OPTIMIZATION OF LAYERED REGENERATOR

OF A MAGNETIC REFRIGERATION DEVICE

A Seminar Report

Submitted by

JOHN THOMAS A

in partial fulfillment for the award of the degree

of
M.TECH DEGREE
IN
MECHANICAL ENGINEERING

Coordinators

Dr.Tide P.S and Mr.Joshy P.J

SCHOOL OF ENGINEERING
COCHIN UNIVERSITY OF SCIENCE AND TECHNOLOGY
COCHIN

November, 2015
 COCHIN UNIVERSITY OF SCIENCE AND TECHNOLOGY

SCHOOL OF ENGINEERING

CERTIFICATE

This is to certify that the seminar report entitled

OPTIMIZATION OF LAYERED REGENERATOR

OF A MAGNETIC REFRIGERATION DEVICE

was submitted by

JOHN THOMAS A

of first semester Mechanical Engineering Division in partial fulfillment

of the requirement for the award of the Degree of Master of
Technology in Thermal Engineering of Cochin
University of Science and Technology
during the year 2015

Coordinators

Dr. Tide P. S.

Mr. Joshy P.J.

Head of Division

Dr. Biju N.

i
 ABSTRACT

Several emerging technologies in the domain of solid-state physics have been investi-
gated as serious alternatives for future refrigeration, heat pumping, air conditioning, or
even power generation applications. These technologies relate to what is called caloric
energy conversion, i.e., barocalorics, electrocalorics, magnetocalorics, and elastocalorics.
Of these technologies, the greatest progress has been observed in the domain of magnetic
refrigeration

Magnetic refrigeration, as an alternative to vapor-compression technology. A technique

to enhance the performance of magnetic refrigerators is using layers of different materials
in the regenerator of such devices. In this study the choice of magnetocaloric materials
in a multi-layered packed bed regenerator is investigated in order to optimize the perfor-
mance. A numerical model has been developed to simulate the packed bed in this study.
Optimized packed bed designs to get maximum temperature span or maximum efficiency
are different. The results indicate that maximum temperature span can be achieved by
choosing the materials with the highest magnetocaloric effect in the working temperature
range, while maximum Carnot efficiency is achieved by choosing materials with Curie
temperatures above the average layer temperature.

ii
 ACKNOWLEDGEMENT

I would like to express my whole hearted thanks to God almighty for His blessings
enabling to complete the seminar.
I wish to express my genuine gratitude to Dr. Biju N, Head of the Mechanical Engi-
neering Department for his unlimited support, direction, advice.
I am very thankful to Prof. Dr. Tide P S and Mr. Joshy P J , Mechanical Engineering
Department, our seminar coordinators for his meticulous guidance.
I would like to express my gratitude to Dr. Ealias Jacob for his wholehearted support
and encouragement.
The help and assistance rendered by the staff in the Mechanical Engineering Depart-
ment are gratefully acknowledged.
I also express my heartily thanks to all my friends for their devoted help and moral
support for the completion of this seminar.

JOHN THOMAS A

iii
Contents

List of Figures v

List of Tables vi

1 INTRODUCTION 1

2 Magnetic Refrigeration 5
2.1 Working . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
2.1.1 Applied Technique . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.2 Active Magnetic Regenerator . . . . . . . . . . . . . . . . . . . . . . . . . 8

3 System model 10

4 Result and Discussion 13

4.1 Optimization Of Layered Regenerators . . . . . . . . . . . . . . . . . . . . 15

5 CONCLUSIONS 18

References 19

iv
List of Figures

1.1 Energy conversion cycle . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

2.1 principal behind magnetocaloric effect . . . . . . . . . . . . . . . . . . . . . 6

2.2 Analogy between magnetic and vapour compression refrigeration . . . . . . 7
2.3 Active magnetic refrigeration cycle . . . . . . . . . . . . . . . . . . . . . . 8

3.1 Flow and field variation for each bed during a cycle . . . . . . . . . . . . . 12

4.1 Average temperature during a cycle at each point along the bed[1] . . . . . 13
4.2 Average temperature of the bed with Curie temperature selected as in fig4.1[1] 14
4.3 Average temperature and carnot efficiency of the bed with Curie temper-
ature selected as in fig4.1[1] . . . . . . . . . . . . . . . . . . . . . . . . . . 14
4.4 curie temperature at T = 47%cycle(maximum temperature span)[1]. . . . . 15
4.5 curie temperature at T = 21%cycle(maximum Carnot efficiency)[1]. . . . . 15
4.6 Carnot efficiency Vs tempertaure span[4]. . . . . . . . . . . . . . . . . . . . 16
4.7 Carnot efficiency Vs cycle time[1]. . . . . . . . . . . . . . . . . . . . . . . . 17
4.8 Temperature span Vs cycle time[1]. . . . . . . . . . . . . . . . . . . . . . . 17

v
List of Tables

3.1 boundary conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

vi
Chapter 1

INTRODUCTION

Refrigeration is the process of removing heat from an enclosed volume or from a sub-
stance. The use of ice for prolonging the storage life of foodstuff dates back many millen-
niums. Ice owes its effectiveness as a cooling agent to its constant melting point of 0℃.
In order to melt, ice must absorb 333 kJ/kg. Up to the middle of the last century all ice
used for cooling was from natural sources. This ice was taken from lakes in winter and
then stored. In 1890, the U.S. still exported 25 million tons of ice per year. However, the
pollution of lakes, the competitiveness of the new technology and no able to follow the
increase of the demand has caused end of this industry.
The world’s energy demands for refrigeration and air conditioning represent nearly 20%
of the energy consumption. The major cooling technology, i.e., the vapor compression of
a gas refrigerant, even though it is now mature, is characterized by a rather low exergy
efficiency, especially for small devices. Despite the fact that other substances are being
used to substitute for existing or already-abandoned harmful refrigerants, many of these
are the subject of future prohibition. Moreover, several alternatives, such as the potential
replacement of existing refrigerants in vapor compression, lead to lower energy efficiency
or problems related to very high pressures, flammability, explosion hazards, etc.
In recent years magnetic refrigeration at room temperature has come up as an alter-
native to the vapor-compression technology. The motivation , mainly, to reach higher
efficiencies and less environmental impacts compared to vapor-compression technology.
The magnetocaloric effect (MCE, from magnet and calorie) is a magneto-thermodynamic
phenomenon in which a temperature change of a suitable material is caused by exposing
the material to a changing magnetic field. This is also known by low temperature physi-
cists as adiabatic demagnetization. In that part of the refrigeration process, a decrease
in the strength of an externally applied magnetic field allows the magnetic domains of a
magnetocaloric material to become disoriented from the magnetic field by the agitating
action of the thermal energy (phonons) present in the material. If the material is isolated
so that no energy is allowed to (re)migrate into the material during this time, (i.e., an
adiabatic process) the temperature drops as the domains absorb the thermal energy to
perform their reorientation. The randomization of the domains occurs in a similar fash-
ion to the randomization at the curie temperature of a ferromagnetic material, except
that magnetic dipoles overcome a decreasing external magnetic field while energy remains
constant, instead of magnetic domains being disrupted from internal ferromagnetism as
energy is added.

1
 Figure 1.1: Energy conversion cycle

One of the improvements in the performance of magnetic refrigerator is to have layers

of different materials in the packed beds instead of only one specific material.While a mag-
netic refrigerator is operating, a temperature span is established between the two ends
of the bed, which means that the temperatures at different positions along the bed are
different at each moment of a cycle. Considering the fact that the magnetocaloric materi-
als show higher magnetocaloric effect about their transition temperature, materials with
different transition temperatures can be used along the bed to enhance the performance.

2
 History and Development

The effect was discovered using nickel in 1917 by French physicist Pierre Weiss and
Auguste Piccard. Originally, the cooling effect was less than 0.5 K/T. Major advances first
appeared in the late 1920s when cooling via adiabatic demagnetization was independently
proposed by Peter Debye in 1926 and chemistry Nobel Laureate William F. Giauque
in 1927. It was first demonstrated experimentally by Giauque and his colleague D. P.
MacDougall in 1933 for cryogenic purposes when they reached 0.25 K.

In 1997, the first near room-temperature proof of concept magnetic refrigerator was
demonstrated by Karl A. Gschneidner, Jr. by the Iowa State University at Ames Labora-
tory. This event attracted interest from scientists and companies worldwide who started
developing new kinds of room temperature materials and magnetic refrigerator designs.

A major breakthrough came 2002 when a group at the University of Amsterdam demon-
strated the giant magnetocaloric effect in MnFe(P,As) alloys that are based on abundant
materials. Among investigations on second order phase transition materials for room
temperature applications with tunable transient temperature, the works of Canepa et al.
(2002) and Dai et al., 2000 who worked on Gd7 P d3−x N ix and (Gd, Dy)1−x N dx alloys can
be mentioned. By changing the composition of Gd7 P d3−x N ix and (Gd, Dy)1−x N dx alloys,
different transition temperatures, both higher and lower than that of Gadolinium, can be
obtained.

Richard it et al. 2004 showed that a two-layered bed consisting of Gd and Gd0.74 T b0.26
(with equal length of layers) can produce larger temperature span at no-load condition
and can give higher cooling capacity at larger spans compared to a single-layered bed
of Gd. Later Rowe and Tura (2006) used three layers of equal length of the materi-
als Gd,Gd0.74 T b0.26 , and Gd0.85 Er0.15 confirming the possibility of improving performance
through layering. However, they concluded that the composition of materials in layers
and their quantity should be optimized for different working conditions. Later Arnoldit et
al. (2011) concluded from their experiments on a two layered bed of Gd and Gd0.85 Er0.15
that the highest temperature spans are achieved when the average temperature during
the whole cycle in a layer is close to the Curie temperature of the layer’s material. Tusek
it et al. (2014) experimentally tested 2-, 4-, and 7-layered parallel plate regenerators
made of LaF e13−x−y Cox Siy . Their experiments highlighted that for different working
temperatures the optimum regenerator composition is different. Among the best per-
forming magnetic refrigerators, the device reported by Jacobs it et al. (2014) works with
six layers of LaFeSiH. The performance of their device is also numerically simulated.

This work investigates optimizing the selection of materials, to get either high effi-
ciency or high temperature span, which requires different selection of the materials, at
design stage using a numerical simulation. The Curie temperatures found through the
optimization can be used as a guide for the desired composition of alloys needed to have
an optimized magnetic refrigerator. Reid it et al. (1994) and Rowe and Barclayit et al.

3
(2003) suggest that the magnetocaloric effect of the materials used in consecutive layers
should ideally increase as the working temperature increases along the bed; neverthe-
less, in practice, such an increase in the magnetocaloric effect is not guaranteed when
the transient temperature is tuned by changing the composition of alloys. In this study
we assumed that the materials used in different layers have similar magnetocaloric effect
although their Curie temperatures are different.

4
Chapter 2

Magnetic Refrigeration

The magnetocaloric effect is a magneto-thermodynamic phenomenon in which a tem-

perature change of a suitable material is caused by exposing the material to a changing
magnetic field. This is known as adiabatic demagnetization. In that part of the refrigera-
tion process, a decrease in the strength of an externally applied magnetic field allows the
magnetic domains of a magnetocaloric material to become disoriented from the magnetic
field by the agitating action of the thermal energy (phonons) present in the material. If
the material is isolated so that no energy is allowed to (re)migrate into the material dur-
ing this time, (i.e., an adiabatic process) the temperature drops as the domains absorb
the thermal energy to perform their reorientation. The randomization of the domains
occurs in a similar fashion to the randomization at the curie temperature of a ferromag-
netic material, except that magnetic dipoles overcome a decreasing external magnetic
field while energy remains constant, instead of magnetic domains being disrupted from
internal ferromagnetism as energy is added.
The magnetocaloric effect can be quantified with the equation below:

(2.1)

where T is the temperature, H is the applied magnetic field, C is the heat capacity of the
working magnet (refrigerant) and M is the magnetization of the refrigerant.
From the equation we can see that magnetocaloric effect can be enhanced by: applying
a large field using a magnet with a small heat capacity using a magnet with a large change
in magnetization vs. temperature, at a constant magnetic field

5
 Figure 2.1: principal behind magnetocaloric effect

2.1 Working
The cycle is performed as a refrigeration cycle that is analogous to the Carnot refrigera-
tion cycle, but with increases and decreases in magnetic field strength instead of increases
and decreases in pressure. It can be described at a starting point whereby the chosen
working substance is introduced into a magnetic field, i.e., the magnetic flux density is
increased. The working material is the refrigerant, and starts in thermal equilibrium with
the refrigerated environment.

Adiabatic magnetization: A magnetocaloric substance is placed in an insulated envi-

ronment. The increasing external magnetic field (+H) causes the magnetic dipoles of the
atoms to align, thereby decreasing the material’s magnetic entropy and heat capacity.
Since overall energy is not lost (yet) and therefore total entropy is not reduced (according
to thermodynamic laws), the net result is that the substance is heated (T + Tad).

Isomagnetic enthalpic transfer: This added heat can then be removed (-Q) by a fluid
or gas gaseous or liquid helium, for example. The magnetic field is held constant to pre-
vent the dipoles from reabsorbing the heat. Once sufficiently cooled, the magnetocaloric
substance and the coolant are separated (H=0).

Adiabatic demagnetization: The substance is returned to another adiabatic (insulated)

condition so the total entropy remains constant. However, this time the magnetic field
is decreased, the thermal energy causes the magnetic moments to overcome the field,
and thus the sample cools, i.e., an adiabatic temperature change. Energy (and entropy)
transfers from thermal entropy to magnetic entropy (disorder of the magnetic dipoles).

6
 Figure 2.2: Analogy between magnetic and vapour compression refrigeration

Isomagnetic entropic transfer: The magnetic field is held constant to prevent the ma-
terial from reheating. The material is placed in thermal contact with the environment to
be refrigerated. Because the working material is cooler than the refrigerated environment
(by design), heat energy migrates into the working material (+Q).

2.1.1 Applied Technique

The basic operating principle of an adiabatic demagnetization refrigerator is the use of
a strong magnetic field to control the entropy of a sample of material, often called the
”refrigerant”. Magnetic field constrains the orientation of magnetic dipoles in the refrig-
erant. The stronger the magnetic field, the more aligned the dipoles are, corresponding
to lower entropy and heat capacity because the material has (effectively) lost some of its
internal degrees of freedom. If the refrigerant is kept at a constant temperature through
thermal contact with a heat sink while the magnetic field is switched on, the refrigerant
must lose some energy because it is equilibrated with the heat sink. When the magnetic
field is subsequently switched off, the heat capacity of the refrigerant rises again because
the degrees of freedom associated with orientation of the dipoles are once again liberated,
pulling their share of equipartitioned energy from the motion of the molecules, thereby
lowering the overall temperature of a system with decreased energy. Since the system
is now insulated when the magnetic field is switched off, the process is adiabatic, i.e.,
the system can no longer exchange energy with its surroundings (the heat sink), and its
temperature decreases below its initial value, that of the heat sink.
The operation of a standard magnetic refrigerator proceeds roughly as follows. First,
a strong magnetic field is applied to the refrigerant, forcing its various magnetic dipoles
to align and putting these degrees of freedom of the refrigerant into a state of lowered
entropy. The heat sink then absorbs the heat released by the refrigerant due to its
loss of entropy. Thermal contact with the heat sink is then broken so that the system
is insulated, and the magnetic field is switched off, increasing the heat capacity of the

7
refrigerant, thus decreasing its temperature below the temperature of the heat sink. In
practice, the magnetic field is decreased slowly in order to provide continuous cooling and
keep the sample at an approximately constant low temperature. Once the field falls to
zero or to some low limiting value determined by the properties of the refrigerant, the
cooling power vanishes, and heat leaks will cause the refrigerant to warm up.

2.2 Active Magnetic Regenerator

There are four processes in an active regenerative thermodynamic cycle. The caloric
materials need to be produced as a porous structure through which the working fluid (in
caloric cooling, this mostly relates to liquids) flows in an oscillatory, counter flow. In the
first process, denoted by A in Fig. 2.3, the caloric regenerator is exposed to an increase
in the field. During this process, the caloric material heats up with a certain temperature
difference (e.g., DTad, which denotes the adiabatic temperature change due to the sudden
change of the field, this process is very close to isentropic).

Figure 2.3: Active magnetic refrigeration cycle

[2]

In process B (Fig. 2.3), the working fluid flows from the cold heat exchanger (CHEX
or heat source heat exchanger) through the porous structure of the caloric regenerator
to the hot heat exchanger (HHEX or heat-sink heat exchanger), where the fluid rejects
the heat to, e.g., the environment. In process C (Fig. 2.3), the caloric regenerator is
exposed to a negative change of the field (demagnetization, depolarization, release and
expansion, respectively) and as a consequence, the temperature of the caloric regenerator
decreases. In the last process D, the working fluid flows from the HHEX through the
caloric regenerator to the CHEX. Since the fluid cools down, it now has the capability to
absorb heat in the heat-source heat exchanger (CHEX), and thus provides cooling. By

8
repeating these processes the temperature gradient will be established along the active
caloric regenerator. Despite the fact that the active caloric regenerative process is one of
the best-known solutions, this approach is very much restricted by the efficiency of the
convective heat transfer in the regenerator (heat-transfer surface, heat-transfer coefficient,
boundary layers, viscosity, fluids and the thermal properties of the materials, etc.). Since
the fluid flow through the regenerator made of caloric material is oscillatory, this somehow
also restricts the efficiency of the active heat regeneration process, especially for high
power densities.

9
Chapter 3

System model

The modeled system consists of a packed bed of magnetocaloric materials and a heat
transfer fluid flowing through the bed. The magnetocaloric material gets warm as it
is magnetized by an external magnetic field and the heat transfer fluid takes the heat
from the magnetocaloric material and rejects it to the ambient in a hot heat exchanger.
Demagnetizing the magnetocaloric material makes it cold, and the fluid exiting the hot
heat exchanger becomes cold when it goes through the demagnetized bed. The cold heat
transfer fluid absorbs heat in the cold heat exchanger and becomes warmer before entering
the packed bed again. The packed bed of the magnetocaloric materials serves as both
refrigerant and regenerator.
To increase the performance, the bed has 6 layers of magnetocaloric materials with
similar magnetocaloric effect but different Curie temperatures. To prevent the layers from
working at temperatures far from their Curie temperature, the temperature span created
by the active magnetic regenerative cycles is divided equally by the number of layers.
Since the temperature gradient along the bed between the cold end and the hot end is
not linear, the lengths of the layers are not necessarily equal. Experimentally measured
properties of Gadoliniumreported by Lozano it et al. (2014) are used as a basis to model
materials with similar magnetocaloric effect but different Curie temperatures. Since mag-
netocaloric effect, temperature, and heat capacity are interrelated through Maxwell equa-
tions (Pecharsky and Gschneidner, 1999), materials with similar magnetocaloric effect but
different Curie temperatures cannot be created by merely shifting the properties along the
temperature axis. In this work, to have more physically correct models of the materials
with Curie temperatures other than that of Gadolinium but with similar magnetocaloric
effect, the heat capacity data are modified in a way that they satisfy Maxwell’s equations.
The variations in the flow of heat transfer fluid and the magnetic field, modeled based
on the device described by Bjrk it et al. (2010) Simpler, but less realistic for rotary
devices, patterns of variation for flow rate and magnetic field, Fig. 2b, are also used to
obtain the results presented in a part of . Positive flow means flow from cold end to hot
end and negative flow rate means flow in the opposite direction.

10
 By applying the first law of Thermodynamics to the solid phase of the packed bed and
the heat transfer fluid going through the bed, equation (1) and (2), which are coupled by
convective heat transfer term, are derived. This is a well established mathematical model
widely used and validated by different authors (Aprea it et al., 2013; Engelbrecht, 2008;
Jacobsit et al. , 2014).

The convection heat transfer coefficient, hsf , is given by Eq. (3) (Kaviany, 1991; Wakao
and Kaguei, 1982), which is valid for the flow through a packed bed with Red below 8500
(Amiri and Vafai, 1998). The effective conductivities, ke.f andke.s , are estimated by Eqs.
(4) and (5) (Amiri and Vafai, 1998).
The pressure drop is calculated using the modified Ergun equation, Eq. (6) (Macdon-
ald it et al., 1979).

Eqs. (1) and (2) are numerically solved through iteration using Backward Time, Cen-
tered Space scheme (60 spatial nodes and 8000 time steps) to find steady state solid and
fluid temperatures along the bed, during a cycle. As initial condition, it is assumed that
both the solid phase and the heat transfer fluid are at thermal equilibrium with heat sink.
The boundary conditions are summarized in Table 3.1.

Table 3.1: boundary conditions

11
 Carnot efficiency used in presenting the results is defined by Eq. (7).

The variations in the flow of heat transfer fluid and the magnetic field, modeled based
on the device described by Bjrk et al. (2015)[1], are shown in Fig.3.1

Figure 3.1: Flow and field variation for each bed during a cycle
[1]

12
Chapter 4

Result and Discussion

Results are obtained with fixed hot side return temperature of 308.15 K, flow and field
variations as shown in Fig. 3.1, cycle duration of 1.5 s, and cooling load of 8.4 W. Cross
section area and length of the packed bed are 2.325X104 m2 and 0.1 m. The particles
in the bed are spheres of 6X10−4 m diameter leaving void fraction of 0.36 for the heat
transfer fluid to flow in between. The heat transfer fluid is ethylene glycol with 20%
volume concentration.

Figure 4.1: Average temperature during a cycle at each point along the bed[1]

At first, the Curie temperature of each layer was chosen as the temperature of the
magnetocaloric material in the layer averaged over both space and time during a cycle as
shown in fig 4.1. For a six-layered bed, as described in section 3, with Curie temperatures
equal to the average temperature of each layer of magnetocaloric materials during one

13
cycle and field and flow variations given by Fig. 3.1, The obtained temperature span and
Carnot efficiency with this approach are 36.4 K and 18.1

Figure 4.2: Average temperature of the bed with Curie temperature selected as in fig4.1[1]

Variations in the temperature suggest that different results can be obtained if choosing
of the Curie temperatures is done based on the bed temperatures at different moments
during a cycle instead of the whole cycle average. Fig. 4.3 shows how the temperature
span and Carnot efficiency vary as different moments of a cycle are used to indicate the
spatial average layer temperatures, chosen as Curie temperatures.

Figure 4.3: Average temperature and carnot efficiency of the bed with Curie temperature
selected as in fig4.1[1]

14
4.1 Optimization Of Layered Regenerators
From Fig. 4.3 it is clear that optimizing for large temperature span and for high Carnot
efficiency need different layer designs since the maximum efficiency is accompanied by a
relatively low temperature span and vice versa.
The Curie temperatures resulting in the highest temerature span and highest carnot
efficirency are shown in fig.4.4 and fig.4.5

Figure 4.4: curie temperature at T = 47%cycle(maximum temperature span)[1].

Figure 4.5: curie temperature at T = 21%cycle(maximum Carnot efficiency)[1].

It can be observed from fig. 4.6 that the Carnot efficiency decreases with the increase
of temperature span. The increased axial diffusion when the temperature span is large
leads to lower efficiency of the cycle.

15
 Figure 4.6: Carnot efficiency Vs tempertaure span[4].

Such an increase is due to the increased temperature gradient, not the insignificant
change in the thermal conductivity of the materials. In addition, the higher viscous
dissipation and pumping power, due to the increased viscosity of the heat transfer fluid at
the lower temperatures reached with larger spans and fixed heat sink temperature, make
the refrigerator less efficient.

Similarities between Figs. 4.1 and 4.4 reveal why the temperature span calculated by
the Curie temperatures equal to the average temperature of layers during whole cycle,
36.4 K, is so close to the maximum temperature span in Fig. 4.3, 36.6 K. It is because
the maximum temperature span, 36.6 K, is calculated when the temperature of the bed
is close to the average temperature of the whole cycle, 289.8 K

The magnetocaloric effect is rather symmetrical about the Curie temperature. There-
fore, choosing Curie temperature of each layer equal to the average temperature of the
layer during the cycle maximizes the magnetocaloric effect. As the results indicate,
the maximized magnetocaloric effect in the layers results in the highest temperature
span achieved by layering, which is in accordance with the conclusions that Arnold it
et.al.(2011)[5] have made from their experimental work.

With flow rate and magnetic field patterns of variation shown in Fig. 3.1, the maximum
Carnot efficiency is obtained when the Curie temperatures are chosen as spatial average
layer temperature plus 1.2 K.

16
Figure 4.7: Carnot efficiency Vs cycle time[1].

Figure 4.8: Temperature span Vs cycle time[1].

17
Chapter 5

CONCLUSIONS

• The optimization is done using a numerical model developed to simulate the mag-
netocaloric effect and regeneration in a multi-layered packed bed of magnetocaloric
materials. Lengths of the layers are determined by the simulation program to divide
the total temperature span equally between the layers. The simulation results for
a fixed heat sink temperature of 308.15 K and cooling load of 8.4 W for various
magnetocaloric material selections are obtained.

• The study shows that the layer design to get maximum temperature span is different
form the layer design for maximizing Carnot efficiency. In fact, higher values of one
of them lead to lower values of the other one

• Maximum temperature span can be achieved by choosing the materials which have
the highest magnetocaloric effect in the working temperature range

• The highest Carnot efficiencies are achieved by choosing materials with Curie tem-
peratures above the average layer temperature during a cycle.

• The maximum Carnot efficiency is achieved with Curie temperatures 1.2 K above
the spatial average layer temperature at the warmest moment of cycle with field
and flow variations shown in Fig. 3.1

• The study shows that the temperature span and efficiency of a magnetic refrigerator
with a fix cooling load are highly sensitive to the working temperatures.

18
References

[1] Bjorn Palm Behzad Monfared. Optimization of layered regenerator of a magnetic

refrigeration device. International Journal of Refrigeration, 57:103 – 111, 2015.

[2] Urban Tomc Alojz Poredo Andrej Kitanovski, Uros Plaznik. Present and future
caloric refrigeration and heat pump technologies. International Journal of Refrig-
eration, 57:288 – 298, 2015.

[3] Mohamed Balli Osmann Sari. From conventional to magnetic refrigerator technology.
International Journal of Refrigeration, 37:8 – 15, 2014.

[4] Urban Tomc Chiara Favero Alojz Poredo Jaka Tuek, Andrej Kitanovski. Experimental
comparision of multi-layered la-fe-co-si and single-layered gd active magnetic regener-
ators for use in a room-temperature magnetic refrigerator. International Journal of
Refrigeration, 2013.

[5] T. Nakagawa T.A. Yamamoto T. Kusunose T. Numazawa K. Mastumoto T. Irie Y. Hi-

rayama, H. Okada and E.Nakamura. Experimental study of active magnetic regener-
ator (amr) composed of spherical gdn. Cryocoolers, 16:531 – 535, 2011.

19