TRL Application Guide AG39

] •H I G H W A Y S
AGENCY
C~
Specification for foamed concrete
by K C Brady, G R A Watts and M R Jones
Application Guide AG39

Prepared for Quality Services, Civil Engineering,
Highways Agency
K C Brady, G R A Watts and M R Jones
Application Guide AG39

First Published 2001
ISSN 1365-6929
Copyright TRL Limited 2001,
This report has been produced by TRL Limited, under/as part

of a Contract placed by the Highways Agency. Any views
expressed are not necessarily those of the Agency.
TRL is committed to optimising energy efficiency, reducing

waste and promoting recycling and re-use. In support of these
environmental goals, this report has been printed on recycled
paper, comprising 100% post-consumer waste, manufactured
using a TCF (totally chlorine free) process.
CONTENTS
Page
EXECUTIVE SUMMARY 1
ABSTRACT 2
1 INTRODUCTION 3
2 FOAMED CONCRETE 3
2.1 Definition, constituents and properties 3
2.2 History and background 4
2.3 Current usage 6
3 QUALITY CONTROL 6
3.1 Foam density and stability 7
3.2 Plastic density 7
3.3 Workability 7
3.4 Segregation 10
3.5 Cube strength 10
3.6 Soundness 10
4 SPECIFICATION 10
4.1 Documentation 11
4.2 Requirements 12
4.2.1 Essential items 12
4.2.2 Optional items 13
5 ACCEPTANCE CRITERIA 14
6 NON-CONFORMITY 15
7 SUMMARY 15
8 REFERENCES 16
APPENDIX A: GLOSSARY OF TERMS A-I
APPENDIX B: GUIDE FOR SPECIFICATION AND

QUALITY CONTROL OF FOAMED
CONCRETE A-2
APPENDIX C: FOAMED CONCRETE A-3
iii
EXECUTIVE SUMMARY
Despite the now relatively common use of foamed concrete as backfill to trench excavations, there is little published material
regarding its properties - in particular its long-term performance and durability. This paucity of data may be restricting the
wider use of foamed concrete for ground engineering and structural applications.
The aim of this project is to address this lack of information and to provide advice on the potential use and specification of
foamed concrete as backfill to substructures.
This report describes the nature of foamed concrete, its composition and properties and how it could be specified for use
in civil engineering works. Because the properties of foamed concrete can vary widely, and it can be used in a wide variety
of applications, it is important to define performance requirements for each case.
Current usage in the UK of foamed concrete is of the order of 250,000 tonnes per year. Despite this fairly significant tonnage
there is, as yet, no definitive guidance on how the material should be tested and specified for use. Furthermore there is a
paucity of information on some of its properties, particularly regarding its long-term performance. This situation might be
restricting the wider use of foamed concrete for ground engineering and structural applications.
The report provides a brief history of the development of foamed concrete and gives examples of contemporary uses. The
quality control tests that might be applied to foamed concrete are described; some of these differ from those used for normal
weight concrete. An example specification for foamed concrete is provided, which includes both mandatory and optional
requirements. Distinction is made between clauses for use with all applications and those required for specific applications.
Guidance on acceptance criteria and actions to be taken in the event of non-conformance is also provided.
Version No: 1 Application Guide A G39 Ocwber 2001

ABSTRACT
Foamed concrete has unique characteristics that can be exploited in civil engineering works. It requires no compaction, but
will flow readily from an outlet to fill restricted and irregular cavities, and it can be pumped over significant distances and
heights. Thus it could be thought of as a free-flowing, self-setting fill. This report provides a conspectus of foamed concrete:
covering its constituents, production, engineering properties and use.
Foamed concrete is simple to produce but, at present, there is a need to provide close control during its production and on-
site supervision during its placement and curing. The need for such special requirements will reduce as industry becomes
more familiar with the character and behaviour of the material. However, within the UK, there are no national specifications
for foamed concrete nor standard tests for measuring its workability, strength or other basic engineering properties.
Guidance on the specification and use of foamed concrete is provided in the report.
Research is still required on some aspects of the properties of foamed concrete - for example there is little published data
on its long-term properties. It seems likely that the increasing use of foamed concrete will lead to a better understanding of
its characteristics, and thus to improvements in methods of testing, specification and perhaps also in its engineering
properties.
Version No: 1 Application Guide A G39 October 2001
2
1 INTRODUCTION
This report describes the nature of foamed concrete, its composition and properties and how it could be specified for use
in civil engineering works. Because the properties of foamed concrete can vary widely, and it can be used in a wide variety
of applications, it is important to define performance requirements for each case.
Current usage in the UK of foamed concrete is of the order of 250,000 tonnes per year, see Pickford and Crompton (1996).
Despite this fairly significant tonnage there is, as yet, no definitive guidance on how the material should be tested and
specified for use. Furthermore there is a paucity of information on some of its properties, particularly regarding its long-
term performance. This situation might be restricting the wider use of foamed concrete for ground engineering and structural
applications.
The report provides a brief history of the development of foamed concrete and gives examples of contemporary uses. The
quality control tests that might be applied to foamed concrete are described; some of these differ from those used for normal
weight concrete. An example specification for foamed concrete is provided, which includes both mandatory and optional
requirements. Distinction is made between clauses for use with all applications and those required for specific applications.
Guidance on acceptance criteria and actions to be taken in the event of non-conformance is also provided.
The work described herein forms part of a research project undertaken by TRL Limited, on behalf of the Highways Agency
(HA), aimed at addressing the shortfall in information and thereby promoting the wider and proper use of foamed concrete.
Assistance in this project has been provided by the University of Dundee, British Cement Association (BCA) and Taywood
Engineering Ltd (TEL) acting under contract to the TRL.
This is the final report of this project; previous reports have addressed:
i materials, methods of production and applications of foamed concrete, Brady and Greene (1997);
ii use of foamed concrete as a backfilI to retaining structures, Watts and Brady (I 999);
iii performance of a masonry arch bridge with a foamed concrete backfill, Watts (2000); and
iv properties of foamed concrete, Brady (2000).
A glossary of terms is provided in Appendix A. A check list for the specification of foamed concrete is provided in Appendix B,
and details of its production, engineering properties and applications are provided in Appendix C.
2 FOAMED CONCRETE
2.1 DEFINITION, CONSTITUENTS AND PROPERTIES

Foamed concrete has been defined in several ways; indeed it has a number of synonyms - such as cellular concrete and
foamcrete and there is confusion in the early literature between foamed concrete and similar materials such as air entrained
concrete. A definition, cited by Van Deijk (1991), is that foamed concrete is a cementitious material having a minimum of
20 per cent (by volume) of mechanically entrained foam in the plastic mortar. This differentiates foamed concrete from (a)
gas or aerated concrete, where the bubbles are chemically formed through the reaction of aluminium powder with calcium
hydroxide and other alkalis released by cement hydration, and (b) air entrained concrete, which has a much lower volume
of entrained air.
For most common uses, the air content is typically between 40 and 80 per cent of the total volume. The bubbles vary in size
from around 0.1 to 1.5 mm in diameter but coalescence might produce voids considerably larger than this, particularly at
the top of pours. The typical appearance of foamed concrete can be seen in Figure 1.
Version No: 1 Application Guide AG39 October 2001

Figure 1 Typical nature of foamed concrete
Foamed concrete requires no compaction, and will flow readily from a pump outlet to fill restricted and irregular cavities:
it can be pumped successfully over significant heights and distances. The 28-day strength and dry density of the material
vary according to its composition, largely its air voids content, but usually they range from 1 to l 0 N/mm2 and from 400
to 1600 kg/m3respectively: lower strengths are associated with lower densities. The most commonly specified strength is
4 N/mm 2. (Strengths of up to 40 N/mm 2 have been produced, but so far this has been limited to laboratory-based research
work.) The plastic density of the material is about 150 to 200 kg/m3higher than its dry density, depending on curing regime
and specimen treatment.
Ordinary Portland cement (PC), to BS 12: 1996, is normally used as the binder, but other types of cement could be used.
Commonly, pulverized-fuel ash (PFA) is used as a partial replacement for PC or as a filler to partially or fully replace sand.
A fine sand, such as that conforming to BS 1200:1976, is normally used as bulk filler but sand to BS 882:1992 can also be
used. Subject to limitations on grading, waste materials such as granite dust and single-sized tailingscan also be used. Small
percentages of coarse aggregate, up to 14 mm in size, can also be added to increase the stiffness of the concrete. Admixtures
have been used to increase the rate of strength gain, particularly for highway trench reinstatement.
Foam is produced using specialised equipment, as shown in Figures 2 and 3. The type varies according to the volume of
the mix and required properties of the foam: further details are provided in Appendix C.
2.2 HISTORY A N D B A C K G R O U N D
Foamed concrete is not a particularly new material, its first patent and recorded use dates back to the early 1920s. According
to Sach and Seifert (1999), limited scale production began in 1923 and, according to Arasteh (1988), in 1924 Linde described
its production, properties and applications. The application of foamed concrete for construction works was not recognised
until the late 1970s, when it began to be used in the Netherlands for filling voids and for ground engineering applications.
Significant improvements in production methods and the quality of foaming agents over the last 15 years have led to
increased production and broadening of the range of applications. An extensive research programme carried out in Holland
helped promote foamed concrete as a building material, see Van Deijk (1991).
4
Figure 2 Small-scale laboratory foam production and typical nature of foam
Figure 3 Large-scale on-site foam production
In 1987, in response to the publication of the Home Report in 1985, a full-scale trial of the use of foamed concrete for trench
reinstatement was undertaken in the UK, details of which have been provided by Taylor (1990). The success of this trial
led to the widespread use of foamed concrete for trench reinstatements in the UK, and other applications followed. The
Highways Authorities and Utilities Committee's (HAUC) 'Specification for the Reinstatement of Openings in Highways'
(1992) recommended the use of foamed concrete within the New Road and Street Works Act (1991). Foamed concrete was
promoted as a replacement to granular fill for highway reinstatement works because of its following advantages:
5
i Ease of application and re-excavation (if necessary).

ii Elimination of compaction plant.
iii Low in-service settlement.
iv Load-distribution properties.
2.3 CURRENT USAGE

Over the past ten years or so, foamed concrete has been used on a large scale in the UK for bulk filling, trench reinstatements
and a variety of other applications: the following is not an exhaustive list:
i Bulk filling, using relatively low strength material, for redundant sewerage pipes, wells, disused cellars and
basements, storage tanks, tunnels and subways etc.
ii Highway trench reinstatement (although it should be noted that controlled low strength mortar is increasingly being
used because it is easier to produce by ready-mixed concrete producers).
iii Infill to the spandrel walls of arch bridges.
iv Backfill to retaining walls and bridge abutments.
v Stabilising soils, for example in the construction of embankment slopes.
vi Various industrial applications and for domestic housing, including as insulation to foundations and roof tiles,
blinding layers, cast in-situ piles, fire protection and high frequency sound insulation.
vii Sandwich fill for precast units.
viii Grouting for tunnel works.
3 QUALITY CONTROL
This section covers the tests that may be specified in the production of foamed concrete: it includes those that should be undertaken
by the supplier and those that might be by the Contractor or Client's representative. Foamed concrete should be sampled in
accordance with the draft BS EN 12350: Part 1: 2000. The most commonly used tests are listed and described as follows:
. Test carried out Tested by
Quality control test At point of delivery Laboratory Contractor Supplier
Foam: density '1 ¢" • ¢"

Foam: stability x ,/ x ¢"
Fresh concrete: plastic density ¢" x ,/" ¢"
Fresh concrete: workability of
base mix and/or foamed concrete • "¢" • ¢"
(several methods available)
Fresh concrete: segregation x ¢' ./" x
Hardened concrete: cube strength x -/ ,/ ¢"
Hardened concrete: soundness ,/ x ./ x
"/ Recommended for most uses and sites
x Not usually recommended, or impractical
• Might be useful for some particular applications or sites
6
3.1 FOAM DENSITY AND STABILITY

The properties of foamed concrete are critically dependent upon the quality of the foam. Measurements of foam density
should be taken as a matter of routine because density can vary according to the volume of the suffactant solution in the
containment vessel and also with the time of storage. Furthermore the level of expansion of the surfactant solution to foam
varies according to the type and details of the equipment, such as the type of the foaming gun, and with the valve settings
controlling the flow and feed pressure of the surfactant and air.
The density of a foam can be determined, quite simply, through weighing a known volume of foam - for example using a
glass measuring cylinder, as shown in Figure 4.
Foam stability can be quantified by measuring the volume of foam which has collapsed into solution at regular time intervals,
as described in Section C1.2.2.
Figure 4 Use of glass measuring cyclinder for determining density of foam
3.2 PLASTIC DENSITY

The plastic density of the base mix and the foamed concrete mix can be determined simply from the weight of a sample
in a container of known volume, of say 5 litres or so as shown in Figure 5. The method is described in BS EN 12350:
Part 6: 2000.
3.3 WORKABILITY
According to McGovem (2000), for most applications the slump of the base mix should be between about 75 and 100 ram.
The workability of the base mix could be assessed using a test developed for low-strength materials. As described by Brewer
(1996), workability can be quantified by the 'spreadability' of a 76.2 mm (3 inch) diameter, 152.4 mm (6 inch) long cylinder
of material as shown schematically in Figure 6. This could also be used to assess the workability of the foamed concrete mix.
7
Workability or 'flowability' can also be assessed from the efflux time of a litre sample through a modified Marsh cone: the
apparatus is shown in Figure 7.
From the tests on spreadability and fiowability, an empirical flow classification was derived by Dhir et al. (1999), and this
is reproduced in Table I.
Figure 5 Measurement of the plastic density of foamed concrete
(76.2 mm) 3"

I I
(152.4 rc
I I
Spread x
Figure 6 Test for the spreadability of the base mix and foamed concrete, from Brewer (1996)
8
152 m m
1.5 litre sample poured
ylindrica[ cone
305 mm
50 mm
Figure 7 Flow cone method for measuring the workability of foamed concrete, as used at Dundee University
Table I Classification of the workability of foamed concrete, from Dhir et al. (1999)
Flow rate, using apparatus

Main class shown in Figure 7 Sub-class Descriptionofflow
I 1 litre in < 1 minute At Constant flow

2 1 litre in > 1 minute B"1" Interrupted flow
3 0.5 litres < effiux < 1 litre C Completion of flow after gentle tamping
4 Effiux < 0.5 litres
5 No flow
t Used with Classes ] and 2 only
9
3.4 SEGREGATION
Segregation of foamed concrete in the fresh state can be detected through foam rising to the surface of the mix (noticeable
in a ready-mix truck or in recently poured concrete), or by the formation of a separate paste/sand mortar at the bottom of
the mixer (noticeable when mixing).
Segregation can be quantified through differences in the density of cores of hardened concrete taken from various depths,
or, as proposed by Dhir et al. (1999), by comparison of the oven-dry densities of 100 mm diameter, 25 mm thick slices
obtained from the top and bottom of a core.
3.5 CUBE STRENGTH

Compressive strength can be measured at 28 and/or 56 days essentially in accordance with BS 1881: Part 116: 1983. But,
because the strength of foamed concrete is relatively low, 150 mm cubes might be required to ensure reasonably accurate
measurements. Disposable polystyrene moulds are widely used for foamed concrete: these can be supplied with lids so that
a specimen can be left in the mould until immediately prior to testing. Steel moulds can be used but they should be lined
with a non-stick plastic film. Mould release oil should not be allowed to come into contact with the foamed concrete because
it can affect the properties of the concrete.
The specimens should be covered and, ideally, left undisturbed for least 3 days to prevent damage by movement and de-
moulding. After demoulding the cubes should be immediately wrapped in an air and watertight film ('cling film') and stored
at 20 +- 2°C in plastic bags (i.e. sealed curing).
3.6 SOUNDNESS
The 'soundness' or 'hardness' of the surface of foamed concrete can be used to assess whether it has developed sufficient
strength to allow additional lifts to be poured, or further site works to commence. The screed tester developed by the Building
Research Establishment (BRE), and as shown in Figure 8, can be used to assess the 'soundness' of a surface. The rig
comprises a weight that slides along a bar and which is allowed to fall freely onto an expansion ring connected to a 6 mm
diameter pin. In testing foamed concrete the penetration of a single drop of the weight should be measured (multiple drops
are normally used with other materials).
4 SPECIFICATION
At present in the UK there are no standard specifications for foamed concrete. Basic requirements for its use for trench
reinstatement exist (as for example in the New Road and Street Works Act, 1991)bat these are not necessarily satisfactory
for other applications.
It is recommended that the basic methods, clauses and format of the draft BS 8500:2000 and BS 5328 (various parts and
dates) are followed when specifying foamed concrete; but it is not possible to adopt all the clauses given in Parts 1 and 2
of the latter. The fundamental problem is that the requirements for coarse aggregate and a minimum strength of 7.5 N/mm 2
preclude the use of foamed concrete: the problem will persist with the replacement of those parts of BS 5328 by BS EN 206:
Part 1 and BS 8500 respectively.
10
Figure 8 View of BRE screed tester
4.1 DOCUMENTATION
Quality assurance
A producer must hold current product conformity certification, based on testing and surveillance, coupled with approval
of their quality system to BS EN ISO 9001:1994 by a certification body accredited by the Secretary of State (or equivalent).
This requires that the technical regulations of the accredited certification body are available for examination. It also requires
the producer to inform the purchaser of the status of the concrete plant at the time of tender, and immediately where there
is any change of status in the period between the time of tender and completion of supply.
Sampling point
Samples of foamed concrete, required for continuous monitoring of production, are required to be taken at the point of
discharge.
Acceptance testing
Acceptance testing by the purchaser is not normally necessary, but purchasers retain the right to carry out such testing.
Sampling, specimen making, curing and testing of fresh concrete shall be certified as conforming to the appropriate part of
the specification. Tests for cube strength shaIl be certified as conforming to the specification and shall be carried out by a
laboratory accredited for the test by the United Kingdom Accreditation Service (UKAS).
11
Deliver3" note
Before discharging a batch of foamed concrete at the point of delivery, the supplier is required to provide the purchaser with
a delivery ticket for the batch, on which is printed, stamped or written the information specified in Section 4.10.4 of BS 5328:
Part 3: 1990.
4.2 REQUIREMENTS
Foamed concrete should be specified as a designed mix. Carbon steel reinforcement should not be used in foamed concrete,
except where corrosion is precluded, or its effects are of no consequence. The specification for foamed concrete should
conform to Parts 1 and 2 of BS 5328:1997, except as defined herein or as agreed by all parties.
4.2.1 Essential items
Identification
The producer shall provide the information as defined in BS 5328: Part 3: 1990.
Permitted materials
Permitted cements should conform to the British Standards listed in Table 2: this is based on Table 1 of BS 5328: Part I :
1997, but note that cements containing slag should not be used unless the results of mixing trials are satisfactory.
Table 2 Types of cements suitable for production of foamed concrete, based on Table 1 of BS 5328: Part I: 1997
Type of bindefl" " Standard
Portland cements
Portland (PC) BS 12:1996
Sulfate-resisting (SRPC) BS 4027:1996
Cements containing pulverised-fuel ash (PFA) or limestone

Portland pulverised-fuel ash BS 6588:1996
Pozzolanic pulverised-fuel ash BS 6610: 199I
Portland limestone BS 7583:1996
Combinations made up in a concrete mixer Combination o f the proportions and properties to

from Portland cement and PFA conform to Clauses 6 to 9 of."
PC conforming to BS 12:1996 with BS 6588:I996 (except Clause 63)
PFA conforming to BS 3892: Part 1:1997 BS 6610:1996 (except Clause 6.2)
"l"Precluding cements containing slag.

* The notes to Table I in BS 5328: Part 1:1997 also apply in this case.
Permitted fine aggregates should conform to the British Standards listed in Table 3: this is based on Table 2 of BS 5328:
Part 1: 1997. Some of the engineering properties of foamed concrete may be improved by the inclusion of a coarse
lightweightaggregate but, because little data are available, it cannot be recommended for unrestricted use at this stage. Some
data have, however, been provided by Regan and Arasteh (1984).
12
Table 3 Aggregates for general use, based on Table 2 of BS 5328: Part 1:1997
Type Designation Standard
Lightweight Lightweight aggregates for concrete BS 3797:1990
Normal weight Sand from natural sources for concrete BS 882:1992

Building sand from natural sources BS 1200:1976
Plastic density
The plastic density of each batch (of say 6 to 8 m3) of foamed concrete should be measured, at the point of delivery or
discharge.
Cube strength
Three cube specimens (preferably I50 mm in size) should be taken for every 50 m 3 of foamed concrete placed.
Pour depth
To prevent undue settlement due to the collapse of the bubble structure, normally the maximum pour depth should not be
greater than 1.5 m within any 16-hour period. Although greater depths and more frequent pours can be placed, the self-
weight of the concrete will increase its plastic density: thus where this is done it might be necessary to reduce the specified
plastic density by 100 to 200 kg/m3 to offset any such increase. Further lifts or other site works may recommence after 16
hours or so when a compressive strength of at least I N/ram 2 has been achieved, or when the soundness indentation of the
material is less than 5 ram, in accordance with the method described in Section 3.5.
4.2.2 Optional items

The following might be requested for a particular end-use or site.
Workability
Workability can be determined using either of the methods described in Section 3.2. Each batch, of say 6 to 8 m3, could be
checked where necessary.
Maximum cube strength

Where required, a maximum acceptable characteristic cube sWength should be specified: this should be agreed with the
supplier.
Resistance to segregation
Segregation can be assessed in accordance with the method described in Section 3.3.
Durability
Foamed concrete is not normally considered to require specification clauses for resistance to frost attack. Where specific
resistance to sulfate and/or acid attack is required, the appropriate type of cement used should be as specified in Table 7 of
BS 5328: Part l : 1997. It might be prudent to undertake trials to confirm the durability of the material in a particularly severe
environment.
13
Sustained core temperature
Where considered necessary, for example to minimise the risk of thermal cracking and delayed ettringite formation
(DEF) ~,the sustained core temperature of a pour could be restricted to less than 65 °C for 12 hours or 100 °C for 3 hours,
see Lawrence (1993).
Soundness
A minimum value of soundness or hardness, as determined from the test described in Section 3.5, might be specified to
ensure that an adequate set has been achieved prior to the placement of overlapping pours.
5 ACCEPTANCE CRITERIA
Density
In most cases the acceptable tolerance for plastic density should be ± 50 kg/m3 of the specified density, but this may be
increased to +- 100 kg/m 3 for particularly dense mixes, i.e. ones having a plastic density in excess of 1600 kg/m3.
The variability of the dry density of the hardened foamed concrete should not exceed +_ I00 kg/m 3 of the mean density.
However such a check would not normally be carded out unless segregation was suspected.
Cube strength
For a particular pour, the mean 28-day strength of all the cubes should be higher than the specified characteristic strength.
Where strength in excess of 10 N/mm 2is required, consideration should be given to specifying the 56-day strength because
this widens the options available to the supplier for the mix design.
Workability
A value of workability could be set as a minimum below which the batch could be rejected because, for example, it might
affect the ability of the concrete to completely fill the void.
Resistance to segregation
The oven-dried density of two 25 mm thick slices taken from towards the top and bottom surfaces of a 100 mm diameter,
300 mm long cylinder, should be less than 50 kg/m ~.
Soundness
The result of an indent test, undertaken say 24 hours followingpouring and using the method described in Section 3.5, should
be less than 5 mm: note however that the timing of the test and the acceptance level can be varied to suit.
i There is no evidence to show that DEF, (fit were to occur, would cause any deterioration offoamed concrete: expansion should be accommodated
within the bubble structure.
14
6 NON-CONFORMITY
The purchaser should determine the action(s) to be taken when the result of a test fails to meet that specified. It might range
from qualified acceptance (in less severe cases) to rejection (and excavation and replacement). In selecting the action to be
taken, due regard should be given to the consequences and degree of non-conformity, and to the economic consequences
of alternative remedial measures. Furthermore, prior to any action, the validity of test results should be confirmed by
checking that the sampling and testing have been carried out in accordance with the specification.
The most common non-conformities will be mismatches in strength and plastic density, excessive temperatures during
curing, and segregation of the mix.
In estimating the quality of substandard foamed concrete and determining the action to be taken, the following should be
asked:
What performance characteristics of the foamed concrete, such as durability, strength, and insulation properties, have
been affected adversely?
ii Are other site works likely to be affected by the non-conformity ?
The purchaser (and/or the supplier) may wish to carry out tests on the hardened concrete. These might include non-
destructive methods (to BS 1881: Part 201: I986) or taking cores (to BS 1881: Part 120: 1983). Provided the test data are
valid the results of any further tests do not annul the original non-conformity.
Advice on the interpretation of the results of non-destructive tests and on the strength of cores is given in B S 6089:198 I.
Tiffs Standard also lists the issues to be considered when deciding the action to be taken with substandard structural concrete,
and further information on relevant Codes of Practice.
7 SUMMARY
Foamed concrete is a versatile low-density free-flowing, self-levellingand self-compacting material. It can be pumped both
horizontally and vertically over long distances and large quantities can be placed quickly. In most cases the maximum depth
that can be poured in a single lift is about 1.5 m: the hardening rate normally allows the concrete to be to be walked on within
24 hours, and for additional lifts of concrete to be poured on top. Considerable quantities of foamed concrete (mainly for
fill) have now been placed in the UK, with few reported problems.
The dry density of foamed concrete is usually between 400 and 1600 kg/m~and its compressive strength, which varies with
density, can typically range between 1 N/mm2 and 25 N/mm 2 at 28 days. Foamed concrete has a satisfactory resistance to
freeze/thaw and sulfate attack (at least in the short term). The penetrability of the material to various gases and liquids is
a function of the constituents and density of the concrete, but it can be dominated by the presence of cracking generated,
for example, during curing. Foamed concrete is, however, reasonably permeable to water vapour and carbon dioxide (CO2),
and so the use of carbon steel reinforcement should only be considered where the risk of corrosion is precluded or its effects
are of no consequence.
Foamed concrete is not covered specifically in BS 5328 (various parts and dates) and so a specification has to be formulated
for its use. This report highlights the key aspects that must be addressed in drawing up such a specification. It is proposed
that the general requirements for normal weight concrete, as specified in BS 5328, should be adopted for foamed concrete
excepting those for compressive strength and plastic density. Acceptance criteria have been proposed in the report, but it
is recognised that these may be adjusted as more feedback from practice is gained.
15
8 REFERENCES
Arasteh A R (1988). Structural application of lightweight aggregate foamed concrete. PhD theseis. London:
Polytechnic of Central London.
Brady K C and Greene M J (1997). Foamed concrete: a review of materials, methods of production and
applications. Project Report PR/CE/149/97. Crowthorne: TRL Limited. (Unpublished report available on direct
personal application only)
Brady K C (2000). An investigation into the properties of foamed concrete. Project Report PR/IS/99/00.
Crowthorne: TRL Limited. (Unpublished report available on direct personal application only)
Brewer W E (1996). Controlled low strength materials (CLSM). Radical concrete technology. (Eds Dhir R K and
P C Hewlett). London: E & F N Spon. pp 655-667.
British Standards Institution. London.
BS 12:1996 Specification for Portland cement.

BS 882:1992 Specification for aggregates from natural sources for concrete.
BS 1200:1976 Specifications for building sands from natural sources.
BS 1881: Part 116:1983 Testing concrete~ Method for determination of compressive strength of concrete cubes.
BS 1881: Part 120:1983 Testing concrete; Method for determination of the compressive strength of concrete cores.
BS 1881: Part 201:1986 Guide to the use of non-destructive methods of test for hardened concrete.
BS 3797:1990 Lightweight aggregates for masonry units and structural concrete.
BS 3892: Part 1:1997 Pulverized-fuel ash. Specification for pulverized-fuel ash for use with Portland cement.
BS 4027:1996 Specification for sulfate-resisting Portland cement.
BS 5328: Part 1:1997 Concrete. Guide to specifying concrete.
BS 5328: Part 2:1997 Concrete. Methods for specifying concrete mixes.
BS 5328: Part 3:1990 Concrete. Specification for the procedures to be used in producing and lransporting concrete.
BS 6089:1981 Assessment of concrete strength in existing structures.
BS 6588:1996 Specification for Portland pulverised-fuel ash cements.
BS 6610:1991 Specification for Pozzolanic pulverised-fuel ash cements.
BS 7583:1996 Specification for Portland limestone cement.
BS 8500:2000 Concrete. Complementary British Standard to BS EN 206: Part 1: 2000. [Draft only]
BS EN 206: Part 1:2000 Concrete. Specification, performance, production and conformity. [Draft only]
BS EN ISO 9001:1994 Quality systems model for quality assurance in design, development, production,
installation and servicing.
BS EN 12350 Part 1:2000 Testing fresh concrete. Sampling.
BS EN 12350 Part 6:2000 Testing fresh concrete. Density.
Dhir R K, Jones M R and Nicol L A (1999). Development of structural grade foamed concrete. DETR Research
Project. University of Dundee.
Highways Authorities and Utilities Committee (1992). Specification for the reinstatement of openings in highways
(A Code of Practice). London: The Stationery Office.
Lawrence C D (1993). Laboratory studies of concrete expansion arising from delayed ettringite formation.
Crowthorne: BCA.
16
McGovera G (2000). Manufacture and supply of ready-mix foamed concrete. One-day awareness seminal" on foamed
concrete: properties, applications and potential. University of Dundee. pp 12-25.
New Roads and Street Works Act (1991). Specification for the reinstatement of openings in highways.
Pickford C and Crompton S (1996). Foamed concrete in bridge construction. Concrete, Nov/Dec.
Regan P E and Arasteh A R (1984). Lightweight aggregate foamed concrete. Low-cost and enel~y saving materials,
Ch. 42, pp 123-138.
Sach J and Seifert H (1999). Foamed concrete technology: possibilities for thermal insulation at high temperatures.
CFI Forum of Technology, DKG 76, No. 9, pp 23-30.
Taylor R W (1990). First interim report on foamed concrete. BCA Report No. RA1.007.00.1. Crowthorne: BCA.
Van Deijk S (1991). Foam concrete. Concrete, July/August, pp 49-54.
Watts G R A and Brady K C (1999). Assessment of the use of foamed concrete as a backfill to retaining structures.
Project Report PR/CE/176/99. Crowthorne: TRL Limited. (Unpublished report available on direct personal
application only)
Watts G R A (2000). Loading tests on a masonry arch bridge baclcfilled with foamed concrete. Project Report PRIIS/41/00.
Crowthome: TRL Limited. (Unpublished report available on direct personal application only)
17
Appendices Specification for foamed concrete
APPENDIX A: GLOSSARY OF TERMS
Base mix The cementitious mortar comprising cement, water, filler and chemical admix-
tures into which foam is incorporated to produce foamed concrete.
Coalescence The joining of two or more bubbles in the foam. (Because larger bubbles have
a tendency to rise to the surface, coalescence can lead to segregation and
reduced strength.)
Compatibility The situation where there is no adverse interaction between a foam and chemical
admixture that reduces the stability of the foam in the mortar.
Sealed curing Curing through the use of an air and watertight film. (Where curing in water is
used, foamed concrete absorbs varying amounts of water.)
Delayed ettringite formation (DEF) This reaction~ which occurs at temperatures in excess of 65°C, involves the
decomposition and reformation of primary ettringite in the presence of mois-
ture: it generates disruptive expansive forces in the concrete matrix.
Flowability The ability of a material to flow under its own weight,
Foam stability The ability of a foam to resist collapse into solution.
Segregation The separation of foam from the mortar, by various interactions between the
components, producing variations in composition and density with pour depth.
Heat of hydration The temperature rise generated by hydration. (Foamed concrete has excellent
insulation properties and so a significant rise in temperature can be attained
through curing and retained for several days. The temperature rise can be
moderated by the use of PFA as a partial replacement for PC.)
Plastic density The density of freshly placed foamed concrete: it can be determined by
measuring file weight of material in a container of known volume.
Plastic viscosity Material characteristicassociated with the energy required to sustain a reason-
able rate of flow of material.
Surfactant A material that affects the properties of an interface between air and liquid such
that it provides a thermodynamically stable environment for foam.
Thermal conductivity Material characteristic indicative of the ease of heat transfer through a material.
Workability Material characteristic indicative of the 'ability to be worked'.
Yield Difference between the target and actual plastic density.
Yield value Minimum stress required to initiate flow.
A-I
Specification for foamed concrete Appendices
APPENDIX B: GUIDE FOR SPECIFICATION AND QUALITY

CONTROL OF FOAMED CONCRETE
Mix constituents
• Where possible avoid the use of superplasticizers to enhance workability. Where they are necessary, tests should be
carried out to confirm compatibilityof the foam and superplasticizer: dosage should not exceed 0.2 per cent by weight
of cement.
• Bear in mind that the w/c ratio of the foamed concrete will be slightly higher than that of the base mix.
• Take account of the retarding effect of PFA on the development of strength and on the reduction in the heat of
hydration.
• Where possible do not specify PC contents greater than 500 kg/m~: the risk of thermal cracking increases substantially
with higher contents and the resulting gain in strength is minimal,
• Carry out trial mixes at small and then large scales to confirm that the required workability can be achieved.
• To minimise shrinkage the w/c ratio should be as low as possible.
Foam production
• Carry out regular checks of the density of the foam: tolerance should be set at about ± 1 0 kg/m 3.
Mixing
Carry out regular checks of the plastic density of foamed concrete. In most cases the tolerance should be set at +-.50
kg/m3, If the plastic density is higher than the target density more foam can be added to the mix, but if it is lower by
more than the tolerance the mix should be rejected.
Carry out spread or flow measurements for both the base and foamed mix. To achieve Class 2 foamed concrete (see
Table 1 of main text), the spread of the PC/PFA base mix should be between 115 and 140 mm and for a PC/sand mix
between 85 and 125 mm. With the latter test, flow should be continuous, flow times should be shorter than 2 minutes;
and the corresponding flow class should be 2B or better (see Table 1 of main text). Note that such tests might only
be undertaken on trial mixes.
Costing
• Cover the exposed surface as quickly as is practically possible with a barrier form of curing membrane,
• Following the pour, erect barriers to isolate the area of the pour for about 24 hours or until an adequate strength has
been achieved.
• Allow sufficient time for the concrete to gain an adequate strength before carrying out work on the surface.
A-2
APPENDIX C: FOAMED CONCRETE

The following provides a brief review of the constituents, method of production, properties and uses of foamed concrete:
further details can be obtained from the references quoted and as listed in the Bibliography. It also presents and discusses
the findings of research work undertaken at the University of Dundee; most of this concerned the properties of higher
strength foamed concrete, i.e. 'structural grades'.
C1 CONSTITUENTS
Low-density foamed concrete, i.e. having a dry density of up to about 600 kg/m 3, is usually formed from cement (to which
other binders could be added), water and foam. Denser foamed concrete will include fine sand, which is suitable for making
concrete or mortar, and/or other fillers.
CI.1 BASE MIX
CI.I.1 Binders
Portland cements
Ordinary Portland cement (to BS 12:1996 or BS EN 197: Part 1: 2000) is usually used as the main binder for foamed
concrete. However rapid-hardening Portland cement to BS 915:1983 has also been used, and there does not seem to be any
evidence why sulfate resisting cement to BS 4027:1980 could not be used.
Usually the total cement content will lie between 300 and 400 kg/m 3, but up to 500 kg/m3 has been used to attain higher
strength concrete. According to Jones (2000) the gain in strength obtained by increasing the content above 500 kg/m 3is
small.
Pulverised fuel ash (PFA)

PFA (to BS 3892: Part 1: 1997 or BS EN 450: 1995) is often added, at levels of up to 80 per cent of the cement content, to
reduce cost, enhance workability and increase the long-term strength of foamed concrete. But, particularly when casting at
a low temperature, account should be taken of the reduced rate of strength gain resulting from its addition to a mix.
Furthermore, because of the increased fineness (and hence greater surface area) of PFA, for equivalent workability PC/PFA
mixes have a greater water demand than PC/sand ones.
Kearsley (1996) has reported on the effects of PFA on the properties of foamed concrete, and more recent research has been
undertaken at the Universityof Dundee. This work indicates that the addition of PFA to a mix leads to a more uniform bubble
structure in the paste, which in turn improves some of the engineering properties of the concrete.
Ground granulated blasO~urnace slag ( GGBS)

For the same advantages as mentioned above for PFA, GGBS (to BS 6699: 1992) has also been used for producing foamed
concrete; typically it would make up 30 to 50 per cent of the total cement content. However with some types of surfactant
the use of GGBS has led to instability (i.e. foam collapse) and segregation of the mix.
Silica fume
Both Taylor (1988) and Kearsley (1996) report that the addition of condensed silica fume, of up to 10 per cent by weight
of cement, significantly increases the strength of foamed concrete where the foam content is less than about 30 per cent.
However the improvement in strength with higher foam contents is small.
A-3
C1.1.2 Aggregates and fillers
Sand
Sach and Seifert (1999) recommend that only fine sands suitable for concrete (to BS 882:1992) or mortar (to BS 1200:1976)
having particle sizes up to about 4 mm and with an even distribution of sizes should be used for foamed concrete. This is
mainly because coarser aggregate might settle in a lightweightmix and lead to collapse of the foam during mixing. The effect
of the grading of the sand on the properties of the foamed concrete is discussed in a BCA (1991a) report.
For practical reasons, most sands can only be used to produce foamed concrete having a dry density in excess of about 1200
kg/m ~.
Coarse PFA
Coarse PFA (to BS 3892: Part 2:1997 or BS EN 450:1995) can be used as a partial or total replacement for sand to produce
foam concrete with a dry density below about 1400 kg/m 3, However for the same workability such mixes have a greater water
demand.
The presence of activated carbon within PFA has been shown, by for example Hoarty (I990), to impair the effectiveness
of air-entraining agents and thereby destabilize the air-void system within air-entrained concrete. However work at the
University of Dundee found no observable degradation of foamed concrete even with a relatively high proportion of coarse
high carbon-content ash.
Limestone fines
Limestone fines have been added, at up to 10 per cent by weight of cement, in conjunction with PFA to accelerate the setting
rate of foamed concrete, see for example De Rose and Morris (1999).
Other
Lightweight aggregates up to 16 mm in size, such as expanded polystyrene granules and Leca expanded clay, have been
used to produce foamed concrete: see for example Regan and Arasteh (1984). Any such aggregates should be of about the
same density of the foamed concrete and have minimal capacity for absorbing water.
C1.1.3 Water
The water used for foamed concrete should be potable. This is crucial when using a protein-based foaming agent because
organic contamination can have an adverse effect on the quality of the foam, and hence the concrete produced.
The water/cement (w/c) ratio of the base mix required to achieve adequate workability is dependent upon the type of
binder(s), the required strength of the concrete, and whether or not a water reducing or a plasticizing agent has been used.
In most cases the value will be between 0.4 and 0.8. The higher values are required with finer grained binders, such as PFA,
and the lower values where either a high strength is required or a superplasticizer has been employed.
Where the water content of the mix would be inadequate to ensure full hydration of the cement, water will be extracted from
the foam and might lead to its disintegration. On the other hand whilst high w/c ratios do not significantly affect the porosity
of the foamed concrete they do promote segregation and increase drying shrinkage.
A-4
C1.1.4 Admixtures
Water reducing and plasticizing agents
Because of its intrinsic high workability and modest strength requirements, plasticizers are not universally added to foamed
concrete. But they have been used to improve the workability of a foamed concrete mix or to achieve the same wm'kability
at a reduced w/c ratio.
Work at the University of Dundee has shown that superplasticizers allow w/c ratios as low as 0.3 to be used with no adverse
effects on the properties of the foamed concrete: indeed their addition improved the stability and flow characteristics of a
mix. But Dransfield (2000) has questioned the use of such additives, and Bartos (i 992) claimed that they might reduce the
stability of the foam. The work at Dundee University showed that problems of compatibility with superplasticizers led to
segregation: the degree varied with the type of surfactant but it tended to be more noticeable with protein-based ones and
was less apparent with anionic ones.
To date no serious or common problems of compatibility have been reported from site works. However little is known about
the detailed effects of plasticizers, or indeed any other admixture, on the properties of the foamed concrete and so dosage
should be based on information from the manufacturer and confirmed from trial mixes.
Set controllers
To restrict the spreading of foamed concrete once placed, thickening agents and waterproofing materials have been added
to the mix. The effects that such viscosity modifiers have on the properties of foamed concrete have been discussed by Bartos
(1992) and by Dransfield (2000).
Accelerators can be used to overcome the rather slow rate of stiffening and strength development of foamed concrete.
Because at this stage the mechanism for accelerated curing is not fully understood their use can only be recommended on
an empirical basis. Dransfield (2000) states that calcium chloride is the most effective accelerator: this can be used in all
but a few cases because carbon steel reinforcement is rarely used with foamed concrete.
C1.1.5 Fibres
Polypropylene fibres have been used to improve the shear strength and stiffness of structural elements formed from foamed
concrete: see for example Kearsley and Mostert (1997) and Mellin (1999). For this purpose Kessler (I998) proposed a
dosage level of 1 kg of fibres per m3 of foamed concrete.
C1.2 FOAM
To produce concrete with a reasonable strength the bubbles in the foam should not be irregular nor thin-walled and there
should not be much coalescence between them. Thus, as reported by Gutmann (1988), a foam composed of small spherical
bubbles is best suited for making foamed concrete.
C1.2.1 Foaming agents

Synthetic or protein-based foaming agents (surfactants) can be used to produce foam. Because of the possibility of
degradation by bacteria and other organisms, natural protein-based agents (i.e. fatty acid soaps) are rarely used to produce
foamed concrete for civil engineering works. However research is underway on the use of protein-based agents for
developing high strength, i.e. 'structural grade', foamed concrete. The chemical composition ofa surfactant must he stable
in the alkaline environment of concrete. Because all surfactants are susceptible to deterioration at low temperatures they
should be stored accordingly.
Version No." 1 Application Guide AG39 October 2001
A-5
There is a wide range of surfactants available, and whilst selection of the most suitable surfactant for a particular application
is difficult, it might prove crucial. Thus it might be prudent to undertake mixing trials prior to usage, particularly for large-
scale or critical applications.
Synthetic
Synthetic surfactants can be classified according to the nature of their hydrophylic group, i.e. that part of the molecule that
is soluble in water, for further details see Myers (1992) and Brady and Greene (1997).
Anionic: about 70 per cent of the surfactants used to produce foamed concrete are anionic, i.e. the active part
of their molecule (the hydrophile) is negatively charged.
Cationic: less than 5 per cent of the surfactants used to produce foamed concrete are cationic, i.e. the
hydrophylic group carries a positive charge.
Non-ionic (orpolar): about 25 per cent of surfactants used to produce foamed concrete are non-ionic, i.e. they
are electrically neutral. This lack of electric charge may give a greater stability to a foamed concrete mix.
Amphoteric and Zwitterionic surfactants are rarely used to produce foamed concrete. Depending on the pH
of the solution the molecules can sustain either a positive or negative charge, or both charges.
The chemical nature of the above is intrinsically and functionally diverse. At present the effect of the nature of the surfactant
on the properties of the foamed concrete is largely unknown, and so surfactants are selected, by and large, on an empirical
basis. The performance of the various types of surfactant might vary with the type of binder. Although cement particles are
known to have both positive and negative charges, the surfaces of particles in a composite mix tend to become negatively
charged with the addition of PFA, see Helmuth (1987). Therefore non-ionic or amphoteric/zwitterionic surfactants might
be more stable in PC/sand mixes whilst single-charge surfactants might be more suitable for PC/PFA mixes. This topic
warrants further study.
Protein-based
According to McGovern (2000), compared to the foams produced by synthetic surfactants, those formed from protein-based
agents have smaller bubble sizes, are more stable (i.e. exhibit less water &'ainage) and have a stronger closed bubble
structure. Protein-based surfactants might therefore be best suited to the production of foamed concrete of relatively high
density and high strength.
C1.2.2 Foam
The surfactant solution usually consists of one part surfactant to between 5 and 40 parts of water. Optimum performance
is commonly attained at a ratio of 1:25, but the optimum value is a function of the type of surfactant and the method of
production. The dilution ratio should be chosen with regard to the critical micelle concentration (cmc) of the foam, this is
the point at which the properties of a foam exhibit a significant change, for example in density as shown in Figure C1.
The surfactant solution is foamed to a consistency similar to that of shaving foam and to a density of between 20 and 90 kg/m3:
the density will vary according to the application but in many cases it will be at the upper end of this range.
The density of foam produced with protein-based surfactants is often about 50 kg/m 3. Dhir et aL (1999) explored the use
of foams formed from such surfactants for manufacturing stmctaral elements: the density of the foams ranged from 30 to
50 kg/m 3 and the w/c ratio ranged between 0.3 and 0.5.
A-6
100 -
90-
80-
70-
60
_Z-
50
o~
E
t~
40
30
20
10
0
20 40 60 80 100 120
Surfactant solution concentration (grns/litre)
Figure C1 Variation of foam density with concentration of surfactant solution, after Anon (2000)
Stability
The stability of foam is a function of its density and the type of surfactant. In general, protein-based surfactants tend to form
a more stable bubble structure than synthetic ones.
The foam has to survive its incorporation into the mortar mix and the chemical environment of the concrete until it has
achieved a reasonable set.
Stability can be affected by a number of external environmental factors including vibration, wind, evaporation and
temperature: some or all of these might be present on a site and so some breakdown in the foam is inevitable.
The inherent stability of a foam can be assessed quite simply by measuring its collapse or drainage with time using a glass
measuring cylinder, as shown in Figure 4 of the main text: typical data are shown in Figure C2.
A-7
90-
80"
70"
o
[ Initial foam density
x 60-
c0
g --O-- 60 kg/m3
o_
o
50-
l
[ ~
X 50 kg/m3
35 kg/rn3
20 kg/m3
E 40" A ,,E
"6
30"
o
:>
20"
.13 'm
10-
r r
20 40 60 80
Time (rain)
Figure C2 Measurements of drainage of a foam, after Anon (2000)
C2 MIX DESIGN
The base mix comprises the solids (cement, sand, any other filler) and water. Its intrinsic strength depends on its constituent
proportions and density. Usually the latter woaid be about 2250 kg/m 3, but it has to be measured to calculate the amount
of foam required.
The strength of the base mix defines the maximum strength of the foamed concrete. By reducing the w/c ratio the strength
of the base mix can be increased but only at the expense of reducing workability. With normal weight concrete, wm'kability
could be restored with a superplasticizer, but their use in foamed concrete increases the risk of segregation. At this stage it
is suggested that the dosage of superplasticizers to foamed concrete is limited to 0.2 per cent by weight of cement,
It should be appreciated that some foam collapse will occur during the mixing, placing and maturing of the concrete on site,
and so the final w/c ratio will be slightly higher than calculated from the batch proportions.
Proportioning
There is at present, no guidance or standard method for proportioning foamed concrete. Because the hardened density of
foamed concrete depends on the saturation level in its pores, it is difficult to obtain an accurate measure of its density on
site. Thus in most cases, foamed concrete is proportioned on a volumetric rather than a weight basis, for example 1 m ~of
base mix to 4 m3 of foam.
A-8
An approach followed at the University of Dundee is as follows:
• Calculate the water and superplasticizer contents from the total cement content, which includes cementitious fillers
such as PFA. Ignore the amount of water contained in the foam.
• Determine the amount of air (kg/m3) in the mix from consideration of a unit volume, and from the target density of
the foam estimate the required quantity of foam.
Assuming a target plastic density (D, kg/m3), water/cement ratio (w/c) and cement content (C, kg/m~), the water (W, kg/m~),
sand (S, kg/m3) or PFA (F, kg/ma), contents of the base mix are calculated as follows:
PC/sand mix:
FromD=C+W+S
water content: W = (w/c) x C
sand content: S = D - C - [(w/c) x C]
PC/PFA mix:
From the above,
water content: W = (w/c) x (C + F)
PFA content: F = [D - C - (w/c) x C] / [(w/c + I]
C3 C H A R A C T E R I S T I C S A N D P R O P E R T I E S
C3.1 FRESH MIX
C3.1.1 Workability and rheology

Foamed concrete is a free-flowing, self-levellingmaterial and should therefore be expected to give a collapse slump. Thus
neither the slump test (to BS EN 12350: Part 2: 2000) for normal weight concrete nor the flow test (to BS EN 12350: Part
5: 2000) for concrete with a high slump are applicable. Usually, therefore, the workability of foamed concrete is evaluated
visually: in most cases it would not be difficult to spot when workability was unacceptably low.
As described by Dhir et al. (1999), the tests used at the University of Dundee to assess the workability of the base mix and
foamed mix include measurement of'spreadability', as described by Brewer (1996) and'flowability', as described by Bartos
(1992). Such measurements can be taken to represent, respectively, the yield and plastic viscosity of the mixes. The tests
are briefly described in Section 3.2 of the main text. Dhir et al. ( 1999 ) recommended that for adequate workability, i.e. for
a Class 2 mix to Table I of the main text, the spread of the base mix shoald be between I 15 and 140 mm for a PC/PFA mix
and between 85 and 125 mm for a PC/sand mix.
As could be expected, and as shown in Figure C3 for examp] e, the workability of foamed concrete increases with increasing
w/c ratio and increasing dosage of superplasticizer.
All other things being equal, of the types of cement used to produce foamed concrete those with GGBS have the highest
workability, but separation of foam and paste has been noted with high GGBS contents and at low w/c ratios.
The effect of the type of surfactant on the properties of the foamed concrete has been investigated at the University of
Dundee, and data from this work are reproduced in Table C1. The 'spreadability' of the various types of concrete are
reasonably similar but, for otherwise identical mixes, concrete formed from a protein-based surfactant has a much shorter
A-9
300 "
zx A
A A 0.4% Suffactant percentage
250 "
200 "
< > / oyosp
"~ 1 5 0 "
CO
100 - []
[]
50-
0 , f , J i
0.2 0.25 0.3 0.35 0.4 0.45 0.5
w / c ratio
Figure C3 Effect of w/c ratio and superplasticizer content on the spreadability of foamed concrete, after Anon (2000)
A-IO
Table CI Effect of the type of surfaetant on the properties of foamed concrete, from Anon (2000)
Surfactant type Synthetic* Amphoteric Anionic Protein-based
w/c ratio 0.35 0.40 0.45 0.35 0.40 0.45 0.35 0.40 0.45 0.35 0.40 0.45
Surfactant content 0.4 0.2 0.0 0.4 0.2 0.0 0.4 0.2 0.0 0.4 0.2 0.0
% wt cement
Spread, mm 185 215 175 305 255 205 240 200 155 345 270 195
(see Section 3.2)
Flow time >4 >4 - 35 25 30 1.5 >2 - 15 15 3

(see Section 3.2) mins mins sees sees sees mins mins sees sees mins
Segregation**,kg/m 3 +20 +13 +3 + 8 7 +65 +37 +78 +14 +13 +1150 + I 8 0 +32
(see Section 3.3)
Compressive"t 1-day 5.0 3.5 3.0 6.0 4.0 3.5 6.5 2.5 3.0 3.5 4.0 2.5
strength, N/mm2 28-day 9.5 10.5 10.0 10.5 9.0 9.0 12.0 5.5 7.0 12.0 11.5 8.0
(see Section 3.4) 56-day 9.0 12.0 8.5 10.0 9.0 9.0 10.0 6.0 7.5 10.5 11.5 8.0
Sorptivity index, 0.077 0.06I 0.096 0.123 0.075 0.084 0.04I 0.079 0.073 0.051 0.0652 0.0476
rnm/min la
(see Section C3.3.3)
Water vapour 2.980 2.430 2.580 2.356 3.244 2.587 2.724 4.626 3.830 1.694 2.285 2.984
permeability,
gms/mm/s x 10-13
(see Section C3.3.3)
~" PC~sand mix with P C content of 500 kg/m 3 : plastic density of 1400 kg/m 3 , all specimens sealed cured
* Commercially available surfactant of unknown electrical charge: the others are, as yet, not commercially available
**Difference between top and bottom of pour
A-11
Specification f o r foamed concrete Appendices
flow time, which is indicative of a much lower plastic viscosity. However mixes formed from protein-based surfactants are
prone to segregation, probably due to incompatibility of the surfactant with the superplasticizer.
Dhir et al. (1999) attempted to quantify the rheology of foamed concrete using a Brookfield RVT viscometer: some of their
data are reproduced in Table C2. The yield values were all less than 2 Nm, which, according to Tattersall (1991), indicates
self-flowing behaviour. The less viscous mixes exhibited lower values of yield and plastic viscosity. It was also found that
PC/sand mixes were more viscous than equivalent PC/PFA mixes, which suggests that the latter have better flow
characteristics.
C3.1.2 Heat of hydration
The cellular structure of foamed concrete provides it with good insulating properties. Consequently the temperature rise
developed through hydration is higher and lasts for longer than for normal weight concrete. The temperature-time relation
is a function of the dimensions of the pour, the cement content and the density of the foamed concrete. According to Ansell
(2000), temperatures of up to 100 °C could be sustained for 3 days. Such persistent high temperature might result in thermal
strains and cracking of the concrete, but such problems have not been reported in the literature. The risk of delayed ettringite
formation, which is promoted by temperatures above 65 °C, is low because ettringite is formed in the pores of the concrete
and expansion can thus be accommodated.
In a series of tests undertaken at the University of Dundee, measurements were made of the temperatures developed during
the maturing of 170 mm cube specimens. The cubes were placed in an insulated box and a thermocouple was installed in
the core of the specimen; maturity curves obtained from a few of these tests are given in Figure C4 and Table C3. The data
show that the peak temperature and the period of time for which the temperature exceeded 65°C tended to reduce with
increasing PFA and GGBS content. Furthermore the time taken to reach the peak temperature tended to increase, and so the
rate of temperature rise was also reduced.
The effect of the density of foamed concrete on the temperatures developed during curing is unclear, but with decreasing
density increasingly higher temperatures might be generated and sustained over a longer period of time due to the greater
volume of air bubbles and hence better insulatingcapacity. However this might be offset by the lower cement content within
a lower density concrete.
Further data from work at the University of Dundee are reproduced in Table C4. These data suggest that neither the type
of surfactant nor the w/c ratio had any effect on the heat of hydration.
C3.1.3 Rate of hardening

There is no standard method for determining the initial and final setting times of foamed concrete. However the methods
given in BS 4550:1978 and ASTM C266-89 for cements might provide the basis of suitable methods for foamed concrete.
Dhir et al. (1999) reported that 'stiffening' of a foamed concrete mix occurred about 5 hours after casting and, as shown
in Figure C5, 9 to 10 hours were required to achieve the set limits given in BS 4550:1978. This slow build-up in hardening,
in comparison with normal weight concrete, is probably due to the retarding properties of the foam.
The 'setting' time of foamed concrete is usually between 12 and 24 hours.
C3.1.4 Plastic settlement
A 1990 BCA report stated that, for a trench reinstatement, within the first year of service foamed concrete settled about 22
per cent less than a typical granular fill. But it should be recognised that the plastic settlement of foamed concrete may be
higher than of normal weight concrete, and special care should be taken where inclusions such as void formers and the like
are used. However no problems of excessive settlement have been reported in practice.
A-12
Appendices Specification f o r foamed concrete
Table C2 Rheology of foamed concrete, from Dhir et al. (1999)
Plastic densiO, Flow time* Yield Plastic viscosiD,**

Mix type "? kg/m 3 seconds Nm Ns/m 2
PC/sand 1400 74 0.82 0.034

1600 76 1.13 0.040
1800 80 1.80 0.05 I
PC/PFA 1400 80 0.44 0.025

1600 77 0.52 0.032
1800 95 0.69 0.075
Assessed using Brool~eld RVT viscometer

¢ w/c ratio = 0.30, PC content = 500 kg/m ~
* See Section C3.2
** For comparison the plastic viscosity of water and syrup are 0.001 and 1 Ns/m 2 respectively
80 ¸
20% PFA
25% PFA
70 ¸
30% PFA
60 ¸
5°1l/
o
~ 40'
E
peak
temperature
30.
20.
Plastic density = 1000 kg/m 3
Total binder c o n t e n t = 6 0 0 kg/m 3
101 , , , ,~
0 20 40 60 80
T i m e (hours)
Figure C4 Effect of the percentage of PFA on the temperature developed by curing of foamed concrete, after Anon (2000)
A-13
Table C3 Effect of PFA and GGBS on temperatures developed during curing of foamed concrete, from
Leptokaridis (2000)
Plastic Peak Time to peak Period when

density temperature temperature temp> 65°C
kg/m ~ Binder I °C hrs hrs
I000 PC / 20% PFA 76 11 12

PC / 25% PFA 67 12 6
PC / 30% PFA 62 13 0
PC / 30% GGBS 92 14 I8
PC / 40% GGBS 74 19 I3
PC / 50% GGBS 74 15 12
1200 PC / 20% PFA 79 11 17

PC / 25% PFA 75 14 14
PC / 30% PFA 81 14 20
PC / 30% GGBS 89 14 21
PC / 50% GGBS 82 19 21
t Total cement (all combinations) content = 600 kg/m ~
Table C4 Effect of surfactant type and w/c ratio on the curing temperatures developed in foamed concrete, from
Anon (2000)
Peak Time to peak

temperature temperature
Type of surfactant w/c ratio °C hrs
Anionic 0.35 61 15
0.40 59 12
0.45 60 11
Amphoteric 0.35 62 12
0.40 62 12
0.45 59 12
Protein-based 0.35 60 15
0.40 63 12
0.45 60 12
PC~sand mix with plastic density of 1400 kg/m ~ and total cement content of 500 kg/m ~
A-14
10
PC/PFA - Flow time 77s, Plastic density 1600 kg/m 3
&. PC/PFA - Flow time 95s, Plastic density 1800 kg/m 3
PO/sand - F]ow time 76s, P]astic density1600 kg/m3
PC/sand - Flow time 80s, Plastic density 1800 kg/m3
E
~7
z Laboratory Temp. = 8 o ° c
~6 9 kg stiffened limit given in BS 4580 (1978)
w/c ratio = 0.30
~-4
"0
t9
0 2
0 ~
0 1 2 3 4 5 6 7 8 9 10 11 12
Time (hours)
Figure C5 Typical hardening times of foamed concrete, after Dhir et al. (1999)
C3.1.5 Plastic shrinkage

The high paste/aggregate ratio and lack of coarse aggregate in foamed concrete makes it vulnerable to shrinkage cracking.
However no particular problems associated with plastic shrinkage cracking on pours with high surface area/volume ratios
have been reported in the literature: further investigation is required. This apparently good performance of foamed concrete
could be attributed to its high air content, but this has yet to be confirmed. Further work is warranted on this important issue.
Curing foamed concrete with a barrier-form of membrane reduces plastic shrinkage. Fibres can also be added for this
purpose but their presence adversely affects the workability of the concrete.
C3.2 HARDENED C O N C R E T E
C3.2.1 Compressive strength

The compressive strength of foamed concrete is a function of the characteristics of both the base mix and foam.
Density of mix
As could have been expected, and as confirmed in Table C5 and Figure C6, the compressive strength of foamed concrete
reduces with decreasing density. For mixes of similar constituents, the density-strength relations should be reasonably
similar. But, because constituents can vary widely, density is not necessarily a reliable indicator of strength (or quality).
A-15
Table C5 Summary of properties of hardened foamed concrete, from BCA (1994)
7-day
Dry compressive Thermal Modulus of Drying
density strength conductivity elasticity shrinkage
kg/m3 N/mm2 W/mK kN/mm2 %
400 0.5- l.O 0.10 0.8- 1.0 0.30-0.35

600 1.0 - 1.5 0.08 - 0.11 1.0 - 1.5 0.22 - 0.25
800 1.5 - 2.0 O. 17 - 0 . 2 3 2.0 - 2.5 0.20 - 0.22
1000 2.5 - 3.0 0.23 - 0.30 2.5 - 3.0 0 . 1 8 - 0. I 5
I200 4.5 - 5.5 0.38 - 0.42 3.5 - 4.0 0.09 - 0.11
I400 6.0 - 8.0 0.50 - 0.55 5.0 - 6.0 0.07 - 0.09
1600 7.5- 10.0 0.62- 0.66 1 0 . 0 - 12.0 0.06- 0.07
16
Plastic density 1200 kg/rn2 Plastic density 1400 kg/m2
14 ...:.....~..0
,. /
.J /
12 f."
E 41' PC
z 10 ~. 30% PFA
<) SRPC /
./"
~ 8 /
~ 6
E / / .~-¢'
0 "" -" ."
~ 4
28 56 28 56
Time (days) Time (days)
Figure C 6 E f f e c t o f c e m e n t t y p e a n d p l a s t i c d e n s i t y o n c o m p r e s s i v e s t r e n g t h o f f o a m e d c o n c r e t e , a f t e r A n o n ( 2 0 0 0 )
A-16
w/c ratio
The effect of the w/c ratio on compressive strength seems to vary with the density of the foamed concrete. The data reported
by Tam et al. (1987) and Dransfield (2000) show decreasing compressive strength with decreasing w/c ratio. But this is the
reverse of the trend reported by De Rose and Morris (1999) for foamed concrete (with a w/c ratio of up to 0.45 ) and for mortar
pastes and normal weight concrete. Data obtained from research undertaken at the University of Dundee me summarised
in Table C6. These data show that strength increased with reducing w/c ratio for a 1200 kg/m3density mix, the reverse was
the case with the 1000 kg/m3 density mix, and there was no consistent trend for the 1400 kg/m3density mix.
Table C6 Effect of w/c ratio on compressive strength of foamed concrete, from Anon (2000)
Plastic Compressive strength, N/mm:

density
kg/m3 w/c ratio 28-day 56-day
1000 0.25 1.5 2.0

0.28 2.0 2.5
0.30 3.0 3.5
1200 0.25 6.5 9.0

0.28 4.5 6.0
0.30 3.0 5.0
1400 0.25 11.5 12.0

0.28 12.0 15.5
0.30 10.5 14.0
All specimens sealed cured

SRPC content = 500 kg/m3
Type offiller
Kearsley (1996) found that the sand content did not have a marked effect on the compressive strength of foamed concrete.
As shown in Figure C7, taken from Dhir (1999), slightly lower early strengths might be obtained with mixes containing
cementitious fillers than ones containing sand. But the strength of the former continues to increase beyond 28 days (due to
pozzolanic reaction), while the strength of the latter does not increase significantly. Increasing the amount of PFA and silica
fume increases the 28-day strength, with a more marked effect at higher foamed concrete densities: see De Rose and Morris
(1999), Kearsley (1996 and 1999) and Taylor (1988).
Type of foam
The compressive strength of foamed concrete is a function of the density of the foam and the type of sarfactant. In general
higher strengths are achieved when the air bubbles are of small, uniform size, see BCA (I994), and when protein-based
surfactants are employed; see Dransfield (2000). The latter can be attributed, in part, to the ability of a protein to hold water
and make it available to the cement mortar at a much slower rate. The hydration process is therefore prolonged, resulting
in a thicker and more compact coating around the air void, which contributes to the development of strengths up to 70 per
cent higher than those achieved with synthetic agents, see Aldridge (2000a). Further, as shown by the data provided in
Table C1 and Figure C8, the strength of foamed concrete formed with different surfactants varies with the w/c ratio.
A-17
50
PFA - plastic density = 1400 kg/ma
- - O ' - Sand - plastic density = 1400 kg/rn 3
E 40
30
(D
> 20
--o o
0
t.)
I ! I [ I
10 20 30 40 50 60
T i m e (days)
50
;~ PFA - plastic density = 1600 kgJm
Sand - plastic density = 1600 kg/m a 3I
E 40
30
20
10
O
{J
I I P I I
10 20 30 40 50 60
T i m e (days)
50
PFA - plastic density = 1800 kg/ma I
--O-- Sand - plastic density = 1800 kg/m 3 I
E 40
z
~ 30
E
.~ 20
~- t0
0
o
10 20 30 40 50 60
T i m e (days)
Figure C7 Development of compressive strength of foamed concrete, after Dhir et al. ( ] 999)
A-18
14 "14
Synthetic Amphoteric
-12
10 J -10
8" 8
6" 6
~" 4- 4
E
E 2 2
z
0 I 0
0 40 60 210 40 60
14- "14
Anionic Protein-based
~- 12- "12
8 10- -10
^
8" 8
6- 6
4- 4
2- 2
0
I 10 I
2•0 40 60 4O 60
T i m e (days)
"0" w/c = 0.35
--D- w/c = 0.40
-~- w/c = 0.45
1400 kg/m3 plastic density

500 kg/m8 plastic density
w/c: 0.35-0.45
Figure C8 Effect of the type of surfactant on the development of strength of foamed concrete, after Anon (2000)
A-19
Curing regime
The curing regime can have a dominant effect on the strength of foamed concrete. And cleady, for quality control and
comparative purposes, it is necessary to establish a standard curing regime. This subject deserves further study.
Kearsley (1996) examined a number of curing regimes. Of the methods, the highest strengths were obtained on specimens
cured at 50 ° C, and on specimens sealed in plastic bags and held at a constant temperature of 22 °C. Kearsley (1996) concluded
that testing water-cured specimens gave low strengths due to the build-up of pore water pressure in the saturated
microstructure of the foamed concrete. Thus he recommended that foamed concrete test specimens be sealed cured, i.e.
wrapped in cling film and stored in plastic bags. Variations of this regime include wetting the specimens after demoulding
and before wrapping, or storing them in a high humidity environment.
Kearsley and Booysens (1998) found that air-cured specimens at 40°C had higher strengths than ones cured in water. As
shown in Figure C9, Dhir et al. (1999) also found that specimens cured in air had a higher strength than those sealed-cured,
the difference was particularly marked with PC/PFA mixes.
Fibres
Mellin (1999) found that the 28-day strength of 'structural-grade' foamed concrete was substantially increased by the
addition of 0.25 and 0.5 per cent by weight of polypropylene fibres: some data are reproduced from Mellin in Table C7.
C3.2.2 Tensile and flexural strength
As with compressive strength, for the same constituents the tensile strength of foamed concrete (subject to bending) will
increase with increasing density of the mix. ASTM C869-91, recommends that foamed concrete should have a minimum
tensile strength of 0.17 N/mm 2. Van Deijk ( 1991 ) reported (56-day) tensile strengths of 0.06, 0.17, 0.32 and 0.51 for mixes
having plastic densities of 600, 900, 1200 and 1500 kg/m 3 respectively.
Jones (2000) reported that the splitting tensile strength (determined to BS 1881 : Part 117:1983) of'structural grade' foamed
concrete was only slightly lower than that estimated for lightweight concrete; data are reproduced in Table C8. As shown
in Figure C 10, foamed concrete formed from PC/sand mixes have slightly higher splitting tensile strengths than PC/PFA
mixes, but the tensile/compressive strength relations of the two are significantly different, possibly due to sand providing
strength through interlocking.
De Rose and Morris (1999) reported that the flexural strength of low-density foamed concrete (determined to BS 1881 : Part
118:1983) reduced with increasing w/c ratio. As shown in Table C7, a substantial increase in the flexural strength of foamed
concrete was generated by the addition of a low percentage of 19 mm long polypropylene fibres. Kearsley and Mostert
(1997) also reported increases in tensile strength through the addition of fibres: in this case higher strengths were obtained
through the addition of increasing amounts of 12 mm long polypropylene fibres. However in practice fibre contents are likely
to be limited by cost considerations.
C3.2.3 Modulus of elasticity

As shown in Table C5, the modulus of elasticity (E) (determined to BS 1881 : Part 121 : 1983) for foamed concrete ranges
from 1 to 12 kN/mm 2for dry densities of 500 to 1600 kg/m 3respectively. By comparison the E value for structural concrete
having a compressive strength of 40 N/mm 2 is about 28 kN/mm 2. This large difference can be attributed to lack of coarse
aggregate in the former.
Relations between E and the 28-day compressive strength for PC/sand and PC/PFA mixes are provided in Figure C 11. For
the same strength, the former have much higher E values; again this is probably due to the interlocking of the fine aggregate.
As shown in Table C7, the addition of 0.5 per cent of polypropylene fibres to foamed concrete beams increased their stiffness
by a factor of between 1.7 and 4.6. These enhanced values are still much lower than for normal weight concrete. Thus much
greater deflection is observed in foamed concrete beams than ones formed from normal weight concrete: furthermore the
former exhibit a slightly more brittle failure but one that was not sudden or explosive.
A-20
50
a) PC/PFA mixes
45
~ 4O
35 Air cured
¢33
30
25 Sealed cured
¢/1
Q. 20
E ~" Watercured
8>. 15
?
(:o
10
0 t I I I I I I I
1100 1200 1300 1400 1500 1600 1700 1800 1900 2000
Hardened density (kg/m3)
30
b) PC/sand mixes
25
E
Z
20
z:
Air cured , O (~ )~O
.> 15 Sealed cured ~ ~k

tO
t//
Q.
E A
8 lO
.¥
¢D
CXl
5
0 I I I I I ~ I
1100 1200 1300 1400 1500 1600 1700 1800 1900 2000
Hardened density (kg/m3)
Figure C9 Effect of curing regime on 28-day compressive strengths of foamed concrete, after Dhir et al. (1999)
A-21
Table C7 Effect of the addition of polypropylene fibres on properties of foamed concrete, from Mellin (1999)
Plastic Fibre Compressive strength, N/mm 2 Flexural Modulus of

density content ~" strength* elasticity "•
kg/m J % 2 day 7 day 28 day N/mm 2 k~V/mm2
1400 0.00 9.0 15.0 25.0 1.3 10.0

0.25 17.0 35.0 62.0 4.1 19.0
0.50 I4.0 22.0 40.0 2.9 17.0
1600 0.00 8.0 16.0 26.0 1.0 4.0

0.25 13.0 30.0 43.0 2.5 I8.0
0.50 16.0 33.0 58.0 3.2 18.5
All specimens sealed cured

"i" 19 mm long polypropylene fibres
• See Section C3.2.2
*" See Section C3.2.3
Table C8 Splitting tensile strength of concrete, from Jones (2000)
28 day Splitting tensile strength 1"i', N/mm 2

Plastic compressive
density strength Normal Lightweight
Mix ~" kg/m 3 N/mm2 Foamed weight" aggregate**
PC/sand 1400 13.5 0.8 1.2 1.3

1600 I9.5 1.8 1.6 1.7
1800 28.5 2.1 2.1 2.2
PC/PFA 1400 21.5 1.5 1.7 1.8

1600 33.5 2.0 2.3 2.4
1800 48.0 2.5 3.0 3.1
t" All specimens sealed cured with PC content of 5OO kg/m s

~'~ See Section C3.2.2
Calculated using the expression f, = 0.20(fc)° 7°by Oluokun (1991)
Calculated using the expression f, = 0.230e) °'67by CEB-FIP (1993)
Version No: 1 Application Guide A(739 October 2001
A-22
0 PC/sandmix
3.5 _a PC]PFAmi
ft. 3
E
2.5
2
¢,
1.5
rJ )
oz.
co 1
0.5
0 I I I 1 I I [ I
5 10 15 20 25 30 35 40 45
28-day compressive strength (N/mm 2)
Figure C10 Relationship between splitting tensile and 28-day compressive strengths, after Dhir et al. (1999)
20
E (sand mixe C)
15
E
"lO
"5
, = • •
I I I I I I I I
5 10 15 20 25 30 35 40 45
28-day compressive strength fc (N/mm2)
Figure C l l Relationship between E-value and 28-day compressive strength, after Dhir et al. (1999)
A-23
C3.2.4 Drying shrinkage

As shown in Table C5, the drying shrinkage of foamed concrete typically ranges from about 0.1 to 0.3 per cent: i.e. between
about 4 and 10 times that of normal weight concrete. The higher values for foamed concrete can be attributed to its (relatively)
high cement content, its high water content and the lack of coarse aggregate in the mix. The amount of drying shrinkage
tends to increase with increasing foam content (i.e. with decreasing density of the concrete) and with increasing temperature,
see Sach and Seifert. (1999).
Drying shrinkage can be reduced by (a) partially replacing PC with PFA (b) adding sand and/or adding lightweight aggregate
to the mix and (c) decreasing the w/c ratio (to about 0.45) but maintaining workability with a superplasticizer. Foaming
agents have also been developed to reduce the settlement/shrinkage of foamed concrete, for example 'Barracell' (FEB/
Master Builders, 1997). Such agents might be used to ensure that a cavity was completely filled.
According to McGovern (2000), shrinkage usually occurs within 20 days or so of casting.
C3.2.5 Creep strain

A thorough search of the literature, prior to this research project, found no data on the creep behaviour of foamed concrete.
The results of tests undertaken as part of this study are provided in Table C9 and Figure C 12. Creep performance is better
expressed in terms of a creep coefficient or specific creep strain rather than simply in terms of creep strain. The creep
coefficient is the ratio of the creep and elastic strains, whilst specific creep strain is the creep strain per unit stress. The creep
coefficients given in Table C9 are broadly consistent with those derived for normal weight concrete, see for example Field
and Bamforth ( 1991 ). Figure C 12 shows, as could be expected, increasing specific creep strain with decreasing density: the
values are higher than for normal weight concrete.
Table C9 Creep coefficient values, after Brady (2000)
Type 1 Type 2 Type 3
Elastic Strain (microstrain) 241 151 117

Creep Strain (microstrain) 508 325 256
Creep coefficient 1.11 1.15 1.19
Density (kg/m3) 1330 1570 1750
28-day compressive strength (N/mm 2) 2.8 6.3 10.8
All specimens sealed cured with a PC content of 270 kg/m 3 (Type 1), 255 kg/m 3(Type 2), and 355 kg/m~ (Type 3).
C3.2.6 Shear capacity

Kearsley and Mostert (1997) found that the shear resistance of reinforced foamed concrete beams was about 35 per cent
lower than the strength of comparable beams formed from normal weight concrete. This difference might be due to the lack
of coarse aggregate in the foamed concrete.
Mellin (1999) found that the addition of fibres and/or stirrups at 125 to 150 mm centres increased the shear strength of the
foamed concrete beams. He found that the shear capacity of fibre reinforced foamed concrete and normal weight concrete
beams were comparable, but for the same load the former exhibited greater deflection.
A-24
600
500
= 400
£
O
300
o 200
100
0
1 10 100 1000
Log time (t+l days)
Figure C12 Specific creep profiles for three foamed concrete mixes, after Brady (2000)
C3.3 DURABILITY
C3.3.1 Freeze-thaw resistance

Because the hollow voids can accommodate the expansive forces resulting from water freezing, foamed concrete has good
freeze/thaw resistance: data on the freeze/thaw resistance of foamed concrete have been provided by BCA (1994) and
Basiursld (2000).
A selection of results from freeze/thaw tests which involvedcycles of 4 hrs freezing in air at - 10 °C followed by 4 hrs thawing
in water at 5 °C (as required by ASTM C666:1990) are provided in Figure C 13. It can he seen that for a dry density of
1400 kg/m3both PC/PFA and PC/sand mixes showed good freeze/thaw resistance, but the performance of a denser mix
(1800 kg/m3) was much poorer, due to the lower void content.
Brady (2000) reported on the resistance of foamed concrete to freeze/thaw in the presence of a salt solution. Assessment
was based on data derived in accordance with RILEM CDC 2 (RILEM, 1977) salt scaling tests in which the exposure regime
is intended to simulate the worst case conditions for structural concrete. The deterioration of foamed concrete in the rather
severe conditions was rapid and extensive, and performance was poorer than for a 'low' resistance concrete.
C3.3.2 Resistance to sulfate attack

BCA (1991b) reported that foamed concrete could be used in sulfate-beating grounds. Some short-term data on the
performance of PC/PFA and PC/sand foamed specimens in a severe environment (Class 5 exposure conditions using a
solution of magnesium sulfate) are reproduced from Dhir (1999) in Figure C 14. Further data, reported by Brady (2000), are
reproduced in Table C 10. Measurements of ultrasonic pulse velocity (UPV) were determined according m B S 1881: Part
203: 1986. These provide a good indicator of changes in the properties of a material: an increase in velocity would usually
indicate an increase in density and vice versa, but velocity would also reduce with cracking of the material. Both sets of data
confirm that foamed concrete has a good resistance to sulfate attack, at least in the short-term. But there is a pressing need
to investigate its longer-term performance.
A-25
PFA mixes Sand mixes
25 25
20'
15
-~ , A-----A A--"--A -&
"~ 10~ "~ 10, A-----~ A A-----
"5 '5
0 I I I o I I I
0 50 100 150 200 50 100 150 200
Cycles Cycles
40
o
to
~ 3s 35
;30 & ~ 30
o
"~ 25 "~ 25
~ 20 20 I [ I
0 50 100 150 200 50 100 150 200
Cycles Cycles
0.2 0.2
0.15
0.1 0.1
0'15I 0.05
0~ 0.oO5
•~ 0
-0.05 ~-&05
©.1
-0.15 -0.15
-O2. I I -O.2 I I I
50 100 150 200 50 100 150 200
Cycles Cycles
Plasticdensity 1400kg/m3 (w/c ratio 0.3) Plasticdensity 1800 kg/ms (w/c ratio 0.3)
Plasticdensity 1400kg/m3 (w/c ratio0.4) Plastic density 1800kg/m3 (w/c ratio 0.4)
F i g u r e 13 Freeze/thaw resistance o f f o a m e d concrete, after D h i r et al. (1999)
A-26
PFA mixes Sand mixes

25 25
El 0 0
20 I 7~=,===,=0=---'----~ ~- -->" 20
0 --0--- o 0
m~
_mc~ 15 ~E 15
~z 10~ ~Z 10~ A A A
?=v
o 13
5 5
I I I I I I I
1 2 3 1 2 3 4
Time (weeks) Time (weeks)
4O 4O
~ 35 o 35
>
3o,, --~ 30
O-
25', ~ 25
2O I I I I I I I 20 I I I I I
0.5 1 1.5 2 2.5 3 3.5 1 2 3 4 5 6
1.6 1.6
1.4 1,4
o~ 1.2 1.2
.c_ 1 .=_ 1
,% 0.8 ,~ 0.8
,% o.e o.e
0.4 0.4
0.2 0.2
0.5 1 1.5 2 2.5 3 3.5 1 2 3 4 5 6

0.05 0.05
0.0(3 0.03
0.01 *~ 0.01
.0.01£ '~ -0.or ~

LU -0.(~3 uJ -0.03
,0.05 I I I I I I I -0.05
0.5 I 1,5 2 2.5 3 3.5 0,5 1 1.5 2 2.5 3 3.5
Plastic density 1400 kg/m 3 (w/c ratio 0.3) --O-- Plastic density 1800 kg/m 3 (w/c ratio 0.3)
Plastlc density 1400 kg/m 3 (w/c ratio 0.4) Plastic density 1800 kg/m 3 (w/c ratio 0.4)
Figure C14 Sulfate resistance of foamed concrete, after Dhir et al. (1999)
A-27
Table C10 Effect of sulfates on properties of foamed concrete, after Brady (2000)
Compressive Hardened
UPV strength density
(krrds) (N/mm 2) (kg/m3)
Time in
Age at solution Mean % of Mean % of Mean % of
test (days) (days) of 3 control of 3 control of 3 control
(a) Foamed concrete type 1

28 0 2.21 100 2.83 100 1330 100
(control)
28 21 2.33 105 2.83 100 1340 101
56 0 2.32 100 3.00 100 1290 100

(control)
56 49 2.50 I08 3.00 100 1370 106
90 0 2.54 100 3.83 I00 I400 100

(control)
90 83 2.56 101 3.67 96 1410 101
(b)Foamedconcretetype2
28 0 2.58 100 6.33 100 1570 100
(control)
28 21 2.71 105 6.50 103 1580 101
56 0 2.71 100 7.17 100 1590 100

(control)
56 49 2.71 100 7.50 105 1590 100
90 0 2.71 100 7.83 100 1600 100

(control)
90 83 2.74 101 7.83 100 1590 100
(c)Foamedconcretetype 3
28 0 3.00 100 10.83 100 1750 100
(control)
28 21 3.04 101 10.33 95 1710 98
56 0 2.78 100 9.67 100 1650 100

(control)
56 49 2.93 105 11.00 114 1710 103
90 0 3.04 100 10.67 I00 1700 100

(control)
90 83 3.09 102 12.83 120 1720 I01
Detai& of foamed concrete given in Table C9
A-28
C3.3.3 P e r m e a t i o n characteristics
Transport processes through foamed concrete include absorption, permeation and diffusion: for any particular agent these
can operate individuallyor, more commonly, in combination. Chemical reactions and electrochemical processes can play
an important role, for example in determining the rate of carbonation. Ions can be transported tlu'ougb the movement of
water. Thus the penetrability of foamed concrete plays an important role in determining its susceptibility to degradation.
The penetrability of foamed concrete is strongly dependent upon its porosity, and the size distribution and connectivity of
the pore spaces. Its permeability to gas or water could therefore be governed by the extent of cracking in the bulk. Where
air voids are interconnected by micro-cracks, permeability will be much higher than for normal weight concrete having the
same w/c ratio. But where the voids are isolated the reverse will be the case.
Through its role in affecting cracking (at least at the surface), the curing regime might have a dominating influence on the
penetrability of concrete. When assessing site performance, tests should be done on cores recovered from the site, or the
in situ curing regime should be replicated (as best can be) on laboratory cured specimens.
Water absorption
Dhir et aL (1999) found that the initial surface absorption (ISA) test, as described in B S 1881 : Part 5:1970, was not suitable
for foamed concrete having a dry density of less than about 1400 kg/m3. This might be due to the coarse porous structure
and perhaps interconnectivity of such concrete giving rise to leakage from under the test cap and/or the sides of the test
specimen. The results of (ISA) tests undertaken on relatively dense foamed concrete are provided in Table C 11 : it can be
seen that the ISA values of a PC/PFA mix are almost an order of magnitude higher than for a PC/sand mix.
Table C l l ISA values of foamed concrete, from Dhir et al. (1999)
ISA* (ml/m2/s)
Mix ¢ w/c ratio 10 min 30 min 60 min
PC/sand 0.3 0.213 0.131 0.079

0.4 0.105 0.051 0.033
PC/PFA 0.3 0.981 0.785 0.703

0.4 1.439 0.458 0.311
t Plastic density = 1800 kg/m3 and PC content = 500 kg/m 3

* See Section C3.3.3
Dhir et al. (I 999) also reported that measurements of the water absorption of totally immersed specimens indicated similar
behaviour and absorption levels for PC/sand and PC/PFA mixes. For both types of mix the absorption rate reduced
significantly with the period of immersion.
The results of water absorption tests (to BS 1881: Part 122: 1983) undertaken on foamed concrete have been reported by
Brady (2000). Some of the data are reproduced in Table C12. The test method requires specimens to be oven-dried prior
to the test but this might lead to excessive cracking of foamed concrete specimens, which would increase absorption.
Nonetheless, the absorption levels, and the values of porosity deft ved from them, are much lower than the air voids calculated
from the mix design information. Thus it can be concluded that a significant number of discrete, or unconnected, air voids
were present in the hardened concrete. By comparison with these data, according to a 1988 report of the Concrete Society
the mean absorption value for concrete is between 3 and 4 per cent. Thus, as could perhaps have been anticipated, foamed
concrete has a much higher capacity for absorbing water than normal weight concrete.
A-29
Table C12 Results of water absorption tests, after Brady (2000)
Foamed concrete t Water absorption (%)
Type 1 30.5 (mean of 3 measurements)

~'See Table C9
Sorptivity indices, calculated using the method given by Hall (1989), from capillary rise tests are given in Table C1. The
values for foamed concrete (with a dry density of 1400 kg/m 3) range from 0.04 to 0.12 ram/taint'Z: this compares with a value
of about 0.020 mm/min v2 for normal weight concrete having a compressive strength of 35 N/ram z, see for example Jones
and Scorey (2000). It can also be seen that foamed concrete formed with a protein-based surfactant had the lowest sorptivity
index, thus confirming the efficiency of the closed cell structure of such concrete.
Water permeability
Brady (2000) reported the results of permeability tests undertaken on specimens of foamed concrete according to Concrete
Society Technical Report 31 (Concrete Society, 1988). Data from these tests are reproduced in Figure C15. Because of
differences in preparing and testing specimens it is difficult to make direct comparisons, but it would seem that the three
types of foamed concrete tested were more than 100 times more permeable to water than normal weight concrete.
Water vapour diffusivity
The results of tests for water vapour diffusivity, using the method described by Dhir et aL (1989a), show, as expected, an
increase in the ease of diffusivity with reducing density. They reported that, for a dry density of 1400 kg/m 3, the rate of
diffusivity for foamed concrete formed from PC/sand was twice that of one formed from PC/PFA.
The values quoted in Table C 1 for foamed concrete are almost double those reported by Jones and Scorey (2000) for normal
weight concrete having a compressive strength of 40 to 50 N/ram 2. The effect of the type of surfactant on the ease of diffusion
of water-vapour has not yet been fully explored, but it could be expected to be in line with the trends observed for other
permeation properties.
Gas permeability
A selection of the results of air permeability tests, using the method described by Dhir et aL (1989b) is given in Table C 13.
The air permeability of foamed concrete is higher than that of normal weight concrete by almost an order of magnitude: the
difference increased with decreasing dry density of the foamed concrete.
Kearsley and Booysens (1998) found that the permeability to oxygen increased with decreasing density of the foamed
concrete. At a dry density of 1500 kg/m 3, foamed concrete had a lower permeability to oxygen than normal weight concrete
having a compressive strength of 25 N/mm 2.
Brady (2000) reported the results of oxygen permeability and diffusion tests undertaken on foamed concrete in accordance
with the TEL in-house procedure and as defined in TP 1303/90/4672 (TEL, 1990a). The data from these tests are reproduced
in Table C 14. The weakest mix, type l, was too permeable for the diffusion tests. The diffusion coefficients for types 2 and
3 were about two orders of magnitude higher than quoted for normal weight concrete having a w/c ratio of 0.7.
A-30
10 .5 ,
t~ 10 .6 . Table C9)
10 ~ .
=
10 -8 .
2Type 3
~5
10 -9 .
10 "1o
1000 1250 1500 1750 2000
Hardened densi~ (kg/m 3)
Figure C15 Relation between water permeability and hardened density, after Brady (2000)
Table C13 Air permeability of foamed concrete, from Dhir et al. (1999)
Inlet pressure (psi) t Intrinsic

permeability*
Mix w/c ratio 110 90 70 50 30 m2xl O ~7
PC/sand 0.3 130 105 75 46 21 18.2

0.3 400 340 245 164 91 25.8
0.4 139 1/8 82 50 25 21.4
0.4 133 i25 90 56 28 18.9
PC/PFA 0.3 25 22 16 8 5.9

0.4 42 35 24 12 I 0.1
Plastic density = 1800 kg/m 3 and PC content = 500 kg/m 3

t 1 psi = 6.89 kN/m 2
* See Section C3.3.3
A-31
Table C14 Permeability o f f o a m e d c o n c r e t e to o x y g e n , after B r a d y (2000)
Oxygen permeability (m 2)
Foamed concrete t" 1 2 3 4 5 6 Mean
(a) Coeff'wients of oxygen permeability

Type 1 1.1xl0 -15 1.1xlO-15 9.4x10 -t6 6.7x10 '6 8.6x10-16 1.1xlO-t~ 9.7x10 -16
Oxygen diffusion coefficient (m2/s)
Foamed concrete ~ 1 2 3 4 5 6 Mean
(b) Coefficients of oxygen diffusion

Type 2 4.6x10-6 4.8x10-6 7.2x10-6 5.1x10-6 6.6x10-6 5.4x10-e 5.6x10-6
Type 3 1.9x]0 -6 1.7x10-6 1.5x10-6 i . 0 x l 0 -6 6.1x10-6 6.1x10 -6 1.2x10-6
t See Table C9
Carbonation
Brady (2000) reported the results of accelerated carbonation tests undertaken on foamed concrete in accordance with pr EN
104-839: 1997. The results are summarised in Figure C16: note that each data point represents the mean of 18 individual
readings. The mean rate of carbonation was reported to be about 5.7 mm/year°~ and at least 50 per cent higher than quoted
by Bamforth (1998) for normal weight concrete having the same cement content.
4.5 ~
4.0- • Type1(seeTableC9)
~- 3.~- Type2
E 3.0-
~ T3 y p e =,~/~r_
.~~" _~,,o
g 2.5-
.o 2.0 -
o= 1.5-
1.0-
0,5 •
0.0
0 5 10 15
Squarerootoftime(days0"5)
Figure C16 Relation between carbonation depth and square root of time for foamed concrete, after Brady (2000)
The carbonation resistance of relatively high strength, i.e. 'structural grade', foamed concrete (using the test method
described by Dhir et al., 1999) is much lower than that of normal weight concrete. As shown in Figure C 17, over a period
of three months or so the depth of carbonation in a PC/sand mix, having a plastic density of 1400 kg/m 3, was in excess of
20 mm. The much higher resistance of the denser foamed concrete might be attributed to its denser stmcture and its higher
cement content.
A-32
25
[ "-(D~ PC/sand, plastic density 1400 kg/m3
PC/sand, plastic density 1800 kg/m3
PC/PFA, plastic density 1400 kg/m3
20 PC/PFA, plastic density 1800 kg/m3
w/c ratio = 0.4 J

g
E
Q 15
E
O
s~
"6 10
0~ ~
0 2 4 6 8 10 12 14 16
Time (weeks)
Figure C17 Carbonation resistance of foamed concrete, after Dhir et al. (1999)
Chloride ion diffusion
Kearsley and Booysens (1998) reported the results of accelerated chloride corrosion tests undertaken using a galvanostatic
system. The results of these tests indicate that the durability of foamed concrete was about equivalent to low-strength normal
weight concrete.
Brady (2000) reported the results of chloride diffusion tests undertaken on foamed concrete in accordance with the TEL in-
house procedure laid down in TP1303/90/4670 (TEL, 1990b). Some of these results are reproduced in Table C 15. In contrast
to the results of the absorption and permeability tests (see above) the results of these tests did not seem to be much influenced
by the hardened density of the concrete. Comparison with the data reported by Bamforth et al. (1997) suggests that the
diffusion coefficient of the three types of foamed concrete at 28-days was about an order of magnitude higher than for normal
weight concrete having a w/c ratio of between 0.5 and 0.6. The coefficient for the foamed concrete reduced about five fold
over a period of about a year, but the chloride ion concentration at the surface increased: such trends are found with normal
weight concrete.
C3.4 O T H E R P R O P E R T I E S A N D C O N S I D E R A T I O N S
C3.4.1 Thermal conductivity

The cellular structure of foamed concrete provides it with a low thermal conductivity, which, according to BCA (1994),
typically is 5 to 30 per cent of that of normal weight concrete. As pointed out by Kessler (1998), in practical terms normal
weight concrete elements would have to be 5 times thicker than foamed concrete ones to achieve similar thermal insulation.
The thermal conductivity of low density foamed concrete can be as low as 0. I 0 W/mK (the lower the value the better the
insulating properties) - see Walker and Clark (1988). This compares to values of between 1.1 and 1.4 W/InK for normal
weight concrete (CIBS, 1980).
A-33
Table C15 Calculated chloride ion diffusion coefficients, after Brady (2000)
Mean
28 day chloride diffusion coefficient (m2/s) surface
chloride
Foamed concrete ~ Specimen 1 Specimen 2 Specimen 3 Mean (%)
(a) 28 days exposure

Type 1 1.22x10 ]o 1.54xi0 -t° 3.73x10 "l° 1.04xI0 to 0.322
Type 2 2.34x 10H~ 4.13xi0 -t° 2.16x10 1° 2.88x10 -t° 0.304
Type 3 1.10xI0 -H~ 1.81xI0 -I° 1.76x10 ]° 1.56x10 t° 0.358
Mean
1 year chloride diffusion coefficient (m2/s) suFfacc
chloride
Foamed concrete "? Specimen 1 Specimen 2 Specimen 3 Mean (%)
(b) 1 year exposure

Type l 5.96x10 ii 5.67x10-]1 4.44xi0 -11 5.36x10-" 0.711
Type 2 7.23x10 "tt 5.16x10 ]] 6.72x10" 6.37x10 -it 0.753
Type 3 6.21x10 "tt 2.70x10 "H 4.33x10 "u 4.41x10 "tt 0.733
t" See Table C9
C3.4.2 Fire and flame spread resistance
The fire resistance of foamed concrete is excellent; at low temperatures it is better in terms of the proportional loss in strength
than normal concrete (Kessler, 1998). But at high temperatures, it suffers from excessive drying shrinkage (Sach and Seifert,
1999). The flame spread resistance of foamed concrete has not been reported but, following from the above, it could be
expected to be better than that of normal weight concrete.
C3.4.3 Embedment of services
Provision for services cast within foamed concrete should he similar to that for well-compacted granular fill. Foamed
concrete is easily re-excavated, and so access to services for future repair is simple.
C3.4.4 R i s k of rodent a t t a c k
Since foamed concrete is a relatively low strength material, there may be a small risk of rodent attack, particularly when used
to embed sewer pipes. Although no such occurrence has been reported the issue might need to be addressed.
C4 P R O D U C T I O N M E T H O D S
Foamed concrete is usually manufactured by mixing together pre-formed cement paste (or mortar) and a relatively stiff
foam, but it can also be formed by adding a foaming agent to the base mix and intimately mixing.
According to Walker and Clark (1988) the rate of production of foamed concrete from a mobile mixer ranges from 3 to
50 m3/hr.
A-34
C4.1 FOAM
A 'wet' or a 'dry' foam can be manufactured. The quality of both types is similar, but dry foam is thought to be more stable,
and hence to produce better quality but lower density concrete, but its rate of production is lower, see Aldridge (2000b) and
McGovern (2000).
C4.1.1 Wet foam

Wet foam can be produced either by compressed air or through the use of a Venturi generator. With the former an aqueous
surfactant solution is pressurised in a vessel and then forced through a small diameter outlet, i.e. the foam gun. With this
technique, the bubbles are quite small and the foam is relatively stiff. With a Venturi generator, the surfactant is sprayed on
a metallic gauze placed inside a metal tube. Air is then sucked through the back of the tube, which combines with the solution
and produces foam with relatively large bubbles. The selection of the foam-generating device is dependent on the volume
output requirements: a Venturi generator is normally used on larger-scale projects.
C4.1.2 D r y foam
Dry foam is produced by spraying a surfactant solution through baffles, this produces a foam with small bubbles. The
resulting foam has a volume of between 20 to 25 times that of the surfactant solution (EABASSOC, 1996) and a density of
between 25 and 80 gms/litre (BCA, 1994). The foaming agents employed can be synthetic compounds (e.g. alkyl sulfates)
or natural organic compounds (e.g. hydrolysed proteins and keratin) that have a stable cellular structure to withstand the
mixing process (Bartos, 1992).
C4.2 MIXING PROCESS
C4.2.1 Direct ready-mix method (preformed)

Following the mixing and testing of the mortar, to ensure that it is homogeneous and of adequate workability, the foam is
prepared and blended gently into the mortar to produce a flowing mix of the required density. (To date little work has been
carried out to determine the minimum workability of a base mix required for adequate incorporation of foam.)
The blending of the foam with the cement paste or mortar can be carried out either at a ready-mix plant or on site, but it must
be undertaken carefully to prevent breakdown of the bubbles of the foam.
The type of mixer, the speed of rotation and mixing time all affect the quality of the foamed concrete. Because they allow
the foam and mortar to flow into each other, rotary drum mixers are effective (Sach and Seifert, 1999). Mixers with paddles
rotating on a horizontal shaft or a screw action in a trough are strongly recommended by DransfieId (2000) because these
encourage 'folding action' of foam into the base mix. The speed of rotation of a ready mix track mixer will affect the foam
dosage required and the properties of the fresh mix. Mixing time is important: short mixing times result in inhomogeneous
foamed concrete, but prolonged mixing at high speeds leads to disintegration of the foam (Kearsley, 1996).
C4.2.2 In-line production

In-line mixing is an alternative to the direct injection of foam into a concrete mixer. In this case the base mix is discharged
into a unit where it is blended with the foam. The density of the material is monitored via an on-board density monitor and
adjustments made to achieve the target plastic density: details of the method have been provided by Aldridge (2000a).
Pumping capability is provided in the unit through a single helical screw pump. According to Aldridge (2000a), the accurate
control of density and the efficiency of this method has led to it being increasingly used to produce foamed concrete on site.
A-35
C5 P R A C T I C A L C O N S I D E R A T I O N S
C5.1 FORMWORK AND MAXIMUM POUR DEPTHS

Unless particular precautions are followed, and the consequences are known and accepted, the depth of a pour should not
exceed 1.5 metres: thicker pours increase the risk of segregation and settlement of the material. Where thicker depths are
required, pours should be carried out in layers of about equal thickness.
C5.2 CURING
As discussed in Section C3.2.1, a number of curing regimes for foamed concrete have been explored. A barrier form of
membrane is recommended for curing on site.
C5.3 ENVIRONMENTAL IMPACT

In terms of environmental issues, there are no additional implications arising from the use of foamed concrete than with
normal weight concrete.
It should be appreciated that because no coarse aggregates are used in the production of foamed concrete, such natural
resources are conserved. Furthermore industrial by-products, such as PFA, are often employed to replace both PC and sand,
particularly in low-density foamed concrete.
C5.4 HEALTH AND SAFETY, CDM REGULATIONS AND COSHH

All health and safety issues raised in the Construction Design and Maintenance (CDM) regulations (i.e. use of goggles,
gloves, ear defenders, etc.) apply to the production, placement and treatment of foamed concrete in both fresh and hardened
states.
Due to the nature of the material special attention must be paid during placing as splashing can occur, and so the use of
goggles should be mandatory. Protective barriers should be placed around the area of the pour to prevent site workers, the
general public and materials from falling in. Such barriers needs to be in place for about 24 hours to allow the foamed
concrete to achieve sufficient strength: the period will vary with the type of cement, the density of foamed concrete, and
the strength, size and depth of the pour.
As regards COSHH, no measures beyond those applying to normal weight concrete production need to be taken for foamed
concrete.
C6 A P P L I C A T I O N S
C6.1 TYPICAL USES

Foamed concrete is used for a variety of applications, ranging from thermal insulation and fire protection to void-fillingand
building elements with successively increasing density and strength requirements.
The Home report (New Roads and Street Works Act, 1991) recommended the use of foamed concrete for reinstating trenches
in roads. This was followed in 1992, by the HAUC specifications, which required minimum strengths of 2 or 4 N/mm 2
(depending on use) and a minimum density of 1050 kg/m3, see Chandler (2000). Such reinstatements are necessary after
pipe laying or repairs are completed. Foamed concrete is self-compacting, does not suffer from excessive internal settlement
A-36
and its load-spreading characteristics prevent the direct transmission of axial loads to the buried services. As a result further
maintenance is seldom required. In addition, foamed concrete develops sufficient strength in a day, and so disruption to
traffic from the street works is low. Because low-density foamed concrete will float, cam must be taken when the trench is
partially filled with water: on the other hand pipes may float where a high density concrete is used and in such cases the pipe
might have to be restrained at the bottom of the trench.
The ability of foamed concrete to flow easily under its own weight has led to its use for void filling,particularly where access
is aproblem. The mostcommon applications are fillingsewerage pipes, wells, cellars and basements ofoId buildings, storage
tanks, tunnels and subways, see for example FEB/Master Builders (I997). However the same precautions as above must
be taken when water is present in the void.
A few bridge abutments in the UK have been built with foamed concrete as the backfill. Because of its strength, the lateral
load imposed on the structure is negligible. And because of its light weight, settlement is much lower than it would be with
a conventional backfilh it therefore allows a substantial reduction in the size of the bridge foundations. For the same reasons,
foamed concrete has been nsed rather thangranularbackfiIltoprovidestabilityon embankment slopes (BCA, 1991b and
to build-up embankments to support overlying bridges.
The light weight and excellent thermal insulation properties of foamed concrete have led to its use as roofing insulation in
the Middle East (EABASSOC, 1996) and South Africa (De Rose and Morris, 1999). Its low density and high tbixotropy
enables roof slopes to be formed (Dransfield, 2000).
Because of its self-levellingproperties, foamed concrete has a]so been used for blinding layers (B asi urski, 2000), and as sub-
floor material to provide thermal and sound insulation to both new and renovated structures (BCA, 1991b).
Other applications include its use as:
• an insulating fill in fire walls or other precast elements;
• a replacement for soils and backfills;
• the construction of cast-in-place piles;
• raft foundations to low rise buildings, such as houses;
• a foundation to sports fields and athletics tracks;
• grouting the interior of tunnels;
• backfilling of voids behind a tunnel lining;
• a foundation base for storage tanks.
C6.2 INNOVATIVE USES

Some innovative applications, as described by Basiurski (2000), include its use as an underfloor heating system, a sprayed
insulating material for concrete domes, decorative panels for wail fences and low-height retaining walls, and for producing
light-weight GRC panels and precast elements (e.g. fence posts, poles, lintels, window and door frames). It has also been
considered for constructing igloos, dog kennels, bullet-absorbing wails, concrete aeroplanes and ornaments.
A-37
C6.3 DETAILS OF PARTICULAR PROJECTS
Approach embankment, Colchester
Foamed concrete was used for the construction of approach embankments to a railway bridge in Colchester, details have
been given by Pickford and Crompton (1996) and Anon (1996). The foundation subsoil at the site was a soft clay and by
using foamed concrete rather than normal weight backfill, the required number of piles to the embankment reduced
substantially and the wall thickness of the abutment was halved: see Aldridge (2000b). As both a relatively low density and
high strength were required, two mixes were used. The upper metre was formed from foamed concrete having a compressive
strength of 4.5 N/mm 2(28-day) and a plastic density of 1450 kg/m3. Below this a lower density (1250 kg/m3), lower strength
(3.0 N/mm 2) mix was specified. The concrete was foamed on site, and measurements of plastic density and specimens for
monitoring compressive strength were taken every few loads. A total of 4000 m3of foamed concrete was placed at this site.
Bridge deck, North Wales
Foamed concrete was used to replace the failed waterproofing membrane of the 25-year old bridge deck of a flyover
connecting Llandudno junction and Deganwy. A PC/sand (34%/66%) mix having a dry density of 1300 kg/m ~and a
28-day strength of 4 N/ram 2 was specified. This was placed to a depth of up to 400 mm and then covered with a black top
wearing course: details have been provided by Pickford and Crompton (1996).
Kingston bridge
The existing 7-arch masonry bridge at Kingston had to be widened to increase traffic capacity: a weight limit was placed
on the existing bridge in 1993. The work involved the use of precast concrete arch shells, to resemble the older structures,
with a concrete saddle cast over the arches and foamed concrete used to fill the voids between the arches and final road
surface. For the upper 700 mm, a PC/sand (31/69) mix was specified with a dry density of 1400 kg/m3and a 28-day strength
of 7 N/mm 2. At greater depths, a PC/PFA/sand (64/12/24) mix having a density of 1600 kg/m 3 and strength of
1 N/mm 2 (28-day) was specified: details have been provided by Aldridge, (2000b).
Access road, Canary Wharf

The Canary Wharf project in London Docklands involved the construction of two access roads, the foundations of which
were formed from foamed concrete to reduce the vertical load on the foundations and thereby reduce settlement. Two mixes
were specified. The upper layer, for use a metre below the road surface, was 0.5 m thick and had a dry density of
670 kg/m 3 and a minimum 28-day compressive strength of 0.87 N/mm 2. Beneath this layer, the concrete had a density of
480 kg/m3 and 28-day strength of 0.27 N/ram2: details have been provided by Van Deijk (1991).
Numerous examples exist of the use of foamed concrete for filling voids as part of stabilisation works. These include:
the Heathrow railway tunnel, part of which collapsed during construction;
an undermined road in Lublin, Poland, which was threatened by collapse when the ground below was washed away
by heavy rain. Old sewer pipes and inspection chambers were also filled with foamed concrete, prior to the
construction of a new sewer system;
iii the placement, in 1986, of 19,500 m3of foamed concrete, having a dry density of 500 kg/m 3, as harbour fill in Holland.
The maximum permissible amount of settlement in the first year was specified to be less than 80 ram: details have
been given by Basiurski (2000).
A-38
C7 R E F E R E N C E S
Anon (1996). UK's largest foamed concrete pour for railway embankment. Quality Concrete, VoI 2, No 2, p 53.
Aldridge D (2000a). Foamed concrete. Concrete, Vol. 34, No. 4, pp 20-22.
Aldridge D (2000b). Foamed concrete for highway bridge works. One-day awareness seminar 'Foamed concrete:
properties, applications and potential' held at the University of Dundee. pp 33-41.
Anon (2000). Interim DETR project report "Development offoamed concretefor insulating trenchfillfoundations and
ground-supported slabs of low-rise domestic buildings'. Dundee: University of Dundee.
Ansell T (2000). Personal communication.
ASTM C266-89. Test method for time of setting of hydraulic cement paste by Gillmore Needles.
ASTM C666-90. Test for the resistance of concrete to rapidfreezing and thawing.
ASTM C869-91. Standard specification for foaming agents used in making preformed foam for cellular concrete.
Bamforth P B, Price W F and Emerson M (1997). An international review of chloride ingress into structural
concrete. Contractor Report CR359. Crowthorne: TILL Limited.
Bamforth P B (1998). Appendix 1: Modelling carbonation rates using the concept of equivalent chemical buffering
capacity. In Guidance on the selection of measures for enhancing reinforced concrete durability. Report prepared under
the DOE Partners in Technology Programme Contract CI 39/3/367 (cc 967). To be published by the Concrete Society.
Bartos P J M (1992). Fresh concrete. Properties and tests. Amsterdam: Elsevier.
Basiurski J (2000). Foamed concrete for void filling, insulation and construction. One-day awareness seminar
'Foamed concrete: properties, applications and potential', held at University of Dundee. pp 42-52.
Brady K C (2000). An investigation into the properties offoamed concrete. Project Report PR/IS/99100. Crowthorne:
TRL Limited. (Unpublished report available on direct personal application only)
Brady K C and Greene M J (1997). Foamed concrete: a review of materials, methods of production and
applications. Project Report PR/CE/149/97. Crowthorne: TRL Limited. (Unpublished report available on
direct personal application only)
British Cement Association (1990). First interim report on foamed concrete. Report RAI.007.00.1. Slough: BCA.
British Cement Association (1991a). Foamed concrete for improved trench reinstatements. Report Ref. 46.043.
Slough: BCA.
British Cement Association (1991b). Foamed concrete. Report Ref. 46.04 i. Slough: BCA.
British Cement Association (1994). Foamed concrete; Composition and properties. Report Ref. 46.042. Slough: BCA.
Brewer W E (1996). Controlled low strength materials (CLSM). Radical concrete technology. (Eds Dhir R K and
P C Hewlett). pp 655-667. London: E & FN Spon.
A-39
British Standards Institution. London
BS 12: 1996. Specification for Portland cement.

BS 882:1992. Specification for aggregates from natural sources for concrete.
BS 915: 1983. Specification for high alumina cement.
BS 1200:1976. Specifications for building sands from natural sources.
BS 1881 : Part 5:1970. Method of testing hardened concrete for other than strength.
BS 1881 : Part 117:1983. Testing concrete; Method for determination of tensile splitting strength.
BS 1881 : Part 118:1983. Testing concrete; Method for determination of fiexural strength.
BS 1881: Part12I: 1983. Testing concrete; Method for determination of static modulus of elasticity in
compression.
BS 1881 : Part 203: 1986. Testing concrete; Recommendations for measurement of velocity of ultrasonic
pulses in concrete.
BS 3892: Part 1:1997. Pulverised-fuel ash; Specification for pulverised-fuel ash to be used as a Type 1
addition.
BS 3892: Part 2:1997. Pulverised-fuel ash; Specification for pulverised-fuel ash for use in cementitious
grouts.
BS 4027: 1980. Specification for sulphate-resisting Portland cement.
BS 4550: 1978. Methods of testing cement.
BS 6699: 1992. Specification for ground granulated blastfurnace slag for use with Portland cement.
BS EN 197: Part 1: 2000. Cement. Composition, specifications and conformity criteria for common cements.
BS EN 450: 1995. Fly ash for concrete. Definitions, requirements and quality control.
BS EN 12350: Part 2: 2000. Method for determination of slump.
BS EN 12350: Part 5: 2000. Method for determination of flow.
CEN (1997). Determination of accelerated carbonation in hardened concrete, pr EN 104-839.
Brussels.
CEB-FIP (1993). Model Code 1990. Thomas Telford, London.
Chandler J W E (2000). Highway reinstatement with foamed concrete. One-day awareness seminar 'Foamed concrete:
properties, applications and potential' held at University of Dundee. pp 26-32.
Chartered Institution of Building Services (1980). C1BS Guide A3. Thermal properties of building structures.
London: CIBS.
Concrete Society (1988). Permeability testing of site concrete - a review of methods and experience. Report 31.
London: The Concrete Society.
De Rose L and Morris J (1999). The influence of the mix design on the properties of micro-cellular concrete.
Specialist techniques and materials for concrete construction, Proc. Int. Conf 'Creating with concrete' (Eds Dhir R K
and N A Henderson) held at University of Dundee. pp 185-197. London: Thomas Telford.
Dhir R K, Levitt M and Wang J (1989a). Membrane curing of concrete: water vapourpermeability of curing
membranes. Magazine of Concrete Research, Vol. 41, No. 149, pp 221-228.
Dhir R K, Hewlett P C and Chan Y N (1989b). Near surface characteristics of concrete: intrinsic permeability.
Magazine of Concrete Research, Vol. 4I, No. 147, pp 87-97.
Dhir R K, Jones M R and Nicol L A (1999). Development of structural grade foamed concrete. DETR Research
Project. Scotland: University of Dundee.
A-40
Dransfield J M (2000). Foamed concrete: Introduction to the product and its properties. One-Day awareness seminar
on 'Foamed concrete: properties, applications and potential' held at University of Dundee, pp 1- I 1.
EABASSOC (1996). Lightweight foamed concrete. EAB Associates Products Information leaflet, 4/96.
FEB/Master Builders (1997). Private communication.
Field S N and Bamforth P B (1991). Long term properties of concrete in nuclear containment structures. Civil
engineering in the nuclear industry. London: Thomas Telford.
G u t m a u n P F (1988). Bubble characteristics as they pertain to compressive strength and freeze-thaw durability.
ACI Materials Journal, September-October, pp 361-365.
Hall C (1989). Water sorptivity of mortars and concretes: a review. Magazine of Concrete Research, Vol. 41, No. 147,
pp 51-61.
Helmuth R (1987). Fly ash in cement and concrete. USA, Illinois: Portland Cement Association.
Hoarty J T (1990). Improved air-entraining agents for use in concretes containing pulverised fuel ashes. In admixtures
for concrete: Improvement of properties (Ed E Vazquez). Proc ASTM Int Symposium, Barcelona, Spain, pp 449-459.
London: Chapman and Hail.
Jones M R (2000). Foamed concrete for structural use. One-day awareness seminar on 'Foamed concrete: properties,
applications and potential' held at University of Dundee, pp 54-79.
Jones M R and Seorey V E (2000). Optimisation of concrete for highway environments. Internal report. Dundee:
University of Dundee.
Kearsley E P (1996). The use of foamcrete for affordable development in third world countries. Appropriate Concrete
Technology. Proc. Int. Conf 'Concrete in the service of mankind' (Ed Dhir R K, M J McCarthy) held at University of
Dundee, pp 233-243. London: E & FN Span.
Kearsley E P (1999). Just foamed concrete - an overview. Specialist techniques and materials for concrete
construction. Proc. Int. Conf 'Creating with concrete' (Ed Dhir R K and N A Henderson) held at University of Dundee,
pp 227-237. London: Thomas Telford.
Kearsley E P and Booysens P J (1998). Reinforced foamed concrete - Can it be durable ? Concrete Beton, No. 91, pp 5-9.
Kearsley E P and Mostert H 17 (1997). The use offoamcrete in Southern Africa. Proc. ACI Int. Conf on high
performance concrete, SP 172-48, pp 919-934.
Kessler H G (1998). Cellular lightweight concrete. Concrete Engineering International, pp 56-60.
Leptokaridis C (2000). Development of insulating foamed concrete foundations for low-rise buildings. MSc
dissertation. Dundee: University of Dundee.
M c G o v e r n G (2000). Manufacture and supply of ready-mix foamed concrete. One day awareness seminm" on "Foamed
concrete: Properties, applications and potential' held at University of Dundee. pp 12-25.
Mellin P (1999). Development of structural grade foamed concrete. MSc dissertation. Dundee: University of Dundee.
M y e r s D (1992). Surfactant science and technology. Second edition. Cambridge: VCH Publishers Inc,.
N e w Roads and Street Works Act (1991). Specification for the reinstatement of openings in highways.
Oluokun F A (1991). Prediction of concrete tensile strength from compressive strength: evaluation of existing
relations for normal weight concrete. ACI Materials Journal, Vol. 88, No. 3, pp 302-309.
A-41
Piekford C and Crompton S (1996). Foamed concrete in bridge construction. Concrete, November/December,pp 14-15.
Regan P E and Arasteh A R (1984). Lightweight aggregate foamed concrete. Low-cost and energy saving materials,
Ch. 42, pp 123-138.
RILEM (1977). Methods of carrying out and reporting freeze~thaw tests on concrete with de-icing chemicals. CDC 2
Salt scaling test.
Saeh J and Seifert H (1999). Foamed concrete technology: possibilities for theru~al insulation at high temperatures.
CFI Forum of Technology, DKG 76, No. 9, pp 23-30.
Tam C T, Lira T Y, Ravindrarajah R S and Lee S L (1987). Relationship between strength and volumetric
composition of moist-cured cellular concrete. Magazine of Concrete Research, Vol. 39, No. 138, pp 13-18.
Tattersall G H (1991). Workability and quality control of concrete. London: E & FN Spon.
Taylor R W (1988). The effect of backfill material on the loading on shallow buried pipelines and on the settlement of
trench reinstatements. Workshop on foamed concrete held at British Cement Association, Slough. pl I.
Taywood Engineering Ltd (1990a). Measurement of oxygen diffusion coefficient of hardened concrete or mortar.
Unpublished testing procedure TP 1303/90/4672.
Taywood Engineering Ltd (1990b). Determination of chloride content of hardened concrete by potentiometric
titration. Unpublished testing procedure TP 1303/90/4670.
Van Deijk S (1991). Foam concrete. Concrete, July/August, pp 49-54.
Walker B and Clark A (1988). Introducing foamed concrete. Concrete Quarterly, Winter, pp 24-25.
A-42
BIBLIOGRAPHY
Anon (1997). Cellular concrete pour relieves earthquake damage. Civil Engineering, November, pp 14-15.
Bartos P J M (1994). Workability of special fresh concretes. Proc. Int. RILEM workshop 'Special concretes;
workability and mixing' (Ed P J M Bartos) held at Paisley, Scotland on 2-3 March 1993. London: E & FN Spon.
Cleland D J and Gilfillan J R (1994). Specifyingflowing concrete - a case study. Proc. Int. RILEM Workshop
'Special concretes; workability and mixing' (Ed P J M Bartos) held at Paisley, Scotland on 2-3 March 1993.
pp 209-214. London: E & FN Spon.
Dhir R K and Yap A W F (1983). Superplasticized high-workability concrete: some properties in the fresh and
hardened states. Magazine of Concrete Research, Vol. 35, No. 125, pp 214-228.
Dimond C R and Bloomer S J (1977). A consideration of the DINflow table. Concrete, December, pp 29-30.
Hobbs D W (1994). Workability and water demand. Proc. Int. RILEM workshop 'Special concretes; workability and
mixing' (Ed P J M Bartos) held at Paisley, Scotland on 2-3 March 1993. pp 55-66. London: E & FN Spon.
Karl S and W~rner J - D (1994). Foamed concrete - mixing and workability. Proc. Int. RILEM workshop 'Special
concretes; workability and mixing' (Ed P J M Bartos), held at Paisley, Scotland, pp 217-224. London: E & FN Spon.
Legrand C (1994). Workability and rheology. Proc. Int. RILEM Workshop 'Special concretes; workability and
mixing' (Ed P J M Bartos) held at Paisley, Scotland on 2-3 March 1993. pp 51-54. London: E & FN Spon.
Malou Z and Cabrinac R (1999). Configuration of pores on the mechanical and thermal characteristics of cellular
concrete through a homogenization method. Specialist techniques and materials for concrete construction, Proc. Int.
Conf 'Creating with concrete' (Ed Dhir R K and N A Henderson) held at University of Dundee. pp 209-218. London:
Thomas Telford.
M B T Admixtures Ltd (undated). Barracell - Rheoplastic lightweight foam concrete. Technical data sheets.
Moorfield G (1994). Filling a gap in the market. Concrete. May/June, pp 12-14.
Reichverger Z (1986). Using an impact device with sliding drop collar for in situ evaluation of compressive strength of
insulating cellular concrete. Journal of Testing and Evaluation, JTEVA, Vol. 14, No. 6, pp 298-302.
Wimpenny D E (1996). Some aspects of the design and production of foamed concrete. Appropriate Concrete
Technology. Proc. Int. Conf 'Concrete in the service of mankind' (Ed Dhir R K and M J McCarthy) held at University
of Dundee. pp 245-254. London: E & FN Spon.
ZolIo R F and Hays C D (1998). Engineering material properties of a fiber reinforced cellular concrete.
ACI Materials Journal, Vol. 95, No. 5, pp 631-635.
A-43
U
ISSN 1365-6929
t~
r.t3
A
V
A
V
A
v
orb
<
O
TRL Limited Q
O]dWokingham Road Crowthorne Berkshire RG456AU United Kingdom
Tel: +44 (0)1344 773131 Fax:+44 (0)1344 770356 O
E-Mail: enquiries@trl.co.uk
www.trl.co.uk r~
J <

TRL Application Guide AG39

Uploaded by

Copyright:

Available Formats

You might also like

TRL Application Guide AG39

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

TRL Application Guide AG39

Uploaded by

Copyright:

Available Formats

] •H I G H W A Y S

Specification for foamed concrete

by K C Brady, G R A Watts and M R Jones

Application Guide AG39

K C Brady, G R A Watts and M R Jones

Application Guide AG39

This report has been produced by TRL Limited, under/as part

TRL is committed to optimising energy efficiency, reducing

APPENDIX A: GLOSSARY OF TERMS A-I

APPENDIX B: GUIDE FOR SPECIFICATION AND

APPENDIX C: FOAMED CONCRETE A-3

Version No: 1 Application Guide A G39 Ocwber 2001

Version No: 1 Application Guide A G39 October 2001

iv properties of foamed concrete, Brady (2000).

2.1 DEFINITION, CONSTITUENTS AND PROPERTIES

Version No: 1 Application Guide AG39 October 2001

Figure 1 Typical nature of foamed concrete

Version No: 1 Application Guide A G39 October 2001

Figure 2 Small-scale laboratory foam production and typical nature of foam

Figure 3 Large-scale on-site foam production

Version No: 1 Application Guide AG39 October 2001

i Ease of application and re-excavation (if necessary).

2.3 CURRENT USAGE

. Test carried out Tested by

Quality control test At point of delivery Laboratory Contractor Supplier

Foam: density '1 ¢" • ¢"

Version No: 1 Application Guide AG39 October 2001

3.1 FOAM DENSITY AND STABILITY

Figure 4 Use of glass measuring cyclinder for determining density of foam

3.2 PLASTIC DENSITY

Version No: 1 Application Guide AG39 October 2001

Figure 5 Measurement of the plastic density of foamed concrete

(76.2 mm) 3"

Version No: 1 Application Guide A G39 October 2001

Flow rate, using apparatus

I 1 litre in < 1 minute At Constant flow

t Used with Classes ] and 2 only

Version No: 1 Application Guide AG39 October 2001

3.5 CUBE STRENGTH

Version No: 1 Application Guide A G39 October 2001

Figure 8 View of BRE screed tester

Version No: 1 Application Guide AG39 October 2001

4.2.1 Essential items

Type of bindefl" " Standard

Cements containing pulverised-fuel ash (PFA) or limestone

Combinations made up in a concrete mixer Combination o f the proportions and properties to

"l"Precluding cements containing slag.

Version No: 1 Application Guide A G39 October 2001

Type Designation Standard

Lightweight Lightweight aggregates for concrete BS 3797:1990

Normal weight Sand from natural sources for concrete BS 882:1992

4.2.2 Optional items

Maximum cube strength

Version No: 1 Application Guide AG39 October 2001

Sustained core temperature

Version No: 1 Application Guide AG39 October 2001

ii Are other site works likely to be affected by the non-conformity ?

Version No: 1 Application Guide AG39 October 2001

British Standards Institution. London.