Polymer Testing 29 (2010) 245–250 Contents lists available at ScienceDirect Polymer Testing journal homepage: www.elsevier.com/locate/polytest Material Properties Mechanical properties of polypropylene layered silicate nanocomposites: Characterization and micro-macro modelling L. Cauvin a, D. Kondo a, *, M. Brieu c, N. Bhatnagar b a Laboratoire de Mécanique de Lille, UMR CNRS 8701, U.S.T.L, Bd. Paul Langevin, 59650 Villeneuve d’Ascq Cedex, France Mech. Eng. Dpt., Indian Institute of Technology, HausKhaz, New Delhi, India c Laboratoire de Mécanique de Lille, UMR CNRS 8701, ECL, 59650 Villeneuve d’Ascq Cedex, France b a r t i c l e i n f o a b s t r a c t Article history: Received 20 September 2009 Accepted 9 November 2009 The present study is devoted to the mechanical behavior of a polypropylene reinforced by clay nano-platelets. Uniaxial tensile tests have been performed to evaluate some mechanical characteristics (stress-strain curves, Young’s modulus, yield stress, strength, strain at rupture) of the nanocomposite. It is shown that the Young’s modulus and the yield stress significantly increase compared to the matrix, even for a very low volume fraction (less than 3%). For the elastic properties, the three-phase model proposed by Ji [1] predicts accurately the impact of nano-platelets on the Young’s modulus, provided that a suitable choice of the interphase properties is made. The offset of the yield stress of the nanocomposite is investigated by means of Pukanszky’s model [2] which provides good results when compared with our experimental data. Ó 2009 Elsevier Ltd. All rights reserved. Keywords: Nanocomposites Montmorillonite clay Polypropylene Mechanical properties Experimental testing Modelling 1. Introduction The first results obtained by Toyota Research [3,4], in the 90s, on the dispersion of nanoscopic platelet silicates (montmorillonite, MMT) in a polyamide 6 matrix have induced considerable worldwide research in the field of nanocomposites consisting of a thermoplastics and clay platelet reinforcement [5–8]. For very low mass fraction of platelets (less than 5%), some remarkable mechanical properties as well as thermal barrier to gases, etc. have led to a great industrial and academic interest (see for instance [6,9–14]). Most of these publications investigated only experimental processing and characterization or theoretical modelling, and often in the linear regime. The present study is concerned with the experimental characterization and the prediction of the elastic properties and yield stress of a class of platelet-reinforced nanocomposites. The main objective is a better understanding * Corresponding author. Tel.: þ33 320337182; fax: þ33 320337153. E-mail address: djimedo.kondo@univ-lille1.fr (D. Kondo). 0142-9418/$ – see front matter Ó 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.polymertesting.2009.11.007 of the effect of the nanoscopic reinforcements on the mechanical behavior: Young’s modulus and yield stress. We first describe the studied material, a polypropylene matrix (PP) reinforced by nanoscopic clay platelets (Montmorillonite, MMT). The experimental characterization of the mechanical behavior is then performed through tensile tests. Concerning the elastic properties, we have proposed in [15] a micromechanical model able to describe the reinforcement effect of the platelets. The model, based on a two phase variational homogenization approach (Hashin–Shtrikhman type bounds) [16], does not account for any size dependence of the elastic behavior. A main objective of the present study is to apply a three-phase model proposed by Ji et al. [1] to the studied nanocomposite. These authors proposed to take into account the size effect of the platelets by means of an interphase. The relevance of this model will be discussed. Concerning the yield stress, we will consider the model proposed by Pukanszky et al. [2] which also takes into account the presence of an interphase. A comparison between experimental data and the model predictions is provided. 246 L. Cauvin et al. / Polymer Testing 29 (2010) 245–250 2. Studied material The studied nanocomposite is a polypropylene (PP) matrix (REPOL H020EG – Reliance make) reinforced by Montmorillonite (MMT) clay platelets (CRYSNANO 1010Southern Clay make). It was manufactured at the Indian Institute of Technology (IIT) of New Delhi in India. The polypropylene (PP) and the clay (MMT) were first thoroughly mixed for different volume fractions of clay in a corotating twin screw extruder (Hake make). The extrudate was granulated for molding purposes. The material samples required for tensile tests were then obtained by injection molding of these nanoclay filled PP granules. The size of the MMT nano-particles was measured with a particle size analyzer (Brookhaven 90Plus) at the IIT. The nano-platelets have an ellipsoidal shape with diameters of 209 nm and 189 nm and thickness 50 nm, hence they can be considered, if required, as a plate. Depending on the processing, three types of material (Fig. 1) can be distinguished. In the first type, the nanoclay particles/platelets are uniformly distributed and dispersed in an aggregate form. This two phase material is expected to enhance mechanical or thermal properties as in classical composites. Therefore, it is classically termed a micro–composite. The second microstructure configuration is made up of intercalated nanocomposites in which polymer chains get into the space between thin platelets of nanometer dimensions, thereby resulting in huge reinforcement. However, in this case, platelets remain isolated in aggregates between which the matrix is intercalated. The last type is composed of nanocomposites with exfoliated structure. All the platelets of the aggregates are separated and dispersed within the matrix. The last two microstructure configurations of materials are the true nanocomposites, partially or fully intercalated and/or exfoliated. Note that the most commonly manufactured nanocomposites are those with a partially or fully intercalated and/or exfoliated structure [17]. Use of X-ray Diffraction (XRD) allows showing whether the obtained nanocomposite has partially intercalated and partially exfoliated structure. 3. Experimental characterization Uniaxial tensile tests have been performed on the studied material (PP þ MMT) using a conventional test machine (INSTRON 4302) with a load cell of 1 kN. Polymers such as polypropylene are generally highly deformable. Therefore, the use of a contact extensometer to measure the strain is not ideal. Measurements are then performed without contact by the mean of a video extensometer (ApollorÒ). The video extensometer focuses a camera on the centre of the tensile test specimen and analyses the displacement of markers (Fig. 3). It allows, by means of the experimental set up shown on Fig. 2, controlling several parameters of test depending on the material response. In classical tensile tests, the overall strain rate can be constant but not the local rate. The system considered here is able to drive the tensile tester and, thus, to modify the strain rate to ensure a constant local rate of strain between the markers throughout the test. Fig. 3 shows the lengths L0 (between the markers before the loading) and L (between the markers during the test) measured by the system to evaluate the local strain (ll) (to ensure it is constant): ll ¼ 1 þ L  L0 L0 (1) For all experiments, the rate of deformation was equal to 103/s in order to guarantee quasi-static conditions. All sample shapes and dimensions were based on ISO 527-1 Type I test piece and were produced at optimized and constant processing parameters in a numerically controlled injection-molding machine (L&T Demag make). Mass fractions of 2%, 3%, 4%, 5%, 6% and 7% of MMT clay reinforcement (embedded in the PP matrix) were considered. Note that these weight ratios correspond to very low volume fraction of MMT (less than 3%). σ,ε v ε Console Video camera Tensile Fig. 1. Different types of nanocomposites. Fig. 2. Video extensometer ApollorÒ. L. Cauvin et al. / Polymer Testing 29 (2010) 245–250 247 2200 E (MPa) 2000 1800 1600 1400 1200 1000 0 1 2 3 4 5 6 7 8 Wt (%) Fig. 5. Young’s modulus of the nanocomposite versus the weight ratio Wt of reinforcements with standard deviations. Fig. 3. Principle of the measure of the local strain ((a) before the test and (b) during the test). Fig. 4 shows the response of the material under tensile loading. To ensure the quality of the experimental results, several tests have been performed for each reinforcement weight ratio and the results shown in Fig. 4 correspond to the average of the measurements. Concerning the elastic regime of the mechanical behavior, a strong effect of the nano-platelets on the variation of the Young modulus as a function of the MMT weight ratio (less than 7%) is observed (see Fig. 5 and also [15]). It is worth noting that the considered weight ratios correspond to very low volume fractions (less than 3%, see Fig. 6). Note that for the elastic properties of the (PP) matrix from analyzing literature data, are E ¼ 900 MPa and n ¼ 0.4. From Fig. 4, it is also possible to determine the yield stress, the strain at rupture and the strength. For the yield stress, we consider that the plastic domain starts at the end of the linear part of the material response as shown on Fig. 7. Fig. 8 indicates the significant effect of the nanoplatelets on yield stress with a fairly constant value with MMT volume fraction (about twice that of the matrix). The strength (Fig. 9) is fairly constant for 2%, 3%, 4% and 5% of MMT clay reinforcement, but increases at 6% and 7% of MMT clay reinforcement. For the strain at rupture (Fig. 10), we observe a decrease as the weight ratio increases. 4. Modeling of the mechanical behavior of the nanocomposite When nanoscopic reinforcements are considered, the matrix–reinforcement interface energy cannot be neglected, as is the case for classical composites. Several authors [18,11] have shown that nanoscopic size of the reinforcement can strongly modify the properties of the matrix which surrounds the reinforcement. This part of the matrix, termed the interphase, has already been modelled in the case of nanocomposites [11,14]. We propose, in this section, 50 Stress (MPa) 40 30 PP pur 2% 20 3% 4% 5% 10 6% 7% 0 0 0,02 0,04 0,06 0,08 Strain 0,1 0,12 0,14 Fig. 4. Tensile test response of the PP-MMT nanocomposite for different MMT weight ratios. L. Cauvin et al. / Polymer Testing 29 (2010) 245–250 2200 35 2000 30 Yield stress (Mpa) E (MPa) 248 1800 1600 1400 1200 1000 25 20 15 10 0 0,5 1 1,5 f(1) (%) 2 2,5 3 0 4.1. Effective elastic properties To analyze the experimental data shown in Fig. 6, the micromechanical model considered is the one proposed by Ji et al. [1] and already used by Aı̈t Hocine et al. [14] for the case of a polyamide-12 reinforced by nano-platelets with a good agreement between experimental data and model predictions. This model is a three-phase model (matrix, inclusion and interphase) and links in parallel and in series the matrix, the reinforcement and the interphase (including its size). For platelet reinforcement having a thickness t and both length and width x (with x > >t), the ~ is expressed as [1]: effective Young’s modulus E 2 ~ ¼ E0 6 E 4ð1  aÞ þ þ ð1  aÞ þ ab aðk  1Þ ð1  aÞ þ lnðkÞ b ða  bÞðk þ 1Þ 2 E1 þb E0 31 (2) 7 5 pffiffiffiffiffiffi where b ¼ fð1Þ , f(1) is the volume fraction of the platelets; qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi  ffi a ¼ 2ðs=tÞ þ 1 fð1Þ , s is the thickness of the interphase; 2 3 4 5 Weight ratio (%) 6 7 8 Fig. 8. Yield stress versus platelet weight ratio of reinforcement. Fig. 6. Young’s modulus of the nanocomposite versus the volume fraction of reinforcements with standard deviations. to analyse the macroscopic elastic and non linear properties of the PP/MMT nanocomposite by taking into account this interphase. 1 k is the ratio between the Young’s modulus on the surface of the platelet Ei and the Young’s modulus of the matrix E0, k ¼ Ei/E0. It must be emphasized that, if the effects of interface are neglected (t ¼ 0), Ji’s model reduces to the two phase model of Takayanagi [19]: " ~ E ¼ E0 ð1  bÞ þ b ð1  bÞ þ b E1 E0 #1 (3) Fig. 11 shows the comparison between the experimental data and the results provided by Eq. (2) and Eq. (3) by considering that E0 ¼ 900 MPa (see Section 3), E1 ¼178 GPa (see [20]) and that the platelets have a thickness of 50 nm (see Section 2). It is noted that the Takayanagi model prediction underestimates the Young’s modulus but the Ji model is able to reproduce the experimental data by considering well chosen interphase thickness t and ratio k, these characteristics being very hard to characterize. We found that t ¼ 9.5  107 m and ratio k ¼ 2.95. These results emphasize the great reinforcement effect of the nanoplatelets. Indeed, the thickness of the platelet is 50 nm but the model leads to an interphase with a thickness of 950 nm and having reinforcing properties compared to the matrix. This particular point needs further research with the help of microscopic observations. 4.2. Yield stress For the prediction of the yield stress as a function of the MMT weight ratio (or volume fraction), the model proposed by Pukanszky et al. [2] is considered: 40 60 35 50 Strength (Mpa) Stress (Mpa) 30 25 20 Tensile Test 15 Yield Stress Tangent 10 40 30 20 10 5 0 0 0 0,05 0,1 0,15 strain Fig. 7. Determination of the yield stress. 0,2 0 1 2 3 4 5 Weight ratio (%) 6 7 Fig. 9. Strength versus weight ratio of reinforcement. 8 L. Cauvin et al. / Polymer Testing 29 (2010) 245–250 0,2 40 0,18 Yield Stress (Mpa) Strain at rupture EXP. 35 0,16 0,14 0,12 0,1 0,08 0,06 0,04 Pukanszky's model 30 25 20 15 0,02 0 0 1 2 3 4 5 6 7 8 10 0 1 Weight ratio (%)     1  fð1Þ s~y  Exp Bfð1Þ ¼  0 sy 1 þ 0; 25fð1Þ sy (4) ! (5) ~y and s0y are the yield stress of the nanocomposite where s and of the matrix, respectively; siy is the yield stress of the interface, r(1) the reinforcements density and S(1) their specific surface area. This model is an empirical correlation developed to describe the tensile yield stress in heterogeneous polymer systems and is a good first step to predict the non linear properties of the considered new class of materials. As shown on Fig. 12, the model of Pukanszky et al. is able to reproduce the strong reinforcement effect of the nano-platelets observed on the yield stress. This model underestimates the reinforcement effect, the average of the error between experimental data and predictions being less than 30%. This can be considered as acceptable. The thickness of the interphase considered in the model is the one obtained during the study of the elastic properties by means of Ji’s model. siy has been taken equal to 2.88  1015 MPa, this value being found to provide the best modelling results. 2300 Exp. 2100 Takayanagi's model Ji's model E (MPa) 1900 1700 1500 1300 1100 900 0 0,5 1 1,5 3 4 5 6 7 Fig. 12. Comparison for the yield stress between the experimental data and the predictions of the Pukanszky’s model. with:   siy 1 þ srð1Þ Sð1Þ ln 0 2 MMT Wt (%) Fig. 10. Strain at rupture versus weight ratio of reinforcement. B ¼ 249 2 2,5 3 Volume fraction (%) Fig. 11. Comparison for the Young’s modulus between the experimental data and the predictions of the two considered models. 5. Conclusions A significant increase of the stiffness of a polypropylene reinforced by MMT nano-platelets has been shown from the experimentation. A previous study [15] had attributed this reinforcement effect to the shape of the platelets. A more physical reason is their nanoscopic size which can induce an interphase in which the properties of the matrix surrounding the platelets is modified. The predictions of the model proposed by Ji [1] have shown good agreement with experimental data, provided that a suitable choice of the properties of the interphase is made. Concerning the yield stress, the model of Pukanszky et al. [2] also provided quite good agreement but, contrary to the Ji’s model, this model is an empirical correlation. It must be noted that several properties of the interphase are hard to measure and have to be chosen: thickness, Young’s modulus and yield stress of the interface matrix/platelet. Current researches consist of directly taking into account, in an homogenization scheme, surface effects (between matrix and platelet) by considering an interface at the place of the interphase, as in [21,22,23]. The consideration of this modelling approach will allow reduction of the number of parameters required. References [1] X.L. Ji, J.L. Jing, W. Jiang, B.Z. Jiang, Tensile modulus of polymer nanocomposites. Polym. Eng. Sci. 42 (5) (2002) 983–993. [2] V. Pukanszky, E. Fekete, Adhesion and surface modification. Adv. Polym. Sci. 139 (1999) 109–153. [3] Y. Kojima, A. Usuki, M. Kawasumi, A. Okada, Y. Fukushima, T. Kurauchi, O. Kamigaito, Mechanical properties of nylon-6-clay hybrid. J. Mater. Res. 6 (1993) 1185–1189. [4] A. Usuki, M. Kawasumi, Y. Kojima, A. Okada, T. Kurauchi, O. Kamigaito, Swelling behavior ot montmorillonite cation exchanged for u-amino acid bi e-caprolactam. J. Mater. Res. 8 (1993) 1174–1178. [5] M. Alexandre, P. Dubois, Polymer-layered silicate nanocomposites: preparation, properties and uses of a new class of materials. Mat. Sci. Eng. 28 (2000) 1–63. [6] J.M. Gloaguen, J.M. Lefebvre, Plastic deformation behaviour of thermoplastic/clay nanocomposites. Polymer 42 (2001) 5841–5847. [7] T.D. Fornes, D.R. Paul, Modelling properties of nylon6/clay nanocomposites using composite theories. Polymer 44 (2003) 4993–5013. [8] Na Wang, Qinghong Fang, Yawei Shao, Jing Zhang, Microstructure and properties of polypropylene composites filled with coincorporation of MCM-41(with template) and OMMT nanoparticles prepared by melt-compounding. Mater. Sci. Eng. A 512 (2009) 32–38. 250 L. Cauvin et al. / Polymer Testing 29 (2010) 245–250 [9] C. Ding, D. Jia, H. He, B. Guo, H. Hong, How organo-montmorillonite truly affects the structure and properties of polypropylene. Polym. Test. 24 (2005) 94–100. [10] J.-I. Weon, H.-J. Sue, Effects of clay orientation and aspect ratio on mechanical behavior of nylon-6 nanocomposite. Polymer 46 (2005) 6325–6334. [11] V. Marcadon, Effets de taille et d’interphase sur le comportement mécanique de nanocomposites particulaires (in French). Département Mécaniques et Matériaux de l’école Polytechnique, Paris, France, 2005. [12] S.A. Gârea, H. Iovu, A. Nicolescu, C. Deleanu, A new strategy for polybenzoxazine-montmorillonite nanocomposites synthesis. Polym. Test. 28 (2009) 338–347. [13] A.S. Luyt, M.D. Dramićanin, Ž Antić, V. Djoković, Morphology, mechanical and thermal properties of composites of polypropylene and nanostructured wollastonite filler. Polym. Test. 28 (2009) 348–356. [14] N. Aı̈t Hocine, P. Médéric, T. Aubry, Mechanical properties of polyamide-12 silicate nanocomposites and their relations with structure. Polym. Test. 27 (2008) 330–339. [15] L. Cauvin, N. Bhatnagar, M. Brieu, D. Kondo, Experimental study and micromechanical modeling of MMT platelet-reinforced PP nanocomposites. C.R. Mecanique 335 (2007) 702–707. [16] P. Ponte Castañeda, J.R. Willis, The effect of spatial distribution on the effective behavior of composite materials and cracked media. J. Mech. Phys. Solid. 43 (1995) 1919–1951. [17] A. Bafna, G. Beaucage, F. Mirabella, S. Mehtab, 3D hierarchical orientation in polymer-clay nanocomposite films. Polymer 44 (2003) 1103–1115. [18] F. Lequeux, L. Monnerie, H. Montes, D. Long, P. Sotta, J. Beriot, Filler-elastomer interaction in model in filled rubbers, a H NMR study 53. J. Non-Crystalline Solid. 307–310 (2002) 719–724. [19] M. Takayanagi, S. Vermura, J.J. Minami, Polym. Sci. Part C 5 (1964) 113. [20] K.S. Alexandrov, T.V. Ryshova, Bull. Acad. Sci. USSR Geophys. Ser. 12 (1961) 1165–1168. [21] H. Le Quang, Q.-C. He, Variationnal principles and bounds for elastic inhomogeneous materials with coherent imperfect interfaces. Mech. Mater. 40 (2008) 865--884. [22] P. Sharma, S. Ganti, Size-dependent Eshelby’s tensor for embedded nano-inclusions incorprating surface/interface energies. J. Appl. Mech. 71 (5) (2004) 663–671. [23] H.L. Duan, J. Wang, Z.P. Huang, B.L. Karihaloo, Size-dependent effective elastic constant of solids containing nano-inhomogeneities with interface stress. J. Mech. Phys. Solid. 53 (2005b) 1574–1596.