{"id":{"repo_id":"msu","oai_identifier":"oai:d.lib.msu.edu:etd_53006"},"canonical_url":"https://search.ndltd.org/etd/msu/oai:d.lib.msu.edu:etd_53006","repository":{"repo_id":"msu","name":"Michigan State University","base_url":"https://d.lib.msu.edu/oai"},"display":{"title":"CLOSING THE PETROCHEMICAL GAP : HOW AI-DRIVEN MODIFICATIONS TO EXISTING CHEMICAL RECYCLING PATHWAYS CAN ENABLE A CIRCULAR PLASTICS ECONOMY","abstract":"Thesis (Ph.D.)--Michigan State University. Chemical Engineering - Doctor of Philosophy, 2026","abstract_html":"Thesis (Ph.D.)--Michigan State University. Chemical Engineering - Doctor of Philosophy, 2026","abstract_has_math":false,"creators":["Behrendt, Michael S."],"institution":null,"degree_name":null,"degree_level":null,"degree_discipline":null,"degree_department":null,"school":null,"contributors":["Dorgan, John R","Calabrese Barton, Scott","Hegg, Eric L","Ferrier, Robert C"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2026,"date_issued":"2026","date_published":"2026","updated_at":"2026-11-18T18:59:34Z","subjects":["Chemical engineering","Macroecology"],"languages":["eng"],"rights":["In Copyright"],"rights_urls":[],"identifier_entries":[{"key":"dc:identifier","label":"Identifier","values":["etd:53006","local:Behrendt_grad.msu_0128D_22009"],"render_values":[{"text":"etd:53006","href":null,"code":true},{"text":"local:Behrendt_grad.msu_0128D_22009","href":null,"code":true}]}],"person_identifier_entries":[]},"links":{"outbound_url":"https://doi.org/doi:10.25335/9b5g-vk27","outbound_label":"DOI","outbound_source":"dc:identifier","record_url":"https://doi.org/doi:10.25335/9b5g-vk27","record_label":"Repository record","record_kind":"doi","record_source":"dc:identifier","download_url":null,"download_label":null,"download_source":null,"access_status":null,"embargo_until":null,"access_note":null},"source_record":{"url":"https://d.lib.msu.edu/oai?verb=GetRecord&metadataPrefix=oai_dc&identifier=oai%3Ad.lib.msu.edu%3Aetd_53006","prefix":"oai_dc"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Dorgan, John R","Calabrese Barton, Scott","Hegg, Eric L","Ferrier, Robert C"]},{"key":"dc:creator","label":"Author","values":["Behrendt, Michael S."]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2026"]},{"key":"dc:relation","label":"Dc Relation","values":["Electronic Theses & Dissertations"]},{"key":"dc:type","label":"Dc Type","values":["text","Electronic dissertations"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Chemical engineering","Macroecology"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["eng"]},{"key":"dc:rights","label":"Dc Rights","values":["In Copyright"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["etd:53006","local:Behrendt_grad.msu_0128D_22009","https://doi.org/doi:10.25335/9b5g-vk27"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["Thesis (Ph.D.)--Michigan State University. Chemical Engineering - Doctor of Philosophy, 2026","The global plastics economy remains predominantly linear because current recycling path-ways are often unable to match the cost and product quality of virgin petrochemical produc- tion. This thesis addresses that gap through three linked studies: (i) process modifications for chemical recycling of polyethylene terephthalate (PET), (ii) low-temperature oxidative depoly- merization of low-density polyethylene (LDPE), and (iii) machine-learning-enabled sorting with contaminant-aware classification. Together, these studies target the major technical bottle- necks of circular plastics systems: feedstock quality, depolymerization selectivity, and sorting accuracy under heterogeneous real-world conditions. For PET, this work demonstrates that a novel base-mediated hydrolysis in ethylene gly- col can provide high terephthalic acid (TPA) recovery under relatively mild conditions. At atmospheric pressure and moderate temperatures, TPA yields greater than 95% were ob- tained. At low temperatures, depolymerization in pure ethylene glycol substantially outper- formed water (65% vs 20% carbon yield to TPA after 168 h). Kinetic analysis was consis- tent with a surface-limited shrinking-core framework, giving an apparent activation energy of Ea = (66.6 ± 0.3) kJ mol−1 and a pre-exponential factor of A = (1.688 ± 0.161) × 108 h−1. Relative to conventional caustic hydrolysis and glycolysis benchmarks, preliminary technoe- conomic comparison indicates reduced reagent cost burden and downstream-separation ad- vantages, supporting the viability of this pathway as a lower-cost route to monomer-grade PET feedstock. For LDPE, a novel aqueous oxidation strategy based on cyclic regeneration of perman- ganate by ozone was developed and experimentally validated at atmospheric pressure and 30 ◦C. The system avoids organic solvents and relies on low-cost oxidant supply from oxygen- derived ozone. Multi-technique characterization (HPLC, ATR-IR, 13C NMR, DSC, and titration) showed progressive oxidation initiated at branch-rich amorphous domains, reaching approxi- mately 10% carbon conversion to soluble carboxylic acids after 72 h. ATR-IR carbonyl signals plateaued after about 48 h, while titration and NMR indicated continued bulk evolution; the final carbon-to-acid ratio in solids approached ∼30, consistent with branch-spaced scission behavior. Reactor engineering studies identified key practical constraints (gas-liquid-solid con- tacting, foaming, particulate transport, and sealing durability) and informed redesign choices that improved mixing and operability. The combined evidence supports a branch-initiated au- tooxidation/scission framework and demonstrates a viable low-temperature route for polyolefin valorization. For sorting, this thesis introduces a contextual paradigm in which contaminant signatures are treated as informative features rather than noise. Using ATR-IR spectra and logistic- regression-based multi-output classification, the model simultaneously predicts polymer type and contaminant classes from 788 spectra spanning four polymers and six contaminant classes. Polymer identification reached 91.14% test accuracy (cross-validation: 92.54 ± 3.56%), while contaminant detection achieved 0.63% Hamming loss on the test set (1.03% in cross-validation) and 98.73% contaminant exact-match accuracy. In a combined seven-output prediction task (polymer + six contaminants), the model achieved an overall exact-match ratio of 89.87%. These results support the hypothesis that contamination-aware sensing can improve decision quality upstream of washing, enabling better routing, reduced processing burden, and higher recovered-material value. Overall, this thesis demonstrates that narrowing the petrochemical gap requires coordi- nated innovation across sorting and chemistry: higher-information feedstock characterization, selective low-severity depolymerization, and process designs aligned with economic realities. The combined framework provides actionable pathways toward scalable, circular recycling of condensation polymers and polyolefins.","Description based on online resource. Title from PDF t.p. (Michigan State University Fedora Repository, viewed ).","Includes bibliographical references."]},{"key":"dc:format","label":"Dc Format","values":["application/pdf","1 online resource (341 p.) : ill."]},{"key":"dc:title","label":"Title","values":["CLOSING THE PETROCHEMICAL GAP : HOW AI-DRIVEN MODIFICATIONS TO EXISTING CHEMICAL RECYCLING PATHWAYS CAN ENABLE A CIRCULAR PLASTICS ECONOMY"]}]}],"canonical_facts":{"dc:contributor":["Dorgan, John R","Calabrese Barton, Scott","Hegg, Eric L","Ferrier, Robert C"],"dc:creator":["Behrendt, Michael S."],"dc:date":["2026"],"dc:description":["Thesis (Ph.D.)--Michigan State University. Chemical Engineering - Doctor of Philosophy, 2026","The global plastics economy remains predominantly linear because current recycling path-ways are often unable to match the cost and product quality of virgin petrochemical produc- tion. This thesis addresses that gap through three linked studies: (i) process modifications for chemical recycling of polyethylene terephthalate (PET), (ii) low-temperature oxidative depoly- merization of low-density polyethylene (LDPE), and (iii) machine-learning-enabled sorting with contaminant-aware classification. Together, these studies target the major technical bottle- necks of circular plastics systems: feedstock quality, depolymerization selectivity, and sorting accuracy under heterogeneous real-world conditions. For PET, this work demonstrates that a novel base-mediated hydrolysis in ethylene gly- col can provide high terephthalic acid (TPA) recovery under relatively mild conditions. At atmospheric pressure and moderate temperatures, TPA yields greater than 95% were ob- tained. At low temperatures, depolymerization in pure ethylene glycol substantially outper- formed water (65% vs 20% carbon yield to TPA after 168 h). Kinetic analysis was consis- tent with a surface-limited shrinking-core framework, giving an apparent activation energy of Ea = (66.6 ± 0.3) kJ mol−1 and a pre-exponential factor of A = (1.688 ± 0.161) × 108 h−1. Relative to conventional caustic hydrolysis and glycolysis benchmarks, preliminary technoe- conomic comparison indicates reduced reagent cost burden and downstream-separation ad- vantages, supporting the viability of this pathway as a lower-cost route to monomer-grade PET feedstock. For LDPE, a novel aqueous oxidation strategy based on cyclic regeneration of perman- ganate by ozone was developed and experimentally validated at atmospheric pressure and 30 ◦C. The system avoids organic solvents and relies on low-cost oxidant supply from oxygen- derived ozone. Multi-technique characterization (HPLC, ATR-IR, 13C NMR, DSC, and titration) showed progressive oxidation initiated at branch-rich amorphous domains, reaching approxi- mately 10% carbon conversion to soluble carboxylic acids after 72 h. ATR-IR carbonyl signals plateaued after about 48 h, while titration and NMR indicated continued bulk evolution; the final carbon-to-acid ratio in solids approached ∼30, consistent with branch-spaced scission behavior. Reactor engineering studies identified key practical constraints (gas-liquid-solid con- tacting, foaming, particulate transport, and sealing durability) and informed redesign choices that improved mixing and operability. The combined evidence supports a branch-initiated au- tooxidation/scission framework and demonstrates a viable low-temperature route for polyolefin valorization. For sorting, this thesis introduces a contextual paradigm in which contaminant signatures are treated as informative features rather than noise. Using ATR-IR spectra and logistic- regression-based multi-output classification, the model simultaneously predicts polymer type and contaminant classes from 788 spectra spanning four polymers and six contaminant classes. Polymer identification reached 91.14% test accuracy (cross-validation: 92.54 ± 3.56%), while contaminant detection achieved 0.63% Hamming loss on the test set (1.03% in cross-validation) and 98.73% contaminant exact-match accuracy. In a combined seven-output prediction task (polymer + six contaminants), the model achieved an overall exact-match ratio of 89.87%. These results support the hypothesis that contamination-aware sensing can improve decision quality upstream of washing, enabling better routing, reduced processing burden, and higher recovered-material value. Overall, this thesis demonstrates that narrowing the petrochemical gap requires coordi- nated innovation across sorting and chemistry: higher-information feedstock characterization, selective low-severity depolymerization, and process designs aligned with economic realities. The combined framework provides actionable pathways toward scalable, circular recycling of condensation polymers and polyolefins.","Description based on online resource. Title from PDF t.p. (Michigan State University Fedora Repository, viewed ).","Includes bibliographical references."],"dc:format":["application/pdf","1 online resource (341 p.) : ill."],"dc:identifier":["etd:53006","local:Behrendt_grad.msu_0128D_22009","https://doi.org/doi:10.25335/9b5g-vk27"],"dc:language":["eng"],"dc:relation":["Electronic Theses & Dissertations"],"dc:rights":["In Copyright"],"dc:subject":["Chemical engineering","Macroecology"],"dc:title":["CLOSING THE PETROCHEMICAL GAP : HOW AI-DRIVEN MODIFICATIONS TO EXISTING CHEMICAL RECYCLING PATHWAYS CAN ENABLE A CIRCULAR PLASTICS ECONOMY"],"dc:type":["text","Electronic dissertations"]},"updated_at":"2026-11-18T18:59:34Z"}