Visualization of the boiling phenomenon inside a heat pipe using neutron radiography.

Experimental Thermal and Fluid Science
AIMS AND SCOPE Experimental Thermal and Fluid Science provides a forum for original, previously unpublished research emphasizing experimental work that
enhances understanding of heat transfer, thermodynamics, and fluid mechanics and their applications. Also covered in the abstract and citation database ScopusÒ.
Full text available on ScienceDirectÒ
ETFS also covers research results in combined heat and mass transfer, multiphase flow, combustion, radiative transfer, porous media, cryogenics, turbulence,
contact resistance, microscale and nanoscale thermal fluid systems and thermophysical property measurements and techniques. Papers related to the
development of new experimental methods, visualization techniques, and instrumentation are an important part of this Journal.
Experience gained and lessons learned in building test facilities and in measuring and reducing test data are important aspects of any experimental work. Some of
these papers include micro-electromechanical systems; medical and biological systems; and thermal and flow control in both the internal and external environment.
Authors are encouraged to report this experience and to summarize the original data.
Journal Homepage Description for Experimental Thermal and Fluid Science
Experimental Thermal and Fluid Science provides a forum for research emphasizing experimental work that enhances basic understanding of heat transfer,
thermodynamics and fluid mechanics, and their applications. In addition to the principal areas of research, the journal covers research results in related fields,
including combined heat and mass transfer, micro and nanoscale systems, multiphase flow, combustion, radiative transfer, porous media, cryogenics,
turbulence, contact resistance, and thermophysical property measurements and techniques. Archival review papers, short communications, invited papers,
letters to the Editor, discussions of previously published papers, and book reviews are regular features of the journal, in addition to full-length articles.

Editors-in-Chief
N. Selc¸uk
Department of Chemical Engineering,
Middle East Technical University,

06531 Ankara, Turkey
Tel.: +90-312-2102603
Fax: +90-312-2102600
E-mail: selcuk@metu.edu.tr

G.P. Celata
ENEA,
Institute of Thermal-Fluid Dynamics,
Via Anguillarese 301,
00123 Santa Maria di Galeria, Rome, Italy
Tel.: +39-06-3048-3905
Fax: +39-06-3048-3026
E-mail: celata@enea.it

Editors
C. Colin
Institut de Me´canique des
Fluides de Toulouse
UMR 5502 CNRS-INP-UPS 2,
Alle´e Camille Soula, 31 400 Toulouse

France
Tel.: 33 (0)5 34 32 28 25
Fax: 33 (0)5 34 32 29 91
E-mail: colin@imft.fr
M.H. Kim
Department of Mechanical Engineering
Pohang University of Science
and Technology
San 31, Hyoja-Dong, Nam-Gu,
Pohang, Kyeongbuk, 790-784,
South Korea
E-mail: mhkim@postech.ac.kr
T.M. Liou
National Tsing Hua University
Taiwan, ROC
Tel.: +886-3-5742607
Fax: +886-3-5729716
E-mail: tmliou@pme.nthu.edu.tw

B.J. McKeon

Professor of Aeronautics
Graduate Aerospace Laboratories
California Institute of Technology
1200 E. California Blvd, MC 301-46
Pasadena, CA 91125, USA
E-mail: mckeon@caltech.edu
D.V. Pence
Professor of Mechanical Engineering
Oregon State University, 204 Rogers Hall
Corvallis, OR 97331-6001, USA
Tel.: 541-737-7018
Fax: 541-737-2600
E-mail: deborah.pence@oregonstate.edu
G. Ribatski
Department of Mechanical Engineering
S~ao Carlos School of Engineering (EESC)
University of S~ao Paulo
Av. Trabalhador S~ao-carlense,
400, Pq Arnold Schimidt
13566-590 S~ao Carlos - SP, Brazil

Tel.: +55 16 3373 9415
Fax: +55 16 3373 9402
E-mail: ribatski@sc.usp.br

S.K. Saha
Mechanical Engineering Department
Bengal Engineering and Science University
Shibpur
Howrah 711 103,
West Bengal, India
Tel.: +919830493430
Fax: +913326682916
E-mail: sujoy_k_saha@hotmail.com

S. Wongwises
Department of Mechanical Engineering
King Mongkut’s University of Technology
Thonburi
126 Bangmod, Tongkru, Bangkok 10140,
Thailand

Tel.: +662-4709115
Fax: +662-4709111
E-mail: somchai.won@kmutt.ac.th

I. Satoh
Japan Tokyo Institute of Technology
Tokyo, Japan
Tel.: +81-3-5734-3238
Fax: +81-3-5734-2945 (Office)/
+81-3-5734-3917 (Department)
E-mail: satohi@mep.titech.ac.jp
C.W.M. van der Geld
Technische Universiteit Eindhoven
Eindhoven, Netherlands
Tel.: +31-40-247-2923
Fax: +31-40-247-5399
E-mail: C.W.M.v.d.Geld@wtb.tue.nl

Editorial Advisory Board
S. Aleekseenko, Russia

N. Brauner, Israel
J.M. Corberan, Spain
P. Di Marco, Italy
I. Golobic, Slovenia
S. Kakac, Turkey
T. Karayiannis, UK
M. Kawaji, Canada
I. Kennedy, USA

K.T. Kiger, USA
J-H. Kim, USA
S.Y. Ko, China
N.W.M. Ko, Hong Kong
I. Marusic, Australia
G. Nathan, Australia
R.E. Nigmatulin, Russia
T. Okawa, Japan
M. Ozawa, Japan

B. Palm, Sweden

D. Pence, USA
L. Smits, USA
J. Soria, Australia
P. Stephan, Germany
S.P. Sukhatme, India
A. van Steenhoven, The Netherlands
K.T. Yang, USA

Founding Editors: R.K. Shah, E.N. Ganic´
Publication information: Experimental Thermal and Fluid Science (ISSN 0894-1777). For 2015, Volumes 60 to 69 are scheduled for publication. Subscription prices
are available upon request from the Publisher or from the Elsevier Customer Service Department nearest you or from this journal’s website (http://www.elsevier.
com/locate/etfs). Further information is available on this journal and other Elsevier products through Elsevier’s website (http://www.elsevier.com). Subscriptions are
accepted on a prepaid basis only and are entered on a calendar year basis. Issues are sent by standard mail (surface within Europe, air delivery outside Europe).
Priority rates are available upon request. Claims for missing issues should be made within six months of the date of dispatch.

Experimental Thermal and Fluid Science 66 (2015) 13–27

Contents lists available at ScienceDirect

Experimental Thermal and Fluid Science

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

Visualization of the boiling phenomenon inside a heat pipe using
neutron radiography
Nandy Putra a,⇑, Ranggi Sahmura Ramadhan a, Wayan Nata Septiadi a,b, Sutiarso c
a

Heat Pipe Technology Research Cluster, Heat Transfer Laboratory, Department of Mechanical Engineering, University of Indonesia, Kampus UI-Depok, 16424, Indonesia
Departement of Mechanical Engineering, University of Udayana, Kampus Bukit Jimbaran, Bali, Indonesia
c
Centre of Science and Technology of Advanced Materials, National Nuclear Energy Agency of Indonesia (BATAN), Indonesia
b

a r t i c l e

i n f o

Article history:
Received 13 August 2014
Received in revised form 28 February 2015

Accepted 28 February 2015
Available online 17 March 2015
Keywords:
Heat pipe
Visualization
Neutron radiography

a b s t r a c t
Heat pipes are effective heat exchangers that have a wide range of applications because of their ability to
passively transfer large amounts of heat. Research into heat pipe technology has dramatically increased
over the last decade and, more recently, has incorporated the use of visualization to help researchers gain
a better understanding of the boiling phenomenon and heat transfer occurring inside a heat pipe.
Neutron radiography is one method of visualization suitable for use in heat pipe investigations due to
unique attenuation characteristics of neutrons attaching to various materials. In this study, an aluminum-based heat pipe was tested using working fluid filling ratios from a 10% to 90% capacity.
Visualization using neutron radiography was conducted at a neutron radiography facility, RN1, under
the supervision of the Centre of Science and Technology of Advanced Materials (PSTBM), National
Nuclear Energy Agency of Indonesia (BATAN). Using temperature and pressure sensors, this study
revealed that the optimum value of working fluid filling ratios directly correlates to the pressure inside
a heat pipe and the size of vapor space available. The neutron radiography facility maintains high neutron
flux at 106–107 n/cm2 s; high quality images were captured utilizing this radiography visualization technology. The captured images demonstrate that the boiling phenomenon inside a pressure-reduced heat

pipe varies when compared with the boiling phenomenon at atmospheric pressure. The visualization
result also shows the importance of wick structure in pumping return condensate from the condenser
to the evaporator.
Ó 2015 Elsevier Inc. All rights reserved.

1. Introduction
Heat pipes are passive heat exchangers operating on the twophase principle. The passive description refers to lack of additional
energy needed during operation; two-phase indicates that the heat
pipe uses the phase change phenomenon of working fluids as the
main mode of heat transfer [1,2]. A heat pipe may have circular
or rectangular cross sections and generally contains a vacuum container with a wick structure and a working fluid. When heat is
applied to the evaporator, the working fluid evaporates and rushes
to fill the evacuated space. As the vapor enters, it contacts the wall
of the lower temperature heat pipe. The vapor then condenses and
releases latent heat. This condensate naturally flows back to the
evaporator, either by gravity or capillarity action induced by the
wick structure. The high amount of latent heat needed during

⇑ Corresponding author. Tel.: +62 812 8781 557.
E-mail address: nandyputra@eng.ui.ac.id (N. Putra).

http://dx.doi.org/10.1016/j.expthermflusci.2015.02.026
0894-1777/Ó 2015 Elsevier Inc. All rights reserved.

the phase change process enables the heat pipe to have higher
effective thermal conductivity [3,4].
Heat pipes used for various cooling systems have been extensively tested for many applications such as central processing units
(CPUs) hard disks, light emitting diodes (LEDs) lamps, avionics and
various medical devices including those used in cryosurgery and
vaccine transport mechanisms [5–11]. Heat pipe components for
each application were investigated for optimum performance.
Wick structures were made of biomaterials, bi-porous structures,
sintered powder, grooved assemblies and metal foam [12–16].
Working fluids such as nanofluids, methanol, acetone, propylene
glycol and refrigerant have also being tested [17–20].
Another parameter affecting heat pipe performance is the working fluid filling ratio. This ratio is the volumetric measure of fluid
injected into the heat pipe compared with the volume of evacuated
space. Naphon et al. [21] tested vapor chambers on CPU cooling
systems with various filling ratios and found that the optimum
ratio of fluid to volume is 20%. Any ratio lower than 20% leads to
the dry-out phenomenon; higher ratios, however, cause the liquid

14

N. Putra et al. / Experimental Thermal and Fluid Science 66 (2015) 13–27

Fig. 1. (a) Picture of the sintered aluminum powder wick structure and (b) SEM image of the wick.

layer to hamper heat transfer at the heat input region. Sukchana
and Jaiboonma [22] investigated the effect of the filling ratio compared with thermal efficiency of a 10 cm, R-134a charged heat
pipe. The investigation also showed the optimum value of filling
ratio relates to the saturation pressure increment and temperature
inside the heat pipe. Lips et al. [23] studied the effect of the filling
ratio of a flat heat pipe and found that this ratio affects the thermal
resistance of a heat pipe. Small filling ratios lead to dry-out at the
evaporator while large filling ratios lead to flooding at the condenser. Hussein et al. [24] investigated wickless heat pipes utilizing various filling ratios and found that a higher filling ratio
provided higher thermal capacity resulting in lower temperature
and lower pressure at the same heat flux. Ong and Alahi [25] tested
a R-134a filled heat pipe employing filling ratios in the range of
35–80% while discovering that the optimum-filling ratio was
80%. Lin et al. [26] studied heat pipes with silver-nanofluids at filling ratios of 20%, 40%, 60% and 80%. The study revealed the optimum-filling ratio for successful heat transfer was 60%. Finally,
Naphon et al. [27,28] tested heat pipes with both Titanium and
R11 nanofluids across various filling ratios. This study established
the optimum value of this ratio was 66% and 50%, respectively.
The heat transfer process ensuing inside a heat pipe with boiling working fluids reduced the pressure and fluid circulation
through the capillary structure. This complex process triggered
the evaporation-boiling phenomenon inside the heat pipe to
appear differently than more commonly assumed processes. The

visualization technique clearly proved these differences and provided important outcomes toward the characterization of heat pipe
operation. The visualization results also show the importance of
wick structure to heat pipe operation [29–31].
Most visualization techniques used during heat pipe operations
employ transparent media such as glass windows [32–35].
Radiography is one method of visualization that uses transmission
imaging and is a technique that utilizes the penetration of radiative
energy to non-destructively study the internal portions of opaque
objects. Finally, neutron radiography uses neutron beam techniques to penetrate heavy materials while beam are absorbed by
lighter materials. This technique is suitable for use in visualization
of the two-phase flow of hydrogen-based fluids inside metal enclosures [36,37].
Mishima et al. [38] visualized and measured two-phase flow
using neutron radiography, providing a result that could be used
not only as a qualitative measurement but also as a quantitative
measure. The need of high-speed video (HSV) technology with capture speeds up to 1000 frame per second was also highlighted in
this study. With an HSV camera, Mishima et al. were able to
observe the characteristics of the two-phase flow and the vapor
fraction. Uchimura et al. [39] stated that the inability of the
glass-transparent pipe to hold high pressure, and temperature
became the main disadvantage of the glass-based visualization
method. They further used real time neutron radiography (RTNR)
to visualize the two-phase flow of liquid metal and to calculate

N. Putra et al. / Experimental Thermal and Fluid Science 66 (2015) 13–27

15

Fig. 2. Conceptual heat pipe design.

Fig. 3. Fabricated heat pipe.

the velocity of a bubble by tracing the centroid of the bubble from
one image to another. Takenaka et al. [40] used neutron radiography to visualize the flow of boiling during the two-phase cryogenic cycle. Using the significant difference in attenuation
coefficient between the fluid nitrogen and the heat exchanger base
material aluminum, the boiling cryogenic fluid flow may be
characterized. Asano et al. [41] visualized and measured the fraction of liquid–gas inside a plate heat exchanger with a CCD camera
capturing images at 30 fps. The distribution of liquid–gas fraction
became the basis of heat exchanger optimizing.
Borgmeyer et al. [42] used neutron radiography to study the
heat transfer capability and fluid flow of oscillating heat pipe.
They found that neutron radiography could help expose the physical phenomenon and characteristics of the two-phase flow of
working fluid inside an oscillating heat pipe. Meanwhile,
Sugimoto et al. [36] evaluated the behavior of a working fluid
inside a variable conductance heat pipe (VCHP), also with neutron
radiography. Video operating at 30 fps captured visualization
images while using an Electron Bombardment Charged Coupled

Device (EB-CCD). The radiography revealed significant changes to
the results by the effect of adding a thin plate, changing the filling
ratio or by varying the inclination angle.
Further investigation into the relationship between the working
fluid filling ratio and the pressure inside a heat pipe are clearly
needed. Visualization techniques using neutron radiography also
needs further examination. This study measures the pressure and
temperature inside an operating heat pipe with various filling
ratios. Neutron radiography is utilized to visualize the activity
inside the heat pipe during experimentation and was conducted
at a neutron radiography facility, RN1 at The Centre of Science
and Technology of Advanced Materials (PSTBM), National Nuclear
Energy Agency (BATAN), Serpong, Indonesia.

2. Methods
The heat pipe sample was made of 2 mm thick aluminum with
an outer diameter of 32 mm. A heat sink was attached at the

16

N. Putra et al. / Experimental Thermal and Fluid Science 66 (2015) 13–27

Fig. 4. Experimental setup.

condenser; the heating surface at the evaporator was
50  40 mm2. Aluminum was chosen as the heat pipe material
due to its lower neutron attenuation coefficient compared with
that of water. The difference of attenuation coefficient resulted in
good contrast between aluminum as the metal enclosure of heat
pipe and water as the working fluid, enabling the researchers to
analyze the fluid within the enclosure more effectively.
A structure of sintered aluminum powder was attached to the
inner wall of the heat pipe, acting as a wick. The Wick structure
is made of aluminum powder with maximum diameter of
100 lm. Picture of the wick structure and SEM image of the wick
are shown in Fig. 1. A second heat pipe was left wickless and is furthermore referred to as the thermosyphon. Both the original heat
pipe and the thermosyphon were charged with water, the working
fluid of choice, at various filling ratios ranging from 10% to 90%.
Each heat pipe maintained a vacuum of 5 kPa. The value is the lowest that can be achieved by the vacuum pump and steadily maintained by the heat pipe. The design sample and fabricated
sample is in Figs. 2 and 3.
An electric heater rated at 250 W was attached to the evaporator to apply varying experimental heating loads of 9 W, 40 W
and 96 W. The heat load is chosen in such range that it represents
a low, medium and high load. The range is also represents the
amount of load that produced by heat source in the real application
such as computer’s CPU. To eject heat, an electric fan was attached
to the heat sink. At the interface of the heater and evaporator, five
type-K thermocouples were attached. A high precision and high
resolution Autonics Digital Pressure Sensor PSA-C01 was attached
at the condenser to maintain reading with an accuracy of 1% F.S.
and adequate pressure range of 101.3 kPa to 101.3 kPa, Both
the temperature and pressure sensors then connected to two high
precision data acquisition units, National Instrument (NI) 9203 and
NI 9211. The experimental setup is shown in Fig. 4.
Neutron radiography was conducted at the neutron radiography facility, RN1, and it used the scintillator detection method.

The neutron flux from the neutron beam tube was 106–107
n/cm2 s. The collimator attached to the beam tube had the ratio
of collimator total length and aperture diameter, L/D, of 83 with
60.95% thermal neutron, 0.78% scattered neutron, 0.48% gamma
and couple production of 2.85% [42]. The neutron radiography
facility can be seen in Fig. 5 shows the configuration of the
scintillator method used in RN1. The neutrons from the beam tube
were transmitted through the sample to the scintillator screen
made of Li6-ZnS. The scintillator is a homogeneous plastic sheet
containing a conversion material, lithium-6 fluoride (6LiF), and
phospor, copper–aluminum and gold activated zinc sulfide (ZnS:
Cu, Al, Au). The lithium absorbs neutron to form energetic helium
and tritium atoms.
6

Li þ n ¼ 3 H þ a þ 4:78 MeV:

ð1Þ

which cause ionization in the phosphor. Upon ionic relaxation,
phosphorescent photons with a wavelength of 455 nm (blue light)
are emitted, stimulating electronic excitations. As the electrons
return to their ground-states, they emit photons with a wavelength
of 565 nm (green light), which are detected by the CCD camera [43]
A mirror made of TiO2 with 95% reflectivity was used to protect
the camera from direct neutron beam, angled at 45° in relation to
the incoming beam. The scintillator screen, mirror and camera
were kept in a dark enclosure to ensure only the light emitted by
the scintillator was captured. The CCD camera was connected to
a computer to read, display and record the data. The CCD camera
used was Andor Technology iKon – M DD934N BV with resolution
of 1024  1024 and microphone size of 13  13. The camera maintains 95% Quantum Efficiency (QE) for the wavelength, 350–800 nm,
a 16 bit (65535) digitation. The experimental neutron radiography
setup is shown in Fig. 6. Images captured by this setup were processed using the processing software, ImageJ 1.47v. The images
were then used to measure the height the working fluid pool and
the gray level.

N. Putra et al. / Experimental Thermal and Fluid Science 66 (2015) 13–27

17

Fig. 5. Neutron radiography facility, RN1.

3. Results and discussion
3.1. Effect of working fluid filling ratio to evaporator temperature and
thermosyphon pressure
This part of study was conducted using only the thermosyphon,
which was made without aluminum wick structure attached to the
inner wall. The thermosyphon was vacuumed to 5 kPa and charged
with water to incrementally vary filling ratios by 10% from 10% to
90%. Each of the thermosyphons were loaded with uniform heating
loads of 9 W, 40 W and 96 W. The evaporator temperature and
pressure were recorded.
Fig. 7 illustrates the evaporator temperatures at variable filling
ratios with different heating loads. Table 1 shows the steady evaporator temperature of the thermosyphon at variable filling ratios
and heating loads. There is an optimum value of working fluid filling ratio that yields the lowest temperature at all of the heating
load variations, as extrapolated from the data found in both
Fig. 7 and Table 1. For all 9, 40 and 96 W heating loads, the 10% filling ratio exhibited the highest evaporator temperatures: 45.7 °C,
74.4 °C and 126.9 °C. The incremental filling ratios then systematically decreased the evaporator temperature, where the lowest
evaporator temperature occurred at the 50% and 70% filling ratios.
In this range, the evaporator temperature showed relatively identical values for all heating load variations at 40 °C, 68 °C and
111 °C for heating load wattage range. When the ratio increased
further to 80% and 90%, the evaporator temperature further

increased with results at the 90% filling ratio yielding 39.7 °C,
71.5 °C and 114 °C for heating load.
Fig. 8 displays the effect of working fluid filling ratio to the pressure inside the thermosyphon. The relation between filling ratio
and pressure has a similar trend to the relation between filling
ratio and temperature. It is clear that the 10% filling ratio produces
the highest pressure. In addition to the incremental filling ratio
change, the pressure systematically decreased until reaching its
lowest value at the ratio between 50% and 70%. The same pressure
was seen to rise again at the filling ratio range between 80% and
90%.
After obtaining the saturation pressure value, the saturation
temperature could be calculated using available software.
Comparing the saturation temperature data to the evaporator temperature data allowed for the estimation of the evaporation phenomenon inside thermosyphon. It must be noted that the
evaporation phenomenon only occurs when the evaporator temperature surpasses the saturation temperature. The relationship
between these temperatures in the thermosyphon at the predetermined filling ratios is depicted in Fig. 9. As shown, at most filling
ratio variations, the evaporator temperature remained above the
saturation temperature shortly after the application of heating
load. This primarily occurred at the 40 W and 96 W heating loads.
These results indicate that the thermosyphon worked with twophase principle and evaporation transpired. An exception occurred
when the highly charged thermosyphon received filling ratios of
80% and 90%. For this variation, as shown in Fig. 9, evaporation only

18

N. Putra et al. / Experimental Thermal and Fluid Science 66 (2015) 13–27

Fig. 6. Experimental setup of neutron radiography.

Fig. 7. Effect of filling ratio to evaporator temperature for various heating loads.

occurred after the application of the 40 W heating load. Therefore,
the two-phase heat transfer phenomenon is harder to achieve at
higher filling ratios.
The relationship between the working fluid filling ratio and the
evaporator temperature and pressure is described in Fig. 10. The
thermosyphon with a 10% filling ratio maintained the smallest

amount of working fluid and caused the thermosyphon to have
the lowest effective heat capacity, ceff. When heat was applied at
a predetermined amount, the sensible heat was used to increase
the temperature of both the thermosyphon and the working fluid.
Because the ceff was small, the temperature rose quickly and
reached the saturation temperature of the working fluid. The fluid
evaporated and expanded, filling up the vapor space and causing
the pressure to rise abruptly. Due to the saturation pressure
increase, the saturation temperature also escalated and caused
the evaporator temperature to rise, as well.
Meanwhile, at the 70% filling ratio, more working fluid caused a
higher heat capacity. This, in turn, produced more heat to be
absorbed and increased the working fluid temperature to instigate
phase change. The fluid in the vapor phase rushed up to the condenser and filled the empty vapor space. It then condensed and flowed back into the evaporator. The vapor circulation kept the
pressure and temperature comparatively low.
At the 90% filling ratio, virtually all of the heat absorbed by the
working fluid caused its temperature to escalate. When the saturation temperature was reached, the working fluid evaporated
within an insufficient amount of vapor space. With this limited
vapor expansion space, the evaporation rate in the evaporator
was decreased. The decrease of evaporation rate meant that the
heat absorbed as sensible heat, unlike the latent heat phase-changing heat at constant temperature, increased the evaporator temperature. This finding substantiates the study conducted by
Naphon et al. [27,28].

Table 1
Evaporator temperature at various heating load and filling ratio.
Filling ratio

10%

20%

Heat load (W)

Evaporator temperature (degree Celcius)

9
40
96

45.7
74.4
126.9

45.3
73.9
121

30%

43.5
73.3
118.6

40%

50%

60%

70%

80%

90%

42.4
71.6
116.9

41.3
68.4
111.4

40.4
68.4
113.5

40.6
65.2
111.4

40.4
67.2
113.6

39.7
71.5
114

N. Putra et al. / Experimental Thermal and Fluid Science 66 (2015) 13–27

Fig. 8. Pressure and filling ratio inside the thermosyphon.

3.2. Analysis of neutron radiography image
Fig. 11 displays the schematic of a heat pipe sample having
undergone neutron radiography. As illustrated, the fan-generated
airflow originated on the right side of the setup and the neutron
beam propagated normally across the pipe. The transparent aluminum enclosure exposed the pool of working fluid at the

19

evaporator, as shown by a dark area in the figure. The capillary pipe
made of stainless steel has a higher attenuation level than aluminum and results in a darker image. The dark zone at the upper
side of the evaporator is a temperature resistant chemical bond.
All the radiography images have exposure time of 4 s and spatial
resolution of 100 lm.
The image result of neutron radiography was analyzed by image
processing software, ImageJ 1.47v; these results are shown in
Fig. 11. The dark zone imagery shows the fluid locations inside
the heat pipe, as do the gray level measurements, regardless of if
the fluid is in a liquid or vaporous state. Image digitation was performed via an 8-bit image with the pixel gray level ranging from 0
to 255 and higher values indicate brighter pixel measures. A dark
zone inside the sample with maximum gray level below 60 is considered as working fluid in liquid state. It appeared inside the sample at the bottom part, in a large amount at initial condition, and in
small fraction at operating condition. The zone considered as the
working fluid liquid pool. The distance from the inner-bottom part
of heat pipe to the upper part of this dark zone was measured and
quantified as liquid pool height, as shown in Fig. 12. Meanwhile at
operating condition, a lighter dark zone (gray level of 90–150)
appeared at the upper surface of the liquid pool. It is considered
as the bubble produced from nucleation, and rise to the upward
direction. The distance from the upper surface of liquid pool to
the upper part of nucleate bubble was measured and quantified
as nucleate bubble height. In the further section, the nucleate bubble height is used to proportionally represent the intensity of the
bubble generation. For the sectional gray level measurement, the

Fig. 9. Evaporator and saturation temperatures inside the thermosyphon.

20

N. Putra et al. / Experimental Thermal and Fluid Science 66 (2015) 13–27

Fig. 10. Relationship between filling ratio, vapor space and pressure inside the thermosyphon.

Fig. 11. Schematic of sample having undergone neutron radiography.

heat pipe was divided into three sections as follows: the evaporator, the adiabatic and the condenser. The condenser is further
divided into three segments one condenser segment and two capillary pipe segments, as shown in Fig. 13.

3.3. Working fluid circulation in the thermosyphon and heat pipe
This study conducted neutron radiography imagery of two passive heat exchangers, the thermosyphon and the heat pipe. The

N. Putra et al. / Experimental Thermal and Fluid Science 66 (2015) 13–27

21

Fig. 12. Quantitative measurement of boiling phenomenon in the terms of liquid pool and nucleate bubble height at initial and operating condition.

identical heating load to the heat pipe, no condensate was
detected. No condensate was visually detected, and there was no
significant change in the gray level measurement at the condenser
section.
These outcomes indicate that the wick structure effectively
worked as a fluid circulation media. The nonexistence of a wick
structure in the thermosyphon caused condensate built up at the
condenser; condensate then could not be rapidly circulated to
the evaporator. The condensate in the thermosyphon took to
become heavy enough to gather after falling down and returning
to the evaporator. The vapor pressure obstructed the condensate
flow from the evaporator. This phenomenon caused the condensate to gather at that location, as shown in Fig. 14. A dissimilar situation ensued at the heat pipe, however. The existence of the wick
structure caused the condensate to return instantaneously to evaporator and was undetectable by radiography imagery.
3.4. Variation of heating load to boiling-evaporation phenomenon

Fig. 13. Sections of a sample in gray level measurement.

main difference between these two devices was the manner in
which working fluid was circulated from the condenser to the
evaporator. Unlike a wickless thermosyphon that used gravity to
circulate a working fluid, the heat pipe used the capillary force provided by the wick structure. The differences in the working principle affecting fluid circulation inside these two heat exchangers; the
circulation was captured by neutron radiography, as shown in
Fig. 14.
Fig. 14 shows both the thermosyphon and the heat pipe with
the heating loads applied at 0 W and 96 W. Even at the 0 W load
application, the heat pipe wick structure was saturated, as indicated by the darker area of the adiabatic and condenser section.
Meanwhile, at the application of a 96 W heating load, the main difference in the two structures is the existence of condensate at the
condenser. In the thermosyphon, the application of a 96 W heating
load produced a small amount of fluid condensate at the condenser, indicated by the dark area seen both in Fig. 14 and by
the gray level measurement. Finally, at the application of an

In this section, visualization using neutron radiography was
conducted on both the thermosyphon and the heat pipe. Heating
loads of 9 W, 40 W and 96 W were applied. A load of 162 W was
also applied to the thermosyphon and heat pipe for a short period
of time to achieve a picture of the boiling-evaporation phenomenon while operating under incredibly high heat flux. The
radiography result is displayed in Figs 15–17, and the quantitative
data of corresponding image is listed in Tables 2 and 3.
Fig. 15 shows the results of the thermosyphon sample with 20%
working fluid and a 5 kPa vacuum in the chamber. Meanwhile,
Table 2 shows the value of the working liquid pool height, the
nucleate bubble height and the gray level of each section of the
thermosyphon, referencing to Fig. 15 and Table 2 show that at
the application of a 9 W heating load (image number 2), evaporation ensued. The evaporation shown by the decrease of both
the liquid pool height and the gray level at the condenser section
were related to the existence of fluid at condenser. At this heating
load value, boiling did not occur, and bubbles were not generated.
At the application of the 40 W heating load (image number 3), the
height of liquid pool decreased further from 1.3 cm to 1.02 cm.
Boiling began at this point, as indicated by the existence of nucleate bubble, a gradation of gray zone above the liquid pool. The
height was measured at 1.02 cm from the surface of liquid pool.
The condensate increased, revealed by the further decrement of
the gray level at the condenser section from 120 to 116. Boiling
occurred more violently at the application of the 96 W heat load

22

N. Putra et al. / Experimental Thermal and Fluid Science 66 (2015) 13–27

Fig. 14. Fluid circulations in a thermosyphon versus a heat pipe.

(image number 4). This indicated further decrement of the liquid
pool to 0.54 cm; the nucleate bubble height increased further to
3.61 cm. Condensate was gathered with greater volume at the condenser, shown by a smaller gray level of 111. Bubble generation
and the condensate are both seen in.
In a short application of the 162 W heat load (image number 5),
the liquid pool height increased while the nucleate bubble height
decreased. These findings indicate that boiling intensity diminished at application of very high heat flux. However, the evaporation continued to occur at high intensity, as indicated by a
decrement of the gray level to 105 at the condenser section.
Additionally, the results clearly show greater condensate accumulation at the condenser during this period.
Meanwhile, Fig. 17 indicates the radiography result of the heat
pipe sample. The heat pipe was water-charged with 20% filling
ratio and vacuumed to 5 kPa. Table 3 shows the height of liquid
pool, the nucleate boiling height and the gray level of the entire
heat pipe section. Fig. 17 further displays a light area at the
lower-left portion of the heat pipe, above the liquid pool surface.
This area is assumed to be a part of the setup not enclosed by
the wick structure and detached from the heat pipe wall.
From Fig. 17, it can be inferred that the application of a 9 W
heating load to the heat pipe produced a quiet evaporation, shown
by the nonexistence of a nucleate bubble above the liquid pool.
Evaporation was detected by the decrement of the gray level at

the condenser, from 107 to 103, indicating the existence of condensate. It may be seen in Table 3 that the application of a 9 W heat
load resulted in the existence of fluid at the adiabatic section.
These findings assumed a condensate flow back to the condenser
from the evaporator via the wick structure. At the application of
a 40 W heating load, boiling began; at the application of the
96 W heating load, boiling ensued more intensely. The liquid pool
height was 1.33 cm and 0.48 cm, and the nucleate bubble height
from the liquid surface was 0.61 and 3.58 for the application of a
40 W and 96 W heating load, respectively. At the condenser, there
was no significant change in the gray level measurement, indicating no condensate remaining in the heat pipe.
Once again, at the short application of a 162 W heating load, a
decrease in boiling intensity was found, indicated by the decrease
of the nucleate boiling height, from 3.58 cm to 1.9 cm. The liquid
pool height also increased to 1.05 cm. The gray level at the condenser did not change significantly during this phase of
experimentation.
As inferred by the radiography visualization testing results,
both the thermosyphon and the heat pipe show an identical boiling-evaporation phenomenon. Fig. 8, depicts the application of
higher heat flux resulting in higher pressure inside the heat pipe.
Higher pressure leads to higher saturation temperature, producing
evaporation at higher temperature. At the application of a 9 W
heating load, the fluid temperature has surpassed the saturation

N. Putra et al. / Experimental Thermal and Fluid Science 66 (2015) 13–27

23

Fig. 15. Neutron radiography result of the thermosyphon.

temperature; however, the temperature was not high enough to
commence the nucleate bubble generation. This generation caused
the phase change phenomenon to remain relatively static without
the occurrence of boiling. At the application of a 40 W heating load,
fluid temperature reached a value high enough to begin bubble
generation. Boiling thus occurred and a bubble was detected. At
the application of a 96 W heating load, the fluid had a higher temperature value, causing rapid boiling and the formation of a large
nucleate bubble.
Finally, at the application of a 162 W heating load, the fluid
temperature has a very high temperature. The pressure, however,
of heat pipe at the 162 W heating load became so high that the saturation temperature was also extremely elevated. This led to the
reduction of boiling phenomenon, resulting in the decrease of the
nucleate bubble height. These findings confirm previous findings
from research conducted by Wong and Kao [35] in which they stated that there is no boiling phenomenon both at the very low and
very high heat fluxes.
3.5. Effect of pressure
The study on effect of pressure was conducted using heat pipe
sample. Heat pipe samples were used with the same filling ratio

and work under same heating load for this section of research;
however, different initial pressures, po, were used. The first sample
was set to atmospheric pressure at 101.3 kPa; meanwhile, the second sample was given a vacuum pressure of 5 kPa. The comparison
of the boiling-evaporation phenomenon of the two samples is
shown in Fig. 18. From the figure, it is realized that boiling did
not occur on the sample at atmospheric pressure and no nucleate
bubble formed above the surface of the liquid pool. Static evaporation was determined from the existence of condensate at the
condenser, especially at the application of a 96 W heating load.
Meanwhile, in the heat pipe operating with the vacuum at atmospheric pressure, boiling was observed after the application of a
40 W heating load. This phenomenon was identified by the existence of a gray zone above the liquid pool, and it again becomes
proof that boiling is harder to achieve in a high-pressure environment due to higher saturation temperatures. Finally, in the reduced
pressure environment, the liquid temperature surpassed the low
saturation temperature and induced boiling.
Those findings agrees in some way with the research conducted
by McGillis, et al. [44,45]. McGillis, et al. compared the nucleate
boiling data from three different pressure which are 4 kPa, 9 kPa,
and 101 kPa. Instead of using closed system as used in this study,
McGillis tested the boiling phenomenon at a controlled pressure

24

N. Putra et al. / Experimental Thermal and Fluid Science 66 (2015) 13–27

system. The study shows that although the reduction in pressure
for saturated water system shifted the boiling curve to higher wall
superheat level, the decrease in saturation temperature due to
lower pressure compensate the effect. Therefore, lower system
pressure leads to lower wall temperature for a given heat flux.
The lower pressure also lower vapor densities, larger bubble
diameter and decrease the minimum heat input required for nucleation to happen.
Comparing to the of McGillis’s, the result of this recent study
could not show the increase of superheat that required for nucleation. However both studies show that, for lower system pressure,
the nucleation achieved at lower heat input. Both also shows that,
in reduced pressure environment, bubble grows with larger
breadth and higher intensity.
3.6. Steady boiling-evaporation phenomenon

Fig. 16. Visual investigation of radiography result.

The study on boiling-evaporation phenomenon was conducted
using heat pipe sample. To determine the phenomenon of boiling-evaporation inside a heat pipe at steady state, a heating load
of 96 W was applied for 100 min to heat pipe samples under initial
vacuum pressure conditions. The fluid, water, was added to the
sample with a filling ratio of 20%. The neutron radiography imagery captured sequential images of the heat pipe temperature rise

Fig. 17. Neutron radiography results of heat pipe.

N. Putra et al. / Experimental Thermal and Fluid Science 66 (2015) 13–27
Table 2
Gray level measurement and height of the liquid pool and nucleate bubble
generation.

Table 3
Gray level measurement and height of liquid pool and nucleate bubble generation.

while the pipe was in a transient state until the temperature was
steady, as depicted in Fig. 19.

25

The figure shows the beginning of temperature raise when the
nucleate bubble was generated at 65 °C (image number 3). Along
with the temperature raise, boiling transpired more intensely
while the nucleate bubble size increased. The maximum nucleate
bubble height observed at the temperature of 105 °C (image number 7) showed a maximum height, hmax, of 3.88 cm. However, when
the rate of temperature change slowed, it was discovered that
when the temperature increased from 105 °C to 115 °C in 60 min,
the height of nucleate bubble also gradually decreased and reached
the minimum height of 0.55 cm.
This phenomenon cannot be attributed to the incremental
increase of pressure inside heat pipe, however, as the data shown
in Fig. 19 indicates, the evaporator temperature occurred at a
higher value than the saturation temperature. Thus, the condition
always fulfills the requirements to induce boiling. The phenomenon may be explained by the heat absorbed by the fluid to
elevate its temperature, as shown by numbers 1–2 in Fig. 19.
When the fluid reached the conditions to induce boiling and
develop a bubble as shown by numbers 3–7 in the same figure,
large amounts of fluid changed phase from liquid to vapor and
grew into a nucleate bubble. This transient boiling created a large
and intense bubble, rushing to fill the evacuated space and reaching maximum height. When the rate of temperature increase
began to decrease and the heat pipe began to operate at a steady
condition, vapor filled the once evacuated space. The vapor-filled
space maintained adequate pressure to suppress the bubble

Fig. 18. Effect of pressure in boiling-evaporation phenomenon for a heat pipe.

26

N. Putra et al. / Experimental Thermal and Fluid Science 66 (2015) 13–27

Fig. 19. Boiling and evaporation phenomenon in a steady-state heat pip.

generation so the nucleate bubble height decreased and reached its
minimum height. At the condenser, condensate then circulated
back to the evaporator. In this final state, heat pipe operation
was complete.
4. Conclusion
There are a number of conclusions resulting from this research,
as follows:
– Pressure measurement plays an important role in supplying
information about evaporation and the heat transfer
phenomenon inside of a heat pipe. The pressure value
may be used to determine the saturation temperature to
further determine the superheated condition at the
evaporator.
– An optimum value of working fluid filling ratio exists as related
to heat capacity and vapor space inside a heat pipe.

– Neutron radiography could be used as a visualization technique
for heat transfer phenomenon and liquid circulation examination inside a heat pipe. Using neutron radiography, it was discovered that static evaporation occurred at the application of
both very low and a very high heat flux. At reduced pressure,
a nucleate bubble developed with larger size and greater intensity due to low vapor density.
– In a steady-state heat pipe, boiling occurred at low intensity due
to the existence of vapor filling the evacuated space, which suppressed the development of the nucleate bubble.
– Neutron radiography showed the functionality of the wick
structure as a circulator of the working fluid.

Acknowledgment
The authors would like to thank the DRPM University of Indonesia
for funding this research through research cluster program.

N. Putra et al. / Experimental Thermal and Fluid Science 66 (2015) 13–27

References
[1] Y.-S. Chen, K.-H. Chien, T.-C. Hung, C.-C. Wang, Y.-M. Ferng, B.-S. Pei, Numerical
simulation of a heat sink embedded with a vapor chamber and calculation of
effective thermal conductivity of a vapor chamber, Appl. Therm. Eng. 29 (2009)
2655–2664.
[2] W.N. Septiadi, N. Putra, M. Juarsa, I.P.A. Putra, R. Sahmura, Characteristics of
screen mesh wick heat pipe with nano-fluid as passive cooling system, Atom
Indones. 39 (2013) 24–31.
[3] S.-S. Hsieh, R.-Y. Lee, J.-C. Shyu, S.-W. Chen, Analytical solution of thermal
resistance of vapor chamber heat sink with and without pillar, Energy Convers.
Manage. 48 (10) (2007) 2708–2717.
[4] H.N. Chaudhry, B.R. Hughes, S.A. Ghani, A review of heat pipe system for heat
recovery and renewable energy applications, Renew. Sustain. Energy Rev. 16
(2012) 2249–2259.
[5] M. Reyes, D. Alonso, J.R. Arias, A. Velazquez, Experimental and theoretical
study of a vapour chamber based heat spreader for avionics applications, Appl.
Therm. Eng. 37 (2012) 51–59.
[6] P. Naphon, S. Wiriyasart, Study on the vapor chamber with refrigerant R-141b
as working fluid for HDD cooling, Int. Commun. Heat Mass Transfer 39 (9)
(2012) 1449–1452.
[7] P. Naphon, S. Wongwises, S. Wiriyasart, Application of two-phase vapor
chamber technique for hard disk drive cooling of PCs, Int. Commun. Heat Mass
Transfer 40 (2013) 32–35.
[8] J.-C. Wang, Thermal investigations on LED vapor chamber-based plates, Int.
Commun. Heat Mass Transfer 38 (9) (2011) 1206–1212.
[9] J.-C. Wang, R.-T. Wang, T.-L. Chang, D.-S. Hwang, Development of 30 Watt
high-power LEDs vapor chamber-based plate, Int. J. Heat Mass Transf. 53 (19–
20) (2010) 3990–4001.
[10] N. Putra, Design, manufacturing and testing of a portable vaccine carrier box
employing thermoelectric module and heat pipe, J. Med. Eng. Technol. 33 (3)
(2009) 232–237.
Adriansyah, W. Sukyono, D. Johansem, F.N. Iskandar, The
[11] N. Putra,
characterization of a cascade thermoelectric cooler in a cryosurgery device,
Cryogenics 50 (11–12) (2010) 759–764.
[12] N. Putra, R. Saleh, W.N. Septiadi, A. Okta, Z. Hamid, Thermal performance of
biomaterial wick loop heat pipes with water-base Al2O3 nanofluids, Int. J.
Therm. Sci. 76 (2014) 128–136.
[13] C.-C. Yeh, C.-N. Chen, Y.-M. Chen, Heat transfer analysis of a loop heat pipe
with biporous wicks, Int. J. Heat Mass Transf. 52 (19–20) (2009) 4426–4434.
[14] D. Deng, D. Liang, Y. Tang, J. Peng, X. Han, M. Pan, Evaluation of capillary
performance of sintered porous wick for loop heat pipe, Exp. Thermal Fluid Sci.
50 (2013) 1–9.
[15] Z. Ming, L. Zhongliang, M. Guoyoan, The experimental and numerical
investigation of a grooved vapor chamber, Appl. Therm. Eng. 29 (2–3) (2009)
422–430.
[16] X. Ji, J. Xu, A.M. Abanda, Copper foam based vapor chamber for high heat flux
dissipation, Exp. Thermal Fluid Sci. 40 (2012) 93–102.
[17] R. Saleh, N. Putra, S.P. Prakoso, W.N. Septiadi, Experimental investigation of
thermal conductivity and heat pipe thermal performance of ZnO nanofluids,
Int. J. Therm. Sci. 63 (2013) 125–132.
[18] N. Putra, W.N. Septiadi, H. Rahman, R. Irwansyah, Thermal performance of
screen mesh wick heat pipes with nanofluids, Exp. Thermal Fluid Sci. 40 (2012)
10–17.
[19] S.-C. Wong, S.-F. Huang, K.-C. Hsieh, Performance tests on a novel vapor
chamber, Appl. Therm. Eng. 31 (10) (2011) 1757–1762.
[20] A.A.A. Attia, B.T.A. El-Assal, Experimental investigation of vapor chamber with
different working fluids at different charge ratios, Ain Shams Eng. J. 3 (3)
(2012) 289–297.
[21] P. Naphon, S. Wongwises, S. Wiriyasart, Application of two-phase vapor
chamber technique for hard disk drive cooling of PCs, Int. Commun. Heat Mass
Transfer 40 (2013). pp. 32.25.
[22] T. Sukchana, C. Jaiboonma, Effect of filling rations and adiabatic length on
thermal efficiency of long heat pipe filled with R-134a, Energy Proc. 34 (2013)
298–306.

27

[23] S. Lips, F. Lefevre, J. Bonjour, Combined effects of the filling ratio and the
vapour space thickness on the performance of a flat plate heat pipe, Int. J. Heat
Mass Transf. 53 (2009) 694–702.
[24] H.M.S. Hussein, H.H. El-Ghetany, S.A. Nada, Performance of wickless heat pipe
flat plate solar collectors having different pipes cross sections geometries and
filling ratios, Energy Convers. Manage. 47 (2006) 1539–1549.
[25] K.S. Ong, Md. Haider-E-Alahi, Performance of a R-134a-filled thermosyphon,
Appl. Therm. Eng. 23 (2003) 2373–2381.
[26] Y.-H. Lin, S.W. Kang, Effect of silver nano-fluid on pulsating heat pipe thermal
performance, Appl. Therm. Eng. 28 (2008) 1312–1317.
[27] P. Naphon, P. Assadamongkol, T. Borirak, Experimental investigation of
titanium nanofluids on the heat pipe thermal efficiency, Int. Commun. Heat
Mass Transfer 35 (2008) 1316–1319.
[28] P. Naphon, D. Thongkum, P. Assadamongkol, Heat pipe efficiency
enhancement with refrigerant-nanoparticles mixtures, Energy Convers.
Manage. 50 (2009) 772–776.
[29] S.-C. Wong, C.-W. Chen, Visual