Vous êtes sur la page 1sur 11

International Journal of Impact Engineering 80 (2015) 13e23

Contents lists available at ScienceDirect

International Journal of Impact Engineering


journal homepage: www.elsevier.com/locate/ijimpeng

Effect of heat treatment on ballistic performance of an armour steel


against long rod projectile
P. Ponguru Senthil*, B. Bhav Singh, K. Siva Kumar, A.K. Gogia
Armour Design and Development Division, Defence Metallurgical Research Laboratory, Kanchanabgh, Hyderabad 500058, India

a r t i c l e i n f o a b s t r a c t

Article history: Ballistic performance of an armour steel at different tempering conditions against a tungsten heavy alloy
Received 29 July 2014 long rod kinetic energy projectile has been investigated. The ballistic performance was found to increase
Received in revised form with decreasing tempering temperature but the performance increments were not proportional to the
13 November 2014
strength increase caused by the tempering. Both target and projectile underwent fragmentation at the
Accepted 12 December 2014
penetration interface in all the three targets. Apart from the strength target material deformation
Available online 19 December 2014
behaviour also showed considerable influence on ballistic performance. The ballistic performance ap-
pears to increase with decreasing toughness for a given strength.
Keywords:
Long rod projectile
© 2014 Elsevier Ltd. All rights reserved.
Penetration velocity
Ballistic performance
Microstructure
Deformation

1. Introduction
P ¼ LU=ðV  UÞ (1)
Penetration of a long rod kinetic energy projectiles in a target
have been found to happen in three phases, transient, steady state where L is the length of the projectile, V is impact velocity of the
primary penetration and a non steady state secondary penetration projectile and U is penetration velocity. The penetration efficiency
[1]. For projectiles with large L/D ratio, the penetration will pre- P/L can be given by,
dominantly happen by steady state penetration and the contribu-
tion from transient and secondary penetration will be very less
[2e4]. During steady state penetration, projectile will be P=L ¼ U=ðV  UÞ (2)
consumed simultaneously with target material by erosion, leading The physical meaning of penetration efficiency during steady
to a decrease in the length of the projectile with increasing pene- state penetration is relative consumption rate of the target and
tration [5,6]. Steady state penetration has been observed in various projectile material. The efficiency of penetration will be determined
experimental studies for different target - projectile combinations by the rate at which the material is removed from the crater (U) and
[7e14]. Penetration velocity and the projectile consumption rate the rate at which the projectile material is utilized (VeU). In fact
will be constant during steady state penetration. Towards the end projectile consumption rate will also be determined by the con-
of the penetration for a small length, the projectile erosion will stop sumption rate of the target material because the impact velocity V
and the non steady state penetration will occur by deceleration of is constant during penetration. The effect of change in penetration
the projectile until the projectile lost its kinetic energy. For large L/ velocity on ballistic performance will be twofold. It can be seen
D ratio projectiles, the total penetration depth can be approximated from equation (1) that the change will directly affect the depth of
by assuming complete steady state penetration. By balancing the penetration because increasing penetration velocity will increase
projectile and target consumption rate the penetration depth (P) the depth of penetration proportionally. The change in penetration
can be given by Ref. [15] velocity will also affect the consumption rate of the projectile there
by the duration of penetration. For instance, the increase in pene-
tration velocity will result in decreasing the consumption rate of
* Corresponding author. Tel.: þ91 40 2458 6332 (office); fax: þ91 40 2434 2252. the projectile and thereby increasing the duration of penetration
E-mail address: pgsenthil@dmrl.drdo.in (P. Ponguru Senthil). which will result in increased depth of penetration.

http://dx.doi.org/10.1016/j.ijimpeng.2014.12.003
0734-743X/© 2014 Elsevier Ltd. All rights reserved.
14 P. Ponguru Senthil et al. / International Journal of Impact Engineering 80 (2015) 13e23

Table 1 density and hardness were 17.5 g/cc and 515 HV respectively. The
Chemical composition. testing was carried out at 0 obliquity. The target plates were
Material Chemical composition stacked in sufficient numbers such that the thickness is sufficiently
Steel 0.28e0.33% C, 0.4e0.6% Mn, 0.1e0.25% Si, 1.3e1.5% Cr,
larger than the depth of penetration in order to give semi infinite
1.5e1.7% Ni, 0.45e0.5% Mo, 0.08e0.12% V, balance Fe thickness. The target plate dimensions with respect to projectile
diameter are 1.7D  17D  40D. Five tests were carried out on
650  C and 500  C tempered targets and three tests were carried
Target and projectile material behaviour plays a major role in out on 450  C tempered target. After ballistic penetration, target
determining the ballistic performance. Apart from strength, mate- and projectile fragments of each target material were collected and
rial properties like strain hardening, thermal softening, ability to studied using optical and scanning electron microscopy. The pro-
accommodate large deformation considerably influence the pene- jectile fragments collected from the craters are shown in Fig. 1. The
tration efficiency [16e18]. In the present study the effect of target cross section of the individual target material craters was also
material properties on the ballistic performance against tungsten examined to understand the deformation behaviour.
heavy alloy (WHA) long rod projectile has been studied. Three
different mechanical properties were obtained by varying heat 3. Results and discussion
treatment cycle in an armour steel. All three target materials were
tested against a fixed L/D projectile at a constant impact velocity. 3.1. Ballistic performance
This enables us to reason the observed difference in performance
with deformation behaviour of the target material. The mechanical properties of the target materials along with
normalized depth of penetration (P/L) are given in Table 2. When
2. Materials and experimental procedure decreasing the tempering temperature yield strength and ultimate
tensile strength of the steel increases whereas the toughness de-
The steel used in the present study is a medium carbon low alloy creases. The tensile strength of steel plates tempered at 450  C
steel. The chemical composition of the steel is shown in Table 1. The (SP450) and 500  C (SP500) is 50% higher than that of the plate
heat treatment consists of quenching and tempering. Hot rolled tempered at 650  C (SP650). The mechanical properties are similar
steel plates were austenized at 910  C and oil quenched. Tempering in case of SP500 and SP450 except percentage elongation and
was done at three different temperatures 650  C, 500  C and 450  C toughness which shows a significant drop in SP450.
in order to attain the expected variation in mechanical properties. The ballistic performance is shown in terms of normalized depth
The samples were air cooled after tempering. Yield strength, Ulti- of penetration in Table 2. SP500 shows only 3% improvement over
mate tensile strength, elongation, hardness and impact toughness SP650 even though the difference in tensile strength is 50%. But
were evaluated for all the three target materials. Ballistic evaluation SP450 with almost similar strength and reduced toughness to that
was carried out using a tungsten heavy alloy projectile with an L/D of SP500 shows a 10% improvement in performance over SP650.
ratio of 16 at an impact velocity of 1600(1588e1620) m/s. Projectile The ballistic performance shows an increasing trend with increase

Fig. 1. The fragments collected from the target crater (a) SP650, (b) SP500, (c) SP450.
P. Ponguru Senthil et al. / International Journal of Impact Engineering 80 (2015) 13e23 15

Table 2
Mechanical properties.

Material ID Tempering Yield Ultimate tensile Elongation % Toughness CVN (J) Hardness VHN P/L (SD)
temperature ( C) strength (MPa) strength (MPa)

SP650 650  C 875 975 20 100 300e320 0.966 (0.025)


SP500 500  C 1351 1455 12 35 440e460 0.936(0.011)
SP450 450  C 1379 1504 7.5 24 440e460 0.872 (0.017)

Table 3 strength. But SP450 which has a similar improvement in tensile


Impact velocity and penetration velocity. strength shows a significant drop in penetration velocity. So it is
Material ID Impact Penetration Penetration apparent that there is also a role of material properties other than
velocity (SD) (m/s) velocity (SD) (m/s) velocity (SD) (mm/ms) the tensile strength in determining the penetration velocity.
SP650 786 (13) 0.786 (0.013)
SP500 1606 (09) 773 (05) 0.773 (0.005)
SP450 745 (07) 0.745 (0.007) 3.2. Role of failure mechanism

The representative microstructures of the cross sections of


in tensile strength but the increase is not proportional. Penetration
fragments are shown in Figs. 2e4 for SP650, SP500 and SP450
velocity for different target materials was calculated from
respectively. The microstructures are shown in unetched condition
normalized penetration using the steady state relationship given by
because it was possible to differentiate the tungsten heavy alloy
equation (2). The penetration velocities are given in Table 3. It
(projectile) and steel (target) fragments easily. In all the three tar-
should also be noted that the penetration velocity was calculated by
gets the fragments consist of a mixture of fine fragments of steel
assuming a complete steady state penetration. But in actual pene-
and tungsten heavy alloy particles as shown in Figs. 2e4. The
tration there is a small portion of the projectile at the tail end which
tungsten heavy alloy fragments were heavily deformed and were
undergoes non steady state penetration. So assuming complete
welded to the steel fragments which are shown in Fig. 3(d) and
steady state leads to over estimation in penetration velocity. Also
4(b). At the projectileetarget interface, both target as well as pro-
during penetration the tail end velocity and penetration velocity
jectile material undergo deformation. The target material flows
decreases over time because of deceleration. So the calculated
outward from the crater. Similarly, projectile material also deforms
penetration velocities actually represent the average penetration
and flows outward from the crater. Due to flow of the projectile and
velocity. However for the purpose of comparison of performance,
target material at the penetration interface the fragments are
the relative differences between penetration velocities can be used.
generated and form a mixture when flowing outward from the
The drop in penetration velocity is relatively very small in SP500
crater. This turbulent flow of projectile and target material causes
target compared to SP650 though there is a 50% improvement in
the temperature to rise in the outer surface of the fragments due to

Fig. 2. (a) & (b) Fragment cross section of SP650 showing mixture of fine target and projectile fragments, (c) & (d) Magnified images of the fragments.
16 P. Ponguru Senthil et al. / International Journal of Impact Engineering 80 (2015) 13e23

Fig. 3. (a) Fragment cross section of SP500 showing mixture of fine target and projectile fragments, (b), (c) & (d) Magnified images of the fragments.

the friction. Because of the heat generated and deformation, the penetration forming a mushroom head (consumption of projectile).
fragments get friction welded on to each other forming lumps of So the ability of the target material to gain momentum determines
the ejected material. This macro mechanism of failure appears to be the penetration efficiency. The slower the target material is in
similar in all three target materials. gaining momentum the more will be the remaining momentum
The typical size of the constituent target and projectile particles carried back by the projectile.
are smaller in SP650 and SP500 lumps compared to SP450 lumps. The target material's flow properties determine the rate at
Microstructure of SP450 fragments contains large fraction of larger which it can absorb momentum. The momentum gain by the
particles and a small fraction of finer particles which are embedded target occurs by the flow (target material gains velocity) of ma-
in between large particles. Size of the particles in SP650 and SP500 terial outwards from the target. For a given mushroom head of
fragments are less than 2 mm whereas the size of the largest par- the projectile, there is a minimum strain required to make the
ticle in SP450 fragment is more than 5 mm. As it can be seen from target material to flow like a fluid around the mushroom head
Fig. 4(a) that only steel particles are larger in SP450 and projectile and to leave the interface. Till the target material is able to
particles are finer. This difference in particle size suggests that there accommodate this strain, the rate of momentum transfer does not
is a difference in micro mechanism of target failure which is likely get affected. If the target material is ductile enough, it will be
cause for the increased ballistic performance of SP450. replaced away from the penetration interface by plastic flow
since it can accommodate large strain. But if the target material is
3.3. Momentum balance at the projectileetarget interface not ductile, it can not accommodate required strain. It resists
deformation and shape change required to make it flow out. Due
During steady state penetration, the stress at the penetration to the geometrical limitations it becomes difficult for the material
interface attains a constant value [1]. With increasing impact ve- to leave the interface without flow. Materials with limited
locity this stress at the targeteprojectile interface increases. ductility undertake only limited deformation and undergo frac-
Generally this stress is several times higher than the target and ture at the projectileetarget interface. In addition to deformation,
projectile strength. For an impact velocity of 1500 m/s numerical need of fracture and fragmentation increases the time required
simulations predict a stress of more than 7 GPa for a zero strength for momentum transfer and reduces the rate of momentum
projectile [4]. Because of the very high stress applied on the target transfer. So with increasing brittleness of the target material the
material compared to its strength the target material at the inter- rate of momentum transfer to target would decrease. It leads the
face does not resist the failure. projectile material to consume large momentum leading to
There is a local momentum balance maintained at the pene- increased consumption rate of the projectile. Penetration velocity
tration interface during steady state penetration. The incoming decreases because of the reduced rate of momentum transfer and
projectile feeds the interface with a constant rate of momentum. causes deviation from the hydrodynamic behaviour. With
Only a part of the projectile momentum is transferred to the target increasing projectile hardness and decreasing strain to failure the
material and the remaining momentum is consumed by the pro- penetration efficiency was found to increase because of its
jectile itself. The unconsumed momentum by target is carried by inability to support a large hydrodynamic (mushroom) head [16].
the projectile material by flowing in the direction opposite to Numerical simulations carried out in semi-infinite target also
P. Ponguru Senthil et al. / International Journal of Impact Engineering 80 (2015) 13e23 17

Fig. 4. (a) Fragment cross section of SP450 showing mixture of target and projectile fragments, (b), (c) & (d) Magnified images of fine fragments.

suggest that with decreasing strain to failure of the projectile the materials resistance to the bulk plastic flow. Appearance of craters
penetration increases [17,18]. also shows significant difference in SP450 compared to SP650 and
It is a well known fact that decreasing tempering temperature SP500 as shown in Fig. 6. The crater shows smooth cylindrical hole
decreases the ductility of the steel. The deformations of the mate- in SP650 and SP500 whereas SP450 an irregular crater. So the
rial around the crater in all three targets are shown in Fig. 5. The possible explanation for the observed difference in the penetration
deformation in SP450 shows more shear localized regions and well velocity can be that SP650 and SP500 were able to accommodate
developed white etching adiabatic shear bands whereas in other the critical deformation required to make the material flow out of
two targets the deformation is more homogeneous. The develop- the crater. SP450, because of its reduced ductility could not
ment of adiabatic shear localization in SP450 is an indication of accommodate the required critical deformation leading to more
18 P. Ponguru Senthil et al. / International Journal of Impact Engineering 80 (2015) 13e23

Fig. 5. Target material deformation adjacent to crater (a) SP650, (b) SP500, (c) SP450.

Fig. 6. Impact craters of targets (a) SP650, (b) SP500 and (c) SP450.
P. Ponguru Senthil et al. / International Journal of Impact Engineering 80 (2015) 13e23 19

Fig. 7. (a) void nucleation in WHA fragment, (b) & (c) Crack propagation through WHA matrix.

deviation from the hydrodynamic behaviour. Also it has resulted in will be higher than the time required for the failure. So, in steady
the formation of large fragments in SP450 target as shown in Fig. 4. state penetration the unit failure mechanism will be defined as a
sequence of micro events that are taking place from the start of the
loading to the removal of material from the projectileetarget
3.4. Role of time scales of failure mechanism interface. Difference in length and time scales of failure mechanism
will affect the time required for the material to leave the penetra-
Another way of looking at this problem is to look at the time tion interface. The time scale of the unit failure and the thickness of
scales of micro mechanisms that are involved in the penetration material removed by the unit failure can be used to compare per-
process of the target materials. Since the penetration velocity is formance of different materials. The time (td) taken for a unit
constant, it is plausible to assume that steady state penetration is a thickness of the material removal from the penetration interface
result of a set of particular deformation and failure events can be determined by the failure mechanism which can be given by
happening repeatedly throughout the penetration. The unit failure
event will be consisting of different deformation and fracture
td ¼ tmech =d (4)
mechanisms occurring in sequence. Depending on the magnitude
and time scale of the loading, the operating mechanism can range where, tmech is the time taken for the failure mechanism and d is
from phase transformation, dislocation nucleation to crack nucle- the thickness of the material removed. The material which is having
ation, fragment flow etc. The different deformation mechanisms higher td ratio will have better ballistic performance. Materials with
operative under different time scales of impulsive loading condi- slowly progressing failure mechanisms will delay the material
tions have been summarized [19,20]. Depending on the material removal and thereby reduce the penetration velocity. In SP500 and
behaviour length scales and time scales of the failure process will SP650 targets the fragments are approximately made of less than
be different in different target materials. This will have direct in- 2 mm fragments. Hence the unit failure mechanism can be
fluence on the penetration velocity. The dimensionless constant F considered to be happening in 2 mm length scales whereas in
introduced by N.K. Bourne can be used to analyse the steady state SP450 the fragment size is approximately 5 mm making the unit
penetration [19]. failure mechanism to be of 5 mm in length scale. So the difference
. in length scales suggests that there must be a difference in the
F ¼ tmech timpulse (3) failure mechanisms happening between SP650, SP500 and SP450
targets. The difference in the time scales of this unit process reflects
where tmech is the time required to complete the unit failure in the penetration velocity difference observed. In situ observation
mechanism and timpulse is the duration of loading. During steady of failure processes are required to get more detailed picture of the
state penetration, F << 1 which means that the duration of loading failure mechanisms.
20 P. Ponguru Senthil et al. / International Journal of Impact Engineering 80 (2015) 13e23

Fig. 8. (a) & (b) Deformed WHA fragments, (c) & (d) intensity peaks obtained by EDX analysis in the marked regions of (a) & (b) respectively.

3.5. Metallurgical changes nucleation. In addition the deformation causes the rise in temper-
ature. Because of the lower melting point of the NieFeeCo matrix
3.5.1. WHA projectile [22], the matrix becomes softened and undertakes more defor-
During penetration tungsten heavy alloy fragments undergo mation compared to tungsten grains. With further deformation, the
bulk severe plastic deformation. The tungsten grains are elongated temperature increases further and void nucleates at random sites
in the direction of deformation. No shear localization is observed in inside the matrix. Cracks propagation happens by the intercon-
the projectile fragments of all three tests. Instead the failure is nection of the voids and leads to fracture as shown in Fig. 7.
found to happen by void nucleation and interconnection. Fig. 7 EDX analysis was carried out at the interior and at the outer
shows void nucleation and crack propagation inside the WHA edges of the tungsten heavy alloy fragments at the marked loca-
fragment. Voids are nucleated in the NieFeeCo matrix and the tions in Fig. 8. The results of EDX analysis are shown in Table 4. It is
growing voids connect with each other and lead to crack propa- found from Table 4 that the matrix inside the deformed fragment is
gation. The elongated tungsten grains stay unaffected during crack made up of Ni, W, Fe and Co. The Nickel content is high in the
propagation. This observation is in contrast to the spall failure matrix as expected for a WHA. However, the composition of the
observed during shock loading which happens by crack propaga- matrix at the outer edges of the fragment has been altered and the
tion through tungsten grains [21]. The above difference can be matrix has high iron content as shown in Fig. 9 and Table 5. The
explained in the following manner. During shock loading the stress presence of high iron can be attributed to the interaction between
suddenly rises to values higher than the strength of both the the WHA and steel fragments. During the penetration process, the
tungsten grain and the matrix causing the spall fracture to propa- failed material which is moving out of the crater experiences tur-
gate through the tungsten grains. During long rod penetration, the bulent motion [23]. This turbulent motion causes friction between
material experiences bulk deformation when it enters the the steel and WHA fragments. This friction would have caused
deforming zone at the projectileetarget interface. Softer phase temperature rise close to the melting of the WHA matrix and steel
undergoes much more deformation and thus causes void fragments leading to enrichment of iron in the periphery of the
WHA fragments. Evidence of melting is also present at the inter-
action zone between WHA and steel fragments as shown in
Table 4 Fig. 10(a) and (b). BSE image of this molten region shows fine
Composition of the matrix (wt %) in deformed WHA fragment (Fig. 8) obtained fragments of tungsten present in the matrix as shown in Fig. 10(c).
through EDX analysis. The EDX analysis carried out on the tungsten grains both at inside
Spot W Fe Ni Co the fragment and at the outer periphery shows presence of only
tungsten. This suggests that the temperature rise was not high
Fig. 8(a) 26.76 12.16 54.94 6.13
Fig. 8(b) 33.97 12.34 48.67 5.02 enough to melt tungsten grains.
P. Ponguru Senthil et al. / International Journal of Impact Engineering 80 (2015) 13e23 21

Fig. 9. (a) & (b): edges of the deformed WHA fragments showing disintegrated W grains, (c) & (d): magnified images of (a) & (b) respectively. (e) & (f) intensity peaks obtained by
EDX analysis in the marked regions of (c) & (d) respectively.

3.5.2. Microstructure of steel fragments


The steel fragments were observed to be heavily deformed. In all
the three target fragments, similar type of phase transformations
was observed. Fig. 11 shows optical microstructure of a steel frag-
ment after etching. The edges of the fragments exhibit grain Table 5
Composition of the matrix (wt%) at the edges of deformed WHA fragment (Fig. 9)
refinement. This could be due to severe plastic deformation and
obtained through EDX analysis.
consequent temperature rise. White grains in Fig. 11(a)e(d) shows
formation of fresh martensite in the prior austenite grain bound- Spot W Fe Ni Co

aries. The temperature rise during deformation appears to be high Fig. 9(c) 11.51 59.94 16.63 10.16
enough to austenitize the martensite. The austenite grains have Fig. 9(d) 32.93 50.99 15.10 e
22 P. Ponguru Senthil et al. / International Journal of Impact Engineering 80 (2015) 13e23

Fig. 10. (a) & (b) molten regions observed in the interaction region between steel and WHA fragments, (c) BSE image showing tungsten rich regions in the molten WHA matrix.

nucleated at the prior austenite grain boundaries. Since the dura- deformation and high temperature at the grain boundaries of the
tion at high temperature is very short (in micro seconds), before fragments, the number of nucleation sites is large leading to finer
completion of austenization, the fragments undergo sudden cool- and higher fraction of fresh martensite at the edges of the frag-
ing resulting in formation of martensite. Because of the heavy ments as shown in Fig. 11(b). However towards interior, the fresh

Fig. 11. (a) Microstructure of a steel fragment, (b) Grain refinement at the edges of the fragment, (c) & (d) formation of fresh martensite at the prior austenitic grain boundaries.
P. Ponguru Senthil et al. / International Journal of Impact Engineering 80 (2015) 13e23 23

martensite content decreases and it is present only as a discon- [7] Behner Th, Anderson Jr CE, Orphal DL, Hohler V, Moll M, Templeton DW.
Penetration and failure of lead and borosilicate glass against rod impact. Int J
tinuous network at the prior austenitic grain boundaries of
Impact Eng 2008;35:447e56.
tempered martensite as shown in Fig. 11(c) and (d). [8] Anderson Jr CE, Walker JD, Bless SJ, Sharron TR. On the velocity dependence of
the L/D effect for long e rod penetrators. Int J Impact Eng 1995;17:13e24.
4. Conclusions [9] Anderson Jr CE, Walker JD, Bless SJ, Partom Y. On the L/D effect for long-rod
penetrators. Int J Impact Eng 1996;18:247e64.
[10] Behner T, Orphal DL, Hohler V, Anderson Jr CE, Mason RL, Templeton DW.
The ballistic performance increases with decreasing tempering Hypervelocity penetration of gold rods into SiCeN for impact velocities from 2
temperature in the steel though the increment is not proportional to 6.2 km/s. Int J Impact Eng 2006;33:68e79.
[11] Orphal DL, Franzen RR, Piekutowski AJ, Forrestal MJ. Penetration of confined
to the increase in strength due to lower temperature of tempering. aluminum nitride targets by tungsten long rods at 1.5e4.5 km/s. Int J Impact
The target deformation behaviour has considerable influence on Eng 1996;18:355e68.
the penetration velocity. With increasing resistance to plastic flow [12] Orphal DL, Franzen RR. Penetration of confined silicon carbide targets by
tungsten long rods at impact velocities from 1.5 to 4.6 km/s. Int J Impact Eng
penetration velocity decreases. Projectile as well as target undergo 1997;19:1e13.
failure at the interface and are removed in the direction opposite to [13] Orphal DL, Franzen RR, Charters AC, Menna TL, Piekutowski AJ. Penetration of
the penetration. The fine fragments of target and projectile gets confined boron carbide targets by tungsten long rods at impact velocities from
1.5 to 5.0 km/s. Int J Impact Eng 1997;19:15e29.
friction welded due to the turbulence and temperature rise. [14] Subramanian R, Bless SJ. Penetration of semi-infinite AD995 alumina targets
by tungsten long rod penetrators from 1.5 to 3.5 km/s. Int J Impact Eng
Acknowledgements 1995;17:807e16.
[15] Birkhoff G. Explosives with lined cavities. J Appl Phys 1948;19:563e82.
[16] Anderson Jr CE, Hohler V, Walker JD, Stilp AJ. The influence of projectile
The authors are grateful to Defence Research & Development hardness on ballistic performance. Int J Impact Eng 1999;22:619e32.
Organization (DRDO) for financial support to carry out this work at [17] Rosenberg Z, Dekel E. On the relation between deformation modes and the
Defence Metallurgical Research Laboratory, Hyderabad. The au- penetration capability of long-rods. In: Int workshop on new models and new
codes on shock wave processes in condensed media. UK: St. Catherines Col-
thors wish to express their gratitude to the Director, DMRL for lege; 15e19 Sept. 1997. p. 334e43.
granting permission to publish this paper. [18] partom Y, Yaziz D. Penetration of L/D ¼ 10 and 20 tungsten alloy projectiles
into RHA targets, shock waves in condensed materials. Colorado: Springs, Co;
1993. p. 1801e4.
References [19] Bourne NK. Materials' physics in extremes: akrology. Metall Mater Trans A
2011;42A:2975e84.
[1] Christman DR, Gehring JW. Analysis of high-velocity projectile penetration [20] Bourne NK. On designing structures for extreme environments. Int J Impact
mechanics. J Appl Phys 1966;37(4):1579e87. Eng 2012;48:107e15.
[2] Orphal DL, Anderson Jr CE. The dependence of penetration velocity on impact [21] Bless SJ, Tarcza K, Chau R, Taleff E, Persad C. Dynamic fracture of tungsten
velocity. Int J Impact Eng 2006;33:546e54. heavy alloys. Int J Impact Eng 2006;33:100e8.
[3] Anderson Jr CE, Hohler V, Walker JD, Stilp AJ. Time-resolved penetration of [22] Wenqi G, Jinxu L, Jia Y, Shukui L. Effect of initial temperature on dynamic
long rods in to steel targets. Int J Impact Eng 1995;16:1e18. recrystallization of tungsten and matrix within adiabatic shear band of
[4] Anderson Jr CE, Orphal DL. Analysis of the terminal phase of penetration. Int J tungsten heavy alloy. Mater Sci Eng A 2011;528:6248e52.
Impact Eng 2003;29:69e80. [23] Pizana C, Murr LE, Baquera MT, Anchondo IA, Putrevu A, Pina CY, et al. Solid-
[5] Lankford J, Anderson Jr CE, Royal SA, Riegel III JP. Penetration erosion phe- state flow, mechanical alloying, and melt-related phenomena for [0 0 1]
nomenology. Int J Impact Eng 1996;18:565e78. single-crystal W ballistic rod penetrators interacting with steel targets. Mater
[6] Gerlach U. Microstructural analysis of residual projectiles e a new method to Sci Eng A 2006;428:301e13.
explain penetration mechanisms. Metall Trans A 1986;17A:435e42.