Engineering and non-engineering adhesives

Adhesives may be classified as either organic or inorganic materials in a number of different ways; for example by origin, by method of bonding, by end use or on a chemical basis (1). Table 2.1 gives a broad classification of the organic adhesives based upon origin under the general headings of animal, vegetable, mineral, elastomeric, thermoplastic and thermosetting adhesives.

Animal glues are generally based on protein either in the form of mammalian collagen, from fish or from milk. They tend to be ‘sticky’ which is useful for applications requiring an instant grab or bond.

The common vegetable glues are based on either starch or cellulose. Unmodified starch dispersed in water is used to form paper pastes but a large proportion is now used in modified forms such as dextrin. Cellulose-based glues are produced by reacting hydroxyl groups present in the polymer chain with different reagents to form a variety of adhesives.

Mineral adhesives include silicates and phosphates for high temperature use and naturally derived products such as bitumen and asphalt.

The elastomeric group of adhesives is based on natural rubber latex and its derivatives or totally synthetic rubber known as SBR (styrene butadiene rubber). There is now a wide range of synthetic rubber adhesives based upon SBR including nitrile and butyl rubber. Another elastomeric adhesive is the versatile polyurethane rubber group.

Thermoplastic adhesives are so called because they may be softened by heating and rehardened on cooling without undergoing chemical changes.

Adhesives can offer substantial economic advantages over more conventional methods of joining. Whilst the building and construction industries represent some of the largest users of adhesive materials, few applications currently involve adhesive joints which are required to sustain large externally-applied forces. However, recent advances in the science and technology of adhesion and adhesives suggest that structural adhesives have enormous potential in future construction applications, particularly where the combination of thick bondlines, ambient temperature curing and the need to unite dissimilar materials with a relatively high strength joint are important. Indeed adhesive bonding, either alone or in combination with other methods of fastening, represents one of the key enabling technologies for the exploitation of new materials and for the development of novel design concepts and structural configurations.

Definitions and bonding

An adhesive may be defined as a material which, when applied to surfaces, can join them together and resist their separation. Thus adhesive is the general term used for substances capable of holding materials together by surface attachment and includes cement, glue, paste, etc. There is no universally accepted definition of a structural adhesive, but in the following chapters the term will be used to describe monomer compositions which polymerise to give fairly stiff and strong adhesives uniting relatively rigid adherends to form a load-bearing joint.

The term adhesion refers to the attraction between substances whereby when they are brought into contact work must be done in order to separate them.

Introduction

The importance of quality assurance in any industry is widely acknowledged. More stringent customer expectations with regard to quality, together with the realisation that continued improvements in quality are often necessary for a company to sustain good economic performance, mean that some level of quality assurance is mandatory for all those associated with adhesive bonding operations. The use of technical specifications and quality control procedures go some way towards fulfilling the requirements of a quality system, and these measures are outlined here.

Most of the aspects relating to the selection and use of adhesives have been discussed in the first four chapters. For example, the properties of adhesive materials have been reviewed, together with the design approach appropriate to structures assembled with them. The importance of surface pretreatment and its influence on joint durability has also been covered. However adhesion science must also interface with fabrication considerations and with structural engineering. To the technical difficulties facing the production engineer in using structural adhesives must be added a number of others for the civil engineer. Primarily these are that:

1. each project tends to be a ‘one-off’ production

2. fabrication commonly takes place on site and so the environment is not easily controlled

3. the success of the bonding operation is very dependent on the diligence employed by the operatives and the work requires skilled supervision

4. the bonded assembly has to last for many decades

5. adhesive polymers have poor fire resistance

Introduction

The truly structural adhesive joint is relatively new. The evolution of the various design approaches follows the empirical development of appropriate joint configurations(l–7) – themselves following on from the long historical development of load-bearing joints in, and between, engineering materials. It must however be emphasized that structural bonded joints existing in engineering disciplines other than those involving civil engineering tend to be formed with thin bondlines, often with relatively high modulus adhesives, whereas the general concern in the construction industry is with thick bondlines – often with lower modulus materials. This is an important difference, since the nature of the resultant bondline stress distributions of loaded joints may be significantly different.

The training normally given to an engineer in the various means of joining materials leaves him at a disadvantage when it comes to using adhesives, with essential choices between the many types available and with the design approach appropriate to structures assembled with these. Naturally the basis for design must stem from the intended function and service environment of the joint, and from a consideration of the loads and stresses which are likely to be encountered in service. As with any fastening method, it follows that the design must be dependent upon the nature of the materials to be joined as well as on the method of joining. It is, for instance, not sufficient simply to substitute adhesive bonding for welding, bolting or riveting.

The science of adhesion is truly multi-disciplinary, demanding a consideration of concepts from such diverse topics as surface chemistry, polymer chemistry, rheology, stress analysis and fracture mechanics. On the other hand adhesives and bonding form a part of our everyday lives and there are very few of us who have not used ‘glues’ in one form or another for DIY jobs at home or work. The difficulty arises in translating an understanding of the complexities of adhesion science to some fundamental ground rules for achieving a strong and durable bonded joint which can be relied upon to transmit significant externally applied forces throughout the life of the structure or component. The construction industry, particularly the civil engineering side, presents a unique situation in respect of the conditions under which adhesive joints may have to be formed, the materials to be bonded and the relatively long design life required as compared with other industries such as aerospace and automotive.

The purpose of this book, therefore, is to provide the practising civil engineer with an insight into the factors involved in the formation of a successful adhesive joint, viz. adhesive selection, surface preparation, joint design, fabrication and protection from hostile conditions in service. The application of these principles is then illustrated by reference to examples of current uses in repair, strengthening and new construction, and the potential for future developments.

Introduction

The successful performance of many every-day products, and many common materials and construction techniques, is dependent upon adequate adhesion between two or more constituents. Most engineers, however, have only the haziest of ideas about the whole concept of adhesion. For to know ‘how’ to prepare substrate surfaces for bonding does not necessarily require a knowledge of ‘why’ adhesive materials should stick to them. It is the intention of this chapter to connect theory with practice, to enable the reader to appreciate ‘why’ before discussing aspects of surface pretreatment pertinent to applications of adhesives in construction.

The strength of bonded assemblies depends not only on the cohesive strength of the adhesive, but also on the degree of adhesion to the bonding surface. One of the disadvantages of adhesive bonding as a method of fastening is that the surfaces need to be clean and, whatever their chemical nature, coherent if a satisfactory degree of contact, and therefore adhesion, is to be obtained. Frequently the adhesive itself is wrongly blamed for ‘not sticking’, but the general source of trouble lies with the surface pretreatment. The use of cold-cure epoxies generally necessitates the careful preparation of metallic adherends in particular, to ensure satisfactory long-term performance. Indeed, whatever the nature of the substrate, or the adhesive to be used with it, its pretreatment is probably the single most important aspect of the bonding operation.

Bearings and expansion joints for bridges

Perhaps the most well known application of resins in civil engineering is in the form of resin mortar for either bridge bearings or expansion joint nosings. In bearings the mortar is used as a bedding compound on which to seat rubber or steel bearing pads. These pads serve to transfer loads from the superstructure to piers or abutments and the stresses they resist are largely compressive in nature.

Expansion joints in concrete bridges were traditionally formed by the use of steel edging angles anchored or bolted into the concrete deck either side of the expansion gap. Such joints have a limited life due to disintegration of the supporting concrete under the action of traffic impact and they are very difficult to repair or replace. As a result resin mortar nosings were introduced on UK highway bridges in the mid 1960s but they too were not entirely free from trouble. The early nosings were based on relatively slow curing epoxy systems and a typical installation procedure is illustrated in Fig. 7.1(1). In due course other epoxy formulations began to appear and faster curing systems were introduced in an attempt to avoid the need for heating and testing in cold weather.

In the late 1960s transverse cracks began to appear in some nosings and in the following years hollowness and cracking became a major nuisance.

Concrete repair

The 1950s and 1960s saw an enormous growth in new construction and many faults are now becoming evident in some of the structures constructed during these years of peak activity. Today, repair and refurbishment is often preferred to demolition and reconstruction with the result that the maintenance and refurbishment market now represents approximately 50% of the construction industry workload of £37bn per annum. Concrete repair is no exception and currently represents a market of at least £500m annually. A variety of concrete repair methods have been developed many of which rely on adhesion between a repair material and the concrete substrate. The more important aspects of concrete repair will now be discussed under the headings of resin injection, patches, coatings and sealants and finally the implications of property mismatch.

Resin injection

The injection of resin-based materials into cracked concrete is a technique widely employed to either restore structural integrity or to seal the cracks for subsequent durability. However, a necessary preliminary to the consideration of resin injection is an assessment to establish the cause of cracking in the first place. Here it is important to distinguish between inactive cracks which no longer move and live cracks which may continue to move with changing loads or temperatures.

In order to assist in establishing the cause of cracking in concrete it is often helpful to distinguish between intrinsic cracking(l) and that caused by an externally applied stress.

Introduction

It has already been observed that structural adhesive bonding, either alone or in combination with other methods of fastening, represents a key enabling technology for the exploitation of new as well as existing materials, and for the development of novel design concepts and structural configurations. This potential has been helped by developments in adhesive materials and simplified surface treatment processes, together with an improved understanding of bonded joint behaviour and design concepts.

Adhesives can offer real technical and economic benefits, providing greater freedom and flexibility for designers. The benefits available may stem particularly from repair and strengthening work, as well as from opportunities with lightweight structures or those in which increased fatigue resistance is desirable. Largely to circumvent some of the potential on-site fabrication difficulties, a number of future development possibilities lie with the off-site prefabrication of bonded assemblies. A number of examples exist currently in building construction, such as sandwich panels, curtain walling and cladding assemblies, structural glazing, stressed skin flooring and framing, and ‘macro’ composite and compound structural elements. It is likely that greater use will be made of high strength alloys and polymer composite materials in the future.

The potential opportunities for making bonded connections in civil engineering structures outlined in this chapter are by no means exhaustive. They represent examples which vary substantially in developmental progress from possibilities to well researched concepts.

Introduction

In the previous chapter, elastic—plastic models for soil were discussed in a general way. Yield loci and plastic potentials were sketched, but no attempt was made to suggest possible mathematical expressions for these curves. In this chapter, a particular model of soil behaviour is described and used to predict the response of soil specimens in standard triaxial tests. In subsequent chapters, this model is used to illustrate a number of features of soil behaviour. This particular model can be regarded as one of the set of volumetric hardening models covered by the discussion in Chapter 4. When the model was originally described in the literature by Roscoe and Burland (1968), it was called ‘modified’ Cam clay to distinguish it from an earlier model called Cam clay (Roscoe and Schofield, 1963). The qualifier modified is dropped here because the modified Cam clay model has probably been more widely used for numerical predictions. The ‘original’ Cam clay model is mentioned in Section 8.4.

The model is described here in terms of the effective stress quantities p' and q which are relevant to the discussion of soil response in conventional triaxial tests. Most of the examples given are for triaxial compression, though it is tacitly assumed that triaxial extension can be accommodated merely by allowing the deviator stress q to take negative values. The extension of the model to other more general states of stress is described briefly in Section 10.6.2.

Introduction: purpose of models

Models of the mechanical behaviour of soils have served two purposes in this book: they have been used primarily to illustrate facets of the observed behaviour of soils which might at first sight be considered extraordinary but which, with even a simple model of soil behaviour, can in fact be anticipated; however, Section 11.3 has shown how such models can be used in finite element analyses of geotechnical problems of practical importance. There are different requirements for models used for purely illustrative purposes and for models used for predicting the response of geotechnical structures. The illustrative model is expected to give a simplified but overall picture of soil behaviour, but the predictive model must be able to match rather closely the behaviour of the elements of real soil which are being deformed in a particular prototype. Some of the simplifying assumptions of the illustrative model are inappropriate for the predictive model if the predictions are to be useful. Many implicit or explicit assumptions have been made in this book in presenting the simple models; the effects of relaxing some of these assumptions are discussed briefly in this chapter.

Effects of time

Loading of a soil with no drainage, or with restricted drainage, in general leads to the generation of excess pore pressures as a result of prevented volume change in the soil. With time, these excess pore pressures tend to dissipate to equilibrium values.