The INSTITUTE OF CONCRETE TECHNOLOGYict.concrete.org.uk/downloads/yearbooks/ict_2007.pdf · 3...

Yearbook: 2007-2008

CONCRETE TECHNOLOGYINSTITUTE OF

The

12th Edition

Published by:THE INSTITUTE OF

CONCRETE TECHNOLOGY4 Meadows Business Park,

Blackwater, Camberley, Surrey GU17 9AB

Tel/Fax: 01276 37831Email: [email protected]

Website: www.ictech.org

ICT YEARBOOK 2007-2008

EDITORIAL COMMITTEE

Professor Peter C. Hewlett (Chairman)GROUP TECHNICAL ADVISOR FOR

JOHN DOYLE GROUP PLCPRINCIPAL CONSULTANT -

BRITISH BOARD OF AGRÉMENT LTD

Peter C. OldhamCHRISTEYNS UK LTD

Dr. Bill PriceLAFARGE CEMENT UK

Graham TaylorINSTITUTE OF CONCRETE TECHNOLOGY

Laurence E. PerkisINITIAL CONTACTS

Rights reserved. No part of this publication maybe reproduced or transmitted in any formwithout the prior written consent of the

publisher. The comments expressed in thispublication are those of the Author and not

necessarily those of the ICT.

ISSN 1366 - 4824£50.00

Engineering CouncilProfessional Affiliate

3

Yearbook: 2007-2008

CONCRETE TECHNOLOGYINSTITUTE OF

The

CONTENTS PAGE

PRESIDENT’S PERSPECTIVE 5By Bryan MarshPresident, INSTITUTE OF CONCRETE TECHNOLOGY

THE INSTITUTE 6

COUNCIL, OFFICERS AND COMMITTEES 7

FACE TO FACE 9 - 11A personal interview with Ian Ferguson

MILESTONES IN THE HISTORY OF CONCRETE TECHNOLOGY: 13 - 17THE HISTORY OF TRAINING AT THE C&CABy Duncan Pomeroy and Graham Taylor

ANNUAL CONVENTION SYMPOSIUM: 19 - 122PAPERS PRESENTED 2006128ADVANCED CONCRETE TECHNOLOGY DIPLOMA 123 - 130SUMMARIES OF PROJECT REPORTS 2006

RELATED INSTITUTIONS & ORGANISATIONS 131

4

55

PRESIDENT’S PERSPECTIVE

This time last year I wrote that we were inmid-stride of a momentous step in theforward march of the Institute. Grand

words indeed but hopefully not hollow ones. If allhas gone to plan, by the time you read this we willhave completed that momentous first step and besettled into a steady pace towards a new era ofstability and development following our mergerwith the Concrete Society.

We are now the ‘professional wing’ of theSociety but, just as importantly, we are still verymuch The Institute of Concrete Technology. Youwon't have seen much change on the outside butbehind the retained façade we are building arobust new structure designed, in the besttraditions of concrete technology, for bothstrength and durability. Probably, the mostobvious difference has been the retirement of ourExecutive Officer, Graham Taylor, after long anddistinguished service to the Institute which hegraciously extended to help us through our periodof transition.

The 2007 Annual Convention was anoutstanding technical success as, indeed, you willsee from the collected papers in this yearbook.The quality of the speakers, the depth and breadthof the topics they covered, and the chairmanshipof the event were first class. Attendance was,however, down on normal which was very

disappointing especially considering the hugeamount of voluntary work put in by the EventsCommittee. Our annual convention is the bestopportunity many members are likely to get in ayear for that vital Continuing ProfessionalDevelopment and ‘networking’, to say nothing ofthe late nights trying to tease industrial secretsfrom colleagues over a few halves of shandy. The2008 Convention promises to be just as good soplease come along and support the main ICTevent of the year - you won't be disappointed.And I'd like to extend a very big welcome to allmembers of The Concrete Society to come alongand find out what the ICT is all about and perhapsconsider membership. Better still, how abouttackling the Diploma in Advanced ConcreteTechnology? This is the foundation stone of theInstitute and still very much the jewel in ourcrown, if you will permit me to indulge in mixedmetaphors. It remains the prime qualification forfull membership and provides a route toprofessional status through the EngineeringCouncil. To become a holder of the diplomarequires demonstration of a thorough knowledgeand understanding of concrete technology in itswidest sense. There is probably no bettereducation in our field than preparation for theACT diploma and success in the examinations is anachievement of which one can be truly proud.

PRESIDENTINSTITUTE OF CONCRETE TECHNOLOGY

6

INTRODUCTIONThe Institute of Concrete Technology was

formed in 1972. Full membership is open to allthose who have obtained the Diploma inAdvanced Concrete Technology. The Institute isinternationally recognised and the Diploma hasworld-wide acceptance as the leading qualificationin concrete technology. The Institute sets higheducational standards and requires its members toabide by a Code of Professional Conduct, thusenhancing the profession of concrete technology.The Institute is a Professional Affiliate body of theUK Engineering Council.

In 2007 the ICT merged with the ConcreteSociety to become the professional wing of theSociety whilst retaining its own identity.

MEMBERSHIP STRUCTUREA guide on ‘Routes to Membership’ has been

published and contains full details on thequalifications required for entry to each grade ofmembership, which are summarised below:

A FELLOW shall have been a CorporateMember of the Institute for at least 10 years andshall have a minimum of 15 years appropriateexperience, including CPD records from the dateof introduction.

A MEMBER (Corporate) shall hold theDiploma in Advanced Concrete Technology andwill have a minimum of 5 years appropriateexperience (including CPD). This will have beendemonstrated in a written ‘Technical andManagerial/Supervisory Experience Report’. Analternative route exists for those not holding theACT Diploma but is deliberately more onerous.

AN ASSOCIATE shall hold the City and GuildsCGLI 6290 Certificate in Concrete Technology andConstruction (General Principles and PracticalApplications) and have a minimum of 3 yearsappropriate experience demonstrated in a writtenreport. An appropriate university degree exempts aGraduate member from the requirement to holdCGLI 6290 qualifications. Those who have passedthe written papers of the ACT course but have yetto complete their Diploma may also becomeAssociate members. All candidates for Associatemembership will be invited to nominate acorporate member to act as SuperintendingTechnologist.

A TECHNICIAN holding the CGLI 5800Certificate in Concrete Practice must also submit awritten report demonstrating 12 monthsexperience in a technician role in the concreteindustry. An alternative route exists for those whocan demonstrate a minimum of 3 yearsappropriate experience in a technician role. Allcandidates for Technician membership will beinvited to nominate a corporate member to act asSuperintending Technologist. There is no minimumage limit in this grade.

A GRADUATE shall hold a relevant universitydegree containing a significant concretetechnology component. All candidates forGraduate membership will be invited to nominatea corporate member to act as SuperintendingTechnologist.

The STUDENT grade is intended to suit twotypes of applicant.

i) The school leaver working in the concreteindustry working towards the Techniciangrade of membership.

ii) The undergraduate working towards anappropriate university degree containing asignificant concrete technology component.

All candidates for Student membership will beinvited to nominate a corporate member to act asSuperintending Technologist. There is no minimumage limit in this grade. There is a limit of 4 years inthis grade.

Candidates are not obliged to attend anycourse (including the ACT course) prior to sittingan examination at any level.

Academic qualifications and relevant experiencecan be gained in any order for any grade ofmembership.

Corporate members will need to be competentin the science of concrete technology and havesuch commercial, legal and financial awareness asis deemed necessary to discharge their duties inaccordance with the Institute’s Code ofProfessional Conduct.

Continuing Professional Development (CPD) iscommon to most professions to keep theirmembers up to date. All members exceptstudents, are obliged to spend a minimum of 25hours per annum on CPD; approximately 75% ontechnical development and 25% on personaldevelopment. The Institute’s guide on ‘ContinuingProfessional Development’ includes a record sheetfor use by members. This is included in theMembership Handbook. Annual random checksare conducted in addition to inspection at times ofapplication for upgraded membership.

ACT DIPLOMAThe Institute is the examining body for the

Diploma in Advanced Concrete Technology. Residential courses are run in South Africa andAustralia. The worldwide web-based course is runfrom the UK, starting in September of alternateyears. Further details of this course can be foundon the website: www.act-course.co.uk and the ICToffice has details of the others.

THE INSTITUTE

7

EXAMINATIONSCOMMITTEE

COUNCILADMISSIONS AND

MEMBERSHIPCOMMITTEE

FINANCECOMMITTEE

MARKETINGCOMMITTEE

EVENTSCOMMITTEE

COUNCIL, OFFICERS AND COMMITTEES - SUMMER 2007

Dr. J.J. ROBERTSChairman

G. TaylorSecretary

R.A. Binns

Dr. P.L.J. Domone

R. Gaimster(corresponding)

J. Lay

Dr. J.B. Newman

B. Raath (corresponding)

R. Ryle

Dr. R.P. West

J.D. Wootten

W. WILDChairman

C.D. Nessfield

B.K. Marsh

Dr. B.J. MAGEEChairman

G. TaylorSecretary

I.R. Berrie

D.G. King(corresponding)

R.J. Majek

C.D. Nessfield

M.S. Norton

Dr. B.K. MARSHPresident

K.C. SutherlandHon Secretary

W. WildHon Treasurer

R. A. Binns

M.D. Connell

I.F. Ferguson

M.G. Grantham

A.M. Hartley

B.J. Magee

R.J. Mangabhai

C.D. Nessfield

P.C. Oldham

M.D. CONNELLChairman

G. TaylorSecretary

Dr. W.F. Price

J.D. Wootten

I.F. FERGUSONChairman

G. TaylorSecretary

R.G. Boult

P.M. Latham

P.L. Mallory

P.C. Oldham

B.C. Patel

S.D. Pepper

G. Prior(corresponding)

R.S. Young

SCOTTISH CLUBCOMMITTEE

R.C. BROWNChairman

G. PriorSecretary

L.R. BakerTreasurer

J.G. Bell

I.A Callander

K.W. Head

J. Wilson

EXECUTIVE OFFICER

G. TAYLOR

TECHNICAL ANDEDUCATIONCOMMITTEE

R.A. BINNSChairman

J.V. TaylorSecretary

L.K. Abbey

M.W. Burton

Dr. A.J. Dowson

R.J. Greenfield

R. Hutton

C.B. Richards

S.M. Walton

A.T. Wilson(corresponding)

8

9

Q: Ian, due to your role as arranger of theannual convention and other events, yourface is now well known to many members.Before we talk about your work with theInstitute and the concrete industry, tell us alittle bit about yourself. You hail from bonnieGlasgow, I believe.

A: Aye laddie, Glasgow born & bred, whichhelped finely hone my gentle, caring, passive andgenerous nature. As well as my mother and sister,my father cared for very several off-licencepremises around Scotland, which helps explain myteetotal beliefs. When I finally grew older (malesget older, we never grow up!) I met the girl of mydreams, married and migrated south to milderclimes. In 1974 I joined Tarmac Construction inthe north east of England constructing heavy-dutyconcrete floors, heavily reinforced walls (thebuilding had to be resistant to the force ofexplosion – helps concentrate the mind!) and theusual foundation and road-base stuff. Theoutdoor life my job offered was great in thesummer, however it was an endurance test duringthe winter months. It wasn’t long before the callof precasting beckoned – the allure of shelteredaccommodation was too great to resist and so Ijoined Dow Mac Concrete Ltd at their Eaglescliffefactory where we cast prestressed bridge beams,railway sleepers, double-T beams together withsteel reinforced box culverts etc. I remained thereuntil 1986 when I joined Marshalls Mono Ltd asTechnical Manager at their Norton works, whichfactory later relocated to Eaglescliffe.

Q: So, you are now comfortably domiciledin Yorkshire. I understand that you haveestablished connections between work andfamily

A: Some 10 years ago, our company wasimplementing an Enterprise Resource Package andI was seconded to the modelling team for 2 years,

working at our Halifax office, after which I finallyrelocated to West Yorkshire, remaining at Halifaxas Technical & Research Manager andsubsequently Group Concrete Technologist.

We have always been a close-knit family; myson-in-law, Roy, joined Marshalls in 1994 workingwithin our production facility at Norton and laterEaglescliffe; once he had qualified in ConcreteTechnology & Construction and achievedAssociate Membership to the ICT, he wasappointed Laboratory Supervisor at Eaglescliffereporting to myself (unenviable position. I’d tellmy daughter if he didn’t do as I asked, she cangive grief big time; something I’d had years ofpreviously!). A position came up at our Halifax siteas Central Laboratory Manager, which Roy appliedfor and was appointed (though I didn’t interviewhim. I’m innocent – thought we had made goodour escape!). Alas the family were once morereunited and living in perfect harmony (Royreporting to me, I reporting to my daughter…)with concrete invariably the centre ofconversation (yes, we are that sad) around thedinner table.

Q: Your position as Group ConcreteTechnologist with a company whichpioneered and lead forward much semi-dryconcrete technology suggests that you havebeen involved in some innovativedevelopments. How important is newtechnology to you, and how do you see theneed for innovation clashing with therequirement to produce to National andEuropean standards?

A: Innovation is at the forefront of everythingwe at Marshalls do, it’s our life-blood and wepositively thrive on it. When I first joinedMarshalls, back in 1986, I had great pleasure inhaving the company of both Allan Dowson, whowas Head of Research & Development, and John

FACE TO FACE

WITH IAN FERGUSON

Ian Ferguson is the Group Concrete Technologist for Marshalls,based at the head office in Halifax. He has considerableexperience and expertise in precast technology and production, forboth wet cast and semi-dry processes.Ian represents Associate Members on the ICT Council, where hevoices opinions on matters which affect Associates and membersin general. Ian also chairs the ICT Events committee, whichorganises the Annual Convention and Technical Symposium, a taskwhich requires much effort in organising the programme,speakers, etc and the generally smooth running of this annual

event, together with the other technical meetings and events held by the Institute.He is also very active within the Concrete Society. Ian’s Face To Face meeting anddiscussion was held with Editorial Committee Member Peter Oldham at an ICT eventin Richmond, Yorkshire.

10

Wilson, our previous Technical Manager based atFalkirk in Scotland. Both colleagues affordedmuch knowledge of semi-dry concrete, thetechnology of which differs greatly fromconventional concrete. I found the subjectfascinating; a code to be cracked, new territory tobe discovered – 21 years on and I’m still learning!

The arrival of new National & Europeanstandards was a positive move for semi-dryconcrete as previous standards proved quiteprohibitive in materials choice. In those days therewere no durability tests identified to prove thatsuch products were fit for purpose and sorestrictions were put in place to ensure that bestpractice was maintained. The introduction of newstandards includes, amongst other tests, adurability test method and so materialsspecification is less prohibitive. This allowedfurther innovation to move forward, especially theneed to reduce CO2 emissions – a subject thatmyself, Marshalls and our wider customers arepassionate about.

Q: You are also obviously involved in thewet-casting and wet pressing of concreteunits as well as semi-dry production. Are anyof the various materials technologies andproduction methods of especial interest toyou?

A: It would be fair to say that I find concretein general, regardless of process type, interesting.I enjoy pushing concrete to its limits, the moreyou probe the more possible the impossiblebecomes. Concrete is a fascinating subject that isdynamic. We live in a constantly changingmodern world which demands ever-increasinginnovation that aids environmental and designrequirements; pressures are often high – evenintense at times, however the elation that successbrings is worth the effort.

Q: You appear to take great interest inthe social and information-sharing sides ofthe industry, being active on ICT Council,chairman of the ICT Events Committee, andwith an active role in the Concrete Society.These activities must take up much of yourtime; you obviously enjoy arranging andorganising?

A: I recall my father saying “if you wantsomething done, ask a busy man” and he wasone very busy man. I guess his lifestyle influencedme to some extent as he was always on the go(which meant the entire family were always onthe go!) and often included our extended familyand friends. I believe knowledge should be sharedand life should never exclude fun! Concrete hasoffered myself and my family a good life over theyears. I have gleaned knowledge from people

across the globe and very many of these havedeveloped into good friends and colleagues. Iview my involvement and activities within theindustry as a vehicle to give something back andto provide the opportunity of giving knowledgeand networking to both young and new entrantsinto our industry.

Q: And how much of this time goes intowhat is the major annual event, theConvention?A: Convention or Annual Symposium is verymuch the main focus of the Events Committee, afine band of dedicated and enthusiastic Institutemembers who, like me, live life on the edge (ofmadness!) and enjoy ever increasing challenges inlife. Firstly the location and date is set, withfurther time spent on developing a theme for theSymposium which should reflect the currentinterest that a wider audience would findattractive, followed by brain-storming sessions toidentify speakers who are experts in fieldsappropriate to the Symposium theme. This cantake up to 3 months; individual committeemembers take it upon themselves to personallycontact proposed speakers and makearrangements as necessary. A programme isprepared followed by marketing operations. Muchwork is carried out in the background in liaisonwith venue management and speakers, securingcopies of papers to be published as handouts onthe day. It does get a little hectic and can test thenerve - to quote Franklin D Roosevelt “happinesslies in the joy of achievement and the thrill ofcreative effort”.

Q: Your involvement with the ICT GolfSociety puts you onto the courseoccasionally. Is this simply a pleasant day out,or a task during which contacts can be madeand information gained?

A: I thought we weren’t going to mentionthis one! The Society meets twice a year, generallyimmediately prior to Convention and then againmid- to late- summer. I firmly believe busy peopleshould make time for fun; golf offers a way torelax and clear the mind with plenty of fresh air,exercise and self-discipline (it don’t mean a thingif you ain’t got that swing). It can be quite aneducation sharing quality time with a variety ofpeople from the various parts of our industry asboth Institute members and non-members alikejoin in. I’ve learnt much by spending three or fourhours with two or three people walking aroundsome fabulous golf courses – life is an educationto embrace regardless of its pursuit.

11

Q: Does your position on Council asrepresentative of the Associate Membersresult in many requests from Associates forinformation or assistance?

A: Oddly enough – no. With exception in part,I don’t believe Associates get involved in theirInstitute anywhere near enough. I see Associatesas highly valuable to the ICT, with much to giveand they represent the future of the Institute. Apersonal disappointment to me was the decisionnot to give them voting rights, which demoralisesAssociates much. I understand and accept thedemocratic view of our Corporate Members,however their number is in decline and I wonderif we have the balance right as “in the council ofmultitude is wisdom”. The Admissions andMembership Committee is tasked with reviewingentry requirement which I do hope is favourabletowards long-standing and practicing Associateswho, in my opinion, may never be afforded theopportunity to have an ACT course funded bytheir employer; a practice widely disposed of inour financially aware employment environment.

Q: Ian, what trends do you see in relationto concrete 2020?A: The sourcing of materials is becoming more

of an issue and, with pressure on ourenvironment, concrete will be expected tobecome carbon neutral. I believe these will besignificant drivers which will radically challengethe convention of concrete production in themedium to long term. It is inevitable thatsuppliers, specifiers and producers will developpartnerships in agreement to maximize secondaryaggregate use, together with multi-blend optionsof binder materials – combinations which willrequire the development of components that willafford such concrete to retain its excellentdurability, finish and aesthetic attributes thatconventional concrete can deliver today.

Q: What is your view on the merger withthe Concrete Society?

A: I believe that the future of the ICT lies inpartnership with the Concrete Society, who arewell established and respected within our industryboth here in the UK and internationally and,indeed, have significant membership, many ofwhom are ICT members. I see synergy betweenthe two organizations, with the ICT bringingprofessional affiliation and education and TCSoffering global marketing, information and servicethrough their excellent web presence, extensivelibrary resource and dedicated staff – a prosperousfuture is envisaged.

Q: And finally, Ian: for the youngermembers and associates, do you see a careerprofile in the area of concrete developmentand application? If so, would it have to bewith a large company such as you are with?

A: I have always considered myself to be quitefortunate to have the privilege my employersafford by way of encouragement, education andopportunity. That said, without enthusiasm,desire, patience and perseverance on my own parttowards concrete development, the former,arguably, may never have been presented. Beingpart of a large and forward thinking companysuch as Marshalls helps nurture innovation;however, I do believe that our younger membershave fabulous opportunities both now and in thefuture, regardless of organization size, so long asthat organization has a desire to flourishorganically. Concrete development is certainlyentering new territory which brings with itdemands for talented, enthusiastic upcomingconcrete technologists.

Ian, many thanks for this interestingconversation. It is apparent that you takemuch joy from, and interest in, the materialand technology we are all involved with, andI hope this continues with your involvementin the Institute’s affairs. On behalf of the ICTCouncil and Members, I would like to thankyou for your time and thoughts

12

13

The British cement makers have collaborated

on matters of mutual interest for many years. By

1917 they had considered how the sales of

Portland cement might be stimulated by the

provision of information and technical advice and

the Concrete Utilities Bureau (CUB) was formed

by a few of the largest cement companies to offer

technical advice to users, by means of leaflets,

brochures and articles in the press. The cement

makers also collaborated on certain industrial and

commercial matters through the Cement Makers’

Federation (CMF).

In 1924 the scope of the CUB was enlarged

and it combined with the British Portland Cement

Research Association (BPCRA) to form the British

Portland Cement Association (BPCA). This

continued to function until 1935 when the CMF

decided that sales of cement would benefit from

a more targeted propaganda department, funded

by a levy on cement deliveries that was not to

exceed four old pence per ton. This decision lead

the CMF members to form, in July 1935, the

Cement and Concrete Association, incorporating

the BPCA but being self-controlled. The original

offices, in Grosvenor Square in London, remained

their headquarters until the mid-1980s. The

organisation comprised technical advisory staff, a

library and a public relations unit which produced

exhibitions and publications and arranged lectures

and similar activities. There were several branch

offices staffed by engineers who could help to

solve problems locally.

The C&CA was a non-profit making

organisation to serve cement users by undertaking

research, developing new uses, improving

techniques, giving advice and encouraging the use

of good designs and methods. In the early years

the Association gave some financial support to

outside bodies undertaking research.

The first Director was Major R A B Smith and

amongst the early staff appointments was Philip

Gooding, who was there for over 40 years and

who was to have a great influence on promotional

and training policies. Stanley Boakes spent his

entire career with the Association and played a

major role in the early educational work and in the

presentation of practical advice to concrete users.

The Association was funded by the levy on

cement sales, which at the time totalled about 4.5

million tons per annum. There were about 20

British cement makers who comprised the entire

membership of the Association. Concrete

producers and users were given advice, usually

without charge, but were not directly involved

with the management of the Association and were

never able to become members. No public

funding was involved and, unlike other industry

bodies, the C&CA was completely independent of

government.

The technology of cement based materials has been developing since the firstconcrete mix was produced. Much of this technology was further improved withtime but much was forgotten (sometimes to be later ‘reinvented’). Somedevelopments have been accidental, such as the discovery of the benefits of airentrainment, some have been the result of foresight and endeavour, or commercialgain, whilst others have been born of necessity such as those for military andstructural reasons.

This series of articles - "Milestones in the history of concrete technology" - hasincluded diverse papers on advances in concrete technology for military and sportingconstruction and nuclear energy generation. Underlining such advances in trainingand education.

This article charts the history of training at the Cement and Concrete Association.Reference is made to other divisions of the C&CA in order to put training in itsproper context – as part of the Association’s overall strategy. We are indebted toDuncan Pomeroy, formerly a Director of C&CA, for permission to use his originalpublication ‘The C&CA and the British Cement Association 1935 – 1992.

MILESTONES IN THE HISTORY OF CONCRETE TECHNOLOGY

THE HISTORY OF TRAINING AT THE C&CA

Adapted from Duncan Pomeroy’s original paper by Graham Taylor

Introduction

14

During its first year, the Association published

21 brochures and 13 leaflets; by 1936 it

employed seventy staff, about one third of whom

were professionally qualified. This ethos of

stimulating cement sales by providing

technologically sound and practical guidance was

to dominate the operations of the C&CA

throughout its existence.

The war yearsDuring the first few years, the emphasis on the

work was little changed; there was considerable

lobbying for a higher proportion of concrete road

construction and for more farm buildings in

concrete. From the outset, C&CA worked with

government laboratories and the construction

industry to build cement sales through co-

operation and not through confidential studies

behind closed doors.

The threat of war affected the Association and

the use of concrete to construct air-raid shelters

and other defences received priority. At the

outbreak of war, the organisation placed itself at

the disposal of the Ministry of Home Security, and

throughout the war, the staff were mainly

engaged on technical work for civil defence. The

Association even went as far as to send sandwich-

board men out onto the streets to promote the

use of concrete in air-raid shelters.

During the war civil projects were not ignored.

There was work on the use of concrete railway

sleepers and the use of concrete for roads and

buildings on farms, to help farmers become more

efficient. Practical lectures on constructing

defences and repairing war-damaged buildings

were given.

At this time there was probably less than a

dozen ready-mixed concrete plants in the country

and most concrete was site-mixed, so there was a

need for training and education of site operatives.

During 1945 staff began to return to the

Association from the Ministry of Home Security,

with 72 in place by the end of the year – a

number little different from the pre-war peak.

The Meynell yearsSir Francis Meynell was appointed Director

General in 1946 and there followed a period of

rapid growth. He was a renowned expert on the

presentation of information, especially in print.

All the regional offices except Edinburgh had

closed during the war and this situation prevailed

for the next decade. In 1946 a small laboratory

was opened in Worcester Park and the internal

technical work of the Association started to grow

rapidly under a new Technical Director. Research

was, at first, primarily materials orientated. The

site at Wexham Springs was bought in 1946 and

the house used as a laboratory. Several outside

bodies co-operated with C&CA on research and

others were formed, such as the FIP (Féderation

Internationale de la Précontrainte).

The period up to 1950 was enormously

stimulating, with the engagement of a team of

young, highly qualified and enthusiastic staff, and

the great changes that were happening in the

concrete construction industry.

The flag-ship publication Concrete Quarterly

was launched in 1948 and the Magazine of

Concrete Research the following year.

New buildings were erected on the Wexham

Springs site from 1946 to 1960 to accommodate

the research and development activities. All the

buildings used different methods of concrete

construction in order to demonstrate the

possibilities of concrete to architects and others.

Fulmer GrangeDuring all of this period the Association was

rapidly expanding its training programme and

from 1950, the year in which Dr Ralph Andrew

joined the staff, was running residential courses in

temporary rented accommodation at Wexham

Place, close to Wexham Springs. In 1957 the first

purpose-built residential block was opened at

Wexham Springs and the building was named

Fulmer Grange’s predecessor - TheMeynell Building at Wexham Springs.

15

after the Director General, Sir Francis Meynell. In

that year there were 24 courses attended by 850

participants. This building remained the training

centre for ten years, until the opening of the

much larger facility at Fulmer Grange, which had

accommodation for 160 residential students.

When the Fulmer Grange complex was built

there was conflict with the local authority who

wanted to impose severe restrictions on the future

building programme of the Association. At one

time there was a threat of demolition of some of

the buildings and this affair was splashed across

the tabloid press. Fortunately, the then Director

General brought his diplomatic skills to bear and

local politicians were persuaded to reverse their

decision.

The existing house on the site became a social

centre for course participants, with a bar, snooker

room, library and sitting room, as well as offices

for the Director of Training and other members of

staff.

It is worth noting that the standard of

accommodation and catering were high – a

deliberate policy since participants on the courses

mainly remembered how well they were looked

after domestically; not that the lectures were

anything but good! Many of those who attended

courses fondly remember their welcome and the

care for their welfare. Early participants were

regularly introduced to the local hostelries, which

must have benefited significantly from the friendly

relationships developed with C&CA.

Colonel Buckmaster, who ran the British spy

service during the war, spent his childhood at

Fulmer Grange and was reputed to regularly run

home for tea when he was at Eton College, about

5 miles away. Later he returned to see what

C&CA had done to his old home and when

invited to go up to the attic, where the staff had

their offices, declined as he had never been to the

servants quarters before!

The golden yearsFurther buildings included the concrete

materials building, with the printing department

upstairs, workshops, stores and photographic

rooms in 1966 and the final one, the computer

block in 1970.

The Hon Leo Russell inherited a thriving

establishment when he took over as Director

General in 1958 and under his shrewd leadership

the Association continued to grow. The building

programme was completed, the regional offices

were re-opened in 1964. The headquarters

advisory staff was built up and research

flourished. Cement sales continued to grow,

reaching a peak of over 20 million tonnes per

annum by 1971. The cement industry was

A Fulmer Grange dormitory block and Fulmer Grange House.

16

changing; by 1958 there were 12 manufacturing

companies compared to the 20 pre-war. This

concentration was set to continue until there

were only three major cement producers in

Britain, who backed the C&CA when it merged

with the CMF in 1987 to form the British Cement

Association (BCA).

In materials research, hydration mechanisms

were seen as fundamental to understanding the

ways in which the properties of concrete could be

controlled to provide the characteristics required

by the user. The techniques established were to

prove invaluable when problems related to

concrete durability occurred. Mix design was

studied in depth, culminating in the joint

publication of Design of normal concrete mixes,

which demonstrated the close working and

mutual respect of the Association and

governmental organisations.

The Association campaigned to raise testing

standards and the testing of test machines,

including its comparative cube testing service.

An iso-thermal conduction calorimeter was

developed to study the effects of factors on the

heat output from cements. Creep and shrinkage

were studied from fundamental and practical

standpoints and the findings were used in British

and CEN Standards. The C&CA was always at the

forefront of development and research, including

high strength concrete and durability.

Many design rules were outlined or refined

following research in the structures division;

notably in bridge design and elements of

structural framed buildings, lightweight concrete,

fibre-reinforced concrete and crack widths.

The construction research section investigated

almost every aspect of concrete on the

construction site and even further – inspection

and maintenance and the re-use of concrete from

demolition. It lead on trench reinstatement, the

use of block paving , bridge corrosion from de-

icing salts, thermal insulation of houses and many

others.

TrainingOne of the original principal objectives of the

association was to inform and educate concrete

workers so that the best technology and practice

would be used to the ultimate benefit of the

cement manufacturing industry. This involved the

provision of lectures and demonstrations at every

level from the making and placing of concrete to

the latest knowledge, frequently based on the

Association’s research, on structural design,

construction practice or materials performance.

In the post-war years, the effort directed to

training grew and by the early 1950s courses of

several days duration were being held at Wexham

Place, together with practical demonstrations,

out-doors on site. The shell roof building

provided an area where these could be done

without the influence of bad weather. By 1954

the number of five-day courses had reached 17,

but there were many more shorter courses, and

lectures were being given throughout the country

by the technical and regional staff.

The first residential training centre, the Meynell

Building, opened in 1957. By 1961 there were 34

training courses a year, attended by over 1200

people, increasing to 40 courses during 1963.

This rapid rise in demand led to the acquisition of

the adjacent Fulmer Grange site and the

construction of the much larger residential

training facility that had 160 single bedrooms,

lecture and seminar rooms and a suite of

laboratories for demonstrations and hands-on

experience. By 1967 the number of courses run

per year reached 100 for the first time, ranging

from one to five days duration.

An important innovation occurred in 1968 –

the establishment of the Advanced Concrete

Technology course; initially of five-weeks duration

but later to last for six weeks. The examinations

were under the auspices of the City and Guilds of

London. (taken over by the Institute of Concrete

Technology in 1985) and successful candidates

C&CA Course brochure showing alecture block.

17

received a diploma. These diploma holders

founded the Association of Concrete

Technologists, which later became the Institute of

Concrete Technology. Participants on this course

were to come from many countries across the

world.

By 1979 over 50 000 participants had attended

C&CA courses and by 1985 this had jumped to

100 000, but by then there was a changing tide

in the industry and the demands for residential

courses began to decline. The Association, as

always, was ready to respond to change and they

had already introduced another innovation, the

correspondence course on ‘Concrete Technology

and Construction’. This proved highly successful

and in the first year 123 students took the course

and sat the City and Guilds examination. The

85% success rate considerably exceeded that for

the average candidate in the UK and overseas.

Sadly, the demand for residential courses

continued to fall and by 1987 a decision was

taken to close Fulmer Grange and to rely on

running training courses around the country.

During the next few years this policy seemed to

have been vindicated as there were more

participants attending the courses than in the

final years at Fulmer Grange. Many of the

training staff were retained in other divisions at

Wexham Springs and continued to provide the

main core of lecturers.

Training was a key undertaking by the BCA

during its early years but eventually it became

economically unviable and was handed over to

Thomas Telford Training in the mid-1990s.

The Advanced Concrete Technology course

continued at Imperial College in London for a few

years before moving to Nottingham University’s

campus. It is now a web-based course run from

the UK by Talent. Residential courses have been

held in Ireland and now run in South Africa and

Australia.

Some courses are still run by Thomas Telford

Training and by one or two individuals, as in-

house events.

Concrete Practice examinations are no longer

available from City and Guilds. Concrete

Technology and Construction (CT&C) distance

learning courses have suffered from lack of

funding but are being re-presented by a

combination of the National Construction College

at Bircham Newton and Isolearn in a more user-

friendly format. In recent times the Institute has

revised the range of syllabuses into four levels, to

reflect industry’s current training needs. Level 1

replaces Concrete Practice and occasional courses

are run for the examination provided by the

Institute, who are keen to ensure the continued

existence of all examination levels. Levels 2 and 3

are intended to replace CT&C – General Principles

and Practical Applications respectively and the

ACT will become Level 4.

As to the future, the merger of ICT with The

Concrete Society opens the door to a possible

expansion of concrete-related training activities.

ConclusionThose who took courses at Fulmer Grange

tend to remember their time there with nostalgia;

the food, the camaraderie and the lectures. Most

of the staff who were there at its closure in 1987

have now retired and it is difficult to envisage

how the valuable results it achieved will be

continued. The glimmer of hope is the occasional

reminiscence from a past student and a further

hope that he will have had an influence on those

following him.

18

ANNUAL CONVENTION SYMPOSIUM: PAPERS PRESENTED 2007

PAPERS: AUTHORS:

A major part of the ICT Annual Convention is the Technical Symposium, where guestspeakers who are eminent in their field present papers on their specialist subjects. Each year papers are linked by a theme. The title of the 2007 Symposium was:

NEW CONCRETE TECHNOLOGY IN CONSTRUCTION Symposium Chairman: Professor Peter Hewlett PhD, LLD, BSc, CChem, FRSC, FIM, HonFICTJohn Doyle Group plc / British Board of Agrément (BBA)

Edited versions of the papers are given in the following pages. Some papers vary inwritten style notwithstanding limited editing.

CHAIRMAN’S INTRODUCTION Professor Peter Hewlett PhD, LLD, BSc, CChem, FRSC, FIM, HonFICTJohn Doyle Group plc / BBA

KEYNOTE ADDRESS Professor Ravindra Dhir OBESUSTAINABLE CONSTRUCTION BSc, PhD, CEng, MIM, HonFICT, FGS, FICEBREAKING DOWN THE CITADELS: University of DundeeCHALLENGES OF FIT-FOR-PURPOSE CONCRETE

LOW CO2 CEMENTS BASED ON Dr. Keith QuillinSULPHOALUMINATE CEMENTS BSc, PhD, MBA

Building Research Establishmentand Dr. Ellis Gartner,BA, MA, PhD, FIMMM, FACerSLafarge Centre de Recherche

UNDER THE COVER: Dr. Phil PurnellNOVEL NDE TECHNIQUES FOR CONCRETE BEng, PhD

University of Warwick

SYNTHETIC MACRO FIBRES Dr. Klaus - Alexander RiederScDGrace Construction Products Ltd

CONCRETE UNDER THE PETROGRAPHIC Dr. Alan PooleMICROSCOPE BSc, DPhil, CGeol, EurGeol, FGS, MAE

Consultant in Geomaterials, Petrology & Engineering Geology

AUTOMATED MONITORING OF THE Professor Nick BuenfieldCONDITION OF CONCRETE STRUCTURES PhD, MSc, BSc(Hons), DIC, CEng, MICE, MICT

Imperial College London

APPLYING FUTURE INDUSTRIALISED Professor Simon AustinPROCESSES TO CONSTRUCTION BSc, PhD, CEng

Loughborough University

A FLEXIBLE CONCRETE ARCH FOR Professor Adrian LongSUSTAINABLE BRIDGES PhD, DSc, FEng, FIAE, FACI, FICE, FIStructE, FIEI, FICT

Queen’s University, Belfast

PRECAST CONCRETE PRODUCTS MADE WITH Dr. Marios SoutsosRECYCLED DEMOLITION AGGREGATES BEng(Hons), PhD, MICT

Liverpool University

AN INNOVATIVE SOLUTION FOR Dr Jean-Philippe ThierryCONCRETE FLOORING Lafarge Centre de Recherche

LIMESTONE FILLER BASED SELF COMPACTING Professor Geert de SchutterCONCRETE: FROM MICROSTRUCTURE TO MSc, PhDENGINEERING PROPERTIES Ghent University

CHAIRMAN’S SUMMARY Professor Peter Hewlett PhD, LLD, BSc, CChem, FRSC, FIM, HonFICTJohn Doyle Group plc / BBA

19

20

It is my privilege to chair your

symposium today. We are here

to listen and learn. This, the

35th ICT Annual Symposium is

something of a landmark

technically.

I say that because there is much new concrete

technology that has been developed in the last

10 - 20 years and much being developed as we

meet. It is right that we should take stock and

adopt change where there is benefit and new

opportunity.

We have 11 presentations involving some 24

authors and co-authors within which there are at

least 8 Professors and 7 PhDs so we should not

be short on erudition. Within that mix we are

well endowed with engineers, and a sprinkling of

chemists, physicists, material scientists and

perhaps an engineering geologist.

The spread of subjects is wide and diverse,

demonstrating the adaptability of concrete on a

global scale.

We are presented with a technical feast today

so let us start the banquet with the keynote

address to be given by Professor Ravindra Dhir.

The subject is ‘Breaking down the Citadels:

Challenges of Fit-for-Purpose Concrete’.

21

CHAIRMAN’S INTRODUCTION

Professor Peter Hewlett PhD, LLD, BSc, CChem, FRSC, FIM, HonFICT

John Doyle Group plc / BBA

22

Ravindra Dhir is a professor at

the University of Dundee and

founding director of the

Concrete Technology Unit at

Dundee.

INTRODUCTIONConcrete has been the world’s premier

construction material since its first modern-day

use in the late 19th century, however, the past 30

years have seen a growing awareness amongst

engineers of the need to ensure durability in

concrete structures. This has arisen largely as a

result of the increasing incidence of premature

deterioration over the period and the substantial

repair and maintenance requirements that have

resulted[1]. While the specific problems

encountered have varied around the world, this

has very much become an issue of global

concern[2]. There has also been a growing

awareness of the importance of sustainability in

concrete construction and in particular of the

need for more effective and efficient use of virgin

and, as a result, recycled materials [3] and,

measures including landfill taxes and aggregate

levies have been introduced, at government level,

in order to promote and encourage these

practices[4].

Concrete is constantly evolving, and practices

and how the material is specified must also

evolve to ensure that it remains the sustainable

and competitive and durable solution to global

infrastructure needs. This paper examines the

current approach to concrete specification and

whether the citadels of concrete technology are

still valid for today’s concretes.

WHAT ARE THE CITADELS OF CONCRETE TECHNOLOGY?

Concrete specification for durability is primarily

governed by three limiting factors which have

been the stronghold of concrete performance for

many years:

• Water/cement ratio,

• Compressive strength,

• Cement content.

Since the patenting of Portland cement by

Joseph Aspdin in 1824, the relationship between

water, air and cement content have been

thoroughly researched. In 1896, Rene Féret

discovered a basic relationship between the ratios

of water and air plus cement to the strength of

concrete. Essentially this was a forerunner to

Abrams Law (1919) which is the governing

inverse relationship between water/cement ratio

and compressive strength still relevant today.

Developments in StandardsWater/cement ratio and compressive strength

have been present in the development of

standards from as far back as 1921[5]. The

introduction of somewhat vague guidance on the

three citadels appeared in CP114 during the

1940s and 50s, citing such terms as “1:2:4 mix

for a very good concrete” and that the water

content should ensure that concrete is

“sufficiently wet”. The standard did, however,

state that it was important to maintain the

water/cement ratio. As standards developed

through the 1960s (CP116) and 1970s (CP110),

the first links between the citadels and durability

were seen. Initially durability was linked to

strength and water/cement ratio; however,

minimum cement content was also seen as being

important. Exposure classes were also introduced

and linked directly to limiting factors. Further

developments in standards for concrete

specification in the 1980s and 1990s (BS8110

and BS5328)[6, 7] lead to the concept of trade-off

between concrete quality (defined by the limiting

factors) and cover depth to steel reinforcement.

However, changes are now taking place across

Europe with the introduction of EN 206-1[8] and

BS8500[9]. In recognition of how concrete has

changed over the past century, these standards

are now moving towards performance-based

specifications with specific degradation

mechanisms assigned to exposure classes.

23

KEYNOTE ADDRESS

SUSTAINABLE CONSTRUCTION BREAKING DOWN THE CITADELS:

CHALLENGES OF FIT-FOR-PURPOSE CONCRETE

Professor Ravindra Dhir OBE,

BSc, PhD, CEng, MIM, HonFICT, FGS, FICE

University of Dundee

24

However, the citadels remain with limiting factors

based on maximum water/cement ratio, minimum

compressive strength and minimum cement

content. The links between the citadels,

exposure conditions and cement type in

standards over the years are shown in Table1[10].

ARE THESE SUITABLE FOR 21st CENTURY CONCRETE?

The limiting factors of maximum water/cement

ratio, minimum compressive strength and

minimum cement content have served concrete

well for over 150 years with buildings such as

Weavers Mill, Swansea (Europe’s first reinforced

concrete structure) surviving for nearly 90 years.

However, there have been a number of instances

of failure through poor specification or

workmanship (eg the Montrose Bridge, built in

1930 and replaced in 2004 due to alkali silica

reaction, and the Tay Road Bridge, built in 1965

and undergoing continuous repairs since 1971).

The limiting factors have worked because they

are easy to specify and measure, can be

determined through the use of well established

and robust testing techniques and can provide

verifiable proof of what is in the concrete.

However, concrete has developed as a

material. Changes have taken place in the types

of constituent materials being used with additions

such as fly ash, ground granulated blastfurnace

slag, metakaolin and silica fume now being

specified in concrete. Recycled aggregates are

also being used as well as admixtures which

significantly modify fresh and hardened concrete

properties. There are now also increasing

demands on the engineering performance of

concrete, including self-compacting concrete,

pumpable and fibre-reinforced concrete as well as

requirements to adapt to changes in

functionalisation such as fire and impact

resistance, offshore environments, seismic loading

and use in nuclear applications.

Moreover, sustainability is now driving many

aspects of construction and concrete must assess

its carbon footprint by addressing important

issues such as embodied energy, use of natural

resources and absorbing recycled materials, whilst

at the same time exploiting some of its

environmentally friendly aspects such as thermal

mass.

MOVEMENT TOWARDS FITNESSFOR PURPOSE CONCRETE

As concrete evolves the question is raised as to

whether the limiting factor approach is suitable

or should we be moving toward ‘fit for purpose’

concrete? In order assess the suitability of the

citadels it is vital to understand the

interdependency of design and performance

when considering fitness for purpose. Design can

be defined as the conception of an idea to ensure

that a structure can withstand the demands

placed on it. Performance, on the other hand, is

a measure of the competency of the structure in

withstanding the demands and meeting the

expected level of service.

Defining DemandsTo determine whether we can achieve fitness

for purpose it is important to define clearly the

demands on concrete which must be met.

Table 1: Developments in standards for specifying concrete durability.

� � � � � � � ��

Demands can come from a variety of areas, the

most obvious being structural integrity,

serviceability and durability. However, economics

will also play an important role and, as we move

towards a more sustainable society, embodied

energy, social impact and carbon footprint will

also be defining demands. It is also vital that the

demands are defined so that concrete can be

assessed for performance to see whether the

demands are being met.

Assessing PerformanceThe level of performance which is deemed to

be successful must also be defined clearly before

performance can be assessed. In terms of

concrete construction, there are a variety of levels

of performance which must be met and many of

these will depend on the demands which have

been defined. In current specification standards

(BS8500), performance levels are defined through

the length of time for the minimum intended

working life (eg 50 years, 100 years). However,

as sustainability is a key driver, performance levels

in the future may also be defined by such aspects

as the total carbon footprint of a concrete

structure, the embodied energy within the

structure or the whole life cost.

Once levels have been defined, we must be

able to measure performance in order to

determine whether they are being met. This is

currently where the limiting factor approach is

used; however, the use of other performance

indicators such as performance models,

sustainability indicators and life-cycle analysis

models may serve as better performance

measurement tools.

DO THE CITADELS ENSURE FITNESS FOR PURPOSE?

The increasing demands being placed on

concrete mean that ensuring the material meets

structural, durability, sustainability and economic

requirements is now paramount to the

construction industry. Although concrete as a

material has evolved significantly, the limiting

factors do little to encourage ‘fit for purpose’

concrete being achieved. Limiting the

water/cement ratio, compressive strength and

cement content may lead to prevention of the

use of sustainable materials such as recycled and

secondary aggregates. Figure 1 shows the

influence of recycled concrete aggregate (RCA)[11]

content on compressive strength; however,

although the current standards do not prevent

the use of recycled materials, the limiting factors

do not allow specifiers the freedom to use

combinations of recycled materials to achieve a

specific performance level.

Placing strict limitations on factors such as

strength, water/cement ratio and cement content

can lead to inhibitions on innovation. Novel

materials such as foamed concrete can often be

specified using factors outwith those in the

standards; however, the specifications limit their

use. Figure 2 (Page 26) examines the effect of

increasing fly ash content on concrete durability;

it shows that, as with many durability aspects,

increasing performance can be gained

independently of the water/cement ratio and/or

strength.

In addition, limitations placed on minimum

cement content can often inhibit increased

durability resistance as shown in Figure 3 (Page

26). This show that reducing the cement content

and optimising the fines content to obtain a

closed structure can lead to lower absorption,

lower permeability and a reduction in chloride

diffusion [12]. Although the limiting factors which

are currently placed on concrete do serve to

maintain a certain level of performance, they may

actually be preventing the development of more

sustainable, economic material.

CONCLUSIONConcrete as a material has changed

significantly over the past 150 years and

standards and specifications have also developed

to accommodate them. However, the industrial

climate is changing rapidly and a movement

towards ‘fit for purpose’ specifications is taking

place. Concrete now must respond to the

demands of a changing world such as (i) climate

change, (ii) carbon footprint issues, (iii) economic

25

Figure 1: Influence of recycled concreteaggregate content on compressivestrength and durability.

26

Figure 2: Influence of fly ash content on durability performance.

Figure 3: Effect on concrete durability of reducing cement content.

PFA Content, %* Assume Part 1 PFA price = 50% of PC price

D is the Chloride ion diffusion coefficient

pressures, (iv) client pressures, (v) increasing

demands from building insurance, and (vi) higher

performance requirements. The question remains

are we moving towards specifying fit for purpose

concrete or are we responding to these changing

demands?

Throughout Europe, there is a drive towards

probabilistic approaches to design in Eurocodes

and in particular performance based

specifications with the publication of the new fib

Model Code for Service Life Design[13]. This

initially focuses on chloride and carbonation-

induced corrosion as well as freeze-thaw attack;

however, can we be completely confident that we

can reliably populate the models for performance

specifications? We must now assess the test

methods in terms of measuring performance,

repeatability and reproduceability, sensitivity to

environmental conditions and cost, so that we

can standardise tools which will give us reliable

data to populate the models.

The current citadels of compressive strength,

water/cement ratio and cement content are in

danger of hindering the progress of concrete as

the world’s premier construction material.

Seismic shifts towards a performance-based

philosophy are now underway throughout

Europe; however, a focussed, coordinated

approach is required to ensure that concrete

emerges as the sustainable solution for new

construction projects.

New visions are required from the concrete

construction industry to embrace and drive the

changes needed in current specifications and

meaningful collaborative research and innovation

are paramount to success. Partnerships in

research across all communities on a large scale

are vital to ensure we move towards the creation

of a ‘fit for purpose’ approach to concrete

construction.

REFERENCES

1. The Concrete Society. Developments indurability design and performance-basedspecification of concrete. Concrete SocietySpecial Publication. CS 109, 1996, 69 pp.

2. SHARP, B.N. and SLATER, D. Design guidesand standards. Concrete in CoastalStructures. (Ed T.R. Allen), Thomas Telford,London, 1998.

3. Department of the Environment. MPG6Mineral planning guide for aggregatesprovision in England, HMSO, London, 1994.

4. UK Government. A better quality of life: astrategy for sustainable development in theUnited Kingdom. The Stationary Office,London, 1997.

5. Everyday Use of Portland Cement,Associated Portland Cement Manufacturers,1921

6. British Standards Institution. BS 8110: Part1. Structural use of concrete: Code ofpractice for design and construction.London, 1985 & 1997.

7. British Standards Institution. BS 5328: Part1. Concrete: Guide to specifying concrete.London, 1991 & 1997.

8. British Standard Institution. BS EN 206-1:Concrete – Part 1: Specification,performance, production and conformity.London, 2000.

9. British Standard Institution. BS 8500:Concrete – Complementary BritishStandard to BS EN 206-1 – Part 2.Specification for constituent materials andconcrete. London, 2002.

10. The Concrete Society. The influence ofcement content on the performance ofconcrete. Crowthorne, 2000, 48 pp.Report No CS125.

11. DHIR, R.K., PAINE, K., DYER, T.D. andTANG, M.C. Value-added recycling ofdomestic industrial and construction arisingsas concrete aggregate, ConcreteEngineering International, March 2004, Vol.8, No. 1, pp 43-48.

12. DHIR, R.K., MCCARTHY, M.J., TITTLE, P.A.J.and ZHOU, S. Role of cement content inspecifications for durability: cement contentinfluences. Proceedings of the Institution ofCivil Engineers., Structures and Buildings (Inpress), 2004.

13. FIB, “Model Code for Service Life Design ofConcrete Structures”, pp180, 2006.

27

28

Keith Quillin is a Principal

Consultant in the Centre for

Concrete Construction and has

worked at BRE since 1990. He

has particular expertise in

cement chemistry, concrete

durability and service life design and prediction of

concrete structures. Other work has included the

development of an environmental code of

practice for the use of pfa as a filling material for

stabilising disused mine workings. Keith has

worked on the long-term properties of

cementitious materials relevant to the

containment of radioactive and toxic waste and

has carried out research and modelling studies to

determine phase equilibria in the CaO-Al2O3-SiO2-

H2O system. Keith has an Honours degree in

Chemistry and a PhD in Physical Chemistry. He

also has an MBA from Henley Management

college.

Dr Ellis Gartner is Principal

Scientist at Lafarge Central

Research, near Lyon, France.

He obtained a PhD in Gas

Phase Reaction Kinetics from

the University of Cambridge in

1974 but his interest in

Construction Materials began at the BRE where

he worked for the three following years, (mostly

under the very tolerant supervision of Dr. Philip

Nixon!) In 1977 he joined the Portland Cement

Association (in Skokie, Illinois) as a Research

Chemist, later becoming Manager of Basic

Research. In 1985 he moved to the Washington

DC area to work in the Central Research

Laboratory of W. R. Grace and Co., where he led

a small research group that created much of

Grace’s new admixture technology (CBA, ADVA,

Eclipse, etc.) In 1996 he moved again, this time

to the Lafarge Central Research Laboratory, to

manage research in cement-admixture

interactions, a critical aspect of the early

development phase of Agilia self-placing

concretes. He is a Fellow of the American

Ceramic Society and the UK Institute of Materials,

and has over 50 scientific publications and 28

patents. His current main interests are in cement

hydration mechanisms and sustainable cement

and concrete technologies.

ABSTRACTThe production of clinkers rich in calcium

sulfoaluminate (CSA) requires less limestone and

less energy than the production of conventional

Portland cement clinkers. Consequently, CO2

emissions per unit of clinker are reduced.

Experimental results also show that CSA-based

cements can be formulated to produce durable

concretes with physical properties comparable to

those of common classes of Portland cement

concrete. Overall CO2 emissions with such

cements can be significantly lower than with

conventional cements based on Portland clinker.

Two main avenues of research were followed

with a view to balancing environmental impact,

manufacturing costs and physical performance:

1. Blends of calcium-sulfoaluminate-rich

cements with non-clinker materials such

as ground granulated blastfurnace slag

(ggbs) and calcium sulfate. Certain

blended cements of this type were found

to give concretes with good physical

properties. They were demonstrated to be

suitable for the manufacture of precast

concrete blocks,

2. Cements based on novel clinkers

containing activated belite, calcium

sulfoaluminate and calcium aluminoferrite.

Such clinkers can potentially be

manufactured from a wide range of raw

materials, emitting 20-25% less CO2 than

Portland cement clinker. Concrete

compressive strength development to 90

days was shown to be comparable with an

otherwise equivalent concrete made using

a conventional Portland cement (PC)

(42.5R).

However, developing new cements like these

for large scale use is an inherently slow process,

and a great deal more work is still required to:

• Confirm that such cements can be

manufactured industrially, with the desired

29

LOW-CO2 CEMENTS BASED ON CALCIUM SULFOALUMINATES

Dr Keith Quillin, BSc, PhD, MBA

Building Research Establishment and

Ellis Gartner, BA, MA, PhD, FIMMM, FACerS

Lafarge Centre de Recherche

30

CO2 emissions savings and low

environmental impact, and at an

acceptable cost

• Better establish the effects of

compositional and processing variables on

performance

• Optimise cements for performance in

various major use categories, in terms of

physical properties, raw materials

availability, manufacturing parameters, etc

• Develop the scientific understanding

necessary to explain the observed

performance, and thus to help predict the

effect of variations in conditions on such

performance

• Clearly establish the long term

performance of these materials in concrete

and provide data that will give users and

specifiers confidence in their durability

• Develop appropriate codes and standards

for their use in construction.

KEYWORDSCarbon dioxide, Calcium sulfoaluminate,

Belite, Durability, Compressive strength.

INTRODUCTIONGlobal cement manufacture (including both

‘pure’ and ‘blended’ Portland cements) has risen

from 594 million tonnes per annum in 1970 to

2284 million tonnes per annum in 2005 with

virtually all of this growth occurring in developing

countries. Global production is likely to increase

significantly over the coming decades as the

global population increases. Recent forecasts

have suggested that global cement production

could exceed 5 billion tonnes per annum by

2050.

Cement manufacture is energy intensive. The

raw materials must be finely ground and

homogenised, and then heated to about 1450ºC

to form Portland cement clinker. Upon cooling,

the hard nodules of clinker must then be finely

ground, with small amounts of other ingredients

such as gypsum, to make the finished cement .

Energy consumption as fuel and electricity

consequently represents about 65-70% of the

variable costs associated with Portland cement

manufacture [1],[2].

Cement manufacture also produces large

amounts of CO2 due directly to the calcination of

CaCO3 (about 1.2 tonnes of limestone are

required to produce one tonne of a typical

modern Portland cement clinker). The amount of

CO2 emitted per unit of Portland cement clinker

manufactured depends on a number of factors

such as clinker type, fuel and raw materials

compositions and the energy efficiency of the

specific kiln system. The amount of CO2

produced per unit of finished cement varies even

more, depending on cement composition (e.g.

content of clinker and non-clinker ingredients),

manufacturing efficiency, etc. For example, the

US Geological Survey quotes[3] a figure for the

USA of about 0.94 tonnes per tonne of Portland

cement clinker. Recent data[4] on global CO2

emissions from cement manufacture (including

the use of non-clinker ingredients) give an overall

average of 0.88 tonnes of CO2 per tonne of

cement. The same data indicate a value for

Europe of about 0.63 tonnes per tonne. These

data imply that about 2 Gt of CO2 per year are

currently emitted directly to the atmosphere due

to cement manufacture.

The pressure to reduce energy consumption

and CO2 emissions during cement manufacture

has led the industry to increase the extent to

which Portland clinker is substituted in

conventional cements by other ingredients that

are currently approved in the existing norms, such

as granulated blastfurnace slag (ggbs), pulverized

fuel ash (pfa), natural pozzolans and limestone.

However, until recently, there have been few

serious attempts to develop novel cements based

on alternative clinkers with intrinsically lower

manufacturing CO2 emissions than conventional

Portland cement clinkers.

A number of alternative cements have been

proposed to deal with perceived CO2 and energy

problems[5],[6],[7]. However, as well as

demonstrating appropriate physical properties

and durability when used in concrete, any

potential replacement cement must be based on

minerals available in quantities comparable to the

demand, and well-distributed world-wide.

Cements based on significantly less abundant

minerals are unlikely to be produced in quantities

large enough to have much impact on global CO2

emissions.

Some of the most promising alternative

cementing systems for general concrete

applications at ambient temperatures currently

appear to be those based at least in part on

calcium sulfates, the availability of which is

increasing due to the widespread implementation

of sulfur dioxide emission controls. These include

calcium sulfoaluminate (CSA) and belite-calcium

sulfoaluminate-ferrite (BCSAF) cements and

31

similar systems that make good use of the

potential synergies between hydrated calcium

silicates, sulfates and aluminates. Some such

cements have been normalised in China under

the generic name “The Third Cement Series”.

In this paper we report some encouraging

preliminary results obtained on two novel

cementing systems of this general type:

• CSA-rich clinkers blended with ggbs and

other non-clinker ingredients

• BCSAF clinkers which can be made in

conventional PC kilns.

BELITE-RICH CEMENTSBelite-rich cements can be made by burning

mixtures of limestone and clay in a similar process

to that used for PC manufacture. The theoretical

energy requirements and process CO2 emissions

for the manufacture of belite-rich cements are

lower than that for PC (the main hydraulic

component of which is alite, an impure form of

C3S).

The formation of a high percentage of alite in

clinker requires that a finely-ground and

homogenised mixture of raw materials be heated

to a minimum temperature of about 1400ºC in

the rotary kiln. However, belite, C2S, is formed

rapidly at above 1200ºC if the raw materials are

sufficiently finely ground and well-mixed, so

cements based mainly on belite can generally be

manufactured at significantly lower kiln

temperatures. Belite-based cements have a

further advantage over PC in that the amount of

CO2 produced by decarbonation is reduced. This

is because the CaO content of belite (65.1% by

weight) is lower than that of alite (73.7% by

weight). In fact, it is this lower CaO content that

really reduces the thermal energy required to

make belite, relative to alite. This is because the

decarbonation of limestone is highly

endothermic, and actually accounts for the

majority of the fuel consumption in a modern

energy-efficient cement kiln system. Differences

in maximum burning-zone temperature have only

an indirect effect on kiln thermal efficiency.

Belite-rich cements can produce concretes with

very good long term properties. However,

strength development is generally very slow in

comparison with alite-rich Portland cements due

to the inherently slow hydration of the usual

belite phase (C2S). In order to produce cements

suitable for most modern applications it is

therefore necessary to either produce a more

reactive form of belite or to add a more reactive

component to contribute to the early age

strength development.

The presence of Fe2O3 in the raw meal can

lead to the formation of ferrite (C4AF) and to the

partial substitution of Al2O3 in other alumina-rich

phases such as calcium sulfoaluminate. The

inclusion of higher levels of Fe2O3 increases the

range of raw materials that can be used in

manufacture and so has the potential to reduce

the cost. Fe2O3 also usually acts as a flux in the

clinkering process, which can help reduce the kiln

residence time.

The compositions of belite-based cements can

be altered through the inclusion of reactive

phases, such as calcium sulfoaluminate, C4A3S

(mineral name: yeelimite), that increase the early

strength. The reactive phase C4A3S (in some

cases together with CS and/or C4AF) hydrates to

produce ettringite as the main hydrate phase

responsible for the early strength. The belite

component hydrates more slowly and is

responsible for the long-term strength.

C4A3S can be formed in a cement kiln at

temperatures between about 1000ºC and

1300ºC. The production of 1 tonne of C4A3S

generates only 216 kg of CO2 from limestone

decarbonation if one assumes that its sulfate

originates from calcium sulfate in the raw mix. (If

this sulfate originates from the combustion of

sulfur in fuel and its reaction with lime, then this

figure increases slightly, to 288kg of CO2 per

tonne of C4A3S; but it is still a low value.)

A simple calculation for a belite-sulfoaluminate

cement (assumed to contain 38% belite and 35%

C4A3S together with calcite calcium aluminoferrite

and calcium sulphate) suggests that limestone

decarbonation would only produce about 300 kg

of CO2 per tonne of cement. However, CO2

emissions arising from energy consumption must

be added to these figures. Whilst it is not

possible to accurately determine CO2 savings

arising from any reductions in energy use prior to

carrying out manufacture on a large scale it is

reasonable to assume that, within limits, energy

use is proportional to CaCO3 content of the raw

feed. Consequently, if such clinkers can be

made in modern energy efficient (preheater) kiln

systems, overall CO2 emissions can be assumed to

be reduced roughly in proportion to the reduction

in process CO2 emissions.

EXPERIMENTALAn experimental programme was carried out

to assess the performance of concretes made

-

-

-

-

-

-

-

32

using various CSA-based cements. The mix

design used in each case was as follows:

Total binder content: 300 kg/m3

Free water: binder ratio: 0.55

Total aggregate: cement ratio: 6.1:1

20-10 mm Thames Valley 35%



No admixtures were used. Concretes were

subjected to a standard curing regime involving

24 hours curing in moulds under damped hessian

prior to storage in the following environments:

• Water at 20ºC

• Water at 5ºC

• Water at 38ºC

• Air (20ºC, 65% RH)

• External sheltered

• External unsheltered

• 4% CO2, 65% RH, 20ºC (accelerated

carbonation)

Compressive strength determinations were

carried out on 100 mm cubes in accordance with

BS1881: Part 116[8]. The hydrated phases present

in the samples were identified by X-ray

diffractometry (XRD), with a Siemens D500

diffractometer using Cu Ka radiation and

operating at 40 KV and 30 mA. Data were

accumulated over one scan of 2ø between 5º and

50º. Assignments of lines were made by

comparisons with JCPDS files. Conduction

calorimetry was carried out using a Wexham

isothermal conduction calorimeter.

Carbonation depth was measured on 75 x 75

x 200 mm prisms made using the same mix

designs as used for compressive strength

development. Specimens for accelerated

carbonation were stored in water to 28 days

followed by 28 days conditioning at 20ºC and

65% RH prior to testing (carbonation occurred

over this period, as shown by the carbonation

depth at 0 days). The samples were then

exposed to a carbon dioxide enriched atmosphere

with a CO2 level of 4% at 20ºC and 65% RH.

Carbonation depths were measured at 2-weekly

intervals as described above. Specimens for

natural carbonation were cured under damp

Hessian and polythene to 1 day prior to placing in

the following exposure conditions:

a. Indoors at 65% RH and 20ºC

b. Outside sheltered

c. Outside unsheltered

Carbonation depths were determined by

spraying freshly fractured surfaces of concrete

specimens with a phenolphthalein indicator

solution and measuring the depth of the coloured

region at 5 points along each edge of the

specimen before averaging.

CSA-GGBS BLENDSThe properties of concretes made using a

range of CSA cements were studied. The

compositions of some of the cements used are

shown in Table 1. Theoretical CO2 savings

relative to pure PC concretes depend on the

composition but can be over 70% for the

compositions studied.

XRD analyses showed that the cements were

composed of various proportions of yeelimite,

belite, anhydrite and gypsum with small amounts

of calcite and possibly tricalcium aluminate.

Studies have included compressive strength

development and durability to 2 years. The CSA

cements were used neat and in blends with other

components such as ggbs, pfa and anhydrite.

Conduction calorimetry outputs on cements A to

D and B2 are shown in Figure 1.

Early age compressive strength data for

concrete made using neat CSA cements in water

at 20ºC are shown in Figure 2. 90-day strengths

are comparable with otherwise equivalent

concrete made using a typical

British PC (type CEM I 42.5R).

Strength development in CSA

cements blended with ggbs and

anhydrite are shown in Figure

3. From these results it is clear

that CSA-rich clinker on its own

is not a good activator for ggbs.

However, combinations of CSA

with calcium sulfate show good

strength development

Oxide content (wt%)

Oxide Cem A Cem B Cem C Cem D B2

SiO2 8.35 4.86 8.19 5.71 9.9

Al2O3 25.6 34.2 24.5 35.9 33.2

Fe2O3 2.84 2.46 2.70 2.47 1.0

CaO 42.0 40.7 42.4 40.9 47.0

SO3 13.8 12.0 13.1 9.30 7.9

Table 1: Compositions of CSA cements.

33

Figure 1: Conduction calorimetry outputs for cements A to D, and for B2 (in eachcase the cell contained 30 g cement mixed with 18.6 g water).

Figure 2: Compressive strength development in concretes made from “pure” CSA-based cements, compared to “pure” PC (cement content = 300 kg/m3; w/c = 0.55).

properties when combined with ggbs (although

the composition of the CSA is important in

determining this). The effects of curing regime on

compressive strength at 180 days are shown in

Figure 4.

Concrete tests of this type of blend show no

significant expansion in water over 9 months and

a rate of carbonation higher than PC but not

extreme[9]. Longer term tests are ongoing.

Precast manufacturing trials have also proved to

be promising.

Although certain CSA/anhydrite/slag blends

give promising results, this raises the issue of the

current limited global supplies of high-CSA

clinkers and of good quality blastfurnace slags.

LAFARGE’S NOVEL BELITE-CSA-FERRITE (BCSAF) CEMENTS

Lafarge has recently developed a novel type of

cement based on an activated belite/CSA/ ferrite

clinker interground with anhydrite. It can be

manufactured from a wide range of raw materials

in conventional Portland cement kilns. With non-

activated clinker, rapid setting and hardening due

to rapid consumption of C4A3 and anhydrite and

formation of ettringite usually occur, with belite

hydration being delayed. However, certain

combinations of minor elements allow significant

activation of the belite (by formation of the γ

phase), while C4A3 hydration is retarded slightly,

giving better set and workability control. The

34

Figure 5: Compressive strength development in equivalent concretes made usingLafarge’s BCSAF cement (B3) and PC (cement content = 300 kg/m3; w/c = 0.55).

Figure 4: Compressive strengths in different environments at 180 days.

Figure 3: Compressive strength development in concretes made fromCSA/ggbs/anhydrite blends, compared to a PC/slag blend (cement content = 300kg/m3; w/c = 0.55).

35

ferrite phase also appears to contribute to

strength development. With these modified

clinkers, belite hydration is significant even at one

day. Manufacture of such clinkers is predicted to

give 20-25% less CO2 emissions than Portland

cement clinker, and similar levels of CO2 savings

are expected for cements made from such

clinkers. The results shown here were obtained

with a simple BCSAF cement (B3) containing only

clinker plus 10% calcium sulfate, but the use of

these cements in more complex blends with

supplementary cementitious materials such as

ggbs is currently being investigated by Lafarge

and will certainly lead to even greater CO2

savings.

Compressive strengthdevelopment

Strength development for B3 in moist-cured

concrete is shown in Figure 5 and is very similar

to the otherwise equivalent PC concrete control.

90-day compressive strengths on storage in water

at 38ºC and at 5ºC were also similar to those of

the PC controls. Longer term strength

development and durability studies are underway.

Dimensional stability75 x 75 x 200 mm concrete prisms were

prepared (with steel inserts cast in to each end to

facilitate dimensional measurements) using the

mix design outlined above. The prisms were

assessed for dimensional stability under the

following conditions:

Figure 6: Dimensional stability of B3 concrete.

Figure 7: Dimensional stability of PC concrete.

36

1. Storage in water at 20ºC

2. Storage in air at 20ºC and 65% RH (after

28 days water curing at 20ºC)

3. Storage in water at 20ºC following heat

curing at 90ºC for 4 hours

The results are shown in Figure 6. Results for

PC concrete stored under similar conditions are

shown in Figure 7.

CarbonationB3 concrete carbonated more rapidly than the

PC control under accelerated carbonation

conditions. Carbonation depths of 6 mm and 3

mm were measured for B3 and PC concretes

respectively after 91 days. However, there was

little difference in the rate of natural carbonation

in dry and external sheltered environments after

91 days. Longer term tests are underway.

Future tests will examine steel corrosion risks

directly.

CONCLUSIONSIt is difficult to change cement manufacturing

technology to greatly reduce CO2 emissions

because few alternative raw materials are widely

available in sufficient quantities. Even if the

materials are available, transportation costs can

be high.

However, our results show that it is possible,

under the right conditions, to manufacture

cements having equivalent performance to

Portland cements but with significantly lower CO2

emissions per unit of clinker, which implies also

the possibility for lower CO2 emissions per unit

volume of concrete.

Cements based on C4A3 plus ferrites, calcium

sulfates and either ggbs or activated belite are a

promising option. They can tolerate high sulfate

contents and hydrate to form mainly ettringite,

C-S-H & AFm phases. Preliminary concrete tests

of two alternative approaches to this have shown

promising strength and durability results, but

much more work is needed to establish sufficient

data to allow for normalisation of such cements

and their acceptance in general construction

applications. In developing a new cement for

large scale use a number of key objectives would

need to be met:

1. CO2 emissions and energy consumption

from manufacture needs to be as low as

possible. Other emissions (e.g. NOx and

SOx) also need to be tightly controlled,

2. The cement must be fit for purpose when

used in concrete, mortar, etc.

For example:

• Concrete made using it needs to have

handling characteristics that are

appropriate for a high-volume

construction material. It must be

consistent with H&S legislation, be

readily mixable using conventional

concrete mixing equipment, have

handling and setting times that allow

easy placement, but yet gain strength at

a reasonable rate (under typical curing

conditions) so to allow subsequent work

to progress rapidly

• Characteristics such as strength

development, durability, etc., must be

comparable to those of existing

construction cements (particularly

Portland cements)

3. Raw materials must be cheap and readily

available locally. Ideally waste and by-

product materials from other sectors

should be consumed in making such

cements

4. Other processing costs also need to be

acceptable, such that the final cost of

such cements is low enough to be

competitive as a means of reducing

cement-related industrial CO2 emissions,

(as opposed, for example, to the proposed

separation of CO2 from cement plant

exhaust gases and its liquefaction for

underground sequestration).

Whilst it is to some degree possible to tailor

the properties of a cement to a particular

application, the aim here is to develop a material

for bulk use as with Portland cement.

Consequently, appropriate codes and standards

for a new class of cements will need to be

developed. This will require a major investment

in collecting and analysing appropriate data and

developing the codes and standards themselves.

REFERENCES

1. LIVESEY,P., Challenges for the cementindustry, p321 Concrete for environmentalenhancement and protection, ed.DHIR,R.K., and DYER,T.D., Spon, London,1996.

2. European cement association (Cembureau);‘Energy is a sensitive factor in cementmanufacture’; www.cembureau.be, 2000.

3. van OSS,H.G., ‘Background Facts and IssuesConcerning Cement and Cement Data’, USGeological Survey Open-File Report 2005-1152, 2005.

4. PRICE,L., & WORRELL,E., Global energy use,CO2 emissions and the potential forreduction in the cement industry, IEACement Energy workshop, Paris, September2006.

5. GARTNER,E., ‘Industrially InterestingApproaches to ‘Low CO2’ Cements’. Cem.Concr. Res. 34 [9] 1489–1498, 2004.

6. QUILLIN,K., ‘Low energy cements’, BREReport 421, 1999.

7. LAWRENCE,C.D., ‘The production of lowenergy cements’, in Lea’s Chemistry ofCement and Concrete, 4th Edition, Ed.HEWLETT,P.C., Arnold, London, 1998.

8. BS 1881: Part 116: 1983: Method fordetermination of compressive strength ofconcrete cubes, British Standards Institution,London, 1983.

9. QUILLIN,K., ‘Low carbon belite-calciumsulfoaluminate cements’, to be published2007.

37

38

Dr Phil Purnell is an Associate

Professor at Warwick University.

His research concerns

composite materials, especially

cementitious composites and

their durability.

ABSTRACTA novel NDE technique – single sided

capacitive imaging, SSCI – has been used to study

concrete components in the lab and on in-situ

highway support structures. The potential for the

technique to overcome some of the limitations of

existing concrete NDE equipment is described. It

was shown that in the lab, SSCI can detect both

rebars and voids. The field studies, scans of 30 m

crossbeams, showed that the technique gives

repeatable results and appears to be able to

detect known anomalies such as repaired areas.

An intrusive investigation using core drilling and

hydro-demolition breakout has been initiated

guided by the findings of the SSCI survey. The

results of this will be used to validate and

calibrate the technique. The potential for further

development of SSCI instruments that can detect

rebars, voids and corrosion in a single system is

described.

KEYWORDSNon-destructive evaluation, Capacitance

imaging.

INTRODUCTIONThe methods for non-destructive evaluation

(NDE) of concrete structures are almost as varied

as the problems and ailments from which these

structures can suffer. Steel corrosion is tracked

using half-cell potentiometers, mystery

reinforcement layouts probed with covermeters

and delamination detected by hammer tap

surveys; the benefits of having a really good look

with the eyes are also not to be underestimated.

Nonetheless, we still frequently find ourselves in

the dark when confronted with needing to know

what might be going on in the interior of an

infrastructure component. There are a number of

reasons for this. First, NDE surveys tend to be

expensive; even if the equipment is relatively

cheap, experts must often undertake the

interpretation of the results and this escalates the

cost. Secondly, much of the available equipment

has severe technical or operational limitations

that prevent us getting a full picture of the

problem; these are discussed in more detail

below. Finally, most concrete NDE equipment can

only be used to assess one problem at a time. A

covermeter cannot detect voids; a half-cell

potentiometer cannot predict bar sizes. Thus the

‘perfect’ concrete NDE technique would be

cheap, easy to use and interpret, and be able to

detect multiple artefacts of different types.

A new NDE method, pioneered at Warwick

University, has the potential to fulfil some of

these objectives. In this paper, we will briefly

review some (not all) of the NDE techniques

currently available and their limitations, give an

overview of the development of the new

technique and its potential advantages and give

the results from a variety of laboratory tests. In

the final part, we will give an account of an

ongoing site investigation using the new

technique. We have examined a number of large

highway support infrastructure components, in

order to try and detect structurally critical sub-

surface voids prior to the design of a major repair

scheme. At the time of writing, surveys were

more or less complete: in the next few weeks,

core drilling will be undertaken to confirm or

deny the accuracy of the technique. Hopefully,

these results should be ready in time for the

presentation.

OVERVIEW OF EXISTING TECHNIQUES

The covermeter is probably the most

commonly encountered NDE equipment in

industry. They work on the electromagnetic

‘induction’ principle. There are two types of

system: the passive system, where the current

induced in an iron-cored coil by an AC source is

affected by the presence of a rebar; or the active

system, where eddy currents are induced in a

rebar by an air-cored coil and the impedance

39

UNDER THE COVER:

NOVEL NDE IMAGING TECHNIQUES FOR CONCRETE*

Phil Purnell, BEng, PhD,

University of Warwick

* This paper was co-authored by Professor Dave Hutchins, Dr. Geoff Diamond and Dr. Matthew Ing.

40

measured. Both systems can successfully

determine cover depth and rebar size in many

situations; but, if the rebar is deeper than about

60 mm into the concrete or the reinforcement

layout is congested then the covermeter can give

misleading readings, or fail to detect the steel at

all. If there are two or more layers of

reinforcement on top of one another, then the

covermeter will generally only detect the top layer

as lower layers are ‘screened’ e.g. [1].

Ultrasonic techniques have also been used to

assess concrete structures for many years [2]. The

time of flight of ultrasonic pulses – or, the speed

of sound in the concrete – can be correlated to

the strength of the concrete fairly well, either in

an absolute sense or more commonly in a relative

sense (i.e. looking for ‘low spots’ or ‘hot spots’).

To work properly, access is required to both sides

of a component so that a signal can be sent from

one transducer to another through the concrete.

This is not always possible. The transducers –

generally based on piezoelectric technology – also

need a good contact with the concrete, which

often requires surface preparation and couplants.

Imaging in concrete using ultrasound is very

difficult as it relies on the acoustic mismatch

between internal features, such as voids and

rebars, and the concrete. The problem is that this

mismatch tends to be similar to that between the

cement paste and the aggregate, causing

scattering of the signals and poor quality images[3]. Some success has been achieved at Warwick

with non-contact scanning ultrasound using

special transducers and powerful signal

processing [4], but from an imaging point of view

this is currently lab based.

More exotic techniques employ technology

based on ground-penetrating radar (GPR)

technology, as used in geophysical exploration.

Radar waves are reflected from boundaries

between materials with different dielectric

constant. Because of the wavelength of radar,

resolution is limited to around 100 mm for

commercial systems. Interpretation of the

complex 2-D colour plots is also something of an

art form and rather subjective [5,6].

The half-cell potentiometer can be used to

infer the probability of corrosion of the steel from

the surface of the concrete, since the degradation

is an electro-chemical process involving corrosion

of steel in an electrolyte (the fluids in the pores of

the concrete). A half-cell is basically a tube

containing an electrode in a saturated solution of

its own ions (usually silver and silver sulphate)

with one end sealed and the other plugged with

a porous material. The porous end is pressed

onto the surface of the concrete and the other

end is connected, via a high-impedance

voltmeter, to either the reinforcing steel (which

needs to be exposed) or a static ‘reference cell’.

An electrolytic connection is therefore made via

the half-cell and the pore fluid in the concrete to

the steel and the voltage measured. This is done

on a grid and a contour map produced. Areas of

high negative voltage gradient dV/dx – not high

voltage itself, as espoused by many practitioners

– indicate high corrosion current and thus areas

of likely steel degradation. Although widely used

and very successful with rapidly corroding steel,

in areas of slow corrosion or heavily carbonated

concrete it can be difficult to get sensible and

stable readings. Furthermore, breaking out of the

concrete cover is required to make the electrical

connection and/or calibrate the potential gradient

Figure 1: Schematic of the capacitive imaging transducer layout.

to the condition of the steel. Variations in the

moisture content of the concrete can also have

very severe effects [7].

The New Technique: Single-sidedcapacitive imaging (SSCI)

Our new NDE method is based on exploiting

the electrical properties of concrete, or more

precisely, the large contrast between its electrical

properties and those of the internal artefacts –

voids and rebars – that we are likely to want to

detect. The basic principle is very simple. A probe

made up of a pair of metal plate electrodes is

positioned parallel to the surface of the concrete.

A small air gap is left between the electrodes and

the surface, controlled by a simple spacer (Fig. 1).

A fixed frequency AC voltage is applied to the

input electrode, causing an electric field to be set

up in the concrete. This induces a charge on the

other electrode, which is measured as the output.

The concrete is effectively acting as the dielectric

in a capacitor: the only difference is that the

plates are coplanar, rather than opposite each

other as in a traditional capacitor. This probe is

then scanned over the surface of the concrete.

The presence of a rebar, void or other internal

detail will change the overall dielectric properties

of the volume through which the electric field is

passing. This change will be detected as a

change in the charge measured at the output

electrode and so a map of local contrasts in

dielectric properties can be produced. Since the

dielectric properties of the items of interest –

concrete, air filled voids, water filled voids and

steel - are relatively large (dielectric constants of

about 5, 1, 80 and zero/undefined, since steel is

a conductor, respectively) this technique should

be able to pick up and tell the difference

between many features of interest.

The penetration depth and shape of the

electric field can be easily modified. The depth is

determined by a combination of electrode

spacing and the voltage frequency used at the

input electrode. The shape of the electrodes and

the frequency of excitation are the main factors

that determine the distribution of the electric

field. Using electrodes shapes with sharp features

e.g. triangles, or concentric electrodes, causes the

field lines to become concentrated vertically into

a highly localised thin filament – the field

becomes like a probing finger – increasing the

resolving power of the technique. As the probe is

scanned over the surface, it should potentially be

possible to get a 3-D picture of the internal

structure. Further technical details of the system,

applied to wide variety of materials systems, can

be found elsewhere [8].

Potential advantages of thetechnique

Although at first glance SSCI may seem similar

to electromagnetic methods (i.e. the covermeter),

the mode of operation is different. No eddy

currents are generated in the sample and only

distortions of electric fields are of interest, as

opposed to measurements of magnetic fields.

This leads to two immediate advantages. First,

the technique is not subject to the same type of

shielding effect, meaning that it may be possible

to detect multiple layers of reinforcement, since

we can easily vary the penetration depth of the

technique. Secondly, we are not reliant on the

features of interest being conductive. As long as

they have some contrast in dielectric properties

compared with concrete, SSCI can detect them.

This means we should be able to detect rebar,

voids, cracks, delaminations etc all with the same

instrument.

The transducers for SSCI are about as simple

as it is possible for a transducer to be: a metal

plate. They can be very cheaply and conveniently

produced in any size or shape using printed

circuit board technology. Some of our prototype

probes are shown in Figure 2. Simple means

cheap; the projected cost for a SSCI transducer is

around five to ten pounds, compared with

hundreds of pounds for e.g. ultrasonic

transducers, so a client would be able to add

extra probes for specialised applications very

easily. It also means that the complexity can be

concentrated at the signal-processing end of the

system, where the ever-increasing computing

power of personal computers and developments

in software can be fully exploited.

The difference in dielectric properties between

details of interest and concrete is very much

larger than that between aggregate and cement

paste and so the scattering effect encountered in

ultrasound imaging is absent. This can be seen in

Figure 3. Note how the SSCI image does not

contain the noise associated with large aggregate

particles seen in the ultrasonic image. Also shown

in Figure 3 is the ability of the technique to image

sub-surface voids.

DEVELOPING THE TECHNIQUEThe preliminary results shown above were

obtained using equipment designed for general-

41

42

purpose use such as composites, aerospace

materials, food processing etc. Since we thought

that SSCI might be particularly useful in civil

applications, a project was set up to try and build

a prototype system. The project combined some

preliminary work by academics in the School with

a MEng final year group project, in both cases

pooling expertise from civil and electronic

engineering. Initially, the focus was on rebar

detection in the lab. In mid-2006, AmeyMouchel,

who provide long-term maintenance,

improvement and management services within

the transportation infrastructure sector, became

involved in the project and the focus shifted to

include detection of voids in-situ in highway

support structures. A number of structures in

their portfolio were identified where the

application of an SSCI survey might be of great

help as a pre-repair survey to ensure that before

deteriorated concrete is removed, any

construction voids that may exist within the

structure are detected and repaired, thus avoiding

any detrimental effect they would have on the

planned repair process. Areas identified as

potential voids or weak spots were to be core-

drilled in order to validate and calibrate the

technique. Time constraints were imposed owing

to access restrictions and contractual issues,

which meant that the laboratory and field trials

were forced to happen in parallel. As such, the

technique is still constantly evolving, and the

results presented here are preliminary; a snapshot

of progress to date.

Laboratory trialsTwo slabs were cast in the laboratory to act as

test beds for the prototype equipment. These are

shown in Figure 4. Both are 1m square and 120

Figure 2: Some prototype SSCI probes. Note sustainable recycling of packagingmaterials.

Figure 3: (a) Comparison between non-contact ultrasonic through transmissionimaging (top) and SSCI (bottom), 10 mm rebar embedded in 50 mm thick concreteslab. (b) Engineered hidden void in Thermalite block and corresponding SSCI scan.

mm thick, cast from C40 concrete with limestone

aggregate. Slab 1 was unreinforced, with a

stepped channel cast into the underside to

simulate a void. The channel ranged from 60 mm

to 90 mm deep (i.e. the cover over the void from

60 mm to 30 mm) with the step occurring in the

centre. Slab 2 was reinforced with two crossing

layers of 8 mm rebars, at 225 mm spacings, 100

mm below the surface (the stub ends of the bars

can be seen in the photos) i.e. the effective cover

was >90 mm, more than the normal 75 mm

maximum in Eurocode 2. Both slabs had lifting

steel cast in place (as can be seen in slab 2),

which was continuous along the length of the

slabs.

The prototype transducers were scanned over

the slabs using a 30 mm grid. Circuitry within the

system converted the charge signal to a voltage

and this was recorded. The results are shown in

figure 5. In the scan of slab 1 (left hand picture),

the wide dark area corresponds to the position of

the void, and the speckled grey areas at the

edges to the position of the lifting steel

reinforcement. Thus the technique appears to be

able to capture both rebars and voids in the same

pass. Some problems remain: the thin black

areas are artefacts caused by the contouring

process smoothing the transition from rebar areas

to plain areas, and the bulls-eye features in the

centre are caused by minor surface irregularities,

but these anomalies can be removed by image

processing: the raw results are presented here.

The scan for the reinforced slab is on the right. In

this case, solid grey areas correspond to

reinforcement and black/speckled areas to plain

concrete. The pattern of the crossing re-bars -

both layers – can clearly be seen and has been

detected at a depth of 100 mm, way beyond that

achievable with standard covermeters. Again, the

raw results are presented to demonstrate proof of

concept and image processing would produce

‘prettier’ images.

Both these results were taken using a 500 kHz

signal. Since then, modelling studies have

suggested that changing the frequency of the

scanning signal, as well as changing the depth of

penetration, also changes the sensitivity of the

technique to certain features; for example, high

43

Figure 4: Laboratory test slabs. Left: Slab 1, unreinforced slab with hidden steppedchannel. Right: Slab 2, Reinforced slab. The prototype equipment is fully containedwithin the toolbox shown.

Figure 5: SSCI results for laboratory slabs. Left: slab 1, Right: slab 2. Both scanstaken at 500 kHz.

44

frequencies are more efficient at finding rebars,

while lower frequencies are better at detecting

voids. Ongoing development of the technique is

attempting to exploit this.

Field InvestigationsThe prototype equipment was used on site to

investigate a number of highway support

structure crossbeams that were scheduled for

investigation and repair. Historical investigations

had identified a large internal void in one of the

crossbeams. While this had been successfully

repaired, further investigative studies were

undertaken. Access to the beams was via a cherry

picker and readings were taken as four horizontal

scans per crossbeam, 250 mm apart vertically, at

150 mm horizontal spacing, which gave suitable

resolution to detect critical voids. Figure 6 shows

a typical crossbeam and the testing kit in action.

Results were plotted as contour plots of

measured voltage against position (in much the

same way as a half-cell survey), effectively

showing up differences in dielectric properties

between the various areas of the beam. These

contour plots were then overlaid onto elevation

drawings of the crossbeams complete with the

results from a previous hammer-visual survey to

check for correlations between the two methods

and to isolate potential voids from surface

delamination. After reviewing the initial surveys, a

series of secondary surveys was started to re-

check areas of interest and confirm repeatability;

these are still ongoing but some results are

included here. Significant development of the

probes occurred between the initial and

secondary surveys and many of the secondary

surveys were carried out at a higher resolution.

Figure 7 gives an overview of a typical survey.

The grey contours are our survey: the grid and

hatchings are a hammer/visual survey carried out

by other investigators (hatchings being suspected

delaminations). Vertical gridlines are

approximately 0.5 m apart; the total length of the

crossbeams is about 30 m. The lab results suggest

that the darker an area is compared to the

‘background’, the more likely it is to contain a

void; lighter areas should correspond to better

quality concrete. Figures 8 and 9 show some

areas of interest in close up. In figure 8a, the area

between the dashed lines and that circled are

areas in which previous repairs are known to have

taken place; these seem to correlate with lighter

areas in the initial and/or secondary surveys. In

figure 8b, the repeatability of the technique is

confirmed: a suspected void imaged on the initial

survey is more prominent on the secondary survey

with optimised equipment. However until

intrusive investigations are undertaken, the

presence of a void cannot be confirmed. Figure 9

gives some further overviews, showing the

Figure 7: Overview of typical crossbeam survey. Top row picture: complete initialsurvey. Bottom row pictures: secondary surveys with improved equipment.

a b

Figure 6: Site investigations. Left: typical crossbeam showing supporting steel. Notethe exposed rebars on the soffit near the column. Right: two-man team in cherrypicker taking readings.

patterns of light and dark areas characteristic of

SSCI scans. Where possible, crossbeams were

scanned twice, once from each side. Major

features could be correlated on both scans,

further enhancing our confidence in the

technique.

DISCUSSIONThe technique has shown promise in the lab:

voids and rebars can definitely be detected with a

single probe, more easily and at greater depth

than with other techniques. A much more

powerful validation of the technique will arise

from the correlations between the site surveys

outlined above and the core drills that will be

taken over the next few weeks. At the moment,

we are recommending that the very darkest areas

on the scans should be drilled first in a

preliminary investigation, along with selected

other anomalies. The cores will be examined

using the array of techniques available at

Warwick and will eventually be tested to

destruction to assess the compressive strength.

Selected parts of the crossbeams will also be

broken out using hydro-demolition to get a better

idea of the internal structure. It is very unusual to

be able to get such real world data, in parallel

with ongoing lab trials, so early in the

development cycle of a new NDT technique and it

is testament to the partnership between the

University and AmeyMouchel that this project has

progressed so far so quickly. We hope that this

combination of field/lab investigation will inspire

other sectors of the industry to help develop the

technique further.

There are a number of avenues for

development for this new technique. The main

thrust of the field study has been to detect voids

of cricket-ball size and upwards, and the

equipment has been optimised accordingly. The

logical next step is to look at the best probe and

signal combinations to examine rebars in-situ. As

well as position and size, it might also be possible

to glean some information on the condition of

the rebars. Given that corrosion is an

electrochemical process, we might expect that

the dielectric properties of the interface between

a rebar and the surrounding concrete could

change once corrosion starts. This might allow us

to tell the difference between good and suspect

reinforcement. With the right combination of

electrode shapes and signal frequencies, it should

theoretically be possible to detect voids, evaluate

rebars and identify corrosion with a single

instrument.

The methods of data collection and

presentation are also a top priority for

development. At the moment, the 2-D data is

collected in a traditional fashion: the probe is

used on a chalked grid, measurements taken with

a voltmeter and plotted later in the office to form

an image. We need to move towards a system

whereby a probe, with a flexible combination of

signal frequencies and electrode size/shapes, is

simply scanned along the surface and a 3-D

image generated immediately on-site. This

45

a b

Figure 8: (Left = inset a, Fig. 7. Right = inset b, Fig. 7)

Figure 9: Selected overview scans of other crossbeams.

46

represents a significant engineering challenge,

but the basic ingredients of position sensing,

automatic data logging, 3-D tomography and

image processing are used today in other NDE

fields such as medical imaging; it is just a matter

of combining them. We have already taken the

first steps towards automatic data logging but

moving to the complete system will require a new

project, in partnership with industry.

CONCLUSIONSThe proof of concept of single sided capacitive

imaging for concrete structures has been proven

in the lab. Voids and rebars can be detected and

imaged with a single system, at cover depths

beyond the reach of many other techniques.

Field trials of the system on in-situ highway

support structures are encouraging in that they

give repeatable results and appear to image

known internal features such as repairs.

Validation and optimisation of the technique

must involve core drilling and breaking out at

locations suggested that the scans suggest to be

deficient; a programme to achieve this is already

underway. The technique has several avenues for

optimisation that could lead to a versatile

instrument for non-destructive evaluation of a

wide range of problems in concrete components.

ACKNOWLEDGEMENTSMuch of this work has been carried out by an

undergraduate Final Year MEng project group,

consisting of a mix of engineers of all

specialisations. Their efforts are greatly

appreciated and special mentions are owed to

Paul Hunt, Tom Carter and Shakil Ismail for their

help with the data for this paper. We would also

like to thank AmeyMouchel, particularly Matthew

Ing and Andrew Wrangles, for allowing us to

access several of the structures in their portfolio,

providing equipment and personnel on site, and

agreeing to align their diagnosis and repair

programme for these structures with our needs.

This represents a significant investment, without

which this project would not have taken place.

The SSCI/ECI technique and subsequent

apparatus has been developed by the University

of Warwick in conjunction with Warwick NDT and

a fuller specification is available in the company

technical technical bulletin which can be found at

www.warwickndt.co.uk.

REFERENCES

1. DEROBERT, X., AUBAGNAC C., ABRAHAM,O. Comparison of NDT techniques on apost-tensioned beam before its autopsy,NDT&E International vol. 35, 2002, p452 etseq.

2. JONES, R., The non-destructive testing ofconcrete. Magazine of Concrete Research 2(June) 1949, 67-78.

3. SCHICKERT, M., Ultrasonic NDE of concrete,Proceedings of the 2002 IEEE UltrasonicsSymposium, pp. 718-727.

4. PURNELL, P., HUTCHINS,D.A. Sounding itout - advances in ultrasonic testing. CivilEngineering (Proceedings of the ICE) vol.157 no.2, 2004 p56

5. BUNGEY, J.H., Sub-surface radar testing ofconcrete – a review. Construction andBuilding Materials vol. 18, 2004 p2 et seq

6. BARRILE V., PUCINOTTI, R. Application ofradar to reinforced concrete structures: Acase study. NDT&E International vol. 38,2005, pp596-604.

7. BAKER, A.F., Structural Investigations, inDurability of Concrete Structures,Investigation, repair, protection ed. G Mays.Spon, London, UK 1992 p63.

8. DIAMOND, G.G., HUTCHINS, D.A., GAN,T.H., PURNELL, P., LEONG, K.K., Single-sidedcapacitive imaging for NDT. Insight vol. 28no. 12, 2006, pp724-730.

Klaus-Alexander Rieder is a

principal scientist at W.R.

Grace, Cambridge, Mass. He

received his ScD in physics from

the Technical University of

Vienna, Austria, in 1995. He is

a member of ACI Committees 209, Creep and

Shrinkage in Concrete; 215, Fatigue of Concrete;

360, Design of Slabs on Ground; 446, Fracture

Mechanics; and 544, Fiber Reinforced Concrete.

He is also member of ASTM and currently

chairing task group E of C09.42. His research

interests include all durability aspects of concrete,

cracking of concrete, fracture mechanics, and the

development of high-performance fibres for

concrete applications. He has authored and co-

authored several papers and holds more than 10

patents in the field of fibre reinforcement.

ABSTRACTThis paper discusses various aspects of recent

developments in the field of synthetic macro

fibres. Different types of synthetic macro fibres

are introduced and their ability to increase the

post-cracking capacity of concrete. A simple

model is presented linking the material properties

of synthetic macro fibres to the material and

structural behaviour of concrete. In addition to

the short-term fibre material properties, some

long-term material property data are presented,

and their importance relative to the durability of

fibre-reinforced concrete is examined. Typical

applications of synthetic macro fibre reinforced

concrete are shown.

KEYWORDSSynthetic macro fibres, Steel fibres, Fibre-

reinforced concrete, Strength, Elastic modulus,

Toughness, Creep, Durability of cracked concrete,

Slab-on-ground, Steel decking.

INTRODUCTIONThe continued pressure to decrease costs

further for buildings and infrastructure, despite

the rising costs of raw materials and labour and

the creativity of architects in coming up with new

building designs, are only two reasons why the

construction industry is constantly looking for

new innovative materials and solutions. In the last

couple of years the well-documented rise in steel

prices, which is the second most used material

behind concrete, put even more pressure on

project managers and engineers to find ways to

reduce cost. Steel is used in beams, girders,

rebars, and mesh or fibres, just to name a few.

Steel fabric is mainly used to control drying

shrinkage and thermal cracking, for example in

slabs. To position the steel fabric properly in a

slab to provide optimum benefit to minimize

drying shrinkage cracking is very labour intense

and time consuming since the mesh needs to be

placed in the top third of the slab and thus needs

to be attached to closely spaced chairs. The

concrete also needs to be pumped from the

ready-mix truck since the already placed steel

fabric prevents the truck from discharging the

concrete where it is actually needed. It is also very

difficult to use laser screeds, which consolidate

and level the concrete and are especially efficient

when large concrete slabs are being poured.

The above-mentioned disadvantages have

lead to increased use of steel fibre reinforced

concrete, where deformed steel fibres are mixed

directly into concrete, replacing the steel fabric. It

has been demonstrated in the past through many

tests conducted by Universities [1, 2, 3, 4, 5, 6] and

thousands of completed projects that steel fibres

used at a sufficient dosage rate provide better or

equal crack control than mesh reinforcement in

applications such as industrial slabs. In addition to

the drying shrinkage crack control that steel fibre

reinforced concrete provides, it was shown that

the flexural capacity of slabs supported on

ground also increased. In order to achieve the

same benefit with mesh, a second layer of mesh

needs to be installed located in the bottom third

of the slab.

Despite the obvious benefits of steel fibre

reinforced concrete compared to conventional

steel fabric reinforcement, this technology also

has several shortcomings: to achieve an

homogeneous fibre distribution in the concrete,

most of the steel fibres need to be added to the

plastic concrete with special dispensers to avoid

fibre balling and nesting. When steel fibre

reinforced concrete needs to be pumped, there is

increased mechanical wear on the pump, hoses

47

SYNTHETIC MACRO FIBRES

Dr Klaus-Alexander Rieder, ScD,

Grace Construction Products Ltd

48

and finishing equipment and the pump pressure

needs to be increased. Although the high pH of

the concrete limits the corrosion of steel fibres

within the concrete, steel fibres close to the

concrete surface and in cracked concrete sections

will start to corrode, if the concrete gets in

contact with water, chlorides or acids. This effect

will lead to surface staining and in the worst-case

scenario in a cracked concrete section with a

crack width of more than 0.5 mm to a loss of

crack control capacity [7, 8, 9], unless of course

stainless steel fibres are used. Another potential

problem is related to steel fibres protruding out

of the concrete surface, which can be caused by

improper concrete placement or by concrete

spalling, abrasion or degradation. This can lead to

damage to vehicle tyres and could cause harm to

pedestrians and animals.

Synthetic Macro FibresRecent innovations in the field of fibre

reinforcement have led to the development of a

new class of fibres – the polymeric macro fibres,

which provide significant amounts of post-

cracking toughness or post-cracking flexural

strength to the concrete. This is in contrast to the

already well-accepted synthetic micro fibres,

which usually have a fibre diameter in the range

of 20 to 50 µm and are used only to control

plastic shrinkage cracking.

The polymeric macro fibres obviously cannot

corrode since most of them are made out of

polypropylene, polyethylene or a blend thereof.

They are usually much easier to handle than steel

fibres since the specific gravity of polypropylene is

about 12% that of steel. They maintain their

mechanical properties in high as well as in low

pH environments. The (equivalent) diameter of

these macro synthetic fibres ranges from 0.5 to 1

mm, with a tensile strength between 350 and

700 N/mm2. Depending on the fibre type used,

different dosage rates are needed to achieve a

residual flexural strength of at least 1 N/mm2. As

with steel fibre reinforced concrete, the more

synthetic macro fibres are needed to achieve a

certain toughness performance target, the higher

the impact of the fibres on the workability and

finishability of the concrete. Hence, the

workability needs to be restored using a

plasticizer or superplasticizer (NOT WATER), in the

concrete mix design.

In addition, due to the relatively low modulus

of elasticity of these polymeric macro fibres,

which is in the range of 3000 to 5000 N/mm2 (3

to 5 GPa), the performance of these fibres at

small crack openings is inferior to the

performance of deformed steel fibres, which have

a modulus of elasticity between 180 000 and 210

000 N/mm2 (180 to 210 GPa). However, since

most of the commercially available steel fibres are

designed to fail in fibre-pull-out mode (fibre pulls

out of the concrete matrix) rather than fibre-

failure mode, where the fibre breaks, the high

modulus of steel is only partly utilized when the

concrete starts to crack. This shortcoming does

not represent a big problem in many applications

such as shotcrete linings for mining tunnels, but it

is a significant shortcoming in applications such

as civil tunnels or slab on ground floors, where

large cracks are not acceptable to the engineer or

owner of the structure.

A New High PerformanceSynthetic Macro Fibre

Recently, a new synthetic macro fibre STRUX®

90/40 developed and patented by W R Grace

,and shown in Figure 1, was introduced to the

German construction market, where it recently

obtained DIBt (German Institute for Civil

Engineering) approval. The main components of

this polymeric fibre type (PP) are polypropylene

and polyethylene. This synthetic macro fibre's

mechanical and geometric properties are

significantly different from existing synthetic

macro fibres. The length of STRUX 90/40 is 40

mm with an aspect ratio (length divided by the

equivalent diameter) of 90 and a specific gravity

of 0.92. The fibre has a rectangular cross-section

with an average width of 1.40 mm and an

average thickness of 0.105 mm. The average

tensile strength of the fibre is 620 N/mm2 (MPa)

with a modulus of elasticity (chord modulus) of

9500 N/mm2 (9.5 GPa).

The ‘flat’ fibre design was chosen in part to

achieve a large fibre surface area to fibre volume

ratio in order to increase the mechanical bond

between the fibre and the cement paste. The

elastic modulus of this synthetic macro fibreFigure 1: Photograph of synthetic macrofibre STRUX 90/40.

nearly matches the elastic modulus of the cement

paste, where the fibre is embedded in, which

allows the fibre to transfer stresses across a crack

immediately after crack initiation has occurred.

Consequently, good crack control can be

achieved in concrete. Synthetic macro fibres with

either an insufficient bond or lower elastic

modulus first need to elongate before they are

able to transfer significant amounts of stresses

across widening cracks leading to larger crack

openings. This can be observed when measuring

the flexural performance (see Figure 2) using a

beam test set-up such as described in ASTM

C1609-06 [10] or JSCE-SF4 [11], which is being used

for designs of fibre reinforced concrete slab-on-

ground in CSTR 34 [12].

In the case of specimens containing lower

modulus crimped PP fibre, the stress drops

significantly right after the peak of the flexural

stress versus beam deflection. At larger beam

deflections, the post-cracking flexural strength

starts to increase again. At the same fibre

49

Figure 2: Flexural stress versus deflection curves measured according to ASTM C1609-06.

Figure 3: Flexural stress versus deflection curves measured according to DIN 1048 part 1.

50

addition rate, the higher modulus STRUX 90/40

fibre shows much better load carrying

characteristics after initial cracking. It nearly

matches the average behaviour of a concrete

reinforced with 20 kg of 50 mm long crimped

steel fibres with a wire diameter of 1.3 mm. After

a beam deflection of around 0.25 mm the STRUX

90/40 reinforced beams maintain their load

carrying capacity until the tests are stopped at a

beam deflection of 3 mm. At larger deflections,

the average residual load carrying capacity of 4.6

kg/m3 of STRUX 90/40 matches the performance

of 40 kg/m3 of that particular steel fibre. It has to

be noted that there are many different types of

steel fibres on the market which all perform

differently in concrete.

One aspect, however, should also not be

forgotten: the standard deviation of the post-

cracking flexural strength of a fibre reinforced

concrete with a high fibre count per volume of

concrete is generally lower compared to a fibre

reinforced concrete with a low fibre count. Since

one kilogram of STRUX 90/40 contains around

185 000 fibres (resulting in better matrix-

reinforcement homogeneity), the standard

deviation even at lower fibre addition rates is

much smaller compared to commonly used steel

fibres. For comparison, one kilogram of the 50

mm steel fibre with a wire diameter of 1.3 mm

contains only around 2000 individual fibres. The

standard deviation is important since the

characteristic value which takes into account the

average value, the standard deviation and the

number of samples tested, is used for design of

fibre reinforced concrete, for example in

Germany. In the DBV (German Concrete

Association) steel fibre technical bulletin [13], the

fibre reinforced concrete is classified in various

fibre classes. Figure 3 shows the performance of

STRUX 90/40 reinforced concrete beams tested

according to DIN 1048 part 1: Testing methods

for concrete.

The concrete mix design used for these tests

was according to the requirements of the

German Institute for Civil Engineering (DIBt) for

‘Fibre products used as concrete additives’ [14] . If

local materials from the UK and/or mix designs

with a higher fines content are used to make

STRUX 90/40 reinforced concrete, significantly

higher residual flexural strength values are

obtained (see Figure 4).

This approval process is now required for all

new fibres introduced into the German market, in

order for the fibre reinforced concrete to be

classified according to EN 206. STRUX 90/40 is

the first approved synthetic macro fibre in

Germany which can be classified in fibre classes

according to the DBV technical bulletin. It was

also approved as an effective fibre to control

plastic shrinkage cracking of concrete. At the

minimum fibre dosage rate needed to achieve 1

N/mm2 of post cracking flexural strength, STRUX

90/40 is as effective as micro fibres added at

typical dosage rates to control plastic shrinkage

cracking.

Figure 4: Flexural stress versus deflection curves measured according to ASTM C1609-06.

During the DIBt approval process additional

tests were carried out, which confirmed the static

effectiveness of STRUX 90/40 in fibre reinforced

concrete particularly with regard to long-term

effectiveness. Some of these findings will be

presented later in this paper.

Large Scale Testing of ConcreteSlabs-on-Ground Reinforced withSTRUX 90/40

Before STRUX 90/40 was introduced

commercially to the North American market, a

large scale-testing programme was conducted at

the University of Illinois [15,16]. One purpose of that

study was to measure and compare the structural

response of plain, steel fibre, steel fabric and

STRUX 90/40 reinforced concrete slabs under

interior and edge loading conditions. Simply

supported small-scale beam tests had

demonstrated that STRUX 90/40 fibres could

significantly increase the post-cracking strength or

toughness of concrete matching the performance

of steel fibres [17]. However, the structural benefits

of STRUX 90/40 added to plain concrete were

unknown at that time. Since the fibre was

specially developed and optimized for such an

application, the structural benefits had to be

investigated.

Large-scale slab tests were also required to

determine if the beam toughness results could be

used to predict the flexural and ultimate capacity

of the synthetic STRUX 90/40 reinforced slabs in

the same manner as previous research had done

with steel fibres [1, 2, 5]. This type of testing was

essential for new fibre types since slab-on-

ground design codes such as the Technical Report

34 (TR 34) published by the Concrete Society [18]

are using beam toughness results such as the Re,3

value to predict the concrete slab’s ultimate

capacity and to calculate the required slab

thickness.

The slab dimensions selected for this study

were 2.2 m x 2.2 m x 0.127 m. The fully

supported slabs were tested under monotonic

loading conditions in displacement control to

better capture the response of the concrete

before and after cracking. Load-induced strains

were measured using embedded strain gauges,

which were placed at various locations before the

slabs were cast. The objectives of the strain

gauge placement were to measure compressive

and tensile stresses at the top and bottom of the

slab. Surface deflections of the concrete slabs

were measured by an array of Linear Variable

Displacement Transducers (LVDTs) as shown in

Figure 5, where the fully instrumented slab can

be seen before the load is being applied. Detailed

information and interpretation of all the test

results obtained from these experiments can be

found elsewhere [14, 15, 16] only a brief summary

and conclusions will be presented in this article.

The ultimate load carrying capacity of the plain

concrete slab is improved significantly with the

addition of STRUX 90/40. Compared to un-

reinforced concrete the flexural capacity under

centre load conditions increased by 25% for a

concrete slab reinforced with 3.0 kg/m3 STRUX

90/40 and 32% for a slab reinforced with 4.4

kg/m3 STRUX 90/40 (see Figure 6). STRUX 90/40

reinforced concrete also outperformed steel fabric

reinforced concrete (As=123 mm2/m).

51

Figure 5: Instrumented concrete slab supported on ground ready for testing.

52

The equivalent flexural strength ratio (Re,3)

indicated a similar increase in the flexural and

ultimate capacity of the fibre reinforced concrete

slabs over the plain concrete slab. Fibres help to

keep the slab in contact with the ground after

flexural cracking is reached enabling a better

redistribution of load. Embedded strain gauges

further confirmed that the load carrying capacity

of the fibre reinforced concrete slabs was

distributed over a larger area before flexural

cracking and thus allowed for a higher flexural

and ultimate capacity relative to the plain and

mesh reinforced concrete slabs.

The equivalent flexural strength ratio (Re,3)

indicated a similar increase in the flexural and

ultimate capacity of the fibre reinforced concrete

slabs over the plain concrete slab. Fibres help to

keep the slab in contact with the ground after

flexural cracking is reached enabling a better

redistribution of load. Embedded strain gauges

further confirmed that the load carrying capacity

of the fibre reinforced concrete slabs was

distributed over a larger area before flexural

cracking and thus allowed for a higher flexural

and ultimate capacity relative to the plain and

mesh reinforced concrete slabs. Figure 7 shows a

failure cracking pattern for a concrete slab

reinforced with STRUX 90/40 (a) and plain

concrete slab (b).

The synthetic macro fibre reinforced slabs and

the steel fibre reinforced slabs had overall similar

fracture behaviour. The load-deflection curves

also indicated that the structural ductility of the

synthetic macro and steel fibre slabs was similar.

Synthetic macro fibres as well as steel fibres did

not increase the tensile capacity of the concrete

slabs, which was consistent with findings from

previously published reports.

The findings from this study confirmed that

the post-cracking flexural strength measured on a

relatively small beam could be used to predict the

Figure 6: Load versus centre slab deflection curves of plain concrete, steel fabric andSTRUX 90/40 reinforced concrete slabs.

Figure 7: Cracking pattern of slabs: a) STRUX 90/40 reinforced concrete,b) plain concrete.

structural performance of STRUX 90/40 as well as

steel fibre reinforced concrete slabs-on-ground.

Since then many hundreds of jobs which utilized

STRUX 90/40 in lieu of conventional steel fabric

or fibres in slab-on-ground applications have

been completed successfully worldwide. A

summary of applications where STRUX 90/40 has

been used is presented at the end of this paper.

Long-Term Behaviour of STRUX90/40 Reinforced Concrete

As mentioned before, during the DIBt approval

process there was a special focus and

investigation on the long-term properties of

synthetic macro fibre reinforced concrete. Since

concrete floors are the main application for

STRUX 90/40, this paper will focus on the

findings relevant to that application. In general, it

has been found that polypropylene is very stable

in the concrete environment. Tensile strength

measurements of synthetic micro fibres

embedded for more than 30 years in concrete

showed minimal loss of strength compared to its

initial value. Accelerated ageing tests can be

performed such as the one described in

Acceptance Criteria for Concrete with Synthetic

Fibres (AC 32), which was published by the ICC

evaluation service. These tests confirmed that

STRUX 90/40 reinforced concrete maintains its

toughness performance long-term. It should be

noted that the above-mentioned evaluation does

not include testing of cracked fibre reinforced

concrete specimens under sustained load.

In the past, there was limited information

available about this topic; neither steel fibre

reinforced concrete nor synthetic macro fibre

reinforced concrete has been tested extensively to

determine the creep coefficient of cracked fibre

reinforced concrete (FRC). Recent studies

conducted by Bernard [19] on pre-cracked round

fibre reinforced panels, which he subjected to

various amounts of sustained loading, showed

that the creep coefficient of steel and STRUX

90/40 reinforced shotcrete were comparable.

However, another type of synthetic macro fibre

exhibited a significant amount of creep during

the test period. MacKay et al. [20] conducted a

study on creep of steel as well as synthetic macro

fibre reinforced concrete based on cracked

beams. The synthetic macro fibre investigated in

this programme had an elastic modulus of about

5 to 6 GPa. Although this macro fibre did not

creep as much as the one Bernard investigated,

the authors found that the creep coefficient of

steel fibre reinforced concrete investigated was

about 50% of that of synthetic macro fibre

reinforced concrete.

In 2005 an initiative was started to revise the

existing Guideline: ´Fibre Reinforced Concrete’

published by the Austrian Society for Concrete

and Construction Technology. It was agreed to

evaluate a standardized test procedure to

measure the creep behaviour of steel and

synthetic macro fibre reinforced concrete. With

the financial support of several fibre suppliers, a

testing programme was initiated. In order to

further investigate the difference in creep

behaviour between steel fibre and STRUX fibre

reinforced concrete, W R Grace started a test

programme to compare the creep of both

materials based on the recommendations of the

Austrian fibre committee.

Beams were cast at the Grace research

laboratory in Cambridge, Massachusetts, USA. A

typical floor mix design was used with a cylinder

compressive strength between 35 and 40 MPa.

STRUX 90/40 was dosed at 2.0 kg/m3, while 20

kg/m3 of a 60 mm long hooked end steel fibres

were added to the concrete. Since these were the

first tests of a larger series, the dosage rates were

not selected based on obtaining similar residual

load carrying capacities, but starting the tests

with the lowest recommended fibre dosage rates.

The 150 mm x 150 mm x 550 mm beams were

cured for 28 days in lime-saturated water before

being tested in four-point bending. Instead of a

final beam deflection of 3.0 mm, the test was

stopped at a beam deflection of 1.75 mm. In

addition, the relaxation of load response of the

cracked fibre reinforced concrete specimen was

recorded to determine the amount of elastic

energy stored in the specimen at a deflection of

1.75 mm. The specimen was then carefully

removed from the testing machine. Three

specimens of each fibre type with approximately

the same amount of residual flexural strengths

were selected and placed in a specially designed

loading frame shown in Figure 8. The calibrated

loading frame was loaded with weights until

50% of the residual flexural load carrying

capacity was reached.

This value was chosen by the Austrian fibre

committee as a starting value for the evaluation

of the creep behaviour of cracked fibre reinforced

concrete, because this stress represents the

maximum stress in the serviceability state, which

will occur in a concrete section according to the

current Austrian design guideline for fibre

reinforced concrete. The load is applied in four

53

54

point bending configuration and the beam

deflection is measured with a mechanical dial

gauge. Figure 9 shows the average deflection (in

mm) under sustained load over time of the two

fibre reinforced materials.

The elastic part of the deflection, which was

measured before, was subtracted from the long-

term creep measurement. As can be seen from

these results, both fibre materials exhibit similar

long-term creep behaviour. The deflection of the

STRUX reinforced concrete beams increases more

in the first couple of hours compared to the steel

fibre reinforced concrete beams. After that the

slope of the curves are nearly identical. Creep

measurement of these samples has been ongoing

for one year and will be continued. In the

meantime, a second creep frame has been

constructed so more tests can be conducted at

the same time.

Contrary to the common belief that steel fibre

reinforced concrete cannot creep since steel has a

low coefficient of creep, the results obtained so

far show that indeed cracked steel fibre

reinforced concrete does creep. This is not caused

by creep of the steel itself but by slow pull-out of

steel fibres out of the concrete matrix and the

interface which creeps. For synthetic macro fibres,

a combination of creep of the fibre material as

Figure 8: Loading frame to measure creep of cracked fibre reinforced concrete beams

Figure 9: Increase of beam deflection under sustained load (without elastic part).

well as the creep of the fibre/matrix interface is

determining the overall creep behaviour of

cracked concrete. Since STRUX 90/40 has a high

elastic modulus, which is achieved by stretching

the film during the extrusion stage nearly to its

limits, the creep coefficient is much smaller

compared to synthetic macro fibres with a lower

elastic modulus, which have not been stretched

as much.

All the test results indicate that cracked

concrete reinforced with STRUX 90/40 loaded up

to 50% of its residual flexural strength will

behave similarly to steel fibre reinforced concrete

in terms of creep behaviour.

ApplicationsThe primary application of STRUX 90/40 has

been the reinforcement of concrete slabs-on-

ground. In more than 100 000 m2 of finished

floors, steel fabric had been replaced with various

amounts of STRUX 90/40 depending on the local

requirements. Especially for use in ready-mix

applications, it is important to provide a fibre that

can be added easily to the concrete with minimal

impact on the normal operation. The fibre's

geometry was also selected such that it would

disperse readily in the plastic concrete and would

not adversely affect finishing characteristics at

normal volume fractions, which range from 0.25

to 0.75% (2.3 to 6.9 kg per m3 of concrete).

55

Figure 10: Pumping and placing of STRUX 90/40 fibre reinforced concrete.

Figure 11: Levelling of STRUX 90/40 reinforced concrete with a laser screed.

56

The fibre can be added to the empty ready-mix

truck prior to loading the concrete constituents or

concrete, therefore minimizing the time the

ready-mix truck has to spend in the concrete

plant. After the concrete constituents are added,

STRUX fibres are then mixed for at least five

minutes at maximum mixing speed, allowing the

fibres to disperse throughout the concrete. Due

to the mixing action with the concrete, the

surface of the fibre is slightly abraded which

increases the bond between the fibre and the

concrete matrix. It is important to note that the

addition of STRUX fibres will decrease the original

workability due to its relatively high fibre surface

area. The workability needs to be restored by

adding additional water reducers or

superplasticizers.

The fibre concrete can be easily pumped, if

needed. Figure 10 shows the placing of concrete

reinforced with 5.3 kg/m3 for a composite steel

decking job in Edinburgh, where more than 1200

m3 of concrete were pumped onto Richard Lees

Steel Decking (RLSD) with ease. Experience shows

that the pump pressure remained low, even for

higher fibre addition rates since the fibres are

very flexible and offer minimum resistance in

shear due to concrete flow. The flexibility of the

fibre also makes the finishing process of the

concrete surface easier. Vibrating screeds are

recommended to level the concrete and bring

enough paste to the surface to embed the fibres.

For larger section of the job, a large laser screed

was used at this job (seen in Figure 11) to

compact and level the concrete. Since there was

no wire mesh placed on the floor the laser screed

could move freely around the job site. For smaller

jobs, a vibrating screed bar also has been used

successfully to level the fibre reinforced concrete.

Figure 12 shows how the STRUX 90/40

reinforced concrete is bull-floated to bring up

additional paste for the final finishing process. For

indoor slabs, the concrete is generally finished

using a power trowel leading to near fibre-free

surfaces seen in Figure 13.

For outdoor applications, a broom finish or

"panned" finish is normally applied (see Figure

14). In this case, some fibres are expected to be

seen on the concrete surface, which will not

cause any harm to the environment, since they

will quickly wear off due to abrasion. The STRUX

fibres close to the surface will provide the plastic

shrinkage crack control needed for these

applications.

Recent projects, where STRUX 90/40 have

been used successfully include an apron repair

project at the airport of San Francisco, where

instead of replacing the complete concrete floor

only the top half of the concrete floor was

removed and replaced with STRUX 90/40

reinforced concrete. Another area of application

includes thin overlays also called whitetopping or

ultra-thin whitetopping for bus stops.

As mentioned before, large outdoor concrete

slabs have been successfully reinforced with

STRUX fibres as well as large industrial floors. In

recent years concrete floors of farm buildings

Figure 12: Finishing of STRUX 90/40 reinforced concrete with a bull float.

such as cow, pig, and chicken stables have been

reinforced with STRUX 90/40. Conventional steel

fabric reinforcement corroded very quickly in such

applications due to the acid environment caused

by the manure. Steel fibres could not be used

either, due to the corrosion issue as well as the

danger that animals might get hurt by steel fibres

sticking out of the degraded/damaged concrete.

In England a new sea defence wall is being

built along 3.2 km of Blackpool's shoreline and is

rapidly becoming a benchmark project for marine

defence authorities [21, 22]. STRUX 90/40 synthetic

macro fibre technology is being used to reinforce

much of the concrete that is being installed along

the length of Blackpool's promenade, as seen in

Figure 15. It is probably the first time in the world

that this technology has been employed for major

precast concrete elements in coastal defences.

The omission of steel reinforcement cages has

proved to be a significant benefit in terms of cost

57

Figure 13: Final passes using a power trowel to achieve near fibre-free concretesurfaces.

Figure 14: Finishing of exterior fibre reinforced concrete slabs.

58

and time. Using STRUX 90/40 also eliminated the

logistical issues of transporting and storing

tonnes of steel fabric on site. Pouring the

concrete into precast moulds off site, without the

problems surrounding the placement of steel

cages, has also speeded up that side of the

operation.

Another application for STRUX 90/40 is to use

it in composite elevated steel deck flooring. In the

USA and the UK, large-scale fire tests of fully

loaded composite steel decking with fibre

reinforced concrete were conducted

demonstrating that STRUX 90/40 can be used as

an alternative to conventional steel fabric

reinforcement in such applications. In the UK,

Richard Lees Steel Decking incorporated STRUX

90/40 in some of their composite steel decking

designs. The benefits for using STRUX fibre

reinforcement instead of steel fabric are obvious:

since no mesh needs to be stored on site, lifted

up several floors and placed onto the deck, the

construction time and labour savings are

significant.

It should be noted at this point that advice

should be sought whether the local building

codes allow the use of fibres before using fibre

reinforced concrete.

CONCLUSIONSA new generation of high performance

synthetic macro fibres are being increasingly used

in the construction market, primarily to replace

steel fabric reinforcement and steel fibre

reinforcement in concrete slabs-on-ground.

Extensive testing and field experience have shown

that STRUX 90/40 can be used in lieu of steel

fabric or steel fibres in most cases. In addition to

the control of plastic shrinkage, drying shrinkage

and thermal cracks, the flexural capacity of slabs-

on-ground is increased by adding STRUX 90/40 to

the concrete allowing either to increase the

allowable loading of the slab or to decrease the

slab thickness while maintaining the same flexural

capacity of the slab. STRUX 90/40 can be easily

added to the concrete, it disperses well in

concrete, and the fibre reinforced concrete can be

placed and finished with conventional methods.

Since STRUX 90/40 does not corrode and does

not provide any safety hazard, this technology is

an innovative solution to problems related to

concrete steel reinforcement in general.

REFERENCES

1. BECKETT, D. and HUMPHREYS, J.,‘Comparative Tests on Plain, FabricReinforced and Steel Fibre ReinforcedConcrete Ground Slabs’, Report No. TP/B/1,Thames Polytechnic School of CivilEngineering, Dartford, 1989

2. BECKETT, D., ‘Comparative Tests on Plain,Fabric Reinforced and Steel Fibre ReinforcedConcrete Ground Slabs’, Concrete, 1990,Vol. 24, No. 3, pp. 43-45

3. BECKETT, D., Van De WOESTYNE, T, andCallens, S., ‘Corner and Edge Loading onGround Floors Reinforced with Steel Fibers’,Concrete, 1999, Vol. 33, No. 3, pp. 22-24

4. BECKETT, D., ‘Concrete Ground Slabs: AnAppraisal of Corner Loading’, Concrete,2000, Vol. 34, No. 8, 3 pp.

5. FALKNER, H. AND TEUTSCH, M.,‘Comparative Investigations of Plain andSteel Fibre Reinforced Industrial GroundSlabs’, Institut Für Baustoffe, Massivbau undBrandschutz, Technical University ofBrunswick, Germany, 1993, No. 102, 70 pp.

Figure 15: STRUX 90/40 reinforced sea defence wall along Blackpool’s shoreline.

6. FALKNER, H., HUANG, Z., AND TEUTSCH,M., ‘Comparative Study of Plain and SteelFibre Reinforced Concrete Ground Slabs’,Concrete International, 1995, Vol. 17, No.1, pp. 45-51

7. NORDSTRÖM, E., ‘Steel fibre Corrosion inCracks, Durability of Sprayed Concrete’,Ph.D thesis, Luleå University of Technology,Sweden, 2000

8. BERNARD, S., ‘Durability of cracked fibrereinforced shotcrete’, Shotcrete – MoreEngineering Developments, Bernard (editor),Taylor and Francis Group, 2004, pp. 59-66

9. GRANJU, J.-L., BALOUCH, S. U., ‘Corrosionof steel fibre reinforced concrete from thecracks’, Cement and Concrete Research,2005, Vol. 35, pp. 572-577

10. American Society for Testing and Materials,ASTM C 1609-06, ‘Standard Test Methodfor Flexural Performance of Fiber-ReinforcedConcrete (Using Beam With Third-PointLoading)’, Vol. 4.02, 2006

11. Japan Society of Civil Engineers, JSCE-SF4,‘Methods of Tests for Flexural Strength andFlexural Toughness of Steel Fiber ReinforcedConcrete’, Concrete Library International,1984, No. 3, Part III-2, pp. 58-61

12. The Concrete Society, ‘Concrete IndustrialGround Floors – A Guide to Design andConstruction’, Technical Report 34, ThirdEdition, UK, 2003

13. Deutscher Beton-Verein E.V. (DBV),‘Merkblatt Stahlfaserbeton’, 2001 (inGerman)

14. Deutsches Institut für Bautechnik (DIBt),“Zulassungsgrundsätze Faserprodukte alsBetonzusatzstoff“, Schriften des DeutschenInstituts für Bautechnik, Reihe B, Heft 18,2004 (in German)

15. ROESLER, J. R., LANGE, D. A., ALTOUBAT, S.A., RIEDER, K.-A., and ULREICH, G.,“Fracture behavior of plain and fiber-reinforced concrete slabs under monotonicloading,” ASCE Journal of Materials in CivilEngineering, 2004, Vol. 16, Sept./Oct., pp.452-460

16. ROESLER, J. R., ALTOUBAT, S. A., LANGE, D.A., RIEDER, K.-A., and ULREICH, G., “Effectof synthetic fibers on structural behavior ofconcrete slabs on ground“, ACI MaterialsJournal, 2006, Vol. 103, No. 1, pp. 3-10

17. ALTOUBAT, S. A., ROESLER, J. R., andRIEDER, K.-A., ‘Flexural capacity of syntheticfiber reinforced concrete slabs on groundbased on beam toughness results’, RilemProceedings PRO 39, Fibre-ReinforcedConcretes, BEFIB 2004, 6th RILEMSymposium on Fibre-Reinforced Concretes(FRC), Vol. 2, pp. 1063-1072

18. The Concrete Society, ‘Concrete IndustrialGround Floors – A Guide to Design andConstruction’, Technical Report 34, ThirdEdition, UK, 2003

19. BERNARD, S., ‘Creep of cracked fibrereinforced shotcrete panels’, Shotcrete –More Engineering Developments, Bernard(editor), Taylor and Francis Group, 2004, pp.47-57

20. MACKAY J., AND TROTTIER, J.-F., ‘Post-crack behavior of steel and synthetic FRCunder flexural loading’, Shotcrete – MoreEngineering Developments, BERNARD(editor), Taylor and Francis Group, 2004, pp.183-192

21. ‘Defender of the Lights, Blackpool’s newseafront’, New Civil Engineer (NCE) NewConcrete Engineering magazine, 2006,Sept., pp. 10-12

22. PERRY, B., ‘Synthetic macro fibres storm tothe front of coastal defence innovation’,Concrete magazine, 2006, Nov., pp. 72-73

59

60

Alan Poole began his career as

a geologist studying at

Nottingham and Oxford

Universities. He became a

lecturer, first at Swansea and

then at London University.

Since 1975 he has also worked as a consultant

specialising in aggregates, concrete and

armourstone and has particular research interests

in alkali-aggregate reaction and deterioration

mechanisms in concrete. He is involved in various

national and international technical committees,

and in working parties producing technical

reports. He is co-author of several technical books

on aggregates, armourstone, concrete and

applied petrography.

ABSTRACTSoon after 1856 when the polarizing

microscope was first developed the polarizing

petrological microscope was being used for the

study of cement mineralogy. Developments of

resins after the Second World War have made the

preparation of thin transparent slices of concrete

and related materials possible so that they are

suitable for examination under a microscope.

Consequently, petrological investigations have

been extended to cements, mortars, renders,

plasters and a wide range of related construction

materials as a matter of routine. These

procedures have been applied to problems of

specification compliance, materials identification

and quality, the quality of workmanship and

investigation of deterioration and failure in

concrete and other materials. A wide range of

petrological techniques from simple tests and

specific staining to more elaborate techniques

such as x-ray diffraction and scanning electron

microscopy are available to assist the

petrographer. Examples illustrating some of the

techniques applied to Portland cement clinkers

and to hardened concrete are presented.

KEYWORDSPetrology, Polarizing microscope, Cement

clinker, Portland cement concrete, Alkali-

aggregate reaction (AAR,ASR), Sulphate attack,

Ettringite, Pulverised fuel ash (PFA), Ground

granulated blastfurnace slag (ggbs),

Water/cement ratio, Petrographic techniques,

Petrographic thin sections.

INTRODUCTIONThe polarizing petrographic microscope,

though it is today a commonplace, economic and

routine method used for the examination of

cements, aggregates, hardened concrete, renders,

mortars and a wide range of related materials, is

a relatively recent addition to the investigative

techniques available to the civil engineer, the

specifier, architect and others involved in using

materials in the construction industry.

As with the development and use of cement

and concrete, the microscope also has a long

history. The magnifying convex lens was known

to the ancients, but the development of the two

lens compound microscope in 1621 attributed to

Cornelius Drebbel in London, perhaps marked the

beginning of microscopical examination of

materials. A landmark in its history was the

publication of ‘Micrographica’ by Robert Hook in

1665. However, it was not until 1856, after the

development of the achromatic lens by Van Deijl

in 1824, that a polarizing microscope was first

described by a mineralogist, S. Highley, in 1856.

The particular value of the polarizing microscope

arises from the property inherent in many

transparent minerals to resolve the plane of

polarized light causing interference colour effects

which are very useful for mineral identification.

THE EXAMINATION OF CEMENTS AND CONCRETE

Initially the polarizing microscope was used for

identification of crystalline chemicals, but by 1880

this type of microscope was in routine use for

mineralogical identification. Early investigators,

though hampered by poor resolution, soon

established the value of the polarizing microscope

for investigating artificial materials such as

cement clinkers and concrete. The first reported

use of the microscope for studying Portland

cement was by Le Chatelier [1] in 1882, a French

chemist interested in metallurgy and cements. He

used reflected light from a cut and polished

surface to identify Alite (tricalcium silicate) as the

principal phase in cement clinker. Later,

61

CONCRETE UNDER A PETROGRAPHIC MICROSCOPE

Alan Poole, BSc, DPhil, CGeol, EurGeol, FGS, MAE

Consultant in Geomaterials, Petrology & Engineering Geology

62

Törnebohm [2] (1897) successfully identified Alite,

Belite (dicalcium silicate), Celite (tetracalcium

alumino ferrite) together with a glassy residual

phase in clinker. More recent improvements in

polishing and etching techniques allowed Tavasci[3] in 1934 to be the first to identify and quantify

the proportions of these phases in polished

sections of clinkers. Figure 1 shows a modern

polished and etched specimen of clinker (the

colours obtained by the HF etching process assist

identification and are indicated on the left hand

side of the figure).

N. C. Johnson [4], as long ago as 1915, realized

the value of the polarizing microscope for the

investigation of concrete and wrote a series of

articles for the journal ‘Concrete and Concrete

Construction’ which was the forerunner of the

modern journal ‘Concrete’ published by the

Concrete Society. The specimen preparation

techniques were limited at that time so he was

only able to examine polished slabs of concrete

which he viewed through the microscope in

reflected light. He was principally concerned with

aggregate type identification, mix proportions

and the compaction of the concrete. Though

limited by the techniques available his papers

demonstrated the value of the microscope for

assessing mix design and concrete quality.

In more recent times improvements in lens

design have improved resolution dramatically, so

that today crystalline materials only a few microns

across can be identified readily. However, the

study of hydrated cements in the matrix of a

concrete remained difficult because of the

problem of preparing a thin transparent section

of concrete or mortar which had to be no thicker

than 30 microns. After the Second World War

various forms of cold-setting epoxy resins were

developed and these were used to impregnate

and so stabilize a thin layer of concrete thus

allowing the friable matrix and aggregate

particles to be firmly held together so that a thin

section of concrete perhaps 25 microns thick

could be manufactured. Large size thin sections

of concrete could be produced such as is shown

in Figure 2 and these enabled the microscope to

be used routinely to provide a wide range of

detailed information about a concrete or mortar

specimen.

Using the petrographic microscope and thin

sections of this type is now a routine method of

obtaining information about mix proportions,

aggregate types, cement type, admixtures if

included, water/cement ratio, hydration state and

concrete quality. It also enables the petrographer

Figure 1: Polished and HF etched Portland cement clinker in reflected light.

Blue: C2S

Brown: C3S

White: Ferrite

Grey: C3A

Figure 2: Large Format Jig and 150x100mm Thin Section.

to identify the type and causative mechanisms of

any deterioration that may be present.

PETROGRAPHIC STANDARDS AND CODES OF PRACTICE

There are relatively few national standards

offering procedures to assist the petrographer.

One is BS 812 part 104 (1994) which is a

thorough statistical methodology for the

evaluation of aggregates. ASTM has a similar

standard in C295 (1998). Worldwide, these

standards are the most important ones dealing

with the petrographic examination of aggregates.

However, standard methods for the petrographic

evaluation of concrete, mortars, renders and

plasters are not available, although BS 4551 part

2 (1998) does offer some guidance on testing

mortars, screeds and plasters. Table 1 gives a

fuller list of the standards that are relevant to

aggregates and cementitious materials.

This lack of modern standards, particularly for

the examination of hardened concrete, mortars

and related materials has been recognised and is

currently being addressed by the Applied

Petrography Group of the Geological Society. This

group of scientists are themselves actively

involved in undertaking petrographic

examinations, or in using the results of

petrographic reports to evaluate specification,

workmanship, quality, durability and the reasons

for deterioration or failure in all kinds of

cementitious materials. Consequently, they are

aware of the limitations of current standard

procedures. They are now in the process of

reviewing existing standards with a view to

recommending improvements and changes that

would update and improve their practical

usefulness for a wider range of professionals

working in the construction industry. They are

also well advanced with developing codes of

practice which will aim to standardise procedures

for the petrographic examination of hardened

concrete and for the examination of mortars,

plasters, renders and similar materials. Both these

codes are based on the

methods that are already in use in materials

testing laboratories in the UK. It is hoped that the

drafts of these codes of practice will be available

for comment in 2007.

A common and necessary requirement in most

of these standard procedure methods is that they

shall be carried out by an accredited, trained or

‘competent petrographer’. Although increasingly,

both in the UK and Europe there is a requirement

that an operative should be appropriately trained

and accredited for many areas of work in

laboratories, there is no scheme currently

available for the accreditation of petrographers

either in the UK or Europe. The Department of

Transportation, Ontario, Canada is one of the few

63

Standard Designation Date Title of the standard Comment

BS 812 Part 104 1994 Testing aggregates. Method for qualitative and Statistically

quantitative petrographic examination of aggregates rigorous

BS 7943 1999 Guide to the interpretation of petrological limited to

examinations for alkali-silica reactivity ASR

ASTM C295 2003 Standard practice for petrograhic examination of Comparable

aggregates for concrete with BS 812

BS EN 932-3 1997 Tests for general properties of aggregates, procedure Limited

and terminology for simplified petrographic description

BS 4551 Part 2 1998 Methods of testing mortars, screeds and plaster Petrography

is limited

CSA A 23.2-15A 2003 Guide for the petrographic examination of aggregates Ontario,

Draft No 6.1

RILEM/TC-ARP/01/03 2001 Petrographic method, AAR 1 Limited to

AAR

ASTM C457 1990 Standard test method for microscopical determination Air voids

of parameters of the air void system in hardened concrete only

Table 1: Standards for Petrogaphic Examination of Aggregates and CementiousMaterials.

64

organisations currently exploring procedures for

the accreditation of petrographers involved in the

examination of aggregates and already has a

draft procedure in place.

The Applied Petrography Group are also

concerned that the highest standard of

petrographic reporting is maintained in the UK,

and they are aware that the quality of a report

will depend, in large measure, on the skill and

experience of the petrographer undertaking the

work. However, the cost of setting up and

maintaining a scheme of accreditation for

petrographers is a cause for concern since it is

unlikely to be met by the industry itself.

Nevertheless, the Group is exploring the

possibility of setting up a register of ‘competent

petrographers’ as a particular specialism within a

wider professional register. This work is ongoing

at the present time and it is hoped that some

means of ensuring the quality of the

petrographer and of a petrographic investigation

will be in place by the end of 2007.

PETROGRAPHIC EVALUATION OF CONCRETE AGGREGATES

The engineer, the architect, the contractor and

the client all require that the materials used

during construction meet the specifications and

are ‘fit for purpose’. This holds true for the

concrete used in a structure and for the materials

making up the concrete. Approximately 75% of

the volume of a concrete is composed of

aggregate. Whether the aggregate is coarse or

fine, a crushed stone, gravel, is recycled, or is

artificial, a simple petrographic examination is

undoubtedly the best way of ensuring that it

meets the quality requirements for its end use.

Provided that the sample selected is

representative of the aggregate to be used, the

petrographer will be able to identify the materials

it contains and provide a sensible appraisal of

other likely properties, such as relative density,

compressive strength, water absorption, particle

shape chacteristics and grading, without further

testing, though additional tests may be required

to comply with specification or contractual

requirements. This information can be obtained

from either a representative sample of aggregate,

or a slice of existing concrete (Poole, 2006) [5].

Figure 3 shows a cut and ground slab specimen

of concrete with the aggregates exposed. The

surface has been stained with acidified Alizarin

Red S, which stains the limestone pink (dark in

this photograph) while the dolomite, which

makes up the bulk of the coarse aggregate in this

slab, remains unstained (light). It is also clear that

the coarse aggregate is a crushed rock, is well

graded and that it tends to be ‘flaky’ in shape.

When a loose coarse aggregate sample is

being examined, the petrographer will examine

the particles directly, perhaps using a binocular

microscope or simple lens to examine individual

particles. However, when examining a fine

aggregate sample with a maximum particle size

of 4 or 5 mm, identification of individual grains is

not always easy. In such cases the petrographer

may mix a representative portion of the

aggregate with epoxy resin and cast it as a small

block. When the resin has set, the block may be

cut and lapped to make a transparent thin

section that allows individual grains to be

identified using a petrological microscope. In this

way, the mineralogy of the fine aggregate and

the proportions of the different mineral species

present can be established.

Petrological examination of loose aggregates

or of aggregate in a mortar or concrete is a

particularly effective method for checking

whether the aggregate contains even very small

amounts of contaminants or of deleterious or

unwanted materials such as soft or friable

particles, clay, coal, pyrite or other materials that

can cause problems when incorporated into a

concrete. Consequently it can provide a most

effective quality control method for concrete

aggregates.

PETROGRAPHIC EXAMINATION OF CONCRETE

As has already been noted no satisfactory

standard methods are currently available for the

petrographic examination of concretes, mortars

or related materials. However, this technique can

provide invaluable information over a wide range

Figure 3: Cut Slab of Concrete ShowingDolomite and Limestone CoarseAggregate.

of applications; aggregate particles can be readily

identified in a thin section of concrete, but

perhaps more importantly the nature of the

cement in the concrete matrix and its state of

hydration can be assessed by such examination.

It is common knowledge that although a

concrete may have set and hardened, some of

the cement grains in the cementitious matrix will

still not be fully hydrated. Old concrete, even

Roman concrete, will retain some un-hydrated

grains of the cement. Though such grains are

small and may be surrounded by rims of

hydration products they are usually sufficient to

enable the petrographer to identify the type of

cement used in the concrete. The hydration

reactions which take place as the cement in a

concrete or mortar sets and hardens produce

calcium silicate gels, small amounts of calcium

aluminate gels and calcium hydroxide. These can

be identified in a thin section and their sizes and

disposition can provide valuable information

about the original water/cement ratio and

indirectly the strength, quality and age of the

particular concrete (see St. John et al. 1998) [6].

Figure 4 shows two unusually large un-hydrated

cement clinker grains, small calcium hydroxide

crystals (portlandite) and hydration rims around

some of the smaller clinker grains.

The determination of the water/cement ratio is

in part related to the size of the calcium

hydroxide crystals present in the cement matrix

but is also related to the porosity of the gel

structure of the cementitious matrix. In most

cases, the larger the hydroxide crystals are, and

the more open and micro-porous the texture of

the matrix, the higher the water/cement ratio. To

assist this determination, the epoxy resin used is

stained with a fluorescent dye (Fluorescein or

Hudson Yellow). The thin section is then viewed

under ultraviolet illumination and an estimate of

the water/cement ratio can then be obtained

from the intensity of the illumination compared

against a series of reference specimens of known

water/cement ratio. The fluorescent dye also

highlights cracks and microcracks present in a

concrete. The crack patterns are often diagnostic

of their cause, whether it is merely natural drying

shrinkage or a more serious problem of

deterioration or failure. A review of non-structural

cracking in concrete is given in the Concrete

Society Technical Report 22, (1982) [7].

Mineral additions are commonly used with

cement in modern concretes since they can

significantly modify the properties of the concrete

and its long term performance. Most additions

are readily identified from an examination of a

thin section of the concrete. Common additions

such as pfa and ggbs are readily identified by the

characteristic shapes of their particles. Admixtures

such as air entraining agents or plasticizers are

more readily identified by their effects on the

microfabric of the concrete matrix. Similarly the

various types of Portland cements and other

cementitious materials such as calcium aluminate

cement or plasters are readily recognised from

their characteristic mineral composition and

hydration products, (St. John et al. 1998) [6].

Careful petrographic examination of the small

scale features in the cement matrix of a concrete

or mortar and the type, size and shapes of the

crystalline components can provide a wealth of

information about the past history of the

particular concrete from the original conditions

that pertained during mixing and placing, to the

conditions the concrete structure has been

exposed to over time.

65

Figure 4: Thin Section View of Part of the Portland Cement Matrix in a Concrete.

Width of field ofview is 0.3 mm

Thin Section ofthe cement matrix

Clinker grains

Crystals of calciumhydroxide

66

PETROGRAPHIC INVESTIGATIONOF CONCRETE DETERIORATION

Petrographic laboratories find that some of the

most common requests for the investigation of

samples of concrete are concerned with

deterioration or failure of the concrete in some

way. The application of petrography to this type

of ‘forensic’ investigation is perhaps its most

appropriate use. The skilled petrographer can

provide definitive information both rapidly and

economically; information that may not be

obtainable by any other means.

Aside from checking uncertainties relating to

materials being within specification, or the need

to determine what materials were used in an

existing structure that is now in need of repair or

replacement, a wide range of deterioration

mechanisms can be investigated. These range

from external problems such as salt ingress,

chemical attack, fire damage, sulphate attack and

frost damage to internal materials problems such

as the incorporation of excessive additions or

admixtures, corrosion of reinforcing steel and the

alkali-aggregate reactions.

At the present time the alkali-silica reaction in

concrete is a cause for concern and is used here

as an illustration of the petrographic approach. In

reality, its occurrence in the UK is rare. However,

if you are the unfortunate owner of a structure

which exhibits this particular problem, its severity

and potential for further damage are important

issues.

There are in fact two types of alkali-aggregate

reaction in concrete. The most common involves

a reaction between alkalis (usually derived from

the cement used) and non-crystalline or poorly

crystalline silica particles in the concrete

aggregate (ASR). Such material in the aggregate

can be recognised by the petographer. They may

be an opaline silica cement holding sedimentary

grains together in an aggregate, some cherty

materials, re-cystallised margins to the silicious

components in a metamorphosed rock, the

silicious glassy matrix within an igneous rock or a

variety of other silicious components which are

poorly crystalline or have crystal lattice defects

and a large surface area. Moisture is important in

facilitating the reaction which typically appears to

be worst on the weather faces of structures.

The effects of the reaction normally take eight

to twelve years to become apparent with cracking

usually being the first surface expression, but

misalignment of structural elements and surface

spalls (‘pop outs’) are other indicators. On

unconstrained concrete elements surface cracking

develops as a seemingly random network of

cracks (map cracking), but on constrained or

loaded elements, cracks tend to develop

perpendicular to the stress loading. An example

of the cracking due to ASR together with an

attempt to constrain the cracks is shown in Figure

5. This shows one of the central support columns

of the Voss Farm Bridge, Devon. The cause of the

ASR and the associated cracking was identified by

examination of thin sections and was due to a

small percentage of chert particles that were

present in the fine fraction of the aggregate and

a high alkali cement. Although a repair was first

attempted by strapping the columns, they were

eventually removed and replaced.

The alkali-silica reaction product is an alkali-

silica gel. This can absorb water, swell and exert

pressures sufficient to disrupt and crack the

concrete. The confirmation of this deleterious

reaction is usually accomplished by a petrographic

examination of thin sections of the affected

concrete. The examination usually enables the

reaction source and the gel to be identified, as

was the case with Voss Farm Bridge.

The second type of alkali-aggregate reaction is

alkali-carbonate reaction (ACR). This is

comparatively rare and although some dramatic

examples are recorded in Canada, there are no

clearly reported examples of it in the UK at the

present time. The reaction again involves alkalis

Figure 5: Part of the Voss Farm BridgeColumn, Showing Vertical Cracks andStrapping.

from the cement used which in this case react

with certain types of carbonate aggregate. The

carbonate aggregate is typically fine grained,

partly dolomitic and contains some clay as an

impurity. The detail of this reaction is still not fully

understood in spite of a considerable research

effort over many years. Moisture is again

necessary for the reaction to proceed but no gel

reaction product is involved, instead there

appears to be a process of de-dolomitization

which produces rims round the margins of

aggregate particles. The result of this reaction is

to produce a rapid and significant expansion of

the concrete with the effects beginning to appear

after a period of months from the time of

casting.

Both these types of reaction involve interaction

between aggregate particle constituents and

alkalis present in the pore fluids of the

cementitious binder of the concrete.

Consequently, these reactions will take place

within the body of a concrete structure and are

not confined to the surface zone. Therefore,

when such deleterious reactions do develop, is

very difficult or impossible to stop them

progressing, though some recent research

techniques involving electrolysis and chemical

treatments appear to be at least partially

successful.

These types of concrete deterioration and

many others such as delayed ettringite formation

and carbonation are most appropriately

investigated by using the petrographic

microscope. The Applied Petrography Group is

well aware of the unique value of this technique

and hope to increase its general acceptance

within the construction industry as an economic

investigative tool. To ensure its quality and value,

they are actively working to improve and develop

appropriate modern standard methods and codes

of practice for use by petrographic investigators

and are assembling a register of fully qualified

petrographers within the UK.

CONCLUSIONSThe petrographic microscope in the hands of a

competent petrographer is a powerful, economic

and perhaps unique tool for the investigation of

building materials. In particular, its application to

hardened concrete and similar materials can

provide information that is not obtainable by

other means.

It may be used in general applications such as:

• Quality control

• Materials compliance with specification

• Mix design

• Aggregate identification, types and

proportions

• Water/cement ratio determination

• Cement type and admixture identification

• Determination of concrete age.

Investigation of particular problems of

concrete deterioration, their cause and the

mechanisms involved might include:

• Carbonation rates

• Salt ingress

• Chemical attack

• Frost damage

• Sulphate attack

• Alkali aggregate reactions

• Corrosion of steel reinforcement.

The Applied Petrography Group are working

towards making the petrographic investigation of

concrete the accepted and reliable first step in

the technical evaluation of such materials.

REFERENCES

1. LE CHATELIER, H. Comptes rendueshebdomandaires des secances del’academie des sciences vol. 94 p13. 1882

2. TÖRNEBOHM, A. E. Ueber die petrographiedes Portland cements Stockholm: VerinSkandinavischer Portland CementFabrikanten. 1897

3. TAVASCI, B. Chim. e Industr. vol. 21. 1939

4. JOHNSON, N.C. The microscope in thestudy and investigation of concrete.Concrete and Concrete Construction p288.1915

5. POOLE, A. A petrographic solution toconcrete problems p34 Concrete. 2006

6. St. JOHN, D. A., POOLE A. B. and SIMS I.Concrete petrography, a handbook ofinvestigative techniques. Elsevier, 1998

7. CONCRETE SOCIETY. Non-structural cracksin concrete. Technical Report 22, 1982

67

68

Nick Buenfeld is a civil engineer

and concrete technologist

specialising in the durability of

concrete structures. Since

2000 he has been Professor of

Concrete Structures at Imperial

College London. He established and heads the

Concrete Durability Group, a multi-disciplinary

group of scientists and engineers aiming to

advance understanding of deterioration processes

and so develop more effective methods of design,

assessment and repair of concrete structures. He

has attracted research grants to the value of over

£4m and has authored/co-authored around 150

publications in refereed journals and conference

proceedings. He served for three two-year terms

on the ICT Council and in 2004 presented The

Seventh Sir Frederick Lea Memorial Lecture to the

Institute. He has been a member of many

national and international technical committees

producing guidance documents for industry and

is a Member of the Editorial Board of Magazine

of Concrete Research. He undertakes

consultancy assignments linked to his research

interests and has provided guidance to the

designers or constructors of many major projects

including the Jubilee Line Extension, the new Los

Angeles Cathedral and the Great Man-Made

River Project in Libya.

ABSTRACTImperial College and CIRIA have recently

investigated the state-of-the-art of automated

and intelligent monitoring of concrete structures

and have produced a Roadmap for the future

development of this field for the UK Department

of Trade and Industry (DTI). This paper discusses

the role of automated monitoring in the

management of concrete structures and describes

the main outputs of the DTI project including a

vision of how monitoring technology and its

application to concrete structures will develop in

the future.

KEYWORDSConcrete, Concrete structures, Automated

monitoring, Intelligent monitoring, Sensors, fibre

optics, MEMS, Wireless, Durability, Degradation,

Deterioration, Corrosion, Life prediction.

INTRODUCTIONA large proportion of the infrastructure of the

developed world is built in concrete and is

required to remain in service for at least 50 years.

Major bridges, dams, heritage buildings and

nuclear storage facilities are often expected to

have lives well in excess of 100 years. With

appropriate design, selection of materials and

construction practice it is possible to produce

concrete structures that will be adequately

durable in most exposure environments.

However, concrete structures do deteriorate and

the complex chemistry of cement, the use of steel

reinforcement and the variety of exposure

environments result in a number of, often

interacting, degradation processes. For most

structures deterioration is so slow that no early

intervention is required. However, deterioration

eventually becomes more rapid and repair or

replacement is required. This is costly in terms of

the materials and labour involved, the disruption

to users and other social costs. Unexpected

failures can have particularly severe financial,

environmental and safety implications.

As deterioration progresses, it becomes

increasingly expensive to rectify. Consequently it

is important to identify deterioration early,

determine its structural significance, and to

monitor the structure so that a timely

intervention is possible and so that more serious

problems may be avoided. Traditionally,

monitoring has been based on visual inspection.

However, increasingly sensors are being mounted

on structures and readings taken automatically to

provide real-time condition information. Where

Automated Monitoring of this kind is used to

explicitly provide information on current condition

and to assist in predicting the remaining life of a

component or structure it is referred to as

Intelligent Monitoring.

In 2004 the Department of Trade and Industry

(DTI) commissioned CIRIA and Imperial College to

investigate the state-of-the-art of the intelligent

monitoring of concrete structures and to produce

69

AUTOMATED MONITORING OF THE CONDITION

OF CONCRETE STRUCTURES

Professor Nick Buenfield, PhD, MSc, BSc(Hons), DIC, CEng, MICE, MICT

Imperial College, London

70

a Roadmap for the future development of this

field. This paper discusses the role of automated

monitoring in the management of concrete

structures and describes the main outputs of the

DTI project including a vision of how monitoring

technology and its application to concrete

structures will develop in the future. References

are provided to more detailed publications.

THE ROLE OF MONITORING IN THE MANAGEMENT OF CONCRETE STRUCTURES

Monitoring systems may be applied to new or

existing structures. For new structures the

emphasis will often be on giving an early warning

of significant deterioration taking place (for

example, the initiation of chloride-induced rebar

corrosion). This is of particular importance where

inspection is difficult or impossible. For existing

structures the emphasis will usually be on

monitoring the rate of an already active

deterioration process. The two cases, however,

may be treated with the same basic approach

once it is recognised that the common aim of

monitoring should be to reliably indicate current

condition and to enable estimation of the residual

life of the structure and/or the time until repairs

are required. Figure 1 shows the relative roles of

the various activities that may be utilised in

assessing the residual life of a concrete structure.

For each potentially sensitive area of a

structure it is usually necessary to carry out non-

destructive testing (NDT) on site, sampling and

laboratory testing to establish the current

condition. A preliminary extrapolation of time to

stages of further deterioration can be made

based on an appropriate model and experience.

At each stage of deterioration the benefits of

remedial action to slow deterioration or locally

repair, replace or reconstruct damaged elements

can be estimated.

At present, by far the most widely practised

form of monitoring is visual inspection. It is

during visual inspections that signs of

degradation, for example concrete cracking or

surface deposits, are most likely to be observed.

Even if an automated monitoring system is

installed, visual inspection should be carried out

wherever possible because visual inspection is

relatively inexpensive and allows a larger area of

concrete surface to be scanned. Some owners

carry out on-site NDT on their structures every 5

or 10 years in an attempt to monitor

deterioration. However, there are several major

benefits of automated monitoring in relation to

on-site testing, as outlined below.

1) The influence of climatic conditions can be

isolated. Many of the measurements of

interest are influenced by the climatic

conditions at, or shortly before, the time

that the measurement is taken. For

example, temperature influences concrete

resistivity and relative humidity influences

half-cell potential. Consequently, single

time readings can be misleading.

Automated monitoring allows

measurements to be made every day, or

indeed many times each day, so that such

influences may be determined and isolated

more effectively.

Figure 1: The role of monitoring in the management of concrete structures.

2) The need for access is reduced. Regular

measurements may be made even if the

structure, or component, is difficult to gain

access to. This may be due to geographic

location (e.g. an offshore structure), local

position (e.g. a buried foundation or the top

of a bridge pier), function (e.g. a nuclear

pressure vessel) or a requirement not to take

the structure out of service (e.g. a

motorway bridge deck or a pipeline).

3) Measurements may be made at the depth

of interest. On-site, NDT usually involves

taking measurements at the surface of the

concrete. In some cases this is too far from

the area of interest to provide reliable data.

With automated monitoring, sensors can be

embedded at the location of interest,

particularly if they are installed at the time

of construction. For example, this allows

corrosion sensors to be placed throughout

the concrete cover (e.g. see Figure 1) or

adjacent to the rebar of interest, even if this

is at the bottom of a deep pile or near the

centre of a thick concrete element. At

present the sensors to monitor most

properties are hardwired to the

instrumentation taking and recording the

measurement. This often limits the number

of sensors deployed. However, major

developments in wireless communications

and related sensor technology mean that a

range of wireless sensors for embedding in

concrete is likely to be available within the

next 5 years.

4) Initiation of deterioration is identified earlier.

Some deterioration processes take many

years to start (initiation phase) and then

lead to rapid deterioration (propagation

phase); for example, chloride-induced

corrosion arising from sea-water

penetration. In such cases it is usually

important to identify, as early as possible,

that the propagation phase has started so

that appropriate remedial measures may be

implemented. Automatic monitoring will

identify that the propagation phase has

started much earlier than on-site NDT

carried out every 5 or 10 years.

THE DTI/CIRIA PROJECT AND ITS OUTPUTS

In 2004 the National Measurement System

Directorate of the DTI commissioned CIRIA and

Imperial College to investigate the state-of-the-art

of the Intelligent monitoring of concrete

structures. This was driven by the belief that the

technology was available, but was not being

widely used. This work comprised a 30-month

contract to investigate both sensors and how

they are used and the available life prediction

models and to prepare a state-of-the-art report

including a Roadmap for the future development

of the technology in this area. The work was

guided by an Industrial Advisory Group made up

of representatives of organisations responsible for

owning and managing structures and those active

in developing and applying the technology. The

main outputs of this project are described below:

DTI Report 1: AutomatedMonitoring of the Deteriorationof Concrete Structures[1]

This report presents the state-of-the-art in

automated monitoring of the deterioration of

71

Figure 2: Anode-ladder system used to monitor corrosion front penetration and topredict time to initiation of rebar corrosion [2].

72

concrete structures. The report focuses on what is

currently practical (summarised in Table 1), but

also refers to methods under development.

Considerable detail is provided on the sensors

available to monitor structural change, reinforcing

steel corrosion, concrete moisture state,

temperature and chemistry and exposure

environment. The measurement principle,

advantages (in relation to competing techniques)

and limitations, the equipment used and potential

application areas are presented for each of the

main methods. The Appendices contain a mass

of useful information including datasheets for the

different generic types of sensor and monitoring

equipment (with a list of equipment

manufacturers/suppliers) and over 30 case studies

reporting on experience in applying monitoring

systems to various structures. Figure 2 is a

typical page from one of these case studies.

DTI Report 2: Service LifePrediction of Concrete Structuresbased on Automated Monitoring[2]

This report reviews a large number of models

for predicting the service life of concrete

structures and discusses which are most

appropriate for use with data derived from

automated monitoring. Models are empirical,

analytical or numerical in nature and it is this

characteristic that dictates the quantity and

availability of the required input parameters. An

overview is presented highlighting the main

differences between modelling approaches and

probabilistic techniques are presented and

discussed. The vast majority of service life models

presented in the literature and used by industry

have been developed independently of advances

made in the field of automated monitoring. In

many cases the role of intelligent monitoring is

restricted to providing only one or two of the

input variables and in a significant number of the

models reviewed, intelligent monitoring cannot

take any active role in the assessment process.

DTI Report 3: A Roadmap for theDevelopment of IntelligentMonitoring of ConcreteStructures[3]

A Roadmap is provides an informed view of

the future development of a particular field. It

may be used to identify bottlenecks and prioritise

future research funding. Developing a Roadmap

Table 1: Properties/behaviour that it would be beneficial to monitor automaticallyand what is currently feasible.

involves defining the current situation,

considering where we want to be in the future

(the Vision) and analysing what needs to be done

to realise this Vision. This report presents a

Roadmap for the future development of the field

of intelligent monitoring of concrete structures.

CIRIA Guide: IntelligentMonitoring of ConcreteStructures[4]

This CIRA Guide provides an entrée into the

subject of intelligent monitoring. It is based on

the DTI Reports and the full DTI Reports are

included on an attached CD.

73

Figure 3: A page from Case Study 22[1] (Jubilee Line Extension corrosion monitoring)courtesy of CAPCIS.

74

A VISION OF THE FUTUREIn developing the Roadmap, a Vision of the

future was required. This had to be a balance

between what is desirable and what is likely,

given that there are many factors involved

beyond the control of those involved with

concrete construction. This involved making

reasonable assumptions, especially about

technology development and incorporating the

requirements and opinions of stakeholders

including owners, consulting engineers and

contractors. This was informed by several

meetings and a workshop, involving the Industrial

Advisory Group, designed to capture these issues.

Ideally the Vision would be described at a specific

time, x years from now. It soon became clear

that the timing is largely dependent on factors

beyond the control of those with a direct interest

in monitoring and/or construction. These factors

include the development of communications

technologies, low power very long-life batteries

and emerging sensor technologies. Consequently

the focus was on identifying likely and desirable

trends, without being unrealistically specific about

their timing. Where it was necessary to be

specific about the Vision date, it was taken as

2020. The paragraphs below summarise key

aspects of this Vision.

In the future, Society is likely to demand that

the condition and risks for all significant

structures are certified, probably once some major

high profile structures have failed due to unseen

deterioration. There will be more pressure on

designers and constructors to produce

demonstrably durable structures and those

responsible for certification will require more

information about the changing condition of

structures. Much of this information will come

from automated monitoring.

Automated monitoring will be considered at

the design stage of most important structures. It

will be applied to structures where the

consequences of reaching a limit state are severe

and where access for visual inspection and testing

is difficult. This will include nuclear waste

repositories, major bridges, city tower blocks,

tunnels, pipelines and offshore platforms. In

many cases monitoring will be focussed on

particular potentially vulnerable details (for

example replaceable bearings) rather than the

whole structure. In some cases structures will be

designed to allow monitoring sensors to be

installed later in the life of the structure, thereby

avoiding sensors degrading or failing before

being useful.

Monitoring will be recognised as a means of

easing the specification of new materials (for

example, new concrete admixtures) with

potentially appropriate properties, but no track

record of real use.

Far more authoritative independent guidance

will be available concerning what is possible and

how to do it. Monitoring the built environment

will be a module on most undergraduate civil

engineering degree courses and specialised MSc

courses are likely to be offered. There will be a

greater number of specialist monitoring

consultants and contractors.

There will be few, if any, aspects of structural

change, rebar corrosion and moisture state that

cannot be measured now that will be measured

by 2020. However, improvements will have been

made to the methods available and their

implementation so that sensors will be more

easily and reliably installed, the measured results

will be more accurate with less drift and sensors

will be more durable. More independently

verified quantitative information will be available

concerning sensor resolution, accuracy, drift and

service life.

Figure 4: MEMS sensors. Left: Wireless tri-axial accelerometer combined with a datalogging transceiver (courtesy of MicroStrain). Right: Micro-sensor for measuringchloride, temperature, moisture and pH in concrete, under development (courtesy ofAdvanced Design Consulting, USA).

Fibre optic sensors will increasingly replace

conventional strain gauge sensors because of

their ability to measure changes at various points

along the fibre length, their immunity to

electromagnetic interference, freedom from drift,

high resolution, high shock resistance, and good

durability. A wider range of chemical sensors is

likely to be available for embedding in concrete

although the perceived need for chloride and

hydroxyl ion sensors is likely to have diminished

with the development of more reliable corrosion

front sensors. Many types of sensor will have

become smaller through new technological

developments. For example, MEMS (miniature

electromechanical system) sensors only a few mm

in size will be available containing an on-board

microprocessor providing intelligence capabilities.

Integrated, stand-alone, intelligent remote

monitoring systems will be commoditised. Wired

connections from sensors to computer will

generally have been replaced with wireless

communications and sensor networks. This will

be a major factor in easing implementation and

reducing the cost of monitoring. These

developments will facilitate the development of

Smart infrastructure with sensors collecting and

transmitting information about the state of the

infrastructure to a central computer and receiving

instructions to actuate devices adjusting the state

of the structure.

ACKNOWLEDGEMENTSThe Department of Trade and Industry and

CIRIA are thanked for allowing this paper to be

presented. The contributions of Dr Ron Davies

and Dr Ali Karimi of the Concrete Durability

Group, Imperial College and Alan Gilbertson (on

behalf of CIRIA) are gratefully acknowledged.

REFERENCES

1. DAVIES, R.D. and BUENFELD, N.R.Automated Monitoring of the Deteriorationof Concrete Structures, DTI, 2007.

2. KARIMI, A. and BUENFELD, N.R. Service LifePrediction of Concrete Structures based onAutomated Monitoring, DTI, 2007.

3. BUENFELD, N.R., DAVIES, R.D. and KARIMI,A. A Roadmap for the Development ofIntelligent Monitoring of ConcreteStructures, DTI, 2007.

4. BUENFELD, N.R., DAVIES, R.D., KARIMI, A.and GILBERTSON, A. Intelligent Monitoringof Concrete Structures, CIRIA, 2007.

75

76

77

Simon Austin is Professor of

Structural Engineering in the

Department of Civil and

Building Engineering at

Loughborough University, one

of the leading research centres

in the built environment. He is also the founder

director of Adept Management, a specialist

management consultancy. Prior to this he worked

for Scott Wilson Kirkpatrick & Partners and

Tarmac Construction.

He has undertaken industry-focused research

for over 25 years into design processes,

modelling, integrated working and management

techniques, information management, process re-

engineering, value management and structural

materials and their design. The latter includes the

building design process, behaviour and design of

structural elements and innovative sprayed and

cast concretes and mortars.

He is the author of over 200 publications and

holder of over 20 EPSRC funded research grants.

He is a member of various BSi and CEN

standards committees and a consultant member

of two trade associations. His research portfolio

has involved collaboration with many leading

companies and organizations in the construction

industry.

ABSTRACTConstruction has traditionally relied on

specifications and 2D drawings to convey material

properties, performance details and location

information. Advanced 3D solid modelling and

digital fabrication methods are growing in

construction. Iconic building design is driving the

industry towards a new era of the Building

Information Model (BIM) where a building is

modelled entirely using 3D solid CAD tools

containing all the required information for

construction. CNC machinery can utilise this

information to manufacture components enabling

highly bespoke and non-repeating components to

be cost competitive. Rapid Manufacturing

machines also use this information to build

components by selectively adding material rather

than the traditional subtractive or formative

processes. The BIM drives current machines for

the production of models for inspection or to

explore assembly issues. Recent developments are

scaling up these processes so that whole building

components can be built using a mega scale,

additive machine. This paper explores some of the

issues relating to the design of building

components and discusses issues on the

implementation of these process.

KEYWORDSFreeform construction, Rapid manufacturing,

Digital fabrication, Construction design.

INTRODUCTIONIn the manufacturing sector, automation using

industrial robots and machines that use direct

numerical control took hold in the 1960s. The

development of microprocessors delivered

computer numerical control in the 1970s and the

IT revolution in the 1980s brought computer

aided design software. In the 1990s advanced

parametric modelling was introduced and the

industry has enjoyed the development of the

integration of design and analysis tools and

machine control. All these technologies can be

found in construction[1,2,3]. The introduction of

CAD/CAM for the creation of large structural

components for freeform buildings is driving the

development of digital manufacturing

technologies for construction. However, cutting

edge designs for buildings are becoming

increasingly unrealisable using the current state-

of-the-art methods - new processes are required[4,5]. The control of material placement and

reducing the number and quantity of materials

will play a key role. The manufacturing sector is

turning to Rapid Manufacturing (RM) for

solutions, especially for the production of highly

personalised products [6]. The construction

industry is waking up to the potential that

automated additive technologies offer for solving

these problems. Ultimately, it is feasible that such

new processes will drive down the cost of existing

construction, while raising the bar of achievable

APPLYING FUTURE INDUSTRIALISED

PROCESSES TO CONSTRUCTION*

Professor Simon Austin, BSc, PhD, CEng,

Loughborough University

* This paper was co-authored by Richard Buswell, Alistair Gibb, Rupert Soar and Tony Thorpe.

78

construction design solutions. New technologies

are most likely to find niche applications initially

and will eventually filter down to the domestic

sector; exemplified by the Tunnelform system[7].

Over the last ten years there have been

attempts to selectively

bond sand and cement to create freeform

structures from traditional building materials[8].

RM has developed large mould making

processes[9] and the first viable largescale freeform

process for construction, Contour Crafting (CC),

has been demonstrated in the laboratory at the

University of Southern California[10]. Khoshnevis is

pushing for the commercialisation of this process

in the USA. It is capable of producing full scale,

freeform wall structures that would replace the

structural concrete block wall similar to that

found in UK house construction.

Contour Crafting, however, cannot take full

advantage of the extended functionality that can

be embedded within the wall structure if the

principles demonstrated by existing RM processes

are applied. Precise control of very small volumes

of build materials would allow the wall to be

constructed, ground up, with all the internal

pipework, conduits and channels in place,

removing structurally redundant material. The

implications of such an approach would lead to:

clever design solutions using geometric freedom;

a smaller number of build materials and a

reduction in the material resource required for the

construction process; simplify on-site operations

with a reduction in complex trade coordination;

force the resolution of interface issues, hence

reducing part count. Design would be complete

up front and this would mean that the structure

could be designed to be more easily disassembled

and recycled at the end of its life. In addition, the

acoustic, permeability and thermal characteristics

can be modified by ‘printing’ appropriate,

optimised geometry[11].

A concept domestic wall structure, designed

using Freeform Construction principles is depicted

in Figure 1.

New UK research at Loughborough University

is developing a process capable of delivering the

‘Wonderwall’ concept at full scale. This paper

gives details on additive processes and a

discussion of the differences between traditional

and Freeform Construction is given as

background information. The issues surrounding

design practice, design tools, and data and

information protocols are highlighted.

ADDITIVE TECHNOLOGIESThere are various names used to describe

essentially the same type of fabrication

technology; Additive Manufacturing, Rapid

Manufacturing, Rapid Prototyping and Solid

Freeform Fabrication. This method of making

physical components is delivered by many types

of processes; Thermojet, Selective Laser Sintering

(SLS), Stereolithography (SLA), 3D printing (3DP),

Fuse Deposition Modelling (FDM), are a few[12].

Typically each of the processes can use a range of

materials and all have advantages and

disadvantages, suiting them to particular tasks.

They all work ‘print’ 3D structures typically up to

500mm in the x, y and z directions. A design is

usually created using 3D CAD solid modelling. A

model is first tessellated in much the same way as

a Finite Element Analysis mesh is generated then

sliced into layers according the specific machine

parameters. Each slice is sent to a machine. The

machine builds the component by sequentially

creating and bonding each layer to its

predecessor to reproduce the 3D artefact.

Applications vary and can be found in the

literature and on the web sties of companies

offering Rapid Manufacturing services. A good

Figure 1: A concept application for Freeform Construction – 'WonderWall'.

79

source of further reading can be found at Castle

Island[13]. Many classification methods exist for

current RM processes. Application descriptions of

common RM processes are offered here for

discussion of the differences of these in a

construction context. Parallels are drawn between

these process, traditional construction methods

and both existing and conceptual Freeform

Construction technologies.

Comparison of additive processFigure 2 depicts the process of slicing the CAD

model and gives diagrams illustrating the

principal features of mentioned processes. Each

works on a layer-by-layer basis. The SLA and SLS

processes employ a laser; the latter cures a liquid

photo-sensitive resin, the former uses the laser to

melt a small area of powdered material that then

sets to form a solid. The surface finish and

fineness of detail is dependent on the material

properties and the width and intensity of the

laser. The laser ‘rasters’ across the build area with

a fine laser. To speed up the build process, both

processes outline the solid/liquid or powder

boundary on each layer with a fine laser beam

profile and then 'fill in' the area with a de-

Figure 2: Diagrams depicting several of the most commonly used Rapid prototypingprocesses.

80

focussed laser or open hatching strategy. As with

any process, control of operating parameters is

crucial for a successful build. SLS requires a

balance of laser intensity, traverse speed

penetration and time required to either melt or

sinter powder particles into a solid. These

parameters are similar with SLA but the phase

change mechanism is different. Even when these

parameters are tightly controlled, issues arise. For

example, the heat generated during the sintering

process leaves solidified components embedded

in a hot but loose powder ‘cake’. If broken open

too soon then thermal distortion will affect

dimensional tolerances. Table 1 compares the

various processes.

The ‘drop-on-demand’ processes of 3DP and

multi-jet deposition epitomise the idea of three

dimensional printing. All employ raster-based

deposition of either phase change building

materials or binder systems onto powders and

use standard inkjet printer technologies (both PZT

and bubble). Fused Deposition Modelling (FDM)

refers to a family of processes which extrude a

range of thermoplastic polymers to build up a

component in much the same way as squeezing a

tube of toothpaste would.

Rapid Manufacturing machines are designed

to build miniature, hand held and desktop sized

items. A site based construction process is

unlikely to use a laser-augmented approach

(SLA/SLS). In addition, the 'vat of material'

approach is unattractive, because of the

impractical issues associated with post-processing

what would be very large components(SLS/3DP).

This leaves processes that deposit material

through a deposition device (Thermojet/FDM). A

key point is that as you increase the build scale,

the volume flow of material will force the design

of a new process; it cannot simply be scaled up.

For build sizes in the order of 1 to 10 mm, the

build material can be deposited by a printer head

and still maintain a reasonable build speed. Using

the printer head to control deposition of a

curable liquid allows incredibly fine feature sizes,

Table 1 summarises the similarities and differences between the mentionedprocesses.

81

up to 600 dpi [14]. Between part sizes on the scale

of 10 to 100 mm it can be cost effective to use a

matrix material that has a larger particle size; the

ZCorp 3DP process uses powdered gypsum [15].

Instead of passing the build material through the

printer head, only a liquid binder is deposited

which binds the matrix material. The volume of

liquid passed through the head is a fraction of

the part volume and hence the build speed is

maintained. Larger processes (100 to 1000 mm)

cannot get enough binder through a printer head

and so only prints a curing agent onto a matrix

material pre-coated with an epoxy based

compound[16].

For conceptual larger scale parts, say 1000 to

10000 mm, a special (nonexistent) agent would

be required so that minute quantities could be

used to reduce the print volume flow.

Additive processes for construction

The top diagram in Figure 3 depicts the

Contour Crafting process and the process

features are in Table 1. Contour Crafting has

been tailored for the production of domestic

housing wall components. The intention is that a

large gantry system will be able to ‘print’ the

entire structure for a house. The process has been

demonstrated to produce large structures in the

order of 3 m long by 1 m high by 0.1 m wide.

This volume is many times that of the

conventional RM processes. In order to deliver

these volumes of material an extrusion and back

fill approach has been adopted: An inner and

outer 'skin' is extruded (~19 by 19 mm) and

forms a permanent shutter. The machine then

backfills with a bulk compound similar to

concrete. One of the key issues is how the build

material maintains its desired form once it is

deposited and while it is curing: Contour Crafting

uses thixotropic materials with rapid curing and

low shrinkage characteristics. The process avoids

post-processing by depositing the material using

similar principles as the Thermojet technique. So

far the process has not been developed to handle

overhanging sections, although the strategy for

creating openings for doors and windows will be

effected by robotic placement of lintels that the

Figure 3: Diagrams of existing and conceptual Freeform Construction processes.

82

deposition head can build off. The shutter

extrusion dimensions limit the feature size that

can be created and the process is given to

producing long, thin walls that can be curved

arbitrarily. The surface finish is trowelled as part

of the process and can achieve very high degrees

of smoothness. The practical wall system that the

Contour Crafted structure would be a part of,

would also need to be clad, insulated, finished

internally, have doors and windows fitted and

mechanical and electrical services, etc. added to

it. Currently the intent is to leave out sections of

the wall as it is ‘printed’ and post-fit services

modules that could be either accessible from the

outside (flush with the surface), or internal, being

built into the wall.

Contour Crafting represents the first

generation of Freeform Construction processes.

The next generation of technologies will be

capable of printing at variable resolutions. A

Multi-Resolution Deposition (MRD) device is

depicted at the bottom of Figure 3. MRD will

build objects at comparable scales and speeds to

Contour Crafting, but will be capable of fine

detailing that gives RM technologies their

strength. The likely specification and process

features will be:

• mineral based compounds (cost)

• selective deposition of material (minimising

post processing)

• feature size down to ~1 mm (control of

surface texture)

• variable deposition resolution (high speed

fill in)

• material shape holding (allow additional

layers while curing)

• high degree of self-supporting features

(minimises post processing)

• inclusion of internal voids and channels

(adding value through function)

• varying material properties through additives

(e.g. for moisture control)

• more freeform surfaces (greater design

freedom for free); and

• more reliable build time and precise

tolerances (machine control).

IMPLICATIONS FOR CONSTRUCTION DESIGN

The MRD Freeform Construction process has

been defined. This section discusses implications

for construction design and issues associated with

design process and information handling. A

conventional UK domestic dwelling wall is cited

here to highlight differences with the concept

wall depicted in Figure 1. This wall might

comprise (typically, from the inside out): 3 mm

coat of finishing plaster to create a hard smooth

finish; 12 mm render to remove imperfections

from the blockwork in preparation for the

finishing plaster; 100 mm load bearing concrete

block wall bedded using a sand/cement mortar;

50 mm cavity filled with an insulation eg, glass

fibre; 100 mm clay brick bedded in mortar, tied

to the internal leaf with steel ties. There are 8

materials listed here and typically 2 trades

required to erect and finish the wall; the

bricklayer and the plasterer. A complete ‘fully

functioning’ wall could include timber and glass

for the doors and windows, employing a

combination of glaziers and joiners; a joiner

would also finish the wall with timber skirting

boards and sills; plastic conduit for electrical

wiring, usually embedded beneath the plaster,

requiring an electrician and labourer; pipework

and panelled heating devices, usually surface

fixed by a plumber. The design now uses nearer

13 materials and 7 trades. In addition, damp-

proofing introduces another material and

openings in the structure require lintels and

usually some sort of temporary former, made in

timber, to guide the brickwork. Scaffold is needed

to elevate the site operatives to access higher

sections of the wall safely. The design becomes a

complicated series of interface resolution issues:

Damp proofing location and draining round

windows; closing cavities; lintel placement;

cutting bricks for openings and defining space;

weatherproofing round windows and doors. The

MRD process would aim to handle many of these

issues with a single operation, utilising reduced

numbers of materials. In the example given, this

process would replace the original 13 materials

with 5; the primary build material, glass, a

framing material, probably an insulating material

and some additive to the primary material for

moisture control. In addition, the thermal

performance could be enhanced[1] and material

resource could be minimised by simply not

printing structurally redundant sections; and there

could be more variation in design because it takes

no more effort or expense to print a curved

section that it does a straight one. The only costs

are; design time, machine setup and run time and

material consumption. It is also likely that self

supporting structures like arches will be employed

to form openings which would reduce the

83

requirement for post-processing. Glazing these

can be achieved using well established CAM/CAD

and CNC technologies. This affects the design of

space and form and would mean more client

choice and greater use of freeform surfaces.

Designing functionality into the interior of the

wall is the real benefit. Designers, architects and

engineers would be required to rethink how

performance can be achieved and enhanced

using solutions based on geometry; using a single

material to realise the design goals. The process

would need to be integrated and increased use of

automated optimisation to derive design solutions

would become more likely. In order to achieve

this, it is conceivable that CAD software tailored

for Freeform Construction design would design

rules. These constraints would ensure that the

designed structures could be successfully built

within the process operating parameters. A

process like MRD simplifies the elemental

operations to achieve a construction component

and limits material options. The key is that

functionality is not compromised, just realised in a

different way. By simplifying the elemental

operations of the construction process, building

design criteria and optimisation routines into such

software are realisable and therefore greater

design variety is afforded with a single process.

Implications for design processThe design processes for a part produced

using RM, MRD and traditional construction are

similar. Table 2 highlights this, comparing RM,

construction and MRD design process. All three

processes describe how to make a physical object

and the stages for all are similar. The actual

operations and the way in which information is

transferred are different. The real difference

between the traditional construction design

process and that used by MRD is how the

building information is gathered, stored and

utilised. Traditional approaches typically use an

‘over-the-wall’ approach between design experts.

The MRD approach will require simultaneous

design because the design of the material

placement will affect the integration of structure,

function and services while having to satisfy the

desired outer form. It is possible that the outside

form could be determined by the wall internal

structure and the space requirements, rather then

working the space design and services integration

around a given wall shape.

Implications for informationand ICT

The MRD process will require a digital

representation of the component to be built and

the assumption here is that this will be a 3D solid

CAD model. The control of RM machines often

uses a common interface Standard Triangulation

Language (STL) file format. STL describes a

faceted surface representation of the CAD model.

Each triangle has a normal associated with it that

indicates which face in 'inside' the component.

Simply, the original CAD model is a geometric

representation of a shape that defines what is

solid and what is not. The format only carries the

object information and is not capable of carrying

other information, such as how to build it. This

can be important because the layers created in

the vertical direction during a part build in RM

can result in non-uniform part material

properties, which can be undesirable. The debate

over standards continues in the manufacturing

sector and the suitability for MRD has yet to be

established. Likely issues include:

• size of data required to define large

structures

Rapid Manufacturing MRD Construction Traditional construction

Specification and brief Specification and brief Specification and brief

Concept and ideas Concept and ideas Concept and ideas

CAD model CAD model Design

STL conversion STL conversion Drawing production

STL testing (buildability) STL testing (buildability) Analysis of site programme

STL Slicing STL Slicing Temporary works

Fabrication Fabrication Build

Post processing Post processing Remove temporary works

Assembly with system 2nd & 3rd fix 2nd & 3rd fix

Table 2: Design process comparison between Rapid Manufacturing, FreeformConstruction and traditional construction.

84

• quality of representation of surfaces

• how to handle multiple materials

• the transition from 2D to solid 3D modelling

• the effort in design and analysis

• machine control

• build information

• distribution of build 'knowledge' to machine

• interfacing with existing design tools

• units and tolerances and repeatability.

CONCLUSIONSAutomating construction will deliver benefits

as has been demonstrated in the manufacturing

sector. There are two options, one of which is to

automate human processes. This approach is

flawed for all except very specific tasks because

encoding the complexity of handling materials

coupled with the highly complex decision making

process exhibited by craftsmen is difficult. The

second option is to simplify the elemental

operations controlled by the computer (Pegna,

1997). Many of these simple operations can be

carried out in such a way as to produce a very

complex product. This is the essence of Rapid

Manufacturing and why the process is so suited

to the construction automation issue.

The industry will have to rethink how

components are designed to maximise the

benefits that a process such as MRD can deliver.

The design process will be computer based which

is a goal the industry is already moving towards.

The MRD device is the focus of ongoing research

at Loughborough University.

REFERENCES

1. HOWE, S. 2000, Designing for automatedconstruction. In Automation inConstruction, vol. 9, 259-276.

2. KOLAREVIC, B., 2003, Architecture in theDigital Age: Design and Manufacturing,New York & London: Spon Press - Taylor &Francis Group.

3. SCHODEK, D., BECHTHOLD, M., GRIGGS,K., KAO, K. and STEINBERG, M., 2005,Digital Design and manufacturing:CAD/CAM Applications in Architecture andDesign, First edn, Wiley and Sons, Hoboken,New Jersey.

4. EGAN, J., 1998, Rethinking Construction,Department of the Environment, London.

5. European Construction TechnologyPlatform, 2005, Strategic Research Agendafor the European Construction Sector:Achieving a sustainable and competitiveconstruction sector by 2030, EuropeanConstruction Technology Platform,www.ectp.org.

6. Invisalign, 2006, Invisalign home page[Online]. Available:www.invisalign.com/generalapp/gb/en/[2006, 03/21] .

7. The Concrete Centre 2004, HighPerformance Buildings: Using Tunnel FormConcrete Construction, The ConcreteCentre, Camberley, Surrey, UK.

8. PEGNA, J., 1997, Exploratory investigationof solid freeform construction. InAutomation in Construction, vol. 5, no. 5,427-437.

9. The American Foundry Society, ModernCasting [Online]. Available:http://www.moderncasting.com/archive/ProductInnos/2005/Prod_099_01.asp [2006,03/21] .

10. KHOSHNEVIS, B., 2002, AutomatedConstruction by Contour Crafting: RelatedRobotics and Information Sciences. InAutomation in Construction Special Issue:The best of ISARC 2002, vol. 13, no. 1, 2-9.

11. BUSWELL, R.A., SOAR, R.C., THORPE, A.and GIBB, A.G.F., 2007, FreeformConstruction: Mega-scale Rapidmanufacturing for Construction. InAutomation in Construction, vol. 16, no. 2,222-229.

12. WOHLERS, T. 2004, Rapid Prototyping,Tooling & Manufacturing: State of theIndustry, Wohlers Associates, Colorado,USA.

13. Castle Island, 2006, 7/6/06-last update,Castle Island's Worldwide Guide toRapidPrototyping [Homepage of Castle IslandCo.], [Online]. Available:http://home.att.net/~castleisland/ [2006,21/6/06] .

14. Objet Geometries Ltd. 2006, Objet Homepage. Available: www.2objet.com/ [2006,06/23] .

15. Z Corp 2006, Z Corp Home Page. Available:www.zcorp.com/ [2006, 06/23].

16. Prometal, 2006, Prometal Home page.Available: www.prometal-rt.com [2006,06/23] .

85

Adrian Long is the former

Dean of the Faculty of

Engineering and held the 1849

Chair of Civil Engineering for

30 years. He has over 40 years

of research experience, mainly

on the behaviour of reinforced and prestressed

concrete structures and concrete technology.

ABSTRACTThis paper describes the further development

of a novel flexible masonry concrete arch system,

which requires no centring in the construction

phase or steel reinforcement in the long-term.

The arch is constructed from a ‘flat pack’ system

by use of a polymer reinforcement for supporting

the dead load but behaves as a masonry arch

once in place. The paper outlines the construction

of a prototype arch and load testing of the

backfilled arch system with some comparisons to

the results from ARCHIE. The arch had a load

carrying capacity far in excess of the current

design loads.

KEYWORDS:Concrete, Masonry arch bridges, Polymer

reinforcement.

INTRODUCTIONOn the basis of aesthetics, durability and

strength the arch form of construction is ideal for

bridges; however its Achilles heel is the need for

centring. Thus if an arch system could be

developed which retained all the above attributes

but does not require centring, which currently

makes arch bridges non-competitive, the concept

would be highly attractive for engineers. This

was the main motivation for taking up the

challenge to develop such a system. A further

goal was to overcome persistent problems of

deterioration in existing bridges. For example

many reinforced concrete beam/beam and slab

bridges, whilst initially cost effective in the 1960s

and 70s, have been found to deteriorate relatively

quickly so that many have had to be replaced

well before half their expected design life. In

addition, this latter form of construction often

fails to meet European Standards for load

carrying capacity [1,2], whereas much older arch

bridges have proven to be very satisfactory both

from the viewpoint of strength and durability.

Therefore, this work supports the stated aim of

the UK Highways Agency [3] whereby

consideration shall be given to all means of

reducing or eliminating the use of corrodible

reinforcement indicating the use of plain concrete

structural elements. It also recommends the use

of the arch form of construction where ground

conditions permit.

In Northern Ireland, the total number of

bridges on motorways, trunk and non-trunk

roads is about 6400, of which 64% are masonry

arch structures. The maintenance of these

masonry structures is an annual exercise requiring

significant funding and staff resources and any

saving that can be achieved has significant

financial implications. Annual expenditure on the

maintenance and strengthening of minor

structures, the majority of which are masonry

arches, now amounts to about £5 million in

Northern Ireland alone. Thus an arch bridge

system with low or zero amount of corrodible

reinforcement, which is easy to construct, would

be highly desirable. In response to this challenge

an arch system which does not require centring

has been developed by staff at Queens University

Belfast working alongside Macrete Ireland Ltd

with the support of Invest Northern Ireland, a

Knowledge Transfer Partnership (KTP), the

Institution of Civil Engineers (ICE) and DRD Roads

Service NI. The prototype arch unit has been

manufactured by Macrete, lifted successfully and

monitored during backfill operations [4, 5]. This

paper provides more detailed information on the

development of the arch system and presents the

results of recent load testing of a prototype arch

with concrete backfill.

METHOD OF CONSTRUCTIONThe arch is constructed and transported in the

form of a flat pack using a polymer grid

reinforcement to carry the self weight during

A FLEXIBLE CONCRETE ARCH FOR SUSTAINABLE BRIDGES*

Professor Adrian Long,

PhD, DSc, FEng, FIAE, FACI, FICE, FIStructE, FIEI, FICT

Queen’s University Belfast

* This paper was co-authored by Su Taylor, Abhey Gupta, Jim Kirkpatrick, and Iain Hogg.

lifting but behaves as a masonry arch once in

place.

Basically there are two options for the

construction of the arch unit. The voussoirs can

be precast individually, laid contiguously

horizontally with a layer of polymer grid material

placed on top. An in-situ layer of concrete,

approximately 40 mm thick, is placed on top and

allowed to harden to interconnect the voussoirs.

Alternatively, the same unit can be made in a

single casting operation by using formwork with

wedge formers spaced to simulate the tapered

voussoirs. Both forms of construction are shown

in Fig. 1. The arch unit can be cast in convenient

widths to suit the design requirement, site

restrictions and available lifting capacity. When

lifted, the wedge shaped gaps close, concrete

hinges form in the top layer of concrete and the

unit is supported by tension in the polymer grid.

The arch shaped units are then placed on precast

footings and all self-weight is then transferred

from tension in the polymer to compression in

the arch.

Of the two options above, the first, based on

precast voussoirs, is preferred because with the

second it has been found that it is difficult to:

• Produce accurately machined wedges

which are sufficiently robust to be

anchored to the base form so that

movement does not take place during

casting,

• Remove the hardened concrete unit from

the complex formwork; e.g. uniform

lifting along the complete length is

required or turning upside down with the

possibility of damage when being turned

back after the formwork has been

removed.

MATERIALSControl samples of the polymeric

reinforcement were tested to ascertain the

material properties. The polymer grid is available

in a range of mesh sizes and strengths. The

manufacturer’s data sheet also gave the

longitudinal and transverse tensile strength in

kN/m. Table 1 summarises the results of material

tests carried out at Queen’s University Belfast. The

differences in the measured and specified values

for the tensile strength may be accounted for by

the differences in the test methods. The initial

0.35 m wide arch used a 100/15 Paragrid® but a

stronger grid, 150/15, was used in the 1 m

prototype arch for the lifting and placing of the

arch. The concrete used for casting the voussoirs

was the standard precast mix used by Macrete

Ltd with a 7-day compressive strength of ~40

N/mm2 and a 28-day compressive strength of ~55

N/mm2.

MANUFACTURE OF THE ARCH UNIT

A prototype arch unit of 5 m span, 2 m rise

and 2.5 m internal radius was constructed and

lifted. This arch required twenty-three voussoirs,

which were 1 m wide and 200 mm deep with a

40 mm slab interconnecting in situ screed. The

arch was lifted at approximately the quarter span

points with additional nominal support at the mid

span region. During lifting, the arch drags at

each end and the mid span point tends to sag;

hence the need for the additional support. When

the end cantilevers are fully effective they

produce a hogging moment in the mid span

region which assists in the formation of the arch.

During this operation a critical case occurs when

the arch is fully formed and suspended at the

lifting points. A maximum bending moment

occurs at the lifting points for the cantilever ends

86

Figure 1(a): Construction of arch unitusing pre-cast individual voussoirconcrete blocks.

Figure 1(b): Monolithic construction ofthe arch unit using a special form ofprecast wedges.

87

and to simulate this condition a series of short

beam elements were tested to establish the

capacity and to investigate the rate of creep in

the low modulus polymer reinforcement. A

summary of the results of these tests is given in

Table 2 and it was found to give an adequate

factor of safety during the lifting procedure. The

lifting sequence is shown in Figures 2(a), 2(b) and

2(c).

Subsequently, an anchor block detail

was designed for the seating of the

arch ring, and the arch unit complete

with tapered seating is shown in

Figure 2c. The anchor block caters for

the slope of the last voussoir,

enabling the arch to form correctly. It

also provided some lateral restraint to

the arch ring during construction,

prior to completion of the arch

system with spandrel walls and

backfill, and in the long-term under

live loading.

STABILITY TEST

Test Set-upTo assess the flexibility of this system, a

stability test under backfilling operations was

conducted and the arch unit was monitored for

horizontal deflections, vertical deflections and

strain at the voussoirs joints. It was originally

intended to backfill the arch with good granular

backfill. However, after a preliminary cost

estimate for this span, it was decided to trial the

use of lean-mix concrete as a backfill option.

Formwork for the backfill was set up independent

Arch Load Applied Ultimate Ultimate Moment: Load: excluding the including the self

self weight weight (kN) (kN.m)

1 14.0 7.52 16.0 8.3

Table 2: Results for Test Beams.

Figure 2: Lifting procedure for the prototype arch.

Figure 2(c): Arch located on tapered seating units

Figure 2(a): “Flat-Pack” arch system. Figure 2(b): Arch unit during lifting.

Sample ID Tensile strength of Tensile strength ofmaterial material from test

from manufacturer results (kN/m width)(kN/m width)

100/15 100 43.6

150/15 150 72.2

Table 1: Paragrid Properties from tests at QUB.

88

of the arch ring to allow free movement during

the placement of the concrete.

Test Procedure and ResultsThe backfilling operation was carried out by

placing approximately 250 mm deep layers of

concrete to each side of the arch and up to 1.5 m

in a horizontal distance from the back face of the

anchor block. The distribution of load to either

side of the arch ring was aimed at minimising the

effects of asymmetric loading. The depth of the

concrete was measured and the transducer and

strain gauge readings were recorded at each

increment of backfill load. The readings showed

a reasonably symmetric response to the

backfilling operation.

The maximum movement in the crown was

0.8 mm upwards in the back face and occurred

at the full height of the concrete backfill. This

was due to the lateral pressure of the wet

concrete creating an inwards movement at the

sides of the arch, which in turn caused the crown

to rise. As the concrete hardened, the lateral

pressure reduced and the deflection response at

the crown reversed in direction; that is, the

upward movement ceased and the crown

deflected very slightly downwards. The results of

the stability tests showed very little movement in

the arch ring and it was concluded that the arch

was stable under the backfilling operations.

LIVE LOAD TESTINGThe 1 m wide prototype flexible concrete arch

system, as shown in Figure 3, was tested in

accordance with the requirements of BS 8110 [6].

A simulated static wheel load was applied at the

mid span and third span of the arch ring. The

single wheel load, for the intended category of

bridge, is 57.5 kN. However, under both loading

conditions, six times the single wheel load, 350

kN, was applied without any signs of distress.

InstrumentationThe instrumentation set-up for the midspan

and third span load tests is shown in Figure 4.

Deflection transducers were used to monitor both

horizontal and vertical deflections and vibrating

wire strain gauges were used to measure crack

openings at the joints between voussoirs.

Testing ProcedureA typical test arrangement is depicted in Figure

4. A circular concentrated load was applied at

the loading position via a 300 mm diameter steel

plate bedded on soft board. An accurately

calibrated 500 kN hydraulic jack was used to

Figure 3: Test set-up for midspan loading.

Figure 4: Instrumentation and test set-up for arch ring and arch with backfill.

apply the load and the test rig was assembled

with the top beam horizontal about both axes,

thus minimizing eccentricity effects. The

simulated static wheel load was applied at mid-

span and then to the third span position. In each

case, the load was applied incrementally and

deflection and strain measurements were

recorded at each increment of load. A service test

load of 100 kN was applied twice at each load

position prior to the application of the full test

load. During the full test load, the behaviour of

the arch and backfill was monitored via digital

photos and the structure was carefully checked

for cracking.

ANALYSIS OF TEST RESULTS

Observed BehaviourFor both the midspan and third span loading,

no cracking was observed under the two test

loads of 100 kN. The first cracking appeared at

an applied load of 200 kN, in the concrete

backfill. In both cases, the crack started on a line

adjacent to the perimeter of the loaded area and

propagated diagonally towards the arch ring

voussoirs. Under midspan loading, the crack

followed the vertical line of the 18 mm plywood

stop end at midspan (Figure 4), which had been

used between the two sides of the backfill to

facilitate demolition. The plywood stop end had

been placed across the whole width of the arch

(i.e. 1 m length) and this probably caused

additional cracking in the concrete backfill. There

was no visible opening in the joints between the

voussoirs and no further cracks developed under

the full test load. Under third span loading and at

the full test load of 352 kN, the joint directly

below the plywood stop end had opened by 2

mm. This opening was partly due to the

differential settlement in the backfill due to the

presence of the plywood across the full width of

the arch unit, which provided a shear plane.

Cracking also occurred horizontally adjacent to

the anchor block. After unloading, there was

visible recovery in the deflections and the cracks

closed in the arch system.

Deflection measurementsThe overall deflections at the maximum

applied loads are more clearly illustrated in

Figures 5 and 6. It can be seen that the maximum

deflection under the midspan load was 2.3 mm

inwards at an applied load of 340 kN. The

maximum deflection under the third span loading

was 10.3 mm outwards at the third span at an

applied load of 352 kN. A deflection of 10.3 mm

is equivalent to the (effective span / 508) and

within acceptable limits for deflection. It is

evident that the plywood stop end caused a

higher degree of cracking at the midspan

compared to a continuous backfill and, coupled

with the polythene liner, probably gave a

conservative prediction of the deflections

compared to an arch ring without a liner or

plywood at midspan.

From the load versus deflection results, it was

noted that there was a shift in the reading at an

applied midspan load of 200 kN. This was due to

cracking in the backfill at the position of the

plywood, as discussed earlier. The rate of change

in deflection also changed at an applied third

span load of 200 kN. This was also due to

cracking in the concrete backfill. However, the

maximum deflections, for both loading

conditions, were less than span/500 at an applied

load of ~340 kN (six times the single wheel load).

The recovery of the arch after loading was

good. For example, for the full test load at

midspan, the maximum deflection at V3 was 2.0

mm at an applied load of 340 kN. After

Figure 5: Test 3. Full test load at midspan: Deflection x 50 at 340 kN load.

89

90

unloading, the permanent deflection was 0.3

mm. This equates to an 85% recovery in

deflection which is within the acceptable limits

given in BS 8110: Pt 2: Section 9 [6].

Strain MeasurementsThe strain measurements were very low with

the midspan load indicating low levels of stress in

the arch ring. Under the third span loading, the

strain measurements were also low. However, the

largest opening occurred at the midspan voussoir

right hand side joint, which did not have a

gauge. The joint opening was approximately 1 –

2 mm at the maximum load at third span.

Analysis of Arch SystemAn analysis of the arch unit was carried out

using ARCHIE[7], a numerical analysis package

which allowed for interaction with the arch

backfill. It is important to note that this software

is also used by the DRD Road Service in Northern

Ireland for load assessment analysis of their arch

bridges. Therefore, validation of the

manufactured arch unit using ARCHIE was a

critical task in the development of the arch

system. The arch unit was analysed under

different wheel loading conditions. A typical case

of the arch unit analysis is shown in Figure 7. An

arch unit of the required geometry can be created

and loaded with the standard wheel loads. A line

of thrust is indicated in Figure 7. Under design

loading, the position of the thrust line in the arch

unit gives information about the stability of the

unit. Furthermore, it was found that the thrust

line is affected by the application of Passive

Pressure (PP) and Backing Material (BM) at the

springing level. Therefore, for a particular loading

condition, using suitable values of PP and BM,

the required thickness of the arch unit can be

found. The predicted deflected shape was similar

to that found experimentally and the predicted

ultimate capacity was very conservative, i.e., less

than the 352 kN applied

when the system was

found to show no signs

of distress or of

impending collapse.

Conclusionsfrom TestsIt can be concluded that

the 1 m arch ring with

concrete backfill was

capable of supporting a

midspan load of 340 kN

and a third span load of

352 kN. The arch ring

showed good recovery in

deflections and cracking

after the removal of all

load. This was despite

Figure 6: Test 3. Full test load at thirdspan: Deflection x 50 at 340 kN load.

Figure 7: Analysis of the arch system using ARCHIE software.

91

the presence of the plywood stop end across the

width of the backfill at midspan and the

polythene layer between the arch ring and the

backfill. These were used to facilitate demolition.

The maximum test loads were six times the

single wheel load and nearly twice the ultimate

load including ULS, dynamic and contingency

factors of safety. The maximum deflection, with

third span load, was 10 mm and is equivalent to

(span/508) which is within acceptable limits for

deflection. The strain values were very low at

maximum applied loads indicating low levels of

stress in the arch ring. The results form ARCHIE

gave a similar deflected shape to the measured

results although the predicted load capacities

were conservative.

OVERALL CONCLUSIONSThe novel arch system has been demonstrated,

in tests reported in this and other papers, to be a

viable alternative to long established methods of

construction and the following advantages have

been identified:

1. As the arch system is cast horizontally it

can be transported to site conveniently in

a “flat pack” form,

2. As centring is not required during

installation this greatly simplifies the

process and enhances the speed of

construction,

3. As there is no corrodible reinforcement

the long-term durability should be

assured,

4. Initial estimates would indicate that the

system is cost competitive with

alternatives such as RC box culverts,

which do not share the aesthetic benefits

or the longevity of an arch,

5. Ease of access to restricted sites. Here the

precast voussoirs can be brought to the

site separately where they can be

transformed into a flexible arch and then

lifted into position in the usual way

A complete 5 m wide arch bridge with a 5 m

span and 2 m rise complete with spandrels and

fill has been constructed at Macrete (Figure 8).

Load testing has been successfully carried out

using two 50 t hydraulic jacks. A replacement

bridge is also planned in conjunction with DRD

Northern Ireland Roads Service.

ACKNOWLEDGEMENTSThe Authors would like to thank the Institution

of Civil Engineers, Invest NI and the Knowledge

Transfer Partnership scheme for supporting the

development of this novel system.

REFERENCES1. BD37/01, Departmental Standard, Loads for

Highway Bridges (used with BS5400: Pt2)Design Manual for Roads and Bridges,Volume 1, Section 3, Part 14, Departmentof Transport, Highway and Traffic, 2001.

2. BD44/95, Departmental Standard, Theassessment of concrete highway bridges,Volume 1, Section 3, Department ofTransport, Highway and Traffic, 1995.

Figure 8: Arch with spandrels.

92

3. UK Highway Agency, BD 57/95 & BA 57/95,Design for Durability Design Manual forRoads and Bridges, Volume 1, Section 3,Department of Transport, Highway andTraffic, 1995.

4. TAYLOR, S.E., GUPTA, A., KIRKPATRICK, J.,LONG, A.E., RANKIN, G.I.B. and HOGG, I.,Development of a novel flexible concretearch system, 11th International Conferenceon Structural Faults and Repairs, Edinburgh,June 2006.

5. GUPTA, A., TAYLOR, S.E., KIRKPATRICK, J.,LONG, A.E. and HOGG, I. A FlexibleConcrete Arch System for Durable BridgesProceeding from IABSE Conference,Budapest, 2006.

6. British Standards Institute, BS 8110: Part 2:Structural use of concrete: Code of practicefor special circumstances Section 9:Appraisal and testing of structures andcomponents for construction, London,1985.

7. ARCHIE-M: Masonry Arch Bridges andViaduct Assessment Software, Version2.0.8, OBVIS Ltd. UK.

93

Dr Marios N Soutsos,

Reader/Lecturer in Engineering

since 1995 following two years

previous industrial experience

relating to concrete repair and

quality control of ready mixed

concrete. Principal research experience is in

construction materials and current interests

include: high strength concrete, cement

replacement materials, chemical admixtures,

concrete rheology, the use of recycled demolition

aggregate in concrete products, repair materials

as well as heat of hydration effects in concrete

structures.

ABSTRACTA study has been undertaken at the University

of Liverpool to investigate the potential for using

construction and demolition waste (C&DW)-

derived aggregate in the manufacture of precast

concrete building blocks. Recycled aggregates

can be used to replace quarried limestone

aggregate, usually used in coarse (6 mm) and fine

(4 mm-to-dust) gradings. Market research has

been carried out to determine the economic

viability of using C&DW-derived aggregates in

blocks. The availability and transportation costs

of quarried and recycled demolition-derived

aggregates have been compared and there

appeared to be scope for investigating the

technical aspects, in addition to the economic

aspects, of the use of recycled demolition

aggregates in block manufacture. The

manufacturing process used in factories, for

large-scale production, involves a “vibro-

compaction” casting procedure, using a relatively

dry concrete mix with a low cement content

(100 kg/m3). Trials in the laboratory have

successfully replicated the manufacturing process

using a specially modified pneumatic hammer to

compact the concrete mix into oversize steel

moulds to produce blocks of the same physical

and mechanical properties as the commercial

blocks. This enabled investigations, i.e. the effect

of recycled demolition aggregate on the

compressive strength, to be carried out in the

University’s laboratory. The physical characteristics

of recycled demolition (C&DW) aggregates may

adversely affect the mechanical properties of the

blocks. However, levels of replacement of

quarried limestone aggregates with C&DW-

derived aggregates have been determined that

will not have significant detrimental effect on the

compressive strength. Factory trials showed that

there were no practical problems with the use of

recycled demolition aggregates. The strengths

obtained confirmed that the replacement levels

selected, based on the laboratory work, did not

cause any significant strength reduction, i.e. there

was no requirement to increase the cement

content to maintain the required strength, and

therefore there would be no additional cost to

the manufacturers if they were to use recycled

demolition aggregates for their everyday concrete

building block production.

KEYWORDSRecycling of materials, Sustainability,

Construction and demolition waste, Concrete

blocks, Aggregates, Environment, Landfill.

INTRODUCTIONIn 1991 the European Commission initiated

the Priority Waste Streams Programme for six

waste streams. One of these was construction

and demolition waste (C&DW) [1]. While C&DW

accounts for 15% of the estimated annual waste

in the UK [2], about 275 million tonnes of new

construction aggregates are extracted annually.

By 2012, if UK demand for aggregates increases

by an expected 1% per annum, an extra 20

million tonnes of aggregates will be needed each

year. About 60% of this is crushed rock and 40%

is sand and gravel [3]. These are essential materials

for buildings and infrastructure, but extraction

causes significant environmental damage.

Government aims are to reduce demand for

primary aggregates by minimising the waste of

construction materials and maximising the use

that is made of alternatives to primary aggregates[4]. An attempt to address the environmental costs

associated with quarrying has been the

PRECAST CONCRETE PRODUCTS MADE WITH RECYCLED

DEMOLITION AGGREGATE

Dr Marios Soutsos, BEng(Hons), PhD, MICT

Liverpool University

* This paper was co-authored by Stephen Millard and Emeritus Professor John Bungey

94

introduction of the Aggregates Levy in April

2002[5].

Although there are many potential uses for

C&DW materials, most are currently used for low-

value purposes such as road sub-base

construction, engineering fill, or landfill

engineering. Only 4% is recycled for high

specification applications; the reason for this

being that while many C&DW materials could be

used for higher-level uses, potential users are

deterred by the perceived risks involved [4].

AIMS AND OBJECTIVES OF THE PROJECT AT LIVERPOOL UNIVERSITY

The study at Liverpool University investigated

the use of crushed C&DW, i.e. recycled

demolition aggregate, in the production of

blockwork and the scope was recently expanded

to include concrete paving blocks and concrete

flags. Blockwork has been selected as a promising

product to begin the investigations because:

• Possible contamination from C&DW

directly affecting reinforcement is not an

issue as common blocks are unreinforced

• Unlike construction projects, blockwork

fabrication is essentially a manufacturing

process where supply of input materials &

storage of output are more easily

managed

• There may be local circumstances that

would make the use of secondary and

recycled materials for high-grade use cost

effective. Merseyside, and more

specifically Liverpool, has been selected as

a realistic illustrative example of a major

UK conurbation undergoing regeneration[6]

• Resource supply or feed material can be

guaranteed in an urban area like Liverpool

where replacement of infrastructure is

occurring, natural aggregate resources are

limited, disposal costs are high, and

environmental regulations encourage

recycling.

In addition to investigating the technicalities of

producing concrete using C&DW, the economics

and practicalities involved were also investigated.

VOLUME AND COST STATISTICSFOR BUILDING BLOCKSThe basic concrete building block is a

commodity product and therefore the profit

margin is low. Large multi-national companies

that generally own quarrying operations

dominate the sector. The raw materials used to

manufacture blocks, whether they are virgin

aggregate, lightweight or man-made aggregates,

are costly to transport and therefore most

manufacturers have been faced with a choice

between being located close to the raw materials

or close to the market. It appears that in the

majority of cases the decision is to have the

precast factory close to or even at the quarry site.

The standard block, because of the low profit

margins, is only sold regionally, within a radius of

50 miles of the precast factory. Table 1 shows the

estimated transport costs for delivery of blocks.

Recent construction statistics by the

Department of Trade and Industry (DTI) indicate

that approximately 360 million blocks are

produced annually[7]. The estimated aggregate

consumption can be based on the assumption

that 90% of each block is aggregate, i.e.

aggregate consumption is 3.6 million tonnes per

year. A single precast factory can use up to 500

tonnes of aggregate on a single day. Based on

the construction output in the North West being

approximately 10% of the total output [7], it is

estimated that 360,000 tonnes of aggregates are

needed annually for the production of blocks in

the North West.

AVAILABILITY ANDTRANSPORTATION OF AGGREGATES

Construction aggregates are a high-volume,

low-unit-value commodity, which makes the

transportation cost a determining factor in

competing sources. Thus the location of resources

may encourage the use of construction and

demolition waste-derived aggregates in certain

areas. For example, past surveys [8] have shown

major movements of quarry materials from one

region to another; for instance, West Midlands

and North Wales to the North West of England,

see Figure 1. In considering future supply patterns

to the North West, assumptions will need to be

made about supplies from Wales, where planning

policies for aggregates are now matters for the

Distance Approx cost

10 miles £20

30 miles £60

50 miles £100

100 miles £200

Table 1: Estimated transport costs for aload of 20 tonnes.

95

devolved administration. It cannot therefore be

assumed that past supply patterns will necessarily

be maintained in the future. It is therefore not

surprising that the Regional Waste Strategy for

the North West [9] aims to “promote the use of

recycled construction and demolition waste in

construction projects and encourage developers

and contractors to specify these materials

wherever possible in the construction process”.

AVAILABILITY AND COST OF C&DW DERIVED AGGREGATES

Liverpool has been selected as a realistic

illustrative example. The Liverpool Housing Action

Trust (LHAT) alone demolished 52 of the 72 tower

blocks in Liverpool between 2001 and 2006.

Figure 2 shows just one of these tower blocks

that was demolished using explosives, a

technique known as “implosion”. 15 000 tonnes

of construction and demolition waste resulted

from the demolition of this building alone. This

waste was transported to a nearby crushing plant

where it was converted to Department of

Transport, Type 1 road sub-base material. Natural

aggregate resources are limited in Liverpool, i.e.

there are no aggregate quarries, but resource

supply or feed material for a crushing plant can

be guaranteed in an urban area where

replacement of infrastructure is occurring. The

feed material, however, may change, i.e. the

tower blocks were mainly constructed of in situ

concrete or precast concrete panels, while most

of the local council housing expected to be

demolished in the near future will be mainly low

rise masonry buildings.

It appears that at least 4.5 million tonnes of

hard C&DW is crushed and/or screened annually

for use as aggregate. Very little evidence was

found of hard C&DW that could be recycled into

aggregate being land filled as waste in the

Merseyside region. Only very modest tonnages

were identified as being used in an unprocessed

form and then it was mainly for landfill

engineering, see Table 2 [10]. However, the cost for

crushing the C&DW, which is estimated to be

approximately £7 per tonne, see Table 3 [10], is not

recovered when it is sold as road sub-base

aggregate. The selling price depends heavily on

the demand and can vary between £2 and £4 per

tonne. The demolition contractors are still

therefore required to include for this difference

and they end up paying the recycling plant

operator to take away the C&DW. Operators of

crushing plants would also welcome not only an

increase in price per tonne but also a guaranteed

constant/regular demand for the C&DW-derived

aggregate. Block making factories appear to be

very interested in C&DW-derived aggregates if

Figure 1: Major movements in 2001 of crushed aggregates from one region toanother[12].

96

the price is lower than that of quarried

aggregate. Approximate aggregate ratios used for

concrete blocks as well as the range of prices per

tonne are shown in Table 4. Normally two sizes of

aggregates are used for concrete block making,

i.e. a 6 mm aggregate and a 4 mm-to-dust

aggregate. A conservative value of £7 per tonne

for 6 mm C&DW-derived aggregates would

satisfy both the operators of crushing plants and

the block making factories. There was therefore

scope for investigating a high-end value market

for C&DW.

PHYSICAL PROPERTIES OF C&DW AGGREGATES

If C&DW-derived aggregates are to be used in

precast concrete products, specific gravity,

absorption, fineness, and angularity are all

important physical properties that need to be

taken into consideration. In the Liverpool study

aggregate gradings have been obtained for

limestone quarried aggregates, supplied by a

block making factory, as well as recycled concrete

and masonry-derived aggregates supplied by local

demolition companies. The concrete C&DW that

was crushed to produce aggregates came from

the foundations of a multi-storey reinforced

concrete building while the masonry C&DW came

from the demolition of low-rise council houses. It

was expected that the detrimental effect of

masonry-derived C&DW aggregates on

compressive strength would be higher than that

of concrete-derived C&DW aggregates. It was

therefore considered prudent to investigate the

effects of concrete- and masonry-derived

aggregates separately, with the possibility of

interpolating to obtain the effects of a mixture of

the two. The percentage of masonry in the

mixture is likely to vary depending on what

contract, whether multi-storey buildings or

masonry houses, the demolition contractor has

secured.

As delivered from the crushing plant, the 4

mm-to-dust recycled masonry was found to be

much finer than natural quarried limestone while

the opposite was found to be true for the

concrete-derived aggregate. In order to obtain a

combined grading similar to that of natural

limestone, the proportion of masonry fines

needed to be reduced from 56% to 43% while

that of concrete fines needed to be increased

from 56% to 61%. However, the initial mixes

indicated that the concrete fines could be

reduced to 45% and still get the same texture on

the blocks as those made with limestone

aggregates. Both the concrete- and the masonry-

derived recycled aggregate had very high water

absorptions: around 10% for the coarse fraction

and up to 18% for the fine fraction.

EXPERIMENTAL WORKThe work described here is divided into three

series of tests; the first series was aimed at

replicating the industrial procedure for making

blocks while the second and third series aimed at

determining the effect of replacing quarried

limestone aggregate with concrete- and masonry-

derived aggregate respectively.

Figure 2: One of fifty-two tower blocks that were demolished in Liverpool between2001 and 2006.

97

Series I – Preliminary TrialsPurpose built moulds were designed and

fabricated in the workshop to enable full height

but half-length block specimens to be made. A

“compaction rig” was designed and fabricated to

produce specimens of the required density, size

and dimensional tolerance. The rig allowed the

vibration/compaction hammer drill to slide down

guide rods to a pre-determined height, see Figure

3. The moulds are oversized in height to allow

the uncompacted material to be placed in them.

The weight of wet material required to produce a

concrete block of approximately 20 kg when dry,

was estimated from previously determined

wet/dry density relationships. The

vibration/compaction hammer drill then

compacted the material to the required height of

215 mm, the height of factory-produced blocks.

Table 2: North West estimates for C&DW Material- Year 2001 [11].

Table 3: Costs for crushing C&DW [14].

Approximately ten tonnes of aggregate were

obtained from a precast concrete factory near

Buxton, Derbyshire. This comprised coarse (6 mm)

and fine (4 mm-to-dust) limestone aggregate.

Specimens were cast, with the proportions of

cement (100 kg/m3), water (30-60 kg/m3), and

coarse and fine aggregate (1250 and 1000 kg/m3

respectively), as given by the factory, to determine

the exact procedure required for replicating the

industrial process. These included:

• Water/cement ratio. The optimum water

cement ratio was found to be the lowest

that allowed a block to be cast and

demoulded immediately after casting

without the block disintegrating

• Curing regime. Laboratory made blocks

were covered in wet hessian and a

polythene sheet for the first 24 hours and

then air cured at ambient laboratory

conditions

• Fibreboard versus mortar capping. The

manufacturers use a predetermined ratio

of 1.06 between the strength of 28-day

mortar-capped blocks and that of 7-day

blocks tested with fibreboard packing on

the ends to simplify and speed up the

testing procedure. The ratio, obtained for

laboratory cast specimens, was very similar

to one used by the manufacturer

• Density. One of the critical parameters in

achieving a target compressive strength

appears to be the density of the concrete

block. A range of acceptable densities, i.e.

1850 to 2000 kg/m3, was provided by a

precast block manufacturer. Blocks made

in the laboratory with a dry density of

2000 kg/m3 achieved strengths above the

target minimum of 7 N/mm2 at 28-days.

Series II – Concrete-derivedaggregates.

After successfully having replicated the

industrial block-making procedure in the

laboratory, the replacement of quarried limestone

with concrete-derived aggregates was

investigated. The mix proportions of

natural limestone aggregate used by

a block making factory, shown in

Table 5 as the “CONTROL”, had to

be converted to volume, replaced by

an equal volume of C&DW recycled

material, and then converted back

into weight. This ensured that the

98

Aggregate Size % of total Approx costaggregates per tonne

used in the mix

4 mm to dust 50-60% £3 – 66 mm 40-30% £8 – 10

Table 4: Prices for quarried aggregates used inconcrete blocks.

Figure 3: Alignment compaction rig.

99

replacement was on a volumetric basis and was

required in order to take into account the

different densities of the recycled aggregates

compared to quarried limestone aggregates.

Blocks made with recycled concrete aggregates

therefore had marginally lower wet densities than

quarried limestone blocks, e.g., 1890 kg/m3 for a

block using 100% replacement of both 6 mm

and 4 mm-to-dust limestone aggregates with

recycled C&DW concrete-derived aggregates

compared to 2125 kg/m3 for a block using only

limestone aggregates.

Each series of mixes started with an initial

cement content of 100 kg/m3. It has been found

from trials that if the concrete mix holds together

after it is squeezed tightly in the hand then the

mix will have sufficient workability to be

compacted into the moulds. A handful of the

concrete mix was taken after mixing for three

minutes. If it did not hold together then

additional water was added. Two or three blocks

were then cast. An increment of additional

cement was then added, the concrete was re-

mixed for another two minutes, and a visual

inspection again determined whether it had

sufficient workability to be compacted into the

moulds. Incremental increase of the cement

content in this manner resulted in blocks with

various cement contents, water-cement ratios,

and therefore compressive strengths. Early age

strengths, i.e. 1-day, were sufficient to allow

moving and stacking of blocks as required in the

manufacturing process. All blocks were tested at

7-days using fibreboard end packing and the

conversion factor of 1.06 was used to convert

this strength to the equivalent 28-day strength.

All the values on the figures are the equivalent

28-day strengths.

Figure 4(a) shows that 100% replacement of

both coarse 6 mm and fine 4 mm-to-dust

quarried limestone aggregate with recycled

C&DW concrete-derived aggregates has a

considerable detrimental effect on the

compressive strength. The cement content would

need to be increased from 100 kg/m3 to

approximately 130 kg/m3 in order for the strength

to be at least 7 MPa.

It was believed that the fines fraction, i.e. 4

mm-to-dust, would be the one that would have

the biggest detrimental effect on the compressive

strength. Studies therefore aimed to replace

either the coarse fraction or the fines fraction

only, but not both, in order to quantify the

relative effects of each fraction. Promising results

were obtained for a 60% replacement of the

coarse fraction with C&DW concrete-derived

aggregate, i.e. there was no detrimental effect on

the compressive strength. However, increasing the

coarse fraction replacement to 100% appears to

have the same detrimental effect as replacing

both the coarse and the fine aggregate fractions

with C&DW concrete-derived aggregates. Figure

4(b) shows that replacing the fine aggregate

fraction only with C&DW concrete-derived

aggregate has more of a detrimental effect on

strength than the coarse aggregate replacement.

Higher than 30% replacement level of fine

aggregate is not recommended. Figure 5 shows

the compressive strength versus the percentage

of replacement of limestone aggregate with

coarse and fine concrete derived aggregates – (All

mixes had 100 kg/m3 of cement). It was

concluded that reasonable replacement levels

would be 60% for the coarse fraction and not

more than 30% for the fine fraction.

Series III – Masonry-derivedaggregates.

The replacement of newly quarried limestone

aggregate with masonry-derived aggregate has

been investigated independently from concrete-

derived aggregates. The lower density of

Table 5: Mix proportions for blocks made with CONCRETE-derived aggregates.

100

masonry-derived aggregates was expected to be

problematic. Replacement of limestone with

masonry-derived aggregates on an equal weight

basis was not possible. The increased volume of

material resulting from the different densities

could not be compacted into a block of the

required dimensions. Replacement again had to

be on a volumetric basis, rather than weight, in

order to take into account the different densities

of the materials. The mix proportions are shown

in Table 6. The effect of replacing newly quarried

limestone with recycled masonry-derived

aggregate can be seen in Figure 6. It appears that

the detrimental effect varies almost linearly with

the percentage replacement level. 20%

replacement level by coarse and fine aggregate

was selected as it still produced blocks with

compressive strengths above 7 MPa.

Factory Trials.In total, 10 tonnes of recycled aggregate were

delivered to the Forticrete Ltd factory at Buxton.

This comprised concrete- and masonry-derived

aggregate, and samples obtained indicated that

the gradings were all comparable to those used

in the laboratory with the exception of the

coarse, i.e. 6 mm, concrete-derived aggregate.

This was not “single-sized” but an “all-in”

aggregate. It was therefore decided that both the

60% coarse fraction and the 30% fine fraction of

concrete-derived aggregate should be replaced by

the “all-in” aggregate. The batch weights and

mix proportions used at the factory are shown in

Table 7.

Figure 4: Compressive Strength Versus Cement Content for Coarse and Fine fractionreplacement (%) of limestone with CONCRETE-Derived Aggregates.

(a) Coarse fraction replacement (b) Fine fraction replacement.

Figure 5: Strength versus percentagereplacement level of limestoneaggregate with coarse and fineconcrete-derived aggregates – (Allmixes had 100 kg/m3 of cement).

101

The factory trials had to be done between

shifts and when there were sufficient numbers of

storage bins empty to hold the recycled material.

After placing the aggregates in the hoppers, the

mix proportions/weights were input into the

computer of the batching plant. The first trial

required three additions of water before approval

was given for the blocks to be cast. This resulted

in the blocks from this first batch being slightly

wetter than the norm. Nonetheless, the same

amount of water was maintained for the higher

cement contents. The cement contents

investigated were approximately 100, 175 and

250 kg/m3. The blocks cast were labelled and one

of the blocks from each batch was weighed. This

enabled an accurate estimate of the cement

content. There were concerns that the red colour

of masonry-derived aggregate would be apparent

in the blocks. However, it was only after careful

inspection of the building blocks that one might

find the odd masonry aggregate particle

appearing on the surface; the cement paste

covered the masonry-derived aggregate

effectively and the colour of the blocks was the

normal dark grey.

The blocks were cured for one day in the

factory’s humidity chamber and six/five were

tested for compressive strength at 7- and 28-

days. The concrete strengths obtained at 28-days

are shown graphically in Figure 7. It is seen that

the industrial vibro-compaction technique was

more efficient than our laboratory technique and

produced higher compressive strengths

throughout. Therefore the relationships between

strength and cement contents were shifted

upwards. The strengths obtained confirmed that

the replacement levels selected, on the basis of

the laboratory work, did not cause any significant

strength reduction, i.e. there was no requirement

to increase the cement content to maintain the

required strength of >7 N/mm2. However, the

curve obtained by the manufacturer for limestone

aggregate appears to rotate rather than shift

upwards. Further investigation is therefore

needed if C&DW-derived aggregates are to be

used in blocks of higher compressive strength.

There would be no additional cost to the

manufacturers if they were to use recycled

aggregates for the common blocks with strength

grades >7 N/mm2. Overall, it was a very

satisfactory factory trial.

Table 6: Mix proportions for blocks made with MASONRY-derived aggregates.

Figure 6: Strength versus percentagereplacement level of limestoneaggregate with coarse and finemasonry-derived aggregates – (Allmixes had 100 kg/m3 of cement).

102

CONCLUSIONSThe location of aggregate resources may

encourage the use of C&DW-derived aggregates

in certain areas. It is believed that Liverpool,

selected as an illustrative example of a city whose

regeneration calls for demolition and major

reconstruction, can benefit from a high value end

use of C&DW-derived aggregates.

Eleven million tonnes of C&DW material arose

in the North West in 2001. About five million

tonnes were crushed and/or screened for use as

aggregates, mainly for road sub-base. The rest of

the C&DW material was used for landfill

engineering, e.g., landfilling of quarry voids. Very

little evidence was found of hard C&DW that

could be recycled into aggregate being landfilled

as waste. However, the cost of crushing the

C&DW, which is estimated to be approximately

£7 per tonne, is not recovered when it is sold as

road sub-base aggregate. The selling price

depends heavily on the demand and can vary

between £2 and £4 per tonne. The demolition

contractors are still therefore required to include

for this difference and they end up paying the

recycling plant operator to take away the C&DW

Table 7: Batch weights and mix proportions used during the factory trials at Buxton.

Figure 7: 28-day strengths of blocks with limestone replacement (%).

(a) “All-in” CONCRETE-derivedaggregates

(b) 20C/20F MASONRY-derivedaggregates

103

(approximately £2 to £3 per tonne). Operators of

crushing plants would also welcome not only an

increase in price per tonne but also a guaranteed

constant/regular demand for the C&DW derived

aggregate. Block making factories will be

interested in C&DW-derived aggregates if the

price is lower than that of quarried aggregate.

Technical aspects of the replacement ofquarried limestone aggregates with concrete-derived aggregates in the production of blocksneed also be considered. The experimental part ofthis work has shown that the industrial “vibro-compaction” technique used by industry forcasting concrete blocks can be replicated using apneumatic hammer drill. This enabledinvestigations, i.e. the effect of C&DW-derivedaggregate on the compressive strength, to becarried out in a laboratory. Levels of replacementof quarried limestone aggregates with C&DW-derived aggregates have been determined thatwill not have significant detrimental effect on thecompressive strength. The levels for concrete-derived aggregate were determined to be 60%for the coarse fraction, i.e. 6 mm, and 30% forthe fine fraction, i.e. 4 mm-to-dust. The levels formasonry-derived aggregate were determined tobe 20% for the coarse fraction, i.e. 6 mm, and20% for the fine fraction, i.e. 4 mm-to-dust.

Factory trials showed that there were nopractical problems with the use of recycleddemolition aggregates. The strengths obtainedconfirmed that the replacement levels selected,based on the laboratory work, did not cause anysignificant strength reduction, i.e. there was norequirement to increase the cement content tomaintain the required strength, and thereforethere would be no additional cost to themanufacturers if they were to use recycleddemolition aggregates for their everyday concretebuilding block production.

ACKNOWLEDGEMENTSThe authors are grateful to the Veolia

Environmental Trust and the Flintshire Community

Trust Ltd (AD Waste Ltd) for funding this project.

The authors would also like to thank Dr. N. Jones,

Mr. K. Tang and Mr. A. Hatzitheodorou for their

assistance with laboratory work. The authors are

also grateful to the following industrial

collaborators for their in-kind contribution to the

project: Clean Merseyside Centre, Marshalls Ltd,

Forticrete Ltd, Liverpool City Council, Liverpool

Housing Action Trust (LHAT), RMC Readymix Ltd,

WF Doyle & Co. Ltd, and DSM Demolition Ltd.

However, the views expressed here are those of

the authors only and do not necessarily represent

those of the funding bodies or collaborators for

this project.

REFERENCES

1. Aggregates Advisory Service, ‘Constructionand Demolition waste – The EuropeanUnion priority waste streams programme’,Digest No. 011, 1999, p. 4.

2. Environment Agency, ‘Waste Production’,Extract from National SWMI report, WasteStatistics 1998-99, www.environment-agency.gov.uk/commondata/105385/swmiwprind.pdf, 1999, p. 14.

3. Office of the Deputy Prime Minister,‘Planning for the Supply of Aggregates inEngland’,www.odpm.gov.uk/stellent/groups/odpm_planning/documents/pdf/odpm_plan_pdf_605804.pdf October 2000, p.70.

4. HM Treasury, ‘Budget Report 2003: Buildinga Britain of economic strength and socialjustice’, Parliament Street, London SW1P3AG, Chapter 7,

5. www.hm-treasury.gov.uk/budget/bud_bud03/budget_report/bud_bud03_repchap7.cfm, 2003.

6. HM Treasury, ‘Summary of responses to theconsultation on the objectives of thesustainability fund under the aggregateslevy package’.

7. www.hm-treasury.gov.uk/consultations_and_legislation/summary_of_responses_ for_the_sustainability_fund_under_the_aggregates_levy_package_consultation/consult_susfund_index.cfm, Sept. 2000.

9. ‘Regeneration means demolition, MPs told’,New Civil Engineer, 30 Nov. 2000, p. 38.

10. Department of Trade & Industry (DTI),‘Construction Statistics Annual 2003’,www.dti.gov.uk/construction/stats/constat2003.pdf, p. 265, 2003.

11. British Geological Society, ReportCR/03/53N, ‘Collation of the results of the2001 aggregate minerals survey for Englandand Wales’, Keyworth, Nottingham NG125GG, p. 103, 2001.

12. Government Office for the North West,‘Regional Waste Strategy For the NorthWest’, Draft for consultation, http://set.iarna.co.uk/afs/downloads/documents/rws_summary.pdf, p. 43, 21 July 2003.

13. DAVIS, LANGDON and EVEREST, ‘Spon’sCivil Engineering and Highway Works PriceBook 2003’, Taylor & Francis Books Ltd, p.xiv, 2003.

104

105

Jean-Philippe Thierry is an

engineer in material scienceand

has worked for Lafarge since

2001 where he first studied the

fire behaviour of concrete from

conventional to UHPC. He then

worked briefly on fibre orientation before putting

the focus on concrete shrinkage. He is now in

charge of a concrete project and is especially

involved in its deployment.

ABSTRACTThis paper presents an innovative solution for

concrete flooring developed by Lafarge. Joint

spacing is pushed up to 20 m x 20 m without any

reinforcement needed in the concrete.

KEYWORDSConcrete, Flooring, Shrinkage, Drying, Joint

spacing, Reinforcement, Abrasion resistance,

Friction, Subbase.

INTRODUCTIONGround-bearing concrete slabs are often

subject to cracking. It is still important to

distinguish two possible causes:

• Differential settlement in the subbase on

which the slab is laying

• Intrinsic deformation of the concrete due

to drying shrinkage and thermal variations

in buildings.

The former is strictly related to subbase quality,

and especially to compaction homogeneity, which

will not be considered here. The latter is related

to the concrete, for which we propose an

innovative solution.

THE PROPOSED SOLUTIONThe main objective was to develop a solution

that would minimize the cracking risk of slabs.

The specification was to push joint spacing up to

20 m x 20 m, and, if possible, to remove

reinforcement.

Cracking implies that the stresses generated in

the slab are greater than the load capacity of the

concrete. There are two ways of solving this

problem.

Improve concrete propertiesFirst, we have to reduce the drying shrinkage

of the concrete. The best way to do this is to

reduce the quantity of free water remaining in

the concrete after hydration, while controlling the

early-age shrinkage. Typical drying shrinkage

values recorded on instrumented slabs are

between 150 and 250 µm/m.

Then we may also improve the load carrying

capacity of the concrete. Here the solution is

quite straightforward: we have to decrease the

water content. Fortunately, implications in terms

of concrete mix design are close to the ones

related to drying shrinkage. Combining these

two aspects allows us to reach high mechanical

performance: typically a flexural strength of

6MPa.

Ease the deformationsThe average failure strain of concrete in

tension is around 100 µm/m, so even if the

concrete exhibits very small drying shrinkage

deformation completely restraining them would

certainly induce cracking. We also have to take

care with preparatory work on the site and some

recommendations are made to reduce the risk of

cracking.

A ground-bearing slab is not free to move.

The subbase on which it sits is neither smooth

nor flat, so a frictional stress is induced at the

concrete/subbase interface. This stress depends

on the nature of the subbase (gravel, sand, plastic

sheeting, etc.) and on the vertical loading. Many

half-scale tests were carried out to optimise that

interface. A blinded, well-compacted gravel

subbase covered with elastic sheeting gave the

best performance. The stresses induced at the

interface were shown to be 2 to 5 times lower

than on a sand subbase.

SUPPLEMENTARY ADVANTAGESDuring the development phase, we had the

opportunity to place several small slabs. This

allowed us to perform several complementary

characterisations. Firstly, abrasion resistance; our

solution achieved AR Special Class without any

dry-shake on top of the concrete. The

measurements were taken several times by GT

AN INNOVATIVE SOLUTION FOR CONCRETE FLOORING

Dr. Jean-Phillipe Thierry

Lafarge Centre de Recherche

106

Certification, who also checked the pull-off

resistance (2.95 and 1.7 MPa) and the slip

resistance, which conformed to BS 8204-2.

In terms of durability, we measured a low

oxygen permeability (100 x 10m-18 m2), as well as

a low water porosity (9.9%). According to

French recommendations, our solution can be

considered as highly durable.

CONCLUSIONSWith all the work done in the last four years,

we are able to propose a solution that leads to a

low cracking risk and which allows thinner

unreinforced slabs to be designed with less

curling and higher abrasion resistance than

conventional reinforced slabs.

107

Geert De Schutter is full

professor at Ghent University,

Belgium. He is doing research

in the field of concrete

technology, at the Magnel

Laboratory for Concrete

Research, Department of Structural Engineering.

He is laureate of several national and

international awards, among which the important

Vreedenburgh Award in 1998 and the prestigious

international RILEM Robert L’Hermite Medal in

2001. Prof. G. De Schutter is chairman of RILEM

Technical Committee TC 205-DSC on Durability of

SCC. In September 2007, he is organising the

5th International RILEM Symposium on SCC, in

Ghent, Belgium.

ABSTRACTThis paper gives a compilation of the effect of

limestone filler on the hydration, microstructural

development, transport properties and durability

of self-compacting concrete. The effect of

limestone filler (ground calcium carbonate) is

studied in detail, and the importance with regard

to the ‘engineering level’ is outlined.

KEYWORDSself-compacting concrete, limestone filler,

ground calcium carbonate, hydration,

microstructure, transport properties, durability

INTRODUCTIONWith the development of self-compacting

concrete (SCC), a new field of applications has

been realised for ground calcium carbonate. In

order to obtain the key properties of self-

compacting concrete (flow ability, passing ability,

and stability), filler materials are added to the

concrete in large quantities. Besides the normal

cement content (around 300 kg per m3 of

concrete), ground calcium carbonate is added in

quantities of 200 to 300 kg per m3 of concrete.

Within concrete research, ground calcium

carbonate is commonly called limestone filler.

Because of the fundamentally new concept of

SCC, the structure of the material, as a result of

the hydration process, can show some differences

compared to traditionally vibrated concrete. In

order to have a good insight in the effect of

ground calcium carbonate on the properties of

the new cementitious material, a fundamental

study is realised concerning hydration,

microstructure development, and durability. The

effect of SCC is far more important that just

yielding self-compactability. High performance

cementitious materials can be obtained, with

excellent properties not only concerning strength

development, but also related to transport

properties and durability.

In this overview paper, some interesting results

are bundled concerning hydration, microstructure

development, and durability. The effect of

ground calcium carbonate (commonly called

limestone filler) is studied in detail, and the

importance with regard to the ‘engineering level’

is outlined. The results were obtained at the

Magnel Laboratory for Concrete Research.

Several researchers have been involved: Dr. A-M

Poppe, Dr G Ye, Dr K Audenaert, Dr V Boel, and

Dr X. Liu. Their contribution is greatly

acknowledged.

PART I: HYDRATION OFFILLER RICH SELF-COMPACTINGCONCRETE

IntroductionTo realise a concrete that is self-compacting, a

high flowability and a high segregation resistance

have to be combined into one concrete. This is,

as is commonly known, enabled by the use of

superplasticizers and viscosity enhancing agents

combined with high concentrations of fine

particles [1, 2]. As a consequence, these high

concentrations of powder materials (cement and

fillers) can lead to the development of a high

heat of hydration, which might cause problems

during hardening.

Experimental determination ofthe heat of hydration: isothermalhydration tests

To evaluate the heat generation in isothermal

conditions, conduction calorimetry is used. This

test is carried out on small samples of cement

LIMESTONE FILLER BASED SELF-COMPACTING CONCRETE:

FROM MICROSTRUCTURE TO ENGINEERING PROPERTIES

Professor Geert De Schutter, MSc, PhD

Ghent University

108

paste, thus excluding any influences of aggregate

particles. An extensive description of the

conduction method can be found in several

national codes, e.g. the Belgian Standard NBN

B12-213.

In order to be able to make a thorough study

of the heat generation, isothermal hydration tests

are carried out on pure cement as well as on

mixtures of cement and filler. Two different fillers

(a limestone filler originating from Marquise,

France and a quartzite filler originating from Mol,

Belgium) are combined with different portland

cements (CEM I 42.5 R, CEM I 52.5 and CEM I

52.5 HSR LA). In this first stage of the research,

these tests are carried out using no chemical

admixtures such as superplasticizers and viscosity

enhancing agents.

Table 1 gives the powder mixtures as used in

the tests. Four different compositions are

considered, each of which is repeated for the

different cement-filler combinations, and on three

different temperature levels (10°C, 20°C and

35°C). In this table some parameters of the

mixtures are given, such as w/c (water/cement

ratio), w/p (water/powder ratio) and c/p

(cement/powder ratio). In this, the powder

content p is the sum of both cement and filler

materials.

Results of isothermal hydrationtests

Figures 1 to 4 show the heat production rate q

(J/gh) as a function of time t, as experimentally

obtained in the isothermal hydration tests at

20°C (after elimination of the first 'wetting peak',[3, 4]). Similar curves are obtained for the tests at

10°C and 35°C. A more fundamental parameter

than time is the degree of hydration ·, defined as

the cement fraction that has reacted. Due to

difficulties in experimentally determining ·, in this

study the degree of reaction r is used as an

alternative parameter. defined as the fraction of

the heat of hydration that has been released at

any point during testing [3, 4, 5, 6].

In order to study the effect of the filler on the

heat development of the hydrating cement, some

numerical results are summarized in Table 2. In

these tables the mixture label starts with a

Mix 1 Mix 2 Mix 3 Mix 4

cement (g) 7.5 4.5 4.5 2.5

filler (g) - 3 3 5

water (g) 3.75 2.25 3.75 3.75

w/c 0.5 0.5 0.83 1.5

w/p 0.5 0.3 0.5 0.5

c/p 1.0 0.6 0.6 0.33

Table 1 : Composition of the powdermixes used in the isothermal tests.

Figure 1: Heat production rate formixtures with CEM I 42.5 R andlimestone filler at 20°C.

Figure 2 : Heat production rate formixtures with CEM I 52.5 and limestonefiller at 20°C.

Figure 3 : Heat production rate formixtures with CEM I 52.5 HSR LA andlimestone filler at 20°C.

Figure 4 : Heat production rate formixtures with CEM I 42.5 R andquartzite filler at 20°C.

109

number, referring to the composition given in

table 1. Furthermore, the mixture label refers to

the cement type and to the filler type (LF =

limestone filler, QF = quartzite filler). Table 2 gives

the maximum heat production rate qmax (J/gh) for

all different mixes at the three testing

temperatures.

DiscussionFrom the results given in the figures 1 to 4,

and from the corresponding results at different

temperatures, it can be clearly noticed that the

reaction mechanism of the hydrating cement in

some cases might be altered due to the presence

of the filler. This is clearly visible when looking

into the results obtained with CEM I 42.5 R

combined with limestone filler. When limestone

filler is added to the cement, the induction period

is shortened considerably and an extra hydration

peak occurs after about 22 hours for mixes 2 and

3, and after about 15 hours for mix 4. In this

latter case, the second, extra peak has a heat

production rate that is even slightly higher than

the first peak. At 35 °C this effect is even more

pronounced. This can lead to the conclusion that

the more limestone filler is added (and c/p is

decreasing) and the higher the testing

temperature is, the higher the production rate of

the extra peak is.

The alteration of the hydration reaction can

also be noticed for the tests with CEM I 52.5,

though much less pronounced. For CEM I 52.5

HSR LA an alteration of the hydration reaction

with the occurrence of a second hydration peak is

not found. Opposed to the limestone filler, the

quartzite filler does not seem to influence the

induction period, and, apart from a slight waving

of the curve of the hydration rate in some cases,

no extra peak is observed. Adiabatic tests carried

out on a few mixes of self-compacting concrete

confirmed the alterations in the reaction

mechanism found during the isothermal tests.

At this moment the driving force for the

change in hydration mechanism and the

occurrence of the extra hydration peak is not fully

clear. A few different hypotheses can be put

forward. In literature the influence of the filler

material, and in particular limestone filler, on the

hydration of the cement is in most cases

considered to be limited to the rate of the

reactions. Several authors mention that the

setting kinetic is improved, the dormant period is

reduced and the hydration process within the first

hours is accelerated [7, 8, 9, 10]. Kadri et al [11]

suppose that the filler particles promote sites of

heterogeneous nucleation to precipitate more or

less crystallized hydrates, and in this way

accelerate the hydration.

For the occurence of the second hydration

peak a few different hypotheses can be followed.

The noticed effect might be related to the

hydration of the C3A in the cement. Bensted [12]

Mixture qmax at 10°C qmax at 20°C qmax at 35°C

1/ 42.5R – LF 4.03 8.01 17.15

2/ 42.5R – LF 4.81 9.54 19.40

3/ 42.5R – LF 5.07 9.14 21.90

4/ 42.5R – LF 5.40 10.48 27.87

1/ 52.5 – LF 7.20 14.32 30.06

2/ 52.5 – LF 8.57 16.42 33.54

3/ 52.5 – LF 8.26 15.52 34.58

4/ 52.5 – LF 8.53 17.49 35.02

1/ 52.5HSR LA – LF 6.42 12.14 23.44

2/ 52.5HSR LA – LF 7.37 13.62 27.44

3/ 52.5HSR LA – LF 7.05 12.93 26.04

4/ 52.5HSR LA – LF 7.07 14.54 23.58

1/ 42.5R – QF 4.03 8.01 17.15

2/ 42.5R – QF 4.76 9.35 14.88

3/ 42.5R – QF 4.99 8.66 15.71

4/ 42.5R – QF 4.40 7.95 16.85

Table 2: Maximum heat production rate qmax (J/gh).

110

indicates that a C3A content of more than 12%

results in a visible extra hydration peak during a

20°C isothermal hydration test. In former

research [3, 4] it was observed that this hydration

peak, related to the transformation of ettringite

into monosulphate can also be noticed for

portland cements with lower C3A content (e.g.

7.5%) when tested at temperatures of 40°C to

50°C. This transformation might be activated by

the presence of the limestone filler, meaning that

the hydration peak can be found at lower

temperatures as well. For the case of portland

cement CEM I 52.5 HSR LA, no extra peaks can

be observed, nor for the pure cement, nor for the

mixes with the limestone filler. For this latter

cement, the C3A content is indeed very low

(2.5%), which might confirm the hypothesis.

Another approach is starting from the principle

that limestone filler is not inert and does not only

acts as an activator for some reactions, but

actually takes part in the hydration reactions.

Research carried out by Bonavetti et al [13]

revealed that in portland cements, the limestone

filler modifies the reactions. Three days after

mixing, monocarboaluminate was detected in the

hydrating paste. This compound was also found

by other researchers at different

times related to the C3A

content of the cement. This

hydration product is probably

supplied by the transformation

of monosulphoaluminate to

monocarboaluminate because

the last compound is more

stable. This reaction might

cause the second hydration

peak detected in the isothermal

hydration tests on CEM I 42.5 R

and CEM I 52.5. More details

about these different

hypotheses can be found in [14].

A more detailed investigation of

the maximum heat production

rates during the isothermal tests

(Table 2) leads to the finding that

this maximum rate also seems

influenced by the addition of the

limestone filler. Addition of the

limestone filler causes an increase

of the maximum hydration

production rate, this opposed to

the addition of quartzite filler,

where the maximum heat

production rate hardly changes

with varying c/p ratio. When

presenting the values of qmax at 20°C for the

cement-limestone combinations in a diagram as a

function of the c/p ratio, there seems to be a

linear relationship (Figure 5).

Hydration modelThe hydration rate of self-compacting concrete

can mathematically be described in an analytical

way, as mentioned in [14,15]. Comparison of the

proposed model with the experimentally obtained

results shows that the mathematical description

of the hydration proces leads to an accurate

prediction (Figure 6). Application of the hydration

model to concrete in adiabatic conditions also

shows a very good prediction capacity of the

temperature rise in the concrete.

PART II: MICROSTRUCTURE OF SELF-COMPACTING CONCRETE

IntroductionIn general, self-compacting concrete (SCC)

often incorporates limestone filler or fly ash to

ensure high fluidity and to reduce the

water/cement ratio (w/c). The major benefits in

using limestone in the SCC are both technical

Figure 5: Maximum heat production rate qmax at 20°C.

Figure 6: Hydration of mix 3, CEM I 42.5 R and LF.

qm

ax

q(J

/gh

)

111

and economical. Limestone filler was believed to

mainly produce a physical effect [16]. However,

several authors also reported that limestone filler

participates actively in the hydration process of

ordinary Portland cement rather than just acting

as a mere physical filler [17]. The aim of this part is

to experimentally examine the role of the

limestone powder in combination with Portland

cement by means of thermometric isothermal

conduction calorimeter, thermogravimetric

analysis and derivative thermogravimetric analysis

(TGA/DTA) and backscattering scanning electron

(BSE) image analysis. In order to do so, two SCC

mixtures with w/c 0.41 and w/c 0.48 both with

water/powder ratio (w/p) 0.27 were tested and

compared to a traditional cement paste (TC) with

w/c 0.48. Based on the experimental results, a

module has been added in the HYMOSTRUC3D[18,19] cement hydration microstructural model in

order to simulate the hydration and

microstructure of SCC. The simulation results

were discussed and compared with experiments.

Hydration and microstructuraldevelopment of SCC

The mix proportions of two SCC and one TC

cement pastes are listed in table 3. The hydration

process of self-compacting cement paste and

traditional cement paste was studied by

thermometric isothermal conduction calorimetry.

The development of microstructure was examined

by scanning electron microscopy and

thermogravimetric analysis and the derivative

thermogravimetric analysis. More details on the

experiments about SEM, TGA/DTA can be found

in [20].

Heat releaseThe rates of heat evolution of all cement

mixtures were measured using 10g samples and

appropriate amounts of mixing water. The heat

evolution and the degree of hydration of 3

mixtures are plotted in Figure 7. The degree of

hydration of SCC was calculated from the

hydration model developed by [4,14] and the

degree of hydration of TC was calculated

according to [18].

From Figure 7 left, the samples of SCC made

with limestone filler (SCC01 and SCC02) shows

higher heat release than traditional cement paste,

TC SCC01 SCC02

Portland cement I 52.5 350 400 400

Water 165 165 192

Limestone powder 200 300

Glenium 51 (liter) 3.2 2.7

Total powder content 350 600 700

Water/powder ratio 0.48 0.27 0.27

Water/cement ratio 0.48 0.41 0.48

Table 3: Mix proportions of the cement paste (kg/m3).

Figure 7: (left) Heat release of 3 different mixtures at 20°C in the first 100 hours and(right) the degree of hydration of 3 different mixtures in the first 200 hours at 20°C.

q(J

/gh

)

g

112

especially in the first 24 hours. The mixtures

made with limestone filler show a shorter

dormant stage and more rapid heat release than

mixtures without limestone filler. An extremum

was found from both SCC mixtures at 12 hours.

The degree of hydration curve (Figure 7 right)

also indicates the rapid chemical reaction of SCC

in the early stage. For the detailed discussion on

the heat release and hydration of SCC can be

found in [4,14,20].

TGA/DTA measurementsThermal investigation by thermogravimetric

analysis and derivative thermogravimetric analysis

can be used to determine the phase composition

in the cement paste. Moreover the TGA curve can

quantitatively provide the mass loss of each

individual phase [21]. Details on the calculation of

thermal decomposition of two SCC samples at

750°C at curing age of 28 days were presented

in[22] and the results are summarized in Table 4.

As earlier discussed in [22], if comparing the

TGA analysis and the theoretical calculations, the

weight loss from the TGA analysis is almost equal

to the weight loss from the theoretical

calculations at 750°C. The main part of the

weight loss at this temperature is due to

decarbonation of the limestone [21]. Thus, it can

be concluded that almost no limestone filler

participated in the chemical reaction during

cement hydration. The limestone powder acts

only as filler in the SCC. This is very important for

developing the hydration microstructural model

for SCC.

BSE image analysisThe backscattering electron images from

samples SCC02 and TC at the ages of 7 days are

shown in Figure 8. From image analysis it is found

that the interface between limestone and

hydrates is quite porous. The evolution of total

porosity and CaCO3 is shown in Figure 9. It can

be found that the amount of CaCO3 almost did

not change during 28 days of hydration. This also

agrees with the results observed by TGA.

TGA analysis Theoretical (mg) calculation (mg)

MIX02 2.98 2.86MIX03 3.78 3.63

Table 4: Weight loss of SCC mixturesfrom TGA and theoretical calculations.

Figure 8: Comparison of BSE images for SCC02 (left) and TC (right) at age of 7 days.

porous interface

limestone

Figure 9: The evolution of total porosity and CaCO3.

%

113

Microstructural model of SCCSimulation of the hydration process and the

microstructure of SCC containing limestone filler

starts from the cement and limestone particle size

distribution. The simulated cubic body of paste

was 100 µm3. The total amount of distributed

cement particles and limestone particles in the

cube is of the order of 125 000 – 180 000. In the

microstructural model, the limestone powder

does not react throughout the hydration process,

in accordance with the experimental results.

Figure 10 shows the degree of hydration of

samples SCC01 and TC, on the one hand as

calculated from experiments on SCC containing

limestone powder, and on the other hand from

the computer simulation model. A good

agreement is found between experiments and

simulations.

The simulated 3D microstructure of SCC02

and TC at the initial stage are shown in Figure

11. The white particles are limestone powder and

the other colours are cement particles. Using the

algorithm developed by [19], the capillary porosity

was calculated from the simulated cement paste

at different hydration stage. The results of

capillary porosity from the simulations and from

the SEM image analysis are presented in Figure

12. Despite the variation of experiments, a good

agreement is found between the simulation

model and the experiments obtained by SEM

image analysis.

Figure 13 shows the 2D simulated

microstructure at hydration stage of 0.60. The

Figure 10 : Comparison of degree of hydration from experiments and fromcomputer simulation model for the samples SCC01 (left) and TC (right).

Figure 11 : 3D simulated microstructure of SCC02 (left) and TC (right) at initial stage.

limestone

cement

Figure 12: Comparison of capillaryporosity from computer simulationmodel of SCC and experiments from BSE analysis.

%

114

particles with dark grey colour are limestone

powder. The calculated porosity of SCC02 and TC

at this hydration stage is 10% and 17.4%,

respectively. From Figure 13, it is noted that

limestone powder fills up the empty pore.

However, the porosity around the limestone filler

particles also is much bigger than the porosity

around the hydrated cement particles. This is

consistent with the experiments where the

interface between limestone filler and hydrates is

quite porous, even on the samples at the age of

28 days hydration.

PART III : EXPERIMENTAL DURABILITY EVALUATION OF SELF-COMPACTING CONCRETE WITH LIMESTONE FILLER

IntroductionAlthough SCC is a very promising cementitious

material, the actual application of SCC might be

somewhat riskful due to a lack of knowledge

concerning its actual durability. The degradation

mechanisms of a cementitious material are

greatly influenced by the permeability of the

material for potentially aggressive media and

there is an important interaction between ‘pore

structure’, ‘transport mechanism’ and

‘degradation’. The permeability itself is strongly

influenced by the pore structure of the material.

Furthermore, the ongoing degradation process

might have an influence on the pore structure of

the material. As the pore structure might be

different for SCC in comparison with traditional

concrete, due to the difference in composition,

some changes in durability behaviour might

occur. At this moment however, it is unclear how

significant these differences will be with regard to

the concrete in practice. A fundamental bottle

neck in this discussion is the lack of fundamental

insight in the transport behaviour of potentially

aggressive media in SCC. As the concept of SCC

is totally different from traditional concrete,

traditional models can not be extrapolated as

such without any verification. This is already

illustrated by the experimental and theoretical

investigation of hydration, carried out at the

Magnel Laboratory for Concrete Research.

According to [14] the hydration models valid for

traditional concrete have to be altered when

applied to SCC.

The transport mechanism in SCC, as well as its

relation with the durability of SCC is studied in

the Belgian research project ‘Transport properties

of potentially aggressive substances in self-

compacting concrete and relation with durability’,

sponsored by the Belgian Fund for Scientific

Research Flanders (F.W.O.-Vl.). The research

project is a cooperation between the Magnel

Laboratory for Concrete Research of Ghent

University, the Catholic University of Leuven, and

the Royal Military Academy of Brussels. The

ongoing research project has the following main

objectives: 1) theoretical and experimental study

of the transport behaviour of potentially

aggressive media in SCC; 2) experimental study of

the durability of SCC, and theoretical correlation

with the fundamental transport mechanisms.

Both objectives should lead to a better and

more fundamental knowledge of the durability

behaviour of the new cementitious material. In

this paper an overview of the first results on three

SCC mixtures and one traditional concrete (TC)

will be presented. An overview of the design and

methodology used for this project is given in[23,24].

Mixture proportionsIn the first stage, 4 concrete mixtures are

studied: 3 SCC and 1 TC. In the SCC mixtures, a

Figure 13: (left) 2D structure of SCC02 at a degree of hydration of 0.62, porosity10.0% (right) 2D structure of TC at a degree of hydration of 0.62, porosity 17.4%.

limestone

porousinterface

115

constant amount of fine particles (cement and

limestone filler) is considered: 600 kg/m3; as well

as a constant amount of sand and gravel,

respectively 853 kg/m3 and 698 kg/m3. The

amount of water is varied in order to obtain a

W/C ratio of 0.40, 0.46 and 0.55. For all mixtures

the same type of cement is used, namely Portland

cement CEM I 42.5 R. The amount of

superplasticizer (polycarboxylic ether) was

determined in order to obtain a suitable

flowability without segregation. Also the values

for the V-funnel (values between 5s and 10s), air

content (values between 1.5% and 3.0%) and

the U-box requiring self-levelling were measured.

For the TC mixture the same amount of cement

and water as for the SCC mixture 1 is taken. In

Table 5 the mixture proportions are given

together with the slump flow and the

compressive strength at 28 days measured on

concrete cubes 150 mm. Besides comparing

traditional concrete to self compacting concrete,

the influence of the W/C ratio is studied.

TRANSPORT PROPERTIESThe degradation mechanisms are greatly

influenced by the permeability of a material for

potentially aggressive substances. The penetration

of liquids and gasses in SCC is studied firstly by

means of basic tests, which should lead to the

determination of the fundamental parameters

governing the penetration behaviour of liquids

and gasses in SCC.

Water permeability Three concrete cubes 150 mm were stored in a

climate room at 20 ± 2°C and at least 90% r.h.

Until the age of 90 days. Then, from the centre

of each cube, one core, diameter 80 mm and

height 25 mm, was taken. Afterwards, these

cores were vacuum saturated in water and put in

the testing device. Water is present at the upper

side of the specimen and on the lower side there

is no water. Hence the pressure head is known

and the coefficient of permeability k [m/s] can be

calculated.

Capillary suctionThree concrete cubes 150 mm were stored in a

climate room at 20 ± 2°C and at least 90 % R.H..

At the age of 28 days, these cubes were stored

for two weeks at 35 ± 5°C and 40 ± 3% R.H..

The test was performed in an environment at 20

± 3°C and 60 ± 3% R.H.. The cubes were placed

on stable supports in water, so that the water

level is 5 ± 1 mm above the lower face of the

specimen. After 3, 6, 24, 72 and 168 hours the

mass of the specimens is determined. From these

results the sorptivity S [g/h] can be calculated.

Water absorption by immersion Three concrete cubes 100 mm were stored in a

climate room at 20 ± 2°C and more than 90%

R.H.. At the age of 28 days, these cubes were

stored for one week at 20 ± 3°C and 60 ± 3%

R.H.. Then the cubes are immersed in water until

the change in mass after 24 hours is less than

0.1%. The cubes are then placed in an oven at a

temperature of 105 ± 5°C until the difference in

mass after 24 hours is less than 0.1% and the

water absorption A [% of dry mass] was

determined.

Water vapour diffusion This method is based on DIN 52615

‘Bestimmung der Wasserdampfdurchlässigkeit von

Bau- und Dämmstoffen’. One cube 150 mm was

SCC1 SCC2 SCC3 TC1

CEM I 42,5 R [kg/m3] 360 360 360 360

limestone filler [kg/m3] 240 240 240

water [kg/m3] 165 144 198 165

river sand 0/5 [kg/m3] 853 853 853 640

river gravel 4/14 [kg/m3] 698 698 698 1225

superplasticizer [l/m3] 2.3 3.6 1.8

W/C [-] 0.46 0.40 0.55 0.46

W/(C+F) [-] 0.27 0.24 0.33 0.46

slump flow [mm] 685 740 820

compressive strength [N/mm2] 55 69 46 48

Table 5: Mix design.

116

made and stored in a climate room at 20 ± 2°C

and at least 90% R.H.. At the age of at least 28

days four cores with diameter 80 mm and height

20 mm were taken from this cube and further

stored at 20 ± 2°C and at least 90% R.H. until

the age of six months. The samples are put above

a reservoir with a saturated solution of a

hygroscopic salt NH4H2PO4 in water giving a layer

of air with a constant R.H. of 93% between the

sample and the water. The sample and the

reservoir are then put in a climate room at 20 ±

3°C and 60 ± 3% R. H.. In this way a constant

gradient of humidity has been created. The mass

of the testing device is measured on a regular

basis. This makes it possible to calculate the

decrease in mass of the water under the sample.

The water vapour diffusion coefficient D [m2/s]

can be calculated by analyzing the mass decrease

as a function of time.

Gas permeability The transport of oxygen is studied

fundamentally by means of a gas permeability

measuring instrument according the specifications

of RILEM TC 116-PCD "Permeability of concrete

as a criterion of its durability" [25]. One prism 400

x 400 x 100 mm is made and stored in a climate

room at 20 ± 2°C and more than 90% R.H.. At

the age of 28 days, three cores of 150 mm

diameter were drilled and from the centre of

these cores, samples with 50 mm height were

taken. Until the beginning of the test, at the age

of three months, the samples are stored in a

climate room at 20 ± 2°C and more than 90

%R.H.. In this paper, only the results of the tests

with an inlet pressure of 3 bar and at saturation

degree of 0% will be discussed. The drying of the

samples is performed according a fixed procedure

based on [26]. The apparent gas permeability kapp

[m2] can be calculated by means of a relationship

based on the Hagen-Poiseuille relationship for

laminar flow of a compressible fluid through a

porous body with small capillaries under steady-

state conditions [27,28].

Experimental results - DiscussionThe results of the tests on transport properties,

described above, are gathered in Table 6.

In general, the water permeability, water

absorption by immersion and the sorptivity of

SCC1 are compared to TC1 somewhat lower, but

the difference is rather small. The coefficient of

water vapour diffusion of SCC1 is lower than the

coefficient of TC1. The somewhat better

transport properties for liquid transport may be

due to the difference in porosity between SCC

and TC. A rather good relationship between the

capillarity porosity and the water permeability has

already been demonstrated [29]. The apparent gas

permeability of TC1 is about 5.5 times higher

than of SCC1. The much lower gas permeability

of SCC can also be noticed for the other SCC

mixtures.

Lowering the w/c ratio by decreasing the

amount of water decreases significantly the

values of the water transport properties. The

higher the amount of water, the bigger the

surplus of water not used by the hydration

process, causing a higher capillary porosity. Also

the gas permeability increases with an increasing

w/c ratio.

DURABILITYFrequently occurring degradation mechanisms

in which transport of aggressive substances is of

importance, are studied by means of several

experiments. Firstly, carbonation and chloride

penetration are investigated by means of tests

using special climate chambers and by immersion

in chloride containing solutions. Chloride

diffusivity was tested by means of an accelerated

method based on electrochemical principles.

Secondly, degradation mechanisms more

aggressive to the concrete itself were

investigated: physical degradation (frost, and frost

in combination with de-icing salts) and chemical

degradation (acid attack). The physical

degradation caused by carbonation and by

TC1 SCC1 SCC2 SCC3

water permeability K [10-11 m/s] 0.98 0.91 0.75 1.20

sorptivity S [g/h] 1.32 1.30 1.18 1.51

water absorption A [% of dry mass] 4.90 4.80 4.20 6.30

water vapour diffusion coefficient D [10-7 m/s] 2.95 2.00 1.71 3.03

apparent gas permeability kapp [m2] 3.018 0.544 0.271 1.382

Table 6: Transport properties.

117

freezing and thawing has also been evaluated by

means of non-destructive techniques: dynamic

and ultrasonic measurements.

Carbonation Two concrete cubes 100 mm were stored in a

climate room at 20°C ± 2°C and more than 90%

R.H.. At the age of 28 days, an epoxy coating

was applied to all surfaces except of the surface

that will be exposed to CO2. The cubes were

permanently stored in a carbonation room at

20°C, 60% R.H. and 10 vol.% CO2. At regular

times (8, 12, 16, 20 and 24 weeks) the

carbonation depth was examined experimentally.

At each time a slice with a thickness of 1 cm was

sawn from each specimen and was sprayed with

a phenolphtalein solution in order to determine

the carbonated zone. After sawing the slices, the

remaining concrete specimen was covered again

with carbon resisting epoxy coating and the

treatment was continued. After sawing a slice,

the carbonation depth is determined at 10 points

and the mean value is calculated. The

carbonation of concrete can be considered as a

Fickian problem and a constant value A,

depending on the diffusion resistance of the

material, can be derived.

Chloride penetration The CTH test method is used, which is based

on an electrical field and gives results in a few

days. This electrical field is applied across the

specimens (3 specimens with height 50 mm and

diameter 100 mm are tested at the same time)

and forces the chloride ions to migrate into the

specimen. After a certain test duration, the

specimen is axially split and a silver nitrate

solution is sprayed on the freshly split section.

The chloride penetration depth can then be

measured after which the chloride diffusion

coefficient D [m2/s] can be calculated. More

details about preconditioning of the sample, the

catolyte and the anolyte solution, the

temperature range, the applied voltage, the test

duration and the calculation method, can be

found in NT BUILD 492. The test specimens were

drilled out of cubes 150 mm at the age of 28

days. Until that moment the cubes are stored in a

climate room at 20°C and more than 90 % R.H.

and after drilling, the cores were placed back in

the climate room. The tests were carried out at a

concrete age of 28 and 90 days.

Freezing and thawing incombination with de-icing salts

The test method is performed according the

Belgian document NTN 018. 150 mm cubes were

made and stored in a climate room at 20 ± 2°C

and more than 90% R.H.. At the age of 28 days,

cores of 100 mm diameter and 100 mm height

were taken and 3 cylinders of each mixture were

prepared for the test. After covering the original

free surfaces of the specimens with a 5 mm thick

layer of a 3% sodium chloride solution, the

specimens went through 28 freeze/thaw cycles.

The temperature in each cycle (24 hours) varied

between 20 ± 2°C and -18 ± 2°C. After 7, 14, 21

and 28 cycles, scaling [mass loss, kg/m2] was

measured.

Acid attack One prism 400 x 400 x 100 mm is made and

stored in a climate room at 20 ± 2°C and more

than 90% R.H.. At the age of 10 months, nine

cores of 80 mm diameter were drilled from the

prisms for each concrete mixture and from the

centre of these cores, samples with 71 mm height

were taken. Until the beginning of the test, at

the age of 1 year, the samples are stored in a

climate room at 20 ± 2°C and more than 90%

R.H.. Three cores of the specimens were placed in

a solution of 1.5 g/l of sulphuric acid (H2SO4) in

water (pH = 1), three cores were placed in a

solution of 30 g/l acetic acid (CH3COOH) and 30

TC1 SCC1 SCC2 SCC3

carbonation A [-] 15.5 19.9 11.9 29.9

chloride diffusion coefficient D at 28 days [10-12 m2/s] 14.2 11.5 10.3 16.9

chloride diffusion coefficient D at 90 days [10-12 m2/s] 8.1 8.2 5.8 11.0

mass loss after 7 freezing-thawing cycles [kg/m2] 1.91 0.15 0.04 0.75

mass loss after 28 freezing-thawing cycles [kg/m2] 5.04 5.01 0.44 9.21

Table 7: Durability tests.

118

g/l lactic acid (C3H6O3) in water (pH = 2.5), and

three cores were placed in water, all at a

temperature of 20°C. The variation of mass, due

to immersion, is followed by weighing the

specimens every week. The samples stay

immersed for a period of six weeks. After this

test, the samples are taken out from the solutions

and are put for two weeks in a climate room at

20 ± 2°C and more than 90 % R.H.. They are

prepared for a compression test in order to be

able to determine the remaining compressive

strength due to deterioration.

Experimental results - discussion The results of the tests on durability, described

above, are gathered in Table 7. SCC1 exhibits a

higher carbonation depth. At 28 days, the

chloride diffusion of SCC1 is lower than the

chloride diffusion of TC1. At 90 days, the chloride

diffusion is more or less the same. After 28

freezing-thawing cycles in combination with de-

icing salts, SCC1 and TC1 have an equivalent

mass loss. However at 7 days, the mass loss of

SCC1 is much lower. The mass loss in time

increased more rapidly for SCC1 than for TC1 for

immersion in a solution of lactic and acetic acid in

water. The mass loss of TC1 is bigger than that of

SCC1 for immersion in a solution of sulphuric

acid in water [30].

The carbonation depth and the chloride

diffusion increase with increasing w/c ratio.

Increasing the w/c ratio increases the mass loss

due to freezing and thawing in combination with

de-icing salts. The mass loss increases with an

increasing w/c ratio for immersion in a solution of

lactic and acetic acid in water. When the samples

are immersed in a solution of sulphuric acid in

water, the biggest mass loss is noticed for the

lowest w/c ratio [30].

Non-destructive testing has been performed

on SCC1 and TC1 subjected to carbonation and

freezing and thawing. For both mixtures, due to

carbonation no significant decrease of modulus

of elasticity was noticed. Due to freezing and

thawing there is a decrease in modulus of

elasticity, which was bigger for TC1 than for

SCC1.

Concerning durability of self-compacting

concrete, in relation with microstructure and

transport properties, more details can be found

in [31,32].

CONCLUSIONSIn some cases, the reaction mechanism of the

Portland cement is clearly influenced by the

addition of the limestone filler. The induction

period is shortened and an extra heat production

peak sometimes occurs, even at the lowest

testing temperatures. These phenomenons are

not found when a quartzite filler is used in the

mixes.

The heat production rate of the second peak is

clearly influenced by the presence of the

limestone filler. The higher the amount of filler

and the higher the testing temperature, the more

pronounced the peak is. A linear dependency of

the maximum heat production rate on the c/p

ratio can be accepted in a first approximation.

The hydration process and the microstructure

of SCC were investigated by heat release,

TGA/DTA and BSE imaging technique. It was

confirmed that the limestone powder in SCC acts

as accelerator during cement hydration at early

stage. However, from BSE image analysis and

TGA analysis, the total amount of limestone

powder did not change throughout the

hydration. Based on HYMOSTRUC3D model, the

microstructure of SCC was simulated and the

pore structure was analyzed. The evolution of

porosity simulated from the model agrees well

with the experiments from the BSE image

analysis. Further research will be paid on the

percolation of capillary porosity of SCC.

A first insight in the transport properties and

the durability behaviour of SCC has been

achieved. Further experimental research is going

on and it will be verified whether traditional

models concerning transport properties and

durability can be applied to the new cementitious

material.

REFERENCES

1. NOGUCHI, T. and MORI, H., State-of-the-artreport : Evaluation of fresh properties ofself-compacting concrete in laboratory andon site, in: OZAWA, K., OUCHI, M. (eds.),Proceedings of the international workshopon self-compacting concrete, Kochi 1998,p. 97-110.

2. OKAMURA, H. and OUCHI, M., Self-compacting concrete. Development, presentuse and future, in: SKARENDAHL, A.,PETERSSON, O. (eds.), Proceedings of thefirst International RILEM Symposium onSelf-compacting Concrete, Stockholm 1999,p. 3-14.

119

3. DE SCHUTTER, G. and TAERWE, L., Generalhydration model for portland cement andblast furnace slag cement, Cem Concr Res25 (3) (1995) 593-604.

4. DE SCHUTTER, G. Fundamental andpractical study of thermal stress inhardening massive concrete elements,Doctoral thesis (in Dutch), Ghent University,1996.

5. Van BREUGEL, K., Simulation of hydrationand formation of structure in cement basedmaterials, Doctoral thesis, TechnicalUniversity Delft, 1991.

6. Van BREUGEL, K., Numerical simulation ofhydration and microstructural developmentin hardening cement-based materials –theory, Cem Concr Res 25 (2) 1995, 319-331.

7. BILLBERG, P. Influence of filler characteristicson SCC rheology and early hydration, in:OZAWA, K., OUCHI, M. (eds.), Proceedingsof the Second International RILEMSymposium on Self-compacting Concrete,Tokyo 2001, p. 285-294.

8. SHARMA, R.L. and PANDEY, S.P., Influenceof mineral additives on the hydrationcharacteristics of ordinary portland cement,Cem Concr Res 29 (9) 1999, 1525-1529.

9. SARI, M., PRAT, E., LABASTIRE, J.F., Highstrength self-compacting concrete - Originalsolutions associating organic and inorganicadmixtures, Cem Concr Res 29 (6) 1999,813-818.

10. Kadri, E.H., Aggoun, S., Duval, R.,Influence of grading and diameter size ofadmixture on the mechanical properties ofcement mortars, in: Bilek, V., Kersner, Z.(eds.), Proceedings of the InternationalSymposium on Non-Traditional cement andconcrete, Brno 2002, p. 306-313.

11. Kadri, E.H., Duval, R., Effect of ultrafineparticles on heat of hydration of cementmortars, ACI Mater J 99 (2) 2002, 138-142.

12. Bensted, J., Some applications ofconduction calorimetry to cementhydration, Adv Cem Res 1 (1) 1987, 35-44.

13. Bonavetti, V.L., Rahhal, V.F., Irassar, E.F.,Studies on the carboaluminate formation inlimestone filler-blended cements, CemConcr Res 31 (6) 2001 853-859.

14. Poppe, A.M., Influence of fillers onhydration and properties of self-compactingconcrete, Doctoral thesis (in Dutch), GhentUniversity, 2004.

15. POPPE, A.M., DE SCHUTTER, G. ‘Analyticalhydration model for filler rich self-compacting concrete’, Journal of AdvancedConcrete Technology, Vol. 4, No. 2, June2006, 259-266.

16. VUK, T., et al. The effects of limestoneaddition, clinker type and fineness onproperties of Portland cement, Cement andConcrete Research, vol. 31 (1) 2001 pp.135-139.

17. PERA, J., et al, Influence of sinely roundlimestone on cement hydration, Cementand Concrete Research, vol. 21 (2) 1999 pp.99-105.

18. Van BREUGEL, K. ‘Simulation of Hydrationand Formation of Structure in HardeningCement-Based Materials’, PhD thesis, DelftUniversity of Technology, The Netherlands,1991.

19. YE, G. ‘Experimental study and numericalsimulation of the development of themicrostructure and permeability ofcementitious materials’, PhD thesis, DelftUniversity of Technology, The Netherlands,2003.

20. YE, G., LIU, X., DE SCHUTTER, G., POPPE,A.M., TAERWE, L., ‘Influence of limestonepowder used as filler in SCC on hydrationand microstructure of cement pastes’,Cement & Concrete Composites, 29, 2007,94-102.

21. BAZANT, Z.P., KAPLAN, M.F., Concrete athigh temperatures. Longman – Addison-Wesley, London, 1996.

22. YE, G., LIU, X., DE SCHUTTER, G., POPPE,A.M. and TAERWE, L., VANDEVELDE, P.,Phase distribution and microstructuralchanges of Self-Compacting Concrete atElevated Temperature submitted to Cementand Concrete Research 2005.

23. DE SCHUTTER, G., AUDENAERT, K., BOEL,V., VANDEWALLE, L., DUPONT, D.,HEIRMAN, G., VANTOMME, J.,D’HEMRICOURT, J., Transport properties inself-compacting concrete and relation withdurability: overview of a Belgian researchproject; 3rd international conference onSCC, 17/8/2003-20/7/2003, Reykjavik,RILEM - 2003. p. 799-807.

24. BOEL, V., AUDENAERT, K., DE SCHUTTER,G., HEIRMAN, G., VANDEWALLE, L.,DESMET, J., VANTOMME, J., ‘Transportproperties of self-compacting concrete withlimestone filler or fly ash’, accepted forpublication in Materials and Structures

120

25. RILEM, T.C., 116-PCD: Permeability ofConcrete as a Criterion of its Durability,Recommendations; Materials andStructures, Vol. 32, April 1999, pp 174-179

26. CARCASSÈS, M., ABBAS, A., OLLIVIER, J.-P.and VERDIER, J.; An optimisedpreconditioning procedure for gaspermeability measurement; Materials andStructures, Vol. 35, January / February2002, pp. 22-27

27. KHAN, M.I., ASCE M.; Permeation of HighPerformance Concrete; Journal of materialsin civil engineering, pp. 84-92; January /February 2003

28. KOLLEK, J.J.; The determination of thepermeability of concrete to oxygen by theCembureau method - a recommendation;Materials and Structures, Vol. 22; 1989, pp.225 - 230

29. AUDENAERT, K., DE SCHUTTER, G.; Waterpermeability of SCC; Advances in concretethrough science and engineering, Evanston,2004, CD-ROM, RILEM, paper 39 in sessionDurability, pp.9

30. BOEL, V., AUDENAERT, K., DE SCHUTTER,G.; Acid attack of self compacting concrete;ICCRRR 2005 (the International Conferenceon Concrete Repair, Rehabilitation andretrofitting), Cape Town, 2005.

31. AUDENAERT, K., Transport mechanisms inself-compacting concrete and relation withcarbonation and chloride penetration,Doctoral thesis (in Dutch), Ghent University,ISBN 90-8578-084-5, 2006.

32. BOEL, V., Microstructure of self-compactingconcrete in relation with gas permeabilityand durability, Doctoral thesis (in Dutch),Ghent University, ISBN 90-8578-113-2,2006.

121

Firstly the subject of new

concrete technology was

addressed excellently by the

speakers, albeit constrained by

the scale of the subject. We are

on a materials and technique

development escalator that will transform

concrete over the next 10 - 20 years. There are

many choices and options and there will be more.

I find that prospect exciting and motivating, I

hope you do too.

Of the 10 papers given, apart from the

keynote address, 4 were on performance

enhancement, 3 on assessment, 2 on

sustainability and 1 on design.

The keynote address by Professor Dhir gave

an historical framework about cement and

concrete. That gave the context of the

symposium and highlighted concrete's

adaptability and wide appeal as well as global

availability.

In addition, milestones in concrete's evolution

were identified, namely:

• Strength and w/c ratio,

• Durability,

• Modern specifications,

• Success and failures,

Removal of historical conventions and

replacement by modern thinking and materials

based on knowledge and proven performance.

It is important to identify the movers for

change within the construction industry so that

one can build confidence based on performance

retained in the long term.

The term citadel is very appropriate - this

stands for fortress, guarding or dominating.

Breaking down the conventions by way of

responsible and proven alternatives giving options

has to be right if concrete is to progress.

As to the other papers, they were of

substance and ongoing relevance and a number

of points should be made.

The search for low energy alternative cements

is addressing a global problem requiring global

solutions. I hope the patenting of such options

does not militate against widespread use. The

progress is noted and welcomed.

The development of site evaluation

techniques is important for performance,

reassurance and liability issues. There are still

problems of definition, interpretation and

quantification. As there were in the early days of

ground radar.

It is desirable to control the development of

the mechanical characteristics of concrete; control

the chemistry (new cement), control the physics -

macro fibres and related additions.

There was a fascinating paper on

petrographics and the development of dynamic

diagnostic techniques; assessing what is occurring

as it is occurring. This underwrites pre-emptive

maintenance. It is clear that this is happening in

the field of corrosion monitoring. We have seen

such instrumentation developments in motorcar

management systems. Might there be a similar

trend in buildings and construction? At this point

in the proceedings I became aware we were

being treated to a different way of thinking;

applying future industrialised processes to

construction - shades of Sir John Egan; rapid

prototyping in 2D and 3D as well as mega-scale

rapid manufacturing - a real flavour of

tomorrow’s world.

The lateral thinking continued with flexible

concrete arches. This is thinking outside of the

box - or rather outside of the arch! The idea is

beautifully simple and could have widespread

application.

Sustainability was raised for the second time in

the proceedings, dealing with recycled

aggregates. The first mention was of course the

low carbon dioxide releasing cements. We saw

the creation of an instant quarry by way of the

explosive demolition of a structure that was then

recycled.

Economics are a key factor with respect to

sustainability; perhaps some form of tax incentive

might improve competitiveness. The absence of

financial wellbeing will require legislation and/or

regulation if we want widespread application.

CEN TC350 and the Code for Sustainable Homes

(published in December 2006) may well demand

materials with a recycled content.

Concrete flooring is an important sector and

problems during use are unwelcome and can be

CHAIRMAN’S SUMMARY

Professor Peter Hewlett, PhD, LLD, BSc, CChem, FRSC, FIM, HonFICT

John Doyle Group plc / BBA

122

very disruptive. Automatic retrieval systems in

warehouses require flat and stable floors. Water

control is a factor but so is optimised particle

packing. This development could have a

significant impact.

It seemed right to finish the symposium with

self-compacting concrete since it has made a

major impact on practical concreting. I am sure

there will be variations of this concept, not least

of all cost containment. Low slump is one

convention or citadel we should discard.

We should not overlook the relationship

between novel and innovative materials that, of

themselves’ have benefit but have to be

incorporated into a design. The cost of running a

building far exceeds the initial cost to build and

therefore the retained performance is essential

and the design implications cannot be ignored.

The demands of modern construction will

require modern materials and concrete can be

that material - there is no alternative which has a

comparable engineering scale of use. Concrete is

no longer a simple composite, it is an interactive

and complex composite that can be manipulated.

The new technologies are only limited by our

imaginations.

I thank the ICT for bringing us together at this

symposium. I thank the speakers for stimulating

us and for producing detailed written papers that

will be published in the 07/08 ICT Yearbook. I

thank the exhibitors and you the audience who

have been engaged and attentive.

The Quality Scheme forReady Mixed Concrete

QSRMC

for quality concrete only the best is good enough

QSRMC’s Governing Board comprises designers, specifiers,purchasers, users and producers of ready mixed concrete

QSRMC’s expert staff have the experience and competence to ensureboth producers and consumers benefit from the highest standards

specify QSRMC by name

benefit from effective assurance

QSRMC 1 Mount Mews High Street Hampton TW12 2SH � 020 8941 0273Visit the quality concrete website: www.qsrmc.co.uk

For our certification services covering other products/systems: www.cpcert.co.uk

The event of the year

2008!

Put these dates in your diary NOW and we’ll see you there.

Full details will be circulated soon.

• Update

• Network

• Socialise

ICT ANNUAL

CONVENTION2008

Chesford Grange Hotel, Warwickshire

31 March to 2 April 2008

PROGRAMME31 March

Members Annual Meeting

1 AprilTechnical Symposium

R&D IN CONCRETE TECHNOLOGY- AN INTERNATIONAL VIEW

Convention dinner

Quattro (UK) Ltd is able to offer a range of servicesfrom convenient locations in West London.

Whatever your needs from aggregates to wastemanagement we have a solution.

We are pleased to offer a ‘one-stop’ service from our Park Royal Depot providing:

• Ready Mixed Concrete, for delivery or collection

• Primary and Reclaimed Aggregates

• Waste Transfer Facilities

For further information please contact our Sales Team

At Quattro (UK) LtdRegency Street

Off Victoria RoadPark Royal

LondonNW10 6NR

Tel: 020 8838 2648Fax: 020 8838 2741

Email: [email protected]

123

ADVANCED CONCRETE TECHNOLOGY DIPLOMA:

SUMMARIES OF PROJECT REPORTS 2006

The project reports are an integral and important part of the ACT Diploma.

The purpose of the projects is to show that the candidates can think about a topic or problem ina logical and disciplined way. The project normally spans some six months. Significant advances can bemade and several of the projects have evolved into research programmes in their own right.

Summaries of a selection of projects submitted during the 2005-2006 course are given in thefollowing pages.

PROJECT TITLE: AUTHOR:

AN EVALUATION OF THE PERFORMANCE OF Amit DawneerangenHIGH VOLUME FLY ASH CONCRETE

AN APPROPRIATE PIN PENETRATION TEST TO ESTIMATE Kevin Lester THE COMPRESSIVE STRENGTH OF CONCRETE

THE EVALUATION OF THREE POLYPROPYLENE FIBRES, AVAILABLE Andries P J Marais IN SOUTH AFRICA, DETERMINING THEIR ABILITY TO REDUCE: PLASTIC SHRINKAGE CRACKING, PLASTIC SETTLEMENT AND BLEEDING

EVALUATING SINGLE STAGE JAW CRUSHER VS. MULTI STAGE Suresh Rao M.CRUSHERS PRODUCED NAVI MUMBAI AGGREGATES PERFORMANCE IN CONCRETE

ANALYSIS OF THE EUROPEAN STANDARD EN 196-1 ‘METHODS David RevueltaOF TESTING CEMENT. PART 1: DETERMINATION OF STRENGTH’: ASSESSMENT OF THE INFLUENCE OF CEN STANDARD SAND

THE EFFECT OF (LETHABO) COARSE FLY ASH GRINDING ON Theo RoelofszSTRENGTH DEVELOPMENT AND WORKABILITY WHEN USED IN FLY ASH ACTIVATED CEMENT

A PRACTICAL APPROACH TO ACCELERATED STRENGTH N SurjooTESTING AND QUALITY CONTROL OF CONCRETE

A full list of earlier ACT projects, dating back to 1971 when the individual project was introduced as arequirement for the Advanced Concrete Technology Diploma examination, was published in the 2000 - 2001edition of the ICT yearbook.

Copies of the reports (except those that are confidential) are held in the Concrete Information Ltd (CIL) Libraryand these can be made available on loan. Subscribers to the CIL’s information service, Concquest, may obtaincopies on loan, free of charge. Requests should be addressed to: Concrete Information Ltd, 4 Meadows BusinessPark, Blackwater, Camberley GU17 9AB

ICT members may address their requests to: The Executive Officer, Institute of Concrete Technology at the sameaddress. Copies can then be obtained from CIL free of charge.

124

AN EVALUATION OF THE PERFORMANCE OF HIGH VOLUME FLY ASH CONCRETE

By: Amit Dawneerangen

SUMMARYThis project is an evaluation of the

performance of concrete mixes containingconventional and high volumes of South Africanfly ash. The stated objective is to “determine themaximum percentage of fly ash that will stillproduce an acceptable concrete in terms ofstrength, strength gain and durability by includinghigh range water reducing admixtures andaccelerators.”

The base cement type is, interestingly, notsimply a CEM I Portland cement but a slagcement with around 20% slag (similar to CEMII/A-S). Consequently, mixing the base cementwith fly ash produces the equivalent of a ternaryblended cement with compositions falling outsideEN 197-1. The fly ash content ranges from 20%to 70% of the total cementitious content.Various concrete mixes were investigated atwater/ binder ratios of 0.67, 0.56 and 0.45(water/binder = water/ (Portland cement + slag+fly ash)).

Concrete was tested for setting time, waterreduction, strength development and durability(as indicated by measurements of water sorptivity,air and oxygen permeability and chlorideconductivity).

It was found that the 28-day cube compressivestrength reduced with increasing fly ash contentfor any given level of water/binder ratio.Consequently, an increase in binder content isrequired compared to the base cement concrete.It was concluded that the cost-optimum mix for aC35 MPa grade concrete contained a binder with60% fly ash.

The early age strength development of thehigh volume fly ash mixes (fly ash content greaterthan 35% of total binder) was slower than thatof mixes with more conventional levels of fly ash.In contrast the strength gain between 7 and 28days and between 28 and 56 days was improved.At water/binder ratios below 0.56 acceptablelevels of compressive strength were produced.

In terms of durability related parameters it wasnoted that the air permeability measurements(using a Torrent Permeability Tester) fell into the‘good’ or ‘excellent’ categories except when thefly ash content was 70% and the water/binderratio was 0.67. Oxygen permeability tests andwater sorptivity measurements confirmed thepoor performance of the high fly ash highwater/binder mixes. The chloride conductivity testresults were quite variable but were again ratedas good or excellent at levels of fly ash below70%.

The report highlights the need for effectivecuring of high volume fly ash concrete (HVFA) inorder to develop optimum strength growth anddurability. It also recommends that tests areundertaken on concrete structures to confirm thefindings of the laboratory study.

The overall conclusion is that “ With thecurrent shortage of materials in the South Africanconstruction industry, consideration could begiven to the use of HVFA with 60% fly ash whichdisplays good compressive strength development,improved setting times using an accelerator, goodair permeability when effectively cured and agood durability test index performance. All thesefactors suggest that HVFA made from SouthAfrican raw materials will produce sound, durableconcrete capable of providing the South Africanconstruction industry with a cost-effectivesolution to its current situation”.

125

AN APPROPRIATE PIN PENETRATION TEST TO ESTIMATE THE COMPRESSIVESTRENGTH OF CONCRETE

By: Kevin Lester

SUMMARYA very large percentage of the cost of

construction is the cost of formwork. Ifformwork can be removed early it is possible tomaximise the number of uses and so makesavings. To be able to strip formwork, theconcrete must have gained sufficient strength tosafely support itself and any loads that it may besubject to.

Whilst a number of tests exist few are suitablefor measuring early strengths up to 10 MPa.Such determinations occur frequently. Forinstance in South Africa shotcrete when used aslining must achieve a compressive strength of 5MPa within eight hours.

Drilling cores is not practicable at low concretestrengths and pullout methods such as the LOKtest have the disadvantage of hiring equipmentand consumable costs.

Alternatives such as the Windsor Probe andWindsor Pin system are not a practical propositionin South Africa. However the Hilti Gun and theSchmidt Hammer are used although the latterrequires a hard surface to produce a reboundreading.

Faced with this situation the author decided tobuild a simple instrument for comparativeevaluation. Called the Leicester Pendulum it wasbased on the principle of a defined mass fallingthrough a given height, driving a pin into theconcrete. The instrument produced the sameenergy for every test, was simple to make anduse and required few consumables.

The project compared the ability of the HiltiGun, Windsor Pin, Leicester Pendulum and theSchmidt Hammer (including the N-type Hammer)to estimate in situ compressive strengths below10 MPa.

The results of the investigation suggest thatthe Hilti Gun, Windsor Pin and LeicesterPendulum may be suitable for estimating the in situ strength of concrete between 2-10 MPa.The N-type Schmidt Hammer was not suitable,because it did not give rebound readings at lowstrengths.

126

THE EVALUATION OF THREE POLYPROPYLENE FIBRES, AVAILABLE IN SOUTH AFRICA,DETERMINING THEIR ABILITY TO REDUCE: PLASTIC SHRINKAGE CRACKING, PLASTICSETTLEMENT AND BLEEDING

By: Andries P J Marais

SUMMARYThis project was carried out to compare the

effectiveness of three polypropylene fibres toreduce the occurrence of:

• Plastic shrinkage cracking

• Plastic settlement

• Bleeding

The fibres used were:

• Un-fluorinated 32µm polypropylene fibre(fibre A)

• Fluorinated 32µm polypropylene fibre(fibre B)

• 18µm polypropylene fibre imported fromAmerica (fibre C).

The testing of the fibres was carried out bycasting concrete slabs 995mm x 180mm x 50mmthick and testing the slabs in a wind tunnel at awind speed of 9m/s and temperature of 30°C forfive hours.

The following measurements were recordedand analysed:

• Time for first crack to appear

• Total number of cracks

• Maximum crack width

• Settlement in the mould

• Bleeding observations and testing.

The conclusions reached were:

1. When evaluated for plastic crack reducingand settlement reducing properties, fibreA had reduced capabilities whencompared to fibres B and C.

2. When evaluated for plastic crack reducingand settlement properties, there wereindications that fibre C was the mosteffective although taking regard of thetest methods used the difference betweenfibres B and C was marginal.

3. All three fibres reduced bleeding of theconcrete by more than 12%. Howeverfibre C performed better than fibres A and B.

4. On balance and under controlledreproducible conditions it was possible torank the fibre types according to theirplastic crack and settlement reducing andbleeding properties.

5. the ranking showed: Fibre C (18µm) > fibre B (32µm Fluorinated) > fibre A (32µm Un-fluorinated)

127

EVALUATING SINGLE STAGE JAW CRUSHER VS. MULTI STAGE CRUSHERS PRODUCEDNAVI MUMBAI AGGREGATES PERFORMANCE IN CONCRETE

By: Suresh Rao M.

SUMMARYThis project examines the effects of aggregates

produced using different types of crusher, on theperformance of concrete. In the Navi Mumbai(New Bombay) area, the majority of smalleraggregate producers utilise single stage jawcrushers (‘Category I producers’), whereas asmaller number of large producers (‘Category IIproducers’) now use multi stage crushers,combining primary jaw crushers with secondarycone crushers and tertiary vertical shaft impactcrushers.

A laboratory test programme was undertakento examine the effects of these different methodsof aggregate production on concreteperformance. For each aggregate (maximumaggregate size 20 mm), a series of concrete mixeswere made at cement contents (Portland cementonly) varying between 300 kg/m3 and 430 kg/m3.The concrete workability was targeted at a slumpof 100 mm and the admixture dosage wasmaintained at a fixed level of 1% by weight ofcement. The water requirement and strengthdevelopment were measured together with theproportion of fine aggregate required to achievea stable mix and uniformly combined aggregategrading.

Examination of the aggregate shapecharacteristics indicated that whilst the elongationindices of both Category I and Category IIaggregates were similar, the flakiness index of theCategory I aggregates was lower.

The results of the concrete testing showed thatthe water demand, at any given cement content,for the Category II aggregates was up to 8%lower than for the Category I aggregates. Alsothe optimum proportion of fine aggregate wasreduced.

In terms of concrete compressive strength, theCategory II aggregates performed better. For a28-day cube compressive strength of 45 MPa, thereduction in cement content when using theCategory II aggregates was typically around 25kg/m3 . However, it was also noted that thisdifference in cement content for a given strengthwas more pronounced at the higher strength,higher cement content end of the range ofcompositions studied.

The author points out that further studiesshould be carried out to confirm these findingswhen applied to other concrete properties such asflexural or tensile strength. It was also suggestedthat improved methods for the shapeclassification of aggregates are needed in order tounderstand better the causes of the observeddifferences in concrete performance.

128

SUMMARYThe inspiration for this project came from an

observation that when Spanish cement producerschanged the source of the standard sand used toproduce samples for cement strength testing,there was a distinct drop in strength. This wasdespite the fact that the new source of sandfulfilled all the requirements for such a sand in EN196-1.

The project report describes a detailedstatistical analysis of the acceptance proceduresfor the CEN Standard Sand in order to determinetheir reliability.

A source of sand for use in producing mortarprisms for cement strength testing must satisfytwo procedures:

• A certification test performed annuallycomparing the strength obtained for agiven cement on mortar prisms made withthe test sand with those made using theCEN reference sand. This test is performedon 20 sets of prisms for each of threecements (One from each of the EN 197-1strength classes 32,5: 42,5: 52,5). Therepeatability of the tests must be less than2%, expressed as coefficient of variation.If this is the case the difference in meanstrength (D) using the test sand and thereference sand must not exceed 5% of themean strength using the reference sand.

ANALYSIS OF THE EUROPEAN STANDARD EN 196-1 “METHODS OF TESTING CEMENT.PART 1: DETERMINATION OF STRENGTH”: ASSESSMENT OF THE INFLUENCE OF CENSTANDARD SAND

By: David Revuelta

• A verification test performed monthlycomparing the performance of the sandproduced each month with that of thesand used in the initial certification test.The test is performed on 10 sets of prismsusing cement from a single strength class.An analysis similar to the certification testis undertaken and D cannot be greaterthan 2.5% more than twice in a 12 monthperiod of sand production.

A statistical model was developed to analysethe certification and verification criteria for astandard sand described in EN 196-1 resulting inan ‘Operating-Characteristic Curve of the Test’(similar to the well known O-C curves of a sampleplan). This tool was used to show that the EN196-1 acceptance criteria were robust and able todetect even slight difference between a new sandsource and the CEN reference sand.

However, it was concluded that there were stillsome slight deficiencies in the requirements ofthe standard:

• The probability of acceptance changedepending on whether the proposed sandgenerates a higher or lower strength thanthe CEN reference sand (ie: the criteria areasymmetric)

• The standard does not state clearly thedifference between conformity criteria andacceptance criteria.

129

SUMMARYWith expanding construction activity in South

Africa, supplies of classified fly ash for use incement manufacture are becoming stretched.There are, however, abundant resources of coarseunclassified fly ash. The objective of this projectwas to investigate whether a ground, coarse flyash could be used as a substitute for classified flyash in the production of Portland-fly ash cement.

A coarse fly ash from Lethabo power station inSouth Africa (typically around 45% retained on45µm, SSA 2230 cm2/g) was ground in alaboratory mill to a fineness similar to that of aclassified fly ash (typically around 10% retainedon 45µm, SSA 2800 cm2/g). This ground materialwas then blended with an ISO EN 197-1 CEM I42,5N Portland cement to produce a CEM IV/APortland-fly cement with 40% fly ash. Bothmortars and concrete were produced using thiscement together with similar cements producedusing classified fly ash and un-ground coarse flyash

The results of the experimental programmerevealed that the grinding process destroyed ahigh proportion of the glassy spheres in the flyash. It also narrowed the particle size distributionof the fly ash (but not to the extent seen inclassified ash). Furthermore the grinding processintroduced ‘gaps’ in the particle size distributionnot seen in either the classified ash or the un-ground ash. These changes were reflected in theincreased water demand of the cementcontaining the ground ash. The standardconsistence of the cement pastes increased by17% when the ground ash was substituted forclassified material. However, there was still adegree of reduction in water demand comparedto a CEM I cement.

THE EFFECT OF (LETHABO) COARSE FLY ASH GRINDING ON STRENGTHDEVELOPMENT AND WORKABILITY WHEN USED IN FLY ASH ACTIVATED CEMENT

By: Theo Roelofsz

No significant effect of ash treatment (grindingor classification) was seen in the measurements ofsetting time.

Standard mortar prism strength tests (ISO EN196-1) carried out at a constant w/c ratioindicated that, compared to the un-ground ash,the cement made using ground fly ash showed a10% improvement in 28-day strength to a levelcomparable with cement containing classifiedash.

A similar situation was seen in the strength ofconcrete made at a fixed water/cement ratio. Theconcrete slump was, as expected, reduced whenthe ground ash cement was used. This clearlyindicated that the increase in cement waterdemand caused by the ash grinding process fedthrough to the concrete.

The overall conclusion from this work was thatthe increased mortar and concrete strengthsobtained using cement containing the ground ashwere not sufficient to overcome the effects of theincreased water demand. Consequently, using aground unclassified fly ash for the manufacture ofPortland-fly ash cement was not seen as anacceptable solution to the shortage of classifiedfly ash.

130

SUMMARYThe speed of modern construction makes the

use of 7-day strength results to predict 28-dayvalues questionable because the concrete towhich it relates might well be covered andinaccessible. As a result there is a need for arapid method of strength testing that can be usedas a quality control tool.

In the UK the current method is contained inBS 1881 Part 112: 1983 but its adoption in SouthAfrica has hardly progressed.

The primary objective of this investigation wasto assess whether a simple accelerated testingmethod could be adopted in South Africa toaccurately predict the 28-day strength of concreteat an earlier age.

This investigation was limited to using the55°C accelerated testing method:

• Two different types of cement/blends(100% CEM1 42.5N and 50-50% blend ofa CEM1 42.5N and ground granulatedblastfurnace slag were used

• The concrete incorporated a water-reducing admixture

• One type of crushed coarse aggregatefrom different batches was used

• A blend of crusher sand and fine fillersand was incorporated

• The strengths ranged from 10 MPa to 35MPa.

Other aspects that were considered were:

• A simple and feasible method that couldbe used under most working environmentswhere water and electricity were readilyavailable

A PRACTICAL APPROACH TO ACCELERATED STRENGTH TESTING AND QUALITYCONTROL OF CONCRETE

By: N Surjoo

• Judgement on the suitability of the regimefor practicable conditions

• The flexibility of the regime so that smalldeviations from the working instructionwould not significantly change thestrength results

• A warm water method (55°C) where thetemperature could be easily maintained

• Temperatures used were modest in orderto minimise health and safety risks.

The project concentrated on acceleratedconcrete specimen at 1 and 3 days. Conclusionsthat may be drawn:

1. Accelerated testing is a reliable method forquality control and estimating strengths atlater ages but requires a greater degree ofcontrol.

2. Cement type correlations may changefrom time to time.

3. The system used is basically simplealthough proper temperature control isnecessary for repeatable results.

4. The size of specimens and use ofadmixtures do not appear to affect theestimated strength relationship.

5. Considerable cost savings can beobtained.

6. It is recommended that the method beused for routine quality control ofconcrete.

ICT RELATED INSTITUTIONS & ORGANISATIONS

ASSOCIATION OFCONSULTING ENGINEERSAlliance House12 Caxton StreetLondon SW1H 0QLTel: 020 7222 6557www.acenet.co.uk

ASSOCIATION OF INDUSTRIALFLOORING CONTRACTORS33 Oxford StreetLeamington SpaCV32 4RATel: 01926 833 633www.acifc.org.uk

ASSOCIATION OF LIGHTWEIGHTAGGREGATE MANUFACTURERSWellington StRipleyDerbyshire DE5 3DZTel: 01773 746111

BRE (BUILDING RESEARCHESTABLISHMENT) LTDBucknalls LaneGarstonWatford WD25 9XXTel: 01923 664000www.bre.co.uk

BRITISH BOARD OF AGRÉMENTP.O.Box 195Bucknalls LaneGarstonWatfordHerts WD25 9BATel: 01923 665300www.bbacerts.co.uk

BRITISH CEMENT ASSOCIATION4 Meadows Business ParkStation Approach BlackwaterCamberley GU17 9ABTel: 01276 608700www.bca.org.uk

BRITISH PRECAST60 Charles StreetLeicester LE1 1FBTel: 0116 253 6161www.britishprecast.org.uk

BSI STANDARDSBritish Standards House389 Chiswick High RoadLondon W4 4ALTel: 020 8996 9000www.bsi.org.uk

BRITPAVEBritish In-Situ ConcretePaving Association4 Meadows Business ParkStation Approach BlackwaterCamberley GU17 9ABTel: 01276 33160www.britpave.org.uk

CEMENT ADMIXTURES ASSOCIATION38a Tilehouse Green LaneKnowleWest MidlandsB93 9EYTel: 01564 776362

CEMENTITIOUS SLAG MAKERS ASSOCIATIONCroudace HouseGoldstone RoadCaterhamSurrey CR3 6XQTel: 01883 331071www.ukcsma.co.uk

CIRIAConstruction Industry Research& Information Association

174-180 Old Street London EC1V 9BPTel: 020 7549 3300www.ciria.org.uk

CONCRETE ADVISORY SERVICE4 Meadows Business ParkStation Approach BlackwaterCamberley GU17 9ABTel: 01276 607140www.concrete.org.uk

CONCRETE BRIDGE DEVELOPMENT GROUP4 Meadows Business ParkStation Approach BlackwaterCamberley GU17 9ABTel: 01276 33777www.cbdg.org.uk

CONCRETE INFORMATION LTD4 Meadows Business ParkStation Approach BlackwaterCamberley GU17 9ABTel: 01276 608770www.concrete-info.com

CONCRETE REPAIR ASSOCIATIONTournai Hall, Evelyn Woods RoadAldershot GU11 2LLTel: 01252 357835www.cra.org.uk

THE CONCRETE CENTRE4 Meadows Business ParkStation Approach BlackwaterCamberley GU17 9ABTel: 01276 606800www.concretecentre.com

THE CONCRETE SOCIETY4 Meadows Business ParkStation Approach BlackwaterCamberley GU17 9ABTel: 01276 607140www.concrete.org.uk

CONSTRUCT4 Meadows Business ParkStation Approach BlackwaterCamberley GU17 9ABTel: 01276 38444www.construct.org.uk

CORROSION PREVENTION ASSOCIATIONTournai Hall, Evelyn Woods RoadAldershot GU11 2LLTel: 01252 357834www.corrosionprevention.org.uk

INSTITUTE OF CORROSIONCorrosion HouseVimy CourtLeighton BuzzardBeds LU7 1FG Tel: 01525 851771www.icorr.org

INSTITUTE OF MATERIALSMINERALS & MINING1 Carlton House TerraceLondon SW1Y 5DBTel: 020 7451 7300www.materials.org.uk

INSTITUTION OF CIVIL ENGINEERSOne Great George StreetLondon SW1P 3AATel: 020 7222 7722www.ice.org.uk

INSTITUTION OF HIGHWAYS& TRANSPORTATION6 Endsleigh StreetLondon WC1H 0DZTel: 020 7387 2525www.iht.org

INSTITUTION OFROYAL ENGINEERSBrompton BarracksChathamKent ME4 4UGTel: 01634 842669

INSTITUTION OFSTRUCTURAL ENGINEERS11 Upper Belgrave StreetLondon SW1X 8BHTel: 020 7235 4535www.istructe.org.uk

INTERPAVEConcrete Block Paving Association60 Charles StreetLeicester LE1 1FBTel: 0116 253 6161www.paving.org.uk

MORTAR INDUSTRY ASSOCIATION38-44 Gillingham StreetLondon SW1V IHUTel: 020 7963 8000www.mortar.org.uk

QSRMC1 Mount Mews High Street, HamptonMiddlesex TW12 2SHTel: 020 8941 0273www.qsrmc.co.uk

QUARRY PRODUCTS ASSOCIATION38-44 Gillingham StreetLondon SW1V IHUTel: 020 7963 8000www.qpa.org

RIBARoyal Institute of British Architects66 Portland PlaceLondon W1B 1ADTel: 020 7580 5533www.architecture.com

SOCIETY OF CHEMICAL INDUSTRY14/15 Belgrave SquareLondon SW1X 8PSTel: 020 7598 1500www.sci.mond.org

UNITED KINGDOM ACCREDITATION SERVICE21-47 High StreetFelthamMiddlesex TW13 4UNTel: 020 8917 8400www.ukas.org.uk

UNITED KINGDOM CAST STONE ASSOCIATION4 Meadows Business ParkStation Approach BlackwaterCamberley GU17 9ABTel: 01276 608771www.ukcsa.co.uk

UNITED KINGDOM QUALITY ASH ASSOCIATIONRegent HouseBath AvenueWolverhamptonWV1 4EGTel: 01902 576 586www.ukqaa.org.uk

TheINSTITUTE OF CONCRETE TECHNOLOGY

4 Meadows Business Park, Blackwater, Camberley, Surrey GU17 9AB

Tel/Fax: 01276 37831Email: [email protected] Website: www.ictech.org

THE ICTThe Institute of Concrete Technologywas formed in 1972 from the Associationof Concrete Technologists. Fullmembership is open to all those whohave obtained the Diploma in AdvancedConcrete Technology. The Institute isinternationally recognised and theDiploma has world-wide acceptance asthe leading qualification in concretetechnology. The Institute sets higheducational standards and requires itsmembers to abide by a Code ofProfessional Conduct, thus enhancingthe profession of concrete technology.The Institute is a Professional Affiliatebody of the UK Engineering Council. In 2007 the ICT merged with theConcrete Society to become theprofessional wing of the Society whilstretaining its own identity.

AIMSThe Institute aims to promote concretetechnology as a recognised engineeringdiscipline and to consolidate theprofessional status of practising concretetechnologists.

PROFESSIONAL ACTIVITIESIt is the Institute's policy to stimulateresearch and encourage the publicationof findings and to promotecommunication between academic andcommercial organisations. The ICTAnnual Convention includes a TechnicalSymposium on a subject of topicalinterest and these symposia are wellattended both by members and non-members. Many other technicalmeetings are held. The Institute isrepresented on a number of committeesformulating National and InternationalStandards and dealing with policymatters at the highest level. TheInstitute is also actively involved in theeducation and training of personnel inthe concrete industry and those enteringthe profession of concrete technologist.

Date post:	28-Mar-2018
Category:	Documents
Upload:	vumien
View:	223 times
Download:	2 times

The INSTITUTE OF CONCRETE TECHNOLOGYict.concrete.org.uk/downloads/yearbooks/ict_2007.pdf · 3...

Documents