Chemoenzymatic Photoreforming: A Sustainable Approach for Solar Fuel Generation from Plastic Feedstocks

Plastic upcycling through catalytic transformations is an attractive concept to valorize waste, but the clean and energy-efficient production of high-value products from plastics remains challenging. Here, we introduce chemoenzymatic photoreforming as a process coupling enzymatic pretreatment and solar-driven reforming of polyester plastics under mild temperatures and pH to produce clean H2 and value-added chemicals. Chemoenzymatic photoreforming demonstrates versatility in upcycling polyester films and nanoplastics to produce H2 at high yields reaching ∼103–104 μmol gsub–1 and activities at >500 μmol gcat–1 h–1. Enzyme-treated plastics were also used as electron donors for photocatalytic CO2-to-syngas conversion with a phosphonated cobalt bis(terpyridine) catalyst immobilized on TiO2 nanoparticles (TiO2|CotpyP). Finally, techno-economic analyses reveal that the chemoenzymatic photoreforming approach has the potential to drastically reduce H2 production costs to levels comparable to market prices of H2 produced from fossil fuels while maintaining low CO2-equivalent emissions.


Supporting Figures
production from enzymatic pre-treatment.(a, b) Enzymatic depolymerization of PCL and PET amorphous polymers under the optimal temperatures of each enzyme, DuraPETase (37 ºC) 1 and LCC (65 ºC) 2 , respectively in carbonate buffer (100 mM, pH 8.5).(c) Monomer yields from enzymatic pre-treatment traced using HPLC-UV (HA from PCL; TPA from PET).The samples were taken after two days (marked in green) for the photoreforming experiments.Alkaline pretreatment with 2 M NaOH at identical temperatures to enzyme pre-treatment samples (i.e., at 37 ºC for PCL and 65 ºC for PET films) yield sub-10 μM monomer amounts (open circles).photocatalysts.While TiO2|Pt absorbs strongly in the UV region, the precious metal-free CNx|Ni2P has an absorption onset around ~460 nm.The FT-IR of TiO2|Pt shows Ti-O vibrational modes between 500-700 cm -1 while in the case of CNx|Ni2P, the vibrations appear at 804 cm -1 (corresponding to the heptazine core), between 1132 and 1411 cm -1 (-CN bending modes), and at ~ 2145 cm -1 (C=N stretch).The data for alkaline pre-treatment is adopted from previous reports. 3,5See Supporting Discussion and Table S10 for details.The recombinant protein LCC-pExp-Bla was expressed overnight at 20 °C and the cells lysed via emulsiflex.After centrifugation and removing cell debris, the supernatant was loaded onto a column containing nickel resin for IMAC purification.To the fraction containing the purified LCC-pExp-Bla recombinant protein (6), TEV protease was added for cleavage overnight.The material was submitted to another IMAC purification step to obtain LCC without fusion tag (8).The numbers correspond to 1: protein ladder (Thermo Fisher Scientific, catalogue number 26616), 2: soluble fraction after protein extraction, 3: insoluble fraction, 4: flow-through, 5 and 6: purification fractions containing 10 and 200 mM of imidazole, respectively, 7: LCC after TEV cleavage overnight, 8: purified LCC, 9: Bla fusion tag and TEV protease.

Techno-economic Analysis
The benefit of the enzyme pre-treatment process for upscaled H2 production and commercialisation is further realised with the help of techno-economic analyses.4] Both economic and environmental feasibility of the process was estimated using the metrics: H2 production cost (₤ kgH2 - 1 ), H2 production cost with TPA revenue (₤ kgH2 -1 ), carbon footprint (gCO2 MJH2 -1 ) and energy returned on energy invested (EROI). 3 consider a base case chemoenzymatic photoreforming plant (plant lifetime assumed to be 20 years) to process 3 tons of PET plastic waste in 60000 L of solution per day with 180 kg TiO2|Pt (1 wt% Pt loading) photocatalyst.This requires the plant to have 6000 m 2 of flat panel photoreactors with a 1 cm reactor depth. 3A reactor depth of 1 cm was chosen to maximise the light-catalystsubstrate interaction. 3Following pre-treatment overnight in a polypropylene tank, the solution is purged with N2 (11250 L day -1 ) and then pumped through the photoreactors.The solar intensity assumed is 1 Sun (100 mW cm -2 ) for 7.5 h day -1 . 3The H2 produced is collected, compressed and stored in a tank (700 bar).The TPA is precipitated from the residual photoreforming mixture by adding sulfuric acid, filtered and crystallised as a secondary product to add economic value to the overall process.
Two cases are considered for our comparative techno-economic analyses: H2 production from enzyme (LCC) pre-treated PET (our work) and a previously reported alkaline pre-treatment approach. 5For the enzyme pre-treatment approach, sequential mechanical processing and overnight enzymatic pre-treatment are considered.According to previous reports, 4 PET depolymerisation efficiency was assumed to be ~69% (gTPA gPET -1 ) overnight.The photocatalytic activities obtained experimentally (our work for enzymatic pre-treatment and previous report 5 for alkaline pre-treatment) were used to estimate the H2 production, assuming a conversion rate of 50% (molH2 molsub -1 ) by the 'base case' models applied for techno-economic analyses in previous reports. 3For the alkaline pretreatment, the base (NaOH) is assumed to be reused 15 times. 3The CO2 production is considered five times the H2 production for enzymatic pre-treatment, assuming that 50% of the CO2 evolved is trapped as small organics as suggested in a previous report for near-neutral media. 3For alkaline pretreatment, no CO2 is evolved as it is trapped as carbonate in the solution. 3The transportation costs are not taken into consideration. 3e production costs, carbon footprint and EROI are calculated using the following equations: Where, Cinv -capital investment cost (₤); Cop -daily operational costs (₤); Ccondaily consumables cost (₤); Hdaily H2 production (kg); toperation life of pilot plant (days).
The results obtained from our techno-economic analyses are shown in Figure S15 and listed in Tables S8-S10.

Major sources of deviations
4] Although detailed calculations and sensitivity analyses is beyond the scope of the current work, it is important to mention the major sources of deviations from the proposed model that may arise depending on the different parameters involved.Factors such as the choice of the (photo)catalyst material (and its efficiency), solar reforming configurations, enzymes utilised, feedstock scale, and pre-treatment method adopted will significantly alter the technoeconomic assessments.The deactivation of materials (e.g., enzymes, etc.), catalyst lifetime and duration of pilot plant operation will also play a vital role.Costs involving land acquisition, transportation, etc. which are not included in our simplified 'base case' also needs to be considered to provide a more holistic picture for future technoeconomic assessments.Other factors that may affect future estimates and commercial viability include sunlight intensities and geographical location of deployment, waste sourcing, competition, partnerships and customers identified, local and governmental regulations, carbon credits, etc.Therefore, all these factors need to be considered and evaluated to realize a viable solar-powered chemoenzymatic reforming plant.Table S1.Monomer yields after enzyme pre-treated plastic substrates in duplicates.Conditions: carbonate buffer (100 mM, pH 8.5) at a total volume of 500 μL.Yields of 6-hydroxyhexanoic acid (HA; from pre-treated PCL) and terephthalic acid (TPA; from pre-treated PET) were measured using analytical HPLC-UV.Note: DuraPETase incubated at 65 °C with both PCL and PET films yielded no detectable monomers using HPLC.'n.m.' indicates "not measured", 'n.q.' indicates "not quantifiable", 'σ' indicates "standard deviation".PET film (20 mg mL -1 ) Dura TiO2|Pt (2 mg mL -1 )
PET nano-plastics (~ 0. --formate, glycolate, oxalate 4.9, n.q., n.q., resp. This work a residence time (in flow) Table S6.Photoreforming of enzyme (LCC) pre-treated PET film (~20 mg mL -1 ) in a custom-made integrated system with TiO2|Pt panel (effective area 3.5 × 3.5 cm 2 ) under benign conditions.Conditions: carbonate buffer (pH ~6-8), AM 1.5G irradiation, 33 ºC.OP indicates "oxidation product(s)". Salt components for carbonate buffer (100 mM, pH 8.5).c Costs of LCC-production was estimated from the industrial production of cellulases through fermentation (case 2).30 Enzyme dosage level was estimated at 27 µg kgPET -1 .d The pilot plant would be paid a gate fee for taking plastic waste, hence the negative cost. 3 Table S9.Technoeconomic analyses for alkaline pre-treatment. Vaues used for assessing photoreforming costs, carbon footprint, EROI, and individual component contributions for alkaline pre-treatment process based on base case of previous reports.[3][4] Only the components/parameters which have been newly added or modified from that in the case of enzymatic pre-treatment are shown in the table (rest remaining the same). Te mechanical processing (required for greater enzyme activity), enzyme and buffer components are not considered for alkaline pre-treatment.Table S10. Thno-economic analyses results and comparison.Summary of parameters/estimates for alkaline pre-treatment used for techno-economic calculations (data adopted from the previous report 5 ) and estimates for our enzyme pre-treatment approach in accordance with base cases of previous reports with minor modifications.[3][4] Our base case Variable Unit Alkaline pre-treatment Figure S1.Monomer production from enzymatic pre-treatment.(a, b) Enzymatic depolymerization of PCL and PET amorphous polymers under the optimal temperatures of each enzyme, DuraPETase (37 ºC) 1 and LCC (65 ºC) 2 , respectively in carbonate buffer (100 mM, pH 8.5).(c) Monomer yields from enzymatic pre-treatment traced using HPLC-UV (HA from PCL; TPA from PET).The samples were taken after two days (marked in green) for the photoreforming experiments.Alkaline pretreatment with 2 M NaOH at identical temperatures to enzyme pre-treatment samples (i.e., at 37 ºC for PCL and 65 ºC for PET films) yield sub-10 μM monomer amounts (open circles).

Figure S4 .
Figure S4.XPS spectra for co-catalysts.(a) Deconvoluted XPS spectra for TiO2|Pt in the Pt 4f region.(b) Deconvoluted XPS spectra for CNx|Ni2P in the Ni 2p region.Traces of Ni (II) on the surface arises from NiO formed by surface aerial oxidation.

Figure S5. 1 H
Figure S5. 1 H NMR analysis of the PCL films after photoreforming.(a) Stacked 1 H NMR spectra of Dura-treated PCL film before (black) and after photoreforming with CNx|Ni2P (red) and TiO2|Pt (blue) photocatalysts.(b) Stacked 1 H NMR spectra of LCC-treated PCL film before (black) and after photoreforming with CNx|Ni2P (red) and TiO2|Pt (blue) photocatalysts.Conditions: Photocatalyst concentration: 2 mg mL -1 ; carbonate buffer (pH 6); AM 1.5G irradiation; 25 ºC; 24 hours; stirring.The panel on the left indicates the magnified region marked in the spectra.The pentanal reference 1 H NMR spectra is shown in green.The structures are shown in right with the corresponding marking of protons.

Figure S6 .
Figure S6.The oxidation schematic for the monomer 6-hydroxyhexanoic acid (derived from PCL plastic after enzymatic pre-treatment) during the photoreforming process.

Figure S10 .
Figure S10.Assembly of the integrated system for photoreforming.(a) Graphics of the individual parts of the photoreactor.(b) Digital images of the frontal part of the photoreactor.(c) Digital image of a typical TiO2|Pt panel.(d) A fully assembled reactor under operation.The PET film is placed in the buffer containing LCC. (e-i) Photographic images showing (e) the plastic film, (f) frosted glass substrate on which the catalyst is deposited as seen in (c), and (g-i) the reactor in three dimensions.The scale bars correspond to 3 cm.

Figure S11 .
Figure S11.(a, b) FESEM images of the PET film (a) before and (a) after 96 h of enzymatic (LCC) pre-treatment inside the integrated system with simultaneous photoreforming.Before placing the film in the solution, it had an unstructured planar surface as shown in (a).However, a clear disfiguration of the PET film surface characterized by inhomogeneous craters caused by the LCC-enzyme was observed after 96 h as shown in (b).

Figure S12. 1 H
Figure S12.1 H NMR analysis of the LCC-treated PET film after photoreforming in the integrated system. 1 H NMR spectra of LCC-treated PET film after photoreforming in the integrated system with a TiO2|Pt panel (effective area 3.5 × 3.5 cm 2 ).Conditions: carbonate buffer (pH 6-8); AM 1.5G irradiation; 33 ºC; 96 h; stirring.The panel on the left indicates the magnified region marked in the spectra.EG indicates 'ethylene glycol' and TPA indicates 'terephthalic acid'.

Figure S13 .
Figure S13.TPA removal protocol for photocatalytic CO2 reduction experiments.(a) LCC-pretreatment step and (b) steps for TPA removal from the pre-treated solution using simulated conditions.(c) The gas-phase IR spectra of the headspace after photocatalytic experiments with 13 CO2 purging at different stages (black: before TPA removal; red: TPA removal without neutralization; green: TPA removal + neutralization) of TPA removal process, showing the existence of both 13 CO and 12 CO.The presence of TPA (black) results in decreased activity (discussed in the main text) and inconclusive IR spectra.(d) The 1H-NMR of the solutions at different stages of TPA removal.(e-g) Photographs of the pre-treated solution at different stages of TPA removal: (e) without TPA removal, (f) TPA precipitation induced by acidic pH, and (g) filtered solution with pH-readjustment used for photocatalytic experiments.

Figure S15 .
Figure S15.Techno-economic analyses results.Comparison of technoeconomic feasibility of alkaline pre-treatment and enzymatic pre-treatment processes for a 'base case' photoreforming pilot plant using the established metrics[3][4] : (a) cost of hydrogen production and (b) carbon footprint.The data for alkaline pre-treatment is adopted from previous reports.3,5See Supporting Discussion and TableS10for details.

Figure S16 .
Figure S16.The amino acid sequence of LCC-pExp-Bla and DuraPETase construct.LCC was cloned into pExp-Bla vector containing 8x histag (in pink) and β-lactamase as a fusion tag (in cyan).LCC was cloned downstream of a TEV site (in red).The full-length recombinant protein has a molecular weight of 69.9 kDa.Dura was cloned into pHAT5 vector containing Strep-tag (in pink).

Figure S17 .
Figure S17.Expression and purification of DuraPETase construct.The recombinant protein DuraPETase was expressed overnight at 20 °C and the cells were lysed via emulsiflex.After centrifugation and removing cell debris, the DuraPETase (29 kDa) was purified using Strep-Tactin ® Sepharose ® resin according to the manufacturer's protocol.

Figure S18 .
Figure S18.Expression, purification and TEV cleavage of LCC-pExp-Bla construct.The recombinant protein LCC-pExp-Bla was expressed overnight at 20 °C and the cells lysed via emulsiflex.After centrifugation and removing cell debris, the supernatant was loaded onto a column containing nickel resin for IMAC purification.To the fraction containing the purified LCC-pExp-Bla recombinant protein (6), TEV protease was added for cleavage overnight.The material was submitted to another IMAC purification step to obtain LCC without fusion tag(8).The numbers correspond to 1: protein ladder (Thermo Fisher Scientific, catalogue number 26616), 2: soluble fraction after protein extraction, 3: insoluble fraction, 4: flow-through, 5 and 6: purification fractions containing 10 and 200 mM of imidazole, respectively, 7: LCC after TEV cleavage overnight, 8: purified LCC, 9: Bla fusion tag and TEV protease.
Enzymatic degradation of PET film.The degradation/hydrolysis of a PET film is shown in the presence of LCC enzyme under benign conditions.Conditions: LCC enzyme (1 μM), 65 ºC, pH ~6-8, no stirring.The change in the shape and transparency of the PET film is visible with time suggesting enzymatic attack.
a Substrates used are synthesised according to protocol detailed in the Methods section other than PET film, which was purchased from Goodfellow.

Table S4 .
EQY determination for photoreforming of enzyme-treated film substrates.Conditions: TiO2|Pt or CNx|Ni2P photocatalyst (2 mg mL ˗1 ) in a sealed quartz cuvette (with active irradiation area of 1 cm 2 ) with N2 purging.Samples irradiated with monochromatic light (λ = 360 nm for TiO2|Pt and λ = 400 nm for CNx|Ni2P, full width at half maximum: 5, intensities taken as average of intensities before and after each measurement); time interval for each experiment: 2 h; room temperature; stirring.

Table S5 .
Comparison of our work with other photoreforming processes reported with relevant polymeric substrates.Unless otherwise mentioned, the experiments were carried out at room temperature (25 ºC).AM 1.5G corresponds to an irradiation intensity of 1000 W m -2 .OP indicates "oxidation product(s)", n.m. indicates "not measured", resp.indicates "respectively", n.q.indicates "not quantifiable".

Table S8 .
[3][4]economic analyses for enzymatic pre-treatment.Values used for assessing photoreforming costs, carbon footprint and EROI, and individual component contributions to the 'base case' model[3][4]for our enzymatic pre-treatment process.