Effect of hemp surface modification on the

3 downloads 0 Views 2MB Size Report
Feb 18, 2016 - Natural fiber reinforced composites (NFC) are re-emerging as alternatives ...... Chemical modification of hemp, sisal, jute, and kapok fibers by.
Composite Interfaces

ISSN: 0927-6440 (Print) 1568-5543 (Online) Journal homepage: http://www.tandfonline.com/loi/tcoi20

Effect of hemp surface modification on the morphological and tensile properties of linear medium density polyethylene (LMDPE) composites Désiré Yomeni Chimeni, Jean Luc Toupe, Charles Dubois & Denis Rodrigue To cite this article: Désiré Yomeni Chimeni, Jean Luc Toupe, Charles Dubois & Denis Rodrigue (2016): Effect of hemp surface modification on the morphological and tensile properties of linear medium density polyethylene (LMDPE) composites, Composite Interfaces To link to this article: http://dx.doi.org/10.1080/09276440.2016.1144163

Published online: 18 Feb 2016.

Submit your article to this journal

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=tcoi20 Download by: [Universite de Lorraine]

Date: 18 February 2016, At: 11:27

Composite Interfaces, 2016 http://dx.doi.org/10.1080/09276440.2016.1144163

Effect of hemp surface modification on the morphological and tensile properties of linear medium density polyethylene (LMDPE) composites Désiré Yomeni Chimenia, Jean Luc Toupea, Charles Duboisb and Denis Rodriguea Department of Chemical Engineering and CERMA, Université Laval, Quebec City, Canada; bDepartment of Chemical Engineering, Polytechnique Montréal, Succursale Centre-Ville, Montréal, Canada

Downloaded by [Universite de Lorraine] at 11:27 18 February 2016

a

ABSTRACT

This work investigates different hemp surface modifications (mercerization, maleated polyethylene (MAPE) addition in solution or in melt blending) to improve the properties of linear medium density polyethylene (LMDPE). From the composites produced, a complete morphological and tensile characterization was performed for a fixed hemp content (30% wt.). The morphological analysis showed that both the direct (melt blending) and solution modifications were able to significantly improve the composites interface quality and therefore the tensile properties (151% increase in modulus and 36% increase in strength over the neat matrix) within the range of conditions tested.

ARTICLE HISTORY

Received 31 August 2015 Accepted 16 January 2016 KEYWORDS

LMDPE; hemp; mercerization; MAPE; solution modification; morphological properties; mechanical properties

1. Introduction Natural fiber reinforced composites (NFC) are re-emerging as alternatives for synthetic fiber reinforced composites in automotive, building, and construction industries.[1] These composites are made from lignocellulosic fibers with polymer matrices. The main idea behind the introduction of natural fibers in a polymer is to reduce raw material costs and environmental issues while producing materials with good mechanical properties. In addition to being relatively low cost, lightweight, biodegradable, and non-abrasive, natural fibers offer good specific properties, are readily available, and do not cause health problems. [2–10] Among all the available natural fibers, hemp is known for its superior strength, durability, resistance to rot, and low lignin content. In a composite, hemp is twice as strong as common wood fibers.[11] For all these reasons, the use of hemp as reinforcement in plastic composites is well justified. But to expand into markets other than automotive (i.e. commercial construction and consumer goods), NFC must achieve high-quality performance, serviceability, durability, and reliability standards.[4] Over the last decade, several studies have been conducted to develop and characterize NFC. For example, the parameters affecting the tensile modulus of short fiber composites are wetting, aspect ratio, and fiber concentration, while those

CONTACT  Denis Rodrigue  © 2016 Taylor & Francis

[email protected]

Downloaded by [Universite de Lorraine] at 11:27 18 February 2016

2 

  D. Y. Chimeni et al.

affecting tensile strength are stress concentration, fiber orientation, and interface quality.[2] The main challenge today is to optimize NFC properties. This can be done by optimizing NFC processing conditions, composition,[12] and morphology. Unfortunately a lack of chemical compatibility between hydrophobic thermoplastic matrices and hydrophilic natural fibers often leads to poor adhesion between both components, hence weak interfacial bonding produces composites with limited mechanical properties.[3,13] Most importantly, fiber dispersion and wetting by the polymer must be improved to enhance fiber-matrix adhesion. On the other hand, the presence of impurities such as waxes at the fibers’ surface are responsible for low fiber wettability by the polymer,[14–16] which increases the compatibility problem. Several studies focused on matrix modification or fiber physical and/or chemical treatment to improve interfacial wettability and adhesion in NFC.[7,14,17] These methods can be divided in two main categories: pretreatment and treatment. The pretreatment cleans and removes some hemicelluloses and lignins from the fibers to allow an easy and better access of the high amount of hydroxyl groups in cellulosic fibers.[14,18,19] On the other hand, the treatment uses a polar agent (coupling agent, etc.) creating covalent bonds between the hydroxyl groups of the fibers and a physical bond with the polymer matrix by chain entanglement.[6] By using this two-step process, improvement of fiber dispersion (reduction of fiber-fiber interactions) and strength of the fiber/matrix interface can be achieved. But most of the modification methods present some limitations weakening the properties of the final product. For example, the use of intensive mixing to obtain good fiber dispersion in the matrix is responsible for high fiber attrition (reduced lengths), while not improving the adhesion with the polymer matrix.[20] Thermal methods can also improve the interfacial properties, but as noted by Qi et al. [21], this method has a major limitation which is fiber degradation because of heat (thermo-oxidation). Priyanka et al. [22] have shown that the use of functionalized polyolefins helps to produce composites with improved interfacial properties without using a plasticizer or fiber pretreatment. Although this method is innovative, it is complex because the fibers are easier and less expensive to treat than the matrix. The use of coupling agents to improve compatibility is the most widely used approach and in this case, the coupling agent is directly added to the polymer during the mixing process (extrusion). But, when a coupling agent is directly mixed with a polymer via melt blending, dispersion is not guaranteed and fiber agglomeration will lead to poor composite properties. Therefore, development of a simple and economic modification process leading to the production of stronger NFC is important. One approach to prevent all these limitations is to increase the amount of active sites (hydroxyl groups) on the fiber surface (by pretreatment) and modify these pretreated fibers before their introduction into a matrix. Recently, this possibility was investigated by Verdaguer and Rodrigue [23] and Raymond and Rodrigue [24] to modify mercerized wood fibers with a compatibilizer (MAPE) in solution (1,2,4-trichlorobenzene). This work is thus a continuing effort to further understand this type of modification. The main objective of this work is to compare different hemp surface modifications and to propose the mechanisms leading to fiber surface modification in solution. To achieve this goal, several characterizations are combined to determine the efficiency of each modification. Then, the tensile properties of the composites are reported and discussed in terms of improvement level.

Composite Interfaces 

 3

Table 1. Codes and compositions of the samples. Codes LMDPE UT UTE3S UTE3D TN TNE3S TNE6S TNE9S TNE3D TNE3S&D

Compositions (fibers codes) Linear medium density polyethylene LMDPE/neat hemp (UH) LMDPE/neat hemp + 3% MAPE in solution (UHS1) LMDPE/neat hemp + 3% MAPE direct mixing LMDPE/mercerized hemp (UT) LMDPE/mercerized hemp + 3% MAPE in solution (AHS1) LMDPE/mercerized hemp + 6% MAPE in solution (AHS2) LMDPE/mercerized hemp + 9% MAPE in solution (AHS3) LMDPE/mercerized hemp + 3% MAPE direct mixing LMDPE/mercerized hemp + 3% MAPE (1.5% in solution + 1.5% direct mixing) (AHS&D)

Downloaded by [Universite de Lorraine] at 11:27 18 February 2016

2. Experimental 2.1. Materials Hemp fibers were obtained from the Hemp Trade Alliance (Quebec, Canada) and sieved. Only the fraction between 60 and 18 mesh Tyler (250 and 1000 μm) was kept. Sodium hydroxide (ACS grade) was obtained from Aldrich Chemicals (Canada) and used without any further purification. The coupling agent used was a maleic anhydride grafted high-density polyethylene (HDPE-g-MAH or MAPE) supplied by Westlake Chemical Co (USA) under the trade name Epolene E-20 (Mw = 7500 g/mol, MFI = 1.24 g/10 min (190 °C/2.16 kg), acid number = 16.9 mg-KOH and a softening point of 113.8 °C). The solvent used was 1,2,4-trichlorobenzene (TCB) HPLC grade from J.T. Baker (USA) and used as received. The matrix used was linear medium density polyethylene (LMDPE) Hival 103538 (Tm = 125 °C, MFI = 3.5 g/10 min (190 °C/2.16 kg), density = 0.936 g/cm3) obtained from Ashland (Canada). 2.2.  Chemical modifications of hemp 2.2.1.  Alkaline pretreatment or mercerization The hemp fibers were pretreated with 8% w/v NaOH solution at room temperature (23 °C) during 3 h. The solid:solution ratio was 1:10 w/v. The alkaline-treated fibers were then washed with distilled water until neutral pH and then dried at 80 °C for 24 h. 2.2.2.  Treatment of hemp with MAPE in solution The TCB solution containing the chosen amount of MAPE (Table 1) was raised to 160 °C with stirring until complete MAPE dissolution. The solution was then cooled down between 80 and 90 °C before the hemp fibers (mercerized or not) were fed in the solution under stirring for 30 min. Finally, the solution modified hemp was filtered and dried in an oven at 80 °C for 48 h. 2.2.3.  Treatment of hemp with MAPE (direct mixing) MAPE direct treatment was performed by dry-mixing of the LMDPE powder with MAPE powder prior to extrusion. All the composites investigated contain 30% wt. of hemp and the amount of MAPE is based on the total amount of material in the compound. The codes and compositions of the different samples and the corresponding fibers investigated are presented in Table 1.

4 

  D. Y. Chimeni et al.

2.3.  Composites fabrication

Downloaded by [Universite de Lorraine] at 11:27 18 February 2016

First, the fibers were oven-dried overnight at 70  °C. To prepare the composites (hemp + LMDPE), a Haake twin-screw extruder Rheomex PTW 16 OS, (L/D = 25) was used with a die diameter of 3.2 mm at a speed of 80 rpm producing a total mass flow rate of 0.5 kg/h. The temperatures profile for the different heating zones of the extruder was set as 150–150–150–150–155–155 °C from the feed hopper to the die. The compounds at the extruder exit were cooled in a water bath and pelletized. Thereafter, the pellets were dried at 80 °C in a heated vacuum oven for 24 h to be molded in a Nissei PS-E injection machine. The injection temperature profile was set as 180–170–170–160 °C (for the nozzle, front, middle, and rear, respectively) with a mold temperature of 30 °C. The mold has four cavities, two dumbbell shapes (type IV of ASTM D638), and two rectangular bars (width and thickness of 12.45 and 3.14 mm with two lengths of 80 and 125 mm). 2.4. Characterizations 2.4.1.  Fourier transform infrared spectroscopy (FTIR) Infrared spectra were obtained with a Nicolet FTIR spectrometer (model 730, Nicolet Instruments, USA) equipped with a mercury-cadmium-telluride detector. The sample absorbance was measured in the IR region 4000–750 cm−1. Each spectrum was obtained from 128 scans at a resolution of 4 cm−1. All spectral operations were executed using the GRAMS/AI 8.0 software (Thermo Galactic, USA). 2.4.2.  Morphological investigation Unmodified and modified hemp, as well as the composites (subjected to cryogenic fracture in liquid nitrogen) were examined using a scanning electron microscope (SEM) JEOL JSM840A at various magnifications. The samples were first coated with a thin layer of gold/ palladium and then examined at 15 kV. 2.4.3.  Thermogravimetric analysis (TGA) Thermogravimetric analysis (TGA) was performed on a model Q5000IR of TA Instruments (USA). Between 6 and 10 mg of material was analyzed by heating up steadily at a rate of 10 °C/min from 50 to 700 °C in nitrogen. 2.4.4.  Density measurements Density data were obtained by a gas pycnometer ULTRAPYC 1200e (Quantachrome Instruments, USA) using nitrogen as the gas phase. The data reported are the average of five measurements while standard deviations were less than 1%. 2.4.5.  Tensile test The tensile properties were measured using an Instron model 5565 universal testing machine (Instron, USA) with a 500 N load cell. Dog-bone shaped samples were prepared according to type IV of ASTM D638. The following conditions were used to perform the characterization: a temperature of 23 °C and a crosshead speed of 5 mm/min. A minimum of seven samples was tested to get an average and standard deviation.

Downloaded by [Universite de Lorraine] at 11:27 18 February 2016

Composite Interfaces 

 5

Figure 1. Infrared spectra of neat hemp, mercerized hemp, and mercerized hemp + 3% MAPE in solution.

Figure 2. Typical SEM micrographs of hemp: (a) untreated hemp; (b) mercerized hemp; (c) mercerized hemp + 3% MAPE in solution and (d) mercerized hemp + 6% MAPE in solution.

  D. Y. Chimeni et al.

Downloaded by [Universite de Lorraine] at 11:27 18 February 2016

6 

Figure 3. (a) TGA curves of the neat, mercerized, and mercerized hemp modified by 6% MAPE in solution and (b) DTG curves of the neat, mercerized, and mercerized hemp modified by 6% MAPE in solution. Table 2. Density of the different hemp. Samples Density (g/cm3)

UH 1.37

UHS1 1.16

AH 1.48

AHS1 1.44

AHS2 1.46

AHS3 1.20

AHS&D 1.39

Note: See Table 1 for sample codes.

3.  Results and discussion 3.1.  Confirmation of hemp modifications FTIR, SEM, TGA, and density tests were used to determine the level of surface modification on the hemp fiand after mercerization and solution treatment. Figure 1 presents the FTIR

Downloaded by [Universite de Lorraine] at 11:27 18 February 2016

Composite Interfaces 

 7

Figure 4. Proposed mechanisms for the surface modification of hemp.

spectra where major differences are seen on the intensity of the following peaks: 3325, 1732, 1506, 1011, 873, and 778 cm−1. The intensity of the broad peak around 3325 cm−1 (–OH) and the peak at 1011 cm−1 (C–C stretching),[1,14] are significantly reduced after mercerization showing that the amount of free –OH groups and C–C bonds decreased. This trend was also observed for mercerized hemp by Kabir et al. [5] as well as Mwaikambo and Ansell [14]. This change is attributed to hemicelluloses and lignins removal. The peak around 1732 cm−1 belongs to the C=O (carbonyl) stretching of the acetate group in hemicelluloses, and carbonyl of pectin and waxes. [1,3,14] This peak is significantly reduced after mercerization indicating that the removal of these elements was successful. The high reduction of the peak at 873 cm−1 corresponding to aromatic C–H,[14] and the disappear peak at 778 cm−1 also assigned to aromatic C–H,[25]

Downloaded by [Universite de Lorraine] at 11:27 18 February 2016

8 

  D. Y. Chimeni et al.

are other evidences of lignins removal after mercerization. But the peak at 1507 cm−1 (C=C stretching), generally associated to lignins,[26] is still visible after mercerization, suggesting only a partial removal of this fiber component. By comparing the spectra of mercerized hemp and mercerized hemp modified by 3% MAPE, the main differences are seen on the intensity of the following peaks: 3325, 2920–2850, 1543, 1373, 1154, 1011, and 895 cm−1. On these spectra, an increase of the hydroxyl peak at 3325 cm−1 can be seen after the modification. This behavior was also reported on wood fibers by Verdaguer and Rodrigue [23] and Raymond and Rodrigue [24] after using the same type of modification. This suggests an additional amount of –OH groups at the surface of those fibers. In addition, a new peak appears at 1373 cm−1 while the peak at 1154 cm−1 increases in intensity after MAPE treatment. The increase of –OH groups and the appearance of the peak at 1373 cm−1 suggest the formation of a tertiary alcohol (C–OH) not yet reported in the literature. The peak at 1154 cm−1 belongs to C–O–C bonds,[27,28] and its increase suggests higher amounts of C–O–C bonds, probably due to some link being created between MAPE and mercerized hemp through the Cell–O–C bonds (see the reaction mechanism proposed later). The intensity increase at 1011 cm−1, as well as at 2920 and 2850 cm−1 belonging to C–C stretching, as well as symmetric and asymmetric aliphatic C–H vibration, respectively,[14,25] confirms the increase of C–C and C–H groups at the hemp surface after solution modification, probably due to the presence of PE in MAPE. All these observations indicate the presence of the coupling agent at the fiber surface. To better see the surface modification of hemp fibers, SEM was used and the results are presented in Figure 2. Figure 2(a) shows the surface of neat hemp, which is very rough and covered by non-cellulosic materials (impurities). The mercerized hemp (Figure 2(b)) has a smoother surface as reported by Aziz et al. [17]. They explained this by the removal of the waxy epidermal tissue, adhesive pectins, and hemicelluloses. On the other hand, the modification of mercerized hemp with 3 or 6% MAPE in solution leads to fiber coating by a thin MAPE layer as seen in Figure 2(c) and (d), respectively. The presence of this coupling agent layer was also reported by Verdaguer and Rodrigue [23] and Raymond and Rodrigue [24] for modified maple wood fibers. The effect of mercerization and the modification of mercerized hemp by MAPE in solution can also be evaluated through the modification of the thermal stability of the corresponding fibers. The TGA and DTG curves of neat, mercerized and mercerized hemp modified by MAPE in solution are shown in Figure 3(a) and (b), respectively. The TGA curves (Figure 3(a)) show that, below 300  °C, mercerization increases the stability of the corresponding hemp. Such trend has already been observed for hemp fibers by Kabir et al. [5], after their modification with different concentration of alkaline solutions. The modification of mercerized hemp with MAPE in solution further increases the stability with respect to neat hemp. In the case of mercerized hemp, increased stability could be attributed to the reduction of hemicelluloses and lignins content leading to a more compact (dense) interlamellar region (see density data presented below) which needed more energy to be broken, hence improved thermal stability below 300 °C. Additional increase in thermal stability after MAPE modification in solution (Figure 3(a)) is another confirmation of the effectiveness of the modification with MAPE in solution. This can be explained by the presence of a thin layer of MAPE (higher thermal stability) at the surface of the fibers. This layer acted as a barrier enhancing the overall thermal stability of the system. This explanation is confirmed by the DTG results presented in Figure 3(b).

Downloaded by [Universite de Lorraine] at 11:27 18 February 2016

Composite Interfaces 

 9

Figure 3(b) reveals that mercerized hemp modified with MAPE in solution presents an additional peak at 448 °C, corresponding to the thermal degradation of MAPE, which is absent in the neat and the mercerized hemp DTG curves. This again confirms the presence of MAPE on the fiber surface. In the 300–400 °C temperature range, the cellulosic interlamellar bonds are broken and the cellulosic fibrils are exposed leading to a degradation peak of the cellulose around 318 °C for neat hemp. This peak is slightly reduced to 315 °C after mercerization and then to 310 °C after the modification of mercerized hemp with MAPE in solution. These observations suggest a slight degradation of the cellulose during mercerization and more after their modification with MAPE in solution due to the conditions used (hot solvents). Another evidence of hemp modification by either mercerization or MAPE in solution can be seen on their density as presented in Table 2. The results show that the density of neat hemp after solution modification (UHS1) drops drastically. Two reasons can explain this behavior: (1) the density difference between the removed components (lignins and hemicelluloses around 1.5 g/cm3) and MAPE addition (0.96 g/cm3), and (2) a degradation of neat hemp due to the hot solvent. The density of mercerized hemp (AH) increases by about 8% with respect to neat hemp (UH). This trend has also been reported by Sawpan et al. [17] and Mwaikambo and Ansell [14]. They attributed this increase to the densification of mercerized hemp cell walls as a result of the removal of impurities (fats, waxes, etc.). The modification of mercerized hemp in solution with 3 as well as 6% of MAPE (AHS1, AHS2) leads to a slight decrease of hemp density compared to mercerized hemp (AH), but the values remaining higher (about 5 and 7%, respectively) compared to neat hemp (UH). This shows that during the solution modification, extra lignin and hemicelluloses are removed while the reaction between hemp and MAPE leads to fiber coating by a thin layer of MAPE with a density lower than that of the removed elements; hence density decreases compared to AH. The 5% increase with respect to UH confirms this explanation by showing that the density increase associated to the densification of the fibers cell walls due to mercerization was more important than their reduction due to the removal of high density component (lignins and hemicelluloses) and/or fibers degradation. When 9% of MAPE was used in solution, the corresponding mercerized hemp modified (AHS3) presents a significant density decrease with respect to AH or UH. In this case, it seems that higher amount of MAPE in solution produced an inverse effect than in the previous cases (AHS1 and AHS2); i.e. for AHS3 the effect associated to the density increase (densification of the fibers cell walls after mercerization) was less important than the density reduction (removal of high-density component (lignins and hemicelluloses) and/or fiber degradation). From the FTIR, SEM, TGA, and density results, a reaction mechanism can be proposed to explain the changes induced by the different treatments. These modifications can be divided into three main steps: mercerization (Step 1), washing to neutralize the alkaline molecules at the hemp surface (Step 2), and the modification of mercerized hemp by MAPE in solution (Step 3). The overall proposed mechanism is presented in Figure 4. In Figure 4 (Step 3), R is a given alkyl group ( ) and represents the HDPE backbone. The structure of MAPE was assumed from its FTIR spectrum (not present here) and the work of Lu et al. [29] reporting that Epolene E20 is a carboxylated HDPE copolymer. From Figure 4, Step 1 (mercerization) is well-known and explained in the literature.[25,30] It is possible to take into account two different cellulose structures (i.e. structure I and II), which can be formed after the modification of mercerized hemp with MAPE in solution

Downloaded by [Universite de Lorraine] at 11:27 18 February 2016

10 

  D. Y. Chimeni et al.

(Figure 4 (Step 3)). Structure II is not in agreement with the FTIR results. From Figure 4 (Step 3), structure II suggests the formation of carbonyl (C=O) groups after modification, while FTIR results for hemp (Figure 1) do not show additional carbonyl groups since the absorption band at 1732 cm−1 for mercerized hemp and mercerized hemp modified in solution are identical. Moreover, the FTIR spectrum of mercerized hemp modified in solution shows an increase of hydroxyl (OH–) group, which do not match this structure. On the contrary, structure I is more likely because it matches the FTIR and SEM results. Indeed, from this structure, additional C–O–C bonds revealed through covalent bonds (Cell–O–C in Figure 4 (Step 3)) is in agreement with the increased peak at 1154 cm−1 of the FTIR spectrum of mercerized hemp modified by MAPE in solution. Moreover, the new peak at 1373 cm−1 corresponding to tertiary –OH revealed by FTIR of mercerized hemp modified in solution is in agreement with the increase of –OH peak at 3325 cm−1 (Figure 1). In addition, structure I is in agreement with the modified hemp SEM picture (Figures 2(c) and (d)) and FTIR (Figure 1) peaks at 2920 and 2850 cm−1 which, respectively, show the presence of a thin layer of MAPE coating the hemp surface. Finally, the unchanged carbonyl absorption peak (1732 cm−1) after modification in solution is in agreement with this structure since there is no additional carbonyl after the reaction. Therefore, the mechanism of hemp modification in solution is the one via structure I in Figure 4 (Step 3). 3.2.  Morphological properties The state of dispersion, wetting, and adhesion of the different hemp inside the polymer matrix was analyzed using SEM. Micrographs were taken at different magnifications and typical results are presented in Figure 5. Figure 5(a) corresponds to the composite based on neat hemp (UT) and reveals debonding areas between the fibers and the matrix (arrows) and holes (circle) resulting from fiber pull-out. This behavior is typical in composites where poor fiber wettability and adhesion occurs because the fibers cannot sustain the applied load and are easily mechanically extracted.[31] This can be related to the presence of impurities and rough hemp surface (Figure 2(a)) as the presence of these elements reduced wettability toward LMDPE. The composite with mercerized hemp (TN) (Figure 5(b)) shows a more homogeneous surface with less gaps and voids. This can be due to increased wettability. As reported by SEM (Figure 2(b)), mercerization led to a smoother surface; i.e. higher contact area (fiber surface). But the presence of some gaps in Figure 5(b) (circle) still suggests fiber pull-out indicating low compatibility. On the contrary, all the composites with MAPE (solution modification in Figure 5(c), direct use in Figure 5(e), or combined direct/solution use in Figure 5d) present a more homogeneous surface with almost no holes and hemp fibers perfectly anchored in the matrix (circles). Moreover, the images do not present any hemp agglomeration indicating good dispersion. This shows that, in addition to better wettability (contact) observed after mercerization, there is better dispersion and compatibility for the composites with mercerized hemp modified by MAPE. This can be explained by the presence of MAPE for which the carbonyl groups are covalently link with the hydroxyl groups of hemp, while the nonpolar part (HDPE) become compatible with the matrix (by chain entanglement). Such effect was also reported by Mohanty et al. [6].

 11

Downloaded by [Universite de Lorraine] at 11:27 18 February 2016

Composite Interfaces 

Figure 5. Typical SEM micrographs of the composites based on: (a) UT, (b) TN, (c) TNE3S, (d) TNE3S&D, and (e) TNE3D.

3.3.  Tensile properties In order to determine whether there are any differences in the mechanical properties of the composites as a result of the different hemp surface treatments, tensile tests were performed. The results, presented in terms of Young modulus E (MPa), tensile strength σ (MPa), and elongation at break ε (%), have been analyzed and presented in Tables 3–5.

12 

  D. Y. Chimeni et al.

Table 3. Effect of mercerization and MAPE solution treatment on the tensile properties of the composites. Samples codes LMDPE UT UTE3S TN TNE3S

MAPE:hemp weight ratio – – 1:10 – 1:10

E (MPa) 241 (21)c 570 (21)a 517 (21)b 589 (19)a 598 (16)a

σ (MPa) 13.1 (0.3)d 13.6 (0.6)d 15.7 (0.4)b 14.8 (0.1)c 16.4 (0.1)a

ε (%) 822 (15)a 14.9 (1.3)c 15.5 (2.5)b,c 17.7 (2.3)b 14.4 (1.6)c

Note: Values with the same letter for each property are not significantly different at the 5% significance level according to the student t-test. The values in parentheses are standard deviations.

Downloaded by [Universite de Lorraine] at 11:27 18 February 2016

Table 4. Effect of the MAPE:hemp weight ratio on the tensile properties of mercerized hemp composites. Sample codes TNE3S TNE6S TNE9S

MAPE:hemp weight ratio 1:10 2:10 3:10

E (MPa) 598 (17)a 582 (22)a 581 (11)a

σ (MPa) 16.4 (0.1)c 17.9 (0.2)a 17.5 (0.3)b

ε (%) 14.4 (1.6)a 11.1 (1.3)b 11.9 (1.0)a,b

Note: Values with the same letter for each property are not significantly different at the 5% significance level according to the student t-test. The values in parentheses are standard deviations.

Table 5. Effect of MAPE introduction method on the composite tensile properties. Sample codes UTE3S UTE3D TNE3S TNE3D TNE3S&D

MAPE:hemp weight ratio 1:10 1:10 1:10 1:10 1:10

E (MPa) 517 (21)c 591 (15)b 598 (17)b 668 (28)a 605 (13)b

σ (MPa) 15.7 (0.4)d 15.5 (0.2)d 16.4 (0 1)c 16.9 (0.1)b 17.8 (0.3)a

ε (%) 15.5 (2.5)a 12.7 (1.1)b 14.4 (1.6)a 10.7 (1.0)c 11.1 (1.3)c

Note: Values with the same letter for each property are not significantly different at the 5% significance level according to the student t-test. The values in parentheses are standard deviations.

3.3.1.  Effect of mercerization and modification with MAPE in solution The effect of mercerization and modification by MAPE in solution on the tensile properties of the composites can be seen in Table 3. The results show that the introduction of 30% of neat hemp in LMDPE leads to a significant increase of the Young’s modulus (about 137%). This is a general trend observed when plastics are reinforced by neat fibers and can be explained by the stiffness imparted by the reinforcing agent (hemp) as its Young’s modulus (70 GPa) is much higher than that of the matrix (0.24 GPa).[31–33] With respect to the composite with neat hemp (UT), the composite with neat hemp modified by 3% of MAPE in solution (UTE3S) leads to a significant decrease of the Young’s modulus (10%) while the composites with mercerized hemp (TN) and mercerized hemp modified in solution (TNE3S) both exhibit a limited increase of Young’s modulus. On the one hand, the decrease observed for the moduli is in agreement with the significant decrease of fiber density (Table 2) and tends to show a decrease in the fiber stiffness in the case of UTE3S and a limited increase of the stiffness in the case of TN and TNE3S probably due to mercerization. On the other hand, the significant decrease of UTE3S modulus and the limited improvement of TNE3S are unexpected results. But given the type of MAPE used (low Mw), this can be simply explained in view of the FTIR, SEM, and TGA results, as well as the reaction mechanism. Indeed, the FTIR (Figure 1), SEM (Figure 2(c) and (d)), TGA (Figure 3(a)), DTG (Figure 3(b)), and the reaction mechanism

Downloaded by [Universite de Lorraine] at 11:27 18 February 2016

Composite Interfaces 

 13

(Figure 4 (Step 3)) showed that a thin layer of MAPE coated the hemp surface after surface modification. Therefore, these results for the moduli suggest that the amount of MAPE attached to the surface of modified hemp is not sufficient to contribute significantly to the composite modulus. For the tensile strength, Table 3 shows that the introduction of 30% of neat hemp in LMDPE leads to a limited improvement of the tensile strength of the composite (UT) with respect to the neat polymer. This trend is due to the lack of wettability coupled to the absence of compatibility between the hydrophilic hemp and the hydrophobic LMDPE (observed in Figure 5(a)). With respect to the composite with neat hemp (UT), the composite with neat hemp modified by 3% MAPE in solution (UTE3S) leads to a significant increase of tensile strength (about 13%), while the composites with mercerized hemp (TN) and mercerized hemp modified in solution (TNE3S) both exhibit a significant increase of the tensile strength (about 9 and 21%, respectively). The slight increase of 9% observed for TN can be associated to both the wettability improvement due to the increased contact area after hemp mercerization (Figure 2(b)) and the absence of compatibility revealed by fiber pull-out (holes) observed in SEM pictures for this composite (Figure 5(b)). The increased tensile strength for UTE3S can be associated to the presence of the thin layer of coupling agent (Figure 2(c) and (d)) which increased their compatibility with the matrix allowing the composite to sustain more stress transfer at the interface compared to UT. The tensile strength increase of 21% for the composite with mercerized hemp modified in solution with 3% of MAPE (TNE3S) is in agreement with SEM (Figure 2(b) and (c)) results and can be explained by the combined effect of mercerization (improved wettability) and the presence of a thin layer of coupling agent at the surface of mercerized hemp modified in solution (increased adhesion). The elongation at break of all the composites drastically drops with respect to the neat matrix (LMDPE) and neither mercerization nor MAPE treatment had significant effects. Earlier studies like Raj et al. [2], Kakroodi et al. [31], Lu et al. [34], as well as Sojoudiasli et al. [32] reported similar trends. This can be attributed to the lower elasticity of hemp fibers (between 2 and 4%),[35,36] compared to LMDPE (822%). Nevertheless, solution modification seems to give a positive trend to this property with respect to neat hemp composite. 3.3.2.  Effect of MAPE/hemp weight ratio in solution The effect of the MAPE:hemp weight ratio on the mechanical properties of the corresponding composites can be seen in Table 4. Variation of the MAPE:hemp ratio during the treatment of hemp in solution leads to no significant changes in term of Young’s modulus between the corresponding composites. This trend for the Young’s modulus is in agreement with the previous results and confirms that the surface MAPE layer do not have a significant contribution on modulus. On the other hand, the variation of MAPE:hemp ratio has a significant effect on tensile strength. Indeed, when the amount of MAPE is doubled in the solution (weight ratio 2:10), this property increases by about 9% for TNE6S compared to TNE3S and when the amount is tripled (weight ratio 3:10) the increase for the corresponding composite (TNE9S) is only 7%; i.e. a decrease of 2%. Increasing tensile strength from TNE3S to TNE6S can be attributed to the fact that the presence of a higher amount of MAPE in solution leads to better hemp coating by MAPE. Thus, the composite interface become stronger (less defects) and tensile strength increased. The decrease observed from TNE6S to TNE9S can be explained by two

14 

  D. Y. Chimeni et al.

Downloaded by [Universite de Lorraine] at 11:27 18 February 2016

reasons: (I) there is a saturation of the fiber surface such as no more MAPE can form covalent bonding. Then, any further MAPE molecule is in excess and is therefore responsible for a plasticizing effect which decreases the quality of the interface of the composite and finally his tensile strength, or (II) the high amount of MAPE in the solution increased the probability of MAPE molecules self-entanglement. As a result, less MAPE molecules are available to bond with hemp, hence a slight decrease of tensile strength. Finally, similar as modulus trends, the variation of MAPE:hemp ratio has no significant effect on the elongation at break of the composites. 3.3.3.  Effect of MAPE introduction: direct, in solution, and their combination The way MAPE was introduced in the composites was studied based on mechanical properties and the results are presented in Table 5. It is useful to keep in mind that in the case of the combined method, the corresponding composite (TNE3S&D) was prepared by using the same amount of MAPE (3%) which was divided into 1.5% in solution and 1.5% directly mixed during extrusion. Table 5 shows that, for the neat or pretreated hemp, the composites with direct use of MAPE exhibit significantly superior modulus than those with hemp modified by MAPE in solution. This can be explained by both the degradation of the hemp cellulosic component during solution modification (which reduced the treated fibers stiffness) and the not significant contribution of the amount of MAPE present at the hemp surface on modulus (for UTE3S and TNE3S as explained in Section 3.3.1), while the direct use of the total amount of MAPE (3%) has (for UTE3D and TNE3D) because of the amount of MAPE. The composite produced by the mixture of both methods (TNE3S&D) presents no significant effect on Young’s modulus with respect to TNE3S, but a significantly lower value (about 9%) compared to TNE3D. These results are unexpected but can be explained, for TNE3S&D, by a significant limitation of the contribution to the stiffness brought by the MAPE modification (directly + solution) probably due to the degradation of the cellulosic component of hemp during this modification. The tensile strength results show that there is no significant difference between both types of modification (direct and in solution) in the case of neat hemp (UTE3S vs. UTE3D), while a significant difference is observed when the hemp is pretreated with NaOH (TNE3D ˃ TNE3S). These trends between the neat and mercerized hemp fibers are due to the effect of mercerization which increased the number of available active sites (hydroxyl) leading to an interfacial quality improvement between mercerized hemp and LMDPE (Figure 5(c)) because of the increased number of MAPE-hemp bonds.[18] This effect is also observed by comparing the tensile strength of UTE3S vs. TNE3S and UTE3D vs. TNE3D. In particular, the tensile strength of TNE3S is significantly lower than TNE3D because of the presence of an insufficient amount of MAPE on the fibers surface to get an interface quality equivalent to that of the composite modified by the direct method since in this case, all the MAPE (3%) is available for bond formation leading to better interfacial improvement. This is more likely to occur than for the case where the amount of MAPE is increased in solution (Table 3), so the tensile strength of the corresponding composite (TNE6S) is higher than that of TNE3D. The composite produced by the combination of both methods (TNE3S&D) presents a tensile strength significantly greater than for the composite TNE3S (about 9%) and TNE3D (about 5%). This result is in agreement with the SEM picture (Figure 5(d)) of this composite and can be explained as follows: the fiber modification in solution acted as

Composite Interfaces 

 15

an impregnation process which facilitated the entanglement between the coated fibers and the matrix containing the other half of MAPE, increasing both the amount and the quality of bonds, hence the quality of the interface in TNE3S&D compared to TNE3S and TNE3D. Such behavior has also been highlighted by Rajabian et al. [19] for HDPE composites based on Kevlar fibers modified by a polyethylene monomer using catalytic compounding where the fibers were coated by a PE layer. Finally, the elongation at break of all the composites is much lower than the neat matrix (LMDPE) as mentioned previously. As expected, the values also decrease with increasing strength (higher σ) due to the low elasticity (elongation at break) of hemp fibers (between 2 and 4%) compared to the LMDPE matrix (822%).[35,36]

Downloaded by [Universite de Lorraine] at 11:27 18 February 2016

4. Conclusion In this study, different approaches of hemp surface modification were investigated. For each case, a complete characterization was performed to determine the level of surface modification and the effect on morphology and tensile properties of 30% wt. filled LMDPE composites was reported. From the results obtained, the main conclusions of this work are: • FTIR, SEM, TGA, and density results showed that during hemp modification in solution, a thin layer of MAPE coated hemp fibers. Based on the data obtained, a reaction mechanism was proposed to explain how MAPE modified the fibers in solution. • SEM of the composites showed that hemp mercerization was responsible for an increase of hemp wettability by LMDPE. This increase slightly improved the interfacial contact area while the combination of mercerization and MAPE in solution increased both wettability and adhesion. • The mechanical properties showed that for modified hemp composites, besides TNE3D and TNE3S&D which tensile modulus were significantly increased (17 and 6%, respectively), all the other composites exhibited no significant increase of this properties with respect to the composite based on neat hemp (UT). Tensile strength increase was significant for all the modified composites (about 20% for TNE3S, 24% for TNE3D, 30% for TNE3S&D, 13% for UTE3D, and 15% for UTE3S) with respect to UT, and greater than that of the tensile modulus suggesting that the MAPE used (E20P) has mostly an effect on interfacial improvement and less contribution on stiffness. Moreover, the combination of hemp mercerization with the use of a coupling agent was successful to significantly improve the strength of the composites. • Finally, MAPE modification in solution was effective for improving the tensile properties of LMDPE/hemp composite. Nevertheless, an optimization of the fiber surface treatment in solution must be done since improving the fiber surface covering by direct contact with MAPE also brings a partial degradation of the cellulosic components of hemp fibers due to the treatment conditions (solvent, temperature, time, and concentration). Thus, more work is necessary to get the full potential of this methodology.

Acknowledgment The technical help of Mr. Yann Giroux was also much appreciated.

16 

  D. Y. Chimeni et al.

Disclosure statement No potential conflict of interest was reported by the authors.

Funding The work was supported by Natural Sciences and Engineering Research Council of Canada (NSERC); Centre de recherche sur les systèmes polymères et composites à haute performance (CREPEC); Centre de recherches sur les matériaux avancés (CERMA); Centre de recherche sur les matériaux renouvelables (CRMR).

Downloaded by [Universite de Lorraine] at 11:27 18 February 2016

References   [1] Lu N, Oza S. Thermal stability and thermo-mechanical properties of hemp-high density polyethylene composites: effect of two different chemical modifications. Composites Part B. 2013;44:484–490.   [2] Raj RG, Kokta BV. Mechanical properties of surface-modified cellulose fiber–thermoplastic composites. ACS Symp. Ser. 1992;476:76–87.   [3] Wang Q, Ait-Kadi A, Kaliaguine S. Catalytic grafting: a new technique for polymer/fiber composites. II. Plasma treated UHMPE fibers/polyethylene composites. J. Appl. Polym. Sci. 1992;45:1023–1033.   [4] Faruk O, Bledzki AK, Fink H-P, et al. Progress report on natural fiber reinforced composites. Macromol. Mater. Eng. 2014;299:9–26.   [5] Kabir MM, Wang H, Lau KT, et al. Mechanical properties of chemically-treated hemp fibre reinforced sandwich composites. Composites Part B. 2012;43:159–169.   [6] Mohanty S, Verma S, Nayak S. Dynamic mechanical and thermal properties of MAPE treated jute/HDPE composites. Compos. Sci. Technol. 2006;66:538–547.   [7] Ouajai S, Shanks RA. Composition, structure and thermal degradation of hemp cellulose after chemical treatments. Polym. Degrad. Stab. 2005;89:327–335.   [8] Pothan LA, Oommen Z, Thomas S. Dynamic mechanical analysis of banana fiber reinforced polyester composites. Compos. Sci. Technol. 2003;63:283–293.   [9] Sewda K, Maiti SN. Dynamic mechanical properties of high density polyethylene and teak wood flour composites. Polym. Bull. 2013;70:2657–2674.   [10] Mechraoui A, Riedl B, Rodrigue D. The effect of fibre and coupling agent content on the mechanical properties of hemp/polypropylene composites. Compos. Interfaces. 2007;14:837– 848.   [11] Twite-Kabamba E, Mechraoui A, Rodrigue D. Rheological properties of polypropylene/hemp fiber composites. Polym. Compos. 2009;30:1401–1407.   [12] Toupe JL, Trokourey A, Rodrigue D. Simultaneous optimization of the mechanical properties of postconsumer natural fiber/plastic composites: phase compatibilization and quality/cost ratio. Polym. Compos. 2014;35:730–746.   [13] Correa CA, Razzino CA, Hage E. Role of maleated coupling agents on the interface adhesion of polypropylene-wood composites. J. Thermoplast. Compos. Mater. 2007;20:323–339.   [14] Mwaikambo LY, Ansell MP. Chemical modification of hemp, sisal, jute, and kapok fibers by alkalization. J. Appl. Polym. Sci. 2002;84:2222–2234.   [15] Aziz SH, Ansell MP. The effect of alkalization and fibre alignment on the mechanical and thermal properties of kenaf and hemp bast fibre composites: part 1 – polyester resin matrix. Compos. Sci. Technol. 2004;64:1219–1230.   [16] Kabir MM, Wang H, Lau KT, et al. Chemical treatments on plant-based natural fibre reinforced polymer composites: an overview. Composites Part B. 2012;43:2883–2892.   [17] Sawpan MA, Pickering KL, Fernyhough A. Effect of various chemical treatments on the fibre structure and tensile properties of industrial hemp fibres. Composites Part A. 2011;42:888–895.

Downloaded by [Universite de Lorraine] at 11:27 18 February 2016

Composite Interfaces 

 17

  [18] Shahzad A. Effects of alkalization on tensile, impact, and fatigue properties of hemp fiber composites. Polym. Compos. 2012;33:1129–1140.  [19]  Rajabian M, Dubois C. Polymerization compounding of HDPE/Kevlar composites. I. Morphology and mechanical properties. Polym. Compos. 2006;27:129–137.  [20] Zhang Y, Rodrigue D, Aït-Kadi A. Tensile properties of polymerized-filled kevlar pulp/ polyethylene composites. Polym. Polym. Compos. 2004;12:1–13.  [21] Chusheng Q, Kangquan G. Effects of thermal treatment and maleic anhydride-grafted polyethylene on the interfacial shear stress between cotton stalk and high-density polyethylene. In Proceedings of the 2012 International Conference on Biobase Material Science and Engineering; 2012 Oct 21–23; Changsha (CN): IEEE; 91–95.   [22] Priyanka PS. Banana fiber/chemically functionalized polypropylene composites with in-situ fiber/matrix interfacial adhesion by Palsule process. Compos. Interfaces. 2013;20:309–329.   [23] Verdaguer A, Rodrigue D. Effect of surface treatment on the mechanical properties of woodplastics composites produced by dry-blending. In Proceedings of the 72th Annual Technical Conference & Exhibition; 2014 Apr 28–30; Las Vegas (NV): SPE; 2021–2025.   [24] Raymond A, Rodrigue D. Effect of surface treatment on the properties of wood–plastics composites produced by rotomolding. In Proceedings of the 72nd Annual Technical Conference & Exhibition; 2014 Apr 28–30; Las Vegas (NV): SPE; 2367–2371.   [25] Jannah M, Mariatti M, Bakar AA, et al. Effect of chemical surface modifications on the properties of woven banana-reinforced unsaturated polyester composites. J. Reinf. Plast. Compos. 2008;28:1519–1531.  [26] Bodîrlău R, Teacă CA. Fourier transform infrared spectroscopy and thermal analysis of lignocellulose fillers treated with organic anhydrides. Rom. J. Phys. 2009;54:93–104.   [27] Kazayawoko M, Balatinecz JJ, Woodhams RT. Diffuse reflectance Fourier transform infrared spectra of wood fibers treated with maleated polypropylenes. J. Appl. Polym. Sci. 1997;66:1163– 1173.   [28] Kaczmar JW, Pach J, Burgstaller C. The chemically treated hemp fibres to reinforce polymers. Polimery. 2011;56:817–822.  [29] Lu JZ, Negulescu II, Wu Q. Maleated wood-fiber/high-density-polyethylene composites: Coupling mechanisms and interfacial characterization. Compos. Interfaces. 2005;12:125–140.   [30] Kalia S, Kaith BS, Kaur I. Pretreatments of natural fibers and their application as reinforcing material in polymer composites – a review. Polym. Eng. Sci. 2009;49:1253–1272.   [31] Kakroodi RA, Kazemi Y, Rodrigue D. Mechanical, rheological, morphological and water absorption properties of maleated polyethylene/hemp composites: effect of ground tire rubber addition. Composites Part B. 2013;51:337–344.   [32] Sojoudiasli H, Heuzey M-C, Carreau P. Rheological, morphological and mechanical properties of flax fiber polypropylene composites: influence of compatibilizers. Cellulose. 2014;21:3797– 3812.   [33] Shahzad A. Hemp fiber and its composites-a review. J. Compos. Mater. 2012;46:973–986.   [34] Shubhashini O, Lu N, Korman T. Effect of alkali treatment on the mechanical properties of hemp-HDPE composites: virgin versus recycled polymer matrix. In Sustainable design and construction. International conference; 2012 Nov 7–9; Fort Worth (TX): ASCE; 818–825.   [35] Mohanty AK, Misra M, Drzal LT. Surface modifications of natural fibers and performance of the resulting biocomposites: an overview. Compos. Interfaces. 2001;8:313–343.   [36] Sathishkumar T, Navaneethakrishnan P, Shankar S, et al. Characterization of natural fiber and composites – a review. J. Reinf. Plast. Compos. 2013;32:1457.