Essais & Simulations n°118

Essais

Essais et Modelisation CTE measured during heating is lower (Figure 4). By increasing the heating rate the gap between the curves is also increasing (Figure 5). On the opposite, if the heating and cooling rate approaches zero, this gap is expected to decrease and the two curves will merge into a single one. This behavior The delta L evolution as a function of temperature for the polyamide resin reinforced with aluminum particles is represented in figure 7. Evolutions are observed cycle after cycle. is plotted as a function of the number of cycles for heating and cooling (Figure 9). Discussion 1- CTE and dependence of temperature direction variation CTE and transitions For polymers, the most significant thermal expansion takes place during phase transitions (glass and liquid-liquid transitions for instance). These phenomena are characterized by macromolecular rearrangements inside the material. Each motion of polymer segment has its own relaxation time and the whole collection gives the time relaxation distribution. For a polymer, relaxation times are following a very wide Debye distribution [6] around the characteristic temperature (the transition is spreading over a broad temperature range whose width is about 100°C). is represented in Figure 4 with a dotted line. This particular behavior can be linked to phase transitions, and therefore the sample was also analyzed by DSC. Results of the pure and reinforced epoxy resin are presented on the same diagram in Figure 6. The shift in the baseline of the diagram observed between -20°C and + 60°C is attributed to the glass transition of the resin. This transition is spreading over a broad temperature range whose width is about 100°C, both for the pure and for the reinforced resin. Resin reinforced with aluminum particles In order to quantify this phenomenon, the evolution is arbitrary studied around 30°C (Figure 8). Unlike the epoxy composite, the thermal dilatation of this composite decreases and stabilizes during the first cycles (the length of red and blue arrows which represents the evolution between two cycles is decreasing cycle after cycle). As already shown, the thermal dilatation is not the same for heating and cooling (the gap between these two values represented by the grey dotted arrows is constant cycles after cycles). To quantify the stabilization of CTE value at 30°C, CTE(T) which the derivative of delta evolution The range of thermal cycle for CTE measurement is [-25°C, 45°C] i.e. in the temperature range of the glass transition for the resin (Figure 6). To measure CTE, temperature ramps of one degree per minute are applied to the material. With such heating and cooling rate, the motions associated with long relaxation times have not the time to occur. The transition is incomplete and internal stresses are generated in the material instead of being totally relaxed. Added carbon nano-tubes must modify the microscopic relaxation mechanisms. They probably prevent or disturb molecular motion and modify the relaxation time distribution associate to the transition. Therefore the CTE depends on the temperature in the transition domain and also the value is not the same on heating and on cooling at the same temperature. A similar behavior has already been observed in a study made on polymer/ montmorillonite nanocomposites [7]. DSC analysis showed an increase in the glass transition temperature of numerous polymers when clay was Essais & Simulations • SEPTEMBRE 2014 • PAGE 35

Essais et Modelisation added. This effect was typically ascribed to the confinement of intercalated polymer within the silicate galleries that prevents the segmental motions of the polymers chain. For HDPE, Dzenis and Ponomarev have also shown that addition of a dispersed filler within the polymer results in a decrease of CTE [8]. Internal thermal stresses effects In composite materials in general, and particularly in the two studied reinforced resins, several phases which have different thermo-mechanical behaviors are coexisting [4]. In this case, the CTE of the matrix is much higher than the one of reinforcements. Quantitatively, the ratio between matrix CTE and particles CTE is 20 for epoxy-carbon composite and 5 for the polyamide-aluminum composite. Also, during heating and cooling intense thermal stresses are generated between matrix and reinforcing particles. Specifically, during heating, the matrix expands more than reinforcing particles and is loaded in compression. Conversely, during cooling the matrix is loaded in traction. Thus, at the end of heating, the thermal stresses are at their maximum: the matrix is compressed and the measured CTE is lower than the “real” CTE (Figure 4) because part of the transformation is prevented by internal stresses between the matrix and reinforcing particles. Just after, the direction of temperature variation is reversing and these compressive stresses are relaxed, the measured CTE is also higher than the References 1 J. PENEL, A. BETTACCHIOLI. Mise au point d’un banc de dilatométrie sous vide et influence de l’histoire thermique sur le coefficient de dilatation thermique d’un matériau sandwich. Essais & Simulation n°111, pp.47-49, octobre 2012 2 R. A. SHANKS. Linear thermal expansion, thermal ageing, relaxations and post-cure of thermoset polymer composites using modulated temperature thermomechanometry. Journal of Thermal Analysis and Calorimetry, (2011) 106:151-158. 3 H. JIANG, B. LIU, Y. HUANG, K.C. HWANG. Thermal Expansion of Single Wall Carbon Nanotubes. July 2004, Vol. 126. 4. K. SONG, X. GUO, S. LIANG, P. ZHAO, Y. ZHANG. Relationship between interfacial stress and thermal expansion coefficient of copper–matrix composites with different reinforced phases. Materials Science and Technology, July 2013. 5 T. CHOTARD, M. HIGER, J. SORO. Caractérisation du comportement mécanique endommageable de réfractaires à haute température par couplage de techniques ultrasonores. 18ème Congrès Français de Mécanique. Grenoble, Aout 2007. Proceedings ? 6 M. STEFENEL. Etude des mouvements moléculaires dans les polymères biphasiques à résilience améliorée polyamides-élastomère par fluage stimulé par la température. PhD Thesis, Université Paul Sabatier, 1984. 7 A. LESZCZYNSKA, K. PIELICHOWSKI. Application of thermal analysis methods for characteri-zation of polymer/ montmorillonite nanocomposites. Journal of Thermal Analysis and Calorimetry, Vol. 93 (2008) 3, 677-687. 8 YU. A. DZENIS, V. M. PONOMAREV. Thermal expansion of a polymer composite with an aggregating disperse filler. Institute of Polymer Mechanics, Academy of Sciences of the Latvian SSR, Riga. Junuary-february, 1989. 9 J-M. HAUSSONNE. Traités des Matériaux, Presses polytechniques et universitaires romandes, 2005 10 G. POMMATAU, M. NIVET-LUTZ. MMC tubes and advanced assemblies for high stability space structures. Meeting Le Bourget 20-06-2001. Proceedings ? 11 D. LUCA MOTOC, J. IVENS, N. DADIRLAT. Coefficient of thermal expansion evolution for cryogenic preconditioned hybrid carbon fiber/glass fiber-reinforced polymeric composite material. Journal of Thermal Analysis and Calorimetry 2013, Vol. 112, p. 1245-1251. 12 TOMPKINS S. S. Thermal Expansion of Selected Graphite-Reinforced Polyimide-, Epoxy, and Glass-Matrix Composite. Journal of Thermophysics, Vol. 8, No 1, 1987. Essais & Simulations • SEPTEMBRE 2014 • PAGE 36