ADIABATIC SHEAR LOCALIZATION IN AISI 1340 AND 4340

STEELS: THE INFLUENCE OF MICROSTRUCTURE AND

GEOMETRY

A Thesis Submitted to the College of

Graduate Studies and Research

in Partial Fulfillment of the Requirements

for the Degree of Master of Science

in the Department of Mechanical Engineering

University of Saskatchewan

Saskatoon

By

Homa Mostaghimi Ghomi

© Copyright Homa Mostaghimi Ghomi, August 2011. All rights reserved.

 PERMISSION TO USE

In presenting this thesis in partial fulfillment of the requirements for a Postgraduate degree from
the University of Saskatchewan, I agree that the Libraries of this University may make it freely
available for inspection. I further agree that permission for copying of this thesis/dissertation in
any manner, in whole or in part, for scholarly purposes may be granted by Professor Akindele
G. Odeshi who supervised my thesis work. It is understood that any copying or publication or
use of this thesis or parts there of for financial gain shall not be allowed without my written
permission. It is also understood that due recognition shall be given to me and to the University
of Saskatchewan in any scholarly use which may be made of any material in my thesis.

DISCLAIMER

The University of Saskatchewan were exclusively created to meet the thesis and exhibition
requirements for the degree of Master of Science at the University of Saskatchewan. Reference
in this thesis to any specific commercial products, process, or service by trade name, trademark,
manufacturer, or otherwise, does not constitute or imply its endorsement, recommendation, or
favouring by the University of Saskatchewan. The views and opinions of the author expressed
herein do not state or reflect those of the University of Saskatchewan, and shall not be used for
advertising or product endorsement purposes.

Requests for permission to copy or to make other uses of materials in this thesis in whole or part
should be addressed to.

Prof. James D. Bugg [Department of Mechanical Engineering]

University of Saskatchewan

Saskatoon, Saskatchewan S7N 5Z9

Canada

OR

Dean

College of Graduate Studies and Research

University of Saskatchewan

107 Administration Place

Saskatoon, Saskatchewan S7N 5A2

Canada

i
Abstract

The mechanical behaviour of AISI 1340 and 4340 steel under high strain-rate loading in
compression and in torsion were investigated using direct impact Hopkinson bar and
torsion split Hopkinson bars. Both alloys contained 0.40 wt. % C, but different amounts
and types of alloying additions. The materials are commonly used in high performance
structural applications, where they could be subjected to dynamic shock loading. The
objective of this study was to study the effects of microstructure, strain rates and
specimen geometry on the occurrence and failure of adiabatic shear bands in these
alloys under dynamic shock loading.

Cylindrical specimens of the AISI 1340 alloys were heat treated to produce martensitic,
dual-phase or pearlitic structure and subjected to impact loading at strain rates ranging
between 1000 and 8000 /s. The martensitic test specimens were tempered at 205, 315
and 425 ºC to determine the effects of tempered condition on the adiabatic shear failure
of the alloy. The effects of geometry on strain localization and adiabatic shear banding
in both alloys were investigated by subjecting cylindrical-, cubical-, and truncated
conical-shaped specimens to high velocity impact. The dynamic torsion test involved
rapidly twisting of heat-treated thin-walled tubular specimens of the alloys and
determining the damage evolution during the high strain torsional loading. Both optical
and scanning electron microscopes were used to evaluate the damage evolution in the
specimens after high strain rate loading.

The types of shear band formed in the alloys depended on the microstructure and strain
rate. Deformed bands were formed at low strain rates and there was a minimum strain
rate required for formation of transformed band in both alloys. This minimum strain rate
was highest in the specimens with pearlitic structure and lowest in the specimen with
martensitic structure. The susceptibility of the martensitic specimens to the occurrence
of transformed shear band decreased with increasing tempering temperature. Cracks
were initiated and propagated along transformed bands leading to fragmentation under
the impact loading. The susceptibility of the adiabatic shear bands to cracking was
markedly influenced by strain-rates, initial microstructure and the specimens geometry.
The geometry of the impacted specimen determined the shape of the adiabatic shear
ii
band and the topography of the fracture surface of fragmented specimens. Fractographic
investigation of fragmented specimens showed ductile shear failure and knobby fracture
mode along the transformed band. Investigations of the transformed band using X-ray
Photo Emission Electron Microscopy and Near Edge X-ray Absorption Fine structure
Spectroscopy showed more nickel and less chromium inside the transformed bands in
impacted AISI 4340 steel than in the region outside the shear band.

iii
Acknowledgements

I would like to acknowledge the people who contributed to a successful completion of

this work. My appreciations go to my parent for their moral and financial support. I am
very grateful to my supervisor, Professor Akindele G. Odeshi, for all his unflinching
support, encouragement and guidance. His deep understanding of situation is greatly
valued and highly appreciated. The assistance of Professor M. Nabil Bassim in the
experimental work of this study in University of Manitoba is also is appreciated. My
gratitude also goes to Dr. Richard Evitts, Dr. Ikechukwuka N. Oguocha, and Dr.
Qiaoqin Yang for their valuable course of materials. To Prof. Jerzy A. Szpunar, Prof.
James Johnston, my committee members, I say big thank you for your time. I also would
like to thank Prof. Leon Wegner for his time as the external examiner. I thank Prof.
James D. Bugg, the Head of Department of Mechanical Engineering for his
understanding, support and help. The financial support the Natural Sciences and
Engineering Research Council of Canada (NSERC) for this research project in form of
Discovery Grant to my supervisor is highly appreciated. I also wish to acknowledge
Robert Peace for his assistance and support in my experimental work. I also would like
to thank Dr. Uday Lanke of the Canadian Light Source (CLS) for his help in my
experimental work. Lastly, lots of thanks to all the department technicians, office
support staff, other graduate students and especially my very supportive friends for their
support.

iv
 Table of Contents
Abstract..............................................................................................................................ii
Acknowledgements.......................................................................................................... iv
Table of Contents...............................................................................................................v
List of Figures.................................................................................................................viii
List of Tables...................................................................................................................хv

CHAPTER ONE: INTRODUCTION................................................................................1

1.1. Overview………………….…………………………………………………...….....1
1.2. Motivation…….………………...………………………………..………………….2
1.3. Research objectives………………….……………………………..………………..3
1.4. Methodology…………………………………………………………..…………….3
1.5. Thesis arrangement………………......……………………………………………...4

CHAPTER TWO: LITRETURE REVIEW.......................................................................5

2.1. Adiabatic shear band...................................................................................................5
2.1.1. Types of adiabatic shear bands.............................................................................7
2.1.2. Multiple adiabatic shear bands..............................................................................9
2.3. Adiabatic shear band initiation…………………………………………………….10
2.4. Mechanisms of formation of adiabatic shear bands..................................................13
2.4.1. Phase transformation...........................................................................................14
2.4.2. Dynamic recovery...............................................................................................15
2.4.3. Dynamic recrystallization……………………………………………………...15
2.4.4. Progressive subgrain misorientation (PriSM).....................................................18
2.5. Microstructure and properties of adiabatic shear bands...........................................18
2.6. Adiabatic shear band failure.....................................................................................26
2.7. Adiabatic shear band recovery..................................................................................31
2.8. Summary…………………………………………………………………………...31

v
CHAPTER THREE: MATERIALS AND EXPERIMENTAL PROCEDURE..............33
3.1. Materials....................................................................................................................33
3.1.1. Specimens...........................................................................................................34
3.2. Experimental procedure............................................................................................36
3.2.1. Heat treatment procedure....................................................................................36
3.2.2. High strain-rate tests...........................................................................................38
3.2.2.1. Compression test...........................................................................................38
3.2.2.2. Torsion test....................................................................................................41
3.2.4. Microstructural investigations.............................................................................44
3.2.5. Microhardness measurements.............................................................................47

CHAPTER FOUR RESULTS AND DISCUSSION.......................................................50

4.1. Stress-strain curve.....................................................................................................50
4.1.1. Overview.............................................................................................................50
4.1.2. The effect of impact momentum.........................................................................51
4.1.3. The effect of microstructure................................................................................58
4.1.4. The effect of alloying element............................................................................64
4.2. Metallographic analysis............................................................................................67
4.2.1. Overview.............................................................................................................67
4.2.2. The effect of impact momentum.........................................................................71
4.2.3. The effect of microstructure................................................................................82
4.2.4. The effect of geometry........................................................................................86
4.3. Fracture of impacted specimens................................................................................91
4.4. X-ray photoemission electron microscopy (XPEEM)............................................103
4.5. Microhardness results.............................................................................................106
4.5.1. Hardness values across the shear band.............................................................106
4.5.2. The effect of microstructure on hardness values..............................................108
4.5.3. The effect of impact momentum on hardness value.........................................111

CHAPTER FIVE: CONCLUSIONS AND RECOMMENDATIONS..........................113

5.1. Conclusions……………………………………………………………………….113

vi
5.2. Limitation and strength of thesis………………………………………….………115
5.3. Recommendation for future works……………………………………………….116

REFERENCES………………………………………………………………………...117

vii
 List of Figures
Figure Page

Figure 2.1. Schematic views of possible mechanisms of adiabatic shear band

initiation…………………………….……………………………... 12
Figure 2.2. Schematic view of dynamic recovery mechanism…........………… 15
Figure 2.3. Schematic view of dynamic recrystallization mechanism, (a)
dislocation multiplication, (b) dislocation patterning forming
elongated cells. (c) & (d) formation of subgrains, and (e)
breakdown of elongated cells into very fine subgrains. Adapted
from [38].….....…............................................................................. 17

Figure 2.4. Schematic view of crack formation inside adiabatic shear band….. 28
Figure 2.5. Schematic view of voids nucleation inside the shear band, (a) in
the absence of shear stress, (b) in the presence of shear stress…..... 29
Figure 3.1. Thin-walled tubular specimens for torsion test………..…………. 35
Figure 3.2. Specimens shape (a) cylinder, (b) cone, (c) cube….........……..….. 35
Figure 3.3. Iron-iron carbide phase diagram…............…....…………..……….. 37
Figure 3.4. Schematic view of the direct impact Hopkinson bar......................... 38
Figure 3.5. Photograph of direct impact Hopkinson bar used in this study......... 39
Figure 3.6. Direct impact test calibration result................................................... 40
Figure 3.7. Schematic view of torsional Kolsky bar testing system…………... 41
Figure 3.8. Photograph of torsional Kolsky bar used in this study...................... 42
Figure 3.9. Torsional Kolsky bar calibration result for incident and reflected
waves……………………………………………………………… 43
Figure 3.10. Torsional Kolsky bar calibration result for transmitted waves…… 43
Figure 3.11. Photograph of pneumatic mounting press used for specimen
mounting…………………………………………………………… 45
Figure 3.12. Photograph of the Nikon Eclipse MA100 optical microscope used

viii
 in this study……………………………………………………….. 46
Figure 3.13. Joel Scanning Electron Microscope (SEM) ……………………… 47
Figure 3.14. Vickers microhardness tester…………………………………….. 48
Figure 3.15. Optical micrograph showing Vickers hardness indentations on
quench-hardened AISI 1340, tempered at 425 C, impacted at 39.1
kg.m/s: (a) inside the shear band, (b) outside the shear band…… 49
Figure 4.1. Typical stress-strain curve for specimens subjected to dynamic
shock loading [martensitic AISI 1340 steel, tempered at 425 C,
cylindrical specimen and subjected to impact loading at 33.2
kg.m/s ( =1146 s-1)]……………………………………………….. 51
Figure 4.2. Dynamic impact stress-strain curves for martensitic AISI 1340
steel specimens (tempered at 315˚C) as a function of the impact
momentum and strain rates..…….………………………………... 53
Figure 4.3. Stress vs. deformation time curves for martensitic AISI 1340 steel
specimens (tempered at 315˚C) as a function of the impact
momentum and strain rates...……...……………………………… 53
Figure 4.4. Dynamic impact stress-strain curves for martensitic AISI 1340
steel specimens (tempered at 205˚C) as a function of the impact
momentum and strain rates...……………………………………... 54
Figure 4.5. Dynamic impact stress-strain curves for martensitic AISI 1340
steel specimens (tempered at 425˚C) as a function of the impact
momentum and strain rates……………………………………….. 54
Figure 4.6. Dynamic impact stress-strain curves for pearlitic AISI 1340 steel
specimens as a function of the impact momentum and strain
rates................................................................................................... 55
Figure 4.7. Dynamic impact stress-strain curves for AISI 1340 steel as a
function of the heat-treatment condition (impacted at 36.1
kg.m/s)……………………………………………………………... 59
Figure 4.8. Stress vs. deformation time curves for martensitic AISI 1340 steel
specimens (tempered at 315˚C), as a function of heat-treatment
condition…………………………………………………………… 60

ix
Figure 4.9. Dynamic impact stress-strain curves for AISI 1340 steel as a
function of the heat-treatment condition (impacted at 32.2
kg.m/s)……………………………………………………………... 61
Figure 4.10. Dynamic impact stress-strain curves for AISI 1340 steel as a
function of the heat-treatment condition (impacted at 33.2
kg.m/s)…………………………………………………………… 61
Figure 4.11. Dynamic impact stress-strain curves for AISI 1340 steel as a
function of the heat-treatment condition (impacted at 34.2
kg.m/s)…………………………………………………………… 62
Figure 4.12. Dynamic impact stress-strain curves for AISI 1340 steel as a
function of the heat-treatment condition (impacted at 38.1
kg.m/s)……………………………………………………………. 62
Figure 4.13. Dynamic impact stress-strain curves for AISI 1340 steel as a
function of the heat-treatment condition (impacted at 39.1
kg.m/s)……………………………………………………………... 63
Figure 4.14. Torsional stress-strain curves for AISI 1340 steel as a function of
the heat-treatment condition (1619 N.m torque)………………… 63
Figure 4.15. Stress-strain curves for martensitic steels, tempered at 315˚C,
cylindrical specimens, impacted at 34.2 kg.m/s…………………. 65
Figure 4.16. Stress-strain curves for martensitic steels, tempered at 315˚C,
cylindrical specimens, impacted at 36.1 kg.m/s………………….. 66
Figure 4.17. Stress-strain curves for pearlitic structure steels, cylindrical
specimen (impacted at 38.1 kg.m/s)……………………………… 66
Figure 4.18. Stress-strain curves for pearlitic structure steels, cylindrical
specimen, (impacted at 39.1 kg.m/s)…………………….……... 67
Figure 4.19. Optical micrographs of AISI 1340 steel specimens before impact
test. (a) austenitized at 850 C, water- quenched and tempered at
315 C – martensitic structure, (b) austenitized at 850 C and air-
cooled – pearlitic structure , (c) austenitized at 740 C, water-
quenched and tempered at 315 C - dual-phase structure…………. 68

Figure 4.20. Optical micrographs of AISI 4340 steel specimens before impact

x
 test. (a) austenitized at 850 C, oil- quenched and tempered at 315
C – martensitic structure, (b) austenitized at 850 C and air-
cooled – pearlitic structure , (c) austenitized at 740 C , oil-
quenched and tempered at 315 C - martensitic structure…………. 70

Figure 4.21. Optical micrographs of martensitic AISI 1340 steel, tempered at

315˚C, impacted at: (a) 34.2 kg.m/s, (b) 36.1 kg.m/s, (c) 39.1
kg.m/s, (d) 42.1 kg.m/s…………………………………….............. 72
Figure 4.22. Optical micrographs of fractured cylindrical martensitic AISI 1340
steel, tempered at 315 C, impacted at 42.1 kg.m/s…………….…... 73
Figure 4.23. Optical micrographs showing deformed bands in the pearlitic AISI
1340 steel, impacted at: (a) 34.2 kg.m/s, (b) 36.1 kg.m/s, (c) 39.1
kg.m/s, (d) 42.1 kg.m/s, (e) 44.1 kg.m/s, (f) 49.1 kg.m/s………….. 75
Figure 4.24. SEM micrographs of pearlitic AISI 1340 steel showing
microstructure inside and outside the shear band (impacted at 44.1
kg.m/s)………………...…………………………………………… 76
Figure 4.25. Optical micrographs showing adiabatic shear bands in dual-phase
structure AISI 1340 steel specimen (tempered at 315˚C), impacted
at: (a) 36.1 kg.m/s, (b) 38.1 kg.m/s, (c) 39.1 kg.m/s, (d) 42.1
kg.m/s……………………………………………………………… 78
Figure 4.26. Optical and SEM micrographs of regions inside and outside the
shear band for dual-phase structure AISI 1340 steel, tempered at
315 C, impacted at 39.1 kg.m/s......................................................... 80
Figure 4.27. Optical micrographs showing adiabatic shear failure of dual-phase
steel (AISI 1340), impacted at 42.1 kg.m/s.……………………...... 81
Figure 4.28. Optical micrographs of martensitic AISI 1340 steel, impacted at
34.1 kg.m/s, cylindrical samples: (a) tempered at 205 C, (b)
tempered at 315 C, (c) tempered at 425 C……………………...... 82
Figure 4.29. Optical micrographs of AISI 1340 steel, impacted at 38.1 kg.m/s,
cylindrical specimens: (a) martensitic, tempered at 315 C, (b)

xi
 Pearlitic, (c) Dual-phase structure………………………………..... 84
Figure 4.30. Optical micrographs of cylindrical specimen of martensitic steel,
(tempered at 315 C), impacted at 34.2 kg.m/s: (a) AISI 1340 steel,
(b) AISI 4340 steel…........................................................................ 85
Figure 4.31. Optical micrographs of specimens partially austenitized at 740 C,
quenched-hardened and tempered at 315 C, impacted at 36.1 kg.
m/s: (a)AISI 1340 steel, (b) AISI 4340 steel……………………..... 86
Figure 4.32. Optical micrographs of martensitic AISI 1340 steel, tempered at
315 C, impacted at 36.1 kg.m/s. (a) cubic sample corner, (b) cubic
sample close to the edge, (c) cylinder……………………...……… 87
Figure 4.33. Optical micrographs of martensitic AISI 4340 steel, tempered at
315 C, impacted at 34.2 kg.m/s: (a) cubic, (b) cylinder, (c)
truncated cone(small cross-section), (d) truncated cone(large
cross-section)……………………………………………….......... 88
Figure 4.34. Optical macrographs for martensitic AISI 4340 steel, Tempered at
315 C, 34.2 kg.m/s: (a) cylindrical, (b) conical, (c) cubical
specimen………………………………………………………….... 89
Figure 4.35. Schematic view of adiabatic shear band in (a) cylindrical, (b)
cubical, (c) conical Specimens……………...………….………… 90
Figure 4.36. Optical micrographs showing cracks along a transformed shear
band. (a) martensitic AISI 1340 steel, tempered at 425 C,
cylindrical specimen, impacted at 42.1 kg.m/s, (b) martensitic
AISI 1340 steel, tempered at 315 C, cylindrical specimen,
impacted at 42.1 kg.m/s, (c) dual-phase structure AISI 1340 steel,
tempered at 315 C, cylindrical specimen, impacted at 44.1 kg.m/s,
(d) martensitic AISI 1340 steel, tempered at 315 C, Cubic
specimen, longitudinal cross-sectional area, (e) martensitic AISI
4340 steel, tempered at 315 C, cylindrical specimen, impacted at
92
38.1 kg.m/s.......................................................................................
Figure 4.37. Optical micrograph of martensitic AISI 1340 steel, tempered at

xii
 315 C, cylindrical specimen, impacted at 42.1 kg.m/s…………….. 93
Figure 4.38. SEM micrographs of martensitic AISI 1340 steel, tempered at 315
C, cylindrical specimen, impacted at 42.1 kg.m/s……….......…….. 94
Figure 4.39. SEM micrographs showing the fracture surface of AISI 1340 with
dual-phase structure (cylindrical specimen)……………………… 96
Figure 4.40. SEM micrographs showing the fracture surface of martensitic
AISI 1340 steel (cubic specimen), impacted at 38.1 kg.m/s, (a)
elongated dimples, (b) knobby features.………………………....... 98
Figure 4.41. SEM micrographs showing the fracture surface of martensitic
AISI 1340 steel (conical specimen), impacted at 36.1
kg.m/s………............................................................……………… 98
Figure 4.42. SEM micrographs showing fracture surface of martensitic AISI
4340 steel, under high strain-rate torsional loading……………….. 99
Figure 4.43. Optical micrograph showing longitudinal section of fracture
cubical specimen of martensitic AISI 1340 steel………………….. 100
Figure 4.44. Optical macrographs of fractured cylindrical specimen of
martensitic, AISI 4340 steel……...……………………….……….. 101
Figure 4.45. Optical macrographs of fractured cubical specimen of martensitic,
AISI 4340 steel…………………………………………………… 102
Figure 4.46. Optical macrographs of fractured conical specimen of martensitic,
AISI 4340 alloy steel, conical specimen…………………….…….. 102
Figure 4.47. Schematic view of crack propagation path, (a) cylindrical, (b)
cubical, (c) conical Specimens…………..……………………….... 103
Figure 4.48. (a) Cr 2p-3d NEXAFS spectra from inside and outside the shear
band, (b) Ni 2p-3d NEXAFS spectra from inside and outside the
shear band………………………………………………………….. 104
Figure 4.49. PEEM imaging using Hg lamp, 150μm FOV…...……………….... 104
Figure 4.50. XPEEM image showing nickel map, using synchrotron light
radiation……………………………………………………………. 105
Figure 4.51. Vickers hardness value across adiabatic shear band of martensitic
AISI 1340 steel, tempered at 315 C, cylindrical specimen,

xiii
 impacted at 39.1 kg.m/s…................................................................. 107
Figure 4.52. Vickers hardness values for AISI 1340 steel before impact
loading, (A) martensitic, tempered at 205 C (B) martensitic,
tempered at 315C, (C) dual-phase structure, tempered at 315 C,
(D) pearlitic structure…………………………………………….... 109
Figure 4.53. Vickers hardness values for AISI 1340 steel, Impacted at 36.1
kg.m/s, (A) martensitic, tempered at 205 C, (B) martensitic,
tempered at 315C, (C) dual-phase structure, tempered at 315 C,
(D) pearlitic structure…………………………………………….... 110
Figure 4.54. Vickers hardness values for martensitic AISI 1340 steel as
function of impact momentum (IM)…………….……………….... 112

xiv
 List of Tables
Table Page

Table 3.1. Chemical compositions of AISI 1340 and 4340 carbon steels…… 34
Table 3.2. Summary of heat treatment procedure…………………………….. 37
Table 4.1. Experimental data sheet for the impact tests……………………… 56

xv
 CHAPER ONE
INTRODUCTION
1.1. Overview

Dynamic shock loading at high strain-rates (103 s-1) leads to local plastic deformation
along narrow bands. This deformation differs from low strain-rate or quasi-static loading
conditions where plastic deformation is relatively homogeneous. An example of
dynamic shock loading is deformation of materials under high impact velocity, and the
narrow bands of extreme shear strain localization are called adiabatic shear bands
(ASBs). Heat generated during impact loading as a result of conversion of kinetic
energy of the impacting object to heat energy during deformation, leads to thermal
softening which occurs simultaneously with the strain hardening effect of plastic
deformation. A near adiabatic condition usually arises in which heat generated along
some narrow paths are not conducted away. This causes local thermal softening and
thermo-viscous instabilities that lead to strain localization and occurrence of adiabatic
shear bands. Adiabatic shear bands are of two types: Deformed and Transformed bands.
Deformed bands appear as greatly distorted regions demonstrating extensive shear
deformation of the grains. Transformed bands appear as white narrow bands in the
impacted materials. Transformed bands are also called white etching adiabatic shear
bands because of their characteristic white colour when observed under an optical
microscope. The microstructure inside the transformed bands cannot be resolved under
optical or Scanning Electron Microscope (SEM). However, observations of transformed
adiabatic shear bands using Transmission Electron Microscope (TEM) in many
materials indicated that they consist of very fine grains of sub-micron size.

The excessive thermal softening and highly localized strain inside adiabatic shear band
can cause fragmentation of even ductile metals under dynamic mechanical loading.
During high strain-rate deformation, tensile stresses are generated inside shear bands as

1
a result of the lower flow stress inside the adiabatic shear bands compared to the outside.
The tensile stresses generated inside the shear band eventually become sufficiently high
enough to open up micro-pores inside these bands. By coalescence of the micro-pores,
voids are formed which elongate and rotate to elliptical shapes. These are finally
connected, initiating microcracks which propagate along the shear band leading
ultimately to fracture. Adiabatic shear bands’ hardness values are usually higher than
that of the bulk material due to the high density of dislocations and very fine nature of
the grains inside the bands. Consequently, they are also very brittle and materials
containing adiabatic shear bands are more susceptible to fracture along shear bands
during subsequent loading. Therefore, fundamental understanding of the mechanism of
formation and failure of adiabatic shear bands, as influenced by materials or process
variables such as composition, strain-rate, microstructure, geometry, etc., is very
important for controlling adiabatic shear failure of engineering components under
dynamic shock loading.

1.2. Motivation
The materials investigated as part of this study were AISI 1340 and AISI 4340 alloy
steels, with both containing 0.40 % C and varying degrees of alloying additions. AISI
4340 steel is one of the most popular High-Strength Low Alloy (HSLA) steels which are
commonly used in high strength applications such as in the automotive industry,
pressure vessels, and gas pipelines. It has considerable amounts of alloying additions,
such as manganese (Mn), chromium (Cr), nickel (Ni), and molybdenum (Mo). Mn, Cr,
Ni and Mo in steels increase the steel hardenability. Moreover, Cr and Mo form carbides
which consequently increase matrix strength. Meanwhile, Mn helps in refining the grain
size. AISI 1340 steel is a low carbon manganese steel that is mostly used in the
automotive body construction. The low amount of carbon and high amount of
manganese make AISI 1340 steel appropriate for formation of steel with a dual-phase
structure, which consist of hard martensite in a soft ferrite matrix. The manganese
addition expand the ferrite + austenite phase region in the iron-iron carbide equilibrium
phase diagram, while the low carbon content, enhances the ability to intercritically
anneal the steel in the ferrite+austenite phase region of the iron-iron carbide equilibrium
phase diagram (shown in Fig. 3.3 on page 37). Moreover, the high amount of manganese

2
and carbon content of the austenite at the intercritical annealing temperature provides
enough hardenability for the austenite within the AISI 1340 steel to form martensite
during quenching. As a result of the unique properties of dual-phase steels, which
include high strength and remarkable ductility, they are widely used in automotive body
construction to save weight and consequently improve fuel consumption efficiency.

Knowledge of the formation and failure of adiabatic shear bands inside AISI 1340 and
4340 steels under dynamic shock loading as influenced by such factors such as heat-
treated condition is a leading point for a fundamental understanding of high strain rate
behaviour of these alloys. The fragmentation of pressure vessel or steel pipe due to
excessive pressure or failure of automobile parts from a car crash can be traced to high
strain-rate loading and adiabatic shear failure of the AISI 4340 and 1340 steels used in
these applications, respectively.

1.3. Research objectives

The research objectives of this study were to:

1. Investigate the effects of microstructure and strain rates on plastic deformation and
failure of AISI 1340 and AISI 4340 steels under dynamic shock loading,

2. Investigate the effect of specimen geometry on the formation of adiabatic shear band
in AISI 1340 and AISI 4340 steels under dynamic impact loading, and

3. Investigate the chemical and bonding changes associated with the occurrence of
transformed bands in alloy steel.

1.4. Methodology
In order to achieve the research objectives, cylindrical, cubical, conical (truncated) and
thin-walled tubular specimens were machined from AISI 1340 and AISI 4340 steels.
These thin-walled tubular and cylindrical specimens were heat-treated to obtain
pearlitic, martensitic or dual phase structures in order to investigate the effects of
microstructure on formation and failure of adiabatic shear band in the alloys under
dynamic shock loading. Another set of cylindrical, cubical and conical specimens of the
alloys were heat treated to obtain martensitic microstructure and used to investigate the

3
effects of specimen geometry on the characteristic of adiabatic shear bands. All the
specimens were subjected to dynamic shock loadings using instrumented direct impact
Hopkinson bar or Torsional Kolsky Bar. The stress-strain curves obtained were
discussed in relation to the sequence of plastic deformation leading to adiabatic shear
failure of the specimens. The effects of strain-rates on deformation behaviour and
microstructural evolution inside the materials were investigated. The characteristics of
adiabatic shear bands observed in the impacted alloys were investigated in relation to
the pre-impact microstructure of the steel specimens. Fractographic studies of the
fractured specimen were undertaken to understand the failure mechanisms in the alloy
under the impact loading and high strain-rate loading in torsion. Synchrotron light
radiation at the Canadian Light Source (CLS) was used to investigate the compositional
changes associated with the occurrence of transformed bands in the AISI 4340 alloy.
This provided further insight into microstructural evolution associated with the
occurrence of transformed bands in this high strength steel and support the design effort
of tailoring materials microstructure in order to make them more resistant to adiabatic
shear failure during dynamic shock loading.

1.5. Thesis arrangement

The current chapter of the thesis introduces the focus, motivation and objective of this
M.Sc. research project. Chapter two consists of a literature review of previous research
investigations on the behaviour of materials under dynamic shock loadings. Relevant
information on materials investigated and experimental techniques are outlined in
chapter three, while chapter four contains results, their analyses and discussion. A
summary of this study and the conclusions drawn from the research investigations are
presented in chapter five.

4
 CHAPTER TWO
LITERATURE REVIEW

The behaviour of materials under dynamic shock loading as reported in previous studies
is reviewed in this chapter. The concept of shear strain localization along adiabatic shear
bands at high strain rates is discussed. The general definition of adiabatic shear band,
types of adiabatic shear bands, mechanisms of formation and failure of adiabatic shear
bands, as well as properties and microstructure of adiabatic shear bands in a variety of
materials are summarized in this chapter.

2.1. Adiabatic shear bands

Plastic deformation at low strain-rates or under quasi-static loadings is relatively
homogeneous. Therefore, the heat generated inside the material dissipates and the
material strain hardens as the shear strain increases. It is governed by slip and twinning
mechanisms [1]. However, at strain rate greater than 103s-1, such as impact loading,
deformation becomes localized along narrow bands which are called adiabatic shear
bands (ASBs) [2,3] . Heat generated during impact loading leads to thermal softening
which occurs simultaneously with the strain hardening effect associated with plastic
deformation. A near adiabatic condition usually arises in which heat generated along
narrow paths are not conducted away, leading to excessive thermal softening and
thermo-viscous instability that leads to the strain localization along the adiabatic shear
bands [2,3]. Adiabatic shear band concepts have been studied experimentally [2-4] and
analytically [5-9]. The concept of an adiabatic shear band was first proposed by Zener
and Holloman [3]. They suggested that extensive shear strain inside the shear band
causes temperature rise within shear band. Staker [10] suggested that a true shear strain
at which onset of adiabatic shear band instability begins is measured experimentally by:
(2.1)

5
where C is volume specific heat, n is strain hardening exponent, and δτ/δT is the slope of
temperature dependence of flow stress. This relationship is applicable for any material
whose stress-strain curve’s first part is parabolic.

Temperature rise within adiabatic shear band depend on the material, composition, and
strain rate. Odeshi and Bassim [11] suggested the heat generated inside the adiabatic
shear band of AISI 4340 steel specimens during high velocity impact caused the
temperature rise within adiabatic shear band close to melting point of the alloy in some
regions. Duffy and Chi [12] measured the local temperature within adiabatic shear band
by small high speed infrared detectors as a function of time and position for a low
carbon cold-rolled steel and three quench-hardened steels (HY-100 and two tempers of
AISI 4340 VAR steel of varying hardness) which were subjected to high strain-rate
deformation using torsional Kolsky bars. The temperature was estimated to reach about
600 C along the adiabatic shear band propagation path [13]. Local temperature within
adiabatic shear band of SS400 steel deformed by a fine blanking process was calculated
to be in the neighbourhood of 600 C [14]. The calculated temperature inside the shear
band of tantalum specimens subjected to high strain rate deformation was determined to
be about 800 C, and onset of recrystallization was observed at high microscopic strains
in isolated areas [15]. Hartley et al. [16] determined the temperature rise within the
adiabatic shear bands of cold rolled AISI 1018 steel and hot rolled AISI 1020 steel to be
450 C based on calculations from the emitted infrared radiation from the surface of the
metals.

The temperature rise within AISI 4340 VAR steel, with the hardness value 44 on the
Rockwell C scale, subjected to dynamic shock loading by Torsional Kolsky Bar testing
was calculated to be 460 C using infra-red detectors [17].The highest temperature rise
inside the adiabatic shear band of C-300 steel (maraging steel) was calculated to be 1400
C (90% of melting point), while that for Ti-6Al-4V was about 450 C [18]. For
polycrystalline zirconium alloy (Zircadine 702, containing 0.7% Hf) which was
subjected to impact loading using a Split Hopkinson Pressure Bar for hat-shaped
specimens, temperature rise within shear band was calculated to be 930K for 100 shear
strain using the Zerilli-Armstrong equation [19]. For low-carbon steel plates with

6
chemical composition of 0.15%C–0.25%Si–1.1%Mn– Fe (wt.%) which were subjected
to high strain-rate torsional loading, the temperature rise within adiabatic shear band
was found to be 540 C with the high temperature reported to promote spheroidization of
cementite inside the shear band. The temperature rise within adiabatic shear band was
calculated by:

(2.2)

where, ρ is density, Cp is heat capacity, and β is a work to heat conversion factor which
is taken as 0.9. Modeling of temperature history inside the shear bands shows that heat
generation inside the shear band occurs at less that 200μs, while the shear band is at
maximum temperature for 19μs [20].

The higher amount of plastic deformation and strain hardening make the shear band
material harder than the bulk material. As a result of high temperature and excessive
thermal softening inside the shear band, the materials inside the shear band are subjected
to more extensive shear strain than those outside shear band region. Therefore, adiabatic
shear bands are usually brittle and their presence in a material makes it more susceptible
to fracture than the material without ASBs. Fragmentation of materials which was
deformed under high velocity impact were observed to be preceded by the formation of
adiabatic shear bands [21, 22]. Adiabatic shear bands have been observed in practical
applications such as high speed machining, explosive fragmentation [2, 13, 23], machine
chips [24, 25], metal forging, and ballistic impact [26].

2.1.1. Types of adiabatic shear bands

There are two types of Adiabatic Shear Bands: deformed and transformed bands [27].
Deformed bands appear as greatly distorted areas demonstrating extensive shear
deformation of the grains. They are commonly observed in non-ferrous materials such
as aluminum and copper alloys. They have also been observed in commercially pure
niobium subjected to dynamic impact loading at low temperature [28]. Transformed
bands appear as white narrow bands in the impacted materials. They are also called
white etching bands, because of the white color within ASBs after etching with nital.
White etching bands are commonly observed in hardened steels [1, 13, 21, 26, 29- 32].

7
Timothy et al. [27] proposed that transformed bands are generally observed in metals
with low thermal diffusivity and low resistance to adiabatic shear localization. These
metals can transform to other phases at high temperatures and then generate a
metastable phase by rapid cooling to room temperature, the quenching medium being
the materials surrounding the shear band. White shear bands have also been reported in
some aluminum alloys [33], tantalum [34], copper [35, 36], and zirconium alloys [37].
Adiabatic shear bands generally have been reported in varieties of engineering materials
including rolled homogenous armour (RHA) [9], RDX/DOS composition
(cyclotrimethylene trinitramine (RDX) to which was added various percentages of the
binder dioctyl sebacate (DOS)) [28], titanium [38], commercially pure niobium [39],
commercially pure vanadium [40], and densified silicon carbide powders [41].

There are different explanations for the white colour of the transformed adiabatic shear
bands. It has been suggested that it is due to the phase transformation occurring during
dynamic shock loading. Zener and Holloman [3], Cho et al. [13], Glenn and Leslie [29],
and Rogers [42] suggested that during adiabatic heating the temperature rise inside the
adiabatic shear band is high enough to cause transformation to austenite in steel along
adiabatic heated path. The austenite region is quenched by the surrounding matrix to
form untempered martensite. Moreover, the hardness of transformed bands in steel has
been found to be close to that of untempered martensite [20].

The theory that phase transformation, leading to the formation of untempered

martensite, occur inside adiabatic shear band during high strain-rate loading, has not
been adequately substantiated in the literature. Meanwhile, high strain-rate deformation
happens in less than 100μs. Therefore, the formation of untempered martensite as a
result of quenching of the austenite inside the white etching band is doubtful in this
short space of time. Cho et al. [13] calculated the temperature reached inside the
transformed shear band of HY-100 steel specimens during a high strain-rate deformation
to be 600 C. This is below the temperature at which austenitization will occur, thereby
ruling out the possibility of martensitic transformation.

White etching bands are difficult to resolve under optical and scanning electron
microscopes. Transmission electron microscopic observation indicates that they consist

8
of very fine grains of sub-micron size. It has been suggested that during massive and
swift deformation, the martensite, and carbides laths break apart into sub-grains with a
few hundreds of nanometer size [1, 13]. Li et al. [44] observed equi-axed nano-sized
grains in adiabatic shear bands which are proposed to form as a result of dynamic
recrystalization. Equi-axed nano-sized structure inside the shear band has been observed
by Meyers et al. [15] for tantalum, Xue et al. [33] for aluminum alloy, Hines and
Vecchio [35] , and Andradea et al.[36] for copper, Li et al.[43] for Monel alloy, Rittel et
al. [45], and Me-Bar and Shechtman[46] for titanium.

The nanosized grains of transformed shear bands offer another explanation for the white
colour and featureless nature of these bands under an optical microscope. Wittman et al.
[20] suggested that the white colour of the adiabatic shear bands is due to the limit of
optical microscopes in resolving the extremely tiny particles making up the shear band.
The white colour of the white etching band after etching with nital on the polished
surface was also attributed to carbide dissolution, which changed the microstructure and
consequently the etching properties of the shear band [20].

Although, white etching bands are commonly observed in hardened steel, depending on
the heat treatment procedure, microstructure and deformation conditions, both types of
shear bands can form in steels [11].

2.1.2. Multiple adiabatic shear bands

A single adiabatic shear band may be formed inside a material under dynamic shock
loading causing fracture of the specimen into two pieces. Multiple adiabatic shear bands
can also form leading to fragmentation into several pieces. The study of multiple
adiabatic shear bands by Xue et al. [40] for thick-walled cylinder (TWC) of AISI 304
stainless steel subjected to explosive loading showed that initiation and propagation of
multiple adiabatic shear bands changed with changes in initial loading condition.
However, these effects did not change the spacing of the shear bands. Average length of
shear bands, as well as the spacing between them was the same in different sections,
while highest shear band length varied in different sections. The middle section with
highest propagating shear band speed had the maximum length. The phenomenon of
multiple adiabatic shear bands formation has been investigated by Xue et al. [47] for

9
pure titanium and Ti-6Al-4V alloy which were deformed under explosive loading of
thick-walled cylinders at a strain-rate of 104s-1. Multiple adiabatic shear bands initiated
at the internal boundary of the tested thick wall cylindrical specimens and exhibited
periodic distribution from early stages of formation. The adiabatic shear band pattern
showed a self-organization property [47, 48]. The propagation velocity of shear band
was higher for Ti-6Al-4V alloy than pure Ti alloy. However, the numbers of adiabatic
shear bands were higher for pure Ti alloy than Ti-6Al-4V. Adiabatic shear bands
showed bifurcation (splitting into two or more bands) when they formed in a spiral
pattern or when the distance between them increased [47]. Loading factors and the type
of material have a significant effect on adiabatic shear band spacing [47]. Spacing of the
multiple adiabatic shear bands also depends on microstructural inhomogeneities which
offer the adiabatic shear band initiation sites [48].

2.3. Adiabatic shear band initiation

The susceptibility to the occurrence of adiabatic shear bands depends on material
properties. Materials with high strain rate sensitivity and low temperature sensitivity of
flow stress are less likely to form adiabatic shear bands [42]. The tendency of a material
to form adiabatic shear bands depends on its original microstructure and its
thermodynamic stability [13]. Materials with high strength and low strain hardening are
more susceptible to adiabatic shear localization [49]. Lee et al. [50] suggested that
metals with higher strength (yield strength and fracture toughness) are more inclined to
form adiabatic shear bands under dynamic mechanical loadings than those with lower
yield strength.

Feng and Bassim [24] suggested that local material defects or any inhomogeneities
inside the material cause adiabatic shear band initiations. These materials defects are
called the initial perturbations. Wright [51] suggested that parametric solutions for the
formation of adiabatic shear bands show that total effective perturbation of the tested
samples can be calculated by simply adding the initial perturbation in strength,
temperature, and wall thickness. Molinari and Clifton [52] suggested that the critical
strain at which the stress collapse happens depends logarithmically on the size of the
initial perturbation. As the size of initial perturbation increases, the collapse strain

10
decreases logarithmically. Investigation by Duffy and Chi [12] on the torsional Kolsky
Bar testing of low carbon cold-rolled steel and three quench-hardened steels (HY-100
and two tempers of AISI 4340 VAR steel of varying hardness) also showed that the
critical strain at which adiabatic shear bands are formed depends on the initial defects.
Mataya et al. [53] suggested that γ' precipitates in the size of 10 nm increases the
susceptibility of γ' strengthened austenitic stainless steel JBK-75 to form shear
localization. It has been suggested that at high strain-rate, localized flow occurs along
high angle grain boundaries in precipitates free zone in the form of recrystallized grains
[53].

Bassim [34] suggested that differences in properties and microstructure cause the
material to form adiabatic shear bands. Local defects and inhomogeneities inside the
material act as preferential sites for shear band initiation. It has been also suggested that
specimen geometry and dimensions affected the formation of adiabatic shear band.
Nesterenko et al. [6] suggested that shear localization initiates along the grains which
are favourably orientated. They also suggested that uneven distribution of grain size can
act as perturbations. Larger grains show lower yield strength, while smaller grains
shows higher yield strength. Therefore, larger grains deform initially and can act as
preferential sites for shear localization. They also proposed that crystallographic
peculiarities have a very important role in shear band initiation in the way that
decreasing the grain size under 100μm increases the number of shear bands [6]. Meyers
et al. [54] investigated the effect of grain size on the high strain rate behaviour of copper
and suggested that localization of plastic deformation is observed for the coarse-grain
sample while the copper sample with smaller grain size exhibited homogenous plastic
deformation.

Figure 2.1 shows the possible mechanisms for shear band initiation suggested as by
Nesterenko et al. [6] which includes: 1) Grain size inhomogeneity, 2) Geometrical
softening, 3) Peirce-Asaro-Neddleman textural localization and 4) Dislocation pile-up
release.

11
Figure 2.1. Schematic views of possible mechanisms of adiabatic shear band initiation,
adapted from [6].

The experimental and numerical study of the effects of notch length and notch root
radius on AM50 and Ti6Al4V alloys showed that adiabatic shear band failure is more
sensitive to the sharpness of the imperfections than the size of them. Numerical
simulation of energy distribution showed a high energy concentration near the notch.
The high amount of energy near the notch motivates nucleation of localized ASBs [45].

Specimen’s geometry can also influence the occurrence of ASBs in an alloy. For
example, a previous study by Li et al. [55] showed that shear strain localization did not
happen for cylindrical tungsten heavy alloy (WHA), while truncated-conic specimens
showed adiabatic shear band formation, which initiated from the upper-right corner to
the centre of the opposite surface with the angle of 45 to the upper surface of the
specimen. It was suggested that adiabatic shearing in the truncated-conic specimens
formed as a result of an uneven stress condition along the specimen as well as shear

12
instability at the critical maximum shear stress. Surface friction was reported to
introduce a higher shear stress in the upper-right corner, which suggests that initiation of
adiabatic shear bands can be influenced by surface friction.

Several mathematical models including Graddy-Kipp (GP) [56] and Wright Ockendon
(OW) models [57] have been proposed for initiation and propagation of adiabatic shear
bands. Nesterenko et al. [6], in comparing these two models, observed that the
experimental results were in a good agreement with theoretical results. They also
suggested that initiation of shear band is mostly better represented by the Wright-
Ockendon model which is based on the perturbation analysis, while propagation of the
shear band is better represented by the Graddy-Kipp model theory which is based on
momentum diffusion. They suggested that there is no evidence of diffusion of
momentum before shear localization initiation. Diffusion momentum only plays a role
during shear localization propagation.

Deformation under dynamic shock loading occurs in stages. Marchand and Duffy [37]
suggested that plastic deformation at high strain-rates consists of three stages:
homogenous strain hardening, inhomogeneous strain hardening, and deformation
localization along narrow adiabatic shear bands. Feng and Bassim [24] and Bassim [34]
also suggested the adiabatic shear band formation in AISI 4340 steel occurs in three
stages: (1) homogenous elastic deformation, (2) simultaneous occurrence of strain
hardening and thermal softening, in which strain hardening is dominant, and (3) thermal
softening dominated deformation. Nesterenko et al. [38] suggested that localization of
deformation occurs once the critical strain for stress collapse is exceeded, which is
accompanied by heterogeneous plastic deformation inside the shear band.

2.4. Mechanisms of formation of adiabatic shear bands

There are numerous uncertainties with regards to the mechanism of formation and
development of microstructure within adiabatic shear bands. However, several theories
have been suggested to explain the mechanism of formation of adiabatic shear bands.
These include phase transformation [3, 13, 12, 29, 42], dynamic recovery [32, 38, 58],

13
dynamic recrystallization [15, 32, 35, 38, 53, 58], and progressive subgrain
misorientation [12, 47, 53, 60].

2.4.1. Phase transformation

Phase transformation is one of the earliest theories proposed to explain the occurrence of
transformed bands in steels. Zener and Holloman [3] believed that formation of white
etching adiabatic shear bands occurs by phase transformation of the room temperature
structure to austenite and its subsequent transformation to untempered martensite as a
result of quenching by the surrounding materials. Cho et al. [61] also considered the
white etching bands that form in AISI 4340 steels to be a product of phase
transformations leading to the formation of untempered martensite. Chen et al. [62]
suggested that, for steels which are deformed under dynamic shock loadings, the
temperature within adiabatic shear bands can reach up to 1500˚C. At this temperature,
phase transformation of ferrite to austenite will occur. Therefore, white etching bands in
high strength steels can form as a result of phase transformation from austenite to un-
tempered martensite. The measured hardness values of the transformed bands have been
observed to be within the range of that for un-tempered martensite [3]. This gives
credence to the theory that the constituent of transformed bands is untempered
martensite.

The theory of phase transformation accompanying formation of transformed bands is in

dispute. Cho et al. [13] observed that, at a strain rate around 103 s-1, the temperature
reaches 600 C inside the transformed band, formed in HY-100 steel during a high strain-
rate deformation. They concluded that, since the temperature inside the shear band did
not reach the austenite phase region, the formation of untempered martensite as a result
of quenching the austenite inside the white etching band is therefore very unlikely.
Zurek [1] observed ultra-fine equi-axed grains inside transformed bands in high strength
low alloy steel and suggested that the higher hardness inside the shear band is due to the
fine grain size of the transformed band rather than phase transformation. TEM
investigation of adiabatic shear bands that form in a Monel alloy under dynamic shock
loading by Li et al. [43] showed that the shear band has the same structure as the matrix

14
(FCC), indicating no phase transformation occurred inside the shear band under
dynamic shock loading.

2.4.2. Dynamic recovery

Dynamic recovery is one of the suggested mechanisms for the formation of transformed
adiabatic shear bands in steels. Chen et al. [14] proposed this mechanism for the
formation of transformed bands in steel during the fine blanking process, as shown in
Fig. 2.2. The mechanism involves elongation and reorientation of grains (ferrite and
pearlite) along the applied shear stress flow direction. A dislocation interaction in the
elongated grains leads to partitioning into very fine grains through formation of cell
walls. The final stage of the dynamic recovery process involves spheroidization
resulting in densely packed very fine subgrains inside shear band [14]. Kad et al. [19]
explained dynamic recovery leading to formation of transformed bands in a zirconium
alloy (Zircadine 702) somewhat differently and proposed it to occur in three stages: 1)
elongation of cells and subgrains, 2) development of misorientation between
neighbouring grains and fragmentation of subgrains to smaller parts, and finally 3) grain
rotation due to rotation of grain boundaries resulting in evolution of a fine equi-axed
grain microstructure.

Figure 2.2. Schematic view of dynamic recovery mechanism, adapted from [14].

2.4.3. Dynamic recrystallization

Another mechanism used to explain the grain refinement inside transformed bands is
dynamic recrystallization. As a result of the transient nature of the adiabatic shear

15
banding process, static recrystallization can not justifiably explain the formation of the
fine equi-axed grains found in transformed bands. Static recrystallization is diffusion
controlled and needs more time to occur than the period of formation of ASB which is
of the order of a few hundred microseconds. Meyers et al. [15], Barry and Byrne [32],
Hines and Vecchio [35], Nesterenko and Meyers [38], Mataya et al. [53], and Hwang et
al. [58] have suggested dynamic recrystallization as the mechanism of formation of
transformed bands in metallic alloys. A dynamic recrystallization model, which involves
grain boundary diffusion has been suggested by Meyers et al. [63]. This model was
called Rotational Dynamic Recrystallization and suggested to be much faster than bulk
diffusion.

The rotational recrystallization mechanism proposed by Lins et al. [60] consists of

dislocation multiplication and patterning to form an elongated cell structure,
transformation of these cells into elongated subgrains surrounded by lamellar boundaries
with high angles (misorientation), and, finally, division of elongated subgrains by
formation of boundaries accompanied by local crystal rotation. This sequence of events
in dynamic recrystallization was also reported by Li et al. [43] and Nesterenko et al.
[38]. Formation and growth of recrystallized grains under dynamic shock loading inside
the shear band depends on the strain-rate and density of dislocations. Atomic thermal
migration also controls recrystallized grain growth if the temperature rise within the
adiabatic shear band is high enough to allow this [43]. Nesterenko et al. [38]
summarised the mechanism for dynamic recrystallization, as shown in Fig. 2.3, to occur
in 4 stages as follows:

1. Multiplication of dislocations, which are dispersed randomly,

2. Arrangement of dislocations in the form of elongated dislocation cells,
3. Generation of subgrains by dislocation pattering as the deformation progresses, and
4. Break-down of subgrains into equi-axed submicron size grains. As the deformation
increases, grain rotation occurs.

Xu et al. [33] suggested that the temperature at which recrystallization occurs in metals
is about 0.4-0.5 Tm. Andradea et al. [36] also observed that temperature rises inside the
shear band of the copper specimens were in the range of 500-800K, which is the

16
temperature range at which recrystallization can occur [36]. However, the calculated
temperature inside the shear bands of tantalum subjected to the high strain-rate
deformation was about 800 C, which is lower than the recrystallization temperature.
Nevertheless, at high microscopic strains in isolated areas, the onset of recrystallization
was observed [15].

Figure 2.3. Schematic view of dynamic recrystallization mechanism, (a) dislocation

multiplication, (b) dislocation patterning forming elongated cells. (c) & (d) formation
of subgrains, and (e) breakdown of elongated cells into very fine subgrains.
Adapted from [38].

17
2.4.4. Progressive subgrain misorientation (PriSM)
Progressive Subgrain missorientation (PriSM) has also been suggested for the formation
of transformed bands. This involves dynamic recrystallization of grains, which occur by
the formation and mechanical rotation of subgrains, followed by boundary refinement
during shear band cooling after deformation. Boundary refinement occurs by decreasing
the surplus dislocation poles developed mainly by dislocation climb during shear band
cooling. During the short period time of deformation, there is not enough time for
dislocation annihilation and boundary refinement for the material inside the adiabatic
shear band. However, the cooling period, during which the temperature within the
adiabatic shear band is still perfectly high, is longer than the deformation period.
Therefore, dislocation annihilation and boundary refinement is kinetically possible
during the cooling of the transformed band after dynamic shock loading [60]. Hines et
al. [64] also proposed that the progressive subgrain misorientation recrystallization
model accounts for evolution of subgrain misorientation inside the adiabatic shear band.
This model is based on a bicrystal approach. Evolution of microstructure inside the
shear band is proposed to proceed in two stages: firstly, recrystallization as a result of
formation and mechanical rotation of subgrains during the deformation, and secondly,
boundary refinement due to the diffusion after deformation and during shear band
cooling.

Perezprado et al. [65] observed no evidence of dynamic recrystallization within the

shear band in hat shaped specimens of Ta and Ta-W alloys that were subjected to
dynamic impact loading. They suggested that the temperature rise inside the shear band
was not high enough to cause dynamic recrystallization. The Progressive subgrain
misorientation (PriSM) mechanism, which is based on the mechanical subgrain rotation,
is considered to be a more suitable mechanism to explain the microstructure evolution
inside the shear band. Under tested conditions of time and temperature, dynamic
recrystallization occurrence is doubtful for this material.

2.5. Microstructure and properties of adiabatic shear bands

Transformed adiabatic shear bands contain very fine, sub-micron size grains with an
average size of tenths of micrometers [42]. It is suggested that carbide lamellae inside

18
the shear bands formed in pearlitic SS400 steel, which was deformed by a fine blanking
process, became elongated as a result of intense shear strain inside the shear bands. The
elongated carbide lath eventually fragmented into smaller grains which underwent
spheroidization forming equi-axed grains. The formation of very fine ferrite subgrains
inside the adiabatic shear band was due to the fragmentation of carbide particles as well
as compression of carbon and iron diffusion paths under hydrostatic pressure and at high
temperatures within the shear band. The hydrostatic pressure inside the adiabatic shear
band of the pearlitic SS400 was suggested to hinder the formation of microcracks inside
the shear band, although the material inside the shear band is subjected to large shear
stresses [14].

TEM investigations of adiabatic shear bands in AISI 4340 steel by Wittman et al. [20]
showed highly deformed carbide particles with the same structure as the surrounding
matrix. They also suggested that partial spheroidization occurs inside the shear band due
to the fragmentation of carbide laths, as well as carbon and iron diffusion, which is
associated with the nucleation of fine ferrite subgrains. High temperature, as well as
shear strain inside the band, cause the fragmentation of carbide particles and accelerate
the shear band formation [1]. The adiabatic shear band investigation in tempered
martensitic structure of HY-100 steel by Cho et al. [13] showed that ASB consisted of
two distinctive regions: the highly elongated grains along adiabatic shear bands, and fine
equi-axed cells. Derep [67] observed that the microstructure inside the adiabatic shear
band in an armour steel after deformation under dynamic shock loading contained a fine
grain equi-axed structure of delta ferrite, mixed with narrow laths of martensite. Lesuer
et al. [68] studied the occurrence of adiabatic shear bands in an ultra-high carbon steel
(UHCS) containing 1.3% C, 3.0% Si, 0.5% Mn, 0.99% Cr and observed a mixture of
nano-sized carbides and ultra-fine grains of ferrite (70μm size) inside the adiabatic shear
band. They suggested that temperature rise inside the shear band caused transformation
to austenite followed by rapid cooling, which resulted in divorced eutectoid
transformation (DET) after severe plastic deformation (SPD) occurrence inside the shear
band. The microstructure inside adiabatic shear bands consisted of nano-sized carbides
in super fine grain ferrites (70 μm). They also observed higher hardness in the shear
band, compared with other parts of the specimen [68].

19
Li et al. [43] observed fine equi-axed subgrains with the average diameter of 150 nm, as
well as low density of dislocations at the center of adiabatic shear bands of a Monel
alloy which was subjected to dynamic shock loading using a Split Hopkinson Pressure
Bar. TEM observation of 8090 Al–Li alloy subjected to dynamic impact loading showed
that the material inside the adiabatic shear band consisted of deformed cells with a high
dislocation density, while the deformed region between the matrix and shear band
consisted of highly elongated cells with well defined boundaries. On the other hand, the
structure at the center of the shear band consisted of very fine equi-axed subgrains with
low density of dislocations with the size of about 0.2μm. It is suggested that these equi-
axed subgrains were formed as a result of dynamic recrystallization [33].

TEM investigation inside the shear band of 304L stainless steels, which were subjected
to dynamic impact loading, indicated the nucleation of ά martensite (ά martensite forms
in metastable austenitic stainless-steel by plastic deformation [69]) and the existence of
a high density of dislocations inside the shear band. As the strain-rate and stress rate
increased, the density of dislocations as well as the fraction of ά martensite inside the
shear band increased. Therefore, the density of dislocations and ά martensite formation
inside the shear band are sensitive to strain-intensity and strain-rate [70, 71]. Meyers et
al. [63] also observed the microstructure of adiabatic shear bands in AISI 304L stainless
steel (Fe–18%Cr–8%Ni) specimens using Electron Backscatter Diffraction (EBSD) and
Transmission Electron Microscopy (TEM). The Electron Backscatter Diffraction
(EBSD) results indicated that regions close to the adiabatic shear band experience grain
subdivision (size 30μm) with angular rotation up to 20 . The microstructure inside the
shear band could not be detected by EBSD. TEM observation of the microstructure
inside the shear band by Meyers et al. [63] revealed two regions: 1) a region with grains
having an average size of 0.1-0.2 μm, which have low dislocation density and well
defined grain boundaries, and 2) a glassy structure region that is formed by a solid-state
amorphization process. This was the first time that an amorphous structure has been
observed inside a shear band. Stacking faults, twinning, martensitic transformation at
twin band intersection and highly plastic deformation were observed outside the shear
band. Therefore, it was suggested that plastic deformation and high temperature inside
the shear band changes the microstructure from a dislocated and twinned structure to a

20
submicron size equi-axed structure with well-defined grain boundaries. For the fine
grains of an average size of 0.2 µm, grain boundary rotation with the rotation angle of
about 30 with respect to the estimated deformation time of approximately 10-50 μs has
been reported [66].

TEM observations of AISI 316L stainless steel specimens by Xue and Gray[23] showed
well-defined adiabatic shear bands with a mixture of equi-axed, rectangular or elliptical,
and elongated subgrains. They observed elongated subgrain laths and a high density of
dislocation at the peripheral region of the transformed band, while fine equi-axed
subgrains with the average size of 100 nm were observed inside the core of the shear
band. The fine equi-axed grains were proposed to form as a result of fragmentation of
elongated laths or subcells reconstruction by patterning of the high density of
dislocations generated inside the shear bands. As the deformation increased, the
adiabatic shear band core width increased and consequently the width of region with
elongated subgrains decreased [23].

TEM observations by Hwang et al. [58] on low-carbon steel plates that were subjected
to dynamic shock loadings using torsional Kolsky bar, showed elongated ferrite and
very fine equi-axed ferrite inside the shear band. Realignment of dislocations inside the
elongated ferrites by dynamic recovery generated cell structures with very fine equi-
axed grains with the average size of 0.05-0.2 μm. The dynamic recrystallization
mechanism was suggested as a possible mechanism which developed these high- angle
sub-grains with relatively low density of dislocations. It was suggested that cementite
particles inside the shear band were fragmented into smaller particles and spheroidized.
Spheroidized cementite particles pinned down ferrite grain boundaries and hindered
ferrite grain growth. Therefore, ferrite particles inside pearlite inside the adiabatic shear
band were finer than those outside the pearlite [58]. The investigation of the effects of
deformation temperature, strain rate, and preshocking condition on the formation of
adiabatic shear bands by Chen et al. [66] in tantalum specimens that were subjected to
dynamic shock loadings showed that as the deformation temperature decreased, shear
localization increased due to the lower heat capacity . It also has been observed that as
the strain rate increased, shear localization increased due to the higher heat generation

21
inside the material. As the temperature decreased and strain rate increased, the width of
the shear band decreased. They also observed that specimens in preshock conditionings
were more incline to form shear localization [66].

TEM investigation of the white etching bands in a Ti–6Al–4V alloy also showed the
existence of equi-axed grains with the diameter of 50 nm inside the shear band [44].
Optical and transmission electron microscopy of adiabatic shear band in Ti-6Al-4V by
Me-Bar and Shechtman [46] showed that the temperature rise inside the adiabatic shear
band as a result of plastic deformation caused melting of the alloy in some regions
besides the transformation from α phase to β phase that is followed by swift cooling at a
rate of 170 Cs-1. The regions, where melting of the alloys occurred, were totally
distorted, while the regions with β phase were elongated along the shear flow direction
[46]. As a result of temperature rise inside adiabatic shear band of Ti-6Al-4V, and
intense localized plastic deformation, completely distorted regions were observed inside
the shear bands which had an entirely different microstructure from the parent alloy.
This microstructure was suggested to be a tempered form of ά parental microstructure
[72]. Small grains with an average size of 0.3-0.5 μm were observed inside the shear
band of titanium, while the regions close to the shear band showed a high density of
dislocations and deformation twinning. The boundary region between matrix and shear
band of commercially pure titanium subjected to dynamic shock loadings consisted of
slip bands and their intersections. Fine equi-axed submicron size grains were observed
inside the shear band [30].

Submicron size grains with the average size of 0.1 μm were observed inside adiabatic
shear bands formed in copper that was rapidly deformed; the fine grain structure of the
shear bands were proposed to form as a result of dynamic recrystallization [36]. Hines
and Vecchio [35] investigated the adiabatic shear bands in copper and observed small
grains with the average size of 0.1-0.2μm inside the adiabatic shear band. Copper
samples that were deformed in liquid nitrogen at sub zero degree temperature exhibited
no temperature-dependence of recrystallization [35].

Lee et al. [70] observed dislocation and twinning deformation inside the adiabatic shear
band of inconel 690 superalloy specimens subjected to dynamic impact loading. They

22
observed that dislocations were arranged in a form of dislocation tangle and caused the
formation of dislocation cell walls. It was observed that, as the test temperature
increased, dislocation annihilation occurred, and the density of dislocation decreased.
On the other hand, increasing strain-rate led to an increase in dislocation and twinning
densities. Therefore, under high strain-rate loading and at low temperature, smaller
dislocation cell size with thicker walls were formed which hindered the dislocation
motion. Lee and his colleagues observed that, as the temperature increased at a constant
strain rate, the flow stress, strain-rate sensitivity, and work hardening coefficient
decreased, while activation volume and temperature sensitivity increased. However, as
the strain-rate increased at a constant temperature, the flow stress and strain-rate
sensitivity increased, while the activation volume and work hardening coefficient
decreased [70].

Nesterenko et al. [38] observed four kinds of microstructure in a deformed specimen of

tantalum: (1) at effective strain values of less than 1 and a calculated temperature less
than 600K, dislocations and elongated dislocation cells were observed, (2) at effective
strain values of between 1 and 2 and a calculated temperature of 600-800 K, subgrain
formation occurred by dynamic recovery, (3) dynamically recrystallized micrograins
were observed for the effective strain between 2 and 2.5 and a calculated temperature of
between 800 and 900K, and they were proposed to form by dynamic recrystallization,
and (4) at effective strains higher than 2.5 and a temperature higher than 1000K post-
deformation recrystallized grains were observed which were suggested to form by static
recrystallization. It has also been suggested that static recrystallization involves an
atomic migrational mechanism, while dynamic recrystallization involves subgrain
rotation and dislocation annihilation. Microscopic analysis of tantalum samples that
were deformed at high strain-rates by Meyers et al. [15] showed subgrains at the centre
of adiabatic shear bands where the shear strain was 5.5. Elongated cells were observed
in the boundary region between the shear band and the bulk material. Under
transmission electron microscope, twinning deformation has been observed inside the
adiabatic shear band of a low carbon steel with pearlite particles of 18 μm. However,
there was no twinning deformation inside the adiabatic shear band of medium and high

23
carbon steel with the pearlite particle size of 25 and 35 μm, respectively. It has been
suggested that grain boundaries restrict twinning deformation [71].

TEM observation of adiabatic shear bands in AISI 4340 VAR steel with hardness of 44
HRC showed highly elongated subgrains with well defined boundaries and varied
dislocation densities, as well as dislocation cells. There was no evidence of phase
transformation inside the adiabatic shear band. The preferred orientation of these
subgrains was calculated at 110 planes which were parallel to the shear direction [17].
Meyers and Wittman [26] reported that hardness of transformed bands observed in
quench-hardened and tempered low-carbon steels (AISI 1018 and 8620) were
significantly higher than that of untransformed matrix. It has been suggested that this
higher hardness of ASB is due to the fine microstructure with supersaturated carbon. It
has been suggested that hardness inside the transformed band of steels depends on the
carbon concentration [61].

Lee et al. [50] studied microstructural evolution inside adiabatic shear bands for three
aluminum alloys: AA 2090 Al-Li alloy, a weldalite 049TM alloy and AA 7039 Al. They
observed that the microstructure of the adiabatic shear band was totally different in the
2090 Al-Li alloy. It contained highly elongated grains and coarse grain boundary
particles. The cracks propagated along the grain boundaries with coarse particles
(intergranular cracking). These intergranular cracks may prevent adiabatic shear band
formation.

Hat shape specimens of a polycrystalline zirconium alloy (Zircadine 702, containing

0.7% Hf), which were impacted under dynamic shock loadings by Split Hopkinson Bar,
exhibited well-defined adiabatic shear bands with the average width of 9-24μm in all
tested specimens. The equi-axed grains with the average size of 200 nm were observed
inside the shear band. Grains near the adiabatic shear band were subjected to moderate
deformation and aligned themselves along the shear direction [19].

Shear localization occurred for granular materials such as silicon carbide powders under
high strain rate deformation. Comminution and rearrangement of particles were the
suggested deformation mechanism of adiabatic shear bands in granular ceramic

24
materials. Plastic deformation of SiC particles caused the temperature to rise inside the
shear band which melted SiC particles inside the shear band and fused them together
[41]. Nesterenko et al. [73] investigated the effect of particle size of densified granular
alumina specimens which were subjected to dynamic shock loadings by thick-walled
cylinders on the adiabatic shear band characteristics. They observed that adiabatic shear
band width was independent of initial particle size and was approximately 10-20μm.
However, the microstructure of adiabatic shear bands significantly depended on the
initial particle size. Specimens with the initial particle size of 4μm, experienced
comminution (break-up) and particle softening, while observations of those with particle
size of 0.4 μm showed a network of cracks parallel and perpendicular to the shear band.
Initial particle size significantly affected displacement, magnitudes and distribution of
adiabatic shear bands due to the different softening mechanism. Hardness of adiabatic
shear bands increased in both specimens with small and big granular particle sizes. They
suggested that specimens with large grain size had higher hardness due to an occurrence
of microfracture, followed by repacking of particles with different sizes. Meanwhile,
small size initial particles just underwent classic repacking and not particle fracturing.

Meyers et al. [63] observed the adiabatic shear bands in metals (Ti, Ta, Ti-6Al-4V, and
stainless steel), granular and pre-fractured ceramics (Al2O3and SiC), a polymer (teflon),
and a metallic glass (Co58Ni10Fe5Si11B16). They suggested that the microstructure
evolution mechanisms for each group were as follows:

1. Metals: The first stage of deformation is associated with thermal softening following
by dynamic recovery and dynamic recrystallization. It was proposed that, during the
high strain-rate deformation, subgrain fragmentation and grain boundary rotation can
occur.

2. Ceramics: Shear localization depends on the particle size. If the particle size is

greater than a critical value ac, comminution and particle break-up will occur.

However, if the particle size is smaller than critical value, a significant amount of
heat that is generated by plastic deformation and particle repacking will cause the
fusion of the particles and bonding.

25
3. Non-crystalline metals and polymers: Local inelastic transformations initiated the
highly localized regions which organized themselves into shear band zones.

The width of the region in which the grain distortions are clear, in comparison with
other parts of the specimen, is called the width of the shear band. The width of the shear
band depends on the materials, chemical compositions and strain rate [71]. Xue-bin [74]
suggested that wide transformed adiabatic shear bands, as well as high local plastic
deformation between deformed and transformed ASB, are formed as a result of a high
heat conversion factor, low density, low heat capacity and high melting temperature.
High local plastic deformation in deformed adiabatic shear band forms when there is a
high work to heat conversion factor, low density, low heat capacity and low melting
point. They also suggested that difference in materials investigated, loading methods,
conditions, and different definitions of the ASB width are three reasons responsible for
the scatter in experimental results on the width of the shear bands [74]. It has been
observed that the width of the shear band in steels decreased as the carbon content and
strain rate increased [71].

The local shear strain within an adiabatic shear band is defined as γ = tanθ where θ is the
distortion angle [33]. As the distance from the centre of the shear band increases, the
distortion angle decreases. Therefore, the maximum local shear strain occurs at the
centre of the adiabatic shear band. The angle of distortion and local shear strains depend
on the material, chemical composition and strain rate. It has been observed that for
steels, as the amount of carbon and strain rate increased, the local shear strain increased
[71]. Engineering strain can be calculated from the true strain using Equation 2.7 [33]:

(2.7)

2.6. Adiabatic shear band failure

Materials fail along the adiabatic shear bands under dynamic impact loadings. Adiabatic
shear bands are preferential sites for crack initiation and propagation [47, 75]. As the
loading rate increased, the temperature inside of the shear band increased, which has
significant effects on material yielding and plasticity. The optical and scanning electron
microscopic observations of roll homogeneous armour (RHA) plate by Raftenberg and

26
Krause [31], perforated by a pointed projectile, showed the existence of some cracks
along the transformed shear bands blended with deformed adiabatic shear bands along
the indentation cavity of the plates. Zurek [1] also observed a ductile fracture mode
inside the white etching adiabatic shear band of pearlitic AISI 4340 steel. Mataya et al.
[53] observed macroscopic, transgranular cracks inside the shear band of γ' strengthened
austenitic stainless steel, JBK-75. Cho et al. [61] suggested that crack formation along
the adiabatic shear band of AISI 1018 cold-rolled steel occurred in three stages: (1)
formation and coalescence of micro-voids and separation along ferrite boundaries, (2)
separation between pearlite and ferrite particles, and (3) breaking apart of pearlite
lamellae. They also observed that, for HY-100 steel, voids initiated at MnS inclusions
inside the adiabatic shear band.

The schematic view of the crack formation and propagation mechanism inside adiabatic
shear band, as proposed by Bassim and Odeshi [76] for quench-hardened and tempered
AISI 4340 steel is shown in Fig. 2.4. It was suggested that the formation of cracks along
adiabatic shear bands starts with the formation of micro-pores inside the adiabatic shear
bands (Fig. 2.4a). These micro-pores cluster together, forming voids, which act as crack
nuclei. The micro-cracks, nucleated at adjacent voids interconnect leading to formation
of crack, as shown in Fig. 2.4(d) [76]. Rittel et al. [45] suggested that voids’ nucleation
occurs because of 1) vacancy coalescence at high strain-rate regions, 2) formation of
micro-voids because of grain boundary sliding, and 3) void nucleation at secondary-
phase particles and ahead of dislocation pile-ups. It has been reported that the existence
of precipitates are not a necessary factor for void nucleation [45].

27
 Fig. 2.4. Schematic view of crack formation inside adiabatic shear band, adapted from
[76].

Xue et al. [77] studied the failure of adiabatic shear bands in Ti–6Al–4V alloys. They
proposed a model for the evolution of a crack inside shear bands as shown schematically
in Fig. 2.5. They suggested that voids nucleate inside the shear band. Odeshi et al. [78]
suggested that micro-voids nucleate inside adiabatic shear bands as result of tensile
stress generation due to the difference between the flow stress inside and outside the
shear bands. These voids grow until they reach the shear band boundary where the
material is harder due to the lower temperature. Xue et al. [77] proposed that these voids
grow along the shear bands, and finally coalesce, causing fragmentation. The schematic
views of the mechanisms proposed by Xue and his colleagues [77] in the absence and

28
presence of shear stress are presented in Fig. 2.5. In the presence of shear stress, the
voids are elongated and rotated as shown in Fig 2.5(b) [77].

Figure 2.5. Schematic view of voids nucleation inside the shear band, (a) in the absence
of shear stress, (b) in the presence of shear stress, adapted from [77].

A ductile fracture mode has been observed inside and outside the shear band. Ductile
fracture inside the shear band occurred as a result of thermal softening which created
tensile stress and opened up micro-voids inside the adiabatic shear band. On the other
hand, ductile fracture outside the shear band occurred as a result of precipitates, which
act as crack nuclei [45]. Nesterenko et al. [38] suggested that ductile fracture along the
adiabatic shear band of deformed tantalum was due to the residual tensile tangential
stresses upon loading. Cho et al. [61] observed elongated dimples and knobby features
on the fracture surface of an AISI 1018 cold-rolled steel (CRS), a structural steel (HY-
100), and an AISI 4340 vacuum arc remelted (VAR) steel tempered to either of two
hardnesses RHC 44 or 55. They observed elongated dimples, indicating ductile shear
fracture mode. They also observed knobby features on the fracture surface, suggesting
melting of alloy in these regions. The fracture surface of the specimen could melt as a
result of heat generation inside the adiabatic shear band as the two fracture surfaces are
rubbed against one another [42, 61].

29
Odeshi and Bassim [11] observed ductile fracture with the evidence of precipitates at
voids’ origin in bulk material of AISI 4340 steel, which was deformed under dynamic
shock loadings. However, both ductile and cleavage fracture were reported to occur
outside the shear band of AISI 4340 steel. Elongated dimples as a result of ductile
fracture mode were observed in the deformed zone between the bulk material and
transformed band. Fracture along transformed bands showed highly elongated dimples
indicating shear ductile fracture and knobby fracture mode. The dimples were deeper
and narrower inside the transformed shear band than dimples formed in deformed bands
in the region adjacent to the transformed band [11]. Lee et al. [70] also observed
dimples on the fracture path through adiabatic shear band in inconel 690 superalloy
subjected to dynamic impact loading.

Lee and Lin [71] also observed both elongated dimples and knobby features on the
fracture surface of adiabatic shear band in low, medium, and high carbon steels. It has
been observed that as the strain-rate increased, the depth of the dimples increased and
the shear band exhibited more ductile fracture. This indicated that as the strain-rate
increased, the temperature within adiabatic shear band increased. Moreover, the
proportion of knobby features increased as strain rate and carbon content increased, and
the size of dimples decreased with increasing strain rate and carbon content [71].

Ryttberg et al. [79] proposed three damage subzones to form below the fracture surface
of 100Cr6 steel that fractured under impact shock loadings: 1) white etching band, 2)
equiaxed grains, and 3) elongated subgrains (The preferred orientation of subgrains was
110 planes parallel to the shear direction). It was also observed that the region below the
fracture surface of tool steel 100CrMn6 was divided into three zone: 1) right under the
fracture surface, white etching adiabatic shear band containing equi-axed ferrite
particles (size. 50-150 nm); 2) within a distance less than 50 μm under the fracture
surface, lamellae cementite and carbide particles within ferrite matrix, and 3) within the
range of 50-100 μm distance under the fracture surface, elongated grains, distorted
carbide lamellae with small lamellar spacing which align themselves parallel to the
shearing direction. Evidence of deformation twinning of each region was reported [79].

30
2.7. Adiabatic shear band recovery
The susceptibility of the material to fracture under subsequent loading can be reduced
by the annealing shear band that developed in the previous dynamic shock loading.
Odeshi et al. [21] investigated the effects of post-impact heat treatment on the adiabatic
shear band in AISI 4340, which were formed under dynamic impact loading. It was
observed that annealing the white etching band at 315 C for 2h did not affect the
microstructure inside the shear band. However, hardness value of the shear band
decreased by soaking the specimen band at 650 C for 20 minutes. By soaking the
specimen at 650 C for 20 minutes the hardness value of the shear band decreased to the
same value as that of the material outside the shear band. By extending the soaking time
at 650 C to 2h, the white ASB changed to a gray colour of the ductile deformed band.
Coarsening of nano-sized particles inside the adiabatic shear band was observed to
occur, which accounted for the reduced hardness of the shear band when soaked at 650
C for 20 minutes.

2.8. Summary
At strain rate greater than 103s-1, deformation become localized along narrow bands in
materials. The narrow paths of intense strain localizations are called Adiabatic Shear
Bands (ASBs). Materials fragmentation under high strain-rate is usually preceded with
the occurrence of adiabatic shear bands, especially transformed bands which contained
nano-sized subgrains and offer preferential sites for crack initiation and propagation.
Therefore, fundamental understanding of the mechanisms of formation these bands and
their failure is very important to prevent adiabatic shear failure of engineering
components under dynamic shock loadings. It is very important to investigate and
document the dynamic impact response of various engineering materials in relation to
adiabatic shear banding as influenced by microstructure, strain rates and geometry.
Although mechanical properties of most engineering alloys under static loading are
readily available in the literature, the same cannot be said for the case of dynamic shock
loading. The need to generate property data sheets for all materials and make them
readily available for the use of design engineers in materials selection for high strain-
rate applications cannot be overemphasized.

31
The current study was conducted to document the dynamic mechanical response and
adiabatic shear banding in AISI 1340 steel as influenced by strain rates, microstructure
and specimen’s geometry. The dynamic mechanical behaviour of the alloy specimens
heat-treated to have martensitic, pearlitic and dual-phase structure will be investigated.
Such a detailed investigation of AISI 1340 steel under dynamic impact loading has not
been previously conducted. The results to be obtained from this investigation will
contribute to knowledge on dynamic impact behaviour of this alloy and to generation of
materials database for dynamic mechanical behaviour of the material. Although AISI
4340, the second alloy intended for investigation in this research study, have been
investigated by a number of researchers [1, 11,12, 20,21], the effects of geometry of the
on the occurrence and trajectory of adiabatic shear bands in the steel has not been
investigated as planned in this research study.

In order to provide further insight to the microstructural evolution that culminates in the
occurrence of transformed bands in steel, chemical analyses of the regions inside and
outside these bands are planned for investigation using the synchrotron light radiations
at the Canadian Light Source (CLS) in Saskatoon. The investigation will be carried out
at the soft x-ray spectromicroscopic (SM) beamline of the CLS using X-ray
Photoemission electron microscopy (XPEEM) and Near edge X-ray absorption Fine
Structure (NEXAFS). Such an in-depth chemical analysis of the nano-grains inside
transformed bands in metallic materials has not been previously undertaken as planned.
Data to be generated from synchrotron study of transformed band will likely open a new
line of discourse on the microstructural evolution associated with the occurrence of
transformed bands in steel in particular and metallic alloys in general.

32
 CHAPTER THREE
MATERIALS AND EXPERIMENTAL PROCEDURE

The materials used in this investigation, the geometry of the specimens, heat treatment
procedure, high strain-rate deformation tests (Direct impact Hopkinson Bar test,
Torsional Kolsky Bar test) are explained in this chapter. Metallographic preparation and
procedures for microstructure analysis using Optical Microscopy (OM), Scanning
Electron Microscopy (SEM), X-ray Photoemission electron Microscopy (XPEEM), and
microhardness testing are also provided in this chapter.

3.1. Materials
The chemical compositions of AISI 1340 and 4340 steel used in this investigation are
shown in Table 3.1. The AISI 1340 steel was supplied by Stelco Inc. in Hamilton
Ontario (now called US steel Canada), which also provided the chemical composition of
the steel. AISI 4340 steel was purchased from steel vendor by mechanical engineering
workshop in college of engineering at University of Saskatchewan. The chemical
composition range of AISI 4340 was looked up in ASM handbook volume 1 [82]. They
both contain 0.40 wt. % C. whereas AISI 1340 steel is manganese steel containing 1.54
wt. % Mn and 0.14 wt. % V, the AISI 4340 steel is a low alloy steel containing
substantial amounts of Mn, Cr and Mo. AISI 4340 steel is one of the most popular High-
Strength Low Alloy (HSLA) steels, and is commonly used in high strength applications
in automobiles and trucks, gas pipelines, pressure vessels, etc. The low carbon content
and high manganese content of AISI 1340 steel makes it suitable for processing into
dual-phase structure. The manganese expands the ferrite + austenite phase region in the
iron-iron carbide equilibrium phase diagram and the low carbon content enhances the
ability to partially austenitize the steel in the ferrite + austenite phase region. On
quenching the steel after intercritical annealing in the ferrite + austenite phase region,
the austenite will transform, leaving a microstructure that consists of hard martensite
embedded in a continuous soft ferrite matrix. The high manganese content and high
carbon content of the austenite in the intercritical temperature range provides enough

33
hardenability for the transforming austenite [82]. As a result of the unique combination
of good strength and ductility in dual-phase steel, it can be rolled into high strength thin
plates suitable for automobile bodywork. Thus, AISI 1340 steel is widely used in
automotive body construction to enable cars to save weight and consume less fuel [82].

Table 3.1. Chemical compositions of AISI 1340 and 4340 carbon steels

Elements AISI 1340 steel (wt%) AISI 4340 steel (wt%)

C .40 .40
Mn 1.54 0.60-0.80
P .022 0.035
S .015 0.04
Si .27 0.15-0.30
Cu .009 -
Ni .012 1.65−2.00
Cr .050 0.70-0.90
Mo .002 0.20-0.30
V .14 -
Cb .068 -
Ti .002 -
Ni .012 1.65-2.00

3.1.1. Specimens
The geometry of the test samples are shown in Figs. 3.1 and 3.2. Samples for torsional
tests were machined from hot-rolled AISI 1340 and 4340 steel rods into thin-walled
tubular specimens with a gage length of 3.8 mm and wall thickness of 0.4 mm (Fig. 3.1).
The compression specimens investigated with direct impact Hopkinson, which were
machined from the hot-rolled rods of AISI 1340 and 4340 steels, were in three shapes:
cylinder, cube, and truncated cone as shown in Fig. 3.2. The cylindrical specimens had
a diameter and length of 9.5 mm and 10.5mm, respectively. The cubical-shape
specimens had dimensions 9.00 mm × 9.00 mm × 10.00 mm. The specimens machined

34
into the shape of a truncated cone had radii of 9.5 mm and 8.5 mm, and a height of 10.5
mm.

Figure 3.1. Thin-walled tubular specimens for torsion test.

Figure 3.2. Specimens shape (a) cylinder, (b) cone, (c) cube.

35
3.2. Experimental Procedure
3.2.1. Heat treatment Procedure
In order to determine the effects of microstructure on the dynamic impact response of
the alloys, machined specimens were heat-treated to obtain pearlitic, martensitic or dual
phase structures. The heat treatment procedures with respect to temperature, time and
quench medium is summarized in Table 3.2. The heat treatment temperature was
designed using the iron-iron carbide equilibrium phase diagram shown in Fig. 3.3. To
obtain martensitic structure, steel specimens having 0.4 wt. % C were austenitized at
850 C for 30 minutes using Linderg blue Furnace located in College of Engineering
(Room 2C24), University of Saskatchewan and then quenched in water, in the case of
AISI 1340 steel or oil in the case of AISI 4340 steel to obtain martensitic structure. The
cylindrical martensitic AISI 1340 steel specimens were divided into three groups,
tempered at 205 C, 315 C or 425 C, respectively, for one hour. AISI 4340 steel
specimens were tempered at only 315 C for one hour. The effect of tempering
temperature on the AISI 4340 has been investigated in a previous study [11, 21, 83]. To
obtain pearlitic structure, specimens were austenitized at 850 C for 30 minutes and then
cooled down to the room temperature in air. The microstructure at room temperature
consisted of ferrite and pearlite. Dual-phase structure was obtained by austenitizing the
specimens at 740 C in the ferrite + austenite phase region (Fig. 3.3) for 20 minutes and
then quenching in water in the case of AISI 1340 steel or oil in the case of AISI 4340
steel. These specimens were then tempered at 315 C for one hour.

36
 Figure 3.3. Iron-iron carbide phase diagram.

Table. 3.2. Summary of heat treatment procedure

AISI 1340 Steel
Microstructure Austenitizing Time Quench Tempering Number of Specimens
Temperature (min) Temperature Impact Torsion
(˚C) (˚C) Test Test
Martensitic 850 30 In Water 205 (1 hr) Cylinder 10 -

850 30 In Water 315(1 hr) Cylinder 10 3

Cube 6
Cone 6

850 30 In Water 425(1 hr) Cylinder 10 -

DP Structure 740 20 In Water 315(1 hr) Cylinder 10 3

Pearlitic 850 30 In air - Cylinder 10 3

AISI 4340 Steel

Microstructure Austenitizing Time Quench Tempering Number of Specimens
Temperature (min) Temperature Impact Impact
(˚C) (˚C) Test Test
Martensitic 850 30 In oil 315(1 hr) Cylinder 4 3
Cube 6
Cone 6

DP Structure 740 20 In oil 315(1 hr) Cylinder 4 3

Pearlitic 850 30 In air - Cylinder 4 3

37
3.2.2. High strain-rate tests
3.2.2.1. Compression test
The dynamic mechanical behaviour of the materials under compressive loading at high
strain rate was investigated using an instrumented Direct Impact Hopkinson Bar. Figure
3.4 shows a schematic view of the direct impact Hopkinson Bar and a photograph of the
direct impact Hopkinson Bar used in this investigation, which is located in Department
of Manufacturing and Mechanical Engineering at University of Manitoba, is shown in
Fig. 3.5. The projectile, which is fired by the light gun, travels through the gun barrel
and strikes the sample at high impact velocity. The projectile was fired at varying firing
pressure to produce projectile velocities that ranged between 20 and 32 m/s at the time it
struck the specimen. The projectile was made of quench-hardened AISI 4340 steel and
weighed 1.905 kg. As the specimen was impacted by the projectile, elastic waves were
produced, which transmitted through the specimen to the output bar. The transmitted
elastic wave signals were captured by a strain gage attached to the output bar. The
captured strain waves were conditioned and amplified by the attached strain pulse
amplifier. The amplified strain pulse data were recorded using a mixed signal digital
oscilloscope.

Figure 3.4. Schematic view of the direct impact Hopkinson bar.

38
 Figure 3.5. Photograph of direct impact Hopkinson bar used in this study.

The elastic waves are captured by the oscilloscope as time (s) vs. voltage (V) data. The
voltage can be converted to the impacted load (N) by calibrating the strain gage on the
out-put bar. In order to calibrate the gage, the output bar was subjected to quasi-static
compressive loads of 1 kN to 7 kN in increments of 1 kN. The voltage corresponding to
each applied load (N) was measured and a plot of load (N) vs. voltage (V) was made, as
shown in Fig. 3.6. The slope of load (N) vs. voltage (V) relationship was found to be
21798 N/V, which gave the conversion factor for voltage measured by the oscilloscope
into the corresponding force in N.

39
 8000
7000
6000
y = 21796x
Load (N) 5000
4000
3000
2000
1000
0
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35
Voltage (V)

Figure 3.6. Direct impact test calibration result.

If it is assumed that specimens have constant volume and a linear variation of

displacement with time, as well as constant strain rate, the true stress and true strain
values can be calculated from the following equations:

(3.1)

(3.2)

where, Li and Lf are initial and final length, respectively. The data for the dynamic
stress-strain curves for the materials under impact loading were computed using
Equations 3.1 and 3.2. The strain rate corresponding to each impact velocity is directly
proportional to the maximum strain and inversely proportional to the length of the
striker bar (l), which was 0.223 m in this case, as shown in Equation 3.3. The strain-rate
was calculated using this equation:

(3.3)

where, C0 is the longitudinal wave propagation velocity in the transmitter bar.

40
3.2.2.2. Torsion test
A torsional Kolsky Bar was used in conducting the high strain rate test in torsion on the
test specimens. A schematic view of the Torsional Kolsky Bar is shown in Fig. 3.7,
while a photograph of the bar used in this study which is located in Department of
Mechanical and Manufacturing Engineering at University of Manitoba, is shown in Fig.
3.8. The system consists of two collinear bars. The incident one can rotate easily within
the support of bearings. The short, thin-walled specimen is loaded between the two bars.
The torsional wave, which is produced by a sudden release of a stored torque in the
loading end of the incident bar, travels through the incident bar and is captured by the
strain gage attached to the incident bar. On reaching the specimen, it loads and deforms
the specimen rapidly. The incident wave is partially transmitted to the transmitter bar
and partially reflected back to the incident bar. The transmitted wave is captured by the
strain gage attached to the transmitter bar while the reflected wave is captured by the
strain gage attached to the incident bar. The amplifier amplifies the detected strain data
and sends them to an oscilloscope, which collects and stores the data. The stress, strain,
and strain-rate can be calculated from the incident, transmitted and reflected wave data.

Figure 3.7. Schematic view of torsional Kolsky bar testing.

41
 Oscilloscope Input Bar

Output Bar Clamp

Torque Pulley
Specimen

Figure 3.8. Photograph of torsional Kolsky bar used in this study.

The corresponding shear strain for the elastic wave data in the form of voltage recorded
by the oscilloscope was determined by loading a connected incident and transmitter bar
in torsion at incremental angles of twist while reading the voltage generated by the strain
gages on the bars. A plot of shear strain against voltage is shown in Figs. 3.9 and 3.10,
which gives the conversion factor for obtaining the shear strain that corresponds to
voltage output from the strain gages on the incident and transmitter bars. By using the
slopes of the lines in Figs. 3.9 and 3.10, the incident strain (I), transmitted strains (T)
and reflected strains (R) were computed from the voltage data from the strain gages.
The shear strain vs. voltage relationships were calculated as follows:

Incident bar. I=1.323x10-4 x Voltage (V) (3.4)

R=1.323x10-4 x Voltage (V) (3.5)

42
Transmitted bar. T=1.314x10-4 x Voltage (V) (3.6)

0.0007
0.0006 y = 0.000132312x
0.0005
Strain

0.0004
0.0003
0.0002
0.0001
0
0 1 2 3 4 5
Voltage (V)

Figure 3.9. Torsional Kolsky bar calibration result for incident and reflected waves.

0.0007
0.0006
0.0005 y = 0.000131367x
0.0004
Strain

0.0003
0.0002
0.0001
0
0 1 2 3 4 5
Voltage (V)

Figure 3.10. Torsional Kolsky bar calibration result for transmitted waves.

43
The strain-rates were calculated as follows [84]:

(3.7)

where rs is the mean radius of the thin-wall specimen, Ls is the length of the specimen, c
is the wave speed propagation velocity in bar, and rb is the radius of the bar. The shear
strain in the specimen was calculated by integrating the strain rate, as shown in Equation
3.5 [84]:

(3.8)

The shear stress was determined using Equation 3.6 as follows [84]:

(3.9)

where G is the shear modulus [84].

3.2.4. Microstructural Investigations

Deformed specimens that had been subjected to dynamic shock loadings were mounted
in phenolic resin using the pneumatic mounting press shown in Fig. 3.11. The mounted
specimens were pre-grinded with water on SiC-220 grit paper mounted on a rotating
polishing wheel for 1 minute. Then, they were grinded on the MD-largo disc on a low
speed polishing wheel with MD-Largo polishing liquid for 5 minutes. The grinded
specimens were then polished in two polishing stages. In stage one, specimens were
polished on a MD-Dac polishing cloth with MD-Dac liquid on a low speed polishing
wheel for 4 minutes. In the final stage, they were polished on MD-Nap polishing cloth
with MD-Nap liquid on low speed polishing wheel for 1 minute. The polished samples
were then etched with 2% nital for about 16 seconds.

44
 Figure 3.11. Photograph of pneumatic mounting press used for specimen mounting.

Prepared specimens were observed under optical microscope (Nikon ECLIPSE

MA100) using four different magnifications: 50X, 100X, 500X, and 1000X located in
Room ENG 2C25 in the Department of Mechanical Engineering at University of
Saskatchewan. Specimens with well formed white shear bands were then cut along the
longitudinal cross-section, polished and etched in 2% nital for 16 seconds. The
specimens were then observed under the optical microscope to give the geometry of the
shear bands along longitudinal cross-section of the impacted specimen. Figure 3.12
shows the Nikon Eclipse MA 100. Fractographic investigation of fragmented specimens
was done using a Scanning Electron Microscope (SEM). Figure 3.13 shows the Joel
JSM 6010L V Scanning Electron Microscope used in this research study which is
located in Department of Mechanical Engineering at University of Saskatchewan. The
SEM was operated using accelerated voltage of 15KV and a secondary electron imaging
technique. X-ray photoemission electron microscope (XPEEM) with Near Edge X-Ray
Absorption Fine Structure (NEXAFS) spectroscopy was used to investigate the chemical
and electronics structure of material inside and outside of the adiabatic shear band for

45
AISI 4340 alloy steel using synchrotron light radiations at the Canadian Light Source
(CLS) in Saskatoon.

Figure 3.12. Photograph of the Nikon Eclipse MA 100 optical microscope used in this
study.

46
 Figure 3.13. Joel Scanning Electron Microscope (SEM).

3.2.5. Microhardness Measurements

Vickers microhardness tests were conducted on the material, inside and outside of the
shear bands for specimens with well-formed adiabatic shear bands. Figure 3.14 shows
the Vickers microhardness tester used in this research which is located in Mechanical
Engineering Department at University of Saskatchewan (Room ENG 2C25).

47
 Figure 3.14. Vickers microhardness tester.

Polished specimens were mounted on the Microhardness tester and indented using a
diamond indenter. The Vickers hardness number is obtained by the ratio of F/A, where F
is the force applied to the square-based pyramid diamond indenter, which was 500gf for
this work. A is the surface area of the resulting indentation in square micrometers, which
can be calculated by the following equation:

(3.10)

where, d is the average length of the indentation diagonal D1 and D2. The Vickers
hardness number was calculated using the following equation [59]:

48
 (3.11)

Figure 3.15 shows optical micrographs of Vickers microhardness test indentations

located inside and outside of the shear band of a quench-hardened AISI 1340 steel
specimen tempered at 425 C, and impacted at 39.1 kg.m/s.

(a) (b)
Indentations
Indentations

Figure 3.15. Optical micrographs showing Vickers hardness indentations on quench-

hardened AISI 1340, tempered at 425 C, impacted at 39.1 kg.m/s, (a) inside the shear
band, (b) outside the shear band.

49
 CHAPTER FOUR
RESULTS AND DISSCUSSION

Stress-strain curve results from the dynamic impact and torsional tests are outlined in
this chapter. The effects of strain rate and microstructure on the dynamic stress-strain
curves are discussed. Results of metallographic and microhardness tests of heat-treated
specimens before and after impact are presented in this chapter. The effects of strain
rate, microstructure, alloying elements, and geometry of the specimen on adiabatic shear
band behaviour of AISI 1340 and 4340 steel are discussed.

4.1. Stress-strain curve

4.1.1. Overview
Figure 4.1 shows a typical stress-strain curve obtained from high strain-rate loading of a
cylindrical steel specimen. After a clear yield point, which shows the onset of plastic
deformation, strain hardening as a result of plastic deformation and thermal softening
due to the heat generated inside the material, occur simultaneously [3]. At the beginning,
strain hardening dominates the plastic deformation until a maximum flow stress is
reached. After the maximum flow stress, thermal softening dominates the deformation
process. Hence, as the strain increases, the flow stress decreases. The sharp stress
collapse due to mechanical instability is a consequence of intense adiabatic heating
along a narrow path leading to strain localization. This point is the critical strain for the
occurrence of an adiabatic shear band ( crit) [3]. It should be noted that assumptions
which has been made to produce dynamic stress-strain curves may not be valid after the
onset of strain localization.

50
 1400
Stain hardening
1200
Adiabatic Heating

1000
True Stress (MPa)

Critical Strain
800
Yield point

600

400

200

0
0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0.16
True Strain

Figure 4.1. Typical stress-strain curve for specimen subjected to dynamic shock loading
[martensitic AISI 1340 steel, tempered at 425 C, cylindrical specimen and subjected to
impact loading at 33.2 kg.m/s ( =1146 s-1)].

4.1.2. The effect of impact momentum

The effects of impact momentum on the dynamic impact response of the martensitic
AISI 1340 steel specimens that were tempered at 315˚C are shown in Figs. 4.2 and 4.3.
The strain-rate generated in each specimen is directly correlated to the applied impact
momentum. As the impact momentum increases, the critical strain and time at which
thermo-mechanical instability occur increases. The increase in critical strain for
instability is due to increasing thermal softening and plastic strain in the specimens as
impact momentum increases. Also, as the impact momentum increases, the specimens
become more work-hardened before thermal softening begins to dominate the
deformation process. The increased plasticity with increase in strain-rates is due to an
increased thermal softening effect during the deformation process. This result agrees
with the submission of Kalthoff and Burgel [75] that the temperature inside shear bands

51
increased as the loading rate increased, leading to significant effects of material yielding
and plasticity. The same behaviour has been observed for the other microstructures
which were impacted at different impact momentums (Figs. 4.2 - 4.6). Lee and Lin [71]
suggested that as the strain-rate increases, the stress increases at any specific strain.
Also, as the strain rate increases, the rate of work hardening decreases due to increased
heat generation during plastic deformation. Increasing the strain rate has dual effects on
plastic deformation. On one hand it increases strain hardening because of the dislocation
multiplication. On the other hand, it increases thermal softening due to the heat
generation during plastic deformation.

It has also been observed that as the impact momentum or strain-rate increases,
maximum flow stress increases. However, for AISI 1340 steels specimens with
martensitic structure, as the impact momentum increases, the maximum flow stress
initially increases, but begins to decrease with increasing impact momentum at high
values of impact momentum. From Table 4.1, this impact momentum beyond which
maximum flow stress begins to decrease with increasing impact momentum is that
momentum above which transformed bands are formed, as shown in the experimental
data presented in Table 4.1. The temperature inside the shear band increases as the strain
rate increases. The impact momentums at which maximum flow stresses begin to
decrease increases with increasing tempering temperature. This indicates that
martensitic specimens that were tempered at a lower tempering temperature are more
inclined to form transformed bands at lower strain-rates. On the other hand, for steel
specimens with pearlitic structure, as the strain-rate increases, the maximum flow stress
increases and there is no evidence of decrease in maximum flow stress, within the range
of applied impact momentum (strain-rate). It can also been seen from Table 4.1 that no
transformed adiabatic shear band was observed in the specimen at highest applied
impact momentum.

52
 30.2 kg.m/s (811 /s)
1800 32.2 kg.m/s (892 /s)
1600 33.2 kg.m/s (1467 /s)
34.2 kg.m/s (1493 /s)
True Stress (MPa) 1400
36.1 kg.m/s (1811 /s)
1200
38.1 kg.m/s (2224 /s)
1000
800
600
400
200
0
0 0.05 0.1 0.15
True Strain

Figure 4.2. Dynamic impact stress-strain curves for martensitic AISI 1340 steel
specimens (tempered at 315˚C) as a function of the impact momentum and strain rates.

30.2 kg.m/s (811 /s)

32.2 kg.m/s (892 /s)
1800 33.2 kg.m/s (1467 /s)
1600 34.2 kg.m/s (1493 /s)
True Stress(MPa)

1400
36.1 kg.m/s (1811 /s)
1200
38.1 kg.m/s (2224 /s)
1000
800
600
400
200
0
0 100 200 300
Time (μs)

Figure 4.3. Stress vs. deformation time curves for martensitic AISI 1340 steel specimens
(tempered at 315˚C) as a function of the impact momentum and strain rates.

53
 2500 30.2 kg.m/s (468 /s)
32.2 kg.m/s (705 /s)

2000 33.2 kg.m/s (1146 /s)

True Stress (MPa) 34.2 kg.m/s (1396 /s)

1500 36.1 kg.m/s (1452 /s)

1000

500

0
0 0.02 0.04 0.06 0.08 0.1
True Strain

Figure 4.4. Dynamic impact stress-strain curves for martensitic AISI 1340 steel
specimens (tempered at 205˚C) as a function of the impact momentum and strain rates.

1600 30.2 kg.m/s (627 /s)

33.2 kg.m/s (1828 /s)
1400
34.2 kg.m/s (2354 /s)
True Stress (MPa)

1200
36.1 kg.m/s (2720 /s)
1000 38.1 kg.m/s (2952 /s)

800 39.1 kg.m/s (3405 /s)

600
400
200
0
0 0.05 0.1 0.15 0.2 0.25
True Strain

Figure 4.5. Dynamic impact stress-strain curves for martensitic AISI 1340 steel
specimens (tempered at 425˚C) as a function of the impact momentum and strain rates.

54
 1200

True Stress (MPa)

30.2 kg.m/s (1436 /s)
1000
32.2 kg.m/s (2047 /s)
800
34.2 kg.m/s (3348 /s)
600 36.1 kg.m/s (3728 /s)
400 38.1 kg.m/s (4769 /s)

200 39.1 kg.m/s (4868 /s)

42.1 kg.m/s (5469 /s)
0
44.1 kg.m/s (6550 /s)
0 0.2 0.4 0.6
49.1 kg.m/s (7889 /s)
True Strain

Figure 4.6. Dynamic impact stress-strain curves for pearlitic AISI 1340 steel specimens
as a function of the impact momentum and strain rates.

55
Table 4.1. Experimental data sheet for the impact test

a. AISI 1340 steel

Impact Impact Total Strain rate ASB
pressure Momentum Strain (s-1)
(kPa) (kg.m/s)
Martensitic Structure ( Temper at 205 C)
60 30.2 0.035 468 Slightly Deformed
80 32.2 0.052 705 Deformed
100 34.2 0.071 965 Slightly Transformed
100 34.2 0.102 1396 Slightly Transformed
120 36.1 0.105 1452 Transformed
140 38.1 Fractured Fractured Transformed
150 39.1 Fractured Fractured Transformed
Martensitic Structure (Temper at 315 C)
60 30.2 0.060 811 Deformed
80 32.2 0.066 892 Deformed
90 33.2 0.106 1467 Deformed
100 34.2 0.108 1493 Intense Deformed
120 36.1 0.129 1811 Deformed and
Transformed
140 38.1 0.156 2224 Transformed
150 39.1 0.176 2534 Transformed
180 42.1 Fractured Fractured Transformed
Martensitic Structure (Temper at 425 C)
60 30.2 0.047 627 None
80 32.2 0.146 2064 None
100 34.2 0.193 2354 Deformed
120 36.1 0.188 2720 Deformed
140 38.1 0.202 2952 Deformed
150 39.1 0.229 3405 Deformed and
Transformed
180 42.1 Fractured Fractured Transformed
Pearlitic Microstructure
60 30.2 0.010 1376 None
60 30.2 0.104 1436 None
80 32.2 0.145 2047 None
100 34.2 0.225 3348 Slightly Deformed
120 36.1 0.247 3728 Deformed
140 38.1 0.305 4769 Deformed
150 39.1 0.310 4868 Deformed
180 42.1 0.341 5469 Deformed
200 44.1 0.393 6550 Deformed
250 49.1 0.452 7889 Deformed

56
 Impact Impact Total Strain rate ASB
pressure Momentum Strain (s-1)
(kPa) (kg.m/s)
Dual-Phase Microstructure
60 30.2 0.057 776 None
80 32.2 0.099 1371 Deformed
100 34.2 0.116 1622 Intense Deformed
120 36.1 0.136 1910 Deformed
140 38.1 0.168 2409 Deformed
160 40.1 0.187 2722 Deformed and
Transformed
180 42.1 Fractured Fractured Transformed
200 44.1 Fractured Fractured Transformed

b. AISI 4340 steel

Martensitic Structure (Temper at 315 C)
100 34.2 0.120 1677 Intense Deformed
120 36.1 0.138 1943 Intense Deformed
140 38.1 Fractured Fractured Transformed
Pearlitic Microstructure
120 36.1 0.110 1528 None
140 38.1 0.144 1999 None
150 39.1 0.140 1872 Slightly Transformed
180 42.1 Fractured Fractured Transformed
Dual-Phase Microstructure
80 32.2 0.089 1228 Slightly Deformed
80 32.2 0.091 1819 Slightly Deformed
100 34.2 0.095 1310 Deformed
100 34.2 0.094 1293 Deformed
120 36.1 0.118 1646 Intense Deformed
120 36.1 0.136 1916 Intense Deformed
140 38.1 Fractured Fractured Transformed
160 40.1 Fractured Fractured Transformed

57
4.1.3. The effect of microstructure
Figure 4.7 shows the effect of microstructure on stress-strain curves for AISI 1340 steel
specimens that were impacted at the same impact momentum of 36.1 kg.m/s. The
corresponding strain-rates generated in each specimen are also indicated in this figure.
The total strains as well as critical strains for the onset of thermo-mechanical instability
are higher for the martensitic AISI 1340 steel specimens that were tempered at 425 C
compared to those that were tempered at 315 C and 205 C. Meanwhile, the AISI 1340
steel specimens with pearlitic structure have the highest total strain and critical strain for
thermo-mechanical instability. The steel with dual-phase structure has intermediate
properties between those with martensitic and pearlitic microstructures. As the
tempering temperature increases for martensitic steel specimens, the initial resistance to
plastic deformation under the impact loading decreases. The strain and the time at,
which thermal softening begins to dominate the deformation process increases, as shown
in Figs. 4.7 and 4.8. The critical strain and time for the onset of adiabatic heating and the
associated thermo-mechanical instability, which triggers shear strain localization, also
increases with increasing tempering temperature. This suggests that the susceptibility of
the martensitic specimens to the adiabatic shear banding will decrease with increasing
tempering temperature.

The amount of strain the specimens can withstand after yielding and before thermal
softening dominate the strain hardening tends to increase with increasing tempering
temperature. This suggests that increasing tempering temperature will result in lower
tendency for the occurrence of adiabatic shear band in quenched and tempered AISI
1340 steel specimens. Hwang et al. [58] investigated the effect of tempering on stress-
strain curves for low-carbon steel plate that was subjected to torsional loading at high
strain rate using Torsional Kolsky Bar. They also observed that as the annealing time
increased, fracture shear strain increased, while the maximum shear stress beyond which
thermal softening plays a major role decreased. As the annealing time increased the
width of the shear band increased.

Tempering temperature and time both have the same effect on the steel microstructure.
By increasing either time or temperature, the internal stresses inside the material are

58
relieved and the material become relaxed. Zhang et al. [8] suggested that tempering
temperature of impacted low alloy steel (40Cr and 40CrNiMo) specimens had a great
effect on the formation of adiabatic shear bands. They observed that at low to medium
tempering temperatures, transformed adiabatic shear bands were formed, while at higher
tempering temperatures, deformed adiabatic shear band were formed.

1800 Quench-hardened
Tempered at 205 C
1600
Quench-hardened
True Stress (MPa)

1400 Tempered at 315 C

1200 Quench-hardened
Tempered at 425 C
1000
Pearlitic Structure
800
600 Dual-Phase Structure
400
200
0
0 0.1 0.2 0.3
True Strain

Figure 4.7. Dynamic impact stress-strain curves for AISI 1340 steel as a function of the
heat-treatment condition (impacted at 36.1 kg.m/s).

59
 1800 Quench hardened, Temper
at 205˚C
1600
Quench hardened, Temper

True Stress(MPa)
1400 at 315˚C
1200 Quench hardened, Temper
1000 at 425˚C

800 Pearlite

600
Dual-Phase Structure
400
200
0
0 100 200 300 400
Time (μs)

Figure 4.8. Stress vs. deformation time curves for martensitic AISI 1340 steel specimens
(tempered at 315˚C), as a function of heat-treatment condition.

Since the AISI 1340 steel specimen with pearlitic structure has the highest ductility, it
experiences the greatest work hardening before thermal softening and has the highest
critical and total strain. The AISI 1340 steel specimens with dual-phase structure have
intermediate properties between the specimens with martensitic and pearlitic structure.
The dual-phase structure contains a soft ductile ferrite matrix that is reinforced with hard
and strong martensite. Therefore, it demonstrates a good mixture of ductility and
strength. Hence, the amount of strain hardening of specimens with dual-phase structure
before thermal softening and critical strain at which they show thermo-mechanical
instability and strain localization are significantly lower than for the steel specimens
with pearlitic structure. On the other hand, it reaches the point at which thermal
softening is the dominant deformation mechanism at a much lower stress level than that
for the martensitic structure. The same behaviour has been observed for the other
martensitic, pearlitic and dual-phase structure specimens impacted at different
momentums [Figs. 4.9- 4.13] or subjected to dynamic torsional loading [Fig. 4.14].
Figure 4.14 shows the effect of microstructure on stress-strain curves for AISI 1340
steel torsional specimens that were subjected to dynamic torsional loadings using an

60
applied torque of 1619 N.m. The maximum flow stress above which thermal softening
becomes dominant is highest in specimens with martensitic structure and lowest in that
with pearlitic structure. The results are in line with those obtained from the impact test.

1800
1600
True Stress (MPa)

1400
1200
1000 Quench-hardened
Tempered at 205 C
800
Quench-hardened
600 Tempered at 315 C
400 Pearlitic Structure

200
0
0 0.05 0.1 0.15
True Strain

Figure 4.9. Dynamic impact stress-strain curves for AISI 1340 steel as a function of the
heat-treatment condition (impacted at 32.2 kg.m/s).

2000
1800
1600
1400
True Stress (MPa)

1200 Quench-hardened
Tempered at 205 C
1000
Quench-hardened
800
Tempered at 315 C
600
Quench-hardened
400 Tempered at 425 C
200
0
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14
True Strain

Figure 4.10. Dynamic impact stress-strain curves for AISI 1340 steel as a function of the
heat-treatment condition (impacted at 33.2 kg.m/s).

61
 2500

2000

Quench-hardened
True Stress (MPa)

1500 Tempered at 205 C

Quench-hardened
Tempered at 315 C
1000
Quench-hardened
Tempered at 425 C
500 Pearlitic Structure

0
0 0.05 0.1 0.15 0.2 0.25
True Strain

Figure 4.11. Dynamic impact stress-strain curves for AISI 1340 steel as a function of the
heat-treatment condition (impacted at 34.2 kg.m/s).

1600
1400
True Stress (MPa)

1200
Quench-hardened Tempered
1000 at 315 C
800 Quench-hardened Tempered
at 425 C
600
Pearlitic Structure
400
Dual-Phase Structure
200
0
0 0.1 0.2 0.3
True Strain

Figure 4.12. Dynamic impact stress-strain curves for AISI 1340 steel as a function of the
heat-treatment condition (impacted at 38.1 kg.m/s).

62
 1400

1200

True Stress (MPa) 1000

800 Quench-hardened
Tempered at 425 C
600 Pearlitic Structure

400
Dual-Phase Structure
200

0
0 0.1 0.2 0.3 0.4
True Strain

Figure 4.13. Dynamic impact stress-strain curves for AISI 1340 steel as a function of the
heat-treatment condition (impacted at 39.1 kg.m/s).

600

500
True Stress (MPa)

400
Quech-hardened, Tempered
300 at 315˚C
Dual-Phase Structure
200
Pearlitic Structure
100

0
0 0.1 0.2 0.3 0.4
True Strain

Figure 4.14. Torsional stress-strain curves for AISI 1340 steel as a function of the heat-
treatment condition (1619 N.m torque).

63
4.1.4. The effect of alloying elements
Figures 4.15 and 4.16 show the effects of alloying elements on the stress-strain curves
for quench-hardened AISI 1340 and AISI 4340 steel specimens subjected to impact
momentums of 34.2 and 36.1 kg.m/s, respectively. At the higher impact momentum,
both martensitic AISI 1340 and AISI 4340 alloy steel specimens show the same general
behaviour (Fig. 4.16) although AISI 4340 steel had a higher strength under quasi-static
loading. For the lower impact momentum of 34.2 kg.m/s, it is observed that martensitic
AISI 1340 steel specimens undergo more strain hardening, before thermal softening
dominates, than martensitic AISI 4340 steel specimens. Thus, AISI 1340 steel
specimens have higher flow stress than AISI 4340 steel at this impact momentum (Fig.
4.15). This suggests that thermal softening has a more significant role in the impact
response of martensitic AISI 4340 steel than martensitic AISI 1340 steel. Therefore,
martensitic AISI 4340 steel specimens will be more inclined to form adiabatic shear
bands than martensitic AISI 1340 steel specimens.

Figs. 4.17 and 4.18 show the effects of alloying elements on stress-strain curves for
pearlitic AISI 1340 and AISI 4340 steel specimens. The AISI 4340 steel specimen has
significantly higher yield strength than the AISI 1340 steel specimen impacted at the
same momentum. This shows that pearlitic AISI 4340 steel specimens are more resistant
to plastic deformation at high strain-rate deformation than AISI 1340 steel specimens.
Moreover, pearlitic AISI 1340 steel specimens are significantly more hardened before
thermal softening than AISI 4340 alloy steel. In addition, the maximum stress above
which thermal softening is dominant is much higher for the AISI 4340 alloy steel than
for the AISI 1340 alloy steel. These observations shows that, although the stress-strain
curves of both alloys in the martensitic heat treatment condition are very close, a
significant difference is noticed in their high strain-rate response when heat treated to
form pearlitic structure.

Both AISI 1340 and AISI 4340 steel contain 0.4% C. However, AISI 4340 has
considerable amounts of alloying addition such as Mn, Cr, Ni and Mo, while AISI 1340
is manganese steel containing 0.14 wt% V. Mn, Cr, Ni and Mo in steels increase the
steel hardenability. Moreover, addition of vanadium, in the case of austenitizing at high

64
temperatures for a long time, provides secondary hardening during tempering [82].
Menzemera et al. [85] suggested that chromium, molybdenum, and manganese assist the
formation of carbides which consequently increase matrix strength. Meanwhile
Molybdenum helps refining the grain size. Moreover, Feng and Bassim [24] suggested
that local materials defects cause adiabatic shear band initiations which are called initial
perturbations. Bassim [34] suggested that any inhomogeneities inside the material
increases the materials’ susceptibility to form adiabatic shear bands. Consequently, AISI
4340 alloy steel with higher amount of overall Cr, Mo, Ni and Mn is more likely to form
carbides which can act as initial perturbations and promote adiabatic shear band
formation. These carbides also increase the internal stresses inside the material which
causes higher temperature rise within adiabatic shear bands during high strain-rate
deformation. Therefore, AISI 4340 steel specimens are more inclined to experience
more intensive adiabatic heating and form adiabatic shear bands than AISI 1340 steel
specimens.

1800
1600
True Stress (MPa)

1400
1200
1000
800 1340
600 4340
400
200
0
0.00 0.02 0.04 0.06 0.08 0.10 0.12
True Strain

Figure 4.15. Stress-strain curves for martensitic steels, tempered at 315˚C, cylindrical
specimens, impacted at 34.2 kg.m/s.

65
 1600
1400

True Stress (MPa)

1200
1000
800
1340
600
4340
400
200
0
0.00 0.05 0.10 0.15
True Strain

Figure 4.16. Stress-strain curves for martensitic steels, tempered at 315˚C, cylindrical
specimens, impacted at 36.1 kg.m/s.

2000
1800
True Stress (MPa)

1600
1400
1200
1000
800 1340

600 4340

400
200
0
0 0.1 0.2 0.3
True Strain

Figure 4.17. Stress-strain curves for steel pearlitic structure steels, cylindrical specimen,
impacted at 38.1 kg.m/s.

66
 2000
1800

True Stress (MPa)

1600
1400
1200
1000
800 1340
600 4340
400
200
0
0 0.1 0.2 0.3
True Strain

Figure 4.18. Stress-strain curves for pearlitic structure steels, cylindrical specimen,
impacted at 39.1 kg.m/s.

4.2.2. Metallographic analysis

4.2.2.1. Overview
Figure 4.19 shows the optical micrographs of the original AISI 1340 steel after the
various heat treatment conditions before the dynamic impact test. The quenched and
tempered steel specimens consist of martensitic plates and retained austenite (Fig.
4.19a). The white phase in the microstructure of the quenched and tempered steel is
untransformed austenite. The normalised steel specimen consists of pearlite and ferrite
(Fig. 4.19b). Phase analysis of the microstructure using Pax-it software shows that the
pearlitic specimen contains 11 vol. % pearlite and 89 vol. % ferrite. The steel subjected
to inter-critical annealing at 740 C before quenching shows the dual-phase
microstructure consisting of 20 vol. % ferrite (bright) and 80 vol. % martensite (dark
phase) as shown in Fig. 4.19c.

67
 Figure 4.19. Optical micrographs of AISI 1340 steel specimens before impact test: (a)
austenitized at 850 C, water- quenched and tempered at 315 C – martensitic structure,
(b) austenitized at 850 C and air-cooled – pearlitic structure, (c) austenitized at 740 C,
water-quenched and tempered at 315 C - dual-phase structure.

68
Figure 4.20 shows the optical micrographs of the original AISI 4340 steel after the
various heat treatment procedures before impact test. The quenched and tempered steel
consists of martensitic plates and retained austenite (Fig. 4.20a). The normalised steel
specimen consists of pearlite and ferrite (Fig. 4.20b). Phase analysis of the
microstructure using Pax-it software shows that the pearlitic specimen contains 35 vol.
% pearlite and 65 vol. % ferrite. The steel subjected to inter-critical annealing at 740 C
before quenching also shows martensitic structure (Fig. 4.20c). Unlike AISI 1340
manganese steel, a dual-phase microstructure could not be formed in AISI 4340 steel at
740 C. The significant amount of manganese in AISI 1340 steel widens the ferrite +
austenite region in the iron-iron carbide equilibrium phase diagram. Therefore, by
austenitizing AISI 1340 specimens inside this region at 740 C mixture of austenite and
ferrite are formed which are then converted to martensite + ferrite on quenching in
water. However, AISI 4340 steel does not have a sufficient amount of manganese,
which can expand ferrite-austenite region in the iron-iron carbide phase diagram.
Moreover, the presence of nickel could expand the austenite phase region [82] to 740 C.
Thereby, complete austenitization can occur at 740 C. This explains why quenching the
AISI 4340 steels from 740 C produced an entirely martensitic microstructure.

69
 Figure 4.20. Optical micrographs of AISI 4340 steel specimens before impact test: (a)
austenitized at 850 C, oil- quenched and tempered at 315 C – martensitic structure, (b)
austenitized at 850 C and air-cooled – pearlitic structure, (c) austenitized at 740 C, oil-
quenched and tempered at 315 C - martensitic structure.

70
4.2.2.2. The effect of impact momentum
During high velocity impact, a high proportion of the kinetic energy of the impacting
projectile is converted to heat inside the material. Since the deformation occurs very
quickly, this heat generated inside the material cannot be conducted away along a
narrow path inside the material. Therefore, the temperature of this localized area
increases, which causes intense plastic deformation, high shear strains and strain-rates
along this narrow path called the Adiabatic Shear Band (ASB). Figure 4.21 shows
optical micrographs of martensitic AISI 1340 steel (tempered at 315 C) impacted at
different impact momentums. At the impact momentum of 34.2 kg.m/s (Fig. 4.21a), the
strain rate inside the adiabatic shear band is just enough to slightly align the martensite
and carbide particles along the adiabatic shear band propagation path and form a light
deformed shear band observed in the specimen. As the impact momentum, and
consequently the strain rate, increases, the intense strain inside the shear band becomes
sufficiently high to break martensite plates and carbide laths into sub-micron size
particles, forming a transformed shear band along which cracking occurs, as shown in
Fig. 4.21d. Figure 4.22 shows the process of crack formation along a transformed
adiabatic shear band in a martensitic AISI 1340 steel specimen impacted at 42.1 kg.m/s.
As the impact momentum reaches 42.1 kg.m/s, the strain-rate and temperature rise
within the adiabatic shear band are high enough to generate tensile stress inside the
shear band and open up micro-voids, which cause crack to form along the adiabatic
shear band propagating path [78].

71
Figure 4.21. Optical micrographs of martensitic AISI 1340 steel, tempered at 315˚C,
impacted at: (a) 34.2 kg.m/s, (b) 36.1 kg.m/s, (c) 39.1 kg.m/s, (d) 42.1 kg.m/s.

72
Figure 4.22. Optical micrographs of fractured cylindrical specimen martensitic AISI
1340 steel, tempered at 315 C, impacted at 42.1 kg.m/s.

73
Figure 4.23 shows optical micrographs of pearlitic AISI 1340 steel specimens impacted
at different impact momentums. It is observed that only deformed adiabatic shear bands
form in steel specimens with pearlitic structure for all the impact momentums
investigated. However, the morphology of the deformed shear bands is somewhat
different from what has been observed in other steels, such as AISI 4340 steel, in which
shearing on the grains was inclined to specimen’s diameter [83]. It is observed in the
current study on AISI 1340 steel that the pearlite and ferrite are aligned radially along
the diameter of circular transverse cross-sectional area. As the impact momentum
increased, the thickness of deformed shear band increased, but a transformed shear band
was not formed. No fracture was observed for pearlitic specimens even at impact
momentums at which fracture occurred for martensitic and dual-phase structure
specimens. At high impact momentums, pearlite and ferrite particles are closely packed
together and form a thick deformed shear band. However, at a low impact momentum of
34.2 kg.m/s, there was no evidence of adiabatic shear band formation in the pearlitic
specimen.

SEM micrographs showing the microstructure of pearlitic AISI 1340 steel within the
regions inside and outside the adiabatic shear band are presented in Fig. 4.24. It is
obvious that pearlite and ferrite grains are smaller inside the deformed band than
outside. This is attributed to the high strain-intensity inside the shear band, which is high
enough to fragment the large pearlite and ferrite grains in this region.

74
 Figure 4.23. Optical micrographs showing deformed bands in the pearlitic AISI 1340
steel, Impacted at: (a) 34.2 kg.m/s, (b) 36.1 kg.m/s, (c) 39.1 kg.m/s, (d) 42.1 kg.m/s, (e)
44.1 kg.m/s, (f) 49.1 kg.m/s.

75
Figure 4.24. SEM micrographs of pearlitic AISI 1340 steel showing microstructure
inside and outside the shear band (impacted at 44.1 kg.m/s).

76
Figure 4.25 shows optical micrographs for AISI 1340 steel specimens with the dual-
phase structure impacted at different momentums. At 36.1 kg.m/s impact momentum, a
deform shear band is formed. As the impact momentum increased, an intense white
etching band, also called transformed adiabatic shear bands, formed. As the impact
momentum increased, shear strain became more intensely localised in the shear bands.
Finally, at an impact momentum of 42.1 kg.m/s, fracture occurred in the dual-phase
structure specimen along the transformed shear band propagation path. As mentioned
above, at a lower impact momentum, the strain rate is just enough to align martensite
and ferrite particles along the shear stress direction. As the impact momentum and
consequently the strain-rate increased, the intensity of shear strain inside the shear band
became intense and caused fragmentation of martensite laths and ferrite particles along
the shear band propagation path.

77
Figure 4.25. Optical micrographs for dual-phase AISI 1340 steel (tempered at 315˚C),
impacted at: (a) 36.1 kg.m/s, (b) 38.1 kg.m/s, (c) 39.1 kg.m/s, (d) 42.1 kg.m/s.

78
SEM micrographs showing the microstructure of the dual-phase structure AISI 1340
steel within regions inside and outside the adiabatic shear band are presented in Fig.
4.26. The shear band imaged here is not a fully formed transformed band that will
normally appear featureless under an optical (Fig. 4.25d) and SEM. It is observed that
materials inside the shear band have less martensite and more ferrite than outside. It is
not very clear at this stage why more ferrite is observed inside the shear band. However,
it has been suggested that the combined action of hydrostatic pressure on the shear band
and high temperature of the shear band can promote a carbon and iron diffusion path
that can lead to the formation of ferrite [42]. Wittman et al. [20] also reported carbon
diffusion aided by dislocation in AISI 4340 steel. Derep [67] observed that the
microstructure of adiabatic shear band in armour steel consists of fine grain equi-axed
structure of delta ferrite, mixed with narrow laths of martensite.

Figure 4.27 shows the optical micrographs for cracks in the dual-phase structure AISI
1340 steel specimens which fractured at 42.1 kg.m/s impact momentum. The crack
formed along the transformed adiabatic shear band. At 42.1 kg.m/s impact momentum,
the strain rate inside the adiabatic shear band of the dual-phase structure AISI 1340 steel
specimen is high enough to generate micro-voids inside the shear band. These micro-
voids interconnect and form cracks.

79
Figure 4.26. Optical and SEM micrographs of regions inside and outside the shear band
for dual-phase AISI 1340 steel, tempered at 315 C, impacted at 39.1 kg.m/s.

80
Figure 4.27. Optical micrographs showing adiabatic shear failure of dual-phase steel
(AISI 1340), impacted at 42.1 kg.m/s.

81
4.2.2.3. The effect of microstructure
Optical micrographs showing the effects of tempering temperature on plastic
deformation of martensitic AISI 1340 steel impacted at 34.1 kg.m/s are represented in
Fig. 4.28. A thin intense white etching band was formed in the specimen that was
tempered at 205 C, while a thicker light deformed band was formed in the specimen
which was tempered at 425 C. The shear band for the specimen that was tempered at
315 C had intermediate properties between the other two specimens. It is a more
diffused form of white etching band.

Figure 4.28. Optical micrographs of martensitic AISI 1340 steel, impacted at 34.1
kg.m/s, cylindrical specimens: (a) tempered at 205 C, (b) tempered at 315 C, (c)
tempered at 425 C.

82
It has been suggested by Rittel et al. [45], that the existence of defects such as abrasive
particles and precipitates or any inhomogeneities inside the microstructure of metallic
materials increases the susceptibility of materials to strain localization and formation of
an adiabatic shear band [45]. These defects and inhomogeneities within the
microstructure of materials are called initial perturbations. The internal stresses and
therefore perturbations are more pronounced in the specimen that was tempered at
205 C. This promotes the intense strain localization and formation of a transformed
adiabatic shear band. As the tempering temperature increases, the stress that may be
introduced during quenching is more relieved and the initial perturbations will decrease.
The uniform dispersion of precipitates, as well as relieved internal strains, decrease the
susceptibility of the material to strain localization. Therefore, the martensite laths are
just capable of aligning themselves and forming a wide deformed band.

The optical micrographs in Fig. 4.29 show the effect of microstructure, as a result of
different heat-treatment procedure, on the plastic deformation of AISI 1340 steel
impacted at 38.1 kg.m/s. Thin intense white etching band is observed for the specimen
with a martensitic structure, while a thick light deform band was formed for specimens
having a pearlitic structure. The specimens with dual-phase structure have intermediate
properties between those having martensitic and pearlitic structures [Fig. 4.29(c)]. A
clear deformed shear band with brighter color than the matrix has been observed for the
dual-phase structure specimen. Stresses that are introduced during quenching into
martensitic microstructure act as an initial perturbation and increase the susceptibility of
the material to shear strain localization. For the pearlitic microstructure specimen, which
was air-cooled to room temperature, no stress is introduced inside the material, which
reduces the susceptibility of the material to shear strain localization along the shear
band. Therefore, shear strain inside the shear band is just high enough to align the
pearlite and ferrite grains along the shear flow direction, causing formation of a
deformed adiabatic shear band. The dual-phase structure has intermediate properties
between transformed band observed in the martensitic specimen and deformed band in
the specimens with pearlitic structure. Dual-phase structure is a mixture of strong
martensite and ductile ferrite. The martensite with internal stresses, as a result of
quenching, increases the susceptibility of the material to shear strain localization. On the

83
other hand, ductile ferrite particles facilitate plastic deformation. Therefore, maximum
temperature rise inside the adiabatic shear band is less for the dual-phase steel than for
the martensitic steel.

Figure 4.29. Optical micrographs of AISI 1340 steel, impacted at 38.1 kg. m/s,
cylindrical specimen. (a) martensitic- tempered at 315 C, (b) Pearlitic, (c) Dual-phase
structure.

84
Figure 4.30 shows the optical micrographs of martensitic AISI 1340 and 4340 steels that
were tempered at 315 C and impacted at 34.2 kg.m/s. Figure 4.30 (a) shows the
micrograph for AISI 1340 alloy steel while Fig. 4.34 (b) is for AISI 4340 alloy steel.
Figure 4.31 shows optical micrographs for cylindrical specimens that were austenitized
at 740 C, quenched-hardened and then tempered at 315 C after impact at a momentum
of 36.1 kg.m/s. Figure 4.31a and 4.31b show the micrographs for AISI 1340 steel and
AISI 4340 steel, respectively. It is evident from both Figs. 4.30 and 4.31 that AISI 4340
steel is more inclined to form an adiabatic shear band than AISI 1340 steel when
subjected to the same heat treat treatment condition. The presence of additional alloying
elements such as Mo, Cr, Ni, and Mn increases the ability of material to form carbide
particles which can act as initial perturbations and increases the susceptibility of the
material to shear localization and the occurrence of adiabatic shear bands. Therefore,
AISI 4340 alloy steel is more inclined to form an adiabatic shear band than AISI 1340
alloy steel. The same results have been observed from stress-strain curves of AISI 1340
and AISI 4340 steels. It should be mentioned that at 740 C, AISI 4340 steel was not able
to form dual-phase structure entirely, as previously explained.

Figure 4.30. Optical micrographs of cylindrical specimen of martensitic steel, (tempered

at 315 C), impacted at 34.2 kg. m/s. (a)AISI 1340 steel, (b) AISI 4340 steel.

85
 Figure 4.31. Optical micrographs of steel specimens partially austenitized at 740 C,
quenched-hardened and tempered at 315 C - impacted at 36.1 kg. m/s: (a)AISI 1340
steel, (b) AISI 4340 steel.

4.2.2.4. The effect of geometry

Optical micrographs of the transverse cross-section of impacted martensitic AISI 1340
steel and AISI 4340 steel specimens showing the effects of specimen shape on the
geometry of adiabatic shear band on the transverse section of the specimen are presented
in Figs. 4.32 and 4.33, respectively. The specimens shown in these figures were
impacted at a momentum of 36.1 kg.m/s. It is observed that the geometry of adiabatic
shear band depends on the geometry of the specimen. The shear band in the cylindrical
specimen appears as a circle on the transverse cross-sectional area [Figs. 4.32 (c), and
4.33 (a)], while in cubic specimens, it appears as a square on the transverse cross-
sectional area [Figs. 4.32 (a), 4.32 (b), and 4.33 (b)]. The shear bands form a circular
path on the transverse section close to the diametrical surface with smaller radius in the
specimens having the shape of truncated cone [Fig. 4.33 (c)], while no shear band was
observed on the transverse section close to the base of the truncated cone with the larger
cross sectional area [Fig. 4.33(d)]. It should be noted that the conical specimen was
impacted on the diametrical surface with the smaller radius. In comparison to the
cylindrical martensitic specimen tempered at the same temperature and impacted at the
same momentum, the cubic samples and the region of the conical specimen with smaller

86
transverse cross sectional area have more intense shear localization. This is attributed to
the higher stress concentration within the sharp corners for the cubic specimen and the
uneven transverse cross sectional area in the conical sample, which could act as the
initial source of perturbation that promotes the occurrence of the intense white etching
bands. The optical macrographs showing the effects of specimen geometry on the
geometry of adiabatic shear bands along the longitudinal cross-section of impacted
specimens are presented in Fig. 4.34.

Figure 4.32. Optical micrographs of martensitic AISI 1340 steel, tempered at 315 C,
impacted at 36.1 kg.m/s: (a) cubic sample corner, (b) cubic sample close to the edge, (c)
cylinder.

87
 Figure 4.33. Optical micrographs of martensitic AISI 4340 steel, tempered at 315 C,
impacted at 34.2 kg.m/s: (a) cubic, (b) cylinder, (c) truncated cone(small cross-section),
(d) truncated cone(large cross-section).

88
 (a)

(b)

(c)

Figure 4.34. Optical macrographs for martensitic AISI 4340 steel, tempered at 315 C,
34.2 kg.m/s: (a) cylindrical, (b) conical, (c) cubical specimen.

89
From the optical micrographs of adiabatic shear bands on both transverse and
longitudinal sections of the steel, geometrical models of adiabatic shear bands in three
dimensions are proposed as presented in Fig. 4.35. For cylindrical specimens, shear
bands form two cones that are mirror images of one another. Two pyramidal shape shear
bands are formed in cubical specimens, which are also mirror images of one another.
For the truncated conical specimen, the adiabatic shear band appears as just one cone
protruding from the smaller transverse cross-sectional area of the specimen.

(a) (b) (c)

Figure 4.35. Schematic view of adiabatic shear band in (a) cylindrical, (b) cubical, (c)
conical specimens.

90
4.3. Fracture of impacted specimens
Figure 4.36 shows the optical micrograph of a crack along transformed adiabatic shear
bands for different specimens with different chemical composition, microstructure, and
geometry. It is observed that failure and fracture occur along the transformed band,
independent of chemical composition, microstructure, and geometry of the specimen. It
should be noted that for pearlitic structure specimens, which were impacted in the strain-
rate range of between 1376 s-1 to 7889 s-1, only deformed bands are observed. Therefore,
no transformed shear band and consequently no failure occur for pearlitic specimens in
this strain-rate range.
Figures 4.37 and 4.38 and show the optical and SEM micrographs of transformed
adiabatic shear bands in martensitic AISI 1340 steel specimen that was tempered at 315
C and was impacted at 42.1 kg.m/s. The existence of elliptical shape holes along the
crack are evidence of the micro-void formation along the transformed adiabatic shear
band. At this impact momentum, the stress flow difference inside and outside of the
white etching adiabatic shear band is high enough to produce tensile stress inside the
shear band [78]. This tensile stress opens up micro-pores along the white shear band.
These micro-pores connect to one another and form cracks along the transformed
adiabatic shear band [76].

91
Figure 4.36. Optical micrograph showing cracks along a white etching band: (a)
martensitic AISI 1340 steel, tempered at 425 C, cylindrical specimen, impacted at 42.1
kg.m/s, (b) martensitic AISI 1340 steel, tempered at 315 C, cylindrical specimen,
impacted at 42.1 kg.m/s, (c) dual-phase structure AISI 1340 steel, tempered at 315 C,
cylindrical specimen, impacted at 44.1 kg.m/s, (d) martensitic AISI 1340 steel, tempered
at 315 C, cubic specimen, longitudinal cross-sectional area, (e) martensitic AISI 4340
steel, tempered at 315 C, cylindrical specimen, impacted at 38.1 kg.m/s.

92
Although the microstructural features of a transformed adiabatic shear band can not be
resolved using optical microscope, SEM micrographs of the transformed adiabatic shear
band suggest fragmentation of martensite laths into submicron size particles, as shown
in Fig. 4.38. The development of cracks from voids nucleated by tensile stress generated
inside the shear band is also evident in the SEM micrograph shown in Fig. 4.38. At high
impact momentum, the strain-rate inside the shear band is high enough to break apart
martensite laths into very small carbide particles which cannot be detected by optical
microscope.

Figure 4.37. Optical micrograph of martensitic AISI 1340 steel, tempered at 315 C,
cylindrical specimen, impacted at 42.1 kg.m/s.

93
 Figure 4.38. SEM micrographs of martensitic AISI 1340 steel, tempered at 315 C,
cylindrical specimen, impacted at 42.1 kg.m/s.

Figure 4.39 shows SEM micrographs of the fracture surface for AISI 1340 steel
specimen with dual-phase structure that fractured under impact loading at a momentum
of 44.1 kg.m/s. Highly elongated dimples and knobby features are observed on the
fracture surface of the adiabatic shear band. Dimples are also observed outside the shear
band, which are thicker and have lower aspect ratio than the dimples inside the shear
bands. The dimples outside the shear band are shallow, unlike the very deep ones
observed inside the shear bands. The highly elongated dimples inside the shear band and
adjacent to the shear band region indicate a ductile shear fracture mode along the
transformed band and in the adjacent deformed band region. Xue et al. [77] studied

94
adiabatic shear bands in Ti–6Al–4V alloys and proposed a schematic plot for void
evolution inside the adiabatic shear bands, as presented in Fig. 2.5. In this schematic
view, the voids nucleate inside the shear band as result of tensile stress generated inside
the shear band. They grow until they reach the shear band boundary, where the material
is harder due to the lower temperature. They proposed that, in the absence of shear
stress, these voids grow in an elliptical shape and finally coalescence and generate
complete separation. However, in the presence of shear stress, these voids are elongated
and rotated [77]. These elongated and rotated dimples are observed as highly elongated
dimples on the fracture surface of adiabatic shear band. Odeshi and Bassim [11] also
observed dimples on the fracture surface of a high strength low alloy steel specimen that
fractured under impact loading. They observed ductile fracture outside the shear band
and reported that dimples were nucleated at the surface of the hard particles in the
microstructure of bulk material. As shown in Fig. 4.39, a knobby fracture mode was
observed on the fracture surface along the adiabatic shear bands, indicating melting of
steel in these regions. Therefore, it could be concluded that the temperature rise inside
the transformed adiabatic shear bands reach the melting temperature of the alloy in some
regions during the dynamic impact loading.

95
 Figure 4.39. SEM micrographs showing the fracture surface of AISI 1340 steel with
dual-phase structure (cylindrical specimen).

Lee et al. [70] observed dimples on the fracture surface of shear bands of inconel 690
superalloy. They also suggested that these intense dimples show ductile failure.
Elongated dimples have been observed in adiabatic bands of other tested materials such
as armour steel [31]. They also suggested that knobby features within adiabatic shear
bands indicate the melting of an alloy in these regions. Rogers [42] suggested that the
knobby features observed on the surface of an adiabatic shear band of a fractured sample
is due to frictional heat generated as a result of fracture surfaces rubbing against one
another, which causes thermal alteration. Cho et al. [13] observed both knobby features
and elongated dimples on the fracture surface of an AISI 1018 cold-rolled steel (CRS), a
structural steel (HY-100), and an AISI 4340 vacuum arc remelted (VAR) steel tempered
to either of two hardnesses RHC 44 or 55. They suggested that the elongated dimples

96
show a ductile shear fracture mode, while knobby features exhibit melting of the alloy in
these regions as a result of heat generation inside the adiabatic shear band as the two
fracture surfaces are rubbing against one another.

Both elongated dimples indicating ductile shear failure and a knobby fracture mode
were also observed in martensitic AISI 1340 steel that fragmented during impact,
irrespective of specimen geometry. For example, Fig 4.40 shows the fracture surface in
the shear band region in a fragmented cubical shaped specimen, showing the elongated
dimples and knobby features. SEM micrographs of a fractured conical specimen [Fig.
4.41] also show smaller dimples on the fracture surface of the bulk material than
dimples on the adiabatic shear band, indicating ductile fracture. Cleavage fracture can
also be observed on the fracture surface of the bulk material. Odeshi and Bassim [11]
observed both ductile and cleavage fracture on the fracture surface of the bulk material
for AISI 4340 steel specimens. They suggested that precipitates open up voids outside
the shear band and cause ductile fracture. Therefore, the features of fracture surface of
specimens are independent of chemical composition and geometry of the impacted steel
specimens. The only difference is in the aspect ratios of the dimples, which are
influenced by the extent of thermal softening or degree of plasticity inside the shear
band and adjoining regions.

97
 Figure 4.40. SEM micrographs showing the fracture surface of martensitic AISI 1340
steel (cubic specimen), impacted at 38.1 kg.m/s, (a) elongated dimples, (b) knobby
features.

Figure 4.41. SEM micrographs showing fracture surface of martensitic AISI 1340 steel,
(conical specimen) impacted at 36.1 kg.m/s.

98
Figure 4.42 shows the SEM micrographs of the fractured surface of torsional specimens
of AISI 4340 steel. Only elongated dimples are observed on the fracture surface. This
shows that the temperature rise within the torsional specimen due to adiabatic heating
was not high enough to cause melting of the alloy at any point during the high strain-rate
torsional loading of the alloy. Evidence of cleavage fracture was also observed on the
fracture surface of the torsion specimen, suggesting a combination of ductile shear and
cleavage fracture in the alloy under high strain-rate torsional loading.

Figure 4.42. SEM micrographs showing fracture surface of martensitic AISI 4340 steel,
under high strain-rate torsion loading.

Figure 4.43 shows the optical micrograph of a longitudinal section of a martensitic AISI
1340 cubic specimen that fragmented under impact loading. The specimen was
sectioned perpendicularly across the fracture surface. Three separate regions can be
recognised under the fracture surface: white etching adiabatic shear band (transformed
adiabatic shear band) on the fracture surface, deformed band showing alignment of
martensite plates and laths along shear direction below the white band layer and the bulk

99
materials outside the shear band region. The high intensity of shear strain inside the
white etching adiabatic shear band caused the martensite to fragment into submicron
size grains which could not be detected by optical microscopy (Region1).

The thermal softening and shear stress in the second region cause the alignment of
martensite plates in shear flow direction indicated by the two arrows in Fig. 4.43 just
below the transformed band on the fracture surface. In region 3 which comprises the
region outside the shear band, i.e. the bulk material, the material just underwent normal
strain hardening and there was no change in microstructure compared to that of the
specimen before impact. This shows that the shear band region consists of a transformed
band surrounded by a deformed band having a lamellar structure.

Figure 4.43. Optical micrograph showing longitudinal section of fracture cubical

specimen of martensitic AISI 1340 steel.

Li et al. [43] also observed fine equi-axed subgrains with an average diameter of 150
nm, as well as low density of dislocations at the centre of the adiabatic shear band as a
result of dynamic recrystallization They also observed the lamellae structure with
elongated grains in a transition region between the centre of the shear band and the
matrix. TEM investigation of dynamic deformation in 8090 Al–Li alloy by Xu et al.
[33] showed that the material outside the adiabatic shear band consists of deformed cells

100
with a high density of dislocations, while the deformed region between the matrix and
shear band consists of highly elongated cells with well defined boundaries. On the other
hand, the structure inside the shear band consists of very fine equi-axed subgrains with a
low density of dislocations with a size of about 0.2 μm. They suggested that these equi-
axed subgrains are formed as a result of dynamic recrystallization.

Figures 4.44 - 4.46 show the optical macrographs of fractured cylindrical, cubical, and
conical AISI 4340 steel specimens. These pictures were taken using a stereomicroscope.
Based on the geometry of these fractured specimens and the three dimensional model
for geometry of the adiabatic shear band in the various impacted specimen shown earlier
(Fig. 4.35) , a schematic representation of crack propagation leading to fragmentation is
proposed as shown in Fig. 4.47. Cracks initiate inside the shear bands and propagate
into the bulk materials and connect with the second conical or pyramidal shape adiabatic
shear band for cylindrical and cubical shaped specimens, respectively. The specimens
eventually fragment into two symmetrical parts, as shown in Fig. 4.44 and 4.45. For the
conical specimen, the crack initiates from the corner of the cone formed on the smaller
side of the cone. It propagates along the adiabatic shear band to the lower point of the
cone and then through bulk material in the shear band free zone into the other end of the
truncated cone with bigger diameter, as shown in Fig. 4.46.

(a) (b)

Figure 4.44. Optical macrographs of fractured cylindrical specimen of martensitic AISI

4340 steel.

101
 (a) (a) (b)

Figure 4.45. Optical macrographs of fractured cubical specimen of martensitic AISI

4340 steel.

(a) (b)

Figure 4.46. Optical macrographs of fractured conical specimen of martensitic AISI

4340 alloy steel.

102
 Figure 4.47. Schematic view of crack propagation path, (a) cylindrical, (b) cubical, (c)
conical Specimens.

4.4. X-ray photoemission electron microscopy (XPEEM)

The results of the investigation of the transformed band in AISI 4340 steel using X-ray
photoemission electron microscopy (XPEEM) and Near Edge X-Ray Absorption Fine
Structure (NEXAFS) spectroscopy (Fig. 4.48-4.50) suggest redistribution of alloying
elements promoted by the intense adiabatic heating and large strain in the shear band
region during impact. Whereas more Nickel was observed inside the shear bands than
outside, less chromium was observed inside the shear bands than in the adjacent region.
Figure 4.49 shows the PEEM images of the adiabatic shear band in an AISI 4340 steel
specimen while Fig. 4.50 shows the Ni map inside and outside of the adiabatic shear
band in the specimen. It is evident that there is more nickel inside the shear band than
outside. During high strain rate deformation, the temperature rises inside the material.
This temperature rise inside the material depends on the chemical composition,
microstructure, and strain rate. Chen et al. [62] suggested that, depending on the
thickness of a target material, the temperature can rise up to 1527 C inside the adiabatic
shear band of the target material. This high temperature inside the shear band, coupled
with high strain, may promote the atomic diffusion of Ni and Cr inside the shear band.
The reason for the observed direction of atom migration across the shear band boundary

103
is not very clear at the moment. The XPEEM and NEXAFS were conducted at the Soft
X-ray Spectromicroscopy (SM) beamline at the Canadian Light Source (CLS) using
synchrotron light radiations.

Figure 4.48. (a) Cr 2p-3d NEXAFS spectra from inside and outside the shear band, (b)
Ni 2p-3d NEXAFS spectra from inside and outside the shear band.

Figure 4.49. PEEM imaging using Hg lamp, 150μm FOV.

104
Figure 4.50. XPEEM Image showing nickel map using synchrotron light radiation.

105
4.3. Microhardness results
4.3.1. Hardness values across the shear band
Figure 4.51 shows the Vickers hardness values across the shear band of a martensitic
AISI 1340 alloy steel specimen that was tempered at 315 C and then impacted at the
impact momentum of 39.1 kg.m/s. The error bars indicate confidence intervals with the
confidence level of 90%. The hardness value increases from the edge of the specimen
and reaches the maximum value at the centre of the shear band. Then it decreases and
become relatively constant. The higher hardness inside shear bands can be attributed to
the change in microstructure inside the shear band, especially grain refinement and
increased dislocation density as outlined in some previous studies [13, 26, 42, 67, 70].

Increased dislocation density inside the shear bands can be one of the contributing
factors to the higher hardness of shear bands compared with the rest of the material. Lee
et al. [70] observed dislocation and twinning deformation inside the shear band. Their
TEM investigation exhibited the nucleation of ά martensite and existence of a high
proportion of dislocations inside the shear band of the deformed 304L stainless steel.
Dislocations are arranged in a form of dislocation tangle which causes dislocation cell
wall formation. Meyers and Wittman [26] suggested that higher hardness of ASB in
low-carbon steels (AISI 1018 and 8620) is due to the fine microstructure with
supersaturated carbon and not a phase transformation. Chen et al. [14] suggested that
carbide lamellae spheroidization occurs inside the adiabatic shear band as a result of the
fragmentation of carbide laths. Moreover, very fine ferrite subgrains form inside the
shear band due to the iron diffusion path under hydrostatic pressure and high
temperatures [14]. Rogers [42] observed that the adiabatic shear band consists of
submicron size grains with the average size of tens of micrometers. TEM investigation
of the white etching band in a TC4 alloy showed the existence of equi-axed grains with
the diameter of 50 nm inside the shear band [44]. Li et al. [43] observed fine equi-axed
subgrains with the average diameter of 150 nm, in the adiabatic shear band for Monel
alloy. Meyers and Pak [30] observed small grains with the average size of 0.3-0.5 μm
inside the shear band of titanium. There have been also fine equi-axed submicron size
grains inside the shear band of tantalum [15]. For Copper samples that were deformed

106
under dynamic shock loadings, the submicron size grains with the average size of 0.1μm
were observed inside the adiabatic shear band [35, 36]. Therefore finer grain size as well
as formation of microstructure containing supersaturated carbon account for the
observed higher hardness values inside the shear band of the investigated steel.

0.12±0.03 mm
530

520

510
Hardness (VHN)

500

490

480

470

460
0 0.2 0.4 0.6 0.8 1 1.2 1.4
Distance from Specimen edge (mm)

Figure 4.51. Vickers hardness value across adiabatic shear band of martensitic AISI
1340 steel, tempered at 315 C, cylindrical specimen, impacted at 39.1 kg.m/s.

107
4.3.2. The effect of microstructure on hardness values
The Vickers hardness values for the original heat treated specimens before the impact
test are shown in Fig. 4.52. The error bars show confidence intervals with the
confidence level of 90%. For martensitic specimens, as the tempering temperature
increases, more of the stress which may be introduced inside the steel during rapid
cooling is relieved. Therefore, hardness value of the pre-impact martensitic specimen
which was tempered at 205 C is 4.5 % higher than that for pre-impact martensitic
specimen which was tempered at 315 C. The hardness values of martensitic specimens
and the dual-phase structure specimen are close to each other. Both were tempered at the
same temperature and the one with the dual-phase structure contains a low fraction of
ferrite (20 % vol.). Therefore, the probability that the micro indentation will be located
on the martensitic structure during the hardness test is about four out of five. This
explains why the hardness of the AISI 1340 steel heat-treated to have dual-phase
structure has the same range of hardness with those that have martensitic structure. The
steel specimens with pearlitic structure have the lowest hardness. Pre-impact hardness
value of the pearlitic specimen is approximately half of that for martensitic specimen
tempered at 315 C. The martensitic structure is formed when the austenitized steel
specimens are rapidly quenched to room temperature. The cooling rate is high enough to
prevent carbon diffusion. These carbon atoms stay in martensitic structure as interstitial
impurities, which effectively hinder dislocation motions. Moreover, martensitic
structure has body-centered tetragonal (BCT) structure. BCT structure has relatively low
slip systems along which dislocations can move [89].

108
 500

400
Hardness (VHN)

300

200

100

0
A B C D

Figure 4.52. Vickers hardness values for AISI 1340 steel before impact loading, (A)
martensitic, tempered at 205 C (B) martensitic, tempered at 315 C, (C) dual-phase
structure, tempered at 315 C, (D) pearlitic structure.

Figure 4.53 shows the results of hardness measurement inside and outside the shear
bands for various AISI 1340 steel specimens (with different microstructures) impacted
at a momentum of 36.1 kg.m/s. The error bars are showing confidence intervals with the
confidence level of 90%. The hardness value inside the shear band of the martensitic
specimen tempered at 205 C is 28% higher than that for the pre-impact specimen, while
the hardness value inside the shear band of the martensitic specimen tempered at 315 C
is 17% higher than that for pre-impact specimen. The higher amount of initial
perturbation for the specimen tempered at 205 C motivates higher a intensity of shear
strain localization in this specimen, which can lead to a higher density of dislocations
and finer microstructure inside the shear band. The hardness value outside of the shear
band for the specimen which was tempered at 205 C is 18% higher than the pre-impact

109
hardness value for the same specimen, while the hardness value outside of the specimen
that was tempered at 315 C is 16% higher than the pre-impact hardness value. Materials
outside the shear band undergo strain hardening as a result of high strain-rate
deformation. Therefore, hardness outside the shear band is greater than hardness values
for the pre-impact specimens.

For the pearlitic steel specimens, the hardness value inside the shear band is 34% higher
than that for the pre-impact specimen. The hardness value outside the shear band of the
pearlitic specimen is 25% higher than that for the pre-impact specimen. This
demonstrated that materials inside and outside the shear band undergo significant
amounts of plastic deformation. The SEM micrographs of the impacted pearlitic
specimen show that pearlite and ferrite grains are smaller inside the shear band than
outside. Therefore, a higher hardness value inside the shear band than outside is due to
the finer nature of pearlite and ferrite grains inside the shear band and to higher strain
hardening inside the bands.

700
Inside ASB
Outside ASB
Pre-impact
600
Hardness (VHN)

500

400

300

200

A B C D

Figure 4.53. Vickers Hardness values for AISI 1340 steel, Impacted at 36.1 kg.m/s, (A)
martensitic, tempered at 205 C, (B) martensitic, tempered at 315C, (C) dual-phase
structure-tempered at 315 C, (D) pearlitic structure.

110
4.3.3. The effect of impact momentum on hardness value
Figure 4.54 shows the effect of an increase in impact momentum on the hardness
measurements inside and outside shear bands in martensitic AISI 1340 steel specimens
that were tempered at 315 C. The error bars show confidence intervals with the
confidence level of 90%. As the impact momentum was increased from 34.2 kg.m/s to
39.1 kg.m/s, the hardness values both inside and outside of the shear bands also
increased. However, while only a marginal increase of 2% in the hardness was recorded
for the region outside the shear band, the hardness increase for the shear band region
was higher (5%). The hardness values inside and outside of the shear band for the
martensitic specimen that was impacted at 34.2 kg.m/s are respectively 15% and 4%
higher than the original hardness of the specimens before impact test. On the other hand,
the hardness value inside the shear band for the martensitic specimen that was impacted
at 39.1 kg.m/s was 19% higher than that for pre-impact specimens, while the hardness
value outside the shear band is 6% higher than that for the pre-impact specimen. As the
impact momentum increases, the material outside the shear band undergoes an increased
amount of work hardening. As the impact momentum increases, the amount of heat
generated inside the shear band increases. Therefore, the temperature rise within the
adiabatic shear band increases as the impact momentum increases. This will translate
into a more intense thermal softening and localised deformation at higher impact
momentum.

111
 600 Inside ASB
Ouside ASB
Pre-impact
550
Hardness (VHN)

500

450

400

350

300
A B
IM=34.2 kg.m/s IM=39.1 kg.m/s

Figure 4.54. Vickers hardness values for martensitic AISI 1340 steel as a function of
impact momentum (IM).

112
 CHAPTER FIVE
CONCLUSIONS AND RECOMMENDATIONS

5.1. Conclusions
The effects of microstructure, strain rates and geometry on plastic deformation and
failure of AISI 4340 and 1340 steel under high strain-rate loading both in torsion and in
compression were investigated. The results of this study show that:

1. Failure of both alloys under dynamic loading in torsion and compression was
initiated by intense shear strain localization along adiabatic shear bands. Both
deformed and transformed bands were observed in the alloys depending on the
microstructure and impact momentum or strain rate.

2. Formation of deformed and transformed adiabatic shear bands occurred in different

stages of deformation. A deformed band formed at low strain rates when the
localized shear strains were only high enough to align grains along the shear flow
direction. As the strain rate increased, transformed adiabatic shear bands formed
which contained fine equi-axed grains of submicron size.

3. The specimens with a martensitic structure were more inclined to form transformed
shear bands that those having a pearlitic structure. The higher the tempering
temperature of the martensitic steel, the lower its tendency to form an adiabatic
shear band under dynamic shock loading.
4. AISI 4340 steel was more susceptible to adiabatic shear failure than AISI 1340 steel
and there was a noticeable correlation between the dynamic stress-strain curves of
both alloys and the microstructural evolution associated with adiabatic shear
banding in the alloys.
5. The trajectory of the shear bands along the transverse section of impacted
cylindrical-, cubical-, and conical-shaped specimens followed the shape of the

113
 cross-section. While circular-shape shear bands were observed in cylindrical
specimens, the shear bands in cubic-shape specimens formed a rectangular path.
The impacted conical-shaped specimens showed circular-shape shear bands on the
smaller transverse-cross sectional area, and no evidence of a shear band on the
larger transverse cross-sectional area at the base of the cone.

6. The overall shape of the adiabatic shear bands in cylindrical-shape specimens was
two cones which were mirror images of one another, while in cubic-shape
specimens, two pyramids that were also mirror images of one another formed. The
shear band in conical-shaped specimens was a cone facing the smaller diametrical
surface of the specimen.

7. The alloys fragmented by ductile shear failure along the adiabatic shear bands.
Fractographic investigations of the fragmented specimens showed both ductile
shear and knobby fracture modes along the transformed shear band region. The
presence of a knobby fracture mode indicates that the temperature rise inside the
shear band during impact was close to the melting point of the steel inside the
transformed band. All the fracture surfaces of fragmented cylindrical-, cubical-, and
conical-shaped specimens showed both ductile shear failure and knobby fracture
modes.

8. The deformed band adjacent to the transformed band also failed by ductile shear
failure. The elongated dimples in the transformed band region were much deeper
than those in the adjacent deformed band. This confirms a higher level of plasticity
in the transformed band.

9. Hardness values of the transformed adiabatic shear bands were higher than that of
the bulk materials outside the shear band region due to the extremely fine structure
of these bands. The proportion of hardness increase inside the shear band over the
hardness of the bulk materials outside the shear band depended on the pre-impact
microstructure.
10. X-ray Photoemission electron microscopic (XPEEM) investigation of the
transformed band in AISI 4340 steel suggests the occurrence of strain-enhanced

114
 atomic diffusion inside the bands as a result in the intense localised adiabatic
heating and high strain intensity. More nickel and less chromium were recorded
inside the transformed shear band than in the region outside the transformed band.

5.2. Limitations and strength of thesis

To study the effects of microstructure, AISI 1340 and AISI 4340 cylindrical and thin-
walled tubular steel specimens were heat treated to provide martensitic, pearlitic, and
dual-phase structures. The number of specimens in each category was limited. In order
to investigate the effect of strain rate on the behaviour of AISI 1340 and 4340,
specimens have to be subjected to different strain rates. Therefore, each test with
specific conditions was conducted once, which is a limitation on the stress strain values
reported. However, at this stage of the research, the primary interest was in the general
trends of the dynamic stress-strain curves as affected by impact momentums and
microstructure.

The transformed band observed in AISI 1340 steel could not be investigated using X-ray
Photoemission Electron Microscopy (XPEEM) and Near Edge X-ray Absorption Fine
structure (NEXAFS) spectroscopy because of a prolonged equipment breakdown. At the
time of writing this thesis, the PEEM beamline is still not fully operational at the CLS.

The effects of geometry of the specimen on the morphology and properties of the
adiabatic shear band has been also studied in this research for AISI 1340 and 4340 steel,
which is a new subject of study. There has been no such previous work on chemical
analysis of the transformed band in these alloys. The chemical analysis of the material
inside and outside the shear band using X-ray Photoemission Electron Microscope
(XPEEM) and Near Edge X-ray Absorption Fine structure (NEXAFS) spectroscopy is a
new approach to understanding the microstructure evolution associated with the
occurrence of transformed bands in steel. The new results of XPEEM and NEXAFS
were presented at the 7th International workshop in LEEM-PEEM on August 8-13, 2010
in New York, and also at Pipeline Materials Reliability Workshop for a NSERC
Strategic Network on April 8, 2011 in Calgary.

115
5.3. Recommendation for future works
The recommendations for future works are as follows:

1. Transmission Electron Microscopy (TEM) investigations to provide more detailed

information on the evolution of microstructure in transformed bands formed in the
two alloys as a function of the investigated heat treatment condition,

2. X-ray Photoemission microscopic (XPEEM) investigation of transformed band in

AISI 1340 steel specimen, and

3. In-situ investigation of composition changes inside adiabatic shear band with

increasing temperature using synchrotron light radiation (XPEEM, and NEXAFS)
at the Canadian Light Source.

116
REFERENCES

[1] A. K. Zurek, ―The Study of Adiabatic Shear Band Instability in a Pearlitic 4340 Steel
Using a Dynamic Punch Test‖, Metallurgical and Materials Transactions 25A
(1994) 2483-2489.

[2] S.E. Schoenfeld, T.W. Wright, ―A failure criterion based on material instability‖,
International Journal of Solids and Structures 40 (2003) 3021–3037.

[3] C. Zener, J. H. Hollomon, ―Effect of Strain Rate Upon Plastic Flow of Steel‖,
Journal of Applied Physics 15 (1944) 22-32.

[4] M. N. Raftenberg, C. D. Krause, ―Metallographic observation of Armor Steel

specimens from plates perforated by shaped charge jets‖, International Journal of
Impact Engineering 23 (1999) 757-770.

[5] X.W. Chena, Q.M. Lib, S.C. Fan, ―Initiation of adiabatic shear failure in a clamped
circular plate struck by a blunt projectile‖, International Journal of Impact
Engineering 31 (2005) 877–893.

[6] V. F. Nesterenko, M. A. Meyers, T. W. Wright, ―Self-Organization in the initiation

of adiabatic shear bands‖, Acta Materialia 46 (1998) 327-340.

[7] S. N. Medyanik, W. K. Liu, S. Li, ―On criteria for dynamic adiabatic shear band
propagation‖, Journal of the Mechanics and Physics of Solids 55 (2007) 1439–1461.

[8] B. Zhang, W. Shen, Y. Liu, R. Zhang, ―Some factors influencing adiabatic shear
banding in impact wear‖, Wear 214 (1981) 259—263.

[9] L. Daridon, O. Oussouaddi, S. Ahzi, ―Influence of the material constitutive models

on the adiabatic shear band spacing. MTS, power law and Johnson–Cook models‖,
International Journal of Solids and Structures 41 (2004) 3109–3124.

[10] M.R. Staker, ―Relation between adiabatic shear instability strain and material
properties‖, Acta Metallurgica 29 (1981) 683–689.

117
[11] A.G. Odeshi, M.N. Bassim, ―High strain-rate fracture and failure of a high strength
low alloy steel in compression‖, Materials Science and Engineering A525 (2009)
96–101.

[12] J. Duffy, Y.C.Chi, ―On the measurement of local strain and temperature during the
formation of adiabatic shear bands‖, Materials Science and Engineering A 157
(1992) 195–210.

[13] K.M. Cho, S. Lee, S.R. Nutt, J. Duffy, ―Adiabatic shear band formation during
dynamic torsional deformation of an HY-100 steel‖, Acta Metallurgica et
Materialia 41 (1993) 923–932.

[14] Z.H. Chen, L.C. Chan, T.C. Lee, C.Y. Tang, ―An investigation on the formation
and propagation of shear band in fine-blanking process‖, Journal of Materials
Processing Technology 138 (2003) 610–614.

[15] M.A. Meyers, Y.J. Chen, F.D.S. Marquis, D.S. Kim, ―High-Strain-Rate Behavior
of Tantalum‖, Metallurgical Transactions 26A (1995) 2493- 2509.

[16] A. Hartley, J. Duffy, R.H. Hawley, ―Measurement of the temperature profile during
shear band formation in steels deforming at high strain rates‖, Journal of
Mechanics and Physics of Solids 35 (1987) 283-301.

[17] C.O. Mgbokwere, S.R. Nutt, J. Duffy, ―Shear band formation in 4340 steel: a TEM
study‖, Mechanics of Materials 17 (1994) 97–110.

[18] M. Zhou, A. J. Rosakis, G. Ravichandran, ―Dynamically Propagating Shear Bands

in Impact-Loaded Pre-notched Plates I. Experimental Investigations of
Temperature Signatures and Propagation Speed”, Journal of the Mechanics and
Physics of Solids 44 (1996) 981-1006.

[19] B.K. Kad, J.M. Gebert, M.T. Perez-Prado, M.E. Kassner, M.A. Meyers, ―Ultrafine-
grain-sized zirconium by dynamic deformation‖, Acta Materialia 54 (2006) 4111–
4127.

118
[20] C.L. Wittman, M.A. Meyeres, H.R. Pak, ―Observation of an Adiabatic Shear
Band in AISI 4340 Steel by High-Voltage Transmission Electron
Microscopy‖, Metallurgical Transactions 21A (1990) 707-716.

[21] A.G. Odeshi, M.N. Bassim, S. Al-Ameeri, ―Effect of heat treatment on adiabatic
shear bands in a high-strength low alloy steel‖, Materials Science and Engineering
A 419 (2006) 69–75.

[22] T.C. Lee, L.C. Chan, B.J. Wu, ―Straining behaviour in blanking process—fine
blanking vs. conventional blanking‖, Journal of Materials Processing and
Technology 48 (1995) 105–111.

[23] Q. Xue, G.T. Gray, ―Development of adiabatic shear bands in annealed 316L
stainless steel. Part II. TEM studies of the evolution of microstructure during
deformation localization‖, Metallurgy and Materials Transaction 37A (2006)
2447–2458.

[24] H. Feng, M.N. Bassim , ―Finite element modeling of the formation of adiabatic
shear bands in AISI 4340 steel‖, Materials Science and Engineering A266 (1999)
255–260.

[25] K.C. Dao, D.A. Schockey, ―A method for measuring shear‐band temperatures‖,
Journal of Applied Physics 50 (1979) 8244-8246.

[26] M.A. Meyers, C.L. Wittman, ―Effect of metallurgical parameters on shear band
formation in low-carbon (approx. 0.20 wt pct) steels‖, Metallurgical Transactions
21A (1990), 3153–3164.

[27] S.P. Timothy, ―The structure of adiabatic shear bands in metals: A critical review‖,
Acta Metallurgica 35 (1987) 301-306.

[28] J.E. Balzer, W.G. Proud, S.M. Walley, J.E. Field, ―High-speed photographic study
of the drop-weight impact response of RDX/DOS mixtures‖, Combustion and
Flame 135 (2003) 547–555.

119
[29] R.C. Glenn, W.C. Leslie, ―The Nature of "White Streaks" in Impacted Steel Armor
Plate‖, Metallurgical Transactions 2 (1971) 2945–2947.

[30] M.A. Meyers, H.R. Pak, ―Observation of an adiabatic shear band in titanium by
high-voltage transmission electron microscopy‖, Acta Metallurgica 34 (1986)
2493-2499.

[31] M.N. Raftenberg, C.D. Krause, ―Metallographic observations of armor steel

specimens from plates perforated by shaped charge jets‖, International Journal of
Impact Engineering 23, (1999) 757–770.

[32] J. Barry, G. Byrne, ―TEM study on the surface white layer in two turned hardened
steels‖, Materials Science and Engineering A 325 (2002), pp. 356–364

[33] Y.B. Xu , W.L. Zhong , Y.J. Chen , L.T. Shen Q. Liu , Y.L. Bai , M.A. Meyers,
―Shear localization and recrystallization in dynamic deformation of 8090 Al–Li
alloy‖, Materials Science and Engineering A299 (2001) 287–295.

[34] M. N. Bassim, ―Study of the formation of adiabatic shear bands in steels‖, Journal
of Materials Processing Technology 119 (2001) 234-234.

[35] J.A. Hines, K.S. Vecchio, ―Recrystallization kinetics within adiabatic shear bands‖,
Acta Materialia 45 (1997) 635–649.

[36] U. Andradea, M.A. Meyersa, K.S. Vecchioa, A.H. Chokshia, ―Dynamic

recrystallization in high-strain, high-strain-rate plastic deformation of copper‖,
Acta Metallurgica et Materialia 42 (1994) 3183-3195.

[37] A. Marchand, J. Duffy, ―An experimental study of the formation process of

adiabatic shear bands in a structural steel‖, Journal of the Mechanics and Physics
of Solids 36 (1988) 251–283.

[38] V.F. Nesterenko, M.A. Meyers, J.C. LaSalvia, M.P. Bondar, Y.J. Chen, Y.L.
Lukyanov, ―Shear localization and recrystallization in high-strain, high-strain-rate
deformation of tantalum‖, Materials Science and Engineering A229 (1997) 23-41.

120
[39] S. Nemat-Nasser, W. Guo, ―Flow stress of commercially pure niobium over a broad
range of temperatures and strain rates‖, Materials Science and Engineering A284
(2000) 202–210.

[40] Q. Xue, V.F. Nesterenko, M.A. Meyers, ―Evaluation of the collapsing thick-walled
cylinder technique for shear-band spacing‖, International Journal of Impact
Engineering 28 (2003) 257–280.

[41] C. J. Shih, M. A. Meyers, V. F. Nesterenko, ―High strain rate deformation of

granular silicon carbide‖, Acta Materialia 46 (1998) 4037-4065.

[42] H.C. Rogers, ―Adiabatic plastic deformation‖, Annual Review of Material Science,
1979, vol. 9, pp. 283–311.

[43] Q. Li, Y.B. Xu, Z.H. Lai, L.T. Shen, Y.L. Bai, ―Dynamic recrystallization induced
by plastic deformation at high strain rate in a Monel alloy‖, Materials Science and
Engineering A276 (2000) 250–256.

[44] G.A. Li, L. Zhen, C. Lin, R.S. Gao, X. Tan, C.Y. Xu, ―Deformation localization
and recrystallization in TC4 alloy under impact condition‖, Materials Science and
Engineering A 395 (2005) 98–101.

[45] D. Rittel, Z.G. Wang, A. Dorogoy, ―Geometrical imperfection and adiabatic shear
banding‖, International Journal of Impact Engineering 35 (2008) 1280–1292.

[46] Y. Me-Bar, D. Shechtman, ―On the adiabatic shear of Ti-6Al-4V ballistic targets‖,
Material Science and Engineering 58 (1983) 181–188.

[47] Q. Xue, M.A. Meyers, V.F. Nesterenko, ―Self-organization of shear bands in

titanium and Ti–6Al–4V alloy‖, Acta Materialia 50 (2002) 575–596.

[48] M. A.Meyers, V. F. Nesterenko, J. C. LaSalvia, Qing Xue, ―Shear localization in

dynamic deformation of materials: microstructureal evolutiona and self-
organization‖, Materials Science and Engineering A317 (2011) 204-225.

121
[49] K.M. Roessig, J.J. Mason, ―Adiabatic shear localization in the dynamic punch test,
part I. experimental investigation‖, International Journal of Plasticity 15 (1999)
241-262.

[50] C.G. Lee, Y.J. Lee, S. Lee, ―Observation of adiabatic shear bands formed by
ballistic impact in Aluminum-Lithium alloys‖, Scripta Metallurgica et Materialia
32 (1995) 821.

[51] T.W. Wright, ―Approximate analysis for the formation of adiabatic shear bands‖,
Journal of the Mechanics and Physics of Solids 38 (1990) 515–530.

[52] A. Molinari, R.J. Clifton, ―Analytical characterization of shear localization in

thermoviscoplastic materials‖, Journal of Applied Mechanics 54 (1987) 806–812.

[53] M. C. Mataya, M. J. Carr, G. Kraus, ―Flow Localization and Shear Band

Formation in a Precipitation Strengthened Austenitic Stainless Steel‖,
Metallurgical Transactions 13A (1982) 1263.

[54] M. A. Meyers, U. R. Andrade, A. H. Chokshi, ―The Effect of Grain Size on the

High-Strain, High-Strain-Rate Behavior of Copper‖, Metallurgical Transactions
26A (1995), 2881.

[55] J.R. Li, J.L. Yu, Z.G. Wei, ―Influence of specimen geometry on adiabatic shear
instability of tungsten heavy alloys‖, International Journal of Impact Engineering
28 (2003) 303–314.

[56] D.E. Graddy, M.E. Kipp, ―The growth of unstable thermoplastic shear with
application to steady-wave shock compression in solids‖, Journal of the
Mechanics and Physics of Solids 35 (1987) 95-119.

[57] T.W. Wright, H. Ockendon, ―A scaling law for the effect of inertia on the formation
of adiabatic shear bands‖, International Journal of Plasticity 12 (1996) 927-934.

[58] B. Hwang, S. Lee, Y. C. Kim, N. J. Kim, D. H. Shin, ―Microstructural development

of adiabatic shear bands in ultra-fine-grained low-carbon steels fabricated by equal

122
 channel angular pressing‖, Materials Science and Engineering A 441 (2006) 308–
320.

[59] G. F. Vander Voort, “ASM handbood”, Volume 8 (2000) 469-495.

[60] J.F.C. Lins H.R.Z. Sandim, H.J. Kestenbach, D. Raabe, K.S. Vecchio, ―A
microstructural investigation of adiabatic shear bands in an interstitial free steel‖,
Materials Science and Engineering A 457 (2007) 205–218.

[61] K. Cho, Y.C. Chi, J. Duffy, ―Microscopic observations of adiabatic shear bands in
three different steels‖, Metallurgical Transactions A 21 (1990) 1161–1175.

[62] X.W. Chen, Q.M. Li, S.C. Fan, ―Initiation of adiabatic shear failure in a clamped
circular plate struck by a blunt projectile‖, International Journal of Impact
Engineering 31 (2005), pp. 877–893.

[63] M.A. Meyers, Y.B. Xu, Q. Xue, M.T. Perez-Prado, T.R. McNelley,
“Microstructural evolution in adiabatic shear localization in stainless steel”, Acta
Materialia 51 (2003) 1307–1325.

[64] J.A. Hines, K.S. Vecchio, S. Ahzi, ―A model for microstructure evolution in
adiabatic shear bands‖, Metallurgical and Materials Transactions 29A (1998)
191–203.

[65] M. T. Perez-prado, J. A. Hines, K. S. Vecchio, ―Microstructural evolution in

adiabatic shear bands in Ta AND Ta–W alloys‖, Acta Materialia 49 (2001) 2905–
2917.

[66] Y.J. Chen, M.A. Meyers, V.F. Nesterenk, ―Spontaneous and forced shear
localization in high-strain-rate deformation of tantalum‖, Materials Science and
Engineering A268 (1999) 70–82.

[67] J.L. Derep, ―Microstructure transformation induced by adiabatic shearing in armour

steel‖, Acta Metallurgica 35 (1987) 1245–49.

123
[68] D.R. Lesuer, C.K. Syn, O.D. Sherby, ―Severe plastic deformation through adiabatic
shear banding in Fe–C steels‖, Materials Science and Engineering A 410–411
(2005) 222–225.

[69] J. Talone, ―Effect of strain-induced α’-martensite transformation on mechanical

properties of metastable austenitic Stainless Steel‖, Doctoral dissertation, Helsinki
University of Technology, Department of Mechanical Engineering, Laboratory of
Engineering Material, 2007.

[70] W.S. Lee, C.Y. Liu, T. N Sun, ―Dynamic impact response and microstructural
evolution of inconel 690 superalloy at elevated temperatures‖, International
Journal of Impact Engineering 32 (2005) 210–223.

[71] W. S. Lee, C. F. Lin, ―Impact properties and microstructure evolution of 304L

Stainless steel‖, Materials Science and Engineering A 308 (2001) 124–135.

[72] S.P. Timothy, I.M. Hutchings, ―The structure of adiabatic shear bands in titanium
alloy‖, Acta Metallurgica 33 (1985) 667.

[73] V. F. Nesterenkot, M. A. Meyers, H. C. Chen, ―Shear Localization in High-Strain

rate Deformation of granular Alumina‖, Acta Materialia 44 (1996) 2017-2026.

[74] W. Xue-bin, ―Phenomenon of transformed adiabatic shear band surrounded by

deformed adiabatic shear band of ductile metal‖, The Transactions of Nonferrous
Metals Society of China, 18 (2008) 1177-1183.

[75] J. F. Kalthoff, A. Burgel, ―Influence of loading rate on shear fracture toughness for
failure mode transition‖, International Journal of Impact Engineering 30 (2004)
957–971.

[76] M.N. Bassim, A.G. Odeshi, ―Shear strain localisation and fracture in high strength
structural materials‖, Archives of Materials Science and Engineering 31 (2008) 69-
74.

[77] Q. Xue, M.A. Meyers, V.F. Nesterenk, ―Self-organization of shear bands in

titanium and Ti–6Al–4V alloy‖, Acta Materialia 50 (2002) 575–596.

124
[78] A.G. Odeshi, S. Al-ameeri, S. Mirfakhraei, F. Yazdani, M.N. Bassim,
―Deformation and failure mechanism in AISI 4340 steel under ballistic impact‖,
Theoretical and Applied Fracture Mechanics 45 (2006) 18–24.

[79] K. Ryttberg, M. Knutson Wedel, P. Dahlman, L. Nyborg, ―Microstructural

evolution during fracture induced by high strain rate deformation of 100Cr6 steel‖,
Journal of Materials Processing Technology 209 ( 2009 ) 3325–3334.

[80] D.M. Turley, E.D. Doyle, S. Ramaligam, ―Calculation of shear strains in chip
formation in titanium‖, Materials Science and Engineering. A 55 (1982) 45.

[81] J.C. Lemaire, W.A. Backofen, ―Adiabatic instability in the orthogonal cutting of
steel‖, Metallurgical Transaction 3 (1972) 477.

[82] G.R. Speich, “ASM handbook”, Vol. 1 (2005) 697-707.

[83] A.G. Odeshi, M.N. Bassim, S. Al-Ameeri, Q. Li, ―Dynamic shear band propagation
and failure in AISI 4340 steel‖, Journal of Materials Processing Technology 169
(2005) 150–155.

[84] Amos Gilat, “ASM handbook”, Volume 8 (2000) 1134-1162.

[85] C.C. Menzemera, T.S. Srivatsanb, R. Ortiza, Meslet Al-Hajrib, M. Petraroli,

―Influence of temperature on impact fracture behavior of an alloy steel‖, Materials
and Design 22 (2001) 659-667.

[86] A.S. Bonnet-Lebouvier, A. Molinari, P. Lipinski, ―Analysis of the dynamic

propagation of adiabatic shear bands‖, International Journal of Solids and
Structures 39 (2002) 4249–4269.

[87] A. Gilat, C.S. Cheng, ―Torsional split Hopkinson bar tests at strain rates above 104
s-1‖, Experimental Mechanics 40 (2000) 54–59.

[88] K. Ryttberg, M. Knutson Wedel, L. Nyborg, ―Electron microscopy of white-

etching band generated by high-velocity parting off of 100CrMn6 steel‖, Materials
Science and Engineering A 480 (2008) 489–495.

125
[89] W. D. Callister, Jr., “Materials science and engineering an introduction”, 7th
edition, John Wiley and Sons, Inc., 2007.

126