This document summarizes a study on the dynamic tensile testing of fabric-cement composites. Three types of fabric-cement composites were tested under high strain rates: AR-glass fabric composite, PE fabric composite, and carbon fiber composite. The testing found that carbon fiber composite exhibited the highest strength and stiffness. Differences in tensile behavior were observed between the composites. Multiple cracking was observed in all composites except the PE composite with plain cement matrix. The results demonstrate the reliability of using high-speed tensile testing for cement-based composites.

1. Dynamic Tensile Testing of Fabric-Cement Composites
Deju ZHU, Barzin MOBASHER
Civil and Environmental Engineering, Arizona State University, USA
Flavio SILVA
Institute of Construction Materials – Technical University of Dresden, Dresden, Germany
Alva PELED
Structural Eng. Dept., Ben Gurion University, Israel
International Conference on Material Science and 64th RILEM Annual Week
Aachen, Germany, Sept. 6– 9, 2010

2. Outline
 Introduction
 Objectives
 Setup of dynamic tensile testing
 Specimen preparation and data analysis
 Failure behavior of composites
 Conclusions

3. Introduction
•HPFRCC, Strain-hardening behavior
•Cases for dynamic loading:
• blast explosions,
• projectiles,
• earthquakes,
• fast moving traffic,
• wind gusts, wind driven objects,
• machine vibrations.
•Inherent brittleness and low tensile strength,
dynamic loading can cause severe damage.
•mechanical properties at high strain rates for
analysis of structural components.
0 0.01 0.02 0.03 0.04
Strain, mm/mm
0
4
8
12
16
20
Stress,MPa
AR Glass Fabric
GFRC
Vf =5%
PE Fabric
E-Glass
Fabric
Mortar
ECC

4. Objectives
Study the role of fabric type on the tensile behavior of textile composites
 Study the strain rate effect on the mechanical material properties
 Investigate the failure behavior (cracking) of different composite system
under dynamic loading

5. High Speed Testing System at ASU
Servohydraulic system, Speed Up to 14 m/s, Load Capacity: 90 kN load

6. High Strain Rate Tensile Test

7. AR-Glass- bonded Carbon FabricPE- Knitted
Textile Materials
Alkali Resistant (AR)-Glass: Bonded fabric, multifilament impregnated with sizing
Polyethylene (PE): Knitted fabric, monofilament
Carbon Fabric: Fiber Bundle
5mm
Yarn type Yarn nature
Strength
(MPa)
Young’s Modulus
(MPa)
Filament size
(mm)
Bundle diameter
(mm)
PE Monofilament 240 1760 0.250 0.25
AR-Glass Bundle 1372 72000 0.014 0.30
Carbon Bundle 2200 240000 0.008 1.15
Table 1 Mechanical properties and geometry of the yarns used in the fabric manufacture

8. Composite Preparation: Pultrusion Process
Cement Composite
Laminates
Squeeze Rollers
Cement
Bath
Fabric
Mix design of the matrix
Cement: 42%
Silica fume: 0% and 5%
Super-plasticizer: 0.1%
water/cement ratio :0.4
Composite: 6 layer fabrics
Curing: in water at room
temperature for 28 days
Cut into specimens: 25mm
x150 mm (width x length).

9. Data Processing

10. Data Processing

11. Data Processing

12. Typical Response under high speed Tension

13. Videos of AR-Glass Fabric and Composites
AR-Glass CompositeAR-Glass Fabric

14. Failure of AR-Glass Composite
(a) (b) (c) (d) (e)
T= 0.4 ms T= 0.6 ms T= 0.8 ms T= 1.5 ms
Yarn
breakage
T= 4.0 ms
AR glass composites
(plain cement):
(a)-(c) multiple micro-
cracking,
(d) main crack widening
and other micro-cracks
closing,
(e) complete failure
(a) (b) (c) (d) (e)
T= 0.5 ms T= 5.8 ms T= 4.0 msT= 0 ms T= 1.5 ms
AR glass composites
(5%silica fume):
(a)-(c) multiple micro-
cracking,
(d) main crack widening
and other micro-cracks
closing,
(e) complete failure

15. Microstructure of AR-Glass Composite
Bundle
with
coating
Matrix
SEM micrographs of AR glass fabric embedded in
cement matrix, side view

16. Stress-Strain Curves of AR-Glass Fabric
and AR-Glass Composite
Stress-strain of AR-Glass
Composite(5% Silica Fume)
Stress-strain of AR-Glass
Fabric
0 0.01 0.02 0.03
Strain, mm/mm
0
2
4
6
8
10
Stress,MPa
Strain Rate, s-1
22
21
13
17
23
AR_Glass_SF
0 0.004 0.008 0.012 0.016
Strain, mm/mm
0
2
4
6
8
Stress,MPa
Strain Rate, s-1
21
15
17
15
AR_Glass_PC
Stress-strain of AR-Glass
Composite (plain cement)
0 0.02 0.04 0.06
Strain, mm/mm
0
1000
2000
3000
4000
Stress,MPa
Strain Rate, s-1
19
18
22
18
23
AR Glass Specimen
(4 yarns)

17. Videos of PE Fabric and PE Composite
PE Composite with Silica FumePE Fabric

18. Failure of PE Composite
(a) (b) (c) (d) (e)
T= 3.8 ms T= 5.8 ms T= 11.8 msT= 0 ms T= 9.8 ms
(a) (b) (c) (d) (e)
T= 0 ms T= 0.5 ms T= 1.0 ms T= 1.5 ms T= 3.5 ms
Crack
PE composites
(plain cement):
(a) intact,
(b) single cracking,
(c)-(d) crack widening,
(e) complete failure.
PE composites
(5% silica fume):
(a)intact,
(b-d) developing of multiple
cracks,
(e) complete failure

19. Microstructure of PE Composite
(a) (b)
SEM micrographs of PE fabric in the cement matrix and (a) cross section
of reinforcing yarn, (b) top view of loop

20. Stress-Strain Curves of PE Fabric and PE
Composite
0 0.05 0.1 0.15 0.2 0.25
Strain, mm/mm
0
1
2
3
Stress,MPa
Strain Rate, s-1
22
21
22
21
23
PE_SF
Stress-strain of PE Composite
(5% Silica Fume)
Stress-strain of PE Composite
(plain cement)
0 0.01 0.02 0.03 0.04
Strain, mm/mm
0
0.4
0.8
1.2
1.6
Stress,MPa
Strain Rate, s-1
26
21
18
26
PE_PC

21. Videos of Carbon Fiber Bundle and
Carbon Composite
Carbon Composite (plain cement)Carbon Fiber Bundle

22. Failure of Carbon Composite
(a) (b) (c) (d) (e)
T= 0 ms T= 1.0 ms T= 1.5 ms T= 2.0ms T= 2.5 ms
Carbon composites:
(a)intact,
(b) few of cracks visible,
(c-e) yarn pullout
(a) (b) (c) (d) (e)
T= 0 ms T= 1.0 ms T= 2.0 ms T= 4.0 ms T= 6.0 ms
Carbon composites
(multiple cracking):
(a)intact,
(b) multiple cracking,
(c)-(e) multiple crack
widening

23. Microstructure of Carbon Composite
(a) (b)
Matrix
Embedded
filaments
SEM micrographs of carbon fiber bundle embedded in cement matrix:
(a) view of cross section and (b) side view

24. Stress-Strain Curves of Carbon Fiber and
Carbon Composite
0 0.005 0.01 0.015 0.02 0.025
Strain, mm/mm
0
400
800
1200
1600
Stress,MPa
Strain Rate, s-1
17
20
18
17
21
Carbon Fibre Bundle
0 0.04 0.08 0.12
Strain, mm/mm
0
4
8
12
16
20
Stress,MPa
Strain Rate, s-1
9
11
9
Carbon Composite
Stress-Strain of Carbon CompositeStress-Strain of Carbon Fiber Bundle

25. Comparison of Different Composites
0 0.02 0.04 0.06 0.08 0.1
Strain, mm/mm
0
4
8
12
16
20
Stress,MPa
Carbon Composite
PE Composite
AR Glass Composite
Comparison of typical stress-strain curves
of different composites (plain cement)
0 0.05 0.1 0.15 0.2 0.25
Strain, mm/mm
0
4
8
12
16
20
Stress,MPa
Carbon Composite
PE Composite
AR Glass Composite
Carbon Composite (plain cement)
AR-glass and PE composites (5% silica fume)

26. Summary Tables
Composites
Strain Rate
(1/s)
Young’s Modulus
(MPa)
Strength
(MPa)
Toughness
(MPa)
Maximum strain
(mm/mm)
PE 23 (4) 140 (24) 1.31(0.17) 0.016(0.007) 0.021(0.007)
AR-Glass 18(3) 1176(323) 5.56(0.51) 0.032(0.009) 0.01(0.002)
Carbon 10(1) 2247(463) 17.86(0.82) 1.21(0.14) 0.10(0.014)
Table 2- Composites Properties (plain cement) under High Rate Loading
Composites
Strain Rate
(1/s)
Young’s Modulus
(MPa)
Strength
(MPa)
Toughness
(MPa)
Maximum strain
(mm/mm)
PE 22 (1) 156 (15) 2.33(0.28) 0.343(0.101) 0.207(0.04)
AR-Glass 19(4) 919(164) 8.27(0.86) 0.127(0.024) 0.025(0.004)
Table 3- Composites Properties (5% silica fume) under High Rate Loading
(the values in parenthesis are standard deviation)
Deju Zhu, Alva Peled, Barzin Mobasher. Dynamic Tensile Testing of Fabric-Cement Composites.
Construction and Building Materials, 2010, in press.

27. Strain Rate Effect on Composites with
plain cement
0 0.005 0.01 0.015
Strain, mm/mm
0
2
4
6
8
Stress,MPa
Strain rate = 2.2x10-5 s-1
Strain rate = 18 s-1
(a) AR Glass composite
0 0.01 0.02 0.03
Strain, mm/mm
0
0.3
0.6
0.9
1.2
1.5
1.8
Stress,MPa
Strain rate = 2.2x10-5 s-1
Strain rate = 23 s-1
(b) PE composite
0 0.02 0.04 0.06 0.08 0.1
Strain, mm/mm
0
5
10
15
20
25
Stress,MPa
Strain rate = 2.2x10-5 s-1
Strain rate = 10 s-1
(c) Carbon composite
Composite
High speed loading
(1000 mm/s)
Quasi-static loading
(0.004 mm/s)a
Strain Rate
(1/s)
Young’s
Modulus (MPa)
Strength
(MPa)
Toughness
(MPa)
Max.
strain
(mm/mm)
Strength
(MPa)
Max. strain
(mm/mm)
PE
23
(4)
140
(24)
1.31
(0.17)
0.016
(0.007)
0.021
(0.007)
1.36
(0.12)
1.98
(0.97)
AR Glass
18
(3)
1176
(323)
5.56
(0.51)
0.032
(0.009)
0.01
(0.002)
5.11
(0.25)
1.03
(0.07)
Carbon
10
(1)
2247
(463)
17.86
(0.82)
1.21
(0.14)
0.10
(0.014)
26.63
(2.87)
0.03
(0.01)

28. Applied Velocity
X (Warp Yarn)
Z
Y (Fill Yarn)

29. Micromechanical modeling of Textile Dry
fabrics
29
•modeling fill and warp yarns and capturing yarn to yarn interaction.
•modeling of contact surfaces as well as mass scaling.
•BCs: Left end of the fabric is fixed and velocity is applied on right end.
• Both contact types (SOFT = 1 & 2) were used

30. Design of Textile Containment system
30
LG689 Experimental
(47%) Single Layer
(37%)
Multi Layer
(31%)

31. Projectile Capture
31
LG657 Experimental
(100%) Single Layer
(100%)
Multi Layer
(100%)

32. Conclusions
 High speed tensile tests of three types of fabric-cement composites
were performed. A fairly uniform tensile behavior was clearly observed
of all composite systems, demonstrating the reliability of the high
speed test method for testing cement-based composites.
 A good correlation was found between the properties of the
composites and the fabrics. The fabric with the highest performance
provides the best composite behavior in the high-rate tensile test, in
this study it is carbon fabric. Differences in tensile behavior of the
various composites were recorded indicating differences in the role of
each fabric as reinforcements under high speed loading.
 A multiple cracking behavior was observed for all the fabric-cement
composites except the PE composite with the matrix of plain cement.

33. Acknowledgements
The authors would like to thank Nippon Electric Glass Co., Ltd. USA,
SAERTEX GmbH & Co. KG, Germany and Polysack Ltd., Israel for their
cooperation for providing the fabrics used in this study. The US Federal
aviation administration, FAA, National Science Foundation NSF, and the
BSF (United States Israel Bi-national Science Foundation) program
2006098 acknowledged for the financial support in this research.