Catalytic Upcycling of Polyolefins and Polystyrene: Integrating Active Site Design With Mass Transfer Considerations

Abstract

Plastic waste accumulates globally due to its chemical inertness and resistance to degradation, creating both an escalating environmental burden and a squandered fossil‐derived carbon resource. Catalytic upcycling provides a promising route to convert these materials into fuels and chemicals; however, development is constrained by the dual challenge of cleaving robust C–C bonds while processing macromolecular substrates subject to severe diffusion limitations. This review assesses the current landscape of catalytic conversion of polyolefins (PO) and polystyrene (PS), including hydrogenolysis, catalytic cracking, hydrocracking, and selective oxidation, and clarifies that catalytic performance cannot be rationalized or optimized solely based on intrinsic active site chemistry. Instead, the spatial distribution of metal and acid sites, catalyst porosity, susceptibility to deactivation and poisoning, and multiscale transport phenomena collectively dictate apparent activity, selectivity, and stability. By integrating active site design with mass‐transfer considerations, from intrapore diffusion to reactor‐level flow behavior, this review identifies design principles that reconcile high intrinsic activity with mass‐transfer efficiency. Overall, tuning the interplay between catalytic functionality and transport properties emerges as a central requirement for advancing efficient, selective, and scalable plastic upcycling technologies.

1. Introduction

The extensive production, widespread use, and improper disposal of plastic products have resulted in persistent environmental pollution and substantial loss of hydrogen‐carbon resources, as most plastics are derived from fossil‐based feedstocks [1, 2]. Among them, polyethylene (PE), polypropylene (PP), and polystyrene (PS) dominate global plastic production. Their backbones consist exclusively of C–C and C–H bonds [3], which confer exceptional chemical stability and durability but also render them highly resistant to natural degradation, often persisting for centuries in ecosystems [4]. Consequently, polyolefin (PO) and PS waste represents both a severe environmental burden and an underutilized carbon reservoir [5].

Addressing this challenge requires efficient plastic upcycling strategies that transform PO and PS waste into value‐added fuels, chemicals, and materials [6, 7]. From a chemical perspective, the principal conversion routes include catalytic cracking, hydrogenolysis, hydrocracking, and selective oxidation [5, 8]. These processes aim to cleave and reorganize the robust polymeric backbones into smaller hydrocarbons or functionalized molecules, spanning aromatics, oxygenates, and liquid transportation fuels [9]. Compared with purely thermal approaches, catalytic routes offer the potential for lower energy input, improved product selectivity, and tunable reaction pathways [10, 11].

Despite significant progress, catalytic upcycling of PO and PS faces several intrinsic challenges. First, the cleavage of strong C–C bonds generally requires highly active catalytic sites and often relies on noble metals or strongly acidic materials, raising concerns regarding cost, scalability, and sustainability [12]. Second, the macromolecular nature of polymers leads to extremely high melt viscosities, introducing pronounced mass transfer limitations [13]. These limitations manifest not only at the nanoscale, where polymer chains struggle to access active sites, but also at the reactor scale, where heat transfer, multiphase transfer, mixing, and feeding of viscous substrates become nontrivial. Third, real‐world plastic waste streams contain additives, fillers, dyes, antioxidants, and external contaminants such as water or oils [14]. These components, together with inevitable coke formation, can severely accelerate catalyst deactivation and complicate long‐term operation.

To overcome these barriers, the rational design of catalysts and catalytic processes is of central importance. In most PO upcycling reactions, two types of catalytic functionalities are essential. Metal sites primarily mediate reactions involving hydrogen or oxygen activation, including hydrogenation, dehydrogenation, hydrogenolysis, and oxidation [15, 16]. Their electronic structure, dispersion, and interaction with support critically determine reaction rates and selectivity. In parallel, acidic sites—typically originating from zeolites or metal oxides—facilitate carbocation‐mediated pathways, enabling C–C bond cleavage through β‐scission and related rearrangements [17, 18, 19, 20]. These acid‐catalyzed steps are indispensable for cracking and hydrocracking processes but must be carefully regulated to avoid overcracking and coke formation.

Beyond the chemical nature of active sites, the physical architecture of catalysts plays a decisive role when processing macromolecular and highly viscous PO. Catalyst porosity, metal–acid spatial distribution, particle morphology, and external surface accessibility strongly influence polymer diffusion, intermediate transport, and product desorption [21, 22, 23]. Unlike small‐molecule reactants, polymer chains interact with catalyst surfaces in a nonuniform and often diffusion‐limited manner [13]. As a result, mass transfer effects across multiple length scales—from molecular diffusion at active sites to macroscopic flow behavior in reactors—can dominate reaction pathways, apparent kinetics, and catalyst stability. Optimizing mass transfer is therefore not merely a process consideration but an integral component of catalyst and reaction design.

This review aims to provide a comprehensive and up‐to‐date overview of catalytic PO and PS upcycling, with an additional focus on the interplay between catalyst design, stability, and mass transfer regulation, which are considered as three main driving forces for catalytic upcycling of plastic waste (Figure 1). The discussion is structured to bridge fundamental reaction mechanisms with practical process considerations. After a brief overview of relevant reaction pathways, subsequent sections examine catalyst design principles for C–C bond cleavage, including metal‐catalyzed hydrogenolysis, acid‐catalyzed cracking, bifunctional metal–acid catalysts for hydrocracking, and selective oxidation routes. Catalyst deactivation mechanisms are then discussed, highlighting the roles of coking, poisoning, and structural evolution under realistic feed conditions. Finally, strategies for enhancing mass transfer—from nanoscale catalyst engineering to reactor‐level design—are summarized, emphasizing their critical impact on efficiency, selectivity, and scalability in plastic waste upcycling.

FIGURE 1.

Toward catalytic upcycling of waste plastics: catalyst design, stability, and mass transfer regulation strategies.

2. Catalyst Design and Applications

2.1. Metal Catalysts for Hydrogenolysis

Hydrogenolysis is a metal‐dominated pathway for PO upgrading, in which C–C bonds in saturated hydrocarbon chains are cleaved under H₂ to form short alkanes. The reaction proceeds through activation of H₂ on metal sites, sequential C–H activation, C–C bond cleavage, and hydrogenation of the resulting fragments. A general mechanism involves partial dehydrogenation of hydrocarbon segments, scission of weakened C–C bonds, and hydrogenation of formed fragments [24]. PO with different backbone structures exhibit distinct product fingerprints: PE, possessing a fully linear –(CH₂) _n – backbone, predominantly yields n‐alkanes with superior activity, whereas PP and PS produce mixtures enriched in branched alkanes or aromatic‐containing fragments due to altered adsorption geometry and intermediate stability. A key mechanistic feature is regioselectivity of C–C cleavage. Terminal and/or excessive C–C bond scission typically generates CH₄, which is a probe product distinguishing with other reaction pathways [25, 26].

Metal‐catalyzed hydrogenolysis has become a key strategy for the chemical recycling of PE waste because it converts long‐chain saturated hydrocarbons directly into short n‐alkanes. The resulting alkanes span gasoline‐ and diesel‐range fuels, lubricant base oils, waxes, and chemical feedstocks whose linear geometry exhibits advantages for cracking to short olefins.

Despite significant progress, hydrogenolysis still faces several key challenges. Many catalysts require relatively high temperatures and H₂ pressures, offer limited selectivity control, and tend to generate excessive light gases instead of liquid‐range alkanes. Moreover, factors such as metal particle size, heteroatom incorporation, and evolution of the metal–support interface exert strong influence on product distribution and catalyst stability. Consequently, precise regulation of metal electronic structure, particle size, and metal–support interactions is essential to improving hydrogenolysis efficiency and selectivity. Current strategies mainly focus on optimizing metal size and coordination environments and utilizing reducible oxide supports rich in oxygen vacancies to stabilize highly dispersed metal sites and tune their electronic properties.

2.1.1. Metal Particle Size

Hydrogenolysis is considered as a metal‐governed C–C scission reaction whose activity and selectivity strongly depend on the structure of the supported metal. Metal particle size especially for Ru, Pt, and Ni is one of the most important structural parameters reported to influence C–C activation pathways and product distributions during PE degradation. The evolution from single atoms to clusters to nanoparticles is a central factor controlling hydrogenolysis depth and product distribution [27].

Numerous studies demonstrate that particle size not only affects activity but is directly correlated with product composition. For Ru‐based catalysts, Ru/ZrO₂ achieves high PE conversion (>80%) with moderate CH₄ formation at ∼2–3 nm Ru size, while increasing particle size to ∼7 nm leads to significantly higher C₁–C₄ gas yields dominated by methane (Figure 2A, Table 1, Entry 1) [28]. A similar volcano‐type trend is also observed on Ru/CeO₂ systems (Figure 2B) [29], where subnanometer to nanoparticle evolution reshapes C–C scission behavior and shifts selectivity between liquid hydrocarbons and light gases. Meanwhile, increasing Ru loading on oxide supports leads to concomitant particle growth and corresponding changes in hydrogenolysis performance (Figure 2C, Table 1, Entry 2) [30]. As Ru loading increases, average particle size grows, which correlates with higher methane and light‐gas yields and diminished liquid alkane formation, highlighting that larger Ru ensembles favor excessive C–C bond scission. Another study reveals that Ru‐based bifunctional catalyst performance depends on the synergy between metal dispersion and support acidity. For PE, optimal Ru dispersion (∼40%–50%) with suitable acidity maximizes liquid yields, whereas for PP, support acidity is far more decisive, suppressing methane formation and promoting internal C–C cleavage (Figure 2D) [16]. Similar size‐dependent trends are reported for Ru on zeolites and SBA [30, 49].

FIGURE 2.

(A) Relationship between Ru particle size and catalytic performance (conversion, TOF, and gas yield). Reproduced with permission [28]. Copyright 2022 Elsevier. (B) Variations in the performance of Ru/CeO₂ in the hydrogenolysis of n‐hexadecane and LDPE with Ru size. Reproduced with permission [29]. Copyright 2022 Wiley. (C) Ru loading versus PE conversion. Reproduced with permission [30]. Copyright 2021 Elsevier. (D) Liquid and gaseous product distributions for the reactions over physical mixture of Ru/SiO₂ and FAU zeolites. Reproduced with permission [16]. Copyright 2022 American Chemical Society.

TABLE 1.

Summary of catalysts and conditions for PE, PP, and PS conversion through hydrogenolysis, catalytic cracking, hydrocracking, and catalytic oxidation pathways.

Entry	Catalyst	Substrate	Reaction conditions	Conv., %	Selectivity, %
Hydrogenolysis					C1	C2–4	C5–20	C21+	Ref.
1	Ru/ZrO2	LDPE	240°C, 6 MPa H2, 4 h	80	∼8.3	3.7	57	15	[28]
2	Ru/Zeolite	LDPE	300°C, 50 bar H2, 3 h	>99	97	3	0	0	[30]
3	Ru9Pt91/C	PE	300°C, 0.5 MPa H2, 8 h	100	∼1.2	∼1.8	∼13	∼84	[31]
4	Ru/C‐EG120	LDPE	240°C, 2 MPa H2, 1.5 h	∼80	∼5.5	∼0.5	∼94	0	[32]
5	Ru/MgAl‐LDO	HDPE	240°C, 4 MPa H2, 4 h	79.4	2.3	3.8	83.2	10.6	[33]
Catalytic cracking					C1–4	C5−20+	Alkene	Aromatic
6	H‐MFI	HDPE	200°C, 1 MPa H2, 5 h	39	20	53	27	n.a.	[34]
7	HY	HDPE	500°C, N2 flow, 7.5 h	99	9	51	26	14	[35]
8	Beta‐N	LDPE	350°C, 3.7 MPa N2, 2 h	82.7	44	41	10	n.a.	[36]
9	LSP‐Z100	HDPE	240°C, 0.1 MPa N2, 4 h	81.8	0.7	85.3	1.8	12.2	[37]
10	s‐ZSM‐5	PE	280°C, Ar and H2 flow, 7 h	99	2.3	33.7	64	n.a.	[38]
11	Zn/meso‐ZSM‐5	PE	400°C, H2, N2 and Ar flow, 7 h	99	34	14	n.a.	50	[39]
12	HZSM‐5(160)	LDPE	380°C, 0.5 MPa CO2, 3.3 h	99.4	27	22.6	n.a.	50.4	[40]
Hydrocracking					C1–4	C5–20	C21+	Aromatic
13	Pt/ZSM‐5DR	LDPE	270°C, 1.5 MPa H2, 12 h	87.7	87.7	20.5	79.5	n.a.	[41]
14	0.1Pt/USY‐CI	PE	280°C, 3 MPa H2, 3 h	64.2	6.5	90.9	2.6	n.a.	[21]
15	Ru/m‐ZSM‐22	PE	250°C, 3 MPa H2, 8 h	95	13.7	86.3	n.a.	n.a.	[42]
16	10Ni/Beta	LDPE	250°C, 3 MPa H2, 2 h	99.1	11.2	88.8	n.a.	n.a.	[43]
17	Ru1/ZrO2	PE	250°C, 3 MPa H2, 8 h	97	28.8	71.2	n.a.	n.a.	[19]
18	1Ni/Nb2O5	PE	260°C, 3 MPa H2, 5h	99	2.1	95.1	2.8	n.a.	[17]
19	1Ni/WO2.72	PE	260°C, 3 MPa H2, 5 h	100	2.5	94.3	3.2	n.a.	[44]
20	1Ni/Nb2O5	PS	280°C, 4 MPa H2, 5 h	96.5	∼1	n.a.	n.a.	74.8	[45]
Oxidation					Ar. acids	Aliphatic diacids	Ar. carbonyls	Multi‐Ar
21	Co‐MCM‐41	PE	125°C, 1 MPa O2, 12 h	99	n.a.	∼59	∼18	n.a.	[46]
22	g‐C3N4	PS	150°C, 1 MPa O2, 5 h	60	74	n.a.	21	n.a.	[47]
23	AlCl3	PS	80°C, air, 6 h	>90	n.a.	n.a.	∼11	∼75	[48]

Comparable size–selectivity relationships appear in Ni and Pt systems. For Ni/SiO₂ catalysts in PE hydrogenolysis, different Ni sizes generate distinct active‐site ensembles and cleavage patterns: at ∼300°C and ∼30 bar H₂, small‐medium Ni particles yield mainly liquid alkanes (typically >60 wt%), whereas larger Ni promotes deeper cracking with higher light‐gas formation, dominated by C₂–C₄ with relatively minor CH₄ [50, 51]. In Pt systems, size‐controlled Pt nanoparticles in mSiO₂/Pt/SiO₂ architectures show that the smallest Pt particles deliver the highest mass‐normalized activity at ∼300°C and ∼1 MPa H₂, achieving high conversion with liquid‐centered product distributions, while light gases remain primarily C₁–C₄ [52]. Overall, Ru is more widely employed because comparable or higher activity and liquid selectivity can be obtained at lower temperatures (often 200°C–250°C), whereas Pt and Ni generally require harsher conditions.

Collectively, these studies have shifted catalyst design from maximizing intrinsic activity toward fine particle‐size control to steer reaction pathways. Practical strategies include controlling metal loading, precursor chemistry, and reduction conditions to govern particle nucleation and growth (e.g., low loading and moderate reduction favor dispersed sites in Ru/CeO₂ or Ru/ZrO₂) [53], exploiting supports that strengthen metal–support anchoring to suppress sintering (e.g., reducible oxides CeO₂, TiO₂, and ZrO₂), and using synthesis routes (in situ nucleation, ion adsorption, and confinement) that yield narrowly distributed, stable particle sizes. Precisely tuning metal size thus enables reduction of low‐value CH₄ formation while maintaining sufficient activity to maximize liquid alkane yield.

2.1.2. Heteroatom Incorporation and Alloying

Compared with monometallic catalysts, alloying or heteroatom doping affords simultaneous electronic and geometric control of hydrogenolysis. Introducing a second metal can shift the primary metal's d‐band center and local electron density, thereby adjusting C–C activation barriers and H adsorption/dissociation energetics; alloy structures also disrupt extended metal surfaces, suppressing deep terminal scission and favoring internal cleavage.

For example, Ru–Pt alloys enable efficient PP hydrogenolysis at low or ambient H₂ pressures with methane selectivity below ∼3.5%: Dilute Ru–Pt catalysts achieve high conversion and predominantly liquid C₅–C₄₀ products at ∼200°C–300°C and low H₂ partial pressures, with significantly suppressed CH₄ relative to monometallic Ru (Figure 3A, Table 1, Entry 3) [31]. Recent advances in Ru–Ni alloy systems further reinforce this concept. Ru–Ni nanoparticles supported on TiO₂ yield up to ∼55 % liquid products under optimized conditions (250°C, 20 bar H₂) with markedly lower CH₄ compared to monometallic Ru, demonstrating enhanced selectivity through alloy formation and tuned Ru/Ni ratios [54]. Ni–Ru bimetallic catalysts similarly show pronounced alloy synergy in PP hydrogenolysis: Compared with Ru/SiO₂, Ni–Ru/SiO₂ increased liquid yield from 65.5% to 83.1% and significantly reduced methane selectivity (Figure 3B) [55]. Ru/C‐EG120 demonstrates that tailoring the electronic structure of Ru to generate coexisting Ru⁰ and Ru^δ+ species provides an additional handle over selectivity. Metallic Ru⁰ sites ensure efficient H₂ dissociation, while electron‐deficient Ru^δ+ centers modulate substrate polarization and lower the barrier for σ‐bond activation, preferentially directing internal C–C scission toward liquid alkane formation while suppressing excessive cracking (Figure 3C, Table 1, Entry 4) [32]. In MgAl‐LDO‐supported ultrafine Ru catalysts for consumer‐grade HDPE hydrogenolysis, the coexistence of metallic Ru and partially oxidized Ru^δ+ species on MgAl‐LDO leads to high selectivity toward C₆–C₂₀ liquid alkanes with suppressed CH₄ formation at moderate temperatures (∼240°C, 30–40 bar H₂), attributed to strong metal–support interactions (SMSIs) (Table 1, Entry 5) [33]. Both MgAl‐LDO‐supported and Ru/C‐EG120 catalysts illustrate that tuning Ru oxidation state and support interactions can enhance backbone scission over terminal C–C cleavage, increasing liquid yields while suppressing CH₄ under comparable hydrogenolysis conditions.

FIGURE 3.

(A) PE hydrogenolysis over Ru₉Pt₉₁/C, Ru₁₄Ni₈₆/C, Ru₆Co₉₄/C, and Ru₉Pd₉₁/C catalysts, and schematic illustration of the reaction mechanism proposed over Ru₉Pt₉₁/C (top) and Ru/C (bottom). Reproduced with permission [31]. Copyright 2024 Nature. (B) Catalytic performance of RuNi alloy catalysts and (C) surface ligand engineering of Ru species by ligandization for boosted polyolefin hydrogenolysis. Reproduced with permission [32]. Copyright 2024 American Chemical Society.

Design principles for alloy catalysts include selecting a second component to tailor the principal metal's electronic structure (Ru–Pt, Ru–Ni, Pt–Sn, and Pt–Ga), controlling alloy composition and atomic arrangement to balance isolated sites versus contiguous ensembles, and combining alloying with support interactions to stabilize desired atomic environments.

2.1.3. Support Facets and Structure

Hydrogenolysis has often been viewed as metal‐centric, with supports serving primarily to disperse and stabilize metal. However, growing evidence shows that support structure crystal facets, defect type, and reducibility actively participate in and modulates hydrogenolysis. Different exposed facets alter metal anchoring, vacancy formation energies, and interfacial electronic structure, thereby steering C–C scission pathways.

In Ru/TiO₂ systems, synthesis pH governs TiO _x overlayer formation and exposed facets, thereby tuning Ru–support coupling and hydrogenolysis performance. At pH 1, Ru/TiO₂ mainly exposes crystalline anatase surfaces with weak overlayer formation, giving comparatively lower PE/PP hydrogenolysis rates and moderate liquid yields. At pH 8, a well‐defined TiO _x overlayer forms with stronger Ru–TiO₂ electronic interaction, enhancing H₂ dissociation and spillover. Under ∼250°C–300°C and 10 bar H₂, this sample shows significantly higher conversion and liquid yields with low CH₄ (<10%). At pH 12, thicker TiO _x modification maintains enhanced activity but with slightly altered hydrogen coverage behavior (Figure 4A) [56]. For CeO₂ supports, facet dependence similarly regulates electronic coupling: Low‐coordination (110)/(100) facets generate oxygen vacancies more readily than (111), and Ru/CeO₂(110) exhibits higher activity and lower methane selectivity owing to vacancy‐mediated adsorption and cleavage favoring internal scission (Figure 4B) [57].

FIGURE 4.

(A) FT‐IR of Ru/TiO₂ adsorbed CO at −196°C at different CO pressures and catalytic performance. Reproduced with permission [56]. Copyright 2022 Nature. (B) Effect of reaction temperature in PP hydrogenolysis for 2 wt% Ru‐supported CeO₂‐SA, CeO₂‐NR and CeO₂‐NC catalysts. Reproduced with permission [57]. Copyright 2023 Elsevier.

Accordingly, support engineering strategies include selecting and synthesizing supports with preferred facet exposure, introducing oxygen vacancies or tuning reducibility to promote hydrogen spillover and interfacial cooperation, and designing nanoscale morphologies to optimize metal anchoring and ensemble behavior. Moderate vacancy concentrations can enhance spillover and stabilize intermediates, suppressing overcracking while promoting midchain cleavage.

2.1.4. SMSIs

SMSI alters not only geometric stabilization of metal particles but also electronic properties and surface chemistry. Under reducing conditions, supports may partially encapsulate metals or transfer charge to metal particles, modifying adsorption energetics and hydrogen activation behavior.

In hydrogenolysis, a moderated SMSI weakens excessively strong adsorption of polymer‐derived alkyl fragments on metal surfaces and limits the formation of large contiguous metal ensembles that otherwise promote deep C–C scission, thereby suppressing methane formation. In Ru/TiO₂ catalysts, synthesis modification using ammonia strengthens Ru–TiO₂ interactions; under H₂, interfacial electron transfer generates Ru^δ+ and Ti³⁺ species, enhances hydrogen spillover, and maintains high surface hydrogen coverage in a reversible manner. Under representative PP hydrogenolysis conditions, this SMSI‐engineered Ru/TiO₂ catalyst achieves ∼74% liquid yield within 6 h and exhibits approximately threefold higher activity compared with conventional Ru catalysts [56]. Similar interfacial‐coupling strategies have been advanced in nonprecious metal systems: Ni species stabilized at near atom‐level dispersion (∼0.9 nm) on SiO₂ display superior H₂ activation capability and controlled C–C cleavage behavior, enabling up to ∼81% liquid fuel yield from PE under relatively mild conditions [50].

Representative systems such as Ru/CeO₂ [29], Ru/TiO₂ [58], and Ni/SiO₂ [50] confirm that catalysts exhibiting clear SMSI signatures generally afford improved selectivity control and stability in hydrogenolysis. Design strategies therefore include selecting reducible supports prone to SMSI (e.g., ZrO₂, TiO₂, and CeO₂), tuning reduction protocols to set SMSI strength in the desirable regime, and engineering interface architectures to sustain favorable metal valence and surface chemistry over time.

2.2. Solid Acid Catalysts for Catalytic Cracking

Catalytic cracking predominantly occurs on solid acid catalysts and proceeds through an acid‐catalyzed mechanism involving carbocation intermediates [59]. The reaction is initiated by the adsorption and protonation of PO chains on Brønsted acid sites, leading to the formation of penta‐coordinated carbocation intermediates. These intermediates may undergo C–C bond or C–H bond cleavage, generating an alkane molecule and a tri‐coordinated carbocation [60]. Subsequently, the resulting carbocations can undergo isomerization via hydride transfer or alkyl migration, followed by β‐scission reactions that produce shorter‐chain olefins and new carbocation species, thereby sustaining the chain‐cracking process [22, 60]. Lewis acid sites (LASs) contribute by interacting with the π electrons or σ bonds of hydrocarbon molecules, enhancing the polarization of C–C and C–H bonds and consequently lowering the activation energy for bond cleavage. In addition, olefins and dienes generated during cracking can be reprotonated on acidic sites and subsequently undergo cycloaddition reactions, followed by dehydrogenation and aromatization, ultimately leading to the formation of aromatic hydrocarbons [20].

Solid acid catalysts, such as zeolites and metal oxides, have been extensively used in the catalytic and hydrocracking conversion of PO owing to their tunable acidity and well‐defined pore structures. The presence of hydroxyl groups and electron‐deficient metal centers exist on solid acid catalysts, giving rise to diverse acidic properties—including acid type, strength, and density—which play a decisive role in determining catalytic performance [60, 61]. In addition, the framework topology and pore architecture of solid acid catalysts collectively regulate the diffusion behavior of reactants and products, the accessibility of acid sites, and the resulting product distribution. It thereby exerts a strong influence on catalytic activity, selectivity, and stability, generating alkanes, alkenes, or aromatics with desired selectivity [62]. Accordingly, this section focuses on the analysis and design of acidity characteristics and pore structures in different types of solid acid catalysts, aiming to provide guidance for the rational development of highly efficient catalysts for PO cracking.

2.2.1. Acid Type, Strength, and Distribution

In zeolite catalysts, acidity primarily originates from Brønsted acid sites (BASs, Si–OH–Al) and LASs (mainly extra‐framework aluminum species), which are commonly characterized by pyridine adsorption infrared spectroscopy (Py‐IR) (Figure 5A) [63, 65, 66]. BASs are the primary active centers responsible for C–C bond cleavage, promoting the conversion of PO into alkanes or olefins. For H‐MFI zeolites (Table 1, Entry 6) subjected to different desilication times, catalytic cracking of PE at 473 K shows that increasing Brønsted acid density significantly enhances solid conversion rising from 17% to 40% and generating C₂–C₃ and C₄–C₇ linear alkanes (Figure 5B) [34]. Solid acid catalysts can also promote the catalytic conversion of PO into short olefins. By integrating short‐range ordered SOD‐type structures into an amorphous aluminosilicate matrix, the accessibility of BAS on the solid acid support to PE is significantly enhanced. This structural optimization enables efficient C–C bond scission. As a result, during the catalytic cracking of low‐density polyethylene (LDPE) at 440°C, an exceptionally high selectivity toward C₃–C₅ olefins of 94.9% and a corresponding yield of 88.1% were achieved [67]. These results indicate that PE isomerization and degradation are closely associated with the ability of BAS to activate C–C bonds under mild conditions. Steam treatment of HY zeolite slightly increased BAS, thereby enhancing the catalytic efficiency of PO [68]. Moreover, MFI zeolites with weaker acidity require higher reaction temperatures to achieve conversion levels comparable to those of strongly acidic counterparts, demonstrating that weak BAS can sustain cracking reactions but with significantly reduced catalytic efficiency [69].

FIGURE 5.

(A) Py‐IR spectra of zeolites Y, Y‐M, Y‐L, and Y‐H. Reproduced with permission [63]. Copyright 2024 American Chemical Society. (B) Solid conversion and rate for PE catalytic cracking on parent and hierarchical MFI catalysts. Reproduced with permission [34]. Copyright 2024 American Chemical Society. (C) Illustration of the possible carbenium‐ion mediated routes catalyzed by LAS and BAS in [C₄Py]Cl‐AlCl₃ and time‐resolved conversion of LDPE in the presence of TBC at different temperatures. Reproduced with permission [10]. Copyright 2024 American Association for the Advancement of Science. (D) Proposed mechanism of tandem PE hydrogenolysis/aromatization via dehydrocyclization and evolution of major product fractions over Pt/Al₂O₃ (orange, CHCl₃‐soluble liquids/waxes; black, insoluble hydrocarbons). Reproduced with permission [64]. Copyright 2020 American Association for the Advancement of Science.

Extra‐framework aluminum species (EFAl, such as penta‐ or hexa‐coordinated Al) and other LASs also play critical roles by cooperating with BAS [70]. For example, ionic liquids such as Al₂Cl₇ ⁻ enable low‐temperature PO degradation, where LASs promote hydride abstraction to generate carbocation intermediates (Figure 5C). As the reaction temperature increases from 30°C to 70°C, PO conversion rises steadily, indicating enhanced catalytic activity of LASs [10]. In n‐hexane cracking over zeolites, the presence of LAS stabilizes BAS, thereby significantly improving overall catalytic efficiency [70]. In Pt/γ‐Al₂O₃ catalysts, LASs promote cyclization, isomerization, and C–C bond cleavage, leading to an increase in soluble products from 21% to 80% with prolonged reaction time and a concurrent decrease in average carbon number (Figure 5D) [64].

Moreover, regulating the distribution and strength of acidity are key factors in directing PO cracking toward aromatics. The presence of BAS markedly lowers the energy barrier for hydrogenation/dehydrogenation of double bonds, highlighting the intrinsic catalytic potential of zeolitic acid sites for hydrogen transfer reactions even under metal‐free conditions [20]. Zn‐modified mesoporous ZSM‐5 catalysts further promote PO cracking and aromatization. ZnO _x species regulate strong acid sites while preserving an appropriate level of Brønsted acidity, thereby favoring aromatization pathways and the formation of light olefins and methylated aromatics [39]. In the co‐conversion of PO and CO₂ to aromatics, a limited number of acid sites facilitate C–C bond cleavage to generate reactive intermediates, whereas excessive acidity promotes hydrogen transfer and condensation reactions, accelerating the formation of polycyclic aromatic hydrocarbons (PAHs) and coke and thus catalyst deactivation [71, 72]. Coating secondary hydrothermally treated ZSM‐5 with an SBA‐15 shell effectively passivates external acid sites while confining strong acidity within the micropores [73]. This strategy not only tunes the acid distribution but also repairs framework defects and improves structural stability and acid durability, thereby enhancing aromatization selectivity [74].

Moreover, Ga‐modified hierarchical ZSM‐5, Ga₂O₃ clusters, and GaO⁺ species interact with zeolitic acid sites to form cooperative Lewis–Brønsted acid centers, which effectively promote olefin cyclization–dehydrogenation–aromatization reactions. The oil product yield over the catalyst reached up to 62.2 wt%, with the selectivity to aromatic hydrocarbons approaching 100% (Figure 6A) [75]. In tandem catalytic systems such as Zn/ZSM‐5 (Table 1, Entry 11) combined with Cu–Fe₃O₄, Zn facilitates dehydrogenation and synergistically enhances aromatization with acid sites, while Cu–Fe₃O₄ regulates hydrogen balance through the reverse water‐gas shift reaction [76]. Likewise, in a tandem catalytic system composed of HZSM‐5 and CuZnZrO₄, efficient solvent‐free co‐conversion of polyethylene (PE) and CO₂ into aromatic hydrocarbons was achieved. Under reaction conditions of 380°C and 5 bar CO₂, both HDPE and post‐consumer plastic waste could be effectively transformed, affording aromatic yields of 54.6–58.6 wt%. Meanwhile, CO₂ conversion exceeded 0.4 mmol⋅g_PE ⁻¹, indicating the important role of the CuZnZrO₄ component in regulating hydrogen balance and promoting tandem aromatization (Figure 6B) [40].

FIGURE 6.

(A) Carbon distribution and the yields of different types of nonsolid products for co‐conversion of LDPE and CO₂ during thermal cracking with Ga/Hie‐ZSM‐5 catalyst. Reproduced with permission [75]. Copyright 2025 Elsevier. (B) Product distributions for raw PE powder and postconsumer PE in the co‐conversion of CO₂ and PE over HZSM‐5 and CuZnZrO₄ catalysts. Reproduced with permission [40]. Copyright 2024 American Association for the Advancement of Science.

Tuning the Si/Al ratio is an effective strategy for regulating the balance between Brønsted and Lewis acidity and thus enhancing cracking performance [77]. Dealumination of H‐Beta via nitric acid treatment yields deAl‐Beta, which exhibits up to 90% selectivity toward C₂–C₄ olefins during PO cracking [78]. This reaction follows a Lewis acid‐mediated pathway involving hydride abstraction, carbocation formation, and subsequent β‐scission to generate light olefins. Emerging self‐supporting zeolitic structures further increase the accessibility of acid sites, particularly surface tri‐coordinated aluminum species, leading to the coexistence of strong Lewis and BAS [70]. This synergistic acidity effectively activates C–H bonds, enables hydrogen self‐transfer, promotes β‐scission and isomerization, and ultimately allows selective conversion of PO into branched alkanes.

Oxide catalysts can also promote PO cracking through intrinsic surface acidity/basicity and defect chemistry, even in the absence of metals. Mesoporous SiO₂ has been shown to facilitate the cracking of PE/PP dissolved in FCC streams at ∼450°C–550°C, increasing conversion relative to thermal pyrolysis and producing gasoline‐range hydrocarbons (C₅–C₁₂) with characteristic paraffinic/olefinic distributions [79]. Alkaline‐earth oxides such as MgO and CaO likewise enhance PE/PP depolymerization at ∼420°C and ∼1 MPa under inert atmospheres, improving liquid yields in the gasoline‐diesel boiling range while suppressing excessive condensation [80]. These results demonstrate that appropriately engineered oxide surfaces can mediate chain scission and tune product selectivity, although typically at higher temperatures.

2.2.2. Microporous and Topological Structures

The microporous structure of zeolites endows them with unique shape‐selective catalytic properties, enabling regulation of reaction pathways and product distributions through spatial confinement effects. Typical zeolite topologies include MFI (ZSM‐5), BEA* (Beta), FAU (Y, USY), and MOR (mordenite), whose pore size and channel dimensionality play decisive roles in reactant diffusion and product selectivity [81].

MFI‐type zeolites (e.g., ZSM‐5) possess intersecting 10‐membered ring channels (∼0.55 nm), whose dimensions are comparable to the kinetic diameters of short‐chain olefins and BTEX molecules, thereby exhibiting strong shape selectivity [82]. However, the MFI‐based catalysts have noticeable preference in product selectivity under different reaction conditions. Specifically, it catalyzes PO into light alkanes and olefins under ambient pressure, while into aromatics under high pressure and temperature in autoclaves. To further improve the diffusion of bulky PO molecules within micropores and enhance the selectivity toward light olefins and alkanes, b‐axis‐shortened s‐ZSM‐5 (Table 1, Entry 10) structures were developed [38]. These materials effectively alleviated diffusion limitations, enabling efficient PO conversion at 280°C, with light hydrocarbon yields (C₁–C₇) reaching 74.6%, of which 83.9% were C₃–C₆ olefins, while coke formation was almost negligible (Figure 7A). Using b‐axis‐shortened HZSM‐5 nanosheets (s‐ZSM‐5) with an average b‐axis length of ∼66 nm, 94.7% LDPE conversion and 90.0% C₂–C₆ olefin selectivity were achieved at 240°C and 1 atm N₂. Within the zeolite micropores, a dynamic isomerization–oligomerization–scission (IOS) cycle, driven by preferential diffusion of short‐chain olefins, continuously promoted LDPE cracking [84]. In Zn‐modified ZSM‐5 catalysts, efficient conversion of PO into aromatic hydrocarbons can be achieved at 390°C. The recovered liquid products exhibit an exceptionally high aromatic content, with monocyclic aromatics accounting for up to 90.0% and BTEX compounds reaching a maximum proportion of 61.5% [85]. In layered self‐pillared zeolites (Table 1, Entry 9) [37], yields exceeding 80% were achieved within 4 h at 240°C (Figure 7B). The liquid products consisted predominantly of branched alkanes (72% selectivity), exhibiting a research octane number (RON) of 88.0, comparable to that of commercial gasoline (RON = 86.6). High‐silica ZSM‐5‐300 exhibited excellent catalytic performance in plastic cracking; under solvent‐free conditions at 280°C for 24 h, it achieved a monocyclic aromatic selectivity of 60.3% [86].

FIGURE 7.

(A) Photograph of the powder PE and SEM image of the s‐ZSM‐5 with a panel morphology and hydrocarbon distribution of the products from conventional thermal depolymerization and catalytic depolymerization route at the s‐ZSM‐5 catalyst. Reproduced with permission [38]. Copyright 2022 American Chemical Society. (B) Conversion of PE to gasoline on LSP zeolites under mild conditions via an SSH strategy. Reproduced with permission [37]. Copyright 2024 Nature. (C) Yield of products in PE cracking over H‐USY and 0.38Na‐USY catalysts at T10 and T97 under programmed‐temperature conditions. Reproduced with permission [83]. Copyright 2024 Elsevier. (D) Liquid product distributions in catalytic PE cracking over Beta‐M and Beta‐N zeolites. Reproduced with permission from ref. [36]. Copyright 2024 American Chemical Society.

Mordenite (MOR), a medium‐pore zeolite featuring one‐dimensional straight channels (∼0.7 nm), exhibits high intrinsic catalytic activity in hydrocarbon cracking. However, its unidirectional pore architecture limits the diffusion of PO and reaction intermediates, often leading to severe coke formation and rapid catalyst deactivation. Pore structure engineering is therefore essential for applying MOR zeolites in PO conversion.

FAU‐type zeolites (e.g., HY and USY) possess a three‐dimensional pore system with large supercages (∼1.2 nm), enabling accommodation of bulky hydrocarbons and facilitating the formation of larger molecules (C₄–C₁₆). H‐USY tends to produce higher fractions of C₃–C₅ and >C₅ olefins, whereas Na‐USY favors the formation of longer‐chain alkanes (>C₅) (Figure 7C). These results indicate that FAU zeolites promote the conversion of PE into smaller molecules, with product distributions strongly biased toward longer hydrocarbon chains (selectivity >80% for >C₅ fractions) [83]. However, the large supercages also prolong the residence time of bulky intermediates, thereby promoting coke formation.

Beta (BEA) zeolites further enhance intracrystalline mass transport owing to their three‐dimensional 12‐membered ring channel system (∼0.7 nm), making them feasible for the diffusion of medium‐sized hydrocarbons (C₃–C₁₄) [87]. Their unique topology also facilitates isomerization and cyclization reactions. Nanosized Beta zeolites (≤200 nm) exhibit outstanding performance in PO cracking, delivering 60%–70% selectivity toward C₅–C₁₂ hydrocarbons, while suppressing the formation of heavier products to below 6% [88]. In contrast to noncatalytic PE pyrolysis, which predominantly yields linear alkanes, the introduction of Beta zeolites promotes the formation of smaller hydrocarbons with fewer than 20 carbon atoms. Notably, Beta‐N (nanosized Beta) shows superior catalytic activity compared to Beta‐M (micrometer‐sized Beta) [36], producing predominantly iso‐ and cycloalkanes in the liquid fraction, with selectivity exceeding 90% (Figure 7D, Table 1, Entry 8).

The pore structure and topology of different zeolites play a decisive role in governing the cracking behavior of PO. MFI‐type zeolites favor the formation of light olefins and aromatics, whereas FAU‐type zeolites facilitate the conversion of larger molecules but are more prone to coke formation. MOR zeolites suffer from rapid deactivation due to diffusion limitations imposed by their one‐dimensional channels, while BEA zeolites exhibit a better balance between molecular diffusion and product selectivity. Therefore, selecting zeolites with appropriate pore architectures is crucial for achieving highly selective conversion of PO into target products.

2.2.3. Mesoporous and Hierarchical Porous Structures

In PO catalytic cracking reactions, molten PO tend to cover the catalyst surface [22], leading to significant diffusion limitations despite the excellent shape‐selective catalytic properties of microporous zeolites (Figure 8A). The narrow pores of microporous zeolites hinder efficient mass transfer, limiting the overall catalytic performance. To address these challenges, hierarchical zeolites, which combine micropores, mesopores (2–50 nm), and macropores (>50 nm), have gained increasing attention in recent years. These hierarchical pore structures are typically achieved through post‐treatment methods that introduce mesoporous or macroporous features to the zeolite surface, significantly improving mass transfer and enhancing PO catalytic cracking efficiency [81].

FIGURE 8.

(A) Polyolefin catalytic cracking system: surface melting, envelopment, and diffusion behavior of PE over zeolites. Reproduced with permission [22]. Copyright 2025 Nature. (B) Schematic illustration of polyolefin cracking over amorphous–crystalline composite catalyst. Reproduced with permission [67]. Copyright 2025 American Chemical Society. (C) Product distribution in LDPE cracking over P/mesoZ5 at different temperatures. Reproduced with permission [89]. Copyright 2022 Elsevier.

The pore size of molecular sieves is positively correlated with PO degradation efficiency, and larger pore sizes can reduce catalyst deactivation. For example, a tunable pore‐size Al‐MCM‐41 catalyst for PP degradation at 220°C achieves a liquid fuel yield of 70.2%, with mesopores facilitating improved mass transfer of PP on the zeolite [77]. Furthermore, other mesoporous‐containing structures enhance mass transfer and allow for precise tuning of pore structure. For instance, hollow ZSM‐5_DR (Table 1, Entry 13) zeolite enables the degradation of LDPE at 270°C, resulting in a conversion rate of 87.7% and a liquid fuel selectivity (C₅–C₂₁) of 77.8% [41]. Hierarchical FAU zeolites synthesized by a templating method enhanced hydrocracking of linear and branched plastics. The HY catalyst showed high activity due to improved mesoporosity. For linear HDPE, conversion plateaued beyond a mesoporosity threshold, while gasoline‐range (C₅–C₁₈) yield increased, achieving 77.3% conversion, 54% selectivity, and a productivity of 8.35 g_C5−18⋅g_cat ⁻¹⋅h⁻¹ [90].

To further enhance mass transfer of PO on solid acid catalysts, various strategies have been developed to modify zeolite structures and improve diffusion efficiency (Figure 8B). For s‐ZSM‐5, primary degradation intermediates can rapidly undergo secondary cracking within the micropores to form smaller olefins, while shortened diffusion pathways facilitate the rapid desorption of these products, thereby suppressing their accumulation and coke formation on the external surface [38]. During LDPE catalytic cracking over P‐meso‐ZSM‐5 (prepared using H₃PO₄ as the precursor) [89], increasing the reaction temperature promotes the formation of methane, ethylene, 1,3‐butadiene, and aromatics, whereas the yield of C₅₊ aromatics decreases, with particularly pronounced changes observed between 600°C and 700°C (Figure 8C). These results indicate that the catalyst‐to‐feed ratio and reaction temperature play critical roles in governing product distribution, enabling flexible tuning toward C₄ olefins, ethylene, propylene, or aromatics while minimizing carbon losses to methane and coke. Similarly, during PS conversion over HZSM‐5, in situ catalytic pyrolysis produces significantly higher solid residue (15.86 C% vs. 4.44 C%) and aromatic yields, with benzene reaching 39.6 C% compared to 23.59 C% under ex situ conditions; aromatics dominate the product distribution (∼80 C%), while alkanes and alkenes remain minor components, indicating enhanced dealkylation and aromatization under in situ conditions. In situ pyrolysis involves directly mixing plastics with HZSM‐5, allowing primary pyrolysis intermediates to immediately enter the zeolite micropores where strong shape‐selective catalysis promotes secondary reactions such as aromatization, typically accompanied by increased coke formation. In contrast, ex situ pyrolysis first thermally decomposes the plastics before the volatile products pass over the catalyst, leading to weaker pore‐controlled secondary reactions and generally lower solid residue. While macroporous structures are less studied, they are widely used in industrial catalysts. While macroporous structures are less studied yet widely implemented in industrial catalysts, the used industrial fluid catalytic cracking catalyst achieves aromatics yields comparable to those of the fresh catalyst while significantly suppressing coke formation, likely due to the presence of meso‐ and macroporous structures within the zeolitic domains that facilitate polyolefin mass transfer [91]. These structures, often formed through particle packing or macrotemplate methods, effectively reduce internal diffusion resistance and bed pressure drop, thereby improving overall mass transfer efficiency.

Overall, hierarchical zeolites significantly enhance the diffusion efficiency of reactants, improve the accessibility of acid sites, and help delay coke formation, thereby improving catalyst stability and performance. These advancements demonstrate the critical role of hierarchical pore structures in optimizing the catalytic cracking of PO.

2.3. Metal–Acid Bifunctional Catalysts for Hydrocracking

The hydrocracking mechanism is built upon the carbocation chemistry of catalytic cracking while incorporating the synergistic hydrogenation–dehydrogenation function of metal sites, thereby establishing an integrated dehydrogenation–cracking–hydrogenation reaction pathway [92, 93]. In bifunctional catalytic systems, metal sites primarily facilitate hydrogenation and dehydrogenation reactions, whereas acidic sites govern carbocation formation, cracking, and isomerization processes [82]. Compared with catalytic cracking, hydrocracking exhibits distinct advantages owing to the presence of metal component, which reduce the reaction temperature via promoting the dehydrogenation step (typically below 300°C) [93]. Under conditions of sufficient metal site density and appropriate hydrogen pressure, olefinic intermediates formed during cracking are rapidly hydrogenated to alkanes. This effectively suppresses excessive cracking into gaseous hydrocarbons and inhibits aromatization, thereby significantly enhancing liquid alkane selectivity and reducing coke formation [94]. Moreover, metal sites can dynamically modulate the surface acidity of the support, working synergistically with acidic sites to further improve the overall efficiency and selectivity of the cracking process.

The performance of bifunctional catalysts for the hydrocracking of PO is influenced by the metal state, acid properties, and the balance between metal species and the support. The interplay between cooperative metal sites and acidic sites effectively transforms PO into value added fuels or chemicals.

2.3.1. State of the Metal Sites

The performance of bifunctional catalytic systems depends on the dispersion and state of the metal species, which exhibit significant efficiency advantages under mild reaction conditions [92]. In single‐metal catalyst systems, noble metal‐based catalysts such as Pt, Pd, and Ru are widely used due to their excellent dehydrogenation and hydrogenation activity. The activation of hydrogen by Pt is the initial and key step in hydrocracking reactions. To enhance reaction efficiency, the Pt loading can be increased, or its deposition location on the zeolite can be adjusted. In Pt/USY catalysts (Table 1, Entry 14), at a platinum loading of 1 wt%, PE is almost completely converted, and increasing the number of hydrogenation sites greatly improves the conversion rate [21]. If Pt nanoparticles are selectively loaded onto the external surface of the zeolite via a sol–gel immobilization method, the hydrocracking reaction can be significantly enhanced. This is because the primary cracking step, which is rate‐determining, is greatly accelerated when accessible Pt nanoparticles are positioned on the zeolite's outer surface. In contrast, limited surface Pt, Pt confined within internal pores, or an increased distance between metal and acid sites slows reaction initiation, introduces diffusion limitations, or promotes undesired overcracking, ultimately reducing the overall hydrocracking efficiency.

Moreover, the size and oxidative state of metal nanoparticles could also influence the performance. In Ru/ZSM‐22 catalysts (Table 1, Entry 15) [42], the prepared Ru nanoparticles are small and exhibit a higher oxidative state. The oxidized Ru species facilitate the adsorption of olefin intermediates. Under reaction conditions of 250°C for 8 h, the liquid product yield exceeds 80% (Figure 9A). Small Ru particles (∼1.6 nm) are rich in oxidized species, which facilitate olefin desorption and subsequent isomerization and β‐scission on Brønsted acid sites, enabling hydrocracking. In contrast, larger Ru particles are mainly metallic, making olefin desorption difficult and promoting consecutive C–C bond cleavage on the metal surface, producing methane and resulting in hydrogenolysis‐dominated reactions.

FIGURE 9.

(A) STEM images and particle size histograms of the Ru/ZSM‐22 samples prepared using different metal precursors, and the product yields and iso/n ratio in the extractables. Reproduced with permission [42]. Copyright 2024 Elsevier. (B) Catalytic performance comparison over Pt/5Ce‐HY, Pt/HY, HY, 5Ce‐HY, Pt/CeO₂, and the proposed hydrocracking route over Pt/5Ce‐HY. Reproduced with permission [95]. Copyright 2024 Wiley.

To reduce the cost of catalyst preparation, non‐noble metal‐based catalytic systems using metals like Ni and Co as active components have been further developed. In Ni/HZSM‐5 catalysts, the addition of Ni enhances the selectivity for low‐boiling hydrocarbons (C₅–C₁₈ = 48.5%), with a maximum activity of 95.3% [96]. Ni‐loaded zeolites have demonstrated repeatability and regeneration ability in continuous operation. Chelating agents can be used to control the dispersion of Ni ions, enabling the preparation of highly dispersed Ni nanoparticles. As the particle size of Ni nanoparticles decreases, the activity of Ni/Beta‐EDTA catalysts gradually increases, reaching 79% at 280°C [97]. MoS _x ‐H‐Beta catalysts can convert PO into highly isomerized liquid alkanes under mild conditions. At 250°C and 20–30 bar of H₂ after 6 h of reaction, an alkane yield of 96% is successfully achieved. Evaporated Mo complexes were deposited onto H‐Beta zeolite, enabling effective confinement of metal species within the zeolite framework and providing highly dispersed active sites for PO degradation [14].

In multimetal‐doped catalytic systems (e.g., Pt–Sn and Pt–Ce), the synergistic effect between the metal components enhances catalytic performance. In the Pt–Sn system, Sn regulates the chemical state of Pt nanoparticles, suppressing excessive cracking and carbon formation while enhancing hydrogenation activity [98]. This multimetal synergy optimizes the balance between hydrogenation and cracking, improving product selectivity and catalyst resistance to carbon deposition. In Pt–Ce/HY catalysts [99], under reaction conditions of 280°C and 2 MPa H₂ for 2 h, high optimized yields (up to 85 wt% C₅–C₁₂) are obtained from low‐density PE. The addition of CeO₂ suppresses excessive cracking into C₁–C₄ gases, and the Pt–O–Ce interaction enhances the dispersion and stability of Pt (Figure 9B) [95]. In catalyst regeneration experiments, even after two cycles of regeneration, high activity is still maintained.

Similar to bifunctional zeolite‐based catalysts, metal oxide‐supported systems can also achieve highly efficient and tunable catalytic performance in PO hydrocracking. Monometallic catalysts comprising Pt, Ru, or Ni supported on reducible oxides such as WO₃, ZrO₂, Nb₂O₅, CeO₂, TiO₂, or WO _x –ZrO₂ effectively utilize the synergistic electronic interaction between the metal and support, playing a crucial role during hydrocracking [17, 18, 100, 101]. On the one hand, oxygen vacancies and the reducibility of these oxides promote enhanced metal dispersion, structural stability, and hydrogen spillover capability, thereby improving H₂ activation efficiency as well as hydrogenation–dehydrogenation cycling of reaction intermediates. On the other hand, structural and electronic modulation from the support can alter the metal's electronic state, making it more favorable for selective cleavage of internal C–C bonds.

Within this framework, the intrinsic state of the metal component particularly its dispersion and characteristic size plays a decisive role in governing the dominant hydrocracking pathway. Reducing metals to highly dispersed or even atomically isolated states can reshape intermediate stabilization and transition‐state energetics, thereby steering the reaction toward internal chain cleavage and narrowing product distributions. A representative example is isolated Ru atoms embedded in ZrO₂, which drive a distinctive carbocation‐like cracking pathway: at 250°C and 3 MPa H₂, this catalyst converts real PE/PP wastes with >98% conversion while delivering ∼70% C₈–C₁₆ fuels and negligible light‐gas production (<0.1%), maintaining similar selectivity even under harsher reaction conditions (Table 1, Entry 17) [19]. Comparable behavior has also been observed for atomically dispersed Pt on reducible oxides, where single‐site Pt ensures efficient hydrogen participation while favoring controlled backbone cracking, yielding >60% liquid hydrocarbons with markedly suppressed light‐gas formation under hydrocracking conditions. These findings collectively demonstrate that tailoring the metal state, especially toward atomic dispersion, is a powerful strategy to redirect hydrocracking pathways and enhance liquid fuel selectivity.

2.3.2. In Situ Generated Brønsted Acidity

In metal oxide‐based hydrocracking catalysts, an important question is how to build BAS with sufficient strength and stability on nonzeolitic supports so that efficient C–C bond cleavage and high liquid‐fuel selectivity can be achieved. Compared with zeolites, where strong Brønsted acidity originates from framework aluminum, most metal oxides only display relatively weak intrinsic acidity. Therefore, creating BAS that can form and function effectively under practical hydrocracking conditions has gradually become a major focus in this area.

For metal oxide‐based catalysts, Brønsted acidity primarily originates from surface metal hydroxyl groups. In Ni/Nb₂O₅ catalysts, in situ pyridine‐IR spectroscopy clearly reveals the dynamic evolution of BAS under hydrogen atmosphere. With increasing reaction temperature and prolonged H₂ exposure, characteristic bands of pyridine adsorbed on BAS (Py‐H⁺) gradually intensify, indicating that surface Nb=O species are progressively protonated to form Nb–OH Brønsted acid centers via a hydrogen spillover induced transformation [17]. This Nb=O to Nb–OH transformation provides a direct structural basis for the emergence of strong, reaction‐relevant acidity rather than relying on weak, static surface hydroxyls. Correspondingly, Ni/Nb₂O₅ exhibits outstanding hydrocracking efficiency toward PO: Under 240°C–260°C and hydrogen atmosphere, PE can be almost completely converted (∼100%), with liquid‐fuel‐range hydrocarbons (C₅–C₂₀) accounting for over 94% of the products. Such performance confirms that dynamically generated Nb–OH BAS effectively stabilize carbocation intermediates and facilitate β‐scission pathways, thereby enabling highly selective and efficient hydrocracking (Figure 10A,B, Table 1, Entry 18).

FIGURE 10.

(A) Schematic illustration and in situ Py‐IR spectra of the transformation of Lewis into Brønsted acid sites on 1Ni/NbO _x ; the liquid formation rate as a function of the in situ generated Brønsted acidity over 1Ni/NbO _x catalysts. (B) The mechanistic illustration of the formation of hydrogen spillover‐induced Brønsted acid and PE hydrocracking pathways over Ni/NbO _x catalyst. Reproduced with permission [17]. Copyright 2025 Wiley. (C) In situ FT‐IR spectra of 1Ni/ WO_2.72 and 1Ni/WO₃ in a D₂ flow. Reproduced with permission [44]. Copyright 2025 American Chemical Society. (D) Mass fractions of recovered hydrocarbons and the rate of C–C bond scission over halogen‐modified Pt/Al₂O₃. Reproduced with permission [102]. Copyright 2023 Elsevier.

A similar mechanistic scenario is observed in Ni/WO _x catalysts, where hydrogen spillover from Ni to the oxide surface induces protonation of W=O species. Operando D₂‐IR spectra provide direct evidence: Upon D₂ exposure, W=O vibrations weaken while characteristic W‐OD stretching bands emerge, confirming the reversible W=O to W‐OD transformation and the generation of in situ Brønsted acidic WO‐OH(D) sites that actively function during hydrocracking (Figure 10C, Table 1, Entry 19) [44]. Several works highlight the WO _x clusters on support surfaces exhibiting strong Brønsted acidity and enhanced hydrocarbon cracking activity [103], WO _x domains with tailored oxygen vacancy densities facilitating proton generation and olefin isomerization/cracking [104], and WO _x anchored on high‐surface‐area oxides demonstrating tunable surface acidity correlated with improved hydrocracking performance [105]. In Pt–Cl/Al₂O₃ systems, analogous cooperative behavior between metal sites and the support enhances surface acidity and cracking capability. Under typical hydrocracking conditions (∼300°C), Pt–Cl/Al₂O₃ catalysts show high conversion with dominant C₅–C₁₂ liquid fractions and reduced formation of light gases, consistent with in situ generation of Brønsted acidic environments that complement metal‐assisted hydrogenation steps (Figure 10D) [102].

From a design perspective, these observations highlight the importance of M=O moieties and their controlled hydrogenation to M–OH for acid functionality: Engineering terminal metal–oxygen double bonds and facilitating their conversion under reaction conditions is a viable strategy to impart strong, regenerable Brønsted acidity on nonzeolitic oxides, bridging the traditional gap between metal and acid functions in hydrocracking catalysts [106, 107].

2.3.3. Metal–Acid Balance

The performance of bifunctional catalysts in polyolefin hydrocracking is fundamentally governed by the synergistic balance between metallic and Brønsted acid sites, which dictates the prevalence of the “ideal” hydrocracking route [108]. This balance ensures the precise kinetic coupling of (de)hydrogenation and C–C scission; an optimal ratio facilitates the skeletal isomerization of olefinic intermediates followed by rapid hydrogenation, thereby maximizing liquid fuel selectivity [15] and overall catalytic activity (Figure 11A) [60, 109]. Conversely, an imbalance, particularly a deficit in metallic sites, promotes secondary cracking and hydrogen transfer, resulting in excessive light gas production and catalyst deactivation via coking. This complex trade‐off is exemplified in Ni/Beta catalysts, where Ni²⁺ ions can occupy exchange sites, reducing Brønsted acidity while providing the necessary hydrogenation capacity to achieve a volcano‐type activity peak [110]. Similarly, in NiMo/ZSM‐5 systems, the redistribution of acid strength through metal incorporation—converting strong sites to weak ones—enhances the synergy between cracking and saturation. Furthermore, the spatial proximity and relative strength of these sites determine the competition between β‐scission and isomerization, effectively steering the product distribution and branching degree. Ultimately, tailoring this metal–acid ratio remains the primary lever for steering reaction pathways toward high‐quality fuel fractions [111].

FIGURE 11.

(A) Reactivity‐mechanism map of the PE depolymerization reaction. Reproduced with permission [15]. Copyright 2024 Nature. (B) Activity‐mechanism map with the 18 tested catalysts positioned according to their activity and mechanism. Reproduced with permission [15]. Copyright 2022 Nature.

The metal–acid balance significantly impacts on the selectivity of the products. In Ni‐HZSM‐5, increasing the Ni metal loading gradually decreases the number of BAS, while significantly increasing the LASs [96]. The reaction activity improves further to 95%, with a marked increase in the selectivity of C₃–C₄ gas products and C₅–C₁₈ hydrocarbons in light oils by approximately 25%. Similarly, Ni–Ru/h‐Beta catalysts exhibit strong metal–acid synergy, where an optimal metal loading favors the production of light diesel fractions (C₁₃–C₁₈) [112].

The metal–acid balance plays a significant role in influencing the reaction pathways. In Ru/MOR catalysts, smaller Ru particle sizes promote the formation and desorption of olefin intermediates through hydrocracking reaction and suppress C–C bond cleavage through hydrogenolysis reaction [113]. Compared to the 1.7% Ru/MOR catalyst, the 0.7% Ru catalyst produces a higher yield of liquid fuel (C₅–C₂₁) at 31.7%, while larger Ru particles tend to generate lower‐value methane. Additionally, the distribution of acid sites on the zeolite surface influences Ru particle aggregation. In Ru/Y catalysts, the Ru loading correlates strongly with the acidity of Y‐zeolite. Lower metal–acid balance and lower C_liq/C_gas ratios correspond to reaction conditions more favorable for hydrocracking, whereas higher metal–acid balance and higher C_liq/C_gas ratios are indicative of hydrogenolysis [92]. At high Ru loading and low acid site concentration, the reaction proceeds through dehydrogenation of the reactant, followed by C–C bond cleavage at the metal sites (Figure 11B).

2.3.4. Spatial Distribution of Active Sites

PO molecules have large molecular sizes, and their initial hydrocracking requires both metal sites and strong acidity. Rational control of the spatial positioning of metal nanoparticles can prevent steric hindrance, which might otherwise limit the accessibility of large molecules to active sites. Studies using C₆–C₂₄ model alkanes on Pt–USY catalysts with spatially distinct Pt locations clarify the role of diffusion constraints (Figure 12A). When Pt is confined inside USY micropores, C₆ achieves higher conversion, reflecting the intrinsically superior activity of these internal sites once accessible. In contrast, C₂₄ exhibits limited access to internal Pt and attains markedly higher conversion when Pt is positioned on the external surface. These trends indicate that, while internal Pt sites possess higher intrinsic activity, their utilization is restricted by size‐dependent mass‐transfer limitations. Consequently, for macromolecular substrates such as PE, the spatial distribution and accessibility of active sites become decisive design parameters, outweighing intrinsic site activity in governing apparent catalytic performance [21, 115]. In the hydrocracking of PO, Pt nanoparticles loaded exclusively on the surface of USY zeolite work synergistically with the zeolite's acidic sites to efficiently initiate the cracking of PO into intermediates, which then enter the zeolite pores for secondary cracking (Figure 12B). The spatial position of metal loading in bifunctional catalysts also influences the efficiency of the hydrocracking reaction. In Pt/Hβ catalysts, trace Pt nanoparticles loaded both on the surface and within the pores achieve deep isomerization and cracking of PO, generating liquid products with high branching characteristics (Figure 12C) [114]. At 250°C, the hydrocracking rate of LDPE is 30,000 g LDPE·g_Pt ⁻¹·h⁻¹, and even at 180°C, a short‐chain alkane yield of 98% is achieved, with a C₅–C₁₂ selectivity of around 80%.

FIGURE 12.

(A) Hydrocracking different n‐alkanes over Pt/USY catalysts with various metal–acid proximity. (B) The proposed mechanism of hydrocracking of PE on Pt/USY catalysts with different metal–acid proximity. Reproduced with permission [21]. Copyright 2023 Royal Society of Chemistry. (C) Schematic illustration of the hydrocracking of LDPE and alkane‐LDPE over Pt _n /Hβ. Reproduced with permission [114]. Copyright 2025 American Chemical Society.

Mesoporous or hierarchical pore structures can effectively enhance the accessibility of acid sites to PO. The diffusion of substrates and intermediates plays a critical role in determining both the reaction rate and product selectivity. For instance, the introduction of mesopores into the one‐dimensional pore channels of MOR zeolites to create HyMOR catalysts resulted in an HDPE conversion rate of 83.6% and a naphtha selectivity of 80% after 4 h at 250°C [116]. The mesopores enhance the accessibility of B‐acid sites to HDPE, facilitating faster diffusion of intermediates on the catalyst surface. Similarly, the incorporation of mesoporous nanosheets on the surface of USY zeolites creates a hierarchical pore structure [63]. As the mesopore content increases, the PE conversion rate improves to 80% at 280°C, with liquid product selectivity exceeding 90%. In this system, the C=C bond cleavage, which is the rate‐limiting step, primarily occurs on the outer surface and mesoporous regions.

Subsequently, intermediate products diffuse into micropores, where further cracking and isomerization reactions occur (Figure 13A). This process enables a synergy between selective catalysis and efficient mass transfer. Hierarchical Hie‐TS‐1 zeolites have also been utilized for the conversion of LDPE into liquid alkanes, achieving a liquid alkane yield of 94.0%, with C₅–C₇ light alkanes contributing 84.8% (Figure 13B). The hierarchical structure ensures that olefins of appropriate size can rapidly diffuse into the micropore–mesopore regions, significantly enhancing catalyst efficiency and producing a narrow product distribution [108].

FIGURE 13.

(A) Hydrocracking of PE over Pt/Al₂O₃ mixed with different Y zeolites and proposed mechanism. Reproduced with permission [63]. Copyright 2024 American Chemical Society. (B) TEM images and corresponding model illustrations of Pt@TS‐1 and Pt@Hie‐TS‐x catalysts, and their catalytic performance. Reproduced with permission [108]. Copyright 2024 Wiley.

Compared with microporous supports, mesoporous and hierarchically porous oxides combine spatial confinement with improved mass transport, which together limit secondary cracking and coke formation while enabling access of long polymer chains to active sites. A clear example is the mSiO₂/Pt/SiO₂ core–shell catalyst: Placing Pt nanoparticles at the base of 1.7–3.5 nm mesopores both suppresses sintering and enforces a processive interaction with PE, producing a narrow liquid distribution centered on C₁₂–C₁₆ (≈40% of liquid species) and, for isotactic PP, a 79% liquid yield (C₉–C₁₈) versus 52.8% for nonporous Pt/SiO₂ (Figure 14A) [117]. Moreover, an oxide‐only embodiment of the same design principle is L‐ZrO₂@mSiO₂ [118], in which ultrasmall amorphous zirconia nanoparticles are confined between mesoporous silica platelets (Figure 14B). Under standard hydrogenolysis conditions (300°C, ≈0.99 MPa H₂), this architecture delivers mass‐specific activity competitive with Pt and a narrow, C₁₈‐centered liquid distribution (C₉–C₂₇ > 90%).

FIGURE 14.

(A) An analogous mechanism proposed for the mSiO₂/Pt/SiO₂ catalyst, in which SiO₂‐supported Pt nanoparticles (orange) are located at the end of nanopores in the mSiO₂ shell. Reproduced with permission [117]. Copyright 2022 Nature. (B) The preparation of L‐ZrO₂@mSiO₂ and PE conversion performance over L‐ZrO₂@mSiO₂ catalysts. Reproduced with permission [118]. Copyright 2023 Nature. (C) Reaction pathway illustration on Ru‐WZr catalysts and relevant catalytic performance. Reproduced with permission [119]. Copyright 2025 American Chemical Society.

Core–shell or dual‐domain oxide architectures further decouple metal and acid functions to steer pathway selectivity. For example, Ru‐WO₃/ZrO₂ at 250°C and 30 bar H₂ produces highly isomerized, branched jet‐fuel–range hydrocarbons (C₈–C₁₆) with high liquid selectivity, and the metal‐to‐acid balance (MAB) controls the degree of isomerization versus cracking (Figure 14C) [119, 120]. Tandem oxide designs such as Ni–WO₃/Al₂O₃ combined with Beta zeolite demonstrate the same principle at scale: A Ni–WO₃/Al₂O₃ with Beta system achieves ∼86% yield to C₅–C₁₆ liquid hydrocarbons from PE at 280°C and 4 MPa H₂ (4 h), with negligible methane, by effecting primary cracking on accessible WO _x ‐modified oxide surfaces and subsequent zeolitic polishing [23]. The large amount of olefin intermediates produced during cracking needs to be promptly hydrogenated by metal sites. If metal particles are aggregated externally while acid sites are located internally, olefins require long‐distance diffusion for hydrogenation, easily leading to coke formation. Therefore, controlling the spatial proximity between metal and zeolite is key to achieving efficient synergy.

2.4. Catalysts for Oxidation of PO

Oxidative conversion of PO represents an alternative approach to thermal cracking and hydroconversion for cleaving inert C–C bonds in polymer backbones, enabling the selective formation of oxygenated products. Thermally or photocatalytically activated oxygen species attack C–H/C–C bonds to produce oxygenated molecules such as carboxylic acids, ketones, and aldehydes, with reactivity largely governed by catalyst redox properties and the availability of surface oxygen species [121, 122]. In recent catalytic oxidative strategies, transition‐metal catalysts and heterogeneous photocatalysts have been shown to facilitate radical‐mediated C–C bond oxidation and subsequent scission, with improved selectivity over simple combustion pathways [47, 123, 124, 125].

The oxidation of PO represents an important upgrading pathway for plastic recycling, which is strongly governed by the synergistic interaction between metal active sites and catalyst supports. Metal species facilitate the activation of molecular oxygen and promote selective oxidative cleavage of polymer chains, while the structural and chemical properties of the support such as porosity, acidity, and confinement effects play a crucial role in regulating mass transfer, stabilizing intermediates, and controlling product selectivity.

2.4.1. Metal Components

Catalytic oxidation is another important strategy for converting PO into oxygenated compounds. Metal components can activate oxygen from the gas phase to achieve selective oxidation. The carrier regulates the chain length of the product, enabling selective oxidation. In Co/MCM‐41, cobalt atoms are atomically dispersed within the zeolite framework, providing oxidation‐active sites [46]. Under mild conditions (125°C), PE shows a high carbon yield of 85.9%, with direct oxidation upgrading into saturated dicarboxylic acids (Table 1, Entry 21). The yield of long‐chain dicarboxylic acids (C₁₀–C₂₀) can reach up to 58.9%. The amount of metal in the catalyst is crucial during oxidation. The Si/Co ratio affects the chain length of the products; the higher the Si content, the more dicarboxylic acids of longer chains (C₁₀–C₂₀) are produced (Figure 15). The regular mesoporous channels of the carrier (2–6 nm) create a nanoconfinement effect on the reaction intermediates, enabling precise control over the product distribution from PE chain scission [47].

FIGURE 15.

Product compositions over Co‐MCM‐41 with different Si/Co ratios and pore diameter and schematic degradation mechanism of PE. Reproduced with permission [46]. Copyright 2024 Wiley.

2.4.2. Photocatalysts

Photocatalysis provides a promising, sustainable pathway for the selective degradation of aromatic plastics such as PS under mild conditions, with the added advantage of potential molecular‐level selectivity control (Figure 16A) [126]. In mechanistic and performance‐oriented studies, this work demonstrated that visible‐light photocatalytic oxidation of PS over graphitic carbon nitride (g‐C₃N₄) catalysts at 150°C under 10 bar O₂ with a 300 W xenon lamp can achieve >90% PS conversion, yielding 0.36 g of aromatic oxygenates from 0.50 g PS feedstock, with benzoic acid as the dominant product (Figure 16B, Table 1, Entry 22) [47]. A complementary photocatalytic strategy employs UV‐assisted degradation to realize efficient PS backbone cleavage under relatively mild conditions. In this system, PS was irradiated using a 300 W UV source at ambient pressure, where time‐resolved analysis revealed rapid molecular‐weight reduction, confirming effective chain scission. The resulting low‐molecular‐weight aromatic intermediates were subsequently transformed into well‐defined aromatic products such as diphenylmethane, highlighting that UV‐induced depolymerization can generate reactive intermediates suitable for selective downstream catalytic upgrading (Figure 16C, Table 1, Entry 23) [48].

FIGURE 16.

(A,B) Photocatalytic PS plastic upgrading over g‐C₃N₄ catalyst; conversion of 500 mg plastic pellets in 20 reaction cycles and effects of pretreatment on PS conversion rate. Reproduced with permission [47]. Copyright 2025 American Chemical Society. (C) Comparison of the conventional depolymerization–repolymerization approach to the tandem Deg‐Up strategy. Reproduced with permission [48]. Copyright 2022 PNAS.

Although PS has demonstrated clear potential for photocatalytic degradation and upcycling, studies on other ester‐containing polyesters such as PLA and PET remain relatively limited, mostly focusing on photocatalytic hydrolysis and oxidative bond cleavage. Several photosensitized catalytic systems have shown that, under UV irradiation, PET ester bonds can be cleaved to generate terephthalic acid and other oxygenated molecules, highlighting the crucial roles of catalyst electronic structure regulation and charge‐separation efficiency in determining polyester degradation performance [127, 128]. Overall, photocatalytic upcycling of PS is evolving from feasibility validation toward mechanistic elucidation and system optimization. Through rational design of photocatalysts, including light‐absorbing components, reactive oxygen species generation pathways, reaction media, and irradiation conditions, highly selective production of aromatic oxygenates can be achieved under mild conditions, while providing valuable methodological insights for the photocatalytic treatment of polyester plastics such as PLA and PET.

3. Catalyst Deactivation

Under realistic reaction conditions involving elevated temperatures, high pressures, and the presence of impurities, catalyst deactivation is inevitable in the catalytic upcycling of PO waste. Deactivation could originate from three typical interrelated factors: metal sintering, poisoning, and coke formation. A comprehensive understanding of these deactivation mechanisms is essential for catalyst design, regeneration, and process optimization [129].

3.1. Changes in the State of Metal

In PO upcycling, metal nanoparticles tend to migrate and sinter under thermal conditions, resulting in particle growth and a corresponding loss of active surface area (Figure 17A) [130]. This structural evolution significantly affects catalytic performance and product selectivity. When metal particles grow from subnanometer or small nanometer sizes to larger aggregates, the number of accessible active sites decreases and hydrogenation efficiency declines.

FIGURE 17.

(A) Conceptual models representing main mechanisms of metal particles sintering. Reproduced with permission [130]. Copyright 2020 Elsevier. (B) HADDF‐STEM images and corresponding Ni size distribution of fresh and reused Ni/Beta. Reproduced with permission [97]. Copyright 2025 Elsevier. (C) Proposed reaction mechanism for PVC oxidation over Ru/HZSM‐5 and Ru‐2Cu/HZSM‐5 catalysts and the concentrations of chlorinated by‐products as a function of temperature over the Ru‐2Cu/HZSM‐5 and Ru‐2Cu‐5Nb/HZSM‐5 catalysts. Reproduced with permission [131]. Copyright 2024 Royal Society of Chemistry. (D) Hydrocracking performance of real‐world plastic systems, including postconsumer plastics and LDPE containing over MoS _x ‐H‐Beta catalysts. Reproduced with permission [14]. Copyright 2024 American Association for the Advancement of Science.

For example, in Ru‐based catalysts, Ru particles were observed to grow from approximately 2–3m to over 4 nm during reaction, leading to a gradual decrease in liquid product yield and a concurrent increase in methane formation [81, 115]. Similar sintering behavior has also been reported for other metals such as Pt and Ni after multiple catalytic cycles [21, 110]. In the case of 0.5Ni/Beta catalysts [97], transmission electron microscopy (TEM) images revealed that after five recycling cycles, metal particles migrated and agglomerated, with average particle sizes increasing from 3.1 to 6 nm (Figure 17B). This aggregation significantly reduced the accessible metal surface area and consequently lowered catalytic activity.

To suppress the migration and sintering of metal nanoparticles during reaction, various strategies can be employed to strengthen metal–support interactions. Common approaches include constructing SMSIs, confining metal species within zeolite pores or cage‐like structures and introducing a second metal to form alloyed or bimetallic sites. The introduction of promoters and chelating agents can effectively enhance metal dispersion and stability. In PtSn‐0.5Ce/SiAl catalysts [132], Ce incorporation suppresses Pt sintering, maintaining particle size and morphology after multiple recycling cycles, whereas Ce‐free PtSn/SiAl shows significant particle growth. This stabilization arises from strong Pt–O–Ce interactions, which weaken Pt–Pt bonds and strengthen Pt–O bonds, preventing metal aggregation under reaction conditions. Similarly, in 0.5Ni/Beta catalysts [133], EDTA chelation during synthesis reduces Ni nanoparticle aggregation, yielding highly dispersed particles (∼2 nm) and promoting uniform metal distribution. These strategies collectively highlight the role of chemical promoters and synthesis control in improving metal dispersion, structural stability, and catalyst durability. Moreover, metal oxides can act as effective antisintering matrices: For example, Mg–Al mixed oxides (as in Ru/Mg₁Al‐LDO) establish SMSIs and oxygen‐mediated anchoring sites that suppress Ru migration and coalescence under reaction conditions. These oxide‐derived anchoring (and SMSI‐like) effects thus provide a general strategy to kinetically and thermodynamically inhibit metal growth while preserving active‐site accessibility [33]. These strategies enhance metal dispersion and thermal stability, thereby prolonging catalyst lifetime during PO conversion.

3.2. Impurity Poisoning

PO feedstocks often contain polymerization catalysts, additives, or contaminants (e.g., Ti, Fe, and Ca), which can promote undesired side reactions and accelerate catalyst deactivation. For instance, in model studies using n‐undecane, increasing the pyrrole concentration to a nitrogen content of 3000 ppm resulted in nearly a 50% decrease in both propane–propylene fraction (PPF) and butane–butene fraction (BBF) yields [134, 135]. This performance loss arises from the strong adsorption of nitrogen‐containing compounds on acidic sites, which suppresses Brønsted acidity [136].

Chlorine contamination, typically originating from PVC, represents another major deactivation pathway. Beyond heterogeneous catalysts, ionic liquid–mediated catalysis has exhibited excellent performance in the conversion of mixed PVC and PET waste into dichloroethane and terephthalic acid, demonstrating their capability in processing Cl‐containing plastics [133]. During thermal processing, chlorine is released as HCl, which can destroy the zeolite framework and irreversibly deactivate active sites. To mitigate this issue, the incorporation of Nb into Ru–Cu/HZSM‐5 catalysts [131] has been shown to suppress the formation of highly toxic by‐products such as trichloroethylene and tetrachloroethylene (Figure 17C). Incorporating MoS _x components into Hβ catalysts demonstrate enhanced tolerance to impurities and resistance to poisoning during the upgrading of PO. MoS _x as a classical hydrodesulfurization catalyst, inherently exhibits remarkable tolerance to heteroatoms such as sulfur and nitrogen. During the hydrocracking of real waste plastics, such as plastic bags, containing impurities, hydrocarbon yields remained above 80% (Figure 17D) [14]. Highly dispersed MoS _x nanoparticles interact strongly with the Hβ zeolite support, which stabilizes the metal species and prevents contaminants from blocking or deactivating active sites.

To mitigate catalyst deactivation caused by impurities, the resistance to poisoning can be improved through rational optimization of catalyst composition and structural design. The introduction of scavenging additives such as alkali metals, rare‐earth elements, or transition metal oxides can preferentially bind contaminant species, thereby protecting active sites. Moreover, constructing corrosion‐resistant catalyst architectures or employing support with buffering capacity helps alleviate framework degradation induced by halogens or nitrogen‐containing compounds.

3.3. Coke Formation

During catalytic conversion of PO, hydrocarbon intermediates probably undergo condensation and cyclization reactions, leading to the coke formation [130, 137]. For porous zeolite catalysts, coke deposition severely restricts pore accessibility and reduces catalytic activity. In catalytic cracking reactions, HZSM‐5 exhibited a 19% decrease in BET surface area and a 15% reduction in micropore area due to coke deposition at 500°C, 15 h. Even more severe deactivation was observed for H‐Beta and HY zeolites, whose BET surface areas decreased by approximately 55%, indicating extensive pore blockage [35]. In particular, the design of b‐axis‐shortened ZSM‐5 [38] or meso‐ZSM‐5 [89] zeolites can effectively enhance mass transfer of PO, thereby reducing coke formation.

TEM images revealed the presence of structured, filamentous coke species on the external surfaces of HY zeolite catalysts (Figure 18A) [35]. Further insights were obtained from ¹H NMR spectroscopy of coke‐related species extracted from spent catalysts. Signals corresponding to aromatic –CH _x (1.7–3.5 ppm) and aromatic protons (6.6–10 ppm) were identified [138]. With increasing zeolite pore size, the aromatic proton signals shifted toward higher chemical shifts, indicating a higher degree of aromatic condensation, particularly in coke formed on HY zeolite (Figure 18B).

FIGURE 18.

(A) TEM images of the different catalysts after cracking polyolefin. a, HZSM‐5 zeolite; b, H‐Beta zeolite; c, HY zeolite. Reproduced with permission [35]. Copyright 2011 Elsevier. (B) ¹H‐NMR of the soluble coke extracted from spent HZSM‐5, H‐Beta, and HY catalysts. (C) Combined FTIR spectra and coke combustion at different temperatures of the spent catalysts. Reproduced with permission [138]. Copyright 2011 Elsevier.

It is critical to recognize that catalyst deactivation pathways rarely occur in isolation but instead act synergistically. For example, metal sintering reduces hydrogenation capacity, thereby accelerating coke formation; accumulated coke can encapsulate metal particles and hinder regeneration [139]; and impurity‐induced framework degradation alters acid‐site distribution, further exacerbating coke deposition. Understanding these coupled deactivation mechanisms is therefore essential for the rational design of more robust catalysts and for improving long‐term performance in PO upcycling processes.

Catalyst regeneration is crucial for maintaining long‐term catalytic performance, and FTIR spectroscopy provides valuable insights into regeneration behavior under different combustion temperatures [140, 141]. The disappearance of characteristic bands associated with aliphatic and aromatic species (1460 and 1590 cm⁻¹) indicates effective removal of hydrocarbon deposits and successful catalyst regeneration (Figure 18C) [138]. To mitigate pore blockage and activity loss caused by coke formation, catalyst pore architecture, acidity distribution, and reaction conditions can be rationally tailored. In particular, hierarchical porosity enhances mass transport and reduces the residence time of bulky intermediates, while optimized acid site density and strength help maintain cracking activity and suppress excessive aromatization. Controlled calcination or reduction further enables efficient coke removal while minimizing structural degradation.

4. Reactor Design and Mass Transfer Enhancement

PO are solid under ambient conditions, and they melt into highly viscous non‐Newton fluids at elevated temperatures. As a result, their efficient conversion requires optimization of mass‐transfer processes across multiple length scales. At the microscopic level, mass transfer is primarily governed by the pore architecture of the catalyst. Constructing hierarchical pore structures and extended external surface area is widely regarded as an effective strategy to enhance molecular accessibility and optimize reaction pathways, as discussed in Sections 3 and 4.

At the macroscopic scale, mass‐transfer efficiency is closely related to catalyst particle morphology (e.g., size and shape), reaction conditions, packing structure, and reactor configuration. Therefore, reactor engineering must be cooperatively integrated with catalyst structural design to simultaneously regulate microscopic accessibility and macroscopic transport behavior. Such multiscale optimization is essential for achieving efficient, stable, and scalable PO conversion processes.

4.1. Reactor Selection and Design

As the core unit in PO catalytic upgrading, reactor configuration, operating mode, and mass‐transfer characteristics critically determine overall process efficiency, product distribution, and economic feasibility [70]. Batch reactors are widely used in lab‐scale studies and serve as essential platforms for catalyst screening and optimization of reaction conditions [132].

In PO conversion, conventional pyrolysis reactors often exhibit limited selectivity toward high‐value products such as light olefins or BTX aromatics [135, 142]. In batch operation mode, mechanically stirred reactors can provide strong shear forces, which effectively reduce the high viscosity of molten polymers, enhance gas–liquid mass transfer, and minimize local concentration and temperature gradients. Moreover, the ability to precisely control gas partial pressures under high‐pressure conditions makes stirred batch reactors particularly suitable for evaluating the intrinsic catalytic activity of materials (Figure 19A,B). Batch reactors demonstrate versatility for both reductive and oxidative transformations, enabling high PO conversion and product selectivity under well‐controlled conditions [46, 94]. Continuous stirred‐tank reactors (CSTRs) allow continuous or semicontinuous feeding and product removal, making them suitable for studying reaction kinetics and maintaining uniform reaction conditions. Their advantages include excellent mixing, temperature control, and steady‐state operation. However, due to the difficulty of operating batch reactors continuously, fixed‐bed reactors—which provide stable flow and uninterrupted operation—are particularly advantageous for evaluating continuous catalytic performance, especially under high‐temperature, low‐pressure conditions. Therefore, fixed‐bed reactors are often employed for the continuous catalytic cracking and aromatization of PE [91].

FIGURE 19.

(A) Schematic summary of the open‐batch reactor design and mechanism. Reproduced with permission [94]. Copyright 2025 Nature. (B) Reaction diagram of catalytic oxidation pf PE over Co/MCM‐41 in pressurized batch reactor. Reproduced with permission [46]. Copyright 2024 Wiley. (C) Modes of the s‐ZSM‐5 and Zn/meso‐ZSM‐5 with different proximity in a fixed‐bed reactor. Reproduced with permission [39]. Copyright 2023 Royal Society of Chemistry. (D) Catalyst particles coated with fused plastic and their movement in the conical spouted bed reactor. Reproduced with permission [35]. Copyright 2011 Elsevier.

To spatially separate different catalytic functions, it is necessary to regulate the packing structure of the catalyst, in which s‐ZSM‐5 and Zn/mesoporous ZSM‐5 are arranged in distinct zones (Figure 19C). This design allows depolymerization and aromatization to proceed independently. At 400°C, such systems achieve up to 76.7% selectivity toward methylated aromatics while effectively avoiding mutual interference between catalytic functions [39]. Similarly, two‐stage reactor systems can isolate catalyst deactivation caused by impurities, achieving aromatic selectivity above 77% and significantly prolonging catalyst lifetime, albeit at the cost of increased system complexity and capital investment [143]. However, fixed‐bed reactors also have some inherent drawbacks. Their heat transfer efficiency is relatively low, making them prone to hot spots in strongly exothermic reactions, which can lead to localized catalyst deactivation or increased side reactions. For processes involving high‐viscosity molten polymers, fixed‐bed reactors may also face issues with excessive pressure drop and uneven distribution.

To further reduce operational costs, conical spouted‐bed reactors have been developed, enabling integrated processes including plastic melting, catalyst coating, and catalytic upgrading. In these systems, N₂ serves as the carrier gas to enhance particle circulation and prevent agglomeration of catalyst‐coated plastic particles during collisions (Figure 19D). For example, using HZSM‐5 as the catalyst, PE cracking at 500°C for 15 h yields approximately 60 wt% of C₂–C₄ olefins (29 wt% propylene, 21 wt% butene, and 10 wt% ethylene), along with ∼10 wt% monoaromatic hydrocarbons [35]. In addition, fluidized catalytic cracking (FCC) reactors, such as circulating fluidized beds (CFBRs), offer enhanced gas‐solid contact and mass‐transfer efficiency. When combined with microspherical catalysts (P‐modified ZSM‐5), continuous reaction–regeneration operation can be achieved, delivering total C₂–C₄ olefin yields exceeding 44% together with high BTX selectivity [144]. Furthermore, dual‐fluidized‐bed systems decouple thermal pyrolysis from catalytic aromatization, increasing BTX yields to above 22.3 wt% while maintaining good operational stability [145].

Reactor choice for polyolefin catalytic upgrading affects reaction control, product selectivity, catalyst stability, and operational efficiency. Batch autoclave reactors, leveraging the advantages of mechanical stirring, enable shorter contact times and are more suitable for converting high‐viscosity polyolefins into olefins or alkanes under high‐pressure conditions. Fixed‐bed reactors offer flexible adjustment of the packing structure and are capable of converting polyolefins into olefins or aromatics under continuous, high‐temperature, and atmospheric‐pressure conditions. Spouted‐bed reactors utilize gas flow to drive particle circulation, enabling the efficient conversion of polyolefins into light olefins and aromatics under moderate temperature conditions. Therefore, the optimal reactor selection should be based on a comprehensive evaluation of the specific operating conditions and process

4.2. Mass Transfer Enhancement

Batch reactors are commonly employed at laboratory scales for PO conversion; however, they are typically operated in a discontinuous mode and are difficult to scale up considering the mass‐transfer limitations, particularly for gas–liquid transport. In systems using nonporous catalysts such as Ru/TiO₂, internal diffusion limitations can often be neglected, whereas for porous materials (e.g., zeolites), internal mass transfer must be carefully considered [13]. Kinetic analyses indicate that the intrinsic reaction rate can be approximately five times higher than the hydrogen diffusion rate, confining the effective reaction zone to a narrow region near the gas–liquid interface, typically on the millimeter scale (Figure 20A).

FIGURE 20.

(A) Characteristic lengths in catalytic polyolefin hydrogenolysis. (B) Influence of catalyst particle motion on catalytic performance. Reproduced with permission [13]. Copyright 2025 Nature.

To enhance hydrogen mass transfer, optimization of catalyst particle size and stirring configuration is essential. For example, the characteristic chain dimension of HDPE (Mw ≈ 200 kDa) is approximately 22 nm according to the freely jointed chain (FJC) model, and diffusion efficiency increases as molecular weight decreases. During PO hydrocracking, smaller catalyst particles (0–0.2 mm) provide about 40% higher C₁–C₄₅ product yields compared with larger particles (0.4–0.6 mm), owing to their improved mobility and more uniform spatial distribution within the reactor. Larger particles tend to settle and distribute unevenly, thereby limiting effective mass transfer. Stirring mode also strongly influences product distribution: although the total yield varies only slightly, the gas–liquid product ratio changes markedly. Impeller‐type stirring favors liquid product formation (≈50% yield), followed by propeller and turbine configurations (Figure 19B). This behavior arises because impeller stirring promotes frequent transport of catalyst particles into hydrogen‐rich zones, whereas turbine stirring restricts such transport, limiting reaction efficiency [13].

During the polyolefin hydrocracking process, macroscopic operating conditions—temperature, pressure, and stirring—collectively regulate mass transfer efficiency through their coupled effects. Temperature is a key factor affecting melt viscosity: As temperature rises, the motion of long‐chain polyolefin molecules intensifies and intermolecular interactions weaken, causing melt viscosity to decrease exponentially [146]. This improved fluidity enables catalyst active sites to access macromolecular chains more easily. Stirring rate plays a crucial role in batch reactors; high shear forces not only further break up the high‐viscosity melt and reduce aggregate size but also substantially increase the gas–liquid–solid contact area, enhancing gas (e.g., H₂) diffusion from the gas phase into the melt and onto the catalyst surface. Although pressure has a relatively minor direct impact on viscosity, it indirectly influences the system by increasing hydrogen solubility in the melt, thereby mitigating coke precursor formation caused by localized hydrogen deficiency [68]. However, at elevated temperatures, if the mass transfer rate cannot keep pace with reaction consumption, the process enters a mass‐transfer‐limited regime. Therefore, practical operation requires optimizing the interplay among these three parameters to match diffusion and reaction rates, preventing catalyst deactivation or selectivity loss from excessive local gradients.

At the microscopic scale, molecular dynamics (MD) simulations provide molecular level insights into mass transfer mechanisms. By constructing complex model systems comprising polyolefin chains, melt environments, and catalyst pore structures, MD simulations track the diffusion trajectories of small molecules within the viscous medium. Results indicate that hydrogen's diffusion coefficient in polyolefin melts is significantly influenced by free volume distribution and segmental dynamics [147]. Furthermore, MD simulations visualize the adsorption, reaction, and desorption behavior of olefin intermediates on catalyst surfaces, revealing how viscosity hinders molecular conformational changes [148]. This provides theoretical support for understanding the microscopic phenomena underlying macroscopic mass transfer coefficients, bridging the gap between operational conditions and fundamental transport mechanisms. For example, molecular dynamics simulations demonstrate that, under high‐pressure conditions, small amounts of ethylene can significantly reduce the density of polymer solutions, thereby providing molecular‐level insight into the intermolecular interactions among polymer, solvent, and monomer species and elucidating their influence on the thermophysical properties of the system [149].

In highly viscous polyolefin systems, mass transfer is often severely constrained by limited convection and extremely low molecular diffusivity. Computational fluid dynamics (CFD) provides a powerful macroscopic tool to visualize flow fields, quantify boundary‐layer thickness, and evaluate gas–liquid–solid phase distribution [150]. For instance, CFD simulations provide valuable insights into reactor design by visualizing hydrogen‐melt interfacial behavior under different stirring modes and speeds [13]. Both propeller and impeller configurations significantly expand the gas–liquid interfacial area, thereby enhancing reaction performance. Simulations also successfully reproduce the characteristic “V‐shaped” hydrogen‐melt interface observed in highly viscous polymer systems, validating the reliability of the model. Similarly, CFD analyses of impeller‐driven systems reveal superior particle circulation and interfacial mass transfer. The stirring power number can therefore serve as an effective descriptor for quantifying mixing intensity and mass‐transfer efficiency, offering guidance for rational reactor optimization [150, 151].

5. Conclusions and Outlook

Catalytic upcycling of PO and PS waste offers a promising pathway to mitigate environmental pollution while recovering value from fossil‐derived carbon resources. Advances to date indicate that neither intrinsic active site chemistry nor process engineering alone dictates performance; rather, catalytic efficiency results from the coordinated optimization of metal and acid functionalities, spatial arrangement of active sites, hierarchical porosity, and resistance to deactivation. Selection among hydrogenolysis, catalytic cracking, hydrocracking, and selective oxidation must be aligned with architectures capable of accommodating macromolecular substrates and alleviating diffusion constraints across multiple length scales. In essence, performance hinges on reconciling high intrinsic reactivity with accessible catalytic environments that sustain activity, selectivity, and stability.

Looking ahead, further progress may arise from innovation across catalyst design, reaction engineering, and system‐level integration. Opportunities include the exploration of earth‐abundant catalytic materials, the tailoring of spatial distributions and transport networks within hierarchical structures, and reactor configurations that alleviate melt viscosity and enhance mass transfer. Addressing real‐world feed heterogeneity, additive‐related interference, and long‐term deactivation will likely benefit from more robust catalyst formulations, adaptive separation strategies, and process schemes aligned with circularity principles. Collectively, these directions suggest potential pathways toward bridging laboratory‐scale advances with commercial implementation, positioning catalytic upcycling as a meaningful contributor to circular carbon economies and sustainable materials management.

Funding

This work was supported by the National Natural Science Foundation of China (22538008, 22208144) and Key Technologies Research and Development Program (2024YFA1510003).

Conflicts of Interest

The authors declare no conflicts of interest.

Han Xinlei, Sun Jie, Zhang Jiuxuan, Qu Zhengyan, Jiang Hong, Tang Zhenchen, Chen Rizhi, ChemSusChem 2026, 19, e202600001. 10.1002/cssc.202600001