A classical method for polymer characterization and polymer process monitoring is making a comeback.

Ingo Alig
Dirk Lellinger

For decades, ultrasonic techniques have been excellent tools for nondestructive testing and imaging. Although the relationships between material properties and acoustical parameters have been studied for a long time (1), ultrasonic devices are not used frequently for material characterization. With the development of high-frequency digital and computer techniques, it was possible to overcome some of the limitations in applying ultrasonic methods to material characterization and process monitoring.

When propagated in polymeric materials, acoustic waves are influenced by the polymer's structure as well as by molecular relaxation processes. It is possible to estimate the viscoelastic properties of polymeric materials from the velocity and attenuation of longitudinal or shear waves (2, 3). One can characterize the viscoelastic properties of polymer melts (4), as well as those of semicrystalline polymers (5). Furthermore, ultrasonic methods have been successfully used to monitor polymer processing (6), chemical reactions (e.g., polymerization or curing of thermosets) (7, 8), film formation from aqueous polymer dispersions (9), glue processes, or crystallization in polymers (10, 11).

Although many investigators have used longitudinal wave experiments, applying shear waves to polymeric materials is rare. In this article, we present some experimental results of using a shear wave reflection technique (12) during isothermal film formation from aqueous dispersions, isothermal crystallization in semicrystalline polymers, and temperature-dependent investigations near the glass transition in amorphous polymer films. We also give examples for applying longitudinal ultrasonic wave techniques to characterize viscoelastic behavior in dispersions and in-line monitoring in the extrusion process (13).

Ultrasonic shear rheology

Film formation and crystallization. The shear wave reflection technique for temperature-dependent measurements of the complex dynamic shear modulus (G* = G' + iG'') was described by Mason et al. (14) as early as 1949. Recently, we adapted the method to the high-frequency digital technique. The principles underlying a conventional single-frequency pulse-burst system using digital analysis of the received echo train are shown in Figure 1. An ultrasonic shear wave is transmitted by a piezoelectric transducer into a fused quartz bar. The shear wave is reflected off the quartz-sample interface and is detected by a second transducer at the other end of the quartz bar. The reflection coefficient r and phase shift of the reflected shear wave will be influenced by the sample. For viscoelastic materials, the complex shear mechanical impedance Z* can be estimated from r and . Using Z* = R + iX , one can calculate the storage modulus G' = (R²–X²)/ and the loss modulus G'' = (2 x RX)/, where is the density. It is noteworthy here that the shear modulus in viscoelastic materials depends on the ultrasonic frequency.

More recently, we have developed a broadband pulsed ultrasonic spectroscopic system (12, 15) that uses Fourier analysis to measure phase and amplitude changes of the reflected signals. All of the electronic components were integrated into a personal computer (PC) to make the system usable in industrial laboratories. Using digital analysis of the received signal, it was possible to simplify the experimental setup and extend the limits of the shear wave reflection technique to higher moduli and a lower film thickness.

The temperature-controlled cell housing shown in Figure 2 contains two reflection cells for simultaneous measurements. With this equipment, we can monitor time-dependent isothermal processes and reactions. We applied the method to fundamental processes (e.g., film formation, reaction kinetics, and phase transitions) and technological assessment (glues, paints, etc.). We have also made temperature-dependent studies of the melting and crystallization of semicrystalline polymers and of the glass-transition behavior of films.

We used the ultrasonic shear wave reflection technique (Figure 3) (12) to investigate isothermal film formation and crystallization kinetics of an amorphous and a semicrystalline polychloroprene sample (13, 14). The polychloroprene films were formed from aqueous latex dispersions supplied by Bayer AG (Leverkusen). Using X-ray scattering, we found that the content of 1,4-cis sequences in the amorphous sample was higher (5.2%) than that in the semicrystalline sample (3.8%). The solids content was 58.0% for the amorphous samples and 46.6% for the semicrystalline samples. The measurements of the dynamic shear modulus have been performed at a frequency of 5.32 MHz. The process of film formation during the evaporation of water is expressed by a stepwise increase of the shear modulus. For the semicrystalline samples, a further increase in shear modulus due to crystallization can be observed. The nucleation and crystallization is triggered by the film formation, but it is suppressed in the dispersed latex particles. From the polarization micrograph (Figure 3, top), it is evident that the semicrystalline polychlorprene crystallizes in spherulites. The isothermal crystallization kinetics can be simulated (10, 11) by combining the Avrami equation (16) with the Kerner model (spherulites in an amorphous matrix) (17).

Comparison of ultrasonic and conventional rheology. In addition to time-dependent measurements, the setup can also be applied for temperature-dependent measurements of the final films. In Figure 4, the temperature dependence of G' and G'' measured with "ultrasonic" (3.5 MHz, samples AC, BS1, and BS2; see next paragraph) and conventional rheology (1 Hz, sample BS2 only) are compared for two acrylic acid-type copolymers.

We prepared the films from commercial aqueous dispersions (BASF, Ludwigshafen). The samples were two n-butyl acrylate-styrene copolymers (BS1 and BS2) with different monomer ratios and the ethylhexyl acrylate-methyl methacrylate copolymer (AC). The glass-transition temperatures (T_g) for these materials were -4 °C and 23 °C for BS1 and BS2, respectively, and -46 °C for AC.

To extend the temperature range, we constructed a "master plot" for the ultrasonic data of BS1, BS2, and AC using T–T_g instead of T. The G' and G'' data from both methods can be fitted by the same set of parameters using the Havriliak-Negami function (18) incorporating the Vogel-Fulcher equation (19, 20) for the temperature dependence of relaxation times. The agreement between both methods is good. This suggests an almost thermo-rheological simplicity of the samples and demonstrates the capacity of the ultrasonic rheometer. For details of the fits and the parameters, see Reference 9.

After we increased the operating temperature range of the instrument, we recently applied the ultrasonic film rheometer for characterization of different rubber mixtures in a temperature range from -35 °C to 60 °C.

Pulse-transmission technique for longitudinal waves change

Ultrasonic spectroscopy. The ultrasonic spectroscopy system for measuring the longitudinal ultrasonic velocity and attenuation in polymer dispersions and in the extrusion process is similar to those described in Reference 15. The longitudinal wave measurements described here have been done with the pulse-transmission technique (Figure 5). Phase and amplitude spectra are calculated after digitizing the transmitted pulse. The receiver and pulse generator are integrated in a PC.

Characterizing polymer dispersions. Usually, dried films have to be prepared to characterize the mechanical properties or the glass-transition temperature of the polymer component in aqueous dispersions. This procedure is time-consuming, and the property of the film may differ from that of the polymer latex particle, because, among other reasons, a small amount of water is dissolved in the polymer.

Two decades ago (21, 22), using an analogue technique, Hauptmann and colleagues showed that the maximum in the temperature dependence of the longitudinal ultrasonic attenuation measured in an aqueous dispersion is related to the main glass-to-rubber transition in the polymer latex particles. The temperature of the ultrasonic loss maximum can be related to T_g measured by calorimetry or dynamic-mechanical analysis using the frequency-temperature superposition principle (3, 19, 20).

Starting from this idea, we developed an ultrasonic system for measuring the glass transition directly in the aqueous polymer dispersions, that is, without film preparation. The system is based on a Fourier-transform pulse-transition technique and records ultrasonic attenuation and sound velocity during a temperature ramp (Figure 6). The maximum in attenuation and the inflection point in the sound velocity can be related to T_g measured by differential scanning calorimetry (DSC). The calorimetric glass-transition temperatures are 5 °C for the butyl acrylate-styrene and 25 °C for butadiene-styrene dispersions. For copolymers or blends, further information may be extracted from a closer analysis of the shape of the curves.

Monitoring the extrusion process. The techniques we have discussed were developed for use in research or industrial laboratories. Several attempts to apply ultrasonic techniques to the monitoring in polymer processing have been reported in recent decades; however, the acceptance and application of those techniques in industry is still limited. The reasons are

• the lack of hardware that is robust enough for processing conditions,
• the insufficient long-term stability of transducers at high temperatures, and
• the need for adaptation to the specific process, for example, customized software.

In Figure 7, an example of the application of a longitudinal ultrasonic technique to the monitoring of filler content, viscosity, or chemical composition in the extrusion process is shown schematically. The excess attenuation relative to a polyethylene sample was measured by an in-line pulse transmission technique in which the two transducers were positioned behind the end of the extruder screw (13). At the same position, the viscosity () was measured by an in-line slot die rheometer. The changes of both quantities for two polystyrene samples with different melt viscosities are shown in Figure 8. A clear correlation between the melt viscosity (expressed by the melt flow index, MFI) and the ultrasonic attenuation of polystyrene is obvious and can be related to the different glass-transition temperatures. As another example, the dependence of the ultrasonic attenuation on the composition is shown in Figure 9 for a poly(methyl methacrylate)–acrylonitrile–butadiene–styrene (PMMA–ABS) blend. Using this technique, we were able to find a relationship between ultrasonic attenuation or velocity and the elastomer or filler content in polymer melts during extrusion.

Ultrasonic studies: an idea whose time has come again

The recent high rate of development in high-frequency digital and computer techniques has opened new possibilities for using ultrasonic equipment for laboratory studies and industrial process control. It is now possible to redesign classical ultrasonic methods to be less expensive, more fully automated, and more robust. By describing some of the method developments performed since 1993 in our laboratory and giving some examples of the application of ultrasonic shear and longitudinal waves to characterize different classes of polymeric materials, it was our aim to demonstrate the feasibility of ultrasonic methods in polymer research, including monitoring and process control. Although the application of ultrasonic methods has a long history, we believe it is time to revive this "old" method.

Acknowledgments
The work has been supported by the Bundesminister für Wirtschaft through the Arbeitsgemeinschaft industrieller Forschungsvereinigungen e.V., AiF-Grant nos. 10835, 10943, 11961N, 12152N, and 12243N. We are also very grateful to the support by the Bayer AG and BASF AG.

This article is adapted from an article written by I. Alig, D. Lellinger, and S. Tadjbakhsch that appeared in Polymeric Materials Science and Engineering 1998, 79, 31-32.

1. In Physical Acoustics; Mason, W. P., Ed.; Academic Press, New York, 1964; Vol. 1, Part A.
2. Alig, I. In Handbuch der Kunststoffprüfung; Schmiedel, H., Ed.; Hanser Verlag: Munich, 1992; Chapter 10.2, p 393.
3. Ferry, J. D. Viscoelastic Properties of Polymers; Wiley & Sons: New York, 1961.
4. Alig, I.; Stieber, F.; Bakhramov, A. D.; Manucarov,Y. S.; Solovyev, V. A. Polymer 1989, 30, 842-848; 1990, 31, 877-881.
5. Perepechko, I. I. In Akusticheskie metody issledovaniya polimerov; Khimiya: Moscow, 1973.
6. Gendron, R.; Dumoulin, M. M.; Piche, L. Polym. Mater. Sci. Eng. 1995, 72, 23-24.
7. Alig, I.; Lellinger, D.; Johari, G. P. J. Polym. Sci., Polym. Phys. Ed. 1992, 30, 791-799.
8. Parthun, M. G.; Johari, G. P. J. Chem. Phys. 1995, 102, 6301-6307.
9. Alig, I.; Tadjbakhsch, S.; Zosel, A. J. Polym. Sci., Polym. Phys. Ed. 1998, 3, 1703-1711.
10. Alig, I.; Tadjbakhsch, S. J. Polym. Sci., Polym. Phys. Ed. 1998, 36, 2949-2959.
11. Alig, I.; Tadjbakhsch, S.; Floudas, G.; Tsitsilianis, C. Macromolecules 1998, 31, 6917-6925.
12. Alig, I.; Lellinger, D.; Sulimma, J.; Tadjbakhsch, S. Rev. Sci. Instrum. 1997, 68, 1536-1542.
13. Alig, I.; Lellinger, D.; Lamour, R.; Ramthun, J. Kunststoffe, submitted.
14. Mason, W. P.; Baker, W. O.; McSkimim, H. J.; Heiss, J. H. Phys. Rev. 1949, 75, 936-946.
15. Alig, I.; Lellinger, D. J. Appl. Phys. 1992, 72, 5565-5570.
16. Avrami, M. J. J. Chem. Phys. 1939, 7, 1103-1112; 1940, 8, 212-224; 1941, 9, 177-184.
    
17. Kerner, E. H. Proc. Phys. Soc. 1956, 69B, 808-813.
18. Havriliak, S.; Negami, S. Polymer 1967, 8, 161-210.
19. Vogel, H. Phys. Z. 1921, 22, 645-646.
20. Fulcher, G. S. J. Am. Chem. Soc. 1925, 8, 339-345.
21. Hauptmann, P.; Herold, H.; Rothe, B.; Zenker, R. Plaste Kautsch. 1979, 26, 307-309.
22. Hauptmann, P.; Herold, H.; Pšge W.; Rothe, B.; Zenker, R. Plaste Kautsch. 1981, 28, 22-25.

Dirk Lellinger (dlellinger@dki.tu-darmstadt.de) has been a senior scientist at the German Plastics Institute in Darmstadt, Germany, since 1995. His primary interests are ultrasonic spectroscopy, dynamic-mechanical spectroscopy, and the development of automated measurement systems. He received his Diploma and a Ph.D. from the Technical University in Merseburg. From 1993 to 1995, he spent two years at the department of physics at the University of Halle.

Return to Top