Ring-opening polymerization of lactides and ε-caprolactone catalyzed by Zn(II) aryl carboxylate complexes supported by 4-pyridinyl schiff base ligands

Abstract

Synthesis and catalytic studies of aryl carboxylate Zn (II) complexes is reported. Reaction of substituted (E)-N-phenyl-1-(pyridin-4-yl)methanimine with a methanolic solution of Zn(CH₃COO)₂ and substituted aryl carboxylate co-ligands gave heteroleptic Zn(II) complexes; [Zn(C₆H₅COO)₂(L1)]₂ (1), [Zn(C₇H₇COO)₂(L1)]₂ (2), [Zn (4-F-C₆H₄COO)₂(L1)]₂ (3), [Zn(C₆H₅COO)₂(L2)]₂ (4), [Zn(C₇H₇COO)₂(L2)]₂ (5), [Zn (4-F-C₆H₄COO)₂(L2)]₂ (6), [Zn(C₆H₅COO)₂(L3)]₂ (7), [Zn(C₇H₇COO)₂(L3)]₂ (8), [Zn (4-F-C₆H₄COO)₂(L3)]₂ (9). The molecular structures of complexes 1 and 4 are dinuclear with the zinc atom in complex 1 adopting a distorted trigonal bipyramidal geometry in a bi-metallacycle while complex 4 is square pyramidal where all four benzoate ligands bridge the zinc metals in a paddle wheel arrangement. All complexes successfully initiated mass/bulk ring-opening polymerization (ROP) of ϵ-caprolactone (ϵ-CL) and lactides (LAs) monomers with or without alcohol co-initiators at elevated temperatures. Complexes 1, 4 and 6 containing the unsubstituted benzoate co-ligands were the most active in their triad; with complex 4 being the most active (k_app) of 0.3450 h⁻¹. The physicochemical properties of the polymerization products of _l-lactide and rac-lactide in toluene revealed melting temperatures (T_m) between 116.58 °C and 188.03 °C, and decomposition temperatures between 278.78 °C and 331.32 °C suggestive of an isotactic PLA with a metal capped end.

1. Introduction

The depletion of petrochemical feedstock as well as the environmental threat posed by non-degradable polymers have led to the increased attention that polycaprolactone (PCL) and polylactides (PLA) have received [1, 2, 3, 4, 5]. Starting material used to produce PLA are acquired from biomass resources making them vital for a sustainable future [6]. The biodegradability of polyesters such as PCL and PLA is a major benefit for their use in agricultural [7, 8], pharmaceutical and biomedical field [9, 10]. These biodegradable aliphatic polyesters are largely obtained by ring opening polymerization (ROP) of respective monomers through a catalyzed route.

Metal complexes of main group elements such as calcium [11], lithium [12], magnesium [13] and aluminum [14] as well as other metals such as tin [15], zinc [16], copper [17], and cobalt [18], even lanthanides [19] have been investigated as catalyst in ROP of cyclic esters monomers. Regardless of the toxicity posed by tin(II) octanoate [20], it has remained a standard catalyst used in industry in ROP studies. The development of new catalyst is hinged on the need to obtain biocompatible metal complexes that will combat the transesterification reactions faced by known catalysts to produce well-defined and neat polymers via different polymerization mechanisms.

Several types of ROP mechanisms have been put forward for different type of catalyst which can be nucleophilic or electrophilic. The most common ROP mechanisms for transition metal complexes are coordination insertion mechanism (CIM) and activated monomer mechanism (AMM). In CIM mechanism chain propagation proceeds via an intramolecular attack of a pre-coordinated lactone by the nucleophilic auxiliary ligand [21, 22]. In contrast, the AMM utilize a binary catalyst-co-initiator system, whereby the lactone is first coordinated to the metal center for activation and subsequently attacked by an external nucleophile which is the co-initiator. These two mechanisms have been complemented by experimental and computational data [23].

The increased attention received by zinc(II) and copper (II) complexes is hinged on their biocompatibility as well as their ease of synthesis. Carboxylate Zn(II) and Cu(II) complexes bearing N-donor ligands are characterized by their unique ability to coordinate in various modes depending on the electronic property of the ligands and the metal ionic character. Metal complexes bearing ancillary ligands such as (pyrazolylmethyl) pyridine [24, 25, 26], Bis(pyrazolylmethyl) pyridine [27, 28], Bis(3,5-dimethyl pyrazole) [29], (benzimidazolylmethyl) amine [30], formamidine [31, 32, 33, 34] have appeared in literature reports as catalyst/initiators for the ROP of ϵ-CL. Zeolitic imidazolate frameworks were also shown to catalyze the conversion of ϵ-caprolactone [35]. These organometallic complexes can be synthesized utilizing two approaches namely solution based and solid state, synthesis.

Solution based approaches utilizes organic solvents where the reactant components are dissolved to facilitate maximum interaction between the ions and the chelating ligands. The major glitch in such methods is long reaction time, high energy, and solvent consumption [36]. There are concerns over the toxicity of organic solvents used. Synthesis can be done at room or at elevated temperatures by heating using various methods which include conventional heating and microwave radiation [37]. Solid state approach involves mechanochemical activation of a physical mixture of the metal salts and the ligands. The merits of the method include fast reactions, simple, no solubility and solvent coordination issues and regarded as “greener” alternative [38].

This report covers comprehensive structural characterization of carboxylate complexes of N-(pyridin-4-yl) methanimine ligands and their catalytic potency in ROP of ϵ-CL, _L-LA and rac-LA. The reaction kinetics and polymer end group analysis catalytic influence of Schiff base ligands and bridging carboxylate co-ligands steric and electronic properties effects were also discussed.

2. Experimental

2.1. Materials

Aniline 99.5%, isonicotinaldehyde 99%, 4-chloroaniline 99%, 4-toluidine 99.6%, 2,6-diisopropylaniline 97%, ethanol 99.5%, CDCl₃ 99.8 atom % of D, diethyl ether 99.8% and dichloromethane 99% were obtained from Sigma Aldrich while argon and nitrogen gases, 5.0 technical grade were purchased from Airflex Industrial Gases, South Africa. All chemicals purchased were used without further purification, but the solvents were dried and distilled by conventional methods under inert atmospheric conditions before they were used.

2.2. Synthesis of Zn(II) aryl carboxylate complexes

The complexes were prepared in a one pot reaction between a metal (II) acetate, benzoic acid and the Schiff base ligand at room temperature (Scheme 1). In a typical reaction the metal (II) acetate (1 mmol) was stirred with the benzoic acid (2 mmol), in ethanol (15 mL) for 40 min at room temperature, after which a solution of the ligand was added as described for metal carboxylates [29, 39, 40].

Scheme 1.

Synthesis of zinc complexes.

2.2.1. [Zn(C₆H₅COO)₂(L1)]₂ (1)

Complex 1 was synthesized using [Zn(OAc)₂] (1.00 mmol), C₆H₅COOH (2.00 mmol) and L1 (1.00 mmol) and isolated as a yellow solid. Yield = 80%, mp 223–224 °C. ¹H NMR (400 MHz, CDCl₃): δ ppm 2.39 [s, 6H, CH₃], 7.22 [q, 8H, Ar], 7.32 [t, J = 7.62 Hz, 8H, Ar], 7.44 [t, 4H, Ar], 7.84 [d, J = 5.92 Hz, 4H, Ar], 8.10 [d, J = 7.37 Hz, 8H, Ar], 8.45 [s, 2H, HC N], 8.89 [d, J = 5.48 Hz, 4H] ¹³C NMR (400 MHz, CDCl₃): δ ppm 174.28, 155.48, 150.09, 147.82, 144.88, 137.77, 132.99, 132.07, 130.54, 129.99, 127.88, 123.09, 121.11, 21.13. . IR: ʋ (cm⁻¹) 3059 ʋ(C–H) str., 1636 ʋ(C=O) asym. str., 1420 ʋ(C=O) sym. str., 720 ʋ (M − O) str., 461 ʋ (M − N) str. ESI-TOF MS: m/z (%); [M + Na]⁺ = 1029.8429. Anal. calcd for C₅₄H₄₄N₄O₈Zn₂: C, 64.36; H, 4.40; N, 5.56%. Found: C, 64.22, H, 4.43, N, 5.62%.

2.2.2. [Zn(C₇H₇COO)₂(L1)]₂ (2)

Complex 2 was synthesized using [Zn(OAc)₂] (1.00 mmol), C₇H₇COOH (2.00 mmol) and L1 (1.00 mmol) and isolated as a yellow solid. Yield = 82%, mp 252–253 °C. ¹H NMR (400 MHz, CDCl₃): δ ppm 2.36–2.39 [s, 18H, CH₃], 7.15 [d, J = 8.00 Hz, 8H, Ar], 7.22 [q, J = 8.37 Hz, 4H, Ar], 7.85 [d, J = 5.54 Hz, 4H, Ar], 8.01 (d, J = 8.06 Hz, 8H, Ar], 8.47 [s, 2H, HC N], 8.87 [d, J = 4.07 Hz, 4H, Ar] ¹³C NMR (400 MHz, CDCl₃): δ ppm 174.43, 155.65, 151.20, 150.15, 147.87, 144.68, 142.59, 137.71, 130.64, 130.15, 129.99, 128.63, 123.01, 121.10, 21.60, 21.12. . IR: ʋ (cm⁻¹) 2924 ʋ(C–H) str., 1602 ʋ(CO) asym. str., 1400 ʋ(C O) sym. str., 778 ʋ (M − O) str., 439 ʋ (M − N) str. ESI-TOF MS: m/z (%); [M+Na]⁺ = 1081.4037. Anal. calcd for C₅₈H₅₂N₄O₈Zn₂: C, 65.48; H, 4.93; N, 5.27%. Found: C, 65.52, H, 4.945, N, 5.43%.

2.2.3. [Zn(4-F-C₆H₄COO)₂(L1)]₂ (3)

Complex 3 was synthesized using [Zn(OAc)₂] (1.00 mmol), 4-F-C₆H₄COOH (2.00 mmol) and L1 (1.00 mmol) and isolated as a yellow solid. Yield = 78%, mp 225–226 °C. ¹H NMR (400 MHz, CDCl₃): δ ppm 8.88 [d, J = 5.52 Hz, 4H, Ar], 8.47 [s, 2H, HC N], 8.08 [q, 8H, Ar], 7.86 [d, J = 5.88 Hz, 4H, Ar], 7.22 [t, 8H, Ar], 6.97 [t, J = 8.59 Hz, 8H, Ar], 2.39 [s, 6H, CH₃] ¹³C NMR (400 MHz, CDCl₃): δ ppm 173.24, 166.65, 164.15, 155.18, 150.00, 147,71, 145.14, 137.93, 132.99, 132.89, 130.03, 129.18, 123.18, 121.13, 114.97, 114.76, 21.12. IR: ʋ (cm⁻¹) 2936 ʋ(C–H) str., 1636 ʋ(C O) asym. str., 1410 ʋ(C=O) sym. str., 778 ʋ (M − O) str., 439 ʋ (M − N) str. ESI-TOF MS: m/z (%); [M+2Na]²⁺ = 1125.3625. Anal. calcd for C₅₄H₄₀F₄N₄O₈Zn₂: C, 60.07; H, 3.73; N, 7.04%. Found: C, 60.17, H, 3.69, N, 6.95%.

2.2.4. [Zn(C₆H₅COO)₂(L2)]₂ (4)

Complex 4 was synthesized using [Zn(OAc)₂] (1.00 mmol), C₆H₅COOH (2.00 mmol) and L2 (1.00 mmol) and isolated as a yellow solid. Yield = 82%, mp 170–171 °C. ¹H NMR (400 MHz, CDCl₃): δ ppm 1.16–1.18 [s, 24H, CH₃], 2.83–2.90 [m, 4H, CH], 7.17 [q, 6H, Ar], 7.33 [d, J = 7.58 Hz, 4H, Ar], 7.45 [t, 2H, Ar], 7.87 [d, J = 5.09 Hz, 4H, Ar], 8.11 [d, J = 7.30 Hz, 4H, Ar] 8.21 [s, 2H, HC N], 8.96 [s, 4H, Ar]. ]. ¹³C NMR (400 MHz, CDCl₃): δ ppm 174.04, 159.58, 159.55, 150.43, 148.20, 144.02, 143.96, 137.06, 131.89, 131.86, 130.45, 127.86, 124.99, 123.21, 122.93, 122.78, 118.60, 28.05, 23.40, 22.46. IR: ʋ (cm⁻¹) 2959 ʋ(C–H) str., 1602 ʋ(C O) asym. str., 1399 ʋ(CO) sym. str., 720 ʋ (M − O) str., 461 ʋ (M − N) str. ESI-TOF MS: m/z (%); [M+H]⁺ = 1145.4896. Anal. calcd for C₆₄H₆₄N₄O₈Zn₂: C, 66.96; H, 5.62; N, 4.88%. Found: C, 66.85, H, 5.624, N, 4.97%.

2.2.5. [Zn(C₇H₇COO)₂(L2)]₂ (5)

Complex 5 was synthesized using [Zn(OAc)₂] (1.00 mmol), C₇H₇COOH (2.00 mmol) and L2 (1.00 mmol) and isolated as a yellow solid. Yield = 82%, mp 176–178 °C. ¹H NMR (400 MHz, CDCl₃): δ ppm 1.17–1.18 [s, 24H, CH₃], 2.38 [s, 12H, CH₃], 2.88 [m, 4H, CH], 7.16–7.18 [q, J = 4.83 Hz, 6H, Ar], 7.93 [d, J = 5.27 Hz, 4H, Ar], 8.06 [d, J = 7.87 Hz, 8H, Ar], 8.24 [s, 2H, HC N], 8.97 [d, 8H, Ar]. ]. ¹³C NMR (400 MHz, CDCl₃): δ ppm 174.57, 159.57, 150.41, 148.21, 144.03, 142.75, 137.07, 130.69, 130.06, 128.68, 125.00, 123.22, 122.96, 28.07, 23.41, 21.62. IR: ʋ (cm⁻¹) 2959 ʋ(C–H) str., 1602 ʋ(CO) asym. Str., 1410 ʋ(C O) sym. Str., 778 ʋ (M − O) str., 439 ʋ (M − N) str. ESI-TOF MS: m/z (%); [M]⁺ = 1203.5505. Anal. calcd for C₆₈H₇₂N₄O₄Zn₂: C, 67.83; H, 6.03; N, 4.65%. Found: C, 67.68, H, 6.00, N, 4.75%.

2.2.6. [Zn(4-F-C₆H₄COO)₂(L2)]₂ (6)

Complex 6 was synthesized using [Zn(OAc)₂] (1.00 mmol), 4-F-C₆H₄COOH (2.00 mmol) and L2 (1.00 mmol) and isolated as a yellow solid. Yield = 78%. Mp 184 °C. ¹H NMR (400 MHz, DMSO): δ ppm 1.16 [s, 24H, CH₃], 2.86 [m, 4H, CH], 6.94 [t, J = 8.46 Hz, 4H, Ar], 7.16 [m, 6H, Ar], 7.88 (d, J = 5.55 Hz, 4H, Ar], 8.05 [t, 4H, Ar], 8.20 [s, 2H, HC N], 8.95 (d, J = 5.14 Hz, 4H, Ar]. ¹³C NMR (400 MHz, CDCl₃): δ ppm 159.29, 150.27, 148.09, 137.03, 132.03, 131.20, 127.26, 125.09, 123.24, 123.07, 28.08, 23.41, 22.46. IR: ʋ (cm⁻¹) 2969 ʋ(C–H) str., 1636 ʋ(CO) asym. Str., 1400 ʋ(C O) sym. Str., 778 ʋ (M − O) str., 427 ʋ (M − N) str. ESI-TOF MS: m/z (%); [M]⁺ = 1219.5040. Anal. calcd for C₆₄H₆₀F₄N₄O₈Zn₂: C, 63.01; H, 4.96; N, 4.59%. Found: C, 62.92, H, 5.04, N, 4.62%.

2.2.7. [Zn(C₆H₅COO)₂(L3)]₂ (7)

Complex 7 was synthesized using [Zn(OAc)₂] (1.00 mmol), C₆H₅COOH (2.00 mmol) and L3 (1.00 mmol) and isolated as a yellow solid. Yield = 82%, mp 222–223 °C. ¹H NMR (400 MHz, CDCl₃): δ ppm7.21 (4H, d, J = 8.52 Hz), 7.32 (8H, t, J = 7.60 Hz), 7.40 (4H, d, J = 8.52 Hz), 7.44 (4H, t, J = 7.34 Hz), 7.84 (4H, d, J = 5.72 Hz), 8.10 (8H, d, J = 7.60 Hz), 8.42 (2H, s), 8.92 (4H, d, J = 5.48 Hz). ¹¹³C NMR (400 MHz, CDCl₃): δ ppm116.43, 122.42, 123.17, 127.88, 129.12, 129.53, 130.53, 132.08, 133.01, 133.24, 144.37, 148.83, 150.22, 151.18, 156.93, 174.34. IR: ʋ (cm⁻¹) 3049 ʋ(C–H) str., 1614 ʋ(C O) asym. Str., 1379 ʋ(CO) sym. Str., 710 ʋ (M − O) str., 416 ʋ (M − N) str. ESI-TOF MS: m/z (%); [M+2Na]²⁺ = 1095.2750. Anal. calcd for C₅₂H₃₈Cl₂N₄O₈Zn₂: C, 59.56; H, 3.65; N, 5.34%. Found: C, 59.47, H, 3.73, N, 5.34%.

2.2.8. [Zn(C₇H₇COO)₂(L3)]₂ (8)

Complex 8 was synthesized using [Zn(OAc)₂] (1.00 mmol), C₇H₇COOH (2.00 mmol) and L3 (1.00 mmol) and isolated as a yellow solid. Yield = 82%, mp 248–249 °C. ¹H NMR (400 MHz, CDCl₃): δ ppm 2.35 (12H, s), 7.13 (8H, d, J = 7.96 Hz), 7.21 (4H, d, J = 8.60 Hz), 7.39 (4H, d, J = 8.60 Hz), 7.86 (4H, d, J = 5.72 Hz), 8.00 (8H, d, J = 8.00 Hz), 8.44 (2H, s), 8.93 (4H, d, J = 5.48 Hz). ¹³C NMR (400 MHz, CDCl₃): δ ppm 21.61, 122.46, 123.30, 128.65, 129.53, 130.02, 130.60, 133.28, 142.63, 144.58, 148.78, 156.86, 174.27. IR: ʋ (cm⁻¹) 3026 ʋ(C–H) str., 1602 ʋ(C O) asym. Str., 1400 ʋ(CO) sym. Str., 766 ʋ (M − O) str., 416 ʋ (M − N) str. ESI-TOF MS: m/z (%); [M+Na]⁺ = 1123.2943. Anal. calcd for C₅₆H₄₆Cl₂N₄O₈Zn₂: C, 60.89; H, 4.20; N, 5.07%. Found: C, 60.89, H, 4.301, N, 4.98%.

2.2.9. [Zn (4-F-C₆H₄COO)₂(L3)]₂ (9)

Complex 9 was synthesized using [Zn(OAc)₂] (1.00 mmol), 4-F-C₆H₄COOH (2.00 mmol) and L3 (1.00 mmol) and isolated as a yellow solid. Yield = 78%. Mp 229–230 °C. ¹H NMR (400 MHz, DMSO): δ ppm 7.21 (8H, t, J = 8.82 Hz), 7.37 (4H, d, J = 8.59 Hz), 7.50 (4H, d, J = 8.58 Hz), 7.93 (4H, d, J = 5.86 Hz), 8.02 (8H, q, J = 4.81 Hz), 8.72 (2H, s), 8.81 (4H, d, J = 5.81 Hz). ¹³C NMR (400 MHz, DMSO): δ ppm 115.08, 115.29, 123.13, 123.59, 129.73, 131.78, 132.54, 143.53, 149.62, 150.79, 151.50, 160.24, 163.20, 165.66, 171.20. IR: ʋ (cm⁻¹) 3083 ʋ(C–H) str., 1636 ʋ(CO) asym. Str., 1400 ʋ(CO) sym. Str., 766 ʋ (M − O) str., 428 ʋ (M − N) str. ESI-TOF MS: m/z (%); [M+2Na]²⁺ = 1161.2722. Anal. calcd for C₅₂H₃₄Cl₂F₄N₄O₈Zn₂: C, 55.74; H, 3.06; N, 5.00%. Found: C, 55.68, H, 3.17, N, 5.10%.

3. General procedure for ROP of cyclic esters

The polymerization reactions were performed in a carousel reaction station, fitted with 12 tubes, a gas distribution system and a reflux unit. ε-CL (1.14 g, 0.01 mol) and the required amount of initiator, depending on the [M]: [I] ratio used, were weighed in a reactor tube and stirred at 110 $°C$. After the required reaction time, the reaction mixture was quenched by rapid cooling to room temperature and the crude product was analyzed by ¹H NMR spectroscopy in CDCl₃. The polymers were cleaned by first dissolving the crude product in CH₂Cl₂, followed by the addition of cold methanol. The white precipitate formed was altered and dried in vacuo. Polymer conversions were analyzed from the peak areas obtained from ¹H NMR spectroscopy and calculated using equation (1) for ε-CL monomer and equation (2) for LAs [41,42].

$$\frac{{\left[Polymer\right]}_t}{{\left[Monomer\right]}_o}x100=\frac{I_{4.0}}{\left(I_{4.0}+I_{4.2}\right)}x100$$ (1)

$$\frac{{\left[Polymer\right]}_t}{{\left[Monomer\right]}_o}x100=\frac{I_{CHpolymer}}{\left(I_{CHmonomer}+I_{CHpolymer}\right)}x100$$ (2)

The catalytic activities of 4 towards ε-caprolactone (ε-CL) and _d,l-lactide polymerizations have been evaluated under different conditions. Polymerization of ε-caprolactone was carried out in bulk using a [monomer]: [initiator] ([M]: [I]) ratio of 100–400. For the polymerization of _d,l-lactide, the reaction was performed in toluene using a [M]: [I] ratio of 200.

3.1. Molecular structure of complexes

Single-crystals of complexes 1 and 4 were analyzed on a ‘Bruker APEX-II CCD’ diffractometer comprising of a low temperature Oxford Cryostream system. The selected crystals were retained at 100.0 K throughout the data collection exercise. Data reduction was performed using the SAINT [43] software and the scaling and absorption corrections were applied using the SADABS [44] multi-scan technique. Using Olex2 [45], the structure was solved with the ShelXT [46] structure solution program using Intrinsic Phasing and refined with the ShelXL [46] refinement package using Least Squares minimization. Non-hydrogen atoms were first refined isotropically and then by anisotropic refinement with the full-matrix least squares method based on F₂ using SHELXL [23]. Crystallographic data and structure refinement parameters are given in Table 1.

Table 1.

Crystal data and structure refinement parameters for complexes 1 and 4.

Complex	1	4
Empirical formula	C54H44N4O8Zn2	C64H64N4O8Zn2
Formula weight	1007.76	1147.92
T/K	150.01	100.01
λ/Å	0.71073	0.71073
Crystal system	Monoclinic	Monoclinic
Space group	C2/c	P21/c
a/Å	29.5190 (6)	16.116 (3)
b/Å	11.0215 (2)	7.5994 (14)
/Å	16.0324 (3)	24.185 (4)
α/°	90	90
β/°	116.4460 (10)	97.820 (9)
γ/°	90	90
V/Å3	4670.20 (16)	2934 (9)
Z	4	4
ρ(calc)/g cm−3	1.4332	1.299
μ/mm−1	1.089	0.875
F (000)	2083.4	1200
Size/mm3	0.34 × 0.21 × 0.14	0.36 × 0.17 × 0.12
θ range/°	4 to 52.74	1.982 to 28.831
Index ranges	−36 ≤ h ≤ 36	−20 ≤ h ≤ 21
	−13 ≤ k ≤ 13	−10 ≤ k ≤ 10
	−19 ≤ l ≤ 19	−31 ≤ l ≤ 31
Reflections collected	22 453	33 458
Independent reflections	4720 [R (int) = 0.0242]	7289 [R (int) = 0.0215]
Data/restraints/parameters	4720/0/308	7289/0/370
GooF	0.833	1.028
R indices [I > 2sigma(I)] R1 wR2	0.0268	0.0322
	0.0901	0.0696
R indices (all data) R1 wR2	0.0334	0.0475
	0.0996	0.0748
Largest diff. peak/hole/e Å−3	0.34/−0.36	0.359/−0.448

4. Results and discussion

4.1. Synthesis and characterization

The Schiff base ligands were obtained by mechanochemically grinding isonicotinaldehyde with substituted anilines in a mortar and pestle. Mechanochemical offers a robust synthetic route where products are obtained in high yields, rapidly and much pure. This solvent free approach via grinding can result in generation of excessive heat which can lead to product isomerization and/or decomposition [36, 47]. Although sometimes this method is referred to as greener approach, in some cases solvents may be used for purification and isolation of the products. Purity of ligands was ascertained by ¹H and ¹³C NMR and the spectra appear in Fig. S1, S2, S3, S4, S5 and S6.

One pot reaction of the pyridinyl Schiff base ligands, substituted and unsubstituted benzoic acids with zinc(II) acetates in methanol produced zinc carboxylates complexes in good yield (78–82%) as depicted in Scheme 1. All complexes were characterized by IR, NMR, mass spectrometry, UV–vis and elemental analysis. Mass spectroscopy and elemental analysis of all the complexes conforms to di-nuclear carboxylate systems [ZnLn (RCOO)₂]₂.

Distinctive ligand peaks after complexation to the zinc metal were observed with IR and NMR spectroscopies. The Schiff base ligands experienced a shift in the proton peaks of the pyridinyl moiety in the ¹H NMR spectra (Fig. S7, S8, S9, S10, S11, S12, S13, S14 and-S15). IR and ¹³C NMR spectroscopies are known to provide valid evidence about the coordination modes of carboxylate complexes [48]. In the ¹³C NMR spectra (Fig. S16, S17, S18, S19, S20, S21, S22, S23 and S24), a peak around 173–174 ppm corresponds to the carboxylate carbonyl carbon coordinated in a bidentate fashion to the zinc metal. Bands corresponding to Zn—N_py were observed between 416 and 461 cm⁻¹ and Zn–O at 710 - 778 cm⁻¹ in the IR spectra (Fig. S25, S26, S27, S28, S29, S30, S31, S32 and S33). The IR spectra of complexes 1–9 revealed intense bands between 1379 and 1420 cm⁻¹ for υ_sym (COO⁻) and 1602–1636 cm⁻¹ for υ_asym (COO⁻) (Fig. S25, S26, S27, S28, S29, S30, S31, S32 and S33) [32,[49], [50], [51]]. The Δυ (υ_asym – υ_sym) values for all complexes are in the range 192–236 cm⁻¹ which is in consonance with asymmetrical bidentate bridging mode [52].

The substituents of the aryl carboxylate co-ligands had a bearing on the magnitude of Δυ obtained in all the complexes. Complexes 3, 6 and 9 with an electron withdrawing fluoro on the carboxylate phenyl ring in their triad have the highest Δυ while 2, 5 and 8 with an electron donating methyl were smaller than 1, 4 and 7 having the unsubstituted benzoate ligands. Absorption electronic spectra of complexes 1–9 in chloroform analyzed between 220 and 440 nm afforded electronic absorption bands between 231 – 266 nm and 322–340 nm allocated to π-π* and n-π* transition, respectively (Fig. 1).

Fig. 1.

UV–Vis spectra of complexes 1–9 in ∼10^-5 M DCM solutions.

4.1.1. Molecular structures of complexes 1 and 4

Complex 1 was formed by slow evaporation of a dcm/hex mixture (1:1) solution and 4 was gotten from recrystallization in methanol at room temperature. The molecular structures of complexes 1 and 4 are shown in Fig. 2a and b, while the selected crystallographic data is given in Table 2. Complexes 1 and 4 are di-nuclear species and the asymmetric unit is occupied by half a molecule with the other half generated by symmetry operation at the center of inversion between the zinc centers.

Fig. 2.

ORTEP drawing of (a) complex 1 and (b) complex 4 with thermal ellipsoids drawn at 50% probability level. Hydrogen atoms are omitted for clarity.

Table 2.

Selected bond lengths (Å) and bond angles (°) for complexes 1 and 4.

Complex	1		4
Zn–O	1.9413 (11)	1.9899 (12)	2.0813 (12)	2.0808 (12)
	1.9547 (12)	2.5542 (13)	2.0624 (12)	2.0792 (12)
Zn—Npy	2.0375 (14)		2.0565 (13)
O–Zn–O	87.12 (5)		88.36 (5)
	132.06 (5)		89.05 (5)
O–Zn—Npy	93.26 (5)		96.99 (5)
	99.33 (5)
	109.78 (5)		102.66 (5)

Symmetry operators: i = 1 − x, +y, 3/2 −s z for 1.

The zinc atoms in complex 1 adopts a distorted trigonal bipyramidal geometry (Fig. 2a) in a bi-metallacycle with the O–C–O of the two carboxylate ligands bridging the zinc metals with bond angles of 108.59 (5)° and 125.02 (14)° for O–Zn–O and O–C–O, respectively [31]. The two bridging carboxylate ligands have a Zn–O bond length of 1.9413 (11) Å and 1.9899 (12) Å [53,54]. The other two carboxylates form a syn-syn asymmetric chelate [30,55] with each zinc metal resulting in a Zn–O bond length of 1.9547 (12) Å and 2.5542 (13) Å. The Zn—N_py bond length and N–Zn–O bond angles in this geometry ranges between 2.029(2) – 2.046(2) Å and 98.59(7) – 112.55(7)°, respectively (Table 1). Complex 4 adopts a square pyramidal geometry where all four benzoate ligands bridged the zinc metals in a paddle wheel arrangement (Fig. 2b). The four O atoms of the carboxylate occupy the base with a Zn–O bond length of 2.0624(12) – 2.0813(12) Å and the ligand pyridinyl N stands at the apex of the pyramid with a Zn—N_py length of 2.0375 (14) Å and 2.0565 (13) Å.

5. Ring opening polymerization of cyclic esters

The initiation efficiency of the zinc complexes was examined for the polymerization of ϵ-CL, _L-LA and rac-LA at 110 °C. All the complexes showed considerable catalytic activity in mass/bulk and solvent polymerization as summarized in Table 3 and the kinetic plots are presented in Fig. 3. The activity of the complexes was compared within and across each triad since they differ basically in the electronic property of the ligand and co-ligand. The outcome of the experiment revealed that the polymerization rates relied on the inductive nature of the substituents on the benzoates within the triads while the induction time experienced by the complexes varied with the Schiff base ligand in the complex. For instance, complexes 7, 8 and 9 containing an electron withdrawing substituent in the Schiff base ligand had an induction period of 30 min while complexes 1, 4 and 6, as well as 2, 3 and 5 which possess electron donating groups on the ligand had an induction period of 4–6 h and about 18 h, respectively.

Table 3.

Summary of polymerization data of ε -CL catalyzed by complexes 1–9^a.

Entry	Complex	t (h	bConv (%)	cMn(calcd) (g mol−1)	bMn(NMR) (g mol−1)	dMn(GPC) (g mol−1)	dÐ	kapp (h−1)
1	1	29	98	22 371	3278	2950	1.95	0.1442
2	e1	10	98	22 371	2267	–	–	0.4144
3	2	49	98	22 371	3222	3848	1.86	0.1156
4	3	49	98	22 371	3162	3429	2.06	0.1173
5	4	15	99	22 599	2856	3160	1.95	0.3450
6	e4	6	99	22 599	1947	–	–	0.7626
7	f4	19	99	22 599	3137	–	–	0.1706
8	g4	5	98	22 371	1078	–	–	0.7806
9	5	48	98	22 371	3925	3811	2.19	0.1215
10	6	31	99	22 599	4144	4262	2.26	0.1782
11	7	15	97	22 143	3137	–	–	0.2369
12	e7	5	98	22 371	1076	–	–	0.7743
13	8	40	99	22 599	3586	–	–	0.1019
14	9	21	99	22 599	4175	–	–	0.2288
15	h4	40	99	22 599	3713	1285	1.72	0.1227
16	i4	40	99	22 599	5451	2760	2.05	0.1048
17	h7	40	99	22 599	2620	–	–	0.1389
18	i7	40	99	22 599	4109	–	–	0.115

^aPolymerization conditions: [M]: [I] = 200:1, 110 °C, Bulk.

^bDetermined from NMR.

^cM_n calculated from the M_wt of monomer x [M]₀: [I]₀ × Conv.,

^dObtained from GPC analysis and calibrated by polystyrene standard considering Mark-Houwink's corrections of 0.56 for PCL and 0.58 for PLAs.

^eBenzyl alcohol.

^fToluene and.

^gEthanol.

^h_L-LA.

ⁱrac-LA.

Fig. 3.

Kinetic plots of ln [CL]₀/[CL]_tvs. t for ε-CL polymerization in bulk using complexes 1–9 at 110 °C, [CL]₀: [I]₀ = 200.

The complexes with unsubstituted benzoate co-ligands were the most active in their triad; with complex 4 having the maximum k_app of 0.3450 h⁻¹ (Table 3, entry 5). A plausible reason for this observation can be traced to the lability of the Zn–O bond. Complexes containing the electron withdrawing fluoro group on the carboxylate ligand (Table 3, entries 4, 9 and 12) were more active than those with electron donating methyl group (Table 3, entries 3, 8 and 11) because they reduce the nucleophilicity at the metal center [56]. Complex 1 having two different zinc centers forming a rigid eight membered ring had a lower rate compared to the paddled-wheel conformation observed with complex 4.

Complexes 4 and 7, which are the most active for ε-CL polymerization were selected for polymerization of _L-LA and rac-LA in toluene under same conditions. The polymerization data and kinetic plots are provided in Table 3 and Fig. 4, respectively. The catalytic activity obtained with complex 7 was higher than that of complex 4 in the polymerization of rac-LA and _L-LA and the order was reversed for ε-CL. Both complexes afforded PLA of higher molecular weight with rac-LA (Table 3, entries 16 and 18) over _L-LA (Table 3, entries 15 and 17) whereas the k_app for _L-LA is greater than for rac-LA. As observed in the polymerization of ε-CL, the induction time of complex 4 in the polymerization of rac-LA and _L-LA was again longer than that of complex 7 because of the bulky isopropyl substituent in complex 4 which is known to hinder monomer access and forthwith decreased the catalytic activity [22,57]. However, the molecular weights of polymers attained with complex 4 are greater than those furnished by complex 7 (Table 3).

Fig. 4.

Kinetic plots of ln [LA]₀/[LA]_tvs. t for LA polymerization in bulk using complexes 4 and 7 at 110 °C, [LA]₀: [I]₀ = 200:1.

6. Effect of solvent and alcohol initiators

The influence of external co-initiator (BnOH) was examined with complexes 1, 4 and 7 using a [M]₀: [I]₀: [BnOH]₀ ratio of 200:1:1 at 110 °C (Fig. 5). The activities obtained after adding BnOH (Table 3, entries 2, 6 and 12) were higher than those obtained in its absence (Table 3, entries 1, 5 and 11). Similarly, the addition of ethanol as co-initiator to complex 4 increased catalytic activity (Table 3, entry 7). The remarkable increase in the catalytic activity with the addition of alcohol co-initiators is as a result of the production of alkoxides species which are known to increase the active sites [58]. However, the solution polymerization using complex 4 in toluene (Table 3, entry 8) resulted in a reduction of the catalytic rates compared to that obtained in bulk (Table 3, entry 5) because the concentration of reactive species involved in the polymerization reduced greatly, a phenomenon that has been reported in literature [25]. The ¹H NMR spectrum of PCL and PLA obtained with complex 4/alcohol systems shown in Fig. S34 and S35 confirmed a co-initiator-initiated ROP.

Fig. 5.

Kinetic plots of ln [CL]₀/[CL]_tvs. t for ε-CL polymerization in the presence of benzyl alcohol as initiator and toluene as solvent at 110 °C, [CL]₀/[I]₀ = 200.

7. Reaction order

The reaction order of complex 4 was examined by changing the catalyst concentration at a constant ε-CL concentration at 110 °C (Table 4)). From the semi-log kinetic plot (Fig. 6) for the metal complex studied, a linear relationship was observed, which corresponds to pseudo-first order reaction kinetics. The polymerization rate is represented with equation (3) below;

$$\frac{d\left[\mathrm{M}\right]}{dt}=k\left[\mathrm{M}\right]$$ (3)

where k = k_r [C]^y, k_r is the chain propagation rate, [C] is the initiator concentration and y, the reaction order.

Table 4.

Summary of polymerization data of ε-CL by complex 4 at different catalyst concentrations.

Entry	Complex	[M:I]	t(h)	bConv (%)	dMn(NMR)	kapp (h−1)
1	4	100	9	99	2689	0.5338
2	4	200	14	99	2856	0.3450
3	4	300	40	99	2903	0.1341
4	4	400	60	99	3131	0.0951

Fig. 6.

Plot of lnk_app against ln [4] for the determination of reaction order.

An increase in k_app as the catalyst concentration increased was observed, alluding to first order kinetics with respect to complex 4. A linear plot of lnk_app vs. ln [4] was obtained and the gradient of the line of best fit corresponds to the order of reaction (Fig. 6). The order with respect to complex 4 was found to be 0.7255 ≈ 0.7 and the rate law is represented by equation (4). The fractional order obtained with catalyst/initiator in melt polymerization is in tandem with literature reports which suggests an accumulation and dissociation of the active groups in the ROP reaction [32,[59], [60], [61]]. The ¹H NMR analysis of the polymer revealed a rise in polymer molecular weight from 2689 g mol⁻¹ to 3131 g mol⁻¹ as the monomer to catalyst ratio was changed from 100:1 to 400:1.

$$\frac{d\left[\mathrm{C}\mathrm{L}\right]}{dt}=k\left[\mathrm{C}\mathrm{L}\right]{\left[\mathbf{4}\right]}^{0.7}$$ (4)

7.1. Effect of temperature on ε-CL polymerization

The mass polymerization of ε-CL was further investigated by complex 4 between 80 °C and 110 °C to obtain the activation parameters. The results presented in Table 5 showed that the temperature significantly influenced the catalytic rate obtained. At low temperature a semi-log plot of ln [CL]₀/[CL]_t vs. time exhibit an extended induction period before a gradual linear increase in rate, and there is a ten-fold decrease in k_app from 0.345 h⁻¹ to 0.0355 as the temperature is changed from 110 °C to 80 °C.

Table 5.

Data for ε-CL polymerization at different temperatures catalyzed by complex 4^a.

Entry	Complex	[M:I]	t(h)	bConv (%)	bMn(NMR)	cMn(GPC)	cÐ	kapp (h−1)
1	d4	200	90	95	2000	–	–	0.0335
2	e4	200	62	99	2284	2152	1.58	0.0700
3	f4	200	33	99	2514	2304	1.80	0.1473
4	g4	200	15	99	2856	3160	1.95	0.3450

^aPolymerization conditions: 110 °C, Bulk.

^bDetermined from NMR.

^cObtained from GPC analysis and calibrated by polystyrene standard considering Mark-Houwink's corrections of 0.56 for PCL.

^d80 °C.

^e90 °C.

^f100 °C.

^g110 °C.

The conversion and molecular weight of PCL obtained at lower temperature were comparably lower than those obtained at elevated temperature. The curves in the chromatogram of PCL obtained from SEC analysis revealed a mono-modal pattern displayed at all the temperatures examined suggest that the polymerization was initiated by similar active sites [62,63]. Although the catalytic activity and obtained molecular weights of PCL increased at higher temperatures, the broadness of their curves is an indication of intra- and inter-molecular transesterification reactions which results in broad polymer dispersity (Fig. 7).

Fig. 7.

An overlay of the GPC chromatogram of PCL obtained by 4 at different temperatures.

The energy barrier ($E_a$) of 86.89 kJ mol⁻¹ was deduced from Arrhenius equation (5) from the gradient of the fitted line of the relationship between lnk and the inverse of the polymerization temperature (1/T) (Fig. 8). From the Eyring equation (6) the computed thermodynamic parameters are enthalpy (${\Delta H}^{\#}$) 83.84 kJ mol⁻¹ and entropy (${\Delta S}^{\#}$) 105.49 J K⁻¹ mol⁻¹ (Fig. 8). These activation parameters obtained conforms to an organized transition state corresponding to a coordination-insertion mechanism.

$$\ln k=\ln A+\frac{-E_a}{RT}$$ (5)

$$\ln \left(\frac{K_{\mathrm{a}\mathrm{p}\mathrm{p}}}{T}\right)=\frac{-{\Delta H}^{\#}}{R}\frac{1}{T}+\ln \frac{K_b}{h}+\frac{{\Delta S}^{\#}}{R}$$ (6)

Fig. 8.

Arrhenius and Eyring plot for the bulk polymerization of ε-CL initiated by 4.

7.1.1. GPC molecular weight and molecular weight distribution of polymers

Corrected GPC molecular weights (M_n) and poly dispersity index (Ð) of chosen catalysts are comparable to those obtained from ¹H NMR (Table 3) with only a few variations. Low molecular weight polymers of 2950–4262 g mol⁻¹ and 1285–2760 g mol⁻¹ were obtained from GPC analysis for PCL and PLAs, respectively. Dispersity between 1.80 – 2.26 and 1.72–2.05 recorded for PCL and PLAs, respectively, presupposes the presence of intra- and inter-molecular trans-esterification reactions. The steric and electronic properties of the Schiff base and carboxylate ligands in the complexes influenced the M_n obtained from GPC analysis. For instance, complexes 4–6 with ortho bulky isopropyl substituent gave M_n of 3160 g mol⁻¹ (4, unsubstituted) < 3811 g mol⁻¹ (5, electron donating) < 4262 g mol⁻¹ (6, electron withdrawing). A slight deviation in the above trend was recorded for complexes 1–3 containing a less bulky substituent; where complex 2 (electron donating) had a slightly higher M_n of 3848 g mol⁻¹ than M_n of 3429 g mol⁻¹ obtained for 3 (electron withdrawing) while 1 (unsubstituted) remained the least in the triad with M_n of 2950 g mol⁻¹ which is traceable to the distorted tetrahedral geometry.

7.2. End group analysis and reaction mechanism

End groups of polymers obtained with complexes 4 and 7 at 110 °C using a [M]₀: [I]₀ ratio of 200:1 were analyzed with ¹H NMR (Fig. S34 and S35) and ESI-MS (Fig. S36). The results support a coordination insertion mechanism as proposed in Fig. 9. The process of monomer enchainment begins by monomer activation at the metal active site. This is preceded by scission of the monomer acyl-oxygen linkage via insertion in the M—O bond. The ancillary organic ligand stay connected to the catalytic center controlling the activity, polymer molecular weights and molecular weights distributions, monomer addition and stereochemistry. Finally, the propagation is terminated by hydrolysis, where the metal is dislodged from the growing chain giving hydroxyl end capped polymers [22,34]. When benzyl alcohol was included as a co-initiator, the ¹H NMR spectrum of PCL (Fig. S34) revealed additional methylene proton (g, 5.12 ppm) and aromatic proton (h, 7.37 ppm) peaks signifying generation of alkoxide initiating species (BnO⁻) responsible initiating of the polymerization reaction via a coordination insertion pathway.

Fig. 9.

Proposed mechanism for the ROP of ε-CL.

8. Stereochemistry of poly-lactides

The thermal and mechanical behavior of synthesized PLA is informed by their microstructure, which is determined by the positioning of methyl groups on the lactides (_L-LA, _D-LA, rac-LA and meso-LA), and the chirality of the catalyst employed. Homo-nuclear decoupled ¹H and ¹³C NMR was employed for the microstructural analysis of the synthesized PLA and the tetrad sequence were assigned according to established reports for PLA in literature [64,65]. The homo-nuclear decoupled ¹H NMR spectra of poly (_L-LA) revealed an essentially isotactic PLA whereas the poly (rac-LA) points to an isotactic stereo-block microstructural conformation (Fig. S37, S38, S39 and S40). The singular methine and carbonyl carbon ¹³C NMR peaks of poly (_L-LA) obtained by 4 at 69.03 and 169.61 ppm, respectively, confirms the isotactic nature of the PLA (Fig. S41a and S42bESI).

However, two methine carbon peaks were observed at 69.19 ppm and 69.02 ppm for poly (rac-LA) were assigned to mrm and (mmm, mmr, mrr, rmr) sequences (Fig. S41bESI). More so, carbonyl carbon peaks at 169.61 ppm, 169.33–169.42 ppm and 169.16 ppm are assigned to rr, (rm, mr) and mm configurations (Fig. S42bESI) [34,65,66]. The physicochemical properties of the synthesized PLA examined by thermal analysis using DSC-TGA is summarized in Table 6. The thermogram had melting temperatures (T_m) between 116.58 and 188.03 °C (Fig. S43) [67]. The detected decomposition temperatures between 278.78 °C and 331.32 °C is suggestive of an isotactic PLA with a metal capped end, which began to decompose at 214.74 °C till the maximum decomposition temperature (T_max) at 436.01 °C resulting in a T_decomp of 221.27 °C (Fig. S43).

Table 6.

TGA data for poly (L-LA) and poly (rac-LA) synthesized by complexes 4 and 7^a.

Entry	Onset (°C)	50% (°C)	Tmax (°C)	ΔTdecomp (°C)	Tm (°C)
bL-LA	201.6	304.3	410.5	208.9	142.96
brac-LA	214.7	326.4	436.1	221.4	188.03
cL-LA	179.7	268.2	402.5	222.8	116.58
crac-LA	181.9	282.7	398.2	216.3	144.41

^aHeating rate of 10 °C/min.

^bComplex 4 and.

^cComplex 7.

9. Conclusion

Nine Zn(II) carboxylate complexes of 4-pyridinyl Schiff base ligands were obtained from a room temperature reaction of Zn(II) acetate with the Schiff base and carboxylate ligands. Their geometry and coordination were established by spectroscopic studies (FTIR, NMR and mass spectroscopy) and elemental analysis. The crystal structures of complexes 1 and 4 were ascertained by single crystal X-ray diffraction studies to be dinuclear with the Schiff base ligands coordinating in an unidentate manner. On the one hand, the carboxylate co-ligands in complex 1 coordinated in monodentate and bridging modes resulting in a distorted tetrahedral geometry around each Zn(II) centers while all the carboxylate co-ligands in complex 4 coordinated in a bridging mode occasioning a paddle wheel arrangement on the Zn(II) atoms. The Zn(II) complexes were catalytic active towards the polymerization of ε-CL, _L-LA and rac-LA affording low molecular weight polymers with or without alcohol initiators. Ligand and co-ligand electronic and steric effects significantly influenced the catalytic activity of the complexes. Homo-nuclear decoupled ¹H NMR and DSC-TGA analysis of the synthesized PLA revealed an isotactically enriched PLA.

Declaration

Author contribution statement

Damilola C. Akintayo: Conceived and designed the experiments; Performed the experiments; Analyzed and interpreted the data; Wrote the paper. Wisdom A. Munzeiwa: Conceived and designed the experiments; Analyzed and interpreted the data. Sreekantha B. Jonnalagadda, Bernard Omondi: Conceived and designed the experiments; Analyzed and interpreted the data; Contributed reagents, materials, analysis tools or data.

Funding statement

This work was supported by National Research Foundation.

Data availability statement

The data that support the findings of this study are openly available in Cambridge Crystallographic Data Centre at http://www.ccdc.cam.ac.uk/conts/retrieving.html, or via e-mail: deposit@ccdc.cam.ac.uk, reference number (CCDC 2191381–2191382).

Declaration of interest's statement

The authors declare no competing interests.

Appendix A. Supplementary data

The following is the supplementary data related to this article: