硅烷偶联剂综述

Effect of Silane Coupling Agents on the Properties of Pine Fibers/Polypropylene Composites Jordi Girone`s,1 Jose´ Alberto Me´ndez,1 Sami Boufi,2 Fabiola Vilaseca,1 Pere Mutje´1 1LEPAMAP group. University of Girona, Campus Montilivi, 17071, Girona, Spain 2LMSE, Faculte´ des sciences de Sfax, BP 802-3018 Sfax, Tunisia Received 4 May 2006; accepted 3 July 2006 DOI 10.1002/app.25104 Published online in Wiley InterScience (www.interscience.wiley.com). ABSTRACT: In the present work, PP-based composites, reinforced with surface modified pine fibers, have been prepared. The surface of the fibers has been treated with several silane derivatives bearing specific functionalities. ��NH2, ��SH, long aliphatic chain, and methacrylic group were chosen as functionalities of the silane derivatives for evaluating the compatibility with the polymer matrix. Me- chanical analysis, contact angle and XPS spectra, SEM mi- croscopy, and water uptake measurements were used as characterization techniques for evaluating the nature of com- posites. XPS as well as contact angle measurements demon- strated that pine fibers and silane derivatives were effec- tively coupled. The mechanical analysis showed an increase in Young’s and flexural moduli, by 12% and 130% respec- tively, and nonsignificant changes in the ultimate tensile strength were noted after surface modification. Water uptake measurements revealed a low water absorption by the mate- rials, always lower than 2 wt %. � 2006 Wiley Periodicals, Inc. J Appl Polym Sci 103: 3706–3717, 2007 Key words: composite; cellulose fibers; silane; interface; fibers treatment; PP INTRODUCTION The recourse of lignocellulosic fibers as a reinforcing element in composite materials still attracts much in- terest at both the academic and industrial level.1–5 Since 1990 composites based on natural fibers, namely those associated with thermoplastic matrices, find application in the automotive and construction indus- try and are more and more used as a substitute of fiberglass composites.6 Besides environmental and eco- nomic concerns, many important factors are now driv- ing the use of lignocellulosic fibers, such as weight saving, good acoustic and thermal insulation, nonabra- sive effect, and good availability. The potential use of plant fibers as a reinforcement of polymer matrices is greatly harmed by their hydro- philic character, high capacity of moisture absorption (7 to 9 wt % at 50% humidity), and poor adhesion between them and the matrix. Indeed, the presence of an adsorbed layer of molecular water on the surface hinders any possible contact or interaction at a mole- cular level between both phases and prevents fiber wetting by the polymer. Likewise, the high density of hydroxyl groups on the surface ensures a high cohe- sion within the fiber network through hydrogen bonding, which is difficult to break up and to bring about efficient and homogenous dispersion of the fiber aggregates on the organic matrix. As a result, a considerable effort is currently being directed toward improving the quality of the interfacial bonding between the polymer and fibers, by surface modifica- tion of lignocellulosic fibers.7–10 Different approaches have been investigated to improve the compatibility between cellulosic fibers and polymeric matrices, i.e., chemical modification,11,12 poly- mer grafting on the surface of the fibers,13 incorpora- tion of compatibilizer, such as maleated polymer13,14 or treatment with coupling agents.9,10,15–17 The for- mer approach is one of the most adopted and stud- ied because of the high density of hydroxyl groups on the surface of the fiber, on which a wide variety of reactions could be undertaken. By means of these reactions, the substitution of this function by the other target group can be carried out to modify the surface structure and bring about better compatibil- ity between both components of the composite. Among the different coupling agents, organo-si- lane (R��Si��(OR0)3) is the most effective, commer- cially available, and low cost product and is widely used for modifying surface properties of inorganic substrates such as fiberglass or oxide filler. Silane coupling chemicals bear alkoxysilane groups, which after hydrolysis are capable of reacting with OH-rich surfaces. Furthermore, there is a wide range of avail- Correspondence to: S. Boufi (sami.boufi@fss.rnu.tn). Contract grant sponsor: NATO; contract grant number: PST.CLG.980373. Contract grant sponsors: Ministry of Education and Sci- ence of Spain (Juan de la Cierva Program) and University of Girona. Journal of Applied Polymer Science, Vol. 103, 3706–3717 (2007) VVC 2006 Wiley Periodicals, Inc. able functionalities for the R group. This R group is responsible for improving the compatibilization between the reinforcing element and the polymer matrix or even establishing covalent bonds between them. Even though the beneficial effect on the adhesion between the fiber and the matrix is well established with fiberglass and inorganic fillers,18 their effects on the properties of lignocellulosic fibers reinforced com- posites are not formally established and still receive considerable attention.9,10,15–17,19–23 Using different coupling agents, Demir et al.15 showed that treatments of luffa fibers with g-amino- propyltrimethoxysilane and g-mercaptopropyltrime- thoxysilane improved both tensile strength and Young’s modulus by more than 40%, compared with untreated fibers. Colom et al.10 modified the surface of aspen wood fibers with g-methacryloxoproyltri- methoxysilane for enhancing interface adhesion. An increase by 37.5% in tensile strength (20 wt %) was reported. However, most of these works were in agree- ment with regard to a gain in mechanical properties and elastic moduli provided by the silane treated fibers, while a discrepancy appeared regarding the evolution of the tensile strength. In a previous work, we have showed that trialkoxy- silane could be effectively anchored on cellulosic fibers through a preliminary physical adsorption from di- luted solution followed by a thermal treatment.19,20 Different spectroscopic techniques have been used to prove the presence of silane, to quantify its amount on the substrate, and to elucidate the structure of the anchored siloxane network on the fiber’s surface. For thermoset-based composites, it has been clearly estab- lished that the reinforcing effect of the fibers could be significantly improved by their treatment with a silane bearing functional group that is able to react with the matrix.21 Conversely, in a recent work an analogue result was shown, where in the presence of LDPE ma- trix, the treatment of fibers with a reactive silane cou- pling agent was performed, obtaining an improve- ment in both the tensile modulus and the tensile strength, and a drop by more than 35% in the compos- ite water uptake.22 The encouraging results obtained for thermoset- based and LDPE composites21,22 motivated the pres- ent study, in which the effect of silane-treated pine fibers, on the properties of polypropylene matrix, was investigated. MATERIALS AND EXPERIMENTAL PROCEDURES Cellulose fibers Bleached softwood (pine) fibers with an arithmetic average length of about 1.1 mm were used as rein- forcement. Silane coupling agents R-trialkoxysilanes with different R functionalization: methacryloxypropyl-(MPS), hexadecyl-(HDS), amino- propyl-(APS) and mercaptopropyl-(MRPS) (Table I) were kindly provided by Degussa and were used without prior treatment. Polymer matrix Poly(propylene) (PP; Isplen 070, Repsol-YPF) was used as polymer matrix. This polymer has a melting temperature of 1698C and a degree of crystallinity around 45%, as determined by differential scanning calorimetry (DSC), with a heating rate of 108C/min. Its density at room temperature was 0.905 g/cm3. The melt flow index of this polymer matrix was 12 g/10 min, measured at 2308C and using a load of 2.16 kg. Fiber treatment The procedure followed for the fiber treatment with silane was optimized to obtain maximum amount of silane coupling agent anchored on the fiber surface. Different steps were preliminary investigated and the following procedure was adopted: in an ethanol/ water (80/20 v/v) solution, chosen silane (1.5 wt %) was added and the pH was adjusted to 4.5–5 by addi- tion of acetic acid (exception made of AMPS because of its auto-catalytic nature), and was kept under stirring for 1 h to ensure complete silane hydrolysis. Then, the hydrolyzed silane solution was added on an ethanol/ water (80/20) pine fiber suspension (5 wt %) and was kept under mechanical stirring for 2 h to reach an effec- tive fiber soaking. After this process, obtained fibers were filtered, dried at room temperature for 2 days and heated at 1108C under a nitrogen atmosphere for 2 h, to promote actual chemical coupling. Preparation of the composites PP and pine fibers (treated and untreated) were mixed at 70/30 ratio (PP/fibers) in a two-roll mill (IQAP LAB) at (190 6 5)8C for 10 min to obtain a well-dispersed material. The mill is equipped with two parallel rolls, turning at two different speeds of 23 and 29 rpm. Obtained blends were cut down to pellets with a particle size in the range of 10 mm, using a pelletizer equipped with a set of knifes and different grids. Pellets were dried and stored at 808C during 24 h. After drying, the pelletized material was injected by an injection-molding machine (Meteor-40, Mateu and Sole´). Processing temperature of heated areas of the injector machine was 1758C, 175 and 1908C being the highest corresponding to the nozzle. First and second injection pressures were 120 and 25 kgf cm�2, respectively. By means of this EFFECT OF SILANE DERIVATIVES ON PP-BASED COMPOSITES 3707 Journal of Applied Polymer Science DOI 10.1002/app procedure, specimens for tensile and flexural tests were obtained with a shape according to ASTM D638 and ASTM D790 standard specifications, re- spectively. Chemical analysis of the surface of the materials XPS spectra were recorded in an XSAM800 (KRATOS) apparatus, operated in the fixed analyzer transmission mode, with a pass energy of 10 eV and nonmonochromatic MgKa and AlKa X-radiations (hu ¼ 1253.7 eV and 1486.7 eV, respectively). A current of 10 mA and a voltage of 13 kV were used. Samples were analyzed in an ultrahigh-vacuum chamber (�10�7 Pa) at room temperature, using 608 analysis angles relative to the surface’s normal. Samples were transferred from the last rinsing solution inside the introduction chamber under argon atmosphere. Spectra were recorded by a Sun SPARC Station 4 with Vision software (Kratos) using a step of 0.1 eV. A Shirley background was subtracted and curve fit- ting for component peaks was carried out using Voigt profiles. Contact angle measurements Dynamic contact angle (CA) measurements were per- formed using a Dataphysics OCA 20 apparatus. Surfa- ces were prepared by mild pressing of fibers to form film-like materials suitable for CA measurements. A calibrated droplet of water was deposited on the sur- faces and the evolution of CA with time was recorded using a CCD camera with an automatic acquisition of 50 images per second. Fiber morphology evaluation Average fiber morphology was determined using an optical microscope equipped with a CCD camera and image analysis software. These measurements gave the fiber length and particle diameter distribu- tion. Thermal analysis DSC was performed with a Perkin–Elmer thermal analysis DSC 8230-B equipment fitted with a cooler system using liquid nitrogen. Samples, around 10 mg, were placed in pressure-tight DSC cells and at least two individual measurements were collected. Each sample was heated from 0 to þ230 8C at a heat- ing rate of 108C/min, held at 2308C for 5 min and finally cooled at the same cooling rate. Melting tem- perature (Tm) and crystallization temperature (Tc) was taken as the peak temperature of the melting endotherm and crystallization exotherm. Mechanical analysis Tensile and flexural tests were performed using a Universal Testing Machine (Instron 1122 according to ASTM D638 and ASTM D790 standard specifica- tions, respectively). Before use, samples were stored at 238C and 50% of relative humidity for 48 h, according to ASTM D618 standard specifications. A minimum of five specimens were used for obtaining each value at room temperature and 50% of relative humidity. Ultimate tensile strength (st, UTS), elastic modulus or Young modulus (Et), flexural strength (sf) and flexural modulus (Ef) were studied. TABLE I The Silane Coupling Agents Used in this Work 3708 GIRONE`S ET AL. Journal of Applied Polymer Science DOI 10.1002/app Evaluation of the fracture surface Surface fractured areas of tensile specimens were observed by scanning electron microscopy (SEM) (Zeiss DMS 960). By means of this technique, it was possible to determine qualitatively adhesion degree between the matrix and reinforcement. Water absorption measurements Sample dimensions for water absorption experiments were 1 cm � 1 cm � 0.5 mm. A minimum of two samples were tested for each material. Samples were weighted and then soaked in distilled water at room temperature. Samples were removed at specific time intervals, blotted to remove the excess of water on the surface, and were immediately weighed. The dif- ference between the mass after a given time of immersion and initial mass compared with initial mass led to determine water absorption. RESULTS AND DISCUSSION To bring about the surface modification, fibers were kept in contact with the silane solution for 2 h fol- lowed by a curing process at 1108C for 1 h after sol- vent removal. During the first stage of the treatment, silanes were physically adsorbed on the surface of the fibers through hydrogen bonding interaction between cellulose hydroxyl groups and silanol groups (Si��OH) of the silanes. The silanol groups are generated by acid hydrolysis of the triethoxy groups. As reported previously, subsequent heat treatments ensured an efficient and irreversible chemical bonding of the sil- ane on the cellulose surface through condensation reaction of BSi��OH with cellulose hydroxyl groups and self-condensation, giving rise to polysiloxane bridges. XPS analyses were carried out to give evidence of silane presence and quantify its amount on the fiber surface. Because of the low amount of silane held on the fiber surface, this technique is the most adapted to provide both qualitative and quantitative informa- tion, regarding the different elements present on the surface and their chemical environment. Figure 1 depicts XPS survey spectra of electron intensity as a function of binding energy for virgin fibers, MPS-, and AMPS-treated fibers. As expected, virgin fibers displayed only two peaks at about 533 and 285 eV attributed to O1s and C1s, respectively. On the other hand, silane treated fibers revealed that, in addition to the peaks associated with oxygen and carbon, emission peaks at 103 and 150 eV characteristic of TABLE II XPS Chemical Quantitative Analysis for Treated and Untreated Cellulose Fibers Sample Element and the corresponding binding energy (eV) C(1s) 285 O(1s) 533 Si(2p) 103 N(1s) 398 S(2p) 165 Virgin pine 54.76 44.8 0.24 0.2 0.0 Pine-MRPS 59.74 37.17 1.54 0.3 1.25 Pine-MPS 46.67 33.96 18.36 1.01 0.0 Pine-AMPS 52.37 35.28 6.77 5.24 0.34 Figure 1 XPS survey spectra for (a) virgin pine fibers, (b) APS, and (c) MPS treated pine fibers. EFFECT OF SILANE DERIVATIVES ON PP-BASED COMPOSITES 3709 Journal of Applied Polymer Science DOI 10.1002/app Si2s and Si2p, respectively, were found. The presence of these peaks confirms the attachment of the silane on cellulose fibers. Elemental composition on the surface layer, determined from the area of each peak normalized with sensitivity factors, for different si- lane modified fibers are summarized in Table II. It is worth noting that the amount of sulfur and nitrogen is close to that of silicon in MRPS and AMPS, respectively, which gives further confirmation of the accuracy of the quantitative analysis. Likewise, the amount of anchored silane varied from 1.5 up to 18%, according to the silane structure. The highest level is observed for MPS-treated fibers, followed by AMPS and MRPS. Better information regarding chemical environment of different elements detected on the surface of fibers could be emerged through an amplification of C1s, N1s, and Si2p peaks related to different fibers [Fig. 2(A)]. C1s peak was deconvo- luted in four different carbon types fitted at 284.8, 286.5, 288, and 289.3 eV and attributed to C1 (C��C/ C��H/C��Si), C2(C��O/C��OH), C3(O��C��O), and C4(O��C¼¼O), respectively. C4 peak appeared only in the presence of MPS. The increase in C1 con- tribution after treatment is in agreement with silane anchoring on cellulose surface, since its presence brings three aliphatic carbons per anchored silane coupling agent. With more amount of bonded sili- con, more was the contribution of the C1 peak. The amino N1s peak could be deconvoluted in two peaks at 399.0 and 400.8 eV corresponding to amino groups (��NH2) and to their protonated form (��NHþ3 ), respectively [Fig. 2(B)]. The higher contribution of the former suggests that the major part of amino groups in adsorbed AMPS is in their unprotonated form. The evolution of the fibers surface properties was investigated by CA measurements with water as liquid probe. Figure 3 depicts dynamic CA versus time for virgin and silane treated fibers after curing treatment and after several weeks of aging. Before treatment, CA did not exceed 308 and dropped rap- idly with time, as a result of the high hydrophilic character of cellulose. Silane treatment brings about an increase in CA which depends on the silane structure; the highest effect was observed with HDS, followed by MPS, MRPS, and AMPS. High CA in the presence of HDS and MPS could be ascribed to their relatively high amount on the surface and to their hydrophobic character. On the other hand, the presence of amino groups on AMPS silane may explain the persistence of the hydrophilic character after treatment. However, the possible interference of BSi��OH groups on the surface character should be taken into account. Indeed, when fibers were sub- Figure 3 Water contact angles change for virgin and treated pine fibers (A) after heat treatment at atmospheric pressure and (B) after heat treatment at partial vacuum. Figure 2 XPS spectra of (A) C1s carbon peaks for virgin and MPS, MRPS, and AMPS pine fibers; (B) N1s peaks for AMPS pine fibers. 3710 GIRONE`S ET AL. Journal of Applied Polymer Science DOI 10.1002/app mitted to a further heat treatment at 1108C for 1 h under partial vacuum (5 � 10�2 mmHg), a signifi- cant raise in CA was noted as shown in Figure 3. This result was imputed to further condensation between BSi��OH groups, which gives rise to a bidimensional polysiloxane network that hinders more efficiently hydroxyl groups on the surface of the fibers. Therefore, for obtaining an efficient anchoring of the silane coupling agent on cellulose substrate, it is advisable to carry out a heat treatment under partial vacuum to ensure a high level of sur- face modification. It is worth noting that treated fibers used for this work in composite preparation were only subjected to heat treatment at 1108C at atmospheric pressure, without recourse to vacuum. Indeed, it was difficult to carry out this procedure efficiently for a relatively high amount of fibers. CA measurements have also been carried out on fibers recovered from composites after complete re- moval of polypropylene matrix by soxhlet extraction with xylene. Results reported on Figure 4 revealed that ensued fibers exhibited a higher hydrophobic character, with a CA exceeding 808, which did not evolve after repeated soxhlet extraction for a longer time. The more the CA of silane treated fibers before their blend with matrix, the more the hydrophobic character of extracted fibers. This phenomenon is probably imputed to th