Plastics-to-syngas photocatalysed by Co–Ga2O3 nanosheets

ABSTRACT Plastics take hundreds of years to degrade naturally, while their chemical degradation typically requires high temperature and pressure. Here, we first utilize solar energy to realize the sustainable and efficient plastic-to-syngas conversion with the aid of water at ambient conditions. As an example, the commercial plastic bags could be efficiently photoconverted into renewable syngas by Co–Ga2O3 nanosheets, with hydrogen and carbon monoxide formation rates of 647.8 and 158.3 μmol g−1 h−1. In situ characterizations and labelling experiments unveil water is photoreduced into hydrogen, while non-recyclable plastics including polyethylene bags, polypropylene boxes and polyethylene terephthalate bottles are photodegraded into carbon dioxide, which is further selectively photoreduced into carbon monoxide. In-depth investigation illustrates that the efficiency of syngas production mainly depends on the carbon dioxide reduction process and hence photocatalysts of high carbon dioxide reduction activity should be designed to promote the efficiency of plastic-to-syngas conversion in the future. The concept for the photoreforming of non-recyclable plastics into renewable syngas helps to eradicate ‘white pollution’ and alleviate the energy crisis simultaneously.


EXPERIMENTAL SECTION Characterization
Transmission electron microscopy (TEM) images were obtained from a JEOL-2010 TEM. Highresolution TEM (HRTEM) images and the corresponding elemental mapping images were measured on a JEOL JEM-ARM200F TEM/STEM. The field-emission scanning electron microscopy (FE-SEM) images were performed by using a FEI Sirion-200 SEM. X-ray diffraction (XRD) patterns were recorded by a Philips X'Pert Pro Super diffractometer with Cu Kα radiation (λ = 1.54178 Å), in which the voltage was 40 kV, the electric current was 100 mA and the scan speed was 35°/min. Atomic force microscopy (AFM) was carried out on a Veeco DI Nano-scope multimode V system with a scan speed of 1.00 Hz. Xray photoelectron spectra (XPS) were measured on an ESCALAB MKII with Al Kα (hυ = 1486.6 eV) as the excitation source. The binding energies obtained in the XPS spectral analysis were corrected by referencing C 1s to 284.8 eV. UV−vis diffuse reflectance spectra were performed on a Perkin Elmer Lambda 950 UV−vis−NIR spectrophotometerand the UV−vis absorption spectrum was recorded on UV-2501PC/2550 (Shimadzu Corporation, Japan). Room temperature photoluminescence (PL) spectra were evaluated by a LabRAM HR Evolution Micro-Raman system, the excitation wavelength was 325 nm, and the PL spectra were accumulated for 30 s with an output power of ca. 1.7 mW. The CO2 temperature programmed desorption (TPD) measurements were carried out on AutoChem II 2920 with a temperature ramp rare of 10 °C/min. In situ FT-IR spectra were executed by a Thermo Scientific Nicolet iS50. In situ ESR spectra were obtained using a JEOL JESFA200 ESR spectrometer. Synchrotron radiation photoemission spectroscopy (SRPES) was performed at the Catalysis and Surface Science Endstation at the BL11U beamline in the National Synchrotron Radiation Laboratory (NSRL) in Hefei, China. The work function (Φ) is determined by the difference between the photon energy and the binding energy of the secondary cutoff edge, according to the formula of Φ = hυ -Ecutoff (hυ refers to the utilized photon energy of 40.00 eV and a sample bias of −5 V was applied to observe the secondary electron cutoff).

Photoconversion of Commercial Plastic Products and Reagent-Grade PE Powders
Commercial plastic products of PE plastic bags, PP plastic boxes and PET plastic bottles were initially crushed into fine powders (<5 mm) by a pulverizer (Benchen Science and Technology, Figures S10-S12).
Reagent-grade PE powders with the size of ca. 5 mm were purchased from Alfa Aesar (China) without any further purification. It should be mentioned that small plastics (≤5 mm) were commonly defined as the microplastics, which were particularly difficult to be recycled and may cause some unpredictable damages to the ecosystem. Then, a closed glass gas-circulation system (Labsolar-ⅢAG, Beijing Perfectlight Technology Co., Ltd) was used to conduct the photoconversion tests. In detail, 50 mg photocatalysts and 100 mg plastic products powders were initially dispersed in 100 mL deionized water under stirring. The dispersion was put into the reaction vessel, which was soaked in an outer vessel so that cooling water could circulate to keep the temperature of the reaction solution at 298 ± 0.2 K. The reaction system was vacuum-treated and then filled with high-purity air to reach an atmospheric pressure.
The light source for the photocatalysis was a 300 W Xe lamp (CEL-HXFUV300, Beijing Ceaulight Technology Co., Ltd., China) with a standard AM 1.5 filter to simulate one-sun irradiation, in which the outputting light density was about 100 mW/cm 2 . During the light irradiation, the evolved gas products were qualitatively and quantificationally analyzed by Agilent GC-7890B gas chromatograph. The liquid products were quantified by NMR (Bruker AVANCE AV III 400) spectroscopy, in which 0.4 mL supernate was mixed with 0.1 mL D2O (deuterated water) and 0.02 mL dimethyl sulfoxide (DMSO, Sigma, 99.99%) was added as an internal standard. The one-dimensional 1 H spectrum was measured with water suppression using a pre-saturation method. To analyze the content of CO2 dissolved in the solution, 2.0 mL supernate was taken immediately after the photocatalysis. Upon adding excess Ba(OH)2 powders to the solution, the white precipitate could be obtained. Taking the Co-Ga2O3 nanosheets as examples, after the following centrifugation and the gravimetric analysis for the white precipitate, the produced CO2 and CO dissolved in the solution after the photoconversion of pure PE or PP or PET could be calculated to be ca. 2.88, 2.53 and 1.65 mmol, respectively. And meanwhile, the amount of gas CO2 and CO after the photoconversion of pure PE or PP or PET detected by gas chromatograph was roughly 0.69, 0.64 and 0.43 mmol, respectively. That is to say, the total carbon content for the photoconversion of all these three plastics over both the two catalysts can exceed 90% (Table S5-6). In addition, a handful of microplastics with smaller size were also detected by the SEM images (Fig. S14), which might be produced during the photoconversion processes, conforming to the principle of carbon balance.

Detection of the Isotope Tracing Gas Products
Synchrotron-based vacuum ultraviolet photoionization mass spectrometry (SVUV-PIMS) was carried out at the combustion endstation at the BL03U beamline at the National Synchrotron Radiation Laboratory at Hefei, China. In detail, 50 mg Co-Ga2O3 nanosheets and 100 mg reagent-grade PE powders were initially dispersed in 100 mL deionized water under stirring. The reaction system was vacuumtreated three times to remove air completely, which was then pumped by high-purity N2 to reach an atmospheric pressure. For the 18 O or H2 18 O labelling experiments, a small amount of 18 O2 or 1 mL H2 18 O were added to the reaction system, and a certain amount of gas was taken immediately after 24 h irradiation and was analyzed at the photon energy of 14.5 eV. For the D2 labelling experiments, D2O were utilized to replace H2O.

In situ FTIR experiments
All FTIR spectra were recorded on Thermo Scientific Nicolet iS50. The spectra were displayed in absorbance units and acquired with a MCT detector and a resolution of 8 cm −1 , using 64 scans. The dome of the reaction cell had two KBr windows allowing IR transmission and a third window allowing transmission of irradiation introduced through a liquid light guide that connects to a 300 W Xe lamp. The Co-Ga2O3 nanosheets and Ga2O3 nanosheets powders were added to the reaction cell. After spraying some water on the powders, the reaction cell was sealed. Then, the reaction system was vacuum-treated three times to remove air completely, which was then pumped by high-purity CO2 to reach an atmospheric pressure. Next, the FTIR spectra were recorded as a function of time to investigate the dynamics of the reactant adsorption in the dark and desorption/conversion under irradiation.

In situ ESR Experiments
(1) Detection of superoxide radical anion (O2 ·-): 5 mg Co-Ga2O3 nanosheets and 15 mg reagent-grade PE powders were dispersed in 1 mL of methanol followed by addition of 20 μL DMPO methanol solution (100 mM). After 10 s irradiation, the mixed reaction solution was analyzed by in situ ESR spectra.
After 10 s irradiation, the mixed reaction solution was analyzed by in situ ESR spectra.

Detection of H2O2
2.0 mL reaction solution was taken immediately after 24 h photoconversion of reagent-grade PE powders. Then, 2 mL 1% o-tolidine in 0.1 M HCl was added to the above reaction solution. Subsequently, the solution was acidified with 1 M HCl (2 mL). The absorption spectrum of the solution was immediately recorded with a UV-vis spectrophotometer and the H2O2 has a characteristic absorption peak at ca. 436 nm.

Detection of H2O2 with o-tolidine
Hydrogen peroxide (H2O2) measurements were performed on an UV−vis−NIR spectrophotometer. 2.0 mL reaction solution was taken immediately after 5 h irradiation of pure PE over the Co-Ga2O3 nanosheets under simulated natural environments. Then, 2 mL 1% o-tolidine in 0.1 M HCl was added to the above reaction solution. Subsequently, the solution was acidified with 1 M HCl (2 mL). The absorption spectrum of the solution was immediately recorded with a UV-vis spectrophotometer. The H2O2 has a characteristic absorption peak at 436 nm.

CO2 Photoreduction Tests
A closed glass gas-circulation system (Labsolar-ⅢAG, Beijing Perfectlight Technology Co., Ltd) was used to conduct the photocatalytic CO2 tests. In detail, 50 mg Co-Ga2O3 nanosheets or Ga2O3 nanosheets were initially dispersed in 100 mL deionized water under stirring. The dispersion was put into the reaction vessel, which was soaked in an outer vessel so that cooling water could circulate to keep the temperature of the reaction solution at room temperature. Then, the reaction system was vacuum-treated three times to remove air completely, which was then pumped by high-purity CO2 ( 13 CO2 was used for the 13 C labelling experiments) to reach an atmospheric pressure. The light source for the photocatalysis was a 300 W Xe lamp (CEL-HXFUV300, Beijing Ceaulight Technology Co., Ltd., China) with a standard AM 1.5 filter, in which the outputting light density was about 100 mW/cm 2 .

Calculation Details
The first-principles calculations were performed with the Vienna ab initio simulation package. 1,2 The interaction between ions and valence electrons was described using projector augmented wave (PAW) potentials, and the exchange-correlation between electrons was treated through using the generalized gradient approximation (GGA) in the Perdew-Burke-Ernzerhof (PBE) form. 3 To achieve the accurate density of the electronic states, the plane wave cutoff energy was 480 eV, a gamma-centre 3 × 9 × 1 for sheet k-point mesh were used. Ionic relaxations were carried out under the conventional energy (10 -5 eV) and force (0.01 eV/Å) convergence criteria. All periodic slabs have a vacuum spacing of 15 Å to avoid the interaction between adjacent layers. The structure model of Co-Ga2O3 nanosheets and Ga2O3 nanosheets contain a supercell size of a = 17.37 Å, b = 5.96 Å, c = 25.66 Å, α = β = γ = 90º.
Gibbs free energies for each gaseous and adsorbed species were calculated at 298.15 K, according to the expression:             As displayed in Figures S10-S12, commercial plastics including PE plastic bags, PP plastic boxes and PET plastic bottles were initially crushed into fine powders by a pulverizer to ensure their good contact with H2O. The corresponding SEM images showed that the sizes of most powders are smaller than 5 mm.  As displayed in Figures S14, a handful of microplastics with smaller size were also detected, which might be produced during the photoconversion processes, conforming to the principle of carbon balance.

Figure S15
The weight losses of commercial PE plastic bags, PP plastic boxes and PET plastic bottles for the Co-Ga2O3 nanosheets and the Ga2O3 nanosheets in a simulated sunlight (AM 1.5G, 100 mW/cm 2 ) at ambient temperature and pressure after 24 h irradiation, respectively. The error bars represent the standard deviations of three independent measurements.
After the photodegradtion processes, the residual plastic powders were collected to calculated their weight losses. To be specific, the plastic powders and photocatalysts were collected by centrifugation after the photoconversion (total residual weight, W1). To lower the deviation, the residual weight of the photocatalysts (W2) were estimated first (when 50 mg photocatalysts were dispersed into 100 mL H2O, around 46 mg samples could be collected by centrifugation). Then the weight losses of plastics were calculated by the following formula: Plastics weight losses (%) = 1 − W1 -W2 initial mass of plastics (Eq. S9) Considering the losses of the plastics during the centrifugation, the calculated plastics weight losses may still be little high. The weight losses of PE plastic bags powders were 53%, 49% and 42% for the Co-Ga2O3 nanosheets, while those were 30%, 29% and 26% for the Ga2O3 nanosheets after 24 h irradiation. More importantly, the weight losses of the PE plastic bags, PP plastic boxes and PET plastic bottles were up to ca. 81%, 78% and 72% after 48 h irradiation over the Co-Ga2O3 nanosheets.

Figure S16
Cycling stability tests: another 100 mg commercial PE plastic bags were added to the system after 24 h irradiation, then the reaction system was vacuum-treated and refilled with high-purity air to reach an atmospheric pressure.
As shown in Figure S16, the formation rates of H2, CO and CO2 for the Co-Ga2O3 nanosheets did not show any obvious drop upon adding another 100 mg commercial PE plastic bags to the system after 24 h irradiation for the cycling measurements, suggesting their superb photocatalytic stability.

Figure S17
Photoconversion of reagent-grade PE powders over the Co-Ga2O3 nanosheets and the Ga2O3 nanosheets in a simulated sunlight (AM 1.5G, 100 mW/cm 2 ) at ambient temperature and pressure: the formation rates of H2, CO and CO2. The error bars represent the standard deviations of three independent measurements.

Figure S18
Photocatalytic water-splitting tests for the Co-doped Ga2O3 nanosheets and the Ga2O3 nanosheets in N2 atmosphere (no PE powders were added): the formation rates of H2 and O2 in 5 h. The error bars represent the standard deviations of three independent measurements.
As shown in Figure S18, H2O could be directly split into O2 and H2 over the Codoped Ga2O3 nanosheets and the Ga2O3 nanosheets when no PE powders were added. The formation rates of H2 and O2 of the Co-doped Ga2O3 nanosheets were 352.0 and 164.9 µmol g -1 h -1 , while those for the Ga2O3 nanosheets were 191.4 and 86.1 µmol g -1 h -1 respectively.

Figure S19
UV-vis absorption spectrum of the reaction solution after the photocatalytic water-splitting tests for 5 h over the Co-Ga2O3 nanosheets.
As shown in Figure S19, some amount of H2O2 was formed during the water splitting process over the Co-Ga2O3 nanosheets.
As revealed in Figure S20, 13 CO was detected after the CO2 photoreduction experiments, which clearly showed that CO came from the reduction of CO2.

Figure S21
The synchrotron-based vacuum UV photoionization mass spectrometry (SVUV-PIMS) of the gas products after 24 h irradiation: blank controlled experiments for the photoconversion of reagent-grade PE powders over the Co-Ga2O3 nanosheets at hν = 14.5 eV. The reaction system was vacuum-treated three times to remove air completely, which was then pumped by high-purity N2 to reach an atmospheric pressure.
As displayed in Figure S21, upon removing H2 18 O and 18 O2 from the reaction system, no C 16 O 18 O was detected in blank controlled experiments, further confirming that H2O and O2 participated in the oxidation of PE powders into CO2.

Figure S22
In situ ESR spectra during the photoconversion of reagent-grade PE powders over the Co-Ga2O3 nanosheets and the Ga2O3 nanosheets with H2O as solvent and DMPO as spin-trapping agent.
As shown in Figure S22, the ESR signals exhibited 1:2:2:1 quartet pattern in water solution, which could be assigned to the ·OH radicals captured by DMPO.

Figure S23
In situ ESR spectra during the photoconversion of reagent-grade PE powders over the Co-Ga2O3 nanosheets and the Ga2O3 nanosheets with methanol as solvent and DMPO as spin-trapping agent.
As displayed in Figure S23, the ESR signals showed quartet pattern with the intensity of nearly 1:1:1:1 in methanol solution, which could be assigned to the superoxide radicals (O2 ·-) captured by DMPO.

Figure S29
The differential charge density maps of *COOH for the Ga2O3 nanosheets, in which the yellow and blue contours manifest electron accumulation and depletion, respectively. The value of isosurfaces is 0.002 eÅ -3 .
As shown in Figure S28, the charge of Ga on the Ga2O3 nanosheets mainly localize on Ga atoms. The introduction of Co atoms causes the charge redistribution of the neighboring Ga atoms, and the charge of Ga atoms would shift to the Co side ( Figure  3D), which helps to stabilize the *COOH and lower its formation energy ( Figure S29, Figure 3B, Figure 3E).

Figure S30
The differential charge density maps of *H for Ga2O3 nanosheets, in which the yellow and blue contours manifest electron accumulation and depletion, respectively. The value of isosurfaces is 0.002 eÅ -3 .
As shown in Figure S30 and Figure 3C, 3F, the introduction of Co atoms results in the increased charge density around Co atoms that would tend to bond with the *H intermediates and hence lowers their formation energies.

Table S1
The formation rates of H2, CO and CO2 during the photoconversion of various commercial plastic products on the Co-Ga2O3 nanosheets and the Ga2O3 nanosheets with a simulated sunlight (AM 1.5G, 100 mW/cm 2 ) at ambient temperature and pressure.

Table S2
The weight losses of commercial PE plastic bags, PP plastic boxes and PET plastic bottles for the Co-Ga2O3 nanosheets and the Ga2O3 nanosheets with a simulated sunlight (AM 1.5G, 100 mW/cm 2 ) at ambient temperature and pressure after 24 h irradiation.

Table S3
Control experiments: the formation rates of H2, CO and CO2 during the photoconversion of reagent-grade PE powders on the Co-Ga2O3 nanosheets and the Ga2O3 nanosheets with a simulated sunlight (AM 1.5G, 100 mW/cm 2 ) at ambient temperature and pressure when light, photocatalysts or PE powders was removed. The "n.d." refers to "not detected".

Table S4
Control experiments: the formation rates of H2, CO and CO2 during the photoconversion of reagent-grade PE powders on the Co-Ga2O3 nanosheets and the Ga2O3 nanosheets with a simulated sunlight (AM 1.5G, 100 mW/cm 2 ) at ambient temperature and pressure when 2 mmol/L AgNO3 solution, or high-purity O2, or N2 was used. The "n.d." refers to "not detected".

Table S5
The calculation of carbon balance after the photodegradation of commercial plastic products on Co-Ga2O3 nanosheets in simulated sunlight (AM 1.5G, 100 mW/cm 2 ) at ambient temperature and pressure.

Table S8
Free energy (eV) of the Co-Ga2O3 nanosheets and the Ga2O3 nanosheets with and without intermediates.