Bis-salicylaldiminato complexes of zinc. Examination of the catalyzed epoxide/co2 copolymerization

Article abstract of DOI:10.1021/ic0006403

A series of salicylaldimine ligands of the general formula (NR²C₇H_5-x(R¹)_xOH) [x = 1 or 2; R¹ = Me, ^tBu, Cl, OMe; R² = 2,6-ⁱPr₂C₆H₃, or 3,5-(CF₃)₂C₆H₃] have been synthesized and characterized via ¹H and ¹³C NMR, elemental analysis, and X-ray crystallography. The concomitant series of zinc bis(salicylaldiminato) complexes of the general formula (NR²C₇H_5-x(R¹)_xO) ₂Zn have been synthesized and characterized in the solid state by X-ray crystallography. All complexes crystallized as four coordinate monomers with distorted tetrahedral geometry about the zinc center. The O-Zn-O angles range between 105 and 112.5°, and the N-Zn-N bond angles were more obtuse spanning the range 122.9-128.9°. The only deviation from distorted tetrahedral geometry occurred when R² = 3,5-(CF₃)₂C₆H₃ which crystallized as a distorted trigonal bipyramidal dimeric species with O_ax-Zn-O_ax bond angles of 165.00(15)°. The equatorial angles approach 120° except for the N_eq-Zn-N_eq angle of 110.54(16)° which is attributed to the strain of the bridging ligands. The zinc bis(salicylaldiminato) complexes showed varying activities as catalyst precursors for the copolymerization of CO₂ and cyclohexene oxide. Activation is proposed to occur via CO₂ insertion in the phenolic Zn-O bond with simultaneous ring-opening resulting in a site for epoxide binding. The difference in activity has been ascribed to the different steric/electronic effects provided by the R¹ and R² substituents on the various steps of the copolymerization mechanism. The activity of the zinc bis(salicyaldiminato) catalyst precursors (<16 g·polym/g·Zn/hr) were similar to the activities of the previously reported zinc phenoxide complexes for this reaction; however, unlike the zinc phenoxide catalysts, the zinc bis(salicylaldiminato) complexes produced poly(cyclohexane carbonate) with greater than 99% carbonate linkages.

Full text of DOI:10.1021/ic0006403

986
Inorg. Chem. 2001, 40, 986-993
Bis-Salicylaldiminato Complexes of Zinc. Examination of the Catalyzed Epoxide/CO2
Copolymerization
Donald J. Darensbourg,* Patrick Rainey, and Jason Yarbrough
Department of Chemistry, Texas A&M University, P.O. Box 30012, College Station, Texas 77842
ReceiVed June 14, 2000
t
A series of salicylaldimine ligands of the general formula (NR2C7H5-x(R1)xOH) [x ) 1 or 2; R1 ) Me, Bu, Cl,
1
OMe; R2 ) 2,6-iPr2C6H3, or 3,5-(CF3)2C6H3] have been synthesized and characterized via H and 13C NMR,
elemental analysis, and X-ray crystallography. The concomitant series of zinc bis(salicylaldiminato) complexes
of the general formula (NR2C7H5-x(R1)xO)2Zn have been synthesized and characterized in the solid state by X-ray
crystallography. All complexes crystallized as four coordinate monomers with distorted tetrahedral geometry
about the zinc center. The O-Zn-O angles range between 105 and 112.5°, and the N-Zn-N bond angles were
more obtuse spanning the range 122.9-128.9°. The only deviation from distorted tetrahedral geometry occurred
when R2 ) 3,5-(CF3)2C6H3 which crystallized as a distorted trigonal bipyramidal dimeric species with Oax-
Zn-Oax bond angles of 165.00(15)°. The equatorial angles approach 120° except for the Neq-Zn-Neq angle of
110.54(16)° which is attributed to the strain of the bridging ligands. The zinc bis(salicylaldiminato) complexes
showed varying activities as catalyst precursors for the copolymerization of CO2 and cyclohexene oxide. Activation
is proposed to occur via CO2 insertion in the phenolic Zn-O bond with simultaneous ring-opening resulting in
a site for epoxide binding. The difference in activity has been ascribed to the different steric/electronic effects
provided by the R1 and R2 substituents on the various steps of the copolymerization mechanism. The activity of
the zinc bis(salicyaldiminato) catalyst precursors (<16 g‚polym/g‚Zn/hr) were similar to the activities of the
previously reported zinc phenoxide complexes for this reaction; however, unlike the zinc phenoxide catalysts, the
zinc bis(salicylaldiminato) complexes produced poly(cyclohexane carbonate) with greater than 99% carbonate
linkages.
Introduction
Recently, much of our attention has been directed at the
syntheses and characterization (both in solution and the solid-
state) of monomeric zinc derivatives.1,2 These efforts are
motivated by the desire to uncover effective catalysts for the
coupling of carbon dioxide with a wide range of epoxides.3,4 In
this report we wish to describe the preparation and solid-state
structures of a variety of zinc complexes derived from the ligand,
salicylaldimine. A specific example of one such cognate is
illustrated in I. These ligands are particularly attractive because
of their ease of preparation, which readily allows for varying
their steric and electronic properties.5 Indeed, Grubbs and co-
Herein, we divulge the synthesis and characterization of a
workers have a short while ago exploited these monoanionic
chelating salicylaldimine ligands in the synthesis of a series of
nickel(II) complexes which were shown to catalyze the polym-
erization of ethylene at low pressures and temperatures.6 On
the basis of the closely related studies of Coates and co-workers,
complexes of the type (salicylaldiminato)Zn(OR), where R )
Me or COMe, should possess the capability to serve as excellent
catalysts for the copolymerization of CO2 and epoxides.7
group of bis-salicylaldiminato complexes of zinc which exhibit
significant steric and electronic diversity. Included in this writing
are the X-ray structures of several of the protonated ligand
precursors, as well as of the zinc complexes. Thus far we have
been unsuccessful in synthesizing the desired 1:1 complexes.
However, these bis(salicylaldiminato)zinc derivatives have been
found to display catalytic activity for the alternating copolym-
erization of CO2 and cyclohexene oxide.
(1) Darensbourg, D. J.; Holtcamp, M. W.; Struck, G. E.; Zimmer, M. S.;
Niezgoda, S. A.; Rainey, P.; Robertson, J. B.; Draper, J. D.;
Reibenspies, J. H J. Am. Chem. Soc. 1999, 121, 107.
(2) Darensbourg, D. J.; Zimmer, M. S.; Rainey, P.; Larkins, D. L. Inorg.
Chem. 2000, 39, 1578.
Experimental Section
Methods and Materials. Unless otherwise specified, all syntheses
and manipulations were carried out on a double manifold Schlenk
vacuum line under an atmosphere of argon or in an argon filled
glovebox. Glassware was flamed out thoroughly prior to use. Toluene,
tetrahydrofuran, diethyl ether, and pentane were freshly distilled from
(3) Rokicki, A.; Kuran, W. J. Macromol. Sci., ReV. Macromol. Chem.
1981, C21, 135.
(4) Beckman, E. Science 1999, 283, 946.
(5) Grubbs, R. H. Science 2000, 287, 460.
(6) Wang, C.; Friedrich, S.; Younkin, T. R.; Li, R. T.; Grubbs, R. H.;
Bansleben, D. A.; Day, M. W. Organometallics 1998, 17, 3149.
(7) Cheng, M.; Lobkovsky, E. B.; Coates, G. W. J. Am. Chem. Soc. 1998,
120, 11018.
10.1021/ic0006403 CCC: $20.00 © 2001 American Chemical Society
Published on Web 01/27/2001

Bis-Salicylaldiminato Complexes of Zinc
Inorganic Chemistry, Vol. 40, No. 5, 2001 987
sodium benzophenone. 3,5-Dichlorosalicylaldehyde, 3,5-di-tert-butyl-
2-hydroxybenzaldehyde, 2-hydroxy-5-methyl-benzaldehyde, 2-hydroxy-
5-methoxy-benzaldehyde, and 2,5-(trifluoro)methylaniline were all
purchased from Aldrich and used as received. 2,6-Di-isopropylaniline
(90% tech) was purchased from Aldrich and was distilled under vacuum
(4 mm Hg) onto molecular sieves prior to use. Bone dry carbon dioxide
supplied in a high-pressure cylinder equipped with a liquid dip-tube
Synthesis of C16H11NOF6, 1e. 2-Hydroxy-3-methyl-benzaldehyde
(1.1 g, 8.4 mmol) and 3,5-(trifluoro)methyl-aniline (1.9 g, 8.4 mmol)
in 30 mL of ethanol were refluxed for 2 h to afford 2.2 g (76% yield)
of the title compound as yellow crystals when the reaction solution
1
was cooled to 0 °C. IR ν(CN) ) 1614 cm-1. H NMR (C6D6 solvent
298 K) δ 2.28 (s, 3H, CH3), 6.65 (t, 1H, aryl), 6.75 (d, 1H, aryl), 6.91
(d, 1H, aryl), 7.01 (s, 1H, aryl), 7.13 (s, 2H, aryl), 7.56 (s, 1H, ketimine-
H), 12.31 (s, 1H, OH).
1
was purchased from Scott Specialty Gases. H and 13C NMR spectra
were acquired on Unity+ 300 MHz and VXR 300 MHz supercon-
ducting NMR spectrometers. The operating frequency for 13C experi-
ments was 75.41 MHz for the 300 MHz instruments. Infrared spectra
were recorded on a Mattson 6021 FT-IR spectrometer with DTGS and
MCT detectors. Analytical elemental analyses services were provided
by Canadian Microanalytical Services Ltd.
General Procedure for the Preparation of Salicylaldimine
Ligands (1a-e). The condensation of salicyaldehydes with aromatic
amines was carried out by refluxing the two reactants in ethyl alcohol
(100%) over molecular sieves for 2 h. Upon cooling the resulting
reaction solution to 0 °C, yellow crystals precipitated from solution
except in the case of 1b, vide infra. The solid product was filtered and
washed with cold ethanol and then dried in vacuo to give the desired
salicylaldimine ligand in excellent yields. Any modifications are
described below for each reaction.
Synthesis of Zn(C20H24NO)[N(Si(CH3)3)2], 2a. A 5 mL yellow
solution of complex 1a (0.15 g, 0.52 mmol) in THF was added via
cannula to a 5 mL colorless solution of Zn[N(Si(CH3)3)2]2 (0.20 g, 0.52
mmol) dissolved in THF. The color of the solution changed to chartreuse
immediately upon addition of 1a to the zinc amide solution. The reaction
solution was allowed to stir at ambient temperature for 2h. All volatiles
were removed via vacuum to produce a yellow powder 0.24 g, (87%
1
yield). H NMR (C6D6 solvent 298 K) δ 0.28 (s, 18H, N(Si(CH3)3)2),
0.97 (d, 6H, CH(CH3)2), 1.29 (d, 6H, CH(CH3)2), 1.39 (m, THF), 2.45
(s, 3H, CH3), 3.05 (sept, 2H, CH(CH3)2), 3.61 (m, THF), 6.50-7.11
(m, 6H, aryl), 7.84 (s, 1H, ketimine-H). Anal. Found (calcd): C, 59.48
(60.03); H, 8.02 (8.14); N, 5.17 (5.39).
Synthesis of Zn(C27H38NO)[N(Si(CH3)3)2], 2b. A 5 mL yellow,
THF solution of complex 1b (0.10 g, 0.25 mmol) was added via cannula
to a 5 mL THF solution of Zn[N(Si(CH3)3)2]2 (0.10 g, 0.25 mmol).
The solution color changed to chartreuse immediately upon addition
of 1b to the zinc amide solution. All volatiles were removed via vacuum
to produce a yellow powder 0.13 g, (81% yield). 1H NMR (C6D6 solvent
298 K) δ 0.90 (d, 6H, CH(CH3)2), 1.22 (d, 6H, CH(CH3)2), 1.29 (s,
9H, C(CH3)3), 1.36 (m, THF), 1.71 (s, 9H, C(CH3)3), 2.99 (sept, 2H,
CH(CH3)2), 3.57 (m, THF), 6.85 (d, 1H, aryl), 7.06 (m, 3H, aryl), 7.78
(d, 1H, aryl), 7.97 (s, 1H, ketimine-H). Anal. Found (calcd): C, 63.72
(64.10); H, 9.02(9.13); N, 4.45(4.53).
Synthesis of C20H25NO, 1a. Upon refluxing 2-hydroxy-5-methyl-
benzaldehyde (2.50 g, 18.4 mmol) and 2,6-diisopropylaniline (3.26 g,
18.4 mmol) in 30 mL of ethanol (100%) over molecular sieves afforded
4.8 g (93% yield) of the title compound as yellow crystals. A drop of
trifluoroacetic acid was used to accelerate the condensation reaction.
1H NMR (C6D6 solvent 298 K) δ 1.05 (d, 12H, CH(CH3)2), 2.34 (s,
3H, 3-CH3), 3.01 (sept, 2H, CH(CH3)2), 6.7 (t, 1H, aryl), 6.9 (d, 1H,
aryl), 7.03 (d, 1H, aryl), 7.1 (s, 2H, aryl), 7.96 (s, 1H, ketimine CH),
13.2 (s, 1H, OH). 13C NMR (C6D6 solvent 298 K) δ 16.2 (3-CH3),
24.3 (CH(CH3)2), 29.0 (CH(CH3)2), 119.4, 124.6, 126.2, 131.2, 134.7,
139.4, 147.9 (aryl), 160.7 (phenolic ipso-C), 167.9 (ketimine C). Anal.
Found (calcd.): C, 81.03, (81.31); H, 8.49, (8.53); N, 4.64, (4.74). IR
Synthesis of Zn(C20H24NO)(CH2CH3), 3a. Complex 1a (0.38 g,
1.3 mmol) was dissolved in 5 mL of hexane to give a yellow-orange
solution. This solution was then cannulated onto a 10 mL colorless
solution of a Zn(CH2CH3)2 (0.79 g, 6.4 mmol) in hexane to give a
chartreuse solution which was allowed to stir for 2 h. Evacuation of
(ethanol) ν(CN) ) 1621 cm-1
.
1
Synthesis of C27H39NO, 1b. 2-Hydroxy-3,5-di-tert-butylbenzalde-
hyde (1.5 g, 6.4 mmol) and 2,6-diisopropylaniline (1.1 g, 6.4 mmol)
in 30 mL of refluxing ethanol (100%) over molecular sieves produced
an oily, yellow residue. This oily product was miscible with pentane.
A drop of trifluoroacetic acid was used to accelerate the condensation
reaction. Separation of the title compound was achieved using a silica
gel column leading to a bright yellow solid after removal of pentane
all volatiles afforded a yellow powder 0.33 g (65% yield). H NMR
(C6D6 solvent 298 K) δ 0.45 (quart., 2H CH2CH3), 1.03 (d, 6H, CH-
(CH3)2), 1.20 (t, 3H, CH2CH3), 1.33 (d, 6H, CH(CH3)2), 1.88 (s, 3H,
CH3), 3.42 (br, 2H, CH(CH3)2), 6.49 (t, 1H, aryl), 6.72 (dd, 1H, aryl),
6.93 (d, 1H, aryl), 7.13 (m, 3H, aryl), 7.94 (s, 1H, ketimine-H).
Synthesis of Zn(C27H38NO)(CH2CH3), 3b. Complex 1b (0.30 g,
0.76 mmol) was dissolved in 5 mL of hexane to give a yellow solution
which was added to a 10 mL colorless solution of Zn(CH2CH3)2 (0.47
g, 3.8 mmol) in hexane to produce a chartreuse solution. Evacuation
of all volatiles afforded a yellow powder 0.28 g (76% yield). 1H NMR
(C6D6 solvent 298 K) δ 0.40 (quart., 2H, CH2CH3), 0.90 (d, 6H, CH-
(CH3)2), 1.196 (s, 18H, C(CH3)), 1.23 (t, 3H, CH2CH3), 1.62 (s, 18H,
C(CH3)), 2.83 (sept, 2H, CH(CH3)2), 6.85 (d, 1H, aryl), 7.05 (m, 3H,
aryl), 7.79 (d, 1H, aryl), 7.95 (s, 1H, ketimine-H).
1
1.9 g (76% yield). IR ν(CN) ) 1623 cm-1. H NMR (C6D6 solvent
298 K) δ 1.24 (d, 12H, CH(CH3)2), 1.40 (s, 9H, (C(CH3)3), 1.56 (s,
9H, C(CH3)3), 3.12 (sept, 2H, CH(CH3)2), 7.21 (m, 2H, aryl), 7.32 (s,
1H, aryl), 7.60 (d, 2H, aryl), 8.35 (s, 1H, ketimine-H), 13.24 (s, 1H,
OH). 13C NMR (C6D6 solvent 298 K) δ 23.91 (s, C(CH3)3), 28.91 (s,
C(CH3)3), 30.14 (s, CH(CH3)2), 31.99 (s, C(CH3)3), 34.69 (s, CH(CH3)3),
35.89 (s, C(CH3)3), 118.96, 123.84, 126.13, 127.48, 138.15, 139.43,
141.35, 147.49 (s, aryl-C), 159.57 (s, phenolic ipso-C), 168.81 (s,
ketimine-C). Anal. Found (calcd): C, 82.38 (82.31); H, 10.00 (9.90);
N, 3.53 (3.51).
Synthesis of C19H21NOCl2, 1c. Employing similar reaction condi-
tions as for complex 1a 2-hydroxy-3,5-dichloro-benzaldehyde (2.00 g,
10.5 mmol) and 2,6-diisopropylaniline (1.86 g, 10.5 mmol) were
refluxed in 30 mL of ethanol (100%) for 2h. Upon cooling, the reaction
solution to 0 °C afforded bright yellow crystals 3.21 g (87% yield) of
1c. IR ν(CN) ) 1624 cm-1. 1H NMR (C6D6 solvent 298 K) δ 1.01 (d,
12H, CH(CH3)2), 2.80 (sept, 2H, CH(CH3)2), 6.65 (s, 1H, aryl), 7.05-
7.15 (m, 3H, aryl), 7.50 (s, 1H, ketimine-H), 14.32 (s, 1H, OH). Anal.
Found (calcd): C, 64.60 (65.15); H, 5.88 (6.04).
Synthesis of Zn(C20H24NO)2, 4a. Complex 1a (0.31 g, 1.0 mmol)
was dissolved in 5 mL of pentane to give a yellow solution. This
solution was added via cannula onto a clear and colorless solution of
Zn[N(Si(CH3)3)2]2 (0.20 g, 0.50 mmol) in pentane to afford a chartreuse
solution. The reaction was allowed to stir for 1.5 h after which all
volatiles were removed via vacuum to give a bright yellow powder
0.30 g (89% yield). Complex 4a could be further purified by
recrystallization as described in the experimental X-ray structural studies
1
section. H NMR (C6D6 solvent 298 K) δ 0.81 (d, 12H, CH(CH3)2),
1.01 (d, 12H, CH(CH3)2), 2.21 (s, 6H, CH3), 3.23 (br, 4H, CH(CH3)2),
6.42 (t, 2H, aryl), 6.7 (dd, 2H, aryl), 6.95-7.10 (m, 8H, aryl), 7.83 (s,
2H, ketimine-H). 13C NMR (C6D6 solvent 298 K) δ 23.42 (s, CH3),
25.34 (s, CH(CH3)2), 28.83 (s, CH(CH3)2), 114.87, 117.41, 124.57,
127.10, 128.39, 128.71, 132.45, 134.70, 136.66, 142.25, 147.38 (s,
aryl-C), 172.23 (s, phenolic ipso-C), 175.36 (s, ketimine-C). Anal.
Found (calcd): C, 72.09 (73.44); H, 7.25 (7.40); N, 4.12 (4.28).
Synthesis of Zn(C27H38NO)2, 4b. Complex 1b (0.20 g, 0.50 mmol)
dissolved in 5 mL of THF to give a yellow solution was added via
cannula onto a colorless 5 mL THF solution of Zn[N(Si(CH3)3)2]2 (0.10
g, 0.25 mmol) to produce a chartreuse solution. Evacuation of all
volatiles afforded 0.18 g (85% yield) of a bright yellow powder.
Synthesis of C20H25NO2, 1d. Similarly, 2-hydroxy-5-methoxy-
benzaldehyde (2.50 g, 16.4 mmol) and 2,6-diisopropylaniline (2.91 g,
16.4 mmol) refluxed in 30 mL of ethanol provided 4.2 g (82% yield)
of bright yellow crystals upon cooling the reaction solution to 0 °C.
1
IR ν(CN) ) 1629 cm-1. H NMR (C6D6 solvent 298 K) δ 1.06 (d,
12H, CH(CH3)2), 3.00 (sept, 1H CH(CH3)2), 3.27 (s, 3H, OCH3), 6.61
(d, 1H, aryl), 6.78 (dd, 1H, aryl), 7.04 (s) 7.10 (s, 4H, aryl), 7.85 (s,
1H, ketimine-H), 12.72 (s, 1H, OH). Anal. Found (calcd): C, 76.99
(77.14); H, 8.07 (8.09), N, 4.35 (4.50).

988 Inorganic Chemistry, Vol. 40, No. 5, 2001
Darensbourg et al.
Table 1. Crystallographic Data for Complexes 1a, 1c, and 1e
Complex 4b could be further purified by recrystallization as described
1
in the experimental X-ray structural studies section. H NMR (C6D6
1a
1c
1e
solvent 298 K) δ 0.90 (d, 6H, CH(CH3)2), 1.19 (s, 18H, CH(CH3)2),
1.22 (d, 18H, C(CH3)3), 1.62 (s, 18H, C(CH3)3), 2.63 (sept, 2H,
CH(CH3)2), 3.94 (sept, 2H, CH(CH3)2), 6.73 (d, 2H, aryl), 6.92 (m,
2H, aryl), 7.00-7.04 (m, 4H, aryl), 7.65 (d, 2H, aryl), 7.85 (s, 2H,
ketimine-H). 13C NMR (C6D6 solvent 298 K) δ 23.12, 24.12, 25.40,
28.45, 28.92 (s, CH(CH3)2), 30.44, 31.78 (s, C(CH3)3), 34.30, 36.25
(s, C(CH3)3), 117.58, 124.53, 127.38, 130.70, 132.02, 136.05, 142.47,
147.94 (s, aryl), 171.73 (s, phenolic ipso-C), 176.62 (s, ketimine-C).
Anal. Found (calcd): C, 75.61 (76.25); H, 9.21 (9.01).
empirical formula
fw
crystal system
space group
V, Å3
Z
a, Å
b, Å
c, Å
C20H26NO
296.42
orthorhombic monoclinic
P212121
1720.6(11)
4
7.541(3)
9.580(4)
23.817(9)
90
90
90
110(2)
1.144
C19H21Cl2NO C16H11F6NO
350.27
347.26
orthorhombic
P212121
1474.5(3)
4
P(1)/n
3665.1(5)
4
11.1221(8)
14.6595(11)
22.4799(16)
90
90.475(2)
90
5.0563(6)
12.8168(15)
22.753(3)
90
90
90
Synthesis of Zn(C19H20NOCl2)2, 4c. Complex 1c (0.18 g, 0.52
mmol) was dissolved in 5 mL of pentane to give a yellow-orange
solution followed by cannulation to a 10 mL pentane solution of Zn-
[N(Si(CH3)3)2]2 (0.10 g, 0.26 mmol) to afford an immediate yellow
precipitate. The precipitate was filtered onto a glass frit and washed
with 3 washings of 5 mL of pentane each to give 0.19 g (94% yield)
yellow solid. Complex 4c could be further purified by recrystallization
R, deg
â, deg
γ, deg
T, K
110(2)
1.270
0.716
110(2)
1.564
0.150
d(calcd), g/cm3
absorp coeff, mm-1 0.069
R,a %
4.91
12.97
6.75
15.91
3.69
11.13
Rw,a %
1
as described in the experimental X-ray structural studies section. H
a R ) ∑||Fo| - |Fc||/∑Fo. Rw ) {[∑w(Fo - Fc2)2/[∑w(Fo )2]}1/2
.
2
2
NMR (C6D6 solvent 298 K) δ 0.71 (br, 12H, CH(CH3)2), 0.97 (d, 12H
CH(CH3)2), 3.1 (br, 4H, CH(CH3)2), 6.37 (d, 2H, aryl), 6.87-7.00 (m,
6H, aryl), 7.26 (d, 1H, aryl), 7.30 (s, 2H, ketimine-H). 13C NMR (C6D6
solvent 298 K) δ 23.49, 25.18, 28.77 (s, CH(CH3)2), 118.65, 124.88,
133.81, 136.02, 142.02, 146.35 (aryl-C), 165.98 (s, phenolic ipso-C),
174.59 (s, ketimine-C). Anal. Found (calcd): C, 59.45 (59.74); H, 5.20
(5.28).
ligands are not sterically bulky enough to inhibit the formation of bis
ligand species. Another route that was pursued in the attempted
formation of mono-salicylaldiminato complexes was through the
reaction of methanol, stoichiometric amounts and in excess, or 1 equiv
of a 2,6-di-substituted phenol with the (salicylaldiminato)Zn(R) (R )
Et or N(Si(Me)3)2) complexes 2 and 3 in THF. These solutions were
allowed to stir at ambient temperature from 1.5 to 12 h after the addition
of protic reagent. In all cases, analysis by NMR showed bis ligand
products or a complex mixture of products whose identity could not
be determined.
Synthesis of Zn(C20H24NO2)2, 4d. Complex 1d (0.49 g, 1.6 mmol)
was dissolved in 10 mL of THF to give a yellow-orange solution
followed by cannulation onto a 10 mL THF solution of Zn[N(Si-
(CH3)3)2]2 (0.30 g, 0.78 mmol) to give a chartreuse solution. Evacuation
of all volatiles produced a sticky yellow solid that was washed several
times with pentane to provide a yellow powder 0.47 g (87% yield).
Complex 4d could be further purified by recrystallization as described
X-ray Structural Studies. Complexes 1a, 1c, 1e, and 4a-4e. Single
crystals suitable for X-ray analysis were obtained for complexes 1a,
1c, 1e, and 4a-4e by dissolving each complex in a minimal amount
of diethyl ether inside a test tube, which was contained inside a Schlenk
tube with approximately 5 mL of toluene under an atmosphere of argon.
The diethyl ether was allowed to slowly evaporate into the toluene
solvent by keeping the tube at 0 °C overnight. Crystal data and details
of data collection for complexes 1a, 1c, and 1e are provided in Table
1. The same information for complexes 4a-4e is provided in Table 2.
The X-ray data were collected on a Bruker CCD diffractometer and
covered more than a hemisphere of reciprocal space by a combination
of three sets of exposures; each exposure set had a different æ angle
for the crystal orientation and each exposure covered 0.3° in ω. The
crystal-to-detector distance was 4.9 cm. Crystal decay was monitored
by repeating the data collection for 50 initial frames at the end of the
data set and analyzing the duplicate reflections; crystal decay was
negligible. The space group was determined based on systematic
absences and intensity statistics.8 The structure was solved by direct
methods and refined by full-matrix least-squares techniques. All non-H
atoms were refined with anisotropic displacement parameters. All H
atoms attached to C atoms were placed in idealized positions and refined
using a riding model with aromatic C-H ) 0.96 Å, methyl C-H )
0.98 Å, and with fixed isotropic displacement parameters equal to 1.2
(1.5 for methyl H atoms) times the equivalent isotropic displacement
parameter of the atom to which they were attached. The methyl groups
were allowed to rotate about their local 3-fold axis during refinement.
For all complexes, data collection, and cell refinement, SMART;8
data reduction, SAINTPLUS (Bruker9); program(s) used to solve
structures, SHELXS-86 (Sheldrick10); program(s) used to refine struc-
tures, SHELXL-97 (Sheldrick11); molecular graphics, SHELXTL-Plus
1
in the experimental X-ray structural studies section. H NMR (C6D6
solvent 298 K) δ 0.85 (d, 6H, CH(CH3)2), 1.04 (d, 18H, CH(CH3)2),
3.29 (s, 10H, OCH3 peak conceals smaller intensity signal of CH(CH3)2
septet), 6.19 (s, 2H, aryl), 7.00 (s, 10H, aryl), 7.68 (s, 2H, ketimine-
H). 13C NMR (C6D6 solvent 298 K) δ 23.20, 25.43, 29.72 (CH(CH3)2),
56.37 (s, OCH3), 116.82, 122.33, 127.69, 142.71, 147.36, 150.42
(aryl-C), 169.34 (phenolic ipso-C), 174.38 (s, ketimine-C). Anal.
Found (calcd): C, 68.05 (70.01); H, 6.98 (7.05); N, 3.67 (4.08).
Synthesis of Zn(C16H10NOF6)2, 4e. Complex 1e (0.45 g, 1.3 mmol)
was dissolved in 10 mL of pentane to give a yellow solution followed
by cannulation onto a 10 mL pentane solution of Zn[N(Si(CH3)3)2]2
(0.25 g, 0.65 mmol) to afford an immediate yellow precipitate. The
precipitate was filtered onto a glass frit and washed with 3 washings
of 5 mL each of pentane to give a yellow solid 0.45 g (92% yield).
Complex 4e could be further purified by recrystallization as described
1
in the experimental X-ray structural studies section. H NMR (C6D6
solvent, 298 K) δ 2.36 (s, 6H, CH3), 6.41 (t, 2H, aryl), 6.52 (d, 2H,
aryl), 6.82 (d, 2H, aryl), 7.01 (d, 2H, aryl), 7.53 (s, 2H, aryl), 7.69
(s, 2H, ketimine-H).
Attempted Syntheses of (Salicyaldiminato)Zn(OR) (R ) Me, Ph,
or COMe) Derivatives. As mentioned earlier, the synthesis of mono-
salicylaldiminato zinc complexes has not yet been accomplished. The
first attempts at making these complexes were modeled after Coates’
synthesis of the highly active diimine zinc acetate catalysts wherein
the sodium or lithium salt of the salicylaldimine ligand was reacted
with Zn(OAc)2. Typically, the sodium or lithium salt was dissolved in
30 mL of THF and allowed to slowly drop onto a 5 equiv excess of
Zn(OAc)2 in 30 mL of THF overnight. After a 12 h reaction time, the
slurry was filtered over a glass frit and the chartreuse solution taken to
1
dryness under vacuum. Analysis of the product by H NMR showed
(8) SMART 1000 CCD; Bruker Analytical X-ray Systems: Madison, WI,
that the bis-salicylaldiminato zinc complexes were produced as opposed
to the targeted (salicylaldiminato)Zn(OAc) complexes, regardless of
which salicylaldimine ligand was used as the starting material. Similar
reaction procedures were used with various zinc starting materials such
as ZnCl2 dissolved in THF, Zn(OTf)2 dissolved in acetonitrile, and Zn-
(CH3CN)4(BF4)2 dissolved in acetonitrile with similar results. In other
words, when soluble zinc starting materials were used in excess, the
bis ligand complexes were produced. Evidently, the salicylaldiminato
1999.
(9) SAINT-Plus, version 6.02; Bruker Analytical X-ray Systems: Madison,
WI, 1999.
(10) Sheldrick, G. SHELXS-86 Program for Crystal Structure Solution;
Institut fur Anorganische Chemie der Universitat: Gottingen, Germany,
1986.
(11) Sheldrick, G. SHELXL-97 Program for Crystal Structure Refinement;
Institut fur Anorganische Chemie der Universitat: Gottingen, Germany,
1997.

Bis-Salicylaldiminato Complexes of Zinc
Inorganic Chemistry, Vol. 40, No. 5, 2001 989
Table 2. Crystallographic Data for Complexes 4a, 4b, 4c, 4d, and 4e
4a
4b
4c
4d
4e
empirical formula
fw
crystal system
space group
V, Å3
C40H50N2O2Zn
656.19
monoclinic
P21/n
3504.0(7)
4
C54H76N2O2Zn‚3PhMe
1126.94
triclinic
P1h
3333.4(5)
4
C38H40Cl4N2O2Zn
763.94
monoclinic
C2/c
7227.2(1)
8
C40H48N2O4Zn
654.21
monoclinic
P21/n
3602.6(9)
4
C64H44F24N4O4Zn2‚PhMe
1611.94
monoclinic
P2/n
3672.8(5)
2
Z
a, Å
b, Å
c, Å
10.3771(13)
17.444(2)
19.524(2)
90
97.494(2)
90
110(2)
1.244
0.737
14.3993(13)
15.3231(13)
15.8728(14)
89.257(2)
73.296(2)
83.682(2)
110(2)
1.614
0.826
7.99
19.40
19.7086(16)
17.3307(14)
22.5847(18)
90
110.4660(10)
90
110(2)
1.492
1.017
6.22
12.0222(17)
14.924(2)
20.888(3)
90
105.989(3)
90
110(2)
1.309
0.726
6.20
13.6459(10)
16.6239(13)
16.9673(13)
90
107.402(2)
90
110(2)
1.523
0.767
R, deg
â, deg
γ, deg
T, K
d(calcd), g/cm3
absorp coeff, mm-1
R,a %
6.11
14.86
7.18
17.16
Rw,a %
14.91
14.25
2
2
a R ) ∑||Fo| - |Fc||/∑Fo. Rw ) {[∑w(Fo - Fc2)2/[∑w(Fo )2]}1/2
.
Scheme 1
version 5.0 (Bruker12); software to prepare material for publication,
SHELXTL-Plus version 5.0 (Bruker).12
molecular weights, and concomitant the polydispersities of the copoly-
mers, were obtained employing gel permeation chromatography in the
laboratories of PAC Polymers, Inc.
Copolymerization of Epoxides and Carbon Dioxide. A typical
copolymerization run was carried out as follows: ca. 0.2 mmol of the
respective zinc bis(salicyaldiminato) complex was dissolved in 20 mL
of cyclohexene oxide or propylene oxide. The solution was loaded via
an injection port into a 300 mL autoclave which had previously been
dried overnight under vacuum at 90 °C. The autoclave was then placed
under 700-800 psi of carbon dioxide and heated to the appropriate
temperature, usually 80 °C. After the allotted time (20 h), the autoclave
was allowed to cool to room temperature, following which the polymer
was extracted. Unreacted monomer was removed by repeated precipita-
tion of the polymer from a dichloromethane solution with methanol.
The poly(cyclohexane carbonate) copolymers were analyzed by 1H
NMR, where the hydrogens adjacent to the carbonate linkages afford
a signal at 4.6 ppm and the absence of polyether linkages was verified
by the absence of a signal at 3.5 ppm. The rest of the aliphatic protons
produce broad signals in the range of 1.2-2.2 ppm. Infrared spectros-
copy was also used to verify the ν(CO) stretch for polycarbonate at
1750 cm-1. The number average (Mn) and weight average (Mw)
Results and Discussion
The salicylaldimine ligands 1a-e were easily prepared by
simple condensations of salicylaldehydes with aromatic amines
in the presence of the catalyst, trifluoroacetic acid, in excellent
yields as outlined in Scheme 1. Compounds 1a, 1c, and 1e
readily crystallized from the reaction solution upon cooling
to 0 °C and their characterization by X-ray diffraction was
accomplished for comparison with the subsequently synthesized
salicylaldiminato zinc complexes. Immediately apparent is the
average CdN bond distance for 1a, 1c, and 1e at 1.281[5] Å,
which is correct for a CdN bond and is consistent with the IR
data obtained where ν(CN) for all three complexes lies between
1620 and 1629 cm-1 13
. It should be noted that the Cimine-Cphenyl
and Nimine-Cphenyl distances are somewhat shorter than expected
for the analogous single bonds (1.54 and 1.51 Å for C-C and
(12) SHELXTL, version 5.0; Bruker Analytical X-ray Systems: Madison,
WI, 1999.
(13) Kovcic, J. E. Spectrochim. Acta 1967, 23A, 183.

990 Inorganic Chemistry, Vol. 40, No. 5, 2001
Darensbourg et al.
cases are colorless liquids, support the assignment of a mono-
adduct species. The best evidence for this assignment is in the
1H NMR chemical shifts of specific resonances on the salicyl-
aldiminato ligand as well as the appearance of new peaks of
appropriate intensities associated with the hexamethylsilylamide
and ethyl functionalities on the complexes, respectively.
1
The H NMR spectra in C6D6 solvent of complexes 2a and
2b exhibit large singlets at 0.28 ppm, which integrates to the
appropriate 18H for the silylamide group. For 2a, the isopropyl
methyl groups appear as two doublets at 0.97 and 1.29 ppm
that integrate to 6H for both doublets. In the starting salicyl-
aldimine ligand 1a, the isopropyl methyl groups appear as one
doublet at 1.05 ppm integrating to 12H. The septet associated
with the lone hydrogen on the isopropyl group for 2a has shifted
to 2.45 ppm (2H) from 3.01 ppm (2H) in 1a. Finally, the
ketimine hydrogen shifts from 7.96 ppm for 1a to 7.84 ppm
for 2a. As discussed in the Experimental Section, similar shifts
occur for the 1H resonances of complexes 2b, 3a, and 3b with
the exception that a quartet and triplet appear in the spectra for
3 at 0.45 and 1.20 ppm corresponding to the remaining ethyl
group on the zinc center in these complexes. It is noteworthy
that the 1H NMR spectra for the bis-salicylaldiminato zinc
analogues 4a and 4b are completely different than that just
described for the monoadducts.
Figure 1. Thermal ellipsoid representation of compound 1a.
The impetus for the synthesis of the mono-salicylaldiminato
zinc complexes 2 and 3 was to use these readily isolated starting
materials in the synthesis of three-coordinate (salicylaldiminato)-
Zn(OR) complexes (where R ) Me, Ph, or COMe). Once
complexes 2 and 3 were isolated, it was hoped to subsequently
react them with a protic reagent such as methanol, phenol, or
acetic acid to protonate the amide or ethyl bases and form the
desired three-coordinate compounds. Unfortunately the synthesis
of (salicylaldiminato)Zn(OR) complexes proved to be nontrivial.
Upon the addition of a protic reagent and workup of the
products, it was discovered through 1H NMR that a mixture of
bis(salicylaldiminato)Zn and Zn(OR)2 was produced. Evidently,
a disproportionation reaction occurred to produce a stable four-
coordinate zinc complex with nitrogen donors and zinc alkoxides
as extended aggregates. Indeed, similar products were observed
when the salicylaldimine ligands were first deprotonated with
NaH or alkyllithium reagents and reacted with various zinc salts
such as Zn(OAc)2, Zn(OTf)2, ZnCl2, and Zn(CH3CN)4(BF4)2.
These reactions are depicted schematically at the bottom of
Scheme 1.
Figure 2. Thermal ellipsoid representation of compound 1c.
Figure 3. Thermal ellipsoid representation of compound 1e.
Complexes 4a-e were conveniently synthesized by reacting
2 equiv of the appropriate salicylaldimine ligand with 1 equiv
of zinc bis-hexamethylsilylamide in either THF or pentane.
Pentane solvent resulted in a cleaner product and higher yields
for the complexes that were insoluble in pentane (e.g., com-
plexes 4c and 4e). All of the complexes have been crystallo-
graphically characterized, and thermal ellipsoid representations
of 4a-e and a simplified drawing of 4e are given in Figures
4-8 and Figure 9, respectively. The relevant bond lengths and
angles are compiled in Table 3 for 4a-e. Complexes 4a-d
crystallized out in distorted tetrahedral geometries with O-Zn-O
angles ranging between 105° and 112.5°. The N-Zn-N bond
angles were typically more obtuse spanning the range 122.9-
128.9° and can most probably be attributed to the larger bulk
of the 2,6-(iPr)2C6H3 group attached to the nitrogen atoms.
Zn-O bond distances for these complexes in general are slightly
longer than the Zn-O bond distances for Zn bis(phenoxides);
this observation is most likely due to the fact that the phenolic
group in 4a-d must accommodate the chelating ring geometry.
Complexes 4a-d contain Zn-O bond lengths ranging between
C-N bonds, respectively) at an average of 1.447[7] and 1.419[6]
Å, respectively. This can be attributed to partial conjugation
between the imine C and N atoms and the aromatic rings.
Thermal ellipsoid drawings of the neutral ligand precursors, 1a,
1c, and 1e, are shown in Figures 1-3, respectively.
As mentioned previously and discussed in detail later, the
synthesis of (salicylaldiminato)Zn(OR) (where R ) Me, Ph, or
COMe) has not yet been accomplished; however, the synthesis
of mono(salicylaldiminato)ZnR (where R ) Et or [N(Si(Me)3)2])
is easily achieved by reacting 1 equiv of the salicylaldimine
ligand with the appropriate zinc reagent, as depicted in Scheme
1, to form complexes 2a, 2b, 3a, and 3b. Solid-state charac-
terization of these derivatives via X-ray diffraction has not yet
been accomplished due to a lack of suitable crystals of these
complexes. Hence, we cannot rule out the possibility that
complexes 3a,b are dimers. On the other hand, because of the
bulky nature of the salicylaldimine ligands and the amide group,
1
dimer formation is unlikely in complexes 2a,b. The H NMR
of complexes 2 and 3, and the fact that both complexes are
bright yellow solids whereas the starting zinc reactants in both

Bis-Salicylaldiminato Complexes of Zinc
Inorganic Chemistry, Vol. 40, No. 5, 2001 991
Figure 7. Thermal ellipsoid representation of Zn(C20H24NO2)2, complex
4d.
Figure 4. Thermal ellipsoid representation of Zn(C20H24NO)2, complex
4a.
Figure 8. Thermal ellipsoid representation of Zn(C16H11NOF6)2,
complex 4e.
Figure 5. Thermal ellipsoid representation of Zn(C27H39NO)2, complex
4b.
Figure 6. Thermal ellipsoid representation of Zn(C19H20NOCl2)2,
complex 4c.
1.908(3)-1.949(3) Å. Interestingly, the shorter bond distances
are observed with more electron donating substituents on the
phenol ring such as Bu and OMe groups and the longer
Figure 9. Simplified ball-and-stick drawing of complex 4e.
t
distances are observed with the electron withdrawing Cl groups
on the phenol ring. The imine C-N bond appears to retain its
double bond character once bound to the zinc center, although
some lengthening outside of error occurs in the CdN bond on
going from the free ligand to the bound ligand as a result of
the possibility of increased resonance throughout the six-
membered chelating ring. The double bond character of the
imine functionality is exemplified in the average Zn-N bond
distances for complexes 4a-d being 1.999[9], which suggests
primarily electron donation from the nitrogen lone pair to the
zinc center.
While complexes 4a-d in the solid-state adapted distorted
tetrahedral monomeric structures, complex 4e was found to be
a dimer with bridging salicylaldiminato ligands resulting in a
distorted trigonal bipyramidal geometry around each zinc center.
This result is likely due to the decreased steric bulk of the
aromatic ring attached to the nitrogen rather than electronic
effects of the CF3 groups. The chelating salicylaldiminato ligand
on the zinc center of this dimeric species contains bond lengths
similar to those observed for the other derivatives 4a-d (see
Table 3). As would be expected, the bond lengths for the
bridging ligands, specifically the Zn-O and Zn-N distances,

992 Inorganic Chemistry, Vol. 40, No. 5, 2001
Darensbourg et al.
Scheme 2
Table 3. Selected Bond Distances (Å) and Bond Angles (deg) for
Complexes 4a-ea
derivatives could be initiated by way of CO2 insertion. This
latter process does not require prior coordination of CO2 to the
metal center. Hence, in instances where the phenolic oxygen
atom is not sterically encumbered, reVersible CO2 insertion into
the Zn-O bond should occur with concomitant chelate ring-
opening leading to an incipient three-coordinate zinc derivative.
This process is represented in Scheme 2, where chelate ring-
opening is driven in the presence of large quantities of CO2
by formation of an eight-membered ring. Indeed, these four-
coordinate complexes do exhibit varying activity for the co-
polymerization of CO2 and cyclohexene oxide. Their effective-
ness for catalyzing this process spans a fairly broad range and
appears to be influenced by the substituents on both the phenolic
and ketimine rings.
Complex 4a
Zn-N(1)
Zn-O(1)
1.990(3)
1.925(3)
Zn-N(2)
Zn-O(3)
1.998(3)
1.925(2)
N(1)-C(12) (imine) 1.302(4)
O(3)-Zn-O(1)
N(2)-Zn-O(3)
111.76(11) N(2)-Zn-N(1)
96.01(11) N(2)-Zn-O(1)
128.77(12)
109.44(12)
Complex 4b
Zn-O(1)
Zn-N(1)
1.908(3)
1.997(3)
Zn-O(2)
Zn-N(2)
1.924(3)
1.984(3)
N(1)-C(1B) (imine) 1.295(5)
O(2)-Zn-O(1)
N(2)-Zn-O(2)
105.59(12) N(2)-Zn-N(1)
96.38(12) N(2)-Zn-O(1)
122.93(13)
118.59(12)
Complex 4c
Zn-O(1)
1.944(3)
2.003(3)
1.287(5)
Zn-O(2)
1.949(3)
2.020(3)
The activities of these bis(salicylaldiminato)zinc complexes
for catalyzing the alternating copolymerization of cyclohexene
oxide and CO2 to provide high molecular weight polycarbonates
were investigated at 80 °C and 55 bar. The efficacy of these
catalysts precursors decreased in the order 4a > 4e > 4b > 4d
. 4c, with turnover frequencies (TOFs) ) 15 > 7.3 > 5.0 >
1.2 > trace (g‚polym/g‚Zn/hr), respectively. The relatively high
activity of 4a can be attributed to the fact that the methyl group
on the phenolic ring is both electron donating and not sterically
hindering, thereby facilitating the insertion of CO2. The remain-
ing salicylaldiminato ligand on the three-coordinate catalyst
formed for 4a must not be, in this context, detrimental to the
zinc center’s ability to bind epoxides. It is important to note
that there is a significant energy barrier to CO2 insertion into
the Zn-O bind due to the attendant disruption of the six-
membered ring formed by the salicylaldminato ligand and the
zinc center. Hence, the position of the equilibrium for formation
of the putative three-coordinate zinc catalyst is strongly de-
pendent on the nature of the Zn-O and Zn-N bonds. When
these complexes are dissolved in C6D6 in an NMR tube and
exposed to an atmosphere of 13CO2 at ambient temperature, the
only 13C resonance observed was that of free 13CO2 (125 ppm).
Nevertheless, this barrier should be easily overcome under the
conditions (80 °C and 55 bar) of catalysis. However, under no
circumstances have we observed formation of significant
quantities of the CO2 insertion product. The three-coordinate
zinc catalyst formed after CO2 insertion is supported by the
absence of polyether linkages within the polymer backbone as
Zn-N(1)
Zn-N(2)
N(1)-C(7) (imine)
O(2)-Zn-O(1)
O(2)-Zn-N(2)
112.37(11) N(1)-Zn-N(2)
93.90(12) O(2)-Zn-N(1)
127.36(14)
114.57(12)
Complex 4d
Zn-O(1)
1.908(3)
2.007(4)
1.300(6)
Zn-O(2)
Zn-N(2)
1.934(3)
1.993(4)
Zn-N(1)
N(1)-C(1) (imine)
O(1)-Zn-O(2)
O(2)-Zn-N(2)
112.56(15) N(1)-Zn-N(2)
113.18(15) O(2)-Zn-N(1)
128.96(16)
94.22(15)
Complex 4e
1.958(3)
2.022(3)
Zn(1)-O(2)
Zn(1)-O(1)
N(2)-C(24) (imine) 1.309(6)
Zn(1)-N(2)
Zn(1)-N(1)
N(1)-C(8) (imine) 1.289(6)
2.049(4)
2.143(4)
O(1)-Zn(1)-O(2)
O(1)-Zn(1)-N(1)
N(1)-Zn(1)-N(2)
N(2)-Zn(1)-O(1A) 128.34(15) N(1)-Zn(1)-O(2)
O(2)-Zn(1)-O(1A) 91.80(14) N(2)-Zn(1)-O(2)
165.00(15) O(1)-Zn(1)-N(2)
102.16(15)
85.14(15) O(1)-Zn(1)-O(1A) 76.75(15)
110.54(16) N(1)-Zn(1)-O(1A) 120.63(15)
92.64(16)
92.53(15)
a Estimated standard deviations in parentheses.
have increased as compared to the corresponding distances in
the monomer complexes. The Oax-Zn-Oax angle is 165.00(15)°
and the equatorial angles are not far from 120° except for the
Neq-Zn-Neq angle which is 110.54(16)°, which can be
1
contributed to the strain in the bridging ligands. The H NMR
spectrum of 4e shows peaks that are slightly broadened, which
could be a result of an equilibrium process between the dimeric
and monomeric species in solution.
Although the bis(salicylaldiminato)zinc complexes were not
the target species for serving as catalysts for the coupling of
carbon dioxide and epoxides, it seemed plausible that these
1
determined by H NMR. In other words, all complexes in this
study produced copolymer with >99% polycarbonate linkages
similar to the three-coordinate Zn(2,6-(tBu)2OC6H3)2(PCy3)

Bis-Salicylaldiminato Complexes of Zinc
Inorganic Chemistry, Vol. 40, No. 5, 2001 993
catalyst2 and unlike the Zn(2,6-(R)2OC6H3)2(solvent)2 complexes
previously studied where the solvent ligands were labile.1
The decrease in catalytic activity in proceeding from the
catalyst precursor 4a to 4e can be the net result of several
contributing factors. First, if the difference in solid-state
structures (monomer 4a and dimer 4e) exists in solution,
dimer f monomer disruption is likely to impose an additional
barrier to reaction. Furthermore, the presence of the electron-
withdrawing CF3 substituents in the 3 and 5 positions of the
ketimine ring should inhibit CO2 insertion into the Zn-O bond.
On the other hand, it would be anticipated that these electron-
withdrawing groups would enhance epoxide binding to the zinc
center and thereby increase catalytic activity. Evidently, these
various factors conspire leading to a decrease in catalytic activity
in going from complex 4a to 4e. The reactivity difference
between catalyst precursors 4a and 4b is more apparent. That
is, the decline in catalytic activity in going from complex 4a to
4b most likely is the result of steric hindrance about the oxygen’s
lone pairs, thereby retarding the initiation step of CO2 insertion.
Further support for this steric argument can be seen by
comparing the space filling models of complex 4a, which has
small methyl groups in the position ortho to the oxygen atom,
and complex 4b, which has larger tertiary butyl groups next to
the oxygen atom. This comparison is shown in Figure 10 and
best seen when viewed down the approximate C2V axis between
the oxygen atoms bound to the zinc center. Clearly shown
in Figure 10 is the increase in steric bulk around the oxygen
atoms on going from methyl substituents A and tertiary butyl
substituents B.
Figure 10. Space filling diagram of complex 4a (A) and 4b (B)
showing the steric hindrance around the oxygen atom associated with
an increase in steric bulk on going from methyl to tertiary butyl
substituents.
and enhance epoxide binding, result in a greatly retarded
catalytic system.
Analysis of the formed poly(cyclohexane carbonate) by gel
permeation chromatography shows that the polymer is rather
polydispersed with Mn ) 41 000 and Mw ) 418 000. This high
polydispersity of 10.3 most likely results from the fact that not
all of the catalyst molecules are initiated at the same time
resulting in polymer chains of various lengths. The fact that
CO2 does not readily insert into the Zn-O bonds of complexes
4a-e supports this possibility.
Acknowledgment. Financial support from the National
Science Foundation (CHE99-10342 and CHE 98-07975 for the
purchase of X-ray equipment) and the Robert A. Welch
Foundation is greatly appreciated.
Originally it was assumed that complex 4d with the small,
more electron releasing OMe substituent on the phenolate ligand
would lead to an enhancement of catalytic activity by accelerat-
ing the CO2 insertion process. Apparently, the concomitant
decrease in epoxide binding leads to a net decrease in catalytic
activity. By way of contrast, the electron-withdrawing chloride
substituents in complex 4c, which should retard CO2 insertion
Supporting Information Available: X-ray crystallographic files
in CIF format for the structure determinations of 1a, c, and e and
4a-e. This material is available free of charge via the Internet at
http://pubs.acs.org.
IC0006403