Coupling of PET waste electroreforming with green hydrogen generation using bifunctional catalyst

Abstract

Cost-effective and high-efficiency bifunctional electrocatalysts for electrooxidation of polyethylene terephthalate (PET) waste and green hydrogen generation are very crucial for practical implementation yet rarely reported. Herein, a bifunctional catalyst of cobalt modified nickel phosphide nanosheet arrays on nickel foam (Co-Ni₂P/NF) for both PET hydrolysate oxidation reaction and hydrogen evolution reaction (HER) is reported, which is obtained by a facile hydrothermal and phosphidation treatment. The electrocatalyst is highly active for both PET hydrolysate oxidation reaction and HER with low overpotentials of 90 and 148 mV, respectively, to achieve a current density of 50 mA cm⁻². By coupling PET hydrolysate oxidation reaction with HER, the assembled electrolyzer with Co-Ni₂P/NF as a bifunctional catalyst only requires 1.43 V to afford 10 mA cm⁻², much lower than that needed for pure water splitting (1.55 V). Complementary DFT study provides an in-depth understanding of HER and electrooxidation of PET on Co-Ni₂P/NF. Our work suggests that electroreforming of abundant PET waste could be an energy-efficient and sustainable strategy for both plastic waste valorization and green hydrogen production via using a cost-effective and active bifunctional catalyst.

---

Introduction

Hydrogen (H₂) has been dubbed “the cleanest energy source in the world” because its combustion product is pure water. Moreover, H₂ is also a stable media for storing intermittent renewable energy, thus is expected to play a pivotal role for a sustainable energy future.^1–3 Nevertheless, only green H₂ is clean and emits negligible carbon in its life cycle. Water splitting is the only method to produce green H₂.^4–6 A thermodynamic voltage of at least 1.23 V is required for H₂O molecular breakage, where substantial amount of energy is used to oxidize water that produces the by-product oxygen.^7,8 Therefore, replacing water oxidation with a new oxidation reaction to reduce energy consumption and to produce higher value product (than oxygen) is attracting ever-increasing interest.^9,10 Electrooxidation of many different organic molecules has been reported to couple with HER for green H₂ generation, such as biomass,^11–13 hydrazine,^14,15 isopropanol,¹⁶ benzylamine,¹⁷ urea,¹⁸ formic acid,¹⁹ and so on.

To this end, polyethylene terephthalate (PET) plastic, as an organic polymer, also can be used to assist the production of green H₂. Due to its excellent tensile strength, thermal stability, electrical insulation, processability, chemical resistance, non-toxicity, and tastelessness, PET is often used in electronic devices, clothing, and product packaging, etc.²⁰ Among them, PET bottles account for about one third of the total amount of PET produced (∼24 million tons),²¹ and are widely used for food, chemical, and pharmaceutical packaging. Although PET bottles have brought a lot of convenience to our daily life, the natural degradation of their waste requires hundreds of years, which causes serious environmental issues, i.e., “white pollution”.²² The recycling of PET waste can not only address environment issue but also recover important resources, being very crucial for sustainable development. Therefore, it is undoubtedly attractive to combine the electroreforming of waste PET bottles with H₂ production.

As far as we know, only few reports to date focus on the plastic electroreforming-assisted green H₂ generation. Duan and coworkers successfully synthesized CoNi_0.25P electrocatalyst for electrocatalytic upcycling of PET to produce formate and H₂. The results reveal that the Faradaic efficiency and selectivity of formate reach more than 80% in a membrane-electrode assembly reactor when the current density is 500 mA cm⁻².²¹ Zhao's group successfully converted PET to formate by using CuO NWs/CF and meanwhile assisted H₂ production. Experimental and density functional theory (DFT) calculation demonstrated that the optimal reaction pathway on CuO NWs is to generate the glyoxal intermediate and obtain formate through the C-C bond cleavage with a high selectivity of 86.5%.²³ However, pure PET plastics were used in these studies, and there is no report on the electroreforming of raw plastic waste coupled with H₂ generation yet.

Herein, we electroreform raw waste PET bottles to assist H₂ production via electrolysis. We firstly explore the maximum solubility of waste PET bottles in sodium hydroxide (NaOH), and then optimize an efficient and active electrocatalyst for electrooxidation of PET bottles hydrolysate, namely cobalt-modified nickel phosphide nanosheets on the nickel foam (Co-Ni₂P/NF). Moreover, Co-Ni₂P/NF serving as a bifunctional electrocatalyst are assembled to form an electrolytic cell, which only requires a voltage of 1.43 V to achieve a current density of 10 mA cm⁻² in 1 M NaOH electrolyte with 5 g L⁻¹ PET, and the Faradaic efficiency of formate can reach 85%, while the recovery of terephthalic acid (TPA) was as high as 80%.

Experimental section

The synthesis of Co-modified Ni₂P nanosheets on Ni foam (Co-Ni₂P/NF)

To produce a series of ultrathin Co_x-Ni₂P/NF nanosheets (x = 0.2, 0.4, 0.6, 0.8), (1 − x) mL 0.5 M nickel(II) chloride hexahydrate (NiCl₂·6H₂O), x mL 0.5 M cobaltous chloride (CoCl₂·6H₂O), and 1 mL ammonium hydroxide (NH₃·H₂O) were mixed to obtain a uniform solution. And then 1 × 1 cm² nickel foam (NF) was immerged into the uniform solution and 1 mL hydrazine hydrate (N₂H₄) solution was added. The solution was then heated at 75 °C for 10 h, obtaining layered cobalt modified nickel hydroxide precursor on NF [termed as Co_x-Ni(OH)₂/NF].

Afterward, a piece of Co_x-Ni(OH)₂/NF and 0.05 g sodium hypophosphite (NaH₂PO₂·H₂O) were placed in two different combustion boats with NaH₂PO₂·H₂O boat located at upstream of a tube furnace and then heated at 300 °C for 2 h with a flow of argon. The excess PH₃ in the product was absorbed by the saturated copper(II) sulfate (CuSO₄) solution after treatment at the downstream. Finally, the product (Co_x-Ni₂P/NF) was obtained after being naturally cooled down to room temperature, followed by repeat water cleaning, and then drying at 60 °C for 10 h.

CoP/NF and Ni₂P/NF were also fabricated as control samples. The synthetic method of CoP/NF nanosheets and Ni₂P/NF nanosheets was similar to that of Co_x-Ni₂P/NF nanosheets, except that the mixture solution of NiCl₂·6H₂O and CoCl₂·6H₂O was substituted by pure CoCl₂·6H₂O or NiCl₂·6H₂O, respectively.

Electrochemical characterizations

HER characterization and PET hydrolysate electrooxidation process were carried out in a typical three-electrode system. Electrocatalyst supported on NF was used as working electrode, Ag/AgCl electrode and carbon rod acted as reference electrode and counter electrode, respectively. LSV curves were compensated according to: E_compensated = E_measured − IRs, where R_s is the solution resistance extracted from the electrochemical impedance spectroscopy (EIS) measurement.²⁴ And according to the equation E_RHE = E_Ag/AgCl + 0.197 V + 0.0591 pH, all electrode potentials have been calibrated as reversible hydrogen electrode (RHE).

In addition, the Faradaic efficiency (FE) for the PET electrooxidation to formate is determined by equation: FE = (3Fn)/Q, where F is Faraday's constant (96 485 C mol⁻¹), n represents the moles of the formate produced, and Q is the total charge passed across the electrode during electrolysis.^21,25

Hydrolysis of waste PET bottles

The mixture of various white waste PET plastic bottles was briefly rinsed with water and dried at 70 °C (for precise weighing). Different amounts of the PET bottle fragment (0.5 g, 1.25 g, 3.75 g) were added to 25 mL 10 M sodium hydroxide (NaOH) solution, and stirred at 80 °C for 10 h. After cooling to room temperature, water was added to increase solution volume to 250 ml, and then it was stirred until the PET fragments were completely dissolved. The electrolytes with different PET contents (marked as 2 g L⁻¹, 5 g L⁻¹, 15 g L⁻¹) were prepared.

Computational method

All the calculations were carried out using the density functional theory (DFT) with the generalized Perdew–Burke–Ernzerhof (PBE)²⁶ with DFT + U (U_eff = U − J = 5.3 for Ni and U_eff = U − J = 4.0 for Co) and the projector augmented-wave (PAW) pseudopotential plane-wave method²⁷ as implemented in the VASP code.²⁸ For the PAW pseudopotential, we included 3d⁸4s¹, 3d⁸4s², 2s²2p⁴, 2s²2p² and 1s¹ were treated as valence electrons for Co, Ni, O, C and H atoms, respectively. A 5 × 5 × 3 Monkhorst–Pack (MP) k-point grid was used for Ni₂P bulk optimization and a plane-wave basis set with an energy cutoff of 400 eV considering the spin polarization. Good convergence was obtained with these parameters and the total energy was converged to 1.0 × 10⁻⁶ eV per atom. The optimized bulk model was used to build the (201) surface unitcell and further expanded to a 6-atom layer 3 × 2 × 1 supercell with a vacuum separation of ∼18 Å for H adsorption and ethylene glycol (EG) reduction process studies. We carried out calculations with the van der Waals (vdW) correction by employing optB86b-vdW functional using a 2 × 2 × 1 MP k-point grid.²⁹

Results and discussion

Synthesis and characterization of electrocatalysts

Co-Ni₂P/NF electrode was successfully obtained through a facile two-step process. As shown in the Scheme 1, cobalt modified nickel hydroxide nanosheet arrays on nickel foam (denoted as Co-Ni(OH)₂/NF) was first achieved (Fig. S1†). Secondly, the final electrocatalyst of Co-Ni₂P/NF was obtained by phosphidation with NaH₂PO₂·H₂O at 300 °C for 2 h.

Scheme 1 Schematic of the synthetic method of Co-Ni₂P/NF.

The morphological characterizations of Co-Ni₂P/NF were analyzed by the scanning electron microscopy (SEM). As depicted in Fig. S2,† the morphologies of four Co-Ni₂P/NF samples are nanosheets, indicating that the variation of Co content does not affect the unique lamellar structure of the electrocatalysts. We chose Co_0.6-Ni₂P/NF for more detailed characterization. As shown in Fig. 1A and B, the Co_0.6-Ni₂P/NF nanosheets are uniformly distributed over the entire surface of 3D porous NF and inherit the highly cross-linked ultrathin layered structure of Co-Ni(OH)₂. The chemical composition of the Co_0.6-Ni₂P/NF were characterized by SEM-mapping, energy-dispersive X-ray spectrum (EDS), and the X-ray diffraction (XRD). The mapping images display the uniform distribution of Ni, Co, and P element (Fig. 1C), indicating the successful synthesis of metal phosphide. And the EDS spectrum further proves the Co element in the product (Fig. 1D). As depicted in Fig. 1E, the XRD pattern suggests that the remaining peaks perfect match with (111), (210), (300) and (211) planes of hexagonal Ni₂P (JCPDS no. 74-1385) expect for the NF substrate (JCPDS no. 04-0850), suggesting the main component of the product is Ni₂P. Due to its low content and the influence of the substrate, the peaks of cobalt species could not be observed. Putting together, the Co modified Ni₂P nanosheets were successfully obtained.

Fig. 1 (A) and (B) SEM images, (C) SEM-mapping, (D) SEM-EDS, and (E) XRD pattern of Co_0.6-Ni₂P/NF.

Transmission electron microscopy (TEM) images further show that Co modified Ni₂P nanosheets have been synthesized successfully (Fig. 2A). More detail about the composition and structures were established by high-resolution transmission electron microscopy (HRTEM). As shown in Fig. 2B, the HRTEM reveals that the heterointerface structure exists in the ultrathin Co_0.6-Ni₂P nanosheets, and the interfaces between CoP and Ni₂P marked with yellow dotted line can be clearly observed. The lattice fringes in different regions can match with two different material of CoP and Ni₂P, respectively. The interplanar spacing of 0.203 nm is agreement with the (201) plane of Ni₂P, and the lattice fringe spacing of 0.189 nm can be indexed to (211) plane of CoP. The EDS elemental mapping analysis of Co_0.6-Ni₂P nanosheets are depicted in Fig. 2C. It can be observed that the elements of Co, Ni and P are uniformly dispersed in the nanosheets. Furthermore, the composition ratio of CoP and Ni₂P in Co_0.6-Ni₂P/NF is about 1 : 4 (Fig. 2D), as obtained from EDS analysis.

Fig. 2 (A) TEM image, (B) HRTEM image, (C) TEM-mapping, and (D) TEM-EDS analysis of Co_0.6-Ni₂P/NF.

As presented in Fig. 3, the chemical valence of Co_0.6-Ni₂P/NF was analyzed by X-ray photoelectron spectroscopy (XPS). The Co, Ni and P elements can be obviously seen from the full spectrum of Co_0.6-Ni₂P/NF (Fig. 3A). As illustrated in Fig. 3B, two narrow peaks at 874.5 eV and 856.7 eV in the Ni 2p XPS spectrum correspond to Ni 2p_1/2 and Ni 2p_3/2 of the oxidized Ni species, while the peaks located at 861.8 eV and 880.3 eV belong to satellite peaks.^30,31 Furthermore, other two peaks centered at 852.9 eV and 869.8 eV are consistent with Ni–P bonds.^32–34 On the Co 2p XPS spectrum, the two peaks located at 778.6 eV and 793.7 eV correspond to Co 2p_3/2 and Co 2p_1/2 of Co–P, respectively. And the two peaks located at 782.0 eV and 798.2 eV are consistent with Co–O bond, with two satellites detected at 803.4 eV and 786.0 eV (Fig. 3C).^35,36 As presented in Fig. 3D, one can observe two peaks located at 129.0 eV and 130.0 eV, which are corresponding to P 2p_3/2 and P 2p_1/2, respectively.³⁷ In addition, a broad peak located at 134.0 eV is originating from a surface oxidization.^38–40

Fig. 3 (A) XPS survey spectrum, and high-resolution XPS spectra of (B) Ni 2p, (C) Co 2p, and (D) P 2p of Co_0.6-Ni₂P/NF.

HER electrocatalytic performance of the prepared catalysts

As shown in Fig. 4A, HER activity was assessed by linear sweep voltammetry (LSV) curves in 1 M NaOH electrolyte. Co_0.6-Ni₂P/NF exhibits the highest HER activity, which only requires an overpotential of 69 mV to achieve a current density of 10 mA cm⁻². As depicted in Fig. 4B and D, the overpotential (148 mV) of Co_0.6-Ni₂P/NF at 50 mA cm⁻² is much smaller than that of Ni₂P/NF (226 mV), CoP/NF (246 mV), Co-Ni(OH)₂/NF (266 mV) and NF (284 mV). To achieve a current density of 100 mA cm⁻², Co_0.6-Ni₂P/NF only requires an overpotential of 200 mV (Fig. 4D). Notably, although the HER activity of Co_0.6-Ni₂P/NF is inferior to that of Pt/C/NF (24 mV), it is still superior to most of reported Ni/Co-based HER electrocatalysts (Fig. 4C).^14,41–52

Fig. 4 LSV curves of (A) Co-Ni₂P/NF with different Co contents in a 1 M NaOH solution and (B) Co-Ni(OH)₂/NF, CoP/NF, Ni₂P/NF, Co_0.6-Ni₂P/NF and Pt/C/NF. Comparison of the overpotential at current densities of (C) 10 mA cm⁻² and (D) 50 and 100 mA cm⁻². (E) Tafel plots of Co-Ni(OH)₂/NF, CoP/NF, Ni₂P/NF, and Co_0.6-Ni₂P/NF.

As illustrated in Fig. 4E, the Tafel slope value of Co_0.6-Ni₂P/NF, Ni₂P/NF, CoP/NF, and Co-Ni(OH)₂/NF are 128, 170, 297, and 275 mV dec⁻¹, respectively, suggesting that Co_0.6-Ni₂P/NF exhibits significantly faster HER kinetics than other electrocatalysts. As presented in Fig. 5A, the electrochemical impedance spectroscopy (EIS) analysis shows that Co_0.6-Ni₂P/NF has the lowest R_ct value (7.6 Ω) among Co-Ni(OH)₂/NF (46.4 Ω), CoP/NF (8.5 Ω), and Ni₂P/NF (10.8 Ω), suggesting Co_0.6-Ni₂P/NF's strongest charge transfer ability and fastest kinetics among all electrocatalysts tested.^53,54 The electrochemically active surface (ECSA) of the electrocatalysts are obtained by calculating the double layer capacitances (C_dl) (Fig. S3†). As depicted in Fig. 5B, Co_0.6-Ni₂P/NF exhibits a highest C_dl (20.1 mF cm⁻²) and thus largest ECSA among Co-Ni(OH)₂/NF (11.7 mF cm⁻²), CoP/NF (19.2 mF cm⁻²), and Ni₂P/NF (19.9 mF cm⁻²). The improved ECSA can be ascribed to the rough surface and 3D porous nanostructure that offers abundant active sites. Undoubtedly, the increase of the active edges/sites is favorable to enhance the interface with electrolyte, the adsorption of water molecule, and the release of H₂, thereby enhancing the HER electrocatalytic activity.^55,56Fig. 5C shows that Co_0.6-Ni₂P/NF exhibits higher intrinsic HER activity (normalized by ECSA), implying that the high total activity of Co_0.6-Ni₂P/NF is not only due to its large ECSA but also to its high intrinsic activity.

Fig. 5 (A) Nyquist plots acquired in 1 M NaOH electrolyte at −0.2 V vs. RHE with the corresponding fitting curves and the equivalent circuit (inset). (B) Plots of the current density at 0.32 V vs. scan rate. (C) ECSA-normalized LSV curves of Co-Ni(OH)₂/NF, CoP/NF, Ni₂P/NF, and Co_0.6-Ni₂P/NF. (D) Chronopotentiometry curve of Co_0.6-Ni₂P/NF in 1 M NaOH solution at constant current density of 10 mA cm⁻² (inset: polarization curves of Co_0.6-Ni₂P/NF before and after 2000 cycles).

We further support the excellent HER performance of Co-Ni₂P/NF with theoretical investigation. The optimized unit cell of Ni₂P was shown in Fig. S4.† The expanded (210) supercell contains 108 Ni and 54 P atoms. Only one Co atom was considered embedded into the Ni₂P surface. Considering three different embedding sites for Co on the surface and the relative energy reveal that the Co prefer to replace Ni at the 4-fold site.

The first step of HER is the electrochemical adsorption process: H⁺ + e⁻ → H*, the second step is the electrochemical discharge process: H* + H⁺ +e⁻ → H₂ or chemical desorption process: 2H* → H₂.⁵⁷ The Gibbs free energy of ΔG_H has been successfully used as a suitable descriptor for evaluating the HER catalytic activity of various materials. Generally, the optimal value of ΔG_H is 0 eV, where hydrogen is adsorbed neither too strongly nor too weakly to facilitate both hydrogen adsorption and desorption.⁵⁷

The adsorption energy (ΔE_H) is computed as

ΔE_H = E(*H) − E(*) − 1/2 E(H₂)

where E(*H) and E(*) are the total energy of a surface with and without hydrogen adsorption, respectively, and E(H₂) is the total energy of a H₂ molecule.

The Gibbs free energy of H (ΔG_H) is defined as:

ΔG_H = ΔE_H + ΔE_ZPE − TΔS_H

where ΔE_H is the adsorption energy, ΔE_ZPE is the difference in zero-point energy, T is the temperature (298 K) and ΔS_H is the entropy difference between H that is adsorbed and in the gas phase. We approximated the entropy of hydrogen adsorption as ΔS_H ≈ 1/2 (), where is the entropy of gas phase H₂ at standard conditions.

The calculated Gibbs free energies of hydrogen (ΔG_H) on the pristine and Co doped Ni₂P surfaces are shown in Fig. S5.† It can be observed that the ΔG_H is 418 meV on the pristine surface and dramatically decreases to 190 meV on Co doped surface. Based on our simulation results, it can be considered that doping Co atoms in Ni₂P can reduce the HER overpotential. It should be noted that only one Co atom was embedded into Ni₂P systems. Since the active site for HER is changed to Co, we consider that more Co doping within the solubility may further reduce the overpotential. Notably, our theoretical results are consistent with the experimental values, in which it was found that Co_0.6-Ni₂P/NF exhibits the highest HER activity with an overpotential of 69 mV.

Finally, the stability of Co_0.6-Ni₂P/NF in 1 M NaOH is accessed by the chronopotentiometric curve. As depicted in Fig. 5D, there was no noticeable decrease of HER current density during a 20 h continuous electrolysis. Indeed, SEM image reveals the morphology of Co_0.6-Ni₂P/NF retained well after 20 h continuous operation thanks to the great structural integrity due to the direct growth of Co_0.6-Ni₂P on NF (Fig. S6A†). Moreover, the HER activity was retained well with a small increase of only 10 mV at 100 mV cm⁻² after 2000 potential cycles (inset of Fig. 5D). It should be noted that the composition of Co_0.6-Ni₂P/NF after HER test showed no noticeable changes (Fig. S6B–F†). The excellent HER performances of Co_0.6-Ni₂P/NF can be explained as follows. Firstly, the in situ growth of Co_0.6-Ni₂P on NF support not only reinforces the interaction between Co_0.6-Ni₂P nanosheet and NF, enhancing the structural integrity, but also improves electrical conductivity, facilitating efficient charge transport.^58–60 Secondly, Co_0.6-Ni₂P possesses ultrathin nanosheet structure and a large electrochemical active area, beneficial for the exposure of more active sites. Also, the 3D porous structure can provide abundant channels to facilitate ion transfer and release of product (H₂), thus giving rise to the boosted catalytic kinetics.^61,62 Thirdly, the phosphorus and metal in Co_0.6-Ni₂P nanosheets play an important role in capturing nearby protons and act as acceptor centers for adsorbed hydrides, promoting HER process.^63,64 Lastly, Co doping and the interface formed between CoP and Ni₂P can further improve the catalytic activity by adjusting the electronic structure of the electrocatalyst.^65–67

The effect of PET hydrolysate concentration on electrooxidation activity

Fig. 6A and B show that Co_0.6-Ni₂P/NF nanosheet exhibits the highest current density for oxygen evolution reaction (OER) at the same voltage in 1 M NaOH. The LSV curves show that the voltage required for the PET hydrolysate electrooxidation of Co_0.6-Ni₂P/NF is significantly lower than that of OER at the same current density (Fig. 6C). To achieve 100 mA cm⁻², Co_0.6-Ni₂P/NF only requires an overpotential of 174 mV in 1 M NaOH with 2 g L⁻¹ PET, which is much lower than that of OER (350 mV). CV curves further show that the PET hydrolysate electrooxidation current density of Co_0.6-Ni₂P/NF at 1.5 V vs. RHE has reached 183 mA cm⁻² while that of OER is only 46 mA cm⁻² (Fig. 6D). These observations indicate that PET hydrolysate electrooxidation is much more favorable than OER thermodynamically.

Fig. 6 (A) LSV curves of Co-Ni₂P/NF with different Co contents in a 1 M NaOH solution. (B) Comparison of the current densities at potential of 1.6 V and 1.8 V vs. RHE. (C) LSV curves of Co_0.6-Ni₂P/NF in 1 M NaOH with or without 2 g L⁻¹ PET. (D) CV curves of Co_0.6-Ni₂P/NF in 1 M NaOH with or without 2 g L⁻¹ PET.

Fig. 7A shows that the current density gradually increases with the addition of 0.5 mL PET hydrolysate (5 g L⁻¹) at 1.5 V. However, the magnitude of each increase in current density gradually decreases. As illustrated in Fig. 7B, one can observe that the current density increases at first and then decreases with increasing PET hydrolysate concentration, and peaks at 5 g L⁻¹. Too high concentration (>5 g L⁻¹) could severely affect the mass transfer and ionic conductivity, and thus decrease the total activity.

Fig. 7 (A) Chronoamperometry curves of Co_0.6-Ni₂P/NF at 1.5 V vs. RHE after adding PET hydrolysate into 1 M NaOH solution. (B) LSV curves of Co_0.6-Ni₂P/NF in 1 M NaOH with different PET hydrolysate content.

PET hydrolysate electrooxidation activity comparison

Next, we investigate the PET hydrolysate electrooxidation performance of the five electrocatalysts under the same experimental conditions. As depicted in Fig. 8A, the PET hydrolysate electrooxidation activity of Co-Ni₂P/NF is much better than other electrocatalysts. Specifically, Co-Ni₂P/NF only requires an overpotential of 90 mV to reach a current density of 50 mA cm⁻², which is much smaller than those of CoP/NF (140 mV), Ni₂P/NF (230 mV), Co-Ni(OH)₂/NF (385 mV), RuO₂/NF (422 mV) and NF (544 mV). The current density of Co-Ni₂P/NF at 1.4 V and 1.6 V are 125 and 402 mA cm⁻², respectively, which are higher than those of the rest electrocatalysts (Fig. 8B). Continuous electrolysis for 1 h at different voltages (Fig. 8C) was conducted, followed by pH adjust using HCl. White substances were precipitated during this process (Fig. 8D). The liquid electrolyte and precipitate were separated by filtration, wherein the precipitate was confirmed to be terephthalic acid (TPA, C₈H₆O₄) by XRD with recovery yield of 80% (Fig. 8E). Additionally, the liquid product was characterized by high-performance liquid chromatography (HPLC). As depicted in Fig. S7,† the content of ethylene glycol (EG) was the highest at 1.2 V vs. RHE, and the electrooxidation rate of EG accelerated when the applied voltage increased. After 1 h electrolysis, the main product was formate, and its maximum FE was about 85% at 1.5 V vs. RHE (Fig. 8F).

Fig. 8 (A) LSV curves of Co-Ni₂P/NF, Ni₂P/NF, CoP/NF, Co-Ni(OH)₂/NF, RuO₂/NF and NF anodes in the electrolyte of 1 M NaOH with 5 g L⁻¹ PET. (B) Comparison of the current density of Co-Ni₂P/NF, Ni₂P/NF, CoP/NF and Co-Ni₂(OH)₂/NF at 1.4 V vs. RHE and 1.6 V vs. RHE. (C) Chronoamperometry curves of Co-Ni₂P/NF at the corresponding potentials. (D) Pictures for separation of TPA and formic acid. (E) The XRD pattern of the recycled solid product with standard TPA spectrum. (F) The Faradaic efficiencies of PET oxidation to formate at each given potential of Co-Ni₂P/NF. (G) Continuous electrolysis using Co-Ni₂P/NF over 20 h in 1 M NaOH electrolyte with 5 g L⁻¹ PET at current density of 200 mA cm⁻².

To further verify the durability of Co-Ni₂P/NF for electroreforming of PET hydrolysate, continuous electrolysis was conducted for 20 h at a constant current density of 200 mA cm⁻², as illustrated in Fig. 8G. The imperceptible potential change shows the excellent stability of Co-Ni₂P/NF electrode for PET electroreforming. Moreover, the morphology of Co-Ni₂P/NF was dominated by nanosheets after 20 h continuous electrolysis (Fig. S8A†). As depicted in Fig. S8B–F,† Co-Ni₂P/NF underwent partial reconstruction and oxidation after PET hydrolysate oxidation, generating metal oxy(hydroxide) analogue. Based on our experimental results and previous studies,^21,25 we proposed the process of waste PET bottle recycling, as shown in Fig. S9,† which mainly includes three steps, namely: (1) waste PET bottle hydrolysis (red part), (2) electrooxidation of EG (blue part) and (3) the separation of TPA and formic acid (black part).

We have also investigated the energy landscape of electrooxidation ethylene glycol (EG). Here, we proposed the possible reaction steps and the sketch of the electrooxidation process of the EG is shown in Fig. S10.†

The electrooxidation reaction energy (ΔEⁱ_EG) of each step is defined as

CH₂OHCH₂OH* + 2OH⁻ → CH₂OCH₂O* + 2 H₂O + 2e⁻ ΔE¹_EG

CH₂OCH₂O* + 2OH⁻ → CHOCHO* + 2 H₂O + 2e⁻ ΔE²_EG

CHOCHO* → CHO* + CHO* ΔE³_EG

CHO* + 2OH⁻ → HCOO* + H₂O + 2e⁻ ΔE⁴_EG

where * represents the adsorption site and the adsorption energy of EG on the surface was considered the reference. It should be noted we only considered the reaction energy ignoring the vibration contribution to the reaction process.

The calculated energy landscapes of EG electrooxidation on the pristine and Co doped Ni₂P surfaces are shown in Fig. S11.† We calculated the reaction energy of each step based on the proposed EG electrooxidation process. In comparison, removing two protons from EG (ΔE¹_EG) needs 2.193 eV on the pristine Ni₂P surface, it dramatically decreases to 1.243 eV on Co doped Ni₂P surface. However, the value of ΔE²_EG is 1.916 eV, slightly lower than 2.043 eV with Co doping. Therefore, we consider that embedding Co atoms can't reduce the energy step for removing two additional protons. Breaking the C–C bond needs the energy of 3.400 eV on the pristine Ni₂P surface, which is much higher than that on Co doped Ni₂P surface of 1.774 eV. The final step of adsorption of OH⁻ is an exothermic reaction on the pristine and Co doped Ni₂P surface. Overall, breaking the C–C bonds is the energy-determining step for the electrooxidation of EG based on our DFT simulations, and chemical doping of Co should be responsible for enhancing the reaction process and reduce the energy consumption for EG electrooxidation reactions.

PET hydrolysate electroreforming-assisted water splitting

Lastly, we explore the waste PET bottle electroreforming-assisted water splitting in a two-electrode system with Co-Ni₂P/NF as the bifunctional electrode in 1 M NaOH electrolyte with 5 g L⁻¹ PET. As depicted in Fig. 9A, the Co-Ni₂P/NF‖Co-Ni₂P/NF electrolyzer requires an overall voltage of 1.43 V to achieve 10 mA cm⁻², which is much less than that of the pure water electrolysis (1.55 V), indicating the PET electrooxidation-assisted water electrolyzer has higher energy conversion efficiency for H₂ fuel than traditional water electrolyzer. During the electrolysis process, a great deal of H₂ gas bubbles was observed from the cathode and the value-added chemicals (sodium formate and disodium terephthalate) were obtained from the anode (Fig. 9B), which effectively eliminated the potential explosion risk of H₂/O₂ mixture in the traditional water electrolyzer. Additionally, the potential of the Co-Ni₂P/NF‖Co-Ni₂P/NF electrolyzer (1.62 V) was much lower than that of commercial electrocatalysts, i.e., Pt/C/NF‖RuO₂/NF electrolyzer (1.92 V) to achieve current density of 50 mA cm⁻² (Fig. 9C). Furthermore, chronopotentiometric curves display that Co-Ni₂P/NF‖Co-Ni₂P/NF electrolyzer is more stable (with unnoticeable voltage change) than the Pt/C/NF‖RuO₂/NF electrolyzer (voltage increase of 40 mV) during 10 h continuous electrolysis (Fig. 9D).

Fig. 9 (A) Polarization curves of Co-Ni₂P/NF‖Co-Ni₂P/NF electrolyzer in 1 M NaOH with or without 5 g L⁻¹ PET, (B) photograph of the Co-Ni₂P/NF‖Co-Ni₂P/NF electrolyzer during electrolysis process. (C) Polarization curves of the Co-Ni₂P/NF‖Co-Ni₂P/NF and Pt/C/NF‖RuO₂/NF electrolyzers in 1 M NaOH with 5 g L⁻¹ PET, (D) chronopotentiometric curves of the Co-Ni₂P/NF‖Co-Ni₂P/NF and Pt/C/NF‖RuO₂/NF electrolyzers at 10 mA cm⁻².

Conclusion

We have successfully developed cobalt modified nickel phosphide electrocatalyst supported on Ni foam (Co-Ni₂P/NF) by a facile hydrothermal and phosphidation treatment. With a large specific surface area, eminent electrical conductivity, and abundant active sites, the resultant Co-Ni₂P/NF catalyst acts as a highly active and robust bifunctional electrocatalyst for HER and PET electrooxidation. The Co element is atomically dispersed in the Ni₂P structure, forming unsaturated atomic heterometallic Co_xNi_1−xP sites that can effectively regulate the electronic structure of Ni and P atoms. As a result, the catalyst exhibits superior activity for electrooxidation PET hydrolysate in alkaline media, affording 50 mA cm⁻² at a low potential of 1.32 V. A two-electrode electrolyzer with the Co-Ni₂P/NF as a bifunctional electrocatalyst can achieve 10 mA cm⁻² current density at a low voltage of 1.43 V for concurrent PET electroreforming and green H₂ generation. This work not only provides a new facile method for the synthesis of bifunctional electrocatalysts through surface and interface engineering, but also opens a pathway for a sustainable recycling of PET bottles.

Author contributions

Ying Li and Li Quan Lee: methodology, investigation, characterization, and writing—original draft. Hu Zhao: investigation and characterization. Zhi Gen Yu and Yong-Wei Zhang: density functional theory (DFT) calculation. Hong Li and Pingqi Gao: supervision, validation, writing—review & editing, and funding acquisition. All authors have given approval to the final version of the manuscript.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was supported by the Nanyang Technological University (Grant no. NTU-ACE2021-02) and Natural Science Foundation for Distinguished Young Scholars of Guangdong Province (Grant no. 2019B151502053), and Chinese Scholarship Council is acknowledged for providing financial support to Ying Li as a Visiting PhD Student at the Nanyang Technological University.

References

1. T. N. Veziroğlu and S. Şahin, Energy Convers. Manage., 2008, 49, 1820–1831.
2. H. Balat and E. Kırtay, Int. J. Hydrogen Energy, 2010, 35, 7416–7426.
3. M. Li, Z. Xu, Y. Li, J. Wang and Q. Zhong, Nano Select, 2021, 3, 147–156.
4. N. Grimaldos-Osorio, F. Sordello, M. Passananti, P. Vernoux and A. Caravaca, J. Power Sources, 2020, 480, 228800.
5. S. Shiva Kumar and V. Himabindu, Mater. Sci. Energy Technol., 2019, 2, 442–454.
6. H. Li, H. Li, Y. Qiu, S. Liu, J. Fan and X. Guo, Nano Select, 2021, 2, 2148–2158.
7. A. Buttler and H. Spliethoff, Renewable Sustainable Energy Rev., 2018, 82, 2440–2454.
8. H. Okada, E. Tsuji, M. Hisada, S. Kitano, H. Habazaki, S. Suganuma and N. Katada, Sustainable Energy Fuels, 2022, 6, 2709–2717.
9. K. Ye, G. Wang, D. Cao and G. Wang, Top. Curr. Chem., 2018, 376, 42.
10. Y. Li, Z. Dang and P. Gao, Nano Select, 2021, 2, 847–864.
11. G. Yang, Y. Jiao, H. Yan, Y. Xie, A. Wu, X. Dong, D. Guo, C. Tian and H. Fu, Adv. Mater., 2020, 32, 2000455.
12. H. Zhao, D. Lu, J. Wang, W. Tu, D. Wu, S. W. Koh, P. Gao, Z. J. Xu, S. Deng, Y. Zhou, B. You and H. Li, Nat. Commun., 2021, 12, 2008.
13. S. Sreekantan, A. A. Balachandran Kirali and B. Marimuthu, Sustainable Energy Fuels, 2022, 6, 3681–3689.
14. Y. Li, Y. Zhao, F. M. Li, Z. Dang and P. Gao, ACS Appl. Mater. Interfaces, 2021, 13, 34457–34467.
15. Q. Sun, Y. Li, J. Wang, B. Cao, Y. Yu, C. Zhou, G. Zhang, Z. Wang and C. Zhao, J. Mater. Chem. A, 2020, 8, 21084–21093.
16. Y. Zhao, S. Xing, X. Meng, J. Zeng, S. Yin, X. Li and Y. Chen, Nanoscale, 2019, 11, 9319–9326.
17. Y. Ding, B.-Q. Miao, S.-N. Li, Y.-C. Jiang, Y.-Y. Liu, H.-C. Yao and Y. Chen, Appl. Catal., B, 2020, 268, 118393.
18. M. Yao, J. Ge, B. Sun, J. Hu, S. W. Koh, Y. Zhao, J. Fei, Z. Sun, W. Hong, Z. Chen, W. Hu and H. Li, Chem. Eng. J., 2021, 414, 128753.
19. Q. Xue, X. Y. Bai, Y. Zhao, Y. N. Li, T. J. Wang, H. Y. Sun, F. M. Li, P. Chen, P. Jin, S. B. Yin and Y. Chen, J. Energy Chem., 2022, 65, 94–102.
20. L. Wang, G. A. Nelson, J. Toland and J. D. Holbrey, ACS Sustainable Chem. Eng., 2020, 8, 13362–13368.
21. H. Zhou, Y. Ren, Z. Li, M. Xu, Y. Wang, R. Ge, X. Kong, L. Zheng and H. Duan, Nat. Commun., 2021, 12, 4679.
22. X. Jiao, K. Zheng, Q. Chen, X. Li, Y. Li, W. Shao, J. Xu, J. Zhu, Y. Pan, Y. Sun and Y. Xie, Angew. Chem., Int. Ed., 2020, 59, 1–6.
23. J. Wang, X. Li, T. Zhang, Y. Chen, T. Wang and Y. Zhao, J. Phys. Chem. Lett., 2022, 13, 622–627.
24. H. Wang, Y. Li, R. Wang, B. He and Y. Gong, Electrochim. Acta, 2018, 284, 504–512.
25. J. Wang, X. Li, M. Wang, T. Zhang, X. Chai, J. Lu, T. Wang, Y. Zhao and D. Ma, ACS Catal., 2022, 12, 6722–6728.
26. J. P. Perdew, K. Burke and M. Ernzerhof, Phys. Rev. Lett., 1996, 77, 3865–3868.
27. P. E. Blöchl, Phys. Rev. B, 1994, 50, 17953–17979.
28. G. Kresse and J. Furthmüller, Comput. Mater. Sci., 1996, 6, 15–50.
29. J. Klimeš, D. R. Bowler and A. Michaelides, Phys. Rev. B, 2011, 83, 195131.
30. J. Liu, C. Li, Q. Ye, L. Lin, Y. Wang, M. Sun and Y. Cheng, Catal. Commun., 2022, 163, 106416.
31. S. Li, Y. Zou, L. Li, Y. Dong and Y. Zhu, ChemistrySelect, 2019, 4, 6295–6303.
32. T. Wang, X. Cao and L. Jiao, eScience, 2021, 1, 69–74.
33. J. Xu, Y. Zhang, Z. Liu, Y. H. Chen and Z. Liu, Mater. Today Commun., 2021, 29, 102731.
34. Y. Liu, B. Wang, K. Srinivas, M. Wang, Z. Chen, Z. Su, D. Liu, Y. Li, S. Wang and Y. Chen, Int. J. Hydrogen Energy, 2022, 47, 12903–12913.
35. Y. Li, F. Li, Y. Zhao, S. N. Li, J. H. Zeng, H. C. Yao and Y. Chen, J. Mater. Chem. A, 2019, 7, 20658–20666.
36. J. Sun, M. Ren, L. Yu, Z. Yang, L. Xie, F. Tian, Y. Yu, Z. Ren, S. Chen and H. Zhou, Small, 2019, 15, 1804272.
37. X. Yang, A. Y. Lu, Y. Zhu, M. N. Hedhili, S. Min, K. W. Huang, Y. Han and L. J. Li, Nano Energy, 2015, 15, 634–641.
38. L. Ji, J. Wang, X. Teng, T. J. Meyer and Z. Chen, ACS Catal., 2019, 10, 412–419.
39. Y. Li, H. Zhang, M. Jiang, Y. Kuang, X. Sun and X. Duan, Nano Res., 2016, 9, 2251–2259.
40. T. J. Wang, Y. C. Jiang, J. W. He, F. M. Li, Y. Ding, P. Chen and Y. Chen, Carbon Energy, 2022, 4, 283–293.
41. C. X. Zhang, D. Li and Y. Xu, J. Mater. Res., 2022, 37, 807–817.
42. T. Kong, Y. W. Sui, J. Q. Qi, F. X. Wei, Y. J. Ren, Z. Z. Zhan, Z. Sun, M. H. Zhou, D. M. Meng, L. J. Zhang, L. Ma and Q. Wang, J. Mater. Sci.: Mater. Electron., 2021, 32, 1951–1961.
43. H. Jin, S. Liu, L. Pei, G. Li, Z. F. Ma, W. F. Bai, S. T. Wu, Y. J. Yuan and J. S. Zhong, RSC Adv., 2021, 11, 22467–22472.
44. W. L. Ding, Y. H. Cao, H. Liu, A. X. Wang, C. J. Zhang and X. R. Zheng, Rare Met., 2021, 40, 1373–1382.
45. W. Zhang, Y. Zou, H. Liu, S. Chen, X. Wang, H. Zhang, X. She and D. Yang, Nano Energy, 2019, 56, 813–822.
46. Z. Zou, X. Wang, J. Huang, Z. Wu and F. Gao, J. Mater. Chem. A, 2019, 7, 2233–2241.
47. F. Wang, Y. M. Liu and C. Y. Zhang, New J. Chem., 2019, 43, 4160–4167.
48. Z. Li, J. Shang, C. Su, S. Zhang, M. Wu and Y. Guo, J. Fuel Chem. Technol., 2018, 46, 473–478.
49. M. Miao, R. Hou, Z. Liang, R. Qi, T. He, Y. Yan, K. Qi, H. Liu, G. Feng and B. Y. Xia, J. Mater. Chem. A, 2018, 6, 24107–24113.
50. L. Zeng, K. Sun, X. Wang, Y. Liu, Y. Pan, Z. Liu, D. Cao, Y. Song, S. Liu and C. Liu, Nano Energy, 2018, 51, 26–36.
51. M. Jiang, Y. Li, Z. Lu, X. Sun and X. Duan, Inorg. Chem. Front., 2016, 3, 630–634.
52. A. Sivanantham, P. Ganesan and S. Shanmugam, Adv. Funct. Mater., 2016, 26, 4661–4672.
53. S. Wang, J. Li, R. Zhang and L. Li, Mater. Lett., 2017, 198, 61–64.
54. X. H. Ji, Z. M. Wang, X. Y. Chen and R. B. Yu, Chem. J. Chin. Univ., 2021, 42, 1377–1394.
55. S. Huo, S. Yang, Q. Niu, F. Yang and L. Song, Int. J. Hydrogen Energy, 2020, 45, 4015–4025.
56. X. Jin, J. Li, Y. Cui, X. Liu, K. Wang, Y. Zhou, W. Yang, X. Zhang, C. Zhang, X. Jiang and B. Liu, Int. J. Hydrogen Energy, 2019, 44, 5739–5747.
57. B. Hinnemann, P. G. Moses, J. Bonde, K. P. Jørgensen, J. H. Nielsen, S. Horch, I. Chorkendorff and J. K. Nørsko, J. Am. Chem. Soc., 2005, 127, 5308–5309.
58. S. Yang, J. Y. Zhu, X. N. Chen, M. J. Huang, S. H. Cai, J. Y. Han and J. S. Li, Appl. Catal., B, 2022, 304, 120914.
59. H. Sun, Z. Yan, F. Liu, W. Xu, F. Cheng and J. Chen, Adv. Mater., 2020, 32, 1806326.
60. Z. B. Li, J. Wang, X. J. Liu, R. Li, H. Wang, Y. Wu, X. Z. Wang and Z. P. Lu, Scr. Mater., 2019, 173, 51–55.
61. J. Liu, D. Zhu, Y. Zheng, A. Vasileff and S.-Z. Qiao, ACS Catal., 2018, 8, 6707–6732.
62. M. Guo, Y. Qu, F. Zeng and C. Yuan, Electrochim. Acta, 2018, 292, 88–97.
63. J. K. Das, A. K. Samantara, S. Satyarthy, C. S. Rout and J. N. Behera, RSC Adv., 2020, 10, 4650–4656.
64. R. Boppella, J. Tan, W. Yang and J. Moon, Adv. Funct. Mater., 2018, 29, 1807976.
65. W. Li, D. Xiong, X. Gao, W.-G. Song, F. Xia and L. Liu, Catal. Today, 2017, 287, 122–129.
66. Y. Ji, J. Xie, Y. Yang, X. Fu, R. Sun and C. Wong, Chin. Chem. Lett., 2020, 31, 855–858.
67. C. Du, L. Yang, F. Yang, G. Cheng and W. Luo, ACS Catal., 2017, 7, 4131–4137.

---

Footnotes

† Electronic supplementary information (ESI) available. See https://doi.org/10.1039/d2se01007k

‡ Equal contribution.

---