<?xml-model href='http://www.tei-c.org/release/xml/tei/custom/schema/relaxng/tei_all.rng' schematypens='http://relaxng.org/ns/structure/1.0'?><TEI xmlns="http://www.tei-c.org/ns/1.0">
	<teiHeader>
		<fileDesc>
			<titleStmt><title level='a'>Polymer Upcycling by Catalytic Hydrogenolysis: The Role of Polyolefin Short-Chain Branching</title></titleStmt>
			<publicationStmt>
				<publisher>American Chemical Society</publisher>
				<date>10/03/2025</date>
			</publicationStmt>
			<sourceDesc>
				<bibl> 
					<idno type="par_id">10681947</idno>
					<idno type="doi">10.1021/acscatal.5c04494</idno>
					<title level='j'>ACS Catalysis</title>
<idno>2155-5435</idno>
<biblScope unit="volume">15</biblScope>
<biblScope unit="issue">19</biblScope>					

					<author>Zachary M Gdowski</author><author>Rishi Raj</author><author>Aditya Bhan</author><author>K Andre Mkhoyan</author><author>Mahesh K Mahanthappa</author><author>Frank S Bates</author>
				</bibl>
			</sourceDesc>
		</fileDesc>
		<profileDesc>
			<abstract><ab><![CDATA[Not Available]]></ab></abstract>
		</profileDesc>
	</teiHeader>
	<text><body xmlns="http://www.tei-c.org/ns/1.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:xlink="http://www.w3.org/1999/xlink">
<div xmlns="http://www.tei-c.org/ns/1.0"><head>&#9632; INTRODUCTION</head><p>Polyolefins such as polyethylene (PE) and polypropylene (PP), which represent a majority of the single-use plastic produced worldwide, are a substantial component in the global plastic waste problem. <ref type="bibr">[1]</ref><ref type="bibr">[2]</ref><ref type="bibr">[3]</ref><ref type="bibr">[4]</ref> One potential means of addressing this issue is to chemically decompose or "upcycle" these relatively chemically inert polymers into useful low molecular weight products for reuse in other applications. <ref type="bibr">[5]</ref><ref type="bibr">[6]</ref><ref type="bibr">[7]</ref> Hydrogenolysis is an especially attractive catalytic upcycling process, wherein carbon-carbon bond cleavage by the addition of hydrogen converts the initial high molecular weight hydrocarbon polymers into value-added lower molecular weight compounds such as lubricants and fuel-range hydrocarbons. <ref type="bibr">[8]</ref><ref type="bibr">[9]</ref><ref type="bibr">[10]</ref> Over the past few years, a wealth of publications have examined the hydrogenolysis of commercial grade polyolefins using various heterogeneous transition metal catalysts, addressing the roles of the active metal, <ref type="bibr">[11]</ref><ref type="bibr">[12]</ref><ref type="bibr">[13]</ref><ref type="bibr">[14]</ref><ref type="bibr">[15]</ref> support architecture, <ref type="bibr">[14]</ref><ref type="bibr">[15]</ref><ref type="bibr">[16]</ref><ref type="bibr">[17]</ref><ref type="bibr">[18]</ref><ref type="bibr">[19]</ref> and other reaction conditions <ref type="bibr">13,</ref><ref type="bibr">16,</ref><ref type="bibr">[19]</ref><ref type="bibr">[20]</ref><ref type="bibr">[21]</ref><ref type="bibr">[22]</ref><ref type="bibr">[23]</ref><ref type="bibr">[24]</ref><ref type="bibr">[25]</ref><ref type="bibr">[26]</ref><ref type="bibr">[27]</ref><ref type="bibr">[28]</ref><ref type="bibr">[29]</ref><ref type="bibr">[30]</ref><ref type="bibr">[31]</ref><ref type="bibr">[32]</ref><ref type="bibr">[33]</ref><ref type="bibr">[34]</ref><ref type="bibr">[35]</ref> to address the upcyclability of various commercial and postconsumer plastic waste streams. <ref type="bibr">[12]</ref><ref type="bibr">[13]</ref><ref type="bibr">[14]</ref><ref type="bibr">[15]</ref><ref type="bibr">[20]</ref><ref type="bibr">[21]</ref><ref type="bibr">[22]</ref><ref type="bibr">[23]</ref><ref type="bibr">34,</ref><ref type="bibr">[36]</ref><ref type="bibr">[37]</ref><ref type="bibr">[38]</ref><ref type="bibr">[39]</ref><ref type="bibr">[40]</ref><ref type="bibr">[41]</ref><ref type="bibr">[42]</ref><ref type="bibr">[43]</ref><ref type="bibr">[44]</ref><ref type="bibr">[45]</ref><ref type="bibr">[46]</ref><ref type="bibr">[47]</ref><ref type="bibr">[48]</ref><ref type="bibr">[49]</ref> In catalytic refining of crude oil into liquid hydrocarbon reformates, bimetallic Pt-Re catalysts supported on inorganic oxide supports (e.g., SiO 2 and Al 2 O 3 ) have long been used for catalytic cracking, dehydrogenation, and hydrogenation reactions. <ref type="bibr">[50]</ref><ref type="bibr">[51]</ref><ref type="bibr">[52]</ref><ref type="bibr">[53]</ref><ref type="bibr">[54]</ref> Pt-Re/Al 2 O 3 modified catalysts are used industrially for naphtha reforming, with alloying characteristics and promoters (Cl, S) being added to maximize rates and limit hydrocracking. The addition of Pt is also known to facilitate reduction of ReO x species at much lower temperatures relative to conditions where no Pt is present. The specific roles of Pt and Re in alkane hydrogenolysis are not known although it is plausible that the mechanisms are similar to those in naphtha reforming. We refer the interested reader to the extensive literature on bimetallic catalysts, notably work by Sinfelt and coworkers, which led to the development of currently employed Pt-Re reforming catalysts. <ref type="bibr">[55]</ref><ref type="bibr">[56]</ref><ref type="bibr">[57]</ref> Approximately 30 years ago, Hucul, Hahn and coworkers at the Dow Chemical Co. <ref type="bibr">58</ref> realized the possibility of using such catalysts supported on ultrawide-pore mesoporous silicas for the complete hydrogenation of poly(styrene) to produce poly(cyclohexylethylene) (PCHE, sometimes called poly-(vinylcyclohexane) or PVCH) <ref type="bibr">[59]</ref><ref type="bibr">[60]</ref><ref type="bibr">[61]</ref><ref type="bibr">[62]</ref> as a high glass transition temperature (T g &#8776; 140 &#176;C) and high modulus engineering thermoplastic with useful optical properties. <ref type="bibr">63,</ref><ref type="bibr">64</ref> In order to effect this polymer saturation reaction in solution with commercially viable kinetics, they developed a precatalyst comprising Pt nanoparticles supported on islands of ReO x in a macroporous silica support with 0.1-1 &#956;m pore diameters. <ref type="bibr">58</ref> The ultrawide-pore silica support is a key feature of this bimetallic catalyst, as it enables facile diffusion of solvated polymers into and out of the catalyst pores that gives rise to commercially useful hydrogenation kinetics. Addition of rhenium to the Pt-based catalyst improves resistance to catalyst deactivation during commercial application in polymer hydrogenation as documented in a patent disclosure. <ref type="bibr">64</ref> Subsequent extension of this Pt-Re/SiO 2 -catalyzed hydrogenation process to cyclohexane solutions of poly(styrene)/ poly(butadiene) triblock and pentablock polymers is now a well-established route commercially practiced by USI Corp. for the production of poly(ethylene)/poly(cyclohexylethylene) copolymers, which are sold under the trade name ViviOn cyclic block copolymers. <ref type="bibr">65</ref> We and others have subsequently used this Pt-Re/SiO 2 catalyst (or a related catalyst with only Pt) for lab-scale hydrogenations of various model hydrocarbon polymers. <ref type="bibr">[66]</ref><ref type="bibr">[67]</ref><ref type="bibr">[68]</ref><ref type="bibr">[69]</ref><ref type="bibr">[70]</ref> However, we found that substitution of D 2 (g) as the saturation reagent leads to catalytic hydrogen-deuterium (H/D) exchange of both backbone and side chain hydrogens of hydrocarbon polymers, including commercial grades of polyethylene and polypropylene. The extent of H/D exchange depends on polymer chemical identity and molecular weight, solvent, catalyst loading, and reaction temperature. <ref type="bibr">[70]</ref><ref type="bibr">[71]</ref><ref type="bibr">[72]</ref><ref type="bibr">[73]</ref><ref type="bibr">[74]</ref> Experiments by Habersberger et al. demonstrated exchange of 32-68% of hydrogen for deuterium atoms on a commercial HDPE reacted over the Pt-Re/SiO 2 catalyst in hydrocarbon solvents. <ref type="bibr">71</ref> SEC-IR measurements showed that the extent of exchange differed significantly across the broad polymer molecular weight distribution. <ref type="bibr">72</ref> Zeng et al. and Kang et al.  reported nonmonotonic trends regarding isotopic exchange as a function of side chain branch density, <ref type="bibr">70,</ref><ref type="bibr">73</ref> and Zeng et al. demonstrated that deuterium exchange in hydrogenated polybutadienes resulted in preferential exchange with side chain versus backbone hydrogen atoms. <ref type="bibr">70</ref> Here, we note that these previous works were conducted without any catalyst activation by heat treatment under vacuum that impacts catalyst reactivity (see below).</p><p>More recently, Ertem et al. noted that Pt-Re/SiO 2 also effects limited degrees of polymer backbone cleavage at 170 &#176;C under 500 psig D 2 (g) at high catalyst loadings (1:1 w/w polymer:catalyst) over 17 h. <ref type="bibr">75</ref> Those studies indicated that solutions of polystyrene (PS) and high-density PE (HDPE) in isooctane did not degrade when reacted with Pt-Re/SiO 2 under D 2 (g), yet poly(ethylene-alt-propylene) (PEP) was reduced to half its starting molecular weight, and commercial linear low-density PE (LLDPE) was degraded to roughly 5% of its initial molar mass. These seemingly inconsistent results suggest that solution hydrogenolysis over this bimetallic catalyst depends on the type and distribution of pendant groups along the carbon backbone, consistent with the trends observed for H/D exchange as a function of side chain content. <ref type="bibr">70,</ref><ref type="bibr">73</ref> Also, Ertem et al. found that the analogous Pt/ SiO 2 catalyst, i.e., without Re, was much less active in LLDPE hydrogenolysis. Ultimately, the detailed atomic structures of these catalysts and the mechanisms by which Pt and Re synergistically act remain obscure due to the limited resolution of available characterization methodsrelevant to examine catalysts under working conditions, despite years of research. <ref type="bibr">[55]</ref><ref type="bibr">[56]</ref><ref type="bibr">[57]</ref> Here, we report experiments based on the series of model polyolefin materials depicted in Figure <ref type="figure">1</ref> that explicitly address the impact of chain microstructure on cyclohexane solution hydrogenolysis reactions catalyzed by Pt-Re/SiO 2 at 140 &#8804; T &#8804; 200 &#176;C under 500 psig H 2 (g). In order to assess how the density of short chain branching influences LLDPE degradation, our model substrates include a series of hydrogenated polybutadienes (hPBs) with narrow molar mass dispersities and 1.5-38 ethyl branches per 100 backbone carbon atoms (0.06 &#8804; x &#8804; 0.86 in Figure <ref type="figure">1</ref>). Additionally, we evaluated the extent of chain cleavage under these reaction conditions for a model HDPE, poly(ethylene-alt-propylene) (PEP), PCHE, and a commercial PCHE-block-PE-block-PCHE (CEC) triblock polymer. Apart from the model HDPE sample, these model polyolefins were derived from a tandem sequence of anionic polymerization and catalytic hydrogenation. We document polymer degradation ranging from virtually no change (PCHE) to a 100-fold reduction (hPB-06) in molecular weight depending on the type and density of side chains in the starting materials. These findings will aid in the design and scale-up of upcycling processes for converting a wide spectrum of commercially available polyolefins and their mixtures into useful low molecular weight chemical feedstocks.</p></div>
<div xmlns="http://www.tei-c.org/ns/1.0"><head>&#9632; EXPERIMENTAL METHODS</head><p>Materials. 1,3-Butadiene (99%, Sigma-Aldrich) was twice stirred over dry nBuLi at -20 &#176;C for 1 h and distilled under vacuum prior to use. Isoprene (99%, Sigma-Aldrich) was analogously purified by twice stirring over dry nBuLi at 0 &#176;C and distilling under vacuum. Styrene (99%, Sigma-Aldrich) was stirred and vacuum distilled at 40 &#176;C from CaH 2 , followed by twice vacuum distilling from nBu(Et)Mg. cis-Cyclooctene (95%, Acros Organics) was filtered through basic alumina prior to use. Anhydrous toluene was obtained from a Glass Contour Inc. solvent purification system, while anhydrous cyclohexane (CHX) and tetrahydrofuran (THF) were obtained from an inhouse purification system by passage through activated alumina and a Q5 deoxygenation catalyst. <ref type="bibr">76</ref> This THF was further purified by vacuum distillation from BuMgCl at ambient temperature. All other reagents were used as received. The CEC triblock copolymer, with M n = 35 kg/mol, &#272; = 1.02, and 62 wt % PCHE and 2.1 branches per 100 backbone carbons in the PE segment, <ref type="bibr">77</ref> was acquired from the Dow Chemical Company (this material is presently sold by the USI Corp., Taipei, Taiwan). <ref type="bibr">65</ref> Model Polydiene Library Synthesis. Air-and water-free anionic polymerizations were performed on a Schlenk line under an argon atmosphere per established procedures, with the detailed reagent amounts reported in Table <ref type="table">S1</ref>. <ref type="bibr">70</ref> Poly(butadiene) (PB) with 94% 1,4 and 6% 1,2 repeat units (referred to as PB-06), was obtained by reaction of 1,3butadiene with sBuLi in cyclohexane at 30 &#176;C for 19 h. Polybutadienes with tailored amounts of 1,2-repeat units denoted PB-x (0.33 &#8804; x &#8804; 0.86) were synthesized from reaction of 1,3-butadiene with nBuLi in cyclohexane at 8 &#176;C for up to 20 h in the presence of THF as a polar modifier with 2 &#8804; [THF]:[nBuLi] &#8804; 290 (see Table <ref type="table">S2</ref> for details). Polymerizations were terminated by addition of degassed methanol. Polystyrene (PS) and polyisoprene (PI) were synthesized by anionic polymerizations in cyclohexane at 40 &#176;C initiated with sBuLi and terminated by addition of either CH 3 OH or slightly acidic CH 3 OH (&#8764;100 &#956;L 37% w/w hydrochloric acid in 20 mL of methanol).</p><p>Catalytic Hydrogenation of Model Polydienes. All polydiene hydrogenation reactions were carried out in a Parr 236HC 1 L reactor under 500 psig H 2 (g) with magnetic stirring by a Teflon stir bar. A solution of PB-06 (&#8764;5.5 g) in toluene (500 mL) was added to the 1 L reactor and sparged with Ar(g) for 30 min. The reactor was then vented, opened to the atmosphere, and 180 mg tris(triphenylphosphine)rhodium(I) chloride (Wilkinson's catalyst) and 1.0 g triphenylphosphine were added. The sealed reactor was again sparged with Ar(g) for 30 min, pressurized with H 2 (g) to a static pressure of 500 psig, and heated to 100 &#176;C for 12-20 h. The resulting polymer, which we designate hPB-06, was recovered by two or three precipitations from hot toluene (&#8764;90 &#176;C) into cold methanol, which was accompanied by a change in polymer color from yellow-orange to white associated with removal of the catalyst. The resulting hPB-06 was dried at 50 &#176;C under vacuum. Since Wilkinson's catalyst is less effective for exhaustively hydrogenating PB-x with x &gt; 0.06, <ref type="bibr">78</ref> Pd/CaCO 3 (5 wt % Pd, reduced, Alfa Aesar) was used per a reported procedure <ref type="bibr">79</ref> to hydrogenate the remaining PB-x samples. Briefly, the catalyst (&#8764;5 g, 1:1 w/w polymer) was dried under vacuum in the Parr reactor overnight at 100 &#176;C followed by activation with 100 psig H 2 (g) at 100 &#176;C for 1 h. Upon cooling to room temperature and venting the H 2 (g), a 1 wt % PB-x in cyclohexane solution (500 mL) was added to the reactor under dynamic Ar(g) pressure. The reactor was then pressurized to a static pressure of 500 psig H 2 (g) and reacted at 100 &#176;C for 37-50 h, to produce hPB-x. PI was hydrogenated using an analogous procedure to produce poly(ethylene-alt-propylene) (PEP), and PS was similarly saturated at 120 &#176;C to yield poly(cyclohexylethylene) (PCHE). Polymers were recovered by pressure-driven membrane filtration (&#8764;80 psig Ar(g) head pressure) through Fisherbrand P8 filter paper and a 0.22 &#956;m Durapore membrane filter to remove the catalyst, followed by precipitation into -20 &#176;C methanol. PEP and hPB-x polymers were dried under vacuum at 50 &#176;C, and PCHE was vacuumdried at 100 &#176;C.</p><p>Reactions produced &gt;99% saturation as assessed by 1 H NMR (see Figures <ref type="figure">S7</ref>, <ref type="figure">S8</ref>, S15, S18, and S21 for representative traces of hydrogenated samples); see below for specific characterization conditions.</p><p>Synthesis of hPB-0. Poly(cyclooctene) of a targeted molecular weight M n = 100 kg/mol was synthesized by Rucatalyzed ring-opening metathesis polymerization (ROMP) of cyclooctene in toluene in the presence of cis-4-octene as a chain transfer agent with [cyclooctene]:[cis-4-octene] = 900:1. <ref type="bibr">80,</ref><ref type="bibr">81</ref> Cyclooctene (12 mL), toluene (5.5 mL), and cis-4-octene (16 &#956;L, diluted in toluene) were added to a roundbottom flask and sparged with Ar(g) for 30 min, followed by addition of Grubbs second generation catalyst (G2, 8.2 mg) in 1 mL toluene. The reaction was then heated to 40 &#176;C for 30 min, after which it was quenched with excess ethyl vinyl ether (100 &#956;L). PCOE was precipitated twice from hot toluene (&#8764;90 &#176;C) into CH 3 OH to yield a white polymer (indicative of catalyst removal) that was dried under vacuum at 50 &#176;C to remove residual solvent.</p><p>PCOE was hydrogenated to form "perfectly linear" poly-(ethylene) (referred to as HDPE or hPB-0) as previously reported. <ref type="bibr">80</ref> The ROMP reaction product was separated into two parts; &#8764;25 vol % of the solution was set aside for molecular characterization of the unsaturated PCOE as previously described. The remainder of the ROMP reaction product was diluted with toluene to reduce the viscosity, and that solution of PCOE, G2, and ethyl vinyl ether in toluene (150 mL total volume) was added to a Parr 452HC 0.3 L reactor and placed under inert atmosphere by sparging with 80 psig Ar(g).</p><p>The reactor was then sealed, pressurized to 500 psig H 2 (g), and heated to 100 &#176;C for 20 h. The resulting polymeric reaction product was isolated by two precipitations from hot toluene (&#8764;90 &#176;C) into cold CH 3 OH, followed by drying under vacuum at 50 &#176;C. Note that the presence of ethyl vinyl ether in the hydrogenation reaction supresses undesired olefin metathesis side reactions of the G2 catalyst during Ru-catalyzed hydrogenation. <ref type="bibr">80</ref> Pt-Re/SiO 2 -Catalyzed Hydrogenolysis Reactions. Cyclohexane solution hydrogenolysis reactions using the Pt-Re/ SiO 2 catalyst provided by the Dow Chemical Co. were performed in the 0.3 L Parr reactor stirred with an "X"-shaped magnetic Teflon stir bar (Chemglass Inc., Vineland, NJ). 1.0 g of macroporous Pt-Re/SiO 2 catalyst was first dried under vacuum in the reactor overnight at 100 &#176;C to remove organic impurities, as acetic acid was apparently used as a surface passivation agent possibly to reduce catalyst handling hazards. The reactor was then pressurized to 50 psig with Ar(g) and vented to ambient pressure for subsequent addition of 1.0 g polymer and 130 mL cyclohexane. Thus, the reaction is conducted with an approximately 1:1 w/w catalyst:polymer ratio. Semicrystalline polymers hPB-06, HDPE, and CEC, which are insoluble in cyclohexane at ambient temperature, were added as solids along with anhydrous cyclohexane (130 mL). hPB-33, hPB-60, hPB-61, hPB-86, PEP, and PCHE were dried overnight under vacuum at 22 &#176;C, dissolved in anhydrous and anaerobic cyclohexane, and added to the reactor with additional solvent to achieve a total solution volume of 130 mL in each reaction. The reactor was sparged with 80 psig Ar(g) for 30 min, then pressurized to a static pressure of 500 psig H 2 (g) and allowed to react for 17 h at 140, 170, or 200 &#176;C. Note that the reactor pressure decreases modestly over the course of the reaction (&#8804;30% decrease from the initial pressure of 500 psig), but never fell below 300 psig. After 17 h, the reactor was cooled in an ice water bath to rapidly decrease its temperature to ambient temperature over the course of &#8764;5 min.</p><p>Hydrogenolytic degradation products obtained from the amorphous polymers hPB-33, hPB-60, hPB-61, hPB-86, PEP, and PCHE remained cyclohexane-soluble and were recovered by pressure-driven membrane filtration as in the polymer hydrogenation reactions. The hydrogenolysis products were subsequently isolated by cyclohexane removal by freeze-drying under vacuum. Note that precipitation of the reaction products was explicitly avoided to minimize loss of low molecular weight products. Hydrogenolysis of semicrystalline HDPE, hPB-06, and CEC instead produced a mixture of semicrystalline and amorphous polymer fragments with limited solubility in cyclohexane. In these cases, the cyclohexane was first removed by freeze-drying, followed by dissolution and hot filtration of the catalyst/polymer mixture in mixed xylenes at 130 &#176;C to avoid crystallization-induced precipitation. The xylene solvent was removed by freeze-drying under vacuum at ambient temperature, and the resulting solid materials were further dried at either 100 &#176;C (for PCHE and CEC) or 50 &#176;C (for all other species) to remove any residual solvent.</p><p>For selected samples, up to four replicate hydrogenolysis batch reactions were performed to assess reaction reproducibility. Batch reactions using the same conditions were also performed, wherein aliquots (&#8776;20 mL) were taken at specified time intervals to monitor the time course of catalytic chain scission reactions. Details regarding the time sampling reactions are provided in Supporting Information, along with a photograph of the sampling apparatus (Figure <ref type="figure">S1</ref>).</p><p>Size Exclusion Chromatography (SEC). Initial model polyolefin and hydrogenolysis reaction product molecular weights were determined by high-temperature size-exclusion chromatography (HT-SEC) in 1,2,4-trichlorobenzene (TCB, stabilized with 0.02 wt % butylated hydroxytoluene) at 135 &#176;C with a flow rate of 1 mL/min. These analyses employed an Agilent PL-GPC 220 SEC system equipped with three PL-Gel Mixed B columns, a differential refractive index (RI) detector, and a Wyatt DAWN HELEOS II LS 18-angle detector for absolute molecular weight quantification. Samples were prepared with concentrations of 1-4 mg polymer/mL of filtered TCB. Light scattering analyses used the refractive index increment dn/dc values for the samples in TCB at 135 &#176;C or a 100% mass recovery assumption for samples where literature values of dn/dc were not available. Alternatively, a Mark-Houwink (M-H) corrected calibration curve based on tabulated values of the M-H constants K and a in TCB at 135 &#176;C was constructed from a conventional calibration curve derived from analyses of narrow dispersity polystyrene standards with M n = 500-123 000 g/mol.</p><p>Nuclear Magnetic Resonance (NMR) Spectroscopy.</p></div>
<div xmlns="http://www.tei-c.org/ns/1.0"><head>1</head><p>H NMR measurements were performed on a 500 MHz Bruker Avance III instrument using a 20 s pulse repetition delay with 16 transients in CDCl 3 for PS and PCHE, or in toluene-d 8 for all other polymers. Samples were prepared with a polymer concentration of &#8764;10 mg/mL and run at 27 &#176;C. In the case of polymers not soluble at 27 &#176;C, spectra were acquired at 90 &#176;C. <ref type="bibr">13</ref> C NMR spectra were acquired using an inverse-gated 1 H decoupled 30&#176;pulse sequence performed with either a 2 or 6 s relaxation delay with &gt;700 scans. For hPB samples, 5 wt % polymer solutions were prepared in 90/10 v/v 1,2-dichlorobenzene/1,1,2,2-tetrachloroethane-d 2 with 0.01-0.02 M chromium(III) acetylacetonate as a spin-lattice relaxation (T 1 ) agent that also minimizes impacts of the nuclear Overhauser Effects (NOEs). <ref type="bibr">82</ref> Samples were measured at either 130 or 27 &#176;C. For PCHE samples, 15 wt % polymer solutions were prepared in 90/10 v/v 1,2-dichlorobenzene/ 1,1,2,2-tetrachloroethane-d 2 with 0.03 M chromium(III) acetylacetonate and measured at 27 &#176;C.</p><p>Differential Scanning Calorimetry (DSC). DSC measurements were conducted with a TA Instruments Q1000 instrument. Samples were heated and cooled at 10 &#176;C/min, with crystallization temperature (T c ) values reported on first cooling, and glass transition and melting temperature (T g and T m ) values reported on second heating. Data analyses were performed using the TA Universal Analysis software.</p><p>Gas Chromatography/Mass Spectrometry. Gas chromatography/mass spectrometry (GC/MS) measurements were performed using a Hewlett-Packard HP-6890 Series GC/QTOF-MS with helium carrier gas. The system was equipped with an Agilent 19091S-133 separation column and an electron impact ionization system. 2 &#956;L of solution was manually injected into the instrument with a separation sequence involving a temperature ramp from 50 to 290 &#176;C at a rate of 12 &#176;C/min with no solvent delay.</p><p>A gas chromatography/flame ionization detection measurement was acquired using manual injection of 0.5 &#956;L of sample into an S/SL inlet at 250 &#176;C with a split ratio of 100. The sample was separated by an HP-PlotQ column, and the FID was set at 400 &#176;C with H 2 , air, and He makeup flows of 40 mL/min, 350 mL/min, and 20 mL/min, respectively. The temperature program involved holding at 40 &#176;C for 3 min, followed by ramping to 270 &#176;C at 20 &#176;C/min and holding at 270 &#176;C for 10 min.</p><p>Catalyst Characterization. Samples for scanning transmission electron microscopy (STEM) studies were prepared using a focused-ion beam (FIB) performed on an FEI Helios NanoLab G4 dual-beam system. The FIB was operated with a Ga-ion beam at 30 kV and an electron beam at 5 kV. An amorphous carbon (am-C) layer of &#8764;50 nm was deposited on the sample as a protective layer prior to FIB milling. Additional 1 &#956;m protective layers of am-C and Pt were deposited on the region of interest. Another set of specimens was prepared by grinding the Pt-Re/SiO 2 catalyst powder into fine particles using a mortar pestle and drop casting from deionized water on a standard Cu TEM grid with am-C support. The specimens were investigated with a ThermoFisher Scientific Talos F200X STEM equipped with a Super-X energy dispersive X-ray (EDX) spectrometer. The microscope was operated at 200 kV with a probe convergence angle of 10.4 mrad and current of 80 pA. Images were acquired with detector inner and outer collection of angles of 46 and 200 mrad, respectively. The STEM-EDX elemental maps were analyzed with ThermoFisher Scientific Velox software. Experimental details for nitrogen adsorption isotherms and X-ray photoelectron spectroscopy measurements are provided in the SI.</p></div>
<div xmlns="http://www.tei-c.org/ns/1.0"><head>&#9632; RESULTS AND ANALYSIS</head><p>Pt-Re/SiO 2 Catalyst Characterization. Despite the extensive use of Pt/SiO 2 and Pt-Re/SiO 2 catalysts for polymer hydrogenations over the past 25 years, minimal characterization is available for these two hydrogenation catalysts developed by the Dow Chemical Co., including no published images of the Pt-Re/SiO 2 catalyst. Bates et al. reported transmission electron microscopy (TEM) and scanning electron microscopy (SEM) images of the Pt/SiO 2 catalyst, which exhibited &#8764;10 nm Pt nanoparticles grafted on a macroporous SiO 2 substrate. <ref type="bibr">63</ref> In addition, Hucul and Hahn reported the development of the ultrawide-pore SiO 2 network that is used as the support in these two catalysts, wherein the average pore diameter is &#8764;0.4 &#956;m as calculated from mercury porosimetry, upon which Pt-Re (or Pt) nanoparticles are deposited. <ref type="bibr">58</ref> We obtained high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM) images to better understand the catalyst structure and the associated length scales relative to the size of individual polymer chains in solution. The catalyst was first dried overnight at 100 &#176;C under vacuum to mimic the procedure used for hydrogenolysis reactions, prior to preparation for imaging using a FIB lift-out technique. STEM images presented in Figure <ref type="figure">2A</ref>,B clearly show the presence of a macroporous SiO 2 network, with pores and cavities &gt;0.1 &#956;m in size, consistent with prior reports of an average pore diameter of &#8764;0.4 &#956;m based on mercury porosimetry measurements. <ref type="bibr">58,</ref><ref type="bibr">64</ref> At higher resolution (inset panel in Figure <ref type="figure">2B</ref>), Pt-Re nanoparticles are observed as bright spots with an average particle size of 3.3 nm, calculated from 2D line scans at five different locations in the micrograph (see Figure <ref type="figure">S2</ref>); note that the oxidation state of the metals cannot be discerned by this technique. While the nanoparticles are largely isolated, a fraction of the particles exist as agglomerates, suggesting a heterogeneous reaction environment.</p><p>Energy-dispersive X-ray (EDX) mapping using STEM was also performed on a separate sample prepared by pulverization and solvent drop-casting to avoid analysis errors stemming from additional Pt deposition during sample preparation. This analysis, shown in Figure <ref type="figure">2C</ref>, shows the presence of both Pt and Re, albeit the Re signal is fainter and appears more broadly distributed than Pt. The EDX maps with the addition of the Si signal are provided in Figure <ref type="figure">S3A</ref>, and fitting of the spectra in Figure <ref type="figure">S3B</ref> indicates the presence of well-dispersed Pt and Re on the macroporous SiO 2 network, at roughly a 2 to 1 ratio of [Pt]: [Re].</p><p>The catalyst surface area was determined using a nitrogen adsorption isotherm. The corresponding Brunauer-Emmett-Teller (BET) plot (Figure <ref type="figure">S4</ref>) produced a surface area of 61.0 &#177; 0.7 m 2 /g, consistent with a previous report on the Pt-Re/ SiO 2 catalyst. 64 X-ray photoelectron spectra (XPS) were obtained (Figure <ref type="figure">S5</ref>), from which the surface composition was estimated as given in Tables <ref type="table">S3</ref> and <ref type="table">S4</ref>. These results indicate surface atomic ratios of 10:1 for Pt:Re and 2.7:1 for O:Si. The platinum peak position corresponds to metallic Pt, <ref type="bibr">83</ref> while the rhenium peak position is consistent with Re 2 O 7 , showing an oxidized rhenium state in the ex situ XPS experiment prior to treatment with H 2 (g). <ref type="bibr">84,</ref><ref type="bibr">85</ref> Model Polyolefin Synthesis and Characterization. Organolithium-initiated anionic polymerization of butadiene proceeds by either 1,4-or 1,2-addition pathways, with variations in solvent and temperature determining the final polymer microstructure. <ref type="bibr">86</ref> Five hPBs with different microstructures were synthesized by anionic polymerization and subsequent hydrogenation with either Wilkinson's catalyst or Pd/CaCO 3 and characterized using NMR, SEC, and DSC. Molecular characteristics for these polymers are presented in Table <ref type="table">1</ref> (see Figure <ref type="figure">1</ref> for chemical structures), with additional details given Figures <ref type="figure">S6-S14</ref>, Tables S1-S2 and S5. In brief, all polymers exhibit relatively low dispersities (&#272; = M w /M n ) and unimodal molecular weight distributions, calculated using SEC with multiangle light scattering and known differential refractive index increment values (dn/dc) for hPB in 1,2,4trichlorobenzene (TCB). <ref type="bibr">87</ref> The ethyl branch contents in the hPBs were determined based on the fraction of 1,2-repeat units in the unsaturated precursors determined by 1 H NMR. A semicrystalline polymer melting peak at T m = 108 &#176;C is evident in the DSC trace for hPB-06 (Figure <ref type="figure">S13</ref>), while all other hPB specimens exhibit no measurable crystallinity and display glass transitions at reduced temperatures (T g &lt; -10 &#176;C, Figure <ref type="figure">S13</ref>).</p><p>Four other polymers were synthesized or acquired to better understand the impact of short-chain branches on polymer degradation: high-density polyethylene (HDPE), poly-(ethylene-alt-propylene) (PEP), poly(cyclohexylethylene) (PCHE), and CEC, a triblock copolymer from the Dow Chemical Co. containing 62 wt % PCHE end blocks and a hydrogenated (1,4-rich) polybutadiene midblock (CEC). <ref type="bibr">77</ref> Characterization data for these polymers are also provided in Table <ref type="table">1</ref> (and Figures S15-S26 and Table <ref type="table">S1</ref>), along with their chemical structures in Figure <ref type="figure">1</ref>. Temperature-Dependent Degradation of hPB-60. We first probed the impact of temperature (T) on hydrogenolysis catalyzed by Pt-Re/SiO 2 by performing degradation experiments with hPB-60 at T = 140, 170, and 200 &#176;C under 500 psig H 2 (g) for 17 h. To characterize the reaction products, SEC data plotted as intensity versus elution time (Figure <ref type="figure">S27</ref>) was recast as intensity versus number-average molecular weight (M n ) using the Mark-Houwink correlation for short-chain branched polyolefins in TCB at 135 &#176;C, <ref type="bibr">90</ref> as shown in Figure <ref type="figure">3</ref>. A detailed description of this conversion and molecular weight correction by ethyl branch content is provided in the SI. By virtue of the near constant value of the refractive index increment dn/dc for the hPB-60 starting material and its degradation products, the intensity as a proxy for the relative mass fractions of chains of a given molar mass. These results revealed that the hydrogenolysis outcome over Pt-Re/SiO 2 strongly depends on temperature, such that hPB-60 is largely unreacted at 140 &#176;C, yet it is degraded to a peak molar mass M peak &lt; 1000 g/mol at 200 &#176;C; note that the lowest M n fragments approach the exclusion limit of the SEC instrument in the latter case. At 170 &#176;C, an intermediate extent of degradation occurred. Consequently, we selected this condition to explore the role of polymer short-chain branching on the hydrogenolytic degradation kinetics. To further characterize the polymer degradation products, [-CH 2 -]:[-CH 3 ] peak ratios were calculated from quantitative 1 H NMR spectra and their T g values were extracted from DSC measurements for comparison to those obtained from the hPB-60 starting material (Table <ref type="table">S6</ref>). The relative methyl peak intensity increased upon degradation (Figure <ref type="figure">S28</ref>), consistent with a greater number of end groups as expected for lower M n products, and T g monotonically decreased (Figure <ref type="figure">S29</ref>) concordant with the expected dependence of T g on molecular weight. <ref type="bibr">91</ref> Polymer Hydrogenolysis Kinetics. Hydrogenolysis was carried out on HDPE, hPB-06, hPB-33, hPB-61, and hPB-86 at 170 &#176;C and for selected cases small aliquots (ca. &#8776; 20 mL) were sampled from the reactor at 1, 3, 5.75, and 7.75 h during each 17 h reaction. The resulting molecular weight distributions are presented in Figure <ref type="figure">4</ref>, where SEC results were converted from retention time to molecular weight from Figure <ref type="figure">S30</ref> or S32 as described above. Reaction replicates were performed to establish reproducibility of the results (Figures S31-S33 and Tables <ref type="table">S10-S12</ref>). These data demonstrate that all of the initial polymer chains, except hPB-86, are consumed during the course of the reactions and yield either M n &#8773; 1000 g/mol for HDPE, hPB-06 and hPB-33, or M n &#8773; 3000 g/mol for hPB-61 after 17 h. Notably, hPB-86 exhibits almost no change in molecular weight over the course of the hydrogenolysis reaction.</p><p>For reactions of hPB-06, hPB-33, and hPB-61 that results in appreciable chain degradation, early time reaction aliquots exhibit bimodal molar mass distributions comprising combinations of unreacted and narrow dispersity starting material and broad distributions of cleaved chains with M n,peak = 10 000-20 000 g/mol (see Figures <ref type="figure">S34</ref> and <ref type="figure">S35</ref> for a description of the deconvolution of the overlapping SEC peaks in these data). This observation crucially implies that adsorption of an unreacted polymer molecule to the heterogeneous catalyst surface is followed by multiple hydrogenolysis reactions prior to surface desorption, reminiscent of a hydrogenation mechanism previously reported by Rosedale and Bates. <ref type="bibr">92</ref> Subsequently, readsorption of the fragmented polymers followed by additional chain cleavage events drives the overall molar mass distribution to a monomodal form centered at the aforementioned lower M n values.</p><p>Here we note that rudimentary calculations of the time required for the polymer coils to diffuse into the catalyst pores from the well-mixed, relatively dilute, solution (seconds) is much shorter than the reaction time associated with the bimodal molecular weight distributions, ruling out mass transfer limitations as the cause of these early time reaction outcomes. (A host of additional factors will influence the overall reaction kinetics including catalyst pore confinement effects, solvent quality, and competition between the solvent and polymer for surface adsorption. Our experiments have been designed to minimize the consequences of these effects, e.g., use of an ultrawide-pore catalyst and a good nonpolar solvent, in determining the relative impact of chain branching in polyolefin hydrogenolysis). Overall, hPB-06 reacts more   <ref type="table">S6</ref> as calculated using a Mark-Houwink correlation (see Experimental Methods for details).</p><p>quickly than hPB-33, which reacts faster than hPB-61, suggesting that the rate of polymer hydrogenolysis depends on the degree of short chain branching in the polymeric starting material. Again, the degraded polymers exhibit expected changes in their 1 H NMR spectra (Figure <ref type="figure">S36</ref>), T g 's, and their melting temperatures T m (Figure <ref type="figure">S37</ref>) as summarized in Tables S7-S9.</p></div>
<div xmlns="http://www.tei-c.org/ns/1.0"><head>Hydrogenolysis of PEP and PCHE.</head><p>In order to assess the influence of side chain chemical structure and density along the all carbon backbone, hydrogenolysis reactions with Pt-Re/ SiO 2 were conducted on PEP and PCHE at 170 &#176;C for 17 h. Figures <ref type="figure">5A (</ref>and <ref type="figure">Table S11</ref>, Figure <ref type="figure">S38A</ref>) shows the results for PEP, which exhibits a pendant methyl group on every fourth carbon atom, along with a statistical distribution of 1.5 isopropyl groups per 100 backbone carbon atoms. On average, chain cleavage in PEP is less extensive than with hPB-61, which displays a similar density of side chains (22 ethyl branches per 100 backbone carbon atoms); &#10216;M n &#10217; PEP &#8773; 9000 g/mol and &#10216;M n &#10217; hPB-61 &#8773; 3000 g/mol. This disparate reactivity is also evident after 17 h at 200 &#176;C, where PEP and hPB-60 are reduced to 1,100 and 600 g/mol, respectively (Figure <ref type="figure">S39</ref>). Also, the results shown in Figure <ref type="figure">5A</ref> include noticeable fractions of unreacted polymer, not present in any of the hPB-61 reaction products. We return to this observation in the Discussion section. Additionally, the variability in the degraded product M n -values across 4 PEP replicates is greater than that obtained with 3 replicates from hPB-61 (Figure <ref type="figure">S31D</ref>); we do not believe this is statistically significant given the limited number of experiments.</p><p>PCHE reacts strikingly differently from all of the other polymers in hydrogenolysis over the Pt-Re/SiO 2 catalyst, with the exception of hPB-86. After 17 h at 170 &#176;C, there is virtually  no decrease in M n of the reaction product as illustrated in Figure <ref type="figure">5B</ref>. However, 1 H NMR revealed the appearance of an additional peak at &#948; 0.9 ppm following the exposure to the catalyst under these reaction conditions (Figure <ref type="figure">S40</ref>), indicative of the formation of additional methyl groups. Inverse-gated decoupled <ref type="bibr">13</ref> C NMR spectra obtained in 90/10 v/v 1,2-dichlorobenzene/1,1,2,2-tetrachloroethane-d 2 with chromium(III) acetylacetonate as a spin-lattice (T 1 ) relaxation agent <ref type="bibr">82</ref> demonstrate the emergence of two weak resonances at &#948; 13.4 and 15.0 ppm after the catalytic reaction, which we associate with the ring opening of &#8764;1 mol % of the pendant cyclohexyl moieties in PCHE (Figure <ref type="figure">S41</ref>). As a control experiment to assess the degree and extent of cyclohexane solvent ring-opening, pure cyclohexane was subjected to the hydrogenolysis conditions for 17 h at 170 &#176;C and static 500 psig H 2 (g) with the same catalyst loading. 1 H NMR and tandem gas chromatography-mass spectrometry analyses of the products revealed that 0.3 mol % and &lt;1 mol % of the solvent had ring-opened to n-hexane, respectively; the explicit experimental protocols and results are described in the Supporting Information, and the results are summarized in Figures <ref type="figure">S42</ref> and <ref type="figure">S43</ref>. Data from gas chromatography with flame-ionization detection (shown in Figure <ref type="figure">S43</ref>) corroborate these results.</p></div>
<div xmlns="http://www.tei-c.org/ns/1.0"><head>&#9632; DISCUSSION</head></div>
<div xmlns="http://www.tei-c.org/ns/1.0"><head>Role of Branching in Hydrogenolysis.</head><p>A compilation of the results of hydrogenolytic degradation experiments of eight differently branched model saturated hydrocarbon polymers at 170 &#176;C provides previously unrecognized insights into the impact of branch content and structure on polymer degradation. Polymer hydrogenolysis is apparently hindered by sterically bulky side chains, which likely prevent backbone adsorption to the catalyst for subsequent chain cleavage. On the other hand, less branched polymers can adsorb sequences of backbone methylene units onto active metal sites for C-C bond scission and addition of hydrogen to yield lower M n products. Figure <ref type="figure">6</ref> summarizes the time-dependent evolution of the reaction product molecular weight and distribution associated with all the hydrogenolysis reactions conducted at 170 &#176;C, as well as the fractional reduction in molecular weight after t = 17 h at 170 &#176;C, defined as</p><p>The final molecular weights M n (t = 17 h) observed during hydrogenolysis of hPB-x and HDPE (i.e., hPB-0), exhibit the trend HDPE &#8776; hPB-06 &lt; hPB-33 &lt; hPB-61 &#8810; hPB-86. We note that similar conclusions can be drawn by considering the fractional reduction in molecular weight, albeit hPB-33 degrades more similarly to hPB-61 when assessed by this metric because its initial molecular weight M n (t = 0 h) is lower.</p><p>For hPB-0 and hPB-06, polymers with short-chain branch contents of either 0 or 1.5 ethyl branches/100 backbone carbons respectively, the final M n and &#272; (see Table <ref type="table">S10</ref>) of degradation products are similar. Thus, low concentrations of ethyl branches minimally impact the overall reaction rate and outcomes. From these findings, we conclude that tertiary carbon centers do not apparently drive the hydrogenolysis reaction. Furthermore, isolated ethyl branches along the all-carbon backbone do not appreciably affect polymer adsorption on the heterogeneous catalyst surface.</p><p>Presumably, runs of backbone methylene units present the active metal sites on the Pt-Re particles with pairs of cleavable backbone carbon atoms as illustrated in Figure <ref type="figure">7</ref>. Random chain cleavage is anticipated to lead to a most-probable molecular weight distribution, i.e., &#272; &#8594; 2, after numerous hydrogenolysis reactions, but prior to the limit of complete reaction, i.e., formation of methane. <ref type="bibr">91</ref> Conversely, the higher ethyl branch contents in hPB-61 and hPB-86 sterically encumber the backbone and prevent its presentation to the catalyst surface for C-C bond activation (Figure <ref type="figure">7</ref>), slowing (hPB-61) and ultimately extinguishing (hPB-86) the rate of degradation. This behavior parallels previous work on H/D exchange with hPBs on Pt-Re/SiO 2 , where hydrogendeuterium exchange by C-H bond activation occurs preferentially on the pendant ethyl groups versus backbone methylene moieties, <ref type="bibr">70</ref> and occurs much more readily with   <ref type="bibr">93</ref> The presence of ethyl branches sterically hinders methylene adsorption and subsequent polymer backbone activation, effectively preventing chain cleavage.</p><p>HDPE relative to iPP. <ref type="bibr">71</ref> However, a direct comparison of the previously reported H/D exchange behavior with the hydrogenolysis reactivity disclosed here is not entirely possible, due to differences in catalyst pretreatment prior to reaction (vide infra).</p><p>To assess the microstructural differences between the last two polymers that exhibit limited reactivities, quantitative <ref type="bibr">13</ref> C NMR spectra were acquired for hPB-61 and hPB-86 (Figure <ref type="figure">S44</ref>). Integration of <ref type="bibr">13</ref> C resonances associated with monomer sequence triads corresponding to backbone methylene units at 30.0 ppm indicate that they are present in lower concentrations in hPB-86 relative to hPB-61, in accordance with the expected statistical distributions of 1,2-and 1,4-butadiene additions in the anionic polymerizations used to produce the model polyolefins for this study. <ref type="bibr">94</ref> The 13 C NMR spectrum of a hydrogenolyzed hPB-61 specimen indicates little change as compared to that of pristine hPB-61 (Figure <ref type="figure">S45</ref>); the &#8764;30 chain cleavage events per reacted macromolecule represent &lt;0.4% of the carbon atoms in the polymer chain, hence significant change in the spectrum after hydrogenolysis is not expected. However, the degraded hPB-61 exhibits five new, small peaks in the chemical shift range &#948; 14-25 ppm arising from additional methyl end groups in the degraded, low molecular weight products. The associated stoichiometry is consistent with the reduction in molecular weight deduced by SEC, confirming limited participation of tertiary carbon sites in the hydrogenolysis reactions.</p><p>Two discrete populations of polymer molecular weights are clearly observed during the initial stages of hPB-06, hPB-33, and hPB-61 hydrogenolysis reactions (Figures <ref type="figure">4</ref> and <ref type="figure">S30</ref>), which offers insight into the reaction mechanism leading to depletion of the starting material. Peak molecular weights between 10 and 20 kg/mol for the degraded chains at early times indicate the initial polymer molecules are cleaved 5 to 10 times on average following adsorption on the surface of the catalyst and desorption of the fragments. This observation concurs with findings by molecular dynamics simulations by Vlachos and coworkers demonstrating that long-chain PE outcompetes short-chain PE in adsorption on a Pt(111) surface. <ref type="bibr">36,</ref><ref type="bibr">95,</ref><ref type="bibr">96</ref> Apparently, multiple bond breaking occurs after this initial macromolecular adsorption step and prior to desorption. At a critical extent of reaction, the resulting low molecular weight polymer (M crit ) is displaced by unreacted high-M n chains (Figure <ref type="figure">S46</ref>), presumably driven by a combination of their lower entropy of dissolution (scales as 1/M n ) and the stronger multidentate binding of the longer chains to the catalyst. <ref type="bibr">91</ref> Once the initial (unreacted) polymer is consumed, further chain cleavage events continually drive down the reaction product M n . We further postulate that sections of the polymer are dynamically presented to the catalyst surface while bound through dehydrogenation and rehydrogenation events, as documented by Zhao et al. using magic-angle spinning NMR during an in situ hydrogenolysis of PE over a Pt/SiO 2 catalyst with a mesoporous silica shell (albeit in the absence of any solvent). <ref type="bibr">97</ref> Such a dynamic macromolecule-catalyst binding mechanism is also consistent with H/D exchange in PE catalyzed by this Pt-Re/SiO 2 documented in earlier publications. <ref type="bibr">71,</ref><ref type="bibr">72</ref> Interestingly, PEP hydrogenolysis is somewhat slower than degradation of hPB-61 in spite of comparable short-chain branch densities: 25 methyl and 22 ethyl branches per 100 backbone carbons, respectively. This difference is exemplified by the presence of 3-24 wt % unreacted PEP chains across four replicates in 17 h reactions, as compared to 13 wt % unreacted hPB-61 chains after only 5.75 h of reaction (Figure <ref type="figure">S47</ref>). Based on <ref type="bibr">13</ref> C NMR characterization of hPB-61, tertiary carbon atoms are separated by runs of either 1 or 5 -CH 2groups (with a small fraction of 9 or more methylenes in a row) along the backbone due to the statistical distribution of 1,4 and 1,2 repeat units in the polydiene precursor PB-61. On the other hand, PEP is characterized by precisely 3 methylene groups between tertiary carbons bearing a methyl side group (neglecting the minor fraction of isopropyl side chains arising from 5% 3,4-addition of isoprene in the otherwise 1,4polyisoprene precursor). We assert that the occurrence of longer sequences of methylene groups in hPB-61 affords greater opportunities for local coordination of -CH 2 -CH 2units with active sites on the catalyst resulting in a faster overall reaction rate, even though PEP has essentially the same number of backbone methylene groups per 100 backbone carbon atoms. Squeezing out virtually all sequences of methylene groups as in hPB-86 nearly extinguishes hydrogenolysis at 170 &#176;C over 17 h, consistent with this hypothesis.</p><p>The resistance to hydrogenolysis exhibited by PCHE agrees well with the above trends and mechanistic deductions derived from the different reactivities of the hPBs and PEP materials. Along with a complete absence of methylene sequences, synpentane interactions between the bulky cyclohexyl rings on every other backbone carbon inhibit the formation of the chain extended trans backbone conformations implicated in PE hydrogenolysis (Figure <ref type="figure">7</ref>). <ref type="bibr">36,</ref><ref type="bibr">98</ref> The stability of this polymer under the conditions used to saturate PS is consistent with the original motivation behind development of the Pt-Re/SiO 2 catalyst for commercial use. Moreover, as we have shown, the pendant cyclohexyl groups are resistant to ring opening, which would detract from the useful physical properties of the resulting PCHE plastic (e.g., elevated T g ). We note that the commercial hydrogenation of PS employs much lower catalyst loadings than used in this work, which renders the possible hydrogenolysis of the cyclohexane group even less significant.</p><p>Overall, our experiments and interpretations shed fresh light on the reaction of polyolefins on heterogeneous catalysts. Figure <ref type="figure">8</ref> summarizes a striking correlation between the number of methylene groups between tertiary carbon atoms in the backbone and the final molecular weight obtained after a 17 h hydrogenolysis reaction over the Pt-Re/SiO 2 catalyst at 170 &#176;C. Although we have restricted our attention to dilute and their products from catalytic hydrogenolysis for 17 h over the Pt-Re/SiO 2 catalyst at 170 &#176;C (f illed circles) as functions of the average number of methylene groups between tertiary backbone carbons (summarized in Table <ref type="table">S13</ref>). Error bars denote the observed data range.</p><p>polymer solutions in cyclohexane, we believe the conclusions of this work regarding the role of branching are likely relevant to upcycling hydrocarbon polymers in the melt state. <ref type="bibr">8,</ref><ref type="bibr">9</ref> Future work will be directed at increasing the polyolefin concentration up to the undiluted melt state, which will require dramatically different reactor designs (e.g., twin screw extrusion) and experimental protocols. Moreover, a host of additional factors will impact the ability of undiluted polymers to be upcycled, including high viscosity associated with entanglements, which restricts pore diffusion, heat transfer limitations, and possibly restricted configurational states on the catalyst surface.</p><p>Hydrogenolysis of Commercial CEC. While studies have thus far focused on the hydrogenolysis of model LLDPE materials with specific branch contents, realistic single-use plastic waste streams are inevitably mixtures of polyolefins with variable structures and short chain branch contents. However, assessing the reactivities of mixtures of the hPB polymers to address this issue is hampered by the fact that the SEC analyses do not readily provide access to the chemical constitution of the chromatographically separated species. In an initial attempt to overcome this problem, we considered an internal competition hydrogenolysis experiment using a CEC triblock polymer (see Figure <ref type="figure">1</ref>), comprising an E block with 2.4 ethyl branches per 100 backbone C atoms (from 1 H NMR) akin to hPB-06 that is flanked by two PCHE end segments. We note that this CEC polymer is commercially produced by saturation of PS-PB-PS triblock using the Pt-Re/SiO 2 catalyst under conditions similar to those reported here, although at much lower catalyst loadings without catalyst pre-treatment.</p><p>Reaction of a 1 wt % cyclohexane solution of this CEC block polymer at 170 &#176;C under 500 psig H 2 (g) for 17 h using a 1:1 w/w polymer to catalyst loading resulted in the chain degradation documented by the SEC data presented in Figure <ref type="figure">9</ref>. The dispersity of the cleaved, lower M n product is considerably greater than that of the starting CEC (&#272; = 1.3 versus &#272; &#8773; 1.02, respectively) yet not as high as that encountered with the model polyolefins.</p><p>Understanding the reaction course for CEC on Pt-Re/SiO 2 requires accounting for the different behaviors of the C versus E blocks in contact with the bimetallic catalyst and in solution. The SEC data in Figure <ref type="figure">9</ref> provide illuminating clues as to the mechanism of triblock copolymer degradation. The molecular weight distribution after reaction (&#272; &#8773; 1.3) is consistent with the formation of CE diblocks by E segment cleavage, wherein the broader &#272; stems from convolution of narrow dispersity C chains (&#272; &#8773; 1.02) connected to a disperse array of E chains (approaching &#272; &#8773; 2 as discussed above) weighted according to the chain composition. We associate the low molecular weight tail on the SEC trace of the degraded material with the presence of a small number of E chains cleaved from the CE diblocks, which form in the initial stages of the hydrogenolysis reaction. Complete decoupling of the C blocks by E midblock cleavage accords with expectations based on the complete digestion of pristine HDPE, hPB-06, hPB-33, and hPB-61 after 17 h (Figure <ref type="figure">4</ref>). Note that if the E blocks were completely consumed, we would expect to see a sharp peak in the SEC trace due to the remaining narrow dispersity C chains, along with a broad peak associated with the E oligomers, neither of which are evident. As discussed below, catalyst pretreatment increases its reactivity. Untreated Pt-Re/SiO 2 catalyst leads to a higher-M n CE diblock product with fewer E fragments (Figure <ref type="figure">S48</ref>). This reinforces our proposed mechanism of rapid chain adsorption and E midblock cleavage, followed by slower hydrogenolysis of the E end blocks, mirroring the behavior documented with the homopolymers. This discrimination of reaction rates between the two stages of CEC degradation is likely amplified by the different propensities for C and E adsorption onto the metal catalyst, and the reduction in molecular weight after the initial bond cleavage that enhances dissolution of the CE diblock copolymer in cyclohexane versus binding to the catalyst surface.</p><p>Catalyst Pretreatment. As outlined in the Experimental section, all catalytic reactions were conducted with Pt-Re/ SiO 2 , pretreated by vacuum drying overnight at 100 &#176;C. This conditioning step was motivated in part by a vinegar-like smell associated with the bulk catalyst. Dried and undried catalyst samples (100 mg) were stirred in THF-d 8 for &#8764;2 h and the resulting solutions were analyzed by 1 H NMR. Figure <ref type="figure">S49</ref> reveals the presence of acetic acid, <ref type="bibr">99</ref> which was largely removed by the vacuum treatment, evidenced by a &#8764;90% decrease in the peak intensity for acetic acid relative to the residual protiated solvent peak in THF-d 8 at &#948; 3.6 ppm. Hydrogenolysis reactions conducted with both forms of the catalyst revealed considerable differences in time-dependent hydrogenolysis outcomes as summarized in Figures <ref type="figure">S50</ref> and <ref type="figure">S51</ref>. Reaction of hPB-60 for 17 h over the untreated catalyst resulted in significant variability in the distribution of product molecular weights between repeat experiments, including the presence of a significant amount of unreacted polymer (16 and 37 wt % in two replicate reactions). Conversely, when subjected to an identical procedure, but using the vacuum treated catalyst, hPB-60 was degraded to roughly a 5-fold lower average molar mass (Figure <ref type="figure">S50</ref>). Moreover, as noted previously, the extent of hydrogenolysis of CEC was enhanced by pretreating the catalyst as illustrated in Figure <ref type="figure">S48</ref>.</p><p>These results aid in resolving puzzling differences in the hydrogenolysis results reported by Ertem et al. versus what we have presented here. <ref type="bibr">75</ref> That earlier study, also associated with deuterium exchange on various saturated hydrocarbon polymers, found limited degradation of HDPE and PEP under the same conditions reported here, albeit in isooctane instead of cyclohexane, in sharp contrast with the results shown in Figures <ref type="figure">4A</ref> and <ref type="figure">5A</ref>, respectively. However, we now believe that this can be attributed to the use of the untreated and thus less reactive Pt-Re/SiO 2 catalyst. Another curious finding in the earlier work was extensive degradation of a linear low-density PE (Exxon Exceed 1018 LLDPE), which appeared to implicate side chain branching in the mechanism of hydrogenolysis, contrary to what we have concluded presently. Exceed 1018 is a copolymer of ethylene and 1-hexene with about 1.3 butyl side chains per 100 backbone carbon atoms as opposed to 1.5 ethyl side chains per 100 backbone carbons in hPB-06. This outcome seems inconsistent with the less reactive untreated catalyst and may reflect a genuine effect associated with butyl branch adsorption onto or C-H activation by the metal surface followed by tertiary carbon bond cleavage possibly through a transient five-membered metallacyclic intermediate. Such reactions could lead to linear &#10216;M n &#10217; &#8776; 10 3 g/mol PE without involving the bond cleavage mechanism associated with the polyolefins presented in this work. Resolving this issue will require additional experiments with the vacuum-treated catalyst.</p></div>
<div xmlns="http://www.tei-c.org/ns/1.0"><head>&#9632; CONCLUSION</head><p>Hydrogenolysis of polyolefins dissolved in cyclohexane sensitively depends on the degree of short-chain branching. Using a small library of model polyethylenes with M n &#8764; 10 5 g/ mol bearing ethyl (hPB), methyl (PEP), cyclohexyl (PCHE), or no (HDPE) side chains produced by anionic polymerization or ring-opening metathesis polymerization and tandem hydrogenation, a correlation between branch content and temperature-dependent susceptibility to hydrogenolysis over a macroporous Pt-Re/SiO 2 catalyst was discovered. Using hPB-60 with 21 ethyl branches per 100 backbone carbons, we found that the hydrogenolysis reaction strongly depends on temperature, with almost no degradation at 140 &#176;C, a modest extent of chain scission at 170 &#176;C, and rapid hydrogenolysis at 200 &#176;C, over a 17 h period. Time-dependent experiments at 170 &#176;C revealed multiple chain cleavage events following initial adsorption onto the catalyst for polymers with 1.5-22 branches per 100 backbone carbon atoms. Little reaction occurred with 38 ethyl side chains per 100 backbone atoms, and there was no observed change in the molecular weight with PCHE. A commercial triblock copolymer CEC, containing hydrogenated polystyrene (C) end blocks coupled through a saturated polybutadiene (E, with 2.2 ethyl branches per 100 backbone atoms), exhibited partial cleavage of the E blocks and no degradation of the C blocks reflecting mixed susceptibility of this hybrid chain microstructure. These findings will aid in developing plastic waste upcycling strategies for converting saturated hydrocarbon polymers and their mixtures into commercially useful low molecular weight products.</p><p>&#9632; ASSOCIATED CONTENT * s&#305; Supporting Information</p></div><note xmlns="http://www.tei-c.org/ns/1.0" place="foot" xml:id="foot_0"><p>https://doi.org/10.1021/acscatal.5c04494 ACS Catal. 2025, 15, 16924-16937</p></note>
			<note xmlns="http://www.tei-c.org/ns/1.0" place="foot" xml:id="foot_1"><p>The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acscatal.5c04494.Reactor setup, reagent amounts for synthesis, HAADF-STEM line scans and EDX fitting, BET and XPS experimental details and results, NMR spectra of precursor polyolefins and select degraded products, SEC calculation details and chromatograms of precursor polyolefins and select degraded products, DSC traces of precursor polyolefins and select degraded products, raw SEC chromatograms of all degraded products, SEC molecular weight distributions and raw chromatograms of hydrogenolysis replicates, tables summarizing M n , &#272;, NMR peak intensities, and DSC thermal transitions for degradations, GC-MS and GC-FID data on ring-opened cyclohexane,<ref type="bibr">13</ref> C NMR spectra of hPBs, NMR spectra of organic impurity in Pt-Re/SiO 2 catalyst, SEC chromatograms for untreated catalyst reactions, calculation of average branch spacing, experimental details and additional characterization data (PDF)</p></note>
		</body>
		</text>
</TEI>
