Rubber analysis made more accurate and informative by special TGA techniques.

Thermogravimetry (TG) according to DIN 51006 or ASTM E 1131 has established itself as an important physical analytical method in the rubber testing laboratory (refs. 1 and 2). As yet, however, there have been restrictions on accurate quantitative separation of mass loss steps which can be related to certain polymers or additives (plasticizers, fillers). Additionally, it was not possible to sufficiently identify individual components.

Using a newly designed, top-loading and vacuum-tight thermogravimetric analyzer (figure 1), the sample temperature is directly measured at the crucible, i.e., where the sample is located, employing a thermocouple from Platinel, a PdPtAu/AuPd alloy. This allows, for the first time, an exact temperature calibration through transition temperatures of reference materials, such as melting of indium or aluminum. Formerly, TGA temperature calibrations could only be performed using time-consuming and relatively inaccurate Curie Point measurements - examining magnetic properties of certain metals while suspended in the heated furnace.

[FIGURE 1 OMITTED]

The evaluated DTA curve (c-DTA, calculated differential thermal analysis), yields information on exothermic (e.g., post-vulcanization) or endothermic effects of the sample (e.g., melting of lubricant wax). Figure 2 depicts the double-step melting (without mass loss) of a filled thermoplastic elastomer at 127[degrees] C and 161d[degrees] C prior to the decomposition of the copolymer at 453[degrees] C (c-DTA curve) and 474[degrees] C (DTG curve), respectively, with a mass loss of 89.1%.

[FIGURE 2 OMITTED]

Using a unique rate-controlled mass loss heating method, three rate-controlled mass change temperature control functions are available in the evaluation software, thus offering a considerably higher separation of the individual mass loss steps compared to conventional measurements at a constant heating rate. For a complete, quantitatively correct rubber analysis, the dynamic mode proved to be successful. With this mode, the heating rate is continually decreased upon exceeding the threshold value, determined by the operator, until the mass loss is finished.

Figure 3 shows the decomposition of a rubber tire mixture (red TG curves) with a considerably better separation of the two elastomer components (dotted red DTG curve), as well as detection of the softener (DTG shoulder at 291[degrees] C) in comparison to a conventional TGA measurement at 20 K/min. (green curves). With this method, the amount of the second elastomer component is more accurately calculated to be the expected value of about 22% (22.1%) instead of only 15.8%.

[FIGURE 3 OMITTED]

An extremely important new method is the use of vacuum conditions during the TGA experiment, referred to as vacu-um-TGA. With this method (ref. 3), better separation of the low-molecular plasticizer from the polymer component is achieved. Due to the boiling point reduction, a considerably better determination of the plasticizer content is now possible. Normally, determination of plasticizer content was only done using time-consuming, and sometimes hazardous wet-chemical procedures such as the Soxhlet extraction method.

The green TG curve with DTG signal (dotted) in figure 4 delivers the first two decomposition steps of the plasticized PVC (PVC-P) to 380[degrees] C. When conducting a conventional TGA measurement in a nitrogen atmosphere at 5 K/min., the release of the plasticizer is overlapped by the separation of hydrochloric acid from PVC. The DTG curve shows only one shoulder at 256[degrees] C, which could be related to a plasticizer amount of approximately 14%. Only by employing the vacuum method (red curves), is it possible to obtain a clear separation of the plasticizer at 214[degrees] C (DTG peak) from the HC1 release at 291[degrees] C. The plasticizer content was determined to 36.8%.

[FIGURE 4 OMITTED]

In addition, special peak separation software (ref. 4) can now be used to identify differential TGA curves (DTG), yielding even more information on the individual components in a rubber mixture. Figure 5 depicts the test results of a carbon black-filled NR/SBR mixture where the plasticizer content was determined to be 11.8%, the NR component 30.1% and the SBR component to 21.0% using the non-linear regression calculation. The correlation coefficient for the curve-fit amounts to 0.9998. The relative error of the individual component is between 2% and 4%.

[FIGURE 5 OMITTED]

Finally, it should be pointed out that a top-loading TGA is well-suited for coupling to a Fourier-Transform Infrared-Spectrometer (FT-IR) and/or a mass spectrometer (MS) through a heated Y adapter, allowing qualitative detection of the gases evolved during decomposition. Top-loading designs are found to better support the natural gas flow path, ensuring more complete collection and analysis of off-gasses during the heating process.

Summary

New and improved thermogravimetric (TGA) measuring techniques, such as rate-controlled mass loss (Super-Res) and the unique vacuum-TGA method, have now been demonstrated to allow better separation of ingredients in a complex rubber mixture. Together with coupling to on-line gas analysis methods such as FT-IR and mass spectrometry,

TGA can perform increasingly comprehensive and accurate analyses of polymers. Employing special evaluation software packages such as c-DTA and Peak Separation, information on TGA test results can be even further increased.

References

(1.) Mohler, H., Stegmayer, A. and Kaisersberger, E., "Chancen und Moglichkeiten fur die thermische analyse nach DIN 51005 bei der qualitatssicherung in der gummiindustrie," Kautschuk Gummi Kunststoffe, 4/91.

(2.) Kaisersberger, E., Knappe, S., Mohler H. and Rahner, S., TA fur die polymertechnik, DSC- TG - DMA - TMA, Band 3, Netzsch-Geratebau, 1994.

(3.) Affolter, S., Schmidt, M. and Wampfler, B., "Ringversuche an polymeren werkstoffen: Thermoanalytische verfahren," Kautschuk Gummi Kunststoffe, 78/99.

(4.) Peak Separation Software, Netzsch-Geratebau, Version 2000.

(5.) Kaisersberger, E., Post, E., Opfermann, J. and Emmerich, W.-D. "Combined thermal analysis and gas-analysis methods and software simulations for the investigation of the potential endangerment to the environment through production and recycling processes," Thermochimica Acta, 325/1999.