Influence of the degree of exfoliation on the thermal conductivity of polypropylene nanocomposites.
Nanocomposites are filled polymers whose particles are nanoscale in at least one dimension and exhibit significantly higher performance than commonly used fillers regarding higher elastic modulus, tensile strength, thermal resistance, lower gas and liquid permeability, reduced flammability, and enhanced rheological properties (e.g., increased melt stiffness). These attributes are easily attained with small amounts of filler (1-11). Layered silicates (especially montmorillonite, MMT) are the most commonly used nanofillers as it is possible to achieve aspect ratios of up to 1000.
Beside the increase of specific material properties, important material parameters change significantly as well. One such material parameter is thermal conductivity, crucial to all polymer processing; the simulation and configuration of heating and cooling procedures; simulation of injection molding processes; dimensions of thermal insulation materials; etc.
It is well known that mineral fillers increase thermal conductivity significantly with greater filler content depending on type, size, and structures. When it comes to layered silicates, thermal conductivity not only depends on the filler content but also on the structure of the particles in the matrix--the so-called degree of exfoliation.
Definition of Thermal Conductivity
Basically, there are three different possibilities of heat transfer: radiation, convection, and conductivity.
Thermal conductivity is the intermolecular transport (transfer) of heat, if there is an unequal temperature distribution within a system (12), (13). The energy transport occurs by interactions of molecules and atoms. Considering Eq. 1, the thermal conductivity [lambda] is the proportionality factor between the heat flow [DELTA]Q and the temperature gradient [DELTA]T (14).
[DELTA]Q/[DELTA]t = [lambada] x A x [DELTA]T/[DELTA]x (1)
The unit of the thermal conductivity is W/mK.
Therefore, the thermal conductivity is a ratio for the energy transport in a volume. Using Eq. 2, it is defined as the heat quantity Q (in Joules), which goes through a specific square A in a specific time--t--if the thermal gradient dT is 1 K (15).
Thermal Conductivity of Polymers
Polymers do not have free moving electrons. Therefore, diffusion mechanisms cannot contribute to thermal conductivity, and the energy transfer is limited to the oscillation of thermally stimulated chain molecules. The thermal conductivity of polymers is based on two different mechanisms. On one hand, heat is transferred through Van der Waals forces, and on the other hand, phonons are actuated through covalent bonding. This, hence, results in an intermolecular heat transport from one macromolecule to another (16).
TABLE 1. Formulation and process conditions of the nanocomposites Filler content (wt %) Screw speed (rpm) 2 100 300 3.5 100 300 5 100 150 200 250 300
This model goes back to Debye, who made following correlation
[lambda] [congruent to] 1/3 x [C.sub.p] x [rho] x V x 1 (2)
where the thermal conductivity is directly proportional to the specific heat capacity cp, density p, sonic speed v, and the mean free length of path 1 of the elastic waves (14).
The thermal conductivity of polymers is influenced by various factors such as temperature, pressure, morphology, molecule orientation, chemical structure, and thermal stress (13-17). Additionally, thermal conductivity can be influenced significantly by the addition of different fillers and additives as the thermal conductivity of most inorganic fillers is around 10 times higher than that of polymers (metal factor 1000). The subsequent change in thermal conductivity depends on the type and amount of the used fillers. For example, quartz powder, glass, metals, carbon, or other mineral and inorganic fillers like talc or calcium carbonate increase thermal conductivity (13), (16), (17). In a heterogeneous system like a filled polymer, the thermal conductivity of the total system is mainly determined by the thermal conductivity of the individual components (18).
Furthermore, factors like particle shape, size, contribution, and orientation influence the overall thermal conductivity as well. Furthermore, some filler like talcum act as a nucleating agent, influencing crystallization, and consequently thermal conductivity (18-20).
Preparation of Polymer Nanocomposites
The used polymer matrix was an isotactic polypropylene homopolymer (BorECO BA212E; MFR 0.3 g/10 min, 230[degree]C/2.16 kg; Borealis/A). A layered silicate intercalated with dimethyl dimethyl ammonium chloride was used as a nanofiller (Nanofill 5, Rock wood/D, [lambda] = 0.6 W/mK). A PP grafted with maleic anhydride (Scona TPPP 2112 Fa. Kometra/D) was taken as a compatibilizer, using the same content as the nanofiller. For the compounding process, a corotating twin screw extruder Theysohn/A TSK30/40D was used. The feed rate of the dosing system was set at 10 kg/h, and the screw speed was varied from 100 to 300 rpm. The extruder temperature profile was set to 160-200[degree]C from the hopper to the die. The different formulations and process conditions are listed in Table 1.
Sniall-Angle X-Ray Scattering
X-ray measurements were performed using Bruker NanoSTAR (Bruker AXS, Karlsruhe, Germany) small angle X-ray scattering equipment. This system was equipped with a two-dimensional X-ray detector. A wave-length of 0.154 nm (Cu K[alpha]) was used. The samples were measured in transmission. All scattering measurements were performed on plate samples to avoid the influence of texture. To obviate statistical effects, the scattering curves recorded at three different positions on the samples were averaged. Scattering curves were corrected for background scatter, and a Lorenz correction was applied to determine the gallery period. Finally, the interlayer distance as well as the area A under the silicate layer peak was calculated (see Fig. I) in order to evaluate the degree of exfoliation. This area is proportional to the number of panicles hit by the X-ray beam. The better the nanocom-posite is exfoliated, the lower the probability of hitting particles and therefore the smaller the area is.
Transmission Electron Microscopy
The TEM experiments were performed using a Zeiss LEO 912 Omega transmission electron microscope (Carl Zeiss, Jena, Germany) using an acceleration voltage of 120 kV. The samples were prepared using a Leica Ultracut UCT ultramicrotome (Leica Microsystems, Wetzlar. Germany) equipped with a cryo chamber. Thin sections of 50 nm were cut using a diamond blade (Diatome AG, Biel, CH) at -120[degrees]C.
TABLE 2. Hot Disk measurement parameters. Hot Disk, Transient Plane Source ISO 22007-2/2008 Mode Heating Temperatures 40-120[degree]C Temperatures steps 20[degree]C Heat output 0.04 W Sensor diameter 4 mm Sensor thickness 70 [micro]m Measurement (heating.) time 5s Time between two measurements for a 20 min constant-temperature environment Repetition 3 X
Thermal Conductivity Measurements
The thermal conductivity in solid state (temperature range 40-120[degrees]C) was measured using the "Transient Plane Source" technique according to ISO 22007-2/2008 using a TPS 2500S, Hot Disk AB, Sweden. The samples were vacuum compression-molded plates with the dimensions 16 X 160 X 4 mm. Table 2 shows the Hot Disk measurement parameters.
A high-pressure capillary rheometer (HPCR; Rheograph 2002, Gottfert, Germany) was used for measuring the thermal conductivity based on the "Line-Source" method according to ASTM D5930-97 in the temperature range from 80 to 200[degrees]C. Table 3 shows the HPCR measurement parameters.
RESULTS AND DISCUSSION
Figures 2 and 3 show the results of the XRD measurements. It is obvious that with increasing screw speed (higher induced energy), the interlayer distance and accordingly the degree of exfoliation increases (see samples with 5 wt% filling degree from 100 to 300 rpm screw speed), whereas the peak area decreases. Additionally, the degree of exfoliation is influenced by the amount of layered silicates shown by the TEM pictures of the samples with 2-5 wt% filling degree at 300 rpm in Fig. 4.
TABLE 3. HPCR Disk measurement parameters. HPCR, Line Source ASTM D5930-09 Mode Cooling Temperatures 80-200[degree]C Temperatures steps 20[degree]C Heat output per unit length 24.36 W Inner diamerter of sample cell 15 mm Diameter/length of people 12/160 mm Measurement (heating.) time 20s Time between two measurements for a 120 min constant-temperature environment Repetition 3 X
It is apparent that with increasing tiller content, the degree of exfoliation also rises. The individual platelets are more isolated and distributed homogeneously. These results will be the basis for the evaluation of the thermal conductivity--degree of exfoliation dependencies.
Figure 5 shows the influence of the clay content on the thermal conductivity of polypropylene nanocomposites. The thermal conductivity in a solid state slightly decreases with rising temperatures and is constant to a large extent in melt state. The isobaric cooling plots show a typical "Z" shape for each of these curves indicating a phase transition. As expected, the higher the MMT content, the higher the thermal conductivity, which is known from conventional fillers. The increase of thermal conductivity, already with a very small amount of filler (2-5 wt%), is substantial. This is particularly surprising, as the thermal conductivity of MMT is only three times higher than that of the polymer (compared to factor 10 of talcum and calcium carbonate).
Figure 6 shows the influence of the screw speed and the degree of exfoliation, respectively, on the thermal conductivity of the 5 wt% nanocomposites processed at different screw speeds (100-300 rpm).
It is evident that an increased screw speed, which is equal to a higher degree of exfoliation, has a significant influence on thermal conductivity. The better the particles are exfoliated, the higher the thermal conductivity (see Fig. 7). Although the microparticles decrease with rising degrees of exfoliation, the overall specific surface for interaction in terms of heat transfer grows. Furthermore, the influence of the silicate layers on the morphology (by, e.g., formation of a 3D physical network or by acting as a nucleation agent) of the polymer increases with rising degrees of exfoliation.
Another interesting fact is the significantly lower increase of thermal conductivity from melt to solid state compared to neat polypropylene. In melt state, the intercalated and exfoliated layers act as physical nodes forming a 3D percolation network (2), (21), (22), which improves the energy transport and the heat transfer caused by elastic waves. We have already shown that intercalated as well as exfoliated layered silicates change the morphology to virtually increased molar mass observable in increased zero-shear viscosity as well as melt stiffness (12). The stiffer and longer the chains, the better the energy transmission is.
In a solid state, crystallization is the dominating factor (almost 20% increase in thermal conductivity) diminishing the disparity in thermal conductivity (16), (23). The influence of the percolation network is essentially lower in a solid system (including crystalline areas) than in a liquid system (without any crystalline areas). The fillers act through their volume resulting in a only minor increase with the nanofilled (<8 vol%) composites. Additionally, the increase of thermal conductivity is much smaller (see Fig. 7). It can be put into perspective that already intercalated structures (e.g., PP 5% at 100 rpm ) lead to a significant increase in thermal conductivity.
Figure 7 shows a comparison of thermal conductivity increase of PP-composites filled with various filters. These results clearly show that using nanoclay leads to a significantly higher increase in thermal conductivity already with a factor of 10 less filling grade. In other words, the increase of thermal conductivity in a molten state with wt% layered silicates is higher than with 30 wt% talcum 30 wt% calcium carbonate, respectively.
Considering the three times lower thermal conductivity of layered silicates compared to talcum or calcium carbonate makes these results even more interesting than presumed. It is obvious that the silicate layers enhance the thermal conductivity not only with their own conductivity but also with their influence onto the morphology by a significantly higher specific surface and virtual chain enhancement (3D percolation network). This model also explains that the thermal conductivity is only increased in melt state as their influence is depleted due to crystallization. The impact of chain enhancement is relativized in a solid state, where the results only reflect the influence due to the filler content, and therefore the increase is much lower than that of the conventional fillers.
Talcum and calcium carbonate influence overall thermal conductivity only with their allocated volume rate with increased thermal conductivity, which is similar in both molten and solid states. Only the energy and heat transfer from polymer to filler as well as within the polymer crystallites is more effective in a solid state. Therefore, the increase in thermal conductivity is almost equal in solid and molten states with a small raise clue to the aforementioned factors.
Another interesting point is the particle shape, as the plate-shaped talcum particles with a lower thermal conductivity than the cubical-shaped calcium carbonate particles (2.1 W/mK compared to 2.5 W/mK) lead to a higher increase in the filler-polymer compound. Layered silicates have similar plate-shaped particles (in the range of nanommeter) resulting in a significantly higher influence on thermal conductivity.
Generally, the results showed that the change of thermal conductivity cannot be neglected when processing polymer nanocomposite melt as already a small amount of filler has a massive influence, for example, on simulation and configuration of heating as well as cooling procedures or injection-molding simulation (percental increase in thermal conductivity corresponds almost linear to percental reduction in cycle time for 1-mm wall thickness parts (4)). Considering the relatively low increase of thermal conductivity in a solid state, applications (e.g., insulation or community heating pipes) are more or less not affected.
The results in melt state clearly show that, beside the filler content, the structure of the layered silicates is a very important factor. There is a major increase of thermal conductivity from agglomerated to intercalated/exfoliated structures, whereas the difference within the intercalated/exfoliated samples is minimal. This indicates that the thermal conductivity (energy transport) is directly linked to the formation of the 3D physical network. If this is established, the further increase of thermal conductivity is minimal.
Compared to conventional fillers, the increase of thermal conductivity is significantly higher as 3.5 wt% ('-8 vol%) of layered silicates achieve almost the same thermal conductivity as 30 wt% vol%) of talcum or calcium carbonate. This is particularly astonishing, as layered silicates have a lower thermal conductivity by a factor of three compared to talcum or calcium carbonate.
In the solid case, the increase of thermal conductivity due to crystallization is the dominating factor. The 3D network has only a very low impact, and the nanoclay only operates through the volume (like conventional fillers) resulting in a small increase of thermal conductivity.
These results are very important for processing polymer nanocomposites as they can be used for faster heating and/or cooling cycles, shorter cooling sections, or the like. Furthermore, a higher thermal conductivity can help save energy by reducing the energy needed for heating and cooling the material. Besides that benefit and considering the relatively low increase of thermal conductivity in a solid state, applications are more or less not affected.
(1.) J.W. Gilman, T. Kashiwagi, and J.D. Lichtenhan, SAMPE .1., 33, 40 (1997).
(2.) S.S. Ray and M. Okamoto, Pro, . Polym. Sci., 28, 1539 (2003).
(3.) S. Laske, M. Kracalik, M. Gschweitl, M. Feuchter, G.
Maier, G. Pinter, R. Thomann, W. Friesenbichler, and G.R.
Langecker, Appl. Polym. Sci., 111, 2253 (2009).
(4.) A. Sanchez-Solis, 1. Romero-Ibarra, M.R. Estrada, F. Calde-ras, and 0. Marrero, Po&tn. Eng. Sci., 44, 1094 (2004).
(5.) H. lshida, S. Cambell, and J. Blackwell, Chem. Mater., 12, 1260 (2000).
(6.) R.A. Vaia and E.P. Giannelis, Macromolecules, 30, 8000 (1997).
(7.) S. Laske, M. Kracalik, M. Feuchter, G. Pinter, G. Maier, W.Marzinger, M. Haberkorn, and G.R. Langecker, Appl. Polym.
Sci., 114, 2488 (2009).
(8.) D.H. Kim, J.U. Park, K.S. Cho, K.H. Ahn, and S.J. Lee, Macromol. Mater. Eng., 291, 1127 (2006).
(9.) N. Hasegawa, M. Kawasumi, M. Kato, M. Usuki, and A.J.
Okada, Appl. Polyin. Sci., 87, 67 (1998).
(10.) M. Maxfield, L.W. Shacklette, R.H. Baughman, B.R. Chris-tiani, and D.E. Eberly, PCT Int. Appl. WO 93/04118 (1993).
(11.) S.S. Lee and J. Kim, Polym. Mater. Sci. Dig., 89, 370 (2003).
(12.) M. Kaviany, Principles of Heat Transfer, Wiley-Interscience, Hoboken NJ (2001).
(13.) C. Kucher, Thermal Conductivity of Filled and Unfilled Polypropylene in Melt and Solid State Under High Pressures, Master thesis, Montanuniversitat Leoben (2008).
(14.) T.A. Osswald and G. Menges, Materials Science of Polymers for Engineers, Carl Hanser Verlag, Munich, 2003.
(15.) H. Schmiedel, Handbuch der Kunststoffpnifthig, Hanser Ver-lag, Munich (1992).
(16.) T.A. Osswald and G. Menges, Materials Science of Polymers for Engineers, Hanser, Munich (2003).
(17.) A. Dawson, M. Rides, and J. Nottay, Polym. Test., 25, 268 (2006).
(18.) Z. Han and A. Fina, Pro,. Polym. Sci., 36, 914 (2011).
(19.) D.M. Bigg, "Thermal Conductivity of Heterophase Polymer Compositions," in Thermal and Electrical Conductivity of Polymer Materials, Y.K. Godovsky and V. P. Privalki, Springer, Berlin/Heidelberg, I (1995).
(20.) I. Duretek, W. Friesenbichler, and C. Kuchcr, in the 25th Polymer Processing Society Annual Meeting, Conference Proceedings, Goa, India (2009).
(21.) B. Hoffmann, C. Dietrich, R. Thomann, C. Friedrich, and R. Miilhaupt, Macromol. Rapid Commun., 21, 57 (2000).
(22.) H. Munstedt, T. Koppl, and C. Triebel, Polymer, 51, 185 (2010).
(23.) H. Lobo and J.V. Bonilla, Handbook of Plastics Analysis, Marcel Dekker Inc., New York (2003).
Stephan Laske, (1) Ivica Duretek, (1) Andreas Witschnigg, (1) Hannelore Mattausch, (1) Daniel Tscharnuter, (2) Clemens Holzer' (1)
(1.) Department of Polymer Engineering and Science, Chair of Polymer Processing, Montanuniversitaet Leoben, Otto Gloeckel-Strasse 2, 8700 Leoben, Austria
(2.) Polymer Competence Center Leoben GmbH, Peter Roseggerstrasse 12, 8700 Leoben
Correspondence to: Stephan Laske; e-mail: stephanlask unileoben. ac.at
Published online in Wiley Online Library ovileyonlinelibrary.com).
[c] 2012 Society of Plastics Engineers
|Printer friendly Cite/link Email Feedback|
|Author:||Laske, Stephan; Duretek, Ivica; Witschnigg, Andreas; Mattausch, Hannelore; Tscharnuter, Daniel; Holz|
|Publication:||Polymer Engineering and Science|
|Date:||Aug 1, 2012|
|Previous Article:||Rheological properties and morphological evolutions of polypropylene/ethylene-butene copolymer blends.|
|Next Article:||In situ synthesis, morphology, and thermal properties of polystyrene--MgAl layered double hydroxide nanocomposites.|