University of Southern California Dissertations and Theses

A comprehensive study of twinning phenomena in low and high stacking fault energy metals

(USC Thesis Other)

A comprehensive study of twinning phenomena in low and high stacking fault energy metals

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

A comprehensive study of twinning phenomena in low and high stacking fault energy
metals

Leonardo Velasco Estrada

Dr. Andrea M. Hodge, Advisor

A Dissertation Presented to the Faculty of the USC Graduate School
In Partial Fulfillment of the Requirements for the Degree
Doctorate of Philosophy
(Materials Science)

University of Southern California

August 2016

I

Acknowledgements
First, I dedicate this thesis to my Parents, Siblings and Family, Emese Maroti, Andrea Hodge,
Oliver Franke, and my friends.
I am very grateful for my advisor, Dr. Andrea Hodge, her guidance, patience, and support during
my time at USC provided me with the right tools to accomplish my goals, grow up personally
and as a scientist. Thank you very much Professor Franke for believing in me, I will always be
grateful to you. I also thank my committee members, Dr. Veronica Eliasson, Dr. Michael
Kassner, and Dr. Jayakanth Ravichandran, for encouraging me to pursue my goals and their
positive critiques.
I want to thank the Materials Science department staff at USC, as well as the Center for Electron
Microscopy (CEMMA), John Curulli and Dr. Matthew Mecklenburg. I am indebted to the
Colombian professors and Dr. Sandra Rodil who encouraged me to pursue this new stage of my
life. I also want to acknowledge the funding from the Defense Advance Research Projects
Agency (DARPA), the Office of Naval Research (ONR), and the National Science Foundation
(NSF) grant programs for making this research possible.
I want to thank Mesike for her love, support, and patience with me, always by my side. Hugo
Burbano, whose unconditional friendship is a symbol of good people in the world. Lastly, but by
far one of the most important things in life, is the friends that I have: Mikhail, Tim, Yifu, I-
Chung, Teri, Nathan, Kamia, Sebastian, Chelsea, Joel, Jianfeng, Joaquin, Patricia, Diego, Javier,
as well as my international and Colombian friends. Thank you, because from every one of you I
have learned a bit more in both academia and life.
II

Table of Contents
ACKNOWLEDGEMENTS ........................................................................................................................................... I
TABLE OF CONTENTS .............................................................................................................................................. II
LIST OF FIGURES ..................................................................................................................................................... V
LIST OF TABLES .................................................................................................................................................... XII
1 ABSTRACT ..................................................................................................................................................... 1
2 INTRODUCTION ............................................................................................................................................ 3
3 BACKGROUND .............................................................................................................................................. 5
3.1 NANOTWINNED METALS AND THEIR PROPERTIES ......................................................................................................... 5
3.1.1 Mechanical properties of nt metals ....................................................................................................... 5
3.1.2 Thermal stability of nt metals ................................................................................................................ 6
3.2 STACKING FAULTS AND STACKING FAULT ENERGY ........................................................................................................ 6
3.3 DESCRIPTION OF TWIN BOUNDARIES ......................................................................................................................... 8
3.4 SYNTHESIS METHODS FOR TWIN BOUNDARIES........................................................................................................... 10
3.4.1 Annealing twins ................................................................................................................................... 10
3.4.2 Deformation twins ............................................................................................................................... 11
3.4.3 Growth twins ....................................................................................................................................... 13
3.5 MAGNETRON SPUTTERING PROCESSING .................................................................................................................. 14
3.5.1 Synthesis of twins by magnetron sputtering ........................................................................................ 18
3.5.2 Nucleation and growth of twin boundaries using magnetron sputtering ........................................... 21
3.5.3 Twins in high SFE materials Al and Ni .................................................................................................. 23
3.6 SUMMARY ........................................................................................................................................................ 24

III

4 SPUTTERING SYSTEM UPGRADES AND CHARACTERIZATION TECHNIQUES ................................................. 25
4.1 SUBSTRATE TEMPERATURE CONTROL ...................................................................................................................... 25
4.1.1 Substrate cooling stage ....................................................................................................................... 25
4.1.2 Substrate heating stage ....................................................................................................................... 27
4.1.3 Further control of film deposition temperature via interrupted sputtering ......................................... 27
4.2 TEMPERATURE MONITORING ................................................................................................................................ 29
4.3 CHARACTERIZATION TECHNIQUES .......................................................................................................................... 30
4.3.1 X-ray diffraction ................................................................................................................................... 30
4.3.2 Focused ion beam ................................................................................................................................ 31
4.3.3 Electron backscattering diffraction ...................................................................................................... 33
4.3.4 Transmission Electron Microscopy ....................................................................................................... 34
5 SYNTHESIS AND CHARACTERIZATION OF NT STRUCTURES IN LOW SFE MATERIALS .................................... 38
5.1 INFLUENCE OF THE SFE ........................................................................................................................................ 38
5.2 INFLUENCE OF THE SPUTTERING DEPOSITION RATE ..................................................................................................... 45
5.3 INFLUENCE OF TEMPERATURE - HEATING AND COOLING THE SUBSTRATE ........................................................................ 49
5.4 INFLUENCE OF THE DEPOSITION RATE IN COMBINATION WITH INTERRUPTED SPUTTERING .................................................. 52
5.5 CONCLUSIONS ................................................................................................................................................... 55
6 TAILORING THE TWIN THICKNESS IN LOW AND INTERMEDIATE SFE MATERIALS: THE MOBILITY OF
GROWTH TWINS SYNTHESIZED BY SPUTTERING ......................................................................................... 56
6.1 MICROSTRUCTURAL CHARACTERIZATION ................................................................................................................. 58
6.2 EFFECT OF THE AVERAGE DEPOSITION TEMPERATURE ................................................................................................. 63
6.2.1 Twin nucleation .................................................................................................................................... 65
6.2.2 Twin boundary mobility ....................................................................................................................... 65
6.2.3 Effect of free surfaces on twin boundary mobility ............................................................................... 68
6.3 PROTEAN REPRESENTATION OF SPUTTERING CONDITIONS: TAILORING Λ ......................................................................... 69
6.4 CONCLUSIONS ................................................................................................................................................... 73
IV

7 SYNTHESIS AND CHARACTERIZATION OF TWIN BOUNDARIES IN HIGH SFE MATERIALS .............................. 75
7.1 MICROSTRUCTURAL CHARACTERIZATION ................................................................................................................. 77
7.1.1 Microstructural features of the films ................................................................................................... 80
7.2 TEXTURE CHARACTERIZATION ................................................................................................................................ 83
7.3 FORMATION OF INCLINED CTBS ............................................................................................................................ 88
7.4 CONCLUSIONS ................................................................................................................................................... 90
8 GENERAL CONCLUSIONS AND FUTURE RESEARCH RECOMMENDATIONS ................................................... 91
9 REFERENCES ............................................................................................................................................... 94
APPENDIX A. SUMMARY OF SPUTTERED SAMPLES ........................................................................................ 99
APPENDIX B. COMPLEMENTARY XRD PATTERNS FOR THE CU, CU ALLOYS, AL, AL ALLOYS,
NI, AND INCONEL600. ..................................................................................................... 110
APPENDIX C. COMPLEMENTARY MICROSTRUCTURAL FIB IMAGES FOR CU, CU ALLOYS, NI,
AND INCONEL600. ............................................................................................................... 112
APPENDIX D. COMPLEMENTARY MICROSTRUCTURAL TEM IMAGES FOR THE CU-ZN ALLOY. ....................... 115
APPENDIX E. COMPLEMENTARY MICROSTRUCTURAL TEM IMAGES. TAILORING Λ ..................................... 116
APPENDIX F. COMPLEMENTARY MICROSTRUCTURAL TEM IMAGES FOR THE AL ALLOYS ............................ 119
APPENDIX G. COMPLEMENTARY MICROSTRUCTURAL IMAGES OBTAINED BY USING CONVENTIONAL EBSD
FOR AL AND AL-5.3WT.%MG .................................................................................................. 120

V

List of Figures
FIGURE 1. (A) PERFECT FCC CRYSTAL STRUCTURE (AB C ABC … … A BC A BC ). (B) INTRINSIC SF, REMOVAL OF AN A LAYER
(ABCABCBCABC). (C) EXTRINSIC SF, INCLUSION OF A B LAYER (ABCABCBABC)................................................................ 7
FIGURE 2. SCHEMATIC REPRESENTATION OF A TWIN, THE TWIN PLANE IS THE C LAYER (ABCABCBACBA). ....................... 9
FIGURE 3. ANNEALING TWINS IN 70:30 BRASS, CTBS ARE PRESENT ACROSS THE GRAIN, WHILE ITBS ARE AT THE EDGE OR AS
STEPS, ADAPTED FROM [47]. ................................................................................................................................ 11
FIGURE 4. TEM IMAGE AND CORRESPONDING INSET SAED PATTERN FOR A CU-AL ALLOY UNDER DYNAMIC PLASTIC
DEFORMATION, THE SAED PATTERN SHOWS THE TYPICAL DOUBLE HEXAGON THAT CORROBORATES THE PRESENCE
OF TWINS [49]. .................................................................................................................................................... 12
FIGURE 5. DEFORMATION TWINNING MECHANISM. (A) NORMAL STACKING SEQUENCE FOR A PERFECT FCC CRYSTAL, THE ARROW INDICATES
A SHEAR FORCE THAT IS APPLIED ON LAYER A, AND WILL GENERATE A PARTIAL SLIP. (B) RESULTING STACKING SEQUENCE AFTER A
PARTIAL SLIP, THE ARROW INDICATES A SHEAR FORCE THAT WILL BE APPLIED ON LAYER C. (C) STACKING SEQUENCE AFTER 2
CONSECUTIVE PARTIAL SLIPS, THE ARROW INDICATES A FORCE THAT IS GOING TO BE APPLIED ON THE LAYER B. (D). A TWIN THAT HAS
BEEN PRODUCED BY A SEQUENCE OF PARTIAL SLIPS. ........................................................................................................ 13
FIGURE 6. (A) TEM IMAGE OF ELECTRODEPOSITED NT-CU WITH A Λ= 36 NM AND EQUIAXED GRAIN SIZE 400-500 NM [50],
(B) FIB IMAGE OF SPUTTERED NT-CU WITH Λ= 40 NM AND COLUMNAR GRAIN SIZE 500-800 NM, THE ARROW
INDICATES THE GROWTH DIRECTION [5]................................................................................................................ 14
FIGURE 7. SCHEMATIC OF A SPUTTERING CHAMBER. ....................................................................................................... 16
FIGURE 8. THORTON STRUCTURAL ZONE DIAGRAM (SZD) THAT DESCRIBES THE MICROSTRUCTURAL CHANGES IN FILMS AS A
FUNCTION OF SYNTHESIS PARAMETERS[53] . ......................................................................................................... 18
FIGURE 9. ILLUSTRATION OF A SMALL NUCLEUS THAT IS EITHER PERFECT OR HAVE A SF. ................................................. 19
FIGURE 10. REDUCTION OF
*
twin
r AS A FUNCTION OF THE TWIN BOUNDARY ENERGY. THIS CALCULATION WAS PERFORMED
BY KEEPING A CONSTANT DEPOSITION RATE OF 0.4 NM/SEC [55]. ........................................................................... 21
FIGURE 11. DIFFERENCE IN THE CRITICAL RADIUS FOR FAULTED NUCLEI AND PERFECT NUCLEI AS A FUNCTION OF THE
DEPOSITION RATE. THIS CALCULATION WAS PERFORMED FOR SS, WHICH HAS LOW SFE (20MJ/M
2
) [55]. ................ 22
FIGURE 12. SUBSTRATE COOLING STAGE AND SUBSTRATE HOLDER .................................................................................. 26
VI

FIGURE 13. SUBSTRATE HEATING STAGE ......................................................................................................................... 28
FIGURE 14. 1 INCH THERMOCOUPLE WAFER. .................................................................................................................. 30
FIGURE 15. XRD PATTERN OF A CU FILM. SEVERAL PEAKS ARE INDEXED AND ARE TYPICAL OF A RANDOM TEXTURE. ........ 31
FIGURE 16. FIB CROSS-SECTIONAL IMAGE OF A MAGNETRON SPUTTERED CU FILM. IT IS POSSIBLE TO OBSERVE SEVERAL
GRAINS THAT HAVE DIFFERENT CRYSTALLOGRAPHIC ORIENTATIONS, NOTE THAT INDIVIDUAL GRAINS HAVE DIFFERENT
GRAY SCALE COLORS, WHILE BEFORE AND AFTER A TB THERE IS A CHANGE IN CONTRAST WITHIN THE GRAIN A
(ENCLOSED BY THE RED SHAPE), AND TBS ARE INDICATED BY THE SMALL ARROWS. THE VERTICAL ARROW INDICATES
THE GROWTH DIRECTION ...................................................................................................................................... 32
FIGURE 17. CONVENTIONAL (A) AND TRANSMISSION (B) EBSD MODES. ADAPTED FROM [69, 70]. .................................... 34
FIGURE 18. SIGNALS THAT CAN BE GENERATED FROM THE INTERACTIONS OF A HIGH ENERGY ELECTRON BEAM AND A THIN
SPECIMEN. ........................................................................................................................................................... 35
FIGURE 19. CU-10WT.%NI CROSS-SECTIONAL TEM VIEW. A) BRIGHT FIELD TEM AND INSET SELECTED AREA ELECTRON
DIFFRACTION (SAED) PATTERN; B) DARK FIELD IMAGE OF A) ACQUIRED BY SELECTING THE DIFFRACTION POINT
ENCLOSED BY A CIRCLE IN THE SAED PATTERN; C) HIGH RESOLUTION IMAGE OF THE SQUARE REGION IN B), NOTE
THE YELLOW LINES HIGHLIGHT THE CHANGE IN THE STACKING SEQUENCE, AND THE RED LINE IS THE TB. ............... 37
FIGURE 20. TEMPERATURE PROFILES COLLECTED DURING SPUTTERING OF CU, CU-4 WT.% AL, AND CU-6 WT.% AL AT 2
MTORR. NOTE THAT THE THREE COMPOSITIONS PRESENT SIMILAR TEMPERATURE PROFILES. .................................. 40
FIGURE 21. NORMALIZED XRD PATTERNS FOR CU AND CU-AL ALLOYS. .......................................................................... 40
FIGURE 22. CROSS-SECTIONAL TEM IMAGES SHOWING THE GRAIN SIZE AND TWIN STRUCTURE OF CU AND CU-AL SAMPLES
SPUTTERED AT 2 MTORR: (A) CU-6 WT.% AL WITH AN AVERAGE GRAIN SIZE OF ~36 NM, (B) ZOOMED IN VIEW OF (A)
SHOWING THE PRESENCE OF MANY SFS INDICATED BY THE STREAKING IN THE INSET SAED PATTERN. (C) CU-4 WT.%
AL WITH AN AVERAGE GRAIN SIZE OF ~51 NM, (D) ZOOMED IN VIEW OF (C), WITH A SIMILAR MICROSTRUCTURE TO
THE ONE OBSERVED IN FIGURE B, THE STREAKING IN THE INSET SAED PATTERN INDICATES MANY STACKING FAULTS
AND THIN TWINS, (E) CU WITH AN AVERAGE GRAIN SIZE OF ~56 NM, (F) ZOOMED IN VIEW OF (E) WITH THICK TWIN
SPACING; THE LACK OF STREAKING IN THE INSET SAED PATTERN INDICATES THE PRESENCE OF FEWER STACKING
FAULTS THAN IN B AND D. THE GROWTH DIRECTION IS VERTICAL FOR ALL THE IMAGES. ........................................... 42
VII

FIGURE 23. TWIN SPACING OR TWIN THICKNESS (Λ) DISTRIBUTIONS MEASURED FROM TEM IMAGES OF: (A) CU-6 WT.% AL,
(B) CU-4WT.% AL, AND (C) CU SAMPLES SPUTTERED AT 2 MTORR. THE AVERAGE ΛS WERE 2, 2.3, AND 11.7 NM,
RESPECTIVELY. ..................................................................................................................................................... 43
FIGURE 24. NORMALIZED XRD PATTERNS FOR CU-4 WT.% AL AT DIFFERENT SPUTTERING RATES. ................................... 46
FIGURE 25. CROSS-SECTIONAL FIB IMAGES FOR CU-4 WT.% AL AT DIFFERENT SPUTTERING RATES. (A) SPUTTERING RATE
0.4 NM/SEC. (B) SPUTTERING RATE 0.86 NM/SEC. C) SPUTTERING RATE 4.12 NM/SEC. THE ARROW INDICATES THE
GROWTH DIRECTION. ............................................................................................................................................ 47
FIGURE 26. CROSS-SECTIONAL TEM IMAGE SHOWING THE TWIN STRUCTURE OF CU-4 WT.% AL SPUTTERED AT A
DEPOSITION RATE OF 4.1 NM/SEC. THE AVERAGE TWIN SPACING IS 9.7 NM; THE INSET SAED PATTERN HAS LITTLE
STREAKING, WHICH INDICATES THE PRESENCE OF FEWER SFS AND THICKER TWINS................................................. 48
FIGURE 27. CRITICAL NUCLEATION RADIUS FOR FAULTED NUCLEI AND PERFECT NUCLEI AS A FUNCTION OF THE
DEPOSITION RATE. THIS CALCULATION WAS PERFORMED FOR SS AT A TEMPERATURE OF 473 K AND 300 K. ............ 50
FIGURE 28. NORMALIZED XRD PATTERNS FOR THE SPUTTERING CONDITIONS EXPLORED BY COOLING AND HEATING THE
SUBSTRATE ........................................................................................................................................................... 52
FIGURE 29. NORMALIZED XRD PATTERNS FOR CU-4.5 WT.% AL SAMPLES FABRICATED AT HIGH SPUTTERING RATE AND
INTERRUPTED SPUTTERING ................................................................................................................................... 53
FIGURE 30. CROSS-SECTIONAL FIB IMAGES SHOWING FULLY NT STRUCTURES WITH COLUMNAR GRAINS IN CU-4.5 WT.% AL
AT A HIGH SPUTTERING RATE AND WITH INTERRUPTED SPUTTERING. THE ARROW INDICATES THE GROWTH DIRECTION
FOR ALL THE SAMPLES. (A) CUAL#35, DUTY CYCLE 100%, GRAIN SIZE 196NM, (B) CUAL#37, DUTY CYCLE 50%,
GRAIN SIZE 213NM, (C) CUAL#38, DUTY CYCLE 33%, GRAIN SIZE 184NM, (D) CUAL#39, DUTY CYCLE 25%, GRAIN
SIZE 190NM, AND (E) CUAL#40, DUTY CYCLE 9%, GRAIN SIZE 74NM, (F) ZOOM IN VIEW OF CUAL#39 (YELLOW
RECTANGLE), DIFFERENCE IN CONTRAST WITHIN THE COLUMNAR GRAIN THAT REPRESENTS REVERSALS IN THE
STACKING SEQUENCE. .......................................................................................................................................... 55
FIGURE 31. REPRESENTATIVE CROSS-SECTIONAL TEM IMAGES OF THE TWINNED STRUCTURES WITHIN A CU-2WT.%AL
ALLOY OBTAINED AT DIFFERENT SPUTTERING CONDITIONS; THE INSET SAED PATTERNS SHOW THE TYPICAL DOUBLE
HEXAGON OF THE (110) ORIENTED TWINNED GRAIN. (A) SAMPLE 2AL#5, NOTING THE OVERALL COLUMNAR GRAIN
STRUCTURE. THE TWIN THICKNESS DISTRIBUTIONS HIGHLIGHT THE VOLUME FRACTION OF TWINS FOR EACH SAMPLE
IN IMAGES (B) TO (E); , (B) SAMPLE 2AL#1, Λ≈5.7 NM; (C) SAMPLE 2AL#4, Λ≈10 NM; (D) SAMPLE 2AL#6, Λ≈27 NM;
VIII

AND (E) SAMPLE 2AL#7, Λ≈35 NM. REPRESENTATIVE TEM IMAGES SHOWING THE Λ ACHIEVED FOR SAMPLES 2AL#2,
2AL#3, AND 2AL#5 ARE PRESENTED IN APPENDIX E). THE GROWTH DIRECTION IS VERTICAL FOR ALL THE IMAGES. . 62
FIGURE 32. REPRESENTATIVE CROSS-SECTIONAL TEM IMAGES OF THE TWINNED STRUCTURES FOR CU-AL AND CU-NI
ALLOYS OBTAINED AT DIFFERENT SPUTTERING CONDITIONS. THE INSET SAED PATTERNS SHOW THE TYPICAL DOUBLE
HEXAGON OF THE (110) ORIENTED TWINNED GRAIN. THE TWIN THICKNESS DISTRIBUTIONS HIGHLIGHT THE VOLUME
FRACTION OF TWINS FOR EACH SAMPLE. FOR CU-4WT.%AL (A) SAMPLE 4AL#2, Λ ~5.1 NM, AND (B) SAMPLE 4AL#5,
Λ≈15 NM; FOR CU-6WT.%AL (C) SAMPLE 6AL#2, Λ≈ 4.8 NM, AND (D) SAMPLE 6AL#7, Λ≈19 NM; FOR CU-
10WT.%NI (E) SAMPLE 10NI#1, Λ≈~15 NM, AND (F) SAMPLE 10NI#4, Λ≈31 NM. REPRESENTATIVE TEM IMAGES
SHOWING THE Λ ACHIEVED FOR SAMPLES 4AL#1, 4AL#3, 4AL#4, 6AL#1, 6AL#3, 6AL#4, 6AL#5, 6AL#6, 10NI#2
AND 10NI#3 ARE PRESENTED IN APPENDIX E. THE GROWTH DIRECTION IS VERTICAL FOR ALL THE IMAGES. SAMPLE
DETAILS ARE PRESENTED IN TABLE 2. .................................................................................................................... 62
FIGURE 33. REPRESENTATIVE TEMPERATURE PROFILES COLLECTED AT DIFFERENT SPUTTERING CONDITIONS FOR CU-AL
AND CU-NI ALLOYS DURING A TIME SPAM OF 60 MINUTES. THE TEMPERATURE PROFILES SHOW THE CHANGE IN
AVERAGE DEPOSITION TEMPERATURE BY THE VARIATION OF ONLY ONE SPUTTERING PARAMETER AS FOLLOWS: (A)
COMPARISON BETWEEN INTERRUPTED SPUTTERING AND CONTINUOUS SPUTTERING; (B) COMPARISON BETWEEN
COOLING, HEATING, AND AS-SPUTTERED; (C) COMPARISON BETWEEN TWO DIFFERENT SPUTTERING POWERS; AND (D)
COMPARISON BETWEEN TWO DIFFERENT DUTY CYCLES. ......................................................................................... 64
FIGURE 34. SCHEMATIC REPRESENTATION OF Λ COARSENING AS A FUNCTION OF COLUMNAR GRAIN GROWTH DURING FILM
DEPOSITION. (A) GRAIN A WILL GROW AT EXPENSE OF GRAIN B; (B) AND (C) CTBS ARE DETACHED FROM A HIGH
ENERGY BOUNDARY AND FORM AN ITB; AND (D) TWO CTBS DISAPPEARED BY THE MIGRATION OF THE ITB DEPICTED
IN FIGURE 4C ....................................................................................................................................................... 68
FIGURE 35. AVERAGE Λ AS A FUNCTION OF TEMPERATURE AND SFE. (A) Λ AS A FUNCTION OF TEMPERATURE FOR THE CU-
ALLOYS WITH A SFE < 13MJ/M2. (B) Λ AS A FUNCTION OF TEMPERATURE FOR THE CU-ALLOYS WITH A SFE >
13MJ/M2. C) PROTEAN Λ CONTOUR ZONE MAP SHOWING THE CHANGE IN TWIN THICKNESS AS A FUNCTION OF THE
SFE AND AVERAGE DEPOSITION TEMPERATURE; THE THREE ZONES DEPICTED IN FIGURES 5A AND 5B ARE
DELINEATED BY THE DASHED LINES IN FIGURE 5C. ................................................................................................ 70
FIGURE 36.REPRESENTATIVE CROSS-SECTIONAL TEM IMAGES OF COLUMNAR GRAINS WITH TBS OF: A) AL, B) AL-MG, AND
C) NI. TBS ARE MARKED BY THE RED DOTTED LINES. AL AND AL-MG HAVE A PREFERENTIAL INCLINED TB ANGLE Α
IX

~70°, IN NI THE INCLINATION OF THE TB ANGLE Α VARIES FROM 0° TO ~75°. THE BLUE ARROWS POINT OUT INCLINED
TBS THAT HAVE THE SAME Α ANGLE IN EACH IMAGE. .............................................................................................. 77
FIGURE 37. LOW MAGNIFICATION AND HIGH RESOLUTION IMAGES OF: AL (A AND B), AL-MG (C AND D), NI (E AND F). THE
COLUMNAR GRAINS ARE MARKED BY WHITE DASHED LINES, WHILE THE RED DOTTED LINES MARK AN INCLINED TB (A,
C, AND E). THE INSET (A, C, AND E) SAED PATTERNS SHOW THE TYPICAL DOUBLE HEXAGON FORMED DUE TO THE
PRESENCE OF TBS, THE SAED PATTERNS ARE DECORATED WITH BLUE AND MAGENTA DOTTED LINES TO FORM THE
DOUBLE HEXAGON AS A CONSEQUENCE OF A REVERSAL IN THE STACKING SEQUENCE OF THE MATERIAL, THE
DIFFRACTION SPOTS ARE INDEXED ACCORDING TO THE PLANE ORIENTATION. HIGH RESOLUTION IMAGES OF THE
SQUARE REGIONS IN A), C) AND D) ARE PRESENTED IN B), D) AND E) RESPECTIVELY. THE CTBS MARKED WITH RED
DOTTED LINES LIE IN A (1 1 -1) OR (-1 -1 1) PLANE, WHILE THE BLUE AND MAGENTA DOTTED LINES SHOW THE (1 1 1)
PLANES AT EACH SIDE OF THE TWO SIDES OF THE CTB. THE INSETS IN B), D) AND F) ARE FAST FOURIER TRANSFORMS
OF THE HIGH RESOLUTION IMAGE, NOTICE THE DOUBLE HEXAGON TYPICAL OF A CTB. .......................................... 79
FIGURE 38. ROTATION OF A COLUMNAR GRAIN WITH A TB INCLINED 70.5°. A) TB ALIGNED IN THE 110 ZONE AXIS, THE TB
PLANE IS PERPENDICULAR TO 110 ZONE AXIS. B) THE COLUMNAR GRAIN IS ROTATED 30 OVER THE [111] DIRECTION
WITH RESPECT TO A). C) THE COLUMNAR GRAIN IS ROTATED 60 OVER THE [111] DIRECTION WITH RESPECT TO A). D)
THE COLUMNAR GRAIN IS ROTATED 90 OVER THE [111] DIRECTION WITH RESPECT TO A). THE CORRESPONDING
SIMULATED DIFFRACTION PATTERNS SHOW THE TYPICAL DOUBLE HEXAGON OF A TB. NOTICE THAT ONLY WHEN THE
TB PLANE IS PERPENDICULAR TO THE 110 ZONE AXIS THE DIFFRACTION PATTERN PROVIDE INFORMATION OF TWO
DISTINGUISHABLE STACKING SEQUENCES. ............................................................................................................. 83
FIGURE 39. NORMALIZED XRD PATTERNS OF THE AL, AL-MG, AND NI FILMS. AL AND AL-MG HAVE A STRONG {111}
TEXTURE, WHILE NI HAS A RANDOM TEXTURE. ....................................................................................................... 84
FIGURE 40. T-EBSD SCANS OF SINGLE COLUMNAR GRAINS. A) AL, B) AL-MG), C) AND D) NI. NOTICE THAT IN AL AND AL-
MG THE CHANGE IN TEXTURE OCCURRED BETWEEN {111} AND {115} PLANES, WHILE IN NI THE CHANGE IN TEXTURE
IS FROM RANDOM PLANE ORIENTATIONS. TWO EXAMPLES ARE GIVEN FOR THE NI FILM TO SHOW HORIZONTAL AND
INCLINED CTBS. .................................................................................................................................................. 86
FIGURE 41. NORMALIZED XRD PATTERNS FOR CU AND CU-ALLOYS. ............................................................................. 110
FIGURE 42. NORMALIZED XRD PATTERNS FOR AL AND AL-ALLOYS. .............................................................................. 111
FIGURE 43. NORMALIZED XRD PATTERNS FOR NI AND INCONEL600. ............................................................................ 111
X

FIGURE 44. CROSS-SECTIONAL FIB IMAGES SHOWING NT STRUCTURES IN CU. A)CU#13, B)CU#15. NOTICE THE CHANGE IN
CONTRAST CHARACTERISTIC OF TWIN BOUNDARIES, AND THE DIFFERENCE IN THE MICROSTRUCTURAL FEATURES
BETWEEN THE TWO IMAGES. ................................................................................................................................ 112
FIGURE 45. CROSS-SECTIONAL FIB IMAGES SHOWING NT STRUCTURES WITH COLUMNAR GRAINS IN CU ALLOYS.A)
CUAL#69,B)CUAL#72, C)CUNI#8, D)CUNI#22. IN E) AND F) CUZN#14 AND CUZN#15, RESPECTIVELY, NOTICE THE
COLUMNAR GRAINS ARE NOT AS SHARP AS IN THE OTHER CU-ALLOYS. NOTICE, IN G)CUAG#7 NO PRESENCE OF TWIN
BOUNDARIES IS OBSERVABLE WITH THE FIB. ....................................................................................................... 113
FIGURE 46 CROSS-SECTIONAL FIB IMAGES SHOWING NT STRUCTURES WITH COLUMNAR GRAINS IN NI AND
INCONEL600:A)NI#1,B)NI#16, C)INCONEL600#1.NOTICE THAT IN NI#1 AND NI#16 THE GRAIN SIZE IS MUCH
SMALLER THAN IN NI#16..................................................................................................................................... 114
FIGURE 47. HORIZONTAL AND INCLINED TBS. A) CUZN#12, B) CUZN#13. THE WHITE ARROW INDICATES THE FILM
GROWTH DIRECTION. .......................................................................................................................................... 115
FIGURE 48. REPRESENTATIVE CROSS-SECTIONAL TEM IMAGES OF THE TWINNED STRUCTURES WITHIN A CU-2WT.%AL
ALLOY OBTAINED AT DIFFERENT SPUTTERING CONDITIONS. (A) SAMPLE 2AL#2, Λ≈6 NM; (B) SAMPLE 2AL#3, Λ≈8 NM;
AND (C) SAMPLE 2AL#5, Λ≈10 NM. THE GROWTH DIRECTION IS VERTICAL FOR ALL THE IMAGES. ........................... 116
FIGURE 49. REPRESENTATIVE CROSS-SECTIONAL TEM IMAGES OF THE TWINNED STRUCTURES IN CU-4WT.%AL, CU-
6WT%AL, AND CU-10WT.%NI ALLOYS OBTAINED AT DIFFERENT SPUTTERING CONDITIONS. (A) SAMPLE 4AL#1, Λ≈2.5
NM; (B) SAMPLE 4AL#3, Λ≈5.1 NM; (C) SAMPLE 4AL#4, Λ≈10 NM; (D) SAMPLE 6AL#1, Λ≈2.1 NM; (E) SAMPLE 6AL#3,
Λ≈4.8 NM; (F) SAMPLE 6AL#4, Λ≈7.4 NM; (G) SAMPLE 6AL#5, Λ≈8 NM; (H) SAMPLE 6AL#6, Λ≈13 NM; (I) SAMPLE
10NI#2, Λ≈17 NM; AND (J) SAMPLE 10NI#3, Λ≈20 NM; THE GROWTH DIRECTION IS VERTICAL FOR ALL THE IMAGES.
......................................................................................................................................................................... 117
FIGURE 50. REPRESENTATIVE CROSS-SECTIONAL TEM IMAGES OF THE SAMPLES THAT PRESENTED HIGH CONCENTRATION
OF ITBS IN A FEW GRAINS. (A) SAMPLE 2AL#5; (B) SAMPLE 4AL#5; (C) SAMPLE 6AL#7; (D) SAMPLE 10NI#1; (E)
SAMPLE 10NI#2; AND (F) SAMPLE 10NI#3. THE ARROWS INDICATE THE LOCATION OF THE ITBS. THE GROWTH
DIRECTION IS VERTICAL FOR ALL THE IMAGES. ..................................................................................................... 118
FIGURE 51. REPRESENTATIVE CROSS-SECTIONAL TEM IMAGES OF COLUMNAR GRAINS WITH TBS OF: A) AL2024#2, AND B)
AL5456#4. TBS ARE MARKED BY THE RED DOTTED LINES. ................................................................................... 119
XI

FIGURE 52. CONVENTIONAL EBSD SCANS FROM THE CROSS-SECTION OF: A) AL#17, B)AL-5.3MG#14. NOTICE THAT BOTH
SAMPLES HAVE STRONG {111} TEXTURE MIXED WITH {115} TEXTURE. SOME Σ3 CTBS ARE HIGHLIGHTED BY THE
THICK RED LINES. ............................................................................................................................................... 120

XII

List of Tables
TABLE 1. SFES FOR DIFFERENT FCC METALS AND ALLOYS................................................................................................ 8
TABLE 2. SPUTTERING CONDITIONS, SFE, GRAIN SIZE AND RESIDUAL STRESS FOR CU AND CU-AL ALLOYS. ..................... 39
TABLE 3. SPUTTERING CONDITIONS, SFE, GRAIN SIZE AND RESIDUAL STRESS FOR CU-4 WT.% AL (SFE=13MJ/M
2
). ....... 46
TABLE 4. SPUTTERING CONDITIONS EXPLORED FOR COOLING AND HEATING THE SUBSTRATE. ........................................... 51
TABLE 5. SPUTTERING CONDITIONS AT A HIGH SPUTTERING RATE AND INTERRUPTED SPUTTERING FOR CU-4.5 WT.% AL. . 53
TABLE 6. SUMMARY OF SPUTTERING CONDITIONS USED TO SYNTHESIZE HIGHLY NT CU ALLOYS. ....................................... 58
TABLE 7. AVERAGE MICROSTRUCTURAL PROPERTIES AND TOTAL STORED ENERGY OF HIGHLY NT CU ALLOYS SAMPLES. ..... 59
TABLE 8. MICROSTRUCTURAL PROPERTIES OF THE AL, AL-MG AND NI FILMS. ................................................................. 81
TABLE 9. T-EBSD ANALYSIS OF THE CTBS IN AL, AL-MG, AND NI FILMS. ........................................................................ 87
TABLE 10. CU SPUTTERED SAMPLES ............................................................................................................................... 99
TABLE 11. CU-AL ALLOYS SPUTTERED SAMPLES .............................................................................................................. 99
TABLE 12. CU-NI ALLOY SPUTTERED SAMPLES .............................................................................................................. 103
TABLE 13. CU-ZN ALLOY SPUTTERED SAMPLES ............................................................................................................. 104
TABLE 14. CU-AG ALLOY SPUTTERED SAMPLES ............................................................................................................. 105
TABLE 15. AL SPUTTERED SAMPLES .............................................................................................................................. 105
TABLE 16. AL-CU ALLOY SPUTTERED SAMPLES ............................................................................................................. 105
TABLE 17. AL-NI ALLOY SPUTTERED SAMPLES ............................................................................................................... 106
TABLE 18. AL2024 ALLOY SPUTTERED SAMPLES ............................................................................................................ 106
TABLE 19. AL6013 ALLOY SPUTTERED SAMPLE ............................................................................................................. 106
TABLE 20. AL5052 ALLOY SPUTTERED SAMPLES ............................................................................................................ 106
TABLE 21. AL5083 ALLOY SPUTTERED SAMPLES ............................................................................................................ 107
TABLE 22. AL5456 ALLOY SPUTTERED SAMPLES ............................................................................................................ 107
TABLE 23. AL-5.3WT.%MG ALLOY SPUTTERED SAMPLES ............................................................................................... 107
TABLE 24. NI SPUTTERED SAMPLES .............................................................................................................................. 109
TABLE 25. INCONEL600 SPUTTERED SAMPLES............................................................................................................... 109
XIII

1

1 Abstract
Nanotwinned (nt) metals are materials that consist of several Σ3 twin boundaries (TBs) located
within multiple grains in the overall microstructure, where the distance between any two
consecutive TBs is within the range of 1 nm to 100 nm. In general, nt metals have attractive
mechanical, thermal, electrical and/or chemical properties. The aforementioned properties of nt
metals can be engineered based on the modification of two nanostructural features: (i) the density
of TBs in the material, and (ii) the average distance between TBs or the twin thickness. These
favorable and tunable properties of nt metals have encouraged extensive research during the last
decade, specifically on fabrication methods that control nanostructural features. Nevertheless, the
synthesis of nt metals has been limited due to stacking fault energy (SFE) restrictions, where
many studies focus only on low SFE materials (< 45 mJ/m
2
) that are prone to twinning. To date,
no studies have shown explicit control of the nanostructural features in nt metals. This
dissertation discusses two main points: (i) the synthesis of nt metals with low, intermediate, and
high SFEs by using magnetron sputtering; and (ii) the control of nanostructural features in nt
metals with low and intermediate SFEs by adjusting the sputtering parameters. The general
results are summarized as follows: 1) a study in low SFE materials (6 - 45 mJ/m
2
) was conducted
by synthesizing thick Cu and Cu-Al films (film thickness ~4 μm). It was revealed that under
identical sputtering conditions the twin thickness increased as the SFE increased. Contrary to
theoretical predictions, the experimental results indicated that the twin thickness can be increased
by using higher deposition rates; 2) a comprehensive study on Cu and Cu alloy films (film
thickness >15 μm) elucidated three main mechanisms to describe twin nucleation and TB
mobility, where the interplay of the mechanisms granted control of the twin thickness in nt
metals. The results from this study were summarized in a protean contour zone map that can be
2

used as a versatile guide to synthesize fully nt metals with tailored twin thicknesses in materials
with a wide range of stacking fault energies (6 - 60 mJ/m
2
); 3) a study in high SFE metals (SFE
> 125 mJ/m
2
) was conducted by synthesizing Ni, Al, and Al alloys thick films (film thickness
>10 μm). Several characterization techniques were used to investigate the nature of twin
boundaries in high SFE materials. The results provide an alternative perspective on the
evaluation of twin boundaries, where new directions can be explored on the synthesis of nt
metals with high SFEs. Overall, this dissertation contributes towards expanding the working
space of nt metals, builds on the characterization of twin boundaries, and provides a guide to
tune nanostructural features of nt metals for potential future applications.

3

2 Introduction
In nanotwinned (nt) metals a high amount of Σ3 twin boundaries (TBs) are present in several
grains of the material microstructure, where the distance between consecutive TBs is < 100 nm.
During the last decade, nt metals have been the subject of extensive research due to the potential
enhancement of mechanical, thermal, electrical and/or chemical properties. For instance, nt Cu
has shown high strength and good ductility compared to UFG or coarse grain Cu [1-3].
Additionally, nt metals have shown to improved corrosion resistance, thermal stability, and
tolerance to radiation damage [4-6].
The attractive properties of nt metals have been shown to stem from the combination of two
nanostructural features: (i) the density of TBs in the material; and (ii) the twin thickness (λ),
which is the average distance between two consecutive TBs. Two particular studies on nt Cu
highlight the influence of the nanostructural features in nt metals. Zhao et al showed that the
corrosion resistance of nt Cu improved as the density of TBs increased [4], while Lu et al
showed that the strength of nt Cu is improved compared to UFG Cu, and that the ductility varies
for different λ [2]. Despite the importance of these favorable properties of nt metals, explicit
control of the nanostructural features has not been achieved. Therefore, to build on the
understanding of nt metals and their tunable properties, it is imperative to control the twin
density and tailor λ. This work can only be achieved by a comprehensive study from a synthesis
perspective.
Different techniques have been utilized to produce materials with a large density of Σ3
boundaries, such as annealing twins by heat treatment [7], deformations twins by plastic
deformation [8], and growth twins by electrodeposition or sputtering [3, 9]. One of the crucial
4

parameters for the synthesis of nt materials by any of the aforementioned methods is the SFE,
which is an intrinsic property of the material. Since low SFE materials are prone to twinning, the
synthesis of nt materials has been limited to metals and alloys systems with low SFEs (<
45mJ/m
2
), such as Cu, Ag, Cu-Al, Cu-Zn, and stainless steel among others [1-3, 8-15].
The present study focuses on the synthesis of TBs by magnetron sputtering in a variety of
materials (Cu-Al alloys, Cu, Cu-Ni alloys, Ni, Al and Al alloys) with a wide range of SFEs (6
mJ/m
2
- 166 mJ/m
2
). Transmission electron microscopy (TEM), electron backscattering
diffraction (EBSD), transmission EBSD, and focus ion beam (FIB) were used to evaluate the
density of TBs and the variation of λ. The main results of the comprehensive study conducted in
this dissertation provide further understanding of TB formation, improved TB density control,
and a strong guide to tailor λ. Moreover, the working space of nt materials can be expanded as
the produced thick films can be used as optimized microstructural configurations to further
investigate the properties of nt metals.

5

3 Background
3.1 Nanotwinned metals and their properties
Improving a material’s properties such as strength, ductility, thermal stability, corrosion
resistance, and conductivity need to be taken into account when designing materials at the
nanoscale [16]. At the macro-scale, the concept of grain boundary engineering is based on the
idea that increasing the fraction of special boundaries can be used to enhance the bulk properties
of structural and functional materials [17]. At the nanoscale, nt metals have been shown to have a
large density of special boundaries (Σ=3 TBs), where λ can play the role of the grain size [2, 18,
19]. Twins affect the properties of the material by reorienting the crystal lattice and introducing a
boundary [20]. The reoriented structures can alter the crystallographic orientation of the material,
which implies that a nt metal can respond differently to an applied force compared to a
nanocrystalline metal without twins [20]. Moreover, the introduced TBs are new obstacles to slip
dislocations and can enhance the mechanical behavior of the material [20]. To highlight the
importance of nt metals and the relevance to this dissertation, a brief overview of some attractive
properties of nt metals is presented, followed by a detailed description of TBs at the atomic and
microscale, a general view of different methods to synthesize nt materials with a special
emphasis on the synthesis of nt metals by magnetron sputtering.
3.1.1 Mechanical properties of nt metals
Several studies have shown that nt metals show a unique combination of mechanical properties,
such as high strength, good ductility, and high hardness [18, 20-23]. For example, the hardness
and tensile strength of nt metals can reach values several times higher than the bulk metal. nt Cu
hardness (~3 GPa) is six times higher than bulk Cu (~500 MPa) [11, 18]. This remarkable
6

behavior of nt metals is attributed to the interaction of dislocations with the TBs, while low λ
limits dislocation motion and dislocation pileups [20]. Therefore, λ can play the role of grain size
and in some studies it has been shown that the hardness of the material increases as λ decreases
[2, 18, 19]. The relationship between hardness and λ follows the Hall-Petch effect, where the
strength of the material increases as the grain size decreases [24].
3.1.2 Thermal stability of nt metals
The thermal stability of nt metals stems from the high density of TBs, which have low boundary
energy. For instance, in Cu, a coherent TB has energy ~30 mJ/m
2
, while a typical GB has ~700
mJ/m
2
. Hence the driving force for GB coarsening is higher than for TB coarsening [25, 26].
Multiple studies have shown experimental evidence of the microstructural thermal stability of
different nt metals compare to coarse or ultra fine grain metals [26-28]. Zhao et al compared
sputtered nt Cu with sputtered ultra fine grain (UFG) Cu [27]. It was observed that after a 3 hour
heat treatment at 200 °C the grain size in UFG Cu was three times higher than the as-sputtered
UFG Cu, while the nt Cu grain size remained unchanged. For the UFG Cu film, the grain size
increased as the heat treatment temperatures were increased (300°C, and 400°C), while the grain
size in the nt Cu film remain stable until a heat treatment at 300°C [27]. Since it has been shown
that TBs can enhance the properties of a material, it is imperative to describe the fundamentals of
TBs and their atomic arrangement. The following sections provide information of stacking faults,
and the role of SFE on TB formation.
3.2 Stacking faults and stacking fault energy
Stacking faults (SFs) are a type of imperfection (planar defect) within a crystalline structure, and
may occur during solid state processing, phase transitions, crystal growth, and during
7

deformation [29]. In a face-centered cubic (FCC) system there are 3 unique "voids" or interstices
between the atoms, where a layer of atoms can be arranged; the 3 "voids" are called A, B, or C.
Figure 1a shows the perfect stacking of an FCC material where there are no SFs and the layer
stacking is denoted by the sequence ABCABC……ABCABC. When a SF occurs, the stacking
sequence changes, as can be seen in Figure 1b and Figure 1c for intrinsic and extrinsic SFs,
respectively. An intrinsic SF can be understood as removing one of the layers from a perfect
stacking sequence (ABCBCABC), while an extrinsic SF is the addition of an extra layer to a
perfect stacking sequence (ABCBABC). The SF is an interruption of the normal stacking
sequence of atomic planes and it is a consequence of the balance between for the repulsive force
between two partial dislocations and the attractive force due to surface tension [30]. Overall, a
SF is needed before a TB can be formed.

Figure 1. (a) Perfect FCC crystal structure (ABCABC……ABCABC). (b) Intrinsic SF, removal of an A layer
(ABCABCBCABC). (c) Extrinsic SF, inclusion of a B layer (ABCABCBABC).
8

The stacking fault energy (SFE) is the increased energy per unit area above that of the unfaulted
stacking sequence. Typical SFE values vary from 1-1000 mJ/m
2
, where the lower the SFE, the
higher the probability of a SF and TB formation [29, 31]. While SFs can occur in different
crystal structures such as body-centered cubic (BCC) and hexagonal close pack (HCP), this study
will focus on SFs and TBs in FCC metals. Table 1 shows the SFEs of some metals and alloys
including materials used in this dissertation.
Table 1. SFEs for different FCC metals and alloys.
Material
SFE
(mJ/m
2
)
Cu-6wt.%Al 6 [32]
Cu-4wt.%Al 13 [32]
Co 15 [33]
Ag 17 [12]
Cu-5wt.%Zn 30 [34]
Cu-2wt.%Al 37 [35]
Cu 45 [36]
Au 45 [33]
Si 60 [37]
Cu-20wt%Ni 74 [34]
Ni-2.5wt%Cu 86 [34]
Mg 125 [33]
Ni 125 [38]
Zn 140 [33]
Al 166 [39]
Pd 175 [34]

3.3 Description of twin boundaries
A twin is composed of a matrix and its mirror, which are two structures separated by a planar
defect (twin plane or coherent twin boundary) [29]. Suppose that a crystal is arranged with a
perfect stacking sequence, as shown in Figure 1a. If, at some point in the stacking sequence, a SF
is formed, it is possible that after the SF occurrence, the subsequence stacking sequence is fully
reversed. The resulting structure is called a twin, as shown in Figure 2 for the stacking sequence
9

ABCABCBACBA, the bolded ABCAB is the perfect stacking sequence, the twin plane or TB is
the underlined C layer, and the reversed stacking sequence is BACBA.

Figure 2. Schematic representation of a twin, the twin plane is the C layer (ABCABCBACBA).

TBs are also a type of grain boundary (GB), and can be categorized by the coincidence site
lattice (CSL) like any other GB. Coincidence sites are the atom positions shared by two
neighboring grains. In general, the density of coincident sites is used to classify the CSL GBs by
using Equation 1 [40].
lattice crystal of cell elementary volume
CSL of cell elementary volume
 
Equation 1

where Σ is the inverse of the density of coincidence sites. The lower the Σ value, the higher the
density of coincident sites. Low angle grain boundaries (<15°) can be classified as Σ=1, while
special GBs, which have high misorientation angle (>15°) and low boundary energy have a Σ
10

value between 3 and 29. In principle, all the special boundaries are TBs, since the atomic
arrangements before and after the boundary are mirrored [41]. However, in practice, only the
CSL that has Σ=3 is referred to as a TB [41]. A Σ3 TB can be coherent or incoherent depending
on the TB family plane, {111} or {211} respectively [40-42], where coherent TBs have a higher
density of coincidence sites per unit GB area than incoherent TBs [41, 43, 44].
Due to the high number of coincident lattice sites and the high density of coincident sites per unit
GB area, coherent TBs (CTBs) are highly ordered structures that are densely packed. When a
significant amount of CTBs are incorporated into the microstructure of a material, low-energy
segments that reduce the overall energy of the GB network are generated. Thus in grain
boundary engineered materials, CTBs play a significant role in the material performance. For
example, CTBs can truncate intergranular corrosion through the GB network, enhance thermal
stability or increase the strength of a material, as has been observed for nt Cu [4, 27, 45]. There
are various synthesis methods used to fabricate materials with CTBs, typically categorized as
top-down and/or bottom-up approaches. Top-down approaches synthesize twins by using solid
state phase transformation, annealing, or plastic deformation, while bottom-up approaches
synthesize twins during solidification from liquid or vapor phase (growth twins) [46]. The
following section describes the synthesis of twin structures in FCC metals by using both top-
down and bottom-up approaches.
3.4 Synthesis methods for twin boundaries
3.4.1 Annealing twins
Annealing twins may form during recovery, primary crystallization, or during grain growth
following recrystallization. Figure 3 shows an image of annealing twins, where CTBs are present
11

across the grain and ITBs are present at the edges of the grains or appear as steps of CTBs that
do not cross the entire grain. Despite the fact that there is not a complete theory to describe the
formation of twins during annealing, their appearance could be attributed to the lowering of grain
boundary energy and/or the dislocation density (twin density increases with initial dislocation
density) [47].

Figure 3. Annealing twins in 70:30 Brass, CTBs are present across the grain, while ITBs are at the edge or as steps,
adapted from [47].

3.4.2 Deformation twins
Twins that occur during deformation are attributed to a shear that is uniformly distributed over a
small volume, where the atoms move a fraction of the interatomic spacing relative to each other
[48]. An example of this type of twinning is shown in Figure 4, where the twin planes appear as
lines and the differences in contrast represent reversals in stacking sequence.
12

Figure 4. TEM image and corresponding inset SAED pattern for a Cu-Al alloy under dynamic plastic deformation,
the SAED pattern shows the typical double hexagon that corroborates the presence of twins [49].

Figure 5 shows the mechanism for obtaining a twinned structure by deformation. This
mechanism involves introducing a succeeding sequence of intrinsic SFs (partial slips of close
packed layers) above an initial SF during plastic deformation. The process can be summarized in
3 steps. Step 1 is shown in Figure 5a, which has a normal stacking sequence for a perfect FCC
crystal. Suppose that an intrinsic SF (partial slip) occurs in the layer A that has been marked with
an arrow; the result is show in Figure 5b, where all the layers above the SF have moved to the
next position. Step 2 introduces a new partial slip on the layer above the one with the previous
intrinsic SF (layer C, marked with an arrow). Figure 5c shows the resulting stacking sequence.
Step 3 is identical to step 2, except the indicated B-layer that will have another partial slip.
Finally, the repetition of partial slips will produce a deformation twin, as shown in Figure 5d.
13

Figure 5. Deformation twinning mechanism. (a) Normal stacking sequence for a perfect FCC crystal, the arrow
indicates a shear force that is applied on layer A, and will generate a partial slip. (b) Resulting stacking sequence
after a partial slip, the arrow indicates a shear force that will be applied on layer C. (c) Stacking sequence after 2
consecutive partial slips, the arrow indicates a force that is going to be applied on the layer B. (d). A twin that has
been produced by a sequence of partial slips.

3.4.3 Growth twins
Growth twins are formed when the stacking sequence is interrupted by the appearance of a SF in
the {111} direction during material deposition or the formation of metal layers. Due to this SF,
the stacking sequence of subsequent layers will be reversed from the original sequence. The
majority of growth twins have been synthesized either by electrodeposition or magnetron
sputtering. Figure 6a shows twins produced by electrodeposition, where the presence of equiaxed
grains can be observed and the twins cross the entire grain [50]. Figure 6b shows twins produced
by magnetron sputtering, where fully columnar grains can be observed and the twins extend
horizontally across the vertical columnar grains [5]. This study will focus on twins fabricated by
magnetron sputtering, and a description of the synthesis principle is presented in section 3.5.
14

Figure 6. (a) TEM image of electrodeposited nt-Cu with a λ= 36 nm and equiaxed grain size 400-500 nm [50], (b)
FIB image of sputtered nt-Cu with λ= 40 nm and columnar grain size 500-800 nm, the arrow indicates the growth
direction [5].

3.5 Magnetron sputtering processing
Sputtering is a process where high-energy particles are used to bombard the surface of a solid
material (referred to as the target) and the atoms from the target material are ejected in all
directions, due to the high-energy collisions. This forms a dense metal vapor in vicinity of the
target. A substrate can then be placed in the metal vapor and be coated with the ejected atoms
from the target [51].
Figure 7 shows a schematic of the sputtering process, where the substrates are facing the target
inside the vacuum chamber. The chamber is evacuated to pressures lower than 10
-6
Torr (1 Torr
≈ 133 Pa), at which point the target, also known as the cathode, is negatively polarized. An inert
gas, such as Ar, is introduced into the chamber and the Ar inflow is adjusted to obtain an
15

atmosphere that is capable of generating a plasma discharge (referred to as the Ar working
pressure). Typical values for the Ar working pressure are in the range of 1 to 15 mTorr. In
magnetron sputtering, a magnetic field is applied around the target, which forms electron traps
and confines the electrons to the vicinity of the target [51]. When the Ar is introduced into the
chamber, the electrons from the magnetic field ionize the Ar gas and generate the plasma. The Ar
ions are then attracted by the negative polarization of the target and the collisions between the Ar
ions and the target induce a cascade reaction, wherein target atoms are ejected in all directions,
generating a condensed metal vapor. Additionally, due to the collisions, more electrons are
generated and trapped in the magnetic field. This reinforces the Ar ionization process and
maintains the bombardment process. As previously stated, the substrate is placed in a region
where the metal vapor is formed. The distance from the substrate to the target is referred to as the
working distance. The shorter the distance between the target and the substrate, the higher the
number of target atoms which can arrive at and be deposited on the substrate [52].
16

Figure 7. Schematic of a sputtering chamber.

The microstructure produced by magnetron sputtering depends on many parameters, such as the
working pressure, control of target polarization (whether by voltage, current, or power), and the
working distance [52, 53]. These are three "independent" variables. Other variables that can be
measured during sputtering, such as film temperature or deposition rate, depend upon the
aforementioned three independent variables, among others. For instance, the deposition rate or
sputtering rate can be increased by either reducing the working distance, decreasing the working
pressure, or increasing the target polarization (higher operational powers). Despite the fact that
magnetron sputtering is a versatile technique, it is challenging to modify any measurable variable
17

without changing another. For example, to reduce the fabrication time of a film, it is possible to
increase the deposition rate by increasing the power. However, a high deposition rate will
increase the energy of the sputtered atoms, which may significantly increase the temperature of
the film, resulting in a drastic change in the film microstructure [53].
To illustrate the possible variations in a film microstructure due to the modification of the
sputtering parameters, the well-known Thornton structural zone diagram (SZD) is displayed in
Figure 8 [53]. The SZD is divided in 4 zones and it is dependent on the substrate temperature and
the Ar working pressure. The microstructure of the film can vary according to the substrate
temperature, while the Ar working pressure allows one to extend the different microstructure
zones to higher temperatures. For instance, thin films with columnar grains belong in zone II.
The SZD is an approximation of some typical features that can be observed in films
microstructures. Recently, in order to generalize the SZD, it has been proposed that the axes in
the SZD need to be modified. For example, since the Ar working pressure affects the film
temperature, it should be modified by a normalized energy that can describe the kinetic energy
during the sputtering synthesis [54].
18

Figure 8. Thorton structural zone diagram (SZD) that describes the microstructural changes in films as a function
of synthesis parameters[53] .

3.5.1 Synthesis of twins by magnetron sputtering
Despite the observation of twins in multiple FCC metals such as Ag, Cu, and stainless steel (SS),
full understanding of their formation and properties, such as twin nucleation, twin density, and λ,
requires further investigation. In general, during sputtering, thin film microstructures develop by
the nucleation, growth, and coalescence of small islands. According to Zhang et al., the amount
of twins and λ in a material that has been grown by sputtering can be correlated to the relative
difference between the radius of a perfect nucleus and a nucleus with a SF defect or TB. Figure 9
depicts a small nucleus that can be perfect or have a SF. In this illustration, only two stacked
atomic layers are displayed.
19

Figure 9. Illustration of a small nucleus that is either perfect or have a SF.

The radii of a perfect nucleus and a twinned nucleus can be calculated using Equation 2 and
Equation 3 [55], respectively, assuming that the nuclei have disk-like shapes. In these equations,
 is the surface energy,
twin
 is the twin boundary energy (TBE  SFE/2), h is the {111}
interplanar spacing, and
v
G  is the bulk free energy per unit volume









 
s
v
P
mKT J KT
G
 2
ln , where K is the Boltzmann constant, T is the temperature,  is
the atomic volume, J is the deposition flux, m is the atomic mass of the interacting species,
and
s
P is the vapor pressure above the solid.
v
perfect
G
r



*

Equation 2
20

h
G
r
twin
v
twin


 

*

Equation 3

By using these equations it is possible to arrive at a simple expression relating the two radii:















v
twin r
G h r
radii critical between difference relative

Equation 4
When the relative difference between the two radii is very small (<10% in the case of stainless
steel), then the probability of either nucleus forming and growing is similar from an energetic
standpoint [56]. By examining Equation 4, it is conspicuous that there are three variables that can
be controlled in order to reduce the difference between the two radii:
twin
 (proportional to the
SFE), J (proportional to the deposition rate during sputtering), and T , which is highly
dependent on sputtering parameters.
The analytical model described above provides an approach for estimating the probability of
growth twins from a thermodynamic point of view, where the idea is to reduce the energy
required to nucleate a faulted nucleus to values that are comparable to those required to nucleate
a perfect nucleus. Therefore, sputtering materials with low SFE values, at high deposition rates,
and at "high" temperatures could lead to the synthesis of nt structures. On the other hand, many
materials such as Al and Ni have high SFEs. As a result, nt structures cannot be easily achieved
in these materials and further control of the synthesis method is needed in order to compensate
for the high SFE.
21

3.5.2 Nucleation and growth of twin boundaries using magnetron sputtering
Previously it was mentioned that a reduction in the energy that is needed to produce a faulted
nuclei is the key parameter to promote SF formation. Since SFE and TB energy (
twin
 ) are
intrinsic properties of each material, their values could be lowered by alloying the original base
metal [57, 58]. For example, in Cu, small additions of a variety of elements, such as Al, Si, Zn,
and Mn, can drastically reduce the SFE [8, 33, 59], although alloying does not always lead to a
reduction of the SFE, as has been shown in the case of Cu-Ni alloys [34]. By selecting materials
with low or reduced SFEs, the probability of developing twins is enhanced. Figure 10 shows
*
twin
r
as a function of
twin
 at a constant sputtering rate of 0.4 nm/sec. It should be noted that reducing
twin
 to values below 10 mJ/m
2
reduces
*
twin
r to values that are comparable to those of
*
perfect
r ,
which remains constant. Thus, low SFE materials are favorable for promoting the growth of a
nuclei with twin defects [55].

Figure 10. Reduction of
*
twin
r as a function of the twin boundary energy. This calculation was performed by keeping
a constant deposition rate of 0.4 nm/sec [55].

22

The effect of the deposition rate on
*
twin
r and
*
perfect
r at a constant SFE=20mJ/m
2
is plotted in
Figure 11. The sizes of the two critical radii are clearly reduced by increasing the deposition rate.
At high deposition rates (1 nm/sec) the difference between the radii is <10%, which means that
the energy for the faulted nucleus is similar to that of a perfect nucleus. Therefore, the formation
of twins is more favorable by increasing the deposition rate [55].

Figure 11. Difference in the critical radius for faulted nuclei and perfect nuclei as a function of the deposition rate.
This calculation was performed for SS, which has low SFE (20mJ/m
2
) [55].

It is assumed that a high probability of TB formation leads to high density of TBs and a low λ.
Thus, low SFE materials synthesized at a high sputtering rate should have a low λ, where λ
decreases as the deposition rate is increased [55]. However, Velasco et al showed that this is not
the case, and contrary to theoretical predictions, λ increases at high deposition rates [60].
Therefore, this analytical model may be useful for estimating the probability of twinning during
deposition but not to determine λ. From an experimental standpoint, the growth of nt structures is
challenging and it is difficult to control the twin formation by only changing the deposition rate,
which means that other variables must be take into account [61]. The two aforementioned studies
23

suggest that further research is needed to understand the twinning phenomena in low SFE
materials. Thus, the influence of other factors must be experimentally investigated and probably
isolated, such as the influence of film temperature, the SFE, and the kinetics of film growth.
3.5.3 Twins in high SFE materials Al and Ni
To date, the synthesis of twins in high SFE materials has been limited and generally includes
techniques which are different from those used for low SFE materials. For example, the use of
intermediate layers of another material and/or epitaxial growth allowed the synthesis of some
TBs in Al [20]. The use of an intermediate ceramic layer of TiN in a configuration Al/TiN/Al,
allowed the Al layers above and below a TiN layer to have a twin configuration when the
thickness of the TiN was less than 2 nm. In this case, the TiN played the role of a semi coherent
twin boundary [62]. Another study using epitaxial growth reported the production of Al nt
structures in a multilayer configuration, Al/Ag. Due to the low mismatch in the lattice parameter
between Ag and Al (>1%), the interfaces between Ag and Al layers are nearly coherent. In
essence, Ag has a low SFE and it is likely to twin, and the coherence at the Ag/Al interface
facilitates the lateral propagation of CTBs from the Ag layer into the Al layer [63]. Based on the
fact that the used techniques require multilayers rather than monolithic films, it is evident that the
synthesis of nt structures in high SFE materials is a challenging process. A recent study by Xue
et al showed the occurrence of few TBs in a low fraction of grains (<9%) in thin (<80 nm) Al
films synthesized by magnetron sputtering [64]. Despite the low density of TBs observed in the
study by Xue et al [64], these results have encouraged the scientific community to search for a
higher TB density in high SFE materials. TBs in pure Ni have been synthesized mainly by
annealing, and to a lesser extent by electrodeposition and sputtering [65-67]. Among these
studies only the electrodeposited Ni presented TBs with a λ < than 100 nm.
24

Overall, the synthesis of TBs in high SFE metals is challenging and requires further
development. The experimental evidence to date suggests that it is possible to overcome the SFE
restrictions, where higher densities of TBs can be synthesized in metals such as Al and Ni.
3.6 Summary
Nanotwinned (nt) metals are materials that consist of several Σ3 twin boundaries (TBs) located
within multiple grains in the overall microstructure, where the distance between any two
consecutive TBs is within the range of 1 nm to 100 nm. Σ3 TBs can be coherent or incoherent
depending on the TB family plane, where CTBs have a higher density of coincidence sites per
unit GB area than ITBs. CTBs have high misorientation angle (60°) and low boundary energy,
making them special types of boundaries. If a significant amount of CTBs are incorporated into
the microstructure, they can generate low-energy segments that reduce the overall energy of the
GB network. Thus in grain boundary engineered materials, CTBs are one of the most important
parameters. For example, CTBs can truncate intergranular corrosion through the GB network,
enhance thermal stability or increase the strength of a material, as has been observed for nt Cu
[4, 27, 45]. Sputtering is a fabrication method that can be used to study the twinning phenomena
at the nanoscale, and the versatility of this method may allow one to overcome SFE restrictions,
vary the twin density, and the control of the twin thickness. The following sections in this
dissertation will discuss the synthesis of nt metals with low, intermediate, and high SFEs, the
control of nanostructural features in nt metals, and the nature of TBs in high SFE metals.

25

4 Sputtering system upgrades and characterization techniques
In the previous section it was shown that, in addition to the SFE of the material, sputtering
parameters play a significant role in the synthesis of nt metals. Therefore, to study the influence
of the sputtering parameters on the resulting nt structures, the sputtering chamber used in this
study was modified. The upgrades allowed for better control of the film deposition temperature,
manipulation of the substrate temperature prior to and during sputtering, further control of the
deposition rate, and in-situ monitoring of the film temperature. Additionally, In this section a
brief description of the relevant characterization techniques used in this dissertation will be
presented.
It should be noted, since the equipment and consumables described in this dissertation were
acquired in the United States, where the standard units of the products are not in the International
System of Units (SI). Therefore, the appropriate SI units are noted in parenthesis.
4.1 Substrate temperature Control
4.1.1 Substrate cooling stage
A substrate cooling system was assembled by modifying a 3 inch (7.62 cm) sputtering gun
(MeiVac). The gun adaptation allows the ability to cool down a substrate holder in the same way
that a sputtering target material is cooled down during sputtering. Figure 12 shows the cooling
stage and a substrate holder, where up to three 1 inch (2.54 cm) substrates can be mounted
simultaneously. The substrate holder faceplate was designed to hold a thermocouple embedded
in a 1 inch Si wafer to monitor the film temperature during deposition. The cooling temperature
was set to 16 °C for the experiments conducted with the cooling stage.
26

Figure 12. Substrate cooling stage and substrate holder
27

4.1.2 Substrate heating stage
A resistivity heating stage (SU-200-IH MeiVac) was used to heat the substrates prior to and
during sputtering to study the influence of high temperatures during the synthesis of the films.
The substrate heater can reach temperatures up to 950 °C. Heating the substrate prior to and
during sputtering should provide better control of the nanostructural features of nt metals. Figure
13 shows the heating stage, where a faceplate was designed to keep the substrate in place and to
hold a thermocouple embedded in a 1 inch Si wafer, and monitor in-situ the film deposition
temperature.
4.1.3 Further control of film deposition temperature via interrupted sputtering
In the production of nt metals, high deposition rates are desired (>5 nm/sec). Therefore, a power
supply (Pinnacle Plus Pulsed-DC Power Supply 5KW) that allows higher deposition rates was
acquired. However, high deposition rates may significantly increase the energy and thus the
temperature of the atoms that arrive to the substrate, possibly leading to significant changes in
the microstructure. Besides the reduction of the film deposition temperature by using a cooling
stage, one way to reduce the temperature due to high deposition rates is by interrupting the
arrival of atoms to the substrate (referred to as interrupted sputtering), which creates a multilayer
microstructure, but within a single material. Although, the existing chamber has an automated
shutter system that physically interrupts the substrate from being coated, it is also desirable to
inhibit target polarization. Thus, an automated system that can turn the target polarization on and
off was implemented and synchronized with the shutter system.

28

Figure 13. Substrate heating stage

29

4.2 Temperature monitoring
To expand the understanding of the role of temperature in the produced microstructures, a
thermocouple (K and T type) embedded in a silicon wafer was used to monitor the film
temperature during deposition. Initially, a 2 inch (5.08 cm) thermocouple was used for
preliminary studies and the subsequent temperature study described in section 5. In this case, the
thermocouple embedded in the Si wafer was placed instead of a substrate and allowed for
monitoring the temperature by replicating the desired sputtering conditions. For the other studies
described in this dissertation, a 1 inch (2.54 cm) wafer thermocouple was used. No difference in
temperature readings or accuracy was observed between the 1 or 2 inch wafer thermocouples.
However, the 1 inch thermocouple occupied less space and matched the size of the sample
substrates. Hence, the 1 inch wafer thermocouple was mounted on the faceplates shown in Figure
12 and Figure 13. Thus, the 1 inch wafer thermocouple can experience the changes in
temperature due to the arrival of atoms in the same way than the 1 inch Si wafer substrates that
are used to growth the films experience temperature changes. In other words, it was possible to
simultaneously grow the films and monitor the film deposition temperature in-situ. Figure 14
shows the 1 inch Si wafer, and the wafer location where the thermocouple is embedded. The
following section provide information on the characterization techniques that were used to study
the films produced with the upgraded sputtering system.
30

Figure 14. 1 inch thermocouple wafer.

4.3 Characterization Techniques
4.3.1 X-ray diffraction
X-rays are radiated electromagnetic waves, which have a short wavelength on the order of
angstroms. When X-rays strike an object they can either be scattered or absorbed. The scattering
of X-rays will be elastic if there is no energy loss in the process and the X-rays wavelength
remains unchanged [68]. In crystalline materials, the diffracted X-rays will have twice the angle
of the incident X-ray. The angle of the diffracted X-rays depend on the Bravais point lattice and
the unit cell dimensions, while the intensity of the diffracted X-rays will depend on the chemical
species (atoms that make up the object) and the arrangement of the atomic structure [68].
31

In this study, X-ray diffraction (XRD) was used to evaluate the structure of the films fabricated
by magnetron sputtering. A Rigaku Ultima IV powder diffractometer was used to collect the
XRD patterns. Figure 15 shows an example of the XRD pattern collected for a Cu film, where
the indexed peaks show that the Cu film has a random texture.
.
Figure 15. XRD pattern of a Cu film. Several peaks are indexed and are typical of a random texture.

4.3.2 Focused ion beam
A focused ion beam (FIB) microscope uses high energy Ga ions to probe a specimen, and works
similarly to an electron microscope. In general, a FIB microscope can be used for two main
purposes: (i) secondary electrons excitation, which consists of the detection of the secondary
electrons that are scattered due to the high energy Ga ions bombardment on the specimen surface
(similar to an scanning electron microscope). When the ion beam is aligned with a prominent
zone axis in the material it is possible to generate crystallographic contrast; (ii) micro-machining,
which consist of ion milling the specimen with an accuracy in the range of hundreds of nm [68].
32

In this study, a JEOL JIB-4500 FIB was used to obtain cross-sectional images of the sputtered
films to observe changes in the crystallographic orientations. Figure 16 shows a cross-sectional
FIB image of a magnetron sputtered Cu film, where it is possible to observe several grains with
different crystallographic orientations. Note that individual grains have different gray scale
shades and a change in contrast is also visible within the grain before and after a TB. The FIB
was also used to extract small pieces from sputtered films followed by a thinning procedure
(thickness of the interest zone is < 100 nm) that allows for transmission electron microscopy
imaging. This procedure is also known as FIB lift-out and a further detailed description can be
found elsewhere [68].

Figure 16. FIB cross-sectional image of a magnetron sputtered Cu film. It is possible to observe several grains that
have different crystallographic orientations, note that individual grains have different gray scale colors, while
before and after a TB there is a change in contrast within the grain A (enclosed by the red shape), and TBs are
indicated by the small arrows. The vertical arrow indicates the growth direction

33

4.3.3 Electron backscattering diffraction
Electron backscattering diffraction (EBSD) is a characterization technique that uses a high
energy electron probe to strike a crystal. Due to the high energy collisions, a fraction of
backscattered electrons can acquire velocities that allow them to escape the solid surface. Even
though the backscattered electrons have been scattered several times within the solid, they retain
energies similar to that of the incident electron beam, which allows determination of their
angular distribution by the diffraction from crystal planes and the formation of Kikuchi
diffraction patterns [68].
In this dissertation, an EDAX hikari detector was used to study the nature of TBs in high SFE
materials. The EBSD scans were performed in two modes: conventional EBSD and transmission
EBSD. Figure 17 shows the sample set up for conventional and transmission EBSD respectively
[69, 70]. Conventional EBSD has a resolution on the order of 30 nm, while transmission EBSD
resolution is on the order of < 10nm, which in principle is more suitable to characterize GBs at
the nanoscale. It is important to highlight that in transmission EBSD the sample has a thickness
that is < 100 nm, and contrary to conventional EBSD, the backscattered electrons that arrive to
the EBSD detector are transmitted through the sample [70].
34

Figure 17. conventional (a) and transmission (b) EBSD modes. Adapted from [69, 70].

4.3.4 Transmission Electron Microscopy
Transmission electron microscopy (TEM) is one of the most popular and powerful techniques
used to explore the structure of materials. Electrons are a type of ionizing radiation, one of the
advantages of using electrons in TEM is the wide range of secondary signals that can be emitted
from the radiated specimen [71]. Figure 18 shows an incident beam of electrons and the
secondary signals that can be generated due to the interaction of the electrons beam with the
specimen [71].
35

Figure 18. Signals that can be generated from the interactions of a high energy electron beam and a thin specimen.

In TEM, a high energy beam of electrons strikes a thin (< 100 nm) specimen similarly to the
process depicted in Figure 18, the electrons that scatter through the specimen (direct beam) are
used to produce different type of images modes such as bright field, dark field, and/or diffraction
patterns [71]. These TEM imaging modes are related to the amount of electrons that are filtered
or selected after they interact with the specimen by using an aperture or a specific electron
detector at a certain angle [71]. The primary takeaway of TEM theory is that by knowing what
causes the scattering of electrons, it is possible to understand and comprehend the image mode
that is being used [71]. Three examples of different TEM imaging modes are given as follows:
(i) a bright field TEM image is acquired by collecting only the scattered electrons from the direct
beam (see Figure 18) [71]; (ii) a diffraction pattern image is collected by the constructive
interference of highly ordered electron waves that are in phase with one another in certain
directions, which then provides a description of the reciprocal relationship between atomic-plane
spacings and scattered angles [71]; and (iii) a dark field image is acquired by collecting scattered
36

electrons from an specific direction [71]. A detailed description of the TEM principles described
in this section and the TEM imaging modes that are used in this dissertation can be found in the
book by Williams and Carter [71].
Different imaging modes acquired with a JEM-2100F JEOL TEM are shown in Figure 19 for a
Cu-10wt.%Ni thin film. Figure 19a shows a bright field image of a typical columnar grain and
TBs crossing the width of the grain, the image contrast comes from only selecting the direct
beam scattered electrons with a small aperture. The inset shown in Figure 19a is the selected area
electron diffraction (SAED) pattern of the area displayed in the bright field image, the SAED
pattern has a double hexagon, which is typical of a twinned structure. The direct beam is the
brightest dot in the center of the SAED pattern, while the other dots are scattered electrons from
specific directions (relationship between atomic-plane spacing). Figure 19b shows a dark field
image of the same columnar grain displayed in Figure 19a, the image was acquired by selecting
the scattered electrons from a specific direction with a small aperture, the scattered electrons
from this specific direction form the diffraction dot enclosed by the blue circle in the SAED
pattern; Figure 19c is a high resolution image of the red square area in Figure 19b, note the TB is
highlighted by the red line, while the yellow lines illustrate the change in the stacking sequence.
37

Figure 19. Cu-10wt.%Ni cross-sectional TEM view. a) bright field TEM and inset selected area electron diffraction
(SAED) pattern; b) dark field image of a) acquired by selecting the diffraction point enclosed by a circle in the
SAED pattern; c) high resolution image of the square region in b), note the yellow lines highlight the change in the
stacking sequence, and the red line is the TB.

38

5 Synthesis and characterization of nt structures in low SFE materials
A variety of sputtering conditions have been identified as being critical in the synthesis by
magnetron sputtering of nt Cu by Hodge et al and Furnish et al [5, 9]. Building on this work and
adjusting the sputtering conditions as a starting point, the influence of SFE, deposition rate, and
temperature on TB formation is presented.
5.1 Influence of the SFE
A description of the influence of the SFE and the sputtering rate was presented in a recently
published paper by Velasco et al [60], where, in order to isolate only the influence of the SFE,
identical sputtering deposition conditions were used to synthesize nt Cu and nt Cu-Al alloys.
Specifically, thick foils (4.2 μm thick) of Cu (99.999%), Cu-4 wt.% Al and Cu-6 wt.% Al
(99.99%) were sputtered onto Si (100) substrates using 2 and 5 mTorr Ar pressures at a power of
65 W. The Cu-Al samples were sputtered using 2" diameter alloy targets with the corresponding
composition. The sputtering rate was approximately 0.88 nm/sec for all samples. The film
thicknesses were measured using a JIB 4500 (JEOL) focused ion beam (FIB) and an XP-2 stylus
profilometer (AMBiOS). Residual stresses were calculated using Stoney's equation by obtaining
the sample curvature before and after deposition [72]. Sample characterization was performed
using a JEM-2100F (JEOL) transmission electron microscope (TEM) and sample preparation
was conducted by FIB lift out. X-ray diffraction (XRD) patterns for the Cu and Cu-Al alloys
sputtered at 2 mTorr were obtained from an Ultima IV diffractometer (Rigaku) using θ-2θ scans.
Table 2 shows a summary of the sputtering conditions, sample characteristics, and corresponding
SFE values for each of the materials used in this study. Residual stresses were small and on the
same order of magnitude for all the films, and thus should not greatly affect the microstructure of
39

the samples. It should be noted that the Cu-6 wt% Al, which has the lowest SFE, presented a
smaller grain size than the other samples, while there is no significant difference in grain size
between the two different working pressures for all the samples. The other results for the Cu-Al
alloys sputtered at 5 mTorr are similar to those observed for samples sputtered at 2 mTorr.
Therefore, only results at 2 mTorr will be shown, which are characteristic for both sputtering
pressures.
Table 2. Sputtering conditions, SFE, Grain Size and Residual Stress for Cu and Cu-Al alloys.
Material Working
Pressure
(mTorr)
Sputtering
rate
(nm/sec)
SFE
a

(mJ/m
2
)
Grain
size
(nm)
Residual
stress
(MPa)
Average Twin
thickness
Experimental
(nm)
Cu-6 wt.% Al
2 0.92 6 36±9 37 2
5 0.85 6 35±9 203
Cu-4 wt.% Al
2 0.86 13 51±11 35 2.3
5 0.83 13 45±13 202
Cu 2 0.92 78, 45 56±14 -28 11.7
a
SFE values from references [32, 33, 36].
In order to verify that our microstructural observations are directly related to the changes in the
SFE only, temperature measurements were taken at all sputtering conditions by using a
thermocouple embedded in a Si wafer (KLA Tencor). Figure 20 shows the temperature profiles
for sputtering at 2 mTorr for the three different materials. The curves are shown over an 80
minute sputtering time span, which is representative of the samples used in this study. It should
be noted that overall, there is a minimal difference between the curves as a function of time.
Therefore, using the results from Table 2 and those of Figure 20, it is concluded that the SFE was
the only tested variable.

40

Figure 20. Temperature profiles collected during sputtering of Cu, Cu-4 wt.% Al, and Cu-6 wt.% Al at 2 mTorr.
Note that the three compositions present similar temperature profiles.

Figure 21 Shows the XRD patterns for the Cu and Cu-Al alloys. The Cu-Al alloys showed a
strong {111} texture in the growth direction, as has been previously observed for highly nt-Cu
[11], while for the current conditions, Cu presented a random texture. Therefore, the difference
in SFE between the Cu and the Cu-Al alloys could be one of the factors that significantly affect
the microstructure.

Figure 21. Normalized XRD patterns for Cu and Cu-Al alloys.

41

Representative TEM images starting from low to high SFE for samples sputtered at 2 mTorr are
shown in Figure 22. The insets in Figure 22b, Figure 22d, and Figure 22f depict representative
selected area electron diffraction (SAED) patterns from the different samples. Figure 22a shows
Cu-6 wt.% Al with highly defined columnar grains and high twin density inside the columns.
Figure 22b shows a zoomed in view noting both SFs and TBs. Figure 22c and Figure 22d depict
Cu-4 wt.% Al with columnar grains and twin densities similar to those found in Figure 22a and
Figure 22b. Overall, the Cu-4 wt.% Al and the Cu-6 wt.% Al samples have similar
microstructures, despite the fact that their SFE differs by a factor of two. The SAED patterns in
Figure 22b and Figure 22d show the typical double-hexagon pattern of the (110) oriented
twinned grain, with streaking which is typical of extremely fine λ and many SFs [12, 55].
In order to highlight the effect of SFE on twinning, Figure 22e shows a Cu sample with columnar
grains and some twins. Although there are many studies showing a highly nt Cu microstructure
[2, 9, 13], in the first part of this study the objective was to perform a direct comparison to the
Cu-Al samples under the same processing conditions, rather than trying to achieve a fully nt Cu
sample. Therefore, in this Cu sample, not all the columnar grains were fully twinned, which is
attributed to its higher SFE compared to the Cu-Al alloys. In Figure 22f, λ can be clearly
observed and noted to be larger than λ found in the Cu-Al alloys. Additionally, the Cu sample
shows only a few SFs, compared to the large amount observed in the Cu-Al samples. This is
reflected in the SAED pattern in Figure 22f, which has the double-hexagon pattern, but not the
streaking which is present in the insets in Figure 22b and Figure 22d.
42

Figure 22. Cross-sectional TEM images showing the grain size and twin structure of Cu and Cu-Al samples
sputtered at 2 mTorr: (a) Cu-6 wt.% Al with an average grain size of ~36 nm, (b) Zoomed in view of (a) showing the
presence of many SFs indicated by the streaking in the inset SAED pattern. (c) Cu-4 wt.% Al with an average grain
size of ~51 nm, (d) Zoomed in view of (c), with a similar microstructure to the one observed in Figure b, the
43

streaking in the inset SAED pattern indicates many stacking faults and thin twins, (e) Cu with an average grain size
of ~56 nm, (f) Zoomed in view of (e) with thick twin spacing; the lack of streaking in the inset SAED pattern
indicates the presence of fewer stacking faults than in b and d. The growth direction is vertical for all the images.

TEM images, such as those shown in Figure 22, were used to determine λ distribution, illustrated
in Figure 23(a-c). The Cu-6 wt.% Al and Cu-4 wt.% Al samples present a sharp λ distribution
with an average λ of 2 nm and 2.3 nm, respectively. In contrast, Cu has a wider λ distribution,
with some thinner twins in the range of 2 nm to 6 nm and a high number of thicker twins larger
than 12 nm; the average λ for Cu was 11.7 nm.

Figure 23. Twin spacing or twin thickness (λ) distributions measured from TEM images of: (a) Cu-6 wt.% Al, (b)
Cu-4wt.% Al, and (c) Cu samples sputtered at 2 mTorr. The average λs were 2, 2.3, and 11.7 nm, respectively.

The experimentally determined values for λ were compared with the theoretical predictions
explained in the Background section of this document. The sputtering conditions described in
Table 2 for the Cu-Al alloys were used to calculate the relative radius difference between the
perfect and twinned nuclei to be less than 4% (Equation 4), where a relative difference of less
than 5% is considered favorable for high twin boundary formation [12]. In order to estimate the
44

theoretical λ, the nucleation rates for perfect and twinned nuclei can be calculated by using
Equation 5 and Equation 6, respectively, where
o
 is a pre-exponential factor and assumed to be
equal for both cases [55].










KT r
o perfect
perfect
e I
) (
*



Equation 5






















 

KT
h r
o twin
twin
twin
e I
 

*

Equation 6
By calculating
twin
perfect
I
I
it is possible to estimate the probability of nucleating a twin with respect
to the amount of perfect nuclei. This probability translates to one twin per some amount of
monolayers of perfectly stacked material. Therefore, λ could be estimated by using Equation 7
and multiplying by h , the {111} interplanar spacing [55].
λ
 
h
G h G KT
h
h
I
I
twin v v
twin
twin
perfect















  
  

 
2
exp
Equation 7

The calculated λs using Equation 7 are ~1.2 nm for the 6 wt% Al sample and ~7 nm for the Cu-4
wt.% Al sample, which are comparable to the experimentally observed λ. These values are
similar to those reported in sputtered nt Ag and SS structures, both of which have a low SFE (<
20 mJ/m
2
) [11, 12, 55]. In contrast, the difference in radius between a perfect and a twinned
nucleus for Cu (45 mJ/m
2
) under the conditions used in this study is ~11%, giving a calculated λ
45

of ~288 m, which is several orders of magnitude higher than the results observed
experimentally. The discrepancy between the theoretical calculations and the experimental
results can be partly attributed to the imprecise values of  and
s
P in Equation 5, Equation 6,
and Equation 7 [55, 73], which lead to exponential changes in the calculated λ values.
Overall, it was possible to isolate the effect of SFE on λ within the Cu and Cu-Al systems. The
experimental results clearly demonstrate the influence of the SFE in the successful production of
nt structures. By comparing the calculated λ from the theoretical predictions and the ones
observed experimentally, it is clear that there are significant differences, which can be attributed
to concepts that are not taken into account by the model. For instance, continuum concepts such
as surface tension and nucleus radius may not be useful for predicting λ at the nanoscale.
5.2 Influence of the sputtering deposition rate
In order to illustrate the influence of the deposition rate on the twin nucleation process and its
effects on λ, a Cu-4 wt.% Al alloy was selected and the deposition rate was varied by increasing
the power density. A detailed description of the sputtering conditions is shown in Table 3. The
samples were sputtered for the same amount of time (80 minutes). However, the increase in the
power density resulted in significant differences in the thicknesses, grain sizes, and residual
stresses of the samples.
The resulting microstructures were studied by XRD and FIB. Figure 24 shows the XRD patterns
for the different sputtering rates. The three samples have a strong {111} texture, which is an
indicator of a highly nt structure. Nevertheless, it is important to highlight that the sample with
the highest sputtering rate presented a small peak from the (200) plane. The peak can be
attributed to the presence of some nanocrystalline grains in the microstructure.
46

Table 3. Sputtering conditions, SFE, Grain Size and Residual Stress for Cu-4 wt.% Al (SFE=13mJ/m
2
).
Sputtering rate
(nm/sec)
Power
density
(W/cm
2
)
Time
(min)
Thickness
(μm)
Grain
size
(nm)
Residual
stress
(MPa)
0.40 1.48 80 1.9 25±1* 6.6
0.86 3.21 80 4.1 51±11 35
4.12 14.8 80 19.7 94±20 144
* Grain size calculated by using FIB images.

Figure 24. Normalized XRD patterns for Cu-4 wt.% Al at different sputtering rates.

Figure 25 shows FIB images of the microstructures resulting from the sputtering conditions in
Table 3. Although the FIB resolution is not adequate to present a quantitative study of the
microstructure, a qualitative interpretation is possible. The three samples have columnar grains
and, due to the strong texture observed in the XRD patterns in Figure 24 and the previous
47

analysis by TEM in Figure 22c, it is assumed that they are fully nanotwinned. Despite the fact
that the sputtering rates between Figure 25a and Figure 25b differ by a factor 2, significant
changes in the microstructures could not be observed.

Figure 25. Cross-sectional FIB images for Cu-4 wt.% Al at different sputtering rates. (a) Sputtering rate 0.4 nm/sec.
(b) Sputtering rate 0.86 nm/sec. c) Sputtering rate 4.12 nm/sec. The arrow indicates the growth direction.

On the other hand, as depicted in Figure 25c, when the sputtering rate was increased by a factor
of ~5, there is a clear change in λ and the microstructure, where wider columnar grains are
present, and it is possible to observe the changes in contrast inside the columns, which are
characteristic of stacking reversals in nt structures. A TEM study was conducted and a
representative TEM image of fully nanotwinned Cu-4 wt.% Al sputtered at 2 mTorr, with a
sputtering rate of 4.1 nm/sec is shown in Figure 26. In this sample, substantially larger twins are
present, as compared to those seen in the other Cu-4 wt.% Al samples sputtered at lower
deposition rates, (Figure 22c and Figure 22d). The inset in Figure 26 shows a SAED pattern with
little streaking, indicating fewer SFs and thicker twins as compared to Figure 22c and Figure
22d. By using equations 1, 2, and 5, the expected λ at this deposition rate is calculated to be ~1
nm. However, the average λ measured using multiple TEM images was ~9.7 nm; with a λ
distribution composed of a few thinner twins (< 2nm) and a large amount of thicker twins on the
48

order of 8-11 nm. These results highlight the interplay between deposition rate and SFE for the
purpose of controlling the twin thickness. It should be noted that increasing the deposition rate
increases the temperature during the sputtering process [52], leading to changes in the grain size
and perhaps influencing the twin nucleation process. It was observed that the grain width of the
sample sputtered at 4.1 nm/sec is twice that of the sample sputtered at 0.86 nm/sec, while λ
increased by a factor of 5. This is in contrast to the expected results from the analytical model,
which predicts a decrease in λ as the deposition rate increases. However, since the SFE for Cu-4
wt.% Al is very low, further reduction of λ by increasing the sputtering rate is not likely to occur;
thus, other factors such as kinetic energy or chemistry may influence the λ and twin formation
during deposition.

Figure 26. Cross-sectional TEM image showing the twin structure of Cu-4 wt.% Al sputtered at a deposition rate of
4.1 nm/sec. The average twin spacing is 9.7 nm; the inset SAED pattern has little streaking, which indicates the
presence of fewer SFs and thicker twins.

49

Overall the results presented in sections 5.1 and 5.2 revealed that λ can be modified by thermal
effects. Therefore, further understanding of the influence of the film temperature (heating or
cooling the substrate), implementation of high deposition rates (>5nm/sec), and the use of
interrupted sputtering need to be explored. Sections 5.3 and 5.4 are preliminary studies that
provide useful information to understand the modification of the nanostructural features of nt
metals
5.3 Influence of temperature - heating and cooling the substrate
Despite the fact that film temperature is an important variable with regard to surface nucleation,
there is little or no discussion about its effects on twin formation during vapor deposition. In the
analytical model description presented in the Background section, Figure 10 and Figure 11 were
plotted by assuming a constant film temperature (300 K) during sputtering. In order to elucidate
the effect of temperature on the sizes of the critical radii and the probability of twin nucleation,
Figure 27 was plotted under the same conditions as Figure 11, but the value for temperature was
raised to 473 K. As a result, both
*
twin
r and
*
perfect
r decrease to approximately half the sizes
observed in Figure 11 for deposition rates lower than 0.1 nm/sec. This is a clear indication that
raising the temperature during sputtering may have a strong influence on the promotion of
twinned structures. For example at 473 K, the deposition rate that is needed to have a small
difference (less than 10%) between
*
twin
r and
*
perfect
r is 10 times less than at 300 K. Therefore,
according to the predictions from the model, high temperatures appear to favor twin nucleation.
50

Figure 27. Critical nucleation radius for faulted nuclei and perfect nuclei as a function of the deposition rate. This
calculation was performed for SS at a temperature of 473 K and 300 K.

A preliminary study has been conducted to determine the influence of the film temperature on
the twin nucleation; the sputtering conditions selected favor a strong {111} texture, which
appears to correspond to a nt structure. Cu, Cu-4 wt.% Al, and Cu-4.5 wt.% Al were sputtered at
the conditions presented in Table 4. Low deposition rates were used when the substrates were
cooled, while a high deposition rate was used for the heated substrate in order to enhance the
effect of heating the substrate.

0.01 0.1 1 10
0
1
2
3
4
5
6
7
8
9
10
11
12
r/r < 10% at 473 K

Critical nucleation radius (nm)
Deposition rate (nm/sec)
r twin 473 K
r perfect 473 K
r twin 300 K
r perfect 300 K
r/r < 10% at 300 K
51

Table 4. Sputtering conditions explored for cooling and heating the substrate.
Type of
sputtering
Material
Sample
name
Power
density
(W/cm
2
)
Deposition
rate
(nm/sec)
Thickness
(μm)
Temperature
(K)
Cooling
Cu Cu#10 22.2 1.55 4.1
289
Cu-4 wt.% Al CuAl#21a 17.27 1.22 6.5
Heating Cu-4.5 wt.% Al CuAl#64 32.89 12.5 7.5 473
No
heating-
No
cooling
Cu-4.5 wt.% Al CuAl#67 32.89 13.5 20 NA

The changes in film texture due to heating and cooling the substrates are shown in the XRD
patterns in Figure 28. Cu#10 and CuAl#21a were cooled during sputtering and presented a
random and a strong {111} texture, respectively. The strong {111} texture in CuAl#21a is
attributed to its low SFE, while in Cu#10, which has a higher SFE it is possible that the use of
higher deposition rates will be necessary to produce strong {111} textures. CuAl#67 was
sputtered at a high deposition rate, without cooling or heating of the substrate and it presented a
strong {111} texture. Despite the fact that CuAl#64 has a low SFE and it was sputtered at a high
deposition rate, it presented a random texture, which can be attributed to the high temperature of
the heated substrate, as well as the amount of energy that was transferred from the high energy
atoms that arrived to the substrate.
This preliminary study provided information with regard to the effect of film temperature on
twinning. It is interesting that increasing the temperature promoted a random texture, which can
52

be indicative of a non-nt structure. The future studies will focus on varying the substrate
temperature from 289 K to 473 K for low and high deposition rates (section 6).

Figure 28. Normalized XRD patterns for the sputtering conditions explored by cooling and heating the substrate

5.4 Influence of the deposition rate in combination with interrupted sputtering
Previously, in section 5.2, it was observed that the increase of the deposition rate (max ~4.1
nm/sec) in Cu-4 wt.% Al increased λ. However, further increase of the deposition rate was not
possible due to the target size limitation (5.08 cm). Therefore, a 7.62 cm Cu 4.5 wt.% Al target
was used to increase the sputtering rate to ~12 nm/sec. It is possible that using a high deposition
rate in combination with interrupted sputtering will lead to further control of λ. Table 5 shows
the conditions selected to evaluate the influence of a high deposition rate and interrupted
sputtering.

53

Table 5. Sputtering conditions at a high sputtering rate and interrupted sputtering for Cu-4.5 wt.% Al.
Sample
name
Power
density
(W/cm
2
)
Interrupted
sputtering
time on / off
(sec)
Duty
cycle
(%)
Deposition
rate
(nm/sec)
Thickness
(μm)
Grain size
(nm)
CuAl#35 32.9 NA 100 12 29 196±12
CuAl#37 32.9 20 / 20 50 12 29 213±20
CuAl#38 32.9 20 / 40 33 11.7 28 184±9
CuAl#39 32.9 20 / 60 25 11.2 27 190±7
CuAl#40 32.9 20 / 200 9 11.2 27 74±2

XRD and FIB were used to study the effect of a high deposition rate and interrupted sputtering
for the samples listed in Table 5. Figure 29 shows the XRD patterns, where all the samples
presented a strong {111} texture, which means that the change in duty cycle did not promote
changes in the film textures.

Figure 29. Normalized XRD patterns for Cu-4.5 wt.% Al samples fabricated at high sputtering rate and interrupted
sputtering
54

The influence of the duty cycle on the microstructure of the samples listed in Table 5 can be
observed in the cross sectional FIB images shown in Figure 30. All the samples presented a nt
structure, and columnar grains. Figure 30f shows a zoomed in view of Figure 30d, where it is
possible to observe the difference in contrast within the columnar grains that represent reversals
in stacking and it is characteristic of twins. The FIB images in Figure 30 were used to estimate
the grain size of the samples. In general, the samples with a duty cycle higher than 25% did not
present significant changes in the grain size, ~190nm, while a duty cycle of 9% decreased the
grain size by a factor of 2.6. Further analysis by using TEM is needed in order to accurately
measure λ and the grain sizes.

55

Figure 30. Cross-sectional FIB images showing fully nt structures with columnar grains in Cu-4.5 wt.% Al at a high
sputtering rate and with interrupted sputtering. The arrow indicates the growth direction for all the samples. (a)
CuAl#35, duty cycle 100%, grain size 196nm, (b) CuAl#37, duty cycle 50%, grain size 213nm, (c) CuAl#38, duty
cycle 33%, grain size 184nm, (d) CuAl#39, duty cycle 25%, grain size 190nm, and (e) CuAl#40, duty cycle 9%,
grain size 74nm, (f) zoom in view of CuAl#39 (yellow rectangle), difference in contrast within the columnar grain
that represents reversals in the stacking sequence.

5.5 Conclusions
It was possible to isolate the effect of SFE on λ within the Cu-Al system. Additionally, by
increasing the deposition rate, a fully nanotwinned structure with larger grain widths and thicker
twins was obtained as compared to samples at low sputtering rates having the same chemical
composition. Therefore, the twin nucleation mechanism shows a dependence on the deposition
rate, which can be correlated to temperature and should be accounted for in future models. It was
also shown that the sputtering techniques can be used for tuning the twin microstructures and
that by decreasing the SFE, one could further tailor λ. The studies in sections 5.3 and 5.4
provided useful sputtering parameter conditions for tailoring λ. The use of high sputtering rates,
combined with interrupted sputtering, and cooling or heating the substrates may allow for
production of a wide range of λs. Additionally, it is possible to extrapolate the developed
synthesis method for other Cu alloys, such as Cu-Ni, Cu-Ag, and/or Cu-Zn, which have a wider
range of SFEs. In section 6, it will be shown how the studies in sections 5.3 and 5.4 allow to
tailor λ.

56

6 Tailoring the twin thickness in low and intermediate SFE materials: The
mobility of growth twins synthesized by sputtering
The current section presents a protean twin thickness contour zone map that illustrates how the
nucleation and the mobility of twin boundaries affects the twin thickness of sputtered films. The
findings described in this section has been recently published in Acta Materialia [25]. The twin
thickness contour zone map can be used as a versatile guide to synthesize fully nanotwinned
films with tailored twin thicknesses in materials with a wide range of stacking fault energies. The
nucleation and mobility of twin boundaries was studied in four Cu alloys of different
compositions (Cu-6wt.%Al, Cu-4wt.%Al, Cu-2wt.%Al, and Cu-10wt%Ni), having stacking fault
energies ranging from 6 mJ/m
2
to 60 mJ/m
2
. The films were synthesized by magnetron sputtering
and characterized by transmission electron microscopy, where the twin thickness varied from 2
nm to 35 nm. The experimental results show that it is possible to control the twin thickness.
Three main mechanisms are explained to describe twin nucleation and twin boundary mobility,
which are correlated to the interplay of specific sputtering conditions and the deposition
temperature.
Four Cu alloys (99.99% purity) were magnetron sputtered (Cu-6wt.% Al, Cu-4wt.% Al, Cu-
2wt.% Al, and Cu-10wt.% Ni) onto 1" (2.54 cm) Si (100) substrates; the Cu alloys targets were
3" (7.62 cm) in diameter. The chamber was evacuated prior to deposition to a base pressure < 1.2
x 10
-6
Torr (1 Torr = 133.3 Pa), while during sputtering, the Ar working pressure was 2 mTorr
and the target-substrate distance 3". In Table 6 the sample name, average deposition temperature,
average deposition rate, sputtering power, substrate temperature, and duty cycle for each of the
57

sputtered films are listed; a value of 100% in the duty cycle represents no sputtering interruption
(refer as continuous sputtering).
For simplicity, the samples were labeled by taking into account the alloy element and its content,
and sequentially label from small to large λ within the same Cu-alloy. The average deposition
temperature listed was varied by three different methods: (i) changing the power to increase or
decrease the deposition rate, (ii) interrupting the flux of sputtered atoms that arrive to the
substrate (refer as interrupted sputtering), and (iii) cooling or heating the substrate during
sputtering. In the last case, a cooling stage was custom made to cool down the substrates to 16
°C, while a resistivity heating stage (SU-200-IH MeiVac) was used to heat the substrates.
Temperature measurements were taken for the different sputtering conditions by means of a K or
T type thermocouple embedded in a Si wafer (Thermo Electric) that it is placed as a substrate to
monitor the deposition temperature.
The films thicknesses were >15 μm, as measured on a JIB 4500 (JEOL) focused ion beam (FIB)
and an XP-2 stylus profilometer (AMBiOS). Transmission electron microscope (TEM) sample
preparation was conducted by typical film cross-sectioning, bright field, dark field and high
resolution TEM images were obtained on a JEM-2100F (JEOL) to characterize the
microstructure. λ was calculated by measuring the distance between consecutive twin
boundaries. The statistical weighted average for λ and grain width was obtained by the
measurement of more than 1100 twin boundaries and more than 200 grains for each sample.

58

Table 6. Summary of sputtering conditions used to synthesize highly nt Cu alloys.
Average deposition rate = deposition rate × duty cycle / 100.
N/A no substrate heating or cooling.

6.1 Microstructural characterization
In Table 7, the SFE for each of the alloys, the average twin thickness, the average grain width,
the probability of twinning (ρ), the total grain boundary energy (GBE) stored in a film, and the
ratio between the total GBE and twin boundary energy (TBE) stored in the film are displayed.
Sample
label
Material
Average
deposition
temperature
(°C)
Duty
cycle
(%)
Average
deposition
rate
(nm/sec)
Substrate
temperature
(°C)
Sputtering
power
(W)
6Al#1
Cu-6wt.%Al
80 100 0.92 N/A 65
6Al#2 50 9 1.26 16 1500
6Al#3 350 100 15.00 N/A 1500
6Al#4 100 9 1.25 N/A 1500
6Al#5 28 9 0.06 N/A 65
6Al#6 290 100 12.93 16 1500
6Al#7 140 9 1.31 100 1500
4Al#1
Cu-4wt.%Al
80 100 0.86 N/A 65
4Al#2 50 9 1.20 16 1500
4Al#3 290 100 11.79 16 1500
4Al#4 350 100 14.49 N/A 1500
4Al#5 100 9 1.21 N/A 1500
2Al#1
Cu-2wt.%Al
36 9 0.74 16 1000
2Al#2 28 9 0.06 N/A 65
2Al#3 50 9 1.10 16 1500
2Al#4 100 9 1.06 N/A 1500
2Al#5 140 9 1.10 100 1500
2Al#6 290 100 12.08 16 1500
2Al#7 350 100 13.49 N/A 1500
10Ni#1
Cu-10wt.%Ni
145 9 1.11 100 1500
10Ni#2 175 9 1.09 150 1500
10Ni#3 190 16.6 2.16 150 1500
10Ni#4 225 9 1.14 200 1500
59

The Cu alloys are listed from low to high SFE, 6 mJ/m
2
for the Cu-6wt.% Al [32], 13 mJ/m
2
for
the Cu-4wt.% Al [32], and 37 mJ/m
2
for the Cu-2wt.% Al [35]. Although, no SFE values were
found in the literature for the Cu-10wt.% Ni alloy, a study conducted by Li et al [34] showed that
the SFE of Cu-Ni alloys is higher than that of pure Cu, Additionally, the SFE of Cu-Ni alloys
increases as the Ni content is increased, for example, for Cu-20wt.% Ni the SFE is 74 mJ/m
2

[34]. Therefore, one can assume that the SFE of Cu-10wt.% Ni alloy is in the range of 47 - 74
mJ/m
2
. For comparison reasons we will use a value of 60 mJ/m
2
throughout this dissertation.
Table 7. Average microstructural properties and total stored energy of highly nt Cu alloys samples.
SFE values from references [32, 34, 35]

Sample
SFE
(mJ/m
2
)
Weighted
average λ
(nm)
Weighted
average
grain width
(nm)
ρ
(%)
Total GBE
(mJ)
Total
GBE/TBE
(J/J)
6Al#1
6
2.1±0.3 40±5 1.1 80 17.7
6Al#2 4.8±0.6 130±20 0.7 93 5.4
6Al#3 6.5±0.8 130±20 0.3 93 5.4
6Al#4 7.4±0.9 130±20 0.6 93 5.4
6Al#5 8±1 110±10 1.5 110 6.4
6Al#6 13±1 190±30 0.4 63 3.7
6Al#7 19±3 300±40 0.5 40 2.3
4Al#1
13
2.5±0.3 60±7 2.3 53 6.7
4Al#2 5.1±0.6 140±20 1.4 86 2.8
4Al#3 5.1±0.6 130±20 0.8 93 3.1
4Al#4 5.9±0.7 100±10 0.7 120 4
4Al#5 15±2 220±30 1.2 54 1.8
2Al#1
37
5.7±0.7 160±20 4.6 75 2.4
2Al#2 6±0.7 130±20 8.6 93 2.9
2Al#3 8±1 160±20 4.1 75 2.4
2Al#4 10±1 260±30 3.5 46 1.4
2Al#5 10±1 210±30 3.1 57 1.8
2Al#6 27±3 200±30 2.3 60 1.9
2Al#7 35±4 230±30 2 52 1.6
10Ni#1
60
15±2 220±30 5.0 54 3.2
10Ni#2 17±2 330±40 4.7 36 2.1
10Ni#3 20±2 260±30 4.4 46 2.7
10Ni#4 31±4 380±50 4.1 31 1.8
60

As shown in Table 7, λ is arranged from low to high for each of the Cu alloys, where λ and grain
width for each sample were calculated by taking several TEM images and measuring at least 500
TBs and 200 grains in different locations. All samples present highly defined columnar grains
and high twin density; where more than 95% of the grains observed are twinned and the TBs are
parallel to the film surface. The parameters showed in Table 7, and the sputtering parameters
displayed in Table 6 will be discuss further in this section together with the methodology applied
to vary λ.
Representative TEM images for the Cu-2wt.% Al alloys presented in Table 7 and their
corresponding λ volume fractions are shown in Figure 31, the insets depict representative
selected area electron diffraction (SAED) patterns, which show the typical double hexagon
pattern of the (110) oriented twinned grain. In the TEM Figure 31a thru Figure 31e, λ variation is
observable for the different sputtering conditions, while λ volume fraction plots show the
increase in λ from ~5.7 nm to ~35 nm. In Figure 32, representative TEM images and λ volume
fractions are shown for the Cu-4wt% Al, Cu-6wt% Al, and Cu-10wt.% Ni. The inset SAED
patterns are typical for twinned grains, and the TEM images highlight λ variation. In Figure 32 it
is evident that λ and λ volume fraction were increased for each of the alloys at different
sputtering conditions; for example, in Figure 32a and Figure 32b the average λ varied from 2 nm
to 19nm for the Cu-6wt.% Al; in a similar way for the Cu-4wt% Al films (Figure 32c and Figure
32d), λ varied from 2.4 nm to 15 nm; and for the Cu-10wt.% Ni (Figure 32e and Figure 32f) λ
varied from 15 nm to 31nm.
61

62

Figure 31. Representative cross-sectional TEM images of the twinned structures within a Cu-2wt.%Al alloy
obtained at different sputtering conditions; the inset SAED patterns show the typical double hexagon of the (110)
oriented twinned grain. (a) sample 2Al#5, noting the overall columnar grain structure. The twin thickness
distributions highlight the volume fraction of twins for each sample in images (b) to (e); , (b) sample 2Al#1, λ≈5.7
nm; (c) sample 2Al#4, λ≈10 nm; (d) sample 2Al#6, λ≈27 nm; and (e) sample 2Al#7, λ≈35 nm. Representative TEM
images showing the λ achieved for samples 2Al#2, 2Al#3, and 2Al#5 are presented in Appendix E). The growth
direction is vertical for all the images.

Figure 32. Representative cross-sectional TEM images of the twinned structures for Cu-Al and Cu-Ni alloys
obtained at different sputtering conditions. The inset SAED patterns show the typical double hexagon of the (110)
oriented twinned grain. The twin thickness distributions highlight the volume fraction of twins for each sample. For
Cu-4wt.%Al (a) sample 4Al#2, λ ~5.1 nm, and (b) sample 4Al#5, λ≈15 nm; For Cu-6wt.%Al (c) sample 6Al#2, λ≈
63

4.8 nm, and (d) sample 6Al#7, λ≈19 nm; For Cu-10wt.%Ni (e) sample 10Ni#1, λ≈~15 nm, and (f) sample 10Ni#4,
λ≈31 nm. Representative TEM images showing the λ achieved for samples 4Al#1, 4Al#3, 4Al#4, 6Al#1, 6Al#3,
6Al#4, 6Al#5, 6Al#6, 10Ni#2 and 10Ni#3 are presented in Appendix E. The growth direction is vertical for all the
images. Sample details are presented in Table 2.

6.2 Effect of the average deposition temperature
The average deposition temperature displayed in Table 6 is calculated by collecting temperature
profiles at the sputtering conditions for each sample during a time spam of 60 min. In Figure 33
representative temperature profiles are displayed, where only one sputtering condition was vary
to change the average deposition temperature; Figure 33a compares continuous sputtering with
interrupting sputtering; Figure 33b compares the influence of substrate temperature; Figure 33c
compares two different sputtering powers; and Figure 33d compares two different duty cycles.
By a combination of interrupted sputtering, cooling or heating the substrate, varying the
sputtering power, and varying the duty cycle it was possible to sputter films at different average
deposition temperatures, which allowed to produce significant changes in λ. Note that in the
present study the effect of residual stresses were not take into account, which were relatively
small and of the same order of magnitude for all samples.
64

Figure 33. Representative temperature profiles collected at different sputtering conditions for Cu-Al and Cu-Ni
alloys during a time spam of 60 minutes. The temperature profiles show the change in average deposition
temperature by the variation of only one sputtering parameter as follows: (a) comparison between interrupted
sputtering and continuous sputtering; (b) comparison between cooling, heating, and as-sputtered; (c) comparison
between two different sputtering powers; and (d) comparison between two different duty cycles.

The methodology used in this study to tailor λ is strongly dependent on the average deposition
temperature. First, we will discuss the three main mechanisms that can affect λ when thermal
energy is added to the system: twin nucleation; TB mobility; and the effect of free surfaces on
65

TB mobility. Second, λ zone plots will be discussed for different SFEs materials at different
sputtering temperatures, followed by a pictorial representation of λ (contour λ zone map), where
different SFEs and sputtering conditions are summarized in terms of the average deposition
temperature (see section 6.3).
6.2.1 Twin nucleation
The synthesis of fully nt films requires sputtering conditions that facilitates a high rate of twin
nucleation. In the case of sputtered growth twins an analytical model developed by Zhang et al
[55] describes from a thermodynamic point of view the probability of twin nucleation (ρ). The
model examines the relative difference between the critical radius of a perfect nucleus
*
perfect
r and
the critical radius of a nucleus with a stacking fault defect
*
twin
r , as shown in section 3.5.1. A
relative difference of less than 5% is considered favorable for high rate of twin nucleation [12].
The probability of twinning (ρ) was calculated by using Equation 4. Note that the average film
deposition temperature displayed in Table 6 was used to perform the calculations with Equation
4, the values of ρ are displayed in Table 7. In general, ρ was < 5%, except for sample 2Al#2.
Low ρ values imply that a large number of TBs are nucleated, and λ should decrease as ρ
decreases.
6.2.2 Twin boundary mobility
To identify whether nucleated TBs can move and alter λ while the films are being sputtered, two
competing mechanisms that can modify the resulting microstructure were examined, (i) grain
growth via grain boundary (GB) mobility; and (ii) λ coarsening via TB mobility. It is possible to
66

identify which mechanism is favored from a thermodynamics point of view, that is by comparing
the total energy stored in the film due to GBs and due to TBs [26].
The total GBE stored in the films can be calculated by making two assumptions: first, the
columnar grains are cylindrical; and second, the boundaries between columnar grains are high
angle grain boundaries (HAGB). For comparison, and due to the lack of HAGB energy values
for the different Cu-alloys used, we use the HAGB energy for nanocrystalline Cu (710 mJ/m2)
[74]. The calculations used: a film surface area of 5.67 x 10
-4
m
2
, a film thickness 15 μm, and the
average grain width of the samples from Table 7. The total energy stored due to grain boundaries
for each of the films is displayed in Table 7 and it is the product of the HAGB energy, the total
number of columnar grains in the film, and the lateral surface area of the cylinder. In a similar
way, the total TBE stored in the films is the product of the TBE, the total number of TBs in the
film, and the surface area that encompasses each TB. To calculate the total number of TBs in the
film, the smallest λ for each of the Cu-alloys was used. The ratio between the total GBE and the
total TBE stored in the films (GBE/TBE) is shown in Table 7. Overall, the total GBE stored is
higher than the total TBE stored, therefore, grain growth is more favorable than λ coarsening. In
other words, from a thermodynamic point of view the TBs should not move. However, when nt
films have been heat treated, it was commonly observed that grain growth and λ coarsening
occurred at the same time [26-28, 75-77].
During grain growth in annealing twins it has been observed that λ coarsening occurs by the
formation and migration of incoherent twin boundaries (ITBs) [78]. In regards to sputtered nt
films, it has been reported that the migration of ITBs promotes the lateral propagation and/or
elimination of CTBs and increase in λ [76, 77]. The ITB migrates through the glide of Shockley
partial dislocations that comprise the ITB [28, 76, 77], where the partials can move in the planes
67

that are perpendicular to the {111} growth direction [76, 77]. Thus, CTBs can laterally displace
across the grain and λ will increase as thin twins will coarsen faster than thick twins [76]. By
combining both the observations for λ coarsening in annealing twins and the migration of ITBs
in sputtered nt films, the microstructure evolution during sputtering can be describe as follows:
Sputtering is a non-equilibrium process, where columnar grains grow during film deposition, and
where the addition of thermal energy increases columnar GB migration [79]. In Figure 34 a
mechanism that displays schematically λ coarsening as the film is being deposited is explained.
Figure 34a shows two columnar grains labeled A and B. In grain A there are four CTBs, a wide
blue line that indicates a boundary of relatively high interfacial energy, and an arrow showing
that grain A will grow at the expense of grain B. As grain A grows, it is possible for a CTB to
detach from the high energy boundary and form an incoherent twin boundary (ITB) (see Figure
34b); this will occur if the free energy of the high energy boundary at the end of the CTB is
sufficiently higher than that of the adjacent boundary [78]; thus, a reduction in the overall free
energy will take place if:
A A A A B A B A ITB ITB AB B A
S S S S
' ' ' ' '
       [78],
where S and  represent the interfacial area and the interfacial energy per unit area,
respectively, of the boundaries indicated by the subscripts [78]. Further increase in the width of
grain A may produce more ITBs, and cause the CTBs to fully detach from the high energy
boundary, as shown in Figure 34c. After the CTBs are completely detached from the high energy
boundary, it is feasible that the addition of thermal energy triggers the migration of the ITB as
noted by the green arrow, then the ITB can migrate and sink in a columnar GB, which leads to
the disappearance of two CTBs [77] (shown in Figure 34d). Therefore, during the film
68

deposition, the formation of ITBs and the lateral propagation of CTBs will increase the average
λ.

Figure 34. Schematic representation of λ coarsening as a function of columnar grain growth during film deposition.
(a) Grain A will grow at expense of grain B; (b) and (c) CTBs are detached from a high energy boundary and form
an ITB; and (d) two CTBs disappeared by the migration of the ITB depicted in Figure 4c

6.2.3 Effect of free surfaces on twin boundary mobility
In this section a mechanism that can affect the mobility of TBs is explained, specifically, when
the TBs are near to the free surface or near to an interface where their mobility can be affected
by the image force. The partial dislocations that comprise ITBs can be drawn to the free surface
because the material is more compliant near to the surface and the dislocations do not experience
forces associated with the bulk of the crystal [80]. Thus, the image force can draw partial
dislocations to the free surface, contribute to ITB migration and increase the mobility of TBs.
The image force can be calculated by
x
A
F
x
  , where A is a constant that relates the shear
modulus, the burgers vector, and the dislocation type, while x is the distance of the dislocation to
69

the surface. Further description of the image force can be find elsewhere [36, 80]. From the
image force expression, one can identify that the image force decreases as dislocations are
deeper inside the film. Therefore the image force will potentially interact with partial
dislocations which are near (a couple hundred nm) to the free surface. In the context of this
dissertation, the image force contributes to an increase in λ by increasing the migration of the
ITBs. This process is enhanced as transient free surfaces are allowed to relax during interrupted
sputtering.

6.3 Protean representation of sputtering conditions: tailoring λ
The interplay of the mechanisms explained in section 6.2 allowed for varying λ within a wide
range (2 - 35 nm). In Figure 35, the sputtering conditions in terms of the average deposition
temperature and corresponding λ (Figure 35a and Figure 35b) are summarized; followed by an
empirical determined λ contour zone map that can be used to tailor λ as a function of temperature
and the SFE (Figure 35c). For visualization purposes, the average λ of the Cu alloys with a SFE
< 13 mJ/m2 is shown in Figure 35a (Cu-6wt.% Al and Cu-4wt.% Al), while the average λ of the
Cu alloys with a SFE > 37 mJ/m2 is displayed in Figure 35b (Cu-2wt.% Al and Cu-10wt.% Ni).
Three zones are depicted in Figure 35a that correspond to the average deposition temperature,
zone I < 90 °C, zone II between 90 and 240 °C, and zone III > 240 °C. Two zones are depicted
in Figure 35b, zone II < 240°C and zone III > 240 °C. The mechanisms described in section 6.2
allow for an explanation of the different λ trends observed in Figure 35a and Figure 35b as
follows:
70

Figure 35. Average λ as a function of temperature and SFE. (a) λ as a function of temperature for the Cu-alloys with
a SFE < 13mJ/m2. (b) λ as a function of temperature for the Cu-alloys with a SFE > 13mJ/m2. c) Protean λ contour
zone map showing the change in twin thickness as a function of the SFE and average deposition temperature; the
three zones depicted in Figures 5a and 5b are delineated by the dashed lines in Figure 5c.

50 100 150 200 250 300 350
10
20
30
40
50
60
c)
Temperature (°C)
SFE (mJ/m
2
)
2.000
6.175
10.35
14.53
18.70
22.88
27.05
31.23
36.00
Twin thickness
Zone III
Zone I
Zone II
71

In zone I, increasing the average deposition temperature allowed λ to decrease monotonically.
Although it is expected that λ will decrease as ρ decreases; the ρ values calculated for zone I
showed a different trend than the one expected. In this case, the use of interrupted sputtering may
allow the image force to trigger ITB migration and increase λ regardless of specific ρ values. At
the lowest temperature, low TB nucleation is expected (Equation 4) and in combination with
interrupted sputtering allows λ to increase.
In zone II, further increase of the average deposition temperature and the use of interrupted
sputtering allowed to increase and control λ as follows: first, increasing the temperature increases
the nucleation of TBs, which favors the growth of a fully nt film; second, during film deposition,
as thermal energy is added to the system, the CTBs can detach from columnar GBs and form
ITBs that will migrate, and increase λ; and third, after each layer is deposited (~200 nm), the ITB
migration is enhanced due to the effect of the image force as transient free surfaces are allowed
to relax. Based on the interplay of the mechanisms proposed to vary λ, and the effect of
temperature in ITBs in zone II [77], the density of ITBs after a sample has been synthesized is
expected to be low. Appendix E shows representative cross-sectional TEM images, where ITBs
are observed in only a few grains for samples in zone II, which is in agreement with the proposed
mechanisms explained in sections 6.2.2 and 6.2.3.
In zone III, when the average deposition temperature is increased, λ showed two different trends:
(i) for the Cu-6wt.% Al and Cu-4wt.% Al, λ remained constant or decreased, this λ trend is
explained as follows: first, the alloys in zone III have the lowest calculated ρ values, which
means that it is expected that a high amount of TBs are nucleated; second, only continuous
sputtering was used in zone III, and there is virtually no free surface until the film reaches its
final thickness. Hence, the effect of the image force on the migration of ITBs will be minimal
72

and limited to few hundred nms below the film free surface. (ii) for the Cu-2wt.% Al alloy, λ
increased which could be to attributed to the influence of the Al content in the alloy. In this case,
the effect of the Zener drag force [81] in the Cu-2wt.% Al is lower than in Cu-4wt.% Al and Cu-
6wt.% Al alloys, and it decreases as the temperature is increased. Therefore, boundary mobility
at elevated temperatures can be higher in the Cu-2wt.% Al alloy, promote grain growth, and λ
coarsening. In general, ITBs were not observed for the samples in zone III, we attribute the lack
of ITBs to the effect of high synthesis temperature on ITB migration in zone III [77]. However,
the overall λ trend in zone III is intriguing and needs further investigation.
It is also important to mention that when the average deposition temperature was increased
beyond zone III, fully nt structures were not obtained for any of the four Cu-alloys. Instead, the
columnar structure was lost, the few twins which formed were not parallel to the film surface,
and the texture was random.
In Figure 35c, the data presented in Figure 35a and Figure 35b are depicted in a protean λ
contour zone map, generated with a Kriging correlation matrix showing λ in correlation with the
average deposition temperature and the SFE of the Cu-alloys. The Kriging correlation is an
optimal prediction method that uses known nearby observations; a detailed description can be
found elsewhere [82]. The zones depicted in Figure 35a and Figure 35b, are delineated by the
dashed lines in Figure 35c. Overall, the λ contour zone map presented in this study can be use as
a versatile guide to synthesize fully nt films with different λ. In the case of low SFE materials it
is possible to achieve low λ values at low or high temperatures (zone I and III), while in zone II
it is possible to obtain higher λ values at intermediate temperatures. As the SFE increases zone II
encompasses the ideal sputtering conditions to vary λ within a wide range (from blue to red).
73

By using interrupted sputtering in combination with low cooling temperatures, it is possible to
increase the zone I area and a larger λ variation. Although, this approach may only work for low
SFE materials. The contour lines highlight that zone III is isolated with respect to the other two
zones, probably due to the effect to the Zener drag force in the Cu alloys, which decreases at
elevated temperatures. Additionally, it is possible to fuse zone I and II. The transition between
the two zones can be marked by an inflection point, where the amount of thermal energy added
to the film significantly increases TB mobility, and in the case of the low SFE materials the
inflection point can be related to the minimum experimentally achieved λ (~2 nm). The λ contour
zone map can be adapted to operate at different sputtering conditions, by using zone I and II as
starting points to increase or decrease λ as desired. For instance, by keeping the same average
film temperature and varying the duty cycle, it is possible to generate variations of the λ contour
zone map presented in this study.

6.4 Conclusions
The experimental results presented in this section show that it is possible to synthesize fully nt
films and vary λ in materials with different SFEs (low/intermediate SFEs). The analytical model
can described the early stages of nucleated TBs and can be used to set proper sputtering
conditions to obtain a high nucleation of TBs (low ρ values). Two mechanisms were used to
explained the mobility of nucleated TBs as the films are synthesized and the mechanisms where
correlated to specific sputtering conditions. The resulting film microstructures were analyzed in a
comprehensive study that allows to present a protean λ contour zone map. The versatility of the λ
contour zone map provides guidance to tailor λ and control the amount of twin boundaries in the
74

film, thus expanding the synthesis space for nanostructure materials. Section 7 discusses the
formation of TBs in high SFE materials using a similar methodology to the one described in
sections 5 and 6.

75

7 Synthesis and characterization of twin boundaries in high SFE materials
The formation of TBs in crystalline solids decays exponentially as the SFE of the material
increases [20]. It has been shown that metals with low SFE are more prone to form TBs. For
instance, in the case of Cu or Cu-alloys (SFE < 45 mJ/m
2
) fully nt films have been synthesized
by a variety of methods, and the inclusion of TBs have shown a direct impact on the properties
of the material, where the mechanical, thermal, and chemical properties are enhanced [1, 4, 20,
22, 23, 27, 83, 84]. These attractive properties increase the working space of nt metals and
contribute to the understanding of materials behavior at the nanoscale. On the other hand, metals
such as Ni, and Al have a high SFE (125 mJ/m
2
and 166 mJ/m
2
, respectively [38, 39] ), and the
formation of TBs in these metals is rare. Al and Ni are widely used and their based alloys have
an extensive range of applications [85, 86]. If TBs are successfully introduced into high SFE
materials, it is probable that the properties of these metals, such as Al and Ni can be enhanced.
During the last decade many efforts have been made concerning the synthesis of TBs in Al and
Al alloys, either at the macroscale and at the nanoscale [62, 63, 87-89]. At the nanoscale only the
study by Xue et al have shown the occurrence of few TBs in a low fraction of grains (<9%) in
thin (<80 nm) Al films, and the study concluded that the formation of growth twins in thicker
films is energetically unfavorable [64]. Regarding Ni, the synthesis of TBs has been mainly
achieved by annealing, and to a lesser extent by electrodeposition and sputtering [65-67]. Among
these studies, only the electrodeposited Ni presented TBs at the nanoscale (λ < 100 nm) [67].
The present study was conducted with two main objectives: (i) synthesize a high density of TBs
in high SFE metals by using magnetron sputtering, specifically, scaling the formation of TBs in
thick films (>10 μm); (ii) utilize complementary characterization techniques to investigate the
76

nature of TBs in high SFE materials, as well as the formation of TBs during continuous grain
growth in magnetron sputtering. The results provide experimental evidence of high frequency of
TB formation in high SFE metals, where the TB plane is inclined with respect to the grain
growth direction. The results also include a comprehensive characterization of the observed
inclined TBs, where as a consequence of the TB occurrence, the grains textures are modified.
The experimental observations allowed discussion on the formation of inclined TBs in high SFE
metals and provide an explanation regarding the formation of TBs during continuous grain
growth in magnetron sputtering.
Ni (99.995% purity), Al (99.99% purity), and Al-5.3wt.%Mg (99.99% purity) targets (7.62 cm in
diameter) were magnetron sputtered onto 2.54 cm Si (100) substrates. For simplicity the Al-
5.3wt.%Mg alloy film will be referred to as Al-Mg throughout this chapter. The sputtering
chamber was evacuated prior to deposition to a base pressure < 1.2 x 10
-6
Torr (1 Torr = 133.3
Pa), during sputtering, the Ar working pressure was 2 mTorr, the target-substrate distance 7.62
cm, and the deposition rate ~7 nm/sec. The thickness of the synthesized films was measured
with an XP-2 profilometer (AMBios). The microstructural characterization of the synthesized
thick films was conducted by transmission electron microscopy (TEM) using a JEOL JEM-
2100F operated at 200Kv. The texture of the films was then analyzed by taking X-ray diffraction
(XRD) patterns using a Rigaku Ultima IV, and by transmission electron backscattering
diffraction (T-EBSD) using a Hikari detector (EDAX). The specimens for TEM were prepared
by mounting a cross section of the film in silicon, dimple grinding, and ion milling using a
Fischione Model 1050 TEM Mill; and by performing focused ion beam (FIB) lift-out using a
JIB-4500 FIB (JEOL). Characterization by T-EBSD was performed only in the FIB lift-out
77

specimens, and the collated data was analyzed with the orientation image microscopy (OIM)
software.

7.1 Microstructural characterization
Representative cross-sectional TEM images of the Al, Al-Mg, and Ni thick films microstructures
are displayed in Figure 36a, Figure 36b, and Figure 36c respectively. The three microstructures
have columnar grains, and several TBs are highlighted by the dotted red lines. In general, the
TBs are not perpendicular to the film growth direction, which is different from what is
commonly observed in the synthesis of TBs in low SFE materials by magnetron sputtering,
where the TBs are usually perpendicular to the film growth direction [9, 25, 56, 60]. In Figure
36, α is the angle between the inclined TB and a plane that is perpendicular to the film growth
direction. In Figure 36a and Figure 36b the TBs are inclined preferentially at an angle α ~70°,
while in Figure 36c the TBs are inclined at different angles and α varies from 0° to ~75°.

Figure 36.Representative cross-sectional TEM images of columnar grains with TBs of: a) Al, b) Al-Mg, and c) Ni.
TBs are marked by the red dotted lines. Al and Al-Mg have a preferential inclined TB angle α ~70°, in Ni the
78

inclination of the TB angle α varies from 0° to ~75°. The blue arrows point out inclined TBs that have the same α
angle in each image.

Low magnification and high resolution TEM images were taken to further study the inclined TBs
observed in Figure 36. In Figure 37a, Figure 37c and Figure 37e, low magnification bright field
images of columnar grains with inclined TBs (marked by red dotted lines) are presented for Al,
Al-Mg, and Ni, respectively. The insets depict the selected area electron diffraction (SAED)
patterns obtained from the columnar grains, the typical double hexagon pattern of the (110) zone
axis oriented twinned grain is highlighted by the dotted blue and magenta lines, each SAED
pattern was also indexed with the corresponding plane orientation. Figure 37b, Figure 37d, and
Figure 37f show zoomed in high resolution TEM images of the square regions in Figure 37a,
Figure 37c and Figure 37e. The TBs are marked by the red dotted lines and lie in a (1 1 -1) or (-1
-1 1) plane, while the blue and magenta dotted lines show the (1 1 1) or (-1 -1 -1) planes at each
of the two sides of the TBs. The yellow lines mark the (1 1 -1) planes that are parallel to the (1 1
-1) TB plane. The insets in Figure 37b, Figure 37d, and Figure 37f show the fast Fourier
transform of the high resolution TEM images confirming the presence of a TB. Since the planes
parallel to the TB plane are also (1 1 -1) (yellow lines in the high resolution images), the TBs
depicted in Figure 37 are symmetrical, tilt, and twist boundaries, which is the definition used to
categorized Σ3 {1 1 1} coherent twin boundaries (CTBs) [41, 44]. Notice that in Figure 37, the
angle α between the TB plane and a plane perpendicular to the film growth direction is ~70.5°
for both Al and Al-Mg, while in the case of Ni α is ~ 50°. The three examples of inclined CTBs
show that there is a change in the texture of the columnar grain, which will be addressed in
section 7.2.
79

Figure 37. Low magnification and high resolution images of: Al (a and b), Al-Mg (c and d), Ni (e and f). The
columnar grains are marked by white dashed lines, while the red dotted lines mark an inclined TB (a, c, and e). The
inset (a, c, and e) SAED patterns show the typical double hexagon formed due to the presence of TBs, the SAED
patterns are decorated with blue and magenta dotted lines to form the double hexagon as a consequence of a
reversal in the stacking sequence of the material, the diffraction spots are indexed according to the plane
orientation. High resolution images of the square regions in a), c) and d) are presented in b), d) and e) respectively.
The CTBs marked with red dotted lines lie in a (1 1 -1) or (-1 -1 1) plane, while the blue and magenta dotted lines
80

show the (1 1 1) planes at each side of the two sides of the CTB. The insets in b), d) and f) are fast Fourier
transforms of the high resolution image, notice the double hexagon typical of a CTB.

7.1.1 Microstructural features of the films
Several TEM images such as the ones in Figure 36 and Figure 37 were used to calculate the
average grain width of the columnar grains, and to estimate the density of TBs in the material.
The data is presented in Table 8, which lists the SFE of the material, the average grain width of
the columnar grains, the density of TBs per unit area, the total area analyzed to count TBs, and
the fraction of twinned grains for the Al, Al-Mg, and Ni films. The SFE of the Al-Mg alloy is
unknown, however, Muzyk et al. estimated qualitative the SFE values for the main alloying
elements in Al commercial alloys by modeling. The calculations in this study predict that the
SFE will decrease by alloying Al with Cu, Mg, Si, or Zn; and can be ordered from high to low
SFE in the following sequence: Al-Cu > Al-Si > Al-Zn > Al-Mg [57]. Therefore the SFE of the
Al-Mg is assumed to be less than Al (166 mJ/m
2
). The average grain width was calculated by
measuring the width of at least 150 grains, while the TB density was calculated by counting at
least 500 TBs in each film. The Al film presented the highest average grain width (290 nm)
among the three films, while the Ni film presented the highest TB density (69 TBs/μm
2
). To
calculate the fraction of twinned grains, it was defined that a grain is consider to be a twinned
grain if it contains at least three TBs. Thus, the Ni film presented the highest fraction of twinned
grains (90%). The TB density in Al-Mg is 13 times higher than that of Al. However, the fraction
of twinned grains is only 1.5 times higher, this difference between these two films can be
attributed to the larger grain width of Al.

81

Table 8. Microstructural properties of the Al, Al-Mg and Ni films.
Sample
name
SFE
(mJ/m
2
)
Average
grain
width
(nm)
Density of
TBs per unit
area
(TBs/μm
2
)
Total area
analyzed
(μm
2
)
Fraction of
twinned grains
(%)
Al 166 290±40 3.4 151 46
Al-Mg <166 90±10 47 11 70
Ni 125 70±10 69 20.7 90

The overall difference in TB density and the fraction of twinned grains between the three
samples can be initially attributed to the fact that Ni has a lower SFE compare to Al or Al-Mg.
Alternatively the low fraction of twinned grains in Al and Al-Mg compared to Ni can be
associated to the inclined angle α of the CTBs, and an explanation is provided as follows: since
the TEM images used to count TBs are from the cross-section of the films, the TB plane of a
high α angle inclined CTB has lower chances to be nearly perpendicular to the 110 zone axis of
the film cross-section than the TB plane of a low α angle inclined CTB. For example, in Figure
38a, a columnar grain with a CTB inclined at α ~70.5° with respect to a plane perpendicular to
the film growth direction was constructed, the columnar grain has similar atomic arrangement to
the one shown in Figure 37d. The CTB plane in Figure 38a is perpendicular to the 110 zone axis
of the columnar grain, which is confirmed by the simulated diffraction pattern. In this case, the
TB will be easily observed by TEM. On the other hand, if the columnar grain was rotated around
the [1 1 1] direction by 30°, and/or 60° it would not be possible to observe the CTB by TEM as
shown in Figure 38b and Figure 38c, respectively. Moreover, a 90° rotation over the [1 1 1]
direction of the columnar grain (Figure 38d) completely blocks the visualization of the CTB. In
Figure 38b, Figure 38c and Figure 38d the simulated diffraction pattern was included to
demonstrate that it is not possible to distinguish the changes in stacking sequence promoted by
82

the CTB when the columnar grain is rotated. Thus, for this example, when the cross-section of
the film is obtained from the material, the CTBs that can be observed by TEM are only the CTBs
with a TB plane that is nearly perpendicular to the 110 cross-section view of the material. The
procedure described for a CTB inclined at α ~70.5° applies to any inclined CTB, and as α
decreases to zero the chances to observe CTBs in the 110 cross-section of the material will
increase. This is typical for CTBs where the TB plane is perpendicular to the [1 1 1] growth
direction of the film, which has been observed in high texture nt metals with low SFE [11, 60,
90]. Overall, the fraction of twinned grains reported in Table 8 for the three films can be
hindered due to the nature of the inclined CTBs, which promote unavoidable geometrical
restrictions when the material is analyzed by TEM.
83

Figure 38. Rotation of a columnar grain with a TB inclined 70.5°. a) TB aligned in the 110 zone axis, the TB plane
is perpendicular to 110 zone axis. b) the columnar grain is rotated 30 over the [111] direction with respect to a). c)
the columnar grain is rotated 60 over the [111] direction with respect to a). d) the columnar grain is rotated 90 over
the [111] direction with respect to a). the corresponding simulated diffraction patterns show the typical double
hexagon of a TB. Notice that only when the TB plane is perpendicular to the 110 zone axis the diffraction pattern
provide information of two distinguishable stacking sequences.

7.2 Texture characterization
The occurrence of the experimentally observed inclined CTBs can be associated with a change in
the texture of the films. For example, in the case of Al and Al-Mg films at one side of the CTB,
84

there are (111) planes that are perpendicular to the [1 1 1] grain growth direction as well as the
film growth direction, while at the other side of the CTB, the (1 1 1) planes are inclined with
respect to the film growth direction (Figure 37b and Figure 37d). Moreover, in the case of the Ni
film at each of the two sides of the CTB the (1 1 1) planes are inclined with respect to the film
growth direction (Figure 37f). XRD and T-EBSD were used to investigate the overall film
texture, and to support the observations by TEM. Specifically, T-EBSD was used to identify the
texture of the grains before and after a CTB. In EBSD, a boundary is considered a Σ3 CTB if the
misorientation angle between two grains is ~60° and if the misorientation axis between the same
two grains is perpendicular to a {111} plane. Figure 39 shows the XRD patterns of the Al, Al-
Mg, and Ni films. The Al and Al-Mg films XRD patterns both show a strong {111} texture,
while the Ni film XRD pattern presents a random texture.

Figure 39. Normalized XRD patterns of the Al, Al-Mg, and Ni films. Al and Al-Mg have a strong {111} texture,
while Ni has a random texture.
85

Figure 40 shows T-EBSD scans of single columnar grains from the films cross-section. In Figure
40a thru Figure 40d, the inverse pole figures of a columnar grain with CTBs are identified by the
change in color in each columnar grain. The legend of the inverse pole figure is included to
identify the change in texture for each of the columnar grains after an inclined CTB. Notice that
in the case of Ni, two scans are included to show CTBs that are perpendicular (Figure 40c) and
inclined (Figure 40d) to the film growth direction. Table 9 lists each of the columnar grains
shown in Figure 40, where the planes before and after the CTBs are labeled by the white
numbered circles. The OIM software analysis was used to find the plane orientation for each of
the white numbered circles with respect to the film growth direction, as well as the
misorientation angle and misorientation axis between consecutive numbered white circles. For
example, in Figure 40a the columnar grain contains three numbered circles, that are also shown
on the right-hand side texture legend at their corresponding orientations, while in Table 9 the
plane that corresponds to the number 1 and number 2 white circles are listed as (-1 1 1) and (1 -1
5) respectively. The misorientation angle and the plane perpendicular to the missorientation axis
between the two planes is 60° and (1 -1 1), respectively. In general, the Al and Al-Mg films
showed mainly two textures before and after the CTB {1 1 1} and {1 1 5}, while Ni showed a
random texture, similar to the observations in the XRD patterns in Figure 39.
86

Figure 40. T-EBSD scans of single columnar grains. a) Al, b) Al-Mg), c) and d) Ni. Notice that in Al and Al-Mg the
change in texture occurred between {111} and {115} planes, while in Ni the change in texture is from random plane
orientations. Two examples are given for the Ni film to show horizontal and inclined CTBs.

87

Table 9. T-EBSD analysis of the CTBs in Al, Al-Mg, and Ni films.
Sample
name
Figure 40
label
CTB between
the white
points in Figure
40
Plane before
CTB
Plane after
CTB
Missorientation
angle
Plane
perpendicular to
Missorientation
axis
Al a)
1 and 2 (-1 1 1) (1 -1 5) 59.8 (1 -1 1)
2 and 3 (1 -1 5) (-1 1 1) 59.1 (-17 17 16)
Al-Mg b) 1 and 2 (9 -8 9) (-5 7 26) 59.9 (1 -1 -1)
Ni c)
1 and 2 (-3 -1 8) (-7 4 3) 59.3 (1 -1 1)
2 and 3 (-7 4 3) (-5 -2 13) 59.8 (17 18 -17)
3 and 4 (-5 -2 13) (-9 5 4) 59.4 (1 -1 1)
4 and 5 (-9 5 4) (-10 4 9) 59.5 (1 -1 -1)
Ni d)
1 and 2 (4 -2 21) (-10 7 15) 60.0 (1 1 -1)
2 and 3 (-10 7 15) (4 -2 21) 60.0 (-1 -1 1)

Despite the strong {1 1 1} texture in Al and Al-Mg observed in Figure 39, the CTBs were not
observed in planes perpendicular to the film growth direction, or to the [111] grain growth
direction. Moreover, in the case of the Ni film, some CTBs are perpendicular to the growth
direction of the film, but due to the fact that the grains are randomly oriented, it was found that
none of the CTBs are lying in either a (1 1 1) or a (-1 -1 -1) plane perpendicular to the [1 1 1]
grain growth direction. The analysis with T-EBSD in Table 9 revealed that the CTBs lie and/or
are nucleated in any {1 1 1} family plane except for the (1 1 1) and (-1-1-1) planes that lie
perpendicular to the [1 1 1] grain growth direction. These results are contrary to the observation
of CTBs in low SFE materials with a strong {1 1 1} texture [9, 11, 25, 55, 60], where the CTBs
preferentially lie in the (1 1 1) plane that is perpendicular to the [1 1 1] grain growth direction
and no change in texture in the columnar grain is observed. Therefore, the inclined CTBs shown
in this manuscript promote a change in the texture of the columnar grains.
88

7.3 Formation of inclined CTBs
Recently, an analytical model proposed by Xue et al. suggests that the formation of inclined TBs
is more favorable than parallel TBs in polycrystalline Al [64]. Note that the terms inclined and
parallel are in reference to the [1 1 1] grain growth direction. The model predicts that during the
earlier stages of sputtering, a nucleus with an inclined CTB can develop and grow if the area of
an incoherent TB (ITB) in contact with the matrix/TB plane is significantly smaller than the area
occupied by the inclined CTB plane. The lower the ratio is between the area of the ITB and CTB
plane, the higher the probability is for a nucleus with an inclined CTB to develop. In addition,
Xue et al. stated that the increasing likelihood of inclined CTBs is truncated as the thickness of
the film is > 80 nm, since the continuous growth of an inclined CTB becomes energetically
unfavorable as the film thickness increases [64].
The thickness of the films presented in this manuscript is ~ 10 μm, the fraction of twinned grains
in the Al film is at least four times higher compared to the study by Xue et al., where the fraction
of twinned grains is 9% [64]. Since the thick films were synthesized at a high deposition rate, it
is possible that the nucleation of inclined CTBs follows the analytical model proposed by Xue et
al at the earlier stages of film deposition. Notably, the length of the inclined CTBs observed in
the thick films is > 80 nm, which suggest that their formation is still energetically favorable even
as the film gets thicker. However, it is unknown why the inclined CTBs occurred during
continuous growth of the films. To address this question, an explanation is presented as follows:
In sputtering, the continuous growth of a columnar grain occurs by the arrival of atoms to the
surface of the grain, where an embryo/nucleus free of defects can form and coalesce, forming
monolayers that follow epitaxial growth [79, 91]. The formation of a nucleus with any type of
89

defect requires higher energy than that of a nucleus without defects. Since the SFE of Al is high,
the formation of a nucleus with a stacking fault or TB defect is rare. It was observed that the
inclined CTBs studied in this manuscript cross the entire width of the columnar grains and that
they lie in a plane that is not perpendicular to the [1 1 1] grain growth direction. Therefore, it is
proposed that during continuous growth of the grain, the formation of a nucleus with an inclined
CTB (similar to the one described at the early stages of sputtering by Xue et al) can occur at the
lateral edges of the columnar grain (grain boundaries are regions of high energy, where the
nucleation of precipitates and second phases is feasible), and in this case the area from a small
portion of the columnar grain boundary can replace the area of the ITB that is in contact with the
matrix/TB plane. Since inside the nucleus, the ratio between the area occupied by the columnar
grain boundary and the area occupied by the inclined CTB plane is very small, it is possible that
the nucleus with an inclined CTB can develop at the lateral edge of a columnar grain. If the
nucleus is stable, the CTB will propagate through the entire width of the grain due to the constant
arrival of atoms, and it will be terminated upon arrival to the other side of the columnar grain.
The formation of the inclined CTB must be accompanied by a change in the grain growth
direction, which is reflected in the texture of the grain. The observations presented in the texture
analysis provide evidence of such changes in the grain growth direction and the change in texture
in a columnar grain, which gives foundation to the formation of inclined CTBs at the edges of
the columnar grains during continuous grain growth in sputtering.
The previous explanation allows understanding of the differences observed in TB density and
fraction of twinned grains between Al and Al-Mg (Table 8). The grain width of Al is three times
wider than that of Al-Mg, which can be attributed to the effect of the Zener drag force, where
impurities or alloying elements can impinge grain boundaries and prevent grain growth [81]. The
90

development of a nucleus with an inclined CTB at the edge of the columnar grain can be
truncated due to the mobility of the lateral grain boundary. As the grain grows laterally, it is
possible that a forming CTB nucleus detaches from the grain boundary, which makes the nucleus
unstable, preventing its development. Therefore, the observed large grain size in the Al film is an
indicator of high grain boundary mobility during continuous grain growth. This effect could
lower the formation of inclined CTBs during the film synthesis. As a consequence, by inhibiting
lateral grain growth in Al it can be possible to enhance CTB nucleation, similar to the
observations in the Al-Mg or Ni film, where the grain width is three times smaller than in the Al
film.
7.4 Conclusions
In this study, thick films of Al, Al-Mg and Ni with CTBs were successfully synthesized. The
density of CTBs and the fraction of twinned grains are higher compared to other studies with
similar materials using magnetron sputtering. It was shown that the observed fraction of twinned
grains can be hindered due to unavoidable geometrical restrictions associated with the inclined
angle of the CTBs. The use of complementary characterization techniques unveils the change in
texture in individual grains that occurs after the formation of a CTB, which is independent of the
overall texture in the films. In general, the experimental observations provide an explanation of
the formation of CTBs during continuous grain growth in the films, and emphasizes the
influence of grain boundary mobility on the nucleation of CTBs in high SFE metals.
Additionally, the results show that is possible to escalate the synthesis of CTBs in high SFE
metals from thin films to thick films, and expand the working space of metals with high SFEs.
91

8 General conclusions and future research recommendations
Previous and current studies on nanostructured materials with low stacking fault energies have
revealed that the formation of high densities of twin boundaries (TBs) enhances mechanical,
thermal, electrical and/or chemical properties of the material. These attractive properties have
encouraged the development of nanotwinned (nt) metals as a promising and growing area of
research. However, the synthesis of such materials has encountered many obstacles such as: (i)
the lack of control over nanostructural features in nt metals (twin density or twin thickness),
which highly impacts the properties of the material; and (ii) the stacking fault energy, which is a
natural barrier to the formation of TBs and has limited the synthesis of nt metals to materials
with a low SFE.
In this dissertation, a comprehensive study of twin boundary formation phenomena was
conducted in a wide range of SFE materials, specifically the synthesis of nt metals by using
magnetron sputtering. The first part of this dissertation addressed the synthesis of nt metals by
isolating the effect of SFE on the nanostructural features of nt metals with low SFE, followed by
studying the effect of sputtering synthesis parameters on the formation of fully nt metals and on
the resulting nanostructural features. These results highlight the influence of specific sputtering
conditions, which then provide the basis for controlling the nanostructural features of nt metals
through sputtering.
The experimental results presented in the second part of this dissertation show the synthesis of
fully nt films and how to control the nanostructural features in materials with low and
intermediate SFEs. The comprehensive study conducted on Cu alloy films (film thickness >15
μm) elucidated the three main mechanisms to describe twin nucleation and TB mobility, where
92

the interplay of the mechanisms granted control of the twin thickness in nt metals and maintained
a high twin density. The results from this study were summarized in a protean contour zone map
that can be used as a versatile guide to synthesize fully nt metals with tailored twin thicknesses
and provides a tool to control the amount of twin boundaries in materials with a wide range of
SFEs (6 - 60 mJ/m
2
), thus expanding the synthesis space for nanostructure materials.
In regards to the formation of TBs in high SFE (SFE > 125 mJ/m
2
) metals, a study was
conducted by synthesizing Ni, Al, and Al alloys thick films (film thickness >10 μm) using
magnetron sputtering. It was shown that the fraction of twinned grains in these films is higher
compared to other studies. The use of complementary characterization techniques shed light on
the investigation of the nature of twin boundaries in high SFE materials, and provided an
alternative perspective on the evaluation and formation of twin boundaries. Additionally, the
results show that is possible to escalate the synthesis of TBs in high SFE metals from thin films
to thick films, explore new directions on the synthesis of nt metals, and expand the working
space of metals with high SFEs.
The expansion and future research of the work presented in this dissertation is vast and
compelling. Some critical research areas that can contribute to the development of nt metals are:
1) exploring the synthesis of nt metals using the protean contour zone map as a guide to tune the
nanostructural features of different alloys systems, for instance Cu-Zn, Inconel, or stainless steel;
2) studying the mechanical, thermal, and/or chemical properties of the nt films produced in this
dissertation, and compare the behavior of the evaluated property by varying the nanostructural
features; 3) expanding the synthesis method developed to enhanced the density of TBs not only
in Al or Ni but to other high SFE metals; and 4) scaling the production of thick nt metal films to
the order of several hundreds of micrometers.
93

Overall, the studies in this dissertation open discussion on the formation of twin boundaries in
materials with a wide range of stacking fault energies, provide tools that can be used for
controlling the nanostructural features of nt materials, build on the characterization of twin
boundaries, and contribute to expanding the working space of nanostructural materials for
potential future applications.

94

9 References
[1] A.M. Hodge, Y.M. Wang, T.W. Barbee Jr. Mechanical deformation of high-purity
sputter-deposited nano-twinned copper, Scripta Materialia 59 (2008) 163-166.
[2] L. Lu, X. Chen, X. Huang, K. Lu. Revealing the Maximum Strength in Nanotwinned
Copper, Science 323 (2009) 607-610.
[3] L. Lu, Y. Shen, X. Chen, L. Qian, K. Lu. Ultrahigh Strength and High Electrical
Conductivity in Copper, Science 304 (2004) 422-426.
[4] Y. Zhao, I.C. Cheng, M.E. Kassner, A.M. Hodge. The effect of nanotwins on the
corrosion behavior of copper, Acta Materialia 67 (2014) 181-188.
[5] T.A. Furnish, J. Lohmiller, P.A. Gruber, J. T. W. Barbee, A.M. Hodge. Temperature-
dependent strain localization and texture evolution of highly nanotwinned Cu, Applied Physics
Letters 103 (2013) 011904.
[6] Y. Chen, K.Y. Yu, Y. Liu, S. Shao, H. Wang, M.A. Kirk, J. Wang, X. Zhang. Damage-
tolerant nanotwinned metals with nanovoids under radiation environments, Nat Commun 6
(2015).
[7] V. Randle. Twinning-related grain boundary engineering, Acta Materialia 52 (2004)
4067-4081.
[8] Y. Zhang, N.R. Tao, K. Lu. Effects of stacking fault energy, strain rate and temperature
on microstructure and strength of nanostructured Cu–Al alloys subjected to plastic deformation,
Acta Materialia 59 (2011) 6048-6058.
[9] A.M. Hodge, Y.M. Wang, T.W. Barbee Jr. Large-scale production of nano-twinned,
ultrafine-grained copper, Materials Science and Engineering: A 429 (2006) 272-276.
[10] X.Y. Li, Y.J. Wei, L. Lu, K. Lu, H.J. Gao. Dislocation nucleation governed softening and
maximum strength in nano-twinned metals, Nature 464 (2010) 877-880.
[11] X. Zhang, H. Wang, X.H. Chen, L. Lu, K. Lu, R.G. Hoagland, A. Misra. High-strength
sputter-deposited Cu foils with preferred orientation of nanoscale growth twins, Applied Physics
Letters 88 (2006) 173116.
[12] D. Bufford, H. Wang, X. Zhang. High strength, epitaxial nanotwinned Ag films, Acta
Materialia 59 (2011) 93-101.
[13] O. Anderoglu, A. Misra, H. Wang, F. Ronning, M.F. Hundley, X. Zhang. Epitaxial
nanotwinned Cu films with high strength and high conductivity, Applied Physics Letters 93
(2008) 083108.
[14] H. Jiang, T.J. Klemmer, J.A. Barnard, W.D. Doyle, E.A. Payzant. Epitaxial growth of
Cu(111) films on Si(110) by magnetron sputtering: orientation and twin growth, Thin Solid
Films 315 (1998) 13-16.
[15] X. Zhang, O. Anderoglu, R. Hoagland, A. Misra. Nanoscale growth twins in sputtered
metal films, JOM Journal of the Minerals, Metals and Materials Society 60 (2008) 75-78.
[16] W.D. Callister. Materials science and engineering: an introduction, John Wiley & Sons,
New York, 2007.
[17] T. Watanabe. Approach to grain boundary design for strong and ductile polycrystals, Res
Mechanica: International Journal of Structural Mechanics and Materials Science 11 (1984) 47-
84.
95

[18] A.M. Hodge, T.A. Furnish, C.J. Shute, Y. Liao, X. Huang, C.S. Hong, Y.T. Zhu, T.W.
Barbee Jr, J.R. Weertman. Twin stability in highly nanotwinned Cu under compression, torsion
and tension, Scripta Materialia 66 (2012) 872-877.
[19] M. Dao, L. Lu, Y.F. Shen, S. Suresh. Strength, strain-rate sensitivity and ductility of
copper with nanoscale twins, Acta Materialia 54 (2006) 5421-5432.
[20] I.J. Beyerlein, X. Zhang, A. Misra. Growth Twins and Deformation Twins in Metals,
Annual Review of Materials Research 44 (2014) 329-363.
[21] A.M. Hodge, T.A. Furnish, A.A. Navid, T.W. Barbee Jr. Shear band formation and
ductility in nanotwinned Cu, Scripta Materialia 65 (2011) 1006-1009.
[22] T.A. Furnish, A.M. Hodge. On the mechanical performance and deformation of
nanotwinned Ag, APL Materials 2 (2014) -.
[23] N.M. Heckman, L. Velasco, A.M. Hodge. Influence of Twin Thickness and Grain Size
on the Tensile Behavior of Fully Nanotwinned CuAl Alloys Advanced Engineering Materials
(2015) n/a-n/a.
[24] Y.F. Shen, L. Lu, Q.H. Lu, Z.H. Jin, K. Lu. Tensile properties of copper with nano-scale
twins, Scripta Materialia 52 (2005) 989-994.
[25] L. Velasco, A.M. Hodge. The mobility of growth twins synthesized by sputtering:
Tailoring the twin thickness, Acta Materialia 109 (2016) 142-150.
[26] O. Anderoglu, A. Misra, H. Wang, X. Zhang. Thermal stability of sputtered Cu films
with nanoscale growth twins, Journal of Applied Physics 103 (2008) 094322.
[27] Y. Zhao, T.A. Furnish, M.E. Kassner, A.M. Hodge. Thermal stability of highly
nanotwinned copper: The role of grain boundaries and texture, Journal of Materials Research 27
(2012) 3049-3057.
[28] D. Bufford, H. Wang, X. Zhang. Thermal stability of twins and strengthening
mechanisms in differently oriented epitaxial nanotwinned Ag films, Journal of Materials
Research 28 (2013) 1729-1739.
[29] C. Hammond. Basics of Crystallography and Diffraction, OUP Oxford, Oxford, GBR,
2009.
[30] J. Pelleg. Mechanical Properties of Materials, Springer Netherlands, 2012.
[31] D. Hull, D.J. Bacon. 1 - Defects in Crystals. in: Hull D, Bacon DJ, (Eds.). Introduction to
Dislocations (Fourth Edition). Butterworth-Heinemann, Oxford, 2001. pp. 1-21.
[32] A. Rohatgi, K. Vecchio, G. Gray. The influence of stacking fault energy on the
mechanical behavior of Cu and Cu-Al alloys: Deformation twinning, work hardening, and
dynamic recovery, Metallurgical and Materials Transactions A 32 (2001) 135-145.
[33] L.E. Murr. Interfacial phenomena in metals and alloys / Lawrence E. Murr ; with a
foreword by Morris Cohen. Addison-Wesley Pub. Co., Advanced Book Program, Reading,
Mass, 1975. pp. 145-148.
[34] W. Li, S. Lu, Q.M. Hu, S.K. Kwon, B. Johansson, L. Vitos. Generalized stacking fault
energies of alloys, Journal of Physics-Condensed Matter 26 (2014).
[35] F.I. Grace, M.C. Inman. Influence of stacking fault energy on dislocation configurations
in shock-deformed metals, Metallography 3 (1970) 89-98.
[36] J.P. Hirth, J. Lothe. Theory of dislocations. Krieger Pub. Co, Malabar, FL, 1992. pp. 839.
[37] E. Aerts, P. Delavignette, R. Siems, S. Amelinckx. Stacking Fault Energy in Silicon,
Journal of Applied Physics 33 (1962) 3078-3080.
[38] M. Rester, C. Motz, R. Pippan. Stacking fault energy and indentation size effect: Do they
interact?, Scripta Materialia 58 (2008) 187-190.
96

[39] T.C. Schulthess, P.E.A. Turchi, A. Gonis, T.G. Nieh. Systematic study of stacking fault
energies of random Al-based alloys, Acta Materialia 46 (1998) 2215-2221.
[40] G. Gottstein, L.S. Shvindlerman. Grain Boundary Migration in Metals: Thermodynamics,
Kinetics, Applications, Taylor & Francis, 1999.
[41] V. Randle. The Measurement of Grain Boundary Geometry, Taylor & Francis, 1993.
[42] D.A. Smith. On the density of coincidence sites in grain boundaries, Scripta Metallurgica
8 (1974) 1197-1199.
[43] G. Dimou, K.T. Aust. Relative energies of grain boundaries near a coincidence
orientation relationship in high-purity lead, Acta Metallurgica 22 (1974) 27-32.
[44] R.C. Pond, V. Vitek. Periodic Grain Boundary Structures in Aluminium. I. A Combined
Experimental and Theoretical Investigation of Coincidence Grain Boundary Structure in
Aluminium, Proceedings of the Royal Society of London A: Mathematical, Physical and
Engineering Sciences 357 (1977) 453-470.
[45] Y.H. Zhao, J.F. Bingert, X.Z. Liao, B.Z. Cui, K. Han, A.V. Sergueeva, A.K. Mukherjee,
R.Z. Valiev, T.G. Langdon, Y.T. Zhu. Simultaneously Increasing the Ductility and Strength of
Ultra-Fine-Grained Pure Copper, Advanced Materials 18 (2006) 2949-2953.
[46] M. Niewczas. Chapter 75 Dislocations and Twinning in Face Centred Cubic Crystals. in:
Nabarro FRN, Hirth JP, (Eds.). Dislocations in Solids, vol. Volume 13. Elsevier, 2007. pp. 263-
364.
[47] F.J. Humphreys, M. Hatherly. Chapter 7 - Recrystallization of Single-Phase Alloys. in:
Hatherly FJH, (Ed.). Recrystallization and Related Annealing Phenomena (Second Edition).
Elsevier, Oxford, 2004. pp. 215-IV.
[48] R.E. Reed-Hill, R. Abbaschian. Physical metallurgy principles. 3rd ed., PWS-Kent Pub.,
Boston, 1992.
[49] C.S. Hong, N.R. Tao, X. Huang, K. Lu. Nucleation and thickening of shear bands in
nano-scale twin/matrix lamellae of a Cu–Al alloy processed by dynamic plastic deformation,
Acta Materialia 58 (2010) 3103-3116.
[50] X.H. Chen, L. Lu. Work hardening of ultrafine-grained copper with nanoscale twins,
Scripta Materialia 57 (2007) 133-136.
[51] R.F. Bunshah. Handbook of hard coatings: deposition technologies, properties and
applications, Noyes Publications; William Andrew Pub., Park Ridge, N.J. Norwich, N.Y., 2001.
[52] J.A. Thornton. Influence of apparatus geometry and deposition conditions on the
structure and topography of thick sputtered coatings, Journal of Vacuum Science and
Technology 11 (1974) 666-670.
[53] J.A. Thornton. Influence of substrate temperature and deposition rate on structure of thick
sputtered Cu coatings, Journal of Vacuum Science & Technology 12 (1975) 830-835.
[54] A. Anders. A structure zone diagram including plasma-based deposition and ion etching,
Thin Solid Films 518 (2010) 4087-4090.
[55] X. Zhang, A. Misra, H. Wang, T.D. Shen, M. Nastasi, T.E. Mitchell, J.P. Hirth, R.G.
Hoagland, J.D. Embury. Enhanced hardening in Cu/330 stainless steel multilayers by nanoscale
twinning, Acta Materialia 52 (2004) 995-1002.
[56] X. Zhang, O. Anderoglu, A. Misra, H. Wang. Influence of deposition rate on the
formation of growth twins in sputter-deposited 330 austenitic stainless steel films, Applied
Physics Letters 90 (2007) 153101.
[57] M. Muzyk, Z. Pakiela, K.J. Kurzydlowski. Ab initio calculations of the generalized
stacking fault energy in aluminium alloys, Scripta Materialia 64 (2011) 916-918.
97

[58] D.J. Siegel. Generalized stacking fault energies, ductilities, and twinnabilities of Ni and
selected Ni alloys, Applied Physics Letters 87 (2005).
[59] A. Rohatgi, K.S. Vecchio. The variation of dislocation density as a function of the
stacking fault energy in shock-deformed FCC materials, Materials Science and Engineering: A
328 (2002) 256-266.
[60] L. Velasco, M.N. Polyakov, A.M. Hodge. Influence of stacking fault energy on twin
spacing of Cu and Cu–Al alloys, Scripta Materialia 83 (2014) 33-36.
[61] S. Mahajan. Critique of mechanisms of formation of deformation, annealing and growth
twins: Face-centered cubic metals and alloys, Scripta Materialia 68 (2013) 95-99.
[62] D. Bhattacharyya, X.-Y. Liu, A. Genc, H.L. Fraser, R.G. Hoagland, A. Misra. Heterotwin
formation during growth of nanolayered Al-TiN composites, Applied Physics Letters 96 (2010).
[63] D. Bufford, Y. Liu, Y. Zhu, Z. Bi, Q.X. Jia, H. Wang, X. Zhang. Formation Mechanisms
of High-density Growth Twins in Aluminum with High Stacking-Fault Energy, Materials
Research Letters 1 (2013) 51-60.
[64] S. Xue, Z. Fan, Y. Chen, J. Li, H. Wang, X. Zhang. The formation mechanisms of growth
twins in polycrystalline Al with high stacking fault energy, Acta Materialia 101 (2015) 62-70.
[65] B. Lin, Y. Jin, C.M. Hefferan, S.F. Li, J. Lind, R.M. Suter, M. Bernacki, N. Bozzolo,
A.D. Rollett, G.S. Rohrer. Observation of annealing twin nucleation at triple lines in nickel
during grain growth, Acta Materialia 99 (2015) 63-68.
[66] R. Wang, S.D. Dahlgren. Observation of twin faults in sputter-deposited high-purity
nickel, J Mater Sci 10 1456-1458.
[67] H. Shan, Y. Zhang, S. Mao, Z. Zhang. Electron microscopy study of five-fold twins in
electrodeposited nano-twin Ni, Acta Metall 50 (2014) 305-312.
[68] D.G. Brandon, W.D. Kaplan. Microstructural characterization of materials, John Wiley,
2008.
[69] EDAX. Electron Backscatter Diffraction (EBSD) Microscopy and Analysis Essential
Knowledge Briefing (2015).
[70] R.R. Keller, R.H. Geiss. Transmission EBSD from 10 nm domains in a scanning electron
microscope, Journal of Microscopy 245 (2012) 245-251.
[71] D.B. Williams, C.B. Carter. Transmission Electron Microscopy: A Textbook for
Materials Science, Springer, 2009.
[72] G.G. Stoney. The tension of metallic films deposited by electrolysis, Proceedings of the
Royal Society of London Series a-Containing Papers of a Mathematical and Physical Character
82 (1909) 172-175.
[73] M. Ohring. Chapter 7 - Substrate surfaces and thin-film nucleation. Materials Science of
Thin Films (Second Edition). Academic Press, San Diego, 2002. pp. 357-415.
[74] Y.K. Huang, A.A. Menovsky, F.R. de Boer. Calorimetric analysis of the grain growth in
nanocrystalline copper samples, Nanostructured Materials 2 (1993) 587-595.
[75] X. Zhang, A. Misra. Superior thermal stability of coherent twin boundaries in
nanotwinned metals, Scripta Materialia 66 (2012) 860-865.
[76] J. Wang, N. Li, O. Anderoglu, X. Zhang, A. Misra, J.Y. Huang, J.P. Hirth. Detwinning
mechanisms for growth twins in face-centered cubic metals, Acta Materialia 58 (2010) 2262-
2270.
[77] L. Xu, D. Xu, K.N. Tu, Y. Cai, N. Wang, P. Dixit, J.H.L. Pang, J. Miao. Structure and
migration of (112) step on (111) twin boundaries in nanocrystalline copper, Journal of Applied
Physics 104 (2008) 113717.
98

[78] R.L. Fullman, J.C. Fisher. Formation of Annealing Twins During Grain Growth, Journal
of Applied Physics 22 (1951) 1350-1355.
[79] M. Ohring. Chapter 9 - Film structure. in: Ohring M, (Ed.). Materials Science of Thin
Films (Second Edition). Academic Press, San Diego, 2002. pp. 495-558.
[80] D. Hull, D.J. Bacon. 4 - Elastic Properties of Dislocations. in: Hull D, Bacon DJ, (Eds.).
Introduction to Dislocations (Fourth Edition). Butterworth-Heinemann, Oxford, 2001. pp. 62-81.
[81] C.V. Thompson. Grain Growth in Thin Films, Annual Review of Materials Science 20
(1990) 245-268.
[82] K. Sobczyk, D.J. Kirkner. Stochastic Modeling of Microstructures, Birkhäuser Boston,
2001.
[83] A. Stukowski, K. Albe, D. Farkas. Nanotwinned fcc metals: Strengthening versus
softening mechanisms, Physical Review B 82 (2010) 224103.
[84] Y. Wei. Scaling of maximum strength with grain size in nanotwinned fcc metals,
Physical Review B 83 (2011) 132104.
[85] E.O. Ezugwu, Z.M. Wang, A.R. Machado. The machinability of nickel-based alloys: a
review, Journal of Materials Processing Technology 86 (1999) 1-16.
[86] R.E. Sanders. Technology innovation in aluminum products, JOM 53 (2001) 21-25.
[87] Z. Xu, N. Li, H. Jiang, L. Liu. Deformation nanotwins in coarse-grained aluminum alloy
at ambient temperature and low strain rate, Materials Science and Engineering: A 621 (2015)
272-276.
[88] F. Zhao, L. Wang, D. Fan, B.X. Bie, X.M. Zhou, T. Suo, Y.L. Li, M.W. Chen, C.L. Liu,
M.L. Qi, M.H. Zhu, S.N. Luo. Macrodeformation Twins in Single-Crystal Aluminum, Physical
Review Letters 116 (2016) 075501.
[89] M.W. Chen, J.W. McCauley, D.P. Dandekar, N.K. Bourne. Dynamic plasticity and
failure of high-purity alumina under shock loading, Nat Mater 5 (2006) 614-618.
[90] P. Zhang, J.Y. Zhang, J. Li, G. Liu, K. Wu, Y.Q. Wang, J. Sun. Combined effect of
texture and nanotwins on mechanical properties of the nanostructured Cu and Cu–Al films
prepared by magnetron sputtering, J Mater Sci 50 (2015) 1901-1907.
[91] K.S.S. Harsha. Principles of Vapor Deposition of Thin Films, Elsevier Science, 2005.

99

Appendix A. Summary of sputtered samples
This appendix contains tables with the list of sputtered films in the course of this dissertation. X-
ray diffraction (XRD), focused ion beam (FIB), transmission electron microscopy (TEM), not
applicable (NA).
Table 10. Cu sputtered samples
Sample name
Working
pressure
(mTorr)
Working
distance
(inch)
Power
density
(w/cm^2)
Thickness
(μm)
Deposition
rate (nm/s)
Characterization
performed or comments
Cu#1-dec-12 5 3 3.2 4 0.8 XRD, FIB
Cu#2-dec-12 5 3 3.2 10 0.8 XRD, FIB
Cu#33-dec-12 2 3 3.2 4 0.9 XRD
Cu#34-dec-12 2 3 3.2 12 0.9 XRD, FIB, TEM
Cu#4-dec-12 2 3 6.4 6 1.7 XRD, FIB
Cu#5-jan-13 2 6 15 10 0.9 XRD, FIB
Cu#6-feb-13 2 6 15 11 1 XRD, FIB
Cu#7-feb-13 2 4 10 11 2 XRD
Cu#8-mar-13 2 3 3.2 8 2 XRD
Cu#9-mar-13 2 3 3.2 8 2 XRD
Cu#10-dec-13 2.5 5.75 22 4 2 XRD, FIB
Cu#11-dec-13 2.5 5.75 30 5 2 XRD, FIB
Cu#12-dec-13 2.5 5.75 7.4 3 0.6 XRD, FIB
Cu#13 2 3 33 22 15 XRD, FIB
Cu#14 2 3 33 22 15 XRD
Cu#15 2 3 33 18 16 XRD, FIB
Cu#16 2 3 33 16 15 XRD, FIB
Cu#17 2 3 33 15 15 XRD, FIB
Cu#18 2 3 33 15 15 XRD, FIB
Cu#19 2 3 33 14 13 XRD, FIB
Cu#20 2 3 33 15 13 XRD, FIB
Cu#21 2 3 33 14 11 XRD, FIB

Table 11. Cu-Al alloys sputtered samples
Sample name
Working
pressure
(mTorr)
Working
distance
(inch)
Power
density
(w/cm^2)
Thickness
(μm)
Deposition
rate (nm/s)
Characterization
performed or
comments
CuAl#1-jul-12 2 6 10 4 0.6 XRD, FIB
CuAl#2-jul-12 5 6 10 3 0.5 XRD, FIB
CuAl#3-jul-12 10 6 10 2 0.3 XRD, FIB
CuAl#4-Aug-12 1.3 6 10 2 0.5 XRD
CuAl#5-Aug-12 5 6 10 21 0.5 XRD, FIB
CuAl#6-oct-12 5 3 2.5 1 0.7 XRD
100

CuAl#7-oct-12 5 3 2.5 3 0.7 XRD, FIB
CuAl#8-oct-12 5 3 2.5 3 0.6 XRD
CuAl#9-oct-12 5 3 2.5 3 0.7 XRD, FIB
CuAl#10-oct-12 5 3 3.2 4 0.8 XRD, FIB, TEM
CuAl#11-oct-12 2 3 3.2 4 0.9 XRD, FIB, TEM
CuAl#12-oct-12 10 3 3.2 4 0.8 XRD, FIB
CuAl#13-oct-12 5 3 3.2 4 0.8 XRD, FIB, TEM
CuAl#14-oct-12 5 3 3.2 4 0.8 XRD, FIB
CuAl#15-oct-12 5 3 3.2 20 0.7 XRD, FIB
CuAl#16-mar-13 2 3 3.2 4 0.9 XRD, FIB, TEM
CuAl#17-jun-13 2 3 3.2 4 0.8 XRD, FIB
CuAl#18-oct-13 2.2 3 3.2 3 0.6 XRD, FIB
CuAl#19-oct-13 2.2 3 1.5 2 0.4 XRD, FIB
CuAl#20-oct-13 2.2 3 15 20 4 XRD, FIB, TEM
CuAl#21a-nov-13 2.5 5.5 17 7 1 XRD, FIB
CuAl#21b-nov-13 2.5 5.5 17 7 1 XRD, FIB
CuAl#22a-nov-13 2.5 5.5 10 4 0.7 XRD, FIB
CuAl#22b-nov-13 2.5 5.5 10 4 0.7 XRD, FIB
CuAl#23-nov-13 2.7 5.5 2.5 2 0.2 XRD, FIB
CuAl#24a-nov-13 2.5 5.5 35 3 1 XRD, FIB
CuAl#24b-nov-13 2.5 8.5 35 NA NA XRD
CuAl#25-dec-13 2.7 5.5 25 NA NA XRD
CuAl#26-dec-13 2.5 5.5 30 5 2 XRD, FIB, TEM
CuAl#27-dec-13 2.5 5.5 22 3 1 XRD, FIB
CuAl#28-may-14 2 3 44 5 12 XRD
CuAl#29-may-14 2 3 44 6 15 XRD, FIB
CuAl#30-may-14 2 3 44 NA NA Experiment failed
CuAl#31-may-14 2 3 33 5 10 XRD, FIB
CuAl#32-jun-14 2 3 33 11 12 XRD, FIB
CuAl#33-jun-14 2 3 33 11 12 XRD, FIB
CuAl#34a-jun-14 2 3 33 10 11 XRD, FIB
CuAl#35-jun-14 2 3 33 29 12 XRD, FIB, TEM
CuAl#36-jun-14 2 3 33 60 12 XRD, FIB
CuAl#37-jul-14 2 3 33 29 12 XRD, FIB
CuAl#38-jul-14 2 3 33 28 12 XRD, FIB
CuAl#39-jul-14 2 3 33 27 11 XRD, FIB
CuAl#40-jul-14 2 3 33 27 11 XRD, FIB
CuAl#41-jul-14 2 3 33 31 13 XRD, FIB
CuAl#42-jul-14 2 3 33 28 15 XRD, FIB
CuAl#43-jul-14 2 3 33 28 15 XRD, FIB
CuAl#44-jul-14 2 3 33 22 14 XRD, FIB
CuAl#45-Aug-14 2 3 33 25 16 XRD, FIB
CuAl#46-Aug-14 2 3 33 30 18 FIB
CuAl#47-Aug-14 2 3 33 NA NA Experiment failed
CuAl#48-Aug-14 2 3 33 24 14 XRD, FIB
CuAl#49-Aug-14 2 3 33 NA NA XRD, FIB
CuAl#50-Aug-14 2 3 33 17 12 XRD, FIB, TEM
CuAl#51-Aug-14 2 3 33 18 12 XRD, FIB
CuAl#52-Aug-14 2 3 33 16 10 XRD, FIB, TEM
CuAl#53-Aug-14 2 3 33 24 10 XRD, FIB
CuAl#54-sep-14 2 3 33 23 10 XRD, FIB
CuAl#55-sep-14 2 3 33 20 11 XRD, FIB
101

CuAl#56-sep-14 2 3 33 19 11 XRD, FIB, TEM
Temp-CuAl#57-sep-14 2 3 33 NA NA Temperature run
Temp-CuAl#58-sep-14 2 3 33 NA NA Temperature run
CuAl#59-sep-14 2 3 33 NA NA Experiment failed
CuAl#60-sep-14 2 3 33 18 12 XRD, FIB
CuAl#61-sep-14 2 3 33 18 11 XRD, FIB
CuAl#62-sep-14 2 3 33 18 12 XRD, FIB
CuAl#63-sep-14 2 3 33 17 11 XRD, FIB
CuAl#64-sep-14 2 3 33 8 12 XRD, FIB
CuAl#65-sep-14 2 3 33 11 12 XRD, FIB
CuAl#66-sep-14 2 3 33 20 15 XRD, FIB
CuAl#67-sep-14 2 3 33 20 15 XRD, FIB, TEM
CuAl#68-sep-14 2 3 33 19 13 XRD, FIB, TEM
CuAl#69-oct-14 2 3 33 19 13 XRD, FIB, TEM
CuAl#70-oct-14 2 3 33 17 12 XRD, FIB, TEM
CuAl#71-oct-14 2 3 33 18 14 XRD, FIB
CuAl#72-oct-14 2 3 33 19 14 XRD, FIB
CuAl#73-oct-14 2 3 33 19 14 XRD, FIB, TEM
CuAl#74-oct-14 2 3 33 19 14 XRD, FIB
CuAl#75-oct-14 2 3 33 NA NA XRD
CuAl#76-oct-14 2 3 33 19 14 XRD, FIB, TEM
CuAl#77-oct-14 2 3 33 20 14 XRD, FIB
CuAl#78a-oct-14 2 3 33 16 11 XRD, FIB
CuAl#78b-oct-14 2 3 33 14 10 XRD
CuAl#79-oct-14 2 3 11 NA NA Temperature run
CuAl#80-oct-14 2 3 33 NA NA Temperature run
CuAl#81-nov-14 2 3 22 NA NA Temperature run
CuAl#82-nov-14 2 3 22 NA NA Temperature run
CuAl#83-nov-14 2 3 1.4 10 0.8 XRD, FIB
CuAl#84-nov-14 2 3 1.4 10 0.8 XRD, FIB
CuAl#85-nov-14 2 3 1.4 10 0.8 XRD, FIB
CuAl#86-nov-14 2 3 1.4 9 0.8 XRD, FIB
CuAl#87a-nov-14 2 4 33 8 7 XRD, FIB
CuAl#87b-nov-14 2 4 33 9 7 XRD, FIB
CuAl#88a-nov-14 2 3 33 14 12 XRD, FIB
CuAl#88b-nov-14 2 3 33 13 11 XRD, FIB
CuAl#89a-nov-14 2 3 33 15 13 XRD, FIB, TEM
CuAl#89b-nov-14 2 3 33 14 12 XRD, FIB, TEM
CuAl#90a-nov-14 2 3 22 17 10 XRD, FIB
CuAl#90b-nov-14 2 3 22 14 8 XRD, FIB
CuAl#91a-nov-14 2 3 11 11 4 XRD, FIB
CuAl#91b-nov-14 2 3 11 11 4 XRD, FIB
CuAl#92a-nov-14 2 3 1.4 9 0.7 XRD, FIB
CuAl#92b-nov-14 2 3 1.4 8 0.7 XRD, FIB
CuAl#93-nov-14 30 3 33 18 15 XRD, FIB
CuAl#94-nov-14 20 3 33 18 15 XRD, FIB
CuAl#95-nov-14 10 3 33 19 16 XRD, FIB
CuAl#96-nov-14 2 3 1.4 10 0.8 XRD, FIB
CuAl#97-nov-14 2 3 1.4 9 0.8 XRD, FIB
CuAl#98-nov-14 2 3 33 16 14 XRD, FIB
CuAl#99-nov-14 2 3 33 20 15 XRD, FIB, TEM
CuAl#100-nov-14 2 3 33 20 14 XRD, FIB, TEM
102

CuAl#101-nov-14 2 3 1.4 19 0.8 XRD, FIB
CuAl#102-dec-14 2 3 1.4 15 0.6 XRD, FIB, TEM
CuAl#103-dec-14 2 3 1.4 16 0.6 XRD, FIB, TEM
CuAl#104-dec-14 2 3 33 17 12 XRD, FIB, TEM
CuAl#105-dec-14 2 3 1.4 16 0.6 XRD, FIB
CuAl#106b-dec-14 2 3 33 14 10 XRD, FIB
CuAl#107-dec-14 2 3 1.4 17 0.7 XRD, FIB
CuAl#108-dec-14 2 3 1.4 19 0.8 XRD, FIB
CuAl#109-jan-15 2 3 1.4 20 0.8 XRD
CuAl#110-jan-15 2 3 33 20 14 XRD
CuAl#111-jan-15 2 3 22 22 12 XRD
CuAl#112-jan-15 2 3 33 19 14 XRD
CuAl#113-jan-15 2 3 22 18 10 XRD
CuAl#114-jan-15 2 3 1.4 20 0.8 XRD
CuAl#115-jan-15 2 3 33 26 14 XRD
CuAl#116-jan-15 2 3 22 18 10 XRD
CuAl#117-jan-15 2 3 33 17 12 XRD
CuAl#118-jan-15 2 3 22 NA NA XRD
CuAl#119-jan-15 2 3 1.4 17 0.7 XRD
CuAl#120-jan-15 2 3 33 17 12 XRD
CuAl#121-jan-15 2 3 33 NA NA XRD
CuAl#122-jan-15 2 3 33 16 12 XRD, FIB, TEM
CuAl#123-jan-15 2 3 22 15 8 XRD
CuAl#124-jan-15 2 3 1.4 17 0.7 XRD
CuAl#125-jan-15 2 3 22 15 8 XRD, FIB, TEM
CuAl#126-jan-15 2 3 33 17 12 XRD, FIB, TEM
CuAl#127-jan-15 2 3 1.4 19 0.7 XRD
CuAl#128-jan-15 2 3 33 19 14 XRD, FIB, TEM
CuAl#129-jan-15 2 3 33 18 13 XRD, FIB, TEM
CuAl#130-jan-15 2 3 22 NA NA XRD
CuAl#131-jan-15 2 3 22 18 10. XRD
CuAl#132-jan-15 2 3 1.4 19 0.8 XRD
CuAl#133-jan-15 2 3 33 18 13 XRD, FIB, TEM
CuAl#134-jan-15 2 3 22 18 10 XRD
CuAl#69 temp run 2 3 33 NA NA Temperature run
CuAl#70 temp run 2 3 33 NA NA Temperature run
CuAl#104 temp run 2 3 33 NA NA Temperature run
CuAl#103 temp run 2 3 1.4 NA NA Temperature run
CuAl#122 temp run 2 3 33 NA NA Temperature run
CuAl#126 temp run 2 3 33 NA NA Temperature run
CuAl#125 temp run 2 3 22 NA NA Temperature run
CuAl#124 temp run 2 3 1.4 NA NA Temperature run
CuAl#135 same as
CuAl#69
2 3 33 NA NA Temperature run
CuAl#136 same as
CuAl#70
2 3 33 NA NA Temperature run
CuAl#137 same as
CuAl#104
2 3 33 NA NA Temperature run
CuAl#138 same as
CuAl#103
2 3 1.4 NA NA Temperature run
CuAl#139 temp run 2 3 1.4 NA NA Temperature run
CuAl#140 same as
CuAl#69
2 3 33 NA NA Temperature run
103

CuAl#141 same as
CuAl#103
2 3 1.4 NA NA Temperature run
CuAl#142 same as
CuAl#70
2 3 33 NA NA Temperature run
CuAl#143 same as
CuAl#104
2 3 33 NA NA Temperature run
CuAl#144 same as
CuAl#16
2 3 3.2 NA NA Temperature run
CuAl#145 same as
CuAl#20
2 3 15 NA NA Temperature run
CuAl#145 same as
CuAl#20
2 3 15 NA NA Temperature run
CuAl#146 same as
CuAl#122
2 3 33 NA NA Temperature run
CuAl#147 same as
CuAl#125
2 3 22 NA NA Temperature run
CuAl#148 same as
CuAl#126
2 3 33 NA NA Temperature run
CuAl#149 2 3 33 20 14 Temperature run
CuAl#150 2 3 33 19 14 Temperature run
CuAl#151 temp run 2 3 33 NA NA Temperature run
CuAl#152 temp run 2 3 33 NA NA Temperature run
CuAl#153 temp run 2 3 1.4 NA NA Temperature run
CuAl#154 temp run 2 3 33 NA NA Temperature run
CuAl#155 temp run 2 3 3.2 NA NA Temperature run
CuAl#156 temp run 2 3 15 NA NA Temperature run
CuAl#157 temp run 2 3 33 NA NA Temperature run
CuAl#158 temp run 2 3 33 NA NA Temperature run
CuAl#159 temp run 2 3 22 NA NA Temperature run

Table 12. Cu-Ni alloy sputtered samples
Sample name
Working
pressure
(mTorr)
Working
distance
(inch)
Power
density
(w/cm^2)
Thickness
(μm)
Deposition
rate (nm/s)
Characterization
performed or
comments
CuNi#1-dec-14 2 3 33 NA NA XRD
CuNi#2-dec-14 2 3 33 17 12 XRD, FIB
CuNi#3-dec-14 2 3 33 15 11 XRD, FIB
CuNi#4-dec-14 2 3 33 17 12 XRD, FIB, TEM
CuNi#5-dec-14 2 3 1.4 16 0.6 XRD, FIB
CuNi#6-dec-14 2 3 1.4 16 0.6 XRD, FIB
CuNi#7-jan-15 2 3 1.4 16 0.6 XRD
CuNi#8-feb-15 2 3 33 15 11 XRD, FIB, TEM
CuNi#9-feb-15 2 3 33 12 9 XRD
CuNi#10-feb-15 2 3 33 17 12 XRD, FIB, TEM
CuNi#11-feb-15 2 3 33 17 13 XRD, FIB, TEM
CuNi#12-feb-15 2 3 33 15 11 XRD
CuNi#13-feb-15 2 3 33 18 13 XRD, FIB, TEM
CuNi#4 TEMP RUN 2 3 33 NA NA Temperature run
CuNi#11 TEMP RUN 2 3 33 NA NA Temperature run
CuNi#10 TEMP RUN 2 3 33 NA NA Temperature run
104

CuNi#13 TEMP RUN 2 3 33 NA NA Temperature run
CuNi#14 TEMP RUN
same as CuNi#10
2 3 33 NA NA Temperature run
CuNi#15 TEMP RUN
same as CuNi#13
2 3 33 NA NA Temperature run
CuNi#16 TEMP RUN
same as CuNi#11
2 3 33 NA NA Temperature run
CuNi#17 2 3 33 NA NA Experiment failed
CuNi#18 2 3 33 NA NA Same as CuNi#11
CuNi#19 2 3 33 16 12 Same as CuNi#11
CuNi#20 2 3 33 17 12 XRD
CuNi#21 2 3 33 16 12 XRD
CuNi#22 2 3 33 19 14 XRD, FIB, TEM
CuNi#23 2 3 33 NA NA Experiment failed
CuNi#24 2 3 33 NA NA Experiment failed
CuNi#25 2 3 33 20 14 XRD, FIB
CuNi#26 TEMP RUN 2 3 33 NA NA Temperature run
CuNi#27 TEMP RUN 2 3 33 NA NA Temperature run
CuNi#28 TEMP RUN 2 3 33 NA NA Temperature run
CuNi#29 TEMP RUN 2 3 33 NA NA Temperature run

Table 13. Cu-Zn alloy sputtered samples
Sample name
Working
pressure
(mTorr)
Working
distance
(inch)
Power
density
(w/cm^2)
Thickness
(μm)
Deposition
rate (nm/s)
Characterization
performed or
comments
CuZn#1-feb-15 2 2 25 9 5 XRD
CuZn#2-feb-15 2.5 2 17 17 7 XRD
CuZn#3-feb-15 2.5 2 17 NA NA XRD
CuZn#4-feb-15 2.5 2 17 11 9 XRD
CuZn#5-feb-15 2.5 2 17 10 9 XRD
CuZn#6-feb-15 2.5 2 17 10 8 XRD, FIB
CuZn#7-feb-15 2.5 2 17 NA NA Temperature run
CuZn#8-feb-15 2.5 2 17 NA NA Temperature run
CuZn#9-feb-15 2.5 4 30 5 5 XRD, FIB
CuZn#10-feb-15 2.5 4 30 5 5 XRD, FIB
CuZn#11-feb-15 2.5 4 30 6 5 XRD, FIB
CuZn#12-feb-15 2.5 4 30 6 5 XRD, FIB, TEM
CuZn#13-feb-15 2.5 4 30 5 4 XRD, FIB, TEM
CuZn#14 2.5 3 33 22 16 XRD, FIB
CuZn#15 2.5 3 33 22 16 XRD, FIB

105

Table 14. Cu-Ag alloy sputtered samples
Sample name
Working
pressure
(mTorr)
Working
distance
(inch)
Power
density
(w/cm^2)
Thickness
(μm)
Deposition
rate (nm/s)
Characterization
performed or
comments
CuAg#1-feb-15 2 3 33 NA NA Experiment failed
CuAg#2-feb-15 2 3 33 22 15 XRD
CuAg#3-feb-15 2 3 33 17 16 XRD, FIB
CuAg#4-feb-15 2 3 1.4 17 0.9 XRD
CuAg#5-feb-15 2 3 1.4 15 0.8 XRD, FIB
CuAg#6-feb-15 2 3 7.6 13 4 XRD, FIB
CuAg#7-feb-15 2 3 3.8 15 2 XRD, FIB

Table 15. Al sputtered samples
Sample name
Working
pressure
(mTorr)
Working
distance
(inch)
Power
density
(w/cm^2)
Thickness
(μm)
Deposition
rate (nm/s)
Characterization
performed or
comments
Al # 1 jan-12 2 6 12 NA NA Experiment failed
Al # 2 jan-12 5 6 23 0.5 0.3 Profilometer
Al # 3 jan-12 10 6 23 0.3 0.2 Profilometer
Al # 4 jan-12 5 6 23 2 0.2 XRD, FIB
Al # 5 jan-12 10 6 23 1 0.1 XRD
Al # 6 jun-12 2 6 10 3 0.4 XRD, FIB
Al # 7 jun-12 5 6 10 1 0.2 XRD, FIB
Al # 8 jun-12 10 6 10 2 0.2 XRD, FIB
Al # 9 Aug-12 1 6 10 2 0.4 XRD
Al#10 4 4 1.4 0.050 0.2 Coating structures
Al#11 4 4 1.4 0.130 0.1 Coating structures
Al#12 4 4 1.4 0.100 0.1 Coating structures
Al#13 4 4 1.4 NA NA Coating structures
Al#14 4 4 1.4 NA NA Coating structures
Al#15 2 2.5 33 12 12 XRD, TEM
Al#16 2 2.5 33 11 11 Extra sample
Al#17 2 3 44 7 12 XRD, TEM

Table 16. Al-Cu alloy sputtered samples

Sample name
Working
pressure
(mTorr)
Working
distance
(inch)
Power
density
(w/cm^2)
Thickness
(μm)
Deposition
rate (nm/s)
Characterization
performed or
comments
AlCu#1-jul-12 2 6 10 2 0.3 XRD, FIB
AlCu#2-jul-12 5 6 10 2 0.2 XRD
AlCu#3-jul-12 10 6 10 2 0.1 XRD
AlCu#4-Aug-12 1.1 6 10 2 0.4 XRD
106

Table 17. Al-Ni alloy sputtered samples
Sample name
Working
pressure
(mTorr)
Working
distance
(inch)
Power
density
(w/cm^2)
Thickness
(μm)
Deposition
rate (nm/s)
Characterization
performed or
comments
Al-2Ni#1 2 2.5 33 12 12 TEM
Al-2Ni#2 2 2.5 33 12 11 Extra sample

Table 18. Al2024 alloy sputtered samples
Sample name
Working
pressure
(mTorr)
Working
distance
(inch)
Power
density
(w/cm^2)
Thickness
(μm)
Deposition
rate (nm/s)
Characterization
performed or
comments
Al2024#1-oct-12 2.1 3 2.4 3 0.4 XRD
Al2024#2-jun-13 1.3 3 3.2 7 0.7 XRD, FIB, TEM
Al2024#3-jun-13 1.5 3 3.2 NA NA XRD
Al2024#4-jun-13 1.3 3 3.2 7 0.7 XRD
Al2024#5-aug-13 0.93 3 3.2 16 0.5 XRD
Al2024#6-aug-13 1.5 3 3.2 20 0.5 XRD
Al2024#7-aug-13 1.2 3 5 19 0.5 XRD
Al2024#8-aug-13 1.2 3 10 5 2 XRD
Al2024#9-aug-13 0.51 3 17 NA NA Experiment failed
Al2024#10-aug-13 1.1 3 20 5 3 XRD

Table 19. Al6013 alloy sputtered sample
Sample name
Working
pressure
(mTorr)
Working
distance
(inch)
Power
density
(w/cm^2)
Thickness
(μm)
Deposition
rate (nm/s)
Characterization
performed or
comments
Al6013#1-jun-13 2 3 3.2 6 0.6 XRD

Table 20. Al5052 alloy sputtered samples
Sample name
Working
pressure
(mTorr)
Working
distance
(inch)
Power
density
(w/cm^2)
Thickness
(μm)
Deposition
rate (nm/s)
Characterization
performed or
comments
Al5052#1-jun-13 2 3 3.2 6 0.6 XRD
Al5052#2-sep-13 2 3 3.2 15 0.4 XRD
Al5052#3-sep-13 2 3 3.2 20 0.5 Extra sample

107

Table 21. Al5083 alloy sputtered samples
Sample name
Working
pressure
(mTorr)
Working
distance
(inch)
Power
density
(w/cm^2)
Thickness
(μm)
Deposition
rate (nm/s)
Characterization
performed or
comments
Al5083#1 20 3 44 7 11 XRD, FIB
Al5083#2 20 3 44 7 12 XRD
Al5083#3 2 3 44 8 13 XRD
Al5083#4 2 2.5 33 12 12 XRD
Al5083#5 2 2.5 33 12 12 Extra sample

Table 22. Al5456 alloy sputtered samples
Sample name
Working
pressure
(mTorr)
Working
distance
(inch)
Power
density
(w/cm^2)
Thickness
(μm)
Deposition
rate (nm/s)
Characterization
performed or
comments
Al5456#1-jun-13 2 3 3.2 6 0.6 XRD
Al5456#2-sep-13 2 3 3.2 17 0.5 XRD
Al5456#3-sep-13 2.8 3 3.2 17 0.4 XRD, TEM
Al5456#4a-nov-13 2.5 3 3.2 19 0.5 XRD
Al5456#4b-nov-13 2.5 3 3.2 17 0.4 XRD, TEM
Al5456#5a-nov-13 2.5 3 3.2 NA NA XRD
Al5456#5-nov-13 2.5 3 3.2 19 0.5 XRD
Al5456#6-nov-13 2.5 3 3.2 18 0.5 XRD
Al5456#7-nov-13 2.7 3 3.2 18 0.6 XRD
Al5456#8-nov-13 2.7 3 3.2 15 0.5 XRD
Al5456#9-feb-14 3 3 25 6 3 XRD, FIB
Al5456#10-feb-14 2 3 49 NA NA Experiment failed
Al5456#11 2 2.5 33 11 11 Extra sample
Al5456#12 2 2.5 33 11 11 Extra sample
Al5456#13 2 2.5 33 NA NA Extra sample

Table 23. Al-5.3wt.%Mg alloy sputtered samples
Sample name
Working
pressure
(mTorr)
Working
distance
(inch)
Power
density
(w/cm^2)
Thickness
(μm)
Deposition
rate (nm/s)
Characterization
performed or
comments
Al-5.3Mg#1-feb-14 2 3 11 6 4 XRD, FIB
Al-5.3Mg#2-feb-14 2 3 22 6 6 XRD
Al-5.3Mg#3-feb-14 2 3 33 8 8 XRD
Al-5.3Mg#4-feb-14 2 3 44 6 10 XRD, FIB, TEM
Al-5.3Mg#5-apr-14 2 3 44 5 8 XRD, FIB, TEM
Al-5.3Mg#6-apr-14 2 3 44 5 8 XRD, FIB, TEM
Al-5.3Mg#7-may-14 2 3 44 4 6 XRD, FIB
Al-5.3Mg#8-apr-15 2 3 44 9 16 XRD
Al-5.3Mg#9-apr-15 2 3 44 10 16 XRD, FIB, TEM
Al-5.3Mg#10-apr-15 2 3 44 9 16 XRD
Al-5.3Mg#11-apr-15 2 3 44 6 11 XRD
Al-5.3Mg#12-apr-15 2 3 44 7 12 XRD
108

Al-5.3Mg#13-apr-15 2 3 44 7 12 XRD
Al-5.3Mg#14-apr-15 2 3 44 7 12 XRD, TEM
Al-5.3Mg#15-apr-15 2 3 44 10 17 XRD
Al-5.3Mg#16-apr-15 2 3 44 10 16 XRD
Al-5.3Mg#17-may-15 2 3 33 13 9 XRD, TEM
Al-5.3Mg#18-may-15 2 3 33 11 8 XRD, FIB, TEM
Al-5.3Mg#19-may-15 2 3 33 13 9 XRD, FIB, TEM
Al-5.3Mg#20-may-15 2 3 33 12 9 XRD
Al-5.3Mg#21-may-15 2 3 33 12 9 XRD
Al-5.3Mg#22-may-15 2 3 33 13 9 XRD, TEM
Al-5.3Mg#23-may-15 2 3 44 7 12 XRD, TEM
Al-5.3Mg#24-TEMP
RUN
2 3 33 NA NA Experiment failed
Al-5.3Mg#25 2 2.5 33 7 11 XRD
Al-5.3Mg#26 2 2.5 33 7 11 XRD
Al-5.3Mg#27 2 2.5 33 13 11 XRD
Al-5.3Mg#28 2 2.5 33 12 10 XRD
Al-5.3Mg#29 2 2.5 33 12 10 XRD
Al-5.3Mg#30 2 2.5 33 12 10 XRD, TEM
Al-5.3Mg#31 2 2.5 33 13 11 XRD, TEM
Al-5.3Mg#32 2 2.5 33 12 10 XRD, TEM
Al-5.3Mg#33 2 2.5 33 12 10 XRD, TEM
Al-5.3Mg#34 2 3 44 7 11 XRD
Al-5.3Mg#35 2 3 44 7 12 XRD
Al-5.3Mg#36 2 3 44 7 11 XRD
Al-5.3Mg#37 2 3 33 NA NA XRD
Al-5.3Mg#37,
repeated
2 3 33 4651 8 XRD
Al-5.3Mg#38 2 3 33 9 7 XRD
Al-5.3Mg#39 2 2.5 33 11 11 XRD, FIB, TEM
Al-5.3Mg#40 2 2.5 33 11 11 XRD, FIB, TEM
Al-5.3Mg#41 2 2.5 33 11 11 XRD, FIB, TEM
Al-5.3Mg#42 2 2.5 33 NA NA TEM
Al-5.3Mg#43 2 2.5 33 9 9 TEM
Al-5.3Mg#44 2 2.5 33 9 9 TEM
Al-5.3Mg#45 2 2.5 33 NA NA Extra sample
Al-5.3Mg#46 2 2.5 33 NA NA TEM
Al-5.3Mg#47 2 2.5 33 12 12 TEM
Al-5.3Mg#48 2 2.5 33 11 11 XRD
Al-5.3Mg#49 2 2.5 33 12 11 XRD

109

Table 24. Ni sputtered samples
Sample name
Working
pressure
(mTorr)
Working
distance
(inch)
Power
density
(w/cm^2)
Thickness
(μm)
Deposition
rate (nm/s)
Characterization
performed or
comments
Ni#1 2 3 33 13 7 XRD, FIB, TEM
Ni#2 2 3 33 14 8 XRD
Ni#3 2 3 33 13 7 XRD, FIB
Ni#4 2 3 33 14 8 XRD
Ni#5 2 3 33 12 7 XRD
Ni#6 2 3 33 17 9 XRD
Ni#7 2 3 33 17 10 XRD
Ni#8 2 3 33 16 9 XRD
Ni#9 2 3 33 16 9 XRD, FIB
Ni#10 2 3 33 NA NA XRD, FIB
Ni#11 2 3 33 NA NA FIB
Ni#12 2 3 33 NA NA FIB
Ni#13 2 3 33 15 9 FIB
Ni#14 2 3 33 NA NA Extra sample
Ni#15 2 3 33 NA NA Extra sample
Ni#16 2 3 33 14 8 XRD, FIB, TEM
Ni#17 2 3 33 13 7 FIB

Table 25. Inconel600 sputtered samples
Sample name
Working
pressure
(mTorr)
Working
distance
(inch)
Power
density
(w/cm^2)
Thickness
(μm)
Deposition
rate (nm/s)
Characterization
performed or
comments
inconel600#1 2 3 33 7 7 XRD, FIB
inconel600#2 2 3 33 28 8 XRD
inconel600#3 2 3 33 28 8 XRD
inconel600#4 2 3 33 NA NA XRD
inconel600#5 2 3 33 NA NA Experiment failed
inconel600#6 2 3 33 NA NA XRD

110

Appendix B. Complementary XRD patterns for the Cu, Cu alloys,
Al, Al alloys, Ni, and Inconel600.
This appendix contains XRD patterns from some of the samples displayed in Appendix A.
Notice that samples presented strong and random textures.

Figure 41. Normalized XRD patterns for Cu and Cu-alloys.

111

Figure 42. Normalized XRD patterns for Al and Al-alloys.

Figure 43. Normalized XRD patterns for Ni and inconel600.

112

Appendix C. Complementary microstructural FIB images for Cu, Cu
alloys, Ni, and Inconel600.
This appendix contains cross-sectional FIB images from some of the samples displayed in
Appendix A. FIB was used to perform a preliminary check to confirm the presence of twin
boundaries.

Figure 44. Cross-sectional FIB images showing nt structures in Cu. a)Cu#13, b)Cu#15. Notice the change in
contrast characteristic of twin boundaries, and the difference in the microstructural features between the two
images.
113

Figure 45. Cross-sectional FIB images showing nt structures with columnar grains in Cu alloys.a)
CuAl#69,b)CuAl#72, c)CuNi#8, d)CuNi#22. In e) and f) CuZn#14 and CuZn#15, respectively, notice the columnar
grains are not as sharp as in the other Cu-alloys. NOTICE, In g)CuAg#7 no presence of twin boundaries is
observable with the FIB.

114

Figure 46 Cross-sectional FIB images showing nt structures with columnar grains in Ni and
inconel600:a)Ni#1,b)Ni#16, c)inconel600#1.Notice that in Ni#1 and Ni#16 the grain size is much smaller than in
Ni#16.

115

Appendix D. Complementary microstructural TEM images for the Cu-Zn alloy.
This appendix contains cross-sectional transmission electron microscopy (TEM) images from
some of the Cu-Zn samples displayed in Appendix A. The TEM images show TBs that are
horizontal and inclined with respect to the film growth direction.

Figure 47. Horizontal and inclined TBs. a) CuZn#12, b) CuZn#13. The white arrow indicates the film growth
direction.

116

Appendix E. Complementary microstructural TEM images. Tailoring λ
Complementary representative TEM images for the Cu alloys presented in Table 7 are shown in
Figure 48 and Figure 49. λ variation is observable for the different sputtering conditions and
different Cu alloys for the study presented in section 6.

Figure 48. Representative cross-sectional TEM images of the twinned structures within a Cu-2wt.%Al alloy
obtained at different sputtering conditions. (a) sample 2Al#2, λ≈6 nm; (b) sample 2Al#3, λ≈8 nm; and (c) sample
2Al#5, λ≈10 nm. The growth direction is vertical for all the images.
117

Figure 49. Representative cross-sectional TEM images of the twinned structures in Cu-4wt.%Al, Cu-6wt%Al, and
Cu-10wt.%Ni alloys obtained at different sputtering conditions. (a) sample 4Al#1, λ≈2.5 nm; (b) sample 4Al#3,
λ≈5.1 nm; (c) sample 4Al#4, λ≈10 nm; (d) sample 6Al#1, λ≈2.1 nm; (e) sample 6Al#3, λ≈4.8 nm; (f) sample 6Al#4,
λ≈7.4 nm; (g) sample 6Al#5, λ≈8 nm; (h) sample 6Al#6, λ≈13 nm; (i) sample 10Ni#2, λ≈17 nm; and (j) sample
10Ni#3, λ≈20 nm; The growth direction is vertical for all the images.

118

Evidence of the interplay of the mechanisms proposed to vary λ in sections 6.2.2 and 6.2.3 can
be presented in terms of the density of remaining ITBs. In Figure 50 representative cross-
sectional TEM images of the samples that belong to zone II are presented. In the TEM images
Figure 50a thru Figure 50f, ITBs are marked by the yellow arrow, these ITBs were only found in
a few grains in the samples. ITBs were not observed in samples that belong to zone III.

Figure 50. Representative cross-sectional TEM images of the samples that presented high concentration of ITBs in a
few grains. (a) sample 2Al#5; (b) sample 4Al#5; (c) sample 6Al#7; (d) sample 10Ni#1; (e) sample 10Ni#2; and (f)
sample 10Ni#3. The arrows indicate the location of the ITBs. The growth direction is vertical for all the images.

119

Appendix F. Complementary microstructural TEM images for the Al alloys

Figure 51. Representative cross-sectional TEM images of columnar grains with TBs of: a) Al2024#2, and b)
Al5456#4. TBs are marked by the red dotted lines.

120

Appendix G. Complementary microstructural images obtained by using
conventional EBSD for Al and Al-5.3wt.%Mg

Figure 52. Conventional EBSD scans from the cross-section of: a) Al#17, b)Al-5.3Mg#14. Notice that both samples
have strong {111} texture mixed with {115} texture. Some Σ3 CTBs are highlighted by the thick red lines.

Abstract (if available)

Abstract Nanotwinned (nt) metals are materials that consist of several Σ3 twin boundaries (TBs) located within multiple grains in the overall microstructure, where the distance between any two consecutive TBs is within the range of 1 nm to 100 nm. In general, nt metals have attractive mechanical, thermal, electrical and/or chemical properties. The aforementioned properties of nt metals can be engineered based on the modification of two nanostructural features: (i) the density of TBs in the material, and (ii) the average distance between TBs or the twin thickness. These favorable and tunable properties of nt metals have encouraged extensive research during the last decade, specifically on fabrication methods that control nanostructural features. Nevertheless, the synthesis of nt metals has been limited due to stacking fault energy (SFE) restrictions, where many studies focus only on low SFE materials (< 45 mJ/m²) that are prone to twinning. To date, no studies have shown explicit control of the nanostructural features in nt metals. This dissertation discusses two main points: (i) the synthesis of nt metals with low, intermediate, and high SFEs by using magnetron sputtering

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Structure and behavior of nano metallic multilayers under thermal and mechanical loading

PDF

Synthesis and mechanical behavior of highly nanotwinned metals

PDF

Understanding the formation and evolution of boundaries and interfaces in nanostructured metallic alloys

PDF

Shock wave response of in situ iron-based metallic glass matrix composites

PDF

Development and characterization of hierarchical cellular structures

PDF

Tailoring compositional and microstructural complexity in nanostructured alloys

PDF

Mechanical behavior and deformation of nanotwinned metallic alloys

PDF

Development and characterization of transparent metal/ceramic and ceramic/ceramic nanomultilayers

PDF

Expanding the synthesis space of 3D nano- and micro-architected lattice materials

PDF

The role of nanotwins and grain boundary plane in the thermal, corrosion, and sensitization behavior of nanometals

$X-ray microbeam diffraction measurements of long range internal stresses in equal channel angular pressed aluminum; & Mechanical behavior of an Fe-based bulk metallic glass$

PDF

X-ray microbeam diffraction measurements of long range internal stresses in equal channel angular pressed aluminum; & Mechanical behavior of an Fe-based bulk metallic glass

PDF

Exploring the thermal evolution of nanomaterials: from nanometallic multilayers to nanostructures

PDF

Mechanical behavior of materials in extreme conditions: a focus on creep plasticity

PDF

Proton kinetics in electrochemistry: new directions and mechanistic analysis

PDF

A comparative study of plasma conditions, microstructure and residual stress in sputtered thin films

PDF

Printed electronics based on carbon nanotubes and two-dimensional transition metal dichalcogenides

PDF

Perovskite chalcogenides: emerging semiconductors for visible to infrared optoelectronics

PDF

Single-wall carbon nanotubes separation and their device study

$Using X-ray microbeam diffraction to study the long range internal stresses in plastically deformed materials$

PDF

Using X-ray microbeam diffraction to study the long range internal stresses in plastically deformed materials

PDF

Transport studies of phase transitions in a quasi-1D hexagonal chalcogenide

Asset Metadata

Creator Velasco Estrada, Leonardo (author)

Core Title A comprehensive study of twinning phenomena in low and high stacking fault energy metals

School Viterbi School of Engineering

Degree Doctor of Philosophy

Degree Program Materials Science

Publication Date 07/06/2016

Defense Date 05/23/2016

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag nanostructured materials,OAI-PMH Harvest,sputtering,stacking fault energy,twinning

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Hodge, Andrea M. (committee chair), Eliasson, Veronica (committee member), Kassner, Michael (committee member), Ravichandran, Jayakanth (committee member)

Creator Email leoveles@gmail.com,velascoe@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c40-263658

Unique identifier UC11280506

Identifier etd-VelascoEst-4516.pdf (filename),usctheses-c40-263658 (legacy record id)

Legacy Identifier etd-VelascoEst-4516.pdf

Dmrecord 263658

Document Type Dissertation

Format application/pdf (imt)

Rights Velasco Estrada, Leonardo

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA