Development of Solar-powered Thermochemical Production of

Hydrogen from Water Al Weimer for the Solar Thermochemical

Hydrogen (STCH) Team17 May 2006

This presentation does not contain any proprietary or confidential information

Project ID No. PD10

OverviewTimeline• 6-25-2003• 12-31-2005• 40%

BarriersAU. High-Temperature ThermochemicalTechnologyAV. High-Temperature Robust MaterialsAW. Concentrated Solar Energy Capital CostAX. Coupling Concentrated Solar Energy and Thermochemical cyclesBudget

•Total Project Funding

$5,614,049 DOE

$752,900 Cost share

•Funds received in FY04

$2,943,232

PartnersThe University of Nevada, Las Vegas The University of Colorado The University of Hawaii

Sandia National Laboratories The National Renewable Energy Laboratory Argonne National

Laboratory General Atomics Arizona Public Service General Electric General Motors

ETH-Zurich MVSystems, Inc. Intematix, Inc.

Objectives

• Identify a cost competitive solar-powered water splitting process for hydrogen production

• Continue experimental cycle studies needed for final quantitative selection

• Numerical and experimental evaluation of solid particle receiver performance

• Optimize heliostat/tower/secondary concentrator characteristics and configurations for various operating temperatures

Approach

• Design and implement a quantitative comparative assessment methodology to screen all known thermochemical cycles and select the top several performers

• Perform literature surveys and laboratory experiments to acquire essential evaluation and design data for the top several concepts

• Develop validated designs for collector system components for integrated system analysis

• Analyze cost and efficiency metrics for integrated cycle performance

• Develop demonstration plant concept design(s) for surviving competitive cycle(s) and provide recommended path forward

Technical Accomplishments/ Progress/Results

• Cycle database and scoring are completed• Experimental work (kinetics, thermodynamics,

and product characterization) underway for five cycles

• Some system designs optimized for various operating temperatures and power requirements

• CFD modeling and simulation carried out to develop understanding of thermal transport in reactors and solid particle receivers

Discovery and assessment of known thermochemical cycles is completed• 353 unique cycles have been discovered and scored• 12 cycles found to be worthy of further experimental study• 5 of those 12 are currently under active study by SHGR

Other•Hybrid copper chloride

Volatile Metal Oxides•Zinc oxide•Hybrid Cadmium•Cadmium Carbonate Metal Sulfates

•Cadmium sulfate•Barium-Molybdenum sulfate•Manganese sulfate

Non-volatile Metal Oxides•Iron oxide•Sodium manganese•Nickel manganese ferrite•Zinc manganese ferrite•Cobalt ferrite

Sulfuric Acid (NHI)•Hybrid sulfur•Sulfur iodine•Multivalent sulfur

Metal Sulfate Cycles were Eliminated from Consideration

“Sulfate cycles”: MO + SO2 + H2O = MSO4 + H2No experimental evidence reported.

Thermodynamically, this reaction is favorable, but side reactions are more favorable.

SHGR studied sulfate cycles with MO = MnO, BaO under a variety of conditions:

25-100oC, makes MSO3 , No H2MO + SO2 = MSO3 for M = Mn, Ba

200-260oC, 100 atm, MnSO3 , No H2MnO + SO2 = MnSO3

230-240oC, 100 atm, BaSO4 + S, No H22 BaO + 3 SO2 = 2 BaSO4 + S

MnSO3*3H2O 7-20-05, 100 C, w500

0

40

80

120

160

200

240

280

320

10 15 20 25 30 35 40 45 50 55 60 65 70

2theta (deg)

MnSO3*3H2O powder 33-907 ref MnSO3*3H2O

We doubt that any “sulfate cycle” is feasible

X-ray diffraction shows MnSO3*3H2O

BaO + SO2 + H2O 230 C, 1440psi burst 9-22-05

0

50

100

150

200

250

300

350

400

450

10 15 20 25 30 35 40 45 50 55 60 65 70

2theta (degrees)

powder recovered BaSO4 24-1035

X-ray diffraction shows BaSO4

Two step ZnO/Zn Process to split water

Concentrated Solar Energy

H2 (product)

Metal Oxide Decomposition

Water Splitting

O2 (vent)

Zn (solid)(stored)

ZnO (solid)

H2O (vapor)

ZnO Zn + 1/2O2

Zn + H2O ZnO + H2

∆H = 557 kJ/mol@2300 K

∆H = -62 kJ/mol@700 K

Solar Reactor

24/7

(both steps demonstrated)

Aerosol ZnO Dissociation of ZnO

•Demonstration of rapid (< 1s) ZnO dissociation for Zn/ZnO cycle•Highest conversion ever (>40%) for thermal dissociation of ZnO•Conversion consistent with kinetic model

-50

0

50

100

150

200

250

300

350

400

450

0 50 100 150 200 250 300 350 400 450 500

Time (s)

Con

cent

ratio

n (p

pm)

OxygenCarbon MonoxideCarbon Dioxide

Feeding stoppedFeeding initiated

(ZnO Zn + 1/2O2; 1600 oC; Al2O3 tube)

9 cm ID x 46 cm longLab Transport Tube

Kinetic Models for Zn/ZnO CycleThermogravimetric Analysis performed to determine kinetic expression

( )23

0 1aE

RTd k edtα α

−⎛ ⎞⎜ ⎟⎝ ⎠= −

k0= 9.30 x 106 ± 0.3 x 106 s-1

Ea= 357.8 ± 12 kJ/mol

L’vov Theory:

160 102.9

/362

1−

∆

×==

=∆

=

− seFaBk

molkJHE

RS

Ta

Tν

ν

γν

o

o

-10.5

-10

-9.5

-9

-8.5

-8

-7.5

-7

-6.5

-60.00049 0.0005 0.00051 0.00052 0.00053 0.00054 0.00055 0.00056 0.00057 0.00058 0.00059

1/T (K^-1)

Arrhenius analysis yielded excellent agreement with prevailing kinetic theory

Hydrogen Production from Zn(Zn + H2O ZnO + H2)

0

2000

4000

6000

0 50 100 150 200 250 300time [min]

conc

entra

tion

[ppm

]

Zinc Feed

Water

Hydrogen

2000 ppmcalibration

360 oC 400 oC

Zn (mp)=419oC

Cadmium Carbonate Cycle might be highly efficient• Flowsheet calculations predict a steady state efficiency of

57.5%(LHV)/68%(HHV)

• High temperature reactions expected to be fast• Experimental program undertaken to investigate

hydrogen production step• High temperature studies required, especially quench

Cd + CO2(g) + H2O = CdCO3 + H2(g) 120oCCdCO3 = CdO + CO2(g) 350oCCdO = Cd(g) + 1/2 O2(g) 1450oC

Hydrogen Generated with Ammonia “Catalyst” • No hydrogen detected without a catalyst• Ammonia, supplied as NH4HCO3, promotes hydrogen

generation– Mechanism probably involves a Cd(NH3)n

+2 complex– Gas phase products determined by mass spectrometry– Final solid product is CdCO3

• Kinetics being studied– No significant change in hydrogen

generation rate between 80 and 140oC for large particle Cd

– Small particle studies underway

Sodium Manganese Cycle(500 mtorr TGA run @ 1310 oC)

-25.0

-20.0

-15.0

-10.0

-5.0

0.0

5.0

10.0

0 100 200 300 400 500 600

Time (min)

% D

elta

Mas

s

0

200

400

600

800

1000

1200

1400

Tem

pera

ture

(C)

3Mn2O3 --> 2Mn3O4+1/2O2

2Mn3O4 --> 6MnO+O2

Product Identifiedas MnO

Co-Ferrite lattice structure after 31 cycles

Test data showing repeatable hydrogen production over 31 cycles

Cast Part#2 - Disc Configuraton 75% PorosityYSZ:Co.67Fe2.33 O4 (3:1) Sintered at 1400C

Hydrogen Generation After 1st, 20th & 31st Cycle

0.00

0.02

0.04

0.06

0.08

0.10

0.12

0.14

0 25 50 75 100 125

Time (min)

H 2(V

ol%

)

Part#2, Cycle 20, 1248C, 3.3cc H2

Part#2, Cycle 31, 1250, 3.41cc H2

Part#2, Cycle 1, 1325C, 2.67cc H2

Current Ferrite Activities

•Evaluate reaction kinetics with on-sun testing

•Optimize reactive material composition

•Perform additional durability testing and characterization

Cast YSZ: Cobalt-Ferrite Lattice DemonstratesRepeatable Hydrogen Production

O2

H2O H2O

O2

x

yx

y

z

Reactive material

Insulation

Concentrated solar flux

Window

Sandia -Invented device uses recuperation of sensible heat to efficiently produce hydrogen in a two -step TC process

H2, H2O

Set of counterrotating rings

Counter Rotating Ring Receiver Reactor Recuperator(CR5) Thermochemical Engine

Hybrid Copper Chloride Progress

Proof of principle demonstrated for all steps in the reaction:

2Cu + 2HCl(g) H2(g) + 2CuCl (425oC)4CuCl 2Cu + 2CuCl2 (electrochemical)2CuCl2 + H2O(g) Cu2OCl2 + 2HCl(g) (325oC)Cu2OCl2 2CuCl + 1/2O2(g) (550oC)

Low temperature Cu-Cl cycle showed promising efficiency(~40% LHV),but some critical data must be determinedexperimentally

• Baseline System– 200m Tower– 100m2 heliostats

• 358 per field• 3 elliptical fields/tower• 3 focal lengths/field• ρ = 0.92• σslope = 1.3 mrad

– CPC Secondary• 23.5o acceptance angle• Cgeometric = 6.1

– Design, size and performance of secondary can be improved with further optimization

Ultra-High Temperature Advanced Tower ConceptMultiple Field & Secondary/Tower, Single Reactor/Tower

Annual Solar Performance

• Yearly Direct Daggett: 2787 kWhr/m2 (2679 kWhr/m2 available)• Yearly Delivered: 1990 kWhr/m2

• Annual Energy Delivered (to drive process): 259 MWhr/tower• Economies of scale for reactor/process components and infrastructure

provide significant cost advantage to advanced tower over dish systems for large scale plants

Based on Daggett TMY Solar Data

0.0

0.2

0.4

0.6

0.8

1.0

1200 1400 1600 1800 2000 2200 2400

Receiver Temperature, oC

Effic

ienc

y To

Pro

cess

Dish, C=7000

Dish, C=4000

Central Receiver

Baseline

Preliminary Solar Receiver ConceptT (K)

8000 W (3000 suns)Tmax = 2341 KTave = 2107 KX = 60.3%ZnO: 80 g/min

0

500

1000

1500

2000

2500

0 0.01 0.02 0.03

Pathlength (m)

Tem

pera

ture

(K)

Particle temperature along the tube with higher temperatures closer to the tube wall

ZnO: 2 g/min

(to be tested)

Solid Particle Advanced Solar Receiver for Thermochemical Processes

Lift

Receiver

Hot Tank

Cold Tank

HeatExchanger

Particle curtain

Aperture

•The solid particle receiver can achieve temperatures in excess of 950 oC

•Falling ceramic particles are directly heated by concentrated solar energy

•The complete solar interface includes two-tank storage and particle lift as well as a heat exchanger to couple the receiver to the thermochemicalprocess -Completed first level component design

-Began cold flow testing

CFD Analysis - Solid Particle Solar Receiver

• Operating condition for high outlet particle temperature (sulfur-iodine thermochemicalprocess, Particle temperature >1200 K)

• Decreasing the mass flow rate• Enlarging the particle size• Increasing the solar irradiation flux

air flow path lines

Particle track

Cavity efficiency as a function of particle size and particle mass flow rate. Solar irradiation flux : 920 kw/m2

Parabolic mirrorsInner reaction tube

Modified SiC insulating tube

Alumina insulation

Windowless Tubular Receiver

High Flux Solar Simulator

ETH-Zurich Solar Simulator

Future Work

• Experimentally resolve the uncertainties of the down selected cycles; close cycles

• Optimize system design for various temperatures and power requirements

• Develop and validate the transport mechanisms and the design of the solid particle receiver

• Update H2A process economics• Investigate materials challenges for solar receiver

and other system components