Work‐hardening Photopolymer from Renewable Photoactive 3,3’‐(2,5‐Furandiyl)bisacrylic Acid

Abstract The design of a photopolymer around a renewable furan‐derived chromophore is presented herein. An optimised semi‐continuous oxidation method using MnO2 affords 2,5‐diformylfuran from 5‐(hydroxymethyl)furfural in gram quantities, allowing the subsequent synthesis of 3,3’‐(2,5‐furandiyl)bisacrylic acid in good yield and excellent stereoselectivity. The photoactivity of the diester of this monomer is confirmed by reaction under UV irradiation, and the proposed [2+2] cycloaddition mechanism supported further by TD‐DFT calculations. Oligoesters of the photoreactive furan diacid with various aliphatic diols are prepared via chemo‐ and enzyme‐catalysed polycondensation. The latter enzyme‐catalysed (Candida antarctica lipase B) method results in the highest M n (3.6 kDa), suggesting milder conditions employed with this protocol minimised unwanted side reactions, including untimely [2+2] cycloadditions, whilst preserving the monomer's photoactivity and stereoisomerism. The photoreactive polyester is solvent cast into a film where subsequent initiator‐free UV curing leads to an impressive increase in the material stiffness, with work‐hardening characteristics observed during tensile strength testing.


Supplementary Figures
Supplementary Figure S1 Semi-continuous oxidation of HMF to DFF with 88% MnO2-packed HPLC column.

. 57
Chemicals, enzymes and UV-curing apparatus 5-(hydroxymethyl)furfural (HMF) was purchased from Fluorochem, manganese dioxide 88%, MnO2 99% (dried at 120 °C overnight in a vacuum oven and stored in a vacuum desiccator) and 1,8octanediol (ODO) were purchased from Acros Organics. Pyridine and piperidine were purchased from Alfa Aesar, 98% H2SO4, 37% HCl, K2CO3, NaCl, MgSO4, chloroform and methanol were purchased from Fischer Scientific. 3Å molecular sieves (activated at 350 °C overnight stored in 80 °C oven) were bought from Honeywell Fluka. Candida antarctica lipase B (CaLB) immobilized onto acrylic resin (iCaLB, product code: L4777), 1,4-butanediol (BDO), diphenyl ether (DPE), Zr(IV) isopropoxide 2-propanol complex, Zn(II) acetate, Ti(IV) tert-butoxide, malonic acid, 2-propanol, CDCl3 and all other chemicals and solvents were purchased from Sigma-Aldrich and used as received if not otherwise specified. 2,2,5,5-tetramethyloxolane (TMO) was prepared as previously described and stored over 3 Å molecular sieves. [1] iCaLB was lyophilized for 48 h and stored in a desiccator before use. The hydrolytic activity of this iCaLB formulation was previously determined using the tributyrin assay, resulting in an activity of 2200 U gdry −1 , further details on the data and protocal for this assay is available from the study by Pellis et al.. [2] UV-lamps used for curing were custom-built by the electronic workshop of the University of York. The 365 nm UV LED gen2 emitter, 1250 mW (max output), LZ1-00UV00 was purchased from LED engine and the 5 W, 253 nm fluorescent lamp (PLS5/TUV 2 Pin 5W Germicidal UVC) from Lamp Specs website. The 22 mm diameter biconvex quartz lens used to collimate the 235 nm beam was purchased from UQG optics.

Synthesis of diformylfuran (DFF, 15):
A spent HPLC column (25 mm length, 4 mm internal diameter) was emptied, cleaned then re-packed with 88% MnO2 (6.2 g) with vacuum connected to one end of the column to ensure effective packing. The column was then connected to a JASCO HPLC pump PU 980 and clamped in a custom-made heating block and heated to 75 ºC. Heating tape was applied to the tubing at the exit of the column to avoid clogging of the capillaries (ESI, Figure S1). Thereafter, neat TMO preheated to 75 ºC was passed through the column at a flow rate of 1 mL.min -1 . Once the first drop of TMO exited the column, a 1% w/v HMF solution in TMO (1 g HMF, 7.9 mmol, in 100 mL TMO) was preheated to 75 ºC and then passed through the system. The solution at the exit of the column was evaporated to afford 2,5diformylfuran (DFF) as white solid (0.771 g, 6.2 mmol, 78% yield). The evaporated TMO was recovered and reused after distillation through a Vigreux column and drying over MgSO4. 1 [3] Synthesis (2E,2'E)-3,3'-(2,5-furandiyl)bisacrylic acid, 16: An oven dried 2-necked 250 mL round bottom flask equipped with a waterless condenser was charged with 2,5-diformylfuran (12. 566 g, 101 mmol), malonic acid (42.146 g, 405 mmol), 150 mL of anhydrous pyridine (stored under 3 Å molecular sieves) and a magnetic stirrer flea. After complete dissolution, piperidine (1.724 g, 20.2 mmol, 0.2 eq.) was added and the system was heated to 85 °C overnight (16 h). Thereafter, the reaction was brought to reflux for 2h. After cooling to room temperature, the solution was poured in a 500 mL beaker and neutralised with careful addition of 37% conc. HCl (150 mL) and 1 M HCl (300 mL) until the pH dropped under 1. The reaction mixture was place on an ice bath and the resulting precipitate was collected via vacuum filtration over fritted funnel. The beige solid further was washed with cold 1 M HCl and dried at 80 ºC in a vacuum oven overnight to afford of 2,5-furan diacrylic acid (17.25 g, 82.9 mmol, 82% yield). 1  Synthesis of (2E,2'E)-3,3'-(2,5-furandiyl)bisacrylic acid dimethyl ester, 1: An oven-dried 250 mL round bottom flask equipped with a waterless condenser was charged with -2,5-furan diacrylic acid (8.144 g, 39.1 mmol) and 80 mL MeOH. Consequently, NaCl (0.984 g, 16.8 mmol, 0.43 eq.) was added to decrease the solubility of the diester in the methanolic solution and 98% H2SO4 (210 μL, 3.91 mmol, 0.1 eq.) was added. The suspension was refluxed for 7h and left to cool with stirring. The solution was further cooled with stirring to 0 °C with an ice bath and the crystalline material was collected by vacuum filtration. Recrystallisation from 2-propanol followed by drying under high-vacuum afforded monomer 1 as pale brown shiny flakes (7.377 g, 31.2 mmol, 82% yield). This monomer was protected from light as much as possible until further use. 1 [4] Chemo-catalytic polycondensation of 1 with 1,8-octanediol: In a custom-made polymerisation flask were added consecutively, the monomer 1 (1 g, 4.23 mmol), 1,8-octane diol (1.24 g, 8.47 mmol) and 5 mol% of the catalyst (K2CO3, Zr(IV) isopropoxide isopropanol complex, Zn(II) acetate or Ti(IV) tert-butoxide e.g. for Ti(IV) tert-butoxide: 72.04 mg, 35 mol% respective to titanium, 0.05 eq.) with a large egg shaped rare earth magnetic stirring bar. The vessel was surmounted with a short path distillation arm to continuously remove the produced methanol, connected to a Schlenk line and the system was purged and filled with Ar or N2 3 times before heating under Ar or N2 at 95 ºC overnight. High vacuum was then applied whilst maintaining the heating, for 4 hours to remove any excess diol and the crude material was analysed by GPC.

Enzymatic polymerizations in high boiling organic media
Based on previously published procedure used for alternative monomers enzymatic polycondensation reactions were carried out as follows. [5] 37 8 mmol of diester (0.2 M) and 8 mmol of the diol (0.2 M) (diester:diol ratio= 1:1) were added together with 4 mL of diphenyl ether (DPE) in a 25 mL round bottom flask. The mixture was then stirred at 85 ºC until complete dissolution of the monomers in the solvent. 10% w w -1 (calculated on the total amount of the monomers) of iCaLB was then added and the reaction was run for 6 h at 1000 mbar. The reaction system was covered with aluminium foil to avoid light-induced side reactions of the components. A vacuum of 20 mbar was subsequently applied for an additional 90 h while maintaining the reaction temperature at 85 ºC. Warm chloroform was added to the reaction mixture to solubilize the polymer product and the biocatalyst was filtered off. The chloroform was then removed under vacuum. The polymer-DPE mixture was subsequently crashed out in ice-cold methanol achieving precipitation of the products. Three methanol triturations were subsequently performed in order to remove the residual DPE. The reactions led to a light-yellow powdery polymerization product. All reactions were conducted in duplicates.

Enzymatic polymerizations in low boiling organic media
Enzymatic polycondensation reactions were conducted in similar manner to that previously described for other polyesters.46 Briefly, 8x10 -4 mol of diester (0.2 M) and 8x10 -4 mol of the aliphatic diol (0.2 M) (diester:diol ratio= 1:1) were added together with 4 mL of TMO as a low boiling, green, organic solvent in a 100-mL round bottom flask. The mixture was then stirred at 85 ºC until complete dissolution of the monomers in the solvent. 10% w/w (calculated on the total amount of the monomers) of iCaLB was then added and the reaction was run for 6 h at 1000 mbar. The reaction system was covered with aluminium foil to avoid light-induced side reactions of the components. A vacuum of 300 mbar was subsequently applied for an additional 90 h while maintaining the reaction temperature at 85 ºC. At the end of the reaction, the biocatalyst was filtered off and the TMO was then removed under vacuum. The reactions led to yellow powdery polymerization products.
ODO-PFAE film casting: The synthesized polymer, ODO-PFAE was dissolved in chloroform (10 mg mL -1 ) and the solution casted on a Kapton polyimide film placed inside a glass petri dish. The solvent was slowly removed at 21 °C keeping the casting system in the dark. The obtained thin film (thickness of 50 μm) was then peeled off the polyimide film and mechanically tested and cured as described below.

UV-irradiation of synthesised material
The beam of a custom-made LED torch or fluorescent lamp (ESI Figure S2 and S3) maximum emission at 365 nm, 1250 mW maximum flux output for the LED and 253 nm, 5W for the fluorescent lamp) was collimated with a biconvex plastic lens (for irradiation at 365 nm) and a biconvex quartz lens (for irradiation at 253 nm) on the sample layer (typically 100 mg neat for monomer 1 and ODO-PFAE 2.5 cm long, 0.5 cm wide strips for film, <1 mm thick, distance from lens ~5 cm for 365 nm lamp, <2 cm for 253 nm lamp) placed on a glass petri dish. The reaction was monitored by FT-IR spectroscopy and mixing with a spatula (for powdery/crystalline) or change of irradiated faces (for film strips) was done every 24 hours to assure homogeneous curing. Typically, UV-curing of a sample was stopped after 72 hours of continuous irradiation unless specified otherwise.

Solubility testing of the cured material
The cured film strips were accurately weighed (9-10 mg) and place in either THF-d8 or CDCl3 (800 L of each) and put on an automatic roller for 1 hour (RT), after which the strip was removed and NMR spectrum of the extract recorded following addition of dimethyl maleate (2 L) as a standard. Comparisons of integration of signals possibly of extracted polymer (CH2s of ester bond) with the standard allowed for determination of extent of extraction (see Figures S37-39). Separately, the recovered strips from above were dried thoroughly under high-vacuum and weighed again to assess via mass balance what proportion of the polymer was extracted into solution, no mass change observed using THF, while for CDCl3 the mass of strip reduced from 10.8 to 10.7 mg.
Nuclear Magnetic Resonance (NMR) Spectroscopy 1 H and 13 C-NMR spectroscopy analysis were performed on a JEOL JNM-ECS400A spectrometer at a frequency of 400 MHz for 1 H and 100 MHz for 13 C. CDCl3 was used as solvent if not otherwise specified.

Gel Permeation Chromatography (GPC)
Samples were dissolved in CHCl3 and filtered through a cotton filter prior to passing into a HPLC vial. Gel permeation chromatography was carried out at 30 °C on an Agilent Technologies HPLC System (Agilent Technologies 1260 Infinity) connected to a 17369 6.0 mm ID × 40 mm L HHR-H, 5 μm Guard column and a 18055 7.8 mm ID × 300 mm L GMHHR-N, 5 μm TSKgel liquid chromatography column (Tosoh Bioscience, Tessenderlo, Belgium) using 1 mL min -1 CHCl3 as mobile phase. An Agilent Technologies G1362A refractive index detector was employed for detection. The molecular weights of the polymers were calculated using linear polystyrene calibration standards 250-70000 Da (Sigma-Aldrich).

Matrix Assisted Laser Desorption Ionization (MALDI)
MALDI-TOF MS analysis were carried out by using a Bruker Solarix-XR FTICR mass spectrometer and the relative software package for the acquisition and the processing of the data. An acceleration voltage of 25 kV, using DCTB as matrix and KTFA as ionization agent were used. 10 μL of sample were mixed with 10 μL of matrix solution (40 mg mL -1 DCTB in CHCl3) and 3 μL of KTFA (5 mg mL -1 ). 0.3 μL of the mixture were applied on the plate and the measurement was conducted in positive mode with the detector set in reflector mode.

Differential Scanning Calorimetry (DSC)
DSC experiments were performed on a TA Instruments Q2000 DSC under an inert gas atmosphere (N2). Heating and cooling rates were set to 5 °C/min over the T range of -60-200 °C. Sample mass was of between 5-10 mg for all measured samples. The Tg values were calculated from the second heating scan.
Thermogravimetric analysis (TGA) TGA was performed on a PL Thermal Sciences STA 625 thermal analyser. ∼10 mg of accurately weighed sample in an aluminium sample cup was placed into the furnace with a N2 flow of 100 mL min −1 and heated from room temperature to 625 °C at a heating rate of 10 °C min -1 . From the TGA profiles the temperatures at 10% and 50% mass loss (Td10 and Td50) were subsequently determined.

Fourier Transformation Infrared Spectroscopy
Fourier Transformation Infrared Spectroscopy (FT-IR) analysis of the synthesized polymers was performed on a PerkinElmer 400 spectrometer using the attenuation total reflectance setting. The same pressure was applied on the outer surface of all analysed samples. A number of 16 scans were recorded using a 1 cm −1 resolution. All spectra were processed using the automated baseline correction and the data auto tune functions.

Computational Details
All the calculations were carried out in Gaussian16 (A.03) program package. [6] Optimizations and frequency calculations were performed without symmetry restrictions at wB97xD/6-31+G(d,p) level due to the good agreement with the experimental UV-VIS absorption of monomer 1. [7] Other functionals were also tested (ESI Table.S10). Solvation was introduced by the CPCM implicit solvent method (chloroform). [8] All the stationary points were characterized as minima or transition state (0 or 1 imaginary frequency) in the corresponding electronic state (ground or excited state) and thermochemistry corrections at standard conditions were added to obtain the corresponding free energies. Both potential energies and free energies (in parentheses) are included in the manuscript in kcal/mol.

Additional methods and information
MestReNova (version, 14.0.0, MestreLab Reasearch) was used for NMR spectra processing. Deconvolution was carried out using the software Global Spectral Deconvolution as described here: https://resources.mestrelab.com/gsd OriginLab2019 (V9.6.5.169, 2019b) was used to compute the full width at height maximum (FWHM) used to calculate the domain frame by the Scherrer equation: Where is the domain frame or crystallite size (in nm), K is the Scherrer constant taken as 0.9, the wavelength used for analysis, here 0.15406 nm, FWMH, the full width at height maximum and the peak position. The asymptotic exponential fitted curved for the conversion followed by FT-IR spectroscopy was also computed with OriginLab using the automatic aymptot1 exponential model available in the software.
XRD was performed using a Bruker AXS D8 Advance controlled by XRD Commander software, with scan type set at locked coupled, operating voltage of 40 kV (current, 40 mA), scan speed of 2 sec/step and the scan scope from 0 to 90 .
Single crystal XRD were recorded on an Oxford Diffraction SuperNova apparatus equipped with dual Mo & Cu sources and structure was resolved by Rachel R. Parker.
GC-FID was recorded on an Agilent GC7890B equipped with Rxi-5HT column The GC method was as followed: initial temperature 50 ºC, holding 0min, Ramp rate 30 ºC/min, Temperature final: 300 ºC, hold 5 min, split ratio 5:1, injector temperature 300 ºC. Blanks containing DCM and TMO was run prior analysis of the HMF-DFF mixture Porosimetry analyses were performed using a Micromeritics ASAP 2010. Samples were degased for at least 6 h at 130 °C prior to analysis. The Brunauer-Emmett-Teller (BET) theory was used to determine the surface area, and data processed using the MicroActive software provided with the instrument.
EDX-SEM data were recorded on a JEOL 7800F Prime SEM, equipped with a Schottky (fieldassisted) thermionic emitter.

Supplementary Figures
Supplementary Figure S1 Semi-continuous oxidation of HMF to DFF with 88% MnO2-packed HPLC column. A: Oxygen input (optional, 2 mL.min -1 ); B: 1% HMF solution in TMO C: JASCO PU-980 HPLC pump D: Heating tape controller E: Heating tape F. MnO2-packed HPLC column (4 mm internal diameter, 25 mm length) clamped in custom-made heating block G: DFF solution obtained at column exit.  The enzyme employed belongs to the serine hydrolase superfamily the mechanism of which has previously been investigated. [9] The conjugation of the monomer 1 is temporarily broken during the formation of a tetrahedral intermediate ( Figure S36, 17) with a serine residue present in CaLB's active site. The transesterification with the diols then occurs with attack of the diol on the acyl enzyme intermediate ( Figure S36, 18) which formed from 17 after release of methanol. The elongated product (20) is released from the second tetrahedral species formed ( Figure S36, 19), which closes the catalytic cycle by reforming the active site. As such the relatively low Mn obtained here is possibly due to the need for the highly stable extended conjugation of 1 to be temporarily partially broken for formation of intermediate 17. [9] The low Mn may also be due to the steric hindrance created by the elongated chains of the polymer. The longer bulky oligomers likely prevent the substrate from entering the CaLB's active site and thus limit the further growth of the polymer. It is in fact well known that fully aliphatic polyesters can reach far higher molecular weights compared to terephthalate, furan and pyridine/based diesters. [5,[10][11][12] Supplementary Figure S37