_I - -- -_____

i his report was prepared as an account of w o ~ k sponsored by an ageocy of the lJniEsd States:Govctilment Neithat theunited StatcsGoverrai,ient norany agency ihereoi, nor any of thcrr employees, makes any warranty, mpress or implied, or assumes any legal liability or responsibility for the accuracy, completcnrss, or usefulness of any inforn-raiion, apparatus orodiict, or process disclosed, or represents that i ts use would not infringe privately owasd rights 3cference herein to any spcc~frl: commercial product, process, or s e ~ ~ ~ i c e by trade name, trademark, rnanufactuict , or othceisriae, ~ O P S not necessarily constitute or imply its endorsement, recornmendation, or favoring by fire United Sta?cs @over nment or any agency thereof Tne vicsals and opinions of authors nxpccsscd herern do riot necessarily state or rcfivct those of the United Sta:es GOVCii-I:iiejat or any agency thetcoi

- - _________ -~

ORNL/TM-10492

Energy Division

COMPUTER TION HEAT PUMP USIN OMIDE AND

TUlEiES

. A

d

OAK ORATORY O a 37831

MARTIN STEMS, INC.

O f f i rograms U, NERGY

under 840R21400

CONTENTS

LIST OF FIGURES . . . . . . . . . . . . . . . . . . . . . . . . . . . . v

ACKNOWLEDGMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii

EXECUTIVE SUMMAEtY . . . . . . . . . . . . . . . . . . . . . . . . . . . ix

ABSTRACT.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi

2.INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

2. EDIA PROPERTIES . . . . . . . . . . . . . . . . . . . . . . 2 IONS DESCRIB X - T D A T A . . . . . . . . . . . . 2 IONS DESCRIBIN PY NTRATION DATA . . . . . . . . . . . . . . . . . . . . . . . 4 ITY, S P E C I F I L CONDUCTIVI . . . . . . . . . . . . . . . . . . . 4

3. COMPUTER SIMULATION. . . . . . . . . . . . . . . . . . . . . . . . . 10

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 ISON BETWEEN L i B r ER UA. . . . . . . . . . . . . . . . . . 16

OF SOLUTION ON COP OF TNM . . . . . . . 18 ENCE OF ABSORB C E . . . . . . . . . . . . . 23

TNM WITH EQUAL

5. CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

REFERENCES. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

A P P E N D I X . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31

iii

LIST OF FIGURES

e-composition-te data (Duhring chart) eous L i B r and a t e mixtures . . . . . . . . . .

era ture

e ra ture LiBr-water m . . . . . . . . . . . . . . . . . . .

aqueous t e r es . . . . . . . . . . . . .

i t r a t e mixture . . . . . . . . mal conducti r e s . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

3

5

3 t comparison and ternary t e mixtures . . . . . . . . . . . . . . . . . . . 1 3

4 a t i c representat io l a t ed s ing le - s t a r a tu re heat pump a t e points . . . . . . . . . 15

tu re boost as t e heat and water tempera ternary m i x t u r e s . . . . . . . . . . . . . . . . . . . . . . . . . 1 7

6 son of pure r e solut ion t e and absorb or L iBr and n i t r a t e mixt . . . . . . . . . . . . . . 1 9

7 son of pure re t e solut ion rate and absorb fo r L i B r and

n i t r a t e mixt . . . . . . . . . . . . . . . . . . . . . 20

8 t of increasing xchanger surface on COP fo r terna u r e s . . . . . . . . . . . . . 21

9 f increasing s o l u t i exchanger surface on flow r a t e for ix tures . . . . . . . . . . 22

10 r i son of COP-lift a heat re lease orber UA fo r t e i s one-half L i B r . . . . . . . . . . . . . . . . . . . . . . . . . . . 24

on of COP-lift and t release when absorber UA f a r t is one-third

a t o f L i B r . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

ACKNOWLEDGMENTS

L

Appreciation is ex Stephen Kaplan, Hr, Richlen, and man (Haifa Te

r support an ation is a 1 for t yp in LeRoy GLlliam Sharon McCona s , and E. W. orial skills.

i a l Programs. document was prep U.S. Department of Ene

v i i

EXECUTIVE SUKMARY

s report presen a computer si at comparing th mance of 11th nary nitrate aqu n a temperature a

In this study, the fallfn heat transfer coe

e-third. Due to a ed thermophy nitrate mixtur be lower than th

t e s relied on e results show t h nitrate mixture may be op

, which is a er than wha ated with

potential for 1 than L i B r .

mental measu i n g Eilm heat per t i e s are r

nitive investiga

i x

ABSTRACT

A new aqueous ter ting of 53 wt % and 19 w t % N

temperature

a computer prog pump with LiB

ncentration-

mixtures as he two f l u i d ere is a regio omparison ca cal data o in the temp In the abse thermophys ical

uess first app w that the t

a 10% advantag advantage in t s L i B r at hi ernary nitrat lization at emperatures an

e l y i n the l o w a t temperature r erforinance a ure is suffic

d corrosion data.

xi

ION

1

2

2. WOWING MEDIA PROPERTIES

The pressure-composition-temperature (P-X-T) and the specific enthalpy-composition-temperature (H-X-T) data for the working fluids are critical pieces of information for heat pump cycle calculations. In the case o f LiBr aqueous mixtures, these thermodynamic properties are well documented between 50 and 350°F. The specific enthalpies of pure refrigerant (water) and steam are available in steam tables. In the case of the ternary nitrate mixtures, Ally has represented the exper- imental B-X-T and H-X-T data by polynomial equations similar in form ts s the ones published €or LiBr in the ASHME Handbook of Fundamentals. These equations €or LiBr and TNMs will be shown later with their atten- dant Duhring and specific enthalpy plots. The main facility provided by these equations is that they can easily be incorporated in a c~mputer program e

5

2.1 EQUATIONS DESCRIBING I?-X-T

The P-X-T data for TNMs and LiBr mixtures are superimposed on one another in Fig. 1. The relationship between the saturation temperature of the pure refrigerant (pure water in each ease) and the temperature o f the mixture of known concentration, X (wt % ) , can be represented for either working fluid by an equation of the form,

t = A(X)tf + B(X) , (1)

where t and t t are the mixture and pure refrigerant saturation tempera- tures expressed in degrees Fahrenheit, respectively, and A axid B are polynomial functions of the concentration, X. Of course, the functional dependence of A and B on X will be different for the two mixtures as is seen by their respective expressions in Fig. 1. The equilibrium vapor pressure of water above the mixture is given by a van’t Hoff type of equation,

E 2 ’ + - D p ~ c + -...-

(tt + 4 5 9 . 6 7 ) (t’ + 4 5 9 . 6 7 ) Loglo

where C, D, and E are constants that must be the same f o r both mixtures because they share a common refrigerant between them. The reasons why E q . (2) is used for both L i B r and TNMs are explained in detail elsewhere. 5

rERNARY SALT

5 2 19 wt B NaNoa

for

4

2.2 EQUATIONS DESCRIBING SPECIFIC ENT

The specific H-X-T data and correlations for aqueous LiBr are reproduced from the ASHFZAE Fund entals Handbook in Fig. 2a. Figure 2b show3 the H-X-T data for the ternary nitrate mixture. The experimental data are correlated with an equation of the form,

4

w =: a ( X ) I- /3(X)t , ( 3 )

where cr and /3 are polynomials in the solute (ternary sa l t ) cancentra- tion, X (wt % ) , and t is the solution temperature. Details regarding this correlation are re orted in ref. 5.

2.3 VISCOSITY, SPECIFIC GRAVITY, AND THERMAL CONDUCTIVITY DATA

The narrow scope of thermaphysical and transport data available in the literature and the sparse amounts o f data available for the ternary nitrate mixture were a serious handicap during the preparation of this report. Viscosity, specific gravity, thermal conductivity, etc., data €or aqueous LiBr mixtures are available up to 100°C (212”F), but proper- ties at temperatures up to 180°C (356°F) were needed for this study. In the absence of experimental data, estimated values were obtained by extrapolation (Figs. 2c,d,e). Perhaps the only source of data far the aqueous ternary nitrate fluid is ref, 3 , which contains adequate data on viscosity, lacks considerable data on specific gravity, and is severely deficient on thermal conductivity, for which only two values of 0.31 and 0.33 W/rn2-K are quoted at salt concentrations of 86.7 wt % and 8 3 . 7 wt % , respectively. These va ues are themselves obtained through a recommended power law equation. The lack of specific gravity data is not very severe because thermal expansion,

4

is small over the temperature range of interest. But the error on ther- m a l conductivity could deviate substantially from the estimated value of 0.31 to 0.33 W/m2*K.

5

60

5 0

40

30

20

0 5 0 60 70 0 io

N , WEIGHT PERCENT

2 . (a) S p e c i f sition-temperature ium ater mixtures.

35

E

W ( X , f ) = a(X) + P ( X ) t

4.516 - 8.649

x 10-5)x2

t = SQLUTIQN TE

F i g . 2 . ( b ) Specific enthalpy-composition-temperature fo r aqueous ternary n i t r a t e mixtures.

7

8

10.0

8.0

"0

4.0

n u)

ro 2.0

W

c 8 3

cn 1.0

0.8

0.6

>

0.4

0.2

F i a . 2 . ( d ) Viscos i ty of L i B r and ternary nitrate mixture.

9

0.2 I I I I I I I I I 1 2 0 20 4Q 120 140 1

RE ("C)

F i g . 2 . ( e ) Thermal. conductivity of L i B r mixtures.

Operation o f a single-stage temperature amplifier heat pump was simulated using a program developed by Grossman and Michekson.

The criteria for evaluating absorption fluids from a practical standpoint are (a) the coefficient of performance, (b) the temperatmre lift and boost temperatures, (c) the flow rates of refrigerant, (d) potential corrosion problems, ( e ) crystallization limits, (f) film heat transfer coefficients, ( g ) mass transfer rate of sorbate into sor- bent, and (h) M area per unit capacity, The coefficient of perfor- mance, refrigerant flow rates, and film heat transfer coefficient are directly related to the size of a particular heat pump unit and are therefore reflected i n capital costs. High COP, heat, and refrigerant transfer rates reduce heat pump size. Corrosion by working fluids is a strong determinant of hardware useful life expectancy, and its reduction OK elimination is obviously preferred. Crystallization is another important operating parameter because it determines the acceptable range of concentration and temperature of the circulating fluid before onset of crystallization. Once crystallization BCCUKS, the system ma be shut down. Mixtures in which crystallization occurs at law ternpera- tures are generally preferred over those in which crystallization occurs at higher temperatures.

First, consider the heat transfer coefficient. The absorber i s the focus of attention because it is here that the sorbate (refrigerant) is absorbed into the sorbent, thereby releasing heat. At least three resistances to heat transfer are involved between the refrigerant vapor and the process stream being produced, the largest of which is resis- tance due ta the film resistance between the sorbent and the tube wall. The other two resistances are the resistance to heat transfer offered by the metal tube wall and that between the wall and the utility stream. These three resistances in series are combined to yield a single overall heat transfer coefficient, U as shown below,

6

0

1 II_

9

DO xw Do 1 *o - +- E- Dihi km r, 0

+ -- --

L

where D and D are the outside and inside diameters of the tube, x i s the tube wall thickness, k is the metal thermal conductivity, D is the logarithm mean diameter of the tube, and ho is the film heat transfer coefficient. The term, represents the resistance to heat transfer

0 i - F B

M L 1

hO

sorber tubes, st two term the f i l m he

e value a€ esis tance . at transfer

ause it implie

than 2000 eous beca

termination efficients i t be determi

ue to the de alue of f i l fficient was u nitrate f l u i approximation. ablished the heat t ransfer

data . The s procedure , Due to the v , and ther the film heat t t for the te

d to be about ha1 lu t ions . Th

a t t ransfer co ated using e

ing the Q V ~ r coefficient n as one-half d a t h i r d set e scenario heat transfer

s t ransfer coe ure of the r:

e the r e f r i ct with the ed will be equilibrium v se that the

the rate of m l y during the orbent and re er. Suppose t h ncentration i

of m a s s t r then equilibr

is absorbed t. If t h i s rat

rbed will be. 1 mount, resuE pressure, eac solut ion (eq nt ra t ions eo erent temper and t h e i r di as subcoolin

ent way to e viscosities nary n i t r a t e s ,

LiBr solution ow t h a t subco g an effect r additive for

used in type I T S . Without an

12

additive, LiBr mixtures typically have 10°C or greater subcooling for low-pressure absorbers. Due to lack of experimental data, the extent of subcooling cannot be quantified; but as explained above, sorbates of higher viscosities will absorb sorbe ts slower than those lower viscosities.

Based on the potential COP and temperature l i f t s shown in Fig. 3 , 10°C of subcooling Ear the ternary nitrate s approximately equal to the advantage that this f l u i d has over lithium romide in terms of tempera- ture lift capability.

Since actual heat, mass transfer, and subcooling data are n o t yet available for the ternary nitrates and since the computer model used in this study does not account €OK any subcooling, the simplifying assump- tion sf no subcooling for the t e r n a ~ g ~ nitrate and LiBr mixtures is used.

having

13

L i B r and tern

1 4

4 . RESULTS

A schematic of the single-stage temperature amplifier heat p u p considered far the simulation study is shown in Fig . 4 . The waste heat source input to the desorber and evaporator at positions 10 and 1, respectively, is stea at a prescribed (input) temperature. This waste heat source temperature is an input parameter in the program which can arbitrarily be changed. The waste heat streams leaving the desorber and evaporator at positions 11 and 2, respectively, is saturated water at

e pressure as their ~ e s p e ~ t i v e inlet streams. Therefore, no pressure gradient is assumed between state points 10 and 11 and 1 and 2, respectively, and all the latent heat of vaporization from the waste stream i s supplied to the desorber and evaporator. In the condenser, cooling water at a prescribed temperature is introduced at 1 3 and exits at 14. The temperature at 1 3 and the cooling water mass flow rate are fixed. In the absorber section, the process stream enters at state paint 3 at a prescribed temperature at its saturated condition (condensed water). As it goes through the absorber, it picks up suffi- cient heat to leave the absorber at 9 as saturated steam, The purpose of the recuperator is to make the absorption cycle more efficient, and the effectiven s s of the recuperator plays a major role in determining the cycle COP.

The overall heat transfer coefficients times area (UA) in each of the five primary equipments used in the simulation appear in Table 1. These were obtained from a prototype test unit at QRNL used to monitor the performance of LiBr aqueous mixtures.

8

Equipment UA

Btu/rn"F W/K 10-3

Desorber 166.7 5.27 Recuperator 8 3 . 3 2 . 6 3 4 Condenser 500 15.81

Ab s orb e r 216.7 -+ 108.3 -3. 7 2 , 2 6.852 -+ 3 . 4 3 + 2 . 2 8 Evaporator 8 3 3 2 6 . 3 4

A cursory inspection of Fig. 1 shows that there is a small region o f overlap between LiBr and ternary mixture solutions, At temperatures below 65°C (=150"F), the TNMs will certainly present crystallization problems, but LiBr is still suitable for heat pump applications without imminent threat of crystallization. It is believed that LiBr/M,O may be used at higher temperatures and concentrations than have been reported.

15

c

ORNL-DWG 85-

-I-

4

4 . Schematic the simulated e heat pump e p o i n t s .

EM LIB% l! B EQUAL ABS0

T h e computer simulated performanees of % i B r and TNMs for two sets o f waste heat and cooling water temperatures are shorn in Pig. 3 . Con- sider the case where the waste heat and cooling water temperatures are 105°C (221°F) and 55°C (131"F), respectively, and the absorber UA is 6852 W/K. With LiBr, one can $0 from 10 to 42°C lifts because the con- centrations and crystallization curves on the Duhring chart (Fig. 1) allow it, With TNMs, the lower lifts of less than 3k"C are not possible because the smallest concentration in the Duhring plat is 70 wt % , and the crystallizatian curve is very near. Hence, with TNMs the range of temperature lifts varies from 34 to 52°C. Similar arguments apply to the case of the waste heat and cooling water temperatures being 140 and 70"C, respectively.

Considering the coefficient of performance, a camparison of the ~ U K V ~ S in Fig. 3 shows that the COP values are higher fer TNM than they are f o r LiBr. However, the question of how much of an advantage in COP is afforded by the TNM over LiBnr becomes quite mbiguous as the tenpera- t u ~ e lift is varied. At low l i f t s ( 3 7 to 4O"C), the predicted COP of the TNM is about 10% higher than that f o ~ LiBr, assuming no subcooling. A s the lift is increased from 42 to S 2 " C , the COB for the LiBr mixture drops off to ze ro ; whereas that far TNM keeps decreasing but is, nevertheless, a finite value until a lift of 52"C, when it too becomes zero. Therefore, between 42 and 52°C temperature lift, the COPS of the TNM are infinitely greater than that of LiBr. The culprit is the pre- cipitous fall in COP with temperature lift because AC approaches zero. Therefore, a valid region of temperature lifts f o r comparison of COPS for the two fluids is one ranging from low lifts up to the point where the precipitous fall in COB begins to OCCUK. Based on the simulation study and in light of the above discussion, it may be appropriate to say that the predicted COP of the T is approximately 10% higher than that for LiBr mixtures.

Two further queries arise from the results shown in Fig. 3 . They are (i) why are the temperature lifts higher f o r TNMs? and (ii) why are the COPS higher than they are f o r LiBr?

The answer to the first query is fairly straightforward. In Fig. 1, the s lope of the equilibrium vapor pressure versus solution tem- perature plot i s flatter for the TNMs than those for LiBr, and this flatness helps to reach out Eurther in the solution temperature field. From the Duhring chart, it is an easy matter to determine the maximum boost temperature possible (AC = 0 ) for a given waste heat and cooling water temperature. This information is shown in Fig . 5. It can be seen from Fig . 5 that in the comon operating temperature regimes, the tem- perature boos t and, consequently, the temperature lift of the TNMs are potentially greater than they are for LiBr mixtures.

The second question, relating to higher COPS for the TNMs, does not seem to have an overtly obvious answer. Since the definition o f COP is Q /Q + Q,, higher COP values clearly indicate higher numerator values A D

110

90

18

with respect to the denominator, Q, + Q,. If the values of QD -+ Q could be held Constant, then the higher COP values are associated wit!

A higher Q values. Then one could investigate the factors affecting Q and from there be able to draw definite conclusions regarding the two working fluids. However, the computer program6 used in the simulation study did not permit the desired 'bevel of freedom in fixing Q and Q,, and could only approximately be held constant f o r the t w o working fluids. The results are shown in Fig. 6 . For waste heat at 140°C (284°F) and a condensing temperature of 70°C (158"F), the dilute solu- tion flow rate (curve 2) for the two fluids is almost identical in the region of overlap. The refrigerant flow rate (curve 1) is slightly higher for the TNMs than it is for L i B r , and this explains why the absorber capacity, Q, (curve 3 ) , is also higher. The same behavior is obta ined f o r a different set of waste heat 105°C (221°F) and condensing water 55°C (131°F) conditions as shown in Fig . 7.

A s mentioned earlier, the primary purpose o f focusing attention on the ternary nitrate mixtures i s because they have the potential to operate at up to 260°C (500°F) with minimal corrosion of mild steel.

A

D they

4 . 2 EFFECT OF SOLUTIQN HI AT EXCHANGER UA ON COP OF 9:

Et is well-known that the coefficient of performance of a heat pump is sensitive to the effectiveness of the solution heat exchanger, other things being equal. The effectiveness increases with the heat exchange surface area. Figure 8 shows the improvement in COP by increasing the surface area in the solution heat exchanger from 7 . 7 4 m2 to 17.0 m2. The waste heat source and cooling water temperature are 140°C (284°F) and 70°C (158"F), respectively. The COP increases by approximately 32% from its value of 0.31 corresponding to a solution heat exchanger area of 7.74 m 2 - Figure 9 shows what effect the change in solution heat exchanger has on the refrigerant. and dilute solution flow rates and the absorber heat release.

The mast dramatic change i s observed in the dilute solution flow rate as the solution heat exchanger area is increased from 7.74 m2 to 1 7 . 0 m 2 * The pure refrigerant too decreases, but the decrease is less than 10%. The significance of this is that the concentration difference between the concentrated and dilute solutions is greater when A = 7 . 7 4 m2 than it is when A 5 17.0 m2.

The change in solution heat exchanger area also influences the total. quantities o f heat delivered to the evaporator and desorber and that rejected in the absorber, Since there is only a 10% change in the pure refrigerant flow rate, the heat input to the evaporator also changes in the same proportion. The desorher and absorber heat increases with increasing solution heat exchanger area. The net result is higher COP as shown in F i g . 8.

C

- 1

8

-*- LITHIUM BROMIDE < - TERNARY NITRATES/H,U

CONDENSING WATER 70°C (158OF)

4 1 O C (19.8OF) AT IN EQUIPMENT -

i on f l o w ra te and absorber heat release for L i B r and ternary n i t r a t e mixtures.

20

F i g . 7 . Comparison of pure re f r igerant , d i l u t e solut ion f l o w r a t e and absorber heat release f o r L i B r and ternary n i t r a t e mixtures.

2 1

h

. Effect of heat excha ternary nitr

22

t

23

e computer simu d s a t i s f a c t o r maximum a t s region is s

that the d i f ation between ed stream a

was not possible t h i s l i m i t a t

hes its maxi oftware in t

4 . 3 NCE OF ABSORB CE

e case of th discussed i n Se ransfer coe d t o be about

and down t o se i n the era11 heat t actual decv er r e s i s t an i t s inf luen n t . For the

t o deal with

cipated decrease t r a t e is s long w i t h the fts below rature <180°C ncentration € w t % , f o r whi Lithium bromi

COP of 0 . 4 , a oost temperat ra ture) o f u

ty. At lift d the absorbe

the absorber ween the t

r i n favor o f TNM has p ro t o bel ieve t h er coe f f i c i n lower. To i

the absorber UA was - t h i r d tha t ted r e s u l t s f o r f l u ids a re

fo r 11 f t s 540 her COP than

er absorber

e r COP than re i s su f f i c i en t

e conclusions t o . 10 and 11 are t h a t (i) L i B

COPS and highe

24

ORNL-DWG 878450

8 0.4 k- z w 2 0.3

0" 00 G o

u.

8 - 0.2 h

$- 0.1

\

--- LITHIiJ

.-.*p TERNARY NITRATE, UA = 6.85 x io3 W / K - TERNARY NITRATE, UA = 3.43 x io3 W / K

WASTE HEAT :I- 140°C: (284°F) - WATER :::I- 7OoC (158°F) ENT - 11 " C (20°F)

HEAT RELEASE

-

EXTRAPOLATION

.I-

O 10 20 30 40 50 60 70 80 90 100

TH - TM, TEMPERATURE LIFT ("C)

7 e-% c .- E

6 3 A W

QY I

' E

0 3 X

F i g . 10. Comparison o f COP-lift and absorber heat release w h e n absorber UA for ternary nitrate i s one-half t ha t of L i B r ,

0.5

0.4

0.3

0.2

0.1

0.

OFM-DWQ 878451

I I I i I 1 -(I- LITHIUM BROMIDE, UA = 6.85 X IO3 1 W/K I

n c (216.7 Btu/min. O F )

% - TERNARY NITRATE, UA = 6.85 X IO3 W/K -/ 6 ‘E

EXTRAPOtATiON ER THAN 70 wt %

UNAVAILABLE

26

heat release through the absorber at lifts greater than 40°C and high boast temperature using w a s t e heat at 140°C (284°F) even when the absar- ber UA is assumed as low as one-third its value for L i B r , and (iii) the film heat transfer coefficient is a strong determinant o f the p e r f o r - mance and needs to be known accurately.

27

NS

U ~ O U S ternar containing a s

s than LiBr rature rang era11 heat nt is reduc Br. In lo etitive w i

ure than has both fluids cap overcome

ired i n futur or both flui sorber film icient and t e critical ation which

t to the othe advantage tha

29

1.

2.

3.

4 .

5 .

6 .

7 .

8.

9 .

10.

B. A . Phillips, '' d Residential

ch a n d Devel

F. C. Hayes and R. J ation of Adva

onference, R ent on Hea

Laboratory, A

W. F. Davidson and ew High Tempe Sub/85-22013/1,

, 1985 Fundame Society.of Hea and Air-Condi , pp. 17.71-1

of Aqueous n t t o C h e m i L/TM-10258,

ssman and E. ion Heat P s , P a r t 1: S imu 1 a t ion RNL/Sub/83-4 ational Lab

t Blab, Chem. , 95-105 (1977) .

ez-Blanco a l e and Perfo t ion Heat P National Lab

rmann, Personal

R. C. DeVault, Persona. S.

31

APPENDIX

Lithium Bromi

solution con

=: 1.63 x 5

e viscosity, v

3 ckness, 6 - Reynolds Nllmbe

e, 6 = 3.206

(Fig. 2e) =: 0

t t ransfer c

ernary Nitra

e r solution tern olution conc

tic viscosity, Y =

kness, 6 - eynolds Number

, 6 = 4.159 x

1 conductivity o == 0.33 W/meK

transfer co 7 9 3 . 4 W/m2.K %

33

1-5, M. 6 , V. 7 . R . 8. R . 9 , F.

21. M. 22. R. 2 3 . F. 24. J .

BUTION

R . Ally H. A. McLain D. Baxter V. C . Mei

D. R. Miller M. Olszewski

. H. Perez-B1

. B. K. Seiver R. €3. Shelt J. J. Tomlin C. D. West T. J. Wilban Central Rese Document Re

C. Maiensch OWL Patent

49. W. J.

UTION

Bierman ve, Fayettev

J . M . C a l m , E arch Ins t i tut 1 0 4 1 2 , Palo A

5 essor of Corn tsburgh, PA

52. G. Douglas Ca lley Authori

53. s Office, U.S.

54. D . C. Erickson, C o . , 627 R i d

lls, I D 8 3 4 0

Annapolis, MD

55. R . J , Fiskum nergy Conve ranch, CE-1 , U.S. Depa 000 Indepen hington, DC

34

56. J. Gilbert, Dames 6 Moore, Inc . , 7101 Wisconsin Avenue, Suite 700, Bethesda, MD 20814

57. S . M. Gillis, Dean, Graduate School, Duke University, 4875 Duke Station, Durham, NC 27706

58. G. Grossman, Technion-Israel Institute of Technology, Faculty of Mechanical Engineering, Haifa, Israel

59. W. T. Hanna, Battelle Columbus Laboratories, 505 King Avenue, Columbus, OH 43201

60. F. C. Hayes, Trane Company, 3600 Pamel Creek Road, LaCrosse, WP 54601

61. F. R. Kalhamer, Vice President, Electric Power Research Institute, B.U. Box 10412, Palo Alto, CA 94303

62. A . Karp, Energy Management Q Utilization Division, Electric Power Research Institute, 3412 Hillview Avenue, P a l o Alto, CA 94303

63. R. E. Kasperson, Professor, Government and Geography, Graduate School o f Geography, Clark University, Worcester, MA 01610

64. R. Macriss, Institute of Gas Technology, 3424 South State Street, Chicago, IL 60616

65. L. A . McNeely, 7310 Steinmeier Drive, Indianapolis, IN 46250

66. D. K. Miller, Borg-Warner Air Conditioning Inc., P.O. Box 1592, York, PA 17'405

67. J. I. Mills, EGG-Idaho Inc., P . O . Box 1625-WCB, Idaho Falls, ID 83415

68. R. L. Perrine, Professor, Engineering and Applied Science, Civil Engineering Department, Engineering I, Room 2066, University o f California, Los Angeles, CA 90024

6 9 . B. A . Phillips, Phillips Engineering Co., 721 Pleasant Street, St. Joseph, MI 49085

7 0 . R . Radermacher, University of Maryland, Department o f Mechanical Engineering, College Park, MD 20342

71. R. C. Reinmamn, Carrier Corp., P , O . Box 4808, Syracuse, Ny 13221

72. S . L. Richlen, Office of Industrial Programs, CE-141, 5F-O35/FORS, U . S . Department of Energy, 1000 Independence Avenue SW, Washington, DC 20585

35

73. U. Rockenfelle Company, 1453 Road, Boulder City,

7 4 . J. D. Ryan, 0 and Comuni CE-132, GF-21 rnent of Ene Independence on, DC 2058

75. D. S. Severso titute, 8800 Avenue, Chica

a 76. S. V. Shelton hanical Engi Institute of , GA 30332

77. J. J. Tuzson, tute, 8800 We Avenue, Chicag

78. G. C. Vliet, T fversity of Te TX 78712

79. PI. Wahlig, La oratory, Univ

fice of Assi Energy Resear

California, B

ent, DOE-ORO, 30

P.O. Box 6 2 ,