Matsumoto 2018

Accepted Manuscript
Investigation on cohesive force of ice particles in ice slurry for

long-term ice storage
Koji Matsumoto , Hiroyuki Kubota , Yuri Umehara , Kohei Ehara ,

Junki Sakamoto , Jun Ueda , Kakeru Sato
PII: S0140-7007(18)30106-3
DOI: 10.1016/j.ijrefrig.2018.03.026
Reference: JIJR 3940
To appear in: International Journal of Refrigeration
Received date: 31 January 2018

Revised date: 28 March 2018
Accepted date: 28 March 2018
Please cite this article as: Koji Matsumoto , Hiroyuki Kubota , Yuri Umehara , Kohei Ehara ,
Junki Sakamoto , Jun Ueda , Kakeru Sato , Investigation on cohesive force of ice particles
in ice slurry for long-term ice storage, International Journal of Refrigeration (2018), doi:
10.1016/j.ijrefrig.2018.03.026
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service
to our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and
all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Highlights
 The trends of ice particle size, porosity, and permeability with time were found out.
 The factors dominating the cohesive force and their effects were found out .
 The cohesive force was found to be maximized at 8 h of storage.
 An inverse correlation between the cohesive force and permeability was identified.
 Appropriate equations to give the dimensionless cohesive forces were proposed.
T
IP
CR
US
AN
M
ED
PT
CE
AC
1
ACCEPTED MANUSCRIPT
Investigation on cohesive force of ice particles in ice slurry for

long-term ice storage
Koji Matsumoto1) Hiroyuki Kubota2) Yuri Umehara3) Kohei Ehara3) Junki Sakamoto3) Jun
Ueda3) Kakeru Sato3)
T
IP
1) Corresponding author. Professor, Department of Precision Mechanics, Chuo University, 1-13-27
CR
Kasuga, Bunkyo-ku, Tokyo 112-8551, Japan
US
Tel: +81-3-3817-1837; Fax: +81-3-3817-1820; E-mail: matsumoto@mech.chuo-u.ac.jp
2) Tokyo Metro Subway Co., Ltd., 3-19-6 Higashi-Ueno, Taito-ku, Tokyo 110-0015, Japan
AN
3) Graduate student, Chuo University, 1-13-27 Kasuga, Bunkyo-ku, Tokyo 112-8551, Japan
M
ED
PT
CE
AC
2
ACCEPTED MANUSCRIPT
Abstract
The use of ice slurry in the cold storage of foods has increased. However, ice particles in ice
slurry easily cohere to each other because of their buoyancy and weight. In this study, ice slurries
with different ice packing factors, supercooling degrees, and ice particle masses were formed in a
T
IP
beaker by supercooling dissolution of water and were stored for various storage times in the beaker
CR
in a temperature-controlled room. The resulting lump of ice particles was removed from the ice
US
slurry, and the cohesive force of the particles was measured at each storage time. The distribution of
the ice particle diameters and the average diameter were measured. The lump’s porosity was
AN
calculated from its dimensions, and the permeability was calculated from the average diameter and
M
the porosity. The trends of the cohesive force and its features were clarified, and appropriate
ED
equations to give the dimensionless cohesive forces were proposed.
Key words: Ice slurry, Ice particles, Cohesive force, Permeability, Long time storage
PT
CE
AC
3
ACCEPTED MANUSCRIPT
Nomenclature
c: Specific heat of pure water [kJ(kgK)–1]
D: Diameter of ice particle lump [mm]
de: Ice particle diameter [mm]
T
IP
F: Non-dimensional cohesive force [–]
CR
Fc: Cohesive force [N]
𝑔: Gravitational acceleration [ms-2]

US
IPF: Ice packing factor defined as (mass of ice particles)/(mass of ice slurry)×100 [%]
AN
K: Permeability [m2]
M
k: Kozeny–Carman constant [–]

ED
L: Height of ice particle lump [mm]
l: Latent heat of fusion of ice [kJkg–1]

PT
m: Mass of ice slurry [g]

CE
mice: Mass of ice particles [g]

AC
mw: Mass of pure water [g]
S: Surface area of ice particle considered to be sphere [mm2]
V (= mice/ice): Volume of ice particles [mm3]
T: Supercooling degree [K]
4
ACCEPTED MANUSCRIPT
ice: Density of ice [km–3]
w: Density of pure water [km–3]
: Porosity of ice particle lump [–]
T
IP
CR
US
AN
M
ED
PT
CE
AC
5
ACCEPTED MANUSCRIPT
1. Introduction
Ice slurry consists of ice particles and water or a solution. The use of ice slurry in not only air
conditioning but also the cold storage of foods has spread because of its good fluidity, rapid response
to changes in a heat load, and high thermal storage density, and its use is expected to expand further.
T
IP
For example, in the dynamic ice storage system for air conditioning, ice slurry is formed using
CR
power overnight, and the formed ice slurry is used for air conditioning during the daytime. Thus, the
US
formed ice slurry must be stored for a long time prior to use in air conditioning or in cold storage of
foods, and during this long-term storage, it is usually stirred. Recently, under the condition without
AN
stirring, a method of separately removing the ice particles and solution from ice slurry was
M
investigated (Egolf et al., 2008). However, during storage, ice particles in ice slurry easily cohere to
ED
one another because of their buoyancy and weight, regardless of stirring. Thus, during storage, a
lump of ice particles forms and interferes with effective extraction of cool heat from ice slurry.
PT
Furthermore, when two ice particles of different sizes adhere as a result of cohesion, the two
CE
particles grow into one through the process of Ostwald ripening (Pronk et al., 2005).
AC
A number of researchers have investigated methods of suppressing the cohesion of ice particles
in ice slurry. For example, it has been clarified that polyvinyl alcohol (PVA), as well as anti-freeze
protein (AFP), inhibits this cohesion, but surfactants and high polymers were found to have no
cohesion inhibition effects (Inada, 2003). Furthermore, it has also been clarified that the Ostwald
6
ACCEPTED MANUSCRIPT
ripening of ices can be suppressed using PVA and AFP (Inada & Modak, 2004). A cetyl dimethyl
betaine, which is an amphoteric surfactant, has been shown to suppress the cohesion of ice particles
and maintain a stable dispersion, but effect of the inhibition of Ostwald ripening was not observed in
this case (Suzuki, 2005). However, previous studies have been limited to the suppression of cohesion,
T
IP
and the cohesive force itself during storage has not been researched.
CR
In this study, the cohesive force arising during long-term storage was assessed; the cohesive
US
force corresponds to the force required to separate ice particles that have adhered to each other
without breaking the ice particles themselves. First, ice slurry was formed by forced supercooling
AN
dissolution of cooled pure water. The cohesive forces of the ice particles in the formed ice slurries
M
were measured under various supercooling degrees, ice packing factors (IPFs), masses of ice
ED
particles, and storage times. Under the same sets of conditions considered in the cohesive force
measurements, the distribution of ice particle diameters with the particles considered as spheres and
PT
the average ice particle diameter were also measured. Because ice particles form a lump during
CE
storage, the porosity of the lump was calculated from its dimensions, and the permeability based on
AC
the Kozeny–Carman equation was calculated from the average ice particle diameter and the porosity.
Finally, the trends of the cohesive force with storage time, its features, the correlation between the
permeability and the cohesive force and equations to give the dimensionless cohesive forces were
investigated, respectively.
7
ACCEPTED MANUSCRIPT
2. Experiment
2.1 Formation of ice slurry
The experimental apparatus to make ice slurry is shown in Fig.1. Pure water in a
T
polymethylpentene (PMP) beaker with a constant inner diameter was placed in a constant
IP
temperature bath at 20 °C, after which the pure water was cooled to a target supercooling degree
CR
while being stirred at 100 rpm to ensure there was no temperature gradient in the pure water. When
US
the pure water reached the target supercooling degree, one ice nucleus with diameter of about 1 mm
AN
formed beforehand was added to the pure water using a syringe so that supercooling of pure water
was forcibly dissolved at a desired supercooling degree and ice slurry was formed. The mass of the
M
formed ice particles was calculated as

ED
PT
CE
𝑐×𝑚𝑤 ×∆𝑇
𝑚𝑖𝑐𝑒 = .
𝑙 (1)
AC
After the formation of ice slurries, their IPFs were exactly adjusted to 10, 20, and 30 %, respectively.
In Eq.(1), the heat capacity of the PMP beaker was considered to be negligible. To confirm validity
of Eq.(1), the water level difference before and after freezing in the beaker was measured by a laser
8
ACCEPTED MANUSCRIPT
displacement meter because the water level rises due to expansion associated with freezing of water
when ice particles is formed from water. And the mass of the formed ice particles was determined
by the measured water level difference, the constant beaker inner diameter and the ice density. The
ratio of the mass of ice particles determined by the water level difference to that calculated by Eq.(1)
T
IP
was over 99%, confirming that the use of Eq.(1) is reasonable.
CR
Ice slurries adjusted to IPFs of 10, 20, and 30 % were stored in an incubator set to 0℃±1.0℃, as
US
shown in Fig.2. To prevent melting and further formation of ice particle, the beaker containing the
ice slurry was set in another larger beaker filled with water containing crushed ice, and the pair of
AN
beakers was set in an insulating container. This container was stored in the incubator for different
M
storage times (1, 3, 6, 8, 9, and 12 h). Because the aim of this study was the successful storage of ice
ED
slurry for 12 h, storage times up to 12 h were considered to confirm that the total mass of the ice
particles did not change during this period. So, the ratio of the mass of ice particles after storage for
PT
12 h to that just after ice formation was also measured using the laser displacement meter, and this
CE
ratio was found to be approximately 99%. Thus, it was also confirmed that the mass of ice particles
AC
did not change even after storage for 12 h.
2.2 Measurement of cohesive force
The experimental apparatus used to measure the cohesive force is shown in Fig.3. The
9
ACCEPTED MANUSCRIPT
apparatus was set in a temperature-controlled room whose temperature was less than 0℃. As shown
in Fig.3, the force was continuously measured by crushing the lump of ice particles on a sponge with
an acrylic plate attached to the tip of a load cell moved downward at a speed 10 mms–1. The
measured force had a tendency to increase linearly in time until the lump of ice particles was
T
IP
completely crushed. Simultaneously, an exact time at which the lump was completely crushed was
CR
determined from images of the crushing process taken by video camera, and the force at that time
US
was defined as the cohesive force. As mentioned in the introduction, the cohesive force corresponds
to the force required to separate ice particles that have adhered to each other without breaking the ice
AN
particles themselves. So, it is important to use a sponge in this apparatus to eliminate the effect of the
M
reaction force from the lab jack under the sponge when the lump is crushed. If the sponge is not used,
ED
the apparent cohesive force including the effect of the reaction force is measured, and the ice
particles themselves may be broken by the reaction force. It was confirmed that ice particles that
PT
adhered to each other were separated without breaking the ice particles themselves after each
CE
measurement run.
AC
2.3 Measurements of ice particle shapes and porosity and permeability of ice particle lump
Before the cohesive force was measured, the shapes of the ice particles and the porosity and
permeability of the lump of ice particles were measured because they may affect the cohesive force.
10
ACCEPTED MANUSCRIPT
The experimental apparatus to measure the ice particle shapes is shown in Fig.4. After the target
storage time, the beaker filled with the ice slurry was removed from the incubator, and a portion of
the ice particles in the ice slurry was immediately moved to a Petri dish for observation in a
temperature-controlled room at less than 0 ℃. The shapes of the ice particles were measured using a
T
IP
microscope with 100 magnifications. To determine an average ice diameter when the ice particle
CR
shape is assumed to be a sphere the following procedure was used. The projected area of each ice
US
particle was calculated through image analysis, a diameter of the circle whose area is equal to the
projected area was determined. From a surface area of a sphere having the determined diameter the
AN
ice particle diameter de was calculated as
M
4𝑆
ED
𝑑𝑒 = √ . (2)
𝜋
PT
Under each set of conditions, the diameters of 500 measured ice particles were calculated,
CE
respectively. The distributions of these sets of 500 ice particle diameters and the average ice particle
AC
diameter of each distribution were obtained.
After the ice slurry had been stored for the desired time, the cylindrical lump of ice particles
was removed from the ice slurry in the PMP beaker, and its height and diameter were measured. The
height and diameter were measured four times, respectively, under each set of conditions. And these
11
ACCEPTED MANUSCRIPT
averages for individual four measurements were used to calculate the porosity of the lump,, which
may affect the permeability, and the porosity was calculated as (Okawa et al., 2003)
4𝑚𝑖𝑐𝑒
𝜙=1− .
𝜌𝑖𝑐𝑒 𝐿π𝐷2 (3)
T
IP
The permeability K based on Kozeny-Charman Equation is shown in Eq.(4) (Okawa et al., 2003,
CR
2004, CARMAN, 1937).
US
AN
𝑑𝑒 2 𝜙3 (4)
𝐾= .
36𝑘 (1 − 𝜙)2
M
Under each set of conditions, the porosity and permeability of the lump and the cohesive force were
ED
measured four times, respectively, and these average values for individual four measurements were
PT
estimated as the results.

CE
The experimental conditions, the uncertainties of the measurements, and the densities of
water and ice (Katayama, 1986) are given in Tables 1, 2, and 3, respectively.
AC
12
ACCEPTED MANUSCRIPT
3. Experimental results
3.1 Average ice particle diameter, porosity, permeability, and cohesive force for fixed ice slurry mass
First, under the condition of a fixed ice slurry mass, the average ice particle diameter, the
porosity and permeability of the ice particle lump, and the cohesive force of the lump were measured
T
IP
for different IPFs, ice particle masses, and supercooling degrees. Under this condition, the IPF is the
CR
same meaning as the mass of the ice particles.
US
From the observation of ice particles, the number of spherical ice particles tended to increase
with increasing supercooling degree and storage time. However, in that of lower supercooling
AN
degree(1 K), flat ice particles were often observed.
M
As mentioned in Sub-section 2.3, under each set of conditions, the average ice particle diameter
ED
was obtained by measuring 500 ice particle shapes. The average ice particle diameter is plotted
against the storage time shown in Fig.5. As shown in Fig.5, the average ice particle diameter
PT
increased with increasing storage time because the ice particle diameter tended to increase with
CE
storage time because of Ostwald ripening. And the average ice particle diameter tended to be larger
AC
at a lower supercooling degree, regardless of the conditions. At the same storage time, the average
ice particle diameter increased with increasing mass (IPF), regardless of the supercooling degree.
Figure 6 shows two examples of lumps of ice particles with different IPFs after storage for 1h at
supercooling degree of 3 K. The diameters of these lumps after storage for 1 h were almost the same,
13
ACCEPTED MANUSCRIPT
but the height of the lump with the higher IPF was greater than that with the lower IPF. A similar
tendency was obtained at each storage time. From Eq.(3), this indicates that the porosity is mainly
dominated by the height of the lump, because there was little variation in the diameter, regardless of
conditions.
T
IP
Figure 7 shows the porosity is plotted against the storage time. As shown in Fig.7, the porosity
CR
tended to gradually decrease with time, regardless of the conditions because the lump’s height,
US
which dominates the porosity given by Eq.(3), decreased with storage time. At a mass of 60 g (IPF =
10 %), the porosity was largest and changed very little with supercooling degree, whereas in the
AN
other cases, the porosity increased with decreasing supercooling degree and decreased with
M
increasing mass of ice particles (IPF).

ED
Next, the permeability calculated from Eq.(4) was investigated. In Eq.(4), the Kozeny–
Carman constant, k , is generally considered to be 5, corresponding to a spherical particle. However,

PT
sometimes, there were many non-spherical ice particles in these experiments. Thus, in this study, the
CE
sphericity of the ice particles was taken into consideration as follows. An ice particle was considered
AC
to be spherical when the ratio of the shortest distance spanning the ice particle to the longest distance
exceeded 90%. When under each condition the ratio of spherical ice particles to 500 ice particles was
over 80%, between 40% and 80%, and less than 40%, the Kozeny–Carman constant, k, was set to 5,
corresponding to spherical ice particles; 9, corresponding to cylindrical ice; and 12, corresponding to
14
ACCEPTED MANUSCRIPT
flat ice, respectively (Carman et al., 1937; Ozgumus et al., 2014; Richardson and Harker, 1955). In
this study, flat ice was often observed under the lower supercooling degree of 1 K, whereas
cylindrical ice may correspond to an intermediate shape in the coalescence process of two ice
particles due to Ostwald ripening. The ratio of spherical ice particle is plotted against the storage
T
IP
time in Fig.8. As shown in Fig.8, the number of spherical ice particles at 3 K was much larger than
CR
that at 1 K. Additionally, the spherical particle tended to increase with increasing storage time,
US
regardless of the condition, although the differences among the three ratios obtained with different
IPFs under the same supercooling degree at each storage time were much smaller. For 3K the
AN
Kozeny–Carman constant k was 5 at 12 h storage, while, k was still 9 until 9 h storage,
M
regardless of the conditions. The ratios showed roughly the same qualitative variation with storage
ED
time at 1 and 3 K, and the differences among the ratios obtained at the three considered IPFs at each
storage time were also much similar at the two supercooling decrees; however, at 1 K, k only
PT
reached 9 at 12 h storage and remained at 12 up to 9 h storage, regardless of the condition.

CE
From the results shown in Figs.5, 7, and 8, the permeability under each set of storage time,
AC
mass of ice particles (IPF) and supercooling degree conditions was obtained, as shown in Fig.9. As
shown in this plot, the permeability showed a convex downward trend when plotted against storage
time, and the permeability was lowest at 6 h and increased after 6 h, regardless of the condition.
15
ACCEPTED MANUSCRIPT
The trend of the permeability was investigated. The term,  3/(1)2, related to the porosity, ,
against the storage time was calculated by Eq.(4), the decrease rate of the term gradually decreased
with storage time, especially, the decrease rate became further smaller after storage for 6 h. While,
the term de2/36k related to the ice particle diameter, de, against the storage time was also calculated,
T
IP
and it increased with storage time. So, this tendency arose because of the above tendencies for both
CR
terms.
US
At each storage time, differences in permeability values in the cases with the same ice particle
mass (IPF) and different supercooling degrees were found to be negligible at ice particle masses of
AN
120 and 180 g (IPF = 20 and 30 %, respectively), considering the uncertainty of the permeability
M
given in Table 2. However, regardless of storage time, this differences were significant at a mass of
ED
60 g (IPF = 10 %), with the permeability at the lower supercooling degree exceeding that at the
higher supercooling degree.

PT
The cohesive force is plotted against the storage time in Fig.10. Figure 10 shows that the
CE
cohesive force was maximized at 8 h, regardless of the condition. This tendency was roughly
AC
opposite that of the permeability shown in Fig.9, although the storage time at which the cohesive
force was maximized was slightly different from that at which the permeability was minimized.
Furthermore, the cohesive force tended to increase with increasing supercooling degree and mass of
ice particles (IPF). However, at each storage time, differences in the cohesive forces in the cases
16
ACCEPTED MANUSCRIPT
with the same mass of ice particles (IPF) and different supercooling degrees were found to be
negligible at ice particle masses of 120 and 180 g (IPF = 20 and 30 %, respectively), considering the
uncertainty of the cohesive force given in Table 2. In contrast, these differences may be meaningful
at a mass of 60 g (IPF = 10 %), with the cohesive force at the higher supercooling degree exceeding
T
IP
that at the lower supercooling degree.
CR
In the present stage, the result that the time at which the cohesive force was maximized
US
differed from that at which the permeability was minimized is not so important. The important
results obtained here are that the cohesive force and the permeability are strongly inversely
AN
correlated, the cohesive force was maximized at 8 h of storage, and the permeability was minimized
M
at a storage time near 8 h. And there remains room for the consideration of the permeability based on
ED
the Kozeny–Carman equation.
As mentioned above, the tendency of the cohesive force was roughly opposite that of the
PT
permeability; that is, the cohesive force tended to increase with decreasing permeability. Thus, the
CE
permeability might be considered to be a physical quantity related to the construction of the lump of
AC
ice particles. A lower permeability corresponds to a denser construction; therefore, the cohesive
force increases with decreasing permeability.
17
ACCEPTED MANUSCRIPT
3.2 Cohesive force under a fixed mass of ice particles
Under the condition of a fixed mass of ice particles, the cohesive forces were measured at
different IPF values and supercooling degrees. Under this condition, the IPF is not the same meaning
as the mass of the ice particles. The cohesive force is plotted against the storage time in Fig.11.
T
IP
Similarly to Fig.10, the cohesive force was also maximized at storage time of 8 h, regardless of the
CR
IPF and supercooling degree. At each storage time, the cohesive force had a tendency to be almost
US
the same, regardless of the IPF or supercooling degree, considering the uncertainty of the cohesive
force given in Table 2; however, at an IPF of 10 %, small but a little meaningful differences between
AN
the cohesive forces at the two considered supercooling degrees (1K,3K) were observed at storage
M
times of 6 and 9 h. The reason for these differences will be investigated in future work.
ED
3.3 Dimensionless cohesive force

PT
In this study, the dimensionless cohesive force was also given to facilitate comparisons between
CE
the present results and results of future studies. The ice volume V in the following equations was
AC
calculated as mice /ice.
In the case of a fixed mass of ice slurry dominating factors are the buoyancy of the lump and the
fixed mass of ice slurry. So, non-dimensional equation was given by dividing the measured cohesive
forces by the dominating factors, and it was expressed as Eq.(5).
18
ACCEPTED MANUSCRIPT
𝐹𝑐
𝐹= . (5)
IPF IPF
(100 + (1 − 100))𝑚𝑔 − 𝜌w 𝑉𝑔
The dimensionless cohesive force calculated using Eq. (5) considering the results of Fig.10 is plotted
T
IP
against the storage time in Fig.12. The qualitative trends shown in Figs.10 and 12 were found to be
CR
in good agreement. Thus, under this case, Equation (5) effectively transformed the dimensional
cohesive forces into dimensionless values.
US
In the case of a fixed mass of ice particles, dominating factors are the buoyancy and the fixed
AN
mass of ice particles. So, similarly to Eq.(5) non-dimensional equation was given, and it was
M
expressed as Eq.(6). The dimensionless cohesive force calculated using Eq. (6) considering the
ED
results of Fig.11 is plotted against the storage time in Fig.13.

PT
𝐹𝑐
𝐹= . (6)
CE
IPF
𝜌w 𝑉𝑔− 100 𝑚𝑔
AC
The qualitative trends shown in Figs.11 and 13 were found to be in good agreement. Thus, under this
case, Equation (6) effectively transformed the dimensional cohesive forces into dimensionless
values.
19
ACCEPTED MANUSCRIPT
4. Conclusions
(1) The dependence of the average ice particle diameter and the porosity and permeability of the ice
particle lump on storage time were found out.
(2) The factors dominating the cohesive force and their effects were found out.
T
IP
(3) The cohesive force was maximized at a storage time for 8 h.
CR
(4) An inverse correlation between the cohesive force and the permeability was identified.
US
(5) Appropriate equations to give the dimensionless cohesive forces were proposed under both
conditions of a fixed ice slurry mass and a fixed mass of ice particles.
AN
M
Acknowledgement
ED
This study was supported by the 2018 Institute of Science and Engineering of Chuo University.
The authors wish to thank Mr. Tezuka and Mr. Morita, graduates of Chuo University, for their
PT
assistance.
CE
AC
20
ACCEPTED MANUSCRIPT
References
CARMAN P.C., 1937, FLUID FLOW THROUGH GRANULAR BEDS, Trans.Inst.Chem.Eng
15,150-66.
Egolf P.W., Kitanovski A., Ata-Caesar D., Vuarnoz D., Meili F., Cold storage with ice slurries,
T
IP
2008, Int. J. Energy Res.,32,187-203.
CR
Inada T., 2003, Inhibition of Agglomeration of Ice Particles by using Substitutes for Antifreeze
Protein, REFRIGERATION, JSRAE, 78-910, 653-657.
US
Inada T., and Modak P.R.,2006, Growth control of ice crystals by poly(vinyl alcohol) and
AN
antifreeze protein in ice slurries, Chem. Eng. Sci., 61,3149-3158.
M
Katayama K.,1986, Heat Transfer, JSME, 4th edition,332,331.

ED
Kumano H., Hirata T., Hagiwara Y., Tamura F., 2012, Effects of storage on flow and heat transfer
characteristics of ice slurry, Int. J. Refrig., 35(5)122-129

PT
Okawa S., Saito A., Eriksson A., Maeda T, Hozumi T., Kumano H., 2003, Study on Permeability
CE
of Ice/Water Mixtures, Trans. of JSRAE,20-3，287-296.

AC
Okawa S., Saito A., Maeda T, Hozumi T., Kumano H.,2004, Influence of Storage on
Permeability of Ice/Water Mixtures, Trans. of JSRAE, 21-1，77-84.
Ozgumus T., Mobedi M., Ozkol U., 2014, DETERMINATION OF KOZENY CONSTANT
BASED ON POROSITY AND PORE TO THROAT SIZE RATIO IN POROUS MEDIUM WITH
21
ACCEPTED MANUSCRIPT
RECTANGULAR RODS, ENG. APPL. COMP. FLUID.,8-2,308-318.
Pronk P., Hansen T. M., Infante Ferreira C. A., Witkamp G. J., A dynamic model of Ostwald
ripening in ice suspensions,2005, J. Cryst. Growth, 275,1355-1361.
Richardson J.F. and Harker J.H., 1955, Chemical Engineering, Vol. 2, Fifth edition, ENG
T
IP
APPL COMP FLUID 1-210.
CR
Suzuki H., 2005, Preventing Agglomeration and Drag Reduction of Ice Slurries Treated with
US
Surfactant, REFRIGERATION, JSRAE, 80-928,110-115.
AN
M
ED
PT
CE
AC
22
ACCEPTED MANUSCRIPT
Stirrer
PMP beaker Impeller
T
IP
Pure water
Cold Brine
CR
US
Fig.1 Experimental apparatus to make ice slurry
AN
Incubator Insulating container
M
ED
Ice slurry
PT
Crushed ices - water

CE
AC
Fig.2 Ice slurries stored in incubator
23
ACCEPTED MANUSCRIPT
Load cell Torque motor
T
IP
Acrylic plate
CR
Ice particles’ lump
Sponge
US Lab jack
AN
Fig.3 Experimental apparatus to measure cohesive force

M
ED
Micro scope
PT
Video camera
CE
Ice slurry
AC
Fig.4 Experimental apparatus to measure ice particle shapes
24
ACCEPTED MANUSCRIPT
Average ice paricle's diameter d[mm]

0.7
0.6
0.5
T
IP
0.4
CR
0.3
1 3 6 8 9 12
Storage time [h]
US
60g,IPF=10%(supercooling degree 1K)
AN
M
Fig.5 Relation between storage time and average ice particle diameter
under fixed ice slurry mass
ED
127mm 128mm
PT
10mm 10mm
CE
IPF=10% IPF=30%
(a) Diameter of ice particles’ lump
AC
52mm
42mm
10mm 10mm
IPF=10% IPF=30%
(b) Height of ice particles’ lump
Fig.6 Diameters and heights of ice particles’ lumps after 1 hour storage under supercooling
degree of 3 K
25
ACCEPTED MANUSCRIPT
Porosity [-] 0.9
0.8
T
IP
0.7
CR
0.6
1 3 6 8 9 12
US
Storage time [h]

AN
M

ED
Fig.7 Relation between storage time and porosity under fixed ice slurry
PT
mass
CE
AC
26
ACCEPTED MANUSCRIPT
Ratio of spherical ice particle [%]

100
k=5
(CARMAN et al.,1937)
80
60
T
k=9
IP
(Ozgumus et al.,2014)
40
CR
20 k=12
(Richardson et al.,1955)
0
1 3 6
US 8 9
Storage time [h]
12
AN
M

ED
Fig.8 Relation between storage time and ratio of spherical ice particles
PT
CE
AC
27
ACCEPTED MANUSCRIPT
×10-9 30
Permeability [m2]
20
T
IP
10
CR
0
1 3
US
6 8 9
Storage time [h]
12
AN
M

ED
Fig.9 Relation between storage time and permeability under fixed ice
slurry mass
PT
CE
AC
28
ACCEPTED MANUSCRIPT
40
Cohesive force [N]
T
30
IP
CR
20
10
1 3 US 6 8 9
Storage time [h]
12
AN
M

ED

PT
Fig.10 Relation between storage time and cohesive force under fixed ice
slurry mass
CE
AC
29
ACCEPTED MANUSCRIPT
40
Cohesive force [N]
T
IP
30
CR
20
10
1 3
US6 8 9 12
AN
Storage time [h]
M

ED
PT
Fig.11 Relation between storage time and cohesive force under fixed mass of ice
particles
CE
AC
30
ACCEPTED MANUSCRIPT
Non-dimensional cohesive force [-]
12
10
T
IP
8
6
CR
4
2
0
US
AN
1 3 6 8 9 12
Storage time [h]
M

ED

PT
CE
Fig.12 Relation between storage time and non-dimensional cohesive force

AC
31
ACCEPTED MANUSCRIPT
Non-dimensional cohesive force [-] 300
200
T
IP
CR
100
0
1 3 US 6 8 9
Storage time [h]
12
AN

M

ED
PT
Fig.13 Relation between storage time and non-dimensional cohesive force

under fixed mass of ice particles
CE
AC
32
ACCEPTED MANUSCRIPT
Table 1 Experimental conditions
Supercooling degree 1K，3K

Ice Packing Factor(IPF) 10%，20%，30%
Mass of ice particles 60g，120g，180g
T
Mass of ice slurry 600g， 1800g
IP
Storage time 1h，3h，6h，8h，9h，12h
CR
Physical quantity
US
Table 2 Uncertainties of measurements
Uncertainty
AN
Mass ±0.12g
Temperature ±0.58℃
Water level ±2.4μm
M
Ice particle size ±1.5μm

Diameter of ice particle lump ±1.3mm
ED
Height of ice particle lump ±1.5mm

Porosity ±0.012
Permeability ±1.7×10-9m2
PT
Cohesive force ±1.5N

CE
AC
Table 3 Densities of water and ice (Katayama, 1986)

Substance Density
Water (0.01℃) 999.78 kgm-3
Ice (0.0℃) 917.0 kgm-3
33

Matsumoto 2018

Uploaded by

Copyright:

Available Formats

You might also like

Matsumoto 2018

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Matsumoto 2018

Uploaded by

Copyright:

Available Formats

Accepted Manuscript

Investigation on cohesive force of ice particles in ice slurry for

Koji Matsumoto , Hiroyuki Kubota , Yuri Umehara , Kohei Ehara ,

To appear in: International Journal of Refrigeration

Received date: 31 January 2018

Investigation on cohesive force of ice particles in ice slurry for

Ueda3) Kakeru Sato3)

equations to give the dimensionless cohesive forces were proposed.

c: Specific heat of pure water [kJ(kgK)–1]

D: Diameter of ice particle lump [mm]

de: Ice particle diameter [mm]

𝑔: Gravitational acceleration [ms-2]

k: Kozeny–Carman constant [–]

L: Height of ice particle lump [mm]

l: Latent heat of fusion of ice [kJkg–1]

m: Mass of ice slurry [g]

mice: Mass of ice particles [g]

mw: Mass of pure water [g]

S: Surface area of ice particle considered to be sphere [mm2]

V (= mice/ice): Volume of ice particles [mm3]

T: Supercooling degree [K]

ice: Density of ice [km–3]

w: Density of pure water [km–3]

: Porosity of ice particle lump [–]

2.1 Formation of ice slurry

formed ice particles was calculated as

did not change even after storage for 12 h.

2.2 Measurement of cohesive force

diameter of each distribution were obtained.

estimated as the results.

increasing mass of ice particles (IPF).

Carman constant, k , is generally considered to be 5, corresponding to a spherical particle. However,

reached 9 at 12 h storage and remained at 12 up to 9 h storage, regardless of the condition.

higher supercooling degree.

the Kozeny–Carman equation.

force increases with decreasing permeability.

3.2 Cohesive force under a fixed mass of ice particles

3.3 Dimensionless cohesive force

calculated as mice /ice.

forces by the dominating factors, and it was expressed as Eq.(5).

cohesive forces into dimensionless values.

results of Fig.11 is plotted against the storage time in Fig.13.

particle lump on storage time were found out.

CARMAN P.C., 1937, FLUID FLOW THROUGH GRANULAR BEDS, Trans.Inst.Chem.Eng

Protein, REFRIGERATION, JSRAE, 78-910, 653-657.

Katayama K.,1986, Heat Transfer, JSME, 4th edition,332,331.

characteristics of ice slurry, Int. J. Refrig., 35(5)122-129

of Ice/Water Mixtures, Trans. of JSRAE,20-3，287-296.

Permeability of Ice/Water Mixtures, Trans. of JSRAE, 21-1，77-84.

RECTANGULAR RODS, ENG. APPL. COMP. FLUID.,8-2,308-318.

ripening in ice suspensions,2005, J. Cryst. Growth, 275,1355-1361.

PMP beaker Impeller

Crushed ices - water

Fig.2 Ice slurries stored in incubator

Load cell Torque motor

Fig.3 Experimental apparatus to measure cohesive force

Fig.4 Experimental apparatus to measure ice particle shapes

Average ice paricle's diameter d[mm]