In situ, quantitative characterization of gelation and fusion mechanism in poly(vinyl chloride) plastisols by small angle light scattering (SALS).

INTRODUCTION

Poly(vinyl chloride) (PVC) plastisol is a suspension of fine particles of PVC resin in a liquid plasticizer (1). The PVC resin used for the plastisol applications (namely, dispersion resin), which is different from the more general grade of PVC (namely, suspension resin), consists of the primary particles and their agglomerates. The former are products of polymerization and the latter result from spray drying and subsequent grinding, which are post-polymerization processes. The typical particle size of the commercial dispersion resins which include agglomerates as well as primary particles ranges from about 15 to 0.2 [[micro]meter] and smaller (2).

Common features of fabricating flexible products from plastisol include i) spreading or coating the plastisol on a substrate and then ii) heating it. As the plastisol is heated, there are penetration and diffusion of the plasticizer molecules into interstitial voids of the agglomerates and amorphous regions of PVC, producing a dry mass. As a result, the plastisol increases in viscosity and becomes translucent. This process is called "gelation." Upon continued heating to a higher temperature, microcrystallites of PVC melt and the polymer molecules fuse together more uniformly. It is noteworthy that during gelation and fusion, the PVC particles become swollen and the particulate structure eventually disappears (3).

The gelation and fusion behaviour of PVC plastisols have been characterized by a number of methods which can be divided into two groups; dynamic methods and static methods. Among various dynamic methods, in situ measurement of buildup of viscoelastic moduli employing a Rheometrics Mechanical Spectrometer (RMS) has successfully offered an effective means to elucidate the gelation and fusion mechanism (3-5). Nevertheless, the above dynamic viscoelastic method has two shortcomings. One is that it is difficult to obtain quantitative information on the progress of gelation and fusion accompanied by the changes in PVC particulate structure. The other is that because this method requires a small amplitude of oscillation, there may possibly be an influence of the mechanical work on the system, recognizing that the gelation and fusion take place under no mechanical stress in most actual commercial practice.

The static methods involve thermal analysis using a differential scanning calorimeter (DSC) and morphological observation using a scanning electron microscope (SEM). DSC measurement has shown to be useful in detecting the plasticizer diffusion into resin particles (6), but this method is less sensitive in characterizing the gelation and fusion mechanism than the dynamic viscoelastic measurement. SEM observation has provided direct information on the progress of the gelation and fusion processes and has been associated with the characteristic changes in the viscoelastic behavior during gelation and fusion (5). However, SEM observation is performed with quenched, freeze-fractured samples, in which there is a chance that the in situ morphology during gelation and fusion may be influenced and altered. In addition, a quantitative evaluation of the morphological changes during gelation and fusion is rather difficult and inaccurate.

The adoption of small angle light scattering (SALS) overcomes the shortcomings of the dynamic and static methods, whenever there is a difference in refractive indices between PVC and plasticizer. This is because the SALS in an in situ, non-destructive technique which is not only free from any mechanical perturbation but also able to provide quantitative information on the phenomena encountered during gelation and fusion of PVC plastisols.

The present work investigates the details of the gelation and fusion behaviour of three model PVC plastisols, accompanied by swelling and eventual disappearance of the PVC particles, through in situ observation with a small angle light scattering (SALS) technique.

EXPERIMENTAL

Preparation of Plastisols

The PVC dispersion resin in this study was a representative commercial product of the Geon Co., Geon 121. Its molecular weight expressed in terms of the inherent viscosity in cyclohexanone at 30 [degrees] C is 1.20 (7).

A homologous series of three plasticizers was di-isohexyl phthalate (DHP), di-2-ethylhexyl phthalate (DOP), and di-isooctyl phthalate (DIOP).

Three model plastisols (i.e., PVC/DHP, PVC/DOP, PVC/DIOP) were prepared by mixing the dry PVC resin with a liquid plasticizer in the proportion of 100 parts resin to 57 parts plasticizer by weight. Also, epoxidized soybean oil and thermal stabilizer (Ba, Cd, Zn stabilizer) were added in the proportion of 100 parts resin to 3 and 2 parts soybean oil and thermal stabilizer, respectively. Mixing was done using a mechanical stirrer with a double blade propeller. After the completion of mixing, air bubbles in the plastisols were removed by vacuuming, but slowly to prevent foaming. Then, the plastisols were aged for two weeks before they were used, during which time their viscosity increased in the beginning and then slowly approached a steady value (8). The major cause of the viscosity increase during aging has been known to be a result of deagglomeration of the PVC particles (8).

Instrument and Procedures

A time-resolved small angle light scattering (SALS) apparatus employed in this study is schematically depicted in Fig. 1, which is a modification of an Optical Multichannel Analyzer (OMA) II system originally developed at the laboratory of Stein (9). This SALS setup is composed of a dark box, a detector controller (EG&G Princeton Applied Research model 1412 Multichannel Detector), and an Optical Multichannel Analyzer (EG&G Princeton Applied Research OMA III system). The box contains a laser light source (He-Ne laser), a shutter, neutral density filters, a polarizer and analyzer, a sample hot stage with a programmable temperature controller, and a two-dimensional Vidicon camera.

As for preparing samples, a droplet of plastisol was mounted on a slide glass and covered by a cover glass. This specimen was de-aired by applying vacuum to remove air bubbles, and then mounted on the sample hot stage. The temperature was raised from 40 to 200 [degrees] C with a controlled heating rate of 2 [degrees] C/min. The heating rate was chosen arbitrarily.

The scattered intensity was monitored by means of the Vidicon camera in conjunction with the detector controller. Then, the analog signal was digitized and analyzed on the OMA III system. The SALS pattern was scanned in a two-dimensional mode. Intensities of scattered light were accumulated as a function of position for each specific temperature. The raw data were further transferred to an off-line computer for post-data treatment such as background correction, data smoothing, rescaling, and so on.

RESULTS

Shown in Figs. 2 and 3 are the variations of scattered intensity as a function of scattering wave number q at several different temperatures and the variations of scattered intensity as a function of temperature at different q for PVC/DHP plastisol, respectively. In these Figures, q = [4[Pi] sin([Theta]/2)]/[Lambda], where [Theta] is the scattering angle and [Lambda] is the wave length of light in the medium. The scattered intensity decreases as q increases, and the magnitude of the scattered intensity is relatively larger for small angles than that for wide angles. As seen in Fig. 3, the scattered intensity increases very slightly at the temperature range of 70 to 90 [degrees] C but rapidly from 110 [degrees] C, reaching a maximum at 134 [degrees] C, and then decreases rapidly. The PVC/DOP and PVC/DIOP systems, although they are not shown, indicate a behavior similar to that of the PVC/DHP system, except for the temperature where the maximum in the scattered intensity appears; 114 [degrees] C for PVC/DOP and 142 [degrees] C for PVC/DIOP.

The relationship between the scattered intensity [I.sub.S] and the scattering wave number q had been proposed by Debye and Bueche (10) and Guinier (11). According to Debye and Bueche, the scattered intensity [I.sub.S] is expressed as

[Mathematical Expression Omitted],

where K is a constant and [a.sub.1] is the short-range correlation distance. The plot of [I.sub.S] against [q.sub.2] is called a Debye-Bueche plot. Figure 4 represents the Debye-Bueche plot for the PVC/DHP system, where the short-range correlation distance [a.sub.1] can be obtained from the intercept of the y-axis and the slope of the straight line by a linear least-square fitting method; [a.sub.1] = [[slope/intercept].sup.1/2]. As seen in Fig. 4, the deviation from a straight line is found in the relatively small angle region associated with larger correlation distances. Thus, the Debye-Bueche approximation is expected to fit better for larger angles which are associated with smaller particles.

Another expression for the scattered intensity proposed by Guinier is as follows;

[Mathematical Expression Omitted],

where K is a constant and [a.sub.2] is a longer-range correlation distance. The plot of ln[I.sub.S] against [q.sup.2] is called a Guinier plot and Fig. 5 is the Guinier plot of the PVC/DHP system. The value of the long-range correlation distance [a.sub.2] can be obtained from the slope of the straight line in the plot; [a.sub.2] = 2[[[absolute value of slope]].sup.1/2]. Because the straight line is sought in a relatively small angle region, the Guinier plot is better for smaller angles which are associated with larger particles.

Both the short-range and long-range correlation distances of three plastisols are calculated and presented as a function of temperature in Figs. 6 through 8. It is noted that although the SALS measurements were performed from 40 to 200 [degrees] C, the data in Figs. 6 through 8 were reproducible only between 80 to 100 [degrees] C and 140 to 150 [degrees] C, where in the former range, the plastisols change from opaque to translucent and in the latter range, the PVC particles begin to lose their boundary identity.

DISCUSSION

Gelation and fusion of PVC plastisols are the processes in which the plasticizer diffuses into the resin particles, accompanied by swelling of the particles, and the system eventually becomes a uniform melt.

As seen in Fig. 3 for the PVC/DHP system, there is a slight change in the scattered intensity between [approximately]70 and 90 [degrees] C and then the scattered intensity goes up sharply from 110 [degrees] C to a maximum at 134 [degrees] C. The slight increase at 70 to 90 [degrees] C indicates that the PVC particles themselves are still not affected significantly by the plasticizer and only the surface of some particles may have been swollen. At this stage, the plasticizer must complete filling the interstitial voids of the agglomerated particles. The more significant increase of the scattered intensity at the temperature range higher than 90 [degrees] C means that the plasticizer diffuses into the interior of the PVC particles and makes them swollen seriously.

The above observation is in agreement with differential scanning calorimeter (DSC) measurements done with the same plastisol. When the PVC/DHP plastisol is heated to 75 [degrees] C in the first scan, the glass transition of PVC remains intact in the second scan [ILLUSTRATION FOR FIGURE 9 OMITTED]. This indicates that the resin particles at 75 [degrees] C are not affected by DHP yet. However, if the plastisol is heated to 85 [degrees] C in the first scan, the glass transition of PVC disappears in the second scan [ILLUSTRATION FOR FIGURE 9 OMITTED]. Apparently, the interior of the PVC particles is about to be affected by DHP diffusing into it.

The maximum in the scattered intensity for PVC/DHP appears at 134 [degrees] C [ILLUSTRATION FOR FIGURE 3 OMITTED], which must be due to the maximum swelling of the particles just before they lose their particle boundaries. In other words, the plasticizer is uniformly distributed inside the resin particles, and the gelation process can be said to be completed at this temperature. As indicated in RESULTS, PVC/DOP and PVC/DIOP plastisols show a maximum in the scattered intensity at 144 and 142 [degrees] C, respectively. Therefore, the completion of the gelation process has occurred at 144 [degrees] C for PVC/DOP and 142 [degrees] C for PVC/DIOP. It should be noted that the observed temperatures corresponding to the maximum swelling (i.e., 134 [degrees] C for PVC/DHP, 144 [degrees] C for PVC/DOP, 142 [degrees] C for PVC/DIOP) are valid only when the heating rate of 2 [degrees] C/min is selected. Furthermore, by relating the maximum in the scattered intensity to the corresponding temperature where it appears for three plastisol systems, it is possible to evaluate the relative solvent power of the plasticizers. The temperature difference of 8 to 10 [degrees] C for the appearance of the maximum scattered intensity in PVC/DHP means that the gelation step is completed earlier compared to the other two systems and thus, DHP has better solvent power than DOP or DIOP. Based on the above consideration, DOP and DIOP show almost a similar solvent power.

The short-range and long-range correlation distances, [a.sub.1] and [a.sub.2] approximated by Debye-Bueche and Guinier represent the distance between the centers of neighboring particles (10, 11). In the case of PVC, because the particles are packed very closely as evidenced in Fig. 10, both correlation distances are regarded as the average values of the diameters of swelling particles. In addition, the short-range correlation distance [a.sub.1] represents an average size of relatively small particles and the long-range correlation distance [a.sub.2] exhibits that of relatively large particles.

At the temperature range of 80 to 100 [degrees] C, all the values of [a.sub.1] for three plastisols are about 0.10 to 0.14 [[micro]meter] and those of [a.sub.2] lie between 0.5 and 0.6 [[micro]meter]. The [a.sub.2] values are in accordance with the median particle size of unswollen resin, 0.36 to 0.7 [[micro]meter], which had been characterized with a sedimentation method (8). Moreover, the sedimentation method had revealed a presence of particles smaller than 0.36 [[micro]meter], i.e., an ultrafine fraction of the resin. Therefore, the values of [a.sub.1] at 80 to 100 [degrees] C indicate the presence of ultrafine particles. Considering both [a.sub.1] and [a.sub.2] values at 80 to 100 [degrees] C and the DSC results shown in Fig. 9, it is thus concluded that although the plasticizers diffused into the interior of the PVC particles at this temperature range, they did not make the particles significantly swollen.

Recognizing that the increase in the correlation distances with increasing temperature corresponds to swelling of the PVC particles with plasticizer, it is interesting to compare the temperatures of individual plastisols at which the correlation distances begin to increase rapidly and the maximum correlation distances are obtained. The reason why the rapid increase in [a.sub.1] and [a.sub.2] occurs at different temperatures for the different plastisols is attributable to the difference in the solvent power of individual plasticizers. Therefore, the lower the on-set temperature of the rapid increase indicates the better solvent power of the plasticizer. The lower the temperature at which the maximum swelling of the particles is attained indicates the better solvent power as well.

Figures 11 and 12 represent the above considerations. As shown in these Figures, the rapid increase in the correlation distances appears at the temperature range of 100 to 110 [degrees] C for PVC/DHP, [approximately]110 [degrees] C for PVC/DIOP, and [approximately]120 [degrees] C for PVC/DOP. Moreover, the temperature at which the maximum values of [a.sub.1] and [a.sub.2] appear is observed in the vicinity of 135 [degrees] C for PVC/DHP, and 143 [degrees] C for PVC/DOP and PVC/DIOP. On the basis of the above observations, DHP proves to have better solvent power for PVC than DOP or DIOP.

CONCLUSION

The small angle light scattering (SALS) method was shown to provide an effective, non-destructive means for characterizing the changes of PVC particulate structures during gelation and fusion of PVC plastisols. In particular, measurements of the short-range and long-range correlation distances based on Debye-Bueche and Guinier approximations revealed not only in situ, quantitative information on the swelling behavior of PVC particles in a plastisol, but also an evaluation of the relative solvent power of the plasticizers.

The SALS method offered a strong potential for real-time monitoring of the gelation and fusion behavior, which is essential for designing and manufacturing flexible PVC products with desired properties.

ACKNOWLEDGMENT

The author would like to thank Dr. N. Nakajima at The University of Akron for his valuable discussion, and Dr. T. Kyu at The University of Akron for his helpful comments on performing the SALS experiment.

REFERENCES

1. J. A. Davidson and K. L. Gardener, "Plastisols," in Encyclopedia of Chemical Technology, 3rd Ed., Vol. 23, Kirk-Othmer, ed., John Wiley & Sons, New York (1983).

2. J. Frados, ed., Plastics Engineering Handbook of the Society of the Plastics Industry, Inc., 4th ed., Van Nostrand Reinhold, New York (1976).

3. N. Nakajima, J. D. Isner, and E. R. Harrell, J. Macromol. Sci.-Phys., B20, 349 (1981).

4. N. Nakajima and E. R. Harrell, Adv. Polym. Technol., 6, 409 (1986).

5. N. Nakajima and S.-Y. Kwak, J. Vinyl Technol., 13, 212 (1991).

6. N. Nakajima, C. M. Yavornitzky, E. J. Roche, and E. R. Harrell, J. Appl. Polym. Sci., 32, 3749 (1986).

7. C. A. Daniels and E. A. Collins, J. Macromol. Sci.-Phys., B10, 287 (1974).

8. N. Nakajima and C. A. Daniels, J. Appl. Polym. Sci., 25, 2019 (1980).

9. R. Tabar and R. S. Stein, J. Polym. Sci.-Polym. Phys. Ed., 20, 2041 (1985).

10. P. Debye and A. Bueche, J. Appl. Phys., 20, 518 (1949).

11. A. Guinier and G. Fournet, Small Angle Scattering of X-Rays, John Wiley & Sons, New York (1955).