University of Southern California Dissertations and Theses

Corrosion studies of engineered interfaces

(USC Thesis Other)

Corrosion studies of engineered interfaces

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content CORROSION STUDIES OF ENGINEERED INTERFACES
by
Karina Hemmendinger
A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(MECHANICAL ENGINEERING)
May 2024
Copyright 2024 Karina Hemmendinger

ii
Dedication
For Ingfrid Haugland and Helene Hemmendinger, with love.

iii
Acknowledgements
There are a lot of people without which this would not have been possible. This is by no
means an exhaustive list, but there are some people that I would like to acknowledge:
I would like to thank my advisor, Prof. Andrea Hodge for the opportunity to do this work.
Thank you for believing in me, and pushing me to grow in all aspects.
Thank you to my committee, Prof. Mitul Luhar, Prof. Paolo Branicio, and Prof. Jahan
Dawlaty for your guidance. I would also like to thank my mentors, Prof. Ananya Renuka
Balakrishna, Prof. Lessa Grunenfelder, Prof. Paul Ronney, and Dr. Zachary Ruff: thank you for
your support, your counsel, and for inspiring me.
To my many USC colleagues, especially Edwin Williams, Dr. Emma Singer, Dr. Saakar
Byahut: it has been a pleasure to work with you and learn from you.
The Office of Naval Research funded the first part of work under the auspices of ONR
grant N00014-18-1-2617. The second phase of this work was partially funded by the National
Science Foundation, through NSF grant OISE-2106597.
Thank you to the members of the Hodge Materials Research Group, both past and
present: Dr. Sebastian Riaño, Dr. Angelica Saenz-Treviso, Dr. Chelsea Appleget, Dr. Joel
Bahena, Dr. Alina Garcia Taormina, Dr. Roya Ermagan, Dr. Daniel Goodelman, Adie Alwen,
Danielle White, Kyle Russell, TJ Oros, Andre Bohn, Ikponmwosa Iyinbor and Ashley
Maldonado.
To Dr. Paul Jaffe: thank you for your enthusiasm and your mentorship.
To Chris, Lucia, Kylie, and Marissa: thank you for all of your help and support, and for
making me laugh. I would not be where I am today without you.
To James, and my family: thank you for your unwavering support, your encouragement,
your faith in me, and the continuous stream of absolutely terrible science puns. I love you.

iv
Table of Contents
Dedication....................................................................................................................................... ii
Acknowledgements........................................................................................................................iii
List of Tables ................................................................................................................................. vi
List of Figures............................................................................................................................... vii
List of Abbreviations .................................................................................................................... xii
Abstract........................................................................................................................................ xiv
Chapter 1. Introduction ................................................................................................................... 1
Chapter 2. Background ................................................................................................................... 5
2.1 Interfaces in Nanostructured Materials............................................................................................... 5
2.1.1 Grains and Grain Boundaries...................................................................................................... 5
2.2 Corrosion Behavior........................................................................................................................... 11
2.2.1 Corrosion Along Grain Boundaries........................................................................................... 13
2.2.2 Dealloying ................................................................................................................................. 16
2.3 Cellular Materials ............................................................................................................................. 19
2.3.1 Structure of Cellular Materials.................................................................................................. 20
2.3.2 Nanoporous Foams.................................................................................................................... 21
2.4 Graded Microstructures.................................................................................................................... 25
2.4.1 Structural Gradients................................................................................................................... 26
2.5 Mechanical Properties of Nanoporous Foams.................................................................................. 30
Chapter 3. Experimental Methods ................................................................................................ 32
3.1 Magnetron Sputtering ....................................................................................................................... 32
3.2 Corrosion Methods ........................................................................................................................... 34
3.2.1 Dealloying ................................................................................................................................. 36
3.3 Characterization Techniques ............................................................................................................ 39
3.3.1 Scanning Electron Microscopy ................................................................................................. 39
3.3.2 Focused Ion Beam..................................................................................................................... 42
3.3.3 Energy Dispersive X-Ray Spectroscopy ................................................................................... 45
3.3.4 Electron Backscatter Diffraction ............................................................................................... 46
3.3.5 Surface-Enhanced Raman Spectroscopy................................................................................... 48
3.4.1 Vickers Hardness Testing.......................................................................................................... 51
Chapter 4. Corrosion Resistance in GB-Engineered 5xxx Series Al Alloys................................ 53
4.1 Sensitization and Corrosion Resistance............................................................................................ 53
4.2 Characterization of the Grain Boundaries and Preliminary Corrosion Analysis.............................. 56
4.3 The Effects of Special Grain Boundary Fraction and Sensitization on Corrosion Resistance......... 58
4.4 Conclusions....................................................................................................................................... 62
Chapter 5. Progression of the dealloying front in bilayer Cu-Al and Cu-Zn nanoporous foams. 64
5.1 Hierarchical and Graded Nanoporous Metallic Materials Through Dealloying .............................. 65
5.2 Analysis of the Dealloying Front...................................................................................................... 68
5.2.1 Characterization of the As-Sputtered Films.............................................................................. 68
5.2.2 The Dealloying Front ................................................................................................................ 70
5.2.3 Foam Morphology..................................................................................................................... 77
5.3 Conclusions....................................................................................................................................... 80
Chapter 6. SERS Response of Gradient Nanoporous Cu ............................................................. 82

v
6.1 Graded Nanoporous Metallic Foams For Sensing............................................................................ 82
6.2 Results and Discussion ..................................................................................................................... 86
6.2.1 Synthesis of Gradient Nanoporous Copper Via Dealloying ..................................................... 86
6.2.2 SERS Response of Gradient Nanoporous Cu............................................................................ 92
6.3 Conclusions....................................................................................................................................... 97
Chapter 7. Conclusions and Future Work..................................................................................... 99
7.1 Conclusions....................................................................................................................................... 99
7.2 Future Work.................................................................................................................................... 101
References................................................................................................................................... 103
Appendices.................................................................................................................................. 116
Appendix A: Summary of Sputtered Samples...................................................................................... 116
Appendix B: Summary of Dealloyed Samples..................................................................................... 118

vi
List of Tables
Table 1: CSL Fractions and Average Grain Size Measured by EBSD......................................... 58
Table 2: Sample Composition, Ligament Size, and Relative Density.......................................... 71
Table 3: Summary of As-Sputtered and Dealloyed Samples........................................................ 85
Table 4: Sample Compositions Before and After Dealloying ...................................................... 88
Table 5: Ligament and Pore Size After Dealloying...................................................................... 91
Table 6: SERS Enhancement Factor............................................................................................. 93
Table 7: Summary of Bilayer Sputtered Samples Sputtered Without Additional Heating......... 116
Table 8: Summary of Bilayer Sputtered Samples Sputtered With Additional Substrate
Heating........................................................................................................................................ 116
Table 9: Summary of Bilayer Cu Alloy Sample Names for Dealloying .................................... 118
Table 10: Summary of Dealloyed Bilayer Cu Samples From Unheated Films.......................... 119
Table 11: Summary of Dealloyed Bilayer Cu Samples From Heat-Treated Films.................... 121
Table 12: Summary of Dealloyed Bilayer Cu Samples From Heated Stage Films.................... 121

vii
List of Figures
Figure 1: Coincident-site lattice map of two grains (blue atoms and black atoms)
demonstrating the geometric fit of the neighboring grain [10]....................................................... 6
Figure 2: Plot of GB energy as a function of misorientation angle in degrees for a variety of
sigma boundaries. Note that there is only a loose correlation between misorientation angle and
boundary energy, and that the boundaries with the same sigma value may possess different
energies [15].................................................................................................................................... 8
Figure 3: Grain orientation maps (a-c) and grain boundaries (d-f) for Ni-based nanostructured
alloys processed via ball-milling, electrodeposition, and magnetron sputtering. In (d-f), Σ1
boundaries are shown in blue, and Σ3 boundaries are shown in red. All other boundaries are
shown in black [17]....................................................................................................................... 10
Figure 4: Summary of the eight forms of corrosion: uniform corrosion, hydrogen damage,
galvanic corrosion, intergranular corrosion, crevice corrosion, dealloying, pitting, erosion
corrosion, and environmentally induced cracking [25]. ............................................................... 12
Figure 5: The sensitization process in a 5xxx series Al-Mg alloy, with Al (blue) and Mg
(green) atoms. From left to right: the alloy as received, with Mg atoms dispersed in the alloy;
the alloy after exposure to heat, with the formation of Mg beta-phase precipitates along the
grain boundary; the alloy after exposure to a corrosive environment, with intergranular
corrosion along the boundaries. .................................................................................................... 14
Figure 6: Continuous ß-precipitates along the grain boundary in 5xxx series Al-Mg [34].......... 15
Figure 7: Electrochemical reduction potentials with associated half-reactions [37]. ................... 16
Figure 8: Dealloying in the Au/Ag alloy system, denoted by orange and grey, respectively.
Vacancies begin to form in (a) as Ag atoms diffuse from the sample surface. The Au atoms
reorganize in (b) to form a passive layer that protects the more accessible Ag atoms. Once
the exposed surface area of Ag atoms is too large to passivate (c, d), new areas are exposed
and ligaments begin to branch [24]............................................................................................... 17
Figure 9: Representative schematic of an electrochemical cell setup for potentiostatic
dealloying [41].............................................................................................................................. 19
Figure 10: 3 nanoporous copper foams of varying relative densities. (a) is a 20% relative
density foam, dealloyed via free corrosion in 85 wt.% H3PO4; (b) is a 30% relative density
foam, dealloyed via free corrosion in 5 wt.% HCl; and (c) is a 35% relative density foam
dealloyed electrochemically in 5 wt.% HCl [46].......................................................................... 20
Figure 11: SEM micrographs of the Cu25Al75/Cu40Al60 (Cu40Al60 deposited first) sputtered
bilayer film from Hu et al. As-sputtered (a) top surface and (b) cross-section; after dealloying

viii
by free corrosion in NaOH, (c) is the top surface and (d) is the cross-section. Note that the
interface between the two dealloyed layers is visible in (d) [59]. ................................................ 23
Figure 12: SEM micrographs of the Cu25Al75/Cu40Al60 (Cu25Al75 deposited first) sputtered
bilayer film from Hu et al. As-sputtered (a) top surface and (b) cross-section; after dealloying
by free corrosion in NaOH, (c) is the top surface and (d) is the cross-section. Note that the
interface between the two different porosities is visible in (d) [59]. ............................................ 24
Figure 13: Types of gradient microstructures. Top row, structural gradients, from left: grain
size, twin thickness, lamellar thickness, mix of grain size and twin thickness. Bottom row,
chemical gradients, from left: phase, solid-solution concentration, chemical compound, mix
of structural and chemical gradients [4]........................................................................................ 25
Figure 14: Templating fabrication method for porous structures. A sacrificial template is
coated via a deposition process, and the template is removed to produce a film [47].................. 28
Figure 15: From left to right: metal powders and space holders are mixed; the mixture is
pressed; the spacers are removed from the compressed mixture; the powder is heated to form
a solid structure [47]. .................................................................................................................... 28
Figure 16: A multi-step dealloying process to produce a hierarchical porous structure. From
left to right: an Ag-Au precursor alloy; the alloy undergoes selective corrosion to form a
porous structure; the structure is annealed to coarsen pores; the material undergoes selective
corrosion once more and exhibits another level of porous structure [47]..................................... 29
Figure 17: Cross-sectional SEM micrograph of hierarchical nanoporous gold,
electrochemically dealloyed through a multi-step dealloying process [67].................................. 29
Figure 18: Schematic of a single-source configuration in a DC magnetron sputtering chamber. 33
Figure 19: Schematic of the two planar sputtering configurations utilized in this study. a)
shows the single target (single composition) configuration that was used to produce the films
in Chapter 4, while b) shows the dual target configuration used to produce the gradient films
in Chapters 5 and 6. ...................................................................................................................... 34
Figure 20: Example of anodic polarization curves 0n 0.1M HClO4 of Ag, Au, Au20Ag80,
Au25Ag75, and Au30Ag70. Resulting critical potentials for Au20Ag80, Au25Ag75, and Au30Ag70
are: 0.8, 0.94 and 1.01 V [84]. ...................................................................................................... 36
Figure 21: The elctrochemical cell setup towards the end of the reaction for an Au30Ag70
precursor in 5M HNO3 at a voltage of 1100 mV.......................................................................... 38
Figure 22: Schematic of a scanning electron microscope. The beam is generated at the
source, and travels through a series of lenses before reaching the sample. Interaction between
the electron beam and the sample produces a variety of signals, which are picked up by
detectors such as the back-scattered electron detector, the secondary electron detector, and

ix
the x-ray detector [91]................................................................................................................... 40
Figure 23: Diagram of the signals produced by incident electrons during SEM for a thin
sample [92].................................................................................................................................... 41
Figure 24: Schematic of a standard SEM/FIB dual-beam system with ion and electron beams,
sample stage, assorted detectors, sample manipulator, and gas injection system for localized
deposition [95] .............................................................................................................................. 43
Figure 25: Milling process for polishing a sample cross-section, shown on corroded copper.
A Pt protective coating is deposited in step a), and the region next to the protective coating is
milled at a medium to high current in b) to expose the cross-section. c) and d) are shown after
milling at progressively lower currents to produce the polished cross-section shown in step d)
[96]................................................................................................................................................ 44
Figure 26: Example EDS composition map of four elements present in shungite rock [100] ..... 45
Figure 27: Conventional EBSD configuration (left) and transmission Kikuchi diffraction
(TKD) configuration (right).......................................................................................................... 46
Figure 28: Example TKD grain orientation map for Al-6060 showing a) TKD pattern quality
as a function of contrast, b) raw orientation data (inverse pole figure not shown) and c) the
combined contrast and cleaned orientation map [102]. ................................................................ 47
Figure 29: Example of the Stokes and anti-Stokes scattering phenomena for electromagnetic
radiation on a nanoparticle substrate with a target molecule [104]. ............................................. 48
Figure 30: Schematic of a common SERS instrument setup [104]. ............................................. 49
Figure 31: Schematic of the charge transfer mechanism between a SERS analyte and the
metallic substrate [109]................................................................................................................. 50
Figure 32: Schematic of the Vickers diamond pyramidal indenter tip, showing the angle
between faces (136˚), and the resulting indent [110].................................................................... 51
Figure 33: Optical micrographs of Vickers indents on optical ceramic multilayer films. From
left: a) AlN(47)/Ag(21), b) AlN(20)/Ag(20), and c) AlN(127)/Ag(10). Layer thickness in
nanometers of each component listed in parentheses [112].......................................................... 52
Figure 34: Cross-sectional TEM micrographs of as-sputtered Al–Mg samples, noting grain
widths, shown as a–c) brightfield and d–f) dark-field sputtered at: 1.0, 3.5, and 7 nm s-1
as
labeled above each column. Growth direction is indicated by the red arrow on the bright-field
micrographs. All samples present a Mg content of 4 to 5 wt%, comparable with that of an
Al5083 alloy as determined by EDS (not shown). ....................................................................... 56

x
Figure 35: Overview of EBSD and grain size of as-sputtered Al-Mg samples, a–c) top-surface
EBSD, d–f ) cross-sectional EBSD (TKD) where red boundaries are special boundaries,
whereas black are unindexed. h–j) The average grain size. Corresponding sputtering rates
noted on top of each column......................................................................................................... 57
Figure 36: Top-surface SEM micrographs post-sensitization and immersion corrosion in
artificial seawater for samples a) 1.0, b) 3.5, and c) 7 nm/s. Cracking is visible in (a) and (b),
denoted by blue arrows; (c) a passive oxide layer. ....................................................................... 61
Figure 37: Graphical abstract, depicting a) the as-sputtered film for configuration 2, and the
subsequent progression of the dealloying front in concentrated H3PO4 after b) 30 minutes,
c) 45 minutes and d) the “complete” time point. .......................................................................... 64
Figure 38: SEM micrographs of the fractured as-sputtered films of (a) configuration 1, (b)
configuration 2, and cut as-sputtered films of (c) configuration 3 and (d) configuration 4. The
orange labels in the top right corner indicate sample configuration. The black arrow on the
right marks the film growth direction. .......................................................................................... 69
Figure 39: SEM micrographs of Cu-Al bilayer films dealloyed in H3PO4 via free corrosion
and corresponding EDS maps for configuration 1 (Cu20Al80/Cu35Al65) and configuration 2
(Cu35Al65/Cu20Al80) at various time intervals. The red dotted line marks the position of the
dealloying front after 30 min (a-d), 45 min (e-h), and “final step” at 105 min (i and j), and
120 min (k and l). EDS map indicates both Al and Cu content.................................................... 74
Figure 40: SEM micrographs of Cu-Al/Cu-Zn bilayer films dealloyed in H3PO4 via free
corrosion and corresponding EDS maps for configuration 3 (Cu20Al80/Cu20Zn80) and
configuration 4 (Cu20Zn80/Cu20Al80) at various time intervals. The red dotted line marks the
position of the dealloying front after 30 min (a-d), 45 min (e-h), and “final step” at 60 min (i
and j). EDS map indicates Al, Cu and Zn content........................................................................ 76
Figure 41: Representative SEM micrographs of foam morphology for top surface (left
column) and cross-section (right column), taken at the final (“final step”) dealloying time
interval. Configuration 1 (a-b), configuration 2 (c-d), configuration 3 (e-f) and configuration
4 (g-h). The orange labels in the top left corner indicate sample configuration........................... 78
Figure 42: Representative SEM micrographs of top-surface Vickers indents performed at the
“final step” dealloying time interval (left column) where the blue boxes in the left column
denote the location of respective zoom-ins (right column) of the edge of the indent. The
orange labels in the top left corner indicate sample configuration as follows: configuration #1
(a-b), configuration #2 (c-d), configuration #3 (e-f) Configuration #4 not shown as sample
was too brittle for testing. ............................................................................................................. 80
Figure 43: SEM micrographs and corresponding EDS maps of as-sputtered Cu20Zn80/Cu20Al80
films (A1, B1) and as-sputtered Cu20Al80/Cu20Zn80 films (C1). A1 and C1 have stage
temperatures of 100˚C, while B1 has stage temperature of 150˚C. Color on EDS map
indicates Al, Cu, and Zn content. Black arrow signifies film growth direction. .......................... 87

xi
Figure 44: Cross-sectional EDS maps of A and B series Cu–Al/Cu–Zn bilayer films dealloyed
in H3PO4 via free corrosion. Top row for a half cycle of 2.5 hours (A2) or 4 hours (B2);
bottom row after a full dealloying cycle of 5 hours (A3) or 8 hours (B3). Left and right
columns correspond to sputtering stage temperature of 100˚C and 150˚C, respectively. Color
on EDS map indicates Al, Cu, and Zn content. Black arrow shows film growth direction. ........ 90
Figure 45: Surface-enhanced Raman spectra of (a) set A samples, (b) set B samples, for
substrate with 4-mercapto benzonitrile in ethanol. Red is the as-sputtered condition, green is
the “half-cycle” dealloy, and blue is the “full-cycle” dealloy. ..................................................... 92
Figure 46: Top-surface SEM micrographs of the (A1, B1) as-sputtered, (A2, B2) “half-cycle”
dealloyed, and (A2, B2) “full-cycle” dealloyed films. Left column and solid border is
for sputtering stage temperature of 100˚C (film set A), right column and dashed border is for
stage temperature of 150˚C (film set B). Red is as-sputtered, green is “half-cycle” dealloyed,
blue is “full-cycle” dealloyed........................................................................................................ 96

xii
List of Abbreviations
at.% Atomic percent
BSE Backscattered electrons
CSL Coincident site lattice
CH4 Methane
C2H4 Ethylene
CO2 Carbon dioxide
DC Direct Current
DI De-ionized
GB Grain boundary
GBE Grain boundary engineering
EBSD Electron backscatter diffraction
EDS Energy dispersive X-ray spectroscopy
EF Enhancement Factor
FCC Face-centered cubic
FIB Focused ion beam
HCl Hydrochloric acid
HCOOH Formic acid
H3PO4 Phosphoric acid
H2SO4 Sulfuric acid
IGC Intergranular corrosion
M Molar
NaOH Sodium hydroxide

xiii
NP Nanoporous
NPC Nanoporous copper
PFIB Plasma focused ion beam
SE Secondary electron
SEM Scanning electron microscopy
SERS Surface-enhanced Raman scattering
SFE Stacking fault energy
SMAT Surface mechanical attrition treatment
TEM Transmission electron microscopy
TKD Transmission Kikuchi diffraction
wt.% Weight percent

xiv
Abstract
Understanding nanoscale interfaces is critical for designing new materials to meet current
and future scientific needs in applications such as radiation shielding, sensing, and catalysis.
These interfaces, such as grain boundaries, comprise a high volume fraction of nanostructured
materials, and changing the type or quantity of interfaces can produce large variations in
properties such as strength or corrosion resistance. While there have been previous studies on the
corrosion response of nanoscale interfaces, much of the work to date has been focused on the
effects on mechanical behavior. Thus, there is a pressing need to explore other aspects of
corrosion behavior in nanoscale interfaces, specifically corrosion resistance and dealloying, in
order to be able to design specific microstructures with the desired corrosion response.
This dissertation investigates the relationship between engineered nanoscale interfaces
and corrosion in order to illuminate the relationship between processing and microstructure. This
was investigated with: (1) the synthesis and evaluation of grain boundary-engineered 5xxx series
Al-Mg films to improve corrosion resistance in seawater, (2) the synthesis of gradient
nanoporous Cu via a dealloying corrosion process and subsequent investigation of the effect of a
sharp relative density or chemical potential interface on the association between processing
protocols and foam morphology, and (3) the investigation of graded interfaces on nanoporous Cu
microstructure and the suitability of these materials for SERS-based sensing. By identifying the
key links between processing and microstructure for these interfaces, these studies provide
critical insight for understanding corrosion in complex material systems and how to leverage this
knowledge for future controlled and reliable synthesis of these materials.

1
Chapter 1. Introduction
Interfaces such as grain boundaries (GBs) are fundamental to understanding a material’s
mechanical behavior and functionality, particularly for nanostructured materials, where GBs
represent a high fraction of the total volume. The number of interfaces and their character can be
engineered to tailor the properties and functionality of the bulk material to a specific application.
These engineered interfaces can include grain boundaries, as well as sharp or graded interfaces
between two or more components that differ in crystallographic orientation, chemical
composition, pore size, or another defining feature of interest. Although there have been many
studies regarding the effect of boundaries and interfaces, they have mostly focused on improving
mechanical properties. In contrast, corrosion studies in this area have been limited to studies of
corrosion along boundaries, rather than leveraging engineered interfaces to minimize corrosion
or to study corrosion as a synthesis method.
Stress corrosion cracking and corrosion along GBs are significant problems in the field of
corrosion-resistant alloys. Metals and alloys with a high fraction of GBs with good geometric fit
between neighboring grains, called special GBs, have been shown to demonstrate improved
material properties including increased corrosion resistance along GBs. The process of grain
boundary engineering (GBE) produces designed microstructures to improve the bulk properties
of a material, and can change the number of grain boundaries and their character. Traditionally,
GBE has been done through heat treatments or plastic deformation. Recent magnetron sputtering
studies have shown the ability to increase the fractions of special GBs significantly, thus
providing a path to improve corrosion resistance. As a novel expansion of this, Al-Mg alloys
were sputtered with interrupted deposition at three fast sputtering rates (1 nm/s, 3.5 nm/s and 7

2
nm/s) in order to determine both the quantity of special grain boundaries and the resulting
corrosion behavior in artificial seawater. The high deposition temperatures and resulting strains
increased the fraction of low-Σ grain boundaries, a type of special grain boundary, which
improved top-surface corrosion resistance.
However, corrosion is multifaceted. Dealloying is a corrosion process based on the
selective corrosion of a metallic precursor, that allows for flexible fabrication of porous
materials. Specific length scales are engineered by balancing the corrosion rate of the less noble
atoms with the reorientation rate of the more noble atoms, via the reaction parameters and
precursor composition. The resulting nanoporous materials have unique properties such as high
surface area, can be manufactured in a variety of geometries, and more recently, can be produced
with gradient pore sizes. This capability is appealing, as the customizability of gradient
structures provides opportunities for novel combinations of chemical and mechanical properties,
although further study is needed to establish the stability and properties of the microstructure at
the interface.
In order to investigate this interface, two distinct case studies with a sharp interface were
conducted. The first, the relative density gradient case, was analyzed with two different Cu-Al
alloys with different atomic percentages of Cu, while the chemical potential case was examined
with a Cu-Al/Cu-Zn interface of the same atomic percentage of Cu. Bilayer films of the
corresponding alloys were dealloyed in concentrated H3PO4 for a set of time intervals to study
the progression of the dealloying reaction front during synthesis. A slower reaction time was
found to correspond to a more uniform foam morphology, while a low (20 at.%) atomic fraction
of Cu yielded nanoporous pillars separated by large voids.

3
This approach to synthesizing gradient nanoporous Cu was then expanded to investigate
the synthesis from a gradient precursor film and the suitability of the resulting gradient
nanoporous material for sensing applications. The compositions from the chemical potential case
were sputtered with additional heating at 100˚C or 150˚C to reduce grain boundary voids and
produce a compositional gradient between layers. After dealloying, the foam morphology and
layer configuration were examined. Utility as a SERS substrate was evaluated with a thiol-based
probe, and the as-sputtered and dealloyed films displayed enhancement factors between 104 and
107
. The SERS enhancement factor (EF) is a measure of the substrate-induced signal
amplification, and all calculated values satisfied the 104 minimum for good enhancement. This
enhancement factor is highly dependent on the local foam morphology at the nanoscale, and its
ability to amplify the Raman signal from the chosen target molecule. The flexible processing of
these gradient nanoporous foams and the capability of selecting for a specific set or size of
nanostructures is key for developing future SERS substrates that can be optimized for sensing
multiple target molecules simultaneously.
This work will focus on synthesizing and evaluating nanoscale interfaces in order to
identify key processing parameters and their relationship to the resulting microstructure and
properties. It will apply well-established techniques such as dealloying, magnetron sputtering,
and energy dispersive X-ray spectroscopy (EDS) to produce and characterize these materials.
Additionally, initial work on the design and synthesis of more robust gradient nanoporous
materials for SERS-based sensing will be shown. Thus, the aim of this research is to investigate
the synthesis and properties of engineered corrosion-resistant and corrosion-fabricated interfaces.
These results will identify key relationships between processing variables and the resulting

4
microstructural features at the interface, to enable more controlled and consistent synthesis of
designed microstructures.

5
Chapter 2. Background
2.1 Interfaces in Nanostructured Materials
Interfaces such as GBs represent a high fraction of the total volume in nanostructured
materials, and are therefore essential to understanding and tailoring their properties [1].
Nanostructured materials, which are defined as having at least one structure or dimension on the
nanoscale, are an important area of current research due to their difference in properties when
compared to single-crystal or larger grained polycrystalline materials of the same composition
[1, 2]. The high fraction of interfaces inhibits dislocation motion during deformation, increasing
the yield strength [3]. The properties and functionality of the bulk material can be tailored by
changing the number and/or type of interfaces. These interfaces can include grain boundaries,
structural gradients and chemical gradients, as well as sharp or graded interfaces between two or
more components that differ in crystallographic orientation, chemical composition, pore size, or
another defining feature of interest [4]. Both the small grain size and the high concentration of
interfaces contribute to changes in deformation mode and properties, but the relationship
between specific grain boundary microstructures and the resulting grain boundaries and
boundary distributions is not understood.
2.1.1 Grains and Grain Boundaries
Grain boundaries (GBs) are defined by five degrees of freedom that form the grain
boundary plane (two degrees) and the misorientation angle (three degrees) [5]. This
misorientation at the grain boundaries may also be defined through Coincident Site Lattice
Theory, which provides an indicator of the geometric fit of neighboring grains [6]. If the atomic
lattices of two neighboring grains are superimposed, the points where the atoms from the two
lattices overlap are considered coincident sites. A higher number of shared coincident sites

6
between two neighboring grains signifies a better geometric fit [7, 8]. This is visible in Figure 1
below, where the atoms of two neighboring grains (blue and black) overlap in the CSL unit cell
box, and the coincident cites are marked in green. The number of coincident sites can be
correlated to a Σ value, where the lower Σ values generally have lower grain boundary energies
[9]. This sigma value is defined as the volume of the coincident site lattice unit cell divided by
the volume of the elementary unit cell, as shown in the following formula [5]:
= �� ℎ� ��
�� ℎ� ��
Figure 1: Coincident-site lattice map of two grains (blue atoms and black atoms) demonstrating
the geometric fit of the neighboring grain [10].
Special grain boundaries, a subcategory of Σ boundaries, are defined by the following equation,
Brandon’s Criterion:
Δ�!"# = 15˚/ Σ
!!
CSL
[100]
[010]
[010]
[1 00]

7
where ∆θmax describes the maximum possible deviation angle for a given coincident-site lattice, -
n is the numerical value 1⁄2, and Σ is defined as the ratio of the volumes of the unit cells [11, 12].
Subsequent criteria derived from the Brandon criterion can provide similar results, and
differ in their value of the exponent n [12]. GBs with sigma values less than or equal to 29 are
considered special GBs, and materials with these special boundaries have been shown to
demonstrate unusual and improved material properties (compared to non-“special” boundaries)
including high or low boundary mobility, decreased boundary segregation, and a decreased
propensity for stress-corrosion cracking and corrosion along grain boundaries [7, 12, 13]. Special
GBs have lower GB energies than the randomly oriented non-CSL boundaries [14]. However,
there is only a loose correlation between misorientation angle and boundary energy, as
boundaries with the same Σ value may not have the same energy. This is demonstrated in Figure
2, which shows a plot of grain boundary energy as a function of misorientation angle for a set of
Σ boundaries in Ni, Cu, and Al. [15]. The fraction of special boundaries in a material can be
tailored through a process called grain boundary engineering (GBE) [16].

8
Figure 2: Plot of GB energy as a function of misorientation angle in degrees for a variety of
sigma boundaries. Note that there is only a loose correlation between misorientation angle and
boundary energy, and that the boundaries with the same sigma value may possess different
energies [15].
2.1.1.1 Grain Boundary Engineering
Grain boundary engineering (GBE) allows for the tuning of both the types of grain
boundaries present in a material and the fraction of special grain boundaries, in order to improve
the properties of the material. GBE microstructures can be produced through several methods,
most of which utilize a deformation combined with an annealing process to promote the
formation of special GBs. Figure 3 shows an example of a GBE material. Sections (a), (b), and
(c) show grain orientation maps and grain boundaries for Ni-based nanostructured alloys
processed via ball milling, electrodeposition, and magnetron sputtering respectively. The special
grain boundaries for the ball milled, electrodeposited, and sputtered samples are superimposed
on the corresponding grain orientation maps in (d), (e), and (f). The more traditional ball milling

9
method produced some Σ1 boundaries (shown in blue) with a few twins (Σ3 boundaries, shown
in red). Sputtering and electrodeposition produced a high number of twins in similar distributions
because of their similar structural kinetics [17].
These results emphasize the link between the GBE processing method and the resulting
microstructure, and highlight the usefulness of deposition-based synthesis techniques such as
sputtering or electrodeposition for GBE microstructures [17-19]. These techniques have been
shown to synthesize materials with a high fraction of low-CSL boundaries with an emphasis on
Σ3 boundaries [14, 16, 20]. GBE has primarily been applied in low stacking fault energy (SFE)
materials, but has also been shown to be effective in high stacking fault energy materials such as
5xxx series aluminum-magnesium alloys [21]. A high fraction of Σ boundaries can increase
corrosion resistance, which is integral to the design and use of materials derived from the 5xxx
series Al-Mg alloys [22]. GBE in the context of sensitization is discussed in further detail in the
following section.

10
Figure 3: Grain orientation maps (a-c) and grain boundaries (d-f) for Ni-based nanostructured
alloys processed via ball-milling, electrodeposition, and magnetron sputtering. In (d-f), Σ1
boundaries are shown in blue, and Σ3 boundaries are shown in red. All other boundaries are
shown in black [17].

11
2.2 Corrosion Behavior
There are eight forms of corrosion in metals (Figure 4): uniform attack, stress corrosion
cracking, erosion-corrosion, selective leaching, intergranular corrosion, galvanic corrosion,
pitting, and crevice corrosion. These mechanisms are interrelated, but have characteristic
differences [23]. The corrosion mechanisms of interest here are stress corrosion, selective
leaching, intergranular corrosion, and galvanic corrosion.
In stress corrosion cracking, the combination of a corrosive environment and tensile
stress produce fine cracks in the metal, with little surface damage. The combination of alloy
composition, stress, corrosion environment, and temperature control whether a given alloy will
corrode [23]. Selective leaching, also called dealloying, is the selective dissolution of a
component (usually an element) from an alloy [23, 24]. This is covered in more detail in Section
2.2.2 (Dealloying) and Section 3.1.2 (Dealloying). Intergranular corrosion is the result of a local
attack at the grain boundary that does not affect the grains themselves. Impurities at the grain
boundary or the segregation of alloying elements to the grain boundary change the local
composition of the grains at the boundary. This protects the grains themselves, but creates a
depletion zone along the boundary that is more corrosion prone than the surrounding grains, and
subject to local attack. In galvanic corrosion, two dissimilar metals in an electrolyte display a
potential difference when connected. The more corrosion-prone metal is anodic while the other is
cathodic, and the resulting electron flow corrodes the anode [23]. Stress corrosion cracking,
intergranular corrosion, and galvanic corrosion are discussed in greater detail in Section 2.2.1
(Corrosion Along Grain Boundaries), Section 2.2.1.1 (Corrosion in 5xxx Series Al Alloys), and
Section 4 (Work to Date).

12
Figure 4: Summary of the eight forms of corrosion: uniform corrosion, hydrogen damage,
galvanic corrosion, intergranular corrosion, crevice corrosion, dealloying, pitting, erosion
corrosion, and environmentally induced cracking [25].

13
2.2.1 Corrosion Along Grain Boundaries
Intergranular corrosion (IGC) is a corrosion process that occurs locally along the grain
boundary, as a result of anodic precipitates, a local decrease in alloying elements or segregation
of alloying elements [26]. The set of grain boundaries in a given sample possess a variety of
grain boundary energies, and as a result, display varying degrees of boundary migration and
segregation, as well as varying degrees of diffusion and precipitate formation [9, 26, 27].
Additionally, boundary connectivity and the mode in which the different types of boundaries
congregate in a material often gather together groups of corrosion-prone or corrosion-resistant
boundaries in separate regions. This is a common occurrence in FCC materials with low or
medium stacking fault energy [26]. IGC will be discussed in the context of the 5xxx series AlMg alloys in the next section, along with corrosion mitigation strategies based in GBE.
2.2.1.1 Corrosion in 5xxx Series Al Alloys
5xxx series aluminum alloys are a corrosion-resistant subset of aluminum alloys
characterized by a super-saturated magnesium content of ≥ 3% Mg by weight. Their weldability
and high strength to weight ratio via solid solution strengthening have made them attractive
materials for naval and transportation applications [28-30]. However, at elevated temperatures
from 50-200 ˚C, the Mg precipitates to the grain boundaries, forming anodic Al3Mg2 ßprecipitates that preferentially corrode [30-32]. This process, termed sensitization, leaves these
materials vulnerable to intergranular corrosion (IGC) and stress corrosion cracking, and it occurs
in 5000 series alloys that contain a Mg content greater than 3 wt% (a supersaturated solution)
[22, 30, 31, 33]. In these systems, Mg diffuses from the solid solution grains to the GBs and
forms an Al3Mg2 metallic compound called a ß-phase precipitate. When exposed to corrosive
environments, the anodic electrochemical difference between the ß-phase and the grain leads to

14
the preferential corrosion of the ß-phase. This causes intergranular corrosion and can lead to
stress corrosion cracking which can leads to a degradation in mechanical properties and
expensive repairs when used in naval applications [30].
Figure 5 below is a visual depiction of sensitization in 5xxx series Al-Mg alloys. Blue
represents the Al atoms, while green represents the Mg atoms. From left to right: Al and Mg
atoms together in the lattice form discrete grains; when heat is applied to the system, some of the
Mg atoms diffuse to the grain boundaries, resulting in the nucleation of ß-precipitates; the
material is exposed to a corrosive environment, the anodic precipitates cause a potential
difference between the grains and precipitates, resulting in intergranular corrosion occurs along
the grain boundaries as the precipitates preferentially corrode.
Figure 5: The sensitization process in a 5xxx series Al-Mg alloy, with Al (blue) and Mg (green)
atoms. From left to right: the alloy as received, with Mg atoms dispersed in the alloy; the alloy
after exposure to heat, with the formation of Mg beta-phase precipitates along the grain
boundary; the alloy after exposure to a corrosive environment, with intergranular corrosion along
the boundaries.

15
Figure 6: Continuous ß-precipitates along the grain boundary in 5xxx series Al-Mg [34]
Previous studies have shown that factors such as the grain boundary misorientation angle
and the orientation of the grain boundary plane can influence the degree of ß-phase precipitation
and as a result, the degree of corrosion along grain boundaries in 5xxx series Al-Mg alloys
specifically [13, 20, 32]. Grain boundaries with a misorientation angle of ≤ 15˚ have been shown
to be resistant to intergranular corrosion [20, 29, 32]. This misorientation angle at the grain
boundaries can be geometrically described using coincident-site lattice (CSL) theory [7, 8], as
described in section 2.1.1. Boundaries with a Σ value of ≤ 29 are called special grain boundaries
and have been shown to improve material properties, including corrosion resistance, compared to
non-special grain boundaries [7, 13] For example, 5xxx series Al-Mg materials synthesized via
magnetron sputtering with higher fractions of low Σ (≤ 29) grain boundaries have shown
improved resistance to sensitization compared to conventional materials fabricated from the
same family of alloys [13]. A previous study on sputtered Al and Ni samples, demonstrated that
the fraction of twin (Σ3) boundaries could be tailored through the use of different sputtering rates
[35]. This concept is further explored in Chapter 4.

16
2.2.2 Dealloying
Corrosion can also be leveraged as a synthesis method, through a selective corrosion
process called dealloying. Precursor materials for dealloying are specially selected binary or
ternary alloys that have a reduction potential difference of at least 0.53V between the two
components [36]. A table of reduction potentials is provided in Figure 7 [37].
Figure 7: Electrochemical reduction potentials with associated half-reactions [37].
Common precursor alloy systems for dealloying include but are not limited to gold/silver,
silver/aluminum, copper/aluminum and copper/zinc. Within these alloy systems, there is a
narrow range of compositions that can be dealloyed, and each composition has a critical potential
value that must be reached for the reaction to proceed through the material [38, 39]. The

17
composition range is defined by the parting limit, which is the minimum quantity of the less
noble component of the system necessary to prevent the formation of a passive layer by the more
noble component on the surface during the initial stages of corrosion [24]. The dealloying
reaction progresses through the bulk of the material if the quantity of the less noble atom in the
precursor material is above this value. Below this value, the more noble of the components can
form a passive layer on the material surface during corrosion, inhibiting the reaction [40].
Figure 8: Dealloying in the Au/Ag alloy system, denoted by orange and grey, respectively.
Vacancies begin to form in (a) as Ag atoms diffuse from the sample surface. The Au atoms
reorganize in (b) to form a passive layer that protects the more accessible Ag atoms. Once the
exposed surface area of Ag atoms is too large to passivate (c, d), new areas are exposed and
ligaments begin to branch [24].

18
During the reaction, the precursor material is introduced to an electrolytic solution. The
less noble component preferentially corrodes, leaving the more noble atoms, which rearrange
themselves into a porous connected structure. If the reaction continues past the time point at
which the less noble component has diffused out of the structure, these atoms will rearrange
themselves to the point that the foam ligaments begin to coarsen [41]. This is demonstrated in
Figure 8, a diagram of an Au-Ag alloy system. In box a, the easily accessible atoms of the less
noble element (silver) disperse into the electrolyte. Then, in box b, the more noble atoms (gold)
begin to rearrange into a passive layer on the surface in contact with the electrolyte. This process
continues until the exposed surface area of silver atoms is too large for a continuous passive
layer (box c), at which point new areas of silver atoms are exposed and the ligaments created by
the gold passive layer begin to branch (box d) [24].
The two main synthesis methods for dealloyed materials are free corrosion and
electrochemically-driven dealloying. In free corrosion, the precursor material is placed on a
fritted glass slide or a noble metal mesh, and the sample and holder are placed in a beaker or
other inert glass container with an electrolytic solution. The reaction is finished when there are
no visible bubbles, and occurs over the course of a few days, though the reaction begins
immediately [42, 43]. In electrochemical dealloying, the sample is attached to an Au working
electrode, and placed in an electrochemical cell with an electrolytic solution, an Ag/AgCl
reference electrode and a Pt counter-electrode. Figure 9 shows a representative schematic of an
electrochemical cell [41].

19
Figure 9: Representative schematic of an electrochemical cell setup for potentiostatic dealloying
[41].
A fixed voltage (potentiostatic dealloying) or a stepped voltage (galvanostatic dealloying)
is run through the system, and the reaction is finished when the dissolution current is negligible
[42, 43]. Both dealloying methods yield continuous porous structures, where the solid portion of
the porous metal framework is continuous through the system [36]. The resulting porous
structures, or foams, are a type of cellular material.
2.3 Cellular Materials
Cellular materials can be categorized in a number of ways. The first is anisotropic versus
isotropic cellular materials. Anisotropic solids have directional properties, which signifies that
there is a change in properties such as strength when the material is loaded on different axes.
These include natural materials such as balsa wood. The properties of isotropic materials are not
directional. Cellular materials can also be described as having open or closed cells. In open cell
materials, the solid portion of the solid/air composite marks the perimeter of the cell-like units,

20
while in closed-cell structures, the solid material surrounds pockets of air. Lastly, cellular
materials can be designated as ordered or stochastic. The structure of ordered or periodic cellular
solids is predictable, and may have a repeating pattern. Stochastic cellular materials, in contrast,
are random in nature [44].
2.3.1 Structure of Cellular Materials
Foams, a subset of cellular materials, can be categorized by both their relative density and
their foam pore/ligament size, where relative density is the foam density divided by the bulk
density of the material. This is inversely related to the foam’s porosity. For example, for a
precursor material of 30 at.% Cu and 70 at.% Al (Al70Cu30), the expected result would be a
dealloyed foam with a relative density of 30% [45]. Figure 10 shows three different nanoporous
copper foams, ranging from 20% to 35% relative density [46].
Figure 10: 3 nanoporous copper foams of varying relative densities. (a) is a 20% relative density
foam, dealloyed via free corrosion in 85 wt.% H3PO4; (b) is a 30% relative density foam,
dealloyed via free corrosion in 5 wt.% HCl; and (c) is a 35% relative density foam dealloyed
electrochemically in 5 wt.% HCl [46].
(a) (b) (c)

21
2.3.2 Nanoporous Foams
Nanoporous foams refer to porous materials with nanoscale ligament sizes, and are
notable for their high surface area, electrical and thermal conductivity, and strength [39].
Precursor materials include Au, Ag, Cu, and Ti-based alloys, along with some bulk metallic
glasses, all with a wide range of starting geometries [47]. These materials are of interest for their
wide variety of practical applications especially in energy-related fields [48]. Nanoporous gold
has applications in microelectronics, ultra-thin electrodes, while nanoporous silver has catalytic,
sensing, and optical applications [49-51]. Nanoporous copper foams have applications as
catalysts to synthesize hydrocarbons such as HCOOH, C2H4 and CH4 from CO2 gas, without the
addition of the high temperatures and pressures required in the traditional reaction pathway [52].
Titanium-based nanoporous materials have applications in gas permeation membranes and gas
sensors, while titanium dioxide specifically is of interest in dielectric, solar cell, and lithium
extraction applications [53, 54]. Lastly, the high surface area and amorphous structure of
nanoporous bulk metallic glasses such as Yr47Cu46Al7 are uniquely suited for gas absorption in
hydrogen storage applications [55]. The most versatile synthesis process for the fabrication of
these nanoporous materials is dealloying, which is described in detail in the next section [47].
Other common synthesis methods are briefly covered in section 2.4.1.2, Porosity Gradients.
2.3.2.1 Fabrication of Nanoporous Foams Via Dealloying
Dealloying is a reproducible and flexible foam fabrication method based on selective
dissolution, with a variety of precursor materials that include metallic nanoparticles, nanotubes,
films, ribbons, and bulk metallic glasses [24, 39, 47]. Ligament and pore sizes ranging from the
nano- to microscale can be achieved through balancing the corrosion rate of the less noble
component with “the rate of reorganization” of the more noble component. In order for the

22
reaction to occur, the atomic percentage of the less noble atoms must be at or above a critical
value called the percolation threshold, a geometric limitation that denotes the minimum atomic
percentage of the less noble atoms necessary to form a continuous structure [24]. The pore and
ligament size of the resulting nanoporous metallic structure increases over time, especially at
elevated temperatures, due to the high surface area [48].
A large fraction of the work on dealloyed nanoporous foams to date has focused on
nanoporous gold structures, as they are both chemically stable and simple to synthesize [56].
However, this synthesis method is not limited to Au-alloys. Cu, Ag, Pd, and Ti-based alloy
precursors are also noted in the literature, as well as select bulk metallic glasses [47, 54, 55, 57].
Recent work with this method includes nanoporous films synthesized from magnetron sputtered
CuxAl100-x alloys to better understand the mechanical properties of nanoporous copper thin films
and outline processing specifics such as electrolyte concentration and film thickness for sputtered
non-noble films [58]. Additionally, work by Z.Y. Hu et al. on sputtered Cu25Al75-Cu40Al60
bilayer copper films produced nanoporous films through a single-step free corrosion dealloying
reaction in NaOH. The films displayed a distinguishable boundary between the two different
compositions due to the difference in pore sizes, for both Cu25Al75/Cu40Al60 (Cu40Al60 deposited
first) and Cu25Al75/Cu40Al60 (Cu25Al75 deposited first) films. In Figure 11, a Cu40Al60 layer was
deposited first, and the columnar grains in that layer remained after dealloying with pores in the
grains (Figure 11(d)), while the Cu25Al75 layer in the same sample formed pores both in the
grains and along the grain boundaries.

23
Figure 11: SEM micrographs of the Cu25Al75/Cu40Al60 (Cu40Al60 deposited first) sputtered
bilayer film from Hu et al. As-sputtered (a) top surface and (b) cross-section; after dealloying by
free corrosion in NaOH, (c) is the top surface and (d) is the cross-section. Note that the interface
between the two dealloyed layers is visible in (d) [59].
When the order of the sputtered layers was reversed (Cu25Al75 deposited first), the Cu25Al75 layer
formed an isotropic nanoporous structure after dealloying (Figure 12 (d)), while the dealloyed
structure of the Cu40Al60 was the same as that in Figure 11.

24
Figure 12: SEM micrographs of the Cu25Al75/Cu40Al60 (Cu25Al75 deposited first) sputtered
bilayer film from Hu et al. As-sputtered (a) top surface and (b) cross-section; after dealloying by
free corrosion in NaOH, (c) is the top surface and (d) is the cross-section. Note that the interface
between the two different porosities is visible in (d) [59].
However, it should be noted that the main focus was on synthesizing a bilayer nanoporous
copper film, and analyzing how the reaction proceeded through the bulk film by imaging the
dealloying front, the location where the reaction was actively occurring [59]. Thus, the properties
of the interface between two similar-composition layers with different porosities, including
interfacial stability and mechanical properties, are still unknown. These properties, as well as
those of interfaces between layers that are dissimilar in both chemical composition and porosity,
will be an important area of future research for their applications in Raman scattering and
sensing applications [60].

25
2.4 Graded Microstructures
Graded microstructures are a subset of heterogeneous microstructures, which are
intentionally anisotropic and allow for enhancements in material properties [61]. Gradient
nanostructured materials in particular have demonstrated unique improvements in mechanical
properties. Depending on the synthesis method, a variation in feature sizes from tens of
nanometers to about 10 microns can be achieved. There are two main categories of gradients
materials: chemical gradients and structural gradients.
Figure 13: Types of gradient microstructures. Top row, structural gradients, from left: grain size,
twin thickness, lamellar thickness, mix of grain size and twin thickness. Bottom row, chemical
gradients, from left: phase, solid-solution concentration, chemical compound, mix of structural
and chemical gradients [4].
Grain size Twin thickness Lamellar thickness Grain size and twin thickness
Phase Solid-solution
concentration
Chemical
compound
Mixed
Structural gradient Chemical gradient

26
Chemical gradients can produce a gradient in both chemical properties and select mechanical
properties such as resistance to impact damage, and include gradients in solid-solution
concentration, chemical composition and phase. They are predominantly of interest in biological
materials. Structural gradients dominate metals and alloys, and include gradients in twin
thickness, grain size, lamellar thickness, or a combination of the above [4].
2.4.1 Structural Gradients
Structural gradients can be synthesized both through top-down methods like surface
mechanical treatments and accumulative roll bonding, as well as through bottom-up approaches
such as 3D printing, electrodeposition, and magnetron sputtering. Bottom-up deposition methods
specifically provide greater specificity in the subsequent gradient as a result of changes in the
deposition parameters, providing more precise control of the resulting microstructure. The
mechanical properties of structurally graded materials do not follow the traditional
strength/ductility tradeoff of their coarse-grained or randomly-grained counterparts [4]. Instead,
the strength difference between nanocrystalline and coarser-grained domains results in “soft” and
“hard” areas in the microstructure. During plastic deformation, the deformation of the softer
coarse grains is constrained by the harder nanocrystalline grains due to their high grain boundary
density. This produces strain and dislocation buildups in the coarser grains, as the dislocations
cannot propagate through the materials until the overall stress is great enough to plastically
deform the nanocrystalline grains [4, 61]. The mixed grain-size microstructure creates local
strain gradients along the border where the nanocrystalline and coarser grains meet, which grow
as dislocations aggregate in coarser grains. This work hardening method is called back-stress,
and contributes to the improved ductility of these materials [61].

27
In addition to work hardening, structural gradients can also provide enhanced friction and
wear resistance, increased strength with good ductility, and improved fatigue resistance. With
regards to the strength-ductility tradeoff, materials with regions of both nanotwins and fine
grains specifically exhibit a smaller loss in ductility than materials with a similarly fine
(submicron) grain size without nanotwins, as the boundaries can both prevent dislocation
propagation and increase the local dislocation density [4]. There have been few studies on
gradient nanostructured materials and their effect on corrosion resistance, another property of
interest [4, 62]. Work on select SMAT-processed stainless steels showed impaired corrosion
resistance at room T as the strain energy at the material surface inhibited a protective passivation
layer [62]. However, this mechanism “is not fully understood”, and the corrosion resistance
depends on a variety of factors including the exact processing parameters, the temperature, and
the starting material [63].
2.4.1.2 Porosity Gradients
There are four main fabrication methods for porous metallic materials with hierarchical
structure: additive manufacturing, templating, sintering, and dealloying [47]. Additive
manufacturing can be used to produce a variety of porous metallic structures from open-celled
porous structures to hierarchical structures, with a minimum pore size on the order of ~300 µm
(Song 34-36, 37) [47]. Templating is a deposition process wherein a metal coating is applied to a
sacrificial template, and the template is removed to produce a thin porous structure. It is more
limited in scope than additive manufacturing, with applications in hierarchical thin-walled lattice
structures or porous films [47].

28
Figure 14: Templating fabrication method for porous structures. A sacrificial template is coated
via a deposition process, and the template is removed to produce a film [47].
During sintering, a mixture of metal powders and space holders are compressed, the spacers are
removed, and the compressed powder is heated to form a porous structure with microscale or
larger pores. Isolated pores of various sizes and shapes, as well as bimodal or hierarchical porous
structures, can be created via changes in the spacers and spacer materials [47, 49, 64].
Figure 15: From left to right: metal powders and space holders are mixed; the mixture is pressed;
the spacers are removed from the compressed mixture; the powder is heated to form a solid
structure [47].
Dealloying, which is the most versatile of the four fabrication methods, is described in detail in
sections 2.2.2 and 2.3.2.1. This process can produce porous structures in a variety of shapes, as
the structure retains the shape of the precursor material, and pore sizes on the micron or
nanoscale [47, 65, 66]. Figure 16 demonstrates a multi-step dealloying synthesis of a hierarchical

29
porous gold structure. A precursor Au/Ag alloy is partially electrochemically dealloyed to form a
porous structure, and then annealed to coarsen the pores. The alloy is then dealloyed again at a
higher potential to produce a second level of porous structure [47, 67]. In this case, shown in
Figure 17, the process produces large pores that facilitate ion transport, and smaller pores within
the larger ligaments that increase the functionalized surface area. This improves the signal
transport by a factor of >2 compared to nanoporous gold of the same specific surface area, from
a controllable synthesis process that can be applied to both bulk and film samples [67].
Figure 16: A multi-step dealloying process to produce a hierarchical porous structure. From left
to right: an Ag-Au precursor alloy; the alloy undergoes selective corrosion to form a porous
structure; the structure is annealed to coarsen pores; the material undergoes selective corrosion
once more and exhibits another level of porous structure [47].
Figure 17: Cross-sectional SEM micrograph of hierarchical nanoporous gold, electrochemically
dealloyed through a multi-step dealloying process [67].

30
2.5 Mechanical Properties of Nanoporous Foams
The mechanical properties of nanoporous foams, specifically nanoporous gold, have been
the focus of both simulation and experimental studies [56]. Experimental methods have included
microtensile and nanoindentation testing to determine yield strength and elastic modulus [68,
69]. Characterization of the mechanical properties of nanoporous foams began with the GibsonAshby model for determining the mechanical properties of foams, according to the following
equations:
�!" = �!"#�!
�!" = 0.3�!"#�!/!
with relative density φ, Young’s modulus E and yield stress σ, for both the ligament and the
nanoporous material [45, 70]. Experimentally derived values for the elastic modulus and yield
strength of nanoporous foams can be greater than those predicted by the Gibson-Ashby model,
especially for foams with lower relative densities, and thus there is disagreement on the
applicability of the unmodified Gibson-Ashby model for deriving the mechanical properties of
nanoporous foams, specifically as the model omits nanoscale-related microstructural factors [45,
56, 71-73]. These microstructural factors include but are not limited to ligament diameter,
ligament connectivity, grain size, and vacancy density [56, 72, 74]. Revised scaling laws have
been developed to incorporate some of these aspects, including ligament connectivity (Liu et al.),
relative density and average ligament size (Hodge et al.), and average ligament size and hardness
(Viswanath et al.) [72, 74-76]. Mechanical testing simulations based on revised Gibson-Ashby
laws have provided insight into deformation mechanisms and the effects of the microstructure in

31
deformation in nanoporous gold and other FCC nanoporous metals. Notably, plastic deformation
during compression preferentially occurs at foam nodes on the ligament surface due to the
aggregation of dislocations, and the resulting buildup of stacking faults [56]. For a set of
nanoporous foams loaded in tension with constant ligament sizes, plastic deformation occurs via
a similar mechanism, but is more concentrated in the nodes for samples with a higher degree of
porosity. This is likely because higher degrees of nanoporosity correlate with a lower degree of
ligament connectivity in the solid portion of the foam structure, which decreases both the
strength and stiffness of the overall structure [45].

32
Chapter 3. Experimental Methods
The following section provides an overview of the experimental methods employed in
this study. Deposition configurations and corrosion methods utilized in the grain boundary
engineering and the gradient nanoporous copper studies are discussed. Additionally, the
microstructural characterization methods used to study 5xxx series Al in Chapter 4 and the
microstructural and spectroscopy characterization techniques used to characterize the gradient
nanoporous copper in Chapters 5 and 6 are outlined.
3.1 Magnetron Sputtering
Magnetron sputtering is a physical vapor deposition process used to synthesize films and
coatings. This process is highly adaptable, and enables films from a wide variety of precursor
materials, with greater control over both film thickness and composition when compared to other
film synthesis methods such as chemical vapor deposition or electroplating. Films synthesized
with this technique also exhibit greater adhesion to the substrate material and better uniformity
[77, 78]. DC magnetron sputtering specifically allows for coatings of conductive materials and
high sputtering (deposition) rates [77]. Figure 18 shows a schematic of a single-source DC
magnetron sputtering configuration.
In this configuration, a target of the material to be sputtered acts as the cathode, and the
substrate is located on the anode [79]. The chamber is placed under vacuum and then flooded
with argon gas. A DC current is then applied to the system, creating Ar+ ions that form an argon
plasma [78]. The sputtering target is generally planar, and in this configuration a ring of magnets
on the reverse side of the target supplies a magnetic field that traps the incoming electrons in the
portion of the field in the ring, creating a closed path of electrons in the plasma. When the ions
from the plasma collide with the plasma region on the target, some of their energy and

33
momentum is transferred to the atoms at the collision site, and a few atoms are emitted from the
target surface. These emitted, or sputtered, ions deposit onto the inside of the chamber,
particularly on the substrate. The sputtering yield for this interaction is the average number of
emitted electrons per collision between an Ar+ ion and the target, and differs with the energy and
type of plasma ion, as well as the specific target material [77]. Other controllable system
variables include the target polarization, distance from the target to the substrate, and working
pressure, and different combinations of these parameters shape the microstructure, deposition
rate, and deposition temperature of the sputtered film [80, 81].
Figure 18: Schematic of a single-source configuration in a DC magnetron sputtering chamber.
The DC magnetron sputtering process can be utilized to fabricate films with tailored
microstructures and compositions. Figure 19 shows the two planar target configurations utilized
in this study: single target (19a) and dual target (19b).

34
Figure 19: Schematic of the two planar sputtering configurations utilized in this study. a) shows
the single target (single composition) configuration that was used to produce the films in Chapter
4, while b) shows the dual target configuration used to produce the gradient films in Chapters 5
and 6.
If two targets are connected to the system, each with their own DC power source, they can be
selectively polarized one at a time to produce a film consisting of repetitive layers composed of
the two different target compositions, or a film comprised of two distinct layers [82].
Additionally, the two sources may also be sputtered simultaneously but at different rates, to
produce a film whose composition is a mix of the two target materials. The sputtering rates of
each individual target, which are controlled by the applied power, can be held constant through
film deposition, or one or both can be varied to create a chemical compositional gradient through
the film thickness [58, 59]. This precise compositional control and the microstructural control
mentioned above make magnetron sputtering the film synthesis method of choice for this study.
3.2 Corrosion Methods
Three corrosion methods are of interest in this study. The first is passive immersion
corrosion, which features in Section 2.2.1.1 and Chapter 4 (Corrosion in 5xxx Series Al Alloys

35
and Corrosion Resistance in Grain-Boundary Engineered 5xxx Series Al Alloys, respectively). In
passive immersion corrosion, a sample is immersed in an aqueous solution such as artificial
seawater (NaCl in water) [13]. After a set period of time, the sample is removed, washed, and
dried. The surface of the sample is then assessed with SEM to determine the presence of features
such as cracks or pits that indicate corrosion has occurred. There is no applied potential in the
system, and the main corrosion mechanisms are intergranular corrosion, galvanic corrosion, and
pitting [23]. If the material is corrosion-resistant, a continuous passive oxide layer forms on the
surface, which protects it from attack by the corrosive environment [83].
The second corrosion method is dealloying by free corrosion. No potential difference is
applied to the material, but a reduction potential difference between components of ≥ 0.53V
drives the reaction, and the less noble element preferentially dissolves into the electrolyte, or
corrosive environment [38]. The third corrosion method, electrochemical dealloying, also relies
on an underlying potential difference and selective corrosion. However, in electrochemical
dealloying there is also an applied potential difference. This applied potential can be either a set
potential (potentiostatic) or a ramp function (galvanostatic) [43]. The applied potential provides
greater control over the reaction than in free corrosion and is determined from the anodic
polarization of the precursor alloy. An example potential curve for Ag, Au, Au20Ag80, Au25Ag75,
and Au30Ag70 is shown below in Figure 20. The resulting critical potentials for the alloys are 0.8,
0.94, and 1.01V, respectively [41, 84].

36
Figure 20: Example of anodic polarization curves 0n 0.1M HClO4 of Ag, Au, Au20Ag80,
Au25Ag75, and Au30Ag70. Resulting critical potentials for Au20Ag80, Au25Ag75, and Au30Ag70 are:
0.8, 0.94 and 1.01 V [84].
3.2.1 Dealloying
Dealloying is a selective corrosion process that is used to synthesize porous materials
from metals or bulk metallic glasses in a large range of starting geometries. Precursor materials
for dealloying are specially selected binary or ternary alloys, generally binary, that have a
reduction potential difference of at least 0.53V between the two components. Common alloy
systems include but are not limited to gold/silver, silver/aluminum, copper/aluminum, and
copper/zinc. Within these alloy systems, there is a narrow range of compositions that can be
dealloyed, and each composition has a critical potential value that must be reached for the
reaction to proceed through the material [38, 39]. The composition range is defined delineated by
the parting limit, representing the minimum amount of the less noble component within the

37
system necessary to prevent the formation of a passive layer by the more noble component on the
surface during the initial stages of corrosion [24]. The dealloying reaction progresses through the
bulk of the material if the quantity of the less noble atom in the precursor material is above this
value. Below this value, the more noble of the components can form a passive layer on the
material surface during corrosion, inhibiting the reaction [40].
The reaction begins at the sample surface in contact with the electrolytic solution. As the
less noble atoms start to dissolve into solution, the more noble atoms rearrange themselves into a
passive layer on the surface in contact with the electrolyte. This process continues until the
exposed surface area of the less noble component is too large for a continuous passive layer,
signifying that the atomic percentage of the noble component is below the parting limit. At this
point, new regions of the less noble component below the sample surface are exposed to the
electrolyte, and the ligaments created by atomic rearrangement into a passive layer begin to
branch [24]. If the reaction continues past the time point at which the less noble component has
diffused out of the structure, these atoms will rearrange themselves to the point that the foam
ligaments begin to coarsen, and change the character of the porosity to a less homogeneous one
[41]. Continued electrolyte exposure in thin film samples will break the structure [85].
There are two main types of dealloying: free corrosion, and electrochemical dealloying.
During free corrosion, the sample is placed on a fritted glass slide or a noble metal mesh in an
inert glass container, with an electrolyte such as a concentrated nitric acid solution. This is
shown in Figure 21 below. The reaction is finished when there are no visible bubbles on the
sample surface, and occurs over the course of a few minutes to a few days, though the reaction
begins immediately [36].

38
In electrochemical dealloying, the sample is attached to an Au working electrode, and
placed in an electrochemical cell with an electrolytic solution, an Ag/AgCl reference electrode
and a Pt counter-electrode. A representative schematic of the electrochemical cell setup is shown
in Figure 9, while Figure 21 shows the laboratory electrochemical cell towards the end of a
dealloying experiment. A fixed voltage above the alloy’s critical potential is run through the
system, and the reaction is finished when the dissolution current is negligible [36, 42].
Figure 21: The elctrochemical cell setup towards the end of the reaction for an Au30Ag70
precursor in 5M HNO3 at a voltage of 1100 mV.
Both dealloying methods yield continuous porous structures where the solid portion of the
porous metal framework is continuous through the system[36]. Free corrosion is simpler to
perform, but the resulting porous structures are prone to cracking, especially for constrained thin

39
films on stiff substrates such as silicon or glass [41, 86]. The decrease in volume for constrained
films during dealloying that occurs to relieved tensile stresses also frequently produces cracks in
the nanoporous films, and is especially prevalent during free corrosion [41, 87]. A decrease in
film thickness to the 45-75 nm range can reduce cracking, but will produce a 2D instead of a 3D
porous structure [88]. Electrochemical methods such as potentiostatic dealloying, while more
complicated to perform, allow more control over the reaction by utilizing the potentiostat to
control the applied cell potential [41]. This yields a way to change the dissolution rate of the less
noble component in the precursor material by changing the potential, instead of the reaction
temperature, electrolyte, and/or electrolyte concentration as in free corrosion [89].
3.3 Characterization Techniques
A variety of electron microscopy, spectroscopy and mechanical testing techniques were
used to characterize as-sputtered solid films and dealloyed specimens.
3.3.1 Scanning Electron Microscopy
A scanning electron microscope consists of an electron source that is modified by a series
of apertures and electromagnetic lenses to produce a beam. The electron source, a special
metallic filament or a field emission gun, produces a stream of electrons that travel down the
vacuum column. The properties of this beam can be tailored by adjusting the lenses and
apertures, which successively change the beam diameter and beam current to focus it on the
sample [90]. A schematic of the scanning electron microscope and its lenses is shown in Figure
22 [91].

40
Figure 22: Schematic of a scanning electron microscope. The beam is generated at the source,
and travels through a series of lenses before reaching the sample. Interaction between the
electron beam and the sample produces a variety of signals, which are picked up by detectors
such as the back-scattered electron detector, the secondary electron detector, and the x-ray
detector [91].

41
As the beam scans over the sample surface, collisions between the incident electrons and the
sample surface produce a number of signals as shown in Figure 23, that supply information on
the elemental composition, crystal orientation and surface features of a sample. Elastic collisions,
where energy is conserved, result in scattering at high angles, producing backscattered electrons.
Inelastic collisions yield characteristic x-rays and secondary electrons. [92].
Figure 23: Diagram of the signals produced by incident electrons during SEM for a thin sample
[92].
Surface topology information in the form of electron images derives from the secondary electron
signal. These electrons have a low energy of less than 50 eV, and are produced just below the
sample surface. Backscattered electrons have a minimum energy of 50 eV, and originate from
scattering events slightly deeper in the sample. Electron backscatter diffraction (EBSD) and

42
transmission Kikuchi diffraction (TKD) convert this backscattered electron signal to
crystallographic information. Characteristic x-rays are produced by inelastic scattering events at
depths ranging from the top surface of the sample to well into the sample cross-section. Energy
dispersive x-ray spectroscopy (EDS) maps these characteristic x-rays to the chemical
composition of the sample [90].
3.3.2 Focused Ion Beam
A focused ion beam (FIB) is a microscope that enables sample milling and imaging in
situ. A liquid-metal ion source (LMIS) provides a stream of ions that scans the sample surface
and produces secondary electrons. Most ion sources for the FIB are liquid-metal ion sources
(LMIS), with gallium as the most common due to its stability and low melting temperature. More
recently, a Xe-based plasma FIB (PFIB) was developed as an alternative to the Ga FIB. These
new systems have similar applications as the Ga FIB such as TEM lamella preparation, are less
reactive with specimens, and reduce sample damage during milling, but may be less precise at
length scales <500 nm [93]. The FIB is frequently a FIB/SEM dual-beam system to facilitate
sample milling and high resolution electron imaging [94]. It is commonly used to prepare TEM
lamella as well as preparing polished sample cross-sections for chemical and structural analysis
such as EDS or EBSD. A schematic of the combined FIB/SEM is shown in Figure 24 [95].

43
Figure 24: Schematic of a standard SEM/FIB dual-beam system with ion and electron beams,
sample stage, assorted detectors, sample manipulator, and gas injection system for localized
deposition [95]
To prepare a sample for milling, a protective coating of Pt, W or C is deposited via gas injection
onto the sample surface [95]. A rough cut is performed at a high current, with subsequent milling
cycles at successively lower currents until the sample is polished to the desired level. An
example of a series of milling operations for a polished cross-section is shown in Figure 25 [96].

44
Figure 25: Milling process for polishing a sample cross-section, shown on corroded copper. A Pt
protective coating is deposited in step a), and the region next to the protective coating is milled at
a medium to high current in b) to expose the cross-section. c) and d) are shown after milling at
progressively lower currents to produce the polished cross-section shown in step d) [96].

45
3.3.3 Energy Dispersive X-Ray Spectroscopy
Energy dispersive x-ray spectroscopy (EDS) uses the scanning electron beam to map
characteristic x-rays across a sample, and compares the energy signal to a database of the unique
x-ray spectra of the elements [97]. This comparison can be used to confirm the composition of a
known sample for quality-control purposes, or to determine the chemical composition of an
unknown material [98]. Chemical composition can be provided as spectra, maps of individual
elements, or as a set of components in atomic or weight percent. The technique allows for both
local area determinations of elemental composition as well as broad surface composition maps
[99]. An example composition map of four elements present in shungite rock is shown below in
Figure 26 [100].
Figure 26: Example EDS composition map of four elements present in shungite rock [100]

46
3.3.4 Electron Backscatter Diffraction
Electron backscatter diffraction (EBSD) is an SEM-based technique that converts the
backscattered electron signal to sample crystallographic information such as grain boundary
character, grain size, grain orientation, and grain boundary angle. Diffraction information from a
set of scattered electrons is unique to a specific crystal orientation and crystal structure [91].
Figure 27 shows the experimental setup for conventional EBSD, which analyzes the top surface
of a sample, and transmission Kikuchi diffraction (TKD), which analyzes the sample crosssection.
Figure 27: Conventional EBSD configuration (left) and transmission Kikuchi diffraction (TKD)
configuration (right).
In conventional EBSD, the sample surface is tilted at a 70˚ angle to the incident beam, and the
resulting backscatter electrons diffract off the sample surface. This technique has a resolution of
~50 nm. For TKD, a sample cross-section must first be cut and thinned to a thickness of less than
100 nm. This is generally performed with FIB milling, as outlined in section 3.3.2. The sample
cross-section is tilted at an angle ranging from 0-15˚ from the incident electron beam, and the

47
scattered electrons scatter through the sample thickness. The spatial resolution limit for TKD is 2
nm [101]. An example of a set of grain orientation maps for Al-6060 obtained with TKD is
shown in Figure 28 [102].
Figure 28: Example TKD grain orientation map for Al-6060 showing a) TKD pattern quality as a
function of contrast, b) raw orientation data (inverse pole figure not shown) and c) the combined
contrast and cleaned orientation map [102].

48
3.3.5 Surface-Enhanced Raman Spectroscopy
Surface-enhanced Raman Scattering (SERS) derives from the Raman inelastic scattering
phenomena [103]. Electromagnetic radiation from a laser or other monochromatic source excites
the molecules on the surface of a material, which emits waves of a different frequency than the
incident wave. The emitted waves exhibit Stokes scattering if they are of a higher frequency than
the original wave, and anti-Stokes for a lower frequency. Figure 29 demonstrates these scattering
events for an incident light wave that strikes a molecule on a nanoparticle surface [104].
Figure 29: Example of the Stokes and anti-Stokes scattering phenomena for electromagnetic
radiation on a nanoparticle substrate with a target molecule [104].
Traditional Raman yields a low scattering signal, thus surface-enhanced Raman scattering was
developed to increase the signal enhancement and sensitivity, whereby the analyte or target
molecule is adsorbed onto a metallic substrate [105]. A schematic of a common SERS setup is
shown in Figure 30 [104].

49
Figure 30: Schematic of a common SERS instrument setup [104].
The dominant mechanism of signal enhancement from the substrate is localized surface plasmon
resonance (LSPR). When the incident electromagnetic wave hits the substrate, the surface
electrons oscillate and local electromagnetic fields form, leading to surface plasmon resonance
when the oscillation frequency and incident wave frequency are equal. The surface plasmons
transfer energy to the adsorbed analyte molecules, which vibrate. As a result, some of that energy
transfers back to the plasmons, and some scatters at a different frequency [106]. For most SERS
substrates, areas with close particle spacing produce hot spots of greater enhancement as the gaps

50
possess a lower local resonant frequency than the surrounding solid material. The critical gap
size is ~1nm [105]. Additionally, surface roughness and local surface topology features such as
sharp particle edges or pore sizes of 150-250 nm can produce signal enhancement hot spots [107,
108]. Charge transfer between the analyte and the SERS substrate can also contribute to the
SERS enhancement. However, this effect is highly dependent on the specific analyte, its ability
to adsorb to the metallic substrate, and the relationship between the Fermi level of the substrate
and the energy gap (difference between the conduction band and the valence band) of the
analyte. A schematic of the charge transfer mechanism is shown in Figure 31 [109].
Figure 31: Schematic of the charge transfer mechanism between a SERS analyte and the metallic
substrate [109].

51
3.4.1 Vickers Hardness Testing
Vickers microhardness testing is a form of indentation hardness testing with a squarepyramidal indenter tip. Microhardness or microindentation refers to the size of the test force (1 gf
to 1000 gf) compared to traditional hardness testing such as Rockwell [110]. A pre-determined
load is applied via the indenter onto the sample surface, and the diagonals of the resulting
indentation are measured as shown in Figure 32. The penetration depth is a function of the
applied load force and the average diagonal length of the indent [111].
Figure 32: Schematic of the Vickers diamond pyramidal indenter tip, showing the angle between
faces (136˚), and the resulting indent [110].
The measured lengths of the diagonals can be converted to a hardness value called the Vickers
Hardness Number (HV) with the following formula:

52
�� = �
� = 1854.4�
�!
where F is the load in grams-force, A is the contact surface area of the indent in µm2
, and 1854.4
is a geometric constant. The quantity d is the average measured length of the diagonals for the
indent in µm [111]. The resulting hardness number HV can also be converted to an approximate
yield strength σy in MPa through the following equation
�! ≈ �
3
where H is defined as the product of 9.807 and the Vickers hardness value HV [111]. Example
indents from Vickers microhardness testing in sputtered ceramic films are shown in Figure 33.
Figure 33: Optical micrographs of Vickers indents on optical ceramic multilayer films. From left:
a) AlN(47)/Ag(21), b) AlN(20)/Ag(20), and c) AlN(127)/Ag(10). Layer thickness in nanometers
of each component listed in parentheses [112].

53
Chapter 4. Corrosion Resistance in GB-Engineered 5xxx Series Al Alloys
The following work was published as Characterization of Grain Boundary-Engineered
Aluminum-Magnesium Alloys in the journal Advanced Engineering Materials. It can be found
in Volume 23, Issue 1 (DOI: 10.1002/adem.202000813) and at reference [21].
This study explores the effect of sputtering rate on the fraction of special grain
boundaries in 5xxx series Al-Mg alloys. Samples are synthesized via interrupted DC magnetron
sputtering with varying deposition rates, and the grain size and grain boundary character are
evaluated with EBSD. The highest sputtering rate (7 nm s-1
) leads to an increase in the total
number of special grain boundaries, ~1.5x greater than that of the lower rates. Increased thermal
energy and enhanced stress relaxation during film growth promote the formation of Σ3 and Σ7
boundaries.
4.1 Sensitization and Corrosion Resistance
5xxx series aluminum alloys are a corrosion-resistant subset characterized by a supersaturated magnesium content of ≥ 3 wt.% Mg, with a high strength-to-weight ratio and good
weldability that have made them attractive materials for naval and transportation applications
[28-30]. However, at elevated temperatures from 50-200˚C, Mg diffuses to the grain boundaries
and forms anodic Al3Mg2 ß-precipitates that preferentially corrode [30-32, 113]. This process,
known as sensitization, leaves these materials vulnerable to intergranular corrosion (IGC) and
stress corrosion cracking [22, 30, 31, 33, 113].
Previous studies have shown that factors such as the grain boundary misorientation angle
and grain boundary plane orientation can influence the degree of precipitation [13, 20, 32]. For
example, grain boundaries with a misorientation angle ≤ 15˚ have been shown to be resistant to
the formation of ß-phase, and materials with coincident site lattice (CSL) boundaries with a Σ

54
value of ≤ 29 have demonstrated improved material properties compared to those with random
angle grain boundaries [7, 13, 20, 29, 32]. The types of boundaries in a material can be tailored
via grain boundary engineering (GBE), a process which allows for designed microstructures that
improve the bulk material properties [14, 16, 20]. While most GBE is performed through
thermomechanical processing, it is also possible to tune the types of grain boundaries through
synthesis techniques like electrodeposition and sputtering [14, 18]. For sputtering in particular, a
previous study demonstrated that the fraction of twin boundaries in Ni and Al could be modified
through the use of different sputtering rates [35]. Furthermore, a sputtered 5xxx series Al-Mg
alloy demonstrated both higher fractions of low Σ (≤ 29) grain boundaries and improved
resistance to sensitization compared to conventional materials [13]. In this study, we go several
steps further by focusing on tailoring the fraction of special grain boundaries in a complex alloy
as well as directly correlating the grain boundary fraction to corrosion morphology.
Specifically, this manuscript examines a range of sputtering rates (1, 3.5, and 7 nm s-1
)
and the resultant boundary fraction of boundaries with a Σ value of ≤ 27 in Al-Mg films.
Characterization via electron backscatter diffraction (EBSD), transmission electron microscopy
(TEM) and scanning electron microscopy (SEM) provide information on special GB fraction,
grain size, sample microstructure, and corrosion morphology as grain boundary engineering at
the nanoscale is explored in the context of sensitization.
Samples were synthesized via interrupted magnetron sputtering with intervals set for 10
seconds on and 100 seconds off onto Si(100) substrates at powers of 1500, 750, and 175 watts,
resulting in sputtering rates of 7, 3.5, and 1 nm s-1
, respectively. The total film thickness was 7
µm for the sample sputtered at 1 nm s-1
, and 10 µm for the samples sputtered at 3.5 and 7 nm s-1
,
as measured by profilometry (Ambios XP-2). Targets for sputtering were obtained from the

55
Naval Research Lab (NRL). An Al5456 (5.3 wt.% Mg) target was used to synthesize the 7 nm s-1
film, while an Al5083 (4.82 wt.% Mg) target was used for both the 3.5 and 1 nm s-1 films. Due to
the higher sputtering rate, there was a decrease in the measured Mg content for the 7 nm s-1
sample compared to the Al5456 target. However, all samples yielded a Mg content of ∼ 4 to 5
wt.%, comparable to that of an Al5083 alloy as confirmed by energy dispersive X-ray
spectroscopy (EDS).
The microstructure of the top-surface and cross-section of the samples were characterized
via EBSD and transmission Kikuchi diffraction (TKD). Cross-section lamellae were prepared via
focused ion beam (FIB) lift-out and characterized with TEM, where bright and dark-field images
were obtained. EBSD and TKD scans were analyzed with the OIM software package where
grain dilation and a neighbor confidence index correlation were performed. The data was
partitioned to remove all points with a confidence index of less than 0.1. The resulting maps
were oriented in the (111) plane and coincident site lattice (CSL) boundaries with values from Σ3
to Σ27 were identified. Average grain size and the grain size distribution were calculated for the
top-surface samples.
To induce sensitization, freestanding films for each sputtering rate were heat treated at
100˚C under vacuum for 7 days [114]. Samples were cleaned both with isopropanol and
ultrasonically in ultrapure deionized water, and then placed between two glass filter disks in
individual beakers of artificial seawater for 10 days. The artificial seawater consisted of 3.5%
NaCl in ultrapure water, with a pH of 8.0 ± 0.1 and a temperature of 19 ± 2˚C. The pH and
temperature were monitored daily and the solution was replaced every two days. Finally,
samples were cleaned with DI water and SEM images were obtained of the top surface of each
sample, along with EDS scans to determine sample composition.

56
4.2 Characterization of the Grain Boundaries and Preliminary Corrosion Analysis
Figure 34 shows characteristic bright-field and dark-field TEM micrographs of the crosssections of the as-sputtered samples. Figure 34a-c present the bright-field micrographs while
Figure 34d-f are the corresponding dark-field micrographs for the samples sputtered at 1, 3.5,
and 7 nm s-1 respectively. The samples, in that order, are henceforth referred to as samples A, B,
and C. The particular columnar grain structure is characteristic of sputtered samples, and the
growth direction is highlighted by the arrow [13]. The measured magnesium compositions of the
as-sputtered samples from slowest to fastest were 4.6, 4.1 and 4.1 wt.% Mg, which are
comparable to the composition of an Al 5083 alloy.
Figure 34: Cross-sectional TEM micrographs of as-sputtered Al–Mg samples, noting grain
widths, shown as a–c) brightfield and d–f) dark-field sputtered at: 1.0, 3.5, and 7 nm s-1
as
labeled above each column. Growth direction is indicated by the red arrow on the bright-field

57
micrographs. All samples present a Mg content of 4 to 5 wt%, comparable with that of an
Al5083 alloy as determined by EDS (not shown).
To further investigate the microstructure, EBSD and TKD were performed on each assputtered sample as shown in Figure 35, where the samples are divided by their corresponding
sputtering rate. The pole figure on the left side identifies overall sample texture as mostly (111),
which can be observed in Figure 35a-f. For all samples, special grain boundaries from Σ3 to Σ27
are shown in red, while black areas of the scan identify regions that were not completely
indexed. Figure 35a-c are top-surface EBSD maps of samples A-C, respectively. Figure 35d-f are
TKD cross-sectional scans, where the columnar grain structure is similar to the TEM shown in
Figure 34.
Figure 35: Overview of EBSD and grain size of as-sputtered Al-Mg samples, a–c) top-surface
EBSD, d–f ) cross-sectional EBSD (TKD) where red boundaries are special boundaries, whereas

58
black are unindexed. h–j) The average grain size. Corresponding sputtering rates noted on top of
each column.
The average grain size (top-surface) and the CSL fractions from all top-surface and crosssectional scans are summarized in Table 1, including percentages for the top 3 most common
Σ boundaries in all samples. Figure 35g-35i show the corresponding grain size distributions, with
average grain sizes of 95 nm (sample A), 210 nm (sample B) and 306 nm (sample C). To
confirm the EBSD-measured grain sizes, cross-sectional grain width measurements were
performed by TEM, with resulting values of 140, 190, and 230 nm for samples A-C respectively.
From these results, there is a clear trend between sputtering rate and grain size, which has been
corroborated by other studies reporting increased boundary mobility at higher sputtering rates
and thus larger grains [115, 116].
Table 1: CSL Fractions and Average Grain Size Measured by EBSD
4.3 The Effects of Special Grain Boundary Fraction and Sensitization on Corrosion
Resistance
Distinct trends were observed in CSL fractions across the three samples in both quantity
and grain boundary type. Samples A and B had similar total special boundary fractions of 0.30
and 0.24 for the top-surface, and the same fraction of 0.20 for the cross-section, while sample C

59
exhibited a marked increase in special boundaries with a fraction of 0.46 for the top-surface, and
0.86 for the cross-section. Overall, the top-surface Σ value for sample C was ~1.5x that of the
fraction of samples A or B, while the fraction for the cross-section increased by more than 4x.
The fractions of specific special GBs varied both between samples and scan types as shown in
Table 1. The most frequent individual special boundaries in sample A were Σ3 and Σ7
boundaries on the top-surface and Σ13b in the cross-section. For sample B, the most prevalent
boundaries were Σ3 on the top-surface, and Σ3, Σ7, and Σ21a in the cross-section. In sample C,
the greatest individual fractions for both EBSD and TKD were for Σ7 boundaries. During grain
growth, the grain boundary mobility and a driving force (surface free energy for films) steer
crystallographic properties such as GB character and texture, along with grain morphology, and
the resulting texture influences the preferred Σ boundaries [18, 27]. For highly textured materials
such as the samples in this study, low misorientation angles at the boundary are statistically more
favorable because individual boundaries are restricted by the limited available crystallographic
orientations of the neighboring grains [117]. In the (111) texture, the particular geometric
constraints make the Σ3, Σ7, Σ13b and Σ21a boundaries some of the most energetically favorable
[18, 118]. However, the high fraction of Σ7 boundaries across all samples was unexpected since
they are highly mobile compared to Σ3 and therefore less likely to be present in the final
sputtered material [119]. In sample C, the prevalence of Σ7 boundaries is tied to increased
boundary migration as the high sputtering rate introduced additional thermal energy into the
system, leading to higher temperatures and greater strain in the sputtered material [35, 120, 121].
Specifically, it is expected that sample C had an average deposition temperature over 125 ˚C
higher than samples A and B [120]. For Σ3 boundaries, higher sputtering rates simultaneously
promote both nucleation and uncoupling from the grain boundaries, and in this instance the

60
resultant boundary migration induced increased boundary decomposition and grain growth,
leading to higher energy Σ7 boundaries in the microstructure [120, 121]. Similarly, this higher
sputtering rate can also account for the significant difference in the observed CSL fractions
between the top-surface and cross-section for sample C: due to mobility as stated above, or
through increased stress relaxation from the interrupted sputtering process, which is more
energetically favorable for the formation of special boundaries [35, 121]. As the film grew, the
high deposition rate increased the temperature at the free surface where new growth was
occurring, promoting even greater boundary mobility and diffusion and activating more
boundary annihilation towards the top of the sample, yielding fewer CSL boundaries [80, 120-
122]. Furthermore, it is also possible that the Σ3 and Σ7 boundaries possessed similar energies
when they appeared together, as has been previously observed by Tschopp et al, as the grain
boundary plane orientation can produce large variations in GB energy even for boundaries with
the same Σ value [9, 123-125].
Beyond the particular types of boundaries observed, it is also important to address the
overall fraction of special boundaries in samples A-C while accounting for both composition and
sputtering rate [35, 120, 126, 127]. For example, the formation of nanotwins and special GBs has
been shown to be inversely proportional to the stacking fault energy (SFE), and is expected to be
limited in Al-and Al-Mg alloys due to their high SFE values (> 160 mJ m-2
) [35, 127, 128]. The
samples in this study have minimal compositional variation (between 4.1-4.6 wt.% Mg) and thus
SFE is not a likely contributor to the observed microstructural variations. To understand the
changes in special GBs amongst the three samples, one can look at a simplified model for Σ3
boundaries only. In Al-alloys, a previous study demonstrated that sputtering rates ≥ 7 nm/s led to
high fractions of twinned grains (46-70%), which agrees with the observed high fraction of

61
special boundaries for sample C. This increase in twin boundaries was attributed to both the high
sputtering rate and the decrease in GB mobility due to stress relaxation during film growth,
present during interrupted sputtering, making twins and special GBs more energetically
favorable [35, 120, 129]. In samples A and B, the changes in sputtering rate did not yield a
significant change in the GB fraction, which is likely due to the probability (r) of twin boundary
nucleation during film growth and its relationship to the deposition flux, where flux is
proportional to the sputtering rate [120]. For Al, an updated twin probability model by Bufford et
al calculated a 3% difference in the value of r between deposition rates of 1 and 3.5 nm s-1
, and a
continued downward trend in r beyond 4 nm s-1 that corresponded to higher probabilities of twin
nucleation [130]. Therefore, the present observed experimental special GB fractions for 1 and 3.5
nm s-1 are in agreement with the model, and the overarching trends for special boundaries are
consistent with previous studies that can be directly correlated to the sputtering rate, although
this is oversimplified for Σ3 boundaries and does not directly correspond to Σ7.
Figure 36: Top-surface SEM micrographs post-sensitization and immersion corrosion in artificial
seawater for samples a) 1.0, b) 3.5, and c) 7 nm s-1
. Cracking is visible in (a) and (b), denoted by
blue arrows; (c) a passive oxide layer.

62
Thus, the results from the microstructural analysis confirm that a high sputtering rate can
increase the special boundary fraction in 5xxx series Al alloys. This, in turn, is expected to
minimize sample sensitization and improve corrosion resistance [13]. To verify this, lowtemperature heat treatments and immersion corrosion studies were performed. Figure 36 shows
representative SEM micrographs of the top surface of each sample after immersion corrosion, for
samples A-C. Note that as-sputtered alloys typically have low surface roughness on the order of
0.5 nm, and SEM micrographs are typically not included due to a lack of contrast on the film
surface and low signal [72, 92, 131, 132]. Figure 36a-b (samples A and B) exhibit cracks and a
few discrete pits along the top surface due to stress corrosion cracking, indicated by blue arrows
[28, 83]. While samples A and B had different deposition rates, they exhibited similar fractions
of special boundaries, which correlates with similar degrees of visible surface corrosion [13, 28].
The observed corrosion results are due to the large percentage of non-special boundaries, which
provide potential nucleation sites for the anodic ß-precipitates that can lead to intergranular
corrosion and stress corrosion cracking [13, 30-32, 113]. In contrast, Figure 36c (sample C)
displays a continuous passive oxide layer in lieu of cracking or pitting, due to its high special
boundary fraction [133, 134]. Other studies have shown that samples with high fractions of low
Σ boundaries tend to have better corrosion resistance [13, 135, 136]. Overall, these trends in
corrosion behavior are consistent with the total measured Σ fraction in each sample, particularly
the high sputtering rate of 7 nm s-1
. Further analysis regarding the types of Σ boundaries under
different conditions would provide insight into the corrosion behavior.
4.4 Conclusions
This study provides direct evidence for the relationship between high sputtering rates and
an increased fraction of low Σ (Σ ≤ 27) GBs for a high SFE Al-Mg alloy. Specifically, a

63
sputtering rate of 7 nm s-1 demonstrated ≥1.5x the special GB fraction as compared to lower
sputtering rates in a 5xxx series Al alloy. The increase in both deposition temperature and strain
in this sample promoted the formation of more energetically favorable low-Σ grain boundaries.
Immersion corrosion tests confirmed improved top-surface corrosion resistance in conjunction
with the improved fraction of special GBs. Thus, through a combination of tuning the magnetron
sputtering parameters and interrupted deposition, the amount of special boundaries in a sputtered
Al sample can be increased to a degree that shows promise in improving the corrosion behavior
of 5xxx series Al-Mg alloys.

64
Chapter 5. Progression of the dealloying front in bilayer Cu-Al and Cu-Zn
nanoporous foams
*A version of this work was published in Journal of Materials Research in 2023 as
Progression of the dealloying front in bilayer Cu-Al and Cu-Zn nanoporous foams. For the
full paper, see reference [137].
The role of interfaces and the controlling synthesis parameters of graded dealloyed
nanoporous metallic materials are investigated, focusing on the dealloying front progression in
complex precursor materials with multiple alloy compositions. Specifically, the effects of
relative density and chemical potential on the dealloying front in sputtered bilayer copper alloy
films are explored with two case studies: Cu-Al/Cu-Al and Cu-Al/Cu-Zn. Cross-sectional
scanning electron (SEM) micrographs and energy dispersive x-ray spectroscopy mapping trace
the dealloying front across three time intervals, while top surface and cross-sectional SEM probe
the final dealloyed foam morphology. Final ligament sizes were found to be independent of the
synthesis parameters (21-28 nm), due to a combination of fast reaction times and phosphateinhibited surface diffusion of Cu atoms. The chemical potential gradient yielded faster reaction
times, whereas slower reaction times and a higher at.% of Cu in the top layer of precursor
material produced a more uniform morphology.
Figure 37: Graphical abstract, depicting a) the as-sputtered film for configuration 2, and the
subsequent progression of the dealloying front in concentrated H3PO4 after b) 30 minutes, c) 45
minutes and d) the “complete” time point.

65
5.1 Hierarchical and Graded Nanoporous Metallic Materials Through Dealloying
Hierarchical porous materials are an emerging field of study for their utility in
applications where the capacity to have a range of feature sizes within the same component is
advantageous [48, 138, 139]. Hierarchical or gradient nanoporous materials, which allow for
multiple pore sizes in the same network of material and possess one or more nanoscale features,
are of high interest for their enhanced Raman scattering, as well as their applications as
multifunctional sensors or catalysts due to their ability to combine ion transport via large pores
with the increased surface area provided by smaller pores [47, 60, 67]. Within the realm of
nanoporous (NP) materials, NP metal foams are of particular interest for properties such as low
relative density and high surface area to volume ratio, making them good candidates for battery
materials [7-10]. While much of the work on dealloyed nanoporous metal foams to date has
focused on noble elements such as Au, Ag, or Pt, cheaper and more available alternatives are
also being explored [11-14]. Dealloyed NP Cu is of particular interest for its applications in areas
of green energy research such as CO2 conversion, where it both increases the reaction efficiency
and selectivity for specific hydrocarbon reaction products [13].
While there are a variety of processing methods able to synthesize gradient NP structures
including additive manufacturing, templating, and sintering, dealloying is the most direct method
for developing nanoporous metallic materials [47]. Dealloying, the selective corrosion of one or
more elements in an alloy to generate a continuous porous structure, is notable because it can be
performed on a variety of precursor shapes (eg. films, ribbons, rods) and can be used to produce
pore sizes on both the micro- and nano-scales, enabling tunability in both connectivity and
specific surface area [65]. Nevertheless, synthesizing graded NP materials via dealloying is
challenging due to factors such as the alloy composition, the initial microstructure, and the

66
changing reaction kinetics. For example, precursor alloys previously used for graded dealloyed
samples contained phases such as intermetallics, which dealloyed in unpredictable ways [47,
140, 141]. Specifically, when a less noble phase preferentially corrodes and the atoms enter
solution, the composition and electrochemical potential of the dealloying medium change,
inducing changes in the reaction kinetics over time [141, 142]. The reaction kinetics, and the
subsequent progression of the dealloying reaction front, can be further altered through changes to
the reaction time or temperature [66, 143]. Additional compositional factors such as the parting
limit of the alloy, or minimum quantity of the more noble component for which the reaction will
occur, and the potential difference between all alloy constituents can influence both the evolution
of the dealloying front and the resulting porous microstructure [38, 39]. Approaches such as
combining dealloying with sintered materials and various multi-step dealloying processes have
also been investigated in attempts to synthesize graded nanoporous materials from less complex
precursors [64, 67, 144]. Preliminary synthesis of NP Cu foams with a pore size gradient through
the thickness has been demonstrated for a bilayer film, where the layer order at the compositional
interface influenced the resulting foam morphology [59]. However, the effects of the layer order
and the type of interface on the dealloying reaction front have yet to be addressed. The
fundamental mechanisms controlling the synthesis of graded NP materials are not understood,
and further work is needed to better define the evolution of the dealloying front for precursor
materials with multiple alloy compositions [52, 58].
In this study, we investigate the evolution of dealloying fronts in similar and dissimilar
alloy compositional interfaces in gradient nanoporous copper foams with precursor materials of
various Cu-Al and Cu-Zn compositions. Specifically, we examined two distinct cases: a change
in relative density (similar interface) and a change in chemical potential (dissimilar interface).

67
Four configurations of bilayer copper alloy films were synthesized via magnetron sputtering due
to its flexibility in precursor materials and control over composition [77]. The as-sputtered films
were imaged with scanning electron microscopy (SEM) before and after dealloying at a range of
time intervals. Subsequently, SEM and energy dispersive x-ray spectroscopy (EDS), were
performed in order to analyze the evolution of the dealloying front and the foam morphology
over time. Vickers indentation was performed to further analyze the effects of the layer order and
type of interface on foam morphology. Overall, this work addresses critical dealloying
parameters and the effect of interfaces for developing gradient nanoporous materials.
Three copper alloys with a reduction potential difference of 0.5V or greater were selected
in this work, including Cu20Al80, Cu35Al65, Cu20Zn80. Bilayer films with individual 1µm layer
thicknesses (total film thickness of 2 µm) were synthesized via DC magnetron sputtering.
Custom alloy targets were sputtered at 50 W, onto Si <110> wafers at a base pressure of 3.1 x
10-6 to 3.3 x 10-6 Pa from 5.08 cm (2 in) sources. Four layer configurations were synthesized:
Cu20Al80/Cu35Al65, Cu35Al65/Cu20Al80, Cu20Al80/Cu20Zn80 and Cu20Zn80/Cu20Al80. These are
referred to in this work as configurations 1-4, respectively (see Table 2).
The cross-sections of the as-sputtered films were prepared by wire saw (Princeton
Scientific WS-25) or by manual fracture and were characterized with SEM (Helios G4 PFIB
UXe DualBeam FIB/SEM). The as-sputtered films were dealloyed by free corrosion in
concentrated H3PO4 at room temperature for 30 minutes, 45 minutes, and until the reaction
reached the “final step,” defined here as the last time interval before sample disintegration. All
dealloyed samples were cleaned three times with DI water and dried for a minimum of 12 hours.
The films were not removed from the substrate, and cross-sections of the dealloyed samples were
prepared by fracture. SEM micrographs were obtained for both the NP foam top surface and

68
cross-sections, while EDS (Oxford UltimMax 170 Silicon Drift Detector) composition maps
were performed solely on the cross-sections. Cross-sectional ligament sizes after the final time
interval were processed by ImageJ.
Vickers indentation was performed with a Leco LM-100 Vickers indenter at 10 grams of
force and for a dwell time of 10s. Indentation testing was performed on the top surface of the
four dealloyed films at the “final step” time for each film. Post-fracture morphology of the
indents was analyzed with top-surface SEM imaging.
5.2 Analysis of the Dealloying Front
5.2.1 Characterization of the As-Sputtered Films
For the dealloying reaction to occur, alloys must have a large electrochemical potential
difference between constituent elements such that one component preferentially dissolves into
solution [145, 146]. Additionally, there must be a high enough atomic fraction (the parting limit)
of the less noble constituent to prevent the more noble atoms from forming a protective passive
layer during the initial stages of corrosion [38, 39]. Thus, three binary copper alloy precursors
were selected for this work: Cu20Al80, Cu35Al65 and Cu20Zn80. Cu, Al and Zn have reversible
reduction potentials of 0.340V, -1.66V and -0.762V, respectively, with potential differences
between any two of the three elements above the threshold necessary for dealloying [37]. These
individual compositions have previously been shown to form different ligament sizes with
relative densities of 20%, 35% and 20%, respectively, in the final dealloyed NP foam [46].
Therefore, the selected alloys were chosen to produce a gradient in relative density after
dealloying or a gradient in chemical potential.
As mentioned in the experimental procedures, bilayer samples in four configurations
were sputtered as solid films, which are shown in the cross-sectional SEM micrographs in Figure

69
38. Two distinct columnar layers are visible in each micrograph, corresponding to the two layers
in each as-sputtered film, where Figures 38a-b are configurations 1 (Cu20Al80/Cu35Al65) and 2
(Cu35Al65/Cu20Al80) and the columnar grains in both layers are packed tightly together. In
contrast, for Figures 38c-d, configurations 3 (Cu20Al80/Cu20Zn80) and 4 (Cu20Zn80/Cu20Al80),
there are noticeable columnar voids between grains. Previous sputtering research on Cu-Zn
alloys observed that alloys with a high fraction of Zn atoms had reduced adatom mobility during
sputtering, inhibiting the growth of a more uniform film and producing a voided columnar
structure [115, 147-149].
Figure 38: SEM micrographs of the fractured as-sputtered films of (a) configuration 1, (b)
configuration 2, and cut as-sputtered films of (c) configuration 3 and (d) configuration 4. The
orange labels in the top right corner indicate sample configuration. The black arrow on the right
marks the film growth direction.

70
5.2.2 The Dealloying Front
In the dealloying process, the dealloying front is the region where the reaction is actively
occurring [59]. Along this boundary, the less noble component preferentially corrodes into
solution while atoms of the more noble component reorganize into a bicontinuous connected
structure and expose the unreacted solid surface. For a reaction in an electrolytic solution, this
begins at the sample/solution interface, and progresses into the sample as the planar interface
between the solution and the solid and porous zones of the sample [24, 150]. In order to capture
the progression of the dealloying front, both HCl (electrochemical dealloying) and H3PO4 (free
corrosion) were tested for the film configurations in this study. Electrochemical dealloying in 5
wt.% HCl yielded a reaction time of less than five seconds, which made it unfeasible to retain
analyzable samples for films of this thickness (2 µm) and precluded examination of the
dealloying front. The accelerated reaction time can be attributed to the chloride ions, which have
been shown to augment surface diffusivity in both Cu-Al and Cu-Zn alloys, increasing the speed
of the reaction front [141, 151]. In contrast, the presence of phosphate ions in the dealloying
solution for Cu alloys has been shown to inhibit surface diffusion of Cu, resulting in both slower
progression of the dealloying front and smaller, smoother ligaments [151]. Consequently, free
corrosion in concentrated H3PO4 was selected as the method of choice for this study as the
reaction time was significantly slower, and samples could be obtained at intervals to examine the
reaction progression.
The evolution of the dealloying front is shown in Figures 39 and 40, which exhibit crosssectional SEM micrographs and their corresponding EDS composition maps across 3 different
time intervals: 30 minutes, 45 minutes, and “final step”. “Final step” was defined as the last time
interval before a sample of a given configuration dissolved in the electrolyte. For all four

71
configurations, the main planar dealloying front fully progressed through the films before either
layer had fully reacted, yielding higher levels of either residual Al or Zn as shown in Table 2.
Table 2: Sample Composition, Ligament Size, and Relative Density
Sample
Name Alloy System Composition at Final Step Ligament Size Relative Density at
Final Step
Top
Layer
Bottom
Layer
Top
Layer
Bottom
Layer
Top
Layer
Bottom
Layer
C1 Cu20Al80/Cu35Al65
89 at.% Cu
11 at.% Al
87 at.% Cu
13 at.% Al 26 nm 21 nm 22 % 40 %
C2 Cu35Al65/Cu20Al80
76 at.% Cu
24 at.% Al
65 at.% Cu
35 at.% Al 28 nm 23 nm 43 % 27 %
C3 Cu20Al80/Cu20Zn80
61 at.% Cu
39 at.% Al
88 at.% Cu
12 at.% Zn 22 nm 28 nm 28 % 22 %
C4 Cu20Zn80/Cu20Al80
30 at.% Cu
70 at.% Zn
52 at.% Cu
48 at.% Al 23 nm 25 nm 34 % 30 %
The relative density �!"#$%&'" for each layer at the last time step was calculated with the
following equation:
�!"#$%&'" = �!" ∗ 1 + �
100
where nCu was the expected number of Cu atoms per 100 atoms of precursor alloy, and R was the
residual fraction of Al or Zn for that layer. The red dashed lines on each SEM micrograph
indicate the position of the dealloying front at that time point. On the EDS maps, Cu and Al are
marked with yellow and red, respectively, and Zn is indicated with blue. In order to isolate the
effects of porosity, composition, and chemical potential, the dealloying front is analyzed for two
cases. Figure 39 presents configurations 1-2 (C1 and C2), bilayer films of different compositions
but the same elements and an expected difference in relative density between layers, while
Figure 40 displays configurations 3-4 (C3 and C4), with an expected difference in potential and
pore/ligament size between layers.

72
5.2.2.1 Effect of Relative Density: Cu-Al/Cu-Al Compositions
Porous materials (foams) can be categorized by both their pore/ligament size and their
relative density (foam density divided by the bulk density of the material), which is inversely
related to the porosity [70]. The expected relative densities based on compositions for this study,
when fully dealloyed, are either 20% (80% porosity) or 35% (65% porosity), with ligament sizes
after free corrosion dealloying in the range of 50-60 nm for a single-composition sample [46,
151]. The calculated relative densities for each layer after the final dealloying time step are in
Table 2, and are greater than the expected relative densities as the objective of this study was to
fully observe the reaction over time, not to have the reaction run to true completion. Thus, the
“final step” dealloying time point was defined as the last time point at which there was a viable
sample that could be retained for further testing, which does not necessarily coincide with when
the dissolution reaction would have run to completion. Thus, the “final step” dealloying time
point was defined as the last time point at which there was a viable sample that could be retained
for further testing, which does not necessarily coincide with when the dissolution reaction would
have run to completion. Dealloying is a diffusion-based process, which is generally slow, and
dealloying reactions may take multiple days to reach completion [152, 153]. However, due to the
fast reaction times, there was not enough time for the residual Al or Zn atoms to exit the
structures, and longer dealloying times did not produce usable samples that could be analyzed.
For simplicity of discussion, samples will be referred to as per the expected relative density. In
Figure 39, configuration 1 has an expected 80% porosity in the top layer and 65% porosity for
the bottom layer, while configuration 2 has the inverse setup. This provides two distinct
scenarios for understanding the reaction front as a function of the porosity, facilitating or
restricting the flow of the electrolyte.

73
The dealloying reaction front between Cu20Al80 and Cu35Al65 proceeded fastest when the
lower-Cu fraction was the top layer, defined here as the layer that was in contact with the
solution first. This can be seen after 30 minutes where the dealloying front progressed ~440 nm
for Figs. 39a-b (C1), compared to ~240 nm for Figs. 39c-d (C2). As the quantity of Cu atoms in
the surface in contact with the electrolyte in Figs. 39a-b was closer to the parting limit, there was
less reorganization of Cu atoms during dealloying, which increased the degree of etching along
the columnar GBs [154]. After 45 minutes, the reaction front in Figs. 39e-f slowed as it entered
the Cu-rich layer (~640 nm traveled in total), while the speed of the reaction front in Figs. 39g-h
increased as it entered the Cu20Al80 layer (~970 nm in total). Here, the decrease in the
concentration of Cu atoms in the bottom layer of configuration 2 facilitated the advancement of
the H3PO4 electrolyte into the film, as the increased porosity provided more channels for the Al
ions to leave the electrolyte/alloy interface. Configuration 1 displayed the opposite [24, 154].
After 105 and 120 minutes, the dealloying front advanced to the bottom of the film for the two
configurations in Figs. 39i-j and 39k-l respectively, though at a slower pace than the earlier time
intervals. At this late stage of the reaction, the presence of Al atoms in solution changed the
electrochemical interactions between the H3PO4 electrolyte and the remaining portion of the film
[47]. Additionally, the foam morphology was expressed differently depending on the layer order.
For configuration 1 the reduced reassembly of Cu atoms during dealloying in the top layer
preserved the macroscopic columnar structure [151]. While for both configurations, the Cu35Al65,
layer (Figs. 39i and 39k) show a more continuous connected porous structure as there was a
greater degree of rearrangement for the Cu atoms. For configuration 2 specifically, the higher
fraction of Cu atoms in the top layer enabled rearrangement to occur earlier in the reaction,
producing a more isotropic nanoporous structure through both layers [154].

74
Figure 39: SEM micrographs of Cu-Al bilayer films dealloyed in H3PO4 via free corrosion and
corresponding EDS maps for configuration 1 (Cu20Al80/Cu35Al65) and configuration 2
(Cu35Al65/Cu20Al80) at various time intervals. The red dotted line marks the position of the
dealloying front after 30 min (a-d), 45 min (e-h), and “final step” at 105 min (i and j), and 120
min (k and l). EDS map indicates both Al and Cu content.
5.2.2.2 Effect of Chemical Potential: Cu-Al/Cu-Zn Compositions
In configurations 3 and 4, the expected relative density (20%) was the same for all layers
but there was a difference in chemical potential between the selected Cu-Al and Cu-Zn
compositions of 2.0V (Cu/Al) and 1.102V (Cu/Zn) [37]. This difference in reduction potential
between constituents is related to the dissolution rate, and is large enough that the Al and Zn are
preferentially soluble and the Cu can diffuse along the electrolyte/film boundary layer [47, 146].

75
After 30 minutes in solution, the reaction front through configuration 3 (Figs. 40a-40b)
had progressed halfway through the Al-rich layer (~530 nm), while the front for configuration 4
(Figs. 40c-d) progressed through the Zn-rich and into the Al-rich layer (~1400 nm). After 45
minutes (Figs. 40e-h), the reaction front in configuration 3 (Figs. 40e-f) had traveled ~650 nm in
total, while the reaction front in configuration 4 (Figs. 40g-h) had advanced through most of the
Al-rich layer, ~1700 nm total. For configuration 4, dealloying times of greater than 45 minutes
led to sample disintegration, whereas for configuration 3, the reaction front progressed almost to
the layer interface after 60 minutes, with no viable samples at longer time intervals. In Figs. 40i
and 40g, both configurations displayed etching along and within the columnar grains of the top
layers, as the spaces between the columnar grains provided a pathway for the H3PO4 etchant to
penetrate the film vertically through the thickness before lateral etching at the surface in contact
with the electrolyte was completed [154]. This was most pronounced in configuration 4 (Fig.
40g), and these results are corroborated by the original as-sputtered microstructure in Fig. 38d,
which is strongly columnar with voids between individual columns. For configuration 3, which
had more voids in the bottom layer of the as-sputtered film (Fig. 38c), the increase in available
etchant pathways once the reaction front reached the compositional interface facilitated the
progression through the bottom layer [154]. Furthermore, the Cu20Zn80 alloy (bottom layer of C3
and top layer of C4) had particularly fast dealloying kinetics and therefore a low critical potential
threshold to initiate the reaction [38].

76
Figure 40: SEM micrographs of Cu-Al/Cu-Zn bilayer films dealloyed in H3PO4 via free
corrosion and corresponding EDS maps for configuration 3 (Cu20Al80/Cu20Zn80) and
configuration 4 (Cu20Zn80/Cu20Al80) at various time intervals. The red dotted line marks the
position of the dealloying front after 30 min (a-d), 45 min (e-h), and “final step” at 60 min (i and
j). EDS map indicates Al, Cu and Zn content.
An additional factor in the reaction progression is the reduction potential difference
between the Al and Zn atoms, which was -0.0898V [37]. Work on ternary alloys for dealloying
has shown that for a one step reaction, the presence of a third element that is electrochemically
active in the selected dealloying medium will change both the dealloying process and the
resulting porous microstructure [155]. While configurations 3-4 in this investigation were
bilayers of two distinct compositions and not ternary alloys, the presence of both Al and Zn
atoms in solution once the reaction front reached the alloy interface changed the electrochemical
potential of the unreacted film in that medium, increasing the speed of the front through the
remaining film [47, 155]. Thus, the reaction mechanism after the dealloying front reached the

77
compositional interface was analogous to that of ternary alloys, and can be applied to complex
dealloying precursors. In the next section we will discuss the effects of both the chemical
potential interface and the relative density interface on the resulting dealloyed foam morphology.
5.2.3 Foam Morphology
To illustrate the effects of relative density and chemical potential on the resulting foam
morphology, top-surface and cross-sectional SEM micrographs were taken of each configuration
after dealloying, at the “final step” time point. Figures in the left column (41a, 41c, 41e, 41g)
show the representative top-surface morphology while figures in the right column (41b, 41d, 41f,
41h) show the corresponding cross-sectional morphology for configurations 1 thru 4. For
configurations with a composition of 20 at.% Cu in the top layer (C1, C3, and C4), the limited
reorganizational capacity of the Cu atoms promoted dealloying along the grain boundaries in
addition to within the grains. For configuration 2, with Cu35Al65 as the top layer, the grain
structure at the top surface is not visible, and the characteristic isotropic porous structure is not
constrained by the original grains as there was more reorganization during the reaction [154].
This isotropic porous structure is also visible in the cross-section of configuration 2 (Fig. 41d). In
contrast, the top surface of configurations 1, 3 and 4 (Figs. 41b, 41f, 41h) exhibited islands of a
more isotropic structure surrounded by voids, corroborating that etching proceeded both along
the GBs and within the grains [58, 154]. The expected ligament size for all samples after free
corrosion was in the range of ~50-60 nm but in actuality the final foam ligament size between
the four sample configurations ranged from 21-28 nm [46, 151]. While the ligament and pore
size of the final nanoporous structure generally changes with composition, it can also be tuned
with the choice of electrolyte [68, 145, 156, 157]. In this case, a combination of a fast reaction

78
time and phosphate-inhibited surface diffusion for the Cu atoms prevented ligament coarsening,
as there was minimal time for the Cu to aggregate, producing smaller ligament sizes [58, 151].
Figure 41: Representative SEM micrographs of foam morphology for top surface (left column)
and cross-section (right column), taken at the final (“final step”) dealloying time interval.
Configuration 1 (a-b), configuration 2 (c-d), configuration 3 (e-f) and configuration 4 (g-h). The
orange labels in the top left corner indicate sample configuration.
In order to further assess the distinctive surface morphologies, Vickers indentation was
performed on the top surfaces as shown in Fig 42. Representative top-surface SEM micrographs
of the indented samples for configurations 1, 2, and 3, respectively are shown in the left column
and zoomed-in micrographs depicted in the right column. Both configurations 1 (Figs. 42a-b)

79
and 3 (Figs. 42e-f), which displayed top-surface intergranular cracking in Fig. 41, showed
deformation that was contained to the indent. The deformation did not propagate, as the material
was not continuous. However, there were slight differences after indentation, and the most
visible cracking was in configuration 1, which had both the largest voids and a mismatch in
relative density between the top and bottom layers [75]. Thus, the bottom layer has a qualitative
effect on the deformation behavior of the bilayer film, and configuration 3, which did not have a
relative density gradient, appears less brittle. There was no comparison for configuration 2 (Fig.
42c-d), which was a better-connected porous material (Fig. 41c) with a more isotropic foam
morphology that showed no intergranular cracking, and instead displayed cracking along the
corners and sides of the indent [56]. Configuration 4 is not shown, as the sample was too brittle
for testing. Future work should include heat treatments prior to dealloying in order to induce a
less sharp compositional boundary between layers, and to reduce the effects of the original
columnar grain structure on the morphology of the dealloyed foams.

80
Figure 42: Representative SEM micrographs of top-surface Vickers indents performed at the
“final step” dealloying time interval (left column) where the blue boxes in the left column denote
the location of respective zoom-ins (right column) of the edge of the indent. The orange labels in
the top left corner indicate sample configuration as follows: configuration #1 (a-b), configuration
#2 (c-d), configuration #3 (e-f) Configuration #4 not shown as sample was too brittle for testing.
5.3 Conclusions
This work explores the effect of relative density and chemical potential interfaces on the
dealloying front and morphology of gradient NP copper foams. Specifically, bilayer nanoporous
films were synthesized via free corrosion dealloying of Cu-Al/Cu-Al (C1, C2) and Cu-Al/Cu-Zn
(C3, C4) precursors in concentrated H3PO4, and analyzed across three different time points. All
configurations with a lower at% of Cu in the top layer yielded faster initial progression of the
dealloying reaction front and etched voids along the columnar GBs, due to reduced
rearrangement of Cu atoms. The configuration containing the Cu-rich top layer produced an

81
isotropic NP structure. In contrast, for chemical potential interfaces, the inherently fast
dealloying kinetics of the Cu20Zn80 alloy and the electrochemical interactions between Al and Zn
atoms in solution facilitated faster reaction times while the large voids between grains in the
precursor films increased etching along the GBs. Slower reaction times produced a more uniform
morphology, which should be considered when selecting both precursor alloys and layer order.
By examining both the effects of relative density and chemical potential, this study highlights
critical factors for the synthesis and development of these graded structures.

82
Chapter 6. SERS Response of Gradient Nanoporous Cu
Nanoporous copper is a promising surface-enhanced Raman scattering (SERS) substrate
alternative to Au, Ag and Pt. Here, gradient nanoporous Cu was synthesized from films sputtered
from Cu-Al and Cu-Zn alloys that were compositionally graded via heat treatment during
deposition. The films were then processed by dealloying. Energy-dispersive X-ray spectroscopy
(EDS) and scanning electron microscopy (SEM) were performed to evaluate the composition,
porosity and foam morphology. SERS enhancement was evaluated with a 4-MBN probe, and
both dealloyed and as-sputtered films exhibited an enhancement factor ≥ 104
. The best
performing sample demonstrated enhancement on the order of 107
, with an optimal pore size
range of ~35-45 nm. Overall, the tailorability of these materials shows promise for
multifunctional Raman-based sensing.
6.1 Graded Nanoporous Metallic Foams For Sensing
Nanoporous metallic foams are notable for their high surface area and good electrical and
thermal conductivity, and are of interest for their range of applications, especially in energy and
sensing-related fields [39, 48, 60]. The unique capacity for microstructural features at multiple
different length scales in hierarchical porous materials enables the design and synthesis of more
complex multifunctional sensors, which require a combination of high surface area and larger
channels for gas and ion transport [67, 158]. For example, nanoporous metal foams are a popular
subset of stable surface-enhanced Raman scattering (SERS) substrates with good enhancement,
where gradient nanoporosity has been shown to further contribute to SERS signal amplification
[60, 159]. Thus, a controlled synthesis process, such as dealloying, is essential for reliable and
reproducible results, especially for complex and gradient nanoporous microstructures [160, 161].
Furthermore, despite many recent advances in SERS substrate development, there is still a need

83
for lower-cost substrates that will amplify chemical signals to readable levels while remaining
physically robust enough for multiple use cycles [106, 108]. Cost-effective alternative substrates
such as nanoporous Cu have shown comparable signal enhancement levels to other commonly
used, more expensive metals such as Au and Ag [162, 163].
SERS generally utilizes a metallic nanostructure to amplify the molecular bond vibration
signal from a nearby target molecule or analyte [108, 164]. A suitable plasmonically active
SERS substrate amplifies the target signal by a factor of 104-108, as the nanoscale surface
morphology enhances and concentrates the electromagnetic field [108, 165, 166]. SERS-based
sensing is of high interest to the scientific community as it enables swift and sensitive in situ
identification of a wide range of target molecules, including drugs, explosives, and toxins [104,
106, 165]. Critically, the SERS enhancement in nanoporous materials is dependent on the local
morphology and topology of the specific substrate, as specific features provide hot spots of
additional enhancement [108, 160]. Work by Qian et al on nanoporous gold demonstrated greater
amplification from rougher ligaments and pore sizes in the 5-10 nm range, while Kucheyev et al
obtained enhancement of ~109
-1011 in nanoporous gold with pore sizes of ~250 nm [108, 167].
Additionally, voids between nanoparticles or islands of porous material can further concentrate
the signal [107, 108]. Thus, gradient nanoporous substrates could lead to significant
enhancements for a more comprehensive selection of target molecules, due to the range of pore
sizes and geometries present in the structure [159].
This study investigates the use of gradient nanoporous Cu as a path forward for cheaper
and more robust SERS substrates, based on bilayer films of dealloyable copper alloys. Fully
dense films were heated during synthesis to produce a graded interface between Cu-Al and CuZn alloys, followed by dealloying at various intervals to produce varied morphology and

84
porosity. Scanning electron microscopy (SEM) and energy-dispersive x-ray spectroscopy (EDS)
were used to study the composition, morphology and surface features, while subsequent analysis
in ImageJ revealed specifics on the nanoporosity and microporosity. The resulting nanoporous
materials were then tested with 4-mercaptobenzonitrile (4-MBN) as the SERS probe, and
showed signal amplification greater than the 104 enhancement threshold [165]. Overall,
composition, layer sequence, dealloying, heat treatments and morphology are evaluated as
factors for developing new multifunctional nanoporous sensors.
The binary alloys used for this study were selected from dealloyable copper alloys with a
reduction potential difference of at least 0.5V between constituent elements, according to the
criteria discussed elsewhere [36, 137]. Bilayer copper films were sputtered from custom Cu20Al80
and Cu20Zn80 alloy targets at 50 W from 5.08 cm (2 in) sources with a base pressure of 1.4 x 10-5
to 1.0 x 10-6 Pa onto Corning Eagle aluminoborosilicate glass substrates. Two substrate
temperatures used during sputtering, 100˚C or 150˚C, with two-layer configurations:
Cu20Zn80/Cu20Al80 (both temperatures) and Cu20Al80/ Cu20Zn80 (100˚C only). The total film
thickness for all sputtered films was 2 µm, with 1 µm individual layer thickness. All films and
dealloyed foams were retained on the glass substrates.
As-sputtered films were dealloyed by free corrosion in 0.1M H3PO4 for times ranging
from 2.5-8 hours, rinsed three times with DI water and dried overnight. A summary of the assputtered and dealloyed samples is provided in Table 3, where sets A and B correspond to a layer
order of Cu20Zn80/Cu20Al80 with stage temperatures of 100˚C and 150˚C, respectively. Set C
corresponds to the inverse layer order (Cu20Al80/Cu20Zn80) and a stage temperature of 100˚C. The
numbers 1, 2, and 3 indicate the dealloying conditions of “as-sputtered”, “half-cycle” and “full
cycle”, respectively. Half-cycle was defined as the time point at which the dealloying reaction

85
front had propagated into the interfacial region, while a full cycle was defined as double that
time.
Table 3: Summary of As-Sputtered and Dealloyed Samples
SEM cross-sections for as-sputtered and dealloyed samples were prepared by plasma FIB
(Helios G4 PFIB UXe DualBeam FIB/SEM) with a protective Pt coating. EDS (Oxford
UltimMax 170 Silicon Drift Detector) composition maps were obtained for the cross-sections,
and SEM micrographs of the top surface and cross-sections were processed with ImageJ to
obtain the average ligament size and top surface microporosity. SERS was performed with a
Horiba XploRA with a 785 nm laser. The test molecule was 4-mercapto benzonitrile (4-MBN) in
ethanol at a concentration of 2mM. As-sputtered and dealloyed films from sample sets A and B
were immersed in the solution overnight to optimize surface coverage of the adsorbed 4-MBN
[168, 169].

86
6.2 Results and Discussion
6.2.1 Synthesis of Gradient Nanoporous Copper Via Dealloying
The dealloying process used to synthesize the selected materials relies on the potential
difference between constituent elements in an alloy to induce preferential corrosion in one
component in an electrolytic solution [24]. Work by Cheng confirmed that a minimum potential
difference of 0.5V between alloy constituents enables preferential corrosion in the dealloying
reaction, and this criterion was used to select the two precursor copper alloys for this study [36,
137]. In a previous study, the dealloying in bilayer films of two different families of alloys: one
Cu-Zn alloy and two Cu-Al alloys (Cu20Al80 andCu35Al65) were examined to study the effects of
relative density and chemical potential differences on the dealloying reaction front. In the
current work, Cu20Al80 and Cu20Zn80, which produced a heterogeneous nanoporous structure,
were selected as the base layers to study the grain boundary etching effects at various
temperatures and to measure the SERS response as a function of layer configuration and
morphology. Specifically, in this study, the film substrates were heated during sputtering to a
constant temperature of 100˚C or 150˚C in order to induce diffusion at the layer interface and
minimize voids, which exacerbate grain boundary etching effects [154, 170, 171]. A substrate
temperature of 300˚C was also attempted, but no compositional gradient was visible in the
resulting film, and it was not pursued further.

87
Figure 43: SEM micrographs and corresponding EDS maps of as-sputtered Cu20Zn80/Cu20Al80
films (A1, B1) and as-sputtered Cu20Al80/Cu20Zn80 films (C1). A1 and C1 have stage
temperatures of 100˚C, while B1 has stage temperature of 150˚C. Color on EDS map indicates
Al, Cu, and Zn content. Black arrow signifies film growth direction.
Figure 43 shows the cross-sectional SEM micrographs and EDS maps of the three
sputtered solid films: A1, B1 and C1. Al, Zn and Cu atoms are shown in red, blue, and yellow,
respectively. Film A1 demonstrated narrow vertical voids with wide columnar grains in the
Cu20Zn80 top layer, while B1 showed both larger voids and grains in the top layer and the
interfacial region. For C1, which had a Zn-rich alloy in the bottom layer, the voids between the
large columnar grains propagated from the initial film adhesion to the substrate, through the
interfacial region and into the top layer of the film. Zn-rich alloys have previously demonstrated
a tendency to form a voided columnar structure due to reduced adatom mobility during the initial
sputtering process [115, 147-149]. However, here, the additional heating on the substrate during
sputtering induced grain growth and increased mobility, leading to fewer voids than previously

88
observed [148]. The sample compositions before and after dealloying are shown in Table 4. All
as-sputtered films showed different compositions than those of the starting alloys, with the
largest differences in A1 and B1. This demonstrated that an interface was successfully formed as
a result of the additional surface diffusion induced by the heated substrate [171]. Note that film
C delaminated during dealloying, making it incompatible with the goal of a stable SERS
substrate.
Table 4: Sample Compositions Before and After Dealloying
In addition to inducing a graded composition in the precursor films, a dilute solution of
the H3PO4 electrolyte (0.1M H3PO4) was also used to further slow the dealloying reaction,
allowing more time for Al and Zn atoms to diffuse out of the film to create a more porous
structure [151]. Cross-sectional SEM micrographs with overlaid EDS maps for the dealloyed sets
A and B films are shown in Figure 44, where A2 and B2 are the half-cycle dealloying condition

89
while A3 and B3 are the full cycle. After dealloying, large columnar voids remained in the top
layer for A2 and A3, interspersed with porous columns. For A3 specifically, the bottom layer
showed a more continuous nanoporous structure with only mild propagation of the large voids
visible in Figure 43. In both A samples, the propagation of larger void columns into the
nanoporous structure showed that grain boundary etching was still the dominant dealloying
mechanism [151]. In sample B2, some of the large columnar voids in the as-sputtered film were
preserved, yielding discrete porous globules. The grey region between the yellow porous
structures in B2 is the Pt coating used in the FIB sample preparation. For B3, which had double
the reaction time of B2, these discrete porous pillars merged into a more continuous porous
structure with lateral voids. The change in void shape and decrease in overall number in sample
set B shows that the main reaction front during etching was successfully changed to a
combination lateral and GB etching mechanism [151]. The Cu fraction in the top layer for the
half cycle dealloyed samples (A2 and B2) was comparable, at just over 70%, as the etching of
discrete porous islands enabled easier dissolution of the Al and Zn [154]. The top layer of the full
cycle dealloyed samples showed a similar pattern for the Cu fraction, while the bottom layer of
B3 had a much higher residual Al fraction of 37% as the addition of lateral etching produced a
slower reaction than that of A3, with fewer pathways for the Al ions to enter solution [151, 154].

90
Figure 44: Cross-sectional EDS maps of A and B series Cu–Al/Cu–Zn bilayer films dealloyed in
H3PO4 via free corrosion. Top row for a half cycle of 2.5 hours (A2) or 4 hours (B2); bottom row
after a full dealloying cycle of 5 hours (A3) or 8 hours (B3). Left and right columns correspond
to sputtering stage temperature of 100˚C and 150˚C, respectively. Color on EDS map indicates
Al, Cu, and Zn content. Black arrow shows film growth direction.
The compositions after dealloying were then used to calculate the average cross-sectional
nanoporosity for A2, A3, B2 and B3 by taking the inverse of the relative density, which was
calculated using the following equation:
�!"#$%&'" = �!" ∗ 1 + �
100
where R is the residual fraction and nCu is the expected atomic fraction of Cu in the
precursor alloy [137]. The cross-sectional nanoporosity, top surface microporosity, and average
ligament sizes were obtained with ImageJ, and are shown in Table 5. Nanopore size can also be

91
interpreted from Table 5 as the size ratio between pore size and ligament size for dealloyed
samples is ∼1:1 [172].
Table 5: Ligament and Pore Size After Dealloying
Average top surface ligament sizes were ~35-45 nm, while cross-sectional ligament sizes were in
the ~20-45 nm range. Specifically, they were smaller than the ~50-60 nm average ligament sizes
obtained from single-composition samples, but larger than the ~20-30 nm average crosssectional ligament size obtained from unheated bilayer films of the same alloys [46, 137, 151].
Here, the 0.1M H3PO4 dealloying electrolyte and wide columnar grains in the precursor alloys
slowed the dealloying reaction such that the ligaments had time to coarsen [58, 137]. All samples
exhibited similar degrees of micro- and nanoporosity, with the highest values for both seen in B3
and A2. Microporosity was defined as the fraction of open space or voids between solid islands
on the top surface of each sample, and depends upon both the original microstructure of the assputtered film and the (if applicable) top surface morphology of the dealloyed sample. These
analyses demonstrate that controlled dealloying of this base set of graded films can be used to

92
produce gradient nanoporous Cu with tunable features, including porosity, grain morphology,
and ligament size.
6.2.2 SERS Response of Gradient Nanoporous Cu
All samples in sets A and B were evaluated as SERS substrates, with 4-MBN as a probe
(target molecule) and an ethanol solvent. The resulting SERS spectra are shown in Figure 45.
The solid lines (Fig. 45a) and dashed lines (Fig. 45b) indicate sample sets A and B, respectively.
Dealloying condition is designated via line color with the as-sputtered condition in red, “halfcycle” dealloying in green and “full-cycle” dealloying in blue.
Figure 45: Surface-enhanced Raman spectra of (a) set A samples, (b) set B samples, for substrate
with 4-mercapto benzonitrile in ethanol. Red is the as-sputtered condition, green is the “halfcycle” dealloy, and blue is the “full-cycle” dealloy.
The calculated enhancement factors (EF), all of which are above the 104 threshold for good
signal enhancement, are displayed in Table 6. The enhancement factor was calculated according
to the equation:

93
�� = �!"#!
�!"#$%&"'
�
�!"#$%&"'
�!"#!
where ISERS and Isolution are the signal intensities for the SERS result and the solution and Nsolution
and NSERS are the number of molecules probed in the bulk solution and the SERS experiment,
respectively [173]. Nsolution and NSERS were calculated following the method used by Phan-Quang
et. al, with a laser spot diameter of 1 mm and an estimated monolayer density of 4.5
molecules/nm2 of 4-MBN on Cu [174].
Table 6: SERS Enhancement Factor
The thiol functional group in this class of target molecules adsorbs well onto Cu surfaces despite
the tendency of Cu to oxidize. Furthermore, it adsorbs in known orientations that aid in signal
enhancement, and can form a linkage in more complex sensors [160, 175, 176]. The exact
structures of organosulfur monolayers adsorbed onto Cu are not completely understood, but have
been shown to display very similar infrared spectra to those on silver, thus an estimated molecule

94
density for a nanoporous silver-substrate was used [160, 175]. This is expected to be a lower
bound on the likely number of adsorbed target molecules per unit area as the bond between the
thiol (sulfur) group and the nanoporous Cu surface may produce a more perpendicular
orientation of the 4-MBN on the substrate surface, resulting in denser packing of the target
molecules [160]. Overall, samples B3 and A2 displayed the best SERS enhancement of 4-MBN,
with EF values on the order of 107 and 106
, respectively. The enhancement effect of the inherent
top-surface microporosity is apparent in the as-sputtered samples A1 and B1, which produced
enhancement factors on the order of 105 and 104, respectively.
SERS enhancement stems from the excitation and trapping of surface plasmons on the
metallic nanostructured substrate. The optimal conditions for a specific substrate and light source
depend on many factors including the metal and the specific local nanoscale topology such as the
size, shape, or geometry of voids [108, 166]. Surface cavities and pores, which have negative
curvature, provide additional enhancement hot spots by electromagnetic field localization, and
signal enhancement correlates with the average pore size [108, 172]. Figure 46 shows SEM
micrographs of the top-surface morphology of the as-sputtered and dealloyed films from sets A
and B, with the same color and labeling conventions as their corresponding SERS spectra in
Figure 45. The micrographs of A1 and B1 (top row, red outlines) show solid, faceted grains with
narrow voids between particles (A1) and grains with more rounded corners and oblong voids
(B1). Although the microporosity of A1 and B1 were quite similar, they had different
compositions (see Table 4). Specifically, the atomic fraction of Zn atoms was 61% for A1 and
48% for B1, leading to higher enhancement in the A1 sample. This finding highlights the effect
of Zn on overall enhancement as previously observed by Sultangaziyev et al Zn has a greater
enhancement effect than either pure Cu or pure Al [177].

95
For the dealloyed samples, both microporosity and nanoporosity played a role in the
SERS enhancement, which shows a distinct correlation with the local surface topology and
surface roughness, including void size, void shape, and nanopore size [107, 108, 166]. Samples
B3 and A2 demonstrated the largest signal amplification, with enhancement factors of 1.7 x 107
and 9.0 x 106
, respectively, while sample A3 came in third with an enhancement factor of 5.1 x
106
. Based on the ~1:1 pore size to ligament size ratio, B2 and B3 possessed slightly larger top
surface pores than A2 and A3, while B3 had the largest cross-sectional nanopores at ~40-50 nm
[172]. This is larger than the optimal pore size of ~5-10 nm determined by Qian et al in np-Au
for crystal violet (CV) and rhodamine 6G (R6G) probe molecules and a laser wavelength of
~500-650 nm [167]. However, this is comparable to the optimal pore sizes for enhancement of
R6G and CV in nanoporous Cu determined by Chen et al, which were in the ~30-35 nm and ~45-
50 nm range, respectively [145].

96
Figure 46: Top-surface SEM micrographs of the (A1, B1) as-sputtered, (A2, B2) “half-cycle”
dealloyed, and (A2, B2) “full-cycle” dealloyed films. Left column and solid border is for
sputtering stage temperature of 100˚C (film set A), right column and dashed border is for stage
temperature of 150˚C (film set B). Red is as-sputtered, green is “half-cycle” dealloyed, blue is
“full-cycle” dealloyed.
With regard to morphology, for the half-cycle samples A2 and B2, shown with green outlines in
the second row of Figure 46, the top-surface microporous morphology from the as-sputtered
films was maintained, with added nanoporosity in the form of smooth ligaments in A2 and
knobbly ligaments in B2. In contrast, after a full dealloying cycle, samples A3 and B3, shown
with blue outlines, yielded more rounded particle edges than their half-cycle counterparts.

97
Overall the addition of nanopores in the ~25-45 nm range increased the resulting SERS
enhancement, with the largest increases for an average combined pore size (top-surface and
cross-sectional) in the ~35-45 nm range. For B3 specifically, the presence of additional surface
roughness from the slightly larger ligaments further contributed to the increased enhancement by
providing more hot spots [145, 167]. Thus, while enhancement of the 4-MBN signal for both assputtered and dealloyed samples was above the minimum threshold, a combination of optimal
void size, nanopore size, and composition produced the best signal amplification for this probe
molecule.
6.3 Conclusions
This work investigates the synthesis of gradient nanoporous Cu and its suitability as a
tunable SERS substrate. The gradient nanoporous structure was fabricated by a combination of
magnetron sputtering with additional heating to generate a gradient composition, and then
subsequent dealloying at various time points to induce a porous structure. Differences in the
resulting nanoporosity and local foam morphology produced SERS enhancement factors for 4-
MBN ranging from 104-107 for all set A and B samples. Composition, layer sequence,
dealloying, heat treatments and foam morphology were evaluated as to their roles in the resulting
SERS enhancement. Microporosity was similar for all samples, while composition and nanopore
size differed. Rougher ligaments and an average combined nanopore size of ~35-45 nm produced
the greatest increase in enhancement. The tunability of these gradient nanoporous materials and
their compatibility with thiol-based target molecules make them great candidates for
multifunctional SERS substrates. Furthermore, the ability to tune processing parameters to
achieve specific nanostructures and morphological features is vital for the development of SERS
substrates with good signal amplification that can be used for multiple target molecules

98
simultaneously.

99
Chapter 7. Conclusions and Future Work
7.1 Conclusions
In nanostructured materials, interfaces such as grain boundaries comprise a high volume
fraction of the material, and both the quantity and the character of these interfaces can be
engineered to alter the material’s functionality. Corrosion studies in nanostructured materials
have generally been focused on corrosion along boundaries, with less emphasis on engineering
interfaces to minimize corrosion, or leveraging corrosion to create interfaces. Interfaces
synthesized through corrosion allow for the fusion of some of the unique properties of both
nanoporous and gradient materials into a single class of materials. The current understanding of
porosity gradients in nanoporous foams is limited, with a focus on synthesis. However, by
synthesizing a variety of sharp and graded porosity interfaces through a selective corrosion
process and analyzing their microstructures, a guide was developed for understanding their
stability, properties and character at the interface. Additionally, by identifying the major
processing parameters and their relationships to the resulting microstructures, this research
enables the synthesis of a specific set of characteristics or nanoscale features more precisely.
The first section of this thesis provides insight into improving corrosion-resistance in
sputtered alloys through the combination of an interrupted deposition process and high sputtering
rates. The resulting higher sputtering temperature and strain aided the formation of more
energetically favorable low-Σ (special) grain boundaries. The types and quantities of low-Σ GBs
were confirmed with electron back-scatter diffraction (EBSD), and immersion corrosion testing
in artificial seawater supported both the increased fraction of special GBs and increased
corrosion resistance on the sample surface in contact with the liquid.

100
In the first part of the second study, two distinct categories of sharp interface were
synthesized via dealloying corrosion: interfaces with a difference in chemical potential, and
interfaces with a difference in relative density. This was achieved with a set of sputtered
precursor films consisting of Cu-Al/Cu-Zn and Cu-Al/Cu-Al, respectively. The reaction front
was analyzed at three time points for each case, and it was determined that both a higher atomic
fraction of Cu and longer reaction times contributed to a more uniform nanoporous structure.
Additionally, voids between grains in the as-sputtered films made grain-boundary etching the
preferred dealloying mechanism, and yielded nanoporous columns with fast reaction times.
Lastly, the final phase of the second study sought to evaluate the utility of gradient
nanoporous Cu as a SERS substrate while identifying the relationships between key processing
parameters and specific foam morphological features. Gradient nanoporous foams have
previously demonstrated good signal enhancement for surface-enhanced Raman scattering
(SERS)-based sensing. Cu specifically is of interest as a cheaper and more robust replacement
for Au and Ag-based foam SERS substrates. Alloys previously selected in Chapter 5 were
sputtered at 100˚C or 150˚C to instigate diffusion, forming compositionally graded films, which
were dealloyed in 0.1M H3PO4. A thiol-based SERS probe, 4-MBN, was used to evaluate the
SERS response of dealloyed and as-sputtered films. Sample composition, microporosity,
nanopore size and ligament morphology were identified as contributing factors to the degree of
signal amplification. All samples exhibited an enhancement effect above the threshold value of
104
, where the best response was on the order of 107
, on par with nanoporous Au. This
understanding will aid in better design of corrosion-fabricated structures of gradient porosity for
catalysis and sensing applications that can combine the necessary chemical properties with the
physical robustness needed for reusable sensor materials.

101
Overall, these studies demonstrated: (1) a high sputtering rate with interrupted deposition
produced a high fraction of special grain boundaries due to stress relaxation during film growth,
and that this improved the corrosion resistance of 5xxx series Al-Mg alloys in artificial seawater,
(2) the synthesis of gradient nanoporous Cu via a dealloying corrosion process and subsequent
investigation of the effect of a sharp relative density or chemical potential interface on the
association between processing protocols and foam morphology, and (3) a pathway for more
robust SERS substrates. By identifying the key links between processing and microstructure for
these interfaces, these studies provide critical insight for understanding corrosion in complex
material systems and how to leverage this knowledge for future controlled and reliable synthesis.
7.2 Future Work
The results of this work have examined the relationship between nanoscale interfaces and
corrosion behavior in multiple contexts. With the rising demand for multifunctional sensors,
there are two future lines of inquiry into gradient nanoporous SERS substrates that should be
pursued. The first is to expand the investigation of the relationship between microstructural
features and SERS response in gradient nanoporous Cu to encompass a broader range of probe
molecules, and the second is to apply this methodology to evaluate other gradient nanoporous
metallic structures.
In order to develop the next generation of multifunctional sensors, there is now a need for
comprehensive studies that analyze the SERS enhancement at the same laser wavelengths for a
broader range of target molecules. Building upon this initial demonstration of gradient
nanoporous Cu structures as SERS substrates for thiol-based analytes is the first step. Thiolbased target molecules, which have consistent adsorption behavior and predictable molecule
orientations, can facilitate more complex sensors. This set of standardized analyses can then be

102
performed on a wider variety of analytes including drug molecules, toxins, and fluorescent dyes.
This will expand understanding of the relationship between microstructural features such as
degree of microporosity, nanopore size and pore shape on signal enhancement for specific
molecules and chemical functional groups. Mapping these relationships between specific
processing variables, microstructural features, and SERS response will provide a method for
synthesizing SERS substrates that are optimized for a set of target molecules and expedite future
substrate development. Additionally, this mapping of porous morphology and SERS response
can be extended to other metals of interest such as Al or Ti that are both cheaper and more
abundant than the conventional Au and Ag, as well as multimetallic substrates. Further
examination of a broader range of substrate materials will enable more low-cost substrates, the
potential for usable ultraviolet wavelengths, and recyclable substrates.
These investigations identified relationships between processing, microstructure and
properties for corrosion at various nanoscale interfaces. Future work on gradient nanoporous
metallic sensors can further explore these relationships for a broader range of molecules in order
to expand this knowledge base and devise a standardized set of testing conditions to facilitate
substrate development.

103
References
[1] H. Gleiter, Nanostructured Materials: Basic Concepts and Microstructure, Acta Mat. 48
(2000) 1-29.
[2] P.I. Dolez, Nanomaterials Definitions, Classifications, and Applications, Nanoeng. (2015) pp.
3-40.
[3] G.E. Fougere, J.R. Weertman, R.W. Siegel, S. Kim, Grain-Size Dependent Hardening and
Softening of Nanocrystalline Cu And Pd, Scr. Metall. Mater. 26 (1992) 1879-1883.
[4] X. Li, L. Lu, J. Li, X. Zhang, H. Gao, Mechanical properties and deformation mechanisms of
gradient nanostructured metals and alloys, Nat. Rev. Mater. 5(9) (2020) 706-723.
[5] L. Priester, Grain Boundaries from Theory to Engineering, first ed., Springer, Les Ulis, 2006.
[6] V. Randle, H. Davies, I. Cross, Grain boundary misorientation distributions, Curr. Opin.
Solid State Mater. Sci. 5 (2001) 3-8.
[7] A.J. Schwartz, W.E. King, The potential engineering of grain boundaries through
thermomechanical processing, JOM (1998) 50-55.
[8] R.E. Smallman, A.H.W. Ngan, Surfaces, Grain Boundaries and Interfaces, in: R.E.
Smallman, A.H.W. Ngan (Eds.), Modern Physical Metallurgy, eighth ed., ButterworthHeinemann, Oxford, 2014, pp. 415-442.
[9] D.L. Olmsted, S.M. Foiles, E.A. Holm, Survey of computed grain boundary properties in
face-centered cubic metals: I. Grain boundary energy, Acta Mater. 57(13) (2009) 3694-3703.
[10] J.A. Bahena, Understanding the formation and evolution of boundaries and interfaces in
nanostructured metallic alloys, Mechanical Engineering, University of Southern California, Los
Angeles, 2020.
[11] D.G. Brandon, The structure of high angle grain boundaries, Acta Metall. 14 (1966) 1479-
1484.
[12] A.H. King, S. Shekhar, What does it mean to be special? The significance and application of
the Brandon criterion, J. Mater. Sci. 41(23) (2006) 7675-7682.
[13] J. Yan, N.M. Heckman, L. Velasco, A.M. Hodge, Improve sensitization and corrosion
resistance of an Al-Mg alloy by optimization of grain boundaries, Sci. Rep. 6 (2016) 26870.
[14] T. Watanabe, Grain boundary engineering: historical perspective and future prospects, J.
Mater. Sci. 46(12) (2011) 4095-4115.
[15] A. Movahedi-Rad, R. Alizadeh, Simulating Grain Boundary Energy Using Molecular
Dynamics, J. Mod. Phys. 05(08) (2014) 627-632.

104
[16] V. Randle, G. Owen, Mechanisms of grain boundary engineering, Acta Mater. 54(7) (2006)
1777-1783.
[17] D.B. Bober, A. Khalajhedayati, M. Kumar, T.J. Rupert, Grain Boundary Character
Distributions in Nanocrystalline Metals Produced by Different Processing Routes, Metall. Mater.
Trans. A 47(3) (2015) 1389-1403.
[18] S. Kobayashi, S. Tsurekawa, T. Watanabe, A new approach to grain boundary engineering
for nanocrystalline materials, Beilstein J. Nanotechnol. 7 (2016) 1829-1849.
[19] K. Harada, S. Tsurekawa, T. Watanabe, G. Palumbo, Enhancement of homogeneity of grain
boundary microstructure by magnetic annealing of electrodeposited nanocrystalline nickel, Scr.
Mater. 49(5) (2003) 367-372.
[20] Y. Zhao, M.N. Polyakov, M. Mecklenburg, M.E. Kassner, A.M. Hodge, The role of grain
boundary plane orientation in the β phase precipitation of an Al–Mg alloy, Scr. Mater. 89 (2014)
49-52.
[21] K.D. Hemmendinger, J. A. Bahena, A.M. Hodge, Characterization of Grain BoundaryEngineered Aluminum-Magnesium Alloys, Adv. Eng. Mater. (2020) 2000813 1-5.
[22] R. Goswami, G. Spanos, P.S. Pao, R.L. Holtz, Precipitation behavior of the ß phase in Al5083, Mater. Sci. Eng., A 527(4-5) (2010) 1089-1095.
[23] G.M. Fontana, Corrosion Engineering, third ed., McGraw-Hill, Singapore, 1987.
[24] I. McCue, E. Benn, B. Gaskey, J. Erlebacher, Dealloying and Dealloyed Materials, Annu.
Rev. Mater. Res. 46(1) (2016) 263-286.
[25] D.A. Jones, Principles and prevention of corrosion, Prentice-Hall, Upper Saddle River,
1996.
[26] D.L. Engelberg, Intergranular Corrosion, Reference Module in Materials Science and
Materials Engineering, in: B. Cottis, M. Graham, R. Lindsay, S. Lyon, T. Richardson, D.
Scantlebury, H. Stott (Eds.), Shreir’s Corrosion, Elsevier, Amsterdam, 2010, pp. 810-827.
[27] D.L. Olmsted, E.A. Holm, S.M. Foiles, Survey of computed grain boundary properties in
face-centered cubic metals—II: Grain boundary mobility, Acta Mater. 57(13) (2009) 3704-3713.
[28] R. Zhang, S.P. Knight, R.L. Holtz, R. Goswami, C.H.J. Davies, N. Birbilis, A Survey of
Sensitization in 5xxx Series Aluminum Alloys, Corros. 72(2) (2016) 144-159.
[29] D.S. D’Antuono, J. Gaies, W. Golumbfskie, M.L. Taheri, Direct measurement of the effect
of cold rolling on beta phase precipitation kinetics in 5xxx series aluminum alloys, Acta Mater.
123 (2017) 264-271.

105
[30] R. H. Jones, D.R. Baer, M.J. Danielson, J.S. Vetrano, Role Of Mg In The Stress Corrosion
Cracking of an Al-Mg Alloy, Metall. Mater. Trans. A 32A (2001) 1699-1711.
[31] J.L. Searles, P.I. Gouma, R.G. Buchheit, Stress Corrosion Cracking of Sensitized AA5083
(Al-4.5Mg-1.0Mn), Metall. Mater. Trans. A 32A (2001) 2859-2867.
[32] R. Zhang, Y. Qiu, Y. Qi, N. Birbilis, A closer inspection of a grain boundary immune to
intergranular corrosion in a sensitised Al-Mg alloy, Corros. Sci. 133 (2018) 1-5.
[33] J.A. Lyndon, R.K. Gupta, M.A. Gibson, N. Birbilis, Electrochemical behaviour of the βphase intermetallic (Mg2Al3) as a function of pH as relevant to corrosion of aluminium–
magnesium alloys, Corros. Sci. 70 (2013) 290-293.
[34] D.L. Foley, A.C. Leff, A.C. Lang, M.L. Taheri, Evolution of beta-phase precipitates in an
aluminum-magnesium alloy at the nanoscale, Acta Mater. 185 (2020) 279-286.
[35] L. Velasco, A.M. Hodge, Growth twins in high stacking fault energy metals:
Microstructure, texture and twinning, Mater. Sci. Eng., A 687 (2017) 93-98.
[36] I.-C. Cheng, Synthesis, Characterization, and Mechanical Properties of Nanoporous Foams,
Materials Science, University of Southern California, Los Angeles, 2013.
[37] D.R. Lide, CRC Handbook of Chemistry and Physics, CRC Press, Boca Raton, 2005.
[38] K. Sieradzki, N. Dimitrov, D. Movrin, C. McCall, N. Vasiljevic, J. Erlebacher, The
Dealloying Critical Potential, J. Electrochem. Soc. 149(8) (2002).
[39] T. Juarez, J. Biener, J. Weissmüller, A.M. Hodge, Nanoporous Metals with Structural
Hierarchy: A Review, Adv. Eng. Mater. 19(12) (2017).
[40] F. Liu, H.-J. Jin, Extrinsic Parting Limit for Dealloying of Cu-Rh, J. Electrochem. Soc.
165(16) (2018) C999-C1006.
[41] O. Okman, Nanoporous Gold: Mechanics of Fabrication and Actuation, Columbia
University, New York, 2012.
[42] O. Okman, D. Lee, J.W. Kysar, Fabrication of crack-free nanoporous gold blanket thin films
by potentiostatic dealloying, Scr. Mater. 63(10) (2010) 1005-1008.
[43] O. Okman, J.W. Kysar, Fabrication of crack-free blanket nanoporous gold thin films by
galvanostatic dealloying, J. Alloys Compd. 509(22) (2011) 6374-6381.
[44] M.F. Ashby, The Mechanical Properties of Cellular Solids, Metall. Trans. A 14 (1983).
[45] N. Beets, D. Farkas, Mechanical Response of Au Foams of Varying Porosity from
Atomistic Simulations, JOM 70(10) (2018) 2185-2191.

106
[46] I.C. Cheng, A.M. Hodge, Morphology, Oxidation, and Mechanical Behavior of Nanoporous
Cu Foams, Adv. Eng. Mater. 14(4) (2012) 219-226.
[47] T. Song, M. Yan, M. Qian, The enabling role of dealloying in the creation of specific
hierarchical porous metal structures—A review, Corros. Sci. 134 (2018) 78-98.
[48] S. Singh, N. Bhatnagar, A survey of fabrication and application of metallic foams (1925–
2017), J. Porous Mater. 25(2) (2017) 537-554.
[49] T. Fujita, Y. Kanoko, Y. Ito, L. Chen, A. Hirata, H. Kashani, O. Iwatsu, M. Chen,
Nanoporous Metal Papers for Scalable Hierarchical Electrode, Adv. Sci. (Weinh) 2(8) (2015)
1500086.
[50] J. Weissmüller, R.C. Newman, H.-J. Jin, A.M. Hodge, J.W. Kysar, Nanoporous Metals by
Alloy Corrosion: Formation and Mechanical Properties, MRS Bull. 34(8) (2011) 577-586.
[51] I.C. Cheng, A.M. Hodge, High temperature morphology and stability of nanoporous Ag
foams, J. Porous Mater. 21(4) (2014) 467-474.
[52] Y. Peng, T. Wu, L. Sun, J.M.V. Nsanzimana, A.C. Fisher, X. Wang, Selective
Electrochemical Reduction of CO2 to Ethylene on Nanopores-Modified Copper Electrodes in
Aqueous Solution, ACS Appl. Mater. Interfaces 9(38) (2017) 32782-32789.
[53] J.L. Plawsky, J.K. Kim, E.F. Schubert, Engineered nanoporous and nanostructured films,
Mater. Today 12 (2009) 36-45.
[54] F.M. Bayoumi, B.G. Ateya, Formation of self-organized titania nano-tubes by dealloying
and anodic oxidation, Electrochem. Commun. 8(1) (2006) 38-44.
[55] W. Jiao, P. Liu, H. Lin, W. Zhou, Z. Wang, T. Fujita, A. Hirata, H.-W. Li, M. Chen,
Tunable Nanoporous Metallic Glasses Fabricated by Selective Phase Dissolution and Passivation
for Ultrafast Hydrogen Uptake, Chem. Mater. 29(10) (2017) 4478-4483.
[56] C.J. Ruestes, D. Schwen, E.N. Millán, E. Aparicio, E.M. Bringa, Mechanical properties of
Au foams under nanoindentation, Comput. Mater. Sci. 147 (2018) 154-167.
[57] I.C. Cheng, A.M. Hodge, Strength scale behavior of nanoporous Ag, Pd and Cu foams, Scr.
Mater. 69(4) (2013) 295-298.
[58] Y.-Z. Lee, W.-Y. Zeng, I.C. Cheng, Synthesis and characterization of nanoporous copper
thin films by magnetron sputtering and subsequent dealloying, Thin Solid Films 699 (2020).
[59] Z.Y. Hu, P.P. Wang, E.G. Fu, X.J. Wang, X.Q. Yan, P. Xu, Z.M. Wu, Y.B. Zhao, Y.X.
Liang, Bilayer nanoporous copper films with various morphology features synthesized by onestep dealloying, J. Alloys Compd. 754 (2018) 26-31.

107
[60] J. Huang, Y. Liu, X. He, C. Tang, K. Du, Z. He, Gradient nanoporous gold: a novel surfaceenhanced Raman scattering substrate, RSC Adv. 7(26) (2017) 15747-15753.
[61] X. Wu, Y. Zhu, Heterogeneous materials: a new class of materials with unprecedented
mechanical properties, Mater. Res. Lett. 5(8) (2017) 527-532.
[62] T. Balusamy, T.S.N. Sankara Narayanan, K. Ravichandran, I.S. Park, M.H. Lee, Influence
of surface mechanical attrition treatment (SMAT) on the corrosion behaviour of AISI 304
stainless steel, Corros. Sci. 74 (2013) 332-344.
[63] R.K. Gupta, N. Birbilis, The influence of nanocrystalline structure and processing route on
corrosion of stainless steel: A review, Corros. Sci. 92 (2015) 1-15.
[64] Z. Esen, Ş. Bor, Processing of titanium foams using magnesium spacer particles, Scr. Mater.
56(5) (2007) 341-344.
[65] Z. Qi, C. Zhao, X. Wang, J. Lin, W. Shao, Z. Zhang, X. Bian, Formation and
Characterization of Monolithic Nanoporous Copper by Chemical Dealloying of Al-Cu Alloys, J.
Phys. Chem. C (2009).
[66] T. Song, Y. Gao, Z. Zhang, Q. Zhai, Dealloying behavior of rapidly solidified Al–Ag alloys
to prepare nanoporous Ag in inorganic and organic acidic media, CrystEngComm 13(23) (2011).
[67] Z.W. Qi, J. Weissmüller,, Hierarchical Nested-Network Nanostructure by Dealloying, ACS
Nano 7 (2013) 5948-5954.
[68] J. Biener, A.M. Hodge, J.R. Hayes, C.A. Volkert, L.A. Zepeda-Ruiz, A.V. Hamza, F.F.
Abraham, Size Effects on the Mechanical Behavior of Nanoporous Au, Nano Lett. 6 (2006)
2379-2382.
[69] A. Mathur, J. Erlebacher, Size dependence of effective Young’s modulus of nanoporous
gold, Appl. Phys. Lett. 90(6) (2007).
[70] L.J. Gibson, M.F. Ashby, The mechanics of three-dimensional cellular materials, Proc. R.
Soc. Lond., A (1982).
[71] N.J. Briot, T.J. Balk, Developing scaling relations for the yield strength of nanoporous gold,
Philos. Mag. 95(27) (2015) 2955-2973.
[72] A.M. Hodge, J.R. Hayes, J.A. Caro, J. Biener, A.V. Hamza, Characterization and
Mechanical Behavior of Nanoporous Gold, Adv. Eng. Mater. 8(9) (2006) 853-857.
[73] H.-J. Jin, J. Weissmüller, D. Farkas, Mechanical response of nanoporous metals: A story of
size, surface stress, and severed struts, MRS Bull. 43(1) (2018) 35-42.

108
[74] R.N. Viswanath, S.R. Polaki, R. Rajaraman, S. Abhaya, V.A. Chirayath, G. Amarendra,
C.S. Sundar, On the scaling behavior of hardness with ligament diameter of nanoporous-Au:
Constrained motion of dislocations along the ligaments, Appl. Phys. Lett. 104(23) (2014).
[75] L.-Z. Liu, H.-J. Jin, Scaling equation for the elastic modulus of nanoporous gold with
“fixed” network connectivity, Appl. Phys. Lett. 110(21) (2017).
[76] L.-Z. Liu, X.-L. Ye, H.-J. Jin, Interpreting anomalous low-strength and low-stiffness of
nanoporous gold: Quantification of network connectivity, Acta Mater. 118 (2016) 77-87.
[77] E.D. McClanahan, N. Laegreid, Sputtering by Particle Bombardment III: Production of Thin
Films by Controlled Deposition of Sputtered Material, Top. Appl. Phys. 64 (1991) 339-377.
[78] S. Swann, Magnetron Sputtering, Phys. Technol. 19 (1988).
[79] J.L. Vossen, J.J. Cuomo, Glow Discharge Sputter Deposition, in: J.L. Vossen (Ed.), Thin
Film Processes, Elsevier, 1978, pp. 11-73.
[80] J.A. Thornton, The microstructure of sputter‐deposited coatings, J. Vac. Sci. Technol. A
4(6) (1986) 3059-3065.
[81] J.A. Thornton, Influence of substrate temperature and deposition rate on structure of thick
sputtered Cu coatings, J. Vac. Sci. Technol. 12(4) (1975) 830-835.
[82] E.G. Fu, N. Li, A. Misra, R.G. Hoagland, H. Wang, X. Zhang, Mechanical properties of
sputtered Cu/V and Al/Nb multilayer films, Mater. Sci. Eng. A 493(1-2) (2008) 283-287.
[83] Z. Szklarska-Smialowska, Pitting corrosion of aluminum, Corros. Sci. 41 (1999) 1743-1767.
[84] A. Dursun, D.V. Pugh, S.G. Corcoran, Probing the Dealloying Critical Potential, J.
Electrochem. Soc. 152(2) (2005).
[85] M.C. Dixon, T.A. Daniel, M. Hieda, D.M. Smilgies, M.H. Chan, D.L. Allara, Preparation,
structure, and optical properties of nanoporous gold thin films, Langmuir 23(5) (2007) 2414-22.
[86] J.W. Kysar, O. Okman, Galvanostatic Dealloying for Fabrication of Constrained Blanket
Nanoporous Gold Films, US Patent 20120077057A1, 20112.
[87] S. Parida, D. Kramer, C.A. Volkert, H. Rosner, J. Erlebacher, J. Weissmuller, Volume
change during the formation of nanoporous gold by dealloying, Phys. Rev. Lett. 97(3) (2006)
035504.
[88] Y. Sun, K.P. Kucera, S.A. Burger, T. John Balk, Microstructure, stability and
thermomechanical behavior of crack-free thin films of nanoporous gold, Scr. Mater. 58(11)
(2008) 1018-1021.

109
[89] J.L. Snyder, K.; Erlebacher, J., Dealloying Silver/Gold Alloys in Neutral Silver Nitrate
Solution: Porosity Evolution, Surface Composition, and Surface Oxides, J. Electrochem. Soc.
155 (2008).
[90] B. Hafner, Scanning Electron Microscopy Primer, University of Minnesota - Twin Cities,
2007, pp. 1-29.
[91] B.J. Inkson, Scanning electron microscopy (SEM) and transmission electron microscopy
(TEM) for materials characterization, in: G. Hübschen, I. Altpeter, R. Tschuncky, H.-G.
Herrmann (Eds.), Materials Characterization Using Nondestructive Evaluation (NDE) Methods,
Elsevier, Amsterdam, 2016, pp. 17-43.
[92] N. Erdman, D.C. Bell, R. Reichelt, Scanning Electron Microscopy, in: P.W. Hawkes, J.C.H.
Spence (Eds.) Springer Handbook of Microscopy, Springer, 2019, pp. 229-318.
[93] S.M. Vitale, J.D. Sugar, Using Xe Plasma FIB for High-Quality TEM Sample Preparation,
Microsc. Microanal. 28(3) (2022) 646-658.
[94] C.A. Volkert, A.M. Minor, Focused Ion Beam Microscopy and Micromachining, MRS Bull.
32 (2007) 389-399.
[95] A. Wolff, Focused ion beams: An overview of the technology and its capabilities, Wiley
Anal. Sci., Wiley, 2020.
[96] J.I. Goldstein, D.E. Newbury, J.R. Michael, N.W.M. Ritchie, J.H.J. Scott, D.C. Joy,
Scanning Electron Microscopy and X-Ray Microanalysis, fourth ed., Springer, New York, NY,
2018.
[97] D.B.C. Williams, C. Barry, Transmission Electron Microscopy: A Textbook for Materials
Science, Springer, New York, NY, 2009.
[98] J. Abraham, B. Jose, A. Jose, S. Thomas, Characterization of green nanoparticles from
plants, Phytonanotechnol. 2020, pp. 21-39.
[99] Y. Leng, Materials Characterization Introduction to Microscopic and Spectroscopic
Methods, John WIley & Sons (Asia) Pte Ltd, Singapore, 2010.
[100] Z.S. Duma, T. Sihvonen, J. Havukainen, V. Reinikainen, S.P. Reinikainen, Optimizing
energy dispersive X-Ray Spectroscopy (EDS) image fusion to Scanning Electron Microscopy
(SEM) images, Micron 163 (2022) 103361.
[101] G.C. Sneddon, P.W. Trimby, J.M. Cairney, Transmission Kikuchi diffraction in a scanning
electron microscope: A review, Mater. Sci. Eng. R. 110 (2016) 1-12.
[102] P.W. Trimby, Orientation mapping of nanostructured materials using transmission Kikuchi
diffraction in the scanning electron microscope, Ultramicroscopy 120 (2012) 16-24.

110
[103] G. Abadias, E. Chason, J. Keckes, M. Sebastiani, G.B. Thompson, E. Barthel, G.L. Doll,
C.E. Murray, C.H. Stoessel, L. Martinu, Review Article: Stress in thin films and coatings:
Current status, challenges, and prospects, J. Vac. Sci. Technol., A 36(2) (2018).
[104] X.X. Han, R.S. Rodriguez, C.L. Haynes, Y. Ozaki, B. Zhao, Surface-enhanced Raman
spectroscopy, Nat. Rev. Methods Primers 1(1) (2022).
[105] E. Smith, G. Dent, Modern Raman Spectroscopy: A Practical Approach, Second ed.,
Wiley, Hoboken, NJ, 2019.
[106] A.I. Perez-Jimenez, D. Lyu, Z. Lu, G. Liu, B. Ren, Surface-enhanced Raman spectroscopy:
benefits, trade-offs and future developments, Chem. Sci. 11(18) (2020) 4563-4577.
[107] F. Diao, X. Xiao, B. Luo, H. Sun, F. Ding, L. Ci, P. Si, Two-step fabrication of nanoporous
copper films with tunable morphology for SERS application, Appl. Surf. Sci. 427 (2018) 1271-
1279.
[108] S.O. Kucheyev, J.R. Hayes, J. Biener, T. Huser, C.E. Talley, A.V. Hamza, Surfaceenhanced Raman scattering on nanoporous Au, Appl. Phys. Lett. 89(5) (2006).
[109] M. Magdy, A Conceptual Overview of Surface-Enhanced Raman Scattering (SERS),
Plasmonics 18(2) (2023) 803-809.
[110] D.W. Hetzner, Microindentation Hardness Testing of Materials Using ASTM E384,
Microsc. Microanal. 9(S02) (2003) 708-709.
[111] ASTM, Standard Test Method for Vickers Indentation Hardness of Advanced Ceramics,
ASTM International, West Conshohocken, PA, 2019.
[112] C.D. Appleget, A.M. Hodge, Optical and Mechanical Characterization of Sputtered
AlN/Ag Multilayer Films, Adv. Eng. Mater. 21(5) (2019).
[113] K.N. Tran, L. Salamanca-Riba, Microstructural Evolution of Severely Plastically
Deformed Sensitized Aluminum 5456-H116 Treated by Ultrasonic Impact Treatment, Adv. Eng.
Mater. 15(11) (2013) 1105-1110.
[114] R. Goswami, R.L. Holtz, Transmission Electron Microscopic Investigations of Grain
Boundary Beta Phase Precipitation in Al 5083 Aged at 373 K (100 °C), Metall. Mater. Trans. A
44(3) (2012) 1279-1289.
[115] V. Chawla, R. Jayaganthan, A.K. Chawla, R. Chandra, Microstructural characterizations of
magnetron sputtered Ti films on glass substrate, J. Mater. Process. Technol. 209(7) (2009) 3444-
3451.
[116] L. Velasco, M.N. Polyakov, A.M. Hodge, Influence of stacking fault energy on twin
spacing of Cu and Cu–Al alloys, Scr. Mater. 83 (2014) 33-36.

111
[117] R.E. García, M.D. Vaudin, Correlations between the crystallographic texture and grain
boundary character in polycrystalline materials, Acta Mater. 55(17) (2007) 5728-5735.
[118] L. Tan, T.R. Allen, J.T. Busby, Grain boundary engineering for structure materials of
nuclear reactors, J. Nucl. Mater. 441(1-3) (2013) 661-666.
[119] R. Aguirre, S. Abdullah, X. Zhou, D. Zubia, Molecular Dynamics Calculations of Grain
Boundary Mobility in CdTe, Nanomater. (Basel) 9(4) (2019).
[120] L. Velasco, A.M. Hodge, The mobility of growth twins synthesized by sputtering:
Tailoring the twin thickness, Acta Mater. 109 (2016) 142-150.
[121] J.L. David B. Bober, Rupalee P. Mulay, Timothy J. Rupert, Mukul Kumar, The formation
and characterization of large twin related domains, Acta Mater. 129 500-509.
[122] G. Stokkan, A. Song, B. Ryningen, Investigation of the Grain Boundary Character and
Dislocation Density of Different Types of High Performance Multicrystalline Silicon, Crystals
8(9) (2018).
[123] M.A. Tschopp, K.N. Solanki, F. Gao, X. Sun, M.A. Khaleel, M.F. Horstemeyer, Probing
grain boundary sink strength at the nanoscale: Energetics and length scales of vacancy and
interstitial absorption by grain boundaries inα-Fe, Phys. Rev. B: Condens. Mater. 85(6) (2012).
[124] E.N. Hahn, S.J. Fensin, T.C. Germann, M.A. Meyers, Symmetric tilt boundaries in bodycentered cubic tantalum, Scr. Mater. 116 (2016) 108-111.
[125] A.J.S. Mukul Kumar, Wayne E. King, Microstructural evolution during grain boundary
engineering of low to medium stacking fault energy fcc materials, Acta Mater. (2002) 2599-
2612.
[126] 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 Mater. 52(4) (2004) 995-1002.
[127] O.A. X. Zhang, R. G. Hoagland, A. Misra, Nanoscale growth twins in sputtered metal
films, JOM 60 (2008) 75-78.
[128] I.J. Beyerlein, X. Zhang, A. Misra, Growth Twins and Deformation Twins in Metals,
Annu. Rev. Mater. Res. 44(1) (2014) 329-363.
[129] D. Flötotto, Z.M. Wang, L.P.H. Jeurgens, E. Bischoff, E.J. Mittemeijer, Effect of adatom
surface diffusivity on microstructure and intrinsic stress evolutions during Ag film growth, J.
Appl. Phys. 112(4) (2012).
[130] D. Bufford, H. Wang, X. Zhang, High strength, epitaxial nanotwinned Ag films, Acta
Mater. 59(1) (2011) 93-101.

112
[131] D.L. Rode, V.R. Gaddam, J.H. Yi, Subnanometer surface roughness of dc magnetron
sputtered Al films, J. Appl. Phys. 102 (2007).
[132] C.G.H. Walker, M.M. El-Gomati, A.M.D. Assa'd, M. Zadrazil, The secondary electron
emission yield for 24 solid elements excited by primary electrons in the range 250-5000 ev: a
theory/experiment comparison, Scanning 30(5) (2008) 365-80.
[133] G. Meng, L. Wei, T. Zhang, Y. Shao, F. Wang, C. Dong, X. Li, Effect of
microcrystallization on pitting corrosion of pure aluminium, Corros. Sci. 51(9) (2009) 2151-
2157.
[134] E.M.L. G. Palumbo, P. Lin, Applications for grain boundary engineered materials, JOM
(1998) 40-43.
[135] Y. Zhao, I.C. Cheng, M.E. Kassner, A.M. Hodge, The effect of nanotwins on the corrosion
behavior of copper, Acta Mater, 67 (2014) 181-188.
[136] J. Yan, A.M. Hodge, Study of β precipitation and layer structure formation in Al 5083: The
role of dispersoids and grain boundaries, J. Alloys and Compd. 703 (2017) 242-250.
[137] K.D. Hemmendinger, A.M. Hodge, Progression of the dealloying front in bilayer Cu–Al
and Cu–Zn nanoporous foams, J. Mater. Res. 38(13) (2023) 3407-3415.
[138] C.M. Parlett, K. Wilson, A.F. Lee, Hierarchical porous materials: catalytic applications,
Chem. Soc. Rev. 42(9) (2013) 3876-93.
[139] J.R. Jones, P.D. Lee, L.L. Hench, Hierarchical porous materials for tissue engineering,
Philos. Trans. R. Soc. Lond., A 364(1838) (2006) 263-81.
[140] X. Wang, J. Sun, C. Zhang, T. Kou, Z. Zhang, On the Microstructure, Chemical
Composition, and Porosity Evolution of Nanoporous Alloy through Successive Dealloying of
Ternary Al–Pd–Au Precursor, J. Phys. Chem. C 116(24) (2012) 13271-13280.
[141] Q. Zhang, Z. Zhang, On the electrochemical dealloying of Al-based alloys in a NaCl
aqueous solution, Phys. Chem. Chem. Phys. 12(7) (2010) 1453-72.
[142] W. Liu, S. Zhang, N. Li, J. Zheng, Y. Xing, Microstructure Evolution of Monolithic
Nanoporous Copper from Dual-Phase Al 35 Atom % Cu Alloy, J. Electrochem. Soc. 157(12)
(2010).
[143] T. Song, Y. Gao, Z. Zhang, Q. Zhai, Influence of magnetic field on dealloying of Al-25Ag
alloy and formation of nanoporous Ag, CrystEngComm 14(10) (2012).
[144] X. Guo, J. Han, P. Liu, L. Chen, Y. Ito, Z. Jian, T. Jin, A. Hirata, F. Li, T. Fujita, N. Asao,
H. Zhou, M. Chen, Hierarchical nanoporosity enhanced reversible capacity of bicontinuous
nanoporous metal based Li-O2 battery, Sci. Rep. 6 (2016) 33466.

113
[145] L.-Y. Chen, J.-S. Yu, T. Fujita, M.-W. Chen, Nanoporous Copper with Tunable
Nanoporosity for SERS Applications, Adv. Funct. Mater. 19(8) (2009) 1221-1226.
[146] J. Erlebacher, R. Seshadri, Hard Materials with Tunable Porosity, MRS Bull. 34 (2009)
561-568.
[147] Z.X. Chen, B. Lu, Q. Huang, L. Wang, B. Huang, Sputtering-growth of Cu/Zn alloy
nanofilms on acrylics substrate, Mater. Sci. Eng. B 117(1) (2005) 81-86.
[148] J.T. Gudmundsson, Physics and technology of magnetron sputtering discharges, Plasma
Sources Sci. Technol. 29 (2020) 1-53.
[149] M. Szymonski, Sputtering of Cu and Zn Atoms from Elemental and Alloy Targets, Appl.
Phys. 23 (1980) 89-92.
[150] Y.-c.K. Chen-Wiegart, S. Wang, W.-K. Lee, I. McNulty, P.W. Voorhees, D.C. Dunand, In
situ imaging of dealloying during nanoporous gold formation by transmission X-ray microscopy,
Acta Mater. 61(4) (2013) 1118-1125.
[151] T. Egle, C. Barroo, N. Janvelyan, A.C. Baumgaertel, A.J. Akey, M.M. Biener, C.M.
Friend, D.C. Bell, J. Biener, Multiscale Morphology of Nanoporous Copper Made from
Intermetallic Phases, ACS Appl. Mater. Interfaces 9(30) (2017) 25615-25622.
[152] M. Graf, B. Roschning, J. Weissmüller, Nanoporous Gold by Alloy Corrosion: MethodStructure-Property Relationships, J. Electrochem. Soc. 164(4) (2017) C194-C200.
[153] T. Juarez, A.M. Hodge, Synthesis of Nanoporous Gold Tubes, Adv. Eng. Mater. 18(1)
(2016) 65-69.
[154] A.A. El Mel, F. Boukli-Hacene, L. Molina-Luna, N. Bouts, A. Chauvin, D. Thiry, E.
Gautron, N. Gautier, P.Y. Tessier, Unusual dealloying effect in gold/copper alloy thin films: the
role of defects and column boundaries in the formation of nanoporous gold, ACS Appl. Mater.
Interfaces 7(4) (2015) 2310-21.
[155] J. Snyder, P. Asanithi, A.B. Dalton, J. Erlebacher, Stabilized Nanoporous Metals by
Dealloying Ternary Alloy Precursors, Adv. Mater. 20(24) (2008) 4883-4886.
[156] M.J. Pryor, J.C. Fister, The Mechanism of Dealloying of Copper Solid Solutions and
Intermetallic Phases, J. Electrochem. Soc. 131(6) (1984) 1230-1235.
[157] H. Ji, X. Wang, C. Zhao, C. Zhang, J. Xu, Z. Zhang, Formation, control and
functionalization of nanoporous silver through changing dealloying media and elemental doping,
CrystEngComm 13(7) (2011).
[158] Q. Kong, L. Lian, Y. Liu, J. Zhang, Hierarchical porous copper materials: fabrication and
characterisation, Micro Nano Lett. 8(8) (2013) 432-435.

114
[159] A. Mukherjee, Q. Liu, F. Wackenhut, F. Dai, M. Fleischer, P.M. Adam, A.J. Meixner, M.
Brecht, Gradient SERS Substrates with Multiple Resonances for Analyte Screening: Fabrication
and SERS Applications, Molecules 27(16) (2022).
[160] P.E. Laibinis, G.M. Whitesides, D.L. Allara, Y.-T. Tao, A.N. Parikh, R.G. Nuzzo,
Comparison of the Structures and Wetting Properties of Self-Assembled Monolayers of nAlkanethiols on the Coinage Metal Surfaces, Cu, Ag, Au, J. Am. Chem. Soc. 113 (1991) 7152-
7167.
[161] W. Liu, C. Xin, L. Chen, J. Yan, N. Li, S. Shi, S. Zhang, A facile one-pot dealloying
strategy to synthesize monolithic asymmetry-patterned nanoporous copper ribbons with tunable
microstructure and nanoporosity, RSC Adv. 6(4) (2016) 2662-2670.
[162] Q. Shao, R. Que, M. Shao, L. Cheng, S.T. Lee, Copper Nanoparticles Grafted on a Silicon
Wafer and Their Excellent Surface‐Enhanced Raman Scattering, Adv. Funct. Mater. 22(10)
(2012) 2067-2070.
[163] A.A. Kowalska, A. Kaminska, W. Adamkiewicz, E. Witkowska, M. Tkacz, Novel highly
sensitive Cu-based SERS platforms for biosensing applications, J. Raman Spectrosc. 46(5)
(2015) 428-433.
[164] C. Li, Y. Huang, X. Li, Y. Zhang, Q. Chen, Z. Ye, Z. Alqarni, S.E.J. Bell, Y. Xu, Towards
practical and sustainable SERS: a review of recent developments in the construction of
multifunctional enhancing substrates, J. Mater. Chem. C 9(35) (2021) 11517-11552.
[165] A.V. Markin, N.E. Markina, J. Popp, D. Cialla-May, Copper nanostructures for chemical
analysis using surface-enhanced Raman spectroscopy, TrAC Trends Anal. Chem. 108 (2018)
247-259.
[166] S. Kundu, A new route for the formation of Au nanowires and application of shapeselective Au nanoparticles in SERS studies, J. Mater. Chem. C 1(4) (2013) 831-842.
[167] L.H. Qian, X.Q. Yan, T. Fujita, A. Inoue, M.W. Chen, Surface enhanced Raman scattering
of nanoporous gold: Smaller pore sizes stronger enhancements, Appl. Phys. Lett. 90(15) (2007).
[168] M.J. Voegtle, T. Pal, A.K. Pennathur, S. Menachekanian, J.G. Patrow, S. Sarkar, Q. Cui,
J.M. Dawlaty, Interfacial Polarization and Ionic Structure at the Ionic Liquid-Metal Interface
Studied by Vibrational Spectroscopy and Molecular Dynamics Simulations, J. Phys. Chem. B
125(10) (2021) 2741-2753.
[169] C. Humbert, B. Busson, C. Six, A. Gayral, M. Gruselle, F. Villain, A. Tadjeddine, Sumfrequency generation as a vibrational and electronic probe of the electrochemical interface and
thin films, J. Electroanal. Chem. 621(2) (2008) 314-321.
[170] S.-J.L. Kang, J.G. Fisher, Grain growth in polycrystalline materials: Current understanding
and future research directions, Open Ceram. 16 (2023).

115
[171] C.M. Eastman, Q. Zhang, J.-C. Zhao, Diffusion Coefficients and Phase Equilibria of the
Cu-Zn Binary System Studied Using Diffusion Couples, J. Phase Equilib. Diffus. 41(5) (2020)
642-653.
[172] A.M. Hodge, J. Biener, J.R. Hayes, P.M. Bythrow, C.A. Volkert, A.V. Hamza, Scaling
equation for yield strength of nanoporous open-cell foams, Acta Mater. 55(4) (2007) 1343-1349.
[173] E.C. Le Ru, P.G. Etchegoin, Quantifying SERS enhancements, MRS Bull. 38(8) (2013)
631-640.
[174] G.C. Phan-Quang, N. Yang, H.K. Lee, H.Y.F. Sim, C.S.L. Koh, Y.C. Kao, Z.C. Wong,
E.K.M. Tan, Y.E. Miao, W. Fan, T. Liu, I.Y. Phang, X.Y. Ling, Tracking Airborne Molecules
from Afar: Three-Dimensional Metal-Organic Framework-Surface-Enhanced Raman Scattering
Platform for Stand-Off and Real-Time Atmospheric Monitoring, ACS Nano 13(10) (2019)
12090-12099.
[175] J.C. Love, L.A. Estroff, J.K. Kriebel, R.G. Nuzzo, G.M. Whitesides, Self-Assembled
Monolayers of Thiolates on Metals as a Form of Nanotechnology, Chem. Rev. 105 (2005) 1103-
1169.
[176] D.T. Kwasnieski, H. Wang, Z.D. Schultz, Alkyl-Nitrile Adlayers as Probes of
Plasmonically Induced Electric Fields, Chem. Sci. 6(8) (2015) 4484-4494.
[177] A. Sultangaziyev, A. Ilyas, A. Dyussupova, R. Bukasov, Trends in Application of SERS
Substrates beyond Ag and Au, and Their Role in Bioanalysis, Biosens. (Basel) 12(11) (2022).

116
Appendices
Appendix A: Summary of Sputtered Samples
This appendix summarizes the films sputtered for this dissertation. Note that all films
used in Chapter 4 (Sensitization and Corrosion Resistance in 5xxx Series Al) were sputtered by
Dr. Joel Bahena during his time at USC, and a complete list of the Al-Mg films can be found in
his dissertation titled Understanding the formation and evolution of boundaries and interfaces in
nanostructured metallic alloys (also available as reference 10) [10].
For all Cu films, four films were synthesized with each sputtering run.
Table 7: Summary of Bilayer Sputtered Samples Sputtered Without Additional Heating
Sample Name Substrate Layer Order Power (W) Ar Pressure
(mTorr)
Thickness
Cu20Al80/Cu20Zn80 Si <110> Cu20Al80/Cu20Zn80 50 W 3 2 um
Cu20Zn80/Cu20Al80 Si <110> Cu20Zn80/Cu20Al80 50 W 3 2 um
Cu20Al80/Cu35Al65 Si <110> Cu20Al80/Cu35Al65 50 W 3 2 um
Cu35Al65/Cu20Al80 Si <110> Cu35Al65/Cu20Al80 50 W 3 2 um
Cu30Zn70/Cu30Al70 Si <110> Cu30Zn70/Cu30Al70 50 W 3 2 um
Cu20Zn80/Cu30Al70 Si <110> Cu20Zn80/Cu30Al70 50 W 3 2 um
Cu35Zn65/Cu20Zn80 Si <110> Cu35Zn65/Cu20Zn80 50 W 3 2 um
Cu20Zn80/Cu35Zn65 Si <110> Cu20Zn80/Cu35Zn65 50 W 3 2 um
Cu30Al70/Cu35Zn65 Si <110> Cu30Al70/Cu35Zn65 50 W 3 2 um
Cu35Zn65/Cu30Al70 Si <110> Cu35Zn65/Cu30Al70 50 W 3 2 um
Cu30Al70/Cu20Al80 Si <110> Cu30Al70/Cu20Al80 50 W 3 2 um
Table 8: Summary of Bilayer Sputtered Samples Sputtered With Additional Substrate Heating
Sample Name Substrate Layer Order Stage T
(˚C)
Ar Pressure
(mTorr) Thickness
Cu20Al80/Cu20Zn80
HS-100˚C
Corning
Eagle glass Cu20Al80/Cu20Zn80 100˚C 5 2 um
Cu20Al80/Cu20Zn80
HS-150˚C
Corning
Eagle glass Cu20Al80/Cu20Zn80 150˚C 5 2 um
Cu20Al80/Cu20Zn80
HS-300˚C
Corning
Eagle glass Cu20Al80/Cu20Zn80 300˚C 5 2 um
Cu20Zn80/Cu20Al80
HS-100˚C
Corning
Eagle glass Cu20Zn80/Cu20Al80 100˚C 5 2 um

117
Cu20Zn80/Cu20Al80
HS-150˚C
Corning
Eagle glass Cu20Zn80/Cu20Al80 150˚C 5 2 um
Cu20Zn80/Cu20Al80
HS-300˚C
Corning
Eagle glass Cu20Zn80/Cu20Al80 300˚C 5 2 um
Cu35Al65/Cu35Zn65
HS-100˚C
Corning
Eagle glass
Cu35Al65/Cu35Zn65 100˚C 5 2 um
Cu35Al65/Cu35Zn65
HS-150˚C
Corning
Eagle glass Cu35Al65/Cu35Zn65 150˚C 5 2 um
Cu35Al65/Cu35Zn65
HS-300˚C
Corning
Eagle glass
Cu35Al65/Cu35Zn65 300˚C 5 2 um
Cu35Zn65/Cu35Al65
HS-100˚C
Corning
Eagle glass Cu35Zn65/Cu35Al65 100˚C 5 2 um
Cu35Zn65/Cu35Al65
HS-150˚C
Corning
Eagle glass Cu35Zn65/Cu35Al65 150˚C 5 2 um
Cu35Zn65/Cu35Al65
HS-300˚C
Corning
Eagle glass Cu35Zn65/Cu35Al65 300˚C 5 2 um
Cu30Zn70/Cu30Al70
HS-100˚C
Corning
Eagle glass
Cu30Zn70/Cu30Al70 100˚C 5 2 um
Cu30Zn70/Cu30Al70
HS-150˚C
Corning
Eagle glass
Cu30Zn70/Cu30Al70 150˚C 5 2 um

118
Appendix B: Summary of Dealloyed Samples
Sample names and labels have been standardized from compositions and dates to the
format “letter-temperature-sample #”, where the letter corresponds to film composition,
temperature is the temperature of the stage, and sample number is the trial number for that film.
Example: “D-HS100-5” would be the fifth dealloyed sample from a film of composition D and a
stage heating condition of 100˚C. “F-HT300-1” would be the first dealloyed sample from a film
of composition F that was heat-treated at a temperature of 300˚C (where the film was heattreated before dealloying).
Table 9: Summary of Bilayer Cu Alloy Sample Names for Dealloying
Film Name Composition Layer Order Stage T
(˚C)
A Cu20Al80/Cu35Al65 Cu20Al80/Cu35Al65 N/A
B Cu35Al65/Cu20Al80 Cu35Al65/Cu20Al80 N/A
C Cu20Al80/Cu20Zn80 Cu20Al80/Cu20Zn80 N/A
D Cu20Zn80/Cu20Al80 Cu20Zn80/Cu20Al80 N/A
E Cu20Zn80/Cu30Al70 Cu20Zn80/Cu30Al70 N/A
F Cu30Al70/Cu20Al80 Cu30Al70/Cu20Al80 N/A
G Cu35Zn65/Cu35Al65 Cu35Zn65/Cu35Al65 N/A
H Cu35Al65/Cu35Zn65 Cu35Al65/Cu35Zn65 N/A
C HS-100 Cu20Al80/Cu20Zn80 Cu20Al80/Cu20Zn80 100˚C
C HS-150 Cu20Al80/Cu20Zn80 Cu20Al80/Cu20Zn80 150˚C
C HS-300 Cu20Al80/Cu20Zn80 Cu20Al80/Cu20Zn80 300˚C
D HS-100 Cu20Zn80/Cu20Al80 Cu20Zn80/Cu20Al80 100˚C
D HS-150 Cu20Zn80/Cu20Al80 Cu20Zn80/Cu20Al80 150˚C
D HS-300 Cu20Zn80/Cu20Al80 Cu20Zn80/Cu20Al80 300˚C
G HS-100 Cu35Zn65/Cu35Al65 Cu35Zn65/Cu35Al65 100˚C
G HS-150 Cu35Zn65/Cu35Al65 Cu35Zn65/Cu35Al65 150˚C
G HS-300 Cu35Zn65/Cu35Al65 Cu35Zn65/Cu35Al65 300˚C
H HS-100 Cu35Al65/Cu35Zn65 Cu35Al65/Cu35Zn65 100˚C
H HS-150 Cu35Al65/Cu35Zn65 Cu35Al65/Cu35Zn65 150˚C
H HS-300 Cu35Al65/Cu35Zn65 Cu35Al65/Cu35Zn65 300˚C
I HS-100 Cu30Zn70/Cu30Al70 Cu30Zn70/Cu30Al70 100˚C
I HS-150 Cu30Zn70/Cu30Al70 Cu30Zn70/Cu30Al70 150˚C

119
Table 10: Summary of Dealloyed Bilayer Cu Samples From Unheated Films
Sample
Name
Composition
Percentage
Electrolyte Applied
Potential
(mV)
Dealloying
Time (hr)
Notes
A-1 Cu20Al80/Cu35Al65 conc. H3PO4 N/A 0.17 retrieved
A-2 Cu20Al80/Cu35Al65 conc. H3PO4 N/A overnight dissolved
A-3 Cu20Al80/Cu35Al65 conc. H3PO4 N/A 1.17 retrieved
A-4 Cu20Al80/Cu35Al65 conc. H3PO4 N/A 1.33 retrieved
A-5 Cu20Al80/Cu35Al65 conc. H3PO4 N/A 2 retrieved
A-6 Cu20Al80/Cu35Al65 conc. H3PO4 N/A 2.5 some broke,
but retrieved
A-7 Cu20Al80/Cu35Al65 conc. H3PO4 N/A 2.5 wrinkled,
retrieved
A-8 Cu20Al80/Cu35Al65 conc. H3PO4 N/A 0.5 retrieved
A-9 Cu20Al80/Cu35Al65 conc. H3PO4 N/A 2 retrieved
A-10 Cu20Al80/Cu35Al65 conc. H3PO4 N/A 2 retrieved
A-11 Cu20Al80/Cu35Al65 conc. H3PO4 N/A 1 retrieved
A-12 Cu20Al80/Cu35Al65 conc. H3PO4 N/A 1 retrieved
A-13 Cu20Al80/Cu35Al65 conc. H3PO4 N/A 0.75 retrieved
A-14 Cu20Al80/Cu35Al65 conc. H3PO4 N/A 0.5 retrieved
B-1 Cu35Al65/Cu20Al80 conc. H3PO4 N/A 2 delaminated
B-2 Cu35Al65/Cu20Al80 conc. H3PO4 N/A 2 delaminated
B-3 Cu35Al65/Cu20Al80 conc. H3PO4 N/A 0.25 retrieved
B-4 Cu35Al65/Cu20Al80 conc. H3PO4 N/A 2 dissolved
B-5 Cu35Al65/Cu20Al80 conc. H3PO4 N/A 2 dissolved
B-6 Cu35Al65/Cu20Al80 conc. H3PO4 N/A 1.75 retrieved
B-7 Cu35Al65/Cu20Al80 conc. H3PO4 N/A 1.83 dissolved
B-8 Cu35Al65/Cu20Al80 conc. H3PO4 N/A 0.5 retrieved
B-9 Cu35Al65/Cu20Al80 conc. H3PO4 N/A 1 retrieved
B-10 Cu35Al65/Cu20Al80 conc. H3PO4 N/A 0.75 retrieved
C-1 Cu20Al80/Cu20Zn80 conc. H3PO4 N/A 1.17 dissolved
C-2 Cu20Al80/Cu20Zn80 conc. H3PO4 N/A 1 retrieved
C-3 Cu20Al80/Cu20Zn80 conc. H3PO4 N/A 0.83 dissolved
C-4 Cu20Al80/Cu20Zn80 conc. H3PO4 N/A 0.67 dissolved
C-5 Cu20Al80/Cu20Zn80 conc. H3PO4 N/A 0.75 dissolved
C-6 Cu20Al80/Cu20Zn80 conc. H3PO4 N/A 0.67 dissolved
C-7 Cu20Al80/Cu20Zn80 conc. H3PO4 N/A 0.5 retrieved
C-8 Cu20Al80/Cu20Zn80 conc. H3PO4 N/A 0.75 retrieved
C-9 Cu20Al80/Cu20Zn80 conc. H3PO4 N/A 1 dissolved
C-10 Cu20Al80/Cu20Zn80 conc. H3PO4 N/A 1.17 dissolved
C-11 Cu20Al80/Cu20Zn80 conc. H3PO4 N/A 1.17 dissolved
C-12 Cu20Al80/Cu20Zn80 conc. H3PO4 N/A 1.08 dissolved
C-13 Cu20Al80/Cu20Zn80 conc. H3PO4 N/A 1.08 dissolved
C-14 Cu20Al80/Cu20Zn80 conc. H3PO4 N/A 1.08 dissolved

120
C
-15 Cu20Al80/Cu20Zn80 conc. H
3PO
4 N/A 1.08 dissolved
C
-16 Cu20Al80/Cu20Zn80 conc. H
3PO
4 N/A 1.08 dissolved
D
-
1 Cu20Zn80/Cu20Al80 conc. H
3PO
4 N/A 0.75 dissolved
D
-
2 Cu20Zn80/Cu20Al80 conc. H
3PO
4 N/A 0.67 dissolved
D
-
3 Cu20Zn80/Cu20Al80 conc. H
3PO
4 N/A 0.75 dissolved
D
-
4 Cu20Zn80/Cu20Al80 conc.
H
3PO
4 N/A 0.83 dissolved
D
-
5 Cu20Zn80/Cu20Al80 conc. H
3PO
4 N/A 0.5 retrieved
D
-
6 Cu20Zn80/Cu20Al80 conc. H
3PO
4 N/A 0.67 retrieved
D
-
7 Cu20Zn80/Cu20Al80 conc. H
3PO
4 N/A 0.75 dissolved
D
-
8 Cu20Zn80/Cu20Al80 conc. H
3PO
4 N/A 0.5 retrieved
D
-
9 Cu20Zn80/Cu20Al80 conc. H
3PO
4 N/A 0.75 retrieved
D
-10 Cu20Zn80/Cu20Al80 conc. H
3PO
4 N/A
1 dissolved
D
-11 Cu20Zn80/Cu20Al80 conc. H
3PO
4 N/A 1.17 dissolved
D
-12 Cu20Zn80/Cu20Al80 conc. H
3PO
4 N/A
1 dissolved
D
-13 Cu20Zn80/Cu20Al80 conc. H
3PO
4 N/A
1 dissolved
D
-14 Cu20Zn80/Cu20Al80 conc. H
3PO
4 N/A 1.08 dissolved
D
-15 Cu20Zn80/Cu20Al80 conc. H
3PO
4 N/A 0.83 retrieved,
larger
starting
sample
D
-16 Cu20Zn80/Cu20Al80 conc. H
3PO
4 N/A 0.92 dissolved
D
-17 Cu20Zn80/Cu20Al80 conc. H
3PO
4 N/A
1 dissolved
D
-18 Cu20Zn80/Cu20Al80 conc. H
3PO
4 N/A 0.83 dissolved
D
-19 Cu20Zn80/Cu20Al80 conc. H
3PO
4 N/A 0.83 dissolved
D
-20 Cu20Zn80/Cu20Al80 conc. H
3PO
4 N/A 0.83 dissolved
D
-21 Cu20Zn80/Cu20Al80 conc. H
3PO
4 N/A 0.83 dissolved
E
-
1 Cu20Zn80/Cu30Al70 conc. H
3PO
4 N/A 15 retrieved
E
-
2 Cu20Zn80/Cu30Al70 conc. H
3PO
4 N/A overnight dissolved
F
-
1 Cu30Al70/Cu20Al80 conc. H
3PO
4 N/A
1 dissolved
F
-
2 Cu30Al70/Cu20Al80 5 wt% HCl N/A 0.5 retrieved
F
-
3 Cu30Al70/Cu20Al80 5 wt% HCl N/A
1 dissolved
F
-
4 Cu30Al70/Cu20Al80 5 wt% HCl N/A 0.33 retrieved
F
-
5 Cu30Al70/Cu20Al80 0.1M NaOH N/A 1.5 retrieved
F
-
6 Cu30Al70/Cu20Al80 0.1M NaOH N/A overnight retrieved
G
-
1 Cu35Zn65/Cu35Al65 conc. H
3PO
4 N/A 0.33 retrieved
G
-
2 Cu35Zn65/Cu35Al65 conc. H
3PO
4 N/A 0.5 retrieved
G
-
3 Cu35Zn65/Cu35Al65 2M
H
3PO
4 N/A
1 retrieved
G
-
4 Cu35Zn65/Cu35Al65 2M H
2SO
4 N/A
1 no change,
retrieved
G
-
5 Cu35Zn65/Cu35Al65
0.01M H3PO4
N/A 18 no change,
retrieved
H
-
1 Cu35Al65/Cu35Zn65 conc. H
3PO
4 N/A 0.33 retrieved
H
-
2 Cu35Al65/Cu35Zn65 conc. H
3PO
4 N/A 0.5 retrieved
H
-
3 Cu35Al65/Cu35Zn65 2M
H
3PO
4 N/A
1 retrieved
H
-
4 Cu35Al65/Cu35Zn65 2M H
2SO
4 N/A
1 retrieved

121
H-5 Cu35Al65/Cu35Zn65 0.01M
H3PO4
N/A 18 delaminated
Table 11: Summary of Dealloyed Bilayer Cu Samples From Heat-Treated Films
Sample
Name
Composition
Percentage
Electrolyte Applied
Potential
(mV)
Dealloying
Time (hr)
Notes
F-HT500-1 Cu30Al70/Cu20Al80 conc. H3PO4 N/A 1 dissolved
F-HT500-2 Cu30Al70/Cu20Al80 5 wt% HCl N/A 0.5 retrieved
F-HT500-3 Cu30Al70/Cu20Al80 5 wt% HCl N/A 1 dissolved
F-HT500-4 Cu30Al70/Cu20Al80 5 wt% HCl N/A 0.33 dissolved
F-HT300-1 Cu30Al70/Cu20Al80 5 wt% HCl N/A 0.33 sample
flipped, nc
F-HT300-2 Cu30Al70/Cu20Al80 5 wt% HCl N/A 0.33 dissolved
Table 12: Summary of Dealloyed Bilayer Cu Samples From Heated Stage Films
Sample
Name
Composition
Percentage
Electrolyte Dealloying
Time (hr)
Starting
Mass (mg)
Notes
C-HS100-1 Cu20Al80/Cu20Zn80 0.1M H3PO4 1 N/A delaminated
C-HS100-2 Cu20Al80/Cu20Zn80 0.1M H3PO4 1.25 N/A delaminated
C-HS100-3 Cu20Al80/Cu20Zn80 0.1M H3PO4 1.5 N/A delaminated
C-HS100-4 Cu20Al80/Cu20Zn80 0.1M H3PO4 0.75 N/A delaminated
C-HS100-5 Cu20Al80/Cu20Zn80 0.1M H3PO4 0.75 N/A dissolved
D-HS100-1 Cu20Zn80/Cu20Al80 0.1M H3PO4 1.5 N/A retrieved
D-HS100-2 Cu20Zn80/Cu20Al80 0.1M H3PO4 2 N/A retrieved
D-HS100-3 Cu20Zn80/Cu20Al80 0.1M H3PO4 2.25 N/A retrieved
D-HS100-4 Cu20Zn80/Cu20Al80 0.1M H3PO4 2 N/A retrieved
D-HS100-5 Cu20Zn80/Cu20Al80 0.1M H3PO4 2 N/A retrieved
D-HS100-6 Cu20Zn80/Cu20Al80 0.1M H3PO4 2 N/A retrieved
D-HS100-7 Cu20Zn80/Cu20Al80 0.1M H3PO4 2.25 N/A retrieved
D-HS100-8 Cu20Zn80/Cu20Al80 0.1M H3PO4 2.25 N/A retrieved
D-HS100-9 Cu20Zn80/Cu20Al80 0.1M H3PO4 2.5 N/A retrieved
D-HS100-10 Cu20Zn80/Cu20Al80 0.1M H3PO4 2.5 N/A retrieved
D-HS100-11 Cu20Zn80/Cu20Al80 0.1M H3PO4 1.5 4.0 retrieved
D-HS100-12 Cu20Zn80/Cu20Al80 0.1M H3PO4 1.5 3.0 retrieved
D-HS100-13 Cu20Zn80/Cu20Al80 0.1M H3PO4 2 5.0 retrieved
D-HS100-14 Cu20Zn80/Cu20Al80 0.1M H3PO4 2.5 8.0 retrieved
D-HS100-15 Cu20Zn80/Cu20Al80 0.1M H3PO4 2.5 12.0 retrieved
D-HS100-16 Cu20Zn80/Cu20Al80 0.1M H3PO4 2 1.7 retrieved
D-HS100-17 Cu20Zn80/Cu20Al80 0.1M H3PO4 2 1.2 retrieved

122
D
-HS100
-18 Cu20Zn80/Cu20Al80 0.1M H
3PO
4 3.17 6.8 retrieved
D
-HS100
-19 Cu20Zn80/Cu20Al80 0.1M H
3PO
4 3.17 3.9 retrieved
D
-HS100
-20 Cu20Zn80/Cu20Al80 0.1M H
3PO
4
2 4.3 retrieved
D
-HS100
-21 Cu20Zn80/Cu20Al80 0.1M H
3PO
4 2.5 4.9 retrieved
D
-HS100
-22 Cu20Zn80/Cu20Al80 0.1M H
3PO
4 2.5 4.9 retrieved
D
-HS100
-23 Cu20Zn80/Cu20Al80 0.1M H
3PO
4
4 4.6 retrieved
D
-HS100
-24 Cu20Zn80/Cu20Al80 0.1M H
3PO
4 4.5 5.2 retrieved
D
-HS100
-25 Cu20Zn80/Cu20Al80 0.1M H
3PO
4
5 5.3 retrieved
D
-HS100
-26 Cu20Zn80/Cu20Al80 0.1M H
3PO
4 5.5 5.2 retrieved
D
-HS100
-27 Cu20Zn80/Cu20Al80 0.1M H
3PO
4 5.5 5.6 retrieved
D
-HS100
-28 Cu20Zn80/Cu20Al80 0.1M H
3PO
4
3 4.6 retrieved
D
-HS100
-29 Cu20Zn80/Cu20Al80 0.1M H
3PO
4 3.5 4.6 retrieved
D
-HS100
-30 Cu20Zn80/Cu20Al80 0.1M H
3PO
4
4 4.8 retrieved
D
-HS100
-31 Cu20Zn80/Cu20Al80 0.1M H
3PO
4 4.5
5 retrieved
D
-HS100
-32 Cu20Zn80/Cu20Al80 0.1M H
3PO
4 2.5 4.7 retrieved
D
-HS100
-33 Cu20Zn80/Cu20Al80 0.1M H
3PO
4 2.5 4.8 retrieved
D
-HS100
-34 Cu20Zn80/Cu20Al80 0.1M H
3PO
4 2.5 4.3 retrieved
D
-HS100
-35 Cu20Zn80/Cu20Al80 0.1M H
3PO
4
5 4.8 delaminated
D
-HS100
-36 Cu20Zn80/Cu20Al80 0.1M H
3PO
4
5 4.9 retrieved
D
-HS100
-37 Cu20Zn80/Cu20Al80 0.1M H
3PO
4 5.5 5.1 retrieved
D
-HS100
-38 Cu20Zn80/Cu20Al80 0.1M H
3PO
4
6 5.1 retrieved
D
-HS100
-39 Cu20Zn80/Cu20Al80 0.1M H
3PO
4
5 5.1 retrieved
D
-HS100
-40 Cu20Zn80/Cu20Al80 0.1M H
3PO
4
5 5.0 retrieved
D
-HS100
-41 Cu20Zn80/Cu20Al80 0.1M H
3PO
4
5 5.0 retrieved
D
-HS100
-42 Cu20Zn80/Cu20Al80 0.1M H
3PO
4
5 5.1 retrieved
D
-HS100
-43 Cu20Zn80/Cu20Al80 0.1M H
3PO
4
5 4.8 retrieved
D
-HS100
-44 Cu20Zn80/Cu20Al80 0.1M H
3PO
4
5 5.0 retrieved
D
-HS100
-45 Cu20Zn80/Cu20Al80 0.1M H
3PO
4
5 5.4 retrieved
D
-HS100
-46 Cu20Zn80/Cu20Al80 0.1M H
3PO
4
5 5.4 retrieved
D
-HS150
-
1 Cu20Zn80/Cu20Al80 0.1M H
3PO
4
1 2.7 retrieved
D
-HS150
-
2 Cu20Zn80/Cu20Al80 0.1M H
3PO
4
1 4.3 retrieved
D
-HS150
-
3 Cu20Zn80/Cu20Al80 0.1M H
3PO
4
1 1.9 retrieved
D
-HS150
-
4 Cu20Zn80/Cu20Al80 conc. H
3PO
4
1 2.8 retrieved
D
-HS150
-
5 Cu20Zn80/Cu20Al80 0.1M H
3PO
4 1.5 1.7 retrieved
D
-HS150
-
6 Cu20Zn80/Cu20Al80 0.1M H
3PO
4
2 1.9 retrieved
D
-HS150
-
7 Cu20Zn80/Cu20Al80 0.1M H
3PO
4 3.17 4.6 retrieved
D
-HS150
-
8 Cu20Zn80/Cu20Al80 0.1M H
3PO
4 3.17 4.4 retrieved
D
-HS150
-
9 Cu20Zn80/Cu20Al80 0.1M H
3PO
4 12 7.3 dissolved
D
-HS150
-10 Cu20Zn80/Cu20Al80 0.1M H
3PO
4 12 6.1 dissolved
D
-HS150
-11 Cu20Zn80/Cu20Al80 0.1M H
3PO
4 12 8.6 dissolved
D
-HS150
-12 Cu20Zn80/Cu20Al80 0.1M H
3PO
4
8 9.6 retrieved
D
-HS150
-13 Cu20Zn80/Cu20Al80 0.1M H
3PO
4
8 12.7 retrieved
D
-HS150
-14 Cu20Zn80/Cu20Al80 0.1M H
3PO
4
8 6.4 retrieved
D
-HS150
-15 Cu20Zn80/Cu20Al80 0.1M H
3PO
4
8 6.1 retrieved
D
-HS150
-16 Cu20Zn80/Cu20Al80 0.1M H
3PO
4
8 6.5 retrieved

123
D
-HS150
-17 Cu20Zn80/Cu20Al80 0.1M H
3PO
4
8 6.9 delaminated
D
-HS150
-18 Cu20Zn80/Cu20Al80 0.1M H
3PO
4 3.5 4.1 retrieved
D
-HS150
-19 Cu20Zn80/Cu20Al80 0.1M H
3PO
4 3.5 4.7 retrieved
D
-HS150
-20 Cu20Zn80/Cu20Al80 0.1M H
3PO
4
4 5.1 retrieved
D
-HS150
-21 Cu20Zn80/Cu20Al80 0.1M H
3PO
4
8 6.7 retrieved
D
-HS150
-22 Cu20Zn80/Cu20Al80 0.1M H
3PO
4
8 6.6 retrieved
D
-HS150
-23 Cu20Zn80/Cu20Al80 0.1M H
3PO
4
8 6.3 delaminated
D
-HS150
-24 Cu20Zn80/Cu20Al80 0.1M H
3PO
4
4 4.9 retrieved
D
-HS150
-25 Cu20Zn80/Cu20Al80 0.1M H
3PO
4
4
5 retrieved
D
-HS150
-26 Cu20Zn80/Cu20Al80 0.1M H
3PO
4
4 5.2 retrieved
G
-HS100
-
1 Cu35Zn65/Cu35Al65 2M H
3PO
4
1 N/A retrieved
G
-HS100
-
2 Cu35Zn65/Cu35Al65 2M H
2SO
4
1 N/A no change,
retrieved
G
-HS100
-
3 Cu35Zn65/Cu35Al65 0.01M H
3PO
4 15 N/A retrieved
G
-HS100
-
4 Cu35Zn65/Cu35Al65 0.01M H
3PO
4 18 N/A no change,
retrieved
G
-HS100
-
5 Cu35Zn65/Cu35Al65 0.1M H
3PO
4
4 N/A retrieved
G
-HS150
-
1 Cu35Zn65/Cu35Al65 conc. H
3PO
4 26 N/A dissolved
G
-HS150
-
2 Cu35Zn65/Cu35Al65 conc. H
2SO
4 26 N/A dissolved
G
-HS150
-
3 Cu35Zn65/Cu35Al65 2M H
3PO
4
1 N/A no change
G
-HS150
-
4 Cu35Zn65/Cu35Al65 2M H
2SO
4
1 N/A no change
G
-HS150
-
5 Cu35Zn65/Cu35Al65 0.01M H
3PO
4 15 N/A retrieved
G
-HS150
-
6 Cu35Zn65/Cu35Al65 0.01M H
3PO
4 18 N/A retrieved
G
-HS150
-
7 Cu35Zn65/Cu35Al65 0.1M H
3PO
4
4 N/A retrieved
G
-HS300
-
1 Cu35Zn65/Cu35Al65 conc. H
3PO
4 26 N/A retrieved
G
-HS300
-
2 Cu35Zn65/Cu35Al65 conc. H
2SO
4 26 N/A retrieved
H
-HS100
-
1 Cu35Al65/Cu35Zn65 0.01M H
3PO
4 18 N/A dissolved
H
-HS100
-
2 Cu35Al65/Cu35Zn65 0.1M H
3PO
4
4 N/A retrieved
H
-HS150
-
1 Cu35Al65/Cu35Zn65 conc. H
3PO
4 26 N/A dissolved
H
-HS150
-
2 Cu35Al65/Cu35Zn65 conc. H
2SO
4 26 N/A dissolved
H
-HS150
-
3 Cu35Al65/Cu35Zn65 2M H
3PO
4
1 N/A delaminated
H
-HS150
-
4 Cu35Al65/Cu35Zn65 2M H
2SO
4
1 N/A delaminated
H
-HS150
-
5 Cu35Al65/Cu35Zn65 0.01M H
3PO
4 15 N/A delaminated
H
-HS150
-
6 Cu35Al65/Cu35Zn65 0.01M H
3PO
4 18 N/A dissolved
H
-HS150
-
7 Cu35Al65/Cu35Zn65 0.1M H
3PO
4
4 N/A retrieved
H
-HS300
-
1 Cu35Al65/Cu35Zn65 conc. H
3PO
4 26 N/A dissolved
H
-HS300
-
2 Cu35Al65/Cu35Zn65 conc. H
2SO
4 26 N/A dissolved
I
-HS100
-
1 Cu30Zn70/Cu30Al70 0.1M H
3PO
4
1 3.3 no change,
retrieved
I
-HS100
-
2 Cu30Zn70/Cu30Al70 0.1M H
3PO
4
1 3.5 no change,
retrieved
I
-HS100
-
3 Cu30Zn70/Cu30Al70 0.1M H
3PO
4
1 3.3 no change,
retrieved
I
-HS100
-
4 Cu30Zn70/Cu30Al70 conc. H
3PO
4
1 2.9 no change,
retrieved

124
I-HS100-5 Cu30Zn70/Cu30Al70 0.1M H3PO4 24 1.7 dissolved
I-HS100-6 Cu30Zn70/Cu30Al70 0.1M H3PO4 2 2.8 no change,
retrieved
I-HS100-7 Cu30Zn70/Cu30Al70 conc. HCl 0.75 4.4 delaminated
I-HS100-8 Cu30Zn70/Cu30Al70 conc. HCl 0.75 5.4 delaminated
I-HS150-1 Cu30Zn70/Cu30Al70 0.1M H3PO4 1 3.5 retrieved
I-HS150-2 Cu30Zn70/Cu30Al70 0.1M H3PO4 1 3.2 retrieved
I-HS150-3 Cu30Zn70/Cu30Al70 0.1M H3PO4 1 5.8 retrieved
I-HS150-4 Cu30Zn70/Cu30Al70 0.1M H3PO4 1 6.6 retrieved
I-HS150-5 Cu30Zn70/Cu30Al70 conc. H3PO4 1 2.6 no change,
retrieved
I-HS150-6 Cu30Zn70/Cu30Al70 0.1M H3PO4 24 2.1 dissolved
I-HS150-7 Cu30Zn70/Cu30Al70 0.1M H3PO4 2 2.2 no change,
retrieved

Abstract (if available)

Abstract Understanding nanoscale interfaces is critical for designing new materials to meet current and future scientific needs in applications such as sensing and catalysis. These interfaces, such as grain-boundaries, comprise a high volume fraction of nanostructured materials, and changing the type or quantity of interfaces can produce large variations in properties such as strength or corrosion resistance. Previous studies on the corrosion response of nanoscale interfaces have mainly focused on the effects on mechanical behavior. Thus, there is a pressing need to explore other aspects of corrosion behavior in nanoscale interfaces, specifically corrosion resistance and dealloying, in order to be able to design specific microstructures with the desired corrosion response.

This dissertation investigates the relationship between engineered nanoscale interfaces and corrosion in order to illuminate the relationship between processing and microstructure. This was investigated with: (1) the synthesis and evaluation of grain boundary-engineered 5xxx series Al-Mg films to improve corrosion resistance in seawater, (2) the synthesis of gradient nanoporous Cu via a dealloying corrosion process and subsequent investigation of the effect of a sharp relative density or chemical potential interface on the association between processing and foam morphology, and (3) the investigation of graded interfaces on nanoporous Cu microstructure and the suitability of these materials for SERS-based sensing. By identifying key links between processing and microstructure for these interfaces, these studies provide critical insight for understanding corrosion in complex material systems and how to leverage this knowledge for future controlled and reliable synthesis of these materials.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

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

PDF

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

PDF

Tailoring compositional and microstructural complexity in nanostructured alloys

PDF

Development and characterization of hierarchical cellular structures

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

PDF

Modeling of bicontinuous nanoporous materials and investigation of their mechanical and hydraulic properties

PDF

Synthesis, characterization, and mechanical properties of nanoporous foams

PDF

Mechanical behavior and deformation of nanotwinned metallic alloys

PDF

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

PDF

Expanding sputtering synthesis domains for the discovery of novel film microstructures and morphologies

PDF

Mechanical behavior and microstructure optimization of ultrafine-grained aluminum alloys and nanocomposites

PDF

Flow and thermal transport at porous interfaces

PDF

Excited state process in perovskites

PDF

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

PDF

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

PDF

An experimental study of the elastic theory of granular flows

PDF

Molecular orientation and structure of hydrogen bonds at water interfaces

PDF

Fabrication and packaging of three-dimensional Parylene C neural interfaces

PDF

Polymer flow for manufacturing fiber reinforced polymer composites

Asset Metadata

Creator Hemmendinger, Karina Dora Haugland (author)

Core Title Corrosion studies of engineered interfaces

School Viterbi School of Engineering

Degree Doctor of Philosophy

Degree Program Mechanical Engineering

Degree Conferral Date 2024-05

Publication Date 02/05/2024

Defense Date 01/12/2024

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

Tag Boundaries,dealloying,gradients,grain boundaries,interfaces,nanoporous,OAI-PMH Harvest,SERS,sputtering

Format theses (aat)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Hodge, Andrea (committee chair), Branicio, Paulo (committee member), Dawlaty, Jahan (committee member), Luhar, Mitul (committee member)

Creator Email khemmend@gmail.com,khemmend@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-oUC113825946

Unique identifier UC113825946

Identifier etd-Hemmending-12654.pdf (filename)

Legacy Identifier etd-Hemmending-12654

Document Type Dissertation

Format theses (aat)

Rights Hemmendinger, Karina Dora Haugland

Internet Media Type application/pdf

Type texts

Source 20240208-usctheses-batch-1125 (batch), 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 author, as the original true and official version of the work, but does not grant the reader permission to use the work if the desired use is covered by copyright. It is the author, as rights holder, who must provide use permission if such use is covered by copyright.

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

Repository Email cisadmin@lib.usc.edu