Technical Note N- 1041

THE EFFECT OF COATINGS AND SURFACES ON DROPWISE CONDENSATION

By

S. C. Garg, Ph.D.

July 1969

This document has been approved for public releaseand sale; its distribution is unlimited.

NAVAL CIVIL ENGINEERING LABO'RATO?Port Hueneme, California 93041

Reproduced by the

CLEARINGHOUSEfor Federal Scientific & TechnicalInformation Springfield Va. 22151

THE EFFECT OF COATINGS AND SURFACES ON DROPWISE CONDENSATION

Technical Note N-1041

ZF 38.512.001.004

by

S. C. Garg, Ph.D.

ABSTRACT

A literature survey of the published information on the effect ofcoatings and surfaces on dropwise condensation has been carried out.The survey has shown that research to date in promoting dropwise conden-sation has been limited mostly to small scale laboratory experiients.There appears to be a serious lack of data on methods of application ofvarious dropwise promoters, useful life of promoters, effective c-oncen-tration, and frequency of promoter renewal. This lack of information ishighlighted by the total absence of dropwise condensation as a fac'tor indesign of condensers in industry.

Recommendations are made to initiate an experimental program toevaluate the effects of vibrating the condenser surface upon dropwisecondensation and the coefficients of heat transfer.

This document has been approved for public release and sale;its distribution is unlimited.

I

CONTENTS

Page, INTRODUCTION 1

MECHANISM OF HEAT TRANSFER IN DROPWISE CONDENSATION 2

LITERATURE SURVEY 7Criteria for Dropwise Condensation 7Dropwise Condensation Using Permanent Coatings 8Dropwise Condensation Using Organic-Promoters' -12Increased Heat Transfer Rate by Mechanical Wiping 24Effect of Condenser Surface 25

COMMERCIAL APPLICATIONS OF- DROPWISE CONDENSATION 27Full Scale Tests 27Effect of Corrosion inhibitors 28Requirements for Practical Use of Dropwise Condensation 28

-CONCLUSIONS 29

RECOMMENDATIONS 31

REFERENCES 32

NOMENCLATURE 36

FIGURES, 38 -

+ •

1, °

1+

INTRODUCTION

In power'generating systems utilizing steam (e.g., electricalgeneration, ship's propulsion, and nuclear power reactor applications),the steam leaving the power unit is reconverted into water in a condenserdesigned to transfer heat from the stean to the cooling water as rapidlyand as efficiently as possible.

The mode of condensation of water vapor in these condensers isimportant since it deiermines the efficiency of the condenser. Thereare two clearly distinguishable ideal modes of condensation, namely (1)filmwise - where the condensate forms a continuous film on the condensingsurface, and (2) dropwise - where the condensate is in the form of smalldrops which are segments of spheres similar in shape, but differing insize. In--practical condensers, however, tie condensate is usuallydistributed over the cooled surface in an irregular manner in which someareas approximare-;dropwise-condensation and others filmwise condensation.This -mode is generally described as mixed condensation.

There is considerable advantage in preserving dropwise condensation,particularly when large heat transfer -rates are desired, since the vaporto condensing surface heat transfer coefficient is of the order of 10times that for filmwise condensation, and the "overall" coefficient ofheat transfer is 2 to 3 times that for filmwise condensation. Increasedrates of heat transfer in dropwise condensation may be ascribed to (1)the conduction of heat through the surfaces of drops being greater thanthat through an equal quantity of liquid spread as a uniform film overthe same surface area, (2) the more rapid removal of liquid from thecondensing surface in drop form due to coalescence, and (3) the high heattransfer rate through the exposed area between the drops or through trixvery small drops.i

Considerable work has been carried out during the past 35 years orso in devising methods for initiating and maintaining dropwise condensa-tion. Recent technological advances (e.g., space power systems andsaline water conversion) have brought about a renewed and vigorousinterest in the phenomenon of dropwise condensation. In these cases ascontrasted with more conventional systems, the resistance to heat transferprovided by the condensing steam is a significant part of the totalresistance so that an increase in the steam side coefficients can have ameasurable influence on the overall efficiency of the system. In alarge desalination plant, for example, approximately 50% of the expenseis in the heat exchanger area and associated equipment, and an increasein the heat transfer efficiency would result in a considerable savingboth in investment and in annual operating cost. Therefore, if methodscould be devised to promote and maintain dropwise condensation continuously

for an indefinite period of time, the increased heat transfer ratesc6uld-be translated directly into a proportionate reduction of space,weight, and. the cost of condensing equipments.

Various system variables such as heat flux, vapor pressure, surface

temperature, -surface finish, nature of surface and wettability, presenceof impurities and noncondensible gases, vapor velocity, surface orienta-tion, thermodynamic and transport pxoperties of the condensing vapor,and presence of dropwise condensation promoters affect the coefficientsof heat transfer in dropwise condensation. This report details thefindings of a comprehensive literature survey carried out to determinethe effects of coatings and surfaces on dropwise condensation.

MECHANISM OF HEAT TRANSFER IN DROPWISE CONDENSATION

Prior to discussing the effect of coatings and surfaces on coefficientof heat transfer in dropwise condensation, it will be appropriate todiscuss various theories on the mechanism of dropwise condensation, ithas been accepted by all investigators that the heat transfer rates -ndropwise condensation are much higher than those in filmwise condensationunder otherwise identical experimental conditions. However, there is agreat deal of disagreement upon the precise m-:chanism by which drops ofcondensation originate. There are two widely different theories proposedin the past, along with a number of variations of these two theories asexplanations of the formation of the droplet

The first main theory was proposed by Jakob2 in 1936 which may besunmarized as follows. The high heat transfer or condensation rate indropwise condensation is due to a direct contact of the dry coolingsurface with hot stean impinging upon it. The condensing steam coversthe cooling surface forming a thin layer of water which continually andquickly grows in thickness until it fractures to form droplets. Thisthickness varies between zero and a small value 6. As the drops coalesceand roll off-the surface, a portion of dry surface is exposed and almostinstantaneously covered by condensation of fresh steam such that theprocess repeats itself. For a given steam condition, the average thick-ness of this layer will depend mainly upon the behavior of the nucleiinitiating the condensation, upon the surface, and slightly upon therate of cooling. if the cooling rate is high (e.g., using a high vaporto surface temperature drop), the drops of a given size are developed bya layer of a given thickness but in shorter intervals of time and surfacearea. In this case, greater cooling effect is brought into and throughthe layer, more steam condenses and more drops form, all without changingthe average thickness of the layer. The temperature drop across a layerof this kind was assumed tc be the same as predicted by the Nusselt theoryfor filmwise condensation- However, the experimental results enable one

to compute two limiting values for the average film thickness, 6.

From the basic equations of heat transfer,

Q = h AT ()

where h = coefficient of heat transfer, Btu/hr fg2 OF

Q = rate of heat flow, Btu/hr

and AT = vapor to surface temperature difference, OF

and for heat conduction through a plane layer,

Q = k ATQ -k - (2)

where k = thermal conductivity of the condensate, Btu/hr ft OF

we obtain:

6 k/h (3)

Using the measured value of h = 14,000 Btu/hr ft2 OF, and assuming,the layer to consist of steam (k = 0.0137 Btu/hr ft OF), the mean layerthickness would be:

8 z 0.98 x l0 6 ft ( 0.0003 m) -

corresponding to about 1,000 diameters of a molecule, this being a verythin layer, since the molecules have a greater distance between themin the steam. If, on the other hand, the layer should consist of waterin liquid form (k= 0.40 Btu/hr ft OF), its mean thickness would be:

6=0.4 -28.6 x 10- 6 ft (0.009 mm)14,000

In view of the above two limiting values, the thickness 6 may beconsidered to be of the order of 0.001 mm. In accordance with this, afilm such as Nusselt's is assumed to exist, but the following essentialdifferences. Nusselt's film is a stable layer of continually increasingthickness whereas the layer suggested by Jakob is unstable and changesquickly, but is of a constant mean thickness. The thickness of Nusselt'sfilm depends on the rate of cooling, the height of the cooling-surface,etc., whereas the thickness of the jakob's layer in dropwise condensationdepends on the properties of the surface, the purity of steam, etc.Only the rapidity with which new layers originate depends on the intensityor rate of cooling.

The second main theory of dropwise condensation was offered by Emmons3

in 1939. He suggested that the molecular processes of evaporation,

3

condensation, and reflection of molecules at surfaces, studied in con-nection with catalytic and electronic emission phenomena, can be appliedin a qualitative way to dropwise condensation.

Consider a vapor in contact with a surface maintained at a lowertemperature than the vapor. When the rate of arrival of vapor moleculesexceeds the rate of their evaporation from the surface, there is a netaccumulation of molecules on the surface. Before the vapor moleculeshave completely covered the surface, some vapor molecules will arriveon top of the first layer. These will arrive at the same rate as before,but, since the range of atomic forces is extremely small, the rate of

evaporation will depend on whether the vapor molecules have a greateror lesser affinity for each other than for the cooling surface.

When the surface has an equal or greater affinity for the vapormolecules than the vapor molecules have for each other, the moleculesaccumulate faster in the first layer than in the second and third layers,and so on. The cooling surface soon becomes completely covered by a

layer of condensed molecules. In order for more layers to form, thelatent heat must now be conducted through the layer already present.Finally, as the number of layers became large, they slip over each otherdue to gravity, and the condensate thus flows off without exposing thecooling surface. This is filmwise condensation.

On the other hand, when the surface has less affinity for the vapormolecules than they have for each other, the second, third and subsequentlayers of molecules form before a first layer is completely built. Thecondensed molecules gather into drops, which grow until their weightbecomes sufficient to move theLa over the surface. Since the adsorptionforce between the first layer and the surface was assumed to be smallerthan the mutual attraction, these molecules are pulled from the coolingsurface when the drop moves downward, leaving behind a bare surface uponwhich the process can repeat itself. This is dropwise condensation.

According to Emmons, the space between the drops is bare in dropwisecondensation and the surface has an important effect on the phenomena.The vapor molecules condense on striking the bare cooled surface and aretherefore at a lower temperature than saturated vapor. The importanceof the process of reevaporation is shown approximately by the equationfor the rate of arrival of vapor molecules at the surface as given bythe kinetic theory of gases. The rate of condensation of saturated vapor

at a pressure of 1-inch of mercury absolute would be 3150 lbs/hr ft2 ifthere were no reevaporation. This is roughly 300 times as great asnormally used in power plant condenser practice.

According to Emmons, therefore, there must be a blanket of super-saturated vapor co4ering the bare cooling surface between the drops.Wherever this supersaturated vapor comes into contact with the surfaceof a drop, condensation occurs very rapidly. A local pzessure reduction

4

takes place which, in turn, sets-:up local eddy currents in the vaporbetween the drops. This mechanism is responsible for the very highheat transfer coefficients in dropwise condensation.

There is a serious disagreement between the above two theories ofdropwise condensatin,. Nagle commented on Emmon's theory as follows.The common experience of every man when he takes a bath is pertinentto dropwise condensation. When he is. in the water and he gets up, thefilm breaks almost at once and he is covered with drops. Therefore,he cannot have any supercooled steam or any reevaporation on him. Thesheet of water breaks and draws up into drops due to the oily natureof the surface rather than that the water that is deposited as conden-sate and is reevaporated. In addition, one would expect that reevapora-.tion of condensate and subsequent condensation on the surface of thedrops would result in an abnormally low heat transfer rate rather thanthe very high one found experimentally. The Emmon's mechanism is alsoin contradiction to that proposed by Eucken4 earlier. Eucken proposedthat liquid nucleation occurs in a thin layer of supersaturated vapornext to the surface, and that in this layer there is a flow towards thedrops because of a density gradient created by coadensation on thesurface of the drops.

In contrast to criticisms of Emmon's theory, Jakob's theory hasobtained support from a microscopic study of dropwise condensationconducted by Welch and Westwater.5 The authors took high speed motion.pictures during dropwise condensation on three different, verticallyaligned, condensing surfaces. Dropwise condensation was-induced byadding 0.005 weight percent cuprice oleate to the feed water in the Csteam-supply boiler. Steam to surface temperature drops varied from0.4 to 470F and the heat flux varied from about 5,000 to 85,000 Btu/hrft2. Their photographs showed that dropwise condensation takes placepreferentially on the swept regions and only slightly on the drops.The film which then forms on the swept regions builds up to an estimatedcritical thickness of about 0.5 micron before fracturing to form newdrops. The creation of essentially a "bare" surface between dropsprovided the explanation of high heat transfer coefficients obtained inthe dropwise condensation process.

The motion pictures5 showed that the drops were segments of spheresand had contact angles of 720 h 80. Numerous coalescences were observedto take place between microscopic drops. Whenever two or more dropscoalesced, a part of the metal surface previously covered by the dropswas exposed, and a distinct flash of light from the lustrous bare metalwas evident. Fresh steam condensed quickly on the cold surface untilthe luster was lost, indicating the presence of liquid, Shortly there-

I after, tiny new droplets became visible. At their first appearance, thedrops were considerably smaller than 0.01 mm and thus were too small topermit accurate diameter measurements. At low heat fluxes, the lustrousbare area was visible for a longer period of time compared to that at

* ,;

!5

high heat fluxes. The time for the first appearance of drops in a bare

area was found to be a function of AT but not a function of the size ofthe area. A definite interdependence of the steam to surface AT, averageheat flux, and the number of coalescences per unit area per unit timewas noted. As heat flux was increased, the heat transfer per coalescencedecreased, the bare area exposed per coalescence became smaller, and thenumber of coalescences between small drops increased. The drops grewfaster and the frequency at which the drops became big enough to rolldown the surface in response to gravity increased. A- large number ofsmall drops appeared in the swept path. The study indicated that about80% of the heat transfer occurred at the bare areas and in the thinliquid film which form on the bare areas.

Based upon the above observations, the authors proposed thefollowing equation to predict the average steam side heat transfercoefficient.

pX Xc ABh = -(4)

tc AT AT

where p = density of liquid, lb/ft3

= enthalpy difference between the vapor and thecondensed liquid, Btu/lb

Xc = critical film thickness at fracture, ft

tc = critical time for film growth, hr

AB = condenser area free of discrete drops, ft2

and AT = total area of the condenser surface, ft2

The values of Xc and AB/AT will depend on the properties of the liquid,

the solid, and the promoter. Since the local metal surface must show

temperature fluctuations as drops form, coalesce, and roll off, thelocal AT is not constant, and varies between zero and a maximum value.In the experimental program, the average values of h and AT were measuredexperimentally and that of tc obtained from the motion pictures.

Their conclusions may be summed up as follows. For water condensingon copper in a dropwise fashion, the condensation directly on dropletsis insignificant for all drops big enough to be seen. The condensationoccurs primarily between drops. The region between drops contain aliquid film which builds up in thickness from zero to the critical valueat which time it fractures spontaneously to form new drops and newlyexposed bare areas. The bare areas cause growth of a new layer which

6

fractures again after reaching the critical thickness. Whenever a filmfractures, the drops on its periphery gain liquid by coalescing withsome of the fractured liquid film. At a AT of 20F, as many as 1000.coalescences in series were observed in the films which resulted in onefinal drop of about 2 mm diameter after about 7 seconds.

In contrast with the above film fracture theory, Umur and Griffith6

concluded from a recent investigation- of dropwise condensation that afilm no greater than a monolayer thick can exist in thearea betweendrops. This finding is based on both theoretical considerations froma thermodynamic point-of-view and on an optical study. The opticalstudy involved determining film thickness by noting changes in theellipticity of plant polarized light when reflected from the metalliccondensing surface. These investigators also proposed a model tor dropgrowth which is consistant with experimentally observed values atatmospheric pressure and at lower pressures which'have shown that growth *rate is a function of vapor pressure.

McCormick and Westwater7 ,8 and Peterson and Westwater9 consideredthe initial formation of drops to be the principal mechanism i. drop-wise condensation. They investigated 7 the identity, behavior, andeffective range of nucleation sites during dropwise condensation-.Furthermore, using photographic techniques,8 they found that the heatflux at a particular pressure and AT was dependent upon the populationof active sites. They observed 8,9 that the number of active sitesalways increased as AT increased and concluded that drop coalescenceplays an active -role in dropwise condensation because newdrops cannucleate only on active sites exposed to the vapor.

The latest theory of dropwise condensation was offered by LeFevreand Rose1 0 in 1966. They considered heat transfer arising from condensa-tion on a single drop and, by means of an assumed distribution of dropsizes, deduced the mean heat transfer rate for the whole surface. Thethree principal physical phenomena considered in the evaluation of theheat transfer through a single drop were: the effect of surface tensionon the relations between the temperature and saturation pressures foreither side of a curved interface, conduction through the drop, anddissipation associated with matter transfer by condensation.

Thus, it appears that while advances have been made, the mechanismof dropwise condensation is far from being clearly understood.

LITERATURE SURVEY

Criteria for Dropwise Condensation

To analyze factors which determinie if the condensation process ina given system will be filwise or dropwise, it will be necessary toexamine the forces acting on a condensed drop. The forces acting upon

7Ii

a droplet adhering to a surface are represented in Figure 1, where71vC sv, and CyI represent the interfacial tensions which are equiva-

andto the surface lecular forces at the liquid-vapor, solid-vapor,- and solid-liquid interfaces. At equilibrium:

"= sv - sl = lvoea v a v cose (5)

where ! is defined as the Adhesion Tension. 6 is the contact angle ofthe droplet. For (,s. - as,) < 0, e > %i2 and the liquid withdrawsfrom the solid surface to form droplets. However, for (oSV - asl) > Glv,Equation (5) cannot be satisfied and there can be no equilibri thereby

j causing filmise condensation. Therefore, the function of a dropwisepromoter is to reduce the surface tension of the solid-vapor interface,while not reducing proportionately the solid-liquid and liqu-d-vaporinterfacial tensions.

Dropwise Condensation Using Permanent Coatings

Investigation of the phenomenon of dropwise condensation bas shownthat it is possible to stimulate and maintain drcirwise condensationusing promoters applied to the condensing surface either as permanenrcoatines or as continually or intermittently applied cocpounds. Ideally,a permanent nonettable surface would be the best method of producingdropwise condensation. This coating could be applied to the condensingsurface at the time of manufacture, and the condenser would continue toyield dropwise condensation, during its lifeti=e. However, the surveyhas shown that there -s no such things as a "peranent coating."Published information on various available permanent-type coatings, theireffectiveness in maintaining dropwise condensation, and their contribu-tion to the thermal resistance of the condensing surface are discussedbelowj.

Under a U. S. Navy Contract, Westinghouse11 investigated the useof permanent, nonwetting agents for the coating of steam condenser tubes.A screening apparatus was set up to check the heat transfer rates andindicate durability of several promoters prior to testing on actualpilot condenser. The promoters investigated in this manner were DowCorning Silicones DC-1107, R-671, XR-4010, R-875 and R-991, and teflon(polytetrafluoro ethylene). The initial check showed that XR-4010,R-875 and R-991 were completely ineffective. The performance of othercoatings is shown in Figure 2. The observed improvement in overallcoefficient of heat transfer over the uncoated plate was approximately507. for DC-1107, 20% for teflon, and 157. for R-671. An examination ofthe duration of effectiveness of these coatings on flat copper platesshowed that teflon was the only coating of sufficient life expectancy(over 125 hours) to warrant further investigation.

The hear exchanger tubes in the final rest were 90-10 copper nickelarranged horizontally in 4 rows and 9 columns, and separated by a vertri-cal baffyle in between the second and third raw. Thbe 18 cubes on one side.of &e baffle were coated with 0.5- =il, thick teflIon and the 18 tubes onthe other side of the baffle were uncoiaced to facilitate conoarison.:-he results are shown in Figure 3.

After approxinately 40 hours of ex.perimnration, dropswise condensa-tion was observed on the tuncoa red tubes a-Aso which was asciibed to.surface conra-inariina by oil carried over with the steam. the resultsfor the oil coataaina--ed tube are also plotted in Figure 3 for conpari-son. It =ay be seen zhat: the overall coefficient of hear transferirproved by 22X and 50% respectively for the teflon coated aid thecontaminated cubes over the uncoated c ues.

Some exoerimenal work has bee=- carried out at the U. S. NavalShip Research and Devrelopcaenc Center, iknnapolis, Maryland, to determinethe effectiveness and e-urability of teflon over a period of time- Forthi--s purpose, a s=al1 3-pass, 74-tube, air-ejector after-condenser wasw)dified to serve as a rest condenser for the teflon coated tubes. 1hetubes were 518-inen- O-D., No. l8MC- of 90~-10 copper-nickel, coated with40.0005-inch teflon on the ou-rside-12

The corAdenser was otterated for a total period of 453-4 hours withhear transfer evaluation tests at four intervals; at the start of theresc, and aftrer 708, 2652, and 4534 hours. The perforc'Mmce of thetefloan coated tubhe condenser after 453-4 hours shoued a reduction in theoverall hear transfer coefficient Of approxinacely 22% of the value atthe start of test- This decrease was z'rtributed to a gradual transitionfron dreawise to fIinwise condensation which increased the -hear flowresistance. T1his apparently occurred due to a thin film of iron oxidethat enveloped the tube condensing surface and affected the noneerringproperty of th-, teflon coating. icr-rever, evaluation tests conductedafter the steen side of the tubes was cleaned shcrzed that son oernanentdeterioration of the nonwerring property of teflon had occurred.Thre=oval of iron oxide increased the heat transfer coefficient, but didnot restore the= to the original value obtained at the beginning of thetests. The results of these tests are tabulated in Table 1.

Topper and 3aerU3 introduced the concept of using teflon on tubularsurfaces to obtain dropwise condensation as early as 1955. They showedconsiderable inoroveziear in the overall coefficient of heat transfer.Later, SwortzeI'-4 used teflon to induce droprwise condensation of steanOn a vertical condensing tube. The condensation was carried out atatmospheric pressure on a 35-inch long, 1/2-inch I.D. copper tube- Themethod of application of teflon consisted of first lightly sandblastingthe tube to assure a clean surface follosred by spraying the tube withone coat of teflon and fusing it in an electric oven at 7000?. Thethickness of the teflon film was found to be less than 0.001 inch. He

9

'A

found the condensing stean fil_ coefficients to be fron 240 to 3300 per-cent over those of the coPper tube before it was coated with the teflonpromoter. 1

Depaw and Reisbig conducted an exuerinea-tal program to study thebehavior of a 0.00025-inch thick filz of ref ion to pro~te dro7.aisecondensation on a 112-iitch 0.D. alunintn tube. The tube was mountedhorizontally and both the dropwise and filzsdse condensations werestudied. They found that curves of heat transfer coefficients versusvapor to surface temperature drop in droowise and filrse condensationcoaverged as the temperature drop bece large. This was caused by therapid formation of drops which rend to blanket the condenser surfacewith liquid at kigh heat fluxes. The authors Also found that highestheat transfer coefficients were obtained with thinnest teflon fil=~because of the high thermal resistance of teflon. A teflon film thick-ness of 0.00025-inch provided 12.5 tim-es as =zch thermal resistance asthe 0.02-inch thick aluninun wall of the condenser tube and about 3rimes more than the condensIng frim itself. This shows thar the thick-ness of the reflon film is an i:nporranc parameter and it should be keptat a rini= to imorove the overall perforrance of the tef omi-coaredcondenser tubes. ir is claimed that teflon films as thin as 0.00005inch can be produ~ced although the productica of such a thin filzi willbe most difficult.

A short experimenzaL_ program to induce dropwise condensation usingteflon w~s also carried out at the Naval Civil Engineering !aboratoryand reported in 19063 by Saternino.J6 He observed that dropwise condensa-tion gave overall heat transfer coeffIcients; which were 2 to 5 tinesgreater than those corresponding to fibsu.ise condensation. The teflonthicilness in these experiment-s vas estimated at 0.001 inch.

Erb and Thelen17.3 8 and Erb)1 9 , 2 0 have been conducting a researchprogram in dropwise condensation at the Franklin institute ResearchLaboratory usinf gold, silver, and ?arylene-N Polymers, as condensationpromoters. Erb-9 reported that a 0.17a thick coating of gold on 2tthick silver on 127v. thick nickel on 90-10 copper-nickel tubes produceddropw.ise condensation even after 3650 hours of continuous operation vithstemz at 1010?. An imnrover;ent in the overall coefficient of heat trans-fer of 50% was obtained. The estimated cost of this coating vas $1.02per sq ft of the tube surface. Erb and Tbelen1 7 have reported that acoatiug of silver on 316 stainless steeal tubes provided good dropwisecondensation from distilled water as well as seawiater. A coating ofsilver sulphide on nild steel tubes-was also claized to provide drop-wisa condensation for over 7500 hours of continuous operation. The useof Parylene-li polymers in romoting drop'wise condensation was also demon-strated by Erb and Thelen. They formed a 0.25p thick film of Parylene-ii polymer coating on chrome-plated 90-10 copper-nickel tubes and obtaineddraywise condensation for over 2600 hours of continuous operation. Usingsteam at 1140 C, an increase in the overall coefficient of heat transferof 387. was obtained.

10

-0

-,4 - 0

- =' -! c m c

C-4 -

0

CEj

0M Ck co

0' -.d

a 0

10 0J 100 C%- 0 0~ S 1 L-

000 -4 -- 0 7 0

00 0-4

aecently, Erb2 0 has been successful in =rintaining dropw.ise conden-

satien coatinuously for over a year on 90-10 copper-nickel which had a5: thick bright silver deposit folowed by a 0.025r thick electro-deposited gold. Co=pared to the unplated tube, the heat transferefficiency was found to be increased by up EC 100%. Also, he hasobserved a 40Z increase in the overall coefficient of heat transfer after2 years of continuous condensation using I- thick deposit of Parylene-1on chromium plated 90-10 copper-nickel. Additional --#d-es are now beingconducted on coatinpbased on gold, silver, Parylene-N pol-ers, teflon,and other fluoro-carbon resins.

Dropwise Condensation Using Organic Promoters

No condenser tube material having the co=bina-ion of ther-al,mechinical and che.ical properties required for sustained dropwisecondensation has yet been found. Therefore, the condenser surfaceproperties must be altered ith the application of a thin layer of so=esubstance having a l1w affinity for the vapor and a very high affinityfor the metal.

Investigators have been testing a wide variety of organic cop-.oundsto produce dropirise condensation. Any co=pound which consists of arelatively active radical would be suitable for this purpose if thezolecules were oriented -wth the active part adsorbed on the coolingsurface, and the inactive part acting as a surface upon which the vapormolecules can condense. in such an arrangement, the active promoterneeds to be only one molecular layer thick. in computing the amountof promoter required to form one molecular thick layer, allc.ance shouldbe made for the relatively rough surface finish of commercially availa-ble condensing surfaces which will result in a corresponding increasein the active surface area.

Drew, Nagle and Snith2l conducted one of the earliest series oftests to determine the effectiveness of various organic compounds inpromoting dropwise condensation. Their findings are detailed in Table 2.

Only those compounds which caused drop-..se condensation to be maintainedfor at least 24 hours without promoter renewal are listed in the Table.Host of these tests were carried out with steam condensing at atmosphericpressure. They concluded that:

- mineral oils are ineffective as promoters for condensation onmetal surfaces

- higher molecular weight fatty acids are very effective promotersfor all metals

- alcohols, organic nitrogen compounds, and halid j are ineffectiveas promoters

12

- only chose substances hicn are strongly absorbed or orherwis=firmly held Co the condensing surface are significant dropwisepromoters

Nagle et al2 2 conduCted cesrs on a 2-11/2-inch ID.. x 2-718-inch O.D. x24-inch long iertically counted copper pipe. Droawise condensation of

stean was carried out on the outer surface of this cube at 5 psig usingaleic acid as the dro~rise promoter. 1ney obtained overall heat transfercoefficients which were 59Z to 350Z greater than chose in fil.tise con-densaEion under identical operating conditions, with an average i-.rave-ment of about S9Z.

Fatica and Katz 2 "3 carried our =easurements of heat transfer, conractangle, and nuber of drops per unit area on a 3-inch x l-Anch copperplate mounted vertically. The dro.rwrise condensation was pro-_z.ed bystearic and oleic acids on copper, chro-ed copper, and nickel placed

copper surfaces. They observed a substantial increase in the overallcoefficient of heat E:ansfer compared to that obtained in fi'; ise con-

densation. Ihe ev.per-ence of Faz=pson2 4 in Britain, however, ha. Shcum-

that the use of oleic acid as a promoter was not always successful -

%-hereas the use of benzyl mrcaptan was always successful on copper or

brass surfaces. Ha-pson2 5 also fourd xanthates, dithiophosphates,synthetic resins such as bakelire and the siliconez r* be effective in

aintaining dropwise condensation. He alsu noted that an excess ofpromoter on the surface, w-nich resulted from its injections into thesteam, caused a considerable reduction in the heat transfer rate indro wise co-ndensation. His work included investigation of the efiects

of surface finish on droFw.ise condensation which will be dealt laterin this survey.

Filming anines, e.g., '.3H 7 "NH2 , have been shown by Tanzola andWied=an26 to induce droFise !;-densa-ion on metal surfaces therebyincreasing the overall coefficient of heat transfer by a factor of

three i.en used in full scale steam plants. This increase is very sub-

stantial in view of the fact that the overall coefficient of heat trans-fer is controlled by a number of thermal resistances only one of which

is controlled by the condensation mode. This may be illustrated by an

example. Consider a practical condenser as represented in Figure 4.

The overall coefficient of heat transfer may then be written:

1 t ' I l(6)U Kh 1 12 h3

where K = thermal conductivity of metal wall of thecondenser, Btu/hr ft OF

t = thickness of the metal wall of condenser, ft

hl, h 2 and h3 = coefficients of heat transfer as shcwn in Figure 4

13

0 a a

$a a z / asi

U 9 is Iao 0 04 -4LI- 0 -

a ~ 0 0 0 0~5 %d udd CI C

a C- on4 go :0

0 aa 0 W 4 aa iC. CA x 10 0- C0 a. U. r-

a- r- 14 3 - - CL IS r- r- mC04 I 40 0 a 0 0 0

*j10 0 C. x 0 u 0 0 0

-n u ~ -9

Si C-4 =) .4 ra fl. q 1-44

UC 0i C) .4 S .4w a 4 0 a (

0 V 0 o0 u -4 SaCVJIA4 Si 6 . aU C ~ 5 ~ U

0 0 a 0 9 V ' 14 14J.1 41 14 40 -4 co eO c) a 00

is P a 0 9 to v0 0 0a~ -4 4 v. w0 0q %4U

02 go 0. aO C) VD aC 0 -

Ia 002a 4 P

S I S

0. c:3 -r4 0s 0 IL 4C)0 Q -

b0. 0 4 100

UI Ii4 C: I 4 3J)

14

Assuming the condensing surface to be 1/4-inch thick cast iron, aconductance of condensing steam of 1000 Btu/hr ft2 OF, a conductanceof heated water of 300 Btu/hr ft2 OF, and a corrosion film conductanceof 150 Btu/hr ft2 OF, the overall coefficient of heat transfer is givenby: 1 4

0. 2 5 1 1 1

320 1000 300 150

= 85 Btu/hr ft2 OF

The use of a promoter generally prevents corrosion of the condensingsurface therby eliminating its resistance to heat transfer. It raynow be noted that to obtain a 3-fold increase in the overall coefficient,a 20-fold increase in condensing coefficient is necessary. To translatethis order of magnitude of improve-ent in the condensing coefficient tothe overall coefficient, it will be necesszry to increase the water sideconductance and decrease the thermal resistance of the metal wall by

similar proportions. 27Blackman and coworkers investigated a number of co-mpounds which

could be used to promote dropwise condensation. They concluded thatactive promoters have in the molecule an unhindered hydrocarbon chainwhich confers water-repellency on the compound and an "anchoring" groupwhich is usually a divalent sulphur atom. This investigation is con-sidered to be the most elaborate study in dropwise promoters to date,and will be discussed in some detail here in or-dir to list the variouspromoters, their effective lives, and their minimum concentrationnecessary to cause dropwise condensation.

In all of the tests, the following precautions and method ofpromoter application to the condensing surfaces were observed. Alltraces of grease and other possible contaminants were removed from theapparatus. The condensing surface consisted of bright polished metaland the condenser tube was completely wettable with cold water. Succes-sive treatments with emery paper, mild abrasive powder, and a mixtureof mild abrasive powder and detergent powder produced the desired con-densing surface. The promoter was generally applied to this condensingsurface in the form of a solution in a low boiling solvent, e.g., ether,acetane or carbon tetrachloride. A solution strength of about 17. wasused.

The compounds listed in Table 3 were found to produce pertect drop-wise condensation (i.e., no filmwise condensation) for at least 500

hours, during which time the condensate showed no change in appearance.For some promoters, condensation was terminated for a few days andrestarted successfully. Some other promoters were tested for even longerperiods of time and found to maintain dropwise condensation over the

entire duration of testing with no deterioration. These compounds are

15

g

listed in Table 4. Three promoters were tested on brass and coppersurfaces to determine minimum promoter concentration necessary to sus-txAn dropwise condensation. The minimum effective concentrations of.these promoters are listed in Table 5. The autb.ors concluded that the

ainiwm effective concentration of the compound might depend upon anyone or more of the following factors: (a) actual concentration of thecompou-d, (b) the molecular concentration of the compound, (c) themolecular concentration of sulphur, and (d) the molecular concentrationof the active hydrocarbon radical.

Table 6 lists a total of 13 compounds which promoted perfect drop-wise condensation but were tested ior only 2 hours. Four other com-pounds tested were found to be either poor or completely ineffectivedropwvse condensation promoters. The ineffectiveness of sodiumdodecylsulphonate (C12H2 5 SO3-Na) led the authors to believe that asulphur atom having all of its valence electrons in use cannot form abond with the metal surface. This conclusion was confirmed by testingdioctadecyl iulphoxide [(C18H3 7 )2S0 ] and sulphane [(C185 7)2SO2] foreffectivemess as dropwise condensation promoters. The former, which hastwo unused valence electrons, was found to be an effective dropwisepromoter whereas the latter, which has no unused valence electrons, didnot promote dropwise condensation-

The authors concluded that compound which, when oriented at themetal surface in a monolayer, presented an "unhindered" hydrocarbonsurface to the steam, were successful in promoting drcpwise condensation.The surface-active groups contained divalent sulpblzr or selenium. Nocorrelation between the structure of the coMpourd and their effectivelife was obtained.

Furman and Hampson2 8 have also reported an experimental investiga-tion in filimise aid dropwise condensation of steam. A strong solutionof dioctadecyl xanteo , C1 8 H3 7 -0-C-S-S-C8H 3 7 , in carbon tetrachloridewas applied to the condensing surface to promote dropwise condensationwhich improved the overall coefficients of heat transfer by a factor ofbetween 2 and 3. Welch and Westwater4 found cupric oleate an excellentpromoter on copper, stainless steel, inconel and copper-nickel surfaces.The promoter was found to be ineffective on aluminum and monel. With

this promoter using a copper surface, perfect dropwise condensationoccurred with ethylene gly,;ol and glycerine.

Bobco and Gosman2 9 used fluorochemicals as dropwise promoters.

Perfluorinated acids were found to produce dropwise condensation ofpure vapors of ico-octane, iso-heptane, cyclo-hexane, carbon disul-phide, and decalin. The use of these compounds in practical condensersis, however, questionable due to their corrosive action on the condensingsurface and their short useful life as condensation promoters.

1

11

ci -0

r-00 a

0 0

0 a 00 = 0Cl .l -j 0i 0 ) c-) r- 0i -'-A4 - a

r 01 Li 414 0. 3 4 0 - M0 S.

0 -A 14 cl -4v 0 o 1 0 = w0 i CJ -0 'o 0> 0l 1 L - A C-I -0 m

pLi 0 i 0 z U i P-. C3 fm 41 C.0 4 0 0 CJ - C 0 :- j a - I %-

0 c i c: r ) c z C 0 - -:i w 0 0 -4 a .0 0 >1L -4 0..-4

"4d r4 r4 "4 0 40 Co -i Li 0 00 4 0.1

U4. V -4 0 Z 0 J -0 0 4 0Zo' i 0 0 0 (1) . 0 v w0.0

0. .C) C 0 r4 r4 va 4 -1 r04 C C* 4.)'-.'--~~ l~ i 0 0 -c 0 4.1 0l

0 a 1:3~ 0. a i 0- U .r a a i u 0 V4 -r 0 0 1.tao 0 C) 00 C) a ) 0 C Li go 0.to0 0>

-ja 0 0 0 a 040 Li 0'~ Li 0 p .0C 0 0t0I0-0 .1 V IJ " 4-I W- 0n 0I

LC40 0 0 -4 0 U 0 C J . r > 0 i ,4 .-4

La Li Li -4 0 >0 0 E4H M )

04 W 4 U U U U c d~4 0

*0u 0 0 a C 0 00 . E4440U4

3-I Li

to w caol

6 /3Ol

. 0 0' L C 0-

0' CO C) j co/200 C)

0/ 04 0l C14 0 C')to c

.0 * 0 0AV 0 . = %.4 4V

0 0 ( 0 0 z' CO %T4 C4 It

0 U' ')C ) W rn 0n C) 0LA4 .- 4 Lf , r 0 C') , t) 1 C4 'Y MO

-C) c)i C14 C14 *l 1:1 c) c .NC =C:'=C'- Ln = -= -= Z = 04 C'C 4 4 00 00

00= C4 4C4 M O)O 00 0 0 C4 r-4 c1-.tC'- rrn C ) C)l C)4 -1 4 ri -4 -1 '

0

17

Table 4. Compounds Promoting Dropwise Condensationfo r Very Long Period of Time (Reference 27).

Compound JDropwise Condensation for at least

C 6H5 *CH2 *S'C2 4C6H5 1730 hours

C 18H37 .0 *CS SC ? 5 1350 hours

C18" 37 .O.CS.S. (C12)10.S.CS.0C1 8 37 1680 hours

CH2 O.CO. (CH ) 10 S-CS-O H021

CH.O.CO.(CH 2)10 .S.CS.0.C2H 3530 hours

OCR -w CH )1 S-CSOC H5

C 2H25S(C1 ) 1850 hours

18

Table 5. Minimum Effective Concentration of Promoterfor'Dropwise Condensation (Reference 27).

Promoter Brass Surface Copper Surface

Dodecane thiol 0l~ .3C 12 H25 SH.1-0.3

Dodecyltrisethyithiosylane 02%015C12 0.25 0.125%C H5

-' ~~Dodecyl octadecanethiolate0.502517 3 500S* 12H25

19

Table 6. Comxmds that Promoted Perfect Rrop~dse Comde~sation(tesede only for 2 hours, reference 27).

1. C~~H 7 SS(e)c0

2. C 18g37S-CS-o. (CH) 10.O.CS -SC18 "37

3.c LH7'S12

5. H

H3 .cO.NH.CS.S.(c 2) 1 0 'C S 1 2 H256. Hs.(CH2 ) 10 .CO.S.(CH2 ) 10.SCO(CH 2) 1 0.SH

7. CH 3 CO*NH*CS*S* (CE 2 )1 0 ' .5. (CH 2 ) 10 .S.-Co. C )0 -SI--C'H

9. SeCN(CH )1 0 .Se-CN

2 10

10. C H 25 SH

11. C18H3, S.C. (NH2', 2 *O'c6 H 2(NO2)A

12. C H..18H37S1C2H25

*13. H 5 -S

20

i!zs as p'e..Thy rea~edt tbe =e of IOMs-ItZi =Mi'aspear 2= MI=N mo .. 7thes ==es =,cIst r ses acids Zaalcobols of Lrabcs wicb c!:21M lC=T'!S Of 23re 20 3:all~s am are ccmp--.rable wihte ibi~ amzme = m33rliekyarocarbers Czaing -C3 =a -CW "a=CE~rs- 1z Oice of te-::a =3 -'slaz-Cbors bat With CO=Sieerabl5 loge Ldr==ab3C CBZ-:ns. Xbey za.to zaiota!= higbest v~alue Of the overall b-ac. Mransfer crcffficlerzs cma copper tcbc, ic was cecessary to inject the gonorer at the rate of0.1 SM per So fr of the C n~!eMSer surIFZCe aM intervalS Of 20~ !=~S-The tim interval had to be reednced to 1l9 hours for scccessftAlccwsckn-2esarion O= a==i, orass and cr-pro-ziickel coeensing& scrfaces.

2--o ad COWDE-kers3 l perf=uned a s--ereE of fe rests C Mimeorganic promr-ers and fcar di;fferent. netels using induscrial sme-~ Ihefour nerals 'uere 70-30 brass, ZZ ab=mdnnzi bras, 70-30 copro-nickel a~=me!. The steam was obtaiced directly from the plat boiler and s~pplieeto the ccadensers at about atmospheric pressure. 7he candenser mbeswere promored by cermplete ine-rsion in a solutiom of the Pr roer for-15 m-inutes. The treated Cuibes were exposed to air for 2, 6, or 24 hursbefore being ficted into the condenser.. Eacrh combination of rnerai,prootrer, and air exposure was tested four tine-s. The promoters restedare listed in Table 7.

Mhe effective lives of the promoters were found to be inconsistentand, generally, conoaratively short, the average life varying bemeeru50 and 300 hours; except for Monel, f&r which all pr~aters gave veryshort lives. Thne general nature of the breakdown strongly suggestedthat failure was caused predominantly by fouling.

The testing was extended to include drop promotion by pr toerinjection into the steam at regular intervals. Using 11)0 c-c of 17.solution of promoter !,o. 9 of Ta2ble 7 in carbon tetrachloride, perfectdropwise condensation on all four types of tubes was obtained whichlasted for 100 hours. Dropwise condensation was maintained continuouslywhen the promoter was injected every 100 hours of operation. However,a 27. solution of the same compound in paraffin maintained perfect drop-wise condensation for a year of continuous opetation w-hen the con entra-tion of the promoter used was 0.05 ppmx of the steam condensed. Similarly,a 17. solution of compound No. 5 of Table 7 produced perfect dropurisezondensation on all 4 metal surfaces and maintained it for a year ofcontinuous operation when the promoter was injected at regular weeklyintervals. The amount of promoter used was 0.01 ppm of the steamcondensed.

21

Ti~1e 7. -O,-rise Cccsi~?ce(referec 3r1)

I. DiO=rae4=7l dlsee=.J-Ie C1!,rJeSeCS%7

2 Diccadecy1 xzrhare Cij3 S-O -C-s-S-CM2 3 7

-3 idacoecyi trithlocarb~ate C-,,S.C-SSC,2.5

4 Diocadecyl - -S-SC 1 8B3

5 Telrakis(edodecanerio)s i2e SI(S Y 27 4

6 Dodeanemis(etane chic)siian 1-iS- 2 H)

7 OSS-tri-ocrtadecai-e phoso~wrodirthloate ( 1 ~ ,SC 8 3I8 Octadecyl seleocyanate C SeCS

Q Motan w'ax

22

7besa erptrimncMs S~nesrc CEMC frer-emt Cleanir- of tr=Bes W~ remaoveSrezTse, scale, etc., may mo.e =ecessary far comirm-us droozise c -

saicm.. ME'rer -ze mrall C efficie=C of beat cransfer =s fodcaecrease& Wicb tim, as Sb~d in Ei-.re 5 fDor brass =6 Cnpr-nickel Mcez4Csir ecaisddeaetT.sln as Uze prczr. ~,it ap;;earsaat alcb=Zci injec:ion= of a promoctr at regniar- fmceruvals may be Suf-e.cient to qr~ome Zad =fmcaimdr_ s conde=nsatio for ce C10=cs

~ercio, he cze scsrface will have to be freqnamtly degreasedor cleazed co ensure the bLenefits Of 1dro-ise cocidensatIo n r= idicghigh co~erll heat transfer races.

Ezapsal ccndczed a srndy to decers-noe the darz~ilit' of v.arloraspronoters in =!Mcalaing drozwise condensaion. He conclcded that mixedprocaoters; =Inafaed dro~4se cone'sacion f~pr loger periods of timecizan sicgle promters, aan different promot~ers have differing effeccive-ness on various azecals..

Brunt and Y~nke-32 comecue ted an invescigation into the effect ofdro,~ise pr cers cc the operations of commrcial seawmter esaporators

-Ad shcrwed Irurowcvmcs in the or-rerall coefficient of beat transfer-.7hey coacluded Chat a reeuction of 30 percent in the capital cosct of Iseawater evaporators couldM be achieved using dropwise condensationpr coers reccamnded in reference 21.. To estimate the effectivenessor dropwrise condensation promoters in ccercial coadensers, Butcherand coworkers 3 3 conducced tests on the conideaser of: a 350 kw; turbo-altern-aror set-. They used solutiLons of nixtures of straight-chain fattyacids of hi~gh molecular weight in liquid paraffin as dronwi--se condensa-cion preotoers; by injecting the= into the turbine exhaust steam_ to givea tocal coverage in the condenser of about 1 =i/fz2 of. the beat transfersurface. The- observed that iz:iro-venent in condenser pe-rfor---nce wasless =arked thnan that obtained i--n sl lborizzry rigs. The i=Prove-meat in the overall coefficient of heat transfer was about 20%. it wassuggested that in~rovenent in the condenser perform-ance was limited bythe water side coefficient and the metal wall resistance.

Wirt and coworkcers 3 4 have also investigated the use of promotersin seawater evaporators. Their full scale rest program show-ed thatimprovements in overall heat transfer coefficients of 2bout 30X% can beachieved by using dropwise promoters. They concluded that injection ofdropwise promoters into steam is a valuable technique for maintaininghigh hear flures.

Bliss znd Harding3 5 used lard oil to promote dropwise condensation.Similar to the observations of Butcher 3 3 te lorpre htipoenents in heat transfer on the condensation side has little effect uponthe overall coefficient of heat transfer due to the metal wall resistanceand the coefficient of heat transfer in evaporation on the other side ofthe condensing surface. They noted substantial improvements in theoverall coefficient of heat traixsfe. when brine was maintained underforced circulation on the cooling side of the condensing surface.

23

3rown and Mweas consieered the effects of lew absolute pressureson the conensacion process. Sucb pressures are used in the steam turbinec=Aeasers. Two 316-n1h dimeter, 42-iaches locg, a&dralty brasi .rcbeswere usd as condersing tubes and coated cc the outside with a layer ofapprximately 0.0001 inch thick teflon to induce and =aintain drcpwisecondmation. 1hey found the o7erall coefficient of beat transfer toIncrease with the absolute pressure for bozh the drc-wise aud the film-vise condensations. The ,werall coefficients in dropwise condensationwere found to be between 2 to 3 tines those for filzwise condensation.The fall-off in the overall coefficieta of heat transfer with fallingpressures was more marked for dropudse coadensation than for filmalse

imara and Natsura3 7 used dimenthylpolysiozane, 1(CR3 ) 2 SIO,as a prmoter on a copper condensing surface and observed the coadensa-

Stion process microscopically. 7he copper place was 17 ca x 4 cam x 0.3 cmthick. Their observation confirmed the estimates by Welch and Vestuater 5

bf the critical thickness of the condensate film to be between 0.4 and0.7 micron (observed r bc-:0.63 micron) and the critical nuclei diametercalculated from Fatica and Katz23 equation, 8.8 nicron, agreed well withthe calculated value of 8.7 micron by McCornick and Baer.3 8

Dolloff and Metzger39 recently com2leted a study of dropwise conden-sation of steam at high pressures. They used cupric oleate at aconcentration of 72 ppm as a promoter. They observed that the heattransfer coefficient had its maximum value at low AT and decreased to anessentially constant value depending an pressure a- high AT. Sizilar tothe observation of Brown and Thomas,3 6 they observed an increase in thecoefficient of heat transfer with an increase in pressure at whichcondensatiom took place.

Increased Heat Transfer Rate by Mechanical Wiping

Two reports were published in 1967 which shoved advantages ofperiodic mechanical removal of steam condensate from a condensing surface.The concept is analogous to the ordinary automobile windshield wiperwhere, with proper blade design, the liquid film may be almost completelyremoved. The wiping action of the blades provides a fresh condensingsurface periodically. Bauman40 conducted a study on a 18 inch diameterx 6 inch high cylinder which was wiped on both the inner and outersurfaces with automobile type wiper blades. He obtained overall coeffi-cients of heat transfer which increased with the wiper speed, and withthe coolant velocity. An increase in the overall coefficient of heattransfer of 60% was obtained at a wiper speed of 180 rpm compared to itsvalue without wipers. Adamson4l extended the work of Bauman to ship-board application of the principle. He has suggested the use of wipersin condensers operating at various load cycles. The wiper blades would

24

zhen be engaged &uring& ;eak or =ear-peak1 loa concitions only- --he

ra~c dezaiis several eesigms or uwiper blad-es for shipBoard :.pliztiz7--

Effect of Coeznser Surface

As early- as 194 Erew, !:Jgle a=Z, Sn-ic!h reporzed the effect olfscrface fl-nish on dr,w.isc condensation. -their experizents showed 6hat

v saweth * or p~lished surfaces in&.sced dropwzise conIdeasation, evenain the absence or oil or farty acic-s. On the other han-d, chey concludedthat fil= cnzensacion can occur on very rougSh or very foul surfaces

I---- ifc edwt[ ail or fatty acids- it therefore appears char inthe absence of zd iti zs or oronocers, ir is necessary to have z scontsurface to induce dropruise dnao.(

Ba=?on 2 5 noted that a =or--or n6iish increased the life of a brasssurface pronored swith oleic acid co 4 ries idi -finishes with numbers3-OM0 grades of ezery paper. Hie also noted that veity zle-an and 'szoorbserface actual!': caused firiecondensation at first rgficPI lAzer can"ecto dropurise cc idensacicn due to zzpurlrties in the stean.

The basic mnchanis-z of :? =otion and maintenance of dropiise conden-sarion lies in the fact that the condenser surface should be nontietcableto0 the cozdensing vapor. A-.y1 surface treatment, including applicationof dropwise preo-orers, whi-ckh u-ill1 cause the surface to become nonwetrablevill be effective in pronmaring drop~rise condensation. TherefOre, -itbecor-es apparent that the nature of the metal surfa:ce and aterial deter-mines to some ex-tent if a given pron-cter will be successful in prcnorin,dropwise condensation. F-!azso. 1 ha-s confirned this in his experimentsin L-Iich he showed that the use of different promorers on the same metalSurfice caused droprwise condensation for different durations in tine,He noted that: usin- a mixture of oleic acid and light lubricating oil afa pronoter on the surfaces of m-irror smooth chrone-plated copper andNo 6J emery paper treated stainless steel, the useful promoter l!fe for

chrom-e-plated copper was twice as long as for stainless steel. He alsofound that a prorzoter of some sort xas necessary for dropw-i~e condensatiznand that highly polished surface alzne will not zcause dropwise condensa-tion. His observatioai appears to be contrary to thoize of Drew, Nagleand Smith. 2 1 However, this difference may well be caused by degree ofsurface polish and cleanlinesr observed during tests.

M4isra and Bonilla4 observed in their experiments with condensingmercury vapor on different surface that it condensed in a dropwise marineron Carbon steel and stainless steel surfaces, and in a filmwise manneron nickel and copper surfaces. They noted with the stainless steelcondensing surface that even extreme care in cleaning the surface did notproduce filmwise condensation.

An elaborate study of surface characteristics to determine thenucleation sites for dropwise condensation has been reported by M!cCormnick

2 and Westwater.7 The nucleation of water drops during dropwise condensation

was scudied on a horizota'. surface of copper coated with a monolayer ofbenzyl nercaptan. .otograpbic evidence showed thvi drops nucleated -

at natural cavities on the condenser surface. Some cavities were nuclea-tici sites because they cortained trapped, liquid water. in addition,the authors found that the cavities produced by needles and by sparkerosion and scratches made on the surface also nucleated drops, Theseexperiments were extended to include the effect of snal solid particleson the condenser surface.

The authors studied the behavior of 37 possibie sites in the photo-graphic study. Out of this total of 37 sites, s r wete fourA to --ucleatLfrom 2 to 5 tines after which these sites became ractive. In a totalof 8 runs, they found six sites which did -or .ucleIte at all, and 4sites which nucleated in 7 runs. Drops also appeared from other locationson the surface which could not be idenLified a5 nucleation sites fromphotographs of the surface. They concluded that drups nucleate at

preferred sites during dropwise condensation.To study the effect of stall particles on the condenser surface,

particles of 24 different materials were tested for effectiveness asnucleation centers. The particle density in these tests varied between800 and 55,000 particles per square centimeter. They found that thenucleation characteristics of a particle was independent of the particledensity whereas nucleation ability among various kinds of particles variedsubstantially. The relative o-der of nucleating abilities of theseparticles was explained by the net heat of adsorption. A positive netheat of adsorption of the vapor on the solid was assumed to mean aspontaneous process. Nucleation of drops ,fas found to occur on particlesof 13 materials having positive net heats cf adsorption. For the insolu-ble, wettable particles, a size range of active nucleation sites ofabout 2t-- 30lwas observed.

McCormick and Westwater showed in a later paper8 that at a particularpressure and temperature difference, the heat flux increases with anincrease in the population of active sites on the surface. However, aathe population of active sites increases, the crowding effect reducesthe growth rate for each new drop added to the total. It was estimatedthat when an active site density of between 3 x 106 t 3 x 10 7 per sq inis reached, the coefficient of heat transfer in dropwise condensationreduces to an extent that it equals the value for f.Llmwise condensation,thereby eliminating the improvements normally expected in dropwisecondensation. In addition, the authors observed that the number ofactive dites depends on the surface preparations and its subcooling.This behavior of sites in condensation seems analogous to their behaviorin nucleate boiling.

26

COMORCIAL APPLICATIONS OF DROkP!ISE CO.1EMSATION

Full Scale Tests

Most of the published work in fildiise or dropwise condensation hasbeen li ited fo single rube or plate condensers under idealized labora-tory conditions. However, a few reports have been published whichdetail tests conducted on full scale commercial condensers.

Tests ising octadecylamine, C-8H3 7NH2 , as a d.opise condensation

prowoter or a full scale steam plant have been reported by Tanzola andWeidman.26 As discussed earlier, an increase of between 100 to 200percent itt the overall coefficient of heat transfer was obtained. Inthe drying machine of the paper mill where these tests were conducted,an average reduction of 10 percent in steam requirements was securedthrough the application of octadecylamine. The author$ also noted asharp reduction in corrosion of the condenser tubes.

Blackman, Dewar and Hampson2 7 conducted tests of three drop-promotingcompounds on a marine condenser. The three compounds, dodecanethiol,C12H25SH, tridodecyl phosphorothioite, (Cl2Hl5S)3PS, and decamethylenedi(ethylxanthate), C2H50-CS-S-(CH2)l0-S-SC-OC2H5 , were applied on thetop surface of about 30 tubes in the condenser. The tubes exhibiteddropwise condensation for over 2 years of operation.

Watson, Brunt and Birt4 0 postulated the requirements for a materialto be used as a drop-Dromoter in a full scale steam plant as follows:

- both the material and any carrier shoald be compatible with

the steam plant,

- it should be inexpensive and nontoxic,

- it should not require specially cleaned surfaces, and shouldmaintain a clean condensing surface during use,

- it should have a long useful life, and

- it should produce a substantial increase in the steam-side Iheat transfer coefficient in a reasonable time. T

Based upon these requirements, they rejected the usefulness ofsimple hydrocarbons, fatty acids, sulphur containing hydrocarbons whichare injected in solvents such as carbon tetrochloride, and filmingamines. They concluded that waxes derived from plants are the bestavailable dropwise condensation promoters, details of which have beengiven earlier.

A series of life tests were reported by Osment et a13 1 usingindustrial steam condensing in a power generating plant and in a marinecondenser, as discussed earlier. Brunt and Minken 2 and Birt et a13 4

27

o-

have reported a reduction of about 30Z in the capital cost of seawaterevaporators if dropwise condensation promoters were used whereas Butcheret a133 reported an increase of about 20% in the overall coefficient, ofheat transfer. In short, savings of up to 30Z could be obtained incommercial condensers with the proper application of the existing know-ledge of dropwise condensation promoters.

Effect of Corrosion Inhibitors

Corrosion films on metal surfaces in-condensation are usuallyformed by oxidation of a thin layer of the metal surface. These metaloxides generally Hiave low thermal conductivity thereby offering asignificant resirtance to heat transfer. The net result is that evenif condensing coefficients are increased by an order of magnitude, thenet increase in the overall coefficient of heat transfer is very small.if the metal surface could be protected from oxidation, a substantialportion of the increase in condensation coefficient due to drop forma-tion could be translated into an increase in the overall coefficientof heat transfer.

Some organic dropwise condensation promoters have been found to becorrosion inhibitors as well. Filming amines afford corrosion controlby the formatiod of a nonwettable film over the condenser surface.

$Since these amines do not function by neutralization of carbonic acid,their performances are independent of dissolved gas concentration.

The commonly employed corrosion inhibitors are straight chaiftprimary amines of 10-18 carbon atoms in their molecules, or their salts.Octadecylamines used by Tanzola and Weidman26 are about the most effee-tive comnercially available filming inhibitors. These amines have beenknown to also lift or loosen and displace existing corrosion depositson the condenser surface. The usual effective concentration is about1-3 ppm of the condensate. Since these amines are also dropwise conden-sation promoters, large increases in the overall coefficients of heattransfer have been obtained with their use. A third, and indirectbenefit of corrosion inhibitors is that they increase the condenser lifesignificantly.

Requirements for Practical Use of Dropwise Condensation

For proper application of dropwise condensation to commercialcondensers, it will be necessary to carry out life test on full scalecondensers under actual operating conditions. To be acceptable, thecondenser must be capable of producing dropwise condensation for thelife of the equipment and require only a minimum of special handlingor servicing. It must yield uniform improvement in heat transfer on acontinuous basis and must not cause corrosion by reason of dropwisecondensation promoters.

28

NOT RIEPRODUCIBLE

Ideally, arop-:ise condensation will be most successful if tnecondenser tube mater al could be inherently capable of producing ir -

wise conensation, or a permanent coating might be app1-ied at the -of manufacture to promote drop:a.ise condensation with a substantial

increase in hear transfer for the life of the equipment tithour b _:affected by steam impingement, fouling or dertrioration. Neither :

these approaches have been zuccessful so far.At this time, the most suitable method appears to be the interm.rten-

application of the dropuise promoter compound. The compound could beintroduced into the condenser either by spraying it onto the tubesurfaces periodically or by addinE it to the steam source as a fix-,e

predetermined proportion of the condensate.

An area of research which appears promising is the effect of vibra-tion of the condenser surface on coefficients of heat transfer in fuln-

wise and dropwise condensation. Although effects of vibration on

Kcoefficients of heat transfer in boiling have been invest-igated byomany authors, no such investigation appears to have been carried out incondensation. It.appears feasible that vibration of the heating surfaccat an appropriate frequency and amplitude may cause break-up of the

liquid film in filiwmise condensation to form drops, thereby increasingthe heat tran.fer rate. Furthermore, vibratian of the condenser surfac?

in aiopwise-condensation may accelerate drop formation, reduce the dropsize necessary for sliding dotm the condenser surface, and/or increasethe sliding velocity of the droplet.

j A saving in the capital cost of condensers of around 30%, observea

earlier, is substaitial and investigation musL be continued to realize

this saving in seawater evaporators, marine condensers, and other com-mercial applications. The investigation should also determine if cliecost of promoting dropwise condensation is offset by the savings:

obtained by reduction of the condenser size.

CONCLUSIONS

1. Heat transfer rates in dropwise condensation are significantlyhigher than those obtained in filmwise or mixed condensation.

2. Condensation on pure, chemically clean metal surfaces is tsna! 'filfnwise. Very smooth and highly polished surfaces can also cov-.e'

filmwise condensation. Compounds which make the condenser su~fi ctnonwettable are good dropwise condensation promoters when applied j!to the condenser surface.

nonwctale re ooddropiseconenstio I~rrno~rswhe aplie

3. Under laboratory conditions, some permanent-type coatings, e.g.,tteflon, -gold, silver, have been found to be effective dropwise

condensation promoters. However, the effective lives of some of.these promoters have been short, possibly due to surface fobling orremoval of the coating in service. Furthermore, the effectivenessof'permanent-type promoters in maintaining dropwise condensation islimited by their low thermal conductivity and the coating thickness.

4. Generally,'organic compounds which, when oriented at the metal surfacein a monolayer and presented an "unhindered"-hydrocarbon surface tothe steam, were successful in promoting dropwise condensation. Thesurface active or anchoring-group usually contains a divalent sulphuror selenium atom. A number of such organic compounds have beenfound which, when applied to the tube surface or.'fed periodicallyinto the steam stream, are effective dropwise condensation promotersfor long periods of tim.

5. Only those substancec which are strongiyadsorbed on the condensingsurface are significant drcpwise condensation promoters. Torexample, mineral oils, alcohols, organic nitrogen compouias, endhalides are ineffective in prbmoting dropwise condeAsation whereashigh molecular weight fatty acids and natural waxes derived fromplants are good promoters of dropwise condensation.

6. Heat transfer coefficient can be increased by mechanically wipingthe condensing surfaces at frequent intervals, the improvement beingproportional to-the wiping frequency.

7. Corrosion inhibitors which also form a nonwettable layer on thecondenser surface are most suitable for achieving dropwise condensa-tion and high overall coefficients of heat transfer.

8. The published information on the durability and behavior of dropwise

condensation promoters is incomplete and inconclusive.

9'. The applications of condensation promoters are enormous. For example,experiments with commercial seawater evaporators have shown thatimprovements in the overall coefficient of heat transfer by usingdropwise condensation 'promoters can reduce the capital cost of suchequipments by 30% or more. However, due to a lack of needed informa-tion on dropwise condensation promoters, they are usually not usedin industrial practices.

10. Further research is necessary to determine durability and effective-ness of dropwise promoters under a varietyof industrial conditions.

30

iti-- - - - -- - - - -- - - - - - - - - - - -_ j

*7

11. It is possible that vibration of the condenser suiface may increaseheat transfer rates by/accelerating drop formation, reducing dropsize necessary for ,sl-.:ding down the condenser surface, and/orincreasing the sliding velocity of the dropiet iin aropwise condensa-ticn. Vibration of the heating suiface in filmwise condensation maycause break up of the liquid film to form drops and'increase heattransfer rates.

3'I

RECOMMENDATIONS

1. Current research nd devdiopment programs to increase the usefullife of permanent type coatings and organic promoters should and nodoubt will be continued. No major breakthrough in the way of newcoatings or promoters is in sight. Consequ-f6rly,, the area of greatestpotential advancement is in the use of vibratidns or wiping techniques.It is, therefore, recommended that an eixerimental program beinitiated to evaluate the effects of vibrating ondenser surfaces.

31

I

REFERENCES

1. Hmpson, H., "Dopwise Condensation on a Metal Surface: A Studyof its Durability," Engineering, Vol. 179, No. 4655, April 15, 1955,pp. 464-469.

2. Jakob, M., "Heat Transfer in Evaporation and Condensation - II,"Mechanical Engineering, Vol. 58, 1936, pp. 729-739.

3. Emonsi H., "The Mechanism of Dropwise Condensation," AmericanInstitute of Chemical Engineers, Trans., Vol. 35, 1939, pp, 109-125.

4. Eucken, A.,,"Energie-und Stoffaustausch an Grenzflachen," DieNaturioissenschaften, Vol. 25, 1937, pp. 209-220.

5. Welch, J. F. and Westwater, J. W., "Microscopic Study of Dropwise

Condensation," Proceedings of the Second International Heat Transfer

Conference, Vol. II, 1961, pp. 302-309.

6. Umur, A. and-Griffith, P., "Mechanism of Dropwise Condensation,"Trans. of ASME, Journal of Heat Transfer, Series C, No. 2, 1965,pp. 275-282.

7. McCormick, J. L. and Westwater, J. W., "Nucleation Sites for Drop-wise Condensation," Chemical Engineering Science, Vol. 20, No. 12,December 1965, pp. 1021-1036.

8. McCormick, J. L. and Westwater, J. W., "Drop Dynamics and HeatTransfer During Dropwise Cond nsation of Water Vapor on a HorizontalSurface,"Clhemical Engineering Progress, Symposium Series, Vol. 26,No. 64, 1966, pp. 120-134.

9. Peterson, A. C. and Westwater, J. W., "Dropwise Condensation ofEthylene Glycol," Chemical Engineering Progress, Symposium Series,No. 64, Vol. 62, 1966, pp. 135-142.

10. LeFevre, E. J. and Rose, J. W., "A Theory of Heat Transfer byDropwise Condensation," Proceedsing of the Third International HeatTransfer Conference, Vol. II, 1966,, pp. 362-375.

i. Cope, E. F., "Investigation of the Use of Non-Wetting Agents forPromoting Dropwise Condensation on Steam Condenser Tubes," WestinghouseElectric Corporation, April 1959, from reference 12.

32

12. Fox, R. M., "A Review of Literature on the Promotion of DropiseCondensation," U. S. Navy Marine Engineering ,Laboratory, Research &Development Report 71-106, 7 July 1964, AD 442895.

13. Topper, L. and Baer, -E.., "Dropwise'Condensation of Vapor and HeatTransfer Rates," Jouiinal of Colloid Science, Vol.-10, 1955, pp. 225-226.

14. Swortzel, F. R., "Film Coefficients for Tetrafluoroetylene PromotedDropwise Condensation," M. S. Thesis, Department of Mechanical Engineering,Raleigh, North Carolina, 1960.

15. Depew, C. A. and Reisbig, R L., "Vapor Condensation on a HorizontalTube Using Teflon to Promote Dropwise Condensation," Industrial andEngineering Chemistry, Process Design & Development, Vol. 3, No. 4,October 1964, pp. 365-36S I

16. U. S. Naval Civil Engineering-Laboratory. Technic4rl Report R-255:Heat Transfer Studies of Dropwise Condensation and Thin Film Evaporation,by,.C. Saturhino. Port Hueneme, California, June 1963.

l i4Erb, R,. A. and Thelen, E., "Gold Plated Condensers," -industrial WaterEngineering, August 1965.

18. Erb, R. A. and Thelen, E., "Promoting Permanent Dropwise Condensa- ition," Industrial and Engineering Chemistry, Vol. 57, No. 10, Oct 1965,pp. 49-52.

19. Erb, R. A., "Wettabilt-ry of Metals under Continuous CondensingCoiditions," Journal of Physical Chemistry, Vol. 69, No. 4, April 1965,

pp. 1306-1309.

20. U. S. Department of the Interior. Saline Water Conversion Reportfor 1967, Office of Saline-Water, 1967, pp. 254-256.

21. Drew, T. B., Nagle, W. M. and Smith, W. Q., "The Conditions forDropwise Condensation of Steam," Trans. of American Institute ofChemical Engineers, Vol. 31, No. 4, 1935, pp. 605-621.

22. Nagle, W. 1., Bays, G. S., Blenderman, L. M. and Drew, T.- B., "HeatTransfer Coefficients During Dropwise Condensation of Steam," Trans. ofAmerican Institute of Chemical Engineers, Vol. 31, No. 4, 1935, pp.593-603.

23. Fatica, N. and Katz, D. L., "Dropwise 'Condensation," ChemicalEngineering Progress, Vol. 45, No. 11, Nov 1949, pp. 661-674.

33

U

i

24. Hampaon, H., '"eat Transfer During Codensation of Stem,"Engineering, Vol. 172, No. 4464, August 17, 1951, pp. 221-223.

25. Hampson, H., "The Condensation of Steam on a Metal Surface," Proc.of the General Discussion on Heat Transfer, Sep 11-13, 1951, London,Institute of Mechanical Engineers, 1951, pp. 58-61.

26. Tanzola, W. A. and eidm* J. G., "Film Forming Corrosion Inhibitorsalso Aid Heat Transfer," Paper Industry, Vol. 36, No. 1, April 1954,pp. 48-50.

27. Blackman, C. F., Dewar, M. J. S. and Haupson, H., "An Investigationof Compounds Promoting the Dropvise Condensation of Steam" Journal ofApplied Chemistry, Vol. 7, April 1957, pp. 160-172.

28. Furman, T. and Hapsma, H., "Experimental Investigation into theEffects of Cross Flow with Condensation of Steam and Steam-Gas Hixtureson a Vertical Tube," Proc. of the Institute of Mechanical Engineers,Vol. 173, No. 5, 1959, pp. 147-169.

-29. Bobco, R. P. and Gosman, A. L., "Promotion of Dropwise Condensationof Several Pure Organic Vapors," ASHE Paper No. 57-S-2, 1957.

30. Watson, R. G. H., Brunt, J. J. and Birt, D. C. P., "DropwiseCondensation of Steam," International Developmer.!-s in Heat Transfer,Part 2, International Heat Transfer Conference, 1961, ASME, pp. 296-301-.

31. Osmiht, B., Tudor, D., Spiers, R. M. and Rugman, W., "Promoters forthe Dropwise Condensat~.on of Steam," Trans. of AIChE, Vol. 40, 1962,pp. 152-160.

32. Brunt, J. J. and Minken, J. W., "The Application of Dropwise Promotersto Seawater Evaporators," The Industrial Chemist:. Vol. 34, No. 399,May 1958, pp. 219-224.

33. Butcher, D. W.i Birt, D. C. P. and Honour, C. W., "Condenser TrialsUsing Dropwise Promoters," Admiralty Materials Laboratory, Poole, England,A.M.L. Report No. A/86(W), Nov 1964.

34. Birt, D. C. P., Brunt, J. J., Shelton, J. T., and Watson, R. G. H.,"Methods of Improving Heat Transfer from Condensing Steam and theirApplication to Condensers and Evaporators," Institute of ChemicalEngineers, Trans., Vol. 37, No. 5, Oct 1959, pp. 289-296.

34

35. Bliss, A. J., and Harding, B. M., "'Torced Circulatioa and DrcpwiseCondensation Techniques for Improving Heat Transfer Ratez for VaporCompression Evaporators," Office of Saline Water, U. S. Department ofthe Interior, R&E Report No. 8.

36. Br-own, A. "R. and Thoras, X. A-, "Filbrise and .Dropvise Condensationof Stea= at Low- Pressures," Proc. of the Third International HeatTransfer Conference, Chicago, Aug 1966, Vol. 1I, pp. 300-305.

37. Sugawara , S. and Katsuta, K., "Fundamental Study of DropwiseCondensation;" Proc. of the Third International Heat Transfer Conference,Chicago, Aug 1966, Vol. II, pp. 354-361.

38. McCormick. iT. L. and Baer, E., "On the Mechanism of Heat Transfer inDropwise Condensation," Journal of Colloid Science, Vol. 18, 1963,pp. 208-216.

39. Dolloff, J. B. and Metzger, N. H., "Dropwise Condensation of Steanat Elevated Pressures," Chemical Engineering Department,-Oniversity ofMassachusetts, ONR Technical Report No. 2, Contract Nonr 3357(02),May 15, 1968.

40. Bauman, J. A., "Wiped Film Condensation," Marine Engineering

Laboratory, Research and Development Report No. 505/66, March 1967.

41. Adamson, W. L., "An Investigation of the Wiped Film Condensationfor Shipboard Application," Marine Engineering Laboratory, Annapolis,Maryland, NSRDC Report No. 2471, Dec 1967. (AD 825359)

42. Misra, B. and Bonilla, C. F., "Heat Transfer in the C3ndensationof Metal Vapors: Mercury and Sodium up to Atmospheric Pressure,"

Chemical Engineering Progress, Symposium Series, Vol. 52, No. 18,1956, pp. 7-21. 4

: 35

I

AS Condenser area free of discrete drops ft 2

AT Total aiea of the ccndenser surface ft 2

h Film coefficient of heat transfer Btu/hr ft 2 OF

Coefficient of heat transfer between condensing Btulhr ft 2 OFvapor and the condenser surface

h2 Equivalent coefficient of heat transfer across Btu/hr ft 2 OFthe thin corrosion layer

Coefficient of heat transfer between cooling Btu/hr ft2 OFwater and--the- c -iden-se-r wall

k Thermal conductivity of condensate Btu/hr ft OF

K Thermal conductivity of metal will Btu/hr ft OF

Q Rate of heat flow Btu/hr ft 2

t Wall thickness ft

tc Critical time for film growth hr

-AT Vap: saturation temperature minus metal OFsurface temperature

U: Overall coefficient of heat transfer Btu/hr ft2 Of

xc Critical film thickness at fracture ft

6 Average film thickness ft

Angle of contact OF

X Enthalpy difference between, the vapor Btu/lban' the condensed liquid

P Density of liquid lb/ft3

a Interfacial tension lb/ft

J 36

u1 v Liquid to vapor interfacial tension lb/fr

O'sv Solid to vapor interfacial tension lb/ft

Solid to liquid interfacial tension lb/ft

k Adhesion tension lb/ft

LH

A,

1

L

37

I __

'1" m u --i i i i m u ..m - " '- i i •. . -m- - ' m . . " i -l ' ..• " -..

II

- -- -- -

2000

Condenser .surfac.,: 0.18" thick copper plate

Curve Coating Thickness (in) -

O 1800 - DC-1107 0.00025_ 2 Teflon 0.00040

3 R-671 0.000254Z Uncoated --

-'1600

M, 1400

411

~~ 1000 [

0

600 I

45 10 15

Water velocity, fps

Figure 2. Performance of' various coatingsin condensation.

39

2200,

0~ 20G0Contamninated tube

£4.1

IWO

S1800

V4

1600

U

oU 1000

1200

04

p

I i-J

ii)I

44

--- h2 ,(corrosion)

Cooling Water

Figure 4. Thermal resistances in atypical condenser system.

41

II - - --.-----,-------- ~ - - - - - - -..... ---- ---- - -- -• --- --- •

tub

1500

AJ

0 150 300 450 600 750 900 1050 1200

Time, hr

yFigure 5. Overall heat transfer coefficientafter injection Of Si(SC 12H25)4in a test condenser.

24

UNCLASSIFIEDSecuzity Clasification

(Sa~tieaeilcaa, 1 f i. e* DOCUMENT CONTROL DATA - R&D s pm e .Mfd

I - ORI5IMATIN 0 ACTIVITY (Co It m author) -2a. REPORT SIECURITY C LASSIFICATION

NF.Val Civil Engin iring Laboratory - UNCLASSIFIED~Por t Hueneme, California 93041 lb. GROUP

2. REPORT TITLE

The Effect of Coatings and Surfaces on Dropwise Condensation

4. DESCRIPTIVE NOTES (Type of report and Mnclualve dae.)

S. AUTHO(S) (Lost nm ine te nae, inttti)

S. C. Garg

G . RE9PORT DATE -7a. TOTAL NO. OF PAGERS 7b.,No. or RIPs- Juy 1969 46 42

i.'CONTRACT OR GRANT NO. Si. ORIGINATORS$ REPORT NUMNER(S)

&PROJECT No. ZIP 38.512.001.004 TN-1041

C. 96, &IN JPORT NO(S) (Anyeother numAItmyb.to 4110

d.

10. A V A IL ANIPITY/LINITATION NOTICES

This document has been approved for public release anld sale; itsdistribution is unlimited.

III SUPPLEMENTARY NOTES 12SOSRN IIAYACTIVITY

Washington, D. C. -20360 t12. S5T AC literature survey of the published information on the effect of

coatings' and surfaces on dropwise cdndensatibn has .eanicarried out,. Thesurvey has-shoWrn\ h'at research to date in promoting dropwise condensationhas been limited mostly to small scale laboratory experiments. There appears-to be a serioui lack of data on methods of application of various dropuise-promoters,- useful life of promoters, effective concentration, and frequencyof proniioter renewal. This lack of information is highlighted by thor totalabsence of dropwise condensation as a factor in design of condensers inindustry.

Recomendations are made to initiate an experiments l;.program toevaluate the effects of vibrating the condenser surface upon d'a-opyisecondensation and the coefficients of heat transfer.( '~~

D D I JA .1473 011z .07480o UNCLASSIFIED

Secuuity Classification

UNCLASSIFIEDLIKALNSINCSecucity Classi-fication _____ ____ ____

KC 035ROLE WYT OL WT MOLZ WY

Coatings

Condensers

IN4STRUCTIONIS

1. O3DGINATING i.CrIVIY: Enter the same sod addr-ss imposed by security classification. using staadard statementsof the contractor, subcomiractor. grantee. Department of I)e- such as:'(ams, actlvity~o other organization (colpogate author) isffi1*g (1) "Qualified requesters may ob~tain copies of thist~,eor. report from DDC."2z. -3S~T SECV3TY CLASSZFIA-fJlN Enter the over -(2) "Foreign announcement and dissemination of thisall securiy claehficatie of ther.epoet. Indce whetherreotb DIsntahrid,"Resticted Don"' Is Locluded MWring is to be inacor-reotbDCisntahrid.amm wiek mpg - lat& socurity regulatloatl (3) "U1. S. Governmtent agencies may obtain copies of

21. ~W~Antugi doaagrdia is~lfid I NDl ~ ,this report directly from DDC. Other qualified DDC2r. tive .1 Autmacdo orcesd at Is sjeiidi i users shall request throughthe grow Sv~er. Ala., when applicabl*;, show that o~ptionalamkiege how& been used for Group 3 sed Group 4 as antheir- (4) 11U. S. military agencies may obtain copies of thisised. report directly from DDC. Other qualified users3. XWORT TITLE Poor the comlete report title In a&H shall request thaoug'scapital letters. Tidles in all cases should be unclassified.TO 4 mosaaul t~itaocumet be silected withott elaosaifics-tioa, shew te classification in all capitals in parenthesis (5) "All distribution of this report is controlled. Qual-Immoditey following the title. ified DDC users shall request through

4. DXBWMMVE NOTE& If appropriate, enter the type of ________________

roport. eg.. interim. progress. summary, annual. of final. If the report has been furnished to the Office of TechnicalGive the Inclusive dates when a specific reporting period is Services, Department of Commerce, for sale to the publIic. indi.cSord. cats this fact and enter the price. if known.

S. AUTIIIRS): Enter the na.,(s) of author(s) as shown on IL SUPPLEMENTARY NOTE& Use for additional *xplana-or is the repett. E06a lost name. fir-t amer mdda Initial. tory notes.It military, show raubt aad beanch of service; The n.,m. ofthe principelaethor is'an absolute minim in requirement. 17- SPONSORING MILITARY ACTIVITY. Enter the name of

the departmental project office or laborsary sponsoring (per6 . REPORT DATE: Enter the date of the report so day. ing for) the research and developmaent. Include addiess.month. year or month. year. If aore than one date appearsn thre report, use dat. of publkcation. 13. ABSTiRACT: Exater an abstract giving a brief and factual

summary of the document indicative of the report, even though7,.L TOT.L. NUMBER OF PAGES. The total page count it may also appear elsewhere in the body of the technical re-

*shotuld follow normal patiiation procensies. L~e., enter the port. If additicmal space is required, a contintiation sheet shellrnmber of pages containing information, he attached.

*7b. NUMBER OF REFERENICE& Enter the total number of I! is highly desirable that the abstract of classified reportsrreferences cit;-J In the report. be unclassified. Each paragraph of the abstract shall end withlla. CONTRC.T OR GRANT NUMBER. If appropriate, enter an indication of the military security classification of the in-the applicable number of the cortract or grant under which formation in the paragraph, represented as (Ts). (3). (C). ot (Ut.the report Was written. There is no limitation on the length nt the abstract. Ho0w-.8b, Ac. & Sd PROJECT NUMBER: Enter the appropriate ever, the suegested length is from 150 to 225 words.

suilietar dunuertiecto. sysch nmestas nuoec ber. c 14. KEY WORDS: Key words are technically meaningful termsatabrojct umbr. ystm cober, tsk umbr. tc.or short phrases that characterize a report and may be used as

9-- OFIGINATOR'S REPORT NUMDER(S): Enter the ofri. index entries for cataloging the report. Key words must becial report number by wisicI, the document will be idetified selected so that no security classification is required. Identi-

ad controlled by the originating activity. This number must tiers, such as equipment model designation, trade name, militarybe unique to this report. project code name, geographic location. may be used as key51 Pb. OTHER REPOkT NUMBER(S): If the repcthsbe words but will be followed by an Indication of technical con-

-assigned ay other report num-bers (either by 0he oripinao test. Thie assignment of links. rle%. and weights is optional.Ior by the speit), alao enter this nuamber(s).

10. AVAUABILITYILIMITAr. ON NOTICES.' Enter any lim-.itations, ce further dlspsmination of tkhe report, cther than those

Secrity Classification