Static and dynamic compressive properties of ultra-high performance concrete (UHPC) with hybrid steel fiber reinforcements

Static and dynamic compressive properties of ultra-high performance concrete (UHPC) with hybrid steel fiber reinforcements

Accepted Manuscript Static and dynamic compressive properties of ultra-high performance concrete (UHPC) with hybrid steel fiber reinforcements Zemei W...

984KB Sizes 2 Downloads 126 Views

Accepted Manuscript Static and dynamic compressive properties of ultra-high performance concrete (UHPC) with hybrid steel fiber reinforcements Zemei Wu, Caijun Shi, Wen He, Dehui Wang PII:

S0958-9465(17)30162-2

DOI:

10.1016/j.cemconcomp.2017.02.010

Reference:

CECO 2786

To appear in:

Cement and Concrete Composites

Received Date: 13 January 2016 Revised Date:

17 January 2017

Accepted Date: 19 February 2017

Please cite this article as: Z. Wu, C. Shi, W. He, D. Wang, Static and dynamic compressive properties of ultra-high performance concrete (UHPC) with hybrid steel fiber reinforcements, Cement and Concrete Composites (2017), doi: 10.1016/j.cemconcomp.2017.02.010. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT 1 2

Static and dynamic compressive properties of ultra-high

3

performance concrete (UHPC) with hybrid steel fiber

4

reinforcements Zemei Wu a, b, Caijun Shi a, *, Wen He a, Dehui Wang a a

b

College of Civil Engineering, Hunan University, Changsha 410082, China; Department of Civil, Architectural and Environmental Engineering, Missouri University of Science and Technology, Rolla, MO, 65401, USA

SC

6 7 8 9 10 11

RI PT

5

Abstract: Ultra-high performance concrete (UHPC) is promising in construction of concrete structures

13

that suffer impact and explosive loads. In order to make UHPC structures more ductile and cost-effective,

14

hybrid fiber reinforcements are often incorporated. In this study, a reference UHPC mixture with no fiber

15

reinforcement and five mixtures with a single type of fiber reinforcement or hybrid steel fiber

16

reinforcements of 6 and 13 mm in length at a total dosage of 2%, by the volume of concrete, were

17

prepared. Quasi-static compressive and flexural properties of those mixtures were investigated. Split

18

Hopkinson press bar (SHPB) testing was adopted to evaluate their dynamic compressive properties under

19

three impact velocities. Test results indicated that UHPC with 1.5% long fiber reinforcements and 0.5%

20

short fiber reinforcements demonstrated the best static and dynamic mechanical properties. The static

21

compressive and flexural strengths of UHPC with 2% long fiber reinforcements were greater than those

22

with 2% short fiber reinforcements, whereas comparable dynamic compressive properties were observed.

23

Strain rate effect was observed for the dynamic compressive properties, including peak stress, dynamic

24

increase factor, and absorbed energy. The reinforcing mechanisms of hybrid fiber reinforcements in UHPC

25

were eventually discussed.

AC C

EP

TE D

M AN U

12

*Corresponding author. Tel./fax: +86 731 8882 3937. E-mail address: [email protected]; [email protected] (C. Shi)

1

ACCEPTED MANUSCRIPT 26 27

Keywords: Ultra-high performance concrete; Dynamic compressive properties; Hybrid fiber

28

reinforcements; Static mechanical properties; Split Hopkinson press bar

29

1. Introduction

RI PT

30

Ultra-high performance concrete (UHPC) was introduced in the mid-1990s [1]. Use of high content

32

of cementitious materials, superplasticizer, steel fiber reinforcement, and special curing, such as heat and

33

autoclave curing, and even nano-particles, are often mentioned in order to achieve high strength and

34

toughness [1-4]. Compared to ordinary concrete, UHPC has a very high compressive strength in the order

35

of 150 MPa and a tensile strength in the order of 10 MPa along with a strain hardening behavior [5, 6]. Its

36

superior performance allows its potential applications in military and protective structures that suffer

37

impact and explosive loads with high impact velocity [7, 8]. However, the cost of steel fiber

38

reinforcements used in UHPC is generally higher than that of the matrix. With the emphasis on

39

sustainability of UHPC, minimizing the fiber reinforcement volume without sacrifice of its performance is

40

of great significance [9]. Blending different types, shapes and/or sizes of fiber reinforcement has been

41

proven as one of the effective methods because of a combination of various features from those different

42

fiber reinforcements [10-12]. For example, the short fiber reinforcements can effectively restrict the

43

development of micro-cracks while the long fiber reinforcements can bridge macro-cracks [9]. Research

44

on the mechanical properties of hybrid fiber-reinforced UHPC has attracted a lot of attention, but most of

45

the literature thus far was focused on quasi-static properties, such as compressive, tensile, and flexural

46

properties, rather than dynamic behavior [5, 12, 13].

AC C

EP

TE D

M AN U

SC

31

47

There are two common methods, drop weight and split Hopkinson pressure bar (SHPB) testing, to

48

measure the dynamic behavior of a composite material. The drop weight testing has the ability to 2

ACCEPTED MANUSCRIPT reproduce conditions under which real life component would be subjected to impact loading [14]. Besides,

50

replication of testing arrangement is possible by using this method. Habel et al. [6] employed this method

51

to evaluate the impact bending performance of ultra-high performance fiber reinforced concrete

52

(UHPFRC) and found that the strength increased with the increase of strain rate. Although this method is

53

simple and easy to be implemented, the experimental results are greatly influenced by the specimen size

54

and configuration, such as drop weight and speed, support stiffness, etc. [15, 16]. The SHPB testing is the

55

most commonly used one for evaluating dynamic properties [17, 18]. Rong et al. [19] used the SHPB

56

testing to investigate the dynamic compression behavior of UHPFRC with different fiber reinforcement

57

volumes and found that the impact resistance of UHPFRC was improved with the increase of fiber

58

reinforcement volume. Under the same compressive loading, the control specimens were crushed, while

59

those with 3% or 4% of steel fiber reinforcements remained essentially intact. Lai et al. [20] indicated that

60

the peak stress, peak strain, and the area under strain-stress curve of UHPFRC increased with the increase

61

of strain rate. Zhang et al. [21] reported that the dynamic tensile strength of C200 green reactive powder

62

concrete increased markedly with the increase of impact velocity, ranging from 4 to 14 m/s. Until now, no

63

information can be found about the effect of fiber hybridization on the dynamic mechanical properties,

64

especially on the dynamic compressive properties of UHPC.

EP

TE D

M AN U

SC

RI PT

49

This research aims at investigating the influences of hybrid steel fiber reinforcements with lengths of

66

13 and 6 mm on flowability, quasi-static compressive and flexural properties, as well as dynamic

67

compressive behavior of UHPC. Split Hopkinson pressure bar testing was conducted to characterize the

68

dynamic compressive behavior, including failure pattern, load-deflection curve, peak stress, and fracture

69

energy, under three high impact velocities. Dynamic increase factor was introduced to compare the

70

quasi-static properties with the dynamic compressive properties. Finally, the reinforcing mechanisms of

71

hybrid fiber reinforcements in UHPC were discussed.

AC C

65

3

ACCEPTED MANUSCRIPT 72 73

2. Materials and experimental program

74

2.1. Materials Portland cement (P.I 42.5) complying with the Chinese Standard GB175-2007 was used [22]. Table 1

76

summarizes the chemical composition and density of the cement, silica fume, and slag. Silica fume has an

77

average particle size of 0.1 to 0.2 µm and a specific surface area of 18,500 m2/kg. Slag has a specific area

78

of 410 m2/kg. Two straight smooth steel fibers with diameters of 0.2 mm but one with a length of 6 mm

79

and the other one of 13 mm were used. The tensile strength is 2,800 MPa. River sand with a maximum

80

particle size of 2.36 mm was used. A polycarboxylate-based superplasticizer (SP) with a solid content of

81

20% was used.

82

Table 1 Chemical composition and density of cementitious materials Chemical CaO SiO2 Al2O3 Fe2O3 SO3 MgO Na2Oeq K2O C composition (%) (%) (%) (%) (%) (%) (%) (%) (%) Cement Silica fume Slag

62.49 21.18 1.85

-

39.11 33.00 13.91

84

2.2. Mixture proportion

Density (kg/m3)

3.41

2.83

2.53

0.56

-

-

3,110

0.59

-

0.27

0.17

0.86

1.06

2,150

0.82

-

10.04

-

-

-

2,900

EP

85

93.90

4.73

TE D

83

M AN U

SC

RI PT

75

Based on the previous study [23, 24], UHPC with a water to binder ratio (w/b) of 0.18 and silica

87

fume content of 20% to 25%, by mass of cementitious materials, can obtain a satisfactory bond to steel

88

fiber reinforcement and mechanical properties. Therefore, UHPC mixture with a w/b of 0.18 and silica

89

fume content of 25% was used in this study. The SP dosage was fixed at 2% by the mass of the binder. To

90

investigate the hybridization of short and long fiber reinforcements on quasi-static and dynamic properties

91

of UHPC, a reference mixture with no fiber reinforcement and five mixtures with a single type of fiber

92

reinforcement or hybrid fiber reinforcements at a total content of 2%, by volume of the concrete, were

93

used. Table 2 shows the five mixture proportions. L0S0 was the reference mixture with no fiber 4

AC C

86

ACCEPTED MANUSCRIPT 94

reinforcement. L1.5S0.5 was referred to the mixture with 1.5% long fiber reinforcements and 0.5% short

95

fiber reinforcements. Table 2 Mixture proportions of UHPC Silica Superplasticizer Cement Slag Sand Water No. fume 3 3 (kg/m3) (kg/m3) (kg/m ) (kg/m ) (kg/m3) (kg/m3) L0S0 472 315 262 1,049 178 21 L2S0 472 315 262 1,049 178 21 L1.5S0.5 472 315 262 1,049 178 21 L1S1 472 315 262 1,049 178 21 L0.5S1.5 472 315 262 1,049 178 21 L0S2 472 315 262 1,049 178 21

Long fiber (%) 0 2.0 1.5 1.0 0.5 0

Short fiber (%) 0 0 0.5 1.0 1.5 2.0

SC

RI PT

96 97

98

2.3. Mixing procedure and sample preparation

M AN U

99

During mixing, dry powders, including cement, silica fume, slag, and river sand, were first mixed for

101

3 min at a high speed. Water and superplasticizer were then added and mixed for approximately 6 min at a

102

low speed. Steel fiber reinforcements were slowly added and mixed for another 6 min until achieving a

103

uniform distribution.

TE D

100

All the mixtures were cast into molds for the testing of flowability and mechanical properties.

105

Specimens were then kept in a room at 20°C and relative humidity (RH) greater than 95% for

106

approximately 24 h. They were then demolded and cured in lime-saturated water at 20°C for 7 and 28 d.

EP

104

107

2.4. Experimental methods

109

2.4.1. Flowability

AC C

108

110

The flowability of all mixtures was measured in accordance with the Chinese Standard GB/T

111

2419-2005 [25]. The mixtures were cast into a mini cone mold and jolted for 25 times. Two diameters

112

perpendicular to each other were determined and mean values were reported.

113

5

ACCEPTED MANUSCRIPT 114

2.4.2. Static mechanical properties Specimens of 40 × 40 × 160 mm were cast for determination of static compressive and flexural

116

strengths. Three samples of each batch were tested. The flexural and compressive strengths measurements

117

were conducted according to GB/T 17671-1999 (similar to ISO 697:1989) [26]. Three-point bending

118

testing was first performed to obtain flexural properties. Six broken specimens measuring 40 × 40 × 40

119

mm were then tested for compressive strength. Two non-casting surfaces of the specimen were used as a

120

bottom and a top surface to ensure complete contact with the platen of the machine. The loading rate was

121

set at 2.4 kN/s. Averages over three samples for flexural strength and over six samples for compressive

122

strength were reported.

M AN U

SC

RI PT

115

123 124

2.4.4. Split Hopkinson pressure bar (SHPB) testing

A schematic illustration of SHPB testing equipment is shown in Fig. 1. The testing is based on the

126

theory of one-dimensional wave propagation in an elastic bar [27]. The specimen is placed between the

127

ends of the input and output bars. A stress wave is created and propagates through the bar toward the

128

specimen at the end of the input bar. This incident wave then splits into a transmitted wave and a reflected

129

wave upon reaching the specimen. The transmitted wave travels through the specimen and goes into the

130

output bar, causing plastic deformation in the specimen. The reflected wave is reflected away from the

131

specimen and travels back to the input bar. The whole process is recorded through a computer.

AC C

EP

TE D

125

132

The SHPB equipment used in this research is made of superior alloy steel with a diameter of 100 mm.

133

The lengths of the input and output bars are 6,000 and 4,000 mm, respectively. The distances of strain

134

gauge on the input and output bars to the specimen are 3,000 and 1,000 mm, respectively. The length of

135

the striker bar is 1,500 mm. Cylinder samples with a diameter of 92 mm and a height of 46 mm were used.

136

Before testing, the two ends of each specimen were ground to make them parallel so as to ensure the 6

ACCEPTED MANUSCRIPT reliability of the results.

RI PT

137

138

(a) Schematic illustration of SHPB equipment [28]

M AN U

SC

139 140 141

142

(b) Picture of SHPB equipment Fig. 1 SHPB testing equipment 3. Results and discussion

147

3.1. Effects of hybrid steel fiber reinforcements on flowability of UHPC

TE D

143 144 145 146

Figure 2 shows the effects of hybrid fiber reinforcements on flowability of the UHPC mixtures. The

149

UHPC mixture with no steel fiber reinforcement had the highest flowability of 248 mm. With the

150

incorporation of a single type of fiber reinforcement or hybrid fiber reinforcements, the flowability

151

decreased. This agrees well with the results reported by Yu et al. [29]. The reduced flowability associated

152

with the fiber reinforcements was attributed to the following reasons [30, 31]: (1) fiber reinforcements are

153

much more elongated and have higher surface area compared to aggregates, which can increase the

154

cohesive force; (2) steel fiber reinforcements can alter the granular skeleton structure and hinder the flow

155

of fresh paste; (3) steel fiber reinforcements tends to be perpendicular to the flow direction, generating

AC C

EP

148

7

ACCEPTED MANUSCRIPT 156

resistance force and moment to the flow velocity of fresh mixtures. It can be also observed from Fig. 2, the UHPC mixtures with the hybrid fiber reinforcements

158

demonstrated higher flowability than those with the single type of fiber reinforcement. The higher the

159

short fiber reinforcement volume in the UHPC mixture with hybrid fiber reinforcements was, the better

160

the flowability was. However, when all the 2% long fiber reinforcements were replaced by the short fiber

161

reinforcement, the mixture (L0S2) obtained the worst flowability. In the case of hybrid fiber

162

reinforcements, the mutual effects between the short and long fiber reinforcements can restrict each

163

other’s rotation to avoid the perpendicular orientation to the flow direction [29-31]. This can reduce the

164

resistance force to some extent, and eventually enhance the flowability when compared to those with the

165

single type of fiber reinforcement (L2S0 and L0S2 mixtures). The worst flowability of the L0S2 mixture

166

might be attributed to the increase in fiber reinforcement volume associated with short length.

M AN U

SC

RI PT

157

250

240

TE D

Flowability (mm)

245

235 230

EP

225

167 168 169 170

L0S0

L2S0 L1.5S0.5 L1S1 L0.5S1.5 L0S2 No.

AC C

220

Fig. 2 Effects of hybrid fiber reinforcements on flowability of UHPC

3.2. Effects of hybrid fiber reinforcements on static compressive and flexural strengths

171

The effects of hybrid fiber reinforcements on static compressive strength of UHPC are illustrated in

172

Fig. 3. The 7- and 28-d static compressive strengths of the reference UHPC mixture (L0S0) were 81.1 and

173

98.3 MPa, respectively. It increased obviously with the incorporation of steel fiber reinforcements. The 8

ACCEPTED MANUSCRIPT compressive strengths of the L2S0 mixture at 7 and 28 d were increased by 35% and 28%, respectively, in

175

comparison to the reference mixture. For the mixtures with hybrid fiber reinforcements, the compressive

176

strength increased first but then decreased with the increase of the short fiber reinforcement volume.

177

When 1.5% long fiber reinforcements and 0.5% short fiber reinforcements were used, the highest

178

compressive strength was achieved. Besides, the static compressive strength of the L2S0 mixture was

179

obviously greater than that of the L0S2 mixture. The decrease in compressive strength with the increase of

180

short fiber reinforcement content might be due to the lower efficiency of short fiber reinforcements in

181

restricting the development of cracks [11, 12, 29]. Therefore, the use of 2% hybrid fiber reinforcements

182

could increase the static compressive strength by 20% to 40% compared to the reference mixture.

M AN U

SC

RI PT

174

150 120 90

TE D

Compressive strength (MPa)

180

60

L0S0

L2S0 L1.5S0.5 L1S1 L0.5S1.5 L0S2 No.

EP

30

183

7d 28d

Fig. 3 Effects of hybrid fiber reinforcements on static compressive strength of UHPC

186

Figure 4 depicts the effects of hybrid fiber reinforcements on static flexural behavior of UHPC. The

187

UHPC reference specimens with no fiber reinforcement showed a sudden drop of load when reaching the

188

initial cracking point at a deflection of 0.25 mm, as observed from Fig. 4(a). UHPC specimens with fiber

189

reinforcement could continue to sustain load after the initial cracking until they reached the peak load. The

190

volume of long and short fiber reinforcements had a significant effect on the peak load. The L1.5S0.5

191

mixture was shown to exhibit the greatest peak load, followed by the L2S0 mixture. As can be also

AC C

184 185

9

ACCEPTED MANUSCRIPT 192

observed from Fig. 4(a), the load-deflection curves showed obvious zigzag patterns after the peak load.

193

This is consistent with the results in literature [11]. It was observed that there were tearing sounds after the

194

peak load, which indicates that the fiber reinforcements were continuously pulled out from the matrix. The change in static ultimate flexural strength of UHPC with hybrid fiber reinforcements showed a

196

similar trend to that of the static compressive strength, as illustrated in Fig. 4(b). The L1.5S0.5 mixture

197

showed the highest flexural strength of 32 MPa at 28 d, which increased by 108% compared to the

198

reference mixture. This might be due to the combined strengthening effects from the short and long fiber

199

reinforcements, which can restrict the crack development in two length scales, both in a scale of 6 mm and

200

13 mm length [12]. However, when more long fiber reinforcements were replaced by short fiber

201

reinforcements, the flexural strength decreased. The L0S2 mixture obtained the worst static flexural

202

strength among the five UHPC mixtures with fiber reinforcement.

M AN U

SC

RI PT

195

15000

40

TE D

Load (N)

12000 9000 6000

0

1

EP

3000 0

Flexural strength (MPa)

L0S0 L2S0 L1.5S0.5 L1S1 L0.5S1.5 L0S2

2

3

4

5

204 205 206

20

10

L0S0

L2S0 L1.5S0.5 L1S1 L0.5S1.5 L0S2 No.

AC C

Deflection (mm)

203

30

0

6

7d 28d

(a) Flexural load-deflection curves at 28 d (b) Flexural strengths at 7 and 28 d Fig. 4 Effects of hybrid fiber reinforcements on flexural behavior of UHPC

207

3.3. Effects of hybrid fiber reinforcements on dynamic compressive properties of UHPC

208

3.3.1. Effects of hybrid fiber reinforcements on dynamic failure pattern of UHPC

209

Figure 5 illustrates the dynamic failure pattern of UHPC with and without fiber reinforcement at 28 d.

210

At the impact velocity of 8.9 m/s, all the specimens exhibited a failure mode of pulverization. The 10

ACCEPTED MANUSCRIPT reference mixture (L0S0) with no fiber reinforcement showed a large quantity of small and irregular

212

broken concrete fragments. The use of steel fiber reinforcements could effectively restrict the transverse

213

deformation of specimens. As can be seen from Fig. 5(b-d), the sizes of the broken fragments of the

214

UHPC specimens varied with the hybrid fiber reinforcement volumes. The L2S0 mixture showed relative

215

large and round broken fragments. For the L1.5S0.5 mixture, balanced broken fragments with both large

216

and small parts could be observed. Many embedded fiber reinforcements were observed in the fracture

217

surface, bridging the small fragments. However, relatively small and elongated broken fragments were

218

observed for the L0S2 mixture, as can be observed from Fig. 5(d).

SC

RI PT

211

(a) L0S0 mixture

(b) L2S0 mixture

AC C

221

EP

220

TE D

M AN U

219

222

11

ACCEPTED MANUSCRIPT 223 224 225 226

(c) L1.5S0.5 mixture (d) L0S2 mixture Fig. 5 Effects of hybrid fiber reinforcements on dynamic failure patterns of UHPC at an impact velocity of 8.9 m/s

227

3.3.2. Effects of strain rates on dynamic compressive load-deflection relationship of UHPC The effects of different strain rates on dynamic compressive load-deflection curves of UHPC with and

229

without fiber reinforcement at 28 d are illustrated in Fig. 6. For all UHPC mixtures, the stress changed

230

almost linearly with the strain up to the peak stress. Afterward, the stress-strain curves showed a

231

decreasing branch. When the strain reached approximately 0.04, a constant stress was observed.

RI PT

228

From Fig. 6, it can be also seen that the dynamic compressive behavior of UHPC was sensitive to

233

strain rate. All the mixtures showed a strain rate effect, that is, the peak stress increased with the increase

234

of strain rate. The peak stress for the L2S0 mixture was 141.2 MPa at the strain rate of 115.5 s-1, and

235

increased to 160.6 and 177.4 MPa, respectively, when the strain rate increased to 158.2 and 195.8 s-1. Li

236

and Cotsovos [17, 32, 33] proposed that this strain rate effect was due to lateral inertia associated with

237

friction to the contact surface under rapid impact loading. Besides, the increasing rate for peak stress

238

decreased gradually with the increase of strain rate. This might be attributed to the Stefan effect associated

239

with the presence of free water in concrete [34].

TE D

M AN U

SC

232

206.1/s 169.8/s

150

AC C

Stress (MPa)

150

180

Stress (MPa)

180

210

EP

210

120

130.0/s

90

120

60

30

30

0.02

0.04 0.06 Strain

0.08

0.10

158.2/s 115.5/s

90

60

0.00

195.8/s

0.00

0.02

0.04

0.06 Strain

240 241

(a) L0S0

(b) L2S0

12

0.08

0.10

ACCEPTED MANUSCRIPT

Stress (MPa)

150

155.7/s 114.8/s

90

107.8/s

120 90

60

60

30

30 0.02

158.7/s

150

120

0.00

195.4/s

180

0.04 0.06 Strain

0.08

0.10

0.00

RI PT

180

210

194.8/s

Stress(MPa)

210

0.02

210 194.7/s 155.6/s

150

Stress(MPa)

Stress (MPa)

111.8/s

90

120

0.08

0.10

156.1/s

109.0/s

90 60

60

30

0.02

0.04

0.00

TE D

30 0.06

0.08

0.10

0.02

0.04

Strain

244

0.06

194.8/s

M AN U

180

150

0.00

0.10

(d) L1S1

210

120

0.08

SC

(c) L1.5S0.5

180

0.06

Strain

242 243

0.04

Strain

(e) L0.5S1.5 (f) L0S2 Fig. 6 Effects of strain rates on dynamic compression stress-strain curves of UHPC at 28 d

248

The dynamic increase factor (DIF) was introduced to give a comparision between the quasi-static and

249

dynamic compressive properties of UHPC , as summerized in Table 3. DIF is defined as the ratio of the

250

dynamic compressive strength (peak stress) to the static compressive strength at curing age of 28 d. An

251

accurate DIF can ensure an appropriate stucture dimension design, reduced maximum amount of materials,

252

and eventually enhanced energy efficiency and environmental protection [28].

AC C

EP

245 246 247

253

As can be seen from Table 3, all the DIFs were larger than 1. This suggests the dynamic compressive

254

strength was greater than the static compressive strength. Besides, the DIFs increased with the increase of 13

ACCEPTED MANUSCRIPT strain rate. It is known that weak zones like pores and micro-cracks intrinsically exist in the UHPC

256

specimens. Under static loading, there is enough time for only a few micro-cracks in local area to develop

257

into macro-cracks, and eventually lead to failure of concrete. However, the loading time is extremely short

258

under impact loading. Before one or two micro-cracks slowly propagate propogates into macro-cracks,

259

more new micro-cracks have been nucleated and enlarged [35, 36]. Such nucleation of a large number of

260

cracks can consume more energy, and hence results in higher ultimate dynamic compressive strength.

261

Moreover, the lateral confinement from both contact surface restriction and lateral inertia can also cause

262

enhancement in dynamic strength during the rapid loading process [36].

SC

RI PT

255

The increased rate for DIF decreased with further increase of strain rate. This is consitent with the

264

results in Ref. [36, 37]. Under low strain rate, cracks in concrete are limited in size and quantity, which

265

can be efficiently prevented by the fiber reinforcements. However, a large quantity of cracks with

266

increasing size would be induced at high strain rate, which is less efficient for fiber reinforcement to exert

267

their influence.

TE D

Table 3 Summarization of static and dynamic compressive properties of UHPC at 28 d Dynamic Static Strain at Dynamic Fracture Impact Strain compressive compressive No. velocity rate peak stress increase energy strength strength -1 -3 (m/s) (s ) (×10 ) factor (MJ/m3) (MPa) (MPa) 8.9 130.0 100.4 6.01 98.3 1.02 1.76 L0S0 11.6 169.8 128.4 5.24 98.3 1.31 1.94 14.0 206.1 146.8 7.54 98.3 1.49 3.10 8.9 115.5 141.2 5.32 134.3 1.05 2.62 L2S0 11.7 158.2 160.6 6.63 134.3 1.20 3.76 13.9 195.8 177.4 6.89 134.3 1.32 3.90 8.9 114.8 159.7 4.06 143.6 1.11 2.72 L1.5S0.5 11.7 155.7 184.3 4.22 143.6 1.28 3.29 13.9 194.8 204.8 7.38 143.6 1.43 4.15 8.9 107.8 151.5 4.28 127.7 1.19 2.59 L1S1 11.7 158.7 177.2 4.67 127.7 1.39 3.27 14.0 195.4 192.0 6.17 127.7 1.50 3.88 8.9 111.8 124.0 3.76 123.2 1.01 2.41 L0.5S1.5 11.6 155.6 171.8 4.50 123.2 1.39 3.17

EP

269

AC C

268

M AN U

263

14

ACCEPTED MANUSCRIPT

L0S2

13.9 8.8 11.6 13.9

194.7 109.0 156.1 194.8

190.4 128.8 161.5 177.9

6.87 3.69 6.48 6.17

123.2 120.5 120.5 120.5

1.55 1.07 1.34 1.48

3.61 2.36 3.06 3.50

270

3.3.3. Effects of hybrid fiber reinforcements on dynamic compressive load-deflection relationship of

272

UHPC

RI PT

271

Figure 7 compares the dynamic compression stress-strain curves of UHPC with hybrid steel fiber

274

reinforcements under the same impact velocity. The reference mixture L0S0 showed the lowest dynamic

275

compressive strength and toughness, whereas the L1.5S0.5 mixture exhibited the highest corresponding

276

values followed by the L1S1 mixture and then the L0.5S1.5 mixture. This agrees well with the results of

277

static compressive properties. Moreover, the UHPC mixtures with the single type of fiber reinforcement

278

(L0S2 and L2S0 mixtures) had comparable dynamic peak stress. Besides, the dynamic compressive peak

279

stress was higher than that of the static compressive strength, especially at higher impact velocity.

M AN U

SC

273

As we know, the impact energy that needed to break the concrete includes two parts: the energy to

281

break the matrix and the energy to pull out the fiber reinforcements. Under impact loading, the impact

282

velocity was extremely fast and the loading time was extremely short. A failure mode of pulverization

283

could occur under impact loading at high impact velocity, which is different from a local failure mode

284

under static compressive loading. Randomly oriented fiber reinforcements can form a network structure in

285

concrete. The internal structure of concrete, including fiber geometry, spacing, and quantity, can have a

286

combined effect on the dynamic properties. A good balanced effect associated with these parameters can

287

provide satisfactory mechanical properties for the composite materials. Generally, for a given amount of

288

fiber reinforcement with single type, the longer the fiber reinforcement is, the less the fiber number is and

289

the larger the fiber spacing is [38]. The contribution from the longer length of long fiber reinforcements in

290

the L2S0 mixture would compensate those from the higher fiber spacing and the less fiber number, and

291

vice versa for the L0S2 mixture. This eventually made the dynamic properties of the L2S0 and L0S0 15

AC C

EP

TE D

280

ACCEPTED MANUSCRIPT mixtures comparable. 210

210

180

180

L1.5S0.5 L1S1 L2S0

150

120 90

L0.5S1.5 L0S2

60 30

120 90

L2S0

L0S0

30 0.02

0.04

0.06

0.08

0.10

0.00

Strain

0.02

0.04

0.06

0.08

0.10

SC

Strain

293

(a) 8.9 m/s

(b) 11.7 m/s

210 180

Stress (MPa)

150 120 90

L1.5S0.5L1S1

L0.5S1.5

L0S0

L2S0

L0S2

TE D

60

M AN U

294

L0S2

60

L0S0

0 0.00

L1.5S0.5 L1S1 L0.5S1.5

RI PT

Stress (MPa)

150

Stress (MPa)

292

30

0.00

0.06

0.08

0.10

Strain

(c) 13.9 m/s Fig. 7 Effects of hybrid fiber reinforcements on dynamic compression stress-strain curves of UHPC at 28 d

EP

296 297 298 299

0.04

AC C

295

0.02

300

Table 3 also summarizes the toughness or fracture energy of UHPC, which can be expressed as the

301

area surrounded by the stress-strain curve to a strain of 0.04. The use of steel fiber reinforcement

302

significantly increased the dynamic compressive toughness. The L1.5S0.5 mixture again showed the

303

highest toughness, which increased by 54.5% and 33.9% at the impact velocities of 8.9 and 13.9 m/s,

304

respectively. With the increase of short fiber reinforcement volume, the toughness decreased.

305

16

ACCEPTED MANUSCRIPT 306

4. Discussion The results above showed that the use of single type of fiber reinforcement and hybrid fiber

308

reinforcements significantly improved the static compressive and flexural properties, as well as the

309

dynamic compressive properties of UHPC. The L1.5S0.5 mixture obtained the highest static and dynamic

310

compressive properties. However, with the increase of short fiber reinforcement content, these properties

311

decreased.

RI PT

307

As we know, there existed intrinsically weak zones, such as micro-cracks and pores, in UHPC, as

313

shown in Fig. 9. The propagation of micro-cracks can be initiated at the weak zones. This can lead to

314

crack tip stress concentration under external forces. To explain the reason for enhanced strength of a

315

composite material associated with fiber reinforcements, Romualdi et al. [38, 39] proposed a model of

316

crack restraining mechanism, as shown in Fig. 8. It was assumed that continuous fiber reinforcements are

317

evenly distributed in the matrix along the loading direction. A crack with radius of a is surrounded by four

318

fiber reinforcements with a fiber spacing of S. Under tensile loading, the region of the crack around the

319

fiber reinforcements can produce cohesive force distribution, as shown in Fig. 8(b). The cohesive force at

320

the crack tip can yield reverse stress field. This can reduce the degree of stress concentration and restrain

321

the propagation of cracks, and hence improving the strength. However, with further increase in the stress,

322

the micro-cracks can propagate into macro-cracks and eventually lead to the failure of concrete members.

AC C

EP

TE D

M AN U

SC

312

S

2a

323 324 325

(a) (b) Fig. 8 Crack restraining mechanism of fiber reinforcements [38, 39] 17

ACCEPTED MANUSCRIPT When hybrid fiber reinforcements were incorporated, the combined effect associated with short and

327

long fiber reinforcements at different length levels could be exerted, as shown in Fig. 9. The initiated

328

micro-cracks can be restrained by short fiber reinforcements. When the micro-cracks propagate into

329

macro-cracks under further loading, the short fiber reinforcements around the cracks are gradually pulled

330

out and the load is mainly sustained by the long fiber reinforcements. This ensures that more time and

331

energy are needed to damage the concrete.

RI PT

326

SC

332 333

M AN U

334 335

337 338 339

Stage II - action of short fibers. Propagation of micro-cracks.

Stage I - no fiber action. Microcracking initiated at weak zones of concrete.

Stage III - action of long fibers. Formation of macro-cracks and eventually cracking of concrete.

Fig. 9 Crack control mechanism for UHPC with hybrid steel fiber reinforcements

TE D

336

The L2S0 and L0S2 mixtures with the single type of fiber reinforcement demonstrated similar

341

dynamic compressive properties, which were different from static compressive properties. This is because

342

the impact velocity was extremely fast and the loading time was extremely short under impact loading. A

343

large number of micro- and macro-cracks can be nucleated and enlarged before the limited micro-cracks

344

developed into macro-cracks under impact loading. Moreover, the quality of the structure, such as fiber

345

geometry, spacing, and quantity, can have a combined effect on the mechanical properties of concrete. For

346

a given volume of single type of fiber reinforcement at a certain fiber reinforcement orientation, the longer

347

the fiber reinforcement is, the less the fiber number is and the larger the fiber spacing is. It was reported

348

that the tensile strength decreasd with the increase of the fiber spacing, but increased with the increase of

AC C

EP

340

18

ACCEPTED MANUSCRIPT the fiber length [40, 41]. Under impact loading, the dynamic properties of UHPC are much more sensitive

350

to the internal strucutre of the composite materials in such a very short time when compared to the static

351

properties. Therefore, only when the combined effect from these parameters reached an optimal state, can

352

the maximum dynamic compressive properties be obtained.

RI PT

349

353 354

5. Conclusions

This study investigated the effects of hybrid steel fiber reinforcements on flowability, static

356

compressive and flexure properties, as well as dynamic compressive behavior of UHPC. A reference

357

UHPC mixture with no fiber reinforcement and five mixtures with a single type of fiber reinforcement or

358

hybrid steel fiber reinforcements at a total content of 2%, by the volume of concrete, were prepared. Based

359

on the results presented in this study, the following conclusions can be drawn:

M AN U

SC

355

(1) The use of steel fiber reinforcement significantly reduced the flowability of UHPC mixtures.

361

UHPC with the single type of fiber reinforcement demonstrated lower flowability than those with hybrid

362

fiber reinforcements. Among the five UHPC mixtures with fiber reinforcements, the mixture with 2%

363

single type of short fiber reinforcements exhibited the worst flowability. For UHPC with hybrid fiber

364

reinforcements, the flowability gradually increased with the increase of the short fiber reinforcement

365

volume.

EP

TE D

360

(2) Fiber hybridization significantly improved the static compressive and flexural properties of UHPC,

367

as well as the dynamic compressive behavior. The use of steel fiber reinforcement could increase the static

368

compressive strength by 20% to 40%. The L1.5S0.5 mixture demonstrated both the best static and

369

dynamic compressive behaviors. The static compressive and flexural properties of UHPC with 2% long

370

fiber reinforcements were significantly higher than those with 2% short fiber reinforcements, while

371

similar dynamic compressive properties were observed. The dynamic compressive properties, including

AC C

366

19

ACCEPTED MANUSCRIPT 372

peak stress, dynamic increase factor, and fracture energy of UHPC, increased gradually with the increase

373

of the strain rate. (3) The reinforcing mechanisms of hybrid fiber reinforcements in UHPC were due to the combined

375

strengthening effects from both the short and long fiber reinforcements, which can restrict the crack

376

development at two length scales. Initially, the initiated micro-cracks were restrained by the short fiber

377

reinforcements. With the micro-cracks propagated further, the short fiber reinforcements were pulled out

378

from the matrix while the long fiber reinforcements began to sustain load. This means that more time and

379

energy are needed to crack the concrete compare to that with the single type of fiber reinforcement. Under

380

impact loading, the impact velocity was extremely fast and the loading time was extremely short. The

381

nucleation and enlargement of cracks, lateral confinement from both the contact surface restriction and

382

lateral inertia could contribute to the dynamic compressive behavior.

M AN U

SC

RI PT

374

383

385

Acknowledgements

TE D

384

Financial support from National Science Foundation of China under contract Nos. 51378196 and U1305243 is greatly appreciated.

387 388

References

389

[1] Richard P, Cheyrezy M. Reactive powder concretes with high ductility and 200-800 MPa compressive

391 392 393 394 395

strength. ACI Mater J 1994; 144(3): 507-518.

AC C

390

EP

386

[2] Shi C, Wu Z, Wang D, Xiao J, Huang Z, Fang Z. A review on ultra high performance concrete: Part I. Raw materials and mixture design. Constr Build Mater 2015; 101: 741-751. [3] Zhang J, Zhao Ya. The mechanical properties and microstructure of ultra-high-performance concrete containing various supplementary cementitious materials. J Sustain Cem Mater 2016: 1-13. [4] Wu Z, Shi C, Khayat KH, Wan S. Effects of different nanomaterials on hardening and performance of

20

ACCEPTED MANUSCRIPT

399 400 401 402 403 404 405 406

reinforced concrete. Cem Concr Compos 2012; 34(2): 172-184. [6] Habel K and Gauvreau P. Response of ultra-high performance fiber reinforced concrete (UHPFRC) to

RI PT

398

[5] Park SH, Kim DJ, Ryu GS, Kohb KT. Tensile behavior of ultra high performance hybrid fiber

impact and static loading. Cem Concr Compos 2008; 30(10): 938-946.

[7] Wang D, Shi C, Wu Z, Xiao J, Huang Z, Fang Z. A review on ultra high performance concrete: Part II. Hydration, microstructure and properties. Constr Build Mater 2015; 96: 368-377.

[8] Tuan NV, Ye G, Breugel KV, Copuroglu O. Hydration and microstructure of ultra high performance

SC

397

ultra-high strength concrete (UHSC). Cem Concr Compos 2016; 70: 24-34.

concrete incorporating rice husk ash. Cem Concr Res 2011; 41(11): 1104-1111.

M AN U

396

[9] Kim DJ, Park SH, Ryu GS, Koh KT. Comparative flexural behavior of hybrid ultra-high performance fiber reinforced concrete with different macro fibers. Constr Build Mater 2011; 25:4144-4155. [10] Hannawi K, Bian H, Prince-Agbodjan W, Raghavan B. Effect of different types of fibers on the

408

microstructure and the mechanical behavior of ultra-high performance fiber-reinforced concretes.

409

Composites Part B: Eng. 2016; 86: 214-220.

412 413

high performance concrete. Constr Build Mater 2016; 103: 8-14.

EP

411

[11] Wu Z, Shi C, He W, Wu L. Effects of steel fiber content and shape on mechanical properties of ultra

[12] Wu Z, Shi C, He W, Wang D. Uniaxial compression behavior of ultra-high performance concrete with hybrid steel fiber. Journal of Materials in Civil Engineering; 2016: 06016017-1-7.

AC C

410

TE D

407

414

[13] Kwon S, Nishiwaki T, Kikuta T, Mihashi H. Development of ultra-high-performance hybrid

415

fiber-reinforced cement-based composites (with Appendix). ACI Materials Journal 2014; 111(3):

416

309-318.

417 418

[14] Ku H, Cheng YM, Snook C, Baddeley D. Drop weight impact test fracture of vinyl ester composites: micrographs of pilot study. Journal of composite materials 2005; 39(18): 1607-1620. 21

ACCEPTED MANUSCRIPT

425 426 427 428 429 430 431 432 433 434 435 436

RI PT

424

[17] Li Q, Meng H. About the dynamic strength enhancement of concrete-like materials in a split Hopkinson pressure bar test. International Journal of Solids and Structures 2003; 40: 343-360. [18] Mohammadi Y, Carkon-Azad R, Singh SP, Kaushik SK. Impact resistance of steel fibrous concrete containing fibres of mixed aspect ratio. Constr Build Mater 2009; 23: 183-189.

SC

423

reinforced lightweight aggregate concrete. Constr Build Mater 2013; 38: 1146-1151.

[19] Rong Z, Sun W, Zhang Y. Dynamic compression behavior of ultra-high performance cement based

M AN U

422

[16] Wang H, Wang L. Experimental study on static and dynamic mechanical properties of steel fiber

composites. International Journal of Impact Engineering 2010; 37(5): 515-520. [20] Lai J, Sun W, Lin W, Jin Z. Static and dynamic mechanical behavior of ECO-RPC. Journal of Southeast University. 2005: 2.

[21] Zhang Y, Sun W, Liu S, Jiao C. Preparation of C200 green reactive powder concrete and its

TE D

421

concrete material with different fibres. Materials & Design 2012; 33: 42-55.

static-dynamic behaviors. Cem Concr Compos 2008; 30(9): 831-838. [22] Chinese national standard. Chinese cement: common portland cement, GB175-2007. Beijing, China, 2007.

EP

420

[15] Xu Z, Hao H, Li H. Experimental study of dynamic compressive properties of fibre reinforced

[23] Shi C, Wang D, Wu L, Wu Z. The Hydration and microstructure of ultra high-strength concrete with cement-silica fume-slag binder. Cem Concr Compos 2015; 61: 44-52.

AC C

419

437

[24] Wu Z, Shi C, Khayat KH. Influence of silica fume content on microstructure development and bond

438

to steel fiber in ultra-high strength cement-based materials (UHSC). Cem Concr Compos 2016; 71:

439

97-109.

440 441

[25] Chinese national standard. Test methods for flowability of cement paste, GB2419-2005, Beijing, China, 2005. 22

ACCEPTED MANUSCRIPT 442 443

[26] China National Standards. Method of testing cements - determination of strength, GB/T 17671-1999, Beijing, China (in Chinese), 1999. [27] Yang H, Li J, Huang Y. Study on the mechanical properties of high performance hybrid fiber

445

reinforced cementitious composite (HFRCC) under impact Loading. in Key Engineering Materials.

446

2015: Trans Tech Publ.

447 448

RI PT

444

[28] Zhang W. Investigation of microstructure formation mechanism and dynamic mechanical behavior of UHPCC. Ph.D. thesis, Southeast University. Nanjing, December 2012.

[29] Yu R, Spiesz P, Brouwers HJH. Static properties and impact resistance of a green ultra-high

450

performance hybrid fibre reinforced concrete (UHPHFRC): experiments and modeling. Constr Build

451

Mater 2014; 68: 158-171.

453

M AN U

452

SC

449

[30] Grünewald S. Performance-based design of self-compacting fiber reinforced concrete. Delft, the Netherlands: Delft University of Technology; 2004.

[31] Boulekbache B, Hamrat M, Chemrouk M, Amzianec S. Flowability of fibre-reinforced concrete and

455

its effect on the mechanical properties of the material. Constr Build Mater 2010; 24(9): 1664-1671.

456

[32] Cotsovos D, Pavlović M. Numerical investigation of concrete subjected to compressive impact

457

loading. Part 1: A fundamental explanation for the apparent strength gain at high loading rates.

458

Computers & structures 2008; 86(1): 145-163.

EP

TE D

454

[33] Cotsovos D, Pavlović M. Numerical investigation of concrete subjected to compressive impact

460

loading. Part 2: Parametric investigation of factors affecting behaviour at high loading rates.

461

Computers & structures 2008; 86(1): 164-180.

462 463 464

AC C

459

[34] Bischoff P, Perry S. Compressive behaviour of concrete at high strain rates. Materials and Structures 1991; 24(6): 425-450. [35] Lee MK, Barr BIG. An overview of the fatigue behavior of plain and fiber reinforced concrete. Cem 23

ACCEPTED MANUSCRIPT

466 467 468 469 470 471

Concr Compos 2004; 26 (4): 299-305. [36] Wang Y, Wang Z, Liang X, An M. Experimental and numerical studies on dynamic compressive behavior of reactive powder concretes. Acta Mechanica Solida Sinica, 2008; 21(5): 420-430. [37] Tai Y. Uniaxial compression tests at various loading rates for reactive powder concrete. Theoretical and Applied Fracture Mechanics 2009; 52(1): 14-21.

RI PT

465

[38] Romualdi JP, Mandel JA. Tensile strength of concrete affected by uniformly dispersed and closely spaced short length of wire reinforcement. ACI J Proc 1964; 61(6):657-671. [39] Romualdi JP, Batson GB. Mechanics of crack arrest in concrete, 1963.

473

[40] Mangat PS. Tensile strength of steel fiber reinforced concrete. Cem Concr Res 1976; 6(2): 245-252.

474

[41] Shah SP, Rangan BV. Fiber reinforced concrete properties. In ACI Journal Proceedings 1971; 68(2):

M AN U

475

SC

472

126-137.

AC C

EP

TE D

476

24