Chapter 1: The Relationship of Structure to Building PDF

CHAPTER 1 The relationship of structure to building The simplest way of describing the function of an architectural structure is to say that it is the part of a building that resists the loads that are imposed on it. A building may be regarded as simply an envelope that encloses and subdivides spa...

CHAPTER 1 The relationship of structure to building The simplest way of describing the function of an architectural structure is to say that it is the part of a building that resists the loads that are imposed on it. A building may be regarded as simply an envelope that encloses and subdivides space in order to create a protected environment. The surfaces that form the envelope, that is the walls, the floors and the roof of the building, are subjected to various types of loading: external surfaces are exposed to the climatic loads of snow, wind and rain; floors are subjected to the gravitational loads of the occupants and their effects; and most of the surfaces also have to carry their own weight (Figure 1.1). All of these loads tend to distort the building envelope and to induce it to collapse; it is to prevent this from happening that a structure is provided. The function of a structure may be Figure 1.1 Loads on the building envelope. Gravitational loads due to snow and to the occupation of the building cause roof and floor structures to bend and induce Facing page: compressive internal forces in walls. Wind causes pressure and suction loads to act on Pantheon, Rome. Painting: all external surfaces. Panini. Figure 1.2 The igloo is a self-supporting compressive Figure 1.3 In the tepee a non-structural skin is supported envelope. on a structural framework of timber poles. Figure 1.4 Exhibition Hall of the CNIT, Paris, 1958; Nicolas Esquillan, engineer. The principal element is a self- supporting reinforced concrete shell. Photo: David Monniaux/Wikimedia Commons.  THE RELATIONSHIP OF STRUCTURE TO BUILDING 11 summed up, therefore, as being to supply the strength and rigidity that are required to prevent a building from collapsing. More precisely, it is the part of a building that conducts the loads that are imposed on it from the points where they arise to the ground underneath the building, where they can ultimately be resisted. The location of the structure within a building is not always obvious because the structure can be integrated with the non-structural parts in various ways. Sometimes, as in the simple example of an igloo (Figure 1.2), in which ice blocks form a self-supporting protective dome, the structure and the space enclosing elements are one and the same thing. Alternatively, the structural and space-enclosing elements can be entirely separate. A very simple example is the tepee (Figure 1.3) in which the protecting envelope is a skin of fabric or hide that has insufficient rigidity to form an enclosure by itself and that is supported on a framework of timber poles. Complete separation of structure and envelope occurs here: the envelope is entirely non-structural and the poles have a purely structural function. Figure 1.5 Modern Art Glass Warehouse, Thamesmead, UK, 1973; Foster Associates, architects; Anthony Hunt Associates; structural engineers. A non-structural skin of profiled metal sheeting is supported on a steel framework, which has a purely structural function. Photo: Andrew Mead. 12 THE RELATIONSHIP OF STRUCTURE TO BUILDING The CNIT exhibition hall in Paris (Figure 1.4) is a sophisticated version of the igloo; the reinforced concrete shell that forms the main element of this enclosure is self-supporting and therefore structural. Separation of skin and structure occurs in the transparent walls, however, where the glass envelope is supported on a structure of mullions. The roof of the Centre Pompidou building at Metz (Figure 11.6), the overall form of which was largely deter- mined by the requirements of its lattice-timber ‘shell’ structure, is similarly configured although in this case the timber structural elements are distinct from the enclosing roof surface that it supports and the building is therefore similar to the tepee in its separation of structural from enclosing elements. The steel frame warehouse by Foster Associates at Thamesmead, UK (Figure 1.5), is almost a direct equivalent of the tepee. The elements that form it are either purely structural or entirely non-structural because the corrugated sheet metal skin is entirely supported by the steel frame, which has a purely structural function. A similar breakdown may be seen in later buildings by the same architects, such as the Sainsbury Centre for the Visual Arts at Norwich, UK (Figures 9.33 and 9.34) and the warehouse and showroom for the Renault car company at Swindon (Figure 3.19). In most buildings the relationship between the envelope and the structure is more complicated than in the above examples and frequently this is because the interior of the building is subdivided to a greater extent by internal walls and floors. For instance, in Foster Associates’ building for Willis, Faber & Dumas (WFD), Ipswich, UK (Figures 1.6 and 5.16) the reinforced concrete structure of floor slabs and columns may be thought of as having a dual function. The columns are purely structural, although they do punctuate the interior spaces and are space-dividing elements, to some extent. The floors are both structural and space-dividing elements. Here, however, the situation is complicated by the fact that the structural floor slabs are topped by non- structural floor finishing materials and have ceilings suspended underneath them. The floor finishes and ceilings could be regarded as the true space- defining elements and the slab itself as having a purely structural function. The glass walls of the building are entirely non-structural and have a space- enclosing function only. The Solaris Building in Singapore by Ken Yeang, with Arups as engineers, (Figures 1.7 and 1.8) is also supported by a reinforced concrete structure. Here the structural continuity (see Glossary) and mouldability that concrete offers were exploited to create a complex juxtaposition of solid and void. The building is of the same basic type as the WFD building however: a structural framework of reinforced concrete supports cladding elements that are non- structural. In the Centre Pompidou in Paris by Piano and Rogers a multi-storey steel framework is used to support reinforced concrete floors and non-loadbearing glass walls. The breakdown of parts is straightforward (Figs 9.28 to 31): identical plane-frames, consisting of long steel columns which rise through  THE RELATIONSHIP OF STRUCTURE TO BUILDING 13 Figure 1.6 Willis, Faber & Dumas Office, Ipswich, UK, 1974; Foster Associates, architects; Anthony Hunt Associates, structural engineers. The basic structure of this building is a series of reinforced concrete coffered slab floors supported on a grid of columns. The external walls are of glass and are non-structural. In the finished building the floor slabs are visible only at the perimeter. Elsewhere they are concealed by floor finishes and a false ceiling. Photo: A. Hunt. the entire height of the building supporting triangulated girders at each floor level, are placed parallel to each other to form a rectangular plan. The concrete floors span between the triangulated girders. Additional small cast-steel girders project beyond the line of columns and are used to support stairs, escalators and servicing components positioned along the sides of the building outside the glass wall, which is attached to the frame near the columns. A system of cross-bracing on the sides of the framework prevents it from collapsing through instability. In this type of building the structure not only provides support but makes a significant contribution to the visual aspects of the architecture. The free form, in both plan and cross-section, of the Riverside Museum in Glasgow by architect Zaha Hadid (Figures 1.9 and 10.25 and 26), make it in some respects a complete contrast to the controlled order of the Centre Figure 1.7 Solaris Building, Singapore, 2011; T. R. Hamzah & Yeang, architects; Arups, engineers. Ecological and sustainability considerations greatly influenced the design of this building. The structural armature is a reinforced concrete framework that allowed the creation of irregular plan-forms and facilitated the inclusion of green corridors and a passively ventilated atrium. Photo: T. R. Hamzah & Yeang SDN. BHD; Photography credit: Albert Lim.  THE RELATIONSHIP OF STRUCTURE TO BUILDING 15 Figure 1.8 Solaris Building, Singapore, 2011; T. R. Hamzah & Yeang, architects; Arups, engineers. Cross-section. The structural continuity offered by the reinforced concrete framework allowed the creation of irregular, curvilinear plan- forms, cantilevered balconies and an internal atrium – all of which contributed to the passive system for environmental control. Graphic: Courtesy of T. R. Hamzah & Yeang SDN. BHD. Pompidou. Architecturally it is quite different, and the structural action is completely suppressed, but structurally the principal part of the building is similar to the extent that a metal skeleton frame-work supports a light, enclosing skin (Figure 1.10). The house with masonry walls and timber floor and roof structures is a traditional form of building in most parts of the world. It is found in many forms, from the historic grand houses of the European landed aristocracy (Figure 1.11) to modern homes in the UK (Figure 1.12). Even the simplest versions of this form of masonry-and-timber building (Figure 1.13) are fairly complex assemblages of elements. Initial consideration could result in a straightforward breakdown of parts with the masonry walls and timber floors being regarded as having both structural and space-dividing functions and the roof as consisting of a combination of the purely supportive trusses, which are the structural elements, and the purely protective, non-structural skin. Closer examination would reveal that most of the major elements can in fact be subdivided into parts that are either purely structural or entirely non-structural. The floors, for example, normally consist of an inner core of timber joists and Figure 1.9 Riverside Museum, Glasgow, 2012; Zaha Hadid, architect; Buro Happold Engineering, engineers. The principal space in this building is an S-plan single exhibition area, of serrated cross-section, that runs through its entire length. The supporting structure is a steel skeleton framework that supports non-structural cladding. The glazed end of the building illustrates its cross-sectional shape. Photos: (a) E. Z. Smith/Hawkeye; (b) Bjmullan/Wikimedia Commons.  THE RELATIONSHIP OF STRUCTURE TO BUILDING 17 floor boarding, which are the structural elements, enclosed by ceiling and floor finishes. The latter are the non-structural elements which are seen to divide the space. A similar breakdown is possible for the walls and in fact very little of what is visible in the traditional house is structural, as most of the structural elements are covered by non-structural finishes. It should not be thought, however, that structural considerations do not make a significant contribution to architectural aspects of traditional loadbearing-wall buildings. Their overall form and general arrangements are in fact largely determined to satisfy structural requirements and the influence of structure on the traditional house, of whatever size and architectural style, was and is profound. To sum up, these few examples of very different building types demonstrate that all buildings contain a structure whose primary function is to support the building envelope by conducting the forces that are applied to it from the points where they arise in the building to the ground below it where they are ultimately resisted. Sometimes the structure is indistinguishable from the enclosing and space-dividing building envelope, sometimes it is entirely separate from it; most often there is a mixture of elements with structural, non-structural and combined functions. In all cases the form of the structure is very closely related to that of the building taken as a whole and usually exerts considerable influence on the nature of that form. The structure may also make a significant contribution to the architectural vocabulary being employed. The elegance with which the structure fulfils its function is considered by many to be something that affects the quality of the architecture. Figure 1.10 Riverside Museum, Glasgow, 2012; Zaha Hadid, architect; Buro Happold Engineering, engineers. Due to its unconventional structural configuration, a massively strong steel framework structure was required. Photo: Hélène Binet. Figure 1.11 Chateau de Chambord, France, 1519–1547; Domenico da Cortona, architect; Pierre Nepveu, engineer. One of the grandest domestic buildings in Europe, the Chateau de Chambord has a loadbearing masonry structure. Most of the walls are structural; the floors are either of timber or vaulted masonry and the roof structure is of timber. Photo: Patricia & Angus Macdonald/Aerographica. Figure 1.12 Local authority housing, Haddington, Scotland, 1974; J. A. W. Grant, architects. These buildings have loadbearing masonry walls and timber floor and roof structures. Photo: P. Macdonald. Figure 1.13 Traditional construction in the UK, in its twentieth- century form, with loadbearing masonry walls and timber floor and roof structures. All structural elements are enclosed in non-structural finishing materials. CHAPTER 2 Structural requirements 2.1 Introduction To perform its primary function of supporting a building in response to whatever loads may be applied to it a structure must possess four properties: it must be capable of achieving a state of static equilibrium, it must be stable, it must have adequate strength and it must have adequate rigidity. The meanings of these terms are explained in this chapter. 2.2 Equilibrium Structures must be capable of achieving a state of static equilibrium under the action of applied load. This requires that the structure, taken as a whole, must be connected to its foundations in such a way that all possible applied loads are balanced exactly by reactions generated at its supports. Similarly, each element in the structure must be connected to the rest of the structure such that equilibrium is established under all possible loading conditions. 2.3 Geometric stability Geometric stability is the property that preserves the geometry of a structure and allows its elements to act together to resist load. The distinction between stability and equilibrium is illustrated by the framework shown in Figure 2.1, which is capable of achieving a state of equilibrium under the action of gravi- tational load, but which is not stable because the frame will collapse if disturbed laterally. Stability can therefore be distinguished from strength or rigidity, because a system can become unstable even though its elements are Facing page: sufficiently strong and rigid to resist the loads that are imposed on them. 30 St Mary Axe, London, This simple arrangement demonstrates that the critical factor, so far as the Foster + Partners/Ove stability of any system is concerned, is the effect on it of a small disturb- Arup & Partners. Photo: ance. In the context of structures this is shown very simply in Figure 2.2 by Muttoo. 22 STRUCTURAL REQUIREMENTS the comparison of tensile and compressive elements. Both are capable of achieving equilibrium but, if the alignment of either is disturbed, the tensile element is pulled back into line following the removal of the disturbing agency but the compressive element, once its initially perfect alignment has been altered, progresses to an entirely new position. The fundamental issue of stability is demonstrated here, which is that stable systems revert to their original state following a slight disturbance whereas unstable systems progress to an entirely new state. The parts of structures that tend to be unstable are the ones in which compressive forces act and these parts must therefore be given special attention when the geometric stability of an arrangement is being considered. The columns in a simple rectangular framework are examples of this (Figure 2.1). The geometric instability of the arrangement in Figure 2.1 would have been obvious if its response to horizontal load had been considered (Figure 2.3) and this demonstrates one of the fundamental requirements for the geometric stability of any arrangement of elements, which is that it must be capable of resisting loads from orthogonal directions (directions at right angles) – two orthogonal directions for plane arrangements and three for three-dimensional arrangements. This is another way of saying that an arrangement must be capable of achieving a state of equilibrium in response to forces from three orthogonal directions. The stability or otherwise of a proposed arrangement can therefore be judged by considering the effect on it of sets of mutually perpendicular trial forces: if the arrangement is capable of resisting all of these then it is stable, regardless of the loading pattern that will actually be applied to it in service. Conversely, if an arrangement is not capable of resisting load from three orthogonal directions then it will be unstable even though the load that it is designed to resist will be applied from only one direction. It frequently occurs in architectural design that a geometry that is potentially unstable must be adopted in order that other architectural requirements can Figure 2.1 A rectangular frame with four hinges is capable of achieving a state of equilibrium under gravitational loading but is unstable because any slight lateral disturbance of the columns will induce it to collapse. The frame on the right here is stabilised by the diagonal element which makes no direct contribution to the resistance of the gravitational load. Figure 2.2 The tensile element on the left here is stable because the loads pull it back into line following a disturbance. The compressive element on the right is fundamentally unstable. Figure 2.3 Conditions for stability of frameworks. The 2-D system is stable if it is capable of achieving equilibrium in response to forces from two mutually perpendicular directions. 24 STRUCTURAL REQUIREMENTS be satisfied. For example, one of the most convenient structural geometries for buildings, that of the rectangular frame, is unstable in its simplest hinge- jointed form, as has already been shown. Stability can be achieved with this geometry by the use of rigid joints, by the insertion of a diagonal element or by the use of a rigid diaphragm that fills up the interior of the frame (Figure 2.4). Each of these has disadvantages. Rigid joints are the most convenient from a space-planning point of view but are problematic structurally because they are difficult to construct and because they can render the structure statically indeterminate (see Glossary). Diagonal elements and diaphragms block the framework and can complicate space planning. In multi-panel arrangements, however, it is possible to produce stability without blocking every panel. The row of frames in Figure 2.5, for example, is stabilised by the insertion of a single diagonal. Where frames are parallel to each other the three-dimensional arrangement is stable if a few panels in each of the two principal directions is stabilised in the vertical plane and the remaining frames are connected to these by diagonal elements or diaphragms in the horizontal plane (Figure 2.6). A three-dimensional frame can therefore be stabilised by the use of diagonal elements or diaphragms in a limited number of panels in the vertical and horizontal planes. In multi-storey arrangements these systems must be provided at every storey level. None of the components that are added to stabilise the geometry of the rectangular frames in Figures 2.1 and 2.6 would make a direct contribution to the resistance of gravitational load (i.e. the carrying of weight), which would normally be the primary load on the structure. These elements are called bracing elements and most structures contain such elements, whose presence frequently affects both the planning and the appearance of the building that it supports. Where, as is normal, a structure is subjected to loads from different direc- tions, the elements that are used solely for bracing when the principal load is applied frequently play a direct role in resisting secondary load. The diagonal elements in the frame of Figure 2.6 for example, would be directly involved in the resistance of horizontal load caused by the action of wind. The diagonal or diaphragm bracing elements that are inserted into rectangular frameworks are often referred to as wind bracing, which gives the impression their sole function is to carry wind load. This is incorrect. These elements would be essential for stability even if wind was not a consideration in the design. It is common practice to provide more bracing elements than the mini- mum number required for stability so as to improve the resistance of three- dimensional frameworks to horizontal load. The framework in Figure 2.6, for example, although theoretically stable would suffer considerable distortion in response to a horizontal load applied parallel to the long side of the frame at the opposite end from the vertical-plane bracing. A load applied parallel to the long side at this end of the frame would also cause a certain amount of distress as some movement of joints would inevitably occur in the transmission Figure 2.4 A rectangular frame can be stabilised by the insertion of a diagonal element or a rigid diaphragm or by the provision of rigid joints. A single rigid joint is in fact sufficient to provide stability. Figure 2.5 A row of rectangular frames is stable if one panel only is braced by any of the three methods shown in Figure 2.4. Figure 2.6 These frames contain the minimum number of braced panels required for stability. 26 STRUCTURAL REQUIREMENTS Figure 2.7 In practical bracing schemes more elements than are strictly necessary to ensure stability are provided to improve the performance of frameworks in resisting horizontal load. The frame at the top here is stable but will suffer distortion in response to horizontal load on the side walls. Its performance is enhanced if a diagonal element is provided in both end walls. The lowest framework contains the minimum number of elements required to resist effectively horizontal load from the two principal horizontal directions. Note that the vertical-plane bracing elements are distributed around the structure in a symmetrical configuration. of it to the vertical-plane bracing at the other end. In practice the performance of the frame is more satisfactory if vertical-plane bracing is provided at both ends (Figure 2.7). This gives more restraint than is necessary for stability and makes the structure statically indeterminate but results in the horizontal loads being resisted close to the points where they are applied to the structure. Figures 2.8 and 2.9 show typical bracing systems for multi-storey frame- works. Another common arrangement, in which floor slabs act as diaphragm- type bracing in the horizontal plane in conjunction with vertical-plane bracing of the diagonal type, is shown diagrammatically in Figure 2.10. Where the rigid-joint method is used it is normal practice to stabilise all panels individually by making all joints rigid. This greatly increases planning freedom by eliminating the need for bracing in the vertical planes. The rigid-joint method is the normal method that is adopted for reinforced concrete frames,  STRUCTURAL REQUIREMENTS 27 Figure 2.8 Typical bracing schemes for multi-storey frameworks. Vertical-plane bracing is provided in a limited number of bays and positioned symmetrically on plan. All other bays are linked to this by horizontal-plane bracing at every storey level. in which continuity through junctions between elements can easily be achieved; diaphragm bracing is also used however in both vertical and horizontal planes in certain types of reinforced concrete frame. Loadbearing wall structures are those in which the external walls and internal partitions serve as the vertical structural elements. They are normally constructed of masonry, reinforced concrete or timber but combinations of these materials are also used. In all cases the joints between walls and floors are normally incapable of resisting bending action (they behave as hinges, in other words) and the resulting lack of continuity means that rigid-frame action cannot develop. Diaphragm bracing, provided by the walls themselves, is used to stabilise these structures. A wall panel has high rotational stability in its own plane but is unstable in the out-of-plane direction (Figure 2.11); vertical panels must, therefore, be grouped in pairs at right angles to each other so that they provide mutual support. Because loadbearing wall structures are normally used for multi- cellular buildings, the provision of an adequate number of vertical-plane bracing diaphragms in two orthogonal directions is normally straightforward Figure 2.9 These drawings of floor grid patterns for steel frameworks show typical locations for vertical-plane bracing. Figure 2.10 Concrete floor slabs are normally used as horizontal-plane bracing of the diaphragm type which acts in conjunction with diagonal bracing in the vertical planes.  STRUCTURAL REQUIREMENTS 29 Figure 2.11 Walls are unstable in the out-of- Figure 2.12 Loadbearing masonry buildings are normally multi- plane direction and must be grouped into cellular structures that contain walls running in two orthogonal orthogonal arrangements for stability. directions. The arrangement is inherently stable. (Figure 2.12). It is unusual therefore for bracing requirements to have a significant effect on the internal planning of this type of building. The need to ensure that a structure is stable is a factor that normally affects the internal planning of buildings. A basic requirement is that some form of bracing must be provided in two orthogonal directions on plan and, if diagonal or diaphragm bracing is used, this will affect wall arrangements. Because vertical-plane bracing is most effective when it is placed symmetrically, either in internal cores or around the perimeter of the building, this can restrict space-planning freedom, especially in tall buildings where the effects of wind loading are significant. 2.4 Strength and rigidity 2.4.1 Introduction The application of load to a structure generates internal forces in the elements and external reacting forces at the foundations (Figure 2.13) and the elements and foundations must have sufficient strength and rigidity to resist these. They must not rupture when the peak load is applied; neither must the deflection that results from the peak load be excessive. The requirement for adequate strength is satisfied by ensuring that the levels of stress that occur in the various elements of a structure, when the peak 30 STRUCTURAL REQUIREMENTS loads are applied, are within acceptable limits. This is chiefly a matter of providing elements with cross-sections of adequate size, given the strength of the constituent material. The determination of the sizes required is carried out either by using geometric rules (such as minimum ratios of span to depth for beams) or by structural calculations (see Chapter 7 for comparison of the two methods). The provision of adequate rigidity is similarly dealt with. Structural calculations – the method most commonly used in the present day to determine suitable sizes for structural elements – allow the strength and rigidity of structures to be controlled precisely and must be preceded by an assessment of the load that a structure will be required to carry. The calculations may be considered to be divisible into two parts and to consist first, of structural analysis, which is the evaluation of the internal forces that occur in the elements of the structure, and second, the element-sizing calculations that are carried out to ensure that they will have sufficient strength Figure 2.13 The structural elements of a building conduct the loads to the foundations. They are subjected to internal forces that generate stresses whose magnitudes depend on the intensities of the internal forces and the sizes of the elements. The structure will collapse if the stress levels exceed the strength of the material.  STRUCTURAL REQUIREMENTS 31 and rigidity to resist the internal forces that the loads will cause. In many cases, and always for statically indeterminate structures, the two sets of calculations are carried out together but it is possible to think of them as separate operations and they are described separately here. 2.4.2 The assessment of load The assessment of the loads that will act on a structure involves the prediction of all the different circumstances that will cause the load to be applied to a building in its lifetime (Figure 2.14) and the estimation of the greatest magnitudes of these loads. The maximum load could occur when the building was full of people, when particularly heavy items of equipment were installed, when it was exposed to the force of exceptionally high winds or as a result of many other eventualities. The designer must anticipate all of these possibilities and also investigate all likely combinations of them. The evaluation of load is a complex process involving the statistical analysis of load data but guidance is normally available to the designer of a structure from loading standards (see Section 7.3.5). 2.4.3 The analysis calculations The purpose of structural analysis is to determine the magnitudes of all of the forces, internal and external, that occur on and in a structure when the most Figure 2.14 The prediction of the maximum load that will occur is one of the most problematic aspects of structural calculations. Loading Standards are provided to assist with this but assessment of load is nevertheless one of the most imprecise parts of the structural calculation process. 32 STRUCTURAL REQUIREMENTS unfavourable load conditions occur. It is a procedure in which the external reactions that act at the foundations of a structure and the internal forces in its elements are calculated from the loads (Figure 2.15). This is a process in which the structure is reduced to its most basic abstract form and considered separately from the rest of the building that it will support. The different types of internal force that can occur in a structural element are shown in Figure 2.16. As these have a very significant influence on the sizes and shapes that are specified for elements they will be described briefly here. In Figure 2.16 an element is imagined to be cut through at a particular cross-section. In Figure 2.16a the forces that are external to one of the resulting sub-elements are marked. If these were indeed the only forces that acted on the sub-element it would not be in a state of equilibrium. For equilibrium the forces must balance and this is clearly not the case here: an Figure 2.15 In structural analysis the complete structure is broken down into 2-D components and the internal forces in these are subsequently calculated. The diagram shows the pattern forces that result from gravitational load on the roof of a small building. Similar breakdowns are carried out for the other forms of load and a complete picture is built up of the internal forces that will occur in each element during the life of the structure.  STRUCTURAL REQUIREMENTS 33 additional vertical force is required for equilibrium. As no other external forces are present on this part of the element the extra force must act on the cross-section where the cut occurred. Although this force is external to the sub-element it is an internal force so far as the complete element is concerned and is called the shear force (see Glossary). Its magnitude at the cross-section where the cut was made is simply the difference between the external forces that occur to one side of the cross-section, i.e. to the left of the cut. Once the shear force is added to the diagram (Figure 2.16b) the question of the equilibrium of the sub-element can once more be examined. In fact, it is still not in a state of equilibrium because the set of forces now acting will produce a turning effect on the sub-element that will cause it to rotate in a clockwise sense. For equilibrium an anti-clockwise moment is required and as before this must act on the cross-section at the cut because no other external forces are present. The moment that acts at the cut and that is required to establish rotational equilibrium is called the bending moment (see Glossary) at the cross-section of the cut. Once this is added to the diagram the system is in a state of static equilibrium, because all the conditions for equilibrium are now satisfied. Shear force and bending moment are forces that occur inside structural elements and they can be defined as follows. The shear force at any location is the amount by which the external forces acting on the element, to one side (a) (b) Figure 2.16 The investigation of internal forces in a simple beam using the device of the ‘imaginary cut’. The cut produces a free- body-diagram from which the nature of the internal forces at a single cross-section can be deduced. The internal forces at other cross- sections can be determined from similar diagrams produced by cuts made in appropriate places. (a) Not in equilibrium. (b) Positional (c) equilibrium but not in rotational equilibrium. (c) Positional and rotational equilibrium. The shear force on the cross-section 1.5 m from the left-hand support is 15 kN; the bending moment on this cross-section is 22.5 kNm. 34 STRUCTURAL REQUIREMENTS of that location, do not balance when they are resolved perpendicular to the axis of the element. The bending moment at a location in an element is the amount by which the moments of the external forces acting to one side of the location, about any point in their plane, do not balance. Shear force and bending moment occur in structural elements that are bent by the action of the applied load. Beams and slabs are examples of such elements. One other type of internal force can act on the cross-section of an element, namely axial thrust1 (Figure 2.17). This is defined as the amount by which the external forces acting on the element to one side of a particular location do not balance when they are resolved parallel to the direction of the element. Axial thrust can be either tensile or compressive. In the general case each cross-section of a structural element is acted upon by all three internal forces, namely shear force, bending moment and axial thrust. In the element-sizing part of the calculations cross-section sizes are determined which ensure that the levels of stress that these produce are not excessive. The efficiency with which these internal forces can be resisted depends on the shape of the cross-section (see Section 4.2). Figure 2.17 The ‘imaginary cut’ is a device for exposing internal forces and rendering them susceptible to equilibrium analysis. In the simple beam shown here, shear force and bending moment are the only internal forces required to produce equilibrium in the element isolated by the cut. These are therefore the only internal forces that act on the cross-section at which the cut was made. In the case of the portal frame, axial thrust is also required at the cross-section exposed by the cut.  STRUCTURAL REQUIREMENTS 35 The magnitudes of the internal forces in structural elements are rarely constant along their lengths and, once calculated, are normally presented graphically in the form of bending moment, shear force and axial thrust diagrams for each structural element (Figure 2.18). The shapes of bending moment, shear force and axial thrust diagrams are of great significance for the eventual shapes of structural elements because they indicate the locations of the parts where greatest strength will be required. Bending moment is normally large in the vicinity of mid-span and near-rigid joints. Shear force is highest near support joints. Axial thrust is usually constant along the length of struc- tural elements. Figure 2.18 The magnitudes of internal forces normally vary along the length of a structural element. Repeated use of the ‘imaginary-cut’ technique yields the pattern of internal forces in this simple beam. 36 STRUCTURAL REQUIREMENTS 2.4.4 Element sizing calculations The size of cross-section that is provided for a structural element must be such as to give it adequate strength and adequate rigidity: in other words, the size of the cross-section must allow the internal forces determined in the analysis to be carried without overloading the structural material and without the occurrence of excessive deflection. The calculations that are carried out to achieve this involve the use of the concepts of stress and strain (see Glossary). In the sizing calculations each element is considered individually and an area of cross-section determined which will maintain the stress at an acceptable level in response to the peak internal forces. The detailed aspects of the calculations depend on the type of internal force and therefore stress involved and on the properties of the structural material. As with most types of design the evolution of the final form and dimensions of a structure is, to some extent, a cyclic process. If the element sizing procedures produce cross-sections that are considered to be excessively large or unsuitable in some other way, modification of the overall form of the structure will be undertaken so as to redistribute the internal forces. Then, the whole cycle of analysis and element sizing calculations must be repeated. If a structure has a geometry that is stable and the cross-sections of the elements are sufficiently large to ensure that it has adequate strength it will not collapse under the action of the loads that are applied to it. It will therefore be safe, but this does not necessarily mean that its performance will be satisfactory (Figure 2.19). It may suffer a large amount of deflection under the action of the load and any deformation that is large enough to cause damage to brittle building components, such as glass windows, or to cause alarm to the building’s occupants or even simply to cause unsightly distortion of the building’s form, is a type of structural failure. The deflection that occurs in response to a given application of load to a structure depends on the sizes of the cross-sections of the elements2 and can be calculated once element dimensions have been determined. If the sizes that have been specified to provide adequate strength will result in excessive deflection they are increased by a suitable amount. Where this occurs it is the rigidity requirement that is critical and that determines the sizes of the structural elements. Rigidity is therefore a phenomenon that is not directly related to strength; it is a separate issue and is considered separately in the design of structures. 2.5 Conclusion In this chapter the factors that affect the basic requirements of structures have been reviewed. The achievement of stable equilibrium has been shown to be dependent largely on the geometric configuration of the structure and is therefore a consideration that affects the determination of its form. A stable  STRUCTURAL REQUIREMENTS 37 Figure 2.19 A structure with adequate strength will not collapse but excessive flexibility can render it unfit for its purpose. form can almost always be made adequately strong and rigid but the form chosen affects the efficiency with which this can be achieved. The provision of adequate strength is accomplished by analysis of the structure to determine the types and magnitudes of the internal forces that will occur in all of the elements when the maximum load is applied and the selection of cross-section shapes and sizes which are such that the stress levels are maintained within acceptable limits. The amount of deflection that will occur under the maximum load can then be calculated and, if this is excessive, the element sizes are increased to bring the deflection within acceptable limits. One final point worth noting is the effect of structural form on structural efficiency. A stable form can always be given adequate strength and rigidity simply by making individual elements large enough to ensure that the stresses are not excessive. It may be thought, therefore, that the provision of adequate strength and rigidity is not something that need be considered when evolving the form of a structure (and therefore of the building that it supports). However, the form selected directly affects the types and magnitudes of the internal forces and therefore of the strengths that are required of the structural elements and the quantities of material that must be provided to realise these strengths. Some forms generate much larger internal forces than others in response to the same applied load. The form that is selected therefore determines the efficiency with which the load can be resisted and is something that should be considered in the preliminary design of a building if wasteful use of material is to be avoided. An inappropriate form may even render the building structurally unviable. Notes 1 Other types of internal force, such as torsion, can occur but are not considered here as they are not normally significant in architectural structures. 2 The deflection of a structure is also dependent on the properties of the structural material and on the overall configuration of the structure. CHAPTER 3 Structural materials 3.1 Introduction The selection of the material for a structure is one of the most crucial aspects of its design. It has consequences for the overall form of the structure, and therefore of the building that it supports, and also for several aspects of its aesthetic make-up. The strength characteristics of a material obviously affect the load-carrying potential of the structure and therefore the maximum height and spans that can be achieved. The physical properties of the material also determine the types of internal force that can be carried and therefore the categories of element for which it is suitable. Unreinforced masonry, for example, may only be used where compressive stress dominates. Reinforced concrete performs well when loaded in compression or bending but not particularly well in axial tension. Steel is the most suitable material for tensile elements. The processes by which materials are manufactured and subsequently fashioned into structural elements also play a role in determining the shapes of elements for which they are suitable and influence the forms in which they are available to the builder. The properties of the structural material also have an important role in determining the visual qualities of a building. The slenderness of elements in high-strength materials such as steel contrasts with the massiveness of a masonry structure. The tactile quality, surface textures and colours of structural components can also affect the ‘materiality’ of architecture. Another vital aspect of the specification of a material is its performance in respect of sustainability. The two principal considerations are the ecological footprint associated with the initial construction – as determined by such Facing page: factors as embodied energy, carbon footprint and embodied water – and the Aspen Art Museum, potential of the material for recycling. The former is notoriously difficult to Aspen, Shigeru evaluate but must nevertheless be considered. It is discussed here in general Ban/KL&A. terms only. Realistic concern for recycling requires that it be given serious Photo: Derek Skalko. 40 STRUCTURAL MATERIALS consideration at the design stage of a building and can affect the choice of material. The various aspects of the influence of material properties on structural design are now discussed in relation to the four principal structural materials of masonry, timber, steel and reinforced concrete. 3.2 Masonry Masonry is one of the ancient building materials that, over the centuries, has been used to construct some of the most spectacular structures of the Western architectural tradition, which have included very tall buildings and wide-span enclosures. Its strength properties are not ideal, however, and for the largest Figure 3.1 Laon Cathedral, France, C12 and C13 CE. The Gothic church incorporates most of the various forms for which masonry is suitable. Columns, walls and compressive form-active arches and vaults are all visible here. Photo: Mattana-Mattis/Wikimedia Commons.  STRUCTURAL MATERIALS 41 structures in particular it has required that structural forms be adopted that eliminate tension. These include the vault and the dome, to achieve large horizontal spans, and thick, buttressed walls to allow great height to be provided safely. Masonry is a composite material in which individual stones, bricks or blocks are bedded in mortar to form columns, walls, arches or vaults (Figure 3.1). The range of different types of masonry is large due to the variety of types of constituent. Bricks may be of fired clay, baked earth, concrete, or a range of similar materials, and blocks, which are simply very large bricks, can be similarly composed. Stone too is not one but a very wide range of materials from the relatively soft sedimentary rocks such as limestone to the very hard granites and other igneous rocks. These ‘solid’ units can be used in conjunction Figure 3.2 Kharraqan Towers, Qazvin, Iran, C11 CE. These 15 m-high late-medieval brickwork structures demonstrate one of the advantages of masonry, which is that very large constructions with complex geometries can be achieved by relatively simple building processes. Photo: Zereshk/Wikimedia Commons. 42 STRUCTURAL MATERIALS with a variety of different mortars to produce a range of masonry types. All have certain properties in common and therefore produce similar types of structural element. Other materials such as dried mud, pisé or even unrein- forced concrete have similar properties and can be used to make similar types of structural element. The fact that masonry structures are composed of very small basic units makes their construction straightforward. They are most commonly used in small-scale structural typologies as in Figures 1.12 and 1.13. Complex geometries can be produced relatively easily, however, without the need for sophisticated plant or techniques and very large structures can be built by these simple means (Figures 0.3, 3.2, 11.13 and 11.14). The only significant constructional drawback of masonry is that horizontal-span structures such as arches and vaults require temporary support until complete. The physical properties of masonry are moderate compressive strength, minimal tensile strength, relatively high density and high thermal capacity. The very low tensile strength restricts the use of masonry to elements in which the principal internal force is compressive. Where horizontal spans are involved, tension can be eliminated by the use of form-active arrangements. Where significant bending movement occurs in masonry elements, for example from side thrusts on walls caused by rafters or vaulted roof structures, from out-of-plane wind pressure on external walls or from the tendency of com- pressive elements to buckle, the level of tensile bending stress is kept low by increasing their thickness. This can give rise to very thick walls and columns and therefore to excessively large volumes of masonry unless some form of ‘improved’ cross-section is used (see Section 4.3). Traditional versions of this are buttressed walls: those of medieval cathedrals or the voided and sculptured walls that support the large vaulted enclosures of Roman Antiquity (Figures 7.4, 10.18 and 10.19) are among the most spectacular examples. In all of these the volume of masonry is small in relation to the total effective thickness of the wall concerned. The fin and diaphragm walls of recent tall single-storey masonry buildings (Figure 3.3) are modern equivalents. Due to the strength characteristics outlined above, the volume of material in a masonry structure is often relatively large and produces walls and vaults that can act as effective thermal, acoustic and weather-tight barriers and also as reservoirs of heat. Other attributes of masonry-type materials are that they are durable, and can be left exposed in both the interiors and exteriors of buildings. They are also, in most parts of the world, available locally in some form and do not therefore require to be transported over long distances. This, together with the fact that brick and block manufacture is generally a low- energy process, gives masonry a relatively small ecological footprint. Masonry is also a material that can be relatively easily recycled. Bricks, stones and blocks can be recovered from demolition processes and even where damaged can be re-used as aggregate for concrete or other building purposes. All of these characteristics make masonry an environmentally-friendly material whose  STRUCTURAL MATERIALS 43 Figure 3.3 Where masonry will be subjected to (a) significant bending moment, as in the case of external walls exposed to wind loading, the overall thickness must be large enough to ensure that the tensile bending stress is not greater than the compressive stress caused by the gravitational load. The wall need not be solid, however, and a selection of techniques for achieving thickness efficiently is shown here. (b) (c) use must be expected to increase in future as the demand for sustainable forms of building become more pressing. It is likely, therefore, that the use of masonry will increase and that it will be ‘reclaimed’ as a structural material that is suitable for large-scale buildings. This, in turn, is likely to have a significant effect on architectural style as the development of overall forms that are compatible with masonry construction are rediscovered and greater emphasis is placed on integrative design in which structure, environmental control and architectural style are evolved together. 3.3 Timber Timber is, with masonry, one of the traditional structural materials but its strength properties are far superior as it can resist tension and compression with equal facility and therefore also bending. It is also a lightweight material with a high ratio of strength to weight. As with masonry, its use involves the acceptance of certain restrictions, such as those imposed by the forms in which it becomes available. It can, however, be relatively easily joined together which allows the build up of large structures, principally in the form 44 STRUCTURAL MATERIALS Figure 3.4 LeMay Museum, Tacoma, 2012; Large Architecture, architects. The principal structural elements here are laminated timber frameworks (1.3 m x 22 mm in cross- section) which span 32 m across the rectangular-plan interior. Photo: Zheng Zhou/Wikimedia Commons. of trussed arrangements, though the structurally weak nature of the joints imposes restrictions that have to be respected. Timber has nevertheless been used to create tall and also wide-span enclosures (Figures 3.4 and 11.5, 11.6). Of the four principal structural materials timber is the only one that is sourced from raw material that is renewable and potentially inexhaustible provided that the forests from which it is extracted are appropriately managed. Timber components can also be re-used, if carefully removed from obsolete buildings, and are suitable for various forms of recycling. It is therefore likely that the importance of timber, as a structural material, will increase with the need to develop sustainable forms of architecture. As with masonry this will require that the vocabulary of timber forms be widened from the fairly limited range that has characterised its use in the Modern period to include more complex and efficient forms of structure (see Section 11.5), within the constraints imposed by the fundamental properties of the material. The fact of timber having been derived from a living organism is responsible for the nature of its physical properties. The material is composed of long fibrous cells aligned parallel to the original tree trunk and therefore to the grain that results from the annual growth cycles. The constituent elements are of low atomic weight, which is responsible for its low density, but the lightness in weight is also due to its cellular internal structure which produces member cross-sections that are permanently ‘improved’ (see Section 4.3).  STRUCTURAL MATERIALS 45 Figure 3.5 Timber connectors are used to increase the load-carrying capacity of bolted connections by reducing the concentration of stress. A selection of different types is shown here. Joint weakness is nevertheless a factor that limits the maximum size of built- up timber structures. Parallel to the grain the strength is approximately equal in tension and compression so that planks aligned with the grain can be used for elements that carry axial compression, axial tension or bending-type loads as noted above. Perpendicular to the grain it is much less strong because the fibres are easily crushed or pulled apart when subjected to compression or tension in this direction. This weakness perpendicular to the grain makes timber intolerant of the stress concentrations such as occur in the vicinity of mechanical fasteners such as bolts and screws and the difficulty of making satisfactory structural con- nections with mechanical fasteners (Figure 3.5) is a factor that limits the load carrying capacity of large-scale timber structures composed of many separate elements. The development, in the twentieth century, of structural glues for timber has to some extent solved the problem of stress concentration at joints but the curing of glue must normally be carried out under controlled conditions of temperature and relative humidity, which is impractical on building sites so that gluing has to be regarded as a pre-fabricating technique. Timber suffers from a phenomenon known as moisture movement. This arises because the precise dimensions of any piece of timber are dependent on its moisture content (the ratio of the weight of water that it contains and its dry weight, expressed as a percentage) which is affected by the relative humidity of the environment, and therefore subject to continuous change. 46 STRUCTURAL MATERIALS It can cause joints made with mechanical fasteners to work loose. The greatest change to the moisture content occurs following the felling of a tree after which it undergoes a reduction from a value of around 150% in the living tree to between 10% and 20%, which is the normal range for moisture content of timber in a structure. The controlled drying out of timber, to avoid damage caused by shrinkage during this phase, is known as seasoning, a process in which the timber is physically restrained to prevent the introduction of permanent twists and other distortions caused by differential shrinkage. The most basic timber elements are of sawn-timber, which is simply timber cut directly from a tree with little further processing other than seasoning, shaping and smoothing. These are normally relatively small (maximum length Figure 3.6 The all-timber house is a loadbearing-wall form of construction in which all of the structural elements in the walls, floors and roof are of timber. An internal wall of closely spaced sawn-timber elements is here shown supporting the upper floor of a two storey building. Photo: Angus J. Macdonald.  STRUCTURAL MATERIALS 47 around 6 m and maximum cross-section around 75 mm × 250 mm) partly because the maximum sizes of cross-section and length are governed by the size of the original tree but also due to the desirability of having small cross- sections for the seasoning process. Sawn-timber elements can be combined to form larger, composite elements such as trusses with nailed, screwed or bolted connections. The scale of structural assemblies is usually modest however due to both the small sizes of the constituent planks and to the difficulty of making good structural connections with mechanical fasteners (Figure 3.6). Timber is also available in the form of products that are manufactured by gluing small elements together in conditions of high quality control. They are intended to exploit the advantages of timber while at the same time min- imising the effects of its principal disadvantages, which are variability, dimensional instability, the restrictions in the sizes of individual components and anisotropic behaviour. Examples of timber products are laminated timber, composite boards such as plywood, and combinations of sawn timber and composite board (Figure 3.7). Laminated timber (Figure 3.7c) is a product in which elements with large rectangular cross-sections are built up by gluing together smaller solid timber elements of rectangular cross-section. The obvious advantage of the process is that it allows the manufacture of solid elements with much larger cross- sections than are possible in sawn timber. Very long elements are also possible because the constituent boards are jointed end-to-end by means of finger joints (Figure 3.8). The laminating process also allows the construction of elements that are tapered or have curved profiles. Arches (Figure 3.9) and portal frame elements (Figure 3.4) are examples of this. Composite boards are manufactured products composed of wood and glue. There are various types of these including plywood, blockboard and particle board, all of which are available in the form of thin sheets. The level of glue impregnation is high and this imparts good dimensional stability and reduces the extent to which anisotropic behaviour occurs. Most composite boards also have high resistance to splitting at areas of stress concentration around nails and screws. Composite boards are used as secondary components such as gusset plates in built-up timber structures. Another common use is as the web elements in composite beams of I- or rectangular-box sections in which the flanges are sawn timber (Figure 3.7b). Because timber possesses both tensile and compressive strength it can be used for structural elements that carry axial compression, axial tension and bending-type loads. Its most widespread application in architecture has been in buildings of domestic scale in which it has been used to make complete structural frameworks, and for the floors and roofs in post-and-beam loadbearing masonry structures. Rafters, floor beams, skeleton frames, trusses, built-up-beams of various kinds, arches, shells and folded forms have all been constructed in timber (Figures 3.4, 3.6 and 3.9 to 3.13). 48 STRUCTURAL MATERIALS Figure 3.7 The I-beam with the plywood web (b) and the laminated beam (c) are examples of manufactured timber products. These normally have better technical properties than plain sawn timber elements such as that (a) shown in (a). The high levels of glue impregnation in manufactured beams reduce dimensional instability, and major defects, such as knots, are removed from constituent sub-elements. (b) (c) Figure 3.8 ‘Finger’ joints allow the constituent boards of laminated timber elements to be produced in long lengths. They also make possible the cutting out of defects such as knots. Photo: TRADA. Timber is, therefore, a material that offers the designers of buildings a combination of properties that allow the creation of lightweight structures that are simple to construct. Its relatively low strength, the small sizes of the basic components and the difficulties associated with achieving good structural joints tend to limit the size of structure that is possible, however, and the majority of timber structures are small in scale with short spans and a small number of storeys. Currently, its most common application in architecture is in domestic building where it is used as a primary structural material either to form the entire structure of a building, as in timber wall-panel construction, or as the horizontal elements in loadbearing masonry structures. Its potential for use in larger structural typologies is, however, considerable (see Section 11.5). Figure 3.9 Olympic Oval, Richmond, Canada; CannonDesign, architects. The principal structural elements of this building, which received a LEED award for its many environmentally sustainable features, are composite arches of laminated timber and steel spanning 100 m. Photo: Duncan Rawlinson/Wikimedia Commons. Figure 3.10 Savill Building, Windsor, UK, 2006; Glen Howells Architects, architects; Buro Happold Engineering and Haskins Robinson Waters, engineers. The primary structural element in this building is a timber grid-shell, spanning 90 x 25 m, constructed from locally sourced larch and oak. Photo: oosoom/ wikimedia commons Figure 3.11 Savill Building, Windsor, UK, 2006; Glen Howells Architects, architects; Buro Happold Engineering and Haskins Robinson Waters, engineers. Individual elements consist of two 80 x 50 mm laths separated by 50 mm or 75 mm shear blocks. The 24 mm-thick plywood skin provides stiffening and is therefore part of the structure. The configuration is similar to that used in aircraft construction (see Figure 4.15) but the fabrication is ‘low tech’. It is an example of the type of innovation required for the creation of sustainable forms of building. Photo: Glen Howells Architects. Figure 3.12 Living Planet Centre, WWF UK, Woking, UK, 2013; Hopkins Architects, architects; Expedition Engineering, engineers. This vaulted timber roof structure spans 80 x 37.5 m. Photo: Morley von Sternberg.  STRUCTURAL MATERIALS 51 Figure 3.13 Living Planet Centre, WWF UK, Woking, UK, 2013; Hopkins Architects, architects; Expedition Engineering, engineers. Individual elements are of laminated timber and are ‘improved’ with built-in ‘lightening’ holes. Photo: Morley von Sternberg. 3.4 Steel Steel is the strongest of the commonly used structural materials, used for the tallest buildings and the longest spans. It has more-or-less equal strength in tension and compression and is therefore able to resist bending well. It can also be relatively easily jointed by welded or bolted connections. This combination of properties has allowed steel to be used in all types of structural configurations and, since its introduction at the beginning of the Modern period, has released architects from many of the constraints on form which had formerly been imposed by the limitations of the traditional structural materials. The glass-clad rectilinear steel skeleton framework has been one of the signature forms of Modern architecture and recent developments in steel fabrication technology has allowed its use to be extended to large-scale and very complex curvilinear shapes. Much of the free-form architecture of recent decades (see Figures 10.23 to 10.27) involves the use of inefficient semi-form- active structural configurations that would have required impossibly bulky elements for support were it not for the use of a very strong structural material. The great freedom of expression that has been enjoyed by certain ‘starchitects’ in the early twenty-first century has therefore been made possible largely by the advent of steel as a structural material that can now be fashioned into complex curvilinear forms. It is also a material that carries a very high environmental cost, with a large carbon footprint and high embodied energy. It can be relatively easily re-used or recycled but the energy input required for the latter is considerable. Its usefulness, in both structural and all other current applications, is such that a 52 STRUCTURAL MATERIALS significant reduction in society’s dependence on steel is unlikely in the near future, but the environmental costs are such that this will come under increasing scrutiny. Given the alternatives that are available for most building applications, the use of steel as a structural material in architecture is likely to decrease in the medium to long term and this, in turn, will increase the influence of structural design on the development of architectural style as forms are adopted that are more suited to environmentally friendly materials (see Section 11.5). The high strength and high density of steel favours its use in skeleton frame type structures in which the volume of the structure is low in relation to the total volume of the building that is supported but a limited range of slab-type formats are also used. An example of a structural slab-type element is the profiled floor deck in which a profiled steel deck is used in conjunction with concrete, or exceptionally timber (Figure 3.14), to form a composite structure. These have ‘improved’ corrugated cross-sections to ensure that adequate levels of efficiency are achieved. Deck units consisting of flat steel plate are uncommon. The shapes of steel elements are greatly influenced by the process that is used to form them – hot-rolling, cold-forming and casting. Hot-rolling is a primary shaping process in which massive red-hot billets of steel are rolled between several sets of profiled rollers. The cross-section of the original billet, which is normally cast from freshly manufactured steel and is usually around 0.5 m × 0.5 m square, is reduced by the rolling process to much smaller Figure 3.14 Hopkins House, London; dimensions and to a particular precise shape (Figure 3.15). The range of Michael and Patty cross-section shapes that are produced is very large and each group requires Hopkins, architects; its own set of finishing rollers. Elements that are intended for structural use Anthony Hunt Associates, structural engineers. The floor structure here consists of profiled steel sheeting that will support a timber deck. A more common configuration is for the profiled steel deck to act compositely with an in-situ concrete slab for which it serves as permanent formwork. The building also illustrates well the slender elements and low volume of structure that the great strength of steel makes possible and that is exploited here for visual effect. Photo: Pat Hunt.  STRUCTURAL MATERIALS 53 Figure 3.15 The heaviest steel sections are produced by a hot-rolling process in which billets of steel are shaped by profiled rollers. This results in elements that are straight, parallel-sided and of constant cross- section. These features must be taken into account by the designer when steel is used in building and the resulting restrictions in form accepted. Photo: Univac Consulting Engineers. have shapes in which the second moment of area (see Glossary) is high in relation to the total area (Figure 3.16). I- and H-shapes of cross-section are common for the large elements that form the beams and columns of structural frameworks. Channel and angle shapes are suitable for smaller elements such as secondary cladding supports and sub-elements in triangulated frameworks. Square, circular and rectangular hollow-sections are produced in a wide range of sizes, as are flat plates and solid bars of various thicknesses. Details of the dimensions and geometric properties of all the standard sections are listed in tables of section properties produced by steelwork manufacturers. The other method by which large quantities of steel components are manufactured is cold-forming. In this process thin, flat sheets of steel, which have been produced by the hot-rolling process, are folded or bent in the cold state to form structural cross-sections (Figure 3.17). The elements that result have similar characteristics to hot-rolled sections, in that they are parallel sided with constant cross-sections, but the thickness of the metal is much less so that they are much lighter but have lower load-carrying capacities. The process allows more complicated shapes of cross-section to be achieved however. Another difference from hot-rolling is that the manufacturing equip- ment is much simpler than that used for hot-rolling and can produce tailor- made cross-sections for specific applications. Due to their lower carrying capacities cold-formed sections are used principally for secondary elements in roof structures, such as purlins, and for cladding support systems. Structural steel components can also be produced by casting, in which case very complex tailor-made shapes are possible. The technique is problematic when used for structural components, however, due to the difficulty of ensuring that the castings are sound and of consistent quality throughout. In the early 54 STRUCTURAL MATERIALS Figure 3.16 Hot-rolled steel elements. Figure 3.17 Cold-formed sections are formed from thin steel sheet. A greater variety of cross-section shapes is possible than with the hot-rolling process. years of ferrous metal structures in the nineteenth century, when casting was widely used, many structural failures occurred – most notably that of the Tay Railway Bridge in Scotland in 1879 – and casting was discontinued as a method for shaping structural elements. The use of the technique for archi- tectural structures was revived in the late twentieth century, largely due to the development of systems for proving the soundness of castings, a spectacular early example being the semi-cantilevered ‘gerberette’ brackets in the Centre Pompidou (Figures 9.30 and 9.31). The development of weldable cast steel, largely in connection with the offshore oil industry, has allowed the technique to be used for complex jointing components in space frameworks (Figures 10.4 and 10.5). Most of the structural steelwork used in building consists of elements of the hot-rolled type and this has important consequences for the layout and overall form of the structures. A consequence of the rolling process is that the constituent elements are prismatic (straight, parallel-sided with constant cross-sections) and this tends to impose a regular, straight-sided format on the structural forms for which it is suitable (Figures 3.19 and 5.10 to 5.14). In recent years, however, methods have been developed for bending hot-rolled structural steel elements into curved profiles and this has extended the range of forms for which it can be used. Because steel structures are pre-fabricated, the design of the joints between the elements is an important aspect of the overall design that affects both the structural performance and the appearance of the frame. Joints are made either by bolting or by welding (Figure 3.18). Bolted joints are less effective for the transmission of load because bolt holes reduce the effective sizes of  STRUCTURAL MATERIALS 55 Figure 3.18 Joints in steelwork are normally made by a combination of bolting and welding. The welding is usually carried out in the fabricating workshop and the site joint is made by bolting. element cross-sections and give rise to stress concentrations. Bolted connec- tions can also be unsightly unless carefully detailed. Welded joints are neater and transmit load more effectively but the welding process is a highly skilled operation and requires that the components concerned be very carefully prepared and precisely aligned prior to the joint being made. For these reasons welding on building sites is normally avoided and steel structures are usually pre-fabricated by welding to be bolted together on site. The need to transport elements to the site restricts both the size and shape of individual components. Two problems associated with steel are its poor performance in fire, due to the loss of mechanical properties at relatively low temperatures, and its high chemical instability, which makes it susceptible to corrosion. Both of these have been overcome to some extent by the development of fireproof and corrosion protection materials, especially paints, but the exposure of steel structures, either internally, where fire must be considered, or externally, where durability is an issue, is always problematic. To sum up, steel is a very strong material with dependable properties. It is used principally in skeleton frame types of structures in which the components are hot-rolled. It allows the production of structures of a light, slender appear- ance and a feeling of neatness and high precision (Figure 3.19). It is also capable of producing very long span structures and structures of great height. The manufacturing process imposes certain restrictions on the forms of steel frames. Regular overall shapes produced from straight, parallel sided elements are the most favoured. 56 STRUCTURAL MATERIALS Figure 3.19 Spectrum building (formerly Renault Sales 3.5 Reinforced concrete Headquarters), Swindon, UK, 1983; Norman Foster Concrete, which is a composite of fragments of stone or other inert material Associates, architects; (aggregate) and cement binder, may be regarded as a kind of artificial masonry Ove Arup & Partners, because it has similar properties to stone and brick (high density, moderate structural engineers. compressive strength, minimal tensile strength). It is made by mixing together An example of steelwork used to create dry cement and aggregate in suitable proportions and then adding water, architectural effect as which causes the cement to hydrolyse and subsequently the whole mixture to well as to provide set and harden to form a substance with stone-like qualities. support. All principal Concrete has one considerable advantage over masonry which is that it is elements are standard available in semi-liquid form during the construction process and this has hot-rolled sections. three important consequences. First, it means that other materials can be Tapering of I-section beams is incorporated into it easily to augment its properties, the most important of achieved by cutting and these being steel in the form of thin reinforcing bars which give the resulting welding of the parallel- composite material (reinforced concrete – Figure 3.20) tensile and therefore sided originals. bending strength as well as compressive strength. Other materials, such as mineral fibre, plastics of various kinds, and fabric, can also be used as tensile reinforcement. Second, the availability of concrete in liquid form allows it to be cast into a wide variety of shapes. Third, the casting process allows very effective connections to be provided between elements and the resulting structural continuity greatly enhances the efficiency of the structure.  STRUCTURAL MATERIALS 57 Figure 3.20 In reinforced concrete, steel reinforcing bars are positioned in locations where tensile stress occurs. This combination of properties, and in particular the combination of high bending strength with mouldability, has allowed reinforced concrete to be used in a very wide range of forms. For the greater part of the Modern period it was the only material in which large-scale, free forms, involving semi-form- active arrangements, could be constructed. In these situations it was normally the difficulties of constructing the formwork on which the concrete would be cast rather than any constraints caused by the material itself that placed restrictions on the forms that were possible. Although concrete can be moulded into complicated shapes, relatively simple shapes are normally favoured for reasons of economy in construction (Figure 3.21). The majority of reinforced concrete structures are therefore post-and-beam arrangements (see Section 5.2) of straight beams and columns, with simple solid rectangular or circular cross-sections, supporting plane slabs of constant thickness. The formwork in which such structures are cast is simple to make and assemble and therefore inexpensive, and can be re-used repeatedly in the same building. These non-form-active arrangements (see Section 4.2) are relatively inefficient but are satisfactory where the spans are short (up to 6 m). Where longer spans are required more efficient ‘improved’ types of cross-section (see Section 4.3) and profile are adopted. The range of possibilities is large due to the mouldability of the material. Commonly used examples are coffered slabs and tapered beam profiles. The mouldability of concrete also makes possible the use of complex shapes and the inherent properties of the material are such that practically any shape is possible. Reinforced concrete has therefore been used for a very wide range of structural geometries. Examples of structures in which this has been exploited are the Wills, Faber & Dumas Building (Figure 1.6), where the mouldability of concrete and the level of structural continuity that it makes 58 STRUCTURAL MATERIALS Figure 3.21 Despite the mouldability of the material, reinforced concrete structures normally have a relatively simple form so as to economise on construction costs. The two most commonly used configurations for multi-storey buildings are shown here: two-way spanning flat slab (left) and beam-column frame (right). The structural armatures of multi- storey reinforced concrete structures are normally variations of one or other of these two generic forms. possible were used to produce a multi-storey structure of irregularly curved plan with floors that cantilevered beyond the perimeter columns, and the Lloyd’s Building, in London (Figs 10.6 to 10.10), in which an exposed concrete frame was given great prominence and detailed to express the structural nature of its function. In recent years the sculptural qualities of reinforced concrete have been exploited in the design of complex structural armatures for large building complexes so as to incorporate features that improve their environmental performance. The buildings of the Malaysian architect Ken Yeang, in which complex ‘green’ corridors are provided in configurations that spiral upwards through buildings, are examples (Figures 1.7 and 1.8). Sometimes the geometries that are adopted for concrete structures are selected for their high efficiency. Form-active shells for which reinforced concrete is ideally suited are examples of this (Figures 1.4 and 9.14). The efficiency of these is very high and spans of 100 m and more have been achieved with shells a few tens of millimetres in thickness. In other cases the high levels of structural continuity have made possible the creation of sculptured building forms that, though they may be expressive of architectural meanings, are not particularly sensible from a structural point of view. A well- known example of this is the roof of the chapel of Notre-Dame du Haut at Ronchamp by Le Corbusier (Figure 10.22), in which a highly individual and inefficient structural form is executed in reinforced concrete. At the time of its construction, it would have been impossible to make this form in any other structural material.  STRUCTURAL MATERIALS 59 3.6 Conclusion This chapter has reviewed the essential structural properties of the four principal structural materials and discussed their applications. In the Modern period they have been used principally in various post-and-beam forms as has suited both the architectural aspirations of Modernism and the economic climate of the age. The need, in future, to evolve building forms that are environmentally sustainable is likely to have a significant effect on the selection of the materials for the structural parts of buildings. This is likely to result in an expansion in the use of the traditional materials of masonry and timber for types of building, such as inner-city office, retail and housing complexes, in place of the more environmentally damaging materials of steel and reinforced concrete. This, in turn, will produce the need for a reconsideration of the forms and styles of ar