International Journal of Heat and Mass Transfer 86 (2015) 65–71

Contents lists available at ScienceDirect

International Journal of Heat and Mass Transfer

journal homepage: www.elsevier.com/locate/ijhmt

On the formation of vapor film during quenching in de-ionized water

and elimination of film boiling during quenching in natural sea water
Shu-Han Hsu a, Yuan-Hong Ho b, Ming-Xi Ho b, Jian-Chiuan Wang b, Chin Pan a,b,c,⇑
a
Institute of Nuclear Engineering and Science, National Tsing Hua University, Hsinchu, Taiwan
b
Department of Engineering and System Science, National Tsing Hua University, Hsinchu, Taiwan
c
Low Carbon Energy Research Center, National Tsing Hua University, Hsinchu, Taiwan

a r t i c l e i n f o a b s t r a c t

Article history: The present study explores the quenching of two metallic spheres, i.e., stainless steel 304 and
Received 14 February 2015 zircaloy-702, respectively, in natural sea water, compared with that in de-ionized water. The diameters
Accepted 16 February 2015 of the spheres are 17.5 mm. In this study, the superiority of sea water compared to deionized water for
Available online 13 March 2015
nuclear power plant quenching was established. By visualization using a high-speed video camera, it was
observed that when deionized water was used for quenching, the formation of a stable vapor film around
Keywords: the metallic quenching test spheres was not immediate as traditionally thought, but occurred
Quenching
through the coalescence of nucleated bubbles at ultra-high temperatures. Conversely, sea water was
Leidenfrost temperature
Transition boiling
observed to enable much more rapid quenching through the prevention of stable film boiling due to a
Dissolved salt zeta-potential effect and quenching starting with transition boiling characterized with moderate bubble
Zeta potential nucleation sites and possible burst out of bubble clusters.
Ó 2015 Elsevier Ltd. All rights reserved.

1. Introduction establish the physical mechanism of the quenching of very hot

Quenching, which is the process of rapidly cooling a hot object
by exposing it to a much cooler liquid, is critical to the nuclear
industry, especially regarding the safety of a nuclear power plant
(NPP). In the event of a loss-of-coolant accident (LOCA), the tem-
perature of the fuel rods in the core of an NPP could rise to over
1000 °C [1]. To ameliorate the situation, the emergency core cool-
ing system of the plant pumps water into the core to cool the fuel
rods containing decay heat. However, if the temperature of the
rods is too high, the heat transfer rate may be limited by film boil-
ing, wherein a stable vapor film blankets the hot surface, resulting
in high thermal resistance. Such might have occurred during the
Fukushima Daiich NPP accident, resulting in ineffective cooling
and eventual core damage due to overheating. Sea water was sub-
sequently used to cool the very hot fuel rods and melts of the plant.
Based on the lessons learned from the accident, the so-called
Ultimate Response Guideline (URG) was recently proposed [2].
Owing to the large potential supply of sea water to NPPs located
on sea coasts, it was recommended for use in emergency cooling
during Fukushima-like accidents. It is therefore of great impor-
tance to the further enhancement of nuclear power safety to
⇑ Corresponding author at: Department of Engineering and System Science,
National Tsing Hua University, Hsinchu, Taiwan.
E-mail address: cpan@ess.nthu.edu.tw (C. Pan).
objects in a liquid, especially natural sea water.
Studies on film boiling and the Leidenfrost temperature, which
is the minimum temperature required for the sustenance of film
boiling, are very abundant in the literature. The mechanism of film
boiling heat transfer is reasonably well understood and there are
well-established models. Bromley’s model [3] for vertical surfaces
and horizontal cylinders and Berenson’s model [4] for horizontal
surfaces are two classical examples. In these two classical models,
it was assumed that heat diffused through the vapor film by con-
duction to the vapor–liquid interface providing entirely the latent
heat of evaporation there. The convective effects and liquid sub-
cooling were not considered. Koh [5] and Sparrow and Cess [6]
adopted a laminar boundary layer type approach for film boiling
on vertical surface to treat the convective effect with some sim-
plifying assumptions at the interface. Sakurai et al. [7] based on
the boundary layer theory, developed a comprehensive model for
pool film boiling from a horizontal cylinder including the con-
tribution of liquid subcooling and radiation from the heating sur-
face. Numerical solutions for the set of non-linear ordinary
differential equations derived were reported and compared well
with experimental data. Sakurai et al. also developed a simplified
model, neglecting convective effects, but considering heat diffusion
through the subcooled liquid boundary layer from the vapor liquid
interface. A very complicated equation for heat transfer coefficient
of film boiling was reported with liquid subcooling effect included.

http://dx.doi.org/10.1016/j.ijheatmasstransfer.2015.02.049
0017-9310/Ó 2015 Elsevier Ltd. All rights reserved.
66 S.-H. Hsu et al. / International Journal of Heat and Mass Transfer 86 (2015) 65–71
The film boiling heat transfer coefficient increased with increasing
liquid subcooling. Although the heat transfer mechanism of film
boiling is well understood and theory is well-established, the real
mechanism of the formation of the vapor film remains unclear. It
is commonly accepted that film boiling is characterized by the
immediate formation of a thin vapor film between the heated
surface and the bulk liquid.
The vapor film of film boiling may become unstable if the
surface temperature becomes too low, and the process would no
longer be sustainable. The minimum temperature required for film
boiling is referred to as the minimum film boiling temperature or
the Leidenfrost temperature as pointed out earlier. Although the
process is not well understood, several propositions have been
made as reviewed in Bernardin and Mudawar [8], including the
hydrodynamic instability hypothesis, metastable liquid (homoge-
neous or heterogeneous) nucleation hypothesis, thermomechani-
cal effect hypothesis, and wettability hypothesis. Such different
propositions result in significant different equations for the
Leidenfrost temperature.
Recently, the effects of heating surface properties, including
wettability, structure, and material, on film boiling and the
Leidenfrost temperature have been intensively reported in the
literature, e.g. Refs. [9–12]. Vakarelski et al. [9] demonstrated
the significant effect of surface wettability on the quenching of
20-mm-diameter stainless steel spheres. They modified the surfaces
of the spheres to be superhydrophilic (contact angle < 10°), hydro-
philic (contact angle < 30°), hydrophobic (contact angle  100°),
and superhydrophobic (contact angle > 160°). They demonstrated
that the temperature dropped rapidly to the pool temperature
during the cooling of the superhydrophilic sphere. The cooling of
the hydrophilic and hydrophobic spheres began with film boiling,
and the Leidenfrost temperatures were about 420 °C and 210 °C,
respectively. In the case of the superhydrophobic sphere, film boil-
ing was maintained throughout the cooling down to the pool tem-
perature, without the collapse of the vapor layer. Lee et al. [10]
investigated the effect of surface condition induced by 2 h-oxidation
procedure and repeated quenching of stainless steel and copper
rodlet in water. They found that the surface morphology and rough-
ness of copper rodlet were significantly changed by 2 h-oxidation.
They also reported that the quenching duration time was signifi-
cantly shorten due to unique feather-like structure on the copper
surface. However, there was no remarkable influence on quenching
performance of stainless steel rodlet resulting from 2 h-oxidation.
Thus, they concluded that the quench cooling might be improved
and controlled by a surface structure design. Kruse et al. [11] demon-
strated the shift of the Leidenfrost temperature from stainless steel
surface with multiscale macro/nano structures by the evaporation of
a single water droplet on a heated surface. They found that the
Leidenfrost temperature of surface with multiscale macro/nano
structures was 175 °C higher than that of the polished surface.
This shift could be attributed to the reductions in contact angle
and substantial capillary wicking resulting from nanoporosity dur-
ing intermittent contacts by the droplet. Sher et al. [12] investigated
that the quenching of spheres with different sizes made by
aluminum, copper or steel in subcooled liquid. They reported the
dependence of the quenching temperature on the liquid subcooling
and material of spheres and increase in the Leidenfrost temperature
with the size of spheres.
The effects of coolant additives including nanoparticles and dis-
solved salt on film boiling and the Leidenfrost temperature have
also been conducted intensively in the literature. For example,
Kim et al. [13] investigated the quenching of steel and zircaloy
spheres using pure water and water-based nanofluids containing
low concentrations of alumina, silica, and diamond nanoparticles.
They found that the first run quenching in a nanofluid was nearly
the same as that in pure water. However, subsequent quenching of
the same spheres was significantly accelerated in water containing
alumina or silica nanoparticles. The critical heat flux (CHF) and
minimum film boiling temperature (MFBT) (i.e., the Leidenfrost
temperature) were significantly increased. The minimum film boil-
ing temperature determined from the boiling curve obtained by
their lumped capacitance model was as high as 400 °C, which is
about 150 °C higher than that for quenching in pure water. Kim
et al. attributed the elevated CHF and MFBT and the accelerated
quenching effect to the deposition of nanoparticles on the surface
of the sphere, which increased the surface roughness and wettabil-
ity. This apparently destabilized the vapor film as proposed by Park
el al. [14] and Takata el al. [15]. Kim et al. also reported that
subsequent quenching in water containing diamond nanoparticles
was very much the same as that in pure water, although the
deposit of the diamond nanoparticles significantly increased the
surface roughness and wettability. They therefore suspected some
other non-negligible effects. Chun et al. [16] conducted quench
experiments of the bare Pt wires and Si/SiC nanoparticle-coated
Pt wires in water and nanofluids containing Si and SiC, respec-
tively. They found that for the bare Pt wires there was no signifi-
cant influence on boiling curve between water, Si and SiC
nanofluids. However, a rapid cooling without stable film boiling
for nanoparticle-coated Pt wires was demonstrated. They also indi-
cated that this phenomenon depended strongly on the macrostruc-
ture and the thickness of nanoparticle deposition layer. Ciloglu and
Bolukbasi [17] investigated the quenching behavior of spheres in
water-based nanofluids containing various volume fractions of
Al2O3, SiO2, TiO2 and CuO nanoparticles. The heat fluxes in film
boiling regime were found to be similar to that of pure water in
the first run for all nanofluids. However, a porous layer depending
on volume fraction and type of nanofluids was observed during
repeated tests. An enhancement of CHF and increase in the
Leidenfrost temperature, which was attributed to the porous layer
resulting from the deposition of nanoparticles, was reported.
Recently, Lee et al. [18] conducted a quenching experiment to
investigate the cooling performance of 35o/oo sea salt solution in
reflood heat transfer through a long vertical tube. They found that
the cooling performance of the sea salt solution was nearly tenfold
that of freshwater, and the CHF was enhanced by 9.7%. Based on
microscopic examination, they concluded that the enhancements
were not due to the superior wettability produced by the sea salt
deposition, but the top quenching caused by the earlier con-
densation of vapor during the reflood of the sea salt solution.
Conversely, Huang and Carey [19] found that the dissolved salt
increased the Leidenfrost temperature owing to the evaporation
of a liquid drop from a surface heated by the formation of nucle-
ation cavities. They reported that the Leidenfrost temperature
was about 10–33 °C higher than that for distilled water, and attrib-
uted this to the suppression of bubble coalescence by the dissolved
salt, salt deposition during the initial contact of the deposition, and
earlier collapse of the vapor film due to increasing salt concentra-
tion at the liquid–vapor interface. Indeed, recently, it has been
extensively studied in the literature that dissolved salt may sup-
press bubble coalescence possibly due to colloidal forces, gas solu-
bility, Gibbs–Marangoni effect, surface rheology and ion-specific
[20]. However, the inhibiting effect of salts on bubble coalescence
is not yet well understood [20]. Abdalrahman et al. [21] investi-
gated the spray cooling of an aluminum alloy using dissolved sin-
gle salt solution and salt mixtures. In their work, it was proposed to
use electrical conductivity instead of concentration to explain the
dissolved salt effect on the boiling phenomena. They indicated that
the Leidenfrost temperature, the temperature corresponding to the
departure form nucleate boiling, maximum heat flux and heat
transfer coefficient all increased with increasing electrical conduc-
tivity. Furthermore, the effect resulted from salt mixtures was
found to be stronger than that of single salt solution.
 S.-H. Hsu et al. / International Journal of Heat and Mass Transfer 86 (2015) 65–71
The objective of the present study was to investigate and
compare the quenching of two types of metal spheres, namely,
stainless steel 304 (S30400) and zircaloy 702 (R60702) spheres,
in sea water, compared with that de-ionized water. It should
be noted that the physical phenomenon that prevail in sea
water is significantly different from those in deionized water.
Furthermore, regarding the application of sea water on to NPP
quenching, there is a variety of cooling water sources that could
be used, such as clean water, raw water, well water, creek water,
and even seawater, provided the source is sufficiently close to
the plant. Water can be obtained from such sources and pumped
into the reactor core in the event of an emergency such as the
Fukushima accident. It is especially relevant to investigate quench-
ing using sea water because many nuclear power plants are located
on sea coasts and use sea water for condenser cooling. It is also of
fundamental interest to understand the mechanism of film boiling
and its destabilization of deionized or sea water.
2. Method
The experimental setup of the present study consists of a test
sphere, pneumatic cylinder, radiant furnace, quench pool,
PC-based data acquisition system (YOKOGAWA/model: MX-100),
and high-speed camera (FASTCAM-Ultima APX12 or IDT
MotionPro X4). Two types of metallic spheres, namely, stainless
steel 304 (S30400) and zircaloy 702 (R60702) spheres, were used
as the objects to be quenched. The spheres were of diameter
17.5 mm and well polished. A 16 mm deep hole was drilled into
the bottom of each sphere for the insertion of a K-type
thermocouple. The thermocouple was attached to the sphere by
graphite glue (AREMCO #669) to ensure good thermal contact.
Because of the relatively high thermal conductivity of the metals
used for the test spheres, it was assumed that the measured tem-
perature was approximately equal to that of the surface. A
650 mm long stainless steel tube with a thermocouple inside
was connected to the test sphere. During the quench experiment,
the sphere with the tube was driven by a 250-mm-stroke pneu-
matic cylinder into the quench pool at a speed of 0.67 m/s. The
pool consisted of a vessel measuring 195 mm long, 195 mm
broad, and 150 mm deep, and was filled with sea or deionized
water to a depth of approximately 110 mm. The pool tempera-
ture during a quenching experiment was monitored by four
T-type thermocouples positioned at the corners. To ensure
approximately the same sphere surface condition, each sample
sphere was used only once.
The sea water was obtained from the Taiwan Strait near the
coast of Hsinchu City, Taiwan. The constitution of the sea water
was not specifically determined but considered to be that of
standard sea water from the Pacific Ocean. The weight percent of
all the ions in the sea water, namely the salinity, was 3.5% or
35o/oo. As could be expected, chloride ion was predominant and
constituted 55.02%, and sodium, sulphate, magnesium, calcium,
and potassium ions respectively constituted 30.66%, 7.71%, 3.65%,
1.17%, and 1.13% [22]. These six most predominant ions accounted
for 99.34% by weight of the constituent ions.
The test sphere was first heated in the 2200 W (220 V, 10 A)
radiant furnace, with the temperature increase monitored by the
data acquisition system. The heating was stopped at 1000 °C and
the pneumatic cylinder was used to drive the sphere into the pool,
at which time filming by the high-speed camera. After completion
of the cylinder stroke, the LED bulb in front of the camera was
turned on to synchronize the filming and sphere temperature mea-
surement for development of the quenching curve.
The temperatures of the sphere and water were recorded at a
frequency of 10 Hz, and the variation of the sphere temperature
67
with time (i.e., the quench curve) was acquired for the cooling pro-
cess. The experiment and recording ended when the sphere
attained thermal equilibrium with the pool water. The process
was repeated five times for each type of metal sphere and quench-
ing water (deionized or sea) using a new sample sphere each time.
The water in the pool was initially at about 33 °C, thereby
representing high subcooling at 1 atm. To eliminate dissolved
gases from the deionized water, it was first boiled for 3 h and sub-
sequently cooled to room temperature for the quench test. The
water was replaced for each cooling procedure.
3. Results and discussion
Fig. 1(a) and (b) shows the quench curves for the stainless steel
and zircaloy spheres, respectively, in deionized water and sea
water. The figures clearly reveal the occurrence of film boiling
and its transition to nucleate boiling in deionized water. The entire
quenching process lasts for about 40 s, out of which film boiling
lasts for about 16 and 12 s for the stainless steel and zircaloy
spheres, respectively. Conversely, in sea water, film boiling seems
to be absent and nucleate boiling occurs after brief transition boil-
ing. For both sphere types, the sphere temperature drops very
quickly, falling from 1000 °C to about 100 °C for sea water in about
13 and 7 s for steel and zircaloy, respectively.
The key to understanding the significant difference between
quenching in deionized water and sea water may be in the early
stages of the quenching. Fig. 2(a) and (b) displays the beginning
of quenching in deionized water after immersion of the sphere in
the pool for 46 and 60 ms, respectively. The very high temperature
of the sphere (about 1000 °C) causes very active bubble nucleation
all over its surface. The nucleated bubbles are small because the
pool water is highly subcooled. It is also noteworthy that the bub-
ble nucleation site density is relatively small. The small bubbles
eventually coalescence to form a vapor blanket around the sphere
with a stable vapor–liquid interface. The interface is initially unsta-
ble due to the transient contact at sphere surface, but stabilizes
within a short period. As indicated by arrows in Fig. 2(c) and (d),
the transient contact happened during the bubbles coalescing pro-
cess. After about 0.22 s, the direct contact between the deionized
water and the metal surface rarely can be observed (Fig. 2(e))
and then a stable film boiling was formed as shown in Fig. 2(f).
The pictures clearly reveal that the vapor film is not instantly
formed when the sphere enters the water, but after rapid bubble
nucleation and subsequent coalescence of the bubbles. The present
study thus demonstrated bubble nucleation at ultra-high tempera-
tures and the formation of a stable vapor film through the coales-
cence of the bubbles. This is in contrast to the traditional view of
immediate formation of a vapor layer between a liquid and a sur-
face at a temperature above the Leidenfrost temperature.
The interfacial temperature between the water and the sphere
during the transient contact may be evaluated using the following
equation [23]:
pffiffiffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffi
kqC p s  T s þ kqC p f  T f
TI ¼ pffiffiffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffi ð1Þ
kqC p s þ kqC p f
where the subscripts I, s, and f represent interfacial, sphere, and
quenching fluid, respectively. Using the above equation and the
sphere temperature of 1000 °C and water temperature of 33 °C,
the interfacial temperature was determined to be 996 °C and
993 °C for the stainless steel and zircaloy spheres, respectively.
These temperatures are much higher than those of homogeneous
and heterogeneous bubble nucleation on a perfectly smooth surface
(300 °C for water at 1 atm), and bubble nucleation at such high tem-
peratures is unusual. The relatively low bubble nucleation density
at such high surface temperatures is also unusual.
68 S.-H. Hsu et al. / International Journal of Heat and Mass Transfer 86 (2015) 65–71

Fig. 1. Cooling curves of SS304 (a), zircaloy 702 (b), in DI water and sea water, respectively.

Regarding the early quenching of both types of spheres in sea
water, similar to the case of deionized water, direct contact
between the sea water and the sphere surface at a high tempera-
ture (1000 °C) resulted in rapid bubble nucleation (Fig. 3(a) and
(b)). The bubble nucleation occurred over the entire surface of
the sphere when it was fully immersed in the sea water.
However, unlike the formation of a stable vapor film through the
coalescence of bubbles in deionized water, there seems to be no
vapor film formed in sea water. The transient contact between
sea water and the sphere surface results in more fierce bubble
nucleations at various locations pointed out by red arrows and,
at certain locations, bubble clusters may be formed and burst
out, as marked by dash-circles in Fig. 3(c–e). Significantly, after
the same period of time of 0.22 s as quench in deionized water,
unlike a clear vapor film shown in Fig. 2(e), the fierce direct contact
between sea water and sphere is persistent with isolated bubble
 S.-H. Hsu et al. / International Journal of Heat and Mass Transfer 86 (2015) 65–71 69

Fig. 2. The snap shot of stainless steel 304 sphere quenching in DI water for selected time of run 4 with a frame rate of 15,000 frame/s. (a and b) Active bubble nucleation after
sphere immersion. (c and d) Transient contact of solid–liquid with bubble nucleations, some of them are marked with an arrow. (e) Transient contact rarely can be observed.
(f) Formation of a stable film boiling.

Fig. 3. The snap shot of stainless steel 304 sphere quenching in sea water for selected time of run 4 with a frame rate of 15,000 frame/s. (a and b) Active bubble nucleation
after sphere immersion. (c and d) More fierce transient contact of solid–liquid with bubble nucleations (some marked with an arrow) and burst of bubble clusters (marked
with a dash circle). (e) Lasting transient contact still exist. (f) Collapse from the top of stainless sphere.

nucleation of bubble clusters bursted out (Fig. 3(e)). This demon-
strates the more lasting, direct liquid–solid contact between the
sea water and the high temperature sphere surface. This region
of moderate bubble nucleation sites and also possible burst out
of bubble clusters is considered to be transition boiling, as also
demonstrated later in Supplementary material for the boiling
curve derived from the quench curve. About 0.43 s after sphere
immersion, very active nucleate boiling mode began on top of
the stainless steel sphere (but began on bottom of the zircaloy
one) and spread over the entire sphere as shown in Fig. 3(f) (A
70 S.-H. Hsu et al. / International Journal of Heat and Mass Transfer 86 (2015) 65–71

detail comparison of quenching for SS304 spheres in deionized

water and sea water, respectively, is provided in Supplementary
video).
To determine the cause of the absence of stable vapor film dur-
ing metal sphere quenching in sea water, the contact angle
between each type of sphere and the quench liquid was measured
before and after heating in the oven and after quenching (see
Table 1). It was observed that the side and bottom surfaces of each
sphere had somewhat different contact angles. Table 1 gives the
mean value of eight measurements, four of which were taken at
the side, and four at the bottom. As can be observed, the steel
sphere was generally more hydrophilic in both deionized water
and sea water than the zircaloy sphere, and both types of spheres
were somewhat more hydrophilic in sea water than in deionized
water. Both types of spheres became more hydrophilic after oven
heating in open air to 1000 °C, possibly due to surface oxidation.
The quenching did not significantly change the contact angle, at
least not sufficiently enough to explain why quenching in sea
water is very much better than in deionized water. It has been pre-
viously proposed [9,13] that a superhydrophilic surface might sig-
nificantly enhance quenching.
The absence of stable film boiling and the lasting transition
boiling region and the consequent rapid quenching of both spheres
in sea water may be attributed to the effect of suppression of bub-
ble coalescence by the dissolved salt. As pointed out in the litera-
ture review, several propositions on the inhibiting effect of
bubble coalescence have been proposed including colloidal forces,
gas solubility, Gibbs–Marangoni effect, surface rheology and ion-
specific. However, the mechanism of inhibiting effect remains
unclear [20]. Besides, another possible effect on the inhibiting
effect on bubble coalescence in sea water may be the zeta potential
of electrical double layer (EDL) in ultra-high temperature and
charged liquid–solid interfaces. Fig. 4 illustrates a schematic dia-
gram of zeta-potential effect on quenching in dissolved salt solu-
tion. The zeta-potential effect created a charge distribution, most
likely negative, over the sphere surface, resulting in the dis-
tribution of ions with opposite charges, most likely including posi-
tively charged sodium ions, near the liquid–solid interface [24].
Literatures [24,25] demonstrated that the zeta potential is directly
proportional to temperature, increases approximately 0.83%–1.75%
per °C over a span of 25–90 °C. However, the entire quenching pro-
cess in sea water, the temperature difference between liquid solu-
tion with ultra-high temperature of charged solid surface, with
temperature above or near 900 °C, resulting in the zeta potential
sufficiently strong to increase the electrostatic force between the
solid and liquid layer near the surface. Thus, the ions in the
Fig. 4. The schematic diagram of zeta-potential effect of quenching in dissolved salt
solution.
seawater might be attracted to the heating surface firmly and pre-
vent the coalescence of nucleated bubbles due to stabilizing the
liquid film between bubbles from forming a stable vapor film.
Indeed, the bubbles are relatively sparsely distributed on the
sphere surface during the initial stage of quenching, as shown ear-
lier in Fig. 3. At this earlier stage of quenching, the distance among
bubbles may be too far to have significant effect of other mecha-
nisms among bubbles proposed in the literature.
Furthermore, the cooling curve of different concentrations of
MgSO4 reported by Abdalrahman et al. [21] for the spray cooling
clearly demonstrates the shift toward left with increasing concen-
tration. It is well-known that the zeta potential increases with
increasing concentration of the solute in the solution. The zeta-po-
tential effect together with other mechanisms proposed in the
literature are thus sufficiently strong to prevent the bubble coales-
cence and formation of stable vapor film when the required con-
centration of the solution is used.
The zeta-potential effect may also adequately explain the rapid
quenching of a superhydrophilic surface by water as reported by
Vakarelski et al. [9] A superhydrophilic surface is generally
regarded as being charged and will therefore produce close contact
with a liquid of opposite charge, resulting in a near-zero contact
angle. Conversely, although the effect also enhances the
hydrophilicity of a surface with diamond nanoparticle deposits
[13], the enhancement may not be sufficiently strong to prevent
the formation of the vapor film considering that diamond is actu-
ally a form of carbon. The film boiling thus remains as usual and
rapid quenching is not achieved [13].

Table 1
Contact angle (degree) between spheres and DI water or sea water drops at room temperature (SS304 stands
for stainless steel sphere; Zr 702 stands for Zircaloy 702 sphere; r is the standard error of the mean of contact
angle).
 S.-H. Hsu et al. / International Journal of Heat and Mass Transfer 86 (2015) 65–71
4. Conclusions
Rapid quenching of stainless steel 304 and zircaloy 702 spheres
at temperatures of about 1000 °C in sea water and was investigated
and adequately explained by the zeta-potential effect created
between the sea water and sphere as well as other forces between
bubbles in water with dissolved salts. Such an effect prevents the
stable film boiling and quenching starts from transition boiling
even the surface is nearly 1000 °C. The rapid quenching of the
hot metal spheres in sea water demonstrated the benefits of NPP
cooling using sea water, although the potential rapid corrosion of
the metals, especially stainless steel, is also of significant concern.
The present study also confirmed the occurrence of rapid bubble
nucleation for a moderate nucleation site density at a temperature
close to 1000 °C, which is much higher than the homogeneous or
heterogeneous temperature of bubble nucleation predicted by
classical nucleation theories. It was further observed that the vapor
layer produced by film boiling was not immediately formed by
contact of the liquid with the hot surface but through the coales-
cence of bubbles nucleated at very high temperature.
Conflict of interest
None declared.
Acknowledgement
This work was supported by the Ministry of Science and
Technology of Taiwan under Contracts No.: NSC 102-3113-P-007-
014.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.ijheatmasstrans-
fer.2015.02.049.
References
[1] W.J. Green, K.R. Lawther, A study of the sensitivity of LOCA heat transfer
analysis for a water-cooled reactor system, Nucl. Eng. Des. 47 (1978) 87–99.
[2] K.S. Liang, S.C. Chiang, Y.F. Hsu, H.J. Young, B.S. Pei, L.C. Wang, The ultimate
emergency measures to secure a NPP under an accidental condition with no
designed power or water supply, Nucl. Eng. Des. 253 (2012) 259–268.
[3] L.A. Bromley, Heat transfer in stable film boiling, Chem. Eng. Prog. 46 (1950)
221–226.
71
[4] P.J. Berenson, Film boiling heat transfer from a horizontal surface, J. Heat
Transfer 83 (1961) 351–356.
[5] J.C.Y. Koh, Analysis of film boiling on vertical surface, J. Heat Transfer 84 (1962)
55–62.
[6] E.M. Sparrow, R.D. Cess, The effect of subcooled liquid on laminar film boiling,
J. Heat Transfer 84 (1962) 149–156.
[7] A. Sakurai, M. Shiotsu, K. Hata, A general correlation for pool boiling heat
transfer from a horizontal cylinder to subcooled liquid, Part 1: A theoretical
pool film boiling heat transfer model including radiation contribution and its
analytical solution, J. Heat Transfer 112 (1990) 430–440.
[8] J.D. Bernardin, J.D. Mudawar, The Leidenfrost point: experimental study and
assessment of existing models, J. Heat Transfer 121 (1999) 894–903.
[9] I.U. Vakarelski, N.A. Patankar, J.O. Marston, D.Y.C. Chan, S.T. Thoroddsen,
Stabilization of Leidenfrost vapour layer by textured superhydrophobic
surfaces, Nature 489 (2012) 274–277.
[10] C.Y. Lee, T.H. Chun, W.K. In, Effect of change in surface condition induced by
oxidation on transient pool boiling heat transfer of vertical stainless steel and
copper rodlets, Int. J. Heat Mass Transfer 79 (2014) 397–407.
[11] C. Kruse, T. Anderson, C. Wilson, C. Zuhlke, D. Alexander, G. Gogos, S. Ndao,
Extraordinary shifts of the Leidenfrost temperature from multiscale
micro/nanostructured surfaces, Langmuir 29 (2013) 9798–9806.
[12] I. Sher, R. Harari, R. Reshef, E. Sher, Film boiling collapse in solid spheres
immersed in a subcooled liquid, Appl. Therm. Eng. 36 (2012) 219–226.
[13] H. Kim, G. DeWitt, T. McKrell, J. Buongiorno, L. Hu, On the quenching of steel
and zircaloy spheres in water-based nanofluids with alumina, silica and
diamond nanoparticles, Int. J. Multiphase Flow 35 (2009) 427–438.
[14] H.D. Park, D. Shiferaw, B.R. Sehgal, D.K. Kim, M. Muhammed, Film boiling heat
transfer on a high temperature sphere in nanofluid, in: Proc. HT-FED04-56471,
2004. pp. 469–476.
[15] Y. Takata, S. Hidaka, J.M. Cao, T. Nakamura, H. Yamamoto, M. Masuda, T. Ito,
Effect of surface wettability on boiling and evaporation, Energy 30 (2005) 209–
220.
[16] S.Y. Chun, I.C. Bang, Y.J. Choo, C.H. Song, Heat transfer characteristics of Si and
SiC nanofluids during a rapid quenching and nanoparticles deposition effects,
Int. J. Heat Mass Transfer 54 (2011) 1217–1223.
[17] D. Ciloglu, A. Bolukbasi, The quenching behavior of aqueous nanofluids around
rods with high temperature, Nucl. Eng. Des. 241 (2011) 2519–2527.
[18] S.W. Lee, S.M. Kim, S.D. Park, I.C. Bang, Study on the cooling performance of sea
salt solution during reflood heat transfer in a long vertical tube, Int. J. Heat
Mass Transfer 60 (2013) 105–113.
[19] C.K. Huang, V.P. Carey, The effects of dissolved salt on the Leidenfrost
transition, Int. J. Heat Mass Transfer 50 (2007) 269–282.
[20] M. Firouzi, T. Howes, A.V. Nguyen, A quantitative review of the transition salt
concentration for inhibiting bubble coalescence, Adv. Colloid Interface Sci.
(2014), http://dx.doi.org/10.1016/j.cis.2014.07.005.
[21] K.H.M. Abdalrahman, Sabariman, E. Specht, Influence of salt mixture on the
heat transfer during spray cooling of hot metals, Int. J. Heat Mass Transfer 78
(2014) 76–83.
[22] C.T Chen, Chemical Oceanography, National Translation and Compilation
Center, Taipei, 1994. 111 (in Chinese).
[23] F.P. Incropera, D.P. Dewitt, T.L. Bergman, A.S. Lavine, Principles of Heat and
Mass Transfer, seventh ed., John Wiley & Sons, Singapore, 2013.
[24] B.J. Kirby, E.F. Hasselbrink, Zeta potential of microfluidic substrates: 1 Theory,
experimental techniques and effects on separations, Electrophoresis 25 (2004)
187–202.
[25] R. Vendittli, X. Xuan, D. Li, Experimental characterization of the temperature
dependence of zeta potential and its effect on electroosmotic flow velocity in
microchannels, Microfluid. Nanofluid. 2 (2006) 493–499.