Printer Friendly
The Free Library
22,710,190 articles and books

Visual observation and numerical studies of N2 vs. CO2 foaming behavior core-back foam injection molding.

INTRODUCTION

Visual observation and numerical studies of [N.sub.2] vs. [CO.sub.2] foaming behavior core-back foam

A reduction of fossil-based materials as well as energy use is desirable from the viewpoint of the sustainability life cycle. There are still many polymers produced and used in industry, however, and not all of these can be replaced by non-fossil-based polymers. Research has therefore been directed toward reducing material consumption and greenhouse gas emission during production processes. The foaming technique can contribute significantly to weight reduction of materials in many fields involving polymers. In the case of vehicular components, for example, the weight reduction from using plastic products improves gas mileage and reduces the consumption of fossil fuel.

Foam injection molding processes are widely used to reduce the weight of plastic products and reduce material use. The use of nitrogen ([N.sub.2]) or carbon dioxide ([CO.sub.2]) as a physical blowing agent in foam molding processes has been investigated to reduce greenhouse gases. Multiple studies investigating foam injection molding have previously been reported. Villamizar and Han (1) and Han and Yoo (2) observed cavity flow in a foam injection process with a chemical blowing agent and established isothermal kinetic models, though these did not incorporate bubble nucleation rate or heat balance. Arefmanesh and S. G (3) proposed a non-isothermal model that integrated bulk flow of the polymer melt into a cell-growth model. However, their model also did not incorporate bubble nucleation rate. Osorio and Tumg (4) developed mathematical models for simulating non-isothermal bubble growth behavior in injection molding of microcellular plastics and compared the predictions with experimental data where [N.sub.2] was used as a blowing agent. Teramoto et al. (5), Barze-gari and Rodrigue (6), and Lee et al. (7) investigated microcellular foam injection processes to optimize process design and operating conditions so as to control the cell structure of the foam. In our previous paper (8), we conducted visual observation and numerical simulation of core-back foam injection molding with [CO.sub.2] and showed the effectiveness of a classical bubble nucleation and growth model of batch physical foaming (the so-called free foaming model), to analyze the experimental results of [CO.sub.2] injection foaming behavior. Core-back foam injection molding is slightly different from traditional structural foam injection molding; once the polymer, which contains a foaming gas, fills up the mold, a portion of the mold is moved to quickly increase the mold cavity volume and rapidly reduce cavity pressure. The sudden pressure drop enhances bubble nucleation and achieves a fine cell structure within the polymer foam. Thus, core-back foam injection molding is effective in achieving a high expansion ratio with uniform fine cell structure.

[FIGURE 1 OMITTED]

In this study, we extended our previous study to investigate foaming behavior in core-back injection molding with [N.sub.2] and to compare it to the use of [CO.sub.2] as a blowing agent.

Few studies have been reported on the performance of [N.sub.2] vs. [CO.sub.2] as a physical blowing agent in the foaming process. Lee and Park (9) and Kim el al. (10-12) performed visual observations of batch as well as extrusion foaming of thermoplastic polyolefin and thermoplastic vulcanizates (TPVs) with [N.sub.2] and reported that, in their foam extrusion process, the largest achievable expansion ratio with [CO.sub.2] was higher than that with [N.sub.2]. This difference arose because [N.sub.2] could not be sufficiently dissolved in the thermoplastic polymer to increase the expansion ratio. They also reported that the cell size achieved with [N.sub.2] was smaller than that with [CO.sub.2] even with relatively high [N.sub.2] content. Di Maio et al. (13) also compared [N.sub.2] with [CO.sub.2] foaming of polycaprolaclones on both batch and extrusion foaming processes. They reported that batch foaming with [N.sub.2] had to be performed at a higher temperature and a higher pressure because of the lower solubility and negligible plasticizing effect of [N.sub.2]. They also mentioned that the lower foam density (i.e., higher expansion ratio) was obtained with [CO.sub.2], but that bubble density was higher with [N.sub.2] in both batch and extrusion foaming. However, a fair comparison of [N.sub.2] with [CO.sub.2] in terms of foaming performance has not been conducted for core-back foam injection molding as well as batch foaming.

In this paper, we applied an observation technique introduced in a previous paper to investigate foaming behavior in the mold cavity of a core-back foam injection molding machine. The foaming performance of [N.sub.2] and [CO.sub.2] as a physical foaming agent was one of the major concerns in the investigation. The accuracy of a classical bubble nucleation and growth model was also investigated. A bubble nucleation and growth model of a batch physical foaming process was employed to analyze the experimental data and to confirm the differential factors in the foaming performances of [N.sub.2] and [CO.sub.2]

EXPERIMENTAL OBSERVATION

The procedure of visual observation and the design of foam injection molding experiments were similar to those of our previous paper (8). An 85-ton foam injection molding machine (J85AD-Mucell, Japan Steel Works) with a gas injection device (Trexel SCF system SII TRJ-IO-A) was used for the experiments. Visual observation was performed through a glass window attached to the mold cavity, which measured 100 mm in width, 180 mm in length, and 1.5 mm in thickness, as shown in Fig. 1. A portion of the cavity was designed to translate with a controlled velocity to expand the cavity volume. Sensors for a data fogging system (Mold Marshaling System EPD-001, Futaba Corp.) were mounted on the opposite side of the prism-shaped glass window to monitor the pressure and temperature in the cavity electronically. A CCD camera (HAS-220, Direct) and zoom lens (VH-Z00R, Keyence) were used to take snapshots of the polymer flow at a frame rate of 200 frames per second. All optical and sensor recordings were synchronized with a trigger signal sent from the injection molding machine.

The polymer used in this study was a polypropylene impact copolymer (ICP, Japan Polypropylene Corp.), with an MFR of 30 g/10 min. Figure 2 shows the viscosity curve of ICP. The shear viscosity was measured with a capillary rheometer (Capilograph IB, Toyo Seiki Seisa-kusho) and the complex viscosity was measured by a rotational rheometer (ARES, TA Instruments). Either [N.sub.2] or [CO.sub.2] was used as a physical foaming agent. The physical foaming agent was provided in the supercritical state to the barrel of the injection molding machine. The feed flow rate and the gas injection time were set to 0.07 kg/h and 0.5 s for [N.sub.2], and 0.3 kg/h and 3.0 s for [CO.sub.2], respectively. The setpoints were determined from the view point of observation and the bubbles could be identified clearly by our visual apparatus in all core-back conditions. The barrel temperature of the injection molding machine was set at 230[degrees]C, and the mold temperature was 40[degrees]C The gas-dissolved ICP was injected by moving the screw at the rate of 100 mm/s until a given stroke was achieved. The shot size ranged from 20.7 to 20.8 s.

[FIGURE 2 OMITTED]

The process conditions of core-back foam injection molding are summarized in Tables 1 and 2. The conditions of foam injection molding with [CO.sub.2] were the same as those specified in our previous paper (8).

NUMERICAL MODELS

Foaming behavior in the mold cavity of the core-back injection molding was similar to that in the batch foaming process (8); the flow front of the gas-dissolved polymer was slightly foamed. However, the bubbles disappeared once the cavity was fully packed with the injected polymer and the cavity pressure reached its maximum. Bubble nucleation was then induced again by the rapid pressure drop during core-back operation. Because foaming occurred after the (low of injected polymer was stopped, bubble nucleation and growth models for a batch foaming process could be applied to as in our previous paper (8), (14), (15)]. The details of models and their assumptions were described in our previous paper (8); a brief summary of the models is given here:
TABLE 1. Core-back injection molding conditions for all cases.

Condition                                Value

Initial cavity size (mm)            100 X 180 X 1.5
Final cavity size (mm)              100 X 180 X 2.5
Injection velocity (mm/s)                       100
Injection temperature ([degrees]C)              230
Mold temperature ([degrees]C)                    40

TABLE 2. Core-back injection molding conditions in individual
experiments.

Experiment   Core-back               Gas injection    Gas flow
no.         rate (mm/s)   Gas type     time (s)     rate (kg/h)

Case 1             10.0  [N.sub.2]             0.5         0.07
Case 2              2.0  [N.sub.2]             0.5         0.07
Case 3              1.0  [N.sub.2]             0.5         0.07
Case 4              2.0  [CO.sub.2]            3.0          0.3


The bubble nucleation rate at time t, J(t), is given by

J(t) = [f.sub.0]exp(-16[pi][[gamma].sup.3] F/3[k.sub.B] T [([P.sub.D](t)-[P.sub.C]).sup.2])) [N.sub.A] [bar.c](t) (1)

where [f.sub.0] and F are correction factors, [gamma] is the surface tension, [N.sub.A] is Avogadro's number, [k.sub.B] is Boltzmann's constant, T is the absolute temperature, [P.sub.D] is the pressure inside the bubble, [P.sub.C] is the pressure of the polymer matrix, and [bar.c](t) is average gas concentration.

[P.sub.D] satisfies the relationship between gas concentration c and solubility parameter [k.sub.H]:

[P.sub.D] = c/[k.sub.H](2)

The bubble growth model is described by momentum balance and mass balance equations.

The momentum balance equation is

dR/dt = R/4[eta]([P.sub.D]-[P.sub.C]-2[gamma]/R) (3)

The mass balance of a bubble is given by

d/dt((4[pi]/3) ([R.sup.3][P.sub.D]/[R.sub.g]T)) = 4[pi] [R.sup.2] D [alpha]c/[alpha]r | r=R (4) The mass transfer into a bubble from its surrounding polymer is given by

[alpha]c/[alpha]t = D[(1/[r.sup.2]) ([alpha]/[alpha]/r) ([r.sup.2] [alpha]c/[alpha]r)], (5)

where D is the gas diffusion constant.

For numerical simulations, we assumed that the pressure of the polymer matrix, [P.sub.c] and its pressure gradient can be approximated by a linear function in experiments:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (6)

RESULTS AND DISCUSSION

Experimental Results

The concentrations of [CO.sub.2] and [N.sub.2] dissolved in injected polymer were measured by weighing; any weight difference can be attributed to the concentration of the foaming agent ([N.sub.2] or [CO.sub.2]) dissolved in the injected polymer. This method assumes that the gas dissolved in the molten polymer was completely diffused into the air during the ten days. Soon after the gas-incorporated polymer was injected into the cavity, the molded product was cooled down without core-backing or foaming, and the product was ejected from the mold and weighed. (The weight of the product was measured again after 10 days from the injection molding; no weight reduction was observed in any case.) The results of the weighing are shown in Table 3 with the standard deviation of measurements, and these results are compared with the setpoint values obtained by multiplying the setpoint value of the gas flow rate by that of the gas injection time. Average values were calculated from measurements of live samples. According to Table 3, the standard deviations were quite small and the initial amounts of foaming agents well controlled. However, in the case of [N.sub.2] injection, the measured value was about 10 times larger than the set-point value. This difference indicates the difficulty in controlling of feed gas flow rate at a low level and the need to measure gas concentration in polymers by weighing when investigating the effect of gas concentration on cell structure.
TABLE 3. Gas concentration: average and standard deviation of
five samples.

Gas type       Gas       Average    Standard    Set   Error  Exp.
            injection      (g)      deviation  point   (%)    no.
            time (s)      (wt%)       (wi%)     (g)

[N.sub.2]         0.5        0.108     0.0050  0.010   91.0  Cases
                       (0.523 wt%)                           1,2,3

[CO.sub.2]        3.0        0.294     0.0044  0.250   15.0   Case
                        (1.42 wt%)                               4


The pressure and temperature profiles were measured during core-back operations. Profiles with different core-back rates are illustrated in Fig. 3. In these experiments, gas-dissolved polymer was injected until the given stroke was reached starting t = 0, where t = 0 indicates the time at which the injection start signal was triggered. Then, the pressure inside the cavity was decreased during the core-back operation and the depressurization rate was changed by the core-back rate. The core-back rate, however, did not affect significantly the temperature profile as can be seen in Fig. 3. A series of snapshots of the foaming behavior with N2 was taken during core-back operation; one of the examples is shown in Fig. 4. When the gas-dissolved polymer was injected into the cavity, bubble nucleation occurred at the flow front due to the pressure gradient between polymer and air in the cavity. However, once the polymer filled up the cavity, the cavity pressure increased, and the bubbles disappeared or became too small to be detected by our visualization device, as shown in Fig. 4a and b. Then, as shown in Fig. 4c-f, bubbles appeared again as the cavity pressure decreased during core-back operation.

[FIGURE 3 OMITTED]

[FIGURE 4 OMITTED]

For comparison, a series of snapshots of the foaming behavior with [CO.sub.2] (8) is shown in Fig. 5. The observed foaming behavior was quite similar in all cases. The major differences between [CO.sub.2] and [N.sub.2] injection foaming were the size and number density of the bubbles; the bubble size (i.e., cell size) was much smaller, and the number of bubbles was much larger in [N.sub.2] injection foaming than those in [CO.sub.2] injection foaming. The number density of bubbles was calculated by the same method described in our previous paper (8) and plotted in Fig. 6 against time elapsed since the injection signal was triggered. The symbols represent the average value of three experiments, and the error bars depict the minimum and the maximum of these measured values. The solid symbols (Cases 1-3) represent the [N.sub.2] injection foaming and the open symbols (Case 4) represent [CO.sub.2] injection foaming. Note that the final bubble density for Case 1 could not be measured because the bubbles increased beyond the counting ability of our observation setup. Comparison of Case 2 with 4 clearly shows that the number density of bubbles blown with [N.sub.2] was much larger than that with [CO.sub.2] even at the same core-back rate. The average diameter of the bubble observed at each time point was measured from the snapshot pictures and plotted in Fig. 7, where the symbols and error bars are given with the same definition in Fig. 6. The data clearly show that the diameter of bubbles foamed with [N.sub.2] was much smaller than that with [CO.sub.2].

[FIGURE 5 OMITTED]

The foaming behaviors of [N.sub.2] and [CO.sub.2] observed in core-back injection molding show some differences from those in batch foaming. Figures 8 and 9 show the result of visual observation of the batch foaming experiments (16). Figures 8 and 9 show the number density and growth rate of bubbles (rate of change in cross-sectional area of bubbles) obtained in batch foaming, respectively. Experimental details for visual observation of batch foaming are given elsewhere (14-17). The experiments were conducted with homo-polypropylene by using either [N.sub.2] or [CO.sub.2] as a physical blowing agent. The disk-shaped samples (500 [micro]m in thickness and 5 mm in radius) were saturated in the view cell by either [CO.sub.2] or [N.sub.2] at 11 MPa and a temperature of 200[degrees]C. Then, the pressure was released at a given rate in the range of 0.63-0.70 MPa/s for both [CO.sub.2] and [N.sub.2] foaming, as shown in Fig. 8. The growth rate of bubbles (i.e., the rate of change in bubble cross-sectional area) was defined for each bubble at the time when the bubble was first detected. The rate calculation scheme is detailed elsewhere (14-17), As can be seen in Figs. 8 and 9, a larger number density of bubbles and higher bubble growth rate were obtained with [CO.sub.2] batch foaming. These results were different from those obtained from core-back foam injection molding, where [N.sub.2] produced a higher number density and smaller bubble size. The difference between batch foaming and the foam injection molding was caused by the initial concentration of the blowing agent dissolved in polymer and their gas saturation pressures. Pressure drop rate and gas saturation pressure could not be changed independently in the batch foaming processes, while they could be tuned independently in the foam injection molding process. Thus, in the batch foaming experiments, the foaming experiment started from the equilibrium state saturated with 11 MPa [CO.sub.2] or [N.sub.2]. The degrees of super-saturation, which corresponds to [P.sub.D]- [P.sub.C] in Eq. 1, were the same in both [CO.sub.2] and [N.sub.2] batch foaming, but the initial concentration of [CO.sub.2] dissolved in polymer was higher than that of [N.sub.2] because of the higher solubility or [CO.sub.2] to PP (18). The higher dissolved gas concentration of [CO.sub.2] at the same degree of super-saturation with [N.sub.2] could enhance the bubble nucleation, increase the driving force of mass transfer from polymer to bubbles, and make the bubble size as well as the number density of bubbles larger in [CO.sub.2] batch foaming compared to [N.sub.2].

In core-back foam injection molding processes, however, the initial dissolved gas concentration did not reach an equilibrium state in the barrel of the injection machine, and super-saturation occurred during depressurization of core-back operation. If the concentrations of [N.sub.2] and [CO.sub.2] dissolved in injected polymer were the same, the saturation pressure at which [N.sub.2] would reach the equilibrium state with the dissolved gas concentration would be higher than that of [CO.sub.2] and the degree of super-saturation of [N.sub.2] during core back foaming would be larger than that of [CO.sub.2], as illustrated In Fig. 10. This larger degree of super-saturation of [N.sub.2] increases the number of nucleated bubbles and decreases bubble size in [N.sub.2] foam injection foaming. Additionally, the level of super-saturation is closely related to the solubility of the gas.

[FIGURE 6 OMITTED]

[FIGURE 7 OMITTED]

[FIGURE 8 OMITTED]

[FIGURE 9 OMITTED]

[FIGURE 10 OMITTED]

Numerical Results

To confirm our understanding of [N.sub.2] and [CO.sub.2] foaming behaviors in core-back foam injection molding, numerical simulation was performed by employing the batch foaming models (8), (14); the foaming behavior in the mold cavity during foam injection molding was also simulated. The physical properties used in the numerical study are summarized in Table 4 and are taken from the literature (14-19). The Newtonian viscosity in Eq. 3 was set at 500 Pa s according to the viscosity curve in Fig. 2. The plasticization-induced variations in the melt viscosity have little effect on bubble growth (20). The process parameters used for the numerical simulations were determined from the experimental results, and the results are listed in Table 5. The temperature was set to a constant value, 200[degrees]C, for all numerical calculations, as shown in Fig. 3. From the time at which the core-back operation started, the cavity pressure decreased in accordance with the core-back rate. The depressurizalion profile was approximated by a line with a constant rate. These profiles are represented by the solid and the broken straight lines in Fig. 11.
TABLE 4. Physical properties of foaming simulation.

Physical properly          [N.sub.2]   [CO.sub.2]

Solubility parameter,     0.21 X       1.15 X
[k.sub.H]                 [10.sup.-4]  [10.sup.-4]
(mol gas/[m.sup.3]/Pa)

Diffusion coefficient, D  7.23 X       8.07 X
([m.sup.2]/s)             [10.sup.-9]  [10.sup.-9]

Surface tension, [gamma]  0.0162       0.0123
(J/[m.sup.2])

Viscosity, [eta] (Pa s)   500          500

Correction factor,        1.00 X       3.27 X
[f.sub.0]                 [10.sup.11]  [10.sup.13]

Correction factor, F      0.0155       0.0155

TABLE 5. Process parameters for foaming simulation.

                                             Time for
                                      Gas    onset of
            Temperature           content  depression
                      T         [c.sub.0]
Exp. no.  ([degrees] C)  (mol/ [m.sup/3])         (s)

Case 1              200             166.9        0.83

Case 2              200             166.9        0.90

Case 3              200             166.9        0.96

Case 4              200             287.3        0.86

             Initial
            pressure  Depression
           [P.sub.c]        rate
                         [DELTA]
                 (0)   [P.sub.c]
Exp. no.        (Pa)      (Pa/s)

Case 1         8.5 X      15.8 X
          [10.sup.6]  [10.sup.7]

Case 2         8.5 X       9.6 X
          [10.sup.6]  [10.sup.7]

Case 3         8.5 X       6.5 X
          [10.sup.6]  [10.sup.7]

Case 4         8.5 X       6.5 X
          [10.sup.6]  [10.sup.7]


The numerical calculation and experimental results are shown in Fig. 12. The solid and broken lines represent the numerical results, and each symbol represents the average value of at least three experiments. The final bubble density of the numerical simulations shows fairly good agreement with experiments, except for the result of Case 1. As was mentioned in a previous section, the final bubble density of Case 1 could not be measured by our experimental technique. The largest discrepancy between experiments and numerical simulations was observed at the onset time of bubble nucleation in Case 3. The numerical calculation shows a slightly earlier onset time than that of experiments. This difference might arise from the linear approximation of the pressure gradient for numerical simplicity and from the truncated maximum pressure.

Figure 13 shows the bubble diameter-time curves for different gas concentrations. The solid and the broken lines represent numerical calculations, while the symbols show experimental data. The numerical calculations show good agreement with experimental data. The numerical simulation also shows that [N.sub.2] can provide a higher number density and smaller bubble size than [CO.sub.2], which agrees with the experimental data.

[FIGURE 11 OMITTED]

[FIGURE 12 OMITTED]

[FIGURE 13 OMITTED]

One of the major concerns in this study was the effect of [N.sub.2] and [CO.sub.2] on bubble nucleation and growth behavior. During experiments, we could see some differences between [N.sub.2] and [CO.sub.2] foaming as well as a difference in foaming behavior in core-back foam injection molding compared to batch foaming. We therefore speculated that the difference between [N.sub.2] and [CO.sub.2] solubility in PP made bubble size and the number of bubbles different in [N.sub.2] foaming compared to [CO.sub.2] and that the time difference at which the system reaches the equilibrium pressure made the foaming behavior in the foam injection molding process different from that in the batch foaming. To confirm the speculation, a parameter simulation study was conducted by using the batch foaming models, Eq(s). 1-6, and by assuming that some of the physical parameters of [N.sub.2]-PP system, such as surface tension [gamma], diffusion coefficient D9 and solubility parameter [k.sub.H], were changed to those of the [CO.sub.2]-PP system.

Figure 14 illustrates the simulation calculation results for two cases with different parameter values and compares the results with Case 2. In the first case, the surface tension of [N.sub.2]-PP drops from 0.0162, the value of [N.sub.2], to 0.0123 J/[m.sup.2], the value of [CO.sub.2] in Table 4; in the second case, the diffusion coefficient increases from 7.23E-9 for [N.sub.2]-PP to 8.07E-9 [m.sup.2]/s for the [CO.sub.2]-PP system. The simulation indicates that the difference in diffusion coefficient between [N.sub.2] and [CO.sub.2] does not affect the bubble nucleation behavior significantly but that a change in surface tension could alter the behavior considerably. If [N.sub.2] had as low a surface tension as [CO.sub.2], [N.sub.2] foaming could provide additional high bubble density. This result indicates that the lower surface tension of [CO.sub.2] was not the key factor causing [CO.sub.2] to have a lower number density of bubbles than [N.sub.2].

[FIGURE 14 OMITTED]

Figure 15 illustrates the simulation calculations with three different solubility parameters, where we assume that the solubility of [N.sub.2] in PP slightly increased toward that of [CO.sub.2]. When the solubility parameter is increased (i.e., the solubility of [N.sub.2] is increased), the number density of bubbles decreases, and the foaming behavior resembles the [CO.sub.2] foaming case. When the solubility parameter of [N.sub.2] is set to the same value as [CO.sub.2], very few bubbles are nucleated. These parametric simulation results confirm our speculations that the solubility of foaming agent in polymer plays a key role of determining the number of bubbles, and that the lower solubility of [N.sub.2] in PP increases the number of bubbles and decreases cell size in [N.sub.2] core-back foam injection molding when compared to [CO.sub.2].

[FIGURE 15 OMITTED]

CONCLUSION

Visual observation and numerical simulation of a core-back foam injection molding process were conducted. Polypropylene foam was prepared by using either [N.sub.2] or [CO.sub.2] as a blowing agent, and the foaming performances of these two blowing agents were compared by using an injection molding machine as well as a batch foaming setup. Both experiments and numerical simulations showed an increase in number density of bubbles and a decrease in bubble growth rate as core-back rate increased for both [N.sub.2] and [CO.sub.2] injection foaming. Furthermore, the lower solubility of [N.sub.2] in PP enabled a higher number density of bubbles and a smaller bubble size in [N.sub.2] core-back foam injection molding compared to [CO.sub.2] foam injection. Even when the initial gas concentration of [N.sub.2] was one-third that of [CO.sub.2], [N.sub.2] injection foaming was capable of producing a higher number density of bubbles with smaller cell sizes than [CO.sub.2]-injection foaming. It was also confirmed that classical bubble nucleation and growth models of batch foaming could successfully simulate the foaming behavior in the mold cavity during core-back foam injection molding.

ACKNOWLEDGMENTS

The authors thank Dr. Y. Fujita of Mitsubishi Chemical Corporation and Mr. M. Arai, Mr. M. Shimouse, and Mr. S. Fukunaga of Japan Polypropylene Corp. for providing the ICP samples and for useful discussion.

REFERENCES

(1.) C.A. Villamizar and CD. Han. Polym. Eng. Sci., 18, 699 (1978).

(2.) CD. Han and H.J. Yoo. Polym. Eng. Sci., 21, 518 (1981).

(3.) A. Arefmanesh and S.G. Advan. Polym. Eng. Sci., 35, 252 (1995).

(4.) A. Osorio and L.-S. Turng. Polym. Eng. Sci., 44, 2274 (2004).

(5.) G. Teramoto, H. Ae, and T. Kanai. Seikei-Kakou, 18, 510 (2006).

(6.) M.R. Barzegari and D. Rodrigue. Polym. Eng. Sci., 49, 949 (2009).

(7.) J.W.S. Lee, J. Wang, C.B. Park, and G. Tao, SPE ANTEC, 743 (2007)

(8.) T. Ishikawa and M. Ohshima. Polym. Eng. Sci., 51, 1617 (2011).

(9.) J.W.S. Lee and C.B. Park. Macromol. Matter. Eng., 291, 1233 (2006).

(10.) S.G. Kim, B.S. Kang, and C.B. Park, SPE ANTEC, 2777 (2005)

(11.) S.G. Kim, C.B. Park, and M. Sain. Cell. Polym., 25, 19 (2006).

(12.) S.G. Kim, J.W.S. Lee, C.B. Park, and M. Sain. J. Applied Poly. Sci., 118, 1691 (2010).

(13.) E. Di Maio, G. Mensitieri, S. Iannace, L. Nicholas, W. Li, and R.W. Flumerfelt. Polym. Eng. Sci., 45, 432 (2005).

(14.) K. Taki, Chem. Eng. Sci., 63, 3643 (2008)

(15.) K. Taki, T. Nakayama, T. Yatsuzuka, and M. Ohshima. J. Cell. Plast., 39, 155(2003).

(16.) K. Taki, Ph. D. Dissertation, Kyoto University (2005)

(17.) K. Taki, T. Yanagimoto, E. Funami, M. Okamoto, and M. Ohshima. Poly. Eng. Sci., 44, 1004 (2004).

(18.) Y. Sato, K. Fujiwara, T. Takikawa, Sumarno, S. Takishima, and H. Masuoka. Fluid Phase Equilibria, 162, 261 (1999).

(19.) S. Areerat, E. Funami, Y. Hayama, D. Nakagawa, and M. Ohshima. Polym. Eng. Sci., 44, 1915 (2004).

(20.) X. Chen, J.J. Feng, and C.A. Bertclo. Polym. Eng. Sci., 46, 97 (2006).

Correspondence to: Takeshi Ishikawa; e-mail: 380528l@cc.m-kagaku.co.jp

Published online in Wiley Online Library (wileyonlinelibrary.com). [c] 2011 Society of Plastics Engineers

Takeshi Ishikawa, (1) Kentaro Taki, (2) Masahiro Ohshima (2)

(1) Advanced Products Laboratory, R&D Center, Mitsubishi Chemical Corporation, Yokkaichi Plant, 1, Toho-cho, Yokkaichi, Mie 510-8530, Japan

(2) Material Processing Engineering Division, Department of Chemical Engineering, Kyoto University, Kyoto 615-8510, Japan

DOI 10.1002/pen.22154
COPYRIGHT 2012 Society of Plastics Engineers, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2012 Gale, Cengage Learning. All rights reserved.

 Reader Opinion

Title:

Comment:



 

Article Details
Printer friendly Cite/link Email Feedback
Author:Ishikawa, Takeshi; Taki, Kentaro; Ohshima, Masahiro
Publication:Polymer Engineering and Science
Article Type:Report
Geographic Code:9JAPA
Date:Apr 1, 2012
Words:4945
Previous Article:Synthesis, characterization of an epoxy-terminated carbosiloxane and its application as a modifier in epoxy systems.
Next Article:The effects of radiation cross-linking and process parameters on the behavior of polyamide 12 in vacuum thermoforming.
Topics:

Terms of use | Copyright © 2014 Farlex, Inc. | Feedback | For webmasters