Sunshine to petrol: Thermochemistry for solar fuels

Transcription

1 Engineering Conferences International ECI Digital Archives CO2 Summit II: Technologies and Opportunities Proceedings Spring Sunshine to petrol: Thermochemistry for solar fuels James Miller Sandia National Laboratories, Eric Coker Sandia National Laboratories Andrea Ambrosini Sandia National Laboratories Anthony McDaniel Sandia National Laboratories Ivan Ermanoski Sandia National Laboratories See next page for additional authors Follow this and additional works at: Part of the Environmental Engineering Commons Recommended Citation James Miller, Eric Coker, Andrea Ambrosini, Anthony McDaniel, Ivan Ermanoski, and Ellen Stechel, "Sunshine to petrol: Thermochemistry for solar fuels" in "CO2 Summit II: Technologies and Opportunities", Holly Krutka, Tri-State Generation & Transmission Association Inc. Frank Zhu, UOP/Honeywell Eds, ECI Symposium Series, (216). co2_summit2/18 This Abstract and Presentation is brought to you for free and open access by the Proceedings at ECI Digital Archives. It has been accepted for inclusion in CO2 Summit II: Technologies and Opportunities by an authorized administrator of ECI Digital Archives. For more information, please contact

2 Authors James Miller, Eric Coker, Andrea Ambrosini, Anthony McDaniel, Ivan Ermanoski, and Ellen Stechel This abstract and presentation is available at ECI Digital Archives:

3 solar. sandia.gov SUNSHINE TO PETROL: THERMOCHEMISTRY FOR SOLAR FUELS James E. Miller, Eric N. Coker, Andrea Ambrosini, Anthony H. McDaniel, Ivan Ermanoski, Ellen B. Stechel* Sandia National Laboratories, *Arizona State University Sandia National Laboratories is a multi-program laboratory managed and operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin Corporation, for the U.S. Department of Energy s National Nuclear Security Administration under contract DE-AC4-94AL85

4 Solar Fuels Impact: Meeting a significant fraction of transportation fuel demand with solar fuels is certainly plausible! High solar to fuel efficiency (>1% Annual Average) is absolutely required. Cost Scale (land, materials of construction (embedded energy)) Water, CO 2 are not limiting Water consumption/cost relatively low (water rights?) High impact opportunity for CO 2 utilization long term requires air capture. Consistent with other human activities occurring over multiple decades. E.B. Stechel and J.E. Miller Re-energizing CO 2 to fuels with the sun: Issues of efficiency, scale, and economics Journal of CO 2 Utilization, 1 (213)

5 Solar Thermal Heat In, Fuel Out Annual Average values Use entire solar spectrum Convert heat directly to chemical energy. Avoid conversion to mechanical or electrical work. Desired outcome defines minimum efficiency. Theoretical value must be higher to account for losses. Capitalize on decades of Synfuel technology, e.g. nco + (2n+1)H 2 C n H 2n+2 + nh 2 O Siegel, Miller, Ermanoski, Diver, Stechel, Ind. Eng. Chem. Res., 213, 52, 3276.

6 A Simple Concept: Heat in, Fuel Out 1) 1/δ MO x 1/δ MO (x-δ) + ½ O 2 2) 1/δ MO (x-δ) + CO 2 1/δ MO x + CO A thermochemical cycle is essentially an engine that converts heat into work in the form of stored chemical energy. Efficiency gains are possible as initial conversion to mechanical work and electricity are avoided. Of interest here are two-step, metal oxide-based processes. 3) CO 2 CO + ½ O 2 Divide an unfavorable endothermic reaction (H 2 O H 2 + ½ O 2, or CO 2 CO + ½ O 2 ) into two thermodynamically favorable reactions. 4

7 The thermodynamic cost implementing a reaction as a cycle DH red endotherm DH oxid exotherm = DH fuel 8 6 (1) (3) (1) Reduction (2) Oxidation (3) Thermolysis G (kcal/mol) (2) T low Temperature (K) T high DH fuel product DH red + C p DT net = Max. thermal eff. Each reaction favorable at a different temperature Some heat will be rejected as an exotherm Temperature gap and exotherm are a function of the active material 5

8 Heat of reaction and thermodynamically defined temperatures. Materials with high reduction temperature, low oxidation temperature (wide spread) minimize reduction enthalpy. Sensible heat considerations favor a narrow spread. G 1 (kcal/mol) % 3% 4% 5% % Temperature (K) CO 2 Thermolysis 7% 8% Note: these use a slightly lower fuel cell-like efficiency 9% B T OX ( C) η max = G 3(T=298) H T TR ( C) TH = TH =.75 Ideal behavior assumed. Thermodynamic T TR and T OX imply DH and DS and vice versa. Not all combinations are realistic. Miller, Ambrosini, Coker, McDaniel, Allendorf, Energy Procedia, 214, 49,

9 Reaction Extent and the Utilization Factor The possible efficiency increases with degree of reaction (d) and/or effectiveness of recuperation. When utilization is low, sensible heat demand becomes a more dominant factor than DH 1. Siegel, Miller, Ermanoski, Diver, Stechel, Ind. Eng. Chem. Res., 213, 52 (9), Efficiency is a function of: Thermodynamics: DH 1 (T high & T low ), d Kinetics: d The reactor: recuperation effectiveness & Pressures, sweep etc. (work input Thermal Efficiency ( H CO Combustion / Net Solar Heat In) The maximum possible efficiency is limited by DH 1. High efficiency (small DH 1 ) corresponds to a large T high -T low H1 = 75 kcal/mol 1 kcal/mol 125 kcal/mol Utilization Factor ( F R /(1- R )) T T T 14 K 85 K 6 K

10 Non-thermodynamic Temperatures? Reduction: Work in the form of Pumping or sweep gas shifts reduction temperature. Heat to Work Conversion Penalty G (kcal/mol) Yes, but at a price! Oxidation: DG -RT*ln{[CO]/[CO 2 ]} Oxidation Reduction 2 Temperature ( C) 25 CDS 3 Separation Work Requirement DH fuel product DH red + C p DT net + Q parasitic + W/ = Thermal eff. 8

11 Optimum Temperature Swing Different lines of similar color represent different recuperation extents for gas and solid Isothermal is possible, but in my opinion inadvisable can we use that electricity to better advantage? Ermanoski, Miller, Allendorf, Phys. Chem. Chem. Phys., 214, 16,

12 CR5 : First-of-a-kind approach and our attempt to apply the lessons. Counter-Rotating-Ring Receiver/Reactor/Recuperator (CR5) Reactorizing a Countercurrent Recuperator Continuous flow, Spatial separation of products, Thermal recuperation

13 Performance Map of Gen-1 Prototype Power Required to Maintain Reduction Temp (Watts) P = 2.9X P = 24.9X Collect data to validate models, guide improvements Ceria-based fins on rings 6 Data Sets: Cold, 2@ 145 C, 2@ 155 C, 162 C 3 ring rotation speeds, 3 CO 2 flow rates for each Constant Ar flow, Pressure =.5 atm Floating Pressure at 155 C P = 23.7X P = 18.8X Total Flow Rate/Rotation Period (Std liters/min 2 ) Data Set 2B, 145 C Data Set 3, 155 C Data Set 4, 155 C Data Set 5, 162 C CO Concentration, Efficiency (%) Efficiency CO Concentration Oxidation Temp.3 13:19: 13:24: 13:29: 13:34: 13:39: Mountain Daylight Time 1/rpm Thermal reduction Temp Temperature ( C) Miller, Allendorf, Ambrosini, Coker, Diver, Ermanoski, Evans, Hogan, McDaniel Development and Assessment of Solar-Thermal-Activated Fuel Production: Phase 1 Summary SAND , July

14 For Your Viewing Pleasure Operating with 22 Rings

15 Materials Challenges 1/δ MO x 1/δ MO (x-δ) + ½ O 2 d extent of reaction 1/δ MO (x-δ) + CO 2 1/δ MO x + CO

16 P(O 2 ), atm Metal Oxide TC Begins with Ferrites 8 Fe 3 O 4 3FeO + ½ O 2 3FeO + H 2 O Fe 3 O 4 + H FactSage Applied to TR ~2 wt% AB 2 O 4 in c-zro Fe 3 O 4 m.p. 18 K CoFe 2 O 4 m.p. 181 K NiFe 2 O 4 m.p. 186 K G (kcal/mol) 4 2 FeO Fe 3 O 4 Extrapolated Region p(o 2 ), air Zn Co Ni Fe Temperature ( C) Ferrite metal oxide cycle (Nakamura 1977) Temperature, K 19 Favorable temperature range (thermodynamics) can be manipulated via metal substitutions in Fe 3 O 4. Partial conversion now possible. 2 14

17 Metal Oxide TC Begins with Ferrites 8 6 Fe 3 O 4 3FeO + ½ O 2 3FeO + H 2 O Fe 3 O 4 + H 2.6 Bulk doped ferrites do not live up to thermodynamic expectations G (kcal/mol) 4 2 FeO Fe 3 O 4 Extrapolated Region H 2 Yield (cc/g ferrite) Co.67 Fe 2.33 O Temperature ( C) Cycle Number Ferrite metal oxide cycle (Nakamura 1977). Porous disk of cobalt ferrite produced small amount of H 2 for only one cycle. 15

18 Flow rate, sccm Ferrites work when you add Zirconia On-Sun Test: Co.67 Fe 2.33 O 4 /YSZ (1:4).6.5 T TR 158 C, T OX 15 C H 2 = scc/g ferrite each cycle H2 O Time, sec Pioneered by Kodama et. al. (ISEC) 24, ISEC , Portland, OR.

19 Fe dissolution and oxygen transport are the keys Mass (%) Inert CO 2 1 wt-% Fe 2 O 3 /8YSZ Time (min.) wt-% Fe 2 O 3 /8YSZ Temperature ( C) 18 O Raw 18 O Raw Beyond the solubility limit additional Fe contributes little to the overall gas yield O Masked 18 O Masked Fe utilization (%) Reaction with 18 O-labelled CO 2 confirms limited utilization of bulk particles relative to Fe/YSZ. Fe EDS Mol % Fe E.N. Coker, J.A. Ohlhausen, A. Ambrosini, and J.E. Miller J. Mater. Chem., 212, 22, DOI:1.139/C2JM15324F.

20 Small Dimension Structures ALD thin film peak production rate ~ 1X faster than bulk Chemically reduced ALD Fe:ZrO 2 nanoparticles Thermally reduced ALD particles Bulk Fe:YSZ CO 2 -splitting activity improves dramatically for particles of ZrO 2 coated with nanometer scale layers of Fe 2 O 3 Phase Segregation Pre-Processing Pre Zr Phase Segregation Fe Pre-Processing Pre-Processing Post-Processing Post Zr Zr Phase Segregation Fe Fe JR Scheffe, MD Allendorf, EN Coker, BW Jacobs, AH McDaniel, AW Weimer Chemistry of Materials 23 (8), October 211 S2P Toyota Visit Post-Processing 18

21 Perspective on Ion Transport 1-2 diffusion length 2 Dt Diffusion Length (m) FeO 127 C Fe 3 O C Fe in Fe 3 O 4 15 C Fe in FeO 11 C 1YSZ 1 C 1YSZ 15 C ceria 8 C ceria 15 C Heat Transfer D (m 2 /sec) t = 3 sec t = 1 min t = 5 min t = 1 hour Heat> Ceria > YSZ >> Fe 3 O 4 Ion (oxide) diffusion lengths are materials- and temperature-dependent. 19

22 The problem with ceria reaction extent G (kcal/mol) Temperature (ºC) Temperature (K) CO 2 thermolysis Hypothetical ideal NiFe 2 O 4-x Reduction NiFe 2 O 4-x Oxidation x=.5 x=.3 x=.1 From a thermodynamic viewpoint, ferrites are superior to ceria (larger d in target temperature range). DH and DS (DG) are functions of redox state (d or x). With each increment of reduction, materials become harder to reduce, easier to oxidize. G (kcal/mol) Temperature (ºC) CO 2 thermolysis Hypothetical ideal CeO 2- Reduction CeO 2- Oxidation Temperature (K) Miller, McDaniel, Allendorf Adv. Energy Mater 213. =.1 =.5 =.1

23 One Path Forward: Tailored MIECs for Thermo and Transport Sr x La 1-x Mn y Al 1-y O 3-d oxidize to split H 2 and CO 2 with lower T TR Comparable kinetics to ceria, but higher utilization. 9 more H 2, 6 more CO compound CO ( mole/g) H 2 ( mole/g) LSAM LSAM LSAM CeO 2-d cycle durability demonstrated A.H. McDaniel, Elizabeth C. Miller, Arifin, A. Ambrosini, E.N. Coker, R. O'Hayre, W.C. Chueh and J. Tong, Energy Environ. Sci., 213,6,

24 Take-home points For any approach to Solar Fuels- Efficiency is key for cost and scalability 1% solar to fuel minimum (lifecycle) Often it is unappreciated that sunlight is a high cost feedstock (capital cost) Low efficiencies increase scale, further challenge efficiency and stretch resources. CO 2 and water (and associated energy costs) are not limiting Thermochemical approaches have potential for high efficiency and thus high impact TE studies support eventual economic viability difficult, but not implausible Small global community has made significant advances in recent years Materials, Reactors, Systems all areas of opportunity and need All impact efficiency, all relatively immature for this technology. Adjacency to other technologies (e.g. solar electric, solar reforming) can help move technology forward, but focused cross-discipline efforts are also needed. Materials are challenging, but we have barely begin to explore the possibilities.

25 Thank You.