Polyhedron 69 (2014) 55–60

Contents lists available at ScienceDirect

Polyhedron
journal homepage: www.elsevier.com/locate/poly

Bis(3,5-dimethylpyrazole) copper(II) and zinc(II) complexes as efficient

initiators for the ring opening polymerization of e-caprolactone
and D,L-lactide
Divambal Appavoo a, Bernard Omondi a, Ilia A. Guzei a,b, Juanita L. van Wyk a, Orpah Zinyemba a,
James Darkwa a,⇑
a
Department of Chemistry, University of Johannesburg, P.O. Box X524, Auckland Park 2006, South Africa
b
Department of Chemistry, University of Wisconsin–Madison, Madison, WI 53706, USA

a r t i c l e i n f o a b s t r a c t

Article history: 3,5-Dimethylpyrazole (L), reacts with [Cu(OAc)2] or [Zn(OAc)2]; and either C6H5COOH, 4-OH–C6H5COOH,
Received 31 July 2013 3,5-NO2–C6H5COOH or 2-Cl–C6H5COOH to afford the corresponding metal complexes: [Cu(C6H5COO)2
Accepted 8 November 2013 (L)2] (1), [Cu(2-Cl–C6H4COO)2(L)2] (2), [Cu(3,5-NO2–C6H3COO)2(L)2] (3), [Cu(4-OH–C6H4COO)2(L)2] (4),
Available online 19 November 2013
[Zn(C6H5COO)2(L)2] (5), [Zn(2-Cl–C6H4COO)2(L)2] (6), [Zn(3,5-NO2–C6H3COO)2(L)2] (7), [Zn(4-OH–C6H4
COO)2(L)2] (8). The diamagnetic zinc complexes were characterized by IR, NMR spectroscopy and elemen-
Keywords: tal analysis; while the paramagnetic copper complexes were only characterized by IR spectroscopy,
Pyrazole
elemental analysis and 1 and 2 were further characterized by single crystal X-ray diffraction studies.
Copper
Zinc
All eight complexes are very good initiators for the ring opening polymerization (ROP) of e-caprolactone
Initiators and D,L-lactide at elevated temperatures, under solvent-free conditions and in toluene respectively; pro-
Ring opening polymerization ducing polycaprolactone (PCL) and poly(D,L-lactide) (PLA) of moderate molecular weights (ca 4757 Da for
e-Caprolactone PCL and 2484 Da for PLA) and narrow polydispersity indices (1.36 for PCL and 1.42 for PLA). The 13C{1H}
D,L-Lactide NMR spectra indicate the stereochemistry of PLA is heterotactic.
Ó 2013 Elsevier Ltd. All rights reserved.

1. Introduction recently, lanthanides [12] and transition metals such as Zn [13]

and Cu [14] have been investigated as initiators. This report pro-
Polycaprolactone (PCL) and poly(D,L-lactide) (PLA) are two very vides a comparison of Zn initiators to their Cu analogues.
important polymers used in agricultural, biomedical and environ- Ligand systems also have the ability to affect the ability of a me-
mental fields due to their tuneable properties and biodegradability tal complex to initiate the ROP of these cyclic esters. In many cases
[1]. These polyesters are mainly synthesized by ring opening poly- ancillary ligands are used to tune metal complexes to aid the poly-
merization (ROP) of the cyclic esters with coordination compounds merization process [15]. Schiff base and b-diiminate compounds
as catalysts. Developing catalysts that are effective and inexpen- are the most popular nitrogen-donor ancillary ligands in initiators
sive for producing these biodegradable polymers would aid in for lactide and lactone polymerization, primarily because they pro-
reducing costs of producing these polymers, especially for agricul- vide electrophilic metal centers for initiators to efficiently coordi-
tural usage. Many metal complexes such as Al [2], Sn [3], Ca [4], Li nate monomers [16]. Pyrazole and pyrazolyl donor compounds
[5], Mg [6], Fe [7], Zn [8] and Zr [9] have been used as initiators/cat- are being actively developed as nitrogen-donor catalysts [17],
alysts for the ROP of cyclic esters. However, in many cases unde- because such compounds offer highly electrophilic metal centers.
sired trans-esterification reactions take place, which hamper Recent reports by Yu et al. [17b], Kang et al. [17c] and Lian et al.
control over the polymerization process. As a result, there is con- [17d] indicate that pyrazolyl zinc complexes are active initiators
tinued search for new initiators/catalysts that can yield well- for the ring opening polymerization of D,L-lactide. In the current re-
defined PCL and PLA. Tin(II) octanoate [10] and aluminium(III) port, we have combined the ability of pyrazole as nitrogen-donor
isopropoxide [11] continue to be the two standard industrial initi- ligands and substituted benzene carboxylate as sources of metal–
ators and also used for the study of the ROP mechanisms. More oxygen bonds to zinc and coppers initiators for the ROP of e-capro-
lactone and D,L-lactide. We have also used Energy Dispersive X-ray
(EDX) analysis as a tool to determine whether the final polymers
⇑ Corresponding author. are free of metals, especially for polymers produced with copper
E-mail address: [email protected] (J. Darkwa). initiators.

0277-5387/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.poly.2013.11.011
56 D. Appavoo et al. / Polyhedron 69 (2014) 55–60

2. Experimental 1.87 BM. Anal. Calc. for C24H24CuCl2N4O4: C, 50.85; H, 4.27; N,

9.88. Found: C, 50.79; H, 4.12; N, 9.78%.
2.1. Materials and instrumentation
2.2.3. Synthesis of [Cu(3,5-NO2–C6H3COO)2(L)2] (3)
Syntheses were performed under a dry nitrogen atmosphere Complex 3 was synthesized using [Cu(OAc)2] (0.20 g,
using standard Schlenk techniques. All solvents were of analytical 1.00 mmol), 3,5-NO2–C6H3COOH (0.43 g, 2.00 mmol) and 3,5-
grade and were dried and distilled prior to use. Toluene and dichlo- dimethylpyrazole (0.19 g, 2.00 mmol) and isolated as a green solid.
romethane were dried and distilled from sodium benzophenone Yield = 0.45 g (66%). IR: 1620 cm 1 (mC@O). Anal. Calc. for C24H22N8-
and P2O5 respectively. 3,5-Dimethylpyrazole, [Cu(OAc)2], [Zn(OAc)2], O12Cu
(1/3 mol)C7H8: C, 44.63; H, 3.51; N, 15.81. Found: C, 44.85;
C6H5COOH (benzoic acid), 4-OH–C6H5COOH (4-hydroxybenzoic H, 3.63; N, 15.74%.
acid), 3,5-NO2–C6H5COOH (3,5-dinitrobenzoic acid), 2-Cl–C6H5
COOH (2-chlorobenzoic acid) and e-caprolactone (98%) were 2.2.4. Synthesis of [Cu(4-OH–C6H4COO)2(L)2] (4)
obtained from Sigma–Aldrich and used as received. D,L-lactide The reaction of [Cu(OAc)2] (0.20 g, 1.00 mmol), 4-OH–C6H4-
was obtained from Purac biochem and used as received. NMR COOH (0.28 g, 2.00 mmol) and 3,5-dimethylpyrazole (0.19 g,
spectra were recorded in chloroform-d (CDCl3) and dimethylsulph- 2.00 mmol) gave 4 as a green solid. Yield = 0.33 g (62%). IR:
oxide-d6 (DMSO) on a Gemini 2000 instrument (300 MHz for 1H 1674 cm 1 (mC@O). Anal. Calc. for C24H26CuN4O6
4CH3OH: C,
NMR and 75 MHz for 13C{1H} NMR) and a Bruker Ultrashield 400 51.09; H, 6.43; N, 8.51. Found: C, 50.74; H, 5.49; N, 8.50%.
instrument (400 MHz for 1H NMR and 100 MHz for 13C{1H} NMR)
at room temperature. 1H and 13C{1H} NMR chemical shifts were 2.2.5. Synthesis of [Zn(C6H5COO)2(L)2] (5)
referenced to the residual signals of the protons or carbons of the Reaction of [Zn(OAc)2] (0.18 g, 1.00 mmol), C6H5COOH (0.24 g,
NMR solvents and are quoted in d (ppm): CDCl3 at 7.24 and 2.00 mmol) and 3,5-dimethylpyrazole (0.19 g, 2.00 mmol) gave 5
77.0 ppm for 1H and 13C{1H} NMR spectra respectively; DMSO at as white crystals. Yield = 0.40 g (80%). 1H NMR (CDCl3): d 2.24 (s,
2.45 and 39.5 ppm for 1H and 13C{1H} NMR spectra respectively. 12H, CH3-pz); 5.91 (s, 2H, CH-pz); 7.36 (t, 3JHH = 14.8 Hz, 4H, 3,5-
Coupling constants are reported in Hertz (Hz). Infrared spectra Bz); 7.43 (t, 3JHH = 14.8 Hz, 2H, 4-Bz); 8.14 (d, 3JHH = 7.2 Hz, 2,6-
were recorded on a Bruker FR-IR Tensor27 spectrometer fitted with Bz). 13C{1H} NMR (CDCl3): d 11.9; 105.8; 127.8; 130.1; 131.2;
an ATP-IR probe. Elemental analyzes were performed on a Vario 135.0; 173.6. ESI-MS m/z (%): 379 [M C6H5COO]+, 50. IR:
Elementar microcube CHNS analyzer at Rhodes University, South 1609 cm 1 (mC@O). Anal. Calc. for C24H26N4O4Zn: C, 57.66; H, 5.24;
Africa. ESI-MS spectra were recorded on a Waters API Quattro Mi- N, 11.21. Found: C, 57.60; H, 5.39; N, 11.24%.
cro spectrometer at the University of Stellenbosch, South Africa.
Magnetic moments of the copper complexes were measured using 2.2.6. Synthesis of [Zn(2-Cl–C6H4COO)2(L)2] (6)
a Magway MSB MK1 magnetic susceptibility balance. The effective Complex 6 was synthesized using [Zn(OAc)2] (0.18 g,
magnetic moment, leff, is reported in Bohr magnetons (BM). 1.00 mmol), 2-Cl–C6H4COOH (0.31 g, 2.00 mmol) and 3,5-dimeth-
Molecular weight and molecular weight polydispersity of polymers ylpyrazole (0.19 g, 2.00 mmol) and isolated as a white solid.
were determined by size exclusion chromatography (SEC), using a Yield = 0.36 g (64%). 1H NMR (CDCl3): d 2.15 (s, 12H, CH3-pz);
WGE Dr. Bures Q1000 Gel Permeation Chromatograph equipped 5.85 (s, 2H, CH-pz); 7.34 (d, 3JHH = 6.1 Hz, 2H, 3-CH-Bz); 7.43 (t,
3
with polystyrene gel columns using THF as an eluent, calibrated JHH = 14.0 Hz, 4H, 4,5-CH-Bz); 7.70 (d, 3JHH = 5.8 Hz 2H, 6-CH-Bz).
13
by polystyrene standards at 40 °C. C{1H} NMR (CDCl3): d 11.8; 104.0; 126.9; 130.1; 131.0; 143.8;
149.2. IR: 1635 cm 1 (mC@O). Anal. Calc. for C24H24Cl2N4O4Zn: C,
50.68; H, 4.25; N, 9.85. Found: C, 50.99; H, 4.31; N, 9.45%.
2.2. Synthesis of zinc and copper complexes
2.2.7. Synthesis of [Zn(3,5-NO2–C6H3COO)2(L)2] (7)
The method used to synthesize the complexes is an adaptation Complex 7 was prepared using [Zn(OAc)2] (0.18 g, 1.00 mmol),
of the procedure described by Baruah for the synthesis of pyrazole 3,5-NO2–C6H3COOH (0.43 g, 2.00 mmol) and 3,5-dimethylpyrazole
metal carboxylates [18]. It is a one pot, room temperature reaction (0.19 g, 2.00 mmol). Compound 7 was isolated as a white solid
between a metal(II) acetate, benzoic acid and the pyrazole com- after filtration and evaporation of the solvent from the filtrate.
pound. In a typical reaction the metal(II) acetate was stirred with Yield = 0.40 g (59%). 1H NMR (DMSO): d 2.12 (s, 6H, CH3-pz); 5.76
the benzoic acid, in a 1:2 mol ratio in methanol (15 mL) for (s, 2H, CH-pz); 8.91 (s, 6H, 2,4,6-Bz). 13C{1H} NMR (DMSO): d
30 min at room temperature, after which a solution of 3,5-dimeth- 11.8; 103.8; 120.6; 129.0; 138.5; 148.1; 166.5. IR: 1625 cm 1
ylpyrazole (L) in toluene (5 mL) was added to the mixture and (mC@O). Anal. Calc. for C24H22N4O12Zn: C, 42.40; H, 3.26; N, 16.48.
stirred for a further 30 min. The ratio of the metal(II) acetate to Found: C, 42.18; H, 3.24; N, 16.44%.
pyrazole was 1:2.
2.2.8. Synthesis of [Zn(4-OH–C6H4COO)2(L)2] (8)
2.2.1. Synthesis of [Cu(C6H5COO)2(L)2] (1) Reaction of [Zn(OAc)2] (0.18 g, 1.00 mmol), 4-OH–C6H4COOH
Complex 1 was obtained as blue crystals from the reaction of (0.28 g, 2.00 mmol) and 3,5-dimethylpyrazole (0.19 g, 2.00 mmol)
[Cu(OAc)2] (0.20 g, 1.00 mmol), C6H5COOH (0.24 g, 2.00 mmol) gave 3 as a white solid. Yield = 0.32 g (60%). 1H NMR (CDCl3): d
and 3,5-dimethylpyrazole (0.19 g, 2.00 mmol). Yield = 0.41 g 2.12 (s, 6H, 5-CH3-pz); 2.43 (s, 6H, 3-CH3-pz); 6.01 (s, 2H, CH-
(82%). IR: 1627 cm 1 (mC@O). leff: 1.67 BM. ESI-MS m/z (%): 376 pz); 8.12 (s,br, 4H, 3,5-Bz); 8.25 (s,br, 4H, 2,6-Bz). 13C{1H} NMR
[M C6H5COO]+, 25. Anal. Calc. for C24H26CuN4O4.H2O: C, 55.86; (CDCl3): d 11.7; 12.4; 105.2; 115.0; 123.8; 131.8; 145.0; 149.3;
H, 5.47; N, 10.86. Found: C, 55.95; H, 5.20; N, 10.71%. 161.1; 170.0; 174.2. IR: 1624 cm 1 (mC@O). Anal. Calc. for C24H26N4-
O6Zn.CH3OH: C, 53.24; H, 5.36; N, 9.93. Found: C, 53.43; H, 5.44; N,
9.49%.
2.2.2. Synthesis of [Cu(2-Cl–C6H4COO)2(L)2] (2)
Complex 2 was prepared by reacting [Cu(OAc)2] (0.20 g, 2.3. Molecular structure determination by single crystal X-ray analysis
1.00 mmol) and 2-Cl–C6H4COOH (0.31 g, 2.00 mmol) with 3,5-
dimethylpyrazole (0.19 g, 2.00 mmol). Complex 2 was obtained Single-crystal X-ray diffraction data for 1 and 2 were collected
as a green solid. Yield = 0.31 g (55%). IR: 1574 cm 1 (mC@O). leff: on a Bruker APEXII diffractometer with Mo Ka (k = 0.71073 Å)
 D. Appavoo et al. / Polyhedron 69 (2014) 55–60 57

radiation and diffractometer to crystal distance of 4.00 cm. The ini- sium sulfate, the solvent was removed in vacuo to give a fluffy
tial cell matrix was obtained from three series of scans at different white solid. Purification of PCL was a lot simpler. The polymers
starting angles. Each series consisted of 12 frames collected at were purified by first dissolving the crude product in CH2Cl2, fol-
intervals of 0.5° in a 6° range about with an exposure time of lowed by the addition of cold methanol to give a white precipitate.
10 s per frame. The reflections were successfully indexed by an In all cases Energy Dispersive X-ray (EDX) analysis was run after
automated indexing routine built in the APEXII program suite. The the purification to determine whether there were remnants of the
data were collected using the full sphere data collection routine catalysts in the polymers. Scanning Electron Micrographs (SEM)
to survey the reciprocal space to the extent of a full sphere to a res- provided data on the morphology of the polymers.
olution of 0.75 Å. Data were harvested by collecting 2982 frames at
intervals of 0.5° scans in x and u with exposure times of 10 s per
frame [19]. A successful solution by the direct methods of SHELXS97 3. Results and discussions
provided all non-hydrogen atoms from the E-map. All non-hydro-
gen atoms were refined with anisotropic displacement coefficients. 3.1. Synthesis of pyrazole metal carboxylates
All hydrogen atoms were included in the structure factor calcula-
tion at idealized positions and were allowed to ride on the neigh- Reactions of copper(II) and zinc(II) acetates with substituted
bouring atoms with relative isotropic displacement coefficients benzene carboxylic acids and subsequently with 3,5-dimethylpyr-
[20]. azole produced bis(pyrazole)metal carboxylates of the respective
metal as depicted in Scheme 1. These reactions are driven by the
elimination of acetic acid from the metal acetate, and most likely
2.4. Polymerization of e-caprolactone and D,L-lactide go through an initial formation of metal benzoates followed by a
reaction with the pyrazole that produces the final product. The
Polymerization of e-caprolactone and D,L-lactide were per- one pot reaction provides a very facile route to making these pyr-
formed on a 12-tube Radley’s parallel reactor, using a [mono- azole metal carboxylates in very good yields.
mer]:[initiator] ([M]/[I]) ratios of 100:1 and 50:1 respectively. The copper complexes (1–4) were characterized by IR, magnetic
While the D,L-lactide polymerizations were typically performed moment measurements, elemental analysis whereas single crystal
with D,L-lactide (1.44 g, 0.01 mol) and in toluene (2.5 mL), the cap- X-ray diffraction studies were used to establish the structures of 1
rolactone reactions were performed with (1.14 g, 0.01 mol) of e- and 2. In all complexes the benzoate carbonyl peaks generally
caprolactone in a solvent-free medium. Both reactions were run shifted by 20 cm 1 compared to the corresponding benzoic acid.
at 110 °C. After the required reaction time, the reaction mixture For example, complex 2 has a carbonyl peak at 1574 cm 1, com-
was quenched by rapid cooling to room temperature with liquid pared to a similar peak at 1591 cm 1 in the corresponding carbox-
nitrogen. The solvent was removed in vacuo and percentage con- ylic acid. All four copper complexes are paramagnetic as they
versions determined from the 1H NMR spectra of the polymers. showed very broad peaks in their 1H NMR spectra because of their
The PLA was purified by first dissolving the crude product in paramagnetic properties. For example, the effective magnetic mo-
dichloromethane (15 mL) and adding a hydrochloric acid solution ment of 1.87 BM for complex 2 at room temperature is as expected
in methanol (0.5 M, 15 mL) to the mixture. The solution was sha- for one unpaired electron located on the Cu(II) ion in this complex.
ken vigorously and allowed to separate and the aqueous layer Single crystal X-ray crystallography confirmed structures
washed with dichloromethane (3  10 mL). The combined organic proposed from elemental analysis data. Crystallographic data and
layer was then washed with distilled water (15 mL), followed by selected bond lengths and angles are in Tables 1 and 2 respectively.
sodium hydroxide solution (1.0 M, 15 mL) and lastly with CH2Cl2 The molecular geometries of 1 and 2 show four-coordinate copper
(3  10 mL). After drying the organic layer over anhydrous magne- complexes with cis-coordination of the two benzoates and the two

O2N

Cl
HN N O NO2
M HN N O
O O M
HN N O
HN N O
O NO2
O
(M = Cu (3), Zn (7))
3,5-NO2-C6H3COOH L Cl
O2N
(M = Cu (2), Zn (6))
L 2-Cl-C6H4COOH
[M(OAc)2]
4-OH-C6H4COOH
L
L
OH C6H5COOH

HN N O HN N O
M M
O O O
HN N HN N O

O NH O
L= N
(M = Cu (4), Zn (8)) OH
(M = Cu (1), Zn (5))

Scheme 1. Synthesis of pyrazole copper and zinc carboxylate complexes.

58 D. Appavoo et al. / Polyhedron 69 (2014) 55–60

Table 1
Crystal data and structure refinement parameters for 1 and 2.

Complex 1 2
Empirical formula C24H24N4O4Cu C24H24Cl2N4O4Cu
Formula weight 498.03 566.91
T (K) 100(2) 100(2)
k (Å) 0.71073 0.71073
Crystal system monoclinic monoclinic
Space group P21/n C2/c
a (Å) 15.968(12) 23.468(4)
b (Å) 10.809(9) 10.1880(16)
c (Å) 16.095(13) 11.6142(18)
a (°) 90 90
b (°) 108.44(5) 119.046(3)
c (°) 90 90
V (Å3) 3635.7(5) 2427.6(7)
Z 4 4
Dcalc (Mg/m3) 1.255 1.551
Absorption coefficient (mm 1) 0.862 1.159
F(0 0 0) 1036 1164
Crystal size (mm3) 0.51  0.19  0.13 0.10  0.08  0.07
h Range for data collection (°) 2.45–27.09 1.99–28.39
Index ranges 20 6 h 6 20, 13 6 k 6 13, 20 6 l 6 20 31 6 h 6 30, 11 6 k 6 13, 15 6 l 6 15
Reflections collected 34 499 12 439
Independent reflections 5755 [R(int) = 0.0334] 3050 [R(int) = 0.0504]
Completeness to h (°, %) 27.09 (99.0) 28.39 (99.0)
Absorption correction numerical with SADABS semi-empirical from equivalents
Maximum and minimum transmission 0.8954 and 0.6670 0.9233 and 0.8929
Refinement method full-matrix least-squares on F2 full-matrix least-squares on F2
Data/restraints/parameters 5755/0/302 3050/0/161
Goodness-of-fit (GOF) on F2 0.981 1.034
Final R indices [I P 2r(I)] R1 = 0.0303, wR2 = 0.0827 R1 = 0.0367, wR2 = 0.0867
R indices (all data) R1 = 0.0349, wR2 = 0.0855 R1 = 0.0532, wR2 = 0.0951
Largest difference peak and hole (e Å 3) 0.386 and 0.383 0.783 and 0.570

Table 2
Selected bond lengths (Å) and bond angles (°) for complexes 1 and 2.

1 2
Cu(1)–O(1) 1.9966(18) N(1)–Cu(1) 1.9815(18)
Cu(1)–N(3) 1.979(2) O(1)–Cu(1) 1.9855(15)
Cu(1)–N(1) 1.986(2) C(6)–O(2) 1.248(3)
Cu(1)–O(3) 1.9944(19) C(6)–O(1) 1.273(3)
O(1)–C(1) 1.267(2) C(6)–O(1)–Cu(1) 102.96(14)
O(2)–C(1) 1.2517(19) N(1)–Cu(1)–N(1)#1 92.39(11)
O(4)–C(8)–O(3) 121.89(14) N(1)–Cu(1)–O(1) 91.22(7)
O(2)–C(1)–O(1) 122.04(15) N(1)#1–Cu(1)–O(1) 161.93(7)
N(3)–Cu(1)–N(1) 90.93(8) N(1)#1–Cu(1)–O(1)# 91.22(7)
N(3)–Cu(1)–O(3) 90.47(8) O(1)–Cu(1)–O(1)#1 90.81(9)
N(1)–Cu(1)–O(3) 177.72(5)
N(3)–Cu(1)–O(1) 176.61(5)
N(1)–Cu(1)–O(1) 92.36(8)
O(3)–Cu(1)–O(1) 86.27(7)

pyrazoles (Figs. 1 and 2). Baruah and co-workers [21] have

reported the accidental co-crystallization of a four-coordinate
complex cis-[(3,5-Me 2 pz) 2 Cu(Bz) 2 ] (A) and a five-coordinate
trans-[(3,5-Me2pz)2Cu(Bz)2(OH2)] (where 3,5-Me2pz = 3,5-dimeth-
ylpyrazole and Bz = benzoate) in a Cu(II) catalysed cleavage of a
Fig. 1. Molecular structure of 1 shown with 50% thermal probability ellipsoids. All
carbon–nitrogen bond in N-benzoyl-3,5-dimethylpyrazole. The H atoms are omitted for clarity.
geometry of complex 1, very similar to structure A reported by
Baruah, crystallizes as a solvate with an unidentified solvent (see
Supporting information). Complexes 1 and A can be superimposed distances to the coordinated oxygen atoms are much shorter at un-
with a rmsd of 0.495 Å. Complexes 1, 2, and A are square planar, der 2.0 Å in the three complexes. The Cu–N distances are also typ-
however the planarity is disrupted in 2 and A. Namely, the dihedral ical and closely group around 1.98 Å. In the crystals of the three
angle between the triangular Cu, O, O plane and the triangular compounds the Cu complexes form similar one-dimensional
plane Cu, N, N is 2.04° in 1, 11.78° in A, and 25.28° in 2. In all three hydrogen-bonded chains.
complexes there are distal Cu


O interactions from the metal cen- The zinc complexes are diamagnetic and could be characterized
ter to the would-be apical sites occupied by the uncoordinated by NMR spectroscopy; showing typical peaks associated with both
oxygen atoms of the carboxylate ligands. These Cu


O distances ligands on complexation to a metal. The characterization of these
are 2.480(6) Å in 1, 2.542 Å in A and 2.562 Å in 2. The Cu–O complexes was further supported by IR and elemental analysis.
 D. Appavoo et al. / Polyhedron 69 (2014) 55–60 59

Fig. 3. ROP of e-CL using 1 and 3 as initiators at [M]/[I] = 50:1.

Table 4
ROP of D,L-lactide in toluene using complexes 1–8.
1
Entry Complex % Conversion kapp (h ) Mn (1H NMR) Mn (SEC)a PDI
1 1 94 0.039 1780 1232 1.84
2 2 93 0.020 1650 – –
3 3 92 0.018 1810 1167 1.87
4 4 92 0.020 1242 – –
5 5 96 0.024 1900 1971 2.05
6 6 95 0.023 1342 – –
Fig. 2. Molecular structure of 2 shown with 50% thermal probability ellipsoids. 7 7 96 0.025 1645 833 2.35
8 8 96 0.023 1342 – –

Conditions: [M]/[I] = 100:1 at 110 °C after 144 h.

3.2. Ring opening polymerization activity a
Using a correction factor 0.58 for Mn.

3.2.1. e-Caprolactone
Complexes 1–8 were all investigated for their activities in the an electron-withdrawing substituent, was at least four orders of
bulk polymerization of e-CL at 110 °C using [M]/[I] = 50:1; leading magnitude slower in initiating the polymerization of the caprolac-
to high conversions of the monomer to polycaprolactone (PCL) tone when compared to the initiator 3 that has two electron-with-
(Table 3 and Fig. 3). Metal-carboxylates as initiators are normally drawing nitro groups (Fig. 3).
used in conjunction with an alcohol to produce a metal–alkoxide 1
H NMR spectroscopy and size exclusion chromatography (SEC)
bond required to initiate the ring opening polymerization of the were used to determine the Mn values of the PCL produced
cyclic ester monomers, according to the mechanistic studies (Table 3). All the Mn from NMR data were lower than Mn from
carried out by several research groups [16c]. We did not add SEC measurement and nearly all the Mn values from NMR are lower
alcohol in order to investigate the effect of substituents on the than Mn determined from SEC and similar to observations were
aromatic ring of the benzoic acid since the metal-carboxylates also made by others using 1H NMR data to estimate molecular weights
have the requisite metal–oxygen bond for initiation of the polymer- of PCL and PLA [22]. The highest Mn, determined by SEC is from 1
ization process. The substituents on the benzoic acid further pro- as initiator (4757 Da, PDI = 1.36) and the lowest Mn was shown
vided us with a way to vary the electrophilicity of the metal center by initiator 2 (3125 Da, PDI = 1.84). These molecular weights
in the initiators. Benzoic acids with electron-withdrawing groups are much lower than compared to molecular weight obtained
were expected to lead to initiators that were more electrophilic. with (pyrazolylphenylimino)phenoxy Al (Mn = 211 000 Da) and Zn
Generally the zinc complexes were better initiators for the poly- (Mn = 181 000 Da) initiators [23]; suggest chain termination in
merization of caprolactone than their copper analogues. The sub- our initiators is very facile. Nonetheless such low molecular weight
stituent on the carboxylate ligand also had a big effect on the polymers can find useful applications where toughness is not a
rate of polymerization. For example the copper initiator 1, without requirement. The formation of low molecular weight polymers is
also indicative of a non-living polymerization process.

Table 3
Bulk ROP of e-CL using Cu and Zn complexes. 3.2.2. D,L-Lactide
a 1 1 b
The ROP of D,L-lactide using the Cu and Zn complexes (1–8)
Entry Complex Time (h) kapp (h ) Mn ( H NMR) Mn (SEC) PDI
were much slower than the ROP of e-caprolactone (Table 4 and
1 1 30 0.090 4242 4734 1.36 Fig. 4), which is not surprising because of the larger ring size of
2 2 29 0.151 2866 – –
the caprolactone. Whereas it took the slowest initiator, 1, to com-
3 3 14 0.286 2514 3450 1.61
4 4 27 0.143 4742 – –
pletely convert CL to PCL in 30 h, it took the same initiator 144 h to
5 5 27 0.185 4405 4757 1.69 have 94% conversion of LA to PLA. However, unlike the CL polymer-
6 6 26 0.182 1817 – – ization the activities of all the initiators for LA polymerization were
7 7 26 0.120 1676 3125 1.84 similar (Fig. 4). The molecular weights of these polymers are much
8 8 25 0.169 2026 – –
lower than PLA from common initiators such as [Al(OiPr)3],
Conditions: [M]/[I] = 50:1, Temp. = 110 °C. (Mn = 90 000 Da) [24] and [Sn(Oct)2] [25] (Mn = 130 000 Da). Poly-
a
Time required for complete polymerization. lactide from 5 had the highest Mn (2484 Da, PDI = 2.05). The stereo-
b
Using a correction factor 0.56 for Mn.
chemistry of the PLA formed was determined from the 13C{1H}
60 D. Appavoo et al. / Polyhedron 69 (2014) 55–60

% Conversion can be found, in the online version, at http://dx.doi.org/10.1016/

100
j.poly.2013.11.011.

80 Iniator 1 References

60 Iniator 8 [1] (a) A. Kowalski, A. Duda, S. Penczek, Macromol. Rapid Commun. 19 (1998)
567;
(b) Z. Gan, T.F. Jim, M. Jim, Z. Yuer, S. Wang, C. Wu, Macromolecules 32 (1999)
40 1218;
(c) B.M. Chamberlain, Y. Sun, J.R. Hagadorn, E.W. Hemmesch, V.G. Young, M.
Pink, M.A. Hillmyer, W.B. Tolman, Macromolecules 32 (1999) 2400;
20 (d) T.M. Ovitt, G.W. Coates, J. Am. Chem. Soc. 121 (1999) 4072;
(e) X. Chen, R.A. Gross, Macromolecules 32 (1999) 308;
(f) R.H. Platel, L.M. Hodgson, C.K. Williams, Polym. Rev. 48 (2008) 11;
0 Time (h) (g) A.C. Albertsson, I.K. Varma, Biomacromolecules 4 (2003) 1466.
0 10 20 30 40 50 [2] (a) A. Duda, Z. Florjanczyk, A. Hofman, S. Slomkowski, S. Penczek,
Macromolecules 23 (1990) 1640;
Fig. 4. OP of D,L-lactide in toluene using complexes 1 and 8. (b) T.M. Ovitt, G.N. Coates, J. Polym. Sci. A 38 (2000) 4686.
[3] H.R. Kricheldorf, M.V. Sumbbl, I.K. Saunders, Macromolecules 24 (1991) 1944.
[4] Z. Zhong, P.J. Dijkstra, C. Birg, M. Westerhausen, J. Feijen, Macromolecules 34
(2001) 3863.
NMR spectra of the polymers. Based on the comparisons with the [5] (a) H.R. Kricheldorf, Makromol. Chem. 194 (1993) 1665;
literature values it was concluded that the PLA produced by com- (b) B.T. Ko, C.C. Lin, J. Am. Chem. Soc. 123 (2001) 7973;
plexes 1–8 are predominantly heterotactic [26]. (c) M.H. Chisholm, C.C. Lin, J.C. Galluccia, B.T. Ko, Dalton Trans. (2003) 406.
[6] (a) H.R. Kricheldorf, M. Berl, N. Scharnagl, Macromolecules 21 (1988) 286;
One of the problems associated with using metal initiators is
(b) M.H. Chisholm, N.W. Eilerts, J.C. Huffman, S.S. Iyer, M. Pacold, K.
the possibility of having trace metals in the polymers isolated. Phomphrai, J. Am. Chem. Soc. 12 (2000) 11845;
Although zinc is bio-friendly and thus having traces of zinc in (c) M.H. Chisholm, J. Gallucci, K. Phomphrai, Inorg. Chem. 41 (2002) 2785;
(d) M.H. Chisholm, K. Phomphrai, Inorg. Chim. Acta 350 (2003) 121.
a polymer for biological use should not be a problem, copper
[7] (a) B.J. O’Keefe, L.E. Breyfogle, M.A. Hillmyer, W.B. Tolman, J. Am. Chem. Soc.
is not bio-friendly. As such we purified all the polymers pro- 124 (2002) 4384;
duced and performed energy dispersive X-ray (EDX) analysis A. Duda, S. Penczek, in: C. Scholz, R.A. Gross, (Eds.), Polymers from Renewable
on these samples to determine if the polymers isolated were Resources: Biopolyesters and Biocatalysis, ACS Symp. Ser. pp. 764.;
(b) P. Dobrzynski, J. Kasperczyk, M. Janeczek, M. Bero, Polymer 43 (2002)
free of metals from the initiators. A typical EDX analysis data 2595.
is given in Fig. S2 in the supplementary materials. In all such [8] B.M. Chamberlaine, M. Cheng, D.R. Moore, T.M. Ovitt, E.B. Lobkovsky, G.W.
measurements we found no metals or only trace amount in Coates, J. Am. Chem. Soc. 123 (2001) 3229.
[9] (a) G. Schwach, J. Coudane, R. Engel, M. Vert, Polym. Int. 46 (1998) 177;
the polymers; confirming that purification processes used pro- (b) M. Bero, J. Kasperczyk, Z. Jedlinski, Macromol. Chem. 191 (1990) 2287;
duced metal-free polymers. (c) M. Bero, J. Kasperczyk, G. Adamus, Macromol. Chem. 194 (1993) 907.
[10] A. Bhaw-Luximon, D. Jhurry, S. Motala-Timol, Y. Lochee, Macromol. Symp. 231
(2005) 60.
4. Conclusion [11] (a) N. Ropson, P. Dubois, R. Jerome, P. Teyssie, Macromolecules 27 (1994)
5950;
(b) P. Dubois, N. Ropson, R. Jerome, P. Teyssie, Macromolecules 29 (1996)
All the copper and zinc complexes synthesized are good initia-
1965.
tors and initiate the ring opening polymerization of both e-capro- [12] (a) M. Labet, W. Thielemans, Chem. Soc. Rev. 38 (2009) 3484;
lactone and D,L-lactide to polycaprolactone and polylactide (b) X. Deng, M. Yuan, C. Xiong, X. Li, J. Appl. Polym. Sci. 73 (1999) 1401.
respectively. The activities of the complexes varied with respect [13] (a) L. Liao, L. Liu, C. Zhang, S. Gong, Macromol. Rapid Commun. 27 (2006)
2060;
to the metal but there is no clear cut variation in the activities of (b) M.H. Chisholm, C.C. Lin, J.C. Gallucci, B.T. Ko, Dalton Trans. (2003) 406;
initiators based on substituents on the benzoic acids that were (c) D.J. Darensbourg, W. Choi, O. Karroonnirun, N. Bhuvanesh,
used as ancillary ligands to introduce a metal–oxygen bond re- Macromolecules 41 (2008) 3493;
(d) Z. Ma, W. Lu, B. Liang, A.J.L. Pombeiro, New J. Chem. 37 (2013) 1529.
quired for the polymerization. Generally the zinc complexes [14] A. John, V. Katiyar, K. Pang, M.M. Shaikh, H. Nanavati, P. Ghosh, Polyhedron 26
showed better activities than their copper analogues. The polymers (2007) 4033.
obtained have low molecular weights, with narrow molecular [15] L.M.A. Roman, B.J. O’Keefe, M.A. Hillymer, W.B. Tolman, Dalton Trans. (2003)
3082.
weight distributions suggesting some control in the polymeriza- [16] (a) A. Arbaoui, C. Redshaw, D.L. Hughes, Chem. Commun. (2008) 4717;
tion process; but, the stereochemistry of the polylactide is predom- (b) J. Libiszowski, A. Kowalski, A. Duda, S. Penczek, Macromol. Chem. Phys.
inantly heterotactic. 203 (2002) 1694;
(c) O. Dechy-Cabaret, B. Martin-Vaca, D. Bourissou, Chem. Rev. 104 (2004)
6147;
Acknowledgements (d) J.B. Gallaway, J.R.K. McRae, A. Decken, M.P. Shaver, Can. J. Chem. 90 (2012)
419. and references therein;
(e) J. Sun, W. Shi, D. Chen, C. Liang, J. Appl. Polym. Sci. (2002) 3312;
We acknowledge financial support for this work from Univer-
(f) A.D. Schwarz, Z. Chu, P. Mountford, Organometallics 29 (2010) 1246.
sity of Johannesburg. We thank Dr. Ravin Narain at the University [17] (a) S.O. Ojwach, J. Darkwa, Inorg. Chim. Acta 363 (2010) 1947;
of Alberta, Canada for the polymer molecular weight (b) X.F. Yu, C. Zhang, Z.X. Wang, Organometallics 32 (2013) 3262;
(c) Y.K. Kang, J.H. Jeong, N.Y. Lee, Y.T. Lee, H. Lee, Polyhedron 29 (2010) 2404;
measurements.
(d) B. Lian, C.M. Thomas, O.L. Casangrande Jr., C.W. Lehmann, T. Roisnel, J.-F.
Carpentier, Inorg. Chem. 47 (2007) 328.
Appendix A. Supplementary data [18] R. Sarma, D. Kalita, B. Baruah, Dalton Trans. (2009) 7428.
[19] Bruker-AXS, APEX2, SADABS, and SAINT Software Reference Manuals, Bruker-AXS,
Madison, Wisconsin, USA, 2009.
CCDC 953410 and 953411 contain the supplementary crystallo- [20] G.M. Sheldrick, Acta Crystallogr., Sect. A 64 (2008) 112.
graphic data for 1 and 2. These data can be obtained free of charge [21] S.K. Deka, M. Laskar, J.B. Baruah, Polyhedron 25 (2006) 2525.
[22] (a) C.-T. Chen, C.-A. Huang, B.-H. Huang, Macromolecules 37 (2004) 7968;
via http://www.ccdc.cam.ac.uk/conts/retrieving.html, or from the (b) Y.-C. Liu, B.-T. Ko, C.-C. Lin, Macromolecules 34 (2001) 6196.
Cambridge Crystallographic Data Centre, 12 Union Road, Cam- [23] C. Zhang, Z.-X. Wang, J. Organomet. Chem. 693 (2008) 3151.
bridge CB2 1EZ, UK; fax: +44 1223 336 033; or e-mail: depos- [24] P. Dubois, C. Jacobs, C. Jérôme, P. Teyssié, Macromolecules 24 (1991) 2266.
[25] A. Duda, S. Penczek, A. Kowalski, J. Libiszowski, Macromol. Symp. 153 (2000)
[email protected]. A typical 13C{1H}NMR spectrum for the
41.
determination of stereochemistry of PLA is provided as supple- [26] (a) J.E. Kasperczyk, M. Bero, Polymer 41 (2000) 391;
mentary material Supplementary data associated with this article (b) J.E. Kasperczyk, Macromolecules 28 (1995) 3937.