Recyclable and Degradable Ionic-Substituted Long-Chain Polyesters

Ionic groups can endow apolar polymers like polyethylene with desirable traits like adhesion with polar compounds. While ethylene copolymers provide a wide range of tunability via the carboxylate content and neutralization with different cations, they lack degradability or suitability for chemical recycling due to their all-carbon backbones. Here, we report ion-containing long-chain polyesters with low amounts of ionic groups (Mn = 50–60 kg/mol, <0.5 mol % of ionic monomers) which can be synthesized from plant oils and exhibit HDPE-like character in their structural and mechanical properties. In the sulfonic acid as well as neutralized sulfonate-containing polyesters, the nature of the cation counterions (Mg2+, Ca2+, and Zn2+) significantly impacts the mechanical properties and melt rheology. Acid-containing polyesters exhibit a relatively high capability to absorb water and are susceptible to abiotic degradation. Enhanced surface wettability is reflected by facilitation of printing on films of these polymers. Depolymerization by methanolysis to afford the neat long-chain monomers demonstrates the suitability for chemical recycling. The surface properties of the neutralized sulfonate-containing polyesters are enhanced, showing a higher adsorption capability. Our findings allow for tuning the properties of recyclable polyethylene-like polymers and widen the scope of these promising materials.

Iso-propanol and acetone used for the precipitation and washing procedure of the polymers were used in technical grade.
Deuterated NMR solvents were supplied by Eurisotop.
The characterization of the soluble intermediate products by NMR spectroscopy was performed in chloroform-d1, 1,1,2,2-tetrachloroethane-d2 or dimethylsulfoxide-d6 as solvent at 25 °C or 110 °C, respectively. A Bruker Avance 400 spectrometer was used. 1 H chemical shifts were referenced to the residual solvent signal (85%). For spectra evaluation, MestReNova software version 12.0.4 was used. Integration of signals was conducted with the manual multiplet analysis tool.
Molecular weights of the polymers containing metal counterions (PEx.x-SO3M with M = Na, Mg, Ca, Zn) were determined by gel permeation chromatography on a SECcurity 2 GPC System from PSS with chloroform as solvent at 35 °C and a flow rate of 1 mL/min, using a PSS SDV linear M column and a refractive index detector. The determination of the molecular weight was performed with a linear calibration against narrow polystyrene standards, obtained from Varian. Analysis was carried out using WinGPC UniChrom software (version 8.30).
For differential scanning calorimetry, a Netsch DSC 204 F1 was used. A heating rate of 10 K/min was employed. All data reported are from the second heating cycles.
Thermogravimetric analysis was performed on a Netzsch STA 429 F3 Jupiter. All measurements were performed with 80 mL/min flow rate and a heating rate of 10 K/min. For measurements under air, a synthetic 80:20 mixture of N2:O2 was used. Samples were dried in a vacuum drying oven for at least 48 h (5 mbar, 50 °C) prior to measurements. X-ray scattering measurements were performed with a Bruker D8 Avance X-ray diffractometer with scintillation counter, Bruker D8 Discover. As a Cu-Kα source, a Bruker IµS Diamond was S4 used. As a detector, a Vantec-500 or a Lynxeye XE-T was used. Measured samples were taken from injection molding.
For tensile testing, polymers were processed in a micro compounder (Xplore MC 5). The melt was filled into a micro-injection-moulder (Xplore IM 5.5). The temperature of the cylinder and mould was varied (detailed in section 2.5). An injection pressure of 16 bar (10 s) and 12 bar (15 s) was applied. Tensile tests were performed on a Zwick/005/1446 Retroline tC II instrument, with a crosshead speed of 5 mm/min (ISO 527-2, type 5A). To determine the Young's modulus, a crosshead speed of 1 mm/min was applied. All samples were stored at room temperature for 24 h before tensile tests were performed. For data evaluation, Zwick Roell testXpert software version 11.0 was used.
To determine the water contact angle, samples were taken from injection moulded specimens and washed with iso-propanol to remove impurities from the surface. They were measured after storage in a vacuum drying oven for 3 days (50 °C, 50 mbar). The contact angle of a sessile drop of Milli-Q water was determined with a Krüss DSA25 drop shape analyzer using KRÜSS ADVANCE software for data evaluation (version 1.8.0.4).
Rheological data was acquired on an ARES-G2 rheometer (TA instruments) using either 13 mm or 8 mm parallel plates made of Invar. All measurements were performed with an ARES-G2 forced convection oven and flushed with nitrogen during the measurements. Oscillation strain sweeps (γ0 = 0.1 -100%) at an angular frequency of 100 rad/s were performed at 140 °C to determine the linear viscoelastic regime, followed by oscillatory frequency sweeps at 160 and 180 °C and a temperature ramp. For the frequency sweeps, a strain amplitude of γ0 = 10 % was applied. For the temperature ramp, a rate of 10 K/min was applied from 80 to 180 °C with a strain amplitude of γ0 = 0.1 %. The master curves for PE18.18-SO3M derivatives were obtained by shifting G' (storage modulus) and G'' (loss modulus) to 180 °C as a reference temperature (by making use of the time-temperature superposition principle). For data acquisition and evaluation, TRIOS software (version 5.5.1.5) was used.
Infrared (IR) spectra were measured on a Perkin Elmer 100 FT-IR spectrometer with an attenuated total reflection (ATR) unit and analyzed with the software Perkin Elmer Inc.
For the water uptake study of PE12.12-SO3H and PE18.18-SO3H, injection moulded specimens were immersed in MilliQ water (10 mL). They were weighed on a Sartorius ME36S balance (error: 1 µg). Samples were patted dry, weighed and re-immersed in water between individual measurements. This was repeated until three values were obtained with a deviation of less than S5 Elemental analysis of polymers PE18.18-SO3M was carried out by MIKROLAB Kolbe.
For printing on films, an ink-jet printer Canon Pixma TS705 with ink cartridges IC-Office XXL was used. Only the blue color of the cartridge was used, the other printer cartridges were masked to ensure no leakage of their color and to prevent color mixing. The printing was performed on films which were produced with a micro-cast film line. The thicknesses of the films ranged from 50 to 100 µm. The respective films were attached to a Sellotape line on a sheet of paper before introduction to the printer. The logo of the University of Konstanz was printed on the film in blue color. After printing, the films were dried for 24 hours in a drying cabinet at 50 °C.
A microfiber cloth was used to wipe the film with even pressure. Similar pressure was applied on the samples to wipe away the design on the films.

Determination of DPn and Mn
The degree of polymerization (DPn) and the number weighted molecular weight (Mn) of the PE12.12-SO3H-x and PE18.18-SO3H-x materials were determined by end group analysis from 1 H NMR spectra.
A signal of the backbone of the oligomethylene segment (H-4 and H-7, multiplet at 4.27 -3.87 ppm) was integrated and referenced to 100H. The end group signals for the hydroxymethyl group (E1, 2H, triplet at 3.63 ppm), the methyl ester (E3 and E3', 3H, singlet at 3.68 ppm) and the acid end group (E2, 2H, triplet at 2.38 ppm) were integrated. In case of the acid end group, the integral cannot be taken directly due to a strong overlap of the signal with the backbone signal H-1. Therefore, only the most downfield shifted and best isolated line of the triplet was integrated and multiplied by a factor of 4. All end groups are assigned representatively for PE18.18-SO3H-1.0 in Figure S1.

Figure S1
Detailed section of 1 H NMR spectra (500 MHz, 323 K, C2D2Cl4) of PE18.18-SO3H-1.0 with end group assignment. Note that the repeat unit can also contain sulfosuccinate units instead of C18-diester units. The respective integrals and calculated values are listed in Table S1.

Estimation of the Sulfonic acid Content in PEx.x-SO3H
The amount of HMSS incorporated into the polymer was calculated from 1 H NMR spectra. It is defined as the molar percentage of sulfosuccinate containing repeat unit in the copolymer.
The molar input into the polymerizationreaction mixture is denoted in the polymer nomenclature (PE12.12-SO3H-1.0 and PE18.18-SO3H-x, respectively). Note that for the lowest amount of HMSS input (PE18.18-SO3H-0.5), the estimation of the sulfonic acid content was impeded by the low signal-to-noise ratio.

Elemental Analysis
Elemental analysis was carried out by MikroLab Kolbe. The samples were dried for 48 h at 50 °C prior to the measurements. Two independent measurements were taken, of which the medium value is listed in Table S2.

TOCSY Results
Since the Brønstedt acid catalyst (HMSS) can catalyze both, esterification and transesterification, the methyl ester end group can be found on either the HMSS unit or the C18diacid after the polycondensation procedure (or C12-diacid, respectively). To further investigate the nature of incorporation of the sulfonic-substituted repeat units into the polymer chain vs.
endgroups, 1 H TOCSY experiments were performed to show that the 1 H signals originating from the sulfonic acid containing unit are not correlating with the ester end group at 3.67 ppm (see Figure S2).
This experiment shows a correlation of the methylene group (H-5) next to the sulfonic acid containing carbon with the methylene groups in the backbone of the C18-diol (H-3). Also, a clear response from the methylene group of the reacted ester is obtained (H-7). This indicates S9 that the chosen measurement conditions are sufficient to capture correlations over the ester groups. Regarding the end groups, a very low response from alcohol end-group E2 is obtained, but no clear response from E3 or E3'. This result supports the assumption that transesterification has occurred to a major extent and that the methyl ester end group is mainly attached to the C18-diacid, which is present to a much larger extent than sulfosuccinic acid. This finding underlines the successful synthesis of sulfonic acid containing all-aliphatic polyesters. A statistical distribution of sulfonic acid containing repeat units in the polyester chains can be assumed from this data.

Simulation of Sulfonate Distribution
To estimate the amount of sulfonate units per polymer chains, a statistical simulation approach was used as previously established by Wu et al. 3 , and applied by Odenwald et al. 4 to S10 acrylate-containing vitrimers. The same calculation method and script as described by Odenwald et al. was used herein.
A Schulz-Flory distribution is assumed in these calculations. Note that for simplicity, a Stirling approximation for large factorials was implemented to limit the amount of numerical calculations.
Considering In case of PE18.18-SO3H-0.8, ~ 62 % of the polymer chains are unfunctionalized. As this material is the basis for the PE18.18-SO3M materials, an alternative calculation under consideration of the end group enrichment of the sulfosuccinate unit was performed.

S11
Note that a full transesterification is assumed for the following calculations, meaning that the methyl ester end groups detected in the 1 H NMR spectra are placed to 99.2 % at the C18-dioate and 0.8 % are placed at the sulfosuccinate unit. This assumption is supported by the TOCSY results discussed above (section 1.5).
For this purpose, a detailed analysis of the ester groups from 1 H NMR spectra is required (see  From these incorporations (accounting for a total of 71% incorporation of the sulfosuccinate ester groups), and the total amounts of acid and ester end groups, the probabilities to obtain a sulfonate ester end group was calculated. A value of 6.7 % of the ester and acid end groups S12 were found to derive from the sulfosuccinate. Upon consideration of the probability to find an acid or ester end group rather than an alcohol end group, it was found that 3.7 % of the total amount of end groups are sulfosuccinate end groups.
From this result, the probabilities of zero, one or two sulfosuccinate end groups were calculated It can be seen that the differences between an entirely statistical incorporation and an end group enrichment is marginal.

Figure S5
Probability (p) of N sulfonate groups per polymer chain with and without consideration of end group enrichment.

Recycling to Monomer
For the recycling to monomer of PE18.18-SO3H-1.0, a stainless-steel reactor with a glass inlet (total volume of 20 mL) was charged with 200 mg of injection moulded specimen and methanol S13 (8 mL, > 99.8 % purity) as well as a stir bar. The reaction mixture was heated to 150 °C for 6 days and stirred at 300 rpm, upon which a pressure of 12 bar developed. After cooling to room temperature without further stirring, a crystallized solid was obtained. The supernatant was removed, and methanol (5 mL) was added. The mixture was heated to reflux until a clear solution was obtained and subsequently was cooled to 4 °C. The crystallized solid was centrifuged off, washed with methanol and dried. A 1:0.99 mixture of C18-diol:C18-diester was obtained according to the 1 H NMR spectrum (cf. section 3.11, Figure S56) in 80 % yield.  Table S3). The total amount of diester/diacid (HMSS and 1,18-octadecanedioic acid, or 1,12-dodecanedioic acid, respectively) accounts for one equivalent with respect to the used 1,18-dodecanediol (or 1,12-dodecanediol, respectively).  The same procedure was followed for polycondensation to PE12.12-SO3H-1.0, with 1,12-dodecanediol (13 g, 64.26 mmol, 1.00 eq.), HMSS (0.01 eq.) and 1,12-dodecanedioic acid (0.99 eq.).

S15
For PE12.12 and PE18.18, the same procedure was applied with the use of 0.3 mol% of [Ti(O n Bu)4] as catalyst. The reaction time at a vacuum of 5 x 10 -3 mbar was 12 hours.
All polymers containing free sulfonic acid were stored in a glovebox under a nitrogen atmosphere.
Assignment of 1 H NMR signals according to Figure S30 - Figure S37.
Assignment of 1 H NMR signals according to Figure S38 - Figure S42.

Processing Conditions
All polymers were processed in a micro-compounder to homogenize the polymer melt and were subsequently injection-moulded into test specimens for tensile testing (ISO 527-2, type 5A). At least five test specimens were prepared for each polymer. The temperature of the microcompounder (Tmelt) and of the injection mould (Tmould) were set to different values depending on the nature of the polymer. An overview of the processing conditions is listed in Table S4.
The stirring speed was set to 10 rpm and the mixing time before the first extrusion to homogenize the melt was 10 min in all cases. S18     Figure S8 ATR-IR spectra of PE12.12, PE12.12-SO3H-1.0 and PE12.12-SO3M. Left graph shows full range; right graph shows the grey marked carbonyl region.

DSC and TGA Thermograms
Figure S12 DSC thermograms of PE12.12 and PE12.12-SO3H-1.0 upon heating (left) and cooling (right). Data from the second cycle with a heating/cooling rate of 10 K/min are displayed. Data was shifted vertically for clarity. Figure S13 DSC thermograms of PE18.18 and PE18.18-SO3H upon heating (left) and cooling (right). Data from the second cycle with a heating/cooling rate of 10 K/min are displayed. Data was shifted vertically for clarity. Figure S14 DSC thermograms of PE12.12 and PE12.12-SO3M upon heating (left) and cooling (right). Data from the second cycle with a heating/cooling rate of 10 K/min are displayed. Data was shifted vertically for clarity.

Figure S24
Water contact angle of PE18.18-SO3H as well as the non-ion containing reference polyester PE18.18. Error bars represent standard deviations calculated from three different droplets (six angles). Samples were surface cleaned with i PrOH and dried for 48 h prior to the measurements. Images show representative measurements of each polymer from the series PE18.18-SO3H.

Figure S25
Water contact angle of PE12.12-SO3M and PE18.18-SO3M as well as their nonion containing reference polyesters, PE12.12 and PE18.18. Error bars represent standard deviations calculated from three different droplets (six angles). Samples were surface cleaned with iPrOH and dried for 48 h prior to the measurements. Images show representative measurements of each polymer from the series PE18.18-SO3M.

Ink Adsorption on Film Surfaces
The results before and after wiping off the color on all PE18.18-SO3M films are depicted in Figure S26.
To test whether the observed effect originates from the excess stearate or from the incorporated ionic groups, PE12.12, PE12.12-SO3Mg and PE12.12-Mgstearate were compared ( Figure S27), as well as PE12.12, PE12.12-SO3Ca and PE12.12-Castearate ( Figure S28). A significant blurring effect in case of PE12.12 is visible, yet the PE12.12-SO3Mg shows a high persistency of the ink. However, the reference polyester PE12.12-Mgstearate, which also contains excess stearate but has no ionic groups incorporated in the polymer backbone, shows a significant blurring effect, comparable to that seen in PE12.12. Therefore, the improved adhesion capability of the investigated polyesters cannot be concluded from the incorporated stearate, but mainly derives from the incorporated ionic groups.

Figure S29
Left: Weight gain of PE12.12 and PE12.12-SO3H-1.0 after storage in water for 12 weeks. Duplicates were investigated. In case of PE12.12-SO3H-1.0, one of the samples embrittled significantly after 4 weeks and could not further be weighed but was incubated further. Right: Images of dried samples after 12 weeks.     Note that protons 5 located next to the sulfonate group are diastereotopic and that the sulfonate group can be oriented either as shown, or the repeat unit can be oppositely arranged.

Figure S39
Detail of the 1 H NMR spectrum (500 MHz, 323 K, C2D2Cl4) of PE12.12-SO3Mg with integrals of end groups as well as protons located close to the sulfonate group (* = 13 Ccoupled satellites). Assignment of signals according to Figure S38. Figure S40 1 H NMR spectrum (500 MHz, 323 K, C2D2Cl4) of PE12.12-SO3Ca. Note that protons 5 located next to the sulfonate group are diastereotopic and that the sulfonate group can be oriented either as shown, or the repeat unit can be oppositely arranged.

Figure S41
Detail of the 1 H NMR spectrum (500 MHz, 323 K, C2D2Cl4) of PE12.12-SO3Ca with integrals of end groups as well as protons located close to the sulfonate group (* = 13 Ccoupled satellites). Assignment of signals according to Figure S40. Figure S42 1 H NMR spectrum (500 MHz, 323 K, C2D2Cl4) of PE12.12-SO3Zn. Note that protons 5 located next to the sulfonate group are diastereotopic and that the sulfonate group can be oriented either as shown, or the repeat unit can be oppositely arranged.

Figure S43
Detail of the 1 H NMR spectrum (500 MHz, 323 K, C2D2Cl4) of PE12.12-SO3Zn with integrals of end groups as well as protons located close to the sulfonate group (* = 13 Ccoupled satellites). Assignment of signals according to Figure S42. Figure S44 1 H NMR spectrum (500 MHz, 323 K, C2D2Cl4) of PE18.18-SO3Mg. Note that protons 5 located next to the sulfonate group are diastereotopic and that the sulfonate group can be oriented either as shown, or the repeat unit can be oppositely arranged.

Figure S45
Detail of the 1 H NMR spectrum (500 MHz, 323 K, C2D2Cl4) of PE18.18-SO3Mg with integrals of end groups as well as protons located close to the sulfonate group (* = 13 Ccoupled satellites). Assignment of signals according to Figure S44. Figure S46 1 H NMR spectrum (500 MHz, 323 K, C2D2Cl4) of PE18.18-SO3Ca. Note that protons 5 located next to the sulfonate group are diastereotopic and that the sulfonate group can be oriented either as shown, or the repeat unit can be oppositely arranged.

Figure S47
Detail of the 1 H NMR spectrum (500 MHz, 323 K, C2D2Cl4) of PE18.18-SO3Ca with integrals of end groups as well as protons located close to the sulfonate group (* = 13 Ccoupled satellites). Assignment of signals according to Figure S46. Figure S48 1 H NMR spectrum (500 MHz, 323 K, C2D2Cl4) of PE18.18-SO3Zn. Note that protons 5 located next to the sulfonate group are diastereotopic and that the sulfonate group can be oriented either as shown, or the repeat unit can be oppositely arranged.