c

I

G

THE KALINA CYCLE AND SIMILAR CYCLES FOR GEOTHERMAL POWER PRODUCTION

C. J. Bliem

DE89 004875

Pub1 ished September 1988 EG&G Idaho, Inc.

Idaho Falls, Idaho 83415

DISCLAIMER

This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency thereof. nor any of their employees, makes any warranty, express or implied, or assumcs any legal liability or mponsi- bility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Refer- ence herein to any specific commercial product, proccss. or service by trade name, trademark, manufacturer, or otherwise docs not necessarily constitute or imply its endorsement, recom- mendation, or favoring by the United States Government or any agency thereof. The views and opinions of authors expr& herein do not necessarily state or reflect those of the United States Government or any agency thereof.

Prepared for the ' U. S. Department o f Energy

Idaho Operations Office Under DOE Contract No. DE-AC07-76ID01570

DISCLAIMER

This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency Thereof, nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof.

DISCLAIMER Portions of this document may be illegible in electronic image products. Images are produced from the best available original document.

4

The Heat Cycle Res rch'Program has as its primary ate temperature

5 . '

is similar to

standard supercritical cycles. These new cycles are considerably more complex than the simple Rankine cycle with a large number of recuperative heat exchangers. The percentage improvement with the Kal ina cycle was .greatest when working fluid mixtures having low performance in the simpler Rankine cycle were used. The supercriti advanced concepts

1 Rankine cycles with the in the Heat Cycle Research uoted in published results for ns with AKT Systems, Inc., the

company responsible for indicated that somewhat

ercialization of the Kalina cycle, have e possible with a water/ammonia mixture.

turbine system.

studied in References 2 and ormance appreciably, the additional

* . A new variation of

i

at most small ing fluid might well

show more substant ia l improvement h i s type of cyc le and the advanced Kal ina cyc le should be considered ser ious ly when advanced heat r e j e c t i o n systems are studied. I

I

Th o r t contains scussion of the mechanics Kal ina cyc le and ideas t o extend the concept

but l i t t l e o r no performance i

appl i ca t i on (360°F

0

b cycles. A modif ied cyc le which ha

1 Scat i on o f the

compared w i t h pub1 i shed r e s u l t sources and estimates about performance a t the geothermal temperatures. F ina l l y , the conclusions o f t h i s sco recommendations o f the d i r e c t i o n o f

I

!

.I

-

4

iii

ACKNOWLEDGMENTS

The author wishes t o thank a number o f people f o r t h e i r he lp w i t h t h i s repor t . F i r s t , the people a t INEL, G. L. Mines a t EG&G, Idaho, Inc. and K. J. Taylor a t DOE’S Idaho Operations O f f i c e were h e l p f u l i n t h e i r review o f the document. The discussions w i t h 0. 3. Demuth, o r i g i n a l l y w i t h and now a consul tant t o the Heat Cycle Research Program, gave i n s i g h t i n t o the problem and helped t o provide a more coherent way t o present the resu l t s . The. review by Dr . D. Y. Goswami o f the Mechanical Engineering Department o f North Carol ina Agr i cu l tu ra l and Technical State Un ive rs i t y gave an outs ide point-of-v iew. Last, the informat ion received from AKT Systems, Inc. and discussions w i t h Dr. M. Tribus and D r . H. M. Leibowitz who conveyed Dr . Kal ina’s th ink ing on t h i s subject have helped t o shape the f i n a l vers ion o f the repor t .

"

c

The Kalina Cycle has recently received a great deal of attention as a power production. The system being

f a conventional unrecuperated a bottoming system using the gas

rbine cycles having

Energy's Energy e 1). The first-ever Kalina cycle plant

ga Park, California. facility as its heat

that the Kalina cycle

le temperature and

at significantly lower

ccordingly, this

scoping study has considered three working f l u i d mixtures: ammonia/water ( the f l u i d used i n the h igh temperature Kal i na appl icat ions) , isobutane/heptane {one o f the best hydrocarbon mixtures w i t h conventional supercr i t i c a l cycles when used w i t and Refr igerant 22/114 ( the f l u i d studied i n the halocarbon study t h a t gave

3) . A mixture with a heavier component (Benzene/Propane) was a lso considered because it was f e l t t h a t a mixture w i th a wider b o i l i n g range might improve the performance.

1

hydrothermal Source a t 36OoF2)

comparable r e s u l t s t o those o f the isobutane/heptane hydrocarbon mixture k

.

-

2. THE KALINA CYCLE . -

e Rankine cycle that uses a alina proposes to use a 70/30 mass percent

vari ab1 e temper heat -source. Th

geofluid) reducing the irreversibil it

heated at pressures above the critical

temperature during condensing will be lower. This will result in possible 1 ower cool i ng wate ure changes in th condenser, fo

ure and, therefor higher cool i the parasitic power requirement to condense the

working fluid. possibly increase the size of the condensers over those in a conventional

lower mean temperature difference will result and may

power cycle.

Figure 2 shows a poss the distillation

encies do not

s it is a substantial fraction of the

VaDor

Initial heal exchange1

Recuperative heat exchanger

no. 2

- ~~

From turbine Possible he exhaust source fluid

1

I I I I I

i I -

-

I I I I 1 I I I I I -I

I I I I I I I I I I

I I

8-0916

1. From condenser :ondenser no. 1 no. 1

Typical Distillation Subsys Kafina Cycle

6

To condenser

addition side modifications incl ude:

expansion through the

cycles. (See Refe

er variation o

?-

I

J

8 0 al a

More volatile mixture

7 Less vo atile mixture

I .

I Condenser 1 I

I Coolant

Distillation subsystem

t

Primary heater

Secondary heater

The system shown in Figure 4 was considered briefly in this study; it 8

1 avoids some of the problems associated with the original Kalina cycle. First, the boiling phase change i s only a partial boiling process. In the original cycle, total boiling was necessary in countercurrent flow. This Mould be difficult to achieve with a reasonable size boiler because .of the low convection heat transfer coefficients when attempting to totally boil and dry out the tube walls. Second, this cycle would allow some heat

air. The original Kalina cycle rejects all of its heat at relatively low temperatures. (This results in higher thermal efficiencies, but gives no opportunity to decrease the cooling water makeup requirements for a wet cool i ng system).

I rejection at a high enough temperature to allow dry cooling to ambient 1

10

assess the potential o f F was chosen .as the heat

em will favor the Kalina

d changes throughout

11

system which was boiled as Kalina did with his bottoming cycle for gas turbines). The i sobutane/hep the best mixed hydrocarbons p (see Reference 2). The fluorocarbon mixture was the best mixture for those two refrigerants in a study reported in Reference 3, and mixture was that proposed by Kalina for the gas turbine

ammonialwater system was studied in Reference 7.

The thermodynamic properties for the hydrocarbon m obtained from the National Bu The refrigerant prope DuPont described in two Freon Technical

rams for Refrigerant Mixt

u of Standards computer determined from a package of codes from

an Institute of Gas Technology Rep0 No. 34 "Physical and Thermodynamic Pro Mixtures. n10 Entropies and superheate the data in this'report and property tables for pure water and pure ammonia assuming ideal gas mixtures (see Reference 7).

s

When no large gain was exhibited by these working fluids, a of propane and benzene was investigated. It was thought that this would have the wide boiling range exhibited by the original ammoni mixture of Kalina, but have a suitable critical temperature for th temperature application.

The second configuration (see Figure 4) was analyzed using normal hydrocarbon mixtures.

o f both the isobutane

ture i n the Kal ina .

bly larger. For the ird column ind tie results that Dr. Kalina

-

--_I . . ~~~ _ _ . _ _ . .. . . . . . _ . _ I

TABLE 1 - CYCLE COMPARISON WITH EVAPORATIVE CONDENSER (TWet Bulb = 6 o o F v Tgeof luid = 36OoF)

Turbine Outl e t

Geofl u i d Outl e t

)--L

P

*** Extrapolated From Other Cycles

increment in performance the Kalina cycle with this working fluid. First, ge is only around 32OF. I For the Kalina cycle rge amount of energy should be available in the

vely high temperature to effect the distillation,

e condensing

turbine exhaust

mixture). Secon system is already in the standard cycle and

work at its o

arge boiling range (lOO°F as with the water-ammonia

id effectiveness.

irectly from the aust, the turbine s an increase in

a net effectiveness increase

back pressure can be reduced

or the Kal ina c is noted in the

system allows an increase in turbine inlet pressure from 200 to 350 psia. The condenser pressure is reduced from 32 to 21 psia. These effects combine to increase the efficiency from 60 to 75%. given, the turbine work, pump work, and cooling parasitic can not be determined exactly.

this cycle can be achieved, the net geofluid effectiveness will be -10.6 w h/lb. This is an increase of 10 to 15% over an advanced supercritical cycle. The heater arrangement and the fact that more than one turbine is required will make, the Kalina cycle less acceptable for small systems; especially, well-head units.

Because only the efficiency is

The efficiency gives the difference in turbine and pump work. The cooling parasitic was extrapolated using the other cycles. if *

Thermal efficien.cies and second law efficiencies are given for each system. The thermal efficiency is the standard net power output divided by the heat input. The net work includes the turbine work as well as the working fluid pumping and heat rejection parasitic. efficiency is defined in the same manner as by Kalina.435 This is the net work output (excluding cooling parasitic) divided by the available energy (exergy) in the heat source stream referred to the average coolant temperature as the environmental temperature. This was performed to compare with the reported results for the Kalina cycle. The thermal efficiency is not very meaningful in this application. A high thermal efficiency is * obtained if less heat is removed from a unit mass of geofluid (therefore, at a higher average temperature). This would increase thermal efficiency at the expense of geofluid utilization that is not desirable. The second law efficiency incorporates the same ideas as the geofluid effectiveness that was used in the early geothermal ~ t u d i e s . ~ , ~ , ~ . Note that for the most part the second law efficiency of the isobutane/heptane and R-22/R-114 mixtures that were optimized without the Kalina modification are relatively high. .The Kal ina modification improves them slightly. improved the most, but still falls short of the performance of the original cycles .

The second law

5

The ammonia/water cycle is

I

0

The geofluid outlet temperature in all o f the Kalina applications

temperature that is 1 ilica in the geofluid, the

cycle application.

d did not show

.

ND RECOMMENDATIONS

gh the study may be drawn:

and scoping in

-

2.

3 .

the Kalina cycle, the use of geofluid as a he distillation subsystem is generally not very

relative to the better bi investigated. Although the turbine power can be incre

cycles presently being

cool i ng parasitic power i ncreas to partially gain. The required condenser s e’ may be incre

er heat load will probably be increased.

Large increments in performance by the additi cycle have been realized for working fluids w condensing ranges. Unfortunately, the w either have small condensing ranges (hyd fluorocarbon mixtures studied) or are not we1 r temperature range in a simple configuration (ammonia mixture). Even the propane/benzene mixture which ha - boiling range did not produce high performance for t appl i cat i on. The advanced Kal i na cycl e overcomes the disadvantages of the ammonia/water mixture in this temperature range, but at the expense of a complex heat addition system.

The combination of the Kalina cycle with different working fluids will give an extra degree of freedom in determining the best performance. There may be some fluorocarbon or hydrocarbon mixtures which combined with the Kalina cycle will give superior performance to all existing cycles studied, although it is suspected that it will be difficult to surpass the geofluid effectiveness of the standard supercri tical systems previously analyzed. There i s a certain amount of irreversibility inherent in a cycle and the cycles previously considered are close to that limit. The primary way to decrease the irreversibility is to find a better heat rejection system.

.

18

October 1984

*

. 2. 0. J. Demuth, -GTH-5753, February

3. C. J. Bliem, Cycles Usinu Mixed-Halocarbon Workinu Fluids, EGG-EP-7312, July 1986.

4. A. I. Kalina and H. M. Leibowitz, "Applying Kalina Tech Bottoming Cycle for Uti1 ity Combined Cycles", ASME PaDer 87-GT-35. Anaheim, Cal i forni a, June 1987.

5. A. I. Kalina and H. M. Leibowitz, "The Design of a 3MW Kalina Cycle

I

,

Experimental Plant. 'I

6. 0. 3. Demuth, Effects of VaDorizer and Evaporative-Condenser Size on I 1, EGG-GTH-6373, October 1983.

7. C. J. Bliem, Preliminary Performance Estimates and Value Analvsis for Binarv Geothermal Power Plants Usinq Ammonia-Water Mixtures as Workinq Fluids, EGG-GTH-6477, January 1984.

j

I I

I

I , 8, J. F. Ely, "Computer Code EXCST--Extended Corresponding States !

Theory," (Version 3.1), National Bureau o f Standards, National Enuineerinq Laboratory, Chemical Enqineerinq Science Division, Boulder Colorado. June 29, 1983.

,

The A.K. Texergy Co.

Systems, Inc. July 18, 1988

Mr. C. J. Bliem Energy Programs Idaho National Engineering Laboratory P.O. Box 1625 IdahoFalls, ID 83415

U

Dear Mr. Bliem:

It was good to talk with you and to leam about your work on the application of the Kalina cycle to geothermal sources. Because Dr. Tribus is traveling extensively, he has asked me to respond to your letter of 30 June 198

Over the past two ears we have developed six variations on the Kalina cycle, as shown in

cycles and assume that a cycle developed for one application will work well in another. For example, as you have found, when our System 1 or 6, which are designed as bottoming cycles, are applied to a geothermal source, they may perfoxm poorly.

Because our limited resources.are completely committed to design support for the

time, to proyide you with a design specifically tailored to your application.

We can, however, outline for you with a broad brush, what is required. Your Table I shows the turbine operating on a 70% mixture between 200 and 32 psia. After briefly reviewing your brine conditions, Dr. Kalina estimated that a well-designed geothermal plant (System 2), using the same 70% mixture, would operate between pressures of 350 psia and 21 psia, respectively. At this expansion ratio the second law efficiency would rise from 52.8% to 75%. We have several suitable designs in conceptual form but at this moment have not the time to demonstrate them to you. If you examine the data we have presented for System 6 (ref. 4) you will find that with a 70/30 mixture and an exhaust pressure of 21.9 psia it has a second law efficiency of 80%. This figure is more characteristic of Kalina type cycles when thev are oD!imized.

Our demonstration plant is based on KCS 1D2 (Kalina Q c l e &stem I Distillation subsystem 2) which is not a very efficient system. This system meets two criteria:

' Table I attached. P t is unfortunate that many people only know about one or two of these

demonstration power plant and certain follow-on business ventures, we are unable, at this J

1) It is the least expensive we can build. 2) It is the most we can afford.

Unfortunately, it only has a second law efficiency of 60%. That value should not be used 1 as the basis for an extrapolation.

24

22320 Foothill Boulevard Suite 540 Hayward, California 94541 Telephone 4151'537-5881 Fax 415/537-8621

Mr. C.J. Bliem

,

.

System

TABLE I

Application .E

OELY VARYING OF LOWEST POSSIBLE

,

EACH WITH A DIFFERENT FROM SEVERAL HEAT SO AT THE SAME TIME.

OLE TURBINE 1

, I I

, , ~

I

1

5.L

26