Novel ultrasonic process for in-situ copolymer formation and compatibilization of immiscible polymers.
Blending and reactive extrusion of mixtures of polymers are useful approaches for the manufacture of new materials with specially tailored performance characteristics that are often not exhibited by the corresponding single component polymers (1-4). Enhanced properties of polymeric materials are achieved by developing multi-component systems in the form of polymer blends composed of two or more homopolymers. However, a majority of the polymer pairs are immiscible with each other and exhibit either very low or no interfacial adhesion and phase-separate upon blending. In addition, immiscible polymer pairs in blends have unstable phase morphologies in the melt state. The mechanical properties of polymer blends are strongly affected by the strength of the interfaces between the different phases, as well as by phase morphology and physical and chemical interactions between the components. Typically, uncompatibilized blends of immiscible polymers are weak and brittle. The present state-of-the-art to achieve compati bilization is to incorporate a third component into a polymer blend leading to a chemical interaction at the interface of the corresponding blend components. Typically, a block copolymer containing blocks chemically identical to each of the components is added. Alternatively, a chemical interaction between the immiscible polymers can be achieved by means of reactive blending which relies on either the in-situ formation of block or graft copolymers or interfacial chemical interactions of polymers using specifically selected reactive chemicals and compatibilizers. Although these approaches for enhancement of the interfaces are common at the present time, these methods are restricted since to make different useful polymer blends they require different, specifically tailored chemicals with functional groups or block or graft copolymers. Therefore, the ability to make virtually any two or more polymers to interact chemically with each other is highly desirable for manufacturing a wide variety of stable polymer ble nds and for polymer recycling.
The present study reports, for the first time, the phenomenon indicating that by means of ultrasonic-assisted extrusion, immiscible polymer blends in the melt under high pressures can be very quickly (on the order of seconds) induced to undergo an in-situ copolymer formation at the polymer interfaces and their vicinities. Surprisingly, copolymer formation can be obtained for pairs of polymers that otherwise would not be expected to react. As a result of the ultrasonic in-situ copolymer formation in immiscible blends, the interfacial adhesion between the polymers is enhanced, causing a desirable stabilization, of phase morphology in the melt state and a significant improvement of the mechanical properties in the solid state.
Polyolefins (high density polyethylene, HDPE, Marlex HMN4550-3, Phillips; or polypropylene. PP, Pro-fax 6523, Himont) were first mixed with an equal amount of an uncured rubber (natural rubber, NR, SMR CV60, Akrochem; solution styrene-butadiene rubber, SBR, Duradene 706, Firestone; or ethylene propylene diene rubber, EPDM, Keltan 2506, DSM) by means of a twin screw extruder (JSW Labotex 30) using a feed rate of 60 g/min. The screw speed was set at 150 rpm and the zone temperatures in the extruder from feed to die section were set at 140/140/145/150/150/155/160/160[degrees]C for HDPE/rubber blends and at 165/165/175/180/180/185/185/190/190[degrees]C for PP/rubber blends. The extrudates were cooled, pelletized and dried. The prepared pellets were extruded and ultrasonically treated using a 38.1 mm single screw extruder with an ultrasonic barrel attachment developed in our laboratory as shown schematically in Fig. 1a (5). Two water-cooled ultrasonic horns were inserted into the barrel and both vibrated longitudi nally at a frequency of 20 kHz with amplitudes of 6 and 10 [mu]m. A 3 kW power supply with a piezoelectric converter and booster was used. The extrusion rate was 37.8 g/min at a screw speed of 20 rpm. The extrusion temperature was set at 150[degrees]C and 190[degrees]C for HDPE/rubber and PP/rubber blends, respectively. The gap between the horns and the screw surface was set at 2 mm. The mean residence time of the melt in the ultrasonic treatment zone was 11.2 sec.
A 50/50 NR/SBR blend was prepared on a two-roll mill (Dependable Rubber Machinery Co.) at 50[degrees]C by mastication for 5 min. The blend was ultrasonically treated in a 38.1 mm single screw extruder with a coaxial ultrasonic die attachment provided with one water-cooled horn as shown in Fig. 1b (6). The horn oscillated longitudinally with a frequency of 20 kHz and amplitudes of 5. 7.5 and 10 [mu]m. The extrusion rate was 37.8 g/min and the extrusion temperature was 120[degrees]C. The gap between the die and horn was set at 2.54 mm. The mean residence time in the treatment zone was 13.6 sec. The untreated and ultrasonically treated NR/SBR blends were compounded on a two-roll mill at 50[degrees]C with 2 phr of sulfur, 5 phr of zinc oxide, 1 phr of stearic acid and 1.1 phr of N-cyclo-hexyl benzothiazole sulfenamide (CBS) for 6 min.
The untreated and ultrasonically treated samples of HDPE/rubber, PP/rubber and NR/SBR blends were compression molded at a pressure of 13.8 MPa and at temperatures 160[degrees]C, 180[degrees]C and 160[degrees]C, respectively, into 127 x 127 X 2 [mm.sup.3] slabs for tensile testing and into 76.2 mm diameter disks having a thickness of 3 mm for impact testing. Compression molding times for plastic/rubber blends and curing times for NR/SBR blends were 5 and 9 min, respectively. Also, the untreated and treated samples of HDPE/rubber and PP/rubber samples were subjected to annealing in the compression molding press for 10 min at temperatures of 160[degrees]C and 190[degrees]C, respectively.
The rheological behavior of the blends was investigated using a Monsanto Processability Tester (MPT). Three capillary dies (diameter of 1.506 mm and length/diameter (L/D) ratios of 5, 10 and 20) were used. End corrections were applied for the calculation of the shear viscosity.
Morphological observations of the annealed and unannealed samples were made using a scanning electron microscope (SEM, Hitachi S-2150). The samples with and without annealing were fractured in liquid nitrogen and the rubber phase was extracted in benzene at 50[degrees]C for 24 hours. After etching, the samples were dried in a vacuum oven at 60[degrees]C for 12 hours and were coated with silver using sputter coater.
The AFM images were obtained in air using a scanning probe microscope Nanoscope IIIa (Digital Instruments) in the tapping mode. Measurements were performed at ambient conditions using rectangular type Si probes with spring constant of 50 [Nm.sup.-1] and resonance frequencies in the 284-362 kHz range. The tip radius was 10 nm. The AFM topographic (height) and elastic (phase) images were simultaneously obtained under normal tapping condition on the microtomed surface.
The Soxhlet extraction experiments using benzene as a solvent were carried out on the untreated and ultrasonically treated PP/NR blends for 72 hours to determine the non-extractable fraction. The molecular weights of the untreated and ultrasonically treated NR, SBR and NR/SBR blends were measured by GPC columns using THF as a solvent with polystyrene calibration standards.
An Instron tensile tester (Model 5567) was used to measure the stress-strain curves of the molded samples at a crosshead speed of 500 mm/min and at room temperature. Impact testing was performed at room temperature using a dart impact tester (Dynatup 8250) with velocity of 4.17 m/s and a dart weight of 5.08 kg. The diameter of the hemispherical probe was 12.7 mm and the samples were held in place on an annular ring with an internal diameter of 38.1 mm. All samples were totally penetrated.
RESULTS AND DISCUSSION
Tensile testing and the dart impact testing of the untreated and ultrasonically treated plastic/rubber blends were performed. As an example, Fig. 2 shows the stress-strain (a) and the force-displacement (b) curves of the PP/NR blends. It is seen, that the stress-strain and impact behaviors of the ultrasonically treated blends were enhanced in comparison with those of untreated blends as evident from a higher tensile strength and elongation to break on the stress-strain curves and by a higher maximum of the force and a higher value of the displacement during impact break. From the stress-strain curves, the tensile strength, elongation at break, secant modulus at 3% strain (except at 100% for NR/SBR blends) and toughness (the area under the stress-strain curve) were determined. From the force-displacement curves the total energy required to fully penetrate samples upon impact break was calculated. All of these values for plastic/rubber and NR/SBR blends are listed in Table 1. Surprisingly, as evident from this table, ultrasonic treatment significantly increased all of these properties for plastic/rubber blends. Similar enhancement of the properties occurs in the case of NR/SBR blends except that the modulus is slightly reduced. The highest increases in toughness (470%) and impact energy (212%) were observed, respectively, in the case of a treated PP/EPDM and PP/NR blends in comparison with those of the untreated samples. It is hypothesized that the ultrasonic treatment of the blends enhances intermolecular interaction possibly through in-situ copolymer formation at the interfaces between dissimilar polymers and their vicinities without the use of any added chemicals. Specifically, the cleavage of the main chain polymer (carbon-carbon) bonds during ultrasonic treatment, as observed earlier in de-vulcanization of rubber vulcanizates (6) and in a polymer melt extrusion (7). may lead to formation of the polymer radicals (8, 9) of each component in the blends. These macro-radicals of both the polymers can undergo a fast recombination leading to in-situ formation of a copolymer during the ultrasonic treatment. It is proposed that the presence of this in-situ formed copolymer is the main reason for the improved mechanical properties of the ultrasonically treated blends.
Copolymer formation takes place at high values of the pressure and ultrasonic energy density. In particular, Table 2 shows the die entrance pressure and ultrasonic energy density during ultrasonic assisted extrusion of various blends at different ultrasonic amplitudes. The ultrasonic energy density is calculated based on the measured power consumption by dividing it by the cross-sectional area of the horn. Ultrasonic power consumption is expended to dissipation losses and breakage of the main chain bonds leading to copolymer formation. One cannot estimate experimentally what fraction of the power is consumed by the bond breakage alone. In addition, the power expended on heat dissipation in the melt and the power transmitted by the traveling waves through the melt cannot be separated. The only measurable losses are the initial power consumption of the acoustic system when the horn works without loading. In the results reported in Table 2, these losses are subtracted from the total power consumption. An increas e in the ultrasonic energy density, seen in Table 2, with the amplitude is due to an increase in the amplitude of strain imposed on the melt. The latter leads to an increase in the stresses generated in the melt and in the forces required for imposing oscillations on the ultrasonic horn by the ultrasonic piezoelectric transducer. It is seen from Fig. 2 that the mechanical performance of the treated samples is more superior at the amplitude of 6[micro]m rather at 10 [micro]m. Evidently, under the action of ultrasound there is a competition between a degradation of the macromolecular chains and copolymerization at the interface. Copolymerization at the interface would cause an improvement of mechanical properties. while degradation would lead to their deterioration. The amount of copolymer formed may possibly be affected by the intensity of ultrasound. Evidently, at the higher amplitude the amount of copolymer is lower than that at the low amplitude. These are possible reasons for the inferior mechanical perfor mance of the samples obtained at the higher amplitude as shown in Fig. 2. As also seen from Table 2, the die pressure decreases substantially during imposition of ultrasound and decreases further as the amplitude of ultrasound increases. There are two possible reasons for the die pressure decrease with amplitude. First, the breakup of polymer chains leads to a decrease of the melt viscosity due to reduction of the molecular weight. To show the presence of this effect, rheological measurements were carried out on the untreated and ultrasonically treated blends after extrusion. As an example, Fig. 3 shows the flow curves of untreated and ultrasonically treated HDPE/NR, HDPE/ EPDM and HDPE/SBR blends obtained at different ultrasonic amplitudes. It is seen that the viscosity of treated samples is lower than that of the untreated samples. Also, the viscosity of ultrasonically treated blends decreases with an increase of ultrasonic amplitude. This decrease of the viscosity may be the main reason for the die pressur e reduction with increase of ultrasonic amplitude in ultrasonically assisted extrusion.
Evidences of Copolymer Formation
There are some strong supporting evidences indicating the possibility of the in-situ copolymer formation. These include the data obtained by the solvent extraction method and the GPC analysis. In particular, extraction experiments were carried out on samples of the PP/NR blends, untreated and ultrasonically treated at amplitudes of 6 and 10 [micro]m. Each mixture was subjected to Soxhlet extraction using benzene as a solvent that is known to selectively dissolve NR but not PP. In case of the untreated sample, 49.8% of the mixture was not extracted. Therefore, the NR content in the untreated mixture remains unchanged (50%) after processing without imposition of ultrasound. However, in the case of the PP/NR mixtures ultrasonically treated at amplitudes of 6 and 10 [micro]m, the amount of the unextracted polymer was higher, respectively, 56.1% and 54.1%. This strongly supports the idea that certain fractions of the PP and NR components in the 50/50 PP/NR mixture undergo chemical interaction causing the in-situ formation of the PP/NR copolymer. Evidently, benzene could not extract the PP and the in-situ formed PP/NR copolymer fraction. Namely, this unextracted PP/NR copolymer created in-situ during ultrasonic treatment led to compatibilization and improved mechanical properties of the ultrasonically treated blends.
Interestingly, the SEM photomicrographs of unannealed PP/NR and PP/EPDM blends, depicted in FIg. 4, show similar sizes of the dispersed domains without an indication of significant differences in the morphologies of the untreated and ultrasonically treated samples. The NR and EPDM components in the blends are present as dispersed phases and the PP component is present as the continuous phase. As seen in Fig. 4, after annealing for 10 min, sizes of the dispersed rubber domains grow significantly in the case of the untreated blends. In contrast, after annealing of the ultrasonically treated blends, the sizes of the dispersed rubber domains increase only slightly in the PP/NR blend and are not affected at all in the PP/EPDM blend. These observations are clear indications that the morphologies of the ultrasonically treated PP/NR and PP/EPDM samples are, respectively, partially and fully stabilized due to ultrasonic in-situ copolymer formation.
The 10 [micro]m phase images of untreated and ultrasonically treated PP/NR blends obtained by the AFM are shown in Fig. 5. It is seen that in the untreated blend a sharp step ranging between 45 and 130 nm is present between the PP and NR phases. However, a smooth step ranging between 6 to 14 nm is observed in the treated blend. This step is almost the order of magnitude lower than that in the untreated blend. Furthermore, the AFM three-dimensional images show some interfacial roughening at the interface between PP and NR in ultrasonically treated blends. Also, the AFM study reveals the presence of a transition interface nanolayer between the plastic and rubber phases in the treated blend. These effects would cause signiflcantl enhancement of adhesion at the interface of immiscible blends.
Further supporting evidence pointing towards the possibility of in-situ copolymer formation could be drawn from Fig. 6 and Table 3 showing the molecular weights and molecular weight distributions of the untreated and ultrasonically treated NR, SBR and NR/ SBR blends. Clearly, after ultrasonic treatment, the number average molecular weights of NR SER components and NR/SBR blends are decreased. Because of degradation, a low molecular weight tail is generated leading to a substantial broadening of the molecular weight distribution. In contrast to this finding, the weight average molecular weight of the ultrasonically treated SBR is slightly increased and that of the NR/SBR blend is increased significantly. Remarkably, a high molecular weight tail is generated in the ultrasonically treated NR/SBR sample which was absent in the untreated sample, as evident from Fig. 6c. The appearance of the high molecular weight tail is also strong evidence that the NR and SER components in the blends experience a chemical transf ormation during the ultrasonic treatment of NR/SBR blends, leading to a possibility of in-situ formation of the NR/SBR copolymer.
Plastic/rubber and rubber/rubber blends are prepared by ultrasonic treatment during continuous extrusion in order to investigate the in-situ compatibilization of the blends without any chemicals. After ultrasonic treatment the viscosity of the plastic/rubber blends was decreased due to the breakup of main chains. However, the tensile strength, elongation at break, Young's modulus and toughness and impact properties of each blend were significantly improved by ultrasonic treatment as compared to the untreated blend. After annealing, the domain size in ultrasonically treated blends is much smaller than that of untreated blends. It is believed that ultrasonic treatment of the blends enhances intermolecular interaction, improves adhesion and possibly makes chemical bonds between dissimilar polymers without use of any chemicals. Also, the results of extraction experiment and the AFM study supported the belief that copolymers are created during ultrasonic treatment of the blends.
Some chemical interactions between NR and SBR were occurred during ultrasonic treatment leading to an increase of the weight and z-average molecular weight and broadening of the molecular weight distribution of NR/SBR blend. The increase in molecular weight is a clear indication of the creation of copolymer during ultrasonic treatment of the NR/SBR blend. The tensile strength and elongation at break of this ultrasonically treated blend were significantly improved. It is believed new copolymers are created during very short time of ultrasonic treatment under high pressures and temperatures. The created copolymer is believed to lead to in-situ compatibilization during extrusion and to be a major reason for enhancing tensile strength and elongation at break of the NR/SBR blend.
It should be noted that in-situ copolymers synthesized by the ultrasonically assisted extrusion process are obtained for pairs of polymers which otherwise could not be simply prepared. Therefore, the ultrasonic assisted extrusion of polymer blends allows one to make new copolymers that could not be readily obtained by the existing techniques for polymer synthesis.
The process of ultrasonic in-situ copolymer formation is very fast, on the order of seconds. It allows one to achieve desirable chemical and physical properties of polymer blends. Because of these tangible attributes, the novel ultrasonic process for in-situ copolymer formation and compatibilization of immiscible polymers in blends is quite unique and very attractive for further studies and industrial applications. In particular. further studies are needed to find the effect of concentration of various polymer pairs on in-situ copolymer formation. Also, additional structural investigations are required to determine the molecular sequence and chemical structure of these in-situ created copolymers. Finally, it should be noted that the observed phenomenon opens the new field of research of ultrasonic-assisted in-situ copolymer formation at polymer interfaces.
[FIGURE 2 OMITTED]
[FIGURE 3 OMITTED]
[FIGURE 5 OMITTED]
[FIGURE 6 OMITTED]
Table 1 The Mechanical Properties of Untreated and Ultrasonically Treated Blends. Tensile Stress Elongation Young's Blend Amplitude MPa at Break, % Modulus MPa PP/NR Untreated 8.39 38.9 191.0 6 [micro]m 11.27 111.7 250.0 10 [micro]m 10.60 126.8 215.0 PP/EPDM Untreated 9.87 29.1 229.0 6 [micro]m 10.51 87.6 261.0 10 [micro]m 10.48 121.2 278.0 HDPE/NR Untreated 4.92 189.0 75.1 6 [micro]m 5.71 300.6 126.5 10 [micro]m 5.78 299.1 127.0 HDPE/EPDM Untreated 4.74 88.2 103.0 6 [micro]m 4.78 116.0 109.0 10 [micro]m 4.84 147.8 125.0 HDPE/SBR Untreated 5.33 15.4 59.0 6 [micro]m 5.46 28.7 101.0 10 [micro]m 5.58 32.5 129.0 NR/SBR Untreated 2.89 378.4 0.89 * 5 [micro]m 4.41 522.1 0.70 * 7.5 [micro]m 7.82 601.7 0.67 * Toughness Impact Energy Blend MPa Joule PP/NR 2.99 2.75 12.19 5.89 13.19 3.52 PP/EPDM 2.68 8.40 9.00 9.46 12.53 8.89 HDPE/NR 8.17 6.09 16.00 8.12 16.35 7.45 HDPE/EPDM 4.24 8.50 5.76 9.63 7.62 9.31 HDPE/SBR 0.59 7.53 1.32 8.76 1.61 9.43 NR/SBR 5.20 -- 8.37 -- 12.73 -- * Modulus at 100% strain, MPa. Table 2 The Ultrasonic Energy Density and the Die Entrance Pressure During Ultrasonic Assisted Extrusion. Blend Amplitude, [mu]m Energy density, Pressure, MPa W/[cm.sup.2] PP/NR 6 32.7 2.14 10 49.9 1.79 PP/EPDM 6 39.6 1.76 10 46.5 1.45 HDPE/NR 6 34.4 1.11 10 39.6 0.90 HDPE/EPDM 6 55.1 2.07 10 65.4 1.42 HDPE/SBR 6 44.8 1.83 108 58.5 1.35 NR/SBR 0 5.66 5 21.9 1.72 7.5 28.5 1.17 10 35.1 0.97 Table 3 The Molecular Weight Characteristics of Untreated and Ultrasonically Treated NR, SBR and and NR/SBR Blend. Material Untreated 5 [mu]m 7.5 [mu]m 10 [mu]m NR Mn 180,400 35,000 69,600 95,600 Mw 1,116,000 689,000 801,100 683,400 Mz 2,616,000 1,981,000 2,532,000 1,911,000 Pl 6.19 19.69 11.85 7.15 SBR Mn 106,800 11,600 14,600 14,300 Mw 387,800 407,900 398,600 438,700 Mz 1,132,000 2,541,000 3,086,000 2,681,000 Pl 3.63 35.16 27.30 30.68 NR/SBR Mn 110,300 53,900 52,200 52,300 (50/50 wt%) Mw 527,000 645,100 608,700 699,400 Mz 1,475,000 2,593,000 2,658,000 5,399,000 Pl 4.78 11.97 11.66 13.37
This work is supported in part by a grant DMI-0084740 from the National Science Foundation and the Hayes Investment Fund, State of Ohio. The authors also wish to express their appreciation to S. Ghose. J. S. Oh, S. E. Shim and J. Yun for their assistance in experiments and Dr. M. Rogunova of the PDI, Inc., for obtaining the AFM images.
(1. ) L. A. Utracki, ed., Two-Phase Polymer Systems, Hanser Publishers. Munich (1991).
(2. ) M. Xanthos, ed.. Reactive Extrusion Principles and Practice. Hanser Publishers. Munich (1992).
(3.) M. J. Folkes and P. S. Hope, eds., Polymer Blends and Alloys, Blackie Academic and Professional, New York (1993).
(4.) D. R. Paul and C. B. Bucknall. eds.. Polymer Blends. John Wiley & Sons, New York (2000).
(5.) J. Yun, J. S. Oh, and A. I. Isayev. Rubber Chem. Technol., 74, No. 2 (2001).
(6.) A. I. Isayev. J. Chen. and A. Tukachinsky. Rubber Chem. Technol., 68, 267 (1995).
(7.) A. I. Isayev. C. M. Wong. and X. Zeng, Adv. Polym. Technol., 10, 31 (1990).
(8.) N. K. Baramboim, Mechanochemistry of Polymers, Maclaren, London (1964).
(9.) A. Casale and R. S. Porter, Polymer Stress Reactions, Academic Press, New York (1977).
A. I. Isayev *
* To whom correspondence should be addressed. E-mail: firstname.lastname@example.org.
|Printer friendly Cite/link Email Feedback|
|Author:||Isayev, A.I.; Hong, Chang Kook|
|Publication:||Polymer Engineering and Science|
|Date:||Jan 1, 2003|
|Previous Article:||A study of the stick-slip phenomenon in single-screw extrusion of linear polyethylene.|
|Next Article:||Application of ultrasound in the determination of fundamental extrusion performance: residence time distribution measurement.|