CO2-facilitated upcycling of polyolefin plastics to aromatics at low temperature

ABSTRACT Plastics are one of the most produced synthetic materials and largest commodities, used in numerous sectors of human life. To upcycle waste plastics into value-added chemicals is a global challenge. Despite significant progress in pyrolysis and hydrocracking, which mainly leads to the formation of pyrolysis oil, catalytic upcycling to value-added aromatics, including benzene, toluene and xylene (BTX), in one step, is still limited by high reaction temperatures (>500°C) and a low yield. We report herein CO2-facilitated upcycling of polyolefins and their plastic products to aromatics below 300°C, enabled by a bifunctional Pt/MnOx-ZSM-5 catalyst. ZSM-5 catalyzes cracking of polyolefins and aromatization, generating hydrogen at the same time, while Pt/MnOx catalyzes the reaction of hydrogen with CO2, consequently driving the reaction towards aromatization. Isotope experiments reveal that 0.2 kg CO2 is consumed per 1.0 kg polyethylene and 90% of the consumed CO2 is incorporated into the aromatic products. Furthermore, this new process yields 0.63 kg aromatics (BTX accounting for 60%), comparing favorably with the conventional pyrolysis or hydrocracking processes, which produce only 0.33 kg aromatics. In this way, both plastic waste and the greenhouse gas CO2 are turned into carbon resources, providing a new strategy for combined waste plastics upcycling and carbon dioxide utilization.

then shaped to pellets with a size of 40-60 mesh.The catalyst was packed into the reactor as a fixed bed, which was supported by a quartz-fiber bed below.It was heated from room temperature to 300 º C at a ramp rate of 5 ℃ /min.CO 2 with 5% Ar as the internal standard for online gas chromatography (GC) analysis was fed through the mixture of polyolefins and catalyst (10 mL/min).Reaction conditions were 1.0 MPa and 300 º C unless otherwise stated.
Thermogravimetric analysis (TG) was performed on Netzsch STA 449 F3.The mass loss was recorded while the catalyst was heated from 40 to 900 °C at a ramp of 10 °C/min in a flowing air (100 mL/min).
Nitrogen adsorption-desorption was carried out on a Quantachrome NOVA 4200e.Before analysis, all samples were pretreated at 300 ℃ for 6 h under vacuum.Isotherms were recorded at liquid nitrogen temperature of 77 K.
Temperature programmed desorption of ammonia (NH 3 -TPD) was performed on a Micromeritics AutoChem 2910 equipped with a TCD.The catalyst was first pretreated in a flowing Ar at 550 ℃ for 1.5 h.After cooling down to 100 ℃ in a flowing Ar, the sample was exposed to 5% NH 3 /He at 100 ℃ .Then the sample was swept by He flow at 100 º C till a stable TCD signal baseline was obtained.Subsequently, the temperature was increased from 100 to 800 ℃ at a ramp of 10 ℃ /min.
Pyridine Fourier transform infrared spectroscopy (FTIR), in situ H-D exchange FTIR and in situ diffuse reflection Fourier transform infrared spectroscopy (DRIFTS) of CO 2 adsorption were performed on a Bruker Tensor 27 with a MCT detector.
For a typical Pyridine FTIR experiments, catalyst powder was pressed into a wafer with a diameter of 14 mm with a thickness less than 0.5 mm.Prior to pyridine adsorption, the sample was degassed under vacuum (<10 -2 Pa) at 450 º C for 1.5 h.The background spectrum was recorded after the cell had been cooled down to 250 ºC .Subsequently, the sample was exposed to pyridine vapor for 5 min at room temperature, followed by evacuation to <10 -2 Pa for 30 min.FTIR spectra were then recorded by accumulating 64 scans at a resolution of 4 cm -1 .
Prior to the in situ H-D exchange FTIR experiments, the sample was pretreated in H 2 at 350 ℃ for 2 h.The background spectrum was recorded after the cell have cooled down to 200 ℃ .Subsequently, the sample was exposed to D 2 at the same temperature.Then the Si-OD-Al band was monitored by accumulating 32 scans at a resolution of 4 cm -1 .
Prior to the in situ DRIFTS of CO 2 adsorption, all samples were pretreated in H 2 at 350 ℃ for 2 h.The background spectrum was recorded after the cell have cooled down to 250 ℃ in Ar.Subsequently, the sample was exposed to CO 2 .DRIFTS were then recorded by accumulating 64 scans at a resolution of 4 cm -1 .After 20 min, the sample was exposed to Ar at 250 ℃ to remove the physically adsorbed CO 2 , followed exposure to H 2 (5 mL/min).DRIFTS were by accumulating 64 scans at a resolution of 4 cm -1 .
The molecular weights and molecular weight distributions of the polyolefin were determined by gel permeation chromatography (GPC) with the PL-GPC220 equipped with a 40-position autosampler and a high-sensitivity refractive index detector at 150 ℃ using 1,2,4-trichlorobenzene as the solvent and calibrated with polystyrene standard.
GC-MS analysis was carried out using (Agilent 7890A-7000B) equipped with a PONA capillary column or FFAP column.
High resolution transmission electron microscopy (HRTEM) characterization was carried out on a JEOL JEM-F200  It can be seen from Figure S1c, thermodynamic equilibrium value of CO 2 consumption under 300 ℃ is 1.9 mol per 1 mol C 20 olefin, the yield of aromatics can be increased from 48% to 82%, and the amounts of aromatics can be increased from 9.7 C mol to 18.2 C mol.     Figure S14.BTX-yield in this work (red stars) in comparison to the previously reported values (black circles) for one pot reactions versus reaction temperatures (corresponding to the data in Table S6).CO 2 turned into 13 CO according to the relative concentration of 13 CO among all CO measured by GC-MS analysis.It indicated that 13 CO accounted for 86% of all CO and the rest 14% CO was 12 CO, attributed to the reaction of 13 CO 2 with carbon residual (Equation S5).
Thus, the fraction of CO 2 entering the aromatics (S CO2-Aro ) was calculated by Equation S6: where n CO2 stands for the amount of converted CO 2 and n CO for the total amount of CO, which were measured by GC.Thus, one can calculate that 90% of the consumed 13 CO 2 had been incorporated into the aromatic products.a: Reaction conditions: 1.0 MPa CO 2 , 300 °C, 0.4 g catalyst, 1.0 g LDPE-1, 100 ml batch reactor, unless otherwise stated.Note that there is a small amount of undetected products.b: The solid residual after reaction, including carbon deposition and unreacted polyolefin, was quantified by TG.

Figure S1 .
Figure S1.Thermodynamic calculation of C 20 -Olefins conversion as a function of reaction temperature using HSC chemistry 9.0 software package.(a) Product distribution in the co-conversion of C 20 -Olefins and CO 2 ; (b) Product distribution in the C 20 -Olefins conversion without CO 2 ; (c) Thermodynamic equilibrium value of CO 2 consumption; (d) Comparison of aromatics yield in C 20 -Olefins with CO 2 (red curve) and without CO 2 (blue curve) assuming C 3 H 8 , H 2 , toluene C 7 H 8 as products.

Figure
Figure S3.HRTEM images of fresh Pt/MnO x (pre-reduced).(a) A typical image; (b) A high resolution image (c) High angle annular dark-field image; (d) Size distribution of Pt nanoparticles.

Figure S13 .
Figure S13.GC analysis of liquid products in the upcycling reaction of LDPE-1 with CO 2 .

Figure S15 .
Figure S15.HRTEM of used Pt/MnO x , which was separated from Pt/MnO x -ZSM-5 after reaction for four cycles.(a) A typical TEM image; (b) A high resolution image; (c) High angle annular dark-field image; (d) Size distribution of Pt nanoparticles.

Figure S16 .
Figure S16.XRD patterns of Pt/MnO x -ZSM-5 catalysts before and after reaction, and after regeneration.After reduction, only MnO crystal phase was observed.The reduced Pt/MnO x was mixed with ZSM-5 as the fresh Pt/MnO x -ZSM-5.After reaction, MnO was converted to MnCO 3 , and the valence state of Mn remains +2, without further oxidation.In comparison, after 4 cycles of reaction and regeneration, no obvious crystallinity change was observed compared to the fresh Pt/MnO x -ZSM-5 catalyst.

Figure
Figure S17.NH 3 -TPD profiles of Pt/MnO x -ZSM-5 catalysts before and after 4 cycles of reaction and regeneration.

Figure S18 .
Figure S18.Pyridine adsorption of Pt/MnO x -ZSM-5 catalysts before and after reaction and after regeneration.

Figure S20 .
Figure S20.Mass spectra of 12 CO and 13 CO detected during the reaction of LDPE-1 with (a) 13 CO 2 in comparison to that with (b) 12 CO 2 at 300 º C.

Figure
Figure S21.FT-IR differential spectra of CO 2 adsorbed Pt/MnO x and MnO x catalysts referenced to the catalysts prior to CO 2 adsorption.Stretching vibration of C-O at 1000 -2200 cm -1 ; stretching vibration of C-H at 2800 -3200 cm -1 .

Figure S22 .
Figure S22.In situ IR differential spectra recorded after introduction of H 2 to the CO 2 adsorbed catalysts as a function of time on stream.(a) Pt/MnO x (pre-reduced); (b) MnO x (pre-reduced).

Figure
Figure S24.A fixed bed reaction with CO 2 flowing through the bed composed of the mixture of LDPE-1 and ZSM-5, with the effluents monitored by an online GC.

Figure S26 .
Figure S26.Detailed product distribution during the model reactions of CH 3 OH/C 6 =

Figure S27 .
Figure S27.Reaction results of CO/LDPE-1 and CO/1-hexene in comparison to that of CO 2 /LDPE-1.A similar product distribution further validates the proposed reaction mechanism in the paper.

Figure S28 .
Figure S28.Mass spectra of alkane represented by isopentane during the reaction of 13 CO/1-hexene in comparison to that of 12 CO/1-hexene.
Transmission Electron Microscope, which was equipped with two EDS analyzers, operating at an accelerating voltage of 200 kV. 20  40 → 1.25  7  8 + 3.75  3  8 S21. FT-IR differential spectra of CO 2 adsorbed Pt/MnO x and MnO x catalysts referenced to the catalysts prior to CO 2 adsorption.Stretching vibration of C-O at 1000 -2200 cm -1 ; stretching vibration of C-H at 2800 -3200 cm -1 .