© 2019

HAO ZHANG

ALL RIGHTS RESERVED

ENERGY ASSISTED-SURFACE SEVERE PLASTIC DEFORMATION

OF HARD-TO-DEFORM METALS

A Dissertation

Presented to

The Graduate Faculty of The University of Akron

In Partial Fulfillment

of the Requirement for the Degree

Doctor of Philosophy

Hao Zhang

December 2019
 ENERGY ASSISTED-SURFACE SEVERE PLASTIC DEFORMATION

OF HARD-TO-DEFORM METALS

Hao Zhang

Dissertation

Approved: Accepted:

Advisor Department Chair

Dr. Chang Ye Dr. Sergio Felicelli

Co-Advisor Interim Dean of the College

Dr. Yalin Dong Dr. Craig Menzemer

Committee Member Acting Dean of the Graduate School

Dr. Xiaosheng Gao Dr. Marnie Saunders

Committee Member Date

Dr. Qixin Zhou

Committee Member
Dr. Jun Ye ii
 ABSTRACT

Surface severe plastic deformation attracts more and more attentions because its

capability to enhance material surface. However, for the hard-to-deform materials, higher

peening intensity leads to surface or subsurface cracks, while lower peening intensity may

not be sufficient to effectively induce beneficial microstructural changes. In this

dissertation, energy assisted-surface severe plastic deformation is proposed to improve

the effectiveness of surface severe plastic deformation on hard-to-deform alloys. Three

different energy input methods, direct current, pulsed current, and laser were used to

integrate with LSP and UNSM in processing metallic materials.

Direct current can rapidly increase the bulk temperature of the 3D-printed Ti64 during

UNSM process. Compared with the conventional UNSM process, better surface finish and

lower subsurface porosities were obtained after DC-UNSM. DC-UNSM also led to a deeper

plastically-affected depth compared with conventional UNSM. Numerical modelling

showed that localized heating occurs near the pores in 3D-printed Ti64 subjected to

electric current, which could potentially facilitate pore closure under ultrasonic striking.

Electropulsing assisted-LSP (EP-LSP) produced higher surface hardness and deeper

hardened layer on conventional Ti64, both of which indicate greater plastic deformation

compared with LSP. Electropulsing assisted tensile test showed that pulsed current can

iii
more effectively decrease the flow stress even though the bulk heating effect is the same.

In addition, the higher the peak current density, the more effective the flow stress

reduction. It is believed that an athermal effect in addition to the thermal effect related

to pulsed current exists in EP-LSP.

Laser assisted UNSM (LA-UNSM) is proposed as an alternative to post-process 3D-

printed Ti64. The localized thermal effect of continuous laser also decreases the flow

stress of the alloy effectively. Smoother and harder surface was obtained after LA-UNSM

due to the enhanced plasticity by pure thermal effect. In addition, LA-UNSM also led to a

thicker plastically-affected layer and lower porosity compared with traditional UNSM

process.

In summary, an innovation concept and surface engineering technology, energy

assisted surface severe plastic deformation was proposed to improve the effectivity and

efficiency of severe plastic deformation on processing hard-to-deform alloys, which

integrates the advantages of SSPD to introduce surface plastic deformation induced

strengthening effect, with the ability of extra energies to improve plasticity through their

thermal and athermal effects.

iv
 ACKNOWLEDGEMENTS

Firstly, I would like to express my sincere gratitude to my advisor Dr. Chang Ye and co-

advisor Dr. Yalin Dong for the continuous support during my PhD program. Under their

supervision, I learned how to define a research problem, find a solution to it, and finally

publish the results. Besides, I benefited greatly from their hardworking and passionate

attitude, which have a profound effect on my current research and future career. I could

not have imagined having better advisors and mentor for my Ph.D. study.

I also would like to thank all my colleagues. Since the topic of my Ph.D. program is

quite different from as what I did in my master’s program, their help made me understand

the research here quickly. I also obtain a lot of new ideas from the stimulating discussion.

I would give my special thanks to Dr Jingyi Zhao, Dr Zhencheng Ren, and Dr Chi Ma, with

whom I accomplish most of my projects. I will always remember the friendship with them.

Last but not least, I sincerely appreciate the support from my parents. It is their

support and encouragements that help me accomplish and eventually obtain the PhD

degree. I am strongly grateful for my wife Renchen Xiao. Thanks for her love,

understanding and sacrificing when I am struggling on the academic road. Appreciate all

people who make contributions to this dissertation work.

v
 TABLE OF CONTENTS

Page

LIST OF FIGURES .................................................................................................................. xi

LIST OF TABLES .................................................................................................................. xvi

CHAPTER

INTRODUCTION ................................................................................................................ 1

Surface severe plastic deformation ........................................................................... 1

Problem statement .................................................................................................... 2

Scope of this research ................................................................................................ 5

Thesis structure.......................................................................................................... 6

II. LITERATURE REVIEW ....................................................................................................... 7

Strengthening mechanism of surface severe plastic deformation ............................ 7

Hard-to-deform alloys ................................................................................................ 9

Dislocation theory of plastic deformation ............................................................... 10

Promotion of dislocation by energy input ............................................................... 11

vi
 Thermal energy input: .................................................................................... 11

Electric energy input ....................................................................................... 13

Material processing at elevated temperature ......................................................... 15

Hot working..................................................................................................... 15

High temperature surface severe plastic deformation .................................. 16

Electrically-assisted material processing ................................................................. 16

Laser-assisted material processing .......................................................................... 17

Chapter summary .................................................................................................... 18

III. MATERIAL CHARACTERIZATION METHODS USED IN THIS DISSERTATION .................. 19

Hardness test ........................................................................................................... 19

Surface topography .................................................................................................. 19

Microstructure ......................................................................................................... 20

XRD characterization ................................................................................................ 21

Porosity by micro-CT and surface profiler ............................................................... 21

Residual stress measurement .................................................................................. 22

Electron backscatter diffraction/orientation imaging microscopy .......................... 24

Transmission electron microscopy (TEM) ................................................................ 24

Tensile tests .............................................................................................................. 25

vii
 Chapter summary .................................................................................................. 26

IV. DICRECT CURRENT ASSISTED UNSM ON 3D-PRINTED TI64 ......................................... 27

Introduction ............................................................................................................. 27

Experimental details ................................................................................................ 28

Sample preparation ........................................................................................ 28

DC-UNSM processing ...................................................................................... 28

Results ...................................................................................................................... 29

Surface finish improvement by DC-UNSM ...................................................... 29

Porosity reduction in the surface layer by DC-UNSM ..................................... 32

Thicker deformation layer by DC-UNSM ........................................................ 35

XRD patterns evolution ................................................................................... 36

Al2O3 generation by DC-UNSM ....................................................................... 37

Hardening effect by DC-UNSM ....................................................................... 39

Discussion ................................................................................................................ 40

Chapter summary .................................................................................................... 44

V. ELECTROPULSING ASSISTED LSP ON TI64 ..................................................................... 46

Introduction ............................................................................................................. 46

viii
 Experimental details ................................................................................................ 47

Materials ......................................................................................................... 47

EP-LSP experiment .......................................................................................... 48

Results ...................................................................................................................... 50

Hardening effect by EP-LSP ............................................................................. 50

Surface roughness increase by enhanced plastic deformation ...................... 51

XRD patterns evolution ................................................................................... 52

Cross-sectional microstructure after EP-LSP .................................................. 53

Misorientation induced by EP-LSP .................................................................. 55

More dislocations and deformation twins induced by EP-LSP ....................... 58

Discussion ................................................................................................................ 61

Chapter summary .................................................................................................... 66

VI. LASER ASSISTED UNSM ON 3D-PRINTED TI64 ............................................................. 67

Introduction ............................................................................................................. 67

Experimental details ................................................................................................ 68

Materials ......................................................................................................... 68

UNSM/LA-UNSM ............................................................................................. 68

ix
 Finite element method (FEM) modelling ........................................................ 69

Results ...................................................................................................................... 70

Further enhanced surface hardness by LA-UNSM .......................................... 70

Surface finish induced by LA-UNSM ............................................................... 71

XRD patterns evolution ................................................................................... 73

Thicker deformation layer by LA-UNSM ......................................................... 75

Porosity reduction in the surface layer by LA-UNSM ..................................... 76

Residual stress induced by LA-UNSM ............................................................. 78

Discussion ................................................................................................................ 80

Local heating effect of continuous laser ......................................................... 81

Nanograins remained at the top and sub-surface layer ................................. 86

Chapter summary .................................................................................................... 87

VII. CONCLUSION .............................................................................................................. 89

REFERENCES ...................................................................................................................... 92

x
 LIST OF FIGURES

Figure Page

1.1 Schematic model of (a) UNSM, (b)LSP. ......................................................................... 2

1.2 Optical micrographs of surface pores (a) before and (b) after UNSM-L treatment.

Images in (c) and (d) are the magnified images of the square areas c and d in the image

shown in (b). ....................................................................................................................... 4

1.3 Cross-sectional optical micrographs (a) a dual-phase Mg-Li alloy after 3 passes of

UNSM processing, (b) the magnified images of the cracks induced by UNSM strikes. ..... 5

2.1 Schematics of (a) working hardening and (b) grain boundary strengthening .............. 9

2.2 Primary slip systems for HCP phase of titanium. ........................................................ 10

2.3 Schematic of hot working processes. ......................................................................... 15

3.1 Zygo NewView 7300 white-light optical profilometer in the Department of Civil

Engineering, University of Akron. ..................................................................................... 20

3.2 Tescan LYRA-3 Model XMU FIB-FESEM system in NCERCAMP, the University of Akron.

........................................................................................................................................... 21

3.3 Skyscan 1172 Micro Computer Tomography Scanner in National Polymer Innovation

Center, the University of Akron. ....................................................................................... 22

xi
3.4 Residual stress measurement: (a) Schematic of measurement positions and (b) a

photograph of the sample. ............................................................................................... 24

4.1 Schematic of the DC-UNSM process. .......................................................................... 29

4.2 The 2D surface profiles and SEM images obtained for 3D-printed Ti64 samples: (a,d)

Control; (b,e) UNSM-treated samples; and (c,f) DC-UNSM–treated samples. ................ 30

4.3 Surface roughness (Ra) of the control, UNSM-treated and DC-UNSM-treated 3D-

printed Ti64 samples......................................................................................................... 31

4.4 Change in porosity of the 3D-printed Ti64 for control, UNSM-treated and DC-UNSM-

treated samples: (a) optical micrograph, (b) processed grayscale image, and (c) histogram

of pores of different sizes. ................................................................................................ 33

4.5 Micro-CT plane view images of the 3D-printed Ti64 samples: (a) control, (b) UNSM-

treated and (c) DC-UNSM–treated samples. Cross sections of 3D-printed Ti64 samples: (d)

control, (e) UNSM-treated and (f) DC-UNSM–treated samples. 3D reconstructed CT

pictures of 3D-printed Ti64 samples: (g) control, (h) UNSM-treated and (i) DC-UNSM–

treated samples. ............................................................................................................... 35

4.6 SEM images of cross sections of 3D-printed Ti64: (a) control, (b) UNSM-treated and

(c) DC-UNSM–treated samples. ........................................................................................ 36

4.7 XRD patterns for the 3D-printed Ti64. ........................................................................ 37

xii
4.8 In-depth element concentration for control, UNSM-treated and DC-UNSM-treated

samples: (a) oxygen, (b) aluminum, (c) titanium, and (d) vanadium. .............................. 39

4.9 (a) Surface hardness and (b) in-depth hardness of control, UNSM-treated and DC-

UNSM-treated 3D-printed Ti64 samples. ......................................................................... 40

4.10 Schematic diagram of a 3D-printed metal subjected to DC-UNSM. ........................ 41

4.11 Ti64 matrix with pores treated using DC-UNSM: (a) The electric current flow, (b) the

electric current density (in A/mm2) and (c) the heating rate (in K/s). ............................. 44

5.1 Schematic of the EP-LSP process. ............................................................................... 48

5.2(a) Surface hardness and (b) in-depth hardness for LSP, DC-LSP, and EP-LSP treatments.

........................................................................................................................................... 50

5.3 Surface roughness at 25x and 100x magnifications after different treatments. ....... 52

5.4 XRD patterns of the Ti64 samples subjected to LSP, DC-LSP, and EP-LSP. ................. 53

5.5 Cross-sectional microstructure of Ti64 samples: (a) control, (b) LSP, (c) DC-LSP, and (d)

EP-LSP. ............................................................................................................................... 54

5.6 Inverse pole figures (IPF) from EBSD observations in the cross sections of the samples

(where all top edges correspond to the top surface of the samples) for (a) controls, (b)

DC-LSP, (c) EP-LSP; KAM charts and maps corresponding to the (d, g) control, (e, h) DC-

LSP, and (f, i) EP-LSP samples; the inset in (g) is the legend for three KAM maps. ......... 56

xiii
5.7 Pole figures from EBSD observation of the cross sections (where all top edges

correspond to the top surface of the samples) of the (a) control, (b) DC-LSP, and (c) EP-

LSP samples. ...................................................................................................................... 58

5.8 Bright-field TEM micrographs of near-surface microstructures of LSP (a–c) and EP-LSP-

processed (d–i) Ti64 samples. In these images, bending contours are indicated by red

arrows, dislocation walls are indicated by yellow arrows, twin boundaries are indicated

by cyan arrows, dislocation tangles are indicated by green arrows, planar dislocations are

enclosed in red-dashed circles, and sub-grain boundaries are indicated by white arrows.

........................................................................................................................................... 60

5.9 Temperature history for samples processed using DC-LSP and EP-LSP. .................... 62

5.10 (a) The temperature profile and (b) strain–stress curves of Ti64 samples subjected to

continuous and pulsed current, and (c) the tensile strength as a function of the peak

current density. ................................................................................................................. 64

6.1 Schematic of the LA-UNSM process. .......................................................................... 69

6.2 FEM model of single-line scanning of UNSM on 3D-printed Ti64 with pore.............. 70

6.3 Surface hardness of 3D-printed Ti64 samples before and after UNSM and LA-UNSM

treatment .......................................................................................................................... 71

6.4 Surface roughness of 3D-printed Ti64 samples before and after UNSM and LA-UNSM

treatment. ......................................................................................................................... 73
xiv
6.5 (a)XRD pattern of 3d-printed Ti64 after UNSM and LA-UNSM process, (b) FWHM of

XRD peaks, (c) 2θ offset between UNSM, LA-UNSM and control .................................... 74

6.6 Cross-section SEM images of the 3D-printed Ti64: (a) control, (b) UNSM and (c) LA-

UNSM ................................................................................................................................ 76

6.7 Plane view images and the cross sections of the 3D-printed Ti64 samples: (a) control,

(b) UNSM-treated and (c) LA-UNSM-treated; (d) histogram of pores before and after

UNSM treatment............................................................................................................... 78

6.8 In-depth residual stress profile of 3D-printed Ti64 samples (a) along the X direction

and (b) along the Y direction before and after UNSM and LA-UNSM processing. ........... 78

6.9 The schematic diagram of LA-UNSM .......................................................................... 81

6.10 The IR image during (a) UNSM treatment and (a) LA-UNSM treatment on 3D-printed

Ti64, (c) Emissivity of 3D-printed Ti64 at different temperatures. .................................. 83

6.11 (a)Temperature distribution along depth direction. Effect of laser on pore reduction

with different depth: (b) 25 µm, (c) 75 µm, (d) 250 µm. Reduced fraction of pore (e) with

different depth and (f) comparison with and with out laser. ........................................... 85

6.12 Effect of laser energy on the pore size. .................................................................... 86

6.13 TEM bright field image of LA-UNSM-processed Ti6Al4V alloys, SAED patterns are

inserted. ............................................................................................................................ 87

xv
 LIST OF TABLES

Table Page

2.1 Slip systems in different crystal structure .................................................................. 10

3.1 Electrical parameters used during the tensile tests ................................................... 26

5.1 Electrical parameters used in the DC-LSP and EP-LSP experiments ........................... 49

5.2 Maximum intensity of different pole figures for Ti64 samples before and after

different LSP treatments. .................................................................................................. 58

xvi
 CHAPTER I

INTRODUCTION

Surface severe plastic deformation

Surface severe plastic deformation (SSPD) is developed from bulk severe plastic

deformation and aim to strengthen the material surface, since most failure of engineering

materials, including fatigue fracture, wear and corrosion etc., are sensitive to the

structure and properties of material surface. With the development of decades, some

advanced SSPD techniques, such as shot peening [1], surface mechanical attrition

treatment [2–5], ultrasonic surface rolling process [6], ultrasonic nanocrystal surface

modification [7,8], and laser shock peening [9] etc., are commonly used to in academic

research and industrial areas. As two representative, UNSM and laser shock peening (LSP),

have been successfully used to improve metallic component mechanical and corrosion

performance in recently several decades [10–21]. UNSM [7,8] is a recently developed

technique (Figure 1.1a) that induces severe plastic deformation (SPD) on a metal surface

by utilizing ultrasonic frequency vibrations (20 kHz) superimposed with a static load.

Significant improvements in fatigue performance [10], wear resistance [15–17], water

erosion resistance [22], and tensile strength [23] have been reported. Compared with

traditional ultrasonic peening, the process parameters in a UNSM process can be precisely

1
controlled, which gives the process greater reliability and repeatability. LSP uses high

energy pulsed laser to introduce high pressure plasma on sample surface (Figure 1.1b).

When the peak stress generated by plasma on sample surface exceed the Hugoniot elastic

limit of the material, plastic deformation occurs. Similar with UNSM, LSP could also induce

compressive residual stress on material surface.

(a) (b)

Figure 1.1 Schematic model of (a) UNSM, (b)LSP.

Problem statement

Since UNSM and LSP are deformation-based surface treatment processes, their

processing efficiency is highly dependent on metal plasticity. For example, due to the low

plasticity of the hard-to-deform alloys, e.g. 3D-printed Ti64, it is challenging to select the

optimal UNSM process conditions to produce the best property enhancement. When

higher striking intensity (i.e., high ultrasonic amplitude) is used, surface or subsurface

cracks will form due to the intrinsic brittleness of the hard-to-deform alloys. While lower

UNSM striking intensity, on the other hand, may not be sufficient to effectively induce

beneficial changes in microstructure. Therefore, when processing hard-to-deform alloys,

increasing the dislocation mobility to improve the plasticity of these alloys is a feasible

approach to further improve the effect of surface severe plastic deformation.

2
 It is known that micro-cracks may form in metals, if the local maximum shear stress

exceeds the shear yield stress [24], cracks can be form easier in alloys with hexagonal

close packed (HCP) crystal structure which has less dislocation slip systems, and alloys

with multiple heterogeneities occurring at the nano/microscale, like voids, inclusions,

pores, and surface roughness.

The primary phase in Ti64 and 3D-printed Ti64 alloy is an HCP structured phase. There

are also some pores distributed in the 3D-printed Ti64 alloy. Figure 1.2 shows the surface

microstructure of the 3D-printed Ti64 samples before and after UNSM treatment [25].

Before UNSM, irregular pores with random shapes and sizes were observed on the sample

surface (Figure 1.2a). After UNSM, pores became smaller and some of them even closed

(Figure 1.2b and c) due to the compression effect induced by UNSM. Some cracks were

observed near a large pore after UNSM treatment (Figure 1.2d).

3
 Figure 1.2 Optical micrographs of surface pores (a) before and (b) after UNSM-L

treatment. Images in (c) and (d) are the magnified images of the square areas c and d in

the image shown in (b).

Another typical HCP structured is Mg alloy. Figure 1.3 shows the microstructure of a

dual-phase Mg-Li alloy. Cross-section of the alloy after 3 passes of UNSM process is shown

in Figure 1.3, part of the α-phases (bright phase) at the top surface layer are refined and

elongated. However, when increase the static load of the UNSM to make the α-phases

finer and more uniform, cracks generate at the phase boundaries on the alloy surface.

4
 (a) (b)

Figure 1.3 Cross-sectional optical micrographs (a) a dual-phase Mg-Li alloy after 3

passes of UNSM processing, (b) the magnified images of the cracks induced by UNSM

strikes.
Scope of this research

Energy assisted surface severe plastic deformation (EASSPD) is SSPD with extra energy

input, such as, direct current, pulsing current, and continuous laser. The combination of

the thermal and mechanical effects appears to work synergistically during the EASSPD

process. By introducing flowing stress decrease and work hardening promotion, it is

expected that EASSPD can generate a thick surface hardening layer and further improve

mechanical properties especially on hard-to-deform alloys. The aim of this research is to

investigate the effect of extra energy on the SSPD processing of hard-to-deform alloys.

In this study, 3D-printed Ti6Al4V (Ti64) was used to evaluate the effect of direct

current assisted UNSM (DC-UNSM) and laser assisted UNSM. Numerical method was used

to solve Kirchhoff’s circuit laws and obtain the current density and heating rate

distribution during DC-UNSM process. Finite element method (FEM) is used to investigate

5
effect of LA-UNSM on pore closing at different depth and different laser energy. Tradition

Ti64 is used to investigate the effect of electropusing assisted laser shock peening (EALSP).

Thesis structure

This dissertation is structured as follows. Chapter two gives a literature review about

hot working and related research work on energy assisted machining. Chapter three

presents the basic characterization methods used in this study. Chapter four presents the

direct current assisted-UNSM (DC-UNSM) work on 3D-printed Ti6Al4V (Ti64) alloy.

Chapter five presents the electropulsing assisted-LSP (EP-LSP) work on traditional Ti6Al4V

(Ti64) alloy. Chapter 6 presents the laser assisted ultrasonic nanocrystal surface

modification (LA-UNSM) work on 3D-printed Ti64. Chapter seven draws the conclusion of

this dissertation.

6
 CHAPTER II2

LITERATURE REVIEW

Strengthening mechanism of surface severe plastic deformation

As stated in Chapter 1, it has been proven that SSPD can enhance the mechanical

properties of materials. In this section, the strengthening mechanism of SSPD will be

briefly introduced.

From the denominate of SSPD, it is explicit that a severe plastic deformation will be

introduced to the top surface layer of the processed materials. At the very top surface,

large and rapid plastic deformation results in high density of deformation induced defects,

dislocations and/or twins, which depends on factors like deformation rate, applied stress,

and stacking fault energy of the materials [9]. Vast of these defects tangle together and

divide the coarse grain into nano-sized grains by forming sub-grain boundaries and grain

boundaries [26], and consuming dislocations and twin boundaries. In the subsurface layer,

attenuation of plastic deformation retards the transformation from dislocation cells to

new grains. Thus, grain size does not decrease remarkably while significant grain

distortion could be observed, and large number of deformation-induced defects reserved.

For deeper region, the transfer of deformation further weakens, and the grains of

materials gradually recovers to the undeformed status.

7
 Theoretically, without considering precipitation and phase transformation, material

strengthening mechanism by plastic deformation can be mainly attributed to two aspects,

working hardening [27] and grain boundary strengthening [28]. For working hardening,

dislocations interact with each other and assume configurations that restrict the

movement of other dislocations, thus flow stress of the material increases with the

dislocation propagation. Larger stress is needed to restart the defect motion, and thus

increase the material strength. For grain boundary strengthening, grain boundaries act as

barriers of dislocation movement. Smaller grains means larger proportion of grain

boundaries and induced larger material strength, until the grain boundary predominate

the deformation process [5,29,30]. These two strengthening mechanisms can be

summarized in Eq. 2.1 [28]:

1 1
−
σf = 𝜎0 + 𝑘(𝑑𝑓𝑝 ) 2
+ 𝛼𝐺𝑏𝜌2 Eq. 2.1

where σf is the final strength of the processed material, 𝜎0 is a friction stress, 𝑘 is the

Hall-Petch constant, 𝑑𝑓𝑝 is the mean free patch for dislocations, 𝛼 is a constant, 𝐺 is the

shear modulus, 𝑏 is the Burgers vector and 𝜌 is the dislocation density.

8
 (a) (b)

Figure 2.1 Schematics of (a) working hardening and (b) grain boundary strengthening

Hard-to-deform alloys

SSPD techniques are deformation-based surface treatment processes, their

processing efficiency is highly dependent on deformation behavior of materials. However,

for hard-to-deform alloys, the strengthening effect of SSPD will be limited to some extent.

The reasons for hard-to-deform mainly originate from two aspects: (1) Low modulus

and high strength, (2) high work hardening tendency. Take titanium alloys as an example

[31,32], a relatively low Young’s modulus and high strength of titanium alloys leads to

spring-back and low plastic deformation remained when the SSPD loading is mild.

However, relatively less slip systems in HCP structured alloys (Table 2.1) [33] make it

easier for slipping dislocations to pile up and thus induced high work hardening. When

aggressive loading applied on the hard-to-deform alloys, blocked dislocations cannot

accommodate further deformation. Local maximum shear stress exceeds the shear yield

stress [24], and hence cracks form at strain concentration region.

9
Table 2.1 Slip systems in different crystal structure

Structures Metals Slip plane Slip direction Slip system

FCC Cu, Al, Ni, Ag, Au {1 1 1} ⟨1 1̅ 0⟩ 12

α − Fe, W, Mo {1 1̅ 0} 12

BCC α − Fe, W {1 2̅ 1} ⟨1 1 1⟩ 12

α − Fe, K {2 3̅ 1} 24

α − Ti, Mg, Zn {0 0 0 1} 3

HCP α − Ti, Mg, Zr {1 0 1̅ 0} ⟨1̅ 2 1̅ 0⟩ 3

α − Ti, Mg {1 0 1̅ 1} 6

Figure 2.2 Primary slip systems for HCP phase of titanium.

Dislocation theory of plastic deformation

It is known that dislocation movement is the most common manifestation of plastic

deformation in crystalline solids. The Orowan equation can be used to describe the

relationship between the plastic strain rate and the average dislocation velocity:

𝜀̇ = 𝑏𝜌𝑚 𝑣̅ Eq. 2.2

10
where 𝜀̇ is the macroscopic strain rate, 𝑏 is the Burgers vector, 𝜌𝑚 is the density of mobile

dislocations, and 𝑣̅ is the average dislocation velocity. Then the relationship between the

plastic strain and the dislocation density:

𝜀 = 𝑏𝜌𝑚 𝑥̅ Eq. 2.3

where 𝑥̅ is the average distance moved by a dislocation. It is emphasized that 𝜌𝑚

appearing in the Eq. 2.2 is the mobile dislocation density, for dislocations which do not

move do not contribute to the plastic strain. According to the equation, in a certain

crystalline solid, the average distance of a dislocation has a maximum value, when there

is higher mobile dislocation density, the plastic strain will be larger. However, during the

plastic deformation, dislocations are continually emitted from dislocation sources. Once

these dislocations meet a barrier such as a grain boundary or sessile dislocations, further

expansion of the loop will be prevented. The entanglement of the dislocations makes the

dislocation not be able to combine or further slipping, micro-cracks initiate as a result.

Promotion of dislocation by energy input

Thermal energy input:

Energy input to metals can help dislocations overcome the obstacle they encounter

during slip. Many expressions defining the dislocation speed for thermally activated

dislocation glides may be found in the literature [34–38]. As summarized in [39], an

increase in temperature will reduce the flow stress and improve plasticity of the metals.

When a dislocation with length of 𝑙 under an applied resolved shear stress 𝜏 ∗ , which

produces a driving force 𝜏 ∗ 𝑏𝑙 on the dislocation line. Suppose each of obstacles produces

11
a resisting force K on the dislocation. To overcome the barrier, the line must move from

position 𝑥1 to 𝑥2 . The isothermal energy change required is equal to the Helmholtz free

energy change
𝑥
∆𝐹 ∗ = ∫𝑥 2 𝐾𝑑𝑥 Eq. 2.4
1

Part of this energy can be provided in the form of mechanical work done by the applied

load and is 𝜏 ∗ 𝑏𝑙(𝑥2 − 𝑥1 ) = 𝜏 ∗ 𝑉 ∗ , where 𝑉 ∗ is known as the activation volume for the

process. The remainder of the energy required the free energy of activation:

∆𝐺 ∗ = ∆𝐹 ∗ − 𝜏 ∗ 𝑉 ∗ Eq. 2.5

If the dislocation is effectively vibrating at a frequency 𝜈 , it successfully overcomes

barriers at a rate of 𝜐exp(−∆𝐺 ∗ /kT)per second. The dislocation velocity is:

𝑣̅ = 𝑑𝜈exp(−∆𝐺 ∗ /kT) Eq. 2.6

where 𝑑 is the distance moved for each obstacle overcome. From Eq. 2.2 and Eq. 2.5, the

macroscopic plastic strain rate is:

𝜀̇ = 𝜌𝑚 𝑏𝑑𝜈exp(−∆𝐺 ∗ /kT) Eq. 2.7

The relation between flow stress and temperature can be described using Johnson-Cook

model as:
𝑚
𝜀̇ 𝑇−𝑇𝑟𝑒𝑓
𝜎 = (𝐴 + 𝐵𝜀 𝑛 )(1 + 𝐶ln )(1 − ( ) ) Eq. 2.8
𝜀̇ 0 𝑇𝑚 −𝑇𝑟𝑒𝑓

where 𝜎 is the flow stress, 𝐴 is the yield stress at reference temperature and reference

strain rate, 𝐵 is the coefficient of strain hardening, 𝑛 is the strain hardening exponent, 𝜀̇0

is the reference strain rate, 𝑇 is the processing temperature, 𝑇𝑚 is the melting

12
temperature of the metal and 𝑇𝑟𝑒𝑓 is the reference temperature (𝑇 ≫ 𝑇𝑟𝑒𝑓 ), m is the

material constant that represents the coefficient of thermal softening exponent.

According to the Eq. 2.6-Eq. 2.8, the when thermal energy is provided to the metals

during the processing, metal temperature will rise and lead to higher dislocation mobility

and lower flow stress, which enhance the plasticity of the metals, which is widely used as

warm/hot working [40–48].

Electric energy input

Electricity is another kind of energy that is widely used to improve metal plasticity in many

manufacturing processes [49]. For example, continuous current has been applied to

improve forgeability and mitigate cracking in Mg alloys [50] by decreasing the required

forging forces. Salandro et al. [51] observed significantly decreased flow stress in

electrically-assisted forming (EAF). Reduced springback in sheet metal bending has also

been reported [52–54]. Very recently, pulsed current was integrated into turning [55,56],

rolling [57–61] and ultrasonic treatment [62–65]. Egea et al. [55] found that pulsed

current can increase machinability and improve the surface finish of carbon steels. Deeper

plastic-affected depth and finer grains were reported in Ti64 under electropulsing [62–

64]. It was argued that electropulsing activates dislocation motion on non-basal planes of

Ti64 and thus promotes more effective recrystallization. In some cases, significant

improvement in plasticity can be achieved while keeping the bulk temperature below

50 °C [66–68]. It was suggested that thermal effect (Joule heating) alone cannot explain

the dramatic change in microstructure and plasticity with electropusing input. Some

13
studies suggest the existence of an athermal effect [60,62,63,65,69–71] without a clear

understanding of the physics.

Extra free energy from heterogeneously distributed electric current is supposed to be

able to improve the dislocation mobility. Dislocation mobility with the presence of the

extra free energy provided by electric current can be calculated using [72–74]:

∆𝐺𝑒
𝑀𝑒 = 𝑀 ∙ 𝑒𝑥𝑝 ( ) Eq. 2.9
𝑘𝑏 𝑇

where 𝑀 is the dislocation mobility without electric current, 𝑘𝑏 is the Boltzmann constant,

𝑇 is temperature, and ∆𝐺𝑒 is the extra Gibbs free energy provided by the electric current

[73]:

𝑥𝑡 2
∆Ge = −∆G + 𝐽 Eq. 2.10
𝑛𝛾

where 𝑛 is the free electron number in a unit volume of metal, 𝐽 is electric current density,

γ is electrical conductivity, 𝑥 is the average number of metallic ions forming a dislocation,

and 𝑡 is the duration of the current. Dislocation velocity can be represented as the

product of dislocation mobility (𝑀𝑒 ) and the total stress (𝜏𝑡 ) acting on a unit dislocation:

𝑣𝑒 = 𝑀𝑒 𝜏𝑡 Eq. 2.11

The relation between flow stress and temperature can be described as [73]:
m
𝜎𝑒 = 𝐶 {2𝜀̇ − 𝜀̇exp [ln ( 0𝑒̇ ) + (
𝜀 𝑥𝑡
𝐽2 )]} Eq. 2.12
𝜀̇ 0 𝑛𝛾𝑘𝑏 𝑇

where 𝜎𝑒 is the flow stress with electricity applied, 𝜀0𝑒̇ is the pre-exponential factor with

electricity applied, 𝜀̇0 is the pre-exponential factor without electricity applied. Higher

14
dislocation mobility and lower flow stress would be induced by electric current, which

enhance the plasticity of the metals.

Material processing at elevated temperature

Hot working

Hot working process metals are plastically deformed at elevated temperature, which

is usually higher than their recrystallization temperature. The higher temperature is used

to promote dislocation annihilation and mitigate work hardening. Therefore, many

traditional cold working techniques are performed at high temperature, such as hot

forging [75], warm rolling [46], warm equal channel angular extrusion [43], and hot

drawing [76]. It has been approved that material processing at elevated temperature

could improve the plasticity of alloys, especially for hard-to-deform alloys. Thus, reduced

the flow stress during the hot working will improve the formability of the processing alloy

and mitigate crack formation.

Figure 2.3 Schematics of hot working processes.

15
 High temperature surface severe plastic deformation

High temperature Ultrasonic Nanocrystal Surface Modification (HT-UNSM) has been

studied by A. Amanov et al. [77–79]. It is found that the surface hardness and compressive

residual stress were further increased. The increase in hardness is mainly attributed to

the grain size refinement. The surface became smoother after HT-UNSM. As a result, HT-

UNSM treatment was found to be beneficial to improving the mechanical and tribological

properties of Ti-6Al-4V alloy more systematically.

C Ye et al. [80–84] investigated effect of warm laser shock peening (WLSP) on plastic

deformation behavior of aluminum alloys and cast copper. According to their

experimental and FEM simulation work, it is found that the surface deformation,

equivalent plastic strain, and depth of residual stress distribution increase with increases

with increasing laser intensity and temperature.

Electrically-assisted material processing

Troisskii [85] reported a phenomenon in which applying an electric current during

deformation reduced the flow stress and increased the elongation of metals in 1960s.

After that, electric current has been used in many manufacturing and processing

techniques [86]. Salandro et al. [51] investigated various factors affecting the electrically-

assisted forming (EAF) process. They found that EAF could reduce the flow stress in

deformed samples due to an electroplastic effect. In addition to the mechanical behavior,

the microstructure evolution of the alloy was also studied after the EAF process. They

16
found that the electroplastic effect is not due solely to the resistive heating, but a more

complex phenomenon happens as electricity assists the deformation.

Wang and Ye et al.[70,87–89] applied pulsed current on ultrasonic surface rolling

process (USRP) on 304 stainless steel, pure Ti, and Ti64 alloys. It is found that the

introduction of pulsed current can facilitate surface cracks healing and increase surface

hardness and compressive residual stress, which is because of further refined grains and

enhanced plastic deformation introduced by coupling effect of USRP and electropulsing.

They attributed these effects to accelerated dislocation mobility and atom diffusion

induced by electropulsing.

Laser-assisted material processing

Laser assisted machining (LAM) [90] has been considered as an alternative process for

hard-to-deform metallic alloys. During LAM processing, laser beam is used to heat only

the under working area. The flow stress decreases in this selected area with elevated

temperature by laser and the material’s behavior changes from brittle to ductile.

Previously, laser has been used in turning, milling, burning, dressing, grinding, and drilling

processes. [91]

Tian and Shin [92] developed a laser-assisted burnishing (LAB) process that integrates

laser heating with burnishing. The localized heating induced by the laser beam was

reported to improve metal plasticity and make the burnishing treatment more effective.

As compared to room-temperature burnishing, LAB results in higher hardness in the

surface layer of the sample due to enhanced plastic deformation and work hardening.

17
 Chapter summary

In this chapter, the strengthening mechanisms of SSPD and issues of hard-to-deform

alloys are reviewed. The dislocation theory of plastic deformation and promotion of

dislocation mobility by energy input are discussed. In order to improve the plasticity of

hard-to-deform alloys and effects of SSPD on hard-to-deform alloys, high dislocation

mobility is need, which requires extra energy input. Finally, the related work on energy-

assisted machining in the literature is reviewed.

18
 3
CHAPTER III

MATERIAL CHARACTERIZATION METHODS USED IN THIS DISSERTATION

Hardness test

The surface hardness was measured from the top surface of the samples using a

Tukon 1202 Vickers hardness tester with a load of 50 g and a holding time of 10 seconds.

Each value of surface hardness represents the average of 5 measurements.

Surface topography

The surface finish after SSPD and EASSPD were characterized using a Zygo NewView

7300 white-light optical profilometer (Figure 3.1). For x25 magnification, the scanning

area was 2.80 mm by 2.1 mm, for x50 magnification, the scanning area was 1.40 mm by

1.05 mm, for 400× magnification, the scanning area was 0.18 mm by 0.13 mm. Each value

of surface roughness represents the average of 5 measurements.

19
 Figure 3.1 Zygo NewView 7300 white-light optical profilometer in the Department of

Civil Engineering, University of Akron.

Microstructure

The cross-section of the samples was mechanically ground to 1200 grits, polished to

1 μm, cleaned with acetone, ethanol, distilled water, respectively and etched with Kroll's

reagent (2 ml of HF, 6 ml of HNO3, and 92 ml of distilled water) before microstructure

characterization. The cross-section microstructure was characterized using a Tescan

LYRA-3 Model XMU FIB-FESEM system.

20
Figure 3.2 Tescan LYRA-3 Model XMU FIB-FESEM system in NCERCAMP, the University of

Akron.

XRD characterization

In order to identify the phase evolution after UNSM and LA-UNSM, X-ray diffraction

(XRD) was performed using a Rigaku Ultima IV X-ray diffractometer with a Cu-Kα radiation

source (λ = 1.5418 Å) operating at 40 kV and 35 mA. The scanning angle ranged from 20 º

to 70 º in 2θ with the scanning rate of 1.0 º / min.

Porosity by micro-CT and surface profiler

Skyscan 1172 Micro Computer Tomography Scanner (Micro-CT) was applied to scan

the samples to analyze the pores size distribution and porosity with different treatment.

The X-ray source worked at 100 kV and 100 µA with an Al-Cu filter and the resolution was

5 µm. All the three samples were cut and ground to around 350 µm of the top surface

layer. More X-ray being absorbed by alloy than pores feasible to be observed. Sample was

mounted on a rotary stage and rotating over 360 ° in steps of 0.4 ° during scanning. The

21
Zygo NewView 7300 surface profiler was also used to complement the Micro-CT

characterization. According to the thickness difference between and after polishing, 100

µm of material on the surface was removed. The polished samples were scanned by the

surface profiler to calculate the porosity.

Figure 3.3 Skyscan 1172 Micro Computer Tomography Scanner in National Polymer

Innovation Center, the University of Akron.

Residual stress measurement

A Proto LXRD system was used to measure the residual stresses. Residual stresses

were analyzed in two orthogonal directions, denoted as X and Y, using the sin2ψ

technique with electrolytic layer removal. The dimension of the sample used for residual

stress measurement was 20 mm × 20 mm × 6 mm. Figure 3.4a shows the measurement

positions, and Figure 3.4b shows an image of the sample during measurement. A 2-mm

aperture size was used in combination with a 3-second exposure time to achieve

adequate intensities. The d-spacing/strain measurements were made using the {213} α-

22
Ti diffraction peak, and the X-ray elastic constants S1213 and S2213/2 were measured as

-2.83*10-6 MPa-1 and 11.89*10-6 MPa-1, respectively, in accordance with ASTM E1426-14

standard (Determining the X-Ray Elastic Constants for Use in the Measurement of

Residual Stress Using X-Ray Diffraction Techniques) [93]. Calibration of the equipment

was performed before each set of measurements using a standard titanium sample in

accordance with ASTM E915-16 standard (Standard Test Method for Verifying the

Alignment of X-Ray Diffraction Instrumentation for Residual Stress Measurement) [94].

Through-the-depth residual stresses were measured by electropolishing the surface

treated region in a layer-by-layer fashion using a solution composed of 86.6 volume

percent methanol, 12.4 volume percent sulfuric acid, and 1.0 volume percent hydrofluoric

acid. A 30-µm step size for layer removal was used in the first 150 µm for detailed mapping

of stress fields in the near surface regions; the step size was increased to a 50-µm step for

the depths beyond. Stress gradient and layer removal corrections were carried out in

accordance with SAE J784a (Residual Stress Measurement by X-Ray Diffraction) [95]. Each

data point represents the average of the three measurements at Points 1, 2, and 3 (Figure

3.4a).

23
 (a) (b)

Figure 3.4 Residual stress measurement: (a) Schematic of measurement positions

and

(b) a photograph of the sample.

Electron backscatter diffraction/orientation imaging microscopy

Cross sections of samples for electron backscatter diffraction (EBSD) and orientation

imaging microscopy (OIM) examination were prepared by mechanical polishing to 1 µm,

followed by finer polishing with 60 nm colloidal silica mixed with 10% hydrogen peroxide

as an oxidizing agent for more than 10 hours. EBSD scans with a step size of 90 nm were

carried out using a Genesis 4040 EDAX/TSL EDS/EBSD system in the Tescan Lyra 3 XMU

microscope operating at 20 kV.

Transmission electron microscopy (TEM)

Transmission electron microscopy (TEM) and selected area electron diffraction (SAED)

were carried out using an FEI Talos F200XTEMworking at 200kV. The TEM samples were

prepared using the lift-out method in an FEI Versa 3D LoVac FIB-SEM Dual Beam system

equipped with an FEI EasyLift system.

24
 Tensile tests

To evaluate the effect of pulsed current on the plasticity of Ti64, tensile tests were

carried out. Dog-bone-shaped tensile samples with a gauge length of 10.5 mm, a width of

3.5 mm, a thickness of 0.83 mm, and a cross-sectional area of 2.905 mm2 were cut from

a plate using electrical discharge machining. The dimension of the tensile samples was

designed according to the ASTM E8/E8M standard with the gauge length and gauge width

scaled down proportionally. The power supply was connected to both ends of the tensile

samples, and electric insulation was used between the tension clamps and the samples.

A FLIR T650sc infrared (IR) camera was used to monitor the temperature of the samples

during the tensile tests. Similar to the LSP experiments, an oscilloscope was used to

monitor the frequency, peak current, pulse duration and RMS current. The RMS current

densities were kept constant for all electropulsing-assisted tensile tests (at 12.1 A/mm2)

to ensure the heating effect of the electric current remained constant. The electropulsing

parameters used in the tensile tests are given in Table 3.1. In tensile tests assisted by

continuous current, to keep the bulk heating effect constant, the continuous current

density was adjusted to 13.0 A/mm2. The electric current was applied to samples 180

seconds before starting the tensile test to ensure that the temperature was stabilized.

The tensile tests were conducted under a constant strain rate of 0.001 s-1. Testing at each

condition was repeated five times to ensure data reliability.

25
Table 3.1 Electrical parameters used during the tensile tests

Frequency Peak current density RMS current density

Sample
(Hz) (A/mm2) (A/mm2)

DC-LSP n/a 13.0 13.0

EP-LSP-1 750 65.4 12.1

EP-LSP-2 400 87.4 12.1

EP-LSP-3 240 113.6 12.1

EP-LSP-4 170 137.7 12.1

EP-LSP-5 140 165.2 12.1

Chapter summary

In this chapter, the material characterization methods used in this study are briefly

reviewed. These characterization methods will be used in the study in the following

chapters.

26
 4
CHAPTER IV

DICRECT CURRENT ASSISTED UNSM ON 3D-PRINTED TI641

Introduction

As introduced in above chapters, UNSM is a deformation-based surface treatment

process, its processing efficiency is highly dependent on metal plasticity. Because of the

low plasticity of the 3D-printed Ti64, it is challenging to select the optimal UNSM process

conditions to produce the best property enhancement. When higher striking intensity (i.e.,

high ultrasonic amplitude) is used, surface or subsurface cracks will form due to the

intrinsic brittleness of the 3D-printed Ti64. Lower UNSM striking intensity, on the other

hand, may not be sufficient to effectively induce beneficial changes in microstructure. An

electric current can provide thermal energy through Joule heating very efficiently.

Compared with traditional heating, metals can be heated by an electric current in less

time, and electric heating can be easily integrated with the UNSM process. In this chapter,

we employed an innovative process called direct current assisted ultrasonic nanocrystal

1 The work in this chapter was published in [139] Zhang, H., Zhao, J., Liu, J., Qin, H., Ren, Z., Doll, G. L., Dong, Y., and Ye,

C., 2018, “The Effects of Electrically-Assisted Ultrasonic Nanocrystal Surface Modification on 3D-Printed Ti-6Al-4V Alloy,”

Addit. Manuf., 22, pp. 60–68. doi:10.1016/j.addma.2018.04.035. Reuse with permission from Elsevier.

27
surface modification (DC-UNSM) to treat 3D-printed Ti64. The combination of UNSM’s

ability to refine the subsurface grain structure and smooth the material surface with the

electric current’s capacity to improve the plasticity of hard-to-deform alloys formed the

basis for the hypothesis that an DC-UNSM process could be an effective means to

eliminate some or all of the mechanical shortcomings of 3D-printed Ti64. In this study,

DC-UNSM was used to treat 3D-printed Ti64, and the effects of this treatment on surface

finish, subsurface porosity and in-depth hardness were investigated.

Experimental details

Sample preparation

3D-printed Ti64 specimens were fabricated using a ProX200 (3D Systems, Inc.) direct

metal laser sintering (DMLS) machine. The Ti64 powders had an average particle size of

38 μm, and the building layer thickness was ~30 μm. Rectangular prism specimens with

dimensions of 15 mm ×15 mm × 2 mm were fabricated.

DC-UNSM processing

Figure 4.1 shows the schematic of the DC-UNSM process. In the UNSM process, a

tungsten carbide tip scans over a sample while striking the material surface at 20 KHz, and

the scan is controlled by a computer program. In this study, the UNSM experiment was

carried out under the following conditions: a static load of 30 N, an ultrasonic vibration

amplitude of 24 μm, a frequency of 20 KHz, a scanning speed (V1 in Figure 4.1) of 2,000

mm/minute, and an interval (the distance between neighbor scans as shown in Figure 4.1)

of 10 μm. These parameters were chosen based on the results of a previous study [25].

28
During the DC-UNSM process, the temperature of the 3D-printed Ti64 samples was held

at 425°C by manipulating the current density. The sample temperature was monitored

using a thermocouple.

Figure 4.1 Schematic of the DC-UNSM process.

4.3 Results
Surface finish improvement by DC-UNSM

(a) (d)

29
 (b) (e)

(c) (f)

Figure 4.2 The 2D surface profiles and SEM images obtained for 3D-printed Ti64

samples: (a,d) Control; (b,e) UNSM-treated samples; and (c,f) DC-UNSM–treated samples.

Figure 4.2 shows the surface profiles and corresponding SEM images of non-treated,

UNSM-treated and DC-UNSM-treated specimens. For the non-processed control sample,

the surface is quite rough (Figure 4.2a), and numerous pores and particles can be

observed (Figure 4.2d). After UNSM treatment, the sample surface become smoother

(Figure 4.2b), although some small particulates can still be observed on the surface (Figure

4.2e). For the DC-UNSM-treated sample, the surface finish was further improved (Figure

4.2), and no particles were observed on the material surface (Figure 4.2f).

Figure 4.3 shows the roughness Ra values obtained for the control, UNSM-treated and

DC-UNSM-treated samples. The mean Ra is 10.6 μm for the control samples. For the

UNSM-treated samples, the mean Ra was 7.1 μm; for the DC-UNSM–treated samples, the

mean surface roughness was was only 1.3 μm. The much lower surface roughness values
30
for the DC-UNSM sample represents a significant improvement in the surface finish.

During UNSM treatment, the sample surface is subjected to pressing and burnishing

forces from the UNSM tip. These forces push the material from high peaks into low valleys

on the material surface, improving the surface finish. However, for hard-to-deform

materials with low plasticity, a low force will induce insufficient surface modification,

while a high force can lead to surface/subsurface cracking. In DC-UNSM, resistive heating

from the electric current can increase the sample temperature and thus its plasticity,

which makes the sample surface easier to deform. The roughness peaks on the surface

can more easily flow towards the valleys, leading to a much better surface finish.

Therefore, DC-UNSM can improve the surface finish of 3D-printed Ti64 more efficiently

than the conventional UNSM treatment.

Figure 4.3 Surface roughness (Ra) of the control, UNSM-treated and DC-UNSM-

treated 3D-printed Ti64 samples.

31
 Porosity reduction in the surface layer by DC-UNSM

In order to compare the effects of UNSM and DC-UNSM on porosity, one third of a

square sample was processed by UNSM, one third was processed by DC-UNSM, and the

remaining third was untreated and used as a control. Next, the entire sample was polished,

and a 100-µm-thick layer of material was removed to reveal any pores. After that, the

sample was characterized by optical microscopy. Figure 4.4a shows a collage of optical

micrographs, with the left portion of the image from the UNSM sample, the control

sample shown in the center, and the DC-UNSM sample shown on the right. This optical

image was processed with the regionprops function in MATLAB to reveal the pores as

block dots in Figure 4.4b. It can be observed that the control sample has many pores and

the UNSM sample has sporadically distributed pores, while the DC-UNSM sample is

apparently free of pores. The image was then processed using the Regionprops function

in MATLAB to produce a histogram of the pores of different sizes. The histogram for the

control, UNSM-treated and DC-UNSM–treated regions is shown in Figure 4.4c. It can be

observed that the number of pores was significantly reduced after UNSM treatment. For

the DC-UNSM-treated region, no pore larger than 5 µm2 was identified, and the number

of pores smaller than 5 µm2 was reduced significantly.

b)

32
 (a) (b)

(c)

Figure 4.4 Change in porosity of the 3D-printed Ti64 for control, UNSM-treated and

DC-UNSM-treated samples: (a) optical micrograph, (b) processed grayscale image, and

(c) histogram of pores of different sizes.

Micro-CT was also used to analyze the porosity in 3D-printed Ti64. Figure 4.5 shows

the micro-CT image with the control sample on the left, the UNSM-treated sample in the

center, and the DC-UNSM-treated sample on the right. According to the CT images, the

number of pores is lowest in the DC-UNSM processed sample and highest in the control

sample, which is consistent with the observations made during optical microscopy. After

reconstruction and calculation, the porosity is about 1.74 % in the control sample, 1.16 %

in the UNSM sample, and 0.82 % in the DC-UNSM sample. It should be noted that it is
33
typically not practical to directly compare the porosity data from micro-CT observations

to the observations from optical microscopy. First, the optical microscopy observations

were only carried out at a single depth (100 µm below the surface) while the micro-CT

data are averages of specific pore volumes. Second, the porosity analysis in micro-CT is

highly dependent on the threshold size for determining whether a pore exists.

Nevertheless, qualitatively, the micro-CT observations are consistent with the optical

microscopy results. From these observations, it is concluded that UNSM can significantly

decrease the porosity in 3D-printed Ti64, but DC-UNSM is more effective. During UNSM

treatment, the compaction effect induced by the peening of the UNSM tip makes larger

pores smaller and smaller pores vanish. This compaction effect is even more pronounced

in DC-UNSM, as the material flowability increases with temperature.

(a) (b) (c)

(D) (e) (f)

34
 (g) (h) (i)

Figure 4.5 Micro-CT plane view images of the 3D-printed Ti64 samples: (a) control,

(b) UNSM-treated and (c) DC-UNSM–treated samples. Cross sections of 3D-printed Ti64

samples: (d) control, (e) UNSM-treated and (f) DC-UNSM–treated samples. 3D

reconstructed CT pictures of 3D-printed Ti64 samples: (g) control, (h) UNSM-treated and

(i) DC-UNSM–treated samples.

Thicker deformation layer by DC-UNSM

Figure 4.6 compares the cross-sectional microstructure of the control, UNSM-treated

and DC-UNSM-treated 3D-printed Ti64 samples. For the control sample, due to the rapid

cooling during the direct metal laser sintering (DMLS) process, the microstructure is

mainly composed of α' martensite, with the typical needle shape of an additive-

manufactured Ti64 [96,97]. The α' martensitic microstructure at the surface of the UNSM-

treated sample is difficult to distinguish, which suggests that the top surface has been

severely deformed. The plastically deformed layer (as indicated by the area between the

black dotted lines) of the DC-UNSM treated sample (Figure 4.6b) is much thicker than in

35
the UNSM-treated specimen (Figure 4.6c). This is probably caused by the high

deformation temperature during the DC-UNSM process. As the deformation temperature

increases, metal plasticity increases. At the same ultrasonic amplitude, DC-UNSM induced

a much deeper plastic-affected depth as compared to that induced by UNSM.

(a) (b)

(c)

Figure 4.6 SEM images of cross sections of 3D-printed Ti64: (a) control,

(b) UNSM-treated and (c) DC-UNSM–treated samples.

XRD patterns evolution

Figure 4.7 shows XRD patterns obtained for the control, UNSM-treated and DC-UNSM-

treated samples. Compared with the patterns for the control sample, all peaks became

broader and lower after UNSM, except for the peak at (002). Plastic deformation leads to

grain refinement and higher dislocation density, resulting in the broadening of the XRD

peaks. The higher (002) peak could be a result of preferred orientation of the grains in

36
this direction following UNSM treatment. For the DC-UNSM sample, the peaks become

narrower and higher, which represents a lower defect density. This could be the result of

the thermal annealing effect caused by the high temperature (420 °C) used in the DC-

UNSM process.

Figure 4.7 XRD patterns for the 3D-printed Ti64.

The width and intensity of the XRD peaks also changed after UNSM and DC-UNSM

treatment. It can be observed that all XRD peaks broadened after UNSM treatment, while

they became narrower and sharper after DC-UNSM treatment. The peak broadening of

UNSM-treated sample is related to the reduction in grain size and the increase in

dislocation density due to severe surface plastic deformation. The sharpening of the

peaks of the DC-UNSM-treated samples is related to the thermal effect of electric current

that drives dislocation annihilation [77,98–100].

Al2O3 generation by DC-UNSM

Figure 4.8 shows the oxygen, aluminum, and titanium contents of the cross sections

for the control, the UNSM-treated and DC-UNSM-treated samples. The high oxygen
37
content (Figure 4.8a) in the control sample may come from oxidation during the DMLS

process. In the UNSM process, a very thin layer of surface material was removed as the

tungsten carbide tip scanned over the material. However, the ultrasonic peening process

can also induce surface oxidation in Ti64, as reported by Vasylyev et al. [101]. The overall

effect makes the surface oxygen content of the UNSM-treated sample slightly lower than

that of the control sample. For the DC-UNSM process, the high process temperature leads

to higher oxygen content on the sample surface due to thermal oxidation. In all three

samples, the oxygen-enriched zone is less than 2 μm thick. The thickness is very similar to

thermally oxidized Ti64, as reported by Garcı ́a-Alonso [102]. Figure 4.8b compares the in-

depth concentration of aluminum. A higher aluminum concentration was observed on the

surface of the DC-UNSM sample. According to Dai et al. [103], the formation and growth

of Al2O3 and TiO2 begin at the same time in Ti alloys. However, the growth rate for TiO2 is

much higher than that for Al2O3 due to the much higher Ti content and the lower growth

activation energy of TiO2. Thus, a titanium oxide layer forms on the very top surface, and

an Al-enriched zone forms beneath the TiO2 oxide layer as a result of titanium depletion.

Due to the lower content of aluminum, the thickness of the Al-enriched zone is only 0.5

μm. Figure 4.8c and 8d compare the Ti and V concentrations for the three samples. Due

to the higher aluminum concentration, the Ti and V concentrations are lower in the top

surface of the DC-UNSM sample, and they gradually increase with the increase in depth.

38
 (a) (b)

(c)

(d)

Figure 4.8 In-depth element concentration for control, UNSM-treated and DC-

UNSM-treated samples: (a) oxygen, (b) aluminum, (c) titanium, and (d) vanadium.

Hardening effect by DC-UNSM

The surface hardness of the control, UNSM-treated and DC-UNSM-treated samples is

shown in Figure 4.9a. The mean Vickers hardness at the surface increased from 359.5 ±

17.3 HV to 437.8 ± 14.2 HV after UNSM treatment. Grain refinement and work-hardening

due to plastic strain probably played a major role in this improvement in hardness [11,22].

For the DC-UNSM-treated sample, the surface hardness further increased to 484.5 ± 11.9

HV. Since it is known that measured hardness is affected by porosity [104], the increased

39
hardness in the DC-UNSM sample can be, at least partially, attributed to the sample’s

lower porosity. In all three samples, the hardness depth profile in Figure 4.9b exhibits

fluctuations (100 HV). These fluctuations may be caused by structural inhomogeneities

created during the 3D-printing process.

(a) (b)

Figure 4.9 (a) Surface hardness and (b) in-depth hardness of control, UNSM-treated

and DC-UNSM-treated 3D-printed Ti64 samples.

Discussion

Ti64 is an alloy that is difficult to deform, and the high defect density and porosity of

3D-printed Ti64 make it even more so. Although the UNSM process can modify the

material, the plastic-affected depth is quite shallow (Figure 4.6b). However, the direct

current applied during the DC-UNSM process increases the plasticity by elevating the

sample temperature. At an elevated temperature of 420°C, the mobility of dislocations

increases, which can promote dislocation movement [89].

40
 Figure 4.10 Schematic diagram of a 3D-printed metal subjected to DC-UNSM.

Figure 4.10 shows a schematic diagram of the DC-UNSM process for a 3-D printed

metal. Under electric current, high-frequency peening of the UNSM tip generates a

thicker surface severe deformation layer on the specimen surface. During processing, the

higher temperature makes it easier for the material in the surface peaks to flow towards

and into the surface valleys, thereby creating a better surface finish. Due to improved

plasticity, a larger number of pores are compressed and closed by the improved atom

diffusivity at elevated temperatures. Smoother surfaces and lower subsurface porosity

will certainly lead to better mechanical properties in the treated material, and this will be

confirmed in a future work.

In order to investigate how resistive heating of DC-UNSM affects the porosity

reduction in Ti64, the electric current flowing through a Ti64 matrix containing several

spherical pore diameters was simulated by a numerical simulation implemented in

MATLAB. An electric current of 200 A was introduced to a Ti64 sample with a length of 20

mm, resulting in an electric field of 50 V/m. The electrical properties of Ti64 used in the

41
simulation are as follows: the conductivity σ is 5.8 × 105 S/m, the density ρ is 4,500 Kg/m3,

and the specific heat capacity cp is 0.54 × 103 J/(Kg∙K). The simulation domain was 500 ×

500 μm with a mesh size of 1 μm. Pores with diameters of 10, 30, and 50 μm were added

to the matrix. The electric current density in the simulated material was obtained by

solving Kirchhoff’s circuit laws,

∇∙𝑱=0 Eq. 4.1

where J is the current density,

𝑱 = σ𝑬 = −σ∇V Eq. 4.2

therefore

∇ ∙ (σ∇V) = 0 Eq. 4.3

where V is the electric potential and σ is the conductivity, which was set to 0 for the pores.

A cellular automata model [105] was applied to track the interface between the pore and

the matrix. After obtaining the current density, the heating rate was obtained using the

equation

𝑱2
𝑆= Eq. 4.4
𝜎𝑐𝑝 𝜌

where cp is the specific heat capacity and ρ is the density.

The current density flow in the vector form and the magnitude of the electric current

density is plotted in Figure 4.11a and b. It is evident that the current flow is distorted by

the pores and the current density becomes unevenly distributed in the surrounding area,

since the current needs to flow around the non-conductive pores. As a result, for the left-

to-right current flow shown in these figures, the current density is enhanced above and
42
below the pores but is weakened in front of and behind the pores. It has been reported

that electric currents can induce crack healing [106]. The critical current density (jc) that

can induce crack healing can be estimated by the following equation [106],

8 2𝑎2 𝜎𝐴
𝑗𝑐 ≈ (𝑐+𝑑)2 ∙ √ Eq. 4.5
𝐸𝜇

where c and d are the length and the width of the sample (two dimensions in Figure 4.11),

a is the half-length of the crack, E is the elastic modulus, and σA is the applied tensile

stress. For a crack with a crack length of 200 μm in a 15 × 15 mm sample under an applied

stress of σA of 1 GPa (which should be the highest approximation for the UNSM-treated

sample), the critical current density is about 3.3 A/mm2. It is clear that the current density

in Figure 4.11b satisfies this criterion. This means that the applied current could

contribute to crack healing in the 3D-printed Ti64 that has been subjected to DC-UNSM.

The enhanced current density also leads to localized joule heating as indicated in

Figure 4.11c, which further enables UNSM to decrease the pore size/number of a treated

metal. Of particular interest is that the localized heating can be further magnified when

two pores are close to each other. In the presence of two adjacent pores, the electric

current is squeezed into the region between the two pores, making the localized heating

more significant. This can also contribute to the compression effect to the pores that is

imposed by DC-UNSM: the heated regions around the pores flow more easily under

UNSM, and the pores close more effectively. This can partially explain why the DC-UNSM

process is more efficient in reducing porosity than using the UNSM process by itself.

43
 (a) (b)

(c)

Figure 4.11 Ti64 matrix with pores treated using DC-UNSM: (a) The electric current

flow, (b) the electric current density (in A/mm2) and (c) the heating rate (in K/s).

Chapter summary

In this chapter, 3D-printed Ti64 samples were treated by UNSM and DC-UNSM.

Compared with conventional UNSM treatment, DC-UNSM was found to be more efficient

in improving the surface finish and eliminating pores in the 3D-printed metals. The severe

plastic deformation layer produced by DC-UNSM was much thicker than the layer

produced by UNSM, which is consistent with the higher metal plasticity at elevated

44
processing temperatures in the DC-UNSM process. Additionally, the surface hardness of

DC-UNSM–treated samples was greater than the hardness values obtained for the control

and UNSM-treated samples. This observation is consistent with the view that the stress

field created by ultrasonic peening can penetrate deeper into hot Ti64 than the stress

field generated in cold Ti64 when using the conventional UNSM process. Modeling

indicated that the electric current density creates inhomogeneous and localized heating

around the pores. The combination of thermal and mechanical effects appears to work

synergistically to close the pores during the DC-UNSM process. Therefore, the DC-UNSM

appears to be a very efficient post-processing technique for enhancing the fatigue

performance of 3D-printed Ti64, even though more robust fatigue testing is still needed

to confirm this finding.

45
 5
CHAPTER V

ELECTROPULSING ASSISTED LSP ON TI642

Introduction

Even though electropulsing has been reported to significantly improve the efficiency
of many deformation processes, including turning [55,56], rolling [58,61] and ultrasonic

treatment [64,65]., the basic physics mechanisms of how pulsed current affects metal

plasticity and thus the deformation process still merits further investigation. It is not clear

whether it is a pure thermal effect [107,108] or if there exists additional effects [109–111].

The aim of this chapter is to investigate the effect of electropulsing-assisted LSP (EP-

LSP) on the microstructure evolution and improvement of mechanical properties of Ti64

alloy to explore the basic physics mechanism in metals subjected to simultaneous

electropulsing and high strain rate plastic deformation. EP-LSP was used to treat Ti64 alloy,

and its effects on surface finish, hardness, in-depth hardness, and microstructure

2 The work in this chapter was published in [140] Zhang, H., Ren, Z., Liu, J., Zhao, J., Liu, Z., Lin, D., Zhang, R., Graber, M.

J., Thomas, N. K., Kerek, Z. D., Wang, G.-X., Dong, Y., and Ye, C., 2019, “Microstructure Evolution and Electroplasticity in

Ti64 Subjected to Electropulsing-Assisted Laser Shock Peening,” J. Alloys Compd., 802, pp. 573–582.

doi:10.1016/j.jallcom.2019.06.156. Reuse with permission from Elsevier.

46
evolution were investigated and compared with direct current-assisted LSP (DC-LSP). To

study whether the effect of pulsed current on the LSP process is purely thermal or not,

the thermal effect from all sets of pulsed current and continuous current when using the

same effective current density are compared. To understand how pulsed current affects

metal plasticity and thus the EP-LSP process, tensile tests were carried out to evaluate the

plasticity of Ti64 subjected to pulsed current and continuous current having the same

thermal effect.

Experimental details

Materials

Samples having dimensions of 10 mm × 10 mm × 0.8 mm were cut from a grade 5

titanium plate purchased from McMaster-Carr. The Ti64 plate has a nominal chemical

composition of 5.50 – 6.75 wt% aluminum, 3.5 – 4.5 wt% vanadium, 0.4 wt% max. iron,

0.08 wt% carbon, 88.10 – 90.92 wt% titanium and 0 – 0.3% other components. The

samples were abraded with silicon carbide sandpaper up to 1200 grit, followed by final

polishing with a 1-μm diamond suspension. Mirror-like polished samples were cleaned in

an ultrasonic cleaner using acetone, ethanol and deionized water in sequence prior to LSP,

DC-LSP or EP-LSP treatment.

47
 EP-LSP experiment

Figure 5.1 Schematic of the EP-LSP process.

Figure 5.1 presents a schematic of the EP-LSP process. All LSP experiments were

carried out using a Continuum Surelite neodymium-doped yttrium aluminum garnet (Nd:

YAG) laser with a wavelength of 1064 nm and a pulse duration (full width half maximum)

of 5 ns. The laser intensity used was 4 GW/cm2, the beam size was 1 mm, the frequency

was 8 Hz, and the overlap was 75% in both the scanning and traverse directions. Nashua

322 multi-purpose foil tape with a thickness of 43 µm was used as the ablative coating.

Borosilicate optical crown glass (BK7 transparent glass) with a thickness of 6.5 mm was

used as the confinement media and was placed over the ablative material.

An in-house generator was used to provide a pulsed current with the appropriate

electropulsing parameters (Table 5.1). The electropulsing parameters – which included

peak current, frequency, pulse duration, and root-mean-square (RMS) current – were

monitored using an oscilloscope. The current pulses were half sine-wave in form, and the

pulse duration was 100 μs. The peak current and frequency were adjusted so that the

48
same RMS current was used. The sample temperature during processing was monitored

using a K-type thermocouple. For DC-LSP, the current density was adjusted to be the same

as the RMS current in EP-LSP to keep the resistive heating effect constant. The electric

current was applied to samples for 180 seconds before LSP to ensure that the

temperature was stabilized.

Table 5.1 Electrical parameters used in the DC-LSP and EP-LSP experiments

Sample Frequency (Hz) iP-P (A/mm2) iRMS (A/mm2)

As-received n/a n/a n/a

LSP n/a n/a n/a

DC-LSP n/a 13.7 13.7

EP-LSP-1 533 85.3 13.7

EP-LSP-2 256 109.7 13.7

EP-LSP-3 221 131.2 13.7

EP-LSP-4 200 145.1 13.7

EP-LSP-5 153 154.7 13.7

49
 Results

Hardening effect by EP-LSP

(a) (b)

Figure 5.2(a) Surface hardness and (b) in-depth hardness for LSP, DC-LSP, and EP-LSP

treatments.

Figure 5.2a compares the surface hardness of the as-received Ti64 alloy subjected to

LSP, DC-LSP and EP-LSP processing. LSP increased the surface microhardness by a

marginal amount, about 1.1%, while DC-LSP treatment increased the surface hardness by

4.5%. Utilizing EP-LSP, the surface microhardness showed a considerable increase and

improved by 13.4% for sample EP-LSP5. In addition, it can be observed from Figure 5.2a

that hardness increases with the peak current density for all EP-LSP cases, which indicates

that increased peak current density could decrease the flow stress and generate greater

plastic strain when the LSP intensity is constant (4 GW/cm2). EP-LSP5 was identified as the

optimal EP-LSP case and will be used for all future discussion. According to the in-depth

hardness shown in Fig 2b, EP-LSP also results in a deeper hardened layer compared with

that for DC-LSP. The hardened layer for the EP-LSP sample has a thickness of 80 µm, while

that for the DC-LSP sample is only about 25 µm. The lower flow stress is responsible for
50
the deeper plastic-affected depth for the EP-LSP sample. Compared with DC-LSP, EP-LSP

results in both higher surface hardness and deeper hardening layer even though the

heating effect from electric current is the same. This means that the EP-LSP effects (i.e.,

higher surface hardness and deeper harden layer) cannot be fully explained by the

thermal effect from resistive heating.

Surface roughness increase by enhanced plastic deformation

Figure 5.3 shows the surface roughness before and after different LSP treatments. It

was found that the LSP-processed Ti64 samples maintained a similar level of roughness

to that of the control samples, which indicates that LSP at room temperature did not

introduce much plastic deformation to the sample surface. However, after DC-LSP and

EP-LSP, the roughness of the samples increased significantly (from ~0.05 µm to ~0.30 µm),

due to peening-induced indents and patterns. This indicates that both continous and

pulse current could enhance the degree of plastic deformation induced by LSP.

51
 Figure 5.3 Surface roughness at 25x and 100x magnifications after different

treatments.

XRD patterns evolution

Figure 5.4 shows the X-ray diffraction (XRD) patterns for Ti64 samples subjected to

different treatments. The as-received material exhibited several α-Ti peaks and β-Ti peaks.

A new peak at 44.8° with very low intensity was observed in the XRD patterns of DC-LSP

and EP-LSP samples. It was not present in the control or the LSP samples. This low

intensity peak indicates the possible presence of precipitates within the material after DC-

LSP and EP-LSP treatment. This is consistent with the study by Liu et al. [112], in which

they found that the application of continuous current to the ultrasonic surface

modification process induced nanoscale precipitates in Ti64. Ye et al. [81,84] also found

that warm laser shock peening could create high density nanoscale precipitates caused

by dynamic strain aging and dynamic precipitation. The thermal energy added to the

52
material increases the chemical driving force for precipitation. The local joule heating at

defects likely assisted the precipitation process.

Figure 5.4 XRD patterns of the Ti64 samples subjected to LSP, DC-LSP, and EP-LSP.

Cross-sectional microstructure after EP-LSP

The cross-sectional microstructure of the Ti64 samples with different LSP treatments

were observed and analyzed using SEM, as presented in Figure 5.5. Typical microstructure

of the cast Ti64 alloy can be observed with dark equiaxed α phase grains and light β phase

embedded in the α phase grain boundaries. Note that no significant microstructure

difference was observed from the SEM images for different LSP treatment. According to

Lou et al. [113], only when the shock pressure generated by LSP is larger than the

nanograin formation threshold or the critical driving force of deformation induced

martensite can LSP induce significant microstructure evolution, such as grain refinement

or phase transformation. Due to the plastic deformation induced by different LSP

processes, a considerable number of dislocations and nano-twins were generated at the

53
top layer of the processed alloy. EBSD and TEM were used to further characterize the

grain distortion and investigate the dislocation structures present in the samples that

were not resolvable by SEM.

(a) Control (b) LSP

(c) DC-LSP (d) EP-LSP

Figure 5.5 Cross-sectional microstructure of Ti64 samples: (a) control, (b) LSP, (c) DC-

LSP, and (d) EP-LSP.

54
 Misorientation induced by EP-LSP

Figure 5.6a–f show the EBSD images and the kernel average misorientation (KAM)

charts of Ti64 samples processed using different LSP treatments. Since no significant

difference in grain size before and after different LSP treatments was observed in the

EBSD images, the KAM method was used to evaluate the degree of plastic deformation.

KAM is a measure of the average misorientation of a point with respect to a selected

number of its nearest neighbors. A larger KAM value indicates a higher misorientation

and, thus, a higher degree of plastic deformation. Figure 5.6d, e and f show the KAM

distribution of the LSP, DC-LSP and EP-LSP samples, respectively. It can be observed that

the KAM values of both the DC-LSP and EP-LSP samples are higher than that of the LSP

sample, with the EP-LSP sample having the highest KAM value (average KAM 1.125).

Higher misorientation in the DC-LSP and EP-LSP samples indicates that more plastic strain

was introduced through DC-LSP and EP-LSP, with the highest plastic strain generated by

EP-LSP. Figs. 6g, 6h, and 6i show the KAM maps for the LSP, DC-LSP and EP-LSP samples,

respectively. Higher and more uniform misorientation can be observed in the EP-LSP

sample as compared with that in the DC-LSP sample, which is consistent with the KAM

charts. This indicates that both continuous current and pulsed current can increase the

plastic strain induced by LSP. In addition, it was found that pulsed current is more

effective in improving the processing effectiveness of LSP compared with continuous

current having the same bulk heating effect.

55
 (a) (b) (c)

(d) (e) (f)

(g) (h) (i)

Figure 5.6 Inverse pole figures (IPF) from EBSD observations in the cross sections of
the samples (where all top edges correspond to the top surface of the samples) for (a)
controls, (b) DC-LSP,

56
 (c) EP-LSP; KAM charts and maps corresponding to the (d, g) control, (e, h) DC-LSP, and

(f, i) EP-LSP samples; the inset in (g) is the legend for three KAM maps.

To further study the effect of EP-LSP on surface texture as compared to DC-LSP, pole

figures in the (0 0 0 1) α and (1 1 0) β directions were plotted in Figure 5.7. The maximum

intensities of different pole figures are listed in Table 5.2. It can be found for α phase, the

maximum intensity of pole figure increases after DC-LSP and EP-LSP. The DC-LSP sample

shows higher maximum intensity for the α phase than the EP-LSP sample. For the β phase,

the maximum intensity of pole figure decreases after DC-LSP and EP-LSP processing, and

EP-LSP induces the lowest maximum intensity in the β phase. It is known that plastic

deformation introduced by LSP can locally change the initial texture in the top surface

layer of Ti64 alloy, and the β phase, which has a body-centered cubic (BCC) crystal

structure, can be more easily deformed as compared with the α phase, which has a

hexagonal close packed (HCP) crystal structure. Thus, grain misorientation in the β phase

is the main reason for the decrease in the maximum intensity. Lower maximum intensity

of the β phase in the EP-LSP sample is attributed to the larger grain misorientation

induced by EP-LSP than the misorientation induced by DC-LSP, which is also the reason

for the lower maximum intensity of the α phase in the EP-LSP sample as compared with

the DC-LSP sample. A similar result was reported for electropulsing-assisted ultrasonic

surface rolling process [70]. As for the increase in maximum intensity of α phase after DC-

LSP and EP-LSP, one possible reason is that the increase in maximum intensity induced by

57
grain recovery and grain growth (at elevated temperatures by using an electric current)

[114] is larger than the decrease in maximum intensity induced by grain distortion, which

results from the low plasticity of the α phase present in the surfaces of these samples.

(a) (b) (c)

Figure 5.7 Pole figures from EBSD observation of the cross sections (where all top

edges correspond to the top surface of the samples) of the (a) control, (b) DC-LSP, and

(c) EP-LSP samples.

Table 5.2 Maximum intensity of different pole figures for Ti64 samples before and after

different LSP treatments.

Control DC-LSP EP-LSP

{0 0 0 1} α 11.399 17.182 13.430

{1 1 0} β 10.766 5.026 4.485

More dislocations and deformation twins induced by EP-LSP

The TEM images of the top surface layers of Ti64 samples processed using LSP and EP-

LSP are shown in Figure 5.8. Dislocation walls and dislocation tangles can be observed in

both samples. Limited deformation features such as dislocations and twin boundaries are

observed Figure 5.8b and c) in the LSP-processed sample, while some bending contours
58
exist at the top layer (as indicated by the red arrow in Figure 5.8a) due to the TEM sample

preparation. The generation of such multi-directional mechanical twins (MTs), which can

be seen in Figure 5.8c, is in good accordance with that observed in pure titanium

subjected to multiple rounds of LSP processing [115]. The low density for the dislocations

and twin boundaries indicates that the degree of plastic deformation is quite low in the

LSP-processed sample. Because the stacking fault energy of the α phase is lower than that

of the β phase, twins can be easily formed in the α phase of the Ti64 alloy [116]. In

contrast, in the images in Figure 5.8g-i, a higher density of dislocations was observed in

the EP-LSP sample. Not only can planer dislocations [117] be observed in the image shown

in Figure 5.8g, but also dense dislocation tangles; dislocation walls and dislocation cells

can be observed in the images in Figure 5.8e and f. The increased dislocation activity

might be caused by the higher dislocation mobility as a result of electropulsing [118–122].

Once the twins are no longer able to accommodate the plastic strain, dense dislocations

are generated near grain boundaries and twin boundaries, which have higher strain and

easily act as dislocation nucleation sites. In addition, similar to the observations of Lainé

et al. [117], sub-grain boundaries formed from complex arrays of dislocations can be

observed in Figure 5.8i, which reveal the higher distortion inside the grains of the EP-LSP

sample as compared with that of the LSP sample. This finding is consistent with the EBSD

results and further confirms that electropulsing can enhance the degree of plastic

deformation when the LSP intensity is the same.

59
 Figure 5.8 Bright-field TEM micrographs of near-surface microstructures of LSP (a–c)

and EP-LSP-processed (d–i) Ti64 samples. In these images, bending contours are

indicated by red arrows, dislocation walls are indicated by yellow arrows, twin

boundaries are indicated by cyan arrows, dislocation tangles are indicated by green

arrows, planar dislocations are enclosed in red-dashed circles, and sub-grain boundaries

are indicated by white arrows.

60
 Discussion

The effect of peak current density is investigated by processing the samples using the

same RMS current density with different peak current densities. By using the same RMS

current density, the temperature profiles during EP-LSP treatment for all EP-LSP samples

should be the same, as shown in Figure 5.9. According to Figure 5.9, it can be seen that

the sample temperatures reached during DC-LSP treatment and the various EP-LSP

treatments are from 25°C to around 361°C, and the maximum temperatures for the

different treatments range between 350°C and 361°C. Very close heating rates,

temperature histories, and maximum temperatures are reasonable, as the same RMS

current density was applied in all treatments. Thus, the thermal effects for the different

treatments should be the same. However, according to the above results, while the

thermal effect is the same, the hardening effect varies according to the current type and

peak current density used in the treatment. These phenomena indicate that the effect of

EP-LSP, once the peak current density exceeds a critical value, goes beyond the thermal

effect.

It is found that when the peak current increases, the surface hardness of the EP-LSP

processed sample increases gradually. As the temperature is the same for all samples, the

thermal contribution to surface hardness increase should also be the same, which would

indicate that the peak current density is responsible for the higher hardness of the EP-LSP

samples. It is possible that pulsed current with a higher peak current density can more

effectively lower the flow stress and thus induce more significant strain hardening that

61
leads to higher hardness. Note that sample EP-LSP1 has a similar hardening effect as

compared to the sample subjected to DC-LSP treatment (Figure 5.2). This suggests that

when the peak current is lower than a critical value, the surface hardness will be close to

the hardness of the samples processed with continuous current using the same RMS

current. It also indicates that when the peak current density is not high enough, its

contribition is purely thermal.

Figure 5.9 Temperature history for samples processed using DC-LSP and EP-LSP.

In order to investigate the effect of pulsed current and continuous current on the

plasticity of the Ti64 alloy, a series of tensile tests were performed. During the tensile

tests, the RMS current densities for electropulsing were kept the same for all tensile tests

(12.1 A/mm2) to ensure the heating effect of the electric current would remain the same

(Figure 5.10a). The peak current densities for the electropulsing varied from 65.4 A/mm2

to 165. 2 A/mm2 to study the effect of peak current density, and the frequency was

adjusted to maintain a constant RMS current density. A continuous current with the same

62
RMS current density (12.1 A/mm2) was also used. However, using a continuous current

with an RMS of 12.1 A/mm2 resulted in a lower temperature than that in the five samples

processed using pulsed current. The lower temperature of the continuous current sample

could result from the lower dynamic electric resistivity for continuous current than that

for pulsed current according to the Drude model. In order to maintain a consistent

thermal effect, the CC current density was adjusted to 13.0 A/mm2. Figure 5.10b shows

the engineering strain–stress curve of Ti64 samples under different peak current densities

with same RMS current density for all EP sets. It was observed that pulsed current can

more effectively decrease the flow stress of Ti64 compared with continuous current, even

though the bulk heating effect is the same. When plotting the tensile strength as a

function of the peak current density in Figure 5.10c, it is interesting to note that the

tensile stress decreases almost linearly with the peak current density. Lower flow stress

indicates higher plasticity and higher dislocation mobility. Even though the RMS current

densities are the same (and thus the thermal effect), higher peak current density can

more effectively increase dislocation mobility and thus more effectively decrease the flow

stress, which is beneficial for deformation-based manufacturing processes. This

observation is consistent with a previous study, in which pulsed current was observed to

induce significantly higher plasticity in AZ31B compared with continuous current [123].

63
 (a) (b)

(c)

Figure 5.10 (a) The temperature profile and (b) strain–stress curves of Ti64 samples

subjected to continuous and pulsed current, and (c) the tensile strength as a function of

the peak current density.

Flow stress is the instantaneous stress required to continue plastically deform the

material and to keep the metal flowing. It is known that plastic deformation in metals is

accomplished predominantly through dislocation movement. Electric current is

considered to be able to reduce the stress needed to generate dislocations from

dislocation sources, such as the grain boundary and pre-existing dislocations

[119,124,125]. It was reported that EP can induce localized high dislocation concentration

around the obstacles and thus dissociate the immobile dislocation junctions in a single-

64
crystalline Ge2Sb2Te5 through in-situ TEM [124]. In addition, the effect of pulsed current

on dislocation mobility also affects the plasticity. According to the studies by Gromov et

al. [118,119], pulsed electric current can increase the dislocation velocity, and higher

current density leads to higher dislocation velocity under the same resolved shear stress

and temperature. Since a large number of crystal defects (e.g. grain boundaries,

precipitates and dislocation tangles) are pre-existing or are being generated during plastic

deformation, localized Joule heating from electron-defect scattering [126] could be

significant during the EP-LSP process. Higher energy absorption at crystal defects can

reduce the resistance for dislocations to slip through obstacles like precipitates and grain

boundaries [127–130]. Based on the results in this study, by controlling the effective

current density and varying the peak current density, it is found that a pulsed current with

high peak current density can more effectively lower the flow stress and can thus improve

the effectiveness of LSP treatment even though the thermal effect is the same. It is

believed that when the effective current density is the same, higher peak current density

can more effectively mobilize the dislocations and thus decrease the flow stress [123],

which results in greater work-hardening and thus higher hardness. It is thus believed that

pulsed current with a sufficiently high peak current density is more efficient for increasing

the effectiveness of LSP compared with a continuous current having the same heating

effect.

65
 Chapter summary

In this chapter, it was observed that EP-LSP results in higher surface hardness and a

deeper hardened layer compared with DC-LSP having the same thermal effect. This means

that the effect of pulsed current on the LSP process goes beyond the thermal effect. In

addition, it was found that the effectiveness of EP-LSP increases with peak current density

when the effective current density is constant. Tensile testing was carried out to study

the effect of pulsed current on the plasticity of Ti64. It was found that pulsed current can

more effectively lower the flow stress of Ti64 compared with continuous current having

the same heating effect. In addition, the degree of flow stress reduction increases with

the peak current density even though the effective current density and thus the thermal

heating effect are the same. It is believed that compared with continuous current, pulsed

current with much higher peak current density can more effectively lower the flow stress

and thus results in greater degree of work-hardening, which leads to greater surface

hardness of deeper hardened layer in EP-LSP. This study has demonstrated that the effect

of pulsed current on the plasticity of Ti64 goes beyond the thermal effect. In addition, the

EP-LSP effect cannot be fully explained by resistive heating alone. Even though more in-

depth study is needed to reveal the fundamental mechanisms of EP-LSP, this study serves

as a first attempt to utilize high-frequency short-duration pulsed current to increase the

effectiveness of LSP, or any other deformation-based manufacturing processes, without

requiring a significant increase in sample temperature.

66
 6
CHAPTER VI

LASER ASSISTED UNSM ON 3D-PRINTED TI64

Introduction

Continuous laser is an effective choice since it can temporarily and locally heat the

processing area. Laser-assisted burnishing (LAB) [92,131] combines laser heating and

burnishing process, which could reduce the ratio of feed force to normal force, and

meanwhile significantly improve surface finish, introduce work hardening and

compressive residual stress in the component.

In the chapter, UNSM and laser heating is combined to integrate the advantages of

UNSM to refine subsurface grains, squash subsurface pores, and smooth surface with the

ability of laser to locally heat to improve plasticity of the 3D-printed Ti64. Surface

roughness, hardness, porosity, microstructure, and residual stress of as-received samples,

UNSM processed samples, and LA-UNSM processed samples were characterized and

compared. a finite element model was constructed to investigate the temperate

distribution and its synergistic effect with UNSM on pore closing.

67
 Experimental details

Materials

In this study, 3D-printed Ti64 alloy fabricated using a ProX200 direct metal laser

sintering (DMLS) machine by 3D System, Inc was used. The particle source used to prepare

for 3D-printed Ti64 has an average size of around 37.5 μm, and the thickness of single

building layer is set as 30 μm. The dimension of the samples for residual stress analysis is

20206 mm3, while the rest samples have a dimension of 15153 mm3. Control

sample indicates the as-received 3D-printed Ti64 samples. UNSM and LA-UNSM are

directly applied to the as-received samples, respectively.

UNSM/LA-UNSM

Figure 6.1 shows the schematic of the LA-UNSM process. During UNSM/LA-UNSM, a

tungsten carbide tip (diameter of 2.4 mm) scans over the material surface following a pre-

programmed path, and meanwhile strikes the surface at an ultrasonic frequency (20 KHz).

The UNSM parameters in this study were set according to our previous research [25]: a

static load of 30 N, an ultrasonic vibration amplitude 24 μm, and scanning speed (V1 in

Figure 6.1) 2000 mm/minute, a distance between neighbor scanning track (interval shown

in Figure 6.1) of 10 μm. During the LA-UNSM process, a continuous laser system is applied

as the laser source. The power of the laser is set from 6, 23, 46 watts, respectively. The

processing temperature during UNSM/LA-UNSM processing was monitored using an IR

camera (FLIR T650sc, FLIR1). The IR images were recorded and analyzed using FLIR

ResearchIR 4.

68
 Figure 6.1 Schematic of the LA-UNSM process.

Finite element method (FEM) modelling

In order to investigate the temperature distribution and its synergistic effects with

UNSM on pore closing, a 2-dimensional FEM control model shown in Figure 6.2 was

constructed to simulate the UNSM/LA-UNSM process. Material properties of Ti64 and

Johnson-Cook Model implemented in the ABAQUS were used to define material model

and adopted as constitutive model. Tungsten carbide UNSM tip was simplified as a rigid

semicircle with diameter of 2.4 mm and the scanning speed of the tip along scanning

direction is 2000 mm/min, which is set the same with the experiment part. The 2-

dimensional movement of the tip is displacement-controlled. The tip vibrates in a

sinusoidal wave [132] with the amplitude calibrated from the single-line experiment and

simulation. A moving Gaussian heat source along scanning direction on the top sample

surface synchronously with UNSM tip was used to simulate the laser beam. The bottom

sample surface plate is fixed during UNSM processing. A 2D plate represented the cross-

section of sample with size of 7 mm by 1.5 mm was constructed. The material within a
69
middle region of radius of 60 µm was removed to induce the pore. Quad elements are

used with free mesh tech near the pore and structured tech for surrounding regions and

meanwhile ensure the smooth transition of element shape near pore region. A biased

mesh size was applied to make the balance between computational accuracy and

efficiency. Eventually, there are 5,734 linear quadrilateral elements CPE4T in the model.

The size of element at the rim of the pore is 4.3 µm while that of the farthest element

apart from the pore at bottom corners is 198.6 µm.

Figure 6.2 FEM model of single-line scanning of UNSM on 3D-printed Ti64 with pore.

6.3 Results

Further enhanced surface hardness by LA-UNSM

Hardness of alloys could be improved through grain refinement, working hardening,

etc. In Figure 6.3, the surface hardness of the control, UNSM-processed and LA-UNSM-

processed samples was compared. It can be observed that the average surface Vickers

hardness increased from 359.5 HV on control sample to 437.8 HV after UNSM treatment.

This hardness increase is mainly attributed to the work-hardening and grain refinement

70
induced by cold working from the UNSM [25,133]. After LA-UNSM, the surface hardness

further increased to 465.4 - 479.5 HV with increasing laser power. It is reported that

measured hardness could be affected by porosity [104]. When there are pores at the

surface layer, the measured macro hardness should be lower. Besides, higher plasticity

induced by laser heating enhanced the plastic strain and more strain hardening can be

generated. Therefore, further increased hardness after LA-UNSM processed sample

because of the lower porosity and higher plastic strain.

Figure 6.3 Surface hardness of 3D-printed Ti64 samples before and after UNSM and

LA-UNSM treatment

Surface finish induced by LA-UNSM

Beside the softening effect, with the increase of laser energy, ablation and oxidation

can also occur on the treated surface, which may weaken the which may weaken the

burnishing effect of UNSM on 3D-printed Ti64 sample. Therefore, surface roughness Ra

71
of the control, UNSM and LA-UNSM-treated samples were characterized and compared,

shown in Figure 6.4. It can be observed that the mean Ra decreased from 10.6 to 7.1 μm

after UNSM treatment. During the UNSM processing, the sample surface will be subject

to pressing and extrusion force from the UNSM tip, which can improve the surface finish

by causing the roughness peaks to flow towards the valleys, thus flatten the waviness and

pores on 3D-printed Ti64 surface. While after LA-UNSM, the surface roughness was

further reduced to 4.4, 3.8, and 5.3 μm with laser power of 6W, 23 W, and 46 W,

respectively. The increase in roughness of DC-UNSM-46W may because of the ablation

and oxidation induced by higher laser power. Although surface roughness could be

further reduced by increasing UNSM load, heavy force leads to high possibility of micro-

crack initiation in hard-to-deform materials, especially at high stress concentration region,

such as edge of pores. Since laser can locally increase surface temperature and plasticity,

which makes the sample surface easier to deform and thus reduces the crack initiation.

Higher plasticity can make is easier for mass flow from roughness peak to valley and lead

to much better surface finish. Therefore, LA-UNSM processing leads to better surface

finish of 3D-printed Ti64 compared to UNSM processing only. DC-UNSM-23W is used in

the subsequent experiments and discussion.

72
 Figure 6.4 Surface roughness of 3D-printed Ti64 samples before and after UNSM and

LA-UNSM treatment.

XRD patterns evolution

XRD patterns of the control, UNSM and LA-UNSM-treated samples were shown in

Figure 6.5a, full width at half maximum (FWHM) of each peak was plotted in Figure 6.5b,

and 2θ offset (peak position difference) of identical peaks between UNSM, LA-UNSM and

control samples was calculated and plotted in Figure 6.5c. It can be observed that most

of the peaks broadened and weakened after UNSM compared to those of control samples.

This peaking broadening is attributed to the grain refinement and dislocation propagation

induced by the severe plastic deformation of UNSM. Only the intensity of (002) peak

became higher, which might be a result of preferred orientation of the grains in this

direction after UNSM treatment [25]. For the LA-UNSM sample, the peaks besides (002)

peak become sharpen again which has a medium FWHM values between those of control

73
and UNSM processed samples. One possible reason is that the elevated temperature

induced by continuous laser promotes the movement of the dislocations, which impedes

the dislocation tangling and further propagation [77]. When comparing the 2θ offset of

treated sample, it can be found that all peaks shift to low angle direction in UNSM treated

samples, which is because of the residual stress and oxidation during UNSM processing

[133]. While for the LA-UNSM processing, thermal effect of laser partially recovered the

grain distortion, and thus reduced the 2θ offset or even reversed the 2θ offset.

(a) (b)

(c)

Figure 6.5 (a)XRD pattern of 3d-printed Ti64 after UNSM and LA-UNSM process, (b)

FWHM of XRD peaks, (c) 2θ offset between UNSM, LA-UNSM and control

74
 Thicker deformation layer by LA-UNSM

SEM images of the cross-sectional microstructure of the 3D-printed Ti64 samples

before and after treatment are shown in Figure 6.6. Large number of typical lath-shaped

α' martensitic can be observed in the control sample, which formed because of the rapid

cooling during the DMLS process [96,97]. A subsurface layer with thickness of about 6 µm

with an approximate boundary shown as the black dotted line could be observed in the

UNSM-processed sample, in which the phase becomes blurry while the α' martensitic

microstructure becomes difficult to distinguish. This microstructural evolution indicates

that the top surface has been refined and distorted because of the severe plastic

deformation introduced by UNSM. Compared with the UNSM-treated sample, this

deformed layer is significantly increased to around 20 µm, as shown Figure 6.6c. The

thicker severe plastic deformation layer attributes to the higher deformation

temperature in the LA-UNSM process. The metal plasticity is improved by increased

working temperature, thus lower flow stress makes plastic deformation easier. As a result,

LA-UNSM has an ability to induce deeper plastic-affected depth compared with that by

UNSM with same processing parameters.

75
 (a) (b)

(c)

Figure 6.6 Cross-section SEM images of the 3D-printed Ti64: (a) control, (b) UNSM

and (c) LA-UNSM

Porosity reduction in the surface layer by LA-UNSM

Figure 6.7a-c shows the micro-CT plane view images and reconstructed cross section

images of the control, UNSM and the LA-UNSM treated sample, respectively. In the micro-

CT plane view images, bright particles represent the pores, while dark particles represent

the pores in the reconstructed cross section images. It can be found that the pore quantity

is least in LA-UNSM processed sample and highest in control sample. After reconstruction

and calculation in the post-processing softer, the porosity is about 0.974 % in control

sample, 0.730 % in UNSM sample, and 0.479 % in LA-UNSM sample. It is concluded that

UNSM can decrease the porosity in the subsurface layer of 3D-printed Ti64, while LA-

UNSM further increase this squashing and closing effect. For UNSM treatment only, the

76
compacting and extrusion effect induced by the peening of the UNSM tip makes larger

pores smaller and smaller pores vanished. Under elevated temperature during LA-UNSM

processing, improved material flowability makes this compacting effect is even more

pronounced and remarkable.

(a) (b) (c)

77
 (d)

Figure 6.7 Plane view images and the cross sections of the 3D-printed Ti64 samples:

(a) control, (b) UNSM-treated and (c) LA-UNSM-treated; (d) histogram of pores before

and after UNSM treatment

Residual stress induced by LA-UNSM

(a) (b)

Figure 6.8 In-depth residual stress profile of 3D-printed Ti64 samples (a) along the X

direction and (b) along the Y direction before and after UNSM and LA-UNSM processing.

78
 The UNSM process has been shown to induce significant subsurface compressive

residual stresses [23,134,135]. Figure 6.8 shows the in-depth residual stress distribution

before and after UNSM/LA-UNSM processing along UNSM scanning direction (Y direction)

and its transverse direction (X direction). For the control sample, the top surface shows a

tensile stress of around 700 MPa along the X direction due to the thermal stress

generated during the DMLS process. UNSM process inverts this tensile residual stress to

an even higher compressive residual stress of 1094 MPa at the top surface layer. The

magnitude of the UNSM induced compressive residual stresses decrease gradually with

depth and convert back to a tensile status beyond a depth of 130 µm. A similar trend has

been observed in the Y direction, as shown in Figure 6.8b, though the magnitude of the

compressive residual stress is lower. This is consistent with another investigation, in

which a lower magnitude of compressive residual stresses was observed in the UNSM

scanning direction compared with that in the transverse direction [8]. Compressive

residual stress is another important factor for improving fatigue resistance, since high

compressive residual stress could hinder the propagation of fatigue cracks [136]. For the

surface layer of LA-UNSM processed sample, the compressive residual stress is further

increased to a magnitude of 1245 MPa in the X direction, which is higher than that of the

UNSM-processed sample. It is known that the tensile residual stresses at the surface of

3D-printed alloys come from the rapid cooling. In this experiment, continuous laser also

locally heats the sample and lead to a rapid cooling in air but does not decrease the

compressive residual stress comes from the UNSM process. There might be two possible

79
reasons. Local heating increases the plastic deform lead by UNSM process. After LA-UNSM

processing, the recovery of elastic deformation at the sub-surface layer will further

compress the surface and thus leads to higher compressive residuals stress. In addition,

the local temperature increased by laser is only 173.8 ºC. Even in air, the cooling rate is

much lower than that of the 3d-printing process, which makes the process will not lead

to significant tensile residual stress.

Discussion

Figure 6.9 shows the schematic diagram of the LA-UNSM process for a porous 3D

printed alloy. Under local heating by continuous laser, high-frequency peening of the

UNSM tip generates a thicker surface severe deformation layer on the thermal softened

specimen surface. During processing, the elevated temperature makes it easier for the

material in the surface peaks to flow towards and into the surface valleys, thereby

creating a better surface finish. Due to improved plasticity and reduced flow stress, a

larger number of pores are compressed and closed by the improved atom diffusivity at

elevated temperatures. Ti64 is a hard-to-deform alloy, and the high defect density and

porosity of 3D-printed Ti64 make it even more so. Although the UNSM process can modify

the material, the plastic-affected depth is quite shallow (Figure 6.6b). As aforementioned,

on one hand, local heating has an ability to improve plasticity by elevating the sample

temperature [38], which is the softening effect; on the other hand, increased plasticity

promotes the plastic deformation, thus enhances the working hardening effect. From the

foregoing experimental results, it could deduce that the hardening effect wins the

80
competition. Here, the temperature should be a key factor that determines the output o

LA-UNSM. Therefore, the surface temperature is determined experimentally and

investigate the effect of temperature distribution interior the material on the pores

squashing and closing using FEM simulation. Grain refinement after LA-UNSM is also

investigated.

Figure 6.9 The schematic diagram of LA-UNSM

Local heating effect of continuous laser

Surface temperature in the experiment

Since working temperature plays a key role in the microstructure evolution of a

material, it is important to determine the temperature during the LA-UNSM treatment.

For temperature measurement using IR camera, surface emissivity is a crucial factor to

measure the temperature accurately. A hot plate with a type-K thermal couple was used

to control the sample temperature and compared with the temperature measure by IR

camera under surface emissivity of 1. The actual surface emissivity at different

81
temperature shown in Figure 6.10b-c was calibrated by calculating the ratio of

temperatures measured by IR camera and thermal couple. It is found that the surface

emissivity of 3D-printed decreases with elevating temperature. When the temperature is

lower than 50 °C, the emissivity is larger than 0.81, while when the temperature is

between 150 °C and 300 °C, the emissivity is settled at around 0.64. Figure 6.10a and b

show the IR digital images during the UNSM treatment and LA-UNSM, while the emissivity

was set as 0.9 for UNSM and 0.64 for LA-UNSM, separately. During the UNSM treatment,

the temperature of the processing area is about 39.2 °C, which is a little higher than the

room temperature (25.4 °C) due to the friction and deformation. While the 23 watts

continuous laser is applied to assist the UNSM treatment, the local temperature is about

173.8 °C. Such a low working temperature with rapid scanning speed (2000 mm/min) can

hardly change the hardness of the Ti64 alloy [137] or induce oxide layer on the top surface

[138].

82
 (a)

(b) (c)

Figure 6.10 The IR image during (a) UNSM treatment and (a) LA-UNSM treatment on

3D-printed Ti64, (c) Emissivity of 3D-printed Ti64 at different temperatures.

Temperature distribution in LA-UNSM to close pores

After simulation of LA-UNSM single-line scanning, the temperature distribution along

depth under heat source (laser beam) is plotted in Figure 6.11a. The in-depth temperature

decreases from 173.8 °C at the top surface to 45 °C at depth of 400 µm. Because of

gradually decreased strain and decreased temperature along depth direction, the ability

of LA-UNSM to squash pores is declining. This declining could be observed in Figure 6.11a

83
to 12d as the pore size difference at depth of 25 µm, 75 µm, and 250 µm, successively. It

is further quantificationally characterized using reduced fraction in pores size as shown in

Figure 6.11d. For the pore with radius of 60 µm, the reduced fraction decreased from 70 %

at depth of 25 µm to 32 % at depth 150 µm. Besides, it could be found that at shallow

layer, the pores squash in all directions, while gradually only squash in one direction along

45 °, which has the maximum shear stress. For the USNM only, the squashing and closing

effect shows a declining trend, but this declining tread decreases along depth direction.

According to Figure 6.11e, when the depth reaches 250 µm, the squashing effects on

pores of UNSM and LA-UNSM show negligible difference, which indicates the

temperature (～ 65 °C) at 250 µm is too low to make any difference to UNSM processing.

84
 (a) (b)

(c) (d)

(e) (f)

Figure 6.11 (a)Temperature distribution along depth direction. Effect of laser on

pore reduction with different depth: (b) 25 µm, (c) 75 µm, (d) 250 µm. Reduced fraction

of pore (e) with different depth and (f) comparison with and without laser.

The effects of different laser powers on the pore squashing at depth of 75 µm are also

investigated and plotted in Figure 6.12. Higher reduced fraction could be found when

increasing the laser power. This phenomenon could also be explained by the temperature

difference as foregoing discussion. Although higher temperature could further assist the

squashing of pores, it may also induce significant thermal softening, surface ablation and
85
oxidation, which will deteriorate the mechanical properties of the 3D-printed alloys. This

will be investigated in a future work.

Figure 6.12 Effect of laser energy on the pore size.

Nanograins remained at the top and sub-surface layer

Thermal softening mainly comes from three aspect, dislocation annealing, residual

stress relief, and recrystallization. Figure 6.13 shows the bright field TEM images and

selected area electron diffraction pattern (SAED) from the top surface of the LA-UNSM-

processed samples. It could be found that the grain sizes are about 50-150 nm after LA-

UNSM. Grain refinement is mainly introduced by the surface severe plastic deformation,

which has been observed by previous reported [11]. Since UNSM can introduce severe

strain and lattice defect distortion to the processed materials [17], the bright-field image

shows complicated non-uniform contrast and SAED pattern shows many rings. The most

valuable information obtained from the TEM bright field image is that the grain size of

3D-printed Ti64 after LA-UNSM still maintains at nanosized level. This indicated that at a

86
low temperature (173.8 °C) did not make grains coarsen significantly, which made the

effect of strengthening from grain refine retained.

Figure 6.13 TEM bright field image of LA-UNSM-processed Ti6Al4V alloys, SAED

patterns are inserted.

6.5 Chapter Summary

In this chapter, 3D-printed Ti64 samples were treated by UNSM and LA-UNSM

techniques. Compared with the conventional UNSM treatment, LA-UNSM shows higher

efficient in process 3D-printed metals. Better surface finish and higher surface hardness

were obtained after LA-UNSM. In addition, the severe plastic deformation layer

produced by LA-UNSM was much thicker than the layer produced by UNSM, which is

consistent with the higher metal plasticity at elevated process temperatures

associated with the LA-UNSM process. Surface local heating shows similar effect

compared with bulk heating, while the difference is this strengthening effect decreases

with increasing of depth, which is because the temperature also shows gradient

distribution. Overall, LA-UNSM is a very

87
efficient post-processing technique for 3D-printed alloys by making the thermal and

mechanical effects work synergistically and energy-efficiently.

88
 7
CHAPTER VII

CONCLUSION

In order to improve the process effectiveness of surface severe plastic deformation

(SSPD) on hard-to-deform alloys, energy assisted surface severe plastic deformation was

proposed and evaluated using experimental and simulation methods. EASSPD is a

synergistic surface processing method taking advantages of advanced heating methods

to improve the plasticity and thus the effectiveness of SSPD processing of hard-to-deform

alloys. Experimentally, DC-UNSM and LA-UNSM have been evaluated using 3D-printed

Ti64 alloy. Extra effect of electropulsing has been evaluated through EP-LSP processing of

Ti64 alloy. To investigate how resistive heating of DC-UNSM affects the porosity reduction

in Ti64, the electric current flowing through a Ti64 matrix containing several spherical

pored was studied using numerical simulation implemented in MATLAB. To investigate

the temperature distribution and its synergistic effects with UNSM on pore closing, a 2-

dimensional FEM control model was constructed to simulate the LA-UNSM process.

Direct current can rapidly increase the bulk temperature of the 3D-printed Ti64 during

UNSM process, which makes the DC-UNSM effect deeper into hot Ti64 compared with

traditional UNSM process. As a result, the severe plastic deformation layer produced by

DC-UNSM was much thicker than the layer produced by UNSM, which is attributed to the

89
higher metal plasticity at elevated processing temperatures in the DC-UNSM process.

Additionally, the surface hardness of DC-UNSM–treated samples was greater than the

hardness values obtained for the control and UNSM-treated samples. Simulation

indicated that the electric current density creates inhomogeneous and localized heating

around the pores. The combination of thermal and mechanical effects appears to work

synergistically to close the pores during the DC-UNSM process.

It was observed that EP-LSP results in higher surface hardness and a deeper hardened

layer compared with DC-LSP having the same thermal effect (same effective current

density), which implies that the effect of pulsed current on the LSP process goes beyond

the thermal effect. Electropulsing tensile testing showed that pulsed current can more

effectively lower the flow stress of Ti64 compared with continuous current having the

same heating effect. In addition, the degree of flow stress reduction increases with the

peak current density even though the effective current density and thus the thermal

heating effect are the same. Compared with continuous current, pulsed current with

much higher peak current density can more effectively increase dislocation mobility and

thus more effectively decrease the flow stress, which results in greater degree of work-

hardening and leads to greater surface hardness of deeper hardened layer in EP-LSP. This

study has demonstrated that the effect of pulsed current on the plasticity of Ti64 goes

beyond the thermal effect. In addition, the EP-LSP effect cannot be fully explained by

resistive heating alone. Improvement of the effectiveness of SSPD without requiring a

90
significant increase in bulk temperature is highly beneficial for deformation-based

manufacturing processes.

LA-UNSM also shows higher effectivity in process hard-to-deform alloys compared

with the conventional UNSM treatment. Locally heating the under-processing area makes

energy utilization more effective for SSPD. Better surface finish and higher surface

hardness were obtained after LA-UNSM. In addition, the severe plastic deformation layer

produced by LA-UNSM was much thicker than that produced by traditional UNSM. Local

heating shows similar influence as bulk heating. The difference is the gradient

temperature distribution in LA-SSPD, while it is more uniformly distributed in current

assisted SSPD. This makes the strengthening effect of LA-SSPD attenuate along the depth

direction. Overall, LA-SSPD can also effectively process hard-to-deform alloys by making

the thermal and mechanical effects work synergistically and energy-efficiently.

The findings in this thesis provide theoretical and practical support on improving the

effectiveness of the surface severe plastic deformation by additional energy input.

Thermal effect from bulk heating and local heating and athermal effect which can lower

the processing temperature from different forms of energy serve as multiple alternatives

to assist SSPD based on specific issues.

91
 REFERENCES

[1] Zhang, Y., Wang, W. H., and Greer, A. L., 2006, “Making Metallic Glasses Plastic by

Control of Residual Stress,” Nat. Mater., 5(11), pp. 857–860.

[2] Tao, N. R., Sui, M. L., Lu, J., and Lua, K., 1999, “Surface Nanocrystallization of Iron

Induced by Ultrasonic Shot Peening,” Nanostructured Mater., 11(4), pp. 433–440.

[3] Liu, G., Lu, J., and Lu, K., 2000, “Surface Nanocrystallization of 316L Stainless Steel

Induced by Ultrasonic Shot Peening,” Mater. Sci. Eng. A, 286(1), pp. 91–95.

[4] Wu, X., Tao, N., Hong, Y., Xu, B., Lu, J., and Lu, K., 2002, “Microstructure and

Evolution of Mechanically-Induced Ultrafine Grain in Surface Layer of AL-Alloy

Subjected to USSP,” Acta Mater., 50(8), pp. 2075–2084.

[5] Tao, N. R., Wang, Z. B., Tong, W. P., Sui, M. L., Lu, J., and Lu, K., 2002, “An

Investigation of Surface Nanocrystallization Mechanism in Fe Induced by Surface

Mechanical Attrition Treatment,” Acta Mater., 50(18), pp. 4603–4616.

[6] Tolga Bozdana, A., Gindy, N. N. Z., and Li, H., 2005, “Deep Cold Rolling with

Ultrasonic Vibrations—a New Mechanical Surface Enhancement Technique,” Int. J.

Mach. Tools Manuf., 45(6), pp. 713–718.

[7] Cherif, A., Pyoun, Y., and Scholtes, B., 2010, “Effects of Ultrasonic Nanocrystal

Surface Modification (UNSM) on Residual Stress State and Fatigue Strength of AISI

304,” J. Mater. Eng. Perform., 19(2), pp. 282–286.

[8] Gill, A., Telang, A., Mannava, S. R., Qian, D., Pyoun, Y.-S., Soyama, H., and

Vasudevan, V. K., 2013, “Comparison of Mechanisms of Advanced Mechanical

92
 Surface Treatments in Nickel-Based Superalloy,” Mater. Sci. Eng. A, 576, pp. 346–

355.

[9] Montross, C., 2002, “Laser Shock Processing and Its Effects on Microstructure and

Properties of Metal Alloys: A Review,” Int. J. Fatigue, 24(10), pp. 1021–1036.

[10] Cao, X. J. J., Pyoun, Y. S. S., and Murakami, R., 2010, “Fatigue Properties of a S45C

Steel Subjected to Ultrasonic Nanocrystal Surface Modification,” Appl. Surf. Sci.,

256(21), pp. 6297–6303.

[11] Ye, C., Telang, A., Gill, A. S., Suslov, S., Idell, Y., Zweiacker, K., Wiezorek, J. M. K.,

Zhou, Z., Qian, D., Mannava, S. R., and Vasudevan, V. K., 2014, “Gradient

Nanostructure and Residual Stresses Induced by Ultrasonic Nano-Crystal Surface

Modification in 304 Austenitic Stainless Steel for High Strength and High Ductility,”

Mater. Sci. Eng. A, 613, pp. 274–288.

[12] Nie, X., He, W., Zhou, L., Li, Q., and Wang, X., 2014, “Experiment Investigation of

Laser Shock Peening on TC6 Titanium Alloy to Improve High Cycle Fatigue

Performance,” Mater. Sci. Eng. A, 594, pp. 161–167.

[13] Cellard, C., Retraint, D., François, M., Rouhaud, E., and Le Saunier, D., 2012, “Laser

Shock Peening of Ti-17 Titanium Alloy: Influence of Process Parameters,” Mater.

Sci. Eng. A, 532, pp. 362–372.

[14] Amanov, A., Penkov, O. V, Pyun, Y.-S., and Kim, D.-E., 2012, “Effects of Ultrasonic

Nanocrystalline Surface Modification on the Tribological Properties of AZ91D

Magnesium Alloy,” Tribol. Int., 54, pp. 106–113.

93
[15] Ye, C., Zhou, X., Telang, A., Gao, H., Ren, Z., Qin, H., Suslov, S., Gill, A. S. A. S. A. S.,

Mannava, S. R. R., Qian, D., Doll, G. L. G. L., Martini, A., Sahai, N., and Vasudevan,

V. K. V. K., 2016, “Surface Amorphization of NiTi Alloy Induced by Ultrasonic

Nanocrystal Surface Modification for Improved Mechanical Properties,” J. Mech.

Behav. Biomed. Mater., 53, pp. 455–462.

[16] Amanov, A., Kim, J., Pyun, Y., Hirayama, T., and Hino, M., 2015, “Wear Mechanisms

of Silicon Carbide Subjected to Ultrasonic Nanocrystalline Surface Modification

Technique,” Wear, 332–333, pp. 891–899.

[17] Amanov, A., Pyun, Y. S., and Sasaki, S., 2014, “Effects of Ultrasonic Nanocrystalline

Surface Modification (UNSM) Technique on the Tribological Behavior of Sintered

Cu-Based Alloy,” Tribol. Int., 72, pp. 187–197.

[18] Hou, X., Qin, H., Gao, H., Mankoci, S., Zhang, R., Zhou, X., Ren, Z., Doll, G. L., Martini,

A., Sahai, N., Dong, Y., and Ye, C., 2017, “A Systematic Study of Mechanical

Properties, Corrosion Behavior and Biocompatibility of AZ31B Mg Alloy after

Ultrasonic Nanocrystal Surface Modification,” Mater. Sci. Eng. C, 78, pp. 1061–1071.

[19] Huang, S., Sheng, J., Zhou, J. Z., Lu, J. Z., Meng, X. K., Xu, S. Q., and Zhang, H. F.,

2015, “On the Influence of Laser Peening with Different Coverage Areas on Fatigue

Response and Fracture Behavior of Ti–6Al–4V Alloy,” Eng. Fract. Mech., 147, pp.

72–82.

[20] KUMAR, S. A., SUNDAR, R., RAMAN, S. G. S., KUMAR, H., KAUL, R., RANGANATHAN,

K., OAK, S. M., KUKREJA, L. M., and BINDRA, K. S., 2014, “Influence of Laser Peening

94
 on Microstructure and Fatigue Lives of Ti–6Al–4V,” Trans. Nonferrous Met. Soc.

China, 24(10), pp. 3111–3117.

[21] Zhou, L., Li, Y., He, W., He, G., Nie, X., Chen, D., Lai, Z., and An, Z., 2013, “Deforming

TC6 Titanium Alloys at Ultrahigh Strain Rates during Multiple Laser Shock Peening,”

Mater. Sci. Eng. A, 578, pp. 181–186.

[22] Gujba, A. K., Ren, Z., Dong, Y., Ye, C., and Medraj, M., 2016, “Effect of Ultrasonic

Nanocrystalline Surface Modification on the Water Droplet Erosion Performance

of Ti6Al4V,” Surf. Coatings Technol., 307, pp. 157–170.

[23] Ye, C., Telang, A., Gill, A. S., Suslov, S., Idell, Y., Zweiacker, K., Wiezorek, J. M. K.,

Zhou, Z., Qian, D., Mannava, S. R., and Vasudevan, V. K., 2014, “Gradient

Nanostructure and Residual Stresses Induced by Ultrasonic Nano-Crystal Surface

Modification in 304 Austenitic Stainless Steel for High Strength and High Ductility,”

Mater. Sci. Eng. A, 613, pp. 274–288.

[24] Dieter, G. E., and Bacon, D. J., 1986, Mechanical Metallurgy, McGraw-hill New York.

[25] Zhang, H., Chiang, R., Qin, H., Ren, Z., Hou, X., Lin, D., Doll, G. L., Vasudevan, V. K.,

Dong, Y., and Ye, C., 2017, “The Effects of Ultrasonic Nanocrystal Surface

Modification on the Fatigue Performance of 3D-Printed Ti64,” Int. J. Fatigue, 103,

pp. 136–146.

[26] Chen, A. Y., Ruan, H. H., Wang, J., Chan, H. L., Wang, Q., Li, Q., and Lu, J., 2011, “The

Influence of Strain Rate on the Microstructure Transition of 304 Stainless Steel,”

Acta Mater., 59(9), pp. 3697–3709.

95
[27] Mott, N. F., 1952, “CXVII. A Theory of Work-Hardening of Metal Crystals,” London,

Edinburgh, Dublin Philos. Mag. J. Sci., 43(346), pp. 1151–1178.

[28] Hansen, N., 2004, “Hall–Petch Relation and Boundary Strengthening,” Scr. Mater.,

51(8), pp. 801–806.

[29] Schioøtz, J., and Jacobsen, K. W., 2003, “A Maximum in the Strength of

Nanocrystalline Copper,” Science (80-. )., 301(5638), pp. 1357–1359.

[30] Zhao, P., and Guo, Y., 2018, “Grain Size Effects on Indentation-Induced Defect

Evolution and Plastic Deformation Mechanism of Ploycrystalline Materials,”

Comput. Mater. Sci., 155(September), pp. 431–438.

[31] Wagner, L., and Wollmann, M., 2013, “Titanium and Titanium Alloys,” Structural

Materials and Processes in Transportation, Wiley-VCH Verlag GmbH & Co. KGaA,

Weinheim, Germany, pp. 151–180.

[32] Khanna, N., and Sangwan, K. S., 2013, “Machinability Analysis of Heat Treated Ti64,

Ti54M and Ti10.2.3 Titanium Alloys,” Int. J. Precis. Eng. Manuf., 14(5), pp. 719–724.

[33] Jackson, A. G., 1991, “Slip Systems,” Handbook of Crystallography, Springer New

York, New York, NY, pp. 83–88.

[34] Stein, D. F., and Low, J. R., 1960, “Mobility of Edge Dislocations in Silicon-Iron

Crystals,” J. Appl. Phys., 31(2), pp. 362–369.

[35] Bammann, D. J., and Aifantis, E. C., 1982, “On a Proposal for a Continuum with

Microstructure,” Acta Mech., 45(1–2), pp. 91–121.

[36] Hirth, J. P., and Nix, W. D., 1969, “An Analysis of the Thermodynamics of Dislocation

96
 Glide,” Phys. status solidi, 35(1), pp. 177–188.

[37] Conrad, H., and Wiedersich, H., 1960, “Activation Energy for Deformation of Metals

at Low Temperatures,” Acta Metall., 8(2), pp. 128–130.

[38] Gibbs, G. B., 1970, “Thermodynamic Systems for the Analysis of Dislocation Glide,”

Philos. Mag., 22(178), pp. 701–706.

[39] Hull, D., and Bacon, D. J., 2001, Introduction to Dislocations, Butterworth-

Heinemann.

[40] Nestel, B., 1980, “Hot Working of Titanium Alloys.”

[41] Bewlay, B. P., Gigliotti, M. F. X., Hardwicke, C. U., Kaibyshev, O. A., Utyashev, F. Z.,

and Salischev, G. A., 2003, “Net-Shape Manufacturing of Aircraft Engine Disks by

Roll Forming and Hot Die Forging,” J. Mater. Process. Technol., 135(2-3 SPEC.), pp.

324–329.

[42] Giarratano, J. G., 1983, “J. G. Giarratano, Foundations of Computector Technology,

Howard W. Sams & Co Indianapolis, IN, 1983.),” Found. Comput. Technol., pp. 1–

200.

[43] Semiatin, S. L., and DeLo, D. P., 2000, “Equal Channel Angular Extrusion of Difficult-

to-Work Alloys,” Mater. Des., 21(4), pp. 311–322.

[44] Valax, M.-F., Rattat, A.-C., Baracat, B., and Cegarra, J., 2016, “Effect of Daylight

Saving Time on Punctuality for Medical Appointments,” Appl. Cogn. Psychol., 30(6),

pp. 911–916.

[45] Semiatin, S. ., and Bieler, T. ., 2001, “The Effect of Alpha Platelet Thickness on

97
 Plastic Flow during Hot Working of TI–6Al–4V with a Transformed Microstructure,”

Acta Mater., 49(17), pp. 3565–3573.

[46] Akbari, G. H., Sellars, C. M., and Whiteman, J. a., 1997, “Microstructural

Development during Warm Rolling of an IF Steel,” Acta Mater., 45(12), pp. 5047–

5058.

[47] Appel, F., Oehring, M., Paul, J. D. H., Klinkenberg, C., and Carneiro, T., 2004,

“Physical Aspects of Hot-Working Gamma-Based Titanium Aluminides,”

Intermetallics, 12(7–9), pp. 791–802.

[48] Huang, C., Hawbolt, E. B., Chen, X., Meadowcroft, T. R., and Matlock, D. K., 2001,

“Flow Stress Modeling and Warm Rolling Simulation Behavior of Two Ti-Nb

Interstitial-Free Steels in the Ferrite Region,” Acta Mater., 49(8), pp. 1445–1452.

[49] Nguyen-Tran, H.-D., Oh, H.-S., Hong, S.-T., Han, H. N., Cao, J., Ahn, S.-H., and Chun,

D.-M., 2015, “A Review of Electrically-Assisted Manufacturing,” Int. J. Precis. Eng.

Manuf. Technol., 2(4), pp. 365–376.

[50] Jones, J. J., Mears, L., and Roth, J. T., 2012, “Electrically-Assisted Forming of

Magnesium AZ31: Effect of Current Magnitude and Deformation Rate on

Forgeability,” J. Manuf. Sci. Eng., 134(3), p. 34504.

[51] Salandro, W. a., Bunget, C. J., and Mears, L., 2011, “Several Factors Affecting the

Electroplastic Effect During an Electrically-Assisted Forming Process,” J. Manuf. Sci.

Eng., 133(6), p. 064503.

[52] Salandro, W. a., Bunget, C., and Mears, L., 2011, “Electroplastic Modeling of

98
 Bending Stainless Steel Sheet Metal Using Energy Methods,” J. Manuf. Sci. Eng.,

133(4), p. 041008.

[53] Sánchez Egea, A. J., González Rojas, H. a., Celentano, D. J., Travieso-Rodríguez, J. A.,

and Llumà i Fuentes, J., 2014, “Electroplasticity-Assisted Bottom Bending Process,”

J. Mater. Process. Technol., 214(11), pp. 2261–2267.

[54] Xu, D. K., Lu, B., Cao, T. T., Zhang, H., Chen, J., Long, H., and Cao, J., 2016,

“Enhancement of Process Capabilities in Electrically-Assisted Double Sided

Incremental Forming,” Mater. Des., 92, pp. 268–280.

[55] Sánchez Egea, A. J., González Rojas, H. A., Montilla Montaña, C. A., and Kallewaard

Echeverri, V., 2015, “Effect of Electroplastic Cutting on the Manufacturing Process

and Surface Properties,” J. Mater. Process. Technol., 222, pp. 327–334.

[56] Wang, H., Chen, L., Liu, D., Song, G., and Tang, G., 2015, “Study on Electropulsing

Assisted Turning Process for AISI 304 Stainless Steel,” Mater. Sci. Technol., 00(0), p.

1743284715Y.000.

[57] Ye, X., Tse, Z. T. H., Tang, G., and Song, G., 2014, “Effect of Electroplastic Rolling on

Deformability, Mechanical Property and Microstructure Evolution of Ti–6Al–4V

Alloy Strip,” Mater. Charact., 98, pp. 147–161.

[58] Liao, H., Tang, G., Jiang, Y., Xu, Q., Sun, S., and Liu, J., 2011, “Effect of Thermo-

Electropulsing Rolling on Mechanical Properties and Microstructure of AZ31

Magnesium Alloy,” Mater. Sci. Eng. A, 529, pp. 138–142.

[59] Xu, Z., Tang, G., Tian, S., Ding, F., and Tian, H., 2007, “Research of Electroplastic

99
 Rolling of AZ31 Mg Alloy Strip,” J. Mater. Process. Technol., 182(1–3), pp. 128–133.

[60] Hu, G., Wang, Z., Zhu, Y., Liu, J., and Tang, G., 2011, “Dynamic Electropulsing

Induced Evolution of Basal Texture and Its Effect on Properties of Magnesium Alloy

AZ61,” Mater. Trans., 52(8), pp. 1565–1568.

[61] Kuang, J., Li, X., Zhang, R., Ye, Y., Luo, A. A., and Tang, G., 2016, “Enhanced

Rollability of Mg-3Al-1Zn Alloy by Pulsed Electric Current: A Comparative Study,”

Mater. Des., 100, pp. 204–216.

[62] Ye, Y., Li, X., Kuang, J., Geng, Y., and Tang, G., 2015, “Effects of Electropulsing

Assisted Ultrasonic Impact Treatment on Welded Components,” Mater. Sci.

Technol., 00(0), p. 1743284715Y.000.

[63] Ye, X., Liu, T., Ye, Y., Wang, H., Tang, G., and Song, G., 2015, “Enhanced Grain

Refinement and Microhardness of Ti – Al – V Alloy by Electropulsing Ultrasonic

Shock,” J. Alloys Compd., 621, pp. 66–70.

[64] Liu, D., Li, X., Tang, G., Chen, L., Wang, H., and Song, G., 2015, “Improvement of

Surface Properties of 2316 Stainless Steel with Ultrasonic Electric Surface

Modification,” Mater. Sci. Technol., 00(0), p. 1743284715Y.000.

[65] Ye, Y., Wang, H., Tang, G., and Song, G., 2018, “Effect of Electropulsing-Assisted

Ultrasonic Nanocrystalline Surface Modification on the Surface Mechanical

Properties and Microstructure of Ti-6Al-4V Alloy,” J. Mater. Eng. Perform., 27(5),

pp. 2394–2403.

[66] Hu, G., Shek, C., Zhu, Y., Tang, G., and Qing, X., 2010, “Effect of Electropulsing on

100
 Recrystallization of Fe-3%Si Alloy Strip,” Mater. Trans., 51(8), pp. 1390–1394.

[67] Xu, Q., Tang, G., Jiang, Y., Hu, G., and Zhu, Y., 2011, “Accumulation and Annihilation

Effects of Electropulsing on Dynamic Recrystallization in Magnesium Alloy,” Mater.

Sci. Eng. A, 528(7–8), pp. 3249–3252.

[68] Jiang, Y., Guan, L., Tang, G., Shek, C., and Zhang, Z., 2011, “Influence of

Electropulsing Treatment on Microstructure and Mechanical Properties of Cold-

Rolled Mg–9Al–1Zn Alloy Strip,” Mater. Sci. Eng. A, 528(16–17), pp. 5627–5635.

[69] Liu, K., Dong, X., Xie, H., and Peng, F., 2015, “Effect of Pulsed Current on the

Deformation Behavior of AZ31B Magnesium Alloy,” Mater. Sci. Eng. A, 623, pp. 97–

103.

[70] Wang, H., Song, G., and Tang, G., 2016, “Evolution of Surface Mechanical Properties

and Microstructure of Ti 6Al 4V Alloy Induced by Electropulsing-Assisted Ultrasonic

Surface Rolling Process,” J. Alloys Compd., 681, pp. 146–156.

[71] Jiang, Y., Guan, L., Tang, G., and Zhang, Z., 2015, “Improved Mechanical Properties

of Mg–9Al–1Zn Alloy by the Combination of Aging, Cold-Rolling and Electropulsing

Treatment,” J. Alloys Compd., 626, pp. 297–303.

[72] Alankar, A., Eisenlohr, P., and Raabe, D., 2011, “A Dislocation Density-Based Crystal

Plasticity Constitutive Model for Prismatic Slip in Alpha-Titanium,” Acta Mater.,

59(18), pp. 7003–7009.

[73] Li, D. L., and Yu, E. L., 2010, “An Approach Based on the Classical Free-Electron

Theory to Study Electroplastic Effect,” Adv. Mater. Res., 148–149, pp. 71–74.

101
[74] Alankar, A., Field, D. P., and Raabe, D., 2014, “Plastic Anisotropy of Electro-

Deposited Pure α-Iron with Sharp Crystallographic <111>// Texture in Normal

Direction: Analysis by an Explicitly Dislocation-Based Crystal Plasticity Model,” Int.

J. Plast., 52, pp. 18–32.

[75] Altan, T., Oh, S.-I., Gegel, G., and Ngaile, G., 2005, Cold and Hot Forging

Fundamentals and Applications, ASM International.

[76] Naka, T., and Yoshida, F., 1999, “Deep Drawability of Type 5083 Aluminium–

Magnesium Alloy Sheet under Various Conditions of Temperature and Forming

Speed,” J. Mater. Process. Technol., 89–90, pp. 19–23.

[77] Amanov, A., Urmanov, B., Amanov, T., and Pyun, Y. S., 2017, “Strengthening of Ti-

6Al-4V Alloy by High Temperature Ultrasonic Nanocrystal Surface Modification

Technique,” Mater. Lett., 196, pp. 198–201.

[78] Amanov, A., and Pyun, Y. S., 2017, “Local Heat Treatment with and without

Ultrasonic Nanocrystal Surface Modification of Ti-6Al-4V Alloy: Mechanical and

Tribological Properties,” Surf. Coatings Technol., 326, pp. 343–354.

[79] Amanov, A., and Umarov, R., 2018, “The Effects of Ultrasonic Nanocrystal Surface

Modification Temperature on the Mechanical Properties and Fretting Wear

Resistance of Inconel 690 Alloy,” Appl. Surf. Sci., 441, pp. 515–529.

[80] Ye, C., Liao, Y., and Cheng, G. J., 2010, “Warm Laser Shock Peening Driven

Nanostructures and Their Effects on Fatigue Performance in Aluminum Alloy 6160,”

Adv. Eng. Mater., 12(4), pp. 291–297.

102
[81] Ye, C., Liao, Y., Suslov, S., Lin, D., and Cheng, G. J., 2014, “Ultrahigh Dense and

Gradient Nano-Precipitates Generated by Warm Laser Shock Peening for

Combination of High Strength and Ductility,” Mater. Sci. Eng. A, 609(0), pp. 195–

203.

[82] Ye, C., and Cheng, G. J., 2010, “Effects of Temperature on Laser Shock Induced

Plastic Deformation: The Case of Copper,” J. Manuf. Sci. Eng., 132(6), p. 61009.

[83] Liao, Y., Ye, C., and Cheng, G. J., 2016, “[INVITED] A Review: Warm Laser Shock

Peening and Related Laser Processing Technique,” Opt. Laser Technol., 78, pp. 15–

24.

[84] Ye, C., Suslov, S., Kim, B. J., Stach, E. A., and Cheng, G. J., 2011, “Fatigue

Performance Improvement in AISI 4140 Steel by Dynamic Strain Aging and Dynamic

Precipitation during Warm Laser Shock Peening,” Acta Mater., 59(3), pp. 1014–

1025.

[85] Troitskii, O. A., 1969, “Electromechanical Effect in Metals,” ZhETF Pisma Redaktsiiu,

10, p. 18.

[86] Nguyen-Tran, H.-D., Oh, H.-S., Hong, S.-T., Han, H. N., Cao, J., Ahn, S.-H., and Chun,

D.-M., 2015, “A Review of Electrically-Assisted Manufacturing,” Int. J. Precis. Eng.

Manuf. Technol., 2(4), pp. 365–376.

[87] Wang, H., Song, G., and Tang, G., 2016, “Effect of Electropulsing on Surface

Mechanical Properties and Microstructure of AISI 304 Stainless Steel during

Ultrasonic Surface Rolling Process,” Mater. Sci. Eng. A, 662, pp. 456–467.

103
[88] Ye, Y., Kure-Chu, S.-Z., Sun, Z., Li, X., Wang, H., and Tang, G., 2018,

“Nanocrystallization and Enhanced Surface Mechanical Properties of Commercial

Pure Titanium by Electropulsing-Assisted Ultrasonic Surface Rolling,” Mater. Des.,

149, pp. 214–227.

[89] Wang, H., Song, G., and Tang, G., 2015, “Enhanced Surface Properties of Austenitic

Stainless Steel by Electropulsing-Assisted Ultrasonic Surface Rolling Process,” Surf.

Coatings Technol., 282, pp. 149–154.

[90] Chryssolouris, G., Anifantis, N., and Karagiannis, S., 1997, “Laser Assisted

Machining: An Overview,” J. Manuf. Sci. Eng., 119(4B), pp. 766–769.

[91] Sun, S., Brandt, M., and Dargusch, M. S., 2010, “Thermally Enhanced Machining of

Hard-to-Machine Materials—A Review,” Int. J. Mach. Tools Manuf., 50(8), pp. 663–

680.

[92] Tian, Y., and Shin, Y. C., 2007, “Laser-Assisted Burnishing of Metals,” Int. J. Mach.

Tools Manuf., 47(1), pp. 14–22.

[93] ASTM E1426 − 14, 2003, “Standard Test Method for Determining the Effective

Elastic Parameter for X-Ray Diffraction Techniques,” ASTM B. Stand., 03(July), pp.

1–4.

[94] ASTM Standard E915, 2010, “Standard Test Method for Verifying the Alignment of

X-Ray Diffraction Instrumentation for Residual Stress Measurement,” ASTM B.

Stand., i(June 1996), pp. 1–4.

[95] Hilley, M. E., Larson, J. A., Jatczak, C. F., and Ricklefs, R. E., 1971, “Residual Stress

104
 Measurement by X-Ray Diffraction,” SAE J784a, 15096.

[96] Xu, W., Brandt, M., Sun, S., Elambasseril, J., Liu, Q., Latham, K., Xia, K., and Qian,

M., 2015, “Additive Manufacturing of Strong and Ductile Ti–6Al–4V by Selective

Laser Melting via in Situ Martensite Decomposition,” Acta Mater., 85, pp. 74–84.

[97] Facchini, L., Magalini, E., Robotti, P., Molinari, A., Höges, S., and Wissenbach, K.,

2010, “Ductility of a Ti-6Al-4V Alloy Produced by Selective Laser Melting of

Prealloyed Powders,” Rapid Prototyp. J., 16(6), pp. 450–459.

[98] Xu, Z. S., Lai, Z. H., and Chen, Y. X., 1988, “Effect of Electric Current on the

Recrystallization Behavior of Cold Worked α - Ti,” Scr. Metall., 22(2), pp. 187–190.

[99] Roh, J.-H., Seo, J.-J., Hong, S.-T., Kim, M.-J., Han, H. N., and Roth, J. T., 2014, “The

Mechanical Behavior of 5052-H32 Aluminum Alloys under a Pulsed Electric Current,”

Int. J. Plast., 58, pp. 84–99.

[100] Kim, M. J., Lee, K., Oh, K. H., Choi, I. S., Yu, H. H., Hong, S. T., and Han, H. N., 2014,

“Electric Current-Induced Annealing during Uniaxial Tension of Aluminum Alloy,”

Scr. Mater., 75, pp. 58–61.

[101] Vasylyev, M. A., Chenakin, S. P., and Yatsenko, L. F., 2016, “Ultrasonic Impact

Treatment Induced Oxidation of Ti6Al4V Alloy,” Acta Mater., 103, pp. 761–774.

[102] García-Alonso, M. C., Saldaña, L., Vallés, G., González-Carrasco, J. L., González-

Cabrero, J., Martínez, M. E., Gil-Garay, E., and Munuera, L., 2003, “In Vitro

Corrosion Behaviour and Osteoblast Response of Thermally Oxidised Ti6Al4V Alloy,”

Biomaterials, 24(1), pp. 19–26.

105
[103] Dai, J., Zhu, J., Chen, C., and Weng, F., 2016, “High Temperature Oxidation Behavior

and Research Status of Modi Fi Cations on Improving High Temperature Oxidation

Resistance of Titanium Alloys and Titanium Aluminides : A Review,” J. Alloys

Compd., 685, pp. 784–798.

[104] Vandenbroucke, B., and Kruth, J.-P., 2007, “Selective Laser Melting of

Biocompatible Metals for Rapid Manufacturing of Medical Parts,” Rapid Prototyp.

J., 13(4), pp. 196–203.

[105] Zhao, J., Wang, G. X., Ye, C., and Dong, Y., 2016, “Cellular Automata Modeling of

Nitriding in Nanocrystalline Metals,” Comput. Mater. Sci., 118, pp. 342–352.

[106] Qin, R., and Su, S., 2002, “Thermodynamics of Crack Healing under Electropulsing,”

J. Mater. Res., 17(08), pp. 2048–2052.

[107] Magargee, J., Morestin, F., and Cao, J., 2013, “Characterization of Flow Stress for

Commercially Pure Titanium Subjected to Electrically Assisted Deformation,” J. Eng.

Mater. Technol., 135(4), p. 041003.

[108] Kinsey, B., Cullen, G., Jordan, A., and Mates, S., 2013, “Investigation of

Electroplastic Effect at High Deformation Rates for 304SS and Ti–6Al–4V,” CIRP

Ann., 62(1), pp. 279–282.

[109] Molotskii, M. I., 2000, “Theoretical Basis for Electro- and Magnetoplasticity,” Mater.

Sci. Eng. A, 287(2), pp. 248–258.

[110] Conrad, H., 2000, “Electroplasticity in Metals and Ceramics,” Mater. Sci. Eng. A,

287(2), pp. 276–287.

106
[111] Ross, C. D., Kronenberger, T. J., and Roth, J. T., 2009, “Effect of Dc on the

Formability of Ti–6Al–4V,” J. Eng. Mater. Technol., 131(3), p. 031004.

[112] Liu, J., Suslov, S., Li, S., Qin, H., Ren, Z., Doll, G. L., Cong, H., Dong, Y., and Ye, C.,

2018, “Electrically Assisted Ultrasonic Nanocrystal Surface Modification of Ti6Al4V

Alloy,” Adv. Eng. Mater., 20(1), p. 1700470.

[113] Lou, S., Li, Y., Zhou, L., Nie, X., He, G., Li, Y., and He, W., 2016, “Surface

Nanocrystallization of Metallic Alloys with Different Stacking Fault Energy Induced

by Laser Shock Processing,” Mater. Des., 104, pp. 320–326.

[114] Liu, Y. J., Liu, Z., Jiang, Y., Wang, G. W., Yang, Y., and Zhang, L. C., 2018, “Gradient

in Microstructure and Mechanical Property of Selective Laser Melted AlSi10Mg,” J.

Alloys Compd., 735, pp. 1414–1421.

[115] Lu, J. Z., Wu, L. J., Sun, G. F., Luo, K. Y., Zhang, Y. K., Cai, J., Cui, C. Y., and Luo, X. M.,

2017, “Microstructural Response and Grain Refinement Mechanism of

Commercially Pure Titanium Subjected to Multiple Laser Shock Peening Impacts,”

Acta Mater., 127, pp. 252–266.

[116] Ren, X. D., Zhou, W. F., Liu, F. F., Ren, Y. P., Yuan, S. Q., Ren, N. F., Xu, S. D., and

Yang, T., 2016, “Microstructure Evolution and Grain Refinement of Ti-6Al-4V Alloy

by Laser Shock Processing,” Appl. Surf. Sci., 363, pp. 44–49.

[117] Lainé, S. J., Knowles, K. M., Doorbar, P. J., Cutts, R. D., and Rugg, D., 2017,

“Microstructural Characterisation of Metallic Shot Peened and Laser Shock Peened

Ti–6Al–4V,” Acta Mater., 123, pp. 350–361.

107
[118] Gromov, V. E., Gurevich, L. I., Kurilov, V. F., and Erilova, T. V., 1989, “Influence of

Current Pulses on the Mobility and Multiplication of Dislocations in Zn,” Strength

Mater., 21(10), pp. 1335–1341.

[119] Gromov, V. E., Gurevich, L. I., Kuznetsov, V. A., and Erilova, T. V, 1990, “Influence

of Electric Current Pulses on the Mobility and Multiplication of Dislocations in Zn-

Monocrystals,” Czechoslov. J. Phys., 40(8), pp. 895–902.

[120] Conrad, H., White, J., Cao, W. D., Lu, X. P., and Sprecher, A. F., 1991, “Effect of

Electric Current Pulses on Fatigue Characteristics of Polycrystalline Copper,” Mater.

Sci. Eng. A, 145(1), pp. 1–12.

[121] Sprecher, A. F., Mannan, S. L., and Conrad, H., 1986, “Overview No. 49: On the

Mechanisms for the Electroplastic Effect in Metals,” Acta Metall., 34(7), pp. 1145–

1162.

[122] Okazaki, K., Kagawa, M., and Conrad, H., 1978, “A Study of the Electroplastic Effect

in Metals,” Scr. Metall., 12(11), pp. 1063–1068.

[123] Zhao, J., Ren, Z., Zhang, H., Wang, G.-X., Dong, Y., and Ye, C., 2019, “Electroplasticity

in AZ31B Subjected to Short-Duration High-Frequency Pulsed Current,” J. Appl.

Phys., 125(18), p. 185104.

[124] Nam, S.-W., Chung, H.-S., Lo, Y. C., Qi, L., Li, J., Lu, Y., Johnson, A. T. C., Jung, Y.,

Nukala, P., and Agarwal, R., 2012, “Electrical Wind Force-Driven and Dislocation-

Templated Amorphization in Phase-Change Nanowires,” Science (80-. )., 336(6088),

pp. 1561–1566.

108
[125] Liao, Y.-H., Liang, C.-L., Lin, K.-L., and Wu, A. T., 2015, “High Dislocation Density of

Tin Induced by Electric Current,” AIP Adv., 5(12), p. 127210.

[126] Zhao, J., Wang, G.-X., Dong, Y., and Ye, C., 2017, “Multiscale Modeling of Localized

Resistive Heating in Nanocrystalline Metals Subjected to Electropulsing,” J. Appl.

Phys., 122(8), p. 085101.

[127] Chen, K., Tamura, N., Valek, B. C., and Tu, K. N., 2008, “Plastic Deformation in Al

(Cu) Interconnects Stressed by Electromigration and Studied by Synchrotron

Polychromatic x-Ray Microdiffraction,” J. Appl. Phys., 104(1), p. 013513.

[128] Budiman, A. S., Hau-Riege, C. S., Baek, W. C., Lor, C., Huang, A., Kim, H. S., Neubauer,

G., Pak, J., Besser, P. R., and Nix, W. D., 2010, “Electromigration-Induced Plastic

Deformation in Cu Interconnects: Effects on Current Density Exponent, n, and

Implications for EM Reliability Assessment,” J. Electron. Mater., 39(11), pp. 2483–

2488.

[129] Rahnama, A., and Qin, R., 2017, “Room Temperature Texturing of Austenite/Ferrite

Steel by Electropulsing,” Sci. Rep., 7(1), p. 42732.

[130] Shen, H., Zhu, W., Li, Y., Tamura, N., and Chen, K., 2016, “In Situ Synchrotron Study

of Electromigration Induced Grain Rotations in Sn Solder Joints,” Sci. Rep., 6(1), p.

24418.

[131] Radziejewska, J., 2011, “Influence of Laser-Mechanical Treatment on Surface

Topography, Erosive Wear and Contact Stiffness,” Mater. Des., 32(10), pp. 5073–

5081.

109
[132] Wu, B., Zhang, L., Zhang, J., Murakami, R., and Pyoun, Y.-S., 2015, “An Investigation

of Ultrasonic Nanocrystal Surface Modification Machining Process by Numerical

Simulation,” Adv. Eng. Softw., 83, pp. 59–69.

[133] Liu, J., Suslov, S., Ren, Z., Dong, Y., and Ye, C., 2019, “Microstructure Evolution in

Ti64 Subjected to Laser-Assisted Ultrasonic Nanocrystal Surface Modification,” Int.

J. Mach. Tools Manuf., 136(September 2018), pp. 19–33.

[134] Wu, B., Wang, P., Pyoun, Y.-S., Zhang, J., and Murakami, R., 2013, “Study on the

Fatigue Properties of Plasma Nitriding S45C with a Pre-Ultrasonic Nanocrystal

Surface Modification Process,” Surf. Coatings Technol., 216, pp. 191–198.

[135] Amanov, A., Pyun, Y.-S., Kim, J.-H., Suh, C.-M., Cho, I.-S., Kim, H.-D., Wang, Q., and

Khan, M. K., 2015, “Ultrasonic Fatigue Performance of High Temperature Structural

Material Inconel 718 Alloys at High Temperature after UNSM Treatment,” Fatigue

Fract. Eng. Mater. Struct., 38(11), pp. 1266–1273.

[136] Murakami, Y., 2002, Metal Fatigue: Effects of Small Defects and Nonmetallic

Inclusions, Elsevier.

[137] Wu, S. Q., Lu, Y. J., Gan, Y. L., Huang, T. T., Zhao, C. Q., Lin, J. J., Guo, S., and Lin, J.

X., 2016, “Microstructural Evolution and Microhardness of a Selective-Laser-

Melted Ti–6Al–4V Alloy after Post Heat Treatments,” J. Alloys Compd., 672, pp.

643–652.

[138] Liu, J., Suslov, S., Li, S., Qin, H., Ren, Z., Ma, C., Wang, G. X., Doll, G. L., Cong, H.,

Dong, Y., and Ye, C., 2017, “Effects of Ultrasonic Nanocrystal Surface Modification

110
 on the Thermal Oxidation Behavior of Ti6Al4V,” Surf. Coatings Technol., 325, pp.

289–298.

[139] Zhang, H., Zhao, J., Liu, J., Qin, H., Ren, Z., Doll, G. L., Dong, Y., and Ye, C., 2018,

“The Effects of Electrically-Assisted Ultrasonic Nanocrystal Surface Modification on

3D-Printed Ti-6Al-4V Alloy,” Addit. Manuf., 22, pp. 60–68.

[140] Zhang, H., Ren, Z., Liu, J., Zhao, J., Liu, Z., Lin, D., Zhang, R., Graber, M. J., Thomas,

N. K., Kerek, Z. D., Wang, G.-X., Dong, Y., and Ye, C., 2019, “Microstructure

Evolution and Electroplasticity in Ti64 Subjected to Electropulsing-Assisted Laser

Shock Peening,” J. Alloys Compd., 802, pp. 573–582.

111