New method for predicting lifetime of seals from compression-stress relaxation experiments Page: 3 of 32
This report is part of the collection entitled: Office of Scientific & Technical Information Technical Reports and was provided to Digital Library by the UNT Libraries Government Documents Department.
The following text was automatically extracted from the image on this page using optical character recognition software:
When oxidative chemical degradation effects are significant during stress-relaxation
experiments, attempts to analyze and model the data are complicated by two phenomena, the
need to separate the physical and chemical relaxation processes and the potential presence of
anomalous diffusion-limited oxidation (DLO) effects. Various workers have suggested methods
for separating physical from chemical effects, including Curro and Salazar , Stenberg and
Jansson , Ito  and Murakami, et. al. . In this study, we utilize a method based on the
Curro-Salazar approach to show that physical effects are unimportant for the temperatures and
times used for probing the chemical effects. Previous work attempting to understand and model
DLO effects on stress-relaxation experiments was mainly due to Murakami and co-workers [10-
13]. These workers tried to quantitatively model tensile stress-relaxation results versus sample
thickness and surrounding oxygen partial pressure using a diffusion-reaction equation that
assumed a linear dependence of reaction rate on oxygen concentration. Although this
assumption may be valid for very low oxygen partial pressures, it is usually inappropriate under
typical application conditions (i.e., air); in their later papers, Murakami, et. al.  acknowledge
this conclusion. A more general diffusion-reaction model for DLO effects on sheet material was
derived by Cunliffe and Davis  using a higher-level reaction rate expression based on the
classic basic autoxidation scheme [16,17]. Although the bimolecular termination reactions used
in the Cunliffe and Davis model are inappropriate for stabilized elastomers, recent work [18,19]
indicates that unimolecular termination variations of the basic oxidation scheme give a DLO
model essentially identical to that derived by Cunliffe and Davis. Careful studies over the past
few years have shown that this general DLO model gives quantitative agreement with
experimental results for elastomers aged in both radiation and thermoxidative environments [18-
20]. In the current paper, we couple this more general oxidation rate expression with diffusion
expressions appropriate to the disk geometry used in most CSR experiments to derive the DLO
model appropriate to CSR studies. We use this modeling coupled with experimental results to
show that large errors are introduced by DLO effects for typical CSR samples. Finally we
introduce a method for eliminating such errors.
Compression molded sheets of a commercial, peroxide-cured butyl rubber were obtained from
Parker Seal Company (Parker compound B612-70) in two thicknesses (-1 mm and -2 mm). To
prepare the cylindrical samples needed for standard compression stress relaxation experiments,
three 12.7 mm diameter disc-shaped samples were cut from the 2 mm thick compression-
molded sheets and stacked on top of one another to obtain an unstrained cylinder of 12.7 mm
diameter and a typical thickness of 6 mm. To make 2 mm diameter disks, a stainless steel
circular die with very thin cutting surfaces was specially prepared. The 2 mm mini-disk samples
were cut from the 1-mm thick sheet, so that the ratio of disk diameter to disk thickness was
approximately the same as for the standard samples. Since the area of a single 2 mm disk leads
to a very small force when compressed, obtaining reasonable CSR force values requires the use
of multiple minidisks strained in parallel (the current series of experiments uses 50). To ensure
that the 50 mini-disks for a given experiment had similar unstrained thicknesses, they were all
cut from a small area of a sheet (-2 cm by 2 cm), chosen such that little thickness variation
occurred in the region used for cutting. To obtain mini-samples whose sides were perpendicular
to their base, it was found necessary to ensure that the 2-mm cutter was kept precisely
Here’s what’s next.
This report can be searched. Note: Results may vary based on the legibility of text within the document.
Tools / Downloads
Get a copy of this page or view the extracted text.
Citing and Sharing
Basic information for referencing this web page. We also provide extended guidance on usage rights, references, copying or embedding.
Reference the current page of this Report.
Gillen, K.T.; Keenan, M.R. & Wise, J. New method for predicting lifetime of seals from compression-stress relaxation experiments, report, June 1, 1998; Albuquerque, New Mexico. (digital.library.unt.edu/ark:/67531/metadc710623/m1/3/: accessed February 23, 2019), University of North Texas Libraries, Digital Library, digital.library.unt.edu; crediting UNT Libraries Government Documents Department.