early age shrinkage cracking of fibre reinforced concrete

EARLY AGE SHRINKAGE CRACKING OF FIBREREINFORCED CONCRETEOlivér Fenyvesi - Zsuzsanna JózsaConventional steel reinforced concrete is often sensitive to cracking, and these cracks could be the result ofearly age shrinkage. Cracks can cause the corrosion of steel reinforcements reducing the lifetime of concretestructures. To reduce this problem fibre reinforced concrete (FRC) is one possible solution. Laboratory testshave been carried out according to the Austrian FRC technical specification, in particular with regard toearly age shrinkage cracking. Four different relatively thin and short fibre types were used: fabricated frompolypropylene, polyacrylnitrile and non alkaline resistant E-glass. In this paper the relationship between thedosage of fibres and early age shrinkage cracking tendency was tested and effectiveness of fibre types arecompared. During the laboratory tests compressive strength was also tested. In case of every fibre type, therelationship between dosage and crack tendency were linear. If we test a fibre type again but with a referencemix with higher crack tendency, the two linear curves will approach each other if dosage is increased.Finally, further possibilities are described with testing more fibre types and higher dosages in the future.Keywords: shrinkage, shrinkage cracking, early age shrinkage, FRC, polymeric fibres, glass fibres1. INTRODUCTIONIn concrete, mortar and cement paste shrinkage takes placefrom the very beginning. This is caused by water movementin the porous and rigid body. During the hydration of cement,while the cement paste is plastic, it undergoes a volumetriccontraction (autogenous shrinkage) which magnitude isof the order of one per cent of the absolute volume of drycement. However, the extent of hydration prior to setting issmall, and once a certain rigidity of the system has developed,the contraction induced by the loss of water by hydration isgreatly restrained (Neville, 1995). Withdrawal of water fromconcrete or mortar or cement paste stored in unsaturated aircauses drying shrinkage. A part of this drying shrinkage isirreversible and should be distinguished from the reversiblemoisture movement caused by alternating storage under wetand dry condition (Neville, 1995; Grube, 2003).Influencing factors of shrinkage in mix design:− cement content of paste− specific surface area of cement− fine aggregate content (under 0.125 particle size)− specific surface area of fine aggregate− water-cement ratio− total aggregate content− type of aggregate− water absorption capacity/water content of aggregate− applied admixtures− compacting rate of paste− porosity− other added components e.g. fibres.Shrinkage of concrete depends on the temperature ofconcrete and its surroundings, on relative humidity and on thevelocity of air movement as well as the curing and compositionof the concrete.The importance of shrinkage in structures is largely relatedto cracking. Time has a two-fold effect from this point ofview: the strength increases, thereby reducing the probabilityof cracking, but on the other hand, the stress induced byshrinkage also increases. If stress reaches the tensile strengthof concrete, cracks appear on the structure or specimen (CEB,1992; Neville, 1995).Fibre reinforced concrete (FRC) is a possible solution toavoid early age shrinkage cracking (Balázs, Lublóy, 2007).Earlier in Hungary, asbestos and celluloid fibres were usedto produce fibre reinforced cementitious products (mainlyroofing elements and pipes) under the trademark “Eternit”.These products have disappeared from the market, but theycan be found in old buildings. Asbestos fibres have very smalldiameter, often smaller than 10 µm, and provide very highstrength, but may cause serious health problems (e.g. lungcancer on inhalation). That is the reason why, from a medicalpoint of view, the smallest allowed diameter of fibres is 7 µm,and on the market there are fibres with a diameter between9 and 500 µm. Nowadays there are fibres made from steel,stainless steel, AR-glass (alkaline resistant), E-glass (nonalkaline resistant), polypropylene, polyacrylnitrile, nylon,carbon, etc. (Józsa et al., 2005).To study crack tendency of different cement types Balázs etal. (1979) have made ring tests, where the time of cracking wasmeasured. Crack tendency of FRCs was investigated by Shah,Weiss, in 2006, Aly, Sanjayan and Collins in 2008 and Alyand Sanjayan in 2009. Restrained shrinkage was also studiedby Salah and Lange in 2001. Recently another test method isused to evaluate restrained early age shrinkage according tothe Austrian fibre-concrete technical specification describedin Faserbeton Richtlinie (2002, 2008). This method was usedby Józsa et al. in 2005, and by Schmidt in 2005 and also inCONCRETE STRUCTURES • 2010 61

86 1966457 1968015 1606089 1984455 Testiert87 1792825 1792634 1776650 1682443 Testiert88 1618080 1659683 1723302 1618886 Testiert89 1964103 1964093 1937004 1963476 Testiert90 1662214 1745513 1722620 1722536 Testiert91 1966868 1794467 1965979 1793183 Testiert92 1660449 1617777 1756298 1713561 Testiert93 1747825 1417472 1425778 1478903 Testiert94 1747566 1747964 1747870 1694693 Testiert95 1721498 1616105 1615818 1691858 Testiert96 1745018 1983948 1984031 1984044 Testiert97 1966379 1761607 1322396 1570047 Testiert98 1644575 1644106 1620515 1687752 Testiert99 1296648 1812420 1794438 1965351 Testiert100 1967812 1967524 1967650 1758610 Testiert101 1983809 1725384 1983621 1726163 Testiert102 1450549 1712041 1726024 1715556 Testiert103 1964394 2018469 1967139 1968549 Testiert104 1667222 1556676 1469242 Testiert105 2005605 2005731 1985441 1821589 Testiert106 1341373 Testiert107 1618530 Testiert108 1747906 TestiertHinweise:Bitte beachten Sie auch die Testatliste des Fachgebiets Wasserbau. Nur wenn beide Übungsteiletestiert wurden wird die Studienleistung als „bestanden“ in TUCaN eingetragen!Kraft 2013-07-18 4/4

Fig. 12: Average summarized crack lengths (of four specimens) of FRCmade with glass and PAN fibres of 12 mm length after curing in tumbler-drierFig. 15: Average summarized crack lengths (of four specimens) of twomixes of FRC made glass fibres of with 12 mm length after curing in windtunnelFig. 13: Average summarized crack lengths (of four specimens) of FRCmade with glass and PAN fibres of 12 mm length after curing in windtunnelFig. 14: Average summarized crack lengths (of four specimens) of FRC madewith glass and PAN fibres of 12 mm length after curing in tumbler-drierthe tested PAN fibre is as effective as the glass fibre, becausein one kg there are more pieces of fibre, but if we study realfibre effectiveness, glass fibre is better. The reason of this effectis the higher Young’s modulus of glass fibres.The specialty of the applied experimental method is that ifwe change the cement type or the aggregate, crack tendencyis changing, so we can not compare our experimental resultswith findings that others measured in other laboratory with adifferent reference mix. If we repeat an experiment sequenceof one fibre type with different mixture with different cracktendency and compare with the first sequence, the results willbe different, as shown in Figs. 15 and 16. In these curvesone point represents the average summarized crack length ofTable 3: Classes of crack tendency according to Faserbeton Richtlinie(2002, 2008)YearClassesReference mixtureAverage crack length [%]FRC mixtureFS 1 100 602002FS 2 100 202008 FS 100 20Fig. 16: Average summarized crack lengths (of four specimens) of twomixes of FRC made with glass fibres of 12 mm length after curing intumbler-drierfour ring specimens. The reference mixtures (made withoutfibre reinforcement) have different crack tendency as theother sequence’s reference mix had. If we compare the resultsof the two sequences, we can see that the two linear curvesapproach each other. This means that theoretically there isa significant point where the two lines are meeting. In ourexperiences we could not reach this point, because one of ourlines had already reached the horizontal axis before crossing,so there were no cracks registered in case of the first mix, atthe highest 1.0 kg/m 3 dosage of fibres. Further investigationsare needed with a reference mix with higher crack tendency toreach this point where the lines are meeting. It could happenthat we cannot reach this point, because the higher porosityof concrete caused by the higher fibre dosage will deform thelinear curves or before this point every mixture will reach thezero crack length.5. CONCLUSIONSIn this paper early age shrinkage cracking (caused by autogenousand drying shrinkage) of different fibre reinforced concretes(FRC) were investigated. Variable parameters were the typeand the dosage of fibres. Four types of fibres were testedfabricated from glass, PAN and PP. To evaluate effectivenessof different types of fibres, a reference mix was producedwithout fibre reinforcement. To measure the early age shrinkagecrack tendency ring test was used according to the Austrianfibre reinforced concrete technical specification described inFaserbeton Richtlinie (2002, 2008). For the ring test specialspecimens were fabricated and cured in wind tunnel in aclimatic room according to the Austrian technical specification.The length of every crack caused by early age shrinkage wasregistered after the test. Further investigations were carried outby testing the specimens in standard tumbler-drier at 60 °C.Afterwards crack lengths were also registered.CONCRETE STRUCTURES • 2010 65

− The relationship between fibre dosage and early ageshrinkage crack tendency was found to be linear in case ofevery tested type of fibres. This means that if we apply ahigher dosage of glass or polymer fibres in concrete, a lowercrack tendency is reached.− There is a significant difference between evaluating theresults by dosage in kg/m 3 or in V%. If we evaluate the earlyage shrinkage crack tendency by V% (and the geometry offibres is the same e.g. the tested PAN and glass fibres) wecan see the real effectiveness of a fibre type. Estimatingthe dosages in V%, glass fibres are better than polymerfibres. But glass has higher density at the same time, so ifwe evaluate by the dosage of fibres in kg/m 3 that meansmore pieces of PAN fibres are compared with a few glassfibres.− If we repeat our experiments with an other reference mixwhich has higher early age shrinkage crack tendency andcompare the two resulted linear curves will be registered,they approach each other by ascendant dosage of fibres.− As an additional result of our tests further investigationsare needed by testing other types of fibres and higherdosages of the tested fibres to evaluate further relationshipsbetween early age shrinkage crack tendency and fibrereinforcement.6. ACKNOWLEDGEMENTSAuthors wish to express their gratitude to Avers Ltd. andBaumbach Metall GmbH for giving free run of materials forpresent research; and to Csaba Gyömbér (Djember) for theimportant help in the laboratory works.7. REFERENCESAly, T., Sanjayan, J. G., Collins F. (2008), “Effect of polypropylene fibers onshrinkage and cracking of concretes” Materials and Structures 22 January2008 41:1741-1753Aly, T., Sanjayan, J. G. (2009), “Shrinkage cracking of OPC-fiber concrete atearly age” Materials and Structures 13 July 2009Balázs, Gy., Borján, J., Cary Silva, J., Liptay, A., Zimonyi, Gy. (1979), “Cracktendency of cement” (A cement repedésérzékenysége) Sci. publications(Tudományos közlemények) in Hungarian, 1979 Budapest, HU ISSN0324-3575Balázs, Gy. L. – Lublóy, É., „Residual compressive strength of fire exposedfibre reinforced concrete” Concrete Stuctures 2007, ISSN 14196441CEB (1992), CEB No. 183 „Durable Concrete Structures” - Design Guide;ISBN 978-0-7277-1620-0EN 206-1 – “Concrete part 1.: specifications, performance, production andcomformity” European CodeFaserbeton Richtlinie (2002), „Austrian fibre-concrete technical specification“2002 March (in German) Österreichische Vereinigung für Beton- undBautechnik, 63 p.Faserbeton Richtlinie (2008), „Austrian fibre-concrete technical specification“2008 (in German) Österreichische Vereinigung für Beton- und Bautechnik,97 p.Fenyvesi, O. (2006), „Early Age Shrinkage Cracking of Fibre ReinforcedLightweight Aggregate Concrete” Proceedings of 6th International PhDSymposium in Civil Engineering (Eds. T. Vogel, N. Mojsilovic, P. Marti),Zürich 23-26 August, 2006 pp. 1-8.Grube, H. (2003), „Definitions of different types of shrinkage – concrete”(Definition der verschiedenen Schwindarten) – Beton, 2003/12 pp. 603Józsa, Zs., Djember, C, Für Kovács, I., Seidl, Á. (2005), “Use of Glassand Synthetic Fibres Preventing Early Age Cracking of Normal andLightweight Concrete” Proc. 1st Central European Congress on ConcreteEngineering, 2006 Graz, Austria pp. 125-130.Neville, A. M. (1995), “Properties of concrete” 1995 ISBN: 0-582-23070-5Salah, A., Lange, D. A. (2001), “Creep, Shrinkage and Cracking of RestrainedConcrete at Early Age” ACI Materials Journal, July-August 2001. pp.323-331Schmidt, M. (2005), „Baumbach Metall GmbH Product informations for PPfibers“ (in German), Baumbach Metall GmbHShah, H. R., Weiss, J. (2006), “Quantifying shrinkage cracking in fiberreinforced concrete using the ring test” Materials and Structures RILEMonline 20 September 2006 39:887-899Wongtanakitcharoen, T., Naaman, A. E. (2006), “Unrestrained early ageshrinkage of concrete with polypropylene, PVA, and carbon fibers”Materials and Structures RILEM online 10 November 2006 40:289-300Olivér Fenyvesi (1981), MSc. in civil engineering (BME 2005). PhDStudent at Budapest University of Technology and Economics Departmentof Construction Materials and Engineering Geology. Main fields of interestare the early age (autogenous and drying) shrinkage and early age shrinkagecracking in normal- and lightweight concretes, the FRC, fiber reinforcedlightweight concretes, self compacting lightweight concretes, buildingdiagnostics, building heritage. Member of Scientific Society of Silicate Industry(Concrete Division).Zsuzsanna Józsa (1950) PhD, associate professor at the Department ofConstruction Material and Engineering Geology at Budapest Universityof Technology and Economics, architect, postgraduate diploma in buildingreconstruction. Her main fields of activities are lightweight aggregate concrete,non-destructive concrete testing, concrete corrosion and repair, thermal andmoisture properties of materials, thermal insulation and water-proofing, rooftiles, wall construction, bricks and tiles, aerated concrete, environmentalcompatible building materials. She is member of IASS WG 18 “EnvironmentalCompatible Structures and Structural Materials”, Association for BuildingBiology, Hungarian Scientific Society of Building, Scientific Society of SilicateIndustry, Hungarian Group of fib, and is member of fib Commission 8 andearlier of TG 8.1: Lightweight aggregate concrete.66 2010 • CONCRETE STRUCTURES