Lewis Acid-Base Pair-Regulated Ring-Opening Copolymerization of Succinic Anhydride and Tetrahydrofuran^†

Wenjun Lu^a, Hongjun Yin^a, Ting Li^a, Jing Huang^a, Xu-Hui Zhang^a, Yang Wang^a, Bihua Xia^a, Shibo Wang*^a,b, Weifu Dong^a,c

Abstract

The cationic ring-opening copolymerization (ROCOP) of succinic anhydride (SA) and tetrahydrofuran (THF) was conducted using aluminium trifluoromethanesulfonate (Al(OTf)₃) via bulk polymerization to synthesize poly(butylene succinate) (PBS). The molecular weight and ester content of products were regulated by constructing Lewis acid-base pairs through the introduction of co-catalysts. Compared with the control group without co-catalysts, the resulting products exhibited significant enhancements in molecular weight. The polyester product with 99.38% ester content was achieved by incorporating gallic acid as co-catalyst. When sodium dihydrogen phosphate (NaH₂PO₄) was employed as co-catalyst, a product with molecular weight of 4.008 kDa was obtained. The employment of Lewis base as a co-catalyst yielded products with 85.93% ester content and molecular weight of 4.570 kDa. The regulatory mechanism of Lewis acid-base pairs in this system was systematically investigated based on experimental results, demonstrating potential application prospects in industrial-scale production.

Keywords

ring-opening copolymerization

Lewis acid-base pairs

succinic anhydride

tetrahydrofuran

Introduction

to existing production equipment and reduced application costs. Its advantages are particularly pronounced in scenarios requiring short-term utilization[7].

Ring-opening copolymerization (ROCOP) is a polymerization methodology involving the ring-opening of cyclic monomers under catalytic conditions to form linear polymers through homopolymerization or copolymerization with other monomers[8–18]. This technique is commonly employed in the copolymerization of epoxides and cyclic anhydrides for polyester synthesis. ROCOP demonstrates high flexibility in monomer selection, allowing precise control over polymer chain structures through strategic selection of cyclic monomers with varying architectures[19–22]. It enables molecular weight and molecular weight distribution regulation via adjustment of monomer feed ratios and catalyst design[23]. The reaction conditions of ROCOP are relatively mild, typically conducted at lower temperatures and pressures. Notably, the absence of small molecule by-products (e.g., water) during the process reduces side reactions and simplifies post-processing, exhibiting atom economy that aligns with green chemistry principles. ROCOP exhibits a propensity for alternating copolymerization, facilitating the formation of alternating structures that enhance ester group content in products. Many ROCOP systems demonstrate characteristics of living polymerization, enabling production of high molecular weight polymers with narrow molecular weight distributions[24]. As a promising polymerization strategy, ROCOP has naturally been applied to PBS synthesis research following extensive investigation.

Tetrahydrofuran (THF), a polymer producible through furfural hydrogenation or 1,4-butanediol dehydration, holds significant potential for environmentally sustainable production due to its biomass-derived origin. As a five-membered cyclic ether with low ring strain, THF maintains chemical stability under standard conditions, facilitating transportation and storage. It demonstrates high selectivity for catalysts and undergoes controllable ring-opening reactions under cationic initiators with mild reaction conditions and high efficiency[25–32]. Succinic anhydride (SA), featuring a five-membered cyclic structure and relatively low melting point, enables solvent-free bulk polymerization in its molten state, thereby minimizing post-processing requirements[33–39]. SA exhibits no homopolymerization tendency after ring-opening, effectively suppressing side reactions. Leveraging these advantages, this study employs THF and SA as monomers for the direct synthesis of PBS via ROCOP, investigating the effects of Lewis acid-base pair systems and multifunctional chain terminators on polymerization outcomes[40–45].

Experimental methods

Instruments and reagents

A JWZFG-100ml mechanical high-pressure reactor (Xi’an Tai-Kang Biotechnology Co., Ltd., China), Nicolet iS50 Fourier-transform infrared spectrometer (FT-IR, Thermo Fisher Scientific, USA), TGA2 thermogravimetric analyser (TGA, Mettler-Toledo, Switzerland), DSC3 differential scanning calorimeter (DSC, Mettler-Toledo, Switzerland), AVANCE NEO 600 MHz nuclear magnetic resonance spectrometer (NMR, Bruker, Switzerland), and LC-20ADXR gel permeation chromatograph (GPC, Shimadzu Corporation, Japan) were utilized.

Succinic anhydride (SA, ≥ 99.5%) was purchased from Xu-Ke New Materials (Shandong) Co., Ltd. Tetrahydrofuran (THF, ≥ 99.5%), calcium hydride (CaH₂, ≥ 99.7%), calcium hydroxide (Ca(OH)2, ≥ 99.7%), benzyl alcohol (BnOH, ≥ 99.7%), ethylene glycol (EG, ≥ 99.5%), glycerol (GLY, ≥ 99.5%), pentaerythritol (PER, ≥ 99.5%), absolute ethanol (≥ 99.5%), and chloroform (≥ 99.7%) were obtained from Sinopharm Chemical Reagent Co., Ltd. Aluminium trifluoromethanesulfonate (Al(OTf)₃, ≥ 98%), anhydrous sodium dihydrogen phosphate (NaH₂PO₄, ≥ 99%), anhydrous calcium dihydrogen phosphate (Ca(H₂PO₄)₂, ≥ 99%), magnesium dihydrogen phosphate (Mg(H₂PO₄)₂, ≥ 98%), aluminium dihydrogen phosphate (Al(H₂PO₄)₃, ≥ 98%), and gallic acid (GA, ≥ 98%) were sourced from Macklin Inc. Petroleum ether (≥ 99.7%) was purchased from General-Reagent.

Experimental procedure

SA was dried in a vacuum oven at 40°C for 48 hours, and THF was dehydrated using 2–3 mm molecular sieves for 72 hours.

All materials were weighed in a glovebox. After removing the reactor from the glovebox, nitrogen was purged for 10 minutes. The reactor was then pressurized to 1.0 MPa with nitrogen at a stirring rate of 200 rpm, and the reaction temperature and time were set according to experimental requirements. Upon reaction completion, the reactor was rapidly cooled in a water bath at ambient temperature, and the product was collected.

An ethanol-water solution (2:3 v/v) was prepared and acidified to pH 3 using hydrochloric acid. The crude product was washed with this acidified solution to remove residual catalysts and unreacted small molecules. After filtration, the product was rinsed with deionized water until the pH reached 7. The filtered solid was dissolved in chloroform under stirring, and the solution was rapidly poured into petroleum ether (three times the solution volume) to precipitate potential polyether by-products. The settled solid was filtered and dried in a vacuum oven at 40°C for 24 hours.

The reaction pathway shown in Fig. 1 suggests two distinct initiation mechanisms. In Fig. 1 (a), initiation starts with SA where Al(OTf)₃, as a strong Lewis acid, preferentially coordinates to the carbonyl oxygen of SA. This coordination polarizes and enhances its electrophilicity, leading to ring-opening. The activated terminal then attacks the oxygen atom of THF, inducing its ring-opening and subsequent polymerization to form ester linkages. The newly formed active terminal may react with either SA or THF. When reacting with THF, ether segments are generated, resulting in a copolymer containing both ester and ether segments. In Fig. 1 (b), initiation begins with THF, which undergoes ring-opening under Al³⁺ activation to form an oxonium ion. This species acts as an active centre to attack the carbonyl group of SA, generating ester linkages and propagating the polymerization. Given the lack of homopolymerization propensity in SA, when its chain end becomes an active centre, it can only undergo ring-opening polymerization with THF. The activated THF terminal retains dual reactivity for both homopolymerization and copolymerization. Despite the distinct initiation pathways, the final polymer structure under both scenarios contains both ester and ether segments, demonstrating analogous compositional features.

Fig. 1

(a) SA-initiated plausible reaction pathway for ring-opening copolymerization; (b) THF-initiated plausible reaction pathway for ring-opening copolymerization.

Characterization

FT-IR analysis: Samples were pressed into thin films and analysed in attenuated total reflectance (ATR) mode. Spectra were recorded over 32 scans at a resolution of 4 cm^-1, covering a wavenumber range of 500–4000 cm^-1.

TGA analysis: Conducted under nitrogen atmosphere, samples were heated from 25°C to 600°C at 10°C/min. Decomposition temperatures and maximum decomposition rate temperatures were derived from the thermogravimetric curves.

DSC analysis: Performed under nitrogen atmosphere, samples were heated from 25°C to 150°C at 10°C/min, held for 5 minutes, cooled to 0°C at 10°C/min, held for 5 minutes, and reheated to 150°C at 10°C/min. Crystallization temperatures were determined from the first cooling curve, while melting temperatures were obtained from the second heating curve.

GPC analysis: Number-average molecular weight (Mn), weight-average molecular weight (Mw), and polydispersity index (PDI, Mw/Mn) were measured using a tetrahydrofuran-based gel permeation chromatograph. Polystyrene standards and THF as the mobile phase were employed, with samples filtered through a 0.22 µm syringe filter prior to analysis.

NMR analysis: Proton nuclear magnetic resonance (¹H NMR) spectra were acquired to elucidate molecular structures. Ester content and molecular weight were calculated based on peak area ratios proportional to hydrogen atom counts.

Results and discussion

Effects of Lewis acid-base Pair systems on polymerization

Through a series of preliminary exploratory experiments, aluminum trifluoromethanesulfonate (Al(OTf)₃) was selected as the catalyst, with Lewis acids such as sodium dihydrogen phosphate (NaH₂PO₄) serving as co-catalysts. The optimal molar ratio of catalyst, co-catalyst, SA, and THF was determined to be 3:3:1000:1100. The polymerization was conducted under a nitrogen atmosphere at 1.0 MPa, with a reaction temperature of 120°C maintained for 2 hours. As illustrated in Fig. 2, a comparative analysis of the infrared spectra among the product synthesized with co-catalyst, the product without co-catalyst, and commercially available PBS are presented in Fig. 2 (a). The product exhibits characteristic peaks consistent with PBS: C-H stretching vibration of methylene groups at 2945 cm^-1, C = O stretching vibration of ester linkages at 1712 cm^-1, and C-O-C stretching vibration of ester bonds at 1155 cm^-1. Notably, all major absorption bands align with those of commercial PBS, and the overall spectral pattern demonstrates analogous features. The ¹H NMR spectrum of the polymer product synthesized at 120°C with co-catalyst is shown in Fig. 2 (b). Peaks are observed at 4.12 ppm, assigned to the -R-CH₂-COO- group, 2.64 ppm attributed to the -COO-CH₂ CH₂-COO- group, and 1.72 ppm corresponding to the -CH₂-R-COO- group. The peak area ratios of these signals were calculated to be approximately 1:1:1, consistent with the predicted ¹H NMR pattern of PBS. Combined with prior characterization results, this confirms the successful synthesis of PBS. Additionally, a resonance peak at 3.44 ppm is assigned to the -O-R-CH₂-CH₂-R-O- group. A distinct peak near 3.70 ppm corresponds to the chain-terminal HO-CH₂-R- group, indicative of residual hydroxyl end groups. These characterization results demonstrate the successful synthesis of PBS through ring-opening polymerization of SA and THF under a Lewis acid-base pair catalytic system.

Fig. 2

(a) FT-IR spectra of the product with co-catalyst, product without co-catalyst, and commercially available PBS; (b) ¹H NMR spectrum of the product obtained at a polymerization temperature of 120°C.

Further experiments were subsequently conducted based on the synthesized PBS. Co-catalysts with weaker Lewis acidity (Lewis acids) or Lewis basicity were employed to preserve the activity of the primary catalyst. The reaction products were characterized through ¹H NMR and GPC analyses, as summarized in Table 1. The data reveal that under equivalent co-catalyst loadings, NaH₂PO₄ significantly enhanced the molecular weight of the product while moderately reducing ester content, accompanied by a noticeable narrowing of molecular weight distribution. Mg(H₂PO₄)₂ and Ca(H₂PO₄)₂ induced marginal reductions in ester content and slight increases in molecular weight, though the variations were less pronounced. Notably, Ca(H₂PO₄)₂ exhibited a broadening of molecular weight distribution. Al(H₂PO₄)₃ demonstrated minimal catalytic efficacy, showing negligible effects on both molecular weight and ester content. Gallic acid slightly increased the molecular weight while concurrently enhancing the ester content. CaH₂, functioning as a Lewis base, provided limited enhancement in molecular weight but substantially elevated the ester content of the product. Ca(OH)₂, while reducing ester content moderately, achieved remarkable molecular weight augmentation. Based on these experimental findings, further investigations were conducted on gallic acid, NaH₂PO₄, CaH₂, and Ca(OH)₂ to systematically examine the influence of co-catalyst loading on reaction outcomes.

Table 1
Effects of Lewis acid-base pair types on the ROCOP of SA and THF ^a
Co-Catalyst	Ester (%) ^b	Ether (%) ^b	Mn (kDa)^c	Mw/Mn ^c	TON (g/mol) ^d
Blank	98.89	1.11	1.829	1.43	31325.77
NaH₂PO₄	95.69	4.31	2.791	1.36	35168.01
Mg(H₂PO₄)₂	96.85	3.15	2.408	1.42	36139.82
Al(H₂PO₄)₃	98.64	1.36	1.789	1.45	41555.40
Ca(H₂PO₄)₂	94.45	5.55	2.075	1.97	38215.10
Gallic acid	99.13	0.87	2.274	1.43	35806.12
CaH₂	96.39	3.61	2.836	1.41	32987.27
Ca(OH)₂	85.93	14.07	4.570	1.52	37438.43

^a The molar ratio of other substances was set as SA: THF: Al(OTf)₃ : co-catalyst = 1000: 1100: 3: 3, with nSA = 100 mmol. The reaction pressure was 0.9 MPa, and the reaction time was 2 h. A control experiment without any co-catalyst under identical conditions for 2 h was conducted for comparison. ^b Ester Unit and Ether Unit were determined via 1H NMR analysis. ^c Number-average molecular weight (Mn) and molecular weight distribution (Mw/Mn) were characterized by GPC. ^d Turnover Number (TON) was calculated as the ratio of the mass of the product to the amount of substance of the primary catalyst Al(OTf)₃.

Effects of Gallic Acid on Lewis acid-base pair systems

Polymerization reactions were conducted with varying gallic acid loadings while maintaining other parameters constant. As shown in Table 2 and Fig. 3, when the additive amount ranged between 0.1 and 0.4 mmol, the ester content exhibited an initial increase followed by a gradual decline, yet consistently remained above 99%—surpassing the performance observed in the absence of a co-catalyst—with the maximum ester content of 99.38% achieved at 0.2 mmol. When the additive amount reached 0.5 mmol, the ester content demonstrated a pronounced downward trend, while the TON increased substantially. The molecular weight of the product displayed an overall upward trajectory with increasing gallic acid dosage, showing minor fluctuations within the 0.1–0.4 mmol range and rising to 2.513 kDa at 0.5 mmol.

Table 2
Effects of Gallic Acid Input Amount on Polymerization
Catalyst : Co-Catalyst	Ester (%)	Ether (%)	Mn (kDa)	Mw/Mn	TON (g/mol)
Blank	98.89	1.11	1.829	1.43	31325.77
3:1	99.22	0.78	2.214	1.33	35864.25
3:2	99.38	0.62	2.069	1.27	37370.07
3:3	99.13	0.87	2.274	1.43	35806.12
3:4	99.26	0.74	2.166	1.38	37598.36
3:5	98.04	1.96	2.513	1.53	41058.31

Fig. 3

Plots of ester content and molecular weight versus Gallic Acid loading.

This phenomenon arises because gallic acid contains three adjacent phenolic hydroxyl groups and one carboxylic acid group, demonstrating strong coordination capacity. Its functional groups can form stable polynuclear complexes with Al³⁺, partially replacing the original coordination structure. The formation of these complexes reduces the concentration of free Al³⁺ and weakens the Lewis acidity. Although the initial initiation efficiency was decelerated, the lifespan of active chains was prolonged. Moreover, these complexes may synergistically catalyze with Al(OTf)₃ to promote polymerization, thereby enhancing molecular weight. The phenolic hydroxyl groups in gallic acid partially neutralize the strongly acidic triflate ions in the system, reducing free proton concentration and consequently inhibiting chain transfer and termination reactions, extending chain propagation duration. The bulky aromatic structure of gallic acid generates spatial effects that hinder local aggregation of active chains and reduce cyclization side reactions, further facilitating linear chain growth. However, excessive gallic acid input creates a weakly acidic environment where partial SA hydrolysis leads to polycondensation participation, decreasing ester bond formation rate. This may transform the original ring-opening polymerization into a combined ring-opening-polycondensation process. The proton acid provided by gallic acid promotes ester bond hydrolysis. THF exhibits Lewis basicity and demonstrates higher susceptibility to co-catalyzed ring-opening by Lewis acid systems compared to SA. With increased input of gallic acid, a competitive relationship emerges between gallic acid and SA, leading to enhanced probability of chain transfer from the active chains of ring-opened THF to gallic acid. These factors collectively promote the propensity of THF toward homopolymerization within the reaction system, ultimately resulting in a decrease in ester content.

Effects of NaH₂PO₄ on Lewis acid-base pair systems

Polymerization reactions were performed with adjusted NaH₂PO₄ loadings while maintaining other parameters constant. The products were analyzed via ¹H NMR and GPC, as summarized in Table 3 and Fig. 4. The data indicate that at low loadings (0.1–0.2 mmol), the ester content of the products showed no significant variation, whereas the molecular weight exhibited a pronounced upward trend, with molecular weight distribution remaining around 1.30. Upon increasing the loading to 0.3 mmol, the ester content displayed a marked reduction, while molecular weight continued to rise. At a loading of 0.5 mmol, the ester content remained above 94%, and the molecular weight increased to 3.421 kDa. Further increasing the loading to 0.7 mmol resulted in a molecular weight of 4.008 kDa, albeit with a decline in ester content below 90%. Notably, the molecular weight distribution consistently remained below 1.5, and the TON exceeded 35000 across all conditions.

Table 3
The Effect of NaH₂PO₄ Input Amount on Polymerization
Catalyst : Co-Catalyst	Ester (%)	Ether (%)	Mn (kDa)	Mw/Mn	TON (g/mol)
Blank	98.89	1.11	1.829	1.43	31325.77
3:1	99.08	0.92	2.249	1.31	37447.93
3:2	99.11	0.89	2.570	1.26	40681.68
3:3	95.69	4.31	2.791	1.36	35168.01
3:5	94.45	5.55	3.421	1.37	39604.48
3:7	88.50	11.50	4.008	1.47	35960.49

As can be seen from Fig. 4 (a), the NMR spectra of products obtained using NaH₂PO₄ as a co-catalyst at different loadings. The spectra reveal that after adding NaH₂PO₄, the resonance peak area of the terminal methylene group at approximately 3.70 ppm significantly diminishes, while the resonance peak of the polyether segment methylene group at 3.44 ppm becomes more pronounced with increasing NaH₂PO₄ loading. Correspondingly, as shown in Fig. 4 (b), the aforementioned trend becomes more evident: the ester content of the product decreases with increasing input amount, exhibiting distinct inflection points, while the molecular weight increase curve demonstrates a smoother and more stable progression.

Fig. 4

(a) ¹H NMR spectra in the range of 3.2-4.0 ppm for products with NaH₂PO₄ as a co-catalyst; (b) Plots of ester content and molecular weight versus NaH₂PO₄ loading.

The underlying reason for this phenomenon may be attributed to the establishment of a Lewis acid-base pair system between NaH₂PO₄ (weak Lewis acid) and Al(OTf)₃, generating synergistic catalytic effects. The molecular weight enhancement mechanism is as follows: During the reaction, the coordination between Al³⁺ and PO₄^3- forms a dynamic equilibrium, continuously releasing free Al³⁺ and PO₄^3- to activate SA and stabilize oxonium ions, respectively. Al³⁺ from Al(OTf)₃ coordinates with H₂PO₄^- (dissociated from NaH₂PO₄) via oxygen atoms, forming a bimetallic complex (Al-O- PO₄-Al). The Al³⁺ retains its coordination capability toward SA's carbonyl oxygen, with enhanced coordination strength due to PO₄^3- bridging. The synergistic effect of Al³⁺ and PO₄^3- in the complex promotes more efficient polarization of SA's carbonyl oxygen, resulting in a highly polarized C = O bond that lowers activation energy for nucleophilic attack, thereby accelerating THF ring-opening and ester bond formation. This increases chain propagation rate and molecular weight. PO₄^3- acts as an electron buffer to modulate Al³⁺ acidity, preventing over-polarization-induced side reactions while modifying Al³⁺ coordination environment and enhancing stability.

The ester content reduction mechanism involves nucleophilic attack by H₂PO₄^- on ester bonds, triggering both intramolecular and intermolecular ester exchange reactions (Fig. 5). As shown in Fig. 5 (a), partial ester bond cleavage under H₂PO₄^- generates carboxylic acid and hydroxyl groups, which may act as new initiators. The carboxylic acid's proton activates THF to form oxonium ions, whose reactive termini further polymerize with SA and THF to extend the main chain. However, polyether segments exhibit lower hydrolysis susceptibility compared to polyester segments under H₂PO₄^- influence, leading to preferential ether segment formation and reduced total ester content. As illustrated in Fig. 5 (b), intermolecular ester exchange redistributes ester groups across chains, forming longer main chains. While this elevates molecular weight, it concurrently decreases overall ester content. Simultaneously, as a Lewis acid, NaH₂PO₄ forms a competitive relationship with SA for the active sites on THF chain segments with increasing dosage, thereby driving the reaction toward THF homopolymerization.

Fig. 5

(a) Hydrolysis reaction after NaH₂PO₄ addition; (b) Ester exchange reaction after NaH₂PO₄ addition.

The addition of NaH₂PO₄ enhances the molecular weight and reduces ester content in the product through dual mechanisms of synergistic catalysis and ester bond reorganization. This result holds potential application value in PBS synthesis, as molecular weight and ester content can be modulated by adjusting NaH₂PO₄ dosage, thereby regulating the material's mechanical properties and degradation rate.

Effects of Ca(OH)₂ on Lewis acid-base pair systems

Polymerization reactions were conducted by adjusting the Ca(OH)₂ input amount under otherwise constant conditions, with results characterized by ¹H NMR and GPC analyses as presented in Table 4 and Fig. 6. The data demonstrate that even a minor Ca(OH)₂ input (0.1 mmol) significantly enhances molecular weight. As the Ca(OH)₂ dosage increases, the molecular weight exhibits a stable improvement, ultimately reaching 4.570 kDa. Concurrently, the ester content decreases, having already declined to 85.93% at 0.3 mmol input. The molecular weight distribution also displays a broadening trend. As shown in Fig. 6 (a), upon addition of 0.1 mmol calcium hydroxide, a distinct resonance peak emerges at 3.44 ppm, with further amplification of this ether segment-associated peak area observed at higher dosages. Correspondingly, As shown in Fig. 6 (b), the data illustrates a pronounced increase in product molecular weight and reduction in ester content with escalating Ca(OH)₂ quantities.

Table 4
The Effect of Ca(OH)₂ Input Amount on Polymerization
Catalyst : Co-Catalyst	Ester (%)	Ether (%)	Mn (kDa)	Mw/Mn	TON (g/mol)
Blank	98.89	1.11	1.829	1.43	31325.77
3:1	98.40	1.60	2.948	1.30	36461.02
3:2	94.45	5.55	3.519	1.42	38152.81
3:3	85.93	14.07	4.570	1.52	37438.43

Fig. 6

(a) ¹H NMR spectra in the range of 3.2-4.0 ppm for products with Ca(OH)₂ as a co-catalyst; (b) Plots of ester content and molecular weight versus Ca(OH)₂ loading.

Upon addition of Ca(OH)₂, both OH^- and Ca²⁺ may react with Al(OTf)₃ to form Al-OH or Al-O-Ca structures, reducing the concentration of free Al³⁺ ions. The formation of these complexes diminishes chain termination and prolongs chain propagation duration. As a Lewis base, Ca(OH)₂ establishes a Lewis acid-base pair system with the strong Lewis acid Al(OTf)₃, enabling synergistic catalysis of SA and THF, thereby enhancing the overall catalytic activity. Even at low Ca(OH)₂ dosages, its catalytic effect on SA ring-opening becomes pronounced, with efficacy further improving as dosage increases. This significantly accelerates SA initiation rates, manifesting as a sharp increase in molecular weight. Additionally, Ca(OH)₂ rapidly neutralizes acidic byproducts generated during the reaction, creating an alkaline environment. This alkalinity induces competitive interactions with THF, reducing ester bond formation between THF and SA. Consequently, even minimal Ca(OH)₂ addition leads to a rapid decline in ester content of the product.

Effects of CaH₂ on Lewis acid-base pair systems

Polymerization reactions were conducted by adjusting the CaH₂ input amount under otherwise constant conditions, with results characterized by ¹H NMR and GPC as shown in Table 5 and Fig. 7. The data indicate that ester content decreases progressively with increasing CaH₂ dosage. The dosage of the co-catalyst in the range of 0.1–0.3 mmol induced a gradual decrease in ester content, whereas a precipitous decline in ester content occurred when the dosage exceeded 0.4 mmol. The molecular weight demonstrated an ascending trend, exhibiting a moderate increase at CaH₂ dosages of 0.1–0.3 mmol, followed by a pronounced enhancement reaching a maximum value of 3.782 kDa at 0.4 mmol. Further increases in CaH₂ dosage resulted in a decline in molecular weight. The molecular weight distribution displayed negligible variation with minimal calcium hydride addition, fluctuating around 1.45. However, excessive CaH₂ loading caused significant broadening of the molecular weight distribution.

Table 5
The Effect of CaH₂ Input Amount on Polymerization
Catalyst : Co-Catalyst	Ester (%)	Ether (%)	Mn (kDa)	Mw/Mn	TON (g/mol)
Blank	98.89	1.11	1.829	1.43	31325.77
3:1	97.92	2.08	2.349	1.55	35256.84
3:2	97.32	2.68	2.674	1.34	33962.26
3:3	96.39	3.61	2.836	1.41	32987.27
3:4	93.13	6.87	3.782	1.73	31711.46
3:5	87.24	12.76	3.152	2.05	38821.87

Fig. 7

Plots of ester content and molecular weight versus CaH₂ loading.

This is likely attributable to altered acid-base conditions induced by the strongly basic CaH₂. As CaH₂ dosage increased, the system transitioned progressively from acidic to neutral. Both H^- and Ca²⁺ may react with Al(OTf)₃, forming Al-H or Al-O-Ca structures, which reduce free Al³⁺ concentration and active site availability. Although the chain propagation rate decreased due to fewer active sites, the lifetime of individual active sites may have been extended. Additionally, trace Ca²⁺ might synergize with Al³⁺ to stabilize propagating chain ends, suppressing chain transfer and termination, ultimately yielding higher molecular weights. When CaH₂ input ranged from 0.1–0.4 mmol, the propensity for ester exchange reactions increased with dosage. Concurrently, dominant chain extension resulted in elevated molecular weight with minimal ester content reduction. However, excessive CaH₂ input (≥ 0.5 mmol) caused complete neutralization and structural alteration of Al(OTf)₃, severely diminishing or even eliminating catalytic activity. Under such conditions, side reactions proliferated, with Ca²⁺ likely dominating to promote ester exchange-induced bond rearrangement or cleavage. For instance, β-elimination occurred, where β-hydrogen abstraction from ester bonds generated carboxylic acid and alkene. Under such conditions, the occurrence of side reactions becomes significantly intensified. Calcium ions (Ca²⁺) in the system may assume a dominant role, promoting ester exchange reactions that induce rearrangement or cleavage of ester bonds, exemplified by β-elimination processes.

Compared with Ca(OH)₂, although CaH₂ exhibits stronger Lewis acidity in anhydrous organic solutions, OH^- can directly undergo neutralization reactions with acidic intermediates generated during polymerization, rapidly reducing system acidity and thus decreasing ester bond formation. In contrast, H^- must first react with free protons or active centers in the system to eliminate acidic species, resulting in a longer reaction pathway. Therefore, CaH₂ only induces a significant reduction in the ester content of the product when its dosage reaches 0.5 mmol.

Conclusions

Upon adding a Lewis acid catalyst, the ester content of the polymeric product exceeded 99%, but higher catalyst dosages decreased ester content while increasing molecular weight, ultimately producing a 4.008 kDa product. The reaction mechanisms of gallic acid and NaH₂PO₄ as cocatalysts were elucidated. Gallic acid stabilized the reaction system via coordination effects while suppressing cyclization through steric hindrance, promoting linear chain growth. NaH₂PO₄ enhanced molecular weight and reduced ester content through dual mechanisms: synergistic catalysis and ester bond rearrangement. When Lewis acids served as cocatalysts, they competed with SA for active centers on ring-opened THF chains, thereby inhibiting SA-THF polymerization and lowering ester content.

Introducing Lewis bases to construct Lewis acid-base pair catalytic systems further reduced ester content and elevated molecular weight, achieving a 4.570 kDa product. Ca(OH)₂ significantly increased molecular weight by forming extensive complexes with the primary catalyst, while its rapid OH^- neutralization minimized acidic byproducts, drastically reducing ester content. When CaH₂ was employed as a cocatalyst, it acted as a desiccant to eliminate residual water in reagents. Limited complexation with the primary catalyst and partial promotion of transesterification resulted in only marginal molecular weight gains and a gradual ester content reduction below 0.3 mmol. Although Lewis bases enhanced SA ring-opening efficiency and substantially increased molecular weight, their strong basicity induced competitive interactions with THF, ultimately decreasing ester content by impeding SA-THF polymerization.

In summary, PBS was synthesized via cationic ring-opening polymerization. By constructing Lewis acid-base pair catalytic systems with cocatalysts and adjusting the molar ratios of acid-base pairs and catalyst-to-monomer, precise control over product molecular weight and ester content was achieved. This approach provides a viable pathway for industrial-scale ring-opening polymerization processes.

Supplementary Information

The online version contains supplementary material available at https://osf.io/mz6b4/?view_only=744a40c3d34b4059ab841eec7de327f0

Author contributions:

Wenjun Lu: Formal analysis and investigation, Writing-original draft preparation, Writing-review and editing; Hongjun Yin: Writing-Supervision; Ting Li: Resources; Jing Huang, Xu-Hui Zhang, Yang Wang, Bihua Xia: Writing-review; Shibo Wang: Writing-review and editing; Weifu Dong: Supervision

Funding

This work was supported by Key Technology for Preparation of Biobased Degradable Packaging Materials from Cheap Biomass (2022YFC2104602).

Data availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Declarations

Competing interests

The authors declare no competing financial interests.

References

Zhang X, Zhang C, Zhang X (2022) A Facile and Unprecedented Route to a Library of Thermostable Formaldehyde-Derived Polyesters: Highly Active and Selective Copolymerization of Cyclic Acetals and Anhydrides. Angew Chem Int Ed 61(13):e202117316. 10.1002/anie.202117316

Haslewood MND, Farmer TJ, North M (2022) Synthesis and chemoselective crosslinking of functionalized polyesters from bio-based epoxides and cyclic anhydrides. J Polym Sci 61(4):311–322. 10.1002/pol.20220552

Bautista M, Martínez de Ilarduya A, Alla A, Vives M, Morató J, Muñoz-Guerra S (2016) Cationic poly(butylene succinate) copolyesters. Eur Polymer J 75:329–342. 10.1016/j.eurpolymj.2015.12.012

Puchalski M, Szparaga G, Biela T, Gutowska A, Sztajnowski S, Krucińska I (2018) Molecular and Supramolecular Changes in Polybutylene Succinate (PBS) and Polybutylene Succinate Adipate (PBSA) Copolymer during Degradation in Various Environmental Conditions. Polymers 10(3):251. 10.3390/polym10030251

Barrino F, De La Rosa-Ramírez H, Schiraldi C, López-Martínez J, Samper MD (2023) Preparation and Characterization of New Bioplastics Based on Polybutylene Succinate (PBS). Polymers 15(5):1212. 10.3390/polym15051212

Georgousopoulou I-N, Vouyiouka S, Dole P, Papaspyrides CD (2016) Thermo-mechanical degradation and stabilization of poly(butylene succinate). Polym Degrad Stab 128:182–192. 10.1016/j.polymdegradstab.2016.03.012

Barletta M, Aversa C, Ayyoob M, Gisario A, Hamad K, Mehrpouya M et al (2022) Poly(butylene succinate) (PBS): Materials, processing, and industrial applications. Prog Polym Sci 132:101579. 10.1016/j.progpolymsci.2022.101579

Xie R, Yang G-W, Zhang Y-Y, Lu C, Li W, Wang J et al (2024) A direct comparison between ring-opening copolymerization and polycondensation to produce polyesters using poly(ethylene succinate) as an example. Polym Chem 15(5):412–421. 10.1039/d3py01314f

Jeske RC, DiCiccio AM, Coates GW (2007) Alternating Copolymerization of Epoxides and Cyclic Anhydrides: An Improved Route to Aliphatic Polyesters. J Am Chem Soc 129(37):11330–11331. 10.1021/ja0737568

10.

Sanford MJ, Peña Carrodeguas L, Van Zee NJ, Kleij AW, Coates GW (2016) Alternating Copolymerization of Propylene Oxide and Cyclohexene Oxide with Tricyclic Anhydrides: Access to Partially Renewable Aliphatic Polyesters with High Glass Transition Temperatures. Macromolecules 49(17):6394–6400. 10.1021/acs.macromol.6b01425

11.

Isnard F, Carratù M, Lamberti M, Venditto V, Mazzeo M (2018) Copolymerization of cyclic esters, epoxides and anhydrides: evidence of the dual role of the monomers in the reaction mixture. Catal Sci Technol 8(19):5034–5043. 10.1039/c8cy01174e

12.

Hirschmann M, Andriani F, Fuoco T (2023) Functional and degradable copolyesters by ring-opening copolymerization of epoxides and anhydrides. Eur Polymer J 183:111766. 10.1016/j.eurpolymj.2022.111766

13.

Zhang K, Bai T, Ling J (2023) Iron(III) Triflate as a Green Catalyst for Janus Polymerization to Prepare Block Polyesters. Macromolecules 56(18):7389–7395. 10.1021/acs.macromol.3c01359

14.

Darensbourg DJ, Poland RR, Escobedo C (2012) Kinetic Studies of the Alternating Copolymerization of Cyclic Acid Anhydrides and Epoxides, and the Terpolymerization of Cyclic Acid Anhydrides, Epoxides, and CO2 Catalyzed by (salen)CrIIICl. Macromolecules 45(5):2242–2248. 10.1021/ma2026385

15.

Suh HS, Ha JY, Yoon JH, Ha C-S, Suh H, Kim I (2010) Polyester polyol synthesis by alternating copolymerization of propylene oxide with cyclic acid anhydrides by using double metal cyanide catalyst. Reactive Funct Polym 70(5):288–293. 10.1016/j.reactfunctpolym.2010.02.001

16.

Laiwattanapaisarn N, Virachotikul A, Chumsaeng P, Jaenjai T, Phomphrai K (2022) Ring-Opening Co- and Terpolymerization of Epoxides, Cyclic Anhydrides, and l-Lactide Using Constrained Aluminum Inden Complexes. Inorg Chem 61(50):20616–20628. 10.1021/acs.inorgchem.2c03532

17.

Huijser S, HosseiniNejad E, Sablong R, de Jong C, Koning CE, Duchateau R (2011) Ring-Opening Co- and Terpolymerization of an Alicyclic Oxirane with Carboxylic Acid Anhydrides and CO2 in the Presence of Chromium Porphyrinato and Salen Catalysts. Macromolecules 44(5):1132–1139. 10.1021/ma102238u

18.

Wood ZA, Assefa MK, Fieser ME (2022) Simple yttrium salts as highly active and controlled catalysts for the atom-efficient synthesis of high molecular weight polyesters. Chem Sci 13(35):10437–10447. 10.1039/d2sc02745c

19.

Xu X, Li H, Mehmood A, Chi K, Shi D, Wang Z et al (2023) Mechanistic Studies on Aluminum-Catalyzed Ring-Opening Alternating Copolymerization of Maleic Anhydride with Epoxides: Ligand Effects and Quantitative Structure-Activity Relationship Model. Molecules 28(21):7279. 10.3390/molecules28217279

20.

Longo JM, Sanford MJ, Coates GW (2016) Ring-Opening Copolymerization of Epoxides and Cyclic Anhydrides with Discrete Metal Complexes: Structure–Property Relationships. Chem Rev 116(24):15167–15197. 10.1021/acs.chemrev.6b00553

21.

Isnard F, Santulli F, Cozzolino M, Lamberti M, Pellecchia C, Mazzeo M (2019) Tetracoordinate aluminum complexes bearing phenoxy-based ligands as catalysts for epoxide/anhydride copolymerization: some mechanistic insights. Catal Sci Technol 9(12):3090–3098. 10.1039/c9cy00806c

22.

Yang Y, Sung K, Lee JD, Ha J, Kim H, Baek J et al (2024) Ultrathin Zn-Gallate Catalyst: A Remarkable Performer in CO2 and Propylene Oxide Polymerization. ACS Sustain Chem Eng 12(10):3933–3940. 10.1021/acssuschemeng.3c06058

23.

Santulli F, Grimaldi I, Pappalardo D, Lamberti M, Mazzeo M (2023) Salen-like Chromium and Aluminum Complexes as Catalysts in the Copolymerization of Epoxides with Cyclic Anhydrides for the Synthesis of Polyesters. Int J Mol Sci 24(12):10052. 10.3390/ijms241210052

24.

Ikeda I, Simazaki Y, Suzuki K (2003) Synthesis of graft polyesters by ring-opening copolymerization of epoxy‐terminated poly(ethylene glycol) with acid anhydrides. J Appl Polym Sci 42(11):2871–2877. 10.1002/app.1991.070421102

25.

Maas LM, Schmid JR, Fasting C, Vossnacker P, Mavroskoufis A, Hopkinson MN (2021) Activation of tetrahydrofuran with 2-((Fluoroalkyl)thio) Benzothiazolium reagents. Tetrahedron. ;101:132512. doi: ARTN 132512

10.

1016/j.tet.2021.132512

26.

Kunnari SM, Oilunkaniemi R, Laitinen RS, Ahlgrén M (2001) An unexpected tetrahydrofuran ring opening: synthesis and structural characterization of Ph3PO(CH2)4TeBr4. J Chem Soc Dalton Trans 233417–3418. 10.1039/b108790h

27.

Li Y, Yu B (2010) Glycosylation initiated cationic ring-opening polymerization of tetrahydrofuran to prepare neo-glycopolymers. Chem Commun 46(33):6060. 10.1039/c0cc00566e

28.

Han Z-Z, Zhang C-P (2021) Phenyl(trifluoroethyl)iodonium-triflate-initiated ring-opening polymerization of tetrahydrofuran. Tetrahedron Lett 73:153146. 10.1016/j.tetlet.2021.153146

29.

Mukherjee D, Osseili H, Truong K-N, Spaniol TP, Okuda J (2017) Ring-opening of cyclic ethers by aluminum hydridotriphenylborate. Chem Commun 53(24):3493–3496. 10.1039/c7cc01159h

30.

McCarthy BJ, Hogen-Esch TE (1996) Synthesis of Narrow Distribution Poly(tetrahydrofuran) Initiated by Alkylboron Triflates. Macromolecules 29(8):3035–3037. 10.1021/ma951314n

31.

Seto R, Yamada S, Matsumoto K, Endo T (2018) Synthesis of block copolymers through umpolung or treatment of propagating end of living cationic polytetrahydrofuran. Polym Bull 76(7):3355–3370. 10.1007/s00289-018-2527-3

32.

Wu Q, Li L, Yu Y, Tang X (2008) The linear relations and living feature in cationic ring-opening copolymerization of epoxy/THF system. Colloid Polym Sci 286(6–7):761–767. 10.1007/s00396-007-1832-4

33.

Ghosh S, Glöckler E, Wölper C, Tjaberings A, Gröschel AH, Schulz S (2020) Active Ga-catalysts for the ring opening homo- and copolymerization of cyclic esters, and copolymerization of epoxide and anhydrides. Dalton Trans 49(38):13475–13486. 10.1039/d0dt02831b

34.

Feng L, Liu Y, Hao J, Xiong C, Deng X (2012) Alternating copolymers with degradability and quantitatively controlled thermosensitivity. J Polym Sci Part A: Polym Chem 50(9):1812–1818. 10.1002/pola.25950

35.

Hosseini Nejad E, van Melis CGW, Vermeer TJ, Koning CE, Duchateau R (2012) Alternating Ring-Opening Polymerization of Cyclohexene Oxide and Anhydrides: Effect of Catalyst, Cocatalyst, and Anhydride Structure. Macromolecules 45(4):1770–1776. 10.1021/ma2025804

36.

Feng L, Yang Z, Liu Y, Hao J, Xiong C, Deng X (2014) Copolymerization of succinic anhydride and epoxide with diethylene glycol side chain. Iran Polym J 23(3):217–226. 10.1007/s13726-013-0217-4

37.

Virachotikul A, Laiwattanapaisarn N, Wongmahasirikun P, Piromjitpong P, Chainok K, Phomphrai K (2020) Ring-Opening Copolymerizaton of Cyclohexene Oxide and Succinic Anhydride by Zinc and Magnesium Schiff-Base Complexes Containing Alkoxy Side Arms. Inorg Chem 59(13):8983–8994. 10.1021/acs.inorgchem.0c00839

38.

Maeda Y, Nakayama A, Kawasaki N, Hayashi K, Aiba S, Yamamoto N (1997) Ring-opening copolymerization of succinic anhydride with ethylene oxide initiated by magnesium diethoxide. Polymer 38(18):4719–4725. 10.1016/s0032-3861(96)01088-9

39.

Isnard F, Lamberti M, Pellecchia C, Mazzeo M (2017) Ring-Opening Copolymerization of Epoxides with Cyclic Anhydrides Promoted by Bimetallic and Monometallic Phenoxy–Imine Aluminum complexes. ChemCatChem 9(15):2972–2979. 10.1002/cctc.201700234

40.

Hu L-F, Chen D-J, Yang J-L, Zhang X-H (2020) An Investigation of the Organoborane/Lewis Base Pairs on the Copolymerization of Propylene Oxide with Succinic Anhydride. Molecules 25(2):253. 10.3390/molecules25020253

41.

Ren Y, Jiang O, Zeng H, Mao Q, Jiang H (2016) Lewis acid–base bifunctional aluminum–salen catalysts: synthesis of cyclic carbonates from carbon dioxide and epoxides. RSC Adv 6(4):3243–3249. 10.1039/c5ra24596f

42.

Ji H-Y, Song D-P, Wang B, Pan L, Li Y-S (2019) Organic Lewis pairs for selective copolymerization of epoxides with anhydrides to access sequence-controlled block copolymers. Green Chem 21(22):6123–6132. 10.1039/c9gc02429h

43.

Ding Z, Wang M, Wang B, Pan L, Li Y (2023) Organo-Catalyzed Cationic Ring-Opening Copolymerization of Cyclic Anhydrides with Oxolanes: Access to Structurally Diverse Polyesters. Macromolecules 57(1):98–109. 10.1021/acs.macromol.3c01899

44.

Wang Y, Kunioka M (2005) Ring-Opening Polymerization of Cyclic Monomers with Aluminum Triflate. Macromolecular Symposia 224(1):193–206. 10.1002/masy.200550617

45.

Zhang X, Sun Y, Zhang C, Zhang X (2023) Upcycling Polytetrahydrofuran to Polyester. CCS Chem 5(5):1233–1241. 10.31635/ccschem.022.202202072