Correlation of Time-Dependent Oxygen Surface Exchange Kinetics

Subscriber access provided by Nottingham Trent University

Energy, Environmental, and Catalysis Applications

Correlation of Time-Dependent Oxygen Surface Exchange Kinetics with Surface Chemistry of La0.6Sr0.4Co0.2Fe0.8O3- Catalysts Doyeub Kim, Jin Wan Park, Byung-Hyun Yun, Jeong Hwa Park, and Kang Taek Lee ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.9b06569 • Publication Date (Web): 13 Aug 2019 Downloaded from pubs.acs.org on August 17, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Title: Correlation of Time-Dependent Oxygen Surface Exchange Kinetics with Surface Chemistry of La0.6Sr0.4Co0.2Fe0.8O3-𝛿 Catalysts

Doyeub Kim†, Jin Wan Park†, Byung-Hyun Yun, Jeong Hwa Park, and Kang Taek Lee*

Department of Energy Science and Engineering, DGIST, Daegu 42988, Republic of Korea

† These

authors contributed equally to this work

* Corresponding Author: Prof. Kang Taek Lee Address: Daegu Gyeongbuk Institute of Science and Technology (DGIST), 333, Techno Jungang Daero, Hyeonpung-eup, Dalseong-Gun, Daegu, 42988, Republic of Korea Email: [email protected] Tel: +82-53-785-6430 Fax: +82-53-785-6409

1 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Abstract The Sr segregation at the surface of a perovskite La0.6Sr0.4Co0.2Fe0.8O3-𝛿 (LSCF) oxygen electrode is detrimental to the electrochemical performance and durability of energy conversion devices such as solid oxide fuel cells (SOFCs). However, a quantitative correlation of degradation of the surface oxygen exchange kinetics with Sr precipitation formation at the LSCF surface is not clearly understood yet. Herein, the correlation of the time-dependent degradation mechanisms of the LSCF catalysts with respect to Sr segregation phenomena at the surface were investigated at 800 ℃ for a prolonged annealing time (~800 h) by combining in-situ electrochemical measurements, and ex-situ chemical and structural analyses at the multiscale. The in-situ monitored surface exchange coefficient (kchem) was found to drastically drop by ~86% over the 800 h, and it was accompanied with the formation of Sr-containing secondary phases on the bulk LSCF surface, as expected. However, the estimated coverage of Sr-segregation on the LSCF surface was only ~15%, even after 800 h of aging time, showing significant deviation from the kchem degradation rate (~86%). The surface chemistry evolution at the clean surface area, which is believed to be electrochemically active, was further analyzed on the nanoscale. The quantified results showed that the Sr elemental fraction of the A-site at the outermost surface of the LSCF samples was becoming deficient from ~4.0 at 0 h to ~0.27 at 800 h annealing. Interestingly, the time-dependent behavioral tendencies between kchem degradation and surface Sr fractional changes were highly analogous. Thus, our results suggest that this Sr deficiency at the clean surface region more dominantly impacts on the degradation process rather than an electrochemical activity passivation by the SrOx-precipitates, which has been shown to be a major degradation mechanism of LSCF performance.


Page 2 of 31

Page 3 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Keywords: solid oxide fuel cells, perovskite, Sr-segregation, surface oxygen coefficient, oxygen reduction reaction



1. Introduction A solid oxide fuel cell (SOFC) is an eco-friendly energy conversion device that directly converts chemical energy into electric power through electrochemical reactions with remarkably high efficiency and low emissions1. The SOFC system typically operates at high temperature ranges of 800 – 1000 oC. For practical commercialization of SOFC technology, lowering the operation temperature to an intermediate temperature (IT) regime of 600 – 800 oC can extend the choice of materials to reduce system cost and improve thermal compatibility as well as long-term durability2-4. However, at reduced operating temperatures, the polarization resistance at the oxygen electrode (cathode) rapidly increases due to the thermally activated nature of the oxygen reduction reactions (ORRs)5-6. For last three decades, a mixed ionic and electronic conducting (MIEC) La0.6Sr0.4Co0.2Fe0.8O3-𝛿 (LSCF) catalyst with a perovskite (ABO3) structure has been most widely and intensively studied as an IT-SOFC oxygen electrode. It is due to its high catalytic activity for the ORR at reduced temperatures7-8. However, upon long-term annealing at SOFC operating temperatures, this LSCF material experiences a crucial drop in the ORR kinetics, which leads to substantial degradation of the electrochemical performance and limits practical applications of this material as an oxygen electrode9-13. As repeatedly reported, this degradation phenomenon is primarily due to Sr segregation, which often forms SrOx-like insulating phases at the LSCF surface14-19. Recently, the origin of Sr segregation at the perovskite oxide surface has been studied using model thin-films with considerably simplified and well-defined geometry, which readily accelerate the surface chemistry changes within tens of hours20-27. For example, Yildiz et al. demonstrated that the Sr segregation at the perovskite thin film surface was driven by the size mismatch between the host and dopant cations, which resulted in A-site cation migration toward


Page 4 of 31

Page 5 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


the perovskite surface to minimize the electrostatic and elastic energy interactions28. They also observed that the evolution of SrO/Sr(OH)2-rich phases suppressed the surface oxygen kinetics of the La0.6Sr0.4CoO3-𝛿 (LSC) catalyst29. In addition, Jung and Tuller reported that the Sr excess at the surface of SrTi1-xFexO3-𝛿 (STF) perovskite thin film was accommodated by the formation of an SrO layer within 5 h at 650 oC, and the island precipitate acted as a passivation barrier to ORR30. On the other hand, Wachsman et al. observed a similar Sr segregation phenomenon at the surface of the ‘bulk’ LSCF perovskite samples, which was conventionally sintered from powder compaction, for relatively long annealing times (100 h) at high temperatures up to 900 oC14. More recently, Jiang et al. reported the formation of a secondary Sr phase on the surface and microstructure coarsening of the practically porous LSCF electrode, resulting in a rapid increase in polarization resistance at the cathode under SOFC operating conditions31. These findings clearly suggest that Sr segregation is a universal phenomenon of the LSCF (or similar perovskite) materials during high temperature operation, irrespective of its morphology and structure, and negatively impacts ORR activity of the host catalysts. Nevertheless, quantitative correlation of Sr segregation at the LSCF electrode surface with surface oxygen exchange kinetics remains unclear, although understanding this relationship is important to overcome its performance degradation issue. In that sense, to the best of our knowledge, the following fundamental questions have not been answered yet. “How do the amount and surface coverage of Sr precipitation at a bulk LSCF surface change quantitatively for long-term operation (>500 h)?” “How does the surface coverage of Sr segregation directly correlate with the degradation of the oxygen surface exchange kinetics over time?” “How does the surface chemistry (e.g., composition) of the ‘clean’ surface region, which is reasonably far from the Sr segregation (thus, is believed to be electrochemically active), change



over time?” “What is the time-dependent relationship between the precipitate-free surface areas and oxygen surface exchange kinetics of the LSCF catalyst?” To attempt to answer these questions for the LSCF material, we systematically investigated the time-dependent behavior of its oxygen surface kinetics and surface chemistry at an elevated operating temperature (800 oC). The oxygen transport properties of the LSCF sample were continuously monitored in-situ using electrical conductivity relaxation (ECR) techniques for > 800 h. Furthermore, morphological and surface chemical changes over time of the LSCF bulk samples were analyzed and quantified on a nanometer scale using a variety of equipment including X-ray photoelectron spectroscopy (XPS), scanning electron microscopy (SEM), atomic force microscopy (AFM), transmission electron microscopy (TEM), and energy dispersive spectroscopy (STEM-EDX) assisted with a focused ion beam (FIB)-SEM dual beam system. Finally, the results from these in-situ and ex-situ analyses were directly compared, and accordingly, their correlations were discussed.

2. Experimental 2.1 Sample preparation To prepare samples for ECR measurement, La0.6Sr0.4Co0.2Fe0.8O3-ẟ powder (LSCF, Kceracell, Korea) was uniaxially pressed into a rectangular-shaped bar at 50 MPa. Pressed green bodies were sintered at 1400 oC for 4 h in air. Relative densities of the sintered samples were measured by the Archimedes method and shown to be > 96 % of the theoretical density. The LSCF bar was then polished using silicon carbide papers (P400/P800/P1200/P2000), followed by diamond suspensions (1 µm grade). Finally, the rectangular LSCF bar was ground to approximate


Page 6 of 31

Page 7 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


dimensions of 1.60 mm x 1.60 mm x 12.90 mm. The prepared bar sample was ultrasonically cleaned by immersion in ethanol prior to measurement. For the surface analysis of the LSCF samples, the LSCF powder was also uniaxially pressed into pellets in a 10 mm diameter die, and then sintered at 1400 oC for 4 h. LSCF pellets were also polished using colloidal silica and placed inside a cylindrical alumina tube with one side clogged. The gas inlet and outlet line were connected to tube to easily adjust the gas conditions, then the pellets were annealed at 800 oC for 0, 200, 400, and 800 h under the same conditions as the atmosphere, respectively.

2.2 Electrochemical measurement The conductivities of LSCF bar samples were measured by a DC four-probe method using a potentiostat (VMP-300, Biologic, France) as shown in Fig. S1. Pt wires of 0.125 mm in diameter were attached to the polished bar sample, while additional Pt wires were connected to Ag mesh at the end of the bar with Pt paste to enhance electrical contact. The bar-type LSCF pellet was loaded on a holder placed in an alumina tube of 25 mm diameter, which was manufactured same design for surface chemistry analysis. The ECR measurement was conducted over the oxygen partial pressure range from 0.01 to 0.21 atm at 800 oC. The pO2 gas stream was adjusted by mixing argon and oxygen with a mass flow controller (MFC, HORIBA, Japan). Before measurement, the pO2 was maintained at 0.01 atm for an hour to achieve equilibrium state and, then step changes in pO2 were introduced by a 4-way valve to rapidly switch between gas pathways and total gas flow rates for each pO2 stream were maintained at 200 sccm. After each measurement, the supply of the gas stream was immediately stopped and the sample was exposed under the atmospheric environments



through the gas outlet line until the next ECR measurement. Fig. S2 shows kchem values measured with different pO2 step sizes in order to verify the reliability of our ECR measurement. The electrical conductivity data of the LSCF bar sample was recorded at intervals of 50 h as a function of exposure time.

2.3 Surface chemistry characterization The surface Sr-containing components of LSCF samples were analyzed using XPS (Thermo Scientific, ESCALAB 250Xi, USA) with a monochromatic Al K𝛼 (1486.6 eV) line of an X-ray source. The binding energy was calibrated with the carbon component of the C 1s peak at 284.8 eV. Surface morphology of LSCF samples was observed using SEM (S4800, Hitachi, Japan) and AFM (NX10, Park systems, Korea). Surface topographic images were obtained using contact cantilever probes (25Pt300B, Park Systems, Korea) with a scan area of 10 × 10 µm2 and 2 × 2 µm2 for the whole surface and precipitates-free regions, respectively. The average surface roughness was calculated as the root mean squares (RMS) of the surface heights and depths at different surface points. Surface cation composition of LSCF samples at localized spots in the nanometer regime were investigated by TEM (HF-3300, Hitachi, Japan). For TEM sample preparation, a FIB-SEM (NB5000, Hitachi, Japan) dual beam system was used at the operating conditions of 6 nA beam current and 30 kV acceleration voltage. The specimen was cut vertically from the bulk within 100 nm thickness, followed by in-situ lifting to the Cu grid. The surface elemental composition on the atomic scale was quantitatively analyzed by TEM combined with EDX in scanning transmission electron microscopy (STEM) mode at an emission current of 9 nA and an acceleration voltage of


Page 8 of 31

Page 9 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


300 kV. The point EDX spectra were obtained by ESPIRIT software (Quantax 200 system, Bruker, USA) with an effective acquisition time of 40 s and a spot size of 1 nm.

3. Results and Discussion 3.1 Time-dependent oxygen transport kinetics Fig. 1a and Fig. S3 show the normalized conductivity curves acquired from polished LSCF bars, which were annealed for 0 to 800 h continuously at 800 oC, followed by oxygen partial pressure changes from 0.01 to 0.21 atm for 2500 s relaxation times. Compared to as-prepared LSCF, the time needed to reach the re-equilibrated state in aged samples increased. Fig. 1b shows the values of kchem and Dchem obtained by fitting based on the Crank model for each sample (see the Supporting Information for details). The initial kchem of as-prepared LSCF (at 0 h) was 5.53×10-4 cm s-1, however, after annealing for 800 h, the value of kchem significantly decreased to 7.89×10-5 cm s-1. In contrast to kchem, the variations of Dchem were almost negligible from 1.31×10-5 to 1.20×10-5 cm s-2. These observations suggest that the degradation of the LSCF cathode material could be strongly related to the change in surface oxygen transport properties rather than the bulk properties of the LSCF sample, which has been consistently reported in previous studies32. Fig. 1c shows the electrical conductivities of LSCF samples as a function of temperature before and after ECR measurements. The electrical conductivities of the two samples were almost identical, again indicating that the total charge transport properties in the LSCF bulk were unchanged during the in-situ ECR measurement for 800 h.

3.2 Surface chemical composition analysis



The surface chemical compositions of the post-annealed LSCF were further investigated by XPS. Fig. 2a-c show the fitting peaks of Sr 3d spectra for LSCF samples after heat treatment for 200, 400, and 800 h respectively. Based on these fitting results, the quantitative concentration of each ingredient was calculated via integration of the XPS intensity data. The deconvolution of Sr 3d spectra in the LSCF indicates that the peaks mainly consisted of two distinctive peaks due to the spin orbit, and the coupling of 3d3/2 and 3d5/2. The pair with lower binding energies (~ 133.5 eV for 3d3/2 and ~ 132 eV for 3d5/2) were recognized as the bulk-bound states of Sr in the lattice. On the other hand, the pair with higher binding energies (~ 135.5 eV for 3d3/2 and ~ 133.9 eV for 3d5/2) were identified as the Sr-containing component on the surface33-34. In this study, the measured 3d spectra were fit into two surface-bound doublets, including the surface SrO (Sr 3d5/2; 133.0 eV) and the surface SrCO3 (Sr 3d5/2 ; 133.9 eV), and a bulk-bound doublet (Sr 3d5/2; 131.7 eV)35. The fractional areas of Sr-containing components are summarized in Table 1. The variations of surface Sr, SrO, and SrCO3 ratios for post-annealed LSCF samples are plotted in Fig. 2d. The result showed that the Sr-containing components at the surface increased with longer annealing period, indicating the gradual formation of SrO and SrCO3 on the LSCF surface upon annealing.

3.3 Quantification of Sr segregation at the surface Fig. 3a-d are SEM images of surface microstructures of LSCF samples with different annealing periods for 0, 200, 400, and 800 h at 800 oC, respectively. The as-prepared LSCF sample (0 h) clearly exhibited a highly dense, smooth and flat surface. (Fig. 3a) Annealing for 200 h, however, induced morphological changes in the form of isolated precipitates on the LSCF surface. These particles appeared to be randomly assembled and highly population up to 800 h annealing. It also indicated that the surface roughness increased upon thermal treatment.


Page 10 of 31

Page 11 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


For further analysis, the surface morphologies were quantitatively investigated via AFM. Fig. 3e-h show the AFM images with low (upper images) and high magnification (lower images) of the post-annealed samples at 0, 200, 400, and 800 h, respectively. Fig. 3e shows that the asprepared LSCF (0 h) exhibited a highly uniform surface with an RMS surface roughness below 1 nm (e. g. 0.68 nm) without any precipitated particles as consistently observed in Fig. 3a. After thermal annealing for 200 h (Fig. 3f), however, the segregated particles on the surface were observed to be ~ 5 % of the total surface area coverage. As annealing time increased to 400 and 800 h, the number of the isolated precipitates increased and the total projected area (i.e., surface coverage) on the LSCF surface was increased to 7 and 14%, respectively (Fig. 3g-h). In addition, the RMS roughness of the surface area without precipitate was reached ~ 6.18 nm after 800 h annealing (Fig. 3h). It is noted that in Fig. 3e-h the vertical (z-) axis of these AFM images (bottom) has a different scale (50 nm) compared to the x- and y-axis scales (2 um) to show the surface roughness changes at different annealing times. (Also, see Fig. S4 for an AFM image where the zaxis has the same scale as the x- and y- axes.)

3.4 Comparison between oxygen surface exchange kinetics and Sr-precipitate coverage Based on the above ECR and AFM analyses, Fig. 4 compares the time-dependent change of the kchem and the Sr-precipitate coverage of the surface area for the aged LSCF samples up to 800 h, respectively. It is noted that the RMS surface roughness from AFM images in Fig. 3e-h was reflected to calculate the more realistic surface area without Sr-segregation (referred to as the ‘clean’ surface in this study) of the LSCF at different annealing times, while the clean surface area was rarely changed within 1 %, even at the maximum RMS value (~ 6.18 nm) (Fig. S4). The rate of kchem degradation changed before and after annealing for ~250 h. The kchem degradation rate of



LSCF during the initial annealing for 250 h (0.26 %/h) was faster compared to that of the remaining 550 h annealing period (0.036 %/h), suggesting possibly different degradation mechanisms. Nevertheless, the kchem value continuously decreased by 85.5% (i.e., 14.5 % of its initial value) during 800 h of annealing. In comparison, the clean surface area of LSCF decreased by 15% (coverage of Sr-segregation) for 800 h aging time, thus showing significant deviation from the kchem degradation value. In previous studies, degradation mechanisms of perovskite catalysts (including LSCFs) for ORRs have been typically explained by the surface segregation of the Sr element and subsequent formation of the insulating Sr-rich phase on the electrode surface (thus, blocking the ORR)36. For example, Jung and Tuller reported that the SrO excess layer on the STF thin film surface acted as a passive barrier for ORR due to its nature as a wide band gap insulator (Eg ~ 6 eV)30. More recently, Barnett et al. observed that kchem of the porous LSCF cathode decreased by ~10 times after aging below 825 oC due to Sr surface segregation on the LSCF surface32. However, in these and other reported studies, the quantitative correlation of the Sr segregation amount (e.g., surface coverage) and surface oxygen exchange properties were not directly explained. Moreover, our results suggest that the tremendous degradation (~86 %) of the kchem value cannot be accounted for only with the surface coverage of the inactive Sr segregation (~ 15 %) on the LSCF surface. Thus, we further investigated the clean surface of the LSCF without Sr-rich precipitates.

3.5 Localized chemical composition analysis at the clean surface The surface chemical compositions of LSCF samples were quantified via atomic-scale EDX by TEM. Fig. 5a shows the schematic steps of the in-situ lift-out technique for TEM specimen fabrication by a FIB-SEM dual beam system. To obtain the TEM sample of the LSCF surface, the


Page 12 of 31

Page 13 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


surface area was covered with protective layers (carbon and tungsten), followed by a FIB etching process to leave a region of interest. The TEM specimen was then completely milled parallel to the surface and attached to the grid for quantitative analysis. Fig. 5b shows a magnified crosssectional TEM image of a representative LSCF surface area. To identify the chemical stoichiometry of the clean surface, the localized regions away from the Sr segregation were selected as the areas marked by a square, as shown in Fig. 5b. For each region, the surface elemental composition was determined at the spot within 2 nm from the surface. Then, the elemental depth profile of the sample was established by analyzing EDX on ten points sequentially at 10 nm intervals, as shown in Fig. 5c. Fig. 6 shows the fractional distribution of the Sr element in the A-site cation lattice ([Sr]:([Sr]+[La])) of the clean surface of the LSCF samples as a function of distance from the sample surface with different annealing times from 0 to 800 h. For the as-prepared LSCF sample (0 h), the distribution of Sr was found to be approximately the ideal stoichiometry value (0.4) of 0.39±0.05 without noticeable changes from the surface to the interior (~100 nm depth), as shown in Fig. 6a. As the annealing time increased to 200, 400, and 800 h, the estimated A-site cation ratio ([Sr]:([Sr]+[La])) of the LSCF samples continuously reduced to 0.32±0.01, 0.30±0.01 and 0.27±0.09, respectively (Fig. 6b-d). In addition, it was observed that the depth of Sr deficiency in the LSCF samples with different annealing times was similar to ~25 nm. This result indicates that the Sr phase can be exsolved from the outermost layer within 25 nm of the LSCF lattice and form isolated SrOx precipitates through preferential lattice and surface diffusion of Sr ions, resulting in Sr-deficiency of the exposed clean surface of the LSCF sample.

3.6 Comparison between the clean surface chemistry and oxygen surface exchange kinetics



Fig. 7 plots the time-dependent changes of the near surface Sr fraction (values of ~ 0 nm depth estimated in Fig. 6 for LSCF samples at different annealing times, showing a continuous decrease in the estimated Sr surface ratio (left y-axis) (counted by [Sr]:([Sr]+[La])) over 800 h. Moreover, it was observed that there were two different slopes before and after 250 h of annealing. Comparing these values with kchem values over time, as shown in Fig. 4 (overlapped in Fig. 7, right y-axis), the behavioral tendency of degradation of these two different properties was highly analogous. Thus, we believe that the Sr deficiency in the A-site cation lattice of the clean surface can be strongly correlated with time-dependent degradation of the oxygen surface exchange kinetics. Based on results in this study, the degradation mechanism on ORR activity of the perovskite LSCF catalyst is schematically illustrated in Fig. 8. Initially, the desired oxygen exchange reactions occur on the surface of the stoichiometric LSCF cathode (left side of Fig. 8). Upon prolonged annealing at high temperatures, however, the catalytic activity of the clean surface oxygen exchange reaction of LSCF continuously decreases due to two major mechanisms: formation of Sr-rich precipitates on the surface and a deficiency of Sr of the LSCF lattice in the surface area (precipitates-free region) (right part of Fig. 8). Furthermore, our study suggests that the latter mechanism (Sr-deficiency on the clean surface, marked as B in Fig. 8) has a more dominant impact on ORR performance degradation compared to the former (Sr-precipitate covering the surface, marked as A in Fig. 8), which, until this study, was more commonly believed to be the dominant mechanism.

4. Conclusions In this study, we investigated the direct correlation of the oxygen surface exchange kinetics evolution over time with the surface chemistry changes related to Sr-segregation of the LSCF


Page 14 of 31

Page 15 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


oxygen electrode. The time-dependent surface oxygen transport kinetics of the bulk LSCF sample was monitored in-situ via the ECR technique during the long-term annealing period at 800 oC for 800 h in ambient air. The kchem was found to be continuously reduced by ~86 % after 800 h, while, the Dchem remained almost unchanged over time. As expected, the observation of the surface morphology by AFM and SEM combined with the surface chemical composition analysis by XPS indicated that Sr-segregation occurred on the LSCF surface, showing a proportional increase in the SrOx-precipitation coverage of the LSCF surface over time. However, the active surface coverage with Sr-precipitates (~15 %) after annealing for 800 h could not reasonably account for the significant drop (~86 %) of kchem for the same period. Thus, the ‘clean’ surface chemistry of the LSCF was further analyzed by high-resolution TEM-EDX assisted by the FIB-SEM sampling technique. At the outermost surface, the quantified fractional amounts of the Sr element in the Asite cation lattice was changed from ideal stoichiometry (0.4) before annealing to a highly Srdeficient state (~0.27) after 800 h, indicating that the Sr-segregation on the LSCF surface lead to Sr-deficiency of the exposed clean surface of the LSCF. Furthermore, this tendency of the redistribution of the Sr element at the LSCF surface was highly analogous to the degradation of the surface oxygen exchange kinetics for 800 h. Thus, our results demonstrate that the deactivation of the LSCF surface is more dominantly affected by the non-stochiometric Sr-deficiency at the LSCF surface rather than an passivation effect of the active surface area by the Sr-rich phase, which is commonly explained as the major degradation mechanism of the perovskite oxygen electrode in previous literature.



Acknowledgement This work was supported by the Global Frontier R&D on Center for Multiscale Energy System funded by the National Research Foundation under the Ministry of Science, ICT & Future Planning, Korea (2014M3A6A7074784). This work was also supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT; The Ministry of Science and ICT) (2019M3E6A1066426).


Page 16 of 31

Page 17 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


References (1) Brett, D. J.; Atkinson, A.; Brandon, N. P.; Skinner, S. J. Intermediate temperature solid oxide fuel cells. Chemical Society Reviews 2008, 37 (8), 1568-1578. (2) Wachsman, E. D.; Lee, K. T. Lowering the temperature of solid oxide fuel cells. Science 2011, 334 (6058), 935-939. (3) Steele, B. C.; Heinzel, A. Materials for fuel-cell technologies. In Materials For Sustainable Energy: A Collection of Peer-Reviewed Research and Review Articles from Nature Publishing Group; World Scientific: 2011; pp 224-231. (4) Wachsman, E. D.; Marlowe, C. A.; Lee, K. T. Role of solid oxide fuel cells in a balanced energy strategy. Energy & Environmental Science 2012, 5 (2), 5498-5509. (5) Lee, K. T.; Lidie, A. A.; Yoon, H. S.; Wachsman, E. D. Rational Design of Lower‐Temperature Solid Oxide Fuel Cell Cathodes via Nanotailoring of Co‐Assembled Composite Structures. Angewandte Chemie International Edition 2014, 53 (49), 13463-13467. (6) Adler, S. B. Factors governing oxygen reduction in solid oxide fuel cell cathodes. Chemical reviews 2004, 104 (10), 4791-4844. (7) Bouwmeester, H. J.; Den Otter, M.; Boukamp, B. A. Oxygen transport in La 0.6 Sr 0.4 Co 1− y Fe y O 3− δ. Journal of Solid State Electrochemistry 2004, 8 (9), 599-605. (8) Murray, E. P.; Sever, M.; Barnett, S. A. Electrochemical performance of (La, Sr)(Co, Fe) O3– (Ce, Gd) O3 composite cathodes. Solid state ionics 2002, 148 (1-2), 27-34. (9) Endler, C.; Leonide, A.; Weber, A.; Tietz, F.; Ivers-Tiffée, E. Time-dependent electrode performance changes in intermediate temperature solid oxide fuel cells. Journal of The Electrochemical Society 2010, 157 (2), B292-B298.



(10) Mai, A.; Becker, M.; Assenmacher, W.; Tietz, F.; Hathiramani, D.; Ivers-Tiffée, E.; Stöver, D.; Mader, W. Time-dependent performance of mixed-conducting SOFC cathodes. Solid state ionics 2006, 177 (19-25), 1965-1968. (11) Simner, S. P.; Anderson, M. D.; Engelhard, M. H.; Stevenson, J. W. Degradation Mechanisms of La–Sr–Co–Fe–O3 SOFC Cathodes. Electrochemical and Solid-State Letters 2006, 9 (10), A478-A481. (12) Oh, D.; Yoo, J.; Gostovic, D.; Jones, K. S.; Wachsman, E. A Kinetic Study of Catalytic Activity Degradation of La0. 6Sr0. 4Co0. 2Fe0. 8O3-δ. ECS Transactions 2009, 16 (51), 97105. (13) Li, Y.; Zhang, W.; Zheng, Y.; Chen, J.; Yu, B.; Chen, Y.; Liu, M. Controlling cation segregation in perovskite-based electrodes for high electro-catalytic activity and durability. Chemical Society Reviews 2017, 46 (20), 6345-6378. (14) Oh, D.; Gostovic, D.; Wachsman, E. D. Mechanism of La 0.6 Sr 0.4 Co 0.2 Fe 0.8 O 3 cathode degradation. Journal of Materials Research 2012, 27 (15), 1992-1999. (15) Druce, J.; Tellez, H.; Burriel, M.; Sharp, M.; Fawcett, L.; Cook, S.; McPhail, D.; Ishihara, T.; Brongersma, H.; Kilner, J. Surface termination and subsurface restructuring of perovskite-based solid oxide electrode materials. Energy & Environmental Science 2014, 7 (11), 3593-3599. (16) Zhao, L.; Drennan, J.; Kong, C.; Amarasinghe, S.; Jiang, S. P. Insight into surface segregation and chromium deposition on La 0.6 Sr 0.4 Co 0.2 Fe 0.8 O 3− δ cathodes of solid oxide fuel cells. Journal of Materials Chemistry A 2014, 2 (29), 11114-11123. (17) Ding, H.; Virkar, A. V.; Liu, M.; Liu, F. Suppression of Sr surface segregation in La 1− x Sr x Co 1− y Fe y O 3− δ: a first principles study. Physical Chemistry Chemical Physics 2013, 15 (2), 489-496.


Page 18 of 31

Page 19 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(18) Niania, M.; Podor, R.; Britton, T. B.; Li, C.; Cooper, S. J.; Svetkov, N.; Skinner, S.; Kilner, J. In situ study of strontium segregation in La0. 6Sr0. 4Co0. 2Fe0. 8O3− δ in ambient atmospheres using high-temperature environmental scanning electron microscopy. Journal of Materials Chemistry A 2018, 6 (29), 14120-14135. (19) Koo, B.; Kim, K.; Kim, J. K.; Kwon, H.; Han, J. W.; Jung, W. Sr segregation in perovskite oxides: why it happens and how it exists. Joule 2018, 2, 1476-1499 (20) Chen, Y.; Jung, W.; Cai, Z.; Kim, J. J.; Tuller, H. L.; Yildiz, B. Impact of Sr segregation on the electronic structure and oxygen reduction activity of SrTi 1− x Fe x O 3 surfaces. Energy & Environmental Science 2012, 5 (7), 7979-7988. (21) Lee, D.; Lee, Y.-L.; Grimaud, A.; Hong, W. T.; Biegalski, M. D.; Morgan, D.; Shao-Horn, Y. Enhanced oxygen surface exchange kinetics and stability on epitaxial La0. 8Sr0. 2CoO3− δ thin films by La0. 8Sr0. 2MnO3− δ decoration. The Journal of Physical Chemistry C 2014, 118 (26), 14326-14334. (22) Chen, Y.; Téllez, H.; Burriel, M. n.; Yang, F.; Tsvetkov, N.; Cai, Z.; McComb, D. W.; Kilner, J. A.; Yildiz, B. Segregated Chemistry and Structure on (001) and (100) Surfaces of (La1–x Sr x) 2CoO4 Override the Crystal Anisotropy in Oxygen Exchange Kinetics. Chemistry of Materials 2015, 27 (15), 5436-5450. (23) Rupp, G. M.; Téllez, H.; Druce, J.; Limbeck, A.; Ishihara, T.; Kilner, J.; Fleig, J. Surface chemistry of La 0.6 Sr 0.4 CoO 3− δ thin films and its impact on the oxygen surface exchange resistance. Journal of Materials Chemistry A 2015, 3 (45), 22759-22769. (24) Ma, W.; Kim, J. J.; Tsvetkov, N.; Daio, T.; Kuru, Y.; Cai, Z.; Chen, Y.; Sasaki, K.; Tuller, H. L.; Yildiz, B. Vertically aligned nanocomposite La 0.8 Sr 0.2 CoO 3/(La 0.5 Sr 0.5) 2 CoO 4



cathodes–electronic structure, surface chemistry and oxygen reduction kinetics. Journal of Materials Chemistry A 2015, 3 (1), 207-219. (25) Tsvetkov, N.; Lu, Q.; Sun, L.; Crumlin, E. J.; Yildiz, B. Improved chemical and electrochemical stability of perovskite oxides with less reducible cations at the surface. Nature materials 2016, 15 (9), 1010. (26) Li, Y.; Zhang, W.; Wu, T.; Zheng, Y.; Chen, J.; Yu, B.; Zhu, J.; Liu, M. Segregation Induced Self‐Assembly of Highly Active Perovskite for Rapid Oxygen Reduction Reaction. Advanced Energy Materials 2018, 8 (29), 1801893. (27) Rupp, G. M.; Opitz, A. K.; Nenning, A.; Limbeck, A.; Fleig, J. Real-time impedance monitoring of oxygen reduction during surface modification of thin film cathodes. Nat Mater 2017, 16 (6), 640-645 (28) Lee, W.; Han, J. W.; Chen, Y.; Cai, Z.; Yildiz, B. Cation size mismatch and charge interactions drive dopant segregation at the surfaces of manganite perovskites. Journal of the American Chemical Society 2013, 135 (21), 7909-7925. (29) Cai, Z.; Kubicek, M.; Fleig, J. r.; Yildiz, B. Chemical Heterogeneities on La0. 6Sr0. 4CoO3− δ Thin Films Correlations to Cathode Surface Activity and Stability. Chemistry of materials 2012, 24 (6), 1116-1127. (30) Jung, W.; Tuller, H. L. Investigation of surface Sr segregation in model thin film solid oxide fuel cell perovskite electrodes. Energy & Environmental Science 2012, 5 (1), 5370-5378. (31) Liu, Y.; Chen, K.; Zhao, L.; Chi, B.; Pu, J.; Jiang, S. P.; Jian, L. Performance stability and degradation mechanism of La0. 6Sr0. 4Co0. 2Fe0. 8O3− δ cathodes under solid oxide fuel cells operation conditions. International Journal of Hydrogen Energy 2014, 39 (28), 15868-15876.


Page 20 of 31

Page 21 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(32) Wang, H.; Barnett, S. A. Degradation Mechanisms of Porous La0. 6Sr0. 4Co0. 2Fe0. 8O3-δ Solid Oxide Fuel Cell Cathodes. Journal of The Electrochemical Society 2018, 165 (7), F564F570. (33) Van Der Heide, P. Systematic x‐ray photoelectron spectroscopic study of La1− xSrx‐based perovskite‐type oxides. Surface and Interface Analysis: An International Journal devoted to the development and application of techniques for the analysis of surfaces, interfaces and thin films 2002, 33 (5), 414-425. (34) Pan, Z.; Liu, Q.; Zhang, L.; Zhang, X.; Chan, S. H. Effect of Sr surface segregation of La0. 6Sr0. 4Co0. 2Fe0. 8O3− δ electrode on its electrochemical performance in SOC. Journal of The Electrochemical Society 2015, 162 (12), F1316-F1323. (35) Yu, Y.; Nikiforov, A. Y.; Kaspar, T. C.; Woicik, J. C.; Ludwig, K. F.; Gopalan, S.; Pal, U. B.; Basu, S. N. Chemical characterization of surface precipitates in La0. 7Sr0. 3Co0. 2Fe0. 8O3δ as cathode material for solid oxide fuel cells. Journal of Power Sources 2016, 333, 247-253. (36) Kubicek, M.; Limbeck, A.; Frömling, T.; Hutter, H.; Fleig, J. Relationship between cation segregation and the electrochemical oxygen reduction kinetics of La0. 6Sr0. 4CoO3− δ thin film electrodes. Journal of The Electrochemical Society 2011, 158 (6), B727-B734.



Page 22 of 31

Table 1. Fractional ratios of Sr 3d5/2 spectrum for bulk, surface SrO, and SrCO3 components of LSCF samples after annealing.

Annealing time

SrO/total

SrCO3/total

total

200h

0.276

0.175

0.451

500h

0.284

0.210

0.494

800h

0.309

0.266

0.575


Page 23 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 1. ECR measurements on LSCF bar samples: (a) normalized conductivity relaxation profiles continuously measured at 800 oC from as-prepared to 800 h over the pO2 range from 0.01 to 0.21 atm, (b) time-dependence of the oxygen surface exchange and diffusion coefficient, (c) electrical conductivities of LSCF samples before/after ECR measurement in atmospheric pressure (pO2 : 0.21 atm)



Figure 2. Deconvolution of peak fitting results of Sr 3d spectra for LSCF samples after (a) 200 h annealing, (b) 400 h annealing, and (c) 800 h annealing. (d) Srx/(Srsurface+Srbulk) ratio of Srsurf, Srsurf(SrO), and Srsurf(SrCO3) components of LSCF as a function of annealing time.


Page 24 of 31

Page 25 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 3. SEM (a-d) and AFM (e-h) micrographs of LSCF surfaces, subsequent annealing at 800 oC

for 0 h (a, e), 200 h (b, f), 400 h (c, g), and 800 h (d, h) and inserted magnified surface areas.



Figure 4. The time-dependent change of active surface area (left y-axis) and relative kchem degradation (right y-axis) for aged LSCF samples up to 800 h. The kchem were continuously measured at 800 ℃ up to 800 h over pO2 range from 0.01 to 0.21 atm.


Page 26 of 31

Page 27 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 5. (a) Sequence for TEM lamella preparation by FIB-SEM lift-out techniques for LSCF samples. (b) An STEM image of the overall cross-sectional surface (red rectangle marks) included desired clean surface area (mint rectangle marks) of LSCF surface. (c) Magnified spot of clean surface area with ten depth points at 10 nm intervals.



Figure 6. TEM-EDX depth-profiles of A-site cation ratios, expressed by [Sr]:([Sr]+[La]), at the near surface region for LSCF samples annealed for (a) 0 h, (b) 200 h, (c) 400 h, and (d) 800 h.


Page 28 of 31

Page 29 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 7. The time-dependent change of estimated Sr surface ratio ([Sr]:([Sr]+[La])) (left y-axis) at the Sr-precipitate-free region and relative kchem degradation (right y-axis) for aged LSCF samples up to 800 h.



Figure 8. Schematic illustration of LSCF catalyst degradation mechanisms at the surface (A: Srsegregation, B: Sr-deficient region) upon long-term annealing.


Page 30 of 31

Page 31 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Graphical abstract


Correlation of Time-Dependent Oxygen Surface Exchange Kinetics

Recommend Documents