Study of Ceramic Tube Fragmentation under Shock Wave Loading Procedia Materials Science 3 ( 2014 ) 592 – 597 Available online at www.sciencedirect.com 2211-8128 © 2014 Published by Elsevier Ltd. Open access under CC BY-NC-ND license. Selection and peer-review under responsibility of the Norwegian University of Science and Technology (NTNU), Department of Structural Engineering doi: 10.1016/j.mspro.2014.06.098 ScienceDirect 20th European Conference on Fracture (ECF20) Study of ceramic tube fragmentation under shock wave loading Irina Bannikova, Sergey Uvarov, Marina Davydova, Oleg Naimark* Institute of Continuous Media Mechanics UB RAS, 1, Ac. Koroleva st., Perm, Russia, 614013 Abstract In this paper we investigate the fragmentation of tubular alumina samples under shock-wave loading. Test of tubular samples was performed on a universal experimental set up of electric explosion conductor Bannikova et al. (2013a). The blast tube destroyed into fragments having as rectangular or irregular geometric shapes. The mass distribution obtained by “method of photography” and “method weighting” are in good agreement Bannikova et al. (2013b). For tubes with wall thickness d = 2.05 mm formed fragments described of by double distribution: small fragments whose size is much smaller than the thickness of the tube - distribution power law, large fragments described as exponential and logarithmic distribution are equally. The inflection point of the distribution curves moves toward smaller scales with increasing energy density. Besides other determined that Weibull function a good description of the distribution of all the fragments by mass. © 2014 The Authors. Published by Elsevier Ltd. Selection and peer-review under responsibility of the Norwegian University of Science and Technology (NTNU), Department of Structural Engineering. Keywords: Fragmentation distribution; electroexplosion; shock wave; alumina ceramic; form factor. 1. Introduction Still during the Second World War by Professor of Physics N. F. Mott were undertaken the first attempts to describe the statistics of fragmentation objects (cylindrical projectiles) subjected to intensive pulsed load. Grady (2006). Since then, the study of the problems of fragmentation continues by our time, therefore significant experimental and theoretical data accumulated. In the world objects are investigated the dimensions of it varies from laboratory specimens crumbling into fragments of weighing less than a gram to the arctic ice fields photographed by satellite or galaxies crumbling meteorites and asteroids. Studied objects of different shapes – spheres, cubes, plates, rings, rods, and of the most diverse materials – copper, lead, glass, concrete, ceramics, rocks, ice, clay, gypsum, soap, wax, potatoes. The objects are investigated under different loading conditions (impact, creep, bending and compression). © 2014 Published by Elsevier Ltd. Open access under CC BY-NC-ND license. Selection and peer-review under responsibility of the Norwegian University of Science and Technology (NTNU), Department of Structural Engineering http://crossmark.crossref.org/dialog/?doi=10.1016/j.mspro.2014.06.098&domain=pdf http://creativecommons.org/licenses/by-nc-nd/3.0/ http://creativecommons.org/licenses/by-nc-nd/3.0/ 593 Irina Bannikova et al. / Procedia Materials Science 3 ( 2014 ) 592 – 597 The main tool of the fragmentation statistical regularities research is determination the fragments distribution by size or masses, in other words, the definition number of fragments N(r, m) with size r or mass m greater than some predetermined. The distribution form depends on many factors: the magnitude of energy expended for the destruction of the sample, the material properties – brittle or ductile, elastic or plastic; dimensionality – 2D for example, plate, rod and 3D – ellipsoid. Depending on the conditions observed distributions of the following types: log-normal, power law (D. L. Turcotte), the distribution of Mott, exponential (D. E. Grady), the Weibull distribution and combination of exponential and power distributions (J. J. Gilvarry). In this paper we investigate the fragmentation of 2D objects - tubular ceramic samples (alumina), in conditions of the shock-wave loading. Nomenclature ms mass of sample d wall thickness of tube h height of tube r1 outer radius of tube r2 inner radius of tube ρo density ceramics WC energy stored in the capacitor battery QW energy expended on evaporation conductor at room temperature w density energy of the explosion N quantity of the fragments of a given mass mf, m mass of the fragment d* character size of the fragment S size of the image area the fragment on photo Sin the inner surface of the tube, of the fragment Sout the outer surface of the tube, of the fragment m* mass of fragment by method photography α form factor 2. Experiment and materials Loading of tubular specimens was carried out on a universal experimental setup of electric explosion conductor Bannikova et al. (2013a). Schematic diagram and appearance of the experimental setup are presented in Bannikova et al. (2013a); Bannikova et al. (2013b). It is a complex that includes cylindrical explosion chamber diameter of 0.24 m and a height of 0.085 m, a high voltage source (HVS, Umax = 5÷15 kV), storage capacitors system (Co = 0.022 ÷ 0.44 mF), discharger and discharge ignition system (manual) on the conductor (copper wire diameter dw = 0.1 mm). Damping material is set on the inner side surface and on the bottom of the cuvette in order to not destroy the wall of the chamber by shock wave. Tubular sample with an axial conductor were placed in the explosion chamber filled with distilled water. The shock wave was initiated in the liquid as a result electric explosion conductor by discharge of the capacitor bank charged from HVS. Water decreased “energy” flying fragments that excluded their further fragmentation by collision fragments with each other and the walls of the chamber. During the tests we performed a series of experiments at different energy of the capacitor bank. Amperage is measured on the conductor with Rogowski coil. The current density was about 1011÷1012 A m-2. Alumina tubes had outer and inner radius r1 = 5.9 mm, r2 = 3.9 mm (the thickness of the tube wall d = 2.05 mm) and a height h was in the range 12.7÷16.7 mm (Fig. 1, (a, b)). The density of the tube material was ρo 2.6 kg m -3. A result conductor explosion the tube fragmented as show in Fig. 1 (c), they had rectangular and irregular geometric forms Bannikova et al. (2013b). Optical microscopy showed that the fragments have a developed cracks structure (Fig. 2). Main cracks are vertical gap cracks formed in result a radial loading of explosive wave, and horizontal cracks are on middle of the 594 Irina Bannikova et al. / Procedia Materials Science 3 ( 2014 ) 592 – 597 sample are crack of reset. Appearance of the last cracks possibly due to that maximum of the velocity profile of the explosive waves necessary on the middle of tube. To collect fragments from liquid used polyethylene film. It was placed on the bottom chamber, under tube. After drying of fragments at room temperature (20оC), they were weighed on an electronic balance HR-202i. Mass of fragments was on average ~ 98% for all test samples. It was decided to consider two methods for determining the mass of fragments, because the part of fragments by mass does not exceed the resolving power of the electronic balance, 0.0001 g. Fig. 1. (a) an external view of the tube; (b) a scheme of the sample; (c) the distribution of fragments in the water after the explosion of the tube. Fig. 2. Cracks on the surface of the fragment, the energy density of the explosion w = 5.9 J g-1: (a) appearance fragment; (b) outer side fragment; (c) the inner surface of the fragment. 2.1. The Method of weighing Fragments were sieved through a laboratory sieves system and were weighed on an electronic balance HR-202i. Distribution of the quantity fragments N in dependence from weight mf exceeding some defined in logarithmic coordinates for all the tested samples is shown in Fig. 3. Since part of the energy stored within the capacitor battery was spent on the processes occurring with the conductor: heating to the melting temperature of the conductor, melting the conductor, heating to the evaporation temperature of the material and evaporation of the conductor, that is some energy Qw, and the mass of the samples ms was different, it was decided to use the term “the density of energy”, w. The density of energy was estimated by Eq. (1): r 2 r 1 h a b c Direction explosive wave b c a 595 Irina Bannikova et al. / Procedia Materials Science 3 ( 2014 ) 592 – 597 S WC m QW w (1) With right from No. of sample shown the corresponding value of the specific energy w (Fig. 3). In Fig. 3 we can distinguish two areas corresponding to the different distribution laws. Since our work loaded ceramic tubes are two- dimensional objects and their fragments have two distributions. For large fragments, whose thickness is commensurate with the thickness of the sample (with the thickness of the tube wall) – one distribution law, for small (volume fragments) – another distribution law. These results agree well with Katsuragi et al. (2005). It has been established that the inflection point of distribution law N(mf) is shifted toward smaller scale with increasing energy density. Fig. 3. The distribution of fragments by weight, logarithmic coordinates. Experimental data processing showed that the distribution of fragments N(mf) weight less than the average weight of mf ≤ mmean and with the size of the fragment much less than the character size of the tube d* mmean, d* = d) is obtained that this fragments distribution described well as an exponential law and the logarithmic law (square deviation R2 = 0.96÷0.98). Eq. (3). Change of coefficients in these distributions depending on the specific energy is shown in Fig. 4, (b, c). The higher energy of the explosion the smaller fragments with a large mass, so the inflection point of the distributions moves toward smaller scales (Fig. 3, black arrow). AmeBN , BmAN )ln( (3) 0 1 2 3 4 5 6 7 8 9 -14 -12 -10 -8 -6 -4 -2 0 ln (N ) ln(m) w incrise w incriseise 596 Irina Bannikova et al. / Procedia Materials Science 3 ( 2014 ) 592 – 597 Fig. 4. (a) change of the coefficients B and A in the distribution power law for small fragments depending on the specific energy; the coefficients A and B: (a) in the exponential distribution; (c) in the logarithmic distribution, Eq. (3). 2.2. The Method of photography In addition to weighing the distribution of fragments by size estimated by digital photography, since the accuracy of 0.0001 g weights not possible to measure the mass of the majority of small fragments. Camera CANON 7D 16.1 Mpix was used to photographing fragments (Fig. 5, (a)). Fragments were placed by hand on black backing. In the right side of image are located those fragments which were individually weighed on an electronic balance, and fragments with height corresponded to tube thickness d* = d (Fig. 5). Fragment image area was calculated using the software package to process digital photographs (Fig. 5, (b)). Mass of large fragments (d* = d) were calculated by formula Eq. (4), where S – the average area of the image area of fragment obtained by processing two photographs, S = (Sout+Sin)/2. Mass small fragments (3D, d* < d) could not be calculated by the method described in Brodskii et al. (2011), since the coefficient “factor of form” α characterizing the geometry of the fragment, was different for tube fragments. “Form factor” for smaller fragments was calculated by the method described below. ρdSm* (4) Mass of large fragments was subtracted from the mass of all fragments of the sample and mass of all small fragments were found, md* 1 gives one exponential distribution, as in the case of shells from the walled cylinders in the works of Grady (2006). Found that the distributions obtained on the “weighting method” and “method of photography” are in good agreement, and subsequently use of them to speed up data processing. Acknowledgements This work was supported by projects of RFBR: No. 14-01-00842, No. 14-01-00370 and No. 14-01-96012, No. 13-08-96025. References Bannikova, I.A., Naimark, O.B., Uvarov S.V., 2013a. Development of methodology for research the relaxation properties in the fluid using the setup of electrical explosion, Extreme states of matter, detonation, shock waves. Proceedings of International conference XX Kharitonov topical scientific readings. Sarov, Russia, p.745–754. Bannikova, I.A., Naimark, O.B., Uvarov S.V., 2013b. Research of the fragmentation of tubular ceramic samples using electrical explosive setup. International Conference “Hierarchically organized systems of animate and inanimate nature”, Proceedings of International Conference. Tomsk, Russia. p. 206–209. (in Russian, CD disk). Brodskii R.Ye., Konevskiy P.V., Safronov R.I.., 2011. Size distribution of sapphire fragments in shock fragmentation. Functional Materials 18, No.2. Grady D., 2006. Fragmentation of rings and shells, Springer – Verlag Berlin Heidelberg, 374. Katsuragi H., Ihara S., Honjo H., 2005. Explosive fragmentation of a thin ceramic tube using pulsed power. Physical Review Letters 95, 095503, p.1–4. I II b 10 mm a