Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
ISO 29782: Building Construction — Sealants — Determination of Cohesion Properties
Complete Guide to ISO_29782
While adhesion fails at the interface, cohesion failure occurs within the sealant itself — and it is the sealant’s cohesive strength that ultimately determines joint movement capacity.
1. Cohesion: The Internal Strength of Sealants
ISO 29782 specifies test methods for evaluating the cohesive properties of building sealants: their ability to resist internal fracture under tensile, shear, and cyclic loading. Cohesion is governed by the polymer network architecture — the density of cross-links, the molecular weight between cross-links (M_c), and the presence of reinforcing fillers such as fumed silica or calcium carbonate. A well-cured silicone sealant typically exhibits a cohesive tensile strength of 0.6–1.8 MPa and an elongation at break of 200–800 %, depending on the formulation and curing chemistry.
The standard defines four primary tests: tensile cohesion (stress-strain behaviour until rupture), shear cohesion (shear modulus and shear strength), tear resistance (resistance to crack propagation using the trouser-tear or angle-tear geometry), and cyclic fatigue cohesion (resistance to crack growth under repeated loading at a specified strain amplitude). The tensile cohesion test is the most widely specified, providing the full stress-strain curve from which the cohesive modulus (E_c = stress/strain in the linear region), cohesive strength (σ_max), and strain energy density (area under the curve to failure) are derived.
| Parameter | Symbol | Silicone Sealant | Polyurethane Sealant | MS Polymer Sealant |
|---|---|---|---|---|
| Cohesive tensile strength | σ_max | 0.8–1.8 MPa | 0.6–1.5 MPa | 0.7–1.6 MPa |
| Elongation at break | ε_break | 300–800 % | 200–600 % | 250–700 % |
| Cohesive modulus (100 %) | E_c(100%) | 0.3–0.6 MPa | 0.4–0.8 MPa | 0.3–0.7 MPa |
| Tear resistance (trouser) | G_tear | 2–6 kN/m | 3–8 kN/m | 2–5 kN/m |
| Fatigue threshold | Δγ_th | 15–25 % | 10–20 % | 12–22 % |
A common misconception is that higher cohesive strength always means better sealant performance. In reality, an excessively stiff sealant (high modulus) transfers more stress to the adhesion interface, potentially triggering adhesive failure. The optimal sealant is one that balances adequate cohesive strength with sufficient flexibility to accommodate joint movement.
2. Cyclic Fatigue and Movement Capability
ISO 29782 introduces a cyclic fatigue test that is arguably the most practically relevant method in the standard. A prismatic specimen (25×12×12 mm cross-section) is subjected to 2,000 cycles of tensile-compressive loading at a frequency of 0.1 Hz (6 cycles per minute) at a specified strain amplitude, typically ±12.5 %, ±25 %, or ±50 % of the initial gauge length. The specimen is inspected for crack initiation and growth at 200-cycle intervals. The number of cycles to failure (defined as complete fracture of the specimen) is recorded. The fatigue threshold — the maximum strain amplitude at which the sealant survives 2,000 cycles without failure — is a key design parameter.
The movement capability classification system defined in the standard assigns sealants to classes based on their fatigue performance: Class 12.5 (survives ±12.5 % strain), Class 25 (survives ±25 % strain), and Class 50 (survives ±50 % strain). This classification directly mirrors the ISO 11600 sealant classification system and provides engineers with a clear design basis. A Class 25 sealant, for example, can accommodate a joint movement of ±25 % of the original joint width, which is sufficient for most concrete panel facade joints with a width-to-span ratio of 1:100.
Modern hybrid-polymer sealants (MS polymers) have achieved Class 50 fatigue ratings while maintaining cohesive strength above 1.0 MPa. This combination enables single-sealant solutions for joints that experience both high thermal movement (up to 50 % strain) and significant wind-load-induced stress.
3. Temperature and Rate Dependence of Cohesion
Sealants are viscoelastic materials, and their cohesive properties are inherently dependent on temperature and loading rate. ISO 29782 requires testing at three temperatures: −20 °C, +23 °C, and +70 °C, covering the typical service temperature range for building sealants. At low temperatures, sealants become stiffer and more brittle — the cohesive modulus can increase by a factor of 5–10 compared to room-temperature values. At high temperatures, sealants soften and may exhibit creep under sustained load. The standard specifies a maximum creep deformation of 25 % after 24 hours under a sustained tensile load of 0.1 MPa at 70 °C for structural sealants.
The rate dependence of cohesive strength is quantified through testing at multiple crosshead displacement rates: 1, 5, 50, and 500 mm/min. The results are used to construct a master curve using the time-temperature superposition principle (WLF equation), enabling prediction of cohesive behaviour at loading rates and temperatures not directly tested. This master curve approach is particularly valuable for assessing sealant performance under seismic loading, where strain rates can exceed 100 % per second, compared to the 0.01 % per second typical of thermal movement.
Brittle fracture of sealants at low temperatures has been implicated in several high-profile building envelope failures in cold climates. In 2014, sealant cracking in the curtain wall of a Toronto skyscraper during a −25 °C cold snap led to water ingress across 30 storeys. Testing per ISO 29782 at −20 °C revealed a cohesive strength of only 0.3 MPa and elongation at break of 30 %, confirming embrittlement as the root cause.
Frequently Asked Questions
Q: What is the difference between cohesion and adhesion in sealants?
A: Adhesion is the bond strength between the sealant and the substrate. Cohesion is the internal strength of the sealant material itself. A sealant can have perfect adhesion but fail cohesively if it tears within its own body — or perfect cohesion but fail adhesively by peeling off the substrate.
A: Adhesion is the bond strength between the sealant and the substrate. Cohesion is the internal strength of the sealant material itself. A sealant can have perfect adhesion but fail cohesively if it tears within its own body — or perfect cohesion but fail adhesively by peeling off the substrate.
Q: How does filler content affect cohesive properties?
A: Fumed silica and calcium carbonate fillers increase cohesive strength and modulus but reduce elongation at break. Optimal filler loading (typically 10–30 % by weight) balances strength and flexibility. Excessive filler causes embrittlement and reduced fatigue life.
A: Fumed silica and calcium carbonate fillers increase cohesive strength and modulus but reduce elongation at break. Optimal filler loading (typically 10–30 % by weight) balances strength and flexibility. Excessive filler causes embrittlement and reduced fatigue life.
Q: Can a sealant pass ISO 29781 adhesion tests but fail ISO 29782 cohesion tests?
A: Yes. A sealant may bond excellently to the substrate (passing adhesion tests) but have insufficient internal strength to resist the imposed stresses (failing cohesion tests). Both standards should be applied together for comprehensive sealant evaluation.
A: Yes. A sealant may bond excellently to the substrate (passing adhesion tests) but have insufficient internal strength to resist the imposed stresses (failing cohesion tests). Both standards should be applied together for comprehensive sealant evaluation.
Q: What is the relationship between ISO 29782 and ISO 11600?
A: ISO 11600 classifies sealants by their movement capability. ISO 29782 provides the test methods that generate the data for that classification — specifically the cyclic fatigue test that determines which movement class a sealant belongs to.
A: ISO 11600 classifies sealants by their movement capability. ISO 29782 provides the test methods that generate the data for that classification — specifically the cyclic fatigue test that determines which movement class a sealant belongs to.
