Francois De Buyl

Silicone sealants and structural adhesives

Abstract The mode of preparation of silicones, their use and typical performances are reviewed. Particular attention is given to their use as sealants and structural adhesives in construction and building applications where adherence properties for assembling two substrates are important. An important part of this paper has been previously published ten years ago in French (Colas, Chimie Nouvelle 8(30) (1990) 847–852), however the purpose with this paper is to bring some more up-to-date informations on the above subject.

Understanding Hydrolysis and Condensation Kinetics of γ-Glycidoxypropyltrimethoxysilane

Monitoring the kinetics of hydrolysis and condensation of γ-glycidoxypropyltrimethoxy-silane (γ-GPS) was carried out by NMR spectroscopy (29Si–, 13C–, and 1H–). The course of these reactions was followed in 2 wt% aqueous dilution conditions (26% D2O/74% H2O), pH 5.4, and temperatures of 26, 50, and 70°C. At ambient temperature, hydrolysis and condensation proceed at very different time scales: a few hours for the hydrolysis versus several weeks for the condensation. Distortionless Enhancement by Polarization Transfer (DEPT) sequences by 29Si– and 13C–NMR spectroscopy were optimized for determining the complete spectral assignment for each hydrolysis step, i.e., RSi(OMe)3–n(OH)n (with R = (C H 2 OCH)C H 2 OC H 2 C H 2 C H 2–;and n = 1, 2, 3). A pseudo-first order rate constant for the first hydrolysis step, T0(OMe)3 + H 2 O → T0(OMe)2OH + MeOH, was calculated to be 0.026 min−1. Simultaneously to the condensation reactions, we have observed epoxy ring opening of the glycidyl- group. All three processes (hydrolysis, condensation, and epoxy ring opening) are dramatically accelerated with temperature increases from 26 to 70°C. The activation energy of the epoxy ring opening leading to the formation of a diol structure at the extremity of the glycidoxypropyl- chain was estimated to be 68.4 kJ/mol.

A generalized cure model for one-part room temperature vulcanizing sealants and adhesives

The cure progression of one-part room temperature vulcanizing (1-RTV) sealants and adhesives is a vital intrinsic material property for their typical end-use applications. Understanding the key factors controlling the cure mechanism becomes more and more important given an ever-increasing demand to improve productivity by allowing assemblies using these materials to move faster through the manufacturing process. For 1-RTV systems, ultimate mechanical and adhesive strengths attain an equilibrium value as their cure proceeds in the presence of atmospheric moisture. A diffusion-controlled, moisture-cure mechanism was established for this family of materials about 10 years ago, most definitively by Comyn et al. The depth z of unidirectional cure with time t was derived to be a function of the vapor pressure p of water in the atmosphere in the proximity of the material surface, and two material properties: the permeability P of moisture through the cured layer and the equivalent material volume V reacting with 1mol of water, and cure time, t, i.e. z= (2VPp t). Hence, the thickness of the cured layer should increase as t. That is, a plot of z vs. t should yield a straight line where, for a particular RTV material, the magnitude of its slope is a function of the temperature and relative humidity (RH) during cure. This was shown for organic hot-melt adhesives and silicone sealants; validation of the moisture-cure model was demonstrated using gravimetrically based measurements to determine V and P separately. However, a deviation from the moisture-cure model was reported from a study using an alkoxy silicone sealant in geometries requiring deep-section cure and, therefore, longer cure times. An increase in the slope of the z vs. t plot was observed after a certain time! interval that was a function of cure conditions. The change in cure rate was attributed to the migration of low molar mass crosslinkers and adhesion promoters towards the cure front and the eventual depletion of these species, which compete with the silicone polymer for reacting with the diffusing water. It was possible to determine the diffusion coefficient, D, of additives that act as adhesion promoters and crosslinkers in uncured alkoxy silicone sealant. From these studies, the utility of the moisture-cure diffusion model was demonstrated. The variables impacting cure performance are reviewed; and, a generalized form of the model is presented to account for variability in cure rates observed over extended time intervals in assembly geometries requiring deepsection cure. By plotting z as a function of (tp), data-sets for each material collected at different conditions of temperature and RH merged into a single straight line with a very high coefficient of determination. For end-users of 1-RTV silicone sealants and adhesives, this latter approach provides a material-specific reference line that demonstrates the correspondence between cure time and vapor pressure; i.e. a certain cured depth can be reached at short cure times with a high RH or at long times with a low RH.

Kinetics of cure, cross link density and adhesion of water-reactive alkoxysilicone sealants

When a silicone sealant is exposed to wet air, a surface skin is formed, which becomes thicker with time. For a 1-part room temperature vulcanization (1-RTV), neutral cure system based on the hydrolysis and condensation reactions of alkoxysilanes, it has been observed that, at constant temperature and humidity, the thickness of the cured layer is initially proportional to the square root of time, but later the gradient of such a plot increases. There are thus two regions of cure, an outer region and an inner one. Swelling in toluene, which has been measured for samples taken at various depths, shows that the cross link density is greater in the outer regions. Single lap joints of aluminium or glass have been cured in these two regions and joints from the outer region are always stronger than the corresponding ones from the inner region. This behaviour has been ascribed to the mobility of low molar mass cross linking and coupling agents. During the initial stages of cure they migrate into the outer region, but once cure has passed into the inner region there is now a paucity of these compounds. Youngs modulus and the concentration profile of silane crosslinking agent have also been measured. All observations confirm migration of crosslinking species, and gradient formation for most physical properties in the final joint. Measurement of the water vapour permeability coefficient (P) of the cured siloxane network has shown that P is independent of humidity and stretching of the material. A plot of log P against reciprocal absolute temperature is non-linear with positive slope, showing that the heat of permeation is negative and varies with temperature. This can be explained by the formation of water-clusters. When the network is formed water is absorbed and alcohol is released.