Peel loads

As described in Section II of this chapter, there are many types of peel tests available to characterize adhesives. These tests are important because peel stresses arise in the loading of many joint geometries, such as lap joints. Peel tests are severe because they constitute a test of the adhesive in its weakest stress mode. However, the peel test is a comparative test for adhesives and is dependent on many parameters. These parameters, such as peel speed, peel angle, bond thickness, and temperature, must be held constant to obtain valid results. The stress analysis of peeling is highly complicated because of these variable dependencies. 

A number of researchers have considered the analysis of the peeling test. Most analyses have involved the peeling of a flexible member from a rigid adherend. One of the first analyses was done by Bikerman. He assumed that both the flexible and rigid substrates behave as perfectly elastic 

The determination of stresses in the peel test is complex and warrants the use of finite element methods. Adams and Crocombe have applied finite element analysis to the peel test. They found that the principal tensile stress in the adhesive is responsible for propagating a crack through the bond, thereby causing failure. 

The uniqueness in the fracture mechanics approach is that it can be applied to a variety of flexible adhesive bond geometries which others have considered. 

In the following section, fracture mechanics methods will be discussed for adhesive bonds consisting of rigid adherends, since these are most common in structural adhesive applications. 

---

Adhesive type Shear load, MPa Peel load, N/m ... 

Adhesive type Slicin load, MPab Peel load. N/irf... 

The rate of peel loading is more important than in lap shear loading, and it should be known and controlled as closely as possible. The rate at which the load is applied is usually specified in the ASTM test procedure. Adhesive thickness also has a significant effect on peel strength values, as does the angle of peeling. 

Typical force versus time traces obtained from T-Peel tests are shown in Fig. 11. As can be seen, the peel load reaches a nearly constant value for both substrate materials, with some minor fluctuations superimposed on the results. These values have been used in an analytical model to calculate the adhesive facture energy, Gc [8]. For all the tests performed the crack propagated cohesively in the adhesive layer. The peel load was found to depend on the alloy type and on the thickness of the substrates, since most of the energy during the test is dissipated by plastic deformation of the arms. Numerical FV work is in progress. 

FIGURE 3.77 Test panel and T-type test (T-peel test) specimen for peel resistance of adhesives (standard test method ASTM D1876-95). The bent, unbonded ends of test specimen are clamped in test grips of tensile testing machine and load applied at a constant head speed of 254 mm (10 in.). Average peeling load (in pounds per inch of specimen width) is determined for the first 127 mm (5 in.) of peeling after the initial peak. 

As expected, the energy release rate J and lump-sum cohesive law can be experimentally determined if the crack tip separation 8, the loadhne rotation Op of the adherends, and the global peel load P are simultaneously recorded during the fracture test. It is noted that this interface constitutive relationship is the equivalent interface cohesive law, not necessarily the intrinsic cohesive law. This is because, in addition to the intrinsic cohesive separation, possible plastic deformation in the adhesive layer contributes to the entire normal separation between the two adherends during the fracture test. Of course, with the decrease of the adhesive thickness, it is expected that this equivalent interface cohesive law will finally approach the intrinsic cohesive law [66]. 

Joints can be weak when subjected to peel load... 

Adhesive layers of bonded joints should primarily be stressed in shear or compression. Tensile, cleavage and peel loads should be avoided, or their effect evaluated with great care. 

P(2) Special attention shall be paid to minimising the peel loads because of the brittle nature of lamination resins, see 5.3.1.3. 

Methods to reduce peel loads and peak stresses described in 5.3.i.3 should be applied when practical. 

Peel and cleavage loadings should always be avoided. When their presence can not be avoided, their effects should be minimised. Peel loads are mainly produced by out-of-plane loads acting on a thin adherend or by the eccentricity of the in-plane loads. When the adherend is thick and stiff, the out-of-plane loads typically produce cleavage loads. The critical areas in the joint with respect to peel and cleavage loadings are the ends of the overlap. 

This relates the test result (the measured peel load F) to a measure of intermolecular forces (the work of adhesion Wa or of cohesion Wc, see Wetting and spreading) and to the rheology via various terms such as energy dissipated in plastic deformation lApiast, in viscoelastic loss iAv/e, in bending Abend, and so on. Here, b is the width of the peeled strip. 

Fig. 4. Comparison of strain energy density at failure for four EVAs with the peel loads measured for backed strips of the polymers peeled from chlorite-oxidized copper ...

These suggest that the peel load at 90° should be twice that at 180°. This point is discussed below. 

Even the best of the latest generation of toughened adhesives can carry loads roughly 100 times greater in standard shear tests than they can in peel. While such tests are not directly comparable, they show that even small peeling loads are particularly destructive. 

In contrast, compression loads are readily borne. If these are compared, then the relative ability of a joint to withstand compression, shear and peel loading is of the order of 1000 100 1. 

Typically, bonded structures are designed so that the structural adhesive will be under shear loads most of the time. Adhesives are stronger in shear than they are in tensile or peel loading. Shear testing is very common because samples are easily fabricated and simple to test. 

The sample is gripped at its free ends and pulled in tension at a crosshead rate of 12 inches/min. At least one-half of the bonded area should be peeled to obtain an average peeling load and at least ten specimens should be tested to calculate the peel strength in pounds per inch of peel width. 

This section will not provide a detailed mathematical derivation of stresses in various joint geometries since this could account for an entire book. However, relevant stress theories will be invoked in the discussion while providing a literature review of the subject. Tensile, shear, and peel loadings will be covered since they are the most basic and common in structural adhesive applications. 

Choose shear loading while avoiding peel loading. 

Classical theories on stress analysis were discussed along with an extensive literature review to provide insight into the complex stresses produced in a variety of joint geometries. Tensile, shear, and peel loads were considered because of their presence in actual bonding applications. 

The loads acting on a bonded joint will result in various types of stresses. Stresses are normally expressed in N/mm. In the case of pure tensile and pure compressive loads the stress distribution over the bond line is very even, so that every part of the bond line carries the same load, and to calculate the stresses, the acting forces are simply divided by the bond area. In reality pure tensile and pure compressive loads are very rare and we are confronted more with shear, cleavage and peel loads. The joint stress distribution, i.e., the location of the stresses across the bond line, is less uniform and more complicated to calculate. 

---