Printer Friendly

Decrosslinking of crosslinked high-density polyethylene via ultrasonically aided single-screw extrusion.

INTRODUCTION

Among various polymer wastes, management of crosslinked plastics is a major environmental problem requiring a solution. Recycling of crosslinked PE is a great challenge due to the presence of a three-dimensional network. Various attempts have been made to recycle crosslinked PE by decomposition of silane-crosslinked low density polyethylene (LDPE) via supercritical alcohol [1-3], size reduction of waste crosslinked LDPE by mechanochemical milling [4, 5], and by modular intermeshing co-rotating twin-screw extrusion [6], In addition, recycling power transmission cable insulated with crosslinked PE and its separation by the thermo-chemical, thermo-mechanical, and microwave-mechanical means have been attempted [7], However, in most of these studies recycling of the crosslinked LDPE was carried out.

During the past two decades, Isayev and co-workers developed ultrasonic-assisted extrusion technology to devulcanize various rubbers [8-11]. Decrosslinking of XHDPE was also attempted, but a severe thermal degradation occurred due to significant overheating from the dissipation of ultrasonic energy leading to inferior properties of the decrosslinked XHDPE [12]. Therefore, recycling of XHDPE requires further development and investigation, especially due to the fact that better controlled ultrasonic-assisted extrusion systems are being developed [13],

The present study is aimed to carry out ultrasonic decrosslinking of XHDPE using novel SSE in which deficiency of earlier design [12] was eliminated. Specifically, extensive comparisons of the properties and structure of the virgin and processed HDPE, XHDPE, and decrosslinked XHDPE obtained from these two extruders are made. It will be shown that at some processing conditions mechanical properties of the decrosslinked XHDPE are close to those of virgin XHDPE. Also, an attempt to correlate the sol-gel behavior and viscosity of the decrosslinked XHDPE is made. An importance of the molecular structure of sol of the decrosslinked XHDPE on its mechanical performance is emphasized.

EXPERIMENTAL

Materials

Polymer used in the present study was HDPE (Paxon 7004, rotational molding grade, 35 mesh powder, Exxon Mobil, Baytown, TX). Crosslinking agent was dicumyl peroxide (DC-40 in pellet form containing 40 wt% peroxide and 60 wt% calcium carbonate as a carrier, Akrochem, Akron, OH). Antioxidant 1010 (tetrakis [methylene (3,5-di-tert-butyl-4-hydroxyhydro cinnamate)] methane, Akrochem, Akron, OH) was used as a stabilizer during the melt rheology test [14]. Xylene (mixture of isomers, ACS Reagent Grade, Sigma-Aldrich, WI) was used as a solvent for the swelling test. Hexane (ACS Reagent Grade, J.T. Baker, Pittsburgh, PA) was used to precipitate the dissolved sol from xylene. The sol was the soluble fraction of samples in the swelling test.

Crosslinking recipe for preparation of XHDPE consisted of 97.5 wt% HDPE and 2.5 wt% DC-40 providing the concentration of active peroxide of 1 wt%. First, a high speed mixer (Vintage Machinery Corp., Lansdale, PA) was used to mix 20 wt% DC-40 pellets with 80 wt% HDPE powder. Then, the prepared mixture was blended with the virgin HDPE powder using the same high speed mixer with the concentration required for crosslinking. In order to avoid the decomposition of the dicumyl peroxide induced by the dissipative heat during mixing and achieve the homogeneity, the mixing was repeated six times with the duration of mixing of 10 s.

XHDPE slabs of dimensions of 26 x 26 x 1.25 [cm.sup.3] were produced by a compression molding press (Carver, Wabash, IN) at 180[degrees]C for 30 min at a hydraulic pressure of 50 MPa. The obtained slabs of XHDPE were crushed by a grinder (WSL180/ 180, Weima, Fort Mill, SC) with sieve having holes of a diameter of 4.5 mm. The obtained particles have an average diameter of 4.0 mm and aspect ratio of 1.8. It was confirmed that the grinding process has no effect on the gel fraction of XHDPE.

Ultrasonic Decrosslinking of the XHDPE

Ultrasonic decrosslinking of the XHDPE was carried out by using SSE. The SSE was a single screw extruder of 25.4 mm (KL 100, L/D = 33, Killion Corporation, Riviera Beach, FL). The schematic design of the ultrasonic SSE and the photograph of the screw are given in Fig. la and b, respectively. The extruder was equipped with two water-cooled ultrasonic horns inserted to the barrel and connected to boosters and converters. Ultrasonic power generators had capacities of 6 kW each operating at a frequency of 20 kHz (Branson 2000bdc, Branson Ultrasonic Co., Danbury, CT). A circular die of a diameter of 2 mm and a length of 60.5 mm was attached to SSE. The ground particles of the XHDPE or powder of the virgin HDPE were starve fed to the extruder by a 21 mm twin screw feeder (K2V-T2, K-Tron Corporation, Pitman, NJ). A temperature profile of the SSE was maintained by electrical heaters attached to the barrel and die. A pressure transducer (PT 435A, Dynisco Instruments, Franklin, MA) located at 730 mm from the feed end of the screw was mounted on the barrel to record the pressure before the ultrasonic treatment zone. The pressure and ultrasonic power consumption were recorded by a data acquisition system (Model DI-715B, DataQ Instruments, Akron, OH). Voltage signals from the pressure transducer and ultrasonic power supply was related to the pressure and ultrasonic power consumption based on calibration provided by manufacturers. The screw contained the union carbide mixer (UCM) of 53.8 mm length located at 418 mm from the start of the screw and melt star mixer (MSM) 63.5 mm long located at 70 mm from the UCM. The diameter of the screw was increased to 38.1 mm from its original diameter before ultrasonic treatment zone and reduced to 33.02 mm to provide a gap of 2.54 mm for ultrasonic treatment. The diameter of the screw was reduced to 25.4 mm at the discharge end.

The ultrasonic decrosslinking of XHDPE by means of SSE was performed at flow rates of 4.5, 7.5, and 15.1 g/min and ultrasonic amplitudes of 5, 7.5, and 10 /un. The mean residence time in the ultrasonic treatment zone at various flow rates was 25.9, 15.5, and 7.7 s. Also, the decrosslinking of XHDPE without imposition of ultrasound was performed at flow rates of 4.5 and 7.5 g/min. Due to an excessive torque, the flow rate of 15.1 g/min was impossible to achieve without imposition of ultrasonic treatment. The SSE screw speed was set at 100 rpm. The temperature profile of the barrel and die was set at 200[degrees]C and the feed port was set at 180[degrees]C. Extrudates of various decrosslinked XHDPE were collected by the same method as that employed for processing of the virgin HDPE.

Preparation of Specimens

The sheet of virgin HDPE and XHDPE and decrosslinked XHDPE were prepared by compression molding in a mold of dimensions of 34 X 16 X 0.185 [cm.sup.3]; 80 g of the virgin HDPE powder was molded at 150[degrees]C for 20 min; 80 g of the mixture containing peroxide was crosslinked by the same method as used for preparation of the XHDPE slabs; 100 g of the decrosslinked XHDPE pellets was molded at 200[degrees]C for 25 min. A hydraulic pressure in compression molding of all the samples was 50 MPa.

Characterization

Swelling Test. Swelling tests of the XHDPE and decrosslinked XHDPE were performed by following ASTM D 2765, test method C. Rectangular shaped specimens (0.5 g) were sliced from the sheet of the XHDPE and decrosslinked XHDPE for swelling test. Specimens were weighed and then immersed in an 8-oz wide round neck glass jar containing 100 ml of xylene. A sealing cap was screwed on the jar. The jar was then immersed in an oil bath at 110[degrees]C for 24 hrs. The swollen specimens were removed from hot xylene and excessive solvent was removed before weighing of the swollen gel. Then, the swollen gel was dried in a vacuum oven at 100[degrees] for at least 16 hrs to completely remove the residual solvent. Then, the dried gel was weighed.

Gel fraction, [zeta], and crosslink density, vv, of the treated and untreated samples were determined as [15]:

[zeta] = [W.sub.gel]/[W.sub.polymer] (1)

[v.sub.e] = - In(1 - [v.sub.ro]) + [v.sub.ro] + [chi][[v.sub.ro].sup.2]/[v.sub.1] (1/[v.sup.3.sub.ro] - 1/2 [v.sub.ro]) (2)

where [w.sub.gel] is the weight of gel, [w.sub.polymer], is the weight of polymer, [v.sub.1] is the molecular volume of the solvent, [chi] is the Flory-Huggins interaction parameter, [v.sub.c] is the crosslink density, and [v.sub.ro] is the volume fraction of the polymer in the swollen network that is in equilibrium with the solvent. For xylene mixture of isomers at 110[degrees]C, [chi] = 0-28 and [v.sub.1] = 128.4 [cm.sup.3]/mol [16].

Morphological Investigation. A field emission scanning electron microscope (JEOL-7401, Japan Electron Optics Laboratory, Japan) was used to investigate the morphology of cryofractured surfaces of the virgin HDPE, XHDPE, and decrosslinked XHDPE obtained by fracturing of tensile bar moldings in liquid nitrogen. The cryofractured surfaces were etched to remove the amorphous phase to reveal the crystalline structure. The etching agent was a mixture of 1 vol. [H.sub.2]S[O.sub.4] : 1 vol. [H.sub.3]P[O.sub.4] and 1 wt% KMn[O.sub.4] [17]. To obtain the best conditions for SEM observations on different samples, the etching time was varied from 0.5 hrs to 3 hrs with the lowest etching time being in case of the virgin HDPE. In addition, the unetched samples of the virgin and decrosslinked XHDPE were used for morphological observations. An acceleration voltage of 1.0 kV or 2.0 kV was used. The sample was mounted on the aluminum stub. A sputter coater (K575X, Quorum Technologies Ltd, UK) was used to coat the surface of samples with silver to provide the conductivity. The obtained SEM images of the cryofractured surface of the decrosslinked XHDPE were analyzed by the ImageJ software to calculate the area average diameter of the gel particles.

Rheological Test. Dynamic test of the XHDPE and decrosslinked XHDPE were conducted by means of the Advanced Polymer Analyzer (APA 2000, Alpha Technology, OH). The 16 [cm.sup.2] square specimen was cut from the sheet for the dynamic test. Frequency sweep tests within a frequency range from 0.1 to 100 rad/s at 160[degrees]C were performed at a strain amplitude of 6.2%. Dynamic tests of the virgin HDPE and the sol from decrosslinked XHDPE were conducted by using the Advanced Rheometric Expansion System (ARES, TA Instruments, New Castle, DE). Circular specimens for dynamic tests of the virgin HDPE and the sol from the decrosslinked XHDPE were prepared by compression molding at same processing parameters as in the case of the virgin HDPE sheet in a disk cavity mold of a diameter of 25 mm and a thickness of 2 mm. Frequency sweep tests within a frequency range from 0.5 to 316 rad/s at temperatures of 150, 160, 180, 190, 200, 210[degrees]C at a strain amplitude of 6.2% were conducted for the purpose of calculating activation energy of viscous flow. It was confirmed that the used strain amplitude was within the linear viscoelasitc region.

Thermal Analysis. Differential scanning calorimetry (DSC, Model Q200, TA Instruments, New Castle, DE) was used to investigate the thermal behavior of the virgin HDPE and XHDPE and decrosslinked XHDPE. Samples were cut from tensile testing specimens by a stainless steel one-sided razor blade. Each sample of weight about 7-10 mg was sealed into the DSC hermetic pan (PS 1007, PS 1010, Instrument Specialist Inc., MI). The heating-cooling-heating cycle at a rate of 10[degrees]C/min was applied for each sample in order to remove thermal history. All samples were characterized by heating from 40 to 200[degrees]C then maintaining at 200[degrees]C for 10 min, cooling from 200 to 40[degrees]C and equilibrating at 40[degrees]C, and finally heating from 40 to 200[degrees]C. The melting point and enthalpy were calculated from the second heating. The temperature corresponding to the melting peak was taken as the melting point. The melting enthalpy was calculated by integrating the area of the melting peak. The baseline of DSC curves after the completion of melting was extended to lower temperatures to define the area of the melting peak. The melting enthalpy of 282 J/mol of the perfect linear polyethylene crystal [18] was used to evaluate the crystallinity. The reported crystallinity and melting point are the average value of at least two measurements.

Tensile Testing. Tensile testing was conducted by using Instron Tensile Tester (Model 5567, Instron, Canton, MA) at a crosshead speed of 25 mm/min. Dumbbell shaped specimens were cut from sheets by using a cutting die having a width of 5 mm in the narrow section and a gauge length of 23 mm. The distance of 23 mm between grips was used as the initial length to calculate the strain. The Young's modulus, yield stress, yield strain, strain at break, and stress at break were determined. The mean values with the standard deviations of at least five specimens were reported.

RESULTS AND DISCUSSION

Processing Characteristics

Figure 2 shows the barrel pressure as a function of the ultrasonic amplitude at various flow rates during the decrosslinking of XHDPE. An increase of pressure with the flow rate is seen which is consistent with general extrusion theory. A significant decrease of the pressure with an increase of the ultrasonic amplitude is seen for the decrosslinking of the XHDPE. It is known that the decrease of the pressure during extrusion with application of ultrasound is due to the thixotropic effects including the shear thinning [19], the decrease of entanglement density [11], the increase of free volume [19], and the permanent effect including the decrease of viscosity [20], In the present study, the decrease of the pressure is mainly caused by the permanent effect during decrosslinking of the XHDPE. The detailed explanation of this effect would be given later.

Figure 3 shows the specific ultrasonic energy (a) and specific mechanical energy (b) as a function of the ultrasonic amplitude at various flow rates in SSE during extrusion of the XHDPE. The specific ultrasonic energy and specific mechanical energy is defined as follows:

Specific ultrasonic energy

ultrasonic power consumption (kJ/s)/flow rate (g/s) (3)

Specific mechanical energy

= Extruder motor power consumption (kJ/s)/flow rate (g/s) (4)

where the ultrasonic power consumption and extruder motor power consumption during the extrusion of XHDPE were measured. As seen in Fig. 3a, an increase of the specific ultrasonic energy with the ultrasonic amplitude and a decrease of the specific ultrasonic energy with the flow rate are captured. The previous experimental studies on the devulcanization of various rubbers [21-27] indicated that the ultrasonic power consumption increases with the ultrasonic amplitude and pressure [21-23], Also, the simulation study indicates that the ultrasonic power consumption increases with the ultrasonic amplitude and pressure in the ultrasonic treatment zone [25, 27]. In this study, the dependence of specific ultrasonic energy on the ultrasonic amplitude is consistent with previous studies [21-27]. The specific ultrasonic energy decreases with an increase the flow rate. However, an increase of the flow rate also leads to an increase of pressure in the ultrasonic treatment zone which results in an increase of power consumption. Apparently, the effect of the flow rate prevails over the effect of the pressure on the specific ultrasonic energy. It should be noted that, as discussed later, the specific ultrasonic energy should not be considered as a measure of the effect of ultrasonic treatment on the properties of the decrosslinked XHDPE. Based on earlier simulation studies on ultrasonic devulcanization [25, 27], it can be suggested that the ultrasonic decrosslinking of XHDPE is the result of ultrasonically induced bubble dynamics. During the oscillation of bubbles, the polymer chains in the network become overstressed eventually causing the breakage of chemical bonds. Thus, the intensity of bubble dynamics determines the effect of ultrasonic treatment. However, the specific ultrasonic energy alone cannot be used to describe the intensity of ultrasonically induced bubble dynamics.

A decrease of the specific mechanical energy with the flow rate and ultrasonic amplitude is seen in Fig. 3b. The decrease of the specific mechanical energy with the flow rate is due to the result of a longer residence time in the extruder. The decrease of the specific mechanical energy on the ultrasonic amplitude is due to the decrease of viscosity of XHDPE as well as the thixotropic effects.

Crosslink Density and Gel Fraction

Figure 4 shows the crosslink density (a) and gel fraction (b) of the decrosslinked XHDPE as a function of the ultrasonic amplitude at various flow rates. The data point for the virgin XHDPE is also shown in this figure. As seen from Fig. 4, decrosslinking takes place even without imposition of ultrasound, as indicated by a decrease of the gel fraction and crosslink density of the XHDPE as indicated by data at zero amplitude. Also, further decrosslinking occurs with an increase of the ultrasonic amplitude and flow rate.

It should be noted that the rupture of network of the crosslinked LDPE via mechanochemical milling was observed before [4, 5], In the present study, the SSE contains a screw with the UCM and MSM. This may lead to high shear stresses and shear rates during extrusion sufficient to rupture the chemical bonds. In fact, as seen from Fig. 4b, the gel fraction is reduced even without ultrasonic treatment by about 25% in SSE at all flow rates, which is due to mechanical decrosslinking induced by extruders.

The effect of ultrasound becomes more significant with an increase of the amplitude which is consistent with earlier studies of devulcanization of rubber [21-27], Previous studies of the devulcanizaion of rubber have shown that the effect of ultrasound increases with the residence time and decreases with the flow rate. However, the present study has shown an opposite trend of the effect of the flow rate and the residence time of ultrasonic treatment on decrosslinking of XHDPE. It should be pointed out that the effect of ultrasound on the decrosslinking or devulcanization diminishes with time. The effect of ultrasonic treatment eventually levels off with time after the sample being exposed to a sufficient length of time. In other words, there is a critical treatment time above which ultrasonically induced decrosslinking shows a weak dependence on time. This phenomenon has been seen in the ultrasonic decrosslinking of the XHDPE [12] and the ultrasonic devulcanization of the SBR vulcanizate [26]. The critical time for the XHDPE is about 10 s [12]. Also, the effect of ultrasonic decrosslinking or devulcanization is increased with the pressure [12, 26]. Therefore, the effect of ultrasound increases with time only if the residence time is less than the critical threshold time. In this study, the gel fraction and crosslink density of the XHDPE is lower than the XHDPE in a previous study [12] and the critical threshold time is less than that of the XHDPE in a previous study [12], Since the shortest residence time for ultrasonic treatment in this study is 7.7 s, which is probably close or longer than the critical threshold time of the XHDPE, all of the samples have been subjected to ultrasonic treatment sufficient. Thus, the effect of the residence time on ultrasonic decrosslinking is insignificant. However, an increase of the flow rate leads to an increase of pressure which results in a stronger effect of ultrasonic decrosslinking. Hence, the gel fraction and crosslink density decrease with the flow rate.

The normalized gel fraction versus the normalized crosslink density with inclusion of effect of mechanically and ultrasonically induced decrosslinking is shown in Fig. 5a for all samples. The normalized gel fraction and the normalized crosslink density are defined with respect to the values of the virgin XHDPE. It is seen that there is an approximate universal linear relation between the normalized gel fraction and normalized crosslink density. A dotted line represents the case that only the chain scission of the main chain is constructed based on the Horikx functions [28]:

1 - [v.sub.e2]/[v.sub.e1] = 1 - [(1 - [s.sup.1/2.sub.2]).sup.2]/[(1 - [s.sup.1/2.sub.1].sup.2] (5)

where [v.sub.e1] is the crosslink density of the undegraded sample, [v.sub.e2] is the crosslink density of the degraded sample, [s.sub.1] is the sol fraction of undegraded sample, and [s.sub.2] is the sol fraction of the degraded sample. It is seen that all the decrosslinked XHDPE lies below the main chain scission curves which indicates the thermal degradation of the network take place. The thermal degradation of the network is probably due to the absence of an antioxidant.

However, in Fig. 5b, by normalizing the gel fraction and crosslink density of decrosslinked XHDPE treated by ultrasound with those of decrosslinked XHDPE without being treated by ultrasound obtained at the same flow rate, the effect of the extruder on ultrasonically induced decrosslinking is revealed. It is seen that the decrosslinked XHDPE lies above the main chain scission curves, which is also plotted based on Eq. 5. This indicates that the ultrasonically induced decrosslinking is more selective than mechanically induced decrosslinking.

Morphology

Figure 6 shows the SEM images of the XHDPE (a) and decrosslinked XHDPE obtained at a flow rate of 7.5 g/min and an amplitude of 7.5 [micro]m (b). A featureless cryofractured surface of the XHDPE is seen in Fig. 6a. However, it is seen that some sub-micron size particles are embedded in the matrix of the cryofractured surface of the decrosslinked XHDPE in Fig. 6b. The statistical analysis of gel particles indicates that their area average diameter is 340 nm. The particles are the gel of the decrosslinked XHDPE. The size reduction of the gel particle in comparison with the virgin XHDPE particles indicates a breakage of the crosslink network.

Figure 7 shows the SEM images of etched cryofractured surfaces of the virgin HDPE (a), virgin XHDPE (b), and decrosslinked XHDPEs obtained at a flow rate of 7.5 g/min and an ultrasonic amplitude of 7.5 [micro]m (c) and at a flow rate of 15.1 g/min and an ultrasonic amplitude of 10 [micro]m (d). These samples are chosen in order to examine the effect of the gel and crosslink density on the morphology of the crystalline structure. As clearly seen from Fig. 7a, there is a well-developed lamellar structure in the virgin HDPE due to its linear molecular architecture. However, as seen from Fig. 7b, the lamellar structure of the virgin XHDPE is distorted by the presence of the high amount of gel (80.5 wt%) and high crosslink density (1.49 X [10.sup.-2] kmol/[m.sup.3]). This observation is similar to that observed on the crosslinked LDPE [29]. It is reported that the presence of crosslinks restricts and distorts the chain folding and growth of the lamellar structure [29], As seen from Fig. 7c, the lamellar structure in the decrosslinked XHDPE of an intermediate gel content (49.9 wt%) and intermediate crosslink density (8.4 X [10.sup.-3] kmol/[m.sup.3]) obtained at a flow rate of 7.5 g/min and an ultrasonic amplitude of 7.5 [micro]m is very similar to that of the virgin XHDPE. Evidently, the gel fraction and crosslink density of the decrosslinked XHDPE shown in Fig. 7c is still too high to allow the growth of a well-developed lamellar structure. In fact, the decrosslinked XHDPE obtained at a flow rate of 15.1 g/min and an ultrasonic amplitude of 10 [micro]m having the lowest gel fraction (41.5 wt%) and lowest crosslink density (3.9 X [10.sup.-3] kmol/[m.sup.3]) exhibits a better developed lamellar structure, as seen in Fig. 7d, than that of the decrosslinked XHDPE shown in Fig. 7c. However, some distortion of the lamellar structure due to the presence of the gel is still evident in Fig. 7d. Accordingly, the gel fraction and crosslink density of the decrosslinked XHDPE significantly affect the morphology of the lamellar structure which in turn significantly influence its Young's modulus, as discussed later.

Dynamic Properties

Figures 8-11 show the complex viscosity, storage and loss moduli, and loss tangent of the XHDPE and decrosslinked XHDPE at various flow rates without ultrasonic treatment (a) and with ultrasonic treatment at ultrasonic amplitudes of 5 [micro]m (b), 7.5 [micro]m (c), and 10 [micro]m (d) as a function of the frequency at 160[degrees]C. In the double logarithmic scale, the complex viscosity of the XHDPE and decrosslinked XHDPE decreases about linearly with the frequency and the storage and loss moduli of the XHDPE and decrosslinked XHDPE increases about linearly with the frequency. It seems that the frequency range used is below the frequencies corresponding to the entanglement plateau region. The loss tangent of the XHDPE exhibits a weak dependency on the frequency, while that of the decrosslinked XHDPE shows a slight decreasing trend with the frequency. In fact, the dependency of complex viscosity of the XHDPE and decrosslinked XHDPE on the frequency can be well described by using a power law model:

[absolute value of [[eta].sup.*]] = K [[omega].sup.n-1] (6)

where [omega] is angular frequency, K is the flow consistency, and n is power law index in Eq. 6. Also, the dynamic moduli of crosslinked XHDPE in a wider vicinity of gel point are governed by a scaling law:

G' [varies] G" [varies] [[omega].sup.m] (7)

where m is power law index in Eq. 7. The power law behavior of the complex viscosity and dynamic moduli in terminal region is characteristic of a polymer near its gelation point and a lightly crosslinked polymer. The average molecular weight ([M.sub.c]) between crosslinks in XHDPE is determined as follows:

[M.sub.c] = [rho]/2[v.sub.e] (8)

where [rho] is the density of the material and [v.sub.e] is crosslink density. According to Eq. 8, the [M.sub.c] of XHDPE is 3.2 X [10.sup.4] g/mol. The dependency of the zero shear rate viscosity on [M.sub.w] of HDPE at 190[degrees]C is given by Raju et al. [30] as

[[eta].sub.o](Pa x s) = 3.4 x [10.sup.-15] [([M.sub.w]).sup.3.6] (9)

based on the measured value of [[eta].sub.o] and the Eq. 9, [M.sub.w] of the virgin HDPE used in the present study is 5.2 X [10.sup.4] g/mol which is close to the [M.sub.c] of XHDPE. This indicates that the XHDPE is a lightly crosslinked polymer in this study. And the decrosslinked XHDPE exhibits a higher [M.sub.c] than the XHDPE. Thus, both of the XHDPE and decrosslinked XHDPE can be considered as a lightly-crosslinked polymer. Therefore, their rheological characteristics can be well described by Eqs. 6 and 7. In fact, similar scaling behaviors are also observed in the crosslinked LLDPE [31] and crosslinked LDPE [32].

It is seen in Figs. 8, 9, and 11 that the decrosslinked XHDPE exhibits a lower complex viscosity and storage modulus but a higher loss tangent compared to those of the XHDPE with the effect becoming more evident with an increase of the ultrasonic amplitude and a decrease of flow rate. This observation is attributed to the rupture of the crosslink network. It is also seen that the slope of the storage modulus versus frequency of the decrosslinked XHDPE is higher than that of the XHDPE. Also, an increase of the loss tangent of the decrosslinked XHDPE in comparison to the XHDPE is consistent with the fact that the decrosslinking indeed takes place. The loss modulus of the decrosslinked XHDPE shows a different behavior compared to other dynamic properties. The loss modulus of the decrosslinked XHDPE is also higher than that of the XHDPE at high frequencies but lower at low frequencies. The decrosslinked XHDPE shows a stronger dependency of the loss modulus on frequency than that of the XHDPE. The difference in behavior of dynamic properties between the XHDPE and decrosslinked XHDPE proves that the decrosslinking indeed takes place during ultrasonic extrusion.

Figure 12 shows that the values of n exhibit an approximate linear dependence on the crosslink density of the decrosslinked XHDPE and the values of K exhibit an approximate power-law dependence on the crosslink density of the decrosslinked XHDPE. These correlations indicate that the crosslink density is the dominant structural factor in the structural-viscosity relationship.

It is seen that the decrosslinked XHDPE is a composite of its gel embedded in its sol parts in its SEM image. Therefore, the molecular characteristics of the sol are of great interest in this study. Since the rheology of polyethylene is very sensitive to the molecular characteristics, the interpretation of the dynamic properties of the sol at various temperatures can reveal the molecular characteristics of the sol. The sol extracted from the decrosslinked XHDPE obtained at a How rate of 7.5 g/min and amplitude of 7.5 [micro]m was studied. The reason for choosing this particular sample is because it exhibits an intermediate crosslink density, gel fraction, and dynamic properties among all the decrosslinked XHDPE samples. The complex viscosity along with their Cross model [33] fit (a) and storage modulus (b) of the virgin HDPE and sol of the decrosslinked XHDPE as a function of frequency at different temperatures are depicted in Fig. 13. The following two equations are used for this fitting:

[eta] * = [[eta].sup.*.sub.o] (T)/1 + [[[eta].sup.*.sub.o] (T) [gamma]./[tau]].sup.1 - n] (10)

[[eta].sup.*.sub.o](T) = A exp ([T.sub.b]/T) (11)

where, A, [T.sub.b], [tau], and n are fitting parameters. The function of [[eta].sup.*.sub.o] (T) is the temperature dependency of the zero-frequency viscosity by the Arrhenius equation. The activation energy of viscous flow of sols is evaluated from the value of [T.sub.b] as

E = [T.sub.b] xR (12)

where R is the gas constant. The parameters of the Cross model and activation energy of the virgin HDPE and the sol are given in Table 1. It is seen that the viscosity of the virgin HDPE is lower than the sol. The slope of the storage modulus versus frequency of the sol of the decrosslinked XHDPE is lower than that of the virgin HDPE within the frequency range studied. The latter indicates different molecular structures of this sol in comparison with that of the virgin HDPE due to the presence of more entanglements in sols of the decrosslinked XHDPE. Also, in this frequency range, the sol shows a strong non-Newtonian behavior indicating that their molecular weights are higher than that of the virgin HDPE. Interestingly, the sol exhibits higher activation energy of viscous flow than the virgin HDPE. Generally, the activation energy of viscous flow of polyethylenes increases with an increase of degree of branching. Therefore, it is dear that the sol has a higher degree of branching than the virgin HDPE. The branched structure of the sol is the result of nonselective decrosslinking during extrusion as indicated in Fig. 5a.

DSC Analysis

Figure 14 shows the DSC traces of the virgin HDPE, XHDPE, and various decrosslinked XHDPEs and a sol extracted from the decrosslinked XHDPE obtained at a flow rate of 7.5 g/ min and an ultrasonic amplitude of 7.5 [micro]m. The XHDPE, decrosslinked XHDPE, and its sol exhibit a broad melting peak with a low melting enthalpy, while the virgin HDPE exhibits a sharp peak with a high melting enthalpy. In a study of melting behavior of the crosslinked LDPE, Kao and Phillips [34] suggested that broadening of the melting peak is due to a decrease of the lamellar thickness and an increase in the internal disorder. Although the crystallization behaviors of HDPE and LDPE are different, the effect of crosslinking on the melting behavior of XHDPE is seen to be similar to that on the crosslinked LDPE.

Figure 15 shows the crystallinity (a) and melting temperature (b) of the decrosslinked XHDPE as a function of the ultrasonic amplitude at various flow rates. For comparative purposes, values of the crystallinity and melting temperature of the virgin HDPE, XHDPE, and a sol extracted from the decrosslinked XHDPE at a flow rate of 7.5 g/min and an amplitude of 7.5 [micro]m are also given in Fig. 15. It is seen in Fig. 15 that the virgin HDPE exhibits the highest crystallinity and melting temperature among all studied materials due to its linear molecular structure. At the same time, the crystallinity and melting temperature of the decrosslinked XHDPE is higher than that of the XHDPE. It is also reported that a decrease of the crosslink density of the crosslinked LDPE [34] and XHDPE [35] increases its crystallinity. However, the crystallinity and melting temperature of the decrosslinked XHDPE do not show a monotonous dependence on the processing conditions, which implies that the crosslink density and gel fraction is not the only structural factor influencing the crystallinity and melting temperature. In fact, the crystallinity also depends on the degree of branching of the sol. The melting temperature and crystallinity of the sol is lower than that of the virgin HDPE as a result of a higher degree of branching. Since the occurrence of the decrosslinking of XHDPE results in both a decrease of the crosslink density and an increase of the amount of sol having a higher degree of branching, the melting temperature and crystallinity of the decrosslinked XHDPE is affected by the competition between these factors. As seen from Fig. 15a, the crystallinities of various decrosslinked XHDPEs obtained at different processing conditions are not significantly different. A decrease of the gel fraction and crosslink density is expected to increase the crystallinity of the decrosslinked XHDPE. However, this effect is reduced due to an increase of the degree of branching of the sol in the decrosslinked XHDPE. Therefore, the melting temperature and crystallinity of the decrosslinked XHDPE is not a function of the crosslink density alone but a function of many variables, including the crosslink density, gel fraction, and molecular structure of the sol. Therefore, a clear trend of the dependency of the crystallinity of the decrosslinked XHDPE on the processing conditions is very difficult to establish. It should be noted that the study on decrosslinking of the crosslinked PE by the pan-milling method reported only an increase of the crystallinity of the decrosslinked PE [5].

Mechanical Performance. Figure 16 illustrates the Young's modulus (a), yield stress (b), yield strain (c), strain at break (d), and stress at break (e) of the decrosslinked XHDPE as a function of the ultrasonic amplitude at various flow rates. These properties for the virgin HDPE and XHDPE are also reported in Fig. 16. Similar to the crystallinity, there is no clear trend of the dependency of tensile properties on the processing conditions, except for an increase of the Young's modulus with ultrasonic amplitudes at most flow rates.

Typical strain-stress curves of the virgin HDPE, XHDPE, and decrosslinked XHDPE are depicted in Fig. 17a. Similar to the virgin HDPE and XHDPE, the decrosslinked XHDPE exhibits typical behavior showing the yielding, necking, and strain hardening behavior before breakage. The stress-strain curve of XHDPE shows a lower yield stress and strain at break but a stronger strain hardening behavior than that of the virgin HDPE. The decrosslinked XHDPE shows a lower stress and strain at break in comparison with the XHDPE. The stress-strain curves of various samples at a strain less than 20% are enlarged and depicted in Fig. 17b to observe their behavior in the yielding region. It is seen that the virgin HDPE shows a sharp peak with a high yield stress in comparison with the XHDPE and decrosslinked XHDPE. It is known that the branched PE may exhibit a phenomenon of double yielding [36]. The presence of a broad yielding region in the stress-strain curves of the XHDPE and decrosslinked XHDPE is a result of the existence of a structural rregularity such as branching and crosslinking.

Comprehensive studies [37-40] on the structure-mechanical property relationship of PE indicate that the tensile properties including the Young's modulus, yield stress and strain and ultimate properties have different dependencies on the molecular characteristics and structural factors including the molecular weight, crystallinity, thickness of noncrystalline region, degree of branching and their length and crosslink density. Therefore, the tensile properties of various samples will be discussed in more detail.

Young's Modulus. Since the Young's modulus is measured in the reversible elastic deformation region of the sample, both the crystalline and noncrystalline regions determine its value. The volume fractions, inherent moduli of these regions, and morphology of the sample determine its Young's modulus. It should be noted that the modulus of the PE crystal is in the order of GPa and that of the noncrystalline region is in the order of 10 MPa [37]. The large difference in the inherent moduli of the crystalline and noncrystalline regions will lead to a strong dependency of the Young's modulus on the crystallinity. In fact, it is observed that an increase of the crystallinity of both the linear PE and branched PE leads to an approximately linear increase of the Young's modulus [38, 39]. However, a discontinuity is seen by extrapolating two fitted lines in the range of the crystallinity where an overlap of the linear PE and branched PE occurs [38, 39]. At this point, the linear PE has much higher Young's modulus than the branched PE. Such a discontinuity indicates the importance of the effect of the noncrystalline region on the Young's modulus, which is explained by taking the morphology such as the thickness of the noncrystalline region into account [38, 39]. Also, an increase of the crosslink density led to a decrease of the Young's modulus, as observed in the crosslinked HDPE [41].

It is seen in Fig. 16a that the Young's modulus of the virgin HDPE is much higher than that of the decrosslinked XHDPE which in turn is higher than that of the XHDPE. This is probably due to the difference in the crystallinity. However, the increase of the Young's modulus of the decrosslinked XHDPE with the amplitude is not a result of the change of the crystallinity. It is more likely caused by morphological changes, shown in Fig. 7c and d, due to the differences in the gel fraction and crosslink density of the decrosslinked XHDPE. As examples, let us consider the decrosslinked XHDPEs obtained at a flow rate of 7.5 g/min and an ultrasonic amplitude of 7.5 [micro]m and at a flow rate of 15.1 g/min and an ultrasonic amplitude of 10 [micro]m. As shown in Fig. 15a, the crystallinity of these two decrosslinked XHDPEs are almost same. However, as seen from Fig. 16a, the Young's modulus of the decrosslinked XHDPE obtained at a flow rate of 15.1 g/min and an ultrasonic amplitude of 10 [micro]m is significantly higher than that of the decrosslinked XHDPE obtained at a flow rate of 7.5 g/min and an ultrasonic amplitude of 7.5 [micro]m. This can only be attributed to the morphological differences in the lamellar structure of these two decrosslinked XHDPEs (see Fig. 7c and d). Since the decrosslinked XHDPE is a semicrystalline polymer, it is not surprising that the Young's modulus of the decrosslinked XHDPE depends on the morphology of its lamellar structure. Thus, the dependency of the Young's modulus on the ultrasonic amplitude is a result of different morphologies.

Yield Stress and Strain. Extensive studies [37-40] indicated that the yield stress of the melt crystallized PE is mainly dependent on its crystallinity. Linear relations between the crystallinity and the yield stress were observed for various linear and branched PEs [37-40], However, at a same level of the crystallinity, the linear PE has a higher yield stress than the branched PE, indicating that there are other factors affecting the yield stress of PE. To our best knowledge, there is no comprehensive study on the yield strain-structure relationship for PE. But Brooks et al. [42] concluded that the yield strain generally increased with a decrease of the crystallinity for the both linear and branched PEs.

It is seen in Fig. 16b that the yield stress of the virgin HDPE is higher than that of the decrosslinked XHDPE which in turn is slightly higher than that of the XHDPE. The yield stress of the decrosslinked XHDPE shows slight dependency on the processing conditions. These findings are opposite to the dependency of the yield strain as seen in Fig. 16c. Since the crystallinity of the decrosslinked XHDPE ranges from 58% to 63% and mainly dependent on the crystallinity, it is expected to see that the yield stress and yield strain of the decrosslinked XHDPE also falls into a small range.

Strain and Stress at Break. The strain hardening behavior of the PEs of various molecular structures govern their strain and stress at break. Presently, the physical nature of the strain hardening behavior is not fully understood. It is believed that there are at least two governing mechanisms. One is the strain-induced oriented crystallization and another is the elastic response of the interlamellar region under applied stresses [38, 39, 43], Entanglements are considered as a major contributor to the second mechanism. An increase of the number of entanglements per chain impedes the chain slippage during deformation resulting in an increase of the modulus, i.e., enhanced strain hardening behavior. However, this effect also decreases the deformability of the polymer resulting in a decrease of the strain at break. Studies on the linear PE [38] in the ductile region of deformation indicated that the strain at break was independent of the supermolecular structure, crystallite thickness, crystallinity, and polydispersity. The strain at break decreases with an increase of Mw. In addition, the strain at break of the branched PE showed a similar dependency on [M.sub.w] and independence on the supermolecular structure, crystallite thickness, and crystallinity. The crosslinking of HDPE results in a decrease of the strain at break in comparison with that of the virgin HDPE due to the fact that crosslinks hinder the chain slippage during deformation. Hence, an increase of the crosslink density leads to a decrease of the strain at break of the XHDPE [41].

The stress at break of the linear and branched PEs showed a dependency on Mw with a maximum value observed at an intermediate molecular weight [39, 40]. An increase of the degree of branching and their length led to a shift of the maximum value to a lower molecular weight and a lower stress on the dependency of the stress at break of the branched PE on [M.sub.w]. The XHDPE with a low value of the crosslink density results in an increase of the stress at break in comparison with that of the virgin HDPE due to an enhanced strain hardening. An increase of the crosslink density of the XHDPE led to a decrease of the stress at break of the XHDPE when the crosslink density is relatively high [41].

It is seen in Fig. 16d that the strain at break of the virgin HDPE is higher than that of the XHDPE which in turn is higher than that of the decrosslinked XHDPE. The virgin HDPE has the highest strain at break due to its linear chain structure favoring the chain slippage during deformation. Also, in the ductile region of deformation of the virgin HDPE having low molecular weight exhibits a lower entanglement density leading to a better deformability in comparison with other samples. It should be pointed out that in considering the strain at break of the XHDPE and decrosslinked XHDPE, these materials should be considered as a composite of gel and sol, rather than a homopolymer. A lower strain at break of the decrosslinked XHDPE in comparison with that of the XHDPE is due to the fact that the decrosslinked XHDPE contains a ruptured network which cannot sustain stress uniformly at a large strain and a sol of higher branching degree which impedes the deformations. However, an absence of a clear trend of the dependency of the strain at break of the decrosslinked XHDPE on processing conditions is seen in Fig. 16d. This is due to the complexity of the effect of different structural factors, including crosslink density, branched structure, molecular weight, and morphology of gel-sol structure, on the strain at break of the decrosslinked XHDPE. This is a reason why a clear dependency of the strain at break of the decrosslinked XHDPE on processing conditions cannot be established.

Similar to the case of the strain at break, values of the stress at break of the virgin HDPE, XHDPE. and decrosslinked XHDPE are shown in Fig. 16e. The stress at break of the XHDPE is higher than that of the decrosslinked XHDPE which in turn is higher than that of the virgin HDPE. Due to the similar features of the stress-strain curves of the XHDPE and decrosslinked XHDPE, the stress at break of these materials is determined by their strain at break. Therefore, the relative ordering of the stress at break of the XHDPE and decrosslinked XHDPE remains the same as in the case of the strain at break. The stress at break of the virgin HDPE is lower than those of the XHDPE and decrosslinked XHDPE which is mainly due to a weak strain hardening behavior of the virgin HDPE. Due to the fact that the strain hardening behavior is affected by many structural factors, including crosslink density, branched structure, molecular weight, and morphology of gel-sol structure, it is very difficult to establish a processing-structure-property relationship with a clear dependency of the stress at break on processing conditions, as seen in Fig. 16e.

CONCLUSIONS

The ultrasonically aided extrusion of the virgin HDPE and decrosslinking of XHDPE are carried out using SSE under various flow rates and ultrasonic amplitudes. Successful decrosslinking of XHDPE is achieved. Processing characteristics during extrusion are obtained. Various characterization techniques are used to determine the gel fraction, crosslink density, morphology, dynamic and mechanical properties, and thermal behavior in order to find the effect of processing conditions.

It is found that the processing characteristics such as the pressure and ultrasonic power consumption are correlated with processing conditions and explained based on the extrusion theory and ultrasonic processing model. Ultrasonic treatment can improve the productivity of the decrosslinking of the XHDPE. The gel fraction and crosslink density of the decrosslinked XHDPE decreases with the ultrasonic amplitude and flow rate. It is found that the high shear stresses generated by the screw rotation lead to mechanically induced decrosslinking of XHDPE. The decrosslinking is further enhanced by imposition of ultrasonic waves. For ultrasonically and mechanically induced decrosslinking, an approximate universal linear relationship between the normalized gel fraction and the normalized crosslink density is found, regardless of the processing conditions. However, by using Horikx function to indentify the type of bond breakage, the difference between ultrasonically induced decrosslinking and mechanically induced decrosslinking by SSE is revealed when the effect of mechanically induced decrosslinking is excluded. The ultrasonic decrosslinking is more selective on scission of the crosslink than mechanically induced decrosslinking.

The SEM study reveals that the decrosslinked XHDPE is a composite of sub-micron size gel particles embedded in the matrix consisted of its sol. This composite structure leads to a complex structure-solid state properties relationship.

The dynamic properties of the virgin HDPE, XHDPE, and decrosslinked XHDPE are measured. The relationship between dynamic properties of the decrosslinked XHDPE and their molecular characteristics are established. It is found that the crosslink density of the decrosslinked XHDPE is the dominant factor affecting its rheology. Dynamic properties of a representative sol sample of the decrosslinked XHDPE are measured and the activation energy of their viscous flow is calculated in order to reveal the molecular structure of the sols. It is found that the sol extracted from the decrosslinked XHDPE has a higher viscosity and more branched structure than the virgin HDPE.

The melting temperature and crystallinity of the decrosslinked XHDPE is determined from its DSC trace. However, the absence of a clear trend of the effect of processing conditions on the crystallinity and melting temperature indicates the significance of the effect of the crosslink density and the branched structure of sol on the thermal behavior of the decrosslinked XHDPE resulting in a complex structure-property relationship.

The mechanical properties of the decrosslinked XHDPE are close to the virgin XHDPE but exhibit no clear trend on the processing conditions. Thus, the processing-structure-property relationship is utilized to explain the absence of this trend based on the published literature on PE. In particular, the latter is explained by the fact that the Young's modulus, yield stress, and yield strain is not solely dependent on the crystallinity but also on other structural characteristics including the gel fraction, crosslink density of the gel, and molecular and branched structure in the sol. The strain and stress at break are also affected by many structural characteristics including the crosslink density, gel fraction, and branched structure in the sol. The complex processing-structure-property relationship is due to the fact that the decrosslinked XHDPE is a composite consisting of the sol and gel exhibiting a complex molecular structure.

ACKNOWLEDGMENTS

The authors wish to thank the ExxonMobil Chemical Company for providing HDPE and the Akrochem Corporation for providing peroxide and antioxidant.

REFERENCES

[1.] I. Okajima, A. Katsuzaki, T. Goto, T. Yamazaki, and T. Sako, J. Chem. Eng. Japan, 43, 231 (2010).

[2.] T. Goto, T. Yamazaki, T. Sugeta, I. Okajima, and T. Sako, J. Appl. Polym. Sci., 109, 144 (2008).

[3.] H. Lee, J.H. leong, H.K. Cho, C.M. Koo, S.M. Hong, H. Kim, and Y.W. Lee, Polym. Degrad. Stability, 93, 2084 (2008).

[4.] X. Zhang, C. Lu, and M. Liang. J. Appl. Polym. Sci., 122, 2110 (2011).

[5.] H. Wu, M. Liang, and C. Lu, J. Appl. Polym. Sci., 122, 257 (2011).

[6.] S.S. Oh, D.S. Bang, and J.K. Lee, Elastomer, 39, 131 (2004).

[7.] C.C. White, J. Wagenblast, and M.T. Shaw, Polym. Eng. Sci., 40, 863 (2000).

[8.] A.I. Isayev and S. Ghose, "Ultrasonic Devulcanization of Used Tires and Waste Rubbers," in Rubber Recycling, S.K. De, A.I. Isayev, K. Khait, Eds., CRC Press, Boca Raton, Chapter 9, pp. 311-384 (2005).

[9.] A.I. Isayev, "Recycling of Rubber," in Science and Technology of Rubber, 3rd ed., J.E. Mark. B. Erman, F.R. Eirich, Eds., Academic Press, New York, Chapter 15, pp. 663-701 (2005).

[10.] A.I. Isayev and J.S. Oh, "Tire Materials: Recovery and Reuse," in The Pneumatic Tire, A.N. Gent, J.D. Walter, Eds., NHTSA D.S. Department of Transportation, Washington, DC, Chapter 18, pp. 670-691 (2005).

[11.] X. Sun and A.I. Isayev, Rubber Chem. Technol., 81. 38 (2008).

[12.] A.I. Isayev and J. Jenkins, SPE ANTEC Technical Paper, 53, 109 (2008).

[13.] A.I. Isayev, PCT Application PCT/US2012/033751, filed April 16, 2012, WO/2012/142562 published 10/18/2012.

[14.] H. Mavridis and R.N. Shroff, Polym. Eng. Sci., 32, 1778 (1992).

[15.] P.J. Flory,./. Chem. Phys., 18, 108 (1950).

[16.] A.F.M. Barton, CRC Handbook of Polymer-Liquid Interaction Parameters and Solubility Parameters, CRC Press, Boca Raton (1990).

[17.] R. Olley and D. Basset, Polymer, 23, 1707 (1982).

[18.] L. Zhu, F. Chiu, Q. Fu, R.P. Quirk, and S.Z.D. Cheng, Polymer Handbook, 4th ed., John Wiley & Sons, pp V-17 (1999).

[19.] A.I. Isayev, C.M. Wong, and X. Zeng, Adv. Polym. Technol., 10. 31 (1990).

[20.] Y. Chen and H. Li, J. Polym. Sci. Part B: Polym. Phys., 45, 1226 (2007).

[21.] A.I. Isayev, S.P. Yushanov, S.H. Kim, and V.Y. Levin, Rheol. Acta, 35, 616 (1996).

[22.] A. Tukachinsky, D. Schworm, and A.I. Isayev, Rubber Chem. Technol., 69, 92 (1996).

[23.] A.I. Isayev, J. Chen: S.P. Yushanov, "Ultrasonic Devulcanization of Waste Rubbers: Experimentation and Modeling," In Simulation of Materials Processing: Theory, Methods and Applications, S.F. Shen, P.B. Dawson, Eds., Balkema, Rotterdam, pp. 77-85 (1995).

[24.] A.I. Isayev, S.P. Yushanov, and J. Chen, J. Appl. Polym. Sci., 59, 803 (1996).

[25.] V.V. Yashin and A.I. Isayev, Rubber Chem. Technol., 72, 741 (1999).

[26.] C.K. Hong, A.I. Isayev, and J. Chen, Rubber Citent. Technol., 75, 1 (2002).

[27.] V.V. Yashin and A.I. Isayev, Rubber Chem. Technol., 73, 325 (2000) .

[28.] M.M. Horikx, J. Polym. Sci., 19, 445 (1956).

[29.] R. Gohil and P. Phillips, Polymer, 27, 1687 (1986).

[30.] V.R. Raju, G.G. Smith, G. Marin, J.R. Knox, and W.W. Graessley, J. Polym. Sci. Polym. Phys. Ed., 17, 1183 (1979).

[31.] E.M. Valles, J.M. Carella, H.H. Winter, and M. Baumgaertel, Rheol. Acta., 29, 535 (1990).

[32.] X. Zhang, H. Yang, Y. Song, and Q. Zheng, Polymer, 53, 3035 (2012).

[33.] M.M. Cross, Rheol. Acta, 18, 609 (1979).

[34.] Y.H. Kao and P.J. Phillips, Polymer, 27, 1669 (1986).

[35.] P.J. Hendra, A.J. Peacock, and H.A. Willis, Polymer, 28. 706 (1987).

[36.] N.W.J. Brooks, R.A. Duckett, and I.M. Ward, Polymer, 33, 1872 (1992).

[37.] R. Popli and L. Mandelkern, J. Polym. Sci. B: Polym. Phys., 25. 441 (1987).

[38.] M.A. Kennedy, A.J. Peacock and L. Mandelkern, Macromole, 27, 5297 (1994).

[39.] M.A. Kennedy, A.J. Peacock, M.D. Failla, J.C. Lucas, and L. Mandelkern, Macromole, 28, 1407 (1995).

[40.] A.J. Peacock and L. Mandelkern, J. Polym. Sci. B: Polym. Phys., 28, 1917 (1990).

[41.] H.A. Khonakdar, J. Morshedian, U. Wagenknecht, and S.H. Jafari, Polymer, 44. 4301 (2003).

[42.] N.W.J. Brooks, R.A. Duckett, and I.M. Ward, Polymer, 40, 7367 (1999).

[43.] J.T. Graham, R.G. Alamo, and L. Mandelkern, Polym. Sci. B: Polym. Phys., 35, 213 (1997).

Avraam I. Isayev, Keyuan Huang

Department of Polymer Engineering, The University of Akron, Akron, Ohio 44325-0301

Correspondence to: Avraam I. Isayev; e-mail: aisayev@uakron.edu

Contract grant sponsor: NSF; contract grant number: CMMI-1131342.

DOI 10.1002/pen.23827

Published online in Wiley Online Library (wileyonlinelibrary.com).

TABLE 1. Cross model parameters and the activation
energy of viscous flow of virgin HDPE and sols
of the decrosslinked XHDPE.

            A,      [T.sub.b],      [tau],
Material   (Pa*s)       (K)          (Pa)

Virgin      1.08       2654         3.6 x
  HDPE                            [10.sup.8]

SSE sol     0.17       5258         8.0 X
                                  [10.sup.8]
                    [[eta].sup.
                     *.sub.0]
                      at 190         E,
Material     n      [degrees]C     (kJ/mol)

Virgin     0. 65      3.30 x         22.1
  HDPE              [10.sup.2]

SSE sol    0. 37      1.46 x         43.7
                    [10.sup.4]
COPYRIGHT 2014 Society of Plastics Engineers, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2014 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Isayev, Avraam I.; Huang, Keyuan
Publication:Polymer Engineering and Science
Article Type:Report
Geographic Code:1U3OH
Date:Dec 1, 2014
Words:9189
Previous Article:On the encapsulation of natural pesticide using polyvinyl alcohol/alginate-montmorillonite nanocomposite for controlled release application.
Next Article:Preparation and electrochemistry of nanostructured PPy/graphite nanosheets/rare earth ions composites for supercapacitor.
Topics:

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters