Pyrolysis of mixed plastics and paper to

produce fuels and other chemicals

Nanta Sophonrat

Doctoral Dissertation

2019

KTH Royal Institute of Technology

School of Industrial Engineering and Management Department of Materials Science and Engineering

Unit of Processes

SE-100 44 Stockholm, Sweden

__________________________________________________________________

Akademisk avhandling som med tillstånd av Kungliga Tekniska Högskolan i

Stockholm framlägges för offentlig granskning för avläggande av teknologie

doktorsexamen, Torsdag den 9 Maj 2019, kl. 10:00 i Kollegiesalen, Brinellvägen 8,

Kungliga Tekniska Högskolan, Stockholm.

ISBN 978-91-7873-148-0

Nanta Sophonrat. Pyrolysis of mixed plastics and paper to produce fuels and other chemicals

KTH Royal Institute of Technology

School of Industrial Engineering and Management Department of Materials Science and Engineering

Unit of Processes SE-100 44 Stockholm, Sweden

ISBN 978-91-7873-148-0 TRITA -ITM-AVL 2019:10

Copyright © Nanta Sophonrat

“We should try to leave the world a better place than when we entered it.

As individuals, we can make a difference, whether it is to probe the secrets of Nature,

to clean up the environment and work for peace and social justice,

or to nurture the inquisitive, vibrant spirit of the young by being a mentor and a guide.”

– Michio Kaku

i

Abstract

As the world population and economy grow, higher consumption results in higher

waste packaging, plastics and paper residues. Pyrolysis offers a way to recover fuels

and other chemicals from this waste fraction. By applying heat to these materials in

the absence of oxygen, pyrolysis process can convert these feedstocks into more

valuable products in the forms of gas, liquid and char.

One important issue in the pyrolysis process which requires an investigation is the

interactions between the feedstocks which consist of cellulose as the main

component of paper and different types of plastics. Regarding this topic, 3

subtopics were investigated which are: the effect of mixing methods on the co-

pyrolysis products, the interactions between the plastics and cellulose, and the

formation of H, OH and water during cellulose pyrolysis. All these experimental

investigations were based on microscale pyrolysis experiments using Py-GC/MS

technique.

In the first work, polyethylene and cellulose were mixed by melting and by putting

side-by-side. It was found that some interactions occurred during co-pyrolysis of

these materials which slightly altered the yields of some anhydrosugars, aldehydes

and ketones when the two feedstocks were mixed together by melting. Nevertheless,

the main pyrolysis products from each feedstock were not affected.

In the second study, the investigation continues on the interactions between

different types of plastics (PE, PP, PS, PET) and cellulose. By using Py-

GC×GC/MS, a good separation of the mixed pyrolysis products could be achieved,

thus assisting the analysis. It was found that although the main pyrolysis produc ts

from each feedstock were not affected by the co-pyrolysis, small interactions

occurred such that the interactions between different plastics were more

pronounced than the interactions between plastics and cellulose. Nevertheless,

some hydrogen transfer reactions occurred when PS was co-pyrolyzed with

cellulose. However, the source of hydrogen was not clear.

Therefore, the investigation on the formation of H and OH radicals during

cellulose pyrolysis was performed. This work combined first -principle calculations

with experimental investigations. The author of the thesis was responsible for the

experimental part. It was found from the first-principle calculations that it is

energetically more favorable for the generation of a pair of H and OH radicals with

subsequence formation of water than to generate a single radical because the

formation of a double bond on the resulting cellulose helps stabilize the structure.

ii

With Gibbs free energy calculations, it was predicted that the water would be

released at 280 °C. This agree well with the experimental findings from multistep

pyrolysis of cellulose in Py-GC/MS which showed that water was generated at two

different temperature ranges with the first peak around 280 °C.

As the interactions between the feedstocks during co-pyrolysis do not much

improve the liquid products’ properties, and the nature of the products produced

from plastics and paper pyrolysis are significantly different; it might be more

beneficial to separate the pyrolysis products from the two feedstocks. Moreover,

the hydrocarbons produced from plastics pyrolysis and the oxygenated products

from paper pyrolysis require different upgradation methods. Stepwise pyrolysis was

then proposed to produce and collect these two products separately. With

simulated feedstock mixtures (PE, PS, cellulose) and real waste fractions which are

paper rejects, it was successfully demonstrated that the stepwise pyrolysis with a

temperature of the first step of 300-350 °C and a temperature of the second step

of 500 °C could be used to produce two products streams as previously described.

However, an optimization of the process and further investigations on product

properties and upgradation are still required.

As a continuation on the investigation of the stepwise pyrolysis, an upgradation of

the products from the first pyrolysis step was studied. When PVC plastic is present

in the feedstock, dehydrochlorination of PVC occurs in the temperature range of

the first pyrolysis step together with the pyrolysis of cellulose. Calcium oxide (CaO)

was then tested for the simultaneous adsorption of HCl and reforming of cellulose

pyrolysis products. The experiments were performed in a two-stage reactor system

which was a pyrolysis reactor connected in series to a catalytic reactor con taining

CaO. It was found that the catalytic temperature should be between 300-350 °C

because the desorption of HCl occurred when the temperature was higher than

400 °C. This was partly due to a reaction between water and CaCl2 which caused

the desorption of HCl.

From all the studies, stepwise pyrolysis has a great potential to produce fuels and

other chemicals from mixed plastics and paper. Further investigations are needed

to develop, evaluate and realize this promising process.

Keywords: Pyrolysis; Mixed plastics and biomass; Cellulose; Refuse derived fuels

(RDF); Paper rejects; Interactions; Stepwise pyrolysis; Calcium oxide (CaO)

iii

Sammanfattning

I takt med en växande världsbefolkning och global ekonomi leder den ökande

konsumptionen till växande avfallsströmmar av bland annat förpacknings-material,

plast och papper. Pyrolys erbjuder en möjlighet att använda dessa strömmar som

råvara för att framställa bränsle och kemikalier. Genom att värma upp dessa

material i frånvaro av syre kan pyrolysprocessen omvandla dessa avfallsströmmar

till högvärdiga produkter i form av gas, vätska och fast kolbaserad material.

En viktig faktor som behöver undersökas i pyrolysprocessen är interaktionen

mellan de blandade materialen som huvudsakligen består av cellulosa från papper

samt olika typer av plast. Inom detta område har vi utfört tre individuella

experimentella undersökningar: påverkan på pyrolysprodukterna utifrån hur

materialen är blandade, interaktionen mellan plaster och cellulosa, samt studerat

produktion av H, OH och vatten under pyrolys av cellulosa. Dessa undersökningar

baserades på pyrolysexperiment i mikroskala genom användning av Py-GC/MS

teknik.

I det första arbetet undersöktes skillnaden mellan sampyrolys av ihopsmält polyeten

och cellulosa jämfört med sampyrolys av polyeten och cellulosa separerade från

varandra. Resultaten visar på att interaktionen mellan materialen under pyrolys

skiljer sig åt mellan de två fallen med variation i utbytet av anhydrosocker, aldehyder

och ketoner. De huvudsakliga pyrolysprodukterna visade ingen signifikant

påverkan i de olika fallena av sampyrolys.

I den andra studien undersöktes interaktionen mellan olika slags plast (PE, PP, PS,

och PET) och cellulosa. Genom användning av Py-GCxGC/MS kunde en god

separation av pyrolysprodukterna genomföras för en djupdykande analys.

Resultaten visar att trots att de huvudsakliga pyrolysprodukterna från respektive

material inte påverkades av sampyrolysen så förekom markanta interaktioner

mellan de olika plastmaterialen jämfört med interaktionerna mellan plast och

cellulosa. Väteöverföringsreaktioner förekom vid sampyrolys av PS och cellulosa

utan att man lyckades identifiera en tydlig vätedonator.

Nästföljande arbete fokuserade på uppkomsten av H- och OH-radikaler under

pyrolys av cellulosa. I detta arbete kombinerades first-principle beräkningar

medexperimentella försök. Författaren till denna avhandling ansvarade för den

experimentella utvärderingen. First-principle beräkningarna visade att bildningen av

ett par H- och OH-radikaler med efterföljande bildning av vatten är energimässigt

fördelaktigt jämfört med att framställa en enstaka radikal eftersom bildandet av en

iv

dubbelbindning i cellulosastrukturen är stabiliserande. Genom beräkningar av

Gibbs fria energi uppskattades att vatten skulle bildas vid pyrolys vid 280 °C. Detta

stämde väl överens med experimentella resultat från flerstegspyrolys av cellulosa i

Py-GC/MS som visar att vatten produceras vid två olika temperaturområden varav

en kromatograftopp uppstod vid pyrolys vid 280 °C.

Eftersom interaktionen mellan de olika materialen under sampyrolys inte påverkar

produkternas kvalitet, samt att sammansättningen av pyrolysprodukter från pyrolys

av plast och papper signifikant skiljer sig åt materialen sinsemellan, kan det vara

fördelaktivt att separera pyrolysprodukterna från respektive material. Dessutom

krävs olika uppgraderingsmetoder för kolvätena från pyrolys av plast respektive de

oxygenerade produkterna från pyrolys av papper. Stegvis pyrolys föreslogs för att

kunna producera pyrolysprodukter från respektive material utifrån en blandning av

materielen. Genom modellerade blandningar av material (PE, PS, cellulosa) och

verkliga avfallsströmmar från pappersåtervinning lyckades man visa att stegvis

pyrolys först i temperatursområdet 300-350 °C följt av pyrolys vid 500 °C kan

användas för att producera två produktströmmar enligt ovanstående beskrivning.

Metoden kräver dock ytterligare processoptimering och undersökning av

produkternas egenskaper och uppgraderingsmöjligheter.

Som fortsättning på arbetet kring stegvis pyrolys studerades uppgradering av

produkter från det första pyrolyssteget. När PVC-plast används som råvara

uppkommer dehydroklorering av PVC i samma temperatursområde som pyrolys

av cellulosa förekommer. Kalciumoxid (CaO) undersöktes baserat på dess

egenskap att adsorbera HCl samt att kunna uppgradera produkter från pyrolys av

cellulosa. Experimenten utfördes i ett system av två seriekopplade reaktorer

bestående av en pyrolysör följt av en katalytisk reaktor innehållande CaO.

Resultaten från den experimentella undersökningen visade att temperaturen i den

katalytiska reaktorn bör vara inom intervallet 300-350 °C då desorption av HCl

förekommer vid temperaturer högre än 400 °C. Detta förklaras delvis genom en

reaktion mellan vatten och CaCl2 som orsakar desorptionen av HCl.

Baserat på utförda forskningsstudier visar stegvis pyrolys på mycket god potential

för att kunna framställa bränslen och andra kemikalier från blandning av plast och

papper. Ytterligare studier krävs för att kunna utveckla, utvärdera och förverkliga

denna lovande process.

Nyckelord: Pyrolys; Blandade plaster och biomassa; Cellulosa; RDF; Pappersavfall;

Interaktioner; Stegvis pyrolys; Kalciumoxid (CaO)

v

บทคดยอ

ในขณะโลกทมประชากรและการเตบโตทางเศรษฐกจมากขน การบรโภคทมากขนกสงผลใหปรมาณขยะบรรจภณฑพลาสตกและกระดาษเพมมากขน ไพโรไลซส เปนวธการหนงทสามารถใชผลตเชอเพลง และสารเคมอนๆจากวสด

เหลอใชเหลาน ดวยการใหความรอนในระบบทไมมออกซเจน กระบวนการไพโรไลซสสามารถเปลยนวตถดบ

เหลาน ใหกลายเปนผลตภณฑทมมลคาเพมได ในรปแบบของ แกส ของเหลว และกากของแขง

ปญหาหนงของกระบวนการไพโรไลซสทตองท าการศกษาเพมเตม คอ ปฏกรยาทเกดขนระหวางวตถดบ ซงไดแก

เซลลโลส ซงแปนสวนประกอบหลกของกระดาษ และพลาสตกประเภทตาง ๆ ส าหรบหวขอน เราไดท าการศกษา 3

หวขอยอย คอ ผลของวธการผสมวสด, ปฏกรยาระหวางพลาสตกและเซลลโลส และการเกดอนม H, OH และน า

ระหวางการไพโรไลซสของเซลลโลส การศกษาเชงทดลองทงสามงานน ไดใชเทคนคไพโรไลซสระดบไมโคร Py-

GC/MS เปนหลก

ในงานแรก พลาสตกโพลเอทลนและเซลลโลส ไดถกผสมโดยการหลอมละลาย และโดยการวางไวขางกน ผลการทดลองพบวา ในการผสมโดยการหลอมละลาย ท าใหเกดปฎกรยาทสงผลใหปรมาณ แอนไฮโดรช แอลดไฮดและค

โตน เพมขนเลกนอย แตผลผลตหลกนนไมไดรบผลกระทบ

ในงานทสอง ไดท าการศกษาตอในเรองปฏกรยาระหวางพลาสตก (PE, PP, PS, PET) และเซลลโลส โดยการใช

เทคนค Py-GC×GC/MS การวเคราะหสารเคมทไดจากการไพโรไลซสของวสดผสมนนสามารถแยกแยะไดงายขนกวา

เทคนคอน ผลการทดลองพบวา ปฏกรยาระหวางพลาสตกดวยกนนนมมากกวาปฏกรยาระหวางพลาสตกและเซลลโลส อยางไรกตาม ในการไพโรไลซสระหวาง PS และเซลลโลส มปฏกรยาการถายโอนไฮโดรเจนเกดขน โดย

แหลงทมาของไฮโดรเจนนนไมสามารถสรปไดชดเจน

ดงนน ในงานตอมา จงไดท าการศกษาการเกดอนพนธ H, OH และน า ระหวางกระบวนการไพโรไลซสของเซลลโลส

งานนเปนการศกษาเชงค านวณรวมกบการทดลอง โดยผจดท าวทยานพนธ เปนผรบผดชอบในสวนของการทดลอง

ผลการค านวณพบวา การเกดค อนมล H และ OH ซงรวมกนในภายหลงเปนโมเลกลของน านน มความนาจะเปนมากกวาการเกด อนมลเดยว เนองจากการเกดค อนพนธสงผลใหโครงสรางของเซลลโลสทเหลอ มความเสถยร

มากกวาโดยการสรางพนธะค จากการค านวณพลงงานของ Gibbs คาดวา น าจะถกปลดปลอยจากโครงสรางท

อณหภมประมาณ 280 °C การทดลองโดยการไพโรไลซสเซลลโลสทหลายอณหภม พบวา มความสอดคลองกบผลการค านวณโดย พบวามการเกดของน าในสองชวงอณหภม และชวงอณหภมแรกมจดสงสดใกล 280 °C

เนองจากปฏกรยาระหวางวสดตางๆในระหวางกระบวนการไพโรไลซสไมไดชวยใหค ณภาพของผลตภณฑดขนนก และของเหลวทไดจากไพโรไลซสของพลาสตกและกระดาษนน มคณสมบตทแตกตางกน การแยกผลตภณฑทงสอง

ออกจากกนนนจงอาจจะเปนทางเลอกทดกวา นอกจากน ไฮโดรคารบอนทไดจากกระบวนการไพโรไลซสของ

vi

พลาสตก และของเหลวทมปรมาณออกซเจนสงทไดจากกระบวนการไพโรไลซสของกระดาษนน ตองใชวธการ

ปรบปรงคณภาพทแตกตางกน เราจงไดเสนอกระบวนการไพโรไลซสแบบขน เพอผลตและเกบเก ยวผลตภณฑทง

สองโดยแยกจากกน โดยการใชวตถดบจ าลอง (PE, PS, เซลลโลส) และกากพลาสตกผสมกระดาษทเหลอจาก

กระบวนการรไซเคลกระดาษ กระบวนการไพโรไลซสแบบขนทอณหภม 300-350 °C ในขนทหนง และ 500 °C ในขนทสอง สามารถผลตผลตภณฑทงสองอยางทไดกลาวไปขางตนได อยางไรกตาม ยงคงตองมการศกษาเพมเตมใน

การหาคาทเหมาะสมของตวแปรในกระบวนการ การตรวจสอบคณสมบตของผลตภณฑ และวธการปรบปรงคณภาพ

ผลตภณฑ

ตอเนองจากการศกษากระบวนการไพโรไลซสแบบขน เราไดศกษาการปรบปรงคณภาพของผลตภณฑของเหลวทได

จากขนทหนงของกระบวนการไพโรไลซสแบบขน เมอมพลาสตก PVC ผสมอยในวตถดบ กระบวนการ ดไฮโดรคลอรเนชนของ PVC จะเกดขนในชวงอณหภมของขนทหนง ซงเปนชวงอณหภมเดยวกบการไพโรไลซสของ

เซลลโลส เราไดทดสอบการใชแคลเซยมออกไซด (CaO) ในการดดซบ HCl และปรบปรงคณภาพของผลตภณฑท

ไดจากการไพโรไลซสของเซลลโลส ผลการทดลองพบวา อณหภมทใชส าหรบ CaO ควรอยในชวง 300-350 °C เพราะจะเกดการคาย HCl ทอณหภมมากกวา 400 °C ซงสวนหนงเกดขนเนองจากน าทมาจากการไพโรไลซสของ

เซลลโลสท าปฏกรยากบ CaCl2 และปลอย HCl ออกมา

การการศกษาทไดกลาวมา กระบวนการไพโรไลซสแบบขน มศกยภาพในการผลตเชอเพลงและสารเคมอน ๆ จาก

พลาสตกผสมกบกระดาษ ยงคงตองมการศกษาเพมเตมเพอพฒนา ประเมน และ ท าใหกระบวนการนใชงานไดจรง

ค ำส ำคญ: ไพโรไลซส; พลาสตกและกระดาษ; เซลลโลส; เชอเพลงขยะ (RDF); กากพลาสตกผสมกระดาษทเหลอจาก

กระบวนการรไซเคลกระดาษ; ปฏกรยา; กระบวนการไพโรไลซสแบบขน; แคลเซยมออกไซด (CaO)

vii

Acknowledgements

I would like to express my gratitude toward my supervisor, Weihong Yang, for

giving me the opportunity to join the research group and giving all the supports

during my study. I would also like to express my gratitude toward my second

supervisor, Linda Sandström, for her supervision and helpful guidance on

experiments and preparation of the manuscripts.

I would like to thank all my co-authors for their helps and guidance on the works

and all the comments for the manuscript: Ann-Christine Johansson, Cláudio M.

Lousada, Ilman Nuran Zaini, Rikard Svanberg, Tong Han, Panagiotis

Evangelopoulos, Sergey Dvinskikh, Ayumu Tagami, Olena Sevastyanova and Pelle

Mellin. And I would also like to thank the anonymous reviewers and the journal

editors for their constructive comments.

I would like to thank my colleagues in energy and furnace technology group for

sharing all the good and bad times with me, Efthymios and Panos for helping me

passing through the tough time during my first and second year, Duleeka,

Chunguang, Pelle, Henry, Tong, Ilman, Devy, Rikard, Shule and Katarzyna for the

fun times we have shared in the office and labs. I feel thankful for your company.

You have made my work life more enjoyable. I am also grateful to have met and

known PhD students in our department, Carrie, Silvia and those not mentioned

here. Thank you for all the funs we have shared and thank you for your supports.

I would like to thank Thai student community in Stockholm for their company, for

sharing foods, and for discussion on both life and works. Also, I am grateful to

have met and known Thai student in the other cities of Sweden through the

TSAC2015 communities.

I would also like to thank all the friends I have met in KTH and in Stockholm, my

landladies – Lotta and Britt, my apartment mates, for sharing good times and

Swedish cultures.

I would like to thank Thai DPST scholarship for the financial support during 4

years for my study and I would like to thank Jernkontoret for the 6 months

scholarship during the preparation of my thesis.

I would like to thank all the supporting system from the department and the ITM

school. Many things have become easier with your helps. Thank you.

I would like to thank all my teachers and friends from my early schools and

universities for their contributions to my education and learning.

viii

Lastly, I would like to thank my family for their understanding and support. You

are the source of power and inspiration for me. I would like to thank my love, P’Eig,

for always being by my side on the good and bad times. I am so grateful that we

are together. I Love you.

Nanta Sophonrat

ix

List of Papers included in the Thesis

I. Sophonrat, N., & Yang, W. (2017). Effect of mixing methods of

polyethylene and cellulose on volatile products from its co-pyrolysis. Energy

Procedia, 142, 315-320.

II. Sophonrat, N., Sandstro m, L., Johansson, A.-C., & Yang, W. (2017). Co-

pyrolysis of Mixed Plastics and Cellulose: An Interaction Study by Py-

GCxGC/MS. Energy&Fuels, 31(10), 11078-11090.

III. Lousada, C. M., Sophonrat, N., & Yang, W. (2018). Mechanisms of

Formation of H, HO, and Water and of Water Desorption in the Early

Stages of Cellulose Pyrolysis. Journal of Physical Chemistry C, 122(23), 12168–

12176.

IV. Sophonrat, N., Sandström, L., Zaini, I. N., & Yang, W. (2018). Stepwise

pyrolysis of mixed plastics and paper for separation of oxygenated and

hydrocarbon condensates. Applied Energy, 229, 314-325.

V. Sophonrat, N., Sandström, L., Svanberg, R., Han, T., Dvinskikh, S.,

Lousada, C. M., & Yang, W. Ex-situ catalytic pyrolysis of mixture of polyvinyl chloride and cellulose using calcium oxide for HCl adsorption

and catalytic reforming of the pyrolysis products. To be submitted to Ind Eng

Chem Res.

Contribution Statement:

Papers I, II, IV, V: Performed experiments, analyzed data, prepared the manuscript

and responded to the reviewers’ comments.

Paper III: Proposed research statement, responsible for the experimental part and

prepared the manuscript on the experimental section and parts of the introduction

section.

x

List of Papers not included in the Thesis

1. Han, T., Sophonrat, N., Evangelopoulos, P., Persson, H., Yang, W., &

Jönsson, P. (2018). Evolution of sulfur during fast pyrolysis of sulfonated

Kraft lignin. J. Anal. Appl. Pyrolysis, 133, 162-168.

2. Evangelopoulos, P., Sophonrat, N., Jilvero, H., & Yang, W. (2018).

Investigation on the low-temperature pyrolysis of automotive shredder residue (ASR) for energy recovery and metal recycling. Waste Management,

76, 507-515.

3. Han, T., Sophonrat, N., Tagami, A., Sevastyanova, O., Mellin, P., & Yang, W. (2019). Characterization of lignin at pre-pyrolysis temperature to

investigate its melting problem. Fuel, 235, 1061-1069.

xi

Contents 1 Introduction 1

1.1 Introduction............................................................................................................ 1 1.2 Objectives................................................................................................................ 5 1.3 Structure of the dissertation................................................................................ 5 1.4 Sustainability aspects of the thesis..................................................................... 6

2 Background 9 2.1 Pyrolysis ................................................................................................................... 9 2.2 Thermal decomposition of plastics ................................................................. 13 2.3 Thermal decomposition of cellulose ............................................................... 15 2.4 Pyrolysis of mixed plastics and paper ............................................................. 16 2.5 Notes on feedstocks recycling .......................................................................... 17

3 Methodology 19 3.1 Feedstock materials ............................................................................................. 19 3.2 Experimental facilities ........................................................................................ 20

3.2.1 Py-GC/MS and Py-GCxGC/MS....................................................... 20 3.2.2 Bench-scale fixed bed reactor ............................................................. 21

3.3 Products analysis.................................................................................................. 22 3.3.1 Liquid analysis techniques.................................................................... 23 3.3.2 Gas analysis techniques ........................................................................ 23 3.3.3 Solid analysis techniques ...................................................................... 23

4 Investigation of interactions between plastics and cellulose using PY-

GC/MS 25 4.1 Introduction.......................................................................................................... 25 4.2 Paper I: Effect of mixing methods.................................................................. 26

4.2.1 Results and discussion .......................................................................... 27 4.3 Paper II: Interactions between plastics and cellulose ................................. 30

4.3.1 Results and discussion .......................................................................... 30 4.4 Paper III: H, OH and H2O formation during the early stages of

cellulose pyrolysis ................................................................................................ 37 4.4.1 Results and discussion .......................................................................... 38

5 Stepwise pyrolysis of mixed plastics and paper 41 5.1 Introduction.......................................................................................................... 41 5.2 Paper IV: Stepwise pyrolysis of mixed plastics and cellulose ................... 42

5.2.1 Results and discussion .......................................................................... 43 5.3 Paper V: Upgradation of products from the first step of stepwise

pyrolysis ................................................................................................................. 48 5.3.1 Results and discussion .......................................................................... 48

xii

6 Conclusions and future work 55 6.1 Conclusions........................................................................................................... 55 6.2 Future work .......................................................................................................... 56

7 References 58

1

Chapter 1

1 Introduction

1.1 Introduction

It is undeniable that plastic materials are essential to our daily lives due to their

properties that can hardly be replaced by other materials combined with their

affordable prices. However, due to the improper management of post-consumer

plastics, plastic waste is a huge environmental problem today causing pollution in

the oceans, rivers, and lands. An estimation of approximately 8 million tons of

plastics entered the ocean in 2010 [1] – equivalent to dumping one garbage truck

into the ocean every minute [2]. This affects not only the marine lives but inevitably

also our lives as microplastics have entered our food chain. It is critical to find

solutions to this problem. Following the circular economy concept, it is encouraged

that the materials are reused, remanufactured and recycled.

The different recycling techniques for plastics are classified as primary (closed-

loop), secondary (mechanical), tertiary (chemical) and energy recovery

(incineration) [3, 4].

• Primary recycling is a mechanical process that recovers materials that can

be used for the same purpose as the original materials. This is usually

possible for pre-consumer plastic residues from industries or well sorted

and cleaned materials, e.g., HDPE milk bottles [5].

• Secondary recycling or downgrading or cascaded recycling recovers plastics

with lower grade than the original plastics. Both primary and secondary

recycling usually involve sorting, grinding, cleaning, melting and remolding.

In general terminology, when refer to recycling, it usually means primary

and secondary recycling.

• Tertiary recycling or chemical recycling produces chemicals which can be

used as monomers for re-manufacturing plastics. Some examples of this

technique are pyrolysis and chemolysis/solvolysis which require sorting

and cleaning of plastics before processing [6, 7]. Although mechanical

recycling is simpler than tertiary recycling, the properties of the plastics

2

degrade after multiple cycles of mechanical recycling. For instance,

injection molding of polyethylene terephthalate (PET) reduces the

molecular weight (Mw) of the polymer by almost 70% after 3 processing

cycles which results in lower ductility of the plastic [8], and structural

changes and deterioration of mechanical properties of high-density

polyethylene (HDPE) have been observed after 30 extrusion cycles [9].

This makes tertiary recycling more attractive to recover monomers from

the degraded plastics, although the efficiencies and the economics of many

processes still need to be improved.

It should be noted that the production of fuels and industrial chemicals

[10] can also be the focus of tertiary recycling as it is not always possible to

recover monomers. It is, however, debatable whether the production of

fuels and other chemicals can be called recycled [11]. Nevertheless, the

production of both help reducing fossil fuel consumption as most of the

industrial chemicals used today are derived from petroleum products.

• Energy recovery by incineration is usually applied to plastics remained after

sorting or for the fraction that is difficult to sort. Although it is not

sustainable in the way that its products cannot be used to make new

materials [11], energy recovery helps reduce fossil fuel consumption.

According to the global flows of plastic packaging materials from the Ellen

MacArthur foundation’s report [2], only 2% of the plastics is recovered by closed-

loop recycling, 8% by cascaded recycling and 14% by energy recovery

(incineration). The rest of 40% is sent to landfill and 32% leaks to the environment.

There is a huge room for improvement regarding collection, sorting and developing

new recycling technology.

The difficult-to-sort plastics are usually mixed with other materials. The mixture is

called the rejected fraction as all the recyclables have been removed. Analyses of

the compositions of the rejected fraction are shown in Table 1 where the main

shares are food waste, paper, plastics and textiles. Table 2 shows the main types of

plastics found in the rejected fraction which are polyethylene (PE), polypropylene

(PP), polyethylene terephthalate (PET), polyvinyl chloride (PVC) and polystyrene

(PS). Although better sorting can reduce the contamination of other fractions in

plastics mixture, some composite materials are difficult to separate, e.g., paper cups

coated with plastic films, plastic-paper-aluminum composite packaging. These

composite materials also contaminate the paper fraction sent for recycling. As the

cellulose fibers are recovered by the paper recycling process, the contaminating

plastics are removed as a rejected fraction. This rejected fraction contains both

3

plastics and paper fiber [12]. Therefore, the future recycling process must be able

to handle mixed materials.

Table 1 Composition (wt%) of the rejected fraction or the non-recyclable fraction

of municipal solid waste from different areas.

Stockholm

Sweden [13]

Goteborg

Sweden

[13]

Onda

Spain

[14]

Castilla y

León,

Spain

[15]

A municipal

in Tondela,

Portugal a

[16]

Bangkok

Thailand

[17]

A university

in Surin,

Thailand

[18]

Food and garden waste

14.6 24.8 16.8 23.9 2.53 9.48 18

Paper 37.6 20.8 32.2 27.9 34.9 9.6 14

Plastics 25.7 15.8 22.2 24.5 34.4 41.1 45

Textiles 4.5 3.6 7.91 8.7 18.4 4.2 19

Wood 1.6 17.3 3.3 2.2 0.6 27.5 c

Absorbent hygiene products

6.7 1.6 4.2 5.8

Composite packaging

2.0 2.2

Other organic matter b

2.0 0.03 10.0 d 0.4 3 d

Metal 3.1 3.8 4.39 3.7 4.8 1.0

Glass 3.4 1.8 1.1 0.5 5.1

0.7 1

Other inorganic 0.4 3.0 1.42 0.48 6.3

Hazardous waste and WEEE

0.2 0.7 2.44 0.34 N/A ND N/A

a from fraction with low food waste. b other organic waste includes rubber, leather, footwear. c wood and leave d did not specify the composition

The previously described tertiary recycling techniques are usually applied to well-

sorted plastics. For mixed plastics, the method of interest is pyrolysis as it does not

require sorting, and the mixed liquid product has a potential to be further upgraded

to fuels and chemicals [19, 20]. Pyrolysis of mixed plastics with other materials as

in municipal solid waste also has a potential to produce fuels and chemicals [21],

but the concerns regarding high water content and complex chemical compositions

should be further investigated [22].

4

It is important to understand the interactions between the feedstock components

during pyrolysis and their effects on the products compositions. The main types of

plastics are PE, PP, PS, PET, PVC as shown earlier in Table 1. For paper, the main

composition is cellulose. In this work, the interactions between cellulose and the

main types of plastics found in the rejected fraction were investigated. Based on the

findings and literature reviews, a pyrolysis process was then proposed to recover

fuels and other chemicals from this waste stream.

Table 2 Plastic wastes compositions from different area in percentage of total

plastic waste (wt%).

Sources

PE

(HDPE,

LDPE)

PP PET PVC PS Other

plastics

Residual a

Mixed plastic

packaging, UK [23] 38.4 22.2 15.3 3.5 4 16.5

Plastic waste fraction

fed to waste-to-energy

plants, Austria [24]

60±17 6±4 8±6 10±7 ~10

Plastic waste in reject

fraction after

mechanical sorting in

Spain [14]

62 5 15 N/A 5 13

Plastic waste in RDF,

Spain [15] 48.4 N/A 7.6 0.3 N/A 43.4 b

Plastic waste left after

source separated in

MSW, Bangkok (SNK

community) [25]

67±6 16±4 5±2 1±1 9±2 3±1

a Include non-packaging plastics and other non-plastic contamination e.g. milk carton b Include all other plastics except, PE, PET, PVC.

5

1.2 Objectives

• To investigate co-pyrolysis of plastics and paper for the production of fuels

and other chemicals.

o To study the effect of mixing methods on co-pyrolysis of PE and

cellulose (Paper I).

o To investigate the interactions between mixed plastics and

cellulose during their co-pyrolysis (Paper II).

o To investigate the mechanism of H and OH radicals’ formation

during cellulose pyrolysis (Paper III).

• To propose a process to recover fuels and other chemicals from the mixed

plastics and paper.

o To investigate the possibility of stepwise pyrolysis of mixed

plastics and paper (Paper IV).

o To investigate the first step of stepwise pyrolysis of PVC and

cellulose mixture by using CaO for the adsorption of HCl and

biooil upgradation (Paper V).

1.3 Structure of the dissertation

This thesis is a compendium of 5 publications prepared during the PhD study as

listed on page x. The papers can be put into two groups based on the objectives of

the work as shown in Figure 1:

• Fundamental investigations: paper I, II, III

• Process development in lab-scale facilities: paper IV, V

The thesis is then organized into 7 chapters. Chapter 2 provides the background

for the thesis. The more specific literature reviews are presented in the introduction

sections of chapter 4 and 5. Chapter 3 presents the methods used in this st udy

which are mainly experimental techniques.

Chapter 4 is based on paper I, II and III where interactions between mixed plastics

and cellulose were investigated using Py-GC/MS techniques. The first paper

discusses the effect of mixing method on pyrolysis products from PE and cellulose

co-pyrolysis. The second paper investigates the interaction between mixed plastics

and paper. It was difficult to explain some of the results on interactions regarding

hydrogen transfer. Therefore, the third paper attempts to explain these results by

investigating the possibility of H and OH radicals’ generation during cellulose

pyrolysis.

6

Chapter 5 is based on papers IV and V. As some understandings were gathered, a

stepwise pyrolysis process of mixed plastics and paper has been proposed and

tested in lab-scale experiments (paper IV). Paper V continues the investigation on

the first-step of the stepwise pyrolysis when PVC is present in the feedstock. The

performance of calcium oxide employed to adsorb HCl and reform biooil was

investigated.

Chapters 6 presents the conclusions and suggestions for future works.

Figure 1 Overview of the thesis work.

1.4 Sustainability aspects of the thesis

One of the objectives of this thesis is to propose a pyrolysis process to recover

fuels and other chemicals from the difficult-to-sort fraction of municipal solid

waste, e.g., plastics and paper composite packaging, and rejected fraction from

paper recycling. If the recovered chemicals can be used to produce new plastics,

then the process can help increase the recycling rate and enhance the sustainable

consumption and production patterns according to the UN sustainable

development goal 12 (SDG 12) [26]. If the recovered chemicals are used as fuels,

then it would help reduce fossil fuels consumption and could be considered as a

part of SDG 7 which is to ensure the access to affordable, reliable, sustainable and

7

modern energy for all. This, however, required further investigation on the

economics of the process.

Moreover, pyrolysis as a chemical recycling process has a potential to help combat

climate change according to SDG 13 by reducing CO2 emission as the carbons in

the original materials are maintained in the liquid chemicals and solid char.

Nevertheless, the process should be self-sustained and/or used renewable energy

sources.

A better recycling technology in general will encourage a better waste management

system. This will result in the cities becoming more sustainable (SDG 11) and the

oceans and lands becoming less polluted (SDG 14 and 15).

8

9

Chapter 2

2 Background

In this chapter, the backgrounds on pyrolysis and thermal decompositions of

cellulose, the main composition of paper, and plastics and are presented, followed

by a literature review on pyrolysis of mixed plastics and paper. The chapter is ended

with a note on feedstock recycling from plastics.

2.1 Pyrolysis

Pyrolysis is a thermal process in which the feedstock materials are thermally

decomposed in a non-oxidizing atmosphere. Depending on the material properties

and heating conditions, when the material is heated, it can undergo physical change

(e.g., deformation, melting) and chemical changes, i.e., breaking of chemical bonds.

The broken bonds generate radicals which can propagate in the chemical structure

and recombine with other radicals. As more heat is applied, more bonds are broken,

and as soon as the products are small enough, they evaporate and leave the bulk

material behind as a solid residue. The volatilized products can further decompose

due to heat or react with other products, resulting in the formation of a wide range

of compounds. Some of these compounds can be condensed into liquid products

and some are permanent gases. The final products are then the solid resid ue or

char, liquids and gases.

Feedstock ℎ𝑒𝑎𝑡→ Gas + Liquid + Solid residue (char)

The pyrolysis process is governed by chemical kinetics and transport phenomena.

It is a complicate process as many reactions occur concurrently and different

reactions respond differently to the heating condit ion applied. Heat and mass

transfer are needed to be considered, e.g., how fast the material is heated, or how

fast the products are swept away from the reaction or heating zone. This makes it

difficult to model pyrolysis reactions and to simulate pyrolysis process.

Experimental investigation is then necessary in many cases.

10

In other words, the quantities and compositions of the pyrolysis products depend

on the properties of the feedstocks and the process conditions. Figure 2 shows how

the feedstocks and process conditions affect the final pyrolysis products. As

previously described, the chemical kinetics and transport phenomena influence the

pyrolysis process in a complex way. Considering the phenomena associated with

thermal decomposition of solid, namely, chemical kinetics, internal and external

heat transfer, the decomposition process can be dominated by one of these

phenomena if the characteristic time of one phenomenon is considerably longer

than the others. This can be classified using a set of dimensionless numbers,

namely, Biot number (Bi), Damkohler number (Da), and Thiele number (Th) [27,

28]. Table 3 shows the process regimes and the calculations of these numbers. As

a rough guide of the phenomena controlling the process, it is important to know if

the process fall into any regime or not for the sake of describing and comparing

the results from different pyrolysis system, e.g., fixed bed, fluidized bed, micro-

pyrolyzer. If the process is in the “external heat transfer controls” or “internal heat

transfer controls” regime, then the reaction(s) (in this case thermal decomposition

reactions) occur at different temperature throughout the bulk of the material

resulting in a mixture of products from different temperature. In determining

kinetic parameters of thermal decomposition reaction of a feedstock, the process

should then be in the “chemical reaction controls” regime.

Although it is useful to know the process regime of the system of interest, it is not

commonly reported in most experimental works. This is due to 1) it is not always

the case that a process would fall into one of the regimes [27], 2) some process

parameters are difficult to evaluate, e.g., kinetic parameters of different feedstocks

[29], heat transfer coefficient in the reactor, h, which is estimated from empirical

and semi-empirical formula [30, 31], 3) the extent of the secondary reactions adds

to the complexity to the final products (refer back to Figure 2). Therefore, it is more

convenient to describe the pyrolysis process with other process parameters.

11

Figure 2 Parameters and phenomena affecting the final pyrolysis products

Table 3 Process regimes of the thermal decomposition of solid [28, 32].

Bi =

ℎ𝐿

𝑘≪ 1

Thin particle: homogeneous

temperature inside particle

Bi =ℎ𝐿

𝑘≫ 1

Thick particle: temperature

gradient inside particle

Da =𝑡𝑟ℎ

𝜌𝑐𝑝𝐿≪ 1 Da =

𝑡𝑟ℎ

𝜌𝑐𝑝𝐿≫ 1 Th =

𝜌𝑐𝑝𝐿2

𝑡𝑟𝑘≪ 1 Th =

𝜌𝑐𝑝𝐿2

𝑡𝑟𝑘≫ 1

𝑡𝑖𝑛 ≪ 𝑡𝑒𝑥,

tr≪ 𝑡𝑒𝑥 𝑡𝑖𝑛≪ 𝑡𝑒𝑥 ≪ 𝑡𝑟 𝑡𝑒𝑥 ≪ 𝑡𝑖𝑛 ≪ 𝑡𝑟

𝑡𝑒𝑥≪ 𝑡𝑖𝑛,

𝑡𝑟 ≪ 𝑡𝑖𝑛

Decomposition

regime

External heat

transfer

controls

Chemical

reaction

controls

Chemical

reaction

controls

Internal heat

transfer

controls

where

𝑐𝑝 is specific heat capacity (J/gK)

ℎ is heat transfer coefficient in reactor (W/Km2)

𝑘 is thermal conductivity of material (W/Km)

𝐿 is characteristic length of particle, e.g., for sphere L = diameter/6

𝑡𝑖𝑛 is internal heat transfer characteristic time, 𝑡𝑖𝑛 = 𝜌𝑐𝑝𝐿2/𝑘

𝑡𝑒𝑥 is external heat transfer characteristic time, 𝑡𝑒𝑥 = 𝜌𝑐𝑝𝐿/ℎ

𝑡𝑟 is reaction time for a specific ratio of conversion e.g. 50% or 90%

conversion

𝛼 is conversion ratio of solid in thermal decomposition

𝜌 is density of material (g/cm3)

12

The process parameters commonly reported are temperature, heating rate and

vapor residence time. Example of other process parameters reported are reactor

dimensions, carrier gas flow rate, types of heating. In many cases, inert gases such

as nitrogen can be used as carrier gases in the pyrolysis process, or the produced

gases can be recycled and used as a carrier gas. Some operating parameters can be

related to vapor residence time as shown in Figure 3. The temperature of the

pyrolysis process is usually in the range of 300-600 °C. The heating rate of the

system, which can be different than that of the particle [33], can vary from slow to

very fast (1-10000 K/min). The vapor residence time can also vary in a wide range:

seconds – days. These three parameters can be used to roughly classify the pyrolysis

process into [34, 35, 36]:

• Slow pyrolysis or carbonization: temperature <400 °C, slow heating rate in

the order of <10 K/min, long vapor residence time (hours to days), high

yield of char, example: charcoal production

• Fast pyrolysis: temperature usually >400 °C, fast heating rate (~1000 K/s),

short vapor residence time (up to 2 s.), high yield of liquid products,

example: fluidized bed reactor.

• Intermediate pyrolysis or conventional pyrolysis: temperature usually >400

°C, moderate heating rate (~10-100 K/min), vapor residence time 10s-

30m, moderate yields of liquid and gas.

In today’s research, one of the main aims of using pyrolysis processes is to produce

liquid products. For biomass pyrolysis, high yield of liquid can be achieved by fast

pyrolysis. While for plastics, the production of liquid and/or wax can be achieved

by both fast and intermediate pyrolysis. The liquid products from pyrolysis of both

biomass and plastics can be upgraded into fuels and other chemicals [19, 35].

Next sections present some backgrounds on thermal decomposition of plastics and

cellulose.

13

Figure 3 Operating parameters affecting the residence time of volatiles

2.2 Thermal decomposition of plastics

As previously described in Chapter 1, the common plastics found in the rejected

fraction are PE, PP, PS, PET, and PVC, which fall into the category of

thermoplastics. Table 4 shows the monomers of each plastics.

The pyrolysis products of these plastics are rather different. Low yields of solid

residue for all the plastics can be seen from thermogravimetric analysis. Pyrolysis

of PE, PP and PS produces almost no solid residue [37]. On the other hand,

pyrolysis of PET and PVC leaves small amount of solid residue (<10%) [37, 38].

The volatile products from PE and PP pyrolysis can be condensed and are mainly

in the form of wax and liquid, while that from PS are mainly in the liquid form.

Some parts of the volatile products from PET pyrolysis can be in the form of solid

due to the nature of the compounds such as benzoic acid and terephthalic acid [39].

For PVC, hydrogen chloride can be emitted at temperatures >300 °C. The main

liquid products from PVC pyrolysis are aromatic hydrocarbons, e.g., benzene, and

PAHs [40, 41].

Although, the investigated plastics usually contain additives which affect the

thermal decomposition of these plastics [19, 42, 40], these components are outside

the scope of this thesis and will be left for future study.

14

Table 4 Plastics and their monomers

Plastics Monomers

Polyethylene (PE) C2H4

Polypropylene (PP) C3H6

Polystyrene (PS) C8H8

Polyethylene

terephthalate (PET)

C10H8O4

Polyvinyl chloride

(PVC) C2H3Cl

The thermal cracking or thermal decomposition mechanisms of plastics involves

chain reactions which can be broadly classified into radical forming, radical

interconverting and radical consuming [29]. For PE and PP, some reactions are

shown as follows [19, 29] where µ is a hydrocarbon radical, β is a radical smaller

than µ, and H is hydrogen.

Chain fission/Recombination µ-H ⇌ β + β'

H-transfer or H-abstraction (both ways) µ-H + β ⇌ µ + β-H

µ-H + µ' ⇌ µ + µ'-H

Decomposition or β-scission/Addition µ ⇌ β + olefin

Isomerization (both ways) µ ⇌ µ'

µ + µ-H ⇌ µ' + µ-H

Disproportionation β + β' → paraffin + olefin

Termination β + µ → products

For PE and PP, the main products from thermal cracking are therefore a mixture

of olefins and paraffins with wide range of carbon number (C1-C>40) [43, 44].

15

The thermal decomposition mechanism of PS is similar to that of PE and PP as

the C-C bond in the main carbon chain is easier to break than the C-C bond

connecting C in the main chain to the phenyl group [45]. The main products from

PS composition is then styrene and styrene oligomers [46].

Thermal decomposition mechanism of PET starts from random scissions of ester

bonds via the transfer of β hydrogen forming a carboxylic acid and olefinic end

groups [47, 48] as shown in Figure 4. The products then further decompose via

ester scission reactions, decarboxylation and decarbonylation. The main volatile

products from PET pyrolysis are CO2, acetaldehyde, 4-(vinyloxycarbonyl) benzoic

acid, benzoic acid, vinyl benzoate and divinyl terephthalate [42, 47].

The thermal decomposition mechanism of PVC has been proposed by many works

as reviewed by Yu et al. [40]. An example of the proposed mechanism is a 4-stage

mechanism [49]: 1) dechlorination with inner cyclization, 2) aromatic chain scission,

3) cyclization/aromatization and 4) degradation from coke formation. This results

in HCl and aromatic hydrocarbons being the main products of PVC pyrolysis.

Figure 4 Decomposition mechanism of PET.

2.3 Thermal decomposition of cellulose

Cellulose is one of the main components of plant cell walls which is used for paper

production. It is a linear polysaccharide consisting of D-glucose units linked by

β(1,4)-glycosidic bonds as shown in Figure 5. The hydrogen bonds between

cellulose chains form a layered structure of cellulose. The different layers then hold

together via Van der Waals forces [50]. The natural crystal structure of cellulose is

Iα and Iβ, the latter being found much in higher plants.

Thermal decomposition of cellulose gives rise to various products such as

levoglucosan, levoglucosenone, hydroxymethyl furfural, furfural,

hydroxyacetaldehyde and acetic acid. Pyrolysis of cellulose mainly generates the

16

same products but with varied yields based on the pyrolysis conditions and the

properties of the cellulose, e.g., degree of polymerization and crystallinity [50].

There are many on-going studies and proposed decomposition mechanisms of

cellulose and the formation of different products [51, 50]. Some proposed reactions

involve breaking of glycosidic bonds by a concerted pathway to produce

levoglucosan, followed by several steps of dehydration and decarboxylation of the

formed products [50].

Figure 5 Basic unit of cellulose polymer.

The liquid products produced by biomass pyrolysis is called biooil. Biooil form

cellulose pyrolysis consists of many oxygenated compounds as suggested above.

Due to the high oxygen content of this liquid, its heating value is lower than that

of hydrocarbon fuels. Attempts to use biooil as fuel requires testing and standards

[52]. Upgradation to obtain better fuel quality is also a topic of current interest [35].

In most of the laboratory works, a trade-off between the high yield and a better

fuel quality is still an issue [35]. Nevertheless, some interesting technologies are

emerging, e.g., the IH2 process [53]. Since cellulose derived from biomass is a

renewable source, the production of fuels or other chemicals from it is considered

sustainable and environmentally friendly.

2.4 Pyrolysis of mixed plastics and paper

As discussed in the introduction section about the rejected fraction consisting of

mixed plastics and paper, the co-pyrolysis of mixed materials has been investigated

with different goals in mind. One is the production of fuel from the already mixed

materials. Another one proposes that the higher H/C ratios in plastics might help

improve the product quality of the biooil from biomass pyrolysis, thus intentionally

mixing plastics with biomass. Some recent literature reviews on co-pyrolysis of

biomass and plastics are presented based on the criteria of heating rate and the

presence of catalyst [54]. In this work, the emphasis is given on the non-catalytic

co-pyrolysis and some reviews have already been presented in the introduction

section of Paper II.

17

2.5 Notes on feedstocks recycling

Recovering monomers from plastics is an idealistic goal of feedstock recycling.

Table 5 shows the chemical building blocks of different plastics. Fundamentally,

the monomers are derived from basic petrochemicals such as hydrocarbon gases

(ethylene, propylene) and aromatic hydrocarbons (benzene, xylene). These

products are derived from petrochemical refineries where different products are

produced from crude oil, e.g., light gas oil, heavy gas oil, liquid petroleum gas. These

products require multiple steps of purification, separation and upgrading to become

final products. This implies the multiple steps required to produce plastic

feedstocks.

On the other hand, liquid fuel as a mixture of hydrocarbons is also an interesting

product to be recovered from pyrolysis of plastics or plastics mixed with biomass.

As the initial products from pyrolysis are mixture of compounds, the most

convenient use is as a liquid fuel for transportation, since less upgradation and

separation processes are required. The properties of the fuel should be comparable

to the petroleum fuel standard [55, 56] or the standard for pyrolysis liquid biofuels

[52].

18

Table 5 Chemical building blocks of some common plastics [57].

Plastics Feedstocks Notes

Polyethylene (PE) C2H4, H2 Some comonomers are 1-

butene, 1-hexene, 1-octene

Polypropylene (PP) C3H6, C2H4, H2 Similar comonomers as PE

Polystyrene (PS) Styrene Styrene is produced from

ethylbenzene which is made by

catalytic alkylation of benzene

and ethylene

Polyethylene

terephthalate (PET)

Monoethylene glycol

(MEG) and purified

terephthalic acid (PTA) or

dimethyl terephthalate

(DMT)

Ethylene glycol (or 1,2-

ethanediol) is derived from

Ethylene oxide + H2O →

ethylene glycol

where ethylene oxide is

produced from direct

oxidation of ethylene by

oxygen.

PTA is produced from xylene

oxidation in air

DMT is produced from p-

xylene and methanol

Polyvinyl chloride

(PVC)

Vinyl chloride (C2H3Cl) Vinyl chloride production

steps are:

1) chlorination reaction

C2H4 + Cl2 → C2H4Cl2

and oxychlorination reaction

C2H4+O2+HCl → C2H4Cl2

2) heat treatment to convert

C2H4Cl2 to vinyl chloride with

HCl removal

19

Chapter 3

3 Methodology

3.1 Feedstock materials

The feedstock used were both neat and mixed materials. Table 6 gives ultimate and

proximate analysis of the materials. The cellulose used in this work is

microcrystalline cellulose purchased from Sigma-Aldridge. It is an alpha cellulose

(type Iβ) with a typical degree of polymerization of less than 400 [58, 59]. The

plastics powder and film were purchased from Good fellow company, except for

the PVC which was acquired from INEOS Chlor Vinyls by a previous lab member

[60].

The pelletized and non-pelletized paper reject was kindly provided by Smurfit

Kappa Roermond paper mill in corporation with RISE Energy Technology Center

(Piteå). It is the rejected fraction from the paper recycling process and mainly

contain plastics and some paper fibers attached to the plastics.

Table 6 Ultimate and proximate analyses of the feedstocks (wt%).

Materials Forms Elemental analysis Proximate analysis

C H N O* Cl S Volatile Ash Moisture

Cellulose Powder, 50 µm

44.1 6.3 < 0.1† 49.5 88.9 <0.3†

Polyethylene (PE, Low density)

Film, 50 µm thick

84.8 13.7 < 0.1 1.5 100 <0.3

Polyethylene (PE) mixed with polyethylene terephthalate (PET)

Powder, <63 µm

69.4 7.5 <0.1 23.1 92.0 <0.3

Polystyrene (PS)

Powder, <63 µm

90.9 7.7 <0.1 1.4 100 <0.3

20

Table 6 Cont.

Materials Forms Elemental analysis Proximate analysis

C H N O* Cl S Volatile Ash Moisture

Polypropylene (PP)

Powder, <63 µm

85.4 13.9 <0.1 0.7 100 <0.3

Polyethylene (PE) mixed with polyethylene terephthalate (PET)

Powder, 300 µm

80.7 12.2 <0.1 7.1 100 <0.3

Polystyrene (PS)

Powder, 250 µm

90.9 7.7 <0.1 1.4 100 <0.3

Polyvinylchloride (PVC)

Powder, 250 µm

38.4 5.0 <0.1† 4.2 52.0 0.04 93.0 0.3

Pelletized paper reject

Pellets 54.5 7.5 0.27 29.3 1.00 0.100 85.3 7.4 2.3

Non-pelletized paper reject

Shredded form

52.9 7.5 0.49 16.6 1.34 0.138 76.5 21 1

* Calculated from 100% difference, † lower limit of detection

3.2 Experimental facilities

3.2.1 Py-GC/MS and Py-GCxGC/MS

Two setups for pyrolysis-gas chromatography connected to mass spectroscopy (Py-

GC/MS) were used in this study. The details for each setup are as follow:

1. Pyrola 2000 connected to GC/MS Agilent 7890A GC and Agilent 5975C MS.

This setup (shown in Figure 6) is at the department of materials science and

engineering, KTH, Stockholm. The Pyrola 2000 [61] is a micro-pyrolyzer capable

of pyrolyzing small amount of samples in the range of ~ 0.1-1 mg. The pyrolyzer

is a platinum filament which is electrically heated. The initial heat-up time is 8 ms,

thereafter the time to reach the set temperature/the pyrolysis time can be adjusted,

e.g., 2-10 s. The temperature is measured by a resistance temperature detector at

the temperature <500 °C and an optical temperature sensor at temperatures of 500-

1100 °C. The calibration of the resistance temperature detector is done by using

the optical temperature sensor at temperatures >500 °C.

The GC can be equipped with many types of column such as HP-5ms (30 m or 60

m × 0.25 mm × 0.25 µm, Agilent), VF-1701 (60 m × 0.25 mm × 0.25 µm, Agilent),

21

DB-1701 (30 m × 0.25 mm × 0.25 µm, Agilent). The methods for each column are

different as shown in paper I-V.

The mass spectrometer scanning range can be adjusted, e.g., between m/z 15 - 550.

NIST-11 library was employed for peaks identification.

Figure 6 Pyrola 2000 - GC/MS

2. Frontier lab PY-3030S single shot pyrolyzer connected to GC×GC/MS

Shimadzu QP 2010 Ultra EI system with liquid nitrogen modulator Zoex ZX1.

This setup is RISE Energy Technology Center in Piteå. The Frontier lab pyrolyzer,

PY-3030S [62], is a micro-pyrolyzer for small amount of sample in the range of

~0.1-1 mg. A small deactivated stainless-steel cup is employed as a sample holder.

To start the pyrolysis, this cup is dropped into the furnace hold at a set temperature.

The pyrolyzer is connected to a GC equipped with 2 columns connected in series

which were, in this work, DB-5 (60 m × 0.25 mm × 0.15 µm, Agilent) and DB-17

(1 m × 0.18 mm × 0.18 µm, Agilent). The modulator, Zoex ZX1, connecting the

two columns uses liquid nitrogen to refocus the compounds emitted from the first

column into the second column [63]. The compounds separated by the first column

is further separated in the second column.

The mass spectrometer scanning range was set to m/z 28.5-320 to avoid the signal

from nitrogen gas. NIST-14 library was employed for peaks identification.

3.2.2 Bench-scale fixed bed reactor

The small bench-scale fixed bed reactor is an in-house facility at KTH, which is

capable of pyrolyzing samples of ~1-15 g. As shown in Figure 7, two setups have

been constructed and used in this thesis: a single-stage reactor and a two-stage

reactor. The schemes of each reactor are shown in paper IV and V. The pyrolysis

reactors are made of 316L stainless-steel tubes. The electric heaters can reach a

temperature of 1000 °C. For the two-stage reactor, a transfer tube connecting the

first and the second stage reactors is heated using heating tape which can reach a

22

maximum temperature of 450 °C. The sample basket is connected to a

thermocouple which can be moved from the water-cooling zone down to the pre-

heated reactor. Nitrogen gas was supplied from the top of the system using either

digital or analog flow meters. A cooling bath was used to condense the pyrolysis

vapors. The non-condensable gas was analyzed using Micro-gas chromatography.

The volume of the gas was read from a volumetric flow meter.

Figure 7 Left: one stage fixed bed reactor with electric heater and water-cooling jacket on the top. Right: two stage reactors (a vertical reactor connected to a

horizontal reactor) with electric heaters and water-cooling jacket on the top part of

the reactor.

3.3 Products analysis

When experiments were performed in the bench-scale fixed bed reactor, the

products were collected and analyzed separately. To obtain mass balances, different

parts of the of the experimental setup were weighed before and after each

experiment including the sample basket, the condensers, and the catalyst reactor.

The liquid, gas, solid residue and catalyst were analyzed according to the following

methods:

23

3.3.1 Liquid analysis techniques

• Liquid composition was analyzed by GC/MS. The liquid sample was

dissolved in a solvent, e.g., methanol, methanol and dichloromethane

mixture. The solution was then analyzed on gas chromatography – mass

spectrometry (GC/MS, Agilent 7890A GC and Agilent 5975C MS)

equipped with automatic liquid sampler. The column used were either HP-

5ms or DB-1701 column. NIST-11 library was used to identify the peaks.

• Water content was analyzed by Karl fisher titration using Mettler Toledo

Excellent titrator T5. A know amount of liquid sample was introduced into

Hydranal – Ketosolver and was titrated with Hydranal composite 5 K.

• Total acid number (TAN) titration was performed using Mettler Toledo

Excellent titrator T5. A known amount of liquid sample was dissolved in

either methanol (for oil) or 1:1 mixture of toluene and 2-propanol (for wax

sample). The solution was then titrated with 0.1 M KOH in EtOH

according to the ASTM D664 method.

• Chlorine content in the liquid was analyzed using Cl35 NMR which is only

sensitive for Cl ion as derived from HCl in aqueous solution.

3.3.2 Gas analysis techniques

• Gas composition was analyzed by a micro gas chromatography - thermal

conductivity detector (Micro GC-TCD, Agilent 490) consisting of 4

columns calibrated for the volumetric percentage of different gases. The

first column, MS5A (Agilent), is for H2, O2, N2, and CO analysis. The

second column, PPU (Agilent), is for CO2, C2H4, C2H6, C2H2 analysis. The

third column, Al2O3/KCl (Agilent), is for C3H8, C3H6 (cyclopropane and

propene), and C4 gases. The fourth column, CP-Sil 5CB (Agilent), is for

higher hydrocarbon gases but was not in used.

• Volume of gas was measured by drum-type gas meter (Ritter TG 1).

3.3.3 Solid analysis techniques

• Elemental analysis was performed by external laboratories (BELAB AB,

Eurofins) using a combustion method following the standard SS-EN ISO

16948:2015 or SS-EN 15407:2011.

• X-ray diffraction (XRD) analysis of catalysts was performed using Bruker

D8 Discover using Co radiation source with Kα1 of 1.79026 Å.

• X-ray fluorescence spectroscopy (XRF) of the catalyst was performed by

an external laboratory (Degerfors Lab).

24

25

Chapter 4

4 Investigation of interactions between plastics and cellulose using Py-GC/MS

4.1 Introduction

This chapter is a compilation of the results from three papers, paper I-III. The main

technique employed was Py-GC/MS which is a comparatively fast technique as

compared to bench-scale or pilot scale experiments and requires small amount of

sample. Therefore, a larger number of investigation cases can be performed.

However, a limitation of this technique is that only a range of volatile pyrolysis

products can be studied. The interpretations of the results are then bound to this

condition.

Mixed feedstocks can be in the form of composite materials where plastics and

paper are attached together tightly. The first paper then investigates if the mixing

methods have any effect on the interactions between polyethylene and cellulose

during co-pyrolysis.

It has been suggested that plastics with higher hydrogen to carbon ratio than

cellulose, when co-pyrolyzed together, would act as hydrogen source and improve

the oil quality [64, 65]. The interaction between plastics and cellulose should be

clarified. Many works have investigated the interactions between different plastics

and cellulose [66]. Nevertheless, when more than one types of plastics are presented

in the mixture, it can be difficult to analyze the products mixture. Therefore, in the

second paper, we investigated the interactions between mixed plastics and cellulose

using Py-GCxGC/MS. With the GCxGC, the separation of pyrolysis products is

better, thus assisting the analysis.

From the second paper, some hydrogen transfer reaction occurred when co-

pyrolyzed plastics and cellulose, e.g., between PS and cellulose. However, it was

difficult to pinpoint the source of hydrogen. We then set out to investigate the

26

formation of hydrogen and hydroxyl radicals during cellulose pyrolysis as presented

in paper III. This paper is a combination of computational (first -principles

calculations) and experimental investigation. As the thesis author was responsible

for the experimental part, details of the experimental section and the overview of

the results and conclusion will be presented.

4.2 Paper I: Effect of mixing methods

The mixing methods tested were 1) mixing by melt ing and 2) not mixing. For

mixing by melting, the cellulose powder was spread equally on top of PE film. The

film and the powder were put between two aluminum foils and clamped between

two ceramic plates. The plates were put into the oven at 105 °C overnight as the

cellulose powder did not attach to the PE film when heating duration in the oven

was too short (2-3 hours). Figure 8 shows the preparation of the melted sample.

The sample was cut into small circular pieces using a cutting mold. For the non-

mixing case, the two feedstocks were put side-by-side on the sample holder. The

sample size was 0.07-0.18 mg. The experiments were performed on Py-GC/MS

using Pyrola 2000 system.

The total ion chromatogram (TIC) area percent are reported for the two cases

together with the calculated values derived from neat materials as shown in Figure

10. The calculated value is equal to the weighted average of the area percent of neat

materials. The weighing factor (W𝑖) is calculated as follows with the assumption

that half of the unknown derived from each material:

W𝑖=Total area of products derived from feedstock 𝑖+ 0.5*(total area of unknown products)

Total area of the whole chromatogram

where 𝑖 is either PE or cellulose. The TIC area instead of area percent could also

be used to report the result; however, due to the uncertainty of the instrument used

for weight measurements, it is more suitable to use area percent.

Figure 8 Preparation of cellulose and PE mixture by melting. Left: sample before

being put into the oven. Right: After taking out of the oven

27

4.2.1 Results and discussion

Figure 9 shows the total ion chromatogram of the neat and mixed materials. The

chromatograms of mixed materials are nearly the same as a direct addition of

chromatogram of neat materials. This shows that the main products derived from

pyrolysis of each material were not affected by co-pyrolysis.

From Figure 10, small changes could be observed in the mixing case, i.e., a decrease

in the formation of aldehydes and increases in the formation of ketones, pyran, and

acetic acid. Similar finding was observed by Xue et al. [67], no explanation was

given on this result. It is obvious that the changes are more pronounce in the case

of close contact mixing as compared to not-mixing case. The main products such

as hydrocarbons and anhydrosugars (mainly levoglucosan) were not affected.

Figure 11 shows the ratios of area percent of the experimental cases to the

calculated values for specific compounds. For the not-mixing case, the ratios were

close to 1, while for the melting case the deviations from the ratio of 1 were more

significant. Higher yields of anhydrosugars and lower yields of small aldehydes

indicate that further decomposition of cellulose pyrolysis products was inhibited

when the cellulose was melted together with PE. Furthermore, the increase of

ketones in the melting case (Figure 10) was attributed to the formation of aliphatic

ketones, e.g., 2-hexanone, 2-oxanone, 2-nonanone, and 2-dodecanone. This might

indicate interaction between the small oxygenates from cellulose pyrolysis products

with unsaturated aliphatics from PE pyrolysis, e.g., acetaldehyde with aliphatic

alkenes.

Overall, the main products are not affected by co-pyrolysis whether the feedstock

were mixed or not. Small changes were observed in the formation of oxygenated

compounds. These changes were more notable in the case of mixing by melting

than in the non-mixing case.

28

Figure 9 Total ion chromatogram from pyrolysis of (a) PE, (b) cellulose, (c)

PE/cellulose not mixed and (d) PE/cellulose mixed by long melting

29

Figure 10 TIC Area percent of different group of compounds from co-pyrolysis of

PE and cellulose when different mixing conditions were applied: long melting, not

mixing and calculated value from individual pyrolysis

Figure 11 TIC area percent and ratio of TIC Area percent between experimental

and calculated value for products derived from cellulose in co-pyrolysis of PE and

cellulose.

0.00

0.00

0.01

0.10

1.00

10.00

100.00

0

1

2

3

4

TIC

Are

a %

Ratio o

f T

IC A

rea p

erc

ent

(Exp./

Calc

.)

PE/Cellulose Long Melting

PE/Cellulose Not Mixed

30

4.3 Paper II: Interactions between plastics and cellulose

Interaction between different plastics (PS, PP and PE/PET) and cellulose were

studied using Py-GC×GC/MS. The pyrolysis temperature was set to 600 °C to

ensure the pyrolysis of all feedstocks. As some previous works have already

revealed interactions between PE [68] or PP [69] with cellulose, the scenarios to be

investigated are mixture of 1) PS and cellulose, 2) PP, PS and cellulose, and 3)

PE/PET, PP, PS and cellulose. The samples were in a powder form with particle

size smaller than 63 µm. The feedstocks for different scenarios were pre-mixed at

different ratios, e.g., PS to cellulose ratios were 70:30, 50:50, 30:70. The sample

amount used was 0.3 ±0.005 mg.

The reported weight-normalized TIC area is the TIC area divided by sample weight.

The theoretical or the calculated value is the weighted average of the weight-

normalized TIC area according to the mixing ratios of the feedstock.

The limitation of Py-GC/MS experiments is that only a specific range of products

could be observed, i.e., compounds within the range of ~C5-C24.

4.3.1 Results and discussion

4.3.1.1 Co-pyrolysis of PS and cellulose

The main products of PS pyrolysis are styrene and styrene oligomers, while that of

cellulose pyrolysis are anhydrosugars (mainly levoglucosan), furans and other

oxygenated compounds. When PS and cellulose were co-pyrolyzed, the main

products from both feedstocks could still be observed. However, the quantities

were different. Figure 12 and Figure 13 show the ratios of experimental values to

the calculated values of some products from co-pyrolysis of PS and cellulose. The

yields of single-ring aromatics increased with a large increase of ethylbenzene, while

the products from cellulose pyrolysis did not change much except for an unknown

sugars compound. The higher yields of benzene, toluene, ethylbenzene,

methylstyrene as seen in Figure 12 could be explained by a promotion of hydrogen

transfer between pyrolysis products. Jakab et al. [70] investigated the pyrolysis of

PS and charcoal mixture in thermogravimetric-mass spectroscopy system (TG/MS)

and Py-GC/MS. They suggested that styrene monomers might be hydrogenated by

the charcoal which contained ~1% hydrogen. Investigation of char properties

should be performed in the future to clarify this suggestion.

31

Figure 12 Ratios of the experimental to the calculated weight-normalized TIC areas of some aromatic compounds from PS and cellulose co-pyrolysis at 600°C (Figure

4 in Paper II).

Figure 13 Ratios of the experimental to the calculated weight-normalized TIC areas

of some anhydrosugars, furans, pyrans, aldehydes and ketones from PS and

cellulose co-pyrolysis at 600°C (Figure S4 in Paper II).

32

4.3.1.1 Co-pyrolysis of PP, PS and cellulose

When PP, PS and cellulose were co-pyrolyzed, the formation of new products were

observed as shown in Figure 14. The new alkylated benzenes are likely formed from

the recombination of hydrocarbon radicals generated from PS and PP pyrolysis.

The formation of (3-methyl-3-butenyl)benzene during mixed plastics pyrolysis has

been reported before in [71]. Moreover, interactions between plastics, e.g., PP and

PS, were previously reported to be due to intermolecular hydrogen abstraction [72].

The comparison between the experimental and the calculated weight -normalized

TIC area of groups of compounds are shown in Figure 15. The yields of single ring

aromatic hydrocarbons largely increase in the co-pyrolysis cases. This increase in

yield affects most of the single ring aromatic components as shown in Figure 16. A

comparison between the PP/PS/Cellulose ratios of 60:10:30 and 40:10:50 shows

that when higher amounts of cellulose was present in the mixture, more

ethylbenzene was formed. This implies that the promotion of ethylbenzene is due

to interaction between PS and cellulose. For other single ring aromatic products,

the trends of the area ratios are not as clear as ethylbenzene, this implies that not

only cellulose but also PP contributes to the formation of these products. It is likely

that these products are formed through radical interactions, e.g., hydrogen transfer,

disproportionation and termination by the mixed radicals generated during

pyrolysis. The main pyrolysis products from cellulose pyrolysis, anhydrosugars,

were not affected by co-pyrolysis. As furans, e.g., furfural and hydroxymethyl

furfural are proposed to form via dehydration of 3-deoxy-glucose intermediate [50],

the small decrease of furans observed implies a suppression of this formation

pathway. Small decreases in the yield of hydrocarbons (alkanes, alkenes, dialkenes)

derived from PP pyrolysis were observed. This can be partly due to the interaction

with PS pyrolysis products to form new alkylated products discussed earlier. The

formation of aliphatic alcohols as observed by Ojha et al. [69] in the co-pyrolysis

of PP and cellulose was not observed in our results which could be due to the

difference in pyrolysis system.

33

Figure 14 2D-Pyrograms showing new compounds formed in the co-pyrolysis of

PP, PS and cellulose (Figure 5 in Paper II).

Figure 15 Weight-normalized TIC area of different groups of compounds emitted from co-pyrolysis of PP, PS and cellulose at three different mixing ratios (60:10:30,

40:10:50 and 20:10:70) at 600°C, along with calculated values based on single

component pyrolysis (Figure 6 in Paper II).

34

Figure 16 Ratios of the experimental to the calculated weight-normalized TIC areas

of some aromatic compounds from PP, PS and cellulose co-pyrolysis at 600°C

(Figure 7 in Paper II).

4.3.1.2 Co-pyrolysis of PE/PET, PP, PS and cellulose

The PE used in this work was contaminated with PET from the beginning. The

amount of PET was estimated to be ~70 wt%. Figure 17 shows the weight-

normalized TIC area of different compound groups. An increase in the yield of

single ring aromatic can still be observed like in the previous cases. The addition of

PE/PET suppressed the formation of ethylbenzene as shown in Figure 18. It is

likely that the interaction was due to PET and not PE because PE was found to

promote ethylbenzene formation by hydrogen donation to PS pyrolysis products

[73]. Figure 19 shows the ratios of the experimental to the calculated TIC areas of

compounds originally derived from PE/PET pyrolysis. It was observed that the

yields of unsaturated products were lower, while the yield of saturated products

were higher as shown in Table 7. This indicates that PET pyrolysis products are

more competitive in hydrogen abstraction from the pyrolysis products of other

feedstocks.

No new compound was observed except for the new alkylated benzenes already

observed in the PP/PS/Cellulose co-pyrolysis case. No significant change was

observed in the yields of anhydrosugars. Moreover, only small changes were

observed in the yields of aliphatic hydrocarbons (alkanes, alkenes, dialkenes) which

1E+0

1E+1

1E+2

1E+3

1E+4

1E+5

1E+6

1E+7

1E+8

1E+9

1E+10

0.0

1.0

2.0

3.0

4.0

5.0

6.0

7.0

8.0

9.0

10.0

Weig

ht-

norm

ali

zed

TIC

A

rea

Rati

o o

f W

eig

ht-

norm

ali

zed

TIC

A

rea E

xp./

Calc

.

PP/PS/C (60:10:30)

PP/PS/C (40:10:50)

PP/PS/C (20:10:70)

PP/PS/C (40:10:50) Area Exp.

PP/PS/C (40:10:50) Area Calc.

35

are likely due to interaction with PS pyrolysis products. From all these observations,

no indication was found that supported the hypothesis that the plastics donate

hydrogen to cellulose pyrolysis products to a significant extent during co-pyrolysis.

However, with different pyrolysis conditions, e.g., slow heating rate [74], different

phenomena can occur.

Figure 17 Weight-normalized TIC area of different groups of compounds emitted

from co-pyrolysis of PE/PET, PP, PS and cellulose different mixing ratio

(30:30:10:30, 30:10:10:50 and 8:8:8:76) at 600°C (Figure 10 in Paper II).

36

Figure 18 Ratios of the experimental to the calculated weight-normalized TIC areas

of some aromatic compounds from PE/PET, PP, PS and cellulose co-pyrolysis at

600°C (Figure 11 in Paper II).

Figure 19 Ratios of the experimental to the calculated weight-normalized TIC areas

of some oxygenated compounds from PE/PET, PP, PS and cellulose co-pyrolysis

at 600°C (Figure 12 in Paper II).

1E+0

1E+1

1E+2

1E+3

1E+4

1E+5

1E+6

1E+7

1E+8

1E+9

1E+10

0

1

2

3

4

5

6

7

8

9

10

Weig

ht-

norm

ali

zed

TIC

A

rea

Rati

o o

f W

eig

ht-

norm

ali

zed

TIC

A

rea

Exp./

Calc

. PE/PP/PS/C (30:30:10:30)

PE/PP/PS/C (30:10:10:50)

PE/PP/PS/C (8:8:8:76)

PE/PP/PS/C (30:10:10:50) Area Exp.

PE/PP/PS/C (30:10:10:50) Area Calc.

1E+0

1E+1

1E+2

1E+3

1E+4

1E+5

1E+6

1E+7

1E+8

1E+9

1E+10

0

1

2

3

4

5

6

7

8

9

10

Weig

ht-

norm

ali

zed

TIC

A

rea

Rati

o o

f W

eig

ht-

norm

ali

zed

TIC

A

rea

PE/PP/PS/C (30:30:10:30)

PE/PP/PS/C (30:10:10:50)

PE/PP/PS/C (8:8:8:76)

PE/PP/PS/C (30:10:10:50) Area Exp.

PE/PP/PS/C (30:10:10:50) Area Calc.

37

Table 7 Changes in the products derived from PET pyrolysis due to the co-pyrolysis

with other feedstocks. (To be considered together with Figure 19)

Products with decreased yield Products with increased yield

4-(Vinyloxycarbonyl) benzoic acid

terephthalic acid

vinyl benzoate

ethyl benzoate

(benzoic acid, ethyl ester)

benzoic acid

1,2-Ethanediol, dibenzoate

1,2-Ethanediol, monobenzoate

4.4 Paper III: H, OH and H2O formation during the early stages of cellulose pyrolysis

From the co-pyrolysis studies of PS and cellulose in paper II, larger yields of

products from hydrogen transfer reactions, e.g., ethylbenzene, toluene, were

observed. Moreover, in the study of co-pyrolysis of PP and cellulose [69], the

formation of aliphatic alcohols was observed which was likely involve the transfer

of OH radicals. It is questionable whether cellulose can act as the source of H and

OH radicals during pyrolysis or whether these products are the result of

hydrogenation by cellulose char [70] or whether it was due to the interaction

between radicals from plastics pyrolysis and water [75]. In this work, first-principle

calculations and Py-GC/MS experiments were performed to investigate the

formation of H, OH and H2O during the thermal decomposition of cellulose.

38

As the author of the thesis was responsible for the experimental section, the results

from the experiments are presented here in detail. The connection to the results

from the calculation and the overall conclusion will also be presented.

Microcrystalline cellulose powder was pyrolyzed in a platinum filament pyrolyzer,

Pyrola2000, which was connected to a GC/MS. The multiple-temperature-step

pyrolysis was performed on the same sample to investigate the evolution of

products, especially water formation.

4.4.1 Results and discussion

Figure 21 shows the total ion chromatogram of the cellulose pyrolysis at different

temperature. The main products from cellulose pyrolysis started to appear at the

temperature of 350 °C. Before that temperature, only the peak at 1.7 min could be

observed. This peak is a mixture of gases and light compounds, e.g., CO, CO2,

water. The identification of each compound in this peak is done by selected mass

spectrum. The abundance of water, CO, CO2 in this peak is shown in Figure 22 as

a function of temperature. Water did not appear before 210 °C. This indicates that

it does not have an origin from the moisture in the sample. The abundance of water

was high at 280 °C, became slightly lower at 320 °C, and increased again after 350

°C. This suggests that the water formation reaction at 280 °C is different from that

at 350 °C.

From the DFT calculations of bond dissociation energies of H and OH radicals at

different positions in the cellulose structure, it was found that the formation of

single radicals (H•, HO•) from any available positions required higher reaction

energy as compared to the generation of two radicals with subsequent formation

of molecular products. The most favorable one being the generation of H from C2

position and OH from C3 position to form a water molecule (see Figure 20 for the

numbering position of carbon atoms). This is due to the formation of a double

bond between C2 and C3 that stabilizes the cellulose product. Furthermore, from

the calculation of Gibbs free energy of this reaction, the temperature at which this

reaction become spontaneous was calculated to be 282 °C, while the temperature

for water desorption from cellulose was estimated to be -153 °C for cellulose with

diradicals at C2 and C3. The generated water at 282 °C is then likely desorbed

spontaneously. This agrees well with the experimental result where the large

amount of water was first released at 280 °C.

As a conclusion, it is not likely that H and OH radicals are formed individually

during cellulose pyrolysis, but they are rather formed together and recombine to

form water. This supports the study that suggested that the interaction between

39

plastics and cellulose was caused by the reaction between radicals from plastics and

water [75]. More investigation should be done in the future where both plastics and

cellulose are present in the calculation.

Figure 20 The structure and carbon numbering of two unit cells of cellulose IβA used in the calculations.

Figure 21 Total Ion Chromatogram of products from cellulose pyrolysis at different

temperature steps.

0 2 4 6 8 10 12 14 16 18 20 22 24

Abundan

ce (ar

bitar

y u

nits)

Retention time (min)

210°C

280°C

320°C

350°C

400°C

170°C

m/z scanningrange:

15 - 250

m/z scanning range: 33 - 550

250°C

40

Figure 22 Abundances of water (m/z 18), CO2 (m/z 44) and CO (m/z 28) in the

first GC peak (RT 1.7 min) at different temperatures. Two tests were performed and denoted as sample A and B. It should be noted that m/z = 28 can have origin

in other species besides CO because small hydrocarbons with m/z of 29-30 were

detected at 350 and 400 °C.

0

1

2

3

4

5

6

7

8

9

150 250 350

Abundan

ce (

×1

07/m

g)

Pyrolysis temperature (°C)

H₂O Sample A

H₂O Sample B

CO₂ Sample A

CO₂ Sample B

CO Sample A

CO Sample B

41

Chapter 5

5 Stepwise pyrolysis of mixed plastics and paper

5.1 Introduction

In the previous chapter, we have investigated the fundamental aspects of plastics

(PE, PP, PS, PET) and paper co-pyrolysis. The microscale pyrolysis has shown that

co-pyrolysis does not affect the main volatile pyrolysis products form each

feedstock. The interactions between plastics and cellulose was small and there was

no indication of hydrogen transfer from plastics to cellulose pyrolysis products.

Further literature reviews on lab-scale [64, 76, 77] and pilot scale [78] showed

similar findings, i.e., small interactions occur but the main products from pyrolysis

of each feedstock are not much affected. Catalytic co-pyrolysis was found to

increase the yield of aromatic hydrocarbons, but not all types of plastics results in

this synergistic effect [67, 79, 80].

The products from biomass and plastics pyrolysis are significantly different.

Pyrolysis of mixed plastics waste usually produces high hydrocarbon content [19]

and can be upgraded to selected products [19, 81]. On the other hand,

biomass/paper pyrolysis produces highly oxygenated liquid . The research on

upgradation of this liquid is on-going with the consistent tradeoff between the yield

and the quality of the products [35]. Since the co-pyrolysis does not provide the

desired significant synergistic effect, it might be more beneficial to separate the two

products from each other, thus allowing different treatment and upgradation

process for pyrolysis products of each feedstock.

Some options for separating products could be to add pre- or post- processes

and/or to modify the pyrolysis process. For the pre-pyrolysis process, we have

already discussed the sorting processes that some materials are difficult to be sorted.

For the post-pyrolysis process, some options are fractional condensation,

distillation, and solvent extraction, e.g., washing the products with water. For

fractional condensation and distillation, some oxygenated and hydrocarbon

42

products have similar boiling points, e.g., levoglucosan (384 °C) and eicosane (wax,

343 °C); therefore, they can be difficult to separate. For solvent extraction, the

products are diluted in solvents and a drying process is required to recover the

products and solvents. Nevertheless, according to the author’s knowledge, not

many works have investigated this option for pyrolysis oil from mixed feedstocks

and it can be interesting to investigate further. For the modification of pyrolysis

process, stepwise pyrolysis is an interesting option to produce different products at

different pyrolysis temperature. This can be beneficial as less post-processing is

required and the upgradation of the products can be performed at the same time if

catalyst is applied. However, the possibility of applying stepwise pyrolysis process

depends very much on the feedstock properties.

Stepwise pyrolysis of mixed plastics has been applied in the past for the

dechlorination of PVC and for producing different products at different

temperature [73]. This is possible due to the different pyrolysis temperature of

different plastics. For cellulose, it is observed that its pyrolysis temperature is lower

than that of most plastics [82] and only few works have investigated stepwise

pyrolysis of mixed plastics and paper [83, 84] (many works on single step pyrolysis,

e.g., [85, 86, 87]). Therefore, we investigate this issue in Paper IV. Initially, as a

proof of concept, mixtures of neat PS, PE and cellulose were used as feedstocks.

Later, real waste samples, the paper rejects, were subjected to stepwise pyrolysis.

In paper V, we continue the investigation on the stepwise pyrolysis when PVC is

presented in the feedstock. The focus was given to the first temperature step as the

products from cellulose pyrolysis would appear together with HCl from PVC

dechlorination. Calcium oxide was proposed to be used for simultaneous HCl

adsorption [88, 89] and upgradation of biooil from cellulose pyrolysis [90]. The

performance of CaO on these duties was investigated.

5.2 Paper IV: Stepwise pyrolysis of mixed plastics and cellulose

The feedstocks were pyrolyzed in a one-stage fixed bed reactor. The reactor was

heated to a fixed temperature before the sample basket was introduced. After

pyrolysis at the first temperature, the basket was restored to a cooling zone. The

basket was weighed, and the condensable products were collected for analysis.

Pyrolysis at a second temperature was performed on the solid residue from the first

pyrolysis step following the same pyrolysis procedure.

43

5.2.1 Results and discussion

Since the thermal decomposition temperature of PS is closed to that of cellulose,

mixtures of PS and cellulose were used in the investigation of the first step

temperature. To pyrolyze PS and cellulose separately, the temperature chosen for

the first step should not be higher than the thermal decomposition of PS which is

350-450 °C [82, 91]. The thermal decomposition temperature of cellulose is 300-

400 °C [82, 91]. Therefore, the temperatures chosen for the investigation are 300

and 350 °C. The temperature of the second step was fixed to 500 °C which is high

enough to pyrolyze most plastics [82]. Figure 23a shows the mass balance of single

step and stepwise pyrolysis of PS and cellulose mixture. Stepwise pyrolysis

produced slightly more char than single step pyrolysis. This is due partly to the

heating rate. When cellulose is subject to higher heating rate at 500 °C as compared

to that at 300 or 350 °C, less char is produced [50]. The first step temperature of

300 °C results in incomplete pyrolysis in a given time (32 min). This can be

observed from the water produced in the second pyrolysis step (also presented in

Table 8) and from the higher amount of cellulose pyrolysis products in the second

pyrolysis step as compared to when the temperature of the first step was 350 °C (as

shown in Figure 24a). Also shown in Figure 24a, when the temperature of the first

step was 350 °C, some amount of multiple-ring aromatic hydrocarbons from PS

pyrolysis was also produced in the first step. This indicate that the temperature of

the first step should not be increased further as more pyrolysis products from PS

will be produced in the first step. The results also indicate that the temperature and

pyrolysis time of the first step is not yet optimized. Nevertheless, the results showed

that the main fraction of pyrolysis products from cellulose and PS can be well

separated into the first and the second pyrolysis step. Table 8 shows the total acid

number (TAN) and water content of the condensable products. The TANs and

water contents of the products from the first pyrolysis step are significantly higher

than those from the second pyrolysis step, supporting the previous discussion.

The first step temperature of 350 °C was chosen for further investigation since

better separation was achieved as compared to that at 300 °C. The main pyrolysis

products from PE is wax or heavy hydrocarbons which cannot be used directly and

should be upgraded into lighter hydrocarbons. One process parameter affecting the

extent of cracking of hydrocarbon chains is the vapor residence time in the heating

zone [92, 93]. The longer the vapor residence time, the lighter the hydrocarbon

products. By adding PE to the feedstock, we investigate the vapor residence time

of the second pyrolysis step by lowering the flowrate of the carrier gas. This resulted

in a longer vapor residence time being approximately 42 s as compare to the regular

case of approximately 5 s. The long vapor residence time case is denoted as long. As

44

shown in Figure 23b, less wax was produced in the long case. However, the liquid

yield was also reduced.

The stepwise pyrolysis with the temperature steps of 350/500 °C and long vapor

residence time was applied to the real waste samples. The mass balances are shown

in Figure 23c. Non-pelletized paper reject contained more ash/metal content thus

producing more char/solid residue as compared to pelletized paper reject. The

compositions of the condensable products are shown in Figure 24c. The

condensable products from the first pyrolysis step consist of large oxygenated

products, while the products from the second pyrolysis step consist of large fraction

of hydrocarbons. Moreover, the total acid number shown of the condensable

products from the first step was much higher than that from the second step as

shown in Figure 24c.

From the all the results, it is shown that stepwise pyrolysis of mixed plastics and

paper can products two separated condensable products streams: one from the first

pyrolysis step containing high acidic and oxygenated compounds, and another one

from the second pyrolysis step containing mainly hydrocarbons. The temperature

of the first step is not yet optimized but should be between 300-350 °C. Further

investigation is needed on the upgradation of the products from both steps and on

the possibility of scaling up the process.

45

Figure 23 Mass balance of single step and stepwise pyrolysis of a) cellulose and PS

mixture (1:1), b) cellulose, PS and PE mixture (1:1:1), c) paper reject samples. The label “long” indicates long vapor residence time in the second pyrolysis step. Diff

indicates the mass fraction which is not accounted for, mainly undetected C4–C7

compounds.

11% 17% 16%10%

8% 8%

60%

7% 13%

1.9%

0.1%

46% 41%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Cell:PS

Single step

500°C

Cell:PS

300/500°C

Cell:PS

350/500°C

Wt

%

a) Cellulose:PS Diff Step 2

Diff Step 1

Gas Step 2

Gas Step 1

Wax Step 2

Liquid Step 2

Water Step 2

Liquid Step 1

Water Step 1

Final Char

4% 9% 11%

57%

10% 12%

21%

39% 27%

18%

13%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Cell:PS:PE

Single step

500 °C

Cell:PS:PE

350/500 °C

Cell:PS:PE

350/500 °C

long

Wt

%

b) Cellulose:PS:PE Diff Step 2

Diff Step 1

Gas Step 2

Gas Step 1

Wax Step 2

Liquid Step 2

Wax Step 1

Liquid Step 1

Water Step 1

Final Char

27%42%

7%

15%12%

6%13%

19%15%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Pellets

350/500°C long

Non-pellet

350/500°C long

Wt

%

c) Real waste samples Diff Step 2

Diff Step 1

Gas Step 2

Gas Step 1

Wax Step 2

Liquid Step 2

Wax Step 1

Liquid Step 1

Water Step 1

Final Char

46

Figure 24 Compositions of liquid products from single step and stepwise pyrolysis analyzed by GC/MS of a) mixture of cellulose and PS, b) mixture of cellulose, PS

and PE, c) paper reject samples.

30.0 21.2 21.7

23.1

1.8

21.7

5.6

16.5

2.2

1.3 1.3

2.0

2.1

0.3

4.4

0.1

2.3

1.2

1.6

1.6

1.1

0

10

20

30

40

50

60

Cell:PS

Single step

500°C

Cell:PS

300/500 °C

Step 1

Cell:PS

300/500 °C

Step 2

Cell:PS

350/500 °C

Step 1

Cell:PS

350/500 °C

Step 2

Weig

hte

d A

rea P

erc

ent (a

rbitary

unit)

Unknown

Other oxygenated compounds

Furan & Pyrans

Anhydrosugars

PAHs

Multiple-ring Aromatic hydrocarbons

Single-ring Aromatic hydrocarbons

a)

5.9 4.9 4.7

21.6 17.6

12.7

15.9

3.5

12.8

6.6

3.1

3.0

0.7

1.4

6.8

1.8

1.6

1.2

0

10

20

30

40

50

60

Cell:PS:PE

Single step

Cell:PS:PE

350/500 °C

Step 1

Cell:PS:PE

350/500 °C

short

Step 2

Cell:PS:PE

350/500 °C

long

Step 2

Weig

hte

d A

rea P

erc

ent (a

rbitary

unit)

Unknown

Other oxygenated compounds

Furan&Pyrans

Anhydrosugars

PAHs

Multiple-ring Aromatic hydrocarbons

Single-ring Aromatic hydrocarbons

Aliphatic hydrocarbons

b)

-2.0

7.8

0.7

0.12

0.4

1.1

1.85 0.61

4.88 1.0

3.89

0.8 6.19

1.4

6.67

2.7

1.53

0

2

4

6

8

10

12

14

16

18

Pellets

350/500 °C

Step 1

Pellets

350/500 °C

long

Step 2

Non-pellet

350/500 °C

Step 1

Non-pellet

350/500 °C

long

Step 2

Weig

hte

d A

rea P

erc

ent (a

rbitary

unit)

Phthalates

Unknown

Other oxygenated compounds

Furan&Pyrans

Anhydrosugars

PAHs

Multiple-ring Aromatic hydrocarbons

Single-ring Aromatic hydrocarbons

Aliphatic hydrocarbons

c)

47

Table 8 Total acid number (TAN, mg KOH/g) and water content (wt%) of

condensable products

Feedstocks Single step

pyrolysis

500 °C

Stepwise pyrolysis

1st step 2nd step

Cellulose:PS

(1:1)

TAN

Water content

18.7

17.4

300 °C 350 °C

36.5 ± 0.0 43.2 ± 0.2

66.7 ± 4.7 50.1 ± 1.5

300/500 °C 350/500 °C

3.0 ± 0.6 3.3 ± 0.5

5.0 ± 1.1 0.15 ± 0.06

Cellulose:PS:PE

(1:1:1)

TAN

Water content

32.0

18.7

350 °C 350 °C

43.5 ± 2.7 49.6

46.8 ± 0.8 45.5

short long

vapor residence time

19.5 ± 3.3 19.6

<0.05 0.3

Paper reject

samples

TAN

Water content

Pelletized Non-pelletized

215.9 NA

58.0 NA

Pelletized Non-pelletized

29.6 10.8

1.7 0.05

48

5.3 Paper V: Upgradation of products from the first step of stepwise pyrolysis

In this study, a mixture of cellulose and 5 wt% PVC was used as feedstock to

investigate the upgradation of the products from the first pyrolysis step. The

temperature used from stepwise pyrolysis are 350 °C for the first step and 500 °C

for the second step. As PVC undergoes dehydrochlorination at the same

temperature with cellulose pyrolysis in the first pyrolysis step, liquid products then

become a mixture of HCl and bio-oil. Calcium oxide (CaO) was then chosen as a

sorbent/catalyst based on its ability to adsorb HCl from PVC dechlorination [88,

89] and to upgrade bio-oil [90].

The feedstocks were pyrolyzed in a two-stage fixed bed reactor. Calcium oxide was

placed in the second reactor. The different ratios of feedstock to CaO (R) used

were 1:0.2, 1:0.4 and 1:1. The different temperature applied on the CaO reactor

were 300, 350, 400 and 600 °C.

5.3.1 Results and discussion

Figure 25 shows the yields of char, liquid, gases, and weight increase on CaO (ΔW)

at different CaO temperature and R. As R increases, ΔW increase and the liquid

yields decrease. This implies the higher formation of coke on the spent CaO and

higher catalytic activity on the products. At a given R, the liquid yields were highest

at 400 °C at the same temperature as when ΔW were lowest.

The yields of water, organic liquids and chloride ion are shown in Figure 26. The

yield of organic liquids was calculated from the total liquid yield subtracted by the

yields of water and chlorine. From the figure, high water and chlorine yields were

observed at 400 °C. The R of 1:1 was able to adsorb all Cl at 300 and 350 °C. At

600 °C, some amount of Cl is also observed in the liquid. The yields of organic

liquids were quite low in all cases (~10%). This is due to the catalytic activity of

CaO which results in the formation of coke and water.

XRD analyses of the spent CaO are shown in Figure 27. XRD pattern of calcium

chloride was not observed even when the feedstock used was neat PVC. This

indicates that the products of CaO and HCl reaction might not be CaCl2 or that

the amount of too small to be observed. When mixed feedstock was employed,

CO2 adsorption by CaO was observed in all cases. Water also react with CaO to

form Ca(OH)2. However, the pattern of Ca(OH)2 disappeared at 400 and 600 °C

due to the decomposition of Ca(OH)2. The correspondence of Ca(OH)2

decomposition and the release of HCl form the CaO at 400 °C implies that the two

49

reactions are related. Further literature reviews show that the presence of water

vapor can cause the reemission of HCl according to reaction (1) which occur at

around 400 °C [94]. Another explanation is the thermal decomposition of Ca(OH)2

which start at 400 °C [95].

CaCl2 (s) + H2O (g) → Ca(OH)Cl (s) + HCl (g) at 410-740 °C (1)

Ca(OH)Cl (s) → CaO (s) + HCl (g) at 640-740 °C (2)

Ca(OH)2 (s) → CaO (s) + H2O (g) start at 400 °C (3)

For the catalytic effect of CaO on the liquid products, Figure 28 shows some of the

main compounds found in the liquid. The weighted area percent was calculated

from the area percent multiplied by the yield of organic compounds. The main

products from non-catalytic pyrolysis of cellulose was levoglucosan (β-D-

Glucopyranose, 1,6-anhydro-) which had the weighted area percent of 4.0 (not

shown in any figures). However, when CaO was applied at different temperature,

the main products shift toward cyclopentenone as shown in Figure 28d-e. The yield

of 2-cyclopenten-1-one was highest at 350 °C indicating higher catalytic activity as

compared to other temperatures. On the other hand, when the temperature

increases to higher than 400 °C, the productions of polycyclic aromatic

hydrocarbons (PAHs) and phenol become dominant.

Another important compound is the acetic acid shown in Figure 28i. When R =

1:1, a complete adsorption of acetic acid was achieved at 300 and 350 °C, while the

yield of acetic acid was high at 400 °C. The adsorption of acetic acid helps reduce

the acidity of the liquid products which is desirable for fuel application.

Overall, the suitable CaO temperature for Cl adsorption and bio-oil reforming is

below 350 °C. The feedstock to CaO ratio of 1:1 have shown to be able to adsorb

all HCl and acetic acid. Nevertheless, it should be further optimized. The CaO

temperature of higher than 400 °C was found to cause the desorption of HCl which

can be observed from the higher Cl ion in the liquid product. This was explained

by the reaction of CaCl2 with water at such temperature.

50

Figure 25 Yield of char, liquid, gases, and weight increase on CaO (ΔW) from the

ex-situ catalytic pyrolysis of mixture of cellulose and 5 wt% PVC at 350 °C at

different catalyst (CaO) temperature and feedstock to catalyst ratio (R) .

Figure 26 Yield of water, organic liquid, and chloride ion found in the liquid

products from the ex-situ catalytic pyrolysis of mixture of cellulose and 5 wt% PVC

at 350 °C.

0%

10%

20%

30%

40%

50%

250 450 650

wt%

CaO temperature (°C)

Char

1:0.2

1:0.4

1:10%

10%

20%

30%

40%

50%

250 450 650

wt%

CaO temperature (°C)

ΔW

1:0.2

1:0.4

1:1

0%

10%

20%

30%

40%

50%

250 450 650

wt%

CaO temperature (°C)

Liquid

1:0.2

1:0.4

1:10%

5%

10%

15%

250 450 650

wt%

CaO temperature (°C)

Gases

1:0.2

1:0.4

1:1

0.0%

0.5%

1.0%

1.5%

2.0%

0%

10%

20%

30%

200 300 400 500 600 700

wt%

Temperature (°C)

1:0.4 Water

1:1 Water

1:0.4 Organic liquid

1:1 Organic liquid

1:0.4 Cl ion

1:1 Cl ion

51

Figure 27 XRD spectra for fresh and spent CaO. Each case is denoted by

feedstock, R, and CaO temperature.

52

Figure 28 Trends for some organic compounds in the liquid products from ex-situ

catalytic pyrolysis of mixture of cellulose and 5 wt% PVC at 350 °C. The CaO

temperature and R were varied.

0.00

0.10

0.20

0.30

0.40

0.50

0.60

0.70

250 350 450 550 650

Wei

ghte

d a

rea

per

cent

(arb

itar

y u

nit

)

CaO temperature (°C)

1H-Inden-1-one, 2,3-dihydro-

1:0.4

1:1

a)

0.00

0.05

0.10

0.15

0.20

0.25

250 350 450 550 650

Wei

ghte

d a

rea

per

cent

(arb

itar

y u

nit

)

CaO temperature (°C)

Naphthalene

1:0.4

1:1

b)

0.00

0.02

0.04

0.06

0.08

0.10

0.12

0.14

250 350 450 550 650

Wei

ghte

d a

rea

per

cent

(arb

itar

y u

nit

)

CaO temperature (°C)

Naphthalene, 2-methyl-

1:0.4

1:1

c)

0.00

0.10

0.20

0.30

0.40

0.50

0.60

0.70

250 350 450 550 650

Wei

ghte

d a

rea

per

cen

t (a

rbit

ary

un

it)

CaO temperature (°C)

2-Cyclopenten-1-one

1:0.4

1:1

d)

0.00

0.05

0.10

0.15

0.20

0.25

250 350 450 550 650

Wei

ghte

d a

rea

per

cent

(arb

itar

y u

nit

)

CaO temperature (°C)

2-Cyclopenten-1-one, 3-methyl-

1:0.4

1:1

e)

0.00

0.20

0.40

0.60

0.80

1.00

250 350 450 550 650

Wei

ghte

d a

rea

per

cen

t (a

rbit

ary

unit

)

CaO temperature (°C)

Phenol

1:0.4

1:1

f)

0.00

0.10

0.20

0.30

0.40

0.50

0.60

0.70

250 350 450 550 650

Wei

ghte

d a

rea

per

cen

t (a

rbit

ary

un

it)

CaO temperature (°C)

1,4:3,6-Dianhydro-α-d-glucopyranose

1:0.4

1:1

m)

0.00

0.02

0.04

0.06

0.08

0.10

0.12

0.14

250 350 450 550 650

Wei

ghte

d a

rea

per

cent

(arb

itar

y u

nit

)

CaO temperature (°C)

Acetic acid

1:0.4

1:1

i)

53

Figure 28 Cont.

0.00

0.20

0.40

0.60

0.80

1.00

250 350 450 550 650

Wei

ghte

d a

rea

per

cent

(arb

itar

y u

nit

)

CaO temperature (°C)

Furfural

1:0.4

1:1

g)

0.00

0.05

0.10

0.15

0.20

0.25

250 350 450 550 650

Wei

ghte

d a

rea

per

cent

(arb

itar

y u

nit

)

CaO temperature (°C)

5-Hydroxymethylfurfural

1:0.4

1:1

h)

0.00

0.20

0.40

0.60

0.80

1.00

250 350 450 550 650

Wei

ghte

d a

rea

per

cen

t (a

rbit

ary

un

it)

CaO temperature (°C)

Levoglucosenone

1:0.4

1:1

j)

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

250 350 450 550 650

Wei

ghte

d a

rea

per

cent

(arb

itar

y u

nit

)

CaO temperature (°C)

β-D-Glucopyranose, 1,6-anhydro-

1:0.4

1:1

k)

54

55

Chapter 6

6 Conclusions and future work

6.1 Conclusions

This thesis investigated how to produce fuels or other chemicals from mixed

plastics and paper by pyrolysis. Both fundamental investigations on interactions

between the mixed feedstocks and the development of pyrolysis process were

performed.

For the fundamental investigation, three works have been conducted to investigate

the effect of mixing methods, the interactions between mixed feedstocks and the

formation of H, OH radicals from cellulose pyrolysis. The conclusions are as

follow:

• The mixing method has only a slight effect on the main volatile products

from co-pyrolysis of PE and cellulose.

• The interactions between mixed plastics (PE, PP, PS, PET) and cellulose

was found to not affect the main volatile products from each feedstock.

Small interactions occur causing hydrogenation of PS and PET pyrolysis

products which can be observed from increased yields of compounds with

less unsaturated bonds, e.g., ethylbenzene.

• A further study then investigated the possibility of H, OH radicals’

formation during cellulose pyrolysis. It was found that these radicals could

form during the early stage of cellulose pyrolysis. However, they are likely

formed together and recombine to form water. The calculations suggest

the formation of water at around 280 °C which agrees with experimental

findings.

As the main products of each feedstock were not affected by co-pyrolysis, it might

be more beneficial to separate the products from pyrolysis of plastics and paper

because the hydrocarbons from plastics pyrolysis and oxygenated products from

paper pyrolysis are different in nature and require different upgradation methods.

56

Stepwise pyrolysis process was proposed and tested with simulated mixtures (PE,

PS, cellulose) and real plastics and paper mixtures.

• It was found that the temperature of the first pyrolysis step of 300-350 °C

and the temperature of the second step of 500 °C can be used to produce

two distinct products streams: one with high acidity and high content of

oxygenated compounds, and one with low acidity and high content of

hydrocarbons.

Further investigation was focused on the first pyrolysis step when PVC was present

in the feedstocks. Calcium oxide was proposed to be used as sorbent/catalyst to

adsorb HCl and to reform the products from cellulose pyrolysis.

• It was found that the suitable temperature for HCl adsorption and

reforming of pyrolysis products was below 350 °C. As when the

temperature was higher than 400 °C, the presence of water in the products

stream cause the reemission of HCl from the reacting CaO. The ratio of

feedstock to CaO was also important and required further optimization.

The ratio of 1:1 was found to be able to adsorb all the HCl from PVC

dehydrochlorination and the acetic acid from cellulose pyrolysis.

Overall, this work has built up a ground for future investigations and developments

of pyrolysis process to recover fuels and other chemicals from mixed plastics and

paper.

6.2 Future work

Some recommendations for future works are as follow:

1) On the experimental investigations of interactions between feedstocks.

Base on paper I and II, more investigation on the interactions between

feedstocks could be conducted in bench-scale reactors where mass

balances can be obtained and the characterizations of gas, liquid, and char

can be performed. Effects of heating rate or interactions between the

different phases, e.g. the melt plastics and cellulose as in [96], should be

further investigated. The feedstock interactions during catalytic co-

pyrolysis is also under current research interests [79].

2) On first-principle investigation of decomposition mechanisms and

interactions between feedstocks during pyrolysis. Based on paper III, more

first-principle investigation on mixed plastics and cellulose could be

57

performed by incorporating more feedstocks in the simulations. It is

important to develop a method for simulating multi-component systems as

this would help guiding and accelerating the experimental design and

process development which is currently done mainly by trial-and-error. An

incorporation of the effect of heating rate and pyrolysis temperature in the

study/calculation is also important.

3) On the process developments for pyrolysis of mixed feedstocks. Based on

paper IV and V, more works could be done in the future to understand and

further develop the stepwise pyrolysis of mixed plastics and paper, for

example, the characterization of wax and char, the application of different

catalysts to produce specific products, the effect of heteroatoms, e.g., S, N,

Cl, the fate of inorganics and its effect on the products and/or the methods

to recover them, and the recovery of metals from the pyrolysis char.

Reactor design is also important to find an efficient way to heat and

pyrolyze the feedstocks. Techno-economic assessment of the process and

life cycle analysis of the materials are also interesting topics to be further

performed.

From the circular economy aspect, more and more work should steer towards the

production of products that can be used to generate new plastics and/or chemicals.

Therefore, the higher purity and selectivity of the products are preferred. This can

be achieved by developing a refinery process capable of handling products from

pyrolysis of rejected fractions.

58

7 References

[1] J. R. Jambeck, R. Geyer, C. Wilcox, T. R. Siegler, M. Perryman, A. Andrady, R. Narayan and K. L. Law, "Plastic waste inputs from land into the ocean,"

Science, vol. 347, no. 6223, pp. 768-771, 2015.

[2] Ellen MacArthur Foundation, "The new plastics economy: rethinking the

future of plastics & catalysing action," Ellen MacArthur Foundation, Worldwide, 2017.

[3] J. Hopewell, R. Dvorak and E. Kosior, "Plastics recycling: challenges and

opportunities," Phil. Trans. R. Soc. B, vol. 364, pp. 2115-2126, 2009.

[4] N. Singh, D. Hui, R. Singh, I. Ahuja and L. Feo, "Recycling of plastic solid

waste: A state of art review and future applications," Composites Part B, vol. 115, p. 409e422, 2017.

[5] F. Welle, "Develop a food grade HDPE recycling," The Waste & Resources Action Programme, Banbury, 2005.

[6] A. Rahimi and J. M. Garcia, "Chemical recycling of waste plastics for new materials production," Nature Reviews Chemistry, vol. 1, p. 0046, 2017.

[7] J. Datta and P. Kopczyńska, "From polymer waste to potential main industrial

products: Actual state of recycling and recovering," Critical Reviews in Environmental Science and Technology, vol. 46, no. 10, pp. 905-946, 2016.

[8] F. L. Mantia, Handbook of Plastics Recycling, Shawbury: Rapra Technology Limited, 2002.

[9] P. Oblak, J. Gonzalez-Gutierrez, B. Zupančič, A. Aulova and I. Emri, "Processability and mechanical properties of extensively recycled high density

polyethylene," Polym. Degrad. Stab., vol. 114, pp. 133-145, 2015.

[10] E. Butler, G. Devlin and K. McDonnell, "Waste Polyolefins to Liquid Fuels via Pyrolysis: Review of Commercial State-of-the-Art and Recent Laboratory

Research," Waste Biomass Valor, vol. 2, p. 227–255, 2011.

[11] E. Royte, "Is burning plastic waste a good idea?," National Geographic, 12

March 2019. [Online]. Available: https://www.nationalgeographic.com/environment/2019/03/should-we-

burn-plastic-waste/. [Accessed 14 March 2019].

[12] P. Bajpai, "Chapter 2 Generation of Waste in Pulp and Paper Mills," in

Management of Pulp and Paper Mill Waste, Switzerland, Springer International

Publishing, 2015, pp. 9-17.

[13] H. Rylander and W. Wiqvist, "Determination of the fossil carbon content in

combustible municipal solid waste in Sweden Report U2012:05 by Avfall

59

Sverige," [Online]. Available: http://www.avfallsverige.se/. [Accessed 26 May 2017].

[14] A. Gallardo, M. Carlos, M. Bovea, F. J. Colomer and F. Albarran, "Analysis of refuse-derived fuel from the municipal solid waste reject fraction and its

compliance with quality standards," Journal of Cleaner Production, vol. 83, pp. 118-125, 2014.

[15] C. Montejo, C. Costa, P. Ramos and M. d. C. Márquez, "Analysis and comparison of municipal solid waste and reject fraction as fuels for

incineration plants," Applied Thermal Engineering, vol. 31, no. 13, pp. 2135-2140,

2011.

[16] I. Brás, M. E. Silva, G. Lobo, A. Cordeiro, M. Faria and L. T. d. Lemos,

"Refuse Derived Fuel from Municipal Solid Waste rejected fractions- a Case Study," Energy Procedia, vol. 120, pp. 349-356, 2017.

[17] J. Nithikul, O. P. Karthikeyan and C. Visvanathan, "Reject management from a Mechanical Biological Treatment plant in Bangkok, Thailand," Resources,

conservation and recycling, vol. 55, no. 4, pp. 417-422, 2011.

[18] T. Weerasak and S. Sanongraj, "Potential of Producing Refuse Derived Fuel (RDF) from Municipal Solid Waste at Rajamangala University of Technology

Isan Surin Campus," Applied Environmental Research, vol. 37, no. 2, pp. 85-91, 2015.

[19] J. Scheirs and W. Kaminsky, Feedstock Recycling and Pyrolysis of Waste Plastics: Converting Waste Plastics into Diesel and Other Fuels, Chichester:

John Wiley & Sons, Ltd, 2006.

[20] D. P. S. Jose Aguado, Feedstock Recycling of Plastic Wastes, The Royal

Society of Chemistry, 1999.

[21] I. Velghe, R. Carleer, J. Yperman and S. Schreurs, "Study of the pyrolysis of municipal solid waste for the production of valuable products," Journal of

Analytical and Applied Pyrolysis, vol. 92, pp. 366-375, 2011.

[22] D. Chen, L. Yin, H. Wang and P. He, "Reprint of: Pyrolysis technologies for

municipal solid waste: A review," Waste Management, vol. 37, pp. 116-136, 2015.

[23] S. Foster, "Domestic mixed plastics packaging waste management options:

Final project report WRAP," The Waste and Resources Action Programme (WRAP), UK, 2008.

[24] T. Schwarzböck, E. V. Eygen, H. Rechberger and J. Fellner, "Determining the amount of waste plastics in the feed of Austrian waste-to-energy

facilities," Waste Management & Research, vol. 35, no. 2, pp. 207-216, 2017.

[25] C. Areeprasert, J. Kaharn, B. Inseemeesak, P. Phasee, C. Khaobang, A.

Kuhavichanun, P. Theerarojprateep and W. Siwakosit, "A comparative study on characteristic of locally source-separated and mixed MSW in Bangkok with

60

possibility of material recycling," J Mater Cycles Waste Manag, vol. 20, pp. 302-313, 2018.

[26] United Nations, "The Sustainable Development Goals Report 2017," United nations, Worldwide, 2017.

[27] D. L. Pyle, "Heat transfer and kinetics in the low temperature pyrolysis of solids," Chemical Engineering Science, vol. 39, no. 1, pp. 147-158, 1984.

[28] C. Di Blasi, "Transition between regimes in the degradation of thermoplastic polymers," Polymer degradation and stability, vol. 64, pp. 359-367, 1999.

[29] R. Vinu and L. J. Broadbelt, "Unraveling reaction pathways and specifying

reaction kinetics for complex systems," Annual review of chemical and biomolecular engineering, vol. 3, pp. 29-54, 2012.

[30] H. Tavassoli, E. Peters and J. Kuipers, "Direct numerical simulation of fluid -particle heat transfer in fixed random arrays of non-spherical particles,"

Chemical engineering science, vol. 129, pp. 42-48, 2015.

[31] W.-C. Yang, Handbook of fluidization and fluid-particle systems, US: Marcel

Dekker, Inc., 2003.

[32] J. Mastral, C. Berrueco and J. Ceamanos, "Modelling of the pyrolysis of high density polyethylene product distribution in a fluidized bed reactor," Journal of

applied and analytical pyrolysis, vol. 79, pp. 313-322, 2007.

[33] J. Lede and O. Authier, "Temperature and heating rate of solid particles

undergoing a thermal decomposition. Which criteria for characterization fast pyrolysis?," Journal of Analytical and Applied Pyrolysis, vol. 113, pp. 1-14, 2015.

[34] A. V. Bridgwater, "Review of fast pyrolysis of biomass and product upgrading," Biomass and bioenergy, vol. 38, pp. 68-94, 2012.

[35] C. Liu, H. Wang, A. M. Karim, J. Sun and Y. Wang, "Catalytic fast pyrolysis

of lignocellulosic biomass," Chemical Society Reviews, vol. 43, pp. 7594--7623, 2014.

[36] IEA (Internaltional Energy Agency), "IEA Bioenergy annual report 2006," IEA, Paris, 2006.

[37] Y. Matsuzawa, M. Ayabe and J. Nishino, "Acceleration of cellulose co-pyrolysis with polymer," Polymer degradation and stability, vol. 71, pp. 435-444,

2001.

[38] A. Brems, J. Baeyens, J. Beerlandt and R. Dewila, "Thermogravimetric pyrolysis of waste polyethylene-terephthalate and polystyrene: A critical

assessment of kinetics modelling," Resources, Conservation and Recycling, vol. 55, p. 772–781, 2011.

[39] S. Kumagai, I. Hasegawa, G. Grause, T. Kameda and T. Yoshioka, "Thermal decomposition of individual and mixed plastics in the presence of CaO or

Ca(OH)2," Journal of Analytical and Applied Pyrolysis, vol. 113, pp. 584-590, 2015.

61

[40] J. Yu, L. Sun, C. Ma, Y. Qiao and H. Yao, "Thermal degradation of PVC: A review," Waste Management, vol. 48, pp. 300-314, 2016.

[41] H. Zhou, C. Wu, J. A. Onwudili, A. Meng, Y. Zhang and P. T. Williams, "Effect of interactions of PVC and biomass components on the formation of

polycyclic aromatic hydrocarbons (PAH) during fast co-pyrolysis," RSC Advances, vol. 5, p. 11371–11377, 2015.

[42] N. Dimitrov, L. K. Krehula, A. P. Sirocic and Z. Hrnjak-Murgic, "Analysis of recycled PET bottles products by pyrolysis-gas chromatography," Polymer

Degradation and Stability, vol. 98, pp. 972-979, 2013.

[43] T. M. Kruse, H.-W. Wong and L. J. Broadbelt, "Mechanistic Modeling of Polymer Pyrolysis: Polypropylene," Macromolecules, vol. 36, pp. 9594-9607,

2003.

[44] A. Marcilla, M. Beltran and R. Navarro, "Thermal and catalytic pyrolysis of

polyethylene over HZSM5 and HUSY zeolites in a batch reactor under dynamic conditions," Applied Catalysis B: Environmental, vol. 86, pp. 78-86,

2009.

[45] T. M. Kruse, O. S. Woo, H.-W. Wong, S. S. Khan and L. J. Broadbelt, "Mechanistic Modeling of Polymer Degradation: A Comprehensive Study of

Polystyrene," Macromolecules, vol. 35, pp. 7830-7844, 2002.

[46] D. K. Ojha and R. Vinu, "Resource recovery via catalytic fast pyrolysis of

polystyrene using zeolites," Journal of Analytical and Applied Pyrolysis, vol. 113, pp. 349-359, 2015.

[47] M. E. Bednas, M. DAY, K. Ho, R. Sander and D. M. Wiles, "Combustion and Pyrolysis of Poly(ethyleneTerephthalate). I. The Role of Flame Retardants on

Products of Pyrolysis," Journal of Applied Polymer Science, vol. 26, pp. 277-289,

1981.

[48] J. N. H. B. J. Holland, "The thermal degradation of PET and analogous

polyesters measured by thermal analysis-Fourier transform infrared spectoscopy," Polymer, vol. 43, pp. 1835-1847, 2002.

[49] B. Gui, Y. Qiao, D. Wan, S. Liu, Z. Han, H. Yao and M. Xu, "Nascent tar formation during polyvinylchloride (PVC) pyrolysis," Proceedings of the

Combustion Institute, vol. 34, no. 2, pp. 2321-2329, 2013.

[50] S. Wang, G. Dai, H. Yang and Z. Luo, "Lignocellulosic biomass pyrolysis mechanism: A state-of-the-art review," Progress in Energy and Combustion Science,

vol. 62, pp. 33-86, 2017.

[51] D. Shen and S. Gu, "The mechanism for thermal decomposition of cellulose

and its main products," Bioresource Technology, vol. 100, p. 6496–6504, 2009.

[52] J. Lehto, A. Oasmaa, Y. Solantausta, M. Kytö and D. Chiaramonti, "Review

of fuel oil quality and combustion of fast pyrolysis bio-oils from lignocellulosic biomass," Applied Energy, vol. 116, pp. 178-190, 2014.

62

[53] CRI Catalyst Company, "IH2 Technology," CRI Catalyst Company, 2018. [Online]. Available: https://www.cricatalyst.com/cricatalyst/ih2.html.

[54] Y. Xue, "Thermochemical conversion of organic and plastic waste materials through pyrolysis," Graduate Theses and Dissertation Iowa State Universit y,

Ames, 2017.

[55] B. K. Sharma, B. R. Moser, K. E. Vermillion, K. M. Doll and N. Rajagopalan,

"Production, characterization and fuel properties of alternative diesel fuel from pyrolysis of waste plastic grocery bags," Fuel Processing Technology, vol.

122, pp. 79-90, 2014.

[56] N. Miskolczi, F. Ates and N. Borsodi, "Comparison of real waste (MSW and MPW) pyrolysis in batch reactor over different catalysts. Part II:

Contaminants, char and pyrolysis oil properties," Bioresource Technology, vol. 144, pp. 370-379, 2013.

[57] I. D. Mall, "NPTEL E-learning coursed from the IITs & IISc: Chemical Enginering - Chemical Technology - I - Lectures," NPTEL, [Online].

Available: https://nptel.ac.in/syllabus/103107082/. [Accessed Febuary

2018].

[58] The meeting of the Joint FAO/WHO Expert Committee on Food Additives

(JECFA), "Residue Monograph prepared by the meeting of the Joint FAO/WHO Expert Committee on Food Additives (JECFA), 84th meeting

2017: Microcrystalline Cellulose," Food and Agriculture Organization of the United Nations (FAO) and World Health Organization (WHO), 2017.

[59] G. Thoorens, F. Krier, B. Leclercq, B. Carlin and B. Evrard, "Microcrystalline cellulose, a direct compression binder in a quality by design environment—A

review," International Journal of Pharmaceutics, vol. 473, no. 1-2, pp. 64-72, 2014.

[60] C. Zhou, "Gasification and pyrolysis characterization and heat transfer phenomena during thermal conversion of municipal solid waste," Doctoral

dissertation KTH Royal Institute of Technology, Stockholm, 2014.

[61] Pyrol AB, "PyroLab - Pyrola 2000," Pyrol AB, 2019. [Online]. Available:

http://www.pyrolab.com/pyrola-2000. [Accessed February 2019].

[62] Frontier Lab, "Single-shot pyrolyzer <PY-3030S>," Frontier Lab, [Online].

Available: http://www.frontier-lab.com/catalog/en/PY-3030S_E.pdf.

[Accessed February 2019].

[63] L. Mondello, "GCxGC Handbook: Fundamental principles of

comprehensive 2D GC," Shimadzu Coporation, Japan, 2012.

[64] M. Brebu, S. Ucar, C. Vasile and J. Yanik, "Co-pyrolysis of pine cone with

synthetic polymers," Fuel, vol. 89, pp. 1911-1918, 2010.

[65] E. Önal, B. B. Uzun and A. E. Putun, "Bio-oil production via co-pyrolysis of

almond shell as biomass and high density polyethylene," Energy conversion and management, vol. 78, pp. 704-710, 2014.

63

[66] X. Zhang, H. Lei, S. Chen and J. Wu, "Catalytic co-pyrolysis of lignocellulosic biomass with polymers: a critical review," Green Chemistry, vol. 18, p. 4145–

4169, 2016.

[67] Y. Xue, A. Kelkar and X. Bai, "Catalytic co-pyrolysis of biomass and

polyethylene in a tandem micropyrolyzer," Fuel, vol. 166, pp. 227-236, 2016.

[68] X. Li, H. Zhang, J. Li, L. Su, J. Zuo, S. Komarneni and Y. Wang, "Improving

the aromatic production in catalytic fast pyrolysis of cellulose by co-feeding low-density polyethylene," Applied catalysis A: General, vol. 455, pp. 114-121,

2013.

[69] D. K. Ojha and R. Vinu, "Fast co-pyrolysis of cellulose and polypropylene using Py-GC/MS and Py-FT-IR," RSC Advances, vol. 5, pp. 66861-66870,

2015.

[70] E. Jakab, M. Blazso and O. Fiax, "Thermal decomposition of mixtures of

vinyl polymers and lignocellulosic materials," Journal of analytical and applied pyrolysis, Vols. 58-59, pp. 49-62, 2001.

[71] R. Miranda, H. Pakdel, C. Roy and C. Vasile, "Vacuum pyrolysis of

commingled plastics containing PVC II. Product analysis," Polymer Degradation and Stability, vol. 73, pp. 47-67, 2001.

[72] R. Miranda, J. Yang, C. Roy and C. Vasile, "Vacuum pyrolysis of commingled plastics containing PVC I. Kinetic study," Polymer Degradation and Stability, vol.

72, pp. 469-491, 2001.

[73] H. Bockhorn, J. Hentschel, A. Hornung and U. Hornung, "Environmenta l

engineering: Stepwise pyrolysis of plastic waste," Chemical Engineering Science, vol. 54, pp. 3043-3051, 1999.

[74] S. Kumagai, K. Fujita, T. Kameda and T. Yoshioka, "Interactions of beech

wood–polyethylene mixtures during co-pyrolysis," Journal of Analytical and Applied Pyrolysis, vol. 122, pp. 531-540, 2016.

[75] D. K. Ojha, S. Shukla, R. S. Sachin and R. Vinu, "Understanding the Interactions Between Cellulose and Polypropylene During Fast Co-Pyrolysis

via Experiments and DFT Calculations," Chemical Engineering Transactions, vol. 50, pp. 67-72, 2016.

[76] Y. Xue, S. Zhou, R. Brown, A. Kelkar and X. Bai, "Fast pyrolysis of biomass

and waste plastic in a fluidized bed reactor," Fuel, vol. 156, pp. 40-46, 2015.

[77] J. Yang, J. Rizkiana, W. B. Widayatno, S. Karnjanakom, M. Kaewpanha, X.

Hao, A. Abudula and G. Guan, "Fast co-pyrolysis of low density polyethylene and biomass residue for oil production," Energy Conversion and Management, vol.

120, pp. 422-429, 2016.

[78] A.-C. Johansson, L. Sandström, O. Öhrman and H. Jilvero, "Co-pyrolysis of

woody biomass and plastic waste in both analytical and pilot scale," Journal of analytical and applied pyrolysis, vol. 134, pp. 102-113, 2018.

64

[79] C. Dorado, C. A. Mullen and A. A. Boateng, "H-ZSM5 Catalyzed Co-Pyrolysis of Biomass and Plastics," ACS Sustainable Chem. Eng., vol. 2, pp. 301-

311, 2014.

[80] X. Li, J. Li, G. Zhou, Y. Feng, Y. Wang, G. Yu, S. Deng, J. Huang and B.

Wang, "Enhancing the production of renewable petrochemicals by co-feedingof biomass with plastics in catalytic fast pyrolysis with ZSM-5

zeolites," Applied Catalysis A: General, vol. 481, pp. 173-182, 2014.

[81] D. K. Ratnasari, M. A. Nahil and P. T. Williams, "Catalytic pyrolysis of waste

plastics using staged catalysis forproduction of gasoline range hydrocarbon

oils," Journal of Analytical and Applied Pyrolysis, vol. 124, pp. 631-637, 2017.

[82] H. Zhou, Y. Long, A. Meng, Q. Li and Y. Zhang, "Classification of municipal

solid waste components for thermal conversion in waste-to-energy research," Fuel, vol. 145, pp. 151-157, 2015.

[83] A. Korkmaz, J. Yanik, M. Brebu and C. Vasile, "Pyrolysis of the tetra pak," Waste Management, vol. 29, pp. 2831-2841, 2009.

[84] A. López, I. d. Marco, B. Caballero, M. Laresgoiti, A. Adrados and A. Torres,

"Pyrolysis of municipal plastic wastes II: Influence of raw material composition under catalytic conditions," Waste Management, vol. 31, p.

197301983, 2011.

[85] P. Rutkowski, "Characteristics of bio-oil obtained by catalytic pyrolysis of

beveragecarton packaging waste," Journal of Analytical and Applied Pyrolysis, vol. 104, pp. 609-617, 2013.

[86] A. Undri, L. Rosi, M. Frediani and P. Frediani, "Fuel from microwave assisted pyrolysis of waste multilayer packaging beverage," Fuel, vol. 133, pp. 7-16,

2014.

[87] J. Haydary, D. Susa and J. Dudáš, "Pyrolysis of aseptic packages (tetrapak) in a laboratory screw type reactor and secondary thermal/catalytic tar

decomposition," Waste Management, vol. 33, pp. 1136-1141, 2013.

[88] M. Brebu, T. Bhaskar, K. Murai, A. Muto, Y. Sakata and M. A. Uddin,

"Removal of nitrogen, bromine, and chlorine from PP/PE/PS/PVC/ABS–Br pyrolysis liquid products using Fe- and Ca-based catalysts," Polymer

Degradation and Stability, vol. 87, no. 2, pp. 255-230, 2005.

[89] H. Zhu, X. Jiang, J. Yan, Y. Chi and K. Cen, "TG-FTIR analysis of PVC thermal degradation and HCl removal," J. Anal. Appl. Pyrolysis, vol. 82, pp. 1-

9, 2008.

[90] D. Wang, R. Xiao, H. Zhang and G. He, "Comparison of catalytic pyrolysis

of biomass with MCM-41 and CaO catalysts by using TGA–FTIR analysis," Journal of Analytical and Applied Pyrolysis, vol. 89, no. 2, pp. 171-177, 2010.

[91] L. Sorum, M. Gronli and J. Hustad, "Pyrolysis characteristics and kinetics of municipal solid wastes," Fuel, vol. 80, pp. 1217-1227, 2001.

65

[92] F. Gao, Pyrolysis of Waste Plastics into Fuels, Canterbury: PhD thesis, University of Canterbury, 2010.

[93] X. Jing, G. Yan, Y. Zhao, H. Wen and Z. Xu, "Study on mild cracking of polyolefins to liquid hydrocarbons in a closed batch reactor for subsequent

olefin recovery," Polymer Degradation and Stability, vol. 109, pp. 79-91, 2014.

[94] G. Fraissler, M. Jöllerö, T. Brunner and I. Obernberger, "Influence of dry and

humid gaseous atmosphere on the thermal decomposition of calcium chloride and its impact on the remove of heavy metals by chlorination," Chemical

Engineering and Processing, vol. 48, pp. 380-388, 2009.

[95] Y. A. Criado, M. Alonso and J. C. Abanades, "Kinetics of the CaO/Ca(OH)2 Hydration/Dehydration Reaction for Thermochemical Energy Storage

Applications," Ind. Eng. Chem. Res., vol. 53, no. 32, p. 12594–12601, 2014.

[96] S. Kumagai, K. Fujita, Y. Takahashi, Y. Nakai and T. Kameda, "Beech wood

pyrolysis in polyethylene melt as a means of enhancing levoglucosan and methoxyphenol production," Scientific reports, vol. 9, p. 1955, 2019.