Amine-bearing cyclic ketene acetals for pH-responsive and degradable polyesters through radical ring-opening polymerisation

Abstract

A series of previously unreported amine-bearing cyclic ketene acetals (CKAs) furnished with different alkyl-substituents were prepared. These CKAs pave the way for a new range of pH-responsive and degradable polyesters from radical ring-opening polymerisation (RROP). CKA 2-methylene-1,3,6-trioxocane (MTC) was copolymerised with the amine-bearing CKAs to obtain a range of new degradable smart copolymers. Varying the type and amount of amine-bearing CKA allowed to fine-tune the value of the polymers, showcasing the versatility of this approach.

---

Following the increasing issues with plastic waste, polyesters like poly(caprolactone) (PCL) or poly(lactic acid) (PLA) and similar materials have become increasingly important materials owing to the biodegradable ester units in the polymer backbone. They found numerous applications in tissue engineering, drug delivery or commodity materials.¹ Traditional synthetic routes towards polyesters are either by polycondensation or ring-opening polymerisation.² However, the former tends to generate polyesters with low molecular weight and the latter method requires rigorous synthetic procedures (especially from ionic polymerization) to afford pure polymers.³ Proposed by Bailey et al., radical ring-opening polymerisation (RROP) of cyclic ketene acetals (CKAs) has received considerable attention and is regarded as an alternative tool to generate polyesters.⁴ RROP possesses the robustness of radical polymerisations, yields degradable polyesters^2,5 and allows for functional groups (e.g. amines) within the polymer backbone.⁶

Homopolymerisations of CKAs are known to generate polyesters (bio)degrading under basic or enzymatic conditions.^7,8 Until now, 2-methylene1,3-dioxepane (MDO), 2-methylene-1,3,6-trioxocane (MTC) and 2-methylene-4-phenyl-1,3-dioxolane (MPDL) are the CKAs with be highest publication record.^9–13 These CKAs are generally characterised by an almost complete degree of ring-opening and a known structure in terms of the main repeating unit and branching positions from inter- or intramolecular H-transfer.^13–16 Some of these polymers already have an additional functionality like the semi-crystallinity of poly(MDO) and the higher hydrophilicity of poly(MTC) as compared to other polyesters from RROP.^12–14 However, the lack of responsive groups limits their range of potential applications like drug delivery.¹⁷

One strategy to circumvent this issue is the copolymerisation of CKAs with various responsive vinylic monomers.^5,18–20 This strategy introduced ester linkages into the vinyl polymer backbone to confer (bio)degradability² and responsive vinylic monomers endowed responsive properties.^21,22 CKAs including MDO and BMDO have been successfully combined with oligoethylene glycol methacrylate and N,N-dimethylacrylamide for water solubility,^23–25N-isopropylacrylamide for temperature responsiveness^26,27 and 2-(diethylamino)ethyl methacrylate for pH-responsivness.^23,28 CKA-bearing copolymers have been developed in the field of functional and degradable nanoparticles,^29–31 hydrogels,³² 3D printing materials,³³ and coatings.^34,35 However, the unfavorable CKA/vinylic copolymerisation parameters caused a composition drift towards gradient copolymer, which prohibited the randomly distributed ester units in the polymer backbone. This will therefore lead to an incomplete degradation into a mixture of oligomers.^36,37 Hence, a reproducible method towards responsive and fully degradable polyesters from RROP is still a challenge.

Our group has focused on fully degradable polyesters purely based on CKAs,^6,8,15,38 including amine-bearing CKAs for pH-responsive polyesters. This includes the amine-bearing N-iso-propyl-2-methylene-1,3,6-dioxazocane (ⁱPr-MAC, 7c), allowing for pH-responsive and biodegradable nanoparticles.⁶ With a cloud point of 6.0, the pK_a of the polymerised ⁱPr-MAC was similar to the well-known pH-responsive poly(diisopropyl aminoethyl methacrylate) (PDPA).³⁹ Copolymerisation of 7cⁱPr-MAC with 4,7-dimethyl-2-methylene-1,3-dioxepane (DMMDO) led to copolymers with an increased molecular weight, but the pH-responsive behaviour shifted from 6 for P(ⁱPr-MAC) to below 4 for the copolymer. Taking advantage of these merits, i.e., pH-sensitivity and biodegradability, these purely CKA-based polyesters and nanoparticles accessed potentials in biomedical field, but lacked a pK_a value closer to neutral pH. Inspired by this notion, we adopted the reported synthetic route to prepare new amine-bearing CKAs with varying substituents (Alk-MACs), including the ethyl (Et-MAC, 7a), n-propyl (ⁿPr-MAC, 7b), n-butyl (ⁿBu-MAC, 7d), sec-butyl (^sBu-MAC, 7e), iso-butyl (ⁱBu-MAC, 7f) and tert-butyl (^tBu-MAC, 7g) derivatives (Fig. 1a). Following monomer synthesis and calculations on the molecule structure, we subsequently copolymerised the Alk-MACs with the more hydrophilic MTC as it ought to have a pK_a closer to neutral pH (Fig. 1b). Hydrolytic degradation will be briefly discussed as well (Fig. 1c).

Fig. 1 (a) Synthetic route towards Alk-MACs through an intermediate carbonate. Labels 3a–g, 5a–g, 7a–g represent alkyl-substituted diethanol amines, intermediate carbonates and CKAs, respectively. (b) Copolymerisation of Alk-MACs with the conventional CKA MTC yields pH-responsive and degradable copolymers. (c) The amines in the Alk-MAC segments get protonated upon acidification and copolymers become hydrophilic and upon hydrolytic degradation the ester bonds get cleaved to yield small molecules.

The first goal of this work was to synthesise the new Alk-MACs following a similar protocol (carbonate route) as for the published ⁱPr-MAC. Starting from diethanolamine 1, alkyl-substituted diols 3 were prepared by alkylation with the corresponding bromoalkene agents 2. Following some optimisation, the previously reported N-propyldiethanolamine 3b, N-iso-propyldiethanolamine 3c, N-sec-butyldiethanolamine 3e and also N-iso-butyldiethanolamine 3f were synthesised (Tables S2–S5 and Fig. S3†). The substances 3a, 3d, and 3g were sourced commercially.

These alkyl-modified diols 3a–g and ethyl chloroformate 4 then reacted in the presence of triethylamine (TEA) to yield the intermediate cyclic carbonates 5a–g. Reaction conditions, including reaction temperature, feeding speed of TEA, feed ratio of reagents, were fine-tuned individually to ensure optimal yield of each carbonate. For instance, the best yield of ethyl carbonate 5a (23%) was achieved with a low feeding speed of TEA (15 μL min⁻¹). Conversely, a faster feeding speed of TEA (40 μL min⁻¹) gave the best yield for n-propyl carbonate 5b and tert-butyl carbonate 5g with 29% each. An even higher feeding speed of 100 μL min⁻¹ then gave the optimal yield of 45% for sec-butyl carbonate 5e and was increased further to 400 μL min⁻¹ for the formation of n-butyl carbonate 5d (25% yield) and iso-butyl carbonate 5f (45% yield, summarized in Tables S6–11†). Hence, larger alkyl substituents mostly demanded a higher feeding speed of TEA for optimal carbonate formation. ¹H NMR and ¹³C NMR spectroscopy with full signal assignment could be achieved for all carbonates 5a–g, always including the signal at around 155 ppm in the ¹³C NMR, (carbonate-bearing carbon atom). HR-ESI-MS measurements with positive mode were used to confirm the successful synthesis of all cyclic carbonates 5a–g (Fig. S4–10†). Molecular models were calculated to get insights into the geometry of the structures (Fig. S20†) and will be discussed separately later.

Methylenation of all cyclic carbonates 5a–g (1.0 equiv.) with the Petasis reagent 6 (2.3 equiv.) were performed in THF at 65 °C for 22 h. All the new CKAs were obtained in a low yield (from 6% for 7d over 11–15% for 7a–b, 7e–g to 25% for 7c (Table S1†)), representing the delicacy of the carbonate route. ¹H, ¹³C NMR and HR-ESI-MS measurements confirmed the successful preparation of these new molecules (Fig. 2 and Fig. S12–18†). As depicted in Fig. 2, signals a–d could be assigned to the CKA ring. From the core CKA functionality, the tertiary carbon atom c between the oxygen atoms was positioned at 164 ppm for the ¹³C NMR spectrum (Fig. S12–S18†). The intense singlet d in the range of 3.53–3.61 ppm in the ¹H NMR spectra and new signal at 69 ppm in the ¹³C NMR spectra represented exo-methylene group and hence proved the formation of the MACs. Signals 1–4 were attributed to the protons on the alkyl-substituents and remained mainly unchanged from the carbonates for all substituents. As compared to other chemical shifts of the starting diols and intermediate substances, protons a, a′ and 2, 2′ of the chiral ^sBu-MAC 7e and the corresponding carbonate 5e showed an additional splitting (see in Fig. 2 and S8†), which is typical for chiral molecules.

Fig. 2 ¹H NMR spectra of all new Alk-MACs 7a–b, 7d–g with their respective alkyl side groups. The inserts are energetic optimal structures of CKAs calculated at 0 K, under vacuum; red balls represent oxygen atoms, grey balls represent carbon atoms, white balls represent hydrogen atoms, blue balls represent nitrogen atoms. The asterisked signals in the ¹H NMR spectra represent the trace of Petasis reagent (black, 6.05 ppm), solvent and impurities (blue, tert-butanol: 1.28 ppm, acetone: 2.17 ppm, THF: 1.85 and 3.76 ppm).

Additionally, theoretical calculations of the molecular structures of all intermediate carbonate and corresponding CKAs were performed using software package GAMESS with basis set STO-6G,⁴⁰ simulating conditions of 0 K and vacuum. These helped us in understanding the energetically optimal structure of the carbonates 5a–g and CKAs 7a–g (see in the insets in Fig. 2 and Fig. S20†). They revealed that, except for the chiral ^sBu derivatives 5e and 7e, the side chain on the amine stretched away from the carbonate or CKA ring. Both chiral derivatives, however, presented an additional structural distinctiveness. Here, the ^sBu-side group tended to bend inwards to the eight-membered ring, which might lead to significant steric hindrance around the tertiary amine. However, as the calculations were performed at simulated 0 K, the effect on the molecule at room temperature remained unknown.

Copolymerisation of the previously reported ⁱPr-MAC 7c and conventional CKA MTC was the starting point to fabricate the pH-responsive polyesters through free RROP. By varying feed ratio of 7c to MTC from 5/95, 10/90, 25/75 to 50/50 (mol%), copolymer composition was modulated. After polymerising either thermally or under UV light for 24 h, purified copolymers were obtained through dialysis. All copolymers were named according to the feed ratio of Alk-MACs and the polymerisation method. For instance, ⁱPr-MAC-5T represents a 7c/MTC mixture with a feed ratio of 5/95 that was subjected to thermal polymerisation.

To determine the monomer conversion and the final molar fractions, crude and purified copolymers were analysed by ¹H NMR spectroscopy (Fig. S21–24†). Signals attributed to protons on the P(MTC) parts were consistent with reported signal assignments.¹⁴ For the P(ⁱPr-MAC) segment, the signal centered at 2.93 ppm appeared in all the samples, which is attributed to the α-proton on the ⁱPr side chain. This signal was significant, as the chemical shift remained virtually unchanged for the monomer and the polymer and confirmed the successful incorporation of the ⁱPr group through all attempts. After workup, the ester signal of P(MTC) (4.20 ppm) and amine signal of P(ⁱPr-MAC) (2.93 ppm) were integrated to calculate the molar fractions in the copolymers (Fig. S23 and section 3.3 of the ESI†). Intriguingly, the final molar fraction of ⁱPr-MAC in the copolymers was always higher than the feed ratio, indicating that ⁱPr-MAC was more reactive than MTC. Additionally, UV-irradiated polymerisation consistently led to a higher ⁱPr-MAC content in the copolymers as compared to thermal polymerization, implying an even greater imbalance of reactivities (Table 1). Copolymers of ⁱPr-MAC and MTC could hence be gradient copolymers, which means copolymers first enriched with ⁱPr-MAC. Considering the omnipresent dispersity within polymers from free radical polymerisation, the partial presence of homopolymers could not be excluded. DOSY measurements of the bimodal copolymer ⁱPr-MAC-5T was carried out (Fig. S26†), and showed that the self-diffusion coefficients related to ⁱPr-MAC and MTC motifs within the main diffusion peak and thus substantiated the successful formation of a copolymer. Additionally, MTC branching units occurred by intramolecular (MTC back-biting) or intermolecular H-abstraction reactions can also be observed. The characteristic signals around 1.20 ppm are related to the methyl end groups from each MTC branching unit (Fig. S24†).^13,41 The SEC chromatograms of the resultant copolymers showed that with increasing ⁱPr-MAC content, copolymers exhibited a broader distribution ranging from 1.3 for 5 mol% to over 11 for 50 mol% (Fig. S29† and Table 1). While thermal initiation showed a consistent increase in dispersity with rising ⁱPr-MAC content, tendency was less clear for polymerizations under UV irradiation, again implying a more reliable reaction under purely thermal conditions. The corresponding results should be treated with caution owing to the complex polymeric structures including ring-opened and ring-retaining structures as well as branched and linear units in all possible combinations from MTC and ⁱPr-MAC. 2D NMR measurements of sample copolymers (Fig. S27 and 28†) still allowed for allocating some major signals. As the ester bonds of MTC and the amines of ⁱPr-MAC could be confirmed, both structural arguments for the targeted degradability and pH-responsiveness were present.

Table 1 Experimental conditions of copolymerisation of MTC with Alk-MACs and the composition of resultant copolymers

Name	f MAC,0/fMTC,0 ^a	% Conv. MAC/MTC^b	F MAC/FMTC ^c	M n ^d (kg mol^−1)	M w ^d (kg mol^−1)	Đ	^e
a Feed ratios of Alk-MACs to MTC. b Conversion of Alk-MACs and MTC, calculated by ^1H NMR spectra. c Final molar fraction of Alk-MACs to MTC in the resulting copolymers, calculated from ^1H NMR spectra. d M n, Mw and polymer dispersity were measured by SEC at 25 °C with a PolarGel-M column (300 × 7.5 mm), using HPLC-Pump 1200 pump system (Agilent technologies) equipped with a multi angle laser light scattering (MALLS) detector (MiniDAWN TREOS II, Wyatt Technology) and a RI detector (K-2301 from KNAUER). DMAc with 3 g L^−1 LiCl was used as an eluent and the flow rate was 1 mL min^−1. e   was obtained through UV-vis method, the transition point was determined when transmittance decreased 50%. It should be noted that shown in the Table 1 was the average value obtained from three measurements, including the corresponding error margin. f Conversion of Alk-MACs was quite high and no presence of vinyl protons can be found by ^1H NMR spectra.
^iPr-MAC-5T	5/95	67/35	7/93	25.8	32.9	1.3	—
^iPr-MAC-10T	10/90	40/28	14/86	26.6	43.5	1.6	—
^iPr-MAC-25T	25/75	48/25	34/66	32.2	155	4.8	7.0 ± 0.1
^iPr-MAC-50T	50/50	35/37	69/31	18.2	193	11	6.5 ± 0.1
^iPr-MAC-5U	5/95	80/67	10/90	16.3	47.4	2.9	—
^iPr-MAC-10U	10/90	100^f/38	19/81	29.3	68.2	2.3	—
^iPr-MAC-25U	25/75	64/15	55/45	9.2	31.5	3.4	6.9 ± 0.2
^iPr-MAC-50U	50/50	100^f/44	80/20	16.5	248	15	6.7 ± 0.2
Et-MAC-25T	25/75	22/17	37/63	17.8	44.8	2.5	5.7 ± 0.3
^nPr-MAC-25T	25/75	27/26	26/74	15.3	42.4	2.8	6.1 ± 0.2
^nBu-MAC-25T	25/75	22/17	36/64	11.8	126	11	6.6 ± 0.3
^sBu-MAC-25T	25/75	29/13	45/55	49.1	420	8.6	6.6 ± 0.3
^iBu-MAC-25T	25/75	30/17	30/70	21.5	183	8.5	6.4 ± 0.2
^tBu-MAC-25T	25/75	60/19	50/50	10.8	65.4	6.1	7.2 ± 0.3

---

Following the more reliable reaction, thermal polymerisation was applied to other Alk-MAC/MTC pairs. A feed ratio of 25/75 (mol%, Alk-MAC/MTC) was applied to yield an Alk-MAC content between 25 and 50 mol% as this would ensure pH-responsiveness and also the MTC content necessary to keep structural diversity within reason. The purified copolymers were analysed by ¹H- and ¹³C NMR spectroscopy and are comprised in Fig. S32–S33.† Similar to the ⁱPr-MAC/MTC copolymers, the high amount of possible substructures prohibited complete signal assignment and hence also quantification of the degree of ring-opening. Branching from MTC units could be confirmed in all cases from the aforementioned signals around 1.20 ppm, but was not quantified due to signal overlap (Fig. S32†). Ester linkages within the copolymer were further confirmed by an exemplary ATR-FTIR spectrum of Et-MAC-25T (Fig. S34†). It showed a strong band 1734 cm⁻¹, showcasing the stretching vibration of the C=O bonds of ester units. The Alk-MAC/MTC content could still be determined as characteristic signals for MTC and the alkyl-substituents on the MACs could be attributed (Table 1 and Fig. S30–31†). Final molar ratios of Alk-MACs/MTC were higher than the feed ratio, highlighting the higher reactivity of all investigated Alk-MACs compared to MTC, again suggesting the formation of gradient copolymers. SEC measurements of all copolymers showed a somewhat constant M_n of 15.4 ± 4.5 kg mol⁻¹ (Fig. S35† and Table 1). Only ^sBu-MAC-25T showed a higher M_n of 49.1 kg mol⁻¹. While Et-MAC-25T and ⁿPr-MAC-25T copolymers had relatively reasonable dispersity values of 2.5 and 2.8, these values went up to 4.8 for ⁱPr-MAC-25T and rose further to 6–11 for the butyl derivatives. It indicated a lack of control over dispersity with increasing length and complexity of the side chain on the amine. A suspected reason for this could be the additional branching sites on the side chain, especially close to the amines. These new propagation centres could then themselves host new polymer chains with branches from all different branching points, leading to a plethora of possible sub-structures and also increased dispersities.

All copolymers contained the aimed-for amount of Alk-MAC and pH-responsiveness was then characterised by dissolving each copolymer in a HCl solution, titrating it from pH 2 towards pH 10. This process deprotonated the ammonium units and gave tertiary amines, changing the solubility of the copolymers from hydrophilic (transparent solution) to hydrophobic (turbid solution) (Fig. 3a). Turbidity pointed towards aggregation and DLS measurement of ⁱPr-MAC-25T/HCl at different pH values confirmed this (Fig. S46†). Volume size distribution traces showed significant increases in hydrodynamic size and polydispersity index (PDI) upon deprotonation, proofing the formation of agglomerates during this process. Turbidity measurements then showed a reversible deprotonation–protonation and hence transparent-turbidity switch of ⁱPr-MAC-25T over three cycles between pH 3 and 10 (Fig. 3b). The titration plots of % transmittance against pH value were then measured at a wavelength of 500 nm and then fitted with a Boltzmann equation (in Fig. S36–45†). When transmittance decreased to 50% of the original value, this pH was defined as cloud point. Copolymer solutions of all substituents were measured as triplicates and first evaluated separately in order to get mean values and error margins (Fig. S36–45† left side, values in Table 1). In a second instance, the values of all three measurements were fitted together in a single Boltzmann fit in order to compare different substituents against each other (Fig. S36–45† right side, comparison in Fig. 3c–e). Within these studies, M_n values were within reason and dispersity values were found to have no measurable impact on this value. Subsequently, impact on the was analysed for (i) amount of Alk-MAC in the polymer, (ii) length of Alk-MAC substituent, (iii) bulkiness of Alk-MAC substituent. For the first category, copolymers with different amounts of ⁱPr-MAC were compared (Fig. 3c). Within this series, the dropped from 7.1 for 34% of ⁱPr-MAC (ⁱPr-MAC-25T) to 6.6 for 69% of ⁱPr-MAC (ⁱPr-MAC-50T). Only the of 6.7 for 80% ⁱPr-MAC-50U does not perfectly fit this trend, but this could be explained with increasing structural irregularities following the low MTC content. In our previous research, the ⁱPr-MAC homopolymer exhibited a pK_a value around 6.0.⁶ All cloud-point values are within the same region, but cannot be compared directly following the different measurement principles. Owing to the higher hydrophilicity of MTC, all values were considerably higher than the one recorded for the ⁱPr-MAC/DMMDO copolymer (found at pH 3.5),^6,13,14 meeting a major goal of this study.

Fig. 3 (a) Transparent copolymer solutions became turbid when the pH value increased. (b) Reversible deprotonation–protonation process of ⁱPr-MAC-25T copolymer solution detected by UV-Vis measurements over three cycles. (c)–(e) Evolution of % transmittance of copolymers solution over pH, showing dependencies of (c) amount of incorporated ⁱPr-MAC, (d) bulkiness of the butyl substitute (ⁿBu, ⁱBu, ^sBu, ^tBu), (e) length of the n-alkyl substituent (Et, ⁿPr, ⁿBu). The values shown in the figure caption were obtained from the single Boltzmann fit of all three measurements of respective copolymer, and therefore were slightly different from the average value listed in the Table 1 (for a comparison of both values, see Table S13 in the ESI†).

For all Bu-MAC copolymers (Fig. 3d), the MAC content ranged from 30% in ⁱBu-MAC-25T to 50% in ^tBu-MAC-25T, being in a comparable range. The bulkiest residue of ^tBu-MAC-25T exhibited the highest of 7.2 despite having the highest amount of Bu-MAC units. All three other Bu-MACs were relatively close to each other with values of 6.4, 6.5 and 6.7 for ⁱBu-MAC-25T, ^sBu-MAC-25T and ⁿBu-MAC-25T, respectively. This implicated that the steric hindrance of the ^tBu residue caused this higher value.

Copolymers with linear substituents showed that Et-MAC-25T exhibited a significantly broader reduction region of % transmittance than ⁿPr-MAC-25T and ⁿBu-MAC-25T. As molecular weights remained within reason, this change could be attributed to the length of the side chain. Corresponding values increased from 5.7 for Et-MAC-25T over 6.1 for ⁿPr-MAC-25T to 6.7 for ⁿBu-MAC-25T. It can hence be concluded that longer alkyl substituents promoted a higher value. This could again be attributed to the increasing steric hindrance of a larger substituent, which goes in-line with the shift of the values observed within the series of butyl-substituents.

Selected copolymers were then subjected to hydrolytic degradation under accelerated condition. The copolymers were dissolved in THF/MeOH (67/33 vol%) with 5 wt% KOH and kept stirring at room temperature for 24 h to ensure complete degradation. Degradation was then proven in multiple ways: (i) ¹H NMR measured in CDCl₃ and D₂O: signals around 4.2 ppm ascribed to the protons adjacent to ester bond of P(MTC) disappeared completely after degradation (Fig. S47 and S49†). (ii) ¹³C NMR: signals around 174 ppm (Fig. S48†), related to carbon atoms in ester bond, also disappeared. (iii) SEC: similar to other RROP studies,^13,29 chromatograms before and after degradation showed a significant shift towards higher retention time after degradation for 24 h (Fig. 4 and Fig. S50†). Altogether, this confirmed the degradation of the copolymers.

Fig. 4 SEC analysis of copolymers before (solid lines) and after hydrolytic degradation (dashed lines) under accelerated conditions. Light scattering signals of (a) ⁱPr-MAC-25T, (b) ⁿPr-MAC-25T and (c) ^tBu-MAC-25T are depicted.

Conclusions

We hence successfully synthesized a series of alkyl-substituted Alk-MACs and provided their calculated molecular structures. Copolymerisation of all Alk-MACs with MTC yielded pH-responsive polyesters best under thermal conditions. The cloud point values varied from 5.7 to 7.2 with all the amount, length and bulkiness of the alkyl residue impacting the cloud points. Degradation of selected polyesters could then be proven by the disappearance of characteristic ¹H and ¹³C NMR signals as well as characteristic shifts in the SEC chromatograms. All presented Alk-MACs are thus a valuable addition to heteroatom-bearing CKA library and copolymerisation with MTC is a facile platform to gain pH-responsive and degradable polyesters.

Author contributions

Yiyi Deng: conceptualisation, formal analysis, investigation, data curation, writing – original draft, writing – review and editing, visualisation, funding acquisition. Anais Frezel: validation, formal analysis, investigation, writing – review and editing, visualisation. Fabian Mehner: validation, formal analysis, investigation, writing – review and editing, visualisation. Peter Friedel: methodology, software, investigation, data curation, writing – review and editing, visualisation. Jens Gaitzsch: conceptualisation, methodology, formal analysis, resources, data curation, writing – original draft, writing – review and editing, supervision, project administration, funding acquisition.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

The manuscript was written by contributions from all authors and all authors gave their approval for the final version. We would like to thank Alissa Seifert for the SEC measurements and Albena Lederer for helpful discussions of the results. The authors thank the Chinese Scholarship Council for funding Y. D., the Evonik Foundation for funding F. M. and the Deutsche Forschungsgemeinschaft (GA2051/7-1) for financial support. We thank the Howdle group at the University of Nottingham for granting access to their NMR facilities.

References

1. D. Goldberg, J. Environ. Polym. Degrad., 1995, 3, 61–67.
2. S. Agarwal, Polym. Chem., 2010, 1, 953–964.
3. J. N. Hoskins and S. M. Grayson, Polym. Chem., 2011, 2, 289–299.
4. W. J. Bailey, Z. Ni and S.-R. Wu, J. Polym. Sci., Part A: Polym. Chem., 1982, 20, 3021–3030.
5. T. Pesenti and J. Nicolas, ACS Macro Lett., 2020, 9, 1812–1835.
6. J. Folini, C.-H. Huang, J. C. Anderson, W. P. Meier and J. Gaitzsch, Polym. Chem., 2019, 10, 5285–5288.
7. G. Hedir, C. Stubbs, P. Aston, A. P. Dove and M. I. Gibson, ACS Macro Lett., 2017, 6, 1404–1408.
8. J. Gaitzsch, P. C. Welsch, J. Folini, C.-A. Schoenenberger, J. C. Anderson and W. P. Meier, Eur. Polym. J., 2018, 101, 113–119.
9. M. R. Hill, T. Kubo, S. L. Goodrich, C. A. Figg and B. S. Sumerlin, Macromolecules, 2018, 51, 5079–5084.
10. J. Tran, E. Guégain, N. Ibrahim, S. Harrisson and J. Nicolas, Polym. Chem., 2016, 7, 4427–4435.
11. M. R. Hill, E. Guégain, J. Tran, C. A. Figg, A. C. Turner, J. Nicolas and B. S. Sumerlin, ACS Macro Lett., 2017, 6, 1071–1077.
12. F. Mehner, M. Geisler, K. Arnhold, H. Komber and J. Gaitzsch, ACS Appl. Polym. Mater., 2022, 4, 7891–7902.
13. T. Pesenti, E. Gillon, S. Ishii, S. Messaoudi, Y. Guillaneuf, A. Imberty and J. Nicolas, Biomacromolecules, 2023, 24, 991–1002.
14. J. Undin, P. Plikk, A. Finne-Wistrand and A.-C. Albertsson, J. Polym. Sci., Part A: Polym. Chem., 2010, 48, 4965–4973.
15. Y. Deng, F. Mehner and J. Gaitzsch, Macromol. Rapid Commun., 2023, 44, 2200941.
16. A. Tardy, J. Nicolas, D. Gigmes, C. Lefay and Y. Guillaneuf, Chem. Rev., 2017, 117, 1319–1406.
17. V. Delplace and J. Nicolas, Nat. Chem., 2015, 7, 771–784.
18. A. Tardy, J.-C. Honoré, J. Tran, D. Siri, V. Delplace, I. Bataille, D. Letourneur, J. Perrier, C. Nicoletti, M. Maresca, C. Lefay, D. Gigmes, J. Nicolas and Y. Guillaneuf, Angew. Chem., Int. Ed., 2017, 56, 16515–16520.
19. Y. Zhang, M. Zheng, T. Kissel and S. Agarwal, Biomacromolecules, 2012, 13, 313–322.
20. T. Pesenti, C. Zhu, N. Gonzalez-Martinez, R. M. F. Tomás, M. I. Gibson and J. Nicolas, ACS Macro Lett., 2022, 11, 889–894.
21. J. Tran, T. Pesenti, J. Cressonnier, C. Lefay, D. Gigmes, Y. Guillaneuf and J. Nicolas, Biomacromolecules, 2019, 20, 305–317.
22. A. Bossion, C. Zhu, L. Guerassimoff, J. Mougin and J. Nicolas, Nat. Commun., 2022, 13, 2873.
23. U. Wais, L. R. Chennamaneni, P. Thoniyot, H. Zhang and A. W. Jackson, Polym. Chem., 2018, 9, 4824–4839.
24. J.-F. Lutz, J. Andrieu, S. Üzgün, C. Rudolph and S. Agarwal, Macromolecules, 2007, 40, 8540–8543.
25. S. Komatsu, T. Sato and A. Kikuchi, Polym. J., 2021, 53, 731–739.
26. D. J. Siegwart, S. A. Bencherif, A. Srinivasan, J. O. Hollinger and K. Matyjaszewski, J. Biomed. Mater. Res., Part A, 2008, 87, 345–358.
27. T.-A. Asoh, T. Nakajima, T. Matsuyama and A. Kikuchi, Langmuir, 2015, 31, 6879–6885.
28. A. W. Jackson, L. R. Chennamaneni and P. Thoniyot, Eur. Polym. J., 2020, 122, 109391.
29. C. Zhu, S. Denis and J. Nicolas, Chem. Mater., 2022, 34, 1875–1888.
30. C. Zhu and J. Nicolas, Polym. Chem., 2021, 12, 594–607.
31. E. Guégain, C. Zhu, E. Giovanardi and J. Nicolas, Macromolecules, 2019, 52, 3612–3624.
32. Y. Hiraguri, K. Katase and Y. Tokiwa, J. Macromol. Sci., Part A: Pure Appl. Chem., 2006, 43, 1021–1027.
33. M. Carlotti, O. Tricinci and V. Mattoli, Adv. Mater. Technol., 2022, 7, 2101590.
34. X. Ai, J. Pan, Q. Xie, C. Ma and G. Zhang, Polym. Chem., 2021, 12, 4524–4531.
35. Q. Xie, Q. Xie, J. Pan, C. Ma and G. Zhang, ACS Appl. Mater. Interfaces, 2018, 10, 11213–11220.
36. D. Gigmes, P. H. M. Van Steenberge, D. Siri, D. R. D'hooge, Y. Guillaneuf and C. Lefay, Macromol. Rapid Commun., 2018, 39, 1800193.
37. A. Tardy, N. Gil, C. M. Plummer, D. Siri, D. Gigmes, C. Lefay and Y. Guillaneuf, Angew. Chem., Int. Ed., 2020, 59, 14517–14526.
38. J. Folini, W. Murad, F. Mehner, W. Meier and J. Gaitzsch, Eur. Polym. J., 2020, 134, 109851.
39. J. Gaitzsch, X. Huang and B. Voit, Chem. Rev., 2016, 116, 1053–1093.
40. M. W. Schmidt, K. K. Baldridge, J. A. Boatz, S. T. Elbert, M. S. Gordon, J. H. Jensen, S. Koseki, N. Matsunaga, K. A. Nguyen, S. Su, T. L. Windus, M. Dupuis and J. A. Montgomery Jr., J. Comput. Chem., 1993, 14, 1347–1363.
41. F. Mehner, T. Meissner, A. Seifert, A. Lederer and J. Gaitzsch, J. Polym. Sci., 2023, 61, 1882–1892.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d3py00929g

---