Back to search

University of Leeds

Novel toolkit for post-production quality assurance of solid recovered fuel (SRF): a focus on sub-sampling uncertainty and composition identification

Abstract

dc:description.abstract

The inherent heterogeneity of waste-derived products is a key limiting factor for sustainable resources recovery and wider circular economy – also for solid recovered fuels (SRF). Confidence in SRF trading necessitates the manufacture of quality-assured SRF. Herein, a novel and adaptable toolkit for SRF characterisation was established through statistically designed experiments (DoE), applicable for quality assurance (QA) to both SRF producers (post-production) and end-users (pre-recovery). A novel machine learning (ML) SRF composition model was devised, able to identify prevalent polymers (cellulose, xylan, lignin, HDPE, PP, PET, and PVC), by implementing thermogravimetric (TG-DTG) and spectroscopic (FTIR) techniques (Objective 4: O4). Following an optimal mixture design, the applicability of the ML model was verified using synthetic mixtures that resemble SRF (cross-validation), waste mixtures with known composition featuring impurities (hold-out validation), and commercially produced SRF samples. The uncertainty arising from laboratory sub-sampling, expressed as relative standard deviation (RSD), was quantified for key quality properties: moisture (RSD: 2.6%), ash (RSD: 6.1%), calorific (RSD: 4.2%), and chlorine (RSD: 18.2%) content, and thermal stability via TG-DTG (RSD: 32.2%) (O3). Uncertainty was quantified through DoE (balanced nested-ANOVA) under a relatively unbiased, yet affordable and practicable sub-sampling plan, benefiting from state-of-the-art equipment, theoretical calculations according to theory of sampling (ToS), and implementation of best sub-sampling practices (O2). This work (O2&3) led to an integrated and practical characterisation scheme, ensuring the validity of any analytical determination based tool, such as the new ML model. A systematic review explored the variability of chlorine in SRF (key QA property) during SRF pre- and co-processing in cement kilns (main end-user) (O1), informing a tolerable level of Cl variability. This research offers the means to obtain a comprehensive set of information on SRF quality with relatively high confidence and low effort; also bearing far reaching wider implications for solid waste characterisation/ waste-derived products QA.

Degree

thesis:*
Name dc:type.qualificationname
Ph.D
Level dc:type.qualificationlevel
doctoral
Grantor dc:publisher.institution
University of Leeds
Year dc:date.issued
2020

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Gerasimidou, Spyridoula
Advisors dc:contributor.advisor
  • Velis, Costas
  • Paul, Williams
ORCID person:identifier.orcid

Rights

Language dc:language.dcterms_iso639_2
eng

Identifiers

dc:identifier.*
Dc Identifier Grantnumber
Full-time University Research Scholarship 2016/17

Chain of custody

source
Harvested from
White Rose University Consortium
Base URL
etheses.whiterose.ac.uk/cgi/oai2
Last updated
2026-11-18
Source record
OAI-PMH GetRecord
related terms
citation

Gerasimidou, Spyridoula. Novel toolkit for post-production quality assurance of solid recovered fuel (SRF): a focus on sub-sampling uncertainty and composition identification. doctoral thesis, University of Leeds, 2020. https://etheses.whiterose.ac.uk/id/eprint/28413/3/Gerasimidou_SPY_CivilEngineering_PhD_Part_Supporting_Information_SI3.xlsx