Hydrocracking of Polyethylene to Gasoline-Range Hydrocarbons over a Ruthenium-Zeolite Bifunctional Catalyst System with Optimal Synergy of Metal and Acid Sites

Abstract

Chemical recycling of plastic waste, especially polyolefins, into valuable liquid fuels is of considerable significance to address the serious issues raised by their threat on environmental and human health. Nevertheless, the construction of efficient and economically viable catalytic systems remains a significant hurdle. Herein, we developed an efficient bifunctional catalyst system comprising γ-Al₂O₃-supported ruthenium nanoparticles (Ru/γ-Al₂O₃) and β-zeolite for the conversion of polyolefins into gasoline-range hydrocarbons. A yield of C_5–12 paraffins up to 73.4% can be obtained with polyethene as the reactant at 250 °C in hydrogen. The Ru sites primarily activate the initial cleavage of C–H bonds of polymer towards the formation of olefin intermediates, which subsequently go through further cracking and isomerization over the acid sites in β-zeolite. Employing in situ infrared spectroscopy and probe–molecule model reactions, our investigation reveals that the optimized proportion and spatial distribution of the dual catalytic sites are pivotal in the tandem conversion process. This optimization synergistically regulates the cracking kinetics and accelerates intermediate transfer, thereby minimizing the production of side C_1–4 hydrocarbons resulting from over-cracking at the Ru sites and enhancing the yield of liquid fuels. This research contributes novel insights into catalyst design for the chemical upgrading of polyolefins into valuable chemicals, advancing the field of plastic waste recycling and sustainable chemical production.

1. Introduction

The surge of plastic waste due to the rapid development of human life has led to gradually increasing threats on the environment and human health [1,2,3]. The predominated polyolefins in the wastes featuring stable C–C and the C–H skeleton are subject to difficult natural degradation [4,5]. The common disposal methods via landfilling and incineration have raised serious concerns regarding the accumulation of microplastics and excessive carbon emissions. In this context, it is urgent to achieve manual recycling of polyolefins, in which chemical upgrading through hydrocracking with green hydrogen is of prominent significance, as it can be efficiently conducted under relatively moderate conditions with considerable production of valuable chemicals and fuels [6,7,8,9].

Noble metals, such as Pt and Ru, are widely used in the hydrocracking of polyolefins due to their ability to efficiently activate hydrogen molecules and catalyze the cleavage of both C–C and C–H, which are prevalent in polyolefin monomers [10,11,12,13]. However, despite their significant hydrogenation capabilities, catalysts based on these noble metals often face challenges in optimizing product selectivity by controlling reaction and diffusion kinetics [14,15,16]. Thus, aluminosilicate zeolite materials featuring ordered porous structures have been commonly introduced into the catalytic systems as a second active component to enhance the preferential production of specific hydrocarbons. The confined microenvironment and abundant acid sites are conducive to modulating the transferring and transformation of certain intermediates via imposing spatial restriction and inducing the formation of specific carbocations, whereby prompting the cracking, oligomerization, isomerization, aromatization, and other complex reactions [17,18,19]. In these cases, a key challenge lies in rationally arranging the metal and acid sites within multifunctional catalytic systems to establish enhanced synergy, ensuring not only high atomic utilization efficiency of noble metals but also precise control over reaction pathways. Several recent studies have employed bifunctional catalysts with Pt metal center and acid sites in close contact for reinforced tandem processes to enhance the production of liquid fuels [20,21,22], whereas Li et al. also reported a Pt@S-1 + Beta zeolite with metal and acid sites spatially separated for well-matched reaction steps delivering narrowed product distribution [23]. Owing to the higher activity of Ru for the C–C scission under desired moderate conditions, the Ru-based bifunctional catalysts for polyolefin hydrocracking have also been studied recently [24,25,26], in which the interplay between Ru and acid sites has not been fully understood, especially regarding the production of liquid fuels. The excessive terminal C–C scission over Ru sites would always lead to more methane with high H₂ consumption that undermines the selectivity of value-added liquid hydrocarbons [27,28], which could be potentially circumvented by constructing metal–acid bifunctional catalysts and rationally tuning the synergy between them.

Herein, we developed a bifunctional catalytic system containing Ru nanoparticles supported on γ-Al₂O₃ and abundant acid sites within β-zeolite for hydrocracking polyolefins. It is disclosed that spatially separated Ru and acid sites with an appropriate proportion are conducive to the effective cleavage of C–H and C–C bonds. The mechanism investigation through in situ infrared spectroscopy and probe–molecule model reactions implies that the optimal synergy of Ru and acid sites considerably contributes to the enhanced tandem process involves initial C–H bond activation and pre-cracking over Ru/γ-Al₂O₃, followed by transferring to the acid sites within β-zeolite for subsequent β-scission and isomerization to produce gasoline-range hydrocarbons, while inhibiting over-cracking of intermediates over Ru sites. This catalytic system not only significantly boosts the yield of C_5–12 gasoline-range hydrocarbons up to 73.4% for the low-density polyethylene (LDPE) hydrocracking at 250 °C but also offers a promising platform for the conversion of other ubiquitous polyolefins, such as linear low-density polyethylene (LLDPE) and polypropylene (PP). This work may offer new insights into the design of Ru-based bifunctional catalysts potentially applied in the chemical upgrading of polyolefin waste into valuable chemicals.

2. Results and Discussion

2.1. Catalytic Performance

A series of Ru-based catalysts were evaluated in a stainless steel batch reactor under a high-pressure hydrogen atmosphere, where a mixture of LDPE and catalyst was placed and heated to a certain reaction temperature for the hydrocracking reactions (details in Section 3). The performances of LDPE hydrocracking and the product distribution over various catalysts are shown in Figure 1A, Figures S6 and S7. The mixture of β-zeolite and SiO₂ exhibits the conversion of LDPE as low as 17% and a broad product distribution with 37.4% C_5–12 branched alkanes, while the mixture of Ru/γ-Al₂O₃ and SiO₂ delivers a significantly higher conversion at 51.8% with predominate straight-chain alkanes beyond C₁₃ (52.8%). Ru/γ-Al₂O₃ combined with β-zeolite, in contrast, not only enhances the conversion of LDPE up to 56.1% but also dramatically narrows the product distribution and promotes the selectivity of C_5–12 gasoline branched liquid fuels as high as 63.9%. Interestingly, despite possessing similar active components, the catalyst via loading Ru onto β-zeolite and mixing with γ-Al₂O₃ (Ru/β(25) + γ-Al₂O₃) results in much lower conversion and selectivity compared with the above Ru/γ-Al₂O₃ + β-zeolite. By calculating the TOF on Ru/γ-Al₂O₃ + β(25) and Ru/β(25) + γ-Al₂O₃, they are 78.8 g_Gasline h⁻¹ g_Ru⁻¹ and 50.5 g_Gasline h⁻¹ g_Ru⁻¹ respectively. This result further demonstrates that separating Ru sites from acid sites can promote Ru-zeolite synergy and correspondingly facilitate the hydrogenolysis of polyolefins. In addition, the bifunctional catalysts comprising Ru/γ-Al₂O₃ and other zeolites (MOR, ZSM-5, and Y) are also evaluated and exbibit inferior performances in both the conversion of LDPE and selectivity of C_5–12 hydrocarbons (Figure S6), suggesting the critical importance of zeolite topology and acid properties.

We also examined the performance of the optimal Ru/γ-Al₂O₃ + β-zeolite system under varying conditions, including stirring speed, pressure, and temperature, all of which significantly influence batch reactions. As shown in Figure S8A, the mass transfer of reactants in the viscous reaction system enhances with the increase of stirring speed from 500 to 600 revolutions per minute (rpm), especially at the initial stage, leading to improved conversion of LDPE. However, further increasing the stirring speed to 700 rpm slightly reduces both conversion and C_5–12 hydrocarbon selectivity, likely due to decreased reactant residence time on the catalyst surface [39], which negatively impacts gas–liquid phase mass transfer. Additionally, raising the hydrogen pressure from 1 MPa to 3 MPa significantly improves the conversion of LDPE from 53.5% to 72.5% and the selectivity of the C_5–12 product from 44.5% to 75.4% (Figure S8B). This may be contributed by a positive shift of the thermodynamic equilibrium and enriched active hydrogen species on the catalyst surface. However, further increasing hydrogen pressure to 4 MPa degrades the conversion of LDPE to 69.1% and the selectivity of C_5–12 to 70.8%, which may be caused by competitive adsorption of hydrogen and reactants on the catalyst surface [40]. Therefore, the optimal stirring speed and hydrogen pressure are 600 rpm and 3 MPa H₂, respectively. The conversion of LDPE almost linearly increases from 58.0% to 98.2% with the reaction temperature rising up from 240 °C to 270 °C, yet the selectivity of C_5–12 hydrocarbons goes through an initial dramatic elevation from 15.7% to 78.3% and a subsequent gradual decrease to 67.0% with a surge of C_1–4 hydrocarbons up to 33.0% (Figure 1B). It is plausible that the increased temperature would kinetically accelerate the cleavage of C–H and C–C bonds, whereas the excessively high temperature could also prompt intensive side reactions due to the iterative and uncontrollable cracking, leading to short-chain products [41].

We further monitored the catalytic performance of Ru/γ-Al₂O₃ + β-zeolite under optimal operation conditions along the extended reaction time from 0.5 h to 8 h. As shown in Figure 1C, approximately 50% LDPE conversion was achieved within 30 min, with C₂₂₊ as the primary product. Subsequently, the conversion keeps increasing with gradually decreased mean carbon number of hydrocarbon products. The highest selectivity of C_5–12 products up to 80.7% is achieved after a 6 h reaction, while further extending the reaction time to 8 h leads to possible secondary cracking of C_5–12 gasoline alkanes towards C_1–4 gaseous hydrocarbons due to the excessive residue time of intermediates and products on the catalysts. These results indicate that the hydrocracking reactions follow a typical tandem pathway involving consecutive C–C bond cleavage, transforming long-chain polymers into hydrocarbons with significantly shorter carbon chains. In this context, controlling the reaction time is crucial for regulating the product distribution under given conditions.

We also examined the recyclability through the cyclic experiments. As shown in Figure 1D, the conversion of LDPE and selectivity of C_5–12 gasoline hydrocarbons in the second use show a little decrease by 2.8% and 6.5%, respectively, while only the selectivity of C_5–12 is subjected to a significant decline by 13.8% in the third use. However, in the fourth use, the conversion of LDPE decreases by ~20% with even more dramatically decreased selectivity of C_5–12 hydrocarbons by ~65% compared to the first use. Instead, the distribution of the products is shifted towards predominantly heavier C₂₂₊ hydrocarbons. We rationally attribute this deactivation to the possible aggregation of Ru particle and carbon deposition on the catalyst (Figures S14 and S15).

The Ru/γ-Al₂O₃ + β-zeolite bifunctional catalyst demonstrates versatility in its application to the hydrocracking of various polyolefin feedstocks, including LLDPE and PP, achieving substantial yields of C_5–12 hydrocarbons, reaching 66.7% and 74.3%, respectively, as evidenced in Figure 1E. In comparison to the cutting-edge Ru-based catalysts recently reported in the literature, our developed catalyst maintains a superior specific activity under analogous moderate reaction conditions, specifically for the synthesis of liquid fuels. This superior performance underscores the high atom utilization efficiency of Ru within our catalytic system, positioning it as a promising candidate for potential industrial applications, as further illustrated in Figure 1F.

2.2. Effects of Metal–Acid Synergy

The above results of catalytic performance have indicated that the cooperation of metal Ru and acid sites of β-zeolite plays a critical role in enhancing the production of gasoline-range hydrocarbons from polyolefin hydrocracking under specific conditions. To gain deep insight into the underlying mechanisms of such synergy between Ru metal acid sites for more rational design and optimization of the bifunctional catalysts, we employed advanced characterization techniques to further investigate the detailed structure–activity relationship.

A series of xRu/γ-Al₂O₃ (x represents the loading of Ru) catalysts with varied Ru loadings from 0.6 wt% to 3.0 wt% were prepared, and the actual Ru contents were quantified by inductively coupled plasma atomic emission spectroscopy (ICP-AES) and summarized in Table S2. Transmission electron microscopy (TEM) tests were conducted to characterize the morphology of xRu/γ-Al₂O₃. As shown in Figure 2A,B and Figure S10, all samples demonstrate Ru nanoparticles uniformly dispersed on γ-Al₂O₃, with the mean particle size increasing from 0.97 nm to 1.39 nm as the Ru loading increases from 0.6 wt% to 3.0 wt%. X-ray diffraction (XRD) patterns (Figure 2C) of all samples exhibit prominent characteristic diffraction peaks of the support γ-Al₂O₃, while no significant diffraction peaks attributed to metallic Ru or Ru oxides can be distinguished, which might be ascribed to the high dispersion and small particle size of Ru species (<2 nm) [42]. We also measured the specific area and pore structure of xRu/γ-Al₂O₃ samples by N₂ physical adsorption and desorption tests (Figure 2D, Table S1). The incorporation of Ru species has a minimal effect on the specific surface area and pore volume of the γ-Al₂O₃ support, with the mesoporous structure remaining intact, as evidenced by the similar hysteresis loops observed at a relative pressure range of P/P⁰ from 0.8 to 1.0.

To elucidate the influence of acid properties on the catalytic performance, we further introduced several β-zeolites (β(y), y represents the ratio of SiO₂/Al₂O₃) of different Si/Al ratios. XRD patterns of all zeolites exhibit characteristic diffraction peaks consistent with the database (Figure 2E). N₂ physisorption–desorption isotherms of all samples show a steep increase at a relative pressure P/P⁰ of approximately 0, indicating the filling of micropores (Figure 2F). The significant hysteresis loop at P/P⁰ = 0.6–1.0 for β(25) manifests the presence of mesopores probably derived from the aggregation of nanosized zeolite crystals, while there are only a few mesopores for β(38) and β(360), as verified by the much smaller hysteresis loops. Correspondingly, the surface area and pore volume of these zeolites also have certain discrepancies, as summarized in Table S1.

First, we employed xRu/γ-Al₂O₃ catalysts with varying Ru loadings, combined with the same β(25) catalyst, to investigate the structure–activity relationship of Ru sites in the conversion of LDPE. The catalytic performance results are shown in Figure 3A. With the Ru loading increased from 0.6 wt% to 3.0 wt%, the conversion of LDPE keeps increasing from 28.6% to 91.0%, while the selectivity of C_5–12 hydrocarbons also grows from 21.8% to the highest value of 80.7%. In contrast, the specific activity normalized by the Ru mass goes through an initial elevation from 32.5 g g_Ru⁻¹ h⁻¹ at 0.6 wt% to 100.7 g g_Ru⁻¹ h⁻¹ at 1.8 wt%, followed by a continuous decrease to 84.1 g g_Ru⁻¹ h⁻¹ at 3.0 wt% (Figure 3A). In situ CO adsorption infrared spectroscopy (CO-IR) was conducted to investigate the structure of surface Ru sites (Figure 3B). All xRu/γ-Al₂O₃ catalysts exhibit three similar ν_CO bands at 2120 cm⁻¹, 2060 cm⁻¹, and 1990 cm⁻¹. The bands at 2120 cm⁻¹ attributed to tricarbonyl CO species adsorption on partially oxidized Ruⁿ⁺ species (Ruⁿ⁺(CO)₃, n = 1–3). The band centered at 2060 cm⁻¹ consists of two characteristic peaks at 2068 cm⁻¹ and 2050 cm⁻¹ (the fitting results are shown in Figure S11), which the peak at 2068 cm⁻¹ is attributed to dicarbonyl CO species adsorbed on the metallic Ru (Ru⁰-(CO)₂), and the peak at 2050 cm⁻¹ is assigned to linearly adsorbed CO on metallic Ru (Ru⁰-CO) [43,44,45,46]. By deconvolving and fitting the four bands (Figure S11 and Table S3), it is evident that metallic Ru species dominate the catalyst surface, with their proportion only slightly increasing from 92.4% to 93.6% as the Ru loading rises from 0.6 wt% to 3.0 wt%. This indicates that pre-reduction in an H₂ atmosphere effectively generates metallic Ru surfaces in all cases, which are likely the active centers responsible for the hydrocracking reactions.

To elucidate the chemical state of Ru in the xRu/γ-Al₂O₃, Ru 3p_3/2 X-ray photoelectron spectroscopy (XPS) characterization was performed. As shown in Figure 3C, the signals of xRu/γ-Al₂O₃ can be deconvolved and fitted into two sets of peaks. The peaks at 462.5 eV can be attributed to the metallic Ru (Ru⁰), while the minor peaks at 465.7 eV correspond to partially oxidized Ru (Ru^δ+) [47,48]. The proportion of Ru species was calculated based on the integrated peak area and summarized in Table S3. With Ru loading increasing from 0.6 wt% to 3.0 wt%, the proportion of metallic Ru⁰ species increased from 50.9% to 79.0%. These values are significantly lower than those measured in CO-IR, which should be ascribed to the deeper detection distance of XPS (~3 nm) in comparison with the CO-IR only involving outermost atomic layers of catalysts. Additionally, quite a few small metallic Ru⁰ species are subject to easier oxidization under atmospheric conditions during the transfer and preparation process of samples for XPS. According to the TEM results, the size of Ru nanoparticles also increases from 0.97 nm to 1.39 nm as the Ru loading increases. We can reasonably correlate the increased metallic Ru⁰ species with the intensified aggregation of Ru particles. Ru⁰ can realize hydrocracking of polyolefin by dissociating hydrogen into active H atoms and reducing the activation energy of C-C bond breakage [49,50]. With the increase of Ru loading, the proportion of Ru⁰ increases, which is more conductive to the activation of C–H and C–C bonds. In this context, the xRu/γ-Al₂O₃ catalysts with low Ru loading of 0.6 wt% and 1.2 wt% exhibit lower specific activity compared to those with higher Ru loading, despite the enhanced accessibility of Ru atoms in Ru/γ-Al₂O₃ with smaller particle sizes. In addition, the emergence of oxidized Ru species can be attributed to the charge transfer at the metal–support interface, which shows an enhancement at low Ru loading. The relatively strong interaction between Ru and γ-Al₂O₃ facilitate the electron transfer from Ru to support Ru-O-Al bonds [51,52], whereas the electron-deficient Ru species may not favor the activation of H₂ molecules and C–H and C–C bonds in polyolefins, as suggested by previous studies [53,54,55]. The Ru^δ+, which is more prevalent at low Ru loading, mainly promotes the fracture of the internal C–C bond and thus inhibits the generation of methane [15,56]. This is because the internal C atoms have a higher electron density than terminal C atoms, making Ru^δ+ more inclined to bind to the internal C atoms. Consequently, the tendency of terminal C–C bond breakage and the selectivity of methane increase with increasing Ru loading.

To investigate the effect of acid sites, we further employed the β-zeolites of different Si/Al ratios with the same 3.0Ru/γ-Al₂O₃ in the conversion of LDPE. As shown in Figure 4A, increasing the Si/Al ratio of β-zeolite from 25 to 360 dramatically reduces the conversion of LDPE from 72.5% to 52.5% and the selectivity of gasoline hydrocarbons (C_5–12) from 75.4% to 12.9%. NH₃-temperature programmed desorption (NH₃-TPD) characterization reveals that all β-zeolites (Figure 4B) exhibit two desorption peaks within a range of 100 °C and 500 °C, corresponding to the weak and strong acid centers. The total acid amounts are calculated to be 1198 mmol g⁻¹, 1059 mmol g⁻¹, and 205 mmol g⁻¹ for β(25), β(38), and β(360), respectively (Table S4). Pyridine-infrared (Py-IR) spectroscopy was employed to discriminate different acid centers in β-zeolites, where the peaks at 1454 cm⁻¹ and 1544 cm⁻¹ are attributed to the adsorption of pyridine on Lewis and Brønsted acid sites, respectively. As shown in Figure 4C and summarized in Table S4, β(360) zeolite displays notably reduced peak intensities at 1454 cm⁻¹ and 1544 cm⁻¹, indicative of a minimal presence of acid sites attributable to the lower proportion of framework Al atoms [57]. In contrast, there is almost the same area of the peaks at 1544 cm⁻¹ for β(25) and β(38), while the peak of β(25) at 1454 cm⁻¹ is significantly higher. This indicates that there is little discrepancy in the number of Brønsted acid sites for β(25) and β(38), corresponding to the similar amount of strong acid sites, as shown in the NH₃-TPD results. Rationally correlating these results with the catalytic activity shown in Figure 4A, it is found that the productivity of C_5–12 hydrocarbons from LDPE hydrocracking is primarily enhanced by the strong Brønsted acid sites, while the moderate Lewis acid sites show minor effect on the apparent performance, as β(25) and β(38) with distinct amounts of Lewis acid sites offer quite similar activity for the production of C_5–12 hydrocarbons. It is plausible that the carbocation-based reactions controlled by the Brønsted acid centers impose a prominent effect on the kinetic behaviors of the intermediates within the β-zeolites.

Besides the optimization of individual Ru/γ-Al₂O₃ and β-zeolites, the synergy could also be influenced by the proportion of dual components. As shown in Figure 4D, the sole 1.8Ru/γ-Al₂O₃ gives the highest conversion of LDPE up to 99.5%, yet exhibiting poor selectivity of C_5–12 hydrocarbons at 23.4%. With the ratio of β-zeolites increasing, the conversion of LDPE demonstrates a gradual decline, while the selectivity of C_5–12 hydrocarbons goes through an initial promotion to 75.4% at the conversion of 72.5% and a subsequent decrease to 64.6% at the conversion of 25.9%. The yield of C_5–12 gasoline hydrocarbons reaches the highest value of 54.7% with 1.8Ru/γ-Al₂O₃ and β-zeolite mixed in a 1:1 ratio. These results imply that Ru/γ-Al₂O₃ primarily contributes to the hydrocracking reaction, as it possesses much higher activity than that of β-zeolite. The introduction of β-zeolite might not only catalyze the carbocation-based cracking reactions but also tailor the products’ distribution through regulating the kinetics behaviors regarding the transferring and transformation of the intermediates.

2.3. Reaction Mechanisms

Now that we have demonstrated that the synergy of metal–acid sites should play a vital role in the catalytic hydrocracking of LDPE, the corresponding reaction mechanisms should also be elucidated for the rational design of catalysts specialized in the critical kinetic process. We first studied the hydrocracking of LDPE at the initial reaction stage to obtain the performance close to kinetic region. As depicted in Figure 5A, during the initial 0.5 h of reaction, the conversion of LDPE and the product distribution on the Ru/γ-Al₂O₃ + β-zeolite and Ru/γ-Al₂O₃ + SiO₂ catalysts are nearly identical. In contrast, the β-zeolite + SiO₂ catalyst exhibits extremely poor activity. These observations indicate that the majority of LDPE is initially activated and transformed on the Ru/γ-Al₂O₃ catalyst via a hydrocracking mechanism, attributable to the significantly superior capability of metallic Ru in activating the H₂, C–H, and C–C bonds within the LDPE molecules. In addition, the pore opening of narrow size in the range of 0.5–1.2 nm greatly restricts the diffusion of reactants into the channels of β-zeolite (Figure S12) and the corresponding accessibility of acid sites to the specific C–H and C–C bonds in bulk molecules. Thus, we speculate that the β-zeolites demonstrate negligible activity until smaller hydrocarbon intermediates are formed from LDPE hydrocracking over Ru/γ-Al₂O₃.

This reaction pathway has been further elucidated through the use of C_20–32 n-alkanes as model reactants, with β-zeolite as the catalyst (Figure 5B). The decreased conversion of alkanes from 48.5% to 38.2% with enlengthened carbon chains from C₂₀ to C₃₂ also implies that the acid-catalyzed behaviors in β-zeolite are highly dependent on the molecular size of reactants. A detailed analysis of the product distribution, using C₂₄ alkane as a representative reactant (Figure S9), reveals the absence of C₁ products and the presence of only minimal C₂ products, alongside a significant yield of C₂₂ and C₂₃ hydrocarbons. These findings lead us to infer that long-chain hydrocarbons initially undergo protonation at the acidic sites of β-zeolite, resulting in the formation of carbocation intermediates. These intermediates are primarily subject to skeletal isomerization and β-scission, while the terminal cleavage of C–C bonds is intrinsically suppressed. The formation of C_22–23 hydrocarbons also suggests that there might be oligomerization reactions of short-chain carbocations at Brønsted acid sites [58].

To further comprehend the reaction pathways of LDPE hydrocracking on Ru/γ-Al₂O₃ and β-zeolite catalysts, in situ infrared experiments were employed to monitor the changes of intermediates in reaction. As shown in Figure 5C,D, there are obvious adsorption bands at 2925 cm⁻¹ and 2854 cm⁻¹ assigned to C–H asymmetric and symmetric stretching vibration (ν_as(CH₂) and ν_s(CH₂)) of the LDPE skeleton, respectively, during the hydrocracking of LDPE on both Ru/γ-Al₂O₃ and Ru/γ-Al₂O₃ + β-zeolite [59,60]. Signals assigned to C=C vibration at 1660 cm⁻¹ also emerge even before the onset of the hydrocracking reaction, suggesting that they might be attributed to the formation of alkene intermediates in the heating period in the absence of a H₂ atmosphere. As the reaction proceeds, both peaks assigned to LDPE and alkene intermediates undergo a gradual decline, indicating that the LDPE is converted into volatile hydrocarbons that diffuse away from the catalyst surface rapidly. The quick consumption of alkene implies that the H₂ atmosphere might kinetically promote the transformation of intermediates. Interestingly, we notice that the decrease of C–H peaks is significantly slower on Ru/γ-Al₂O₃ + β-zeolite in comparison with Ru/γ-Al₂O₃, while the descent of C=C peaks almost has an identical rate (Figure 5E,F). This suggests that Ru/γ-Al₂O₃, primarily adept at hydrogen activation, plays a pivotal role in the elimination of unsaturated alkene intermediates. In contrast, β-zeolite is in favor of the transferring of certain intermediates into the channels for more moderate and controllable cracking and isomerization reactions, thereby prolonging the residue time of hydrocarbons in the Ru/γ-Al₂O₃ + β-zeolite and preventing the excessive cracking over Ru/γ-Al₂O₃.

Based on the preceding discourse, we posit a potential reaction mechanism for the bifunctional catalytic system comprising Ru/γ-Al₂O₃ and β-zeolite, as depicted in Figure 6. Ru/γ-Al₂O₃ firstly transform LDPE into long-chain olefin intermediates with its prominent capability to activate H₂ molecules and C–C and C–H bonds. Once the intermediates are tailored to specific dynamic diameters close to the pore opening of β-zeolite, they are diffused into the channels and undergo isomerization, β-scission, and oligomerization reactions at the strong Brønsted acid sites. The optimal acid properties and the confinement effect of the pores and channels in β-zeolite are harnessed to selectively produce C_5–12 hydrocarbons. Within this context, the unsaturated intermediates are subjected to hydrogenation at the Ru metal sites, ultimately resulting in the production of gasoline-ranged alkanes. The meticulously regulated synergy between the Ru and acid sites is chiefly responsible for the high selectivity of the target products, achieved through a controlled cleavage of C–C bonds, thereby mitigating excessive cracking and the resultant formation of C_1–4 gaseous byproducts. This proposed mechanism highlights the sophisticated interplay between the catalytic components and their collective contribution to the enhanced efficiency and selectivity of the chemical recycling process.

3. Materials and Methods

3.1. Materials

RuCl₃∙xH₂O (35.0–42.0% Ru basis), NH₃∙H₂O, cyclohexane (HPLC grade, ≥99.9%), mesitylene (GC standard, ≥99%), C₇–C₄₀ saturated alkane mixture (certified reference material, 1000 μg ml⁻¹ each component in hexane), n-eicosane (AR grade, ≥99%), n-tetracosane (AR grade, 99%), n-octacosane (AR grade, ≥97%), n-dotriacontane (AR grade, ≥98%), and nano-alumina (99.99% metals basis, γ-phase, 20 nm) were purchased from Aladdin Co., Ltd. (Fukuoka, Japan) β zeolites (Ammonium, S.A.) with Si/Al molar ratios of 25, 38, and 360 were purchased from Alfa Co., Ltd. (Shanghai, China). Low-density polyethylene, linear low-density polyethylene, and polypropylene (average Mw250,000 by GPC) were purchased from Macklin Co., Ltd. (Shanghai, China).

3.2. Synthesis of Catalysts

Synthesis of xRu/γ-Al₂O₃ (x represents the loading of Ru): RuCl₃∙xH₂O was dissolved in an aqueous solution containing a small amount of HCl via sonication, which was then mixed with 20 mL of water to obtain a homogeneous solution (denoted as solution A). Then, 2 g of γ-Al₂O₃ (after calcined in air at 400 °C for 2 h) was mixed with 50 mL of water and uniformly dispersed by stirring, and 1.8 mL of ammonia solution was added (denoted ad solution B). The RuCl₃ solution was pumped into solution B using a constant flow pump at a flow rate of 5 mL min⁻¹ and 25 °C under vigorous stirring. The suspension was further stirred at room temperature for 3 h, followed by filtering and drying at 120 °C overnight. The sample was reduced at 400 °C for 2 h in a 10% H₂/Ar atmosphere to remove NH₄Cl and reduce the active metal Ru.

Synthesis of Ru/β-zeolite: β-zeolites were initially calcined in air at 550 °C for 6 h and then dispersed in 100 mL of water, followed by mixing with a certain amount of RuCl₃ aqueous solution under 25 °C stirring in a water bath at 60 °C overnight. The solvent was removed using a rotary evaporator at 60 °C, and the sample was dried in an oven at 120 °C overnight. Before use, it was reduced at 400 °C for 2 h in a 10% H₂/Ar atmosphere.

3.3. Reaction Test

The hydrocracking of polyolefin was conducted in a high-pressure stainless steel vessel reactor (50 mL). The inner diameter and height of the reactor used in the experiment were ϕ34 mm and 78 mm, and the distance between the bottom of the stirring paddle and the thermocouple and the bottom of the reactor was about 5 mm, as shown in Figure S1. The catalysts were pre-ground to less than 60 mesh and the particle size was small enough to eliminate possible internal diffusion effects [61,62]. About 4 g of polyolefin powder was mixed with Ru/γ-Al₂O₃ and zeolite, according to the ratio of reactant to catalyst (10:1). The mixture was ground uniformly in a mortar and then placed into the reactor vessel and sealed. The atmosphere inside the reactor was flushed by the reaction gas (H₂) five times, pressurized to required reaction pressure, and held for a period. The reactor temperature was ramped to 240–270 °C within 0.5 h, while mechanical stirring was performed at a speed of 500–700 rotations per minute (rpm).

After the reaction was complete, the reactor was cooled in an ice-water bath rapidly until the temperature inside the reactor dropped below 20 °C. The gaseous products were collected using a gas sampling bag, analyzed using gas chromatography-flame ionization detection (GC-FID), and quantified with C₁-C₈ n-alkane standard gas. The liquid products were dissolved with cyclohexane as a solvent and mesitylene as an internal standard. The mixture was sonicated and then centrifuged. The supernatant was taken for gas chromatography-mass spectrometry (GC-MS) analysis. The residue was dried at 85 °C and then collected, and the solid was weighed.

For the recyclability test, the cyclohexane extract was centrifuged to obtain the used catalyst, which was then dried at 85 °C to remove cyclohexane. Before each use, the catalyst was reduced at 400 °C for 2 h in a 10% H₂/Ar atmosphere in a tubular furnace.

The conversion, selectivity, degree of branching, carbon balance, and mass balance in this work were calculated as follows. The carbon balance and mass balance were above 80% for all tests. The difference value from 100% is possibly ascribed to the partial carbon deposition on the solid catalyst and inevitable loss of products in the sample transferring steps.

$$\mathrm{C}\mathrm{o}\mathrm{n}\mathrm{v}\mathrm{e}\mathrm{r}\mathrm{s}\mathrm{i}\mathrm{o}\mathrm{n}\left(\%\right)={\displaystyle \frac{\mathrm{m}\mathrm{a}\mathrm{s}\mathrm{s} \mathrm{o}\mathrm{f} \mathrm{i}\mathrm{n}\mathrm{i}\mathrm{t}\mathrm{i}\mathrm{a}\mathrm{l} \mathrm{P}\mathrm{E}-\mathrm{m}\mathrm{a}\mathrm{s}\mathrm{s} \mathrm{o}\mathrm{f} \mathrm{r}\mathrm{e}\mathrm{s}\mathrm{i}\mathrm{d}\mathrm{u}\mathrm{a}\mathrm{l} \mathrm{P}\mathrm{E}}{\mathrm{m}\mathrm{a}\mathrm{s}\mathrm{s} \mathrm{o}\mathrm{f} \mathrm{i}\mathrm{n}\mathrm{i}\mathrm{t}\mathrm{i}\mathrm{a}\mathrm{l} \mathrm{P}\mathrm{E}}}\times 100\%$$

(1)

$$\mathrm{S}\mathrm{e}\mathrm{l}\mathrm{e}\mathrm{c}\mathrm{t}\mathrm{i}\mathrm{v}\mathrm{i}\mathrm{t}\mathrm{y}\left(\%\right)\hspace{0.17em}\mathrm{o}\mathrm{f} {\mathrm{C}}_{\mathrm{i}}={\displaystyle \frac{\mathrm{m}\mathrm{a}\mathrm{s}\mathrm{s} \mathrm{o}\mathrm{f} {\mathrm{C}}_{\mathrm{i}}}{\mathrm{m}\mathrm{a}\mathrm{s}\mathrm{s} \mathrm{o}\mathrm{f} \sum {\mathrm{C}}_{\mathrm{i}}}}\times 100\%$$

(2)

$$\mathrm{D}\mathrm{e}\mathrm{g}\mathrm{r}\mathrm{e}\mathrm{e} \mathrm{o}\mathrm{f} \mathrm{b}\mathrm{r}\mathrm{a}\mathrm{n}\mathrm{c}\mathrm{h}\mathrm{i}\mathrm{n}\mathrm{g}\left(\%\right)={\displaystyle \frac{\mathrm{m}\mathrm{a}\mathrm{s}\mathrm{s} \mathrm{o}\mathrm{f} \sum \mathrm{i}\mathrm{s}\mathrm{o}{\mathrm{C}}_{\mathrm{i}}}{\mathrm{m}\mathrm{a}\mathrm{s}\mathrm{s} \mathrm{o}\mathrm{f} \sum {\mathrm{i}\mathrm{s}\mathrm{o}\mathrm{C}}_{\mathrm{i}}+\mathrm{m}\mathrm{a}\mathrm{s}\mathrm{s} \mathrm{o}\mathrm{f} \sum {\mathrm{n}\mathrm{C}}_{\mathrm{i}}}}\times 100\%$$

(3)

$$\mathrm{C}\mathrm{a}\mathrm{r}\mathrm{b}\mathrm{o}\mathrm{n} \mathrm{b}\mathrm{a}\mathrm{l}\mathrm{a}\mathrm{n}\mathrm{c}\mathrm{e}\left(\%\right)={\displaystyle \frac{\mathrm{m}\mathrm{a}\mathrm{s}\mathrm{s} \mathrm{o}\mathrm{f} \mathrm{c}\mathrm{a}\mathrm{r}\mathrm{b}\mathrm{o}\mathrm{n} \mathrm{i}\mathrm{n} \mathrm{p}\mathrm{r}\mathrm{o}\mathrm{d}\mathrm{u}\mathrm{c}\mathrm{t}}{\mathrm{m}\mathrm{a}\mathrm{s}\mathrm{s} \mathrm{o}\mathrm{f} \mathrm{c}\mathrm{a}\mathrm{r}\mathrm{b}\mathrm{o}\mathrm{n} \mathrm{i}\mathrm{n} \mathrm{i}\mathrm{n}\mathrm{i}\mathrm{t}\mathrm{i}\mathrm{a}\mathrm{l} \mathrm{P}\mathrm{E}}}\times 100\%$$

(4)

$$\mathrm{M}\mathrm{a}\mathrm{s}\mathrm{s} \mathrm{b}\mathrm{a}\mathrm{l}\mathrm{a}\mathrm{n}\mathrm{c}\mathrm{e}\left(\%\right)={\displaystyle \frac{\mathrm{m}\mathrm{a}\mathrm{s}\mathrm{s} \mathrm{o}\mathrm{f} \mathrm{p}\mathrm{r}\mathrm{o}\mathrm{d}\mathrm{u}\mathrm{c}\mathrm{t}+\mathrm{m}\mathrm{a}\mathrm{s}\mathrm{s} \mathrm{o}\mathrm{f} \mathrm{r}\mathrm{e}\mathrm{s}\mathrm{i}\mathrm{d}\mathrm{u}\mathrm{a}\mathrm{l} \mathrm{P}\mathrm{E}}{\mathrm{m}\mathrm{a}\mathrm{s}\mathrm{s} \mathrm{o}\mathrm{f} \mathrm{i}\mathrm{n}\mathrm{i}\mathrm{t}\mathrm{i}\mathrm{a}\mathrm{l} \mathrm{P}\mathrm{E}+\mathrm{m}\mathrm{a}\mathrm{s}\mathrm{s} \mathrm{o}\mathrm{f} {\mathrm{H}}_2 \mathrm{c}\mathrm{o}\mathrm{n}\mathrm{s}\mathrm{u}\mathrm{m}\mathrm{p}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}}}\times 100\%$$

(5)

3.4. Catalyst Characterization

Inductively coupled plasma atomic emission spectrometry (ICP-AES): The content of different elements in the catalysts was determined using a Avio 550 Max (PerkinElmer, Waltham, MA, USA) inductively coupled plasma optical emission spectrometer.

X-ray diffraction (XRD): XRD patterns were recorded using a Empyrean (Malvern Panalytical, Shanghai, China) diffractometer with a Cu Kα (λ = 0.154 nm) radiation source, operating at 40 kV and 40 mA, with a 2θ range of 5–90°.

NH₃-temperature programmed desorption (NH₃-TPD): The samples were pretreated at 400 °C for 1 h under a helium flow of 30 mL min⁻¹, then cooled to 100 °C. After the baseline stabilized, NH₃ was adsorbed to saturation. The samples were then heated to 800 °C at a rate of 10 °C min⁻¹ under a helium flow for TPD experiments. The effluent was detected using mass spectrometry.

CO-pulse adsorption: Before CO injection, the samples were reduced at 400 °C for 1 h under a 10% H₂/Ar flow of 30 mL min⁻¹ with a heating rate of 5 °C min⁻¹. The samples were then purged with helium at 30 mL min⁻¹ for 1 h. CO pulses were sent to the catalyst, and the CO adsorption curves were measured using a thermal conductivity detector until no further CO was adsorbed.

N₂ physical adsorption and desorption: The specific surface area and porosity of the catalysts were determined using Quantachrome instruments through N₂ adsorption–desorption isotherms. The catalysts were degassed at 300 °C for 4–6 h, followed by the acquisition of the adsorption isotherms.

Transmission electron microscopy (TEM): TEM images were obtained using a JEM-2100 F (JEOL Ltd., Tokyo, Japan) microscope with an acceleration voltage of 100 kV. The samples were dispersed in ethanol and dropped onto carbon-coated copper TEM grids.

X-ray photoelectron spectroscopy (XPS): XPS was performed on a Escalab 250 Xi⁺ (Thermo Fisher, Waltham, MA, USA) spectrometer using Al Kα radiation at an operating voltage of 15 kV and a current of 20 mA. The binding energy scale was calibrated by setting the C1s transition to 284.8 eV.

Thermo-gravimetric analysis (TGA): TGA was conducted using an STA449F5-Thermoster (Netzsch, Shanghai, China) thermogravimetric analyzer with air as the carrier gas. Approximately 10 mg of the sample was loaded into a ceramic crucible and placed on the balance. The temperature was programmed to increase, and the weight changes were recorded as a function of temperature and time.

Infrared spectroscopy of pyridine adsorption (Py-IR): The samples were pressed into pellets and treated under vacuum at 450 °C for 0.5 h. Pyridine was introduced into the sample cell at room temperature until adsorption saturation was reached. Then, under high vacuum, the samples were heated at a rate of 10 °C min⁻¹ to 150 °C, 250 °C, 350 °C, and 450 °C for 0.5 h each to desorb unstable adsorbed species at the corresponding temperatures, followed by scanning of the infrared signals of the samples.

CO adsorption–desorption in situ diffuse reflectance infrared Fourier transform spectroscopy (CO-DRIFTS): CO-DRIFTS was collected on a NICOLET iS20 (Thermo Fisher, Waltham, MA, USA) FTIR spectrometer equipped with a mercury cadmium telluride detector. Before testing, the samples were pretreated in 5% H₂/He at 200 °C for 0.5 h and cooled to room temperature under the same atmosphere. Background spectra were collected with a resolution of 4 cm⁻¹ and 32 scans under a He atmosphere. Then, 10% CO/He was introduced into the chamber, and spectra were collected at 20 °C. After adsorption saturation, CO flow was stopped, and the chamber was purged with He while continuing to collect spectra until no bands attributable to gaseous CO remained.

In situ IR Spectroscopy: In situ infrared cell in diffuse reflection mode was used in this experiment, as shown in Figure S2. Infrared spectra were collected by the NICOLET iS20 (Thermo Fisher, Waltham, MA, USA) FTIR spectrometer equipped with a mercury cadmium tellurium detector. The crucible in the in situ cell is ϕ6 × 4.5 mm in size, and the temperature is measured by a thermocouple placed inside the in situ cell and controlled by an external temperature controller. The gas enters the in situ cell through the intake pipe and then directly exits the outlet pipe and is discharged into the tail gas without detection and analysis. LDPE was dissolved in toluene at 100 °C to obtain a solution with concentration of 5 mg LDPE mL⁻¹. A layer of KBr of 5 mm height was placed at the bottom of the crucible, followed by the addition of 10 mg of catalyst. The mixture was then pressed flat and placed in the in situ cell. The catalyst was reduced in situ at 250 °C for 30 min. After reduction, the catalyst background was collected, and the temperature was decreased to 100 °C. The crucible was taken out, and 20 μL of the LDPE solution was gently pipetted onto the catalyst surface in the crucible. The crucible was then reinserted into the in situ cell, which was purged with helium gas at 100 °C for 1 h. The atmosphere was then switched to hydrogen, and the temperature was increased to 250 °C at a rate of 5 °C min⁻¹. In situ IR spectra of LDPE degradation were collected.

4. Conclusions

In this work, we developed a bifunctional catalytic system containing Ru nanoparti-cles supported on γ-Al₂O₃ and abundant acid sites within β-zeolite for hydrocracking of polyolefins. It is disclosed that spatially separated Ru and acid sites with an appropriate proportion is conducive to the effective cleavage of C–H and C–C bonds. The mechanism investigation through in situ infrared spectroscopy and probe–molecule model reactions implies that the optimal synergy of Ru and acid sites considerably contributes to the enhanced tandem process involves initial C–H bond activation and pre-cracking over Ru/γ-Al₂O₃, followed by transfer to the acid sites within β-zeolite for subsequent β-scission and isomerization to produce gasoline-range hydrocarbons while inhibiting over-cracking of intermediates over Ru sites. Through fully leveraging the unique properties of Ru/γ-Al₂O₃ and β-zeolite catalysts, this catalytic system not only achieves a top-level yield of C_5–12 gasoline-range hydrocarbons up to 73.4% for the LDPE hydrocracking at 250 °C but also holds a promise in the conversion of other ubiquitous polyolefins. This work may offer new insights for the design of Ru-based bifunctional catalysts and contribute to the chemical upgrading of polyolefin waste into valuable chemicals.