Interaction of Environmental Variables

Thus far, the effect of individual target factors on the decrease of the modulus has been examined over various robustness factor combinations. In investigat­ing the interaction of robustness factors, a comparison among external block – to-block differences is performed. From the plot of average modulus ratios targeting RH [Fig. 10(b)], the effect of RH in decreasing the modulus ratio is seen for exposures without UV, and vice versa. For instance, high RH levels always reside at the bottom of each bar for 50°C/25 % movement/UV, whereas low RH levels are located at the bottom of each bar for 50°C/25 % movement/no UV. Furthermore, a close examination of the results for 50°C over all levels of cyclic movement and RH reveals that with UV exposure, the modulus decrease depends on the level of cyclic movement, but the humidity effect is small. With­out UV, humidity has a greater effect than cyclic movement. Indeed, high RH levels are found at the bottom of each bar in all six settings involving combina­tions of temperature and movement without UV. Ignoring the bars showing a small local RH level effect, low RH levels with the presence of UV are located at the bottom of each bar in five of six different settings involving different temper­atures and movement with UV. These results suggest that the interaction between UV and RH is significant in that UV seems to overwhelm the effect of moisture.

Copyright by ASTM Int’l (all rights reserved); Tue May 6 12:07:08 EDT 2014

Downloaded/printed by

Rochester Institute Of Technology pursuant to License Agreement. No further reproductions authorized.

The negligible effect of humidity on modulus decrease for specimens exposed to UV prompts an examination of the interactions between tempera­ture and motion. The block plots in Figs. 8 and 11 reinforce the two conclusions drawn from Fig. 6. First, the individual effect of temperature or cyclic move­ment is significant, as shown by the progressive decrease in modulus with increasing magnitudes of cyclic movement or temperature. Second, the combi­nation of temperature and cyclic movement produces even greater effects. In particular, a modulus decrease of >75 % is observed for 50°C/25 % movement. This also can be seen in Table 2, which is simply a rearrangement of the data that groups temperature and cyclic movement on the vertical axis and RH level and UV on the horizontal axis. The combinations of elevated temperatures and 25 % movement always are located at the top left corner of the table, showing the deleterious effect of combining UV radiation, cyclic movement, and elevated temperature. Indeed, this combination of environmental variables is the worst setting for modulus decrease. This is consistent with a prior study on two seal­ants with different formulations, which showed that sealant joints were able to resist the individual influence of cyclic movement, high temperature, or RH but degraded substantially when exposed to a combination of cyclic movement and high temperature or RH, or a combination of these three factors [1].

The simultaneous application of UV and cyclic movement has a pro­nounced effect on crack formation in sealants. As shown in Fig. 13, small, deep cracks are visible on the surface of specimens exposed to 50°C/75 % RH/UV/25 % cyclic movement, but fewer shallow cracks ae observed in the specimens under the same conditions but without cyclic movement. These observations imply that exposure without cyclic movement results in only surface degrada­tion, and that simultaneous cyclic movement during weathering accelerates the bulk deterioration of sealants. Notably, specimens under similar conditions without UV did not exhibit any cracking. This indicates the significant deleteri­ous impact of UV exposure and cyclic movement on the aging of this particular

TABLE 2—Data rearrangement grouping temperature and cyclic movement on the verti­cal axis and RH level and UV on the horizontal axis.

Temperature (3)

and Movement (4) Relative Humidity (4) and UV (2)

FIG. 13—Appearance of representative test specimens exposed to the combination of 50°C, 75 % RH, and UV radiation (a) with 25 % cyclic movement and (b) without cyclic movement. Exposure without cyclic movement results in only surface degrada­tion, whereas simultaneous cyclic movement during weathering accelerates the bulk deterioration of sealants.

sealant. Cyclic tensile stress might promote the penetration of UV into the seal­ants and enable microcrack initiation in the degraded polymer surface layer to propagate into the bulk sealants. In contrast, cyclic compressive loads might lead to the alignment of molecular segments and the packing of molecules such that the ease of UV penetration into the sealant is reduced and sealant degrada­tion is minimized. This is a subject for further study.

It is recognized that the relative effects of the various environmental factors on the properties of the sealant can differ with the type of base polymer in the sealant, other components included in the formulation, and impurities and reaction products. Each ingredient reacts to the various environmental factors in different ways, so the relative effects of stresses will vary from one sealant formulation to another. The present study presents findings based on tests with a single sealant; however, the experimental protocol could be applicable to broad classes of materials.

Conclusions

The accurate prediction of in-service performance in less time than is required for field tests and tests on structures has been hindered by a poor understanding of the failure modes in the two environments, a lack of methods for accurately quantifying the effects of environmental degradation factors, and crude techni­ques for monitoring sealant degradation. In this study, a reliability-based approach was implemented in order to systematically assess the individual and synergistic impacts of four major environmental factors (temperature, cyclic movement, UV exposure, and RH) on a sealant system. This methodology uti­lized laboratory exposure devices that allowed accurate control and monitoring of these environmental factors, as well as a quantitative measurement proce­dure for characterizing sealant degradation. Changes in the modulus were used as an indicator of the effects of environmental factors because a decrease in the modulus was found to be a precursor to cracking and debonding in our sealant system, which allow moisture penetration (the usual definition of failure).

Copyright by ASTM Int’l (all rights reserved); Tue May 6 12:07:08 EDT 2014

Downloaded/printed by

Rochester Institute Of Technology pursuant to License Agreement. No further reproductions authorized.

Because each material responds differently to environmental factors, the modu­lus should not be viewed as a universal performance indicator for all sealants. The main conclusions are summarized as follows:

1. Elevated temperature is statistically significant for decreasing modulus irrespective of the robustness factor settings (i. e., cyclic movement, RH, and UV radiation).

2. Cyclic movement at 25 % strain is statistically significant for decreasing modulus irrespective of the robustness factor settings (i. e., temperature, RH, and UV radiation). However, unlike with 25 % strain, cyclic move­ment at 8 % and 15 % strain does not have any impact on the modulus. This observation suggests that there might be a threshold value of cyclic strain below which the effect of strain is negligible.

3. UV is statistically significant for decreasing modulus irrespective of the robustness factor settings (i. e., cyclic movement, RH, and temperature).

4. RH seems less important for decreasing modulus irrespective of the robustness factor settings (i. e., cyclic movement, temperature, and UV radiation).

5. UV radiation suppresses the effect of RH on modulus decrease. In the absence of UV, RH has a greater effect than cyclic movement.

6. Among various environmental factors, cyclic movement at 25 % strain is the most important factor for modulus decrease, followed by UV radi­ation, temperature, and RH.

7. The combination of cyclic movement and temperature produces a syn­ergistic effect leading to a larger reduction in modulus than that seen with either variable alone.

Acknowledgments

The support from a NIST/industry consortium on Service Life Prediction of Sealant

Materials is greatly appreciated. Participating companies include DAP, Degussa,

Dow Corning, Kaneka Texas, SIKA, Solvay, Tremco, and Wacker Silicones.