Solid Oxide Electrochemical Systems: Material Degradation Processes and Novel Mitigation Approaches

materials

Review Solid Oxide Electrochemical Systems: Material Degradation Processes and Novel Mitigation Approaches

Michael Reisert, Ashish Aphale and Prabhakar Singh * Department of Materials Science and Engineering, University of Connecticut, Storrs, CT 06269, USA; [email protected] (M.R.); [email protected] (A.A.) * Correspondence: [email protected]; Tel.: +1-860-486-8379; Fax: +1-860-486-8378

Received: 2 August 2018; Accepted: 30 October 2018; Published: 2 November 2018

Abstract: Solid oxide electrochemical systems, such as solid oxide fuel cells (SOFC), solid oxide electrolysis cells (SOEC), and oxygen transport membranes (OTM) enable clean and reliable production of energy or fuel for a range of applications, including, but not limited to, residential, commercial, industrial, and grid-support. These systems utilize solid-state ceramic oxides which offer enhanced stability, fuel flexibility, and high energy conversion efficiency throughout operation. However, the nature of system conditions, such as high temperatures, complex redox atmosphere, and presence of volatile reactive species become taxing on solid oxide materials and limit their viability during long-term operation. Ongoing research efforts to identify the material corrosion and degradation phenomena, as well as discover possible mitigation techniques to extend material efficiency and longevity, is the current focus of the research and industrial community. In this review, degradation processes in select solid oxide electrochemical systems, system components, and comprising materials will be discussed. Overall degradation phenomena are presented and certain degradation mechanisms are discussed. State-of-the-art technologies to mitigate or minimize the above-mentioned degradation processes are presented.

Keywords: corrosion; electrode poisoning; solid oxide; interconnect; electrode; oxide scale

1. Introduction

1.1. Solid Oxide Electrochemical Systems There is an ever-growing push to utilize alternative, clean forms of energy and deviate from a dependency on nonrenewable fossil fuels. One way this has been achieved is through the exploitation of electrochemical devices, which convert the chemical energy from fuels into electricity or vis-versa without any combustion of the fuels [1]. Some systems are currently integrated in electrical grids and used in automotive/aerospace applications as liquid or polymer electrolytes. However, a different variety is comprised of a solid metal oxide electrolyte and electrodes. These solid oxide electrochemical systems are favorable as they provide high energy conversion efficiencies, flexibility in design, and flexibility in fuel choice [2–4]. This review will discuss three particular solid oxide electrochemical systems: Solid oxide fuel cells, solid oxide electrolysis cells, and oxygen transport membranes, as well as related degradation processes associated with them. To achieve wide-scale industrial use of solid oxide electrochemical devices, the materials that comprise them and any material degradation processes must be well understood. Major degradation phenomena can occur at every system component. In this review, the solid–gas interactions which play a large role in system performance degradation will be discussed [5,6]. This includes poisoning of both fuel and air electrodes by volatilized species, as well as oxidation and

Materials 2018, 11, 2169; doi:10.3390/ma11112169 www.mdpi.com/journal/materials Materials 2018, 11, 2169 2 of 16 metal loss, resulting from simultaneous exposure of metallic components to different gaseous species. KnowledgeMaterials 2018 of these, 11, x FOR phenomena PEER REVIEW and a review of the current state-of-the-art technology and2 of 15 novel mitigation approaches will be useful in prolonging material lifespans for efficient operation of solid oxidation and metal loss, resulting from simultaneous exposure of metallic components to different oxidegaseous electrochemical species. Knowledge systems. of these phenomena and a review of the current state-of-the-art technology and novel mitigation approaches will be useful in prolonging material lifespans for 1.1.1.efficient Solid Oxide operation Fuel of Cells solid oxide electrochemical systems. Fuel cells are open systems that can be continuously fueled. This makes fuel cells optimal for 1.1.1. Solid Oxide Fuel Cells grid applications, as they can be intermittently refueled to continuously provide electricity with no system replacementFuel cells are necessary open systems [7,8]. that They can operate be continuously more efficiently fueled. This than makes thermomechanical fuel cells optimal means for of energygrid production applications, as directas they energy can be conversionintermittently eliminates refueled to the continuously need for combustion provide electricity [4]. Fuel with cells no have beensystem designed replacement with various necessary materials, [7,8]. They specifically operate electrolytemore efficiently materials, than thermomechanical to yield a variety means of types. of A energy production as direct energy conversion eliminates the need for combustion [4]. Fuel cells have proton exchange membrane fuel cell (PEMFC), for example, utilizes a polymer electrolyte such as been designed with various materials, specifically electrolyte materials, to yield a variety of types. A Nafion® to foster protonic movement. A platinum catalyst is used as an anode to split the hydrogen proton exchange membrane fuel cell (PEMFC), for example, utilizes a polymer electrolyte such as fuel intoNafion protons® to foster and protonic electrons. movement. The electrons A platinum are forced catalyst to is anused external as an anode circuit to duesplit tothe the hydrogen electrically insulatingfuel into properties protons ofand the electrons. electrolyte, The electrons while the are protons forced areto an conducted external circuit through due the to the electrolyte electrically toward the cathode.insulating At properties the cathode, of the the el protonsectrolyte, are while reunited the protons with electronsare conducted that havethrough travelled the electrolyte the external circuit,toward as well the ascathode. oxygen At fromthe cathode, an oxidizing the protons gas flownare reunited to the with cathode. electrons This that results have travelled in the chemical the formationexternal of watercircuit, vapor, as well which as oxygen is filtered from outan oxidiz of theing cell gas as flown waste. to This the electrochemistrycathode. This results is thein the driving forcechemical behind fuelformation cell operation of water vapor, and has which been is adheredfiltered out to of in the developing cell as waste. new This types electrochemistry of fuel cells. is Onethe driving promising force varietybehind fuel of fuel cell celloperation is the and solid has oxide been fueladhered cell to (SOFC), in developing which new uses types a solid of fuel ceramic cells. electrolyte. This is advantageous as the ceramic electrolyte is very stable and offers a long operating One promising variety of fuel cell is the solid oxide fuel cell (SOFC), which uses a solid ceramic lifetime, whereas polymer electrolytes can dry out or flood if they are not hydrated in the precise electrolyte. This is advantageous as the ceramic electrolyte is very stable and offers a long operating amount,lifetime, at which whereas point polymer they lose electrolytes efficiency can or dry stop out working or flood altogetherif they are [not9]. Theyhydrated can in also the operate precise on a rangeamount, of hydrogen-based at which point fuels they likelose hydrocarbons,efficiency or stop whereas working PEMaltogether fuel cells[9]. They must can use also a pureoperate hydrogen on fuel sourcea range [ 4of]. hydrogen-based Solid oxide fuel fuels cells like operate hydrocarbons, at high temperatures whereas PEM byfuel reducing cells must an use oxidizing a pure hydrogen gas (usually air) atfuel the source cathode [4]. into Solid oxygen oxide fuel ions. cells Unlike operate PEM at fuelhigh cells, temperatures which reduce by reducing a fuel andan oxidizing move ions gas from anode(usually to cathode, air) at a the SOFC cathode moves into oxygen oxygen ions ions. from Unlike cathode PEM fuel toanode, cells, which where reduce they a meet fuel hydrogenand move ions to formions water from vapor anode as to wastecathode, and a releaseSOFC moves an electron oxygen to ions an external from cathode circuit. to Thisanode, is becausewhere they of themeet nature of thehydrogen solid electrolyte, ions to form which water is vapor typically as waste an oxygen-ion and release conducting an electron to yttria-stabilized an external circuit. zirconia This is (YSZ) because of the nature of the solid electrolyte, which is typically an oxygen-ion conducting yttria- ceramic [2,3]. The cathode, also known as the air electrode (AE), utilizes electronically conducting stabilized zirconia (YSZ) ceramic [2,3]. The cathode, also known as the air electrode (AE), utilizes or mixed ionically and electronically conducting, oxygen-reducing ceramics, which have a thermal electronically conducting or mixed ionically and electronically conducting, oxygen-reducing expansionceramics, coefficient which have that a closelythermal matchesexpansion the coefficient electrolyte that toclosel reducey matches thermal the stresseselectrolyte under to reduce dynamic operatingthermal conditions. stresses under A comparative dynamic operating schematic condit of theions. operating A comparative principles schematic of a SOFC of the and operating a PEMFC is picturedprinciples below of in a FigureSOFC and1. a PEMFC is pictured below in Figure 1.

FigureFigure 1. Comparative 1. Comparative operative operative schematic schematic ofof a solid ox oxideide fuel fuel cell cell (SOFC) (SOFC) and and proton proton exchange exchange membranemembrane fuel fuel cell cell (PEMFC). (PEMFC).

Materials 2018, 11, 2169 3 of 16

Materials 2018, 11, x FOR PEER REVIEW 3 of 15 While typically operating on the principle of oxygen ion conduction, some SOFC systems can produce energyWhile typically through operating proton conductionon the principle similar of oxygen to that ion shown conduction, in the some PEMFC SOFC operation systems can [10 –13]. The materialsproduce energy which through enable thisproton are conduction discussed similar later in to this that review. shown in the PEMFC operation [10–13]. The materials which enable this are discussed later in this review. 1.1.2. Solid Oxide Electrolysis Cells 1.1.2. Solid Oxide Electrolysis Cells The governing principles of electrochemistry that allow fuel cells to efﬁciently produce electricity can also beThe used governing to produce principles fuel orof oxygenelectrochemistry by means that of electrolysis.allow fuel cells Electrolysis to efficiently is the produce splitting of gaseouselectricity species, can such also as be water used vaporto produce or carbon fuel or dioxide, oxygen intoby means their constituentsof electrolysis. through Electrolysis the application is the splitting of gaseous species, such as water vapor or carbon dioxide, into their constituents through of an external voltage. Essentially, a fuel cell can be operated in reverse to create what is called an the application of an external voltage. Essentially, a fuel cell can be operated in reverse to create what electrolysis cell [14]. The reactions that occur within an electrolysis cell are non-spontaneous redox is called an electrolysis cell [14]. The reactions that occur within an electrolysis cell are non- reactionsspontaneous and require redox electrical reactions energyand require in the electrical form of energy an applied in the voltage form of to an proceed. applied Therefore,voltage to the functionproceed. of these Therefore, systems the function is to convert of these electrical systems is energy to convert to chemical electrical energy energy. to Inchemical the case energy. of water electrolysis,In the case which of water is carried electrolysis, out to which create is hydrogencarried out for to fuel,create bothhydrogen polymer for fuel, exchange both polymer membrane electrolysisexchange cells membrane (PEMEC) electrolysis and solid cells oxide (PEMEC) electrolysis and solid cells oxide (SOEC) electrolysis can cells be used (SOEC) [15 can]. In be a used PEMEC, water[15]. is split In a PEMEC, at the anode water inis split what at isthe called anode an in oxygenwhat is called evolution an oxygen reaction evolution (OER). reaction It is oxidized (OER). It into oxygenis oxidized gas, protons, into oxygen and electrons. gas, protons, The and cathode electrons. undergoes The cathode a hydrogen undergoes evolution a hydrogen reaction evolution (HER), wherereaction the protons (HER), from where the the anode protons travel from through the anod thee electrolytetravel through and the meet electrolyte the supplied and meet electrons the to supplied electrons to create hydrogen gas [15]. For the SOEC, the electrolyte is oxygen conducting, create hydrogen gas [15]. For the SOEC, the electrolyte is oxygen conducting, as it is in the solid as it is in the solid oxide fuel cell. Therefore, water is split at the cathode and oxygen ions travel to oxide fuel cell. Therefore, water is split at the cathode and oxygen ions travel to the anode where they the anode where they release electrons [14,15]. The supply of electrons to the cathode deprotonates release electrons [14,15]. The supply of electrons to the cathode deprotonates the hydrogen and creates the hydrogen and creates hydrogen gas. The co-electrolysis of CO2 and H2O using SOEC technology hydrogenis a promising gas. The means co-electrolysis of hydrocarbon of CO fuel2 and production, H2O using while SOEC recycling technology greenhouse is a promisinggas emissions means of hydrocarbon[14,16]. A comparative fuel production, schematic while of the recycling operating principles greenhouse of agas SOEC emissions and PEMEC [14 ,is16 pictured]. A comparative below schematicin Figure of the 2. operating principles of a SOEC and PEMEC is pictured below in Figure2.

FigureFigure 2. Comparative 2. Comparative schematics schematics of of a a polymer polymer exchangeexchange membrane membrane electrolysis electrolysis cellcell (PEMEC) (PEMEC) and a and a solidsolid oxide oxide electrolysis electrolysis cell cell (SOEC) (SOEC) (Based (Based on on schematics schematics from [15]). [15]).

1.1.3.1.1.3. Oxygen Oxygen Transport Transport Membranes Membranes OxygenOxygen transport transport membranes membranes (OTM) (OTM) are are usedused in separating high-purity high-purity oxygen oxygen from from air. This air. This is usedis used in the in the oxy-combustion oxy-combustion process,process, where where hi highgh purity purity oxygen oxygen is used is usedto ignite to fuel. ignite The fuel. oxy- The oxy-combustioncombustion process process is more is more efficient efﬁcient and cleaner and cleaner than the than standard the standard means of using means air of as usingan oxidant. air as an oxidant.Solid Solid oxide oxide membranes membranes can be caneither be ionic either conducto ionic conductorsrs or mixed ionic or mixed electronic ionic conductors electronic (MIEC), conductors (MIEC),as shown as shown in Figure in Figure 3. Ionic3. Ionicconductors conductors are called are passive called passivemembranes membranes as they are as electrically they are electricallydriven by an outside source [17]. The MIEC membrane is called an active membrane and it relies on the driven by an outside source [17]. The MIEC membrane is called an active membrane and it relies on difference of oxygen partial pressure (PO2) on either side of the membrane [17]. Oxygen-rich air yields the difference of oxygen partial pressure (PO ) on either side of the membrane [17]. Oxygen-rich air a high PO2 on one side of the membrane. A natural2 gas flown on the opposing side results in a much yieldslower a high POP2,O and2 on thermodynamically, one side of the membrane. oxygen ions A are natural driven gas toward ﬂown the on lower the opposingPO2. side results in a much lower PO2, and thermodynamically, oxygen ions are driven toward the lower PO2.

Materials 2018, 11, 2169 4 of 16 Materials 2018, 11, x FOR PEER REVIEW 4 of 15

Materials 2018, 11, x FOR PEER REVIEW 4 of 15

FigureFigure 3. Comparative 3. Comparative schematic schematic of: of: ( a()a) a a mixed mixed ionic-electronicionic-electronic conductor conductor (MIEC) (MIEC) and and(b) an (b )ionic an ionic conductor,conductor, both both used used as oxygenas oxygen transport transport membranes,membranes, adapted adapted from from [17], [17 with], with permission permission from from Elsevier,Elsevier, 2018. 2018. Figure 3. Comparative schematic of: (a) a mixed ionic-electronic conductor (MIEC) and (b) an ionic 1.2. Device1.2. Device and Systemand System Materials Materials conductor, both used as oxygen transport membranes, adapted from [17], with permission from SolidElsevier,Solid oxide oxide 2018. electrochemical electrochemical systems systems areare mostmost often comprised comprised of ofthree three major major components; components; an an anode, a cathode, and an electrolyte. In the case of OTM’s, configurations with these three anode, a cathode, and an electrolyte. In the case of OTM’s, configurations with these three components 1.2.components Device and or System one single Materials membrane acting as an electrolyte are possible. System components work or one single membrane acting as an electrolyte are possible. System components work harmoniously harmoniously during chemical reactions to provide the desired end goal of either electrical energy or Solid oxide electrochemical systems are most often comprised of three major components; an duringpure chemical gaseous reactions species. The to providemetals and the metal desired oxides end used goal for of eitherthese system electrical components energyor must pure fulfill gaseous anode, a cathode, and an electrolyte. In the case of OTM’s, configurations with these three species.several The criteria metals for and efficient metal oxidescell operation, used for such these as systemconductivity components of either mustions, electrons, fulfill several or both criteria components or one single membrane acting as an electrolyte are possible. System components work for efficientspecies, cellmatching operation, coefficients such of as thermal conductivity expansion, of eitherand strong ions, catalytic electrons, activity. or both species, matching harmoniously during chemical reactions to provide the desired end goal of either electrical energy or coefficientspure gaseous of thermal species. expansion, The metals and and strong metal catalyticoxides used activity. for these system components must fulfill 1.2.1. Electrolytes several criteria for efficient cell operation, such as conductivity of either ions, electrons, or both 1.2.1. Electrolytes species,The matching most important coefficients electrochemical of thermal expansion, system componen and strongt is the catalytic electrolyte. activity. In solid oxide fuel and Theelectrolysis most importantcells, ceramic electrochemical metal oxides are system used to component conduct either is the oxygen electrolyte. ions or protons. In solid Yttria- oxide fuel and1.2.1. electrolysisstabilized Electrolytes zirconia cells, (YSZ) ceramic is valued metal as oxides an oxyge aren ion used electrolyte to conduct material, either and oxygenis commonly ions used or protons. in high-temperature applications [2–4]. This value is due to the ceramic’s high oxygen ion conductivity, Yttria-stabilizedThe most zirconia important (YSZ) electrochemical is valued system as an componen oxygen iont is electrolytethe electrolyte. material, In solid oxide and isfuel commonly and diffusivity, and thermal stability. These properties all arise from the cubic fluorite structure usedelectrolysis in high-temperature cells, ceramic applications metal oxides [are2–4 ].used This to conduct value is either due tooxygen the ceramic’sions or protons. high oxygenYttria- ion developed when yttria (Y2O3) is used as a dopant within the zirconia (ZrO2) structure. The cubic phase stabilized zirconia (YSZ) is valued as an oxygen ion electrolyte material, and is commonly used in conductivity,of zirconia diffusivity, is not stable and at room thermal temperature stability. because These Zr properties4+ ions are too all small arise to from stabilize the this cubic phase, fluorite high-temperature applications [2–4]. This value is due to the ceramic’s high oxygen ion conductivity, structureand instead developed the monoclinic when yttria phase (Y2 Ois 3observed.) is used asHoweve a dopantr, doping within the the structure zirconia with (ZrO a slightly2) structure. larger The diffusivity, and thermal stability. These properties all arise from4+ the cubic fluorite structure cubiccation, phase most of zirconia often yttrium is not stable ions (Y at3+ room), promotes temperature the formation because of the Zr cubicions fluorite are too structure small to at stabilize much this developed when yttria (Y2O3) is used as a dopant within the zirconia (ZrO2) structure. The cubic phase phase,lower and temperatures instead the monoclinicdown to room phase temperature. is observed. This doping However, creates doping oxygen the vacancies structure which with allow a slightly of zirconia is not stable at room temperature because Zr4+ ions are too small to stabilize this phase, for transport of oxygen ions, following 3+the Kröger-Vink reaction in Equation (1) [18]: largerand cation, instead most the oftenmonoclinic yttrium phase ions is observed. (Y ), promotes Howeve ther, doping formation the structure of the cubic with a fluorite slightly structure larger at 3+ muchcation, lower most temperatures often yttrium down ions to(Y room), promotesY2O temperature.3 → 2Y the′Zr +formation 3Ox ThisO + V ∙∙ dopingofO the cubic creates fluorite oxygen structure vacancies at much(1) which allowlower for transport temperatures of oxygen down to ions, room following temperature. the Kröger-VinkThis doping creates reaction oxygen in Equation vacancies (1) which [18]: allow A pictorial representation of the defect structure of YSZ is shown below in Figure 4. for transport of oxygen ions, following the Kröger-Vink reaction in Equation (1) [18]: 0 x ·· Y2O3 → 2Y Zr + 3O O + V O (1) Y2O3 → 2Y′Zr + 3OxO + V∙∙O (1) A pictorialA pictorial representation representation of of the the defect defect structurestructure of of YSZ YSZ is is shown shown below below in Figure in Figure 4. 4.

Figure 4. Cubic structure of yttria-stabilized zirconia (YSZ).

FigureFigure 4. 4.Cubic Cubic structure structure of yttria-stabilized zirconia zirconia (YSZ). (YSZ).

Materials 2018, 11, 2169 5 of 16

Other oxygen-conducting ceramics used as electrolytes include doped cerium oxides like samarium doped ceria (CSO or SDC) and gadolinia-doped ceria (CGO or GDC), which offer high oxygen ion conductivities at intermediate temperatures (500–700 ◦C) [4]. However, they are not suited for high temperatures as reducing atmosphere at elevated temperature causes a partial reduction of ceria to Ce3+, which makes the electrolyte electronically conductive and greatly reduces efﬁciency [19]. For oxygen transport membranes, mixed electronic-ionic conductors like many lanthanum-based perovskites are used as electrolytes, as they allow pure oxygen ion conduction driven by an oxygen partial pressure (PO2) difference [17]. These perovskites, often used as electrode materials, will be further described in the following section. The passive or electrically driven OTM uses a pure ionic conductor like YSZ and operates much like a SOEC. Proton-conducting SOFC/SOEC systems typically use barium-based perovskites of the structure Ba(Ce,Y)O3-δ (BCY) as electrolyte materials [10,12,13]. Typical dopants of this structure include Sr, Zr, Yb, and Fe [13]. These dopants can increase protonic conductivity and stability for implementation within proton-conducting SOFC/SOEC systems [20].

1.2.2. Electrodes SOFC and SOEC electrodes refer to the anode, or fuel electrode, and cathode, or air electrode. These electrodes are the sites of certain reactions which allow ionic transfer through the electrolyte. Both electrode types must maintain favorable reactivity at the triple-phase boundary (TPB) of the electrode, electrolyte, and gaseous species [21,22]. The cathode must have adequate porosity to allow gaseous oxygen to diffuse toward the electrolyte, as well as high electronic conductivity, to allow reduction of oxygen [21]. Cathode materials are often perovskite-type oxides of the general formula ABO3 [21,23]. The A site cation can be a mix of rare and alkaline earth metals, whilst the B site is a transition metal which enables catalysis for the redox (reduction-oxidation) reaction at the cathode [21]. Therefore, doping of these cation sites can enable better electronic conductivity and electrocatalytic properties [21]. Many variations of lanthanum-based perovskites exhibit good cathodic properties. These lanthanum-based oxides are often doped at the A site with Sr, which produces increased electron-hole concentrations and reduces unwanted reactivity of La with electrolyte materials [21]. They are further classified by the B site dopant, which can include Mn, Fe, or Co. In conventional SOFC systems, the fuel electrode must catalyze the reaction between fuel and oxygen ions from the electrolyte, whilst fostering electronic transfer to the external circuit [24]. Platinum has favorable catalytic properties, which warranted its use and integration into both cathodes and anodes for various solid oxide electrochemical systems [4,13,24]. However, the extremely high cost of platinum has spurred efforts to develop cheaper electrode alternatives [4,24]. Furthermore, nickel metal as an anode material offers lower polarization resistance and a cheaper price compared to platinum [2]. Nickel mixed with the electrolyte material (Ni-YSZ or Ni-BCY) is used to create a porous mixed ionic-electronic conducting cermet that efficiently allows fuel flow toward the electrolyte [2,24]. Since Ni is susceptible to carbide formation, other material choices include composite cermets based on Cu and CGO, which have been considered and show promise in carbonaceous atmospheres [22,24]. Some perovskite-structured ceramics are also used as anode materials because of their mixed conductivities [24].

1.2.3. Interconnects/Sealants/Balance-of-Plant For proper assembly and gas flow for multiple fuel and electrolysis cell stacks, structural components known as interconnects are relied on. At elevated operating temperatures (800+ ◦C), LaCrO3 ceramics were the primary interconnect material choice [2,3,25,26]. However, advancements in electrolyte materials and dimensions have allowed solid oxide fuel cell operating temperatures to fall to an intermediate range of 600–800 ◦C[27]. This has opened a window for new materials to be used for interconnects, specifically cheaper and more easily manufactured metal alloys. The ceramic components of a solid oxide fuel cell have coefficients of thermal expansion of around Materials 2018, 11, 2169 6 of 16

10−12 × 10−6 K−1 [28]. To ensure the cell does not fail under thermal stress, the interconnect materials contactingMaterials the 2018 ceramic, 11, x FOR components PEER REVIEW must match or nearly-match the thermal expansion of the6 ceramics.of 15 Interconnects are also relied on to assist in the circuit, which ultimately yields the fuel cells electrical voltageceramic output. components For this, interconnects must match must or nearly-mat be conductivech the and thermal resist any expansion ohmic losses of the during ceramics. operation. To deviateInterconnects from the are expensive also relied ceramicon to assist lanthanum-chromite in the circuit, which ultimately interconnects yields that the havefuel cells previously electrical been voltage output. For this, interconnects must be conductive and resist any ohmic losses during used, certain iron, chromium, and nickel alloys have been more easily and more cost-effectively operation. To deviate from the expensive ceramic lanthanum-chromite interconnects that have manufacturedpreviously tobeen meet used, these certain requirements. iron, chromium, The and development nickel alloys have of these been materialsmore easily specifically and more cost- for use as interconnectseffectively manufactured has drawn to a lotmeet of these attention requirements and inspired. The development many research of these efforts materials to maximizespecifically their efficiencyfor use and as interconnects compatibility. has drawn a lot of attention and inspired many research efforts to maximize Interconnectstheir efficiency andand othercompatibility. cell components must be hermetically sealed to ensure no fuel or oxidant loss duringInterconnects operation. and In other planar cell configurations components must of be SOFC/SOEC/OTM hermetically sealed to systems, ensure no the fuel outer or oxidant edges of interconnects,loss during electrodes, operation. andIn planar electrolytes, configurations as well of as SOFC/SOEC/OTM between individual systems, cells, arethe sealedouter edges to bond of the entireinterconnects, stack [29]. State-of-the-art electrodes, and electrolytes, seal types as include well as rigidlybetween bonded individual silica cells, glass are sealed seals to which bond providethe hermeticentire sealing, stack [29]. flexible State-of-the-art design integration, seal types cheapinclude costs, rigidly and bonded tailored silica properties glass seals [29 which]. provide hermetic sealing, flexible design integration, cheap costs, and tailored properties [29]. Balance-of-plant (BOP) materials include external system components, which mix, heat, reform, Balance-of-plant (BOP) materials include external system components, which mix, heat, reform, and flowand flow gases gases to the to the cell cell stack stack [30 [30].]. These These components components are are needed needed for for all all solid solid oxide oxide electrochemical electrochemical systems,systems, as they as they allow allow properly properly heated heated gases gasesto to infiltrateinfiltrate respective respective cell cell stack stack components components and andensure ensure properproper cell operation. cell operation. A flow A flow chart chart highlighting highlighting the the components components of BOP with with respect respect to to the the cell cell stack stack is shownis shown below, below, in Figure in Figure5. 5.

Figure 5. Flow chart of balance-of-plant (BOP) components/system operation (based on schematic Figure 5. Flow chart of balance-of-plant (BOP) components/system operation (based on schematic from Reference [30]). from Reference [30]). 2. Material Corrosion/Degradation Phenomena 2. Material Corrosion/Degradation Phenomena Owing to the complexity of solid oxide electrochemical system materials and operating conditions, Owing to the complexity of solid oxide electrochemical system materials and operating a varietyconditions, of degradation a variety processesof degradation continue processes to plague continue some to plague full-ﬂedged some full-fledged grid integrations grid integrations and industrial uses.and These industrial can occur uses. asThese solid–solid, can occur solid–gas,as solid–solid, or solid–liquidsolid–gas, or solid–liquid degradation degradation reactions, reactions, as shown in Figureas6 .shown As previously in Figure 6. mentioned, As previously the mentioned, interactions the betweeninteractions solid between materials solid andmaterials the oxidizing and the or reducingoxidizing gases or in reducing the system gases are in extremely the system problematic. are extremely Two problematic. detrimental Two gas/solid detrimental reactions gas/solid are those that occurreactions at active are those areas that in occur the cell, at active and those areas thatin the lead cell, to and the those breakdown that lead of to structural the breakdown components. of Thesestructural degradation components. phenomena These will degradation be discussed phenomena in detail. will be discussed in detail.

Materials 2018, 11, 2169 7 of 16 Materials 2018, 11, x FOR PEER REVIEW 7 of 15 Materials 2018, 11, x FOR PEER REVIEW 7 of 15

FigureFigure 6. Schematic 6. Schematic representation representation of of thethe overalloverall degrad degradationation processes processes in (solid in (solid oxide oxide fuel fuelcells) cells) SOFC/SOEC systems. SOFC/SOECFigure 6. Schematic systems. representation of the overall degradation processes in (solid oxide fuel cells) SOFC/SOEC systems. 2.1. Electrode2.1. Electrode Poisoning Poisoning and and Degradation Degradation An2.1. Electrode ongoingAn ongoing Poisoning issue issue that andthat hasDegradation has plagued plagued the viability viability of ofmany many solid-state solid-state electrochemical electrochemical devices devices is is thethe poisoning poisoningAn ongoing of of electrodeissue electrode that species has species plagued by volatile by the volatile viability and atmospheric and of many atmospheric solid-state species. These species.electrochemical species These are devicesvolatized species is are volatizedfromthe poisoning frominterconnect interconnect of electrode and sealant and species sealant materials by volatile materials at antypicald at atmospheric typical operating operating species. temperatures temperaturesThese species and durationsare and volatized durations of of SOFC/SOEC/OTMSOFC/SOEC/OTMfrom interconnect systems systemsand sealant [ 5[5,17].,17]. materials at typical operating temperatures and durations of SOFC/SOEC/OTM systems [5,17]. 2.1.1.2.1.1. Sulfur Sulfur 2.1.1. Sulfur Sulfuric impurities in hydrocarbon fuels such as H2S degrade Ni-YSZ and Ni-BZCYYb (Ni-BCY Sulfuric impurities in hydrocarbon fuels such as H2S degrade Ni-YSZ and Ni-BZCYYb (Ni-BCY doped with Zr, Yb) electrodes by forming nickel sulfide, which forms as large particles that reduce doped withSulfuric Zr, Yb) impurities electrodes in hydrocarbon by forming nickelfuels such sulﬁde, as H2S which degrade forms Ni-YSZ as large and Ni-BZCYYb particles that (Ni-BCY reduce the thedoped triple with phase Zr, Yb)boundary electrodes where by formingactive ionic nickel specie sulfide,s transfer which at forms the electrode/electrolyte as large particles that interface reduce triple phase boundary where active ionic species transfer at the electrode/electrolyte interface [31,32]. [31,32].the triple This phase degradation boundary may where also occuractive dueionic to speciesulfurs chemisorption transfer at the on electrode/electrolyte the electrode and saturation interface This degradation may also occur due to sulfur chemisorption on the electrode and saturation of the of[31,32]. the electrode, This degradation which can may also also reduce occur active due tosites su lfur[33]. chemisorption In these poisoning on the conditions, electrode andnickel saturation has also electrode,beenof the shown whichelectrode, to can oxidize, which also reducecanwhich also leads activereduce to siteslarger active [ 33polari sites]. In [33].zation these In andthese poisoning cell poisoning voltage conditions, drop,conditions, greatly nickel nickel reducing has has also also the been shownnickelbeen to oxidize,shown catalyst’s to whichoxidize, efficacy leads which (Figure to leads larger 7) [33]. to polarizationlarger polarization and cell and voltage cell voltage drop, drop, greatly greatly reducing reducing the the nickel catalyst’snickel efﬁcacy catalyst’s (Figure efficacy7)[ (Figure33]. 7) [33].

Figure 7. Mechanisms of sulfur poisoning in: (a) Low partial pressure of hydrogen sulfide (PH2S) fuel

FigurewhereFigure 7. Mechanisms sulfur7. Mechanisms adsorption of sulfurof sulfurdecreases poisoning poisoning some in:active in: ((aa)) sites LowLow bypartial partial forming pressure pressure NiS; ofand ofhydrogen hydrogen(b) High sulfide PH sulﬁde2S (fuelPH (2whereS)PH fuel2S) fuel wherenickelwhere sulfur insulfur the adsorption anode adsorption oxidizes. decreases decreases some some active active sitessites by forming forming NiS; NiS; and and (b) ( bHigh) High PHP2SH fuel2S fuelwhere where nickelnickel in the in anodethe anode oxidizes. oxidizes.

Sulfur can also exist as an air impurity in the form of SO2, and therefore, affect the cathode air stream [34]. At the lanthanum strontium cobalt ferrite (LSCF) air electrode, SO2 impurities in the oxidant react with strontium to form SrSO4 deposits on the electrode surface, which prevent oxygen Materials 2018, 11, 2169 8 of 16 diffusion [34]. This has resulted in research and approaches to prevent sulfurous contamination of the cell.

2.1.2. Chromium Chromium has been shown to volatilize (evaporate) from chromium-containing ferritic steels used for interconnects and in BOP components [35]. Over prolonged operation in a SOFC/SOEC, ferritic steels which form a chromia scale can lose chromium through volatilization (evaporation). Furthermore, chromium species can diffuse through the triple phase boundary and deposit on the air electrode, limiting the oxygen reduction reaction at the air electrode. The chromium species can also react with electrode constituents and create complex oxides, which has been shown to cause cell degradation [36–38]. In systems with strontium and manganese, such as LSM cathodes, gaseous chromium species can cause the growth of solid SrCrO4 and spinel-type (Cr,Mn)3O4, which also reduce electrochemical activity by building on active boundary area [35]. Jiang et al. found that deposition of chromium occurred on the LSCF electrode used, while in an interconnect/LSM/YSZ cell, chromium deposited on the electrolyte surface [39]. Each area of deposition was found to reduce cell efﬁciency by blocking the active area for oxygen reduction at the electrolyte/electrode interphase. In humid air, CrO2(OH)2 is the most abundant chromium gaseous species [40]. This gaseous species forms in the following reaction process:

3 Cr O + O + 2H O → 2CrO (OH) (2) 2 3 (s) 2 2 (g) 2 (g) 2 2 (g) Kurokawa et al. found that the enthalpy of reaction favored this species under the given conditions, which matched previous works and proved that this gaseous chromium species is most abundant and to blame for electrode poisoning [40]. This gaseous species reduces at the triple phase boundary and forms solid Cr2O3, thereby reducing the active area at the boundary and preventing oxygen reduction [35].

2.1.3. Silicon Silicon impurities are a result of volatilization from glass sealants used to hermetically seal cell stacks. These seals are used in many SOFC/SOEC/OTM systems as separation of gaseous species and leak-proof stacks are critical for efﬁcient system operations. Volatile Si species can deposit at active triple phase boundary sites of the Ni-YSZ electrode/YSZ electrolyte interface as silica (SiO2) in SOFC and SOEC systems, and at electrocatalytically active sites in OTM systems [5,17,29,41,42]. Glassy-phase solid SiO2 deposits with the equilibration of the following equation, as in References [5,42]:

Si(OH)4(g) ↔ SiO2(s) + 2H2O (g) (3)

Gaseous silicon hydroxide (Si(OH)4) forms SiO2 deposits and steam at the cell areas where steam is reduced to hydrogen. This is most prominent where the hydrogen electrode is closest to the electrolyte, leading to high concentrations of deposited SiO2 at the electrode/electrolyte interface which block electrocatalytic sites [5,42]. SiO2 deposition at active sites simultaneously leads to increased non-ohmic polarization and ohmic losses at the fuel electrode [29].

2.2. Corrosion of Metallic Components Interconnect materials in SOFC/SOEC systems are exposed to a so-called dual atmosphere of air and hydrogen-based fuel on opposing sides. They must separate these gases and ensure the gases flow to their respective electrodes for proper cell function. Therefore, interconnects must remain stable in this dual atmosphere for the entire cell lifespan to maintain cell efficiency. However, at the intermediate cell operating temperature, interconnects undergo aggressive corrosion via iron oxidation [43,44]. More specifically, it has been shown experimentally that this oxidation occurs rapidly (within a 50 h test) and Materials 2018, 11, 2169 9 of 16

Materials 2018, 11, x FOR PEER REVIEW 9 of 15 is more pronounced on the air-exposed side of the interconnect alloy (Figure8(1,2)). In comparison, an interconnectrapidly (within alloy a 50 left h test) to oxidize and is more in a pronounced single atmosphere on the air-exposed of air (Figure side 8of(3,4)) the interconnect or hydrogen-based alloy fuel for(Figure the same8(1,2)). operating In comparison, conditions an interconnect does not showalloy left nearly to oxidize the expanse in a single and atmosphere severity of oxidation,of air highlighting(Figure 8(3,4)) the phenomenon or hydrogen-based is a result fuel offor the the speciﬁcsame operating dual atmosphere conditions condition.does not show The nearly phenomenon the expanse and severity of oxidation, highlighting the phenomenon is a result of the specific dual occurs as a result of hydrogen permeation in the metal and its movement through the metal and atmosphere condition. The phenomenon occurs as a result of hydrogen permeation in the metal and oxide scales, where it contributes to nodule iron oxide growth, and ultimately platelet-like outward its movement through the metal and oxide scales, where it contributes to nodule iron oxide growth, growth.and Itultimately also appears platelet-like to occur outward more severelygrowth. It at also lower appears operating to occur temperatures. more severely Alnegren at lower et al. ◦ testedoperating AISI 441 temperatures. in a dual atmosphere Alnegren et exposure al. tested atAISI 600, 441 700, in a anddual 800 atmosphereC, and exposure found that at 600, more 700, severe ◦ dual atmosphereand 800 °C, and corrosion found that occurred more severe at 600 dualC. atmo Thesphere samples corrosion at higher occurred temperatures at 600 °C. formedThe samples the more preferredat higher chromia/MnCr temperatures2O formed4 spinel the scales more with preferred little iron chromia/MnCr oxide overgrowth.2O4 spinel This scales “inverse with little temperature” iron effectoxide is not overgrowth. well known This [44 “inverse]. temperature” effect is not well known [44].

FigureFigure 8. Comparison 8. Comparison of ferriticof ferritic stainless-steel stainless-steel samplessamples in dual dual atmosphere atmosphere (1 (and1 and 2) and2) and single/dry single/dry air atmosphereair atmosphere (3 and (3 and4). 4).

3. Mitigation3. Mitigation Approaches Approaches To suppressTo suppress material material degradation,degradation, certain certain state-of-the-art state-of-the-art methods methods have been have developed. been developed. These Thesemethods, methods, along along with with some some newer newer means meansof prolonging of prolonging material lifetimes material will lifetimes be discussed. will be Current discussed. Currentstate-of-the-art state-of-the-art mitigation mitigation methods methods include include coating coating metallic metallic components components to reduce to reduce species species volatilization,volatilization, and and doping doping electrode electrode materials materials to toprevent prevent poisonous poisonous species species deposition. deposition. Other more Other morenovel novel approaches approaches include include the use the of use getters of gettersto remove to removecertain volatile certain species. volatile These species. forms of These degradation mitigation will now be further discussed with respect to certain component forms of degradation mitigation will now be further discussed with respect to certain component degradations aforementioned. degradations aforementioned.

3.1. Electrode Poisoning Mitigation ◦ In pre-oxidizing AISI 441 stainless steel in a CO2/CO mix at 850 C, Wongpromrat et al. found that the pre-oxidized alloy did not evolve as much gaseous chromium species due to the promotion of a Materials 2018, 11, x FOR PEER REVIEW 10 of 15

3.1. Electrode Poisoning Mitigation Materials 2018, 11, 2169 10 of 16 In pre-oxidizing AISI 441 stainless steel in a CO2/CO mix at 850 °C, Wongpromrat et al. found that the pre-oxidized alloy did not evolve as much gaseous chromium species due to the promotion singleof n-type a single chromia n-type scalechromia inﬂuenced scale influenced by the low by oxygen the low partial oxygen pressure partial pre-oxidizing pressure pre-oxidizing atmosphere of −10 −10 10 atmospherebar PO2 [45 of]. 10 The bar defect PO2 scale[45]. The of this defect chromia scale of type this lowered chromia adherencetype lowered of wateradherence vapor of water chromium speciesvapor and chromium ultimately species lowered and theultimately volatilization lowered rate the [45volatilization]. This is because rate [45]. the This possible is because point the defects withinpossible the n-type point structuredefects within are oxygenthe n-type vacancies structure andare oxygen chromium vacancies interstitials and chromium [45,46]. interstitials Therefore, the [45,46]. Therefore, the OH- groups can become dissolved in the oxide which is preferred, as the OH- OH- groups can become dissolved in the oxide which is preferred, as the OH- groups adsorbed on the groups adsorbed on the outside are easily combined with chromia to form the gaseous species outside are easily combined with chromia to form the gaseous species CrO2(OH)2. Chromia scales CrO2(OH)2. Chromia scales grown in air or oxygen, with a higher oxygen partial pressure, were a mix grownof inp-type air orand oxygen, n-type within defect a higher structure, oxygen and partialtherefore, pressure, exhibited were more a mixadsorbed of p-type OH- groups and n-type as in - defectdepicted structure, in Figure and therefore, 9 [45,46]. exhibited more adsorbed OH groups as depicted in Figure9[45,46].

Figure 9. Comparison of defect structures of p-type and n-type chromia with regards to surface Figure 9. Comparison of defect structures of p-type and n-type chromia with regards to surface adherence of OH− groups, adapted from [45], with permission from Elsevier, 2018. adherence of OH− groups, adapted from [45], with permission from Elsevier, 2018. Reactive element oxides are not sufficient in preventing evaporation as they are typically porous Reactive element oxides are not sufficient in preventing evaporation as they are typically porous andand thin thin in in nature nature [ 47[47].]. Likewise, Likewise, perovskite perovskite coatings coatings have proven have provenineffective ineffective as a chromium as a barrier chromium barrier[48]. [48 The]. spinel-type The spinel-type oxides have oxides proven have to proven be a much to more be a effective much more coating effective variety in coating suppressing variety in suppressingchromium chromium evaporation. evaporation. Yang et al. used Yang a etslurry-coating al. used a slurry-coating technique to deposit technique a spinel to Mn deposit1.5Co1.5 aO spinel4 Mn1.5coatingCo1.5O on4 coating the ferritic on stainless the ferritic steels stainless AISI 430 steelsand E-brite AISI [49]. 430 andIn Figure E-brite 10, [both49]. AISI In Figure 430 and 10 E-brite, both AISI 430 andwere E-brite coated wereand the coated area-specific and the resistance area-specific (ASR) resistance was measured (ASR) at was800 °C measured for 400 h. at 800 ◦C for 400 h. At theAt conclusionthe conclusion of of the the test, test, SEM SEM andand EDS were were us useded to to characterize characterize the theinterconnect interconnect materials. materials. It wasIt was found found that that no no chromium chromium species species reachedreached th thee LSCM LSCM cathode, cathode, revealing revealing the theefficacy efficacy of the of the coating [49]. coating [49]. In an attempt to mitigate the effects of sulfur poisoning on cell anodes, Marina et al. added In an attempt to mitigate the effects of sulfur poisoning on cell anodes, Marina et al. added antimony and tin to Ni-YSZ electrodes and showed a reduction in sulfur adsorption. This can been antimonyaccredited and to tin Sb to and Ni-YSZ Sn taking electrodes up active and sites showed at the aelectrode reduction surface, in sulfur weakened adsorption. sulfur adsorptive This can been accreditedbonds, toincreased Sb and sulfur Sn taking oxidation, up active or secondary sites at ph thease electrodeformation [50]. surface, What weakened seems most sulfur likely is adsorptive that, bonds,since increased Sb and Sn sulfur segregate oxidation, to surface or grain secondary boundaries phase (the formation more active [50 surface]. What sites), seems it may most directly likely is that,prohibit since Sb adsorption and Sn segregate of sulfur toat those surface sites, grain much boundaries like sulfur (the prohibits more hydrogen active surface from adhering sites), it may directlyduring prohibit poisoning adsorption [50]. More of sulfur work at is those needed sites, to muchdetermine like sulfurthe exact prohibits mechanism hydrogen of these from added adhering duringmetallic poisoning impurities. [50]. MoreHowever, work despite is needed an initial to determine decrease thein hydrogen exact mechanism oxidation of of these the Sb added and Sn- metallic impurities.doped electrodes, However, which despite recovered an initial after these decrease impuriti ines hydrogen diffused into oxidation the bulk ofNi-YSZ, the Sb the and inclusion Sn-doped of Sb and Sn in the electrode has a positive mitigating effect against sulfur poisoning. electrodes, which recovered after these impurities diffused into the bulk Ni-YSZ, the inclusion of Sb Recently, mitigation approaches to chromium poisoning of cell cathodes have led to the and Sn in the electrode has a positive mitigating effect against sulfur poisoning. development of chromium getters. Aphale et al. tested the stability of a SrO-NiO, or SrxNiyO, solid Recently, mitigation approaches to chromium poisoning of cell cathodes have led to the development of chromium getters. Aphale et al. tested the stability of a SrO-NiO, or SrxNiyO, solid solution getter, which was shown to remain stable up to 900 ◦C[51]. This getter composition had been shown to be effective in capturing gaseous chromium species through the following reaction, as in Reference [52]:

Sr9Ni7O21(s) + CrO2(OH)2(g) → SrCrO4(s) + NiO(s) + H2O(g) (4) Materials 2018, 11, x FOR PEER REVIEW 11 of 15

solution getter, which was shown to remain stable up to 900 °C [51]. This getter composition had been shown to be effective in capturing gaseous chromium species through the following reaction, Materials 2018, 11, 2169 11 of 16 as in Reference [52]:

SrNiO() +CrO(OH)() → SrCrO() +NiO() +HO() (4) This concept of using cheap material getters for vaporous chromium has shown potential in This concept of using cheap material getters for vaporous chromium has shown potential in mitigating the effects of chromium poisoning for 40,000 to 50,000 h of system operation [52]. The mitigating the effects of chromium poisoning for 40,000 to 50,000 h of system operation [52]. The science of getters may expand to the mitigation of other poisonous species like sulfur or silicon [51]. science of getters may expand to the mitigation of other poisonous species like sulfur or silicon [51]. This may prove to be the best mitigation approach for SiO deposition, as other approaches are not This may prove to be the best mitigation approach for SiO2 2deposition, as other approaches are not well researchedwell researched yet yet and and only only replacement replacement of of tectosilicate tectosilicate albite albite (NaAlSi (NaAlSi₃O3₈O) sealant8) sealant at the at hydrogen the hydrogen electrodeelectrode has beenhas been proposed proposed [5 ,[5,42].42].

(1) (2)

FigureFigure 10. SEM/EDS 10. SEM/EDS analysis analysis of: of: (1 ()1 AISI430) AISI430 andand ((22)) E-briteE-brite with with Mn Mn1.51.5CoCo1.5O1.54 Oprotection4 protection layer layer after after the contactthe contact ASR ASR measurement measurement at at 800 800◦ C°C in in airair forfor about 400 400 h. h. (a ()a SEM) SEM cross-section cross-section and and(b) EDS (b) EDSline line scan alongscan along line line A–A A–A in (ina), (a adapted), adapted from from [ 49[49],], withwith permissionpermission from from Elsevier, Elsevier, 2018. 2018.

3.2. Metallic3.2. Metallic Component Componen Corrosiont Corrosion Suppression Suppression ApproachesApproaches similar similar to theto the growth growth of of preferential preferential oxide oxide scales scales and and deposition deposition of coating of coating varieties varieties usedused to suppress to suppress cathodic cathodic poisoning poisoning havehave alsoalso been considered considered for for dual dual atmosphere atmosphere corrosion corrosion suppression. Preferential scale development of chromia and (Cr,Mn)3O4 spinel have positive effects suppression. Preferential scale development of chromia and (Cr,Mn)3O4 spinel have positive effects in suppressing corrosion of metal interconnects [53]. When pre-oxidized, the ferritic stainless steels in suppressing corrosion of metal interconnects [53]. When pre-oxidized, the ferritic stainless steels with substantial chromium (16–23%) form a dense, passivating chromia layer if the partial pressure with substantial chromium (16–23%) form a dense, passivating chromia layer if the partial pressure of of oxygen (PO2) of the oxidizing atmosphere is controlled. Forming the passivating Cr2O3 scale during oxygena pre-oxidation (PO2) of the has oxidizing shown to atmosphere be beneficial is in controlled. mitigating expansive Forming thecorrosion passivating during Crdual2O atmosphere3 scale during a pre-oxidationexposure. In has air shown or high to PO be2 atmospheres, beneﬁcial in all mitigating constituents expansive of the steel corrosion can oxidize, during which dual often atmosphere results exposure.in a chromia In air orscale high withPO cuboidal2 atmospheres, (Cr,Mn)3O all4 crystallites constituents and of points the steel of iron can oxide oxidize, growth which [53]. oftenDriving results in a chromiathe PO2 to scale a lower with limit cuboidal will allow (Cr,Mn) only3O chromia4 crystallites and (Cr,Mn) and points3O4 spinel of iron to form. oxide When growth exposed [53]. Driving to the PO2 to a lower limit will allow only chromia and (Cr,Mn)3O4 spinel to form. When exposed to dual atmosphere, the pre-oxidized samples out-performed the as-received in mitigating iron oxide overgrowth [54]. 0 Using the Gibbs free energy for metal oxide formation (∆G ), the PO2 in atmospheres can be calculated using the following relation:

0 − ∆G = −RTln(PO2) (5) Materials 2018, 11, 2169 12 of 16

In Equation (5), R is the universal gas constant in J/mol*K and T is temperature in K. This relation is the backbone of the Ellingham diagram of metal oxide formation, which is used to determine the equilibrium partial pressure of oxygen at a speciﬁc temperature and the ease of reduction of a metal oxide [54]. Oxygen activity can be approximated by its partial pressure using the following equation, as in Reference [54]: 1 2∆G0 PO = exp ( ∗ ) (6) 2 y RT

In Equation (6), y is a coefficient respective of the reaction for a certain metal oxide MxOy, highlighted in Equation (7) [54]: y xM + O → M O (7) 2 2 x y Equation (6) yields the range of oxygen partial pressure that will form a given oxide on its respective metal within the Ellingham diagram. Using the diagram, one can determine whether, at a certain temperature and PO2, a certain metal oxide will form or whether a metal will remain stable in the given conditions. Chromia is considered a dense, passivating scale that hinders cationic mobility compared to the defect-heavy p-type FeO [54]. Cr and Fe are relatively similar in size, making their relative mobility through their respective oxides similar to mobilities in the opposing oxides. Sabioni et al. found that iron diffusion in chromia was hindered by thermodynamics, as chromia appeared to lower the oxygen potential at the metal/scale interface, which prevented iron in the metal from oxidizing [55]. According to Sabioni, the bulk cationic diffusion coefficient should vary with the oxygen pressure as 3/16 (PO2) [55]. With PO2 considered, this would yield a diffusion coefficient in 1 atm oxygen equal to 5.6 times the diffusion coefficient in 10−4 atm oxygen [55]. It is clear from this that oxygen partial pressure works as a major driving force in oxide scale growth via influence on diffusion. Talic et al. investigated the doped-spinel oxides of MnCo2O4, MnCo1.7Cu0.3O4, and MnCo1.7Fe0.3O4 deposited on interconnect steel Crofer 22 APU samples using electrophoretic deposition. The goal was to suppress corrosion in air while maintaining a higher conductivity than the commonly formed Cr2O3 and MnCr2O4 scales [56]. The spinel coatings reduced the parabolic rate of oxidation at the higher end of testing temperatures (800–900 ◦C); however, their mitigating effects diminished with a decrease in temperature. This was inconclusive on whether the coatings would ultimately improve corrosion resistance over cell operating time, as the area-specific resistance of the coated Crofer 22 APU was significantly lower than the uncoated alloy [56]. The concept of doped-spinel coatings is also used to combat Cr evaporation [57]. Coating varieties using reactive elements have been beneficial in providing corrosion resistance, much like the addition of alloying elements such as manganese and chromium within the alloy itself [58]. The reactive elements are rare earth in nature, usually zirconium, lanthanum, cerium, and yttrium [58]. These elements have a high oxygen affinity and a larger ion size than chromium, making them effective in promoting certain scale development. They have also been shown to improve chromia scale conductivity [59]. Their exact mechanism, known as the reactive element effect (REE), is unknown; however, Pint has discussed hypotheses in great detail based on early observations and hypotheses from Whittle and Stringer in 1980. These hypotheses include improved chemical bonding between oxide and alloy or a possible change in oxygen vacancy-assisted diffusion due to reactive element addition. However, these hypotheses are not yet conclusive. It was also noted by Pint that alloying elements may have the same effect and should also be considered in optimizing interconnect materials [60].

4. Conclusions Solid oxide electrochemical systems serve as a promising means of providing clean and sustainable forms of energy in the near future. However, research in material development is needed to improve performance stability and increase system lifetime. The ongoing research efforts have proven effective Materials 2018, 11, 2169 13 of 16 in developing ways to mitigate the degradation of materials in these systems. Citing the nature of their operation, electrochemical systems are subjected to conditions which take a toll on the materials comprising them. Such degradations can be the result of harsh atmospheres, long operating times, elevated operating temperatures, and chemical compatibilities. These include poisoning of electrodes by gaseous species and corrosion of metallic cell interconnects caused by speciﬁc atmosphere exposure. This has led to various mitigation approaches that have examined materials used in systems, including material states, compositions, and coatings. Adsorbent gaseous species that threatened electrode operation were blocked by certain metal additions to the electrodes or captured by getters. Corrosion of metallic interconnects was mitigated through protective scale development and certain coatings. While materials challenges persist and degradation/corrosion continue to plague many electrochemical systems, research efforts and the overall potential of these systems has made them favorable in the quest for clean energy.

Author Contributions: All authors contributed significantly to this paper. Funding: This research was funded by the US Department of Energy under grant # DE-FE 0023385 and Nissan Motor Corporation. Acknowledgments: The authors acknowledge the University of Connecticut and the Center for Clean Energy Engineering for providing experimental and characterization facilities and laboratory support. Technical discussions with Patcharin Burke (NETL), Nilesh Dale (Nissan Motor Corporation), and Jeff Stevenson (PNNL) are also gratefully acknowledged. Conflicts of Interest: The authors declare no conflict of interest.

References

1. Bard, A.J.; Faulkner, L.R. Electrochemical Methods Fundamentals and Applications, 2nd ed.; John Wiley & Sons, Inc.: Hoboken, NJ, USA, 2001; ISBN 978-0-471-04372-0. 2. Mahato, N.; Banerjee, A.; Gupta, A.; Omar, S.; Balani, K. Progress in Material Selection for Solid Oxide Fuel Cell Technology: A Review. Prog. Mater. Sci. 2015, 72, 141–337. [CrossRef] 3. Minh, N.Q.; Mizusaki, J.; Singhal, S.C. Advances in Solid Oxide Fuel Cells: Review of Progress through Three Decades of the International Symposia on Solid Oxide Fuel Cells. ECS Trans. 2017, 78, 63–73. [CrossRef] 4. Stambouli, A.B.; Traversa, E. Solid Oxide Fuel Cells (SOFCs): A Review of an Environmentally Clean and Efﬁcient Source of Energy. Renew. Sustain. Energy Rev. 2002, 6, 433–455. [CrossRef] 5. Yildiz, B. Reversible Solid Oxide Electrolytic Cells for Large-Scale Energy Storage: Challenges and Opportunities. In Woodhead Publishing Series in Energy; Kilner, J.A., Skinner, S.J., Irvine, S.J.C., Edwards, P., Eds.; Woodhead Publishing: Cambridge, UK, 2012; pp. 149–179. 6. Yokokawa, H.; Tu, H.; Iwanschitz, B.; Mai, A. Fundamental Mechanisms Limiting Solid Oxide Fuel Cell Durability. J. Power Sources 2008, 182, 400–412. [CrossRef] 7. Adams, T.A.; Nease, J.; Tucker, D.; Barton, P.I. Energy Conversion with Solid Oxide Fuel Cell Systems: A Review of Concepts and Outlooks for the Short- and Long-Term. Ind. Eng. Chem. Res. 2013, 52, 3089–3111. [CrossRef] 8. Kirubakaran, A.; Jain, S.; Nema, R.K. A Review on Fuel Cell Technologies and Power Electronic Interface. Renew. Sustain. Energy Rev. 2009, 13, 2430–2440. [CrossRef] 9. Ji, M.; Wei, Z. A Review of Water Management in Polymer Electrolyte Membrane Fuel Cells. Energies 2009, 2, 1057–1106. [CrossRef]

10. Bi, L.; Da’as, E.H.; Shaﬁ, S.P. Proton-Conducting Solid Oxide Fuel Cell (SOFC) with Y-Doped BaZrO3 Electrolyte. Electrochem. Commun. 2017, 80, 20–23. [CrossRef]

11. Bi, L.; Fabbri, E.; Traversa, E. Solid Oxide Fuel Cells with Proton-Conducting La0.99Ca0.01NbO4 Electrolyte. Electrochim. Acta 2018, 260, 748–754. [CrossRef] 12. Zhang, X.; Jiang, Y.; Hu, X.; Sun, L.; Ling, Y. High Performance Proton-Conducting Solid Oxide Fuel Cells

with a Layered Perovskite GdBaCuCoO5+xCathode. Electron. Mater. Lett. 2018, 14, 147–153. [CrossRef] 13. Lefebvre-Joud, F.; Gauthier, G.; Mougin, J. Current Status of Proton-Conducting Solid Oxide Fuel Cells Development. J. Appl. Electrochem. 2009, 39, 535–543. [CrossRef] Materials 2018, 11, 2169 14 of 16

14. Chen, K.; Jiang, S.P. Review—Materials Degradation of Solid Oxide Electrolysis Cells. J. Electrochem. Soc. 2016, 163, F3070–F3083. [CrossRef] 15. Schmidt, O.; Gambhir, A.; Staffell, I.; Hawkes, A.; Nelson, J.; Few, S. Future Cost and Performance of Water Electrolysis: An Expert Elicitation Study. Int. J. Hydrogen Energy 2017, 42, 30470–30492. [CrossRef] 16. Zheng, Y.; Wang, J.; Yu, B.; Zhang, W.; Chen, J.; Qiao, J.; Zhang, J. A Review of High Temperature

Co-Electrolysis of H2O and CO2 to Produce Sustainable Fuels Using Solid Oxide Electrolysis Cells (SOECs): Advanced Materials and Technology. Chem. Soc. Rev. 2017, 46, 1427–1463. [CrossRef][PubMed] 17. Gupta, S.; Mahapatra, M.; Singh, P. Lanthanum Chromite Based Perovskites for Oxygen Transport Membrane. Mater. Sci. Eng. R Rep. 2015, 90, 1–36. [CrossRef] 18. Eichler, A. Tetragonal Y-Doped Zirconia: Structure and Ion Conductivity. Phys. Rev. B 2001, 64, 174103. [CrossRef] 19. Ormerod, R.M. Solid Oxide Fuel Cells. Chem. Soc. Rev. 2003, 32, 17–28. [CrossRef][PubMed]

20. Fabbri, E.; D’Epifanio, A.; Bartolomeo, E.; Licoccia, S.; Traversa, E. BaCe1-x-YZrxYyO3-d Protonic Conductor for Intermediate Temperature Solid Oxide Fuel Cells (IT-SOFCs). ECS Trans. 2008, 6, 23–28. 21. Sun, C.; Hui, R.; Roller, J. Cathode Materials for Solid Oxide Fuel Cells: A Review. J. Solid State Electrochem. 2010, 14, 1125–1144. [CrossRef] 22. Sun, C.; Stimming, U. Recent Anode Advances in Solid Oxide Fuel Cells. J. Power Sources 2007, 171, 247–260. [CrossRef] 23. Skinner, S.J.; Kilner, J.A. Oxygen Ion Conductors. Mater. Today 2003, 6, 30–37. [CrossRef] 24. Shaikh, S.P.S.; Muchtar, A.; Somalu, M.R. A Review on the Selection of Anode Materials for Solid-Oxide Fuel Cells. Renew. Sustain. Energy Rev. 2015, 51, 1–8. [CrossRef] 25. Sakai, N.; Yokokawa, H.; Horita, T.; Yamaji, K. Lanthanum Chromite-Based Interconnects as Key Materials for SOFC Stack Development. Int. J. Appl. Ceram. Technol. 2004, 1, 23–30. [CrossRef] 26. Yang, Z. Recent Advances in Metallic Interconnects for Solid Oxide Fuel Cells. Int. Mater. Rev. 2008, 53, 39–54. [CrossRef] 27. Yang, Z.; Xia, G.G.; Walker, M.S.; Wang, C.M.; Stevenson, J.W.; Singh, P. High Temperature Oxidation/Corrosion Behavior of Metals and Alloys under a Hydrogen Gradient. Int. J. Hydrogen Energy 2007, 32, 3770–3777. [CrossRef] 28. Quadakkers, W.J.; Piron-Abellan, J.; Shemet, V.; Singheiser, L. Metallic Interconnectors for Solid Oxide Fuel Cells—A Review. Mater. High Temp. 2003, 20, 115–127. 29. Mahapatra, M.K.; Lu, K. Glass-Based Seals for Solid Oxide Fuel and Electrolyzer Cells—A Review. Mater. Sci. Eng. R Rep. 2010, 67, 65–85. [CrossRef] 30. Atkinson, A.; Aguiar, P.; Brandon, N.P.; Brett, D.J.L. Fuel Cells for Autonomous Vehicles, SEAS DTC Annual Technical Conference; SEAS DTC: Edinburgh, UK, 2006; p. C5. 31. Grgicak, C.M.; Green, R.G.; Giorgi, J.B. SOFC Anodes for Direct O00000xidation of Hydrogen and Methane Fuels Containing H2S. J. Power Sources 2008, 179, 317–328. [CrossRef]

32. Sun, S.; Cheng, Z. H2S Poisoning of Proton Conducting Solid Oxide Fuel Cell and Comparison with Conventional Oxide-Ion Conducting Solid Oxide Fuel Cell. J. Electrochem. Soc. 2018, 165, F836–F844. [CrossRef] 33. Sasaki, K.; Susuki, K.; Iyoshi, A.; Uchimura, M.; Imamura, N.; Kusaba, H.; Teraoka, Y.; Fuchino, H.; Tsujimoto, K.; Uchida, Y.; et al. H2S Poisoning of Solid Oxide Fuel Cells. J. Electrochem. Soc. 2006, 153, A2023. [CrossRef] 34. Xiong, Y.; Yamaji, K.; Horita, T.; Yokokawa, H.; Akikusa, J.; Eto, H.; Inagaki, T. Sulfur Poisoning of SOFC Cathodes. J. Electrochem. Soc. 2009, 156, B588. [CrossRef] 35. Aphale, A.; Liang, C.; Hu, B.; Singh, P. Chapter 6—Cathode Degradation from Airborne Contaminants in Solid Oxide Fuel Cells—A Review. Solid Oxide Fuel Cell Lifetime Reliab. Crit. Chall. Fuel Cells 2017, 102–114. [CrossRef] 36. Falk-Windisch, H.; Svensson, J.E.; Froitzheim, J. The Effect of Temperature on Chromium Vaporization and Oxide Scale Growth on Interconnect Steels for Solid Oxide Fuel Cells. J. Power Sources 2015, 287, 25–35. [CrossRef] Materials 2018, 11, 2169 15 of 16

37. Falk-Windisch, H.; Claquesin, J.; Sattari, M.; Svensson, J.-E.; Froitzheim, J. Co- and Ce/Co-Coated Ferritic Stainless Steel as Interconnect Material for Intermediate Temperature Solid Oxide Fuel Cells. J. Power Sources 2017, 343, 1–10. [CrossRef] 38. Badwal, S.P.S.; Deller, R.; Foger, K.; Ramprakash, Y.; Zhang, J.P. Interaction between Chromia Forming Alloy Interconnects and Air Electrode of Solid Oxide Fuel Cells. Solid State Ion. 1997, 99, 297–310. [CrossRef] 39. Jiang, S.P.; Zhang, J.P.; Zheng, X.G. A Comparative Investigation of Chromium Deposition at Air Electrodes of Solid Oxide Fuel Cells. J. Eur. Ceram. Soc. 2002, 22, 361–373. [CrossRef] 40. Kurokawa, H.; Jacobson, C.P.; DeJonghe, L.C.; Visco, S.J. Chromium Vaporization of Bare and of Coated Iron–chromium Alloys at 1073 K. Solid State Ion. 2007, 178, 287–296. [CrossRef] 41. Haga, K.; Adachi, S.; Shiratori, Y.; Itoh, K.; Sasaki, K. Poisoning of SOFC Anodes by Various Fuel Impurities. Solid State Ion. 2008, 179, 1427–1431. [CrossRef] 42. Hauch, A.; Jensen, S.H.; Bilde-Sorensen, J.B.; Mogensen, M. Silica Segregation in the Ni/YSZ Electrode. J. Electrochem. Soc. 2007, 154, A619–A626. [CrossRef] 43. Yang, Z.; Xia, G.; Singh, P.; Stevenson, J.W. Effects of Water Vapor on Oxidation Behavior of Ferritic Stainless Steels under Solid Oxide Fuel Cell Interconnect Exposure Conditions. Solid State Ion. 2005, 176, 1495–1503. [CrossRef] 44. Alnegren, P.; Sattari, M.; Svensson, J.E.; Froitzheim, J. Severe Dual Atmosphere Effect at 600 ◦C for Stainless Steel 441. J. Power Sources 2016, 301, 170–178. [CrossRef] 45. Wongpromrat, W.; Berthomé, G.; Parry, V.; Chandra-ambhorn, S.; Chandra-ambhorn, W.; Pascal, C.; Galerie, A.; Wouters, Y. Reduction of Chromium Volatilisation from Stainless Steel Interconnector of Solid Oxide Electrochemical Devices by Controlled Preoxidation. Corros. Sci. 2016, 106, 172–178. [CrossRef]

46. Promdirek, P.; Lothongkum, G.; Chandra-Ambhorn, S.; Wouters, Y.; Galerie, A. Oxidation Kinetics of AISI441 Ferritic Stainless Steel at High Temperatures in CO2 Atmosphere. Oxid. Met. 2014, 81, 315–329. [CrossRef] 47. Shaigan, N.; Qu, W.; Ivey, D.G.; Chen, W. A Review of Recent Progress in Coatings, Surface Modiﬁcations and Alloy Developments for Solid Oxide Fuel Cell Ferritic Stainless Steel Interconnects. J. Power Sources 2010, 195, 1529–1542. [CrossRef] 48. Mah, J.C.W.; Muchtar, A.; Somalu, M.R.; Ghazali, M.J. Metallic Interconnects for Solid Oxide Fuel Cell: A Review on Protective Coating and Deposition Techniques. Int. J. Hydrogen Energy 2017, 42, 9219–9229. [CrossRef]

49. Yang, Z.; Xia, G.-G.; Li, X.-H.; Stevenson, J.W. (Mn,Co)3O4 Spinel Coatings on Ferritic Stainless Steels for SOFC Interconnect Applications. Int. J. Hydrogen Energy 2007, 32, 3648–3654. [CrossRef] 50. Marina, O.A.; Coyle, C.A.; Engelhard, M.H.; Pederson, L.R. Mitigation of Sulfur Poisoning of Ni/Zirconia SOFC Anodes by Antimony and Tin. J. Electrochem. Soc. 2011, 158, B424. [CrossRef] 51. Aphale, A.; Uddin, M.A.; Hu, B.; Heo, S.J.; Hong, J.; Singh, P. Synthesis and Stability of SrxNiyOz Chromium Getter for Solid Oxide Fuel Cells. J. Electrochem. Soc. 2018, 165, F635–F640. [CrossRef] 52. Uddin, M.A.; Banas, C.J.; Liang, C.; Pasaogullari, U.; Recknagle, K.P.; Koeppel, B.J.; Stevenson, J.W.; Singh, P. Design and Optimization of Chromium Getter for SOFC Systems through Computational Modeling. ECS Trans. 2017, 78, 1063–1072. [CrossRef] 53. Rufner, J.; Gannon, P.; White, P.; Deibert, M.; Teintze, S.; Smith, R.; Chen, H. Oxidation Behavior of Stainless Steel 430 and 441 at 800 ◦C in Single (Air/Air) and Dual Atmosphere (Air/Hydrogen) Exposures. Int. J. Hydrogen Energy 2008, 33, 1392–1398. [CrossRef] 54. Faust, R. Dual Atmosphere Corrosion of Ferritic Stainless Steel Used for Solid Oxide Fuel Cell Applications; Chalmers University of Technology: Göteborg, Sweden, 2017.

55. Sabioni, A.C.S.; Huntz, A.M.; Silva, F.; Jomard, F. Diffusion of Iron in Cr2O3: Polycrystals and Thin Films. Mater. Sci. Eng. A 2005, 392, 254–261. [CrossRef] 56. Talic, B.; Molin, S.; Wiik, K.; Hendriksen, P.V.; Lein, H.L. Comparison of Iron and Copper Doped Manganese Cobalt Spinel Oxides as Protective Coatings for Solid Oxide Fuel Cell Interconnects. J. Power Sources 2017, 372, 145–156. [CrossRef] 57. Bateni, R.; Wei, P.; Deng, X.; Petric, A. Spinel Coatings for UNS 430 Stainless Steel Interconnects. Surf. Coat. Technol. 2007, 201, 4677–4684. [CrossRef] 58. Grolig, J.G.; Froitzheim, J.; Svensson, J.-E. Coated Stainless Steel 441 as Interconnect Material for Solid Oxide Fuel Cells: Evolution of Electrical Properties. J. Power Sources 2015, 284, 321–327. [CrossRef] Materials 2018, 11, 2169 16 of 16

59. Fontana, S.; Amendola, R.; Chevalier, S.; Piccardo, P.; Caboche, G.; Viviani, M.; Molins, R.; Sennour, M. Metallic Interconnects for SOFC: Characterisation of Corrosion Resistance and Conductivity Evaluation at Operating Temperature of Differently Coated Alloys. J. Power Sources 2007, 171, 652–662. [CrossRef] 60. Pint, B.A. Progress in Understanding the Reactive Element Effect since the Whittle and Stringer Literature Review. In John Stringer Symposium on High Temperature Corrosion; Asm Intl: Materials Park, OH, USA, 2003; pp. 9–19.

© 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).