers
tion drastically. This may be the consequence of
5
Ž
.
radical formation, e.g., C5H5 and Ti h -Cp Cl2,
Ž
.
increase 7–25% by adding acetonitrile. The
selectivity to the hydroperoxide 11a increases
from about 60% to 70–80% as well. An en-
hanced catalyst stability because of the axial
which can react with chlorinated solvents to
5
Ž
.
w
x
yield Ti h -Cp Cl3 and further radicals 51,52 .
In the presence of excess TBHP, the activity of
Ž .
Ti O TPPF20 4b is significantly lower and the
Ž
.
Fig. 6. Photooxygenation of cyclohexene 10a 20.1"0.4 mmol . Comparison between the activity of oxo- and peroxotitanium porphyri-
Ž
.
nates 8 h, 40 ml dichloromethane, TQ 150 Z2, 2br2c and 4br4c: 0.019 mol; 3br3c and 7br7c: 0.036 mmol .

(
)
20
P.E. Esser et al.rJournal of Molecular Catalysis A: Chemical 140 1999 13–24
Table 5
Ž
.
Photooxygenation of 22.5 mmol cis-cyclooctene 10b using peroxo complexes Ti O 2 P as catalysts
0.019 mmol Catalyst, TQ 150 Z2, 8.1 h, 40 ml CH2Cl2 , 168C.
hydroperoxide 11a is formed selectively. In
general, the hydroperoxide decomposition in-
creases with increasing concentration.
curred. Within 4 h of irradiation, the complexes
Ž .
Ž .
Ž .
Ti O TPPF20 4b, Ti O TMP 5b and Ti O TP-
PCl4 6b were completely destroyed. The por-
Ž .
phyrinate Ti O TPPF8 7b, on the other hand,
was stable for 6 h. The product formation was
3.2.5. ActiÕation of hydrogen peroxide and t-
butylhydroperoxide
low with a total yield between 2 and 5%. Hy-
Ž
.
Ž
.
droperoxide 11a 50% and ketone 13a 30%
Because of the above discussed titanium por-
phyrinates reactivity, one can postulate a new
way of hydrogen peroxide activation according
were the main products present in the reaction
mixture, together with alcohol 12a and epoxide
14a. The desired activation path for hydrogen
peroxide was not possible under these reaction
conditions, but there must be another oxidising
species which is responsible for the rapid cata-
5
to Fig. 10. A mixture of 3.6=10y mol
Ž .
O5Ti P , 20.5 mmol cyclohexene 10a, 1.1 g
Ž
.
H2O2 30%, 9.5 mmol , 30 ml CH2Cl2 and 10
ml CH3CN was, therefore, stirred under argon
at 158C for 1 h before the irradiation with the
high-pressure mercury lamp TQ 150 Z2 was
started. A rapid bleaching of the catalyst oc-
Ž
lyst bleaching hydroxyl radicals; OH, for ex-
.
ample . However, all light with wavelenghts
shorter than 290 nm were filtered off by Pyrexe
glass. No significant direct hydrogen peroxide
Ž
.
Ž
.
Ž
.
Fig. 7. Activity of different catalyst 0.037 mmol during the photooxygenation 6 h of cyclohexene 10a 20 mmol .

(
)
P.E. Esser et al.rJournal of Molecular Catalysis A: Chemical 140 1999 13–24
21
Ž .
a Influence of additives on the catalyst activity and selectivity to the hydroperoxide 11a during the photooxygenation of
Fig. 8.
Ž
.
Ž
.
Ž
.
cyclohexene 10a 20.5"0.3 mmol with Ti O TMP 3b as catalyst in 40 ml CH2Cl2 8 h, TQ 150 Z2 : run 1: 0.018 mmol 3b; run 2: 0.036
mmol 3b; run 3: 0.037 mmol 3br0.21 mmol Imidazol; run 4: 0.037 mmol 3br0.15 mmol DABCO; run 5: 0.037 mmol 3br0.16 mmol
Ž
.
Ž .
Ti O acac2; and run 6: 0.037 mmol 3br0.16 mmol Cp2TiCl2. b Influence of additives on the catalyst activity and selectivity to 11a
Ž
.
Ž
.
Ž
.
during the photooxygenation of cyclohexene 10a 20.1"0.4 mmol with Ti O TPPF20 4b as catalyst in 40 ml CH2Cl2 8 h, TQ 150 Z2 :
run 7: 0.036 mmol 4b; run 8: 0.019 mmol 4b; run 9: 0.019 mmol 4br0.21 mmol Imidazol; run 10: 0.019 mmol 4br0.21 mmol DABCO;
and run 11: 0.037 mmol 4br20.2 mmol TBHP.
splitting can, therefore, occur. It is obvious that
the Ti P moiety is oxidised by hydrogen per-
catalyse the epoxidation of cis-cylooctene 10b
with TBHP in dichloromethane at 408C. If the
temperature is raised to 828C, selective forma-
IIŽ .
oxide with loss of the porphyrinate ligand.
Addition of t-butylhydroperoxide to the pho-
tooxygenation of cyclohexene decreases the
overall yield with a reduced decomposition of
the hydroperoxide 11a in the case of
Ž
.
tion of the epoxide 14b occurred Table 6 .
During the reaction, partial formation of the
corresponding peroxo complexes takes place
Ž
.
TLC . No catalyst bleaching could be ob-
Ž .
Ti O TPPF20 4b as catalyst, which is stable
served. The porphyrinate ligand has only a mi-
under these circumstances for 8 h. In the dark,
at ambient temperature, no decomposition of the
Ž .
hydroperoxide 11a occurred. Ti O P does not
Ž .
Fig. 9. Acetonitrile and imidazole as axial ligands: P sTMP,
TPPF20
.
Fig. 10. Photocatalytic activation of hydrogen peroxide.

(
)
22
P.E. Esser et al.rJournal of Molecular Catalysis A: Chemical 140 1999 13–24
Table 6
Thermal epoxidation of cis-cyclooctene 10b with TBHP in toluene
Ž
.
Ž
.
Ž
.
Ž
.
Catalyst
n Catalyst
n 10b
n TBHP
Conversion
TON
Yield
S 11b
w
x
w
x
w
x
Ž . w x
10b %
w
x
Ž . w x
11b %
w x
%
mmol
mmol
mmol
molrmol
5b
6b
3b
4b
0.021
0.020
0.020
0.020
2.3
2.5
2.7
2.8
2.3
3.1
3.0
3.0
9
8
8.1
90
84
90
88
12
12
10
12
14
12
10.1
10.8
8.8
828C, 20 h.
Ž .
nor influence on the catalyst activity. The best
mation of Ti O TPP 1c. In contrast, the phenyl
2
Ž .
Ž .
result is reached with Ti O TMP 3b. After a
substituted O5Ti P 3b– 6b show better cat-
reaction time of 60 h, the conversion reached
30% without loss of selectivity. The relative
reactivities of the cyclic olefins during the epox-
alytic performances.
3.2.6. Mechanism
Ž .
idation with the system TBHPrTi O TPP 1b
The titanium porphyrinate-catalyzed pho-
tooxygenation of cyclohexene 10a and cis-
cyclooctene 10b results in the formation of hy-
droperoxide. Singlet oxygen is produced via the
well-known energy transfer mechanism from
w
x
were described by Agarwal 53 . A big excess
of the olefin, with respect to the TBHP
n1brnTBHPrnOlefin s1r10r50 , was used be-
cause of catalyst deactivation through the for-
Ž
.
Ž
.
Fig. 11. Photo catalytic mechanism of titanium-catalyzed decomposition of allylic hydroperoxide.

(
)
P.E. Esser et al.rJournal of Molecular Catalysis A: Chemical 140 1999 13–24
23
excited porphyrinate to dioxygen. The product
yields and the selectivity towards hydroperoxide
formation are lower than in the case of free
porphine bases. This indicates lower quantum
peratures in the dark. The synthesis of a teflon
Ž .
porphyrinate Ti O TPPF28 could lead to prepar-
ative useful epoxide yields and a stable complex
for photochemical activation of hydrogen perox-
ide.
Ž
.
yields and a photo catalytic property of the
titanyl group. Fig. 11 shows a summary of
possible hydroperoxide decomposition routes.
The oxotitanium porphyrinate can be trans-
References
Ž
.
formed to cis-hydroxy-alkylperoxytitanium IV
w x
1
R.A. Sheldon, G. Franz, Ullmann’s Encyclopedia of Indus-
trial Chemistry, Vol. A18, Verlag Chemie, Weinheim, 1991,
p. 261.
L.I. Simandi, Dioxygen Activation and Homogeneous Cat-
alytic Oxidation, Amsterdam, 1991.
Ž
.
porphyrinates, peroxotitanium IV porphyri-
nates and titanium II porphyrinates. The path-
Ž .
w x
2
ways b and e could be the rate-determing steps
in this sequence. The epoxide formation occurs
only for high catalyst concentrations and espe-
cially, when titanium porphyrinates with halo-
gen substituents on the phenyl ring were used.
N-oxides as intermediates, which were postuled
during the photolyses of peroxotitanium por-
phyrinates in ethanolic solution 54 , could also
be responsible for the epoxide formation and
catalyst decomposition. However, doting the re-
action mixture with photolabile titanium com-
plexes also indicates an influence of impurities
on catalyst bleaching.
w x
Ž .
3
R.A. Sheldon, J. Dakka, Catal. Today 19 1994 215.
w x
Ž .
M. Prein, W. Adam, Angew. Chem. 108 1996 519.
M. Prein, W. Adam, Angew. Chem. Int. Ed. Engl. 35 1996
4
w x
Ž
.
5
495.
w x
6
Ž
.
A.A. Frimer Ed. , Singlet Oxygen, Vol. 2, CRC Press, Boca
Raton, FL, 1985.
w x
7
Ž
.
H. Quast, T. Dietz, A. Witzel, Liebigs Ann. Chem. 1995
1495.
w
x
w x
8
S.G. DiMagno, P.H. Dussault, J.A. Schultz, J. Am. Chem.
Ž
.
Soc. 118 1996 5312.
w x
9
M. Vogt, PhD thesis, RWTH, Aachen, 1991.
w
w
w
x
Ž
.
10 I.T. Horvath, J. Rabai, Science 266 1994 72.
x
11 I.T. Horvath, J. Rabai, US Patent 5463082, 1995.
x
12 R.D. Chambers, G. Sandford, A. Shah, Synth. Commun. 26
Ž
.
1996 1861.
w
w
w
x
13 D.N. Hendrichson, M.G. Kinnaird, K.S. Suslick, J. Am.
Ž
.
Chem. Soc. 109 1987 1243.
x
14 A. Maldotti, R. Amadelli, C. Bartocci, V. Carassiti, E. Polo,
Ž
.
G. Varani, Coord. Chem. Rev. 125 1993 143.
4. Summary
x
15 A. Maldotti, C. Bartocci, R. Amadelli, E. Polo, P. Battioni,
Ž
.
D. Mansuy, J. Chem. Soc., Chem. Commun. 1991 1487.
In a comparitive study, the significant influ-
ence of substituents, oxotitanium and peroxoti-
tanium groups on meso-tetraarylporphines in
the course of the photosensitized singlet oxygen
ene reaction could be demonstrated. Free por-
phines show the best catalytic activity. Good
yields of the allylic hydroperoxides starting from
cyclohexene and cis-cyclooctene can be ob-
w
w
x
Ž
.
16 Y. Ito, Oxid. Commun. 15 1992 149.
x
17 M.W. Peterson, D.S. Rivers, R.M. Richman, J. Am. Chem.
Ž
.
Soc. 107 1985 2907.
w
w
w
w
w
w
w
w
w
w
x
Ž
.
18 T. Berthold, D. Rehorek, H. Hennig, Z. Chem. 26 1986
183.
x
19 C.R. Guest, K.D. Straub, J.A. Hutchinson, P.M. Rentzepis, J.
Ž
.
Am. Chem. Soc. 110 1988 5276.
x
20 L. Weber, G. Haufe, D. Rehorek, H. Hennig, J. Chem. Soc.,
Ž
.
Chem. Commun. 1991 502.
x
21 L. Weber, J. Behling, G. Haufe, H. Hennig, J. Prakt. Chem.
Ž
.
334 1992 265.
Ž
.
tained. meso- 2,6-Difluorophenyl porphine is
the sensitizer of choice for singlet oxygen. The
replacement of hydrogen by oxotitanium or per-
oxotitanium group causes a reduction of the
singlet oxygen production and an enhanced
x
22 L. Weber, J. Behling, G. Haufe, H. Hennig, J. Am. Chem.
Ž
.
Soc. 116 1994 2400.
x
23 H. Hennig, J. Behling, R. Meusiger, L. Weber, Chem. Ber.
Ž
.
128 1995 229.
x
24 S. Mosseri, P. Neta, P. Hambright, D.Y. Sabry, A. Harriman,
Ž
.
J. Chem. Soc., Dalton Trans. 1988 2705.
Ž
.
photocatalytic hydroperoxide decomposition.
x
25 Y. Matsuda, S. Sakamoto, H. Koshima, Y. Murakami, J. Am.
Ž
.
Chem. Soc. 107 1985 6415.
Complexes with fluorinated ligands are superior
x
26 Y. Matsuda, H. Koshima, K. Nakamura, Y Murakami, Chem.
sensitizers. The photochemical epoxidation ac-
Ž
.
Lett. 1988 625.
Ž .
tivity of Ti O TPPF20 4b at room temperature is
x
27 L. Weber, G. Haufe, D. Rehorek, H. Hennig, J. Mol. Catal.
Ž
.
still lower than the epoxidation at elevated tem-
60 1990 267.

(
)
24
P.E. Esser et al.rJournal of Molecular Catalysis A: Chemical 140 1999 13–24
w
x
w
w
x
Ž
.
28 H.J. Ledon, M. Bonnet, D. Galland, J. Am. Chem. Soc. 103
41 J.S. Lindsey, R.W. Wagner, J. Org. Chem. 54 1989 828.
Ž
.
x
1981 6209.
42 P. Fournari, R. Guilard, M. Fontesse, J.-M. Latour, J.-C.
w
w
w
x
Ž
.
29 E.M.K. Mansour, P. Maillard, P. Krausz, S. Gaspard, C.
Marchon, J. Organomet. Chem. 110 1976 205.
Ž
.
w
x
Giannotti, J. Mol. Catal. 41 1987 361.
43 R. Guilard, J.-M. Latour, C. Lecomte, J.-C. Marchon, J.
x
Ž
.
30 C.J. Boreham, J.-M. Latour, J.-C. Marchon, Inorg. Chim.
Protas, D. Ripoll, Inorg. Chem. 17 1978 1228.
Ž
.
w
x
Acta 45 1980 L69.
44 V.A. Ganzha, G.P. Gurinovich, B.M. Dzhagarov, G.D.
´
x
Ž
.
31 P.E. Esser, U. Englert, W. Keim, Chem. Ber. 129 1996
Egorova, E.I. Sagn, A.M. Shulga, J. Appl. Spectrosc. 50
Ž
.
833.
1989 402.
w
w
x
w
w
w
x
Ž
.
32 P.E. Esser, PhD thesis, RWTH, Aachen, 1995.
45 C.B. Hansen, W. Drenth, Catal. Lett. 20 1993 359.
x
x
46
33 G.H. Barnett, M.F. Hudson, K.M. Smith, J. Chem. Soc.,
.
Ž
.
x
Perkin Trans. 1 1975 1401.
47 P. Bergamini, S. Sostero, O. Traverso, P. Deplano, L.J.
w
w
w
w
w
w
w
x
Ž
.
34 G. H. Barnett, M.F. Hudson, K.M. Smith, Tetrahedron Lett.
Wilson, J. Chem. Soc., Dalton Trans. 1986 2311.
Ž
.
w
x
1973 2887.
48 M. Hoshino, M. Imamura, S. Watanabe, Y. Hama, J. Phys.
x
Ž
.
35 J.B. Kim, J.J. Leonard, F.R. Longo, J. Am. Chem. Soc. 94
Chem. 88 1984 45.
Ž
.
w
w
x
x
Ž
.
1972 3986.
49 T. Ozawa, A. Hanaki, Inorg. Chim. Acta 141 1989 49.
50 L. Weber, G. Haufe, D. Rehorek, H. Hennig, J. Mol. Catal.
x
36 J. Dalton, L.R. Milgrom, S.M. Pemberton, J. Chem. Soc.,
Ž
.
Ž
.
Perkin Trans. 2 1980 370.
60 1990 267.
x
w
w
x
37 W.J. Albery, P.N. Bartlett, C.C. Jones, L.R. Milgrom, J.
51 R.W. Harrigan, G.S. Hammond, H.B. Gray, J. Organomet.
Ž
.
Ž
.
Chem. Res., Synop. 1985 364.
Chem. 81 1974 79.
x
x
Ž
.
38 R.W. Wagner, D.S. Lawrence, J.S. Lindsey, Tetrahedron
52 Z.-T. Tsai, C.H. Brubaker, J. Organomet. Chem 166 1979
Ž
.
Lett. 28 1987 3069.
199.
x
w
w
x
x
Ž
.
39 H. Volz, S. Schneckenburger, J. Prakt. Chem., Chem.-Ztg.
53 D.D. Agarwal, J. Ind. Chem. B 31 1992 517.
54 M. Hoshino, K. Yamamoto, L.P. Lillis, T. Chijimatsu, J.
Ž
.
335 1993 283.
x
Ž
.
40 A.W. van der Made, E.J.H. Hoppenbrouwer, R.J.M. Nolte,
Uzawa, Inorg. Chem. 32 1993 5002.
Ž
.
W. Drenth, Rec. Trav. Chim. Pays-Bas 107 1988 15.