Page 658 - Handbook of Thermal Analysis of Construction Materials

626 Chapter 15 - Roofing Materials

Changes greater than these values indicate changes in the material resulting
from the exposure. A standard test procedure for DMA/TPA was also
recommended.
The conclusions from the DSC analysis were that the technique is
not universally applicable to the analysis of roofing materials. Since the
variability of the method was not investigated, recommendations were not
made. The use of newer techniques such as modulated DSC (MDSC) and
oscillating DSC (ODSC) to characterize the T range of roof membrane
g
materials was beyond the scope of the study. Therefore, the Committee did
not recommend the use of DSC as a universal method. However, it was
recommended that the applicability of MDSC and ODSC for determining
glass transition of roofing membrane materials be investigated.
The applicability of thermoanalytical techniques for the character-
ization of roofing membrane materials is gaining popularity in the roofing
industry. Several papers have been published in the last five years. [53]–[59]
For example, more and more manufacturers, contractors, consultants, and
building owners are turning to these techniques to solve in-service roofing
membrane problems as well as to evaluate new roofing membrane materi-
als. A typical example is the paper on the problem of EPDM membrane
shrinkage. [55] The study was conducted in 1995 to investigate the causes of
EPDM shrinkage. The performance of EPDM roofs, typically, had been
quite good, however, instances of flashing problems attributed to mem-
brane shrinkage had occurred. The laboratory study included TG, DMA,
tensile strength, elongation, and thermally induced load. The TG results
showed that the oil loss of the EPDM membrane overlapped with the region
where the polymer degraded. Therefore, it was difficult to measure the total
oil content for each sheet.
Since the EPDM rubber polymer is relatively stable and not
expected to decompose when exposed to natural weathering, Oba and Paroli,
et al., [54][55] compared the difference in oil content between the top and
bottom layers by subtracting the total weight loss below the polymer
decomposition region (500°C) for the top sheet from that of the bottom
sheet. It was found that for some membranes, the oil loss during exposure
to natural weather was as high as 8% when compared to the bottom sheet.
In addition, the TG results showed that for one sample taken from near a
flashing, the weight loss was not typical of an EPDM membrane.
The glass transition measured by DMA also showed similar results.
An infrared spectrum taken from the sample showed that the sample taken
from the flashing was neoprene-based. These thermoanalytical techniques