Effect of Strain on the Modulus of Sealants Exposed to the Outdoors

ABSTRACT: The effects of applied strain on sealants exposed to outdoor weathering were examined for two sealant formulations, Sealants A and C. Both static and dynamic strain was applied to the sealants during the sum­mer in a Gaithersburg, MD outdoor location. Both sealants exhibited a re­versible change in equilibrium distance. Stress relaxation studies on all samples revealed that, for Sealant A, two mechanisms affected modulus change; exposure without applied strain increased the modulus while addi­tionally applied strain decreased the modulus. Only one mechanism that decreased the modulus was found for Sealant C. A 7 % dynamic strain and a 25 % static strain were observed to produce equivalent modulus changes in both systems.

KEYWORDS: sealant, strain, modulus, outdoor weathering,

characterization

Introduction

Sealants are expected to seal gaps between dissimilar materials in the building envelop that change dimension with changes in either temperature (differential thermal expansion coefficients) or humidity (wood frame construction). The ability of a sealant to maintain its physical integrity, thus maintaining a seal as the gap dimensions change, is a critical function of the sealant. The time de­pendent response to applied strain is critical in the in-service environment and yet is often neglected in the test methods employed to evaluate the sealant.

The current evaluation methods for sealants are either threshold-based or based on long-term exposure outdoors. The most typical example of a threshold-based test is the ASTM Standard Test Method for Adhesion and Co­hesion of Elastomeric Joint Sealants under Cyclic Movement (Hockman Cycle) (ASTM C719). In this test, the sealant is exposed to a series of environmental

stresses, such as exposure to elevated temperature, immersion in water, and cyclic mechanical testing, each conducted in sequence. At the end of the serial exposures, the sealants are evaluated visually for failure in adhesion or cohe­sion. ASTM C719 based testing has been effective in creating a minimal – performance threshold; however, it does not have the ability to predict in­service performance or differentiate between products that perform above this minimal threshold.

Typical commercial outdoor exposures are performed with no strain on the sealant. If strain is included in the outdoor exposure experiment, a static strain is developed using gage blocks to strain the sealant to a set tension level or clips for a set compression value. These strains do not reflect the dynamic strain that occurs in the in-service environment on the sealant.

There are several examples of outdoor weathering devices that impose strain on the sealant that are constructed of two materials with dissimilar co­efficients of thermal expansion. Upon thermal change, these devices impose a strain on the sealant. Examples include unplasticized polyvinyl chloride (PVC) and steel [1], wood and aluminum [2], concrete and aluminum [3], and steel and aluminum [4] devices. Manually operated devices have also been used to create cycling effects [5]. Results from these devices have indicated that joint movement is a predominate factor in sealant failure [6]. From these limited results, it is clear that imposed strain on the sealant is a critical part of the exposure environment.

These studies illustrate that applied strain strongly affects the response of the sealant [7]. What is not clear is what the expected response of the sealant would be under long-term exposure to a dynamically or static applied strain. Additionally, a modulus characterization of the sealant was used to track the changes in the sealant. Changes in the time dependent modulus can be specifi­cally attributed to molecular level mechanisms [8].

In previous studies, the sealant was evaluated both before and after expo­sure by visual inspection [6]. While useful for determining cracking and mor­phological changes, such evaluation does not anticipate the fundamental changes of the sealant that occur at the molecular level.

An additional complication is the nonlinear viscoelastic nature of the seal­ant. The sealant responds to deformation by dissipating (viscous) or storing (elastic) the resulting stress that is imparted to the sealant during the applied strain. The viscoelastic response is both time – and strain-dependent. To sepa­rate this time and strain dependence and to characterize the viscoelastic nature of the sealant, stress relaxation experiments were performed to evaluate the sealant both before and after exposure.

Experimental