ThammasatInt. J. Sc. Tech.,Vol. 9, No.

3, July-September
2004

Toughness Evaluation
of Steeland PolypropyleneFibre
ReinforcedConcreteBeams
under Bending
Piti Sukontasukkul
AssistantProfessorDepartmentof Civil Engineering,
Facultyof EngineeringKing Mongkut's Institute
of Technology-NorthBangkok
l5l8 PibulsongkramRoad,Bang Sue,Bangkok 10800
T e\ 02-9 | 3 -2500 ext. 862| -29, F ax.02- 587-4337, Email: piti @kmitnb.ac.th

Abstract
Generally, the performance of most materials is characterized by parameters based on the
mechanical properties such as strength,strain, or stiffness etc. However, in fibre reinforced concrete
(FRC), unlike other materials,strengthor stiffrressalone is not sufficient to characterizeits behaviour,
and the value of toughnessis often used instead.In this study,two different methods(ASTM C1018
and JSCE SF-4) are used to measurethe toughnessof steel and polypropylene fibre reinforced
concretesubjectedto bending.Resultsindicatedthat in the JSCEmethod,the informationobtainedby
only one specifieddeflectiontoughnessseemedto be insufficientin reflectingthe characteristics of the
load-deflection curves of both FRCs. On the other hand, in the ASTM method, the obtained
information using the four toughnessvalues at different deflections appearedto better clariff the
characteristics
of both FRCs.

Keywords: fibre reinforcedconcrete,toughness,ASTM Cl0l8, JSCE SF-4

1. Introduction small because the matrix cracks essentially at

Short fibres have been known and used for
centuries to reinforced brittle materials like
cement or masonry bricks. At that time, fibres
were natural fibres, such as horse hair, straw, etc.
Now, there are numerous fibre types available
for commercialuse, the basic types being steel,
glass, synthetic materials (polypropylene,
carbon, nylon, etc.) and some natural fibres. As
for steel fibres, wire and metal clips were used
first in l9l0s to improve the properties of
concrete. Extensive research on steel fibres
began in the 1960's ll,2f, and,since then a
substantial amount of research, development,
application and commercialization have
occurred[3-7].
Typically, the fibre volume fraction in Steel
Fibre Reinforced Concrete (SFRC) is in the
range of 0.5ohto 1.5. At the practical volume
fraction used in SFRC (<1%), the increasein
compressive,tensile, or flexural strength is
the samestressand strainas in plain concrete[8-
I ll. The real advantageof adding fibres is that,
after matrix cracking, fibres bridge these cracks
and restrain them. In order to further deflect the
beam, additional forces and energies are
required to pull out or fracture the fibres. This
process, apart from preserving the integrity of
concrete, improves the load-carrying capacity
beyond cracking. This improvement createsa
long post-peak descendingportion in the load-
deflectioncurve.
Since fibres do not significantly improve the
properties of concrete or strength prior to the
peak (concrete cracks) as they do on the post-
peak response, it would be more practical to
evaluate the performance of FRC based on the
energy absorption (area under the load-
deflection curve). The area under the load-
deflection curve is usually referred to as the
toughnessobtained from a static test of a beam

J)
 2004
Int.J. Sc.Tech.,Vol. 9,No. 3,July-September
Thammasat

specimenup to a specifieddeformation(not to : fine : coarse) were cast in the form of beams

be confusedwith the fracture toughness,K16).
The two most common methods to
determineflexural toughnessare based ASTM
C l 0 l 8 a n d J S C ES F - 4 [ 2 ] . I n A S T M C 1 0 1 8 ,
toughness is specified in terms of toughness
indices (Is Iro and I2e),which refers to the area
under the load-deflection curye calculatedout to
three different specified deflections.While, in
the caseof JSCE SF-4. the areaunder the load-
deflection curve up to a specified deflection
(L/150) is measured and referred to as the
toughness.
In this study, both ASTM and JSCE
standardswere used to determine the toughness
of plain concrete as well as steel and
polypropylene reinforced concrete beams.
Resultsfrom both methodswere then compared
and discussed.
2. ExperimentalProgram
Plain and fibre reinforced concretewith the
mix proportionsof 1:0.45:I .7'.2.7(cement:water
(dimensions of 100x 100x 350 mm).
Two fypes of fibres were used: Steel and
Polypropylene at 2 different volume fractions:
loh and 2o/o.Geometrical details and properties
of each fibre are given in Table l. The casting
schedule is given in Table 2. After being
removed from the molds, the specimenswere
then cured in water for 28 days before being
subjectedto test.
All tests were carried out at the Department
of Civil Engineering,King Mongkut's Institute
of Technology-NorthBangkok,using a 1500kN
universal testing machine. During the test,
specimenswere placed on a simple support with
a clear span of 300 mm, and then subjectedto a
third-point loading at the rate 0.05 in/min (two
point-loads at 113 of the clear span, Fig. 1).
Results in terms of load-deflection curve were
collectedby a PC-baseddataacquisitionsystem.

Table l: FibreGeometry

Length SectionShape and Tensile

Type Material Shape
(mm) Dimension(mm) Strength (MPa)
-\--
HookedEnd Steel 60 Circle D i a .0 . 5 0 1000
Crimped Polypropylene /'\-/V\a\,4\ 58 Rectangle 1 . 0x 0 . 5 450

Table2: CastineSchedule

Concrete Volume No. of

Designation
type Fraction (7o) Sample

Plain concrete PLN 5

SteelFRC ISFRC t% 5
SteelFRC 2SFRC 2% 5
PolypropyleneFRC IPFRC t% 5
PolvoropyleneFRC 2PFRC 2% 5

36
 Thammasat
Ini. J. Sc.Tech.,Vol. 9,No. 3,July-September
2004

Loading Head

Fig. 2: Fracturetoughnessand Indices according

Fig 1: Test Setup to ASTMC1018

In addition, the terms of toughnessindices

3. Determining Fracture Toughness
(I5, I16and 126)are also calculated.Each index is
Two methodswere usedfor determiningthe
the ratio between the post-peak toughnessand
flexural toughnessnamely: ASTM C1018 and
the pre-peak(elastic)toughness(Fig. 2).
JSCESF-4.
I5 Area OACD/Area OAB
3.1 ASTM CTOT8 T -
rl0 Area OAEF/Area OAB
In ASTM Cl0l8, toughness(or energy
Izo Area OAGH/Area OAB
absorptiondefined as the area under the load-
deflectioncurve) is calculatedout at 4 specified
The residual strength represented by the
deflections(6, 36, 5.56 and 10.56,Fig. 2). The
averagepost-crackingload that the specimen
toughnessis calculatedat the deflection6 which
may carry over a specificdeflectioninterval,are
is consideredthe elastic or pre-peaktoughness
(first-cracktoughness),while the other three (at usuallydeterminedas follows:
36,5.56and 10.56)are considered the post-peak
toughness. Rs,,'o=20(1rc_ Is)
R r o . u=o t 0 ( i r o - l , o )
Area OAB : Toughnesscorrespondingto a
deflectionof6, (T6)
3.2 JSCE SF-4
: Toughnesscorrespondingto a Unlike the ASTM Cl0l8, JSCE SF-4
Area OACD
providesjust a single value of toughness.For a
deflectionof 35, (T:a)
given load-deflectioncurve, toughnessis the
: Toughnesscorrespondingto a area under the load deflection curve measured
Area OAEF
up to a specified deflection, 6,6 : L/150, as
deflectionof 5.56,(Tssa)
referredto Area OABC in Fig. 3.
Area OAGH : Toughnessconespondingto a
The toughness factor (equivalent to the
deflectionof 10.55(Tro.s)
average residual strength) can also be
: The deflectionat the linear determinedas:
where 5
elasticlimit
Toughness
factor: toughness
x L/(BH2x 6,6)

where L -- Span length

B : Width of the specimen
H : Height of the specimen
66 : Deflectionat L/\50

J I
 2004
ThammasatInt. J. Sc. Tech.,Vol. 9, No. 3, July-September

In the caseof polypropylenefibre reinfbrced

concrete (PFRC) as shown in Fig. 5, the
'double-peak'
responsefound herewas a typical
response. The first response was a typical
responseof concrete under bending. The load
increasedproportionallywith the deflection up
to the peak, and then the tailure occurreddue to
the matrix cracking. The secondresponsewas
the load recovery due to the eftect of fibre
bridging.

25.0
accordingto JSCE
Fig.3: Fracturetoughness
SF.4 20.0

! 15.0
;
4. Results and Discussion 3 ro.o
5.0
4.1 Load-Deflection Response
Typical load-deflectionresponsesof plain 0.0
0.0 2.5 5.0 7.5 10.0 12.5 15.0
concrete,SFRC and PFRC beamsare given in
Deflection(mm)
Figs.4 and -5.
Comparingbetweenplain concreteand FRC, Fig. 4: Typical Load-DetlectionResponsesof
it was the post-peak response that really Plain and SteelFibre ReinforcedConcrete
diff'erentiatedthe plain concretefrom the FRC.
For plain concrete,the behaviourwas more in a 14.0
brittle manner.Oncethe strainenergywas high 12.0
enough to cause the crack to self-propagate,
^10.0
fracture occurred almost instantaneouslyonce z
5 8.0
the peak load was reached. due to the
tremendousamount of energy being released. 3 5.0
-
4.0
For FRC. the fibre bridging eff'ect helped to
2.O Plain concrete
control the rate of energy release.Thus, FRC
maintainedits ability to carry load after the peak. 0.0
0.0 2.5 5.0 7.5 10.0 12.5 15.0
Comparing between each type of FRC, it
Deflection (mm)
could be seenclearly that the flexural responses
Fig. 5: Typical Load-DeflectionResponsesof
of both FRCs were quite different.ln the caseof
steel fibre reinforced concrete (Fig.a), the Plain and PolypropyleneFibre Reinfbrced
'single-peak'response. Concrete
response was a so-called
Prior to the peak, the load increased
proportionally with the increasing deflection. The differences in the load-deflection
and then at the peak (concretecracking).there responsesof both fibres were essentiallydue to
was a slight drop of load before fibres beganto the propertiesof the fibres themselves.In SFRC,
take over and this led to a graduallydrop of load. becauseof its high strength and stiffness,the
Unlike plain concrete where the point of fibres were highly effective in terms of bridging
concretecracking indicatedthe point of tailure, instantaneously over the cracks at a very small
in FRC, with the effect of fibres bridging across deformation or crack opening once the crack
the crack surface.FRC was able to maintainthe started to form. However. in the case of the
load carrying ability even after the concretehad polypropylenefibres which were low in strength
been cracked as shown in the descendinglong and stiffhess, they required much larger
post peakresponse. deformationor crack opening before the fibres
couldrespondto the load.As a resultof this,the

38
 Int.J. Sc.Tech.,Vol. 9, No. 3, July-September
Thammasat 2004

recovery of load was found late in the post-peak
response ofPFRC (13).
4.2 Toughness
In this section,toughnessesfor both FRCs
were measured according to the methods
described before; the results are compared and
discussed.
4.2.1 ASTM C1018
According to this method, the toughnessis
measuredat four different deflections: one prior
to the peak (T) and three after the peak (T36,
(Table 3). Once obtained,they
T5.56and T1e.56)
were then used to determine the toughness
indicesas shownin Table 3 and Fig. 6.
Table 3: ToughnessAccording to ASTM
and this resulted in a large drop of strength in
the load-deflection curve. Thus, the first peak
responseprior to the low recovery ofPFRC got
no action from the fibre and was, in fact, the
responseofthe plain concrete.
As for the steel fibre, the elastic toughness
(T6) of SFRC changed and dependedon the
contentofthe fibre as seenby the increasingT6
with the increasingfibre content. Becauseof the
high strength and elastic modulus of steel fibre,
immediately after the concrete first cracked,
fibres startedto take action. At low fibre content
(lo/o),a small drop of load was found, and then
followed by a quick recovery of load almost
immediately.With the volume fraction of 2%o,
larger (twice) numbers of fibres were
intercepted at the crack surface. This allowed
fibres to pick up the load as soon as the concrete
cracked,achieving no sign ofstrength drop.

Concrete Toughness(N-m) ToushnessIndices

Tvpe 6 l3d l s . s ol r o . s o I I Irn
lPFRC 2.7 8.3 16.3 30.7 3.1 6.1 11 . 6
2PFRC 2.6 11.2 27.2 65.4 4.2 10.3 24.7
lSFRC 12.9 )4) 49.9 5.5 10.4 21.3
2SFRC 3.5 28.4 55.5 92.6 8.2 15.9 26.6

ASTMToughnessIndices
30 Corrsiderthe post-peaktoughness(36, 5.56
and 10.56)by looking at the toughnessindices,
Ezs
.9 at small deflection(36 and 5.56), SFRC seemed
E 2 0 to be tougher than PFRC at both volume
8 1 5 fractions as indicated by the larger values of 15
o
-E 10 and 116.This was because,in PFRC, with a large
ct
B s drop of load immediately after the peak, the area
F under the curve (as well as toughness) of the
PFRC became smaller than that of SFRC.
1PFRC 2PFRC 1SFRC 2SFRC
However, at large deflection (10.56), once the
trt ar, trr,l polypropylene fibres in the PFRC came into
Fig. 6: ToughnessIndicesaccordingto ASTM play, the load started to recover and the
c1018 toughness of the PFRC was found to be
significantly increased,especially at 2o/ovolume
Considerthe pre-peak(or elastic)toughness fractions where the toughnessindex of 2PFRC
(T5). In the caseof the polypropylenefibre, the increased to almost the same level as that of
pre-peak toughness(T6) of PFRC was found to 2SFRC,
be similar to that of plain concreteand remained
constant even with the increasing fibre content 4.2.2 JSCE SF-4
from lo/oto 2o/o.The reasonfor this performance The toughnessvalues and factor according
was partly due to the material properties of the to JSCE SF-4 of both FRCs are siven in Table 4
polypropylene fibre itself. With low strengthand andFie. 7.
elastic modulus, after the first concrete crack,
polypropylenedid not take action immediately

39
 2004
Int.J. Sc.Tech.,Vol. 9, No. 3, July-September
Thammasat

Table 4: Toughnessaccordingto JSCESF-4 curve, rough descriptions of the behaviour of

both FRCs could be achieved.For example,the
Concrete Toughness Toughness small value of Is of the PFRC indicated that the
Tvne (N-m) Factor PFRC did not perform well at the small
lPFRC 8.3 1.24 deflection. However, the increasevalue of Izoof
2PFRC 9.8 1.47 the PFRC to almost the same level as that of
lSFRC 18.5 2.78 SFRC indicated that the performance of the
2SFRC 30.8 4.62 PFRC was quite well at larger deformation.

JSCEToughness
?
i 25
i,o ,o
-E 1s
CD
e 1 0
F
5 of the SFRCwas a single-peakresponse.
1PFRC 2PFRC 1SFRC 2SFRC
Fig. 7: Toughnessaccordingto JSCE SF-4
Based on this method, for a given load-
deflectioncurve, a value oftoughnessmeasured
up to a deflectionof L/l50 (2 mm) is calculated
and then used to determinethe toughnessfactor.
The obtained results indicated that the
performanceof SFRC is more superior than that
of PFRC at both volume fractions.
4.2.3 ASTM C1018vs. JSCE SF-4
Comparing between these two methods,
singlevalue toughnessmeasuredusing the JSCE
method did not have any difficulty reflecting the
toughnessproperfy of the SFRC. However, in
the case of PFRC. JSCE did not seem to be
sufficient to reflect the true toughnessproperfy.
For instance,ifthe toughnessvalue providedby
JSCE is consideredalone without looking at the
load-deflection curve, it will lead to the
conclusion that the performance of PFRC was
much poorer than that of SFRC which was not
quite right. It is true that the toughnessof PFRC
at small deflectionwas poorerthan that of SFRC,
but, at the larger deflection, the performanceof
PFRC was increasedto almost the same as that
ofSFRC.
On the other hand, the toughnessprovided
by ASTM at different deflections seemed to
work out well in term of capturing and reflecting
the true toughnessproperties of both SFRC and
PFRC. By considering the toughnessindices
alone without looking at the load-deflection
5. Conclusions
From this study, the conclusions can be
drawn as follows:
5.1 Both SFRC and PFRC behaveddifferently
under bending. Due to the properties of the
fibres, the behaviour of PFRC was clearly a
double-peakresponsewhile the behaviour
5.2 Because of the way each FRC behaved
differently, the toughnessof each FRC was
also different.
5.3 According to the ASTM method, with
toughness measured out at 4 different
deflections, the obtained information
seemed to capture and reflect the true
toughness properties of both SFRC and
PFRC quite well.
5.4 On the other hand, the JSCE method,with a
single value toughness, was not quite
sufficient in reflecting the real toughness
propertiesof the PFRC.
6. Acknowledgements
The author would like to thank the SR Fibre,
Co., Ltd. and the Bekeart Onesteel Fibre
Australasia Co., Ltd. for providing both stee-
and polypropylene fibres. Special thanks are
also due to the graduate students in the
AdvancedConcreteTechnologycourse(185241)
for their assistance.
7. References
[1] Romualdi, J. P., and Batson, G. B,
Mechanics of Crack Arrest in Concrete,
Proceedings,ASCE, Vol. 89, pp. 147'168,
EM 3, Jun 1963.
[2] Romualdi,J. P., and Mandel, J. A., Tensile
Strength of Concrete Affected by
Uniformly Distributed Closely Spaced
Short Lengths of Wire Reinforcement,ACI
joumal, Proceedings,Vol. 61, No. 6, pp.
657-61| Jun I 964.

40

Int.J. Sc.Tech.,Vol. 9, No. 3, July-September
Thammasat
t3l Shah, S., Concreteand Fibers Reinforced
Concrete Subjected to Impact Loading,
Cement-Based Composites: Strain Rate
Effects on Fracture. Materials Research
Society Symposia ProceedingsVol. 64,
1986.
t4l Cotteral, B, Fracture Toughness and The
Charpy V Notch Impact Tests, British
Welding Joumal,Vol. 9, No. 2, pp. 83-90,
1962.
t5l Manfore, G. 8., A Review of Fiber
Reinforcement of Portland Cement Paste,
Mortar, and Concrete, Joumal, PCA
Research and Development Laboratories,
Vol. 10,No. 3, pp. 36-42,Sep 1968.
t6l Shah, S. P., and Rangan, B. V., Fiber
Reinforced Concrete Properties, ACI
Joumal, Proceedings,Vol. 68, No. 2, pp.
1 2 6 - l 3 5 , F e b1 9 7 1 .
U] Shah,S. P., and Rangan,B. V., Ductility of
Concrete Reinforced with Stimrps, Fibers
and CompressionReinforcement,Journal,
Structural Division, ASCE, Vol. 96, No.
5 T 6 , , p p . l 1 6 7 - 11 8 5 ,F e b 1 9 7 0 .
4l
2004
[8] Johnston, C. D., Steel Fibre Reinforced
Mortar and Concrete-A Review of
Mechanical Properties, Fiber Reinforced
Concrete, SP-44, American Concrete
Institute,Detroit,, pp. 124-142,1914.
[9] Williamson, G. R., The Effect of Steel
Fibers on the Compressive Strength of
Concrete,ACI SpecialPublication,SP-44:
Fiber Reinforced Concrete, American
Concrete Institute, Detroit, pp. 195-207,
1974.
[10] Johnston,C. D., and Gray, R. J., Uniaxial
Tension Testing of Steel Fibre Reinforced
Cementitious Composites, Proceedings,
Intemational Symposium on Testing and
Test Methodsof Fibre-CementComposites,
zuLEM, Sheffield,pp.451-461,Apr. 1978.
!l] Mindess, S., Fibre Reinforced Concrete:
Challenge and Prospects,Fiber Reinforced
Concrete:Modem Development,pp. 1-12,
19 9 5 .
[2] Mindess, S., Young, J. F., Darwin, D.,
Concrete 2no Edition. Prentice hall. 2002.