Accelerated Ageing of Alkali Treated Olive Husk Flour Reinforced Polylactic Acid (PLA) Biocomposites: Physico-Mechanical Properties

Abstract

In this study, olive husk flour was added to poly(lactic acid) (PLA) to produce fully biosourced and biodegradable composites. In particular, untreated and alkali treated particles were used to produce the biocomposites at 20 wt.% via melt extrusion followed by injection moulding. The samples were then subjected to accelerated ageing (UV irradiation and water spray at 50°C) for different amounts of time (120, 240, 360 and 480 h). The results show that accelerated ageing decreased the tensile strength (TS) and Young's modulus (YM) for both untreated and alkali treated biocomposites, but the treated particles presented a lower reduction. Further comparison was made via differential scanning calorimetry (DSC) and Fourier transform infrared spectroscopy (FTIR) to detect any changes in the samples.

Keywords

Olive husk flour Poly(lactic acid)Accelerated ageing Mechanical properties

1. Introduction

Polymers based on petrochemicals resources are facing several issues related to oil price (raw material costs) and availability (limited supplies). This is why biobased polymers were developed because they are more sustainable, especially in terms of biodegradation and carbon life cycle analysis. But as their petrosourced counterparts, biosourced polymers properties can be improved by adding different types of particles. To keep their green origin, composites based on natural fibres attracted the attention of both industrial and academic research. This led to the development of biocomposites having properties similar to usual ones made from inorganic fillers^1,2. This is even more interesting since these particles are locally available (agro-forestry residues), low cost, low density and easily modified. So far, several types of natural fibres were introduced into different matrices to produce fully biodegradable materials^3,4. These materials are expected to have a positive global effect by using renewable resources, but they are new with limited applications and their long terms properties (stability) and degradation rates are still not fully understood.

One of the most known biopolymer having a commercial success is polylactic acid (PLA). Despite having similar mechanical properties than industrial polyolefins, PLA sensitivity to atmospheric conditions such as UV irradiation (sunlight), humidity and temperature limits its use for outdoor applications. Thus, the weathering behaviour of PLA and its biocomposites should be more studied to extend their application.

Olive husk, one of several lignocellulosic materials, is an agricultural residue produced as by-products of the olive milling process in olive-producing countries such as Algeria. Olive husk is commonly used as a biofuel in the form of pellets due to its high combustion energy or simply burned causing major environmental concerns. For this reason, this study was undertaken to find other applications for this waste by developing biocomposites based on poly(lactic acid).

The resistance of natural fibre composites upon UV light exposure is of particular concern since polymer surface chemistry changes, commonly known as photodegradation, occurs^5,6. This degradation ranges from slight surface discoloration affecting the aesthetic appeal for indoor applications to extensive loss of mechanical properties for outdoor applications^7–9. Furthermore, the combined effect of light, humidity, and temperature can totally destroy lignocellulosic networks limiting the performances of unprotected materials such as wood, especially for outdoor applications⁹.

The photodegradation of natural fibres is attributed to the degradation of its components namely cellulose, hemicelluloses, and lignins¹⁰. It is also known that lignins and hemicelluloses are more inclined to degradation than cellulose¹¹. But lignins degrade upon UV-light exposure, while hemicelluloses degrade upon moisture absorption and biological means¹². The UV degradation process is known to start by the formation of free radicals and possibly with the oxidation of phenolic-hydroxyl groups¹³. Moreover, singlet oxygen developed by oxygen quenching of photoexcited lignins plays a significant role in lignocellulosic natural fibres degradation¹³. The formed singlet oxygen is a peroxides source¹⁴, which can initiate carbohydrate auto-oxidation and lignins cleavage^15,16. In PLA, active oxygen species can initiate the degradation reaction by attacking neighbouring chains and the degradation process extends into the polymer through the diffusion of these reactive oxygen species¹⁷.

It is well known that lignocellulosic materials surface treatments create some structural modifications influencing interfacial bonding with the matrix. An alkali treatment is the most commonly used chemical modification to decrease the fibres hydrophilic nature, thus improving their compatibility with most matrices⁵. On the other hand, some other changes on the lignocellulosic fibres surface (morphological, chemical and physical) can also be obtained by alkali treatments.

Similar as for lignocellulosic materials, organic polymers degrade when exposed to UV light in the presence of air, and a question arises whether filler addition accelerates or not the oxidative processes. So composite degradation is not only important in terms of material's resistance within its useful lifetime, but is also paramount for post-consumer uses as it determines the possibility of recycling or reusing these materials.

Based on the literature, the degradation of PLA based biocomposites mainly consists of hydrothermal degradation and photo-degradation¹⁸. In hydrothermal degradation, the PLA-natural fibres interface is one of the main factors affecting the hydrothermal degradation rate¹⁹. Associated with natural fibres water absorption and swelling properties, the fibre-matrix interface is the main parameter controlling the hydrothermal degradation process via water molecules diffusion²⁰. Combined with photo-degradation, a complex degradation process can be obtained, especially taking into account the reflectivity and deep penetration of UV radiation as the main factors affecting the photo-degradation rate of biocomposites.

So based on the limited amount of information available in the literature, the main objective of this work is to study the effect of different degradation (UV, humidity, temperature) on a biocomposites. In particular, olive husk flour, with or without alkali surface treatment, was introduced into PLA to determine its effects on mechanical properties and degradation rates.

2. Experimental Procedure

2.1 Materials

The matrix used was PLA grade 2003D in the form of pellets and supplied by NatureWorks LLC (USA). NaOH (Sigma-Aldrich, USA) was used for the surface treatment. The olive husk flour (OHF) used in this study was supplied by local olive manufacturers (Algeria) after the olive oil extraction process. The material was sieved to keep only the particles having a diameter less than 100 μm.

2.2 Methods

2.2.1 Alkali Fibre Treatment

The OHF was treated with 5% NaOH for one hour. The particles were then washed several times with distilled water containing 1% acetic acid to clean their surface and neutralise the sodium hydroxide excess. Finally, they were washed with distilled water until neutral pH to be dried in ambient air for 12 h and then in an oven at 80°C for 6 h.

2.2.2 Biocomposites Processing

Untreated and alkali treated biocomposites with 20 wt.% of particles were prepared by extrusion/injection moulding.

According to the PLA 2003D data sheet, the pellets were dried for 12 h at 50°C under vacuum before being extruded and injection moulded.

A twin-screw extruder (Haake Rheomex OS PTW16) was used to perform melt compounding. A uniform temperature (180°C) and a constant screw rotation speed (50 rpm) were used. After extrusion, the material was cooled in a water bath and cut into pellets. The samples were then injection moulded on a Nissei model PS60E9ASE machine. The injection temperature was set at 190°C with a mould temperature at 40°C. The specimens were directly moulded to perform the mechanical characterisation.

2.2.3 Accelerated Ageing

Accelerated ageing was carried out using an accelerated weathering tester (QUV by Q-LAB) following ASTM G154. A fluorescent bulb (UVA) with a 0.68 W/m² irradiance (at 340 nm) was used with cycles of 1 h UV irradiation, followed by 1 min of spray with de-ionised water and a subsequent 2 h condensation, while maintaining a temperature of 50°C. Five specimens from each batch of tensile, impact and morphological testing were subjected to the ageing process for different amounts of time (120, 240, 360 and 480 h).

3. Characterisation

3.1 Tensile Properties

Tensile tests were carried out using a mechanical tester model 5565 (Instron, USA) with a 500 N load cell. Type V dog bone samples according to ASTM D638 were used to perform the tests at room temperature with a rate of 10 mm/min. For each material, five specimens were tested to get an average and standard deviation.

3.2 Differential Scanning Calorimetry (DSC)

Thermal characterisations using DSC were performed using a DSC7 from Perkin Elmer. A heat/cool/heat procedure was applied over a temperature range from 40 to 250°C at 10°C/min under a nitrogen atmosphere (N₂). The glass transition temperature (Tg), crystallisation temperature (Tc), and melting temperature (Tm) were determined, while the melting degree of crystallinity (x) was determined as:

x = \frac{Δ H m}{Δ H ° m} \times \frac{100}{φ}

(1)

where ΔH_m represents the heat of fusion (J/g), ΔH°_m represents the theoretical heat of fusion of 100% crystalline PLA (93.1 J/g)²¹ and φ represents the weight fraction of polymer in the biocomposites.

3.3 Colour Changes and Weight Loss

Colour changes after each accelerated weathering period were determined via observation made by optical images of the weathered specimens and comparison with unweathered ones.

After each accelerated weathering period, the samples were kept in a dry desiccator until analysis. Average weight loss (WL) was determined as:

W L = (\frac{m_{t} - m_{0}}{m_{0}}) 100

(2)

where m₀ is the initial mass of the sample before accelerated weathering and m_t is the mass after a certain time (t) of accelerated weathering. An average of three measurements was used to report the results.

3.4 Morphology

Scanning electron micrographs (SEM) were taken to characterise the morphology of the OHF before and after treatment and the biocomposites. Firstly, the samples were broken under cryogenic conditions (liquid nitrogen). Then, a Au/Pd layer was sputtered onto the exposed surfaces to take images at different magnifications on a JEOL JSM-840A microscope operated at 15 kV.

3.5 Fourier Transform Infrared (FTIR) Spectroscopy

FTIR spectra were recorded using a SHIMADZU FTIR-8400S in the 4000–400 cm⁻¹ range with a resolution of 4 cm⁻¹.

4. Results and Discussion

4.1 Morphological Studies

Figure 1a presents typical micrographs of the PLA and biocomposites before and after the accelerated ageing. PLA presents an apparently smooth surface, while the presence of particles in the biocomposites leads to a different morphology with voids, holes and exposed particles.

Figure 1

(a) Micrographs of the cryofractured surface of PLA and the biocomposites before and after accelerated ageing (b) Micrographs of untreated and treated OHF and (c) Micrographs of the cryofractured surface of PLA and the biocomposites before and after accelerated ageing

During processing, the crystallisation might have induced internal rearrangements leading to a more fragile behaviour. Also, limited particle-matrix adhesion produced particles pull-out. Although the treated particles may have better contact (wettability) with the matrix, mechanical adhesion might not be present.

The SEM images of untreated and alkali treated OHF is shown in Figure 1b where surface roughness differences can be seen. According to Prasad et al.²², alkali treatment removes waxy epidermal tissues, adhesive pectin and hemicelluloses. A comparison between the untreated and alkali treated OHF reveals topographical changes because the removal of these low molecular weight compounds leads to the formation of a rough surface. This treatment will influence moisture absorption and provides a more effective surface area for better adhesion with the matrix.

But after accelerated ageing ( Figure 1a ), the PLA matrix presents a rougher texture in the biocomposites. This polymer texture change might be related to crystallinity variations during accelerated ageing, coupled with particles swelling due to water absorption disrupting the chain structure and packing, thus leading to higher free volume content (porosity).

As shown in the magnified images of untreated and treated composites before and after ageing ( Figure 1c ), the interface between OHF and PLA has been severely damaged as several voids after ageing can be observed, especially without chemical treatment.

4.2 Weight Loss and Composites Appearance

Weight loss during accelerated ageing and visual aspect of the samples are presented in Figure 2 . Initially, the sample weight is increasing with lower values for neat PLA than the biocomposites ( Figure 2a ). This weight gain can be associated to water absorption during water spray and condensation cycles. The lower moisture absorbed by the neat PLA is expected due to its lower hygroscopic nature compared to OHF; i.e. the particles are more hydrophilic because of their polar groups such as −OH and −COOH leading to higher affinity towards water molecules. On the other hand, cellulosic materials are hydrophilic in nature, and therefore, they can absorb water²³.

Figure 2

a) Weight loss curves as a function of weathering time and b) variations of the visual aspect

As Figure 2a shows, treated and untreated biocomposites have similar behaviour. The biocomposites might also have some preferential channel at the particle-matrix interface leading to increased water absorption and diffusion in the samples²⁴. But the weight loss after weight gain at longer time (20 days) might be caused by molecules (extractives) leaching out upon exposure to the accelerated ageing.

Colour change is also a significant problem for polymer composites. Colour changes were first evaluated by comparing images of the specimens after each accelerated weathering period. Figure 2b shows that neat PLA is colourless; i.e. mostly transparent due to the high degree of amorphous structure related to the very fast cooling rate of injection moulding impeding PLA chains to crystallise (see Table 1 ). No significant colour change for the PLA specimens were observed, even after 480 h, which is in agreement with a previous study²⁵. When 20 wt.% OHF was added, Figure 2b shows that the biocomposites initially have a glossy/creamy colour. But after each accelerated weathering period, a slight colour change can be seen, especially in the loss of the glossy character²⁶. Nevertheless, the difference is again less important for the treated particles.

Table 1

Thermal parameters of the biocomposite materials

Sample	ΔHcc (J/g)	ΔHm (J/g)	x (%)	Tg (°C)	Tc (°C)	Tm (°C)
PLA 0 h	19.5	19	20.4	60	112	147
PLA 480 h	25.0	28	30.0	63	123	149
Untreated 0 h	23.0	25	33.5	62	112	147
Untreated 480 h	25.0	29	38.9	62	110	145
Alkali treated 0 h	19.0	18	24.1	62	115	145
Alkali treated 480 h	24.5	27	36.2	60	114	143

4.3 Fourier Transfer Infrared Spectroscopy (FTIR)

Neat PLA and the biocomposites before and after different weathering time were analysed by FTIR to detect any changes in specific chemical groups to determine possible composition variations²⁷.

Figure 3a presents the neat PLA spectrum with the assigned peaks. According to the literature^28–31, it is known that PLA can be identified by five characteristic bonds³²: C–C stretching at 867 cm^–1, C–O stretching at 1079, 1126, 1181 and 1264 cm^–1, C–H deformation at 1377 and 1452 cm^–1, C=O ester carbonyl groups at 1748 cm^–1, and C–H stretching at 2853 and 2924 cm^–1.

Figure 3

FTIR spectra of the samples: (a) neat PLA, (b) untreated biocomposites and (c) treated biocomposites

It has been reported that the main degradation mechanism associated to UV irradiation is known as “photolysis” leading to “chain scission”, especially in the C–O and C–C bonds of the PLA ester backbone structure by photon absorption. Effectively, the intensity of these IR bands is decreasing after each weathering cycle. Apart from photolysis, another effect of UV irradiation is “photo-oxidation” that might initially result in organic peroxide formation and their subsequent degradation into carboxylic acid and ketone end groups having C=O bonds³³. It is also observed that the intensity of this IR band either increases or shifts to lower wavenumbers with time.

It is known that hydrolysis also contributes to main chain scission, especially in the C–O bonds of the PLA ester structure³⁴. It was also reported that, similar as for photo-oxidation, hydrolysis leads to the formation of carboxylic acid and ketone end groups. Therefore, changes in the IR spectrum due to hydrolysis would be very similar as for the changes discussed above.

For the untreated biocomposites, Figure 3b shows an increased CO absorption intensity for the 480 h weathered samples over the control sample around 1734 cm⁻¹ and 1600 cm⁻¹ ³⁵. The increase in the carbonyl absorption band indicates some modifications in the lignins structure. Increased intensity of the 1600 cm⁻¹ band may indicate the formation of quinone by irradiation during accelerated ageing which was found around 1650 cm⁻¹ by others³⁵. The bands at 1734 and 1600 cm⁻¹ are characteristic absorptions of carbonyl stretching vibrations of non-conjugated (in xylan) and conjugated (in lignins) esters and carboxylic acids, and their concentration increases as carbonyl groups are produced from lignins and/or carbohydrates due to chemical degradation³⁶. Neat PLA has a C-H deformation band at 1350 and 1460 cm⁻¹, which was found to be at 1400 cm⁻¹ for both control and aged samples³⁷.

4.4 Differential Scanning Calorimetry (DSC) Analysis

DSC measurements were used to characterise the thermal properties and the thermograms are presented in Figure 4 , while the properties are reported in Table 1 . The curves were also used to determine the glass transition temperature (T_g), melting temperature (T_m) and crystallinity degree (x).

Figure 4

DSC traces for: (a) neat PLA, as well as (b) untreated and (c) alkali treated biocomposites before and after accelerated ageing (480 h)

As observed in Figure 4 , the presence of a melt-peak in the thermograms indicates the presence of crystalline domains, while a double melting endotherm was detected for the second melting of aged biocomposites having a main melting temperature of 147°C and a shoulder at 150°C, indicating a melt recrystallisation and remelting of PLA in the biocomposites. This behaviour has been reported for different PLA systems and has been linked to the presence of two different crystal structures: the α form (pseudo-orthorhombic) melting at higher temperature, while the β form (orthorhombic), also known as imperfect crystal, melts at a lower temperature within the melting temperature range³⁸.

The biocomposites have higher Tg compared with neat PLA. Similar increases for PLA based materials were reported by Acioli-Moura and Sun³⁹, Hablot et al.⁴⁰, Moura et al.⁴¹ and Spiridon et al.⁴². This trend is associated to the nucleating effect of the particles, as well as their limiting effect on molecular chain motion (steric hindrance).

The combined action of temperature, humidity and UV led to surface embrittlement of the exposed samples, explaining the slight increase in the melting temperatures, with the exception of PLA/olive husk flour biocomposites. It seems that in this biocomposites the higher crystallinity level can be attributed to the scission reactions occurring at the lamellar surfaces leading to higher crystal surface free energy⁴³. Another reason for the increased crystallinity during accelerated weathering is chain scission (decreased molecular weight) from photolysis and hydrolysis as shorter chains are more mobile, so conformational requirement for the creation of an ordered crystalline structure is easier.

4.5 Effect of Accelerated Ageing on Mechanical Properties

Figure 5 presents the mechanical properties changes due to accelerated ageing. As expected, the tensile strength decreases with exposure time, but the treated particles have slightly higher values than untreated ones. Nevertheless, the effect of alkali treatment is better seen on Young's modulus. Finally, the elongation at break is also slightly better with treated particles.

Figure 5

Effect of accelerated ageing on: (a) tensile strength, (b) Young's modulus and (c) elongation at break of neat PLA, as well as the untreated and alkali treated biocomposites

The lower tensile strength (TS) for the biocomposites is usually related to poor fibre/matrix interfacial adhesion. As expected, the alkali treatment was able to improve tensile strength. This can be associated to better interfacial contact/bonding between the olive husk flour and PLA matrix.

As expected, tensile strength (TS) decreases with increased weathering time. The most important TS reduction is observed for the untreated biocomposites. For example, the tensile strength was found to decrease from 54 to 39 MPa (28%) for untreated biocomposites, while the decrease is lower for treated ones (22%) ( Figure 5a ). The reduction for neat PLA with increasing accelerated ageing time is considered to be related to plasticisation and swelling effects⁴⁴, as well as photochemical degradation⁴⁵.

As seen in Figure 5b , the Young's modulus (YM) was found to decrease from 1593 MPa to 1134 MPa (29%) for untreated biocomposites, while the reduction is much lower for treated ones (17%). Although weight loss by photodegradation and leaching would be expected to decrease the molecular weight of the neat PLA leading to increased crystallinity, these effects lead to lower YM. Nevertheless, the plasticisation effect by humidity absorption in the neat PLA could increase the matrix toughness and outweigh the effect of increased crystallinity. Also, as the alkali treatment decreased the relative amount of lignins⁴⁶, it is highly possible that the untreated biocomposites would be less susceptible to UV radiation leading to more mechanical properties retention than for untreated biocomposites.

Finally, lower elongation at break with OHF addition in PLA was observed ( Figure 5c ). This is attributed to the particles rigidity. Again, the highest overall reduction was observed for the untreated biocomposites. In all cases, the values decreased with weathering time, while the surface treatment gave higher values for the biocomposites.

5. Conclusion

In this work, biocomposites based on poly(lactic acid) and olive husk flour (OHF) were produced via melt processing (extrusion followed by injection). In particular, a chemical modification of the particles was performed (alkali treatment). The main objective of the work was to determine the effect of particle addition and surface treatment on the long term properties and degradation (acceleration weathering) of these biocomposites.

The results showed that the addition of olive husk flour into PLA had a positive effect on the biocomposites Young's modulus and the treated particles had higher mechanical properties.

After accelerated ageing, tensile strength and Young's modulus were found to decrease for both untreated and alkali treated biocomposites. However, untreated biocomposites had the highest overall mechanical properties reduction compared to alkali treated biocomposites. After accelerated aging, the neat PLA became yellow, while the biocomposites had a darker brown colour (loss of surface gloss).

FTIR and DSC analyses were used to confirm that degradation occurred with time leading to changes in crystallinity content and chemical groups present to explain the mechanical properties loses upon exposure to accelerated ageing. These results were also confirmed by morphological analysis (SEM).

Finally, our results showed that PLA/OHF is a suitable combination to produce biocomposites derived from fully natural resources, allowing the development of environmentally friendly materials having better properties than the neat matrix. Nevertheless, more work needs to be done to improve on the results obtained, especially in terms of interfacial adhesion by using a “natural” coupling agent.

Footnotes

Acknowledgement

The authors would like to thank the Bejaia oil mill for the olive husk used in this study. We also thank the research centre on advanced materials (CERMA) and the research centre on renewable materials (CRMR) for technical help on processing and characterisation.

References

Mohanty

, Misra

, Drzal

Journal of Polymers and the Environment. 10 (2002) 19–26.

Elsawy

M. A.

, Kim

K.-H.

, Park

J.-W.

Renewable and Sustainable Energy Reviews. 79 (2017) 1346–1352.

Mittal

, Chaudhry

, Matsko

N. B.

Journal of Applied Polymer Science. 131 (2014) 40658.

Di Giuseppe

, Castellani

, Budtova

Composites Part A: Applied Science and Manufacturing. 95 (2017) 31–39.

Adhikary

K. B.

, Pang

, Staiger

M. P.

Composites Part B: Engineering. 39 (2008) 807–815.

Zivkovic

, Pavlovic

, Fragassa

International Journal for Quality Research. 10 (2016).

Evans

P. D.

Wood Material Science and Engineering. 4 (2009) 2–13.

Muñoz

C. M.

, Egsgaard

, Landaluze

J. S.

Journal of the American Institute for Conservation. 53 (2014) 236–251.

Dixit

American Journal of Polymer Science & Engineering. 2 (2014) 1–11.

10.

Dence

C. W.

, The determination of lignin, in Methods in lignin chemistry. Springer, pp. (1992) 33–61.

11.

Harper

Composites. 13 (1982) 123–128.

12.

Anglès

, Ferrando

, Farriol

Biomass and Bioenergy. 21 (2001) 211–224.

13.

Hon

D. N.-S.

, Shiraishi

Wood and cellulosic chemistry, revised, and expanded, CRC press (2000).

14.

Graupner

, Rößler

, Ziegmann

Composites Part A: Applied Science and Manufacturing. 63 (2014) 133–148.

15.

Schmidt

, Kimura

, Gray

Research on Chemical Intermediates. 21 (1995) 287–301.

16.

Yun

, He

BioResources. 12 (2017) 5236–5248.

17.

Jamshidian

, Tehrany

E. A.

, Imran

Comprehensive Reviews in Food Science and Food Safety. 9 (2010) 552–571.

18.

Le Duigou

, Davies

, Baley

Polymer Degradation and Stability. 94 (2009) 1151–1162.

19.

C.-S.

Polymer Degradation and Stability. 94 (2009) 1076–1084.

20.

Shukla

N. B.

, Daraboina

, Madras

Polymer Degradation and Stability. 94 (2009) 1238–1244.

21.

Deroiné

, Le Duigou

, Corre

Y.-M.

Polymer Degradation and Stability. 108 (2014) 319–329.

22.

Prasad

, Pavithran

, Rohatgi

Journal of Materials Science 18 (1983) 1443–1454.

23.

Hammiche

, Boukerrou

, Djidjelli

Construction and Building Materials 47 (2013) 293–300.

24.

Martucci

J. F.

, Ruseckaite

R. A.

Polymer Degradation and Stability. 116 (2015) 36–44.

25.

Kaynak

, Dogu

International Polymer Processing. 31 (2016) 410–422.

26.

Peng

, Cao

Photochemical Behavior of Multicomponent Polymeric-based Materials. (2016) 291–346.

27.

Sandler

S. R.

, Karo

, Bonesteel

Polymer synthesis and characterization: a laboratory manual, Academic Press (1998).

28.

Bocchini

, Fukushima

, Blasio

A. D.

Biomacromolecules. 11 (2010) 2919–2926.

29.

Ozdemir

, Lekesiz

T. O.

, Hacaloglu

Polymer Degradation and Stability. 134 (2016) 87–96.

30.

Persson

, Lorite

G. S.

, Kokkonen

H. E.

Colloids and Surfaces B:

Biointerfaces., 409 (2014) 416–121.

31.

Janorkar

A. V.

, Metters

A. T.

, Hirt

D. E.

Journal of Applied Polymer Science. 106 (2007) 1042–1047.

32.

Barczewski

, Matykiewicz

, Krygier

Journal of Material Cycles and Waste Management. (2017) in press.

33.

Lesaffre

, Bellayer

, Vezin

Polymer Degradation and Stability. 139 (2017) 143–164.

34.

Journal of Biomedical Materials Research. 48 (1999) 342–353.

35.

Pandey

K. K.

, Nagaveni

BioResources. 4 (2009) 257–267.

36.

Copinet

, Bertrand

, Longieras

Journal of Polymers and the Environment. 11 (2003) 169–179.

37.

Nakayama

, Hayashi

Polymer Degradation and Stability. 92 (2007) 1255–1264.

38.

Arpitha

, Sanjay

, Yogesha

Science. 2 (2017) 59–65.

39.

Acioli-Moura

, Sun

X. S.

Polymer Engineering & Science. 48 (2008) 829–836.

40.

Hablot

, Dharmalingam

, Hayes

D. G.

Journal of Polymers and the Environment. 22 (2014) 417–429.

41.

Moura

, Botelho

, Machado

Journal of Polymers and the Environmen.t 22 (2014) 148–157.

42.

Spiridon

, Paduraru

O. M.

, Zaltariov

M. F.

Industrial & Engineering Chemistry Research. 52 (2013) 9822–9833.

43.

Cho

, Kim

, Yoon

Macromolecules. 36 (2003) 7652–7660.

44.

Boudrahem

, Belbah

, Kirati

, Journal of Materials, Processes and Environment. 2 (2014) 20–26.

45.

Azwa

, Yousif

Thermal degradation study of kenaf fibre/epoxy composites using thermo gravimetric analysis, in 3rd Malaysian postgraduate conference (MPC2013), pp. 4–5 (2013).

46.

Islam

M. S.

, Pickering

K. L.

, Foreman

N. J.

Journal of Adhesion Science and Technology. 23 (2009) 2085–2107.