Outstanding Properties and Performance of CaTi0.5Mn0.5O3–? for Solar-Driven Thermochemical Hydrogen Production

•Large entropy of reduction and intermediate enthalpy reduction•pO2-dependent reversible structural transition: orthorhombic ? tetragonal cubic•Outstanding H2 yield 10 mL g?1 upon at 1,350°C a cycle time 1.5 h•H2 production rate limited by gas-phase mass transport rather than material kinetics A global transition away from fossil fuel energy to sustainable sources, in particular solar energy, requires breakthroughs storage. Two-step thermochemical hydrogen (STCH) production, which utilizes the entire spectrum, functions absence precious metal catalysis, yields oxygen separately, has emerged as an attractive route for meeting this demand. Here, we report properties CaTi0.5Mn0.5O3??. The combination large moderate reduction, along with rapid kinetics, results outstanding productivity temperature just short h. furthermore displays excellent thermal stability. Beyond performance metrics, thermodynamic data connect chemistry arbitrary cycling conditions, critical step toward designing materials suitable widespread commercial adoption. Variable valence oxides perovskite crystal structure have promising candidates via two-step cycling. exceptional efficacy CaTi0.5Mn0.5O3–? (CTM55) process. enthalpy, ranging between 200 280 kJ (mol-O)?1, entropy, 120 180 J (mol-O)?1 K?1, CTM55 create favorable conditions water splitting. oxidation state changes are dominated Mn, Ti stabilizing cubic phase increasing its enthalpy. 10.0 ± 0.2 is achieved (reduction) 1,150°C (water splitting) total h, exceeding all previous reports. gas evolution suggests and, higher, process primarily magnitude driving force. storage.1Tachibana Y. Vayssieres L. Durrant J.R. Artificial photosynthesis water-splitting.Nat. Photon. 2012; 6: 511-518Crossref Scopus (1417) Google Scholar,2Young K.J. Martini L.A. Milot R.L. Snoeberger R.C. Batista V.S. Schmuttenmaer C.A. Crabtree R.H. Brudvig G.W. Light-driven fuels.Coord. Chem. Rev. 256: 2503-2520Crossref PubMed (300) Scholar demand.3Scheffe Steinfeld A. Oxygen exchange splitting H2O CO2: review.Mater. Today. 2014; 17: 341-348Crossref (273) Scholar,4Bulfin B. Vieten J. Agrafiotis C. Roeb M. Sattler Applications limitations two oxide redox cycles; review.J. Mater. 2017; 5: 18951-18966Crossref Key features approach include utilization operation catalysts, temporal differentiation steps, simplifying separation. readily implemented using non-stoichiometric because such can retain their integrity throughout process.5Chueh W.C. Falter Abbott Scipio D. Furler P. Haile S.M. High-flux solar-driven dissociation CO2 nonstoichiometric ceria.Science. 2010; 330: 1797-1801Crossref (1091) half-cycles (1) high-temperature carried out TTR, Equation 1, releases portion phase, (2) usually lower temperature, steam TWS, 2, “split” released.Thermal reduction,TTR1??ABO3-?i?1??ABO3-?f+12O2(Equation 1) Water splitting,TWS1??ABO3-?f+H2O?1??ABO3-?i+H2(Equation 2) ?i ?f non-stoichiometry values initiation completion, respectively, half-cycle, ?? = – change through To date, major classes been evaluated STCH applications, fluorites perovskites. largely constrained ceria derivatives,5Chueh Scholar,6Scheffe Thermodynamic analysis cerium-based production.Energy Fuels. 26: 1928-1936Crossref (190) whereas perovskites,7Kubicek Bork A.H. Rupp J.L.M. Perovskite oxides—a review on versatile class solar-to-fuel conversion processes.J. 11983-12000Crossref Scholar,8Vieten Bulfin Huck Horton Guban Zhu Lu Persson K.A. Materials design solid solutions applications.Energy Environ. Sci. 2019; 12: 1369-1384Crossref explicitly represented Equations 1 offer enormous range possible chemistries. Ceria-based require excessively high temperatures, ideally >1,500°C, generate non-negligible quantities process, chemical modifications produced only gains goal or reduced operation.6Scheffe Thus, attention increasingly turned perovskites, identifying that display characteristic, i.e., so drive both steps accessible temperatures. Two-temperature, benefits monotonically increases oxidation, value required, depending conditions.9Meredig Wolverton First-principles framework evaluation H2O- CO2-splitting materials.Phys. 2009; 80: 245119Crossref (83) Recent computational surveys suggested perovskites Mn4+ B-site will enthalpies.8Vieten Scholar,10Wolverton Transformative high-efficiency fuels.https://www.hydrogen.energy.gov/pdfs/review19/p167_wolverton_2019_o.pdfDate: 2019Google Experimental studies validated these predictions indicate sufficiently entropies.10Wolverton Driven considerations, examine here Ca(Ti0.5Mn0.5)O3 (CTM55), compound not stands meet requirements applications due incorporation Mn4+, but also composition formed entirely earth-abundant elements. end-member CaMnO3 potential storage material.11Bulfin Starr D.E. Azarpira Zachäus Hävecker Skorupska K. Schmücker Redox Ca0.8Sr0.2MnO3 perovskites.J. 7912-7919Crossref Scholar,12Mastronardo E. Qian X. Coronado J.M. favourable Fe-doped heat storage.J. 2020; 8: 8503-8517Crossref At ambient it adopts distorted (the GdFeO3 type), undergoing successive transitions then phases heating loss oxygen.12Mastronardo Scholar,13Rormark Morch A.B. Wiik Stolen S. Grande T. Enthalpies CaMnO3–?, Ca2MnO4–? SrMnO3–? deduced properties.Chem. 2001; 13: 4005-4013Crossref (63) Introduction inactive elements recently proven be highly effective tool tuning properties. In particular, introduction Al into (La,Sr)MnO314Deml A.M. Stevanovi? V. Holder Sanders O’Hayre R. Musgrave C.B. Tunable vacancy formation energetics complex SrxLa1–xMnyAl1–yO3.Chem. 6595-6602Crossref (67) Scholar,15Ezbiri Takacs Theiler Michalsky activity LaxSr1?xMnyAl1?yO3?? (0?x?1, 0?y?1) synthesis.J. 4172-4182Crossref 10% Fe CaMnO312Mastronardo (where strictly 3+) shown modify while retaining surprising characteristic CaMnO3. Distinct Fe, dopant work expected remain 4+ during case, configurational associated reaction might scale -ln(?3-?), opposed smaller term -ln(?3-0.5x??), where x 3+ concentration argument ln number ways arranging vacancies structure, motivating consideration tetravalent species. Furthermore, similar CaMnO3, structure,16Shi C.-Y. Hao Y.-M. Hu Z.-B. Structural magnetic single Ca(Ti1/2Mn1/2)O3.J. Magn. 2011; 323: 1973-1976Crossref (3) anticipated transform heating. Depending boundaries presumed phases, content, enhancing productivity, may transitions. work, present comprehensive study evolution, thermodynamics, water-splitting capacity CTM55. tracked situ X-ray diffraction (XRD) measurements, complemented absorption near edge spectroscopy (XANES) thin-film CTM55, providing insight relative Mn simultaneous differential scanning calorimetry (DSC) thermogravimetric (TGA) detect obtain function partial pressure (pO2) temperature. From data, determined ?. Although detailed model developed properties, were found particularly well suited applications. Thermochemical experiments performed directly assess efficacy. profiles compared computed case remains quasi-equilibrium (i.e., limited), roles kinetic macroscopic production. Overall, stability, fast render competitive high-capacity Starting powders prepared solid-state methods involving final 1,400°C air, completion sample was furnace cooled room Rietveld XRD pattern resulting (Figure S1A) revealed type distortion random arrangement B-site. refined lattice constants 5.3603(1), b 7.5506(2), c 5.3262(1) Å. Porous self-supporting monoliths (?50% porosity) TGA measurements S1B). open ensures easy access sample, surface area step, bulk diffusion lengths migration. For pulsed laser deposition (PLD) thin films, dense target (?97% density) fabricated uniaxial isostatic pressing ball-milled fine powder sintering S1C). Elemental dispersive (EDS) collected multiple positions polished yielded cation ratio Ca:Ti:Mn 50.03 0.05 : 25.09 0.04 24.88 0.10 S1D). Similarly, quantitative inductively coupled plasma-optical emission spectrometry (ICP-OES) molar 50.00 0.01 25.15 0.02 25.35 0.02. Both targeted stoichiometry had attained. Further cross-validation derived complete experiment measurement reference described below. Simultaneous thermogravimetry (TG-DSC) under three pO2 (0.208, 0.028, 4.20 × 10?5 atm) relatively ramp 10°C min?1 1). DSC higher 1B) reveal transitions, correspond narrow 30°C–40°C window stability phase. shift temperatures decreasing pO2. lowest condition, single, irreversible event detected, suggesting direct retained cooling. Mass 1A) occurs predominantly, entirely, despite 10°C–20°C hysteresis measured DSC, cooling almost coincident one another, system invariant 1C), indicating equilibrium uptake. equilibration further evident invariance weight isothermal hold 1,200°C 1D), sharply contrasts low measurement. behavior exactly analogous what CaMnO3,12Mastronardo similarly undergoes pO2, very producing transition. summarized Table S1. It noted typical expose no more ?1.2 10?6 atm 800°C ?4.8 1,000°C, would leave within regime More oxidizing half-cycle required take advantage step-change crystallographic High-temperature patterns air 100°C 600°C 1,400°C, (ramp min?1). accounting peaks higher. Representative refinement presented Figure S2. No attempt made capture over exists. behaved, showing expansion heating, presumably S3), slight increase slope These analyses thermally stable least 1,400°C. Ex after exposure (Table S2) minimum 1,600°C 1,450°C S4). Before values, content specified selected T 0.075 atm, S5A). used corroborate composition. brief, ex products stoichiometric CaTiO3 Ca0.5Mn0.5O S5B), implying completely 2+ state, even extremely reducing conditions. fractions implied 50.0 0.5 25.2 0.1 24.8 0.2. Evaluation consistent quantity material. Thermogravimetric up maximal 1,500°C ten different 0.208–4.20 included each suite experiments. challenge arises competing needs sweep evolved quickly enough environment does experiment, situation exacerbated when hence large, attain sensitivity small, additionally, absolute small. Experiments designed address challenges. Specifically, 2.14 10?4 (1.604 g), those (0.945 g). 0.028 slow 2°C min?1, utilized, recorded Agreement S6) confirm obtained. 0.0087 lower, stepped protocol 1–5 h times S7) assured most cases return inlet value. Under extrapolation (Figures S8 S9). Shown 3A TGA, lines discrete points continuous respectively. Dashed fits calculated thermodynamics Evident small anomaly ?980°C. This feature reflects observed CaMnO3–?.12Mastronardo Transition obtained derivative good agreement S1), again show expressed provides means quantifying functions. limit infinitesimal extent (???0), Gibbs ? tends zero (?redG(T,?)?0). constant, Kredeq, given byKredeq=pˆO212=exp??redG?T,?RT=exp??redH??+T?redS??RT(Equation 3) pˆO2 pO2/pref, equal referenced standard pressure, pref atm. R universal ?redG?(T,?), ?redH?(?), ?redS?(?) ?, per mole oxygen. Rearranging 3, yieldsRlnpˆO212=??redH??T+?redS??(Equation 4) While strongly dependent includes defect-laden oxide, ?redH?(?) weakly terms typically negligible dependence Following van't Hoff method,20Hao Yang C.-K. Ceria–zirconia (Ce1–xZrxO2??, ? 0.2) splitting: study.Chem. 6073-6082Crossref (136) Arrhenius plot T-pO2 pairs prepared, extracted slopes intercepts, datasets displayed 3B. Figures 3C 3D, Fit describing provide S12. exists (? < 0.015) implies uncertainty reasons, region (0.023–0.038). Nevertheless, larger respective cycling, relevant captured region, experimental (and phase). An interesting raw 3A) apparent onset plateau 0.25, reflected turn upward trends Presuming fully oxidized, coincides mean 3+. Anomalies variable known occur electronic (e.g., n-type p-type),21Mizusaki Yoshihiro Yamauchi Fueki defect perovskite-type solution La1?xSrxFeO3??.J. Solid State 1987; 67: 1-8Crossref (111) Scholar, 22Park C.Y. Jacobson A.J. Electrical conductivity nonstoichiometry La0. 2Sr0. 8Fe0. 55Ti0. 45O3–?.J. Electrochem. Soc. 2005; 152: J65Crossref (62) 23Merkulov O.V. Markov A.A. Leonidov I.A. Patrakeev M.V. Kozhevnikov V.L. SrFe1–xSnxO3–?.J. 2018; 262: 121-126Crossref (5) explain observations here. perspective long significant extrapolations avoided, any set calculations S13. put context, several additional materials, specifically, SrTi0.5Mn0.5O3–? (STM55),18Qian He Mastronardo Baldassarri Favorable SrTi0.5Mn0.5O3?? splitting.Chem. 32: 9335-9346Crossref (14) CaMnO3–? (CM),12Mastronardo La0.9Sr0.1MnO3–? (LSM91),17Ignatowich M.J. Davenport T.C. Yamazaki Impact enhanced reducibility rates production.MRS Commun. 7: 873-878Crossref (12) La0.6Sr0.4MnO3–? (LSM64),17Ignatowich La0.6Sr0.4Mn0.4Al0.6O3–? (LSMA6446),15Ezbiri La0.6Sr0.4Mn0.6Al0.4O3–? (LSMA6464),15Ezbiri CeO2–?,19Panlener R.J. Blumenthal R.N. Garnier J.E. cerium dioxide.J. Phys. Sol. 1975; 36: 1213-1222Crossref (346) 3D. Higher CM,12Mastronardo CaMn0.9Fe0.1O2.95–? (CMF91)12Mastronardo (not shown) trend comparisons approaching CeO2–? LSMA6446, moderate, comparable LSMA6464 LSM91. combination, fact above. Within STM55. wide range, somewhat La1–xSrxMnO3–? La1–xSrxMnyAl1–yO