Tensile Properties at Different Pulling Speeds and Temperature Conditions

Ten test specimens of the ASTM C1135 type were cured for 28 days by condi­tioning them at (23 ± 2)°C and (50 ± 5)% relative humidity (RH). They were then pulled in an extensometer at a rate of 5 mm/min (ISO/FDIS 28278-1) and 50 mm/min (ASTM C1135) at 23°C. Figure 3 shows cohesive failure mode of locally compounded one-part structural silicone sealant.

Figure 4 shows for each of the sealants tested some results obtained as an average over three tested specimens.

According to the ISO/FDIS 28278-1 standard, the proposed criteria for suc­cessfully passing the tensile testing is a characteristic breaking stress giving 75 % confidence that 95 % of the test results will be higher than this value of 0.5 MPa or larger at 23°C and a rupture pattern that shows equal to or more than 90 % cohesive failure. Additionally, the tensile values after exposure to harsh conditions should be measured, and its delta mean value (DXmean) as a ra­tio of the initial value shall be equal to or larger than 75 %. For definition of DXmean and the characteristic value Ru,5 see Eqs 1 and 2, respectively:

DXmean — [38]mean. c/Xmean. n (1)


Xmean — the average breaking stress, either under tension or shear,

Xmean. n — the average breaking stress, either under tension or shear in the initial state (23°C) and,

Xmean. c — the average breaking stress, either under tension or shear after conditioning or ageing.

FIG. 4—Initial tensile values prior to exposure to weathering at different pulling speeds.

where S is the standard deviation of the series under consideration.

The parameter saf is the eccentricity of 5 % with 75 % confidence and this statistical parameter is well elaborated in ISO 3207 [12]. Table 2 shows this pa­rameter as a function of the number of test pieces in a study.

For a country like Korea, which has four distinctive seasons, the class T1 requirements of the ISO/FDIS 28278-1 standard (shown in Table 3) can be applied for tensile strength at different temperatures.

Figure 5 displays the average tensile test results obtained at —20 and 80°C as a percentage of the initial average test result obtained at 23°C (DXmean, see Eq 1) for pull rate of 5 mm/min.

As can be seen, some products showed a larger mechanical property differ­ence at 80°C compared to the standard condition at 23°C. These results are linked to poor adhesion (prominent failure mode) of these specimens (Fig. 6). This is the case, for instance, for B-4 which is a one-part IG secondary sealant which displayed adhesive failure and only achieved a tensile strength of 0.2 MPa.

As shown in Fig. 4, with 5 mm/min extension rate called for in ISO/FDIS 28278-1, which has previously been noted as being a slower pulling rate than that of ASTM C1135 (50 mm/min), somewhat lower tensile values were observed than for the faster pulling speed, but the difference was insignificant. Some products, like B-4 and B-2, only showed tensile values around 0.2 and 0.3 MPa, which are far below the minimum requirement of 0.5 MPa defined in ISO/FDIS 28278-1.

TABLE 2—The variable sab as a function of the number of test pieces (see ISO 3207).

Number of pieces




















Test temperature, °C



AXmean > 75%


Ru,5 > 0.5 MPa


AXmean > 75%

TABLE 3—Tensile strength requirements as defined in ISO/FDIS 28278-1.

Although the initial tensile strength of C-1-a in standard conditions was lower, the property changes at low and high temperatures were reasonable. In case of B-3 and B-4, both sealants showed significant differences when tensile strength were measured at high temperature. There were failure mode changes from 100 % cohesive to some adhesive failure after heat storage conditioning.

Some products such as A-1, A-3, B-4, B-3, B-2, C-1-a, and C-3-b showed ad­hesive failure mode at 80°C, which means high temperature in a real application has a very critical impact on the long term adhesion durability of these struc­tural silicone sealants. Previous studies have documented that the most detri­mental condition for silicone durability is water immersion compared to UV radiation and heat (see, for instance, the paper by Bergstrom [7] and literature cited therein). But it is contrarily noted from this evaluation that for adhesion on float glass and anodized aluminum, high temperature exposure condition could be one of the critical factors to impact negatively on adhesion perform­ance (see Fig. 4). As can be seen, adhesive failures occurred on the glass side as well as the aluminum side.

Load perpendicular to the glass surface, such as wind loads, generate ten – sile/compression stresses in the structural seal. ISO/FDIS 28278-2 suggests that the maximum tensile stress is considered to develop at the center of the longest side of the pane (trapezoidal loading) and it can be calculated as shown in


h — structural sealant bite height, a — smallest edge of a rectangular pane, and F — wind load.

The selected bite height should ensure that the actual tensile stress acting on the structural sealant stays well below a certain strength, which is deter­mined from the characteristic ultimate limit state value of the structural sealant (Ru;5 in tensile loading) by applying a certain safety factor, as shown in

Rtensile, u,5

ftensile S (4)


In Eq 4 ytot is a safety factor which should be set by national rule. In the case of absence of national value, a ctot of 6 should be assumed. As ISO/FDIS 28278-2 states in one of its notes, “The use of high value for the partial factor ytot to­gether with inaccurate calculation model results in structural seal dimensions, which ensure an acceptable safety reliability level as demonstrated by experi­ence during the last twenty years." Lower ytot values are acceptable only when accurate calculation models are used together with an appropriate defined safety reliability level.

Therefore, the value of Ru,5 is an important number required for the calcu­lation of the structural bite for glazing as defined in ISO/FDIS 28278-2 standard using the current global industry consensus of limiting the design tensile strength of a structural sealant to a maximum of 139 kPa (20 psi) [1]. Figure 7 displays the safety factor calculated based on the Ru,5 values determined for the individual sealants. As can be seen, A-3, B-2, B-3, C-3-a, C-3-c, and C-1-b meet the safety factor of 6 requirement as suggested by the ISO/FDIS 28278-2 stand­ard or have a high possibility of meeting it.

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.

FIG. 7—Calculated safety factor for each sealant based individual on Ru5.

When comparing the strain-stress curves of the evaluated products, the dif­ferent behaviors of the products become apparent. For example, one-part struc­tural sealant from company C showed very high modulus compared with the products from other companies. In general, the products from company C showed only a small difference between the tensile values at the standard condi­tion and those at harsh temperature conditions. It is assumed that the differ­ence in behavior of the various sealant might be caused by differences in their compositions, for instance due to differences in sealant formulations such as different filler loading level, or the quality of the formulation ingredients, such as different polymer chain length or branching.