Organocatalytic Ring-Opening Polymerization of Methyl-Substituted Glycolides

Abstract

Aliphatic polyesters derived from the ring-opening polymerization (ROP) of cyclic esters offers access to sustainable, depolymerizable, and renewable materials. Herein, we report the first organocatalytic ROP of dimethyl glycolide (DMG) and tetramethyl glycolide (TMG), synthesized from biobased α-hydroxy acids via an acylation/cyclization pathway. Using a metal-free catalytic system comprising phosphazene base (P₂-Et), thiourea (TU), and benzyl alcohol (BnOH) as the initiator, racemization-free poly­(dimethylglycolide) (PDMG) and poly­(tetramethylglycolide) (PTMG) were synthesized at room temperature. The resulting polyesters exhibited predictable molecular weights, low dispersity indices (Đ ≤ 1.25), and no transesterification, confirmed by ¹H NMR, SEC, and MALDI-TOF analyses. Mechanistic studies revealed distinct activation pathways: PDMG polymerization proceeds via TU imidate anion activation mechanism, while PTMG follows the conventional initiator/chain-end activation mechanism. Both polymerization processes demonstrated typical first-order kinetics. Computational modeling identified two key transition states (TSs) in the ROP mechanism: TS-1, involving the nucleophilic attack of BnOH on the carbonyl carbon of DMG or TMG, and TS-2, which involves the subsequent ring opening of the cyclic ester. Importantly, PDMG and PTMG can be quantitatively depolymerized into their respective monomers, enabling complete material recycling. This study establishes a sustainable approach for designing renewable polyesters with potential lifecycle management.

1. Introduction

The growing demand for sustainable and biocompatible materials has sparked significant interest in biobased polyesters, particularly those derived from renewable resources such as lactide and glycolide monomers. − Polymers obtained from these biobased monomers offer a reduced environmental footprint compared to conventional petroleum-based polymers. Moreover, they provide several advantages, including enhanced biodegradability, comparable mechanical and processing characteristics. − Aliphatic polyesters synthesized from these monomers, such as polylactic acid (PLA), polyglycolic acid (PGA), and their copolymer poly­(lactic-co-glycolic acid) (PLGA), combine excellent biodegradability and biocompatibility, making them highly appealing for diverse applications. These include packaging, the food industry, adhesives, and a range of biomedical uses such as medical sutures, drug delivery systems, and tissue engineering. ,− Among these materials, PLGA stands out as a vital component in the biomedical field, primarily due to its exceptional biodegradability and biocompatibility. − Its versatility and favorable properties have solidified its role in advancing medical and pharmaceutical technologies. Its safety profile and effectiveness have led to regulatory approvals from major agencies, including the U.S. Food and Drug Administration (FDA) and the European Medicines Agency (EMA). − However, despite their promising applications, PLA and PGA have certain limitations, such as a relatively low glass transition temperature (T _g) (35–60 °C), limited solubility (particularly for PGA), inherent brittleness, limited thermal resistance, and challenges in functionalization.

To expand the range of physicochemical properties and broaden the potential applications of PLA and PGA, various strategies have been explored, including blending and copolymerization. , Another promising approach involves substituting one, two, three, or four hydrogen atoms of glycolide with alkyl, aryl, or other functional groups. Structural modifications to the glycolide monomer, while preserving the degradable polyester backbone, provide a means to fine-tune polymer hydrophobicity, adjust the T _g, and introduce novel chemical functionalities into degradable polymers. ,, For example, the first reported substituted glycolide was described in 1971 by Tohara and co-workers, who synthesized poly­(diisobutyl glycolic acid) through the ring-opening polymerization (ROP) of diisobutyl glycolide in bulk at 180–190 °C for 30–150 h, using metal catalysts such as Na₂CO₃, MgO, CaO, PbO, Zn, and ZnO. However, the high polymerization temperature and extended reaction time likely caused racemization. Diisobutyl glycolide offers distinct advantages, including simple synthesis from leucine, the presence of a β-carbon atom, and a short side chain that facilitates targeted polymerization. Its morphology and hydrophobic properties make it particularly suitable for drug delivery, enabling controlled drug release from nanoparticles.

In their seminal work on the ROP of substituted glycolides, Baker and coworkers established a variety of symmetric and asymmetric substituted glycolide monomers, including ethyl, hexyl, isobutyl, isopropyl, cyclohexyl, diisobutyl, diethyl, dihexyl, and dicyclohexyl derivatives. − Symmetrically substituted glycolides are readily synthesized via direct cyclization of α-hydroxy acids in toluene or xylene using p-toluene sulfonic acid (pTSA) as a catalyst. In contrast, acylation/cyclization pathways yield both symmetric and asymmetric substituted glycolides. The latter pathway involves a two-step process: esterification of α-hydroxy acid with α-bromo acyl bromide to form a bromo-substituted carboxylic acid, followed by cyclization under basic conditions to produce the desired substituted glycolide. These substituted glycolides were polymerized via melt-polymerization using catalysts such as Sn­(Oct)₂, SnO, SnBr₂, SnBr₄, PbO, and Ph₄Sn ,, or through solution polymerization with Sn­(Oct)₂ catalysts. Thermal studies revealed that increasing the side chain length on the glycolide ring reduced the T _g, while branching within the side chain increased T _g. For instance, the Bayer group synthesized a high T _g (98 °C) polymer from dicyclohexylglycolide monomer. More recently, Mert and co-workers employed symmetric and asymmetric substituted glycolides for ROP using Sn­(Oct)₂, producing a biodegradable block copolymer with poly­(ethylene oxide) (PEO) suitable for drug delivery applications. ,, Despite these advances, most ROP reactions rely on metal-based catalysts. However, trace metal residues in substituted polyglycolides pose toxicity concerns for biomedical applications. ,

Substituted glycolides, such as dimethyl glycolide (DMG) and tetramethyl glycolide (TMG), remain relatively underexplored in the literature. To the best of our knowledge, no studies have reported the polymerization of DMG. However, Haruo Nishida and co-workers were the first to investigate the polymerization of TMG, employing a lithium-based catalyst under bulk conditions at 130 °C (Scheme A). , The resulting polytetramethyl glycolide (PTMG), exhibited a high T _g of 70 °C and a melting temperature (T _m) of 206 °C. Despite these promising thermal properties, PTMG did not achieve the desired molecular weight and displayed a broad molecular weight distribution. Notably, its recyclability was demonstrated via selective reduction, successfully regenerating the TMG monomer along with methacrylic acid.

1. (A) Previous Report on Synthesis of Tetramethyl Glycolide (TMG) and Its Organometallic Ring Opening Polymerization (ROP); (B) Synthesis of Methyl-Substituted Glycolide Monomers, Dimethyl Glycolide (DMG), and TMG, via an Acylation/Cyclization Pathway, Followed by Their Organocatalytic ROP Using BnOH as an Initiator and P₂-Et as a Catalyst.

In this work, we focus on modifying the substitution pattern at the α-carbon of glycolide to synthesize two unexplored depolymerizable polyesters, with their physical properties influenced by the number of methyl groups at the α-carbon. Specifically, the absence of a chiral center in DMG and TMG monomers resulted in racemization-free polyesters, which represent the simplest structural variations of methyl-substituted polyglycolides. Notably, the ROP of DMG and TMG was performed at room temperature using an organocatalyst. Organocatalysts have gained significant attention as eco-friendly alternatives to metal-based catalysts across diverse fields, including biomedicine, electronics, and food packaging. Their advantages include straightforward synthesis, ease of use, stability, and efficient removal from final products, an essential feature in applications where residual catalysts could pose risks or reduce material performance. ,−

2. Experimental Section

2.1. Materials and Characterization

2-Hydroxyisobutyric acid (Alfa Aesar, 99%), Bromoacetyl bromide (Acros, 98%), 2-Bromopropionyl bromide (Acros, 98%), 2-Bromo-2-methylpropionyl bromide (Acros, 98%), and N-Bromosuccinimide (NBS, Alfa Aesar, 99%) were used directly without further purification. Triethylamine (TEA, Aldrich, 99.5%) was distilled over CaH₂ before use. Solvents like Hexane (VWR 97%), ethyl acetate (VWR, 99.5%), and diethyl ether (VWR, 99%) were used as received, while; dichloromethane (DCM VWR), acetonitrile (Alfa Aesar) distilled over CaH₂ under reduced pressure. Tetrahydrofuran (THF, VWR), Benzene (VWR), and Toluene (VWR) were distilled over sodium metal under reduced pressure. Benzyl alcohol (BnOH, Aldrich, 99.8%) was distilled over CaH₂ before using as an initiator. 1-Cyclohexyl-3-(3,5-difluorophenyl)­thiourea (TU) was synthesized as per previous reports and purified before use, 1-Ethyl-2,2,4,4,4-pentakis­(dimethylamino)-2λ5,4λ5-catenadi­(phosphazene), (P₂-Et, Aldrich, 98%) was used as received without further purification. Deuterated chloroform (CDCl₃, Aldrich, 99.8%) was used as received. trans-2-[3-(4-tert-Butylphenyl)-2-methyl-2-propenylidene]­malononitrile (DCTB, Aldrich, 99%) and Sodium trifluoroacetate (Na-TFA, Aldrich, 98%) were used as received.

All ¹H NMR and ¹³C NMR spectra were conducted on either a 500 MHz Bruker AV NEO or 600 MHz Bruker AV NEO NMR spectrometers in CDCl₃. Size exclusion chromatography (SEC) was performed on a VISCOTEK VE2001 equipped with columns using two identical Agilent PLgel-Mixed C columns (5 μm) in connected series and refractive index as a detector array using THF as an eluent at a flow rate of 1.0 mL/min. A calibration curve to determine the molecular weight was obtained using a Polystyrene standard. MALDI-TOF MS experiments were carried out by using trans-2-[3-(4-tert-butylphenyl)-2-methyl-2-propenylidene]­malononitrile (DCTB) as the matrix in THF and Na-TFA as ionizing agent on a Bruker Ultrafex III MALDI-TOF mass spectrometer (Bruker Daltonik, Bremen, Germany). The thermal behavior of PDMG and PTMG was performed on differential scanning calorimetry (DSC) in the temperature range 0 to 220 °C at a heating rate of 1.0 °C/min in a nitrogen atmosphere. Thermogravimetric analysis (TGA) was performed under a nitrogen atmosphere at a heating rate of 10 °C/min, from 25 °C to 600 °C.

2.2. Computational Details

2.2.1. Ring Strain for Substituted Glycolides

All geometry optimizations were performed using B3LYP/6-31G+(d) level of theory with DFTD3­(BJ) dispersion correction − by the Gaussian 16 C.02 program. The harmonic vibrational frequency analyses were performed on all structures at the same level. Single-point energies were further refined using M062X/6-311+G­(d,p) level of theory with DFT-D3 dispersion correction. The reported energies were gained by adding the single-point energies and zero-point energy Correction at 298.15 K.

2.2.2. Computational Methods and Results for Mechanism Study

All geometry optimizations were performed using B3LYP/6-31G++(d,p) level of theory with DFTD3­(BJ) dispersion correction 1–4 by the Gaussian 16 C.02 program.5 The harmonic vibrational frequency analyses were performed on all structures at the same level. Single-point energies were further refined using M062X/6-311++G­(d,p) level of theory with DFT-D3 dispersion correction. The reported energies were gained by adding the single-point energies and zero-point energy Correction at 298.15 K.

2.3. Monomer Synthesis: (DMG and TMG)

2.3.1. Synthesis of 2-(2-Bromoacetoxy)-2-methylpropanoic Acid

N-bromosuccinimide (NBS) (3.45 g, 19.21 mmol) was dissolved in 500 mL of anhydrous dichloromethane (DCM) in a 1000 mL flame-dried round-bottom flask under an inert atmosphere at 0 °C. 2-bromoacetyl bromide (16.8 mL, 192.1 mmol) solution in Dry DCM (50 mL) was added dropwise to the continuously stirring reaction mixture at 0 °C, resulting in the formation of yellow precipitate. A solution of 2-hydroxy-2-methylpropanoic acid (20 g, 192.1 mmol) and triethylamine (TEA) (26.9 mL, 192.1 mmol) in dry DCM (50 mL) was added dropwise to the activated NBS-acetyl bromide reaction mixture. The reaction temperature was raised to 25 °C and stirred at the same temperature for 16 h. The solvent was evaporated at a reduced pressure and redissolved in diethyl ether (500 mL), followed by washing with water. The organic layer was concentrated to yield a yellow-brown crude product. The crude product was purified by column chromatography using a 30% hexane/ethyl acetate mixture. Yield 90%.

2.3.2. Synthesis of 3,3-Dimethyl-1,4-dioxane-2,5-dione (Dimethyl Glycolide)

Triethylamine (TEA) (28.6 mL, 202.6 mmol) was dissolved in 500 mL anhydrous acetonitrile in a flame-dried 1000 mL two-neck round-bottom flask in an argon atmosphere. The reaction temperature was raised to 60 °C, followed by the dropwise addition of 2-(2-bromoacetoxy)-2-methylpropanoic acid (38 g, 168.8 mmol) solution in 100 mL anhydrous acetonitrile. The reaction mixture was stirred at the same temperature for 3 h. The reaction was quenched by adding 16 mL of acetic acid to the cooled reaction mixture. The solvent was evaporated at reduced pressure to yield crude product. The crude product was purified by column chromatography using a 30% hexane/ethyl acetate mixture. Yield 80%

2.3.3. Synthesis of 2-((2-Bromo-2-methylpropanoyl)­oxy)-2-methylpropanoic Acid

2-((2-bromo-2-methylpropanoyl)­oxy)-2-methylpropanoic acid was synthesized by same reaction procedure used for the synthesis of 2-(2-bromoacetoxy)-2-methylpropanoic acid. N-Bromosuccinimide (NBS) (5.2 g, 28.81 mmol), 2-bromo-2-methylpropanoyl bromide (35.5 mL, 288.15 mmol), and solution of 2-hydroxy-2-methylpropanoic acid (30 g, 288.15 mmol) and triethylamine (TEA) (40.2 mL, 288.15 mmol) in dry DCM (50 mL) was used. The reaction was allowed to proceed at 25 °C for 16 h. The crude product was obtained by following the same workup procedure. The crude product was purified by column chromatography using a 30% hexane/ethyl acetate mixture. Yield 86%

2.3.4. Synthesis of 3,3,6,6-Tetramethyl-1,4-dioxane-2,5-dione (Tetramethyl Glycolide)

3,3,6,6-Tetramethyl-1,4-dioxane-2,5-dione was synthesized by following the same reaction conditions used to synthesize 3,3-dimethyl-1,4-dioxane-2,5-dione (DMG). Where Triethylamine (TEA) (9.6 mL, 94.82 mmol) and 2-((2-bromo-2-methylpropanoyl)­oxy)-2-methylpropanoic acid (20 g, 79.90 mmol) were used. The reaction was allowed to complete at 70 °C. Reaction progress was determined with the help of ¹H NMR and stopped after 72 h by adding 6 mL of acetic acid. The crude product was obtained following the same workup procedure and purified by column chromatography using a 30% hexane/ethyl acetate mixture. Yield 80%.

2.4. Polymer Synthesis

2.4.1. General Procedure for ROP of DMG Using P₂-Et Organocatalyst and TU

Considering entry 4 in Table , the experimental procedure is described below. The ROP was conducted by applying the Schlenk technique under an argon atmosphere. In a glovebox, P₂-Et (21 μL, 0.0208 mmol, 1.0 Molar solution in THF) followed by the addition of TU (23.2 mg, 0.0625 mmol). To this reaction mixture, BnOH (14 μL, 0.0139 mmol, 1.0 Molar solution in THF) was added and finally 50.0 equiv of DMG (100 mg, 0.693 mmol) and 2.0 mL of THF were added. The flask was sealed with a Teflon stopper, and the reaction was carried out for 8 h at RT. A few drops of acetic acid solution (containing 5% acetic acid in THF) were added to terminate the reaction. The polymer was precipitated twice in methanol and finally dried at 40 °C in a vacuum oven.

1. Characterization Data of PDMGs and PTMGs Synthesized via ROP of DMG and TMG Using P₂-Et-Based Organocatalyst.

entry	monomer	[ROH]/[P2-Et] [TU]/[M]	temp. (°C)	time (h)	conv. (%)	M n theo.	M n NMR	M n SEC/(Đ)
1	DMG	1/1/0/50	RT	72	74	5400	5600	5400/1.30
2	DMG	1/1/0/50	40	48	73	5300	5100	5800/1.32
3	DMG	1/1.5/0/50	RT	8	90	6500	6400	6200/1.38
4	DMG	1/1.5/4.5/50	RT	8	97	7000	6600	6900/1.22
5	DMG	1/1.5/4.5/100	40	24	87	12,600	12,800	12,900/1.26
6	TMG	1/1/0/50	RT	36	63	5500	4900	6900/1.12
7	TMG	1/1.5/0/50	RT	24	83	7300	7500	9400/1.14
8	TMG	1/2/0/50	RT	16	86	7500	7700	10,200/1.14
9	TMG	1/2.5/0/50	RT	12	90	7800	8000	10,500/1.17
10	TMG	1/2/0/100	40	48	73	12,700	12,800	15,200/1.13

^aCalculated from ¹H NMR spectroscopic analysis of resulting polymer (600 MHz, 298 K, CDCl₃).

^bTheoretical molar mass calculated using [(DMG or TMG/Initiator) × conversion × 144.1 or 172.2 + Mol. wt. of initiator].

^cCalculated from ¹H NMR spectra (600 MHz, 298 K, CDCl₃) by comparison of repeating unit signal intensity to the initiator signal intensity.

^dDetermined by SEC in THF using linear PS standards (35 °C), (all the reactions were performed in THF using BnOH as the initiator).

2.4.2. General Procedure for ROP of TMG Using P₂-Et Organocatalyst

Considering entry 8 in Table , the experimental procedure is described below. The ROP was conducted by applying the Schlenk technique under an argon atmosphere. In a glovebox, P₂-Et (23.2 μL, 0.0232 mmol, 1.0 Molar solution in THF) followed by addition of BnOH (11.6 μL, 0.0116 mmol, 1.0 Molar solution in THF). To this reaction mixture, 50.0 equiv of TMG (100 mg, 0.580 mmol) and 1.0 mL of THF were added. The flask was sealed with a Teflon stopper and the reaction was carried out for 16 h at RT. A few drops of acetic acid solution (containing 5% acetic acid in THF) was added to terminate the reaction. The polymer was precipitated twice in methanol and finally dried at 40 °C in a vacuum oven.

2.5. Depolymerization of PDMG and PTMG

Selective ring-closing depolymerization was conducted on synthesized polymers (PDMG and PTMG). In this procedure, 100 mg of the polymer was dissolved in chloroform in a sublimation apparatus. Subsequently, 2–3 mg of Tin­(II) 2-ethylhexanoate was added as a catalyst. The solution was thoroughly mixed, and the chloroform was evaporated, forming a thin polymer film. This film was subjected to gradual heating under reduced pressure, with the onset of melting observed at approximately 200 °C. Depolymerization was initiated at around 240 °C, and the complete conversion was reached within 1 h. A solid crystalline cyclic monomer with roughly 90% purity was successfully collected via sublimation. The purity of the cyclic monomer was further enhanced by precipitation in hexane to effectively remove aliphatic impurities.

3. Results and Discussion

The racemization-free dimethyl glycolide (DMG) and tetramethyl glycolide (TMG) monomers were synthesized via an acylation/cyclization pathway, , as these monomers are not commercially available. In the acylation step, an α-hydroxy acid was first esterified with an α-bromo acyl bromide to produce a bromo-substituted carboxylic acid. This intermediate was subsequently cyclized under basic conditions to yield the desired DMG and TMG monomers (Scheme B). This synthetic approach proved to be versatile, straightforward, and reliant on relatively inexpensive starting materials. The reaction progress, including the quantitative formation of the intermediate ester and the consumption of the initial α-hydroxy acid, was effectively monitored using ¹H NMR spectroscopy. For DMG, the −CH₂ protons of the intermediate compound shifted from 3.85 to 5.0 ppm in the cyclic monomer, confirming complete cyclization (Figure S1). Additionally, ¹³C NMR analysis confirmed ester formation, as the acid carbonyl signal shifted from 177 ppm in the intermediate to 167 ppm in the DMG cyclic monomer (Figure S2). Similarly, the synthesized TMG monomer and its intermediate were characterized using ¹H NMR and ¹³C NMR, with comparable shifts observed (Figures S3 and S4).

It was observed that an increase in substitution on the glycolide ring leads to a decrease in polymerization ability. This trend is consistent with reported reductions in ring strain values. , According to Saiyasombat et al., increasing the number or size of substituents diminishes polymerization ability due to stronger intramolecular steric repulsions in the polymer chain compared to the ring. This steric hindrance is so pronounced that highly substituted glycolides, such as the TMG monomer, do not undergo polymerization. In our study, density functional theory (DFT) calculations were used to determine the ring strain values by measuring the change in enthalpy of polymerization (ΔH _p). The results align with these observations, showing that the ΔH _p of DMG (−9.8 kcal/mol) and TMG (−8.1 kcal/mol) is significantly higher than that of glycolide (−15.2 kcal/mol). This difference in ring strain energy explains TMG’s resistance to polymerization, as lower strain renders the monomer less reactive (Figure S5).

Our initial ROP experiments for DMG focused on optimizing the solvent, temperature, and catalyst-to-initiator ratio. The ROP of DMG was carried out in benzene at room temperature for 48 h using benzyl alcohol (BnOH) as the initiator and P₂-Et as the catalyst at varying concentrations (0.5, 1.0, and 1.5 equiv) with respect to the initiator. These conditions resulted in monomer conversions of 19, 41, and 67%, respectively (Table S1, entries 1–3). Increasing the catalyst concentration to 2.0 equiv significantly improved monomer conversion to 92% after 48 h (Table S1, entry 4). However, the use of a higher catalyst concentration promoted transesterification reactions, as indicated by the SEC traces, which showed an increased dispersity index (Đ) of 1.59. Additionally, MALDI-TOF spectra revealed two extra series of peaks alongside the main polymer series. These additional peaks correspond to mass losses of dimethyl glycolyl (86 g/mol) and glycolyl (58 g/mol) units, consistent with the formation of transesterification products (Figure S6A,C). To mitigate this issue, thiourea (TU) was added to the reaction, effectively suppressing transesterification. The MALDI-TOF MS spectrum displayed a single population, with peak differences of 144.1 g/mol corresponding to DMG monomer units, confirming the suppression of side reactions (Figure S6D). Despite this improvement, the SEC traces showed a higher Đ of 1.39, which was attributed to monomer solubility issues during the initial stages of the reaction.

To further optimize the polymerization conditions, we switched the reaction solvent to THF, identified as the optimal solvent due to its polarity and the solubility of the reactants during the initial stage. Polymerization was performed using a 1.0 equiv catalyst at room temperature (RT) and 40 °C (Table , entries 1 and 2). Under these conditions, the reactions proceeded slowly, with monomer conversion reaching 74% after 72 h at RT and 73% after 48 h at 40 °C. Despite the slow polymerization rates, the resulting PDMG exhibited moderate control over molecular weight, low dispersity (Đ ≤ 1.3), and minimal transesterification (10–20%), as confirmed by combined ¹H NMR, SEC, and MALDI-TOF analyses (Figure S7). Increasing the catalyst concentration from 1.0 to 1.5 equiv significantly enhanced the monomer conversion, achieving 90% in just 8 h. However, this higher catalyst concentration led to increased transesterification, as evidenced by characterization results (Table , entry 3 and Figure S8).

The ROP of the DMG monomer, conducted with 1.5 equiv catalyst and 4.5 equiv TU (Table , entry 4), yielded transesterification-free PDMG, as illustrated in Figure B. The molecular weight of the resulting PDMG closely matched the targeted M _n values, indicating that the polymerization proceeded as intended, producing the desired polymer structure (Figure A,D). Additionally, the MALDI-TOF MS spectrum of PDMG showed a single series of peaks corresponding to the formula: 144.1n + 108.1­(BnOH) + 23 (Na⁺), confirming that polymerization was successfully initiated by BnOH (Figure B). Importantly, no signs of transesterification were observed, as the spectrum lacked peaks indicative of molar mass losses of 86 or 58 g/mol, which would correspond to dimethyl glycolyl or glycolyl units, respectively. High molecular weight PDMG was successfully synthesized at 40 °C by optimizing the DMG to BnOH feed ratio, as the reaction carried out at room temperature resulted in lower monomer conversion (Table , entry 5 and Figure S9). Furthermore, kinetics experiments conducted with varying concentrations of the P₂-Et catalyst (Figures E and S10A) demonstrated that the polymerization rate depended on the catalyst concentration, as evidenced by the resulting first-order kinetic plot. It is also worth noting that adding TU has only a minor effect on the polymerization rate. This was confirmed by kinetics experiments comparing reactions with and without TU (Figure S10B). While TU does not significantly alter the rate, it plays a crucial role in modulating the basicity of P₂-Et, thereby selectively promoting ring-opening over transesterification.

1.

(A) ¹H NMR spectra (600 MHz, 298 K, CDCl₃) for PDMG and PTMG (Table , entries 4 and 8), (B and C) MALDI-TOF mass spectra of PDMG (DP = 50), and PTMG (DP = 50), (D) SEC traces (THF at 35 °C, PS standard) of PDMG and PTMG (Table , entries 4 and 8), (E and F) first-order kinetics plot for ROP of DMG (catalyzed by P₂-Et/TU) and TMG (catalyzed by P₂-Et) at different times at room temperature, [DMG]₀ = 0.35 mmol, and [TMG]₀ = 0.48 mmol.

The ROP of the DMG monomer involves two potential ring-opening sites: the glycolyl site (Site A) and the dimethyl glycolyl site (Site B), as depicted in Figure S11A. To identify the specific site of ring opening, experiments were conducted under conditions promoting a low degree of polymerization ([BnOH]:[P₂-Et]:[TU]­[DMG] = 1:1:3:2) to isolate and analyze the ring-opening products. The ¹H NMR spectrum of the PDMG dimer (Figure S11B) revealed products with dimethyl glycolyl end groups, indicating that polymerization predominantly proceeds via ring opening at the glycolyl site (Site A). Notably, no peaks were detected in the 4.2–4.3 ppm region, which corresponds to −CH₂OH end groups, confirming the absence of ring opening at the dimethyl glycolyl site (Site B). Further evidence was provided by the ¹H NMR analysis of PDMG with a degree of polymerization of 22, which displayed end-group signals corresponding to two methyl groups attached to a tertiary carbon [−C­(CH₃)₂OH] (Figure S11C). These findings collectively confirm that ring opening occurs exclusively at the less sterically hindered glycolyl site (Site A).

After the successful synthesis of PDMG, our next objective is to polymerize the highly substituted TMG monomer, which exhibits significant steric hindrance. In this work, the TMG monomer was also polymerized for the first time using an organocatalyst. This achievement is particularly noteworthy because the monomer exhibits very low ring strain, making it challenging to polymerize due to increased steric hindrance. When the ring size remains constant, as in the glycolide series, increasing substitution on the α-carbon significantly reduces polymerizability, primarily due to a decrease in enthalpy of polymerization (ΔH _p). The ROP of TMG conducted under conditions identical to those used for DMG, including the same catalyst loading (1.0 equiv of P₂-Et) and a monomer concentration of 0.29 mmol, resulted in a relatively low monomer conversion (28%) even after 72 h (Figure S12). To improve the polymerization efficiency, we increased the TMG monomer concentration to 0.58 mmol while keeping the catalyst loading constant. Under these modified conditions, a significantly higher monomer conversion of 63% was achieved within 36 h (Table , entry 6), demonstrating the positive impact of increased monomer concentration on the reaction rate. Increasing the P₂-Et concentration to 1.5 equiv improved the monomer conversion to 83% within 24 h (Table , entry 7). Importantly, in both cases, the resulting polymers showed no signs of transesterification, as confirmed by characterization data (Figures S13 and S14).

For optimal catalyst loading, we determined that 2.0 equiv P₂-Et achieved 86% monomer conversion within 16 h (Table , entry 8), providing good control over molecular weight and no signs of transesterification (Figure A,C,D). However, increasing the catalyst concentration to 2.5 equiv resulted in noticeable transesterification in the resulting polymer (Table , entry 9 and Figure S15). Similar to DMG, the polymerization of TMG followed first-order kinetics across varying catalyst concentrations, as shown in Figures F and S10A. High molecular weight PTMG was successfully synthesized at 40 °C by optimizing the TMG to BnOH feed ratio, as the reaction carried out at room temperature resulted in lower monomer conversion (Table , entry 10 and Figure S16). After an extended reaction time of 48 h at 40 °C, a monomer conversion of 73% was achieved for a targeted degree of polymerization (DP) of 100. The resulting PTMG exhibited an M _n,_SEC value of 15,200 g/mol with a low dispersity index (Đ) of 1.13, with no evidence of transesterification in the polymer (Figure S16).

Lin and Waymouth have identified two distinct ROP mechanisms based on the relative pK _a values of the thiourea (TU) and the base: the cooperative hydrogen-bond activation mechanism and the TU imidate anion activation mechanism. , In the cooperative hydrogen-bond activation mechanism, the base has a lower pK _a than TU and is unable to deprotonate it. In this scenario, TU activates the carbonyl carbon of the monomer via hydrogen bonding, while the weak base activates the alcohol initiator or chain end, facilitating the ROP process. In contrast, the TU imidate anion activation mechanism occurs when the base has a higher pK _a than TU, enabling it to deprotonate TU. This generates a phosphazenium thioimidate ion pair. In this mechanism, the anionic component of TU activates the alcohol initiator or chain end, while the remaining N–H group in TU activates the carbonyl carbon of the monomer through hydrogen bonding.

In the present study, the reaction mechanism was confirmed to follow the TU imidate anion activation pathway, as evidenced by ¹H NMR analysis and supported by DFT calculations. The observed pK _a values of TU (13.2 in DMSO) and P₂-Et (32.9 in MeCN) confirm that TU is significantly more acidic, allowing P₂-Et to deprotonate TU and generate TU imidate anions. This phenomenon was validated through ¹H NMR analysis, which showed that an equimolar mixture of P₂-Et and TU resulted in the deprotonation of one -NH group in TU. This was evident from the disappearance of the 5.8 ppm peak in the ¹H NMR spectrum of the P₂-Et and TU mixture (Figure S17A), indicating the formation of a phosphazenium thioimidate ion pair. Upon adding equimolar BnOH to this mixture, a phosphazenium alkoxide (BnO^–) ion pair was formed, as demonstrated in Figure S17B. This transformation was confirmed by the shift in the −CH₂ peak of BnOH from 4.56 to 4.67 ppm, indicating the generation of the alkoxide anion. The alkoxide anion acts as a potent nucleophile, capable of attacking the carbonyl carbon of the monomer and initiating polymerization. This mechanistic pathway aligns with the dual activation strategy characteristic of the TU imidate activation mechanism, as illustrated in Figure A.

2.

(A) Proposed reaction mechanism for the organocatalytic ROP of DMG and (B) Gibbs free energy profile diagram, illustrating energy values corresponding to Gibbs free energies (ΔG, kcal/mol).

DFT calculations were conducted to elucidate the reaction mechanism of the ROP of DMG. The activation of TU by P₂-Et occurs through the deprotonation of one of the −NH hydrogens in the TU backbone, forming an active TU imidate anion, as confirmed by ¹H NMR data (Figure S17). This activated species facilitates the ROP of DMG, which progresses from the initial intermediate (IN-0) to the final stable intermediate (IN-5) through two transition states (TS-1 and TS-2). As shown in Figure B, the TU imidate anion, initiator, and DMG monomer form a complex (IN-0) stabilized by noncovalent interactions. In the subsequent step (IN-1), a strong hydrogen bond interaction develops between the initiator and the active TU imidate anion, with a relative Gibbs free energy (ΔG) of 7.1 kcal/mol. The reaction proceeds through the first transition state (TS-1), where proton transfer from the initiator to the TU imidate anion occurs simultaneously with the nucleophilic attack on the carbonyl carbon of the DMG monomer to form tetrahedral intermediates. The calculated Gibbs free energy barrier (ΔG ^‡) for this nucleophilic attack is 22.5 kcal/mol, representing the rate-determining step in the ROP of DMG. The next step involves the ring-opening transition state (TS-2), where cleavage of the bond between the carbonyl carbon and the oxygen atom at the α-position of the DMG monomer occurs, with a ΔG ^‡ of 6.5 kcal/mol. The resulting intermediate (IN-4) spontaneously converts to a more stable intermediate (IN-5), where the oxyanion coordinates with TU, exhibiting a relative Gibbs free energy of −2.7 kcal/mol. The propagation step involves the nucleophilic attack of the oxyanion from the linear, opened DMG on a new DMG molecule, forming a growing polymer chain with two opened monomeric units. This process continues through successive additions of DMG monomers, enabling chain propagation.

The ROP of TMG follows the conventional initiator/chain-end activation mechanism. In this process, the P₂-Et base deprotonates the initiator, generating an alkoxide anion (BnO^–), which acts as a strong nucleophile. This alkoxide anion initiates the polymerization by attacking the carbonyl carbon of the monomer. The formation of the alkoxide anion was confirmed by ¹H NMR analysis, where the −CH₂ peak of BnOH shifted from 4.56 to 4.63 ppm, indicating the generation of BnO^– (Figure S18). A similar computational methodology was applied to the ROP of the TMG monomer. The calculated free energy profile diagram and corresponding reaction mechanism are shown in Figure S19. The ROP of TMG progresses through intermediate complexes (IN-0′ to IN-3′) via two transition states, TS-1′ and TS-2′, with calculated ΔG ^‡ of 14.4 and 11.0 kcal/mol, respectively. As observed with DMG, TS-1′ represents the rate-determining step, involving nucleophilic attack by the alkoxide anion, while TS-2′ corresponds to the ring-opening step.

The thermal stability of the synthesized polymers, PDMG and PTMG, was evaluated using thermogravimetric analysis (TGA) in a nitrogen atmosphere. Figure A presents typical TGA curves, showing the percentage weight loss as a function of temperature. The analysis was performed at a heating rate of 10 °C/min. Both polymers exhibited single-step degradation. The T _1O values (temperature at 10% weight loss) for PDMG and PTMG were 265 and 270 °C, respectively, with both undergoing complete weight loss around 320–340 °C. Differential scanning calorimetry (DSC) was used to analyze the thermal transitions of both polymers. Two consecutive heating and cooling cycles were recorded for PDMG (Figure C) and PTMG (Figure D) at 1 °C/min in a nitrogen atmosphere. Although the first cycle is typically excluded due to thermal history, both cycles are presented to provide insights into phase behavior changes in the polymers. The first heating cycle of PDMG (Figure C, black) showed no glass transition (T _g) but revealed two endothermic melting transitions (T _m) at 140.5 and 155.5 °C. These dual melting transitions may result from different crystalline phases or variations in lamellar crystal thickness. − Interestingly, no crystallization peak (T _c) was observed during the first cooling cycle, even at the slow cooling rate of 1 °C/min. In the second heating cycle, a single T _g was observed at 32.1 °C, confirming that PDMG is amorphous. The absence of a T _c during the cooling cycle suggests that PDMG’s asymmetric structure hinders crystallization. This behavior is consistent with other polymers possessing asymmetric structures, such as poly­(lactide-co-glycolide) (PLGA) and polyisobutyl lactide (PIBL)-based polyesters, which also exhibit an amorphous nature. ,

3.

(A) Thermogravimetric analysis, (B) XRD pattern for pristine and annealed polymer samples, (C and D) DSC thermograms for PDMG and PTMG.

In contrast, both the first and second heating cycles of PTMG, with its symmetrical repeating unit, display distinct endothermic melting transitions (T _m) without a glass transition (T _g) peak (Figure D). The first heating cycle shows melting transitions at 178.5, 184.5, and 190 °C, while the second heating cycle exhibits two melting transitions at 174 and 188 °C. Additionally, both the first and second cooling cycles reveal a crystallization transition (T _c) at 148 °C, indicating that PTMG retains its crystallinity after two heating cycles. These observations are further supported by XRD analysis. For PDMG, a reduction in crystallinity was observed upon annealing, with the % crystallinity decreasing from 38.6% (pristine PDMG; Figure B, solid red line) to 23% (annealed PDMG; Figure B, dotted red line). Conversely, PTMG maintains a consistent % crystallinity of 47.6% in both its pristine and annealed states (Figure B, black solid and dotted line).

Finally, the depolymerization experiments for PDMG and PTMG (Table , entries 4 and 8) were performed at 240 °C using a sublimation apparatus with 1.0–2.0 wt % stannous octoate as the catalyst (Figure A). A pictorial representation of the depolymerization process is provided in Figure S20A. PDMG and PTMG, with respective M _n values of 6900 and 10,200 g/mol, were efficiently depolymerized back into their monomers, DMG and TMG. In both cases, a monomer recovery of over 95% was achieved within 2 h. The sublimed product was analyzed using ¹H NMR (Figure B,C) and ¹³C NMR (Figure S20B,C). For PDMG, the shifts of the −CH₃ peak from 1.63 to 1.71 ppm and the −CH₂ peak from 4.68 to 4.98 ppm indicate effective depolymerization into the DMG monomer. Similarly, for PTMG, the −CH₃ proton peak shifted from 1.56 to 1.73 ppm, corresponding to the formation of the TMG monomer. The resulting DMG and TMG monomers were purified by washing with hexane to remove catalytic impurities, making them suitable for subsequent repolymerization. As a representative example, a repolymerization experiment was carried out using the TMG monomer recovered from depolymerization. The resulting polymer exhibited characteristics comparable to the originally synthesized PTMG, confirming the preservation of monomer integrity during the depolymerization-repolymerization cycle. The repolymerized PTMG achieved a high monomer conversion of 90% and showed no evidence of transesterification, as confirmed by MALDI-TOF analysis (Figure S21), further supporting the efficiency and fidelity of the closed-loop polymerization process.

4.

(A) Schematics for depolymerization of PDMG and PTMG polymers using Sn­(Oct)₂, (B) ¹H NMR spectra (600 MHz, 298 K, CDCl₃) of PDMG (blue), after depolymerization (green, DMG + Sn­(Oct)₂) and after purification (brown, DMG), (C) ¹H NMR spectra (600 MHz, 298 K, CDCl₃) of PTMG (blue), after depolymerization (green, TMG + Sn­(Oct)₂) and after purification (brown, TMG).

4. Conclusions

In conclusion, we report the first organocatalytic ROP of DMG and TMG cyclic diester monomers to synthesize racemization-free PDMG and PTMG at room temperature, using a phosphazene base and thiourea as an organocatalyst. The resulting polyesters exhibit precise molecular weight control, low dispersity, and no transesterification. Mechanistic studies revealed that DMG polymerizes via the TU imidate anion activation pathway, whereas TMG follows the conventional initiator/chain-end activation mechanism, as confirmed by ¹H NMR analysis and supported by DFT calculations. Both polymerizations adhere to first-order kinetics. Notably, PDMG and PTMG are chemically recyclable, yielding their respective DMG and TMG monomers with high selectivity and efficiency. These polyesters offer a potential alternative to commercial, nonrecyclable polymers. This work provides a comprehensive study of the synthesis, polymerization, and depolymerization of DMG and TMG, elucidating their structural properties and polymerization mechanisms. These hydrophobic, sustainable polyesters have potential applications in modern materials, including controlled and targeted drug delivery systems and thermoresponsive injectable hydrogels when combined with hydrophilic blocks.

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acspolymersau.5c00059.

• NMR spectra, SEC data, MALDI-TOF data, and DFT calculations data for PDMG and PTMG polymers (PDF)

CRediT: Shrikant Babanrao Nikam data curation, formal analysis, investigation, methodology, writing - original draft; Prakash Alagi conceptualization, data curation, formal analysis, investigation, methodology, writing - original draft; Jiaxi Xu software, writing - review & editing; Safa Alkhamis formal analysis, investigation; Nikos Hadjichristidis conceptualization, funding acquisition, methodology, project administration, resources, supervision, validation, writing - review & editing.

The authors declare no competing financial interest.