Dinuclear zinc complexes based on parallel β-diiminato binding sites: Syntheses, structures, and properties as CO2/epoxide copplymerization catalysts

Article abstract of DOI:10.1021/om070221e

With the background that β-diketiminato zinc complexes efficiently catalyze the CO₂/epoxide copolymerization via a mechanism involving two catalyst molecules, a ligand system containing two parallel β-diiminato binding sites linked by a xanthene backbone ([^RXanthdim] ^2- with residues R = 2,3-dimethylphenyl and 2,4-difluorophenyl at the iminato units, respectively) was investigated with respect to its zinc coordination chemistry. The corresponding diimines [^RXanthdim]H ₂ were treated with diethylzinc to yield the complexes [ ^Me2C6H3Xanthdim](ZnEt)₂, 4, and [ ^F2C6H3Xanthdim](ZnEt(thf))₂, 5, respectively. In order to convert these compounds into polymerization catalysts, they were subsequently treated with SO₂, which indeed resulted in the corresponding ethylsulfinates. Due to aggregation via intermolecular bridging of ethylsulfinate ligands, the product after the reaction of 5 represents an insoluble coordination polymer. Aggregation does take place also for the primary product obtained from the reaction of 4, as evidenced by the isolation of the tetramer [{[^Me2C6H3Xanthdim]Zn₂(μ-O₂SEt)} μ-O₂SEt]₄, 6, and the precipitation of an insoluble solid on storing of such mixtures. 4, 5, and 6 display moderate activities as catalysts for die copolymerization of cyclohexene oxide and CO₂. A bimodal molecular weight distribution points to two effective mechanisms, one of which probably involves two cooperating Zn centers as anticipated. Possible structural reasons for these catalytic results are discussed.

Full text of DOI:10.1021/om070221e

3668
Organometallics 2007, 26, 3668-3676
Dinuclear Zinc Complexes Based on Parallel â-Diiminato Binding
Sites: Syntheses, Structures, and Properties as CO2/Epoxide
Copolymerization Catalysts
Maurice Frederic Pilz,† Christian Limberg,*,† Boyan B. Lazarov,‡ Kai C. Hultzsch,*,‡ and
Burkhard Ziemer†
Institut fu¨r Chemie, Humboldt-UniVersita¨t zu Berlin, Brook-Taylor-Strasse 2, 12489 Berlin, Germany, and
Institut fu¨r Organische Chemie, Friedrich-Alexander UniVersita¨t Erlangen-Nu¨rnberg, Henkestrasse 42,
91054 Erlangen, Germany
ReceiVed March 8, 2007
With the background that â-diketiminato zinc complexes efficiently catalyze the CO2/epoxide
copolymerization via a mechanism involving two catalyst molecules, a ligand system containing two
parallel â-diiminato binding sites linked by a xanthene backbone ([RXanthdim]2- with residues R )
2,3-dimethylphenyl and 2,4-difluorophenyl at the iminato units, respectively) was investigated with respect
to its zinc coordination chemistry. The corresponding diimines [RXanthdim]H2 were treated with diethylzinc
to yield the complexes [Me2C6H3Xanthdim](ZnEt)2, 4, and [F2C6H3Xanthdim](ZnEt(thf))2, 5, respectively.
In order to convert these compounds into polymerization catalysts, they were subsequently treated with
SO2, which indeed resulted in the corresponding ethylsulfinates. Due to aggregation via intermolecular
bridging of ethylsulfinate ligands, the product after the reaction of 5 represents an insoluble coordination
polymer. Aggregation does take place also for the primary product obtained from the reaction of 4, as
evidenced by the isolation of the tetramer [{[Me2C6H3Xanthdim]Zn2(µ-O2SEt)}µ-O2SEt]4, 6, and the
precipitation of an insoluble solid on storing of such mixtures. 4, 5, and 6 display moderate activities as
catalysts for the copolymerization of cyclohexene oxide and CO2. A bimodal molecular weight distribution
points to two effective mechanisms, one of which probably involves two cooperating Zn centers as
anticipated. Possible structural reasons for these catalytic results are discussed.
that is, research of the last years has increased the productivity
of this reaction by ca. 100 times and also has expanded the
range of applicable oxiranes. While a large number of catalysts
are known today, only a few of them reach the activity of zinc
complexes.3 G. W. Coates and co-workers discovered a highly
active, living polymerization system at low pressures and
temperatures based on â-diketiminato-zinc catalysts. In order
to prepare these, a mononuclear ethyl zinc precursor complex
(Scheme 1a) is treated with alcohols or acetic acid to give
dinuclear alkoxides and acetates, respectively.2e,h,4 Activation
of the ethyl precursors can also proceed via treatment with SO2,
which gives rise to ethylsulfinates.5 Solution studies hinted at
monomer/dimer equilibria dependent on the counteranions, and
on the basis of stoichiometric insertion studies and kinetic
investigations a mechanism was suggested involving two zinc
diketiminate units (Scheme 1b).3a This and other examples where
two diketiminato metal units cooperate in the activation of a
certain substrate,6 as well as the fact that bimetallic catalysis is
also utilized by many metalloenzymes,7 stimulated us to think
Introduction
CO2 is an inexpensive, nontoxic, totally renewable feedstock,
whose reduction would have a positive impact on the global
carbon balance, and these facts alone are sufficient motivation
for producing chemicals from CO2 whenever possible.1 After
early work in the area of polymers, recent years gave rise to a
number of new homogeneous catalysts for the copolymerization
of CO2 and oxiranes to form polycarbonates (eq 1).2
These discrete catalysts allow for more rapid and selective
polymerizations in comparison to the heterogeneous matches;
* Corresponding authors. Reprint requests to C.L. Fax: +49 30 2093
6966. E-mail: Christian.limberg@chemie.hu-berlin.de.
† Humboldt-Universita¨t zu Berlin.
‡ Friedrich-Alexander Universita¨t Erlangen-Nu¨rnberg.
(1) Arakawa, H.; Aresta, M.; Armor, J. N.; Barteau, M. A.; Beckman,
E. J.; Bell, A. T.; Bercaw, J. E.; Creutz, C.; Dinjus, E.; Dixon, D. A.; Domen,
K.; Dubois, D. L.; Eckert, J.; Fujita, E.; Gibson, D. H.; Goddard, W. A.;
Goodman, D. W.; Keller, J.; Kubas, G. J.; Kung, H. H.; Lyons, J. E.; Manzer,
L. E.; Marks, T. J.; Morokuma, K.; Nicholas, K. N.; Periana, R.; Que, L.;
Rostrup-Nielson, J.; Sachtler, W. M. H.; Schmidt, L. D.; Sen, A.; Somorjai,
G. A.; Stair, P. C.; Stults, B. R.; Tumas, W. Chem. ReV. 2001, 101, 953.
(2) (a) Darensbourg, D. J.; Holtcamp, M. W. Macromolecules 1995, 28,
7577. (b) Darensbourg, D. J.; Holtcamp, M. W. Coord. Chem. ReV. 1996,
153, 155. (c) Super, M. S.; Berluche, E.; Costello, C. A.; Beckman, E. J.
Polym. Mat. Sci. Eng. Prepr. 1996, 74, 430. (d) Super, M. S.; Beckman, E.
J. Macromol. Symp. 1998, 127, 89. (e) Cheng, M.; Lobkovsky, E. B.; Coates,
G. W. J. Am. Chem. Soc. 1998, 120, 11018. (f) Beckman, E. Science 1999,
283, 946. (g) Nazaki, K.; Nakano, K.; Hiyama, T. J. Am. Chem. Soc. 1999,
121, 1108. (h) Cheng, M.; Darling, N. A.; Lobkovsky, E. B.; Coates, G.
W. J. Chem. Soc., Chem. Commun. 2000, 2007.
(3) (a) Review on Zn-based catalysts: Coates, G. W.; Moore, D. R.
Angew. Chem., Int. Ed. 2004, 43, 6618. (b) Cr-Salen complexes with high
activity: Darensbourg, D. J.; Mackiewicz, R. M.; Phelps, A. L.; Billodeaux,
D. R. Acc. Chem. Res. 2004, 37, 836. Darensbourg, D. J.; Mackiewicz, R.
M.; Rodgers, J. L.; Phelps, A. L. Inorg. Chem. 2004, 43, 1831. Darensbourg,
D. J.; Mackiewicz, R. M.; Billodeaux, D. R. Organometallics 2005, 24,
144.
(4) (a) Cheng, M.; Moore, D. R.; Reczek, J. J.; Chamberlain, B. M.;
Lobkovsky, E. B.; Coates, G. W. J. Am. Chem. Soc. 2001, 123, 8738. (b)
Moore, D. R.; Cheng, M.; Lobkovsky, E. B.; Coates, G. W. Angew. Chem.
2002, 114, 2711; Angew. Chem., Int. Ed. 2002, 41, 2599. (c) Allen, S. D.;
Moore, D. R.; Lobkovsky, E. B.; Coates, G. W. J. Am. Chem. Soc. 2002,
124, 14284.
(5) Eberhardt, R.; Allmendinger, M.; Luinstra, G. A.; Rieger, B.
Organometallics 2003, 22, 211.
10.1021/om070221e CCC: $37.00 © 2007 American Chemical Society
Publication on Web 06/14/2007

Dinuclear Zinc Complexes
Organometallics, Vol. 26, No. 15, 2007 3669
Scheme 1. (a) â-Diketiminato Ethyl Zinc Precursor, (b)
Suggested Transition State of the Epoxide Ring Opening (R
) Polymer)
about the design of a ligand containing two diiminato moieties
in close proximity. Some ligands fulfilling this requirement have
already been described; in most of these examples the diiminato
moieties are incorporated into a macrocyclic ring system,8 in
other cases two N atoms belonging to different diiminato units
are linked,9 and in all these compounds the chelating functions
are arranged face to face. Inspired by the above-mentioned work
by Coates et al., some of them, for instance o-phenylene-bridged
bis(anilido-aldimine) compounds, were employed to synthesize
dinuclear zinc complexes, which indeed proved to be signifi-
cantly more active than mononuclear versions, at least under
very diluted conditions.8c
Figure 1. Structure of one of the two independent molecules of
[
cocrystallized molecule of chloroform and hydrogen atoms [apart
from the ones at N(1) and N(1)′] are omitted for clarity.
F2C6H3Xanthdim]H2 found in the elemental cell of its crystals. A
We were interested in synthesizing a ligand where the two
diiminato units are oriented parallel to each other, to investigate
the difference this makes for the cooperative behavior of the
two metal centers.
Table 1. Selected Bond Lengths [Å] and Angles [deg] for
[
F2C6H3Xanthdim]H2
N(1)-H(1)
C(15)-N(1)
C(14)-C(13)
C(21)-N(2)
0.88(4)
N(2)-H[N(1)]
C(14)-N(1)
C(13)-C(21)
C(22)-N(2)
1.987(4)
1.343(5)
1.436(5)
1.410(5)
Hence, recently we reported the synthesis of
1.405(5)
1.373(5)
1.294(5)
[
Me2C6H3Xanthdim]H2 (Scheme 2), where this idea is realized.10
Here we describe the derivatization of [RXanthdim]H2, the
syntheses of corresponding zinc complexes, and the results of
tests concerning their catalytic activity in the epoxide/CO2
copolymerization.
C(14)-N(1)-H(1)
C(14)-C(13)-C(21)
C(21)-N(2)-C(22)
116.4(4)
122.8(3)
121.2(4)
C(13)-C(14)-N(1)
C(13)-C(21)-N(2)
124.0(4)
122.0(4)
C(3)-C(2)-C(13)-C(14) 46.8(5) C(1′)-C(2′)-C(13′)-C(21′) 50.6(5)
Results and Discussion
C(13)-C(14)-N(1)-C(15) 2.7(4) C(13)-C(21)-N(2)-C(22)
N(1)-C(15)-C(16)-F(1) 3.5(6) N(2)-C(22)-C(23)-F(3)
2.5(3)
3.5(10)
Ligand Synthesis. The synthesis of [Me2C6H3Xanthdim]H2
reported10 starts from the diol 2, which can be prepared from
the dibromide 1 (Scheme 2). 2 then has to be converted into
the dicarboxylic acid 3 via initial OH/Br and Br/CN exchange
followed by acidic hydrolysis. A Vilsmeier type of reaction then
leads to a double vinamidinium salt, which is hydrolyzed, and
the last step in the reaction sequence consists of an anilation
reaction through which the residues R are introduced. So far
only one derivative containing 2,3-dimethylphenyl substituents,
[
Me2C6H3Xanthdim]H2, has been described. Here we report a
second one, [F2C6H3Xanthdim]H2 (see Scheme 2), that can be
obtained analogously and might be an alternative, if electron-
withdrawing residues are required. Figure 1 shows the molecular
structure of [F2C6H3Xanthdim]H2 adopted in the crystal (one of
two independent molecules present in the unit cell), and Table
1 lists its relevant bond lengths and angles. The crystallographi-
cal C2-axis goes right through the center of the molecule (C(9)-
Scheme 2

3670 Organometallics, Vol. 26, No. 15, 2007
Pilz et al.
Figure 2. Molecular structure of 4. Hydrogen atoms are omitted for clarity.
Scheme 3
O(1)). As in the molecular structure of [Me2C6H3Xanthdim]H2,10
the CdN and CdC bonds are localized within the â-diimine
units, and this also becomes evident in the solution NMR spectra
of [F2C6H3Xanthdim]H2. Within each of the diimine moieties the
distance between the fluorine atom in ortho-position of one of
the 2,4-difluorophenyl rings and the amine proton H(N) is less
than the sum of van der Waals radii, giving rise to hydrogen
bonding (F1-H[N(1)]: 2.276(3) Å). This and the sterically
undemanding nature of the fluorine atoms (in comparison to
the methyl groups in [Me2C6H3Xanthdim]H2) are the reason for
the almost perfectly coplanar â-diimine units. The distance
between them amounts to ca. 4.5 Å.
Complex Synthesis and Structures. In order to prepare dinu-
clear Zn complexes of the potential ligands [Me2C6H3Xanthdim]2-
and [F2C6H3Xanthdim]2-, the protonated forms were treated with
2 equiv of ZnEt2. This led to the complexes [Me2C6H3Xanthdim]-
(ZnEt)2, 4, and [F2C6H3Xanthdim](ZnEt(thf))2, 5 (Scheme 3),
which can be isolated in pure form after removing all volatiles
under vacuum and recrystallization. 4 and 5 were characterized
by means of IR and NMR spectroscopy as well as by elemental
analysis. In addition, single crystals could be grown, and the
results of the X-ray diffraction analyses are shown in Figures 2
and 3 (Tables 2 and 3 list relevant bond lengths and angles).
In 4 each Zn atom has a trigonal planar coordination
environment consisting of the two N atoms of the â-diiminato
ligand and the terminal ethyl ligand (sum of angles around the
Zn atoms: Zn(1): 360.1° and Zn(2): 359.5°). These planes are
somewhat tilted with respect to the planes defined by the
diiminato units in a way that brings the two Zn atoms closer
together. The Zn-Zn separation of 4.922 Å is significantly
shorter than that reported for a dinuclear zinc complex,8b 6.010
Å, containing one of the above-mentioned macrocyclic bis-â-
diiminato ligands. As the two Zn diiminato units are not oriented
perfectly parallel to each other and as additionally the xanthene
backbone is somewhat distorted in the crystal structures
probably due to packing effectssthe Zn-Zn distance can be
expected to be even shorter in a relaxed structure in solution.
Within each of the individual units the Zn-N bond distances
are equivalent [Zn(1)-N(1): 1.995 Å, Zn(1)-N(2): 1.997 Å,
Zn(2)-N(3): 2.002 Å, and Zn(2)-N(4): 2.000 Å], and the
same is true for the corresponding C-N and C-C bonds, which
points to an efficient delocalization of the π-electrons. As in
the case of the lithium salt of [Me2C6H3Xanthdim]2-, the diiminato
moieties are twisted10 with respect to the xanthene backbone
(C(1)-C(2)-C(45)-C(55): 44.1° and C(11)-C(12)-C(24)-
C(34): 62.7°).
The structure of 5 is similar to that of 4, but each Zn ion
coordinates a thf molecule in addition, thereby reaching asin
Zn chemistry rarely observed8b,11sdistorted trigonal pyramidal
coordination sphere, where the thf ligand occupies the axial
position [sum of angles for the Zn-N and Zn-C bonds around
the zinc ions: Zn(1), 353.5°; Zn(2), 348.3°]. In 5 the two Zn

Dinuclear Zinc Complexes
Organometallics, Vol. 26, No. 15, 2007 3671
Figure 3. Molecular structure of 5. Hydrogen atoms and two cocrystallized molecules of thf are omitted for clarity.
Table 2. Selected Bond Lengths [Å] and Angles [deg] for
Compound 4
for the xanthene backbone, but the Zn diiminato units are
directed away from each other so that the Zn-Zn distance is
larger than in 4 (5.6038 Å). The Zn-N and Zn-C bond lengths
of both compounds 4 and 5 are comparable to those previously
described for zinc atoms coordinated by â-diketiminato and alkyl
ligands.4a,c,5,12
As outlined in the Introduction, for catalytic applications zinc
alkyl complexes are not ideally suited as precursors, so that
they are usually converted to the corresponding acetates,
alkoxides, or alkylsulfinates for this purpose. We decided to
follow the alkylsulfinato approach and therefore reacted 4 and
5 dissolved in toluene or thf with SO2. In the case of 4 this
leads to a yellow solution, and after removal of all volatiles a
yellow solid is obtained that can be redissolved readily in thf.
Storing this solution at rt leads to the precipitation of a yellow
solid that analyzes as [Me2C6H3Xanthdim](Zn(O2SEt))2 and cannot
be brought back into solution.
Zn(1)-Zn(2)
Zn(1)-N(1)
Zn(2)-N(3)
Zn(1)-C(43)
N(1)-C(26)
C(25)-C(24)
C(34)-N(2)
C(47)-N(3)
C(46)-C(45)
C(55)-N(4)
4.922(1)
1.995(2)
2.002(2)
1.978(3)
1.425(1)
1.398(4)
1.334(3)
1.429(4)
1.391(4)
1.331(3)
Zn(1)-N(2)
Zn(2)-N(4)
Zn(2)-C(64)
N(1)-C(25)
C(24)-C(34)
N(2)-C(35)
N(3)-C(46)
C(45)-C(55)
N(4)-C(56)
1.997(2)
2.000(2)
1.963(3)
1.325(3)
1.402(4)
1.428(3)
1.333(3)
1.396(4)
1.434(4)
N(1)-Zn(1)-N(2)
C(26)-N(1)-C(25)
C(25)-C(24)-C(34)
C(34)-N(2)-C(35)
N(3)-Zn(2)-C(64)
N(3)-C(46)-C(45)
C(45)-C(55)-N(4)
92.75(9)
117.7(2)
124.0(2)
116.9(2)
134.0(2)
127.4(2)
127.9(3)
N(1)-Zn(1)-C(43)
N(1)-C(25)-C(24)
C(24)-C(34)-N(2)
N(3)-Zn(2)-N(4)
C(47)-N(3)-C(46)
C(46)-C(45)-C(55)
C(55)-N(4)-C(56)
128.0(2)
127.1(2)
126.7(3)
93.63(9)
115.7(2)
124.2(2)
115.8(2)
C(24)-C(25)-N(1)-Zn(1) 13.3(4) C(45)-C(46)-N(3)-Zn(2) 10.4(4)
C(11)-C(12)-C(24)-C(34) 62.7(3) C(1)-C(2)-C(45)-C(55) 44.1(3)
Addition of hexane to the above-mentioned thf solution leads
to yellow crystals, which were investigated also by means of
Table 3. Selected Bond Lengths [Å] and Angles [deg] for 5
Zn(1)-Zn(2)
Zn(1)-N(1)
Zn(2)-N(3)
Zn(1)-C(58)
C(26)-N(1)
C(25)-C(24)
C(32)-N(2)
C(41)-N(3)
C(40)-C(39)
C(47)-N(4)
5.6038(8)
2.005(2)
2.033(2)
1.973(3)
1.421(4)
1.400(4)
1.322(4)
1.420(4)
1.395(4)
1.329(3)
Zn(1)-N(2)
Zn(2)-N(4)
Zn(2)-C(64)
N(1)-C(25)
C(24)-C(32)
N(2)-C(33)
N(3)-C(40)
C(39)-C(47)
N(4)-C(48)
2.017(3)
2.032(2)
1.972(3)
1.326(4)
1.407(4)
1.421(4)
1.336(4)
1.400(4)
1.422(4)
(6) (a) Dai, X.; Warren, T. H. J. Am. Chem. Soc. 2004, 126, 10085. (b)
Smith, J. M.; Lachicotte, R. L.; Holland, P. L. Chem. Commun. 2001, 1542.
(c) Smith, J. M.; Lachicotte, R. L.; Pittard, K. A.; Cundari, T. R.; Lukat-
Rodgers, G.; Rodgers, K. R.; Holland, P. L. J. Am. Chem. Soc. 2001, 123,
9222.
(7) McCleverty, J. A., Meyer, T. J., Eds. (Que, L., Jr., Tolman, W. B.,
Vol. Eds.) ComprehensiVe Coordination Chemistry II; Elsevier: Oxford
2004, Vol. 8, Chapters 8.1, 8.10, 8.13, 8.15, 8.19, 8.24.
(8) (a) Curtis, N. F. In ComprehensiVe Coordination Chemistry II;
McCleverty, J. A., Meyer, T. J., Eds. (Lever, A. B. P., Vol. Ed.); Elsevier:
Oxford, 2004; Vol. 1, Chapter 1.20. (b) Lee, S. Y.; Na, S. J.; Kwon, H. Y.;
Lee, B. Y.; Kang, S. O. Organometallics 2004, 23, 5382. (c) Lee, B. Y.;
Na, S. J.; Kwon, H. Y.; Lee, S. Y.; Na, S. J.; Han, S.; Yun, H.; Lee, H.
Park, Y.-W. J. Am. Chem. Soc. 2005, 127, 3031.
(9) (a) Bourget-Merle, L.; Hitchcock, P. B.; Lappert, M. F. J. Organomet.
Chem. 2004, 689, 4357. (b) Vitanova, D. V.; Hampel, F.; Hultzsch, K. C.
J. Organomet. Chem. 2005, 690, 5182. Also compare: Allen, S. D.; Moore,
D. R.; Lobkovsky, E. B.; Coates, G. W. J. Organomet. Chem. 2003, 683,
137, for an alternative approach, involving, however, oxo imido units.
(10) Pilz, M. F.; Limberg, C.; Ziemer, B. J. Org. Chem. 2006, 71, 4559.
(11) Sun, X. X.; Qi, C.-M.; Ma, S.-L.; Huang, H.-B.; Zhu, W.-x.; Liu,
Y.-C. Inorg. Chem. Commun. 2006, 9, 911.
N(1)-Zn(1)-N(2)
C(26)-N(1)-C(25)
C(25)-C(24)-C(32)
C(32)-N(2)-C(33)
N(3)-Zn(2)-C(64)
N(3)-C(40)-C(39)
C(39)-C(47)-N(4)
93.69(9)
118.7(2)
125.6(3)
118.2(3)
131.4(2)
127.1(2)
126.6(3)
N(1)-Zn(1)-C(58)
N(1)-C(25)-C(24)
C(24)-C(32)-N(2)
N(3)-Zn(2)-N(4)
C(41)N(3)-C(40)
C(40)-C(39)-C(47)
C(47)-N(4)-C(48)
131.8(2)
127.7(2)
125.6(3)
93.17(9)
115.0(2)
126.0(2)
116.2(2)
C(11)-C(12)-C(24)-C(25) 43.7(4) C(1)-C(2)-C(39)-C(40) 51.1(5)
ions are positioned almost exactly within the planes defined by
the diiminato units (C(24)-C(25)-N(1)-Zn(1): 2.4(4)°, C(39)-
C(40)-N(3)-Zn(2): 6.9(4)°), and the high symmetry within
the diiminato units again points to an effective delocalization
of the π-electron density. Only a slight folding can be observed
(12) (a) Moore, D. R.; Cheng, M.; Lobkovsky, E. B.; Coates, G. W. J.
Am. Chem. Soc. 2003, 125, 11911. (b) Chisholm, M. H.; Gallucci, J.;
Phomphrai, K. Inorg. Chem. 2002, 41, 2785.

3672 Organometallics, Vol. 26, No. 15, 2007
Pilz et al.
Figure 4. One monomeric unit of the tetrameric structure found for 6 (compare Figure 5). Hydrogen atoms are omitted for clarity.
Figure 5. View of the 32-membered ring of the tetramer 6. Hydrogen atoms are omitted for clarity.
X-ray diffraction. The result is shown in Figures 4 and 5 as
well as in Scheme 4. As intended, SO2 has inserted into all
Zn-C bonds so that ethylsulfinate anions were formed, one of
which is found in a bridging position between the two Zn ions
coordinated to a certain Xanthdim ligand (Figure 4). Within
these anions the S-O bond lengths are quite similar [S(2)-
O(4): 1.48(2) Å, S(2)-O(5): 1.53 (2) Å, S(4)-O(8): 1.498(9)
Å, S(4)-O(9): 1.49(2) Å], and the same is true for the Zn-O
bonds generated by their coordination, so that altogether the
intramolecularly bridging ethylsulfinate anions are bonded
symmetrically with delocalized charge. So far the principal
connectivities are identical to the ones observed for the products
of reactions between mononuclear diketiminato zinc alkyl
complexes and SO2. However, while in the latter case the
remaining alkylsulfinato anion forms a second bridge between
the two Zn centers, so that a dinuclear complex with a Zn-
(RSO2)2Zn entity is stabilized,3,5 the special design of the
Xanthdim ligand system seems to make intermolecular interac-
tions more favorable. In principle these could have led to a
coordination polymer, but a ring structure is formed here (see
Figure 5) involving four [Me2C6H3Xanthdim](Zn(O2SEt))2 units
that are linked via ethylsulfinato moieties to give the tetramer

Dinuclear Zinc Complexes
Organometallics, Vol. 26, No. 15, 2007 3673
Scheme 4
(2) Å, S(1)-O(10): 1.46(2) Å). Crystallographically the tet-
ramer consists of a dimer of a dimer. The xanthene backbones
are pointing outward and the N-bonded aryl rings are located
above and below the 32-membered ring. The Zn-Zn distances
within the individual [Me2C6H3Xanthdim]Zn2(µ-O2SEt) units are
quite similar (Zn(1)-Zn(2): 5.093(3) Å, Zn(3)-Zn(4): 5.182-
(3) Å) and much shorter than those found between such moieties
(Zn(1)-Zn(4): 5.61(1) Å, Zn(2)-Zn(3): 5.96(1) Å). All Zn
ions involved are coordinated tetrahedrally by the O atoms of
two ethylsulfinate ligands and the two N atoms of a diiminato
unit.
If an analogous reaction with SO2 is performed starting from
5, a yellow solid precipitates that cannot be redissolved or
purified sufficiently to allow unequivocal identification. It can
only be assumed that in the case of 5 a coordination polymer is
formed that is insoluble and thus not employable as a homo-
geneous catalyst. According to LDI-ToF-MS measurements (see
Figure 6), the product formed initially in the reaction between
4 and SO2 is monomeric; no aggregated species could be
detected in solution, but a degradation during the laser desorption
cannot be excluded (in fact the broad signals observed in the
NMR spectra and their behavior on variation of the temperature
point to multiple dynamic processes and species in solution).
Precipitation of an insoluble solid on aging of such solutions is
probably caused by an aggregation process comparable to the
one observed to occur spontaneously on treatment of 5 with
SO2. However, the rate of polymerization is much slower in
the case of the primary product obtained from 4, and 6 may be
regarded as an intermediate (accordingly, storing of a solution
of 6 also led to an insoluble solid).
Table 4. Selected Bond Lengths [Å] and Angles [deg] for 6a
Zn(1)-Zn(2)
Zn(1)-O(4)
Zn(1)-N(2)
Zn(2)-O(5)
Zn(4)-N(4)
S(2)-O(5)
5.093(3)
1.95(2)
1.95(2)
1.951(9)
1.95(2)
1.54(2)
1.58(2)
1.71(3)
1.561(9)
1.43(2)
1.39(2)
1.31(2)
1.46(2)
1.40(3)
1.42(2)
1.46(2)
1.942(8)
1.49(2)
1.936(8)
Zn(1)-O(3)
Zn(1)-N(1)
Zn(2)-O(6)
Zn(2)-N(3)
S(2)-O(4)
S(2)-C(64)
S(1)-O(10)
S(3)-O(6)
1.933(8)
1.953(9)
1.89(2)
1.950(9)
1.48(2)
1.79(2)
1.46(2)
1.39(2)
1.71(2)
1.33(2)
1.40(2)
1.47(2)
1.30(2)
1.40(3)
1.43(2)
1.934(9)
1.498(9)
1.98(2)
S(1)-O(3)
S(1)-C(62)
ul;1S(3)-O(7)
C(30)-N(1)
C(25)-C(24)
C(26)-N(2)
C(46)-N(3)
C(28)-C(27)
C(29)-N(4)
S(1)-O(10)
Zn(3)-O(8)
S(4)-O(9)
S(3)-C(127)
N(1)-C(25)
C(24)-C(26)
N(2)-C(38)
N(3)-C(28)
C(27)-C(29)
N(4)-C(54)
Zn(3)-O(7)
S(4)-O(8)
Zn(4)-O(9)
Zn(4)-O(10)
S(1)-O(3)-Zn(1)
Zn(1)-O(4)-S(2)
S(2)-O(5)-Zn(2)
Zn(2)-O(6)-S(3)
N(1)-Zn(1)-N(2)
O(3)-Zn(1)-N(1)
O(5)-Zn(2)-N(3)
117.6(6)
131.1(8)
125.3(5)
140.4(7)
97.9(4)
O(3)-Zn(1)-O(4)
O(4)-S(2)-O(5)
O(5)-Zn(2)-O(6)
O(4)-S(2)-C(64)
N(3)-Zn(2)-N(4)
O(3)-Zn(1)-N(2)
O(5)-Zn(2)-N(4)
99.2(5)
107.6(7)
98.6(5)
101.0(8)
97.4(4)
Catalytic Activity. The bis(â-diiminatozinc) complexes 4,
5, and 6 were tested in the copolymerization of cyclohexene
oxide (CHO) and CO2 (Table 5). Complex 4 displayed moderate
activity at 50 °C and 8 atm CO2 in neat CHO, and the polymer
product contained only 50% of carbonate linkages (Table 5,
entry 1; compare for instance the results4b for one of the most
active â-diketiminato zinc methoxide complexes under com-
parable conditions (entry 8): TOF ) 2290 h-1, 90% carbonate
linkages, Mw/Mn ) 1.09). Increasing the reaction temperature
to 75 °C improved the content of carbonate linkages to 77%
(Table 5, entry 2), but led to the formation of 3% cyclic
carbonate byproduct and broadening of the molecular weight
distribution. However, dilution of the reaction mixture with
toluene (1:1 v/v) improved the carbonate linkage content to 91%
and produced a polymer with a higher molecular weight and a
narrower polydispersity of 1.7 (Table 5, entry 3 vs entry 2).
116.5(4)
114.6(4)
117.5(5)
120.0(4)
C(24)-C(25)-N(1)-Zn(1)
C(3)-C(2)-C(24)-C(26)
2(2) C(27)-C(28)-N(3)-Zn(2)
55(2) C(13)-C(12)-C(27)-C(29) 47(2)
7(2)
a Underlined bonds are part of the 32-membered ring.
[{[Me2C6H3Xanthdim]Zn2(µ-O2SEt)}µ-O2SEt]4, 6. As can be seen
in Table 4, the intermolecularly bridging ethylsulfinato ligands
are partially asymmetric, as indicated by two significantly
different S-O and Zn-O bond distances within a certain Zn-
O-S-O-Zn moiety (e.g., S(3)-O(6): 1.39(2) Å, S(3)-O(7):
1.561(9) Å) and partially symmetric (e.g., (S(1)-O(3): 1.58-
Figure 6. Measured M+ peak in LDI-ToF-MS (left) and calculated spectra for [[Me2C6H3Xanthdim]Zn2(O2SEt)]+ (right).

3674 Organometallics, Vol. 26, No. 15, 2007
Pilz et al.
Table 5. Copolymerization of Cyclohexene Oxide (CHO)
oxide/CO2 copolymerization. While treatment with SO2 was
meant to activate the Zn-ethyl complexes for this type of
reaction, interestingly the Zn-ethyl complexes proved to be more
active than the Zn-O2SEt tetramer (considering that mononuclear
â-diketiminato Zn-ethyl complexes usually do not show any
activity before activation; current research will therefore also
deal with the CO2 reactivity of 4 and 5). A bimodal molecular
weight distribution hints at a cooperation of the two Zn centers
during the catalysis, as intended, but nevertheless the activity
observed is only moderate. One explanation could be that the
steric bulk within the reaction pocket overcompensates the
positive effect the cooperation could have on the efficiency of
the catalysis. Alternatively, the finding might indicate that the
accessibility of a cooperative interaction as shown in Scheme
1b is a prerequisite for a high activity. Future studies will address
this and other questions via further variation of the ligand
system.
and CO2 Catalyzed by Bis-â-diiminato Zinc Complexesa
%
% car- cyclic
T
t
yield
TOF bonate car-
Mw/
Mw Mn
entry cat. [°C] [h] [mg] TON [h-1] linkages bonate Mn
1
4
4
50 72
75 24
1600 666 9.3
450 170 7.1
250 90 3.8
450 220 6.1
800 380 5.3
300 110 1.15
750 285 5.9
382 2290
50
77
91
8
15
85
76
90
96
0.5 16 760 45 100 2.7
2
3
9
1
1
1
2
5970 23 300 3.9
10 300 17 900 1.7
3
4b 75 24
4
5
50 36
75 72
50 96
75 48
50 0.17
nd
nd
nd
nd
5
5
6
6
4840 15 800 3.3
4640 12 800 2.8
22 900 24 961 1.09
49 600 64 480 1.3
7
6
8c
9d
A
B
60
3
651 217
a Reaction conditions: 40 mmol of CHO, 20 µmol of cat ) 40 µmol of
Zn ([CHO]:[Zn] ) 1000:1); 8 atm of CO2. b Reaction mixture was diluted
with toluene (1:1 v/v). nd ) not determined. c Ref 4b. d Ref 5. A ) [{(2,6-
Me2C6H3)NC(Me)C(CN)C(Me)N(2,6-i-Pr2C6H3)}Zn(OMe)]2. B ) [{(2,6-
Et2C6H3)NC(Me)CHC(Me)N(2,6-Et2C6H3)}Zn(O2SEt)]2.
Experimental Section
General Procedures. Apart from the ligand synthesis, all
manipulations were carried out in a glovebox or by means of
Schlenk-type techniques involving the use of a dry argon atmo-
The fluorine-substituted complex 5 on the other hand produced
a polymer containing predominantly ether linkages, presumably
as a result of an increased Lewis acidity of the metal center
due to the electron-withdrawing 2,4-difluorophenyl substituents,
which supports the ring-opening polymerization of CHO. Most
notably, the ethylsulfinate complex 6 displayed catalytic activity
comparable to the ethyl zinc complex 4 only at 75 °C (Table 5,
entry 7 vs entry 2), whereas a significantly lower activity, but
higher carbonate linkage content was observed at 50 °C (Table
5, entry 6 vs entry 1, also compare entry 9, illustrating the results
obtained for a highly active â-diketiminato zinc ethylsulfinate
complex). The lower activity of 6 in comparison to 4 might
result from the higher degree of aggregation within complex 6,
which might hamper the initiation of the polymerization. At
higher temperature, a more rapid dissociation of the tetramer
on the other hand should lead to more active species. In all
cases the molecular weight distributions are bimodal, which
could be interpreted in terms of two effective mechanisms, one
of which probably involves two cooperating Zn centers as
anticipated (the competing one might comprise individual,
independent Zn centers without cooperation). However, probably
the steric bulk within the reaction pocket overcompensates the
positive effect of the cooperation supported by the dinucleating
ligand system, so that the latter does not increase the efficiency.
1
sphere. The H and 13C NMR spectra were recorded on a Bruker
AV 400 NMR spectrometer (1H 400.13 MHz) or on a Bruker DPX
300 MHz (1H 300.13 MHz, 13C 75.5 MHz) with CDCl3, C6D6, or
1
thf-d8 as solvent at 25 °C. The H and 13C NMR spectra were
calibrated against the residual proton and natural abundance 13C
resonances of the deuterated solvent (CDCl3 δH 7.24 ppm, δC 77.0
ppm; C6D6 δH 7.15 ppm δC 128.0 ppm; thf-d8 δH 3.58 ppm, δC
67.7 ppm). Microanalyses were performed on a Leco CHNS-932
elemental analyzer. Infrared (IR) spectra were performed using
samples prepared as KBr pellets on a Shimadzu FTIR-8400S
spectrometer. LDI-ToF mass spectra were recorded with an Applied
Biosystems Voyager-DE mass spectrometer without matrix using
a laser wavelength of λ ) 337 nm. The mass spectrum was
calculated with the program Molecular Weight Calculator, Version
6.31, by Matthew Monroe. Solvents were dried using a Braun
solvent purification system, degassed by a vacuum-freezing cycle,
and stored above 4 Å molecular sieves. SO2 gas was dried by
passing it through a short column of P4O10, condensed in a Young-
Schlenk tube at -30 °C that contained molecular sieve 4 Å, and
stored for at least 24 h before use. Molecular weights of the
polycarbonates were determined by GPC (Agilent 1100 Series using
RI detector, 3 SDV linear M 5 µm columns with 8 × 50 mm, 8 ×
300 mm, and 8 × 600 mm dimensions) versus polystyrene standards
using THF at 35 °C as eluent.
2,7-Bis(1,1-dimethylethyl)-9,9-dimethyl-4,5-bis(1-(2,4-difluoro-
phenylamino)-3-(2,4-difluorophenylimino)isopropenyl)xan-
thene, [F2C6H3Xanthdim]H2. The raw material (3.30 g, 6.74 mmol)
obtained starting from 3 in the Vilsmeier-type reaction (Scheme
1) according to ref 10 was placed in a round-bottomed flask,
dissolved in 150 mL of dry cyclohexane. Subsequently, 2,4-
difluoroaniline (7.00 mL, 68.78 mmol) and p-toluenesulfonic acid
(1.90 g, 9.98 mmol) were added, and the flask was connected to a
Dean Stark apparatus with a drying tube on top. The mixture was
heated under reflux for 5 days and then neutralized with a
concentrated NaHCO3 solution. The aqueous solution was extracted
several times with a mixture of diethyl ether/tetrahydrofuran in a
2:1 ratio, and after drying the combined organic phases with
magnesium sulfate all volatiles were removed from the filtrate under
vacuum. To the resulting amber-colored oil was added 30 mL of
absolute ethanol, and the mixture was heated to reflux. The yellow
precipitate formed was recrystallized several times from absolute
ethanol, and after sufficient drying in Vacuo [F2C6H3Xanthdim]H2
was obtained (3.00 g, 3.31 mmol; solid; 43.8% yield with respect
to 3 employed).
Conclusions
A novel ligand system has been used to prepare dinuclear
versions of previously known â-diketiminato zinc complexes
that had proved efficient catalysts for the epoxide/CO2 copo-
lymerization. Two ligand versions were employedsone with
electron-donating methyl substituents at the N-aryl donor
functions and the other containing electron-withdrawing F
atomssto produce two complexes, each of which comprises
two Zn-ethyl units. The latter can be readily converted into Zn-
ethylsulfinato moieties via reaction with SO2. Due to the
bidentate character of the ethylsulfinato ligands, they give rise
to aggregation reactions, which in the case of the more electron-
deficient, fluorinated complex leads to an insoluble coordination
polymer. In the case of the more electron-rich representative
aggregation leads to a soluble tetramer with an interesting
structural motif. This compound and the Zn-ethyl complexes
were investigated as potential catalysts for the cyclohexene

Dinuclear Zinc Complexes
Organometallics, Vol. 26, No. 15, 2007 3675
Table 6. Crystal Data for Compounds [F2C6H3Xanthdim]H2, 4, 5, and 6
F2C6H3Xanthdim]H2
[
(without solvent)
[
F2C6H3Xanthdim]H2
4
5
6
formula
fw
space group
a, Å
b, Å
c, Å
R, deg
â, deg
γ, deg
V, Å3
C53H46F8N4O
906.94
P21/n
15.584(2)
14.072(3)
20.847(3)
90
95.16(2)
90
4553(2)
4
C54H47Cl3F8N4O
1026.31
Pnna
16.073(5)
16.716(5)
18.411(7)
90
90
90
4947(3)
4
1.378
C65H78N4OZn2
1062.05
P1h
C73H86F8N4O5Zn2
1382.2
P21/n
15.955(2)
18.698(3)
22.696(2)
90
91.21(2)
90
6769(2)
4
1.356
Mo KR (λ ) 71073 Å)
0.784
C67H82N4O5.50S2Zn2
1226.23
P1h
16.975(3)
17.897(3)
24.587(4)
97.68(2)
101.63(2)
91.66(2)
7239(2)
16.806(3)
17.110(3)
20.756(3)
102.13(2)
93.16(2)
99.35(2)
5733(2)
4
Z
4
density, g/cm3
radiation
µ, mm-1
abs corr
1.323
1.230
1.125
Mo KR (λ ) 71073 Å) Mo KR (λ ) 71073 Å) Mo KR (λ ) 71073 Å)
0.102
integration
(X-Shape)
Mo KR (λ ) 71073 Å)
0.766
integration
(X-Shape)
52 224
0.259
0.881
integration
(X-Shape)
61 574
integration
(X-Shape)
38 787
integration
(X-Shape)
44 805
no. of reflns collected 19 302
no. of unique reflns
no. of obsd refln
5797 [R(int) ) 0.1715] 5405 [R(int) ) 0.1594] 19 517 [R(int) ) 0.0620] 11 915 [R(int) ) 0.0496] 26 555 [R(int) ) 0.1682]
1578
4449
12 322
8213
5322
no. of data; params
R1 , wR2
5797, 615
0.0591, 0.0822
5405; 353
0.1134, 0.2277
19 517; 1334
0.0390, 0.0762
11 915; 840
0.0433, 0.1039
26 555; 1457
0.0941, 0.2233
a
b
2
2
2
a R1 ) ∑||Fo| - |Fc||/∑|Fo|. b wR2 ) {∑[w(Fo - Fc )2]/∑[w(Fo )2]}1/2
.
Mp: > 300 °C. IR (KBr, cm-1): υ˜ 3424 (ν(NH)) (w), 3064
(w), 2956 (m), 2906 (m), 2867 (w), 1640 (vs), 1612 (m), 1595
(m), 1552 (vs), 1516 (vs), 1499 (vs), 1453 (s), 1436 (s), 1390 (s),
1362 (w), 1311 (s), 1263 (vs), 1230 (s), 1183 (m), 1140 (m), 1096
7.75 (s, 4 H, CHN). 13C{1H} NMR (thf-d8, 75 MHz, 25 °C): δ 0.0
(CH2(ethyl)), 12.4 (CH3(ethyl)), 15.2 (CH3(N-aryl)), 20.4 (CH3(N-aryl)),
31.6 (C(CH3)3), 33.2 (C(CH3)2), 34.6 (Cq), 35.4 (Cq), 104.1 (Cq),
121.2 (CH(1/8)), 121.9 (CH(N-aryl)), 126.3 (CH(N-aryl)), 126.5
(CH(N-aryl)), 127.8 (CH(3/6)), 129.5 (Cq), 130.4 (Cq), 130.4 (Cq), 137.9
(Cq), 145.0 (Cq), 146.7 (Cq), 152.9 (Cq), 161.9 (CHN). Anal. Calcd
for C65H78N4OZn2: C, 73.69; H, 7.42; N, 5.29 Found: C, 73.73;
H, 7.48; N, 5.23.
1
(m), 1066 (w), 997 (w), 962 (m), 851 (s). H NMR (CDCl3, 300
MHz, 25 °C): δ 1.36 (s, 18 H, C(CH)3)3), 1.76 (s, 6 H, C(CH)3)2),
6.53 (m, 4 H, CH(5A)), 6.72 (m, 4 H, CH(3A)), 6.93 (m, 4 H, CH(6A)),
4
4
7.11 (d, J(H,H) ) 2.4 Hz, 2 H, CH(3/6)), 7.39 (d, J(H,H) ) 2.4
Hz, 2 H, CH(1/8)), 7.97 (s, 2 H, CHN), 7.98 (s, 2 H, CHN), 12.66
(ps-t, J ) 4.7 Hz, 2 H, NH). 13C{1H} NMR (CDCl3 75 MHz, 25
°C): δ 31.5 (C(CH3)3, 33.5 (C(CH3)2), 34.5 (Cq), 34.8 (Cq), 104.4
(m, CH(3A)), 106.8 (Cq), 110.9 (d, 2J(C,F) ) 22.6 Hz, CH(5A)), 116.9
(d, 3J(C,F) ) 9.1 Hz, CH(6A)), 121.8 (CH(3/6)), 125.2 (CH(1/8)), 126.7
(Cq), 129.2 (Cq), 130.1 (m, C(1A)), 144.9 (Cq), 145.6 (Cq), 149.4
[
F2C6H3Xanthdim](ZnEt(thf))2, 5. 5 could be synthesized via the
same procedure as described for 4 but employing [F2C6H3Xanthdim]-
H2 as the starting material with quantitative yield. In contrast to 4,
5 is soluble in diethyl ether. Large crystals could be obtained by
slow evaporation of a tetrahydrofuran solution of 5.
IR (KBr, cm-1): υ˜ 3067 (w), 2962 (m), 2901 (w), 2867 (w),
1612 (m), 1598 (s), 1503 (s), 1474 (vs), 1447 (s), 1429 (m), 1376
(m), 1327 (s), 1261 (s), 1196 (m), 1140 (m), 1099 (m), 967 (m),
3
1
(CHN), 153.7, 153.8 (dd, J(C,F) ) 11.5 Hz, J(C,F) ) 254.2 Hz
C(2A)C(1A)-NH, 3J(C,F) ) 11.5 Hz, 1J(C,F) ) 250.1 Hz, C(2A) (1A)d
C
N), 148.6 (d(ps)-t, 3J(C,F) ) 4.53 Hz, 1J(C,F) ) 250.1 Hz, C(4A)).
19F NMR (CDCl3, 282 MHz, 25 °C): δ -121.94 (s, br, F2),
-115.02 (m, F4). EI-MS: m/z (%) 906.3 (27) [M+], 793.3 (13)
[M+ - C6H3F2], 767.3 (23) [M+ - C7H3NF2], 614.3 (22), 599.3
(21), 307.2 (18), 140.0 (19), 129.0 (100), 101.0 (26), 82.0 (20),
57.1 (43), 41.0 (26). EI-HRMS: m/z calcd for C53H46F8N4O,
906.3543; found, 906.3543. Anal. Calcd for C53H46F8N4O (906.94):
C, 69.93; H, 5.56; N, 5.72. Found: C, 69.72; H, 5.85; N, 5.51.
1
847 (m), 810 (m). H NMR (thf-d8, 400 MHz, 25 °C): δ 0.11 (q,
3J(H,H) ) 8.0 Hz, 4 H, CH2(ethyl)), 1.04 (t, 3J(H,H) ) 8.0 Hz, 6 H,
CH3(ethyl)), 1.31 (s, 18 H, C(CH3)3), 1.69 (s, 6 H, C(CH3)2), 6.69
4
(m, 4 H, CH(5A)), 6.92 (m, 4 H, CH(3A)), 7.01 (d, J(H,H) ) 2.4
4
Hz, 2 H, CH(3/6)), 7.05 (m, 4 H, CH(6A)), 7.35 (d, J(H,H) ) 2.4
Hz, CH(1/8)), 7.9 (ps-d, J(H,H) ) 2.0 Hz, 4 H, CHN). 13C NMR
(thf-d8, 75 MHz, 25 °C): δ -1.39 (t, 5J(C,F) ) 5.0 Hz, CH2(ethyl)),
13.0 (CH3(ethyl)), 31.5 (C(CH3)3), 33.8 (C(CH)3)2), 34.7 (Cq), 35.2
[
Me2C6H3Xanthdim](ZnEt)2, 4. [Me2C6H3Xanthdim]H2 (300 mg,
2
(Cq), 104.7 (ps-t, J(C,F) ) 25.8 Hz, CH(3A)), 104.9 (Cq), 111.9
0.342 mmol) was dissolved in tetrahydrofuran, and diethyl zinc
(450 mg, 3.64 mmol) was added. After stirring the resulting bright
honey-colored solution overnight all volatiles were removed under
vacuum and the remaining yellow oil was dissolved in a small
amount of diethyl ether. The product precipitated, and after
removing all volatile compounds under vacuum again pure 4 was
obtained in quantitative yield. Crystals suitable for X-ray analysis
could be grown by storing a suspension of 4 in diethyl ether for 3
weeks.
(dd, 2J(C,F) ) 22.4 Hz, 4J(C,F) ) 3.7 Hz, CH(5A)), 121.5 (CH(3/6)),
2
2
126.3 (dd, J(C,F) ) 9.8 Hz, J(C,F) ) 3.0 Hz CH(6A)), 127.0
(CH(1/8)), 129.5 (Cq), 129.7 (Cq), 137.6 (dd, 2J(C,F) ) 9.7 Hz,
4J(C,F) ) 3.7 Hz, C(1A)), 145.4 (br, 2 × Cq), 155.6 (dd, 1J(C,F) )
246.2 Hz, J(C,F) ) 11.7 Hz, C(2A)), 159.6 (dd, J(C,F) ) 246.2
Hz, 3J(C,F) ) 11.7 Hz, C(4A)), 162.8 (ps-d, CHN). 19F NMR
(thf-d8, 282 MHz, 25 °C): δ -123.42 (s, br, F2), -117.30 (m,
F4). Anal. Calcd for C65H70F8N4O3Zn2‚(C4H8O)2: C, 63.43; H, 6.27;
N, 4.05. Found: C, 63.01; H, 6.37; N, 4.15.
3
1
IR (KBr, cm-1): υ˜ 2962 (m), 2855 (m), 1599 (s), 1576 (s), 1557
(m), 1460 (vs), 1440 (vs), 1382 (s), 1365 (s), 1309 (vs), 1261 (vs),
1245 (vs), 1224 (s), 1184 (s), 1162 (s), 1124 (s), 1092 (s), 1057
(s), 1016 (s), 987 (s), 955 (m), 877 (w), 857 (m), 797 (s), 781 (vs),
746 (m), 726 (m), 685 (m), 668 (w), 651 (w), 609 (w), 559 (w),
[
Me2C6H3Xanthdim](Zn(SO2Et))2, 6. In a Young-Schlenk tube
4 (364 mg, 0.343 mmol) was dissolved in 5 mL of tetrahydrofuran.
The tube was connected to a high-vacuum apparatus, cooled to -30
°C, and evacuated. An excess of dry SO2 gas was condensed in
portions to the solution, leading to a color change to red-orange.
After stirring for 30 min at -30 °C the mixture was warmed to
room temperature and all volatiles were removed under vacuum,
yielding quantitatively a yellow solid with the stoichiometry
1
504 (w), 493 (w). H NMR (thf-d8, 300 MHz, 25 °C): δ 0.03 (q,
3
3J(H,H) ) 8.1 Hz, 4 H, CH2), 0.846 (t, J(H,H) ) 8.1 Hz, 6 H,
CH3), 1.32 (s, 18 H, C(CH3)3), 1.67 (s, 6 H, C(CH3)2), 2.11 (ps-d,
4
12 H, CH3), 6.74-6.84 (m, 12 H, CH(N-aryl)), 7.04 (d, J(H,H) )
4
2.4 Hz, 2 H, CH(3/6)), 7.34 (d, J(H,H) ) 2.4 Hz, 2 H, CH(1/8)),
[
Me2C6H3Xanthdim](Zn(SO2Et))2. Crystals suitable for X-ray analyses

3676 Organometallics, Vol. 26, No. 15, 2007
Pilz et al.
could be obtained by slow evaporation of the solvent from a solution
of 6 in a mixture of tetrahydrofuran (5 mL) and hexane (20 mL).
IR (KBr, cm-1): υ˜ 3063 (w), 2961 (s), 2911 (m), 2868 (w),
1602 (s), 1576 (m), 1481 (vs), 1463 (vs), 1442 (s), 1374 (m), 1321
(vs), 1254 (w), 1221 (m), 1072 (m), 989 (νas(SO)) (s), 963 (νs-
and quickly frozen to -93 °C. Centered reflections were refined
by least-squares calculations to indicate the unit cells. Unit cell
and collection parameters for the complexes are listed in Table 6.
Diffraction data were collected in the appropriate hemispheres and
under the conditions specified also in Table 6. Two crystal structure
determinations were performed for [F2C6H3Xanthdim]H2: the first
one was performed for a crystal containing additional disordered
chloroform solvent molecules that led to relatively high R values.
In the second case the crystal did not include any solvent molecule,
so that the R values are significantly lower. However, the resolution
was higher in the former case (it thus forms the basis for the
structure discussion), so that both results are listed in Table 6.
Solution and Refinement of the Structures of [F2C6H3Xanthdim]-
H2, 4, 5, and 6. The structures were solved by direct methods
(program: SHELXS-97)13 and refined versus F2 (program: SHELXL-
97)14 with anisotropic temperature factors for all non-hydrogen
atoms. The final residuals are indicated in Table 6. The crystal-
lographic data (apart from structure factors) of [F2C6H3Xanthdim]-
H2, 4, 5, and 6 were deposited at the Cambridge Crystallographic
Data Center as supplementary publication nos. CCDC 625190/
637291, CCDC 625192, CCDC 625191, and CCDC 625193. Copies
of the data can be ordered free of charge from the following address
in Great Britain: CCDC, 12 Union Road, Cambridge CB2 1EZ
(fax: (+44)1223-336-033; e-mail: deposit@ccdc.cam.ac.uk).
1
(SO)) (s), 856 (w), 782 (m), 721 (w), 670 (w). H NMR (C6D6,
300 MHz, 25 °C): δ 0.47 (m, 3 H, CH3), 0.92 (s, br, 3 H, CH3),
1.35 (s, 18 H, C(CH3)3), 1.65 (s, 6 H, C(CH3)2), 1.96-2.51 (m, br,
12 H, CH3(N-aryl), 4 H, CH2(ethyl)), 6.72-7.11 (m, br, 8 H, CH(N-aryl)),
7.38 (s, br, CH(3/6, 1/8)), 7.45-7.72 (m, br, 4 H, CH(N-aryl)), 7.91-
8.10 (m, br, 4 H, CHN). 13C NMR (C6D6, 75 MHz, 25 °C): δ 5.7
(br, CH3(ethyl)), 15.3 (br, CH(N-aryl)), 20.7 (br, CH(N-aryl)), 31.5
(C(CH3)3), 34.2 (C(CH3)2), 34.9 (2 × Cq), 54.0 (br, CH2(ethyl)), 101.6
(Cq), 121.0 (CH(3/6)), 123.8 (CH(N-aryl)), 123.9 (CH(N-aryl)), 126.3
(CH(N-aryl)), 128.1 (CH(1/8)), 129.5 (Cq), 130.4 (Cq), 137.6 (br, Cq),
144.5 (Cq), 145.7 (br, Cq), 151.7 (br, Cq), 164.5 (br, CHN). MALDI-
MS [C65H78N4O5S2Zn2] m/z (%): 1093 (83%), 1099.4 (100%),
1101.0 (76%), isotopic pattern for M+ - C2H5SO2. Anal. Calcd
for C65H78N4O5S2Zn2: C, 65.59; H, 6.61; N, 4.71; S, 5.39. Found:
C, 65.14; H, 6.84; N, 4.61; S, 5.27.
General Procedure for Copolymerization of Cyclohexene
Oxide and CO2 Using Catalysts 4, 5, and 6. The copolymerization
reactions were performed in a Bu¨chi tinyclave reactor. Cyclohexene
oxide (3.92 g, 40 mmol) was added to the reactor in the glovebox
under an inert atmosphere of argon. The catalyst (20 µmol,
corresponding to 40 µmol of Zn) was placed on a glass boat in
order to prevent premature mixing. Then the reactor was closed
and transferred to a Schlenk line, where the atmosphere was
exchanged for 8 atm CO2. After saturation of the cyclohexene oxide
with CO2 was complete, the solution and the catalyst were mixed
by tilting the reactor. The system was then heated to the desired
temperature. The pressure of CO2 was maintained at 8 atm. After
the specified reaction time the heating was discontinued and the
reactor was cooled in an ice bath to 5-10 °C, then the pressure
was released and the reaction mixture was quenched by addition
of CH2Cl2 and MeOH. All volatiles were evaporated in Vacuo at
80 °C to obtain the polymer product as a crispy solid.
Acknowledgment. We are grateful to the Fonds der
Chemischen Industrie, the BMBF, and the Dr. Otto Ro¨hm
Geda¨chtnisstiftung for financial support. K.C.H. is a DFG Emmy
Noether fellow (2001-2007). We also would like to thank P.
Neubauer for crystal structure analyses and Prof. J. Okuda and
Dr. K. Beckerle (RWTH Aachen) for GPC measurements.
Supporting Information Available: This material is available
free of charge via the Internet at http://pubs.acs.org.
OM070221E
(13) Sheldrick, G. M. SHELXS-97, Program for Crystal Structure
Refinement; University of Go¨ttingen, 1997.
(14) Sheldrick, G. M. SHELXL-97, Program for Crystal Structure
Solution; University of Go¨ttingen, 1997.
X-ray Diffraction Data Collection for [F2C6H3Xanthdim]H2,
4, 5, and 6. Single crystals of the compounds were selected in a
cold stream of nitrogen. They were then mounted on top of a fiber