Experiments to Establish Adhesive Strength and Load-Bearing Capacity

Series of Tests

Numerous parameters have an influence on the load-bearing capacity and serv­iceability of an adhesive joint. First of all, the adhesive itself with its material properties and their complex dependence on external factors, such as tempera­ture, and the materials to be joined together, as well as the geometry and thick­ness of the layer of adhesive, and the loads on it. Verifying the durability is also important, which must take into account the degradation of the adhesive and the adhesive bond as a result of various environmental effects or adjacent materials.

Therefore, once the material parameters of the UV – and light-curing acry­late adhesive AC4 selected had been determined with material specimens, fur­ther tests on small-scale in situ specimens and sample components were necessary. Load-carrying tests on specimen components enable a realistic assessment of the load-bearing behavior and serve to back up a static analysis. The test setup and the results of the component tests, which were called for

TABLE 2—Glass transition temperatures according to ASTM D 4065.


Glass transition temperature (°C)













within the scope of the approval procedure for the glass corridor, are summar­ized below.

In addition, shear strength values had to be obtained from small-scale specimens (glass-glass adhesive joints in single shear). Compression shear tests were carried out at various temperatures between —25 and + 75°C, as well as fol­lowing accelerated aging.

Methods and Test Setup

Block Shear Test—The shear strengths were determined using glass-glass adhesive joints in single shear at temperatures of —25, 0, + 25, + 50, and + 75°C in a block shear test (Fig. 10), according to the works standard of the adhesive manufacturer DELO-Norm 5 [20,21]. Besides taking into account different tem­peratures, some of the test specimens were subjected to accelerated aging and subsequently tested at room temperature following conditioning for 24 h. The parts joined by the adhesive were made from annealed glass, but chemically toughened glass for the tests at —25, 0, and + 25°C. Toughened glass became essential to avoid glass failure during the tests at lower temperatures. A prelimi­nary study has proven that the bonding strength was not affected by using the two different glass types. The glass components measured 20 x 20 mm and were 5 mm thick. The adhesive joint was set to a thickness of 1.5 mm, with a 5-mm overlap between the two parts. Loading in a universal testing machine was car­ried out with displacement control at a rate of 10 mm/min until failure of the specimen. The load at failure was recorded. Ten specimens were tested per parameter.

Downloaded/printed by

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

Climatic cycle tests plus immersion in water and cleaning agent (commer­cial dish liquid W5 Power Blue, basic) have proved to be relevant aging scenar­ios for acrylate adhesive joints [22]. The specimens were, thereafter, stored for 21 days in a climate chamber with cyclic variation of humidity and temperature. The conditions at the start of the test were + 40°C and 95 % relative humidity. After maintaining these conditions for 15 h, the temperature was lowered to —20°C and kept constant at this level for 2 h. Afterward, the climatic conditions were changed to + 80°C and 50 % relative humidity. A weathering cycle lasted a total of 24 h, including the ramp up/down times, which means that within the period of the test, the cycle was repeated 21 times. Other test specimens were stored for21 days at + 45°Cina0.1 % detergent solution or in pure water.

Loading Test—Both the dimensions and the glazing of the specimens— frame corners with short legs 0.75 m long—corresponded with those of the orig­inal component. The laminated safety glass of the elements consists of four plies of fully tempered glass each 10 mm thick. The two outer layers of PVB are 1.90 mm thick, the inner layer 1.52 mm. The depth of the cross section was 300 mm throughout. A total of ten non-aged sample components were tested at room temperature. Five tests were carried out with a transparent plastic setting block, but the setting block was omitted from the other five to study the influence of the use and positioning of the setting block in more detail. The transparent plas­tic setting block carries the vertical load in the event of failure of the adhesive to meet the requirements of a fail-safe concept. Additionally, the block prevents creeping of the bonded joint under long-term loading, which otherwise may lead to large deformations and unwanted glass-glass contact.

Each sample component was clamped in a test rig and loaded by a hydraulic cylinder at the outermost end of the cantilever (Fig. 11(a)). The base of the sample components was cast into the frame using an injection mortar to reduce slip to a minimum. The load was applied incrementally up until failure or until a maxi­mum load, governed by the test setup, of about 95 kN was reached (Fig. 11(b)). The load increment was 10 kN up to a total load of 40 kN, thereafter, the

FIG. 11—(a) Loading test setup (schematic), and (b) load and deflection in component testing frameSC7.

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.

increments were reduced to 5 kN. Every loading increment was held for 2 min. A numerical simulation to check the loading level in the test revealed that the char­acteristic strengths of toughened safety glass were far exceeded at maximum load. If the joint does not fail, then the glass, therefore, becomes relevant to the design.

The deflection at the end of the cantilever essentially depends on the rota­tion at the adhesive joint. The deformation of the glass itself; on the other hand, is small. The vertical displacement of the frame beam was, therefore, measured by transducers at two points along its top edge. The vertical deflection at the end of the cantilever was obtained from the two measurements by linear extrap­olation. A specific fixture was developed to support the displacement trans­ducers. The device was clamped onto the specimen directly above the restraint. The transducers measured only the relative deformation between the post and the cantilever beam. Hence, potential slip at the fixing of the specimen compo­nent did not impair the results. The stresses in the glass were recorded with strain gauges (Fig. 12). The critical points for this were determined in a numeri­cal model. A photoelastic analysis was carried out during the tests on the com­ponents to obtain a qualitative statement regarding the change in the stress distribution within the corner zone.

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.