Fill - Free fillable Ageing of Flame Retarded Polylactic Acid PDF form

N°: 42158

THESE DE DOCTORAT

Ageing of Flame Retarded Polylactic Acid

Présentée et soutenue publiquement à

L’UNIVERSITE LILLE 1 SCIENCES ET TECHNOLOGIES

Pour obtenir le grade de

DOCTEUR

Spécialité : Molécules et Matière Condensée

Par

Nicolas Lesaffre

Titulaire du Master Ingénierie des Systèmes Polymères de l’Université Lille 1

Thèse dirigée par

Prof. Maude Jimenez et Prof. Gaëlle Fontaine

Soutenue le 10 Novembre 2016 devant la Commission d’Examen composée de :

Prof. Serge Bourbigot, Université Lille 1 Président du Jury

Prof. Carl-Eric Wilén, Abo Akademi Univeristy Rapporteur

DR. Sandrine Thérias, Université Clermont-Ferrand II Rapporteur

Dr. Xavier Couillens, Solvay Examinateur

Prof. Maude Jimenez, Université Lille 1 Directeur de thèse

Prof. Gaëlle Fontaine, Université Lille 1 Co-directeur de thèse

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

À Virginie

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Acknowledgements

First, I would like to thank Prof. Alexandre Legris, head of the UMET laboratory, and

Prof. Serge Bourbigot, head of the team ISP-R2F (Reaction and Resistance to Fire). I would like

to thank them for giving me the opportunity to join their labs, work on this project and achieve

this PhD in excellent job conditions. I extend my thanks to Mr. Bernard Fontaine, head of the

Ecole Nationale Supérieure de Chimie de Lille (ENSCL), where the laboratory is located.

I specially would like to express my gratitude to Prof. Maude Jimenez and Prof. Gaëlle

Fontaine who supervised me during my PhD thesis in the R2Fire lab. Working with them was

a great pleasure throughout the whole project. They provided me with an unbelievable

scientific and moral support that I could ever have desired. Their expertise, their patience and

guidance experienced during the whole period of the PhD, combined with continuous and

stimulating participation, kept me highly motivated and enthusiastic.

I would like to acknowledge Prof. Serge Bourbigot who agreed to chair the jury but also

Prof. Carl-Eric Wilén, DR. Sandrine Thérias and Dr. Xavier Couillens who accepted to take on

their time, to bring their expertise as reviewers of this manuscript and to be part of the jury.

I am very grateful to any technique expert such as Bertrand Revel, Bertrand Doumert, Hervé

Vezin, Aurélie Malfait, Grégory Stoclet and Séverine Bellayer for their support and expertise

on different apparatus or techniques I used outside the lab.

Further, I would like to thank everyone in the lab for just being their selves and for the

wonderful atmosphere in the lab. First, thanks to those who left before me but brought me a

real pleasure to work in the beginning of this PhD: Bastien, Antoine, Marianne, Marion, Nico

Renaud (my dear homonymous), Gwen, Carmen and in particular my friend Andrea alias

Francis. Thanks to all the lab’s permanent members: Maude, Gaëlle, Sophie, Serge, Fabienne,

Mathilde, Michel, Charaf’, Catherine and Severine. Of course, I am very thankful to Bridget

(the small red hair crazy woman with a big heart) who really acted as a mother for all of PhD

students in the lab. Thanks to Agnès, Anil, Audrey, Ben (PlaymoBen), Chi, Hirak, JoJo, Laurie,

Maryem, Nitayooooo, Pauline (Mother Christmas), Redgy, Sawsen, Sarah, and Tatenda for all

great moments we shared together. A very special thanks to Mathieu (our little Corky), el

Midgetator, Pierlouse, Berthy “Enrico” la menace and Benji “Ballon d’or” Balavoine for being

who they are. More than colleagues they became a real friend. I will never forget all the

unbelievable moments we shared together. A page had been turned but as the poet said,

“Show must go on”…

Finally, I would like to thank my friends for having supported me during these last three

years. Of course I am also so grateful to my family, my mother, my father and my sister for

their invaluable support. Last but not least, thank you to Virginie to whom I dedicate this

manuscript, for her patience, her love and her invaluable daily support in the best and worst

moments.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Table of contents

vii

Table of contents

ACKNOWLEDGEMENTS ..................................................................................................... V

TABLE OF CONTENTS ....................................................................................................... VII

ABBREVIATIONS .............................................................................................................. XI

GENERAL INTRODUCTION ............................................................................................... 17

CHAPTER I: STATE OF THE ART ........................................................................................ 23

1. Poly(lactic acid) ........................................................................................................ 24

PLA synthesis ....................................................................................................................................... 24

Properties of PLA ................................................................................................................................. 25

1.2.1. Structure ..................................................................................................................................... 26

1.2.2. Optical properties ....................................................................................................................... 28

1.2.3. Thermal properties ..................................................................................................................... 30

1.2.4. Crystallization properties ............................................................................................................ 32

1.2.5. Mechanical properties ................................................................................................................ 35

Conclusion ........................................................................................................................................... 36

2. Flame retardancy of PLA .......................................................................................... 36

Basics on flame retardancy of polymers .............................................................................................. 36

2.1.1. Physical action ............................................................................................................................ 38

2.1.2. Chemical action ........................................................................................................................... 38

Flame retardancy of PLA by conventional flame retardants ............................................................... 38

Flame retardancy of PLA by nanoparticles .......................................................................................... 40

Intumescent PLA .................................................................................................................................. 41

Conventional flame retardants and nanoparticles: synergy in PLA ..................................................... 42

Conclusion ........................................................................................................................................... 43

3. Ageing of polymers .................................................................................................. 43

Chemical ageing of polymers ............................................................................................................... 44

3.1.1. Chemical degradation mechanisms ............................................................................................ 44

3.1.2. Photo-chemical degradation ....................................................................................................... 46

3.1.2.1. Photo-initiation step ............................................................................................................... 46

3.1.2.2. Propagation step .................................................................................................................... 48

3.1.2.3. Termination step .................................................................................................................... 50

3.1.3. Thermal degradation .................................................................................................................. 50

3.1.4. Hydrolytic degradation ............................................................................................................... 52

3.1.5. Bio-chemical degradation ........................................................................................................... 54

Ageing of Poly(lactic) acid .................................................................................................................... 55

3.2.1. Photo-degradation of PLA ........................................................................................................... 55

3.2.2. Hydrolysis and bio-chemical degradation of PLA ........................................................................ 57

3.2.3. Thermal degradation of PLA ....................................................................................................... 59

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Table of contents

viii

3.2.4. Synergy between temperature, relative humidity and/or UV light exposure ............................ 60

Ageing of flame retarded polymers ..................................................................................................... 60

3.3.1. Thermal degradation of FR polymers ......................................................................................... 61

3.3.2. Hydrolytic degradation of FR polymers ...................................................................................... 62

3.3.3. Photo-chemical degradation of FR polymers .............................................................................. 62

Conclusion ........................................................................................................................................... 65

4. Conclusion ............................................................................................................... 65

CHAPTER II: MATERIALS AND METHODS ......................................................................... 67

1. Materials preparation and processing ...................................................................... 68

Materials .............................................................................................................................................. 68

1.1.1. Poly(lactic acid) (PLA) .................................................................................................................. 68

1.1.2. Flame retardants and nanoparticles ........................................................................................... 68

Material preparation and processing .................................................................................................. 69

1.2.1. Extrusion ..................................................................................................................................... 69

1.2.2. Molding ....................................................................................................................................... 71

1.2.3. Powders ...................................................................................................................................... 71

2. Accelerated ageing methods .................................................................................... 71

Ageing in presence of temperature and relative humidity ................................................................. 71

Ageing in presence of UV-rays ............................................................................................................. 72

3. Physico-chemical characterization ............................................................................ 73

Microscopies ........................................................................................................................................ 73

3.1.1. Digital microscopy ....................................................................................................................... 73

3.1.2. Electron probe micro analysis (EPMA) ........................................................................................ 73

3.1.3. Scanning electron microscopy (SEM) .......................................................................................... 74

3.1.4. Transmission electron microscopy (TEM) ................................................................................... 74

Thermal analysis .................................................................................................................................. 74

3.2.1. Thermo-gravimetric analysis (TGA)............................................................................................. 74

3.2.2. Differential scanning calorimetry (DSC) ...................................................................................... 75

Color changes measurements ............................................................................................................. 76

Gel permeation chromatography (GPC) .............................................................................................. 77

Melt flow index (MFI) .......................................................................................................................... 78

X-ray diffraction (XRD) ......................................................................................................................... 78

X-ray photoelectron spectroscopy (XPS) ............................................................................................. 79

Fourier transform infrared spectroscopy (FTIR) .................................................................................. 79

Chemical derivatisation ....................................................................................................................... 80

Solid state nuclear magnetic resonance (NMR) .................................................................................. 80

Electron paramagnetic resonance spectroscopy (EPR) ....................................................................... 81

4. Fire testing methods ................................................................................................ 83

4.1.1. Mass loss calorimeter (MLC) ....................................................................................................... 83

4.1.2. Limiting oxygen index (LOI) ......................................................................................................... 83

5. Conclusion ............................................................................................................... 84

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Table of contents

CHAPTER III: IMPACT OF AGEING ON PLA ........................................................................ 85

1. Visual measurement of ageing ................................................................................. 86

Visual evaluation ................................................................................................................................. 86

Surface morphology of PLA after ageing ............................................................................................. 87

Color changes ...................................................................................................................................... 88

2. Physico-chemical properties after ageing ................................................................. 89

Impact of ageing on molecular mass ................................................................................................... 90

Thermal behavior ................................................................................................................................ 93

2.2.1. Thermogravimetric analysis ........................................................................................................ 93

2.2.2. Glass transition and melting temperature .................................................................................. 94

Effect of ageing on crystallization behavior ......................................................................................... 98

Conclusion ......................................................................................................................................... 103

3. Elucidation of the mechanisms of degradation ........................................................ 104

Degradation under T/RH exposure .................................................................................................... 104

Degradation under T/UV and T/UV/RH exposure. ............................................................................ 106

3.2.1. Identification of radical species ................................................................................................ 106

3.2.2. Secondary products formed during degradation ...................................................................... 111

4. Conclusion .............................................................................................................. 116

CHAPTER IV: INFLUENCE OF FLAME RETARDANTS ADDITIVES ON THE AGEING OF PLA .... 119

1. Impact of ageing on the morphology of FR-PLAs ...................................................... 120

Visual observations ............................................................................................................................ 120

Surface morphology of FR-PLAs after ageing .................................................................................... 122

Conclusion ......................................................................................................................................... 125

2. Physico-chemical properties of FR-PLAs during ageing ............................................. 125

Molecular mass .................................................................................................................................. 125

2.1.1. Impact of fillers on molecular mass of FR-PLAs ........................................................................ 125

2.1.2. Impact of ageing on the molecular mass of FR-PLAs ................................................................ 127

Thermal properties ............................................................................................................................ 130

2.2.1. Effect of fillers on glass transition and melting temperature ................................................... 130

2.2.2. Behavior of glass transition and melting temperatures during ageing ..................................... 130

Crystallization behavior ..................................................................................................................... 135

2.3.1. Crystallization comportment of unaged FR-PLAs ..................................................................... 135

2.3.2. Influence of ageing exposure on crystallization behavior of FR-PLAs....................................... 137

Conclusion ......................................................................................................................................... 140

3. Mechanisms of degradation of PLA in presence of fillers ......................................... 141

Surface chemical analysis of aged FR materials ................................................................................ 142

Role FR additives during hydrolysis ................................................................................................... 144

3.2.1. Hydrolysis of flame retardant additives .................................................................................... 144

3.2.2. Evaluation of the migration of flame retardant additives ........................................................ 148

Impact of FR additives during UV irradiation of PLA ......................................................................... 152

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Table of contents

4. Conclusion .............................................................................................................. 155

CHAPTER V: IMPACT OF AGEING ON THE FLAME RETARDANT PROPERTIES OF FR-PLAS... 159

1. Flame retardant performances of FR-PLAs ............................................................... 160

Thermal stability of FR-PLAs .............................................................................................................. 160

Reaction to fire .................................................................................................................................. 161

1.2.1. Behavior under mass loss cone calorimeter (MLC) ................................................................... 161

1.2.2. Limiting oxygen index (LOI) test results .................................................................................... 164

Mechanism of action of FR-PLAs ....................................................................................................... 164

Conclusion ......................................................................................................................................... 166

2. Thermal stability and flame retardant properties after ageing ................................. 167

Effect of ageing on thermal stability of FR-PLAs ................................................................................ 167

Limiting Oxygen Index (LOI) test results after ageing ........................................................................ 170

Viscosity of FR-PLAs ........................................................................................................................... 172

Mass loss cone performances after ageing ....................................................................................... 176

Mass loss cone behavior of FR-PLAs loaded at 20 wt.-%. .................................................................. 178

Molecular mass of FR-PLAs loaded at 20 wt.-% ................................................................................. 182

3. Conclusion .............................................................................................................. 185

GENERAL CONCLUSION .................................................................................................. 187

OUTLOOK ...................................................................................................................... 191

LIST OF FIGURES, TABLES AND EQUATIONS .................................................................... 195

REFERENCES .................................................................................................................. 205

APPENDIX ...................................................................................................................... 219

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Abbreviations

ABS Acrylonitrile butadiene styrene

ATH Aluminium trihydroxide

ATR Attenuated total reflectance

APP Ammonium polyphosphate

BDP Bisphenol A bis(diphenyl phosphate)

BSE Back scattered electron

C Carbon

C20A Cloisite 20A

C30B Cloisite 30B

CIE Commission internationale de l'éclairage

CCD Charge-coupled device

CDCl

Deuterated chloroform

CNa

Cloisite Na

Cp Heat capacity

DecaBDE Decabromodiphenylether

DSC Differential scanning calorimetry

DTG Differential Thermogravimetric analysis

Activation energy

E&E Electrical and electronic

EG Expanded graphite

EPMA Electron probe micro-analysis

EPR Electron paramagnetic resonance

ESEEM Electron spin-echo envelope modulation

EVA Ethyl vinyl acetate

FR Flame retarded

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Abbreviations

xii

FR-PLA PLAMEL

APP

FR-PLA-C30B PLAMEL

APP

C30B

FTIR Fourier transform infrared spectroscopy

FTT Fire testing technology

GPC Gel permeation chromatography

HIPS High impact polystyrene

HRR Heat release rate

K Kelvin

LDPE Low density polyethylene

LIG Lignin

LOI Limiting oxygen index

Critical molecular mass

Entanglement molecular mass

Number average molecular mass

Weight-average molecular weight

MAH Maleic anhydride

MAS Magic angle spinning

MC Melamine cyanurate

MDH Magnesium dihydroxide

MEL Melamine

MFI Melt flow index

MLC Mass loss calorimeter

MMT Montmorillonite

MP Melamine polyphosphate

MWNT Multi-walled carbon nanotubes

N Nitrogen

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Abbreviations

xiii

NFPA National Fire Protection Association

Ammonia

NMR Nuclear magnetic resonance

number of chain scission

O Oxygen

P Phosphorus

PA Polyamide

PA6 Polyamide 6

PC Polycarbonate

PC2 Multicouche Ni/C

PCL Polycaprolactone

PE Polyethylene

PEA Poly(esteramide)

PEEK Poly(ether ether ketone)

PER Pentaerythritol

PET Poly(ethylene terephthalate)

PHAs Polyhydroxyalkanoates

pHRR Peak of heat release rate

PI Polydispesity index

PLA Poly(lactide) / Poly(lactic acid)

PMMA Poly(methyl methacrylate)

PP Polypropylene

PS Polystyrene

PVC Poly(vinyl chloride)

Py-GC/MS Pyrolysis – gas chromatography - mass spectroscopy

r Rotational deformation

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Abbreviations

xiv

RDP Resorcinol bis(deiphenyl phosphate)

RI Refractive index

ROP Ring opening polymerization

Si Silicon

SEM Scanning electron microscopy

ST Starch

Crystallization temperature

Glass transition temperature

Melting temperature

TAP Thallium acid phthalate

TBP-DP 2,4,6-tribromophenyl-2’,3’-dibromopropylether

TEM Transmission electron microscopy

TGA Thermo-gravimetric analysis

THR Total heat release

TiO

Titanium dioxide

THF Tetrahydrofuran

TMS Tetramethylsilane

TPP Triphenyl phosphate

TTI Time to ignition

UV Ultraviolet

WAXD wide-angle X-ray diffraction

XRD X-ray diffraction

XPS X-ray photoelectron spectroscopy

ZFS Zero field splitting

2D HYSCORE 2D Hyperfine sublevel correlation spectroscopy

ΔH

Enthalpy of crystallization

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Abbreviations

ΔH

Enthalpy of melting

Χ Crystallinity

δ Chemical shift

asym

Asymmetric deformation (scissoring)

sym

Symmetric deformation (scissoring)

ν Elongation vibration

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

General Introduction

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

General Introduction

Plastics are present in the human life since Antiquity, when the ancient Mesoamericans

first processed natural rubber into balls or figurines, but it is most of all at the end of XIX

century that they have been developed. During this period, synthetic plastics were developed

to be used in a broad range of applications. However, the improvements in chemical

technology after the First World War permitted the expansion of new forms of plastics and

mass production which began around the 1940s. The production of plastics has grown up and

has never decreased since that time (Figure 1), except during the oil shocks in 1973 and 1979,

and during the crisis in 2008. The European production of synthetic plastics has reached about

58 million tons in 2011 compared to less than half a million tons in the 50’s. The production

market of synthetic plastics is dominated by China, Europe and North American Free Trade

Agreement (NAFTA) with ca. 64 % of the worldwide plastic production (Figure 1) [1, 2].

Figure 1: World plastic production from 1950 to 2012 [1, 2].

For more than 50 years, plastics have thus gradually replaced traditional materials such as

wood or metal in many applications. They present a great variety of properties and a

processing versatility, which make them part of our everyday life. Most of these plastic

materials are produced from non-renewable and natural gas-resources. Thus, as the use of

synthetic polymers has grown during the past decades, the environmental pollution has

extended in dangerous proportions last few years mainly due to (i) the production of these

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

General Introduction

plastics and (ii) their disposal in landfills after use [1]. Therefore, a move from fossil-based

polymers to new polymers exhibiting further properties, especially bio-based polymers, would

be appropriate. The worldwide capacity of bio-based plastics production, according to

company announcements, will increase from 1.58 Mt in 2013 to 7.85 Mt in 2019 [3, 4]. It is

projected that the most important representatives by 2020 will be starch based plastics

(1.3 Mt), polylactic acid (PLA) (0.8 Mt) and bio-based polyethylene (PE) (0.6 Mt) [3]. Of the

above mentioned bio-based plastics, PLA has drawn attention these last years as an

interesting material in several applications areas, particularly in packaging, electrical and

electronic (E&E) and automotive [3, 5]. PLA is very easy to process and its production requires

20 to 50% less fossil fuel resources than the production of petroleum based polymers [3].

Figure 2: Global production capacities of bioplastics in 2014 (by material type) [4].

In spite of their properties and their numerous applications, plastics exhibit one major

drawback: most of them are highly flammable. In case of fire, these materials will tend to

release fuel and toxic smokes. The US National Fire Protection Association (NFPA) report that

1.300.000 fires occurred in the United States in 2014 [6], corresponding to a fire every 25

second, and more tragically caused civilian fire deaths and casualties. Plastics are heavy

contributors to the development and propagation of a fire. Hence, to be used in certain

sectors (e.g. E&E), plastics have to be flame retarded and their flame retardant (FR) properties

as well as their properties of use must be kept for the product lifetime. Looking at bio-based

plastics, the interest for flame retarded PLA kept growing these past few years (Figure 3); more

than 145 papers are dedicated to flame retardancy of PLA in 2015.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

General Introduction

Figure 3: Evolution of number of publications versus time with keyword “flame retardancy”

and “PLA” (Scifinder Sept 2016).

Plastics materials can be fire retarded by different methods [7, 8]: (i) by physical methods,

i.e. the incorporation of flame retardants (FRs) into polymers via melt blending, (ii) by chemical

methods, i.e. the incorporation of FRs into the chemical structure of polymers, e.g. via

copolymerization or grafting and (iii) by the coating of a FR layer on the surface of the material.

Various kinds of FRs are used in polymers, such as mineral fillers and boron-, phosphorus-,

halogen- and nitrogen-based FRs. Concerning PLA, flame retardancy can be obtained by

incorporating conventional flame retardants such as metal hydroxides (e.g. Aluminium

trihydroxide (ATH), Magnesium hydroxide (MDH) …), phosphorus-containing compounds

(e.g. Ammonium polyphosphate (APP), Melamine polyphosphate (MP) …) halogenated

(e.g. Decabromodiphenylether (DecaBDE), Triphenylphosphate (TPP) …) and nitrogenated

compounds (e.g. Melamine (Mel), Melamine cyanurate (MC) …) which can be combined with

nanoparticles (e.g. organoclays, carbon nanotubes …) [8-24].

Fire standard tests for polymer-based materials are however usually carried out on newly

manufactured samples. However, exposure to various natural ageing conditions (e.g. sun, rain,

temperature …) can impair flame retardant properties throughout the lifetime of the material.

Therefore, it is necessary to evaluate the influence of different kinds of ageing conditions on

the FR properties of these FR polymers. Numerous studies can be found in the literature

concerning the effect of temperature and relative humidity or light (UV-rays) on the

degradation of neat PLA [25-35]. The degradation of a PLA matrix containing fillers (sepiolite,

fluorohectorite, calcium sulphate, nanoclays …) has for example been investigated [32, 34-39].

It is demonstrated that PLA is extremely sensitive to hydrolysis and photo-degradation

phenomenon, which are accelerated when temperature increases [29]. These kinds of ageing

exposure involve the scission of the polymer chains, leading to the complete degradation of

the material. Most of the time hydrolysis and photo-degradation lead to a decrease (i) of the

molecular mass of the material, (ii) of its glass transition temperature and (iii) of the

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

General Introduction

mechanical properties of the polymer [29, 33-35, 38]. The incorporation of fillers such as

nanoclays or calcium sulphate could accelerate the degradation of the PLA matrix due to the

affinity of fillers with water and UV radiations [35, 37]. Few papers however deal with the

ageing of flame retarded materials [40-48]. In all the works reported in the literature, the

effect of accelerated ageing on the FR properties is mainly evidenced, but the authors do not

establish clear relations between these properties and the changes in thermo-physical

properties of the polymers during ageing process. This relationship is however needed to

elucidate the ageing mechanism and thus design long-term durable flame retardant

formulations.

A recent study of our group reported the effectiveness of the combination of Melamine,

Ammonium polyphosphate (APP) and organomodified montmorillonite (Cloisite 30B) on the

FR properties of PLA [12]. In this way, two formulations were developed, i.e. FR-PLA

(containing Melamine and APP) and FR-PLA-C30B (containing Melamine, APP and

Cloisite 30B). Thus, based on this study, this work deals with the impact of accelerated ageing

on flame retarded PLA and the role of flame retardant additives during ageing of this polymer.

Three different kinds of ageing will be investigated: (i) temperature and UV exposure (T/UV),

(ii) temperature and relative humidity exposure (T/RH) and (iii) a combination between

temperature, relative humidity and UV radiations (T/UV/RH).

This work is divided into five chapters. The first chapter gives a general background about

PLA. The synthesis, structure and the different properties of this polymer are reported. Then,

a literature review investigates the flame retardancy of polymers and of PLA in particular. The

last section of this chapter gives an overview of the chemical ageing of polymers and some

examples are given concerning the impact of ageing on PLA and flame retarded materials in

general.

The second chapter reports the materials and experimental methods used in this study.

The polymer (PLA) and flame retardant additives as well as the material preparation process

are described. Then, accelerated ageing tests and characterization methods are presented

followed by experimental techniques permitting to evaluate fire retardant properties of the

materials.

The third chapter reports the impact of ageing tests on neat PLA as a reference material.

The objective of this chapter is to identify the effects of T/UV, T/RH and T/UV/RH exposure on

the physico-chemical properties of neat PLA in order to evidence the mechanisms of

degradation involved. First the visual modifications as well as the impact on the molecular

mass, glass and melting temperatures, crystallinity and thermal stability are investigated for

each kind of exposure. Furthermore, hydrolytic and photo-oxidative mechanisms of

degradation occurring during ageing are proposed.

Chapter four studies the impact of the incorporation of FR fillers on the properties of PLA.

Then, the impact of ageing on the physico chemical properties of flame retarded PLAs, i.e. FR-

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

General Introduction

PLA and FR-PLA-C30B is evaluated. Finally, the role of fillers on the mechanisms of degradation

of PLA is investigated.

The last chapter deals with the impact of ageing on the flame retardant properties of PLA.

The fire properties and the mechanism of action of FR-PLAs are firstly reminded. Thereafter,

fire retardant properties after ageing are investigated. Finally, the relationship between the

change in both physico-chemical and flame retardant properties is examined. The role of

ageing on the behavior of flame retardant properties is thus detailed.

At the end of this work a general conclusion is given and outlook for future studies on this

subject are proposed.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter I: State of the Art

This first chapter starts by an overview of a biodegradable polymer, i.e. PLA. In particular

its production process and properties, are fully commented. As this study mainly deals with

ageing of flame retarded materials, the different ways to obtain flame retardant PLA are

described. At the end of the chapter, the different types of chemical ageing affecting

polymeric materials and especially PLA are discussed with focus on the mechanisms of

degradation occurring during ageing. Moreover, some examples of aged flame retarded

systems are presented.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter I: State of the Art

1. Poly(lactic acid)

The aim of this study is to evidence and understand the impacts of ageing on flame retarded

polymers, and thus determine the FR durability of such materials. In this way we’ve decided

to focus this study on polylactic acid (PLA) which has recently received some great deal of

attention [3, 5] and especially in the field of flame retardancy [8, 10, 12-15, 20-23, 49-51].

Before investigating in details the ageing of flame retarded PLAs, the strategy is to first

understand the effect of ageing on raw PLA. Hence, determining the mechanisms of

degradation occurring during ageing as well as the ageing effect on the physico-chemical

properties of the raw PLA. Then the effect of the incorporation of flame retardant fillers on

the (i) FR behavior, (ii) the physico-chemical properties and (iii) degradation of PLA during

ageing will be investigated.

PLA is nowadays the first commodity biodegradable polymer produced from annually

renewable resources. It is a potential good candidate to replace traditional petroleum derived

polymeric materials [3]. PLA is relatively cheap and has some remarkable properties that make

it suitable for various applications such as packaging, textile or transportation like automobile

[3, 5, 52]. PLA can be modified, plasticized, filled, chemically modified, reactive blended and

processed like many other conventional polymers. Nevertheless, PLA must be studied

carefully due to its unique properties: structural, thermal, crystallization and rheological

properties of PLA must be taken into account during its converting process [52]. Hence, an

overview of this biodegradable polymer (i.e. the synthesis pathways, its structure and

properties) is reported in this part, in order to have a wide background knowledge of the

material, before investigating its ageing behavior.

PLA synthesis

Using an appropriate catalyst and heat, two major routes exist to produce PLA (Figure 4):

(i) direct condensation polymerization of lactic acid or (ii) conversion of lactic acid to the cyclic

lactic dimmer and ring opening polymerization (ROP) through the lactide intermediate [5, 52].

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter I: State of the Art

Figure 4: Synthesis of PLA from L- and D-lactic acids, adapted from Lim et al. [52].

ROP is the most common way to produce high molecular weight PLA. In fact, water must

be eliminated during polymerization of lactic acid to obtain high molecular weight polymer.

As water remains during the direct condensation polymerization, the molecular mass of PLA

is decreased [5]. Thus, it is necessary to add a step during the process including chain coupling

agents in order to extend the chain length of the polymer and then obtain high molecular

weight PLA [53].

The ROP of lactide monomer avoids using coupling agents. Indeed, poly(lactic acid)

oligomers, also called pre-polymers (with low molecular weight) are formed by

polycondensation in a first step. Then, in a second step, the depolymerization of

poly(lactic acid) oligomers produces a cyclic dimer called lactide monomer. Lactide monomer

is then finally polymerized by ROP to form high molecular weight polylactide (Figure 4).

Among numerous processes allowing obtaining polylactide, ROP is the most commonly

used in the industry. However, this kind of process requires the use of a catalyst in order to

initiate the polymerization [53]. Numerous catalysts can be used for ROP but previous

researches have demonstrated the efficiency of tin octanoate S

(Oct)

as catalyst [51]. In fact,

tin octanoate is preferred as catalyst for the bulk polymerization of lactide due to its solubility

in molten lactide, high catalytic activity, and low rate of racemization of polymer [51, 54-56].

Properties of PLA

The ability to control both PLA production and structural architecture permits to predict

and evaluate its different physico-chemical properties (e.g. crystallinity, mechanical, thermal

or optical properties). As these latter play a key role on the lifetime of the material, the next

section is dedicated to the different properties of PLA.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter I: State of the Art

1.2.1. Structure

The polymer architecture and its molecular mass are responsible of the different properties

of the high molecular weight PLA. According to Auras et al. [5] the ability to control the

stereochemical architecture permits precise control over the speed of crystallization and

finally the degree of crystallinity, the mechanical properties and the processing temperatures

of the material. It is thus of primary importance to study the structure of PLA.

PLA is a linear aliphatic polyester (Figure 5) that is composed of repeating units of lactic

acid (72 g/mol) containing an asymmetric carbon atom. This stereo center can be either in (S)L

or (R)D configuration. L-lactic acid and D-lactic acid, the two isomers of lactic acid are shown

in Figure 6.

Figure 5: Poly(lactic acid) structure.

Figure 6: The stereoisomers of lactic acid.

Most of the biological organisms producing lactic acid provide mainly L-lactic acid and no

major source of D-lactic acid is available [5]. Theoretically, these two isomers of lactic acid are

able to produce four distinct materials; poly(D-lactic acid) (PDLA), poly(L-lactic acid) (PLLA),

poly(D,L-lactic acid) (PDLLA) and meso-PLA, obtained by the polymerization of meso-lactide

(L,D-Lactide).

Depending on the optical configuration of the lactic acid three main stereoisomers of

lactide result from the depolymerization of poly(lactic acid) pre-polymers: L,L-lactide, D,D-

lactide and L,D-lactide (Figure 7).

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter I: State of the Art

Figure 7: Chemical structures of L,L-, L,D- and D,D-lactides [5].

PLA can also be found in stereocomplex form. Stereocomplexation of polymers

corresponds to the formation of complex between two kinds of stereoisomers. This is for

example the case of poly(methyl methacrylate) (PMMA) [57], poly(γ-benzyl glutamate) [58],

and PLA [59, 60]. In the case of PLA, stereocomplexation happens between PLLA and PDLA.

Hydrogen bonds are formed between hydrogen and carbonyl group, contributing to the

formation of crystalline structure (Figure 8) different from the α-, β- and γ-forms commonly

observed [60]. These stereocomplex can be formed in solution or by melt blending.

Figure 8: Stereocomplex of PLLA and PDLA [61].

Multi-step polymerization is another way to achieve stereocomplex, in order to obtain

PDLLA diblock or multiblock poly-(L,L)-(D,D)-lactide. Depending on the process used, it is

possible to obtain different PDLLA chains (Figure 9) [60].

Figure 9: PDLLA multiblocks as function of D and L content, adapted from Kakuta et al. [60].

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter I: State of the Art

1.2.2. Optical properties

Figure 10 presents the transmission percentage versus wavelength for different polymers,

including PLA in the wavelength range of 180 – 600 nm [5]. As reported by Auras, Harte and

Selke [5], PLA does not transmit ultra-violet (UV) rays in the range of 190 – 220 nm whereas

at 225 nm the amount of UV light transmitted by PLA increases significantly. Indeed, 85 % of

the UV light is transmitted at 250 nm compared to 95 % at 300 nm. Thus, while UV-C is not

transmitted, nearly all the UV-B and UV-A light passes through the PLA films. Therefore, the

application of transparent PLA films to, for example, dairy products may require additives to

block UV light transmission.

Figure 10: Transmission percentage versus wavelength for PLA, PS, LDPE, PET and cellophane

films, adapted from Auras et al. [5].

Effective UV stabilizers are able to absorb UV light and thus to prevent damage to light

sensitive packaged foods, resulting in retention of taste and appearance, extension of shelf

life and improvement of product quality. PLA transmits less UV-C light than LDPE (Figure 10),

but PET, PS and Cellophane transmit less UV-B and UV-A than PLA and LDPE. Moreover, PET

does not transmit UV-C and UV-B wavelengths, which are the most damaging for food.

PLA also shows some characteristics bands by Fourier Transform Infrared (FTIR)

spectroscopy (Figure 11). For this polymer, maximum absorbance occurs at 240 nm and is

attributed to ester group present in the backbone (Figure 5). The main PLA absorption bands

in the infrared range are summarized in Table 1 [29, 32, 34, 35, 62, 63].

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter I: State of the Art

Figure 11: FTIR (ATR) spectrum of PLA in the range 500 to 4000 cm

-1

Table 1: Infrared spectroscopy data for PLA.

Assignment

Peak position (cm

-1

)

-OH stretch (free)

3571

-CH- stretch

2997 (asym), 2946 (sym), 2877

-C=O carbonyl stretch

1748

-CH

bend

1456

-CH- deformation including symmetric and

asymmetric bend

1382, 1365

-CH- bend

1315 (not shown), 1300 (not shown)

-C=O bend

1257

-C-O- stretch

1211, 1130, 1093

-OH bend

1047

-CH

rocking modes

956, 921 (not shown), 912 (not shown)

-C-C- stretch

926 (not shown), 870

-C=O deformation

755

These different bands correspond to characteristics bonds of PLA such as –OH groups

located at the extremity of the polymer chain (3571 cm

-1

and 1047 cm

-1

corresponding to

stretching and bending regions). The bands at 2997, 2946 and 2877 cm

-1

are assigned to the

CH stretching region, ν

, ν

and ν CH modes. The C=O stretching region appears as a

large band at 1748 cm

-1

. The region between 1500 cm

-1

and 1360 cm

-1

is characterized by the

1456 cm

-1

band. The CH deformation and asymmetric bands appear at 1382 cm

-1

and

1365 cm

-1

, respectively. The bands at 1315 cm

-1

and 1300 cm

-1

are due to CH bending modes.

Furthermore, in the region of 1300 cm

-1

to 1000 cm

-1

, it is possible to observe the C-O

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter I: State of the Art

stretching modes of the ester groups at 1211 cm

-1

and the ν C-O asymmetric mode at

1093 cm

- 1

. Between 1000 cm

-1

and 800 cm

-1

, peaks can be observed at 956 cm

-1

921 cm

-1

and

912 cm

-1

which can be attributed to the characteristic vibrations of the CH

rocking modes

(peaks of PLA that had the α- and β-form crystals were observed at 921 and 912 cm

-1

respectively). Bands related to the crystalline and amorphous phase of PLA can be observed.

The peaks at 956 cm

-1

and 871 cm

-1

can be assigned to the amorphous phase and the ones at

921 cm

-1

and 756 cm

-1

to the crystalline phase [64, 65].

1.2.3. Thermal properties

In the solid state, PLA can be either amorphous or semi-crystalline, depending on the

overall optical composition, primary structure, thermal history, and molecular weight. Semi-

crystalline PLA exhibits a glass transition temperature (T

) and a melting temperature (T

Most of the physical characteristics (density, heat capacity, mechanical and rheological

properties) of PLA are to a great extent dependent on its glass transition temperature.

The T

(50-60°C) and T

(130°C-230°C) of PLA are important in order to define the

commercial applications. Above T

(around 58°C) PLA is rubbery, while below T

, it is still able

to creep until it is cooled to its β-transition temperature at around -45°C, below which it

behaves as a brittle polymer [52, 64, 66]. As displayed in Figure 12 the T

of PLA is very

dependent on both the molecular weight and stereoisomer content of the polymer. According

to Dorgan et al. [67], T

increases with molecular weight until a maximum value at infinite

molecular weight. These maximum values reach 60°C, 56°C and 54°C for PLAs with 100, 80

and 50 % L-stereoisomer respectively. It may be noted that, for similar content of L-lactic acid

and D-lactic acid, the PLA containing L-lactide has higher T

values [67].

Figure 12: Glass transition temperatures for PLAs of different L-contents as a function of

molecular weight, adapted from Dorgan et al. [67].

The relationship between T

and molecular weight of a polymer can be represented by the

Flory-Fox Equation 1:

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter I: State of the Art

= T

∞





Equation 1

Where T

∞

is the T

at the infinite molecular weight, K is a constant representing the excess

free volume of the end groups of polymer chains, and M

is the number average molecular

weight. The values of T

∞

and K are around 57 and 58°C and 55000 and 73000 as reported in

the literature for PLLA and PDLLA, respectively [68]. The thermal history of the polymer also

plays an important role concerning the T

value of PLA. Quenching the polymer from the melt

at a quick cooling rate (>500°C/min, such as during injection molding) will lead to a highly

amorphous PLA. The PLA obtained has a tendency to rapidly undergo ageing in matter of days

under ambient conditions [69, 70] and this phenomenon contributes in an important manner

to the embrittlement of PLA.

As well as T

, the melting temperature (T

) of PLA also depends on its stereoisomer

content. Hence, the T

for stereo-chemically pure PLA (either L or D) is around 180°C with an

enthalpy between 40-50 J/g. The presence of meso-lactide in the PLA structure can depress

the T

by as much as 50°C, depending on the amount incorporated to the polymer. Figure 13

shows the variation of the T

as a function of % meso-lactide introduced in the PLA based on

data from Witzke and Hartmann [52, 71, 72].

Figure 13: Peak melting temperature of PLA as a function of % meso-lactide [52]. (○)

represents values reported by Witzke [71]; (●) represents values reported by Hartmann [72].

The relationship between T

and meso-lactide content can be approximated reasonably

well by the following Equation 2 [71]:

(°C) ≈ 175°C – 300W

Equation 2

Where W

is the fraction of meso-lactide below 18 % level, and 175°C is the melting

temperature of PLA made of 100% L-lactide. Typical T

values for PLA are in the range of

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter I: State of the Art

130 - 160°C. The T

decrease induced by meso-lactide has several important implications as it

helps (i) expanding the process window, (ii) reducing thermal and hydrolytic degradation, and

(iii) decreasing lactide formation [52].

1.2.4. Crystallization properties

PLA produced from more than 93 % L-lactic acid can be semi crystalline whereas PLA from

50 to 93 % of L-lactic acid is strictly amorphous. It is well known that commercial PLAs are

copolymers of PLLA and PDLLA, which are produced from L-lactide and D,L-lactide (maxi 10%

of D), respectively [52, 64, 65]. The L-isomer constitutes the main fraction of PLA derived from

renewable sources since the majority of lactic acid from biological sources exists in this form.

Depending on the compositions of the optically active L- and D-L enantiomers as well as on

the preparation conditions, PLA can crystallize principally in three forms (α, β and γ).

The α-form crystal is the most stable and exhibits a well-defined diffraction pattern [73]. It

has a melting temperature (T

) of 185°C compared to 175°C for the β-form [5, 52]. As

described in Table 2, the α-form has an orthorhombic space group, with a unit cell containing

two antiparallel chains. The lattice parameters are a = 10.66 A°, b=6.16 A° and c (chain axis) =

28.88 A°, with a crystal density of 1.26 g.cm

-3

[74].

The chain conformation of the β-form is a left-handed 3-fold helix [74]. It has an

orthorhombic or a trigonal unit cell containing a 3 1 (3 A° rise/1 monomeric unit) polymeric

helix. The unit cell dimensions are as follows: a = 10.31 A°, b = 18.21 A° and c = 9.0 A° for the

orthorhombic and a = 10.52 A°, b = 10.52 A° and c = 8.8A° for the trigonal (Table 2) [73].

Finally, the γ-form, found by epitaxial crystallization, contains two antiparallel (s(3/2)

helices in the pseudo orthorhombic unit cell (a = 9.95 A°, b = 6.25 A° and c = 8.8 A°) and it

assumes the known three-fold helix of PLA (Table 2) [75].

Table 2: Unit cell parameters for PLLA and stereocomplex crystals [5].

Crystalline

Phase

Space Group

Helix

Helical

conformation

a (nm)

b (nm)

c (nm)

α (°)

β (°)

γ (°)

P-

orthorhombic

1.07

0.645

2.88

P-

orthorhombic

1.06

0.61

2.88

Orthorhombic

1.05

0.61

Orthorhombic

1.031

1.821

0.90

Trigonal

1.052

0.88

120

Orthorhombic

0.995

0.625

0.88

Stereo

complex

Triclinic

0.916

0.87

109.2

109.8

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter I: State of the Art

Generally, the faculty of PLLA or PDLA to crystallize is highly linked to the thermal process

implementation [76], the nature and quantity of fillers that they contain as well as the ratio

between L-lactide/D-lactide [61].

Many researchers investigated the crystallization behavior of PLA [64, 65, 72, 77].

Accordingly, Differential Scanning Calorimetry (DSC) is commonly used to determine the

crystallinity of PLA. By measuring the heat of fusion ΔH

and heat of cold crystallization ΔH

the crystallinity can be determined based on the following Equation 3:

Crystallinity (%) =





Equation 3

ΔH

(100%) is the melting enthalpy for 100% crystalline PLLA or PDLA homopolymers and

is 93.1 J/g [52]. Quenching the optically pure PLA from the melt at quick cooling rate leads to

a decrease in crystallinity (%) and results in a quite amorphous polymer. The quenched PLA

(Figure 14) resulted in an exothermic crystallization peak and a glass transition temperature

upon subsequent reheat, while slow cooling produces a polymer with higher crystallinity with

much lower enthalpy of crystallization [77].

Figure 14: DSC thermograms of water quenched, air-annealed (cooled from 220°C to ambient

temperature in 5 min), and full-annealed (cooled from 220°C to ambient temperature in 105

min) PLLA samples. DSC scans were performed at a heating rate of 10°C/min, adapted from

Sarasua et al. [77].

The tendency for PLA to crystallize upon reheat also depends on the heating rate. In fact,

Figure 15 illustrates a set of the heat capacities of amorphous poly(L-lactide) after cooling

from the melt at 10 K/min, followed by heating with heating rates from 0.3 to 30 K/min. These

results indicate that the quenched poly(L-lactide) crystallizes to different degrees on heating

rates slower than 30 K/min, and undergoes additional reorganization, melting, and

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter I: State of the Art

recrystallization between the glass transition and final melting. The stereoisomer content

plays as well an important role in the tendency for PLA to crystallize upon reheat (Figure 16).

As shown in Figure 16, PLA polymers containing more than 8% D-isomer remain amorphous

even after 15 h of isothermal treatment at 145°C. In contrast, at 1.5% D-isomer level, although

the quenched PLA has a minimal crystallinity, the isothermal treatment at 145°C resulted in a

large endothermic melting peak around 450 K.

Figure 15: DSC scans for 1.5% D-lactide PLA samples cooled from the melt at 10K/min and

then reheated at different heating rates from 30 to 0.3 K/min, adapted from Pyda et al. [78].

Figure 16: DSC scans at 20 K/min for PLA with 1.5% (PLA-L), 8.1% (PLA-M) and 16.4% (PLA-H)

D-isomers. All samples were cooled quickly from the melt and isothermally crystallized at

145°C for 15h. The quenched PLA-L sample was cooled similarly from the melt but did not

undergo the 15 h isothermal crystallization, adapted from Pyda et al. [78].

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter I: State of the Art

In general, the half time crystallization of PLA increases about 40% for every 1% meso-

lactide (L,D-Lactide) in the polymerization mixture, which is mainly driven by the reduction of

the melting point for the copolymer [79]. It also has been demonstrated by Vasanthakumari

et al [80, 81] that the kinetics of crystallization depends on the molecular mass of the polymer.

In fact, authors reported that the rate of crystal formation increases with a decrease of the

molecular weight.

Overall, the formation of crystals may or may not be favorable depending on the end-use

requirements of PLA articles. As an example, high crystallinity will not be optimal for injection

molded preforms that are intended for further blow molding, since rapid crystallization of the

polymer would hamper the stretching of the preform and optical clarity of the resulting bottle.

On the contrary, an increased crystallinity will be desirable for injection molded articles for

which good thermal stability is important. Crystallization of PLA articles can be initiated by

annealing at temperatures higher than T

and below T

to improve their thermal stability.

1.2.5. Mechanical properties

The mechanical properties of PLA have been widely studied over the years [5, 64, 72, 82].

Despite its numerous advantages such as high strength and high modulus, the inherent

brittleness is problematic in many commercial applications. Compared to Polystyrene (PS),

PLA not only has comparable tensile strength and modulus but also exhibits very similar

inherent brittleness as shown in Table 3. In fact, T

of PLA is between 55 and 65°C, tensile

strength at break and modulus are 53 MPa and 3.4 GPa, respectively, whereas elongation at

break is 6%.

Table 3: PLA mechanical properties compared to those of most common polymers used in

commodity applications [83].

PLA

PET

HIPS

(°C)

55-65

105

Tensile strength

at break (MPa)

Tensile modulus

(GPa)

3.2

2.8

3.0

2.1

0.9

Elongation at

break (%)

130

120

These last years PLA toughening has become the focus of numerous investigations. Many

strategies have been more or less successfully developed to improve the toughness of PLA,

including plasticization, copolymerization, addition of fillers and blending with a variety of

flexible polymers or rubbers [36, 77, 84-86].

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter I: State of the Art

Conclusion

In this part, it was reported that PLA is undoubtedly one of the most promising candidates

in the market of biodegradable polymers, for further developments. In fact, depending on its

structure and chemical composition, PLA can reach good physical, thermal and mechanical

properties. This makes it a good candidate to be used in several applications areas such as

packaging, electrical and electronic (E&E) and automotive.

However, to be used in some of these sectors (e.g. E&E), PLA has to be flame retarded,

which implies the incorporation of fire retardant fillers during PLA processing. Of course, these

flame retardant properties must be kept for the product lifetime. The next part is thus

dedicated to the different types of additives used to enhance flame retardant properties of

PLA.

2. Flame retardancy of PLA

Nowadays, PLA is primarily used in packaging and textile sectors, but its expected market

should be susceptible to extend to transportation and E&E sectors in order to substitute

thermoplastics like PP, PA or PET. For instance, Fujitsu is making injection molded computer

keys and computer housing made from polycarbonate (PC)/PLA blends [16]. Another recent

development which should enable wider applications of PLA in electronics products is NEC’s

process for imparting flame resistance to PLA without the use of halogen or phosphorous

compounds [8]. The product is reported to have heat resistance, moldability and strength

comparable to fiber-reinforced polycarbonate used in desktop type electronic products.

According to these PLA applications, fire hazard is an issue and flame retardancy of PLA is

required. That is why the interest for developing flame retardant PLA kept growing these past

few years; more than 110 papers are dedicated to FR-PLA in 2014. Literature revealed that the

flame retardancy of PLA can be obtained by different ways such as: (i) the use of conventional

FR additives, (ii) nanoparticles, (iii) combination of both, and (iv) intumescent additives [8]. In

this section, some basics on flame retardancy as well as the different approaches used to

flame retard PLA will be developed.

Basics on flame retardancy of polymers

As polymeric materials are made up mainly of carbon and hydrogen, most of them are

highly combustible [87]. This is why it is primordial to flame retard them to decrease risks in

limiting fire spread. The combustion of polymers is a very complex process which has already

been widely studied in the literature [88, 89].

The combustion reaction of polymers involves three components: one or more combustible

materials (polymers), a combustion agent which can be an oxidizing agent (generally oxygen

in air) and heat. The whole process usually starts with an increase in the temperature of the

polymeric material due to an external heat source, coming from either an inner (e.g.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter I: State of the Art

exothermicity of the combustion reaction) or outer heat source. Polymer bond scissions are

then induced and the resulting volatile fractions of polymer fragments diffuse into air and

create a combustible gaseous mixture. Then ignition occurs. The combustion of a polymer is

the result of a combination of the effects of heat, represented by two mechanisms [87] (Figure

17):

- Non oxidizing thermal degradation (pyrolysis): initiated by chain scission under the

effect of temperature increase

- Oxidizing thermal degradation: reaction with oxygen to produce a variety of low

molecular weight products.

Figure 17: Combustion of a polymer [90].

The concept of “flame retardancy” was introduced in order to inhibit or stop the

combustion process of polymeric materials, by modifying the “actors” of fire development.

Flame retardants are efficient at improving the fire safety of plastics; they can increase

resistance to ignition, reduce the rate of burning, flame spread, smoke emission and minimize

dripping during combustion [91]. On the contrary, flame retardants can negatively affect

material properties, producing unwanted side effects and add unnecessary cost. Hence, this

highlights the difficulty of developing an efficient flame retarded system.

To flame retard a polymer, various methods can be used like the chemical modification of

the polymer itself. As an example, it is possible to copolymerize the flame retardant directly

with the monomers to yield an intrinsic FR polymer [92, 93]. The use of inherently flame

retardant polymers such as poly(tetrafluoroethylene) or polyimides is another possibility to

inhibit the combustion of a material [94, 95]. Surface coating is also used to protect polymeric

materials against the influence of an external heat source [96-99].This kind of flame retardant

method has some benefits, as the material bulk is not modified during treatments, but

adhesion issues can reduce the FR properties of the coating. The last and most widely used

method to protect polymers consists in incorporating flame retardants into the polymer

during its processing (e.g. extrusion). This is the cheapest method to fire retard polymers. The

following part will thus only deal with bulk fire retarded PLAs. The fire retardants added in the

PLA matrix can have physical and/or chemical mode of action and can act in the gas or

condensed phase. These modes of action are detailed hereafter.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter I: State of the Art

2.1.1. Physical action

The physical action of fire retardant additives takes place through the formation of a

protective layer, by cooling and/or dilution. In the first case, a protective layer of low thermal

conductivity is formed when the material is submitted to an external heat source, which

protects the material through reduction of the heat and mass transfers between the external

heat source and the material [50, 100]. In the second case, materials can be protected by

cooling through the endothermically decomposition of fire retardant fillers, leading to a

decrease in temperature by absorbing the external heat [101]. Concerning the dilution, the

protection takes places through the addition of compounds releasing inert gases (e.g. carbon

dioxide, water) upon decomposition. These kinds of additives dilute the fuel in the solid and

in the gaseous phases and thus lower the concentration of combustible gases in the

surrounding atmosphere.

2.1.2. Chemical action

Chemical mode of action is the other possibility by which polymers can be protected. This

mode takes place in the condensed and/or gas phase. Two kinds of reactions can occur in the

condensed phase: (i) the formation of a carbon layer (charring) on the polymer surface,

allowing to limit the volatilization of the fuel but also the oxygen diffusion and insulating the

polymer underneath from the external heat [102]; or (ii) the acceleration of the polymer

decomposition leading to a pronounced flow of the polymer. Hence, a withdrawal from the

sphere of influence of the flame that breaks away. Charring is the most common condensed

phase mechanism [103, 104].

In the gaseous phase, the reaction takes place via interruption of the radical mechanism of

the combustion process by flame retardants or by their decomposition products. Therefore,

the exothermic processes that occur in the flame are stopped and the system cools down.

Hence, the supply of flammable gases is reduced or even completely suppressed. In particular,

halogens can act as flame inhibitors [105].

It is possible that flame retardants such as magnesium hydroxide (MDH) show exclusively

a physical mode of action. However, flame retardants acting exclusively through a chemical

mode of action are rare. In fact, one or several physical mechanisms generally take part in

chemical mechanisms, which are commonly endothermic dissociation or dilution of fuel.

Flame retardancy of PLA by conventional flame retardants

FRs additives are mainly constituted by metal hydroxides (aluminium trihydroxide (ATH),

magnesium hydroxide (MDH) …), phosphorus-containing compounds, halogenated

compounds and nitrogenated compounds which act in gas and/or condensed phase to provide

low flammability of the material [8, 102]. These conventional FRs have been widely evaluated

in PLA over the last few years [8, 9, 12-18].

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter I: State of the Art

Some researchers studied the use of aluminium trihydroxide (ATH) in PLA to make it flame

retardant [13, 14, 16]. Even if the mechanism is not completely elucidated by the authors, the

fire behavior of the material is similar to typical thermoplastics flame retarded with ATH. For

instance, Kimura et al.[16] demonstrated that the use of ATH allows increasing the UL 94-V

(3.2 mm) value of PLA from non-classified to V-2. Flame retardancy of the material is in his

case due to physical effects requiring relatively large amount of ATH (between 50 and

65 wt.- %). Bourbigot et al. [101] reported that the activity of ATH consists in:

a) The dilution of the polymer in the condensed phase

b) Decreasing the amount of available fuel

c) Increasing the amount of thermal energy needed to raise the temperature of

decomposition to the pyrolysis level, due to the high heat capacity of the fillers

d) Emission of water vapor - enthalpy of decomposition

e) Dilution of gaseous phase by water vapor – decrease of amount of fuel and oxygen in

the flame

f) Possible endothermic interactions between the water and decomposition products in

the flame

g) Decrease of feedback energy to the pyrolysing polymer

h) Insulative effect of the oxides (Al

) remaining in the char

i) Charring of the materials

Others researchers such as Yanagisawa et al. [14] have recently studied the flame

retardancy of PLA (at UL 94-V (3.2 mm)) using high load ATH (50%) in combination with

phenolic resins. Applications was housings of electronic products where the required level of

flame retardancy is high. The action mechanism of the flame retarded system is reported to

occur via the formation of a homogenous char layer, resulting from the phenolic resins

degradation at the surface of the material upon heating. This latter is not fully described in

this paper but an assumption concerning the mechanism of action is that phenolic resin should

provide high char yield, which could be reinforced by the formation of aluminum oxide from

ATH. Hence the ceramic char becomes then protective enough to yield good FR properties to

PLA.

Textile is a major application for PLA and it is also required to achieve good level of flame

retardancy. In this way, Kubokawa et al [15, 17, 18] have investigated the flame retardancy of

PLA fabrics using bromine and triphenylphosphate (TPP), as flame retardants. It was found

that limiting oxygen index (LOI) increases from 24 vol.-% to 28 vol.-% with TPP and to 26 vol.- %

with the bromine-containing FR. According to the authors, TPP FR treatment promotes

charring, suggesting a mechanism of action in the condensed phase. In the case of brominated

treatment, action in gaseous phase should occur.

Other phosphorus based FRs are also efficient to achieve low flammability of PLA. For

instance, PLA material filled with 30 wt.-% of phosphinate-based FR (OP1311 of Clariant,

Germany) is V-0 classified at UL 94-V (3.2 mm) with a LOI value superior to 30 vol.-% and a

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter I: State of the Art

decrease of the peak of heat release rate (PHRR) of around 15% observed by cone calorimetry.

As well as OP1311, Melamine polyphosphate (MP) presents good fire performances in PLA

matrix [10]. In fact, V-0 rating (3.2 mm) is achieved at UL 94-V for 20 wt.-% of MP and the peak

of heat release rate (PHRR) determined by cone calorimetry (35 kW/m²) is decreased by 33%.

As described in this section, many conventional flame retardants can be used to achieve

flame retardancy of PLA. However, literature report that the interest for nanoparticles to

flame retard PLA kept growing these past few years. The next section is thus dedicated to the

flame retardancy of PLA using nanoparticles.

Flame retardancy of PLA by nanoparticles

One interest in the use of nanoparticles is the development of efficient

polymer/nanoparticles systems at very low loading of nanoparticles (less than 10 wt.-%). The

nanocomposites exhibit remarkably improved properties (e.g. mechanical …) as compared

with those of virgin polymer [106]. It was reported that PLA nanocomposites exhibit a high

storage modulus both in solid and melt states, an increased flexural properties, a decrease in

gas permeability, an increased heat distortion temperature, an increase in the rate of

biodegradability of PLA, and also a decrease of reaction to fire of some polymers [107].

Indeed, it was demonstrated that adding a small amount of clay into the polymer matrix

allows reducing pHRR for numerous polymer nanocomposites [108, 109], especially PLA. The

use of nanoparticles is then an interesting way to flame retard plastics while reducing both

cost and quantities of FR fillers used.

Bourbigot et al. [20] reported that 3 wt.-% of montmorillonite (MMT) platelets dispersed

into PLA matrix reduced significantly the pHRR by about 40%. In the case of pure PLA, vigorous

bubbling is observed when the polymer is burning and no residue is left at the end of the

experiment. But when the clay nanocomposite burns, char formation is observed. This char

serves as a barrier limiting both mass and energy transport. In addition to clay, multi-walled

carbon nanotubes (MWNT) are also known to significantly improve the fire behavior of

polymers when they are nanodispersed in the polymer matrix [10, 21, 110]. PLA/MWNT

exhibits limited enhancement in cone calorimetry [20] but the flame spread is much lower

compared to virgin PLA when testing samples in vertical position at UL-94 [61]. Virgin PLA

flows, drips, and burns quickly while PLA/MWNT does not flow and does not drip. It is

suspected that MWNT incorporated in PLA increases the apparent viscosity of the material

and so, it avoids dripping as it was previously evidenced by Kashiwagi et al. [21]. PLA/MWNT

burns very slowly, keeping in an upright position but the ignition does not stop. This test then

shows the interest to incorporate MWNT in PLA.

Graphite that has a two dimensional structure as clays, is another nanoparticle that has

already shown great potential when incorporated in polymers. Dispersed in the nanosize

range in polymeric matrix, graphite can improve overall properties of the system as clay does

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter I: State of the Art

[111, 112] but can also provide its own characteristic properties of thermal and electrical

conductivity. Murariu et al. [22] incorporated expanded graphite (EG) at different loadings

into PLA. Whereas PLA easily burns and drips without charring when exposed to a flame in

horizontal position, PLA/EG nanocomposites at 6 wt.-% loading burn with no dripping and

present a decrease by 30% of pHRR in cone calorimetry experiment. It is shown that the

nanocomposite swells during burning in order to form a porous foamed carbonaceous char

that acts as a barrier to limit the combustion. The high amount of char formed during burning

is ascribed to additional expansion of graphite nanolayers. This expansion is provoked by both

the heat and gases/products of combustion. Hence, authors reported that the PLA/EG system

presents an “intumescent-like” behavior.

Intumescent PLA

“Intumescence” comes from Latin “intumescere” meaning “to swell up”, which describes

rightly the behavior of an intumescent material. Indeed, when heating an intumescent system

beyond a critical temperature, the material begins to swell and then to expand resulting in the

formation of a foamed cellular charred layer on the surface. This latter protects the underlying

material from the action of the heat flux or the flame [50, 113]. The concept of intumescence

was applied to flame retard polyesters such as PLA and it was proven to be efficient for this

kind of polymers.

An intumescent system is composed of different ingredients which are usually (i) an acid or

a material yielding acidic species upon heating, (ii) a char former and (iii) a component that

decomposes at the right temperature and at the right time to enable the blowing of the

system. Concerning the acidic source, acids, ammonium salts, and phosphates can be used.

Char formers are mainly hydroxyl-containing compounds, such as polyols and starch and

blowing agents can be Melamine (MEL) compounds which can release gases upon heating, in

the temperature range corresponding to the development of the intumescence process.

A typical example of intumescent system is the combination of ammonium polyphosphate

(APP) with pentaerythritol (PER) in which APP plays the role of both acid source (yielding

phosphoric acid) and blowing agent (releasing NH

) and PER is the char former [114].

Researchers such as Reti et al. investigated the substitution of PER by lignin (LIG) or starch (ST)

in the APP/char former system [9]. At 40 wt.-% loading of APP/char former (PER, LIG or ST),

high LOI values are observed (up to 58 vol.-% with PER) and in all cases an intumescent

protective layer was formed at the surface of the material. The interesting features of the bio-

based systems are V-0 classification at the UL-94 test (3.2 mm) while PLA-APP/PER is only V-2

rated. Under an external heat flux of 35 kW/m² exposure on cone calorimetry experiment,

those systems exhibit a reduction of pHRR by 65% for APP/PER and by 40% for APP/LIG and

APP/ST. An intumescent layer protecting the underlying material is formed under the heat

flux, limiting heat and mass transfers.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter I: State of the Art

In order to reinforce mechanical properties of the polymers, fibers can also be incorporated

in the intumescent material. Shumao et al. [23] combined natural fibers (ramie fibers) with

APP to make an intumescent system for PLA. At 40 wt.-% loading, LOI reaches 35 vol.-% and

the FR system exhibits V-0 rating at UL-94 (3.2 mm). Authors report that APP releases

phosphoric acid, polyphosphoric acid and non-flammable gases when exposed to flame.

Hypothesis of mechanism is that the resulting acid contributes to intramolecular or

intermolecular ramie dehydration, then dehydrogenizing, charring and rupture of chemical

bonds.

Conventional flame retardants and nanoparticles: synergy in PLA

Polymer nanocomposites exhibit low flammability in terms of cone calorimetry but they

fail to pass other tests, in particular when samples are in vertical position (e.g. LOI, UL-94) [8].

The mechanism of protection involves the formation of a char layer covering the entire sample

surface, acting as insulative barrier and reducing volatiles escaping to the flame. The formation

of such a layer which does not crack during burning is critical to obtain low heat release rate

from nanocomposites [102]. Hence, nanoparticles have been combined with flame retardants

fillers in order to see if a synergistic effect is obtained.

As shown in the section above, intumescence is one of the efficient methods existing to

limit the flammability of PLA. In our group, Fontaine and Bourbigot [12] investigated the

combination of APP with Melamine (MEL) (30 wt.-% in a 5 to 1 ratio respectively) as efficient

intumescent combination in PLA (namely FR-PLA). They demonstrated that the incorporation

of an additional nanofiller such as organomodified montmorillonite (e.g. cloisite 30B (C30B))

presents positive effects on reaction to fire of the materials meaning that APP, MEL and C30B

have a positive synergistic effect in term of fire retardancy in PLA. The intumescent PLA

containing C30B does not burn and the small peak of heat release is only due to a very short

ignition of the material which goes off very quickly. At 30 wt.-% loading of APP/MEL/C30B in

PLA (namely FR-PLA-C30B), the material achieves very good fire performances. Indeed, high

LOI values up to 52 vol.-% and V-0 rating (3.2 mm) were measured. Moreover, FR-PLA-C30B

exhibits a large decrease of about 90% of the pHRR measured by cone calorimetry (35 kW/m²).

These great performances were partially explained by the increase in thermal stability of the

FR-PLA-C30B formulation compared to FR-PLA. Cloisite 30B can react with APP to form

alumino- and silicophosphates stabilizing the intumescent structure at high temperature [115,

116]. The intumescent PLA was also evaluated with a total loading of 10 wt.-%. This loading is

still sufficient to get relatively high LOI (33 vol.-%) and a synergistic effect is also observed with

C30B (35 vol.-%). It is outstanding that V-0 at UL-94 is achieved for the formulation with and

without Cloisite 30B. This study exhibiting very good results in terms of flame retardancy

shows that combination of nanoparticles and intumescent agents offers promising way to

design flame retarded PLA.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter I: State of the Art

Conclusion

This part has reviewed the state of the art of the flame retardancy of PLA. The basics of

flame retardancy were firstly reminded, and a focus of the methods providing flame

retardancy to PLA was detailed. Very high performances in terms of fire behavior can be

reached in certain cases (e.g. intumescence).

PLA was mainly used for non-durable applications but it is turning to durable ones. Hence,

because of the growing market for PLA for durable applications, flame retardancy is required

and of course these flame retardant properties must be kept for the product lifetime.

However, due to its structural and chemical composition as well as its bio-degradable

character, PLA is very sensitive to ageing and chemical degradation such as hydrolysis or

photo-degradation. Whereas many papers are related to the ageing of raw PLA under various

constraints, ageing of filled PLA has not been widely studied yet.

The next part is therefore focused on the ageing of polymeric materials, especially chemical

ageing and associated mechanisms of degradation. A particular attention is paid to the ageing

of PLA and its consequences on the properties of this material. Moreover, some examples of

flame-retarded polymers are given in order to understand the effects of ageing on the fire

behavior and thereafter correlate then to our FR systems.

3. Ageing of polymers

During their lifetime, polymeric materials are submitted to different constraints depending

on their environment of use. In outdoor applications, polymers undergo principally the action

of light, oxygen, water, pollution, heating, etc. An important number of papers deals with

ageing issues related to polymeric materials. This is explained by the necessity of

understanding the failure observed, in order to prevent (or ensure) the lifetime of the

materials. This is also explained by the multiplicity of systems, ageing tests (natural and

accelerated) as well as analytical techniques.

In materials science the term ageing is used when engineering properties, physical or

chemical characteristics of a material change over a period of time. It may be useful to

distinguish physical and chemical ageing. In the first case, there is no alteration of the chemical

structure of the macromolecules, only the spatial configuration or composition of the material

is affected. In the second case, the chemical structure of the macromolecules is modified. All

ageing mechanisms at molecular level lead to changes in morphology and macroscopic

properties. Sometimes a number of age-related phenomena operate simultaneously and/or

interactively [117, 118]. Usually, effects of ageing are not desired in advanced materials

(except maybe for biodegradation) and therefore understanding and fighting ageing

phenomena play an essential part in polymer science.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter I: State of the Art

This part gives an insight in the common mechanisms involved in chemical ageing of

polymeric materials, including change of properties due to external stimuli such as

temperature, humidity and ultra-violet (UV) radiations. Then, as the objective of these

researches is to study the ageing behavior of fire retarded PLA, a special focus will be done on

the ageing of PLA and flame retarded polymers. The first section is dedicated to the

generalities of chemical ageing of polymers.

Chemical ageing of polymers

Chemical ageing is certainly the most important cause of ageing, involving irreversible

changes in the molecular structure of the polymer chains. In the case of chemical ageing, two

different situations can be distinguished, whether it occurs in presence or absence of oxygen.

For instance, chemical ageing without oxygen involves reactions such as hydrolysis,

depolymerization or elimination, radiochemical and photochemical processes. In presence of

oxygen, one can mention thermo-oxidative or photo-oxidative ageing. In that case, the role of

oxygen is really crucial. In fact, it contributes to the chemical reactions leading to the ageing

of the material. Consequently, chemical ageing generally leads to chain scission phenomena,

depolymerization/elimination or cross linking. Figure 18 presents the most usual parameters

involving chemical ageing as well as their consequences. These one then lead to the

deterioration of the characteristics of polymers (e.g. mechanical, physical …), to a color change

and to the appearance and growth of cracks at the surface that sometimes result in

destruction of the polymer [118-121]. Others parameters such as immersion in liquids, acids,

oils, etc can induce chemical ageing [30, 122, 123]. These additional factors are not

investigated in this PhD thesis and thus are not detailed further in this chapter.

Figure 18: Main chemical ageing factors and consequences on polymeric materials.

3.1.1. Chemical degradation mechanisms

There are many different types of possible chemical reactions causing degradation of

polymers, however the most common ones are photo-chemical, thermal, hydrolytic and bio-

chemical reaction. These different types of chemical degradation mechanisms are reported

and detailed hereafter.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter I: State of the Art

Generally, four common mechanisms of polymer degradation are identified (Figure 19).

These reactions can be divided into two groups: (i) those involving atoms in the main polymer

chain (chain scission and crosslinking) and (ii) those involving principally side chains or groups

(side chain elimination and side chain cyclization). The degradation of polymers can be

explained by one of these general mechanisms or by a combination of some of them [124].

Figure 19: General degradation mechanisms, adapted from Beyler, at al. [124].

The most common reaction mechanism involving the break of bonds in the main polymer

chain is usually called “chain scission”. It may occur at the end or at random locations in the

polymer chain. End chain scission results in the production of monomer, and the process is

commonly called “unzipping”. Random-chain scission generally results in the generation of

both monomers and oligomers as well as in some other chemical species.

Another general chemical degradation mechanism is “Cross-linking” that involves the main

polymer chain. It generally occurs after some stripping of substituents and involves bonding

between two adjacent polymer chains [124].

Elimination and cyclization are the two main reaction types involving side chains or groups.

In the case of elimination reactions, the bonds connecting side groups of the polymer chain to

the chain itself are broken, the side groups often reacting with other eliminated side groups.

The products resulting from elimination are usually small and thus volatile. In cyclization

mechanism, two adjacent side groups are bound together, resulting in a cyclic structure.

These degradation mechanisms lead to “a deterioration of any property of the polymer”

[125]. In fact, the degradation of polymers by chain scission generally leads to a decrease of

the molecular weight of the material as the polymer chains are broken down to smaller units

[121]. Thus leading to a decrease of some properties such as glass transition and melting

temperatures of the material [27, 29, 34, 35, 52, 123, 126]. Moreover, the crystallinity of the

polymer can be affected by degradation during ageing. For instance, in semi-crystalline

polymers such as PLA, the amorphous phases are more susceptible to be degraded than

crystalline ones. The degradation of amorphous phases leads to an increase in crystallinity of

the material. In fact, chain scission occurring in amorphous phases of PLA produces molecule

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter I: State of the Art

segments which can integrate crystalline phase if they have sufficient mobility and will thus

increase the crystallinity of the material. This phenomenon is called “chemi-crystallization” [5,

38, 127]. Mechanical properties of polymers such as tensile strength or Young’s modulus are

also usually affected by the ageing of the material [29, 52, 123, 128]. Some other effects of

ageing are reported in literature such as yellowing [129] or formation of cracks at the surface

of the material [32, 121] but generally the effects of ageing are dependent on the nature on

the material but also of the external stimuli applied to the polymer.

3.1.2. Photo-chemical degradation

Photo-degradation is a phenomenon inducing degradation of a photodegradable molecule

by the absorption of photons, particularly those from wavelengths found in sunlight, such as

infrared radiation, visible light and ultraviolet (UV). Photons are considered as one of the

primary sources of damages exerted upon polymeric substrates at ambient conditions.

It is reported that photo-degradation is able to change the physical, chemical and optical

properties of polymers. Thereby the loss of mechanical properties, the changes in molecular

weight and visual effects like “yellowing” are the most damaging effects which can be

observed [28, 34, 35, 129-132].

A common photo-degradation reaction is photo-oxidation, which is the process of

decomposition of the material by the combined action of light and oxygen, which can be

accelerated at elevated temperature [29, 129]. Most of synthetic polymers may be degraded

by UV and visible light. The near-UV radiations (between 290 and 400 nm) in the sunlight

determine the lifetime of polymeric materials in outdoor applications [133-135]. Photo-

oxidation process can also be photo-induced by chromophoric impurities (catalyst,

peroxides …).

UV radiations have sufficient energy to cleave C-C bonds [125]. The most damaging UV

wavelength for a specific plastic depends on the bonds present and the maximum degradation

therefore occurs at different wavelengths for different types of plastics, e.g. around 300 nm

for PE and around 370 nm for PP.

Focusing on the mechanism of action, photo-oxidative degradation of polymers includes

various processes such as chain scission, cross-linking and secondary oxidative reactions which

take place via a radical process [129, 136]. Many researches detail the mechanism of photo-

oxidation of polymers by involving the production of radicals and subsequent reactions with

oxygen [125, 129, 137, 138]. This mechanism proceeds through a chain reaction into steps

gathering initiation, propagation and termination. These steps are detailed below.

3.1.2.1. Photo-initiation step

The absorption of UV light that has sufficient energy to break the chemical bonds in the

main polymer chain is responsible for the initiation of the photo-oxidation mechanism. In fact,

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter I: State of the Art

the polymer (PH) absorbs light and produces low molecular weight radicals (R

●

) and/or

polymeric macro radicals (P

●

) as follows (Equation 4).

Equation 4

The photo-initiation step may be initiated by various physical factors such as UV radiation,

heat, ionization and chemical factors (e.g. direct reaction with O

or atomic oxygen …). In

photo-oxidation process, hydroperoxides (POOH) are the most significant and most

widespread initiators. Whatever the initial mechanism of radical formation, hydroperoxides

are created just after reaction with oxygen and are key intermediates in the oxidation of

polymers (Equation 5 and Equation 6) [129].

Equation 5

Equation 6

As shown in Equation 7, hydroperoxides usually decompose to form radicals (PO

●

and

●

OH)

that can abstract hydrogen atoms from the polymer and thus initiate photo-oxidation. In this

way, hydroperoxides are extremely photolabile [136].

Equation 7

The formation of hydroperoxide and its photolysis illustrate the formation of other

effective functional groups such as carbonyls during β-scission of the alkoxy radical [129].

Besides the initiation step by photolysis of hydroperoxide groups, ketone photolysis is the

second major contributor to the photo-degradation of polymers. It proceeds through two

major reactions known as Norrish I and Norrish II (Figure 20) [125, 129, 138-140].

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter I: State of the Art

Figure 20: Norrish I and II mechanisms.

As described in Figure 20, ketones present onto the polymers backbones are able to absorb

photons of appropriate energy when they are exposed to light. Thus, C-C bonds break and

scission of the polymer backbone occurs [129, 138].

3.1.2.2. Propagation step

According to Rabek et al. [141] as well as Yousif et al. [129], propagation can be divided

into six different steps:

1. Subsequent reaction of low molecular radicals (R

●

) or polymer alkyl radical (P

●

) in a

chain process similar to the abstraction of hydrogen from the polymer molecule

(Equation 8).

Equation 8

2. Reaction of the polymer macro radicals with oxygen which forms polymer peroxy

radicals POO

●

(Equation 5).

3. Abstraction of a hydrogen from the same or from another polymer molecule (‘PH) by

polymer alkyl-peroxy radical (POO

●

) with the formation of a hydroperoxide group

(Equation 9).

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter I: State of the Art

Equation 9

4. Photo-decomposition of hydroperoxide groups with the formation of polymer alkyloxy

(PO

●

), polymer peroxy (POO

●

), and hydroxyl (

●

OH) radicals (Equation 7 and Equation

10).

Equation 10

5. Abstraction of hydrogen from the same or another polymer molecule by polymer

alkyloxy radical (PO

●

) with formation of hydroxyl function groups (-OH) in polymer

(Equation 11).

Equation 11

6. Scission process of polymer alkoxy radicals (CH

CHO

●

) with the formation of aldehyde

end groups and end polymer alkyl radicals (Equation 12).

Equation 12

Generally, the overall processes of photo-oxidative degradation can be simplified by photo-

oxidation cycle as described in Figure 21.

Figure 21: simplified photo-oxidation cycle of polymers, adapted from Yousif et al [129].

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter I: State of the Art

3.1.2.3. Termination step

The termination of photo-degradation is achieved by “mopping up” the free radicals to

create inert products [125]. In fact, the radicals can be terminated by numerous different

combination reactions between two polymer radicals, as detailed in Equation 13, Equation 14

and Equation 15:

Equation 13

Equation 14

Equation 15

The polymer radicals may combine to give crosslinked or branched product [125]. If the

oxygen pressure is relatively high, the termination follows Equation 13 and Equation 14. In

comparison, if the oxygen pressure is low, other termination reactions take place to some

extent. When sufficient oxygen content cannot be maintained during the degradation,

Equation 14 becomes significant. Finally, as shown in Equation 15, polymer radicals can

recombine together [129].

As developed in this part, photo-oxidative degradation takes place via a radical process. As

well as photo-chemical degradation, thermal degradation proceeds through a radical process

[129, 136]. In fact, under normal conditions, photo-chemical and thermal degradations are

similar and classified as oxidative degradation [125]. The next part will be focused on the

thermal degradation of polymers as well as the description of mechanisms of degradation.

3.1.3. Thermal degradation

Generally, temperature is one of the parameters that plays a major role in chemical ageing.

Thermal degradation of polymers can be defined as “molecular deterioration as a result of

overheating”. At high temperatures the components of the long backbone chain of the

polymer can start to break up (molecular scission) and react with another molecule to change

the polymer properties [142].

Under normal conditions, photochemical and thermal degradations are similar and are

classified as oxidative degradations. The main difference between both is the sequence of

initiation steps leading to auto-oxidation cycle. Another difference is that thermal reactions

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter I: State of the Art

occur throughout the bulk of the polymer sample, whereas photochemical reactions generally

occur at the surface [143].

Another important fact concerning the thermo-oxidation is the chain scission

phenomenon. In fact, crosslinking and chain scission have both consequences on the polymer

properties [118, 144]. The degradation via crosslinking includes hardening, skinning, gel

formation, a decrease in tack, and an increase in viscosity. Degradation via chain scission

results in softening, a decrease in viscosity, an increase in tack, and a loss of cohesive strength.

In addition, discoloration also might occur upon oxidation. Even though discoloration normally

does not lead to significant changes in engineering properties, it might be not wanted for

application.

From a mechanistic point of view, as for photo-oxidative degradation the thermal

degradation in presence of oxygen is divided into three steps, i.e. initiation, propagation and

termination [125, 145]. The initiation step of degradation follows either chain end degradation

(also known as unzipping route) as shown in Equation 16, or random degradation route as

shown in Equation 17.

Equation 16

Equation 17

The chain-end degradation (depolymerization reaction) presented in Equation 16 starts

from the end of the polymer chain and consecutively releases monomer units from the chain-

ends. Similar to photo-oxidative degradation, it occurs via free radical mechanism. This kind

of reaction is the opposite of the propagation step in addition polymerization. During

depolymerization, the molecular weight of the polymer decreases and large quantity of

monomers is released simultaneously [125, 137]. The random degradation route presented in

Equation 17 occurs at any random point along the polymer chain. As for random degradation

to happen, the polymer chain does not require necessarily to carry active sites (sensitive to

radical reaction) such as for chain-end degradation [125]. During propagation step, the

monomer is formed as depicted in Equation 18 and Equation 19.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter I: State of the Art

Equation 18

Equation 19

Finally, termination takes place through radical coupling [125, 146] as previously reported

for photo-oxidative degradation (Equation 13, Equation 14 and Equation 15, p.50) but also

following Equation 20, Equation 21 and Equation 22.

Equation 20

Equation 21

Equation 22

During termination, crosslinking can form some larger molecules induced by reaction of

radicals P

•

with other radicals P

•

, alkoxy radicals PO

•

or peroxy radicals POO

•

. Alkoxy radicals

•

and peroxy radicals POO

•

can also react together to form POOP or POP molecules.

As photo and thermal degradation, hydrolytic degradation (hydrolysis) of polymers is one

of the most common and important ageing mechanism.

3.1.4. Hydrolytic degradation

Hydrolytic degradation occurs in polymers that have water sensitive active groups,

especially those that are very hygroscopic. Polymers that are concerned by hydrolytic

destruction usually have heteroatoms in the main or side chain of the polymer. In that case

we can also talk about hydrolysis of polymers, which involves the scission of chemical

functional groups by reaction with water as an example is shown in Figure 22. In fact,

hydrolysis of ether compounds leads to the formation of alcohols functional groups. The

degree of degradation will depend on both exposure time and temperature.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter I: State of the Art

Figure 22: Example of hydrolysis of polyether.

Hydrolytic degradation can be harmful but also helpful in some cases. For example,

hydrolytic chain scission of the ester linkage of polyester is the fastest and the most “harmful”

degradation process, as it causes a considerable reduction in the molecular weight and in the

mechanical properties of the polymer. However, in the case of drug delivery system, the

polymer degradation is useful [147].

Two basic types of hydrolytic degradation are discussed: surface erosion and bulk erosion.

During surface erosion mechanism, the mass loss of the polymer occurs at the water/polymer

interface (Figure 23 (a)) and in that case, the rate of hydrolysis is higher than the water rate

diffusion [147, 148]. As an example, most of the enzymatically resorbable polymers show

surface erosion. Surface-eroding polymers have a better ability to achieve zero-order release

kinetics and are therefore ideal candidates for developing drug delivery devices [149].

During bulk erosion, water diffuses rapidly into a polymer structure, leading to hydrolysis.

Then, the subsequent mass loss occurs throughout the bulk of the material (Figure 23 (b)). In

this case, the water diffusion rate into the polymer is higher than the hydrolysis rate. Hence,

the water that penetrates the substrate leads to hydrolysis from the inside to the surface [147,

148]. The sudden and rapid loss of strength and structural integrity as the resorption continues

over time is a characteristic behavior of bulk eroding polymers.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter I: State of the Art

Figure 23: Illustration of the hydrolytic mechanisms: (a) surface erosion, (b) bulk erosion

without autocatalysis, and (c) bulk erosion with autocatalysis, adapted from Viera et al.

[150].

Autocatalytic degradation is another factor that complicates the degradation behavior of

polymers [151]. The phenomenon of autocatalysis occurs when the reaction product is itself

the catalyst for the same reaction. In the case of bulk eroding polymers, the oligomeric

hydrolysis products (usually carboxylic and other acids) are retained within the material,

causing a localized decrease in pH which accelerates the rate of degradation [152]. As a result

of this autocatalytic hydrolysis, hollow structures are formed within the polymeric material,

which induce a rapid deterioration of the mechanical properties of the material as well as a

sudden loss of its structural integrity, as shown in Figure 23 (c). For instance, in the case of

polyesters such as PLA, when the thickness of the substrate is larger than 2 mm, the

entrapment and accumulation of acidic oligomers and monomers result in autocatalytic

degradation [153].

3.1.5. Bio-chemical degradation

The bio-chemical degradation, which is also known as “biodegradation”, is a biochemical

transformation of compounds by micro-organisms. Mineralization of organic compounds

yields carbon dioxide and water under aerobic conditions, and methane and carbon dioxide

under anaerobic conditions [154, 155]. Hydrolysis, photo-oxidation and physical disintegration

may enhance biodegradation of polymers by increasing their surface area for microbial

colonization or by reducing their molecular weight [154-156].

Biodegradability is also defined as the propensity of a material to get breakdown into its

constituent molecules by natural processes (often microbial digestion). The metabolites

released by degradation are also expected to be non-toxic to the environment and

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter I: State of the Art

redistributed through the carbon, nitrogen and sulfur cycles. Biodegradation is chemical in

nature (catalytic nature e.g. enzymes) but the source of attacking chemicals is from micro-

organisms. Researchers such as Singh et al. [125] or Lucas et al. [154] have reported that

biodegradation of polymers occurs via four different mechanisms: (i) solubilization, (ii) charge

formation followed by dissolution, (iii) hydrolysis and (iv) enzyme-catalyzed degradation.

Petrochemical-based plastic materials are not easily degraded in the environment because of

their hydrophobic character, additionally the three-dimensional structure interferes with the

formation of a microbial bio-film, leading to a reduced biodegradation extent [125].

After having identified and developed the different kinds of chemical ageing, this chapter

will now focus on the ageing of Poly(lactic) acid. Indeed, this researches are dedicated to PLA

and more precisely to the ageing of flame retarded (FR) PLA. In this context, it is crucial to

examine the ageing of the raw PLA material in order to understand the phenomena occurring

during ageing of filled PLA.

Ageing of Poly(lactic) acid

Aliphatic polyesters such as PLA currently deserve a particular interest as environmentally

degradable polymeric materials. However, due to its chemical composition, PLA is very

sensitive to ageing, leading to degradation of the polymer, such as hydrolysis [153], photo-

degradation [29], etc. Hence, the development of PLA based materials must find a

compromise between sustainability and performances. Therefore, this part will deal with the

ageing behavior of PLA under different kinds of chemical degradation.

3.2.1. Photo-degradation of PLA

Many studies have been led over the years, investigating the influence of photo-

degradation on PLA properties. It is reported for example that UV irradiation light decreases

the physical integrity and enhances the degradation of the material. Thus leading to a

decrease of both M

and mechanical properties (e.g. stress at break, percentage elongation

at break and strain energy). Focusing on thermal properties, a decrease of both T

and T

observed for PLA exposed to UV treatment [29, 33, 157]. As already mentioned, PLA chains

are photodegradable in both crystalline and amorphous regions but their photo-degradability

is higher in the amorphous regions [27] thus leading to an increase of the crystallinity of PLA

when irradiated by UV light.

At the state of the art, two main photo-degradation mechanisms of PLA have been

reported, depending on the UV irradiation wavelength. In fact, for irradiation under 300 nm,

the photo-degradation of Poly(lactic) acid proceeds via the Norrish II mechanism at ester and

ethylidene groups adjacent to ester oxygen, as shown in Figure 24 [27, 32, 33, 158, 159].

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter I: State of the Art

Figure 24: Norrish II mechanism for PLA photo-degradation: (a) PLA chain under UV

irradiation, (b) photophysical excitation and (c) oxidation/scission reactions in PLA chains,

adapted from Belbachir et al. [33].

This kind of mechanism causes the chain cleavage of the material, the decrease of the

molecular weight and the formation of carbon-carbon double bonds [159]. The photo-

degradation of PLA films under UV irradiation proceeds via a bulk erosion rather than a surface

erosion [159]. During photo-degradation, no formation of specific low-molecular-weight

peaks are observed, which strongly suggests that the chains in crystalline regions are

photodegradable even at early stage of degradation, in marked contrast with the case of

hydrolytic degradation [159].

Bocchini et al. [34] reported that Norrish II mechanism occurs when UV irradiation is under

300 nm, in a region where the carbonyl groups of aliphatic polyester can absorb energy and

consequently lead to photo-reaction. Conversely, Bocchini as well as many other researchers

[28, 29, 34, 35] evidenced a second mechanism of photo-degradation of PLA proceeding when

UV irradiation is in the range 300 to 400 nm [34]. In that, photo-degradation usually starts by

radicals formed from impurities during UV-irradiation or thermal decomposition (Figure 25).

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter I: State of the Art

Figure 25: Photo-degradation (radical oxidation) process of irradiated PLA: hydroperoxides

chain propagation and formation of anhydrides by photolysis of hydroperoxides, adapted

from Gardette et al. [35].

In this mechanism, the most probable reaction is the abstraction of tertiary hydrogen from

PLA chain, leading to the formation of a radical P

●

(Figure 25 (1)). This radical can react with

oxygen to form a peroxide radical (Figure 25 (2)), which may easily abstract another hydrogen

from a tertiary carbon with formation of a hydroperoxide and the initial radical P

●

(Figure

25 (3)). Then, the hydroperoxide undergoes photolysis (Figure 25 (4)) with the formation of

the

●

OH and PO

●

radicals that can further evolve by β-scission (Figure 25 (5)). Furthermore,

taking into account the stability of the different fragments the most probable β-scission

appears to be the reaction leading to the formation of anhydride groups (5a) [34, 35].

As well as photo-degradation, hydrolysis is another kind of degradation known to be very

harmful for the durability PLA based materials.

3.2.2. Hydrolysis and bio-chemical degradation of PLA

PLA like other polyesters is naturally hydrophilic due to its polar oxygen linkages. Its affinity

for water makes it very sensitive to hydrolysis (Figure 26) via an autocatalytic mechanism of

degradation [5, 31, 125, 153, 160]. In presence of water, ester is hydrolyzed into carboxylic

acid and alcohol according to Figure 26.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter I: State of the Art

Figure 26: Hydrolysis of PLA.

Hydrolysis of PLA is the first step of the biodegradation of the material which is composed

of two steps (Figure 27) [5].

Figure 27: Biodegradation of PLA in 60°C, adapted from Auras et al. [5].

As reported in Figure 27, PLA degradability is driven in a first place by hydrolysis and

cleavage of ester linkages in the polymer backbone. The hydrolysis of PLA is auto-catalyzed by

the carboxylic acid end groups. The process follows first order kinetics [5]. The duration of the

degradation is determined by the initial molecular weight of the polymer as well as by its

chemical structure [125]. As hydrolysis degradation mainly takes place in the amorphous

regions, degradation results in an increase in the crystallinity of PLA. A decrease of mechanical

and thermal properties (T

and T

) as well as of molecular weight is also observed during

hydrolysis of PLA [29-31, 38, 52, 121, 123, 128, 161, 162]. It can be accelerated by acids or

bases and it is also affected by both temperature [29] and moisture levels [66].

In the second step (Figure 27), the low molecular weight PLA resulting from the hydrolysis

is then able to diffuse out of the bulk polymer and be used by micro-organisms, yielding carbon

dioxide, water and humus [5, 125, 162].

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter I: State of the Art

3.2.3. Thermal degradation of PLA

As previously mentioned, temperature has a direct influence on the degradation of PLA.

Temperature is one of the main factors responsible for a fast reduction of M

. This decrease

leads to a reduction in both glass transition temperature and mechanical properties of the

polymer film. Whereas the degree of crystallinity increases due to chemi-crystallization

process [26, 29, 127, 157, 163, 164].

Mechanism of thermal degradation of PLA (from around 200 to 450°C) can be divided into

4 parts as follows: (i) the dominant reaction such as intra- and intermolecular ester exchange,

which leads to the formation of lactide and cyclic oligomers. (ii) The cis-elimination for

polyesters that yields less than 5% of acrylic acid and acrylic oligomers also occurs, but is not

at all a dominant reaction even at high pyrolysis temperatures. (iii) Unzipping

depolymerization is also observed. In fact, the lower the molecular weight, the more

concentrated the terminal hydroxyl groups, which accelerate the unzipping depolymerization

and the intermolecular exchange. (iv) Pyrolytic elimination of PLA which results in species

containing conjugated double bonds due to the carbonyl group [52].

Kopinke et al. [165] proposed that above 200°C, PLA can degrade through intra- and

intermolecular ester exchange, cis-elimination, radical and concerted non-radical reactions,

leading to the formation of CO, CO

, acetaldehyde and methylketene. In contrast, researchers

such as McNeil and Leiper [166] emitted the hypothesis that thermal degradation of PLA is a

non-radical “backbiting” ester interchange reaction involving the –OH chain ends. Depending

on the point in the backbone at which the reaction occurs, the product can be lactide

molecule, an oligomeric ring, or acetaldehyde plus carbon monoxide (Figure 28).

Figure 28: Thermal degradation mechanism of PLA, adapted from McNeill et al. [166].

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter I: State of the Art

Many researches have been led on the combination of temperature and other factors like

humidity, UV irradiation, oxidation and generally, the combination of these parameters

enhance the degradation rate of PLA [26, 29, 127, 157, 163, 164].

3.2.4. Synergy between temperature, relative humidity and/or UV light exposure

According to literature, temperature is a key parameters in chemical ageing of polymers as

it is able to catalyze reactions such as photo-degradation and/or hydrolytic degradation of PLA

[26, 29, 37, 38, 121, 123, 128, 167]. In fact, it is known that the combination between

temperature (50 to 60°C) and humidity cause PLA to degrade rapidly [29, 121, 123, 128]. In

this case, the increases of temperature favors the diffusion of water and thus increases the

hydrolysis of PLA as well as the loss of its physico-chemical properties (M

, T

…).

Moreover, researchers such as Copinet et al. [29] demonstrated the synergistic effect

obtained when combining temperature, UV-light and relative humidity. They emitted the

hypothesis that in conditions of humidity and temperature, UV treatment catalyzed the

hydrolysis of ester linkage. When temperature and relative humidity increase, effect of UV

treatment on the degradation of PLA matrix increases too, as it causes a faster autocatalysis

process during hydrolysis.

The ageing of PLA under temperature, humidity and/or UV irradiation generally leads to

some complex phenomena, as several ageing mechanisms can occur simultaneously. Whereas

ageing of raw PLA have been widely studied in this section, it is now useful to discuss

concerning the ageing of flame retarded polymers as our study is focused on the ageing of

flame retarded PLA.

Ageing of flame retarded polymers

Nowadays, the fire standard tests for polymer-based materials are currently carried out on

newly manufactured samples. However, various ageing conditions can impair flame

retardancy throughout the lifetime of a material [40]. Therefore, it seems necessary to

evaluate the influence of different kinds of ageing on the flammability of flame retarded

polymers.

It was previously reported that many ageing conditions promote structural modifications

of a polymer, like crosslinking and more generally chain scissions. For example, this latter leads

to a decrease of the molecular weight and then of the material viscosity. However, viscosity is

an important parameter controlling flammability. Indeed, in some fire tests, materials are

rated according to their ability to drip. Dripping is obviously more severe for low-viscosity

materials. Viscosity influences many other phenomena involved in flame retardancy:

accumulation and migration of fillers at the surface [168], formation of a protective layer

[169], intumescence [170], bubbling [171], etc. Moreover, low-molecular-weight molecules

are volatile and can decrease thermal stability and promote early ignition. Nevertheless, this

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter I: State of the Art

significant effect of ageing is rarely taken into account in studies dealing with the influence of

ageing on flame retardancy.

Thus, in this part, the influence of major chemical ageing scenarios (thermal, hydrolytic and

photo-chemical degradation) on the flame retardancy of polymeric materials is discussed.

3.3.1. Thermal degradation of FR polymers

The migration of low molecular-weight FR fillers to the surface is the first effect induced by

thermal ageing of FR polymers. Researchers such as Inata et al. [41] investigated the influence

of thermal ageing (at 110, 130 and 145°C) on the migration of 16 halogenated compounds

incorporated into polypropylene (PP). They reported that the weight loss of filled PP during

ageing is attributed to the migration of FRs. However, the migration is not only dependent on

the temperature and the molecular weight of the FR but also on its structure. Migration is

higher for aliphatic brominated additives, followed by molecules containing only one aromatic

ring. Molecules containing double aromatic rings, phosphate or isocyanuric groups hardly

migrate. In these experiments, thermal ageing was carried out at high temperatures.

Nevertheless, the authors have calculated the activation energy of migration to estimate the

migration rate at lower temperatures (corresponding to service life). In some cases, migration

can be significant. For 2,4,6-tribromophenyl-2’,3’-dibromopropymether (TBP-DP), which

exhibits the highest migration rate, the half-value period (corresponding to the loss of 50% of

additive) at 30°C is only 5000h, i.e. 200 days. LOI was also measured after ageing for some FRs.

It is reported that during ageing, the FR content decreases due to exudation and thus LOI

decreases too. These observations seem to indicate that the relation between LOI and FR

content is roughly linear.

Nanoparticles incorporated into FR polymers may also be concerned by the migration

under the effect of temperature. In fact, Colonna et al. [42] studied the effect of thermal

ageing on the combustion behavior of an ethylene–vinyl acetate (EVA) containing organically

modified clay nanocomposite. Thermal ageing was carried out close to, but below the melting

point of the polymer matrix (90°C). It was evidenced that during ageing, the nanoparticles

dispersion changed from intercalated to an exfoliated structure. Consequently, the fire

behavior of the EVA system was modified. It was observed that nanoclays move rapidly to the

surface, promoting the formation of a physical barrier preventing the development of flames

and thus delaying ignition on MLC experiment. However, the pHRR was increased significantly

from 430 to 657 kW/m² but no explanation is given in the paper to explain this phenomenon.

Thermal ageing was also reported by Zuo et al. [120] who studied the effect of thermo-

oxidative ageing on flammability of long glass fiber-reinforced PA6 composite containing

tris(tribromophenyl) cyanurate and antimony trioxide. The samples were aged for 10, 30 and

50 days under thermal-oxidative conditions at 160°C. UL-94 (3.2 mm) rating remains V-0 even

after 50 days of ageing. A positive effect of ageing is observed for LOI test (from 25.6 to

37.4 vol.-% after 50 days of ageing) and the pHRR observed in cone calorimetry decreases

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter I: State of the Art

from 108 to 72 kW/m². From the author’s point of view, the improvement of the flame

retardancy is explained by the enrichment of the surface in FRs due to migration. Thus, it

allows FRs to act sooner, leading to the reduction of the pHRR. Moreover, the charring effect

is promoted after ageing (from 31.5 to 35 % after 50 days of ageing) but the mode of action in

the gas phase is not modified even if the effective heat of combustion appears to slightly

decrease for 50 days of ageing.

3.3.2. Hydrolytic degradation of FR polymers

As previously reported, the presence of water can lead to dramatic changes in molecular

weight and viscosity, which are important factors controlling the flame retardancy. Recent

work investigating the influence of ageing in presence of water/moisture evidenced water can

modify chemically FR materials but also affect their flame retardancy.

Oztekin et al. [43] recently highlighted the influence of water on the flame retardancy of

poly(ether ether ketone) (PEEK). When a small amount of moisture is present in PEEK, water

is released just after the polymer melts (343°C). This release leads to vigorous bubbling and

results in shorter time to ignition (from 207 to 110 s at an irradiance of 50 kW/m² in cone

calorimeter) but lower pHRR (from 355 to 280 kW/m²).

Levchik et al. [44] studied the impact of hydrolytic degradation induced by artificial ageing

on the fire performance of flame retarded polycarbonate (PC)/acrylonitrile butadiene styrene

(ABS) blends. Three aryl phosphates, triphenyl phosphate (TPP), resorcinol bis(deiphenyl

phosphate) (RDP) and bisphenol A bis(diphenyl phosphate) (BDP) were tested as flame

retardants. They demonstrated that these three additives show comparable FR efficiency at

the same phosphorus level. However, under artificial ageing at 70°C and in the presence of

water (85% RH), hydrolysis of aryl phosphates leads to the release of acidic species, which

attack the polycarbonate. It results in a decrease of the fire retardant properties of the PC/ABS

blend compared to those before artificial ageing. This demonstrates that the integrity of FR

structure also plays a crucial role in the lifetime of the flame retardancy.

3.3.3. Photo-chemical degradation of FR polymers

It was reported that thermal ageing may promote migration of FRs. But while service

temperature is generally much lower than processing temperature (which sets the minimal

temperature of stability of FRs), it is doubtful that thermal ageing can degrade a FR structure.

On the contrary UV radiation can induce such modifications. For instance, it was pointed out

that viscosity is an important parameter controlling flammability. Therefore, flame retardancy

can significantly decrease after photo-ageing due to the detrimental effect of FRs on the

photo-stability of polymers [40].

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter I: State of the Art

It is known that FRs, and particularly halogenated ones, can promote fast ageing of

polymers. Papers have been devoted by Sinturel et al [45, 46] concerning the study of the

photo-oxidation of fire retarded PP. Authors reported that oxidation rate of PP increases in

the presence of 5 wt.-% decabromodiphenylether (decaDBE) and the induction period is

dramatically decreased. These effects are observed from 5% of FR and do not increase very

much for higher contents. It is evidenced that when exposed to UV radiation, decaDBE

degrades through the hemolytic cleavage of the carbon-bromine bond [46]. Hence, the

formation of a reactive bromine radical leads to hydrogen abstraction on the backbone and

promotes the degradation of the system.

Other types of FR do not show the same trends. Researchers such as Chantegraille et al.

[172] investigated a flame retarded PP with an intumescent system based on APP and a

synergist containing nitrogen. UV-light irradiation was carried out under polychromatic light

with wavelengths higher than 300 nm in a SEPAP 12.24 unit, in the presence of oxygen at 60°C.

The authors reported that the FR can be photo-oxidized but not degraded by photolysis in the

absence of oxygen. Moreover, even if the oxidation of FR starts much earlier than the

oxidation of PP, both oxidations occur independently and no influence of the FR on the photo-

oxidation of PP matrix was detected. Lonkar et al. [173] also studied the effect of 5 and

10 wt.- % of layered double hydroxide and organomodified montmorillonite on PP photo-

oxidation. Whereas good dispersion of nanoparticles was achieved, it is shown that photo-

oxidation of PP is significantly enhanced by the presence of organomodified montmorillonite.

The influence of layered double hydroxide depends on divalent cations. In fact, Mg

has a

degrading effect while Zn

does not accelerate the oxidation of the polymer. This study

demonstrates that metallic compounds can exhibit a catalytic effect on the degradation of

flame retarded polymers. These metallic impurities increase the decomposition rate of

hydroperoxide, thus, accelerating the overall radical process [34]. However, a high ratio of

metallic compounds should be needed to observe this catalytic effect.

Vahabi et al. [40] explained that flammability of a FR polymer may be mainly driven by the

properties of a thin surface layer and in all these cases, photo-chemical ageing may have an

influence on flame retardancy. It was demonstrated by Diagne et al. [47, 48] that PP grafted

maleic anhydride (MAH) exhibits better fire performances measured in a cone calorimeter

after UV irradiation. Ageing scenarios may promote the crosslinking of PP-MAH in a thin

surface layer (oxidation profile shows that UV radiation modifies only a 150 μm thick layer).

However, when nanoclays are incorporated into PP-g-MAH, flame retardancy is poorer after

ageing. Indeed, nanoclays can reduce the formation of radicals by attenuating the UV

penetration and by preventing radical mobility and thus, inhibiting crosslinking. Nevertheless,

it must be pointed out that the cone calorimeter results are unclear due to the large standard

deviation in the data. Others researchers such as Tidjani and Wilkie [174] also have

investigated the flammability and ageing of PP-g-MAH containing nanoclays. They showed

that nanoclays enhance the photo-oxidation of the material. The absorbance related to

ketones and carboxylic acids observed at 1715 cm

-1

by Fourier Transform Infrared

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter I: State of the Art

Spectroscopy (FTIR) increases faster than the one of PP-g-MAH without nanoclays. Moreover,

no significant changes were observed in cone calorimeter data after ageing. It is possible that

the enhanced photo-degradation of the material observed in presence of nanoclays results

from the degradation of the polymer matrix during the incorporation of nanoparticles, instead

of a catalytic effect of nanoparticles. Indeed, if the incorporation of nanoparticles decrease

the molecular mass of the material even before ageing, it will lead to a faster degradation of

the material.

Generally, UV ageing is combined with other sources of ageing like heat and water. Hence,

it is sometimes difficult to identify the main source of ageing. Chen et al. [175] observed a

strong decrease of LOI and UL-94 rating of PLA/ramie composites flame retarded by APP after

several days of weathering combining UV, heat and relative humidity. Ageing was carried out

under UV irradiation (UV-B313 fluorescent UV lamp) at 0.6 W/m², ageing cycle in two phases:

8 h at 60°C and 4h at 50°C for 7, 14 and 21 days (exposure to humidity is not specified). Even

if the reasons of the lower fire performances after ageing are not explained, it appears at least

that water as well as UV irradiation play a major role in degradation. Indeed, migration of APP

to the surface is assigned to its hydrolysis. APP is initially water-insoluble and becomes soluble

when its molecular weight decreases. The migration of APP leads to a heterogeneous

distribution of APP into the material. Moreover, authors reported that the coating of fibers by

APP is decreased, thus promoting the candelwick effect. Nevertheless, the synergistic effect

of UV radiation with water seems to be well proven [29]. Almeras et al [176] investigated

effect of ageing on the flame retardancy of a PP/polyamide 6 (PA6)/APP/ethyl vinyl acetate

(EVA

) blend. Ageing was carried out under a xenon lamp exposure (1.1 W/m²), rain and

temperature cycling for 200 h (18 min of wetting and 102 min of drying at 65 °C, 50% relative

humidity). Fire performances were measured using cone calorimeter (50 kW/m²), LOI and UL-

94 tests (3.2 mm). Whereas the unaged samples exhibited a LOI of 33 and were V0 rated in

UL-94 test, LOI decreasesd to 23 after ageing and aged samples were not classified at UL-94.

Concerning cone calorimetry, pHRR increases from 200 kW/m² for the initial sample to

300 kW/m² after ageing. The formation of a smaller intumescent protective structure is

reported. The decrease in terms of fire performances after ageing is attributed to the

degradation of APP by hydrolysis and to its migration (or to the migration of its degradation

products, e.g. ortho, pyro and polyphosphates of shorter chains). In fact, water diffusion into

the material and migration are both facilitated by the surface degradation. However, these

different modifications need the combination of UV and water.

The studies reported above show that UV irradiation can induce a decrease in flame

retardancy of polymers but other studies suggest that is not always the case. Larché et al.

[177] who studied the photo-oxidation of crosslinked EVA filled with aluminium trihydroxide

(ATH) and kaolin show that UV irradiation leads to a strong enrichment of the surface in fillers.

The hypothesis of fillers migration is excluded by the authors who explained this phenomenon

by the formation and release of low-molecular-weight products due to the EVA oxidation. The

composition profile shows that the enrichment occurs in the first 350 μm. In the first 100 μm,

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter I: State of the Art

the polymer content is less than half of its value in the bulk after 500 h of ageing (60°C,

irradiance in the range of 300-400 nm, 100 W/m²). Flammability was not measured but

authors were convinced that the enrichment of the surface would promote a barrier effect

and improve the flame retardancy. Some works have already shown that EVA filled with

mineral fillers exhibited better flame retardancy if these fillers migrate quickly to the surface

during cone calorimeter test [178].

Conclusion

In this part, it was reported that polymers and particularly PLA performances in terms of

durability are limited because of multiple chemical ageing mechanisms such as photo-

degradation, hydrolysis or thermal degradation. Chemical ageing is responsible for the

degradation of the material into low-molecular weight products (carboxylic acids, alcohols …)

and thus the decrease of the thermal and mechanical properties. Many papers are related to

the ageing of raw PLA under various constraints, but few of them describe ageing of filled PLA.

Concerning flame retarded polymers, published papers investigate mainly fire behavior

evolution but correlations between physico-chemical properties and fire performances after

ageing are rarely made. Nevertheless, the relationship between physico-chemical properties

and fire performances is a key parameter to investigate in order to develop durable flame

retardant systems. Actually, only few studies have been devoted to the ageing of FR polymers

generally filled with decaBDE or APP, which is clearly very few among all the FR fillers.

According to the wide range of ageing conditions, polymers and FRs, several phenomena can

be highlighted, impacting or not the flame retardancy of FR polymers.

4. Conclusion

Poly(lactic) acid is the first commodity biodegradable polyester produced from annually

renewable resources. Due to its low cost production and its interesting mechanical and

thermal properties, PLA is one of the best candidates among bioplastics. However, to be used

in transportation fields for example, PLA has to be flame retarded, and these flame retardant

properties must be kept for the product lifetime. Besides, due to its chemical composition and

its biodegradable character, PLA is sensitive to ageing and its environment of use. That is why

the development of PLA based materials must be based on a compromise between

sustainability and performances.

In this chapter, it was highlighted that many possibilities exist to fire retard PLA. The use of

conventional flame retardants, nanoparticles, and combination of both in PLA was detailed.

FR PLA (loaded at 30 wt.-%) usually exhibits good fire properties with (i) LOI increase up to

58 vol.-% [114], (ii) V- 0 classification achieved at UL-94 (3.2 mm) [10, 12, 114] and (iii) PHRR

decrease by 90% as observed by cone calorimetry [12].

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter I: State of the Art

It was also showed that polymeric materials and especially polyesters such as PLA are

extremely sensitive to ageing. Hence, an overview was given about the ageing and general

mechanisms of degradation of PLA under hydrolysis, thermal or photo-degradation. It appears

that during ageing, PLA turn into to low-molecular-weight products, leading to the decrease

in both thermal and mechanical properties but tending to increase the crystallinity of the

material [27, 29, 34, 35, 38, 125, 165, 166]. A focus was also done concerning the effects of

ageing on the fire behavior of flame retarded polymers, which is a complex subject depending

on the ageing conditions as well as on the nature of both polymer and FRs. Different

phenomena were evidenced during ageing of FR polymers such as migration [41, 42, 120] or

degradation and hydrolysis of fillers and nanoparticles [44, 175, 179]. Moreover, depending

on the authors, FR can protect or accelerate UV degradation of the FR material [45-48, 172,

173, 176].

As the development of PLA based materials must be a compromise between fire

performances and sustainability, the goal of this study is to characterize and understand the

ageing of flame retarded PLAs containing 30 wt.-% of FR additives (i.e. APP, MEL and C30B).

Up to now, only a few scenarios have extensively studied the flame retardancy of polymers.

Concerning PLA, whereas this polymer has been widely studied in terms of flame retardancy,

literature reveals that investigations about ageing of flame retarded PLA are still needed.

The strategy of this work is to study the effect of accelerated ageing in a raw PLA in

presence of temperature (T), relative humidity (RH), UV light and oxygen. Hence, the effects

on the properties of the raw material as well as on the mechanisms of degradation will be

determined. Then, the impact of the addition of FR fillers on the PLA properties, ageing and

degradation will be investigated. As previously evidenced by our team, Melamine, APP and

organoclays (C30B) have demonstrated their efficiency as FR in PLA intumescent system. The

impact of ageing on the fire behavior of both PLA and FR-PLAs will be finally assessed and

elucidated.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter II: Materials and Methods

This chapter aims at describing the presentation of the materials, techniques and

experimental protocols used in this work. In a first part, the materials, namely PLA, fire

retardants and nanoparticles are presented. The different formulations process is also

described. Apparatus used to investigate the behavior of materials under accelerated ageing

are then presented. Finally, the techniques related to ageing characterization and fire tests

are detailed.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter II: Materials and Methods

1. Materials preparation and processing

Materials

1.1.1. Poly(lactic acid) (PLA)

The polymer used during this study is an Ingeo PLA 4032D, a polyester supplied by

NatureWorks LLC (Ravago distribution center, Arendonk, Belgium). Before used, PLA was dried

overnight at 70°C. Properties of this PLA are reported in Table 4.

Table 4: Properties of Ingeo PLA 4032D.

Properties

PLA 4032D

Units

Residual monomer content

0.2

D-isomer content

1.5

Density

1.24

g/cm

Glass transition

temperature (T

)

61.5

°C

Melt Flow Index (MFI)

(190°C, 2.16 kg)

g/10 min

Melt density (at 230°C)

1.08

g/cm

± 68 000

g/mol

Tensile Strength, psi

7.7 (53)

MPa

Tensile Yield Strength, psi

8.7 (60)

MPa

Tensile Modulus, kpsi

500 (3.5)

GPa

Tensile elongation

Melting Point

155 - 170

°C

1.1.2. Flame retardants and nanoparticles

As previously mentioned, this study is based on the ageing of flame retarded PLA. Hence,

the flame retarded systems chosen are based on a previous work done in R

FIRE team by

G.Fontaine and S.Bourbigot [12]. A combination of Melamine, ammonium polyphosphate with

and without organomodified montmorillonite Cloisite 30B is incorporated into the PLA matrix.

Table 5 summarizes all additives that have been used in this study, their supplier and their

chemical nature.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter II: Materials and Methods

Table 5: Summary of additives used.

Name

FR type

Supplier

Chemical structure

Melamine 99%

(Mel)

Nitrogen FR

Sigma Aldrich

Exolit

AP422 (APP)

Ammonium

polyphosphate FR

Clariant

Cloisite 30B (C30B)

surfactant

Sheet-like

nanoparticle

(MMT modified

with methyl

tallow bis-2-

hydroxyethyl

ammonium

chloride)

Southern Clay

Product Inc.

Melamine (99%) was obtained from Sigma Aldrich (St Louis, MO). Ammonium

polyphosphate (APP, Exolit AP 422, soluble fraction in water < 1 wt.-%) in powder was supplied

by Clariant (Charlotte, NC). Nanoparticles, Cloisite 30B, a montmorillonite modified by bis-(2-

hydroxyethyl)methyl tallowalkyl ammonium cations, was supplied by Southern Clay products

(San Antonio, TX). The FRs and nanoparticles were dried for 24h at 70°C before use.

Material preparation and processing

1.2.1. Extrusion

Three different formulations, i.e. PLA, FR-PLA and FR-PLA-C30B (Table 6) were prepared by

using a co-rotative twin screw extruder from Thermo Fischer Scientific (Figure 29) (HAAKE

Rheomix OS PTW 16 twin screw extruder) with a barrel length of 400 mm and a screw diameter

of 16 mm (L/D = 25). Both FR-PLA and FR-PLA-C30B were prepared by incorporating 30 wt.-%

of FR fillers. Moreover, for the comprehension of the effects of ageing on flame retardancy of

PLA, two additional formulations were prepared with 20 wt.-% of FR fillers, i.e. FR-PLA-20%

and FR-PLA-C30B-20% (Table 7).

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter II: Materials and Methods

Figure 29: Thermo Scientific extruder.

Table 6: Composition and name of the formulations PLA, FR-PLA, FR-PLA-C30B at 30 wt.-%

PLA (wt.-%)

FR loading

(wt.-%)

Formulation

name

Mel (wt.-%)

APP (wt.-%)

C30B (wt.-%)

100

PLA

FR-PLA

FR-PLA-C30B

4.83

24.17

Table 7: Composition and name of the formulations FR-PLA, FR-PLA-C30B at 20 wt.-%

PLA (wt.-%)

FR loading

(wt.-%)

Formulation

name

Mel (wt.-%)

APP (wt.-%)

C30B (wt.-%)

FR-PLA-20%

3.18

16.14

FR-PLA-C30B-20%

3.18

16.14

0.66

The extruder is composed of a heater barrel divided in ten zones, in which the polymer

melts. The polymer, fire retardant additives and nanoparticles were added using a volumetric

side feeder. Before each experiment, a calibration of the volumetric feeders was performed

in order to obtain a good control of the composition of the final blends. Temperature profile

of the ten heating elements was set as reported in Table 8 and the rotational speed of the

screw was fixed at 100 rpm.

Table 8: Temperature profile of the extruder from hopper to die.

Zone

5**

Temperature

(°C)

205

200

195

190

185

180

175

165

*PLA was fed in zone 1, **FR and nanoparticles mixed together were fed in zone 5

At the end of the extruder, PLA and FR-formulations forms a strip which is cut into pellets.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter II: Materials and Methods

1.2.2. Molding

Mass loss cone calorimeter (MLC) plates of 100x100x3 mm

were obtained by pressing the

material, previously extruded and pelletized, with a hydraulic hot press in specific molds. Two

plates heated at 185°C in between a mold containing 35 g of formulation were used. A force

of 20 kN was applied for 2 minutes and then a force of 40 kN was applied for 8 minutes. Prior

being removed from the mold, each sample was cooled down with water cooling system until

it reaches 50°C. LOI samples size was 100x10x3 mm

were obtained by cutting them into MLC

plate using a bandsaw. Dimension of samples were selected according to the ISO-4589-2 and

ISO-13927 standards [180, 181]. Barrels and plates where then stored at room temperature

in sealed bags before tests.

1.2.3. Powders

A high speed rotor mill (Retch – Ultra centrifugal Mill ZM 200) was used to obtain fine

powder (< 500 μm) from the extruded pellets. The material remains in the grinding chamber

for a very short period and liquid nitrogen is used to cool down the sample, which means that

the characteristic features of the sample (i.e. thermal properties) to be determined are not

altered. Then the formulations were stored at room temperature in sealed bags before

analyses.

2. Accelerated ageing methods

As previously mentioned, one objective of this work is to screen the impact of ageing on

virgin PLA and both FR-PLA and FR-PLA-C30B, by submitting them to accelerated ageing tests

under constraints of their environment of use. Temperature (T), relative humidity (RH) and

ultra-violet light (UV) are the main parameters investigated during experiments. PLA as well

as both FR-PLAs were submitted for four months to three different kinds of ageing

experiments respectively; (i) temperature and relative humidity, (ii) temperature and UV, and

(iii) temperature, UV and relative humidity. Accelerated ageing tests are described and

detailed below.

Ageing in presence of temperature and relative humidity

In order to evaluate the impact of temperature and relative humidity on our materials, PLA

and both FR-PLAs were subjected to constant heating 50°C and relative humidity (75%) for

four months. Swelling, oxidation and hydrolysis of the materials should be accelerated by

combining temperature and relative humidity. This temperature/relative humidity treatment

will be further addressed to as T/RH treatment. For this purpose, a humidity chamber test

HCP 108 (Figure 30) supplied by Memmert was used to simulate ageing conditions.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter II: Materials and Methods

Figure 30: Humidity chamber test HCP 108.

The experiments were carried out on barrels, plates and extruded pellets which were

regularly (i.e. each week) taken out of the oven to be characterized. All samples removed from

the oven were dried overnight at 50°C before testing, in order to eliminate the excess of

humidity and to perform all characterizations in the same and controlled conditions. Effects

of each period of ageing on the behavior of PLA and FR-PLAs specimens were investigated by

comparing the changes of properties of aged materials to those of unaged material.

Ageing in presence of UV-rays

The accelerated weathering under UV-rays of PLA and both FR-PLAs was carried out by two

distinct experiments. The first ageing test was realized by exposing the samples to both

temperature and UV-rays (T/UV), and the second one consisted in exposing the samples to

temperature, UV-rays and relative humidity (T/UV/RH). For this purpose, two accelerated

weathering chambers from Q-LAB (Figure 31) were used: (i) QUV/se for T/UV and (ii)

QUV/spray for T/UV/RH.

Figure 31: Q-LAB, QUV/spray weathering chamber.

The weathering conditions chosen are derived from a ISO 4892-3 standard [182]. Both Q-

UV devices are equipped with 4 UV lamps (UVA 351) with 0.76 w/m² irradiance. Wavelengths

under 300 nm are filtered using these kinds of lamps. The Q-UV/spray tester is also equipped

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter II: Materials and Methods

with humidification unit, distilled water vessel, pump and piping system for circulation of the

water to the humidification unit.

Cycles of 4 hours UV irradiation at 50°C, followed by 4 hours dark at 50°C for T/UV or

4 hours dark and water condensation at 50°C for T/UV/RH were applied as listed in Table 9.

The specimens (extruded pellets, plates and barrels) attached to the test panels were exposed

to these consecutive cycles without interruption for four months. As for T/RH ageing, effects

of accelerated weathering were investigated each week. Samples aged under T/UV/RH were

dried overnight at 50°C after ageing before testing, in order to eliminate the excess of humidity

and to perform all characterizations in the same conditions

Table 9: Conditions in accelerated weathering chambers for T/UV and T/UV/RH ageing.

Experiment

Cycle period

Cycle time

(minutes)

Temperature

(°C)

Relative

humidity

T/UV

240

Dark

240

T/UV/RH

240

Dark and

condensation

240

Yes

3. Physico-chemical characterization

Microscopies

3.1.1. Digital microscopy

Digital microscopes are a variation of a traditional optical microscope that uses optics and

a charge-coupled device (CCD) camera to output a digital image to a monitor. In our case, the

microscope enables to get a clear view of the surface of PLA and FR-PLAs plates thanks to the

high depth of field. The expansion of the char after mass loss cone experiment was also

evaluated using the digital microscopy. Images of plates and chars were taken with a VHX-

1000 microscope (Keyence) at 50x magnification and stitched together in order to get a clear

“3D” image of the whole plate.

3.1.2. Electron probe micro analysis (EPMA)

Electron probe microanalysis (EPMA) is an analytical technique that is used to establish

the composition of small areas on specimens. This is one of several particle-beam techniques.

Technically, a beam of accelerated electrons is focused on the surface of a specimen using a

series of electromagnetic lenses, and these energetic electrons produce characteristic X-rays

within a small volume (typically 1 to 9 μ

) of the specimen. The characteristic X-rays are

detected at particular wavelengths, and their intensities are measured to determine

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter II: Materials and Methods

concentrations. Excepting H, He and Li, all elements can be detected, they emit at a specific

set of X-rays. EPMA has a high spatial resolution (10 kV-1 µA-150 nm) and sensitivity (around

3.0 microamperes), and this analytical technique allows obtaining highly magnified secondary

and backscattered electron images of a sample.

In this work, FR-PLA and FR-PLA-C30B were analyzed by EPMA to assess a possible

migration of fillers during ageing. In this regard, the samples were first carbon coated with a

Bal-Tec SCD005 sputter coater. A camera SX100 electron probe micro analyzer was used to

perform elemental analysis on the surface of the samples (electron interaction 1μ

). Back

scattered electron (BSE) images were carried out at 6 kV and 15 kV, 40 nA and phosphorus,

nitrogen and silica X-ray mappings were carried out at 6 kV 15 kV, 40 nA. On BSE pictures, the

darkest parts correspond to the “lightest” elements. For mappings, the crystals used to detect

the Kα of phosphorus, nitrogen and silica were respectively, a PER (pentaerythritol) crystal, a

PC2 (multicouche Ni/C) crystal and a TAP (thallium acid phthalate) crystal.

3.1.3. Scanning electron microscopy (SEM)

Scanning electron microscopy (SEM) is a technique that uses the interactions occurring

between electrons and matter. An electron beam is applied on the sample, leading to the

emission of secondary electrons and backscattered electrons (among other species), which

are then used to obtain topographic and chemical information respectively. The surface of PLA

and both FR-PLAs was observed before and after ageing using a Scanning Electron Microscope

FEG field emission gun JEOL JSM-7800F LV. SEM (Secondary electrons) experiments were

performed on carbon coated samples at 6 kV. Images were taken at x75 x400, x700, x1300

and x2500 magnifications.

3.1.4. Transmission electron microscopy (TEM)

Transmission electron microscopy (TEM) is a technique used to qualitatively evaluate the

dispersion of additives in a polymeric matrix. This technique was used to analyze the

dispersion of organomodified montmorillonite, i.e. Cloisite 30B in the PLA matrix. All the

samples were ultra-microtomed with a Leica Ultracut UCT ultramicrotome with a Diatome

diamond knife at room temperature to give section with a nominal section of 100 nm. The

section were transfered to Cu/Rh grid of 400 mesh. Bright-field TEM images of the samples

were obtained at 200 kV under low-dose conditions with a FEI Technai G2 electron

microscope.

Thermal analysis

3.2.1. Thermo-gravimetric analysis (TGA)

Thermo-gravimetric analysis (TGA) is a technique allowing to monitor the weight loss of a

sample versus temperature while the sample is heated following a defined temperature ramp.

The measurements can be carried out in oxidative (air) atmosphere.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter II: Materials and Methods

TGA measurements were carried out using a SETARAM (TG92-16). Balance and purge flow

rates were set at 15 and 100 mL/min respectively. Experiments were performed under air.

Samples of about 10 mg were positioned in open vitreous silica pans with gold foil and

submitted to an isotherm at 50°C for 10 minutes, then followed by a heating ramp of 10°C/min

up to 800°C. The precision on the temperature measurements is ± 1.5°C in the temperature

range of 50-800°C. Each TG analysis was repeated at least twice to ensure the repeatability of

the measurements. Theoretical TGA curves of FR-PLA and FR-PLA-C30B were calculated by

linear combination according to Equation 23.

wt.-% (theo) =



   







   







   



    

Equation 23

Where h, i, j and k are the mass fraction of PLA, Melamine, APP and Cloisite, 30B,

respectively (h + i + j + k =1) and wt.-% is the residual weight in percentage.

3.2.2. Differential scanning calorimetry (DSC)

DSC is widely used for the characterization of polymers. All experiments in this work have

been performed on a TA Instrument Discovery DSC and monitored with Trios software. This

technique is used to study the behavior of polymers when exposed to a temperature gradient

(rise and / or decrease of the temperature at a controlled rate) under nitrogen atmosphere.

This type of analyses allows the measurement of glass transition temperature (T

characterized by a change in the slope of the thermogram), the heat capacity (Cp) and, for

semi-crystalline polymers, the melting temperature (T

), the melting enthalpy (ΔH

), the

crystallization temperature (T

) and the crystallization enthalpy (ΔH

). Experimentally, two

pans are heated during the analysis: one pan containing 5.0 ± 0.5 mg of the sample previously

grounded, and a second empty pan used as a reference. Each sample has been exposed to

2 cycles of heating/cooling (Figure 32). Experiments were carried out at 10°C/min from room

temperature to 180°C, then an isotherm of 3 minutes was performed, after what a decrease

of temperature of 1°C/min until -20°C was followed by an isotherm of 3 minutes. At the end

of the experiment, the device returns to ambient temperature.

Figure 32: Temperature/time ramp used for the DSC analysis.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter II: Materials and Methods

The thermal history of the sample is suppressed during the first cycle. Concerning the

characterization of the different samples, only the second cycle was taken into account. All

information such as melting temperature (Tm), melting enthalpy (ΔHm), crystallization

temperature (Tc), crystallization enthalpy (ΔHc) and cold crystallization enthalpy (ΔHcc) were

obtained from these thermograms. The degree of crystallinity (χ) was calculated from the

melting enthalpy by the following Equation 24.

 

    









  

Equation 24

In this equation, χ allows measuring the proportion of material in the crystalline state. ΔHm

(100%) corresponds to the melting enthalpy of a 100% crystalline PLA (93.1 J/g) and  is the

additive content (FRs, nanoparticles). The temperature precision is estimated to 5.10

-3

°C. The

enthalpy precision is ± 0.04%.

Color changes measurements

The color changes and reflectance of unaged and aged samples were recorded using a

Datacolor CHECK 3 portable spectrophotometer from Datacolor Industry. The CIE (French

“Commission internationale de l'éclairage”) L*a*b color system was employed for the color

measurement. This system is commonly used in textile characterization [183]. In the CIE L*a*b

system the three coordinates presented in Figure 33 represent (i) the lightness of color (L* = 0

corresponds to black and L* = 100 corresponds to diffuse white; specular white may be

higher), (ii) the color position between red/magenta and green (negative values of a* indicate

green and positive values indicate magenta), and (iii) the color position between yellow and

blue (negative values of b* indicate blue and positive values indicate yellow).

Figure 33: color space of CIE Lab system.

The difference between unaged and aged samples (∆E) was calculated according to the

Equation 25 where ∆L* represents the lightness difference; ∆a* and ∆b*, the differences in a

and b values.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter II: Materials and Methods

 





 









 





  

Equation 25

The a* and b* coordinates are close to zero for neutral colors (white and gray) and increase

in magnitude for more saturated or intense colors. The advantage of CIE Lab system is that

color differences can be expressed in units that can be related to visual perception [184].

Three reflectance measurements were made on each sample; the samples were rotated 90°

before each measurement.

Gel permeation chromatography (GPC)

In order to determine the number average molecular mass (M

) of materials, all samples

were characterized by gel permeation chromatography (GPC). GPC is a type of high

performance liquid chromatography used to determine the molecular weight distribution of

polymers which can be performed in a wide range of solvents, from non-polar organics to

aqueous ones. Experimentally, columns packed with very small, round and porous particles

are used, allowing separation of molecules contained in the solvent that is passed through

them. Molecules are separated on the basis of their size, hence “size exclusion”. The first GPC

columns were packed with materials referred to as gels, hence “gel permeation”. The particles

in the columns are made from polymers that have been cross-linked to make them insoluble,

or from inorganic materials, such as spherical silica.

Experimentally, GPC is performed in tetrahydrofuran (THF) at 40°C, with a flow rate of

1 mL/min and a polymer concentration of 2 mg/mL after filtration through a 0.2 μm pore-size

membrane, to eliminate FR additives and nanoparticles. Measurements were performed on a

Waters system (Separation Module Waters e2695) equipped with three columns (Stryragel

HR1, Styragel HR3 and Styragel HR4) placed in series and coupled with a differential refractive

index (RI) Wyatt detector (WYATT Optilab T-Rex). It was established, however, that in order

to get the correct M

value of PLA, the experimental value obtained from the GPC traces using

polystyrene standards have to be multiplied by 0.58 [185]. In fact, when using polystyrene

standards for GPC analysis of Polylactic acid, the PLA molecular weights obtained by GPC are

typically larger than the actual molecular weights. Multiplication of the GPC data by a factor

of 0.58 has been suggested to get a more accurate number average molecular mass. The

polydispersity index (PI) was calculated as M

. Moreover, in order to analyze an

autocatalysis process that can occur during degradation of PLA based materials, the number

of chain scission n

, was calculated according to Equation 26 [186].

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter II: Materials and Methods











Equation 26

Where M

and M

are the number average molecular masses at time t and at the initial

time, respectively. Considering the precision of the equipment and all the multiple stages of

purification and dissolution of compounds, the error of measurement is estimated at 10%.

Melt flow index (MFI)

In order to characterize the viscosity of the formulations as a function of ageing, the melt

flow index (MFI) has been determined. In fact, when the MFI of a blend increases, its viscosity

decreases. The device used is a Melt Indexer MP1200 provided by Tinius Olsen (Figure 34) and

all the experiments were done in accordance to ISO 1133 Procedure B [187], i.e. under a

weight constraint of 2.16 kg and at 165°C. Using this procedure, the time taken by the piston

to run through 5 mm in the barrel is measured with the help of a sensor and with the weight

of the different collected rods, it allows obtaining directly the Melt Flow Index (MFI, g/10min),

the density (d, g/cm

) and the dynamic viscosity (η, Pa/sec) from the device. The error of

measurement is estimated to 15%.

Figure 34: Melt flow Index tester.

X-ray diffraction (XRD)

X-ray diffraction is a rapid analytical technique widely used as to determine and

characterize one or several crystalline phases of a material, providing information about the

structure of the material at the molecular scale. Besides, the recent technical and computing

developments allow obtaining precise information on the structure of the amorphous phase

of a polymer.

In this work, XRD was used in order to identify crystalline phases of our PLA based materials.

To this end, the experimental device was equipped with a Genix micro-source (Xenocs, France)

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter II: Materials and Methods

which produces a parallel beam wavelength of 1.54 Å (K

Cu line). A charge coupled device

(CCD) Very High resolution camera (Photonic Sciences) was used as a detector. Distance

between sample and detector was 8 cm, binning 4x4, and the samples were exposed for

15 minutes.

X-ray photoelectron spectroscopy (XPS)

X-ray photoelectron spectroscopy (XPS) measurements were carried out on an AXIS

Ultra

DLD

Kratos analytical apparatus using a monochromatic Al KR X-ray source (hν =

1486.6 eV)) in the following conditions: the emission voltage was set to 15 kV, the current

10 mA, the pressure in the analyzing chambers was maintained at 10

-9

Torr or lower during

the analysis, and the size of the analyzed area was 300 μm x 700 μm with a depth of 10 nm.

High-resolution C1s spectra were recorded at pass energies of 160 eV with a step of 1 eV. Data

treatment and peak-fitting procedures were performed using Casa XPS software. Obtained

spectra were rescaled by shift of C1s C–C at 285 eV. Thus, it is possible to determine the atomic

percentage of carbon, oxygen nitrogen, phosphorus and silica at the surface of the plates. XPS

is an analytical technique permitting to quantify the amount of targeted elements and to

identify their chemical or electronic state to get detailed information about the bonding of

elements at the sample surface. Samples are irradiated by an X-ray beam (E = hν) generated

by an high power aluminum monochromator. The energy of the X-rays is adsorbed by an atom

leading to emission of a core photoelectron. The emitted photoelectron exhibits a kinetic

energy (E

kinetic

) depending on the incident X-ray and the binding energy (E

binding

) of the atomic

orbital (Equation 27).

hν = E

kinetic

+ E

binding

Equation 27

Fourier transform infrared spectroscopy (FTIR)

Room temperature infrared spectra of the initial and degraded thin films of PLA were

recorded between 500 and 4000 cm

-1

using a Fourier Transformed Infrared spectrometer

(ThermoScientific) Nicolet IS 50 in ATR and transmission. Final spectra in ATR resulted from

32 scans using a resolution of 2 cm

-1

. Final spectra in transmission resulted from 16 scans using

a resolution of 2 cm

-1

and 32 cm

-1

. Both ATR and transmission spectra were processed by

OMNIC software. Electromagnetic waves belonging to the infrared domain are sent towards

the molecules composing the sample, inducing transitions between the vibration levels. Links

between atoms are targeted and information about these links are obtained. Thus, FTIR allows

to evidence chemical modifications on the structure of PLA based materials after ageing and

then permit to determine mechanisms of degradation. To avoid the differences due to the

thickness of the films, spectra were normalized according to the band at 2997 cm

-1

νC-H

characteristic vibration stretching band of PLA [34].

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter II: Materials and Methods

Chemical derivatisation

The T/UV and T/UV/RH aged samples were submitted to chemical treatment in order to

identify the photoproducts formed and to establish the mechanism of degradation. The

irradiated films were exposed to a reactive gas such as ammonia (NH

) at room temperature.

saturated atmosphere was obtained under glass bell in presence of concentrated NH

solution. The ammonia used in the experiment was provided from Sigma Aldrich. In parallel

to these treatments, we verified that the polymeric samples non-irradiated did not react with

the gas. As illustrated in Figure 35, the reaction of NH

with anhydride leads to the formation

of ammonium carboxylates and amide groups.

Figure 35: Reaction of anhydride with ammonia.

The ammonium carboxylates and primary amide groups are characterized by an infrared

absorption band above 1600 cm

-1

and 1675 cm

-1

respectively.

Solid state nuclear magnetic resonance (NMR)

The solid-state nuclear magnetic resonance (NMR) is an effective tool to analyze the

changes in the chemical environment of an atom inside a material. In the solid state (on the

contrary to liquid state), the chemical shift anisotropy (anisotropy of magnetic moment, CSA),

has a strong effect on the spectra, by broadening peaks. Through a tensoral analysis of the

magnetic moments in a molecule, it is possible to demonstrate that a “Magic Angle” exists,

with respect to the applied magnetic field at which the spinning of the sample leads to a

minimization of absorption line broadening due to CSA. This method of analysis is so called

“Magic Angle Spinning” (MAS).

In the case of carbon, the resonance of atoms is only possible in their isotopic form (

C).

However their low abundance leads to poor signal. This poor signal can be overcome by an

excitation of protons in the sample which makes them resonate at the same frequency than

that of target atom (Hartman-Hahn condition). This process is called cross-polarization (CP),

and the time of polarization is also called contact time. CP leads to a high enhancement of the

excitation of studied nuclei. However an interference exists between the large number of

protons and the decay of isolated nuclei, due to weak spin interactions. In this case, the

dampening of the signal can be removed with a strong radiofrequency signal which holds the

protons in a high resonance state so that they do not absorb resonance from the nuclei. This

is called

H decoupling.

P NMR measurements were performed on a Bruker Avance II 400 spectrometer with a 4 mm

probe. The Larmor frequency was 40.5 MHz (9.4 T). Measurements were performed with

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter II: Materials and Methods

dipolar decoupling at MAS of 10 kHz. 120 s delays was used between two pulses. Phosphoric

acid (H

) in aqueous solution (85%) was used as reference.

Electron paramagnetic resonance spectroscopy (EPR)

Electron paramagnetic resonance (EPR) spectroscopy is a very powerful and sensitive

method for the characterization of the electronic structures of materials. In EPR spectroscopy,

microwave radiation is used to probe species with unpaired electrons, such as radicals, radical

cations, and triplets in the presence of an externally applied static magnetic field. This

technique is used to study: (i) free radicals, i.e. atoms, molecules or ions containing on

unpaired electron either in the solid, liquid or gaseous phases, (ii) transition ions including

actinide ions (these routinely may have up to five or seven unpaired electrons), (iii) various

“point” defects in solids / localized imperfections, (iv) systems with more than one unpaired

electron such as triplets state systems, biradicals and multiradicals and (v) systems with

conducting electrons such as semiconductors and metals.

Theoretically, when a molecule or compound with an unpaired electron is placed in a strong

magnetic field, the spin of the unpaired electron can align in two different ways creating two

spin states, m

= ± ½. The alignment can either be along the direction (parallel) to the magnetic

field which corresponds to the lower energy state m

= - ½ or opposite (antiparallel) to the

direction of the applied magnetic field ms = + ½. The two alignments have different energies

and this difference in energy lifts the degeneracy of the electron spin states. The energy

difference is given by Equation 28.

ΔE = E

- E

= hν = gmβB

Equation 28

Where h is the Planck’s constant (6.626 x 10

-34

J/s), ν is the frequency of radiation, β is the

Bohr magneton (9.274 x 10

-24

J/T), B represents the strength of the magnetic field in Tesla and

g is the g-factor. During the experiment the values of h, ν and β does not change and g value

decrease as B increases. The g-factor is a unit measurement of the intrinsic magnetic moment

of the electron, and its value for a free electron is 2.0023. The concept of g can be roughly

equated to that of chemical shift in NMR. EPR spectrum is the absorption of microwave

frequency radiation plotted against the magnetic field intensity.

In an EPR experiment the field of the spectrometer magnet is swept linearly to excite some

of the electrons in the lower energy level to the upper energy level while the sample is

exposed to fixed microwave irradiation. The free of the unpaired electrons have a small

magnetic field and orient themselves parallel to the larger field produced by the

spectrometer’s magnet. At a particular magnetic field strength the microwave irradiation will

cause some of the free electrons to “flip” and orient against the spectrometer’s magnetic field.

This separation between the lower and the higher energy level is exactly matched by our

microwave frequency. The condition where the magnetic field and the microwave frequency

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter II: Materials and Methods

are “just right” to produce an EPR resonance (or absorption) is known as the resonance

condition is detected by the spectrometer. EPR spectroscopy can be carried out either by (i)

varying the magnetic field and holding the frequency constant or (ii) varying the frequency

and holding the magnetic field constant (as in the case of NMR spectroscopy). It exist two

typical methods to record an EPR spectra: (i) the continuous wave method where the sample

is irradiated continuously with microwave radiation of fixed frequency while the magnetic

field is slowly swept and the microwave absorption is measured for each field position. The

second one (ii) is the pulse EPR method where a short pulse of high microwave radiation are

sent to the sample and the response in the absence of radiation is recorded.

Experimentally, all chosen samples were measured at room temperature on a Bruker

E500 A/A/X TMHS spectrometer using X-band continuous wave mode. Converse time and

time constant were set to be 40.96 ms and 20.48 ms respectively. Microwave power and

modulation amplitude were respectively set to 2 mW and 1 G. The first order baseline

correction was applied to all recorded EPR. The weight of samples and measuring Q values

were normalized for the spectra to analyze the intensity of all paramagnetic species.

Formation of radical species under irradiation was performed in situ using Schimadzu

irradiation system with a filter operating in the range of 300-400 nm. For each samples a

kinetic was measured by recording spectrum each minute.

EPR image collection: The signals were acquired with a field-of-view of 4mm and gradient

strength of 175 Gcm

-1

. The sample was placed closest to the center of the resonator to

minimize the inhomogeneity of the B1 field of the resonator. The two-dimensional (2D)

images were acquired with a size of 512x512 pixels resulting in a pixel size of 5 µm.

EPR image processing: The image processing involves deconvoluting the whole signal

acquired under a magnetic field gradient from the reference signal collected without gradient.

After deconvolution process signals were backprojected using Fourier Transform, giving the

spatial distribution of paramagnetic species

Pulsed-EPR measurements were carried out at X-band (~ 9.7 GHz) with a Bruker ELEXSYS

E580 equipped with an Oxford Helium flow cryostat for low temperature measurements,

typically 4 K. Echo field sweep were achieved with /2 and pulse length values of 10 ns and

22 ns respectively. The hyperfine sublevel correlation spectroscopy (2D-HYSCORE)

measurements were also carried at 4.2 K with the four pulse sequence /2--/2-t

--t

-/2-

echo, and a time delay is 136 ns. This echo time were chosen by previously performing a 2D

3 pulses electron spin echo envelope modulation (ESEEM) versus t in order to minimize the

“blind spot” effect. The principle of the 2D HYSCORE is given in Appendix 5, p.228.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter II: Materials and Methods

4. Fire testing methods

4.1.1. Mass loss calorimeter (MLC)

The mass loss calorimeter (MLC) allows simulating the fire conditions at a small bench scale

according to ISO 13927 [181]. A schematic representation of the mass loss calorimeter is given

in Figure 36. The core of the instrument is a radiant electrical heater in the shape of a

truncated cone, irradiating a flat horizontal sample (100 x 100 x 3 mm

) placed beneath it, at

a chosen heat flux (e.g. 35 kW.m

-2

to simulate a mild fire in this study). The distance between

the sample holder and the heat source can be chosen as well (e.g. 25 mm in this study).

Ignition is provided by an intermittent spark igniter located 13 mm above the sample.

Figure 36: Schematic representation of mass loss calorimeter.

The mass loss calorimeter measures the temperature of the evolved gases using a

thermopile located at the top of the chimney. The calibration of the heat release rate (HRR) is

performed with methane. A methane flow of 0 to 6.7 mL.min

-1

is burnt above the sample

holder to obtain a calibration curve of the heat release as a function of temperature.

The values obtained by mass loss calorimeter testing are: the heat release rate (HRR), the

peak of heat release rate (pHRR), the total heat release (THR), the time to ignition (TTI) and

the mass loss of the sample during combustion (ML). In this work, the MLC measurements

were performed on a Fire Testing Technology mass loss calorimeter device at 35 kW/m² and

25 mm, according to the ISO 13927, on specimens of 100x100x3 mm

. All measurements were

performed three times. The presented curves are the worst case of repeatable results. The

acceptable error of measurement is estimated at 10 % for all values.

4.1.2. Limiting oxygen index (LOI)

The limiting oxygen index (LOI) is a small heat source ignition test. With the LOI test (ISO

4589-2 [180]), the relative flammability of materials, their ignitability and inflammation can

be evaluated. The experimental set-up of the LOI device is shown in Figure 37. LOI

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter II: Materials and Methods

measurements are performed on a Fire Testing Technology (FTT) device with barrels of

100x10x3 mm³ at room temperature. The specimen is clamped vertically into a glass cylinder

in a controlled oxygen-nitrogen mixture atmosphere. The measured LOI value corresponds to

the minimal oxygen concentration required to sustain the combustion of a material. It is

calculated according to Equation 29 and expressed as the percentage of oxygen in an oxygen-

nitrogen mixture.

Figure 37: Experimental set-up of the LOI test.

Materials having a LOI value below 21 vol.-%O

are called combustible, those with a LOI

value above 21 vol.-%O

are flame retarded. Thus, the higher the LOI value, the better the fire

retardancy of the material.

   















 





Equation 29

5. Conclusion

In the first part of chapter II, the polymer, the additives as well as the preparation methods

of the materials used in this study were presented. Afterwards, materials and protocols used

to simulate accelerated ageing of materials were described. Experimental techniques

investigated to characterize the impact of ageing on PLA and FR-PLAs, as well as methods to

assess fire retardancy before and after ageing were finally discussed.

The following Chapter III will further detail the impact of accelerated ageing on raw PLA.

Properties of this material before and after ageing, as well as its mechanisms of degradation

will be investigated. Comprehension of the degradation of raw PLA is a key parameter to

investigate, before investigating ageing of FR-PLAs in Chapter IV, so as to understand

thereafter the role of fire retardant additives on the degradation of PLA.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter III: Impact of ageing on PLA

This chapter deals with the impact of ageing on neat PLA. As reported in chapter I, PLA is

extremely sensitive to its environment, leading to a degradation of the material. In this way,

before investigating ageing of FR-PLAs, it is necessary to understand the effect of ageing

(temperature, UV and relative humidity) on neat PLA. The first section of this chapter presents

the visual and surface changes occurring on PLA over time as a function of ageing conditions.

Then, the impact of ageing on physico-chemical properties of the material is investigated.

Finally, mechanisms of degradation implied during ageing are proposed.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter III: Impact of ageing on PLA

1. Visual measurement of ageing

Even if the effects of ageing on PLA are widely reported in literature as stated in chapter I

(p.23), the objective of this chapter is to compare the already published results with the PLA

used in this study. Aim is to (i) evidence change in physico-chemical properties and (ii)

elucidate the mechanisms of degradation while bringing a complementary insight through the

use of unconventional experimental techniques such as electron paramagnetic resonance

spectroscopy (EPR). Hence, the first part of this chapter will be dedicated to the effects of

temperature, UV and relative humidity (i.e. T/UV, T/RH and T/UV/RH) on the physico-chemical

properties of PLA: the surface topography of the materials will be examined using digital

microscopy and SEM and the color changes and reflectance of aged samples will be compared

to the unaged ones using L*a*b* device.

Visual evaluation

The impact of the three kinds of ageing experiments on raw PLA was first evaluated visually.

The numerical pictures of PLA before ageing and after 60 days ageing under T/UV, T/RH and

T/UV/RH exposure (Figure 38) show that ageing promotes the “bleaching” of the plates. This

effect is even more pronounced for PLA aged for 60 days under T/UV/RH (Figure 38 (d)),

followed by T/RH (Figure 38 (c)) and T/UV (Figure 38 (b)) conditions.

Figure 38: Pictures of PLA at 0 d (a), 60 d T/UV (b), 60 d T/RH (c) and 60 d T/UV/RH (d)

exposure.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter III: Impact of ageing on PLA

In order to investigate in more detail these surface modifications, digital microscopy,

scanning electron microscopy (SEM) and L*a*b* were performed on PLA plates before and

after ageing.

Surface morphology of PLA after ageing

Whereas PLA surface is completely smooth before ageing, holes and cracks, slightly visible

to the naked eyes, appear during ageing. Digital microscopy was thus first used to compare

neat PLA before ageing (Figure 39 (a)) and aged PLA. Under T/UV/RH exposure cracks are

formed in the entire surface of the PLA after 60 days (Figure 39 (d)), whereas under T/UV and

T/RH they are dispatched in low amount at the surface (Figure 39 (b) and Figure 39 (c),

respectively).

Figure 39: Digital microscopy pictures of unaged PLA (a), 60 d T/UV (b), 60 d T/RH (c) and

60 d T/UV/RH (d).

Pictures of unaged and aged PLA obtained by SEM (Figure 40) support these observations.

In fact, holes and cracks are observed by SEM (circled in red in the pictures) and the

phenomenon is more pronounced for PLA aged for 60 days under T/UV/RH (Figure 40 (d))

where cracks are observed on the entire surface of the plate. As previously reported, PLAs

aged for 60 days under T/UV (Figure 40 (b)) and T/RH (Figure 40 (c)) are less damaged.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter III: Impact of ageing on PLA

Figure 40: SEM pictures obtained for unaged PLA (a), 60 d aged under T/UV (b), T/RH (c) and

T/UV/RH (d) conditions.

Color changes

The “bleaching” of PLA evidenced when submitted to environmental constraints is not in

accordance with data reported in literature. Indeed, as described in chapter I (p.44), ageing of

PLA usually leads to a “yellowing” of the material [28, 34, 35, 129].

Hence, in order to quantify the color changes during ageing, L*a*b* measurements were

performed on PLA plates before and after 60 days ageing under T/UV, T/RH and T/UV/RH

conditions. Results reported in Table 10 typically show an evolution of the color of the plates

during ageing. In fact, ∆E, which represents the difference between two colors, is 7.5 after 60

days exposure to T/UV and reaches 12.7 and 17.5 after 60 days exposure to T/RH and T/UV/RH

respectively. L*, which represents the lightness of color, increases from 52.0 before ageing to

58.8, 64.4 and 69.4 for PLA aged for 60 days under T/UV, T/RH and T/UV/RH respectively. The

increase in both ΔE and L* values indicates that plates become whiter after ageing, which

corroborates the “bleaching” phenomenon. For a* which represents the color between

red/magenta and green, values vary from -2.1 for PLA before ageing to -1.9, -1.6 and -1.3 for

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter III: Impact of ageing on PLA

PLA aged for 60 days under T/UV, T/RH and T/UV/RH respectively. Concerning b*, which

represents the color between yellow and blue, values increase from -1.0 to -4.0, -3.4 and -2.8

after 60 days ageing under T/UV, T/RH and T/UV/RH conditions, respectively. Due to the

relatively low values of a* and b*, no conclusion can be drawn. Indeed, generally, a* and b*

coordinates are close to zero for neutral colors (i.e. white and grey).

Table 10: L*a*b* values of unaged and 60 days aged PLA under T/UV, T/RH and T/UV/RH

conditions.

L*a*b* values

PLA

PLA T/UV 60 d

PLA T/RH 60 d

PLA T/UV/RH 60 d

52.0

58.8 (+6.8)

64.4 (+12.4)

69.4 (+17.4)

-2.1

-1.9 (+0.2)

-1.6 (+0.5)

-1.3 (+0.8)

-1.0

-4.0 (-3.0)

-3.4 (-2.4)

-2.8 (-1.8)

∆E

7.5

12.7

17.5

Hence, L*a*b* permits to quantify the change in PLA color when exposed to T/UV, T/RH

and T/UV/RH conditions and then corroborates the “bleaching” of materials. However, one

can notice that this “bleaching” phenomenon do not support literature which mainly report a

“yellowing” of PLA during ageing [28, 34, 35, 129]. This case provides an excellent example of

the complexity of the problem of quantifying for color changes. It is possible that the

difference observed against literature is due to the experimental parameters used during

ageing exposure (kind of device, light …). As an example, exposure to one light source may

cause principally bleaching and another, yellowing. Generally, exposure to the Xenotest

apparatus source results in increased yellowing. A fluorescence blue light, with radiation

peaked at 420 nm results chiefly in bleaching. Sunlight, which contains both ultraviolet and

visible radiation, also results in bleaching [188].

Results obtained by digital microscopy, SEM and L*a*b* measurements demonstrate that

whatever ageing conditions, visual impacts of ageing on PLA are similar (i.e. bleaching and

cracking), but dependent on the external stimuli applied to the polymer. Indeed, the effects

of ageing on the visual appearance of PLA are more pronounced when exposed to T/UV/RH

than when exposed to T/RH and T/UV conditions. As it was demonstrated that ageing impacts

the appearance of PLA, it is now necessary to understand what can be the effects of T/UV,

T/RH and T/UV/RH exposure on the physico-chemical properties of PLA. In fact, these latter

are responsible for the integrity of the material during its lifetime.

2. Physico-chemical properties after ageing

This part details the impact of ageing on the physico-chemical properties of PLA, especially

through the number average molecular mass (M

), the thermal properties such as T

and T

as well as the crystallinity. These properties are investigated until 125 days of ageing under

T/UV, T/RH and T/UV/RH conditions.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter III: Impact of ageing on PLA

Impact of ageing on molecular mass

PLA, like many other polyesters, is known to be extremely sensitive to ageing conditions,

leading to hydrolysis [5, 31, 125, 153] or photo-oxidation [27, 29, 33, 34]. These reactions

involve the scission of ester bonds, leading to the decrease of the number average molecular

mass (M

) of the material. Thus, M

was studied as a function of ageing exposure using gel

permeation chromatography (GPC). The evolution of M

and polydispersity index (PI) as a

function of time for different ageing conditions is plotted in Figure 41 and data are reported

in Table 11.

Table 11: M

and polydispersity index (PI) data as a function of ageing of PLA exposed to

T/UV, T/RH and T/UV/RH conditions.

Exposure

time

(days)

T/UV

(g/mol)

T/UV

T/RH

(g/mol)

T/RH

T/UV/RH

(g/mol)

T/UV/RH

68000

1.8

68000

1.8

68000

1.8

64600 (-5%)

1.7 (-0.1)

64250 (-6%)

1.7 (-0.1)

61600 (-11%)

1.8

61200 (-10%)

1.6 (-0.2)

59500 (-13%)

1.6 (-0.2)

51400 (-24%)

1.7 (-0.1)

57600 (-15%)

1.6 (-0.2)

52100 (-23%)

1.7 (-0.1)

39300 (-42%)

1.8

52100 (-23%)

1.7 (-0.1)

42000 (-38%)

1.7 (-0.1)

30950 (-54%)

1.8

47000 (-31%)

1.7 (-0.1)

34200 (-50%)

1.7 (-0.1)

21500 (-68%)

1.9 (+0.1)

44000 (-35%)

1.8

30300 (-56%)

1.8

15900 (-77%)

1.9 (+0.1)

38600 (-43%)

1.7 (-0.1)

25300 (-63%)

1.8

9100 (-87%)

1.8

105

33400 (-51%)

1.8

22600 (-67%)

1.7 (-0.1)

5100 (-92%)

1.9 (+0.1)

125

30000 (-66%)

1.7 (-0.1)

21500 (-71%)

1.8

3200 (-95%)

1.9 (+0.1)

It appears that M

decreases as a function of time for all samples (Figure 41 (a) and Table

11). In fact, M

decreases from 68000 g/mol before ageing to 30000 g/mol (-66%), 21500

g/mol (- 71%) and 3200 g/mol (-95%) after 125 days ageing in T/UV, T/RH and T/UV/RH

conditions, respectively. This drop of M

suggests that chain scission mechanism is largely

predominant during ageing under T, UV and/or RH. One can notice that the degradation of

PLA does not exhibit an induction period, i.e. that the hydrolysis and/or photo-oxidation start

as soon as the sample is exposed to ageing constraints. Moreover, these results indicate that

the decrease of M

is even more pronounced when UV is combined with T and RH compared

to T/RH and T/UV.

The polydispersity index of PLA (Figure 41 (b)) is independent of ageing conditions, as it

remains stable between 1.6 and 1.9. This result indicates that chain scissions occur randomly

along the polymer macromolecules under hydrolysis and/or photo-oxidative conditions [124,

127, 189] and suppose the fact that degradation is homogeneous through the plates.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter III: Impact of ageing on PLA

Figure 41: M

(a) and polydispersity index PI (b) versus ageing time of PLA samples aged

under T/UV (●), T/RH (▪) and T/UV/RH (Δ) conditions.

According to previous observations concerning PI, the drop of M

observed in Figure 41 (a)

is attributed to a chain scission process: indeed if depolymerisation process had occurred at

the same time, it would have caused a remarkable PI increase, which is clearly not the case in

this study.

The number of PLA chain scission (n

) as a function of time and ageing conditions (Figure

42) determined according to Equation 26 (p.78), increases as a function of ageing time. This

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter III: Impact of ageing on PLA

phenomenon depends on the ageing conditions: it is even more pronounced for PLA aged

under T/UV/RH conditions compared to T/RH and T/UV.

Figure 42: Average number of random chain scissions per unit mass (n

) for PLA as function of

ageing time for different ageing conditions; T/UV (●), T/RH (▪) and T/UV/RH (Δ).

Many studies led on PLA reveal that chain scission occurring during degradation leads to

the formation of carboxylic acid, which is reported to be responsible for an autocatalysis

process [5, 31, 125, 150, 153]. Indeed, autocatalysis occurs when the reaction product of

degradation is itself the catalyst for the same reaction [151]. In the case of PLA, carboxylic

acids formed during degradation are retained within the material, causing a localized decrease

in pH which accelerates the rate of degradation and induces a sudden loss of the structural

integrity of the material [152, 153]. Taking this into consideration, the behavior of n

(Figure

42) indicates an increase in the number of chains broken during ageing, thus increasing the

formation of secondary products such as carboxylic acid in the medium. These ones are thus

able to promote the auto accelerated character of the degradation, which is in agreement

with literature [123, 128, 153]. In addition, the combination between UV, T and RH accelerates

the degradation of PLA. Indeed, under T/UV/RH n

is around 3.2x10

-5

after 60 days against

1.7x10

-5

and 6.3x10

-6

under T/RH and T/UV, respectively.

Investigations led by GPC reveal that the molecular mass of PLA decreases when the

material is submitted to T/UV, T/RH and T/UV/RH exposure. Moreover, it is evidenced that

this behavior depends on the ageing conditions. Indeed, the decrease of M

is even more

important for PLA aged under combination of T, UV and RH compared to T/RH and T/UV. It is

well known that the decrease of molecular mass of PLA due to chain scission is followed by a

decrease of the thermal properties of the material [29, 33, 38, 121, 123, 157]. Hence, the

evolution of thermal properties versus ageing is studied in the following section, using

thermogravimetric analysis (TGA) and differential scanning calorimetry (DSC).

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter III: Impact of ageing on PLA

Thermal behavior

2.2.1. Thermogravimetric analysis

The thermal stability of PLA as a function of ageing was evaluated using TGA. Data collected

from the TGA of PLA aged under T/UV, T/RH and T/UV/RH exposure are reported in Table 12:

the onset temperature (T

10%

) corresponding to 10 wt.-% degradation of the material, the

temperature corresponding to its maximum rate of degradation (DTG), and the residual

weight obtained at 800°C are considered in this table.

Table 12: TGA data (10°C/min under air) obtained after ageing of PLA exposed to T/UV, T/RH

and T/UV/RH conditions.

Exposure / time

10%

(°C)

Max DTG (°C)

Residual weight (%)

0 d

332

358

T/UV 125 d

322 (-10)

355 (-3)

T/RH 125 d

300 (-32)

347 (-11)

T/UV/RH 125 d

279 (-53)

337 (-21)

Unaged PLA decomposes in one step from 280°C (Figure 43) and shows a maximum

temperature of degradation at 358°C with no residue left at 800°C. Aged PLAs also decompose

in one step and lead no residue as well. However, it appears that the onset temperature of

degradation (T

10%

) decreases upon ageing. Indeed, T

10%

decreases from 332°C before ageing

to 322°C, 300°C and 279°C after 125 days ageing under T/UV, T/RH and T/UV/RH exposure,

respectively. The DTG also decreases after ageing, from 358°C to 355°C, 347°C and 337°C after

125 days under T/UV, T/RH and T/UV/RH, respectively.

Figure 43: TGA curves as a function of time for unaged PLA (-), T/UV (-), T/RH (-) and

T/UV/RH (-) 125 days aged (10°C/min under air).

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter III: Impact of ageing on PLA

This decrease in onset temperature of degradation could be linked to the decrease in

molecular mass occurring during hydrolysis and photo-oxidation [174]. Indeed, the thermal

stability of PLA depends on the strength of its bonds. If the molecular mass decreases, it

creates new species with thermally weak bonds (e.g. hydroperoxides, carboxylic acids and

peroxides). These ones can quicker initiate the thermal degradation and then decrease the

thermal stability of PLA [174, 190-192]. Overall, TGA investigations demonstrate that ageing

decreases the thermal stability of PLA until 125 days. This phenomenon depends on the ageing

conditions. Indeed, the decrease of thermal stability is reported to be more significant under

T/UV/RH exposure compared to T/RH and T/UV. Under T/UV exposure, thermal stability is

only slightly affected.

2.2.2. Glass transition and melting temperature

The thermal properties of PLA before and after ageing were also evaluated by DSC. The DSC

data of PLA aged until 125 days under T/UV, T/RH and T/UV/RH conditions were compared to

those obtained for non-aged PLA. The glass transition temperatures (T

) and melting

temperatures (T

) are listed in Table 13.

It appears that both T

and T

are affected by ageing (Figure 44, Figure 45 as well as Table

13). In fact, T

decreases from 61.5°C to 58.8°C (-2.7°C), 57.4°C (-4.1°C) and 30.2°C (- 31.3°C)

after 125 days under T/UV, T/RH and T/UV/RH conditions, respectively. T

decreases from

160.2°C to 158.4°C (-1.8°C), 156.2°C (-4°C) and 122.3 (-37.9°C), after 125 days under T/UV,

T/RH and T/UV/RH conditions, respectively.

Table 13: DSC data of PLA reported as a function of ageing exposure to T/UV, T/RH and

T/UV/RH conditions.

Ageing

exposure

(days)

T/UV

g(°C)

T/UV

m(°C)

T/RH

g(°C)

T/RH

m(°C)

T/UV/RH

g(°C)

T/UVRH

m(°C)

61.5

160.2

61.5

160.2

61.5

160.2

61.1

160.2

61.1

160.1

61.0

159.9

61.0

159.9

60.9

159.7

60.8

159.7

61.0

159.4

60.8

158.5

59.9

158.2

60.9

158.9

59.9

158

58.9

157.1

60.7

158.7

59.5

157.9

55.5

154.4

60.4

158.9

59.4

157.4

45.5

150.6

105

59.8

158.9

58.1

156.6

40.7

129.9

125

58.8

158.4

57.4

156.2

30.2

122.3

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter III: Impact of ageing on PLA

Figure 44: Evolution of glass transition temperature (T

) of PLA as a function of exposure to

T/UV (●), T/RH (▪) and T/UV/RH (Δ) conditions.

Figure 45: Evolution of melting temperature (T

) of PLA as a function of exposure to T/UV

(●), T/RH (▪) and T/UV/RH (∆) conditions.

As evidenced it appears that both T

and T

of PLA are only slightly affected when the

material is exposed to T/UV or T/RH for 125 days. However, in the case of exposure to

T/UV/RH, the decrease of both T

and T

is quite obvious. After 125 days under T/UV and T/RH

exposure, PLA matrix is certainly not sufficiently degraded as under T/UV/RH to observe the

same decrease of its thermal properties (i.e. T

and T

). Hence, to confirm this assumption,

investigations were led on the impact of the molecular mass on the behavior of thermal

properties during ageing.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter III: Impact of ageing on PLA

It is well known that M

has a direct influence on the T

of a polymer [29, 32]. In fact, in the

case of linear polymers, T

and M

are linked according to the Fox and Flory relationship

(Equation 30), where k (K.kg.mol

-1

) is the Fox-Flory constant and Tg,∞ the glass transition

temperature of PLA having an infinite molar mass.

     

Equation 30

Thus, if the degradation of PLA under T/UV, T/RH and T/UV/RH exposure fits this equation,

when the molecular mass of the material decreases, the T

should decrease too. Hence, T

PLA was plotted against M

for the three ageing conditions (Figure 46). It appears that the

decrease of T

as a function of M

is quite stable during exposure to T/UV and T/RH conditions.

However, in the case of T/UV/RH exposure, the drop of T

is obvious from M

± 20000 g/mol.

Figure 46: T

as a function of M

for PLA aged under T/UV (●), T/RH (▪) and T/UV/RH (∆)

exposure.

This suggests that the drop of T

occurs from a critical M

of approximatively 20000 g/mol.

The term of critical molecular mass (M

) is considered as a characteristic parameter for any

polymer as an indicator of entanglement [193-195]. Indeed, the concept of entanglement in

polymer system is commonly accepted to interpret viscoelastic properties of high molecular

weight polymers. An important parameter for entangled macromolecules is the average

molecular mass spacing between entanglement junctions, i.e. the entanglement molecular

mass M

or the critical molecular mass M

, where some polymer properties of viscoelastic

nature are changed. Both M

and M

are interrelated according to Equation 31:

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter III: Impact of ageing on PLA

= 2.M

Equation 31

The value of M

is considered as a characteristic parameter for any polymer, as a kind of

“passport number”. Entanglement behavior of polymers is correlated with parameters of their

molecular geometry. The M

parameter denotes the transition from which the material

becomes fragile, starting to lose its structural integrity [193-195]. This critical molecular mass

reported at 20000 g/mol for PLA is in accordance with the literature [66, 196-199]. Indeed,

many researchers reported a M

of ± 9000 g/mol which evidences a M

of ± 18000 g/mol

according to Equation 31.

Hence, results reported in Figure 46 evidenced that after 125 days of ageing under T/UV/RH

the critical M

of PLA is reached, leading to the rapid degradation of PLA. Due to this low M

the mobility of the polymer chain is really high, explaining why the T

of the material

decreases. After 125 days exposure to T/UV/RH, both M

and T

of PLA reach very low values,

i.e. 3200 g/mol and 30.2°C, respectively. Taking into account this critical molecular mass

around 20000 g/mol, PLA is not sufficiently degraded under T/UV (30000 g/mol) and T/RH

(21500 g/mol) even after 125 days, to start the complete loss of its properties. It appears that

follows the same evolution as T

during ageing. The high mobility of the polymer chain

under T/UV/RH exposure affects the backbone flexibility and intermolecular interactions,

leading to the decrease of T

. [38]. As for T

, PLA is not sufficiently degraded under T/UV and

T/RH exposure, even after 125 days, to start the decrease of T

As expected, it was demonstrated that the degradation of the polymer, due to random

chain scission, leads to a decrease of the molecular mass of the material. The polymer chains

are broken into smaller units, thus leading to a decrease of some properties such as glass

transition and melting temperature of the material. Generally this decrease of thermal

properties also leads to changes in crystallinity (e.g. increase/decrease of crystallinity,

formation of new crystal-forms …) [5, 26, 38, 127, 157]. Hence, the next section is dedicated

to the crystallization behavior of PLA after ageing.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter III: Impact of ageing on PLA

Effect of ageing on crystallization behavior

As mentioned in the state of the art, PLA can be either amorphous or semi-crystalline,

depending on its stereochemistry and thermal history [5, 52]. In the case of a semi-crystalline

material, PLA can crystallize into three forms, i.e., α, β and γ. The evolution of the crystallinity

(χ) of PLA as a function of ageing for the three ageing conditions was measured by DSC

according to Equation 3 (p.33). Data are reported in Table 14 and Figure 47. First observation

is that the PLA used in this study is almost amorphous with a relatively low crystallinity of

about 6.5%.

Table 14: Crystallinity of PLA as a function of ageing duration under T/UV, T/RH and T/UV/RH

exposure.

Ageing time (days)

χ (%) – T/UV

χ (%) – T/RH

χ (%) – T/UV/RH

6.5

7.4 (+0.9)

8.1 (+1.6)

13 (+6.5)

9 (+2.5)

10.6 (+4.1)

16.5 (+10)

10.5 (+4)

15.1 (+8.6)

28.7 (+22.2)

12.3 (+5.8)

19.8 (+13.3)

37.8 (+31.3)

16.1 (+9.6)

21.3 (+14.8)

41.3 (+34.8)

14.8 (+8.3)

22.3 (+15.8)

43.5 (+37)

105

17.4 (+10.9)

24.5 (+18)

46.7 (+40.2)

125

19.8 (+14.3)

30.1 (+23.6)

52.9 (+46.4)

It is evidence that crystallinity increases when PLA is exposed to T/UV, T/RH and T/UV/RH.

Indeed, crystallinity is enhanced by 14.3% and 23.6% after 125 days exposure to T/UV and

T/RH, respectively. However, the most significant increase is observed for PLA aged under

T/UV/RH conditions, where 52.9% of crystallinity is achieved after 125 days. The higher

crystallinity of PLA after ageing is related to a high state of degradation of the material [5, 26,

27, 29, 38, 157]. Hence, the high crystallinity value reported after 125 days exposure to

T/UV/RH, indicates that this kind of exposure is more detrimental for the lifetime of PLA

compared to T/RH and T/UV conditions. This relationship between crystallinity and

degradation is further developed in this section.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter III: Impact of ageing on PLA

Figure 47: Evolution of crystallinity of PLA as a function of ageing time, for T/UV (●), T/RH (▪)

and T/UV/RH (∆) exposure.

To support these results and determine the kind of crystalline form of PLA, FTIR analyses

were performed on thin films (± 0.1 mm) of PLA aged under T/UV, T/RH and T/UV/RH

conditions, similarly to the plates. FTIR is one of the simplest ways to evidence the evolution

of the crystallinity of a polymer. Indeed, the bands at 870 cm

-1

and 950 cm

-1

are assigned to

the amorphous phase of PLA, whereas the ones at 755 cm

-1

and 694 cm

-1

are linked to the

crystalline phase [161, 200]. Moreover, some studies report that during crystallization of PLA,

the intensity of the band located at 950 cm

-1

decreases and a new one appears at 918 cm

-1

linked to the formation of crystals [200, 201]. FTIR analyses were performed until 60 days since

it was enough to evidence the evolution of the crystallinity. Moreover, after 60 days exposure

to T/UV, T/RH and T/UV/RH conditions, films exhibit the same M

as those of 60 days aged

plates, reported in p.90. Visually, films exposed to T/UV, T/RH and T/UV/RH conditions show

the same behavior as the plates (p.86).

FTIR results obtained for PLA before and after ageing are presented in Figure 48. For the

three spectra, peak observed at 950 cm

-1

corresponds to the δ

asym

+ νC-C. The one at

918 cm

-1

is linked to the rCH

. Peaks at 870 cm

-1

, 755 cm

-1

and 694 cm

-1

correspond to νC-COO,

δC=O and νCH, respectively.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter III: Impact of ageing on PLA

100

Figure 48: FTIR spectra as a function of ageing time for PLA aged under T/UV (a), T/RH (b)

and T/UV/RH (c).

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter III: Impact of ageing on PLA

101

It is obvious that during ageing, the intensity of the band at 950 cm

decreases, whereas

the one at 918 cm

-1

increases, confirming the evolution from an amorphous phase to a

crystalline one. Moreover, the intensity of the two bands at 694 and 755 cm

increases too,

referring to the formation of new crystals in the polymer. This phenomenon quickly appears

when PLA is aged under T/UV/RH compared to T/RH and T/UV. However, in the case of T/UV

exposure, the formation of crystalline phase seems to be very low. These results are in

accordance with the ones reported by DSC where crystallinity is enhanced by 14.3%, 23.6%

and 52.9% after 125 days exposure to T/UV, T/RH and T/UV/RH, respectively.

The rCH

mode at 918 cm

-1

is assigned to the 10

helix conformation of the α-form crystals

[30, 65, 161]. This can demonstrate that during ageing the PLA used in this study crystallizes

into α-form. However, this FTIR characteristic band is also present in the spectrum of α’-form

crystal [161, 202]. Recently Zhang et al. [202, 203] evidenced the existence of an α’ crystal

form, considered as a distorted α-form, with a 10

helix conformation but with a different

lateral arrangement than the one of the α form [161]. In order to determine in which

crystalline form PLA crystallizes during ageing, XRD experiments were performed.

The wide-angle X-ray diffraction (WAXD) patterns obtained by XRD (Figure 49) show that

unaged PLA and PLA exposed to T/UV present a typical amorphous profile. Due to the

relatively low crystallinity of this PLA (i.e. 6.5%, Table 14), the characteristic diffraction peaks

of crystallized PLA are not observed. In the case of T/UV exposure, XRD corroborates the weak

formation of crystals related to the band observed at 918 cm

-1

by FTIR. However, new

diffraction peaks appear at 2θ = 16.6° and 18.8° after ageing under T/RH and T/UV/RH

conditions, with peaks of higher intensity for this last condition. This can be explained by an

increase in crystallinity and the formation of new crystals structure. According to literature

[202, 203], reflection observed at 2θ = 16.6° and 18.8° in the WAXD profiles of aged PLAs is

typically characteristic of α’-form crystals. Indeed, in presence of α-form crystals, reflections

at 2θ = 16.6° and 18.8° are also observed but with additional ones including 2θ = 12.5°, 14.8°,

22.5°, 27.3° and 29.2°.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter III: Impact of ageing on PLA

102

Figure 49: WAXD patterns of PLA obtained by XRD, before (●) and after 60 of ageing under

T/UV (●), T/RH (●) and T/UV/RH (●).

In accordance with studies published on the crystallization behavior of PLA during ageing

[5, 29, 38, 127], it was demonstrated by DSC, FTIR and XRD that exposure to T/UV, T/RH and

T/UV/RH conditions has a direct influence on the crystallization of PLA. Indeed, an increase in

crystallinity occurs during ageing, with the formation of α’-form crystals. These phenomena

depend on the ageing conditions: whereas it is quite obvious for PLA aged under T/UV/RH and

T/RH conditions, the crystallinity is only slightly increased when exposed to T/UV conditions

and no formation of new crystalline structures is evidenced. Indeed, the degradation of the

material under T/UV exposure is certainly not sufficient after 125 days to achieve high

crystallinity and α’-form crystals structure. Hence, the impact of ageing on crystallization is

more important when PLA is aged under T/UV/RH, followed by T/RH and T/UV, respectively.

This increase in crystallization can be explained by the following mechanisms:

 When submitted to hydrolysis, the amorphous phases of PLA are more susceptible

to be degraded than crystalline phases, leading to an increase in crystallinity. In fact,

chain scissions occurring in amorphous phase of PLA produce low molecular chains

which can integrate crystalline phase if they have sufficient mobility. Indeed, they

can rearrange easier and thus giving crystallites leading to a weakening of the

system. Thus, an increase in crystallinity of the material occurs. This phenomenon

is well known in literature and called “chemi-crystallization” [5, 29, 38].

 Literature reports that when submitted to photo-oxidation, on the contrary to

hydrolysis, PLA chains are photodegradable in both crystalline and amorphous

regions. Although PLA chains are photodegradable even in crystalline regions, their

photo-degradability is higher in the amorphous regions [27, 204]. This is why as well

as during hydrolysis, it leads to an increase in crystallinity of PLA. However, it

appears that the increase of crystallinity is much lower than in the case of T/RH

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter III: Impact of ageing on PLA

103

exposure. As reported by Fayolle et al. [204], one possibility to explain this

difference is that radiochemical events occurring into the crystalline phase induce

a decrease of the crystallinity ratio which can compensate, at least partially, the

effect of chemi-crystallization. However, in presence of oxygen, the radiochemical

yield of crystal destruction is low compared to the crystallization yield, so

crystallization is observable. Another assumption to explain this difference is that

the kinetics of degradation and thus of crystallization of PLA is lower in the case of

T/UV compared to T/RH exposure.

 The combination between T/UV/RH leads to a faster degradation of the material

and thus the chemi-crystallization effect is more significant, explaining the higher

crystallization after 125 days of ageing compared to T/UV and T/RH exposure.

Conclusion

It has been demonstrated in this part that ageing affects PLA during its service lifetime. In

fact, external stimuli applied on the material such as temperature, UV light or relative

humidity have a direct influence on the physico-chemical properties of the material. Indeed,

it was highlighted that during ageing:

 The molecular mass (M

) of the material decreases

 The thermal stability decreases certainly due to the formation of thermally weak

bonds during ageing.

 Both T

and T

of PLA decrease and this phenomenon occurs from a critical

molecular mass around 20000 g/mol

 An increase in crystallinity is observed associated to the formation of α’-form

crystals

 These phenomena depend on the ageing conditions. Indeed, the impacts of

ageing are even more important for PLA aged under T/UV/RH compared to T/RH

and T/UV.

These effects of ageing on the physico-chemical properties are in adequacy with the wide

range of papers related to the ageing of PLA [29, 31, 33-35, 38, 121, 123, 153, 157-160, 162].

However, while the effects of ageing have been evidenced, mechanisms of degradation from

which these phenomena arise, have to be elucidated. Especially because many different

mechanisms have been proposed all over the years depending on the ageing conditions and

duration. In this way, the next part of this chapter is dedicated to the understanding of

mechanisms of degradation occurring during exposure to T/UV, T/RH and T/UV/RH.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter III: Impact of ageing on PLA

104

3. Elucidation of the mechanisms of degradation

Depending on the external stimuli applied, many different types of mechanisms can cause

degradation of PLA. Whereas literature reveals that hydrolysis is the main phenomenon

occurring in presence of water [5, 29, 31, 38, 52, 121, 123, 125, 153, 160, 161], researches

dealing with the photo-degradation mechanism of PLA are more contrasted. Indeed, photo-

degradation is reported to occur via two distinct mechanisms which are (i) a Norrish II

mechanism [27, 32, 33, 158, 159] and (ii) a radical mechanism [28, 29, 34, 35, 39]. In this

regard, this section will investigate mechanisms occurring during ageing under T/UV, T/RH and

T/UV/RH exposure.

Degradation under T/RH exposure

FTIR analyses were performed on thin films of PLA along exposure to T/RH in order to

identify structural changes occurring during ageing (Figure 50). These investigations were

performed at a resolution of 32 cm

-1

. The spectra are classified into four distinct regions,

corresponding to specific bands assignment: -OH stretch (3655 and 3671 cm

-1

, Figure 50 (a)),

-CH stretch (2877, 2946 and 2997 cm

-1

, Figure 50 (b)), -C=O ester carbonyl (1762 cm

-1

, Figure

50 (c)), -OH bend (1060 cm

-1

), -C-O stretch (1093, 1130, 1211, cm

-1

), and –C=O bend

(1257 cm

- 1

) and –CH- deformation (1367, 1454 cm

-1

) in Figure 50 (d).

These four regions as well as the one between 1560 to 1680 cm

-1

, corresponding to

carboxylate ions (COO

) (Figure 50 (e)), were chosen to follow PLA degradation versus ageing

duration. In the -OH region (Figure 50 (a)), an increase in hydroxyl absorption is observed as

function of ageing. In the case of the -C-H- stretch (Figure 50 (b)) no significant changes are

observed between virgin and degraded samples. It appears that the band at 1762 cm

-1

as well

as the bands in the range 1090-1300 cm

-1

are decreased during ageing. Indeed, those

corresponding to -C=O and -C-O stretching ester respectively shift towards lower wavelength

in range 1090-1300 cm

-1

attributed to -C-O- stretching alcohol. One can notice the apparition

of a new peak around 1630 cm

-1

when PLA is aged (Figure 50 (e)), corresponding to

carboxylate (COO

), which can result from the hydrolysis of PLA ester bond [31]. A decrease of

-OH band at 1060 cm

-1

is also observed (Figure 50 (d)).

Confirming literature [5, 29, 31, 38, 52, 121, 123, 125, 153, 160, 161], observations done by

FTIR analyses show that the degradation mechanism of PLA exposed to T/RH conditions is

predominantly driven by an hydrolysis of ester linkage, which is accelerated by temperature

(Figure 51) [29, 205]. As long as the polymer is exposed to this external stimulus, it absorbs

water, resulting in hydrolysis of ester linkages and breaking down of long macromolecular

chains. Thus, as previously described, this leads to the decrease of both molecular mass and

thermal properties of the material, but also to an increase in the crystallinity of the polymer

due to a “chemi-crystallization” process [5, 26, 38].

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter III: Impact of ageing on PLA

105

Figure 50: FTIR spectra of PLA before and after various ageing durations under T/RH

conditions.

Figure 51: degradation of PLA by hydrolysis under T/RH exposure.

The hydrolysis of PLA has been reported to take place mainly in the bulk of the material

rather than on the surface and has been assumed as an autocatalytic process when

temperature increases [25, 123, 147, 148, 150, 152]. This phenomenon leads to random chain

scission responsible for the decrease of the molecular mass of PLA. As previously reported in

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter III: Impact of ageing on PLA

106

Chapter I (p.52), autocatalysis occurs when the reaction product is itself the catalyst for the

same reaction. In the case of PLA, carboxylic acid chains formed in the bulk during the

degradation facilitate autocatalysis in the heart of the sample [152]. Moreover, the hydrolytic

chain cleavage is known to proceed preferentially in amorphous regions, which, as previously

explained, play a role on the crystallization of the polymer along ageing [5, 29, 38].

Degradation under T/UV and T/UV/RH exposure.

The effects of UV light on the degradation of PLA have been widely studied over the years

[27-29, 32-35, 157-159]. Until now, two photo-degradation mechanisms of PLA have been

evidenced, i.e. the Norrish II mechanism and a radical mechanism. These mechanisms depend

on the UV irradiation wavelength. Indeed, Norrish II occurs when PLA is exposed to UV-rays

under 300 nm whereas radical mechanism occurs in the range of 300 to 400 nm. In this study,

PLA based materials are exposed to UV-rays in the range 300-400 nm, and thus the Norrish II

mechanism does not seem to be the most probable one. Hence, the next part will investigate

mainly the radical mechanism.

The radical mechanism involves the abstraction of a tertiary hydrogen from PLA chain,

leading to the formation of radical species [34, 35]. Hence two techniques were investigated

to prove the existence of this mechanism: Electron Paramagnetic Resonance Spectroscopy

(EPR) and FTIR. EPR is known to be one of the most effective technique to detect and identify

free radicals generated by a chemical system during UV irradiation. Secondly, FTIR was

performed to identify secondary products formed during PLA irradiation.

3.2.1. Identification of radical species

EPR was firstly used to identify the radical species formed during exposure of PLA under

T/UV and T/UV/RH conditions. The EPR spectrum displayed in Figure 52 results from the

irradiation of virgin PLA at room temperature in the range 300 to 400 nm. The measurements

were carried out in situ while the sample was irradiated. This spectrum shows that the signal

is centered at an effective g value of 2.0063, which is typical of carbon radical species close to

an oxygen heteroatom [206, 207]. Moreover, the 4 lines splitting of the signal seem to indicate

the presence of a biradical specie [208, 209].

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter III: Impact of ageing on PLA

107

Figure 52: EPR spectrum of PLA irradiated with wavelengths between 300 and 400 nm at

room temperature.

EPR analyses performed on PLA suggest that PLA degradation proceeds via a radical

mechanism. In fact, it was demonstrated the formation of a carbon biradical specie when

exposed to UV-rays in the range 300 to 400 nm. According to the EPR results and literature

[28, 34, 35] the biradical formed can be schematized as depicted in Figure 53.

Figure 53: Biradical specie observed by EPR for PLA.

Researches have been already led by Babanalbandi et al. [207], focusing on the effect of

gamma irradiation on polylactic acid by EPR. However, contrary to this study, they reported

the formation of a monoradical when exposed to gamma irradiation (wavelength < 0.01 nm).

Indeed, due to the gamma irradiation, C-C bonds were broken, leading to the formation of

monoradical species. Irradiation in the range of 300 to 400 nm is not able to break C-C bonds,

thus explaining why only biradicals are observed.

In order to study the influence of temperature on UV ageing, kinetic evolution of the

formation of biradical species as a function of temperature was considered (Figure 54).

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter III: Impact of ageing on PLA

108

Figure 54: Kinetic evolution of EPR signal for PLA as a function of temperature, 25°C (●),

50°C (●) and 75°C (●).

These curves show that the number of biradical species formed under photochemical

ageing is maximum at 75°C and decreases when temperature decreases. These data from EPR

signal as a function of temperature were used to determine the activation energy (E

) needed

to initiate the formation of biradical species. This approach is based on an Arrhenius law.

Indeed, a linear relation (R²=0.98) is noted between ln (dХ”/dB) and the inverse of

temperature (Figure 55), suggesting that these coefficients follow an Arrhenius law (Equation

32):

(dX”/dB) = (dX”/dB)

exp(-





)

Equation 32

Where dX”/dB can be assigned to the susceptibility of radical formation, (dX”/dB)

is the

pre-exponential factor, E

is the activation energy in J/mol, R is the universal gas constant in

J.K

-1

.mol

-1

and T is the temperature is Kelvin (K).

The activation energy (E

) is obtained from the slope of the linear fitting of ln (dХ”/dB)

versus T

-1

and is estimated at 20 kJ/mol. The pre-exponential factor is equal to 3.4x10

. There

is actually no literature dealing with the activation energy of radical formation in PLA. Indeed,

the Arrhenius approach is not widespread in the study of radical formation kinetics by EPR.

However, few published values are available concerning the E

of water diffusion coefficient

into PLA from 40 to 80 kJ/mol [123, 210] or E

of thermal degradation from 72 to 125 kJ/mol

[127, 166, 211]. Even if it is not comparable, the E

determined in this study is very low

compare to the literature values. It appears that PLA is highly sensitive to photo-degradation

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter III: Impact of ageing on PLA

109

because of its small value of E

. These investigations demonstrate that temperature promotes

the activation of radicals.

Figure 55: Identification of the activation energy of biradical formation in PLA for different

temperatures.

In order to study the distribution of radicals during UV irradiation, EPR imaging experiments

(Figure 56) were realized on pellets of PLA under irradiation (300 – 400 nm). The pictures show

the formation and migration of the biradical species from the bulk to the whole sample.

Figure 56: 2D YZ EPR imaging of PLA 50°C under irradiation at 30 min (a) and 4 hours (b).

Intensity of color represents the local spin concentration from blue (min) to red (max).

After 30 min irradiation (Figure 56 (a)), it appears that micro domains of 50-100 µm are

preliminary formed at the center of the PLA sample. These domains extent to whole sample

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter III: Impact of ageing on PLA

110

when irradiation duration increases, with a migration of radicals observed along the X and Y

axes (Figure 56 (b)). These results are in accordance with literature [159].

The degradation mechanism of PLA samples exposed to both UV-rays and relative humidity

is relatively difficult to understand. In fact, there is no practical way to identify which one of

these two constraints catalyzes the other one. In order to estimate the additional influence of

relative humidity, kinetic evolution of the formation of biradical species as a function of ageing

exposure to T/UV/RH was studied at 50°C (Figure 57) and compared to results obtained for

T/UV exposure.

Figure 57: Kinetic evolution of EPR signal for PLA as a function of ageing exposure, T/UV (●)

or T/UV/RH (●) at 50°C.

This graph shows that the formation of radicals under T/UV increases during the first hour

of irradiation and then reaches a plateau till the end of irradiation. However, it appears that

the amount of radicals formed under T/UV/RH exposure is smaller than that under T/UV

conditions. The maximum amount of radicals formed is rapidly reached. One assumption to

explain this phenomenon is that hydrolysis and radical mechanism occur simultaneously

during T/UV/RH exposure. In this case, less radicals are formed because hydrolysis can disrupt

their formation by degrading PLA, thus forming secondary products sooner. Hence, the

combination between hydrolysis and UV-degradation is more detrimental for PLA lifetime.

FTIR analyses were then performed in order to identify the secondary products formed

during exposure to T/UV and T/UV/RH. Hence the next section is dedicated to these

investigations.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter III: Impact of ageing on PLA

111

3.2.2. Secondary products formed during degradation

FTIR analyses were performed at a resolution of 32 cm

-1

, on thin films of PLA after various

durations of exposure to T/UV and T/UV/RH, in order to identify secondary products formed

during irradiation of PLA (Figure 58).

Figure 58: FTIR spectra (x32 cm

-1

) of PLA before and after ageing under T/UV and T/UV/RH as

a function of ageing time.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter III: Impact of ageing on PLA

112

Spectra are classified into four regions, which correspond to the peaks assigned to the

following bands: -OH stretch (3655 and 3671 cm

-1

, (Figure 58 (a) and (b)), -C=O ester carbonyl

(1762 cm

-1

, (Figure 58 (c) and (d)), -OH bend (1060 cm

-1

), -C-O stretch (1093, 1130 and

1211 cm

-1

), –C=O bend (1257 cm

-1

) and -CH-/-CH

deformation (1367, 1454 cm

-1

) in Figure

58 (e) and (f), -COO

carboxylate (1630 cm

-1

) in Figure 58 (g) and (h). These peaks assignments

and corresponding positions are given in Table 1 (p.29).

It is clearly observed in Figure 58 (a) and (b) that T/UV and T/UV/RH exposure lead to an

increase in hydroxyl absorption between 3400 to 3750 cm

-1

, which can correspond to products

of degradation such as hydroperoxides or alcohols [34]. The higher intensity absorption

reported after ageing for T/UV/RH exposure (Figure 58 (b)) compared to T/UV (Figure 58 (a))

suggests that the combination of both humidity and UV enhances the degradation of PLA due

to a synergistic effect. The band at 1762 cm

-1

corresponding to -C=O stretching ester slightly

decreases when PLA is aged under T/UV exposure (Figure 58 (c)). As for hydroxyl bands, the

decrease of the -C=O band is even more pronounced when PLA is aged under T/UV/RH (Figure

58 (d)). In the range 1000-1300 cm

-1

(Figure 58 (e) and (f)), -C=O (1257cm

-1

) and –OH

(1060 cm

- 1

) bends are also decreasing along ageing time whatever ageing conditions.

Evolution of bands of -C-O- stretching ester in the range of 1093-1211 cm

-1

shows that the

bands are decreased and shifted towards lower wavelength after exposure to T/UV and

T/UV/RH conditions, corresponding to -C-O- stretching alcohol and carboxylic acid. The

increase in carboxylate (COO

) observe as a function of ageing at 1630 cm

-1

(Figure 58 (g) and

(h)) tends to confirm the formation of carboxylic acid during T/UV and T/UV/RH exposure.

Furthermore, this phenomenon is even more pronounced for PLA aged under T/UV/RH

conditions. These observations confirm that both radical mechanisms and hydrolysis occur

when PLA is exposed to T/UV/RH conditions and their combined action enhances the

degradation of PLA compared to T/UV and T/RH exposure.

Generally, the photo-degradation of PLA via a radical mechanism induces the formation of

anhydride groups [34, 35, 39]. However, at a resolution of 32 cm

-1

these chemical compounds

were not observed. Hence, in order to evidence their presence, FTIR analyses were also

performed at a resolution of 2 cm

-1

. The analyses of PLA photo-oxidized under both T/UV and

T/UV/RH show the formation of a band with a maximum at 1845 cm

-1

(Figure 59 (a) and (b)),

attributed to the formation of anhydride groups [34, 35, 39]. The intensity of this band

increases versus exposure duration for both ageing tests (Table 15). However, it appears that

the intensity of the anhydride band increases faster when PLA is exposed to T/UV compared

to T/UV/RH conditions. The intensity of the band increases by 0.08 after 60 days under T/UV

exposure compared to 0.03 under T/UV/RH exposure. In fact, due to the presence of water

under T/UV/RH exposure, anhydrides can be easily decomposed into carboxylic acid by

hydrolysis [34], which corroborate FTIR results previously reported.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter III: Impact of ageing on PLA

113

Figure 59: FTIR spectra (x2 cm

-1

) of PLA before and after ageing under T/UV and T/UV/RH as

function of ageing time.

Table 15: Evolution of the band at 1845 cm

-1

of PLA before and after ageing under T/UV and

T/UV/RH as a function of ageing time.

Ageing duration (days)

Abs 1845 cm

-1

(T/UV)

Abs 1845 cm

-1

(T/UV/RH)

0.24

0.27 (+0.03)

0.25 (+0.01)

0.30 (+0.06)

0.26 (+0.02)

0.32 (+0.08)

0.27 (0.03)

The presence of the anhydrides was also confirmed by exposing the unaged and 60 days

photo-oxidized films (T/UV and T/UV/RH) to ammonia (NH

) saturated atmosphere for

3 hours. Indeed it is known that the reaction of NH

with anhydride leads to the formation of

ammonium carboxylates and amide groups [34, 35]. FTIR analysis (resolution 2cm

-1

, Figure 60)

shows that the intensity of the absorption band at 1845 cm

-1

decreases after NH

treatment

whereas new bands appear at 1600 cm

-1

and 1675 cm

-1

. According to literature [34, 35], these

modifications indicate a reaction between photoproducts (e.g. anhydrides) and NH

and prove

the formation of both amide groups and ammonium carboxylate ions (Figure 35, p.80).

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter III: Impact of ageing on PLA

114

Figure 60: FTIR spectra (x2 cm

-1

) of 60 days aged PLA films to T/UV and T/UV/RH, before and

after treatment with NH

EPR and FTIR allow elucidating the degradation mechanism of PLA under UV irradiation: a

radical mechanism. Degradation is accelerated in presence of temperature and relative

humidity. This mechanism leads to the formation of secondary products such as anhydrides.

According to literature [28, 29, 34, 35, 39, 212] and experiments done in this study, a

summarizing mechanism of degradation is proposed in Figure 61.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter III: Impact of ageing on PLA

115

Figure 61: Proposed photo-oxidative mechanism of degradation for PLA, in presence of

water.

The mechanism is as follows: due to the action of UV, hydrogen is removed from the PLA

chain, leading to the formation of radical species. Then, oxygen reacts with the radical leading

to the formation of hydroperoxides. Photolysis of the hydroperoxides forms radicals (R1) that

can further evolve by β-scission into three different species: two possible C-C bond scissions

leading to P1 and P2, and one scission of C-O bond leading to P3. Considering the stability of

these different fragments and FTIR analyses previously reported, the most probable β-scission

appears to be P1, leading to the formation of anhydride groups. In presence of water, the

formation of anhydrides is disturbed because they can be easily hydrolyzed into carboxylic

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter III: Impact of ageing on PLA

116

acid such as acetic acid. The presence of an acidic medium, evidenced with a pH paper and

the characteristic odor of acetic acid reported during ageing makes this assumption

reasonable. P1 and P3 products can then further evolve thanks to the action of UV, dioxygen,

water and scission of C-C bond to finally yield formic, acetic, oxalic and carbonic acids, as well

as methyl acetate. As previously, pH paper allowed evidencing the acidity of the medium.

Moreover, the characteristic odors of both acetic acid and methyl acetate were reported

during ageing. There is certainly the formation of other products due to the action of

combined water and UV during T/UV/RH ageing test but for now, no investigations such as

pyrolysis GC/MS were led to determine which ones. Of course, when temperature increases

in the medium, the degradation of the material is accelerated, and even more when water is

present.

The UV degradation of PLA has been demonstrated to mainly take place in the bulk of the

material rather than on the surface. Molecular mass investigations evidenced an autocatalytic

process in the presence of UV-rays, leading to random chain scission responsible of the

decrease of the molecular mass. Both thermal properties and crystallinity are slightly affected

under UV-rays exposure, except in combination with relative humidity. Indeed, the

degradation via radical mechanism is enhanced in presence of relative humidity. The

combination of both UV and RH leads to a synergistic effect between hydrolysis and radical

degradation. Thus, the impacts of ageing are even more important compared to T/UV and

T/RH, explaining the higher degradation level of PLA when aged under T/UV/RH conditions.

4. Conclusion

Chapter III was dedicated to effects of different kinds of ageing conditions on neat PLA. The

investigations led on (i) the impacts of ageing on the physico-chemical properties and (ii) the

mechanisms of degradation are in accordance with the literature [29, 31, 33-35, 38, 121, 123,

153, 157-160, 162].

The appearance of cracks and holes at the surface of the plates, as well as a bleaching of

PLA samples was visually reported after 60 days of ageing. An influence of the ageing

conditions on the physico-chemical properties of PLA was identified. In fact, it was highlighted

that:

 The molecular mass (M

) of PLA decreases due to a random chain scission

 The thermal stability slightly decreases under T/UV exposure, but this decrease

is much more obvious under T/RH and T/UV/RH exposure.

 When exposed to T/UV/RH conditions, both T

and T

of PLA decrease and this

phenomenon occurs from a critical molecular mass of 20000 g/mol. In the case

of T/UV and T/RH exposure, both T

and T

are only slightly affected because

the critical molecular mass is still not reached after 125 days.

 Due to a “chemi-crystallization” phenomenon, an increase in crystallinity is

observed during ageing, associated to the formation of α’-form crystals. These

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter III: Impact of ageing on PLA

117

latter are not observed under T/UV exposure. Hence, the rate of

degradation/crystallization is certainly lower than under T/RH and T/UV/RH

exposure explaining why crystals are not observed during T/UV ageing.

 After 125 days exposure, the combination between T, UV and RH appears to be

the more drastic for lifetime of PLA, due to a synergistic effect. Hence, according

to literature and to our different experiments it is evidence that the effect of

T/UV/RH > T/RH > T/UV.

In any cases, the degradation of PLA has been reported to take place in the bulk of the

material rather than on the surface, assuming an autocatalytic process. However, the

mechanism of degradation depends on the exposure conditions:

 Under T/RH exposure, PLA is degraded predominantly by hydrolysis of ester

bonds. This phenomenon is even more pronounced when temperature

increases. Long macromolecular chains are breaking down as long as PLA is

exposed to T/RH.

 Under T/UV exposure, PLA is degraded via a radical mechanism. In presence of

UV-rays, radicals are formed and evolved into secondary products, reducing the

macromolecular chains into smaller unit.

 Under T/UV/RH exposure, it is reported that the combination between UV-rays

and RH creates a synergistic effect. Indeed, both hydrolysis and radical

mechanism occur during exposure and thus enhance the degradation of the

material compared to T/UV and T/RH exposure. However, it is quite difficult for

now to determine which of the two phenomena is predominant (i.e. hydrolysis

or radical mechanism).

 In any exposure conditions, the temperature increase enhances the degradation

of PLA.

 It appears that EPR is an innovative technique to investigate the mechanism of

degradation of polymers. Indeed, it offers new perspectives on both

mechanistic and kinetic points in the field of photo-degradation of polymers.

Since the mechanisms of degradation of neat PLA were elucidated, it is of interest to

investigate the impact of FR fillers when PLA based formulations are aged. Hence, in chapter

IV, the influence of FR additives on the physico properties of PLA and ageing behavior will be

analyzed in details.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

118

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter IV: Influence of flame retardants additives on the ageing of PLA

119

Chapter IV: Influence of flame retardants

additives on the ageing of PLA

In this chapter, the influence of the addition of flame retardants (i.e. Melamine, APP and

Cloisite 30B) on (i) the physico-chemical properties, (ii) the ageing behavior and (iii) the

mechanisms of degradation of the PLA matrix is investigated. Hence, two formulations are

studied and compared: FR-PLA, containing APP and Melamine, and FR-PLA-C30B, containing

APP, Melamine and Cloisite C30B.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter IV: Influence of flame retardants additives on the ageing of PLA

120

The ageing mechanism of PLA under T/UV, T/RH and T/UV/RH conditions has been

investigated and elucidated in chapter III. To fire retard PLA, Melamine, APP and Cloisite 30B

are incorporated in the polymer matrix, leading to an intumescent system [12]. However, even

if the impact of organoclays on the ageing of PLA has already been studied [32, 34, 37, 39, 162,

213], the impact of FR fillers on the ageing behavior of PLA has never been investigated. This

study is of primary importance to predict the lifetime of the FR-PLA and FR-PLA-C30B

intumescent systems for potential use in everyday life. In this way, the following part is

dedicated to the impact of ageing on the morphology of FR-PLAs.

1. Impact of ageing on the morphology of FR-PLAs

Visual observations

The numerical pictures of PLA and both FR-PLAs plates taken before ageing (Figure 62)

show a color change of the material when FR additives are incorporated into the PLA matrix.

In fact, whereas PLA (Figure 62 (a)) is almost transparent, FR-PLA (Figure 62 (b)) and FR-PLA-

C30B (Figure 62 (c)) samples tend to a cream color. This color change was quantified by L*a*b*

(Appendix 1, p.220). Indeed, the combination between L*, b* and ΔE values obtained during

ageing confirm the creamy color observed when FR are incorporated into PLA.

Figure 62: numerical pictures of PLA (a), FR-PLA (b) and FR-PLA-C30B (c) plates before ageing.

Then, in order to evaluate the impact of T/UV, T/RH and T/UV/RH ageing conditions on

both FR-PLAs, visual evaluation was also done all along the experiments, as long as possible

before the samples degrade (this is why the observation times are different depending on

samples). The numerical pictures taken before and after ageing under T/UV, T/RH and

T/UV/RH conditions are shown in Figure 63 and Figure 64 for FR-PLA and FR-PLA-C30B,

respectively. Compare to raw PLA, the main difference is the formation of a rough white layer

as well as holes and fractures at the surface of both FR-PLAs plates during ageing: these

phenomena are reported after 28 days ageing under T/RH exposure (Figure 63 (c) and Figure

64 (c)). Regarding (Figure 63 (d) and Figure 64 (d)), similar phenomena are observed for both

FR-PLA and FR-PLA-C30B plates after only 6 days ageing under T/UV/RH conditions. On the

contrary, it appears that after 60 days ageing under T/UV, only a laundering of FR-PLA and FR-

PLA-C30B occurs ((Figure 63 (b) and Figure 64 (b)). The laundering and the white layer formed

during ageing were also evidence and quantified using L*a*b* device (Appendix 2, p.221).

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter IV: Influence of flame retardants additives on the ageing of PLA

121

According to the color space of CIE lab system, the increase in ∆E and L* values confirm the

color change during ageing and the tendency for laundering.

Figure 63: Visual evaluation as a function of ageing for FR-PLA 0 d (a), 60 d T/UV (b), 28 d

T/RH (c) and 6 d T/UV/RH (d).

Figure 64: Visual evaluation as a function of ageing for FR-PLA-C30B 0 d (a), 60 d T/UV (b),

28 d T/RH (c) and 6 d T/UV/RH (d).

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter IV: Influence of flame retardants additives on the ageing of PLA

122

Focusing on ageing in presence of humidity (i.e. T/RH and T/UV/RH), it appears that FR-

PLA and FR-PLA-C30B are completely disintegrated into powder form (Figure 65) after 35 days

and 10 days under T/RH and T/UV/RH exposure, respectively.

Figure 65: Visual evaluation of FR-PLA (a) and FR-PLA-C30B (b) until disintegration into

powder (c).

These observations demonstrate that the presence of FR fillers leads to new phenomena

that were not observed in the case of virgin PLA (p.86), in particular, the formation of this

rough white layer at the surface of FR materials which appears very fast when UV and humidity

are combined.

In order to investigate in more detail these surface modifications, scanning electron

microscopy (SEM) analyses were performed on FR-PLA and FR-PLA-C30B plates before and

after ageing.

Surface morphology of FR-PLAs after ageing

Scanning electron microscopy (SEM) was used to observe the holes, fractures and white

spots at the surface of aged FR-PLA (Figure 66) and FR-PLA-C30B (Figure 67).

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter IV: Influence of flame retardants additives on the ageing of PLA

123

Figure 66: SEM pictures obtained for unaged FR-PLA (a), 60 d T/UV (b), 28 d T/RH (c) and 6 d

T/UV/RH (d).

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter IV: Influence of flame retardants additives on the ageing of PLA

124

Figure 67: SEM pictures obtained for unaged FR-PLA-C30B (a), 60 d T/UV (b), 28 d T/RH (c)

and 6 d T/UV/RH (d).

It appears that the surfaces of FR-PLA and FR-PLA-C30B are relatively smooth before ageing

(Figure 66 (a) and Figure 67 (a)). On the contrary, few cracks are observed after 60 days under

T/UV exposure (Figure 66 (b) and Figure 67 (b)) and impact of ageing is much more

pronounced for FR-PLA and FR-PLA-C30B when aged under T/RH and T/UV/RH conditions.

Indeed, holes, spots and fractures are observed at the surface of the materials after 28 days

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter IV: Influence of flame retardants additives on the ageing of PLA

125

exposure to T/RH conditions (Figure 66 (c) and Figure 67 (c)). Moreover, phenomena are even

more pronounced for FR-PLA-C30B (Figure 67 (c)) compared to FR-PLA (Figure 66 (c)). It

appears that after 6 days under T/UV/RH exposure holes, cracks and spots are more apparent

on the surface of both FR-PLAs (Figure 66 (d) and Figure 67 (d)) and that fillers are less

embedded by PLA matrix on the surface. As under T/RH, the degradation seems to be more

pronounced for FR-PLA-C30B compare to FR-PLA.

Conclusion

The visual observations and SEM results reported in this part demonstrate the impact of FR

fillers on the ageing of PLA. In fact, only a slight bleaching and cracking were reported for virgin

PLA when submitted to T/UV, T/RH and T/UV/RH exposure (p.86). However, holes, fractures

and a rough white layer appear at the surface of the T/RH and T/UV/RH aged materials until

their complete disintegration when FR fillers are incorporated into the PLA matrix. Only the

exposure to T/UV conditions appears not to be drastic for visual integrity of FR-PLA and FR-

PLA-C30B. Indeed, the material is certainly not sufficiently degraded after 60 days under T/UV

exposure. Hence in order to determine the causes of these phenomena, the impact of the

addition of fillers on the physico-chemical properties of PLA and their behavior during ageing

are investigated in the next part. In particular, molecular mass, thermal properties (i.e. T

and

) and crystallinity are studied into details.

2. Physico-chemical properties of FR-PLAs during ageing

This part investigates the physico-chemical properties of both FR-PLA and FR-PLA-C30B

before ageing, in terms of molecular mass (M

), thermal properties (i.e. T

, T

) and

crystallinity. The impact of ageing on these physico-chemical properties is then investigated

and compared to that of PLA, studied in chapter III. As for PLA, physico-chemical behavior of

FR-PLA and FR-PLA-C30B was studied until 125 days under T/UV. However, due to the quick

degradation of materials, FR-PLAs were studied until 105 days under T/RH and 90 days under

T/UV/RH conditions. Moreover, as FR-PLAs are disintegrated into powder after 10 and 35 days

under T/UV/RH and T/RH, respectively, investigations were led using this powder beyond

these durations.

Molecular mass

2.1.1. Impact of fillers on molecular mass of FR-PLAs

The M

of PLA and FR-PLAs, before ageing, obtained by GPC (Table 16) show that when FR

fillers are incorporated into PLA, M

is reduced by 57% and 60% for FR-PLA and FR-PLA-C30B,

respectively. However, it appears that the polydispersity index is not affected by the

incorporation of FR fillers. Indeed, PI remains stable between 1.8 and 1.9.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter IV: Influence of flame retardants additives on the ageing of PLA

126

Table 16: Number average molecular mass (M

) of PLA, FR-PLA and FR-PLA-C30B materials.

Material

PLA

FR-PLA

FR-PLA-C30B

(g/mol)

68 000

29 000 (- 57%)

27 000 (- 60%)

1.8

1.9 (+0.1)

This M

decrease is explained by the preparation process of the samples. The incorporation

of fillers into PLA via the extrusion process creates mechanical (shear, strain …) and thermal

stresses, decreasing molecular mass of PLA [84, 214-220]. Moreover, due to the presence of

fillers, chemical reaction can also occur during extrusion, breaking polymer chains and

decreasing M

. In fact, Melamine and APP contain amine functions that are able to break PLA

chains through aminolysis reaction on ester (Figure 68) [221-223].

Figure 68: Aminolysis reaction of PLA.

In addition, it is known that the surfactant contained in Cloisite 30B can undergo

degradation during the melting process at around 200°C, in presence of residual water,

according to Hofmann elimination mechanism (Figure 69) [224].

Figure 69: Hofmann elimination reaction on the surfactant of Cloisite 30B.

During Hofmann elimination, the quaternary ammonium salt decomposes into an amine,

which can thus be responsible for the acceleration of the polymer decomposition in the initial

stage [224-227]. However, due to the temperature used during extrusion process (i.e. 185°C

maximum in the zone of incorporation of fillers) and the relatively low loading of organoclays

(i.e. 1%), it is doubtful that Hofmann elimination is very meaningful to explain the decrease of

in this case . Moreover, considering the measurement uncertainty (i.e. 10%) the decrease

of M

is similar for both FR materials indicating that the 1 wt.-% of organoclays do not

particularly impact the M

of FR materials.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter IV: Influence of flame retardants additives on the ageing of PLA

127

Knowing that M

of raw PLA is affected by (i) ageing exposure and (ii) the incorporation of

FR fillers, it is useful to investigate concerning the impact of ageing on M

of FR-PLAs.

2.1.2. Impact of ageing on the molecular mass of FR-PLAs

It was previously reported in chapter III (p.90) that the molecular mass of PLA decreases

when the exposure duration to T/UV, T/RH and T/UV/RH increases. In order to determine if

this phenomenon is also observed in the case of FR-PLAs, GPC analyses were performed during

T/UV, T/RH and T/UV/RH exposure. M

and PI of FR-PLAs as a function of ageing time are

compared in Table 17 to the values obtained for neat PLA. The curves obtained by GPC,

plotting M

as a function of ageing conditions and duration, are presented in Figure 70.

Table 17: M

and PI data as function of ageing of PLA, FR-PLA and FR-PLA-C30B samples

aged under T/UV, T/RH and T/UV/RH.

Exposure / time

(days)

PLA

(g/mol) / PI

FR-PLA

(g/mol) / PI

FR-PLA-C30B

(g/mol) / PI

68 000 / 1.8

29 000 / 1.9

27 000 / 1.9

T/UV 125 d

30 000 (-55%) / 1.7

13 500 (-53%) / 1.8

12 500 (-54%) / 1.9

T/RH 105 d

17 000 (-72%) / 1.7

4 300 (-84%) / 1.8

4 100 (-84%) / 1.8

T/UV/RH 90 d

9 100 (-86%) / 1.8

2 800 (-90%) / 1.8

2 500 (-91%) / 1.9

Figure 70: M

versus ageing time of FR-PLA (a) and FR-PLA-C30B (b) samples aged under

T/UV (●), T/RH (▪) and T/UV/RH (Δ).

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter IV: Influence of flame retardants additives on the ageing of PLA

128

As for virgin PLA, it is evidenced that M

of FR-PLA and FR-PLA-C30B decreases versus time

for all ageing exposure conditions. In fact, regarding FR-PLA, M

decreases from 29000 g/mol

before ageing to 13500 (-53%), 4300 (-84%) and 2800 g/mol (-90%) after 125 days under T/UV,

105 days under T/RH and 90 days under T/UV/RH, respectively. Concerning FR-PLA-C30B, M

decreases from 27000 g/mol before ageing to 12500 (-54%), 4100 (-84%) and 2500 g/mol

(- 91%) after 125 days under T/UV, 105 days under T/RH and 90 days under T/UV/RH,

respectively. Furthermore, after 6 days of ageing under T/UV/RH and 28 days under T/RH

conditions, the M

of both materials is around 20000 g/mol (Figure 70), which corresponds to

the appearance of the white layer at the surface of the plates in presence of relative humidity.

As in the case of neat PLA, the drop of M

suggests that chain scission occurs during ageing

of FR-PLAs. Moreover, it seems that M

of both FR materials decreases at the same rate and

that this rate is higher when UV is combined with T and RH compared to T/RH and T/UV.

Indeed, after 90 days, M

is decreased by 90% and 91% under T/UV/RH exposure for FR-PLA

and FR-PLA-C30B, respectively. On the contrary, after 90 days exposure to T/RH, M

decreased by 75% and 81% for FR-PLA and FR-PLA-C30B, respectively, and by 37% for both

materials under T/UV exposure. Moreover, the kinetics of degradation of PLA seems not to be

impacted by the presence of FR fillers. As an example, after 90 days under T/UV/RH, M

of FR-

PLAs is reduced by around 90% compared to 86% for neat PLA. These reductions can be

considered as similar regarding the measurement uncertainty estimated to 10% and tend to

confirm that the degradation rate is similar for PLA and FR-PLAs.

Polydispersity index (PI) of FR-PLA and FR-PLA-C30B is independent of ageing, as it remains

stable between 1.8 and 1.9 (Table 17). Hence, chain scissions occur randomly along the

polymer macromolecules [124, 127, 189].

The number of chain scissions of FR-PLA and FR-PLA-C30B (n

) as a function of time and

ageing conditions (Figure 71) shows that whatever the nature of the sample, n

increases as a

function of ageing duration. This phenomenon depends on the ageing conditions: it is even

more important for FR-PLAs aged under T/UV/RH conditions than to those submitted to T/RH

and T/UV. Moreover, for the same ageing condition, there is no significant difference between

FR-PLA and FR-PLA-C30B.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter IV: Influence of flame retardants additives on the ageing of PLA

129

Figure 71: Average number of random chain scissions per unit mass (n

) for FR-PLA and FR-

PLA-C30B as a function of ageing time at different ageing conditions; T/UV (●), T/RH (▪) and

T/UV/RH (Δ).

The combination between UV, T and RH accelerates the degradation of FR-PLAs: regarding

FR-PLA, under T/UV/RH, n

is around 3.2x10

-4

after 90 days against 1.9x10

-4

and 3.8x10

-5

after

105 days under T/RH and 125 days under T/UV, respectively. For FR-PLA-C30B, under

T/UV/RH, n

is around 3.6x10

-4

after 90 days against 2.8x10

-4

and 4.3x10

-5

after 105 days under

T/RH and 125 days under T/UV, respectively. These results confirm that FR fillers do not

disturb the degradation of the materials: (i) the presence of fillers doesn’t disturb the chain

scission process of PLA and thus (ii) the autocatalytic process still occurs due to the increase

of carboxylic acid end groups, as previously reported in chapter III (p.90) [5, 31, 125, 150, 153].

The ageing of FR-PLA and FR-PLA-C30B leads to the decrease of the molecular mass of the

material, whatever ageing conditions. However, it appears that the M

of FR materials reaches

lower values after ageing compared to those of PLA. Due to the fact that the degradation rate

seems to be the same for unfilled or filled PLAs, the hypothesis is that the already low M

before ageing of FR-PLA and FR-PLA-C30B is the main cause explaining why such low Mn are

reached for FR materials. This also explains why the degradation of FR-PLAs samples is more

pronounced compared to unfilled ones. This can also partially explain the phenomena

observed at the surface of both FR-PLAs, i.e. appearance of a white layer, when the critical

molecular mass (i.e. 20000 g/mol) is reached.

As the impact of fillers on the behavior of the molecular mass during ageing was

demonstrated in this section, the next section will investigate the influence of FR fillers on (i)

glass transition temperature (T

) and (ii) melting temperature (T

) of PLA during ageing, using

DSC analyses.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter IV: Influence of flame retardants additives on the ageing of PLA

130

Thermal properties

2.2.1. Effect of fillers on glass transition and melting temperature

The thermal properties of FR-PLA and FR-PLA-C30B were evaluated by DSC in order to

investigate the effect of the incorporation of fillers on the T

and T

of materials. The DSC data

of FR-PLA and FR-PLA-C30B (Table 18) were compared to those of PLA.

Table 18: DSC data of PLA, FR-PLA and FR-PLA-C30B as a function of M

Material name

(g/mol)

(°C)

PLA

68000

61.5

160.0

FR-PLA

29000 (-57%)

57.5 (-6.5%)

159.8 (-0.1%)

FR-PLA-C30B

27000 (-60%)

57.4 (-7.0%)

159.2 (-0.5%)

Whereas FR additives do not affect the T

of materials, it appears that the T

of FR-PLA and

FR-PLA-C30B is somehow lower (loss of 4°C) compared to that of PLA. Incorporation of fillers

into a polymer can affect the molecular mobility of the polymer chains due to a plasticizing

effect [228-230]: the fillers can increase the local flexibility of the material, thus affecting the

free volume of the polymer. Therefore, it can cause the decrease of T

. On the other hand, the

decrease of T

can also be explained by the degradation of the PLA matrix during extrusion

process, which is in accordance with the decrease of M

observed when FR fillers are

incorporated into the material (p. 125). Indeed, the molecular mass of a polymer is known to

be a parameter affecting T

[231-235]. In the case of linear polymer such as PLA, the glass

transition temperature decreases when the number average molecular mass of the polymer

decreases [231, 234-236]. Generally, this dependence can be quantified using the Fox and

Flory relationship (cf. Equation 30, p.94). In reality, this phenomenon is linked to the

concentration of chain ends of the material. Indeed, the increase of chain ends (low M

) leads

to an increase in free volume of the polymer, thus decreasing T

[231].

As both T

and T

of neat PLA are impacted by (i) ageing exposure and (ii) the incorporation

of FR fillers, the next section focus on the effect of ageing on thermal properties (T

and T

)

of FR-PLAs.

2.2.2. Behavior of glass transition and melting temperatures during ageing

The glass transition (T

) and melting temperatures (T

) of FR-PLA and FR-PLA-C30B were

also evaluated by DSC all along the exposure to T/UV, T/RH and T/UV/RH conditions and

compared to those obtained for non-aged materials (Table 19 and Table 20 for FR-PLA and

FR-PLA-C30B, respectively).

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter IV: Influence of flame retardants additives on the ageing of PLA

131

Table 19: DSC data of FR-PLA as a function of M

and ageing time under T/UV, T/RH and

T/UV/RH.

T/UV/RH

(°C)

159.8

151.9

(-7.9)

149.4

(-10.4)

140.1

(-19.7)

128.8

(-31)

110.6

(-49.2)

94.3

(-65.5)

T/UV/RH

(°C)

57.5

50.3

(-7.2)

46.9

(-10.6)

40.7

(-16.8)

29.9

(-27.6)

25.2

(-32.3)

21.5

(-36)

T/UV/RH

(g/mol)

29000

17500

(-39%)

13000

(-54%)

8800

(-69%)

5300

(-80%)

4000

(-86%)

2800 (

-90%)

T/RH

(°C)

159.8

153.3

(-6.5)

150.7

(-9.1)

148.4

(-11.4)

138.5

(-21.3)

131.7

(-28.1)

120.9

(-38.9)

101.1

(-58.7)

T/RH

(°C)

57.5

54.8

(-2.7)

52.8

(-4.7)

50.2

(-7.3)

48.6

(-8.9)

44.8

(-12.7)

40.8

(-16.7)

36.2

(-21.3)

T/RH

(g/mol)

29000

24300

(-16%)

18400

(-34%)

15800

(-44%)

12400

(-56%)

10000

(-65%)

6900

(-75%)

4300 (

-84%)

T/UV

(°C)

159.8

158.4

(-1.4)

156.5

(-3.3)

156.3

(-3.5)

155.1

(-4.7)

154.5

(-5.3)

154.1

(-5.7)

153.5

(-6.3)

152.5

(-7.3)

T/UV

(°C)

57.5

56.4

(-1.1)

56.3

(-1.2)

55.1

(-2.4)

53.9

(-3.6)

52.8

(-4.7)

50.9

(-6.6)

49.5

(-8)

48.3

(-9.2)

T/UV

(g/mol)

29000

26000

(-9%)

24100 (

-14%)

21900 (

-22%)

213

00 (

-26%)

19600 (

-30%)

17400 (

-37%)

15500 (

-45%)

13500 (

-53%)

Ageing

exposure

(days)

105

125

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter IV: Influence of flame retardants additives on the ageing of PLA

132

Table 20: DSC data of FR-PLA-C30B as a function of M

and ageing time under T/UV, T/RH

and T/UV/RH.

T/UV/RH

(°C)

159.2

148.9 (

-10.3)

146.1 (

-13.1)

132.8 (

-26.4)

115.4 (

-43.8)

96.1 (-63.1)

83.9 (

-75.3)

T/UV/RH

(°C)

57.4

43.1 (

-14.3)

40.0 (

-17.4)

34.9 (

-22.5)

22.5 (

-34.9)

16.2 (

-41.2)

11.5 (

-45.9)

T/UV/RH

(g/mol)

27000

15000 (

-45%)

9700 (

-64%)

5800 (

-75%)

4300 (

-84%)

3400 (

-88%)

2500 (

-91%)

T/RH

(°C)

159.2

152.0 (

-7.2)

148.9 (

-10.3)

145.9 (-13.3)

134.9 (

-24.3)

122.7 (

-36.5)

107.5 (-51.7)

89.4 (

-69.8)

T/RH

(°C)

57.4

52.9 (

-4.5)

50.4 (

-7)

46.6 (

-10.8)

39.6 (

-17.8)

35.2 (

-22.2)

28.9 (

-28.5)

23.0 (-33.4)

T/RH

(g/mol)

27000

21600 (

-20%)

17600 (

-35%)

12200 (

-55%)

9800 (

-64%)

7900

(-71%)

5100 (

-81%)

4100 (

-85%)

T/UV

(°C)

159.2

157.4 (

-1.8)

154.3 (

-4.9)

154.0 (

-4.6)

153.3 (

-5.9)

152.8 (

-6.4)

152.3 (

-6.9)

151.3 (

-7.9)

150.0 (

-9.2)

T/UV

(°C)

57.4

56.2 (

-1.2)

55.3 (

-2.1)

55.1 (

-2.3)

53.2 (

-4.2)

52.3 (

-5.1)

49.0 (

-8.4)

48.6

(-8.8)

46.7 (10.7)

T/UV

(g/mol)

27000

25100 (

-7%)

23700 (

-12%)

216

00 (

-20%)

211

00 (

-23%)

19100 (

-29%)

17000 (

-37%)

14700 (

-54%)

12500 (

-54%)

Ageing

exposure

(days)

105

125

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter IV: Influence of flame retardants additives on the ageing of PLA

133

Regarding FR-PLA material (Table 19, Figure 72 and Figure 73) it is evidenced that both T

and T

are affected by ageing under T/UV, T/RH and T/UV/RH conditions. In fact, T

decreases

from 57.5°C to 48.3°C (-9.2°C), 36.2°C (-21.3°C) and 21.5°C (-36°C) after 125 days under T/UV,

105 days under T/RH and 90 days under T/UV/RH exposure, respectively. T

decreases from

159.8°C to 152.5°C (-7.3°C), 101.1°C (-58.7°C) and 94.3°C (-35.5°C) after 125 days under T/UV,

105 days under T/RH and 90 days under T/UV/RH, respectively.

FR-PLA-C30B (Table 20, as well as Figure 72 and Figure 73) exhibits the same behavior as

FR-PLA: both T

and T

are affected by ageing under T/UV, T/RH and T/UV/RH exposure.

Indeed, T

decreases from 57.4°C to 46.7°C (-10.7°C), 23°C (-33.4°C) and 11.5°C (-39.3°C) after

125 days under T/UV, 105 days under T/RH and 90 days under T/UV/RH, respectively. T

decreases from 159.2°C to 150°C (-9.2°C), 89.4°C (-69.8°C) and 83.9°C (-75.3°C) after 125 days

under T/UV, 105 days under T/RH and 90 days under T/UV/RH, respectively.

These results demonstrate once ageing, that the impact of ageing on T

and T

dependent on the ageing conditions: the decrease of T

and T

is more important when

samples are exposed to T/UV/RH conditions compared to T/RH and T/UV, respectively.

Figure 72: Evolution of glass transition temperature (T

) of FR-PLA and FR-PLA-C30B as a

function of exposure time under T/UV (●), T/RH (▪) and T/UV/RH (Δ).

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter IV: Influence of flame retardants additives on the ageing of PLA

134

Figure 73: Evolution of melting temperature (T

) of FR-PLA and FR-PLA-C30B as function of

exposure time under T/UV (●), T/RH (▪) and T/UV/RH (Δ).

Moreover, it is quite obvious regarding Table 19, Table 20 and Figure 74 that the decrease

of both T

and T

of FR materials is even more important for low molecular masses:

 For FR-PLA: T

and T

are 48.3°C and 152.5°C, respectively, for a molecular mass

of 13500 g/mol compared to 21.5°C and 94.3°C, respectively, for a molecular

mass of 2800 g/mol.

 For FR-PLA-C30B: T

and T

are 46.7°C and 150°C, respectively, for a molecular

mass of 12500 g/mol compared to 11.5°C and 83.9°C, respectively, for a

molecular mass of 2500 g/mol.

 The lower M

reported during ageing for FR-PLA-C30B compared to FR-PLA can

be held responsible for the lower T

and T

values observed for this intumescent

formulation.

This clearly demonstrates the importance of the molecular mass of FR-PLAs on the behavior

of T

and T

during ageing (cf. Fox and Flory relationship Equation 30, p.94). Regarding the

graph plotting T

as a function of M

of FR-PLA and FR-PLA-C30B for different ageing exposure

conditions and durations (Figure 74), it appears that for both formulations T

is stable at the

beginning of the exposure and then decreases when M

reaches the critical molecular mass

(i.e. 20000 g/mol). T

follows the same evolution as T

during ageing. The high mobility of the

polymer chains due to ageing affects the backbone flexibility and intermolecular interactions,

leading to the decrease of T

[38]. Compared to PLA, the lower M

before ageing of both FR-

PLAs permits to reach quickly the critical M

from which T

and T

start to decrease.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter IV: Influence of flame retardants additives on the ageing of PLA

135

Figure 74: Correlation between the decrease of T

and M

of FR-PLA and FR-PLA-C30B as a

function of ageing under T/UV (●), T/RH (▪) and T/UV/RH (∆).

It is also observed that whatever ageing conditions, the decrease of both T

and T

is more

pronounced in the case of FR-PLA-C30B compared to FR-PLA. The most probable explanation

is that the lower M

reported during ageing for FR-PLA-C30B can be held responsible for the

lower T

and T

values. However, another assumption to explain this phenomenon is that

Cloisite 30B catalyzes the degradation of FR-PLA-C30B due to the presence of hydroxyl groups

in the structure of the clay [237, 238]. This last hypothesis will be investigated in the next part.

Knowing the behavior of the molecular mass and thermal properties (i.e. T

and T

) of FR-

PLAs during ageing, the crystallization behavior is the last physico-chemical property to

investigate in order to compare the performance of FR-PLAs and PLA during ageing. Hence,

the next section is dedicated to the effect of fillers on the crystallization of FR-PLA and FR-PLA-

C30B as well as to their behavior during ageing.

Crystallization behavior

2.3.1. Crystallization comportment of unaged FR-PLAs

The crystallization behavior of FR-PLA and FR-PLA-C30B, studied using DSC, is compared to

the one of PLA. Crystallinity (χ) of the different formulations are reported in Table 21 and were

calculated according to Equation 24 (p.76).

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter IV: Influence of flame retardants additives on the ageing of PLA

136

Table 21: Crystallinity (χ) data for PLA, FR-PLA and FR-PLA-C30B as a function of M

Material name

(g/mol)

χ (%)

PLA

68000

6.5

FR-PLA

29000 (-57%)

11.5 (+5)

FR-PLA-C30B

27000 (-60%)

12.5 (+6)

Unfilled PLA is almost amorphous with a relatively low crystallinity of about 6.5%. However,

it appears that when fillers are added into PLA, the crystallinity of the material increases,

reaching 11.5% and 12.5% for FR-PLA and FR-PLA-C30B, respectively. It seems that Melamine,

APP in particular and Cloisite 30B to a less extent, promote the crystallization of the FR-

materials. Two assumptions can be made to explain this effect on crystallinity:

 FR fillers act as nucleating agents by promoting the crystallization of the

materials [25, 36, 39, 239-242].

 The lower M

of FR-PLA and FR-PLA-C30B before ageing promotes the

crystallization of the materials. Indeed, polymers of low molecular weight can

crystallize more easily compared to those of high molecular weight [66, 81, 231,

243-245]. This is probably due to the fact that smaller molecules have a higher

mobility and can thus easily form crystals

XRD experiments were performed in order to determine if fillers have an influence on the

kind of crystals formed during crystallization (Figure 75).

Figure 75: WAXD patterns of PLA (●), FR-PLA (●) and FR-PLA-C30B (●).

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter IV: Influence of flame retardants additives on the ageing of PLA

137

PLA presents a typical amorphous WAXD patterns due to its low crystallinity (i.e. 6.5%). FR-

PLA and FR-PLA-C30B exhibit similar WAXD patterns. Characteristic diffraction peaks of FR

fillers are identified: (i) Melamine peaks at 2θ = 17.9°, 20.1°, 22.1°, 26.1°, 27.5°, 29.1°, 30.6°

and 32.8° [246, 247]; (ii) APP peaks at 2θ = 14.8°, 15.5°, 20.1°, 22.1°, 26.1°, 27.5°, 29.1°, 30.6°

and 32.8° [248-250]; (iii) The characteristic peak of Cloisite 30B at 2θ = 5.0° [36, 251] is not

observed due to the relatively low loading of organoclay (i.e. 1 wt.-%). No characteristic peak

of crystallized PLA is observed in the WAXD patterns of the two FR formulations. Indeed, the

crystallinity of PLA versus FR fillers is certainly too low to see these peaks and thus it is not

possible to determine under which form PLA crystallizes.

2.3.2. Influence of ageing exposure on crystallization behavior of FR-PLAs

The evolution of crystallinity (χ) for FR-PLA and FR-PLA-C30B was then studied as a function

of ageing time for T/UV, T/RH and T/UV/RH exposure (Table 22 and Table 23 for FR-PLA and

FR-PLA-C30B, respectively). Focusing firstly on FR-PLA, results reported in Table 22 and Figure

76 plotting the behavior of χ as a function of ageing duration both evidence that crystallinity

increases when FR-PLA is exposed to T/UV, T/RH and T/UV/RH. Indeed, χ increases by 11.5%

and 43% after 125 days exposure to T/UV and 105 days exposure to T/RH, respectively.

However, the most significant increase is observed for FR-PLA aged under T/UV/RH conditions,

where a 53.5% increase in crystallinity is achieved after 90 days, corresponding to a χ of 65%

and to the lower M

of the material (i.e. 2800 g/mol).

Table 22: Evolution of Crystallinity for FR-PLA as a function of M

for different ageing

durations under T/UV, T/RH and T/UV/RH exposure.

Ageing

time

(days)

T/UV

(g/mol)

T/UV

χ (%)

T/RH

(g/mol)

T/RH

χ (%)

T/UV/RH

(g/mol)

T/UV/RH

χ (%)

29000

11.5

29000

11.5

29000

11.5

26000 (-9%)

14.0 (+2.5)

24300 (-16%)

15 (+4.5)

17500 (-39%)

27.0 (+15.5)

24100 (-14%)

16.0 (+4.5)

18400 (-34%)

19.0 (+7.5)

13000 (-54%)

36.0 (+24.5)

21900 (-22%)

17.0 (+5.5)

15800 (-44%)

28.0 (+16.5)

8800 (-69%)

43.0 (+31.5)

21200 (-26%)

18.0 (+6.5)

12400 (-56%)

31.0 (+19.5)

5300 (-80%)

49.0 (+37.5)

19600 (-30%)

20.0 (+8.5)

10000 (-65%)

38.0 (+26.5)

4000 (-86%)

60.0 (+48.5)

17400 (-38%)

21.5 (+10)

6900 (-75%)

43.0 (+31.5)

2800 (-90%)

65.0 (+53.5)

105

15500 (-45%)

22.0 (+10.5)

4300 (-85%)

54.5 (+43)

125

13500 (-52%)

23.0 (+11.5)

Regarding FR-PLA-C30B formulation, an increase in crystallinity is also observed (Table 23

and Figure 76). Indeed, χ is enhanced by 14.5% and 52.5% after 125 days exposure to T/UV

and 105 days exposure to T/RH, respectively. However, as for FR-PLA, the most significant

increase is observed for exposure under T/UV/RH conditions, where an increase in crystallinity

of 63.5% is achieved after 90 days, corresponding to a χ of 76%. This value of crystallinity

corresponds to the lower M

observed for the material (i.e. 2500 g/mol) when exposed to

T/UV/RH.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter IV: Influence of flame retardants additives on the ageing of PLA

138

Table 23: Evolution of Crystallinity for FR-PLA-C30B as a function of M

for different ageing

duration under T/UV, T/RH and T/UV/RH exposure.

Ageing

time

(days)

T/UV

(g/mol)

T/UV

χ (%)

T/RH

(g/mol)

T/RH

χ (%)

T/UV/RH

(g/mol)

T/UV/RH

χ (%)

27000

12.5

27000

12.5

27000

12.5

25100 (-7%)

14.5 (+2)

21600 (-20%)

15.5 (+3)

15000 (-45%)

32.0 (+19.5)

23700 (-12%)

18.0 (+5.5)

17600 (-35%)

23.0 (+10.5)

9700 (-64%)

41.0 (+28.5)

21500 (-20%)

19 (+6.5)

12200 (-55%)

34 (+21.5)

5800 (-79%)

54.0 (41.5)

20900 (-23%)

21.0 (+8.5)

9800 (-64%)

42 (+29.5)

4300 (-84%)

59.0 (+46.5)

19100 (-29%)

22.0 (+9.5)

7900 (-71%)

50 (+37.5)

3400 (-88%)

71.0 (+58.5)

17000 (-37%)

23.0 (+10.5)

5100 (-81%)

57 (+44.5)

2500 (-91%)

76.0 (+63.5)

105

14700 (-54%)

26.0 (+13.5)

4100 (-85%)

65 (+52.5)

125

12500 (-54%)

27.0 (+14.5)

Figure 76: Evolution of crystallinity of FR-PLA (a) and FR-PLA-C30B (b) as a function of ageing

time, for T/UV (●), T/RH (▪) and T/UV/RH (∆) exposure.

These results seem to indicate that the increase in crystallinity is linked to the decrease of

the molecular mass, which is in accordance both with literature [5, 27, 29, 38, 204] and the

results reported for PLA in chapter III (p.98). Indeed, during degradation, the molecular mass

decreases and due to a chemi-crystallization process [5, 27, 29, 38, 204] the crystallinity

increases. It is also observed that the increase in crystallinity is more pronounced for FR-PLA-

C30B than for FR-PLA (Table 22 and Table 23). In fact, χ increases by 53.5% and 63.5% for FR-

PLA and FR-PLA-C30B after 90 days exposure to T/UV/RH, respectively. This phenomenon can

be explained by the lower M

reached by FR-PLA-C30B during ageing. Indeed, when plotting

crystallinity as a function of M

for FR-PLA and FR-PLA-C30B (Figure 77) it appears that the

lower the M

, the higher the crystallinity: for example, after 105 days exposure to T/RH χ

reaches 54.5% and 65% when M

is 4300 and 4100 g/mol for FR-PLA and FR-PLA-C30B,

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter IV: Influence of flame retardants additives on the ageing of PLA

139

respectively. On the other hand, it is possible that organoclays acts as strong nucleating

agents, altering the materials crystallization behavior during ageing by promoting the

crystallization of FR-PLA-C30B [25, 38, 39].

Figure 77: Correlation between the decrease of crystallinity and M

of FR-PLA and FR-PLA-

C30B as a function of ageing under T/UV (●), T/RH (▪) and T/UV/RH (∆).

It was previously demonstrated that the increase in crystallinity of PLA during ageing is

linked to the formation of α’-form crystals (p.98). Hence, XRD was performed to identify if the

presence of FR fillers change the nature of crystals formed during ageing exposure. The WAXD

patterns obtained by XRD for unaged and 60 days aged FR-PLA and FR-PLA-C30B under T/UV,

T/RH and T/UV/RH conditions are shown in Figure 78. Before ageing, both FR-PLA and FR-PLA-

C30B exhibit the characteristic diffraction peaks of Melamine and APP that have been

previously reported (p.135). Even if these peaks are also observed for both materials after

60 days ageing, the appearance of two new diffraction peaks is observed for both FR-PLAs.

These two new peaks that appear at 2θ = 16.6° and 18.8° are characteristic of α’-form crystals

of PLA [202, 203]. As for PLA, no additional peaks characteristic of α-form crystals are observed

during ageing at 2θ = 12.5°, 14.8°, 22.5°, 27.3° and 29.2°.

The two α’-form characteristic peaks show higher intensity for materials exposed to

T/UV/RH conditions compared to those exposed to T/RH and T/UV. Moreover, regarding one

kind of exposure, the intensities of α’-form crystals peaks are higher for FR-PLA-C30B

compared to FR-PLA. This is in accordance with the evolution of the crystallinity observed by

DSC.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter IV: Influence of flame retardants additives on the ageing of PLA

140

Figure 78: WAXD patterns of FR-PLA and FR-PLA-C30B obtained by XRD, before (●) and after

60 of ageing under T/UV (●), T/RH (●) and T/UV/RH (●).

DSC and XRD analyses both evidence the increase in crystallinity when FR-PLA and FR-PLA-

C30B are exposed to T/UV, T/RH and T/UV/RH conditions. As for PLA, α’-form crystals appear

during ageing which formation is linked to (i) the decrease of M

and (ii) chemi-crystallization

process occurring during ageing. This means that presence of FR fillers doesn’t impact the

nature of crystal formed during ageing. The lower M

of FR-PLAs before ageing and the

nucleating effect of fillers promote the crystallization of both FR materials. Moreover, the

lower M

of FR-PLA-C30B during ageing, compared to the one of FR-PLA, leads to a material

of higher crystallinity for the same ageing period.

Conclusion

It has been demonstrated in this part that the incorporation of FR additives (i.e. Melamine,

APP and C30B) impact the physico-chemical properties of PLA at two different levels: (i) before

and (ii) after ageing of the materials.

Indeed, even before ageing these FR additives can modify some properties of PLA:

 The incorporation of fillers reduces the molecular mass of PLA, due to physical

and/or chemical processes.

 The molecular mass is known to be a parameter affecting T

. Thus the decrease

of M

can be held responsible for the decrease of T

observed when PLA is filled

with FR fillers. Moreover, the plasticizing effect of fillers can also be a

reasonable assumption to explain the lower T

of FR-PLAs.

 As well as T

, the crystallinity is also dependent on the molecular mass of the

material. Thus, the lower M

of FR-PLA and FR-PLA-C30B before ageing

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter IV: Influence of flame retardants additives on the ageing of PLA

141

promotes the crystallization of the materials, explaining why the crystallinity of

FR-PLAs are higher than those of PLA. In addition, the nucleating character of

FR fillers can slightly promotes the crystallization of the materials

Secondly, it was also reported that effects of ageing are more pronounced when FR fillers

are incorporated into the PLA matrix:

 Voids, fractures and a white layer are observed at the surface of FR-PLAs until

their complete disintegration when exposed to T/RH and T/UV/RH conditions.

These phenomena are observed when M

(± 20000 g/mol) is reached.

 Only the 60 days exposure to T/UV appear not to be too damaging for the visual

integrity of FR-PLA and FR-PLA-C30B. In fact, after 60 days M

is still not

reached.

 As for neat PLA, it was evidenced that the molecular mass of FR-PLA and FR-

PLA-C30B decreases when exposed to T/UV, T/RH and T/UV/RH conditions.

 The behavior of the molecular mass has a direct impact on the physico-

chemical properties of FR materials during ageing. Indeed, the decrease of M

leads to a decrease in both T

and T

as well as an increase in crystallinity.

 Due to the lower M

of FR-PLAs before ageing, FR-PLA and FR-PLA-C30B reach

more rapidly the critical molecular mass (M

± 20000 g/mol) from which the

material starts to disaggregate and to lose its thermal properties. Thus

explaining why FR-PLAs are more degraded after 125 days of ageing compared

to PLA.

Hence, it is obvious that FR fillers impact PLA before and after ageing. Furthermore, the

molecular mass appears to be a key parameter in the durability of PLA and FR-PLAs as it

governs the structural integrity as well as the behavior of thermal properties and crystallinity

during ageing. However, while the mechanisms of degradation of neat PLA have been

elucidated in the previous chapter (p.104), the impact of the addition of fillers on the

mechanisms of degradation of PLA still has to be investigated. This is why the next part of this

chapter is dedicated to the understanding of the impact of fillers on the mechanisms of

degradation occurring during exposure to T/UV, T/RH and T/UV/RH.

3. Mechanisms of degradation of PLA in presence of fillers

Literature as well as our previous investigations have demonstrated that mechanism of

degradation of PLA depends on the kind of accelerated ageing conditions (Chapter III, p.104)

[5, 27-29, 31-35, 38, 39, 52, 121, 123, 125, 153, 158-161]. In fact, when exposed to T/RH,

hydrolysis is the predominant mechanism. On the contrary, under T/UV and T/UV/RH, it was

reported that PLA is degraded via a radical mechanism, which is in competition with hydrolysis

in the case of T/UV/RH. Objective of the following part is to determine if flame retardant

additives influence the mechanisms of degradation of PLA.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter IV: Influence of flame retardants additives on the ageing of PLA

142

Surface chemical analysis of aged FR materials

Chemical surface analyses of the layer formed at the surface of both FR-PLAs (before and

after T/RH and T/UV/RH conditions) have been carried out by electron probe micro analysis

(EPMA) X-Ray mappings (1 μm

diffusion volume) and X-Ray photoelectron spectroscopy (XPS)

experiments (few μm

diffusion volume).

EPMA X-Ray mappings were carried out to characterize the surface of FR-PLA and FR-PLA-

C30B plates, looking at (i) nitrogen, present in both Melamine and APP, (ii) phosphorus,

characteristic of APP and (iii) silicon, characteristic of Cloisite 30B. The evolution of nitrogen

and phosphorus concentrations at the surface of FR-PLA plates (Figure 79) is reported before

ageing and after 60 days under T/UV (M

± 21300 g/mol), 28 days under T/RH (M

20000 g/mol) and 6 days under T/UV/RH exposure (M

± 20000 g/mol). Whereas no significant

evolution is noticed concerning FR-PLA aged under T/UV (Figure 79 (b)), pictures indicate that

the concentrations of nitrogen and phosphorus increase slightly at the surface of the plate

after 28 days exposure to (T/RH) (Figure 79 (c)) compared to unaged material (Figure 79 (a)).

This phenomenon is even more obvious (in particular for phosphorus) at the surface of plates

aged for 6 days under T/UV/RH (Figure 79 (d)).

Figure 79: EPMA pictures for unaged (a), 60 d T/UV (b), 28 d T/RH (c) and 6 d T/UV/RH (d)

aged FR-PLA.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter IV: Influence of flame retardants additives on the ageing of PLA

143

Regarding FR-PLA-C30B, the X-Ray mappings of silicon, nitrogen and phosphorus

(Figure 80) are reported before and after 60 days under T/UV (M

± 21100 g/mol), 28 days

under T/RH (M

± 20000 g/mol) and 6 days under T/UV/RH exposure (M

± 20000 g/mol). As

well as for FR-PLA, no significant evolution is noticed concerning FR-PLA-C30B aged under

T/UV condition (Figure 80 (b)). Pictures after 28 days exposure to T/RH show that the

concentrations of silicon, nitrogen and phosphorus increase at the surface of the plates (Figure

80 (c)) compared to the unaged FR-PLA-C30B (Figure 80 (a)). This increase in fillers

concentration at the surface of the plates is quite obvious after 6 days exposure to T/UV/RH

(Figure 80 (d)) and much more visible than the one for FR-PLA (Figure 79 (d)).

Figure 80: EPMA pictures for unaged (a), 60 d T/UV (b), 28 d T/RH (c) and 6 d T/UV/RH (d)

aged FR-PLA-C30B.

These X-Ray mappings confirm that when the critical molecular mass (± 20000 g/mol) is

reached, the combination of water and FR fillers leads to the formation of the white layer

observed at the surface of aged FR-PLAs (cf. p.122). Moreover, this effect is enhanced by UV

exposure. The layer is composed of nitrogen, phosphorus and silicon which are characteristic

elements of FR fillers incorporated into the PLA matrix (i.e. Melamine, APP and Cloisite 30B).

In addition to EPMA analysis, X-Ray photoelectron spectroscopy was also performed on FR-

PLA and FR-PLA-C30B plates before and after 60 days exposure to T/UV, 28 days to T/RH and

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter IV: Influence of flame retardants additives on the ageing of PLA

144

6 days to T/UV/RH conditions. The atomic percentages (At %) of carbon (C), oxygen (O),

nitrogen (N), phosphorus (P) and silicon (Si) at the extreme surface of FR-PLAs plates were

then quantified (Table 24 and Table 25 for FR-PLA and FR-PLA-C30B, respectively). Whereas

the concentration of carbon decreases at the surface of aged materials when exposed to T/UV,

T/RH and T/UV/RH conditions, it appears that the concentration of oxygen, nitrogen,

phosphorus and silicon increases. As an example, C % at the surface of the plates is decreased

by 17.6% and 22.7% for FR-PLA and FR-PLA-C30B, respectively, after 6 days exposure to

T/UV/RH conditions. Meanwhile, (i) O % increases by 13.8 and 17.9%, (ii) P % increases by

2.5% and 2.7%, (iii) N % is increased by 1.8% and 2.1% for FR-PLA and FR-PLA-C30B,

respectively and (iv) Si % reaches 0.8 % for FR-PLA-C30B after 6 days exposure to T/UV/RH

conditions.

Table 24: At % on the surface of FR PLA determined by XPS as a function of ageing.

Ageing

(type/days)

C 1s (At %)

O 1s (At %)

N 1s (At %)

P 2p (At %)

83.4

16.2

0.4

T/UV 60 d

78.9 (-4.5)

19.5 (+3.3)

0.7 (+0.3)

0.9

T/RH 28 d

67.4 (-16)

29.4 (+13.2)

1.4 (+1)

1.8

T/UV/RH 6 d

65.8 (-17.6)

30 (+13.8)

1.8 (+1.4)

2.5

Table 25: At % on the surface of FR PLA-C30B determined by XPS as a function of ageing.

Ageing

(type/days)

C 1s (At %)

O 1s (At %)

N 1s (At %)

P 2p (At %)

Si 2p (At %)

83.9

15.1

0.8

0.2

T/UV 60 d

77.5 (-6.4)

20 (+4.9)

1.3 (+0.5)

1.2 (+1)

T/RH 28 d

65 (-18.9)

30.9 (+15.8)

1.7 (+0.9)

2 (+1.8)

0.4

T/UV/RH 6 d

61.2 (-22.7)

33 (+17.9)

2.1 (+1.3)

2.9 (+2.7)

0.8

These results, which demonstrate the higher concentration of nitrogen, phosphorus and

silicon at the surface of aged materials are in accordance with the observations done by EPMA

X-Ray mappings. It also appears that relative humidity exposure increases the concentration

of fillers at the surface of aged samples. Hence, the role of fillers during hydrolysis is

investigated in the next section.

Role FR additives during hydrolysis

3.2.1. Hydrolysis of flame retardant additives

Whereas the presence of water was evidenced to play a key role during the formation of

the white layer during ageing, it is not clear if this layer corresponds to FR fillers on the surface

or to secondary products due to their degradation (hydrolysis). Hence, in order to identify

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter IV: Influence of flame retardants additives on the ageing of PLA

145

structural changes of additives during ageing in presence of relative humidity, Melamine, APP

and Cloisite 30B were exposed to T/RH atmosphere for 125 days. FTIR-ATR analyses were

performed on powder all along the exposure to T/RH.

Firstly, focusing on Melamine (Figure 81), the spectra can be classified into different

regions, which correspond to the following bands assignment [252, 253]: -N-H (primary

amine), stretching and bending from 2800 cm

-1

to 3500 cm

-1

, -N-H (primary amine) stretching

located at 1650 cm

-1

, C=N ring vibration observed at 1546 cm

-1

, -C-N triazine stretching and

bending at 1460 cm

-1

, 1430 cm

-1

and 1020 cm

-1

and C=N bending of trizaine ring located at

800 cm

-1

. These characteristics peaks are also observed after 60 and 125 days of exposure to

T/RH with no appearance of new peaks, indicating that Melamine is not hydrolyzed in

Ammeline, Ammelide or Cyanuric acid in these relative humidity conditions [254-256].

Figure 81: FTIR spectra (ATR) of Melamine as function of ageing exposure under T/RH.

Spectra of APP before ageing and after 60 days and 125 days exposure to T/RH conditions

are shown in Figure 82. APP like many other polyphosphates is known to be sensitive to

hydrolysis [257-260]. In fact, APP is initially water insoluble and becomes soluble when its

molecular weight decreases, thus leading to the formation of polyphosphoric acid (Figure 83)

[175]. However, in our study, it seems that APP is not hydrolyzed (Figure 82), as similar

characteristics peaks of APP are observed [261, 262] before and after ageing: -N-H (primary

amine) bending from 2800 cm

-1

to 3200 cm

-1

, NH

located at 1710 cm

-1

, 1670 cm

1 and

1470 cm

-1

, -P=O bands identified at 1240 cm

-1

, the region between 800 cm

-1

and 1100 cm

-1

assigned to asymmetric and symmetric vibration of P-O. The peak located at 795 cm

-1

assigned to P-O-P bands and should decrease during hydrolysis, which is not the case here.

Moreover, the characteristic peaks of polyphosphoric acid [259, 260], i.e. -OH in the range of

3200-3600 cm

-1

and -P-OH at 898 cm

-1

are not observed.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter IV: Influence of flame retardants additives on the ageing of PLA

146

Figure 82: FTIR spectra (ATR) of APP as function of ageing exposure under T/RH.

Figure 83: Hydrolysis reaction of APP, according to Chen et al. [175].

As it is quite difficult to distinguish -P-O-P- and -P-OH or-N-H and -OH regions in FTIR,

solid state NMR analyses were carried out on FR-PLA and FR-PLA-C30B to support the FTIR

results. Experiments were performed before ageing, on APP after 105 days exposure to T/RH

and on the white layer collected on the surface after 28 days exposure to T/RH conditions and

6 days exposure to T/UV/RH conditions (Figure 84). Whereas APP generally shows two peaks

located at -22 and - 24 ppm, corresponding to a well crystallized structure with Q

phosphorus

units contained in polyphosphate chains, the APP spectrum depicted in Figure 84(a) and (b)

exhibits one single band located at -22 ppm (P-O-P links, here in polyphosphates) [258, 263].

Moreover, this spectrum show an amorphization due to the form of the band. Before ageing,

both FR-PLA and FR-PLA-C30B exhibit this single band at - 22ppm. After 28 days of ageing

under T/RH and 6 days under T/UV/RH, APP is also visible on the white layer formed on FR-

PLA and FR-PLA-C30B and the two peaks characteristic of Q

units of polyphophoric acid

substituting NH

) located at -25 and -27 ppm [258] are not observed (Figure 84 (a) and

(b)). This tendency is confirmed by the spectrum of APP exposed 105 days to T/RH ageing,

where only the band at -22 ppm is observed. Hence, this tends to confirm that (i) APP is not

degraded or very partially, explaining why polyphosphoric acid is not detected and that (ii) FR

fillers are found at the surface of FR-PLAs plates after ageing. It is possible that after 125 days

of ageing APP is still not sufficiently degraded to observe its hydrolysis.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter IV: Influence of flame retardants additives on the ageing of PLA

147

Figure 84:

P Solid-state NMR spectra of FR-PLA (a) and FR-PLA-C30B (b) before and after

105 days and 90 days exposure to T/RH and T/UV/RH conditions, respectively.

FTIR spectra of Cloisite 30B before and after 60 days and 125 days exposure to T/RH

conditions are shown in Figure 85. Characteristic peaks of this organomodified

montmorillonite [264, 265] are similar before and after ageing: -OH structural hydroxyl group

stretching at 3610 cm

-1

, -C-H asymmetric and symmetric stretching at 2920 cm

-1

and

2850 cm

- 1

, respectively. The band located at 1610 cm

-1

can be attributed to -C-OH

deformation, the one at 1120 cm

-1

corresponds to Si-O stretching. The bands in the range of

880 to 980 cm

-1

correspond to Al-FeOH deformation and the band located at 790 cm

-1

is also

attributed to Si- O stretching. Some papers report the catalytic effect of Cloisite 30B on the

degradation of polymers due to the structure of the clay [237, 238]. However, (i) the behavior

of the clay reported by FTIR during ageing and (ii) previous results concerning the evolution of

the physico-chemical properties of FR-PLAs (p.125) seem to indicate that these clays do not

accelerate the hydrolysis of PLA.

Figure 85: FTIR spectra (ATR) of Cloisite 30B as function of ageing exposure under T/RH.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter IV: Influence of flame retardants additives on the ageing of PLA

148

Moreover, even if the results are not presented, a parallel study (Appendix 3, p.221) was

also led during this work concerning the effect of the type of nanoparticle on the hydrolytic

degradation of flame retarded PLA. In fact, literature reveals that the hydrophilic character of

clays, due to the type of surfactant, can enhance or not the degradation rate of PLA [25, 37,

38]. Hence, three kinds of Cloisite, i.e. 20A, 30B and Na

(hydrophilic character of C-Na

> C-

30B > C-20A) were incorporated into FR-PLA materials and then aged for 60 days under T/RH

conditions. The formation of voids, fractures and of a white layer was reported after 28 days

ageing for each formulation. Moreover, physico-chemical properties of the materials were

studied and it appears that the molecular mass, T

, T

and crystallinity of the three materials

exhibit the same behavior during ageing, whatever the formulation. All these results seem to

indicate that the hydrophilic character of organoclays does not play a particular role during

the degradation of the PLA matrix.

The behavior of FR fillers during ageing was confirmed by FTIR spectra recorded for FR-PLAs

during exposure to T/RH and T/UV/RH conditions (Appendix 4, p. 226)). As it was reported

that FR fillers are not degraded (or partially in the case of APP) during ageing in presence of

RH, it is confirmed that the white layer formed at the surface of FR materials during RH

exposure is mainly composed of Melamine, APP and C30B in the case of FR-PLA-C30B.

However, it is still not so obvious to determine if (i) the flame retardant additives migrate

through the PLA matrix during ageing or if (ii) fillers are just more accessible at the surface due

to the degradation of the matrix (both chain mobility and structural integrity of the material

are no longer sufficient to maintain the cohesion between fillers and polymer matrix).

3.2.2. Evaluation of the migration of flame retardant additives

In order to assess the possible migration of FR fillers to the surface of FR-PLAs, SEM and

EPMA analysis were carried out. Cross section SEM analyses have been performed on FR-PLA-

C30B exposed for 28 days to T/RH (Figure 86 (a)) and 6 days to T/UV/RH conditions (Figure

86 (b)). Fillers seem to be uniformly dispersed in the PLA matrix with no migration during

ageing. Thus, to validate this assumption, EPMA analyses were also performed on these

samples.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter IV: Influence of flame retardants additives on the ageing of PLA

149

Figure 86: SEM pictures in cross section obtained for FR-PLA-C30B aged 28 days under

T/RH (a) and 6 days under T/UV/RH (b).

Cross section EPMA X-Ray mappings performed on FR-PLA-C30B aged for 28 days under

T/RH and 6 days under T/UV/RH conditions are reported in Figure 87. Mappings in nitrogen,

phosphorus and silicon contained in T/RH aged (Figure 87 (a)) and T/UV/RH aged FR-PLA-C30B

(Figure 87 (b)) corroborate the assumption that fillers do not migrate to the surface of the

plates during ageing. In fact, for both materials, the concentration of nitrogen, phosphorus

and silicon is quite the same from the bottom to the top of the plates. The high concentration

of fillers on the surface of FR-PLAs is certainly related to the fact that fillers are more

accessible, which is confirmed by SEM pictures (Figure 88). In fact, SEM picture of FR-PLA-

C30B exposed for 28 days to T/RH conditions reveals that APP particles are clearly visible and

not completely embedded in the PLA matrix anymore.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter IV: Influence of flame retardants additives on the ageing of PLA

150

Figure 87: EPMA pictures in cross section for FR-PLA-C30B aged 28 days under T/RH and 6

days under T/UV/RH conditions.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter IV: Influence of flame retardants additives on the ageing of PLA

151

Figure 88: SEM pictures obtained for FR-PLA-C30B after 28 days exposure to T/RH.

As previously demonstrated (p.104), the hydrolytic degradation of PLA starts from the bulk

of the material. Investigations led during this chapter reveal that in presence of FR additives,

small voids, fractures and a layer composed of FR fillers appear at the surface of the

formulations exposed to RH. Moreover, those fillers such as APP are clearly visible and are not

completely embedded in the PLA matrix anymore. To explain these phenomena, the impact

of FR fillers on the hydrolysis of PLA can thus be described as follow:

 The M

of FR materials is lower than the one of PLA before ageing, due to the

incorporation of fillers into the polymer matrix.

 This lower M

before ageing allows reaching more rapidly the critical molecular

mass from which the materials starts to disaggregate and to lose its properties (T

, …), explaining why FR-PLAs are more degraded compared to PLA.

 When the critical molecular mass (M

) is reached during degradation, both chain

mobility and structural integrity of the material are no longer sufficient to maintain

the cohesion between fillers and polymer matrix. Moreover, it seems that

hydrolysis of PLA occurs at the interface between polymer matrix and fillers. This

can be explained by accumulation of water at this interface due to the hydrophilic

character of clays and APP [25, 38, 266]. Indeed, it is possible that these fillers

attract water, resulting in a higher local water concentration, increasing the

degradation of the PLA matrix. Additives are thus more accessible at the surface,

leading to the formation of a layer composed of Melamine, APP and C30B (in the

case of FR-PLA-C30B) until complete disintegration of the material into powder.

 On the other hand, it is possible that APP is partially hydrolyzed (even if results

suggest otherwise) [175, 258, 267] and thus explaining why particles are not

completely embedded in the PLA matrix anymore.

As for neat PLA, investigations led on the ageing of FR-PLAs show that the combination of

T, UV and RH is more drastic compared to sole T and RH exposure. In fact, whereas FR-PLAs

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter IV: Influence of flame retardants additives on the ageing of PLA

152

disintegrate into powder after 35 days exposure to T/RH, they need only 10 days when

exposed to T/UV/RH. This demonstrates the synergistic effect of the combination between UV

and RH on the degradation of FR-PLAs. The next step consists in elucidating the degradation

mechanism of FR-PLAs exposed to UV rays and to understand the role of fillers during such a

degradation.

Impact of FR additives during UV irradiation of PLA

It was demonstrated in chapter III (p.106) that UV degradation of PLA proceeds via a radical

mechanism. This latter leads to the formation of secondary products such as anhydrides,

alcohols or carboxylic acid until the complete degradation of PLA. The objective is now to

determine the impact of FR fillers on UV degradation of PLA.

Since the UV irradiation of PLA leads to the formation of biradical species, EPR experiments

were led both on PLA and FR-PLAs in order to study the kinetics of formation of these radicals

when FR fillers are incorporated into the PLA matrix (Figure 89). EPR spectra recorded at 50°C

for the three materials clearly show that less radicals are formed during UV irradiation when

FR fillers are incorporated into the materials. Indeed, whereas the sensitivity of radicals’

formation is around 300000 for PLA after 6 hours of exposure, it is only around 125000 for

both FR-PLAs. Hence, it appears that the incorporation of flame retardant additives into PLA

matrix has a direct impact on the photo-oxidation of PLA. However, the presence of

Cloisite 30B into FR-PLA-C30B formulation does not seem to impact the formation of radical

species compared to FR-PLA.

Figure 89: Kinetic evolution of EPR signal as a function of time, for PLA (●), FR-PLA (●) and

FR-PLA-C30B (●).

To support these results, 2D hyperfine sub-level correlation (2D HYSCORE, Appendix 5,

p.228) measurements were performed on PLA (Figure 90) and both FR-PLAs (Figure 91) at

50°C. Neat PLA (Figure 90) shows a typical spectrum of the polylactide structure. In fact, the

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter IV: Influence of flame retardants additives on the ageing of PLA

153

spectrum shows a

H pattern centered at 14.5 MHz corresponding to CH

with two protons

coupled at respectively 10 MHz and 18 MHz corresponding to CH. At low frequency region, a

weak coupling with

C of 2 MHz (CH

) and one at 5 MHz (CH) can also be observed.

Figure 90: 2D HYSCORE recorded at 50°C for PLA.

Focusing on the 2D HYSCORE experiments carried out for FR-PLA and FR-PLA-C30B, the two

materials exhibit the same spectrum. Thus only FR-PLA-C30B is reported in Figure 91. The CH

protons of PLA at 10 and 18 MHz are not further observed. The second change occurs in the

C region with a coupling of one carbon at 4 MHz and two carbons of PLA coupled at 2 MHz

(CH

) and 5 MHz (CH). Additionally, a weak coupling of 1.8 MHz with

N and 6 MHz with

is observed at Larmor frequency of 2.6 MHz.

Figure 91: 2D HYSCORE recorded at 50°C for FR-PLA and FR-PLA-C30B.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter IV: Influence of flame retardants additives on the ageing of PLA

154

These results suggest an interaction between fillers and PLA that disrupts the formation of

radical species. In fact, it is possible that fillers inhibit and stabilize the formation of radicals,

thus preventing radical mobility. Some authors have already reported this kind of

phenomenon, particularly in presence of Melamine (as UV absorber) and organoclays [47, 48,

268-270]. The presence of FR-fillers in FR-PLAs can attenuate the light absorption (UV

penetration) within the polymer matrix, preventing the material from degradation. Moreover,

FR-fillers can avoid oxygen diffusivity in the mass of a specimen, thus leading to a starvation

of oxygen in the core of the material. The mobility of radicals created during irradiation is thus

reduced.

The effect of the combination between UV-rays and relative humidity during exposure of

PLA to T/UV/RH was previously elucidated (p.106). It was reported that in presence of water,

hydrolysis and radical mechanism occur simultaneously, reducing the amount of radicals

formed. Indeed, hydrolysis can disrupt their formation by degrading PLA, thus forming

secondary products sooner. In order to estimate the additional influence of relative humidity

during ageing of FR-PLA and FR-PLA-C30B, kinetic study of the formation of radical species for

both FR-PLAs as a function of exposure to T/UV/RH was carried out by EPR (Figure 92) and

compared to the results obtained for sole T/UV exposure.

Figure 92: Kinetic evolution of EPR signal for FR-PLA and FR-PLA-C30B, as a function of ageing

exposure to T/UV (●) or T/UV/RH (●) at 50°C.

As for PLA, this graph shows that the same quantity of radicals is susceptible to be formed

for both kinds of exposure until approximatively half an hour of irradiation. The quantity of

radicals then formed under T/UV exposure is higher than that formed under T/UV/RH

conditions. As in the case of neat PLA, UV and water are in competition during degradation.

Indeed, hydrolysis and radical mechanism occur simultaneously during exposure to T/UV/RH,

secondary products are thus formed sooner and can be easily decomposed. As expected, the

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter IV: Influence of flame retardants additives on the ageing of PLA

155

presence of relative humidity enhances the degradation of FR-PLAs due to the combination

between hydrolysis and radical mechanism. Whereas fillers protect the material against UV

degradation, the lower M

of both FR-PLA and FR-PLA-C30B before ageing is responsible for

the more pronounced degradation of the formulations.

The impact of FR fillers on the UV degradation of PLA can be summarized as follow:

 The M

of FR materials are lower than those of PLA before ageing, due to the

incorporation of fillers into the polymer matrix.

 FR additives protect the material against UV degradation, thus limiting radical

formation and mobility. This also explains why T/UV exposure is less drastic

compared to T/RH

 The exposure to T/UV/RH appears to be the most detrimental for both FR-PLAs.

Indeed, the degradation of the material is enhanced as hydrolysis and UV

mechanism occur simultaneously.

 For a given kind of exposure, the lower M

before ageing of FR-PLAs is responsible

for the sooner degradation than the one of neat PLA. Indeed, the critical molecular

mass from which the FR-materials start to disaggregate and lose their properties is

reached more rapidly.

4. Conclusion

Chapter IV was dedicated to the study of the impact of fillers on the ageing of PLA. It was

demonstrated that FR fillers impact PLA at three different levels: (i) before ageing, (ii) during

ageing and (iii) on the mechanism of degradation.

Even before ageing, it was showed that the incorporation of FR additives impact the

different properties of PLA:

 The most obvious effect is the reduction of the molecular mass of around 60% when

fillers are incorporated due to physical and/or chemical process (mechanical shear,

thermal stress, aminolysis …)

 T

is decreased due to the M

decrease and/or plasticizing effect of fillers

 Crystallinity is increased due to the lower M

and/or nucleating character of fillers

It has been evidenced that both kinds of ageing exposure and flame retardant additives

have a direct impact on the ageing of PLA (Figure 93). The effects of ageing are indeed more

pronounced when FR fillers are incorporated into the PLA matrix:

 Whereas only slight bleaching and cracking are observed for virgin PLA, voids,

fractures and a white layer appear at the surface of the T/RH and T/UV/RH aged

FR-PLAs when the critical molecular mass (M

) of 20000 g/mol is reached

(Figure 93).

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter IV: Influence of flame retardants additives on the ageing of PLA

156

 FR-PLAs are completely disintegrated into powder after 10 and 35 days of

exposure to T/UV/RH and T/RH conditions, respectively. Only the exposure to

T/UV appears not to be drastic for the visual integrity of FR-PLA and FR-PLA-

C30B as the critical molecular mass is still not reached (M

> M

) (Figure

93).

 The causes of these phenomena can be explained by the evolution of physico-

chemical properties of the materials during ageing. Indeed, the decrease of the

has a direct impact on the behavior of T

, T

and crystallinity (decrease of

and T

and increase in crystallinity).

 The most important point is that due to the already low molecular mass of FR-

PLAs before ageing (M

), these materials reach more rapidly the critical

molecular mass, thus explaining the faster degradation of the materials. As well

as for unfilled PLA, T/UV/RH exposure conditions are the most drastic

conditions for the ageing of FR materials followed by T/RH and T/UV (Figure

93).

Figure 93: Schematic representation of the effect of fillers and ageing exposure on the

degradation of FR-PLA-C30B.

It was then demonstrated that the incorporation of FR fillers has a direct impact on the

hydrolysis and radical mechanism during degradation:

 FR fillers protect the material against UV degradation, thus limiting the formation

of radical species and of secondary degradation products.

 In presence of water the critical molecular mass (M

) is quickly reached, at the

expense of the cohesion between filler and polymer matrix. Hydrolysis of FR-PLAs

occurs at the interface between matrix and fillers. This is certainly due to an

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter IV: Influence of flame retardants additives on the ageing of PLA

157

accumulation of water between the fillers-matrix interface, thus leading to the

formation of a layer composed of Melamine, APP and C30B (in the case of FR-PLA-

C30B) at the surface of FR-PLAs (Figure 93).

 The potential hydrolysis of APP could also explain why particles are not completely

embedded in the PLA matrix anymore.

 As demonstrated in the case of virgin PLA, combination between T, UV and RH

enhances the degradation of both FR-PLAs due to hydrolysis and radical mechanism

that occur simultaneously.

Finally, the faster degradation of FR-PLAs compared to the one of PLA is explained by the

lower molecular mass of FR-PLAs before ageing, demonstrating that molecular mass is a

crucial parameter for PLA and FR-PLAs as it governs the other properties of the material (T

, crystallinity …) and its durability.

The main goal of incorporating FR fillers into PLA is to improve its flame retardant

properties. The impact of these fillers on the mechanisms of degradation and properties of

PLA without and with ageing has been studied in this last chapter. It is now necessary to

determine and understand the effect of ageing on the fire properties of our materials. Hence,

the chapter V is dedicated to the fire retardancy of PLA and a special attention will be given

to the impact of ageing on the fire properties of materials.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

158

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter V: Impact of ageing on the flame retardant properties of FR-PLAs

159

Chapter V: Impact of ageing on the flame

retardant properties of FR-PLAs

Melamine, APP and Cloisite 30B were incorporated into PLA to increase its flame retardant

properties. In the first part of this chapter, the flame retardant properties of PLA and both FR-

PLAs are examined and the mechanism of intumescence is reminded. However, these

properties must be kept for the lifetime of the product designed, this is why impact of ageing

on FR properties is of primary importance. The second part of this chapter is thus dedicated

to the study of impact of ageing on the FR properties of PLA and FR-PLAs. In regards to

literature, this work is the first fundamental study on this topic.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter V: Impact of ageing on the flame retardant properties of FR-PLAs

160

1. Flame retardant performances of FR-PLAs

The two flame retarded materials studied in our work (i.e. FR-PLA and FR-PLA-C30B) have

been developed during previous researches carried out in our lab [12]. The aim of the project

was to develop and optimize an intumescent polylactide exhibiting a non-burning behavior

upon heating. The intumescent mechanism as well as FR properties of such materials have

already been elucidated [12, 24, 61]. However, it is necessary to evaluate the thermal and fire

properties of our own formulations, before investigating their behavior during ageing. Hence,

this part is focused on the study of the thermal stability and flame retardant properties of PLA

and both FR-PLAs using TGA, MLC and LOI. Furthermore, the FR mechanism of action of the

intumescent PLAs will be reminded.

Thermal stability of FR-PLAs

Previous researches have demonstrated that the intumescent mechanism mainly takes

place in condensed phase [61]. Hence, thermal stability plays a fundamental role in the

mechanism of protection. In this way, the thermal stability of our FR-PLAs was studied using

TGA, and compared to that of neat PLA.

Data collected from the TGAs of PLA and FR-PLAs are reported in Table 26. The onset

temperature (T

10%

) corresponding to 10 wt.-% degradation of each material, the temperatures

corresponding to its maximum rates of degradation (DTG) and the residual weight obtained

at 800°C are considered in these tables. The thermal stability of pure compounds, already

investigated by Fontaine et al. [12], is reported in Appendix 6, p.229.

Table 26: TGA data (10°C/min under air) obtained for PLA, FR-PLA and FR-PLA-C30B.

Material

name

(g/mol)

10%

(°C)

Max DTG

(°C)

Max

DTG

(°C)

Max

DTG

(°C)

Residual weight

(wt.-%) at

800°C

PLA

68000

332

358

FR-PLA

29000

(-57%)

312 (-20)

346 (-12)

555

730

9.5

FR-PLA-C30B

27000

(-60%)

308 (-24)

343 (-15)

560

740

The TGA curves obtained for PLA (Figure 94) show that the polymer decomposes in one

step from 280°C with a maximum of degradation at 358°C with no residue left at 800°C.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter V: Impact of ageing on the flame retardant properties of FR-PLAs

161

Figure 94: TGA curves of PLA (●), FR-PLA (●) and FR-PLA-C30B (●) under air flow at 10°C/min.

FR-PLA and FR-PLA-C30B degrade in three steps (Figure 94). For FR-PLA, the first

degradation step occurs between 250°C and 400°C with 70% weight loss. The second step

occurs between 400°C and 650°C with 15% weight loss. The last one starts from 650°C until

750°C with 9.5 wt.-% residue at 800°C. The TGA curve of FR-PLA-C30B is similar to that of FR-

PLA until 400°C. However, a significant difference is observed after 400°C. Indeed, after 400°C,

FR-PLA-C30B degrades by forming a larger amount of stable residue, resulting in a final residue

of 12 wt.-% against 9.5 wt.-% for FR-PLA. The different compounds interact together (chemical

reaction and/or physical interaction, lower M

of FR-PLAs ), leading to a thermal degradation

of the PLA matrix at lower temperature in a first time [12, 61] (T

10%

is 312°C and 308°C for FR-

PLA and FR-PLA-C30B, respectively, against 332°C for PLA) but in a second time, FR additives

lead to a thermal stabilization of PLA with higher residues at the end of the experiment. The

destabilization is similar for both FR-PLAs, but stabilizing effect is more important in the case

of FR-PLA-C30B. Indeed, nanoparticles in PLA/Melamine/APP formulations can stabilize the

intumescent char [12, 24, 61]. C30B acts as a reinforcing agent of the intumescent coating,

allowing to increase its efficiency.

Reaction to fire

1.2.1. Behavior under mass loss cone calorimeter (MLC)

The aim of incorporating flame retardant fillers was to design a PLA with excellent fire

retardant properties (i.e. FR-PLA and FR-PLA-C30B). In order to investigate the fire retardant

properties of such materials, the mass loss cone calorimeter (MLC) is one of the well

established tools, providing data on the surface ignition and penetrative burning of a material.

Hence, the heat release rate (HRR) of the two intumescent formulations was measured by

MLC and compared to unfilled PLA (Figure 95 and Table 27).

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter V: Impact of ageing on the flame retardant properties of FR-PLAs

162

Figure 95: RHR curves as a function of time of PLA (●), FR-PLA (●) and FR-PLA-C30B (●).

Table 27: MLC values obtained for PLA, FR-PLA and FR-PLA-C30B.

Material name

Time to ignition (s)

pHRR (kW/m²)

THR (MJ/m²)

PLA

420

72.0

FR-PLA

45 (+6)

82 (-80%)

4.8 (-93%)

FR-PLA-C30B

43 (+4)

48 (-89%)

3.1 (-96%)

The virgin PLA ignites 39 s after the beginning of the experiment and reaches a peak of heat

release rate (pHRR) of 420 kW/m² with a total heat release (THR) of 72 MJ/m². For both FR-

PLAs, an intumescent char is obtained (Figure 96) and in all the cases the addition of fillers

decreases both pHRR and THR compared to the neat PLA.

Figure 96: Char residues of FR-PLA and FR-PLA-C30B.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter V: Impact of ageing on the flame retardant properties of FR-PLAs

163

Indeed, pHRR reaches 82 kW/m² (-80%) and 48 kW/m² (-89%) and THR reaches 4.8 MJ/m²

(-93%) and 3.1 MJ/m² (-96%) for FR-PLA and FR-PLA-C30B, respectively. However, it appears

that the time to ignition is similar to that of unfilled PLA (39 s for PLA compared to 45 s and

43 s for FR-PLA and FR-PLA-C30B, respectively). These results indicate that the intumescent

shield developed by FR-PLA and FR-PLA-C30B when exposed to a radiative heat flux is very

efficient, leading to an important decrease of both pHRR and THR. Concerning the char

residues (Figure 97), it seems that the one of FR-PLA-C30B is more compact and cohesive, but

less expanded than the one of FR-PLA. Moreover, whereas few flames were observed at the

surface of FR-PLA during MLC experiment, FR-PLA-C30B exhibits a non-burning behavior,

confirming that C30B acts as a good synergist for intumescent PLA [12, 24, 61].

Figure 97: Digital microscopy pictures and cross section of char residues of FR-PLA and FR-

PLA-C30B.

MLC results demonstrate that the intumescent char formed under radiative heat flux plays

an impressive protective role, with strong reductions of HRR and THR. The addition of a low

percentage (1 wt.-%) of nanoparticles such as C30B improves further the efficiency of the

protection by decreasing the pHRR by 89% compared to PLA. Regarding these results, we also

investigated the flammability of PLA and FR-PLAs using limiting oxygen index (LOI).

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter V: Impact of ageing on the flame retardant properties of FR-PLAs

164

1.2.2. Limiting oxygen index (LOI) test results

Flammability is one of the most important characteristics of materials in the field of flame

retardancy. In fact, when a material exhibits good performances during flammability scenario,

it should limit propagation of flames during a real fire. Limiting oxygen index (LOI) is one of

the scenarios commonly used to evaluate the flammability of a material. The results obtained

for PLA and both FR-PLAs are presented in Table 28.

Table 28: LOI values of PLA, FR-PLA and FR-PLA-C30B.

Material name

LOI (vol.-%)

PLA

FR-PLA

50 (+30)

FR-PLA-C30B

56 (+36)

The results obtained show that the combination of Melamine and APP in PLA provides very

high LOI values. LOI of FR-PLA reaches 50 vol.-% compared to 20 vol.-% for unfilled PLA.

Moreover, the incorporation of C30B into FR-PLA-C30B allows an increase of LOI values to

56 vol.-% due to the intumescent phenomenon. Intumescent PLAs exhibit high LOI values

compared to the unfilled PLA, which demonstrate their lower flammability compared to PLA

and proves the efficiency of FR fillers. In addition, these results show once again that

organoclay (C30B) has a synergistic effect with APP and Melamine when only a small amount

is incorporated into FR-PLA.

Mechanism of action of FR-PLAs

Previous researches reported by Fontaine et al. [24] and Gallos et al [61] have elucidated

the mechanism of action of the intumescent FR-PLA and FR-PLA-C30B. Indeed, the

combination between Melamine and APP into PLA (i.e. FR-PLA) leads to the formation of an

intumescent structure that is able to enhance the fire properties of the material during fire

scenarios. It is reported that PLA is thermally degraded in order to form aromatic and aliphatic

carbonyl compounds in the condensed phase. Other compounds such as lactide, carbon

dioxide, ester and aldehyde are released in the gas phase. APP is reported to be thermally

degraded by chain scission mechanisms forming phosphates. These latter present an acidic

character and react with the carbonyl compounds in the condensed phase. Phosphates also

permit the development of an intumescent phenomenon when combined with Melamine

(Figure 98).

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter V: Impact of ageing on the flame retardant properties of FR-PLAs

165

Figure 98: Intumescent mechanism, adapted from Gallos et al [61].

Viscosity of polymers during burning is a key point to control the development of

intumescent barriers [271]. Gallos et al. [61] reported that the viscosity of the FR-PLA upon

heating is adequate to lead to the formation of this intumescent thermal barrier. The addition

of Cloisite 30B allows increasing the viscosity of the blend and this increase is enough to

significantly impact the swelling kinetics. Indeed, both swelling rate and expansion of the char

of FR-PLA-C30B are decreased compared to FR-PLA. MLC and LOI results reveal that FR-PLA-

C30B exhibits better fire performances compared to FR-PLA. Indeed, whereas FR-PLA present

flames at the surface of the char when exposed to the radiative heat flux, FR-PLA-C30B

presents a non-burning behavior. Flames are extinguished 10 s maximum after ignition. This

is due to the fact that the structure of the two chars of FR-PLA and FR-PLA-C30B is different,

thus the release of gases is also different (Figure 99).

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter V: Impact of ageing on the flame retardant properties of FR-PLAs

166

Figure 99: Release of gas through the char of FR-PLA and FR-PLA-C30B, adapted from Gallos

et al [61].

In the clay containing formulation, C30B reacts with APP and forms aluminophosphate

species [24, 61]. These species can thermally stabilize the structure of the char, allowing to

obtain a more cohesive char. The role played by the clay in the improvement of the FR

performances can be summarized as follows: the Cloisite 30B permits to obtain a better

protection by an intumescent and more insulating char (less swelling but lower temperature).

The char also limits the “fuel” release due to its better cohesion. This limitation of fuel release

is a consequence of the diminution of the HRR since there is less fuel to feed the flame and

the sporadic flames observed extinguish spontaneously and quickly. This intumescent PLA

becomes then nonflammable.

Conclusion

In conclusion, the combination of APP and Melamine is an effective method to improve the

fire properties of PLA and to obtain a thermal shield via an intumescent phenomenon. The

combination of FR-PLA with nanoparticles such as Cloisite 30B permits to obtain a synergistic

effect which enhances the efficiency of the intumescent PLA process. Indeed, the

organomodified montmorillonite should increase the viscosity of the formulations which plays

a key role on the swelling kinetic and reacts with APP to form aluminophosphates contributing

to the modification of the char structure. These differences in terms of structure lead to a

more cohesive char. More homogeneous combustible gases are liberated at the surface of the

char, reducing their local concentration and preventing their combustion, while maintaining

the protection efficiency in terms of thermal isolation. As the mechanism of intumescence of

our FR materials as well their associated fire performances have been evidenced, the next part

is dedicated to the effect of ageing on the thermal stability and the reaction to fire of PLA and

both FR-PLAs.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter V: Impact of ageing on the flame retardant properties of FR-PLAs

167

2. Thermal stability and flame retardant properties after ageing

In this part, the effect of T/UV, T/RH and T/UV/RH ageing on the thermal stability of FR-PLA

and FR-PLA-C30B is investigated. Moreover, fire retardant properties (MLC and LOI) of PLA

and both FR-PLAs are examined all along the exposure to T/UV, T/RH and T/UV/RH conditions.

Finally, the influence of ageing on the mode of action of flame retarded materials will be

studied.

One may notice that it was not possible to perform LOI and MLC measurements after 35

days and 10 days ageing under T/RH and T/UV/RH, respectively, as the materials were

completely disintegrated into powder. Thus, the fire performances of FR-PLAs have been

evaluated until 60 days (M

± 21300 g/mol), 28 days (M

± 20000 g/mol) and 6 days (M

20000 g/mol) exposure to T/UV, T/RH and T/UV/RH conditions, respectively.

Effect of ageing on thermal stability of FR-PLAs

As well as for neat PLA in chapter III (p.93), the thermal stability of FR-PLAs as a function of

ageing conditions was evaluated using TGA. Data obtained are reported in Table 29 and Table

30 for FR-PLA and FR-PLA-C30B, respectively. The onset temperature (T

10%

), corresponding to

10 wt.-% degradation of the material, the temperatures corresponding to its maximum rates

of degradation (DTG), and the residual weight obtained at 800°C are considered in these

tables.

Table 29: TGA data (10°C/min under air) obtained after ageing of FR-PLA exposed to T/UV,

T/RH and T/UV/RH conditions.

Exposure /

time (days)

(g/mol)

10%

(°C)

Max DTG

(°C)

Max DTG

(°C)

Max DTG

(°C)

Residual

weight (wt.-%)

at 800°C

29000

312

346

555

730

9.5

T/UV 125 d

13 500

(-53%)

304 (-8)

345 (-1)

557 (+2)

732 (+2)

9.7 (+0.2)

T/RH 105 d

4 300

(-84%)

297 (-15)

333 (-13)

555

729 (-1)

9.0 (-0.5)

T/UV/RH 90 d

2 800

(-90%)

277 (-35)

325 (-21)

556 (+1)

732 (+2)

9.5

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter V: Impact of ageing on the flame retardant properties of FR-PLAs

168

Table 30: TGA data (10°C/min under air) obtained after ageing of FR-PLA-C30B exposed to

T/UV, T/RH and T/UV/RH conditions.

Exposure /

time (days)

(g/mol)

10%

(°C)

Max DTG

(°C)

Max DTG

(°C)

Max DTG

(°C)

Residual

weight (wt.-%)

at 800°C

27000

308

343

560

740

T/UV 125 d

12 500

(-54%)

305 (-3)

340 (-3)

561 (+1)

740

12.5 (+0.5)

T/RH 105 d

4 100

(-84%)

295 (-13)

327 (-16)

561 (+1)

742 (+2)

13.0 (+1)

T/UV/RH 90 d

2 500

(-91%)

258 (-50)

311 (-32)

562 (+2)

742 (+2)

13.0 (+1)

As well as unaged materials, aged FR-PLAs (Figure 100 (a) and (b)) degrade in three steps

and lead to a residue at 800°C. However, as for neat PLA (p.93), the onset temperature of

degradation of both FR-PLAs decreases upon heating. Indeed, T

10%

of FR-PLA (i) decreases

from 312°C before ageing to 304°C, 297°C and 277°C after 125 days exposure to T/UV,

105 days to T/RH and 90 days to T/UV/RH conditions, respectively. For FR-PLA-C30B (ii), T

10%

decreases from 308°C before ageing to 305°C, 295°C and 258°C after 125 days exposure to

T/UV, 105 days to T/RH and 90 days to T/UV/RH conditions, respectively.

Same phenomenon is reported concerning the behavior of DTG

. In fact, DTG

of FR-PLA (i)

decreases from 346°C before ageing to 345°C, 333°C and 325°C after 125 days exposure to

T/UV, 105 days to T/RH and 90 days to T/UV/RH conditions, respectively. DTG

of FR-PLA-C30B

(ii) decreases from 343°C before ageing to 340°C, 327°C and 311°C after 125 days exposure to

T/UV, 105 days to T/RH and 90 days to T/UV/RH conditions, respectively. However, it appears

that whatever the kind of ageing exposure, DTG

and DTG

remain similar after ageing for both

materials. Indeed, DTG

is 555°C ± 2°C and 560 C ± 2°C for FR-PLA and FR-PLA-C30B,

respectively, whereas DTG

is 730°C ± 2°C and 740°C ± 2°C for FR-PLA and FR-PLA-C30B,

respectively.

The residue left at 800°C is quite similar for FR-PLA and FR-PLA-C30B after ageing: whatever

the kind of ageing exposure it is around 9 wt.-% ± 0.5 wt.-% for FR-PLA and 12 wt.-% ± 1 wt.- %

for FR-PLA-C30B. One can notice that like unaged materials, the final residue of FR-PLA-C30B

is higher than that of FR-PLA after ageing.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter V: Impact of ageing on the flame retardant properties of FR-PLAs

169

Figure 100: TGA curves as function of time for FR-PLA (a) and FR-PLA-C30B (b), before ageing

(●), 125 days T/UV (●), 105 days T/RH (●) and 60 days T/UV/RH (●) exposure.

All these results indicate that the thermal stability of FR-PLAs materials is impacted during

ageing. The decrease of T

10%

and DTGA

, which correspond to the degradation of PLA, suggests

the formation of thermally weak bond (e.g. hydroperoxides, carboxylic acids and peroxides)

during the ageing exposure of the PLA matrix (chapter III, p.93). These ones can quickly initiate

the thermal degradation and then decrease the thermal stability of PLA [174, 190-192],

explaining the behavior of both T

10%

and DTGA

. On the other hand, knowing that DTGA

and

DTGA

are linked to the presence of FR fillers into FR-PLAs, their behavior reported during

ageing seems to indicate that FR fillers are not affected by ageing, which is in accordance with

observations previously reported in chapter IV (p.141).

Theoretical TGA curves of FR-PLA and FR-PLA-C30B calculated according to Equation 23

demonstrate that the residues obtained at 800°C are mainly due to the presence of FR fillers

into the PLA matrix (Table 31). Indeed, whereas no residue is left at 800°C for neat PLA, the

residues obtained by TGA experiment are 9.5 wt. % and 12 wt. % at 800°C for FR-PLA and FR-

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter V: Impact of ageing on the flame retardant properties of FR-PLAs

170

PLA-C30B, respectively. These values are close to the theoretical ones of 9.1 wt. % and

11.5 wt. % calculated for FR-PLA and FR-PLA-C30B, respectively.

Table 31: Comparison between theoretical and experimental residual weight (wt. %)

obtained at 800°c for FR-PLA and FR-PLA-C30B.

Formulation name

Residual weight (wt.-%) at 800°C

FR-PLA (theo)

9.1

FR-PLA (exp)

9.5 (+0.4)

FR-PLA-C30B (theo)

11.5

FR-PLA-C30B (exp)

12 (+0.5)

Thus, if the residue left at 800°C is mainly due to the presence of FR fillers, and considering

that only PLA matrix is affected by ageing, it is quite obvious that this residue does not change

during ageing. Hence, the limitation of fuel release due to the insulating char is still efficient

and the thermal protection is thus maintained during ageing.

As the effect of ageing on the thermal stability of intumescent PLAs is now evidenced, the

next section is dedicated to the effect of ageing on the fire performances of the materials

studied, in particular thanks to LOI test.

Limiting Oxygen Index (LOI) test results after ageing

The impact of ageing on flame retardancy of PLA was firstly assessed by LOI test. The values

obtained are summarized in Table 32, Table 33 and Table 34 for PLA, FR-PLA and FR-PLA-C30B,

respectively.

Table 32: LOI ranking of PLA as a function of exposure to T/UV, T/RH and T/UV/RH

conditions.

Ageing time (days)

LOI (vol.-%) T/UV

LOI (vol.-%) T/RH

LOI (vol.-%)

T/UV/RH

PLA

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter V: Impact of ageing on the flame retardant properties of FR-PLAs

171

Table 33: LOI ranking of FR-PLA as a function of exposure to T/UV, T/RH and T/UV/RH

conditions. The (+x) are calculated using the unaged formulation as reference.

Ageing time (days)

LOI (vol.-%) T/UV

LOI (vol.-%) T/RH

LOI (vol.-%)

T/UV/RH

FR-PLA

57 (+7)

52 (+2)

60 (+10)

52 (+2)

59 (+9)

54 (+4)

Table 34: LOI ranking of FR-PLA-C30B as a function of exposure to T/UV, T/RH and T/UV/RH

conditions. The (+x) are calculated using the unaged formulation as reference.

Ageing time (days)

LOI (vol.-%) T/UV

LOI (vol.-%) T/RH

LOI (vol.-%)

T/UV/RH

FR-PLA-C30B

64 (+8)

58 (+2)

67 (+11)

58 (+2)

67 (+11)

60 (+4)

Results show that whatever the kind of exposure, ageing does not impact the LOI value of

unfilled PLA which remains at 20 vol.-% (Table 32). However, regarding FR-PLA (Table 33) and

FR-PLA-C30B (Table 34), LOI values surprisingly increase when the formulations are exposed

to T/UV, T/RH and T/UV/RH conditions:

 Under exposure to T/UV: LOI increases by 4 vol.-% for both FR-PLAs after

60 days, from 50 to 54 vol.-% for FR-PLA and from 56 to 60 vol.-% for FR-PLA-

C30B.

 Under exposure to T/RH: LOI reaches 59 vol.-% (+9) and 67 vol.-% (+11) after

28 days for FR-PLA and FR-PLA-C30B, respectively.

 Under exposure to T/UV/RH: LOI rises by 10 vol.-% and 11 vol.-% to reach 60

and 67 vol.-% after 6 days for FR-PLA and FR-PLA-C30B, respectively.

It appears that during ageing, the LOI values of FR-PLA and FR-PLA-C30B surprisingly

increase. This increase clearly depends on the ageing conditions: it is more important for

ageing under T/UV/RH conditions than for T/RH and T/UV exposure. In fact, as an example,

LOI of FR-PLA-C30B slightly increases until 60 vol.-% after 60 days exposure to T/UV, whereas

it reaches 67 vol.-% after 28 days exposure to T/RH or only 6 days to T/UV/RH conditions.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter V: Impact of ageing on the flame retardant properties of FR-PLAs

172

It is noteworthy that a change in behavior of FR-PLA and FR-PLA-C30B is reported during

LOI tests (Figure 101). Indeed, it appears that the charring effect decreases, while dripping

occurs when ageing exposure time increases. Nevertheless, this behavior is more noticeable

for materials aged under T/UV/RH conditions compared to T/RH and T/UV.

Figure 101: Evolution of the behavior of LOI samples of FR-PLA and FR-PLA-C30B as a function

of ageing exposure to T/UV, T/RH and T/UV/RH conditions.

These investigations demonstrate that ageing has a positive effect on LOI values, by

reducing the flammability. This phenomenon is certainly due to a “runaway effect”. Indeed,

the combustible material escapes from the flame by dripping phenomenon which stops the

flaming of the samples. In order to better understand this phenomenon, the behavior of the

viscosity of materials during ageing was investigated. As mentioned previously, ageing leads

to a decrease of the molecular mass. The decrease in M

of a polymer is known to usually lead

to a decrease in the viscosity of the material [272] and this parameter is an important factor

that impacts the general fire retardant behavior of a material, particularly on LOI test [50, 61,

102, 273, 274]. In this aim, melt flow indexer (MFI) was performed on FR-PLAs before and all

along ageing exposure.

Viscosity of FR-PLAs

MFI was performed in order to investigate the changes in viscosity of FR-PLAs during

ageing. Indeed, the dynamic viscosity is known to be inversely proportional to MFI (viscosity

decreases when MFI increases) [272, 275]. Data obtained are reported in Table 35 and Table

36 for FR-PLA and FR-PLA-C30B, respectively.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter V: Impact of ageing on the flame retardant properties of FR-PLAs

173

Table 35: Viscosity data obtained for FR-PLA exposed to T/UV, T/RH and T/UV/RH conditions.

Ageing

time (days)

FR-PLA

(T/UV)

(g/mol)

(T/UV)

Viscosity

(Pa/s)

(T/RH)

(g/mol)

(T/RH)

Viscosity

(Pa/s)

(T/UV/RH)

(g/mol)

(T/UV/RH)

Viscosity

(Pa/s)

29000

1240

29000

1240

29000

1240

27100

(-3%)

1100

(-11%)

25800

(-11%)

750

(-40%)

21500

(-23%)

400

(-68%)

26000

(-10%)

910

(-27%)

21300

(-24%)

370

(-70%)

17500

(-39%)

120

(-90%)

25300

(-13%)

720

(-42%)

18400

(-37%)

160

(-87%)

24100

(-17%)

460

(-63%)

Table 36: Viscosity data obtained for FR-PLA-C30B exposed to T/UV, T/RH and T/UV/RH

conditions.

Ageing

time (days)

FR-PLA-

C30B

(T/UV)

(g/mol)

(T/UV)

Viscosity

(Pa/s)

(T/RH)

(g/mol)

(T/RH)

Viscosity

(Pa/s)

(T/UV/RH)

(g/mol)

(T/UV/RH)

Viscosity

(Pa/s)

27000

2010

27000

2010

27000

2010

26000

(-4%)

1600

(-20%)

24800

(-8%)

1000

(-49%)

22300

(-18%)

600

(-70%)

25100

(-7%)

1300

(-35%)

21200

(-22%)

450

(-78%)

15000

(-45%)

100

(-95%)

24500

(-10%)

800

(-60%)

15400

(-43%)

110

(-94%)

21800

(-19%)

430

(-79%)

One can firstly notice that the dynamic viscosity of FR-PLA-C30B is higher than that of FR-

PLA before ageing, thus even if the M

is lower. As already reported (p.164), this is due to the

incorporation of organomodified montmorillonite into the material [24, 61]. Generally,

intercalation, exfoliation and dispersion of clays in a polymer matrix can display an increase in

viscosity [276-281]. In order to evaluate the dispersion of organoclays in our materials,

transmission electron microscoscopy (TEM) analyses were thus performed on FR-PLA-C30B. It

appears that FR-PLA-C30B is constituted of many big agglomerates of clay platelets with width

superior to 0.5 μm (Figure 102 (a)). However, the material also present platelets of smaller

size with width of 200 nm (Figure 102 (b)). Due to their length and width, these agglomerates

are of intermediate size between micrometrical and nanometrical fillers. Indeed, although

these agglomerates have a dimension lower than one micron, they have a shape factor

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter V: Impact of ageing on the flame retardant properties of FR-PLAs

174

relatively low compared to the one that can be obtained when clay platelets are well

exfoliated.

Figure 102: TEM pictures of big (a) and small (b) agglomerates of Cloisite 30B in FR-PLA-C30B.

FR-PLA-C30B contains different sizes of agglomerates of Cloisite 30B but no isolated

platelets. Thus, the organoclay is not well dispersed and FR-PLA-C30B is not a nanocomposite

but a microcomposite. Nevertheless, this state of dispersion seems to be enough to display an

increase in viscosity of the FR-PLA-C30B material compared to FR-PLA.

The dynamic viscosity of FR-PLA and FR-PLA-C30B appears to be dependent on the ageing

exposure time (Figure 103 (a)) and on the molecular mass of the material (Figure 103 (b)).

Indeed, the viscosity decreases when ageing exposure increases (M

decreases) whatever

ageing conditions:

 Under T/UV, the dynamic viscosity decreases from 1240 Pa/s

= 29000 g/mol) and 2010 Pa/s (M

= 27000 g/mol) to 460 Pa/s

= 24100 g/mol) and 430 Pa/s (M

= 21800 g/mol) after 30 days for FR-PLA

and FR-PLA-C30B, respectively.

 Under T/RH, the dynamic viscosity reaches 160 Pa/s (M

= 18400 g/mol) and

110 Pa/s (M

= 15400 g/mol) after 21 days for FR-PLA and FR-PLA-C30B,

respectively.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter V: Impact of ageing on the flame retardant properties of FR-PLAs

175

 Under T/UV/RH the most noticeable decrease of viscosity is observed. Indeed,

it decreases to 120 Pa/s (M

= 17500 g/mol) and 100 Pa/s (M

= 15000 g/mol)

after only 14 days for FR-PLA and FR-PLA-C30B, respectively.

Figure 103: Evolution of viscosity of FR-PLA and FR-PLA-C30B as a function of ageing

exposure to T/UV (●), T/RH (▪) and T/UV/RH (Δ) conditions (a) and molecular mass (b).

As for the molecular mass (p.90 and p.127), the decrease of viscosity is more important

when materials are aged under T/UV/RH compared to T/RH and T/UV conditions: after

14 days ageing, the dynamic viscosity is only 100 Pa/s for FR-PLA-C30B aged under T/UV/RH,

compared to 450 Pa/s and 1300 Pa/s for T/RH and T/UV exposure, respectively.

Hence, the decrease in viscosity observed during ageing is responsible for the “runaway

effect” previously described. This one thus leads to an increase in LOI values for both FR-PLAs

during ageing. Knowing the impact of ageing on the LOI performance of our materials, it is

now useful to study it impact on MLC performances. The next section is then dedicated to

these investigations.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter V: Impact of ageing on the flame retardant properties of FR-PLAs

176

Mass loss cone performances after ageing

As demonstrated in the last part (p.160), the use of Melamine, APP and organoclay (C30B)

leads to the development of an efficient intumescent shield protecting PLA against fire. It is

suitable to keep these performances during the lifetime of the material. Thus, MLC properties

of PLAs were examined as a function of ageing in order to evaluate the impact of T/UV, T/RH

and T/UV/RH exposure on the flame retardancy.

MLC measurements were performed until 60 days for PLA under each exposure conditions

(Table 37 and Figure 104). As for LOI test, results depicted for PLA show that whatever the

kind of exposure, ageing does not impact the MLC performances. Indeed the time to ignition

(TTI) as well as both pHRR and THR remain unchanged. Indeed, TTI is around 39 s ± 4s, pHRR

is 420 ± 2% and THR is 72 ± 6%. Whereas the physico-chemical properties (M

, T

and

crystallinity) of unfilled PLA are impacted during ageing, the LOI and MLC results demonstrate

that the fire performances of the material remain unchanged.

Table 37: MLC values obtained for PLA as a function of exposure to T/UV, T/RH and T/UV/RH

conditions.

PLA

(g/mol)

Time to ignition (s)

pHRR (kW/m²)

THR (MJ/m²)

t = 0

68000

420

72.0

T/UV t= 60 days

47000 (-31%)

43 (+4)

424 (+2%)

71.1 (-2%)

T/RH t = 60 days

34200 (-50%)

41 (+2)

422 (+1%)

69.4 (-4%)

T/UV/RH t = 60 days

21500 (-68%)

38 (-1)

416 (-1%)

67.7 (-6%)

Figure 104: RHR curves of unaged PLA (●) and 60 days exposure to T/UV (●), T/RH (●) and

T/UV/RH (●) conditions.

Focusing on the two intumescent PLAs, measurements were performed until 6, 28 and

60 days under T/UV/RH, T/RH and T/UV, respectively. MLC data obtained for FR-PLA are

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter V: Impact of ageing on the flame retardant properties of FR-PLAs

177

reported in Table 38 and Figure 105. Contrary to LOI it seems that MLC performances are not

impacted during ageing. Indeed, TTI is 45 s for unaged material and remains quite similar after

ageing. The pHRR remains at 82 kW/m² after 60 days ageing to T/UV and decreases to

81 kW/m² (-2%) and 79 kW/m² (-4%) after 28 days under T/RH and 6 days under T/UV/RH

conditions, respectively. The only difference is observed for the THR which is decreased by

33%, 34% and 55% to reach 2.9 MJ/m², 2.8MJ/m² and 1.9 MJ/m² after 60 days under T/UV,

28 days under T/RH and 6 days under T/UV/RH, respectively. However, due to the relatively

low values of THR and the limits of the MLC device, the decrease of THR cannot be considered

as significant.

Table 38: MLC values obtained for FR-PLA as a function of exposure to T/UV, T/RH and

T/UV/RH conditions.

FR-PLA

Mn (g/mol)

Time to ignition (s)

pHRR (kW/m²)

THR (MJ/m²)

t = 0

29000

4.8

T/UV t= 60 days

21300 (-26%)

43 (-2)

2.9 (-33%)

T/RH t = 28 days

20000 (-30%)

81 (-2%)

2.8 (-34%)

T/UV/RH t = 6 days

20000 (-30%)

43 (-2)

79 (-4%)

1.9 (-55%)

Figure 105: RHR curves of unaged FR-PLA (●) and 60 days exposure to T/UV (●), T/RH (●) and

T/UV/RH (●) conditions.

The FR-PLA-C30B formulation shows the same behavior as FR-PLA (Table 39 and Figure

106). Indeed, The TTI which is 43 s before ageing reaches 46 s, 45s and 44s after 60 days under

T/UV, 28 days under T/RH and 6 days under T/UV/RH, respectively. Thus, the TTI can be

considered as unchanged after ageing. The pHRR of the material is slightly decreased during

ageing but is in the experimental error of 10%. Indeed, pHRR is 48 kW/m² before ageing and

reaches 47 kW/m² (-2%), 46 kW/m² (-3%) and 43 kW/m² (-8%) after 60 days under T/UV,

28 days under T/RH and 6 days under T/UV/RH, respectively. As well as for FR-PLA, the THR is

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter V: Impact of ageing on the flame retardant properties of FR-PLAs

178

reported to decrease during ageing, from 3.1 MJ/m² before ageing to 1.9 MJ/m² (-36%),

1.8 MJ/m² (- 42%) and 1.1 MJ/m² (-63%) after 60 days under T/UV, 28 days under T/RH and

6 days under T/UV/RH, respectively. However, due to the relatively low values of THR and the

limits of the MLC device, the decrease of THR cannot be considered as significant.

Table 39: MLC values obtained for FR-PLA-C30B as a function of exposure to T/UV, T/RH and

T/UV/RH conditions.

FR-PLA-C30B

(g/mol)

Time to ignition (s)

pHRR (kW/m²)

THR (MJ/m²)

t = 0

27000

3.1

T/UV t= 60 days

21100 (-23%)

46 (+3)

47 (-2%)

1.9 (-36%)

T/RH t = 28 days

20000 (-26%)

45 (+2)

46 (-3%)

1.8 (-42%)

T/UV/RH t = 6 days

20000 (-26%)

44 (+1)

43 (-8%)

1.1 (-63%)

Figure 106: RHR curves of unaged FR-PLA-C30B (●) and 60 days exposure to T/UV (●), T/RH

(●) and T/UV/RH (●) conditions.

Considering the experimental error and the low values obtained in terms of pHRR and THR,

no significant changes in MLC performances were evidenced after ageing for FR-PLAs loaded

at 30wt.-%. Hence, even if ageing has a direct impact on the physico-chemical properties of

FR-PLAs, it appears that it does not affect fire performances of such materials measured by

MLC. It was thus decided to investigate the impact of ageing on the MLC performances of a

less effective system. Hence, PLAs filled at 20 wt.-% FR loading were investigated. In fact, the

MLC performances of a less effective system may be affected during ageing exposure.

Mass loss cone behavior of FR-PLAs loaded at 20 wt.-%.

This study was led on two FR-PLAs filled with 20 wt.-% of FR fillers in the same ratio as for

the formulations at 30 wt.-%. The composition and names of these two formulations are

reported in chapter II (Table 7, p.70). FR-PLA-20% and FR-PLA-C30B-20% were aged under

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter V: Impact of ageing on the flame retardant properties of FR-PLAs

179

T/UV/RH condition since this latter was previously reported to be the more drastic for PLA

materials (p.106 and p.152), thus the effects of ageing should be quickly observed.

As FR-PLA-20% and FR-PLA-C30B-20% were completely disintegrated into powder after

28 days, MLC investigations were led on both materials before ageing, after 17 and 22 days of

exposure to T/UV/RH conditions. Indeed, after 17 days the M

is reached and the high

concentrated FR layer is formed at the surface of the plates. As expected, FR-PLA-20% (Table

40 and Figure 107) exhibits lower fire properties than FR-PLA-30% before ageing, pHRR and

THR are 186 kW/m² and 46.3 MJ/m², respectively, compared to 82 kW/m² and 4.8 MJ/m² for

the FR-PLA-30%. Regarding their fire properties after ageing, it appears that the TTI which is

88 s ± 2 s all along the exposure is quite stable for the FR-PLA-20%. The THR is surprisingly

decreased by 20% and 30% after 17 days and 22 days exposure. Same phenomenon is

observed for the pHRR which is 186 kW/m² before ageing and reaches 124 kW/m² (-33%) and

94 kW/m² (-48%) after 17 days and 22 days of ageing, respectively. These results obtained

show that MLC fire performances of the FR-PLA-20% material are clearly improved after

ageing.

Table 40: MLC values obtained for FR-PLA-20% as a function of exposure time to T/UV/RH

conditions.

FR-PLA-20%

Time to ignition (s)

pHRR (kW/m²)

THR (MJ/m²)

t = 0

186

46.3

T/UV/RH t= 17 days

90 (+2)

124 (-33%)

37.5 (-20%)

T/UV/RH t = 22 days

86 (-2)

94 (-48%)

32.5 (-30%)

Figure 107: RHR curves of unaged FR-PLA-20% (●), 17 days (●) and 22 days (●) aged under

T/UV/RH conditions.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter V: Impact of ageing on the flame retardant properties of FR-PLAs

180

As well as in the case of FR-PLA, FR-PLA-C30B-30% presents higher fire performances

compared to FR-PLA-C30B-20% (Table 41 and Figure 108). Whereas pHRR and THR are

48 kW/m² and 3.1 MJ/m² for FR-PLA-C30B-30%, they respectively reach 88 kW/m² and

11.6 MJ/m² for FR-PLA-C30B-20%. Focusing on the performances of FR-PLA-C30B-20% after

ageing, the TTI remains unchanged until 22 days exposure. Indeed, TTI is 92 s ± 3 s.

Furthermore, the pHRR decreases after ageing from 88 kW/m² to 65 kW/m² (-26%) and

51 kW/m² (-42%) after 17 and 22 days exposure to T/UV/RH, respectively. The same trend is

observed for THR which is reduced from 11.6 MJ/m² to 8.9 MJ/m² (-23%) and 7.6 (-35%) after

17 and 22 days exposure, respectively. Like FR-PLA-20%, the behavior of FR-PLA-C30B-20%

demonstrates that MLC fire performances are improved after ageing.

Table 41: MLC values obtained for FR-PLA-C30B-20% as a function of exposure time to

T/UV/RH conditions.

FR-PLA-C30B-20%

Time to ignition (s)

pHRR (kW/m²)

THR (MJ/m²)

t = 0

11.6

T/UV/RH t= 17 days

93 (+1)

65 (-26%)

8.9 (-23%)

T/UV/RH t = 22 days

95 (+3)

51 (-42%)

7.6 (-35%)

Figure 108: RHR curves of unaged FR-PLA-C30B-20% (●), 18 days (●) and 22 days (●) aged

under T/UV/RH conditions.

The study led on FR-PLAs loaded at 20 wt.-% show that MLC performances are enhanced

during ageing. A reduction of pHRR and THR along with unchanged TTI is observed. Moreover,

a change in intumescent behavior is reported when samples are aged. Indeed, during the

experiments, the thermal shield was observed to be formed sooner, with a higher swelling,

resulting in a more expanded char. The cross section of the chars obtained after MLC

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter V: Impact of ageing on the flame retardant properties of FR-PLAs

181

measurements were thus analyzed by digital microscopy. The expansion of the char as a

function of ageing exposure to T/UV/RH is reported in Figure 109 and Figure 110 for FR-PLAs-

20%. The thickness of the char was measured on three different chars of FR-PLA-20% and FR-

PLA-C30B-20%.

Figure 109: Evolution of the char expansion of FR-PLA-20% as a function of ageing under

T/UV/RH exposure.

Figure 110: Evolution of the char expansion of FR-PLA-C30B-20% as a function of ageing

under T/UV/RH exposure.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter V: Impact of ageing on the flame retardant properties of FR-PLAs

182

The thickness of FR-PLA-20% char (Figure 109) is 0.5 cm before ageing and reaches 1.3 cm

after 17 days exposure to T/UV/RH. Moreover, the thickness of the FR-PLA-C30B-20% char

(Figure 110) is 0.8 cm before ageing and grows up to 1.5 cm after 17 days exposure to

T/UV/RH. These observations confirm the higher expansion of the char after ageing, for both

FR-PLA and FR-PLA-C30B filled at 20 wt.-%.

Knowing the positive impact of ageing on the thermal protection of intumescent PLA, it is

of interest to understand how ageing can improve the MLC performances. For this purpose,

investigations were led on the molecular mass of FR-PLAs loaded at 20 wt.-%.

Molecular mass of FR-PLAs loaded at 20 wt.-%

The change in molecular mass of FR-PLAs-20% was studied as a function of ageing exposure

to T/UV/RH and compared to FR-PLA and FR-PLA-C30B filled at 30 wt.-% (Table 42 and Figure

111). The first observation is that before ageing, the M

of FR-PLAs-20% is higher than that of

FR-PLAs-30%. Indeed, M

of FR-PLA-20% and FR-PLA-C30B-20% are 32500 g/mol and

31000 g/mol, respectively, compared to 29000 g/mol for FR-PLA-30% and 27000 g/mol for FR-

PLA-C30B-30%. Hence the incorporation of 20 wt.-% of fillers is less impacting for the

molecular mass compared to the incorporation of 30 wt.-%.

Table 42: M

data of FR-PLAs-20% and FR-PLAs-30% as a function of ageing exposure to

T/UV/RH conditions.

Exposure

time (days)

FR-PLA-20%

FR-PLA-30%

FR-PLA-C30B-20%

FR-PLA-C30B-30%

32500

29000

31000

27000

27900 (-14%)

23500 (-17%)

26000 (-16%)

21000 (-20%)

21000 (-35%)

17500 (-38%)

20200 (-35%)

15000 (-40%)

16500 (-49%)

13000 (-53%)

15300 (-51%)

9750 (-60%)

11000 (-67%)

8800 (-69%)

9250 (-70%)

5800 (-75%)

Furthermore, as for FR-PLAs-30%, the exposure to T/UV/RH conditions leads to the

decrease of the molecular mass of both intumescent materials. Indeed, M

of FR-PLA-20%

decreases from 32500 to 11000 g/mol (-67%) whereas for FR-PLA-C30B-20% it decreases from

31000 to 9250 g/mol (-70%) after 45 days exposure to T/UV/RH conditions. Considering the

experimental error of 10%, one can notice that the M

of FR-PLAs-20% and FR-PLAs-30%

decreases at the same rate. M

decreases by 67% and 70% for FR-PLA-20% and FR-PLA-C30B-

20%, respectively compared to 68% and 75% for FR-PLA-30% and FR-PLA-C30B-30%,

respectively, after 45 days exposure to T/UV/RH. However, due to the higher M

of FR-PLAs-

20% before ageing, compared to that of FR-PLAs-30%, more time is required to reach the

critical molecular mass (M

= 20000 g/mol) from which the material starts to disaggregate.

This explains why the materials need 17 days to reach the critical M

compared to 6 days for

PLAs filled at 30 wt.-% loading.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter V: Impact of ageing on the flame retardant properties of FR-PLAs

183

As previously reported in chapter IV (p.120 and p.127), when the critical molecular mass is

reached, a high concentrated FR layer appears at the surface of the materials. In the case of

intumescent FR-PLAs-20%, this layer is formed after 17 days when M

reaches the M

20000 g/mol (the complete disintegration of the plates occurs after 28 days approximatively).

This phenomenon is in accordance with the improvement of MLC performances reported in

this section.

Figure 111: M

versus ageing time of FR-PLA 20% and 30% (●) and FR-PLA-C30B (●) 20% and

30% materials aged under T/UV/RH conditions.

All along the different investigations led during our work, some phenomena were

evidenced both in terms of physico-chemical properties or fire performances after ageing:

 The decrease of the molecular mass of FR-PLAs when exposed to artificial

accelerated ageing (T/UV, T/RH and T/UV/RH).

 The decrease of thermal stability of FR-PLAs as a function of ageing exposure.

 When the critical molecular mass (M

) is reached, a high concentrated FR layer is

formed at the surface of the materials.

 In terms of fire performances, a reduction of pHRR and THR is observed along ageing

exposure.

 A change of the intumescent behavior is reported when samples are aged. Indeed,

it was observed that the intumescent shield is formed sooner accompanied by a

higher expansion of the char.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter V: Impact of ageing on the flame retardant properties of FR-PLAs

184

According to these different observations a simple mechanism can be proposed to explain

the impact of ageing on the MLC performances (Figure 112):

 The decrease of the molecular mass occurring during ageing of FR-PLAs impacts the

dynamics of molecular chains. In fact, ageing can increase the mobility of the chains

and have an effect on the strength of polymer bonds. The degradation of the

polymer matrix could lead to the formation of new or more reactive species that

can initiate quicker the thermal degradation of the polymer [174, 190-192] (Figure

112 (a)).

 The decrease of both pHRR and THR is observed from the moment when the critical

molecular mass (M

) is reached. Thus, when the high concentrated FR layer is

formed at the surface of the material, i.e. when FR fillers are more readily available

on the surface (Figure 112 (a)).

 It is thus possible that the more reactive species formed during PLA degradation

and fillers layer formed at the surface of FR-PLAs both promote the intumescent

mechanism. Thus, when exposed to the radiative heat flux of the MLC, PLA and

fillers can quickly react, thus promoting the intumescent mechanism. The insulative

barrier is formed sooner and both pHRR and THR are decreased (Figure 112 (b)).

 On the other hand if more reactive species are formed during ageing, they are able

to react with fillers in the condensed phase. Thus explaining the higher expansion

of the char reported during MLC test and demonstrating that the polymer itself

participate to its own protection.

 Hence, it seems that ageing promotes the intumescence mechanism through the

kinetics of formation of the protective shield.

 However, further investigations are still needed as to prove our different

assumptions and in particular determine if the mechanism of intumescence itself is

modified during ageing. Hence it will be possible to evidence a complete mechanism

of action in terms of MLC fire performances of aged FR-PLAs.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter V: Impact of ageing on the flame retardant properties of FR-PLAs

185

Figure 112: Schematic representation of the effect of ageing on the intumescent mechanism

of FR-PLA-C30B-20%.

In the case of FR-PLA and FR-PLA-C30B filled at 30 wt.-% loading, the MLC performances

remain similar after ageing maybe because the two intumescent formulations already exhibit

very high fire performances before the formation of the FR concentrated layer at the surface.

On the other hand, the values of pHRR and THR obtained during the experiments are very low,

thus considering the limits of measurements of the MLC device no effect of ageing can be

considered. One possibility to avoid this problem is to increase the heat flux of the MLC device.

Indeed, whereas our experiments were performed at 35 kW/m², by increasing the heat flux

until 50 or 75 kW/m², it should be possible to obtain higher values of THR and pHRR and thus

evidence the effect of ageing on the MLC performances of FR-PLAs-30%.

3. Conclusion

In Chapter V, the fire properties of intumescent PLAs as well as their behavior during ageing

were studied. Firstly, it was confirmed that the incorporation of fillers such as Melamine and

APP allows the development of an efficient intumescent system improving the fire

performances of PLA [12, 24, 61]. Moreover, the addition of nanoparticles (i.e. Cloisite 30B)

shows a synergistic effect enhancing the protection by intumescence, with a more insulating

char that limits the “fuel” release due to its better cohesion, thus decreasing pHRR and THR.

In the second part, flame retardant properties of PLA and FR-PLAs after ageing were

investigated under T/UV, T/RH and T/UV/RH conditions. During ageing it appears that flame

retardant properties of virgin PLA are not modified until 60 days of ageing and thus whatever

the kind of exposure. However, concerning FR-PLA and FR-PLA-C30B, investigations led for LOI

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Chapter V: Impact of ageing on the flame retardant properties of FR-PLAs

186

and MLC tests reveal that ageing impacts the flame retardancy of intumescent PLAs. It was

demonstrated that the LOI of FR-PLAs increases during ageing. The increase of LOI was

reported to be higher in the case of T/UV/RH exposure compared to T/RH then T/UV

conditions. Action mechanism was found to be due to a “runaway effect”. Indeed, during

ageing, the degradation of the material leads to a decrease in viscosity of the system. This

phenomenon creates a “runaway effect”, which permits to the combustible material to escape

from the flame by dripping, thus, leading to the increase in the LOI values for both FR-PLAs.

Furthermore, ageing was reported to have no effect on the fire performances observed by

MLC for PLA and FR-PLAs. Even before ageing the two intumescent PLAs exhibit very high fire

performances (very low pHRR and THR), due to the optimization of the formulation. Thus the

effect of ageing on the MLC performances cannot be observed for these formulations at

35 kW/m².

However, by studying a less effective intumescent PLAs (i.e. FR-PLA-20% and FR-PLA-C30B-

20%) it was found that ageing improves the MLC performances at 35 kW/m², by decreasing

both pHRR and THR, whereas TTI remains stable. The degradation of the materials during

ageing is responsible for these surprising performance. Indeed, the enhanced degradation of

FR-PLAs (the high decrease of molecular mass) can leads to the formation of new or more

reactive products and of a concentrated layer of nitrogen and phosphorus (also Cloisite 30B

in the case of FR-PLA-C30B formulations) at the surface of the sample (p.120, p.142). Both

phenomena promote the kinetics of intumescence through the development of a more

efficient shield when the material is exposed to a heat source, favoring the intumescent

phenomenon and enhancing the flame retardancy of the materials.

Investigations reported during this work demonstrate that physico-chemical properties

, viscosity …) and FR fillers both play an important part in the improvement of fire

properties during ageing. It is obvious that the molecular mass plays a crucial role in the

durability of intumescent PLAs. Indeed it is mainly responsible for the behavior of the other

physico-chemical properties (viscosity, T

, T

, thermal stability, crystallinity …) and thus

indirectly to the formation of a high concentrated FR layer at the surface of the materials

during ageing. Hence, the degradation level of the materials governs the molecular mass and

thus the behavior of the physico-chemical properties and fire performances during FR-PLAs

lifetime.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

General conclusion

187

General conclusion

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

General conclusion

188

PLA is nowadays the first commodity biodegradable polymer produced from annually

renewable resources. It is a potential good candidate for replacement of traditional petroleum

derived polymeric materials. Indeed PLA is relatively cheap and has some remarkable

properties that make it suitable for various applications such as packaging, textile or

transportation like automobile. However, to be used in certain sectors (e.g. automobile), PLA

has to be flame retarded and its flame retardant properties as well as its properties of use

must be kept for the product lifetime. Literature review revealed that the interest for flame

retarded PLA kept growing these past few years. Indeed, many conventional flame retardants

which can be combined with nanoparticles (metal hydroxides, phosphorus-containing

compounds, halogenated and nitrogentaed compounds, organoclays …) are generally

incorporated with efficiency into PLA matrix, to design flame retardant formulations.

However, regardless of its interesting properties, PLA is known to be extremely sensitive to its

environment of use, impacting the lifetime of the material. Ageing generally leads to the

degradation of the material via various phenomena (e.g. hydrolysis, thermo-degradation,

photo-degradation …). Ageing which is detrimental for the integrity of the material over time

can also affect the durability of the flame retardancy. The aim of this PhD was to evaluate the

effect of ageing on the lifetime of flame retarded PLAs.

In this context, the first purpose of this work was to evidence the impact of ageing on neat

PLA, a commercially available grade used in this study. Indeed, the understanding of the

behavior of neat PLA during ageing was of primary importance to apprehend the ageing of

flame retarded PLAs. The ageing of the material was carried out under three different

accelerated ageing tests under T/UV, T/RH and T/UV/RH exposure. The physico-chemical

properties (M

, T

, thermal stability, crystallinity) and visual appearance of PLA during

ageing were assessed by a various number of techniques, such as GPC, DSC, TGA, XRD, FTIR,

digital microscopy, SEM and L*a*b*. This step of characterization permitted to correlate the

effects of ageing with the ones already reported in literature. Indeed, it was found that ageing

has a direct influence on the physico-chemical properties of the material by decreasing M

, T

and thermal stability of PLA. On the contrary, the crystallinity of the material increased.

Visually, cracks, holes and a bleaching of the material were reported during ageing. It was

reported that these phenomena depend on the ageing conditions, as the impacts of ageing

are even more important for PLA aged under T/UV/RH compared to T/RH and T/UV. The

influence of flame retardant additives on the ageing of PLA was then studied. Both visual and

physico-chemical characterizations performed during exposure revealed that flame retardant

additives have a direct impact on the ageing of PLA. Indeed, the effects of ageing appeared to

be more pronounced when FR fillers are incorporated into the PLA matrix. Even if cracks and

holes are also observed during ageing of FR-PLAs, the formation of a white layer was

evidenced at the surface of the T/RH and T/UV/RH aged materials when a critical molecular

mass of 20000 g/mol is reached. Moreover, it appeared that both FR-PLA and FR-PLA-C30B

were surprisingly disintegrated into powder after 10 and 35 days of exposure to T/UV/RH and

T/RH conditions, respectively. On the contrary, after 60 days, the exposure to T/UV appeared

not to be drastic for the visual integrity of FR-PLAs as it was demonstrated that the M

was still

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

General conclusion

189

not reached. The different characterization techniques used all along the ageing exposure

(GPC, DSC, XRD, FTIR …) suggested that these phenomena can be explained by the evolution

of physico-chemical properties of the materials. The impacts of ageing on the physico-

chemical properties of FR-PLAs were found to be similar to those reported for neat PLA

(decrease of M

, T

, increase in crystallinity …). However, the already low molecular mass

before ageing explained why FR-PLAs reached quickly the M

(20000 g/mol) and thus the faster

degradation of FR materials. This proved that the molecular mass is a crucial parameter for

polymers as it governs most of the physico-chemical properties of the material (T

, T

crystallinity …) and thus its durability.

Strategically, it was also fundamental to elucidate the mechanisms of degradation

occurring during T/UV, T/RH and T/UV/RH exposure. Especially as many different mechanisms

have been proposed all over the years depending on the ageing conditions and duration. It

was evidenced that the degradation of neat PLA takes place in the bulk of the material

assuming an autocatalytic process, but mechanism of degradation depends on the exposure

conditions. FTIR evidenced that hydrolysis is the predominant phenomenon occurring during

T/RH exposure, whereas under T/UV, EPR experiments revealed that PLA is degraded via a

radical mechanism. On the other hand, the increase in temperature was reported to be

responsible for a faster degradation. The combination between UV-rays and RH creates a

synergistic effect under T/UV/RH exposure. Both hydrolysis and radical mechanism occur

simultaneously, accelerating the degradation of the polymer matrix. However, only few

studies reported in literature deal with the exposure to T/UV/RH conditions and it is quite

difficult up to now to determine which of the two phenomena is predominant. In presence of

FR fillers, EPR experiments indicated that FR fillers protect the material against UV and O

penetration. On the contrary, hydrolysis mechanism appeared to be modified when FR fillers

are incorporated into the PLA matrix. Indeed, in presence of FR fillers, hydrolysis seems to

occur at the interface between matrix and fillers, certainly due to an accumulation of water.

As a consequence the formation of the white layer composed of Melamine, APP and C30B (in

the case of FR-PLA-C30B) at the surface of FR-PLAs. Moreover, the partial hydrolysis of APP

could also be held responsible for this phenomenon, explaining why APP particles are not

completely embedded in the PLA matrix anymore. During this work, the use of the EPR device

was revealed to be an innovative technique to use, as it offers new perspectives on both

mechanistic and kinetics points in the field of photo-degradation of polymers.

The main goal of incorporating FR fillers into PLA was to improve its flame retarded

properties. Knowing the impact of fillers on the physico-chemical properties and mechanisms

of degradation of PLA, it was thus crucial to investigate the durability of the flame retardant

properties. It was firstly demonstrated that the incorporation of FR fillers (i.e. Melamine and

APP) permitted the development of an efficient intumescent system improving the fire

performances of PLA. Furthermore, the addition of nanoparticles (i.e. Cloisite 30B) showed a

synergistic effect. The intumescence of the system was enhanced, with a more insulating and

a more cohesive char. After ageing, it was showed that the fire performances of neat PLA

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

General conclusion

190

remained similar whereas those of FR-PLAs were improved, especially in LOI test. Indeed, it

was demonstrated that the LOI of FR-PLAs increases during ageing due to a “runaway effect”.

Due to the already very high fire performances before ageing, ageing was reported to have no

effect on the fire performances observed by MLC for FR-PLAs filled at 30 wt.-%. However, it

was demonstrated that ageing improves the MLC performances of a less effective intumescent

PLA (i.e. FR-PLA-20% and FR-PLA-C30B-20%) by decreasing both pHRR and THR. The

degradation of the matrix during ageing and the high concentrated FR layer formed at the

surface of the materials were held responsible for this surprising behavior. They both

accelerate the kinetics of intumescence, thus promoting the development of a more efficient

shield able to enhance the flame retardancy of the materials. This demonstrates the key role

of ageing in the behavior and the durability of flame retardancy during lifetime of FR-PLAs.

This PhD aimed to evidence the impact of ageing on the durability of a flame retardant

system, based on PLA. It was shown that PLA is a very promising polymer in the field of flame

retardancy, even if it is extremely sensitive to its environment of use. It was demonstrated

that both FR fillers and physico-chemical properties (especially the molecular mass) play an

important role in the improvement of fire properties during ageing. This could be used as a

basis for further work, dealing with the ageing of flame retarded polymers

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Outlook

191

Outlook

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Outlook

192

As many topics were approached in this manuscript, many outlook emerged along the

dissertation: optimization of extrusion process, mechanistic outlook, kinetics degradation,

modelling…

The first and obvious outlook is related to the molecular mass of FR-PLAs. In this

manuscript, it was evidenced that the molecular mass is a crucial parameter for the durability

of the material. It could be interesting to limit the decrease of M

during the extrusion process

of FR-PLAs (chain extender, master batch…), as to study PLA and FR-PLAs with similar

molecular mass. Hence, avoiding the effect of the molecular mass on degradation of FR-PLAs,

it would be possible to focus solely on the contribution of FR fillers on the ageing of the

intumescent materials.

Most papers dealing with ageing of polymers, especially PLA, demonstrate the effect of

ageing on the mechanical properties of the material. The mechanical properties of a polymer

involve its behavior under stress. These properties are essential when considering how a

polymer can be used. Hence, the impact of the addition of FR fillers on the mechanical

properties of PLA (using tensile tests, dynamic mechanical analysis…) should be studied as well

as their behavior under exposure to T/UV, T/RH and T/UV/RH exposure.

In terms of mechanistic approach, it was reported in this manuscript that PLA is mainly

degraded into carboxylic acids, such as formic acid, acetic acid, oxalic acid… During this work,

the use of pH paper and the identification of their characteristic odor were the only way to

evidence these products. However, the pyrolysis – gas chromatography - mass spectroscopy

(Py-GC/MS) is a chemical analysis that could evidence the secondary products formed during

the degradation of PLA. Indeed, Py-GC/MS is a method of thermal decomposition in which

smaller molecules produced can be separated by gas chromatography and detected using

mass spectrometry. Hence, it is a useful techniques for the identification of chemical species.

As already reported, electron paramagnetic resonance (EPR) is another innovative technique

that should be further used in order to study the kinetics and mechanism of degradation of

polymeric materials. Most papers focusing on the photo-degradation of polymers, especially

PLA, suggest mechanism of degradation based on FTIR investigations. However, this technique

should be coupled with EPR in order to elucidate the intermediate mechanisms (radicals)

occurring during photo-degradation and not only evidence the final products. Moreover, EPR

is an appropriate technique as to evidence the role of FR fillers during photo-degradation of

FR-PLAs. Indeed, this could be interesting to further investigate the protective role of FR fillers

during photo-degradation of FR-PLAs. Thus determining how FR fillers prevent the

degradation of PLA and evidencing the mechanism of protection.

In the field of fire testing, it has been reported that the MLC performances of FR-PLAs filled

at 30 wt.-% loading remain similar after ageing, maybe because the two intumescent

formulations already exhibit very high fire performances before ageing. Hence, the values of

pHRR and THR obtained during the experiments are very low, thus considering the limits of

measurements of the MLC device no effect of ageing can be considered. The question raised

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Outlook

193

from these results is “Could that be possible to avoid the limitation of the MLC device?”. For

instance, one possibility could be to increase the heat flux of the MLC device. Indeed, whereas

our experiments were performed at 35 kW/m², by increasing the heat flux until 50 or

75 kW/m², it should be possible to obtain higher values of THR and pHRR and thus evidence

the effect of ageing on the MLC performances of FR-PLAs-30%. This phenomenon have been

demonstrated in the case of neat PLA (Figure 113 and Table 43).

Figure 113: RHR curves as a function of time of PLA at 35 kW/m²(●), 50 kW/m² (●) and

75 kW/m² (●).

Table 43: MLC values obtained for PLA at 35 kW/m², 50 kW/m² and 75 kW/m².

Heat flux

(kW/m²)

Time to ignition (s)

pHRR (kW/m²)

THR (MJ/m²)

420.0

72.0

37 (-2)

513.0 (+22%)

80 (+11%)

40 (+1)

557.0 (+32%)

82 (+14%)

Indeed, at 35 kW/m², pHRR and THR of PLA are 420 kW/m² and 72 MJ/m², respectively. By

increasing the heat flux until 50 kW/m², pHRR and THR reach 513 kW/m² (+22%) and 80 MJ/m²

(+12%), respectively. For an external heat flux of 75 kW/m², pHRR and THR of PLA are

increased until 557 kW/m² (+32%) and 82 MJ/m² (+14%). Finally, it was shown that during MLC

experiment, the intumescent shield of aged FR-PLAs is formed sooner, with a higher swelling,

resulting in a more expanded char. In this manuscript, no investigations were led on the

structure of the chars during ageing. Knowing the role of the char structure on the

intumescent mechanism of FR-PLAs [12, 24], more studies should be led on this topic. Using

techniques such as 3D X-Ray tomography or rheometer, it could be interesting to study the

structure, the porosity and the resistance of the char before and after ageing, in order to

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Outlook

194

understand the role of the char in the improvement of fire performances during ageing. In

general, more investigations are needed to evidence if ageing impacts the intumescent

mechanism itself or if it only plays on the kinetics of intumescence.

Some issues still exist to design long term commercially applications for flame retarded PLA

systems, as addressed in this Ph.D work, but the current efforts will probably help to resolve

them. In particular all the methods developed and the key parameters detailed in this Ph.D

work are really required to go further in the research on the ageing of flame retarded systems.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

List of figures, tables and equations

195

List of figures, tables and equations

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

List of figures, tables and equations

196

List of Figures

Figure 1: World plastic production from 1950 to 2012 [1, 2]. ................................................. 18

Figure 2: Global production capacities of bioplastics in 2014 (by material type) [4]. ............. 19

Figure 3: Evolution of number of publications versus time with keyword “flame retardancy”

and “PLA” (Scifinder Sept 2016). .............................................................................................. 20

Figure 4: Synthesis of PLA from L- and D-lactic acids, adapted from Lim et al. [52]. .............. 25

Figure 5: Poly(lactic acid) structure. ......................................................................................... 26

Figure 6: The stereoisomers of lactic acid. ............................................................................... 26

Figure 7: Chemical structures of L,L-, L,D- and D,D-lactides [5]. .............................................. 27

Figure 8: Stereocomplex of PLLA and PDLA [61]. ..................................................................... 27

Figure 9: PDLLA multiblocks as function of D and L content, adapted from Kakuta et al. [60].

.................................................................................................................................................. 27

Figure 10: Transmission percentage versus wavelength for PLA, PS, LDPE, PET and cellophane

films, adapted from Auras et al. [5]. ........................................................................................ 28

Figure 11: FTIR (ATR) spectrum of PLA in the range 500 to 4000 cm

-1

. ................................... 29

Figure 12: Glass transition temperatures for PLAs of different L-contents as a function of

molecular weight, adapted from Dorgan et al. [67]. ............................................................... 30

Figure 13: Peak melting temperature of PLA as a function of % meso-lactide [52]. (○)

represents values reported by Witzke [71]; (●) represents values reported by Hartmann [72].

.................................................................................................................................................. 31

Figure 14: DSC thermograms of water quenched, air-annealed (cooled from 220°C to

ambient temperature in 5 min), and full-annealed (cooled from 220°C to ambient

temperature in 105 min) PLLA samples. DSC scans were performed at a heating rate of

10°C/min, adapted from Sarasua et al. [77]. ........................................................................... 33

Figure 15: DSC scans for 1.5% D-lactide PLA samples cooled from the melt at 10K/min and

then reheated at different heating rates from 30 to 0.3 K/min, adapted from Pyda et al. [78].

.................................................................................................................................................. 34

Figure 16: DSC scans at 20 K/min for PLA with 1.5% (PLA-L), 8.1% (PLA-M) and 16.4% (PLA-H)

D-isomers. All samples were cooled quickly from the melt and isothermally crystallized at

145°C for 15h. The quenched PLA-L sample was cooled similarly from the melt but did not

undergo the 15 h isothermal crystallization, adapted from Pyda et al. [78]. .......................... 34

Figure 17: Combustion of a polymer [90]. ............................................................................... 37

Figure 18: Main chemical ageing factors and consequences on polymeric materials. ........... 44

Figure 19: General degradation mechanisms, adapted from Beyler, at al. [124]. .................. 45

Figure 20: Norrish I and II mechanisms. ................................................................................... 48

Figure 21: simplified photo-oxidation cycle of polymers, adapted from Yousif et al [129]. ... 49

Figure 22: Example of hydrolysis of polyether. ........................................................................ 53

Figure 23: Illustration of the hydrolytic mechanisms: (a) surface erosion, (b) bulk erosion

without autocatalysis, and (c) bulk erosion with autocatalysis, adapted from Viera et al.

[150]. ........................................................................................................................................ 54

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

List of figures, tables and equations

197

Figure 24: Norrish II mechanism for PLA photo-degradation: (a) PLA chain under UV

irradiation, (b) photophysical excitation and (c) oxidation/scission reactions in PLA chains,

adapted from Belbachir et al. [33]. .......................................................................................... 56

Figure 25: Photo-degradation (radical oxidation) process of irradiated PLA: hydroperoxides

chain propagation and formation of anhydrides by photolysis of hydroperoxides, adapted

from Gardette et al. [35]. ......................................................................................................... 57

Figure 26: Hydrolysis of PLA. .................................................................................................... 58

Figure 27: Biodegradation of PLA in 60°C, adapted from Auras et al. [5]................................ 58

Figure 28: Thermal degradation mechanism of PLA, adapted from McNeill et al. [166]. ....... 59

Figure 29: Thermo Scientific extruder. ..................................................................................... 70

Figure 30: Humidity chamber test HCP 108. ............................................................................ 72

Figure 31: Q-LAB, QUV/spray weathering chamber. ............................................................... 72

Figure 32: Temperature/time ramp used for the DSC analysis. .............................................. 75

Figure 33: color space of CIE Lab system. ................................................................................ 76

Figure 34: Melt flow Index tester. ............................................................................................ 78

Figure 35: Reaction of anhydride with ammonia. .................................................................... 80

Figure 36: Schematic representation of mass loss calorimeter. .............................................. 83

Figure 37: Experimental set-up of the LOI test. ....................................................................... 84

Figure 38: Pictures of PLA at 0 d (a), 60 d T/UV (b), 60 d T/RH (c) and 60 d T/UV/RH (d)

exposure. .................................................................................................................................. 86

Figure 39: Digital microscopy pictures of unaged PLA (a), 60 d T/UV (b), 60 d T/RH (c) and

60 d T/UV/RH (d). ..................................................................................................................... 87

Figure 40: SEM pictures obtained for unaged PLA (a), 60 d aged under T/UV (b), T/RH (c) and

T/UV/RH (d) conditions. ........................................................................................................... 88

Figure 41: M

(a) and polydispersity index PI (b) versus ageing time of PLA samples aged

under T/UV (●), T/RH (▪) and T/UV/RH (Δ) conditions. ........................................................... 91

Figure 42: Average number of random chain scissions per unit mass (n

) for PLA as function

of ageing time for different ageing conditions; T/UV (●), T/RH (▪) and T/UV/RH (Δ). ............ 92

Figure 43: TGA curves as a function of time for unaged PLA (-), T/UV (-), T/RH (-) and

T/UV/RH (-) 125 days aged (10°C/min under air). ................................................................... 93

Figure 44: Evolution of glass transition temperature (T

) of PLA as a function of exposure to

T/UV (●), T/RH (▪) and T/UV/RH (Δ) conditions. ...................................................................... 95

Figure 45: Evolution of melting temperature (T

) of PLA as a function of exposure to T/UV

(●), T/RH (▪) and T/UV/RH (∆) conditions. ............................................................................... 95

Figure 46: T

as a function of M

for PLA aged under T/UV (●), T/RH (▪) and T/UV/RH (∆)

exposure. .................................................................................................................................. 96

Figure 47: Evolution of crystallinity of PLA as a function of ageing time, for T/UV (●), T/RH (▪)

and T/UV/RH (∆) exposure. ...................................................................................................... 99

Figure 48: FTIR spectra as a function of ageing time for PLA aged under T/UV (a), T/RH (b)

and T/UV/RH (c). .................................................................................................................... 100

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

List of figures, tables and equations

198

Figure 49: WAXD patterns of PLA obtained by XRD, before (●) and after 60 of ageing under

T/UV (●), T/RH (●) and T/UV/RH (●). ..................................................................................... 102

Figure 50: FTIR spectra of PLA before and after various ageing durations under T/RH

conditions. .............................................................................................................................. 105

Figure 51: degradation of PLA by hydrolysis under T/RH exposure. ..................................... 105

Figure 52: EPR spectrum of PLA irradiated with wavelengths between 300 and 400 nm at

room temperature. ................................................................................................................ 107

Figure 53: Biradical specie observed by EPR for PLA. ............................................................ 107

Figure 54: Kinetic evolution of EPR signal for PLA as a function of temperature, 25°C (●),

50°C (●) and 75°C (●). ............................................................................................................ 108

Figure 55: Identification of the activation energy of biradical formation in PLA for different

temperatures. ......................................................................................................................... 109

Figure 56: 2D YZ EPR imaging of PLA 50°C under irradiation at 30 min (a) and 4 hours (b).

Intensity of color represents the local spin concentration from blue (min) to red (max). .... 109

Figure 57: Kinetic evolution of EPR signal for PLA as a function of ageing exposure, T/UV (●)

or T/UV/RH (●) at 50°C. .......................................................................................................... 110

Figure 58: FTIR spectra (x32 cm

-1

) of PLA before and after ageing under T/UV and T/UV/RH as

a function of ageing time. ...................................................................................................... 111

Figure 59: FTIR spectra (x2 cm

-1

) of PLA before and after ageing under T/UV and T/UV/RH as

function of ageing time. ......................................................................................................... 113

Figure 60: FTIR spectra (x2 cm

-1

) of 60 days aged PLA films to T/UV and T/UV/RH, before and

after treatment with NH

. ...................................................................................................... 114

Figure 61: Proposed photo-oxidative mechanism of degradation for PLA, in presence of

water. ..................................................................................................................................... 115

Figure 62: numerical pictures of PLA (a), FR-PLA (b) and FR-PLA-C30B (c) plates before

ageing. .................................................................................................................................... 120

Figure 63: Visual evaluation as a function of ageing for FR-PLA 0 d (a), 60 d T/UV (b), 28 d

T/RH (c) and 6 d T/UV/RH (d). ................................................................................................ 121

Figure 64: Visual evaluation as a function of ageing for FR-PLA-C30B 0 d (a), 60 d T/UV (b),

28 d T/RH (c) and 6 d T/UV/RH (d). ........................................................................................ 121

Figure 65: Visual evaluation of FR-PLA (a) and FR-PLA-C30B (b) until disintegration into

powder (c). ............................................................................................................................. 122

Figure 66: SEM pictures obtained for unaged FR-PLA (a), 60 d T/UV (b), 28 d T/RH (c) and 6 d

T/UV/RH (d). ........................................................................................................................... 123

Figure 67: SEM pictures obtained for unaged FR-PLA-C30B (a), 60 d T/UV (b), 28 d T/RH (c)

and 6 d T/UV/RH (d). .............................................................................................................. 124

Figure 68: Aminolysis reaction of PLA. ................................................................................... 126

Figure 69: Hofmann elimination reaction on the surfactant of Cloisite 30B. ........................ 126

Figure 70: M

versus ageing time of FR-PLA (a) and FR-PLA-C30B (b) samples aged under

T/UV (●), T/RH (▪) and T/UV/RH (Δ). ...................................................................................... 127

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

List of figures, tables and equations

199

Figure 71: Average number of random chain scissions per unit mass (n

) for FR-PLA and FR-

PLA-C30B as a function of ageing time at different ageing conditions; T/UV (●), T/RH (▪) and

T/UV/RH (Δ). ........................................................................................................................... 129

Figure 72: Evolution of glass transition temperature (T

) of FR-PLA and FR-PLA-C30B as a

function of exposure time under T/UV (●), T/RH (▪) and T/UV/RH (Δ). ................................ 133

Figure 73: Evolution of melting temperature (T

) of FR-PLA and FR-PLA-C30B as function of

exposure time under T/UV (●), T/RH (▪) and T/UV/RH (Δ). ................................................... 134

Figure 74: Correlation between the decrease of T

and M

of FR-PLA and FR-PLA-C30B as a

function of ageing under T/UV (●), T/RH (▪) and T/UV/RH (∆). ............................................. 135

Figure 75: WAXD patterns of PLA (●), FR-PLA (●) and FR-PLA-C30B (●). .............................. 136

Figure 76: Evolution of crystallinity of FR-PLA (a) and FR-PLA-C30B (b) as a function of ageing

time, for T/UV (●), T/RH (▪) and T/UV/RH (∆) exposure. ....................................................... 138

Figure 77: Correlation between the decrease of crystallinity and M

of FR-PLA and FR-PLA-

C30B as a function of ageing under T/UV (●), T/RH (▪) and T/UV/RH (∆). ............................. 139

Figure 78: WAXD patterns of FR-PLA and FR-PLA-C30B obtained by XRD, before (●) and after

60 of ageing under T/UV (●), T/RH (●) and T/UV/RH (●). ..................................................... 140

Figure 79: EPMA pictures for unaged (a), 60 d T/UV (b), 28 d T/RH (c) and 6 d T/UV/RH (d)

aged FR-PLA. ........................................................................................................................... 142

Figure 80: EPMA pictures for unaged (a), 60 d T/UV (b), 28 d T/RH (c) and 6 d T/UV/RH (d)

aged FR-PLA-C30B. ................................................................................................................. 143

Figure 81: FTIR spectra (ATR) of Melamine as function of ageing exposure under T/RH. .... 145

Figure 82: FTIR spectra (ATR) of APP as function of ageing exposure under T/RH. .............. 146

Figure 83: Hydrolysis reaction of APP, according to Chen et al. [175]. ................................. 146

Figure 84:

P Solid-state NMR spectra of FR-PLA (a) and FR-PLA-C30B (b) before and after

105 days and 90 days exposure to T/RH and T/UV/RH conditions, respectively. ................. 147

Figure 85: FTIR spectra (ATR) of Cloisite 30B as function of ageing exposure under T/RH. .. 147

Figure 86: SEM pictures in cross section obtained for FR-PLA-C30B aged 28 days under T/RH

(a) and 6 days under T/UV/RH (b). ......................................................................................... 149

Figure 87: EPMA pictures in cross section for FR-PLA-C30B aged 28 days under T/RH and 6

days under T/UV/RH conditions. ............................................................................................ 150

Figure 88: SEM pictures obtained for FR-PLA-C30B after 28 days exposure to T/RH. .......... 151

Figure 89: Kinetic evolution of EPR signal as a function of time, for PLA (●), FR-PLA (●) and

FR-PLA-C30B (●). .................................................................................................................... 152

Figure 90: 2D HYSCORE recorded at 50°C for PLA. ................................................................ 153

Figure 91: 2D HYSCORE recorded at 50°C for FR-PLA and FR-PLA-C30B. .............................. 153

Figure 92: Kinetic evolution of EPR signal for FR-PLA and FR-PLA-C30B, as a function of ageing

exposure to T/UV (●) or T/UV/RH (●) at 50°C. ...................................................................... 154

Figure 93: Schematic representation of the effect of fillers and ageing exposure on the

degradation of FR-PLA-C30B. ................................................................................................. 156

Figure 94: TGA curves of PLA (●), FR-PLA (●) and FR-PLA-C30B (●) under air flow at 10°C/min.

................................................................................................................................................ 161

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

List of figures, tables and equations

200

Figure 95: RHR curves as a function of time of PLA (●), FR-PLA (●) and FR-PLA-C30B (●). ... 162

Figure 96: Char residues of FR-PLA and FR-PLA-C30B............................................................ 162

Figure 97: Digital microscopy pictures and cross section of char residues of FR-PLA and FR-

PLA-C30B. ............................................................................................................................... 163

Figure 98: Intumescent mechanism, adapted from Gallos et al [61]. ................................... 165

Figure 99: Release of gas through the char of FR-PLA and FR-PLA-C30B, adapted from Gallos

et al [61]. ................................................................................................................................ 166

Figure 100: TGA curves as function of time for FR-PLA (a) and FR-PLA-C30B (b), before ageing

(●), 125 days T/UV (●), 105 days T/RH (●) and 60 days T/UV/RH (●) exposure. .................. 169

Figure 101: Evolution of the behavior of LOI samples of FR-PLA and FR-PLA-C30B as a

function of ageing exposure to T/UV, T/RH and T/UV/RH conditions. .................................. 172

Figure 102: TEM pictures of big (a) and small (b) agglomerates of Cloisite 30B in FR-PLA-

C30B. ....................................................................................................................................... 174

Figure 103: Evolution of viscosity of FR-PLA and FR-PLA-C30B as a function of ageing

exposure to T/UV (●), T/RH (▪) and T/UV/RH (Δ) conditions (a) and molecular mass (b). .... 175

Figure 104: RHR curves of unaged PLA (●) and 60 days exposure to T/UV (●), T/RH (●) and

T/UV/RH (●) conditions. ......................................................................................................... 176

Figure 105: RHR curves of unaged FR-PLA (●) and 60 days exposure to T/UV (●), T/RH (●) and

T/UV/RH (●) conditions. ......................................................................................................... 177

Figure 106: RHR curves of unaged FR-PLA-C30B (●) and 60 days exposure to T/UV (●), T/RH

(●) and T/UV/RH (●) conditions. ............................................................................................ 178

Figure 107: RHR curves of unaged FR-PLA-20% (●), 17 days (●) and 22 days (●) aged under

T/UV/RH conditions................................................................................................................ 179

Figure 108: RHR curves of unaged FR-PLA-C30B-20% (●), 18 days (●) and 22 days (●) aged

under T/UV/RH conditions. .................................................................................................... 180

Figure 109: Evolution of the char expansion of FR-PLA-20% as a function of ageing under

T/UV/RH exposure. ................................................................................................................ 181

Figure 110: Evolution of the char expansion of FR-PLA-C30B-20% as a function of ageing

under T/UV/RH exposure. ...................................................................................................... 181

Figure 111: M

versus ageing time of FR-PLA 20% and 30% (●) and FR-PLA-C30B (●) 20% and

30% materials aged under T/UV/RH conditions. ................................................................... 183

Figure 112: Schematic representation of the effect of ageing on the intumescent mechanism

of FR-PLA-C30B-20%. .............................................................................................................. 185

Figure 113: RHR curves as a function of time of PLA at 35 kW/m²(●), 50 kW/m² (●) and

75 kW/m² (●). ......................................................................................................................... 193

Figure 114: FR-PLA-C20A (a), FR-PLA-C30B (b) and FR-PLA-CNa

T/RH conditions. ..................................................................................................................... 222

Figure 115: M

versus ageing time of FR-PLA-C20A (●), FR-PLA-C30B (▪) and FR-PLA-C30B (Δ)

samples aged under T/RH conditions. ................................................................................... 223

Figure 116: Evolution of glass transition temperature (T

) of FR-PLA-C20A (●), FR-PLA-C30B

(▪) and FR-PLA-CNa

(Δ) as a function of exposure time under T/RH conditions. ................. 224

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

List of figures, tables and equations

201

Figure 117: Evolution of melting temperature (T

) of FR-PLA-C20A (●), FR-PLA-C30B (▪) and

FR-PLA-CNa

(Δ) as a function of exposure time under T/RH conditions. ............................. 224

Figure 118: Evolution of crystallinity of FR-PLA-C20A (●), FR-PLA-C30B (▪) and FR-PLA-CNa

(∆) as a function of ageing time under T/RH conditions. ....................................................... 225

Figure 119: FTIR (ATR) spectra of PLA, FR-PLA, FR-PLA 105 days (T/RH) and FR-PLA 90 days

(T/UV/RH) in the range of 600 t0 3600 cm

-1

. ......................................................................... 226

Figure 120: FTIR (ATR) spectra of PLA, FR-PLA-C30B, FR-PLA-C30B 105 days (T/RH) and FR-

PLA-C30B 90 days (T/UV/RH) in the range of 600 t0 3600 cm

-1

. ........................................... 227

Figure 121: Example of 2D HYSCORE microwave pulses. ...................................................... 228

Figure 122: 2D HYSCORE spectrum of a coat standard sample. ............................................ 228

Figure 123: TGA curves of pure compounds, PLA (●), Mel (●), APP (●) and C30B (●) under air

flow at 10°C/min. ................................................................................................................... 229

Figure 124: DTG of pure compounds, PLA (●), Mel (●), APP (●) and C30B (●) under air flow at

10°C/min. ................................................................................................................................ 230

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

List of figures, tables and equations

202

List of Tables

Table 1: Infrared spectroscopy data for PLA. ........................................................................... 29

Table 2: Unit cell parameters for PLLA and stereocomplex crystals [5]. ................................. 32

Table 3: PLA mechanical properties compared to those of most common polymers used in

commodity applications [83].................................................................................................... 35

Table 4: Properties of Ingeo PLA 4032D. .................................................................................. 68

Table 5: Summary of additives used. ....................................................................................... 69

Table 6: Composition and name of the formulations PLA, FR-PLA, FR-PLA-C30B at 30 wt.-%

loading. ..................................................................................................................................... 70

Table 7: Composition and name of the formulations FR-PLA, FR-PLA-C30B at 20 wt.-%

loading. ..................................................................................................................................... 70

Table 8: Temperature profile of the extruder from hopper to die. ......................................... 70

Table 9: Conditions in accelerated weathering chambers for T/UV and T/UV/RH ageing. ..... 73

Table 10: L*a*b* values of unaged and 60 days aged PLA under T/UV, T/RH and T/UV/RH

conditions. ................................................................................................................................ 89

Table 11: M

and polydispersity index (PI) data as a function of ageing of PLA exposed to

T/UV, T/RH and T/UV/RH conditions. ...................................................................................... 90

Table 12: TGA data (10°C/min under air) obtained after ageing of PLA exposed to T/UV, T/RH

and T/UV/RH conditions. ......................................................................................................... 93

Table 13: DSC data of PLA reported as a function of ageing exposure to T/UV, T/RH and

T/UV/RH conditions.................................................................................................................. 94

Table 14: Crystallinity of PLA as a function of ageing duration under T/UV, T/RH and T/UV/RH

exposure. .................................................................................................................................. 98

Table 15: Evolution of the band at 1845 cm

-1

of PLA before and after ageing under T/UV and

T/UV/RH as a function of ageing time. .................................................................................. 113

Table 16: Number average molecular mass (M

) of PLA, FR-PLA and FR-PLA-C30B materials.

................................................................................................................................................ 126

Table 17: M

and PI data as function of ageing of PLA, FR-PLA and FR-PLA-C30B samples aged

under T/UV, T/RH and T/UV/RH. ........................................................................................... 127

Table 18: DSC data of PLA, FR-PLA and FR-PLA-C30B as a function of M

. ............................ 130

Table 19: DSC data of FR-PLA as a function of M

and ageing time under T/UV, T/RH and

T/UV/RH. ................................................................................................................................ 131

Table 20: DSC data of FR-PLA-C30B as a function of M

and ageing time under T/UV, T/RH

and T/UV/RH. ......................................................................................................................... 132

Table 21: Crystallinity (χ) data for PLA, FR-PLA and FR-PLA-C30B as a function of M

.......... 136

Table 22: Evolution of Crystallinity for FR-PLA as a function of M

for different ageing

durations under T/UV, T/RH and T/UV/RH exposure. ........................................................... 137

Table 23: Evolution of Crystallinity for FR-PLA-C30B as a function of M

for different ageing

duration under T/UV, T/RH and T/UV/RH exposure. ............................................................. 138

Table 24: At % on the surface of FR PLA determined by XPS as a function of ageing. .......... 144

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

List of figures, tables and equations

203

Table 25: At % on the surface of FR PLA-C30B determined by XPS as a function of ageing.. 144

Table 26: TGA data (10°C/min under air) obtained for PLA, FR-PLA and FR-PLA-C30B. ........ 160

Table 27: MLC values obtained for PLA, FR-PLA and FR-PLA-C30B........................................ 162

Table 28: LOI values of PLA, FR-PLA and FR-PLA-C30B. ......................................................... 164

Table 29: TGA data (10°C/min under air) obtained after ageing of FR-PLA exposed to T/UV,

T/RH and T/UV/RH conditions. .............................................................................................. 167

Table 30: TGA data (10°C/min under air) obtained after ageing of FR-PLA-C30B exposed to

T/UV, T/RH and T/UV/RH conditions. .................................................................................... 168

Table 31: Comparison between theoretical and experimental residual weight (wt. %)

obtained at 800°c for FR-PLA and FR-PLA-C30B. ................................................................... 170

Table 32: LOI ranking of PLA as a function of exposure to T/UV, T/RH and T/UV/RH

conditions. .............................................................................................................................. 170

Table 33: LOI ranking of FR-PLA as a function of exposure to T/UV, T/RH and T/UV/RH

conditions. The (+x) are calculated using the unaged formulation as reference. ................. 171

Table 34: LOI ranking of FR-PLA-C30B as a function of exposure to T/UV, T/RH and T/UV/RH

conditions. The (+x) are calculated using the unaged formulation as reference. ................. 171

Table 35: Viscosity data obtained for FR-PLA exposed to T/UV, T/RH and T/UV/RH conditions.

................................................................................................................................................ 173

Table 36: Viscosity data obtained for FR-PLA-C30B exposed to T/UV, T/RH and T/UV/RH

conditions. .............................................................................................................................. 173

Table 37: MLC values obtained for PLA as a function of exposure to T/UV, T/RH and T/UV/RH

conditions. .............................................................................................................................. 176

Table 38: MLC values obtained for FR-PLA as a function of exposure to T/UV, T/RH and

T/UV/RH conditions................................................................................................................ 177

Table 39: MLC values obtained for FR-PLA-C30B as a function of exposure to T/UV, T/RH and

T/UV/RH conditions................................................................................................................ 178

Table 40: MLC values obtained for FR-PLA-20% as a function of exposure time to T/UV/RH

conditions. .............................................................................................................................. 179

Table 41: MLC values obtained for FR-PLA-C30B-20% as a function of exposure time to

T/UV/RH conditions................................................................................................................ 180

Table 42: M

data of FR-PLAs-20% and FR-PLAs-30% as a function of ageing exposure to

T/UV/RH conditions................................................................................................................ 182

Table 43: MLC values obtained for PLA at 35 kW/m², 50 kW/m² and 75 kW/m². ................ 193

Table 44: L*a*b* values of PLA, FR-PLA and FR-PLA-C30B. ................................................... 220

Table 45: L*a*b* values of unaged and aged FR-PLA under T/UV, T/RH and T/UV/RH

conditions. .............................................................................................................................. 221

Table 46: L*a*b* values of unaged and aged FR-PLA-C30B under T/UV, T/RH and T/UV/RH

conditions. .............................................................................................................................. 221

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

List of figures, tables and equations

204

List of equations

Equation 1 ................................................................................................................................ 31

Equation 2 ................................................................................................................................ 31

Equation 3 ................................................................................................................................ 33

Equation 4 ................................................................................................................................ 47

Equation 5 ................................................................................................................................ 47

Equation 6 ................................................................................................................................ 47

Equation 7 ................................................................................................................................ 47

Equation 8 ................................................................................................................................ 48

Equation 9 ................................................................................................................................ 49

Equation 10 .............................................................................................................................. 49

Equation 11 .............................................................................................................................. 49

Equation 12 .............................................................................................................................. 49

Equation 13 .............................................................................................................................. 50

Equation 14 .............................................................................................................................. 50

Equation 15 .............................................................................................................................. 50

Equation 16 .............................................................................................................................. 51

Equation 17 .............................................................................................................................. 51

Equation 18 .............................................................................................................................. 52

Equation 19 .............................................................................................................................. 52

Equation 20 .............................................................................................................................. 52

Equation 21 .............................................................................................................................. 52

Equation 22 .............................................................................................................................. 52

Equation 23 .............................................................................................................................. 75

Equation 24 .............................................................................................................................. 76

Equation 25 .............................................................................................................................. 77

Equation 26 .............................................................................................................................. 78

Equation 27 .............................................................................................................................. 79

Equation 28 .............................................................................................................................. 81

Equation 29 .............................................................................................................................. 84

Equation 30 .............................................................................................................................. 96

Equation 31 .............................................................................................................................. 97

Equation 32 ............................................................................................................................ 108

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

References

205

References

1. PlasticsEurope. Plastics - the Facts 2014/2015 An analysis of European plastics production,

demand and waste data. 2014 [cited 2015 06-2015]; Available from:

http://www.plasticseurope.fr/.

2. PlasticsEurope. Plastics - the Facts 2013 - An analysis of European plastics production demand

and waste data. 2013 [cited 2015 06-2015]; Available from: http://www.plasticseurope.fr/.

3. Shen, L., Haufe J, Patel M K, Product overview and market projection of emerging bio-based

plastics. Copernicus Institute for Sustainable Development and Innovation, Utrecht University,

2009.

4. EuropeanBioplastics. Global production capacities of bioplastics 2015 [cited 2016 02-2016];

Available from: http://www.european-bioplastics.org/.

5. Auras, R., Harte B, Selke S, An overview of polylactides as packaging materials. Macromolecular

Bioscience, 2004. 4(9): p. 835-864.

6. Association, N.F.P. Fires in the US. 2013 [cited 2016 04-2016]; Available from:

http://www.nfpa.org/news-and-research/fire-statistics-and-reports/fire-statistics/fires-in-

the-us.

7. Levchik, S., V., Weil E D, Flame retardancy of thermoplastic polyesters—a review of the recent

literature. Polymer International, 2005. 54(1): p. 11-35.

8. Bourbigot, S., Fontaine G, Flame retardancy of polylactide: an overview. Polymer Chemistry,

2010. 1(9): p. 1413-1422.

9. Réti, C., Casetta M, Duquesne S, Bourbigot S, Delobel R, Flammability properties of intumescent

PLA including starch and lignin. Polymers for Advanced Technologies, 2008. 19(6): p. 628-635.

10. Bourbigot, S., Duquesne S, Fontaine G, Bellayer S, Turf T, Samyn F, Characterization and

reaction to fire of polymer nanocomposites with and without conventional flame retardants.

Molecular Crystals and Liqid Crystals, 2008. 486: p. 325.

11. Solarski, S., Ferreira M, Devaux E, Fontaine G, Bachelet P, Bourbigot S., et al, Designing

polylactide/clay nanocomposites for textile applications: effect of processing conditions,

spinning and characterization. Journal of Applied Polymer Science, 2008. 109: p. 841.

12. Fontaine, G., Bourbigot S, Intumescent polylactide : A nonflammable material. Journal of

Applied Polymer Science, 2009. 113(2009): p. 3860-3865.

13. Nishida, H., Fan Y, Mori T, Oyagi N, Shirai Y, Endo T, Feedstock recycling of flame-resisting

poly(lactic acid)/aluminum hydroxide composite to L,L-lactide. Industrial & Engineering

Chemistry Research, 2005. 44: p. 1433-1437.

14. Yanagisawa, T., Kiuchi Y, Iji M, Enhanced flame retardancy of polylactic acid with aluminium

tri-hydroxide and phenolic resins. Kobunshi Ronbunshu, 2009. 66(49-54).

15. Kubokawa, H., Ohta M, Hatakeyama T, Flame-retarding of polylactide fabrics. Sen'i gakkaishi,

1999. 55: p. 290-297.

16. Kimura, K., Horikoshi Y, Bio-based polymers. Fujitsu Scienific and Technical Journal, 2005(41):

p. 173-180.

17. Kubokawa, H., Komatsu H, Flammability and thermal decomposition behavior of polylactide -

cellulose union cloths. Sen'i gakkaishi, 2000. 56: p. 497-500.

18. Kubokawa, H., Hatakeyama T, Thermal decomposition behavior of polylactide fabrics treated

with flame retardants. Sen'i gakkaishi, 1999. 55: p. 349-355.

19. Isitman, N., A., Dogan M, Bayramli E, Kaynak C, The role of nanoparticle geometry in flame

retardancy of polylactide nanocomposites containing aluminium phosphinate. Polymer

Degradation and Stability, 2012. 97(8): p. 1285-1296.

20. Bourbigot, S., Fontaine G, Duquesne S, Delobel R, PLA nanocomposites: quantification of clay

nanodispersion and reaction to fire. International Journal of nanotechnology, 2008. 5: p. 683-

692.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

References

206

21. Kashiwagi, T., Du F, Douglas J F, Winey K I, Harris R H, Shields J R, Nanoparticle networks reduce

the flammability of polymer nanocomposites. Nature Materials, 2005. 4(12): p. 928-933.

22. Murariu, M., Dechief A L, Bonnaud L, Paint Y, Gallos A, Fontaine G, Bourbigot S, Dubois P, The

production and properties of polylactide composites filled with expanded graphite. Polymer

Degradation and Stability, 2010. 95(5): p. 889-900.

23. Shumao, L., Jie R, Hua Y, Tao Y, Weizhong Y, Influence of ammonium polyphosphate on the

flame retardancy and mechanical properties of ramie fiber-reinforced poly(lactic acid)

biocomposites. Polymer International, 2010. 59(2): p. 242-248.

24. Fontaine, G., Gallos A, Bourbigot S, Role of montmorillonite for enhancing fire retardancy of

intumescent PLA. Fire Safety Science, 2014. 11: p. 808-820.

25. Fukushima, K., Tabuani D, Dottori M, Armentano I, Kenny J M, Camino G, Effect of temperature

and nanoparticle type on hydrolytic degradation of poly(lactic acid) nanocomposites. Polymer

Degradation and Stability, 2011. 96(12): p. 2120-2129.

26. Ho, K., Pometto A L, Hinz P N, Effects of temperature and relative humidity on polylactic acid

plastic degradation. Journal of Environmental Polymer Degradation, 1999. 7(2): p. 83-92.

27. Tsuji, H., Echizen Y, Nishimura Y, Photodegradation of biodegradable polyesters: A

comprehensive study on poly(l-lactide) and poly(ɛ-caprolactone). Polymer Degradation and

Stability, 2006. 91(5): p. 1128-1137.

28. Janorkar, A.V., Metters A T, Hirt D E, Degradation of poly(L-lactide) films under ultraviolet-

induced photografting and sterilization conditions. Journal of Applied Polymer Science, 2007.

106(2): p. 1042-1047.

29. Copinet, A., Bertrand C, Govindin S, Coma V, Couturier Y, Effects of ultraviolet light (315 nm),

temperature and relative humidity on the degradation of polylactic acid plastic films.

Chemosphere, 2004. 55(5): p. 763-773.

30. Badia, J., D., Santonja-Blasco L, Martinez-Felipe A, Ribes-Greus A, Hygrothermal ageing of

31. Sambha’a, E.L., Lallam A, Jada A, Effect of hydrothermal polylactic acid degradation on polymer

molecular weight and surface properties. Journal of Polymers and the Environment, 2010.

18(4): p. 532-538.

32. Zaidi, L., Kaci M, Bruzaud S, Bourmaud A, Grohens Y, Effect of natural weather on the structure

and properties of polylactide/Cloisite 30B nanocomposites. Polymer Degradation and Stability,

2010. 95(9): p. 1751-1758.

33. Belbachir, S., Zaïri F, Ayoub G, Maschke U, Naït-Abdelaziz M, Gloaguen J M, Benguediab M,

Lefebvre J M, Modelling of photodegradation effect on elastic–viscoplastic behaviour of

amorphous polylactic acid films. Journal of the Mechanics and Physics of Solids, 2010. 58(2):

p. 241-255.

34. Bocchini, S., Fukushima K, Di Blasio A, Fina A, Frache A, Geobaldo F, Polylactic acid and

polylactic acid-based nanocomposite photooxidation. Biomacromolecules, 2010. 11(11): p.

2919-2926.

35. Gardette, M., Thérias S, Gardette J L, Murariu M, Dubois P, Photooxidation of

polylactide/calcium sulphate composites. Polymer Degradation and Stability, 2011. 96(4): p.

616-623.

36. Fukushima, K., Tabuani D, Camino G, Poly(lactic acid)/clay nanocomposites: effect of nature

and content of clay on morphology, thermal and thermo-mechanical properties. Materials

Science and Engineering: C, 2012. 32(7): p. 1790-1795.

37. Paul, M.A., Delcourt C, Alexandre M, Degée P, Monteverde F, Dubois P.,

Polylactide/montmorillonite nanocomposites: study of the hydrolytic degradation. Polymer

Degradation and Stability, 2005. 87(3): p. 535-542.

38. Zhou, Q., Xanthos M, Nanoclay and crystallinity effects on the hydrolytic degradation of

polylactides. Polymer Degradation and Stability, 2008. 93(8): p. 1450-1459.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

References

207

39. Araujo, A., Botelho G L, Silva M, Machado A V, UV Stability of Poly(Lactic Acid) Nanocomposites.

Journal of Materials Science and Engineering, 2012. B(3 (2)): p. 75-83.

40. Vahabi, H., Sonnier R, Ferry L, Effects of ageing on the fire behaviour of flame-retarded

polymers: a review. Polymer International, 2015. 64(3): p. 313-328.

41. Inata, H., Maki I, Ishikawa T, Takeda K, Diffusion of additives and deterioration with passage of

time in polypropylene. Journal of Applied Polymer Science, 2006. 99: p. 2152-2162.

42. Colonna, S., Cuttica F, Frache A, Aging of EVA/organically modified clay: Effect on dispersion,

distribution and combustion behavior. Polymer Degradation and Stability, 2014. 107: p. 184-

187.

43. Oztekin, E., S., Crowley S B, Lyon R E, Stoliarov S I, Patel P, Hull T R, Sources of variability in fire

test data: A case study on poly(aryl ether ether ketone) (PEEK). Combustion and Flame, 2012.

159(4): p. 1720-1731.

44. Levchik, S.V., Bright D A, Alessio G R, Dashevsky S, New halogen-free fire retardant for

engineering plastic applications. Journal of Vinyl and Additive Technology, 2001. 7(2): p. 98-

103.

45. Sinturel, C., Philippart JL, Lemaire J, Gardette JL, Photooxidation of fire retarded polypropylene.

I. Photoageing in accelerated conditions. European Polymer Journal, 1999. 35(10): p. 1773-

1781.

46. Sinturel, C., Lemaire J, Gardette JL, Photooxidation of fire retarded polypropylene: III.

Mechanism of HAS inactivation. European Polymer Journal, 2000. 36(7): p. 1431-1443.

47. Diagne, M., Guèye M, Vidal L, Tidjani A, Thermal stability and fire retardant performance of

photo-oxidized nanocomposites of polypropylene-graft-maleic anhydride/clay. Polymer

Degradation and Stability, 2005. 89(3): p. 418-426.

48. Diagne, M., Gueye M, Dasilva A, Vidal L, Tidjani A, The effect of photo-oxidation on thermal and

fire retardancy of polypropylene nanocomposites. Journal of Materials Science, 2006. 41: p.

7005-7010.

49. Reti, C., Casetta M, Duquesne S, Bourbigot S, Delobel R, Flammability properties of intumescent

PLA including strach and lignin. Polymers for Advanced Technologies, 2008. 19: p. 173.

50. Bourbigot, S., Le Bras M, Duquesne S, Rochery M, Recent advances for intumescent polymers.

Macromolecular Materials and Engineering, 2004. 289(6): p. 499-511.

51. Bourbigot, S., Fontaine G, Gallos A, Bellayer S, Reactive extrusion of PLA and of PLA/carbon

nanotubes nanocomposite: processing, characterization and flame retardancy. Polymers for

Advanced Technologies, 2011. 22(1): p. 30-37.

52. Lim, L.T., Auras R, Rubino M, Processing technologies for poly(lactic acid). Progress in Polymer

Science, 2008. 33(8): p. 820-852.

53. Kricheldorf, H., R., Kraiser-Saunders I, Jurgens C, Wolter D, Polylactides synthesis,

characterization and medical application. Macromolecular Symposia, 1996. 103(85).

54. Hyon, S., H., Jamshidi K, Ikada Y, Synthesis of polylactides with different molecular weights.

Biomaterials, 1997. 18(22): p. 1503-1508.

55. Kaihara, S., Matsumura S, Mikos A G, Fisher J P, Synthesis of poly(L-lactide) and polyglycolide

by ring-opening polymerization. Nat. Protocols, 2007. 2(11): p. 2767-2771.

56. Dubois, P., Degee P, Jerome R, Teyssie P, Macromolecular engineering of polylactones and

polylactides. 8. Ring-opening polymerization of iε-caprolactone initiated by primary amines

and trialkylaluminum. Macromolecules, 1992. 25(10): p. 2614-2618.

57. Miyamoto, T., Inagaki H, The stereocomplex formation in poly(methyl methacrylate) and the

stereospecific solymerization of Its monomer. Polym J, 1970. 1(1): p. 46-54.

58. Fukuzawa, T., Uematsu I, Uematsu Y, Physical properties of mixtures of poly([gamma]-benzyl-

L-glutamate) and poly([gamma]-benzyl-D-glutamte). Polymer Journal, 1974. 6(6): p. 537-541.

59. Ikada, Y., Jamshidi K, Tsuji H, Hyon S H, Stereocomplex formation between enantiomeric

poly(lactides). Macromolecules, 1987. 20: p. 904-906.

60. Kakuta, M., Hirata M, Kimura Y, Stereoblock polylactides as high-performance bio-based

polymers. Polymer Reviews, 2009. 49:2: p. 107-140.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

References

208

61. Gallos, A., Polylactides stéréocomplexés et ignifugés : Élaboration par extrusion réactive et

caractérisations, 2011, Ph.D. Thesis, Université Lille 1, Sciences et Technologies. p. 221.

62. Nakayama, N., Hayashi T, Preparation and characterization of poly(l-lactic acid)/TiO2

nanoparticle nanocomposite films with high transparency and efficient photodegradability.

Polymer Degradation and Stability, 2007. 92(7): p. 1255-1264.

63. Bocchini, S., Frache A, Comparative Study of Filler Influence on Polylactide Photooxidation.

eXPRESS Polymer Letters, 2013. 7(5): p. 431-442.

64. Garlotta, D., A literature review of poly(lactic acid). Journal of Polymers and the Environment,

2001. 9(2): p. 63-84.

65. Kister, G., Cassanas G, Vert M, Effects of morphology, conformation and configuration on the

IR and Raman spectra of various poly(lactic acid)s. Polymer, 1998. 39(2): p. 267-273.

66. Drumright, R., E., Gruber P R, Henton D E, Polylactic acid technology. Advanced Materials,

2000. 12(23): p. 1841-1846.

67. Dorgan, J.R., Janzen J, Clayton M P, Hait S B, Knauss D M, Melt rheology of variable L-content

poly(lactic acid). Journal of Rheology, 2005. 49(3): p. 607-619.

68. Jamshidi, K., Hyon S H, Okada Y, Thermal characterizaton of polylactides. Polymer, 1988.

29(2229).

69. Celli, A., Scandola M, Thermal properties and physical ageing of poly (l-lactic acid). Polymer,

1992. 33(13): p. 2699-2703.

70. Cai, H., Dave V, Gross R A, McCarthy S P, Effects of physical aging, crystallinity, and orientation

on the enzymatic degradation of poly(lactic acid). Journal of Polymer Science Part B: Polymer

Physics, 1996. 34(16): p. 2701-2708.

71. Witzke, D.R., Introduction to properties, engineering, and prospects of polylactide polymers.

PhD thesis. East Lansing, MI: Michigan State University, 1997.

72. Hartmann, M.H., High molecular weight polylactic acid polymers, in Biopolymers from

Renewable Resources, D.L. Kaplan, Editor 1998, Springer Berlin Heidelberg: Berlin, Heidelberg.

p. 367-411.

73. Hoogsteen, W., Postema A R, Pennings A J, Ten Brinke G, Zugenmaier P, Crystal structure,

conformation and morphology of solution-spun poly(L-lactide) fibers. Macromolecules, 1990.

23(2): p. 634-642.

74. Sasaki, S., Asakura T, Helix distortion and crystal structure of the α-form of poly(l-lactide).

Macromolecules, 2003. 36(22): p. 8385-8390.

75. Cartier, L., Okihara T, Ikada Y, Tsuji H, Puiggali J, Lot B, Epitaxial crystallization and crystalline

polymorphism of polylactides. Polymer, 2000. 41(25): p. 8909-8919.

76. Tsuji, H., Ikada Y, Properties and morphologies of poly(l-lactide): 1. Annealing condition effects

on properties and morphologies of poly(l-lactide). Polymer, 1995. 36(14): p. 2709-2716.

77. Sarasua, J.R., Arraiza A L, Balerdi P, Maiza I, Crystallinity and mechanical properties of optically

pure polylactides and their blends. Polymer Engineering & Science, 2005. 45(5): p. 745-753.

78. Pyda, M., Bopp R C, Wunderlich B, Heat capacity of poly(lactic acid). The Journal of Chemical

Thermodynamics, 2004. 36(9): p. 731-742.

79. Kolstad, J.J., Crystallization kinetics of poly(L-lactide-co-meso-lactide). Journal of Applied

Polymer Science, 1996. 62(7): p. 1079-1091.

80. Kishore, K.V.R., Nucleation parameters for polymer crystallization from non-isothermal

thermal analysis. Colloid and Polymer Science, 1988. 266(11): p. 999-1002.

81. Kishore, K., Vasanthakumari R, Pennings A J, Crystallization kinetics of poly(l-lactic acid).

Polymer, 1983. 24(2): p. 175-178.

82. Perego, G., Cella D, Bastioli C, Effect of molecular weight and crystallinity on poly(lactic acid)

mechanical properties. Journal of Applied Polymer Science, 1996. 59: p. 37-43.

83. Liu, H., Zhang J, Research progress in toughening modification of poly(lactic acid). Journal of

Polymer Science Part B: Polymer Physics, 2011. 49(15): p. 1051-1083.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

References

209

84. Al-Itry, R., Lamnawar K, Maazouz A, Improvement of thermal stability, rheological and

mechanical properties of PLA, PBAT and their blends by reactive extrusion with functionalized

epoxy. Polymer Degradation and Stability, 2012. 97(10): p. 1898-1914.

85. Yang, S.L., Wu Z H, Yang W, Yang M B, Thermal and mechanical properties of chemical

crosslinked polylactide (PLA). Polymer Testing, 2008. 27(8): p. 957-963.

86. Montse, C.H., Del Valle S, Hentges E, Bleuet P, Lacroix D, Planell J A, Mechanical and structural

characterisation of completely degradable polylactic acid/calcium phosphate glass scaffolds.

Biomaterials, 2007. 28(30): p. 4429-4438.

87. Laoutid, F., Bonnaud L, Alexandre M, Lopez-Cuesta J M, Dubois Ph, New prospects in flame

retardant polymer materials: From fundamentals to nanocomposites. Materials Science and

Engineering: R: Reports, 2009. 63(3): p. 100-125.

88. Vovelle, C., Delfau J L, Combustion des plastiques. Technique de l'Ingénieur, 1997.

89. Bourbigot, S., Delobel R, Duquesne S, Comportement au feu des composites. Technique de

l'Ingénieur, 2006.

90. European_Flame_Retardants_Association, How do flame retardants work ? 2006.

91. Müller, M., Systemic approach of the synergism in flame retarded intumescent polyurethanes,

2012, Ph.D. Thesis, Université Lille 1, Sciences et Technologies. p. 268.

92. Wang, H., Wang Q, Huang Z, Shi W, Synthesis and thermal degradation behaviors of

hyperbranched polyphosphate. Polymer Degradation and Stability, 2007. 92(10): p. 1788-1794.

93. Price, D., Pyrah K, Hull T R, Milnes G J, Ebdon J R, Hunt B J, Joseph P, Flame retardance of

poly(methyl methacrylate) modified with phosphorus-containing compounds. Polymer

Degradation and Stability, 2002. 77(2): p. 227-233.

94. Lammers, M., Klop E A, Northolt M G, Sikkema D J, Mechanical properties and structural

transitions in the new rigid-rod polymer fibre PIPD (`M5') during the manufacturing process.

Polymer, 1998. 39(24): p. 5999-6005.

95. Cheng, T., W., Chiu J P, Fire-resistant geopolymer produced by granulated blast furnace slag.

Minerals Engineering, 2003. 16(3): p. 205-210.

96. Wang, G., Yang J, Influences of expandable graphite modified by polyethylene glycol on fire

protection of waterborne intumescent fire resistive coating. Surface and Coatings Technology,

2010. 204(21–22): p. 3599-3605.

97. Wang, Z., Han E, Ke W, An investigation into fire protection and water resistance of intumescent

nano-coatings. Surface and Coatings Technology, 2006. 201(3–4): p. 1528-1535.

98. Duquesne, S., Magnet S, Jama C, Delobel R, Intumescent paints: fire protective coatings for

metallic substrates. Surface and Coatings Technology, 2004. 180–181: p. 302-307.

99. Gardelle, B., Duquesne S, Vandereecken P, Bellayer S, Bourbigot S, Resistance to fire of curable

silicone/expandable graphite based coating: Effect of the catalyst. European Polymer Journal,

2013. 49(8): p. 2031-2041.

100. Jimenez, M., Duquesne S, Bourbigot S, Intumescent fire protective coating: Toward a better

understanding of their mechanism of action. Thermochimica Acta, 2006. 449(1–2): p. 16-26.

101. Bourbigot, S., Le Bras M, Leeuwendal R, Shen K K, Schubert D, Recent advances in the use of

zinc borates in flame retardancy of EVA. Polymer Degradation and Stability, 1999. 64(3): p.

419-425.

102. Bourbigot, S., Duquesne S, Fire retardant polymers: recent developments and opportunities.

Journal of Materials Chemistry, 2007. 17(22): p. 2283-2300.

103. Morgan, A., B., Wilkie C A, Fire retardancy of polymeric materials. CRC Press, Taylor and Francis

Group, 2010.

104. Troitzsch, J., Plastics flammability handbook. Munich, Carl Hanser Verlag Munich, 2004.

105. Lewin, M., Unsolved problems and unanswered questions in flame retardance of polymers.

Polymer Degradation and Stability, 2005. 88(1): p. 13-19.

106. Alexandre, M., Dubois P, Polymer-layered silicate nanocomposites: preparation, properties and

uses of a new class of materials. Materials Science and Engineering: R: Reports, 2000. 28(1–2):

p. 1-63.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

References

210

107. Sinha Ray, S., Okamoto M, Biodegradable polylactide and its nanocomposites: opening a new

dimension for plastics and composites. Macromolecular Rapid Communications, 2003. 24(14):

p. 815-840.

108. Bourbigot, S., Vanderhart D L, Gilman J W, Bellayer S, Stretz H, Paul D R, Solid state NMR

characterization and flammability of styrene–acrylonitrile copolymer montmorillonite

nanocomposite. Polymer, 2004. 45(22): p. 7627-7638.

109. Zheng, X., Wilkie C A, Nanocomposites based on poly (ϵ-caprolactone) (PCL)/clay hybrid:

polystyrene, high impact polystyrene, ABS, polypropylene and polyethylene. Polymer

Degradation and Stability, 2003. 82(3): p. 441-450.

110. Kashiwagi, T., Grulke E, Hilding J, Harris R, Awad W, Douglas J, Thermal degradation and

flammability properties of poly(propylene)/carbon nanotube composites. Macromolecular

Rapid Communications, 2002. 23(13): p. 761-765.

111. Wang, S., Tambraparni M, Qiu J, Tipton J, Dean D, Thermal expansion of graphene composites.

Macromolecules, 2009. 42(14): p. 5251-5255.

112. Steurer, P., Wissert R, Thomann R, Mülhaupt R, Functionalized graphenes and thermoplastic

nanocomposites based upon expanded graphite oxide. Macromolecular Rapid

Communications, 2009. 30(4-5): p. 316-327.

113. Vandersall, H., J., Intumescent coating systems, their development and chemistry. Journal of

Fire Flammability, 1971. 2: p. 97-140.

114. Bourbigot, S., Le Bras M, Delobel R, Carbonization mechanisms resulting from intumescence

association with the ammonium polyphosphate-pentaerythritol fire retardant system. Carbon,

1993. 31(8): p. 1219-1230.

115. Bourbigot, S., Le Bras M, Delobel R, Decressain R, Amoureux, J P, Synergistic effect of zeolite in

an intumescence process: study of the carbonaceous structures using solid-state NMR. Journal

of the Chemical Society, Faraday Transactions, 1996. 92: p. 149-158.

116. Bourbigot, S., Le Bras M, Dabrowski F, Gilman J W, Kashiwagi T, PA-6 clay nanocomposite

hybrid as char forming agent in intumescent formulations. Fire and Materials, 2000. 24(4): p.

201-208.

117. White, J.R., Polymer ageing: physics, chemistry or engineering? Time to reflect. Comptes

Rendus Chimie, 2006. 9(11–12): p. 1396-1408.

118. Bonten, C., Berlich R, Aging and chemical resistance. Hanser Verlag, 2001.

119. Pillai, S., K., Sinha Ray S, Scriba M, Ojijo V, Hato M J, Morphological and thermal properties of

photodegradable biocomposite films. Journal of Applied Polymer Science, 2013. 129(1): p. 362-

370.

120. Zuo, X., Shao H, Zhang D, Hao Z, Guo J, Effects of thermal-oxidative aging on the flammability

and thermal-oxidative degradation kinetics of tris(tribromophenyl) cyanurate flame retardant

PA6/LGF composites. Polymer Degradation and Stability, 2013. 98(12): p. 2774-2783.

121. Gil-Castell, O., Badia J D, Kittikorn T, Strömberg E, Martínez-Felipe A, Ek M, Karlsson S, Ribes-

Greus A, Hydrothermal ageing of polylactide/sisal biocomposites. Studies of water absorption

behaviour and physico-chemical performance. Polymer Degradation and Stability, 2014. 108:

p. 212-222.

122. Perry, M., C., Vail M A, Devries K L, Effect of NOx environments and stress on degradation of

mechanical properties of Kevlar 49 and nylon 6. Polymer Engineering & Science, 1995. 35(5):

p. 411-418.

123. Deroiné, M., Le Duigou A, Corre Y M, Le Gac P Y, Davies P, César G, Bruzaud S, Accelerated

ageing of polylactide in aqueous environments: Comparative study between distilled water and

seawater. Polymer Degradation and Stability, 2014. 108: p. 319-329.

124. Beyler, C.L., Hirschler M M, Thermal decomposition of polymers. SFPE Handbook of Fire

Protection Engineering 2, Section 1, Chapter 7, 2002: p. 111-131.

125. Singh, B., Sharma N, Mechanistic implications of plastic degradation. Polym Degrad Stab, 2008.

93: p. 561 - 584.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

References

211

126. Azapagic, A., Emsley A, Hamerton I, Polymers: The environment and sustainable development.

2003.

127. Rasselet, D., Ruellan A, Guinault A, Miquelard-Garnier G, Sollogoub C, Fayolle B, Oxidative

degradation of polylactide (PLA) and its effects on physical and mechanical properties.

European Polymer Journal, 2014. 50(0): p. 109-116.

128. Le Duigou, A., Davies P, Baley C, Seawater ageing of flax/poly(lactic acid) biocomposites.

Polymer Degradation and Stability, 2009. 94(7): p. 1151-1162.

129. Yousif, E., Haddad R, Photodegradation and photostabilization of polymers, especially

polystyrene: review. SpringerPlus, 2013. 2(1): p. 398.

130. Martin, J.W., Chin J W, Nguyen T, Reciprocity law experiments in polymeric photodegradation:

a critical review. Progress in Organic Coatings, 2003. 47(3–4): p. 292-311.

131. Hamid, S.H., Prichard W H, Mathematical modeling of weather-induced degradation of

polymer properties. Journal of Applied Polymer Science, 1991. 43(4): p. 651-678.

132. Andrady, A.L., Pegram J E, Tropsha Y, changes in carbonyl index and average molecular weight

on embrittlement of enhanced photo-degradable polyethylene. J Environ Degrad Polym, 1993.

1: p. 171-179.

133. Rånby, B., Photodegradation and photo-oxidation of synthetic polymers. Journal of Analytical

and Applied Pyrolysis, 1989. 15(0): p. 237-247.

134. Jensen, J.P.T., Kops J., Photochemical degradation of blends of polystyrene and poly(2,6 -

dimethyl-1,4-phenylene oxide). Journal of Polymer Science: Polymer Chemistry Edition, 1980.

18(8): p. 2737-2746.

135. Sheldrick, G.E., Vogl O, Induced photodegradation of styrene polymers: A survey. Polymer

Engineering & Science, 1976. 16(2): p. 65-73.

136. Weiss, P., The effects of hostile environments on coatings and plastics. Journal of Polymer

Science: Polymer Letters Edition, 1984. 22(5): p. 296-296.

137. Ito, M., Nagai K, Degradation issues of polymer materials used in railway field. Polymer

Degradation and Stability, 2008. 93(10): p. 1723-1735.

138. Feldman, D., Polymer weathering: photo-oxidation. Journal of Polymers and the Environment,

2002. 10(4): p. 163-173.

139. Davidson, R., S., Goodwin D, Fornier de Violet P, The mechanism of the norrish type II reaction

of α-keto-acids and esters. Tetrahedron Letters, 1981. 22(26): p. 2485-2486.

140. Wang, Z., Norrish type I reaction, in Comprehensive Organic Name Reactions and

Reagents2010, John Wiley & Sons, Inc.

141. Rabek, J., F, Mechanism of photophysical process and photochemical reaction in polymers.

John wiley and Sons, New York, 1987.

142. Zhou, Q., Xanthos M, Nanosize and microsize clay effects on the kinetics of the thermal

degradation of polylactides. Polymer Degradation and Stability, 2009. 94(3): p. 327-338.

143. Taylor, D., R, Mechanistic aspects of the effects of stress on the rates of photochemical

degradation reactions in polymers. Journal of Macromolecular Science, Part C: Polymer

Reviews, 2004. 44(4): p. 351 - 388.

144. Pizzi, A., Mittal K L, Handbook of adhesive technology, revised and expanded. CRC Press, 2003:

p. 1036.

145. Murata, K., Hirano Y, Sakata Y, Azhar-Uddin MD, Basic study on a continuous flow reactor for

thermal degradation of polymers. Journal of Analytical and Applied Pyrolysis, 2002. 65(1): p.

71-90.

146. Soto-Oviedo, M.A., Lehrle R S, Parsons I W, De Paoli M A, Thermal degradation mechanism and

rate constants of the thermal degradation of poly(epichlorohydrin-co-ethylene oxide), deduced

from pyrolysis-GC-MS studies. Polymer Degradation and Stability, 2003. 81(3): p. 463-472.

147. Engineer, C., Parikh J, Raval A, Review on hydrolytic degradation behavior of biodegradable

polymers from controlled drug delivery system. Trends Biomaterials and Artificial Organs, 2011.

25(2): p. 79-85.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

References

212

148. Hofmann, D., Entrialgo-Castaño, M., Kratz, K., Lendlein, A., Knowledge-based approach

towards hydrolytic degradation of polymer-based biomaterials. Advanced Materials, 2009.

21(32-33): p. 3237-3245.

149. Göpferich, A., Tessmar J, Polyanhydride degradation and erosion. Advanced Drug Delivery

Reviews, 2002. 54(7): p. 911-931.

150. Viera, A., C., Viera J C, Ferra J M, Magalhaes F D, Guedes R M, Marques A T, Mechanical study

of PLA–PCL fibers during in vitro degradation. Journal of the Mechanical Behavior of Biomedical

Materials, 2011. 4(3): p. 451-460.

151. Siparsky, G., L., Voorhees K J, Miao F, Hydrolysis of polylactic acid (PLA) and polycaprolactone

(PCL) in aqueous acetonitrile solutions: autocatalysis. Journal of environmental polymer

degradation, 1998. 6(1): p. 31-41.

152. Göpferich, A., Mechanisms of polymer degradation and erosion. Biomaterials, 1996. 17(2): p.

103-114.

153. Li, S., Vert M., Biodegradation of aliphatic polyesters, in Degradable Polymers2002, Springer

Netherlands. p. 71-131.

154. Lucas, N., Bienaime C, Belloy C, Queneudec M, Silvestre F, Nava-Saucedo J E, Polymer

biodegradation: Mechanisms and estimation techniques – A review. Chemosphere, 2008.

73(4): p. 429-442.

155. Katarzyna, L., Lewandowicz G, Polymer biodegradation and biodegradable polymers – a review

Polish Journal of Environmental Studies, 2010. 19(2): p. 255 - 266.

156. Palmisano, A., C., Pettigrew C A, Biodegradability of plastics. BioScience, 1992. 42(9): p. 680 -

685.

157. Ho, K., L, G., Pometto A L, Effects of electron-beam irradiation and ultraviolet light (365 nm) on

polylactic acid plastic films. Journal of environmental polymer degradation, 1999. 7(2): p. 93-

100.

158. Sato, S., Ono M, Yamauchi J, Kanehashi S, Ito H, Matsumoto S, Iwai Y, Matsumoto H, Nagai K,

Effects of irradiation with vacuum ultraviolet xenon excimer lamp at 172 nm on water vapor

transport through poly(lactic acid) membranes. Desalination, 2012. 287: p. 290-300.

159. Tsuji, H., Echizen Y, Nishimura Y, Saha S K, Photodegradation of poly(L-lactic acid): Effects of

photosensitizer. Macromolecular Materials and Engineering, 2005. 290(12): p. 1192-1203.

160. Gorrasi, G., Pantani R, Effect of PLA grades and morphologies on hydrolytic degradation at

composting temperature: Assessment of structural modification and kinetic parameters.

Polymer Degradation and Stability, 2013. 98(5): p. 1006-1014.

161. Pan, P., Inoue, Y, Polymorphism and isomorphism in biodegradable polyesters. Progress in

Polymer Science, 2009. 34(7): p. 605-640.

162. Stloukal, P., Kalendova A, Mattausch H, Laske S, Holzer C, Koutny M, The influence of a

hydrolysis-inhibiting additive on the degradation and biodegradation of PLA and its

nanocomposites. Polymer Testing, 2015. 41: p. 124-132.

163. Rasmussen, K., Grampp G, Van Eesbeek M, Rohr T, Thermal and UV degradation of polymer

films studied in situ with ESR spectroscopy. ACS Appl. Mater. Interfaces, 2010. 2(Copyright (C)

164. Badía, J., D., Strömberg E, Ribes-Greus A, Karlsson S, Assessing the MALDI-TOF MS sample

preparation procedure to analyze the influence of thermo-oxidative ageing and thermo-

mechanical degradation on poly (Lactide). European Polymer Journal, 2011. 47(7): p. 1416-

1428.

165. Kopinke, F.D., Remmler, M., Mackenzie, K., Thermal decomposition of biodegradable

polyesters—II. Poly(lactic acid). Polymer Degradation and Stability, 1996. 53(3): p. 329-342.

166. McNeill, I., C., Leiper H A, Degradation studies of some polyesters and polycarbonates—2.

Polylactide: Degradation under isothermal conditions, thermal degradation mechanism and

photolysis of the polymer. Polymer Degradation and Stability, 1985. 11(4): p. 309-326.

167. Speranza, V., De Meo A, Pantani R, Thermal and hydrolytic degradation kinetics of PLA in the

molten state. Polymer Degradation and Stability, (0).

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

References

213

168. Kashiwagi, T., Shields JR, Harris RH, Davis RD, Flame-retardant mechanism of silica: Effects of

resin molecular weight. Journal of Applied Polymer Science, 2003. 87(9): p. 1541-1553.

169. Kashiwagi, T., Mu M, Winey K, Cipriano B, Raghavan S, Pack S, rafailovich M, Yang Y, Grulke E,

Shields J, Harris R, Douglas R, Relation between the viscoelastic and flammability properties of

polymer nanocomposites. Polymer, 2008. 49: p. 4358-4368.

170. Carpentier, F., Bourbigot S, Le Bras M, Delobel R, Rheological investigations in fire retardancy:

application to ethylene–vinyl-acetate copolymer–magnesium hydroxide/zinc borate

formulations. Polymer International, 2000. 49(10): p. 1216-1221.

171. Batistella, M., Otazaghine B, Sonnier R, Caro-Bretelle AS, Petter C, Lopez-Cuesta JM, Fire

retardancy of ethylene vinyl acetate/ultrafine kaolinite composites. Polymer Degradation and

Stability, 2014. 100: p. 54-62.

172. Chantegraille, D., Morlat-Therias S, Gardette, JL, Photochemical behaviour of fire-retarded

polymers. Polymer Degradation and Stability, 2010. 95(3): p. 274-277.

173. Lonkar, S., P., Therias S, Capera N, Leroux F, Gardette JL, Photo-oxidation of

polypropylene/layered double hydroxide nanocomposites: Influence of intralamellar cations.

European Polymer Journal, 2010. 46: p. 1456-1464.

174. Tidjani, A., Wilkie C A, Photo-oxidation of polymeric-inorganic nanocomposites: chemical,

thermal stability and fire retardancy investigations. Polymer Degradation and Stability, 2001.

74(1): p. 33-37.

175. Chen, D., Li J, Rena J, Combustion properties and transference behavior of ultrafine

microencapsulated ammonium polyphosphate in ramie fabric-reinforced poly(L-lactic acid)

biocomposites. Polymer International, 2011. 60: p. 599-606.

176. Almeras, X., Le Bras M, Hornsby P, Bourbigot S, Marosi G, Anna P, Delobel R, Artificial

weathering and recycling effect on intumescent polypropylene based blends. Journal of Fire

Sciences, 2004. 22(2): p. 143-161.

177. Larché, J., F., Gallot G, Boudiaf-Lomri L, Poulard C, Duemmler I, Meyer M, Evidence of surface

accumulation of fillers during the photo-oxidation of flame retardant ATH filled EVA used for

cable applications. Polymer Degradation and Stability, 2014. 103: p. 63-68.

178. El Hage, R., Viretto A, Sonnier R, Ferry L, Lopez-Cuesta JM, Flame retardancy of ethylene vinyl

acetate (EVA) using new aluminum-based fillers. Polymer Degradation and Stability, 2014. 108:

p. 56-67.

179. Wang, L., L., Wang Y C, Li G Q, Experimental study of hydrothermal aging effects on insulative

properties of intumescent coating for steel elements. Fire Safety Journal, 2013. 55(0): p. 168-

181.

180. ISO-4589-2:1996, Plastics -- Determination of burning behaviour by oxygen index -- Part 2:

Ambient-temperature test, 1996.

181. ISO-13927:2001, Plastics -- Simple heat release test using a conical radiant heater and

thermopile detector, 2001.

182. ISO-4892-3:2013, Methods of exposure to laboratory light sources -- Part 3: Fluorescent UV

lamps, 2013.

183. Ghoreishian, S., Badii K, Norouzi M, Malek K, Effect of cold plasma pre-treatment on

photocatalytic activity of 3D fabric loaded with nano-photocatalysts: Response surface

methodology. Applied Surface Science, 2016. 365: p. 252-262.

184. Colleoni, C., Massafra M R, Migani V, Rosace G, Dendrimer finishing influence on CO/PES

blended fabrics color assessment. Journal of Applied Polymer Science, 2011. 120(4): p. 2122-

2129.

185. Kowalski, A., Duda A, Penczek S, Polymerization of l,l-Lactide initiated by aluminum

isopropoxide trimer or tetramer. Macromolecules, 1998. 31(7): p. 2114-2122.

186. Foulc, M., P., Bergeret A, Ferry L, Ienny P, Crespy A, Study of hygrothermal ageing of glass fibre

reinforced PET composites. Polymer Degradation and Stability, 2005. 89(3): p. 461-470.

187. ISO-1133:2005, Plastics -- Determination of the melt mass-flow rate (MFR) and the melt

volume-flow rate (MVR) of thermoplastics, 2005.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

References

214

188. Feller, R.L., Accelerated ageing: Photochemical and thermal aspects1994: Marina del Rey, CA :

Getty Conservation Institute, 1994.

189. Keles, H., Naylor A, Clegg F, Sammon C, Investigation of factors influencing the hydrolytic

degradation of single PLGA microparticles. Polymer Degradation and Stability, 2015. 119: p.

228-241.

190. Manring, E., L., Blume R C, Simonsick J, Adelman D J, Thermal degradation of Poly( 2,2-dialkyl-

3-hydroxypropionic acid). 2. Thermal degradation initiated by random scission

Macromolecules, 1992. 25: p. 4863-4870.

191. Grassie, N., McNeill I C, Thermal degradation of polymethacrylonitrile. Part V. The mechanism

of the initiation step in coloration reactions. Journal of polymer science, 1959. 39(135): p. 211-

222.

192. de Oliveira, H., P., Rieumont J, Nogeuiras C, Souza D, Sanchez R, Thermal degradation behavior

of a carboxylic acid-terminated amphiphilic block copolymer poly(ethylene)-b-poly(ethylene

oxide). Journal of Thermal Analysis and Calorimetry, 2011. 103(2): p. 443-451.

193. Dobkowski, Z., Determination of critical molecular weight for entangled macromolecules using

the tensile strength data. Rheologica Acta, 1995. 34(6): p. 578-585.

194. Cho, K., S., Ahn K H, Lee S J, Simple method for determining the critical molecular weight from

the loss modulus. Journal of Polymer Science Part B: Polymer Physics, 2004. 42(14): p. 2724-

2729.

195. Lyu, S., Untereker D, Degradability of polymers for implantable biomedical devices.

International Journal of Molecular Sciences, 2009. 10(9): p. 4033-4065.

196. Dorgan, J., R.,, Rheology of Poly(Lactic Acid), in Poly(Lactic Acid)2010, John Wiley & Sons, Inc.

p. 125-139.

197. Inkinen, S., Hakkarainen M, Albertsson AC, Södergård A, From lactic acid to poly(lactic acid)

(PLA): Characterization and analysis of PLA and its precursors. Biomacromolecules, 2011.

12(3): p. 523-532.

198. Stloukal, P., Pekařová S, Kalendova A, Mattausch H, Laske S, Holzer C, Chitu L, Bodner S, Maier

G, Slouf M, Koutny M, Kinetics and mechanism of the biodegradation of PLA/clay

nanocomposites during thermophilic phase of composting process. Waste Management, 2015.

42: p. 31-40.

199. Rathi, S., Coughlin E, Hsu S, Golub C, Ling G, Tzivanis M, Maintaining structural stability of

poly(lactic acid): effects of multifunctional epoxy based reactive oligomers. Polymers, 2014.

6(4): p. 1232.

200. Liu, X., Zou Y, Li W, Cao G, Chen W, Kinetics of thermo-oxidative and thermal degradation of

poly(d,l-lactide) (PDLLA) at processing temperature. Polymer Degradation and Stability, 2006.

91(12): p. 3259-3265.

201. Meaurio, E., López-Rodríguez N, Sarasua J R, Infrared Spectrum of Poly(l-lactide): Application

to Crystallinity Studies. Macromolecules, 2006. 39(26): p. 9291-9301.

202. Zhang, J., Duan Y, Sato H, Tsuji H, Noda I, Yan S, Ozaki Y, Crystal modifications and thermal

behavior of poly(l-lactic acid) revealed by infrared spectroscopy. Macromolecules, 2005.

38(19): p. 8012-8021.

203. Zhang, J., Tashiro K, Tsuji H, Domb A J, Disorder-to-order phase transition and multiple melting

behavior of poly(l-lactide) investigated by simultaneous measurements of WAXD and DSC.

Macromolecules, 2008. 41(4): p. 1352-1357.

204. Fayolle, B., Richaud E, Colin X, Verdu J, Review: degradation-induced embrittlement in semi-

crystalline polymers having their amorphous phase in rubbery state. Journal of Materials

Science, 2008. 43(22): p. 6999-7012.

205. Torres, A., Li S M, Roussos S, Vert M, Degradation of L- and DL-lactic acid oligomers in the

presence of Fusarium moniliforme and Pseudomonas putida. Journal of environmental

polymer degradation, 1996. 4(4): p. 213-223.

206. McCain, D., C., Hayden D W, Theory of electron spin g-values for carbonyl radicals. Journal of

Magnetic Resonance (1969), 1973. 12(3): p. 312-330.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

References

215

207. Babanalbandi, A., Hill D J T, O'Donnell J H, Pomery P J and A. Whittaker, An electron spin

resonance study on γ-irradiated poly(l-lactic acid) and poly(d,l-lactic acid). Polymer

Degradation and Stability, 1995. 50(3): p. 297-304.

208. Laurent, B., Roskosz M, Remusat L, Leroux H, Vezin H, Depecker C, Isotopic and structural

signature of experimentally irradiated organic matter. Geochimica et Cosmochimica Acta,

2014. 142: p. 522-534.

209. Laurent, B., Roskosz M, Remusat L, Robert F, Leroux H, Vezin H, Depecker C, Nuns N, Lefebvre

JM, The deuterium/hydrogen distribution in chondritic organic matter attests to early ionizing

irradiation. Nat Commun, 2015. 6.

210. Codari, F., Lazzari S, Soos M, Storti G, Morbidelli M, Moscatelli D, Kinetics of the hydrolytic

degradation of poly(lactic acid). Polymer Degradation and Stability, 2012. 97(11): p. 2460-

2466.

211. Gupta, M., C., Deshmukh V G, Thermal oxidative degradation of poly-lactic acid. Colloid and

Polymer Science, 1982. 260(3): p. 308-311.

212. Bussiere, P.O., Etude des mécanismes de phototransformation des polymères au cours du

vieillissement photochimique oxydatif: de l'échelle moléculaire à l'échelle macroscopique, in

Institut de Chimie de Clérmont-Ferrand2012, Ecole nationalle Supérieure de Chimie de

Clermont-Ferrand. p. 164.

213. Cevdet, K., Burcu S, Accelerated weathering performance of polylactide and its montmorillonite

nanocomposite. Applied Clay Science, 2016. 121–122: p. 86-94.

214. Liu, X., Wang T, Chow LC, Yang M, Mitchell JW, Effects of inorganic fillers on the thermal and

mechanical properties of poly(lactic acid). International Journal of Polymer Science, 2014.

2014: p. 8.

215. Capone, C., Di-Landro L, Inzoli F, Penco M, Sartore L, Thermal and mechanical degradation

during polymer extrusion processing. Polymer Engineering & Science, 2007. 47(11): p. 1813-

1819.

216. El'darov, E., G., Mamedov F V, Gol'dberg V M, Zaikov G E, A kinetic model of polymer

degradation during extrusion. Polymer Degradation and Stability, 1996. 51(3): p. 271-279.

217. Pielichowski, K., Njuguna J, Thermal degradation of polymeric materials2005: Rapra

Technology Limited.

218. Vera-Sorroche, J., Kelly A L, Brown E C, Gough T, Abeykoon C, Coates P D, Deng J, Li K, Harkin-

Jones E, Price M, The effect of melt viscosity on thermal efficiency for single screw extrusion of

HDPE. Chemical Engineering Research and Design, 2014. 92(11): p. 2404-2412.

219. González-González, V., A., Neira-Velázquez G, Angulo-Sánchez J L, Polypropylene chain

scissions and molecular weight changes in multiple extrusion. Polymer Degradation and

Stability, 1998. 60(1): p. 33-42.

220. Goetze, K., P., Porter R S, Extrusion degradation of a high molecular weight polystyrene. Journal

of Polymer Science Part C: Polymer Symposia, 1971. 35(1): p. 189-200.

221. Hong, Y., Gao C, Xie Y, Gong Y, Shen J, Collagen-coated polylactide microspheres as chondrocyte

microcarriers. Biomaterials, 2005. 26: p. 6305-6313.

222. Rangelov, M., A., Petrova G P, Yomtova V M, Vayssilov G N, Catalytic role of vicinal OH in ester

aminolysis: proton shuttle versus hydrogen bond stabilization. The Journal of Organic

Chemistry, 2010. 75(20): p. 6782-6792.

223. Vakhitova, L.N., Taran N A, Lapushkin M P, Drizhd V L, Lakhtarenko N V, Popov A F, Solid-phase

aminolysis in the ammonium polyphosphate– pentaerythritol–amine system. Theoretical and

Experimental Chemistry, 2012. 48(3): p. 176-181.

224. Leszczyńskaa, A., Njugunab J, Pielichowskia K, Banerjeec JR, Polymer/montmorillonite

nanocomposites with improved thermal properties. Part II: Thermal stability of

montmorillonite nanocomposites based on different polymeric matrixes. Thermochimica Acta,

2007. 454(1): p. 1-22.

225. Zheng, X., Gilbert, M., An investigation into the thermal stability of PVC/montmorillonite

composites. Journal of Vinyl and Additive Technology, 2011. 17(2): p. 77-84.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

References

216

226. Isfahani, A.P., Mehrabzadeh M, Morshedian J, The effects of processing and using different

types of clay on the mechanical, thermal and rheological properties of high-impact polystyrene

nanocomposites. Polymer Journal, 2013. 45(3): p. 346-353.

227. Romanzini, D., Piroli V, Frache A, Zattera A J, Amico S C, Sodium montmorillonite modified with

methacryloxy and vinylsilanes: Influence of silylation on the morphology of clay/unsaturated

polyester nanocomposites. Applied Clay Science, 2015. 114: p. 550-557.

228. Lewitus, D., McCarthy S, Ophir A, Kenig S, The effect of nanoclays on the properties of PLLA-

modified polymers part 1: mechanical and thermal properties. Journal of Polymers and the

Environment, 2006. 14(2): p. 171-177.

229. Chieng, B., Ibrahim N, Yunus W, Hussein M, Poly(lactic acid)/Poly(ethylene glycol) polymer

nanocomposites: Effects of graphene nanoplatelets. Polymers, 2014. 6(1): p. 93.

230. Shi, X., Zhang G, Vu-Phuong T, Lazzeri A, Synergistic effects of nucleating agents and plasticizers

on the crystallization behavior of poly(lactic acid). Molecules, 2015. 20: p. 1579-1593.

231. Oudet, C., Polymères: structure et proprietés, introduction1994: Masson.

232. Blanchard, L., P., Hesse J, Lal-Malhotra S, Effect of Molecular Weight on Glass Transition by

Differential Scanning Calorimetry. Canadian Journal of Chemistry, 1974. 52(18): p. 3170-3175.

233. Tadlaoui, K., Pietrasanta Y, Michel A, Verney V, Influence of molecular weight on the glass

transition temperature and the melt rheological behaviour of methyl methacrylate telomers.

Polymer, 1991. 32(12): p. 2234-2237.

234. Omelczuk, M., O., McGinity J W, The influence of polymer glass transition temperature and

molecular weight on drug release from tablets containing poly(DL-lactic acid). Pharmaceutical

Research, 1992. 9(1): p. 26-32.

235. Narladkar, A., Balnois E, Vignaud G, Grohens Y, Bardeau J F, Morphology and glass transition

of thin polylactic acid films. Polymer Engineering & Science, 2008. 48(9): p. 1655-1660.

236. Montserrat, S., Colomer P, The effect of the molecular weight on the glass transition

temperature in amorphous poly(ethylene terephthalate). Polymer Bulletin, 1984. 12(2): p. 173-

180.

237. Fukushima, K., Abbate C, Tabuani D, Gennari M, Camino G, Biodegradation of poly(lactic acid)

and its nanocomposites. Polymer Degradation and Stability, 2009. 94(10): p. 1646-1655.

238. Sinha Ray, S., Yamada K, Okamoto M, Ueda K, New polylactide-layered silicate

nanocomposites. 2. Concurrent improvements of material properties, biodegradability and

melt rheology. Polymer, 2003. 44(3): p. 857-66.

239. Marosi, G., Anna P, Balogh I, Bertalan G, Tohl A, Maatoug M A, Thermoanalytical study of

nucleating effects in polypropylene composites. Journal of thermal analysis, 1997. 48(4): p.

717-726.

240. Krikorian, V., Pochan D J, Unusual crystallization behavior of organoclay reinforced poly(L-lactic

acid) nanocomposites. Macromolecules, 2004. 37: p. 6480-6491.

241. Qian, J., Zhu L, Zhang J, Whitehouse R S, Comparison of different nucleating agents on

crystallization of poly(3-hydroxybutyrate-co-3-hydroxyvalerates). Journal of Polymer Science

Part B: Polymer Physics, 2007. 45(13): p. 1564-1577.

242. Ujhelyiova, A., Slobodova M, Ryba J, Borsig E, Vencelova P, The effect of inorganic

nanoadditives on the thermal, mechanical and UV radiation barrier properties of polypropylene

fibres. Open Journal of Organic Polymer Materials, 2012. 2: p. 9.

243. Gopalan, M., R., Mandelkern L, Effect of crystallization temperature and molecular weight on

the melting temperature of linear polyethylene. The Journal of Physical Chemistry, 1967.

71(12): p. 3833-3841.

244. Stack, G., M., Mandelkern L, Voigt-Martin I G, Crystallization, melting, and morphology of low

molecular weight polyethylene fractions. Macromolecules, 1984. 17(3): p. 321-331.

245. Tin-Sin, L., Rahmat A R, Rahman W A W A, Polylactic acid: PLA biopolymer technology and

applications. PDL handbook series2012: Elsevier. 352.

246. Yu, D., L., He J L, Liu Z Y, Xu B, Li D C, Tian Y J, Phase transformation of melamine at high pressure

and temperature. Journal of Materials Science, 2008. 43(2): p. 689-695.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

References

217

247. Liao, Y., Zhu S, Ma J, Sun Z, Yin C, Zhu C, Lou X, Zhang D, Tailoring the morphology of g-C3N4

by self-assembly towards high photocatalytic performance. ChemCatChem, 2014. 6(12): p.

3419-3425.

248. Levchik, G., F., Levchik S V, Sachok P D, Selevich A F, Lyakhov A S, Lesnikovich A I, Thermal

behaviour of ammonium polyphosphate-inorganic compound mixtures. Part 2. Manganese

dioxide. Thermochimica Acta, 1995. 257: p. 117-125.

249. Uma, T., Tu H Y, Warth S, Synthesis and characterization of proton conducting poly phosphate

composites. Journal of Materials Science and Engineering, 2005. 40: p. 2059-2063.

250. Sherief, M., A., Hanna, A A, Abdelmoaty A S, Synthesis and characterization of nanosized

ammonium polyphosphate. Canadian Journal of Applied Sciences, 2014. 3(4): p. 94-99.

251. Sahoo, D., Nayak P L, Synthesis and characterization of chitosan/cloisite 30B film for controlled

release of ofloxacin. Journal of Applied Polymer Science, 2012. 123(5): p. 2588-2594.

252. Wang, Y., L., Mebel A M, Chung-Jen W, Chen Y T, Ching-Erh l, Jyh-Chiang j, IR spectroscopy and

theoretical vibrational calculation of the melamine molecule. Journal of the Chemical Society,

Faraday Transactions, 1997. 93(19): p. 3445-3451.

253. Mircescu, N., E., Oltean M, Chiş V, Leopold N, FTIR, FT-Raman, SERS and DFT study on

melamine. Vibrational Spectroscopy, 2012. 62: p. 165-171.

254. Kokotou, M., Thomaidis N, Behavior and retention models of melamine and its hydrolysis

products. Chromatographia, 2012. 75(9): p. 457-467.

255. Bernhard, A., Peitz D, Elsener M, Wokaun A, Kröcher O, Hydrolysis and thermolysis of urea and

its decomposition byproducts biuret, cyanuric acid and melamine over anatase TiO2. Applied

Catalysis B: Environmental, 2012. 115–116: p. 129-137.

256. Gong, H., Tang S, Zhang T, Catalytic hydrolysis of waste residue from the melamine process and

the kinetics of melamine hydrolysis in NaOH solution. Reaction Kinetics, Mechanisms and

Catalysis, 2016. 118(2): p. 377-391.

257. Venugopalan, M., V., Prasad R, Hydrolysis of ammonium polyphosphate in soils under aerobic

and anaerobic conditions. Biology and Fertility of Soils, 1989. 8(4): p. 325-327.

258. Jimenez, M., Bellayer S, Revel B, Duquesne S, Bourbigot S, Comprehensive study of the

influence of different aging scenarios on the fire protective behavior of an epoxy based

intumescent coating. Ind. Eng. Chem. Res., 2013. 52: p. 729-743.

259. Arai, Y., Sparks D L, ATR–FTIR spectroscopic investigation on phosphate adsorption mechanisms

at the ferrihydrite–water interface. Journal of Colloid and Interface Science, 2001. 241(2): p.

317-326.

260. Rudolph, W., W.,, Raman- and infrared-spectroscopic investigations of dilute aqueous

phosphoric acid solutions. Dalton Transactions, 2010. 39(40): p. 9642-9653.

261. Lizheng, S., Kefu C, Preparation and characterization of ammonium polyphosphate/diatomite

composite fillers and assessment of their flame retardant effects on paper. BioResources, 2014.

9(2): p. 3104-3116.

262. Shao, Z., Deng C, Tan Y, Yu L, Chen M J, Chen L, Wang Y Z, Ammonium polyphosphate

chemically-modified with ethanolamine as an efficient intumescent flame retardant for

polypropylene. Journal of Materials Chemistry A, 2014. 2(34): p. 13955-13965.

263. Gérard, C., Fontaine G, Bellayer S, Bourbigot S, Reaction to fire of an intumescent epoxy resin:

Protection mechanisms and synergy. Polymer Degradation and Stability, 2012. 97(8): p. 1366-

1386.

264. Hadj-Hamou, A., S., Matassi S, Abderrahmane H, Yahiaoui F, Effect of cloisite 30B on the

thermal and tensile behavior of poly(butylene adipate-co-terephthalate)/poly(vinyl chloride)

nanoblends. Polymer Bulletin, 2014. 71(6): p. 1483-1503.

265. Debasish, S., Nayak P L, Synthesis and characterization of chitosan/cloisite 30B film for

controlled release of ofloxacin. Journal of Applied Polymer Science, 2012. 123(5): p. 2588-2594.

266. Martos, J., Osinaga P W R, De-Olivera E, Suita-de-Castro L A, Hydrolytic degradation of

composite resins: Effects on the microhardness. Materials Research, 2003. 6(4): p. 599-604.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

References

218

267. Braun, U., Wachtendorf V, Geburtig A, Bahr H, Schartel B, Weathering resistance of halogen-

free flame retardance in thermoplastics. Polymer Degradation and Stability, 2010. 95(12): p.

2421-2429.

268. Jahromi, S., Moosheimer U, Oxygen barrier coatings based on supramolecular assembly of

melamine. Macromolecules, 2000. 33(20): p. 7582-7587.

269. Purser, D., Fire retardant materials. 3 - Toxicity of fire retardants in relation to life safety and

environmental hazards, in Fire Retardant Materials2001, Woodhead Publishing. p. 69-127.

270. Lahtela, V., Kärki T, Improving the UV and water-resistance properties of Scots pine (Pinus

sylvestris) with impregnation modifiers. European Journal of Wood and Wood Products, 2014.

72(4): p. 445-452.

271. Gérard, C., Contribution of nanoparticles to the flame retardancy of epoxy resins, 2011,

Université des sciences et technologies de Lille.

272. Sanches, N., B., Dias M L, Pacheco E B A V, Comparative techniques for molecular weight

evaluation of poly (ethylene terephthalate) (PET). Polymer Testing, 2005. 24(6): p. 688-693.

273. Schartel, B., Bartholmai M, Knoll U, Some comments on the main fire retardancy mechanisms

in polymer nanocomposites. Polymers for Advanced Technologies, 2006. 17(9-10): p. 772-777.

274. Matzen, M., Kandola B, Huth C, Schartel B, Influence of flame retardants on the melt dripping

behaviour of thermoplastic polymers. Materials, 2015. 8: p. 5621-5646.

275. Dealy, J., M., Wissbrun K F, Role of rheology in extrusion, in Melt Rheology and Its Role in

Plastics Processing: Theory and Applications1999, Springer Netherlands: Dordrecht. p. 441-

490.

276. Mishra, A., K., Mushtaq S, Nando G B, Chattopadhyay S, Effect of Cloisite and modified Laponite

clays on the rheological behavior of TPU–clay nanocomposites. Rheologica Acta, 2010. 49(8):

p. 865-878.

277. Bashir, M., A.,, Effect of nanoclay dispersion on the processing of polyester nanocomposites, in

Department of Mechanical Engineering2008, McGill University, Montreal, QC, Canada. p. 118.

278. Huang, X., Lewis S, Brittain W J, Vaia R A., Synthesis of polycarbonate-layered silicate

nanocomposites via cyclic oligomers. Macromolecules, 2000. 33(6): p. 2000-2004.

279. Ishida, H., Campbell S, Blackwell J, General approach to nanocomposite preparation. Chemistry

of Materials, 2000. 12(5): p. 1260-1267.

280. Hyun, Y., H., Lim S T, Choi H J, Jhon M S, Rheology of poly(ethylene oxide)/organoclay

nanocomposites. Macromolecules, 2001. 34(23): p. 8084-8093.

281. Vaia, R., A., Ishii H, Giannelis E P, Synthesis and properties of two-dimensional nanostructures

by direct intercalation of polymer melts in layered silicates. Chemistry of Materials, 1993. 5(12):

p. 1694-1696.

282. Camino, G., Costa L, Trossarelli L, Study of the mechanism of intumescence in fire retardant

polymers. V. Mechanism of formation of gaseous products in the thermal degradation of

ammonium polyphosphate. Polymer Degradation and Stability, 1985. 12: p. 203-212.

283. Bellucci, F., Camino G, Frache A, Sarra A, Catalytic charring–volatilization competition in

organoclay nanocomposites. Polymer Degradation and Stability, 2007. 92(3): p. 425-436.

284. Xie, W., Gao Z, Pan W P, Hunter D, Singh A, Vaia R, Thermal degradation chemistry of alkyl

quaternary ammonium montmorillonite. Chemistry of Materials, 2001. 13(9): p. 2979-2990.

285. Xie, W., Xie R, Pan W P, Hunter D, Koene B, Tan L S, Vaia R, Thermal stability of quaternary

phosphonium modified montmorillonites. Chemistry of Materials, 2002. 14(11): p. 4837-4845.

286. Cervantes-Uc, J., M., Cauich-Rodríguez J V, Vázquez-Torres H, Garfias-Mesías L F, Paul D R,

Thermal degradation of commercially available organoclays studied by TGA–FTIR.

Thermochimica Acta, 2007. 457(1–2): p. 92-102.

287. Solarski-Samyn, F., Compréhension des procédés d’ignifugation du polyamide 6 : Apport des

nanocomposites aux systèmes retardateurs de flamme phosphorés, 2007, Université des

sciences et technologies de Lille.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Appendix

219

Appendix

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Appendix

220

Appendix 1: Quantification of color change for FR-PLAs.

To quantify the color change observed when FR fillers are incorporated into the PLA

matrix, L*a*b* measurements (Table 44) were performed on all samples (i.e. PLA, FR-PLA and

FR-PLA-C30B). L* corresponds to color position between black and white, a* the color position

between red/magenta and green and b* the color position between yellow and blue. ΔE

corresponds to the differences between PLA and FR-PLAs, calculated according to Equation

25 (cf. chapter II, p.76).

Table 44: L*a*b* values of PLA, FR-PLA and FR-PLA-C30B.

L*a*b* values

PLA

FR-PLA

FR-PLA-C30B

-2

0.5

2.5

-1

∆E

Data evidence the color change of PLA when fillers are incorporated into it. L* increases

from 52 for PLA to 83 for FR-PLA and 76 for FR-PLA-C30B. a* increases by 2.5 and 4.5 for FR-

PLA and FR-PLA-C30B, respectively, compare to neat PLA. Finally, b* increases by 13 for both

FR-PLA and FR-PLA-C30B compare to neat PLA. ΔE reaches 35 and 28 for FR-PLA and FR-PLA-

C30B, respectively. These results demonstrate that when FR fillers are incorporated the color

of PLA material changes: the high L* value of both FR-PLAs specifies that the materials tend to

whiter color and the positive value of b* for both FR materials indicates a tendency to

yellowing. The combination between L* and b* values evidence the creamy color observed

when FR are incorporated into PLA. a* cannot be taken into account due to the value close to

zero.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Appendix

221

Appendix 2: Color changes of FR-PLA and FR-PLA-C30B after ageing

It was demonstrated that when FR-PLA and FR-PLA-C30B are submitted to environmental

constraints (i.e. T/UV, T/RH and T/UV/RH conditions), some phenomena are observed on the

surface of the materials. Indeed, a laundering of plates is reported when samples are aged

under T/UV whereas under T/RH and T/UV/TH, the appearance of a white layer at the surface

of the samples is evidenced. Thus L*a*b* measurements were performed in order to quantify

the color changes of FR-PLAs before and after ageing (Table 45 and Table 46 for FR-PLA and

FR-PLA-C30B, respectively).

Table 45: L*a*b* values of unaged and aged FR-PLA under T/UV, T/RH and T/UV/RH conditions.

L*a*b* values

FR-PLA

FR-PLA T/UV 60 d

FR-PLA T/RH 28 d

FR-PLA T/UV/RH 6 d

86 (+3)

89 (+6)

95 (+12)

0.5

0.3 (-0.2)

0.2 (-0.3)

0.01 (-0.4)

9 (-3)

8 (-4)

4 (-8)

∆E

Table 46: L*a*b* values of unaged and aged FR-PLA-C30B under T/UV, T/RH and T/UV/RH

conditions.

L*a*b* values

FR-PLA-C30B

T/UV 60 d

FR-PLA-C30B

T/RH 28 d

FR-PLA –C30B

T/UV/RH 6 d

80 (+4)

84 (+8)

93 (+17)

2.5

2 (-0.5)

1.9 (-0.6)

11 (-1)

6.5 (-5.5)

∆E

Data obtained by L*a*b* display that when FR-PLA and FR-PLA-C30B are aged, the color of

the materials changes. Regarding at first the FR-PLA formulation, ∆E is 4 after 60 days exposure

to T/UV and reaches 7 and 15 after 28 days and 6 days exposure to T/RH and T/UV/RH,

respectively. For FR-PLA-C30B, ∆E is 4 after 60 days exposure to T/UV and reaches 8 and 18

after 28 days and 6 days exposure to T/RH and T/UV/RH, respectively. The increase in the ∆E

demonstrates the alteration of colors during ageing. Moreover, for both materials, it appears

that L* increases as function of ageing time and conditions. This shows that materials become

whiter after ageing and corroborates the bleaching phenomenon visually observed.

Concerning a*, it appears that the values does not present significant changes for both FR-

formulations. In the case of b*, the values are decreased as function of ageing time and kind

of exposure. Indeed, the higher decrease of b* for FR-PLA and FR-PLA-C30B is observed when

materials are aged under T/UV/RH: b* is decreased by 8 for FR-PLA and by 5.5 for FR-PLA-

C30B after 6 days exposure. The negative tendency value of b* for both FR materials indicates

a propensity for the blue color. For a*, the close zero value designate neutral colors. L*a*b*

results clearly prove the bleaching observed when FR-materials are aged.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Appendix

222

Appendix 3: Effect of the type of nanoparticle on the hydrolysis of FR-PLAs

In order to determine if the hydrophilic character of clays plays a role during the hydrolytic

degradation of flame retarded PLA, three kinds of Cloisite, i.e. 20A, 30B and Na

(hydrophilic

character of CNa

> C30B > C20A) were incorporated into FR-PLA materials and then aged for

60 days under T/RH conditions.

The formation of voids, fractures and of a white layer is reported (Figure 114) after 28 days

of ageing under T/RH exposure for the three materials (i.e. FR-PLA-C20A, FR-PLA-C30B and FR-

PLA-CNa

Figure 114: FR-PLA-C20A (a), FR-PLA-C30B (b) and FR-PLA-CNa

T/RH conditions.

Physico-chemical properties of the materials were also studied all along the exposure to

T/RH, i.e. the molecular mass (Figure 115), T

(Figure 116), T

(Figure 117) and crystallinity

(Figure 118).

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Appendix

223

Figure 115: M

versus ageing time of FR-PLA-C20A (●), FR-PLA-C30B (▪) and FR-PLA-C30B (Δ)

samples aged under T/RH conditions.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Appendix

224

Figure 116: Evolution of glass transition temperature (T

) of FR-PLA-C20A (●), FR-PLA-C30B

(▪) and FR-PLA-CNa

(Δ) as a function of exposure time under T/RH conditions.

Figure 117: Evolution of melting temperature (T

) of FR-PLA-C20A (●), FR-PLA-C30B (▪) and

FR-PLA-CNa

(Δ) as a function of exposure time under T/RH conditions.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Appendix

225

Figure 118: Evolution of crystallinity of FR-PLA-C20A (●), FR-PLA-C30B (▪) and FR-PLA-CNa

(∆) as a function of ageing time under T/RH conditions.

Before ageing, M

, T

and T

are higher with a lower crystallinity value for FR-PLA-C20A

compared to FR-PLA-C30B then FR-PLA-CNa

. It appears that the molecular mass, T

, T

and

crystallinity of the three materials exhibit the same behavior during ageing, whatever the

formulation. Indeed, M

, T

and T

of the three formulations decrease with increasing ageing

exposure to T/RH. Crystallinity increases when ageing duration increase. Moreover, after

ageing, M

, T

and T

are still higher with a lower crystallinity value for FR-PLA-C20A compared

to FR-PLA-C30B then FR-PLA-CNa

. This seem to indicate that the hydrophilic character of

organoclays does not play a particular role during the degradation of the PLA matrix compared

to the M

. Indeed, the lower the M

before ageing, the higher the impact of ageing on T

, T

and crystallinity.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Appendix

226

Appendix 4: FTIR (ATR) spectra of FR-PLAs aged under T/RH and T/UV/RH

exposure.

The behavior of FR fillers during ageing is confirmed by FTIR spectra recorded for both FR-

PLAs during exposure to T/RH and T/UV/RH conditions (Figure 119 and Figure 120). Indeed,

characteristic bands of PLA (cf. Table 1, p.29), Melamine and APP (previously reported p.144)

can be identified on the spectrum of unaged FR-PLA and FR-PLA-C30B. Furthermore, similar

spectra are observed for both FR-PLAs after 35 days exposure to T/RH and 10 days exposure

to T/UV/RH conditions (just before disintegration into powder) with no appearance of new

peaks. As characteristic bands of Cloisite 30B are not observed in the FR-PLA-C30B spectra due

to the relatively low loading of clays (i.e. 1 wt.-%), both FR-PLAs exhibit the same spectra.

Figure 119: FTIR (ATR) spectra of PLA, FR-PLA, FR-PLA 105 days (T/RH) and FR-PLA 90 days

(T/UV/RH) in the range of 600 t0 3600 cm

-1

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Appendix

227

Figure 120: FTIR (ATR) spectra of PLA, FR-PLA-C30B, FR-PLA-C30B 105 days (T/RH) and FR-

PLA-C30B 90 days (T/UV/RH) in the range of 600 t0 3600 cm

-1

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Appendix

228

Appendix 5: 2D HYSCORE principle

The 2D HYperfine Sub-level CORrElation (2D HYSCORE) is essentially a two dimentional

electron spin-echo envelope modulation (ESEEM) experiment in which correlation is

tranferred from one electron spin manifold to another. A sequence of four microwave pulses

is applied to the sample and the stimulated spin-echo produced is measured (Figure 121)

Figure 121: Example of 2D HYSCORE microwave pulses.

In a HYSCORE experiment, the time between the second π/2 and π pulse is varied in one

dimension and the time between the π and third π/2 pulse is varied in a second dimension. A

two dimensional Fourier transform gives the sample data below (Figure 122).

Figure 122: 2D HYSCORE spectrum of a coat standard sample.

As in an ESEEM experiment, the peaks observed are essentially an NMR spectrum of nuclei

that are coupled to the electron. HYSCORE allows one to take a complicated ESEEM spectrum

and extend the data into a second dimension. Peaks appearing in the upper right and lower

left quadrants typically arise from nuclei in which the hyperfine coupling is less than the larmor

frequency. They appear at the larmor frequency, separated by the hyperfine coupling. Peaks

from nuclei in which the hyperfine interaction is greater than the larmor frequency appear in

the upper left and lower right quadrants. The HYSCORE spectrum can get very complicated for

nuclei with a spin of I > 1/2 with many peaks arising from the complication of additional

nuclear zeeman and quadrupole levels. Even with the complexity of the spectra, HYSCORE on

systems with multiple nuclei can make ESEEM spectra that would be difficult or impossible to

interpret much more manageable.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Appendix

229

Appendix 6: Thermal stability of pure compounds

The TGA curves of the pure polymer (PLA) and additives (Melamine, APP and Cloisite 30B)

are depicted in Figure 123. PLA starts to decompose in one step around 280°C with no residue

left at the end of the experiment, i.e. 800°C. APP decomposes in two steps, the first one

between 250°C and 450°C corresponding to 14% weight loss due to water and ammonia

release [282], leading to the formation of phosphoric acid. The second weight loss between

450°C and 700°C is linked to a 65% weight loss attributed to the sublimation of P

-H

O [12,

282] with 13 wt.-% residue left at the end of the experiment. Melamine begins to decompose

at 240°C, in one step with no residue at the end of the experiment, i.e. 800°C. It results in the

formation of melam (stable up to 400°C), melem (stable up to 500°C) and melon (stable up to

600°C) [61]. The condensation of Melamine is accompanied by a release of ammonia [12, 61].

Figure 123: TGA curves of pure compounds, PLA (●), Mel (●), APP (●) and C30B (●) under air

flow at 10°C/min.

Decomposition of C30B is well known and documented [283-285]. According to literature,

this organoclay can decompose in 4 steps. These four steps of degradation can be retrieved in

Figure 123 and in particular in Figure 124 which shows the rate of degradation (DTG) of pure

compounds.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Appendix

230

Figure 124: DTG of pure compounds, PLA (●), Mel (●), APP (●) and C30B (●) under air flow at

10°C/min.

The first step starts before 200°C and is linked to the desorption of water, oxygen and

nitrogen absorbed in C30B. The second step (200°C to 450°C) results from the thermal

degradation of the surfactant of Cloisite 30B. This second step is visible on the TGA curve in

Figure 123 (15% weight loss) and is normally constituted of two consecutive stages as

observed in the DTG curve (Figure 124). From 200°C to 320°C, C30B releases water, carbon

dioxide, alkane, alkene, aldehyde, carboxylic acid and amine. From 320°C to 450°C, water,

alkane, alkene and alcohol are released [286]. From 450°C to 800°C, the 15% weight loss

corresponds to the dehydroxylation of the aluminosilicate structure and to the release of

degraded products of the surfactant. Researches in our lab have already confirmed this

degradation process [287].

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

VIEILLISSEMENT D’ACIDE POLYLACTIQUE IGNIFUGÉ

RÉSUMÉ - Le but de ces recherches est d’étudier le vieillissement de l’acide polylactique (PLA)

ignifugé. Trois vieillissement différents ont été étudiés i.e. température/ultra-violet (T/UV),

température/humidité relative (T/RH) et température/ultra-violet/humidité relative (T/UV/RH). Afin

de comprendre le rôle des retardateurs de flamme lors du vieillissement, l’étude de la dégradation du

PLA vierge à tout d’abord été mené. Des méthodologies innovantes ont été développées et les

propriétés physico-chimiques du polymère ont été caractérisées en fonction de la durée et du type

d’exposition. Les mécanismes de dégradation dépendants du vieillissement (T/UV, T/RH et T/UV/RH)

ont été élucidés et comparés à la littérature. L’influence des retardateurs de flamme (i.e. Mélamine,

Polyphosphate d’ammonium et Cloisite 30B) lors du vieillissement des formulations ignifugées a

ensuite été étudiée. Il a été prouvé que ces additifs ont une influence directe sur le vieillissement, au

niveau des propriétés physico-chimiques et des mécanismes de dégradation. La masse molaire s’est

révélée être cruciale dans la mesure où elle gouverne l’évolution des propriétés physico-chimiques du

matériau durant le vieillissement, et donc sa durabilité. Sachant que le PLA est ignifugé par

incorporation d’additifs retardateurs de flamme, il était primordial de comprendre l’effet du

vieillissement sur ces additifs et donc leurs effets sur les propriétés feu du PLA. Il a ainsi été montré

que les performances feu de ce matériau sont améliorées au cours du vieillissement, jusqu’à sa

dégradation complète. Ces performances ont été corrélées avec l’évolution des propriétés physico-

chimiques survenant au cours du vieillissement, présentant un rôle prépondérant sur la cinétique

d’intumescence.

AGEING OF FLAME RETARDED POLYLACTIC ACID

ABSTRACT - This work deals with the ageing of flame retarded (FR) Polylactic acid (PLA). The impact

of three accelerated ageing conditions i.e. temperature/ultra-violet (T/UV), temperature/relative

humidity (T/RH) and temperature/ultra-violet/relative humidity (T/UV/RH) was studied. In order to

understand the role of fire retardant additives on ageing of flame retarded PLAs, the first study was

focused on neat PLA. Innovative methodologies were developed, the change in physico-chemical

properties of the polymer was characterized as a function of ageing exposure and ageing duration.

Moreover, the mechanisms of degradation occurring during T/UV, T/RH and T/UV/RH exposure were

elucidated and compared to the literature. Then, the influence of flame retardants (i.e. Melamine,

Ammonium polyphosphate and Cloisite 30B) on the ageing behavior of FR-PLAs was investigated. FR

fillers were evidenced to have a direct influence on physico-chemical properties and mechanisms of

degradation of the material during ageing. The molecular mass was reported to be a crucial parameter,

as it is related to the physico-chemical properties and thus to the durability of the material. The main

goal of flame retardants is to improve the flammability of PLA thus the effect of ageing on the fire

properties of PLA was determined. It is noteworthy that the fire properties of flame retarded PLA are

improved during ageing, until the complete degradation of the materials. These surprising

performances were found to be correlated to the change in physico-chemical properties which play a

key role on the kinetics of intumescence.

Thèse de Nicolas Lesaffre, Lille 1, 2016

lilliad.univ-lille.fr

Download
Save PDF to desktop

Email
Email completed PDF

Send for Signing
Email for others to sign

Print
Print document

NAME

Ageing of Flame Retarded Polylactic Acid

View Audit Log
Print With Audit Log
Duplicate Document

Fill Online, Printable, Fillable, Blank Ageing of Flame Retarded Polylactic Acid Form

Related forms

Fillable Ageing of Flame Retarded Polylactic Acid

Fill Online, Printable, Fillable, Blank Ageing of Flame Retarded Polylactic Acid Form

Related forms

First 20 documents completely FREE!