1'1'1'11111 (“111W """?"""" 1111 17?, ""T""""1,1""""" {111.1111 "111 “W ?"""?1"""1""1" ?' 111' """ 1'1" 111??" ?""i""1j.-""1"'? in "1 ""1" M1 ? i’ """i i1""?""""""d%i""i"" "F 1&1“? g? . 1111‘} :12?" 1111‘??? "i? jii'" 1111 "’1 """li"?""??""1"]"""""?iii"i3" i111 "I"?""""""" - ' 1 ..?""'1"~"11”" 1i?" "? 1""1’1'I""" "'""?'????1"""" " ' .‘ -- '. s" ; '{1 11u1 j1—11111‘1‘? 11' "?"V"” 1“, 11.!“ 1 ?I. 1 n ‘ . 1 1 1 . 1-11.11.111111 1" ,3: 11111-11111111 i" .. 1' = 1...11111111111111 ‘ 11 ? 1;? 1 1 "”ng "i M" ""131? "'"1?1"?""' ""1"?" " "" ?"“?"1 ‘" i511"? -' ? "1???" 1 «11111 1 ' L :1» iii? ‘1."r‘:'".~1 "1"!"1i‘111 if???" H’1113‘ his-'11 11 "" if !1 i1 ?i?17"' 1.1131 11??” ”1‘1 111% ii """"'?-; 114' . 11??"‘1111" "i1? 1 If; 11:13?" "ii *1? jg" I?" “13‘"; f" """"" 4‘1?" 11 i ii?" 1 .1 1.1? ?i 1311 1? 1h" 11""? """"?""":"'""""" I'M , i1 '1 1111111: "5&4? at ‘11:???" “"11 " ? .11 * " 1."?1 ""1?" '1? 1” ;w.3:s.: 3 mi: 15% ”i" 1‘1.':‘1? ? f "R1?” J11? ”’11” 1'1"?" 1?1?'l1""1 11‘ 1H1}? MM??? ”Mimi? i"""" {511i}? ”i Pfljaiili' "ii”??? fji‘ 3 11'3" I?" i111 ”i? '131iij “11111“? 1" "ii???" iii"? ?:ii 1“ "f "in???” W 11"!"""" """"""" ""1"" "1?” I?" 1"?! "ii"? ""11""iiii"? '1; "l": 1 .1 1 1" 1! .‘I? '1"°""';1." . ~. .1. "‘1‘- 1 1 1 -1~1111.11..111-1~111. '" 414'11'1'11" 1113111111111. "i1111111111111'1?:: "1 't> ~. - 1 u ,‘ ..‘ 1 I" . .. ., 111; j 1 ‘1 1251111111} 11‘ ’3 V "i 1"“? '1 I"? .1‘1 " r F. 1. .1131, 1:15: I _ . ;‘"’"?"?""."",""".‘?"""' "in. ? -!,1pi-'1'-1 151111111117! """""1"?1?111.111 1 11? ’t w ‘1. ‘ ya. to. ma." \ 1‘ v- k... _. v , “1......“J -—..._. I: v n m.— l O o. - . ‘ _ _ ~ _ a I rr11 111? i111" "i?!" ""~ "" “an: 111111" .4 .‘ s4.- . a. w... . -ho: ' O - v. - -1 . . ’ ' tr. ‘ V u. MCI—v ‘ § . , M m -m;~ Van; “m: ‘ Ct ...__ . . — .. ‘3 __ A .. ._ . . ,. . ? . . 1: “ 3‘ . 1"?" '" 1"? 11"???"111‘11 '.';"?"11‘""" 1" 1. 'iii" 1" 3331 1' ' 11‘ ""i?"i""??""11‘1 11‘?“ r.‘ .111 .. 1"?" 1" "I " I ? "?""?"??1??"""?1"1""1"1" i??? ""' «11"?1: " 1"“? 11 ."!1' .- 1-11. '1" 1'11 - ?1‘ii1??i??1?1iii111i. ' - i1“ ( i131, ", 1' in" "i? _..""'11‘ 1,???1?" """"""??' I A" "1;," ?_| " f. "i - 11."?11;??ii1jjjii1?‘1'1'1??"~’ ' "1:12???" . ‘Lz... W‘— 345515 Date 0-7639 LIBRARY L333”: Stan University This is to certify that the thesis entitled ENHANCED NUCLEATE 3mm N6 presented by JAVED ARSHAD has been accepted towards fulfillment of the requirements for Ms. A --- . M €( mm «(A L- m w m; ( degree in Major professor MS U is an Affirmative Action/Equal Opportunity Institution MSU LIBRARIES m RETURNING MATERIALS: Place in book drop to remove this checkout from your record. FINES will be charged if book is returned after the date stamped below. ENHANCED NUCLEATE BOILING BV (a Javed Arshad A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE DEPARTMENT OF l‘v’lECHANI CAL EI‘JGINEERING 1982 ABSTRACT ENHANCED NUCLEATE BOILING Bv V Javed Arshad The heat transfer mechanism responsible for the order of magnitude increase in the boiling heat transfer coefficient for enhanced boiling surfaces compared to plain smooth surfaces has been investigated experi- mentally. A video tape recorder and a motor-driven still camera were used to observe the evaporation process in grooves of three cross—sectional geometries with a cover plate over the grooves containing microdrilled holes. Inception of boiling, formation of a thin liquid evaporation layer and dry- out were observed in the groove. It was concluded that thin film.eva- poration is the principal heat transfer mechanism in nucleate boiling on enhanced surfaces as suggested by several previous investigators. Dry- out of the groove explained the test surface's reversion to the smooth surface pool boiling curve at intermediate heat fluxes. TABLE OF CONTENTS LIST OF TABLES ............................................... LIST OF FIGURES .............................................. LIST OF SYMBOLS .............................................. 1.0 INTRODUCTION .................................... 2 o 0 STATE’OF-IIY'IE-ARI‘ mm o o o o 00000000000000 o 00000000000000 2.1 Preliminarywork. ................ 2.2 Nucleation criteria and cavity geometry . . . . . . . . . . . . 2.3 Mechanism of heat transfer in nucleate boiling . .. . . 2. 14 Commercial enhanced surfaces . . . ....... . . . . . . . ...... 2.5 Analytical work .......... . ......................... 3. 0 EXPERIMENTAL DESIGN AND PROCEDURES ...................... 3.1 Pool boiling ....................................... 3.2 Visualization experimental design and procedures . . . “.0 RESULTS AND DISCUSSION .................................. 14.1 Pool boiling curves .. . ............................ A . 2 Visualization of the evaporation phenomenon inside the grooves ....................................... ll. 3 Thin film evaporation .............................. A. A Comments ........................................... 5. 0 CONCLUSION .............................................. APPE’DICES ................................................... LIST OF REFERENCES iv xi \OUOUU U.) 13 27 27 36 38 38 A6 55 60 7O 71 LIST OF TABLES Structural dimensions of the tested surfaces [17] ...... Structural dimensions of the test surfaces ............. Data for the first pool boiling curve for smooth SWface .000 ...... 0 ..... 00.00.000.000 00000 00 000000000000 Data for the second pool boiling curve for smooth surface ... ................................... . ......... Data for the first pool boiling curve for enhanced Surface No. l ......... ...... ........................... Data for the second pool boiling curve for enhanced Surface No. l ........... ... ............................ Data for the first pool boiling curve for enhanced Surface No. 2 ...... . .................... ............... Data for the second pool boiling curve for enhanced Surface No. 2 ............................... . .......... iv 19 30 71 72 73 7A 75 76 10. ll. l2. 13. LIST OF FIGURES Effect of emery paper roughening for acetone bOileIg on copper [3] 00000.00..000...00000000.. ......... (a) States of the liquid-vapor interface in a reentrant cavity. (b) Reciprocal (l/r) vs. vapor volume for liquid having a 90° contact angle [6] ........ (a) Effect of several surface treatments on sodium boiling at 65 mm Hg [7]. (b) Cross section of doubly reentrant cavities tested ................ ..... (a) Enhancement for water boiling at 101 kPa (1 atm) on a stainless steel surface having minute nonwetted spots (30-60 spots/cm2, 0.25 rrm diameter or less). (b) Enlarged photo of Teflonespoted smooth surface [8] .. A typical photograph of the oscilloscope face showing the surface temperature behavior [13] .......... ......... Sketches of two ways the microlayer might vaporize ...... (a) Cross section of COpper sintered porous boiling surface. (b) Enhancement provided by porous HIGH- FLUX surface for three fluids boiling at 101 kPa (1 am) [1“] oooooooooooooo 0000000 000000. 0000000000 00000. Some patented boiling surfaces applied to circular tubes [1] ......... ......... .. .............. ...... ....... Comparative single—tube pool boiling test results for p-xylene at 101 kPa (latm). From [16] .................. Geometry of the surface structure [17] .................. Boiling curves of water [17] ............................ Contribution of latent heat transport to total heat flux [17] ............................................... Apparatus for observation experiment [17] ............... l2 12 1M 16-17 l8 19 21 21 22 14. 15. 16. l7. l8. 19. 20. 21. 22. 23. 2A. 25. 26. 27. 28. 29. 30. 31. 32. 33. 3A. 35. Comparison between the measured heat flux and the predicted heat flux for water and nitrogen [17] ......... Measured boiling data of 1A fluids on HIGH-FLUX sintered surface compared with values predicted by Eq. (2—12) [14] ....... ..... ............ ................. Magnified view of the surface [17] ...................... Test surface; (a) Side View, (b) Top View (0) and (d) Groove cross-sections ....................... U-tube pool boiling facility ............................ Power supply circuit .................................... Visualization experimental setup ........................ First pool boiling curve for the smooth surface ......... Second pool boiling curve for smooth surface ............ First pool boiling curve for Surface No. l .............. Second pool boiling curve for Surface No. l ............. First pool boiling curve for Surface No. 2 .............. Second pool boiling curve for Surface No. 2 ............. Pool boiling curves ..................................... Schematic sequence of activation of a vapor pocket and film dryout at medium heat flux ..................... Schematic of shape of film in three geometries .......... (Tl) Evaporation starts from right corner ............... (Tl) Vapor front advancing in the groove ................ (Tl) vapor starts coming out of the groove, liquid inside forms of a film on the groove walls .............. (Tl) At low heat flux liquid film is apparent ........... (Tl) At intermediate heat flux lower half of the groove seems to have dried-out ................................. (Tl) Flooded groove ..................................... 22 26 28 29—30 32 33 37 39 A0 A1 A2 143 an A5 A8 149 51 51 52 52 53 53 36. 37. 38. 39. 40. Al. A2. “3. AA. 45. A6. A7. 48. A9. 50. 51. 52. 53. SA. 55. 56. 57. 58. 59. (T1) Liquid-vapor front in the groove as the heating is stOpped and groove is getting flooded ..... ...... ..... (Tl) Groove flooding .................................... (T2) Flooded groove .. ................................... Evaporation starts in the groove ........................ Vapor starting to ccme out through the pore ............. Liquid forms a film on groove walls ..................... Nucleation at lower heat flux ........................... Nucleating groove at intermediate heat flux ...... .. ..... (R1) Flooded groove ..... ................................ (R1) Evaporation starting from lower right corner ....... (R1) vapor front advancing in the groove ................ (R1) Liquid being pushed out of the groove .............. (R1) Nucleating groove at lower heat flux ............... (R1) Nucleating groove at intermediate heat flux ........ (R1) Nucleating groove at intermediate heat flux ........ (R1) Nucleating groove (higher heat flux), liquid film is less apparent ........... . ..... . ................. (Cl) Flooded groove ..................................... (Cl) Nucleating groove (a uniform thin film is seen along the periphery). Low heat flux is applied ......... (Cl) Nucleating groove (low heat flux) .................. (Cl) Groove flooding as the liquid-vapor front moves in when heating is stopped .............................. (C2) Flooded groove ..................................... (C2) Nucleating groove .................................. (C2) Nucleating groove (intermediate heat flux) ......... (C2) Nucleating groove (low heat flux). Droplets in groove and bubble are fbrmed by condensation on the glass plates .................................................. vii 5A 5A 56 56 57 57 58 58 59 6O 6O 61 61 62 62 63 63 6U 6U 65 65 66 66 60. (C2) Groove flooding ....... . ........... . ................ 67 61. (C2) Evaporation starting at different spots along the periphery ........................................... 67 viii List of Symbols Nomenclature A area [m2] db bubble departure diameter [m] h heat transfer coefficient [W/(mgK)] hfg heat of vaporization [J/kg] k thermal conductivity [kW/m—°C] n1 slope of vapor pressure curve NA number of active boiling sites P pressure [N/m2] q heat flux [W/m2] qex heat flux on the outer surface [W/m2] r cavity mouth radius [m] T temperature [C°] Vfg volume of vaporization [m3/Kg] 0 surface tension [dyne/m] Subscripts L latent 1 liquid max maximum opt optimum sat saturation V vapor N wall 1.0 INTRODUCTION Heat transfer enhancement has received much attention recently, in part due to rising energy costs. The heat transfer enhancement tech- niques are employed to economize on equipment cost and size, handling large and concentrated heat fluxes and improving the performance of the heat exchange equipment. Special enhanced boiling surfaces, now largely in use for building heat exchange equipment, incorporate a special sur- face microgeometry which promotes high performance nucleate boiling with heat transfer coefficients an order of magnitude greater than those of a conventional smooth surface. On a typical pool boiling curve when entering the boiling region from natural convection there is a large increase in the heat transfer coefficient. Two mechanisms are basically responsible for this increase: latent heat transport in the form of bubbles leaving the nucleating sur- face and agitation caused by growing and departing bubbles enhancing nat- ural convection over the inactive area. Enhanced boiling surfaces are a 5 to 10 fold improvement (in terms of heat transfer coefficients) over conventional smooth surfaces for mode- rate heat fluxes up to about 200 kN/mE. A number of different techniques have been employed to prepare these special surfaces. In the early inves— tigations surfaces roughened with different grades of emery paper were tested. These surfaces showed a temporary improvement before reverting to smooth surface behavior. The later investigations concentrated more on artificially formed nucleation sites whose performance did not deteriorate with time. These surfaces with a high pOpulation density of stable nuclea— tion sites are commercially available under different trademarks. A high performance of these special surfaces has been thought to be due a) to a lower superheat required for initiation of boiling because of large radii of the vapor nuclei trapped in the cavities of special geometry b) to a high density of such boiling sites and c) to thin film.evaporation within the structured porous matrix. Although a number of researchers have manufactured and tested a variety of enhanced surfaces, only a very few have done a detailed study of the heat transfer mechanism responsible for the order of magnitude in- crease in the boiling heat transfer coefficient for enhanced surfaces. Such a study is vital before we can model the boiling process to look for an analytical expression for predicting the heat transfer coefficient. In the present study two special surfaces were designed and manu- factured. These surfaces had grooves machined into them (one surface with triangular cross-sectional grooves and the other with rectangular) and a cover plate containing microdrilled holes placed over the grooved surface. Nucleate pool boiling curves were obtained to ascertain the performance of these two surfaces compared to a plain smooth surface. In a separate set up for a visual study, a videotape recorder and a motor-driven still camera were used to observe the evaporation process in grooves of three different cross—sectional geometries (triangular, rectangular and circular). Chapter 2 deals with the state-of-the-art review of enhanced boil— ing. Chapter 3 describes the experimental design and procedures for the present study followed by a discussion of the results in Chapter A. Chapter 5 presents the conclusions of the study. 2.0 STATE—OF-THE—ART REVIEW 2.1 Preliminary work Previous investigators have tested heat transfer surfaces incor— porating a wide variety of enhancement techniques. An excellent and recent review by Webb [1] describes the historical development and test- ing of these surfaces for a wide range of liquids. Early investigators [2,3] tested several liquids on surfaces roughened with different grades of emery paper. However, only a temporary improvement was shown before the surface behavior reverted back to that of a conven- tional smooth surface. Khurihari and Myer's [3] results (see Figure 1) showed that the heat transfer coefficient was proportional to the surface roughness. They noted that boiling site density (the number of boiling sites per unit area) increased with increasing roughness. The later in- vestigators concentrated more on artificially formed nucleation sites whose performance would not deteriorate with time. The artificially formed sites also allowed incipience to occur at a lower degree of wall superheat, AT, which is the temperature difference between the heated wall and the bulk liquid temperature. 2.2 Nucleation criteria and cavity geometry Bankoff [A] examined the mechanism of bubble nucleation in detail and arrived at the conclusion that boiling bubbles originate from pre- existing vapor pockets in cavities on the surface. Nestwater [5] exa- ililii.|-l, v J mwuzxogc UZBGUCUZ. i N in c f. o o o R. 4 Tu . . o 0 ll 2 0 MS 4 3 2 H i O l23l2l|23ll2316 MI E m I V V .1 [4/0 ooOOOAoooovVvaW '00 0 700°] 1 0 I I O [I o o, 0 CV iJAAJL 00/0 _ w el- ll. .1 _ _ olu. . o- . no I? "' 'III' I F 'v! v lva’vleI. to '0 OILO m 11"” - v ,0 o o v of 00/. L P b l b > b , Ly IP'IFI P . R l b w 0 O m 0 0 0 O O O O m 0 O O o m m m m 8 6 4 3 2 I 2.:Edmzéxtdpm ...; .pzuacbou oz:_om 232 tan,“ Effect of emery paper roughening for acetone boiling on COpper [3]. Figure l . 5 mined the boiling surface under a microscope and confirmed Bankoff's conclusion. Griffith and wallis [6] made a detailed study of boiling from ar- tifically formed sites. They correlated their observations with the following nucleation criteria. The Gibb's equation for a static, mechanical equilibrium across a curved interface which is a segment of a sphere can be written as _ -22 - Pv - P1 — AP - r (2 1) When the bubble is also in a state of thermal equilibrium the tempera- ture of the vapor inside the bubble must be identical to that of the surrounding liquid. Now the vapor is at its saturation temperature- corresponding to PV. Since PV > P the liquid surrounding the bubble 23 must be superheated. The excess temperature in the liquid can be re- lated to the pressure difference AP through the Clapyron equation. The Clapeyron equation is given as AP _ fg _ ETT- — T v (2 2) sat fg Eliminating AP from eqs. (2-1) and (2—2) we get 20 T V _ _ sat fg _ (Tw _ ism) _ AT — I, hfg (2 3) If the cavity mouth radius r, also called the critical radius, is substituted into eq. (2—3), the temperature difference (AT) becomes the nucleation superheat. This is the minimum wall superheat needed to start the bubble growing from a cavity with a mouth radius of r. Griffith and wallis [6] compared the experimental values of AT (nucleation superheat) for different artificial cavities with those cal- culated from eq. (2-3). The results showed that the nucleation criteria embodied in eq. (2—3) are correct. In one of their tests on artificial cavities they noted that the measured wall superheat was 20°F (11°C) as opposed to 3°F (1.7°C) as calculated from eq. (2—3). To explain this they argued that the mean surface temperature is not equal to the temperature felt by an active cavity i.e., the surface is much cooler in the vicinity of an active cavity than elsewhere. They con- cluded that the cavity mouth radius determines the wall superheat needed to initiate boiling and its shape determines its stability once boiling has begun. Figure 2(a) shows a reentrant cavity proposed by Griffith and Wallis. It is called "reentrant" because of its special internal shape. Figure 2(b) shows a plot of the reciprocal of radius vs. volume. When the radius of curvature becomes negative, then accord- ing to eq. (2-3), the liquid must be subcooled in order to flood the cavity. Therefore, such a cavity should be a very good vapor trap to serve as a nucleation site for the initiation of boiling. Marto and Rohsenow [7] conducted their tests on a 75mm diameter block with twelve reentrant cavities. Figure 3(b) shows a cross—sec- tional view of their artificial cavities which produced much better en— hancement in terms of heat transfer coefficient compared to some other en-5 hancement techniques they tested (see their results Figure 3(a)). Hummel [8] tested a stainless steel surface sprayed with scattered spots of teflon (a non wetting material) for the boiling of water. A nonwetting coating establishes a larger liquid—vapor interfacial curva- ture which in turn would require a lesser degree of superheat for its /—T\‘- (a, I, //‘1\ \ ‘1 (3)’/‘T\~. .‘ 83904 {/to) g r. \ \—_-L— -’ " (l) :anmd ’ " ...LBLL Effie V ' ‘ ‘3) (OVA)(B)(”/(]‘]\F\TF“~» ¢Vol me HH3) (4: U lb) Figure 2. (a) States of the liquid-vapor interface in a reentrant cavity. (b) Recriprocal (1/r) vs. vapor volume for liquid having a 90° contact angle [6]. Io‘ ' 1 T1 IfiWTYTIT— rm 1T1 Y‘T'T'TT] E V Run 2! wanton IIMSN : . unnoouau n:- : DURAN! CAvntt 53 L J I show :2 mason mus" I P mtu POROUS ". P / Inns 1 .- s? N C *- co nun 29.30 ronous ‘ m mcxu coumc (‘SpPORC z .05 ._ 9m 3 I 3 S * co ouon uou-oo:u~o~ Q. : O ‘0 DATA j 3 ‘ J . v ‘ i L. o .. A. _‘I C <5 .1 »- a I '0‘ 1411 1111111 1 111 1 11111,“ I0 I00 I000 YI; -tIAY .°' Ia) -1 r-OJOmm 0.64mm 0.35 l (b) Figure 3. (a) Effect of several surface treatments on sodium boiling at 65 mm Hg [7]. (b) Cross section of doubly reentrant cavities tested. existence. Hummel's results (see Figure A) show a dramatic enhance— ment where h increases an order of:magnitude for a given AT. 2.3 Mechanisms of heat transfer in nucleate boiling There are several hypothesis about the mechanism of heat transfer in boiling. First, a growing bubble is thought to be surrounded by a superheated thermal boundary layer. Heat is absorbed by the bubble from this boundary layer on its periphery and the heat is transported away as the bubble departs from the heated surface. Another mechanism.is known as thermal boundary layer stripping. Adjacent to the heated surface a thermal boundary layer exists in the fluid. Heat is thought to be transported by cyclic stripping of the thermal boundary layer by departing bubbles. Also, as the bubbles leave the surface, they allow the cold liquid to contact the surface. The cold liquid readily conducts heat away from.the hot surface. Such a mechanism appears to be more dominant in subcooled boiling as shown by Rohsenow and Clark [9]. The above ideas are also shared by a number of other researchers [10,11]. Yet another hypothesis is that the liquid at the base of a bubble is rapidly vaporized, as advocated by Edwards [12] and Moore and Mesler [13]. They visualize that evaporation takes place continuously from w thin liquid film.at the base of the bubble. The temperature of such a filniwould be expected to drOp significantly after the mass transfer has begun due to the associated very high heat transfer coefficient for thin film evaporation (similar to film condensation). 10 Ar‘l‘ o .F o w 20 x: a), I 50_ l I 1 j‘C Water boning on stainless steel Pulled stainless steel 1 wuth Teflon in pits 4, 20_ /—Smooth stanless steel I u wath Teflon SPOIS . III‘ I Is a 5 I0— J :3 m i? . «E C. a ‘1 ‘°’ " x 5 A j E o d 9 2| 1 8 Smooth "‘ I stantess steel _. .9 2__ surface ‘8 l ‘5 Putted stainless E steel Surface — ‘ I— g i 0 5 I0 IS 20 Afs‘y , .C Ia) - 5 —~ —.— — M 1 b l } . f3» 8 (3 Q ’ \ <4 {3) 1 ) ' {7:7}- A Figure A. (a) Enhancement for water boiling at 101 kPa (1 atm) on a stainless steel surface having minute nonwetted spots (30-60 spots/cmz, 0.25 mm diameter or less). (b) Enlarged photo of Teflon- spoted smooth surface [8]. 11 Moore and.Mesler [13] measured the surface temperature during nu— cleate boiling with a special thermocouple so designed as to measure the surface temperature of a small area and to have an extremely rapid re— sponse time. The thermocouple voltage was read on a potentiometer and the temperature fluctuations were viewed on an oscilloscope. The sur- face temperature was found to drop occasionally by 20 to 30°F (11 to 17°C) in about 2 milliseconds and then return to its previous level during the boiling of water. This indicated a rapid extraction of heat dur- ing a short time interval. Figure 5 shows a typical photograph of the oscilloscope face showing the surface temperature behavior. Moore and Mesler calculated the heat transfer rate during one tem- perature drop and then multiplied it with the frequency of the tempera- ture drop and showed that it accounted for about 70-90% of the average heat flux for many calculations. They also argued that the very high heat transfer coefficient (between 165—267 kW/m2°C) obtained during the temperature excursion would not be expected if convection was an important factor. They proposed that the only hypothesis that appears to be consis- tent with their observations is the process of vaporization at the base of the bubble. They visualize the details of what happens as follows. As a bubble grows on the surface it exposes the heating surface wet with a microlayer of liquid to the interior of the bubble. This microlayer rapidly vaporizes, removing heat rapidly from the surface until it is completely vaporized. This simple sequence of events is the only way that they were able to explain the rapid removal of heat occurring dur- ing the short period of time when they observed the surface temperature dropping rapidly. Figure 6 shows two extreme possibilities of bubble growth and the formation of a microlayer. They suggested it might be l2 O a c’l'lmolcur dlvlnlon I «autumnal/max! dmum Figure 5. A typical photograph of the oscilloscope face showing the surface tem- perature behavior [13]. 90000 DUO-Ll“. co-rumv Vivi-ll! .8 “1 mt “0'5 4:: 4:5 (II) (1::) (”r—I)_ KOO-Ll“. B I" WV WI“? an; m 5 m "I Fm an Figure 6. Sketches of two ways the microlayer might vaporize [l3]. possible to achieve better correlations if consideration was given to microlayer vaporization. 2.A Commercial enhanced surfaces Based on the past experience with special surfaces and the funda— mental studies mentioned in the previous sections, a number of techni- ques have been employed to manufacture a variety of commerical enhanced surfaces. A category of "porous surfaces" incorporates microholes (pores) con— nected to a matrix of interconnected reentrant cavities under the surface. These surfaces have proved to be a.mejor breakthrough. Figure 7a shows a cross-section of the HIGH—FLUX* surface which consists of a porous me- tallic matrix bonded to a metallic subtrate. The Gottzmann et. a1. [1A] results for pool boiling tests on a HIGH-FLUX surface are shown in Figure 7b. At a given heat flux the heat transfer coefficient is seen to in- crease by a factor of 20 due to the smaller AT required by the enhanced surface to sustain the heat flux. The average pore size within a cer- tain range has been reported to be the most significant dimension. Mbst of these pores are of reentrant shape according to Bergles and Chyu [15]. The pores within a matrix are interconnected such that vapor formed in one pore can activate adjacent pores. Cold metal working is also one of the techniques being employed to form a high density of interconnected reentrant cavities in the form of grooves or tunnels below the surface. These surfaces have openings all along the top of the grooves connecting them with the bulk liquid. Fig— *Trademark Union Carbide HEAT FLUX (BIu/hr-II2 lxlO-S Lunnolw ‘liiii’ l‘ VAPOR our [70 8‘; ZS M. \ \ VAPOR BUBBLE TRAPPED IN ACTIVE SITE (.1) 50 5 I (II I / //°' HIGH FLUX 1 J l J 05 I0 5.0 I00 500 Tu ' Tmt PF) (b) Figure 7. (a) Cross section of copper sintered porous boiling surface. (b) Enhancement provided by porous HIGH-FLUX surface for three fluids boiling at 101 kPa (1 atm) [1A]. p4 \Il ure 8 shows some of the commercially available boiling surfaces of this type. Figure 8a depicts a reentrant grooved surface formed by bending fin tips to form narrow cavity openings. Figure 8b shows the THERMO- EXCEL—E (trade mark) surface consisting of reentrant grooves with holes at the top of enclosed grooves. Figure 8c shows the GEWA—T surface con- sisting of narrow finned tubing with the fins flattened down. Figure 8d is the ECR-AO surface made by cross knurling through helical fins and then compressing the surface to form reentrant cavities. Yilmaz and westwater [l6] conducted single tube pool boiling tests on some of these commercial enhanced surfaces. Their results are shown in Figure 9. The ECR—AO and the HIGH FLUX surfaces gave the best results. These tubes also exhibited a critical heat flux (CHF) about AOZ greater than that of a conventional smooth tube. 2.5 Analytical work While a variety of enhanced surfaces have been manufactured and tested and are in fact commercially available, very little work has been done to mathematically model the enhancement mechanism to look for some heat transfer correlation for these surfaces. Nakayama et. al. [17] have proposed an analytical model of the dyna- }mic cycle of bubble formation. They conducted experiments on structured surfaces composed of interconnected internal cavities in the form of tun- nels and small pores connecting the bulk pool of liquid and the tunnels. Figure 10 shows the geometry of the structured surface they used in their experiments. Structural dimensions are given in Table l for three dif- ferent surfaces used for boiling water. The surface material was oxygen free copper. The most widely varied among the geometrical dimensions 16 H IYI - n "I. r _ ~ ———_— — ‘i‘ Pore Tunnel (bl Figure 8. Some patented boiling surfaces applied to circular tubes [1]. 17 7.5 fin/cm fin tube eaterpfslattemng M6909 O .O / ll 'Il; l II it Figure 8. (continued) Heal Flux, Clo, W/m2 18 ATE = Tw-TL' F 2 4 6 8 IO 2 4 6 a lo2 2 , 1 1'1 JIJITJIJITTIIIA lfingl‘IJllJTTTTIIJ 1 0 PLAIN - csas 4- o GEWA-T o ECR4o “ 6 _. v GEWA IZOD PINS/M o HIGH FLUX .[Io 1 THERMOEXCEL-E f8 2” 7+6 +4 '06: j. 8—- ' ~.. 6— «+2 I — " f 4"— ‘l 2 L 4.403 m. l8 (0 2‘ “+6 8 +4 5 '05::- —. u 8: . 3 E 6... . ° \ .32 I 4— ° - ‘ _ -rlO f8 2“ “l6 4 .04 1 1111111] 1 llllllll * 10° 2 468l0' 2 468l02 2 ATazTW-TL'C Figure 9. Comparative single-tube pool boiling test results for p-xylene at 101 kPa (latm). From [16]. 19 A J A la)r_—EZZS§s )5“! AéESS. ° A. 4% ......\..... «A A A li A & mnne d\‘w v V111 / l Figure 10. Geometry of the surface structure [17]. Table 1. Structural dimension of the tested surfaces [17]. fluid NO. do (o y "t Ar/A 20 was the pore diameter dO which was considered crucial for the ejection of vapor bubbles from the tunnel. Boiling curves are shown in Figure 11. Dependence of boiling performance on pore diameter dO is notable. To determine the latent heat contribution to the overall heat trans- fer rate, they measured boiling site density NA/A, the bubble departure frequency f , and the diameter of departing bubbles db for R-ll. The latent heat flux qL, was then computed from = 3 - qL (NA/Awb hfg pv(1r cab/6) (2 u) where pv is vapor density and hfg latent heat of vaporization. The ratio of qL to the total heat flux q is shown against q in Figure 12. The latent heat contribution for the structured surface is seen to be dominating at low heat fluxes. This is thought to imply that a significant role is played by the vaporization process in the tunnels. Nakayama et. a1. [17] have also conducted visual experiments to ob— serve the vaporization process in the tunnels. The apparatus shown in Figure 13 consists of a base block, two glass plates (30mm x 30mm, 1mm thick) forming a pool space above the base block, and a thin metal lid haveing a row of through holes (pores). A tunnel space is formed between the lid and the base block. Heat was supplied through the direct heating f the base block or by passing a d.c. current through the lid. The eva- poration phenomenon in the continuous tunnel was observed from the side. Based on their observations and experimental results they modeled a se- quence of events of the bubble growth and departure cycle at the pores in an effort to develop an analytical solution to find qL. 31 q ___.__ toml heat flux latent heat: flux 21 I 5 r valet Surfac- No. 6. [ horIzomaI surlare O "4 0.20 L o 9—2 0.1:. A 11—) 0.08 A 3 ’ ’— plaIn mtlace 3 ‘ .’ \ '0':— a. . 1 . F J 5 e I I .2 i F "‘ :- : 2» 1‘ L 0 I I - I I0 :- 5 11111 1 1 14 111111 1 1 5 LG 2 5 IO 2 Hall superheat 51' (K) Figure 11. Boiling curves of water [17]. 1.0 ~ Surface No. , A O RIlll-l A RUN-3 . x plain surface 0.5 "‘ 0 11111 J “3131771111 1 J heat flux qW/cn’) Figure 12. Contribution of latent heat transport to total heat flux [17]. ' . gou-————. glass plates R _” \ Ilqmd $1109ng 000' 0801070 Ci] Figure 13. Apparatus for observation experiment [ l7 ] . 3 [Surface No ,L (DU-I 5‘ [JV-2 3 10’ 5- A H.) 08 A \ > a, s:__IILN—2 o l‘ * A.LN-3 : 2; 013-4 ‘ D O 0'3 10‘:- D .. .E ‘ a > 5.0 F‘ r u. ‘0 U 1 ’- ‘ . z 100 E— A. ‘ 3 S:- h a F- O - P. v e 2 11 1 11111J .1 111111u1 1 1 11111J 11 .11 N 3 10° 2 s 10’ a s 107 a s predicted heat flux q (U/en’) pred. Figure 1A. Comparison between the measured heat flux and the predicted heat flux for water and nitrogen [17]. 23 Basically they have made use of conservation equations of mass and momentum in terms of meniscus height at the active and inactive pores. These equations were derived by referring to [18]. They derive expres- sions for the mass of liquid evaporated in the tunnels, the time for completing a cycle, and the fraction of total number of pores active 8. Then, the latent heat transport flux is given as .355 of liquid q = (evaporated in )h N/(Time)(Area) (2-5) L fg tunnels They also gave an empirical correlation for qex’ the heat flux from the outer surface of structured surface. l/y -x/y 98X = (AT/Cg) (NA/A) (2-6) where x,y and Cq are constants which could vary from one fluid to another and NA = 8N, where N is the total number of pores. Now q = qL + qex (2-7) In the process of arriving at Eq. (2—7) five empirical constants have been introduced to correlated their experimental results (see Figure 1A). A procedure of predictive computations for geometrically uniform surface structures (such as their own) is outlined. They noted that more experi- ments on similar surfaces were necessary to test the validity of their analytical model. The Gottzmann et. a1. [1A] model is strongly based on the assumption that effectively all vaporization occurs within the porous matrix or tun- 2A nels, see Figure 7a. They also make use of Gibb's equation and the well known Clapyron relation; p = __. - A- I, (2 l) h A:.=.__;Q£__ (2-2) Al [t 88.0 fg where AT = Tw - TS; letting %§-= m, the slope of the vapor pressure curve and substituting (2-2) in (2-1) gives AT = — (2-8) Neglecting the temperature drOp across the metallic matrix, the total wall superheat AT will be the sum of ATS, eq. (2-8), and the tem— B perature drOp ATLP across the thin liquid film covering the metallic walls. Therefore AlB = ATS + ATLF (2—9) and 2 AT =-2—9-+ 89? B rm ki (2-10) where a geometric factor B is introduced. The conduction of heat through the liquid from the hot solid surface to the liquid—vapor interface is called thin film evaporation (It is similar to film condensation). The optimum.radius is found by differentiating eq. (2—10) giving ok = __i 1/3 _ ropt. (8mg) (2 ll) Substituting back into eq. (2—10) gives kfimz AT; 15.15: = ——2—— (2-12) 27 Bo Gottzmann tested this equation for several surfaces with opthmmn pore sizes. Results are shown in Figure 15. The data is seen to be rather scattered but good qualitative agreement is found. 26 I T T '9 Theoreticok "L m2 V3 2,3 0 3: h '0.042l — q :L "2 \ O E A I A 33 D E ' - IO— 9' v -~ 2“ A v & ,. _ g ’ if n ‘3 SYMBOL FLUID PSIA 2 A o Nitrogen 6 o 4_ V A Oxygen [5 w A n Pr lone l5 '— ._. ‘ . R'Z '5 g V A Ammonia 15 w I Ethanol l5 E v R-ll IS A v Water IS I FPHB 6 I Toluene 5 ‘ R-ll 5.7 0 Water 5.7 v Water 2.l ' l J A R'll3 1 2.0 l,000 4,000 I0,000 40.000 HIGHEST MEASURED he Figure 15. Measured boiling data of 14 fluids on HIGH-FLUX sintered surface compared with values predicted by Eq. (2—12) [1“]. 3.0 EXPERIMENTAL DESIGN AND PROCEDURES 3.1 Pool boiling An experimental program was designed to study the enhancement mech— anism in porous structured surfaces in continuation of the Nakayama et. al. [17] work. However the exact geometry of the structured surface was known in the present study as opposed by the previous work by Nakayama where the geometry was only approximately rectangular as shown in Figure 16. To study the effect of a change in groove geometry on enhancement, boiling curves for two different groove cross-sections (triangular and rectangular) were obtained. The two special enhanced boiling test surfaces made of brass were designed, manufactured and tested in the laboratory. The test section (25.“mm dia) contained circular machined grooves: one surface with a triangular cross-section and the other with rectangular, coded surface No. l and No. 2 respectively for convenience. A thin copper lid (thick- ness 5pm) was soldered onto the surface which covered the machined grooves forming tunnels. The lid contained microdrilled holes (pores) all along the centerline of the grooves, thus connecting the tunnels with the bulk pool of water. Dimensions were carefully selected to have a uniform distribution of pores over the entire test section. Figure 17 shows detailed drawings of the test surfaces and Table 2 gives the dimensions of the surface structures. The outer section of the test surface was made very thin (0.3mm) to minimize the conduction heat 27 28 Figure 16. Magnified View of the surface [17]. 29 ( 0) COPPER LID J U /%===‘===I- J __—-—/ 31:33.:22353 /E======s== THREE / THERMOCOUPLE HOLES . . (b) TEST 5 ECT ION Figure 17. Test surface; (a) Side View (b) Top View 3O (C) SECTION A-A COPPER LID PORE DIA a, K71“? 1* his? 3 / jafi'r 'g, (d) 4153—? \\ \\‘ Figure 17 (continued) (0) and (d) Groove cross—sections. Table 2. Structural dimensions of the test surfaces “Iii?“ d o )‘o A Di “3 1 0.18 0.60 0.60 0.50 0.25 2 0.25 0.60 1.20 0.50 0.25 ‘ 31 loss in the radial direction. A high temperature resistant and low ther- mal conductivity adhesive was used to stick a circular ring at the bottom of the thin section of the test surface. The ring is in turn screwed to a cylindrical shell. Figure 18 shows the pool boiling setup. A specially designed nichrome filament heater is tightly screwed to the bottom of the test surface and is used to heat the test section. The assembly is then sealed off with a flanged U—tube. The heater supply and thermocouple wires run through the U—tube and out to the instrumentation panel. The whole assembly is then immersed in a large beaker containing double distilled water, maintained at saturation temperature through di- rect heating from a hot plate. Another smooth (ordinary machined) brass surface was also manufactured and tested to compare the enhancement ef- fect of the special surfaces. An adjustable D.C. power supply, monitored by a digital multimeter (DMM) was used for the test section heater. Figure 19 shows the power supply circuit. Copper—constantan thermo- couples and a digital thermometer were used to measure temperatures. Three thermocouples under the test section were used to record the tem- perature at distances of 2, M and 6mm from the top of the surface. Another thermocouple was used to record the bulk liquid temperature. The beaker and the test section were cleaned with acetone before each test. The water level in the beaker was maintained about 100m above the test section by periodic replenishment during the test while the pool of water was maintained at saturation temperature by a hot plate. The maxi- mum subcooling noted was O.3°C. The test section heater power supply was increased slowly in steps and then decreased following the same steps. J11. :Us l0 r J FigurellS: U-tube pool boiling facility (legend: l— test section, 2. test surface 3. thermocouple wires, 4. power leads 5. heater, 6. circular ring attached to the test surface using low conductivity adhesive, 7. cylindrical shell, 8. flauged U-tube, 9. thermometer for measuring bulk temperature, l0. beaker, ll. hot plate). U) POWER SUPPDY HEATER U) SHUNT DhflNl Figure 19. Power Supply Circuit. 3“ The following readings were taken after steady state conditions were reached at each step: 1. Temperature at 2mm (from the top of the surface), T2 2. Temperature at 4mm (from the top of the surface), Tu 3. Temperature at 6mm (from the top of the surface), T6 u. VOltage drop across test section heater, V" with the [FTC F 5. Voltage drop across the shunt, VS with the Dfifi. m 6. Bulk water temperature, isat The following calculations are performed to find the heat transfer coefficient, h. The electrical current, I, flowing through the circuit is obtained from <1 _ s I - fig. (3-1) where Rs is a known shunt resistance and V5 is the voltage drop across the shunt measured with the DMM. The power input to the heater is calculated as Power = V? I (3-2) and then the heat flux, q, is determined from q = Power/A (3.3) where As is the projected surface area of the test section. 35 The wall temperature, Tw, is obtained from TW = T2 — (Tu - T2) (3‘)“) The wall superheat then is AT = Tw - sat (3-5) and the heat transfer coefficient is calculated from h = q/AT (3—6) It is assumed in the above calculations that all the input heat flows through the test section i.e., negligible radial heat conduction to the outer section of the surface and negligible radiation and natural convection losses inside the cylindrical shell. Also, assuming a linear temperature gradient, the temperatures at 2 and 4mm are used to extra- polate to the surface temperature, T“. The data recorded and listed in appendix A show that the temperature gradient between 6, A and 2mm from the surface is very close to being linear at lower heat fluxes. At higher heat fluxes it is somewhat off because of possible radiation and natural convection losses inside the cylindrical shell and radial conduction los— ses to the outer section of the test surface. Total heat losses were es- timated to be about 5%. Appendix B gives a discussion of the experimental errors involved in the measurements. 36 3.2 Visualization experimental design and procedures The apparatus was designed to visualize the evaporation phenomenon in the grooves. The observation was along the axis of the groove rather than from the side as in Nakayama's case. Also, the apparatus was designed with grooves made with brass walls to exactly simulate the grooves on the actual test surface while in the Nakayama study the continuous grooves had glass walls. The apparatus, shown in Figure 20, consists of a base block made from brass with grooves machined across the axis of the base block. The grooves are covered with a copper lid (shim stock) containing micro— drilled holes, and two thin glass plates (24mm x 30mm) forming a pool space above the base block. These components are pasted together care- fully with a high temperature resistant epoxy so that no microscopic paths for the fluid existed between the tunnels and the pool other than the holes (pores) in the lid. The pore sizes used were 0.15mm and 0.25mm. There were three pores on top of each groove. The pool was filled with double distilled water, its top Open to atmosphere. Water was replenished with a dropper during the experiment. Heat was supplied through a D.C. power controlled heater under the base block. The evaporation process in the grooves of three different cross- sectional geometries (triangular, rectangular and circular) was recorded using a videotape camera and recorder and a motor driven still camera. The groove size was also varied to study the effect of change in size on the evaporation process inside the grooves. Heat transfer measurements could not be made with the visualization test section because of high heat losses to the environment from the base plate and the glass walls. (d) 37 Figure 20. Visualization experimental setup. (a) schematic diagram (Legend: 1. brass base block, 2. c0pper lid, 3. glass plates, A. heater, 5. camera, 6. florescent light, 7. diffusely reflecting screen, 8. pores); (b) triangular grooves cross-section; (c) rectan— gular grooves cross-section; (d) circular grooves cross-section. “.0 RESULTS AND DISCUSSION “.1 Pool boiling curves The data recorded from the pool boiling tests is included in the appendix A. Separate curves for each experimental run are plotted in Fig— ures 21 through 26. The data averaged over two experiments on each sur- face is plotted in Figure 27 along with two curves for the Nakayama en- hanced surfaces (coded surfaces No. 3 and No. A). Also, the smooth sur— face pool boiling curve obtained by Nakayama is shown. As shown in Figure 27, surface No. l performs well at low heat fluxes but reverts towards the smooth surface behavior at intermediate heat fluxes. Surface No. 2 is better over a higher range of heat flux and shows about a seven-fold improvement in the boiling heat transfer coef- ficient over the smooth surface of Nakayama. Nakayama and coworkers performed their tests over a relatively high heat flux range. Their curves show an improved behavior as the pore dia- meter is decreased approximately to the range of 0.1-0.25mm. However, the present surface No. 2 (pore dia. 0.25mm) did behave better than Nakayama's surface No. 3 (pore dia. 0.2nn) at least over the available common part of the heat flux range. It is also interesting to note that the tunnel pitch, At’ in the present surface No. 2 was 1.2mm compared to 0.6mm in the Nakayama's surface No. 3. This means that the groove spacing is rather a flexible dimension (within a certain range) and as long as a continuous supply of liquid enters through the pores and subsequent (thin fihh) eva- poration in the grooves is assured the surface will give good results. 38 39 0 Increasing MT! flux 0 Decreasing heat flux 2 2X I0 2 0‘ O inclpl¢nce q (KW/m) U/ .9 HEAT FLUX 9 / / 04 0-5 I 5 no 15 WALL SUPERHEAT AHK) Figure 21. First pool boiling curve for the smooth surface. 0 Increasing heat flux ODecreosing heat flux 2 2X“) 2 10 NE so , § incipience \\ as / 0' \\\$‘/ x 9/ :3 .J ... y“ 10 I.— <1 w I 5 // liO * / o/ 0.1 0-5 | 5 WALL SUPERHEAT ATiK) Figure 22. Second pool boiling curve for smooth surface. l0 A Increasing heat flux A Decreasing heat flux 2 2XIO 2 ID A A E 50 4 A g / 3‘ :6 a / A x D .J u. I0 ... A < Lu I A 5 I O-I 0-5 I 5 IO WALL SUPERHEAT ATIK) Figure 23. First pool boiling curve for Surface No. 1. 141 l5 A2 A Increasing heat flux A Decreasing hpoi flux 2 2XHO .o / A (ME; 5c) £5 ~c 1/ E: j/‘7AI U’ x :3 .A _J . I K) ... A <1 Lu I 5 I 04 (>5 I 5 I0 WALL SUPERHE AT AT (K) Figure 2A. Second pool boiling curve for Surface No. l. “3 I Increasing he I qux o Decreasing II at flux 2 2XIO 2 CI 2 50 i/ \ / 3 7 ‘7 35 U )// o / x 3 J . 0 u I0 I... <1 m I 5 I I O-I .0-5 I 5 IO wALL SUPERHEAT ATIK) Figure 25. First pool boiling curve for Surface No. 2. I Increasing he I flux 0 Decreasing n at flux 2 2XIO 2 I0 I 0 (mg; 5C) //,0 E 25 / ’6 U - / 7/ X / :3 .1 LL I0 [.— <1 w I 5 I I O! 05 I 5 I0 MI Figure 26. WALL SUPERHEAT AT (K) Second pool boiling curve for Surface No. 2. SUN%CE POREENA CODE d.(mm) A I PRESENT 0-I8 a 2 PRESENT 0.25 f [03 i O 3MI ' (co: II:‘--'/r:‘-CC> 99.9 1.A 100.6 100.5 100.A 100.3 0.3 3.6 99.9 5.A 101.5 101.2 101.1 101.1 1.1 9.1 100.0 12.0 103.5 103.1 102.7 102.5 2. A.8 99.9 22.1 105.6 10A.8 10A.1 103.8 3.9* 5.7 99.9 27.6 106.5 105.3 10A.6 10A.5 A.6 6.0 99.9 3A.3 107.7 106.3 105.A 105.0 5.1 7.6 99.8 A8.A 110.0 108.0 106.8 106.2 6.A 7.6 99.7 86.2 11A.6 111.1 108.9 107.8 8.1 10.6 99.7 A32.1 119.0 113.7 110.6 109.0 9.3 1A.2 99.7 fl89.7 12A.A 116.9 112.3 110.0 10.3 18.A 99.7 H33.3 119.0 113.7 110.u 108.8 9.2 1A.5 99.7 8A.8 11A.3 110.8 108.7 107.7 8.1 10.5 99.8 A8.0 109.9 107.8 106.7 106.2 6.A 7.5 99.8 33.8 107.1 105.7 10A.8 10A.A A.6 7.3 99.8 27.5 105.5 10A.A 103.8 103.5 3.8 7.2 99.8 22.0 10A.A 103.A 102.9 102.7 3.0 7.3 99.8 11.9 102.6 102.1 101.8 101.7 2.0 6.0 99.7 5.A 100.9 100.8 100.7 100.6 0.9 6.0 3" . . inc1pience 72 Table A. Data for the second pool boiling curve for smooth surface (plotted in Figure 22). 75a: c T10 '36 :2 TV -37 N r = 2% Ir.‘ sat (CO, (kw/n2) (0°: (0°? (00‘ (0°: Ikw/mz-CCF 100.0 3 0.4 : 100.2 § 100.1 i 100.1 I 100.1 0.1 ' 4.0 100.0 I 1.4 100.5 I 100.4 I 100.3 E 100.25 . 0.25 5.8 ' 100.0 ‘ 5.5 101.3 I 101.2 ' 101.1 I 101.1 I 1.1 5.0 100 0 12.1 103 8 I 103.2 102.7 i 102.5 ' 2.5 I 4.8 100.0 16.4 105.0 ' 104.4 103.8 103.4 3.5 4.7 100.0 22.3 106.4 105.5 104.6 104.2 4.2* 5.3 99.9 48.8 110.1 108.1 106.6 105.9 6.0 8.1 99.9 86.6 114.3 110.7 108.4 107.3 7.4 11.7 99.9 132.1 118.7 113.3 110.0 108.4 8.5 15.5 99.9 190.3 124.2 116.5 111.8 109.5 9.6 19.8 99.9 I132.7 118.9 113.4 110.0 108.3 8.4 15.8 100.0 85.8 114.2 110.7 108.4 107.3 7.3 11.7 100.0 47.0 110.1 108.2 106.8 106.1 6.1 7.7 100.0 21.0 106.2 105.3 104.4 104.0 4.0 5.3 100.0 16.0 104.8 104.0 103.4 103.1 3.1 5.2 100.1 11.9 103.5 103.0 102.6 102.4 2.3 5.2 100.1 5:4 101.6 101.4 101.2 101.1 1.1 4.9 100.2 1.4 100.5 100.5 100.4 100.4 0.2 7.1 *incipience L~_ 73 Table 5. Data for the first pool boiling curve for enhanced Surface No.1 (plotted in Figure 23). 0 55a: c .10 '16 :2 TV =§?_w r = ET w sat .. (0° (kw/n2) (00‘ 0°? (00‘ (c0, Ira <‘—CCI 100.0 5.3 7 100.8 I 100.6 100.4 100.3 I 0.3 i 17.7 I 99.9 8.4 . 101.0 100.7 100.5 5 100.3 I 0.4 i 19.2 I 99.9 12.0 101.4 101.0 100.7 . 100 5 I 0.6 I 20.1 99.9 21.6 102.4 101.7 101.2 ‘ 100.8 0.9 24.0 100.0 34.2 103.8 102.7 102.0 101.5 1.5 22.8 100.0 48.6 105.3 103.8 102.8 102.0 2.0 24.3 100.0 67.6 107.4 105.3 104.0 103.0 3.0 22.5 100.0 85.5 109.4 106.8 105.1 103.8 3.8 22.5 99.9 133.8 115.2 111.1 108.0 105.6 5.7 23.5 99.8 ' 195.5 122.2 116.5 110.9 106.5 6.7 29.2 99.9 I 138.6 116.0 1 111.9 108.6 106.0 6.1 22.7 99.9 I 88.3 110.2 107.4 105.6 ' 104.3 4.4 20.1 99.9 65.1 107.5 105.5 104.2 103.2 3.3 19.7 99.9 49.3 105.7 104.2 103.2 102.4 2.4 . 20.5 100.0 34.0 104.0 102.8 102.2 101.7 1.7 20.0 100.0 22.3 102.6 101.8 101.4 101.1 1.1 20.3 100.0 12.2 101.4 101.0 100.8 100.6 0.6 20.3 100.0 .5 101.0 100.7 100.5 100.4 0.4 21.3 100.0 5.5 100.8 100.6 100.4 100.2 0.25 22.1 1__- _Il_1 Table 6. No. 1(plotted in Figure 2A). 7A Data for the second pool boiling curve for enhanced Surface T55: q 110 'T6 T2 Tw AT r =IZ% =T —T - w sat H (CO (kl-J/rr.‘/ (cm (to I . (0° (0° I: 1.;I.-:/.~f—~Cn 100.0 3 5.5 100.8 100.6 I 100.4 I 100.3 I 0.3 18.3 99.9 8.5 101.0 100.7 I 100.5 I 100 3 I 0.4 21.2 99.9 12.1 101.2 100.8 ' 100.6 I 100.4 0.5 24.4 99.9 16.5 101.7 101.2 100.8 ‘ 100.5 0.6 27.5 99.9 21.4 102.3 101.6 101.2 100.9 1.0 21.4 99.9 34.6 103.9 102.8 102.1 101.5 1.6 21.6 99.9 48.9 105.7 104.2 103.0 102.1 2.2 22.2 99.9 66.3 107.6 105.6 104.0 102.8 2.9 22.8 99.9 86.0 110.0 107.4 105.2 103.6 3.7 23.2 99.9 135.8 111.8 111.5 108.0 105.3 5.4 25.1 99.9 I 196.6 122.2 116.2 111.0 106.9 7.1 27.7 99.9 135.5 116.0 111.7 108.0 105.1 5.2 26.1 100.0 85.9 110.4 107.8 105.5 103.7 3.7 23.2 99.9 66.2 108.1 106.0 104.4 103.2 3.3 20.1 99.9 49.0 106.1 104.5 103.3 102.4 2.5 19.6 99.9 34.4 104.3 103.2 102.4 101.8 1.9 18.1 99.9 21.3 102.6 101.9 101.4 101.0 1.1 19.4 99.9 16.5 101.9 101.4 101.0 100.7 0.8 20.7 99.9 12.1 101.2 100.9 100.7 100.5 0.6 20.2 99.9 8.5 101.0 100.7 100.5 100.3 0.4 21.3 99.9 5.5 100.8 100.6 100.4 100.2 0.3 18.3 _ __ -_-I-,_--I_-__.L-. ___..I 75 Table 7. Data for the first pool boiling curve for enhanced Surface No. 2 (plotted in Figure 25). Tsat Q T10 T6 T2 TV =§T_T h z'Z% 1: sat (CO) (km/m2) (0°: (0°) (0°, (0°) LkW/mz-CO) 100.2 5.1 100 4 100.3 100 3 100.3 0.1 50.8 100 1 12.7 100.8 100.9 100.6 100.3 0.2 63.6 100.1 22.3 101 7 101.6 101.1 101.5 0.6 37.2 100.1 37.0 102.6 102.2 101.6 100.9 0.8 46.3 100.1 49.0 103.4 102.8 101.9 101.0 0.9 54.3 100.1 65.0 104.6 103.7 102.5 101.4 1.3 50.0 100.1 85.3 106.0 104.8 103.2 101.6 1.5 56.9 100 1 110.7 107.7 106.0 104.1 102.2 2.1 52.7 100 1 131.0 108.9 106.8 104.6 102.4 2.4 54.6 100.1 191.7 112.6 109.4 106.1 102.9 2.9 66.1 100.1 137.0 109.6 107.3 104.9 102.5 2.4 57.1 100.1 109.2 107.9 106.2 104.2 102.3 2.2 49.6 100.0 84.3 106.4 105.1 103.6 102.1 2.1 40.1 100.0 69.3 105.4 104.4 103.2 102.0 2.0 34.7 100.0 51.4 104 2 103.5 102.6 101.6 1.6 32.1 100.0 35.9 103 1 102.8 102.0 101.2 1.2 30.0 100.0 22.5 102.2 102.0 101.6 101.2 1.2 18.8 100.1 12.2 101.6 101.5 101.2 100.9 0.8 15.3 100.1 6.3 100.6 100.6 100.4 100.2 0.1 63.1 Table 8. No. 2 (plotted in Figure 26). 76 Data for the second pool boiling curve for enhanced Surface Tsat q T10 T6 T2 Tw =$T_T h z'Ef w sat (0°, (kW/m2) (0°: (0° (0°\ (0°) ka/mZ—C°> 100.2 5.7 100.4 100.3 100.3 100.3 0.1 57.0 100 2 12.3 100.8 100.8 100.6 100.4 0.2 61.5 100.2 21.7 101.6 101.6 101.1 100.6 0.4 54.3 100.1 35.0 102.5 102.2 101.5 100.8 0.7 47.0 100 1 49.6 103.4 102.9 102.0 101.1 1.0 49.6 100.1 66.9 104.6 103.7 102.5 101.3 1.2 55.7 100.1 86.5 105.8 104.5 103.0 101.5 1.4 61.8 100.1 107.1 107.3 105.5 103.7 101 9 1.8 59.5 100.2 132.2 108.9 -106.7 104.4 102.1 1.9 69.6 100.2 195.6 113.1 109.8 106.4 103.0 2.8 69.9 100.2 132.6 109.3 107.1 104.8 102.5 2.3 57.7 100.2 110.0 107 9 106.2 104.2 102.2 2.0 55.0 100.2 89.0 106.7 105.4 103.7 102 0 1.8 49.4 100.2 67.0 105.2 104.4 103.0 101.6 1.4 47.9 100.2 50.4 104.2 103.6 102.5 101.4 1.2 42.0 100.2 34.7 103.0 102 7 102.0 101.3 1.1 31.6 100.2 22.8 102.2 102 2 101.6 101.0 0.8 28.5 100.2 12.5 101.4 101.4 101.1 100.8 0.6 20.9 100.2 5.7 100.4 100 4 100.3 100.2 —- -— 77 Appendix B Errors involved in the measurements A digital temperature readout device was used which read tempera- ture to one tenth of a degree (°C). At lower heat fluxes when AT was of the order of 0.1 to O.3°C an error of up to 100% could be involved. At higher heat fluxes the temperature gradient in the test section below the test surface is not quite linear due to the effect of the elec— tric heater bolt hole; thus, the two thermocouples nearest the test sur- face are used for extrapolation purposes. The radial conduction heat losses to the outer section of the test surface tend to increase at high- er temperatures but the higher boiling heat transfer coefficient from the test surface reduces its thermal resistance correspondingly. Total heat losses were estimated to be about 5% over the heat flux range at which experiments were performed. Therefore, the error should not be significantly high. It should be noted that for comparison purposes the two surfaces (i.e., a smooth surface vs enhanced) are of the same design and hence their errors are similar. LIST OF REFERENCES 10. ll. 12. 13. l“. 78 LIST OF REFERENCES Webb, R.L., The Evolution of Enhanced Surface Geometries for Nu— cleate Boiling, Heat Transfer Engineering, Vol. 2, No. 3—8, pp. H6- 69, 1981. Jakob, M., Heat Transfer, Wiley, New York, pp. 636-638, 1949. Kurihari, H.M. and Myers, J.E., Effects of Superheat and Roughness on the boiling Coefficients, AIChE J., v01. 6, No. 1, pp. 83-91, 1960. Bankoff, S.G., Trans. Am. Soc. Mech. Engrs., 79, 735 (1957). Westwater, J.W., Clark, H.B., and Strange, P.S., Chem. Eng. Progr. Symp. Ser. No. 29, 55, 103 (1959). Griffith, P., and wallis, J.D., The Role of Surface Conditions in Nucleate Boiling, Chem. Eng. Prog. Symp. Ser., v01. 55, No. 49, pp. [749—63, 19590 Marto, P.J., and Rohsenow, W.M., Effects of Surface Conditions on Nucleate Pool Boiling of Sodium, J. Heat Transfer, V01. 88, pp. 196— 20“, 1966. Hummel, R.L., and Young, R.K., Improved Nucleate Heat Transfer, Chem. Eng. Prog. Symp. Ser., v01. 61, No. 59, pp. 26u-H70, 1965. Rohsenow, W.M., and Clark, J.A., idid., 73, 609 (1951). Jakob, Max, "Heat Transfer", V01. 1, p. 61“ Wiley, New York. Forster, K. and Grief, R., Trans. Am. Soc. Mach. Engr., J. Heat Transfer, 81, H3 (1959). Edwards, D.K., M.S, Thesis, Univ. of Calif., Berkeley, California (1956). Moore, F.D. and Mesler, R.B., The Measurement of Rapid Surface Temperature Fluctuation during Nucleate Boiling of Water, AIChE J., Vol. 7, No. 4, pp. 620—629, 1971. Gottzmann, C.F., Wulf, J.E., and O'Neill, P.S., Theory and Applica- tion of High Performance Boiling Surfaces to Components of Absorp— tion Cycle Air Conditoners, Proc. Conf. Hat'l. Gas Hes. Tech., Session V, paper 3, Chicago, February 28, 1971. 15. 16. 17. 18. 19. 20. 7.9 Bergles, A.E. and Chyu, M.C., Characteristics of Nucleate Pool Boiling From Porous Metalic Coatings, Advances in Enhanced Heat Transfer, 1981, HTD4Vol. l8. Yilmaz, S. and.Westwater, J.W., Effect of Commercial Enhanced Sur— faces on the Boiling Heat Transfer Curve, 20th Nat'l Heat Transfer Conf., Milwaukee AIChE Paper, August 2—5, 1981. Nakayama, w., Daikoku, R., Nakajima, T., Dynamic Model of Enhanced Boiling Heat Transfer on Porous Surfaces Part 1,11, ASME J. of Heat Transfer, Vbl. 102, No. 3, 1980, pp. 445-456. L'Ecuyer, M.R.L., and Murtby, S.N.B., "Energy Transfer from a liquid to Gas Bubbles Forming at a Submerged Orifice", NASA TN D-2547, 1965. Arshad, J. and Thome, J.E., "Enhanced Boiling Surfaces: Heat Trans- fer Mechanism and Mixture Boiling", Submitted for the ASME-JSME Thermal Engineering Joint Conference, Honololu, March 1983. Thome, J.E., "Prediction of Binary Mixture Boiling Heat Transfer Co— efficient Using Only Phase Equilibrium Data; accepted for publication by Int. J. Heat Mass Transfer. "'741317777171117777777715’ 129