Falling Film Ammonia Evaporators

J. M. Gonzalez G., J. M. S. Jabardo, Politecnica de Madrid, Spain

ACRCTR-33

For additional information:

Air Conditioning and Refrigeration Center University of Illinois Mechanical & Industrial Engineering Dept. 1206 West Green Street Urbana, IL 61801

(217) 333-3115

W. F. Stoecker University of Illinois

December 1992

Prepared while Professors Gonzalez and labardo were Visiting Professors at the ACRe.

.,'

The Air Conditioning and Refrigeration Center was founded in 1988 with a grant from the estate of Richard W. Kritzer, the founder of Peerless of America Inc. A State of Illinois Technology Challenge Grant helped build the laboratory facilities. The ACRC receives continuing support from the Richard W. Kritzer Endowment and the National Science Foundation. Thefollowing organizations have also become sponsors of the Center.

Acustar Division of Chrysler Allied-Signal, Inc. Amana Refrigeration, Inc. Carrier Corporation Caterpillar, Inc. E. I. du Pont de Nemours & Co. Electric Power Research Institute Ford Motor Company General Electric Company Harrison Division of G M ICI Americas, Inc. Johnson Controls, Inc. Modine Manufacturing Co. Peerless of America, Inc. Environmental Protection Agency U. S. Army CERL Whirlpool Corporation

For additional information:

Air Conditioning & Refrigeration Center Mechanical & Industrial Engineering Dept. University of Illinois 1206 West Green Street Urbana IL 61801

2173333115

FALLING FILM AMMONIA EVAPORATORS

by

Juan M. Gonzalez G. and Jose M. Saiz Jabardo

Politecnica de Madrid, Spain

w. F. Stoecker, University of Illinois at Urbana-Champaign

ABSTRACT

This paper is a literature review and correlation of data on

falling film and spray evaporators. This type of evaporator offers

the potential advantages of improved heat-transfer coefficients and

a reduction in charge of evaporating fluid. The challenge in

reviewing papers by various authors is that the data appears in

different forms, so an attempt has been made to achieve some

uniformity in order to critically compare the data. Another

objective of this paper is to provide guidelines for how a

developer and manufacturer of this type of evaporator can improve

the performance of the model under development.

One of the important factors in achieving the maximum

improvement in evaporator performance is the proper choice of the

rate of refrigerant spray. A circulation rate that is too high

increases the refrigerant charge and also increases the pumping

power for the circulation of the refrigerant. On the other hand,

if the spray rate is too low, some areas on the tube become dry and

the heat-transfer coefficient drops off rapidly. Also, the optimum

tube placement in a falling-film evaporator may not be the same as

in a flooded ~vaporator. This paper also explores research studies

on enhanced surfaces for ammonia and the potential combination of

enhanced surfaces with the falling film concept.

INTRODUCTION

The reduction in the thickness of the ozone layer in recent

years has compelled the refrigeration and air conditioning industry

to intensify their efforts to find refrigerants that can be

-2-.,'

substituted for CFC and certain HCFC refrigerants. One of the

possible alternatives to the banned refrigerants is ammonia which

has excellent thermodynamic properties for a refrigerant and causes

no depletion of the ozone layer. The low concentrations of ammonia

that are considered toxic, however, necesitate special care to

avoid and/or minimize the effect of its release, particularly in

public and semipublic areas. One approach for improving the

ability to handle ammonia safely is to build refrigeration packages

with only small charges of refrigerant. The development of

low-charge chillers in the next few years will facilitate the use

of ammonia water chillers for air conditioning service and possibly

antifreeze chillers for supermarkets and other applications. One

attractive concept for reducing the refrigerant charge is the use

of falling film evaporators as a sUbstitute for flooded and even

direct-expansion chillers. Falling film evaporators are expected

to provide a high ratio of refrigeration capacity to refrigerant

charge.

This paper provides a description of the state of the art of

falling-film evaporators based on a bibliographic study. The first

type of evaporator studied is one with a vertical orientation and

with planar surfaces that are enhanced by means of fluted geometry.

The problem of "dryout" is analyzed with the objective of

determining the minimum quantity of liquid necessary to guarantee

the complete wetting of the tubes.

Described for horizontal evaporators are the hydrodynamics of

several types of flow patterns, the mechanism of entrainment of

liquid in the vapor flow, and several parameters that affect the

quantity of liquid flow rate in'order to avoid the break-down

condition. Several correlations to determine the heat-transfer

coefficients of the film for both the case of superficial

evaporation as well as nucleate boiling are described. In the past

few years enhanced heat-transfer surfaces have come into wide use,

particularly with copper tubes used for halocarbon refrigerants.

The availability of aluminum tubes with enhanced surfaces opens

similar opportunities for ammonia. Results obtained from

experimental tests of industrial-scale heat exchangers permit a

-3-

comparison with predictions from laboratory- scale models.

FALLING-FILM EVAPORATORS

Falling film evaporators are shell-and-tube heat exchangers in

which the refrigerant is distributed by means of a spray

arrangement above the tube bundle, forming a thin film of liquid

around the tubes. This system permits, at the same time, operation

with small temperature differences between the refrigerant and tube

and with high heat-transfer coefficients. These advantages have

been used frequently in applications in the chemical industry to

maintain a stable temperature and avoid decomposition of products.

In the desalination of sea water, falling film evaporators can

reduce the energy cost and they have been investigated for OTEC

applications that use the temperature gradients in the ocean to

generate power. In the OTEC application a reduction in the

temperature difference between the sea water and working fluid will

increase the energy efficiency of the system.

One of the most prominent characteristics of this equipment is

the low refrigerant charge which will be an attractive feature in

the refrigeration industry both in the short- and medium-term

because of the high costs of HFCs _and HCFCs or because of the

objective of increased safety in the case of ammonia.

One of the advantages of this type of equipment in comparison

to flooded evaporators is that the elevation of the boiling

temperature caused by hydrostatic head in the lower part of the

evaporator is avoided. The existence of hydrostatic head and its

effect of reducing the heat transfer rate in the evaporator does

not occur in spray chillers. This effect is most important in

shell and tube evaporators of 'large diameters, tube-type

evaporators of large vertical dimensions, evaporators operating at

low temperatures, and evaporators using refrigerants of high liquid

density, such as halocarbons.

Spray chillers recirculate the refrigerant that does not

evaporate back to the spray nozzles by means of a pump. The

systems should achieve a uniform distribution of the fluid on the

-4-

exterior of the tube bundle to guarantee a complete wetting of the

tubes. An incomplete wetting results in a drastic reduction in the

heat-transfer coefficient at the surface. The design must take

into consideration the interaction of the falling liquid with the

flow of vapor generated within the tube bundle. The circulation

rate of refrigerant should be minimized to the extent possible with

the objective of reducing the refrigerant charge as well as the

cost of pumping and the distribution system.

The publications on which this article has especially focused

deal principally with water and ammonia. Most of the experieces

with water evaporators derive from the work of Fletcher (1974,

1975), Canizzaro (1972), and Liu (1975) and were stimulated

principally by projects on desalination plants and were supported

by the Office of Saline Water. Meanwhile most of the work on

ammonia was promoted by OTEC especially that of Conti (1978),

Lorenz (1979-1982), and Owens (1977). The major investigations

with halocarbons were conducted by Danilova (1976) and Spegele

) with R-11 and R-12. Chyu and Bergles were very active in

studies of enhanced surfaces (1979 to 1982).

VERTICAL EVAPORATORS

Evaporators installed with a vertical orientation may not be

of as great a commercial interest as horizontal evaporators, but

certain of the insights gained from the research on vertical

evaporators may also apply to horizontal evaporators. In both

vertical and horizontal evaporators the boiling fluid is outside

the tubes. Essentially the thickness of the liquid layer will be

greater for long vertical tubes which decreases the coefficient of

heat transfer at the top of the tube because of the large thickness

needed there. Nevertheless, a long tube can generate a turbulent

regime that intensifies the coefficient of heat transfer (Dukler

(1960), Chun & Saban (1971, 1972). The hydrodynamic process of the

liquid on the exterior of the tube controls the heat transfer. An

early work by Nusselt considered the formation of a plane film in

the condensation process that increases with the length of tube.

-5- ....

With respect to the heat transfer, Chun and Seban (1971)

determined experimentally that the values of the heat-transfer

coefficients across the film in laminar as well as turbulent

regimes were greater with wavy flow. Other authors such as Kapitza

(1948) and Dukler (1960) studied the reduction in the thickness of

the liquid film due to the wavy motion of the fluid along the

length of the tube. The thickness was less than predicted by means

of the famous Nusselt theory, just as Kapitza had previously

described.

In the laminar regime using the thickness developed by Nusselt

the nondimensional coefficient of the film h* is

h* = (4/3)1/3 Re- 1/ 3 (1)

where the coefficient of the nQndimensional film is

h* = h /[(k3g~2/~2)]1/3 (2)

While if the reduction in thickness of the film because of the

effect of the ripples is considered, Chu and Seban propose: -0.22

h* = 0.606 (Re/4) Laminar regime (3) -3 0.4 .65

h* = 3.8xl0 (Re) (Pr) Turbulent regime (4)

Several authors such as Sinek and Young (1962) and Ganchev

(1972) confirm these correlations experimentally, however the data

reported for water by Fujita and Ueda (1978), although exhibiting

the same trends, have values 10% higher in the range of heat flows

from 30,000 to 70,000 w/m2 since they consider an additional

reduction in the liquid film that increases the flow of heat. They

propose for water the following correlations:

h* = 0.90 (Re)-0.22 For Re < 3200 (5 )

h* -3 (Re) .4 (6) = 6.0xl0 For Re > 3200

In the case of heat flows that ar~ predominantly nucleate boiling,

the process that controls the heat transfer and the heat-transfer

coefficient is independent of 'the velocity of the fluid. It is

thus possible to estimate the heat-transfer coefficient by means of

the following equation h = 1.24 q 0.741 w/m2oc (7)

o :1. A. Shmerler and I. Mudawar (1988) experimentally studied the

development of the turbulent sublayer with its wavy motion in the

falling film and proposed the following equation for values of Re

-6- "~'

between 2480 and 39,430:

h* = 0.0106 (Re)0.3 (pr)·63 (8)

The comparison of results of several authors is shown in Fig. 1.

17000 ....----------------,

u2000 ,.. c E ! ~7000

2000!---~~~~------~ o . 3000

RE 1000

• NUSS8..T

• OIJ

• FWITA

• SlMRER

• q.1.11()A5

a . q.2.8 1 ()A5

Fig. 1. Comparison of the film coefficients h reported by

various authors for vertical falling film evaporators.

DRYOUT IN VERTICAL FALLING FILMS

The dryout of a film of liquid flowing over a heated surface

produces dry areas that drastically reduce the transmission of

heat. Fujita has described two types of dry-out in vertical

surfaces:

(a) Dry patch in which the appearance of ~he dry area in the lower

part of the tube is due to reduction in the thickness of liquid due

to evaporation and entrainment.

(b) Flooding in which the thickness of the film is reduced by the

flow of vapor rising to the upper portion of the tube. This flow

of vapor is accompanied by a sudden reduction in the thickness of

the film of the liquid on the heated surface.

Once the film has been broken the dry surface no longer

disipates the heat rapidly and the temperature of the surface

rises. When the liquid returns to wet the surface, nucleate

boiling is produced that violently drives off the liquid, repeating

the process. Several configurations of dryout are shown in Fig. 2.

Col

"0'0

-7-

Hcallransler 10 ral6nlliquid films .nd film brakdown-'

StOIlI. ,I" ,Gteft

..... tlOn 01 Initial ,I" poren

(III

Fig. 2. Dryout in vertical tubes

.. '

...

Several authors (3. Bond and M. Donald (1957» have investigated

the quantity of liquid necessary to avoid breakdown. Their

approach was to explore the consequences of a modification of the

surface tension of the liquid film in the case of an absorption

column and determining the minimum thickness necessary to maintain

the wet conditions on it.

Later Norman and McIntyre (1960) have studied the effect of the

formation of the dry patch as a consequence of the change in the

surface tension with the temperature along the length of the liquid

surface (Maragoni effect), which is, "when irregularities occur in

the film thickness, the region of greater thickness will have a

surface of lower temperature, so subsequent higher surface tension

~orce will draw liquid from the thinner region. Finally, the

liquid deficiency in the thinner region will occur, forming a dry

patch on the wall surface."

Fujita (1976) has studied the phenomenon for the cases of saturated

-8-

and subcooled liquid films flowing on the exterior of a uniformly

heated vertical column. He found that thedryout occurred at water

flow values of 0.01 to 0.02 kg/m-s which correspond to Re of 140 to

280. In his work, J. Tang (1991) developed an analytical method to

determine the minimum quantity of liquid necessary to guarantee the

wetting of the tubes with vertical surfaces, both with and without

shear forces. Tang's theoretical predictions were confirmed by the

experimental data of Hewitt (1965). He found that the minimum flow

necessary was a function of the parameter M and the angle of

contact (Fig. 3). 4/5

Re = 8.6 M (1 min

- cos ~)3/5 (8 )

where M = ~/4 /( j ~1/4 gl/4) (9 )

In the case of water, M = 498 while for ammonia at 22°C the value

of M is 1010. Considering that the value of the angle of wetting

depends on the nature of the surface and the type of fluid, large

discrepancies in the values are found in the literature. Thus, for

water Hartley and Murgatroyed (1964) report for stainless steel a

value of 45° while Fujita (1978) present values between 8.5° - 15°

and Norman and McIntyre propose 17°. Other authors report for

ammonia on stainless steel values between 20° and 30° that

corresponds to a value of Re of the order of 100. min

1000

II •• 10 100

.-20 • I: • .-30 i • • .-40 a: 10 • .-so

~ •• 70

1 100 300 500 700 900 1100

..

Fig. 3. Minimum Re for wetting of vertical tubes.

-9- ....

Figure 3 shows the values of Re necessary to guarantee the min

wetting of the tubes as a function of the value of M and the angle

of wetting~. One observes that small modifications in the angle

of wetting have a great influence on the value of Re . min

INTENSIFYING HEAT TRANSFER

The transfer of heat in heat exchangers is intensified

considerably by means of axially-fluted surfaces. Alexander and

Hoffman suggested that the enhancement is largely the result of

naturally occurring waves which intermittently wash over the crests

of the flutes as shown in Fig. 4. When the trough between waves

passes by, the surface tension dries the crest of the flute by

pulling the liquid into the rill, thus producing a higher heat

transfer coefficient.

S

t ~ ,

.{ .' ~ l i

1 ~;

,

.' ~ . . \. " :!" , .'

'.

L· ~ •• 7.

, :; .. , , ~

·1 .' I '. ':" '.

I; ~ f. ~: tl

i f r ! ;

LIOUID FLOWS DOWNWARD

FLUTE lAVED BY WAVE

FLUTE EXPOSED IN TRpUGH OF WAVE

Fig. 4. Enhancement on fluted vertical tubes.

-10- .. '

The literature shows no equation that permits a correlation of the

transfer of heat with the dimensions and shape of the flutes with

the physical properties of the fluid. However, for a certain

geometry Rothfus and Newman (1978)present graphs that permit

calculation of the film coefficient for the case of a 26 mm radius

fluted aluminum tube on which ammonia is boiling (Fig. 5). The

film coefficient depends on the Reynolds number in the upper part

of the tube and on the recirculation of liquid, sigma. It can be

observed that an increase in the recirculation rate increases the

coefficient of heat transfer, however, an increase in the liquid

rate also increases the energy requirement of the pump providing

the circulation. On the other hand, reducing the circulation rate

of liquid too low increas.es the probability of break down .

~ ::: !!! u

~

~ 11.1 :: 11.1 Q

CIS .. :: z z ~ 11.1 :I

• I I

NtAN "tAT TItAN!'EIt COE,. ... CIENT "011 ,.,NON'A CONDENSING ON .HE OUTSI:lE SUR,.ACE O,..toN EXTERNAU,Y 'LUTED l-INCH ALUMINUN rust.

U'NIL CONSTANT' IIAOIUS 'LUTES.

........ ...-__ ~~;;7,"~(:-h~-I~ OO! CONDENSER

2

MASS FLOW AT aOfTOW OF TUBE a-:. -:-:~=-=~=-=~~~~~..:..:.::

MASS RATE OF EVAPORATION

10: •••. I

Col Z .. & a 10' 2 .. 6 a 10· REYNOLDS ~UNeER OF LIOUID NH

FED TO fOP OF FLUTED fUeE (f:'1APORATOR) MEASUREO AT BOTTOM OF TUBE (CONDENSER)

Fig. 5. Heat-transfer coefficient with ammonia on a fluted

ammonia tube.

-11-

The experiments conducted on this type of surface by Rothfus

and Newman with ammonia revealed that at low Reynolds number (400)

the film coefficients are lower than those obtained for plain tubes

owing to the low flow rate of liquid not wetting the flutes. On

the other hand, in the transition range from laminar to turbulent

(Re from 1500 to 2000), the maximum value of the heat-transfer

coefficient for ammonia occurs. Mostly the values are higher than

for a plain tube. Starting from the maximum point, the curve for

the film coefficient decreases with an increasing Reynolds number,

although later it begins to increase again as shown in Fig. 6

(Alexander & Hoffman).

In h*

~ Smooth

T"\~ - ; Ln Re

Fig. 6. Heat transfer coefficients on vertical fluted tubes.

The majority of the studies on this type of heat exchanger were

developed from ideal conditions of the laboratory on a single tube.

Carnegie Mellon University, however, constructed an evaporator heat

exchanger of 9400 kW capacity with 240 vertical fluted tubes. The

tests were conducted by Argonne National Laboratories and permit us

to analyze the performance of a real heat exchanger and compare

this performance with the theoretical predictions found in the

literature.

-12- ....

An evaporator designed for 100 gpm of ammonia (Re = 4700) 2 gives coefficients of the order of 1800 Btu/(hr-ft -OF). For

the same conditions the coefficient of heat transfer calculated for

a plain tube from the equation of Chun and Seban gives a value of 2

690 Btu/(hr-ft -OF). Figure 7 shows values of the film

coefficient h obtained experimentally for various values of the

Reynolds number as well as the values of h calculated by Chun and

Seban. Also shown are the theoretical values of Rothfuss. The

comparison shows that the heat transfer rate can be intensified by

a factor of 2.6 while the type of surface increases the coefficient

only by a factor of 1.57. However, the experimental values

obtained were less than those reported by Rothfus.

I&.

N C = ~

:s ~ GI

~

10000

1000

100 100 -

~

1000

r.

.. ~ ~ ... .. hChu&S • hargone

II h Rothfus

10000

Fig. 7. Comparison of heat-transfer coefficients from three

sources.

If the film coefficients obtained experimentally are compared

with the values of the curve of Rothfus, the following observations

may be made:

The maximum value of the heat-transfer coefficient appears,

according to Rothfus, in the interval of Reynolds number from

1500 to 3000 which corresponds to a circulation rate of 32 to 64

gpm of ammonia. However, in this range of flow rates the heat­

transfer coefficient is actually reduced because of the

-13-

emergence of dry out because of the poor distribution of liquid

in the heat exchanger.

The values obtained by means of the Rothfus curve for a

recirculation factor of 5 are considerably higher than the

values obtained from tests.

HORIZONTAL EVAPORATORS

Horizontal evaporators utilizing the principle of falling

films have certain advantages over vertical evaporators:

*the coefficient of heat transfer is higher

*the flow distribution is more uniform that permits a reduction

in the recirculation rate needed to avoid dryout, and this

feature permits a reduc~ion in refrigerant charge and allows

lower pumping cost

*vertical evaporators must be oriented precisely vertical, while

there is greater tolerance of deviation of the orientation

from the horizontal in the case of the horizontal evaporator

On the other hand, the horizontal evaporator is more vulnerable to

dryout because of the interaction of the liquid flow with that of

the vapor currents.

In the horizontal evaporator where liquid evaporates on the

exterior of the tubes, the hydrodynamics of the flow controls the

heat-transfer process. It is necessary, then, to design the

distribution system to distribute the liquid uniformly over the

tube bundle so as to permit wetting of all the tube surfaces.

In general, the distribution system feeds the upper tubes of

the bundle in such a way that the liquid drops down from one tube

to another. The tubes most likely to suffer dry out are those in

the lower portion of the bundle. In the case of large heat

exchangers it may be possible to provide some distribution streams

in the upper part of the bundle and some originating over the tubes

in the intermediate section as shown in Fig. 8.

AUXILIARY SPRAY TUBES (PLUGGED)

~!! 93 93 94

NO. OF FUIICTlOIIAL TUBES PER PASS

CARBOtl STEEL SHELL a TUBESHEETS

-14- ,.'

71"1.0.

I· + 72 ~4 0.0,

5"SIGHT ..L GLASSrg (

*1.5"0.0. TI TUBE EXTERtlAlLY COATED WITII !Iff 'I ltIf1rlr III"" ",

Fig. 8. Distribution system providing liquid supply both

at the top and the intermediate section of a

large heat exchanger (Hills, et aI, 1979).

HYDRODYNAMICS

When the. liquid flows from one tube to another the flow can

take the form of drops, columns, or sheets, as shown in Fig. 9,

depending on the flow rate of the liquid, the physical properties

of the liquid, and the separation of the tubes. When the flow rate

is very low, as in Fig. 9a the liquid flows in drops from one tube

to another. When the flow rate is increased (Fig. 9b) the pattern

converts to columns. Mitrovic (1986) indicates that the Reynolds

number corresponding to the transition between drops and columns is

between 150 and 200. A further increase in flow rate converts the

flow pattern to sheets (Fig. 9c). The transition from columns to

sheets occurs at Reynolds numbers between 315 and 600 according to

Kocacomustafaogullari. No publications have reported the

boundaries between laminar and turbulent flow when liquid spills

-15-..•.

from one tube to another, although Parker (1975) reports laminar

regimes occurring at Reynolds numbers of 6000. Horizontal falling

film evaporators generally operate with Reynolds numbers between

800 and 4000 which correspond to flow patterns of liquid sheets in

the laminar region (Kocacomustafaogullari). A. Jacobi is currently

studying the hydrodynamics of falling films at the University of

Illinois at Urbana-Champaign.

a b c

AI OlscnETE onOPLET DILIOUIO COLUI.IIIS CllIQUIO SHEET

Fig. 9. Possible flow patterns of the liquid falling

from one tube to another.

To study the phenomena of dryout, entrainment and the

transport properties in the droplet regime, it is necessary to

determine the dimensions of the drops, the distance between them

and the size frequency.

Sideman (1978) studied the phenomenon of the dropwise pattern

and concluded from experiment~ that the heat transfer in this

regime is at least two times higher than in the film regime. He

determined the frequency of various drop sizes and the extent of

separation between the drops for several values of the Reynolds

number. He found that with a Reynolds number less than 100 the

drop size is greater than 6 mm and by means of increasing the flow

the frequency of the drops also increases while the distance

between the points of formation decreases. In the case of Reynolds

-16- "~'

numbers less than 100, drops larger than 6 mm in diameter coexist

with drops of smaller size and greater frequency originating from

satellite locations. The large drops constitute 80% of the total

volume flow. C. M. Sabin (1978) studied the dropwise regime for

ammonia and determined the diameter of the drop of 0.1 inch and a

separation of 3/4 inch. He found that the frequency was irregular

along the length of the tube of the order of two times the diameter

and when not receiving drops for several seconds produced dryout on

plain tubes.

Yung and Lorenz (1980) studied the separation of the drop from

the film of liquid. They found that at the moment of separation

the large drop drags a small wake that breaks off and forms

satellite drops. They propose an equation for the diameter of the

main drop d p' 1/2

d = C(cr" /~ g) (10) p 1

For a mixture of alcohol and water the experimentally determined

value of C is 3. The diameters d of the drops from the s

satellite stream are in the range:

0.24 < d /d < 0.46 (11) s p

The volume flow in the streams of small drops constitute 19% of the

total volume.

The distance between drops has been studied by 3. Tang (1991)

K. Taghavi (1980), Dhir (1981), Mitrovic ,(1986), and D. Yung

(1980). They found that the film thickness on the underside of the

tube is in a wave form and the distance between drops coincided

with the longitude of the wave in accordance with the famous

instability equation of Taylor: '1/2

~ = 2(n"/~ g) (12) 1

The value of n is 3 if the liquid layer is relatively thick and its

value is 2 if the layer is thin. In tests with ammonia with a

value of n = 2 the equation permits the prediction of ~ with a

margin of error less than 4%. However, Ganic and Roppo (1980)

demonstrated experimentally that the value of A is not affected by

the flow rate of liquid nor by the separation between tubes.

-17-

ENTRAINMENT

An important process in falling film evaporators is the

influence of entrainment. In the absence of a current of vapor,

the liquid that does not evaporate on a tube passes directly to the

tube below it. However, when a vapor current prevails that

impinges transversely on the falling liquid, the drops or columns

of liquid can be deflected which results in an incomplete wetting

of the lower tubes resulting in a reduction in the heat transfer

coefficient.

In falling film evaporators this entrainment can result in

tubes not receiving the quantity of liquid necessary to keep the

tubes wet. Entrainment can result from several mechanisms:

deflection of the flow of liquid, atomization, stripping, and by

nucleate boiling.

Deflection Entrainment

In the situation that exists in a heat exchanger, a vapor

velocity moves perpendicularly to the flow of liquid and produces a

deflection of the path of the drops. If the angle of deflection a

in Fig. 10 is greater than the angle ~ then the drops do not strike

the lower tube .. D. Yung and Lorenz (1980) determined the value of

velocity u for the angle of deflection a to equal the angle ~: g

1/2 -1/4 u = [(3~ dg)/(2~)] [(P/D)(P/D-1)] (13)

g 1 g where P is the pitch of the tubes and D is the diameter

The maximum velocity of ~he vapor to prevent failure of the

drop from striking a lower tube is a function of the diameter of

the drops d. In particular, it is the diameter of the small drops,

d that controls the maximum value of (u ) . s g m

Figure 11 shows the magnitudes of maximum velocity of the

vapor as a function of the diameter of the drops for the case of an

ammonia evaporator operating at OOC calculated from Eq. 13. for

-18-....

o

--vapor - • • -

Fig. 10. Deflect.ion of the path of falling liquid because

of vapor velocity.

values of P/D = 1.25, 1.33, and 1.5. The dimension of the primary

drops according to Eq. 10 is 6 mm while the diameter of the

secondary drops varies between 0.14 mm and 0.28 mm. The maximum

velocity of the gas, (U) is between 1.3 and 1.7 mIs, g m

depending on the value of P/D.

-• -E -at :::I

4

3

2

1

o4---~~~~-r~~~~~~--~ 0.000 0.001 0.002 0.003 0.004 0.005 0.006 0.007

dp(m)

II ug(pld-1.8)

• ug(pld=1.vr , • ug(pld= 1. Ji

dp1 •••••••••• ds2

ds3

Fig. 11. Maximum velocities of the gas to avoid entrainment

during dropwise regimes with ammonia at OOC.

-19- "~'

In the case of the overflow of liquid that forms a column of

liquid, Yung and Lorenz (1980) proposed the following equation for

the calculation of the maximum velocity of the gas:

tan (flj)

4 A. r g

In which the effective diameter d* is:

8 A. r d* _ (, ___ )1/2 (2 g S)-1/4

7t PI

and the vapor drag coefficient C over the range of Reynolds d

numbers of interest is 1, Yung and Lorenz (1980).

Figure 12 represents the values of the maximum velocity of the gas

as a function of tbe Reynolds number obtained by means of Equation

13 for ammonia at OOC with tubes of 1.5 inch diameter and PIn = 1.25. For values of the Reynolds numbers between 300 and 700,

which is typical of the column regime, the maximum velocity of the

gas that still permits the drops to strike the lower tube varies

between 2.2 and 2.8 m/s.

A. M. Czikk (1978) et al determined experimentally for the

sheet regime the deflection angle of liquid ammonia in a heat

exchanger. They used a bundle of 1-inch diameter tubes arranged in

a triangular pattern with a 1.5 pitch to diameter ratio. The

ammonia was sprayed on the tubes at several different flow rates

and the drops were subjected to vapor velocities between 0.5 and 4

m/s. Figure 13 shows the angles of deflection, obtained as a

function of the velocity and the Reynolds number. It is observed

how for vapor velocities lower than 1 mls the angle of deflection

-20- ....

is negative. This indicates that the liquid flows around the

downstream (with respect to the vapor) side of the tube, passes its

low point, and then drips off with a component in the upstream

vapor direction. This effect is accentuated as the Reynolds number

increases and the velocity diminishes. With vapor velocities

higher than 1 mls the deflection increases with vapor velocity to

values between 30 0 and 40 0 at velocities of about 3 m/s. The

horizontal lines in Fig. 13 represent the maximum permitted values

of the angle ~ for various PIn values.

4 -.. 3

-• -E 2 -m ::I

1

o o

~ ~/

500

- l...--""

1000 Re

-~

ug

1500 2000

Fig. 12. Maximum velocity of gas as a function of Reynolds number.

50,-----------------------------~

40

30

.. 20

10

P/D~1.5

Re400

Re 1200

Re 2100

-10;---~~--_r----~--~----~--~ o 1 2 3

u(m/s)

Fig. 13. Experimentally-determined angles or der~eC~1on or ~ne

liquid at various vapor velocities.

-21-

ENTRAINMENT BY ATOMIZATION

With high velocities of the gas in contact with the drops, it

is possible to produce atomization of the drops. Wallis (1969)

found that the atomization is produced by values of the Weber

number higher than 12. Therefore, the velocity of the gas should

be less than the value calculated by means of the following

expression:

Wea 12 a )1/2

pg d pg d \....

From Eq. 16 it is seen that the maximum velocity of the gas for

atomization, u decreases as a result of the increase in the g,a

diameter of the drop as is represented in Fig. 14. Having,

therefore, drops that are larger in diameter than d limits the p

maximum velocity of the gas. For ammonia at OOC the diameter of

the primary drops d is 6 mm requiring that the velocity of the p

gas be less than 2.4 mls in order not to cause atomization.

-• -E -• Q ::s

6~----------------------------_

5

4

3

2~~-r~-'--~~ __ ~~~ __ ~ ·0.001 0.002 0.003 0.004 0.005 0.006 0.007

~~,--

lEI uga

Fig. 14. Maximum velocities of the gas in order not to

cause atomization of ammonia.

-22- ..•.

ENTRAINMENT BY STRIPPING

If the velocity of the vapor is sufficiently high, the film of

liquid becomes unstable, separating into drops that are dragged by

the flow of vapor. Ishii and Grolmes (1975) developed a criterion

that permits establishing the velocity of the vapor current that

produces stripping in the case of concurrent flow of vapor and

liquid. However, owing to the low liquid velocity in a falling

film evaporator this velocity drops out of the equation resulting

in the following:

PI III PI 2 0'.1 ( )1/2 ( )115 = 1

PI ( PI - PI) g

In the case of ammonia at OOC the minimum velocity u that

produces a loss of liquid due to stripping is 4.7 mIg, a value

higher than that which satisfies the conditions of deflection.

ENTRAINMENT BY NUCLEATE BOILING

The bubbles of vapor that form during nucleate boiling in the

liquid film tear the film, producing small drops. These small

drops are conveyed by the vapor current and result in a mist.

During the tests of an ammonia heat exchanger of 1 MW capacity for

the OTEC project, Thomas, Lorenz et al (1~79) experienced

entrainment which in this mode amounted to a fraction of liquid

less than 1% of the flow of vapor. This mode of entrainment is

rare in industrial equipment and is thus of not serious interest.

MINIMUM WETTING RATE

The minimum wetting rate (MWR) is the minimum quantity of

liquid necessary to maintain a wetted condition of the tubes.

Values lower than the MWR cause dryout and the coefficient of heat

transfer is reduced drastically. The value of the MWR depends on

several parameters: the physical properties of the fluid, nature of

-23- ..•.

the surface, the angle of wetting, the type of distribution system,

the separation of the tubes, the heat flux and the velocity of the

vapor. The magnitude of the minimum Reynolds number that

guarantees the wetting of the tubes can be theoretically calculated

by means of the equation proposed by J. Tang and S. Lin (1991) as a

function of the properties of the fluid and the angle of wetting ,:

Re M4/5 (1 _ ~)3/5 = 4.854 cos y min

(18)

where

In the case of ammonia at 22°C, the value of M is 1010 and

assuming an angle of wetting between 20° and 30°, the value of

Re varies between 180 and 350. As can be seen in Fig. 15, min

the value of Re i is very senstive to the value of the angle m n

of wetting , and a small variation in this parameter changes the

value of the MWR considerably.

c E • a::

1000,-------------------------------~

100

? "

~ .. ~)

If ' -

o 200 400 600 800 1000 1200 1400

M

III ,,10

• " 20

• " 30

• " 40

• " 50

Fig. 15. Minimum Reynolds numbers that avoid causing

dryout, as a function of M.

The difficulty in evaluating the angle of wetting, on the

surface restricts the use of the analytical equations for the

determination of the MWR. Therefore in any particular situation it

-24- ..•.

is necessary to experimentally determine this parameter. Table 1

summarizes some of the published works that have studied the

conditions of wetting in this type of evaporator.

Table 1. Summary of studies reporting on wetting.

SOURCE FLUID Minimum ruBE OBSERVATIONS Wetting REYNODl.S

J.Tang Ammonia 180 - 350 Plain Correlation * A.M.Czikk Ammonia 200-300 Plain Influence of

vapor velocity Conti Ammonia 220 Plain Comparison of

the wetting of various fluids

Ganic- W'ater 70-200 Plain Influence of Roppo Heat Flux

Tube Separation C.Sabin Ammonia 17 -170 Smooth Dry- Out

Rough Nickel Turner & Stripe Burnished, Diamon Knurled, Strainght Knurl

C.Sabin Ammonia 17 -170 Hi-Flux Wetting on the Nickel Plate tubes

aIEC Ammonia 500 Plain aIEC Ammonia 220 Hi-Flux . 32 MBTU/h

An analysis of the bibliography permits us to make the

following observations:

1. The wetting of the surface is improved by low viscosities and

surface tension. The effects of these properties could be

observed from the experiments of Conti (1978) who compared the

wetting conditions of water with ammonia on a stainless steel

tube, 5 cm in diameter. In the case of water with a flow rate

of 41 mL/(min-cm) (Re = 290) complete wetting of the tube did

-25- ..•.

not result. However, in the case of ammonia it was possible to

achieve a complete wetting with much smaller flow rates,

namely, 8 mL/(min-cm) corresponding to Re = 220.

2. The type of surface affects the formation of the draining of

the liquid surrounding the tubes and the value of the MWR.

C. Sabin (1978) demonstrated in a study of seven different

types of surfaces the behavior of ammonia in the dropwise

regime (7 < Re < 170). The types of surfaces were smooth

finish, rough, nickel plate, Linde Hi Flux, Turner and stripe

burnished, diamond knurled and straight knurled. It was found

that the nickel plate and Hi Flux surfaces favor the uniform

distribution of the liquid with the resulting complete wetting

of the tube, while the remaining surfaces form small canals of

liquid (rivelets) and form dry zones along the length of the

tubes.

3. The value of Re. increases with the heat flux and the m1n

separation between tubes as can be seen from Fig. 16 based on

experiments conducted on water by Ganic and Roppo (1980).

4. The circulation of the vapor around the tube changes the form

of the draining of the liquid on the tube and therefore the

MWR. For low velocities the vapor tends to equalize the

thickness of the film of liquid and reduces the value of the

Re. required to guarantee the wetting of the tube. m1n . However, from a certain value of the velocity, this effect is

reversed due to the entrainment. Figure 17 shows the results

of experiments conducted by A. M. Czikk with ammonia in which

is observed that Re. is achieved with a velocity of the m1n

gas at approximately 2 m/s.

-N E -;: =-a

2 4

-26-

r (kg/m·s)

6 8 . Re

2 4

Fig. 16. Film breakdown heat flux for water versus

liquid film flow rate.

1200

1000

800

• iii As a:: 600

400

200 0 2 4 6 8

v in/.

Fig. 17. Values of Re as a function of the min

velocity of the gas.

-27-

The experimental determination of the value of the MWR in

falling film evaporators in the OTEC project has allowed

observation of a similitud between the results obtained in

laboratory conditions with only one tube with the values obtained

from industrial evaporators of large dimensions. In the case of a

falling film evaporator of 1 MW capacity with plain titanium tubes,

the value of the MWR in the tubes was 45 lb /hr-ft (Re. = m m~n

500) having a value that is independent of the liquid supply system

used. Meanwhile with another evaporator of 32,000,000 Btu/hr

capacity with 388 Hi Flux tubes of 1.5-in in diameter and 55 inches

long, the value of the MWR was 20 lb/hr-ft (Re = 220).

HEAT TRANSFER

Most of the experiments conducted on predicting the transfer

of heat in falling film evaporators have centered on the

development of facilities for desalination or for the OTEC program

with water or ammonia (Liu, Fletcher, Parken, Conti, Seban, Owen).

On the other hand some experiments have concentrated on halogenated

refrigerants (Danilova). Table 2 describes the conditions of these

heat-transfer studies.

These experiments have permitted the development of several

correlations on the transfer of heat in falling film evaporators

(Owen, Lorenz and Yung, Sabin, Chyu and Bergles, Danilova) as well

as well as checking the validity of analy~ical and numerical models

(Parken and Fletcher, Chyu and Bergles, and Kocacomustafaogullari).

SURFACE BOILING

There are two main mechanisms of falling film evaporation-­

surface evaporation that occurs at low heat fluxes and bubble

formation-that takes place at high heat flux. Surface evaporation

is dominant in most falling film evaporators and in this regime the

local heat transfer coefficient is independent of heat flux.

-28-

Table 2. Summary of research studies on falling film evaporation.

SOURCE FLUIIl Test Oem Lan T (C) q w/m2 Re BOlLI secction 10- 3 m

LIU(1975) H2O stainles 2.5 30.48 55-100 28.-55 1200- No 5000

LIU(7195) H2O stainles 5 60.9 62-81 21 1300- No 2050

FLETCHER&SERNA H2O Cu/Ni 2.5 25 49-127 35-63 1500- Yes I (1974) 4100 FLETCHER&SERNA H2O Cu/Ni 5 25 49-127 23-52 2950- Yes S(1974) 7500 .FLETCHER& H2O Groves 5 25 49-127 30-80 700-7000 Yes& HAN(1985) knurle No

d PAKER &Al (1990) H2O BRASS 2.5 & 25 49-127 30-80 1000- Yes&

5 6800 No PARKEN(1977) H2O BRASS 2.5 15 49-127 49-79 2000- YES

10700 PARKEN,(1977) H2O BRASS 5 15 49-127 30-80 1000- YES&

6700 No SERNAS(1979) H2O BRASS 5 15 45-122 47-79 1680- No

6050 SERNAS(l979) H2O. BRASS 2.5 15 45-117 47-79 1150- No

4600 OWENS(1978) NH3 STEEL 5 30 22 5 16 110-9500 No CONTI(1978) NH3 S1F.EL 5 30 22 5-16 200- No

10500 ROGFRS&GOINDI H2O ALUMIN 13 45.7 17 100 800-2000 No

I (J989) U SABIN(J978) NH3 S1F.EL 2.5 30 12-24 3-25 17-170 No CHYU& H2O ill 2.5 15 100 10-200 300-1750 Yes BERGLESfI987) DANILOVA(1976) R12 S1F.EL 1.8 18 -40 0.5-25 250-6000 Yes

R22 50 R113

SPEGELE(1975) Rll PORCEL 1.2 23 5 13.7 180 .630 Yes AIN

In situations where nucleate boiling does not occur, the

saturated liquid film in contact with the tube is heated and a

surface evaporation is produced. The value of the film coefficient

h* depends on the values of the Reynolds number, Prandtl number and

on the ratio HID, the ratio of the distance between the tubes H to

the diameter D. Several authors have developed correlations that

can be written in the general form by means of the equation:

h* = (A1) (Re)A2 (pr)A3 (H/D)A4 ~ ( 19)

Values of the constants and exponents are shown in Table 3.

-29-,.'

Table 3. Constants and exponents for surface boiling equation

Source Re Ai A2 A3 A4 D, cm - ---

Owens(78) <1680 Pr- I .S 2.2 -1/3 0 0.1 1 Owens(78) >1680 Pr-1.S 0.185 0 0.5 0.1 1 Sernas(79) 1000<Re<6000 0.01925 0.24 0:66 0 1 2.5 Sernas(79) 1000<Re<6000 0.01729 0.24 0.66 0 1 5 Parken(90) 1000<Re<6000 0.042 0.15 0.53 0 1 2.5 Parken(90) 1000<Re<6000 0.038 0.15 0.53 0 1 5 Mitrovic86 240< Re<1100 0.0108 0.349 0.5 0.158 ~ 1.8 Danilova76 250 < Re < 6000 0.03 0.22 0.32 0.48 Re*J\·04

where the value 'of ~ and Re are obtained by means of the following

equations:

1 J3 =

1 + exp( -49 .86 * 10-3 Re 1.32 ) (.:to)

q"

Re* = { ( VI2/g)l/3 ) } hfg pv VI

Falling film evaporating coefficients calculated from the

several equations are shown in F~g. 17.

-30-

1~00,-----------------------------~

10000

II

6: ~BOOO c

• E -:l SOOO • --0-.c

4000 .** •• nnnlllill

~~--~--r-~~~---P--~--__ ~ o 2000 4000

Re 8000 8000

Owen

Semas Parker Danilova Mitovic

Fig. 17. Falling film heat transfer coefficients obtained

from several different correlations for surface

boiling with ammonia at OOC, diameter of 25 mm with

H/D = 1.25

ANALYTICAL MODELS

In addition to the empirical correlations, several authors

have developed mathematical models that predict the heat transfer

coefficients in the falling film evaporator. Rogers (1981)

developed a model that permits the prediction of the behavior of

the liquid film around the tube in the laminar regime. Two regions

are proposed: (1) the developing region, ~nd (2) the developed

region. The film coefficient is determined by means of the

boundary layer integral method. The results obtained in evaluating

the film coefficient in the developed region were very close to

those obtained using the equation of Fujita (1978) for the case of

vertical tubes.

M. C. Chyu and A. E. Bergles (1987) developed an analytical

model for predicting the heat transfer for the case of surface

evaporation, considering four separate zones around the

circumference of the tubes (Fig. 18). The four zones are (1)

stagnation flow, (2) impingement, (3) thermal developing/developed

-31-,.,.

with heating of the liquid, and (4) fully developed region with

surface evaporation .

• sr"""TlOII nOlI ItGlCIII

II ''''IIGIII[IIT nOllIlGlCIII

III M_ arVEL.., •• IiGiOll

II ,. ..... U mELCIfID 1£51011

Fig. 18. Model for falling-film evaporation on a horizontal

tube.

The coefficient of transmission of heat can be calculated as a

function of the film coefficient in each of the regions by the

equation:

On the other hand, Kocacomustafaogullair (1988) developed a

mathematical model that predicts the surface evaporation in a

horizontal tube bundle for the range of Reynolds number between 800

and 4000. He proposed that the liquid falls from one tube to

another as a thin liquid jet impinging on the top of the tube at a

uniform free-falling velocity. For this situation by means of

.finite differences, the profile of the velocities and temperatures

around the tube can be obtained.

-32-

NUCELATE BOILING

An increase in the flow of heat per unit area q" forms bubble

nuclei that considerably intensify the heat transfer. In this

case, the film coefficient h* depends on the heat flux q" in

addition to the values of the Reynolds and Prandtl numbers.

Owens (1976) developed a correlation to determine the film

coefficient with the assumption that the ratio HID has the same

influence that it has in the case of surface evaporation. However,

other authors (Danilova (1976), Parken) determined experimentally

that this factor does not influence the transfer of heat during

nucleate boiling.

W. H. Parken, et al (1990) considered that in the case of the

falling film with nucelate boiling that the film coefficient

depends on the flow of heat q". They developed correlations valid

for the values of Prandtl number between 1.3 and 3.6 and with heat 2

fluxes between 30 and 80 kW/m. These correlations can be

written in a general form by means of the following Eq. 23 with the

constants and coefficients shown in Table 4.

(23)

Table 4. Constants and coefficients for Eq. 23.

Source Re A1 A2 A3 A4 A5 D,cm

Parken(90) lOOO<Re<7000 0.00082 0.1 0.65 0 0.4 2.5 Parken (90) lOOO<Re<7000 0.00094 0.1 0.65 0 0.4 5 Owens(78) 0.0175 0 0.5 0.1 0.25 5

However, Danilova (1976) proposed for halogen fluids the following

equation

.72 Nu* = 1.32X10-3 Re 0.63 K

f P

.48 Pr (24)

-33-.'

where

Nu* = (h/k)( 0 )1/2 \2.~ ') --g(p(py)

(1~ ) Rc* = (q"/hf,PyV)( _0_ )1/2

g(p(py)

Kp= (Pl/CS)( _0---.;. )1/2 lZ1 ) g(p(Py)

On the other hand, Lorenz and Yung (1979) calculated the film

coefficient for the case of nucleate boiling in a tube super­

imposing the effects of convection and nucleate boiling. For this

purpose they imagined unwrapping the tube to form a vertical

surface of a length L = nD/2 and defined within the length L two

distinct regions:

(a) Developing region of the length designated L in which the d

saturated liquid is heated until a linear velocity profile is

achieved, and all of the heat transferred from the wall super-

heats the liquid film and no evaporation occurs, and

(b) Fully developed temperature profile in which the fluid is

evaporated from the surface.

These regions are shown in Fig. 19

The mean film coefficient h is calculated by superposition of

the coefficients of convective transmission and the surface

evaporation in each of the regions by means of the following

expression:

Ld TltERHAl

THIN fiLM ON HORIZONTAL TUBE

DEVELOPING REGION

-34-

r .J T • TS

r · r~~i~ING • REGION

FULLY DEVELOPED REGION

EVAPORATION + BOILING

SIMPLIfiED GECHETRT USED IN MODEL

Fig. 19. Regions in falling film on a horizontal tube.

(28)

where hb is the value of the film coefficient in nucleate

boiling. From data by Bergles and Rohsenow (1964) an expression

for hb is:

hb -' _______ _ ( Cp/hfg PfS)3 4 T2

For the film coefficient h in Zone a: d

h d

(30)

In Zone b it is possible to calculate the film coefficient by means

of the Chun and Seban equation

-35-

he= 0.821 ( k3g Jv2)113 (Re,l) "().22 in the laminar regime

'he= 3.8 x 10-3 ( k3g Jv2)113 (Re)O.4 (Pr)O.65 turbulent regime

The length Ld can be calculated by means of the Nusselt

equation:

3J.L Let = ('-__ -1)1/2

(31)

(32)

Heat transfer coefficients for ammonia at two different

Reynolds numbers are shown in Figs. 20a and 20b. The curves are

essentially identical indicating that the results are independent

of the Reynolds number.

1C1C1DOO 1C1C1DOO

• Owen

• Parka' 0 Daniiova -- 0

0 I • .. .. c 10000~ • C 1aooo ~ i 0 • I • • E g • • • Q • -- a • • • • • - • • • • - • • .c • .c

1000 \ 1000 .

i i I

\ 20000

. \ 20000 50000 80000 10000 80000 .. -,w/m A 2) I .. - W/mA2

I l--) (b)

Fig. 20. Falling film heat-transfer coefficients with Reynolds

numbers of (a) 1000 and (b) 5000.

-36-

ENHANCED SURFACES IN COMBINATION WITH FALLING FILM EVAPORATION

The use of enhanced surfaces for heat transfer in falling film

evaporators permits an increase of between 2 and 8 times in the

heat-transfer coefficients with respect to the values achieved with

plane tubes. The enhancement of the heat transfer can occur

through the mechanism of convection as well as having a favorable

influence on the nucelation.

Convection enhancement. Intensification of the convection

mechanism in low heat-flow regimes enhances turbulence, increases

the heat-transfer surface. and/or changes the angle of wetting. For

this mode of enhancement several types of surfaces have been used:

microfins, elliptical tubes, longitudinal rib, and spiral and

longitudinal flutes.

Nucleation enhancement. This method of enhancement is used

with high rates of heat flow and uses surfaces with microcavities

of very small diameter with the objective of trapping bubbles of

vapor in the interior and

of the pores are selected

properties of the fluid

assisting in nucleation. The dimension

in accordance with the physical

,~,k) with the object of opposing the

liquid from penetrating within the microcavities and destroying the

vapor embryos. The enhancement of the heat transfer can be

achieved using several types of surfaces, (a) those with a coating

applied to the base surface, and (b) using integral roughness.

(a) Coating of very porous material. Such a coating may be

formed by sintering, electrolytic deposition, brazing, flame

spraying, or foaming. The coating consists of a proous matrix with

large numbers of connecting pores composed of metal particles that

are bonded to each other and to the substrate. In this surface the

heat is transmitted in the first place by conduction by means of

the particles of the matrix and by continued conduction by means of

the film of liquid where the evaporation occurs. To the

-37- .. '

interconnected pores of the matrix can be supplied the liquid at

the same time that the vapor passes by means of the matrix toward

the surface of the liquid, as shown in Fig. 21 from Webb (1983).

By means of the generation of vapor in the interior of the pore,

the pressure increases until it achieves a value needed to overcome

the forces attributable to the surface tension. Then the vapor

escapes by means of the interconnecting pores and passes to the

surface. On the other hand, vapor formed in a pore activates

adjacent pores producing a chain of nucleation events.

LIQUID IN

VAPOR OUT

VAPOR BUBBLE TRAPPED IN ACTIVE SITE

Fig. 21. Cross-section of sintered porous coating.

The principal requirement of these surfaces is that they have

a porous matrix of the dimensions appropriate for the fluid, and

these dimensions depend on the physical properties of the fluid.

The mean radius of the pore varies from 0.01 mm to 0.1 mm. For

those fluids with significant surface tension and thermal

conductivity, such as water and ammonia, large radii are preferred

while for fluids with low surface tension and thermal conductivity

such as halocarbon refrigerants, small diameters are chosen. Webb

(1981) reports that the performance is not significantly affected

by the thickness of the coating nor the dimensions and form of the

particles.

-38-.. '

The other parameter that must be taken into consideration is

the coefficient of thermal conductivity of the particle because the

heat must propagate by means of conductivity through the particle.

In the Hi-Flux enhanced surface commercialized by Union Carbide the

surface consists of a large number of connecting porous metal

particles bonded to each other and to the substrate by brazing or

sintering.

(B) Integral Roughness. Th,e nucleation sites are formed by

mechanically working.,or chemically etching of the base of the

surface. Typically these sites are reentrant groves or tunnels

rather than discrete cavities. Within this configuration are

included T-shaped surfaces and the Thermoexel E surface, as shown

in Fig. 22.

Fig. 22. (a) T-shaped fins developed flattening the tips

of spiral fins of an integral-fin tube, and (b) Thermoexel E

surface with reentrant grooves and minute-spaced hole atop

the enclosed groove, from ~ebb (1981).

Por.

The T-shaped surfaces consist of tubes formed by flattening

the tips of spiral fins of an integral fin tube to restrict the

mouth of the space between two neighboring fins, as shown in Fig.

22a. The parameters that characterize this surface are the number

of fins per unit length (fins/inch) and the gap between the fins.

-39-

Among this type of surface are the GEWA T manufactured by Wieland

Werke. Some characteristics of the surfaces are:

Name Fins/in Gap width Fluid References

GEWA T 19C 19 0.25 mm H2o, R-114 Chyu & Bergles

GEWA T 26B 26 0.15 mm H2O, R-114 " 24 T Shaped 12 0.258 mm NH Huang (1990) 3

The Thermoexel surface of Hitachi (Fig. 22b) has minute

parallel tunnels with tiny holes located at regular intervals

communicating with the outside. The vapor generated is ejected

from the pores in the form of bubbles, and the liquid is drawn into

the tunnels to replenish the evaporated" liquid.

Table 5 presents a summary of some of the surfaces used for

enhancing the heat transfer in falling film evaporators.

For the case of a falling film evaporator with water, Chyu and

Bergles (1989) studied the behavior of three commercial surfaces,

Hi-Flux, Thermoexel E, GEWA T using heat fluxes less than 2

10 kW/m. The T-shaped fin tubes are those that produce the

greatest enhancement of heat transfer--by a factor of 3. while for - 2

heat fluxes of the order of 100 kW/m the Thermoexel E surface

achieves higher coefficients of heat transfer by a factor of 5

with respect to the values obtained for plane surfaces. Figure 23

presents the results of the experiments by Chyu and Bergles (1989).

However, in the case of the vertical falling film evaporator

with Refrigerant-114, Fagerholm (1985) found that the Thermoexel

surface produced greater enhancement than the GEWA T and Hi-Flux

surfaces in the complete range of heat fluxes between 3 and 30

kw/m2. He obtained an enhancement of a factor of 8, as shown

in Fig. 24.

-40-

Table 5. Enhanced surfaces studied by various experimenters.

Source Surface Manufacturer Fluid

Conti Hi2ht Flux Linde U.Carbide NH3 Conti Oroved Surface NH3 N .Fagerholm Thermoexel E Hitachi R 114 N.Fa2erholm Thermoexel EC Hitachi R 114 N .Fa2erholm Oewa-T Wieland Werke R 114 N .Fagerholm Hi2h Flux Linde U.Carbide R 114 N .Fa2erholm Sand Blasted R 114 Chyu& Bergles Oewa Tl9 C Wieland Werke H2O Chvu& Bergles Oewa T26 C Wieland Werke H2O Chyu& Ber2les Thermoexel-E Hitachi H2O Chyu& Bergles Hi2h Flux Linde U .Carbide H2O Flecher Kumled Cu/Ni tube H2O Lorenz& Yun2 High Flux Linde U.Carbide NH3 Lorenz& Yung Wolverine St 19 NH3 Lorenz& Yun2 Tubochill Fin NH3 Lorenz& Yung C MU Fluted NH3 Sabin& Straight knurled NH3 PODDendiek 2rooves Sabin& Diamond knurled NH3 PODDendiek grooves Nakavama Dorous Dlale Hitachi R11 Nakayama Verticallv 2rove Dlate Hitachi R11 Nakayama Horizontally grove Hitachi R11

Dlate Kajikawa Fine Metal Fiber Nh3 /R22

F.M.F. Bukin Danilova Porous Coating RI2/R22

Danilova Metal Coating Rll/RI2 . R22

Yingke Tan Mechanical porous R1l3 surface

Ouoqing Wang Mechanical porous R1l3 laver Structure

-41-

100000,-------------~Mn~~----------~

10000

1000~--~~~~~~~~~~~~~~~~

.1 1 10 100

DT IIC

PLAIN

--0-- GEWAT

• *

HIFLUX

1E

Fig. 23. Results of experiments by Chyu and Bergles (1989).

100000~------------------------~

# PLAIN

N

~ * THERM:> EXEL < 10000 ": E - --0-- GEWAT ~ -r:T • HI FLU

1000 ' . . 1 1 10

TS-TSAT (IIC)

Fig. 24. Results of experiments by Fagerholm (1985).

,.'

The porous metal coating has been used for enhancement of heat

transfer in falling film evaporators with horizontal tubes both

with ammonia and fiuorocarbons.

-42-

In the case of ammonia boiling on horizontal tubes covered with

the Hi-Flux surface, Conti (1978) obtained an enhancement of the

order of three times for heat flows of 10 kW/m2 resulting in

film coefficients of 17 kw/m2-oc which was practically constant

for values of circulation ratios higher than 5. On the other hand,

this type of coating was used in a heat exchanger of 32,000,000

Btu/hr with 388 titanium tubes in the OTEC project. The film

coefficient determined in the Argonne Laboratory for various flows 2 of ammonia was 21.5 kW/m _oC, a value that remained constant

and independent of the flow of ammonia for values of Reynolds

number higher than 220, but falling abruptly from this value owing

to the appearance of dryout on the tubes.

The enhancement of heat transfer for Refrigerants-12 and -22

in falling film evaporators was studied by Buking and Danilova

(1982). They compared the heat-transfer coefficients on different

types of tubes with coatings produced by spraying, sintering and

jacketing and observed that the the highest film coefficients were

obtained using tubes of stainless steel 1KHl18N97 by sintering

powdered metal to the tube surface with a porosity of 46-52% and a

grain dimension of 0.06 to 0.25 ~m. They achieved an enhancement

factor of from 3 to 5 times in compar~son to plain tubes in the 2 band of heat flows between 1 and 25 kW/m. On the other hand,

they observed that the film coefficient was independent of the

magnitude of the circulation ratio over a wide interval.

INFLUENCE OF THE BUNDLE EFFECT IN HEAT TRANSFER IN

FALLING FILM EVAPORATORS

Most of the experiments conducted to determine the correlations

for calculating the heat-transfer coefficients in falling film

evaporators were conducted on single tubes. Few reports exist on

the performance of bundles, but several experiments conducted with

ammonia, R-12 and R~22 indicate that so long as the tubes are kept

wetted the value of the film coeffient of a tube bundle is similar

to that of a single tube. This important conclusion permits

-43-

falling film evaporators to be designed using empirical values

obtained from tests of a single tube.

For R-12 and R-22, Bukin (1982) showed that values of the film

coefficients experimentally obtained were not affected by the

number of tubes. He used an evaporator of six tubes in a vertical

row. On the other hand, for ammonia, Lorenz and Yung (1982)

determined the heat-transfer coefficients on various tubes of a

heat exchanger with 6304 tubes operating with a capacity of 1 MW

developed for the OTEC Project and observed that the average value

of the film coefficient for the full heat exchanger was similar to

that obtained for one tube. In addition, they observed a uniform

distribution of the film coefficients in the interior of the bundle

with a margin of ± 10% with respect to the value of the film

coefficient both for plain tubes and for Hi-Flux tubes.

COMPARISON BETWEEN THE HEAT TRANSFER COEFFICIENT OF

FALLING FILM EVAPORATORS AND FLOODED EVAPORATORS

The coefficients of heat transfer in falling film evaporators

are generally higher than those obtained in flooded evaporators.

The difference that exists between the coefficients of heat

transfer in the two types depend on the rate of heat flow and the

type of surface.

In falling film evaporators with low rates of heat transfer,

the transfer of heat is obtained by superficial boiling and the

film coefficient depends on the value of the Reynolds number. In

flooded evaporators, on the other hand, the mechanism of heat

transfer is free convection and the value of the film coefficient

depends on the rate of heat flow. During high heat flow rates

nu~leate boiling is started and from that point on the transfer of

heat is intensified more strongly in pool boiling than in falling

film evaporation. The difference in the heat-transfer coefficient

progressively decreases as the heat transfer rate increases.

Figure 25 shows results obtained by Chyu and ~ergles (1989) for

-44-"~'

water boiling on plane tubes and with tubes of enhanced surfaces

(Thermoexel E) in pool boiling and in falling film. It can be

observed how at low heat flow rates the coefficient of heat

transfer in falling film evaporation is several times higher than

that of pool boiling, but the difference decreases at high rates of

heat flow. On the other hand, in the case of enhanced surfaces the

difference between the curves of pool boiling and falling film is

less than that of the plain surfaces owing to the intensification

factor of the enhanced surfaces. In regimes of high rate flows of

heat the curves for the heat-transfer coefficients converge for all

1000000y---------------------------------__ -,

--1:1'-- Thermo Excel -Falling Film

----- Plane Falling Film 100000 , ,

N ,,f

c JS E 10000 , " - ,rS , ~ " " , "

rS '

1000

• Thermo Excel. Pool Boiling .

100 Plane Pool Boiling

.1 1 10 100

DT Ie

Fig. 25. Comparison of heat transfer coefficients of falling

film evaporators with pool-boiling evaporators for

plane and enhanced surfaces.

REFRIGERANT CHARGE IN FALLING FILM EVAPORATORS

One of the most interesting characteristics of falling film

evaporators-is that they can operate with a lower refrigerant

charge than flooded evaporators. Figure 26 compares the estimated

refrigerant charges for flooded and falling film ammonia

-45-

evaporators operating at O°C and cooling water with a mean

temperature difference of 7.5°C. For flooded evaporators the

heat-transfer coefficient of 1050 w/m2_oc was chosen in

"~'

accordance with data from a manufacturer's catalog. The depth of

the refrigerant was assumed to be 3/4 of the shell diameter. For

the falling film evaporator, a heat-transfer coefficient of 4300

w/m2_oc was used, a value obtained by Argonne Laboratories in

the OTEC project. To calculate the charge of refrigerant in

falling film evaporators, all the tubes are covered with a film of

refrigerant, and the"' liquid at the bottom of the shell was assumed

to be at a depth of 1/4 the radius of the shell. In Fig. 26 it can

be observed that it is possible, for example, for a heat exchanger

with a 700 kW capacity to operate with a charge of about 30 kg of

ammonia. Even after adding the additional charge existing in the

condenser, the total charge is likely to be low enough that many

building codes will allow installation where ammonia is not now

permitted.

1000~----------------------------~

~ • J

100

u

10~--------~--~~~~~~~-.~

100000 1000000 , "(w,

• •

Flooded

FatIng FiJN\

Fig. 26. Estimated comparative ammonia refrigerant charges in

flooded and falling film evaporators.

....

1. Horizontal Spray Film Evaporator aque Chemical Corporation,. Office of Saline Water,:

2. Experimental & analityticla Studies of Heat Transfer in a Falling Film System(Part 1). Office of Saline Water:

3. Horizontal Spray Film evaporator Aqua Chemical Corporation). Office of Saline Water,:

4. Second Report on Horizontal Tube Multiple Effect Process Pilot Plant Test and Desing. 1970, Office Saline Water:

5. Third Report on Horizontal Tube Multiple Effect Process Pilot Plant Test and De sing. 1971, Office Saline Water:

6. Third Report on Horizontal Tube Multiple Effect Process Pilot Plant Test and Desing. 1971, Office 'Saline Water:

7. Fourth Report on Horizontal Tube Multiple Effect Process Pilot· Plant Test. 1972, Office Saline Water:

8. Final Report horizontal Spray Film (HTEX) and Vertical Tube (VTEX) Evaporator O.S.W. 1973, Office of Saline Water,Contract14-3014 ,Thermal Process Division, San Diego Test Facility U.S. Department of Interior,Burs and Row Construction Corp, Paramus N .0.:

9. Fifth Report on Horizontal Tube Multiple Effect Process Pilot Plant Test. 1973, Office Saline Water:

10. A.E., D., Fluid Mechanics and Heat Transfer in Vertical Falling Film System. Chem.EngProg Symps. Ser, 1960. 56 N 10: p. 1 10.

11. Alexander, L.G. and H.W. Hoffman, Performance Characteristic of Advanced Evaporator Tubes for long Tubes Evaporator. 1971, ORNL:

12. Aly G., AI-Haddad, and A.-J. m, Parametric Study on Falling Film Seawater Desalination. Desalination, 1978. 65: p. pp43.

13. Andersson, H.I., Diffusion from a Vertical Wall into an Accelerating Falling Liquid Film. Int. J. Heat Mass Transfer, 1987. 30: p. 683.

";"'.". -. 47

14. Ayub and knewitz. Limited Inventory Ammonia Falling Film Spray Evaporator. in I.I.A.R. 1991. San Francisco(California):

15. Bergles and Jensen. Enhancement Single Phase Heat Transfer for OTEC system. in OTEC. 1987.

16. Berntson, Berntson, and Pan holzer, Heat Transfer of Nonazeotropic Mixtures in a Falling Film Evaporator. : p. 1337 1350.

17. Bond J. and M.B. Donald, The effect of Absortion on the wetted area of absortion tower. Chemic~1 Engineering Science, 1957. 9886: p. 237 - 244.

18. Bukin, Danilova, and Dyundin, Heat Transfer form Freons in a Film Flowing over Bundles of Horizontal Tubes That carry a Porous Coating. Heat Transfer Soviet Research, 82. 14.: p. 98 103.

19. C.P.Newman, S.A.H.&., Dynamic models of Falling Film Evaporators for Ocean Thermal Energy Convertions Plants. 1976, Carnegie Mellon University:

20. C.Panchal, Energy Efficiency comparation of advanced Ammonia Heat Exchanger Types. 1991 ..

21. Cannizzaro, Fourth Report on Horizontal Tube Multiple Effect (HTME) Process Pilot Plant Test Program, 1972,. 1972, Universal Desalting Corporation,:

22. Cavallini A., Frizzerin S., and Rossetto L. Condensation of R11 Vapor Flowing Downward outsdie a Horizontal Tube Bundle. in 8th International Heat Transfer Conference. 1986.

23. Chang and Witte, Liquid Solid Contact. in Pool Film Boiling From a Cylinder. Journal of Heat Transfer, 1990. 112: p. 263266.

24. Chen, M.M., An Analytical Study of Laminar Film Condensation: Part I -Flat Plates. American Society of Mechanical Engineers Journal of Heat Transfer, 1961. 83: p. 48-54.

25. Chen, I., Y., Heat Transfer Analysis of a Falling Film Horizontal Tube

48

Evaporator. 1984, Univ.Wisconsin, Milwaukee,WI:

26. Chepurnoy , S., Berkuta" Heat Transfer in Turbulent Flow of Liquid Flim over Vertical Surfaces. Heat Transfer Soviet Research, 76. 8: p. 62 65.

27. Chun and Seban, Heat Tansfer in Falling Film. Journal of Heat Transfer,1971. VOL: p. 391 396.

....

28. Chun and Seban, Performance Prediction of Falling Film Evaporators. J.Heat Transfer, 72. 94: p. 432 436.

29. Chyu M.C. and BerglesA.E., Falling Film Evaporation on a Horizontal Tube. : p. 39 48.

30. Chyu M.C., Boiling Heat Transfer From a Structured Surface. 1979, Iowa State University: .

31. Chyu M.C. and BerglesA.E., Characteristc of Nucleate Pool Boiling from Porous Metallic Coatings. Advances in Enhanced Heat Transfer, ASME, New York, 1981, . 66-71.

32. Chyu M.C. and Bergles A.E., An analytical and Experimental Study of Falling Flim Evaporatartion on a Horizontal Tube. Journal of Heat Transfer, 87. 109: p. 983 .. 990.

33. Chyu M.C. and Bergles A.E., Thermal Analysis on the Electric Cylindtical Test Heater for Heat Transfer Experiment. Experimental Thermal and Fluid Science, 1988. 1: p. 19-27.

34. Chyu M.C. and Bergles, Horizontal Tube Falling Film Evaporation With Structures Surfaces. Transation of ASME, 89. 111: p. 518 524.

35. Chyu M.C .. and Bergles. Enhancement of Horizontal Tubes Spay Film Evaporators. in 7th Int Heat Transfer Conference. 1982.

36. Conti. Experimental investigation of Horizontal Tube Ammonia Film Evaporators with small Temperature Differentials. in Proceeding of the Fifth Ocean Thermal Energy Convertion OTEC, Miami Beach. 1978. Miami Beach:

49 "~'

37. Czikk A.M., Ocean Thermal Power Plants Heat Exchangers". 1977, UCC -Linde Division,Tonawanda, NY:

38. Czikk A.M. Fluid Dymamic and Heat Transfer Studies of OTEC Heat Exchangers. 1978. Proceeding of the Fifth Ocean Thermal Energy Convertion OTEC, Miami Beach:

39. Danilova, Burkin, and Dyundin, Heat transfer in Spray-Type Refrigerator Evaporators. Heat Transfer Soviet Research, 76. 8: p. 105 113.

40. Danilova, Guygo, and Boreshanskaya, Enhancement of Heat Transfer during Boiling of Liquid Refrigerants at low Heat fluxes. Heat Transfer Soviet Research, 1976. 8 N2 4: p. 1 .. 8.

41. Dhir, V.K., Hydrodinamic Transition During Dripping of a Liquid From Underside of a Horizontal Tube. 1981, ASME:

42. Dukler, A.E., Chem. Eng. Prog. Symp. Ser, 1960. 56 N 10: p. 1 -10.

43. Fagerholm, Kivioja, and Ghazanfari. Using Structured Surfaces to Enhancement Heat Transfer in Falling Film Flow. in I.I.R. 85. Trondheim (Norway):

44. Faghari A. and Seban R.A., Heat and Mass transfer to a turbulent Liquid Film. Int. J. Heat Mass Transfer, 1988. 31: p. 891.

45. Faghari A. and Seban R.A., Heat and Mass transfer to a turbulent Liquid Film II. Int. J. Heat Mass Transfer, 1989. : p. 1796.

46. Fletcher L.S., Valentinas Sernas , and L. S. Galowin, Evaporation from thin water films on horizontal tubes. Ind. Eng. Chem. Proccss.Des.Dev, 1974. 13 N 3: p. 265 .. 269.

47. Fletcher L.S., Sernas V., and W. P.H., Evaporation Heat Transfer Coeffitcients for Thin Sea Water Film on Horizontal tubes. Ind. Eng. Chem.Procc.Des.Dev., 1975. 14: p. 411.

48. Fletcher L.S. and J.-C. Han, Falling Film evaporation and Boiling in

50

Circumferential and Axial Grooves on Horizontal tubes. Ind.Eng.Che.Process.Des.Dev., 1985. 24: p. 570 575.

.'

49. Frivic and Christensen, An Experimental Study on the Influence of oil on Heat Transfer in Gravety Fed, Flooded Evaporator. 85, I,I,F.- I,I,R.:

50. Fujita, T. and T. Ueda, Heat Transfer to Falling Liquid Films and Film Breakdown-1 /Subcooled Liquid Film. J.Heat Mass Transfer, 1978. 21: p. 97 108.

51. Fujita. T. and T. Ueda, Heat Transfer to Falling Liquid Films and Film Breakdown-II Saturated liqud Films with Nucleate Boiling. J.Heat Mass Transfer, 1978. 21: p. 109 118.

52. Fujita, Takahama, and Asano, Heat and Mass Transfer in countercurrent Flow of Air and Wate Film{ Effects of Cylindes Istalaled in the Air Flow. Transation of The Japan Society of Mechanical Engineering, 1982. 48: p. 518 525.

53. Ganchev, B.G., Heat Transfer Soviet Research, 1972. 4: p. 102.

54. Ganic, Firts, Second, Third Quarterly Report Submittal to ERDA. 1977, UCC-Linde Division, Tonawanda, NY:

55. Ganic and Roppo, An Experimental Study of Falling Film Breakdown on a Horizontal Cylinder During Heat Transfer. J.Heat Transfer, 1980. 102: p. 342.

56. Goindi, S.S., Experimental Study of the Flow ~nd Heat Transfer Characteristics of Thin Water Films on a Large Diameter Horizontal Tube. 1985, Carleton University Ottawa:

57. Han J.C. and F. L.S. Influence of Concurrent Steam Shear Velocity on Falling Film Evaporation and Boiling in a Horizontal Tube. in Proceedings of the Joint ASME-JSME Thermal Engineering Conference, HI. 1985.

58. Hartley, D.E. and W.Murgatroyed, Criteria for the break up of thin liquid layers flowing isothermally aver a solid surface. Int.J.Heat Mass Transfer, 1964. 7: p. 1013 1015.

51 ....

. 59. Hewitt, G.F. and ·P.W.C.Lacey, The break down of the liquid film·in annular two phase flow. Int. J. Heat Mass Transfer" 1965. 8: p. 781-788.

60. Hillis, Lorenz, Yung, and Sather, OTEC Performance Tests of the Union Carbide Sprayed Bundle Evaporator. 79, Argonne National Laboratory:

61. Holmberg, Berntssons, and Perssons. Heat Transfer in a Falling Film Lithium Bromide Water Evaporator - An Experimental Study. in XVIII th Internationa Congress of Refrigeration,. 1991. Montreal:

62. Ishii M and Grolmes M.A., Inception Criteria for Droplet Entrainement in Two Phase Concurrente Film Flow.". AIChE Journal, 1975. 21 N2: p. 308 318.

63. Jang M.J. and J.f. Makki, A Model For Predicting Heat Transfer to a Turbulent Falling Film. Chem. Eng. Comm., 1990. : p. 153 -165.

64. Kocacomustafaogullari G., Two Fluid Modeling in Analyzing the interfacial Stability of Liquid Film Flows. Int. J. Multifase Flow, 1985. 11: p.63.

65. Kocacomustafaogullari G. and Chen, Fallinf Film Heat Transfer Analysis on a bank of Horizontal Tube Evaporator,. AICHE Journal, 1988. 34 N 9: p. 1539 1549.

66. Kutadeladze, Fundamentals of Heat Transfer, . 1963, London. p. 307.

67. Kutaladze S.S. and Gogonin 1.1., Heat Transfer in Film Condensation of Slowly Moving Vapor. Int. J. Heat Mass Transfer, 1979. 22: p. 1593.

68. Kutaladze S.S. and Gogonin 1.1., Investigation of Heat Transfer in Film Condensation of Flowing Vapour on Horizontal Tube Banks. Int. J. Heat Mass Transfer, 1985. 28: p. 1831. .

69. Lewis and Sather, OTEC Performance Tests of the Union CarbideFlooded Bundle Evaporator. 79, Argonne National Laboratory:

70. Lienhard, J.H. and Wong, The Dominant Unstable wavelength and minimum heat flux during film boiling on a horizontal cylinder. j.Heat Transfer, 1964. : p. 220.

52 ..•.

71. Lienhard J.H., Burnout on Cylinders. Journal of Heat Transfer, 1988. : p. 1271.

72. Liu, The Evaporating Falling Film on Horizontal Tubes. 1975, University of Wisconsin,Madison:

73. Lorenz and Yung, Anote on combined boiling and evaporation of liquid films on Horizontal Tubes. Transation of A.S.M.E., 1979. VOL 101 FEB: p. 178 .. 180.

74. Lorenz, Yung, Hillis, and Sather, OTEC Performance Tests of the Carnegie Mellon University Vertical Fluted Tube Evaporator. 79, OTEC:

75. Lorenz, Yung, Howard, Pachal, and Poucher, OTEC 1 Power System Test Program: Performance of One Megawatt Heat Exhanger. 81, OTEC (Argone National Laboratory):

76. Lorenz and D. Yung, Film Breakdown and Bundle Depth Effect in Horizontal Tube Falling Film Evaporators. J.Heat Transfer, 82. 104: p. 569 571.

77. Lukach, Radchenko, Tananayko, and Petukhov, Heat Transfer to Thin Liquid Films Flowing Down a Poymer Surface. Heat Transfer Soviet Research, 76. 8: p. 20 24.

78. M. Cerza and V.Sernas, Nucleate Boiling in thermally developing and fully developed laminar falling water films. Journal of Heat Transfer, 1988. 10: p. 221 .. 228.

79. Marto P.J., Wanniarachchi A.S., and Pulido R.I., Augmenting the Nucleate Pool Boiling Characteristics of GEWA -T Finned tubes in R 113. ASME HTD, 1985. 52: p. 63-67.

80. Marto P.J., Zebrowski D., and Rose J.W., An experimental Study of R113 Film Condensation on Horzontal Integral Fin Tubes. J.of Heat Transfer, 1990. 112: p. 758.

81. Masahiro Ishimatsu, Takehiro Nosoko, and T. Nagata, A Model For Predicting Heat Transfer to a Turbulent Falling Film. Chern. Eng. Comm.,

53

1990. 97: p. 602 616.

82. Mitrovic, k.S.a.J., Heat Transfer in Natural Convective Boiling of Refrigerants and Refrigerant Oil Mixtures in Bundles of T Shaped Finned Tubes. Advances in Enhanced Heat Transfer, ASME, New York, 1981, . 136-141.

83. Mitrovic, Influence of tube Spacing and flow rate on Heat Transfer form a Horizontal .. Tube a Falling Film Liquid. 86. 4: p. 1946 .. 1956.

84. Miyasaka Y. and I. S., The effect of pure forced convection on the boiling heat transfer between a two- dimensional subcooled water jet and a heated surface. J.Chem: Eng. of Japan, 13: p. 22.

85. Moalem D. and S. s., Theorical Analysis of a Horizontal Condenser-Evaporator Elliptical Tube. Transation Of ASME, 75. : p. 352 359.

86. Moalem D. and S. Sideman, theorical Analysis of a horizontal condenser evaporator. J.Heat &Mass Transfer, 1976. 19: p. 259 .. 270.

87. Moalem D., Sideman S., and Dukler A.E., Dripping Characteristics in a Horizontal Tube Film Evaporators. Desalination, 1978. 27: p. 117 127.

88. Moalem D., Sideman S., and Horn H., Enhanced Film Mass Transfer Coefficient on Grooved Horizontal Conduits. Chem. Eng. Sci., 1979. 34: p. 420.

89. Mudawwar I.A. and E.M.M. A., Momentum and Heat Transfer Across Greely Falling Turbulent Liquid Flims. Int. Mult. Flow, 1986. 12: p. 771.

90. Nakayama, Daikoku, and Nakajima. Enhancement of Boiling and Evaporation on Structured Surfaces with gravity driven Film Flow of R-11. in Heat Transfer. 82. Munich:

91. Norman W.S. and Mcintyre V., Heat Transfer to a Liquid Film On a Vertical Surface. Trans. Instn. Chem, Eng., 1960. 38: p. 301 307.

92. Norman W.S. and MC Intyre V., The effect of surface Tension Changes n the minimum wetting rates in a wetted rod distillation. Trans.lnstn. Chem.

.'

.;-": ., 54 .. '

Engrs., 1960. 38: p. 294 300.

93. Owens. Comments on Sea Water side Enhancement for OTEC Heat Exchangers. in Fourth Ocean Thermal Energy Conference. 1977. New Orlean:

94. Owens W.L., Cost comparison for threee OTEC Heat Exchanger Designs. 1977 ..

95. Owens W.L. Correlation of Thin Film Evaporation Heat Transfer Coefficients for Horizontal tubes. in the Fifth Ocean Thermal Energy Convertion OTEC, Miami Beach. 1978.

96. Parizhskiy, Chepurnenko, Lagota, and Taranets, Study of Booiling Heat Transfer with a Falling Film of Refrigerant. Heat Transfer Soviet Research, 72. 4: p. 43 47.

97. Parken W.H., Heat Transfer te thin Water Films on Horizontal tubes. 1975, Rutgers University:

98. Parken W.H. , Fletcher L.S., V.Sernas , and J.C.Han, Heat Transfer through Falling Flims Evaporation and Boiling on Horizontal tubes. Journal of Heat Transfer, 1990. 112: p. 744-749.

99. Parken W.H .. and F. L.S., An Experimental and Analytical Investigation of Heat Transfer to Thin Water Films on Horizontal Tubes. 1977, Univ. Virginia, Charlottesville, VA,:

100. R.L.Wett, The Evolution of Enhanced Surface Geometries for Nucleate Boiling. Heat Transfer Eng., 1981. 2: p. 46-69.

101. Rautenbach R. and Gebel J., On the Concetration of RoBrines by Seeded Horizontal Tube Falling Film Evaporation (HTFE). Desalination, 1989. 76: p. 107.

102. Rhinesmith, B.Y.&.R., Horizontal Spray Film Evaporation. CHEM. Eng Prog., 1981. 77: p. 69-73.

103. Rifert, Zadiraka, Barabash, and Andreyev, The Mechanism and Rate of Heat Transfer in Vaporization of Film Flowing Down a Specially Shaped Surface. Heat Transfer Soviet Research, 1976. 8: p. 84 87.

55 .. '

104. Rifert, Barabash, Zadiraka, and Andreyev, The Mechanism and Rate of . Heat Transfer in Vaporization of a Film Folwing Down a Spaecially Shaped Surface. Heat Transfer Soviet Research, 76. 8: p. 84 87.

105. Rogers J.T., Laminar Falling Film Flow and Heat Transfer Characteristic on Horizontal tubes. The Canadian Journal of Chemical Engineering, 1981. 59: p. 213.

106. Rogers J.T. and Goindi S.S., Experimental Laminar Falling Film Heat Transfer Coefficients on a Large Diametrer Horizontal Tube. The Canadian Journal of Chemical Engineering, 1989. 67: p. 561.

107. Rohtfus, R. and C.P. Newman .. in Proceeding of the Fifth Ocean Thermal Energy Convertion OTEC. 1978. Miami Beach:

108. Sabin C.M., Nucleate vaporization and extended surface heat transfer at low temperatura difference. 1976, Geoscience Ltd:

109. Sabin C.M. and Poppendiek. Heat Transfer Enhancement for evaporator. in Fourth Ocean Thermal Energy Conference. 77. New Orlean:

110. Sabin C.M. and Poppendiek. Film Evaporator of Ammonia over Horizontal round Tubes. in Proceeding of the Fifth Ocean Thermal Energy Convertion OTEC, Miami Beach. 78. Miami Beach:

111. Seban A.R. Transport to Falling Film and Heat Transfer Characteristic on Horizontal Tubes. in Sixth International Heat Transfer Conference. 1878. Toronto:

112. Sergeyev and Kovalenko, Heat and Mass Transfer in Falling Films wiht Low Frequency Pertubations of the interface. Heat Transfer Soviet Research, 81. 113: p. 96 104.

113. Sernas, Heat Transfer correlation for subcooled water films on horizontal tubes. Transation of ASME, 1979. VOL 101 FEB 79: p. 176 .. 178.

114. Shen Vin Qui and Guo VI Hu. A Study of Horizontal Tube Spray Film Evaporation And Its Enhancement. in XVII International Congress of Refrigeration. 1987.

56

115. Shmerler and Mudawwar, Local Heat Transfer Coefficient in Wavy Free Falling Turbulent Liquid Films Undergoing uniform sensible heating. Int. Journal Heat Mass Transfer, 1988. 31: p. 66 77.

116. Shmerler and Mudawwar, Local Evaporative Heat Transfer Coefficient in Turbulent free falling Liquid Films. Int. Journal Heat Mass Transfer, 1988. 31: p. 731 742.

117. Sideman, Horn, and Moalen, Transpor Characteriscs of Films Flowing over horizontal Tubes. Int. J. Heat. Mass Transfer, 1978. 21: p. 285 293.

118. Sideman S., Some Design Aspect of Thin Film Film Evaporators /Condesers in Water Desalination. 1981. : p. 681-703.

119. Sinek, J.R. and E.H. Young, Chem.Eng.Progr, 1962. 58 N12: p. 74.

~ .. "

120. Siu Ming Yih and C.C. Kuo, Desing and Testing of a New Type of Falling Film Gas Liquid contancing Device. AUChE, 1988. 34 N3: p. 499 501.

121. Spegele. Horizontal Tube Thin Film Boiling. in XIVth Inst. Refrigeration. 1975. Moscow:

122. Swart, R.H., Capillary Tube Heat Exchangers. Refrigerating Engineering, 1946. September: p. 221-224, 248-249.

123. Taghavi , K., Efffect of Surface Curvature on Laminar Film Condensation. J of Heat Mass Transfer, 1980. 23: p. 1433 1445.

124. Tan, Y., G. Wang, S. Wang, and N. Cui. An Experimental Research of Spray Falling Film Boiling on The Second Generation Mechanically Made Porous Surface Tubes. in Ninth Interantional Heat Transfer Conference. 1990. Jerusalen ( Israel):

125. Tang.J., Lin S., Lu Z., and Yu-chi.' Droplet Spacing of Falling Film Flow on Horizontal Tube Bundles. in XVIII th Internationa Congress of Refrigeration,. 1991. Montreal:

126. Taubman and Kalishevich, Experimental and Theorical Study of Feat Transfer in Evaporation of a Liquid in a falling film. Heat Transfer Soviet

57

Research, 76. 8: p. 1 8.

127. Thomas ,A., J.J. Lorenz ,D.T. Young, and N.F. Sather. Performance Tests of the 1 Mwt Shell and Tube and Compact Heat Exchanger for OTEC. in Six Ocean Thermal Energy Conversion (OTEC). 79. Washignton:

128. Tolubinskiy, Antonnenko, and Ostrovskiy, Heat Transfer during Vaporization in Thin Films. Heat Transfer during Vaporization in Thin Films, 1976. 6: p. 109 114.

129. Toshihiko Fujita and T. Ueda, Heat Transfer to Falling Films and Flims Breakdowns. Int J. Heat & Mass Transfer., 78. 21: p. 97 .. 108.

~ ...

130. Tree, D.L. and W.A. Helmer, Experimental Heat and Mass Transfer Data for Condensing Flow in a Parallel Plate Heat Exchanger. AHSRAE Transactions, 1976. 82 Part 1: p. 289-299.

131. Tsay, Y.L., Lin T.F., and Yan W.M., Cooling of a Falling Liquid Film Through interfacial Heat and Mass Transfer. Int.J. Multiphase Flow, 1990. 16: p. 853.

132. Ueda T. and T. H., Measurement of velocity, Temperature and Velocity Fluctuation Distribution in Falling Film Evaporator. Int. J. Multiphase flow, 1975. 2: p. 261.

133. W.H.Parken and L.S.Fletcher. Heat T!ansfer in Thin films flowing over Horizontal Tubes. in Proc. Int. Heat Transfer Conf. 1982.

134. Wallis, G.B., One Dimension Two Fase Flow, , M. Hill, Editor. 1969, New York. p. 377.

135. Wang, G., Y. Tan, and S. Wang. A Study of Spay Fallign Film Boiling on Horizontal Mechaincally Made Porous Surface Tubes. in International Symposium of Heat Transfer Enhancement and Energy Conservaton (ISHTEEC). 1988. (ISHTEEC) China:

136. Wang, Youchum, Chen, and Huanzhuo. Rayleigh Taylor Stability Analysis of a Condensate Film Superposed on an Infinitely Thick Layer of Vapor. in Advances in Phase Change Heat Transfer, Chohgqing. 1988. Sichuan China:

58

137. Webb L.R., The evolution of Enhanced Surface Geometries for Nucleate Boiling. Heat Transfer Engineering, 81. 2: p. 46 69.

138. Webb. L.R., Nucleate Boiling on Porous Coated Surfaces. Heat Transfer Engineering, 1983. 4: p. 71 82.

139. Wenzel, R.N., Resistance of Solid Surfaces to Wetting by Water. Ind. and Eng Chemistry, 1936. 28: p. 988-944.

140. Westwater, S.Y.a.J./., Effect of Comercial Enhancement Surfaces on the Boiling Heat Transfer. Curve. Advances in Enhanced Heat Transfer, ASME, New York, 1981, . 73-91.

141. Westwater, S.Y.a.J./., Effect of Comercial Enhancement Surfaces on the Boiling Heat Transfer Curve. Advances in Enhanced Heat Transfer, ASME, New York, 1981, .73-91.

142. Yundt, B., Horizontal Spary Film Evaporation. Chemical Engineering Progress, 1981. : p. 69.

143. Yung, Hillis, Lorenz, and Saytther, OTEC Performance Tests of the Union Carbide Enhancement Tube Condenser. 79, Argonne National Laboratory:

144. Yung D., Lorenz J.J., and Ganic E.N., Vapor liquid interaction an Entrainment in Falling Film Evaporators. J. Heat Transfer, 1980. VOL 102 FEB: p. 20 25.