NOTICE CONCERNING COPYRIGHT RESTRICTIONS

This document may contain copyrighted materials. These materials have been made available for use in research, teaching, and private study, but may not be used for any commercial purpose. Users may not otherwise copy, reproduce, retransmit, distribute, publish, commercially exploit or otherwise transfer any material.

The copyright law of the United States (Title 17, United States Code) governs the making of photocopies or other reproductions of copyrighted material.

Under certain conditions specified in the law, libraries and archives are authorized to furnish a photocopy or other reproduction. One of these specific conditions is that the photocopy or reproduction is not to be "used for any purpose other than private study, scholarship, or research." If a user makes a request for, or later uses, a photocopy or reproduction for purposes in excess of "fair use," that user may be liable for copyright infringement.

This institution reserves the right to refuse to accept a copying order if, in its judgment, fulfillment of the order would involve violation of copyright law.

(el/<

INTRODUCTION

HIGH TEMPERATURE UNDERGROUND THERMAL ENERGY STORAGE

R. E. Collins, J. R. Fanchi and G. 0. MorrellThe University of Houston, Houston, Texas 77004

K. E. Davis, T. K. Guha and R. L. HendersonSubsurface, Inc., Bellaire, Texas 77401

A major problem associated with the efficientutilization of energy sources is the storage of heat, in a

system that minimizes heat loss. One possible solution1-8

to this problem is to store the heat underground . Theunique aspects of the systems discussed in this paper are

their capability' of achieving thermal storage at suffi-ciently high te mperature and pressure, and with suffi-cient transfer rates, that electric power generation.on alargg commercial scale using heat from ' storage is

possible. The two methods under study aredeep aquiferstorage of very high pressure hot water and deep cavernstorage of hot oil utilizing solution caverns. in massive

salt deposits. Of these two methods the cavern storageappears to be the more feasible on the basis of our

preliminary studies.

Currently computer simulators are being developedfor detailed studies of thermal losses and pumpingrequirements associated with deep aquifer storage(superheated water, 650' F, 2700 psi) and deep cavern

storage. These studies will also address other aspects ofsuch geothermal storage systems, in particular the

solution and transport of minerals, in the ease.of aquiferstorage, and thermo-mechanical stresses on earth andwell co mponents. The present status of these ongoing

studies will be discussed below.

STORAGE SYSrEMS

In order to achieve maximum heat storageefficiency it is necessary to minimize. heat loss.

Previouslyl;2 it has been shown that deep underground

storage of thermal energy, using a working fluid at hightemperature and pressure, can be achieved with rela-tively small conduction losses for cyclic injection andwithdrawals if the system is sufficiently large. Inparticular, the asymptotic loss ratio is estimated as,

Heat lost per cycle = 31<rc (1)

Total Heat Stored R2

where K is the thermal diffusivity of the rock, TC is the

period of injection-withdrawal, and R is the equivalentradius of the cycliely heated region. Thus only larEestorage syste ms should be considered.

If the working fluid is to re main in the liquid phase,an important requirement for maximizing th9 energy

density, a high pressure in the storage region is neees-sary. This, in turn, dictates a deep storage zone if onlythe earth overburden pressure is to confine the fluid; in

·other words, the pressure in the storage region must be

less than the fracture pressure which, for high pressurestorage, can only be achieved at great .depths. A roughrule of thumb assigns 0.75 psi per foot of depth as the

fracture pressure.

Storage syste ms meeting the constraints necessi-tated by these considerati6ns can be realized by using

water as the working fluid and injection through a wellinto a porous aquifer or using oil as a working fluid and

injection · through a well into a solution eavern within a

massive salt deposit.

GEOLOGICAL FEASIBILITY

Geologically two types of aquifer storage• can be

considered: shallow, for aquifer depth ranging from 3000ft. to' 5500 ft.; and deep, for depths exceeding 5500 ft.The 5500 ft. division has been set based on a chargedaquifer pressure of 2700 psi: Areas where:the · thickness

of the sedimentary column above the basement rock( igneous and metamorphic rocks 'of all ages) are 'over

5500 ft. are potential regions for deep aquifer storage.Shallow aquifer storage is the same as: deep aquifer

storage except that the temperature and pressure of theinjected fluid will be comparatively less depending on

depth. This system is suggested where no other

alternative is available.

As with aquifer storage, two types of cavern

htorage' are distinguishable: solution cavity and 9;ceava-tion cavity. Artifieially created solution cavities insubsurface salt deposits are made using water to dissolve

the salt. The fact that most rock salt has a low porosity

and per meability, and exhibits se mi-plastic properties,tending to close small fraetures and openings made in it,

makes massive salt' formations ' ideal storage sites.Consequently, caverns in i. salt ; should be useful for

thermal storage using some'fluid other than water as the

working fluid.

Recent.advances · in drilling and mining, techniquesnow pemit the use of deeply buried rock cavities forstorage of water or steam. · Rock types suitable forconstructing underground exeavations are crystalline and

massive rocks. Dooley, et.al" have shown that econo-mics, environmental effects, and safety are favorable tothis thermal storage system · with power output being

quite satisfactory.

On the basis of our geological study• it has beenfound that underground storage of high temperature andhigh pressure fluid is geologically feasible in approxi-mately 80 % of the United States, with .the exception of

the West Coast and areas of mountain intrusions. Manyof the areas suitable for underground storage coincidewith regions of good solar insolation or sites of signifi-

cant heat supply from industries or nuclear power plants.

AQUIFER STORAGE

A mathematical ,model of, steam: injection into apermeable earth stratum containing brine has beenformulated and programmed.for computation on a highspeed digital computer to ,evaluate thermal losses,

' Ii \,16 2

thermal degradation of re trieved heat, and injection and

retrieval pumping requirements in various operationalmodes. The model is similar to the two-phase model of

Faust and Mercerlo, significant differences being that

here gravity is included and the model represents theaxially symmetric ,geometry of a single well. This

program is n6w in the final debugging stage.

In addition to the comprehensive numerical modeldescribed above a simple analytical calculation has been

peformed to provide a preliminary evaluation of theback-flow capability of an aquifer storage well. On thebasis of these calculations it appears that for a single

aquifer storage. well in an unbounded aquifer, pumping

would be required for backflow. Further calculationswith the d mple analytical model indicate . artesianbackflow is only possible fro m totally enclosed' aquifers;a result which diminishes the applicability of the aquifer

storage system. However, a downhole pump could beinstalled to achieve backflow conditions. This techno-logy is currently being developed in the ge6thermalindustry.

In addition to the backflow problem, the pr6blemofmineral solubility and scale formation, particularly sili-cate minerals, appears to be a major problem confrontingaquifer storage systdms. Silica scale· deposition,

especially if occurring in high te mperature, high pressure

steam turbines, could have detrimental economic effectsin brine systems associated with deep sandstone aquifers.

It is anticipated2, however, that through careful startuptechniques (i.e. repeated · flushing with pure hot water,chenical treatments, etc.) and the 6yclic nature of thesyste m whereby the retrieved water is reinjected afterheating, such scale problems can be minimized withsubsequent enhancement of the viability of aquiferstorage syste ms.

CONSTRUCTION OF SrORAGE CAVERNS

Solution caverns in salt domes have been used forstorage: of a variety of products .- from crude oil to

radioactive waste. Massive salt is an ideal storage

medium for bulk liquids because of its very low porosity

11and permeability . The essential geologic requirementsfor suitab16 and safe underground storage in salt massesare known and easily satisfied, for example, by the manysalt domes of the Gulf region of the United States. Thecavern construction requires an adequate supply of freshwater and an acceptable means of brine disposal. Freshwater is injected into the well and circulated,. causingdissolution of the salt to form brine. As this brineapproaches saturation, it is displaced from the' cavity byincoming water. The dissolution of salt and removal ofbrine creates the cavern. Using these techniques, manyeaverns of multi-million barrel capacity have been

12developed .

The actual construction of a salt dome hot oilstorage eavern would require approximately 9 months,

with the well being constructed in about 60 days and the

remaining -time spent creating the eavern. : The welldesign ig relatively complex in that it would be installedunder ambient conditions and operate at temperatures in

excess of 600: F.'' This. creates tremendous stresses onthe cement to easing bond, as well as the cement to.

formation bond. Experience with steam'stimulation ofoil wells indicates that these problems cah be overcome

through the use of special cements and expansion joints

, A eavern storage well would be configured as aseries of.concentric pipes set at various depths as in

conventional injection wells. The first pipe is called theconductor. pipe and is generally driven into place with.adiesel hammer. ., The conductor pipe usually extends to

approximately:100 feet below ground level and protectsthe shallow fresh water aquifers as well· as supplying aconduit for the drilling fluid to return to the mud tanks.The second pipe installed is the surface easing. This

easing is set below the base of the fresh water aquifersand prevents their contamination from drilling fluid, saltwater, and ·possibly.oil and gas. For a heat storage well

this easing would be cemented to the surface with a high

quality heat resistant cement. The next pipe installed is

the protection easing. It is usually set and cementedapproximately, 200 feet into the salt. This easingprotects the caprock and other salt water aquifers.

There. will.=be two additional pipes installed, howeverthese will not' be. permanently fixed with cement. Theinnermost tubing string will be used to obtain the propercavity configuration by moving it up and down during the

leaching process. The other easing will be used toprotect the roof of the salt cavity. A hydrocarbon suchas kerosene or diesel fuel is pldeed in the annular spacebetween this easing and the protection easing. Thisprevents water contact with' the salt cavern roof andthus prevents it from being leached.

After the cavern is developed, the free hangingcasings will be repositioned so that the innermost easingis near the , bottom of the cavern and the other easingnear the top. The hot oil would be pumped into one andthe resident fluid displaced through the other.

The configuration of the cavity can be confirmed

through the use of a sonar. log. From this data, we can

calculate.,the configuration of the 'cavity and its exact

volume.

j Once the well' and cavity are'drilled and completed,the maintenance problems and costs will be , minimal.

However, routine' checks on the integrity of the wellsshould be made ' at least once , a year. This can be

accomplished with various geophysieal logging techni-ques.

CAVERN STORAGE OPERATION

We have used a simple mathematical model tostudy the 6peation of a cavity heat storage system. Theeavern is partially filled with gas (nitrogen) which iscompressed during injection. Expansion of thecompressed gas forces the fluid out during retrieval. Forpreliminary studies of the thermal• losses and pumpingrequire ments the eavern' is approximated as a sphericallysymmetric heat source embedded in an infinite earth of

uniform-.thermal diffusivity. Numerical integration ofthe heat' conduction equation and a coupled heat balanceequation for the cavern is used to compute the tempera-

ture, the heat· loss rate and the pressure as functions of,time.

-:A complete. eavern storage system would consist.ofthe high temperature .eavern,.an additional cold fluid

reservoir, which might also be a eavern, and appropriateheat exehangers'and circulating equipment.

The optimum operational parameters for a eavernstorage system have not yet been determined, but

on the outer. easing.

63

preliminary numerical results have been generated basedupon the parameters in Table I using the eomputersimulator described above. These parameters correspondto a heat transfer rate of 74 megawatts. The assumedoperational proeedure eonsisted of displacing the brinefrom the completed eavern with nitrogen then injeetinghot oil, compressing the nitrogen, and subsequentlyinitiating cyclic injection and withdrawal of hot oil.

TABLE I. Cavern Operation Parameters(These parameters correspond to a transfer rate

of 74 MEGAWATTS.)

Q = 2000 gal/min Injeetion flow rateT = 480 min. Injection PeriodT.. = 648'F Injeetion TemperatureinJTo = 130° F Ambient Temperature

of Rock Salt2K = 0.00533 em /see Thermal Diffusivity of3 Rock Salt

Co Po = 0.49 cal/em ' C Specific Heat Densityof Oil

CAVERN STORAGE SIMULATOR RESULTS

Minimum heat losses from the eavern are ofcritical importanee in eavern heat storage. The lossratio, defined as the ratio of the heat lost per eyele tothe total heat stored in the cavern, must be kept small,The loss ratio is defined by the equation

Heat Loss Rate �042TLoss Ratio - e (2)

Total Heat Storedwhere the heat loss rate and total heat stored aredetermined by computer simulator and Te is the timepdriod of one storage cycle. The loss ratio as a functionof time is shown in Figure 1 for a 100 ft. radius cavernand a one day storage period. The different curveseorrespond to initiating cyclic storage and retrieval atdifferent tim es. The loss ratio at 70 weeks of cyclicopeation is still well above the limiting, asymptoticvalue as given by Equation (1) while the eavern tempera-ture has risen to between 20 F and 100 below injectiontemperature. However, the loss ratio does fall to areasonably small value after just a few weeks, there-after it decreases more slowly with time while caverntemperature continues to rise slowly.

We also determined the ratio of the heat lost percycle to the heat input per eyele given by

Heat Loss per Cycle = Loss Rate = T e (3)Heat Input pe Cycle f Q Co po(Tinj - To).

Values of this ratio at the end of 6ne year of cyclicoperation are plotted in Figure 2 versus the dimension-less ratio of oil volume transferred pe eyele to totaleavern volume. The heat lost per cycle is seen to beless than 4% of the heat input per eyele for any cavernradius less than 100 ft. at an injection flow rate of 2000gpm. The asymptotic value of the ratio of the heat lostper cycle to the heat input per cycle goes approximately,as R/Q. Therefore, the minimum cavern size consistentwith desired storage performance should be used in orderto minimize heat losses and provide for maximumelectric power production.

The ideal fluid for use as in a high te mperature,eavern storage syste m must have a high boiling point,high te mperature stability, reasonable cost, and a rea-sonably high value of specific heat (per unit volum e) tomaximize the heat stored per volume of fluid. Ofcourse, these fluids must not dissolve the salt as

10.0

• 8.0 -

%• 7.0 -

• 6.0 -P0i 5.0

%2 4.0-li• 3.0.0J

R c =IO O f t.25 % GAS CAP

+ I, -0 10 20 30 40 50

TIME (WEEKS)Figure 1. Loss Ratio as a function of time.

4.000X

w W 3.03 -100ttI z 2.0-WWa.a..1-:3M 0- 1.0 -0Za-

3 13ZZO

--,460 70

I0 0.05 0.10 0.15 0.20 0.25

Q T/ 34 11 R•, DIMENSIONLESSFigure 2. Ratio of heat lost per eyele to heat injectedper eyele after 52 weeks of eyelie storage as a functionof the ratio of oil injected per cycle to eavern volum e.

CAVERN STORAGE WORKING FLUID

9.0 074TrRB·064XIO-2C] 3NO. CHARGED TEMP.hNJ.TEMR

I 0.65C' 0 0.75

M 0.85rr 0.95

THEORETICAL LIMITAT EQUILIBRIUM

2.0

-"1.0 -

-

-

...I

. 64

continued cavern growth is undesirable. Some commer-cially available fluids being considered are listed in Table

II along with values for so me of their properties.

13TABLE II. Working Fluids

Liquid UsuableTemp.

° F

Dowtherm A 55 to 750

Therminol 44 60 to 700

Dowtherm G 12 to 650

Therminol 66 25 to 650

Humbletherrm500 -5 to 600

Mobiltherm600 -5 to 600

Therminol 55 25 to 600

650

650

BoilingPoint

° F

Specific Heat Costx Density $/gal

cal/em3° Cat 1000 F/7000 F

495 .405/.440

572 .430/.467

.370/.497

720-950 .412/.476

.365/.494

/.428

FUTURE CONSIDERATIONS FOR CAVERN STORAGE

Some of the topics now being studied include:1. - The gas pressures and pumping requirements for

various operating conditions.2. The effects of shutdown on heat losses from the

storage system.3. Feasibility of multi-day storage and seasonal stor-

age.4. Thermo-mechanical stresses on the well, salt, and

above ground equipment.5. Introduction of gravel or sand into the cavern in

order to mduce the volume of storage fluidrequired and possibly raise the density of heatstored.

6. Requirements for the cold fluid storage reservoir.7. Above ground equipm ent for cavern storage opera-

tions.8. . Economic considerations, especially in regard to

best operating conditions, overall scale, and sur-face equipment required.

9. Environmental considerations.

ECONOMIC CONSIDERATIONS

An aquifer storage well capable of handling 2000gpm would cost approximately $1,200,00.00. This pricedoes not include the support surface facilities such as

pumps, heat exchangers, etc. However, this equipmentwould be similar, if not the same, for a cavern storage

system. A cavern storage well would cost approximately$700,000.00 to drill and complete, $100,000.00 to leach

out the cavity, and $400,000.00 for a disposal well todispose of the salt water created during the leachingprocess. Thus the total cost for a cavern well is also$1,200,000.00. A 75 foot radius cavern filled with gravel

would require about 2.5 million gallons of oil. At$.60/gallon this volume of oil would cost $1,500,000.00.

Operational costs will probably be lower for thecavern storage system than for an aquifer storagesystem. We. anticipate that an aquifer well will have tobe pumped in most cases. Also, there is a strongpossibility that there will be a silica scale problem

associated with an aquifer storage well. These factorswould add considerable costs to aquifer storage systems.

.61

6.60

.67

.72

The preliminary studies completed to date indicate

that underground thermal storage can be interfaced.witha yariety of high te mperature heat generating systems,e.g. nuclear power plants and producers of industrialwaste heat as well as large central focus solar collectors.Furthermore, much of the technology necessary for the

design and construction of underground thermal storagefacilities already exists. Consequently the undergroundstorage of thermal energy, particularly cavern storage,

appears to offer a promising near-term method of storingheat at temperatures high enough to permit reasonably

efficient electric power generation.

REFERENCES

1. Collins, R.E.; "Geothermal Storage of SolarEnergy," in "Governors Energy Advisory CouncilReport, Project NT-3," A.E. Dukler, ExecutiveDirector, The Univ. of Houston; Houston, Texas

(Nov. 1974).

7 2. Collins, R.E. and K.E. Davis; "Geothermal Storageof Solar Energy for Electric Power Generation,"Proc. Int. Conf. on Solar Heating and Cooling,Negat Veziroglu, Ed., University of Miami (Dee.

1976).(in press).

3. Meyer, C.F. and D.K. Todd; "Conserving Energywith Heat Storage Wells," Environmental Sei. &Tech. Vol. 7,#6 0973).

4. Meyer, C.F. and W. Hausz; "A New Concept inElectric Generation and Energy Storage," presentedat Frontiers of Power Tech. Conf., Oklahoma State

U. (October 1975).

5. Meyer, C.F.; "Status Report on Heat StorageWells," Water Resources Bulletin, Vol. 12, #2 (1976).

6. Dooley, J.L., G.P. Frost, L.A. Gore, R.P. Ham-mond, D. L. Rawson, and S.L. Ridgeway; "AFeasibility Study of Underground Energy Storage

Using High-Pressure, High-Temperature Water,"ERDA Contract No. 5(04-3)4243, RDA-TR-7100-001, National Technical Information Service, U.S.

Dept. of Commerce, 5285 Port Royal Road, Spring-field, VA. 22162 (Jan. 1977).

7. Davison, R.R., W.B. Harris, and J.H. Martin;"Storing Sunlight Underground," Chemtech (Dec.1975).

8. Tsang, C.F., C.B. Goranson, M.J. Lippmann, and

P.A. Witherspoon; "Modeling Underground Storagein Aquifers of Hot Water from Solar PowerSystems," · presented at Int. Solar Energy Soe.(American See.) "Solar World" meeting Orlando,Florida (June, 1977).

9. Guha, T.K., K.E. Davis, R.E. Collins, J.R. Fanchi,

and A.C.Meyers III; "Manmade GeothermalEnergy", Proc. of Miami Int. Conf. on Alternative

Energy Sources, Negat Veziroglu, Ed., University ofMiami (Dec. 1977) (in Press).

10. Faust, C.R. and J.W. Mercer; "Finite-Difference

Model of Two-Dimensional, Single, and Two-Phase

Heat Transport in a Porous Medium -Version I,"U.S.G.S. Open-File Report 77-234 0977).

11. Halbouty, M.T.; "Salt Domes Gulf Region, UnitedStates and Mexico", Gulf Publishing Company0967).

CONCLUSIONS

.13

65

12. Third Symposium on Salt. Vol. I & II, Editors: J.L.

Rau and L.F. Dellwing, The Northern Ohio Geolo-gieal Society, Inc., Cleveland, Ohio (1970).

13 Bundy, F.P., Herrick, C.S.,and Kosky, P.G,; "The

Status of Thermal Energy Storage", General Elee-tric Company, Corporate Research and Develop-ment, P.O. Box 8, Schenectady, New York 12301,Report No. 76CRD041 (April 1976).

t

66