Lactide as the Playmaker of the ROP Game ... - ACS Publications

Jan 20, 2017 - Lactide as the Playmaker of the ROP Game: Theoretical and. Experimental ... metal center, and OR is the initiating group. L creates a t...
0 downloads 0 Views 7MB Size
Article pubs.acs.org/IC

Lactide as the Playmaker of the ROP Game: Theoretical and Experimental Investigation of Ring-Opening Polymerization of Lactide Initiated by Aminonaphtholate Zinc Complexes Dawid Jędrzkiewicz,† Grazẏ na Adamus,‡ Michał Kwiecień,‡ Łukasz John,† and Jolanta Ejfler*,† †

Faculty of Chemistry, University of Wroclaw, Fryderyka Joliot-Curie 14, 50-383 Wroclaw, Poland Centre of Polymer and Carbon Materials Polish Academy of Science, Marii Curie-Skłodowskiej 34, 41-819 Zabrze, Poland



S Supporting Information *

ABSTRACT: A family of homo- and heteroleptic zinc complexes bearing aminonaphtholate ligands was synthesized and fully characterized. Using NMR spectroscopy and DFT calculation, bisalkoxy-bridged complexes [LZn(μ-OR)]2 were confirmed to have dimeric structures in solution, analogous to those obtained via X-ray crystallography. Surprisingly, a detailed experimental and theoretical study of the catalytic activity of [LZn(μ-OR)]2 in the ring-opening polymerization (ROP) of lactides showed that although well-defined alkoxy dimers possess a single-site structural motif, the most active initiator is obtained during in situ alcoholysis of the alkylzinc precursor. These results indicate that rational ancillary and alkoxy ligand design that takes into account its mutual interaction on monomer coordination may be key to the synthesis of new highperformance ROP catalysts.



INTRODUCTION The so-called green polyesters have multiple biomedical and environmental commercial applications. The academic and industrial interest in these polyesters stems from two traits: the possibility to synthesize them with “by design” properties and their renewable, biodegradable nature.1 Polylactide (PLA), probably the most prominent of them, has found its major applications in ecological packaging, agricultural materials, active drug delivery, and medical devices. The controlled synthesis of PLA is accomplished through metal-catalyzed/-initiated ring-opening polymerization (ROP) of lactide. This is a widely used method to prepare well-defined polyesters with competitive applications. The most attractive so far are the so-called single-site initiators with the general formula L−M−OR, where L is the ancillary ligand, M is the metal center, and OR is the initiating group. L creates a tunable environment, and the stability of M prevents aggregation and undesired side reactions. In the context of medical applications, biometal complexes (Zn, Mg, Ca) are welcome due to their innocuous nature, ready availability, and effectiveness for polymerization, in terms of both activity and stereoselectivity.2 However, the desired heteroleptic complexes (L−M−OR) for Zn, Mg, and Ca are kinetically labile and may undergo deactivation through Schlenk-type equilibrium, redistribution, or bis-chelatation reactions.4e,h Therefore, judicious choice of metal and ligand combinations is crucial both to optimize the activity and stereocontrol of initiators in ROP and to ensure the appropriate cost and biological tolerance. © XXXX American Chemical Society

Among the zinc initiators that have been studied in greatest detail are those involving the N,O-donor ligands: tris(pirazolyl)borate, β-diketiminate, and amino-/iminophenolate.3−5 Recently, aminophenolate ligands have become the most trendy in the construction of single-site initiators. Welldefined zinc complexes containing these ligands usually present dimeric structures. Monomeric structures are considerably less common. Most catalytically active compounds are generated in situ in the direct reaction between alkyl zinc complexes (LZnR) and alcohols. The potential differences in catalytic activity between alkyl (LZnR) and alkoxy (LZnOR) compounds are commonly explained as the time-consuming transformation during an alcoholysis reaction. However, the answer is not always unequivocal, because heteroleptic alkylzinc compounds have a tendency to undergo dynamic behavior in solution, with the equilibrium position depending on the properties of the ancillary ligands, the polarity of the solvent, and so on. Therefore, structural information about compounds crystallized from such a mixture is not always representative of the structure of the active species present in solution. We recently reported on the influence of the architecture of the aminophenolate/-naphtholate ligands on the design of the structural motif of zinc complexes. All of our earlier obtained that alkyl dimeric aminophenolate zinc complexes mediate the Received: October 11, 2016

A

DOI: 10.1021/acs.inorgchem.6b02439 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

Anal. Calcd (Found) for C36H56N2O2Zn: C, 70.40 (70.06); H, 9.19 (9.30); N, 4.56 (4.72). 1H NMR (500 MHz, C6D6, RT): δ = 7.37 (dd, JHH = 8.5, 2.6 Hz, 2H, ArH), 7.17 (d, JHH = 7.0 Hz, 2H, ArH), 7.00 (d, JHH = 2.5 Hz, 2H, ArH), 3.93 (br, 2H, N−CH2−Ar), 3.35 (br, 2H, N− CH2−Ar), 2.71 (s, 2H, N−CH), 1.95 (s, 6H, N−CH3), 1.74−0.76 (m, 20H, CH2), 1.41 (s, 18H, C(CH3)3). 13C NMR (75 MHz, C6D6, RT): δ = 165.4 (ArC−OH, 2C), 136.2 (ArC−C, 2C), 128.0 (ArCH, 2C), 127.6 (ArCH, 2C), 120.7 (ArC−CH2, 2C), 120.0 (ArCH, 2C), 64.3 (N−CH, 2C), 60.9 (N−CH2, 2C), 37.2 (N−CH3, 2C), 33.9 (C(CH3)3, 2C), 32.2 (C(CH3)3, 6C), 26.2 (CH2, 10C). (L2)2Zn. To a stirred solution of (L2ZnEt)2 (0.72 g, 1 mmol) in toluene (30 mL) MeOH (80 μL, 2 mmol) was added dropwise at ambient temperature. The solution was stirred for 1 h, and then it was placed at −15 °C until colorless crystals appeared as product. It was filtered off and dried in vacuo. Yield 77% (0.46 g, 0.8 mmol, based on the ligand). Anal. Calcd (Found) for C36H44N2O2Zn: C, 71.81 (71.65); H, 7.37 (7.40); N, 4.65 (4.91). 1H NMR (500 MHz, C6D6, RT): δ = 9.01 (dd, JHH = 14.0, 8.4 Hz, 2H, ArH), 7.85 (d, JHH = 8.0 Hz, 2H, ArH), 7.49 (dd, JHH = 13.8, 6.9 Hz, 2H, ArH), 7.43 (dd, JHH = 10.7, 3.9 Hz, 2H, ArH), 7.23 (d, JHH = 7.9 Hz, 2H, ArH), 7.08−7.03 (m, 2H, ArH), 4.51 (s, 2H, N−CH2−Ar), 3.19 (s, 2H, N−CH2−Ar), 2.87 (s, 2H, N− CH), 1.77 (s, 6H, N−CH3), 1.97−0.53 (m, 20H, CH2). 13C NMR (75 MHz, C6D6, RT): δ = 164.5 (ArC−OH, 2C), 136.9 (ArC, 2C), 130.4 (ArCH, 2C), 129.8 (ArCH, 2C), 127.5 (ArCH, 2C), 126.4 (ArC, 2C), 125.1 (ArCH, 2C), 123.6 (ArCH, 2C), 113.4 (ArCH, 2C), 112.3 (ArC−CH2, 2C), 65.0 (N−CH, 2C), 60.6 (N−CH2, 2C), 36.6 (N− CH3, 2C), 25.8 (CH2, 10C). (L2ZnOBn)2. To a stirred solution of (L2ZnEt)2 (0.72 g, 1 mmol) in toluene (30 mL) benzyl alcohol (0.21 mL, 2 mmol) was added dropwise at ambient temperature. The solution was stirred for 1 h, and then it was placed at −15 °C until colorless crystals appeared as product. It was filtered off and dried in vacuo. Yield 87% (0.77 g, 0.9 mmol). Anal. Calcd (Found) for C50H58N2O4Zn2: C, 68.11 (68.23); H, 6.63 (6.87); N, 3.18 (3.25). 1H NMR for four major sets of signals (600 MHz, C6D6) δ = 9.70 (d, J = 8.3 Hz, 1H, Ar−H), 9.24 (d, J = 8.8 Hz, 1H, Ar−H), 9.16 (d, J = 8.2 Hz, 1H, Ar−H), 8.63 (d, J = 8.5 Hz, 1H, Ar−H), 8.03−6.62 (m, 38H, Ar−H), 6.15−6.10 (m, 2H, Ar−H), 5.55 (d, J = 11.1 Hz, 1H, O−CH2−Ar), 5.45 (d, J = 10.8 Hz, 1H, O−CH2− Ar), 5.43 (d, J = 11.1 Hz, 1H, O−CH2−Ar), 5.25 (d, J = 11.5 Hz, 1H, O−CH2−Ar), 4.91 (d, J = 11.1 Hz, 1H, O−CH2−Ar), 4.56 (d, J = 11.1 Hz, 1H, O−CH2−Ar), 4.47 (d, J = 11.4 Hz, 1H, O−CH2−Ar), 3.10 (d, J = 11.3 Hz, 1H, O−CH2−Ar), 3.93 (d, J = 12.4 Hz, 1H, N− CH2−Ar), 3.85 (d, J = 12.1 Hz, 1H, N−CH2−Ar), 3.80 (d, J = 12.4 Hz, 1H, N−CH2−Ar), 3.79 (d, J = 11.8 Hz, 1H, N−CH2−Ar), 3.46 (t, J = 13.6 Hz, 1H, N−CH), 3.31−3.18 (m, 1H, N−CH), 3.08 (d, J = 11.7 Hz, 1H, N−CH2−Ar), 3.07 (d, J = 11.8 Hz, 1H, N−CH2−Ar), 3.02 (d, J = 12.4 Hz, 1H, N−CH2−Ar), 2.89 (d, J = 11.9 Hz, 1H, N− CH2−Ar), 2.24−2.16 (m, 1H, N−CH), 1.90−1.80 (m, 1H, N−CH) 3.30−0.37 (m, 40H, CH2), 1.57 (s, 3H, N−CH3), 1.53 (s, 3H, N− CH3), 1.38 (s, 3H, N−CH3), 1.33 (s, 3H, N−CH3). 13C NMR for four major sets of signals, based on HMQC (75 MHz, C6D6, RT): δ = 130.3 (ArCH, 2C), 129.9 (ArCH, 22C), 129.7 (ArCH, 1C), 129.4 (ArCH, 1C), 129.0 (ArCH, 2C), 128.2 (ArCH, 2C), 127.4 (ArCH, 2C), 126.0 (ArCH, 4C), 124.4 (ArCH, 2C), 123.2 (ArCH, 2C), 113.6 (ArCH, 1C), 113.2 (ArCH, 1C), 113.0 (ArCH, 1C), 112.6 (ArCH, 1C), 69.3 (O−CH2−Ar, 1C), 69.0 (O−CH2−Ar, 1C), 68.8 (O− CH2−Ar, 1C), 68.4 (O−CH2−Ar, 1C), 65.6 (N−CH, 1C), 64.3 (N− CH, 1C), 64.1 (N−CH, 1C), 63.8 (N−CH, 1C), 60.1 (N−CH2−Ar, 1C), 60.0 (N−CH2−Ar, 1C), 59.9 (N−CH2−Ar, 1C), 59.8 (N− CH2−Ar, 1C), 35.7 (N−CH3, 1C), 35.1 (N−CH3, 1C), 35.0 (N− CH3, 1C), 34.2 (N−CH3, 1C), 26.2 (CH2, 10C), 24.5 (CH2, 10C), 23.5 (CH2, 10C), 22.3 (CH2, 10C). (L2ZnOPe)2. To a stirred solution of (L2ZnEt)2 (0.72 g, 1 mmol) in toluene (30 mL) 1-phenylethanol (0.24 mL, 2 mmol) was added dropwise at ambient temperature. The solution was stirred for 1 h, and then it was placed at −15 °C until colorless crystals appeared as product. It was filtered off and dried in vacuo. Yield 82% (0.75 g, 0.8 mmol).

ROP of lactide with high efficiency at room temperature in a controlled fashion.5p Appropriate structural and catalytic studies have now been performed using an alkoxy-related series of zinc complexes. In comparative studies, the ROP of LA for alkoxide analogs was expected to be faster but the order was found to be reversed. The reaction time needed for the formation of alkoxide species may be the answer. A plausible solution may be that in-situ-generated initiator is more active and is not a classical heteroleptic zinc alkoxide (LZnOR)2 but rather a compound with a more sophisticated structure that was obtained exclusively in lactide companionship. Here, we report the details of the synthesis and characterization of new heteroleptic dimeric zinc complexes. The structural motifs of the complexes have been confirmed by Xray diffraction studies and detailed DFT analysis and compared with the corresponding species in solution. We aimed to establish whether the obtained single-site initiators display a higher level of activity than the corresponding active complex generated via direct alcoholysis reaction. The experimental and theoretical data presented here should allow for a more precise design of effective initiators based on an evaluation that ensures the fit of both ancillary and alkoxy ligands during an ROP reaction.



EXPERIMENTAL SECTION

General Materials, Methods, and Procedures. All reactions and operations were performed under an inert atmosphere of N2 using a glovebox (MBraun) or standard Schlenk apparatus and vacuum line techniques. The solvents were purified by standard methods: toluene, distilled from Na; CH2Cl2, distilled from P2O5; hexanes, distilled from Na; methanol, distilled from Mg; deuterated solvents (C6D6, toluened8), distilled from CaH2. Benzyl alcohol (>99% Aldrich) and 1phenylethanol (98% Aldrich) were freshly distilled under dry N2 and freeze/thraw degassed prior to use; L-LA ((3S)-cis-3,6-dimethyl-1,4dioxane-2,5-dione) (98% Aldrich) was sublimed and recrystallized from toluene prior to use. ZnEt2 (1.0 M solution in hexanes), Nmethylcyclohexylamine (98%), formaldehyde (37% solution in H2O), 1-naphthol, and 4-tert-butylphenol were purchased from Aldrich and used as received. 1H and 13C NMR spectra were detected at the temperature range from 233 to 333 K using a Bruker ESP 300E or 500 MHz spectrometer. Chemical shifts are reported in parts per million and referenced to the residual protons in the deuterated solvents. The number-average molar mass (Mn) and the molecular weight distribution index of the samples were determined by gel-permeation chromatography (GPC). The system was composed of a Viscotek VE 1122 solvent delivery system and a Shodex SE-61 refractive index detector. The analyses were performed in chloroform at 35 °C at a flow rate of 1 mL/min with the aid of a PLgel 3 μm MIXED-E (Polymers Laboratories) high-efficiency column (300 mm × 7.5 mm). The injection volume was 100 μL of the sample in chloroform (0.3% w/v). Polystyrene standards (Polymer Laboratories) with narrow molar mass distributions were used to generate a calibration curve and to compute the average molar masses and dispersities of the sample. Microanalyses were conducted with an ARL Model 3410+ ICP spectrometer (Fisons Instruments) and a VarioEL III CHNS (inhouse). Syntheses. N-[Methyl(2-hydroxy-5-tert-butylphenyl)]-N-methylN-cyclohexylamine (L1-H), N-[Methyl(1-naphthol)]-N-methyl-N-cyclohexylamine (L2-H), (L1ZnEt)2, and (L2ZnEt)2. The named compounds were synthesized according to literature procedures.5p (L1)2Zn. To a stirred solution of (L1ZnEt)2 (0.73 g, 1 mmol) in toluene (30 mL) MeOH (80 μL, 2 mmol) was added dropwise at ambient temperature. The solution was stirred for 1 h, and then it was placed at −15 °C until colorless crystals appeared as product. It was filtered off and dried in vacuo. Yield 73% (0.45 g, 0.7 mmol, based on the ligand). B

DOI: 10.1021/acs.inorgchem.6b02439 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

in C6D6. After reaction was completed an excess of hexanes was added to the reaction mixture. Filtration and vacuum drying yielded a white polymer. The resulting solid was dissolved in dichloromethane, and the polymer was precipitated with excess cold methanol. The polymer was collected by filtration, washed with methanol to remove unreacted monomer, and dried under reduced pressure. The reaction mixtures were prepared in a glovebox; then subsequent operations were performed by means standard Schlenk techniques. Details of X-ray Data Collection and Reduction. X-ray diffraction data for a suitable crystal of each sample were collected using a KUMA KM4 CCD and Xcalibur CCD Onyx or Ruby (see Supporting Information Table S1) with ω scan technique. Data collection and processing utilized the CrysAlis suit of programs.9 The space groups were determined based on systematic absences and intensity statistics. Lorentz polarization corrections were applied. The structures were solved by direct methods and refined by full-matrix least-squares on F2. All calculations were performed using the SHELXTL-2013 suite of programs.10 All non-hydrogen atoms were refined with anisotropic displacement parameters. Hydrogen atom positions were calculated with geometry and not allowed to vary. Thermal ellipsoid plots were prepared with 30% of probability displacements for non-hydrogen atoms by using the Mercury 3.1 program.11 All data have been deposited with the Cambridge Crystallographic Data Centre CCDC-1495734 for (L2ZnOBn)2, CCDC-1495735 for (L2ZnOPe)2, CCDC-1495736 for (L2)2Zn, and CCDC-1495737 for (L1)2Zn. Copies of the data can be obtained free of charge by application to CCDC, 12 Union Road, Cambridge CB21EZ, UK or e-mail: [email protected]. Computational Details. All density functional theory (DFT) calculations were performed with the program suite Gaussian 03.12 The geometries of the complexes were optimized by using B3LYP density functional theory and the 6-31G** basis sets, implemented in Gaussian 03 on all atoms.13,14 The starting geometries of complexes (LZnOR)2 and L2Zn were generated from their crystal structures, whereas the starting geometries of adequate species were derived from their optimized complexes. Structures of all isomers of calculated complexes were optimized in vacuo. Frequency calculations confirmed the stationary points to be minimal.

Anal. Calcd (Found) for C52H62N2O4Zn2: C, 68.65 (68.82); H, 6.87 (7.12); N, 3.08 (3.11). 1H NMR for four major sets of signals (500 MHz, C6D6) δ = 9.33 (d, J = 6.6 Hz, 1H, Ar−H), 9.32 (d, J = 7.7 Hz, 1H, Ar−H), 9.24 (d, J = 7.6 Hz, 1H, Ar−H), 9.20 (d, J = 8.3 Hz, 1H, Ar−H), 7.92−6.54 (m, 40H, Ar−H), 5.29 (q, J = 6.3 Hz 1H, O−CH− Ar), 5.24 (q, J = 6.2 Hz, 1H, O−CH−Ar), 5.18 (q, J = 6.3 Hz, 1H, O− CH−Ar), 5.03 (d, J = 6.3 Hz, 1H, O−CH−Ar), 3.95 (d, J = 12.3 Hz, 1H, N−CH2−Ar), 3.62 (d, J = 12.5 Hz, 1H, N−CH2−Ar), 3.59 (d, J = 12.5 Hz, 1H, N−CH2−Ar), 3.32 (d, J = 12.7 Hz, 2H, N−CH2−Ar), 3.22 (d, J = 12.5 Hz, 1H, N−CH2−Ar), 3.20 (d, J = 12.5 Hz, 1H, N− CH2−Ar), 3.03 (d, J = 12.5 Hz, 1H, N−CH2−Ar), 2.75−2.67 (m, 1H, N−CH), 2.67−2.58 (m, 1H, N−CH), 2.43−2.35 (m, 1H, N−CH), 2.14−2.07 (m, 1H, N−CH), 2.23 (d, J = 6.4 Hz, 3H, C−CH3), 2.02 (d, J = 6.4 Hz, 1H C−CH3), 1.94 (d, J = 6.4 Hz, 1H C−CH3), 1.61 (d, J = 7.9 Hz, 1H C−CH3), 2.96−0.38 (m, 40H, CH2), 1.85 (s, 3H, N− CH3), 1.79 (s, 3H, N−CH3), 1.57 (s, 3H, N−CH3), 1.42 (s, 3H, N− CH3). 13C NMR for four major sets of signals, based on HMQC (75 MHz, C6D6, RT): δ = 130.3 (ArCH, 2C), 130 (ArCH, 1C), 129.7 (ArCH, 1C), 129.3 (ArCH, 1C), 128.6 (ArCH, 2C), 128.3 (ArCH, 2C), 127.4 (ArCH, 2C), 127.3 (ArCH, 20C), 126.0 (ArCH, 2C), 124.9 (ArCH, 2C), 124.6 (ArCH, 2C), 123.8 (ArCH, 2C), 123.1 (ArCH, 2C), 112.9 (ArCH, 1C), 74.2 (O−CH−Ar, 2C), 73.7 (O− CH−Ar, 1C), 73.0 (O−CH−Ar, 1C), 64.7 (N−CH, 2C), 63.8 (N− CH, 2C), 60.2 (N−CH2−Ar, 1C), 59.9 (N−CH2−Ar, 1C), 59.6 (N− CH2−Ar, 1C), 59.1 (N−CH2−Ar, 1C), 36.3 (N−CH3, 2C),, 34.9 (N−CH3, 2C), 25.8 (CH2, 2C), 25.3 (CH2, 2C), 23.6 (C−CH3, 1C), 22.6 (C−CH3, 1C), 22.4 (C−CH3, 1C), 22.0 (C−CH3, 1C). Representative Procedure for Solution Polymerization. ROP of L-Lactide Initiated by “Single-Site” Zinc Alkoxides: (LZnOR)2 (where OR = OBn, OPe). The solution of zinc complex (LZnOR)2 in CH2Cl2 (20 mL) was placed in a Schlenk flask, and next L-lactide in a molar ratio of zinc center [Zn] to L-lactide [Zn]/L-LA = 1/100 was added. The resulted solution was stirred at room temperature for a prescribed time. At certain time intervals (80 min), about 1 mL aliquots were removed, precipitated with hexanes, and dried in vacuo. Representative procedure for (L2ZnOBn)2: [I]/L-LA = 0.5/100; (L2ZnOBn)2 (0.03 g, 0.03 mmol), L-LA (0.86 g, 6.00 mmol), time: 8 h. ROP of L-Lactide Initiated by in-Situ-Generated Initiators: (LZnEt)2/ROH (where OR = OMe, OBn, OPe). Procedure A. The solution of zinc precatalyst (LZnEt)2 in CH2Cl2 (20 mL) was placed in a Schlenk flask, and next L-lactide and appropriate alcohol were added simultaneously. The fixed molar ratio of zinc center [Zn] to L-lactide and alcohol: [Zn]/L-LA/ROH = 1/100/1. The resulted solution was stirred at room temperature for a prescribed time. At certain time intervals (10 min), about 1 mL aliquots were removed, precipitated with hexanes, and dried in vacuo. Procedure B. The solution of zinc precatalyst (LZnEt)2 in CH2Cl2 (20 mL) was placed in a Schlenk flask, and appropriate alcohol was added. Next, after 15 min L-lactide was added. The fixed molar ratio of zinc center [Zn] to L-lactide and alcohol: [Zn]/L-LA/ROH = 1/100/1. The resulted solution was stirred at room temperature for a prescribed time. At certain time intervals (10 min.), about 1 mL aliquots were removed, precipitated with hexanes, and dried in vacuo. Representative procedure for (L2ZnEt)2/BnOH: [I]/L-LA/BnOH = 0.5/100/1; (L2ZnEt)2 (0.02 g, 0.03 mmol), L-LA (0.86 g, 6.00 mmol), BnOH (6 μL, 0.06 mmol), time 60 min. ROP of L-Lactide Initiated by Homoleptic Compounds with External Alcohol: L2Zn/ROH (where OR = OMe, OBn, OPe). The solution of L2Zn in CH2Cl2 (20 mL) was placed in a Schlenk flask, and next L-lactide and appropriate alcohol were added simultaneously. The fixed molar ratio of zinc center [Zn] to L-lactide and alcohol: [Zn]/LLA/ROH = 1/100/1. The resulted solution was stirred at room temperature for a prescribed time. At certain time intervals (10 min), about 1 mL aliquots were removed, precipitated with hexanes, and dried in vacuo. Representative procedure for (L2)2Zn/BnOH: [I]/LLA/BnOH = 1/100/1; (L2)2Zn (0.04 g, 0.06 mmol), L-LA (0.86 g, 6.00 mmol), BnOH (6 μL, 0.06 mmol), time 60 min. The conversion was determined while observing 1H NMR resonances of the polymer and monomer by dissolving the precipitates



RESULTS Synthesis and Characterization of the Complexes. The new heteroleptic alkoxide zinc complexes were synthesized in a two-step process: reaction of diethyl zinc with the proligand L2H to form the dimeric ethyl zinc complexes (L2ZnEt)2 followed by alcoholysis with selected ROH (methanol MeOH, benzyl alcohol, BnOH, 1-phenylethanol, PeOH) to yield the planned heteroleptic compounds (LZnOR)2, corresponding to a singlesite initiator structure for RO = BnO and PeO (Scheme 1). Benzyl alcohol is the most popular end group of PLA formed in the presence of aminophenolate metal complexes. The choice of methanol for the alcoholysis reaction was not accidental: we wished to verify the premise from the literature that the synthesis of homoleptic compounds is the most probable pathway for this reaction.5n In the presence of MeOH, the (L2ZnEt)2 complex undergoes a transformation to a mixture of compounds, but homoleptic (L2)2Zn in crystallized form is the only product that can be identified. Furthermore, the reaction of methanol with ethyl-aminophenolate zinc dimer (L1ZnEt)2, which is similar to the (L2ZnEt)2 complex that we studied here, also only yielded a homoleptic compound (L1)2Zn, i.e., it was the only product that could be isolated from the reaction mixture (Scheme 2). All synthesized zinc compounds are colorless, air and moisture sensitive, and soluble in chlorinated hydrocarbons, THF, and toluene. The complexes were all assessed by standard elemental analysis, NMR spectra, and single-crystal X-ray C

DOI: 10.1021/acs.inorgchem.6b02439 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry Scheme 1. Synthesis of Heteroleptic Zinc Complexes

Scheme 2. Synthesis of Homoleptic Zinc Complexes

diffraction. Suitable monocrystals for single-crystal X-ray analysis were crystallized from crude reaction solution. The solid-state structures of (L2ZnOBn)2, (L2ZnOPe)2, (L2)2Zn, and (L1)2Zn are shown in Figures 1, 2, 3, and 4, respectively, and Table 1 with summarized crystal data. The single-crystal structures of the heteroleptic complexes (L2ZnOBn)2 and (L2ZnOPe)2 reveal the expected dinuclear nature for the alkoxy-bridged compounds with four coordinated zinc centers. The metal centers in these complexes adopt distorted tetrahedral geometries, bridged by the oxygen atoms of the alkoxide ligands to form the core rhomboid Zn2O2. A comparison of the angles within the Zn2O2 core indicates that the O−Zn−O and Zn−O−Zn bond angles are similar, while the angle of O1−Zn−O2 in (LZnOBn)2 is slightly narrower than that in (L2ZnOPe)2: 113.93(7)° and 117.79(7)°, respectively. The bond distances and angles between zinc and O, N, and C atoms are all comparable to those observed in a related alkoxy-bridged zinc species described previously.5h,j It is noteworthy that while a significant set of ethyl zinc aminophenolate/-naphtholate complexes (LZnEt)2 have been confirmed through X-ray diffraction study, there are only a few examples in the corresponding set of related alkoxy dimers (LZnOR)2.5d,h,j Methanolysis of the starting dimers (L2ZnEt)2 and (L1ZnEt)2 with a stoichiometric quantity of methanol yielded the mononuclear homoleptic compound. The molecular structure of the homoleptic complexes (L2)2Zn and (L1)2Zn are similar to previously reported aminophenolate homoleptic zinc complexes.5f,h,n The compounds contain two O,N-chelating ligands that form six-membered rings around a

Figure 1. Molecular structure of (L2ZnOBn)2. Thermal ellipsoids are drawn at a 30% probability level. H atoms are excluded for the sake of clarity.

zinc atom. Selected geometrical parameters for these compounds are given in Table 1. We recently reported the structural details of heteroleptic alkylzinc dimers (LZnEt)2 that were from a study aiming to establish guidelines for rational ancillary ligands designed in the context of catalytic activity in the ROP of cyclic esters. The role of alkoxy ligands in single-site initiators is well established and clear in the monomeric complex L−M−OR. However, the distinct role of the OR group in dimeric species remains rather D

DOI: 10.1021/acs.inorgchem.6b02439 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

alcoholysis reaction. The most desired structure for single-site initiators (L−Zn−OR) has been confirmed here through X-ray analysis, but these complexes that are well defined in the solid state are not as clear as in solution (Figure 5). A distinctive fragment of the 1H NMR spectroscopic feature of (L2ZnOBn)2 includes four sets of doublets of methylene backbones: h mark for OBn fragment, i-mark for L2 (Figure 5). The latter signals (lower intensities) are obscured and cannot be distinguished in 2D NMR spectra (see Figure 5 and Figures S1−4). The pattern of the spectrum suggests the existence of an equilibrium between the different forms of (L2ZnOBn)2 complex. This surmise is based on our earlier results obtained for complexes bearing this ligand family, where the behavior observed in solution arises from the coexistence of isomers of the complex generated by the chirality on the nitrogen atoms located on an ancillary ligand. The substrate used for the alcoholysis reaction, alkylzinc dimeric (L2ZnEt)2, which is well defined in the solid state, undergoes a transformation in solution into a mixture of isomers, two of them visible in NMR spectra in a ratio of ca. 3:1. Five of them have been generated by the DFT method through the transformation of zinc and nitrogen atoms.5p,r Here again, despite the detailed analysis, both 2D NMR and NOESY spectra are unable to discern whether intramolecular rearrangement between a mixture of dimers or monomer/dimer equilibrium (or both) is involved. Furthermore, the DOSY data showed species with different nuclearity present in solution (Figure 6). In order to understand the transformation of this complex in solution, we calculated theoretical diffusion coefficients for monomeric and di/trimeric forms generated in a DFT study according to the procedure described in our previous paper.5p The higher Df values calculated for monomers for both van der Waals and Connolly surfaces indicate that the monomer L2ZnOBn is probably absent in solution. Comparable Df values for all possible dimers generated in the DFT study and estimated via NMR suggest that the solid state dimeric structure is retained in solution (Supporting Information, Table S22). The diffusion coefficients obtained from DOSY experiments are comparable to published values for similar dimeric complexes.5d,p The lowest and highest values for the experimentally obtained diffusion coefficient (7.30 ×10−10, 5.27 × 10−10) are close to the Df SES calculated for linear trimeric species and homoleptic (L2)2Zn compounds, respectively. Although the linear trimeric species [(L2)2Zn3(OBn)4] and homoleptic (L2)2Zn compounds are probably present in solution, the fraction of anti/syn-(L2ZnOBn)2 dimers is dominant. Despite the accuracy in the structural motif of (L2ZnOBn)2 dimers according to single-site initiators, the linear trimeric species detected in solution is also potentially active in ROP entities. The homoleptic one could be active as a couple with ROH. Therefore, under these condition, it should be a dormant species. The complex (L2ZnOPe)2 possesses additional chiral carbon atoms complicating the characterization of the behavior in solution (Supporting Information, Figures S5−9), but the conclusion is compatible: the solution contains a mixture of dimers with a nonconsiderable addition of trimeric and homoleptic compounds. The 1H NMR spectra of homoleptic (L2)2Zn and (L1)2Zn compounds are clear and anticipated. There is only one set of well-resolved signals that is consistent with the structure in the solid state (Supporting Information, Figures S10−13 and S15−

Figure 2. Molecular structure of (L2ZnOPe)2. Thermal ellipsoids are drawn at a 30% probability level. H atoms are excluded for the sake of clarity.

Figure 3. Molecular structure of (L2)2Zn. Thermal ellipsoids are drawn at a 30% probability level. H atoms are excluded for the sake of clarity.

Figure 4. Molecular structure of (L1)2Zn. Thermal ellipsoids are drawn at 30% probability level. H atoms are excluded for the sake of clarity.

undefined because it is unclear whether bridging or terminal modes seem to be a better option in improving the catalytic behavior of metal alkoxide initiators. Unfortunately, there are only limited reports of (LMOR)2 complexes determined by Xray and DFT studies. Thus, the next stage in this exploratory program was to obtain single-site initiators that are potentially at a higher level than the corresponding active complexes generated in a direct E

DOI: 10.1021/acs.inorgchem.6b02439 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry Table 1. Selected Bond Distances (Ångstroms) and Angles (degrees)a atoms bond distance Zn1−O1 Zn1−O2 Zn1−O2i Zn1−N1 Zn1−N2 Zn1−Zn1i,a O2−O2i,a angles O1−Zn1−O2 O1−Zn1−O2i O2−Zn1−O2i O1−Zn1−N1 O1−Zn1−N2 O2−Zn1−N1 O2−Zn1−N2 O2i−Zn1−N1 Zn1−O2−Zn1i

(L2ZnOBn)2

(L2ZnOPe)2

1.9195(19) 1.9707(18) 1.9643(18) 2.066(2)

1.9249(17) 1.9662(16) 1.9699(16) 2.079(2)

2.9164(7) 2.642(3)

2.9067(8) 2.654(3)

113.93(7) 114.15(8) 84.34(7) 101.59(8)

117.79(7) 110.72(7) 84.80(7) 100.87(7)

121.39(8)

121.61(7)

121.91(8) 95.66(7)

121.63(7) 95.20(7)

(L2)2Zn

(L1)2Zn

1.918(3) 1.932(3)

1.9265(13) 1.9184(13)

2.126(4) 2.097(4)

2.1006(15) 2.0909(16)

107.08(14)

112.05(6)

99.59(15) 114.39(15) 115.42(15) 100.71(14)

98.73(6) 111.32(6) 112.37(6) 99.86(6)

i = −x + 1, −y + 1, −z + 1 for (L2ZnOBn)2. i = −x + 1/2, −y + 1/2, −z + 1 for (L2ZnOPe)2. Values in parentheses refer to the nonbonding interactions.

a

Figure 5. Fragment of the 1H NMR spectrum for (L2ZnOBn)2 in benzene-d6 (*).

18). The diagnostic methylene protons of the Ph−CH2−N linker are present at room temperature in broad singlets at 4.51 and 3.19 ppm. The variable-temperature 1H NMR spectrum indicated the presence of three species in solution identified by three sets of coupled doublets similar to zinc compounds that were obtained earlier (Supporting Information, Figure S14).5c The DOSY experiment measuring for isolated (L2)2Zn only indicated the homoleptic monomeric species, but the most exciting DOSY spectrum was obtained with a postmethanolysis reaction mixture, before L2Zn precipitation (Figure 7). In this case, two possible entities with different nuclearity coexist in solution. The theoretical diffusion coefficients for possible structures including monomer, dimers, trimers, and tetramers, both in linear and in cyclic versions, were calculated using DFT (Figure 8). A good agreement between all theoretical and experimental Df SES values was found for (L2)2Zn and the linear tetramer [(L2)2Zn4(μ-OMe)6] (Figure 7). These combined experimental and computational data suggest that the possible synthetic

pathway to a methanolysis reaction leads to the main product: the homoleptic monomer L2Zn, which is easy to crystallize. The second product, which is difficult to isolate, is a linear oligomer containing methoxy-bridged zinc atoms flanked by terminal ancillary ligands. Usually the appearance of the bis-chelatation product L2Zn during the alcoholysis reaction of alkyl−metal complexes is bad news because this process is considered to represent a deactivation pathway of an in-situ-formed LZnOR initiator. Zinc alkoxides Zn(OR)2 formed as coproducts of the ligand redistribution reaction sometimes readily react with still unreacted LZnOR monomers, causing linear hetero-oligomeric species [L2Zn3(OR)4] to appear (Scheme 3, Path A, described in earlier literature4j). The assumption of these mechanistic scenarios is that the monomer is operative and the dimer is unable to engage in such a transformation. On the basis of experimental observation indicating the dominance of dimeric species in solution, we investigated an alternative reaction pathway (Scheme 3). To distinguish F

DOI: 10.1021/acs.inorgchem.6b02439 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

Figure 6. 1H DOSY spectrum of (L2ZnOBn)2 in benzene-d6. Geometrically optimized structures and (Df homoleptic (L2)2Zn, dimers anti-(L2ZnOBn)2 and syn-(L2ZnOBn)2, and linear trimer [(L2)2Zn3(OBn)4].

SES

) calculated for monomer L2ZnOBn,

Figure 7. 1H DOSY spectrum for (A) in situ methanolysis reaction of (L2ZnEt)2 and (B) (L2)2Zn complex.

between the two reaction pathways (A and B), we evaluated the energies of appropriate structural motifs via DFT (Figure 9). The energies for monomers L2ZnOR are over 209.30 and 204.39 kJ/mol larger than for dimeric compounds (LZnOR)2

for methoxy and benzyloxy species, respectively. The large energy differences between dimers and monomers additionally confirm that monomers are absent in solution. On the other hand, the difference between the values of the relative energies G

DOI: 10.1021/acs.inorgchem.6b02439 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

methoxy species can also be explained as a reaction of the dimer with a previously generated trimer (Scheme 3, Path C). The calculations for corresponding structural motifs containing fixed alkoxy ligands OR = OBn and a different ancillary ligand framework (L) were also performed (Figure 10). The data obtained again suggest that a dimeric species with a potential admixture of the homoleptic monomer L2Zn and heteroleptic trimer L2Zn3(OR)4 is mostly preferred. A similar trend has been observed for the potential monomer (LZnOBn; L = L1,2,3). The relative energy value between homoleptic compounds (L2Zn + Zn(OBn)2, L = L1,2,3) is again greater than for the favorable mixture (L2Zn + L2Zn3(OBn)4). In all of the considered zinc complexes, theoretical studies confirm the experimental data, indicating that zinc compounds of the single-site type (LZnOR)2 are stable dimers in the solid state but after dissolution yield a mixture of dimers (rather incapable of monomerization) with small additions of homoleptic monomers L 2 Zn and heteroleptic trimers L2Zn3(OR)4. The identification of the homoleptic L2Zn product during/after the alcoholysis reaction of alkyl−zinc compounds is not an indicator of a simple ligand redistribution reaction but a more advanced transformation-formed mixture of heterooligomeric zinc complexes. Therefore, the yield and quality of the desired main product LMOR/(LMOR)2 for kinetically labile metals relates to the molecular fitting of ligands (L and OR), compound solubility, and dynamic behavior in the solution. That may be the overall reason for the clean synthesis of the (L2ZnOBn)2 and (L2ZnOPe)2 complexes and difficulties when methanol was used for alcoholysis. Even high-quality crystals of an ideal structural motif yield a small fraction of unexpected products after dissolution.

Figure 8. DFT-optimized structures of methoxy zinc species.

between dimers and the mixture of both homoleptic (L2)2Zn and heteroleptic trimers is small enough to suggest another more energy favorable pathway (Scheme 3, Path B). The formation of a linear tetramer experimentally verified for

Scheme 3. (A) Hypothetical Deactivation Pathway of Single-Site Initiators; (B and C) Suggested Scheme of L2Zn Formation during an Alcoholysis Reaction

H

DOI: 10.1021/acs.inorgchem.6b02439 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

Figure 9. Relative energies for the DFT-optimized zinc structures with ligand L2 and various alkoxy groups (R = Me, Bn, Pe).

Figure 10. Relative energies for the DFT-optimized zinc structures with various ligands (L = L1, L2, L3) and benzyloxy groups.

Lactide Polymerization. Our aim is a more in-depth mechanistic understanding of the fitting process between the metal center in an active complex and lactide monomers as well as the way in which the structural flexibility of metal complex influences its catalytic activity. The focus of our study was the verification of the reactivity of single-site initiators (LZnOR)2; OR = OBn, OPe versus similar ones obtained during an in situ alcoholysis reaction of metalorganic precursors (L2ZnEt)2. We investigated the catalytic activity of structurally analogous dimers under comparable polymerization reaction conditions (Table 2). The growth of polylactide and the loss of monomers were monitored throughout the ROP reactions using NMR spectroscopy. All polymerization reactions were carried out under mild conditions at room temperature and achieved high conversion (>96%) in 8−12 h for single-site initiators and in 40−60 min for L2ZnEt/ROH systems. The living ROP of cyclic esters for single-site initiators proceeds via a coordination/insertion mechanism and involves the coordination of this ester to the metal center through the

exocyclic carbonyl oxygen. The formation of this adduct leads to a nucleophilic attack of the active alkoxy ligand onto the carbonyl carbon of the cyclic ester. This reaction is followed by the spontaneous ring opening, i.e., cleavage of the acyl−oxygen bond. The polymer subsequently formed is of the type HO− [PLA]n−OR, presented here by initiators n = 100, OR = OBn or OPe, respectively. The longer time for ROP reactions achieved by (LZnOBn)2 and (LZnOPe)2 could be accepted, but the PDI values (1.60−1.64) are unfortunately not appropriate for the precise growth of the polymer chain guaranteed by single-site initiators (Table 2, entries 1 and 2). The polymerization results obtained for (LZnEt)2/ROH systems indicated that real initiators formed during an alcoholysis reaction are not a classic single site based on an LZnOR motif. Furthermore, the sequence of addition of the reactants is crucial. The best PDI value (∼1.2) was observed for the ROP reaction when transformation of alkyl−zinc precursor was carried out in the presence of lactide monomers (Table 2, entries 3−5). Delaying lactide addition to the (L2ZnEt)2/ROH I

DOI: 10.1021/acs.inorgchem.6b02439 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry Table 2. ROP of LA Initiated by Zinc Complexes [I]a entry/[I]

ROH

[I]/[L-LA]/ROH

time

C [%]b

103 Mn,calc

103 Mnd

Mw/Mne

MeOH BnOH BnOH PeOH MeOH BnOH PeOH MeOH BnOH PeOH MeOH BnOH PeOH

0.5/200/0 0.5/100/0 0.5/100/0 0.5/100/0 0.5/100/1 0.5/100/1 0.5/100/1 0.5/100/1 0.5/100/1 0.5/100/1 0.5/100/1 1/100/1 1/100/1 1/100/1 1/100/1 1/100/1 1/100/1

18 h 8h 10 h 12 h 60 min 60 min 60 min 80 min 60 min 60 min 60 min 40 min 60 min 60 min 60 min 60 min 80 min

96.6 99.6 99.2 98.2 96.0 96.0 99.0 99.5 99.9 99.8 99.8 100 99.8 99.5 97.8 98.8 96.2

27.95 14.46 14.40 14.15 13.87 13.94 14.38 14.46 14.43 14.49 14.51 14.44 14.49 14.46 14.13 14.35 13.99

33.02 15.12 16.21 14.03 14.22 15.05 15.89 14.92 13.98 14.63 16.11 14.56 14.93 13.88 15.18 14.80 14.44

1.66 1.60 1.60 1.64 1.28 1.22 1.20 1.17 1.64 1.56 1.59 1.14 1.16 1.17 1.20 1.21 1.18

2

1/(L Zn−OBn)2 2/(L2Zn−OBn)2 3/(L2Zn−OBn)2h 4/(L2Zn−OPe)2 5/(L2ZnEt)2f 6/(L2ZnEt)2f 7/(L2ZnEt)2h 8/(L2ZnEt)2 9/(L2ZnEt)2g 10/(L2ZnEt)2g 11/(L2ZnEt)2g 12/(L2)2Zn 13/(L2)2Zn 14/(L2)2Zn 15/(L1)2Zn 16/(L1)2Zn 17/(L1)2Zn

Reaction conditions: Vsolvent = 25 mL, CH2Cl2; T = 25 °C. bObtained from 1H NMR. cCalculated from Mn,cal = [L-LA]0/[ROH]0 × C × 144.13 + MROH unless otherwise specified. dDetermined by GPC calibrated versus polystyrene standards and corrected by a factor of 0.58 according to literature recommendations.7 eObtained from GPC. fPublished results.8 gSequences of substrate addition [(L2ZnEt)2 + ROH] and L-LA after 15 min. h rac-LA was used to give atactic PLAs (Pi ≈ 0.5 according to 1H {1H} NMR). a

Figure 11. Relative energies for the geometrically optimized methoxy zinc species.

system influences PLA quality (see Table 2, entries 6−8). The catalytic system based on ethyl−zinc complexes and alcohol combination gave a stock of initiators that were more or less effective in ROP reactions, but the lactide “knew” which one was best. In contrast to well-defined initiators (LZnOR), the homoleptic compounds (L2)2Zn and (L1)2Zn (in the presence of alcohols) show adequate control over ROP parameters: the obtained PLAs showed narrow PDIs (1.14−1.20) and molecular weight values close to those expected for 100-PLAOR polymers. To gain a more detailed picture of the real initiator structure, the initiation step of the ROP catalytic cycle was observed by employing theoretical methods. The described ROP of lactides

typically involved monomeric derivatives L−M−OR working as initiators.6 One detailed DFT computational analyses of the coordination−insertion mechanism for ROP of lactides was determined for magnesium and later for the zinc species with βdiketiminate ancillary ligands.6a Similarly, there are only a few examples of DFT calculations for dimeric complexes, but the assumption is that one metal center actively works during the ROP process.6b,h,j The complexes are sometimes simplified for DFT computations upon replacing the benzyloxy moiety (see experimental data) with a methoxy group. The experimental data revealed that the most active initiator is accessible during stoichiometric alcoholysis reaction. J

DOI: 10.1021/acs.inorgchem.6b02439 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

Figure 12. Structures of (LZnOMe)2_A−C-type species in the presence of two L-lactide molecules (schematic on left, DFT-optimized on right).

To track it, we first performed a gas-phase geometrical optimization of all possible structures that appeared during an in-situ alcoholysis reaction between heteroleptic ethylzinc dimers (L2ZnEt) and the tested alcohols: MeOH, BnOH, and PeOH. The energy profile of the different transition states and intermediates for methoxy−zinc species accessible during alcoholysis is shown in Figure 11, and key data are summarized in Table S12. For zinc entities with PeO and BnO, see Supporting Information Figures S32−35 and Tables S13 and S14. The ideal example of initiators described earlier is our theoretically generated monomeric L2ZnOMe complex, which easily coordinates the lactide monomer (Supporting Information). However, the high energetic cost of monomer formation indicates that the ROP reaction in our examples cannot be mediated by this species. The alcoholysis reaction, which is carried out on the dimeric complex (L2ZnEt)2 and followed by the formation of potential intermediate species, is presented in Figure 11. Likewise, the ROP of lactide could be initiated by

these dimeric species containing all terminal/bridging (A/C) or both terminal and bridging (B) alkoxy ligands. For the sake of clarity, the discussion of the coordination of lactide to the modeled zinc complexes is delineated separately for the most stable isomers anti-A−C(R,S). Figure 12 presents adducts with lactides marked A−C(R,S)_2L-LA. For details of the isomers (S,S)/(R,R), see Supporting Information Figures S37−39. The distances between carbonyl O atoms of lactide molecules and zinc centers are in the range 2.10−2.27 Å for optimized structures of all (R,S) isomers (Figure 12). However, the bond lengths between alkoxide O atoms and the C atom of the lactide is shorter (2.60−2.62 Å) for the lactide molecules coordinated to zinc centers surrounded by terminal alkoxy ligands (type A) than the corresponding distances for zinc atoms in the neighborhood of bridged alkoxy ligands (3.97− 4.09 Å). It can be considered as the indicator of the easy insertion reaction accompanied by an alkoxy ligand terminal bonded to a zinc center. K

DOI: 10.1021/acs.inorgchem.6b02439 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

Figure 13. Energy profile for the first cycle in the ROP of lactide initiated by zinc compounds (energies in kJ/mol and optimized structure of intermediates (I−V) for dimers A−C).

Scheme 4. Schematic Representation for the First Cycle in ROP of Lactide Initiated by Zinc Dimers with One Working Metal Center

LZnOR monomer is similar to that reported for other zinc initiators, involving the dissociation of the dinuclear compound to yield active monomeric species.6 The parent complexes of the presented family of initiators are dimeric (LZnEt)2. These complexes and (LZnOR)2 were shown to be dinuclear both in the solid state and in solution. The dynamic behavior of these complexes in solution enabled the creation of a dangling amine arm of ancillary ligands. This process is crucial during coordination of the lactide molecule. The predisposition of easy coordination or decoordination of the hemilabile arm of stabilizing ligands is emphasized more by other research groups as essential for an ROP reaction.5d,g The assumption of dangling aminonaphtholate ligands gives effective coordination

The energy profile of the ROP initiation reaction using all relevant intermediates (L2ZnOMe)2 of types A−C(R,S) is shown in Figure 13. For separate diagrams for all discussed dimers see Supporting Information Figures S42−46. The initiation steps for all dimers (types A−C(R,R/S,S/R,S)) and the monomer L2ZnOR were investigated for this study. However, the simultaneous tracking of the differences between them is crucial for the explanation of the polymerization results. Although the experimental data indicates that the monomeric compounds LZnOR are absent in solution, DFT calculations were performed to explore these initiators (Supporting Information, Figure S41 and Table S15). The possible mechanism for initiation of ROP of lactides by our putative L

DOI: 10.1021/acs.inorgchem.6b02439 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

Figure 14. Energy profile for the first cycle in ROP of lactide initiated by zinc compounds (energies in kJ/mol and optimized structure of intermediates (I−V) for dimers A: A′ gray, zinc centers working simultaneously; A″ black, zinc centers working alternately.

Scheme 5. Schematic Representation for the First Cycle in ROP of Lactide Initiated by Zinc Dimers with Two Simultaneously (A′) or Alternately (A′′) Working Metal Centers

tetrahedral coordination with a flexible amine arm of ancillary ligands coordinated to zinc atoms by both nitrogen and oxygen donor atoms. The creation of the binding position on the metal center for lactide coordination required decoordination of the nitrogen atom of ancillary ligand. This modification of the coordination sphere around the zinc center leaves metals coordinated by four O atoms of lactide, both alkoxy and dangling aminonaphtholate ligands. The schematic representation and optimized structures of formed intermediates I−V(A− C) are shown in Scheme 4. The first step with the lowest energy is for C (i.e., from I−II coordination lactide to the metal center), which leads to an overall energy slightly below 13.50 kJ/mol. However, the energy difference between I−II intermediates is 47.71 and 23.08 kJ/mol for B and A, respectively. The dimeric structure of the A−C species is maintained by the bridging O atoms. In the next step, the alkoxide O atom attacks the carbonyl carbon atom of the coordinated lactide molecule with the lowest energy of only 6.61 kJ/mol for C. The steps with the highest barrier are the ones from III−IV involving opening of the lactide ring. The final decoordination of the opened lactide

of lactide to the zinc centers. However, simultaneous coordination/decoordination of nitrogen atoms followed by the next stages of ROP (i.e., coordination of lactide, insertion, ring opening, decoordination) is impossible for optimization of the transition stages between them. We first examined dinuclear pathways for polymerization of L-lactide by dimers of type A−C with one working metal center. The reaction sequence between appropriate species and the lactide molecule and energy profiles are separately shown in Figures S42−44. The summary energy profile of the insertion step for the most stable species A−C(R,S) is presented in Figure 13 with C species considered as the reference for the energy profile. The next stages of the first ROP insertion are marked I−V (I, active species; II, coordination; III, insertion; IV, ring opening; V, decoordination), and the structures of complex intermediates are I(A−C)−V(A−C), respectively. An in-depth analysis of Figure 13 shows that the different microsteps of the first ROP insertion are similar for all species A−C. As shown in Figure 12, each zinc atom in all of the types of potentially active initiators I(A−C) adopts a slightly distorted M

DOI: 10.1021/acs.inorgchem.6b02439 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

single-site initiators (L 2ZnOR)2 operating through the coordination−insertion mechanism. The alkylzinc precatalyst in the presence of an alcohol coinitiator is operative in the ROP reaction when the zinc complex, alcohol, and lactide monomer are mixed together in the same moment. To catalyze the ROP process, LZnEt underwent the alcoholysis reaction, leading to the replacement of the terminal ethyl group by the alkoxy ligand. The new zinc alkoxides bridged by oxygen atoms of pincer aminonaphthol ligands constitute a key intermediate species that is the most favored during the coordination of the lactide molecule. The initiation step of ROP of lactides was carried out by alternately working zinc centers through the coordination of the monomer and the insertion of terminal alkoxy ligands, followed by the ring opening of the lactide, and finally decoordination of the lactyl−lactate arm (Scheme 6).

leading to terminal linear lactyl−lactate ester bonded to zinc atoms yielded stable species (A−C)-V (Scheme 4). This first insertion step of the ROP cycle is relatively similar to that reported using a dimeric intermediate with two centers binding one lactide molecule. Moreover, the experimental results suggest simultaneous/alternate reactivity of the two zinc centers. Therefore, we next examined dinuclear pathways with two working metal centers for dimers appearing as the first potentially active species during alcoholysis. The energy profile for simultaneously and alternately working metal centers is presented in Figure 14, while the corresponding schematic structures of intermediates are shown in Scheme 5. Interestingly, the height of the energy for alternate coordination of lactide A″-II is inverted, and the next microsteps are more facile than for a simultaneously or singly working metal center (A′). Subsequently, from A-II dimers, the insertion (step II−III) process takes place followed by the coordination of the next lactide; the energy of intermediate AIII is lower than intermediate A-II. However, the next stage of the ring opening and insertion (III−IV) of the second monomer for the alternately process is preferred, and in consequence, the intermediate A″-IV is much more stable than intermediate A′-IV, see Figure 14. These steps for similar initiators of types A−C showed the energy difference for the first initiation step to be smaller in the presence of dimeric species A, emphasizing cooperation between two zinc centers. In addition, the simultaneously working zinc centers A′ (gray lines, Figure 14) reached the highest energy for the ring-opened stage. The same behavior was observed for mixed dimer B and alkoxy-bridged dimer C (green lines, Figure 13). PLA chain growth is possible to start from all species A′ or A″(V) on the zinc centers with acyclic lactyl−lactate ester substituent, but initiators with alternately working centers are the best.

Scheme 6. Suggested First Cycle in the ROP of Lactide Initiated by a Zinc Complex



CONCLUSIONS A family of homo- and heteroleptic zinc complexes with aminonaphtholate ligands was synthesized and characterized. The alkyl−zinc (LZnEt)2 complexes undergo alcoholysis reaction with benzyl alcohol or 1-phenylethanol to yield the planned (LZnOR)2 complexes corresponding to a single-site initiator structure. The methanolysis of (LZnEt)2 dimers proceeds via a different reaction pathway, and homoleptic L2Zn is the only isolated product. In all of the considered zinc complexes, theoretical studies confirm the experimental data, indicating that zinc (LZnOR)2 compounds are dimers in the solid state but after dissolution the solution is a mixture of a dominant fraction of dimers with a small, nonconsiderable addition of trimers and homoleptic L2Zn. Our study additionally suggests that the appearance of the L2Zn product during alcoholysis is not a ligand redistribution reaction but a more advanced transformation that forms a mixture of homoleptic L2Zn and hetero-oligomeric zinc complexes. Therefore, the precise synthesis of desired LMOR/(LMOR)2 complexes with kinetically labile metals relates to the molecular fitting of both ancillary (L) and initiating (OR) ligands. The focus of our study was the verification of the reactivity of single-site-type (LZnOR)2 complexes versus similar ones obtained during in-situ alcoholysis reaction of (LZnEt)2. DFT methods were used to study the initiation step of ring-opening polymerization (ROP) of lactides mediated by aminonaphtholate zinc complexes (LZnEt)2 acting as precatalysts or

Earlier released results proved the flexibility of the aminaphtolate ligands for easily reversible nitrogen atom decoordination, which facilitates the fitting process between ligands and coordinating monomers. When lactide is present during the alcoholysis reaction, the next intermediate could be formed. It contains both terminal and bridged alkoxy ligands, but a single-site-type complex is the only isolable product after alcoholysis. Unexpectedly, the guidelines for rational structural motif design of new initiators do not have to be the most desired type of initiators. The new highlighted improvement of the old catalyst and the key to the design of the new one is still the undefined role of the alkoxy ligand in this “molecular recognition”. We used a soccer match as an analogy for our discoveries (Scheme 7). While a predetermined structure is generally N

DOI: 10.1021/acs.inorgchem.6b02439 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry Scheme 7. ROP Game



important for the synthesis of the initiator, in this “game” it takes adding the lactide to the reaction at the right moment to obtain the appropriate structure in a timely manner. The striker position is dimer A, the midfielder is B, and the defender is C, while the central playmaker is lactide. Through precise passes to the striker and through successive goals, lactide helps to builds PLA. In this analogy, the adduct of homoleptic L2Zn with alcohol is a player that can hold every position.



ACKNOWLEDGMENTS The authors express their gratitude to the National Science Centre in Poland (Grant NN 204 200640), KNOW (under the project for Wrocław Biotechnology Center), and the Wrocław Centre for Networking and Supercomputing (http://www. wcss.wroc.pl).



ASSOCIATED CONTENT

(1) For key references on application of PLA, see: (a) Langer, R.; Tirrell, D. A. Designing Materials for Biology and Medicine. Nature 2004, 428, 487−492. (b) Wang, Z.; Hu, H.; Wang, Y.; Wang, Y.; Wu, Q.; Liu, L.; Chen, G. Fabrication of Poly(3-Hydroxybutyrate-co-3Hydroxyhexanoate) (PHBHHx) Microstructures Using Soft Lithography for Scaffold Applications. Biomaterials 2006, 27, 2550−2557. (c) Murariu, M.; Doumbia, A.; Bonnaud, L.; Dechief, A. L.; Paint, Y.; Ferreira, M.; Campagne, C.; Devaux, E.; Dubois, P. High-Performance Polylactide/ZnO Nanocomposites Designed for Films and Fibers with Special End-Use Properties. Biomacromolecules 2011, 12, 1762−1771. (d) Kim, H.; Park, H.; Lee, J.; Kim, T. H.; Lee, E. S.; Oh, K. T.; Lee, K. C.; Youn, Y. S. Highly Porous Large Poly(Lactic-co-Glycolic Acid) Microspheres Adsorbed with Palmityl-Acylated Exendin-4 as a LongActing Inhalation System for Treating Diabetes. Biomaterials 2011, 32, 1685−1693. (e) Zhang, P.; Wu, H.; Wu, H.; Lù, Z.; Deng, C.; Hong, Z.; Jing, X.; Chen, X. RGD-Conjugated Copolymer Incorporated into Composite of Poly(Lactide-co-Glycotide) and Poly(L-Lactide)Grafted Nanohydroxyapatite for Bone Tissue Engineering. Biomacromolecules 2011, 12, 2667−2680. (f) Liao, L.; Dong, J.; Wang, G.; Fan, Z.; Li, S.; Lu, Z. Microstructure−Property Relationship of L-Lactide/ Trimethylene Carbonate/Glycolide Terpolymers as Cardiovascular Stent Material. Eur. Polym. J. 2015, 66, 429−436. (g) Hu, Y.; Darcos, V.; Monge, S.; Li, S. Thermo-Responsive Drug Release from SelfAssembled Micelles of Brush-Like PLA/PEG Analogues Block Copolymers. Int. J. Pharm. 2015, 491, 152−161.

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.inorgchem.6b02439. Spectroscopic and computational data: NMR spectra of (L2ZnOBn)2, (L2ZnOPe)2, (L2)2Zn, and (L1)2Zn; schematic and gas-phase-optimized structures of zinc isomers; energies of structures optimized in vacuo; selected bond distances and angles for DFT-optimized structures (PDF) X-ray crystallographic CIF files for (L2ZnOBn) 2, (L2ZnOPe)2, (L2)2Zn, and (L1)2Zn (CIF)



REFERENCES

AUTHOR INFORMATION

Corresponding Author

*E-mail: jolanta.ejfl[email protected]. ORCID

Jolanta Ejfler: 0000-0002-7467-1312 Notes

The authors declare no competing financial interest. O

DOI: 10.1021/acs.inorgchem.6b02439 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry (2) For leading and recent reviews, see: (a) O’Keefe, B. J.; Hillmyer, M. A.; Tolman, W. B. Polymerization of Lactide and Related Cyclic Esters by Discrete Metal Complexes. J. Chem. Soc., Dalton Trans. 2001, 2215−2224. (b) Dechy-Cabaret, O.; Martin-Vaca, B.; Bourissou, D. Controlled Ring-Opening Polymerization of Lactide and Glycolide. Chem. Rev. 2004, 104, 6147−6176. (c) Wu, J.; Yu, T.-L.; Chen, C. T.; Lin, C.-C. Recent Developments in Main Group Metal Complexes Catalyzed/Initiated Polymerization of Lactides and Related Cyclic Esters. Coord. Chem. Rev. 2006, 250, 602−626. (d) Platel, R. H.; Hodgson, L. M.; Williams, C. K. Biocompatible Initiators for Lactide Polymerization. Polym. Rev. 2008, 48, 11−63. (e) Wheaton, C. A.; Hayes, P. G.; Ireland, B. J. Complexes of Mg, Ca and Zn as Homogeneous Catalysts for Lactide Polymerization. Dalton Trans. 2009, 25, 4832−4846. (f) Ajellal, N.; Carpentier, J. F.; Guillaume, C.; Guillaume, S. M.; Helou, M.; Poirier, V.; Sarazin, Y.; Trifonov, A. Metal-Catalyzed Immortal Ring-Opening Polymerization of Lactones, Lactides and Cyclic Carbonates. Dalton Trans. 2010, 39, 8363−8376. (g) Stanford, M. J.; Dove, A. P. Stereocontrolled Ring-Opening Polymerisation of Lactide. Chem. Soc. Rev. 2010, 39, 486−494. (3) For key references on trispyrazolylborate, see: (a) Chisholm, M. H.; Eilerts, N. W. Single-Site Metal Alkoxide Catalysts for RingOpening Polymerizations. Poly(dilactide) Synthesis Employing {HB(3-Butpz)3}Mg(OEt). Chem. Commun. 1996, 7, 853−854. (b) Chisholm, M. H.; Eilerts, N. W.; Huffman, J. C.; Iyer, S. S.; Pacold, M.; Phomphrai, K. Molecular Design of Single-Site Metal Alkoxide Catalyst Precursors for Ring-Opening Polymerization Reactions Leading to Polyoxygenates. 1. Polylactide Formation by Achiral and Chiral Magnesium and Zinc Alkoxides, (η3-L)MOR, Where L = Trispyrazolyl- and Trisindazolylborate Ligands. J. Am. Chem. Soc. 2000, 122, 11845−11854. (c) Chisholm, M. H.; Gallucci, J.; Phomphrai, K. Well-Defined Calcium Initiators for Lactide Polymerization. Inorg. Chem. 2004, 43, 6717−6725. (4) For key references on β-diketiminate, see: (a) Cheng, M.; Attygalle, A. B.; Lobkovsky, E. B.; Coates, G. W. Single-Site Catalysts for Ring-Opening Polymerization: Synthesis of Heterotactic Poly(lactic acid) from rac-Lactide. J. Am. Chem. Soc. 1999, 121, 11583− 11584. (b) Chamberlain, B. M.; Cheng, M.; Moore, D. R.; Ovitt, T. M.; Lobkovsky, E. B.; Coates, G. W. Polymerization of Lactide with Zinc and Magnesium β-Diiminate Complexes: Stereocontrol and Mechanism. J. Am. Chem. Soc. 2001, 123, 3229−3238. (c) Chisholm, M. H.; Gallucci, J.; Phomphrai, K. Coordination Chemistry and Reactivity of Monomeric Alkoxides and Amides of Magnesium and Zinc Supported by the Diiminato Ligand CH(CMeNC6H3-2,6-iPr2)2. A Comparative Study. Inorg. Chem. 2002, 41, 2785−2794. (d) Allen, D. S.; Moore, D. R.; Lobkovsky, E. B.; Coates, G. W. Structure and Reactivity of Mono- and Dinuclear Diiminate Zinc Alkyl Complexes. J. Organomet. Chem. 2003, 683, 137−148. (e) Chisholm, M. H.; Phomphrai, K. Conformational Effects in β-diiminate Ligated Magnesium and Zinc Amides. Solution Dynamics and Lactide Polymerization. Inorg. Chim. Acta 2003, 350, 121−125. (f) Dove, A. P.; Gibson, V. C.; Marshall, E. L.; White, A. J. P.; Williams, D. J. Magnesium and Zinc Complexes of a Potentially Tridentate βDiketiminate Ligand. Dalton Trans. 2004, 4, 570−578. (g) Drouin, F.; Oguadinma, P. O.; Whitehorne, T. J. J.; Prud’homme, R. E.; Schaper, F. Lactide Polymerization with Chiral β-Diketiminate Zinc Complexes. Organometallics 2010, 29, 2139−2147. (h) Hsueh, L.-F.; Chuang, N.T.; Lee, C.-Y.; Datta, A.; Huang, J.-H.; Lee, T.-Y. Magnesium Complexes Containing η1- and η3-Pyrrolyl or Ketiminato Ligands: Synthesis, Structural Investigation and ϵ-Caprolactone Ring-Opening Polymerisation. Eur. J. Inorg. Chem. 2011, 2011, 5530−5537. (i) Chen, H.-Y.; Peng, Y.-L.; Huang, T.-H.; Sutar, A. K.; Miller, S. A.; Lin, C.-C. Comparative Study of Lactide Polymerization by Zinc Alkoxide Complexes with a β-Diketiminato Ligand Bearing Different Substituents. J. Mol. Catal. A: Chem. 2011, 339, 61−71. (j) Balasanthiran, V.; Chisholm, M. H.; Choojun, K.; Durr, C. B. Ethyl 2-Hydroxy-2Methylpropanoate Derivatives of Magnesium and Zinc. The Effect of Chelation on the Homo- and Copolymerization of Lactide and εCaprolactone. Dalton Trans. 2014, 43, 2781−2788.

(5) For key references on amino/imino−phenolate ligands, see: (a) Ejfler, J.; Kobyłka, M.; Jerzykiewicz, L. B.; Sobota, P. Highly Efficient Magnesium Initiators for Lactide Polymerization. Dalton Trans. 2005, 2047−2050. (b) Ejfler, J.; Krauzy-Dziedzic, K.; Szafert, S.; Jerzykiewicz, L. B.; Sobota, P. Synthesis, Characterization, and Catalytic Studies of (Aryloxido)magnesium Complexes. Eur. J. Inorg. Chem. 2010, 2010, 3602−3609. (c) Ejfler, J.; Szafert, S.; Mierzwicki, K.; Jerzykiewicz, L. B.; Sobota, P. Homo- and Heteroleptic Zinc Aminophenolates as Initiators for Lactide Polymerization. Dalton Trans. 2008, 46, 6556−6562. (d) Williams, C. K.; Breyfogle, L. E.; Choi, S. K.; Nam, W.; Young, V. G., Jr.; Hillmyer, M. A.; Tolman, W. B. A Highly Active Zinc Catalyst for the Controlled Polymerization of Lactide. J. Am. Chem. Soc. 2003, 125, 11350−11359. (e) Chen, H. Y.; Tang, H. Y.; Lin, C. C. Ring-Opening Polymerization of Lactides Initiated by Zinc Alkoxides Derived from NNO-Tridentate Ligands. Macromolecules 2006, 39, 3745−3752. (f) Zheng, Z.; Zhao, G.; Fablet, R.; Bouyahyi, M.; Thomas, C. M.; Roisnel, T.; Casagrande, O., Jr.; Carpentier, J.-F.; Zinc. and Enolato-Magnesium Complexes Based on Bi-, Tri- and Tetradentate Aminophenolate Ligands. New J. Chem. 2008, 32, 2279−2291. (g) Labourdette, G.; Lee, D. J.; Patrick, B. O.; Ezhova, M. B.; Mehrkhodavandi, P. Unusually Stable Chiral Ethyl Zinc Complexes: Reactivity and Polymerization of Lactide. Organometallics 2009, 28, 1309−1319. (h) Chuang, H.-J.; Weng, S.-F.; Chang, C.-C.; Lin, C.-C.; Chen, H.-Y. Synthesis, Characterization and Catalytic Activity of Magnesium and Zinc Aminophenoxide Complexes: Catalysts for Ring-Opening Polymerization of L-Lactide. Dalton Trans. 2011, 40, 9601−9607. (i) Poirier, V.; Roisnel, T.; Carpentier, J.-F.; Sarazin, Y. Zinc and Magnesium Complexes Supported by Bulky Multidentate Amino-Ether Phenolate Ligands: Potent Pre-Catalysts for the Immortal Ring-Opening Polymerisation of Cyclic Esters. Dalton Trans. 2011, 40, 523−534. (j) Ikpo, N.; Saunders, L. N.; Walsh, J. L.; Smith, J. M. B.; Dawe, L. N.; Kerton, F. M. Zinc Complexes of Piperazinyl-Derived Aminephenolate Ligands: Synthesis, Characterization and Ring−Opening Polymerization Activity. Eur. J. Inorg. Chem. 2011, 2011, 5347−5359. (k) Song, S.; Zhang, X.; Ma, H.; Yang, Y. Zinc Complexes Supported by Claw-Type Aminophenolate Ligands: Synthesis, Characterization and Catalysis in the RingOpening Polymerization of Rac-Lactide. Dalton Trans. 2012, 41, 3266−3277. (l) Wang, H.; Ma, H. Highly Diastereoselective Synthesis of Chiral Aminophenolate Zinc Complexes and Isoselective Polymerization of rac-Lactide. Chem. Commun. 2013, 49, 8686−8688. (m) Wang, H.; Yang, Y.; Ma, H. Stereoselectivity Switch between Zinc and Magnesium Initiators in the Polymerization of rac-Lactide: Different Coordination Chemistry, Different Stereocontrol Mechanisms. Macromolecules 2014, 47, 7750−7764. (n) Farwell, J. D.; Hitchcock, P. B.; Lappert, M. F.; Luinstra, G. A.; Protchenko, A. V.; Wei, X.-H. Synthesis and Structures of Some Sterically Hindered Zinc Complexes Containing 6-Membered ZnNCCCN and ZnOCCCN Rings. J. Organomet. Chem. 2008, 693, 1861−1869. (o) Wojtaszak, J.; Mierzwicki, K.; Szafert, S.; Gulia, N.; Ejfler, J. Homoleptic Aminophenolates of Zn, Mg and Ca. Synthesis, Structure, DFT Studies and Polymerization Activity in ROP of Lactides. Dalton Trans. 2014, 43, 2424−2436. (p) Jędrzkiewicz, D.; Ejfler, J.; Gulia, N.; John, Ł.; Szafert, S. Designing Ancillary Ligands for Heteroleptic/Homoleptic Zinc Complex Formation: synthesis, structures and Application in ROP of Lactides. Dalton Trans. 2015, 44, 13700−13715. (r) Jędrzkiewicz, D.; Ejfler, J.; John, Ł.; Szafert, S. Zebra Reaction or the Recipe for the Synthesis of Heterodimeric Zinc Complexes. Dalton Trans. 2016, 45, 2829−2838. (6) For key references on theoretical calculation of ROP initiated by metal compounds, see: (a) Marshall, E. L.; Gibson, V. C.; Rzepa, H. S. A Computational Analysis of the Ring-Opening Polymerization of racLactide Initiated by Single-Site β-Diketiminate Metal Complexes: Defining the Mechanistic Pathway and the Origin of Stereocontrol. J. Am. Chem. Soc. 2005, 127, 6048−6051. (b) Yu, T.-L.; Wu, C.-C.; Chen, C.-C.; Huang, B.-H.; Wu, J.; Lin, C.-C. Catalysts for the RingOpening Polymerization of ε-Caprolactone and L-Lactide and the Mechanistic Study. Polymer 2005, 46, 5909−5917. (c) Dove, A. P.; Gibson, V. C.; Marshall, E. L.; Rzepa, H. S.; White, A. J. P.; Williams, P

DOI: 10.1021/acs.inorgchem.6b02439 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry D. J. Synthetic, Structural, Mechanistic, and Computational Studies on Single-Site β-Diketiminate Tin(II) Initiators for the Polymerization of rac-Lactide. J. Am. Chem. Soc. 2006, 128, 9834−9843. (d) Liu, J.; Ling, J.; Li, X.; Shen, Z. Monomer Insertion Mechanism of Ring-Opening Polymerization of ε-Caprolactone with Yttrium Alkoxide Intermediate: A DFT Study. J. Mol. Catal. A: Chem. 2009, 300, 59−64. (e) Del Rosal, I.; Brignou, P.; Guillaume, S. M.; Carpentier, J.-F.; Maron, L. DFT Investigations on the Ring-Opening Polymerization of Cyclic Carbonates Catalyzed by Zinc-{β-Diiminate} Complexes. Polym. Chem. 2011, 2, 2564−2573. (f) Borner, J.; Dos Santos Vieira, I.; Pawlis, A.; Doring, A.; Kuckling, D.; Herres-Pawlis, S. Mechanism of the Living Lactide Polymerization Mediated by Robust Zinc Guanidine Complexes. Chem. - Eur. J. 2011, 17, 4507−4512. (g) Chen, H.-Y.; Mialon, L.; Abboud, K. A.; Miller, S. A. Comparative Study of Lactide Polymerization with Lithium, Sodium, Magnesium, and Calcium Complexes of BHT. Organometallics 2012, 31, 5252− 5261. (h) Fang, J.; Yu, I.; Mehrkhodavandi, P.; Maron, L. Theoretical Investigation of Lactide Ring-Opening Polymerization Induced by a Dinuclear Indium Catalyst. Organometallics 2013, 32, 6950−6956. (i) Lawan, N.; Muangpil, S.; Kungwan, N.; Meepowpan, P.; Lee, V. S.; Punyodom, W. Tin (IV) Alkoxide Initiator Design for Poly(DLactide) Synthesis Using DFT Calculations, Computational and Theoretical Chemistry. Comput. Theor. Chem. 2013, 1020, 121−126. (j) Fliedel, C.; Vila-Vicosa, D.; Calhorda, M. J.; Dagorne, S.; Aviles, T. Dinuclear Zinc−N-Heterocyclic Carbene Complexes for Either the Controlled Ring-Opening Polymerization of Lactide or the Controlled Degradation of Polylactide Under Mild Conditions. ChemCatChem 2014, 6, 1357−1367. (k) Tabthong, S.; Nanok, T.; Sumrit, P.; Kongsaeree, P.; Prabpai, S.; Chuawong, P.; Hormnirun, P. Bis(pyrrolidene) Schiff Base Aluminum Complexes as Isoselective-Biased Initiators for the Controlled Ring-Opening Polymerization of racLactide: Experimental and Theoretical Studies. Macromolecules 2015, 48, 6846−6861. (l) Sattayanon, C.; Sontising, W.; Limwanich, W.; Meepowpan, P.; Punyodom, W.; Kungwan, N. Effects of Alkoxide Alteration on the Ring-Opening Polymerization of ε-Caprolactone Initiated by n-Bu3SnOR: a DFT Study. Struct. Chem. 2015, 26, 695− 703. (m) Del Rosal, I.; Brignou, P.; Guillaume, S. M.; Carpentier, J.-F.; Maron, L. DFT Investigations on the Ring-Opening Polymerization of Substituted Cyclic Carbonates Catalyzed by Zinc-{β-Diketiminate} Complexes. Polym. Chem. 2015, 6, 3336−3352. (n) Marlier, E. E.; Macaranas, J. A.; Marell, D. J.; Dunbar, C. R.; Johnson, M. A.; DePorre, Y.; Miranda, M. O.; Neisen, B. D.; Cramer, C. J.; Hillmyer, M. A.; Tolman, W. B. Mechanistic Studies of ε-Caprolactone Polymerization by (salen)AlOR Complexes and a Predictive Model for Cyclic Ester Polymerizations. ACS Catal. 2016, 6, 1215−1224. (7) Kowalski, A.; Libiszowski, J.; Duda, A.; Penczek, S. Polymerization of L,L-Dilactide Initiated by Tin(II) Butoxide. Macromolecules 2000, 33, 1964−1971. (8) Jędrzkiewicz, D.; Czeluśniak, I.; Wierzejewska, M.; Szafert, S.; Ejfler, J. Well-Controlled, Zinc-Catalyzed Synthesis of Low Molecular Weight Oligolactides by Ring Opening Reaction. J. Mol. Catal. A: Chem. 2015, 396, 155−163. (9) CrysAlisRED Software; Oxford Diffraction: Wrocław, Poland, 1995−2004. (10) Sheldrick, G. M. A Short History of SHELX. Acta Crystallogr., Sect. A: Found. Crystallogr. 2008, 64, 112−122. (11) Macrae, C. F.; Bruno, I. J.; Chisholm, J. A.; Edgington, P. R.; Mccabe, P.; Pidcock, E.; Rodriguez-Monge, L.; Taylor, R.; Van De Streek, J.; Wood, P. A. Mercury CSD 2.0 − New Features for the Visualization and Investigation of Crystal Structures. J. Appl. Crystallogr. 2008, 41, 466−470. (12) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A., Jr.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Keith, T.; Kobayashi, R.; Normand, J.;

Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega, N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A.; J. Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, O.; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09, Revision B.01;Gaussian, Inc.: Wallingford, CT, 2010. (13) Becke, A. D. Density-Functional Thermochemistry. III. The Role of Exact Exchange. J. Chem. Phys. 1993, 98, 5648−5652. (14) Lee, C.; Yang, W.; Parr, R. G. Development of the Colle-Salvetti Correlation-Energy Formula Into a Functional of the Electron Density. Phys. Rev. B: Condens. Matter Mater. Phys. 1988, 37, 785−789.

Q

DOI: 10.1021/acs.inorgchem.6b02439 Inorg. Chem. XXXX, XXX, XXX−XXX