VUIC-2019

International Conference on Innovation and Societal Transformation

(VUIC-2019)
th st
28 February& 1 March 2019, Pune

Review on Eutectic Freeze Crystallisation and Auxiliary Cooling Systems

Vedang Nagre , Kausik Ware, Shivam Baldota, Prajna Paramita Choudhuri, Krishna Patil
Guided by: Prof. Pranav Dhaneshwar, Prof. Gayatri Gawande
1, Department of Chemical Engineering, Vishwakarma Institute of Technology, Bibwewadi, Pune
E-mail addresses: vedang.nagre16@vit.edu kaushik.ware16@vit.edu
Contact No.: +919987532513
--------------------------------------------------------------------------------------------------------------------------------------------
Abstract:
As compared to the membrane separation techniques or freeze crystallisation, conventional process
such as evaporation yield higher concentration. Although due to low energy consumption and energy
requirement in the process, freeze crystallisation technique yields comparatively higher quality product
, which makes it very appropriate technique for the processing of RO reject water . Eutectic freeze
crystallization (EFC) separates inorganic aqueous solutions into pure salt and pure water. Operating at
the eutectic point, ice and salt can be formed concurrently as two separate phases. In this review paper,
various types of eutectic fluids systems and their treatment supplemented by the study of refrigeration
technologies which consist of conventional gas compression systems and magnetic refrigeration sys-
tem are studied.

Keywords: Membrane Separation Techniques, Eutectic Freeze Crystallisation, Reverse Osmosis,

Magnetic Refrigeration

Introduction:

Earth comprises 75% water, and 97.5% of all
water available of the Earth's surface is saline
while the rest of it manifests as fresh water.
However, only 0.3% of this fresh water can be
directly used for day-today purposes [1–3].
Global warming, climate change, human popu-
lation growth, and rapid industrialization have
led to continuous increase in the demand for
water supply, thus, cornering the issue of water
scarcity. Human population on Earth is pro-
jected to rise by a whopping 2.6 billion by the
year 2025, and two-thirds of these people would
be living under severe water shortage while the
rest would have to face total water scarcity
[3,4]. Desalination is a major technology that
serves to fulfil water-supply demands by pro-
cessing the saline water available in plenty on
the Earth's surface. As of 2017, capacities of de-
salination plants have achieved values of the or-
der of 99.8 million m3/day [3,5]. Desalination,
in general, can be categorized into two
groups— (1) involving phase change process;
and (2) involving water-of-salt extraction [6].
Desalination processes involving phase
changes, which were first developed, include
thermal desalination processes, such as multi-
effect distillation (MED), multi-stage flash dis-
tillation (MSF), and humidification–dehumidi-
fication desalination (HDH). The second cate-
gory of desalination processes that require use
of membranes include reverse osmosis (RO),
forward osmosis (FO), and electro dialysis
(ED). The RO process has been widely adopted
as a popular desalination technology and has
been used 80% of the times in over 15,000 de-
salination plants spread across the world. RO is
superior to thermal desalination owing to its
lower energy requirements, and development in
membrane materials and manufacturing pro-
cesses have also led to lower operating costs of
RO processes. Nowadays, yet another phase
change based desalination technique the freez-
ing melting process has been introduced. Freez-
ing desalination (FD) starts to gain recognition
due to several favourable points compared to
thermal and membrane-based desalination.
This paper provides a comprehensive explana-
tion and understanding of the FD process as a
newcomer in desalination technology by high-
lighting the basic concept underlying the pro-
cess and its development as well as describing
its various applications and potential to be inte-
grated along with other desalination technolo-
gies, thereby creating a hybrid system. Thence-
forth, the paper illustrates promising future ap-
plications of FD from other aspects.
1.1 Cooling Technology
Heat machines are widely used as heat pumps
providing cold-to-warmer-level heat transfer
from the point of view of cost efficiency, usage
of heat pumps is more efficient than direct heat-
ing by fuel burning or by electricity (the ad-
vantages of heat pumps are considered in de-
tails, in particular, in [8]. Vapour-compression-
type heat machines are at present time most of-
ten applied for artificial refrigeration using
evaporation-condensation and compression-ex-
pansion thermodynamic cycles. Currently, ab-
sorption refrigeration [9], thermoelectric cool-
ing [10], and air cycle refrigeration [11] are
considered to be alternative to the vapour-com-
pression refrigerating technology. Power effi-
ciency problems of the currently used vapour-
compression refrigeration technology can be
solved by a magnetic heat (refrigeration) ma-
chine that can be used in the same areas where
the vapour-compression heat machines are ap-
plied. Magnetic refrigerating machine is a cool-
ing device which implements magnetic cooling
technology based on magneto-calorific effect
(MCE). MCE manifests itself in an adiabatic
temperature change or an isothermal change in
the magnetic part of the entropy of a magnetic
substance with a change in the external mag-
netic field. MCE reaches its maximum value at
temperatures close to the temperature of the
magnetic phase transition of the magnetic ma-
terial. As a working body in the magnetic re-
frigerating machine a magnetic material in solid
phase is used that changes its temperature (heat-
ing-cooling) and magnetic part of entropy re-
lated to the magnetic system of the material
while magnetizing-demagnetizing (because of
MCE). This enables to replace the vapour com-
pression cycles of evaporation-condensation
and compression expansion with the solid body
magnetization-demagnetization cycles.
2 Limitation of major desalination technol-
ogy
Several characteristics of FD can overcome
drawbacks of the major desalination technol-
ogy, thermal desalination and membrane desal-
ination. Freeze desalination is insensitive to
fouling phenomena which is the handicap of
membrane-based desalination processes [6,12–
14]. Unlike the RO process, FD does not require
intensive pre-treatment and chemical require-
ment when performing the process. In addition,
RO essentially produces concentrated brine,
which is harmful to environment. FD, on the
other hand, possesses the potential to treat con-
centrated brine, produced by the RO process,
close to zero liquid discharge (ZLD) [15].
Freeze desalination suffers minimum scaling
and corrosion problems owing low operating
temperatures when compared against thermal
desalination process. In terms of energy re-
quirement, thermodynamically, latent heat of
ice fusion is 333 kJ/kg while evaporation of wa-
ter requires 2500 kJ/kg [16,17]. The FD pro-
cess, therefore, requires only one-seventh (ap-
proximately) the energy required during ther-
mal desalination. The competitiveness of FD
has been growing by facing the obstacles in the
main desalination technology. Detailed infor-
mation on the limitation of membrane-based
and thermal desalination will be further ex-
plained below.
2.1 Thermal Desalination Disadvantages
Treating concentrated brine under high opera-
tional temperature leads corrosion problems,
and it significantly effects on production rate
[18,19]. The expensive capital cost is also dis-
advantage for installing more efficient MSF
process because the water recovery depends on
the number of distillation stages [20]. MED also
known as multiple effect evaporation (MEE) is
the advanced design to mitigate the high energy
demand of MSF. MED utilizes lower operating
temperature than MSF so that the process con-
sumes 15 kWh/m3 to produce desalinated water
[22]. Though MED has advantages in relatively
low specific capital cost and short start-up pe-
riod, the general thermal desalination problems
derived from operating at 60–700C was not
solved [20,21]. Lower unit capacity when boil-
ing point rises considerable than MSF process
is another shortcoming of MED [23]. Another
advanced thermal desalination such as HDH,
multi effect humidification has been invented
[24,25], and some elements of the process has
been modified, such as thermal or mechanical
hybrid vapour compressor [26] and natural vac-
uum [27]. Adsorption desalination (AD) then
developed due to its low energy consumption
compared to other thermal desalination
method—1.5 kWh/m3 [23]. AD requires high
number of various adsorbent material [28–30]
low production rate even with repeated cycle
[31] and space-demanding system [32]. In gen-
eral, thermal desalination still requires high
electrical and mechanical energy which come
from fossil fuel, and results in high CO2 emis-
sion, another problem [33–35]. To achieve
higher efficiency of the thermal process, some
researches started to focus on utilization of re-
newable thermal energy. As example, in Africa
and Middle East, the utilization of solar energy
has grown tremendously [36]. Though there
were many works performed to increase energy
efficiency, the developed thermal process still
consumes more energy than membrane process.
To summarize, freezing requires much less heat
than evaporation.
2.2 Membrane Base Desalination Disad-
vantages
utilizing membrane generates some drawbacks
of membrane-based desalination. The main ob-
stacles are high costs on maintenance, multi-
steps of pre-treatment and brine problems. Sev-
eral factors are known to drive the maintenance
cost of membrane-based desalination in water
treatment systems, such as power requirements,
labour, material, chemical needs for membrane
cleaning and scale inhibition, as well as mem-
brane life cycle [37]. The major factors that
raise the cost of the maintenance of membrane
processes is the membrane fouling resulting in
the rapid decline of permeate flux during the
process [38,39]. This tends to increase number
of cleaning and membrane replacement, affect-
ing the process efficiency as well as economical
consideration. It leads to high operating and
maintenance costs, thus, it is hard to apply this
technology in low and middle income countries
[40]. Backwash is necessary during membrane
operation to remove some accumulation of re-
versible foulants, thus some researches focused
on optimizing backwash process [40,41]. At
some point, chemical cleaning has to be applied
to control foulants that are not removed just by
application of backwash [43]. The usage of
chemicals is both not economically and envi-
ronmentally friendly. This is another plus point
for FD since it is not required any chemical ad-
dition during the process. Some studies are still
going on to find the most effective way of mem-
brane cleaning process to deal with economic
and environmental impact. From an economic
perspective, a simulation of the FD process was
performed [44] for treating water drawn from
the Dakota Aquifer for augmentation of the
Grand Forks, North Dakota. Based on results of
this simulation, water-production costs, includ-
ing capital, maintenance, and operational costs,
were calculated. The water production cost of
FD is compared to that incurred when using
other desalination methods, and the comparison
is illustrated in Fig. 1. It can be clearly seen that
the FD process ($0.34/m3) is more economical
when compared to the other desalination meth-
ods.
Fig: 1 Water-production costs of various desal-
ination methods [44]
3. Principle and general concept of eutectic
freeze desalination
the FD process represents a desalination tech-
nique involving a phase change—which in this
case, constitutes a freezing process—from liq-
uid to solid. Liquid, in this case, refers to sea-
water or saline water (i.e. brine) while solid re-
fers to ice. Theoretically, a major part of ice
crystals comprises pure water. [45]
A major part of ice crystals comprises pure wa-
ter. During the freezing process, ice crystal will
form, thus, fresh water will be extracted in the
form of ice and the liquid remained will be more
concentrated. The FD process, therefore, pos-
sesses a high separation factor. [45] The FD
process heavily relies on use of refrigerants, as
it needs lower temperature to operate. In terms
of refrigerant position, therefore, the FD pro-
cess can be classified into two types—direct FD
and indirect FD [46]. Fig. 2a illustrates the di-
rect FD process, wherein the refrigerant is
maintained in directly in contact with saline wa-
ter. The refrigerant enters the crystallizer, ab-
sorbs heat from saline water, and thus, vapor-
izes. Simultaneously, saline-water temperature
decreases, and nucleation occurs.[45] On the
other hand, major drawbacks of direct FD in-
clude refrigerant contamination of water prod-
ucts and concentrated brine as well as probabil-
ity of hydrate formation[45]. Fig. 2b depicts a
simplified block diagram of the indirect FD
process. Compared to direct FD, the main dif-
ference here is that the refrigerant and saline
water are separated by the crystallizer wall. The
advantage of indirect FD is the production of
refrigerant-free water, which is important to de-
termine the future usage of product water. How-
ever, indirect FD induces complexity in nucle-
ation and crystal growth as well as increases op-
erating costs. Since heat transfer in this case oc-
curs through a solid barrier, the process is not
as efficient as in direct FD. The FD process
comprises minimum three steps—(1) precool-
ing, (2) crystallization, and (3) separation[6].
The precooling step is required to prevent sud-
den crystallization of saline water caused by the
massive temperature drop. The solution, in-
stead, is cooled down until it attains a tempera-
ture close to its freezing point. The crystalliza-
tion step can be divided into nucleation and ice-
growth steps. Nucleation refers to the phenom-
enon wherein aggregate of water molecules be-
come more ordered, thereby forming a rela-
tively stable lattice structure that continuously
grows until it attains a state of thermodynamic
equilibrium with the solution. At some point,
the lattice structure exceeds the energy barrier
to nucleate, then a not readily re-dissolved
structure will form, called nuclei or nucleus
[46,47]. For a nuclei to grow spontaneously,
there exists a minimum crystal size, called crit-
ical radius, which varies with changes in tem-
perature. Kelvin equation is shown below[49]
$%&'
𝑟∗ = (1)
(' ∆&
Here, r∗ represents the critical radius; σ denotes
the interfacial tension between water and ice
and has an estimated value of 22 erg/cm2; Tf
denotes saline-water freezing temperature in
Kelvin; Lf denotes latent heat of fusion; and ΔT
represents temperature difference with respect
to the freezing point. Post completion of the
crystallization step, ice slurry is separated from
concentrated brine. Subsequently, the ice slurry
is melted to obtain fresh water as the final prod-
uct Crystallization is the most important of the
three steps involved in the FD process. Nuclea-
tion is the critical step in FD, since it denotes
the starting point of crystallization. Whether the
process is already having crystal or not, nuclea-
tion can be divided into two types— (1) primary
and (2) secondary. Primary nucleation forms
crystals within solutions not containing any pre-
existing crystals while secondary nucleation oc-
curs in the presence of pre-existing crystals
[50,51]. Primary nucleation covers homoge-
nous as well as heterogeneous nucleation. How-
ever, primary nucleation in industries is more
likely heterogeneous nucleation rather than ho-
mogenous nucleation [52]. Spontaneous nu-
cleus form within the solution during homoge-
nous nucleation, and this is caused by molecular
accretion [46,50]. Furthermore, heterogeneous
nucleation occurs on foreign particles and/or
impurities present within the solution [51]. In
the presence of solutes and/or impurities, a
freezing-point depression occurs, which can be
expressed using the following equation
salt solubility line, a small amount of salt can be
observed to start precipitating [54]. If solution
“A” keeps being cooled down, the solution will
become supersaturated with salt upon cooling,
and thus, the eutectic point is reached. At this
point of temperature and concentration, the
largest amount of salt is possibly dissolved in
the solution [53]. Heat transfer between the so-
lution and refrigerant and mass transfer of salt
molecules across the solution–ice-crystal inter-
face plays critical roles in the FD process. Fig.
4 illustrates the simplest indirect FD process
conducted inside a vessel equipped with a cool-
ing jacket. As seen, a solid–liquid interface ex-
ists during the freezing process. Assuming that
crystallization and heat transfer occur in one di-
mensional direction, crystallization or phase
change begins at x=0 (representing solid–liquid
interface) and is a function of time. Heat trans-
fer, ice growth rate, and nucleation rate model-
ing are summarized in Table 1. Meanwhile,
during the direct FD process, refrigerant droplet
size must also be considered, since it directly
affects the heat transfer process. Equations to be
used are summarized in Table 2.

∆𝑇 = −𝐾-. 𝑚 (2)

Here, ΔTf denotes the depression in freezing

point; KFP denotes the cryoscopic constant,
which equals 1.86 kg/mol for water; and m de-
notes molality of the solution. Typical seawater
contains an NaCl concentration of 35 g/kg,
which in turn, leads to freezing-point depres-
sion of the order of approximately −2.1oC. The
more the ice crystals that form within the solu-
tion, the higher is the concentration of the re-
maining solution. This behaviour tends to fur-
ther depress the freezing point of the solution.
For example, consider an existing saline solu-
tion “A” having a certain concentration and
maintained at a certain temperature (Fig. 3). It
Fig 3: Salt-solution phase diagram [53,54]
is allowed to cool until it reaches the ice line,
and at that instant, a small portion of ice begins
to form. If there exists another saline solution
“B” that being cooled down until it reaches the

4. Freeze Crystallisation Process
This section provides an account of extant re-
searches performed concerning direct FD, indi-
rect FD, and hybrid systems coupled with the
FD process.
4.1 Direct freeze Desalination
mass production of fresh water from saline so-
lution seemed largely impractical at first. With
a view to making FD an economical yet effi-
cient process, Gilliland suggested recovery of
latent heat of fusion of ice and elimination of
indirect FD process through use of water-im-
miscible refrigerants, such as Freon, propane,
or butane. Thenceforth, the direct FD process
attracted significant research interest. Hirakawa
and Kosugi [55] considered direct contact FD
as the most effective FD mode, and employed
butane and Freon as refrigerants, which are in-
soluble in seawater but evaporate when brought
in contact with seawater, which in turn, forming
ice crystals.
4.1 Gas hydrate
The gas hydrate or hydrate freezing process was
envisioned as being very similar to direct con-
tact FD [56]. The process essentially involves
nonstoichiometric crystalline aggregation of
hydrogen-bonded water molecules around a
central gas molecule [56,57]. Feed water and
gas are mixed at a certain condition of temper-
ature and pressure to facilitate hydrate for-
mation. Addition of gaseous or liquid hydrocar-
bon and CO2 as formation or guest gases causes
hydrates to form at temperatures above the
freezing point of ice [58]. It has been stated that
gas hydrate formation yields solution freezing
points as high as 12 oC [56]. The state-of-the-
art of gas hydrate desalination processes is sum-
marized in Table 4.
Table 1: Equations used in heat transfer and rate
calculation in indirect FD
4.2 Indirect Freeze Crystallisation
and refrigerant. In other words, there exists a
physical barrier tending to hinder the heat trans-
fer process between the refrigerant and saline
solution. Indirect FD, in general, can be classi-
fied into suspension freeze crystallization and
layer freeze crystallization [17,59]. Suspension
freeze crystallization is a well-known process
employed in the food industry; however, it is
not efficient in terms of cost and simplicity of
process and equipment [17,60]. However, inde-
pendent control over ice nucleation and crystal
growth is difficult to achieve [6]. Section 4.2.1
describes enhancement of suspension freeze
crystallization processes through experiments.
Meanwhile, with regards to layer freeze crystal-
lization, formation of ice crystals on the cold
layer results in layer-by-layer crystallization
[17,60]. Layer freeze crystallization employs
simple equipment design and is more beneficial
in terms of separation owing to smaller surface
area per unit mass of ice obtained[59]
4.3 Suspension Freeze Crystallisation
The suspension freeze crystallization concept is
usually presented in the form of a scraped-sur-
face heat exchanger or jacketed kettle vessel
[6,61]. Table 5 summarizes latest research
trends concerning suspension freeze crystalli-
zation.
4.4. Layer freeze crystallization
The layer-by-layer crystallization phenomenon
in cold-plate freezing leads to formation of
purer ice crystals, since lesser impurities are
trapped between ice layers. This method can
further be classified into three types—progres-
sive freezing, static layer crystallization and dy-
namic layer crystallization.
4.5. Progressive freezing.
Solutes at the ice-solution interface are moving
from one end of a vessel which is contacted
with refrigerant to the other end, [62]. The feed
solution is contained within a tube, and the tube
is slowly immersed into the refrigerant (refer
Fig. 5). Consequently, this method enables con-
centration of impurities in the liquid phase and
facilitates extraction of pure ice crystals by con-
trolling one-dimensional ice growth [63]. Table
4.6. Static layer crystallization.
When crystallization is achieved through a stag-
nant solution, it is called static layer crystalliza-
tion [144,145]. Basically, the cold surface is im-
mersed into the solution, without any move-
ment until the ice layer forms on the cold sur-
face. The cold surface itself can be vary from
cylindrical to a flat surface. Experiments that
had been conducted on static layer crystalliza-
tion are shown in Table 7.
4.7. Dynamic layer crystallization.
Falling film freezing, refers to a technique
wherein the solution to be concentrated is main-
tained in contact with a cooled vertical surface
[17,59,150]. In other words, fluid is made to
flow down while maintaining contact with a
chilled surface, over which crystallization and
crystal growth occur. Few experiments per-
formed with focus on the falling film freezing
method However, not many experiments have
been performed concerning falling film freez-
6 summarizes current research trends concern-
ing this method.
ing desalination. This method is more com-
monly employed in food industries and related
researches.
4.8 FD hybrid system
Rahman, Ahmed and Chen [45] stated that there
exists high desalination potential in combining
the freezing process with other desalination
methods, thereby developing a synergistic hy-
brid system. Madani [66] proposed use of a
zero-discharge direct-contact freezing desalina-
tion method combined with a solar evaporator
(DCF–SE) to reduce environmental impact of
concentrated brine solutions rejected from sea-
water desalination plants. Products extracted
from this hybrid system included fresh water,
sodium, magnesium and potassium salts, and
bromine. Bromine was recovered by means of a
stripping process involving steam prior to the
extraction of pure water by employing the
DCF–SE process. For fresh-water recovery of
up to 80%, 12.31–13.78 kWh/m3 of energy was
consumed depending on the amount of heat re-
jection. Additionally, an area measuring nearly
116 km2 was required to evaporate the blow
down of the Arabian Gulf desalination plant.
Hanafi [67] proposed a wind freezing–melting
process to reduce the energy consumed during
a desalination process. Wind power was used to
supply energy to the compressor and cooling-
coil system, which in turn, supported a vac-
uum–freeze desalination system. This hybrid
system was expected to produce fresh water
supply limited to 3000m3/ day.
Wang and Chung [68] suggested a hybrid de-
salination process comprising indirect contact
FD (ICFD) combined with direct contact mem-
brane distillation (DCMD). DCMD can be
readily integrated with existing desalination
processes, since it is not a pressure-driven pro-
cess and involves minimum capital cost. In ad-
dition, compared to other desalination pro-
cesses that require membrane installation,
DCMD is less affected by high salt concentra-
tions. In this method, feed seawater was first
treated using the ICFD system and resulted in
extraction of fresh water and brine rejection.
The ICFD brine was subsequently treated by
means of the DCMD process. Consequently,
use of this hybrid process enabled attainment of
71.5% water recovery closely matching drink-
ing water standards.
5. Current challenges and development
Given the present state-of-the-art, up-scaling
the FD process to the industrial level would
cause several problems, such as excessive en-
ergy requirements, lower quality of water prod-
uct, as well as difficulty in achieving precise
nucleation control. Development of a solution
to each of these problems is discussed is this
section.
5.1 Energy consumption
The refrigeration requires high grade energy
consumption such as electricity, FD process
could, therefore, become much more attractive
if alternate sources of energy involving lower
costs could be utilized. Antonelli [69] was the
first to propose utilization of LNG cold
energy for desalination processes, thereby
providing mechanical energy by means of an
organic Rankine cycle while also serving as a
cooling system for N-butane during secondary
nucleation. Liquefied natural gas is re-gasified
prior to its utilization. Regasification of LNG
absorbs approximately 837 kJ/kg energy.
Wang and Chung [68] studied desalination per-
formed using the hybrid FD–MD system pow-
ered by LNG cold energy. In this case, LNG
was employed to cool the secondary refrigerant,
placed inside the FD crystallizer, and MD dis-
tillate. By means of the HYSYS software, Cao,
Lu, Lin and Gu [70] simulated the FD process
while utilizing LNG cold energy. A dynamic
model was also developed and simulated using
the gPROMS software. It was concluded that 1
kg equivalent LNG cold energy was roughly
equivalent to extraction of 2 kg of fresh water.
Recently, Lin, Huang and Gu [71] investigated
a FD system utilizing LNG cold energy. In their
study, R401A and liquid nitrogen were em-
ployed as the secondary refrigerant and LNG
replacement, respectively. Production capacity
of the process corresponded to 150 L of fresh
water per hour and 2 kg of fresh water per kg of
LNG. Salt rejection was observed to be 50% for
a single cycle.
5.2 Quality Enhancement
For the purpose of purification or quality en-
hancement, three posttreatment techniques—
centrifugation, sweating, and washing—are
commonly employed. Han, Shin, Rhee and
Kang [72] demonstrated that use of a centrifuge
is capable of lowering salinity to as low as
0.25%. Implementation of this method, how-
ever, is too expensive to be suitable for mass
desalination applications.
Vaessen [73] was able to attain ice with purity
parameters equivalent to<50 ppm of K+
and<0.1 ppm of H+ and Mg2+ from KNO3–
HNO3 solution after three washing cycle.
Reddy, Kramer, Lewis and Nathoo [74] con-
cluded that the purity of ice could be improved
by up to 90% through crystallization of 5%
Na2SO4 solution. Besides, washing is expected
to be effective in enhancing purity by removing
impurities on crystals surface. Washing, there-
fore, comes across as one of the most promising
post treatments for the FD processes. However,
the amount of fresh water required during
washing to achieve potable water standards re-
mains relatively high till date. In order to release
pockets of concentrated brine trapped within ice
crystals, sweating can also be considered as an
effective post-treatment technique[15]. Mandri,
Rich, Mangin, Abderafi, Bebon, Semlali, Klein,
Bounahmidi and Bouhaouss [75] investigated
the sweating process through one cycle of indi-
rect FD. It was observed that salinity of the sa-
line solution was substantially reduced from 35
g/kg to 0.5 g/kg at 00C post optimization of the
freezing condition and sweating step. Rich,
Mandri, Mangin, Rivoire, Abderafi, Bebon,
Semlali, Klein, Bounahmidi, Bouhaouss and
Veesler [76] stated that sweating serves sub-
stantially improve ice quality (purity), and that
sweating temperature and duration play a sig-
nificant role in the purification process. In their
study, the decrease in salinity was observed to
be significant at the beginning but gradually
settled down to a nearly constant value after 3
or 4 h. Operation at a sweating temperature
higher than optimum yielded similar reduction
in salinity. Han, Shin, Rhee and Kang [72] ap-
plied sweating at 9 0C for 3 min and observed a
10% increase in the efficiency of salt removal.
One drawback observed during his study was
that a large number of ice crystals had melted
(owing to higher temperature) to achieve lower
salinity in remaining ice crystals. Rich, Mandri,
Mangin, Rivoire, Abderafi, Bebon, Semlali,
Klein, Bounahmidi, Bouhaouss and Veesler
[76] melted nearly half of the ice gained to pro-
duce drinking water. Thus, further investiga-
tions to achieve major improvements in the
sweating process need to be performed. Addi-
tion of ice seeds to induce nucleation is often
considered to enhance product quality as well
as achieve better control over nucleation. With-
out ice seed addition, the temperature of the
system quickly drops below the freezing point
and the system attains a super cooled state. At
large degrees of super cooling, dendritic ice
crystals should be formed to release the latent
heat. At the end of the super cooling stage,
greater impurities get trapped between dendritic
ice crystals, which in turn, imply lower quality
of water. Addition of ice seeds causes super
cooling state to terminate immediately, and ice
crystals are formed shortly thereafter
5.3 Nucleation Control
Super-cooled crystallization is spontaneous and
stochastic, thereby making it difficult to predict
and control crystallization time and product
quality[77-79]. With regards to direct seeding,
there exist some physical methods to have a bet-
ter control over nucleation, such as application
of power ultrasound, electric fields, and micro-
wave irradiation. Power ultrasound has fre-
quency in the range of 20–100 kHz and propa-
gates through a medium in cycles of compres-
sions and rarefactions; the local pressure, there-
fore, continuously varies. Gas bubbles form
owing to liquid fracture, and subsequently, dur-
ing creation of a local vacuum, the air that first
dissolved into water diffuses to form bubbles.
During cycles of negative and positive pres-
sures, bubbles tend to expand and contract, re-
spectively [77]. However, after few cycles, the
bubbles finally collapse. This phenomenon is
referred to as acoustic cavitation. Hickling's
theory states that collapse of cavitation bubbles
generates high pressure exceeding 5 GPa in lo-
cal zones for very short period [80], thereby re-
sulting in higher degree of super cooling in lo-
cal zones. Nonetheless, under bulk temperature,
nucleation occurs under lower degrees of super
cooling. Application of an electric field to in-
duce nucleation, or electro freezing, serves to
precisely adjust nucleation temperature in pure
water to a desired value between the freezing
point and temperature corresponding to occur-
rence of spontaneous nucleation. Orlowska,
Havet and Le-Bail [81] demonstrated that suffi-
ciently high electrostatic fields possess 100%
probability of inducing ice nucleation when ap-
plied for 30 s at various super cooling degrees.
Kacker, Radoiu and Kramer [82] applied mi-
crowave heating as a means of nucleation con-
trol. By manipulating the number of particles in
the system, measured using focused beam re-
flectance measurement (FBRM), crystalize dis-
tribution could also be controlled.
6. Future Promising Applications
The term “desalination” on FD does not mean
that freezing process is limited on desalination
area. The core of freezing process in saline so-
lution treatment is highly relied on fresh water
and its contaminant separation, and since FD
has high tolerance for feed concentration, thus
it is favourable to broaden FD application area.
Its application has high potential to be more de-
veloped, eventually. More application of FD
will be described below.
6.1 Brine Management
Desalination is major technology to counter wa-
ter scarcity. However, concentrated brine is
produced from both thermal and membrane-
based desalination. RO as major reliable desal-
ination method to produce fresh water, occu-
pied 44% share in capacity and 80% share in
number of desalination plant [83]. However,
RO generates concentrated brine in the range of
1 to 25% of the feed [84]. It was estimated the
cost of brine disposal is 5–33% of total cost of
Desalination[85]. sometimes limited by fouling
[189,196]. Beside to desalinate water, freezing
process—FD or freezing crystallization can
also be applied to treat concentrated brine from
various sources. Highly saline solution is not a
problem for freeze crystallization, because it
will even closer to the eutectic concentration
[83]. Coupling direct FD with RO to treat RO
brine from inland desalination plant had been
proposed and successfully reduced 90% of
brine at same rate of water production, com-
pared to RO plant without further treatment of
FD [86].
6.2 Resource Mining
Freeze crystallization can be a solution to the
problem of metal recovery from seawater,
brine, or other saline solutions. As depicted in
Fig. 3, if a saline solution is continuously
cooled after reaching its eutectic point, ice and
salt crystals form simultaneously. Thus, valua-
ble minerals from the concentrated brine or
other saline solutions could be effectively re-
covered. Separation of ice from salt becomes
very easy due to significant density difference
[87]. Randall, Nathoo and Lewis [88] investi-
gated freezing crystallization to treat waste
from e Malahleni Water Reclamation Plant in
South Africa. In addition to potable water re-
covery, they succeeded in obtaining pure sul-
phates of calcium (98% purity) and sodium
(96.4%). Certain brines, however, contain more
salts that crystallize at the same temperature. In
such a case, additional treatment would be re-
quired to achieve salt separation. Randall,
Nathoo and Lewis [46] attempted to manipulate
the system to crystallize only one type of salt by
adding seed salt crystals. However, in their re-
search, only sodium sulphate seeding was ob-
served to promote the crystallization of the de-
sired product while magnesium sulphate seed-
ing failed. In the light of the above results, use
of freeze crystallization must be promoted in
brine and/or seawater mining processes; how-
ever, there still exists great potential for further
research concerning the application of this tech-
nique on an industrial scale.
8. Conventional Cooling Technology
Refrigeration and cryogenic technology play a
very important role in improving modern living
standards and work environment. Cooling tech-
nology has covered a wide range of aspects
from domestic air conditioning and refrigerator
to industrial gas liquefaction and scientific re-
search. Vapour-compression-type heat ma-
chines are at present time most often applied for
artificial refrigeration using evaporation-con-
densation and compression-expansion motor-
driven mechanical compressor provides com-
pression and expansion in these devices. Today
vapour-compression technology that has al-
ready been improved for over one hundred
years has practically been achieving its upper
limit of development, and further essential im-
provement of its power efficiency could hardly
be expected thermodynamic cycles, the electri-
cal power consumption for cooling in its vari-
ous forms accounts for more than 15% of total
use of electric energy [1]. ]. This consumption
can reach 30% in developed countries [2,3]. son with the traditional gas-compression refrig-
even the best commercial conventional refrig-
erator units can only attain 40% of ideal Carnot
efficiency [5,6]. Therefore, improving the en-
ergy efficiency of refrigeration technology is of
vital importance for the control of energy con-
sumption. On the other hand, the traditional
cooling techniques use some harmful refriger-
ants. For example, chlorofluorocarbons (CFCs)
and hydro chlorofluorocarbons (HCFCs) were
used as refrigerant gases, but they were found
to be highly damaging to the ozone layer that
protects the earth from the intense rays of the
sun. In order to achieve this goal, the participat-
ing countries sought to limit the emission of
greenhouse gases by using more zero-carbon
technologies. Consequently, the search for new
refrigeration technology with high energy effi-
ciency and environmental friendly is urgent.
9. Magneto calorific effect
The magneto-caloric effect (MCE) is an in-
trinsic advanced magneto-thermodynamic phe-
nomenon of the magnetic material in which the
temperature is changed by exposure of the ma-
terial to a magnetic field [11–13] Total entropy
S of the magnetic material is contributed by
three parts, i.e., the magnetic entropy DSM, the
electronic entropy DSe , and the lattice entropy
DSl [14,15]. The spins will align parallel to the
magnetic field when a magnetic material is
magnetized adiabatically which, in turn, lowers
the contributions of DSM and DSe . Because to-
tal entropy of the system stays constant under
an adiabatic condition, the lattice contribution
DSl increases oppositely to keep total S con-
stant. This leads to the temperature increase due
to the enhancement of lattice vibrations. Vice
versa, the removal of the field causes spin ran-
domization that results in an elevation of mag-
netic entropy and a decrease of lattice entropy;
then the temperature of magnetic material de-
creases [16]. Based on MCE, magnetic refriger-
ation has been developed as a promising novel
cooling technology, including high efficiency
air conditioning in large buildings. In compari-
eration, magnetic refrigeration is advantageous
in several aspects [14,17,18]: (i) the magnetic
refrigeration materials and water-based heat
transfer medium have no serious environmental
problems; (ii) greater efficiency of a magnetic
refrigerator comparing to conventional ones
[6,19] (in case of maximum cooling power val-
ues); for example, in [20], for a magnetic refrig-
erator working at RT the value of cooling power
(768 W) and 60% of ideal Carnot cycle effi-
ciency are reached when a temperature span is
about 20 K (see Table 2 in [20]); (iii) the mag-
netic refrigerator can be more compact and less
noisy due to the use of solid materials and evad-
ing the compressor. Therefore, magnetic refrig-
eration has attracted worldwide interest. His-
torically, MCE has been discovered by Weiss
and Piccard in 1917 who found that the change
of magnetic field caused a temperature change
in nickel around its Curie temperature (TC)
[23,24]. Later, Debye [25] and Giauque [26] in-
dependently proposed that MCE can attain very
low temperature by adiabatically demagnetiz-
ing paramagnetic salts. In 1933, the first MCE
application was experimentally demonstrated
by Giauque and MacDougall to achieve a tem-
perature of 0.25 K by adiabatic demagnetization
of Gd2(SO4)3•8H2O [27]. Although the mag-
netic refrigeration gained much progress at low
temperatures, the near-room temperature mag-
netic refrigeration had not been reported until
1976, when Brown demonstrated that a large
temperature gradient of 46 K can be obtained
under the magnetic field change of 7 T in pure
Gd around 294 K [28]. Since then, efforts have
been put to the room temperature magnetic re-
frigeration. In 1990, Nikitin et al. [29] reported
MCE of Fe49Rh51 alloy which is the largest
MCE ever achieved (12.9 K in Δµ0H = 1.95 K)
[30]. The closest to this result values, obtained
from direct measurements, are presented in [31]
(9.2 K in 1.9 T). In [31] the estimations of ΔT
(10,5 – 12 K in 1,9 T), obtained from ΔSM data
i.e. from indirect measurements are also pre-
sented. It should be noted that the magnetic
properties and MCE in FeRh alloys strongly de-
pend on structural defects, preparation proce-
dures, heat treatment, measurements conditions
[32–35], therefore, the available experimental
data presented in literature can vary greatly.
Two major breakthroughs occurred in 1997.
The Ames laboratory and Astronautics Corpo-
ration of America developed a proof-of-princi-
ple room temperature magnetic refrigeration
device with maximum cooling power of 600 W
at temperature span 10 K, magnetic field 5 T
and flow rate 5 L/min [18]. It is worth noting
that with a further increase of temperature span,
the cooling power decreases linearly, thus at a
temperature span of 22 K the cooling power is
only 150 W. Furthermore, Pecharsky and
Gschneidner reported the gigantic MCE in
Gd5(Si, Ge)4 compounds around room temper-
ature which exhibit a significantly higher MCE
and tunable working temperature compared
with Gd [36,37]. The discovery of large MCE
in Gd5(Si, Ge)4 alloys dramatically accelerated
the field of magnetic refrigeration worldwide
Though the adiabatic demagnetization of para-
magnetic salts has become standard technique
to attain very low temperatures, the magnetic
refrigeration is still unavailable for liquefaction
of nitrogen and hydrogen due to the lack of suit-
able magneto-caloric materials [53]. In recent
years, some materials have been found to pre-
sent large MCEs around the liquefaction tem-
perature of nitrogen and hydrogen. Importantly,
excellent working bodies for this temperature
range are rare earth orthoaluminates with a per-
ovskite structure, such as DyAlO3 and DyxEr1-
xAlO3, that have advanced properties than gar-
nets such as Gd2Ga3O12, Dy2Ga3O12,
Dy2Ga3O12 [54–58]. Herein we review recent
progress on the magnetic refrigeration in the
temperature range of nitrogen and hydrogen liq-
uefaction (20-77 K) and about magnetic cooling
The review does not consider the helium tem-
perature range which will be analysed else-
where.
10. Magnetic Cooling Advantages
The development of a high-efficiency magnetic
refrigerating machine (magnetic refrigerator)
with a superconducting magnetic field source
operating in the room temperature range is cur-
rently an actual scientific and engineering prob-
lem. high power efficiency (according to theo-
retical estimates, the magnetic refrigerator effi-
ciency is about 85 % of Carnot cycle in the tem-
perature range from 150 to 300 K),
- less overall dimensions and weight as com-
pared to vapour-compression devices due to
considerably greater density of solid body ver-
sus gas,
- low wearability due to low operating fre-
quency of magnetic heat machine,
-safety and ecological compatibility (working
body is a nontoxic material, no chemical
coolants are used).
11. Magnetic Cooling Disadvantages
The disadvantages of magnetic cooling devices
include:
- higher cost of magnetic refrigerators (expen-
sive sources of magnetic field and, possibly,
magneto-caloric material of the working body);
- the presence of a magnetic field source may
impose restrictions on possible applications of
magnetic refrigerators.
Mentioned advantages will allow solving the
problems of vapour-compression machines (re-
frigerators) and entirely replacing with time the
vapour-compression refrigerating technology
with the magnetic one. Being appropriately
adapted, the magnetic refrigerating machines
can be used in the same areas, where the va-
pour-compression machines are applied. Cur-
rently, R&D on magnetic heat (refrigerating)
machine designing take place at research cen-
tres in most of the developed countries such as:
USA (University of Ames, Iowa (Lee et al.
2002; Zarkevich et al.,2018; Zimm et al., 1998).
12. Magnetic field sources for a magnetic re-
frigerator: permanent magnets,
LTSC,HTSC
The designs of magnetic heat machines that
have been suggested and patented by now are
based on the schemes with a magnetic field
source assembled using permanent magnets or
implemented using a superconducting solenoid
made of traditional superconductors operating
in the helium temperature range (low-tempera-
ture superconductors – LTSC) (Yu et al., 2010).
It should be noted that the systems based on per-
manent magnets provide high-intensity fields
(of about 1.5 T level) only in a narrow operating
gap of about 1 to 2 cm width (Bjørk et al.,
2010).This makes it impossible to place suffi-
cient quantity of working body to the working
space of the magnetic field source, which re-
duces refrigerating capacity of the whole de-
vice.
For generating magnetic field of high strength
(over 10 T), superconducting magnetic systems
have been used for over half a century manu-
factured based on superconducting Nb-Ti and
Nb-Sn wires which transition to superconduct-
ing state requires cooling to the temperatures
corresponding to liquid helium, i.e. these are
LTSC. However, from the point of view of
overall expenditures, usage of LTSC field
sources to create a magnetic refrigerating ma-
chine does not seem to be promising. Liquid he-
lium is quite an expensive consumable and it re-
quires to be gathered to gaseous state after its
transition. In addition to this, refrigerating sys-
tems based on two-stage cryogenic refrigerators
(cryocoolers) consume essential power, which
brings to nothing the advantages of magnetic
refrigeration. Today superconducting wires are
actively developed in the world that are based
on so called “high-temperature” superconduc-
tors (HTSC) characterized by relatively high
temperatures of transition to superconducting
state (critical temperatures) – from 90 K to 110
K.
HTSC materials based on perovskite ceramics
make it possible to provide critical current den-
sity of up to 10 MA/cm2 at 77K temperature.
Most promising are second-generation HTSC
wires in the form of thin (40 to 100 µm thick
and 4 to 12 mm wide) metal carrier ribbon with
highly textured buffer and superconducting ox-
ide layers of up to 5 µm total thickness depos-
ited on it. These ribbons are covered with cop-
per layer to provide capability of soldered elec-
trical connections. Usage of HTSC ribbons en-
ables to create solenoids with field intensity of
1.5 to 2 T in considerably greater working gap
(of about 1 l volume) using liquid nitrogen cry-
ogenic system, or using a single-stage
cryorefrigerator. Only one attempt has been
made till now to use HTSC field source in mag-
netic refrigerating machines (Blumenfeld et al.,
2002a). Currently, there are no commercial pro-
totypes of magnetic refrigerating machines with
HTSC magnetic field sources in the world mar-
ket.
13. Active Magnetic Regenerator
Different thermodynamic cycles that require
appropriate design solutions are used in mag-
netic heat machines. The analysis of available
literature devoted to magnetic heat machines
(Allab et al., 2005; Chiba, 2017; Kitanovski et
al., 2015; Kulkarni, 2015; Luiz Dutra et al.,
2017; Yu Figure 1 Kinds of working body ge-
ometry (regenerator types): (a) regenerator with
close packing of particles; (b) regenerator with
longitudinal channels in a massive block (ma-
trix); (c) regenerator composed of a set of flat
plates (plate regenerator); (d) regenerator con-
sisting of a set of perforated plates. zet al.,
2010,2010; Zimm et al., 2018, 1998) shows that
all currently known devices designed for cool-
ing from the room temperature range operate
following active magnetic regenerator (AMR)
cycle. In this cycle, the magnetic material itself
is used as a regenerator, and heat exchange dur-
ing regeneration process takes place between
the magnetic material and heat carrier, which
prevents technical difficulties and losses due to
the use of an individual regenerator as it takes
place in the devices with Erickson and Brighton
regenerative cycles.
Active magnetic regenerator is a container
filled with a magnetic material in the form that
allows passing of heat carrier through the re-
generator. Magnetic material which geometry
corresponds to the regenerator kind is com-
monly called a working body. Figure 1 depicts
possible kinds of working body and the types of
magnetic regenerators corresponding to them.
Figure 1 Kinds of working body geometry (re-
generator types): (a) regenerator with close
packing of particles; (b) regenerator with longi-
tudinal channels in a massive block (matrix);
(c) regenerator composed of a set of flat plates
(plate regenerator); (d) regenerator consisting
of a set of perforated plates.
Figure 1a shows the regenerator with close
packing of particles. The container of such a
regenerator is densely filled with magnetic ma-
terial particles. Here the particles of both spher-
ical and arbitrary shape can be used. Heat car-
rier flow in the regenerator with close packing
of particles passes through hollow space of the
close packing. Figure 1b shows a regenerator
with longitudinal channels in the massive block
of material. Here liquid flow passes through the
channels along the long regenerator face. In the
plate regenerator (Figure 1c) the working body
consists of flat plates with long edges parallel to
the regenerator longitudinal dimension
stackedin such a way that to keep gaps between
the plates for heat carrier to pass through. Re-
generator consisting of a set of perforated plates
(Figure 1d) has the plates with the holes located
across the heat carrier movement. Regenerator
container shall not necessarily be of a cylindri-
cal shape as itis shown in Figure 1 Regular-
shape containers of uniform cross-section are
selected as a rule.
Cylindrical and rectangular regenerator con-
tainers are used in AMR heat machines (Yu et
al.,2010). Some results of theoretical and exper-
imental investigations of the various AMR fill-
ings are presented in (Chiba, 2017; Keawkam-
rop et al., 2018; Nakashima et al., 2018a,
2018b; Shi etal., 2018; You et al., 2018).Be-
sides regenerators for AMR machines shown in
Figure 1, regenerators with working material in
the form of a grid (grid regenerator) and thin
ribbon (foil) winded as a coil (ribbon regenera-
tor) are applied in cryogenic refrigerating de-
vices – Figure 2 (Ackermann, 1997).
Figure 2 Regenerators of cryogenic refrigerat-
ing devices: (a) with working material in the
form of grid (grid regenerator); (b) grid element
of grid regenerator (– inner diameter of regen-
erator container); (c) with working material in
the form of coiled ribbon (ribbon regenerator)
(Ackermann, 1997) As it can be seen in Figure
2, the grid regenerator is essentially similar to
the perforated plate regenerator shown in Fig-
ure 1d.
Regenerator geometric configuration can be
characterized using a number of parameters,
including the following (Ackermann, 1997):
- matrix surface area;
- regenerator clear opening area;
- regenerator working material (matrix) poros-
ity (porosity factor);
- regenerator matrix surface density ;
- regenerator matrix hydraulic radius ;
- regenerator matrix hydraulic diameter .
Non-regenerative Carnot cycles are used only
in temperature range below 20 K, but even for
this temperature range AMR cycles are used
(Bézaguet et al., 1994; Hakuraku and
Ogata,1985; Jeong, 2014). Creation of a spe-
cific magnetic refrigerator prototype requires
implementation of its components and assem-
blies (magnetizing-demagnetizing system,
magnetic regenerator with working body made
of magnetic material, heat carrier transfer sys-
tem).
14. AMR Working Body
Literature data analysis (see Tables 1 - 4) shows
that gadolinium and its alloys with terbium,
dysprosium, erbium, and yttrium are mainly
used in currently existing prototypes of mag-
netic refrigerating machines operating in room
temperature area. Compositions based on lan-
thanum-ferrum-silicon intermetallic compound
are applied in several prototypes, and in one
prototype - Pr0,65Sr0,35MnO3 manganite (Le-
gait et al., 2014).Today lanthanum-ferrum-sili-
con-based compounds (La(Fe13−xSix)) are
considered to be quite promising from the point
of view of their application in magnetic refrig-
erating machines operating in the room temper-
ature range (GschneidnerJr et al., 2005).
Gd5(Si1−xGex)4 and MnFe(P1−xAsx) com-
pounds are also promising (GschneidnerJr et
al., 2005). However, these materials have the
first-order magnetic phase transition; such ma-
terials reveal hysteresis both in field depend-
ences of magnetization and changes of the mag-
netic part of entropy ΔSM (H). Nevertheless, it
is important to note that work is underway to
obtain such alloys with partial substitution of
elements (Fu et al., 2018; Gębara et al., 2017),
which can lead to a change in the mechanisms
of transition and reduce the hysteresis value.
For example, such materials as La(FeSiCo)H
that are only slightly into the first order transi-
tion regime have been proposed (Fuet al.,
2018), and that’s why these materials can have
very small hysteresis. Hysteresis in M(H) -
causes additional heating due to hysteresis loss,
and hysteresis in ΔSM (H) – cycle equilibrium
disturbance. These both processes degrade re-
frigerating machine performance. Therefore,
materials with the first-order phase transition
are undesirable to be used as working body of
magnetic heat machines. In addition, the tech-
nology of manufacturing such materials is quite
difficult, and the materials themselves possess
high hardness and brittleness, which makes dif-
ficult their processing in order to have them in
the form needed for their use in magnetic refrig-
erators. In the course of magnetic refrigerating
machine usage brittle materials will crumble,
and their highly hard particles will be carried
away by heat carrier flow and get into valves
and pump leading to their rapid wear. To pre-
vent this effect, brittle materials are coated with
protective polymer film, but such coating dete-
riorates heat exchange between the material and
heat carrier bringing to nothing the advantages
of the applied compounds. It is worth noting
that the magnitude of the heat transfer blockage
can be decreased by reducing the thickness of
polymer coating.
Currently, magneto-caloric effect in manga-
nite’s, including Pr1-xSrxMnO3, is rather inten-
sively investigated (GschneidnerJr et al., 2005).
At the same time, it should be noted that this
system does not feature essential values of mag-
neto caloric effect (anyhow, it does not
exceed that observed in gadolinium) (Chen et
al., 2000; Chen and Du, 2001) though the se-
cond order magnetic phase transition takes
place in it; so, application of Pr0,65Sr0,35MnO3
compound by the authors of (Legait et al., 2014)
is reasoned most likely by research interest. In
addition, manganite are ceramics having high
hardness, so they have the above-mentioned
disadvantages typical for very hard and brittle
materials.
15. Heat Carrier Transfer System
Data show that both liquids and gases are used
as a heat carrier in AMR heat machines operat-
ing in the range of room temperature values.
Most often water as well as water-based mix-
tures (water – ethyl alcohol, water – antifreeze,
water – ethylene glycol) are used as a liquid. If
the device is designed to be operated at below 0
°С temperature, additives are used to decrease
water freezing temperature. Anticorrosive addi-
tives were also added to water to prevent work-
ing material oxidation (1 % water solution of
caustic (NaOH) was used for this purpose in
(Cheng et al., 2013)). Besides water, oils were
also used – silicone and olive. Liquid provides
quite good thermal contact with magnetic ma-
terial surface in regenerator and hence high heat
transfer factor between heat carrier and mag-
netic material necessary for the efficient opera-
tion of regenerator and heat machine as a whole.
It was noted in (Kitanovski and Egolf, 2010)
that liquid metals have maximum heat-transfer
factor for metal surface – liquid system. Due to
this fact it was suggested to use them as a heat
carrier in AMR heat machines. It should be
noted that the only metal being in liquid state at
room temperature is mercury (melting temper-
ature is -38.8 °С), however due to its extreme
toxicity practical use of this substance is rather
difficult. Currently, intensive research in the
field of the development of nano-liquids con-
taining nanoparticles of various substances is
carried out, including nanotubes (Saidur et
al.,2011). Such liquids feature increased heat
conductivity and heat-transfer factors in com-
parison with traditional liquids used as heat car-
riers. In particular, it is reported that adding of
0.4 % fraction of nanoparticles to initial liquid
can increase its thermal conductivity by 40 %
(Saidur et al., 2011). So, usage of Nano liquids
in AMR heat machines as a heat carrier should
be recognized to be promising.
As it follows from Tables 1 - 4, besides liquids
used as a heat carrier in AMR heat machines
there are also applied gases – nitrogen and he-
lium, the latter is used most often. It should be
noted that gases are used at sufficiently high
pressures – up to 10 atm, because of their low
heat capacity and heat transfer at atmospheric
pressure; so, use of gases assumes additional
technical difficulties connected with high pres-
sure. Therefore, use of water-based liquids and
water solutions of antifreeze substances should
be acknowledged to be preferable. As it can be
seen from data given if Tables 1 through 4, two
types of devices are applied to generate heat
carrier flow through the AMR heat machine op-
erating loop: displacer and circulating pump.
Displacer is a plunger that is mechanically
moved using a rod in the cylinder divided by the
plunger into two parts. Today the displacers are
broadly used in cryocoolers. The advantage of
the displacer is high speed of operation and the
possibility to arrange AMR heat machine based
on the scheme without any additional switching
valves decreasing the reliability of the device as
a whole and leading to extra heat losses. At the
same time essential disadvantage of the operat-
ing scheme of AMR heat machine with dis-
placer is some “dead volume” of heat carrier
that constantly exists in the regenerator while
the machine operation and thus does not partic-
ipate in heat exchange processes in hot and cold
heat exchangers. Presence of such a dead vol-
ume can essentially decrease the efficiency of
the device as a whole. There is no such disad-
vantage for the schemes where heat carrier is
transferred along the loop in one direction by a
pump, and flow switching during operation cy-
cle is implemented using flow direction valves.
16. Magnetizing-Demagnetizing of a Work-
ing Body
Magnetic heat machines can be divided into
two groups as to the method of magnetizing de-
magnetizing of the working body. The first
group includes the devices where magnetiza-
tion (demagnetization) of the working body is
implemented by placing the working body into
the operating gap of the magnetic system (tak-
ing out of the working body from the operating
gap of the magnetic system). The second one –
the devices where magnetization-demagnetiza-
tion is implemented by switching on/off of cur-
rent in the magnetic field source winding (the
working body itself rests here). In the first
group, placing of the working body into the area
of exposure to magnetic field can be executed
either by the field source movement (here the
working body rests), or by the working body
moving (here the field source rests).

17. Switching operation mode of a supercon-

ducting field source.

The method where magnetization-demagneti-

zation is provided by means of switching the
current in the magnetic field winding with the
rest (stationary) working body was applied in
magnetic refrigerating devices operating based
on Carnot magnetic cycle in the range of cryo-
genic temperatures below 20 K – such devices
are called refrigerators with adiabatic demag-
netization (AD refrigerators). Figure 3 depicts
AD refrigerator general block diagram. The de-
vice consists of a switched magnetic field
source that is a superconducting solenoid made
of a low-temperature superconductor (LTSC)
and two thermal valves that provide thermal
contact at isothermal segments of Carnot cycle
with cooled heat load and heat sink and isolat-
ing the working body at the segments corre-
sponding to adiabatic magnetization and de-
magnetization.
Such scheme was implemented, in particular, in
(Hakuraku and Ogata, 1985), where a supercon-
ducting Nb-Ti solenoid was used to generate
the field of 3 T intensity. Solenoid operated in
pulse mode and provided 0.3 Hz operating fre-
quency of the device. And the average rate of
field build-up was of about 3.8 T/s, maximum
value at the initial segment was about 9 T/s.
LTSC solenoid was used in AD refrigerator
suggested in (Bézaguet et al., 1994). Solenoid
provided maximum field strength of 3.5 T at the
average rate of field increase/decrease while
magnetizing/demagnetizing of 17.5 T/s. Con-
siderably less field rates of change were used in
AD refrigerator in (Kashani et al., 1996), where
LTSC solenoid with maximum field of 6.5 T
provided field change while demagnetization at
the average rate of 0.04 T/s, and the complete
AD refrigerator cycle achieved 690 s. Low
value of the field rate of change (0.07 T/s for
maximum field value of 2 T) generated by
LTSC superconducting solenoid was used in
AD refrigerator in (Bartlett et al., 2015).
Figure 3 General block diagram of a magnetic
refrigeration device with a stationary working
body and switched magnetic field source oper-
ating based on Carnot magnetic cycle (AD re-
frigerator).
18. Magnetic field change in a working body:
motion of a working body or a field source.
The next group of magnetic heat machines in-
cludes the devices where magnetization/de-
magnetization is implemented by placing/tak-
ing out of magnetic working body (regenerator)
to/from the area of exposure to magnetic field
source. Such devices can be implemented by
two methods:
- working body rests (stationary working
body), field source moves relative to the work-
ing body (Figures 4a and 4c);
- field source rests (stationary field source),
working body moves relative to the source (Fig-
ures 4b and 4d). Two types of movement are
possible in both cases:
- linear (back-and-forth motion), when the
field source or the working body executes
periodic linear motion along a certain direction
(Figures 4a and 4b); - rotational, when the field
source or the working body executes rotational
movement.

Figure 4 Magnetic heat machines with magnet-
izing-demagnetizing by means of moving
working body (regenerator) and magnetic field
source: (a) machine with stationary working
body and linear movement of magnetic field
source; (b) machine with stationary magnetic
field source and linear movement of working
body; (c) machine with stationary working
body and rotational movement of magnetic
field source; (d) machine with stationary mag-
netic field source and rotational movement of
working body.
19. Linear motion of a working body and a
stationary magnetic field source: examples
Superconducting field sources require cooling
of solenoid winding to cryogenic temperatures
(4.2 K for LTSC and 40 K and higher for
HTSC) that is provided either using cryogenic
liquids or cryocooler (sometimes cryocooler is
used in closed cooling systems in combination
with cryogenic liquids). A cryostat is required
in both cooling methods for thermal insulation
of the cooled solenoid from the environment.
So, a superconducting field source with a refrig-
erating system is quite a bulky device which it
is rather difficult to move especially when it is
rotational movement (difficulties grow if cryo-
genic liquid is used for cooling). The analysis
of data on magnetic heat machines available in
literature (Arnold et al., 2011; Hirano et al.,
2002; Kitanovski et al., 2015; Kulkarni, 2015;
Richard et al., 2004; Rowe and Tura, 2006;
Rowe and Barclay, 2002; Tura, 2005; Yu et al.,
2010; Zimm et al., 2018, 1998) and patent bases
showed that the superconducting sources are
not used in the machines with a moving field
source. Stationary superconducting field
sources have also not yet been used in the ma-
chines based on turning (rotational) schemes.
Thus, the superconducting field sources are cur-
rently applied only in the magnetic heat ma-
chines built based on the scheme of the linear
back-and forth motion of the working body (re-
generator) with a stationary field source.
Several AMR magnetic heat machine proto-
types implemented based on linear back-and-
forth motion of regenerator and stationary su-
perconducting magnetic field source are de-
scribed in literature (Arnold et al., 2011; Hirano
et al., 2002; Richard et al., 2004; Rowe and
Tura, 2006; Rowe and Barclay, 2002; Tura,
2005; Zimm et al., 1998).
Figure 5 depicts the block diagram of AMR
heat machine described in (Zimm et al.,1998).
The machine includes a stationary magnetic
field source with a warm operating gap based
on LTSC (Nb-Ti) cooled by liquid helium, a
pneumatic linear drive, the system of switching
the direction of heat carrier flow, a pump, cold
and hot heat exchangers, composed regenerator
(R1 and R2), and a pipeline connecting all the
machine components and forming its operating
loop where heat carrier circulates. A composite
regenerator is used in this work that includes
two identical containers that are by turns placed
into the working gap of the field source. Such
an approach enables to compensate partially the
force needed to remove the regenerator from
the field source. In particular, while removing
R1 regenerator from the operating gap of the
field source when moving upward, R2 regener-
ator that becomes exposed to the field source
impact will be drawn into it under the effect of
ponder motive forces that will at the same time
hinder R1 regenerator from being removed
from the operating gap. So, ponderomotive
forces acting on R1 and R2 will be partially
compensated decreasing the force that the drive
must provide in the course of the device opera-
tion and reducing the power consumed by the
drive. It is possible to optimize the effect of
ponderomotive forces by means of selection of
a distance between the containers. Besides
compensation of ponderomotive forces, the
scheme with the composite regenerator includ-
ing two containers which dimensions fit the op-
erating gap of the magnetic field source enables
to double refrigerating capacity versus a single-
container regenerator scheme. Each of the con-
tainers of AMR machine (Zimm et al., 1998)
contained 1.5 kg of spherical gadolinium parti-
cles of 0.15 to 0.3 mm diameter; and maximum
force gained by the drive when moving regen-
erators was 2500 N. Field rate of change while
remagnetization was about 5 T/s.
Figure 5 Block diagram of AMR heat machine
presented in (Zimm et al., 1998). AMR heat
machine suggested in (Hirano et al., 2002) was
of approximately similar design. The device
also used a composite regenerator consisting of
two containers (8 cm long, of 6.2 cm inner di-
ameter) each of that contained 1.1 kg of spheri-
cal gadolinium particles of 0.3 mm, diameter
moved by a hydraulic linear drive. Magnetic
field of 4 T value was generated by LTSC mag-
netic field source with a warm operating hole of
10 cm diameter cooled by contact method using
Gifford-McMahon cryocooler. Maximum force
while moving regenerators was 1600 N. Unlike
the above-mentioned machines from (Hirano et
al., 2002; Zimm et al., 1998), the back-and-
forth motion in AMR heat machine described in
(Richard et al., 2004; Rowe and Tura, 2006;
Rowe and Barclay, 2002) was provided using
an electric motor-driven crank-and-rod mecha-
nism
Figure 6 Block diagram of back-and-forth
AMR heat machine with a crank-and-rod mech-
anism.
A magnetic field source was designed based on
a solenoid made of LTSC Nb-Ti wire
(solenoid inductance was 105 H), which tem-
perature is maintained at the level of 4.2 K by
the contact method using Gifford-McMahon
cryocooler. The solenoid was placed into a vac-
uum cryostat. Current leads were made of
HTSC. The field rate of change while magneti-
zation remagnetization was 2.5 T/s. The dimen-
sions of one regenerator were as follows: 2.5
cm diameter, 8.8 cm length. The majority of
back-and-forth AMR heat machines are the de-
vices with magnetic field sources designed
based on permanent magnets (Allab et al.,
2006; Balli et al., 2012; Bouchekara et al.,
2012; Bour et al., 2009; Cheng et al., 2013; Clot
et al., 2003; Czernuszewicz et al., 2014; Dupuis
et al., 2009; Engelbrecht et al., 2009, 2011;
Hirano et al., 2009; Kawanami et al., 2006; Kim
and Jeong, 2009; Legait et al., 2014; Lu et al.,
2005; Romero Gómez et al., 2013; Sari and
Balli, 2014; Tagliafico et al., 2009, 2013; Tre-
vizoli et al., 2011; Tušek et al., 2013, 2014; Yao
etal., 2006; Zheng et al., 2009).
20. Rotational motion of a working body or
a magnetic field source: examples.
One more large group of AMR heat machines
are those with turning (rotational) movement of
field source or working body (turning (rota-
tional) AMR heat machines) (Aprea et al.,
2014; Arnold et al., 2014; Bahl et al., 2014;
Bohigas et al., 2000;Engelbrecht et al., 2012;
He et al., 2013; Jacobs et al., 2014; Lozano et
al., 2014, 2013; Okamura etal., 2007a, 2005b;
Shir et al., 2005; Tura and Rowe, 2011, 2009b,
2007a; Tušek et al., 2010; Vasile and Muller,
2006, 2005; Zimm et al., 2007, 2006, 2005). As
it can be seen, operating frequencies of the de-
vices on permanent magnets are within the
same range as for the machines with supercon-
ducting magnet field sources: 0.025 – 1.5 Hz.
Field rates of change in the permanent magnet-
based devices are within the range from 0.2 to
3.5 T/s, which also agrees with the values for
the devices with superconducting field sources
(0.2 to 5 T/s, And the devices with permanent
magnets are built both based on the scheme
with stationary magnet field source and with
stationary regenerator. The latter ones are pos-
sible in the case of permanent magnet-based
sources due to the compactness of such sources.
It should be noted that field sources based on
permanent magnets are characterized by less
field values in operating gaps , maximum field
value does not exceed 1.65 T. In addition, high
fields in the permanent magnet-based sources
can be generated only in rather narrow operat-
ing gaps (Bjørk et al., 2010). In particular, op-
erating gap of the magnetic field source with
maximum magnetic field strength (1.65 T) is 1
cm. Less volume of working material can be
placed into narrow gaps, which leads to less re-
frigeration capacity values at the same values of
operating frequency.
The analysis of available literature revealed nei-
ther rotational-type device with superconduct-
ing magnetic field source used. At the same
time, a rotational-type device with a stationary
electromagnet as a magnetic field source was
suggested in (Coelho et al., 2009). Figure 7
shows the block diagram of the device.
Figure 7 Block diagram of rotation AMR heat
machine with electromagnet.
Figure 7 shows that in rotational AMR heat
machines both stationary and rotating field
sources based on permanent magnets are used
(field sources or regenerators in rotational
AMR heat machines are electric motor driven),
and the field in the operating gap of the source
does not exceed 2 T (is within 0.77 to 2 T inter-
val). At the same time operating frequency of
such machines is higher than that of the back-
and-forth devices (0.25 to 10 Hz versus 0.025
to 1.43 Hz, see Table 3), which provides their
higher efficiency (Yu et al., 2010).
The analysis of available literature revealed nei-
ther rotational-type device with superconduct-
ing magnetic field source used. At the same
time, a rotational-type device with a stationary
electromagnet as a magnetic field source was
suggested in (Coelho et al., 2009). Figure 7
shows the block diagram of the device. Figure
7 Block diagram of rotation AMR heat machine
with electromagnet. It should be noted that an
electromagnet is not an acceptable field source
for magnetic heat machines since it consumes
too much power to generate sufficient fields in
the operating gap for the machine operation,
and the operating gap remains here quite nar-
row. In particular, about 3 kW power is required
to generate the field of about 1.5 T in about 2
cm gap.
21. Conclusion
Increasing water demand with limited fresh wa-
ter availability source would soon lead our
planet into acute water scarcity. Since seawater
forms the major source of water, desalination
comes across as the leading technology with a
great potential to meet future water supply de-
mands. As of today, RO is the most reputable
technology in a majority of desalination plants
around the world. FD is an emerging technol-
ogy that offers significant advantages over ther-
mal and membrane- based desalination pro-
cesses. Based on refrigerant location during the
FD process, it can be classified into direct and
indirect FD processes. Major investigations
preformed concerning both FD processes have
laid major focus on operating conditions and
use of the right apparatus to achieve potable wa-
ter standards. Investigations concerning post-
treatment methods to enhance ice-crystal purity
have also been performed. Upscaling the FD
process requires consumption of large amounts
of high quality energy to produce the low tem-
perature by the refrigeration cycle. Combining
FD with LNG regasification plant could solve
the problem concerning energy consumption,
thereby making FD more attractive by lowering
its operational cost. In terms of quality enhance-
ment, centrifugation, washing, and sweating are
techniques proposed to be applied post crystal-
lization to boost product quality. However,
these post-treatment techniques still require fur-
ther investigations, since these are not yet fully
effective in terms of cost, large water-handling
capacity, and ice loss that tends to affect overall
plant productivity. Upscaling the FD process
also requires better control, since nucleation
demonstrates a probabilistic occurrence. Vari-
ous methods, such as seeding and application of
power ultrasound, electric fields, and micro-
wave heating, have been investigated as means
of effective nucleation control to enhance end-
product quality by effectively reducing the de-
gree of super cooling. Apart from desalination
(which simply involves fresh water recovery),
freeze crystallization can also be employed in
applications concerning mineral recovery from
concentrated brine produced in desalination
plants. Some minerals contained in seawater are
available in large concentrations, and post de-
salination, these minerals tend to get further
concentrated. Thus, energy needed will be re-
duced due to less volume to be treated. In addi-
tion, mineral recovery in desalination plants
tends to reduce the amount of waste generated
as well as the cost of water recovery. However,
there still exist problems associated with selec-
tive mineral recovery, which must be resolved
before the process could be employed at an in-
dustrial level. The FD process would, therefore,
become more attractive if it can be proven to
perform efficiently for both, desalination as
well as metal recovery.
21.1 Magnetic Cooling
Use of superconducting magnetic system
(SMS) as a magnetic field source for a solid
state magnetic heat pump provides a number of
advantages. It can be stated that currently the
niche of SMS-based magnetic heat machines
remains poorly developed, and new technical
solutions that do not require the development of
sophisticated devices but at the same time ena-
ble to extend essentially the range and capacity
of refrigeration are of great practical interest.
The most optimal ways of magnetization-de-
magnetization are the following: switching of
operating current in the winding of the mag-
netic field source and mechanical moving of re-
generator with working body relative to

Fig5: Progressive Freezing apparatus

Fig 6: Mineral content comparison

 Table 3:
Reference Equipment Sample Performance

Landau and Martindale
Gibson, Emmermann,
Grossman, Johnson,
Modica and Pallone
Antonelli
Xie, Zhang, Liu, Lv, Ruan and
Hosseini
Stirred crystallizer, 9-inch diameter cylindrical
tank with discharge port located 6 in. above
base. Tank was equipped with an impeller
comprising two sets of four blades. Orifices
were installed at tank bottom inside an annular
duct to serve as butane inlet.
Sectional crystallizer, 6-in. diameter column,
separators were placed at 3-in. intervals. The
setup comprised six major compartments and
one overflow compartment. Each compartment
was equipped with an impeller comprising four
removable baffles.
Draught-tube crystallizer, 9-inch diameter
vessel measuring 17 in. high, 6-inch diameter
tube suspended 1 in. above vessel base.
Spray freezer measuring 0.9 m and 1.2 m in
diameter and height, respectively.
Cylindrical vessel
(mechanical energy required was supplied by
means of an organic Rankine cycle (ORC))
Jacketed cylinder equipped with nozzles
(leaned at 45 °C) and air diffusers.
Brine water prepared using rock salt and
town water.
Brine water prepared using rock salt and
town water.
Brine water prepared using rock salt and
town water
Seawater
Seawater
Inclined nozzle positions led to spiral flow of
the refrigerant, thereby promoting
turbulence and enhancing heat transfer,
which ultimately led to better separation of
ice on the surface of refrigerant droplets. Ice,
so produced, was collected from the upper
part of the ice generator.
Ice salinity was found dependent on stirring rate
and turbulence caused by refrigerant and crystal
dispersion. Reduced local sub-cooling led to
lower nucleation rates.
Baffle installation led to decreased salinity with
increase in stirring rate by breaking down
crystals, thereby increasing their surface area.
This implied higher contamination of
concentrated brine on crystal surface.
Excessive nucleation rates were observed along
with poor crystal quality. Super cooling of the
feed was realized 4 °C prior to the occurrence of
nucleation. To avoid this situation, crushed ice
(as ice seed) was added as solution temperature
approached the freezing point. This helped
reduce the degree of super cooling to 0.2 °C, and
overall ice quality was improved.
Large-sized ice crystals were formed in the first
compartment, and it was, therefore, necessary
to double the size of the first compartment to
ensure bulk vaporization of butane.
A tendency to form agglomerates on continuous
operation was observed.
With appropriate washing, water with salinity
measuring < 500 ppm could be achieved.
High production rate under low driving force
during low-power operation, absence of moving
parts, and relative ease during scaling up owing
to design simplicity and efficient unit.
Each hour, 113.4–136 kg of ice could be
produced with low Freon residues remaining
within the slurry.
An ice-cake layer was formed at the bottom
owing to insufficient turbulence.
3
The system ($0.96/m ) was more economical
compared to an SRF plant supported by an
3
electrical drive ($1.28/m ).
The design prevented formation of ice cakes at
the bottom.
Critical velocity of droplets measured 0.35 m/s.

Table 4
Latest research trends concerning gas hydrate desalination.

References Systems Results

McCormack and Andersen,
McCormack and Niblock
Park, Hong, Lee, Kang, Lee, Ha
and Lee
Niu
Kang, Linga, Park, Choi and Lee
Karamoddin and Varaminian
Han, Rhee and Kang
Fakharian, Ganji and Naderifar
Preliminary study of a clathrate (gas hydrate) desalination plant using
HCFC R141B (CH 3 CCl2 F) as refrigerant, which demonstrates a critical
decomposition temperature of 11.6 °C (0 Pa), along with calculated
production cost.
A new design was developed to continuously produce gas-hydrate
pellets. The apparatus capacity measured 1.8 L, and CO 2 was used as
forming gas.
Desalination of produced water containing NaCl, CaCl2 , and MgCl2 by
forming gas hydrate and subsequently dewatering it. Operating
temperatures ranged from −1 °C to 1 °C (3.1–5.5 MPa) while using
CO2 as the forming gas.
Evaluated the effect of hydrate-based desalination of actual seawater
samples (obtained from southeast coast of Korea) on removal
efficiency of cations and anions for the first time using CO 2 and CH 4 as
guest gases.
Gas hydrate desalination was performed using tetrahydrofuran (THF)
at 17:1 M ratio.
Gas hydrate treatment of seawater obtained from the east coast of the
Jeju Island in Korea. Cyclopentane (3% mol) was used as hydrate
former, and simple filtration and washing processes were
subsequently performed.
CO2 gas hydrate formation at 3.5 MPa and 1.05 °C with different
produced water samples.
3
Cost of operation was observed to be $0.46–0.52/m with public
3
financing and $0.59–0.68/m with private financing.
Up to 78.7%, 76.6%, 72%, 80.4% and 73.3% of Na, Mg, Ca, K, and B
were removed, respectively.
99% of NaCl and MgCl2 were removed from the solution.
CO 2 , when used as the guest gas, yielded better results compared to
CH4; cation and anion removal efficiencies of the order of 71–94%
and 73–83%, respectively, were observed.
THF hydrate formation demonstrated removal efficiencies of the
order of 52.6% (4.5 wt%) and 58.5% (3.0 wt%) for NaCl solution
and 70.6% (1.5 wt%) and 66.2% (1.9 wt%) for KCl solution.
By means of simple filtration, 63% of salt ions were removed, and
subsequent washing yielded a further 42% increase in efficiency.
A three-stage gas hydrate formation process resulted in a removal
efficiency of the order of 82–89.2% with the aid of an appropriate
washing process. The temperature of washing water was
maintained at 1.05 °C.

Table 5: Current research
trends in suspension freeze
crystallization.

References Equipment Samples Performances

Vaessen
Lewis, Nathoo, Thomsen,
Kramer, Witkamp, Reddy
and Randall
Reddy, Lewis, Witkamp,
Kramer and van Spronsen
Mtombeni, Maree,
Zvinowanda, Asante,
Oosthuizen and Louw
Liu, Hao, Guo and Rao
Lin, Huang and Gu
Scraped cooled wall crystallizer (SCWC) having
a capacity of 115 L and equipped with four
rotating scrappers attached to the cooling wall.
Scraped cooled wall crystallizer (SCWC) having
a capacity of 12 L.
Scraped cooled wall crystallizer (SCWC) having
a capacity of 12 L.
HybridICE unit comprising a refrigeration unit
and scraped surface heat exchangers (SSHEs).
The first heat exchanger was designed for
nucleation while the second was designed for
the ripening process.
Scraped surface heat exchanger (SSHE)
comprising two concentric cylinders of different
sizes with scraped blades installed between
them.
Flake ice maker, wherein a rotary ice blade was
used to scrape ice layers off the heat exchanger
surface.
KNO3 and HNO3 solutions
RO retentates in South Africa containing
high amounts of sodium, chlorine,
sulfate, and ammonia.
Hypersaline brine wastewater in South
Africa containing 4% NaSO4 along with a
synthetic RO retentate mimicked by
concentrated NaCl solution
Synthetic brine sample containing
common salt
NaCl solution
Seawater
Ice crystals extracted from the KNO3 and HNO3
solutions, after washing each solution three times,
+ +
contained < 50 ppm of K and < 0.1 ppm H
2+
and Mg ions, respectively.
From the Na2SO4–brine solution, pure ice crystals
containing < 20 ppm impurities were obtained
after seven washing steps; very pure
Na2SO4.10H2O was also recovered.
Pure ice crystals and sodium sulfate stream
containing 20 wt% NaCl were obtained. Potential
for sequential removal of salts from brine was also
reported owing to recovery of > 90% pure
Na2SO4 extracted from the concentrated NaCl
solution.
Average salt removal measured 96% in accordance
with chloride analysis and 98.5% as per
conductivity measurements. Average energy
consumption measured 7.45 kWh per ton of ice
extracted.
Four different temperature stages that matched the
theoretical model were observed during
crystallization.
A mathematical correlation between phase-change
latent heat and solution concentration was also
determined.
Salt removal rate of only 50% was observed, and
more cycles of the FD process or further treatment
with RO were required to enhance salt removal.

Table 6: Recent research trends concerning progressive freezing.

References Equipment Samples Performances

Wakisaka, Shirai and Several aluminum cylinders containing
Sakashita treated solution and packed inside the
crystallizer. Flon C318 as coolant was made
to flow outside cylinders.
Miyawaki, Kato and Tubular ice system along with several
Watabe aluminum cylinders containing treated
solution and packed inside the crystallizer.
Fujioka, Wang, A 52-mm of diameter cylindrical vessel
Dodbiba and circulated with ethylene glycol maintained
Fujita at −20 °C.
Glucose solution as a substitute for actual
wastewater
Sucrose, diethyl ether, butyl aldehyde, ethyl
acetate, ethanol, methyl butanoate, 1-propanol,
butyl acetate, hexyl aldehyde, 1-butanol, hexyl
acetate, 1-hexanol, and diethyl ether.
NaCl solution
Coolant temperature was set at −2.5 °C for the first
10 min and subsequently adjusted to −11 °C for the
next 30 min; washing water was then pumped for 30 s
followed by flow of gaseous coolant around cylinders
to detach the ice formed.
2
Approximately 578.6 kg/m h of productivity was
gained in a single batch operation lasting 70 min.
Effects of operating conditions on the partition
constant of a solute (K) were determine via
concentration polarization modeling, thereby
demonstrating that slower ice-growth and higher
circulation rates are preferable for efficient tubular ice
systems.
Ice-front speed of 0.5 cm/h and stirring velocity of
1.45 m/s were determined as optimum.

Table 7
Researches on static layer crystallization.

Reference Equipment Sample Performance

Kuiper, Lankin, Gauthier, Walker and Davies
Hirata, Matsuzaki and Ishikawa
Rich, Mandri, Bendaoud, Mangin, Abderafi,
Bebon, Semlali, Klein, Bounahmidi,
Bouhaouss and Veesler
Eisenbart and Ulrich
Htira, Cogné, Gagnière and Mangin
Brass hollow cold finger AFP
solution
Vertical cooled polyvinyl-
chloride plate
Stainless steel tube immersed
in a jacketed glass tank
Stainless steel cold finger
inside a 600 ml double-
walled beaker
Vertical stainless steel
immersed in cylindrical tank
E. coli lysate supernatant
Ethylene glycol solution
NaCl solution and seawater
from the Rabat
1–10 wt% glycerol in
water, with addition of 1-
butanol
Water – acetone
Concentrated antifreeze protein in concentration of 1.78 mg/L
from initial solution of 0.77 mg/L.
Vertical plate performed as better as horizontal plate, but the ice
was easier to remove thus it became preferable for practical use.
There was liquid inclusion contaminated the ice and this can be
handled by applying lower growth rate. Rabat seawater was
almost purified to drinking water standard in 31 h.
Solvent-aided layer crystallization was successful to purify
glycerol from water without any post-treatment (washing and
sweating).
Acetone content was reduced to 3.92 g/L from initial 50 g/L with
growth rate of 0.1 mm/h.
Table 8
Researches on dynamic layer crystallization.
Reference
Rich, Mandri, Mangin, Rivoire,
Abderafi, Bebon, Semlali, Klein,
Bounahmidi, Bouhaouss and
Veesler
Jiang, Hou, He and Wang
Williams, Ahmad and Connolly
Zambrano, Ruiz, Hernández, Raventós
and Moreno
Equipment
Stainless steel tube immersed in a jacketed
glass tank
Crystallizer installed with sparger on the top
and equipped with circulator bath
The ice-maker machine comprised a
refrigerated plate (maintained at −14 °C)
inclined at 195° above the collecting tank
having a capacity of 3.5 L.
The system comprised refrigerated plate,
thermally insulated chamber, and cylindrical
jacketed vessel.
Sample
NaCl solution and seawater
from the Atlantic Ocean and
Mediterranean Sea.
Analytical grade phosphoric
acid (85–91 wt%)
NaCl solution, Arabian Gulf
seawater, and RO rejected
brine.
Saline solution prepared using
commercial grade salt.
Performance
The three experiments achieved purity levels up to
drinking water standard over an 8-h cycle (5 h
freezing and 3 h sweating at 0 °C) with the initial feed
temperature set at −2.4 °C; the sweating step was
applied to melt impure zones.
Half of ice layer thickness grew during the first 1/3
crystallization time.
39.3% salinity reduction in terms of concentrating RO
brine and maximum water recovery of 50% were
achieved.
Salt removal efficiency of the order of 98.5% was
achieved through five stages of falling freezing and
sweating and three stages of the block freezing
process.
References:
[1] P.G. Youssef, R.K. Al-Dadah, S.M.
Mahmoud, Comparative analysis of desalina-
tion technologies, Energy Procedia 61 (2014)
2604–2607.
[2] L. Agno, M. Juanico, Physical Geography,
Goodwill Trading Co., Inc., Sta. Cruz, Manila,
1987.
[3] A. Al-Karaghouli, L.L. Kazmerski, Energy
consumption and water production cost of con-
ventional and renewable-energy-powered de-
salination processes, Renew. Sust. Energ. Rev.
24 (2013) 343–356.
[4] M. Barlow, T. Clarke, Blue Gold: The Fight
to Stop the Corporate Theft of the World's Wa-
ter, New Press, 2011.
[5] I. GWI, IDA Desalination Year Book 2017–
2018, (2018).
[6] M.S. Rahman, M. Ahmed, X.D. Chen,
Freezing-melting process and desalination: I.
Review of the state-of-the-art, Sep. Purif. Rev.
35 (2006) 59–96.
[7] J. Morillo, J. Usero, D. Rosado, H. El
Bakouri, A. Riaza, F.-J. Bernaola, Comparative
study of brine management technologies for de-
salination plants, Desalination 336 (2014) 32–
49.
[8] Rognon, F., 2005. Efficiency potentials of
heat pumps with combined heat and power. For
maximum reduction of CO2 emissions and for
electricity generation from fossil fuels with
CO2 reduction in Switzerland (No. SFOE--
250179). Swiss Federal Office of Energy
(SFOE).
[9] Srikhirin, P., Aphornratana, S., Chungpai-
bulpatana, S., 2001. A review of absorption re-
frigeration technologies. Renew. Sustain. En-
ergy Rev. 5, 343–372.
https://doi.org/10.1016/S1364-0321(01)00003-
X [21] H. El-Dessouky, I. Alatiqi, S. Bingulac, H.
[10] Riffat, S.B., Ma, X., 2003. Thermoelec-
trics: a review of present and potential applica-
tions.Appl.Therm.Eng.23,913–935.
https://doi.org/10.1016/S1359-4311(03)00012-
7
[11] Gigiel, A.J., 1996. Air cycle refrigeration.
Proc. Inst. Refrig. 92, 1.
[12] J. Chang, J. Zuo, K.-J. Lu, T.-S. Chung,
Freeze desalination of seawater using LNG cold
energy, Water Res. 102 (2016) 282–293.
[13] W. Cao, C. Beggs, I.M. Mujtaba, Theoret-
ical approach of freeze seawater desalination
on flake ice maker utilizing LNG cold energy,
Desalination 355 (2015) 22–32.
[14] A. Brunetti, F. Macedonio, G. Barbieri, E.
Drioli, Membrane engineering for environmen-
tal protection and sustainable industrial growth:
options for water and gas treatment, Environ.
Eng. Res. 20 (2015) 307–328.
[15] P.M. Williams, M. Ahmad, B.S. Connolly,
Freeze desalination: an assessment of an ice
maker machine for desalting brines, Desalina-
tion 308 (2013) 219–224.
[16] R.D.C. Shone, The freeze desalination of
mine waters, J. South. Afr. Inst. Min. Metall. 87
(1987) 107–112.
[17] K. El Kadi, I. Janajreh, Desalination by
freeze crystallization: an overview, Int. J. of
Thermal & Environmental Engineering 15
(2017) 103–110
[18] I.S. Al-Mutaz, I. Wazeer, Comparative
performance evaluation of conventional multi-
effect evaporation desalination processes,
Appl. Therm. Eng. 73 (2014)
1194–1203.
[19] H. Baig, M.A. Antar, S.M. Zubair, Perfor-
mance evaluation of a once-through multistage
flash distillation system: impact of brine heater
fouling, Energy Convers. Manag. 52 (2011)
1414–1425.
[20] T. Mezher, H. Fath, Z. Abbas, A. Khaled,
Techno-economic assessment and environmen-
tal impacts of desalination technologies, Desal-
ination 266 (2011)
263–273.
Ettouney, Steady-state analysis of the multiple
effect evaporation desalination process, Chem.
Eng. Technol. 21 (1998)
437–451.
[22] M. Al-Sahali, H. Ettouney, Developments
in thermal desalination processes: design, en-
ergy, and costing aspects, Desalination 214
(2007) 227–240.
[23] J.W. Wu, A study of silica gel adsorption
desalination system, School of Mechanical En-
gineering, The University of Adelaide, Aus-
tralia, 2012.
[24] K. Bourouni, R. Martin, L. Tadrist, M.T.
Chaibi, Heat transfer and evaporation in geo-
thermal desalination units, Appl. Energy 64
(1999) 129–147.
[25] I. Houcine, M. Benamara, A. Guizani, M.
Ma.lej, Pilot plant testing of a new solar desali-
nation process by a multiple-effect-humidifica-
tion technique, Desalination 196 (2006) 105–
124
[26] H. Sharon, K.S. Reddy, A review of solar
energy driven desalination technologies, Re-
new. Sust. Energ. Rev. 41 (2015) 1080–1118.
[27] A. Midilli, T. Ayhan, Natural vacuum dis-
tillation technique—part I: theory and basics,
Int. J. Energy Res. 28 (2004) 355–371.
[28] D. Zejli, R. Benchrifa, A. Bennouna, O.K.
Bouhelal, A solar adsorption desalination de-
vice: first simulation results, Desalination 168
(2004) 127–135.
[29] X. Wang, K.C. Ng, Experimental investi-
gation of an adsorption desalination plant using
low-temperature waste heat, Appl. Therm. Eng.
25 (2005) 2780–2789.
[30] K. Thu, H. Yanagi, B.B. Saha, K.C. Ng,
Performance analysis of a low-temperature
waste heat-driven adsorption desalination pro-
totype, Int. J. Heat Mass Transf. 65
(2013) 662–669.
[31] G.P. Narayan, M.H. Sharqawy, J.H.
Lienhard V, S.M. Zubair, Thermodynamic
analysis of humidification dehumidification de-
salination cycles, Desalin. Water Treat. 16
(2010) 339–353.
[32] Y.-D. Kim, K. Thu, K.C. Ng, Adsorption
characteristics of water vapor on ferroalumino-
phosphate for desalination cycle, Desalination
344 (2014) 350–356
[33] S.A. El-Agouz, Desalination based on hu-
midification–dehumidification by air bubbles
passing through brackish water, Chem. Eng. J.
165 (2010) 413–419.
[34] E. Chafik, A new type of seawater desali-
nation plants using solar energy, Desalination
156 (2003) 333–348.
[35] E. Chafik, Design of plants for solar desal-
ination using the multi-stag heating/ humidify-
ing technique, Desalination 168 (2004) 55–71.
[36] P.F. Varadi, F. Wouters, A.R. Hoffman,
The Sun is Rising in Africa and the Middle
East: On the Road to a Solar Energy Future, Pan
Stanford Publishing, 2018.
[37] A. Bennett, Membranes in industry: facili-
tating reuse of wastewater, Filtr. Sep. 42 (2005)
28–30
[38] W. Guo, H.-H. Ngo, J. Li, A mini-review
on membrane fouling, Bioresour. Technol. 122
(2012) 27–34.
[39] H.S. Abid, D.J. Johnson, R. Hashaikeh, N.
Hilal, A review of efforts to reduce membrane
fouling by control of feed spacer characteris-
tics, Desalination 420 (2017) 384–402.
[40] S.G. Arhin, N. Banadda, A.J. Komakech,
I. Kabenge, J. Wanyama, Membrane fouling
control in low pressure membranes: a review on
pretreatment techniques for fouling abatement,
Environ. Eng. Res. 21 (2016) 109–120.
[41] J.-J. Qin, M.H. Oo, K.A. Kekre, B. Liber-
man, Development of novel backwash cleaning
technique for reverse osmosis in reclamation of
secondary effluent, J. Membr. Sci. 346 (2010)
8–14.
[42] M.M. Motsa, B.B. Mamba, J.M. Thwala,
A.R.D. Verliefde, Osmotic backwash of fouled
FO membranes: cleaning mechanisms and
membrane surface properties after cleaning,
Desalination 402 (2017) 62–71.
[43] L.K. Wang, J.P. Chen, Y.T. Hung, N.K.
Shammas, Membrane and Desalination Tech-
nologies, Humana Press, 2010.
[44] J.E. Boysen, J.A. Harju, C. Rousseau, J.
Solc, D.J. Stepan, Evaluation of the natural
freeze-thaw process for the desalinization of
groundwater from the North Dakota aquifer to
provide water for Grand Forks, North Dakota,
U.S. Bureau of Reclamation Water Treatment
Technology Program Report No. 23, 1999.
[45] M.S. Rahman, M. Ahmed, X.D. Chen,
Freezing-melting process and desalination: re-
view of present status and future prospects, Int.
J. Nucl. Desalin. 2 (2007) 253–264.
[46] H.M. Curran, Water desalination by indi-
rect freezing, Desalination 7 (1970) 273–284.
[47] W. McCabe, J. Smith, P. Harriott, Unit Op-
erations of Chemical Engineering, McGraw-
Hill Education, 2005.
[48] N.E.R. Zimmermann, B. Vorselaars, D.
Quigley, B. Peters, Nucleation of NaCl from
aqueous solution: critical sizes, ion-attachment
kinetics, and rates, J. Am. Chem. Soc. 137
(2015) 13352–13361.
[49] B.W. Tleimat, Chapter 7 - freezing meth-
ods A2 - SPIEGLER, K.S, in: A.D.K. Laird
(Ed.), Principles of Desalination, Second Edi-
tion, Academic Press, 1980, pp. 359–400.
[50] D.G. Randall, J. Nathoo, A. Lewis, Seed-
ing for selective salt recovery during 2009, pp.
639–646.
[51] M. Dalvi-Isfahan, N. Hamdami, E. Xan-
thakis, A. Le-Bail, Review on the control of ice
nucleation by ultrasound waves, electric and
magnetic fields, J. Food Eng. 195 (2017) 222–
234.
[52] X. Liu, K. Zhuang, S. Lin, Z. Zhang, X. Li,
Determination of supercooling degree, nuclea-
tion and growth rates, and particle size for ice
slurry crystallization in vacuum, Crystals 7
(2017) 128.
[53] B.J.J. Verbeek, Eutectic freeze crystalliza-
tion on sodium chloride, Master Thesis, TU
Delft, Netherlands, 2011.
[54] M.J. Fern.ndez-Torres, F. Ruiz-Bevi., M.
Rodr.guez-Pascual, H. von Blottnitz, Teaching
a new technology, eutectic freeze crystalliza-
tion, by means of a solved problem, Educ.
Chem. Eng. 7 (2012) e163–e168.
[55] S. Hirakawa, K. Kosugi, Utilization of
LNG cold, Int. J. Refrig. 4 (1981) 17–21.
[56] J.E. Miller, Review of water resources and
desalination technologies, Sandia National
Labs Unlimited Release Report SAND-2003-
0800, 2003.
[57] K.-n. Park, S.Y. Hong, J.W. Lee, K.C.
Kang, Y.C. Lee, M.-G. Ha, J.D. Lee, A new
apparatus for seawater desalination by gas hy-
drate process and removal characteristics
of dissolved minerals (Na+, Mg2+, Ca2+, K+,
B3+), Desalination
274 (2011) 91–96.
[58] S. Han, Y.-W. Rhee, S.-P. Kang, Investi-
gation of salt removal using cyclopentane
hydrate formation and washing treatment for
seawater desalination, Desalination
404 (2017) 132–137.
[59] M. Wakisaka, Y. Shirai, S. Sakashita, Ice
crystallization in a pilot-scale freeze
wastewater treatment system, Chem. Eng. Pro-
cess. Process Intensif. 40 (2001) 201–208.
[60] M. Gunathilake, M. Dozen, K. Shimmura,
O. Miyawaki, An apparatus for partial ice-melt-
ing to improve yield in progressive freeze-con-
centration, J. Food Eng. 142 (2014) 64–69.
[61] M.B. Lakhdar, R. Cerecero, G. Alvarez, J.
Guilpart, D. Flick, A. Lallemand, Heat transfer
with freezing in a scraped surface heat e
changer, Appl. Therm. Eng. 25 (2005) 45–60
[62] L. Liu, O. Miyawaki, K. Nakamura, Pro-
gressive freeze-concentration of model liquid
food, Food Sci. Technol. Int. 3 (1997) 348–352.
[63] R. Fujioka, L.P. Wang, G. Dodbiba, T. Fu-
jita, Application of progressive freeze concen-
tration freeze concentration for desalination,
Desalination 319 (2013) 33–37.
[64] T. Htira, C. Cogn., E. Gagni.re, D. Mangin,
Experimental study of industrial waste water
treatment by freezing, Journal of Water Process
Engineering 23 (2018) 292–298.
[65] Y. Shirai, M. Wakisaka, O. Miyawaki, S.
Sakashita, Conditions of producing an ice layer
with high purity for freeze wastewater treat-
ment, J. Food Eng. 38 (1998) 297–308
[66] A.A. Madani, Zero-discharge direct-con-
tact freezing/solar evaporator desalination com-
plex, Desalination 85 (1992) 179–195.
[67] A. Hanafi, Desalination using renewable
energy sources, Desalination 97 (1994) 339–
352.
[68] P. Wang, T.-S. Chung, A conceptual
demonstration of freeze desalination–mem-
brane distillation (FD–MD) hybrid desalination
process utilizing liquefied natural gas (LNG)
cold energy, Water Res. 46 (2012) 4037–4052.
[69] A. Antonelli, Desalinated water production
at LNG-terminals, Desalination 45 (1983) 383–
390
[70] W.-s. Cao, X.-s. Lu, W.-s. Lin, A.-z. Gu,
Parameter comparison of two small-scale natu-
ral gas liquefaction processes in skid-mounted
packages, Appl. Therm. Eng. 26 (2006) 898–
904.
[71] W. Lin, M. Huang, A. Gu, A seawater
freeze desalination prototype system utilizing
LNG cold energy, Int. J. Hydrog. Energy 42
(2017) 18691–18698.
[72] W.-s. Cao, X.-s. Lu, W.-s. Lin, A.-z. Gu,
Parameter comparison of two small-scale natu-
ral gas liquefaction processes in skid-mounted
packages, Appl. Therm. Eng. 26 (2006) 898–
904.
[73] R.J.C. Vaessen, Development of scraped
eutectic crystallizers, Doctoral Thesis, Mechan-
ical Maritime and Materials Engineering, TU
Delft, 2003
[74] S.T. Reddy, H.J.M. Kramer, A.E. Lewis, J.
Nathoo, Investigating factors that affect separa-
tion in a eutectic freeze crystallization process,
Abstracts of the International Mine Water Con-
ference, 19th–23rd October 2009, Pretoria,
South Africa, 2009.
[75] Y. Mandri, A. Rich, D. Mangin, S. Abder-
afi, C. Bebon, N. Semlali, J.-P. Klein,T. Boun-
ahmidi, A. Bouhaouss, Parametric study of the
sweating step in the sea water desalination pro-
cess by indirect freezing, Desalination 269
(2011) 142–147.
[76] A. Rich, Y. Mandri, D. Mangin, A.
Rivoire, S. Abderafi, C. Bebon, N. Semlali, J.-
P. Klein, T. Bounahmidi, A. Bouhaouss, S.
Veesler, Sea water desalination by dynamic
layer melt crystallization: parametric study of
the freezing and sweating steps, J. Cryst.
Growth 342 (2012) 110–116.
[77] Z. Zhang, D.-W. Sun, Z. Zhu, L. Cheng,
Enhancement of crystallization processes by
power ultrasound: current state-of-the-art and
research advances, Compr. Rev.
Food Sci. Food Saf. 14 (2015) 303–316.
[78] A.E. Delgado, D.-W. Sun, Ultrasound-as-
sisted freezing, in: H. Feng, G. Barbosa-
Canovas, J. Weiss (Eds.), Ultrasound Technol-
ogies for Food and Bioprocessing, Springer
New York, New York, NY, 2011, pp. 495–509.
[79] R. Chow, R. Blindt, R. Chivers, M. Povey,
A study on the primary and secondary nuclea-
tion of ice by power ultrasound, Ultrasonics 43
(2005) 227–230.
[80] H. Kiani, Z. Zhang, A. Delgado, D.-W.
Sun, Ultrasound assisted nucleation of some
liquid and solid model foods during freezing,
Food Res. Int. 44 (2011) 2915–2921.
[81] M. Orlowska, M. Havet, A. Le-Bail, Con-
trolled ice nucleation under high voltage DC
electrostatic field conditions, Food Res. Int. 42
(2009) 879–88
[82] R. Kacker, M. Radoiu, H.J.M. Kramer,
Novel design integrating a microwave applica-
tor into a crystallizer for rapid temperature cy-
cling. a direct nucleation control study, Cryst.
Growth Des. 17 (2017) 3766–3774.
[83] L.F. Greenlee, D.F. Lawler, B.D. Free-
man, B. Marrot, P. Moulin, Reverse osmosis
desalination: water sources, technology, and
today's challenges, Water Res. 43
(2009) 2317–2348.
[84] D.G. Randall, J. Nathoo, A succinct re-
view of the treatment of reverse osmosis
brines using freeze crystallization, Journal of
Water Process Engineering 8 (2015)
186–194.
[85] ESCWA, Water desalination: the experi-
ence of GCC countries, Regional Symposium
on Water Use and Conservation, Amman, Jor-
dan, 1993.
[86] A.A. Madani, S.E. Aly, A combined
RO/freezing system to reduce inland rejected
brine, Desalination 75 (1989) 241–258.
[87] M. Hasan, N. Rotich, M. John, M. Louhi-
Kultanen, Salt recovery from wastewater
by air-cooled eutectic freeze crystallization,
Chem. Eng. J. 326 (2017) 192–200.
[88] D.G. Randall, J. Nathoo, A.E. Lewis, A
case study for treating a reverse osmosis
brine using eutectic freeze crystallization—ap-
proaching a zero waste process,
Desalination 266 (2011) 256–262.
Magnetic Cooling
[1] A. Zhukov, ed., Novel Functional Magnetic
Materials: Fundamentals and Applications,
Springer International Publishing,2016.
//www.springer.com/us/book/9783319261041
(accessed June 21, 2018).
[2] S.F. Pearson, in 17th Informatory Note on
Refrigerating Technologies: How to Improve
Energy Efficiency in Refrigerating Equipment
(International Institute of Refrigeration).
(2003).
[3] M. Balli, S. Jandl, P. Fournier, A. Kedous-
Lebouc, Advanced materials for magnetic cool-
ing: Fundamentals and practical aspects, Ap-
plied Physics Reviews. 4 (2017) 021305.
doi:10.1063/1.4983612.
[4] M.A. McNeil, V.E. Letschert, Future air
conditioning energy consumption in develop-
ing countries and what can be done about it: the
potential of efficiency in the residential sector,
in: European Council for an Energy Efficient
Economy Summer Study Proceedings, 2007: p.
1311.
[5] A. Kitanovski, J. Tušek, U. Tomc, U. Plaz-
nik, M. Ozbolt, A. Poredoš, Magnetocaloric
Energy Conversion: From TheorytoApplica-
tionsSpringer InternationalPublishing,2015.
//www.springer.com/la/book/978331908740 5
(accessed June 21, 2018).
[6] V. Franco, J.S. Bl´azquez, B. Ingale, A.
Conde, The Magnetocaloric Effect and Mag-
netic Refrigeration Near Room Temperature:
Materials and Models, Annual Review of Ma-
terials Research. 42 (2012) 305–342.
doi:10.1146/annurev-matsci-062910-100356.
[7]See https://treaties.un.org/doc/Publica-
tion/UNTS/Volume%201522/volume-1522-I-
26369-English.p df for Montreal Protocol On
Substances that Deplete the Ozone Layer, Con-
cluded at Montreal on 16 September 1987,
United Nations-Treaty Series.
[8] O. Sari, M. Balli, From conventional to
magneticrefrigeratortechnology, International
Journal of Refrigeration. 37 (2014)8–15.
doi:10.1016/j.ijrefrig.2013.09.027.
[9]See http://unfccc.int/resource/docs/con-
vkp/kpeng.pdf for Kyoto Protocol to the United
Nations Framework Convention on Climate
Change, United Nations, 1998.
[10] Y. Gao, X. Gao, X. Zhang, The 2 °C
Global Temperature Target and the Evolution
of the Long-Term Goal of Addressing Climate
Change—From the United Nations Framework
Convention on Climate Change to the Paris
Agreement, Engineering. 3 (2017) 272–278.
doi:10.1016/J.ENG.2017.01.022.
[11] S.B. Roy, in: Handbook of Magnetic Ma-
terials, Edited by K. H. J. Buschow, North-Hol-
land Publishing Company, Amsterdam, The
Netherlands, 2014.
[12] K.A. Gschneidner, V.K. Pecharsky, Thirty
years of near room temperature magnetic cool-
ing: Where we are today and future prospects,
International Journal of Refrigeration. 31
(2008) 945– 961. doi:10.1016/j.ijre-
frig.2008.01.004.
[13]E.Brück,Developmentsin magneto-caloric
refrigeration, J. Phys. D: Appl. Phys. 38
(2005)R381.doi:10.1088/0022-
3727/38/23/R01.
[14] A.M. Tishin, Y.I. Spichkin, The Magneto-
caloric Effect and its Applications, IOP Pub-
lishing, Bristol, 2003
[15] A.M. Tishin, Y.I. Spichkin, Relationship
of adiabatic, isothermal and field constant
changes of a magnetic entropy, AIP Conference
Proceedings. 614 (2002) 27–34.
doi:10.1063/1.1472522.
[16] A.M. Tishin, Y.I. Spichkin, V.I. Zverev,
P.W. Egolf, A review and new perspectives for
the magneto-caloric effect: New materials and
local heating and cooling inside the human
body, International Journal of Refrigeration. 68
(2016) 177–186. doi:10.1016/j.ijre-
frig.2016.04.020.
[17] O. Gutfleisch, M.A. Willard, E. Brück,
C.H. Chen, S.G. Sankar, J.P. Liu, Magnetic Ma-
terials and Devices for the 21st Century:
Stronger, Lighter, and More Energy Efficient,
Advanced Materials. 23, 821–842.
doi:10.1002/adma.201002180.
[18] C. Zimm, A. Jastrab, A. Sternberg, V.
Pecharsky, K.G. Jr, M. Osborne, I. Anderson,
Description and Performance of a Near-Room
Temperature Magnetic Refrigerator, in: P. Kit-
tel (Ed.), Advances in Cryogenic Engineering,
Springer US, 1998: pp. 1759–1766.
doi:10.1007/978-1-4757-9047-4_222.
[19] N.A. Zarkevich, D.D. Johnson, V.K.
Pecharsky, High-throughput search for caloric
materials: the CaloriCool approach, J. Phys. D:
Appl. Phys. 51 (2018) 024002.
doi:10.1088/1361-6463/aa9bd0. [20] B.
CHAUDRONJ.,C.MULLER,M. HIT-
TINGER,EtAl.,Performance –measurement on
a large-scale magneto-caloric cooling applica-
tion at room temperature.,(2018).
doi:10.18462/iir.thermag.2018.0022.
[21] A. Rowe, Configuration and performance
analysis of magnetic refrigerators, International
Journal of Refrigeration. 34 (2011) 168–177.
doi:10.1016/j.ijrefrig.2010.08.014.
[22] B. Yu, M. Liu, P.W. Egolf, A. Kitanovski,
A review of magnetic refrigerator and heat
pump prototypes built before the year 2010, In-
ternational Journal of Refrigeration. 33 (2010)
1029– 1060. doi:10.1016/j.ijre-
frig.2010.04.002.
[23] P. Weiss, A. Piccard, Le phénomène mag-
nétocalorique, J. Phys. Theor. Appl. 7 (1917)
103–109.
doi:10.1051/jphystap:019170070010300.
[24] A. Smith, Who discovered the magnetoca-
loric effect?, EPJ H. 38 (2013) 507–517.
doi:10.1140/epjh/e2013-40001-9.
[25] P. Debye, Einige Bemerkungen zur Mag-
netisierung bei tiefer Temperatur, Annalen Der
Physik. 386 (n.d.) 1154–1160.
doi:10.1002/andp.19263862517.
[26] W.F. Giauque, A THERMODYNAMIC
TREATMENT OF CERTAIN MAGNETIC
EFFECTS. A PROPOSED METHOD OF
PRODUCINGTEMPERATURES CONSID-
ERABLY BELOW 1° ABSOLUTE, J. Am.
Chem. Soc. 49 (1927) 1864–1870.
doi:10.1021/ja01407a003.
[27] W.F. Giauque, D.P. MacDougall, Attain-
ment of Temperatures Below 1 Absolute by De-
magnetization of Gd2(SO4)3*8H2O, Phys.
Rev. 43 (1933) 768–768.
doi:10.1103/PhysRev.43.768.
[28] G.V. Brown, Magnetic heat pumping near
room temperature, Journal of Applied Phys-
ics.47(1976)3673–3680.
doi:10.1063/1.323176. [29] S.A. Nikitin, G.
Myalikgulyev, A.M. Tishin, M.P. Annaorazov,
K.A. Asatryan, A.L. Tyurin, The magnetoca-
loric effect in Fe49Rh51 compound, Physics
Letters A. 148 (1990) 363–366.
doi:10.1016/0375-9601(90)90819-
[30] J. Liu, T. Gottschall, K.P. Skokov, J.D.
Moore, O. Gutfleisch, Giant magnetocaloric ef-
fect driven by structural transitions, Nat Ma-
ter.11(2012)620–626. doi:10.1038/nmat3334.
[31] A. Chirkova, K.P. Skokov, L. Schultz,
N.V. Baranov, O. Gutfleisch, T.G. Woodcock,
Giant adiabatic temperature change in FeRh al-
loys evidenced by direct measurements under
cyclic conditions, Acta Materialia. 106 (2016)
15–21. doi:10.1016/j.actamat.2015.11.054.
[32] J.B. Staunton, R. Banerjee, M. dos S. Dias,
A. Deak, L. Szunyogh, Fluctuating local mo-
ments, itinerant electrons, and the magneto-ca-
loric effect: Compositional hypersensitivity of
FeRh, Phys. Rev. B. 89 (2014) 054427.
doi:10.1103/PhysRevB.89.054427.
[33] M. Takahashi, R. Oshima, Annealing Ef-
fect on Phase Transition of Equiatomic FeRh
Alloy, Materials Transactions, JIM. 36 (1995)
735–742. doi:10.2320/matertrans1989.36.735.
[34] M.P. Annaorazov, K.A. Asatryan, G. My-
alikgulyev, S.A. Nikitin, A.M. Tishin, A.L.
Tyurin, Alloys of the Fe-Rh system as a new
class of working material for magnetic refriger-
ators, Cryogenics. 32 (1992) 867–872.
doi:10.1016/0011-2275(92)90352-B.
[35] V.I. Zverev, A.M. Saletsky, R.R. Gimaev,
A.M. Tishin, T. Miyanaga, J.B. Staunton, Influ-
ence of structural defects on the magnetocaloric
effect in the vicinity of the first order magnetic
transition in Fe50.4Rh49.6, Applied Physics
Letters. 108 (2016) 192405.
doi:10.1063/1.4949355.
[36] V.K. Pecharsky, J. Gschneidner K.A., Gi-
ant Magnetocaloric Effect in Gd5Si2Ge2, Phys.
Rev. Lett. 78 (1997) 4494–4497.
doi:10.1103/PhysRevLett.78.4494.
 [37] V.K. Pecharsky, K.A. Gschneidner, Tun-
able magnetic regenerator alloys with a giant
magnetocaloric effect for magnetic refrigera-
tion from ∼20 to ∼290 K, Appl. Phys. Lett. 70
(1997) 3299–3301. doi:10.1063/1.119206.
[38] X.X. Zhang, J. Tejada, Y. Xin, G.F. Sun,
K.W. Wong, X. Bohigas, Magnetocaloric effect
in La0.67Ca0.33MnOδ and
La0.60Y0.07Ca0.33MnOδ bulk materials,
Appl. Phys. Lett. 69 (1996) 3596–3598.
doi:10.1063/1.117218.
[39] Z.B. Guo, Y.W. Du, J.S. Zhu, H. Huang,
W.P. Ding, D. Feng, Large Magnetic Entropy
Change in Perovskite-Type Manganese Oxides,
Phys. Rev. Lett. 78 (1997) 1142–1145.
doi:10.1103/PhysRevLett.78.1142.
[40] H. Feng-xia, S. Bao-gen, S. Ji-rong, Z. Xi-
xiang, Great magnetic entropy change in La(Fe,
M ) 13 ( M =Si, Al) with Co doping, Chinese
Phys. 9 (2000) 550. doi:10.1088/1009-
1963/9/7/016. [41] A. Fujita, S. Fujieda, Y. Ha-
segawa, K. Fukamichi, Itinerant-electron meta-
magnetic transition and large magneto-caloric
effects in La(FexSi1-x)13 compounds and their
hydrides, Phys. Rev. B. 67 (2003) 104416.
doi:10.1103/PhysRevB.67.104416.
[42] H. Wada, Y. Tanabe, Giant magneto-ca-
loric effect of MnAs1−xSbx, Applied Physics
Letters. 79 (2001) 3302–3304.
doi:10.1063/1.1419048. [43] O. Tegus, E.
Brück, K.H.J. Buschow, F.R. de Boer, Transi-
tion-metal-based magnetic refrigerants for
room-temperature applications, Nature. 415
(2002) 150–152. doi:10.1038/415150a.
[44] F. Hu, B. Shen, J. Sun, Magnetic entropy
change in Ni51.5Mn22.7Ga25.8 alloy, Appl.
Phys. Lett. 76 (2000) 3460–3462.
doi:10.1063/1.126677.
[45] T. Krenke, E. Duman, M. Acet, E.F. Was-
sermann, X. Moya, L. Mañosa, A. Planes, In-
verse magneto-caloric effect in ferromagnetic
Ni–Mn–Sn alloys, Nature Materi-
als4(2005)450–454. doi:10.1038/nmat1395.
[46] E. Liu, W. Wang, L. Feng, W. Zhu, G. Li,
J. Chen, H. Zhang, G. Wu, C. Jiang, H. Xu, F.
de Boer, Stable magneto-structural coupling
with tunable magnetoresponsive effects in hex-
agonal ferromagnets, Nature Communications.
3 (2012) 873. doi:10.1038/ncomms1868.
[47] Y. Li, H. Zhang, K. Tao, Y. Wang, M. Wu,
Y. Long, Giant magnetocaloric effect induced
by reemergence of magnetostructural coupling
in Si-doped Mn0.95CoGe compounds, Materi-
als & Design. 114 (2017) 410–415.
doi:10.1016/j.matdes.2016.11.002.
[48] V. Provenzano, A.J. Shapiro, R.D. Shull,
Reduction of hysteresis losses in the magnetic
refrigerant Gd5Ge2Si2 by the addition of iron,
Nature. 429 (2004) 853–857. doi:10.1038/na-
ture02657.
[49] J. Lyubina, R. Schäfer, N. Martin, L.
Schultz, O. Gutfleisch, Novel design of
La(Fe,Si)13 alloys towards high magnetic re-
frigeration performance, Adv. Mater. Wein-
heim. 22 (2010) 3735– 3739.
doi:10.1002/adma.201000177.
[50] F. Guillou, G. Porcari, H. Yibole, N. van
Dijk, E. Brück, Taming the First-Order Transi-
tion in Giant Magneto-caloric Materials, Ad-
vanced Materials. 26, 2671–2675.
doi:10.1002/adma.201304788.
[51] V.I. Zverev, R.R. Gimaev, A.M. Tishin, Y.
Mudryk, K.A. Gschneidner, V.K. Pecharsky,
The role of demagnetization factor in determin-
ing the ‘true’ value of the Curie temperature,
Journal of Magnetism and Magnetic Materials.
323 (2011) 2453–2457.
doi:10.1016/j.jmmm.2011.05.012. [52] H.
Zhang, F. Hu, J. Sun, B. Shen, Effects of inter-
stitial H and/or C atoms on the magnetic and
magnetocaloric properties of La(Fe, Si)13-
based compounds, Sci. China Phys. Mech. As-
tron. 56 (2013) 2302–2311.
doi:10.1007/s11433-013-5357-1.
[53] W.F. Giauque, D.P. MacDougall, The Pro-
duction of Temperatures below One Degree
Absolute by Adiabatic Demagnetization of
Gadolinium Sulfate, J. Am. Chem. Soc. 57
(1935) 1175–1185. doi:10.1021/ja01310a007.
[54] V.K. Pecharsky, K.A. Gschneidner, S.Y.
Dan’kov, A.M. Tishin, Magneto-caloric Prop-
erties of Gd3Al2, in: Cryocoolers 10, Springer,
Boston, MA, 2002: pp. 639–645.
doi:10.1007/0-306-47090-X_76.
[55] M.D. Kuz’min, A.M. Tishin, Magnetic re-
frigerants for the 4.2-20 K region: garnets or
perovskites, J. Phys. D: Appl. Phys. 24
(1991)2039.doi:10.1088/0022 3727/24/11/020.
[56] M.D. Kuz’min, A.M. Tishin, DyAlO3: A
more promising refrigerant than Dy3Al5O12?,
Journal of Applied Physics. 73 (1993) 4083–
4085. doi:10.1063/1.352835.
[57] A.M. Tishin, L.P. Bozkova, New magnetic
refrigerants for low temperature region, Journal
of Applied Physics. 81 (1997) 1000–1001.
doi:10.1063/1.364191. [58] S.Y. [58] Dan’kov,
V.V. Ivtchenko, A.M. Tishin, K.A.Gschneid-
ner,V.K.Pecharsky, Magnetocaloric Effect in
GdAl2 and Nd2Fe17, in: Advances in Cryo-
genic Engineering Materials, Springer, Boston,
MA, 2000: pp. 397–404. doi:10.1007/978-1-
4615-4293-3_51. Ackermann, R.A., 1997. Cry-
ogenic Regenerative Heat Exchangers.
Springer Science + Business Media LLC.
[59] Allab, F., Kedous-Lebouc, A., Fournier,
J.M., Yonnet, J.P., 2005. Numerical modeling
for active magnetic regenerative refrigeration.
IEEE Trans. Magn. 41, 3757–3759.
https://doi.org/10.1109/TMAG.2005.854757
[60]Allab, F., Kedous-Lebouc, A., Yonnet,
J.P., Fournier, J.M., 2006. A magnetic field
source system for magnetic refrigeration and
its interaction with magneto-caloric material.
Int. J. Refrig., Magnetic Refrigeration at Room
Temperature 29, 1340–1347.
https://doi.org/10.1016/j.ijrefrig.2006.07.025
[61]Aprea, C., Cardillo, G., Greco, A., Ma-
iorino, A., Masselli, C., 2016. A rotary perma-
nent magnet magnetic refrigerator based on
AMR cycle. Appl. Therm. Eng. 101, 699–703.
https://doi.org/10.1016/j.a
plthermaleng.2016.01.097
[62]Aprea, C., Greco, A., Maiorino, A.,
Mastrullo, R., Tura, A., 2014. Initial experi-
mental results from a rotary permanent magnet
magnetic refrigerator. Int. J. Refrig. 43, 111–
122. https://doi.org/10.1016/j.ijr
frig.2014.03.014
[63]Arnold, D.S., Tura, A., Rowe, A., 2011.
Experimental analysis of a two-material active
magnetic regenerator. Int. J. Refrig. 34, 178–
191. https://doi.org/10.1016/j.ijre-
frig.2010.08.015
[64]Arnold, D.S., Tura, A., Ruebsaat-Trott, A.,
Rowe, A., 2014. Design improvements of a
permanent magnet active magnetic refrigera-
tor. Int. J. Refrig., New Developments in Mag-
netic Refrigeration 37, 99–105.
https://doi.org/10.1016/j.ijrefrig.2013.09.024
[65]Bahl, C.R.H., Engelbrecht, K., Eriksen,
D., Lozano, J.A., Bjørk, R., Geyti, J., Nielsen,
K.K., Smith, A., Pryds, N., 2014. Develop-
ment and experimental results from a 1 kW
prototype AMR. Int. J. Refrig., New Develop-
ments in Magnetic Refrigeration 37, 78–83.
https://doi.org/10.1016/j.ijrefrig.2013.09.001
[66]Balli, M., Sari, O., Mahmed, C., Besson,
C., Bonhote, P., Duc, D., Forchelet, J., 2012. A
preindustrial magnetic cooling system for
room temperature application. Appl. Energy
98, 556–561. https://doi.org/10.1016/j.apen-
ergy.2012.04.034
[67] Bartlett, J., Hardy, G., Hepburn, I.D.,
2015. Performance of a fast response minia-
ture Adiabatic Demagnetisation Refrigerator
using a single crystal tungsten magnetoresis-
tive heat switch. Cryogenics 72, 111–121.
https://doi.org/10.1016/j.cryogen-
ics.2015.10.004
[68] Bézaguet, A., Casas-Cubillos, J., Lebrun,
P., Losserand-Madoux, R., Marquet, M.,
Schmidt- Ricker, M., Seyfert, P., 1994. Design
and construction of a static magnetic refrigera-
tor Operating between 1.8 K and 4.5 K. Cryo-
genics, Fifteenth International Cryogenic En-
gineering Conference 34, 227–230.
https://doi.org/10.1016/S0011-2275(05)80049-
9
[68]Bjørk, R., Bahl, C.R.H., Smith, A., Pryds,
N., 2010. Review and comparison of magnet
designs for magnetic refrigeration. Int. J. Re-
frig. 33, 437–448.
https://doi.org/10.1016/j.ijrefrig.2009.12.012
[69] Blumenfeld, P.E., Prenger, F.C., Stern-
berg, A., 2002a. High temperature supercon-
ducting magnetic refrigeration. Advances in
cryogenic engineering: proceedings of the cry-
ogenic
engineering conference 47.
[70]Blumenfeld, P.E., Prenger, F.C., Stern-
berg, A., Zimm, C., 2002b. High temperature
superconducting magnetic refrigeration. AIP
Conf. Proc. 613, 1019–1026.
https://doi.org/10.1063/1.1472124
[71] Bohigas, X., Molins, E., Roig, A., Tejada,
J., Zhang, X.X., 2000. Room-temperature
magnetic refrigerator using permanent mag-
nets. IEEE Trans. Magn. 36, 538–544.
https://doi.org/10.1109/20.846216
[72] Bouchekara, H.R.E.H., Kedous-Lebouc,
A., Yonnet, J.P., 2012. Design of a new mag-
netic refrigeration field source running with
rotating bar-shaped magnets. Int. J. Refrig. 35,
115–121. https://doi.org/10.1016/j.ijr
frig.2011.08.006
[73] Bour, S., Hamm, S.J., Michot, H., 2009.
Experimental and numerical analysis of a re-
ciprocating room temperature active magnetic
regenerator, in: Proc. 3rd International Confer-
ence on Magnetic Refrigeration at Room Tem-
perature (Thermag III) 11/05/2009 -
15/05/2009
[74]Des Moines, WA, United States. p. 415.
[75]Chen, P., Du, Y.W., 2001. Large magnetic
entropy change in (Pr 1- x Nd x ) 0.5 Sr 0.5
MnO 3. J. Phys. Appl. Phys. 34, 1868.
https://doi.org/10.1088/0022-3727/34/12/316
[76]Chen, P., Du, Y.W., Ni, G., 2000. Low-
field magneto-caloric effect in
Pr0.5Sr0.5MnO3. EPL Europhys. Lett. 52,
589. https://doi.org/10.1209/epl/i2000-00478-8
[77]Cheng, J., Liu, G., Huang, J., Liu, C., Jin,
P., Yan, H., 2013. Refrigeration effect of
La(FeCoSi)13B0.25 compounds and gadolin-
ium metal in reciprocating magnetic refrigera-
tor. J. Rare Earths 31, 1163–1167.
https://doi.org/10.1016/S1002-
0721(12)60421-9
[78]Chiba, Y., 2017. Experimental and N mer-
ical Study of the Behavior of a Multilayer for
Active Magnetic Refrigerator Based on La-Fe-
Co-Si. IEEE Trans. Magn. 53, 1–4.
https://doi.org/10.1109/TMAG.2017.2703100
[79]Chiba, Y., Smaïli, A., Mahmed, C., Balli,
M., Sari, O., 2014. Thermal investigations of
an experimental active magnetic regenerative
refrigerator operating near room temperature.
Int. J. Refrig., New Developments in Magnetic
Refrigeration 37, 36–42.
https://doi.org/10.1016/j.ijrefrig.2013.09.038
[80]Christiaanse, T.V., Trevizoli, P.V., Misra,
S., Carroll, C., Asten, D. van, Zhang, L., Tey-
ber, R., P
[81]Govindappa, Niknia, I., Rowe, A., 2018.
Experimental study of 2-layer regenerators
using Mn–Fe–Si–P materials. J. Phys. Appl.
Phys. 51, 105002.
https://doi.org/10.1088/1361-6463/aaaba7
[82]Clot, P., Viallet, D., Allab, F., Kedous-
Lebouc, A., Fournier, J.M., Yonnet, J.P., 2003.
A magnetbased device for active magnetic re-
generative refrigeration. IEEE Trans. Magn.
39, 3349–3351.
https://doi.org/10.1109/TMAG.2003.816253
[83]Coelho, A.A., Gama, S., Magnus, A.,
2009. Prototype of a Gd-based rotating mag-
netic refrigerator for work around room tem-
perature, in: Proc. 3rd International Confer-
ence on Magnetic Refrigeration at Room Tem-
perature (Thermag III) 11/05/2009 15/05/2009
[84] Des Moines, WA, United States. p. 381.
Czernuszewicz, A., Kaleta, J., Królewicz, M.,
Lewandowski, D., Mech, R., Wiewiórski, P.,
2014. A test stand to study the possibility of
using magnetocaloric materials for refrigera-
tors. Int. J. Refrig., New Developments in
Magnetic Refrigeration 37, 72–77.
https://doi.org/10.1016/j.ijrefrig.2013.09.017
[85] Domanski, P.A., 1999. Evolution of re-
frigerant application., in: International Con-
gress of Refrigeration, May 4, Milan, Italy. Pr
sented at the International Congress on Refrig-
eration.
[86] Dupuis, C., Vialle, A.J., Legait, U., 2009.
New investigations in magnetic refrigeration
device, AMR cycle and refrigerant bed perfor-
mance evaluation, in: Proc. 3rd International