Materials and Structures (2009) 42:1025–1038 DOI 10.1617/s11527-008-9441-3
ORIGINAL ARTICLE
Toughness testing of ultra high performance fibre reinforced concrete Marijan Skazlic´ Dubravka Bjegovic´
Received: Received: 8 April 2008 / Accepte Accepted: d: 10 October October 2008 / Published Published online: online: 22 October 2008 Ó RILEM 2008
Abstract In this paper an investigation is made of the applicability of the ASTM C 1609 procedure for testing testing toughne toughness ss of ultra ultra high high perfor performan mance ce fibre fibre reinfo reinforce rced d concre concretes tes contai containing ning a large large amount amount of fibre fibre (C2% by volum volume) e) and and exhi exhibi biti ting ng defle deflect ction ion hardening behaviour. All mixtures exhibited deflection hardening behaviour, and the parameters varied included (1) the amount of steel fibres, (2) the type of steel fibres, (3) the size of the longest fibre, (4) the addition of polypropylene fibres, and (5) the size of the maximum aggregate grain in the concrete matrix. Based Based on compar compariso ison n of the curves curves obtained obtained from from flexural flexural toughness tests with the evaluation of the test results obtained according to ASTM C 1609 and with the statis statistic tical al analys analysis, is, the author authorss rec recomm ommende ended d additional additional toughness toughness parameters parameters (P100,3.00, P100,4.00, P100,6.00, T100,3.00, T100,4.00, a n d T100,6.00) for the eval evalua uatio tion n of tough toughne ness ss resu result lts. s. Such Such addi additi tion onal al toughness parameters are calculated using a similar procedure as that specified in ASTM C 1609.
M. Skazlic´ (&) Á D. Bjegovic´ Materials Department, Faculty of Civil Engineering, University of Zagreb, Kaciceva 26, Zagreb, Croatia e-mail:
[email protected] [email protected] D. Bjegovic´ Institute of Civil Engineering Croatia, Rakusina 1, Zagreb, Croatia e-mail:
[email protected] [email protected]
Keywords Deflection hardening behaviour Á Fibre Á ASTM C 1609 Á Ultra high performanc performancee fibre reinforced concrete Á Toughness
1 Introduct Introduction ion One of main advantages gained from fibre reinforcement ent in concr oncret etee is an incr increa ease se in toug toughn hnes esss properties properties [1–3]. The The most most ofte often n used used me meth thod od for for test testing ing toug toughne hness ss of fibre fibre rein reinfo forc rced ed conc concre rete te is flexural toughness testing [3 [3]. The The ASTM ASTM C 1018 1018 stan standa dard rd has has been been used used for for toughness tests of fibre reinforced concrete for more than than a deca decade de.. Acco Accord rdin ing g to this this stan standa dard rd,, the the evaluation of toughness test results is made on the basi basiss of dimen dimensio sionl nles esss para parame mete ters rs of toug toughn hnes esss indexes and residual strength factor [4 [4]. Major complaints from researchers about ASTM C 1018 relate to difficulties in the determination of first crac crack k and and to the the prob proble lems ms occu occurr rrin ing g at accu accura rate te measurement of deflection [3 [3, 5–7]. It was found out that, because of errors arising in determination of first crack, toughness indexes and residual strength factors are not quite quite approp appropria riate te for the evalua evaluation tion of the behaviour of fibre-reinforced concretes with a small amount of fibres and different fibre volume fraction, and of those concretes tested on different specimens [6, 8–12 12]. ]. Numerous Numerous researches researches have carried carried out toughness toughness tests tests acc accord ording ing to proced procedure uress outlin outlined ed in ASTM ASTM C
1026
1018, but have evaluated the test results according to the Japanese standard JCI-SF 4 [13, 14]. The procedure for evaluating toughness test results specified in JCI-SF 4 has proved to be more reliable than that laid down in ASTM C 1018 in the case when the performances of fibre-reinforced concretes with a small amount of fibres and different fibre volume fractions were to be distinguished [2, 11, 14]. This is because of the disadvantages mentioned above that in the year 2005 the ASTM C 1018 standard was replaced with a new standard, i.e. ASTM C 1609 [15]. Thus, any matter contained in ASTM C 1018 with which the researches often found faults were excluded from ASTM C 1609. According to ASTM C 1609, toughness tests are carried out on concrete beams of 100 9 100 9 400 mm or of 150 9 150 9 600 mm. Flexural load is applied under constant rate of displacement at one-third of test specimen spans. The evaluation procedure for toughness test results is very similar to the evaluation Fig. 1 Definition of toughness indexes according to ASTM C 1609
Materials and Structures (2009) 42:1025–1038
procedure set down in JCI-SF 4. Specifically, in the evaluation of the test results, first-peak load, peak load, residual load, and the areas below the load– deflection curve are calculated (Fig. 1). Ultra high performance fibre reinforced concrete (UHPFRC) is a composite construction material with a cement matrix having a typical compressive strength of not less than 150 MPa and to which fibres are added to improve tensile strength and to ensure deflection hardening behaviour in flexural tests [16, 17]. Up to today, ASTM C 1609 has been primarily used for toughness tests of fibre-reinforced concretes containing small amount of steel fibres (\2% by volume). In contrast, the applicability of ASTM C 1609 for toughness tests of UHPFRC containing a large amount of steel fibres ( C2% by volume) and exhibiting deflection hardening behaviour has not been investigated so far [6]. In this paper an analysis is made of the applicability of the ASTM C 1609 procedure for testing
Materials and Structures (2009) 42:1025–1038
toughness of UHPFRC containing a large amount of fibre (C2% by volume) and exhibiting deflection hardening behaviour. Based on the results obtained, the authors recommended additional toughness parameters for the evaluation of toughness behaviour of UHPFRC. Further research in this area should be done.
2 Experimental investigation Flexural toughness and compressive strength tests were carried out on seven different UHPFRC mixtures. Each mixture was prepared three times. All the mixtures had the same water-binder ratio and fresh concrete workability. Plain and hybrid steel fibres were used. Hybrid fibres include steel fibres of various length and shape, and they are used to achieve synergetic effects in fresh and hardened concrete [18, 19]. Concrete mix compositions were varied as for the following factors: (1) the amount of steel fibres (2%, 3% and 5% by volume), (2) the type of steel fibres (each having 3% by volume of hybrid and ordinary steel fibres), (3) the size of the longest fibre 30 mm and 40 mm in the mixture containing 5% by volume of hybrid steel fibres), (4) the addition of polypropylene fibres (0.8% by volume of polypropylene fibres to 3% by volume of steel fibres), and (5) the size of a maximum aggregate grain in the concrete matrix 0.5 mm and 4 mm. The concrete mix compositions are given in Table 1. Considering that the concrete mix compositions did not include coarse aggregate, they can be also called ultra high performance fibre reinforced mortars; however, according to the accepted definition found in the literature, this type of the material is still termed ultra high performance fibre reinforced concretes (UHPFRC) [6]. 2.1 Materials The concrete components used in this experimental work had been found suitable for production of UHPFRC in previous investigations carried out by the authors [20]. The components used were only those available in the Croatian market. They included Portland cement, silica fume, quartz sand, superplasticizer, water, and steel and polypropylene fibres.
1027
Table 1 Compositions of concrete mixtures Mixture components, kg/m3
M1
M2
Cement
1115 1115 1115 1115 1115 1115 1115
Silica fume
169
Quartz sand, 0–0.5 mm
1073 1073 –
Quartz sand, 0–4 mm
–
–
1073 –
–
–
–
Water
204
204
204
204
204
204
Superplasticizer
30.8 32.1 34
37.6 37
39
38.5
169
M3
169
M4
169
M5
169
M6
169
M7
169
1073 1073 1073 1073
204
Steel fibers SF 1 –
–
–
–
39
156
156
Steel fibers SF 2 1 56
234
234
234
117
156
156
Steel fibers SF 3 –
–
–
–
–
–
78
Steel fibers SF 4 –
–
–
–
78
78
–
Polypropylene fibers PP1
–
–
–
8
–
–
–
Water/binder ratio
0.16 0.16 0.16 0.16 0.16 0.16 0.16
Physical and chemical properties of the cement and silica fume are given in Table 2. The aggregates used in this study contained quartz sand ranging in size from 0–0.5 mm and 0–4 mm fractions. The specific gravity and water absorption of the 0–0.5-mm quartz sand were 2.68 g/cm3 and 0.76% respectively, and those of the 0–4-mm quartz sand were 2.66 g/cm3 and 1.26% respectively. Properties of superplasticizer are shown in Table 3. Four types of steel fibres (SF 1, SF 2, SF 3, and SF 4) and one type of fibrillated polypropylene fibres (PP 1), whose characteristics are shown in Table 4, were used in this study. 2.2 Specimens Toughness was tested on 100 9 100 9 400 mm beams, while compressive strength was tested on 40-mm cubes. Flexural toughness tests were conducted on a set of six specimens for each mixture and compressive strength tests were conducted on a set of eighteen specimens for each mixture. A total of 42 specimens and 126 specimens for toughness tests and compressive tests respectively were used. The specimens were prepared in a 70-l laboratory mixer. The overall duration of the mixing was between 10 and 13 min. In all the mixtures, cement, aggregate and silica fume were mixed dry for 4 min before the
1028
Materials and Structures (2009) 42:1025–1038
addition of water and then, 2 min after water was introduced, superplasticizer was added. About 1 min after the addition of the superplasticizer, fibres were
loaded manually. The mixing was completed after a homogenous concrete mixture was obtained. The mixing time was the longest in the case of the mixtures with 5% by volume of steel fibres (M 6 and M 7). The samples were vibrated on a vibrating table vibrating at a rate of 150 Hz. The specimens were demoulded at the age of 14 h. The specimens were tested at the age of 28 days after being cured in water at the water temperature of 20°C. In previous investigations it was found that the specimens cured using heat steaming method exhibit higher strength [6, 16]. Considering that fibre-reinforced concrete sampleswhich are normally tested using the ASTM C 1609 procedureare not cured employing heat steaming method, in this experimental work only water curing method was applied.
Table 2 Physical and chemical properties of cement and silica fume Physical and mechanical properties
Ordinary Portland cement
Silica fume
Specific gravity (g/cm 3)
3.12
2.22
Blaine fineness (cm /g)
5,030
18,595
Residual material on the 0.09 mm sieve (%)
3.93
–
Residual material on the 0.045 mm sieve (%)
–
69.8
SiO2 (%)
19.71
93.02
Al2O3 (%)
5.02
1.37
Fe2O3 (%)
3.00
0.64
CaO (%)
63.51
1.35
Loss by burning (%)
1.22
2.08
SO3 (%)
3.82
0.38
Non-soluble residual in HCl and Na2O3 (%)
0.33
75.05
MgO (%)
2.17
0.75
Free lime (%)
1.09
–
Chlorides (%)
0.006
0.027
Na2O (%)
0.28
–
K 2O (%)
0.75
–
2
Chemical properties
2.3 Items of investigation Flexural toughness specimens were loaded in a four point loading configuration with two supports spaced a distance of 300 mm and two top loading points spaced at 100 mm according to the ASTM C 1609 standard. The evaluation of toughness test results was made as specified in ASTM C 1609 as well as by using the approach recommended by the authors. The rate of loading during toughness tests was 0.1 mm/ min. The test was conducted on a testing machine having the flexural capacity of 200 kN. The toughness test results were collected at a frequency of 1 Hz.
– Denotes not measured items
Table 3 Properties of superplasticizer
Table 4 Properties of steel and polypropylene fibers
Mass volume (g/cm3)
pH
Solid content (%)
Main component
1.07
5.9
24
Polycarboxylate ether
Characteristics
Fiber SF 1
SF 2
SF 3
SF 4
PP 1
Fiber length (mm)
6
13
30
40
6
Fiber diameter (mm)
0.15
0.15
0.4
0.5
0.015
Fiber aspect ratio
40
87
75
80
400
Density (g/cm )
7.8
7.8
7.8
7.8
0.9
Tensile strength (MPa)
2,590
2,059
2,193
1,725
256
Elongation at break (%)
3.2
3.3
3.3
3.3
8.3
Modulus of elasticity (GPa)
210
210
210
210
8
Fibre type
Straight
Straight
Hooked ends
Hooked ends
Fibrillated
3
Materials and Structures (2009) 42:1025–1038
1029
Compressive strength was determined using a press of 3000 kN capacity with a controlled gain in force.
3 Experimental results and discussion 3.1 Toughness
2.4 Statistical analysis The statistical method used for evaluating the toughness and compressive strength test results was an analysis of variance of hierarchy models [21–23]. Although the hierarchy model may have an arbitrary depth, this paper discusses the case when the specimen consists of several groups, each group having several sub-groups, and each sub-group having various numbers of variants. The group means a certain concrete mixture (M 1 to M 7). Considering that each mixture was prepared three times, each of these three tests made a sub-group, and each subgroup had variants, that is, the test results of a specific property. Two zero hypotheses (H 0) were considered; the first hypothesis that the groups belong to the same specimen, and the other hypothesis is that the subgroups, within the groups, belong to the same specimen. The zero hypothesis (H 0) about variances is checked using F - test. The procedure for the analysis of variance of hierarchy models starts with calculating experimental F -factors: F exp1 ¼ F exp2 ¼
s2 between groups s2 between sub-groups s2 between subÀgroups s2 within sub-groups
ð1Þ ð2Þ
Table F -factor (F tabl) is read from the tables for free variants of the two respective variances with the selected probability of error of 0.05 [22]. The F exp and F tabl are compared and a decision made as to whether to accept or reject the zero hypotheses. If F exp1 \ F tabl1, it is concluded that there are no significant differences between the groups. This means that, between the mixtures analyzed, there is no significant difference as for toughness properties tested. If F exp1 [ F tabl1, a conclusion is made that the differences between the groups are significant; this indicates that there are significant differences between the mixtures investigated with respect to toughness property tested. If F exp2 \ F tabl2, it is considered that there are no significant differences between the sub-groups, or specifically that the test results are repeatable. When F exp2 [ F tabl2, the test results are not repeatable.
Table 5 and Fig. 2 present all the mean values of the results obtained from toughness tests. As the toughness test results were collected at the same frequency, the curves shown in Fig. 2 were obtained by calculating mean values of the force and deflection in a specific time interval. The analysis of the results obtained from the toughness tests was made using toughness parameters defined according to ASTM C 1609 (P1, PP, P100,0.50, P100,2.00, T100,2.00). The authors’ recommendations to the evaluation of toughness test results obtained for UHPFRC specimens described in this paper includes, besides toughness parameters defined in ASTM C 1069, taking into account additional toughness parameters, i.e. P 100,3.00, P100,4.00, P100,6.00, T100,3.00, T100,4.00, a n d T100,6.00. These additional toughness parameters are obtained using the same procedure as the one specified for the toughness parameters given in ASTM C 1609. The reason for their inclusion in the analysis of toughness test results is the fact that UHPFRC exhibits good behaviour and high toughness also at large deflections. The test results were statistically analyzed by an analysis of variance of hierarchy models in order to establish the existence of a significant difference among the mixtures tested. In this process, the toughness parameters defined in ASTM C 1609 and those from the authors’ recommendations for toughness evaluation were used. 3.1.1 Amount of steel fibres
In Fig. 3 the curves of the average values obtained from toughness tests are illustrated. Concrete mixtures M 1, M 5 and M 6 differed according to the amount of fibres. Specifically, the mixtures M 1, M 5 and M 6 contained 2%, 3% and 5% by volume of steel fibres respectively. The analysis of toughness test results using both the procedure specified in ASTM C 1609 and the authors’ recommendations for the evaluation of toughness test results is illustrated in Figs. 4 and 5. From the diagrams obtained from both the tests performed and toughness parameters calculated, it can be concluded that with an increase in the amount of steel fibre toughness properties are also increased.
1030
Materials and Structures (2009) 42:1025–1038
Table 5 Mean values of the results obtained from compressive strength tests and flexural toughness tests Mixture Compressive strength tests Compressive strength (MPa)
Flexural toughness tests First-peak deflection (mm)
First-peak strength (MPa)
Net deflection at peak Peak strength Peak strength/First-peak load (mm) (MPa) strength ratio
M1
182.9
0.06
11.10
0.91
22.30
2.01
M2
213.6
0.05
11.41
0.66
21.28
1.87
M3
197.1
0.07
12.24
0.73
23.09
1.89
M4
190.0
0.07
13.10
0.84
22.67
1.73
M5
211.3
0.06
12.62
0.97
22.91
1.82
M6
223.8
0.13
21.80
0.92
34.21
1.57
M7
212.0
0.15
19.29
0.88
28.59
1.48
Fig. 2 The curves of mean values obtained from flexural toughness for all concrete mixtures
120
M6 100
M7 80
) N k ( d 60 a o L
M1
40
M5 M3 M2
20
M4
0 0
1
2
3
4
5
6
7
8
Deflection (mm)
The results of the statistical analysis of the toughness test results (Table 6) illustrate that the mixtures M 1, M 5 and M 6 differ significantly in toughness parameters obtained using the procedure specified in ASTM C 1609 (P1, PP, P100,0.50, T100,2.00) and those toughness parameters obtained using the procedure proposed by the authors (P100,4.00, P100,6.00, T100,3.00, T100,4.00, T100,6.00). The obtained results proved that the introduction of additional toughness parameters for the evaluation of toughness behaviour was justified. In the case when the amounts of steel fibres are varied, toughness properties change with an increase in deflection and therefore the behaviour of these materials should be taken into account also at the deflection exceeding 2 mm. The toughness parameters up to the deflection of 2 mm are calculated according to ASTM C 1609.
3.1.2 Size of the longest fibre
The mixtures M 6 and M 7 contained the same amount of hybrid steel fibres (5% by volume), but they differed as for the size of the longest steel fibre 40 mm and 30 mm respectively. The mixture M 6, as shown in the diagrams in Fig. 3, exhibits better toughness behaviour than the mixture M 7 does. The results obtained from the calculation of toughness parameters are presented in Figs. 6 and 7. The statistical analysis of the toughness test results showed that the mixtures considerably differ only in respect of the toughness parameter P100,0.50. From Fig. 3 it is evident that there are no important differences in toughness tests being carried out either at large or at small deflections, and it is reasonable that the results obtained from the toughness parameters specified in
Materials and Structures (2009) 42:1025–1038
Fig. 3 The curves of mean values obtained from flexural toughness tests for concrete mixtures M 1, M 5, M 6 and M 7
1031
120
M6 100
M7 80
) N k ( d 60 a o L 40
M1
M5
20
0 0
1
2
3
4
5
6
7
8
Deflection (mm) Fig. 4 Comparison of the mean values of first-peak load, peak load and residual load for concrete mixtures M 1, M 5 and M 6
120.00 100.00
) 80.00 N k ( d 60.00 a o l 40.00 20.00 0.00 1 P
P P
0 5 . 0 , 0 0 1 P
0 0 . 2 , 0 0 1 P
M1
ASTM C 1609 and from those mentioned in the authors’ recommendations correspond. 3.1.3 Type of steel fibre
The mixtures M 2 and M 5 were prepared to contain the same amount of fibres (3% by volume). However, the difference in their composition was that M 2 had steel fibres of the same type, while M 5 had hybrid steel fibres. The curves obtained from toughness tests are illustrated in Fig. 8, while the evaluations of test results according to the ASTM C 1609 and authors’ recommendations are given in Figs. 9 and 10. The toughness curves illustrated in Fig. 8 show that the mixture with hybrid steel fibres, i.e. M 5 has
M5
0 0 . 3 , 0 0 1 P
0 0 . 4 , 0 0 1 P
0 0 . 6 , 0 0 1 P
M6
better behaviour than the mixture M 2. With an increase in deflection, the mixture M 5 shows far better behaviour than M 2 because it also contains, besides short fibres, long steel fibres that are more effective at large deflections than short fibres. The statistical analysis of the parameters according to ASTM C 1609 showed that there is no significant difference in behaviour between M 2 and M 5. In contrast, the statistical analysis of the toughness parameters recommended by the authors showed an important difference in these two mixture when the parameters P100,4.0, P100,6.0 and T100,6.0 are taken into account. This example illustrates that the evaluation of toughness behaviour according to ASTM C 1609 is adequate but incomplete in the case of UHPFRC with
1032
Materials and Structures (2009) 42:1025–1038
Fig. 5 Comparison of the mean values of toughness for concrete mixtures M 1, M 5 and M 6 obtained from the calculation of the area under the load–deflection curve up to a certain deflection
450,00 400,00 350,00 ) m300,00 (N s 250,00 s e n 200,00 h g u o t 150,00
100,00 50,00 0,00 0 0 . 2 , 0 0 1 T
0 0 . 3 , 0 0 1 T
0 0 . 4 , 0 0 1 T
M1
Table 6 The statistical analysis of the toughness test results obtained for the mixtures M 1, M 5 and M 6 Analyzed mixtures
M 1, M 5, M 6
Parameters
F exp1
F tabl1
F exp2
F tabl2
P1
62.51
5.14
2.77
3.37
PP
21.79
5.14
2.87
3.37
P100,0.50
69.59
5.14
1.55
3.37
P100,2.00
3.13
5.14
3.15
3.37
P100,3.00
3.24
5.14
2.14
3.37
P100,4.00
5.91
5.14
1.27
3.37
P100,6.00
35.46
5.14
0.34
3.37
T100,2.00
34.84
5.14
0.38
3.37
T100,3.00
7.93
5.14
2.73
3.37
T100,4.00
7.99
5.14
2.29
3.37
T100,6.00
9.05
5.14
1.90
3.37
different type of steel fibres when the ratio of a minimum cross-section size of the specimen to the fibre length is lower than 5; for this reason, additional toughness parameters that more adequately describe the behaviour at larger deflections should be used. 3.1.4 Addition of polypropylene fibres
The mixtures M 2 and M 4 each contained 3% by volume of steel fibres except that M 4 also had 0.8% by volume of polypropylene fibres. The curves obtained from toughness tests, as presented in Fig. 8, illustrate
M5
0 0 . 6 , 0 0 1 T
M6
that the mixture M 4 exhibits better behaviour up to deflection of about 2 mm, while M 2 exhibits better toughness performance beyond this deflection point. This can be explained by the fact that the addition of polypropylene fibres owing to their hydrophobic properties results in reduced adhesion of steel fibres to cement matrix, and consequently poorer behaviour at larger deflections. Figures 11 and 12 show the toughness parameters calculated for the mixtures M 2 and M 4. The analysis of toughness test results obtained according to ASTM C 1609 may lead to erroneous interpretation of the results obtained from testing. ASTM C 1609 takes into account the behaviour up to the deflection point of 2 mm, and this is the deflection up to which the mixture M 4 exhibits better behaviour. In contrast, the mixture M 2 shows better behaviour beyond this deflection point. However, if the test results are analyzed according to the recommendations given by the authors, toughness parameters at larger deflections can also be obtained. The statistical analysis of the test results shows that the mixture M 2 exhibits much better behaviour than the mixture M 4 with respect to the toughness parameters which are not defined in ASTM C 1609, i.e. P 100,4.0 and P100,6.0. 3.1.5 Size of the maximum aggregate grain
The mixtures M 2 and M 3 contained the same amount of fibres (3% by volume); however, the
Materials and Structures (2009) 42:1025–1038
Fig. 6 Comparison of the mean values of first-peak load, peak load and residual load for concrete mixtures M 6 and M 7
1033
120,00
100,00
) 80,00 N k ( d 60,00 a o l 40,00
20,00
0,00 1 P
P P
0 .5 0 , 0 0 1 P
0 .0 2 , 0 0 1 P
M6
Fig. 7 Comparison of the mean values of toughness for concrete mixtures M 6 and M 7 obtained from the calculation of the area under the load–deflection curve up to a certain deflection
0 .0 3 , 0 0 1 P
0 .0 4 , 0 0 1 P
0 .0 6 , 0 0 1 P
M7
450,00 400,00 350,00 ) m300,00 N ( s 250,00 s e n 200,00 h g u 150,00 o t 100,00 50,00 0,00 0 .0 2 , 0 0 1 T
0 .0 3 , 0 0 1 T
0 .0 4 , 0 0 1 T
M6
mixture M 2 had smaller maximum aggregate grain (0.5 mm) than the mixture M 3 (4 mm). The toughness curves (Fig. 13) illustrate that up to the deflection of about 2 mm better toughness behaviour is exhibited by the mixture M 3 and beyond this point by the mixture M 2. This is due to the fact that the mixture M 2 had a higher ratio of fibre length to a maximum aggregate size than the mixture M 3 (26 and 3.25 respectively), and this parameter is crucial for mixture behaviour under flexural load at larger deflections. On the other hand, the results given in Table 5 illustrate that the mixture with a larger maximum aggregate size had higher flexural strength by 9%. This result can be explained by better distribution of steel fibres in the case of a larger maximum aggregate size.
0 .0 6 , 0 0 1 T
M7
The evaluation of toughness tests according to ASTM C 1609, as shown in Figs. 14 and 15 indicates that the mixture M 2 shows better behaviour. The introduction of the additional toughness parameters illustrated that the mixture M 2 displays better behaviour at deflection exceeding 2 mm. The statistical analysis of toughness test results showed that M 2 has markedly better behaviour than M 3 with respect to the toughness parameter P 100, 6.0 that is not specified in ASTM C 1609. 3.1.6 Discussion
From the above discussion it is apparent that the recommendations given by the authors for UHPFRC toughness tests have some advantages over the
1034
Fig. 8 The curves of mean values obtained from flexural toughness tests for concrete mixtures M 2, M 4 and M 5
Materials and Structures (2009) 42:1025–1038 80 70 60
M5
) 50 N k ( d 40 a o L 30 20
M2 M4
10 0 0
1
2
3
4
5
6
8
7
Deflection (mm) M2
Fig. 9 Comparison of the mean values of first-peak load, peak load and residual load for concrete mixtures M 2 and M 5
M4
M5
0 0 . 2 , 0 0 1 P
0 0 . 3 , 0 0 1 P
90,00 80,00 70,00 60,00 ) N 50,00 (k d a 40,00 o l
30,00 20,00 10,00 0,00 1 P
P P
0 5 . 0 , 0 0 1 P
M2
procedure described in ASTM C 1609. The authors’ recommendations for the evaluation of toughness test results, besides toughness parameters defined in ASTM C 1069, taking into account additional toughness parameters, i.e. P100,3.00, P100,4.00, P100,6.00, T100,3.00, T100,4.00, and T100,6.00. These additional toughness parameters are obtained using the same procedure as the one specified for the toughness parameters given in ASTM C 1609. In comparison to UHPFRC, conventional FRC has less quantity of fibres and a lower quality concrete matrix, which results in poorer bond between the fibres and the matrix and lower flexural toughness. In all load–deflection diagrams obtained from flexural
0 0 . 4 , 0 0 1 P
0 0 . 6 , 0 0 1 P
M5
toughness tests, deflection hardening after first-peak strength can be noticed. Such deflection hardening was accompanied by multiple cracks and absorption of a large amount of energy. The tested fibrereinforced concretes behaved in such a way because they contained a large amount of fibres exhibiting good adhesion to the dense and compact matrix. Owing to their improved properties in comparison with those of conventional fibre-reinforced concrete, this concrete type is termed ultra high-performance fibre-reinforced concrete. The authors recommend that, when the tests of UHPFRC containing 2% to 5% in volume of steel fibers are conducted according to ASTM C 1609,
Materials and Structures (2009) 42:1025–1038
Fig. 10 Comparison of the mean values of toughness for concrete mixtures M 2 and M 5 obtained from the calculation of the area under the load–deflection curve up to a certain deflection
1035
300,00
250,00 ) 200,00 m N ( s s e 150,00 n h g u o t 100,00
50,00
0,00 0 .0 2 , 0 0 1 T
0 .0 3 , 0 0 1 T
0 .0 4 , 0 0 1 T
M2
Fig. 11 Comparison of the mean values of first-peak load, peak load and residual load for concrete mixtures M 2 and M 4
0 .0 6 , 0 0 1 T
M5
80.00 70.00 60.00
) 50.00 N k ( d 40.00 a o l 30.00 20.00 10.00 0.00 1 P
P P
0 .5 0 , 0 0 1 P
0 .0 2 , 0 0 1 P
M2
additional toughness parameters (P100,3.00, P100,4.00, P100,6.00, T100,3.00, T100,4.00, and T100,6.00) should be used in any of the following cases: – – –
The ratio between a minimum cross-section size of the specimen and the fiber length is below 5; A maximum aggregate size used is larger than or equal to 4 mm; and Polypropylene fibers are used in combination with steel fibers.
0 .0 3 , 0 0 1 P
0 .0 4 , 0 0 1 P
0 .0 6 , 0 0 1 P
M4
3.2 Compressive strength The results of compressive strength tests are summarized in Table 5. All the mixtures have compressive strength higher than 180 MPa. By statistical analysis it was established that the results obtained from compressive strength tests are repeatable for all mixtures from M 1 to M 7. The analysis of the results showed that the mean values of compressive strength exhibited by
1036
Fig. 12 Comparison of the mean values of toughness for concrete mixtures M 2 and M 4 obtained from the calculation of the area under the load–deflection curve up to a certain deflection
Materials and Structures (2009) 42:1025–1038 250.00
) 200.00 m N ( s150.00 s e n h g100.00 u o t 50.00
0.00 0 .0 2 , 0 0 1 T
0 .0 3 , 0 0 1 T
0 .0 4 , 0 0 1 T
M2
Fig. 13 The curves of mean values obtained from flexural toughness tests for concrete mixtures M 2, and M3
0 .0 6 , 0 0 1 T
M4
80 70
M3
60
) 50 N k ( d 40 a o L 30
M2
20 10 0 0
1
2
3
4
5
6
7
8
Deflection (mm)
UHPFRC with 3% and 5% in volume of steel fibres were higher by 16% and by 17% respectively and by 16% and by 22% respectively compared with UHPFRC with 2% in volume of fibres. An increase in the length of steel fibres contained in the concrete with 5% in volume of fibres caused a 6% increase in mean compressive strength. The use of hybrid fibres instead of plain steel fibres in the quantity of 3% in volume resulted in a reduction in mean compressive strength by 1%. The addition of polypropylene fibres to UHPFRC with 3% in volume of fibres caused a 12% decrease in mean compressive strength. The increase of maximum aggregate size from 1 mm to 4 mm in the case of UHPFRC with 3% volume of fibres resulted in a decrease in compressive strength by 8%. Because the compressive strength tests were carried out on the 40-mm cubes, which are normally
used for testing mortar specimens, comparative compressive strength testing on 100-mm cubes was also performed [20]. The obtained test results illustrated that mean compressive strength values obtained from tests on larger cube specimens are smaller by about 20%.
4 Conclusions An experimental investigation into toughness and compressive strength, and a statistical analysis of the results were carried out on ultra high performance fibre-reinforced concrete specimens containing a large amount of fibre (2–5% by volume) and exhibiting deflection hardening behaviour. The parameters varied were the following: (1) the amount
Materials and Structures (2009) 42:1025–1038
Fig. 14 Comparison of the mean values of first-peak load, peak load and residual load for concrete mixtures M 2 and M 3
1037
90,00 80,00 70,00 60,00 ) N 50,00 k ( d a 40,00 o l
30,00 20,00 10,00 0,00 1 P
0 5 . ,0 0 0 1 P
P P
0 0 . ,2 0 0 1 P
M2
Fig. 15 Comparison of the mean values of toughness for concrete mixtures M 2 and M 3 obtained from the calculation of the area under the load–deflection curve up to a certain deflection
0 0 . ,3 0 0 1 P
0 0 . ,4 0 0 1 P
0 0 . ,6 0 0 1 P
M3
250,00
200,00 ) m 150,00 N ( s s e n h g 100,00 u o t
50,00
0,00 0 0 . 2 , 0 0 1 T
0 0 . 3 , 0 0 1 T
0 0 . 4 , 0 0 1 T
M2
of steel fibres, (2) the type of steel fibres, (3) the size of the longest steel fibre, (4) the addition of polypropylene fibres, and (5) the maximum aggregate size in the concrete matrix. In this investigation, which was focused on the analysis of the applicability of the existing toughness test methods specified in ASTM C 1609 to the UHPFRC specimens, the following conclusions were made:
•
0 0 . 6 , 0 0 1 T
M3
The authors recommendations to the analysis of toughness test results obtained according to ASTM C 1609 is based on the introduction of additional toughness parameters (P100,3.00, P100,4.00, P100,6.00, T100,3.00, T100,4.00, and T100,6.00)in addition to those given in ASTM C 1609in order to make this standard fully applicable for UHPFRC. When testing UHPFRC
1038
•
•
•
containing 2% to 5% in volume of steel fibers according to ASTM C 1609, it is recommended that additional toughness parameters should be used in any of the following cases: The ratio of a minimum cross-section size of the specimen to the fiber length is lower than 5; A maximum aggregate size is larger than or equal to 4 mm; and Polypropylene fibers in combination with steel fibers are used.
Such additional toughness parameters are calculated using a similar procedure as that specified in ASTM C 1609. This is due to the fact that ASTM C 1609 is primarily designedand has been used so farfor fibre-reinforced concretes with smaller amount of steel fibres (\2% by volume) and for fibre-reinforced concretes with matrices of lower quality than those of UHPFRC and whose behaviour in toughness tests is much lower than that of UHPFRC. Further research should be done to verify the advantage and disadvantage of this standard for toughness testing of UHPFRC. Acknowledgements The results presented in this paper originate from scientific projects (Modern methods for testing building materials, 082-0822161-2996, Principal researcher Marijan Skazlic´, PhD, Assistant Professor, and The Development of New Materials and Concrete Structure Protection Systems, 082-0822161-2159, Principal researcher Dubravka Bjegovic´, PhD, Professor), supported by the Ministry of Science, Education and Sports of the Republic of Croatia.
References 1. Kovacs I, Balazs GL (eds) (2004) Structural performance of steel fibre reinforced concrete. Publishing Company of Budapest University of Technology and Economics, Budapest 2. Shah SP, Brandt AM, Ouyang C, Baggott R, Eibl J, Glinicki MA et al (1995) Toughness characterization and toughening mechanisms. In: Naaman AE, Reinhardt HW (eds) Proceedings of the second international RILEM workshop. E&FN SPON, Ann Arbor, USA, pp 193–228 3. Johnston CD (2001) Fiber-reinforced cements and concretes. Gordon and Breach Science Publishers, Singapore 4. ASTM C 1018 (1997) Standard test method for flexural toughness and first-crack strength of fiber-reinforced concrete (using beam with third-point loading). ASTM International, West Conshohocken 5. ACI Committee 544 (2006) Measurement of properties of fiber reinforced concrete. In: ACI manual of concrete practice, ACI 544.2R-89, reapproved 1999, American Concrete Institute, USA
Materials and Structures (2009) 42:1025–1038 6. Naaman AE, Reinhardt HW (1996) High performance fiber reinforced cement composites 2 (UHPFRCC 2). RILEM Proceedings 31, E & FN Spon 7. Chanvillard G (1999) Characterisation of fibre reinforced concrete’s performance after a flexural test – part 1: on subjectivity of toughness indices. Mater Struct 32:601– 605. doi:10.1007/BF02480495 8. Johnston CD, Skarendahl A (1992) Comparative flexural performance evaluation of steel fibre-reinforced concretes according to ASTM C 1018. Mater Struct 25:191–200. doi: 10.1007/BF02473063 9. Mindness S, Chen L, Morgan DR (1994) Determination of the first-crack strength and flexural toughness of steel fiberreinforced concrete. Adv Cement Base Mater 1:201–208. doi:10.1016/1065-7355(94)90025-6 10. Johnston CD (1986) Toughness of steel fibre reinforced concrete. In: Johnston CD (ed) Steel fibre concrete. Elsevier Applied Science Publishers Ltd., Amsterdam 11. Cangiano S, Cucitore R, Plizzari GA (2002) A new proposal for the evaluation of fracture properties of steel fiber reinforced concrete. In: Proceeding of 6th international symposium on utilization of high strength/high performance concrete, Leipzig, Germany, pp 873–886 12. Balaguru P, Narahari R, Patel M (1992) Flexural toughness of steel fibre reinforced concrete. ACI Mater J 89(6): 541–546 13. JCI Standard SF-4 (1984) Method of tests for flexural strength and flexural toughness of fiber – reinforced concrete. Japan Concrete Institute Standards for Test Methods of Fiber Reinforced Concrete, Tokyo 14. Banthia N, Yan C, Bindiganavile V (2000) Development and application of high performance hybrid fiber reinforced concrete. In: Rossi P, Chanvillard G (eds) Proceedings of the fifth international RILEM symposium on fibre-reinforced concrete. Lyon, France, pp 471–480 15. ASTM C 1609 (2005) Standard test method for flexural performance of fiber-reinforced concrete (using beam with third-point loading). ASTM International, West Conshohocken 16. Naaman AE, Reinhardt HW (2003) High performance fiber reinforced cement composites 4 UHPFRCC-4: international RILEM workshop. Mater Struct 36:710–712 17. Naaman AE, Reinhardt HW (2006) Proposed classification of UHPFRC composites based on their tensile response. Mater Struct 39:547–555. doi:10.1617/s11527-006-9103-2 18. Banthia N, Gupta R (2004) Hybrid fiber reinforced concrete (HyFRC): fiber synergy in high strength matrices. Mater Struct 37:707–716 19. Pons G, Mouret M, Alcantara M, Granju JL (2007) Mechanical behaviour of self-compacting concrete with hybrid fibre reinforcement. Mater Struct 40:201–210. doi: 10.1617/s11527-006-9131-y 20. Skazlic´ M (2002) High performance hybrid fibre-reinforced concretes. Master’s Thesis, Faculty of Civil Engineering, University of Zagreb 21. Triola MF (1989) Elementary statistic. The Benjamin/ Commings Publishing Company, New York 22. Steel RGD, Torrie JH (1960) Principles and procedures of statistics. McGraw-Hill Book Company, New York 23. Pausˇe Z (1993) An introduction to mathematical statistics. Sˇkolska knjiga, Zagreb