Chemical modification of regenerated cellulose fibres by cellulose nano-crystals: Towards hierarchical structure for structural composites reinforcement

Industrial Crops and Products 100 (2017) 41–50

Contents lists available at ScienceDirect

Industrial Crops and Products journal homepage: www.elsevier.com/locate/indcrop

Chemical modification of regenerated cellulose fibres by cellulose nano-crystals: Towards hierarchical structure for structural composites reinforcement Abdelghani Hajlane a,b , Hamid Kaddami b,∗ , Roberts Joffe a,c a

Composite Centre Sweden, Luleå University of Technology, S-97187 Luleå, Sweden Laboratory of Organometallic and Macromolecular Chemistry-Composite Materials, Faculty of Sciences and Techniques, Cadi Ayyad University, Av. Abdelkrim el khattabi B.P. 549, Marrakech, Morocco c Group of Materials Science, Swerea SICOMP, S-94126 Piteå, Sweden b

a r t i c l e

i n f o

Article history: Received 28 September 2016 Received in revised form 3 February 2017 Accepted 4 February 2017 Keywords: Regenerated cellulose fibres CNC Chemical modification Cerium ammonium initiation Hierarchical structure Mechanical properties

a b s t r a c t A simple and innovative new route, with less negative impact on the environment, for depositing and hope-grafting cellulose nano-crystals onto the surface of regenerated cellulose fibres (Cordenka 700 Super 3), using ␥-methacryloxypropyltrimethoxysilane as coupling agent, is presented. Hierarchical cellulosic structure involving micro-scale fibres and nano-scale cellulose crystal network was created as verified by the scanning electron microscopy. The fibres were initially oxidised by optimized concentration of cerium ammonium nitrate to generate radicals on the cellulose backbone in order to polymerize the coupling agent at the surface. Infrared spectroscopy and scanning electron microscopy confirmed the chemical polymerisation of MPS onto regenerated cellulose fibres without enabling to show the chemical bonding between silane and nano-crystals. However, tensile test which was performed to study the impact of different treatments on mechanical properties of regenerated cellulose fibres, revealed that the modification by silane decreased the stiffness and strength of fibres (22% and 10% decrease, respectively) while the strain at failure was increased. These changes were attributed to the treatment conditions which may have induced the disorder and the misalignment of the structure of cellulose fibres (e.g. axial orientation of molecular chains and crystalline phase of the fibre has been reduced). This assumption is supported by the results from successive loading-unloading test of the fibre bundle. However, after depositing cellulose nano-crystals onto the fibre’s surface, the stiffness was recovered (20% increase in comparison to MPS treated fibres) while the strength and strain at failure remained at the same order of magnitude as for fibres treated only by the coupling agent. © 2017 Elsevier B.V. All rights reserved.

1. Introduction Due to high mechanical properties combined with chemical/environmental resistance and low weight, synthetic polymer composites reinforced with continuous synthetic fibres are widely used in structural applications such as aerospace and aeronautics, marine, oil/gas industries, energy, transportation, sporting equipment, construction. However, because of environmental concerns and public awareness of potential problems associated with growing consumption of oil and petroleum based materials (such as polymers, for example) there is rising demand for sustainable and eco-friendly materials. In latest years this demand massively

∗ Corresponding author. E-mail addresses: [email protected], [email protected] (H. Kaddami). http://dx.doi.org/10.1016/j.indcrop.2017.02.006 0926-6690/© 2017 Elsevier B.V. All rights reserved.

boosted research on natural fibre composites and bio-based polymers. In particular plant fibres with high content of cellulose are of great interest due to their high stiffness and strength. Cellulose is the most abundant renewable resource polymer on Earth where the nature produces more than 7.5 × 1010 tons a year (Habibi, 2014). The remarkable physical and chemical properties allow this sustainable polymer to be used in various forms for different applications that cover huge domains ranging from products that do not require high mechanical properties (e.g. papers, cellophane films, etc) up to the applications that require load bearing capabilities, for instance structural polymer composites (e.g. precursor for carbon fibres, regenerated cellulose fibres). It is well known that composites usually exhibit excellent in-plane properties (e.g. tensile stiffness and strength) which are controlled by fibres that may be assembled in various 2-D reinforcements (e.g. uni-directional fabric, weave, non-crimp fab-

42

A. Hajlane et al. / Industrial Crops and Products 100 (2017) 41–50

ric). However, through-thickness properties are relatively weak because they are governed by the properties of the resin and in large extend by the strength of fibre/matrix interface (Tong et al., 2002). In general, the reinforcing function of fibres is mostly governed by the fibre/matrix interaction (interfacial adhesion) that can be enhanced through chemical or physical modification of the polymer and/or the fibre. Natural fibres bear hydroxyl groups from cellulose as well as lignin and there are numerous papers published regarding the modification and the use of coupling agents to promote the adhesion between the fibre surface and the matrix. Such modifications use coupling agent depending on the physico-chemical nature of the matrix. Maleic anhydride graft PP homopolymers or copolymers have been widely used as coupling agent for thermoplastic composites (Angles et al., 1999; Bendahou et al., 2008; Caulfield et al., 1999). On the other hand, silicon alkoxides such as alkyl triethoxysilanes (Devi et al., 1997; Gassan and Bledzki, 1999; Pothan et al., 1997; Raj et al., 1990), pure organic coupling agents (Felix and Gatenholm, 1991; Hassan and Nada, 2003; Kaddami et al., 2006; Sbiai et al., 2008; Singh et al., 1998) or simple oxidation of the fibres (Dai et al., 2013; Moharana et al., 1990; Sbiai et al., 2011; Sbiai et al., 2010; Wei et al., 2016) have been frequently used to modify lignocellulosic fibres to improve the composites processing with thermoset matrices. Moreover, and even if these modifications usually decrease the own mechanical properties of the fibres, they upgrade the adhesion fibre/matrix and then enhance not only the mechanical properties of composites and reduce the water uptake. Among other methods to increase adhesion between fibre and matrix the development of multi-scale hierarchical composite is rapidly growing. This approach joins nano-scale reinforcement with micro-scale fibres and has been tried out on various combinations of substrates (micro-scale fibres) with different nano-reinforcement (Qian et al., 2010). For instance, carbon nanotubes (CNT) have been deposited on various types of fibres (carbon fibres, quartz, alumina), the deposition can be achieved by growing CNT directly on fibres, by electrophoretic deposition or by incorporating CNT into sizing applied on fibres (Qian et al., 2010). In the recent work concerning CNTs grown on woven fabrics of quartz and alumina fibres (Li et al., 2015) it was demonstrated that addition of nano-phase on the surface of reinforcement resulted in an increase of the in-plane shear strength of epoxy based composite. Similar approach of improving adhesion between natural fibres and polymer matrix can be employed in bio-based composites. However, the use of nano-reinforcement, such as CNT, is not possible with bio-based materials because it would diminish environmental friendliness of these materials. Therefore, in case of micro-sized cellulosic fibres one should also consider use of biobased nano-reinforcement, such as cellulose nano-crystals for instance. Pure cellulose can be extracted from cellulosic materials and be highly valorized and transformed under an acid hydrolysis into rod-like crystalline particles: cellulose nanocrystals (CNC) (Habibi, 2014; Lin et al., 2012) with an elastic modulus between 120 ˇ et al., 2005). Regarding the surface chemistry and 150 GPa (Sturcová of CNC, where a large number of hydroxyl groups are present, they represent a unique substrate for significant surface modification to graft molecules with different functional groups by applying various chemical processes outspreading therefore their use in wide range of applications (Habibi et al., 2010; Peng et al., 2011; Zhou et al., 2011). Earlier two routes of chemical modification of the surface of cellulose nano-crystals extracted from palm tree (Bendahou et al., 2014) have been developed. It was also demonstrated in previous work (Hajlane et al., 2013) that cellulose nano-crystals extracted from date palm tree can be successfully deposited onto micro-sized continuous regenerated cellulose fibres (RCF) in order to create hierarchical structure. It resulted in composite with improved

fibre/matrix adhesion which delayed initiation of transverse cracks in material and increased the transverse properties of a unidirectional (UD) laminate. The approach to design composites with hierarchical structure to enhanced intralaminar and interlaminar performance is currently developed by number of researchers. Within the same framework of creating hierarchical structure, this paper describes another route to deposit cellulose nano-crystals onto RCF with less negative impact on the environment by using water as solvent in the chemical process and optimized concentrations of reagents. This makes it more suitable for industrial process. On the other hand, the introduction of “polymerizable” coupling agent will generate interface with new properties that will certain improve the mechanical properties by more pronounced delay of cracks initiation in the resulting structural composites. 2. Experimental 2.1. Materials The micro-scale reinforcement used in this study was RCF Cordenka 700 Super 3 twisted Z100 (CORDENKA GmbH, Germany). Fibres were supplied in a form of bundles on bobbin, the summary of the materials characteristics are given in Table 1 (Wang et al., 1988). Cellulose nano-crystals (CNC) were extracted from date palm tree in the laboratory according to the procedure described in previous work (Bendahou et al., 2009). These nano-crystals were deposited using Methacryloxypropyltri-methoxysilane (MPS) as a coupling agent (purity ≥ 98%). This later was purchased from Sigma Aldrich along with Sodium hydroxide, Nitric Acid and Ethanol. The initiator Cerium Amonium Nitrate (CAN) was purchased from Fluka. All these chemicals were used as-received without any further purification. The water was distilled in laboratory in order to be used as co-solvent with ethanol. 2.2. Chemical modification of fibres 2.2.1. Silane treatment A 1600 ml mixture of ethanol and water (50/50 v%) was poured into a reactor and heated up to 65 ◦ C. After stabilizing the temperature, NaOH and Nitric acid solutions were added in order to adjust the pH to 7. When the pH was stabilised (takes approximately 2 h), 16 g of fibres previously wound on a holder and fixed on mechanical stirrer were introduced in the reactor. The ratio of fibres to solvent was adjusted to be 1% (w/v). The graft polymerisation of MPS was carried out by adding first cerium ammonium nitrate (CAN) 10−3 mol L−1 in the reactor and stirred for 30 min. Meanwhile, the mixture was purged with Nitrogen gas for 15 min to remove any dissolved oxygen gas. At this stage, the free-radicals are supposed to be created at the surface of the cellulose backbone and ready to react with vinyl monomers (MPS). To initiate the graft polymerization, 10−3 mL L−1 of MPS was added to the reactor and the nitrogen gas flow was maintained until the end of the reaction (5 h); the mixture was stirred at medium constant rate. It’s worthy noticing that the pH and the concentrations of CAN and MPS used were chosen after optimization of the reaction conditions by varying the ceric ion concentration, reaction time, pH as well as MPS concentration. Fibres were then washed two times by ethanol and once by water to remove unreacted products and other impurities trapped in the grafted fibres. 2.2.2. NCC deposition After polymerization of MPS onto the surface of cellulose fibres, different concentrations of cellulose nano-crystals were prepared in order to create hierarchical structure combining the micro-sized

A. Hajlane et al. / Industrial Crops and Products 100 (2017) 41–50

43

Table 1 Characteristics of Cordenka 700 Super 3 fibres provided by the manufacturer in the material datasheet. Linear density (dtex)

Breaking force (N)

Elongation at break (%)

Number of filaments in the bundle

Twist (t/m)

Diameter of fibre (␮m)

2440

127.9

12.2

1350

100

12.5

fibres and cellulose nano-crystals. These nano-crystals were prepared according to the method described in earlier work (Hajlane et al., 2013). The process of depositing the nano-crystals was performed by adding three different concentrations of cellulose nano-crystals 0.1, 0.2 and 0.4 wt%, respectively. The distilled water (1600 mL) was poured in the reactor and the pH was adjusted to be between 4 and 5 by adding sulphuric acid (H2 SO4 ). When the pH was stabilised, the fibres modified by MPS were re-placed in the reactor and a specific concentration of nano-crystals was added. Prior to sonication, a solution of NaOH (0.1 M) was added to cellulose nano-crystals to reduce the amount of sulfonate groups and release the hydroxyl (or alcoholate) groups of CNC. CNC suspensions were first sonicated for 5 min prior to the addition into the reactor. The reaction was maintained at medium stirring rate at room temperature for 5 h. Afterwards, the fibres were washed three times by distilled water and dried at 110 ◦ C for 4 h. Final product was stored in desiccator to protect the fibres from dust. 2.3. Characterisation of fibres 2.3.1. Fourier transform infrared spectroscopy Fourier Transform Infrared (FTIR) spectroscopy was used to analyze the functional groups of RCF and also to investigate the changes brought by the chemical modifications of fibres. The measurements were performed with PerkinElmer 100 spectrometer. In order to obtain a good resolution of the spectrum, the fibres were cut to length of 1 cm (sample holder diameter). Then the fibres were centred in the holder and subjected to the infrared light. The spectra of the samples were obtained after 32 scans from 400 to 4000 cm−1 with a resolution of 4 cm−1 . 2.3.2. Wide angle X-ray diffraction (WXRD) Wide Angle X-ray powder diffraction was carried out at D500 Siemens diffractometer, using a radiation wavelength of 1.54 å (CuK␣) at ambient moisture and temperature. The scattering data was collected in the angular range 2␪ from 5 to 60◦ at a rate of 2◦ min−1 . Crystallinity index (CrI) was determined using the Segal et al. (Segal et al., 1959) empirical equation adopted for cellulose II (Azubuike et al., 2012). CrI =

I002 − Iam × 100 I002

Where I002 is the maximum intensity of diffraction of the (002) lattice peak at 2␪ angle between 22 and 23◦ , and Ia is the intensity of diffraction of the amorphous part of the fibres, which is taken at a 2␪ angle between 18 and 19◦ , where the intensity is at its minimum (Roncero et al., 2005). 2.3.3. Scanning electron microscopy (SEM) The topography of the surface of untreated and treated fibres either by MPS or by nano-crystals was observed by SEM operating in high vacuum pressure (5 × 10−2 mbar) at accelerating voltage of 10–15 kV. Prior to scanning process, the samples were sputtered with gold using BLAZER SCD 050 Sputter coater in argon atmosphere. In order to obtain high quality images, the sputtering was performed at 60 mA for 50 s.

Fig. 1. Typical results from loading-unloading test of Cordenka 700 Super 3 fibre bundles.

2.3.4. Mechanical properties of fibres Tensile specimens were prepared by gluing bundles between wooden tabs using a two components epoxy: Araldite 2011. The final fibre length between the tabs was 100 mm (schematic drawing of specimen can be found in (Hajlane et al., 2013)). Tensile tests were performed on fibre bundles using Instron 4411 machine with 500 N load-cell. At least 5 successful tests (specimen loaded until the failure) from each fibre batch were performed. In order to obtain comparative results, all tests were run at the same strain rate of 10% min−1 (cross-head speed set in “mm/min” according to the specimen length) and at 23 ◦ C while the relative humidity was around 14%. Additionally to the 100 mm long specimens three different bundle lengths (50, 150, and 200 mm) were tested in order to determine the machine compliance which should be accounted for measurements of strain (similarly as described in standard ASTM D 3379-75 for single fibre tensile tests). The stress was obtained by dividing the applied force by the total cross section area of fibre bundle (combined cross-sections area of 1350 single fibres). Strain was obtained from the displacement of cross-head of machine (system compliance was accounted for). Stiffness was determined from the linear part of the stress-strain curve within strain interval of 0.4 and 0.9%. This interval was chosen based on a typical stress-strain curve of untreated fibre bundle presented in Fig. 7a. Strength and strain at failure were obtained from the stress-strain curve at the point of bundle failure. The loading-unloading test was conducted on all fibres to study the impact of treatment on stiffness evolution. In fact, stiffness evolution is directly related to the microstructure and therefore each change is expected to appear “macroscopically” on the stress-strain behaviour. This test was carried out on the same bundle configuration and at the same conditions as simple tensile tests. In this case, the fibre bundles were loaded to certain load level and unloaded to 0N, the load was increased by 5N in each consecutive step. The typical ensemble of stress-strain curves obtained in such test is shown in Fig. 1. The bundle stiffness was calculated in the linear part of each loading cycle for stiffness evolution analysis. The stiffness was calculated within stress interval of 40 MPa to 90 MPa which corresponds to the strain levels used in simple tensile test (0.4–0.9%).

44

A. Hajlane et al. / Industrial Crops and Products 100 (2017) 41–50

Fig. 2. Schematic representation of chemical reactions of MPS on cellulose fibres.

2.4. Chemical grafting of MPS on cellulose fibres: preliminary considerations Different mechanisms had been proposed to describe the creation of radicals on cellulose backbone (McDowall et al., 1984). In the present work the grafting procedure of MPS was based on the method described by N.Gaylord (Gaylord, 1972). The polymerisation reaction of MPS has been reported to be a process in which the transfer of electrons from the hydroxyl groups of cellulose to the CAN ions happens to create a free radical of anhydroglucose units of the cellulose molecule. In fact, this mechanism supposes that the grafting site is the cleaved C2-C3 bond of the anhydroglucose unit generating a carbonyl group and a free-radical which initiates the polymerisation of MPS monomers. A simplified mechanism of this reaction is schematically shown in Fig. 2. This mechanism has been inspired from some works reported by (Yilmaz et al., 2007; Zhang et al., 2003). During the process of modification by MPS, the methoxy groups of MPS react with water and transform to silanol groups. These groups are expected to interact with hydroxyl groups of cellulose nano-crystals units and therefore react after removing water. Thus, the depositing process of cellulose nano-crystals was carried out in distilled water at pH between 4 and 5. At this particular pH, the silanol groups tend to condense in linear siloxane chain (2D network) at low temperature (Ying et al., 1993). Simplified schematics of the reaction mechanism are presented in Fig. 3. After CNC deposition the fibres were heated at 105 ◦ C under vacuum for 1 h to promote chemical reaction between silanol groups of MPS and hydroxyl groups of CNC.

Fig. 3. Simplified illustration for depositing cellulose nano-crystals on modified fibres by MPS.

3. Results and discussion 3.1. Microstructure characterization 3.1.1. FTIR FTIR was applied to characterize the chemical structure of cellulose and to elucidate the changes after chemical treatments. Spectra

A. Hajlane et al. / Industrial Crops and Products 100 (2017) 41–50

45

Fig. 4. Infrared spectra of fibres: a) unmodified; b) modified by MPS; c) deposited CNC.

of modified fibres are shown in Fig. 4 along with spectrum of unmodified fibres for comparison. The progress of the chemical treatment of fibres resulted in modification and appearance of peaks revealed by FTIR. A decrease in the intensity of the large band at 3347 cm−1 characteristic for the stretching vibration of the hydroxyl groups (OH) was observed. In fact, during the treatment by MPS, the hydroxyl groups, particularly, OH-groups in position 3 of the glycosidic unit were transformed to aldehyde during the oxidation process by CAN (indicated by an arrow in Fig. 3). Moreover, it might happen that the penetration of water molecules is reduced and subsequently the peak intensity is decreased. A new band appeared around 2974 cm−1 which corresponds to the stretching vibration of the C H alkane. The presence of this new band is due to the increase of amount of CH-molecules after grafting MPS onto the surface of fibres. Increase in intensity of the asymmetric stretching vibration of the band at 2892 cm−1 , which corresponds to the CH2 group indicating the presence of the aliphatic chains grafted on fibres, was observed. These bands’ attributions are in agreement with previous works reported using the same or similar coupling agents (Abdelmouleh et al., 2004; Abdelmouleh et al., 2002; Bach et al., 2005; Botaro and Gandini, 1998; Botaro et al., 2005; Gandini et al., 2001). The polymerised MPS on fibres can be characterized by the presence of the band 1720 cm−1 corresponding to the stretching vibration of the C O functions which is characteristic to COO group from MPS. The presence of this new stretching band provided sufficient evidence of the successful chemical modification. However, it was difficult to prove the presence of CNC by FTIR since the bands corresponding to Si O Si and Si O C are overlapping with C O C and are hidden by the large band from 1000 to 1200 cm−1 .

Fig. 5. WAXD patterns of the cellulose fibres for different treatments, pattern of the untreated fibres is presented for comparison.

3.1.2. WAXD analysis X-ray analyses were performed in order to study the effect of the chemical treatments on the microstructure of fibres. Any change in the microstructure would influence the overall mechanical properties of fibres. Fig. 5 shows WAXD patterns of fibres before and after treatments. Untreated fibres (VF) present a peak around 22◦ corresponding to the crystallographic plane (002). Treated fibres (MPS) show, however, in addition to the peak located around 22◦ , pronounced peaks around 12.5 and 20◦ which correspond to the crystallographic planes (011) and (011), respectively (Li et al., 2010). This indicates that the treatment misaligned the orientation

46

A. Hajlane et al. / Industrial Crops and Products 100 (2017) 41–50

Fig. 6. SEM micrographs of fibres: a) unmodified; b) modified by MPS; c) deposited CNC with concentration C1 = 0.1%; d) deposited CNC with concentration C2 = 0.2% and e) deposited CNC with concentration C3 = 0.4%.

of the crystalline phase of the fibre structure. It’s worth noticing to remind that these types of regenerated cellulose fibres (Rayon) were preferentially oriented with its crystallographic c-axis pointing along the fibre direction. In fact, the chemical treatments applied misoriented the crystalline region with the fibre axis given rise to the observed new peaks (at 12.5 and 20◦ ). Further, the fibres sized with CNC (MPSCNC2 and MPSCNC3) show the same crys-

talline structure as MPS. The absence of peaks corresponding to the crystallographic planes of CNC (which have a crystalline structure of cellulose I −native cellulose) could be explained by the low initial concentration and the very low fraction deposited on fibres that prevent CNC to be observed in WXRD spectra. On the other hand, the crystallinity indices calculated showed that the treatments did

A. Hajlane et al. / Industrial Crops and Products 100 (2017) 41–50

47

Fig. 7. Typical stress-strain curves for fibres with different treatments.

not affect the crystallinity of the fibres and were found to be around 61% (around what had been found in (Smole et al., 2003)). 3.1.3. SEM The comparison of SEM images of modified and untreated fibres (see Fig. 6) supports statement about success of modification procedure. Untreated fibres present a uniform and clean surface as it is shown in Fig. 6a. The treatments increased the roughness of the surface of fibres by particles resulting from the polymerization of the coupling agent as well as by the cellulose nano-crystals (see Fig. 6b–e). Fig. 6b exhibits a quasi-homogeneous distribution of the “poly-MPS” along the surface of fibres. After depositing cellulose nano-crystals, a network binding the single fibres together is created. This network was observed to be expanded by increasing the concentration of CNC suspension (See Fig. 6c and e). At low concentration (Fig. 6c), CNC are scarcely present between the fibres and showing non-homogeneous distribution. The higher concentration of CNC results in denser CNC network (as it is shown in Fig. 6d and e). Moreover, the CNC network shown in Fig. 6d (C = 0.2%) seems to be more distributed on the fibres’ surface in comparison with higher concentration (Fig. 6e). In general, the dispersion of cellulose nano-crystals is very challenging because of i) their affinity to each, ii) their high specific surface and iii) the electrostatic interactions they develop. In this case, the CNC are less aggregated which indicates that the sonication process was efficient and also that the attraction between the CNC and the coupling agent is high. However, observations along the fibre length showed that the CNC were not covering the whole surface of filaments but only act as a binder holding the fibres together. This would improve the fibre to fibre stress load transfer in further application in structural composites. 3.2. Mechanical properties of fibre bundles

Fig. 8. Effect of the treatment on mechanical properties of the fibre bundles. a) Stiffness; b) Strength; c) Strain at failure.). VF, untreated fibres; MPS, modified by MPS; MPSCNC1, MPSCNC2 and MPSCNC3 are deposited fibres by cellulose nanocrystals at different concentrations C1 = 0.1%, C2 = 0.2% and C3 = 0.4%, respectively.

3.2.1. Tensile tests Since it is intended to develop RCF composite for structural application, it is important to study the effect of the treatments on mechanical properties of fibres and relate these changes to the microstructure of material. The general behaviour of fibres can be represented by typical stress-strain curve of untreated fibres as shown in Fig. 7 (curve a). The two well defined regions on the curve can be distinguished, before and after yielding, indicating non-linear behaviour of these fibres. Each fibre exhibits the same performance with an initially linear section, a yield point where the slope of the curve decreases and an almost linear behaviour until failure.

A detailed description of the shape of tensile curve is reported in previous study (Hajlane et al., 2013). Typical stress-strain curves for fibres with different treatments are presented in Fig. 7. Fig. 8 shows the effect of different treatments on mechanical properties of fibres. The same results are summarized in Table 2. The stiffness of bundles was decreased (∼20%) after modification by MPS and increased again back to the reference values after depositing CNC at different concentrations (Fig. 8a). Whereas strain at failure is significantly increased due to modification by MPS (∼60%) and then reduced back to the reference values after depositing CNC on fibres (Fig. 8c). However, the tensile strength was decreased after

48

A. Hajlane et al. / Industrial Crops and Products 100 (2017) 41–50

Table 2 Summary results of mechanical properties for bundles with different fibre treatment.

VF MPS MPSCNC1 MPSCNC2 MPSCNC3

Stiffness (GPa)

Strength (MPa)

Strain (%)

20.3 ± 0.8 16.1 ± 0.6 19.3 ± 0.3 20.0 ± 0.6 19.3 ± 0.5

654 ± 42 590 ± 42 545 ± 13 526 ± 7 537 ± 11

8.0 ± 0.8 13.0 ± 1 8.5 ± 0.7 7.5 ± 0.2 9.0 ± 0.4

all types of treatments (Fig. 8b), by MPS (∼10%) and by depositing CNC (∼18%). It should be noted that the effect of CNC concentration appears to be rather irrelevant and the mechanical properties are similar for all concentrations. The process of modification by MPS seems to induce a plasticizing effect even though the oxidation by CAN was expected to induce defects and make fibres more brittle. It can be assumed that reduction of mechanical properties of fibres due to treatment occurs because of changes occurred in the internal structure of material. For example, during the chemical modification using solvent ethanol/water and relatively high temperature, the chains oriented along the fibre-axis may relax and disorder is introduced in both, amorphous and crystalline regions. This disorder results in, not only, an increase of the angle between the fibre axis and the crystallites orientation but also in misalignment of the amorphous part of the cellulose fibres. Those assumptions are supported by the results previously reported on orientation of the microstructure of regenerated cellulose. In fact, Hermans and co-workers (Hermans et al., 1948) have correlated the orientation of the crystalline and the amorphous portions of regenerated cellulose with the drawing ratio. It has been shown that during the drawing process the orientation factors of both crystalline and amorphous regions were influenced by the stretching of fibres. By comparing the evolution of those factors obtained by birefringence and by X-ray, it was possible to conclude that the orientation of the crystallites runs ahead of that of the amorphous portion. Another paper presents a study on the deformation mechanism at molecular level by combining Raman spectroscopy analyses and mechanical tests performed on cellulose fibres (Kong and Eichhorn, 2005). In this work it was not possible to identify within which part of cellulose (crystalline or amorphous) the orientation is taking place. However, a pronounced and continuous shift towards lower wave number of the band 1414 cm−1 which corresponds to the C O H was observed during the mechanical elongation of the fibres until the yield point. In this study, in the tensile tests curves the yield point is followed by a plateau at higher strains (instead of another increasing line as in our case). This plateau was attributed to a breaking down in the hydrogen bonding. It was also shown for different draw ratios of the fibres that the higher draw ratio and the corresponding Raman spectra showed more significant shift of the 1414 cm−1 band. It is worth noting that the elastic modulus of the fibre increases by increasing the draw ratio (as it is reported in (Kong and Eichhorn, 2005)). The current work shows that chemical treatment acts in the opposite direction (compare to the fibre stretching) and introduces misalignment on the molecular level. After depositing CNC, their network spreads over the surface of fibres and evolves the stiffness bringing it back to the same level as for untreated fibres. This preservation of the stiffness is very important for further application of these fibres in structural composite materials. 3.2.2. Loading-unloading tests Successive uniaxial cyclic loadings of fibres until failure were performed to identify the potential damage of fibres resulting from the chemical modifications. As shown in the previous section, treatments affected the mechanical properties of fibres. Indeed, fibre

Fig. 9. Fibre stiffness evolution with increasing load for fibres with different treatments.

stiffness and strength were decreased and in order to comprehend the mechanism by which properties were altered, it was necessary to study the fibre stiffness evolution/degradation after each straining step. Fig. 9 shows representative evolution of the stiffness with respect to the applied load for treated and reference fibres. It should be noted that results are presented in normalized form (normalization is done with respect to the stiffness value E0 obtained in the first loading ramp at low strain <0.5%). The curves shown in Fig. 9 can be divided into three different regions with respect to the applied strain ␧: i) ␧ = 0–1.5%; ii) ␧ = 1.5–3.5%; iii) ␧ = >3.5% – until failure. The stiffness of all fibres increases within Region-1 (up to ␧ = 1.5%), more significant increase is observed for treated fibres compare to the reference material. Then stiffness drops in the Region-2 followed by stabilisation afterwards for untreated fibres and stays at almost the same level until fibres fail. It is worth noting that the transition between the first and second region corresponds to yielding where the behaviour changes from viscoelastic to viscoplastic. However, the stiffness of the treated fibres increases after yielding and stabilise until failure at a value 35% higher than the initial modulus E0 . This behaviour is exhibited by all treated fibres (by both MPS and CNC) indicating that it is not caused by depositing cellulose nano-crystals on fibres. Hence, it can be concluded that the depositing CNC does not affect the microstructure of fibres. Rather, the changes in the non-crystalline and crystalline phases are induced during the treatment by MPS. It is a reasonable assumption, since the treatment by MPS was performed in ethanol/water solvent and at elevated temperature (65 ◦ C). At these conditions, solvent diffuses into the fibre structure and likely to promote the breakdown of inter-chain interactions giving rise to a disordered structure where the crystalline and amorphous regions are reoriented with respect to the fibre axis (misorientation). Upon straining, the crystalline and noncrystalline domains in a treated fibre, which are less oriented parallel to the loading direction, experience higher reorientation compared to untreated fibre. Due to reorientation, the stiffness of the fibre increases with increasing the applied strain (Northolt and Baltussen, 2002). According to simulation study, when amorphous cellulose is loaded, inter-chain hydrogen bonds of cellulose break and only one third of the newly formed hydrogen bonds recover to their old conformation upon releasing the load (Chen et al., 2004). Therefore, the reorientation of the amorphous phase is largely persistent, as the fibre stiffness increases, which was the case in the current study.

A. Hajlane et al. / Industrial Crops and Products 100 (2017) 41–50

When crystalline and non-crystalline domains experience reorientation, the amorphous phases (randomly oriented cellulose), compared to crystalline domains, receive the largest part of the reorientation and this is due to their low elastic modulus (Gindl and Keckes, 2006). Indeed, in the current study, the misorientation detected by WAXD in treated fibres would not alone explain the increase in fibre stiffness upon straining as observed in cyclic loading experiment. 4. Conclusions Cellulose nano-crystals were successfully deposited onto the surface of Cordenka 700 Super 3 fibres using MPS as coupling agent. A route with less effect on the environment was investigated to polymerize the coupling agent by creating free radicals on cellulose backbone. It was possible to retain fibre length (continuous filaments) in order to be used in composites for high performance applications. The nano-crystal network was created and deposited on fibre surface along with poly-MPS particles as validated by SEM. Multi-scale hierarchical fibres for structural composites were created by depositing cellulose nano-crystals onto regenerated cellulose fibres. This enables recovering the loss of stiffness and the failure strain lowered down after the chemical treatment by MPS. On the other hand, a significant increase of stiffness during the successive uniaxial cyclic loadings has been observed for the treated fibres in comparison to untreated ones. This increase is due to the misorientation occurred in the amorphous phase of RCF during the chemical treatment. Acknowledgments The authors would like to thank Hassan II Academy of Science and Technology the Moroccan National Centre for Scientific and Technical Research, the Swedish Research Council (ref. nr. 20096433), and EXCEL project (funded by local government Norrbotten, SWEDEN) for their financial supports. References ˇ Sturcová, A., Davies, G.R., Eichhorn, S.J., 2005. Elastic modulus and stress-transfer properties of tunicate cellulose whiskers. Biomacromolecules 6, 1055–1061. Abdelmouleh, M., Boufi, S., ben Salah, A., Belgacem, M.N., Gandini, A., 2002. Interaction of silane coupling agents with cellulose. Langmuir 18, 3203–3208. Abdelmouleh, M., Boufi, S., Belgacem, M., Duarte, A., Salah, A.B., Gandini, A., 2004. Modification of cellulosic fibres with functionalised silanes: development of surface properties. Int. J. Adhesion Adhesives 24, 43–54. Angles, M.N., Salvadó, J., Dufresne, A., 1999. Steam-exploded residual softwood-filled polypropylene composites. J. Appl. Polym. Sci. 74, 1962–1977. Azubuike, C.P., Rodríguez, H., Okhamafe, A.O., Rogers, R.D., 2012. Physicochemical properties of maize cob cellulose powders reconstituted from ionic liquid solution. Cellulose 19, 425–433. Bach, S., Belgacem, M.N., Gandini, A., 2005. Hydrophobisation and densification of wood by different chemical treatments. Holzforsch 59, 389–396. Bendahou, A., Kaddami, H., Sautereau, H., Raihane, M., Erchiqui, F., Dufresne, A., 2008. Short palm tree fibers polyolefin composites: effect of filler content and coupling agent on physical properties. Macromol. Mater. Eng. 293, 140–148. Bendahou, A., Habibi, Y., Kaddami, H., Dufresne, A., 2009. Physico-chemical characterization of palm from phoenix dactylifera-L, preparation of cellulose whiskers and natural rubber-based nanocomposites. J. Biobased Mater. Bioenergy 3, 81–90. Bendahou, A., Hajlane, A., Dufresne, A., Boufi, S., Kaddami, H., 2014. Esterification and amidation for grafting long aliphatic chains on to cellulose nanocrystals: a comparative study. Res. Chem. Intermed., 1–18. Botaro, V.R., Gandini, A., 1998. Chemical modification of the surface of cellulosic fibres. 2. Introduction of alkenyl moieties via condensation reactions involving isocyanate functions. Cellulose 5, 65–78. Botaro, V.R., Gandini, A., Belgacem, M.N., 2005. Heterogeneous chemical modification of cellulose for composite materials. J. Thermoplast. Compos. Mater. 18, 107–117. Caulfield, D.F., Feng, D., Prabawa, S., Young, R., Sanadi, A.R., 1999. Interphase effects on the mechanical and physical aspects of natural fiber composites. Angew. Makromol. Chem. 272, 57–64.

49

Chen, W., Lickfield, G.C., Yang, C.Q., 2004. Molecular modeling of cellulose in amorphous state. Part I: model building and plastic deformation study. Polymer 45, 1063–1071. Dai, D., Fan, M., Collins, P., 2013. Fabrication of nanocelluloses from hemp fibers and their application for the reinforcement of hemp fibers. Ind. Crops Prod. 44, 192–199. Devi, L.U., Bhagawan, S., Thomas, S., 1997. Mechanical properties of pineapple leaf fiber-reinforced polyester composites. J. Appl. Polym. Sci. 64, 1739–1748. Felix, J.M., Gatenholm, P., 1991. The nature of adhesion in composites of modified cellulose fibers and polypropylene. J. Appl. Polym. Sci. 42, 609–620. Gandini, A., Botaro, V., Zeno, E., Bach, S., 2001. Activation of solid polymer surfaces with bifunctional reagents. Polym. Int. 50, 7–9. Gassan, J., Bledzki, A.K., 1999. Effect of cyclic moisture absorption desorption on the mechanical properties of silanized jute-epoxy composites. Polym. Composites 20, 604. Gaylord, N., 1972. A proposed new mechanism for catalyzed and uncatalyzed graft polymerization onto cellulose. J. Polym. Sci. Part C: Polym. Symp., 153–172, Wiley Online Library. Gindl, W., Keckes, J., 2006. Strain hardening in regenerated cellulose fibres. Composites Sci. Technol. 66, 2049–2053. Habibi, Y., Lucia, L.A., Rojas, O.J., 2010. Cellulose nanocrystals: chemistry, self-assembly, and applications. Chem. Rev. 110, 3479–3500. Habibi, Y., 2014. Key advances in the chemical modification of nanocelluloses. Chem. Soc. Rev. 43, 1519–1542. Hajlane, A., Kaddami, H., Joffe, R., Wallström, L., 2013. Design and characterization of cellulose fibers with hierarchical structure for polymer reinforcement. Cellulose 20, 2765–2778. Hassan, M.L., Nada, A.A., 2003. Utilization of lignocellulosic fibers in molded polyester composites. J. Appl. Polym. Sci. 87, 653–660. Hermans, P., Hermans, J., Vermaas, D., Weidinger, A., 1948. Deformation mechanism of cellulose gels. IV. General relationship between orientation of the crystalline and that of the amorphous portion. J. Polym. Sci. 3, 1–9. Kaddami, H., Dufresne, A., Khelifi, B., Bendahou, A., Taourirte, M., Raihane, M., Issartel, N., Sautereau, H., Gerard, J.-F., Sami, N., 2006. Short palm tree fibers-thermoset matrices composites. Composites—Part A: Appl. Sci. Manuf. 37, 1413–1422. Kong, K., Eichhorn, S., 2005. The influence of hydrogen bonding on the deformation micromechanics of cellulose fibers. J. Macromol. Sci. Part B: Phys. 44, 1123–1136. Li, R., Chang, C., Zhou, J., Zhang, L., Gu, W., Li, C., Liu, S., Kuga, S., 2010. Primarily industrialized trial of novel fibers spun from cellulose dope in NaOH/urea aqueous solution. Ind. Eng. Chem. Res. 49, 11380–11384. Li, R., Lachman, N., Wagner, H.D., Wardle, B.L., 2015. Hierarchical carbon nanotube carbon fiber unidirectional composites with preserved tensile and interfacial properties. Compos. Sci. Technol. Lin, N., Huang, J., Dufresne, A., 2012. Preparation, properties and applications of polysaccharide nanocrystals in advanced functional nanomaterials: a review. Nanoscale 4, 3274–3294. McDowall, D., Gupta, B., Stannett, V., 1984. Grafting of vinyl monomers to cellulose by ceric ion initiation. Prog. Polym. Sci. 10, 1–50. Moharana, S., Mishra, S., Tripathy, S., 1990. Chemical modification of jute fibers. I. Permanganate-initiated graft copolymerization methyl methacrylate onto jute fibers. J. Appl. Polym. Sci. 40, 345–357. Northolt, M., Baltussen, J., 2002. The tensile and compressive deformation of polymer and carbon fibers. J. Appl. Polym. Sci. 83, 508–538. Peng, B., Dhar, N., Liu, H., Tam, K., 2011. Chemistry and applications of nanocrystalline cellulose and its derivatives: a nanotechnology perspective. Can. J. Chem. Eng. 89, 1191–1206. Pothan, L.A., Thomas, S., Neelakantan, N., 1997. Short banana fiber reinforced polyester composites: mechanical, failure and aging characteristics. J. Reinf. Plast. Compos. 16, 744–765. Qian, H., Greenhalgh, E.S., Shaffer, M.S., Bismarck, A., 2010. Carbon nanotube-based hierarchical composites: a review. J. Mater. Chem. 20, 4751–4762. Raj, R., Kokta, B., Grouleau, G., Daneault, C., 1990. The influence of coupling agents on mechanical properties of composites containing cellulosic fillers. Polym. Plast. Technol. Eng. 29, 339–353. Roncero, M.B., Torres, A.L., Colom, J.F., Vidal, T., 2005. The effect of xylanase on lignocellulosic components during the bleaching of wood pulps. Bioresour. Technol. 96, 21–30. Sbiai, A., Kaddami, H., Fleury, E., Maazouz, A., Erchiqui, F., Koubaa, A., Soucy, J., Dufresne, A., 2008. Effect of the fiber size on the physicochemical and mechanical properties of composites of epoxy and date palm tree fibers. Macromol. Mater. Eng. 293, 684–691. Sbiai, A., Maazouz, A., Fleury, E., Souterneau, H., Kaddami, H., 2010. Short date palm tree fibers/polyepoxy composites prepared using RTM process: effect of tempo mediated oxidation of the fibers. BioResources 5, 672–689. Sbiai, A., Kaddami, H., Sautereau, H., Maazouz, A., Fleury, E., 2011. TEMPO-mediated oxidation of lignocellulosic fibers from date palm leaves. Carbohydr. Polym. 86, 1445–1450. Segal, L., Creely, J., Martin, A., Conrad, C., 1959. An empirical method for estimating the degree of crystallinity of native cellulose using the X-ray diffractometer. Text. Res. J. 29, 786–794. Singh, B., Verma, A., Gupta, M., 1998. Studies on adsorptive interaction between natural fiber and coupling agents. J. Appl. Polym. Sci. 70, 1847–1858.

50

A. Hajlane et al. / Industrial Crops and Products 100 (2017) 41–50

Smole, M.S., Perˇsin, Z., Kreˇze, T., Kleinschek, K.S., Ribitsch, V., Neumayer, S., 2003. X-ray study of pre-treated regenerated cellulose fibres. Mater. Res. Innov. 7, 275–282. Tong, L., Mouritz, A.P., Bannister, M., 2002. 3D Fibre Reinforced Polymer Composites. Elsevier. Wang, Y., Backer, S., Li, V.C., 1988. A special technique for determining the critical length of fibre pull-out from a cement matrix. J. Mater. Sci. Lett. 7, 842–844. Wei, J., Chen, Y., Liu, H., Du, C., Yu, H., Ru, J., Zhou, Z., 2016. Effect of surface charge content in the TEMPO-oxidized cellulose nanofibers on morphologies and properties of poly (N-isopropylacrylamide)-based composite hydrogels. Ind. Crops Prod. 92, 227–235.

Yilmaz, E., Adali, T., Yilmaz, O., Bengisu, M., 2007. Grafting of poly (triethylene glycol dimethacrylate) onto chitosan by ceric ion initiation. React. Funct. Polym. 67, 10–18. Ying, J.Y., Benziger, J.B., Navrotsky, A., 1993. Structural evolution of alkoxide silica gels to glass: effect of catalyst pH. J. Am. Ceram. Soc. 76, 2571–2582. Zhang, J., Yuan, Y., Shen, J., Lin, S., 2003. Synthesis and characterization of chitosan grafted poly (N: N-dimethyl-N-methacryloxyethyl-N-(3-sulfopropyl) ammonium) initiated by ceric (IV) ion. Eur. Polym. J. 39, 847–850. Zhou, C., Wu, Q., Yue, Y., Zhang, Q., 2011. Application of rod-shaped cellulose nanocrystals in polyacrylamide hydrogels. J. Colloid Interface sci. 353, 116–123.