Development of a Molten-salt Thermocline Thermal Storage System for Parabolic Trough Plants

James E. Pacheco Steven K. Showalter William J. Kolb Sandia National Laboratories

Forum 2001: Solar Energy: The Power to Choose April 24, 2001

SOLAR

THERMAL

ENERGY

Sun Lab
Sandia National Laboratories, Albuquerque, NM National Renewable Energy Laboratory, Golden CO

Need for Thermal Storage for Trough Plants

Enables dispatchability without fossil fuel Provides load shifting Increases revenue from plants Allows higher solar fraction in combined cycle
(CC) plants

Improves economics of solar in CC plants Decouples collection from electricity production

SOLAR THERMAL ENERGY

Sun Lab
Sandia National Laboratories, Albuquerque, NM National Renewable Energy Laboratory, Golden CO

How Thermal Storage Can be Integrated into a Solar Plant

Direct Heat transfer fluid is the same as the storage media Examples: SEGS 1 (mineral oil), Solar Two (molten salt)

Indirect Heat transfer fluid is transferred to another media Examples: SEGS plant (synthetic oil transfers heat to molten salt in a two tank system).
Hot Tank Turbine Steam Generator

Estimated Capital Cost Of a Two-Tank Molten Salt System: $31/kWht Long-term Goal: $10/kWht
SOLAR THERMAL ENERGY

Oil-to-Salt Heat Exchanger

Thermal Storage Cold Tank

Condenser Oil Salt Steam/Water

Sun Lab
Sandia National Laboratories, Albuquerque, NM National Renewable Energy Laboratory, Golden CO

Collector Field

Indirect Thermal Storage System

Objective of Thermocline Development Project

Reduce the capital costs of storage for

parabolic trough plants relative to the baseline (indirect two-tank molten-salt system).

Address the technical risks associated with

thermocline storage:
Compatibility of filler materials with molten-salt Unproven concept Safety issues of fuel (Therminol) next to oxidizer (nitrate salt)
SOLAR THERMAL ENERGY

Sun Lab
Sandia National Laboratories, Albuquerque, NM National Renewable Energy Laboratory, Golden CO

Description of Thermocline System

HTF from Field

A thermocline molten salt system: - Uses a single tank to storage energy - Has a thermal gradient that separates the hot from cold fluid. - Uses a low-cost filler material to displace higher-cost molten nitrate salt.
Salt-to-Oil Heat Exchanger

HTF Return Thermocline Tank

SOLAR THERMAL ENERGY

Sun Lab
Sandia National Laboratories, Albuquerque, NM National Renewable Energy Laboratory, Golden CO

Development Activity Divided into 3 Areas

Model thermocline system:

and thermal cycling tests

Simulate performance and economics Isothermal

Evaluate candidate filler materials:

Test a small pilot-scale thermocline:

2.3 MWh capacity to validate performance and operating characteristics

SOLAR

THERMAL

ENERGY

Sun Lab
Sandia National Laboratories, Albuquerque, NM National Renewable Energy Laboratory, Golden CO

Modeling a Thermocline System

Transient thermal behavior simulated by the Schumann equations which describe the heat transfer between a fluid and a packed bed. Finite difference representation of these equations enabled calculations of the thermal gradient and inlet and outlet temperatures as a function of time. Also enabled calculation of extracted energy and capacity.
( C p ) f T f = (mC p ) f T f A y + hv (Tb T f )

SOLAR

THERMAL

ENERGY

Tb ( C p )b (1 ) = hv (T f Tb )

Sun Lab
Sandia National Laboratories, Albuquerque, NM National Renewable Energy Laboratory, Golden CO

Model Results
Charging
425 400 375 350 325 300 275 0.00
425

Discharging
In
400 375 350 325

Out

Out
0.50 1.00 1.50 2.00 2.50 3.00 3.50

300 275 0.00 0.50 1.00 1.50 2.00 2.50 3.00

In
3.50

Time, hours

Time, hours

18 16 14 12 10 8 6 4 2 0
SOLAR THERMAL ENERGY

0.0 hrs 0.5 1.0 1.5

2.0

2.5

3.0

290 300 310 320 330 340 350 360 370 380 390

Sun Lab
Sandia National Laboratories, Albuquerque, NM National Renewable Energy Laboratory, Golden CO

Temperature, deg C

Thermal Gradient During Charging

Screening Potential Filler Materials in Nitrate Salt

Seventeen commonly mined minerals and crushed rock were evaluated for their compatibility with nitrate salts The ideal filler material would have the following properties:
Inexpensive Widely available Low void fraction Compatible with nitrate salts Have a high heat capacitance
SOLAR THERMAL ENERGY

Chemical Name Aluminum oxide Aluminum oxyhydroxide Barium carbonate Barium sulfate Calcium carbonate Calcium fluoride Calcium sulfate Iron (II,III) oxides Iron titanate Magnesium carbonate Silicon carbide Tin oxide Calcium hydroxyphosphate Calcium flurophosphate Calcium carbonate Silicon dioxide

Mineral Name Corundum Bauxite* Witherite Barite Marble Fluorite Anhydrite Taconite Ilmenite Magnesite Carborundum Cassiterite Hydroxylapatite Fluorapatite Limestone Quartzite

Chemical Formula Al2O3 AlOx(OH)z BaCO3 BaSO4 CaCO3 CaF2 CaSO4 Fe2O3,Fe3O4 FeTiO3 MgCO3 SiC SnO2 Ca5(PO4)3 OH Ca5(PO4)3F CaCO3H2O SiO2

Sun Lab
Sandia National Laboratories, Albuquerque, NM National Renewable Energy Laboratory, Golden CO

Isothermal Tests of Filler Materials

Evaluated candidate materials in nitrate salts at 400 C. Samples were removed at 10, 100, 400 and 1000 hours of exposure. Samples were washed, weighed, and photographed. The salt was analyzed for contaminants. Results indicated the most promising materials were: quartzite, taconite, marble, NM limestone,apatite, corrundum, scheelite, and cassiterite.

SOLAR

THERMAL

ENERGY

Sun Lab
Sandia National Laboratories, Albuquerque, NM National Renewable Energy Laboratory, Golden CO

Thermal Cycling Tests

Evaluated how well materials held up to thermal cycling condition expected in a thermocline system. At least 350 cycles between 290 and 400 deg C were conducted on each sample. Samples tested: taconite, marble, NM limestone, quartzite, and silica sand. Results:
NM Limestone fell apart Marble softened and individual grains appeared to enlarge Taconite pellets held together well. Absorbed some salt in pores Quartzite rock and silica sand held up remarkable well and were selected as filler material for pilot-scale test
THERMAL ENERGY

Cold Tank

Hot Tank

Chamber
Thermal Cycling System

SOLAR

Sun Lab
Sandia National Laboratories, Albuquerque, NM National Renewable Energy Laboratory, Golden CO

Thermocline Test

Evaluate on a larger (pilot-scale) a molten-salt thermocline concept. Fabricated a 6 m tall by 3 m diameter carbon steel tank. Filled tank with a 2:1 mixture of quartzite rock and silica sand to a level of 5.2 m. Sodium nitrate and potassium nitrate were melted and added to tank. A propane heater simulated the heat input from the solar field (via the salt-to-oil heat exchanger). A forced-air cooler simulated heat rejection to the steam generator (through the salt-to-oil exchanger).
SOLAR THERMAL ENERGY

2.9 MW Propane Fired Salt Heater TC

FCV 201

25 ft
Carbon steel cold pump HV 201

HV 101 Stainless steel hot pump

TC

28 ft

22 ft
2.2 MW Salt to Air Cooler
Flow meter

17 ft
FCV 101

14 ft
TC
TC Flow meter

15 ft

3ft 3in

HV 301

7 in

TC

-14 in
Drain sump

Thermocline Storage Tank 10 diameter x 20 high

Sun Lab
Sandia National Laboratories, Albuquerque, NM National Renewable Energy Laboratory, Golden CO

Tests Conducted

SOLAR

Verification of heat capacity of system Evaluation of size and shape of thermal gradient Evaluation of change of shape of gradient over time Measurement of heat loss over time
Pilot-Scale Thermocline System
THERMAL ENERGY

Sun Lab
Sandia National Laboratories, Albuquerque, NM National Renewable Energy Laboratory, Golden CO

Thermocline Test Results

6.000 5.000 4.000 3.000 2.000 1.000 0.000 280
SOLAR

Capacity measured to be 2.44 MWh (slightly higher than 2.3 MWh design estimate.) Height of gradient during charging matches modeled profile. Gradient tended to become more tapered after 41 hours, but can still yield useful energy at a reasonable temperature potential. Heat loss was higher than modeled (likely due to heat loss through pump penetrations at the top (which werent modeled.)
7.0

13:30 13:00 12:30 12:00 11:30

6.0 5.0 4.0 3.0 2.0 1.0 0.0 41 hrs 31 21 11 0

300
THERMAL

320
ENERGY

340

360

380

400

280

300

320

340

360

380

400

Bed Temperature, Deg C

Temperature, deg C

Sun Lab

Profile during discharging and stagnant with heat loss

Sandia National Laboratories, Albuquerque, NM National Renewable Energy Laboratory, Golden CO

Economic Analysis for a 688 MWht Two-tank and Thermocline Systems

Component Nitrate Solar Salt, $k Filler Material, $k Tank(s), $k Salt-to-oil Heat Exchanger, $k Total, $k Specific Cost, $/kWh Two-Tank Molten Salt 11800 0 3800 5500 21100 31 Thermocline with Quartzite 3800 2200 2400 5500 13900 20

Thermocline Molten-Salt System is 65% the cost of a Two-tank Molten-Salt System

SOLAR THERMAL ENERGY

Sun Lab
Sandia National Laboratories, Albuquerque, NM National Renewable Energy Laboratory, Golden CO

Summary

A molten salt thermocline system has been

developed that is lower cost than a two-tank molten salt system.

Isothermal and thermal cycling tests showed

that silica sand and quartzite rock as well as taconite were compatible with nitrate salts.

The feasibility of a molten-salt thermocline

system was proven on a pilot scale 2.3 MWh storage system.
SOLAR THERMAL ENERGY

Sun Lab
Sandia National Laboratories, Albuquerque, NM National Renewable Energy Laboratory, Golden CO