Bio‐based and Degradable Block Polyester Pressure‐Sensitive Adhesives

Abstract A new class of bio‐based fully degradable block polyesters are pressure‐sensitive adhesives. Bio‐derived monomers are efficiently polymerized to make block polyesters with controlled compositions. They show moderate to high peel adhesions (4–13 N cm−1) and controllable storage and loss moduli, and they are removed by adhesive failure. Their properties compare favorably with commercial adhesives or bio‐based polyester formulations but without the need for tackifier or additives.


General Procedure
All solvents and reagents were purchased and used as obtained from commercial sources (Sigma Aldrich) unless stated otherwise. The synthesis of [H2L] macrocyclic ligand, was carried out in air and performed according to literature procedures. [1] The synthesis of the catalyst, [LMgZn(C6F5)2], monomer purification and subsequent polymerizations were carried out under inert conditions using standard Schlenk line techniques and a nitrogen-filled glovebox.
Limonene oxide (LO) was purified by first stirring over NaH followed by addition of MeI and fractional distillation at 40 °C (1 mbar). This procedure was repeated twice more (three distillations in total) then kept under a nitrogen atmosphere. ε-decalactone (DL) had been dried over CaH2, distilled under reduced pressure twice and kept under a nitrogen atmosphere. 1,4-Benzenedimethanol (BDM) had been recrystallised from toluene three times and kept under nitrogen. The tricyclic anhydrides (TCAs) were synthesised and purified based on modified literature procedures. [2] Sublimation of the TCAs (1-3) was performed at least twice and column chromatography of TCA 4 was performed four times in order to enhance the purity of the TCA monomers.
NMR spectra were obtained using a Bruker AV 400 instrument. SEC data was obtained using a Shimadzu LC-20AD instrument with HPLC grade THF as the eluent flowing at 1.0 mL/min at 30 °C and monodisperse polystyrene standard was used for calibration. Elemental analysis was performed by Mr. Stephen Boyer at London Metropolitan University, North Campus, Holland Road, London, N7. The thermal properties were measured using DSC3+ (Mettler Toledo, Ltd). A sealed empty crucible was used as a reference, and the DSC was calibrated using indium. Samples were heated from 25 °C to 150 °C, at a rate of 10 °C/min, under N2 flow (80 ml/min), and were kept at 150 °C for 2 min to erase the thermal history. Subsequently, the samples were cooled to -80 °C, at a rate of 10 o C/min, and kept at -80 °C·min -1 for a further 2 mins, followed by a heating procedure from -80 °C to 150 °C, at a rate of 10 °C/min. Each sample was run for two heating−cooling cycles. The glass transition temperatures (Tg) reported are taken from the third heating cycle. TGA was measured using a TGA/DSC 1 system (Mettler-Toledo Ltd). Samples were heated from 25 °C to 500 °C, at a rate of 5 °C/min, under N2 flow (100 cm 3 /min). Viscoelastic properties were measured using an Anton Paar Physica MCR 301 rheometer. Temperature sweep tests were performed at a constant shear strain of 0.5% and a frequency of 1 Hz from 20 °C to 150 °C at 2 °C/min. Frequency sweep tests were performed at a constant shear strain of 0.5% from 0.01 Hz to 30 Hz at 25 °C. The 180 ° Peel test was performed according to ASTM D3000 standard testing methods using a Shimadzu EZ-LX Universal Testing Instrument at a peeling rate of 305 mm/min (ISO 29862:2018). The adhesive material was placed on a PET sheet (0.5 inch width) and adhered to a polished stainless steel test panel with a constant pressure provided by a 4.5 lb weight. The average peel force was collected and reported from three samples.

Synthesis of [LMgZn(C6F5)2] (1)
Under inert conditions, the [H2L] macrocyclic ligand (500 mg, 0.90 mmol) and magnesium bis(hexamethyldisilazide) (Mg(N(SiMe3)2)2)(312 mg, 0.90 mmol) were stirred in anhydrous THF (10 mL) at 25 °C for 1 h. A solution of bis(pentafluorophenyl) zinc (361 mg, 0.90mmol), in anhydrous THF (5 mL), was then added dropwise to the ligand/MgHMDS2 mixture which was stirred overnight at 25 °C to afford a cloudy orange solution. The solvent was filtered and washed with pentane (2 x 20 mL) to yield a white solid (0.72 mg, 81%). End group analysis by 31 P{ 1 H} NMR spectroscopy Literature procedure for hydroxyl end group analysis by 31 P { 1 H} NMR spectroscopy was followed. [3] A mixture of stock solution (40 µL), excess 2-chloro-4,4,5,5-tetramethyl dioxaphospholane (40 μL) and polymer sample (20 mg) was added to a NMR tube and shaken. The mixture was allowed to react for 6 h before measurement. The stock solution consists of Bisphenol A (400 mg) and of Cr(acac)3 (5.5 mg) in pyridine (10 mL). . After the desired reaction time, the solution became viscous and the mixture was exposed to air to quench the polymerization. The crude sample was analysed by 1 H NMR spectroscopy. SEC analysis (THF as eluent, 1 mL/min, 30 °C) of the crude polymer was carried out following removal of excess solvent under reduced pressure. The polymer was purified by precipitation of a DCM solution (<1 g/mL) into MeOH (50 mL) at -78 °C. 100/300). Thereafter, the mixture was exposed to air to quench the polymerization. The crude sample was analysed by 1 H NMR spectroscopy. SEC analysis (THF as eluent, 1 mL/min, 30 °C) of the crude polymer was carried out following removal of excess LO under reduced pressure. The polymer was purified by precipitation of a DCM solution (<1 g/mL) into MeOH (50 mL). Note: Interestingly, the methine proton in the LO unit on the backbone of the polyester was observed to be a multiplet from 4.78 ppm to 5.18 ppm which is a feature not observed in previous reports of polymers utilizing LO. [4] Nevertheless, HSQC NMR spectroscopy confirmed these peaks correlated to carbon atoms in very similar environments, thus suggesting these peaks are likely the due the same proton (Fig. S7).

Exemplar terpolymerization of DL, LO and TCA
Under anaerobic conditions, LO (3.3 mL, 20.6 mmol), 1 (10.0 mg, 0.01 mmol), BDM (5.6 mg, 0.04 mmol) and DL were added to a vial charged with a stirring bar and allowed to react at 60 °C for a desired amount of time. After an aliquot was taken, the selected TCA was added and the mixture was allowed to react further at 140 °C until complete conversion of the anhydride (see table for molar ratios). Thereafter, the mixture was exposed to air to quench the polymerization. 1 H NMR spectroscopy and SEC analysis (THF as eluent, 1 mL/min, 30 °C) were performed at key intervals. The polymer was purified by precipitation of a DCM solution (<1 g/mL) into MeOH (50 mL) three times.
Note: Mn, NMR and the Wt%Hard are calculated by examining the degree of polymerization (DP) based on a comparison between the initiator signals and the peaks for the PE and PDL blocks. For instance, for the triblock copolyester prepared using TCA 1 with 41 wt% hard block, the integration of the benzylic protons of the initiator (proton 'a' (7.34 ppm) in Fig. S12) is compared with the methylene protons on the anhydride unit (proton '20' (5.74 ppm) in Fig.  S12) and the methine proton on the lactone unit (proton '5' (4.85 ppm) in Fig. S12). A DP of 10 and 33 for the PE and PDL blocks are found, respectively. As each PE and PDL unit is 386.53 g mol -1 and 170.25 g mol -1 , this amounts to 37.9 kg mol -1 for the block copolyester. Similarly, the wt% is calculated by comparing the total molecular weight for the PE and the PDL and converting the ratio for PE into a percentage. The theoretical renewable content of the block copolymer is calculated by dividing the total mass of the renewable monomers (i.e. limonene oxide, ε-decalactone, α-phellandrene and citraconic anhydride) by the total mass of all the monomers used in the reaction.

Degradation experiments
A modified literature procedure was followed. [5] pTSA.H2O (10 mg, 0.05 mmol) was added to a solution of P3 (100 mg) in THF (3 mL) and heated with stirring to 60 °C in a vial.
Scheme S9. In situ activation of the catalyst from reaction between 1 and 1,4-Benzenedimethanol (BDM) Figure S1. 1 Figure S3. 1 H NMR spectrum of the polyester from the ROCOP of LO/TCA 1 in CDCl3. Note: The polyester LO methine proton was observed as a multiplet (4.78 -5.18 ppm), which is a feature different to previous polymers utilizing LO.
[4] The multiplet may have arisen due to a lack of regioselectivity in the ROCOP process. The HSQC NMR spectrum confirms these peaks correlate to carbon atoms in very similar environments, thus suggesting these peaks are likely the due the same proton (Fig. S7). As such, the multiplets are proposed to be due to regio-irregularity in the ROCOP process.                                   Polyester from the polycondensation of isosorbide, long chain dimer diol and dimer fatty acid Yes Yes Adhesive 0.5 -10 [16] Polyester partly derived from plant oils Polyester from polycondensation of dimer fatty acids in combination with a dimer fatty diol, butane diol or isosorbide; mixed with epoxy plant oil No Yes Adhesive 0.6 -6.25 [17] Poly(βhydroxyorganoate) based composites Mixture of various random copolymers of poly(βhydroxyorganoate)s with unsaturated side chains (in 40-70 wt%) Yes (10-60 wt%)