Structural Sealant Glazing Testing and Results for Seismic Work

Through the course of the project, data generated from ASTM C1135 [12] Standard Test Method for Determining Tensile Adhesion Properties of Struc­tural Sealants was used to help develop the finite-element model and predict

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. 4—Typical details of horizontal and vertical stack joints to be used on the building (SI unit conversion: 1 in. — 25.4 mm).

sealant behavior. Also the environmental conditions found in ASTM C1184 [13], The Standard Specification for Silicone Sealants, were applied and tested to demonstrate sealant durability.

The results of this testing generated stress-strain curves for the two sealants being utilized for the project, Dow Corning 983 Silicone Glazing Sealant (DC 983 SGS) and Dow Corning 995 Structural Silicone Sealant (DC 995). The two – part DC 983 SGS sealant is used for the in-shop glazing of the curtain-wall pan­els because of its quick setup time, allowing panels to be fabricated and subse­quently moved to the job site after 24 h. Any field glazing or re-glazing required would be completed with the one-part DC 995 sealant, as it comes in packaging that is usable in the field, while still maintaining the high sealant strength neces­sary for the application.

A summary of the sealant properties for DC 983 SGS sealants and DC 995 sealants are shown in Figs. 5 and 6, respectively. Both of these graphs include the sealant properties when tested in tension per ASTM C1135 [12] criteria and sealant properties when tested in shear per AC 45 testing criteria [14]. The dis­placement (strain) rate for both tension and shear testing was 0.5 in. (12.7 mm) per minute according to ASTM C1184 [13]. In both tension and shear coupon

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.

DOW 983SGS — Tension (ASTM C1135) & Shear (AC45) Summary
(GRAPH #1)

5:1 Safety Factor For Tension Loading Allowable Sealant Stress = 28 psi (193 kPa)

70 80 90 100 110 120 130 140 150 160 170 180 190 200 210 220 230 240

Strain (%)

88‘C ♦ 3hr cooling Tension — • — • -29* C Tension 5000hr QlJV Tension

88’C * 3hr cooling Shear

FIG. 5—Dow Corning 983 stress—strain curves.

tests, the failure was cohesive. Testing for both tension and shear were per­formed for a number of different environmental conditions. These graphs indi­cate a couple of key components of SSG sealant behavior performance. The first item that becomes apparent is that SSG sealants are much stiffer in tension than in shear. Based on the test results, the sealants are more than 4 times stiffer in tension than in shear. This is consistent with what is documented in ASTM C1401 [1] and supports the argument that it is conservative and appro­priate to design SSG sealants based on the tested tension properties. A higher tension modulus will generate higher stresses in the sealants at a given strain. As an example, at approximately 10% strain, a bead of DC 983 SGS sealant would experience a 28-psi (193-kPa) stress if loaded in tension, but only about 10-psi (69-kPa) stress if loaded in shear. The second issue that the graphs illus­trate is that whereas the overall ultimate stresses and strains the sealant can support are affected by different environmental conditions, there is much less variation in sealant behavior at allowable stress levels based on a 5:1 safety fac­tor between those same variety of environmental conditions. Whereas ASTM C1401 documents the use of a 2.5:1 safety factor when originally establishing the 20-psi (138-kPa) allowable sealant stress in the 1970s and 1980s, OSHPD has requested a 5:1 safety factor for this project.

Although the SSG sealant bead will experience a combination of tension and shear behavior when subjected to seismic in-plane racking, to model seal­ants in finite-element analysis, the most conservative tensile sealant properties

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. 6—Dow Corning 995 stress-strain curves.

were used. For DC 995 sealant, the most conservative results were obtained from the sample exposed to 5000 h of UV radiation, which is a simulation of accelerated aging, including the following cycles: 8 h UV at 60° C and 4 h con­densation at 50°C. The sealant test data shown in Fig. 6 also includes shear test­ing data for comparison purposes and documents that the use of tension properties is conservative. For DC 983 SGS sealant, the most conservative results were obtained from a sealant sample that was cyclically tested in tension from 0 to 50 psi (0-345 kPa), and then pulled to failure, showing a 140-psi (965-kPa) ultimate strength (Fig. 5) after cyclic testing. The average across five samples cyclically tested was 149 psi (1027 kPa). The same cyclic testing was performed with DC 995 sealant, resulting in an ultimate strength average between five samples of 185 psi (1276 kPa). To be very conservative, it was chosen to use the 5000-h UV results at 145 psi (1000 kPa) instead of the cyclic testing data for DC 995 sealant.

Another aspect of structural sealant that needs to be determined for seismic analysis is deformability that is defined [2] as the ratio of the ultimate deforma­tion to the limit deformation. The ultimate deformation is defined as “The defor­mation at which failure occurs and that shall be deemed to occur if sustainable load reduces to 80% or less of the maximum strength." The limit deformation is defined as “Two times the initial deformation that occurs at a load equal to 40% of the maximum strength." For DC 983 SGS sealant, the ultimate and limit deformations were obtained as 96% and 42%, respectively, giving the ratio of 96/42 = 2.3. This ratio falls between 1.5 and 3.5 that according to ASCE 7-05 is

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.

defined as limited deformability material. The ratio for DC 995 sealant referenc­ing the 5000 h QUV curve is 1.9. Both these ratios define the sealants as limited deformability and per Table 13.5-1 [2]. Component amplification factor, ap =

2.5 and component response modification factor, Rp = 2.5.

According to ASTM C719 [15], DC 983 SGS 2-part sealant is rated for 25% movement, whereas DC 995 SSS 1-part sealant is rated for 50% movement as allowable (serviceability) sealant strains for wind and thermal loading. ASTM 1401 [1] sets the allowable stress because of wind and thermal movement as 20 psi (138 kPa). This allowable stress is based on a SSG sealant with an ultimate strength of no less than 50 psi (345 kPa) or a 2.5:1 factor of safety against fail­ure. Based on the sealant test results shown in Fig. 5 for DC 983 SGS sealant, this represents a safety factor of 140 psi (965 kPa)/20 psi (138 kPa) = 7.0:1. Sim­ilarly, based on the sealant test data in Fig. 6 for DC 995 sealant, this represents a safety factor of 146 psi (1007 kPa)/20 psi (138 kPa) = 7.3:1. Therefore, both sealant types meet the requirements of C1401 for using a 20 psi (138 kPa) allow­able stress for wind and thermal loading. Based on consultation with Dow Corn­ing, a proposed allowable sealant stress for seismic design is 50 psi (345 kPa). However, the building permit officials (OSHPD) requested a safety factor of 5:1 be adhered to for this specific project. Based on available sealant test results by Dow Corning, the most conservative condition is the cyclic tension testing of 983 SGS sealant at a strain rate of 0.5 in. (12.7 mm) per minute with an ultimate tensile strength of 140 psi (965 kPa), which results in an allowable stress of 28 psi (193 kPa) for seismic loading for a factor of safety of 5:1. Dow Corning

2- part DC 983 SGS sealant was chosen as the preferred sealant for shop glazing, whereas Dow Corning 1-part DC 995 sealant was chosen for field glazing. Typi­cal properties for these sealant types are listed in Table 1.

Based on the curves generated in Figs. 5 and 6 the Young’s modulus could be calculated at different ranges of sealant stress. The results are listed in Table

2. The modulus can change slightly depending on the stresses between which the slope of the curve is arrived at, but upon examination of Figs. 5 and 6, it can be seen that both sealants behave very linearly from 0 to 30 psi (207 kPa). Although 50 psi (345 kPa) is an industry-accepted level of sealant stress for seis­mic designs, this project limited the stress to be at a minimum 5:1 safety factor as requested by OSHPD. With an ultimate sealant strength of 140 psi (965 kPa), conservatively, the resultant design stress level is then 28 psi (193 kPa).

TABLE 1—DC 983 SGS and DC 995 sealant properties.


2-part DC 983 SGS

1-part DC 995

Tensile adhesion modulus

300-800 psi

175-325 psi

(ASTM C1135)

(2069-5515 kPa)

(1207-2241 kPa)

Tensile strength (ASTM D412)

250 psi (1724 kPa)

350 psi (2413 kPa)

Movement capability (ASTM C719)


± 50%

Durometer (ASTM C661)

40-50 Shore A

40 Shore A

Shear adhesion modulus (AC 45)

75 psi (517 kPa)

50 psi (345 kPa)

TABLE 2—Dow Corning 983 SGS modulus and ultimate tensile properties.

Young’s modulus,

Ultimate strength,

psi (kPa)

psi (kPa)

21-day room temperature


165 (1138)

88° C elevated temperature

485 (3344)

146 (1007)

5000-h UV exposure

401 (2765)

170 (1172)

Additionally, seismic movement primarily results in shear stresses in the structural sealants. Based on tensile properties of uniform behavior, linear materials such as these two sealants can be used to develop equations and pre­dict shear behavior. During the course of this study, the same dimension tensile adhesion joints were also tested in shear. As discussed earlier per test results and also shown by Zarghamee et al. in 1996 [16], the shear modulus is approxi­mately 1/4 that in tension. The ultimate strengths in either mode are very simi­lar. For example, as shown in Figs. 5 and 6, at 28 psi (193 kPa) in tension, the strain on the sealant is approximately 10% in tension and close to 40% in shear. This is actually favorable for seismic testing in that for a given strain or dis­placement during a seismic load, the associated stress on the sealant is lower in a shear mode than in a tensile mode. Thereby, the primary data by which this project was designed for, being tensile data, is a conservative approach.

The other important sealant properties to account for when considering seismic design are durability of the sealant and consistency of the sealant strength and modulus properties over time, which can be seen in the 5000 h QUV exposure and high-temperature exposure conditions. Upon review of these results, it can be seen that extended weathering has very little effect on silicone sealants, which is critical to long-term reliable seismic performance. The shear results are summarized in Table 3.

Hot, cold, and long-term weathering exposure conditions were tested to ver­ify the consistency and durability of the silicone sealant properties. Although

TABLE 3—Summary shear data; 21-day room-temperature (RT) cure (+ environmental exposures).


Ultimate shear stress psi (kPa)

Ultimate shear strain (%)

Young’s modulus psi (kPa)

Dow Corning 995 RT

135 (931)


45 (310)

Dow Corning 995 RT + 1 h 88°C + 3 h RT dwell

140 (965)


47 (324)

Dow Corning 995 RT + 1 h -29° C

148 (1020)


50 (345)

Dow Corning 983 SGS RT

136 (938)


71 (490)

Dow Corning 983 SGS RT + 1 h 88°C + 3 h RT dwell

143 (986)


73 (503)

Dow Corning 983 SGS RT + 1 h -29°C



65 (448)

silicone sealant durability is well documented in the industry, a quick overview of weathered sealant properties under tensile testing is summarized in Table 4.

SSG sealant is designed to transfer wind, thermal, and seismic loads and/or deformations through the glazing to the curtain-wall framing. For seismic design of building curtain-wall systems, to minimize racking-induced sealant shear stresses, and keep them within the allowable shear capacity, a unitized glazing system was chosen for the project. Within the unitized system, the sili­cone sealant remains adhered and absorbs stresses from building movements and wind load. When the silicone is adhered to the aluminum frame, it is then a part of an operating joint. Joint properties are tested per ASTM C1135 [12] and the resulting modulus of the system joint (not sealant alone) is 300—500 psi (2069—3448 kPa). The ultimate strength of such a joint is between 140—165 psi (965—1138 kPa) in tension and 135—170 psi (931 — 1172 kPa) in shear.