DYMAT 2009 (2009) 503509 EDP Sciences, 2009 DOI: 10.

1051/dymat/2009072

A study of dynamic behavior of the autoclaved aerated concrete

R. Belouettar1 and J.R. Klepaczko2
rie Laboratory of Civil Engineering, Universite de Annaba, BP. 12, Annaba, Alge Laboratory of Physics and Mechanics of Materials, UMR CNRS 75-54, Universite de Metz, Ile du Saulcy, 57045 Metz Cedex, France
2 1

Abstract. This work presents an experimental study on the dynamic behavior of autoclaved aerated concrete. The paper reports a series of tests with a Hopkinson Pressure Bar [1,2] applied to the direct impact technique [5]. Several strain rates were applied varying from quasi-static to impact between 1.4r102 sx1 to 103 sx1. Two states of the aerated concrete were tested, that is dry and saturated with water. In general, an increase of strain rate causes an increase of the critical failure stress for such materials. In the quasi-static range and up to certain critical impact velocity VC the autoclaved aerated concrete shows a positive strain rate sensitivity However, when a critical impact velocity VC is reached a very specic behavior is observed, that is the cell structure is pulverized near the plane of impact. This eect is completely new.

1. INTRODUCTION Cellular concretes are mostly used in the building industry as a material with a low thermal conductivity. However, this material is not so strong and can be easily crushed. It appears that during an impact loading it can be pulverised without diculty. Such behavior is important, for example, in case of earthquakes or short-time loadings caused by impact. Because the cell structure of such material is very brittle it leads to instantaneous crushing near the loading zone. In order to nd the rate sensitivity and an impact resistance to crushing a study was carried out in quasi static compression tests with a servo-hydraulic fast testing machine. In order to cover the widest possible range of strain rates, the range of quasi-static strain rates was applied varying from 10x4 sx1 to 10 sx1 and the high strain range was applied up to 103 sx1. Thus in the quasi-static range of strain rates the sensitivity could be found. The next step was to test such materials using specially designed experimental set-up designed and constructed at LPMM. 2. DIRECT IMPACT TESTS The rst mechanical tests at high strain rate were carried out by B.Hopkinson [1] based on the concept of ballistic pendulum and elastic wave propagation in a bar, called since Hopkinson bar. It was only since 1948 that the Hopkinson bar has been instrumented by Davies [2]. In both cases the bars served as specic gages to determined short pressures. One year later, Kolsky developed a technique called the Split Hokinson Pressure Bar or Kolsky apparatus [3]. There were two instrumented Davies bars with a small cylindrical specimens inserted in between. The Kolsky apparatus was modied later on by Lindholm [4]. In general, its principle is based on the theory of the elastic wave propagation in bars and makes it possible to obtain a large range of high strain rates varying from 5.102 sx1 to 5.103 sx1 [3, 4, 7]. In this study, however, only one Hopkinson bar instrumented with SR gages was applied in combination with non-contact optical displacement gage [5]. The principle of experimental set-up is shown in Figure 1.

504

DYMAT International Conferences

Hopkinson bar Ve Vp

X Gauge Specimen Projectile

Figure 1. Principle of the Hopkinson bar in direct impact arrangement.

3. DETERMINATION OF THE OUTPUT QUANTITIES In order to obtain the critical compression stress at failure in the setup shown in Figure 1. A simple theory of wave propagation in bars was applied. During contact between the projectile and the specimen a damage occurs on a part of the thickness of the specimen. The critical force of failure is transmitted into the Hopkinson bar and it can be measured as a function of time by the SR gage An equilibrium of forces is assumed between the specimen of autoclaved aerated concrete and projectile: Fep Fp , where Fep is the force in the specimen Fp is the force of the projectile, thus Aep sep Ap s p , where Aep is the cross section surface of specimen and Ap is the cross section surface of the projectile. Three sources of light, ber optics and three photodiodes, Figure 2, is split into two mass velocities in the specimen and projectile, V0 Vep Vp . This assumption is valid only at the rst instant of the contact specimen/projectile. The well known relation between stress and the mass velocity leads to the following relations: s ep rep C0ep Vep and s p rp C0p Vp , where C0s and C0p are respectively the elastic wave speeds in the specimen and projectile and rep and rp are the densities of the specimen and projectile. The relation between stresses and the impact velocity is given by sep sp V0 1 rep C0ep rp C0p The assumption of the force equilibrium during rst instant of crushing leads to the following relation " # Aep rep C0ep rB C0B eT t V0 1 2 rep C0ep Ap rp C0p where s B rB C0B eT t and rB , C0B and eT(t) are respectively the density, the elastic wave speed and the transmitted wave recorded by SR gages as a function of time, all in the Hopkinson bar. This simplied formula allows to estimate the critical failure stress as well as the critical impact velocity V0C for various quasi-brittle materials loaded at various impact velocities when the critical conditions are detected in the eT (tC) signal, where tC is the instant of failure. 4. DESCRIPTION OF THE SET-UP FOR DIRECT IMPACT LOADING The arrangement applied for the direct impact tests is shown in Figure 2. It makes possible to determine not only the critical impact velocity of crushing but also, due to application of the optical displacement gage, to determine the whole stress versus strain relation at high strain rates. This set-up has been developed at the Laboratory of Physics and Mechanics of Materials (LPMM) [5]

DYMAT 2009

505

Figure 2. Arrangement of Hopkinson bar with direct impact [5].

0.12

0.1

0.08

Force (volt)

0.06

0.04

0.02

0 B CK -0 . 0 2 0 .0 012

0.0013

0.0014

0.0015

0 .0 016

0.0017

0.0018

0.0019

0.002

Time (s)

Figure 3. Digital record representing a typical test of autoclaved aerated concrete of BCK type.

A gas gun launches a projectile at a desired velocity V0. The specimen is supported by the Hopkinson tube. The tube is closed in front by a thin steel plate to support a specimen. Application of the Hopkinson tube instead of a bar substantially increased sensitivity of measurement of the transmitted waves. This improvement was caused by a low velocity of crushing found for the aerated concrete being tested. The displacement of the projectile during the contact with specimen is measured by the rst channel of the optical displacement gage, the second channel can measure displacement of the tube front. The complete set of measurements consists of the impact velocity V0 (three photodiodes and two time counters) and the displacement of projectile in contact with specimen UA(t), the displacement of the tube front UB(t) and the transmitted elastic wave in the Hopkinson tube eT(t). All electric signals after conditioning are stored in a digital oscilloscope with maximum sampling rate 1 GHz. Later the digital records are stored in a PC for further analyses. 4.1 Analysis of stored signals The impact velocity V0 is directly related to the pressure in the chamber of the gas launcher, and _ the nominal strain rate is dened as e V0 =lS , where lS is the specimen length. Because both the

506

DYMAT International Conferences

displacement of the contact projectile/specimen UA(t) and the transmitted wave eT(t) are recorded, it is possible to determine completely the behavior of a specimen, [5]. The net displacement of the specimen is calculated by the dierence between the displacement UA(t) and displacement of the tube UB t , thus the nominal strain can be also found Uep t UA t UB t and en 1 UA t UB t ls 3

The displacement of the tube at the time of loading is related directly to the transmitted wave by the following relation UB C0
Z
t 0

eT jdj

whereC0 is the elastic wave speed. The stress measurement is possible by the means the SR gauges via the transmitted wave eT(t). The stress as well as the force can be given according to the signal of the transmitted wave [5]. sT t rB C2 eT t 0 5 where EB rB C2 is the modulus of elasticity of the tube. In the case of compression, the force 0 which acts on the surface of contact of the bar and of the specimen is assumed to be the same. Thus the mean force loading the specimen, and consequently the nominal stress s n, can be written in the following way: Fep t FB t and Fep t rB C2 AB eT t 0 Fep t Aep 6

s n t

where AB is the cross section of the Hopkinson tube. 5. RESULTS OF COMPRESSION TESTS The tests were carried out on two types of aerated concrete. The specimens were tested in two dierent states: the dry state and the wet state (saturated with water). For the reason of dispersion of the results, mainly stresses, ve tests were performed at each impact velocity V0. The assumed impact velocities of the projectiles are: 5 m/s, 10 m/s, 20 m/s, 30 m/s and 35 m/s. The cylindrical projectiles of diameter 49.8 mm were machined out of aluminum alloy 2017A, of density and modulus of elasticity respectively equal to 2.79 g/cm3 and 74 GPa. The specimens were in the form of cubes with the side dimension 40.0 mm. Thus the ve programmed velocities for the dynamic tests yield strain rates from 1.43r102 sx1 to 8.75r103 sx1 After analysis of all records the values of the critical stress of crushing were obtained. The results are presented in the form of sC versus log(_ ) in Figures 4a and 4b for dry and wet BCK e aerated concrete, Figures 5a and 5b show similar results for BCM aerated concrete. The wet materials show higher critical stresses as compared to dry materials. The sensitivity is observed between speeds 10 sx1 and 1.43r102 sx1. At lower strain rates the rate sensitivity of the critical stress is positive, but at certain strain rate, typically higher than 102 sx1, one notices a reduction in the critical stresses when the strain rate is increased beyond of this critical value. In some cases a minimum of the critical stress is obtained. Then an increase in the critical stresses is observed anew. The mean values of the critical stresses are denoted by dark triangles. The results for dry

DYMAT 2009
Critical stress (MPa) Critical stress (MPa)

507

6 5

Cri ti ca l s tr ess (MPa )

4 3 2 1 0 -0.5

0.5

1.5

2.5

3.5

Log of strain rate (1/s)

Figure 4a. Critical stress versus logarithm of strain rate for dry autoclaved aerated concrete BCK.

Critical stress (MPa) Critical stress (MPa)

Cri ti ca l s tr ess (MPa )

5 4 3 2 1

-0.5

0.5

1.5

2.5

3.5

Log of strain rate (1/s)

Figure 4b. Critical stress versus logarithm of the strain rate for wet autoclaved aerated concrete BCK.

and wet autoclaved aerated concrete BCK are shown in Figure 4, whereas the results for dry and wet BCM aerated concrete are shown in Figure 5. 6. CONCLUSIONS The scheme of the direct impact experiment seems to be very useful in materials testing at high strain rates [5, 6]. Application of the non-contact optical displacement gages enabled to measure the real mean strain in a specimen leading to the complete measurements of stresses and strains. Elimination of time leads to estimation of the stress-strain characteristics at relatively high strain rates.

508

DYMAT International Conferences

Critical stress (MPa) Critical stress (MPa)

Cri ti cal st ress (MPa)

5 4 3 2 1 0 -0.5 0 0.5 1 1.5 2 2.5 3 3.5

Log of strain rate (1/s)

Figure 5a. Critical stress versus logarithm of strain rate for dry autoclaved aerated concrete BCM.

Critical stress (MPa) Critical stress (MPa)

Cri ti cal st ress (MPa)

5 4 3 2 1 0 -0.5 0 0.5 1 1.5 2 2.5 3 3.5

Log of strain rate (1/s)

Figure 5b. Critical stress versus logarithm of strain rate for wet autoclaved aerated concrete BCM.

It is concluded after the results obtained during the dynamic tests of the autoclaved aerated concrete BCK and BCM that the autoclaved aerated concrete has a positive rate sensitivity at low and medium strain rates. But at the high strain rates a critical impact velocity V0C occurs, this velocity is around 5.0 m/s the two materials tested. The BCM reaches a maximum value of the critical stress of 5.3 MPa in dry state and 4.5 MPa in the wet state (strain rate 1.43r102 s1). Whereas for the same strain rate the values of the critical stress for the BCK is only 3.75 MPa (in dry state) and 4.00 MPa (in wet state). Both materials showed a higher rate sensitivity in the state saturated with water. The eect of saturation of pores with water intervenes in this behavior. At strain rates higher than y 1.4r102 sx1 ( y 5.0 m/s) the stress of failure decreases. This decrease of the failure stress is due mainly due pulverization of the specimen material close to the impact surface. When the strain rate increases beyond the critical value the material undergoes further damage on a certain thickness and the waves can be transmitted again. Anyway, such behavior proves that the transmission of the waves is disturbed around the critical impact velocities. This eect should be tested with more details in the future.

DYMAT 2009 References

509

[1] Hopkinson B., A method of measuring the pressure produced in the detonation of explosives or by impact of bullets, Phil. Trans. Royal Soc. London, A 213. (1913), pp. 437. [2] Davies R.M., A critical study of the Hopkinson pressure bar, Phil. Trans. Royal Soc. London, A240 (1948), pp. 375. [3] Kolsky H., An investigation of the mechanical properties of materials at very high rates of loading Proc. Phys. Soc., London, B62, (1949), pp. 672. [4] Lindholm U. S., Some experiments with the Split Hopkinson Pressure Bar J, Mech. Phys. Solids, Vol. 12, (1964), pp. 317. [5] Klepaczko J.R., Advanced experimental techniques in materials testing, Ch.6 in New Experimental Methods in Material Dynamics and Impact, eds. W.K.Nowacki and J.R.Klepaczko, IFTR, Warsaw, (2002), pp. 223. [6] Belouettar R., Klepaczko J. R., Abadlia M. T., Etude du comportement mecanique du beton cellulaire autoclave : inuence de la vitesse de deformation , Revue Francaise de Genie Civil, vol 5 nx 5 (2001), pp.667. [7] Klepaczko J. R., Quasi-static and Dynamic Compression Behavior of Coal, Report nx 1, Department of Mech. Eng. The university of Manitoba, Canada, (1982)