Corrosion of High-Entropy Alloys in Chloride Solutions
THESIS Presented in Partial Fulfillment of the Requirements for the Degree Master of Science in The Graduate School of The Ohio State University
By Orion Swanson Graduate Program in Materials Science and Engineering The Ohio State University 2018
Master's Examination Committee: Dr. Gerald Frankel, Advisor Dr. Christopher Taylor
Copyrighted by Orion J. Swanson 2018
ABSTRACT
High entropy alloys (HEAs) are a new class of alloys composed of five or more major alloying elements, each making up between 5 at% and 35 at% of the total composition. HEAs have shown promising characteristics of superior corrosion resistance, and have a huge possible compositional space, which is largely unexplored. The intent of this project is to develop greater understanding of how composition influences the corrosion resistance of HEAs in chloride solutions. An integrated computational materials engineering (ICME) approach has been adopted to produce single phase corrosion resistant HEAs. A combined empirical and computational approach was employed to generate HEAs with variations in composition and corrosion resistance. Initial investigation focused on a Ni-rich HEA, Ni38Cr21Fe20Mo6W2Ru13, which has shown exceptional corrosion resistance in concentrated acidic chloride environments. The alloy was tested potentiodynamically in strong HCl solutions, showing spontaneous passivity until high voltages, where corrosion was transpassive and no signs of localized corrosion.
The study also investigated the influence of Cr-content on pitting/crevice corrosion of
Ni38Fe20Crx(MnCo)42-x HEAs in chloride solutions. The range of Cr content was from 20 to 5 wt%. Arc-melted button samples of four non-equimolar single-phase HEAs were synthesized and homogenized. As expected, initial electrochemical testing showed a trend of decreasing corrosion resistance as Cr content decreased, with corresponding increase in Mn and Co content.
ii
Interestingly, even the low Cr HEAs (below 12 wt.% Cr) were passive at the open circuit potential in 0.6 M NaCl solution with an extensive passive region, suggesting enhanced corrosion resistance compared to conventional corrosion resistant alloys. On the other hand, these alloys are extremely susceptible to crevice corrosion during anodic polarization, to the extent that it is difficult to measure pitting potentials in the absence of crevice artifacts. Thus, a novel droplet cell was utilized to be able to measure the pitting potentials. Pitting with no crevicing was seen with the use of the novel droplet cell, and the pits were crystallographic in all alloys in the series. It is hypothesized that the crystallographic nature of these pits are caused by the inability of HEAs to generate a salt film, as well as weak passive films, which do not adequately prevent metal cation diffusion. The source of the passivity of the low chromium alloys remains a subject for further study.
iii
This work is dedicated to the Swansons.
iv
ACKNOWLEDGEMENTS I would like to thank my advisor, Prof. Jerry Frankel for giving me the opportunity to study the field and gain experience that I know will prove valuable to me in the future. I also appreciate all of the support, which I have received over the course of pursuing this degree. I would like to thank all FCC members, and all members of the WastePD EFRC. Particularly, Dr.
Tianshu Li, who has guided me through experimentation and research.
I would also like to thank all of my family and friends, who have always been there to support me. Specifically, I would like to thank my mother, father, and sister for all of their support throughout this process. Thank you, for giving me a home and place where I have always been supported to follow my dreams. For welcoming me back into your home to pursue my dreams to grow as a student and as a person. I would like to thank all of my friends, who have supported me every time I was in need, or just needed a great person with whom to have a conversation. I am in debt to all of you, and everyone who I have not mentioned here, but played a role in my success during this endeavor.
v
VITA May 2016………………………………………………..B.S (Biomedical Engineering) The Ohio State University August 2016 – present…………………………………..M.S. (Material Science and Engineering) The Ohio State University
Field of Study Major Field: Materials Science and Engineering
vi
Table of Contents
ABSTRACT ...... ii
ACKNOWLEDGEMENTS ...... v
VITA ...... vi
LIST OF TABLES ...... xi
LIST OF FIGURES ...... xii
INTRODUCTION ...... 1
LITERATURE REVIEW ...... 3
2.1 INTRODUCTION ...... 3
2.2 DEFINING HIGH ENTROPY ALLOYS ...... 3
2.2.1 Alloy Structure ...... 3
2.2.2 Corrosion Resistance ...... 4
2.3 OVERVIEW ...... 6
2.4 LOCALIZED CORROSION OF HEAs ...... 6
2.4.1 Pitting Corrosion...... 6
2.5 CURRENT ALLOY PROCESSING AND GENERATION ...... 7
2.5.1 Processing of single phase alloys ...... 8
2.6 CORROSION BEHAVIOR OF HEAs ...... 10
2.6.1 Corrosion resistance of HEAs in chloride solutions ...... 10
2.6.2 Corrosion Resistance of HEAs in sulfide solutions...... 11
vii
2.7 EFFECTS OF ALLOYING ON HEA CORROSION RESISTANCE ...... 12
2.7.1 Effects of Cr content on HEAs ...... 12
2.7.2 Effects of Mo alloying ...... 13
2.7.3 Effects of Al, Co, and Mn...... 14
2.7.4 Effects of Ni alloying ...... 14
CORROSION BEHAVIOR OF Ni38Cr21Fe20Mo6W2Ru13 HIGH ENTROPY ALLOY IN
CHLORIDE ENVIRONMENTS ...... 18
3.1 INTRODUCTION ...... 18
3.2 METHODS...... 22
3.2.1 Alloy Characterization ...... 22
3.2.2 Electrochemical Corrosion Characterization ...... 23
3.3 RESULTS...... 25
3.3.1 Microstructure of Ni38Cr21Fe20Mo6W2Ru13 ...... 25
3.3.2 Corrosion behavior of Ni38Cr21Fe20Mo6W2Ru13 as cast and homogenized alloy,
Immersion ...... 26
3.3.3 Potentiodynamic polarization measurements and corrosion morphologies ...... 27
3.3.4 Linear polarization resistance and 24h corrosion potential measurements in 12 M HCl
...... 28
3.3.5 Critical Pitting Temperature ...... 29
3.4 DISCUSSION ...... 29
viii
3.4.1 Superior pitting resistance based on passive film breakdown ...... 30
3.4.2 Superior pitting resistance based on pit growth stability ...... 31
3.4.3 Critical dissolution current density ...... 32
3.4.4 Maximum dissolution current density ...... 32
CORROSION BEHAVIOR OF Ni38Fe20Crx(MnCo)42-x HIGH ENTROPY ALLOYS IN 0.6M
NACL SOLUTION ...... 46
4.1 INTRODUCTION ...... 46
4.2 EXPERIMENTAL ...... 48
4.2.1 Specimen preparation ...... 48
4.2.1 Microstructure characterization ...... 49
4.2.2 Electrochemical Corrosion Characterization ...... 50
4.3 RESULTS AND DISCUSSION ...... 52
4.3.1 Characterization ...... 53
4.3.2 Conventional immersion cell testing ...... 53
4.3.3 Droplet cell experiments ...... 56
4.3.4 Crystallographic pits in HEAs ...... 58
4.3.5 Mechanism for formation of crystallographic pit morphologies on canary alloys ...... 60
4.4 CONCLUSIONS ...... 63
CONCLUSIONS AND FUTURE WORK ...... 77
5.1 CONCLUSIONS ...... 77
ix
5.2 FUTURE WORK ...... 78
BIBLIOGRAPHY ...... 80
x
LIST OF TABLES Table 3.1: Chemical Composition (wt.%) of material ...... 34
Table 4. 1: Chemical Composition (wt%) of materials...... 65
xi
LIST OF FIGURES Figure 3.1: CALPHAD generated pseudo-binary phase diagram, provided by J. Saal and P. Lu from QuesTek...... 34
Figure 3.2: Image of as cast Ni38Cr21Fe20Mo6W2Ru13 HEA as cast HEA with dendritic regions
(DR) and interdendritic regions (IR) labeled...... 35
Figure 3.3: The large grain size of the HEA after 1250°C 120 h heat treatment...... 35
Figure 3.4: Typical SEM micrograph of the As Cast Ni38Cr21Fe20Mo6W2Ru13 alloy, segregation and pores visible...... 36
Figure 3.5: EDS mapping of as-cast sample, displaying typical dendritic microstructure...... 36
Figure 3.6: Typical SEM micrograph of the heat treated Ni38Cr21Fe20Mo6W2Ru13 alloy with uniform single phase microstructure and no porosity...... 37
Figure 3.7: EDS mapping results of homogenized Ni38Cr21Fe20Mo6W2Ru13 sample...... 37
Figure 3.8: XRD pattern of as-cast HEA with peaks corresponding to typical FCC structure. ... 38
Figure 3.9: Optical images of HEA after 24-h immersion in 12 M HCl, ~25°C a) as-cast b) homogenized...... 38
Figure 3.10: Potentiodynamic Polarization of HEA in 0.6 M NaCl, 30 oC...... 39
Figure 3.11: Optical images of corrosion morphology of as-cast sample...... 39
Figure 3.12: Stepped polarization corrosion morphology of as-cast HEA showing the preferential attack at the interdendritic regions...... 40
Figure 3.13: Optical image of homogenized HEA1 corrosion morphology...... 40
Figure 3.14: Typical polarization curves of Ni38Cr21Fe20Mo6W2Ru13 at 30°C in solution 1 M
HCl, 6 M HCl, 12 M HCl. The curve in 0.6 M NaCl is included for comparison...... 41
Figure 3.15: Polarization resistance over 24 h immersion in 12 M HCl at 30°C, as-cast and homogenized HEA...... 41
xii
Figure 3.16: 24 h OCP in 12 M HCl...... 42
Figure 3.17: Potentiostatic CPT measurement at 700 mV in 3.5% NaCl on both as-cast and homogenized HEAs...... 42
Figure 3.18: Magnified potentiostatic CPT curves...... 43
Figure 3.19: Current transients compared to commercial SS316L...... 43
Figure 4.1:CALPHAD pseudo-binary phase diagram for Ni38Fe20Crx(MnCo)42-x Information provided by J. Saal and P. Lu, QuesTek...... 65
Figure 4.2: SEM images of homogenized a)Ni38Fe20Cr22Mn10Co10 b) Ni38Fe20Cr14Mn14Co14 c)Ni38Fe20Cr10Mn16Co16 d)Ni38Fe20Cr6Mn18Co18...... 66
Figure 4.3: EDS mapping of CanCr22 after homogenization, showing even distribution of element...... 66
Figure 4.4: XRD peak analysis of CanCr22, 14, 10, and 6 all with representative FCC structure peaks...... 67
Figure 4.5: Potentiodynamic polarization curves of the canary alloys in 0.6M NaCl at 30°C in a typical immersion cell...... 67
Figure 4.6: Typical SEM images of crevice corrosion found after polarization in 0.6M NaCl at
30°C in a typical immersion cell...... 68
Figure 4.7: Typical cyclic polarization curves of the canary alloys in 0.6M NaCl at 30°C in a typical immersion cell...... 68
Figure 4.8: Graphical representation of chromium alloying effect on critical crevice corrosion metrics, Ecorr, Ecrev, and Erp, in Ni38Fe20Crx(MnCo)42-x...... 69
Figure 4.9: CPT measurements scanning from an initial temperature of 4°C to 16°C...... 69
Figure 4.10: Potentiodynamic polarization of CanCr22 in 3.6M NaCl at -15°C...... 70
xiii
Figure 4.11: 30 min OCP of the canary alloys in 0.6M NaCl at ~25°C performed with the hanging droplet cell...... 70
Figure 4.12: Typical potentiodynamic polarization curves of the canary alloys in 0.6 M NaCl at
~25°C performed with the hanging droplet cell...... 71
Figure 4.13: Pit morphologies observed on canary alloys after polarization with use of the hanging droplet cell. a) Pits along the waterline; CanCr6 b) Small pits near the waterline; CanCr6 c) A large number of pits initiated near at the waterline edge; CanCr10 d) Multiple pits in close proximity near the waterline edge; CanCr10 e) Single large crystallographic pit; CanCr6 f)
Single large crystallographic pit; CanCr10 g) Single large crystallographic pit; CanCr14 h)
Single large crystallographic pit; CanCr22 ...... 72
Figure 4.14: Typical cyclic polarization curves of the canary alloys in 0.6 M NaCl at ~25°C performed with the hanging droplet cell...... 73
Figure 4.15.: Graphical representation of chromium alloying effect on critical pitting corrosion metrics, Ecorr, Epit, and Erp, in Ni38Fe20Crx(MnCo)42-x...... 73
Figure 4.16: Potentiodynamic polarization curve of CanCr22 at a surface temperature of -6°C in
0.6M NaCl performed with the hanging droplet cell...... 74
Figure 4.17: OM images of A) lacy pit cover prior to sonication formed during polarization of
CanCr22 at a surface temperature of -6°C in 0.6M NaCl B) Pit after sonication and cleaning. . 74
xiv
Chapter 1
INTRODUCTION
High entropy alloys (HEAs) are a new family of alloys that are now a focus of study in the fields of corrosion and metallurgy. HEAs rely on high configurational entropies because of multiple principal alloying elements, as opposed to traditional alloying where elements are alloyed with a single principal alloying element. This new alloying methodology has greatly expanded the study of new alloys due to the vast compositional space now available for study.
High entropy alloys have shown enhanced physical and chemical properties, making them of interest in a wide range of industrial uses.
The goal of this study is to investigate the underlying corrosion and electrochemical science that governs the dissolution of HEAs. One objective was to characterize the corrosion nature of Ni38-Cr21-Fe20-Mo6-W2-Ru13 in chloride solutions and study effects of alloy composition and processing. A subsequent objective was aimed at understanding the effect of chromium content in HEAs on the localized corrosion behavior.
This thesis consists of 5 chapters including this chapter as an introduction to the work.
Chapter 2 is a literature review, which provides a review of literature relating to the structure, processing, and corrosion characteristics of HEAs, including their general and pitting corrosion characteristics. Chapter 3 focuses on the corrosion of the Ni-based HEA,
Ni38Cr21Fe20Mo6W2Ru13, tested in chloride environments ranging from moderately to extremely aggressive environments. The alloy is shown to be extremely resistant to corrosion in solutions with high acidity and chloride content as well as at high temperature. Chapter 4 examines the effects of chromium content and its effect on the corrosion resistance of Ni38Fe20Crx(MnCo)42-x
1
HEA. A new method of avoiding crevice corrosion to allow examination of the pitting corrosion properties is applied.
Chapter 5 concludes the findings of the study with regards to the successes of corrosion resistant alloy design of HEAs. The effects of chromium alloying in HEAs, and new findings regarding localized corrosion of HEAs.
2
Chapter 2
LITERATURE REVIEW
2.1 INTRODUCTION Corrosion affects materials in a variety of industries, leading to material failure during the service life of products and equipment. The failure and replacement of current materials in service comes at great economic cost, the US alone spent $1 trillion in 2013 or 6.2% of the gross domestic product (GDP) on corrosion related costs, while globally corrosion accounts for about
3% of the global GDP[1]. The development of corrosion resistant materials and coatings capable of extending the lifespan of equipment and structures can greatly reduce current economic expenses. Previous research of corrosion resistant materials has been largely focused on stainless steels, nickel-based alloys, as well as titanium alloys using conventional metallurgy, where low concentrations of beneficial alloying additions such as Cr, Mo, Ni, W, N have been used to improve the corrosion resistance of the primary alloy. The enhanced corrosion resistance of these alloys is the result of the ability of these alloying additions to generate a protective oxide layer, which prevents corrosion of the underlying metal surface. The enhanced corrosion resistance of these alloys has made them staples in a wide range of industries, from food and chemical production to nuclear power and waste containment.
2.2 DEFINING HIGH ENTROPY ALLOYS 2.2.1 Alloy Structure
Multiple-element systems have been an area of interest since they were proposed by Yeh et al. [2]. These alloy systems have since been known as high-entropy alloys (HEA). HEAs are traditionally considered to be single phase solid solutions of five or more elements at or near
3 equimolar composition [2, 3]. However, the literature considers a wide variety of multiphase alloys as meeting the criteria for an HEA. HEAs have been of interest because of their unique properties, such as high strength and hardness, exceptional wear properties, as well as high temperature strength, all of which could prove to be beneficial in a variety of industrial applications. In contrast to conventional corrosion-resistant alloys, HEAs consist of multiple principal alloying elements. In theory, multiple principal alloying elements allow for extensive compositional freedom as well as greater property control compared to traditional primary element alloys. Due to the high entropy mixing effect [2], where the tendency of an alloy to order or segregate is greatly reduced due to the entropy of mixing, alloys composed of greater numbers of primary elements are therefore, less likely to form ordered structures. Consequently,
HEAs have a tendency to form disordered solid solutions of face-centered cubic (FCC), body- centered cubic (BCC), or hexagonal close-pack (HCP) structures [4-6], instead of more complex intermetallic phases. The locally disordered chemical environment, which results from the random arrangement of constituent elements within these solid solutions has been theorized to lead to the unique corrosion properties of HEAs [4].
2.2.2 Corrosion Resistance
Conventional corrosion resistant alloys, such as stainless steels, rely on formation of a protective surface oxide layer. Therefore, HEAs with compositions similar to stainless steel or other corrosion-resistant alloys will show similar corrosion resistance and oxide formation.
Stainless steels and their alloying have been studied for the past several decades due to their corrosion resistance and the nature of their passive films. Several elements such as Cr and W, have been identified as strong oxide formers, which when added to stainless steels, contribute to
4 their corrosion resistance. Other elements, such as Mo and N, also improve corrosion resistance, but likely by dissolving and then altering the local pit environment. Stainless steels and other alloys have been extensively modeled by empirical methods, which have generated tools for industry design of corrosion-resistant alloys. One such tool, the pitting resistance equivalent
(PRE) number (or PREN) [7-11], is described by [7]:
푃푅퐸푁 = 푤푡. %퐶푟 + 1.6푤푡. %푊 + 3.3푤푡. %푀표 + 16푤푡. %푁 퐸푞. (1)
This equation was created from large amounts of empirical data to guide the selection of stainless steel alloys for applications where stainless steels would be susceptible to localized corrosion. The knowledge of all possible alloying elements, which are key factors in the determination of an alloy’s PREN, have not been studied to the same degree, leaving a large knowledge gap in the prediction of HEA corrosion resistance. HEAs possessing similar oxide forming elements as conventional corrosion resistant alloys such as Cr, W etc., which are capable of oxide layer formation, may greatly reduce the corrosion rate of a material. However, high-entropy alloys, in contrast to conventional alloys, have the ability to be comprised of considerably larger compositional fractions of beneficial elements. The properties of HEAs are considered to be closely related to the properties of their constituent elements. However, these alloys also rely on the interactions between compositional elements [12], which can play a critical role in the overall properties. Chen et al. studied the CuNiAlCoCrFeSi HEA and compared it to 304 stainless steel, which has shown that the HEA is more noble than what would be predicted from the individual elements [13]. High-entropy alloys have unique corrosion properties making them of interest for use in coatings and as possible replacements for current commercial corrosion resistant alloys.
5
2.3 OVERVIEW
This chapter will focus on the current state of knowledge of HEA production and alloying which affects the corrosion properties of HEAs in aqueous environments. The effects of both composition and microstructure will be critically analyzed. The corrosion resistance of HEAs in both chloride- and sulfide-rich solutions will be discussed in detail, as well as the effects of alloying of well-known beneficial elements and their effects on HEA resistance to all forms of corrosion. The current methods being used in the field to gather critical electrochemical data on the corrosion of HEAs in accelerated testing conditions will also be analyzed.
2.4 LOCALIZED CORROSION OF HEAs 2.4.1 Pitting Corrosion
Passive films, their breakdown, and subsequent contribution to pit formation have been studied in detail [14, 15]. Similarly, pitting corrosion of alloys, caused by a break in the passive film, has also been studied extensively for stainless steels and other corrosion-resistant alloys.
One of the major disadvantages of passive materials is their susceptibility to localized corrosion.
In pitting corrosion, the passive film breaks resulting in a surface that is susceptible to pitting.
The nature of HEA pitting is still relatively unexplored. The general methods used to study the phenomena are largely electrochemical tests consisting of potentiodynamic polarization as well as critical pitting temperature testing. Potentiodynamic polarization allows for the determination of an alloy’s pitting potential [16]. The pitting potential can be described as the potential at which a stable pit begins to grow. The nobility of this potential, along with repassivation potentials that can be obtained from cyclic polarization experiments, have been used to describe the susceptibility of alloys to pitting corrosion. The nobility of this pitting
6 potential represents the alloy’s resistance to pitting. The nobler the pitting potential, the less likely the alloy is to undergo pitting under a specific electrolyte [15].
Pitting is known to be associated with metallographic features such as second phases particles, inclusions, and defects [17]. The resulting segregation caused by these features, decreases an alloy’s resistance to pitting corrosion, due to their greater susceptibility to pit initiation [18].
Temperature has been studied as a critical factor in pitting corrosion of alloys [16]. There is often a critical temperature below which an alloy will not pit, and above which pitting is observed. At temperatures below the critical pitting temperature (CPT) an alloy will undergo transpassive dissolution at high potentials, while above the CPT the alloy will be susceptible to pitting at potentials below the transpassive potential. The CPT is commonly used to describe an alloy’s resistance to pitting corrosion—the higher the CPT, the greater the resistance to pitting corrosion [15, 19].
2.5 CURRENT ALLOY PROCESSING AND GENERATION Production and processing of HEAs has largely been focused on the ability of HEAs to present desirable properties, such as making conventional alloys lighter and stronger. High- entropy alloys have been studied largely for their enhanced properties and capability to consist of large compositional fractions of desired elements, which allow these materials to be extremely tunable [2, 20], thereby generating novel alloys with improved properties unattainable by conventional alloying and processing methods.
7
2.5.1 Processing of single phase alloys
Alloying elements play a significant role in the properties and microstructures of HEAs.
Not much is known in the way of designing and processing these alloys. Therefore most alloys are either left unprocessed in the as-cast condition or held at high temperatures for long periods of time in order to generate homogenized and heterogeneous phase alloys preferable for study.
Li et al. [21] studied the effect of alloying elements on the microstructure of the
FeNiCrCu composition by adding alloying elements Co,Al,Mo,Mn, and Zr in equiatomic mole fractions to the base composition. This study shows that these alloying additions can greatly change the microstructure and properties of the HEA alloy. The HEAs were compared using optical microscopy, scanning electron microscopy, and x-ray diffractometry techniques. Results show that when alloy composition is comprised of elements near the IIIV family group of the periodic table (i.e. FeNiCrCuCo and FeNiCrCuMn), a single phase FCC structure is observed.
When Zr is added in the case of FeNiCrCuZr the alloy is a BCC solid solution with intermetallics present. It was found that the typical nature of an HEA microstructure is a combination of dendrites and interdendric regions throughout the crystal structure. There are complex interactions present in HEA alloy systems that create a diverse number of microstructures, which can be highly tuned to produce ideal microstructures for a variety of applications.
The alloying of HEAs with Cu and Al is a primary concern when looking at the corrosion resistance of HEA. The effects of Cu alloying have been studied by Hsu et al. in the
FeCoNiCrCux high-entropy alloy system in 3.5% NaCl solution [22]. Evaluation of corrosion properties was done by both immersion and potentiodynamic polarization tests. The microstructural analysis done with SEM/EDX shows segregated interdendric regions with high
8 amounts of Cu. The copper segregation into these retions leads to major corrosion of the interdendrites due to galvanic coupling with the dendritic regions. The alloying of Cu and subsequent segregation leads to poor corrosion resistance of these copper rich HEA compositions.
Al has been seen to have similar negative effects on the microstructure and phase composition of HEAs. Increased alloying of Al has been studied by Kao et al. in AlxCoCrFeNi
HEA to change the microstructure from a purely FCC microstructure at low Al contents between
0 and 0.25 atomic fraction, while possessing both FCC and BCC phases when the atomic fraction of Al is greater (0.5 and 1) [23]. The change in microstructure corresponds to sharp decline in corrosion resistance seen in various chloride-containing sulfuric acid solutions at 25ºC.
A strong decrease in breakdown potential as well as an increase in passive current density are seen when the microstructure changes from FCC to FCC+BCC [23].
The segregation of many different elements has been seen to have detrimental effects on the corrosion resistance of HEA alloys. Currently the HEAs that are most resilient to the effects of corrosion have been single phase alloys. Single phase alloys have a number of beneficial properties that make them of primary interest when generating a corrosion resistant HEA [12,
17]. Corrosion resistance of single phase HEAs has been largely attributed to their inclusion free, uniform solid solution nature. Subsequent work has been done to predict and generate single- phase HEAs by thermodynamic calculations, and phase predictions.
9
2.6 CORROSION BEHAVIOR OF HEAs 2.6.1 Corrosion resistance of HEAs in chloride solutions
The corrosion resistance of HEAs in chloride solutions has been studied to better understand the resistance of HEAs compared to conventional corrosion resistant alloys of similar compositions. The Co1.5CrFeNi1.5Ti0.5Mo0.1 HEA has been studied by measuring both pitting potential (Epit) and critical pitting temperature (CPT) in 1M NaCl, and 1M NaCl + (0.1,1M)
NaNO3 chloride solutions [24]. The Co1.5CrFeNi1.5Ti0.5Mo0.1 alloy exhibits an FCC solid solution structure [23]. The effect of Cl- concentration was determined by simultaneously testing
Epit and CPT using potentiodynamic polarization tests at temperatures ranging from 25 to 80ºC.
Through the range of temperatures in deaerated 0.001 and 0.01M NaCl solutions the alloy shows a very high transpassive potential, which varies very little with temperature. The alloy CPT was tested using a potentiodynamic technique in solutions of 0.1, 0.5, and 1 M NaCl with temperatures ranging from 25 to 80ºC. The potentiodynamic curves can be used to monitor CPT by observing when there is a change from transpassive dissolution, which is denoted by a gradual increase in the current, while pitting corrosion is seen by a steep decrease in breakdown potential, occurring within the conventional passive potential range. The CPT was found to depress by increases in [Cl-]; the CPT of Co1.5CrFeNi1.5Ti0.5Mo0.1 was found to be 70, 60, and 60ºC for solutions 0.1, 0.5, and 1 M NaCl respectively. The influence of [Cl-] on corrosion of HEAs has been shown to be very complex. Chou et al. reported that Leckie et al., Wang et al, and Meguid et al. have seen that pitting potential was proportional to the logarithm of the chloride concentrations at a constant temperature [24]. The result seems to be in agreement that increases in both temperature and [Cl-] decrease the pitting resistance of the alloy [24]. Newman
10 has attributed this effect to the increased porosity of the passive film at high temperatures allowing for the formation of chloro-complexes of chromium [25].
Chen et al. compared CuNiAlCoCrFeSi HEA and compositionally similar 304SS with near equal Cr content in deaerated 1 M NaCl solution at room temperature [13]. The HEA was shown to be more noble and have a lower corrosion current density than 304SS in NaCl. The
HEA can be described as having greater resistance to uniform corrosion, due to the nearly amorphous HEA structure, which contains little to no grain boundaries, allowing for greater congruence of the passive film. The passive region of the HEA is smaller than that of 304SS and when the HEA passive film is ruptured the alloy is more susceptible to nucleation and growth than 304SS.
2.6.2 Corrosion Resistance of HEAs in sulfide solutions
- The corrosion resistance of HEAs in sulfide [SO4 ] solutions was studied in an identical manner to [Cl-]. Potentiodynamic polarization measurements of Co1.5CrFeNi1.5Ti0.5Mo0.1 were made in solutions of 1 M NaCl containing 0.1, 0.25, 0.5, 0.75, and 1 M Na2SO4 in temperatures ranging from 25 to 80ºC. All potentiodynamic curves show an active-passive corrosion behavior over the entire temperature range [13]. The pitting potential increased with
2- 2- the increase of [SO4 ]. The increased of [SO4 ] had a positive effect on the pitting resistance of the alloy in the solution, with the pitting potential and the CPT of the alloy being higher in the 1
2- M NaCl + 1 M Na2SO4 solution. The addition of SO4 ions has shown to be an inhibitor of
2- pitting corrosion. The difference between Ecorr and Epit increases with increasing SO4
2- - concentration, therefore it can be determined that SO4 in Cl solutions act as a pitting inhibitor.
Chen et al. tested CuNiAlCoCrFeSi HEA in H2SO4 and compared potentiodynamic polarization curves to 304SS [13]. In deaerated H2SO4 solutions at room temperature, the HEA
11 was shown to have a greater resistance to general corrosion than 304SS. However, the range of potentials in which the HEA remains passive is much smaller than 304SS. Pitting corrosion is not observed in either the 304SS or HEA in 1 N H2SO4 [13]. Therefore, it can be seen that pitting is either decreased or more often absent after accelerated testing of HEAs in sulfate containing solutions. However, the corrosion resistance of an HEA alloy system in sulfate is dependent on the alloying composition. It has been seen that these HEA alloys have improved corrosion properties compared to the commercial alloys. HEA properties have shown to be advantageous when comparing the total alloy to the sum of its parts from an electrochemical viewpoint.
2.7 EFFECTS OF ALLOYING ON HEA CORROSION RESISTANCE 2.7.1 Effects of Cr content on HEAs
Chromium has conventionally been relied on in the production of stainless steels and other corrosion resistant alloys such as nickel-based alloys to induce formation of a chromium- oxide rich passive film. The amount of Cr alloyed in a steel is critical in ensuring a stainless quality and behavior. Newman et al. prepared high purity Fe-Cr polycrystals from which 7.3, 9.4, and 10.9 wt.% plate electrodes were prepared [25]. Anodic polarization was performed in deaerated 1 M H2SO4 at room temperature. Incomplete passivity was confirmed in the 10.9 wt.%
Fe-Cr samples, providing sufficient evidence that a Cr concentration between 12-13% is required to be considered a passive stainless steel. Lee et al. studied AlXCrFe1.5MnNi0.5 with varying Al content by potentiodynamic polarization in 0.5 M H2SO4 solutions. All alloy compositions show an active-passive corrosion nature with high resistance to general corrosion at high potentials.
The Al free alloy CrFe1.5MnNi0.5 demonstrated the lowest passive current density [26].
12
Potentiodynamic polarization in 1M NaCl at room temperature shows spontaneous passivity with higher Epit associated with lower Al concentrations. These alloys and most HEA alloys where chromium is a primary alloying element have chromium compositions greater than 13wt.%, which can be associated with the enhanced corrosion resistant properties of most HEAs.
However, there is still a large knowledge gap in the area of the oxide film that forms on the surface of these HEAs [26]. It is not necessarily true that a HEA high in chromium will have a chromium-oxide surface film or if other oxides or un-oxidized element will be present in the oxide layer of any particular HEA.
2.7.2 Effects of Mo alloying
Mo has been conventionally used in corrosion resistant alloys as it is known to assist in the repassivation of pits in acid and chloride environments [27]. However in many HEAs Mo and
Cr segregate forming a sigma phase rich in both elements [28]. The strong metal-metal bond strength between Mo and Cr that causes this segregation can be detrimental to the overall corrosion resistance of such Mo-containing HEAs. The effect of Mo concentration has been studied by Chou et al. in a Co1.5CrFeNi1.5Ti0.5Mox alloy series by potentiodynamic polarization testing in 0.5 M H2SO4, 1 M NaOH, 1 M NaCl. The Mo composition was varied from 0 to 19.96 wt.%. In deaerated 0.5 M H2SO4 all alloys presented similar corrosion resistance
[28], while in 1M NaOH the Mo containing alloys showed a decreased resistance to corrosion in both an increase icorr and a decrease in breakdown potential. However, the Mo-containing alloys showed improved resistance to pitting in 1M NaCl. The Mo-containing alloys showed a much wider passive region as well as Epit between 1.1-1.2 V SHE, far more noble than Epit, 0.33 V
SHE, of the Mo-free composition. Molybdenum has been shown to be a beneficial alloying addition, which increases an alloy’s resistance to pitting in Cl- rich solutions.
13
2.7.3 Effects of Al, Co, and Mn
The effects of Al alloying was studying by both Lee et al. [26] and Kao et al. [23] using
AlxCoCrFeNi and AlxCrFe1.5MnNi0.5 alloy series, respectively. Both alloy systems were tested potentiodynamically in 0.5 M H2SO4, giving a basis for comparison when substituting Co for
Mn. The alloying of Al in both cases has been seen to have little to no effect on the corrosion response in sulfuric acid solutions in the Co-containing alloy. However, the increase in Al content has a deleterious effect at higher temperatures, which can be seen in an increase in passive current density. The Mn-containing alloy had increased passive current density with the increase of Al. Both compositions show a negative effect on corrosion resistance, which corresponds to the addition of Al. The replacement of Co with Mn in the Al-free CoCrFeNi alloy weakens the corrosion resistance of the alloy in the corresponding 0.5 M H2SO4 solution. The resistance of the MnCrFeNi alloy is lower than common 304SS. The effect of Al alloying is closely related to the formation of a multiphased FCC-BCC alloy, which corresponds to a reduction in corrosion resistance of the alloys.
2.7.4 Effects of Ni alloying
In the stainless steel industry, nickel has conventionally been used as an austenite phase stabilizer. Ni-Mo and Ni-Cr alloy systems have shown to be highly resistant to corrosion in acidic solutions, both sulfuric and hydrochloric acids. The alloying effect of Ni on the corrosion resistance of HEAs has been studied by Qiu et al., by potentiodynamic polarization testing of the
Al2CrFeCoCuTiNix series, x ranging as 0, 0.5, 1, 1.5 and 2, in 3.5wt. % NaCl and 1 M NaOH.
The alloy of greatest corrosion resistance was seen to be a molar ratio of 1, suggesting that there is a critical content of Ni at which point if the Ni content is increased there are detrimental impacts on the corrosion resistance [29]. The microstructures of alloys containing a greater
14 concentration of Ni were seen to form a detrimental Al, Ni-rich B2 phase, which may negatively affect the corrosion resistance of the alloy studied. The Ni alloy composition and its effects on corrosion resistance are still relatively unknown.
15
References
[1] Cost of Corrosion Study, NACE, 2014. [2] J.W. Yeh, S.K. Chen, S.J. Lin, J.Y. Gan, T.S. Chin, T.T. Shun, C.H. Tsau, S.Y. Chang, Adv Eng Mater, 6 (2004) 299-303. [3] M.-H. Tsai, J.-W. Yeh, Materials Research Letters, 2 (2014) 107-123. [4] Shi, Y., Yang, B., & Liaw, P. K. (2017). Corrosion-resistant high-entropy alloys: A review. Metals, 7(2), 43. [5] Zhang, Y.; Zuo, T.T.; Tang, Z.; Gao, M.C.; Dahmen, K.A.; Liaw, P.K.; Lu, Z.P. Microstructures and properties of high-entropy alloys. Prog. Mater. Sci. 2014, 61, 1–93. [6] Gao, M.C. Chapter 11: Design of high-entropy alloys. In High-Entropy Alloys: Fundamentals and Applications; Gao, M.C., Yeh, J.W., Liaw, P.K., Zhang, Y., Eds.; Springer: Cham, Switzerland, 2016. [7] R.F.A. Jargelius-Pettersson, Application of the Pitting Resistance Equivalent Concept to Some Highly Alloyed Austenitic Stainless Steels, CORROSION, 54 (1998) 162-168. [8] K. Lorenz, G. Medawar, Über das Korrosionsverhalten austenitischer Chrom-Nickel- (Molybdän-) Stähle mit und ohne Stickstoffzusatz unter besonderer Berücksichtigung ihrer Beanspruchbarkeit in chloridhaltigen Lösungen, Thyssenforschung, 1 (1969) 97-108. [9] J.H. Cleland, What does the pitting resistance equivalent really tell us?, Engineering Failure Analysis, 3 (1996) 65-69. [10] G. Herbsleb, Der Einfluß von Schwefeldioxid, Schwefelwasserstoff und Kohlenmonoxid auf die Lochkorrosion von austenitischen Chrom-Nickel-Stählen mit bis zu 4 Massen-% Molybdän in 1 M Natriumchlorid-Lösung, Materials and Corrosion, 33 (1982) 334-340. [11] H. Okamoto, The effect of tungsten and molybdenum on the performance of super duplex stainless steels, in: Proc. Conf. Application of Stainless Steels ‘92, 1992, pp. 360-369. [12] Tsai, M., & Yeh, J. (2014). High-entropy alloys: A critical review. Materials Research Letters, 2(3), 107-123. [13] Chen, Y., Duval, T., Hung, U., Yeh, J., & Shih, H. (2005). Microstructure and electrochemical properties of high entropy alloys—a comparison with type-304 stainless steel. Corrosion Science, 47(9), 2257-2279. [14] Frankel, G. (1998). Pitting corrosion of metals a review of the critical factors. Journal of the Electrochemical Society, 145(6), 2186-2198. [15] Schmuki, P. (2002). From bacon to barriers: A review on the passivity of metals and alloys. Journal of Solid State Electrochemistry, 6(3), 145-164. [16] G.S. Frankel, Pitting corrosion of metals - A review of the critical factors, Journal of the Electrochemical Society, 145 (1998) 2186-2198. [17] Jones, D. A. (1996). Principles and prevention of corrosion prentice hall. Saddle River, NJ [18] Park, C., Rao, V. S., & Kwon, H. (2005). Effects of sigma phase on the initiation and propagation of pitting corrosion of duplex stainless steel. Corrosion, 61(1), 76-83. [19] Arnvig, P.E., & Bisgard, A.D. (1996). Determining the potential independent critical pitting temperature (CPT) by a potentiostatic method using the Avesta Cell. United States: NACE International. [20] Zhang, Y., Zuo, T. T., Tang, Z., Gao, M. C., Dahmen, K. A., Liaw, P. K., & Lu, Z. P. (2014). Microstructures and properties of high-entropy alloys [21] Li, C., Li, J. C., Zhao, M., & Jiang, Q. (2009). Effect of alloying elements on microstructure and properties of multiprincipal elements high-entropy alloys
16
[22] Hsu, Y., Chiang, W., & Wu, J. (2005). Corrosion behavior of FeCoNiCrCux high-entropy alloys in 3.5% sodium chloride solution [23] Kao, Y., Lee, T., Chen, S., & Chang, Y. (2010). Electrochemical passive properties of AlxCoCrFeNi (x=0, 0.25, 0.50, 1.00) alloys in sulfuric acids [24] Chou, Y. L., Wang, Y. C., Yeh, J. W., & Shih, H. C. (2010). Pitting corrosion of the high- entropy alloy Co1.5CrFeNi1.5Ti0.5Mo0.1 in chloride-containing sulphate solutions [25] Newman, R. C., Foong Tuck Meng, , & Sieradzki, K. (1988). Validation of a percolation model for passivation of Fe-Cr alloys: I current efficiency in the incompletely passivated state [26] Lee, C. P., Chang, C. C., Chen, Y. Y., Yeh, J. W., & Shih, H. C. (2008). Effect of the aluminium content of AlxCrFe1.5MnNi0.5 high-entropy alloys on the corrosion behaviour in aqueous environments [27] P. Crook, Corrosion of Nickel and Nickel•Base Alloys, ASM International, Materials Park, OH (2005). [28] Chou, Y. L., Yeh, J. W., & Shih, H. C. (2010). The effect of molybdenum on the corrosion behaviour of the high-entropy alloys Co1.5CrFeNi1.5Ti0.5Mox in aqueous environments [29] Qiu, X., & Liu, C. (2013). Microstructure and properties of Al2CrFeCoCuTiNix high- entropy alloys prepared by laser cladding
17
Chapter 3
CORROSION BEHAVIOR OF Ni38Cr21Fe20Mo6W2Ru13 HIGH ENTROPY ALLOY IN CHLORIDE ENVIRONMENTS
3.1 INTRODUCTION Corrosion is the deterioration or degradation of a metal as the result of a chemical or electrochemical reaction between it and the surrounding environment. As a reverse process of metallurgy, metal corrosion is a spontaneous process that inevitably causes great economic loss across the globe. The global cost of corrosion is estimated to be $2.5 trillion annually, which is equivalent to roughly 3.4 percent of the global Gross Domestic Product (2013) [1]. Moreover, corrosion is not only considered a financial issue, but also one of health and safety. Corroded metal structures can and do cause injury and death. Therefore, development of new corrosion resistant alloys (CRAs) that possess superior intrinsic corrosion resistance, especially to localized corrosion in harsh environment, is of enormous economic and safety benefits.
High entropy alloys (HEAs) are a newly developed family of multicomponent alloys [2] that are generally defined as alloys that consist of five or more principal elements with very high configuration entropy. By breaking the conventional alloy design limitation based on one single majority host element, HEAs have exhibited exceptional properties, such as combined high strength and ductility [3, 4], improved fatigue resistance [5, 6], high fracture toughness [7, 8] and high thermal stability [9]. As one of the key properties for practical application, the corrosion performance of HEAs has been a focus of investigation during the last decade. In one of the earliest studies, Chen et al. reported that, relative to SS304, the general corrosion resistance of a
Cu0.5NiAlCoCrFeSi HEA was superior, but the pitting resistance of the HEA was inferior [10].
18
In contrast, Co1.5CrFeNi1.5Ti0.5Mo0.1 HEA was found to exhibit a very high pitting resistance with a critical pitting temperature (CPT) ranging from 60 ~ 70oC in chloride solution, which is much higher than that of the commercial SS304 and SS316[11]. It can be seen that the corrosion performance of the different HEAs is not the same.
To understand these phenomena, the influence of heterogeneity in microstructure and beneficial elements on the corrosion performance of HEAs has been investigated. In the as-cast
AlCoCrFeNi HEA, it was reported that dendritic regions suffer from selective corrosion in 0.6 M
NaCl solution due to the segregation of Cr in the interdendritic region and small precipitates in the dendrites [12]. In the FeCoNiCrCux (x=0, 0.5 and 1) system, it was found that the increase of
Cu content increases the tendency of localized corrosion, which was attributed to the microgalvanic corrosion caused by segregation of Cu in interdendritic regions [13]. In the
AlxCoCrFeNi (x=0.3, 0.5 and 0.7) system, Shi et al. reported a decreased pitting resistance with the increase of Al content, which was attributed to the increased volume fraction of the (Al, Ni)- rich and Cr-depleted BCC phase [14]. Similar results have also been reported by Kao et al. [15] and Lee et al. [16]. In the Al0.5CoCrCuFeNiBx (x=0, 0.2, 0.6 and 1) system, it was found that the corrosion resistance of the HEA in 1 M H2SO4 decreases with the increase of boron content due to the formation of Cr, Fe and Co borides [17]. The effect of beneficial elements, such as Mo and V, on the pitting resistance of the HEAs has also been reported. In the
Co1.5CrFeNi1.5Ti0.5Mox (x=0, 0.1, 0.5 and 0.8) system, Chou et al. reported that the Mo- containing HEAs are less susceptible to pitting corrosion in 1 M NaCl solution compared with the Mo-free HEA as indicated by cyclic potentiodynamic polarization tests in 1 M NaCl [18]. In
TiZr0.5NbCr0.5VxMoy HEAs, it was found that the addition of V and Mo increases the pitting resistance of the HEAs in 3.5 wt.% NaCl solution [19]. Based on the previous studies, any
19 heterogeneities in the microstructure, such as inclusions, second phases, segregations, will decrease the corrosion performance of the HEAs, especially the resistance to localized corrosion, either though galvanic attack or pitting corrosion. On the other hand, the commonly believed beneficial elements also increase the pitting resistance of HEAs. In other words, although the
HEAs have exhibited exceptional mechanical and thermal properties, their corrosion performance still follows the fundamental understanding of corrosion developed over years of experience. This allows us to design corrosion resistant HEAs (CR-HEAs) in the framework of the empirical paradigms and intuition based on prior knowledge.
Microstructure and composition are the two key factors to the corrosion performance of an alloy. Commonly, the presence of heterogeneities in microstructure decreases the corrosion resistance. For pitting corrosion, the heterogeneities, such as inclusions or impurities, can promote the pitting evolution by accelerating the pitting initiation process [20]. Therefore, a single phase HEA should be an ideal candidate for high corrosion resistance. However, most
HEAs reported in the literature are multi-phase, which usually decreases their corrosion resistance, either through galvanic attack or pitting corrosion. In addition to the microstructure, the chemical composition is another significant determinant of corrosion resistance. On the one hand, the chemical composition of an alloy determines the composition of passive film, which thereby influences the initiation process of pitting corrosion. On the other hand, it also influences the dissolution rate of the metal inside the pit, which determines the propagation process of a pit.
The influence of chemical composition on pitting corrosion of Fe-Cr-Ni alloys can be described by the pitting resistance equivalent number (PREN), which is empirically fit of corrosion parameters, such as pitting potential and critical pitting temperature, to the composition. Various
20 equations for PREN have been developed during the last few decades [21-25]. One commonly used PREN equation is [25]:
푃푅퐸푁 = 푤푡. %퐶푟 + 1.6푤푡. %푊 + 3.3푤푡. %푀표 + 16푤푡. %푁 퐸푞. (1)
Generally, an alloy with a higher PREN value has a higher pitting resistance. Based on the above two notions, one can expect that a single phase and a high PREN value are two key factors that should be taken into consideration when designing a CR-HEA. However, PREN equations, such as the one in Eq.1, are only valid for the Fe-Cr-Ni system.
HEAs reported previously are equiatomic or near equiatomic composition. Although the configuration entropy has the maximum value, the corrosion resistance is commonly not optimized because of the presence of the multi-phases in the alloy. Therefore, the non- equiatomic composition space should be explored to identify the most corrosion-resistant single- phase HEAs. However, the conventional approach to finding such alloys, i.e. trial and error, is usually costly and time-consuming and is not suitable for designing and synthesizing CR-HEAs in the vast compositional space. Integrated computational materials engineering (ICME) is a modern systems-based approach to design materials that meet a specific need for performance by linking computational materials models across multiple length scales [26]. The alloy discussed in this paper was designed using conventional intuition of corrosion resistant alloy design as well
Calculation of Phase Diagrams (CALPHAD) modeling. In this paper, this new method of corrosion resistant alloy design for HEAs and the corresponded alloy produced are evaluated.
21
3.2 METHODS The high-entropy alloy, NiCrFeMoWRu, which will be referred to as HEA1, was prepared by vacuum arc melting of the constituent elements in a water cooled copper crucible.
The Ni, Mo, and W starting materials were in the form of small diameter wire with purity of
99%, 99.95%, and 99.99%, respectively, while Fe, Cr, and Ru were in the form of chips with purities of 99.98%, 99.99%, 99.95%, respectively. The constituent elements, adding up to 10-12 g per button, were arranged in order of lowest melting point to highest melting point from the bottom of the crucible to the top. The arc melt chamber was brought under vacuum after being purged with argon (Ar) three times, and the combination of elements was melted under a pressure of 0.5 atm. Flipping and remelting of the alloy button was repeated at least 5 times to improve the homogeneity of the alloy. The alloy was prepared in the form of buttons of ~ 1 cm thick and ~2 cm diameter. The buttons had lustrous surfaces indicating that little oxidation occurred during the arc melting process. A second button of identical composition was cast via arc melting and encapsulated under vacuum and backfilled to 0.125 atm Ar gas. Fig. 3.1 shows the CALPHAD generated phase diagram of HEA1 with a large FCC region. The encapsulated button was heat treated at 1250°C for 120 h in the large single FCC phase field and water quenched to fully homogenize the alloy. Table 1 shows the chemical composition of the test material used in the work.
3.2.1 Alloy Characterization
The grain size and structure of NiCrFeMoWRu was characterized using optical microscopy (OM) after both as-cast and homogenized samples were mechanically ground with
SiC paper to 1200 grit, cleaned by ethanol, and dried with compressed air. The OM was performed under both polarized and non-polarized light. Further microstructure characterization
22 was performed with a Quanta 200 scanning electron microscope (SEM) and a Rigaku SmartLab type X-ray diffraction (XRD). The semi-quantitative analysis of the alloy’s chemical composition was conducted using an SEM equipped with energy dispersive spectrometry (EDS).
3.2.2 Electrochemical Corrosion Characterization
The corrosion resistance of the alloy was evaluated by immersion tests, potentiodynamic polarization, and critical pitting temperature testing. Before immersion tests, the samples were cleaned in acetone and dried with condensed air. The weight of each specimen was measured before and after immersion with a precision balance. The specimens were immersed in 12 M HCl for 24 h. Subsequently, the specimens were ultrasonically washed to remove any loose corrosion product. The corrosion morphology of the specimens was observed using OM.
Prior to electrochemical testing, each button was cut into 2 mm thick sections, and each section was ground with SiC paper to 1200 grit, and immediately prepared for electrochemical testing. The samples were connected to a lead wire, wrapped with polytetrafluoroethylene
(PTFE) tape and coated with black wax leaving an exposed working area ~ 0.2 cm2. The test solutions used in this work include 0.6 M NaCl as well as 1, 6, and 12 M HCl, prepared from reagent grade NaCl and HCl, and dissolved in DI water. Potentiodynamic experiments were conducted under the thermostatic condition of 30°C. The electrochemical cell used in this work was a 250 ml jacketed beaker and about 200 ml of solution. All solutions were left in the aerated condition. The electrochemical polarization experiments were performed with a conventional three-electrode cell, consisting of a working electrode, a platinum counter electrode, and a saturated calomel electrode (SCE) as the reference electrode.
23
The potential was controlled and the current was measured using a Gamry Reference 600 potentiostat with a computer-controlled electrochemical interface, allowing for monitoring of the total current (I), potential (V), and time (t). All experiments were conducted at a scan rate of 20 mV/min from the initial potential of -0.25 V versus the open circuit potential (OCP) and terminated when the current reached 1 mA. Prior to the experiment, the OCP was recorded for 10 min to acquire a steady state potential. Characteristic electrochemical parameters, such as corrosion potential (Ecorr), corrosion current density (icorr), primary passive potential (Epp), passive current density (icrit), breakdown (Eb) or transpassive potential (Etrans), and repassivation potential (ERP) can be extracted from the evaluation of polarization curves. An average corrosion current density was found by dividing corrosion current I by the working area of the electrode.
Eb or Etrans was determined by observing the potential at which a continuous increase of the anodic current initiated, indicating the sustained breakdown of the passive film. Following the polarization experiment, the specimens were washed and then dried with compressed air. The morphology of the corroded surface was then immediately examined by OM. To determine reproducibility, tests were repeated three times under each condition.
The open circuit corrosion potential (OCP) and linear polarization resistance (LPR) were monitored for 24 h in 12 M HCl. OCP was recorded every second over a 24 h period. LPR tests were performed by the scanning the working electrode potential ±0.02V(SCE) from cathodic potentials to the anodic direction around the OCP. The scan rate was 0.125 mV s-1. One polarization resistance measurement was made every 1800 s for a total of 35 measurements. The polarization resistance (Rp) was determined from the slope of the potential vs. current plots.
Potentiostatic critical pitting temperature (CPT) measurements were performed using a jacketed standard three-electrode cell connected to a programmable temperature-controlled
24 circulator. A temperature probe was placed near the sample to monitor the solution temperature.
The testing solution for each measurement was about 300 mL of 3.5wt.% NaCl solution that was exposed to air. The electrochemical experiments were carried out with a Gamry Reference 600 potentiostat. The samples were immersed in the test solution at the initial temperature of 10.3 oC.
After 10 min of stabilization, the test solution was heated by the water circulator at a rate of 1 oC min-1. The actual heating rate of the testing solution at the initial stage was non-linear and smaller than 1 oC min-1, but it increased gradually and was stabilized at 1oC min-1 after a certain period. Prior to starting the temperature scan, the sample was anodically polarized to 700 mV
SCE. This potential was held constant during the whole temperature scan. The current was collected at a rate of 1 Hz and the CPT was defined as the temperature at which the current increased rapidly and continuously exceeded 100 µA for 60 s.
3.3 RESULTS 3.3.1 Microstructure of Ni38Cr21Fe20Mo6W2Ru13
Optical microscopy was used to approximate the grain size of the HEA before and after heat treatment. Fig. 3.2 shows that the as-cast sample exhibited a typical dendritic structure with dendritic regions (DR) being segregated by connected interdendritic regions (IR). This segregation is typical of arc melted samples, where some elements solidify before other elements, leading to a cored microstructure with compositional segregation. Fig. 3.3 shows the
HEA after heat treatment. The grains were uniform, much larger than 1mm in size, and free of dendritic structure. Without etching, the as-cast and homogenized samples show segregation and grains boundaries respectively. The contrast in the as-cast structure can be due to a significant difference in hardness between the Ru rich dendrites and the Ru depleted interdendrites, which caused topological variations during polishing.
25
Fig. 3.4 shows the secondary electron SEM image of an as-cast sample. The as-cast
DR/IR microstructure was much clearer, showing the clear segregation of alloying elements.
Elemental segregation of the alloy is observed further by SEM/EDS, shown in Fig. 3.5, where the dendrite interiors are seen to be enriched in Ru, while the inter-dendritic regions are enriched in Mo and Cr. Fig. 3.6 shows secondary electron SEM images of the heat treated/homogenized sample. These are single-phase and free of inclusions. The EDS mapping in Fig. 3.7 indicates the absence of any secondary phases as well as a uniform chemical composition. Fig. 3.8 shows the
XRD result from the as-cast HEA. Analysis of the spectra using conventional peak indexing rules indicated that the as-cast sample had a single-phase FCC crystal structure despite the coring and elemental segregation between DR and IR. This is evidence of the wide range in solid solubility of the FCC phase at high temperatures and slow diffusion kinetics at lower temperatures as the sample solidified and cooled.
3.3.2 Corrosion behavior of Ni38Cr21Fe20Mo6W2Ru13 as cast and homogenized alloy,
Immersion
The corrosion rate of as-cast and homogenized Ni38Cr21Fe20Mo6W2Ru13 in 12 M HCl solution for the 24-h immersion tests at room temperature ~25°C can be calculated by Equation
(2) [27]:
8.76×104×푊 Corrosion rate(mm/yr) = 퐸푞. (2) 푡 ×퐴 ×퐷 where W is specimen weight loss (in g) after the time t (in h) exposed to the electrolyte, A (in cm2) is the exposed area, and D (in g/cm3) is the alloy’s density. The density of HEA1 was calculated based on elemental densities to be 9.81 g/cm3. The mass loss data resulted in measured corrosion rates for the as-cast and homogenized of 0.714 mm/yr and 0.193 mm/yr,
26 respectively. Both as-cast and homogenized HEA1 displayed remarkably low corrosion rates in an extremely aggressive solution. The homogenized HEA1 displays better general corrosion resistance compared to the as-cast HEA1.
Fig. 3.9 shows typical surface images of both the corroded as-cast and homogenized
HEA after immersion. The corroded surface of the as-cast HEA1 displays a system of dendrites as well as connected interdendritic regions. In contrast, the homogenized HEA1 specimen displays a single uniform surface with exposed scratches that remain after the grinding process.
Therefore, the major corrosion types of the as-cast HEA1 are general corrosion along with interdendritic corrosion, while the major type of corrosion of the homogenized HEA1 is general corrosion.
3.3.3 Potentiodynamic polarization measurements and corrosion morphologies
The potentiodynamic polarization curves of as-cast and homogenized samples in 0.6 M
NaCl at 30°C, Fig. 3.10, show that both samples exhibit similar corrosion resistance, being spontaneously passive with little to no metastable events and very high breakdown potentials.
After polarization, the corrosion morphologies were examined with OM. Fig. 3.11 displays the optical image of corrosion morphology of the as-cast sample. The surface was etched, with no evidence of pitting morphology, indicating that the current increase at the breakdown potential reflected the onset of transpassive dissolution. Fig. 3.11(b) shows that inter- dendritic regions were preferentially dissolved. To assess the evolution of the dissolution process, the polarization tests were stopped at different potentials above breakdown and the surface was examined. A phenomenon of preferential confined transpassive dissolution is observed at the inter-dendritic regions occurring at potentials above the transpassive potential.
27
The corrosion morphology suggests a four-stage corrosion evolution process with increasing potential (Fig. 3.12): 1) Transpassive dissolution in IR. 2) Transpassive dissolution within DR.
3) Cracking of surface product on IR. 4) Peeling off of “transpassive film” from IR, leaving fresh surface exposed.
Fig. 3.13 shows optical images of the homogenized sample after polarization. The surface was uniformly etched, again with no indication of localized corrosion. The passive film appears to have uniformly dissolved during polarization at high potentials. The surface was free of any indication of pits or crevices that would suggest a localized breakdown of the passive film.
Fig. 3.14 displays the potentiodynamic polarization curves of both as-cast and homogenized samples in 1 M, 6 M, and 12 M HCl at 30⁰C. The curves in all three solutions show transpassive breakdown, which was also observed in 0.6 M NaCl. However, the homogenized sample in 12 M HCl shows a large reduction in the passive region and the as-cast sample exhibits nearly active dissolution. No pits were observed in any of these experiments.
3.3.4 Linear polarization resistance and 24h corrosion potential measurements in 12 M HCl
Fig. 3.15 shows the trend in the polarization resistance with time during exposure at the OCP in
12 M HCl. Rp decreased after 20,000 and 10,000 s for as-cast and homogenized samples, respectively. The plot contains a number of clear outlier points that do not match the trend of Rp; these points are likely caused by errors of the analysis software that was used to automatically calculate polarization resistance. The decrease in Rp cannot be explained by crevice corrosion because, given the severity of the environment, a crevice environment should not be different than the bulk environment, and no crevice was seen after testing. Other factors must therefore explain this decrease in Rp such as a change in the surface composition or depassivation by
28 dissolution of the oxide surface layer, or change in the solution environment. Fig. 3.16 displays the OCP of the HEA1 samples in 12 M HCl over the 24 h period. The OCPs for the as-cast and homogenized samples were approximately 145 mV SCE and 175 mV SCE, respectively. The stable OCP is representative of a stable steady state process, suggesting that there is no depassivation in 12 M HCl. A change in OCP corresponding to the decrease in Rp would be expected, but is not seen.
3.3.5 Critical Pitting Temperature
Fig. 3.17 and the magnified version in Fig. 3.18 show potentiostatic CPT curves for both as-cast and homogenized HEA1. The initial dramatic decrease in current indicates the formation of the passive film on both alloy surfaces. During the whole heating process to reach the temperature limit of the equipment, 85oC, no stable pitting breakdown was observed. The CPTs of both samples are thus higher than this temperature. Only very small metastable pitting transients (~0.2 uA) can be observed at the high temperatures, while no obvious metastable pitting transients were observed at low temperatures. The currents at low temperatures are compared with the commercial SS316L in Fig. 3.19, which shows strong metastable pitting activity. In contrast, the HEAs show no current noise, indicating that the passive films on both samples have great stability. It can be concluded that both as-cast and homogenize HEAs possess excellent pitting resistance.
3.4 DISCUSSION Pitting corrosion is influenced by two major aspects: nucleation of the pit, the local breakdown of the passive film, and the transition of a metastable pit into a stable pit. The parameters used to evaluate the pitting susceptibility of HEA were potentiodynamic polarization, critical pitting temperature, and polarization resistance and could all be affected by pit nucleation
29 and growth. In this section, both critical factors are considered in describing the resistance of
HEA1 to pitting corrosion.
3.4.1 Superior pitting resistance based on passive film breakdown
The studied HEA, Ni38Cr21Fe20Mo6W2-Ru13 demonstrates remarkable corrosion resistance. According to the recently proposed perspective on critical step of pitting corrosion
[28], such a superior corrosion resistance could be either attributed to the properties of the passive film or the resistance to formation of stable pits, or both. From the perspective of passive film breakdown as a critical step, the continuity, formability, and stability of the passive film are major contributing factors to corrosion resistance.
Passive film continuity is governed largely by phase composition of the alloy [28].
Inclusions, impurities and second phases are known to be detrimental to the continuity since the oxides formed at these sites are generally easily ruptured and less protective. The composition of
HEA1 is rich in oxide forming elements; Ni, Cr, Fe, and Ru are all capable of forming oxides in a variety of corrosive environments. To form a continuous oxide, the oxide-forming elements must be well distributed throughout the alloy. HEAs rely on the high-entropy effect, which lowers the tendency of elements to order and segregate, leading to the solid solution nature of
HEAs [2]. The solid solution and random positioning of elements throughout the solid-solution lattice enhances the continuity of the oxide layer. Both the as-cast and homogenized HEA samples show outstanding passive film continuity due to their lack of both inclusions and secondary phases. The corrosion resistance of this alloy can be seen in the polarization plots where only transpassive dissolution is displayed at potentials above 1V SCE for both 1 M and 6
M HCl solutions.
30
Passive film stability of HEA1 was evaluated by both polarization data and CPT testing.
The polarization curves of HEA shown in a 0.6 M NaCl as well as 6 M HCl shown in Fig. 3.10 and Fig. 3.14, respectively, show no occurrences of current transients that might be associated with metastable pitting. The passive film also shows great stability as it is polarized to high polarization potentials, remaining stable until +1 V SCE in both solutions when it then undergoes transpassive dissolution. Similarly, the CPT tests suggest a very stable passive film; the HEA is observed to be free of metastable events until high temperatures (60-80°C) where metastable events on the order of 0.1 µA are present for short lifespans. The absence of metastable events in both polarization and CPT measurements describes a very stable oxide that formed on the metal surface. Elements with high oxygen affinities are known to have similarly high metal-oxygen
(M-O) bond strength [30]. The strength of the HEA passive film in aggressive solutions suggests very high M-O bonding across the passive film.
3.4.2 Superior pitting resistance based on pit growth stability
Pit stability is determined by the competition between the diffusion of metal cations and dissolution metal at the pit surface [28]. In general, a pit is stabilized by a locally aggressive environment, generated by the active dissolution of metal from the pit surface. The pit will either become stable when the solution is sufficiently aggressive, or otherwise the surface will repassivate. When diffusion of the metal cations out of the pit into the bulk solution is fast enough to prevent this local aggressive environment, a stable pit is unable to form. Maintaining an environment suitable for stable pit formation is dependent on the balance between dissolution and diffusion of metal ions [28].
31
3.4.3 Critical dissolution current density
Considering a hemispherical pit with a pit depth and radius of r, the local environment inside the pit is influenced by two crucial diffusion current densities, imax, and ilim [28]. Here imax is the maximum dissolution current density at the pit bottom, and ilim is the diffusion-limited current density, which is achieved when localized solution is saturated with metal ions while pit growth is maintained. In order for an open pit to be stable, icrit ≤ imax. In other words the critical concentration of metal ions at the pit surface to maintain a stable pit must be achieved. When considering these factors, pit depth is crucial, because as pit depth increases the icrit needed to sustain pit growth decreases. The icrit is very large for small pits but can be very small when a pit has a very large depth. The low imax of HEA1 is unable to meet criteria for stable pitting, and pitting does not occur. Metastable pit size and lifetime can be observed in the collected CPT data, where a conventional CRA (316L SS) has very large and long-lasting metastable events.
These transients correlate to large metastable pits. In contrast, HEA1 shows only very small metastable events of ~0.2 µA. HEA1 has no inclusions that could promote deep pit nuclei.
3.4.4 Maximum dissolution current density
The maximum dissolution current density describes the rate of dissolution in a pit at a given potential in the absence of any ohmic potential drop [28]. A solution of 5 M HCl has been suggested to be an adequate simulation of the aggressive environment inside of a pit [31].
Therefore, the 6 M HCl test solution would simulate an environment more aggressive than any environment that is generated in a pit. HEA1 in this extreme environment demonstrates strong, spontaneous passivity, with a current density of 10-6 A/cm2 enduring to high potentials. Thus, this HEA demonstrates extreme resistance to pitting corrosion. In fact, no evidence of pitting was found at potentials to 1 V SCE and temperatures to 85°C. Analogous to metal-oxygen bond
32 strength in affecting passivity, metal-metal (M-M) bond strength is critical to the dissolution of the metal surface. Passivity promoters such as, Ti, Cr, and Ni have very high M-M bond strength. The high concentrations of passivity promoter elements in the HEA result in a very high M-M bond strength throughout the alloy, contributing to low rates of dissolution.
3.5 CONCLUSIONS
(1) A high corrosion resistant single-phased HEA was designed and successfully fabricated.
(2) HEA1 does not exhibit pitting even in the most aggressive pit condition.
(3) The high pitting resistance is derived from: (1) the continuous and protective passive film, which inhibits the pitting nucleation process; (2) the decreased pit growth stability caused by a low imax.
(4) The segregation in the as-cast sample does not decrease the pitting resistance, however, it accelerates the transpassive dissolution.
33
Figures
Table 3.1: Chemical Composition (wt.%) of material
Designation Composition (atomic ratio) Ni(58.69) (%) Cr(51.99) (%) Fe(55.84) (%) Mo(95.95) (%) W(183.84) (%) Ru(101.07) (%)
HEA1 Ni38-Cr21-Fe20-Mo6-W2-Ru13 33.3 16.3 16.7 8.6 5.5 19.6
Figure 3.1: CALPHAD generated pseudo-binary phase diagram, provided by J. Saal and P. Lu from QuesTek.
34
Figure 3.2: Image of as cast Ni38Cr21Fe20Mo6W2Ru13 HEA as cast HEA with dendritic regions (DR) and interdendritic regions (IR) labeled.
Figure 3.3: The large grain size of the HEA after 1250°C 120 h heat treatment.
35
Figure 3.4: Typical SEM micrograph of the As Cast Ni38Cr21Fe20Mo6W2Ru13 alloy, segregation and pores visible.
Figure 3.5: EDS mapping of as-cast sample, displaying typical dendritic microstructure.
36
Figure 3.6: Typical SEM micrograph of the heat treated Ni38Cr21Fe20Mo6W2Ru13 alloy with uniform single phase microstructure and no porosity.
Figure 3.7: EDS mapping results of homogenized Ni38Cr21Fe20Mo6W2Ru13 sample.
37
Figure 3.8: XRD pattern of as-cast HEA with peaks corresponding to typical FCC structure.
Figure 3.9: Optical images of HEA after 24-h immersion in 12 M HCl, ~25°C a) as-cast b) homogenized.
38
Figure 3.10: Potentiodynamic Polarization of HEA in 0.6 M NaCl, 30 oC.
Figure 3.11: Optical images of corrosion morphology of as-cast sample.
39
Figure 3.12: Stepped polarization corrosion morphology of as-cast HEA showing the preferential attack at the interdendritic regions.
Figure 3.13: Optical image of homogenized HEA1 corrosion morphology.
40
Figure 3.14: Typical polarization curves of Ni38Cr21Fe20Mo6W2Ru13 at 30°C in solution 1 M HCl, 6 M HCl, 12 M HCl. The curve in 0.6 M NaCl is included for comparison.
Figure 3.15: Polarization resistance over 24 h immersion in 12 M HCl at 30°C, as-cast and homogenized HEA.
41
Figure 3.16: 24 h OCP in 12 M HCl.
Figure 3.17: Potentiostatic CPT measurement at 700 mV in 3.5% NaCl on both as-cast and homogenized HEAs.
42
Figure 3.18: Magnified potentiostatic CPT curves.
Figure 3.19: Current transients compared to commercial SS316L.
43
References [1] G. Koch, J. Varney, N. Thompson, O. Moghissi, M. Gould, J. Payer, International Measures of Prevention, Application, and Economics of Corrosion Technologies Study, NACE, (2016). [2] J.W. Yeh, S.K. Chen, S.J. Lin, J.Y. Gan, T.S. Chin, T.T. Shun, C.H. Tsau, S.Y. Chang, Nanostructured High-Entropy Alloys with Multiple Principal Elements: Novel Alloy Design Concepts and Outcomes, Advanced Engineering Materials, 6 (2004) 299-303. [3] O.N. Senkov, S.V. Senkova, C. Woodward, Effect of aluminum on the microstructure and properties of two refractory high-entropy alloys, Acta Materialia, 68 (2014) 214-228. [4] Z. Li, K.G. Pradeep, Y. Deng, D. Raabe, C.C. Tasan, Metastable high-entropy dual-phase alloys overcome the strength–ductility trade-off, Nature, 534 (2016) 227. [5] M.A. Hemphill, T. Yuan, G.Y. Wang, J.W. Yeh, C.W. Tsai, A. Chuang, P.K. Liaw, Fatigue behavior of Al0.5CoCrCuFeNi high entropy alloys, Acta Materialia, 60 (2012) 5723-5734. [6] Z. Tang, T. Yuan, C.-W. Tsai, J.-W. Yeh, C.D. Lundin, P.K. Liaw, Fatigue behavior of a wrought Al0.5CoCrCuFeNi two-phase high-entropy alloy, Acta Materialia, 99 (2015) 247-258. [7] B. Gludovatz, A. Hohenwarter, D. Catoor, E.H. Chang, E.P. George, R.O. Ritchie, A fracture-resistant high-entropy alloy for cryogenic applications, Science, 345 (2014) 1153-1158. [8] Z. Zhang, M. Mao, J. Wang, B. Gludovatz, Z. Zhang, S.X. Mao, E.P. George, Q. Yu, R.O. Ritchie, Nanoscale origins of the damage tolerance of the high-entropy alloy CrMnFeCoNi, Nature communications, 6 (2015) 10143. [9] L.J. Santodonato, Y. Zhang, M. Feygenson, C.M. Parish, M.C. Gao, R.J.K. Weber, J.C. Neuefeind, Z. Tang, P.K. Liaw, Deviation from high-entropy configurations in the atomic distributions of a multi-principal-element alloy, Nature Communications, 6 (2015) 5964. [10] Y.Y. Chen, T. Duval, U.D. Hung, J.W. Yeh, H.C. Shih, Microstructure and electrochemical properties of high entropy alloys—a comparison with type-304 stainless steel, Corrosion Science, 47 (2005) 2257-2279. [11] Y.L. Chou, Y.C. Wang, J.W. Yeh, H.C. Shih, Pitting corrosion of the high-entropy alloy Co1.5CrFeNi1.5Ti0.5Mo0.1 in chloride-containing sulphate solutions, Corrosion Science, 52 (2010) 3481-3491. [12] Q.H. Li, T.M. Yue, Z.N. Guo, X. Lin, Microstructure and Corrosion Properties of AlCoCrFeNi High Entropy Alloy Coatings Deposited on AISI 1045 Steel by the Electrospark Process, Metallurgical and Materials Transactions A, 44 (2013) 1767-1778. [13] Y.-J. Hsu, W.-C. Chiang, J.-K. Wu, Corrosion behavior of FeCoNiCrCux high-entropy alloys in 3.5% sodium chloride solution, Materials Chemistry and Physics, 92 (2005) 112-117. [14] Y. Shi, B. Yang, X. Xie, J. Brechtl, K.A. Dahmen, P.K. Liaw, Corrosion of Al xCoCrFeNi high-entropy alloys: Al-content and potential scan-rate dependent pitting behavior, Corrosion Science, 119 (2017) 33-45. [15] Y.-F. Kao, T.-D. Lee, S.-K. Chen, Y.-S. Chang, Electrochemical passive properties of AlxCoCrFeNi (x=0, 0.25, 0.50, 1.00) alloys in sulfuric acids, Corrosion Science, 52 (2010) 1026-1034. [16] C.P. Lee, C.C. Chang, Y.Y. Chen, J.W. Yeh, H.C. Shih, Effect of the aluminium content of AlxCrFe1.5MnNi0.5 high-entropy alloys on the corrosion behaviour in aqueous environments, Corrosion Science, 50 (2008) 2053-2060. [17] C.P. Lee, Y.Y. Chen, C.Y. Hsu, J.W. Yeh, H.C. Shih, The Effect of Boron on the Corrosion Resistance of the High Entropy Alloys Al0.5CoCrCuFeNiB x, Journal of The Electrochemical Society, 154 (2007) C424-C430.
44
[18] Y.L. Chou, J.W. Yeh, H.C. Shih, The effect of molybdenum on the corrosion behaviour of the high-entropy alloys Co1.5CrFeNi1.5Ti0.5Mox in aqueous environments, Corrosion Science, 52 (2010) 2571-2581. [19] J. Li, X. Yang, R. Zhu, Y. Zhang, Corrosion and Serration Behaviors of TiZr0.5NbCr0.5VxMoy High Entropy Alloys in Aqueous Environments, Metals, 4 (2014) 597. [20] G.S. Frankel, Pitting corrosion of metals - A review of the critical factors, Journal of the Electrochemical Society, 145 (1998) 2186-2198. [21] R.F.A. Jargelius-Pettersson, Application of the Pitting Resistance Equivalent Concept to Some Highly Alloyed Austenitic Stainless Steels, CORROSION, 54 (1998) 162-168. [22] K. Lorenz, G. Medawar, Über das Korrosionsverhalten austenitischer Chrom-Nickel- (Molybdän-) Stähle mit und ohne Stickstoffzusatz unter besonderer Berücksichtigung ihrer Beanspruchbarkeit in chloridhaltigen Lösungen, Thyssenforschung, 1 (1969) 97-108. [23] J.H. Cleland, What does the pitting resistance equivalent really tell us?, Engineering Failure Analysis, 3 (1996) 65-69. [24] G. Herbsleb, Der Einfluß von Schwefeldioxid, Schwefelwasserstoff und Kohlenmonoxid auf die Lochkorrosion von austenitischen Chrom-Nickel-Stählen mit bis zu 4 Massen-Molybdän in 1 M Natriumchlorid-Lösung, Materials and Corrosion, 33 (1982) 334-340. [25] H. Okamoto, The effect of tungsten and molybdenum on the performance of super duplex stainless steels, in: Proc. Conf. Application of Stainless Steels ‘92, 1992, pp. 360-369. [26] C.J. Kuehmann, G.B. Olson, Computational materials design and engineering, Materials Science and Technology, 25 (2009) 472-478. [27] ASTM G1-03(2017)e1 Standard Practice for Preparing, Cleaning, and Evaluating Corrosion Test Specimens, ASTM International, West Conshohocken, PA, 2017 [28] Frankel, G. S., Li, T., & Scully, J. R. (2017). Perspective—localized corrosion: Passive film breakdown vs pit growth stability. Journal of the Electrochemical Society, 164(4), C180-C181. [29] Jones, D. A. (1996). Principles and prevention of corrosion prentice hall. Saddle River, NJ [30] Marcus, P. (1998). Surface science approach of corrosion phenomena. Electrochimica Acta, 43(1-2), 109-118. [31] V.M. Salinas-Bravo, R.C. Newman, Corrosion Science, 36 (1994) 67-77.
45
Chapter 4:
CORROSION BEHAVIOR OF Ni38Fe20Crx(MnCo)42-x HIGH ENTROPY ALLOYS IN 0.6M NACL SOLUTION 4.1 INTRODUCTION High entropy alloys (HEAs) are a new and developing branch of alloys having a composition consisting of five or more principal elements, giving the alloy very high configurational entropy. HEAs were first presented by Yeh et al.[1], showing that this new alloy system, which uses unconventional alloying techniques, can produce alloys with exceptional properties, such as high strength and ductility, improved fatigue resistance, high fracture toughness and thermal stability. High entropy alloys have also exhibited enhanced corrosion resistance, making them a candidate for a wide range of practical applications.
In an early study of HEA corrosion resistance, Chen et al. studied the general corrosion resistance of SS304 and Cu0.5NiAlCoCrFeSi HEA, showing that the general corrosion resistance of the HEA was superior to SS304 [2]. However, SS304 was shown to have the superior resistance to pitting corrosion. Further, SS304 was compared to
Co1.5CrFeNi1.5Ti0.5Mo0.1 HEA, which showed superior corrosion resistance for both general and localized corrosion. The HEA was determined to have a critical pitting temperature (CPT) in the range of 60 ~ 70°C in chloride solution, which is far higher than conventional commercial corrosion resistant alloys i.e. SS304 and SS316 [2]. The corrosion performance for HEAs is seen to largely vary according to their composition and processing.
The HEAs of interest in this study are capable of forming a single phase as well as being primarily composed of nickel, and iron. Previous studies on the AlxCoCrFeNi and
AlxCrFe1.5MnNi0.5 HEA systems were conducted by Kao et al. [3] and Lee et al. [4], respectively. The studies were both conducted with the alloy of each of their series free of Al, the
46 results show that the Co containing alloy has greater corrosion resistance to general corrosion than the Mn containing alloy. While both alloys show a single phase nature, they differ greatly in overall corrosion resistance due to their compositions.
Nickel alloy pitting potentials show a strong dependence on chloride concentration, as well temperature. The higher the temperature, the more likely is pitting to occur, and the lower will be the pitting potential. Pitting in Ni-based alloys is also highly influenced by the structure and composition of the passive film. When passive films are amorphous, or defects such as grain boundaries or dislocations are abundant, these act as pathways for ions to diffuse through the passive film, lowering the pitting potential of the alloy. [5,6] Nickel alloys are highly susceptible to crevice corrosion, which has often been found to be more of a problem than pitting for Ni- based alloys. Kehler et al. found that high temperatures and chloride concentrations resulted in more active crevice stabilization for the Ni-based alloys 625 and C-22 [7].
The chemical composition of HEAs plays the major role in determining the nature of the passive film, which influences the initiation of both pits and crevices. Composition also strongly influences the progression of a pit or crevice, by influencing the dissolution rate of metal ions from the pit/crevice. The influence of chemical composition on pitting corrosion of Fe-Cr-Ni alloys can be described by the pitting resistance equivalent number (PREN), an empirical fit of corrosion metrics, such as pitting potential or critical pitting temperature, to the composition.
Various equations for PREN have been developed during the last few decades [8-12]. One commonly used PREN equation is [12]:
푃푅퐸푁 = 푤푡. %퐶푟 + 1.6푤푡. %푊 + 3.3푤푡. %푀표 + 16푤푡. %푁 퐸푞. (1)
Generally, alloys with higher PREN values have higher pitting resistance. From the information presented above, one can expect that a single phase alloy as well as a high PREN value are two
47 factors to be taken into consideration when designing a corrosion resistant HEA (CR-HEA).
However, PREN equations, such as the one in Eq.1, are only valid for the Fe-Cr-Ni system, and cannot be extended to alloys of other elements.
HEAs reported previously are of equiatomic or near equiatomic concentration, commonly with multi-phase, and dendritic structures, as well as a few single-phase structures. These equiatomic alloys possess high configurational entropy, but are not optimized for corrosion resistance due to their typical multi-phase structure. Using conventional trial and error approaches to alloy design of single-phase corrosion resistance HEAs would be costly and time consuming, to explore such a large compositional space. The use of integrated computational materials engineering (ICME), which is a modern systems-based approach to design materials that meet a specific need for performance by linking computational materials models across multiple length scales [13], can be applied for the design and production of corrosion resistant, single-phase, non-equiatomic HEAs. This study focuses on the corrosion resistance of HEAs designed by combining known corrosion resistant elements and computation to predict single phase structures, with the goal of understanding the traits of alloying additions in an HEA system.
4.2 EXPERIMENTAL 4.2.1 Specimen preparation
The goal of this work was to study a series of HEAs with a wide range of expected corrosion resistance by varying the Cr content from well above to well below 12%, which is known to be a critical value for passivity. As many of the alloyus were expected to exhibit much worse corrosion resistance than the Ni-based HEA described in the last chapter, they are referred to as
48 canary alloys. The chemical compositions of the high-entropy alloys used in this work are presented in Table 1. The naming convention of the canary alloys corresponds to the Cr concentration of each alloy, thus Ni38Fe20Cr22Mn10Co10, Ni38Fe20Cr14Mn14Co14,
Ni38Fe20Cr10Mn16Co16, and Ni38Fe20Cr6Mn18Co18 will further be referred to as CanCr22,
CanCr14, CanCr10, and CanCr6, respectively. Fig. 4.1 shows the distribution of compositions in a CALPHAD generated pseudo-binary phase diagram, provided by J. Saal and P. Lu from
QuesTek. The phase diagram has a large single-phase FCC region at high temperatures, within which a variety of single phase FCC compositions can be selected. The alloy series composition was manipulated by replacing Cr concentration with equal amounts of Mn and Co to generate alloys with expected varying corrosion resistance. The alloys were prepared by arc melting of constituent elements starting from basis metals of purities 99%, 99.98%, 99.99%, 99%, 99.5% for Ni, Fe, Cr, Mn, and Co respectively, under a 0.5 atm Ar atmosphere. An extra 1 wt.% of Mn was added to the initial materials to compensate for the expected evaporation of Mn, which occurs during arc melting of Mn-rich alloys[14]. Each specimen was flipped and melted at least five times to improve the homogeneity of the specimen. The alloys were prepared as buttons about 1 cm thick and 2 cm in diameter. Buttons of all four alloy compositions were vacuum encapsulated and backfilled to 0.125 atm of inert Ar gas. The specimens were then heat treated at
1100°C for 96 h followed by water quenching to generate single-phase FCC alloys with uniformly distributed compositions.
4.2.1 Microstructure characterization
Alloys were polished to a 1µm finish using diamond paste and cleaned with ethanol before microstructure characterization. Microstructure characterization was performed with a
Quanta 200 scanning electron microscope (SEM) and a Rigaku SmartLab type X-ray
49 diffractometer (XRD). The semi-quantitative analysis of the chemical compositions was conducted using an SEM equipped with energy dispersive spectroscopy (EDS)
4.2.2 Electrochemical Corrosion Characterization
The corrosion resistance of the alloys was evaluated by potentiodynamic polarization.
Prior to testing, each button was cut into 2 mm thick sections, and each section was ground with
SiC paper to1200, and immediately prepared for electrochemical testing. The samples were tested using two separate preparation methods. Initially, the samples were connected to a lead wire, wrapped with polytetrafluoroethylene (PTFE) tape and coated with black wax, leaving an exposed working area of about 0.2 cm2. Potentiodynamic experiments were conducted under the thermostatic condition of 30°C. The electrochemical cell used in this work was a 250 ml jacketed beaker and approximately 200 ml of solution was used for each experiment. All solutions were exposed to air. The polarization experiments were performed with a conventional three-electrode cell, consisting of a working electrode, a platinum counter electrode, and a saturated calomel electrode (SCE) as the reference electrode. The potential was controlled and the current was measured using a Gamry Reference 600 potentiostat with a computer-controlled electrochemical interface, allowing for monitoring of the total current (I), potential (V), and time (t). All experiments were conducted at a scan rate of 20 mV/min from the initial potential of -0.25 V versus the open circuit potential (OCP) and terminated when the current reached 1mA. Prior to the experiment, the specimen was cathodically polarized at -1 V SCE for 180 s. The OCP was then measured for 10 min to acquire a steady state potential. Following the polarization experiment, the specimens were washed and then dried with compressed air. The morphology of the corroded surface was then examined by a scanning electron microscope. To determine reproducibility, tests were repeated three times under each condition.
50
In subsequent experiments, a microcell was used as a novel hanging droplet cell to prevent crevice corrosion caused by sample masking. The microcell is capable of maintaining constant droplet sizes, allowing for reproducible contact areas for the electrolyte on the electrode surface. In a typical microcell the electrolyte is typically kept under vacuum via a syringe that is able to maintain pressure to keep the droplet a constant droplet size for the duration of testing.
Similar to a wire loop electrode [15], part of the sample is exposed to a liquid/air interface, but no masking is required. The exposed test area can be approximated to the opening of the microcell, which is 1.5 mm, which resulted in exposed areas of about 0.13 cm2. The exposed area can be determined more accurately after testing with the use of an optical microscope and image analysis. These droplet experiments were performed at room temperature, ~25°C. The sample was placed beneath the microcell capillary and the microcell stage was raised until the droplet extending from the capillary opening made contact with the surface of specimen. The typical three electrode set up was contained in the microcell with the Pt counter and the SCE reference electrode maintaining close proximity within the microcell and the working electrode being connected through contact with the hanging droplet. All experiments were conducted at a scan rate of 1 mV/s from the initial potential of -0.1 V versus the OCP, and terminated when the current reached 100 µA. Cyclic polarization tests were also conducted with scan direction reversal at 0.1 mA, and using forward and reverse scan rates of 1 mV/s. Prior to the experiment, the OCP was recorded for 100 s to acquire a steady state potential.
Tests of the canary alloys for CPT or critical crevice temperature (CCT) were performed using samples masked with PTFE tape and black wax in a jacketed cell connected to a programmable temperature-controlled circulator and standard three-electrode set up. A temperature probe was placed in solution next to the sample to monitor solution temperature.
51
The testing solution for each measurement was approximately 300 mL of 0.01 M NaCl solution under aerated conditions. A 0.01 M NaCl test solution was used to ensure that crevicing could be clearly identifiable by increase in the current density of each alloy at a potential of 0.7 V SCE.
The electrochemical experiments were carried out using a Gamry Reference 600 potentiostat.
The samples were immersed in the test solution at an initial temperature of 4oC and polarized to a potential of 700 mV SCE. This potential was held constant during the entire temperature scan.
After 10 min of stabilization, the test solution was heated by the water circulator at a rate of 1oC min-1. The actual heating rate of the testing solution at the initial stage was non-linear and smaller than 1oC min-1, but it increased gradually and was stabilized at 1oC min-1 after a certain period. The current was collected at a rate of 1 Hz, and CPT was defined as the temperature at which the current increased rapidly and continuously exceeded 100 A/cm2 for 60 s.
Measurement of critical pitting temperature was attempted using the hanging droplet cell and a copper cooled stage, which is capable of sustaining a minimum alloy surface temperature of -6°C. T The alloy was tested potentiodynamically in 0.6 M NaCl solution droplets at low temperatures with the goal of identifying a temperature where the alloy will undergo only transpassive dissolution with no evidence of pitting.
4.3 RESULTS AND DISCUSSION Most HEA alloys have been produced without attempting to form a single-phase structure, instead exhibiting typical as-cast dendrite and interdenritic structures. The equatomic
CoCrFeNiMn alloy was studied by Stepanov et al. [16]. It was heat treated at between 600 to1100°C, and was shown to have a single-phase FCC nature at higher treatment temperatures i.e. 1100°C. The alloys generated in this work are non-equatomic but all present single phase
52
FCC solid-solution structures. It is a breakthrough in alloy design to be able to produce alloys with predictable structures, with a goal of improving alloy properties.
4.3.1 Characterization
Fig. 4.2 shows secondary electron SEM images of the four homogenized samples. Pores can be observed, but they are small in size, ~1-5 µm in diameter. The four homogenized samples all appear to be single phase with no inclusions or segregation of alloying elements, but some residual surface contaminants from the grinding process are visible as white spots. Fig. 4.3 shows the EDS maps of the CanCr22 section, revealing uniform distribution of alloying elements over a large area of the sample. The other three alloy compositions also exhibited homogeneous structures, thus only CanCr22 is shown here. Fig. 4.4 shows the XRD spectra of all four canary alloys. XRD patterns were analyzed through conventional peak identification to identify the crystalline structures, indicating that all four alloys after homogenization are FCC crystalline structures. The diffraction peaks of the high-chrome alloys are very well defined, indicating a solid-solution FCC structure, while the diffraction peaks of the low-chrome alloys are noisier.
This is common for alloys that have a greater amorphous nature to their structure. The greater amorphous structure might be beneficial to the corrosion performance by allowing these alloys to more easily form a passive oxide film [1, 2, 17].
4.3.2 Conventional immersion cell testing
A conventional immersion cell allowing for the complete immersion of masked samples in testing solution was used for electrochemical testing of the canary alloys. Fig. 4.5 displays typical potentiodynamic polarization curves of the canary alloys in 0.6 M NaCl at 30C. The
53 polarization curves were generated while these specimens were immersed in solution with a selected area exposed. All four alloy systems exhibit passivity in this solution, which is remarkable for the alloys with 10 and 6% Cr. The passive current density is a weak function of
Cr content on the order of 1 A/cm2 for the alloys with 10-22% Cr. The CanCr6 alloy exhibits a much higher passive current density, about 10 A/cm2, but the passive nature is still clear. All alloys also show a large amount of metastable events in the passive region, starting at potentials slightly higher than their OCPs. The unusually large amount of metastable events at potentials only slightly higher than the OCP suggest that the passive films were undergoing a large amount of breakdown and repassivation events. Investigation of the samples after testing showed that the breakdown of all alloys was associated with crevice corrosion; Fig. 4.6 shows typical corrosion images of the CanCr alloys. There was little to no general corrosion, as crevice corrosion was the dominant corrosion form. These alloys are very susceptible to crevice corrosion, which could not be avoided by standard masking methods.
The large number of metastable events can be partly attributed to the poor quality of the passive film on the alloys. A poor passive film is generally a factor of both metal-metal (M-M) bond strength as well as metal-oxygen (M-O) bond strength [18]. From this perspective, the canary alloys seem to have relatively low M-M and M-O bond strength in the 0.6 M NaCl solution, even though the bonding is strong enough to form a stable oxide at room temperature in
Cl- solutions. Meanwhile, these oxide films at higher temperatures or under applied potential are easily ruptured [19]. The lack of strong passivity is surprising for the CanCr22 alloy as an alloy with 22% Cr would be expected to exhibit a strong passive film. Furthermore, it was a clean alloy absent of susceptible particles such as MnS or second phases that might be expected to degrade the corrosion performance. So the CanCr alloys exhibit an interesting combination of
54 unexpected passivity (in the 10 and 6% Cr alloys) and unexpected metastable breakdown (in the
22% Cr alloy).
Fig. 4.7 displays typical cyclic polarization of the four alloys, which all show the presence of positive hysteresis loops associated with propagation of localized corrosion. As the alloy composition decreased in Cr content, the hysteresis loops became larger and the repassivation potentials decreased, as would be expected. The positive hysteresis as well as the corresponding low repassivation potentials are representative of a high susceptibility to localized corrosion. The critical potentials are compiled in Fig. 4.8. These measurements have a great deal of scatter in them, but they do point towards some trends that have been widely accepted, such as the increase in Ecrev with an increase in Cr content. The repassivation potential also increased with higher Cr composition.
Fig. 4.9 displays the currents recorded during CCT measurements of the canary alloys in
0.01 M NaCl solution. CCT testing shows that these alloys are all highly susceptible to crevice corrosion. The curve for CanCr22 alloy allows for assessment of a critical temperature of about
11°C using the ASTM-150 standard of greater than 100 µA/cm2 current density for more than 60 seconds. All other alloy systems surpass this threshold at the initial starting temperature of 4°C, indicating that the critical temperature was lower. Further low temperature testing was done in a attempt to how much lower the crevice temperature might be.
Fig. 4.10 displays the potentiodynamic polarization curve for CanCr22 at the lowest temperature possible using the temperature control equipment, -15°C. The solution concentration was increased to 3.6 M NaCl to prevent the solution from freezing during testing at -15°C. The concentration of a solution has been shown to have very little effect on the temperature at which an alloy will present active localized corrosion [20]. The curve in Fig. 4.10 and the evidence of
55 crevice corrosion after the experiment indicated that the CanCr22 alloy was unable to reach high potentials for transpassive dissolution without exhibiting localized corrosion. The critical crevice temperature must be yet lower than -15°C.
Potentiodynamic testing of the CanCr22 at -15°C in 3.6 M NaCl resulted in crevice corrosion breakdown at 0.4 V SCE. However, the results of temperature scanning of the same alloy starting at 4°C with an applied potential of 700 mV SCE in 0.01 M NaCl could be interpreted to show that the CCT was 11°C using the ASTM-G150 threshold [21]. The two results are in stark contradiction. While the chloride concentrations in these solutions are very different, it has been reported that there is only a slight relationship between the solution concentration and critical pitting temperature [20]. Therefore, there is an obvious gap in the criteria of both CCT and CPT where the alloy is undergoing pitting or crevicing at lower temperatures than the CCT suggested by the current ASTM standards.
4.3.3 Droplet cell experiments
Crevice corrosion is an artifact of testing for many corrosion resistant alloys that has plagued the corrosion community for many years. Prevention measures, such as the application of different masking to assess the corrosion of selected areas, has proven to be difficult for a wide range of specimens. SS304, which is known to be highly susceptible to crevice corrosion, is commonly analyzed and tested for its resistance to pitting corrosion. Nickel-based alloys and super alloys are also known for their limited resistance to crevice corrosion. When the masking on the surface of these alloys is applied in a less precise manner, they will often show artifacts of crevice corrosion. To prevent crevice corrosion in this series of varying Cr-content HEAs, the alloy was first tested using traditional methods of masking with conventional black wax masking and then tested under complete immersion. All samples were potentiodynamically tested with
56 multiple repeats. After removal of the mask, all samples show evidence of moderate to severe crevice corrosion. The novel hanging droplet cell described above has the ability to prevent crevice corrosion associated with masking. The cell allows for a selected area to be defined and limited to the area of contact between the cell and the droplet, therefore, there is no longer a need for masking of the sample. The samples prepared here were ground to 1200 grit with SiC paper and water, cleaned with ethanol and then electrically connected to a lead. This preparation was much shorter and easier than for conventional masking methods. The samples are free from any and all crevice corrosion artifacts as there is no masking material. Only pits and no crevices were formed using this testing method.. This approach allows the study of pitting characteristics for alloys with high susceptibility to crevice corrosion.
Fig. 4.11 displays OCP measurements in for samples in 0.6 M NaCl in the droplet call.
The OCP of each alloy was monitored for 30 min. The microcell was used as a hanging droplet cell to produce droplets of nearly uniform sizes between 0.12-0.15 cm2. The OCPs of the alloys are shown to become more stable as the concentration of Cr increases. Both CanCr14 and
CanCr22 present stable oxides, while CanCr6 and 10 present oxides of a less stable nature, with numerous potential drops indicative of metastable pitting at the OCP. Nonetheless, the potential always returned to the higher value, indicating passive behavior.
Fig. 4.12 displays the potentiodynamic polarization curves for the four alloy systems using the hanging droplet cell method. The curves now show a much more distinct and consistent separation of pitting potentials between the four alloys compared to previous testing where crevice corrosion was present. The pitting potentials are higher for the high-Cr alloys. This aligns with conventional knowledge of the benefits of Cr alloying in HEAs [4].
57
Looking at all relevant measurements, the passive films forming on both CanCr6 and
CanCr10 are weak when compared to CanCr14 and CanCr22. However, these passive films are still strong enough to be considered passive, and they break down to form pits at high potentials in the 0.6 M NaCl solution. The formation of pits requires an alloy to be passive, otherwise general corrosion is observed. The passive nature of these low chromium HEAs contradicts conventional knowledge for corrosion resistant alloys. Newman et al. has studied the passivity in stainless steels [22], showing that there is a minimum Cr content for a passive stainless steel between 12-13wt.% Cr alloying. Both CanCr6 and CanCr10 are below the minimum chromium composition required to demonstrate passivity, however both alloys have proven to be passive despite their low chromium compositions. This remarkable property can be attributed to a number of unique HEA qualities. In regards to alloying composition, the other elements in the alloys other than Cr (i.e. Ni, Fe, Mn, Co) are not known to be strong passive film formers [23].
The relative corrosion resistances of the elements other than Cr are broadly similar to that of Fe.
However, the combined alloying effect of HEAs, as reported by Yeh et al.[24] and Zhang et al.[25], has shown that the alloy components as well as their interactions have beneficial properties, related to many physical properties as well as alloy aqueous dissolution [24, 25].
HEAs are known to be solid-solution formers, in the case of this study all alloys are single phase
FCC alloys free of segregation and inclusions [1].
4.3.4 Crystallographic pits in HEAs
Fig. 4.13 displays SEM images the pits seen for the four HEAs, showing that the pits in all alloys are crystallographic in nature. Small pits are seen near the water line, at the meniscus of the droplet. These pits appear in for both the CanCr6 (Fig.4.13 a), b)) and CanCr10 (Fig.4.13 c), d)) alloys, some of these pits seem to have a phase present in the center where dissolution in
58 not occurring as rapidly as the remainder of the pit. Fig.4.13 a) and c) show coalescence of these pits at the waterline edge. Fig.4.13 e), f), g), and h) show large crystallographic pits found on
CanCr6, 10, 14, and 22 respectively. These large pits appear away from the waterline edge. The
CanCr14 and CanCr22 alloys show fragmented pit covers. Pits in most alloys are polished when formed during cyclic polarization experiments in which the sample is scanned to high potentials to initiate breakdown and pit growth proceeds during the period of backscanning in the positive hysteresis potential region. Pits and crevices formed at low potentials, for instance during long term exposure testing, can be crystallographic in nature. It is generally accepted that a critical pit solution concentration at the pit surface, Ccrit, is needed to prevent repassivation, and that Ccrit is high, but less than the saturation concentration, Csat, which would be associated with polishing.
However, the conditions must be right to maintain a concentration between Ccrit and Csat and result in crystallographic dissolution.
A new theory for pit stability indicates that the maximum rate of dissolution in a pit should exceed the rate of transport out of the pit to maintain Ccrit at the pit surface, i.e. imax>icrit
[26]. From the aspect of pit dissolution kinetics, there are relationships observed between alloy composition and imax, imax, and pit stability, as well as pit stability and pitting tendency. From the aspect of diffusion kinetics, the alloy composition may influence icrit and icorr through their effect on the robustness of pit covers. Generally, robust pit covers allow metastable pits to grow deeper to increase the pit stability. So passive film properties affect both pit initiation and pit stabilization. Additionally, the alloy composition might also influence the Ccrit, and the relationship should be between alloy composition and Ccrit, Ccrit and icrit, icrit and pit stability, pit stability, and pitting tendency [27].
59
Furthermore, although a salt film is not essential to pit stability, it can promote the transition from metastable to stable pit growth by providing a reservoir of metal cations to maintain the aggressive environment under transient conditions [26]. For example, when the rupture of the pit cover exceeds the critical value (a*) before the depth of a metastable pit reaches the critical depth to sustain pit growth in the uncovered state (r*), a salt film, if present, can dissolve to replenish the metal cations to maintain the sufficiently aggressive environment required for active dissolution. This might allow the metastable pit to keep growing to reach r* and then transition to stable growth. Therefore, it can be predicted that an alloy that can easily form a salt film on a pit surface will have a higher likelihood to produce stable pits and thus exhibit an increased pitting tendency [27]. In contrast, if the salt film is hard to form on the pit surface, the pitting tendency will decrease. Salt films can precipitate only if the Csat is reached
(actually supersaturation is often required). Thus, the relationship between the alloy composition and pitting tendency can also be established in this way:
Alloy composition Csat salt film pit stability pitting tendency
To predict pitting tendencies, an alloy’s pit cover robustness, pit dissolution kinetics, and pit critical concentration and saturation concentration must be well understood [28].
4.3.5 Mechanism for formation of crystallographic pit morphologies on canary alloys
Fig. 4.12 shows that metastable pitting events in the potentiodynamic measurements occur very frequently on each alloy. Thus, the explanation of the increase in Epit with Cr concentration can be explained from the perspective of pit stability. Fig. 4.13 shows pit morphologies after the test. Interestingly, the pit morphologies on all of these different samples are crystallographic; no hemispherical polished pits are observed. Crystallographic morphologies
60 are also observed in crevices in these alloys, as shown in Fig. 4.2. While pits and crevices in
SS304 formed under the same conditions of cyclic polarization to high currents always have electropolished surfaces.
It is generally accepted that polished pit morphologies are associated with the precipitation of the salt film, while crystallographic morphologies indicate the pit growth is controlled by charge transfer and ohmic effects without the presence of a salt film. Hence, one possible explanation for these crystallographic pits is that the Csat of these canary alloys is high and cannot easily be reached. Because chloride salts containing multiple transition metal cations do not exist [29], a single cation salt will form as the pit concentration in an alloy increases.
After the first salt forms, the dissolution rate of all elements will be limited, not allowing for the formation of subsequent or secondary salt films. Thus, the Csat for a pit should be defined as the total concentration of metal cations for precipitation of the first salt. Csat for a conventional alloy with a dominant host element should be similar to that for a pure metal of the dominant element.
For example, the Csat for stainless steel is essentially equal to the solubility of FeCl2, 4 mol/L.
However, the elements in HEAs are in near-equal atomic concentration. Therefore when the first salt precipitates, the concentrations of other dissolved cations should also be considerable (based on the assumption that dissolution occurs congruently). Thus, it can be predicted that the Csat for
HEAs should be higher than the solubility of any pure metal chloride salt.
A high Csat indicates a high ilim. As a result, while imax can exceed icrit during pit growth in these alloys, it cannot reach ilim, so the pit morphology remains crystallographic. Since reaching ilim and precipitating a salt film increases the likelihood for metastable pits to survive pit cover rupture events, the lack of a salt film promotes metastable pit repassivation in these alloys.
61
Higher imax values and correspondingly higher applied potentials are required for the stabilization of metastable pits in these alloys.
Fig. 4.14 displays cyclic polarization measurements obtained with the hanging droplet cell. The cyclic polarizations show similar increasing positive hysteresis and decreasing repassivation potentials in the series of alloys from CanCr22 to CanCr6. The same was seen in the cyclic testing under immersion conditions that resulted in crevice corrosion presented above,
Fig. 4.7. The CanCr22 alloy consistently exhibits metastabilty on the downward scan after repassivation. This is an interesting observation, which has not been seen in other alloys, and can be similarly explained by the nature of the multicomponent alloy system. As these passive films are reforming, there is still a high enough potential where this alloy would typically be susceptible to pitting. However, it is not able to reach the Csat necessary to generate a sufficiently corrsosive environment, which could sustain a stable pit. Therefore metastabilty is also observed after passivation of the alloy. The lower current on the downward scan also provides a smaller background allowing for the observation of small metastable pits.
The droplet cell data are compiled in Fig. 4.15 showing the effects of Cr content on Ecorr,
Epit, and Erp. There are clear, near linear, trends between Cr content and both Epit and Erp. On the other hand, there is no representative trend between Cr content and Ecorr. Studies on other HEA alloys have also found that the corrosion resistance of an alloy can be greatly increased by the addition of chromium alloying [4].
The hanging droplet cell was used in an attempt to determine the CPT. Because of the setup, the lowest achievable alloy surface temperature was -6°C. Even though it was previously shown that CCT was below -11°C, testing at -6°C was still of interest because it was expected that CPT would be higher than CCT. Fig. 4.16 shows the resultant polarization curve recorded
62 for CanCr22 in 0.6 M NaCl with the hanging droplet cell at -6°C. The polarization curve shows a pitting at a potential of 0.4 V SCE, indicating that the critical pitting temperature of the alloy is below -6°C. Fig. 4.17 shows the resulting pit had a lacy cover and after removal of the cover the pit is seen to be a conventional hemispherical polished pit. The formation of a lacy pit at low temperatures suggests that the passive film is more robust at low temperatures, promoting the formation of a lacy cover. The lacy cover acts as a diffusion barrier, which enhances the likelihood that the saturation concentration can be reached and a salt film to form [28]. Similarly, the polarization curve shows much less metastabilty at low temperatures, suggesting that the passive film has far greater stability. The more stable passive film, acting as a diffusion barrier, may allow for the formation of a stable pit at low temperatures. However, the critical pitting temperature is still far below -6°C, and cannot be studied using current methods and equipment.
4.4 CONCLUSIONS (1) The low chromium alloys are seen to be uniquely passive below the conventional
standards required for passivity of a stainless steel.
(2) There is a great deal of metastable activity seen, which corresponds to the inability of the
alloy to reach a critical concentration of metal cations needed for formation of stable pits
or crevices.
(3) The novel hanging droplet cell has shown utility in its ability to prevent crevice corrosion
on highly-susceptible materials, allowing for greater understanding and future research of
alloy pitting.
(4) The HEA crystallographic pitting nature can be attributed to the inability of multiple
element salts to form. This leads to the inability to generate stable pits without large
enough potentials to generate current densities greater than icrit.
63
Acknowledgements: This work was supported as part of the Center for Performance and Design of Nuclear Waste Forms and Containers, an Energy Frontier Research Center funded by the U.S.
Department of Energy, Office of Science, Basic Energy Sciences under Award # DE-SC0016584
64
Figures
Table 4. 1: Chemical Composition (wt%) of materials.
Designation Composition (atomic ratio) Ni(58.69) (%) Fe(55.84) (%) Cr(51.99) (%) Mn(54.94) (%) Co(58.93) (%)
CanCr22 Ni38Fe20Cr22Mn10Co10 39.62 19.84 20.32 9.76 10.47
CanCr14 Ni38Fe20Cr14Mn14Co14 39.34 19.70 12.84 13.57 14.55
CanCr10 Ni38Fe20Cr10Mn16Co16 39.20 19.63 9.14 15.45 16.57
CanCr6 Ni38Fe20Cr6Mn18Co18 39.07 19.56 5.46 17.32 18.58
Figure 4.1:CALPHAD pseudo-binary phase diagram for Ni38Fe20Crx(MnCo)42-x Information provided by J. Saal and P. Lu, QuesTek.
65
Figure 4.2: SEM images of homogenized a)Ni38Fe20Cr22Mn10Co10 b) Ni38Fe20Cr14Mn14Co14 c)Ni38Fe20Cr10Mn16Co16 d)Ni38Fe20Cr6Mn18Co18.
Figure 4.3: EDS mapping of CanCr22 after homogenization, showing even distribution of element.
66
Figure 4.4: XRD peak analysis of CanCr22, 14, 10, and 6 all with representative FCC structure peaks.
0.3 CanCr22 0.2 CanCr14 CanCr10 0.1 CanCr6
0.0 )
SCE -0.1
-0.2
-0.3 Potential(V
-0.4
-0.5
-0.6 1E-10 1E-9 1E-8 1E-7 1E-6 1E-5 1E-4 1E-3 Current Density (A/cm2)
Figure 4.5: Potentiodynamic polarization curves of the canary alloys in 0.6M NaCl at 30°C in a typical immersion cell.
67
Figure 4.6: Typical SEM images of crevice corrosion found after polarization in 0.6M NaCl at 30°C in a typical immersion cell.
CanCr22 0.4 CanCr14 CanCr10 CanCr6
0.2 )
SCE 0.0
-0.2 Potential(V
-0.4
-0.6 1E-10 1E-9 1E-8 1E-7 1E-6 1E-5 1E-4 1E-3 Current Density (A/cm2)
Figure 4.7: Typical cyclic polarization curves of the canary alloys in 0.6M NaCl at 30°C in a typical immersion cell.
68
0.4 Ecorr (V) Ecrev (V) 0.3 Erp (V)
0.2 )
SCE 0.1
0.0
Potential(V -0.1
-0.2
-0.3
4 6 8 10 12 14 16 18 20 22 24 Chrome Content (at%)
Figure 4.8: Graphical representation of chromium alloying effect on critical crevice corrosion metrics, Ecorr, Ecrev, and Erp, in Ni38Fe20Crx(MnCo)42-x.
0.004
) 2 0.003
0.002
CurrentDensity(A/cm 0.001
0.000 4 6 8 10 12 14 16 18 Temperature (oC)
Figure 4.9: CPT measurements scanning from an initial temperature of 4°C to 16°C.
69
0.6
0.4
0.2
0.0 PotentialV(SCE) -0.2
-0.4
1E-10 1E-9 1E-8 1E-7 1E-6 1E-5 1E-4 Current Density (A/cm2)
Figure 4.10: Potentiodynamic polarization of CanCr22 in 3.6M NaCl at -15°C.
Figure 4.11: 30 min OCP of the canary alloys in 0.6M NaCl at ~25°C performed with the hanging droplet cell.
70
CanCr22 0.4 CanCr14 CanCr10 CanCr6
0.2
) SCE 0.0
Potential(V -0.2
-0.4
1E-9 1E-8 1E-7 1E-6 1E-5 1E-4 1E-3 Current Density (A/cm2)
Figure 4.12: Typical potentiodynamic polarization curves of the canary alloys in 0.6 M NaCl at ~25°C performed with the hanging droplet cell.
71
Figure 4.13: Pit morphologies observed on canary alloys after polarization with use of the hanging droplet cell. a) Pits along the waterline; CanCr6 b) Small pits near the waterline; CanCr6 c) A large number of pits initiated near at the waterline edge; CanCr10 d) Multiple pits in close proximity near the waterline edge; CanCr10 e) Single large crystallographic pit; CanCr6 f) Single large crystallographic pit; CanCr10 g) Single large crystallographic pit; CanCr14 h) Single large crystallographic pit; CanCr22
72
0.6 CanCr22 CanCr14 0.4 CanCr10 CanCr6
0.2
) SCE
0.0 Potential(V -0.2
-0.4
1E-10 1E-9 1E-8 1E-7 1E-6 1E-5 1E-4 1E-3 Current Density (A/cm2)
Figure 4.14: Typical cyclic polarization curves of the canary alloys in 0.6 M NaCl at ~25°C performed with the hanging droplet cell.
0.6 Ecorr (V) Epit (V) 0.4 Erp (V)
0.2
) SCE
0.0 Potential(V -0.2
-0.4
4 6 8 10 12 14 16 18 20 22 24 Chrome Content (at.%)
Figure 4.15.: Graphical representation of chromium alloying effect on critical pitting corrosion metrics, Ecorr, Epit, and Erp, in Ni38Fe20Crx(MnCo)42-x.
73
0.8
0.6
0.4 )
SCE 0.2
0.0 Potential(V
-0.2
-0.4
1E-9 1E-8 1E-7 1E-6 1E-5 1E-4 Current Density (A/cm2)
Figure 4.16: Potentiodynamic polarization curve of CanCr22 at a surface temperature of -6°C in 0.6M NaCl performed with the hanging droplet cell.
Figure 4.17: OM images of A) lacy pit cover prior to sonication formed during polarization of CanCr22 at a surface temperature of -6°C in 0.6M NaCl B) Pit after sonication and cleaning.
74
References
[1] Yeh, J., Chen, S., Lin, S., Gan, J., Chin, T., Shun, T., . . . Chang, S. (2004). Nanostructured high‐entropy alloys with multiple principal elements: Novel alloy design concepts and outcomes. Advanced Engineering Materials, 6(5), 299-303. [2] Chen, Y., Duval, T., Hung, U., Yeh, J., & Shih, H. (2005). Microstructure and electrochemical properties of high entropy alloys—a comparison with type-304 stainless steel. Corrosion Science, 47(9), 2257-2279. [3] Kao, Y., Lee, T., Chen, S., & Chang, Y. (2010). Electrochemical passive properties of AlxCoCrFeNi (x=0, 0.25, 0.50, 1.00) alloys in sulfuric acids [4] Lee, C. P., Chang, C. C., Chen, Y. Y., Yeh, J. W., & Shih, H. C. (2008). Effect of the aluminium content of AlxCrFe1.5MnNi0.5 high-entropy alloys on the corrosion behaviour in aqueous environments [5] Hur, D. H., & Park, Y. (2006). Effect of temperature on the pitting behavior and passive film characteristics of alloy 600 in chloride solution. Corrosion, 62(9), 745-750 [6] Stellwag, B. (1997). Pitting resistance of alloy 800 as a function of temperature and prefilming in high-temperature water. Corrosion, 53(2), 120-128. [7] Kehler, B., Ilevbare, G., & Scully, J. (2001). Crevice corrosion stabilization and repassivation behavior of alloy 625 and alloy 22. Corrosion, 57(12), 1042-1065. [8] R.F.A. Jargelius-Pettersson, Application of the Pitting Resistance Equivalent Concept to Some Highly Alloyed Austenitic Stainless Steels, CORROSION, 54 (1998) 162-168. [9] K. Lorenz, G. Medawar, Über das Korrosionsverhalten austenitischer Chrom-Nickel- (Molybdän-) Stähle mit und ohne Stickstoffzusatz unter besonderer Berücksichtigung ihrer Beanspruchbarkeit in chloridhaltigen Lösungen, Thyssenforschung, 1 (1969) 97-108. [10] J.H. Cleland, What does the pitting resistance equivalent really tell us?, Engineering Failure Analysis, 3 (1996) 65-69. [11] G. Herbsleb, Der Einfluß von Schwefeldioxid, Schwefelwasserstoff und Kohlenmonoxid auf die Lochkorrosion von austenitischen Chrom-Nickel-Stählen mit bis zu 4 Massen-Molybdän in 1 M Natriumchlorid-Lösung, Materials and Corrosion, 33 (1982) 334-340. [12] H. Okamoto, The effect of tungsten and molybdenum on the performance of super duplex stainless steels, in: Proc. Conf. Application of Stainless Steels ‘92, 1992, pp. 360-369. [13] C.J. Kuehmann, G.B. Olson, Computational materials design and engineering, Materials Science and Technology, 25 (2009) 472-478. [14] Otto, F., Yang, Y., Bei, H., & George, E. P. (2013). Relative effects of enthalpy and entropy on the phase stability of equiatomic high-entropy alloys. Acta Materialia, 61(7), 2628-2638. [15] Stockert, L., Hunkeler, F., & Bohni, H. (1985). A crevice-free measurement technique to determine reproducible pitting potentials. Corrosion, 41(11), 676-677. [16] Stepanov, N. D., Shaysultanov, D. G., Yurchenko, N. Y., Zherebtsov, S. V., Ladygin, A. N., Salishchev, G. A., & Tikhonovsky, M. A. (2015). High temperature deformation behavior and dynamic recrystallization in CoCrFeNiMn high entropy alloy [17] Li, X., Zheng, Z., Dou, D., & Li, J. (2016). Microstructure and properties of coating of FeAlCuCrCoMn high entropy alloy deposited by direct current magnetron sputtering. Materials Research, (AHEAD), 0-0.
75
[18] Marcus, P. (1998). Surface science approach of corrosion phenomena. Electrochimica Acta, 43(1-2), 109-118. [19] Carranza, R. M., & Alvarez, M. G. (1996). The effect of temperature on the passive film properties and pitting behaviour of a FeCrNi alloy [20] Ernst, P., & Newman, R. C. (2007). Explanation of the effect of high chloride concentration on the critical pitting temperature of stainless steel [21] ASTM G150-13 Standard Test Method for Electrochemical Critical Pitting Temperature Testing of Stainless Steels, ASTM International, West Conshohocken, PA, 2013, [22] Newman, R. C., Foong Tuck Meng, , & Sieradzki, K. (1988). Validation of a percolation model for passivation of Fe-Cr alloys: I current efficiency in the incompletely passivated state [23] P. Crook and D. Klarstrom, Introduction to Alloys Resistant to Aqueous Corrosion, ASM International, Materials Park, OH (2005). [24] Yeh, J. (2013). Alloy design strategies and future trends in high-entropy alloys. Jom, 65(12), 1759-1771 [25] Zhang, Y., Zuo, T. T., Tang, Z., Gao, M. C., Dahmen, K. A., Liaw, P. K., & Lu, Z. P. (2014). Microstructures and properties of high-entropy alloys. Progress in Materials Science, 61, 1-93. [26] Frankel, G. S., Li, T., & Scully, J. R. (2017). Perspective—localized corrosion: Passive film breakdown vs pit growth stability. Journal of the Electrochemical Society, 164(4), C180-C181. [27] Frankel, G. (1998). Pitting corrosion of metals a review of the critical factors. Journal of the Electrochemical Society, 145(6), 2186-2198. [28] Laycock, N. J., & Newman, R. C. (1997). Localised dissolution kinetics, salt films and pitting potentials [29] A. Anderko, OLI, personal communication
76
Chapter 5 CONCLUSIONS AND FUTURE WORK 5.1 CONCLUSIONS Current understanding of HEA corrosion resistance do to chemical composition along with ICME was used to design corrosion resistant HEA1 as well as a series of corrosion susceptible alloys, here given the name Canary alloys, used to further the understanding of the effects of HEA composition and processing on corrosion resistance. The major results are as follows:
1) High corrosion resistant single-phased HEA1 was designed and successfully fabricated.
The alloy uses an uncommon non-equatomic high entropy composition with high
concentrations of elements which benefit corrosion resistance.
2) Under accelerated potentiodynamic testing of HEA1 in aggressive solutions, only
observations of transpassive corrosion were identified. Further, HEA1 does not
demonstrate pitting nature even in the most aggressive pit condition.
3) The high pitting resistance is derived from: (1) the continuous and protective passive film
which inhibits the pitting nucleation process; (2) the decreased pit growth stability caused
by a low imax.
4) The segregation of alloying elements in the dentritic and interdenritic regions of the as-
cast HEA1 sample do not decrease the pitting resistance, however, it accelerate the
transpassive dissolution. There was found to be only slight difference in corrosion
resistance between the as-cast and homogenized HEA1 samples.
5) Varying chromium alloys were designed and produced, known as Canary alloys.
Remarkably, the low chromium alloys are seen to be uniquely passive below the
conventional standards required for a stainless steel to be passive.
77
6) The polarization curves of the canary alloys show a great deal of metastable pitting
activity, which coincides with to the inability of the alloy to reach a critical concentration
of metal cations needed for formation of stable pits or crevices. Typically stable pits are
found below the transpassive potential due to the production of an aggressive chemistry
at the pit surface generated by the release of metal cations and subsequent influx of
hydrogen ions.
7) The novel microcell droplet cell has shown utility in its ability to prevent crevice
corrosion on highly-susceptible materials, allowing for greater understanding and future
research of alloy pitting on a variety of metal which currently take very complex
electrochemical cells to study.
8) The canary HEAs crystallographic pitting nature can be attributed to the inability of a
multiple element salts to form. This leads to the inability to generate stable pits without
large enough potentials to generate current densities greater than icrit.
5.2 FUTURE WORK
From information presented in chapter 4, alloys with low pit stabilization tendency should have weak pit covers, low pit dissolution rates, high critical concentration, and high saturation concentrations. It should be noted that the passive film has two opposing roles in pitting. A strong passive film is desired to minimize or eliminate passive film breakdown, causing pit initiation to be the rate controlling step. However, once initiated, a weak passive film is desired to reduce the robustness of the undermined passive film pit cover. The latter three aspects described above: pit dissolution rates, critical and saturation concentrations. Pit dissolution rates, Ccrit and Csat can be assessed by testing bulk samples in simulated pit solutions, 1-D pit experiments. Csat can also be addressed by OLI calculations. Further, we need to see if there is a correlation between imax/Csat and Epit/CPT. Future research work is as follows:
1) OLI point calculations for solutions having increasing concentration of cations at ratios equal to
78
alloy compositions; determine Csat or pit concentration for precipitation of first salt film.
2) Alloys to test given in Table 1: canary alloys, variations of CanCr14 with increasing Ni, Ni-Cr
and Fe-Cr binaries with Cr content equal to canaries, pure Fe and Ni. OLI calculation will be
performed on all alloy systems. The binary alloys and unitary metal are used for comparison. The
alloys series based on CanCr14 is used to determine if there is a threshold content of the primary
element when Csat is equal to that of the pure metal.
3) 1D pitting experiments for selected alloys based on OLI results to determine pit dissolution
kinetics DCcrit and DCsat. Assuming that D values for different metal cations are similar in
aqueous solution, the Ccrit and Csat can be determined.
4) Cyclic polarization testing of alloys, to see if there is a positive correlation between the Ccrit/Csat
and Epit, and a negative correlation between the imax and Epit
79
BIBLIOGRAPHY
A. Anderko, OLI, personal communication Arnvig, P.E., & Bisgard, A.D. (1996). Determining the potential independent critical pitting temperature (CPT) by a potentiostatic method using the Avesta Cell. United States: NACE International. ASTM G1-03(2017)e1 Standard Practice for Preparing, Cleaning, and Evaluating Corrosion Test Specimens, ASTM International, West Conshohocken, PA, 2017 ASTM G150-13 Standard Test Method for Electrochemical Critical Pitting Temperature Testing of Stainless Steels, ASTM International, West Conshohocken, PA, 2013. Carranza, R. M., & Alvarez, M. G. (1996). The effect of temperature on the passive film properties and pitting behaviour of a FeCrNi alloy Chen, Y., Duval, T., Hung, U., Yeh, J., & Shih, H. (2005). Microstructure and electrochemical properties of high entropy alloys—a comparison with type-304 stainless steel. Corrosion Science, 47(9), 2257-2279. Y.Y. Chen, T. Duval, U.D. Hung, J.W. Yeh, H.C. Shih, Microstructure and electrochemical properties of high entropy alloys—a comparison with type-304 stainless steel, Corrosion Science, 47 (2005) 2257-2279. Y.L. Chou, Y.C. Wang, J.W. Yeh, H.C. Shih, Pitting corrosion of the high-entropy alloy Co1.5CrFeNi1.5Ti0.5Mo0.1 in chloride-containing sulphate solutions, Corrosion Science, 52 (2010) 3481-3491. Y.L. Chou, J.W. Yeh, H.C. Shih, The effect of molybdenum on the corrosion behaviour of the high-entropy alloys Co1.5CrFeNi1.5Ti0.5Mox in aqueous environments, Corrosion Science, 52 (2010) 2571-2581. J.H. Cleland, What does the pitting resistance equivalent really tell us?, Engineering Failure Analysis, 3 (1996) 65-69.
Cost of Corrosion Study, NACE, 2014. P. Crook and D. Klarstrom, Introduction to Alloys Resistant to Aqueous Corrosion, ASM International, Materials Park, OH (2005). Ernst, P., & Newman, R. C. (2007). Explanation of the effect of high chloride concentration on the critical pitting temperature of stainless steel Frankel, G. (1998). Pitting corrosion of metals a review of the critical factors. Journal of the Electrochemical Society, 145(6), 2186-2198.
80
Frankel, G. S., Li, T., & Scully, J. R. (2017). Perspective—localized corrosion: Passive film breakdown vs pit growth stability. Journal of the Electrochemical Society, 164(4), C180-C181. Gao, M.C. Chapter 11: Design of high-entropy alloys. In High-Entropy Alloys: Fundamentals and Applications; Gao, M.C., Yeh, J.W., Liaw, P.K., Zhang, Y., Eds.; Springer: Cham, Switzerland, 2016. B. Gludovatz, A. Hohenwarter, D. Catoor, E.H. Chang, E.P. George, R.O. Ritchie, A fracture- resistant high-entropy alloy for cryogenic applications, Science, 345 (2014) 1153-1158. M.A. Hemphill, T. Yuan, G.Y. Wang, J.W. Yeh, C.W. Tsai, A. Chuang, P.K. Liaw, Fatigue behavior of Al0.5CoCrCuFeNi high entropy alloys, Acta Materialia, 60 (2012) 5723-5734. G. Herbsleb, Der Einfluß von Schwefeldioxid, Schwefelwasserstoff und Kohlenmonoxid auf die Lochkorrosion von austenitischen Chrom-Nickel-Stählen mit bis zu 4 Massen-Molybdän in 1 M Natriumchlorid-Lösung, Materials and Corrosion, 33 (1982) 334-340. Y.-J. Hsu, W.-C. Chiang, J.-K. Wu, Corrosion behavior of FeCoNiCrCux high-entropy alloys in 3.5% sodium chloride solution, Materials Chemistry and Physics, 92 (2005) 112-117. Hur, D. H., & Park, Y. (2006). Effect of temperature on the pitting behavior and passive film characteristics of alloy 600 in chloride solution. Corrosion, 62(9), 745-750 R.F.A. Jargelius-Pettersson, Application of the Pitting Resistance Equivalent Concept to Some Highly Alloyed Austenitic Stainless Steels, CORROSION, 54 (1998) 162-168.
Jones, D. A. (1996). Principles and prevention of corrosion prentice hall. Saddle River, NJ Kao, Y., Lee, T., Chen, S., & Chang, Y. (2010). Electrochemical passive properties of AlxCoCrFeNi (x=0, 0.25, 0.50, 1.00) alloys in sulfuric acids Kehler, B., Ilevbare, G., & Scully, J. (2001). Crevice corrosion stabilization and repassivation behavior of alloy 625 and alloy 22. Corrosion, 57(12), 1042-1065. G. Koch, J. Varney, N. Thompson, O. Moghissi, M. Gould, J. Payer, International Measures of Prevention, Application, and Economics of Corrosion Technologies Study, NACE, (2016). C.J. Kuehmann, G.B. Olson, Computational materials design and engineering, Materials Science and Technology, 25 (2009) 472-478. Laycock, N. J., & Newman, R. C. (1997). Localised dissolution kinetics, salt films and pitting potentials C.P. Lee, Y.Y. Chen, C.Y. Hsu, J.W. Yeh, H.C. Shih, The Effect of Boron on the Corrosion Resistance of the High Entropy Alloys Al0.5CoCrCuFeNiB x, Journal of The Electrochemical Society, 154 (2007) C424-C430. Lee, C. P., Chang, C. C., Chen, Y. Y., Yeh, J. W., & Shih, H. C. (2008). Effect of the aluminium content of AlxCrFe1.5MnNi0.5 high-entropy alloys on the corrosion behaviour in aqueous environments
81
Li, C., Li, J. C., Zhao, M., & Jiang, Q. (2009). Effect of alloying elements on microstructure and properties of multiprincipal elements high-entropy alloys Z. Li, K.G. Pradeep, Y. Deng, D. Raabe, C.C. Tasan, Metastable high-entropy dual-phase alloys overcome the strength–ductility trade-off, Nature, 534 (2016) 227. J. Li, X. Yang, R. Zhu, Y. Zhang, Corrosion and Serration Behaviors of TiZr0.5NbCr0.5VxMoy High Entropy Alloys in Aqueous Environments, Metals, 4 (2014) 597. Q.H. Li, T.M. Yue, Z.N. Guo, X. Lin, Microstructure and Corrosion Properties of AlCoCrFeNi High Entropy Alloy Coatings Deposited on AISI 1045 Steel by the Electrospark Process, Metallurgical and Materials Transactions A, 44 (2013) 1767-1778. Li, X., Zheng, Z., Dou, D., & Li, J. (2016). Microstructure and properties of coating of FeAlCuCrCoMn high entropy alloy deposited by direct current magnetron sputtering. Materials Research, (AHEAD), 0-0. K. Lorenz, G. Medawar, Über das Korrosionsverhalten austenitischer Chrom-Nickel-(Molybdän- ) Stähle mit und ohne Stickstoffzusatz unter besonderer Berücksichtigung ihrer Beanspruchbarkeit in chloridhaltigen Lösungen, Thyssenforschung, 1 (1969) 97-108. Marcus, P. (1998). Surface science approach of corrosion phenomena. Electrochimica Acta, 43(1-2), 109-118. Newman, R. C., Foong Tuck Meng, , & Sieradzki, K. (1988). Validation of a percolation model for passivation of Fe-Cr alloys: I current efficiency in the incompletely passivated state H. Okamoto, The effect of tungsten and molybdenum on the performance of super duplex stainless steels, in: Proc. Conf. Application of Stainless Steels ‘92, 1992, pp. 360-369. Otto, F., Yang, Y., Bei, H., & George, E. P. (2013). Relative effects of enthalpy and entropy on the phase stability of equiatomic high-entropy alloys. Acta Materialia, 61(7), 2628-2638. Park, C., Rao, V. S., & Kwon, H. (2005). Effects of sigma phase on the initiation and propagation of pitting corrosion of duplex stainless steel. Corrosion, 61(1), 76-83. Qiu, X., & Liu, C. (2013). Microstructure and properties of Al2CrFeCoCuTiNix high-entropy alloys prepared by laser cladding
V.M. Salinas-Bravo, R.C. Newman, Corrosion Science, 36 (1994) 67-77. L.J. Santodonato, Y. Zhang, M. Feygenson, C.M. Parish, M.C. Gao, R.J.K. Weber, J.C. Neuefeind, Z. Tang, P.K. Liaw, Deviation from high-entropy configurations in the atomic distributions of a multi-principal-element alloy, Nature Communications, 6 (2015) 5964. Schmuki, P. (2002). From bacon to barriers: A review on the passivity of metals and alloys. Journal of Solid State Electrochemistry, 6(3), 145-164. O.N. Senkov, S.V. Senkova, C. Woodward, Effect of aluminum on the microstructure and properties of two refractory high-entropy alloys, Acta Materialia, 68 (2014) 214-228.
82
Y. Shi, B. Yang, X. Xie, J. Brechtl, K.A. Dahmen, P.K. Liaw, Corrosion of Al xCoCrFeNi high- entropy alloys: Al-content and potential scan-rate dependent pitting behavior, Corrosion Science, 119 (2017) 33-45. Shi, Y., Yang, B., & Liaw, P. K. (2017). Corrosion-resistant high-entropy alloys: A review. Metals, 7(2), 43. Stellwag, B. (1997). Pitting resistance of alloy 800 as a function of temperature and prefilming in high-temperature water. Corrosion, 53(2), 120-128. Stepanov, N. D., Shaysultanov, D. G., Yurchenko, N. Y., Zherebtsov, S. V., Ladygin, A. N., Salishchev, G. A., & Tikhonovsky, M. A. (2015). High temperature deformation behavior and dynamic recrystallization in CoCrFeNiMn high entropy alloy Stockert, L., Hunkeler, F., & Bohni, H. (1985). A crevice-free measurement technique to determine reproducible pitting potentials. Corrosion, 41(11), 676-677. Z. Tang, T. Yuan, C.-W. Tsai, J.-W. Yeh, C.D. Lundin, P.K. Liaw, Fatigue behavior of a wrought Al0.5CoCrCuFeNi two-phase high-entropy alloy, Acta Materialia, 99 (2015) 247-258. Tsai, M., & Yeh, J. (2014). High-entropy alloys: A critical review. Materials Research Letters, 2(3), 107-123.U Yeh, J. (2013). Alloy design strategies and future trends in high-entropy alloys. Jom, 65(12), 1759-1771 J.W. Yeh, S.K. Chen, S.J. Lin, J.Y. Gan, T.S. Chin, T.T. Shun, C.H. Tsau, S.Y. Chang, Nanostructured High-Entropy Alloys with Multiple Principal Elements: Novel Alloy Design Concepts and Outcomes, Advanced Engineering Materials, 6 (2004) 299-303.Z Z. Zhang, M. Mao, J. Wang, B. Gludovatz, Z. Zhang, S.X. Mao, E.P. George, Q. Yu, R.O. Ritchie, Nanoscale origins of the damage tolerance of the high-entropy alloy CrMnFeCoNi, Nature communications, 6 (2015) 10143. Zhang, Y., Zuo, T. T., Tang, Z., Gao, M. C., Dahmen, K. A., Liaw, P. K., & Lu, Z. P. (2014). Microstructures and properties of high-entropy alloys. Progress in Materials Science, 61, 1-93.
83