The effect of phenyl substituents on the activity of some zirconocene photoinitiators

Article abstract of DOI:10.1002/ejic.200390044

In this paper we report the synthesis of several unbridged zirconocenes, 1a-c, 2 and 5, each bearing a phenyl substituent in the 2-position of the cyclopentadienyl ring. We also report their photochemical behaviour and we compare the results with those obtained with the corresponding unsubstituted metallocenes 3a-c and 4. Study of their electronic spectra and EPR/spin trapping experiments showed photogeneration of ligand- and zirconium-centred radicals, thus making these complexes suitable as photoinitiators for radical polymerisation processes. The substitution of the hydrogen atom in the 2-position with a phenyl group had positive effects both on the absorption coefficients, improving the light absorption efficiency, and on the reaction yields of tert-butyl acrylate polymerisation. These results can be explained in terms of the combined effects of greater persistence of the radical in solution and/or increased efficiency of the radical in initiating the polymerisation process. Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim, Germany, 2003.

Full text of DOI:10.1002/ejic.200390044

FULL PAPER
The Effect of Phenyl Substituents on the Activity of Some
Zirconocene Photoinitiators[‡]
Eleonora Polo,*[a] Andrea Barbieri,[a] and Orazio Traverso[b]
Keywords: Homogeneous catalysis / Metallocenes / Photolysis / Polymerisation / Zirconium
In this paper we report the synthesis of several unbridged
zirconocenes, 1a−c, 2 and 5, each bearing a phenyl substitu-
atom in the 2-position with a phenyl group had positive ef-
fects both on the absorption coefficients, improving the light
ent in the 2-position of the cyclopentadienyl ring. We also absorption efficiency, and on the reaction yields of tert-butyl
report their photochemical behaviour and we compare the
results with those obtained with the corresponding unsubsti-
acrylate polymerisation. These results can be explained in
terms of the combined effects of greater persistence of the
tuted metallocenes 3a−c and 4. Study of their electronic spec- radical in solution and/or increased efficiency of the radical
tra and EPR/spin trapping experiments showed photogen- in initiating the polymerisation process.
eration of ligand- and zirconium-centred radicals, thus mak-
ing these complexes suitable as photoinitiators for radical
polymerisation processes. The substitution of the hydrogen
( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2003)
Introduction
tracted increasing interest due to the ‘‘tuneability’’ of their
structures and because they leave few toxic residues and
photobleach with high efficiency, thus leaving clear or col-
ourless coatings after exposure to argon lasers.
In the context of our research into metallocenes suitable
as photoinitiators in radical polymerisation processes, we
have recently[1] explored the possible use of several known
and new zirconocenes. In the light of the encouraging re-
sults obtained, we then investigated possible improvements
to their activity as photoinitiators through the introduction
of particular substituents in the cyclopentadienyl moiety to
modify the electronic distribution.
In this paper we report the photochemical behaviour of
several unbridged zirconocenes, each bearing a phenyl sub-
stituent in the 2-position of the cyclopentadienyl ring. This
substitution should help in stabilising the radical of the
polymerisation initiator and enhance the polymerisation ef-
ficiency, due to the potential delocalisation of the spare
electron offered by the phenyl ring, thus prolonging its life
in solution. The introduction of this substituent should in-
crease the absorption coefficients, thus improving the light
absorption efficiency.
A growing area of interest in photochemistry is the de-
sign of new photoinitiators for use in pigmented coatings,
photoresists, adhesives, and holography;[2Ϫ4] of these, initi-
ators with absorption peaks at wavelengths longer than
300 nm are of particular interest since they can be used with
the visible light produced by continuous wave lasers.[5]
Since the absorption of a photon generates an electronic-
ally excited state that does not occur thermally or occurs
only at very high temperatures, photoinitiated reactions and
thermally activated processes can induce different reaction
pathways. Light can be used as a clean, tuneable and versat-
ile ‘‘reagent’’, thus allowing selective activation of a photo-
initiator, also in the presence of organic solvents and chem-
ical contaminants. Another advantage offered by photoini-
tiation is that the light can be spatially directed and turned
on or off simply.
The desirability of expanding the application of photoini-
tiated chemistry has recently resulted in the discovery of
several classes of inorganic and organometallic
photoinitiators.[6Ϫ9] Among these, metallocenes have at-
Several zirconium complexes were synthesised and
studied: bis[η5-(2-phenyl-4,5,6,7-tetrahydro-1h-indenyl)]zir-
conium dichloride [1a, (C6Ph)2ZrCl2], bis[η5-(2-phenyl-
4,5,6,7,8-hexahydroazulenyl)]zirconium dichloride [1b,
(C7Ph)2ZrCl2], bis[η5-(2-phenyl-4,5,6,7,8,9-hexahydro-1H-
[‡]
Zirconocenes
as
Photoinitiators
for
Free
Radical
Polymerisation of Acrylates, 2. Part 1: Ref.[1]
C.N.R., I.S.O.F. sez. Ferrara,
Via L. Borsari 46, 44100 Ferrara, Italy
Fax: (internat.) ϩ 39-0532/240709
E-mail: tr3@unife.it
[a]
cyclopenta[8]annulenyl)]zirconium
dichloride
[1c,
tr1@unife.it
Dip. di Chimica dell’Universita di Ferrara,
Via L. Borsari 46, 44100 Ferrara, Italy
Fax: (internat.) ϩ 39-0532/240709
E-mail: tr2@unife.it
[b]
(C8Ph)2ZrCl2] and bis[η5-(2-phenylindenyl)]zirconium
dichloride [2, (IndPh)2ZrCl2]. The results were compared
with those obtained with the corresponding unsubstituted
`
324
 2003 WILEY-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim 1434Ϫ1948/03/0102Ϫ0324 $ 20.00ϩ.50/0 Eur. J. Inorg. Chem. 2003, 324Ϫ330

The Effect of Phenyl Substituents on the Activity of Some Zirconocene Photoinitiators
FULL PAPER
metallocenes: bis[η5-(4,5,6,7-tetrahydro-1H-indenyl)]zircon-
2-Phenyl-substituted ligands can easily be obtained in
ium dichloride [3a, (C6)2ZrCl2], bis[η5-(4,5,6,7,8-hexahy- three steps from the enones 6aϪc. Treatment of 6aϪc with
droazulenyl)]zirconium dichloride [3b, (C7)2ZrCl2], bis[η5- phenyllithium gave the allyl alcohols 7aϪc, which were de-
(4,5,6,7,8,9-hexahydro-1H-cyclopenta[8]annulenyl)]- hydrated under very mild conditions by treatment at room
zirconium dichloride [3c, (C8)2ZrCl2], and bis(η5-indenyl)- temp. with Amberlyst 15, a strongly acidic ion-exchange
zirconium dichloride (4, Ind2ZrCl2).
resin. Since the resulting dienes 8aϪc polymerise quite eas-
ily, it is crucial to use low temperatures and short workup
times for the dehydration step in order to improve the reac-
tion yields. The use of an ion-exchange resin gave better
yields than the usual acid workup, since the reagent can be
easily and rapidly removed from the reaction medium by
simple filtration. Furthermore, in the case of these phenyl-
substituted ligands, filtration through Florisil afforded only
one of the three possible isomeric dienes, thus making the
hydrogen abstraction easier, since only the isomer bearing
an allylic ϪCH2Ϫ group can easily be dehydrogenated.[11]
Treatment of the dienes with a slight excess of n-butylli-
thium gave the corresponding lithium salts 9aϪc, with over-
all yields over the three preceding steps ranging from 60 to
70%. Treatment of the lithium salts 9aϪc with ZrCl4·2THF
in THF in a 2.3:1 molar ratio resulted in the formation of
the corresponding metallocenes 1aϪc, in yields ranging
from 68 to 72%.
Another similar complex, bis(4-methyl-1,2-diphenylcy-
clopentadienyl)zirconium dichloride [5, (CpMePh2)2ZrCl2],
was also synthesised and tested. It should combine the
above advantages with a shorter and cheaper synthetic pro-
cedure.
The synthesis of the complex 5 was performed according
to Scheme 2, starting from 4-hydroxy-3,4-diphenylcyclo-
pent-2-en-1-one.
Results and Discussion
Ligand and Complex Synthesis
The synthetic approach is outlined in Scheme 1. The
starting enones 6aϪc were obtained by a synthetic proced-
ure previously designed in our laboratories.[10]
Scheme 2
The synthesis of the ligand 14[12,13] and of the corres-
ponding metallocenes have already been described by
others,[14,15] but we have found that the reaction yields were
hardly reproducible. Our modified procedure proved more
reliable and gave higher yields both of the intermediates
and of the metallocene 5.
Scheme 1
Eur. J. Inorg. Chem. 2003, 324Ϫ330
325

E. Polo, A. Barbieri, O. Traverso
FULL PAPER
The starting compound was converted into the enone 11 transitions, in order to test their photoreactivity in solution
by treatment with the reagent Me3SiCl/NaI in acetonitrile and whether they might be possible candidates for free rad-
(a synthetic equivalent of Me3SiI)[16,17] under very mild ical photopolymerisation. The nature of the complexes has
conditions. Treatment with methyllithium, followed by de- a significant influence on their photoreactivity in solution;
hydration with Amberlyst 15, gave the diene 13 as the sole in particular, the introduction of the phenyl substituent
isomer, if Florisil treatment was used. Treatment with n- seems to accelerate the initial decomposition of the com-
butyllithium and metallation with ZrCl4·2THF gave the plexes, but only to a certain extent. In the cases of com-
metallocene 5.
plexes 1aϪc, 4 and 5, we observed percentages of degrada-
tion (estimated by UV/Vis spectroscopy) varying from 10
to 40% with respect to the initial values after only 10 min
(Figure 1). After that time, though, a sort of plateau was
reached, and this persisted for the rest of the irradiation
time, as if a photostationary equilibrium had been estab-
lished. The photodecomposition was reversible; after 24 h
of storage in the dark, the initial spectral pattern and ab-
sorbance values were almost completely recovered. In con-
trast, only 15% of the unsubstituted complexes 3aϪc was
degraded even after 1 h of irradiation.
Photochemistry
UV/Vis spectra of the complexes 1aϪc, 2 and 5 were re-
corded in benzene and compared with those obtained previ-
ously[1] for the complexes 3aϪc and 4. While we did not
observe significant variations in the position of the λ max-
ima (just a slight shift towards shorter wavelengths), we did
see that the substituents in the cyclopentadienyl ring in-
creased the extinction coefficients by more than one order
of magnitude (Table 1), a feature particularly interesting for
the efficiency of the process.
Table 1. Spectroscopic properties of the complexes in benzene
Complex
λmax [nm] εmax [molϪ1·L·cmϪ1
]
1a
1b
1c
2
(C6Ph)2ZrCl2
(C7Ph)2ZrCl2
(C8Ph)2ZrCl2
(IndPh)2ZrCl2
320
300
300
300
310
330
310
350
310
345
305
345
310
380
23500
18000
14000
60500
60500
41000
1600
1100
2380
1540
2920
1720
1475
720
3a
3b
3c
4
(C6)2ZrCl2
(C7)2ZrCl2
(C8)2ZrCl2
(Ind)2ZrCl2
Figure 1. Percentage variation of the absorption band maxima of
complexes in benzene as a function of the irradiation time
A particular case is represented by complex 2, which
shows a similar photostability to the complexes 3aϪc, des-
pite its phenyl substituent. This effect can be attributed to
the extra stabilisation given by the indenyl aromatic system.
5
(CpMePh2)2ZrCl2 290
12000
These zirconium() complexes possess d0 configuration,
so any dominant effect of Ligand-Field (LF) and/or Metal-
to-Ligand Charge Transfer (MLCT) transitions can reason-
ably be excluded. Since the free ligand shows absorption
bands at lower energies (λ Ն 330 nm), the observed shift
supports the hypothesis of a prevalent Ligand-to-Metal
Charge Transfer (LMCT) character for the electronic trans-
ition, consistent with the formation of ligand- and metal-
centred radicals (Scheme 3). The photochemistry of the
complexes strongly supports the LMCT assignment.[18,19]
EPR Spin Trapping Investigation
To establish the nature of the species produced upon irra-
diation, EPR/spin trapping experiments were carried out by
direct irradiation of benzene solutions of the zirconocenes
inside the spectrometer cavity. The spin trap N-tert-butyl-α-
phenylnitrone (PBN) was used to identify the nature of the
radical species (Scheme 4).[20,21]
Scheme 4
Scheme 3
Besides the signals already recorded with the reference
Selective photolysis was carried out on the complexes in complexes 3aϪc and 4, the spectra of the phenyl-substi-
benzene at the wavelengths corresponding to their LMCT tuted complexes 1aϪc, 2 and 5 showed the presence of a
326
Eur. J. Inorg. Chem. 2003, 324Ϫ330

The Effect of Phenyl Substituents on the Activity of Some Zirconocene Photoinitiators
FULL PAPER
second radical species, consistent with the presence of a studies, since it combines good monomer conversion with a
cyclopentadienyl radical.[22] Since the hyperfine coupling shorter synthetic preparation.
constant (hfcc) values are common to all the phenyl-substi-
Another interesting feature is that the polymerisation
tuted complexes, including the metallocene 5, which does yields are linear with the irradiation time. This result could
not possess a second ring, they can reasonably be assigned be a possible sign of a living radical process.[23] Further
to a radical delocalised on the phenyl ring.
studies in order to ascertain the nature of this process un-
ambiguously are currently in progress.
Table 2. hfcc values of the radical spin adducts with PBN in ben-
zene
As was to be expected, the stereochemical outcome of
the reaction was quite independent of the complex used and
the data were quite similar to those reported with the un-
substituted complexes, with a content of syndiotactic triads
[rr] ranging from 30 to 35% and of isotactic triads [mm]
around 6%. This result is consistent with the radical nature
of the process.
Complex
Species I
Species II
aN [G] aH [G] [%] aN [G] aH [G] [%]
1a
1b
1c
2
(C6Ph)2ZrCl2
(C7Ph)2ZrCl2
(C8Ph)2ZrCl2
(IndPh)2ZrCl2
(C6)2ZrCl2
(C7)2ZrCl2
(C8)2ZrCl2
14.6 2.4
14.2 2.1
14.6 2.7
14.5 2.6
14.6 3.2
14.5 2.9
14.7 3.0
14.6 2.5
61 15.3 2.6
60 15.0 4.7
70 15.0 3.8
66 15.4 3.3
100 Ϫ
100 Ϫ
100 Ϫ
100 Ϫ
39
40
30
34
Ϫ
Ϫ
Ϫ
Ϫ
29
Table 3. tert-Butyl acrylate polymerisation data
3a
3b
3c
4
Ϫ
Ϫ
Ϫ
Ϫ
[a]
Complex
20% (v/v) tBA in C6H6
(Ind)2ZrCl2
(CpMePh2)2ZrCl2 14.6 2.3
Yield [%][b] Stereochemistry [%][c]
5
71 14.9 3.5
[rr]
[mr]
[mm]
1a
1b
1c
2
(C6Ph)2ZrCl2
(C7Ph)2ZrCl2
(C8Ph)2ZrCl2
(IndPh)2ZrCl2
(C6)2ZrCl2
(C7)2ZrCl2
(C8)2ZrCl2
46
46
39
56
24
22
24
25
30
34
36
35
37
40
35
36
35
65
60
58
59
52
45
50
51
60
05
06
06
06
11
15
15
13
05
Polymerisation Activity
The polymerisation of acrylates has great industrial signi-
ficance and has been the subject of intensive experimental
3a
3b
investigations and practical applications. On the basis of the 3c
4
(Ind)2ZrCl2
(CpMePh2)2ZrCl2 40
photochemical findings, polymerisation tests were per-
formed on tert-butyl acrylate (tBA), the least reactive
5
monomer in the acrylates series. For this reason, the com-
mercial product does not contain stabilisers as the other
[a]
Polymerisation conditions: a 20% (v/v) monomer solution in
benzene in a 4-mL cuvette was irradiated (λ Ͼ 320 nm) for 1 h in
[b]
the presence of 7 mg of the metallocene.
Degree of polymeris-
ation: (weight of polymer obtained/weight of starting monomer) ϫ
100. [c] Percentage composition of syndiotactic [rr], atactic [mr], and
isotactic [mm] triads in the polymer, calculated from deconvolution
of the 13C NMR peaks at δ ϭ 41Ϫ43 ppm.
acrylates do, and its chemical sluggishness makes it a good
terminating agent for the polymerisation of the other acryl-
ate monomers. A positive result obtained in tBA polymeris-
ation can be considered evidence of a high efficiency of a
metallorganic complex as a radical initiator and can rule
out the possibility that the polymer obtained may have
come from direct photochemical activation of the mono-
mer.
Conclusion
The polymerisation tests were carried out in benzene and
the tacticity of the polymer formed was estimated by decon-
On the basis of their photochemical behaviour in solu-
volution of the methine peaks of poly(tBA) at δ ഠ 42 ppm tion, known and new phenyl-substituted metallocene com-
in the 13C{1H} NMR spectra.
plexes 1aϪc, 2 and 5 have been used here for the first time
The final result of the polymerisation might be due to as photoinitiators for radical polymerisation processes.
the combined effects of two factors: i) the photoreactivity An interesting feature is the high extinction coefficients
of the complex in solution, and ii) the efficiency of the rad- of these new complexes, which improve the light absorption
ical in promoting the polymerisation process. efficiency of the irradiation process, and the possibility to
The polymer yields obtained with the complexes 1a,b, 2 shift their λ maxima towards lower energies by a judicious
and 5 were at least twice as high as those obtained with the choice of phenyl substituents without the necessity to mod-
corresponding unsubstituted metallocenes 3aϪc and 4. This ify the entire synthetic procedure.
improvement in the polymerisation outcome can be ex-
The radical species involved in the photoinduced cleavage
plained by better persistence of the radical initiating species of the metalϪligand bond have been identified and tBA
in solution. The extra stabilisation provided by the phenyl polymerisation has been studied in detail and the results
ring can slow down recombination and fragmentation pro- compared with those obtained with the complexes 3aϪc
cesses, thus guaranteeing higher concentrations of radical and 4, already described in the first paper of this series.[1]
initiators in solution. Particularly interesting is the bis(2-
The substitution of the hydrogen atom in position 2 of
phenylindenyl)zirconium dichloride (2), which offers the the cyclopentadienyl ring with a phenyl group increased the
highest percentage in conversion combined with the best polymerisation yields significantly, due to the combined ef-
photostability in solution. Complex 5 also deserves more fects of a longer persistence of the radical in solution and/
Eur. J. Inorg. Chem. 2003, 324Ϫ330
327

E. Polo, A. Barbieri, O. Traverso
FULL PAPER
Preparation of the Complexes: The enones 6aϪc,[11] 4-hydroxy-3,4-
diphenylcyclopent-2-en-1-one (10),[13] the metallocenes Ind2ZrCl2
(3)[25] and (2-phenylindenyl)2ZrCl2 (4)[26] were prepared as de-
scribed in the literature.
or increased efficiency of the radical in initiating the poly-
merisation process.
Experimental Section
Allylic Alcohols 7. General Procedure: Enone 6 (8.80 mmol) in an-
hydrous diethyl ether (20 mL) was cooled to 0 °C under nitrogen.
Phenyllithium (6.5 mL, 10.4 mmol, 1.6  in cyclohexane/ether) was
slowly added, and the yellow solution became brown. The cooling
bath was removed and the mixture was stirred at room temp. until
TLC (eluent: diethyl ether/petroleum ether, 1:1) showed only traces
of the starting material (about 2 h). The reaction mixture was then
cooled to 0 °C, the excess of alkylating agent was decomposed by
careful addition of iced water, and the separated organic phase was
washed several times with diethyl ether. The collected organic frac-
tions were dried and the solvent was evaporated under reduced
pressure. The crude alcohol 7 was obtained in quantitative yield as
a mixture of two diastereoisomers.
General Remarks: All manipulations of air- and/or moisture-sensit-
ive materials were carried out under an inert gas by use of a dual
vacuum/argon line and standard Schlenk techniques. All solvents
were thoroughly dehydrated and deoxygenated under argon before
use. They were dried and purified by heating at reflux under argon
in the presence of a suitable drying agent (pentane, petroleum ether,
diethyl ether, benzene, toluene and THF: Na/benzophenone ketyl
or potassium; CH2Cl2: CaH2) followed by distillation and storage
˚
under argon in Young’s ampoules. CH3CN was dried with 4 A
molecular sieves. Solvents and solutions were transferred under
positive argon pressure through stainless steel cannulas (diameter
0.5Ϫ2.0 mm) and mixtures were filtered in a similar way with
modified cannulas that could be fitted with glass fibre filter disks
(Whatman GFC). Unless otherwise specified, all reagents were pur-
chased from commercial suppliers (Aldrich and Fluka) and used
without further purification. Reaction courses and product mix-
tures were routinely monitored by thin layer chromatography
(TLC) on pre-coated silica gel F254 plates. Preparative flash chro-
2-phenyl-2,4,5,6,7,7a-hexahydro-1H-inden-2-ol (7a). Major Isomer:
1H NMR (CDCl3): δ ϭ 1.20Ϫ1.50 (m, 3 H), 1.60 (s, 1 H, exchanged
on D2O addition, OH), 1.70Ϫ2.20 (m, 6 H), 2.40Ϫ2.65 (m, 2 H,
CH2Cp), 5.40 (m, 1 H, ϭCHCp), 7.10Ϫ7.60 (m, 5 H, ArH) ppm.
Minor Isomer: 1H NMR (CDCl3): δ ϭ 1.20Ϫ1.50 (m, 3 H), 1.60
(s, 1 H, exchanged on D2O addition, OH), 1.70Ϫ2.20 (m, 6 H),
2.40Ϫ2.65 (m, 2 H, CH2Cp), 5.50 (m, 1 H, ϭCHCp), 7.10Ϫ7.60
(m, 5 H, ArH) ppm.
˚
matography was carried out with 60 A silica gel (230Ϫ400 mesh
ASTM). Anhydrous magnesium sulfate was used for drying. Ele-
mental analyses were performed with a Carlo Erba 1106 Elemental
Analysis apparatus.
2-phenyl-1,2,4,5,6,7,8,8a-octahydroazulen-2-ol (7b). Major Isomer:
1H NMR (CDCl3): δ ϭ 1.35Ϫ1.95 (m, 11 H), 2.35Ϫ2.60 (m, 2 H),
2.65Ϫ2.81 (m, 1 H), 5.48 (m, 1 H, ϭCHCp), 7.15Ϫ7.60 (m, 5 H,
NMR Spectra: 1H (200.13 MHz) and 13C (50.32 MHz) NMR spec-
tra were recorded at room temperature with a Bruker AC 200 spec-
trometer. Spectra were referenced internally to the residual protio
solvent resonance relative to tetramethylsilane (δ ϭ 0 ppm). Deu-
terated solvents were dried and distilled under argon in the pres-
ence of a suitable drying agent (CDCl3: CaH2; C6D6: K) and stored
under argon in Young’s ampoules.
1
ArH) ppm. Minor Isomer: H NMR (CDCl3): δ ϭ 1.35Ϫ1.95 (m,
11 H), 2.35Ϫ2.60 (m, 2 H), 2.65Ϫ2.81 (m, 1 H), 5.58 (m, 1 H, ϭ
CHCp), 7.15Ϫ7.60 (m, 5 H, ArH) ppm.
2-phenyl-2,4,5,6,7,8,9,9a-octahydro-1H-cyclopenta[8]annulen-2-ol
1
(7c). Major Isomer: H NMR (CDCl3): δ ϭ 1.38Ϫ2.40 (m, 13 H),
2.40Ϫ2.60 (m, 2 H), 2.62Ϫ2.70 (m, 1 H), 5.52 (m, 1 H, ϭCHCp),
1
7.10Ϫ7.65 (m, 5 H, ArH) ppm. Minor Isomer: H NMR (CDCl3):
EPR Studies: The spin trap PBN was purchased from
SigmaϪAldrich Chem. Co. and used without further purification.
The EPR/spin trapping experiments were performed with a Bruker
EMX spectrometer, operating in the X-band (microwave fre-
quency ϭ 9.74 GHz, microwave power ϭ 20 mW, magnetic field
modulation frequency ϭ 100 kHz, magnetic field modulation am-
plitude ϭ 1 G). Sample solutions were freeze-pump-thaw-degassed
and then irradiated with a 350-W medium-pressure Hg lamp dir-
ectly in the EPR spectrometer cavity (Bruker ER-4104OR, TE102).
Hyperfine coupling constants (hfccs) were calculated by best-fit
simulation of experimental spectra with NIEHS WinSim soft-
ware.[24]
δ ϭ 1.38Ϫ2.40 (m, 13 H), 2.40Ϫ2.60 (m, 2 H), 2.62Ϫ2.70 (m, 1
H), 5.60 (m, 1 H, ϭCHCp), 7.10Ϫ7.65 (m, 5 H, ArH) ppm.
Dienes 8. General Procedure: Crude alcohol 7 (max, 8.80 mmol),
partially dissolved in pentane (30 mL), was stirred at room temp.
with Amberlyst 15 (1.27 g). After 1 h, the reaction was complete
(TLC: diethyl ether/petroleum ether, 2:1). Magnesium sulfate was
added and, after a further 5 min of stirring, the suspension was
filtered through a short column (h ϭ 6 cm; i.d. ϭ 3 cm) of Florisil
(100Ϫ200 mesh) and the solvents were evaporated under reduced
pressure, keeping the temperature below 30 °C. A yellow solid
was obtained.
Photochemical Experiments: Irradiation was performed with an
Oriel 500-W high-pressure Hg lamp equipped with cut-off filters
for wavelength selection and a water filter to prevent thermal pro-
2-phenyl-4,5,6,7-tetrahydro-1H-indene (8a): 1H NMR (CDCl3): δ ϭ
1.60Ϫ1.80 (m, 4 H, CCH2CH2), 2.20Ϫ2.45 (m, 4 H, CCH2CH2),
3.25 (d, J ϭ 1.17 Hz, 2 H, CH2Cp), 6.70 (d, J ϭ 1.17 Hz, 1 H, ϭ
CHCp), 7.10Ϫ7.70 (m, 5 H, ArH) ppm.
cesses. UV/Vis spectra were recorded with
Lambda 40 double-beam spectrophotometer.
a PerkinϪElmer
2-phenyl-4,5,6,7,8-hexahydroazulene (8b): 1H NMR (CDCl3): δ ϭ
1.55Ϫ1.82 (m, 6 H, CCH2CH2), 2.40Ϫ2.52 (m, 4 H, CCH2CH2),
3.25 (d, J ϭ 1.08 Hz, 2 H, CH2Cp), 6.70 (d, J ϭ 1.08 Hz, 1 H, ϭ
CHCp), 7.15Ϫ7.20 (m, 1 H, ArH), 7.20Ϫ7.35 (m, 2 H, ArH),
7.35Ϫ7.50 (m, 2 H, ArH) ppm.
Polymerisation Tests: tert-Butyl acrylate was dehydrated with CaH2
and vacuum-distilled immediately before sample preparation. In a
typical experiment, a 20% (v/v) monomer solution in benzene in a
4-mL cuvette was irradiated at λ Ͼ 320 nm for 1 h in the presence
of 7 mg of the metallocene. All polymerisations were carried out at
20 °C. The reaction was quenched by addition of a small amount
of aqueous hydrochloric acid, and the polymer was precipitated by
pouring CH3OH/HCl (10:1) into the reaction mixture. The polymer
was collected by filtration and dried overnight under high vacuum.
2-phenyl-4,5,6,7,8,9-hexahydro-1H-cyclopenta[8]annulene (8c): 1H
NMR (CDCl3): δ ϭ 1.40Ϫ1.60 (m, 4 H, CCH2CH2), 1.60Ϫ1.80
(m, 4 H, CCH2CH2), 2.45Ϫ2.60 (m, 4 H, CCH2CH2), 3.25 (s, 2 H,
CH2Cp), 6.72 (s, 1 H, ϭCHCp), 7.10Ϫ7.65 (m, 5 H, ArH) ppm.
328
Eur. J. Inorg. Chem. 2003, 324Ϫ330

The Effect of Phenyl Substituents on the Activity of Some Zirconocene Photoinitiators
FULL PAPER
Lithium Salts 9. General Procedure: Diene 8 (max, 8.80 mmol) was washed with aqueous sodium thiosulfate, water and brine and
placed in a Schlenk flask, degassed, dissolved in anhydrous petro- dried. The solvent was evaporated under reduced pressure to give
leum ether (30 mL) and cooled to 0 °C. n-Butyllithium (5 mL, 2.5 a deep yellow solid, which was purified by flash chromatography
 solution in hexanes, 12.5 mmol) was added dropwise, and the
solution was stirred at 0 °C for 30 min, then allowed to reach room
temp. and stirred overnight. Lithium salt 8 separated from the solu-
tion as powdery solid, which was filtered through a cannula and
washed twice with petroleum ether. The residual solvent was
pumped off, leaving the lithium salt 9 as yellow pyrophoric powder.
(eluent: diethyl ether/petroleum ether, 1:1.5) to give the enone 11
(0.44 g, 1.86 mmol). (Yield 93%). M.p. 108Ϫ110 °C. 1H NMR
(CDCl3): δ ϭ 2.46 (dd, J ϭ 2.0, 18.8 Hz, 1 H, cis-PhCHCH2), 3.14
(dd, J ϭ 7.3, 18.8 Hz, 1 H, trans-PhCHCH2), 4.65 (dt, J ϭ 7.3,
1.7 Hz, 1 H, PhCHCH2), 6.77 (d, J ϭ 1.5 Hz, 1 H, COCH),
7.11Ϫ7.42 (m, 8 H, ArH), 7.47Ϫ7.61 (m, 2 H, ArH) ppm. 13C
NMR (CDCl3): δ ϭ 46.68, 46.75, 126.96, 127.06, 127.90, 128.70,
129.06, 129.11, 130.79, 133.18, 141.22, 142.41, 175.12, 208.12 ppm.
2-Phenyl-4,5,6,7-tetrahydro-1H-indenyllithium (9a): 1.07 g. Yield
over three steps: 60%; average yield per step 84%.
1-Methyl-3,4-diphenylcyclopent-2-en-1-ol
(12):
Methyllithium
2-Phenyl-4,5,6,7,8-hexahydroazulenyllithium (9b): 1.24 g. Yield over
(2.15 mmol, 1.5  in diethyl ether) was added dropwise to a solu-
tion of enone 11 (0.44 g, 1.86 mmol) in anhydrous THF (20 mL),
cooled to 0 °C. The yellow solution turned orange. The mixture
was stirred for 30 min at 0 °C and for 30 min at room temperature
(TLC: diethyl ether/petroleum ether, 1.5:1). The reaction mixture
was then again cooled to 0 °C, the excess of alkylating agent was
decomposed by careful addition of iced water, and the separated
organic phase was washed several times with diethyl ether. The col-
lected organic fractions were dried and the solvent was evaporated
under reduced pressure. The crude alcohol 12 was obtained in
quantitative yield as a mixture of two diastereoisomers. Major Iso-
mer: 1H NMR (CDCl3): δ ϭ 1.59 (s, 3 H, CH3), 1.68 (s, 1 H,
OH), 1.96Ϫ2.12 (m, 1 H, cis-PhCHCH2), 2.63Ϫ2.82 (m, 1 H, trans-
PhCHCH2), 4.27Ϫ4.42 (m, 1 H, PhCHCH2), 6.34 (d, J ϭ 1.5 Hz,
three steps: 65%; average yield per step 87%.
2-Phenyl-4,5,6,7,8,9-hexahydro-1H-cyclopenta[8]annulenyllithium
(9c): 1.44 g. Yield over three steps: 71%; average yield per step 89%.
Metallocene Dichlorides 1. General Procedure: ZrCl4·2THF
(4.5 mmol) in THF (20 mL) was added dropwise to solutions of the
lithium salts 9 (9.6 mmol) in anhydrous THF (20 mL). The reaction
mixtures were stirred overnight. The solvent was then removed un-
der vacuum, and the residues were extracted with toluene, filtered,
concentrated and layered with pentane. Fine, deep yellow solids
precipitated and were isolated by filtration, washing with pentane
and concentration under high vacuum.
Bis[η5-(2-phenyl-4,5,6,7-tetrahydro-1H-indenyl)]zirconium Dichlor-
ide (1a): 1.80 g. Yield 68%. C30H30Cl2Zr (552.69): calcd. C 65.2, H
1
1 H, COCH), 7.18Ϫ7.41 (m, 10 H, ArH) ppm. Minor Isomer: H
NMR (CDCl3): δ ϭ 1.50 (s, 3 H, CH3), 1.72 (s, 1 H, OH),
1.96Ϫ2.12 (m, 1 H, cis-PhCHCH2), 2.63Ϫ2.82 (m, 1 H, trans-
PhCHCH2), 4.27Ϫ4.42 (m, 1 H, PhCHCH2), 6.78 (d, J ϭ 1.5 Hz,
1 H, COCH), 7.08Ϫ7.41 (m, 10 H, ArH) ppm.
1
5.5, Cl 12.8; found C 65.8, H 5.7, Cl 12.4. H NMR (CDCl3): δ ϭ
1.2Ϫ1.5 (m, 4 H, CCH2CH2), 1.6Ϫ1.9 (m, 4 H, CCH2CH2),
2.1Ϫ2.4 (m, 8 H, CCH2CH2), 6.1 (s, 4 H, CHCp), 7.0Ϫ7.4 (m, 10
H, ArH) ppm. 13C NMR (CDCl3): δ ϭ 22.0 (CCH2CH2), 23.8
(CCH2CH2), 113.7 (CH Cp), 123.9 (CϭCH Cp), 125.2 (mCH Ar),
127.3 (pCH Ar), 129.0 (oCH Ar), 130.8 (CH2CϭCCp), 133.3 (CCϭ
Ar) ppm.
4-Methyl-1,2-diphenylcyclopentadiene (13): The crude alcohol 12
(max. 1.56 mmol) in diethyl ether (30 mL) was stirred at room
temp. with Amberlyst 15 (1.27 g). After 1 h, the reaction was com-
plete (TLC: diethyl ether/petroleum ether, 2:1). Magnesium sulfate
was added and, after a further 5 min of stirring, the suspension was
filtered and the solvents were evaporated under reduced pressure.
The residue was dissolved in petroleum ether and filtered through
a short column (h ϭ 6 cm; i.d. ϭ 3 cm) of Florisil (100Ϫ200 mesh)
and the solvents were evaporated under reduced pressure, keeping
the temperature below 30 °C. A yellow solid (0.30 g, 1.31 mmol)
was obtained (yield over two steps: 80%; average yield per step:
89%). 1H NMR (CDCl3): δ ϭ 2.19 (d, J ϭ 1.5 Hz, 3 H, CH3), 3.49
(d, J ϭ 1.1 Hz, 2 H, CH2Ph), 6.37 (d, J ϭ 1.3 Hz, 1 H, CϭCH),
7.11Ϫ7.46 (m, 10 H, ArH) ppm. 13C NMR (CDCl3): δ ϭ 16.2,
48.8, 126.0, 126.9, 127.7, 128.2, 128.35, 132.1, 137.6, 137.5, 136.6,
141.4, 144.0 ppm.
Bis[η5-(2-phenyl-4,5,6,7,8-hexahydroazulenyl)]zirconium Dichloride
(1b): 2.17 g. Yield 78%. C32H34Cl2Zr (580.74): calcd. C 66.2, H 5.9,
Cl 12.2; found C 65.8, H 6.2, Cl 12.0. 1H NMR (CDCl3): δ ϭ
1.6Ϫ2.1 (m, 12 H, CCH2CH2), 2.2Ϫ2.7 (m, 8 H, CCH2CH2), 6.3
(s, 4 H, CH Cp), 7.1Ϫ7.6 (m, 10 H, ArH) ppm. 13C NMR (CDCl3):
δ ϭ 28.6 (CH2), 30.0 (CH2), 32.0 (CH2), 116.4 (CH Cp), 124.4 (Cϭ
CH Cp), 125.3 (mCH Ar), 127.2 (pCH Ar), 129.0 (oCH Ar), 133.3
(CCϭAr), 134.6 (CH2CϭC Cp) ppm.
Bis[η5-(2-phenyl-4,5,6,7,8,9-hexahydro-1H-cyclopenta[8]annulenyl)]-
zirconium Dichloride (1c): 2.10 g. Yield 72%. C34H38Cl2Zr (608.80):
calcd. C 67.1, H 6.3, Cl 11.7; found C 66.8, H 6.8, Cl 11.3. 1H
NMR (CDCl3): δ ϭ 1.0Ϫ1.5 (m, 12 H, CCH2CH2), 1.6Ϫ1.9 (m, 4
H, CCH2CH2), 2.0Ϫ2.3 (m, 4 H, CCH2CH2), 2.3Ϫ2.6 (m, 4 H,
CCH2CH2), 5.6 (s, 4 H, CH Cp), 7.0Ϫ7.5 (m, 10 H, ArH) ppm.
13C NMR (CDCl3): δ ϭ 25.9 (CH2), 26.4 (CH2), 31.1 (CH2), 32.3
(CH2), 107.6 (CH Cp), 124.3 (mCH Ar), 124.7 (CϭCH Cp), 126.0
(pCH Ar), 127.5 (CH2CϭC Cp), 128.6 (oCH Ar), 134.9 (CCϭ
Ar) ppm.
4-Methyl-1,2-diphenylcyclopentadienyllithium (14): The diene 13
(0.30 g, 1.31 mmol) was placed in a Schlenk flask, degassed, dis-
solved in anhydrous petroleum ether (30 mL) and cooled to Ϫ20
°C. n-Butyllithium (2.20 mmol, 2.5  solution in hexanes) was ad-
ded dropwise, and the solution was stirred at Ϫ20 °C for 2 h and
then allowed slowly to reach room temp. and stirred overnight. The
lithium salt 14 separated from the solution as a powdery solid,
which was filtered through a cannula and washed twice with petro-
leum ether. The residual solvent was pumped off, leaving the li-
thium salt 14 (0.30 g, 1.24 mmol) as a pale yellow pyrophoric pow-
der (yield 95%).
3,4-Diphenylcyclopent-2-en-1-one (11): The alcohol 10 (0.5 g,
2 mmol), dissolved in dry benzene (10 mL), and Me3SiCl
(12 mmol, a slightly exothermic reaction occurs) were added to a
suspension of NaI (12 mmol) in anhydrous CH3CN (0.6 mL,
12 mmol). The mixture was stirred at room temp. in the dark until
TLC (eluent: diethyl ether/petroleum ether, 1.5:1), showed only
traces of the starting alcohol (about 1 h). The reaction mixture was
Bis(1,2-diphenyl-4-methylcyclopentadienyl)zirconium Dichloride (5):
then cooled to 0 °C, quenched with water and extracted several The lithium salt 14 (0.30 g, 1.24 mmol) was dissolved in anhydrous
times with diethyl ether. The collected organic fractions were
THF (20 mL), and a solution of ZrCl4·2THF (4.5 mmol) in THF
Eur. J. Inorg. Chem. 2003, 324Ϫ330
329

E. Polo, A. Barbieri, O. Traverso
FULL PAPER
Green, J. Organomet. Chem. 1999, 577, 211Ϫ218. For other
synthetic strategies, see references cited herein.
W. A. Herrmann, R. Anwander, H. Riepl, W. Scherer, C. R.
Whitaker, Organometallics 1993, 12, 4342Ϫ4349.
R. Guy, C. Mahmoud, Bull. Soc. Chim. Fr. 1970, 10,
3585Ϫ3598.
H. Plenio, D. Burth, Organometallics 1996, 15, 4054Ϫ4062.
F. Zhang, Y. Mu, L. Zhao, Y. Zhang, W. Bu, C. Chen, H. Zhai,
H. Hong, J. Organomet. Chem. 2000, 613, 68Ϫ76.
F. Zhang, Y. Mu, J. Wang, Z. Shi, W. Bu, S. Hu, Y. Zhang, S.
Feng, Polyhedron 2000, 19, 1941Ϫ1947.
G. A. Olah, A. Husain, B. P. Singh, A. K. Mehrotra, J. Org.
Chem. 1983, 48, 3667Ϫ3672.
T. Sakai, K. Miyata, M. Utaka, A. Takeda, Tetrahedron 1987,
28, 3817Ϫ3818.
G. L. Geoffroy, M. S. Wrighton, Organometallic Photochem-
istry, Academic Press, 1979.
A. Barbieri, A. Droghetti, S. Sostero, O. Traverso, J.
Photochem. Photobiol., A 1999, 129, 137Ϫ142.
S. Sostero, D. Rehorek, O. Traverso, E. Polo, Inorg. Chim. Acta
1993, 209, 171Ϫ176.
S. Di Martino, S. Sostero, O. Traverso, D. Rehorek, T. J. Kemp,
Inorg. Chim. Acta 1990, 176, 107Ϫ112.
D. R. Duling, J. Magn. Reson. Ser. B 1994, 104, 105Ϫ110.
B. A. Howell, J. Polym. Sci., Part A: Polym. Chem. 2000, 38,
1727Ϫ1729.
A. S. W. Li, K. B. Cummings, H. P. RoethLing, G. R. Buetner,
C. F. Chignell, J. Magn. Reson. 1988, 79, 140Ϫ142.
E. Samuel, Bull. Soc. Chim. Fr. 1966, 11, 3548Ϫ3564.
G. W. Coates, R. M. Waymouth, Science 1995, 267, 217Ϫ219.
Received July 2, 2002
(20 mL) was added dropwise. The reaction mixture was stirred
overnight. The solvent was then removed under vacuum, and the
residue was extracted with toluene, filtered, concentrated and
layered with pentane. A powdery, deep yellow solid (0.62 g,
0.99 mmol) precipitated and was isolated by filtration and washing
with pentane (yield 80%). C36H30Cl2Zr (624.35): calcd. C, 69.2, H
4.8; Cl, 11.4; found C 69.1, H 4.9; Cl, 11.3. 1H NMR (CDCl3): δ ϭ
1.73 (s, 6 H, CH3), 6.38 (s, 4 H, CpH), 7.11Ϫ7.49 (m, 20 H, ArH)
ppm. 13C NMR (CDCl3): δ ϭ 15.2, 118.9, 127.4, 128.0, 128.8,
129.4, 133.2 ppm.
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[1]
E. Polo, A. Barbieri, S. Sostero, M. L. H. Green, Eur. J. Inorg.
Chem. 2002, 8, 405Ϫ409.
N. S. Allen (Ed.), Photopolymerization and Photoimaging Sci-
[2]
ence and Technology, Elsevier, Amsterdam, 1989.
S. P. Pappas (Ed.), Radiation Curing: Science and Technology,
[3]
Plenum Press, New York, 1992.
A. B. Scranton, C. N. Bowman, R. W. Peiffer (Eds.), Photopoly-
[4]
merization: Fundamentals and Applications, ACS Symposium
Series, no. 673, American Chemical Society, Washington, DC,
1997.
[22]
[23]
[5]
Photoinitiators for UV curing. Key Products Selection Guide.
Additives, Ciba, Basel, 1998, available at http://www.cibasc.com
[6]
[24]
B. M. Monroe, G. C. Weed, Chem. Rev. 1993, 93, 435Ϫ448.
[7]
R. S. Davidson, J. Photochem. Photobiol.,
263Ϫ275.
A 1993, 69,
[25]
[26]
[8]
[9]
R. S. Davidson, J. Photochem. Photobiol., A 1993, 73, 81Ϫ96.
C. Kutal, Coord. Chem. Rev. 2001, 211, 353Ϫ368.
[10]
E. Polo, R. M. Bellabarba, G. Prini, O. Traverso, M. L. H.
[I02359]
330
Eur. J. Inorg. Chem. 2003, 324Ϫ330