Concentrated Solar Power on Demand CSPond: Solar Harvesting and Storage Charles Forsberg Department of Nuclear Science and Engineering (NSE) Massachusetts Institute of Technology 77 Massachusetts Ave; Bld. 42-207a; Cambridge, MA 02139 Tel: (617) 324-4010; Email: cforsber@mit.edu November, 2011 CSPond Faculty: Alexander Slocum (ME), Jacopo Buongiorno (NSE), Charles Forsberg (NSE), Thomas McKrell (NSE), Alexander Mitsos (ME), Jean-Christophe Nave (ME) CSPond Students: Daniel Codd (ME), Amin Ghobeity (ME), Corey J. Noone (ME), Stefano Passerini (NSE), Jennifer Rees (ME), Folkers Rojas (ME)
Joint Mechanical and Nuclear Science and Engineering Project Shared Liquid-Salt Technology Base 2 Solar beam Aperture (closes at night or in bad weather) Ground Molten Salt Pond Salt to/from salt in loop at steam generator L D Z Graphite Insulation Fluoride-Salt High- Temperature Reactor (FHR) Concentrated Solar Power on Demand (CSPonD)
3 Outline Existing solar systems CSPond Base Case Design Experimental Validation Alternative Design Options Path Forward A. Slocum, J. Buongiorno, C. W. Forsberg, T. McKrell, A. Mitsos, J. Nave, D. Codd, A. Ghobeity, C. J. Noone, S. Passerini, F. Rojas, Concentrated Solar Power on Demand, J. Solar Energy
4 Existing Solar Power Towers Mirrors reflect sunlight to boiler Boiler tubes on top of tall tower absorb light Heat water and convert to steam Steam turbine produces electricity Poor economics High capital cost Low thermal efficiency PS-10, 11MWe peak, image courtesy of N. Hanumara
5 The Challenge is Cost Low efficiency system In theory: high efficiency In practice Low steam temperatures to avoid boiler-tube thermal fatigue from variable light Wind and sunlight always changing energy fluxes High heat loses from exposed boiler tubes High costs Mirrors Largest cost component Incentives for efficient light to electricity system Tall tower PS-10, Spain, 11MWe peak, image courtesy of N. Hanumara
6 CSPond Base-Case Design
7 CSPond Characteristics Combining Many Technologies in a New Way Concentrated solar thermal power system Built-in thermal storage Heliostat field similar to solar power tower Radically different light receiver to: Boost light-to-electricity efficiency Provide thermal heat storage Unique features: Light volumetrically absorbed in liquid salt bath Salt bath could operate to 1000 C
8 CSPond Description Figure Next Page Mirrors shine sunlight to receiver Receiver is a high-temperature liquid salt bath inside insulated structure with open window for focused light Light volumetrically absorbed through several meters of liquid salt Building minimizes heat losses by receiver Enables salt temperatures to 900 C Small window minimizes heat losses but very high power density of sunlight through open window Power density would destroy conventional boiler-tube collector Light absorbed volumetrically in several meters in salt Requires high-temperature (semi-transparent) salt Similar salt requirements as for FHR heat transfer loop
9 Two Component System Non-Imaging Refractor Lid Lid Heat Extraction Hot Salt to HX Cold Salt from HX Light Reflected From Hillside Heliostat rows to CSPonD System Light Collected Inside Insulated Building With Open Window (Not to scale!)
10 Advantages of Hillside Heliostat Field Eliminate tower-based receiver heavy equipment on ground Avoid remote storage and high pressure pumps Downward focused light Potentially lower land costs
11 CSPond Light Receiver Efficient light-to-heat collection Concentrate light Focus light through small open window Minimize heat losses Challenge Light energy per unit area very high Will vaporize solid collectors Cold Salt from HX Non-Imaging Refractor Lid Lid Heat Extraction Hot Salt to HX Light Volumetrically Absorbed in Liquid Salt Bath
Light Focused On Transparent Salt Light volumetrically absorbed through several meters of salt Molten salt experience Metal heat treating baths (right bottom) Molten salt nuclear reactor Advantages No light-flux limit No thermal fatigue Can go to extreme temperatures Molten Chloride Salt Bath (1100 C) 12 12
Salt Vapor Condenses On Ceiling 13 Cooled ceiling: Lid Heat Extraction Salt buildup until liquid salt layer with flow back to salt bath Self-protecting, self-healing ceiling Highly reflective Non-Imaging Refractor Lid Cold Salt from HX Lid Heat Extraction Hot Salt to HX
Capture Efficiency: Energy Balance Q in System Losses: Q rad, lid-aperture Heat to lid: Q conv, pond-lid Q rad, pond-lid Q vaporization Q reflected 14 Q conv, lid-aperture Q rad, pond-aperture Q tank Q salt η capture = Q system Q in = Q in Q losses Q in = Q in (Q rad, l-a + Q conv, l-a + Q rad, p-a + Q tank ) Q in
15 Two Classes Of Molten Salts Appearance of molten NaCl-KCl salt at 850 C Near-term: Nitrates Used in some concentrated thermal energy solar systems Off the shelf Temperature limit of ~550 C (Degradation) Longer-term: Chlorides and Carbonates Thermodynamically stable Peak temperatures > 1000 C
1.0 System Design Enables Efficient Light Collection and High Temperatures 16 fraction of incident energy at aperture 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 0 500 1000 1500 2000 2500 3000 3500 NREL (2003) NREL Solar I: steam 500C NREL Solar II: salt 565C Peak & average values Nitrate Salt: 550C lid temp 240C Aperture to Illuminated Pond Area ratio Peak Concentration (kw/m^2, 'suns') Chloride Salt: 950C lid temp 660C Compares favorably with measured values for CSP Power Tower Systems system output (MWe): 4 nominal pond size diameter (m): 25.0 depth (m): 5.0 avg beam down angle (deg): 21.4 Nitrate Salt, Lid peak temp (C): 550/240 Chloride Salt, Lid peak temp (C): 950/660 16/29
17 CSPond Integral Heat Storage Salt tank has insulated separator plate Plate functions Separates hot and cold salt Bottom light absorber Storage role If excess heat input, plate sinks to provide hot salt storage volume If power demand high, plate raised with cold salt storage under plate Daytime Divider plate (moves down) Divider plate (moves up) Nighttime
4 MW e System Sizing: 2500 m 3 salt 40 hours storage Virtual Two-Tank Concept Hot salt Divider plate (moves down) 18 5m 28m Daytime = charging Cold salt 24/7 hot salt as the average temperature of the tank decreases when the sun is not shining Divider plate (moves up) Nighttime
System Performance Nitrate (550/240) Chloride (950/660) Overall System Useful Energy Salt Useful Energy Lid Useful Energy system output (MWe): 4 nominal pond size Uses for lid heat: Low temp (Nitrate) Power cycle pre/reheat RO feedwater heat MED feedwater heat High temp (Chloride) Power cycle primary heat diameter (m): 25.0 depth (m): 5.0 avg beam down angle (deg): 21.4 Nitrate Salt, Lid peak temp (C): 550/240 Chloride Salt, Lid peak temp (C): 950/660 Lid α vis /ε ir 0.44 Low-temp heat rejection (C): 25 19
20 CSPond Experimental Testing and Analysis
Molten Salt Optical Characterization 21 Solar Irradiance Attenuation of NaCl- KCl (50-50wt%) salt at 850 C NaNO 3 -KNO 3 (60-40wt%) (l) Variable optical path length transmission apparatus NaCl-KCl (50-50wt%) (r) Appearance of molten NaCl-KCl salt at 850 C Experimental Range Stefano Passerini, Dr. Tom McKrell, Prof. Jacopo Buongiorno, MIT
60-Sun Solar Simulator 22 MIT CSP Solar Simulator 10.5 kw e Calculated Optical Power (kw/m 2 ) Spectral Intensity (arbitrary units) 80 70 60 50 40 30 20 10 MIT CSP Simulator 7x 1500 W metal halide lights Terrestrial solar spectrum 60 kw/m 2 peak 38 cm aperture Adjustable: wavelength (nm) 300 400 500 600 700 800 900 1000 height (0-1 m) ε n = α solar = 0.11 angle (0-90 tilt) ε n = 0.05 α solar = 0.11 ε n = 0.06 α solar = 0.14 Commercial Xenon Solar Simulator 0 0 2 4 6 8 10 X = radial offset from aperture center (cm) x
Volumetric Light Absorption 23 1 Height (x/l) 0.9 t = 0 0.8 8.3 h 0.7 0.6 Increasing Increasing time time 0.5 0.4 0 h salt 0.3 0 h tank 2.8 h salt 0.2 2.8 h tank 0.1 8.3 h salt 8.3 h tank 0 200 220 240 260 280 300 320 340 360 Temperature ( C) 2.8 h Temperature distribution of NaNO 3 - KNO 3 (60-40wt%) heated optically
Virtual Two-Tank System Testing 24 Salt tank @ 300 C with nitrate salt Ceramic (high temp) insulation Fiberglass insulation Inside Solar Simulator
Divided Thermocline Storage 25 UP DOWN Temperature distribution of NaNO 3 -KNO 3 (60-40wt%) heated optically Enables 24/7 power 25
Tank Wall Design 26 Flexible alloy liner Reduces thermal shock in refractory lining Internal firebrick insulation allows for mild steel tank shell Flexible protective liner made of AISI 321H stainless steel from Kolb (1993) and Gabbrielli (2009)
27 Solar Flux Distribution Modeling Incident Reflected Beam-down angle = ϕ θ i θ r θ t Transmitted θ i = 90 φ θ r = θ i n salt sin θ = n t air sinθ i R s R p sin( θ t θ i ) = sin( θ t + θi ) tan( θt θi ) = tan( θt + θi ) 2 2 Reflected intensity 1 CSPonD beam-down systems are in this range Flux distribution in receiver from a single central heliostat ( )/ 2 R = R s + R p 0 0 45 90 Beam-down angle
28 CSPond Alternative Design Options
29 Heliostat Field Placement Options Mirrors to Hilltop Collector Tower Reflects Light Downward Hillside Mirrors to Collector
30 Multiple Power Cycle Options Salt Temperature: 500 C, 700 C, and 700+ C Steam power cycles--today Supercritical carbon dioxide power cycle High efficiency Very compact and potentially low cost Advanced technology Air Brayton power cycle Existing technology No cooling water options Requires 700 C salt temperatures
Carbon Dioxide Properties Result in Very Small Equipment Main compressor wheel: 85kW 31 Manufactured by Barber & Nichols for SNL
32 50-MWe Power Conversion Unit Small Units Shop Fabricated
33 Air-Brayton Power Cycles Air Brayton power cycles have low cooling requirements relative to other power cycles Viable at salt peak temperatures of ~700 C Significant efficiency penalties at lower temperatures Several different options
34 Stack Open-Air Recuperated Brayton Liquid-Salt Power Conversion Cycle Air Inlet Recuperator Heater salt Reheater salt Liquid Salt Air Compressor Turbines Generator
Stack Open-Air Brayton Liquid-Salt Combined Cycle Power Cycle 35 Heat Recovery Steam Generator High Temperature Salt Air Water or Steam Air Inlet Heater salt Reheater salt Reheater salt Compressor Turbines Generator
36 Comparison of Brayton Power Cycles 700 C 1 Salt; 100 MW(t) Plant Cycle Air Brayton Combined Cycle Efficiency 40% 44% Condenser Heat Rejection* None (No water requirement) 28 MW(t) 1 Efficiency drops rapidly with peak temperature 2 Traditional closed power cycles (Steam, Carbon Dioxide, Helium) with 50% efficiency reject 50 MW(t) to Condenser
37 CSPond Status Patents Pending
38 Two Parallel or Sequential Paths Forward Small 100 kw systems integration test using nitrate salts Uses proven existing solar salt Rapid testing possible Develop higher-temperature chloride or carbonate CSPond Higher efficiency with potentially lower costs Robust against salt degradation Follow-on integration test with different salts
Next Step: 100 kw t Research System 39 salt loop to/from HX Lid geometry T.B.D. for 1-bounce down Overall receiver size: ~ 4m dia x 3m (w/o lid) Hot salt Salt Tank: 2m dia x 2m depth 6.2 m 3 salt capacity (11.2 metric tons) Nitrate salt (550C/275C) 15 h storage (1.5 MWh) 2.4 MWh daily solar input required for continuous operation Cold salt Not shown: aperture cover, concentration booster, lid heat rejection system and divider plate actuator
Next Step: Alternative Salts Insufficient Data for Non-Nitrate Systems 40 Solar Irradiance NaNOAttenuation 3 -KNO 3 of NaCl- KCl (50-50wt%) (60-40wt%) salt at 850 C NaCl-KCl (50-50wt%) Higher temperature salts more robust (no possibility of thermal decomposition) Higher efficiency with open air Brayton power cycles and no water requirements Experimental Range Stefano Passerini, Dr. Tom McKrell, Prof. Jacopo Buongiorno, MIT
41 Conclusions Analysis and experiments indicate significantly better economics than existing concentrated solar power-tower systems (Higher efficiency) Significant uncertainties (Path forward) No small pilot plant under realistic conditions Limited review (Wider review underway now that patent filings complete) Large incentives for higher-temperature salt than nitrate (more robust system and dry cooling) but limited experimental data Large incentives to determine commercial viability of CSPond
42 Questions 42
43 Biography: Charles Forsberg Dr. Charles Forsberg is the Executive Director of the Massachusetts Institute of Technology Nuclear Fuel Cycle Study, Director and principle investigator of the High- Temperature Salt-Cooled Reactor Project, and University Lead for Idaho National Laboratory Institute for Nuclear Energy and Science (INEST) Nuclear Hybrid Energy Systems program. Before joining MIT, he was a Corporate Fellow at Oak Ridge National Laboratory. He is a Fellow of the American Nuclear Society, a Fellow of the American Association for the Advancement of Science, and recipient of the 2005 Robert E. Wilson Award from the American Institute of Chemical Engineers for outstanding chemical engineering contributions to nuclear energy, including his work in hydrogen production and nuclear-renewable energy futures. He received the American Nuclear Society special award for innovative nuclear reactor design on salt-cooled reactors. Dr. Forsberg earned his bachelor's degree in chemical engineering from the University of Minnesota and his doctorate in Nuclear Engineering from MIT. He has been awarded 11 patents and has published over 200 papers.