International Journal of Heat and Mass Transfer 53 (2010) 1015–1022

Contents lists available at ScienceDirect

International Journal of Heat and Mass Transfer

journal homepage: www.elsevier .com/locate / i jhmt

An experimental study on CHF enhancement in flow boiling using Al2O3 nano-fluid

Tae Il Kim *, Yong Hoon Jeong, Soon Heung ChangKorea Advanced Institute of Science and Technology, 373-1, Guseong-dong, Yuseong-gu, Daejeon 305-701, Republic of Korea

a r t i c l e i n f o

Article history:Received 4 April 2009Received in revised form 1 October 2009Accepted 1 October 2009Available online 11 December 2009

Keywords:Nano-fluidCHFFlow boilingDeposition

0017-9310/$ - see front matter � 2009 Elsevier Ltd. Adoi:10.1016/j.ijheatmasstransfer.2009.11.011

* Corresponding author. Tel.: +82 42 350 3856; faxE-mail addresses: skyimgf@kaist.ac.kr (T.I. Kim

Jeong), shchang@kaist.ac.kr (S.H. Chang).

a b s t r a c t

The critical heat flux (CHF) is one of the most important thermal hydraulic parameters in heat transfersystem design and safety analyses. CHF enhancement allows higher limits of operation conditions suchthat heat transfer equipment can be operated safely with greater margins and better economy. The appli-cation of nano-fluids is thought to have strong potential for enhancing the CHF. In this study, zeta poten-tials of Al2O3 nano-fluids were measured and flow boiling CHF enhancement experiments using Al2O3

nano-fluids were conducted under atmospheric pressure. The CHFs of Al2O3 nano-fluids were enhancedup to �70% in flow boiling for all experimental conditions. Maximum CHF enhancement (70.24%) wasshown at 0.01 vol% concentration, 50 �C inlet subcooling, and a mass flux of 100 kg/m2 s. Inner surfacesof the test section tube were observed by FE–SEM and the zeta potentials of Al2O3 nano-fluids were mea-sured before and after the CHF experiments.

� 2009 Elsevier Ltd. All rights reserved.

1. Introduction

The critical heat flux (CHF) is defined as the heat flux when theboiling heat transfer coefficient between a heated surface and fluidis dramatically dropped as the phase of fluid near the heated sur-face changes from liquid to vapor due to a rise of heat flux or sur-face temperature, or a change of flow rate, pressure, etc. [1]. WhenCHF occurs in a heat flux control system, the heater surface tem-perature is sharply increased as a result of a drop of the boilingheat transfer coefficient, and this could lead to catastrophic failureof the heated surface. CHF enhancement allows higher limits ofoperation conditions such that heat transfer equipment can beoperated safely with greater margins and better economy. Accord-ingly, many researchers have been developing methods to enhancethe CHF, including induction of a swirl flow by twisted tape or aribbed tube, increase of the heat transfer surface by the implemen-tation of fin structures, enlargement of turbulent flow, flow vibra-tion, nano-fluids, etc.

Nano-fluids are a new type of engineered fluids that containnano-sized particles less than 100 nm. These nano-sized particlescan improve thermal hydraulic properties of a fluid such as ther-mal conductivity, heat transfer coefficient, wettability, etc. [2].Modern nano-technology provides many opportunities to utilizenano-sized particles in this size range and thus nano-fluids haveenormous potential for practical application.

While numerous studies on CHF enhancement using nano-flu-ids in pool boiling have been reported, there is little data about

ll rights reserved.

: +82 42 869 3810.), jeongyh@kaist.ac.kr (Y.H.

CHF in flow boiling using nano-fluid. The main objective of thepresent study is to conduct CHF experiments at low flow and lowpressure using a nano-fluid and confirm CHF enhancement. Possi-ble mechanisms underlying CHF enhancement via application of aof Al2O3 nano-fluid are also explored.

2. Previous works

Most CHF experiments using nano-fluids were conducted inpool boiling conditions, and CHF was enhanced by up to 200%.The mechanism underlying this enhancement has yet to be clari-fied and is still under discussion. Meanwhile, relatively few CHFexperiments using nano-fluids in flow boiling condition have beenconducted, primarily due to stability and cleaning problems.

Kim et al. conducted an experimental study on the CHF charac-teristics of nano-fluids in pool boiling. Their results showed thatthe CHFs of nano-fluids containing TiO2 or Al2O3 were enhancedup to 100% over that of pure water. Also, SEM observations subse-quent to the CHF experiment revealed that nano-particles weredeposited on the wire surface. They concluded that the CHFenhancement could be attributable to enhanced wettability viathe deposition of nano-particles [3].

Bang et al. also investigated CHF characteristics of nano-fluids inpool boiling. They found that CHFs of nano-fluids containing alu-mina, zirconia, and silica nano-particles dispersed in water were en-hanced. They concluded that nano-particles are deposited on theheater surface, forming a porous layer during nucleate boiling. Thisporous layer improves the wettability of the surface considerably,as measured by a marked reduction of the static contact angle [4].

Jeong et al. investigated the wettability of heated surfaces inpool boiling using surfactant solutions and nano-fluids. Contact

1E-3 0.01 0.1 115

20

25

30

35

40

45

50

55

60

65

Avg

. zet

a po

tent

ial(m

V)

Concentration of Nano fluid(vol%)

Fig. 2. Average zeta potentials (1–24 h) with different concentrations.

1016 T.I. Kim et al. / International Journal of Heat and Mass Transfer 53 (2010) 1015–1022

angle measurements showed that both surfactant solutions andnano-fluids exhibited enhanced wettability. In addition, the surfac-tant solutions exhibited a greater decrease in contact angle, andhence further increased wettability [5].

Kim et al. conducted flow boiling CHF experiments using nano-fluids incorporating alumina, zinc-oxide, diamond. Their resultsshowed that the CHF values of the nano-fluids were enhanced byup to 40–50% with respect to pure water at 2000–2500 kg/m2 smass flux whereas the CHF was not enhanced at 1500 kg/m2 s massflux. They attributed the enhancement to nano-particle deposition,and the enhancement appeared to be weakly dependent on nano-particle concentration for the alumina nano-fluids, whereas it in-creased more pronouncedly with nano-particle concentration forthe zinc-oxide and diamond nano-fluids [6].

Many researchers have investigated the effects of time, temper-ature, concentration, particle type, dispersion medium and pH onthe stability of nano-fluids. It has been found that pH is the mostimportant factor affecting the dispersion stability of nano-fluids.

Wäsche et al. showed that the zeta potentials of nano-fluidscontaining Al2O3, SiC, and Si3N4 were changed by the pH of thenano-fluids. They also showed that the zeta potentials of thenano-fluids were slightly increased with increasing time [7].

Zhu et al. showed that the zeta potential and absorbency wereimportant bases for selecting conditions for dispersing particles.They also found that pH has a critical effect on the stability ofthe alumina suspension [8].

While there is abundant data on the effects of pH on the zetapotentials of Al2O3 nano-fluids, there is relatively little data regard-ing the effects of time and concentration on the zeta potentials ofAl2O3 nano-fluids.

3. Dispersion stability of Al2O3 nano-fluids

3.1. Preparation and measurement of Al2O3 nano-fluids

Al2O3 nano-particles were purchased from Nanostructured &Amorphous Materials, Inc. Vender specified size and purity of thenano-particles is 50 nm and 99.7% respectively. The Al2O3 nano-fluids were subjected to 2 h of sonication in an ultrasonic bath.The zeta potentials of the Al2O3 nano-fluids were measured usingan ELS-Z2 produced by Otsuka Electronics. The ELS-Z2 measureselectrophoretic mobility by electrophoretic light scattering andthe zeta potential is then calculated using the Smoluchowski Equa-tion. Also, the pHs of the Al2O3 nano-fluids were measured to eval-uate the reliability of the zeta potential results.

1001010.120

30

40

50

60

Zeta

pot

entia

l(mV)

Time(hours)

0.001vol% 0.01vol% 0.05vol% 0.1vol% 0.5vol%

Fig. 1. Zeta measurements according to increasing time; KAIST experimental data.

3.2. Zeta potential of Al2O3 nano-fluids

All of the zeta potentials of the Al2O3 nano-fluids were in arange of 30–60 mV. These results indicate that the Al2O3 nano-flu-ids were stable as colloidal fluids. The zeta potentials of Al2O3

nano-fluids slightly increased within the margin of error with anincrease of time within 1 day (Fig. 1). Wäsche et al. also showedthat the zeta potential increased after 24 and 48 h of ageing time.They concluded that the ageing time plays a role in the develop-ment of an electrical double layer, either due to suspension effectswith regard to homogenization of powder particles or due to solu-tion effects, which may lead to a change in the composition orstructure of the electrical double layer [7]. From the present re-sults, it is verified that the dispersion stability of the Al2O3 nano-fluids during CHF experiment will be adequate if the experimenttime is less than 1 day.

Also, the zeta potentials of Al2O3 nano-fluids slightly increasedwith an increase of particle concentration from 0.001 vol% to0.5 vol% (Fig. 2). Note, however, that Wäsche et al. showed thatthe zeta potentials of Al2O3 nano-fluids decreased with an increaseof particle concentration from 1 vol% to 10 vol%. This discrepancy

Fig. 3. Relation between pHs and zeta potentials of Al2O3 nano-fluids from Isobe’data [9].

Fig. 4. Schematic diagram of experimental loop.

Table 1Test matrix of CHF experiment

Test matrix

Uniformly heated cylindrical tubeOuter diameter 12.78 mmInner diameter 10.98 mmL/D ratio 45.53Heated length 500 mm

Vertically upward flowPressure 101.3 kPaMass flux 100–300 kg/m2 sInlet subcooling 25, 50 �C

Working fluidTotal fluid 53 lDI waterNano fluid/Al2O3 0.001–0.1 vol%

T.I. Kim et al. / International Journal of Heat and Mass Transfer 53 (2010) 1015–1022 1017

may be due to the difference in the concentration range. Furtherinvestigation should be conducted to elucidate the effect of con-centration on the zeta potential. Nonetheless, from the present re-sults, it is verified that the dispersion stability of the Al2O3 nano-fluids during the CHF experiment will be sufficiently high if theconcentration of the nano-fluid is in a range of 0.001 vol%–0.5 vol%.

The pH values of Al2O3 nano-fluids having a concentration of0.05 vol% were also measured and the results showed little changewith increasing time within a period of 3 days. The relation be-tween pH and zeta potential corresponds well with Isobe’s data(Fig. 3) [9]. Thus, it is found that the zeta potentials of the Al2O3

nano-fluids are reliable.

4. CHF experiments with Al2O3 nano-fluids

Fig. 5. Schematic diagram of test section.

4.1. Experimental apparatus and procedure

4.1.1. Experimental apparatusFlow boiling CHF experiments using Al2O3 nano-fluids were

conducted in KAIST’s low pressure water loop, which is shown inFig. 4. This experimental loop consists of an electromagnetic flow

50 100 150 200 250 300 350300

400

500

600

700

800

900

1000

1100

1200

1300

1400

1500

DI water: 50'C inlet subcooling DI water: 25'C inlet subcooling Groeneveld: 50'C inlet subcooling Groeneveld: 25'C inlet subcooling

CH

F(k

W/m

2 )

Mass flux (kg/m2 s)

Fig. 6. CHF with pure water compared with 1995 CHF look up table of Groeneveld[11].

0.1 1 10 100 1000 10000 100000 10000000.1

1

10

100

1000

10000

100000

300 kg/m2s

200 kg/m2s

100 kg/m2s

Bubbly

Bubbles sulgs

Slugs

Churn

Wispy-annular

DI water : 50 'C inlet temp.DI water : 75 'C inlet temp.0.01% Al

2O

3 Nano-fulid : 50 'C inlet temp.

0.01% Al2O

3 Nano-fulid : 75 'C inlet temp.

Mom

entu

m F

lux(

vapo

ur)

(kg/

s2 m)

Momentum Flux(liquid) (kg/s 2 m)

Annular

Fig. 7. Flow regimes of CHF experiments using Hewitt and Roberts map [12].

1018 T.I. Kim et al. / International Journal of Heat and Mass Transfer 53 (2010) 1015–1022

meter, two pre-heaters to control the inlet temperature of theworking fluid, a condenser for cooling of the working fluid, a surgetank with an overhead water reservoir, a CRN2-40 centrifugalpump, a needle valve to provide throttling, and a test section tube.A test section tube was directly heated by an electrical DC powersupply unit with 32 V rated output voltage and 2000 A rated out-put current. The working fluids were circulated by a centrifugalpump with a variable speed driver.

The working fluid flows vertically upward in the test sectiontube. The dimensions of the test section tube and flow parametersare listed in Table 1. The test section is a SS-316 circular tube of12.78 mm outer diameter, 0.90 mm thickness, and 500 mm length.The L/D ratio of the test section tube is 45.53, which exceeds the Le/

Vapor region

Normal LFDmechanism

Liquid region Dryout

Fig. 8. Normal CHF mechanisms in an annular

D ratio of 20, the minimum value for a fully developed flow underthe test conditions. Five Type-K thermocouples with 1.5 mm outerdiameter were attached on the outer surface of the test sectiontube to measure outer surface temperatures and detect the onsetof CHF. The current and voltage between both electrodes weremeasured. The temperatures of the working fluid at the inlet andoutlet of the test section tube were measured by in-stream T-typesheathed thermocouples. A schematic diagram of the test section isshown in Fig. 5.

4.1.2. Experimental procedureThe experimental procedure is as follows. The working fluid is

circulated by a centrifugal pump and heated by pre-heaters to re-move non-condensable gas. Degassing is performed for an hourunder atmospheric pressure. After the degassing process, a sampleof the working fluid is extracted for measuring the zeta potentialand pH in order to confirm the dispersion stability of the Al2O3

nano-fluid. The heating power in the test section is increased grad-ually by slowly increasing the voltage of the test section. The volt-age of the test section is increased stepwise with thermalequilibrium of the working fluid in the loop. At least two consecu-tive runs were conducted for each condition. The increment of heatflux used near the CHF is �20 kW/m2. The CHF condition is definedas a sudden increase in the temperature of the test section tubesurface. Heat flux in the test section is calculated as

q00 ¼ VIpDiL

where V and I are the measured voltage and current, and Di and Lare the test section inner diameter and length, respectively. AfterCHF experiments, a sample of working fluid is extracted for measur-ing the zeta potential and pH to confirm the dispersion stability ofthe Al2O3 nano-fluid. After each CHF experiment, the test sectiontube is replaced with a new tube.

To confirm the dispersion stability of Al2O3 nano-fluid in theloop during the CHF experiments, zeta potentials of the Al2O3

nano-fluid sample extracted from in the loop were measured byan ELS-Z2 before and after the CHF experiments. Also, the pH val-ues of the Al2O3 nano-fluid sample are measured to evaluate thereliability of the zeta potential results.

Uncertainty analyses were carried out by the method reported byMoffat [10]. Mass flux uncertainty was estimated as 3% at 100 kg/m2 s, 2.5% at 200 kg/m2 s and 1.7% at 300 kg/m2 s. Temperature mea-surement uncertainties were primarily estimated considering thethermocouple calibration and temperature correction from the ther-mocouple reading to the reference surface. The maximum variationof the five measured wall temperatures (K-type thermocouples) was±0.5 �C. The uncertainty in the inlet and outlet working fluid temper-atures (T-type thermocouples) was estimated to be less than ±1 �C.The maximum error in controlling the inlet temperature was±0.5 �C. Uncertainty in heat flux was estimated by taking into ac-count the voltage and current. Heat losses to surroundings were lessthan 1.5% for heat flux conditions of 100–2000 kW/m2, assuming auniform temperature distribution with a heater surface temperature

Local dryout due tounstability of annular flow

Vapor region

Liquid region

dryoutLocal

flow and local dryout in an annular flow.

T.I. Kim et al. / International Journal of Heat and Mass Transfer 53 (2010) 1015–1022 1019

of 180 �C. The heated surface area also contributed to the uncer-tainty. Considering all these factors, the overall uncertainty in theheat flux was 4%; taking into account the uncertainties of heat fluxand power, the uncertainty in the CHF was around 5%.

4.2. Results and discussion

All experiments were conducted in flow boiling under atmo-spheric pressure at fixed inlet conditions of temperature and massflux. The CHF results of DI water agreed well with the results pro-vided in the 1995 CHF look up table prepared by Groeneveld(Fig. 6) [11].

The CHFs of the Al2O3 nano-fluids were enhanced, by as muchas 70%, for all experiment conditions. Maximum CHF enhancement(70.24%) was shown at 0.01 vol% concentration, 50 �C inlet subco-oling, and a mass flux of 100 kg/m2 s.

Using the plot of Hewitt and Roberts [12] for a vertical upwardflow, flow patterns were explored at different mass flux levels andinlet temperatures. The flow regimes of all conditions were annularflow (Fig. 7). In the annular flow regime, the mechanism of CHF isliquid film dryout [1]. If CHF occurs by the normal liquid film dry-out mechanism, nano-particle deposition cannot account for theCHF enhancement. The test section of the CHF experiment has aL/D ratio of 45. This number is higher than the minimum valuefor a fully developed flow, but lower than other normal values(>100). Because of this, the annular flow may be not stable, andthus some local dryout can occur [13] (Fig. 8). When local dryoutoccurs, rewetting would take place easily owing to the effects of

Fig. 9. FE–SEM observations of test section tube inner surfaces after CHF experiment(c) 0.01 vol% Al2O3 nano-fluid and (d) 0.1 vol% Al2O3 nano-fluid.

nano-particle deposition, ultimately leading to enhanced CHF.The deposition of nano-particles enhances the wettability of theliquid film to the heated surface, and hence promotes the CHF.FE–SEM observation showed deposition of Al2O3 nano-particleson the inner surface of test section tube (Fig. 9). Many researchersalso concluded that the deposition of nano-particles on the heatedsurface is the main reason for CHF enhancement in pool boilingbecause wettability of a fluid onto a heated surface is enhancedwhen nano-particles are deposited on the heated surface [7–10].

Sarwar et al. showed that CHF was enhanced by up to about 30%in flow boiling using a micro-porous coating. They also attributedthe CHF enhancement to enhanced wettability [14].

The effect of deposition can be clarified by conducting experi-ments using a nano-particle deposited tube with DI water as aworking fluid.

The CHFs of Al2O3 nano-fluids were increased with increasingmass flux at inlet subcooling of 50 �C and 25 �C (Fig. 10). However,the CHF enhancement ratios of Al2O3 nano-fluids did not show atrend toward increased mass flux at inlet subcooling of 50 �C and25 �C (Fig. 11). The effects of flow characteristics may duplicatethe effect of enhanced wettability by the deposition of nano-parti-cles with increased mass flux. Consequently, the maximum CHFenhancement can be obtained at the lowest mass flux. Jeonget al. also showed that CHF enhancement was more pronouncedat very low mass flux (100 kg/m2 s) and concluded that it is dueto an increasing wettability of the heater surface and promoting li-quid supply under bubbly or churn flow regime [15]. However themass flux condition in this experiment is very low (100–300 kg/

s (a) as-received, and after CHF experiment with (b) 0.001 vol% Al2O3 nano-fluid,

50 100 150 200 250 300 350

500

1000

1500

2000

2500 DI water : 50'C inlet subcooling DI water : 25'C inlet subcooling 0.001vol% Al

2O

3 N.F. : 50'C inlet subcooling

0.001vol% Al2O

3 N.F. : 25'C inlet subcooling

CH

F(kW

/m2)

Mass flux (kg/m2s)

500

1000

1500

2000

2500 DI water : 50'C inlet subcooling DI water : 25'C inlet subcooling 0.01vol% Al2O3 N.F. : 50'C inlet subcooling

0.01vol% Al2O3 N.F. : 25'C inlet subcooling

CH

F(kW

/m2 )

Mass flux(kg/m2s)

50 100 150 200 250 300 350

50 100 150 200 250 300 350

500

1000

1500

2000

2500 DI water : 50'C inlet subcooling DI water : 25'C inlet subcooling 0.1vol% Al2O3 : 50'C inlet subcooling

0.1vol% Al2O3 : 25'C inlet subcooling

CH

F(kW

/m2 )

Mass flux(kg/m2s)

Fig. 10. CHFs with different mass flux levels of (a) 0.001 vol% Al2O3 nano-fluid,(b) 0.01 vol% Al2O3 nano-fluid and (c) 0.1 vol% Al2O3 nano-fluid.

1.0

1.2

1.4

1.6

1.8

2.0

Enh

ance

men

t Rat

io

Mass flux(kg/m2s)

0.001vol% Al2O

3 N.F. : 50'C inlet subcooling

0.001vol% Al2O

3 N.F. : 25'C inlet subcooling

1.0

1.2

1.4

1.6

1.8

2.0

Enh

ance

men

t Rat

io

Mass flux(kg/m2s)

0.01vol% Al2O

3 N.F. : 50'C inlet subcooling

0.01vol% Al2O

3 N.F. : 25'C inlet subcooling

100 150 200 250 300

100 150 200 250 300

100 150 200 250 3001.0

1.2

1.4

1.6

1.8

2.0

CH

F e

nhan

cem

ent R

atio

Mass flux(kg/m2s)

0.1vol% Al2O

3 N.F. : 50'C inlet subcooling

0.1vol% Al2O

3 N.F. : 25'C inlet subcooling

a

b

c

Fig. 11. CHF enhancements with different mass flux levels of (a) 0.001 vol% Al2O3

nano-fluid, (b) 0.01 vol% Al2O3 nano-fluid and (c) 0.1 vol% Al2O3 nano-fluid.

1020 T.I. Kim et al. / International Journal of Heat and Mass Transfer 53 (2010) 1015–1022

m2 s), and thus further investigation under high mass flux is neces-sary in order to understand the effect of the flow and deposition onCHF.

While FE–SEM observations showed that the deposition of nano-particles increased with increasing concentration, the CHFs of theAl2O3 nano-fluids were virtually unchanged within the margin of er-ror with an increase of Al2O3 nano-particle concentration from0.001 vol% to 0.1 vol% at inlet subcooling of 50 �C and 25 �C(Fig. 12). The effect of deposition may already be saturated at a con-

centration of 0.001 vol%. Kim et al also showed that the CHF of Al2O3

nano-fluids in pool boiling is increased at very low concentrations(610�4 vol%) and is nearly unchanged above 10�3 vol% [6]. Thishypothesis can be confirmed by results of the very low concentration(610�4 vol%) CHF experiment.

The zeta potentials and pHs of Al2O3 nano-fluids were almostunchanged before and after the CHF experiments within the mar-gin of error (Table 2). Also, particle sizes of the Al2O3 nano-fluidswere nearly unchanged before and after CHF experiments. Fig. 13

600

800

1000

1200

1400

1600

1800

2000

2200

Inlet subcooling : 50'C

G=300kg/m2s

G=200kg/m2s G=100kg/m2s

CH

F(k

W/m

2 )

Concentraion of Al 2O3 nano-fluid (vol%)

0.00 0.02 0.04 0.06 0.08 0.10

0.00 0.02 0.04 0.06 0.08 0.10500

600

700

800

900

1000

1100

1200

1300

1400

1500

1600

1700

1800

1900

2000

Inlet subcooling : 25'C

G=300kg/m2s

G=200kg/m2s

G=100kg/m2s

CH

F(k

W/m

2 )

Concentraion of Al2O

3 nano-fluid (vol%)

Fig. 12. CHFs of nano-fluids with different conc. at inlet subcooling of (a) 50 �C and(b) 25 �C.

Table 2Zeta potentials and pHs before and after CHF experiments at 100 kg/m2 s.

Zeta 0.001vol%

Zeta 0.01vol%

Zeta 0.1vol%

pH 0.001vol%

pH 0.01vol%

pH 0.1vol%

Before 37 ± 2 48 ± 2 45 ± 4 6.07 5.01 5.23After 35 ± 1 53 ± 2 49 ± 1 6.15 4.97 5.07

Fig. 13. Size measurements (a) before and (b) after CHF experiments.

T.I. Kim et al. / International Journal of Heat and Mass Transfer 53 (2010) 1015–1022 1021

shows the 0.001 vol% Al2O3 nano-fluid particle sizes before andafter the CHF experiment at an inlet subcooling of 50 �C and a massflux of 200 kg/m2 s. These results verify that the Al2O3 nano-fluid inthe experimental loop was stable during the CHF experiments.

Experiments were conducted using Al2O3 nano-fluids as work-ing fluids under atmospheric pressure in flow boiling in order toevaluate possibility of enhancing critical heat flux (CHF). The sig-nificant findings can be summarized as follows:

� The zeta potentials of Al2O3 nano-fluids were slightly increasedwithin the margin of error with increasing time within a periodof 1 day. Also, the zeta potentials of the Al2O3 nano-fluids wereslightly increased with an increase of particle concentrationfrom 0.001 vol% to 0.5 vol%. From these results, it was verifiedthat the dispersion stability of Al2O3 nano-fluids during theCHF experiment was sufficient when the concentration of thenano-fluid was in a range of 0.001 vol%–0.5 vol%.

� The CHFs of Al2O3 nano-fluids were enhanced, up to about 70%,in flow boiling for all experiment conditions. This is attributed toenhanced wettability of the liquid film onto the heater surfacedue to the deposition of Al2O3 nano-particles on the inner sur-face of the test section tube. FE–SEM observations showed thedeposition of Al2O3 nano-particles on the inner surface of thetest section tube.

� The CHFs of the Al2O3 nano-fluids were increased with increas-ing mass flux at inlet subcooling of 50 �C and 25 �C. However,the CHF enhancement ratios of Al2O3 nano-fluids did not showa trend toward increased mass flux at inlet subcooling of 50 �Cand 25 �C. Further investigation under high mass flux is neces-sary in order to understand the effects of flow and depositionon the CHF.

� The CHFs of Al2O3 nano-fluids were almost unchanged withinthe margin of error with an increase of Al2O3 nano-particle con-centration from 0.001 vol% to 0.1 vol% at inlet subcooling of50 �C and 25 �C. The effect of deposition may already be satu-rated at a concentration of 0.001 vol%, and this hypothesis canbe confirmed by the results of a very low concentration(610�4 vol%) CHF experiment.

1022 T.I. Kim et al. / International Journal of Heat and Mass Transfer 53 (2010) 1015–1022

Acknowledgements

The authors would like to express their gratitude to Dr. T.H.Chun of KAERI.

References

[1] S.H. Chang, W.P. Baek, first ed., , Seoul, 1997, pp. 1–116.[2] S. Choi, Enhancing thermal conductivity of fluids with nanoparticles, in:

D.A. Siginer, H.P. Siginer, H.P. Wang (Eds.), Developments and Applica-tions of Non-Newtonian Flows, FED-Vol. 231/MD-Vol. 66, ASME, 1995, pp.99–105.

[3] M.H. Kim, J.B. Kim, H.D. Kim, Experimental studies on CHF characteristicsof nano-fluids at pool boiling, Int. J. Multiphase Flow 33 (2007) 691–706.

[4] I.C. Bang, S.H. Chang, Boiling heat transfer performance and phenomena of Al2

O3–water nano-fluids from a plain surface in a pool, Int. J. Heat Mass Transfer48 (2005) 2407–2419.

[5] Y.H. Jeong, W.J. Chang, S.H. Chang, Wettability of heated surfaces under poolboiling using surfactant solutions and nano-fluids, Int. J. Heat Mass Transfer 51(2008) 3025–3031.

[6] S.J. Kim, T. McKrell, J. Buongiorno, L.W. Hu, Experimental study of flow criticalheat flux in alumina–water, zinc-oxide–water and diamond-water nanofluids,J. Heat Transfer 131 (4) (2009).

[7] R. Wäsche, M. Naito, Experimental study on zeta potential and streamingpotential of advanced ceramic powders, Powder Technol. 123 (2002) 275–281.

[8] D. Zhu, X. Li, N. Wang, X. Wang, J. Gao, H. Li, Dispersion behavior and thermalconductivity characteristics of Al2 O3–H2O nanofluids, Curr. Appl. Phys. 9 (1)(2008) 131–139.

[9] T. Isobe, Y. Hotta, K. Watari, Dispersion of nano- and submicron-sized Al2 O3

particles by wet-jet milling method, Mater. Sci. Eng. B 148 (2008) 192–195.[10] J. Robert, Moffat, Describing the uncertainties in experimental results, Exp.

Therm. Fluid 1 (1988) 3–17.[11] D.C. Groeneveld et al., The 1995 look-up table for critical heat flux in tubes,

Nucl. Eng 163 (1996) 1–23.[12] G.F. Hewitt, D.N. Roberts, Studies of two phase flow patterns by simultaneous

X-ray and flash photography, AERE-M2159, 1969.[13] G.F. Hewitt, Burnout, in Handbook of Multiphase Systems, Hemisphere,

Washington, 1982. pp. 6.66–6.141.[14] M.S. Sarwar, Y.H. Jeong, S.H. Chang, Subcooled flow boiling CHF enhancement

with porous surface coatings, Int. J. Heat Mass Transfer 50 (2007) 3649–3657.[15] Y.H. Jeong, M.S. Sarwar, S.H. Chang, Flow boiling CHF enhancement with

surfactant solutions under atmospheric pressure, Int. J. Heat Mass Transfer 51(2008) 1913–1919.