View PDF VersionPrevious ArticleNext Article

Open Access Article
This Open Access Article is licensed under a
Creative Commons Attribution 3.0 Unported Licence
DOI: 10.1039/D4RA07747D (Paper) RSC Adv., 2024, 14, 38079-38084

Simple divalent metal salts as robust and efficient initiators for the ring-opening polymerisation of rac-lactide†

Phoebe A. Lowy and Jennifer A. Garden*
EaStCHEM School of Chemistry, University of Edinburgh, Edinburgh, EH9 3FJ, UK. E-mail: j.garden@ed.ac.uk

Received 30th October 2024 , Accepted 12th November 2024

First published on 29th November 2024

Abstract

Simple divalent metal benzoxides are reported as robust and efficient catalysts for lactide (LA) polymerisation. Following a “pre-stir” step to aid solubility, the best performing catalyst, Zn(OBn)2, gave quantitative monomer conversion in just 30 seconds and performed well under industrially relevant settings with high monomer loadings, bulk polymerisation conditions and non-anhydrous conditions using technical grade LA.

Given the high societal reliance of plastics, there is a major push to develop sustainable alternatives to petrochemical-derived polymers.1 One of the most promising replacements is poly(lactic acid) (PLA), a bio-derived polyester that is biodegradable, recyclable, and has diverse applications in packaging, drinking cups, bottles, fibres and medical sutures/stents.2,3 PLA is typically produced via the ring-opening polymerisation (ROP) of lactide (LA) using tin(II) octanoate (Sn(Oct)2, Fig. 1 A),6–8 which is the standard industry catalyst due to its high activity.4,9 However, the drawbacks include toxicity concerns that limit certain applications within the biomedical field, and the need for high monomer purity because water and other impurities can reduce the polymerisation control.
image file: d4ra07747d-f1.tif
Fig. 1 Examples of the industrial standard Sn(Oct)2, highly active and/or robust Zn-alkoxide initiators, and the simple salts explored for LA ROP in this work.4,5,14,15

While many alternatives to Sn(Oct)2 have been reported, most rely on carefully designed ligands to boost the activity, by improving the solubility, fine-tuning the electronics or changing the metal coordination environment.10 Experimental and theoretical studies have shown that ligand-supported monometallic catalysts for cyclic ester ROP generally proceed via a coordination–insertion mechanism (Scheme 1).7,11 Polymerisation occurs via monomer coordination to the Lewis acidic metal centre with nucleophilic attack, ring-opening and insertion of the cyclic ester into a metal-Oalkoxide bond. Some of the most efficient metal-based initiators reported for cyclic ester ROP are complexes based on magnesium, calcium and especially zinc (see Fig. 1 B, C and S13† for some representative examples).4,12–16 Zinc is particularly attractive because it is inexpensive, colourless, non-toxic, and relatively robust towards air and moisture compared to Mg and Ca.12,13,17 However, such divalent metal complexes typically require multi-step syntheses and are air- and moisture-sensitive, thus need to be handled under extremely anhydrous conditions. This limits their use in large scale, industrially relevant settings, and so the development of robust, air- and moisture-tolerant analogues is highly appealing. While there are relatively few examples of robust LA ROP catalysts based on low toxicity metals,18 Wu and co-workers reported two air-stable heterometallic Li/Zn and Na/Zn catalysts (Fig. 1 D and E).5 Notably, E was active for L-LA ROP when the reactions were performed in air using an unpurified monomer source, giving PLA with controllable molecular weight and high LA conversions of 94%, albeit after a relatively long reaction time (48 h, Đ = 1.4, 90 °C, toluene, [cat][thin space (1/6-em)]:[thin space (1/6-em)][L-LA] 1[thin space (1/6-em)]:[thin space (1/6-em)]175).


image file: d4ra07747d-s1.tif
Scheme 1 Proposed coordination insertion mechanism for LA ROP using a monometallic catalyst.

In terms of simple, non-ligated initiators for cyclic ester ROP, metal-alkyl reagents such as BuLi and Bu2Mg have been reported. While these are efficient initiators, and can even deliver partial heterotacticity control in rac-LA ROP, they are also highly pyrophoric. Alternatively, simple metal-alkoxide salts, including those based on Li,19 Mg,20 Ca,21 Al,22,23 Y24 and La25 have been exploited as efficient initiators for LA ROP, including those added as activity boosters to harness heterometallic cooperativity in mixed-metal catalyst systems.26 Beyond lactide,27 the use of simple zinc salts has been demonstrated in the ROP of other cyclic esters, including the use of Zn lactate in the ROP of 1,4-dioxan-2-one,28 and Zn(OBn)2 for the ROP of gem-disubstituted valerolactones.29 Herein, we report the use of three simple metal salts based on low toxicity metals, Mg(OBn)2, Ca(OBn)2, and Zn(OBn)2, as highly active initiators in LA ROP upon incorporation of a “pre-stir” into the reaction set-up, and probed the robustness of these systems towards industrially relevant conditions.

Metal-alkoxides Mg(OBn)2, Ca(OBn)2, and Zn(OBn)2 were generated from the addition of 2 equiv. of benzyl alcohol (BnOH) to Mg(HMDS)2, Ca(HMDS)2(THF)2 or Et2Zn in THF solvent at room temperature.26 BnOH was selected because it is inexpensive, commercially available, and ligated metal-benzoxides have shown good activities in cyclic ester ROP. The complexes were characterised by NMR spectroscopy and MALDI-ToF mass spectrometry but unfortunately, attempts to crystallise these three complexes from a range of solvents including THF, toluene and chloroform were unsuccessful. 1H NMR characterisation revealed the presence of 1 equiv. of residual HMDSH in isolated Mg(OBn)2 (which was factored into the MW), and trace HMDSH in Ca(OBn)2 (∼0.1 equiv.). As removal of residual HMDSH has previously been reported to decompose group 2 complexes, Mg(OBn)2 and Ca(OBn)2 were employed directly as catalysts.17 While Mg(OBn)2, Ca(OBn)2, and Zn(OBn)2 have been previously reported for rac-lactide polymerisation (r.t., THF solvent), Zn(OBn)2 was completely inactive.26 Here, all three M(OBn)2 salts were tested for rac-LA ROP at 70 °C in toluene with 1 M LA concentration; conditions that have been successful for other ROP catalysts.17 Under these conditions, all three M(OBn)2 salts were active in rac-LA ROP, with reasonable polymerisation control (Table 1, entries 1–4). Zn(OBn)2 and Mg(OBn)2 converted 50 equiv. and 140 equiv. of rac-LA in 40 min, respectively (entries 1 and 2), whilst Ca(OBn)2 was the most active and converted 90 equiv. of LA in just 2 min (entry 3).

Table 1 Optimisation of polymerisation conditions using M(OBn)2 salts for rac-LA ROPa
Entry Cat. LA equiv. Pre-stir (min) Time (min) Conv.b (%) kobs (min−1) Mn calcc (kg mol−1) Mn obsd (kg mol−1) Mn obs/Mn calc (2 chains) Đ
a [rac-LA] = 1 M in toluene, 70 °C. For Mg(OBn)2, the mass of residual HMDSH (1.0 equiv.) was factored into the catalyst mass.b Conversion calculated using 1H NMR spectroscopy.c Mn calc of polymers calculated from Mn calc = M0 × ([M]/[I]) × conversion assuming 2 chains per catalyst.d Mn obs and Đ determined by SEC using polystyrene standards in THF. Values corrected by a correction factor of 0.58.21e [HMDSH][thin space (1/6-em)]:[thin space (1/6-em)][BnOH][thin space (1/6-em)]:[thin space (1/6-em)][rac-LA] = 1[thin space (1/6-em)]:[thin space (1/6-em)]1[thin space (1/6-em)]:[thin space (1/6-em)]100.
1 Zn(OBn)2 100 — 40 50 0.025 3.6 4.8 1.35 1.18
2 Mg(OBn)2 167 — 40 84 0.07 10.1 9.0 0.88 1.40
3 Ca(OBn)2 100 — 2 90 — 6.5 6.8 1.04 1.62
4 Ca(OBn)2 100 — 5 97 — 7.0 4.3 0.62 2.65
5 Zn(OBn)2 100 30 0.5 96 6.0 6.9 7.0 1.00 1.47
6 Mg(OBn)2 167 30 5 91 0.49 10.9 8.3 0.68 1.99
7 Mg(OBn)2 100 30 5 93 0.49 6.7 7.1 1.05 1.52
8 Ca(OBn)2 100 30 2 87 0.81 6.3 5.4 0.90 1.53
9e HMDSH 100 — 1440 0 — — — — —

Performing kinetic studies with the slower Zn(OBn)2 and Mg(OBn)2 catalysts revealed an induction period of approximately 10 min in both cases (Fig. S4†). The analysis of induction periods is important because catalysts can undergo changes during this time, such as a change in the aggregation state, solubilisation of the catalyst species, structural rearrangement of a ligand to facilitate metal coordination,15 or alcoholysis of a metal-alkyl precursor to generate an active metal-alkoxide initiator. Introducing a “pre-stir” step, in this case to aid solubulity of both the catalyst species and monomer, can overcome the induction period, giving a better representation of the true propagation rate.30 Therefore, two separate solutions of M(OBn)2 in toluene and rac-LA in toluene were stirred for 30 min at 70 °C, using a “pre-stir” step to ensure rac-LA dissolution and aid catalyst solubility before the catalyst was added to the LA solution (Table 1, entries 5–8). All three salts gave first order kinetics with respect to monomer concentration (Fig. S5†). For Zn(OBn)2, there was a remarkable increase in rate after introducing this 30 minutes “pre-stir”; the kobs value increased by a factor of 240 (Table 1, entries 5 vs. 1). The activity of Mg(OBn)2 also improved, albeit to a lesser extent, with the observed rate constant increasing by a factor of 7 (entries 6 vs. 2). In contrast, Ca(OBn)2 gave similarly high rac-LA conversions with or without a pre-stir (90% vs. 87% in 2 min, respectively, entries 3 and 8). Identical polymerisation rates were observed whether a 1[thin space (1/6-em)]:[thin space (1/6-em)]167 or 1[thin space (1/6-em)]:[thin space (1/6-em)]100 loading of Mg(OBn)2[thin space (1/6-em)]:[thin space (1/6-em)]rac-LA was used (both kobs = 0.49 min−1, Table 1 entries 6 and 7), where the 1 equiv. of residual HMDSH present was factored in to the catalyst MW in both cases. Note that if the mass of HMDSH was not included in the catalyst MW, the 1[thin space (1/6-em)]:[thin space (1/6-em)]167 ratio becomes a 1[thin space (1/6-em)]:[thin space (1/6-em)]100 ratio. The identical kobs values therefore imply that, under these conditions, factoring residual HMDSH into the calculations does not significantly impact the catalyst activity. No HMDS-capped PLA was detected by MALDI-ToF analysis, and a control reaction confirmed that HMDSH does not initiate rac-LA ROP in the presence of BnOH (entry 9 and Fig. S8†). Taken together, the data shows that HMDSH does not initiate rac-LA ROP in these systems, which is consistent with literature reports for other divalent metal catalysts.17,26

For all three M(OBn)2 salts, Mn obs generally increased linearly with conversion (Fig. S7†), although Zn(OBn)2 gave some lower-than-expected Mn obs values in the late stages of the polymerisation. MALDI-ToF mass spectrometry was used to analyse the PLA end groups, and all three M(OBn)2 salts generated α-benzoxy, ω-hydroxy (major series) as well as α-hydroxy, ω-hydroxy end-capped and/or cyclic PLA as the minor series (Fig. S8†). The minor series were attributed to chain transfer or transesterification reactions, increasing in the later stages of LA polymerisation as has been reported for other catalysts.10,31 Taken together, the presence of OBn/H end groups and the first-order reaction in monomer supports a coordination-insertion mechanism, which has been reported for other metal salt catalysts.32,33 Due to the impressive activities of the M(OBn)2 salts in LA ROP following a “pre-stir”, these salts were further explored targeting industrially relevant conditions. The activities of the M(OBn)2 salts were therefore investigated at higher monomer loadings of up to 2500 equiv. of rac-LA in toluene at 70 °C (Table 2). Interestingly, Zn(OBn)2 retained excellent activities at high monomer loadings (Table 2, entries 1–7, Fig. 2a). For example, at [cat][thin space (1/6-em)]:[thin space (1/6-em)][rac-LA] loadings of 1[thin space (1/6-em)]:[thin space (1/6-em)]1000, Zn(OBn)2 converted 960 equiv. rac-LA in 3.5 min (kobs = 0.94 min−1) achieving a high Mn value of 49.3 kg mol−1 (entry 5). Furthermore, Zn(OBn)2 tolerated even higher monomer loadings of 1[thin space (1/6-em)]:[thin space (1/6-em)]2500, converting 2250 equiv. of LA in 10 min (kobs = 0.23 min−1), and delivered high molecular weight PLA (157.5 kg mol−1, entry 7). Zn(OBn)2 also provided excellent control with a linear increase in Mn obs and narrow dispersities (Đ = 1.14–1.23, entries 6–7, Fig. 2b; [rac-LA] = 2 M). Notably, PLA with molar mass >100 kg mol−1 is attractive for industrially relevant settings,34,35 and the dispersities achieved with Zn(OBn)2 are competitive with those generated using industrial catalysts (Đ < 2).18,36 While increasing the monomer loading led to a decrease in the observed rate constant, kobs, the activity of Zn(OBn)2 remains extremely high and is competitive with some of the most active ligand supported Zn catalysts [Fig. 2a and Table S1†].12,13,15,16,37 For example, under similar conditions, Zn(OBn)2 outperformed B (Fig. 1), with respective kobs values of 1 × 10−1 s−1 and 3.5 × 10−2 s−1 ([cat][thin space (1/6-em)]:[thin space (1/6-em)][rac-LA] = 1[thin space (1/6-em)]:[thin space (1/6-em)]100, [rac-LA] = 1 M in toluene, 70 °C or 60 °C, respectively).15 Additionally, at higher monomer loadings, the activity of Zn(OBn)2 is approximately 20 times greater than that of the dimeric Zn complex C reported by Hillmyer, Tolman and co-workers, albeit under different reaction conditions (kobs = 1.5 × 10−2 s−1 vs. kobs = 0.8 × 10−3 s−1, respectively; [cat][thin space (1/6-em)]:[thin space (1/6-em)][rac-LA] = 1[thin space (1/6-em)]:[thin space (1/6-em)]1000, [rac-LA] = 1 M in toluene, 70 °C vs. CH2Cl2, 25 °C, Fig. 1).16 Zn(OBn)2 also performs well compared to other bimetallic Zn catalysts reported for higher monomer loadings (Tables 2, S1 and Fig. S13†). At a [cat][thin space (1/6-em)]:[thin space (1/6-em)][rac-LA] ratio of 1[thin space (1/6-em)]:[thin space (1/6-em)]600, Zn(OBn)2 gave 50% conversion of LA in 30 seconds at 70 °C in toluene, whereas bimetallic F (Fig. S13 and Table S1†) gave 89% conversion after 24 h in dichloromethane at 25 °C. While Zn(OBn)2 is highly active, it is less active than some of the best-performing Zn complexes reported to date.12,13 For example, under the same monomer loading of [cat][thin space (1/6-em)]:[thin space (1/6-em)][rac-LA] = 1[thin space (1/6-em)]:[thin space (1/6-em)]1000, the bis-Zn complex reported by Brooker, Williams and co-workers (H, Fig. S13 and Table S1†) converted 730 equiv. of LA in just 30 seconds at room temperature vs. 290 equiv. at 70 °C by Zn(OBn)2. Zn(OBn)2 is also less active than the guanidine hydroquinoline ligand-supported Zn complex reported by Herres-Pawlis and co-workers (I, Fig. S13 and Table S1†).13 However, the use of the simple Zn(OBn)2 salt avoids the need for extensive ligand synthesis, bringing significant advantages in terms of cost, time and ease of catalyst synthesis.

Table 2 Probing the robustness of M(OBn)2 salts in rac-LA ROP at high monomer loadingsa
Entry Cat. LA equiv. Time (min) Conv.b (%) kobs (min−1) Mn calcc (kg mol−1) Mn obsd (kg mol−1) Mn obs/Mn calc (2 chains) Đ
a rac-LA pre-stirred in toluene for 30 min prior to addition of M(OBn)2, [rac-LA] = 1 M in toluene, 70 °C.b Conversion calculated using 1H NMR spectroscopy.c Mn calc of polymers calculated from Mn calc = M0 × ([M]/[I]) × conversion assuming 2 chains per catalyst.d Mn obs and Đ determined by SEC using polystyrene standards in THF. Values corrected by a correction factor of 0.58.21e [rac-LA] = 2 M in toluene.f Uncorrected Mn reported as correction factor only valid at lower Mn (<20 kg mol−1).22
1 Zn(OBn)2 300 0.5 73 2.53 15.8 15.2 0.96 1.19
2 Zn(OBn)2 600 0.5 50 1.29 21.6 20.7 0.96 1.18
3 Zn(OBn)2 1000 0.5 44 0.90 31.8 19.0 0.60 1.22
4e Zn(OBn)2 1000 0.5 29 0.94 20.6 20.8 1.01 1.14
5e Zn(OBn)2 1000 3.5 96 0.94 68.9 49.3 0.72 1.33
6e Zn(OBn)2 2500 1 37 0.23 66.1 58.0f 0.88 1.14
7e Zn(OBn)2 2500 10 90 0.23 161.4 157.5f 0.98 1.23
8 Mg(OBn)2 500 5 46 0.14 16.6 12.4 0.75 1.37
9 Mg(OBn)2 1670 5 11 0.03 13.2 — — —
10 Mg(OBn)2 1670 80 88 0.03 105.8 25.0 0.24 1.70
11 Ca(OBn)2 300 1 4 — 0.85 — — —
12 Ca(OBn)2 300 25 17 — 3.6 — — —


image file: d4ra07747d-f2.tif
Fig. 2 (a) Semi-logarithmic plot of rac-LA conversion vs. time at 70 °C with Zn(OBn)2 in toluene solvent with various [cat][thin space (1/6-em)]:[thin space (1/6-em)][rac-LA] loading ratios ([rac-LA] = 1 M for 1[thin space (1/6-em)]:[thin space (1/6-em)]100 (circles); 1[thin space (1/6-em)]:[thin space (1/6-em)]300 (triangles); 1[thin space (1/6-em)]:[thin space (1/6-em)]500 (diamonds) and 1[thin space (1/6-em)]:[thin space (1/6-em)]1000 (pentagon) loadings, and 2 M for 1[thin space (1/6-em)]:[thin space (1/6-em)]2500 (square) loadings), with kobs and R2 values presented in the figure legend. (b) Evolution in Mn and Mw/Mn with monomer conversion for a [Zn(OBn)2][thin space (1/6-em)]:[thin space (1/6-em)][rac-LA] ratio of 1[thin space (1/6-em)]:[thin space (1/6-em)]2500 ([rac-LA] = 2 M) (refer to ESI, Fig. S4,† for additional plots of Mn vs. Mw/Mn).

The high activities of these M(OBn)2 salts, upon the incorporation of a pre-stir, also makes them highly competitive compared to other reported alkoxide salts. For example, reported Ge alkoxides were found to have extremely low activities, requiring 1 week to reach 90% conversion of LA under harsh reaction conditions (120 °C in chlorobenzene).38 In situ generated Ca(OR)2 (R = iPr, Me) and Y(OR)3 (R = iPr) exhibited high activities, where Ca(OR)2, formed from Ca[N(SiMe3)2]2(THF)2 and 2-propanol converted 100 equiv. LA in 30 min (r.t., THF, Đ = 1.05).21,25 Additionally, Y(OR)3 formed via the exchange of bulky phenolate ligands in yttrium tris(2,6-di-tert-butylphenolate) with 2-propanol, converted 150 equiv. L-LA in 8 min (r.t. in dichloromethane, Đ = 1.37). By comparison, Zn(OBn)2 performs well converting 925 equiv. of LA after just 1 minute (70 °C in toluene, Table 2 entry 6).

Compared to Zn(OBn)2, Mg(OBn)2 and Ca(OBn)2 were less robust at higher monomer loadings. While Mg(OBn)2 retained good activities up to a monomer loading of 1670 equiv. rac-LA, converting 1470 equiv. in 80 min (Table 2, entries 8–10, Fig. S10†), the Mn obs values were significantly lower than the Mn calc values and around 20% of the targeted Mn values. This may arise from protic impurities, which are more prevalent at higher monomer loadings and can act as chain transfer agents or cause early termination.36 For Ca(OBn)2, the catalyst activity was rapidly reduced, with a higher [cat][thin space (1/6-em)]:[thin space (1/6-em)][rac-LA] loading of 1[thin space (1/6-em)]:[thin space (1/6-em)]300 giving only 17% conversion after 25 min, compared to 87% in 2 min for a 1[thin space (1/6-em)]:[thin space (1/6-em)]100 loading (Table 2, entry 12 vs. Table 1 entry 8). The sensitivity of group 2 metal alkoxides, especially the heavier analogues, towards higher monomer loadings has been highlighted in literature. While commercially available Sr(OiPr)2 converted 20 equiv. of LA in 1 min at a 1[thin space (1/6-em)]:[thin space (1/6-em)]20 [cat][thin space (1/6-em)]:[thin space (1/6-em)][L-LA] loading (r.t., toluene),39 the activity was significantly lower at higher loadings of 1[thin space (1/6-em)]:[thin space (1/6-em)]100, or 1[thin space (1/6-em)]:[thin space (1/6-em)]200, converting ∼100 equiv. in 15 min or 90 min, respectively. Unlike the alkaline earth metal analogues, Zn(OBn)2 shows promise for maintaining high activities and good polymerisation control at high monomer loadings. While none of the M(OBn)2 salts displayed any stereocontrol, producing only atactic PLA from rac-LA (Pi ≈ 0.5), Zn(OBn)2 generated isotactic PLA from L-LA (Pi = >0.99, Table 3, entry 1). This lack of epimerisation is industrially relevant because isotactic PLA delivers superior mechanical properties compared to atactic PLA.

Table 3 Employing Zn(OBn)2 in LA ROP under industrially relevant conditions
Entry LA equiv. Time (min) Conv.a (%) Mn calcb (kg mol−1) Mn obsc (kg mol−1) Mw (kg mol−1) Đ
a Conversion calculated using 1H NMR spectroscopy.b Mn calc of polymers calculated from Mn calc = M0 × ([M]/[I]) × conversion assuming 2 chains per catalyst.c Mn obs, Mw and Đ determined by SEC using polystyrene standards in THF with a correction factor of 0.58.22d [L-LA] = 1 M in toluene, 70 °C.e [rac-LA] = 1 M in toluene, 70 °C, non-anhydrous conditions.f [rac-LA] = 1 M in toluene, 70 °C, non-anhydrous conditions, technical grade rac-LA.g 150 °C, neat, recrystallised rac-LA.h Uncorrected values reported as correction factor only valid at lower Mn (<20 kg mol−1).22
1d 100 0.5 95 6.8 9.8 — 1.34
2e 100 0.5 96 6.9 4.5 — 1.46
3f 100 3 75 5.4 6.3 — 1.20
4g 2500 30 60 108.0 88.0h 154.2 1.75
5g 2500 60 72 129.6 90.2h 167.0 1.85
6g 2500 90 75 135.0 78.6h 147.0 1.87

Further demonstrating the industrial relevance of Zn(OBn)2, excellent rates for rac-LA ROP were maintained when performing the polymerisations in air (see ESI† for details). In 30 s, 96 equiv. of rac-LA were converted to PLA, which was identical to the polymerisations performed under an inert atmosphere ([cat][thin space (1/6-em)]:[thin space (1/6-em)][rac-LA] = 1[thin space (1/6-em)]:[thin space (1/6-em)]100, toluene, 70 °C; Table 3, entry 2 vs. Table 1, entry 5). In LA ROP, it is common academic practise to purify the monomer via multiple recrystallisations followed by sublimation. Here, Zn(OBn)2 tolerated unpurified, technical grade rac-LA (96% purity), giving a high conversion of 75% in 3 min under air ([cat][thin space (1/6-em)]:[thin space (1/6-em)][rac-LA] = 1[thin space (1/6-em)]:[thin space (1/6-em)]100, toluene, 70 °C; Table 3, entry 3 and Fig. S12†). The high catalyst activities, similar Mn values and moderate dispersities suggest that Zn(OBn)2 is relatively robust towards oxygen, water and other impurities (Table 3, entry 2 and 3). A particular challenge in LA ROP is to achieve high Mn and well controlled polymerisation under industrial ‘bulk’ conditions, i.e. in the absence of solvent. The current industrial standard, Sn(Oct)2, polymerises LA in a few hours at 180 °C under bulk conditions.36 Some simple iron salts have been reported, including halides, acetates, or alkoxides, yet these require unfavourably low monomer to initiator ratios (0.12–1.20 wt% catalyst), high temperatures (∼180 °C) and long reaction times (<48 h) to achieve high monomer conversions.40,41 Interestingly, studies with Zn(OBn)2 under bulk polymerisation conditions showed high activities, appreciable Mn values and moderate dispersities, indicating the potential of Zn(OBn)2 to be utilised in industrially relevant settings (entries 4–7). For instance, Zn(OBn)2 gave 60% conversion of 2500 equiv. of rac-LA in 30 minutes, even when the LA was purified solely through recrystallisation (and not via sublimation), to reduce and simplify purification procedures. This generated PLA with high Mn values (88.0 kg mol−1) and reasonable dispersity (Đ = 1.75, 150 °C, Table 3, entry 4). Notably, PLA with an industrially relevant Mw value of 167 kg mol−1 was produced (Table 3, entry 5).

These studies describe the use of simple Mg(OBn)2, Ca(OBn)2 and Zn(OBn)2 salts as highly active catalysts for LA ROP following the incorporation of a “pre-stir” step into the reaction set-up. While simple metal alkoxide salts often suffer from low polymerisation rates,42 here we show that this can be overcome by incorporating a pre-stir step to improve the solubility. In all cases, the M(OBn)2 catalysts reached high conversions at a 1[thin space (1/6-em)]:[thin space (1/6-em)]100 catalyst[thin space (1/6-em)]:[thin space (1/6-em)]monomer loading (>90%) in less than 5 minutes, and this took less than 30 seconds for the most active catalyst, Zn(OBn)2. Furthermore, Zn(OBn)2 retained catalytic activity at high monomer loadings (up to 2500 equiv.), delivering industrially relevant PLA with high molecular weight (Mn > 150 kg mol−1) and good polymerisation control (Đ = 1.14–1.23). The robustness of Zn(OBn)2 allowed technical grade rac-lactide to be polymerised in air, thus offering ample potential for application on larger scale in industrially relevant settings.

Data availability

The data supporting this article have been included as part of the ESI.†

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

We thank the EPSRC SOFI2 CDT (EP/S023631/1, P. A. L.), UKRI Future Leaders Fellowship (MR\T042710\1, J. A. G.) and the Royal Society (RSG\R1\180101, J. A. G.) for funding. We also gratefully acknowledge Jack Hughes for performing MALDI-ToF analysis on the metal-based complexes.

Notes and references

  1. M. Okada, Prog. Polym. Sci., 2002, 27, 87–133 CrossRef CAS.
  2. R. Auras, B. Harte and S. Selke, Macromol. Biosci., 2004, 4, 835–864 CrossRef CAS PubMed.
  3. S. Inkinen, M. Hakkarainen, A.-C. Albertsson and A. Södergård, Biomacromolecules, 2011, 12, 523–532 CrossRef CAS PubMed.
  4. R. Mehta, V. Kumar, H. Bhunia and S. N. Upadhyay, J. Macromol. Sci. Polymer Rev., 2005, 45, 325–349 CrossRef.
  5. L. Wang, X. Pan, L. Yao, N. Tang and J. Wu, EurJIC, 2011, 5, 632–636 Search PubMed.
  6. E. Castro-Aguirre, F. Iñiguez-Franco, H. Samsudin, X. Fang and R. Auras, Adv. Drug Deliv. Rev., 2016, 107, 333–366 CrossRef CAS PubMed.
  7. E. Fazekas, P. A. Lowy, M. Abdul Rahman, A. Lykkeberg, Y. Zhou, R. Chambenahalli and J. A. Garden, Chem. Soc. Rev., 2022, 51, 8793–8814 RSC.
  8. U. Yolsal, P. J. Shaw, P. A. Lowy, R. Chambenahalli and J. A. Garden, ACS Catal., 2024, 14, 1050–1074 CrossRef CAS PubMed.
  9. A. Kowalski, A. Duda and S. Penczek, Macromolecules, 2000, 33, 689–695 CrossRef CAS.
  10. M. J. Stanford and A. P. Dove, Chem. Soc. Rev., 2010, 39, 486–494 RSC.
  11. R. H. Platel, L. M. Hodgson and C. K. Williams, Polym. Rev., 2008, 48, 11–63 CrossRef CAS.
  12. A. Thevenon, C. Romain, M. S. Bennington, A. J. P. White, H. J. Davidson, S. Brooker and C. K. Williams, Angew. Chem., Int. Ed., 2016, 55, 8680–8685 CrossRef CAS PubMed.
  13. A. Hermann, T. Becker, M. A. Schäfer, A. Hoffmann and S. Herres-Pawlis, ChemSusChem, 2022, 15, e202201075 CrossRef CAS PubMed.
  14. M. T. Martello, D. K. Schneiderman and M. A. Hillmyer, ACS Sustainable Chem. Eng., 2014, 2, 2519–2526 CrossRef CAS.
  15. W. Gruszka, L. C. Walker, M. P. Shaver and J. A. Garden, Macromolecules, 2020, 53, 4294–4302 CrossRef CAS.
  16. C. K. Williams, L. E. Breyfogle, S. K. Choi, W. Nam, V. G. Young, M. A. Hillmyer and W. B. Tolman, J. Am. Chem. Soc., 2003, 125, 11350–11359 CrossRef CAS PubMed.
  17. W. Gruszka, H. Sha, A. Buchard and J. A. Garden, Catal. Sci. Technol., 2022, 12, 1070–1079 RSC.
  18. M. C. D'Alterio, I. D'Auria, L. Gaeta, C. Tedesco, S. Brenna and C. Pellecchia, Macromolecules, 2022, 55, 5115–5122 CrossRef.
  19. M. Bero, P. Dobrzyński and J. Kasperczyk, J. Polym. Sci., Part A:Polym. Chem., 1999, 37, 4038–4042 CrossRef CAS.
  20. D. Wannipurage, S. D'Aniello, D. Pappalardo, L. W. Kulathungage, C. L. Ward, D. P. Anderson, S. Groysman and M. Mazzeo, Dalton Trans., 2023, 52, 8077–8091 RSC.
  21. Z. Zhong, P. J. Dijkstra, C. Birg, M. Westerhausen and J. Feijen, Macromolecules, 2001, 34, 3863–3868 CrossRef CAS.
  22. A. Kowalski, A. Duda and S. Penczek, Macromolecules, 1998, 31, 2114–2122 CrossRef CAS.
  23. P. Dubois, C. Jacobs, R. Jerome and P. Teyssie, Macromolecules, 1991, 24, 2266–2270 CrossRef CAS.
  24. W. M. Stevels, M. J. K. Ankoné, P. J. Dijkstra and J. Feijen, Macromolecules, 1996, 29, 6132–6138 CrossRef CAS.
  25. W. M. Stevels, M. J. K. Ankoné, P. J. Dijkstra and J. Feijen, Macromolecules, 1996, 29, 3332–3333 CrossRef CAS.
  26. W. Gruszka and J. A. Garden, Chem. Commun., 2022, 58, 1609–1612 RSC.
  27. H. R. Kricheldorf and D.-O. Damrau, Macromol. Chem. Phys., 1997, 198, 1753–1766 CrossRef CAS.
  28. H. R. Kricheldorf and D.-O. Damrau, Macromol. Chem. Phys., 1998, 199, 1089–1097 CrossRef CAS.
  29. X.-L. Li, R. W. Clarke, J.-Y. Jiang, T.-Q. Xu and E. Y.-X. Chen, Nat. Chem., 2023, 15, 278–285 CrossRef CAS PubMed.
  30. P. A. Lowy, M. Abdul Rahman, G. S. Nichol, C. A. Morrison and J. A. Garden, ChemCatChem, 2024, e202301338 CrossRef CAS.
  31. A. Rae, A. J. Gaston, Z. Greindl and J. A. Garden, Eur. Polym. J., 2020, 138, 109917 CrossRef CAS.
  32. W. M. Stevels, M. J. K. Ankoné, P. J. Dijkstra and J. Feijen, Macromolecules, 1996, 29, 8296–8303 CrossRef CAS.
  33. Y. Wang and T.-Q. Xu, Macromolecules, 2020, 53, 8829–8836 CrossRef CAS.
  34. A. Buchard, C. J. Chuck, M. G. Davidson, G. Gobius du Sart, M. D. Jones, S. N. McCormick and A. D. Russell, ACS Catal., 2023, 13, 2681–2695 CrossRef CAS PubMed.
  35. T. A. Swetha, A. Bora, K. Mohanrasu, P. Balaji, R. Raja, K. Ponnuchamy, G. Muthusamy and A. Arun, Int. J. Biol. Macromol., 2023, 234, 123715 CrossRef CAS PubMed.
  36. P. V. S. Nylund, B. Monney, C. Weder and M. Albrecht, Catal. Sci. Technol., 2022, 12, 996–1004 RSC.
  37. B. M. Chamberlain, M. Cheng, D. R. Moore, T. M. Ovitt, E. B. Lobkovsky and G. W. Coates, J. Am. Chem. Soc., 2001, 123, 3229–3238 CrossRef CAS PubMed.
  38. A. Finne, Reema and A.-C. Albertsson, J. Polym. Sci., Part A:Polym. Chem., 2003, 41, 3074–3082 CrossRef CAS.
  39. D. Bandelli, C. Weber and U. S. Schubert, Macromol. Rapid Commun., 2019, 40, 1900306 CrossRef CAS PubMed.
  40. M. Stolt and A. Södergård, Macromolecules, 1999, 32, 6412–6417 CrossRef CAS.
  41. H. R. Kricheldorf and A. Serra, Polym. Bull., 1985, 14, 497–502 CrossRef CAS.
  42. O. Dechy-Cabaret, B. Martin-Vaca and D. Bourissou, Chem. Rev., 2004, 104, 6147–6176 CrossRef CAS PubMed.

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d4ra07747d
This journal is © The Royal Society of Chemistry 2024
Click here to see how this site uses Cookies. View our privacy policy here.