Rheological and viscoelastic properties of multiphase acrylic rubber/fluoroelastomer/polyacrylate blends.
The development of high performance polymer blends for industrial applications involves careful selection and characterization of constituent polymers (1, 2). Miscibility or compatibility is considered one of the important requirements for useful applications (3). The use of fluorocarbon rubber (FKM) and polyacrylate rubber (ACM) for automotive, aerospace and other industrial applications Is widespread because of their characteristic high temperature and fluid resistance properties (4). Blending these polymers is an alternate route to combat high cost and poor processablity of fluoroelastomers. Adding a third polymer for further improving performance of these blends has been a subject of research for some time.
The miscibility of many commercially important rubber-rubber and rubber-plastics blends has been reviewed in the literature (5-8). ACM/FKM blends show complete miscibility at all compositions characterized by nuclear magnetic resonance spectroscopy [H.sup.1]NMR), infrared spectroscopy (IR), dynamic mechanical thermal analysis (DMTA) and differential scanning calorimetry (DSC) (9). The processability and die swell characteristics of the same blends with and without fillers reveal good processing behavior of the filled blends (10). Addition of a reactive monomer during the mixing of polymer/polymer blends is a novel technique to enhance the physico-mechanical properties of the system through grafting, polymerization and partial crosslinking. Polyfunctional monomers including multifunctional acrylates have been used to accelerate grafting of styrene onto polyethylene, polypropylene and cellulose substrates (11). The grafting of multifunctional acrylate on the fluorocarbon rubber increases the mechanical properti es after radiation curing (12). Thermoplastic elastomers derived from blending of ACM/FKM and polyfunctional acrylates show improved mechanical, solvent resistant, and heat aging properties (13). The dynamic viscoelastic properties of fluoroelastomers and acrylics used for soft denture liners show more sensitivity towards frequency and exhibit viscoelasticity (14). Transmission electron microscopic study reveals that the ACM/FKM/polyacrylate blend has a multiphase morphology (15).
The rheological and viscoelastic properties of multiphase polymer blends show complex behavior due to the difference in the flow behavior of the individual polymer in the blend (16-21). For such systems, multiple phase morphology directly influences the rheological properties. The viscosity of the components, composition, interfacial interaction between phases, and processing parameters are found to be the main factors that influence the morphology of multiphase polymer blends (22).
Though the rheological and viscoelastic properties of many blends showing multiphase morphology have been discussed, the correlation between rheological properties and morphology resulting from phase separation for an ACM/FKM/polyacrylate blend has not been established. In this study, by employing rheometry, we have focused on the influence of composition and amount and nature of polyfunctional acrylates on the rheological properties of the ACM/FKM/polyacrylate blends. The viscoelastic properties of some representative samples have also been studied through stress relaxation phenomena.
Acrylate rubber (ACM), NIPOL AR 51 (density-1100 kg [m.sup.-3] at 25[degrees]C, Mooney viscosity ML (1 + 4) at 100[degrees]C - 55, Tg = -4[degrees]C) containing epoxy cure site monomer was obtained from Nippon Zeon Co. Ltd., Tokyo, Japan. Fluorocarbon rubber (FKM), Viton B 50 (density 1850 kg [m.sup.-3], 68% F, 1.4% H, Tg = -0[degrees]C) was supplied by DuPont Dow Elastomers. The acrylate monomers-(i) hexanediol diacrylate (HDDA, density 1050 kg [m.sup.-3]), (ii) trimethylolpropane triacrylate (TMPTA, density 1110 kg [m.sup.-3]) and (iii) dipentaerythritol hexacrylate (DPHA, density 1110 kg [m.sup.-3]) were procured from UCB Chemicals, Belgium. The molecular structures of ACM, FKM and multifunctional acrylates have been given in our earlier publication (13).
Preparation of the Samples
ACM was mixed with FKM in the following blend ratios-100/0, 70/30, 50/50, 30/70 and 0/100 (w/w) in a Brabender Plasticorder, PLE 330 at 100[degrees]C for 5 minutes at a rotor speed of 60 rpm. Benzoyl peroxide, an initiator for polymerization of polyfunctional acrylates, at a level of 0.1% was added and the mixing was continued for 5 minutes. The liquid multifunctional acrylate monomer at an appropriate level was then added slowly in the mixer at the same temperature until the mass became homogeneous in about 3 minutes. After mixing, the mass was sheeted out at 30[degrees]C using a laboratory two roll-mill (6" X 13" Schwabenthan, Berlin). The samples were then molded at 170[degrees] for 5 minutes between DuPont Teflon sheets in a two plate hydraulic press (Moore Press, U.K.) provided with cooling circuits and then cooled for 5 minutes at a pressure of 5 MPa. After each molding, the samples were cooled in the press itself to 50[degrees]C. Test specimens of 25-mm diameter were cut from the molded slabs of approx imately 1.8-mm thickness using cutting dies.
Dynamic Rheological Measurements
The rheological and stress relaxation measurements were performed using an ARES RDA III rheometer (Rheometric Scientific, Inc.) with parallel plate geometry. The tests were carried out in steady rate and dynamic frequency modes at three temperatures -160, 180 and 200[degrees]C. Dynamic shear properties were determined as a function of angular speed of deformation 0.1 to 100 rad/s. For all the experiments, the strain amplitude was maintained constant at 1%. The stress relaxation measurement was carried out using the same instrument at 200[degrees]C.
Dynamic Mechanical Analysis
Dynamic mechanical properties of the solid samples were determined by means of stress-strain oscillation measurements using a dynamic mechanical thermoanalyzer ARES RDA III (Rheometric Scientific Inc.). The tests were carried out at a frequency of 1 Hz, and the temperature programs were run from -100 to 200[degrees]C under a controlled sinusoidal strain at a heating rate of 4[degrees]C/min in a flow of nitrogen. The oscillating dynamic strain was 0.1%. The viscoelastic properties. the storage modulus (G'), loss modulus (G"), and the mechanical loss factor (damping), tan [delta] (G"/G') were recorded as a function of temperature.
Transmission Electron Microscopy (TEM)
The morphological studies of the blends were performed using a transmission electron microscope (Hitachi-HT300) operating at 100 kV. The specimens used for TEM were cryo-microtomed at -45[degrees]C and stained with Os[O.sub.4] for better contrast.
RESULTS AND DISCUSSION
Effect of Blend Composition on Rheological Properties
The compositions of the binary and the ternary blends of ACM, FKM and ACM/FKM blends with PTMPTA at a constant level of 30 phr are represented in Table 1. The rheological properties of these blends are characterized by dynamic test under shear mode for storage modulus (G') and complex viscosity ([[eta].sup.*]) at different oscillatory angular frequencies. Figure 1 shows variation of G' with frequency for all blends at 180[degrees]C. It is seen that the G' is strongly dependent on the blend composition. Thus, the storage modulus of AT30 is the highest one among all blends at all frequencies. However, the variation of G' with frequency of this blend is minimal. When the level of FKM is increased, the G' values decrease correspondingly and the binary blend of FKM and PTMPTA has the lowest G' value and the change in its value with frequency is high. This blend and the blends with lower proportions of ACM show a second terminal elastic plateau at the lower frequency side due to creation of additional chains either through polymerization or grafting. In order to verify the trend in G' value with composition, the model proposed by Kerner (23) correlating the effect of composition on storage modulus is used and the curves are also included in the graph. It is seen that the deviation of experimental values from the theoretical model is very low for all the blends with varying compositions. The effect of composition on rheological properties is more clearly pronounced if one considers the variation of viscosity with composition at 180[degrees]C, which is given in Fig. 2. The plot is also provided with a curve obtained from the logarithm rule of mixtures that applies at constant temperature and shear rate (24).
[sigma] [[psi].sub.i] log [[eta].sup.*.sub.i] (1)
where [eta] and [[eta].sub.i] are the viscosities of the blends and of the components respectively. [[phi].sub.i] is the volume fractions of components. The blends have lower values of [[eta].sup.*] than predicted by Eq 1 and they are classified as negative deviation blends (24). This deviation is larger at low frequencies compared to higher frequencies. The change in dynamic modulus and dynamic viscosity with varying composition of ACM and FKM in the blend at constant loading of polyacrylate is given in Figs. 3a and 3b respectively. Both G' and [[eta].sup.*] decrease on increasing the FKM content in the blend. This indicates that FKM is not compatible with the polyacrylate phase and also extracts part of the ACM from the ACM/polyacrylate phase because of the miscibility of ACM and FKM. As a result, the modulus and viscosity decrease with increasing content of FKM in the blend. All the blends show the pseudoplastic behavior, i.e., decrease of viscosity with increasing frequency (which is a measure of shear ra te), and obey the power law (shear dependency) model
The power law coefficients for different compositions at different temperatures obtained from the log-log plot of complex viscosity vs. frequency are given in Table 2. The power law index, [eta], representing the pseudoplasticity of all blends, decreases with increasing temperature, suggesting more non-Newtonian behavior of the systems at higher temperature. By comparing the values of n with composition of the blends, it is seen that the blends with higher proportion of ACM have very low values, indicating a higher amount of grafting and crosslinking between the components of the blends. This, in fact, has very close similarity with the storage modulus values. AT30 blends containing only PTMPTA undergo a high amount of grafting because of the similar chemical natures of the functional groups, i.e., acrylate moiety, and thus form a single-phase system as explained in our earlier work (15).
[[eta].sup.*] = k[([omega]).sup.n-1] (2)
On the contrary, the FT30, containing FKM and PTMPTA, shows a two-phase system due to dissimilar chemical identities (Fig. 4). The effect of temperature on the rheological behavior shows anomalous results. As increase of temperature slightly increases the melt viscosity of all blends, which is more pronounced for FT3O. Therefore the power law coefficient (k), a measure of viscosity at unit frequency, increases with a rise in temperature. When the temperature is raised, the unreacted portion of TMPTA in the blend undergoes polymerization and partial crosslinking in the presence of peroxide added during mixing, leading to an increase in chain entanglement, which restricts the flow of the polymer chain. This may lead to increased phase separation.
The phase separation is influenced by a number of factors, such as the concentration of reactive polyacrylate, compatibility between phases, temperature, presence of residual peroxide and processing conditions (22). The thermally induced crosslinking process is also one of the causes for phase separation in a partially miscible blend. So the compatibility between phases can be studied by understanding the complex viscosity function comprising of viscous and elastic components (25)
[[eta].sup.*] = [eta]' - i[eta]" (3)
where [eta]' is the in-phase elastic component and i [eta]" [eta]" is the out-of-phase viscous component of the dynamic complex viscosity. Moreover, these components are related to the energy stored and the oscillatory frequency by the equation
[eta]' = G"/[omega]; [eta]" = G'/[omega] (4)
where G' is the storage modulus (elastic component) and G" is the loss modulus (viscous component). It is also well known that the representation of dynamic shear data or viscosity data in the Cole-Cole plot (Fig. 5) gives information about the relaxation processes taking place in the multiphase blends. It is assumed that when a blend is miscible or compatible, the Cole-Cole plot is an almost semicircular curve/arc, and we can deduce the average relaxation time of a phase, which is equal to the inverse of the maximal frequency corresponding to the horizontal tangent at the top of the circle/arc. This plot can also be used to determine the zero shear viscosity, [[eta].sub.o] (obtained from the extrapolation of the Cole-Cole plot to the x axis, [eta]' at [eta]" = 0). The present system does not show the miscibility curve due to complex temperature and shear rate dependent rheological behavior and the relaxation time could not be determined. However, it gives an idea about the phase immiscibility at higher tempe rature due to initiation of polymerization and crosslinking. Moreover, the viscosity is increased monotonically for all systems, indicating the absence of an additional relaxation mode as a result of the presence of a polyacrylate phase.
Figure 6 illustrates a log-log plot of G' vs. G" (Han plot) of the blends AT30, AFT30 and FT30 at three temperatures. These blends exhibit both composition and temperature dependent correlation. However, there is closeness of these curves at higher dynamic moduli. This type of dependency shows the presence of immiscibility mainly caused by the presence of polyacrylate phase. It is seen from the figure that the variation is great for FT30 because of complete incompatibility at all temperatures.
As there is a marginal increase in the viscosity with an increase in temperature for most of the blends at lower frequencies, the activation energy calculated from the Arrhenius plot of viscosity vs. inverse of temperature shows the value to be very close, in the range of 1-10 kcal/mole, and a distinction could not be made between the blends.
Variation of Level of Polyacrylate
Figures 7 and 8 show the effect of the level of TMPTA on G' and [[eta].sup.*] of 50/50 (w/w/) ACM/FKM at different frequencies. As the proportion of TMPTA is increased, the storage modulus is also increased in a linear fashion. The corresponding loss modulus has the same trend. However, it is seen that in the low-frequency region, the difference between G' and G" increases with decreasing frequencies, leading to a lowering of tan 8. These low-frequency viscoelastic features are characteristics of a pseudosolid like behavior. True solid like behavior would be said to occur if both G' and G" became independent of frequency with G' vastly exceeding G". This pseudosolid like response dampens the terminal relaxation. Similar features have been observed for ordered block copolymers and have been attributed to the onset of micro domain and defect related dynamics dominating over single-chain relaxation (26-28). The plot of [[eta].sup.*] versus frequency (Fig. 8) shows the pseudoplastic behavior of the blends irrespe ctive of TMPTA loadings. The increase in complex viscosity values on increasing TMPTA level shows the influence of TMPTA in restricting the flow of the polymer chain by undergoing partial crosslinking.
Variation of Nature of Polyfunctional Acrylates
Polyacrylates of different functionalities viz. di-, tri-and hexa-functional reactive groups are chosen to study the effect of functionality. Figure 9 depicts the variation of [[eta].sup.*] with frequency for different acrylates. As the functionality is increased from di-functional to hexa-functional, [[eta].sup.*] is also increased; however, the difference is very low between di- and ti-functional acrylates compared to tri- and hexa-functional acrylate (DPHA). The polyacrylate containing polyfunctional reactive groups normally undergo, in addition to polymerization and grafting, a crosslinking reaction leading to gel formation depending upon the number of reactive groups. In our earlier study we reported the complete phase separation of DPHA from the rubber phase, leading to lower mechanical properties (15).
Dynamic Mechanical Analysis (DMA) of the Blends
DMA was used to investigate the miscibility or compatibility of polymer systems through some kind of specific interaction by various groups (28). Generally for an immiscible blend, the loss tangent (tan [delta]) curves show the presence of two peaks corresponding to the [T.sub.g]'s of individual polymers. For a miscible blend, the curves show only a single peak corresponding to the transition of blend, which may be in between the transition peak of the component polymers, whereas broadening or shifting of transition peak occurs in the case of partially miscible systems. The dynamic mechanical properties of representative samples of AFH30 and AFT30, studied over the temperature range of -100 to 200[degrees]C, are given in Fig. 10. The tan [delta] curve shows a sharp peak at 4[degrees]C due to glass transition of ACM/FKM and another broad peak at around 70[degrees]C corresponding to [T.sub.g] of PTMPTA or PHDDA arising from segmental motion.
The area under the tan [delta] curve is related to the activation enthalpy of relaxation of the backbone motion (29). The average activation enthalpy of transition for a polymer can be calculated from the dynamic mechanical data, as proposed before using the following equation (30)
TA = (ln [G'.sub.G] - ln [G'.sub.R])[pi] [RT.sup.2.sub.g]/2[([delta][H.sub.a]).sub.avg]
where tA is the area under the tan [delta] curve, [G'.sub.G] and [G'.sub.R] are the storage modulus at glassy phase and rubbery phase, respectively, [([delta][H.sub.a]).sub.avg] is the activation enthalpy of the transition process. Accordingly, for a given tA, the height of the tan 8 peak is dependent on the breadth of the transition and hence the range of different relaxation processes contributing to the overall transition. AFT30 shows lower peak height than AFH30. This implies that there is more restriction in the main chain mobility in the AFT30. While considering the storage modulus of these blends, the AFT30 shows higher G' compared to the AFH30. TMPTA, being a tri-functional acrylate, imparts more grafting and partial crosslinking, leading to more restriction of molecular motion at the transition.
Stress Relaxation Measurement
Stress relaxation experiments provide information on the dynamics of the relaxation processes. Experiments were conducted at a single value of strain amplitude (1%), and the corresponding time dependent relaxation modulus, G(t), was measured. Figure 11 shows the stress relaxation curves of binary and ternary blends of ACM, FKM and 50/50 (w/w) ACM/FKM with PTMPTA at 200[degrees]C. The equilibrium stress relaxation is not reached for all three blends, showing the viscoelastic nature of the blends. However, there is a plateau in the initial relaxation period, and then the stress decays exponentially with time at the beginning and follows logarithmic decay at the latter time. The slope of the decay curve of FT30 is steeper than for the other blends, indicating a more damping nature of this blend. This is due to the non-interacting nature of FKM with polyacrylate. ACM undergoes more interaction with polyacrylate and forms single-phase structure as evidenced from TEM photomicrography (15). In a dynamic test, stress relaxes at early times because of beta-relaxation believed to involve rotation of the side-groups about the chain backbone as well as other fast rearrangements on the atomic scale. At later times and in highly entangled melts or In a crosslinked structure, mass transport through diffusion of whole chains induces additional relaxation.
* The storage modulus of AT30 is higher than that of FT30 at all frequencies because of higher grafting of multifunctional acrylate with ACM. When the level of FKM is increased in the blend, the G' value is decreased. The same trend is also observed in the case of complex viscosities of the blends.
* All the blends show the pseudoplastic behavior and obey the power law. By comparing the values of power law coefficient with composition of the blends, It is seen that the blends with higher proportion of ACM have very low values indicating more amount of grafting and crosslinking between the components of blends.
* The effect of temperature on the rheological behavior shows anomalous results. With increase of temperature, there is a slight increase of the melt viscosity of all blends due to continuation of residual polymerization, grafting and crosslinking at higher temperature.
* A higher proportion of polyacrylates increases the shear modulus and dynamic viscosity. The polyacrylates with higher functionality also show the same trend.
* While considering the storage modulus of the blends, the AFT30 shows higher G' compared to the AFH30 and the tan [delta] value of AFT30 is lower than that of AFH30, indicating more gelling in AFT30.
* Stress relaxation of the blends shows higher elastic characteristics with linear relaxation process with time.
[FIGURE 1 OMITTED]
[FIGURE 2 OMITTED]
[FIGURE 3a OMITTED]
[FIGURE 3b OMITTED]
[FIGURE 4 OMITTED]
[FIGURE 5 OMITTED]
[FIGURE 6 OMITTED]
[FIGURE 7 OMITTED]
[FIGURE 8 OMITTED]
[FIGURE 9 OMITTED]
[FIGURE 10 OMITTED]
[FIGURE 11 OMITTED]
Table 1 Composition of ACM, FKM and 50/50 (w/w) ACM/FKM Blend Containing Polyfunctional Acrylates Composition AFT AFT (w/w) AT30 (70/30/30) AFT30 (30/70/30) FT30 AFT10 AFT40 ACM 100 70 50 30 -- 50 50 FKM -- 30 50 70 100 50 50 HDDA -- -- -- -- -- -- -- TMPTA 30 30 30 30 30 10 40 DPHA -- -- -- -- -- -- -- Composition (w/w) AFH30 AFD30 ACM 50 50 FKM 50 50 HDDA 30 -- TMPTA -- -- DPHA -- 30 Table 2 Power Law Coefficients of Various Blends Measured at Different Temperatures. Temperature ([degrees]C) n k x [10.sup.4] [right arrow] 160 180 200 160 Sample [down arrow] AT30 0.05 0.04 0.04 97 AFT70/30 0.07 0.05 0.05 82 AFT30 0.08 0.07 0.06 70 AFT30/70 0.09 0.09 0.08 35 FT30 0.21 0.16 0.10 11 AFT10 0.14 0.12 0.11 26 AFT40 0.05 0.06 0.05 167 AFH30 0.16 0.13 0.10 71 AFD30 0.08 0.06 0.05 76 Temperature ([degrees]C) k x [10.sup.4] [right arrow] 180 200 Sample [down arrow] AT30 133 143 AFT70/30 96 108 AFT30 73 79 AFT30/70 50 63 FT30 19 34 AFT10 26 26 AFT40 178 186 AFH30 85 100 AFD30 89 103
(1.) L. A. Utracki, Polymer Alloys and Blends, Hanser Publishers (1989).
(2.) C. Koning. M. V. Duin, C. Pagnoulle, and R. Jerome, Prog. Polym. Sci., 23, 707 (1998).
(3.) O. Olabisi. L. M. Robeson, and M. T. Shaw, Polymer-Polymer Miscibility, Academic Press, New York (1979).
(4.) Daniel L. Hertz Jr., in Handbook of Elastomers--Revised Ed Won. A. K. Bhowmick and H. L. Stephens, eds., Marcel Dekker, New York (2001).
(5.) D. R. Paul and C. B. Bucknall. Polymer Blends, Vol. 2, John Wiley & Sons, New York (1999).
(6.) M. Xanthos, Polym. Eng. Sci., 28, 1392 (1988).
(7.) M. Saleen and W. E. Baker, J. Appl. Polym. Sci., 39, 655 (1990).
(8.) J. G. Bonner and P. S. Hope. in Polymer Blends and Alloys. p. 46, M. J. Folkes and P. S. Hope. eds., Blackie Glassgow, U.K. (1993).
(9.) M. A. Kader and A. K. Bhowmick, Rubber Chem. Technol., 73, 889 (2000).
(10.) M. A. Kader. A. K. Bhattacharya, and A. K. Bhowmick, Polym. & Polym. Composites, 9, 263 (2001).
(11.) P. A. Dworjanyn, Field and J. L. Garnett, "Effect of radiation on high technology polymer," ACS symposium series, Vol. 38 p.1 (1989).
(12.) I. Banik and A. K. Bhowmick, Radiat. Phys. and Chem., 54. 135 (1999).
(13.) M. A. Kader and A. K. Bhowmick, Rubber Chem. Technol, 74, 662 (2001).
(14.) H. Murata, N. Taguchi, T. Hemado, and J. F. McCabe, Biomaterials, 21, 1421 (2000).
(15.) M. A. Kader and A. K. Bhowmick, T. Inoue and T. Chiba, J. Mat. Sci., 37, 1503 (2202).
(16.) P. Piersoni, D. Ercoll, G. Goizueta, and N. Capati, J. Reinforced Plast. Compos., 20, 182 (2001).
(17.) Y. J. Jim, G. S. Shin, I. T. Lee, and B. K. Kim, J Appl. Polym. Sci., 47, 295 (1993).
(18.) M. Memmati, H. Zazokdast, and H. S. Panahi, J. Appl. Polym. Sci., 82, 1129 (2001).
(19.) A. K. Gupta, A. K. Jain, B. K. Ratnam, and S. N. Maiti, J. Appl. Polym. Sci., 39, 515 (1990).
(20.) W. Zheng. Y. Leng, and X. Zhu, Plast. Rubber Composite Process Appl., 25, 490 (1996).
(21.) T. Kojima, Y. Kikuchi, and T. Inoue, Polym. Eng. Sci., 32, 1863 (1992).
(22.) S. Y. Hobbs. M. E. Dekkers, and V. H. Watkins, Polymer, 29, 1598 (1988).
(23.) E. H. Kerner, Proc. Phys. Soc., 69, 808 (1956).
(24.) L. A. Utracki and M. R Kamal, Polym. Eng. Sci., 22, 96 (1982).
(25.) J. D. Ferry, Viscoelostic Properties of Polymers, Wiley, New York (1970).
(26.) R. G. Larson, K. I. Winey, S. S. Patel, H. Watanabe, and R. Bruinsma, Rheol. Acta, 32, 245-253 (1993).
(27.) D. C. Morse and F. S. Bates, J. Rheology, 43, 167 (1999).
(28.) D. J. Walsh and J. S. Higgins, Polymer, 23, 336 (1982).
(29.) M. C. O. Chang, D. A. Thomas, and L. H. Sperling, J. Appl. Polym, Sci., 34, 407 (1987).
(30.) M. Lin and S. Lee, Polymer, 38, 53 (1997).
M. ABDUL KADER #
ANIL K. BHOWMICK *
* Corresponding author.
# Present Address: Crescent Engineering College, Vandalur. Chennai - 600 048. India.
|Printer friendly Cite/link Email Feedback|
|Author:||Kader, M. Abdul; Bhowmick, Anil K.|
|Publication:||Polymer Engineering and Science|
|Date:||Apr 1, 2003|
|Previous Article:||Tyrosine imprinted polymer beads with different functional monomers via seed swelling and suspension polymerization.|
|Next Article:||Effects of molecular weight and molecular structure of low profile additives on the properties of Bulk Molding Compound (BMC).|