See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/289870606

Foundation engineering

Article in The Structural Engineer · July 2008

CITATIONS READS

2 11,048

1 author:

John Boscawen Burland

Imperial College London
161 PUBLICATIONS 11,596 CITATIONS

SEE PROFILE

All content following this page was uploaded by John Boscawen Burland on 17 August 2017.

The user has requested enhancement of the downloaded file.

Burland:Layout 1 10/7/08 11:35 Page 44

BURLAND

‘...difficulties with soils are

almost exclusively due, not
to the soils themselves, but
to the water contained in
their voids. On a planet
without any water there
would be no need for soil
mechanics’.

44 Centenary Issue 21 July 2008 The Institution of Structural Engineers

Burland:Layout 1 10/7/08 11:36 Page 45

BURLAND

Foundation engineering

Prof. John B. Synopsis by tier, each tier consisting of a single layer of blocks of uniform
This paper traces the development of the craft and science of thickness across the full width of the structure. Greater care was
Burland, CBE foundation engineering in the UK during the last 100 years also taken in dressing and placing the blocks.
FREng, FRS, PhD, drawing mainly on papers published in The Structural Engineer. Kerisel2 describes how in Mesopotamia, over a period of
MSc(Eng), DSc(Eng), The story is told of how Terzaghi coupled engineering geology three millennia, the art of building ziggurats was developed, in
CEng, FIStructE, FICE, with the science of soil mechanics to provide the necessary most cases on very weak alluvial soils. Because of the scarcity
FCGI rigour for modern geotechnical modelling and analysis. In the of stone these high massive structures were made of sun-baked
Professor, Dept of Civil 1960s two innovative construction techniques combined to bricks laid out in successive courses. As construction proceeded
Engineering, Imperial transform foundation engineering world-wide – large diame- the underlying alluvium soon yielded under the weight causing
College London
ter bored piles and diaphragm walls. In recent years the use of the base to spread laterally. Work progressed very slowly with
numerical methods of geotechnical analysis has become wide- long pauses in between so that little by little the rate of settle-
spread. The method opens the way for the instructive model- ment and spreading diminished. Eventually it was possible to
ling of complex soil-structure interaction problems provided build a small temple at the top (Fig 1). Around 2100 BC the
Keywords: Foundations, History,
Journals, Structural Engineer, Soil
the inherent uncertainties are recognised. However, easy access Sumerians began to place thick layers of woven reeds every six
mechanics, Karl von Terzaghi, to the computer packages poses grave dangers of inappropri- to eight courses of sun-dried brickwork. In this way the hori-
Biographies, Structural engineering ate use of the methods. The risks and hazards posed by the zontal tensions caused by the tendency of the foundations to
ground are discussed and it is concluded that their effective spread were resisted. As a result ziggurats could be built with
© J. B. Burland CBE
management requires well planned and executed site investi- nearly sheer sides and massive temples on top. This innovation
gations for all structures and the structural engineer has a key is often cited as the earliest example of reinforced earth.
role to play. Two early examples of successful foundation engineering in
China are given by Kerisel2. The elegant early 7th Century
Introduction Zhaozhou (otherwise known as Anchi) Bridge (Figs 2 and 3) is
Foundation engineering is as old as the art of building and, like founded on clay which was treated by digging it out beneath
building, it developed largely on the basis of accumulated expe- the abutments and recompacting it in layers interspersed with
rience and empirical procedures. Because ground conditions compacted layers of broken bricks. The late 10th Century 44m
vary so much from one locality to another, foundation practice high Pagoda of Longhua is founded on a thick layer of soft clay
varied widely. Moreover extrapolation of experience from one extending to a depth of about 30m. The foundations are of brick
locality to another was fraught with uncertainty. laid on a wooden raft which in turn rests on wooden piles
Parry1 describes how the ancient Egyptians learned from the driven at very close spacing – perhaps one of the earliest exam-
foundation failure of the South Dahshur Pyramid. This was built ples of a piled raft. The foundations remain unchanged since
during the reign of Pharaoh Snofru (2575-2551 BC) using a they were constructed over 1000 years ago.
central core supported by a series of inclined buttress walls, as
had become the tradition. The pyramid was founded on a clay Development of the subject in the early 20th
layer. The large differential foundation movements resulted in Century
significant structural distress in the tomb chambers and their Many of the early papers published in The Structural Engineer
access passages. As a result the pyramid was finished off with describe various techniques of foundation construction.
a considerably reduced slope of the upper portions giving rise McCarthy3 describes the wide range of sheet pile sections and
to the structure being named ‘The Bent Pyramid’. Abandoning pile types that were available at the time for the construction
the concept of buttress walls, future pyramids were built up tier of excavations, coffer dams and foundations. Brooke-Bradley4

Opposite: Montage showing ground conditions beneath the leaning tower of Pisa (Courtesy Dr David Edwards whilst at Imperial College
London) Fig 1. Construction of an early ziggurat. (1) Fill, (2) soft alluvium, (3) Temenos – sacred enclosure (Kerisel2). / Fig 2. Zhaozhou
Bridge – early 7th Century (Kerisel2).

The Institution of Structural Engineers Centenary Issue 21 July 2008 45

Burland:Layout 1 10/7/08 11:36 Page 46

BURLAND

published a major paper on bridge foundations the contents of which was are described of obtaining undisturbed soil samples and of measuring their
spread over five issues. The paper describes the wide range of foundations and compressibility and their permeability in the laboratory. The paper goes on to
abutments that were in common use at the time. These and other papers make explain how, making use of the stress changes beneath a foundation using the
little reference to the mechanical properties of the ground and how its response Boussinesq elastic theory, it was now possible to make reasonable estimates of
can be assessed. For example Brooke-Bradley4 states that: the magnitude and distribution of settlement across a building together with its
‘If the bearing power of sub-soil should prove to be inadequate to carry the time dependence.
proposed loads, it must be artificially strengthened’.
Methods of doing this are then described together with the various types of The birth of soil mechanics
piles available for this purpose. No where does one find how the ‘bearing power’ Because of his work in developing the theoretical basis of soil mechanics and
of the ground can be assessed in the first place. It is also stated that ‘all settle- foundation engineering, Terzaghi is often regarded as essentially a theoretician.
ment should be avoided if possible’, examples are given of damaging settlement Nothing could be further from the truth. In the Centenary year of this Institution
but little guidance is given on how it could be estimated. it is relevant to reflect on his struggles to develop the discipline, the art and the
In their paper marking the 50th Anniversary of The Structural Engineer, Bowen science of geotechnical engineering as described by Burland13.
and Measor5 refer to the outstanding contribution made at this time by Hiley on Karl von Terzaghi (Fig 4) was born in Prague in 1883. Though he read mechan-
the driving of piles. In 1925 he published a paper in Engineering on a ‘Rational ical engineering at the Technical University of Graz he was much more interested
Pile-driving Formula’ and his experience in the use of this formula appeared in in geology. He switched to civil engineering and went to work for a firm special-
his classic paper ‘Pile-driving calculations, with notes on Driving Forces and ising in hydroelectric power generation. Although his main activity was in the
ground resistance’ which was published in The Structural Engineer in 19306. design of reinforced concrete, the planning of the structures was of course inti-
In the early issues of The Structural Engineer some space is given to the mately involved with geology. But frequently he found the guidance of expert
design and construction of retaining walls. In 1915 Wentworth-Shields7 read a geologists unhelpful. He encountered many cases of failure – mainly due to lack
paper to The Concrete Institute on ‘The Stability of Quay Walls on Earth of ability to predict and control groundwater – piping failures were abundant.
Foundations’. He opens with the following memorable statement: Also slope failures, bearing capacity failures and excessive settlements.
‘In spite of the large amount of experience which has been gained in the Recognising the obvious influence of geological factors he concluded that it
construction of quay walls, it is still one of the most difficult problems in engi- was necessary to collect as many case records as possible so as to correlate fail-
neering to design a wall on an earth foundation with confidence that it will be ures with geological conditions. It is well known that he then spent 2 intense
stable when completed. . . . Even if the designer of such a wall is assured that years (1912 – 1914) in the western United States observing and recording. Two
it will stand, he cannot with any confidence tell you what factor of safety it years that ended in disillusion and depression. The following quote from his
possesses’. Presidential Address to the 4th International Conference on Soil Mechanics and
In the paper he summarising the current methods of estimating horizontal Foundation Engineering sums up his mood at that time14:
earth pressures. Much use is made of the approach developed by Arthur Langtry ‘At the end of two years I took my bulky collection of data back to Europe,
Bell8 for clay soils. In essence this approach extends Rankine’s classic theory for but when I started separating the wheat from the chaff I realised with dismay
dry granular material possessing no cohesion, to clay soils possessing both fric- that there was practically no wheat. The net result of two years of hard labour
tion and cohesion, although Langtry Bell did not use these terms. At that time was so disappointing that it was not even worth publishing it’.
the angle of shearing resistance ϕ was defined as the angle of` repose. Based So much for geology on its own. So much for precedent and case histories on
on some laboratory tests in a type of shear box, Bell lists values of k (cohesion) their own. To quote Goodman15, the problem lay in the fact that:
and α (angle of friction) for a variety of clays ranging from very soft puddle clay ‘. . . . the names geologists give to different rocks and sediments have devel-
to very stiff boulder clay and demonstrates their use in earth pressure calcula- oped mainly from a scientific curiosity about the geologic origin of these mate-
tions. rials, whereas Terzaghi was aiming towards discerning the differences in their
In 1928 Moncrieff9 published a major paper in The Structural Engineer on engineering properties’.
earth pressure theories in relation to engineering practice. He summarises the Shortly after his appointment to the Royal Ottoman Engineering University in
various approaches to calculating earth pressures from Coulomb (1773) and Constantinople in 1916 Terzaghi began to search the literature for insights into
Poncelet (1840) through Rankine (1856), Resal (1910) to Langtry Bell (1915). the mechanical behaviour of the ground. He became increasingly frustrated. What
In this paper he refers to the difficulty of determining the angle of repose for he witnessed was a steady decline from 1880 in recorded observations and
clayey soils. He cites a cutting in clay in which the side slopes varied from verti- descriptions of behaviour. This was replaced by myriads of theories postulated
cal to 1 vertical in 1½ horizontal while in parts the clay was ‘running down like and published without adequate supporting evidence. This experience must have
porridge’. been uppermost in his mind when, in his Presidential Address to the First
It is all too clear from these early papers that, in spite of significant, even International Conference on Soil Mechanics and Foundation Engineering16 he
heroic, engineering achievements in the construction of major foundations, stated the following:
retaining structures and tunnels, there was little understanding of the factors that ‘In pure science a very sharp distinction is made between hypothesis, theo-
control the mechanical behaviour of soil in terms of its strength and stiffness. ries, and laws. The difference between these three categories resides exclusively
Moreover there is almost no reference to the influence of ground water on in the weight of sustaining evidence. On the other hand, in foundation and earth-
strength, stability or earth pressures. It is hardly surprising that there were work engineering, everything is called a theory after it appears in print, and if
frequent failures, particularly of slopes and retaining walls. the theory finds its way into a text book, many readers are inclined to consider
Two papers in The Structural Engineer in particular mark the beginning of a it a law’.
change. In 1933 Sandover10, in a paper on ‘Foundations’, makes reference to the Thus Terzaghi was emphasising the enormous importance of assembling and
importance of obtaining as much information as possible about the nature of the examining factual evidence to support empirical procedures. He was also bring-
underlying strata and their ‘bearing power’ that will have to support the load. ing out the importance of instilling rigour. This is often equated with mathematics
He advocates the use of boreholes and trial pits. Moreover he makes reference but there is at least as much rigour in observing and recording physical phenom-
to the new science of soil mechanics in explaining those factors that control the ena, developing logical argument and setting these out on paper clearly and
magnitude and rate of settlement of a foundation. He briefly cites the work of precisely.
Terzaghi and his book Erdbaumechanik11, a copy of which was available in the In 1918 Terzaghi began to carry out experiments on forces against retaining
Institution Library. walls. He then moved on to piping phenomena and seepage beneath embank-
Then in 1934 the topic of settlement of structures on clay soils is discussed ment dams. Over this period Terzaghi came to realise that geology could not
in detail by Tschebotareff12. He makes detailed reference to the new science of become a reliable and helpful tool for engineers unless and until the mechani-
soil mechanics and its application in the prediction of the settlement. Methods cal behaviour of the ground could be quantified – this required systematic exper-

46 Centenary Issue 21 July 2008 The Institution of Structural Engineers

Burland:Layout 1 10/7/08 11:36 Page 47

BURLAND

3 4

imentation. On a day in March 1919, and on a single sheet of paper, he wrote Towards the end of his lecture he made the following important assertion:
down a list of experiments which would have to be performed. ‘Experience alone leads to a mass of incoherent facts. But theory alone is
Terzaghi then entered an intense period of experimental work in which he equally worthless in the field of foundation engineering, because there are too
carried out oedometer (confined compression) tests and shear tests on clays and many factors whose relative importance can be learned only from experience’.
sands. From these tests he developed his physical understanding of the princi- On 2 May 1939 Terzaghi delivered the 45th James Forrest Lecture at the
pal of effective stress (often referred to as the corner stone of soil mechanics), Institution of Civil Engineers, London with the title ‘Soil Mechanics – A New
excess pore water pressures and time-rate of consolidation – this was the birth Chapter in Engineering Science’19. The lecture summarised in simple terms the
of soil mechanics. To make headway with modelling the consolidation phenom- basic elements of the discipline of soil mechanics and its application to a number
enon analytically he turned to the mathematics of heat conduction. We see here of engineering problems ranging from earth pressure against retaining walls, the
the interplay between experiment and analytical modelling. failure of earth dams due to piping through to the phenomenon of consolida-
tion and the settlement of foundations. Early on in the lecture Terzaghi made the
Impact of soil mechanics on structural engineering memorable statement that:
In 1933, a Soil Physics Section was established at the Building Research Station ‘. . , in engineering practice difficulties with soils are almost exclusively due,
in the UK and Dr Leonard Cooling was put in charge of it. He set up the first not to the soils themselves, but to the water contained in their voids. On a planet
proper soil mechanics laboratory in Britain, equipped with the apparatus neces- without any water there would be no need for soil mechanics’.
sary to classify soils, measure their basic mechanical properties and carry out He was a forceful and charismatic figure and this lecture made a very profound
sampling. By 1935 the first investigations of civil engineering problems had impact on the structural and civil engineers in the UK. Peter Dunican, Past
begun and the group was moved to the Engineering Division of BRS being President of IStructE, attended the lecture as a young man and told of how
renamed the Soil Mechanics Section. It was in August 1937 that the well known Terzaghi had electrified the audience. Many leading geotechnical engineers,
Chingford embankment dam failure occurred and the team from BRS carried out including the late Sir Alec Skempton, emphasise what a pivotal role this lecture
the investigation. Terzaghi was called in to re-design the embankment and the played in the development of soil mechanics in the UK. As with his earlier lecture
necessary testing and analysis was carried out at BRS. This gave great impetus to the IStructE, Terzaghi emphasised very strongly the importance of retaining a
to the acceptance of soil mechanics as a key discipline in civil engineering in the balance between theory and practice in soil mechanics.
UK. We have seen that before Terzaghi’s work, foundation engineering was essen-
On 6 December 1934 Terzaghi delivered a lecture before IStructE in London tially a craft developed largely on experience but with little understanding of the
with the title ‘The actual factor of safety in foundations’17. He illustrated his physical factors that determine the properties of the ground and almost no
lecture with a large number of case histories of measured distributions of settle- means of measuring those properties experimentally. By the late 1940s and early
ment across buildings and their variation with time. He was able to explain the 1950s all that had changed and this change was reflected in the papers in The
broad features of behaviour using the basic principles of soil mechanics and foun- Structural Engineer. In particular Capper20 published a paper on ‘Soil Mechanics
dation analysis, demonstrating how vital it is to establish the soil profile with in Relation to Structural Engineering’ which is a model of clarity.
depth and across the plan area of the building. Even so, he showed that local However there were, and still are, misunderstandings about the application
variations in soil properties and stratification make it impossible to predict the of soil mechanics to structural engineering. Gower Pimm21 published a deliber-
settlement patterns with any precision. Without actually using the term, he drew ately provocative, even mischievous, paper attacking soil mechanics as a valid
attention to the important concept of ground-structure interaction pointing out aid to foundation design. This paper, together with the extensive discussion of
that the structure of a building should not be treated in isolation from its foun- it makes interesting reading. The essence of his thesis is that the ground is too
dations. He even drew attention to the fact that reinforced concrete beams can variable to make meaningful calculations and samples are too disturbed and vari-
yield plastically without impairing the stability or appearance of a frame build- able to make meaningful measurements. It seems that geology is to be relied
ing provided the cracking is not excessive. It is of interest to note that, in their upon but not soil mechanics. When read alongside the story of Terzaghi’s efforts
seminal paper on the allowable settlement of buildings, Skempton and to place foundation engineering on a rational scientific basis, with geology as a
MacDonald18 drew extensively on the case histories provided by Terzaghi in this key supporting discipline and soil mechanics as the basis for understanding the
lecture. mechanical response of the ground, it is clear that Gower Pimm’s paper is a trav-

Fig 3. Zhaozhou Bridge / Fig 4. Karl von Terzaghi (By kind permission of the Norwegian Geotechnical Institute)

The Institution of Structural Engineers Centenary Issue 21 July 2008 47

Burland:Layout 1 10/7/08 11:36 Page 48

BURLAND

5 6

esty of Terzaghi’s contribution. But a careful reading of the paper suggests that drilling a shaft of 2m or more in diameter and over 30m deep. When the shaft
what was really upsetting Gower Pimm was the fact that, in their zeal for the has been sunk to its full depth the bottom can be belled out to a diameter of
application of the new found science, some exponents of soil mechanics were 5m or more. The authors state that they ‘. . . believe that the introduction of cylin-
overstating its ability to make precise predictions of behaviour. This is still the case der piles in the construction of heavy foundations is an important development
today. As emphasised previously, Terzaghi never lost an opportunity to stress that in foundation engineering practice and that it will have a wide field of applica-
in most situations high precision of prediction is not possible because of the tion in the future’.
inherent uncertainties. One of the earliest projects on which large diameter bored piles were used
As an aid to communication between structural and geotechnical engineers, was for the Shell Centre on the South Bank of the Thames23, 24. Their use spread
Burland22 recently attempted to explain the key aspects that have to be consid- rapidly. Very soon debates began to rage as to the mechanisms of load transfer
ered in tackling a ground engineering problem drawing on the views expressed from the pile shaft and base to the ground. Experience with small diameter bored
by Terzaghi. Geotechnical modelling involves at least four distinct but inter- piles showed that the settlement of a single pile under working load was very
linked activities, each having their own rigour: small. But load testing carried out on these large diameter underreamed piles
• The establishment of the ground profile including the groundwater conditions. showed much larger settlements under working load. In a paper published in The
• nterpretation of the measured or observed behaviour of the ground both in Structural Engineer in 1960 Kirkland25 expressed the view that the presence of
the laboratory and/or in the field. an underream reduced substantially the skin friction carried by the shaft. In the
• Prediction using appropriately idealised physical or analytical models. discussion that followed divergent views were expressed but it was evident that
• The application of empirical procedures and judgment based on case histo- many, including district surveyors in London, were very unsure as to how such
ries and relevant experience. piles should be designed. In 1962 the results of two large scale pile test were
reported by Frischman and Fleming26 in The Structural Engineer. The discussion
The boundaries between these four activities often become confused and one to this paper revealed that a number of key questions still remained un-
or more of them is often completely neglected. The first three activities may be answered.
depicted as forming the apexes of a triangle with well-winnowed experience In June 1960 a group of piling contractors approached the Building Research
occupying the centre as shown in Fig 5. The paper went on to demonstrate that Station to see if field experiments at full-scale could be put in hand to resolve
there are very close similarities between the approach followed routinely by the the many design problems of large diameter bored piles. A major research
geotechnical engineer and that followed by a structural engineer working on an programme was undertaken jointly funded by industry and the BRS, the results
historical building. This analogy is helpful in gaining an understanding of the of which were published by Whitaker and Cooke27 in the Conference on Large
approach of the geotechnical engineer to analysis and design. Bored Piles held in London in February 1966. Thirteen full-scale tests were
carried out on straight shafted and underreamed bored piles in London clay in
Some key developments in foundation engineering which the load carried by the base was measured by means of a specially
The development of soil mechanics during the first half of the 20th Century designed load cell. Every pile was first loaded in increments up to about twice
provided the rigour to foundation engineering that had been missing up till that the working load and then the pile was loaded to failure by means of a constant
time. During the second half of the 20th Century there have been major devel- rate of penetration test. The results of this major research programme have
opments in methods of construction that have completely transformed founda- proved of outstanding importance and have clarified the key mechanisms of
tion engineering. The two key ones took place in the 1950s and 1960s. The first behaviour of large diameter bored piles in stiff clay. As the pile is loaded the shaft
is the introduction of large diameter bored piles and the second is the technique resistance is fully mobilised at small settlements and thereafter remains approx-
of diaphragm wall construction. imately constant while the base resistance requires much larger movements to
mobilise, as illustrated in Fig 6. It became clear that the settlement and carry-
Large diameter bored piles ing capacity of large diameter bored piles could be explained in terms of basic
In their 50th Anniversary paper Bowen and Measor5 referred briefly to the recent soil mechanics principles and this left the way open for the development of`
introduction of large diameter bored piles to the UK. In particular they described rational design principles28, 29.
the system that originated in the USA that employs an auger and is capable of

Fig 5. The geotechnical triangle / Fig 6. Results of a load test on an instrumented underreamed bored pile in London Clay showing
ductile behaviour of the shaft27

48 Centenary Issue 21 July 2008 The Institution of Structural Engineers

Burland:Layout 1 10/7/08 11:36 Page 49

BURLAND

7 8 9

Diaphragm retaining walls ter bored piles and diaphragm walls, together with numerous innovative devel-
Dr Christian Veder invented the concept of the continuous concrete diaphragm opments in each, has completely transformed foundation engineering and under-
wall in 193930. The wall is formed in a trench which itself is supported by drilling ground construction world-wide over the last 30–40 years.
mud during excavation. The idea remained dormant during the Second World
War. It was not until 1951 that it was used for the first time for the construction Foundation–structure interaction and the advent of numerical
of a deep watertight cut-off wall 1500m long at Venafro, near Naples31. analysis
Thereafter the use of concrete diaphragm walls spread spectacularly. The first As mentioned previously, Terzaghi, in his seminal 1934 lecture to IStructE17,
application in Britain appears to be for the construction of the retaining walls drew attention to the effects of foundation structure interaction. In 1947
for Hyde Park Corner underpass in 1961. Meyerhof36 obtained results for simple frames founded on pad footings under-
In its simplest form the process consists of excavating a trench, or slot, in short going differential settlements. He showed that the stresses induced by differen-
panels, the sides being supported by bentonite slurry (Fig 6). A reinforcement tial settlements can be significant. The paper illustrates how complex such an
cage is then lowered into the trench followed by tremied concrete which interactive analysis can be. Prior to the 1970s design practice tended to consider
displaces the bentonite slurry (Fig 7). Work on the adjoining panel then the ground and the structure in relative isolation. In 1971 the Institution formed
commences with successive panels following on progressively. a Special Study Group to investigate the need for recognition to be given to inter-
The rapid adoption and growth of the method can be attributed to a number action effects and in 1978 an Institution Report was published on Structure-soil
of factors. The technique has a wide range of applications including cut-off interaction under the Chairmanship of Sam Thorburn, Past President. In 1989 this
walls, earth retaining walls for a variety of applications such as underpasses, deep Report was revised and extended to include a wide range of structures in addi-
basements, underground stations, tunnels, docks and pump houses. Diaphragm tion to buildings37.
elements can be formed in a variety of plan shapes including cruciforms and T- The above Report deals with important issues such as allowable total and
shapes. It has been demonstrated that, properly installed, the elements can carry differential settlements, serviceability limits, and the classification of damage. A
significant vertical loads both in end bearing and side friction. Therefore, when clear and important distinction is drawn between the ‘general’ approach to soil-
used as retaining walls, they allow effective transfer of vertical load from the structure interaction and a ‘detailed analysis’. A ‘general’ approach entails repre-
superstructure to the subsoil. Also they are frequently used as foundation senting the global stiffness of the structure in a simple way in order to assess its
elements in their own right as ‘barrettes’. influence on the total and differential movements induced in the structure as it
The technique is particularly suited for the urban environment as the noise interacts with the ground. For example, in assessing the impact of subsidence
levels and vibration are no more than that associated with normal civil engi- induced by tunnelling beneath a building, Potts and Addenbrook38 represent the
neering plant. Moreover ground movements associated with installation and well building as a simple beam. The stiffness of the structure modifies the form of
designed support can be kept to an acceptable level so that the impact on adja- subsidence movements induced by the tunnel by interacting with the stiffness
cent buildings and services can be minimised. Methods of reconditioning and of the ground. Simple modelling of this type can be very instructive and can lead
treating the bentonite slurry have been developed to improve quality control and to substantial reductions in differential settlements compared with ‘green field
eliminate risks of pollution. There have been rapid developments in trench exca- site’ values as confirmed by numerous field measurements on actual struc-
vation equipment so that a wide range of ground conditions can be tackled. tures39.
Perhaps the first project making use of diaphragm retaining walls to be A high order of sophistication is needed if ‘detailed analysis’ of forces and
described in The Structural Engineer is the Underground Car Park at the Palace stresses acting on the structural members due to soil-structure interaction is
of Westminster32 (Figs 8, 9). This was closely followed by a paper on the geot- required. The recent advent of powerful methods of numerical analysis and their
echnical aspects of the Barbican Arts Centre33. Since then numerous papers wide availability has made such analyses accessible to all. It is now possible to
have been published on projects in which this technique has been adopted. More obtain solutions to many complex problems which a few years ago would have
recently there have been papers describing the design of the deep box for been out of`reach. If used sensibly and with discernment these powerful numer-
Westminster Station, London34 and the application of barrettes as foundation ical methods can be of considerable assistance in enabling a designer to gain a
elements for a Mega Tower in Hong Kong35. feel for the behaviour of a soil-structure system. However, if used blindly, they
It is clear from these and other papers that the introduction of large diame- are a menace and can be extremely misleading.

Fig 7. Grab excavation equipment for diaphragm walls / Fig 8. Diaphragm wall cage during insertion into a slurry trench wall
Fig 9. Underground car park at the Palace of Westminster32

The Institution of Structural Engineers Centenary Issue 21 July 2008 49

Burland:Layout 1 10/7/08 11:36 Page 50

BURLAND

It is particularly relevant here to quote from Section 2.3 of the Institution’s methods of physics should be used to measure the mechanical properties of soils
Report on Soil-Structure Interaction37. After highlighting the many idealisations and rocks, so as to provide data for rational analysis and design. The geotech-
that have to be made in modelling soil-structure interaction it is concluded that: nical triangle illustrated in Fig 4 was developed from a study of Terzaghi’s
‘. . .even if engineers were in possession of unlimited analytical power the methods and it is significant that the ‘ground profile’ and its exploration is
uncertainties in the soil, the structure and the precise excavation/construction placed at the top with ‘observed behaviour’ in the form of laboratory or field
procedure are so great that precision in the prediction of behaviour would be testing at the lower left hand corner. These two aspects are essential to devel-
unlikely to improve significantly. Analysis is only one of the facilities required in oping an appropriate geotechnical model for analysis and can be summed up
designing for soil-structure interaction. In most circumstances the real value of succinctly by the following questions:
the analysis will be in assisting the engineer to place bounds on overall behav- • What is there and how well defined is it?
iour or in assessing the influence of various construction features e.g. a local stiff- • How and when did it get there?
ening because of a deep beam or a shear wall. Recognition of these inevitable • What are the mechanical properties and their likely range?
uncertainties can lead only to improved design’. • Are there any details that could have a critical influence on response?
The above statement echoes very strongly Terzaghi’s persistent message that It is the author’s opinion22 that these key questions that govern geotechnical
inherent variability of the ground make precise prediction of behaviour impos- site investigation are broadly the same as those that govern the approach of a
sible. Of far greater importance than illusory precision is that an appropriate structural engineer to an existing historical building or indeed to any existing
numerical analysis can provide a most valuable understanding of key mechanisms structure. To answer the questions adequately requires considerable expertise,
of behaviour. For example it is doubtful that the successful stabilisation of the ‘well-winnowed experience’ and costs money.
Leaning Tower of Pisa could have been achieved without first identifying the key Previous papers published in The Structural Engineer have drawn attention to
mechanisms governing the movements of the tower by means of a calibrated the importance of carrying out an adequate site investigation, the most notable
numerical model40. being Capper20 and McFarlane & Tomlinson43. Both of these papers are specif-
In April 2004 the temporary retaining system for a 30m deep excavation ically aimed at the practising structural engineer who has no specialist geot-
adjacent to the Nicoll Highway in Singapore collapsed13, 41 (Fig 10). It is sober- echnical expertise and both are still relevant today. The Introduction to Capper’s
ing to reflect that one of the prime causes of the collapse was identified by the paper, which places soil mechanics in the context of structural engineering, is well
Committee of Inquiry as: worth reading. One detects that it was written in part to counter the attack on
‘Use of an inappropriate soil simulation model which over-estimated the soil soil mechanics made a few years earlier by Pimm21. McFarlane and Tomlinson
strength at the accident site and (as a consequence) underestimated the forces stress the importance of appropriate site supervision emphasising that ‘the
on the retaining walls within the excavation’. employer gets what he pays for’. They also raise the important issue of compet-
With regard to numerical modelling in geotechnical design, The Committee itive tendering outlining the arguments in favour of a negotiated contract while
of Inquiry went on to recommend that: at the same time recognising that many public and government authorities
‘Generally, numerical analysis or modelling should not be over relied upon. It require competitive prices to be obtained for all types of work.
can only be used to supplement and not supplant sound engineering practice Recent years have seen a growing emphasis on the importance of recognis-
and judgement. It must be well undertaken by competent persons. Those who ing geotechnical hazards and the management of geotechnical risk. This
perform geotechnical numerical analysis must have a fundamental knowledge approach is aptly encapsulated by the phrase ‘Without site investigation ground
of soil mechanics principles and a clear understanding of numerical modelling is a hazard’ which is the title of a document44 written for clients and their profes-
and its limitations’. sional technical advisors in the construction industry. The document summarises
This wise recommendation is summed up in the pithy remark attributed to Dr briefly a number of case histories illustrating the consequences of inadequate site
Hugh Golder and which is quoted in the Institution’s Report on Soil-Structure investigation and the benefits of adequate site investigation. Inadequate site
Interaction: investigations can arise from a lack of awareness of the hazards associated with
‘Any design that relies for its success on precise analysis is a bad design’. ground, inadequate focus of finance, insufficient time and a lack of geotechni-
cal expertise. The document concludes that it is vital for financial decision makers
Site investigation to appreciate that you pay for a site investigation whether you have one or not,
Glossop42 has suggested that Terzaghi’s contribution is founded on two princi- and you are likely to pay considerably more if you do not. Hence a site investi-
ples: (1) that every ground engineering problem is basically geological requiring gation should be undertaken for every site, since without a properly procured,
an exploration of the geological structure (i.e. the ground profile) and ground supervised and interpreted site investigation, hazards which lie in the ground
water conditions at a site bearing in mind that it is often minor geological beneath the site cannot be ascertained.
details that can have a critical influence on engineering problems; and (2) that The findings of the above document were dramatically confirmed in a recent

Fig 10. The Nicoll Highway collapse, Singapore in April 2004

50 Centenary Issue 21 July 2008 The Institution of Structural Engineers

Burland:Layout 1 10/7/08 11:36 Page 51

BURLAND

survey carried out by the Federation of Piling Specialists and reported in The have previously been used for industrial or other purposes. As a consequence
Structural Engineer in a viewpoint article by Egan45. In a survey of over 220 proj- there will be an increasing use of ground improvement methods involving a
ects it was found that in a third of the cases the site investigation was so poor variety of novel geotechnical processes and new innovative equipment and
that optimum design of foundations or ground improvement could not be carried plant. With the world-wide trend towards urbanisation and the development of
out. In particular there was an almost total lack of preliminary geotechnical desk mega-cities there will also be an increase in the construction of underground
studies – a trend identified in the earlier report. It is concluded that the struc- infrastructure and the consequent need to protect existing buildings and serv-
tural engineer is uniquely placed to ensure that ground risk is properly managed ices. This will require advances in our understanding of ground-structure inter-
by (1) ensuring that the client is advised on the risks, (2) obtaining the right action. These two developments, ground improvement techniques and urban
specialist advice early in a project and (3) ensuring that the ground is adequately underground construction, will entail careful construction control by means of
investigated recognising that procurement of a site investigation on the basis of measurements and real-time monitoring. Significant developments in these tech-
price alone will almost inevitably lead to poor value. niques can be expected. The need for high quality site investigation and ground
exploration will remain and it is to be hoped that the benefits of improved
The future? management of ground risk will come to be more widely recognised and imple-
There will be a trend towards developing more sites on poor ground or ones that mented.

References

1. Parry, R. H.: Engineering the pyramids. Sutton Publishing Limited, 2004 diameter piles in London Clay’, The Structural Engineer, 40/4, p 123-131,
2. Kerisel, J.: Down to Earth. Foundations past and present: the invisible art of 1962
the builder. Balkema, Rotterdam, 1987 27. Whitaker, T. and Cooke, R. W.: ‘An investigation of the shaft and base resist-
3. McCarthy, M. J.: ‘Piling in the service of Structural Engineering’, The ances of large bored piles in London Clay’, Proc. Symp. on Large Bored Piles,
Structural Engineer, 5/12, p 369-380, 1927 p 7-49, Institution of Civil Engineers, London, 1966
4. Brooke-Bradley, H. E.: ‘Bridge Foundations – Parts I and II’, The Structural 28. Burland, J. B., Butler, F. G. and Dunican, P.: ‘The behaviour and design of
Engineer, 10/10, p 417-426, 1932; 11/12, p 508-521, 1933; 12/1, p 18- large diameter bored piles in stiff clay’, Proc. Symp. on Large Bored Piles, p
26; 12/2, p 96-105; 12/3, p 130-140, 1934 51-71, Institution of Civil Engineers, London, 1966
5. Bowen, F. M. and Measor, E. O.: ‘Foundations and sub-structures’, The 29. Skempton, A. W.: ‘Summing up’, Proc. Symp. on Large Bored Piles, p 155-
Structural Engineer, 36/13, 57-65, Jubilee Issue 1958 157, Institution of Civil Engineers, London, 1966
6. Hiley, A.: ‘Pile-driving calculations with notes on driving forces, and ground 30. Veder, C.: ‘Closing address’, Proc. Conf. on Diaphragm Walls and Anchorages,
resistance’, The Structural Engineer, 8/7, p 246-259; 8/8, p 278-288, 1930 p 221-222, Institution of Civil Engineers, London, 1974
7. Wentworth-Shields, F. E.: ‘The stability of quay walls on earth foundations’, 31. Veder, C.: ‘Excavation of trenches in the presence of bentonite suspensions
Proc. Conc. Inst., XX/2, p 173-222, 1915 for the construction of impermeable and load bearing diaphragms’, Proc.
8. Bell, A. L.: ‘The lateral pressure and resistance of clay and the supporting Symp. on Grouts and Drilling Muds in Engineering Practice, p 181-188,
power of clay foundations’. Min. Proc. Inst. C. E., 199, p 233-272, 1915 Butterworths, London, 1963
9. Moncrieff, J. M.: ‘Some earth pressure theories in relation to engineering 32. Burland, J. B. and Hancock, R. J. R.: ‘Underground car park at the House
practice’, The Structural Engineer, 6/3, p 59-84, 1928 of Commons, London: geotechnical aspects’, The Structural Engineer, 55/2,
10. Sandover, J. A. M.: ‘Foundations’, The Structural Engineer, 11/8, p 338-351, p 87-100, 1977
1933 33. Stevens, A., Corbett, B. O. and Steele, A. J.: ‘Barbican Arts Centre: the design
11. Terzaghi, K.: Erdbaumechanik auf bodenphysikalischer grundlage, Leipzig und and construction of the substructure’, The Structural Engineer, 55/11, p 473-
Wien: Franz Deuticke, 1925 485, 1977
12. Tschebotarioff, G.: ‘Modern methods of`clay analyses, forecasting settle- 34. Hankin, R.: ‘The design of Westminster Station (Jubilee Line extension)’, The
ments of structures, and allied problems’, The Structural Engineer, 12/8, p Structural Engineer, 78/19, p 19-22, 2000
350-360, 1934 35. Chan, G., Lui, J., Lam, K., Yin, K. K., Law, C. W., Lau, R., Chan, A. and Hasle,
13. Burland, J. B.: ‘Ground-structure interaction: designing for robustness’, R.: ‘Shaft grouted friction barrette piles for a super high-rise building’, The
Proc. IStructE Centenary Conf., Hong Kong, p 211-234, 2008 Structural Engineer, 82/20, p 32-36, 2004
14. Terzaghi, K.: ‘Presidentail address’, 4th Intl. Conf. Soil Mechanics and 36. Meyerhof, G.G.: ‘The settlement analysis of building frames’, The Structural
Foundation Engineering, London, Vol. 3, p 55-58, 1957 Engineer, 25/9, p 369-409, 1947
15. Goodman, R. E.: Karl Terzaghi, The Engineer as Artist, ASCE Press, Virginia, 37. Soil-structure interaction; the real behaviourof structures, The Institution of
USA, 1999 Structural Engineers, London, 1989
16. Terzaghi, K.: ‘Presidentail address’, 1st Int. Conf. Soil Mechanics and 38. Potts, D. M and Addenbrooke, T. L.: ‘The influence of and existing surface
Foundation Engineering, Harvard, Vol. 3, p 13-18, 1936 structure on ground movements due to tunnelling’, Proc. Instn Civ. Engrs,
17. Terzaghi, K.: ‘The actual factor of safety in foundations’, The Structural Geotechnical Engineering, 125, p 109-125, 1997
Engineer, 13/3, p126-160, 1935 39. Burland, J. B.: ‘Results of the research’. Chapter 21 in: Building response to
18. Skempton, A. W. and MacDonald, D. H.: ‘Allowable settlement of build- tunnelling. Case studies from the Jubilee Line Extension, London, Volume 1,
ings’, Proc. ICE, part 3, 5, p 727-784, 1956 Projects and methods. Burland, J. B., Standing, J. R. and Jardine, F. M. (eds):
19. Terzaghi, K.: ‘Soil Mechanics- a new chapter in engineering science’, Proc. CIRIA Special Publication 200, CIRIA and Thomas Telford, London, p 315-
Inst. Civ. Eng., 12, p 106-141, 1939 344, 2002
20. Capper, P. L.: ‘Soil mechanics in relation to structural engineering’, The 40. Potts, D. M.: ‘Numerical analysis: a virtual dream or practical reality’,
Structural Engineer, 31/2, p 47-61, 1953 Géotechnique, 53/6, p 535-573, 2003
21. Pimm, G. B, R.: ‘The limitations of soil mechanics as applied to foundations’, 41. Ministry of National Development, Singapore. (2005). Press release avail-
The Structural Engineer, 26/10, p 589-604, 1948 able from
22. Burland, J. B.: ‘Interaction between structural and geotechnical engineers’, http://www.mnd.gov.sg/Newsroom/newsreleasees/2005/news170505.ht
The Structural Engineer, 84/8, p 29-37, 2006. m. Accessed 3 November 2007
23. Williams, G. M. J.: ‘Design of the foundations of the Shell Building, London’, 42. Glossop, R.: Eighth Rankine Lecture, ‘The rise of geotechnology and its influ-
Proc. 4th Int. Conf. Soil Mechanics and Foundation Engineering, Vol. 1, p 457- ence on engineering practice’, Géotechnique, 18/2, p 107-150, 1968
461, 1957 43. McFarlane, I. H. and Tomlinson, M. J.: ‘Site investigations for structural foun-
24. Measor, E. O. and Williams, G. M. J.: ‘Features in the design and construc- dations’, The Structural Engineer, 52/2, p 57-65, 1974
tion of the Shell Centre, London’, Proc. Inst. Civ. Eng., 21, p 475-502, 1962 44. ‘Without site investigation ground is a hazard’, Site Investigation in
25. Kirkland, G. W.: ‘Tall buildings – problems in the design and construction Construction, Vol. 1, Thomas Telford, London, 1993
as affecting the structural engineer’, The Structural Engineer, 38/6, p 177- 45. Egan, D.: ‘The ground – how structural engineers can manage the risk’,
184, 1960 The Structural Engineer, 85/22, p 16-17, 2007
26. Frischmann, W. W. and Fleming, W. G. K.: ‘The use and behaviour of large

The Institution of Structural Engineers Centenary Issue 21 July 2008 51

View publication stats