Florida State University Libraries
Electronic Theses, Treatises and Dissertations The Graduate School
2007 Experimantal and Theoretcal Study of Magnetic Hyperthermia Saleh Saad Al-Hayek
Follow this and additional works at the FSU Digital Library. For more information, please contact [email protected]
THE FLORIDA STATE UNIVERSITY
FAMU-FSU COLLEGE OF ENGINEERING
EXPERIMANTAL AND THEORETCAL STUDY OF MAGNETIC HYPERTHERMIA
By
SALEH SAAD AL-HAYEK
A Dissertation submitted to the Department of Mechanical Engineering in partial fulfillment of the requirements for the degree of Doctor of Philosophy
Degree Awarded: Summer Semester, 2007
Copyright © 2007 Saleh S. AL-Hayek All Rights Reserved The members of the Committee approve the Dissertation of Saleh Saad Al-Hayek defended on June, 29th, 2007.
Ching-Jen Chen Professor Directing Dissertation
Jim P. Zheng Outside Committee Member
Peter N. Kalu Committee Member
Chifu Wu Committee Member
Approved:
Chiang Shih, Chair, Department of Mechanical Engineering
Ching-Jen Chen, Dean, FAMU-FSU College of Engineering
The Office of Graduate Studies has verified and approved the above named committee members.
ii
To my mother and late father, Aminah and Saad Al-Hayek,
To my Wife, Ikhlas,
To my children, Umar and Rhunda.
iii
ACKNOWLEDGEMENTS
The writing of a dissertation can be a lonely and isolating experience, yet it is obviously not possible without the help of God Almighty: ALLAH; The Creator, The Source of Peace, The Sovereign Lord, The Merciful, The Beneficent, The Provider, The Truth, The Wise, The One, The Eternal Owner of Sovereignty, The Reckoner, The First and The Last, and then, the personal and practical support of numerous people. Thus my sincere gratitude goes first to the Creator of the heavens and earths and what’s in between; ALLAH, glorified be He, then to his final messenger, and the seal of prophets; prophet Muhammad - peace be upon him, then to my parents, my wife and children, and all my friends, for their love, support, and patience over the last few years.
This dissertation would not have been possible without the expert guidance of my esteemed advisor, Dean / Prof. Ching-Jen Chen. Not only was he readily available for me, as he so generously is for all of his students, but he always read and responded to the drafts of each chapter of my work more quickly than I could have hoped. His oral and written comments are always extremely perceptive, helpful, and appropriate.
I wish to thank Profs. Jim P. Zheng, Peter N. Kalu, and Chifu Wu, of the FAMU- FSU college of Engineering, at Florida State University, for serving in my committee, and to my brother, Prof. Yousef Haik, for inspiring and encouraging me to pursue a higher education in mechanical engineering, and for enabling me to do so.
My research for this dissertation was made more efficient but also much more extensive through the use of several characterization resources. Thus I gladly express my gratitude to Dr. Eric Lochner, the staff, and my good friend Shahid, at MARTECH, especially for the training on SQUID and X-Ray diffractometer, which I have used so frequently. Also, I would like to thank Mr. Veenu, from Quartek Int. inc., for his
iv nanoparticle synthesis insight, and Dr. Yan Xin for her help in nanoparticles TEM characterization.
Many people on the faculty and staff of the FAMU-FSU College of Engineering, Florida State University assisted and encouraged me in various ways during my course of studies. I am especially grateful to Prof. Chiang Shih, the department chairman, for providing me with the financial support and for believing in me. Many thanks go to Mr. George Green, Mrs. Humose and Liza for their departmental support, and I was also greatly inspired by Profs. Patrick Hollis, William Oates, and Peterson Hruda for whom I was a Teaching Assistant for two years, and I thank the students whom I was privileged to teach and from whom I also learned much.
v
TABLE OF CONTENTS
List of Tables ...... viii
List of Figures ...... x
Abstract ...... xiv
1. INTRODUCTION ...... 1
1.1 Motivation...... 1 1.2 Proposed Study ...... 3 1.3 Outline ...... 5
2. BACKGROUND ...... 6
2.1 Definition ...... 6 2.2 Types of Hyperthermia ...... 8 2.3 Magnetic Materials ...... 14 2.3.1 Magnetic Properties ...... 16 2.3.2 Magnetic Domains...... 20 2.3.3 Curie Temperature ...... 22 2.4 Magnetic Hyperthermia ...... 26 2.4.1 Hyperthermia Using Bulk Magnetic Materials ...... 27 2.4.2 Intracellular Hyperthermia ...... 29 2.4.3 Magnetic Fluid Hyperthermia ...... 31 2.5 Magnetic Loss Processes ...... 32 2.5.1 Hysteresis Losses………………………………………………..32 2.5.2 Relaxation……………………………………………………….35 2.5.2.1 Mechanisms of Rotation of Magnetic Moments ...... 37 2.5.2.2 Physical Basis of Heating SPM Particles……………….39 2.5.3 Eddy Current Loss…………………………………………… ...40
3. MATERIALS AND METHODS...... 43
3.1 Magnetic Nanoparticles ...... 43 3.2 Nanomagnetic Particles Fabrication and Encapsulation Methods...... 47
vi 3.3 Inorganic Nanoparticles Synthesis by Chemical Co-Precipitation ...... 56 3.3.1 Synthesis of MnZn Ferrite Nanoparticles...... 56 3.3.2 PEG Encapsulated MnZnFe Using Polymer Emulsion Process ..63 3.3.3 Synthesis of ZnGd Ferrite Nanoparticles ...... 66 3.3.4 HSA Encapsulated ZnGdFe Nanoparticles ...... 66 3.3.5 Synthesis of ZnNd Ferrite Nanoparticles ...... 67 3.3.6 Polyvinyl alcohol Encapsulated ZnNdFe Nanoparticles ...... 68 3.3.7 Synthesis of GdZnCe Ferrite Nanoparticles ...... 69 3.3.8 Ethyle Cellulose Encapsulated GdZnCeFe by Polymer Emulsion ...... 70 3.4 Size Distribution and Morphology...... 71 3.4.1 X-Ray Diffraction ...... 71 3.4.2 Transmision Electron Microscopy...... 74 3.4.1 Scanning Electron Microscopy...... 77 3.5 Nanoparticles Surface Characterization...... 79 3.6 Magnetic Heating Equipment and Instruments ...... 82 3.7 SQUID ...... 86
4. RESULTS AND OBSERVATIONS ...... 92
4.1 Powder XRD Pattern and TEM Magnetic Particles Characterization 92 4.2 Properties of Particles after Co-precipitation and Heating 107 4.3 Magnetic Properties of Particles and Magnetic Solution...... 110 4.4 Magnetic Nanoparticles SQUID Charactarization Results...... 121 4.5 Magnetic Heating Tests and Results….……...... 133
5. ANALYSIS AND ESTIMATE OF MAGNETIC HEATING ...... 142
5.1 Dominant Magnetic Heating Mechanisms ...... 142 5.2 Power Calculation Model ...... 151
6. CONCLUSIONS AND FUTURE WORK ...... 162
APPENDIX ...... 167
A Table of Conversions ...... 167
REFERENCES ...... 168
BIOGRAPHICAL SKETCH ...... 182
vii
LIST OF TABLES
Table 2-1: Typical hyperthermia strategies in oncology ...... 11
Table 2-2: Comparison of the three main techniques of magnetic hyperthermia using magnetic particles as mediators. 13
Table 2-3: Type of change of the magnetic properties of a ferromagnet with a decrease in the substance dimensions from macroscopic to atomic. 15
Table 2-4: Summary of different types of magnetic behavior...... 19
Table 3-1: Critical particle diameters for single magnetic domains in magnetic metals ...... 44
Table 3-2: Mean sizes of particles DXR obtained by the Scherrer formula from X-rays powder diffraction...... 59
Table 3-3: Chemical composition of particles co-precipitated and heated in alkaline solution...... 60
Table 3-4: The associated water content in particles coprecipitated and heated in alkaline solution. 61
Table 3-5: Properties of particles heated in alkaline solution for different time periods after co-precipitation with NaOH. 62
Table 4-1: Main structural features of the prepared GdZn-Ferrite NP’s ...... 98
Table 4-2: Characteristics of the GdZnCe-ferrite inorganic nanoparticles...... 104
Table 4-3: Main structural features of the prepared ZnNd-Ferrite NP’s ...... 107
Table 4-4: Properties of particles after coprecipitation and heating...... 108
Table 4-5: Properties of fractions of the surface treated particles separated by centrifugation for two samples: (x=0.5) and (x=0.2)...... 109
Table 4-6: Properties of ZnNd-ferrite magnetic nanoparticles...... 109
viii
Table 4-7: Properties of ZnGd-ferrite magnetic nanoparticles...... 110
Table 4-8: Main magnetic properties of the prepared MnZn-ferrite NP’s...... 122
Table 4-9: Thermocouple testing without sample ...... 135
ix
LIST OF FIGURES
Figure 2-1: Hypothetical magnetic responses associated with different classes of magnetic material ...... 18
Figure 2-2: Magnetization as a function of field at T < Tc, after cooling the sample from above Tc ...... 20
Figure 2-3: The magnetization process for a ferromagnetic sample that has cooled from above its Tc ...... 21
Figure 2-4: Above the Curie temperature (Tc) a ferromagnet is paramagnetic exhibiting Curie-Weiss behavior [χ = C/(T-θ)]...... 23
Figure 2-5: Schematic showing the atomic moments in a ferromagnet 24
Figure 2-6: Classification of Magnetic Hyperthermia ...... 27
Figure 2-7: Bone heating and corresponding temperature rise. 28
Figure 2-8: Development of magnetic materials for hyperthermia . 30
Figure 2-9: Hysteresis loop ...... 33
Figure 2-10: Schematic of: (a) Neel relaxation and (b) Brown relaxation ...... 35
Figure 2-11: Inductive heating via Eddy currents ...... 41
Figure 3-1: Behavior of superparamagnetic particles with and without the presence of an applied external magnetic field. 44
Figure 3-2: Setup used for synthesizing different nanoparticles ...... 48
Figure 3-3: Schematic of nanoparticles encapsulated in a polymer ...... 69
Figure 3-4: X-Ray Diffractometer at MARTECH ...... 72
Figure 3-5: TEM similar to the one at NHMFL ...... 75
x Figure 3-6: JEOL JEM-100C Transmission Electron Microscope at NHMFL ..77
Figure 3-7: Sample is placed insides microscope column through door ...... 78
Figure 3-8: SEM system working principle ...... 78
Figure 3-9: JEOL JSM-840 scanning electron microscope at NHMFL ...... 79
Figure 3-10: Zeta potential ...... 80
Figure 3-11: Picture of coil setup used for nanoparticles magnetic heating ...... 84
Figure 3-12: Circuit diagram for the resonance circuit ...... 86
Figure 3-13: Magnetic Property Measurement System (MPMS), SQUID ...... 87
Figure 3-14: The output of the SQUID ...... 89
Figure 3-15: A typical plot of temperature dependence of magnetization ...... 90
Figure 4-1a: Powder X-ray diffraction pattern of MnZnFe2O4 ...... 93
Figure 4-1b: Powder X-ray diffraction reference pattern for MnZnFe2O4 ...... 94
Figure 4-2: TEM micrograph for: (a) MnZn-ferrite nanopaticles and (b) PEG polymer encapsulated MnZn-ferrite particles formed by emulsion ...... 95
Figure 4-3: SEM images for two different resolutions of PEG encapsulated manganese zinc ferrite nanoparticles ...... 96
Figure 4-4a: Powder X-ray diffraction pattern of ZnGdFe2O4 ...... 99
Figure 4-4b: Powder X-ray diffraction reference pattern for ZnGdFe2O4 ...... 100
Figure 4-5: HSA encapsulated Gd-Zn-Ferrite ...... 101
Figure 4-6a: Powder X-ray diffraction pattern of GdZnCe-ferrite ...... 102
Figure 4-6b: Powder X-ray diffraction reference pattern for GdZnCe-ferrite ....103
Figure 4-7: Ethyle cellulose encapsulated GdZnCe-Ferrite ...... 101
Figure 4-8a: Powder X-ray diffraction pattern of GdZnCe-ferrite ...... 105
Figure 4-8b: Powder X-ray diffraction reference pattern for GdZnCe-ferrite ...... 106
xi Figure 4-9: Polyvinyl alcohol encapsulated ZnNd-ferrite particles ...... 104
Figure 4-10: Magnetization curves for samples S1 and S3 of MnZn-ferrite nanoparticles at T=294 K. 111
Figure 4-11: Spectra of magnetic moments of samples S1&S3 of MnZnFe nanoparticles at T=294 K ...... 111
Figure 4-12: Magnetization curves of sample S1 & S3 at T=294 K ...... 112
Figure 4-13: Spectra of magnetic moments of sample S1 & S3 at T=294 K .....113
Figure 4-14: Magnetization curves of sample S4 & S5 at T=294 K ...... 113
Figure 4-15: Spectra of magnetic moments of sample S4 & S5 at T=294 K ...... 114
Figure 4-16: Magnetization vs. temperature of samples S1, S2 and S3 115
Figure 4-17: Magnetization vs. temperature of samples S4 & S5 ...... 117
Figure 4-18: Thermomagnetic coefficients kT =-δδδσ/δσ/σ/σ/δδδδΤΤ of samples S1 & S3 (a) and S4 & S5 (b) in a temperature range of 294-373 K vs H strength ...... 119
Figure 4-19: Magnetization Curve of MnZn-ferrite nanopaticles taken by SQUID magnetometry at 300 K 124
Figure 4-20: Magnetization Curve of MnZn-ferrite at 5 K 125
Figure 4-21: ZFC&FC magnetizations as a f (T) for MnZnFe ...... 125
Figure 4-22: FcZFc at 100 Oe for ZnGd-ferrite ...... 127
Figure 4-23: Magnetization Curve of ZnGd-ferrite nanopaticles at 300 K 128
Figure 4-24: Magnetization Curve of ZnGd-ferrite nanopaticles at 5 K 128
Figure 4-25: Fc & ZFc at 100 Oe for ZnNd-ferrite ...... 130
Figure 4-26: Magnetization Curve of ZnNd-ferrite nanopaticles at 300 K 131
Figure 4-27: Magnetization Curve of ZnNd-ferrite nanopaticles at 5 K 131
Figure 4-28: Temperature dependence of magnetization for GdZnCe-ferrite ....132
Figure 4-29: Magnetization Curve of GdZnCe-ferrite nanopaticles at 5 K 132
xii
Figure 4-30: Magnetization Curve of GdZnCe-ferrite nanopaticles at 300 K 133
Figure 4-31: The heating pattern of MnZn-ferrite [Zn= 0.9 conc.] ...... 137
Figure 4-32: The heating pattern of MnZn-ferrite [Zn = 0.5 conc.] ...... 137
Figure 4-33: The heating pattern of ZnGd-ferrite, at 961 KHz ...... 139
Figure 4-34: The heating pattern of ZnGd-ferrite, at 433 KHz ...... 139
Figure 4-35: The heating pattern of GdZnCe-ferrite [Gd, x = 0.2 conc.] ...... 140
Figure 4-36: The heating pattern of ZnNd-ferrite [Zn, x = 0.2 conc.] ...... 140
Figure 5-1: Normalized plots of χ”(ω) against logf (Hz) for three samples ...... 144
Figure 5-2: Specific magnetization of the ZnNd-ferrite powder ...... 146
Figure 5-3: ACS spectra of the real part of ZnNd-ferrite fluid sample ...... 147
Figure 5-4: ACS spectra of the imaginary part of ZnNd-ferrite fluid sample ....147
Figure 5-5: ACS spectra of the real part of ZnNd-ferrite immobile sample ...... 148
Figure 5-6: ACS spectra of the imaginary part of ZnNdFe immobile sample.....148
Figure 5-7: ACS spectra of ZnNd-fe. fluid and the immobilized samples ...... 149
Figure 5-8: Specific heat power data of ZnNd-ferrite fluid by coil ...... 150
Figure 5-9: Hysteresis loops of ZnNdFe at temperatures below Curie point .....154
Figure 5-10: Calculated values of heating power from hysteresis loop ...... 157
Figure 5-11: Theoretical calculations of temperature dependence of heating power for ZnNdFe particles at different frequencies ...... 161
Figure 5-12: Theoretical line representing power calculated from Eq.21 ...... 161
xiii
ABSTRACT
Localized magnetic particle hyperthermia heating treatment using ferromagnetic and superparamagnetic nanoparticles continue to be an active area of cancer research. Magnetic hyperthermia is a promising therapeutic method for treatment of cancer. It’s based on the intratumoral deposition of biocompatible magnetic nanoparticles followed by exposure to a high-frequency electromagnetic field. The dissipation of energy connected especially with magnetic hysteresis losses, Neel and Brown relaxations results in a local heating of the active particles and consequently leads to the destruction of the cancer cells. Magnetic nanoparticle materials used has to have high specific power loss and a suitable temperature dependence of power loss allowed by an adjustment of the Curie temperature to about 315 K (43 °C). Overheating is ruled out due to a decrease of the magnetic hysteresis losses in the vicinity of the Curie temperature. One way to solve this task is the use of magnetic nanoparticles with the magnetic properties suitably modified by compositional variations. This dissertation, reports on localized magnetic hyperthermia studies using newly fabricated, as-synthesized, self-heating magnetic nanoparticles. Exposed to an alternating magnetic field, these nanoparticles act as localized heat sources at certain target regions inside the human body. Superparamagnetic nanoparticles provide attractive biotechnical and physiological advantages such as: direct injection through blood vessel due to ease of control of particle size, remote controlling of transport to tumor cells by externally applied magnetic field gradients, and resonant response to a time varying magnetic field resulting in heating up nanoparticles. In this dissertation, a report of the very promising and successful self-heating temperature rising characteristics of MnZn-ferrite, ZnGd-ferrite, GdZnCe-ferrite and ZnNd-ferrite nanoparticles obtained by chemical methods, mainly, co-precipitation process and under different applied magnetic fields and frequencies to confirm the effectiveness as hyperthermia agents. Magnetic and structural properties of these
xiv nanoparticles were analyzed in order to study the physical nature of self-heating characteristics and to investigate the effectiveness as hyperthermia agents in biomedicine. All four types of nanoparticle systems showed both superparamagnetic and ferromagnetic behaviors depending on particle sizes. Dominant magnetic heating mechanisms were studied and qualitatively identified, and a newly developed mathematical model to calculate the magnetic heating power was derived. The derived model proved to be in good agreement with the experimental results.
xv
CHAPTER 1
INTRODUCTION
1.1 Motivation
Current development in nanotechnology has provided a new set of systems, methodologies and materials for biomedical research and applications. This study is devoted to the application of nanomagnetics as hyperthermia agents for cancer therapy. The ability to control the nanomagnetic particles composition provided a unique tool to utilize their unique physical properties that have derived many biomedical applications. Nanomagnetics are currently used in hyperthermia, cell separation, diagnostics, and as contrast agents for magnetic resonance imaging - MRI.
Clinical feasibility of treating cancer by hyperthermia alone has been previously investigated [1, 2]. Most of these methods have limitations either due to the invasive thermometry or in their inability to reach optimal temperature for the tumor sites when noninvasive techniques are used especially to treat the deep seated tumors. Use of heat along with other modalities -Radiation and Chemotherapy- has been reported to improve these therapies. For example, when hypoxic cancer cells are subjected to heating and oxidant agent during irradiation the response was enhanced by a factor of 4.3 [3]. Thus the preferred hyperthermia system will need to have the ability to localize the heating at the tumor site without generating a spot overheating as well as ability to couple with other modalities of treatment.
Localized heating utilizing nanomagnetics has been studied. The particles will generate heat when subjected to alternating magnetic filed. In order to avoid spot heating the particles are synthesized with controlled Curie temperature of less than 43oC. The self control heating of the particles by limiting their Curie temperature provides a safe guard against necrosis. If these particles were to be encapsulated in a polymer along with
1 a chemo-drug or radiation-sensitizer, this will provide a unique self controlled system of localized heating.
The particles can be localized at the tumor site using a steady magnetic field either applied non-invasively or surgically at the tumor site. Once the magnetic nanoparticles system is deposited on site then an alternating magnetic field will be applied. The alternating field will cause the particles to dissipate heat to the surroundings and at the same time heat the tumor. This integrated system of self controlled magnetic hyperthermia therapy has a great potential in cancer treatment. However, the research on different components of the system is still evolving.
This study details the research that was carried out to develop such an integrated system. This included the following areas:
o Synthesis of new nanomagnetics that posses high magnetic energy, uniform size distribution - less than 20 nm - and with the preferred Curie temperature. These particles were also, be biocompatible.
o Encapsulating these particles with various polymers or shells that are able to hold the particles on site and address toxicity issues.
o Experimental and theoretical heat estimation of power dissipation from the deposited nanoparticles to the surroundings.
o Analytical formulation of magnetic heating produced in particles.
This study focuses on several issues involving use of magnetic nanoparticles in hyperthermia therapy. First, the use of hyperthermia in cancer treatment is reviewed. Then the concept of magnetic hyperthermia is introduced leading to the development of nanomagnetic hyperthermia in which the current research in magnetic hyperthermia invoking heating of nanomagnetic particles with electromagnetic waves is presented. Understanding of magnetic properties of the nanostructured magnetic materials and its fabrication is important to magnetic hyperthermia cancer treatment.
2 1.2 Proposed Study
This study lays out, both, experimental and theoretical investigations of the phenomenon of magnetic hyperthermia and its possible applications in medical treatment of malignant tumors. In order to achieve optimal and safe operational hyperthermic conditions, it is necessary to investigate what heating model or magnetic loss processes are dominant in the ensemble of nanoparticles which are injected at the cancerous tumor site. Or, if there were two or more heat loss mechanisms present what is the dominant mechanism of heating. This dissertation is, also, focused on the development of hyperthermia physics and technologies associated with deep local magnetic particle hyperthermia the goal of which is to heat tissues or tumors with therapeutic temperature elevations of 420-430C. Physically, magnetic hyperthermia in relation with medical applications is a heating phenomenon whereby a localized area’s or tissue’s temperature is elevated via electromagnetic waves. Magnetic hyperthermia has the advantage of elevating the temperature by an energy emitter located remotely outside of the body. Moreover, the heating may be localized by focusing and tuning the electromagnetic wave transmission. In order to successfully aid in constructing a magnetic hyperthermia device there are two important fundamental principles that must be studied. The first is the physical principle of delivering the electromagnetic energy to a given volume of tissues or cells which may also contain substance of enhancing the heat absorption. Second, is the understanding of the magnetic loss process of converting the electromagnetic energy into heat and elevating the temperature of the body such as tumors or tissues. This includes the selection of optimal electromagnetic wave frequency and energy transfer. In addition to the understanding of the physical principles there are several additional engineering challenges must be solved before magnetic hyperthermia can be implemented clinically. The first one is the control of the temperature and energy conversion. The second is the ability to conduct site specific magnetic heating. The technique proposed for magnetic hyperthermia consists of injecting locally the biologically compatible magnetic nanoparticles to a specific site in the body followed by heating the site with an alternating current electromagnetic field. For the best medical
3 treatment the nano-size magnetic particles encapsulated with biocompatible material are considered. In addition, for safety and practical application, the magnetic nanoparticles are best to have the Curie temperature at ~ 315K. The magnetic hyperthermia system must have a device equipped with the control of temperature and power input. Not only the temperature of the site needs to be monitored but also the applied power needs to be regulated for heating the nanoparticles. From this set up a uniform and safe heat distribution may be achieved with the use of synthesized nanoparticles having Curie point about 430C. A part of this study is devoted to synthesizing magnetic nanoparticles having a suitable Curie point. Thus, when the elctromagnetic heating reaches the Curie temperature the magnetic properties of the nanoparticles changes and the injected material is no longer heated since magnetic nanoparticles are self-regulated. The specific aims of this study, thus, involve: • Synthesis and characterization of newly developed, in-situ., as-synthesized magnetic nanoparticles. • Investigation of the optimal electromagnetic wave frequencies for magnetic hyperthermia application. • Engineering analysis of procedure for achieving a steady state temperature of 430C at the target site. • Demonstration of proof of concept experiments in terms of heating the magnetic nanoparticles via AC magnet. Control and measurement to verify that nanoparticles achieve the desired heating of (430C). • Analytical formulation of magnetic heating and comparison with experimental results.
4
1.3 Thesis Outline
This dissertation is organized as follows. Chapter 1 serves as an introduction for the dissertation motivation, outline, and its specific aims. Chapter 2 reviews the relevant issues about hyperthermia in general and magnetic hyperthermia in particular and the use of magnetic nanoparticles in its administration. The synthesis methods and characterization techniques relevant to magnetic hyperthermia are presented in chapter 3. This chapter also deals with the issue of finding adequate materials of biocompatible nature that are capable of self regulating their temperatures to up to the desired Curie temperature of 430C. Chapter 4 explains the results and observations and provides proof of concept experimentations of magnetic heating. Chapter 5 deals with how energy is imparted to the embedded magnetic nanoparticles in tumors and also with providing an estimate of power deposition and the calculation of heat generation using an AC magnet that generates electromagnetic waves. It also lays out a newly derived formula for quantifying magnetic heating and gives predictions on dominant magnetic heating mechanisms pertaining to different scenarios. Conclusions and future work of this research are devoted to Chapter 6.
5
CHAPTER 2
BACKGROUND
2.1 Definition
Hyperthermia is a heat treatment approach in cancer therapy. Cancerous cells are vulnerable at high temperatures. By rising the temperature of the target tissue to between 42°C and 46°C the viability of the cancerous cells is reduced and their sensitivity to chemotherapy and radiation is increased [3]. It has been suggested that hyperthermia may additionally stimulate activities of the host immune system against growing cancer cells. The vascular and nervous systems are not fully developed in a tumor, the tumor is thus not effectively cooled by blood flow, increasing its susceptibility to temperature. Cancer cells are destroyed at about 43°C while normal cells can survive at higher temperatures of 46°C. This offers us a window for hyperthermia therapy. Hyperthermia is one of the most promising approaches in cancer therapy: it consists in heating and destroying the target tissue to temperatures between 420C and 460C. Various methods are employed in hyperthermia, such as the use of hot water, capacitive heating, and inductive heating of malignant cells [4-7].The possibility of treating cancer by artificially induced hyperthermia has led to the development of many different devices designed to heat malignant cells while sparing surrounding healthy tissue [8-10]. Experimental investigations of the application of magnetic materials for hyperthermia date back to 1957 when Gilchrist [11], heated various tissue samples with
20-100 nm size particles of γ -Fe2O3 exposed to a 1.2 MHz magnetic field. Since then, there have been numerous publications describing a variety of schemes using different
6 types of magnetic materials, different field strengths and frequencies and different methods of encapsulation and delivery of the particles [12-30]. In broad terms, the procedure involves dispersing magnetic particles throughout the target tissue, and then applying an AC magnetic field of sufficient strength and frequency to cause the particles to heat. This heat conducts into the immediately surrounding diseased tissue whereby, if the temperature can be maintained above the therapeutic threshold of 42˚C for 30 min or more, the cancer is destroyed. Whereas the majority of hyperthermia devices are restricted in their utility because of unacceptable coincidental heating of healthy tissue, magnetic particle hyperthermia is appealing because it offers a way to ensure only the intended target tissue is heated. The problem with hyperthermia is the difficulty of delicate local heating of only the tumor region until the required temperature is reached without damaging the surrounding normal tissue. Magnetic particle hyperthermia is appealing because it offers a way to ensure that only the intended target tissue is heated. The concept is based on the principle that a magnetic particle can generate heat by hysteresis loss when placed in a high-frequency ~1 MHz magnetic field [5]. Intracellular hyperthermia using dextran nanoparticles dates from 1979 [31]. The principle of heating with superparamagnetic particles - that show no magnetic hysteresis at low frequencies - by an AC field has been reviewed by Rosensweig [32]: the dissipation results from the orientational relaxation of the particles having thermal fluctuations in a viscous medium [33]. A number of studies have demonstrated the therapeutic efficacy of this form of treatment in animal models [34], but the application of this technology to human patients is just starting [35].
7
2.2 Types of Hyperthermia
The application of hyperthermia to treat malignant tumors is not a new concept. To Hippocrates -470-377 B.C., the Father of Medicine is attributed the statement that: “Those who cannot be cured by medicine can be cured by surgery. Those who cannot be cured by surgery can be cured by heat. Those who cannot be cured by heat are probably incurable” [36]. Interest in hyperthermia was revived when Dr. W. Busch treated a patient with a facial sarcoma. This lesion disappeared in the presence of erysipelas in 1866. Since then, numerous basic studies have shown hyperthermia to be an effective modality for treating cancer. Hyperthermia treatment alone, however, will not suffice in many cases. Combined chemotherapy, irradiation, and hyperthermia has synergistic effects and the benefits are substantial [37]. The objective of hyperthermia for cancer patients is to destroy the malignant tissues while minimizing damage to the normal tissues. It is a challenge for physicists and engineers to design heating systems which allow safe application of heat to deep seated tumors at temperatures of 420-460C. The required temperatures should be maintained for a certain period of time and include the complete tumor volume regardless of its shape. In order to advance hyperthermia treatment, methods of deep heating have to be improved, the possibility of noninvasive thermometry has to be considered, and heat sensitivity needs to be defined. This chapter provides a literature preview on all types of hyperthermia in general and magnetic hyperthermia in particular. In therapeutic hyperthermia treatment a wide variety of modalities exist with corresponding technical demands that vary widely as well. The majority of technical demands of a hyperthermia treatment can be identified by classifying the style of treatment along three dimensions - the magnitude of the temperature elevation, the location of the treatment region, and the size of the treatment region. In terms of temperature elevation on the one end of the spectrum are gentle elevations of 10-20C, typically used to stimulate blood flow to muscle, while at the other
8 end of the spectrum are temperatures exceeding 1000C that are used to ablate or directly remove tumors. Sizes of treatment regions vary from the entire body -systemic- to treatment regions with dimensions on the order of centimeters - localized. With respect to location tumors within 5cm of the surface of the skin are typically classified as superficial, whereas tumors more than 5cm from the surface of the skin are considered deep. This thesis is focused on the development of hyperthermia physics and technologies associated with deep local magnetic hyperthermia the goal of which is to heat tissues or tumors with therapeutic temperature elevations of 420-430C. Although the nature of the causal link between hyperthermia and tumor response is still a subject of debate, several postulated mechanisms are in agreement with clinical observations. First, tumors may respond directly to thermal doses, either immediately - through cell membrane protein denaturation - or in a delayed manner - through unsuccessful reproduction due to genetic material damage. Alternatively, response may be generated through indirect mechanisms; tumor malnutrition induced by thermal damage to the blood vessels that supply the tumor serving as one example. An important class of indirect response mechanisms seems to be associated with the elevated oxygen levels found in tumors exposed to hyperthermia. This is one of the mechanisms through which hyperthermia is believed to provide synergistic benefit when used in conjunction with radiation and chemotherapies. In electromagnetic hyperthermia energy may be deposited in the patient through capacitive coupling, inductive coupling, or wave propagation. Capacitive and inductive hyperthermia involve relatively low frequencies - less than 25 MHz - and are sometimes referred to in the literature as RF hyperthermia. Both techniques suffer from a relative lack of selectivity in heating in comparison to the third technique, wave propagation, which enjoys the flexibility associated with the constructive and destructive interference of spatially distributed fields. Systems that exploit electromagnetic wave propagation in this way are sometimes referred to as microwave hyperthermia systems and can range in frequency from 40 MHz to 2.45 GHz. The focus of study presented herein is on electromagnetic, deep local hyperthermia. The primary advantage of this modality is the potential to exploit the photonic and wavelike natures of the energy to generate selective power deposition.
9 Additionally, electromagnetic wave propagation does not face the sharp impedance contrast offered by, for example, a bone-muscle interface to ultrasonic waves. Finally, because electromagnetic fields are vectorial there is an additional degree of freedom; the polarization of the energy which is not available in ultrasonic hyperthermia. Deep local electromagnetic hyperthermia -magnetic hyperthermia- faces two challenges. As frequency increases, the minimal achievable focus size decreases which is a desirable result. However, at the same time absorption increases which requires more power be deposited in the tissue containing the magnetic nanoparticles and the tumor; clearly an undesirable result. Therefore, depending on tumor depth, size, and location, it is essential to choose the optimal frequency before attempting treatment. Also it is important to choose magnetic nanoparticles of Curie temperature of 420C to act as temperature switch once the desired temperature is achieved. The second major difficulty facing magnetic hyperthermia stems from the fact that human tissue is heavily inhomogeneous electromagnetically. At typical ultrasound frequencies the body can be modeled as homogeneous with the exception of strongly scattering interfaces such as junctions between bone and muscle. For electromagnetic radiation at microwave frequencies no such simplifying assumption can be made; increasing the difficulty in producing accurate predictive models. Therefore, to perform a relevant electromagnetic analysis in this frequency range a detailed anatomical model - together with the corresponding electromagnetic properties - of the patient must be known. Current modalities for cancer hyperthermia may be classified according to the nature of the heating source and the heated target from whole-body to tumoral cell level (Table 2-1), [38-40]. Main heating sources fall into three categories: contact with externally heated liquid, contact-less applicator- e.g. ultrasound, microwave, radiofrequency and infrared devices, and inserted heating source - e.g. probes, antennas, laser fibers and mediators such as nanoparticles. Unlike other inserted heating sources, e.g. optical fibers, radiofrequency and microwave antennas, mediators convert the electromagnetic energy into heat when exposed to an external electrical or magnetic field. Macroscopic mediators are inserted within the body by surgical intervention. Micro- or nano-scale mediators are injected as
10 particle dispersion. They are heated either by capacitive applicators, i.e. designed for favoring the electric component of electromagnetic fields (E-field), or by inductive applicators where the electric component is lowered to the benefit of the magnetic one (H-field), [39].
Table 2-1. Typical hyperthermia strategies in Oncology [29,30] (in italics: experimental techniques which are currently undergoing preclinical evaluation)
Overheated region Hyperthermia strategy Whole-body Organ Tumor Tumor Cell Contact with a hot source Hot bath, air, Isolated organ Direct injection of wax, blanket, perfusion (e..g. hot water (960C) suits, etc. liver)
Ultrasound applicator Scanned focused ultrasound monitored by MRI
Electro- Radiative Water- Interstitial laser magnetic applicator filtered photocoagulation hyperther infrared (direct insertion of mia exposur laser fibers) Capacitive Radio- Focalized - RF- antennas based applicator frequency microwave beam on unipolar or bipolar capacitance through one interstitial electrodes. hyperthermia single element - Interstitial through two applicator microwave anntennas electrodes coupled at the coupled at the body surface body surface Inductive - Magnetic interstitial Intracellular applicator implants hyperthermi (magnetic hyperthermia a (IH) hyperthermia) - Arterial through embolization ligand- hyperthermia (AEH) mediated or direct injection magnetic hyperthermia (DIH) particle (i.v. of magnetic particles. administrati- on)
11
For capacitive hyperthermia mediators would have to be materials with high electric conductivity - heating via eddy currents, and for inductive hyperthermia they have to be magnetizable. Capacitive applicators may lead to uncontrolled heating of the body because of the tissue’s intrinsic electrical conductivity and/or to electrical field heterogeneities due to differences in tissue dielectrical permeabilities. Therefore inductive mediators seem currently more useful because tissues do not contain intrinsic magnetic materials which could deliver heat in an AC magnetic field. Interstitial macroscopic mediators for magnetic hyperthermia are generally ferromagnetic rods or seeds directly inserted into tumor tissues [40]. These thermoseeds are typically of the order of 1 mm in diameter and 1-7 cm in length. Various alloys have been used and corrosion is prevented by a protective coating or gold platting. Even if this technique has been demonstrated to work in a wide variety of human tumor types in vivo, its main limitations are stressful surgical intervention, difficult accessibility to some tumors, potential thermoseed migration and non-uniform temperature patterns. There is, therefore, a possible thermal under dosage of critical regions. It may be noticed that for a similar purpose thin sticks - 0.5 cm in length and 0.6 mm in diameter - made of carboxymethylcellulose and magnetite nanoparticles -10 nm in diameter - have been inserted stereotactically into the brain tumor of rats [41]. The advantage of such thermoseeds is their heat dissolution after AC magnetic field application allowing the magnetic nanoparticles to diffuse through the tumor. The latest magnetic hyperthermia modalities are based on micro- and nano-scale mediators in the form of an injectable colloidal dispersion of magnetic particles and may be performed according to three strategies: arterial embolization hyperthermia - AEH, direct injection hyperthermia - DIH, and intracellular hyperthermia - IH [40], (Table 2-2). Their use appears as the most promising cancer hyperthermia therapy in particular because of the better temperature homogeneity [42]. Prior to heating their distribution in tissues may be determined by MRI; taking advantage of their magnetic properties. Moreover, the intracellular route which is based on i.v.-administered stealth magnetic nanoparticles designed for selective uptake by tumor cells would be the optimal
12 method permitting to selectively overheat tumor cells even in disseminated metastases in any region of the body.
Table 2-2. Comparison of the three main techniques of magnetic hyperthermia Using magnetic particles as mediators, [30]. Arterial embolization Direct injection Intracellular hyperthermia (AEH) hyperthermia (DIH) hyperthermia (IH) Magnetic Through the arterial supply of Directly injected in the Arterial embolization or particle the tumor tumor direct injection or ideally administration i.v .injection Heat origin Intravascular (within the Extracellular Intracellular blood vessels) Nanoparticle Mono-or multi-domain Mono-or multi-domain Mono-or multi-domain design particles particles nanoparticles with stealthy corona and ligand for i.v. injection Expected - A more effective tissue - Not dependent on an -Probable improvement advantages temperature distribution due artiery pathway to the of the treatment efficacy. to the concentration gradient tumor: could be -Treatment of scattered of particles through the tumor applicable to a wide tumors and metastases. and the surrounding tissues, range of tumor types. -Lower and therefore with the highest concentration -No consequent risk from safer required magnetic in the tumor (no sharp drop in arterial catheterization field. temperature at the tumor edge) Probable -Not applicable to tumors -Restricted to tumors drawbacks without a good supply which would be (e.g. micrometastases) accurately visualized and and tumors outside the accessible under liver. radiological guidance. -Risk of embolization -Necessity of repeated and subsequent ischemic injections for large or necrosis of normal irregularly shaped tissues tumors. -Increased risk of needle track implantation or local tumor spread.
13 2.3 Magnetic Materials
The magnetic properties which show the difference between a massive - bulk - material and a nanomaterial are very important. In particular, magnetisation per atom and magnetic anisotropy of nanoparticles can be much greater than those of a bulk specimen, while differences in the Curie (TC) or Neel (TN) temperatures, i.e., the temperatures of spontaneous parallel or antiparallel orientation of spins between nanoparticle and the corresponding microscopic phases reach hundreds of degrees. The magnetic properties of nanoparticles are determined by many factors; the key of these including the chemical composition, the type and the degree of defectiveness of the crystal lattice, the particle size and shape, the morphology, the interaction of the particle with the surrounding matrix and the neighboring particles. By changing the nanoparticle size, shape, composition and structure, one can control to an extent the magnetic characteristics of the material based on them. However, these factors cannot always be controlled during the synthesis of nanoparticles nearly equal in size and chemical composition, hence, the properties of nanomaterials of the same type can be markedly different. The unique magnetic properties are usually inherent in the particles with a core size of 2-30 nm. For magnetic nanoparticles, this value coincides in the order of magnitude with the theoretical estimate for the smallest dimensions of a magnetic domain in most magnetic materials as shown in Table 2-3 below.
14
Table 2-3. Type of change of the magnetic properties of a ferromagnet with a decrease in the substance dimensions from macroscopic to atomic.
Object Characteristic size Specific magnetic properties
Macroscopic ≥ 1µm spontaneous magnetization below Tc. The (bulk) sample appearance of a nonzero magnetic moment is suppressed by the formation of a domain structure Microscopic 50-1000 nm magnetic characteristic strongly depend on sample the sample pre-history, preparation and processing method
Single-domain 1-30 nm the presence of a blocking temperature Tb particles) in a hysteresis. At a temperature higher than Tb, diamagnetic the particle transferes into the matrix superparamagnetic state. In the Tb < T < Tc region, the particle has a spontaneous magnetization and a nonzero total magnetic moment, which easily changes the orientation in the external field. Single atom ~ 0.2 nm usual paramagnetic properties (ion) 15 2.3.1 Magnetic Properties The applications of magnetic nanoparticles can be seen throughout biomedicine and other biologically related fields. Their magnetic properties allow them to be exploited in the presence of a magnetic field and used to function in areas such as hyperthermia, drug delivery and enhanced imaging for MRI. Before describing magnetic hyperthermia application a brief discussion of some of the basic concepts of magnetism is needed to better support the use of magnetic nanoparticles to drive certain processes. If a magnetic material is placed in a magnetic field of strength H, the individual atomic moments in the material contribute to its overall response, the magnetic induction: B = o (H+M) (1) where o is the permeability of free space, and the magnetization M = m/V is the magnetic moment per unit volume, where m is the magnetic moment on a volume V of the material. All materials are magnetic to a certain extent, with their response depending on their atomic structure and temperature. They may be conveniently classified in terms of their volumetric magnetic susceptibility,χ, where: M =χH (2) describes the magnetization induced in a material by H. In SI units, χ is dimensionless and both M and H are expressed in Am-1. Most materials display little magnetism and even then only in the presence of an applied field. These are classified as either paramagnetic or diamagnetic. However, some materials exhibit ordered magnetic states and are magnetic even without a field applied and these are classified as ferromagnetic, ferrimagnetic, and antiferromagnetic - prefix refers to the nature of the coupling interaction between the electrons within the material, [43]. The susceptibility in ordered materials depends not just on temperature, but also on H, which gives rise to the characteristic sigmoidal shape of the M-H curve, with M 16 approaching a saturation value at large values of H. Moreover, in ferromagnetic and ferrimagnetic materials hysteresis is commonly observed. This is the irreversibility in the magnetization process that is related to the pinning of magnetic domain walls at impurities or grain boundaries within the material as well as intrinsic effects such as the magnetic anisotropy of the crystalline lattice. This gives rise to open M-H curves called hysteresis loops, [44]. The shapes of these loops are determined in part by particle size. In large particles - micron or more - there is a multi-domain ground state that leads to a narrow hysteresis loop since it takes relatively little field energy to make the domain walls move; while for smaller particles, there is a single domain ground state which leads to a broad hysteresis loop. For nanometer sized particles, superparamagnetism can be observed; where the magnetic moment of the particle as a whole is free to fluctuate in response to thermal energy while individual atomic moments maintain their ordered state relative to each other. This leads to an anhysteretic, but sigmoidal M-H curve, [44]. Figure 2-1 shows different magnetic field response curves and is an example of how magnetic particle of varying size might behave in the blood stream. Table 2-4 summarizes the different types of magnetic behavior. Magnetic particles, such as maghemite and magnetite, have been in use since the 1970s in areas of bioscience and medicine. Their unique magnetic properties allow them to move in high magnetic field gradients making them useful in such areas as magnetic hyperthermia, drug targeting and bioseparations. Magnetic particles generate a magnetic field and influence the local area around them making for excellent contrast agents in magnetic resonance imaging -MRI, [45]. Potential applications for biosensing techniques may also employ magnetic particles by selectively binding the particles to species of interest and mobilizing them under a magnetic field. 17 Figure 2-1. Hypothetical magnetic responses associated with different classes of magnetic material ranging in size that have been injected into a blood vessel. Diamagnetic (DM), paramagnetic (PM), ferromagnetic (FM), and superparamagnetic (SPM). (after [44] ) 18 Table 2-4: Summary of different types of magnetic behavior. (http://www.aacg.bham.ac.uk/magnetic_materials/type.htm) Example / Magnetism χ Atomic / Magnetic Behaviour Susceptibility Atoms have Dia- Small & Au -2.74x10-6 no magnetic magnetism negative. Cu -0.77x10-6 moment Atoms have randomly β-Sn 0.19x10-6 Para- Small & oriented Pt 21.04x10-6 magnetism positive. magnetic Mn 66.10x10-6 moments Large & positive, Atoms have function parallel Ferro- of applied aligned Fe ~100,000 magnetism field, magnetic microstruc moments ture dependent. Atoms have mixed parallel and Antiferro- Small & anti-parallel Cr 3.6x10-6 magnetism positive. aligned magnetic moments Large & positive, Atoms have function anti-parallel Ferri- of applied Ba aligned ~3 magnetism field, ferrite magnetic microstruc moments ture dependent 19 2.3.2 Magnetic Domains When a ferromagnetic material is cooled from a temperature above its Curie temperature in H = 0, it will usually show very little evidence of having a large magnetization value. This is due to domain formation. Instead of all the magnetic moments in the sample lining up in the same direction in one single domain, the lower energy configuration is for the sample to be divided up into several or many magnetic domains. Within a domain all of the moments are aligned in the same direction - the magnetization is saturated within a domain. It is important to note that the magnetic domains are not the same as crystallographic domains - the magnetic domains cannot be seen without magnetic imaging techniques. In the border region between different magnetic domains the direction of magnetization changes. This border region is called the domain wall. It is the way that domain walls move that causes much of the irreversibility in ferromagnets. When a magnetic field is applied to a sample that has been cooled in H = 0 from above Tc, the sample will initially seem to be nonmagnetic. As a magnetic field is initially applied, the M(H) behavior looks like the initial curve identified in figure 2-2. This is called the virgin curve and the behavior associated with this curve is shown in figure 2-3. Figure 2-2. Magnetization as a function of field at T < Tc, after cooling the sample from above Tc. Identified are the initial magnetization curve (or, virgin curve), the saturation magnetization (Ms), the remanent magnetization (Mr) and the coercive field (Hc). 20 Figure 2-3. The magnetization process for a ferromagnetic sample that has cooled from above its Tc. (A) In H = 0 the magnetic domains form a closed loop with the moments aligned along preferred crystallographic directions. Within each domain the magnetization is saturated. (B) As H is applied the domain walls move to allow the domains aligned with the field to grow. (C) At a sufficiently high field only a small number of domains exist. (D) The last stage of the magnetization process is the rotation of the moments away from the preferred crystallographic direction and parallel with the applied field to form one saturated domain. Initially, the domains are directed so that the spontaneous magnetic fields form a closed magnetic loop. When a magnetic filed is applied more of the moments start to align with the field. This does not happen randomly within all domains, but rather the 21 domains with a component directed along the field direction preferentially grow in size. In figure 2-3 this is pictured as the domain wall moving. This selective domain growth continues until the whole sample is one domain. Even though the moments in a domain are parallel to each other they still may not be aligned with the magnetic field but along some preferred crystallographic direction. This is due to magnetic anisotropy. To reach full saturation requires that the moments turn from their preferred crystallographic direction to be parallel to the magnetic field. As the field is reduced from saturation the moments will first return to the preferred crystallographic direction. As the field is decreased further the domain walls will reform – or, nucleate - and try to move. The domain wall motion can, however, be strongly impeded as in the case of hard frromagnets. This impedance to domain wall motion is caused by domain wall pinning. Domain walls can get stuck at various defects, like grain boundaries and inclusions, producing the characteristics of remanence and coercivity. 2.3.3 Curie Temperature The magnetization of a ferromagnet does not return to zero by reducing H to zero. However, by applying a sufficiently large magnetic field in the opposite direction, the magnetization can be returned to zero. The magnetic field required to return the magnetization to zero is called the coercive field, Hc. The coercive field is not an intrinsic property and the value of Hc has an additional dependence on the rate of change of the magnetic field, dH/dt. The size of the coercive field determines how useful the material is for various applications. Although Hc is related to the remanent magnetization - i.e., if Mr were zero, Hc would also be zero - it is a different property. A major distinction between ferromagnets and paramagnets is that the ferromagnetic state is a state of long range order. This long range order sets in at a phase transition which occurs at the Curie temperature. The Curie temperature is close to the same point where a (1/χ vs. T) plot will extrapolate to zero for a Curie-Weiss paramagnet that becomes ferromagnetic. An example is shown in figure 2-4, where Curie-Weiss paramagnetic behavior [χ = C / (T-θ)] with θ > 0 is observed above the transition, while 22 below Tc the system is ferromagnetically ordered. Below Tc, χ is no longer a useful parameter since χ is, both, field and history dependent, and instead it is the saturation magnetization Ms that is an important intrinsic property. Figure 2-4. Above the Curie temperature (Tc) a ferromagnet is paramagnetic exhibiting Curie- Weiss behavior [χ = C/(T-θ)] with θ > 0. Below Tc, χ is no longer a useful parameter since χ is both field and history dependent. Instead, Ms is an important intrinisic property. In a ferromagnet below Tc, the individual magnetic moments of the atoms are all lined up in the same direction and essentially locked together as shown schematically in figure 2-5. Instead of magnetic moments acting individually they act together like one very large magnetic moment. The term long range order means that if we know the orientation of one moment at a particular position we can determine the orientation of any other moment a long distance away. For a ferromagnet all of the moments within a domain are aligned in the same direction. On the other hand, for a paramagnet the orientation of any moment is random and even though there is a higher probability for a moment to align with the magnetic field each moment acts nearly independently of the others. 23 Figure 2-5. Schematic diagram showing how the atomic moments in a ferromagnet are locked together and aligned in the same direction below Tc. In conclusion, Ferromagnetic materials have a definite temperature of transition at which the phenomenon of feromagnetism disappears and the material becomes paramagnetic. This temperature of transition is called the Curie temperature. Many materials will lose essentially all of their magnetism after being heated above their Curie point and then cooled. Below the Curie temperature, the ferromagnet is ordered and above it disordered. Also, the saturation magnetization Ms goes to zero at the Curie temperature. Curie temperature can be utilized to overcome the problems of uneven heating and temperature regulation. If the magnetic nanoparticles material has Curie temperature in the optimum heating range 42-43ºC, then once subjected to oscillating magnetic field the temperature will rise only up to Curie temperature. If they are further subjected to the magnetic field of any intensity they won’t be heated up thereafter as beyond Tc nanoparticles become paramagnetic. This will ensure uniform heating because now the magnetic field can be kept on till all the particles irrespective of their depth inside the body reach the optimum temperature which corresponds to their Tc. If nanoparticles with Curie temperature of 42- 43ºC are used then there would be no need to regulate the applied field which results in equipment cost reduction. 24 A note about units. When M(H) is linear and reversible it is possible to define the magnetic susceptibility as (χ = M / H). The MPMS SQUID reports the magnetic moment, m, in units of emu, a cgs unit. If this is divided by the volume (in cm3) one will have the volume magnetization M (in emu/cm3). Dividing this M by the applied field H (in Oe) gives the volume susceptibility which is dimensionless, but often expressed as emu/cm3 or emu/(cm3Oe). Similarly dividing m by the mass in grams (g) gives the mass magnetization Mg (or, σ). Dividing Mg by H in Oe gives the mass susceptibility which is expressed as emu/g. Dividing m by the number of moles gives the molar magnetization Mm in emu/mole. Dividing by H in Oe gives the molar susceptibility which is also exoressed as emu/mole. These can be converted to SI units using the table in Appendix A. It is necessary to be able to measure and express quantitatively the various characteristics of magnetism. The magnetization is a measure of the extent to which an object is magnetized. It is a measure of the magnetic dipole moment per unit volume of the object. Magnetization carries the same units as a magnetic field: amperes/meter. Conversion between CGS and SI magnetic units 25 2.4 Magnetic Hyperthermia Magnetic hyperthermia refers to the introduction of ferromagnetic or superparamagnetic particles into the tumor tissue followed by the application of an external varying magnetic field. The particles transform the energy of the magnetic field into heat by several mechanisms: eddy current loss, hysteresis loss and relaxation loss which includes Brownian relaxation and Neel relaxation. The efficiency of the transformation of energy is strongly dependent on the strength and frequency of the magnetic field, the properties of the magnetic particles and the cooling capacity of the blood flow [46,47]. Superparamagnetic materials have zero remanence and energy transformation is mainly attributed to Brownian and Neel relaxation losses. Hyperthermia treatment in cancer has definite benefits particularly when it is combined with radiation therapy or chemotherapy. This leads to many studies in developing safe and effective means of hyperthermia. One of the leading mechanisms for hyperthermia is magnetic hyperthermia. Magnetic hyperthermia is heating treatment that invokes conversion of electromagnetic energy into thermal heating and hence elevating the tumor temperature which enhances the cancer treatment when combined with radiation or chemotherapy. This section discusses several approaches in developing magnetic hyperthermia. Magnetic materials have been widely used for hyperthermia of biological tissues. This is based on the occurrence of magnetic energy losses during the alternative magnetization processes. In principle the magnetization process determines the magnetic losses. These losses, depending upon the thermal conductivity and heat capacity of the surrounding medium, can be dissipated in the form of heat raising the temperature of the surrounding. The losses of magnetic energy are of different kinds. They are determined both by the intrinsic and extrinsic properties and the particle sizes. Besides the hysteresis losses and eddy current losses for larger grains and relaxation losses for superparamagnetic particles - Neel relaxation - and frictional losses of particles - Brownian movements - have been extensively exploited for hyperthermia. 26 Depending on the approach of investigation - characteristics of magnetic heat sources - the magnetic hyperthermia can be classified as in Figure 2-6. Magnetic Hyperthermia (Magnetic Materials as Heat Source) Bulk Materials Magnetic Nanoparticles (Thermoseed, Rod, etc.) (Intracellular Hyperthermia) Colloidal Dispersion of S Encapsulated Ferro/ Ferri (Magnetic Fluid Hyperthermia) Magnetic Particles Figure 2-6. Classification of Magnetic Hyperthermia (modified from Sadhana Vol.28, parts 3&4) 2.4.1 Hyperthermia Using Bulk Magnetic Materials Magnetic hyperthermia can be induced by two different methods. In the first method finite size magnetic implants are surgically placed within the tumor site. These implants absorb energy from externally applied AC magnetic field, then they dissipate it in the form of heat to the surrounding tissues. These tissues can be destroyed if the temperature rises above 420C. A large number of bioactive/ biocompatible glass and glass ceramics have been studied for such investigations [48-52]. These are known to form bonds by the formation of an apatite layer on the surface. It is, however, difficult to get homogenous heat distribution through this method. In such a method it is expected that temperature rise will be observed close to the implanted material and there will be non-uniformity in the 27 temperature distribution in the tumor region. Figure 2-7 shows a schematic diagram for application of alternating magnetic field to a bone packed with glass ceramics. Figure 2-7. Schematic diagram showing bone heating and corresponding temperature rise (from [52]). Three different fluro-optic thermometers have been used for temperature measurements. There is a danger that the temperature may rise more than the requirement and the normal tissues may get affected. This can be avoided if the transition temperature, 0 Tc, of these thermo seeds can be tuned between 42 and 50 C. The tuned Curie temperature would act as a temperature switch during treatment and a constant temperature is maintained in the tumor region [53-55]. A major drawback of the procedure using bulk materials is that it is an invasive method and requires surgical removal after hyperthermia treatment. Therefore, repeated surgery may be required which could be traumatic. 28 2.4.2 Intracellular Hyperthermia The alternative approach is to use fine particles as heat mediators instead of needles or rods such that hyperthermia becomes noninvasive. When solutions containing magnetic nanoparticles (1-100 nm) [11], are injected, these particles are easily incorporated into the cells as their diameters are in the nanometer range. These magnetic particles selectively heat up tissues by coupling AC magnetic field to targeted magnetic nanoparticles. As a result, the whole tumor can be heated up uniformly. This is called intracellular hyperthermia. It has been shown that malignant cells take up nine times more magnetic nanoparticles than normal cells [49]. Therefore the heat generated in malignant cells is more than that in normal cells. Also, as blood supply in the cancerous tissues is not normal the heat dissipation is much slower. Hence, the temperature rise in the region of tumor is higher than in the surrounding normal tissues. It is therefore expected that this therapy is much more concentrated and localized. Hyperthermia with small particles started in 1957. Gilchrist [56], found high concentrations of magnetite particles in lymph nodes near the injection site after it was administered into subserosa of the intestine of dogs. Gilchrist observed a temperature rise of 4.70C/min, when lymph node (47mg ferrite /gm of tissue.) were subjected to AC magnetic field (15–20 kA/m, frequency 1-2 MHz). Best power absorption was reported for particle core size distribution between 20 and 100 nm. Medal [57], used an AC magnetic field at 120 kHz, with field amplitude of 37kA/m. The result was not so satisfactory since most of the animals died within 7 minutes of the application of the AC field. Later, in 1965 they published another paper [58]. The frequency used was 55 kHz and field strength was increased to 40 kA/m. The report was encouraging as no side effects were observed. Gordon [58], in 1997 treated mammary tumor-bearing rats by injecting ferrofluid of dextran magnetite. He used an AC applicator of frequency of 450 kHz - 38kA/m field strength. A ferrofluid of 100 mg was slowly injected into the tail vein over 10 minutes. After 48 hours an AC field was applied for 12 min. He observed a temperature increase of 80C/min. After one week of therapy magnetic particles were seen 29 in the liver, spleen, and kidney. In electron microscopic investigation some intracellular uptake of particles was reported. Prior to 1990, this mode of hyperthermia did not gain importance. In figure 2-8, the general flow diagram for developing suitable nano magnetic materials for magnetic intracellular hyperthermia is given. Figure 2-8. Chart for development of suitable magnetic materials for clinical hyperthermia. The larger magnetic particles show ferri- or ferromagnetic properties. Comparable heating effects may be achieved with relatively larger particles as with superparamagnetic particles [59]. These particles can be exploited for the intracellular hyperthermia treatment of cancer. Optimization of magnetic parameters of particles is essential which includes synthesizing particles having high value of specific absorption 30 rate [59, 60]. This minimizes the required dose of magnetic particles. It is, however, important that the intracellular delivery of particles is maximized. 2.4.3 Magnetic Fluid Hyperthermia Magnetic fluids can be defined as fluids consisting of ultramicroscopic particles (~100 nm) of magnetic oxide [61]. These particles are stabilized by using surfactant to prevent their agglomeration and make stable colloidal suspension in suitable medium - water or hydrocarbon. They behave like true homogeneous fluids and are highly susceptible to magnetic fields, [62]. Ferrofluid consisting of superparamagnetic particles of Fe3O4 and other magnetic particles, modified / coated with different types of biopolymer or synthetic polymer are used for hyperthermia applications, [61]. During magnetic fluid hyperthermia administration the cells may be loaded with magnetic nanoparticles by virtue of comparable dimension. That’s why it is also known as intercellular hyperthermia. Although there had been several in vitro and in vivo studies clinical applications were not thought of. Jordan in 1993, demonstrated that magnetic nanoparticles -which are superparamagnetic - can be exploited more usefully to absorb power using Brownian and Neel relaxations. It was observed that these relaxations could generate much more heat compared to conventional ferro / ferri magnetic particles exhibiting hysteresis losses. He had developed ferrofluid consisting of nanoparticls modified by aminosilan which has 10-fold higher uptake by glioblastoma cells than the normal cells, [55]. Since then, there had been number of reports which brought this therapy close to clinical trials. 31 2.5 Magnetic Loss Processes The magnetic hyperthermia is achieved through conversion of magnetic energy absorbed by the material and loss due to heat dissipation during the alternative magnetization processes. In principle the magnetization process of the magnetic material determines the magnetic energy absorption and hence the energy conversion into heating. The actual increase in temperature at a targeted site due to the conversion of magnetic energy losses also depends on the local blood circulation and the thermal conductivity and heat capacity of the surrounding medium which can diffuse the heat. The losses of magnetic energy to become thermal heating have several types. They are the hysteresis losses within the grain, eddy current losses, and relaxation losses for superparamagnetic particles known as Neel relaxation and frictional losses of particles due to Brownian movements, [44]. The magnetic energy losses also depend on the intrinsic and extrinsic properties, the particle size, and the frequency and amplitude of the electromagnetic waves. 2.5.1 Hysteresis Losses Ferromagnetic particles possess hysteretic properties when exposed to a time varying magnetic field which in turn gives rise to magnetically induced heating. The amount of heat generated per unit volume is given by the frequency multiplied by the area of the hysteresis loop (figure 2-9): PFM = µ0 f ∫ H d M (3) Equation 3 ignores other possible mechanisms for magnetically induced heating, such as eddy current heating and ferromagnetic resonance, but these are generally irrelevant in the present context. The particles used for magnetic hyperthermia are much too small and the AC field frequencies much too low for the generation of any substantial eddy currents. Ferromagnetic resonance effects may become relevant but only at frequencies more than those generally considered appropriate for magnetic hyperthermia. 32 Figure 2-9. Hysteresis Loop. For ferromagnetic particles above the suparparamagnetic size limit there is no implicit frequency dependence in the integral of Eq. 3, so PFM can be readily determined from quasi-static measurements of the hysteresis loop using, for example, a VSM or SQUID magnetometer. As mentioned earlier, the re-orientation and growth of spontaneously magnetized domains within a given ferromagnetic particle depends on both microstructural properties such as vacancies, impurities or grain boundaries, and intrinsic properties such as the magnetocrystalline anisotropy as well as the shape and size of the particle. In principle, substantial hysteresis heating of the ferromagnetic particles could be obtained using strongly anisotropic magnets; however, the constraints on the amplitude of H that can be used mean that fully saturated loops cannot be used. Minor - unsaturated - loops could be used and would give rise to heating but only at much reduced levels. In 33 fact, from equation 1, the maximum realizable PFM should involve a rectangular hysteresis loop. However, this could only be achieved with a system of uniaxial particles perfectly aligned with H; which is unfeasible. For a system of randomly aligned ferromagnetic particles the maximum that could be achieved is ~ 25% of this ideal maximum. Hysteresis losses may be determined by integrating the area of hysteresis loops, a measure of energy dissipated per cycle of magnetization reversal, [63]. It depends on the field amplitude as well as the magnetic prehistory. Above a critical particle size domain walls exist and hysteresis losses of those multi-domain particles depend on the type and configuration of wall pinning centers given by the particle structure, [64]. For low field amplitudes losses per cycle depend on the field according to a third order power law - Rayleigh law - which is explained by wall pinning. With decreasing particle size a transition to single domain particles occurs; the simplest process of magnetization reversal of which is the uniform mode, [65]. The equilibrium magnetization direction is determined by anisotropy contributions due to crystal structure, particle shape and surface. In the simplest case of uniaxial anisotropy, for an ensemble of randomly oriented, non-interacting, ellipsoidal single domain particles, hysteresis loss energy density is given by twice the anisotropy energy density K. It may be enhanced by nearly a factor of four if particle axes are aligned with the external field. Besides this uniform mode more complicated modes of magnetization reversal - e.g. curling, buckling, fanning - result from micromagnetic theory, [66]. Recently modern methods of electron microscopy offered possibilities to directly depict the processes of magnetization reversal in individual particles. For maghemite particles, Eagle and Mallinson [67], found a strong decrease of Hc with increasing particle size in the range 30-100 nm which was interpreted as an indication of a non- uniform reversal mode. For magnetite crystals in the mean diameter range above ~ 50 nm, Hc and Hr were found to decrease with increasing particle size d according to the emperical d−0.6 power law [68]; which implies a decrease of hysteresis losses in the multi-domain size range. Since, on the other hand, magnetite particles below ~ 20 nm become superparamagnetic a 34 maximum of hysteresis losses may be expected for single domain iron oxide particles near to a mean diameter of 30 nm. Comparing different types of magnetic iron oxide particles - multidomain as well as superparamagnetic one - with respect to their specific loss power [69], it was shown that hysteresis losses of different types of magnetite particles may differ by orders of magnitude in the range of field amplitudes below 10 kAm−1 due to differences of particle size, shape and microstructure. 2.5.2 Relaxation The field of magnetic particle hyperthermia has been revitalized by magnetic fluid hyperthermia where the magnetic particles are superparamagnetic nanoparticles suspended in water or a hydrocarbon fluid to make a magnetic fluid or ferrofluid [70- 73]. When a ferrofluid is removed from a magnetic field its magnetization relaxes back to zero due to the ambient thermal energy of its environment. This relaxation can correspond either to the physical rotation of the particles themselves within the fluid, or rotation of the atomic magnetic moments within each particle. Rotation of the particles is referred to as Brownian rotation while rotation of the moment within each particle is known as Neel relaxation; as depicted in figure 2-10. Figure 2-10. Schematic of: (a) Neel relaxation and (b) Brown relaxation. Each of these processes is characterized by a relaxation time: τB for the Brownian process depends on the hydrodynamic properties of the fluid; while τN for the Neel process is determined by the magnetic anisotropy energy of the superparamagnetic 35 particles relative to the thermal energy. Both Brownian and Neel processes may be present in a ferrofluid, whereas only τN is relevant in fixed superparamagnetic particles where no physical rotation of the particle is possible. The relaxation times τB and τN depend differently on particle size; losses due to Brownian rotation are generally maximized at a lower frequency than are those due to Neel relaxation for a given size. With decreasing particle size the energy barrier for magnetization reversal decrease too, hence, thermal fluctuations lead to relaxation phenomena. Therfore, quasistatically measured hysteresis loops - e.g. by VSM, or SQUID – narrow, and specific loss power (SLP) determined becomes smaller than data measured directly by calorimetry. In the case of Neel relaxation - the fluctuation of the magnetic moment direction across an anisotropy barrier - the characteristic relaxation time of a nanoparticle system is given by the ratio of the anisotropy energy KV to the thermal energy kT: −9 τ N = τ 0 exp[KV /(kT )] ( τ 0 ~ 10 s). (4) For a characteristic time of measurement τm a critical particle volume Vc may be defined by τN(Vc) = τm. For a measuring frequency of 300 kHz and a magnetic anisotropy energy density of 104 J m−3, the critical diameter is about 20 nm. In a fluid suspension of magnetic particles being characterized by a viscosity η, additionally a second relaxation mechanism occurs due to reorientation of the whole particle - Brown relaxation - with the characteristic relaxation time given by: 3 τ B = 4πηrh /(kT) (5) which was first derived by Debye [73], for rotational polarization of molecules - rh is the hydrodynamic radius which due to particle coating may be larger than the radius of the magnetic particle core. In the general case, the faster of the relaxation mechanisms is dominant and an effective relaxation time may be defined by: τ eff =τ Nτ B /(τ N +τ B ) (6) 36 A general treatment of the relaxation in ferrofluids based on a Fokker-Planck equation was given by Shliomis and Stepanov, [74]. The frequency dependence of relaxation of the particle ensemble may be well investigated experimentally by measuring spectra of the complex susceptibility. The imaginary part χ " (f) ; which is related to magnetic losses may be described by, [75]: " 2 2 χ ( f ) = χ0φ 1/( +φ ), φ = fτ ,BN χ 0 = µ0 M s V /(kT) (7) where Ms is the saturation magnetization. Accordingly, depending on the size distribution width, the spectra show more or less pronounced peaks related to Neel or Brown relaxation the position of which is given by φ = 1. The loss power density P is related to χ " (f) according to, [76]: " 2 P f ,( H) = µ0πχ ( f )H f (8) The loss power density, P (Wm−3) is related to the specific loss power SLP (W g−1) by the mean mass density of the particles. According to these equations at low frequencies (φ << 1) - in the superparamagnetic regime - losses increase with the square of frequency while for φ >>1 losses approach a frequency independent saturation value P 2 = 0πH (χ0/τ ). The strong size dependence of the relaxation times leads to a very sharp maximum of the loss power density in dependence on particle size, [69]. Accordingly, a remarkable output of heating power occurs only for particle systems with narrow size distribution with the mean diameter being well adjusted in relation to the treatment frequency. The effect of size distribution on loss power density was elucidated theoretically by Rosensweig [32]. 2.5.2.1 Mechanisms of Rotation of Magnetic Moments The typical size of the magnetic particles in a ferrofluid is on the order of 10 nm; sufficiently small for them to be magnetic monodomains. The particles have to have non- zero magnetic moments for the ferrofluid to show its magnetic properties. In presence of 37 a non-homogeneous magnetic field B(r), ferrofluids are attracted to the region where the field intensity is maximum. This happens because the magnetic moments µ, rotate to the minimum energy direction (U = -µ.B) which is parallel to the field. Then it is pulled by the force F =∇B.µ in the direction of the field gradient. Two distinct mechanisms exist for the rotation of the magnetic moments in magnetic fluids. One is the rotation of the magnetic particle inside the liquid carrier, known by the names of Debye rotation or Brownian rotation; this last name is because, even in absence of a magnetic field, the particle rotates due to the Brownian torques - molecular collisions, which causes rotational Brownian motion; the relaxation time for this rotation is, for spherical particles: τ B = 3Vη / kBT (9) where V is the particle’s volume and η the liquid’s viscosity. The other mechanism is the rotation of the magnetic moment with respect to the particle, known as Neél rotation. The relaxation time for this rotation is strongly dependent on the particle’s volume and on the temperature, namely: −1 τ N = f0 exp(KV = k BT ) (10) where f0 is the Larmor frequency and K is the anisotropy constant of the particle. In the 4 3 case of magnetite particle, K = 1.1×10 J/m , at room temperature, τN increases from 4×10-9s to 7×10-5s upon increasing the particle’s diameter from 10 nm to 20 nm. When the Neél rotation is the dominant mechanism, i.e., when the magnetic moment is quasi-free to rotate, the particle is superparamagnetic. By lowering the temperature one comes to a temperature, TB, known as blocking temperature, below which τN is larger than the typical observation times. Below TB the particle is not anymore superparamagnetic, but the magnetic fluid is still superparamagnetic because the particle and so also µ, continues to be quasi-free to rotate. Some hypothesis, which are usually made in theoretical proposals for the rotational dynamics of the superparamagnetic particles and their magnetic moments in ferrofluids are: 38 I) The particle has a symmetry axis of easy magnetization; a unit vector along which is usually denoted by c. II) The magnetic moment µ has constant modulus, µ, and rotates inside the particle in a uniaxial potential modeled by V = -K (µ.c)2 where K is known as asymmetric constant. III) The particles’ moment of inertia has a negligible contribution to the equations of rotational motion, in comparison with the Brownian torque and rotational dissipation terms. 2.5.2.2 Physical Basis of Heating Superparamagnetic Particles The physical basis of the heating of superparamagnetic particles by AC magnetic fields was reviewed by Rosensweig [32]. It is based on the Debye model and the recognition that the finite rate of change of M in a ferrofluid means that it will lag behind H. For small field amplitudes assuming small interactions between the constituent superparamagnetic particles, the response of the magnetization of a ferrofluid to an AC field can be described in terms of its complex susceptibility: χ = χ ' + i χ " (11) where both χ ' and χ " are frequency dependent. The out-of-phase χ " component results in heat generation given by, [32]: " 2 PSPM = µ0πfχ H (12) which can be interpreted physically as meaning that if M lags H there is a positive conversion of magnetic energy into internal energy. This simple theory compares favourably with experimental results, for example, in predicting a square dependence of " PSPM on H [77] and the dependence of χ on the driving frequency, [78-80]. Measurements of the heat generation from magnetic particles are usually quoted in terms of the specific absorption rate - SAR - in units of Wg−1. Multiplying SAR value by the density of the particle yields PFM and PSPM, hence, the parameter allows comparison of the efficacies of magnetic particles covering all the size ranges, [81-84]. 39 It is clear from such comparisons that most real ferromagnetic materials require applied field strengths of ~ 100 kAm−1 before they approach a fully saturated loop and therefore, only minor hysteresis loops can be utilized given the operational constraint of 15 kAm-1; giving rise to low SARs. In contrast, superparamagnetic materials are capable of generating impressive levels of heating at lower fields. For example, the best of the ferrofluids reported by Hergt [84], has a SAR of 45Wg−1 at 6.5 kAm−1 and 300 kHz which extrapolates to 209Wg−1 for 14 kAm−1, compared to 75Wg−1 at 14 kAm−1 for the best ferromagnetic magnetite sample. While all of these samples would be adequate for magnetic particle hyperthermia, importantly, it seems clear that ferrofluids and superparamagnetioc particles are more likely to offer useful heating using lower magnetic field strengths. 2.5.3 Eddy Current Loss Ferromagnetic alloys become heated by eddy current loss when placed in an oscillating magnetic induction field, as depicted in Figure 2-11. Heat is transferred to the surrounding tissue resulting in localized interstitial hyperthermia. The temperature at which the alloy changes from the ferromagnetic to the nonmagnetic or paramagnetic state is termed its Curie point. A needle-shaped alloy used in the induction of localized interstitial hyperthermia is known as a thermoseed, and magnetic induction heating of thermoseed implants can be used to produce highly localized hyperthermia in deep-seated tumors. Figure 2-11 depicts the inductive heating via Eddy currents working principle. Thermoseeds absorb energy from an externally applied magnetic induction field without being in direct contact with the source of power. Most thermoseeds lose their ferromagnetic properties at a predetermined temperature Tc within the therapeutic temperature range; and automatic temperature regulation can thereby be achieved. Hyperthermic treatment using thermoseed implants requires their correct insertion into the target tissue which can be difficult, but these implants have several advantages: the procedure does not require any wire connection between the implants and external devices; it produces automatic temperature control in the target tissue if self-regulating 40 thermoseeds are used; and target tissue can be heated repeatedly without the need for any additional devices or surgery. Figure 2-11. Inductive heating via Eddy currents. Metal thermoseeds generate heat by eddy current loss and tissues near the thermoseeds are then heated via thermal conduction. In the hyperthermic treatment of tumors by thermoseeds thermal self-regulation should be particularly beneficial. Self- regulating thermoseeds adjust their rate of heat production so as to maintain a temperature close to the Tc. Thus, if an array of thermoseeds were used for interstitial hyperthermia treatment of a tumor the temperature attained by each implant would depend on its particular environment, [85]. In summary, this chapter surveyed the current view and technology of various types of hyperthermia treatments, and in particular the magnetic hyperthermia methodology. There are several types of hyperthermia including, but not limited to, whole body, local and regional hyperthermia. Magnetic hyperthermia which is mainly focused on cancerous tissues, tumors, and tumor cells, either can be administered intravenously - through ligand mediated magnetic particles, or through direct injection. In fact, there are three main techniques of magnetic hyperthermia using magnetic particles as mediators. They are: arterial embolization hyperthermia AEH, direct injection 41 hyperthermia DIH, and Intracellular hyperthermia IH. Magnetic hyperthermia, whereby magnetic materials are used as heat source, may use bulk materials as in case of thermoseeds, rods, etc., or it may utilize the use of magnetic nanoparticles as in the case of Intracellular hyperthermia. These magnetic nanoparticles are either colloidal dispersion of superparamagnetic particles - magnetic fluid hyperthermia - or encapsulated ferro/ ferri magnetic particles. 42 CHAPTER 3 MATERIALS AND METHODS This chapter presents the chemical methods that were used to synthesize and encapsulate various magnetic nanoparticles in this study, as well as various experimental methods used for characterization of magnetic nanoparticles. Elements chosen for the material synthesis of nanoparticles were bio-compatible, and those which were not they were encapsulated with a biocompatible material. Each of these elements is more or less, present in the human body in large quantities or as trace elements. 3.1 Magnetic Nanoparticles Some basic material properties change significantly as overall size decreases from bulk to nanosize. Magnetism is one such property. Typically, macroscopic magnetic materials are separated into domains or sections where magnetic spins are cooperatively oriented in the same direction. In the presence of an external magnetic field, these domain spins will tend to align with that field producing an overall magnetic moment. Ferromagnets are materials that retain a residual magnetic moment in the absence of an external magnetic field. A critical particle diameter exists for each type of material below which domain walls cease to exist. Particles smaller than this critical particle diameter will be comprised of one single domain. Estimates for the critical particle diameters for single magnetic domains - critical size below which a particle should be single-domain - are illustrated in Table 3-1. When single domain particles are subjected to an external magnetic field the particle magnetic moments align with the field. If there is complete randomization of the 43 orientations of the particle’s magnetic moments when the applied magnetic field is removed the material is considered to be superparamagnetic, Figure 3-1. Table 3-1. Critical particle diameters for single magnetic domains in magnetic metals [86,87]. Metal Critical particle diameter (nm) Cobalt (Co) 70 Iron (Fe) 14 Nickel (Ni) 55 Iron oxide (Fe3O4) 128 Figure 3-1. Behavior of superparamagnetic particles with and without the presence of an applied external magnetic field. Magnetic nanoparticles are especially important for applications such as information storage [88], contrast agents for magnetic resonance imaging [89], and magnetic hyperthermia [90-96], e.g. ferrofluids. A ferrofluid is a colloidal suspension of single-domain magnetic particles with typical dimensions of about 10 nm dispersed in a liquid carrier, [97–99]. The liquid carrier can be polar or nonpolar. A frrofluid keeps its fluidity even if subjected to strong magnetic fields; (>> 10 kG). Ferrofluids are optically isotropic but in the presence of an external magnetic field exhibit induced birefringence, [100]. 44 In the context of the present discussion, however, only those treatments will be considered for which some special inorganic material placed inside the body is additionally exposed to the alternating magnetic field and used as a more or less localized heat source. This material is always in a solid state even if some special examples are described as fluids because the primarily heated portion in the fluid consists of small solid particles suspended in a liquid. In the case of magnetic fluids, single domain particles with high saturation magnetization, high magnetic susceptibility, and low or zero coercivities are ideal. To synthesize single domain particles control over particle size and the absence of particle aggregation is desirable. The magnetic behavior of aggregated particles deviates significantly from that of isolated single particles; resulting in inconsistent and uncontrollable behavior. Amphiphilic block copolymer micelles may be used as nano-reactors because reasonable control over particle size can be achieved and the solvated polymer serves to sterically stabilize the particles. Interestingly, magnetic nanoparticles are not only found in modern technology but also in biology. One particular species of bacteria known as magnetotactic bacteria, contain iron-rich nanoparticles that presumably allow for geomagnetic orientation as a guiding system, [101]. It can be induced that some sort of biomolecular self-assembly process promotes the synthesis of these inorganic particles within the living organism. Temperature control by means of invasively applied sensors is a serious drawback of all kinds of hyperthermia. A self-regulating regime of temperature control, however, is possible by making use of the temperature dependence of magnetic properties. Near a certain temperature Tc , the magnetization changes to zero and no magnetic losses exist above this temperature. This effect was utilized with implants made of Ni-Cu alloys [102,103], Ni-Pd alloys [103,104], non-metallic ferrite material [105], and other substances. Furthermore, there are investigations [105,106], concerning ferromagnetic amorphous metal flakes with respect to their usefulness for temperature controlled hyperthermia. All these materials, however, are not biocompatible because they contain elements, e.g. Cu and Ni, which are cytotoxic or even cancerogenic. Therefore, a reliable coating must be applied which prevents any contact with biological material. Unfortunately, no similar 45 temperature-control effect seems to be possible for the non-toxic, well known, iron-oxide material because its Curie-point is well above the therapeutic range. Difficulties experienced in providing adequately localized tissue heating have led to the development of a variety of focused radiative and interstitial hot source techniques, [107]. One approach has been to develop intravenously injected biocompatible, nanoparticles which are heated by heat losses induced in them by an externally applied alternating magnetic field. These techniques allow excellent site-specific heating and in the case of Tc ferromagnetic nanoparticles [108-110], provide automatic temperature regulation. Individual ferromagnetic substances with Tc in the range suitable for magnetic hyperthermia -e.g. 41-45 0C- do not exist. But, by combining several elements and by varying the composition of the mixture it is possible to produce alloys, amorphous structures, ferrites and other multi-component systems consisting of metals and metalloids. In any case, a large heating power of the material is desirable in order to reduce the amount of material to be applied to the malignant tumor suffering patient. The heating effect, i.e., the spatial distribution and temporal development of temperature in the tumor region must be precisely predetermined by choosing correct external technical parameters. Therefore, for a reliable dosage the underlying physics has to be thoroughly understood. The very important biological and medical problems -e.g., targeting of the particles, temperature influence on several tissues or specific cells and combination of heating with other therapeutical means- are beyond the scope of this dissertation. When cells are exposed to chemical or physical stresses they undergo alterations in the patterns of protein expression. For example, when some microbial cells are exposed to high temperatures a set of heat shock proteins is transcriptionally upregulated. In addition to heat shock, other environmental factors such as osmotic shock or nutrient limitation can play major roles in how cells respond to these stress factors. Colloid stability is another important factor that dictates the fate of nanoparticles in aqueous and physiological solutions. The stability and, more generally, the microstructure and the macroscopic states of dispersions are determined by kinetic and thermodynamic 46 considerations. Thermodynamics dictates what the equilibrium state will be whereas the kinetics determines if the equilibrium state will be reached and how quickly, [111]. The role of polymers on colloid stability is more complicated than electrostatic stability due to low molecular weight electrolytes, such KCl. According to the literature, if the added polymer moieties are polyelectrolytes then there will be a combination of electrostatic effects as well as effects that arise from the polymeric nature of the additive; this combined effect is referred to as electrostatic stabilization. Even in the case of nonionic polymers, addition of the polymer to dispersions can promote stability or destabilize the dispersion depending on the nature of the interactions between the polymer and the solvent, and between the polymer and the dispersed particles, [111]. In the case of very low polymer concentrations, bridging flocculation may occur as a polymer chain forms bridges by adsorbing on more than one particle, [111]. At higher concentrations of a polymer brush-like layers can form on the particles. These brushes can be extended over sufficiently large distances to mask out the influence of van der Waals attraction between the particles thereby, imparting stability to the dispersion. This is known as steric stabilization; and for this mechanism to occur the polymer molecules must be adsorbed or anchored on the particle surfaces. At moderate to high polymer concentrations the free polymer chains in solution may cause depletion flocculation due to the influence of the exclusion of polymer chains in the region between two particles when the particles are very close to one another, [111]. 3.2 Nanomagnetic Particles Fabrication and Encapsulation Methods The preparation of magnetic nanoparticles has been an area of considerable study. Quantum size effects and the large surface area of magnetic nanoparticles dramatically change some of the magnetic properties, and the particles exhibit superparamagnetic phenomena because each particle is considered as a single magnetic domain. A difficulty related to the nature of magnetic fluids is that the nanoparticles, which have a large ratio of surface area to volume, tend to agglomerate in order to reduce their surface energy by strong magnetic dipole- dipole attractions between particles. As a result, one of the main 47 problems associated with producing stable magnetic fluid is prevention of agglomeration during the synthesis and coating process, [112]. Figure 3-2 shows a schematic representation of the apparatus setup used for preparation of the various nanoparticles in this research. The resulting particles are hydrophilic; negatively charged in alkaline medium and positively charged in acidic medium. Their stability depends on the nature and the concentration of the counter ions. In an alkaline medium - pH > 9, polarizing or highly charged cations, such as ammonium alkaline, or alkaline-earth ions, give rise to flocculation while low polarizing cations, such as tetramethylammonium –TMA- ion, favor solution stability, [113]. Mixing is also a major factor in the preparation of stable magnetic solutions. Synthesis under vigorous mechanical stirring results in the formation of small particles by reducing the tendency to agglomerate. The use of nonmagnetic stirring is also important in preparing magnetic nanoparticles because of the influence of magnetism on particle formation. Figure 3-2. Diagram of the reactor setup used for synthesizing different nanoparticles. 48 Nanoparticles Fabrication Methods One objective of this study is to find suitable magnetic materials which will exhibit Curie temperature in the optimum range 42-43 ºC. For this a wide variety of magnetic compounds were investigated. They were encapsulated and synthesized using, mainly, chemical processes and were then tested for their Curie temperature followed by their magnetic heating analysis. Surface modification of nanoparticles is important not only for stability purposes, but also to increase their functionality. Finding a means to control size as well as maintaining stable solutions is the subject of intense study. There are two steps to fabricate nanomagnetic particles for magnetic hyperthermia application: 1. Synthesis of inorganic nanomagnetic particles. 2. Encapsulation of nanomaterials by polymer or with a suitable surfactant. In the first step, nanoparticles are mostly formed by wet chemical methods such as chemical co-precipitation [114-116], borohydride reductioin [117-119], sonochemical [120,121], vapor trapping [122], and molecular self assembly [123]. Dry methods may be used such as laser ablation, microwave plasma vapor deposition [124], carbon arc process [125]. For encapsulated magnetic particles chemical methods are mostly used. Magnetic nanoparticles that possess high magnetic susceptibility and saturation magnetization are suitable candidates for magnetic hyperthermia medical application. Taking that into consideration, magnetite is extremely suitable as it is ferromagnetic and is much stable compared to other transition metals such as Co, Ni and Fe. For in vivo applications, in particular, it is important that stable organic coating surround the magnetic particles. Mostly, Fe3O4 and γFe2O3 are being used as magnetic materials to make magnetic encapsulated particles for biomedical applications. Physical methods usually form dry powders and wet methods give rise to particles suspended in water or in any solvent but they can be transformed into dry powder also, depending on the requirement for the further encapsulation procedure. Maghmite particles are formed by co-precipitation of acidic mixture of ferrous and ferric chloride by ammonium hydroxide and with citrate ions particles were synthesized as small as 2 nm, [126]. Many of the nanoparticles presented in this study were synthesized by coupling of chemical co-precipitation [127], and ultrasonication. The particles were further coated 49 with a cationic surfactant. Many particles were very well dispersed and are in the size range of 6-10 nm as it is the case of as-synthesized MnZn-ferrites, GdZn-ferrites and ZnGdCe-ferrites, etc. In order to avoid agglomeration the magnetic particles have to be coated with a shell of an appropriate material. According to the coating, the ferrofluids are classified into two main groups: surfacted –SFF; if the coating is a surfactant molecule, and ionic - IFF; if it is an electric shell. There are essentially two methods to prepare nanoparticles by size reduction - physical method, [128] and chemical precipitation -chemical method, [129]. In size reduction magnetic powder of micron size is mixed with a solvent and a dispersant in a ball mill in order to grind for a period of several weeks. Chemical precipitation is probably the most used method to prepare magnetic nanoparticles nowadays. Different procedures have been developed to achieve this goal. In general, these procedures start with a mixture of FeCl2 and FeCl3 and water. Co- precipitation occurs with the addition of ammonium hydroxide and then the system is subjected to different procedures to peptization, magnetic separation, filtration and finally dilution. The procedure involves reactions in aqueous or non aqueous solutions containing soluble or suspended salts. Once the solution becomes supersaturated with the product the precipitate is formed by either homogeneous or heterogeneous nucleation. The formation of nuclei after formation usually proceeds by diffusion. In any case, concentration gradients and reaction temperatures are very crucial parameters in determining the growth rate of the particles, e.g. to form monodispersed particles. As an example, in order to prepare unagglomerated particles with a very narrow size distribution all the nuclei should form at nearly the same time and subsequent growth must occur without further nucleation or agglomeration of the particles. Generally, particle size and particle size distribution, physical properties such as crystallinity and crystal structure, and degree of dispersion can be affected by reaction kinetics [130, 131]. Moreover, concentration of reactants, reaction temperature, pH and the order of addition of reactants to the solution are very crucial aspects. Even though a multielement material is often made by co-precipitation of batched ions, it is not always 50 easy to co-precipitate all the desired ions simultaneously as different species may only precipitate at different pH. Hence, control of chemical homogeneity and stoichiometry requires a very precise control of reaction conditions, [130]. The major advantage of chemical synthesis is its versatility in designing and synthesizing new materials that can be refined into final product. The main merit that chemical processes offer over physical methods; as chemical synthesis offers mixing at the molecular level, is excellent chemical homogeneity. Molecular chemistry can be designed to prepare new materials by understanding how matter is assembled on an atomic and molecular level and the consequent effects on the desired material macroscopic properties. A basic understanding of the principles of crystal chemistry, thermodynamics, phase equilibrium and reaction kinetics becomes really important to take advantage of the many benefits that chemical processing has to offer. However, there are certain hurdles in chemical processing. In some synthesis, the chemistry is very hazardous and complex. Contamination may also result from byproducts being generated or side reactions in the chemical process. It ought to be minimized in order to obtain preferred properties in the final product. Agglomeration can also be a major drawback in any phase of the synthetic process and it can terribly alter the properties of the materials. Chemical Co-precipitation Method The chemical co-precipitation process is the most widely used for the commercial production of inorganic nanoparticles. In this process the salt solution of the required metallic elements is reduced by NaOH solution. The reactants when mixed are at a temperature of 90ºC. After mixing the reaction is continued for ~ 40 minutes along with heating at 90ºC. Co-precipitation is widely used to synthesize magnetite ferrite, e.g. (Fe3O4) nanoparticles. It is also used to make many other ferrites such as zinc ferrite, manganeasen-zinc ferrite, copper ferrite, etc. Ferrite nanoparticles are obtained by the co- precipitation from aqueous solutions of trivalent Fe3+ and bivalent metal Me2+, whereby 51 Fe2+, Mn2+, Co2+, Ni2+ and/or Zn2+ may serve as Me2+. The initial molar proportion (Me2+/Fe3+) is always taken as the stoichiometric ½. The co-precipitation reaction takes place in two steps: firstly, solid hydroxides of metals in the form of colloidal particles are obtained by the co-precipitation of metal cations in alkaline medium. This is referred to as Co-precipitation step. For the case of MnZn-Ferrite, this reaction takes place as follows: 2+ 2+ 3+ − (1− x)Mn + xZn + 2Fe + 8OH (1− x)Mn(OH)2 .xZn(OH)2 .2Fe(OH)3 Secondly, the product is subjected to heating in the precipitation alkaline solution to provide the transformation of solid solution of metal hydroxides to the MnZn-Ferrite. This step is referred to as ferritisation step: (1− x)Mn(OH)2 .xZn(OH)2 .2Fe(OH)3 Mn(1-x)Znx Fe2 O4 .nH2 O + (4-n) H2O An interesting feature of co-precipitation method is that the product will contain some amount of associated water even after heating in alkaline solution for a long period of time. The rate of mixing of reagents plays a vital role in the size of the resultant particles. Co-precipitation comprises two processes: nucleation i.e. formation of centers of crystallization followed by growth of particles. Relative rates of these two processes decide the size and polydispersity of precipitated particles. Polydispersed colloids are obtained as a result of simultaneous formation of new nuclei and growth of the earlier formed particles. A less dispersed in size colloid is formed when the rate of nucleation is high and the rate of particles growth is low. This situation corresponds to a rapid addition and a vigorous mixing of reagents in the reaction. Slow addition of reagents in the co-precipitation reaction leads to formation of bigger nuclei than rapid addition. Also, in the case of slow addition of base to solution of metal salts a separate precipitation takes place due to different pH of precipitation for different metals. Separate precipitation may increase chemical inhomogenity in the particles. To obtain ferrite particles of a smaller size less dispersed in size and more 52 chemically homogeneous, the mixing of reagents must be performed as rapid as possible, [131]. An increase in temperature, e.g. in the range 20-100°C, may significantly accelerates formation of ferrite particles. The activation energy for formation of ferrites of different metals is not equal. It should be concluded from this study that heating at temperatures close to 90°C is preferable for an easier and more rapid formation of the MnZn-ferrite particles. Other chemical methods that may be found in literature include but are not limited to: borohydried reduction and refluxing in polyol. Briefly, in borohydried reduction process, salts of desired metallic elements are reduced by sodium borohydride (NaBH4). The procedure involves a drop-wise addition of aqueous solution of metallic salts to NaBH4 solution accompanied with vigorous stirring. The pH value of the salt solution should be kept at 6 while that of NaBH4 at 12. NaOH can be added to the NaBH4 solution to increase the pH to this level. Also the reaction should be carried out in argon atmosphere by passing argon into the flask during reaction. In refluxing in polyol process, liquid polyols such as ethylene glycol or diethylene are used both as a solvent and as a reducing agent for the chemical preparation of metallic powders from various inorganic precursors, [132]. The basic reaction scheme for the synthesis of these metal powders by the polyol process involves: dissolution of the solid precursor followed by reduction of the dissolved metallic species by the polyol itself. Then, nucleation of the metallic phase and growth of the nuclei occur. In order to obtain metallic powders with a narrow size distribution, two conditions must be satisfied: first, a complete separation of the nucleation and growth steps is required, and, second, the aggregation of metal particles must be avoided during the nucleation and growth steps. General procedure for synthesis of different metallic powder systems comprised of suspending corresponding metal precursors in ethylene glycol or tetraethylene glycol and subsequently bringing the resulting mixture to refluxing temperature ~ 120 to 200°C for one to three hours. During this reaction time, the metallic moieties are precipitated out of the mixture. The metal-glycol mixture is then cooled to room temperature, filtered, and the collected precipitate is dried in air. Polyol approach results in synthesis of metallic 53 nanoparticles protected by surface adsorbed glycol thus minimizing the oxidation problem, [133]. Nanoparticle Encapsulation Methods In many fields of biomedical applications, a drug, protein, magnetic material hormone, peptide, or other agents is encapsulated in a polymer matrix and delivered to a site either instantaneously or in a controlled manner in response to some external stimulus - i.e. Ph, temperature, magnetic field, radiation, concentration gradients, etc. Many micro-encapsulation techniques exist which can produce a variety of particle types and sizes under various conditions. Methods, typically involve solidifying emulsified liquid polymer droplets by changing temperature, evaporating solvent, or adding chemical cross-linking agents. Physical and chemical properties of the encapsulant and material to be encapsulated can sometimes suggest the most suitable method for encapsulation; making only certain methodologies useful in certain circumstances. A brief description of many existing methods for making micron/sub-micron particles is presented, herein. 1) Solvent evaporation: A polymer is dissolved in a volatile organic solvent, such as methylene chloride. A substance to be incorporated is added to the solution, and the mixture is suspended in an aqueous solution that contains a surface active agent such as poly -vinyl alcohol. The resulting emulsion is stirred until most of the organic solvent evaporated; leaving solid microspheres. Both, nano and micron size particles can be produced by this process. This method is useful for relatively stable polymers like polyesters and polystyrene, [134]. 2) Phase inversion: A polymer is dissolved in a good solvent, fine particles of a substance to be incorporated are dissolved in the polymer solution then, the mixture is poured into a strong non-solvent for the polymer; to spontaneously produce polymer particles. Polymer is either coated on the particles or the particles are dispersed in the polymer. In general, droplets of polymer solution are formed by forcing them through a spinner and letting them in contact with a polymer non 54 solvent which is highly miscible with the polymer solvent; thereby causing rapid precipitation of the outer layer of the droplet. The microcapsules should be left in contact with the nonsolvent until substantially all of the solvent has been replaced with nonsolvent. This process requires formation of a droplet with dimensions established prior to contacting the nonsolvent. The method can be used to produce microparticles in a wide range of sizes, say, 100 nm-10µm. Examplary polymers which can be used include polyvinylphenol and polyactic acid. 3) Solvent-nonsolvent temperature induced crystallization: Coupling of ultrasonication and nonsolvent temperature induced crystallization process to synthesize magnetic nanoparticles encapsulated by polymers is a feasible technique. In this process, high yield of crystalline polymers that are insoluble at room temperature can be obtained, [127]. A polymer and a magnetic material are dissolved in a solvent with high boiling point and then added to a nonsolvent at high temperature. Then the mixture is cooled suddenly to 00C; resulting in formation of an emulsion separating two liquids. The polymer and the magnetic material form a brown layer at the junction of the two liquids. Top and bottom layers of solvent-nonsolvent mixture pipetted out and the particles are washed with acetone and ultrasonicated. This method is usful in making nanomagnetic encapsulated particles with a crystalline polymer. Magnetic nanoparticles may be coated using polymer/proteins by forming nano/micro spheres in a suspension containing them. Upon spheres formation, magnetic particles are captured inside and encapsulation occurs. The colloidal dispersions will posses a vast large interface area between the dispersed phase and the continuous phase. Colloids may display significant kinetic stability that prevents their aggregation. Nanoparticles production depends upon: the chemical production of colloidal dispersions, their kinetic stabilization, and effective recovery of the final formulates. Polymer materials consist of large molecules whose solution allow for preparation of 55 stable and size controlled colloidal dispersions. Several polymers can be used as stabilizers of colloidal dispersions as they provide a surface coating of the metastable microphase lowering its tendency to phase-aggregation. Coacervation is the externally- induced separation of at least two phases; a common feature of all methods for the preparation of nanoparticles. Many encapsulation methods, such as: solvent displacement, salting out, emulsion diffusion, solvent evaporation and polymer emulsion method were utilized in this study and will be discussed in the context of encapsulating the various as-synthesized magnetic nanoparticles. 3.3 Inorganic Nanoparticle Synthesis by Chemical Co-Precipitation The following nanoparticle systems were studied and synthesized using chemical co-precipitation method and finally, encapsulated with a suitable polymer. The choice of encapsulation method relied upon suitability and mere exploitation of the process. 3.3.1 Synthesis of MnZn Ferrite Magnetic Nanoparticles Spinel ferrites, for example, MnZn-ferrite, are soft-magnetic materials characterized by a high magnetic polarization and a high electrical resistivity, [135]. When these materials are prepared in the form of nanoparticles, they are superparamagnetic, and can be used for the preparation of magnetic fluids. Such fluids are prepared by distributing nanoparticles with a aprticle size of the order of 10 nm, and with a narrow size distribution in a carrier liquid. These magnetic fluids have the potential to be used in a variety of modern technologies: from medicine and pharmacy to magnetic hyperthermia. Precipitation in a water-in-oil microemulsion has been shown to be a very promising technique for preparing monodisperse, ultrafine particles of controlled size and morphology. In this method, co-precipitation occurs in tiny droplets of water embedded with a surfactant; so called reverse micelles which are distributed in an oil phase. The water pools of reverse micelles act as micro-reactors for the synthesis of the particles in which the particle size of the product is controlled by the size of these pools. The size of 56 the reverse micells is thermodynamically determined in particular by the water -to- surfactant molar ratio, [136]. The preparation of nanosized, superparamagnetic spinel MnZn-ferrite particles using coprecipitation in microemulsions is a two-step process. In the first step, precursor hydroxides of Mn, Zn and Fe are precipitated in the environment of the microemulsion. In the second step, hydroxides can be calcinated to obtain the final ferrite product, [137,138]. Alternatively, the final ferrite product can be obtained directly in the microemulsion in situ with oxidation of precursor hydroxides, when Fe (II) hydroxide is used [139,140], according to Shkorr’s reaction [141]: 2+ 2+ 2+ - ½ Mn + ½ Zn + 2Fe + 6OH + ½ O2 Mn0.5Zn0.5Fe2O4 + 3H2O In this study, MnZn-ferrite nanoparticles were prepared in situ by the oxidation of Fe (II) hydroxide in a microemulsion consisting of hexanol as the oil phase, n-hexadecil trimethylamonium bromide –CTAB- as the surfactant, and various solutions of reactants as the aqueous phase. In previous studies, it was determined that the pH value after the precipitation of hydroxides should be above approximately 8 to produce the spinel ferrite product after the oxidation of the Fe (II) hydroxide; at lower pH values, FeOOH was obtained as the major product, [139]. Also, the influence of the composition of the microemulsion and the concentration of the reactants in the aqueous solutions of the microemulsions on the nature of the produced MnZn-ferrite nanoparticles were studied. MnZn-ferrites has been pursued due to their high sensitivity of magnetization to temperature. Magnetic nanoparticle solutions constituted by these ferrites should be good candidates for magnetic hyperthermia application. They retain the properties of a fluid even in the presence of high magnetic fields and particles do not separate from the carrier liquid. An easy and convenient chemical synthesis of ionic magnetic nanoparticle solutions was proposed by Massart in the early 80s; nanoparticles are precipitated and then peptized using an appropriate particle surface treatment, [142]. Surfacted magnetic nanoparticle solutions based on low-evaporating non-polar liquids keep their fluid properties over a rather wide temperature range which is important for practical magnetic hyperthermia application. Details of a preparation process for fabrication of MnZn-ferrites, as well as investigation of some physical and 57 chemical properties of colloidal ferrite particles with different degrees of Zn substitution, are presented, herein. Preparation of magnetic nanoparticle solutions requires magnetic nanoparticle synthesis and then the formation of stable colloidal solution. Magnetic particles must be chemically stable in the liquid carrier and have a convenient size to provide colloidally stable ferrofluid. For Mn and Co ferrites, suitable particle sizes must not exceed 11–12 nm, [143]. As MnZn-ferrite has approximately the same specific gravity as Mn and Co ferrites (~ 5 g/cm3), this size range is transferable to the MnZn-ferrite particles. Synthesis of MnZn-ferrite nanoparticles with size in the range 6–20 nm by chemical co- precipitation method was achieved. However, control of MnZn-ferrite nanoparticles size by variation of synthesis conditions still remains a serious drawback. MnZn-ferrite nanoparticles are obtained by chemical co-precipitation and ferritization. These are two sequential processes that include co-precipitation of metal salts into hydroxides which occurs immediately followed by transformation of hydroxides into ferrite which starts at co-precipitation but requires certain time and heating to complete, [144, 145]. Samples after co-precipitation and ferritization are called coprecipitated and heated because an ionic ferrofluid formation of a stable colloidal solution is realized by surface treatment and stabilization, [143]. This involves coating particles with a suitable surface charge density to prevent agglomeration. Henceforth samples are named surface treated. MnZn-ferrite particles are prepared by co-precipitation of aqueous solutions of MnCl2, ZnCl2 and FeCl3 mixtures in an alkaline medium. Initial molar proportion ([Mn]+[Zn]) / [Fe]) is always taken as the stoichiometric 1/2. Initial concentration of salts is taken 0.15 mol/liter of total metal content Me ([Me] = [Fe]+[Mn]+[Zn]). Three different bases can be taken as co-precipitating agents: sodium hydroxide - NaOH, methylamine -CH3NH2- and ammonia -NH3. For NaOH, the reaction time is 40 minutes, for CH3NH2 and NH3 the reaction times are 30 and 10 minutes respectively. Formation of MnZn-ferrite particles requires the heating of the solution at elevated temperatures. For co-precipitation with NaOH, conditions close to boiling are preferable. 58 0 0 For CH3NH2 and NH3, reaction temperatures may not exceed 900 and 600 C, respectively, due to high evaporation of these bases. Zn Substitution Degree 2+ 2+ Compositions with different initial degrees of Zn substitution x [Mn ]1-x [Zn ]x 3+ [Fe ]2.0, where x is 0; 0.2; 0.4; 0.5; 0.6; 0.8; 1.0, are taken for initial mixtures of metal salts. For NaOH and CH3NH2, it was found that increased x leads to formation of particles of smaller sizes. Mean sizes of particles determined by the Scherrer formula from X-ray powder diffraction data are summarized in Table 3-2. Mean size of Mn-ferrite particles (x = 0) exceeds that of Zn ferrite (x = 1) by a factor of ~ 3 ; co-precipitation with NaOH. Mn ferrite particles co-precipitated with NaOH have mean size ~ 20 nm and are too large to form a colloidally stable ferrofluid. Table 3-2: Mean sizes of particles DXR obtained by the Scherrer formula from X-rays powder diffraction. DXR values are shown for coprecipitated and heated in an alkaline solution particles as well as for particles subjected to additional surface treatment. Coprecipitation base NaOH CH3NH2 Sample after Sample after Coprecipitation Coprecipitation and heating in an and heating in an alkaline solution alkaline solution Zn-Substitution Degree (initial) x DXR (nm) DXR (nm) 0.0 18.8 9.9 0.2 11.0 6.8 0.4 9.9 6.5 0.5 8.6 - 0.6 7.3 5.6 0.8 6.6 - 1.0 6.2 - 59 Chemical compositions of co-precipitated particles are found to be close to the initial chemical compositions of the mixtures of metal salts, as shown in Table 3-3. Table 3-3. Chemical composition of particles co-precipitated and heated in alkaline solution. Initial chemical composition of a reacting mixture: [Mn2+] = 50(1-x), [Zn2+]=50x, 2[Fe3+] = 50 in mol %, where x is the initial Zn substitution degree Initial mixture of reagents Sample after co-precipitation and heating Zn Metal content (mol %) Zn substitution substitution MnO ZnO Fe O degree x 2 3 degree x 0.20 39.0 10.0 51.0 0.20 0.40 29.5 19.5 51.0 0.40 0.50 24.5 22.0 53.5 0.47 0.60 20.0 27.5 52.5 0.58 0.80 10.5 37.5 52.0 0.78 0.0 50.9 0.0 49.1 0.0 0.20 41.2 7.7 51.1 0.16 0.50 26.1 22.3 51.6 0.49 Increasing the degree of Zn substitution affects the content of associated water in the particles. Associated water content in particles grows with increasing Zn substitution x, as shown in Table 3-4. This agrees with observations in [144] and may be related to the higher affinity of Zn with respect to water. Reduction in the particle size with increasing x must also be taken into account. 60 Table 3-4. The associated water content in particles coprecipitated and heated in alkaline solution. Data for samples with different Zn substitution degree x coprecipitated with two different bases NaOH or CH2NH3 are presented. Water content is determined by calculation from metals content. Content of associated Co-precipitation base Zn substitution degree x water (w %) via calculation NaOH 0.0 6.0 NaOH 0.40 5.6 NaOH 0.47 10.0 NaOH 0.58 10.6 NaOH 0.78 12.7 CH2NH3 0.0 6.5 CH2NH3 0.47 13.9 With regard to the nature of the co-precipitating base, three different bases were used as co-precipitating agents: NaOH, CH3NH2 and NH3. In the case of NaOH, concentration of the base is ~ 0:5 mol/l and provides a pH =13 after co-precipitation. For the two other bases, CH3NH2 and NH3, concentration of the base is 1 mol/l and pH after co-precipitation is 11 and 9, respectively. The nature of the co-precipitating base has a significant influence on the size of the ferrite particles. For Mn-ferrite, it is shown that three bases result in particles with decreasing size in the following sequence: DNaOH > DCH3NH2 > DNH3 , [143]. For MnZn- ferrite, the same trend holds - see Table 3-2 .Co-precipitation with NaOH leads to the formation of ferrite for the entire range of x . For CH3NH2 and NH3, formation of ferrite particles takes place only within a limited range of x. When the initial Zn substitution exceeds a definite threshold, a nonmagnetic roentgenamorphous product is obtained. This 2+ 2+ may be due to formation of soluble complexes of Zn and partly Mn by NH3 and CH3NH2, [146]. 61 Formation of an amorphous precipitate was also reported for the case of Fe3+ and Zn2+ 0 co-precipitated with NH3 and heated at 60 C for several hours [147]. Ferrite particles obtained under boundary conditions of ferrite formation zones, for example, sample at x=0.5 co-precipitated with CH3NH2, demonstrate a reduced resistance to acid attack. The formation reaction of MnZn-ferrite requires certain conditions which depend mainly on temperature and pH value, [144]. The influence of heating time on the properties of resulting particles is studied for samples co-precipitated by NaOH with an initial Zn substitution of x=0.5. Samples are boiled for different times at 95–1000C in a precipitation solution - pH ~ 13- under vigorous stirring. No any significant changes in the chemical composition of particles are found in the process of boiling as depicted in Table 3-5. Table 3-5. Properties of particles heated in alkaline solution for different time periods after co- precipitation with NaOH. Chemical composition, calculated content of associated water in the particles, particles mean size DXR, obtained from X-rays powder diffraction by the Scherrer formula and size limits from transmission electronic microscopy DTEM Content of metals (mol %) Content of Time of associated MnO ZnO Fe2O3 DXR DTEM heating water (nm) (nm) (min) (w %) 1 23.3 22.2 54.5 22.0 4.9 2.0-4.0 20 24.0 23.3 52.7 9.8 9.0 2.5-9.5 40 23.1 23.3 53.6 11.8 8.2 3.0-10.0 90 24.0 23.3 52.7 12.5 8.6 5.0-10.0 Particles after 1 min of boiling show hardly pronounced widened peaks of spinel structure whereas already after 20 min of boiling these peaks become well pronounced. Mean sizes DXR, estimated from X-rays powder diffraction and size limits estimations DEM made from electronic microscopy pictures, show that significant growth of particle 62 size occurs in the first 20 min, Table 3-5. It should be mentioned that DXR is near upper limit of DTEM. Associated water content in particles is found to significantly decrease in the first 20 minutes. For 20 minutes boiled particles evaporation of about 70% of the associated water takes place in the temperature range of 20-2000C, whereas the remaining 30% evaporates at higher temperatures. Further heating of precipitate in an alkaline solution at 1000C for up to 90 min fails to change the associated water content within the particles. 3.3.2 Peg Encapsulated MnZnFe Using Polymer Emulsion Process Polymers and paraffin wax have been traditionally considered as good carrier matrices for nanoparticles. Paraffin wax has been the carrier of choice for studies exploring relaxation phenomena with varying particle concentrations. On the other hand, several advanced polymer composites have been synthesized with a wide variety of inclusions like metals, semiconductors, carbon nanotubes and magnetic nanoparticles. Many attractive properties of polymers like non-corrosiveness, light weight, mechanical strength and dielectric tunability can be utilized along with novel magnetic and optical properties of nanoparticles to make multifunctional materials. The inclusion of ferromagnetic or superparamagnetic nanoparticles in polymers is especially important as magnetic nanoparticles have shown promise in various potential applications like carriers for drug delivery, magnetic recording media, highfrequency applications, etc. The composite system of conducting polymers and ferrite nanoparticles can be used as an effective coating for both magnetic and high frequency electromagnetic shielding. Ferrite particles have an added advantage of reducing the eddy current losses due to their high electrical resistance. Some of the soft ferrites like manganese zinc ferrites was extensively studied for RF applications such as magnetic hyperthermia. Polymer emulsion process comprises of dispersing a liquefied polymer phase in an aqueous liquid medium phase containing at least one nonionic, anionic or cationic oil- in-water functioning emulsifying agent, in the presence of a compound selected from the group consisting of those hydrocarbons and hydrocarbyl alcohols, ethers, alcohol esters, amines, halides and carboxylic terminal aliphatic hydrocarbyl group of at least 8 carbon atoms, and mixtures thereof, and subjecting the resulting crude emulsion to the action of 63 comminuting forces sufficient to enable the production of an aqueous emulsion containing polymer particles averaging less than about 0.5 micrometer in size. The solvent is preferably one which may be readily removed. To remove the solvent a vacuum evaporation technique is usually employed. The solvent should be devoid of an aliphatic hydrocarbyl group of 8 or more carbon atoms and inert in the emulsion. The oil- in-water functioning emulsifying agents employed in the process are generally surface active agents also useful in the detergents field. The emulsifying agent is operative in the aqueous medium in relatively low concentrations, being generally included therein in proportions of about 0.1 to about 5%, desirably ~ 0.2 - 3% by weight of water in the aqueous phase. The higher aliphatic hydrocarbyl-containing additive compound, or mixture thereof, required in carrying out the process is generally employed in proportions of about 0.2 % - 12%, preferably about 0.4% - 6%, by weight of the polymer phase. These additives should be relatively highly water-insoluble. They should not have too much high a molecular weight preferable not more than about 2,000. Furthermore, when the additive is included in the aqueous phase, its weight ratio relative to the emulsifying agent should be more than 1:1, i.e. ~ 4:1. When included in the polymer phase the ratio is about 0.3:1 - 1:1 in many cases. These additive compounds increase the stability of these fine-sized particle emulsions by inhibiting sedimentation or degradation caused by the tendency of the small particles or droplets to coalesce molecularly. They should hence be inert, and resistant to diffusion into the aqueous medium phase and to any solvent removal procedures applied after the comminuting step. In this process the polymer phase is admixed with stirring or other agitation into the aqueous medium phase containing the emulsifying agent and the magnetic particles desirably at temperatures above room temperatures to below the boiling point of the aqueous medium. It is preferable to produce emulsions with as high a solid content as possible therefore, the ratio of the liquefied polymer phase to the aqueous medium phase should be as high as possible without of course introducing inversion possibilities, i.e. emulsification of the aqueous phase in the polymer phase. The weight ratio of the polymer phase to the aqueous phase generally will range from about 0.2 - 1:1. 64 The resulting crude emulsion of coarse polymer phase droplets containing the magnetic nanoparticles is then subjected to the action of comminuting forces sufficient to enable the production of an aqueous emulsion containing polymer particles containing the magnetic nanoparticles. An ultrasonicator was used in this study to supply this force. The above described crude emulsion is passed through such comminuting device a sufficient number of times, until an emulsion is obtained containing small size polymer phase particles. This is followed by the stripping or removal of the solvent from the emulsion e.g. by vacuum evaporation technique. In this study the polymer emulsification process was used to synthesize PEG and Ethyl cellulose coated magnetic nanoparticles. The Polymer emulsion method described above was used to encapsulate MnZn- ferrite nanoparticles inside polyethylene glycol having a melting point of 40°C. The following ingredients were used: Polymer Phase: Polymer: Polyethylene glycol (PEG) MW: 1,540: 2gm. Solvent: Methylene Chloride: 10 ml, 13.2 gm. Aqueous medium phase: Water: 40 ml Emulsifying agent: Sodium dodecyl sulphate: 0.33 gm. Inhibitor compound: 1-Octanol: 1.1 ml, 1.32 gm Magnetic particles: MnZn-ferrite: 50 mg. The magnetic particle: polymer ratio was approximately 1:40. The sodium dodecyl sulphate and 1-octanol were dissolved in 40 ml of distilled water using a magnetic stirrer. Then 50 mg of MnZn-ferrite was added. The polymer phase was prepared by dissolving 2 gm of PEG into 10 ml of methylene chloride. A crude emulsion was formed by adding the polymer phase to the aqueous medium phase. The crude emulsion was sonicated using an ultrasonicator 5 times in steps of 3 minutes. The resultant emulsion was then stirred inside a round bottom flask for 12 hours at 700 rpm. The solvent was then removed using vacuum evaporation method. The polymer encapsulated particles thus formed were washed with acetone and stored under PBS buffer solution. The PEG encapsulated particles as observed under a TEM are shown in chapter 4. The images show PEG spheres of varying sizes from 20 nm to 200 nm. The 65 light shaded circles are the PEG spheres whereas the dark regions inside the PEG spheres are the MnZn-ferrite nanoparticles. 3.3.3 Synthesis of ZnGd Ferrite Magnetic Nanoparticles To prepare ZnGd-ferrite nanoparticles co-precipitation process was, also, adopted as the synthesis protocol. Ferrous sulfate hepta-hydrate (FeSO4), Zinc sulfate ZnSO4.7H2O and Gadolinium chloride hexa-hydrate GdCl3.6H2O were used to obtain Zn2+, Gd3+ and Fe3+ ions in aqueous solutions. The solutions containing these ions were mixed in an appropriate molar proportion in distilled water, filtered and added to 8 M sodium hydroxide NaOH solution at the temperature ~ 900C. Washing with distilled water and acetone and then drying procedures were, then followed. It was observed from the study of the MnZn-Ferrite particles and Gd substituted MnZn-Ferrite nanoparticles that addition of Gd results in an increase in the Curie temperature as well as its pyromagnetic coefficient. This result was also confirmed by Upadhyay, [148]. Thus in this study, Gd substituted Zn Ferrite nanoparticles were synthesized using chemical co-precipitation method so as to increase the Curie temperature with the preferred 315 K as well as to increase its pyromagnetic co-efficient. 3.3.4 ZnGdFe Nanoparticles Encapsulation with Protein (HSA) ZnGd-ferrite nanoparticles were encapsulated with a protein, e.g. human serum albumin (HSA). That was performed to ensure that they stay in the body for long time and are not eliminated by the reticoendothelial system. HSA is a single chain polypeptide of 585 residues, which comprises about 60% of the plasma protein. In human, albumin is synthesized in the liver and posses a half-life in circulation of 19-20 days, [149]. Briefly, 1ml aqueous solution of 250 mg HAS and 250 mg ZnGd-ferrite nanoparticles were added drop wise to a mixture of 40 ml-n-hexane, 10 ml light bmineral oil and 0.5 ml of sorbitan sesquioleate to form water in oil inverse emulsion system and it was sonicated for a period of 10 minutesat 50% amplitude. A 10 ml of 25% glutaraldehyde saturated with toluene was added to this mixture and mechanically stirred 66 with Teflon paddle stirrer at 200 rpm for ~ 15 minutes. The supernatant was decanted and the nanoparticles were washed repeatedly with petroleum ether and acetone and finally they were dispersed in distilled water. The process comprised the following steps: for preparation of glutaraldehyde saturated with toluene, 10 ml of 25% glutaraldehyde and 10 ml. of toluene were dispersed ultrasonically at 50% amplitude. The mixing was done in an ice-bath and in two five min steps. Afterwards the mixture was taken in a separating funnel and the milky white phase is separated. The phase separation process was repeated thrice by adding 10 ml of toluene for each repetition to ensure complete separation. For preparation of HSA - GdZn-ferrite particles in water, 250 mg. of HSA and about 75 mg. of GdZn-ferrite were dissolved in 1 ml. of distilled water and this mixture was added drop wise to a mixture of 40 ml. of n-hexane, 10 ml. of light mineral oil and 0.5 ml. of sorbitan sesquioleate. This inverse (water-in-oil) emulsion is sonicated for three times at 50% amplitude. The ultrasonication is carried out in an ice-water bath to prevent heat denaturation of albumin protein. For crosslinking of HSA micro/nano spheres with glutaraldehyde, the above dispersion is then mixed with 10 ml. of GST and the mixture is stirred at 1500 rpm for 2 hrs using a motor operated, teflon coated stirrer. The micro/nano spheres thus formed are centrifuged and washed with petroleum ether repeatedly followed by washing with PBS and stored in acetone. TEM images in chapter 4 show that the magnetic nanoparticles were well encapsulated within HSA using this method. Also, it was observed that each HSA shell encapsulated on an average one magnetic particle. The empty light shaded circles indicate the excess HSA compared to the amount of magnetic particles. 3.3.5 Synthesis of ZnNd Ferrite Magnetic Nanoparticles MnNd-ferrite nanoparticles with a narrow size distribution were prepared using a two-step process: the precipitation of the corresponding hydroxides, followed by the oxidation of the Fe2+. When these materials are prepared in the form of nanoparticles they are superparamagnetic, and can be used for the preparation of magnetic fluids for hyperthermia purposes. 67 ZnNd-spinel ferrite nanoparticles were prepared using the coprecipitation method. The starting materials for ZnNdFe were: Ferrous sulfate hepta-hydrate (FeSO4), manganese chloride hepta-hydrate NdCl2.4H2O, and ZnSO4.7H2O were used to obtain Zn2+, Nd2+ and Fe3+ ions in aqueous solutions. The solutions containing these ions were dissolved and mixed in distilled water with concentrations of 0.8 and 0.4, respectively. The aqueous mixture was added into 8 M NaOH solution to form a precipitate. The slurry was then placed in a boiling water bath and digested for one hour. Later, the slurry was filtered and washed until the solution pH became neutral. The resultant powder was heated in air in a tube furnace at various temperatures. ZnNd-ferrite fluids were prepared by distributing nanoparticles with a particle size of the order of ~ 10 nm, and with a narrow size distribution, in a carrier liquid. 3.3.6 Polyvinyl Alcohol Encapsulated ZnNdFe Nanoparticles ZnNd-ferrite nanoparticles were used for encapsulation within Polyvinyl Alcohol (PVA). Poly-vinyl alcohol is linear flexible molecule with no charge. Therefore, it adsorbs non-specifically on the surface of nanopoarticles. The interaction with the surface takes place through hydrogen bonds between polar functional groups of the polymer chain and hydroxylated and protonated groups on the surface. Though the interaction energy between surface and each chain segment is smaller than kT, chains adsorb very well because of large number of contact points. The affinity of the macromolecule for the surface usually increases with its molecular weight. The conformation of the adsorbed polymer remains similar to that of free macromolecule and exhibit tails and loops between contact points. The adsorbed layer provides an excellent steric protection against aggregation, [150,151]. This can be schematically depicted as below. 68 Figure 3-3. Schematic of nanoparticles encapsulated in a polymer. In this method, 250 mg of ZnNd-ferrite magnetic nanoparticles fabricated by co- prcipitation method was suspended in 40 ml distilled water. 2 gm PVA (Molecular Weight : 125,000) (purchased from Polysciences Inc.) was then added to this suspension. The PVA is water soluble henceforth, dissolves with ease. The solution was then sonicated using an ultrasonicator at 50% amplitude for 5 minuts in an ice bath. After sonication the resulting solution was washed repeatedly with acetone to remove the excess polymer content. The encapsulated nanoparticles were stored in acetone. The morphology of these PVA encapsulated ZnNd-ferrite particles, is shown in chapter 4. The dark shaded regions -ZnNd-ferrite- within the lightly shaded circles -PVA spheres- indicate that the ZnNd-ferrite nanoparticles were well encapsulated within the PVA shell. The large quantity of empty circles are due to the excess amount of Polyvinyl Alcohol as compared to the ZnNd-ferrite particles. 3.3.7 Synthesis of GdZnCe Ferrite Magnetic Nanoparticles Gd-Zn Ferrite has recently gained a lot of interest mainly because of the ability to change its properties by varying the proportions of its constituents. In this study, we have synthesized GdZnCe-ferrite nanoparticles using chemical co-precipitation method in which the proportions of Zinc and Cerium were varied and the effects on the properties were studied. The Curie temperatures of their particles were ~ 315 K. In this study several GdZnCe-ferrite particles with various constituent proportions were synthesized to check whether by adding Ce, it is possible to bring down the Curie temperature to 315 K. 69 Several samples of the form GdZnCeFe were synthesized using chemical co- precipitation method. In this method a 0.1 M solution of the metal salts GdCl2, Fe2SO4, CeCl3 and ZnSO4 was added to an 8 M solution of NaOH. The mixture was stirred vigorously at 90ºC for 40 minutes. Thereafter the synthesized magnetic nanoparticles were filtered and washed 3 times with distilled water and 3 times with acetone. The particles were then allowed to dry in air at room temperature. 3.3.8 Ethyle cellulose Encapsulated GdZnCeFe Using Polymer Emulsion The Polymer emulsion method described earlier was used to encapsulate GdZnCe-ferrite nanoparticles inside ethyl cellulose which has a glass transition temperature of 42°C. The following ingredients were used: Polymer Phase: Polymer: Ethyl Cellulose: 2gm. Solvent: Methylene Chloride: 10 ml Aqueous medium phase: Water: 40 ml Emulsifying agent: Sodium dodecyl sulphate: 0.2 gm. Inhibitor compound: 1-Octanol: 0.5 ml, 0.4 gm Magnetic particles: GdZnCe-ferrite: 50 mg. The magnetic particle: polymer ratio was approximately 1:40. The sodium dodecyl sulphate and 1-octanol were dissolved in 40 ml of distilled water using a magnetic stirrer. Then 50 mg of GdZnCe-ferrite nanoparticles was added. The polymer phase was prepared by dissolving 2 gm of ethyl cellulose into 10 ml of methylene chloride. A crude emulsion was formed by adding the polymer phase to the aqueous medium phase. The crude emulsion was sonicated using an ultrasonicator 5 times in steps of 3 minutes. The resultant emulsion was then stirred inside a round bottom flask for 12 hours at 700 rpm. The solvent was then removed using vacuum evaporation method. The polymer encapsulated particles thus formed were washed with acetone and stored under PBS buffer solution. 70 3.4 Size Dstermination and Morphology 3.4.1 X-Ray Diffractometer (XRD) X-ray powder diffraction is a method used to determine the crystal structure and analyze the phase of a particular material. Diffraction occurs as waves interact with a regular structure whose repeat distance is about the same as the wavelength. The phenomenon is common in the natural world, and occurs across a broad range of scales. X-rays have wavelengths on the order of a few angstroms, the same as typical interatomic distances in crystalline solids, which means X-rays can be diffracted from minerals that, by definition, are crystalline and have regularly repeating atomic structures. When certain geometric requirements are met, X-rays scattered from a crystalline solid can constructively interfere, producing a diffracted beam. A diffraction pattern appears, and this diffraction pattern can be analyzed to determine various structural properties of a material in question. W. L. Bragg recognized a predictable relationship among several factors: a) The distance between similar atomic planes in a mineral - the interatomic spacing- called the d-spacing and measured in angstroms. b) The angle of diffraction called the theta angle and measured in degrees. For geometrical reasons the diffractometer measures an angle twice that of the theta angle, 2θ. 3) The wavelength of the incident X-radiation, symbolized by λ and measured in angstroms (equals 1.54 Å for copper which is commonly used). These relationships are expressed in an equation defining Bragg’s law, and can be written as follows: nλ = 2d (sinθ) (1) 71 Figure 3-4. X-Ray Diffractometer at MARTECH. X-ray diffraction is used to determine the identity of crystalline solids based on their atomic structure. During x-ray diffraction analysis, x-ray beams are reflected off the parallel atomic layers within the nanoparticle material over a range of diffraction angles. Because the x-ray beam has a specific wavelength, for any given 'd-spacing' - distance between adjacent atomic planes- there are only specific angles at which the exiting rays will be 'in phase' and therefore, will be picked up by the detector producing a peak on the diffractogram. Just like a fingerprint, every material has its own distinct set of diffraction peaks that can be used to identify it. MARTECH diffractometer system uses CuKα radiation that has a wavelength of 1.54Å. Analyses are commonly run using a 40kV 45mA x-ray tube voltage, a 0.04° soller slit, 1° divergence and antiscatter slits, and a 1/2° (for powder) or 1/4° (for clays) receiving slit. The mechanical assembly that makes up the sample holder, detector arm and associated gearing is referred to as goniometer. A typical diffraction spectrum consists of a plot of reflected intensities versus the detector angle 2θ ,[152]. Most of all solid materials can be described as crystalline. When X-rays interact with a crystalline substance -Phase- one gets a diffraction pattern. Every crystalline substance gives a pattern; the same substance always gives the same pattern; and in a 72 mixture of substances each produces its pattern independently of the others. The X-ray diffraction pattern of a pure substance is, therefore, like a fingerprint of the substance. The powder diffraction method is thus ideally suited for characterization and identification of polycrystalline phases. The main use of powder diffraction is to identify components in a sample by a search/match procedure. Furthermore, the areas under the peak are related to the amount of each phase present in the sample, [152]. Solid matter can be described as: amorphous, whereby atoms are arranged in a random way similar to disorder found in a liquid, e.g. Glasses, and crystalline wherby atoms are arranged in a regular pattern, and there is as smallest volume element that by repetition in three dimensions describes the crystal. This smallest volume element is called a unit cell. Dimensions of a unit cell are usually described by three axes: a, b, c and the angles between them: α, β, γ. When an X-ray beam hits an atom the electrons around the atom start to oscillate with the same frequency as the incoming beam. Destructive interference will occur in almost all directions i.e., the combining waves are out of phase and there is no resultant energy leaving the solid sample. However the atoms in a crystal are arranged in a regular pattern, and in a very few directions constructive interference occurs. Waves are in phase and there will be well defined X-ray beams leaving the sample at various directions. Hence, a diffracted beam may be described as a beam composed of a large number of scattered rays mutually reinforcing one another. The X-rays are reflected from a series of parallel planes inside the crystal. The orientation and interplanar spacings of these planes are defined by the three integers h, k, and l called indices. A given set of planes with indices h, k, and l cut the a-axis of the unit cell in h sections, the b axis in k sections and the c axis in l sections. A zero indicates that the planes are parallel to the corresponding axis, e.g. the 2, 2, 0 planes cut a- and b-axes in half, but are parallel to the c-axis. 73 3.4.2 Transmission Electron Microscope (TEM) Transmission electron microscopy -TEM- is a procedure that is used to characterize the morphology of materials such as nanoparticles. These instruments are used because of the limited image resolution in light microscopes imposed by the wavelength of visible light. Electrons have wave-like characteristics with a wavelength substantially less than visible light. Since electrons are smaller than atoms TEMs are capable of resolving atomic level detail. Samples are prepared for TEM imaging by inserting a TEM grid - copper coated with formvar - into dry or wet powder - usually dried overnight using tweezers to hold the grid. The sample grid is then lightly tapped to remove any excess particles and the grid is placed in the TEM for imaging. This procedure can be used to characterize the coated and uncoated magnetic particles. Transmission electron microscopy is the premier tool for understanding the internal microstructure of materials at the nanometer level. Although x-ray diffraction techniques generally provide better quantitative information than electron diffraction techniques, electrons have an important advantage over x rays in that they can be focused using electromagnetic lenses. This allows one to obtain real-space images of materials with resolutions on the order of a few tenths to a few nanometers depending on the imaging conditions, and simultaneously obtain diffraction information from specific regions in the images, e.g., small precipitates as small as 1 nm. Variations in the intensity of electron scattering across a thin specimen can be used to image strain fields, defects such as dislocations and second-phase particles, and even atomic columns in materials under certain imaging conditions. Transmission electron microscopy is such a powerful tool for the characterization of materials that some microstructural features are defined in terms of their visibility in TEM images. In addition to diffraction and imaging the high-energy electrons - usually in the range of 100 to 400 keV of kinetic energy - in TEM cause electronic excitations of the atoms in the specimen. Two important spectroscopic techniques make use of these excitations by incorporating suitable detectors into the transmission electron microscope, 74 energy-dispersive x-ray spectroscopy -EDS, and electron energy loss spectroscopy - EELS. A fully equipped transmission electron microscope has the capability to record the variations in image intensity across the specimen using mass thickness or diffraction contrast techniques to reveal the atomic structure of materials using high-resolution - phase-contrast- imaging or Z-contrast –incoherent- imaging to obtain electron diffraction patterns from small areas of the specimen using a selected-area aperture or a focused electron probe, and to perform EELS and EDS measurements with a small probe. Additional lenses can be installed in conjunction with an EELS spectrometer to create an energy filter; enabling one to form energy-filtered TEM images. These images enable mapping of the chemical composition of a specimen with nanometer spatial resolution. In addition to the main techniques of (1) conventional imaging, (2) phase-contrast imaging, (3) Z-contrast imaging, (4) electron diffraction, (5) EDS, and (6) EELS, in TEM many other analyses are possible. A JEOL-2010 TEM located at NHMFL was used in this study to observe the morphology and structure of the nanoparticles synthesized. A more technical explanation of a typical TEMs working is as follows, Figure 3-5: Figure 3-5. TEM similar to the one at NHMFL. 75 A light source at the top of the microscope emits the electrons that travel through vacuum in the column of the microscope. Instead of glass lenses focusing the light in the light microscope the TEM uses electromagnetic lenses to focus the electrons into a very thin beam. The electron beam then travels through the specimen under study. Depending on the density of the material present some of the electrons are scattered and disappear from the beam. At the bottom of the microscope the unscattered electrons hit a fluorescent screen; which gives rise to a shadow image of the specimen with its different parts displayed in varied darkness according to their density. The image can be studied directly by the operator or photographed with a camera. The darker areas of the image represent those areas of the sample that fewer electrons were transmitted through; they are thicker or denser. The lighter areas of the image represent those areas of the sample that more electrons were transmitted through ; they are thinner or less dense, [153]. For the observation of the nanoparticles a dilute solution of the particles in acetone was prepared. Two drops of this solution were then dropped on the copper grids which acted as the specimen in TEM. The JEOL JEM-100C Transmission Electron Microscope was used for analysis, figure 3-6 . This 100 kV microscope has a resolution of 0.4 nm using a tungsten filament. The maximum magnification is 800,000x. Several single and double tilt specimen stages can be used in the operation. This microscope is easy to operate and is used for Initial examination of TEM samples, standard diffraction contrast imaging and standard electron diffraction techniques. 76 Figure 3-6. JEOL JEM-100C Transmission Electron Microscope at NHMFL. 3.4.3 Scanning Electron Microscope Scanning electron microscopy -SEM- is used for the imaging of bulk, thin film, patterned and nanoparticle systems. Energy analysis of the secondary X-rays produced by the electron beam can be used to characterize the elemental composition of the regions imaged. Conventional light microscopes use a series of glass lenses to bend light waves and create a magnified image. The Scanning Electron Microscope creates the magnified images by using electrons instead of light waves. The SEM shows very detailed 3-D images at much higher magnifications than is possible with a light microscope. The images created without light waves are rendered black and white. Samples have to be prepared carefully to withstand the vacuum inside the microscope. Biological specimens are dried in a special manner that prevents them from shriveling. Because the SEM illuminates them with electrons they also have to be made to conduct electricity. The sample is placed inside the microscope's vacuum column through an air-tight door. 77 Figure 3-7. Sample is placed insides microscope vacuum column through door. After the air is pumped out of the column, an electron gun -at the top- emits a beam of high energy electrons. This beam travels downward through a series of magnetic lenses designed to focus the electrons to a very fine spot. Near the bottom, a set of scanning coils moves the focused beam back and forth across the specimen row by row. Figure 3-8 . SEM system working principle, [154]. 78 As the electron beam hits each spot on the sample secondary electrons are knocked loose from its surface. A detector counts these electrons and sends the signals to an amplifier. The final image is built up from the number of electrons emitted from each spot on the sample. Figure 3-9 . JEOL JSM-840 scanning electron microscope at NHMFL. The JEOL JSM-840 scanning electron microscope at NHMFL was used. This scanning electron microscope operates at accelerating voltage from 200 V to 40 kV. Its magnification ranges from 10x to 300,000x, with a resolution of 4 nm. This instrument was used for conducting morphological characterization of the ultra fine microstructures of biological and metallic samples and performing standard SEM imaging analysis. 3.5 Nanopartcle Surface Characterization Zeta Potential Measurements The potential at the surface of shear for a particle is defined as the zeta potential. Derivations show that the zeta potential is the double-layer potential close to the particle surface and one of its applications is the measurement of surface charges of particle surfaces, such as nanoparticles. The liquid layer of a particle in suspension migrating in an electric field moves at the same velocity as the surface - shear surface. This shear 79 surface occurs well within the double layer, likely at a location roughly equivalent to the Stern surface. Although the precise location of the surface of shear is unknown, it is assumed to be within a couple of molecular diameters of the actual particle surface for smooth particles. This thickness is associated with the zeta potential and defines the ion atmosphere near a surface, [155]. The magnitude of the zeta potential gives an indication of the potential stability of a colloidal system. If all the particles have relatively large negative or positive zeta potentials they will repel each other and create dispersion stability. If the particles have low zeta potential values there is no force to prevent the particles from agglomerating and there is dispersion instability. A dividing line between stable and unstable aqueous dispersions is generally taken at either +30 or -30mV. Particles with zeta potentials more positive than +30mV are normally considered stable. Particles with zeta potentials more negative than -30mV are normally considered stable , Figure 3-10. Figure 3-10. Zeta potential. Taken from Silver Colloids (www.silver-colloids.com/Tutorials/Intro/pcs1.html). The BIC 90Plus/BI-MAS (Zeta Potential), located in our lab., was used for the magnetic nanoparticles surface characterization. The multi angle particle sizing option in 80 a BIC ZetaPals was used to measure the approximate diameter of the particles and their size distribution. The Mas Option is an automatic particle sizer designed for use with either concentrated suspensions of small particles or solutions of macromolecules. In general, sizes from 2nm-3 m may be measured using this device. The technique involved - photon correlation spectroscopy-PCS- of quasi elastically scattered light -QELS- is based on correlating the fluctuations about the average, scattered, laser light intensity. The specifications for this machine were as follows: Size Range : 2nm to 3 m Accuracy : ±1% to 2% with monodisperse samples Repeatability : ± 1% to 2% with dust free samples Sample Volume : 0.5 to 3ml Measurement Time : Typically 1 to 2 mins Results : Mean and Standard Deviation calculated for size distribution by weight assuming a Lognormal distribution. Light may be treated as an electromagnetic wave. The oscillating electron- magnetic field induces oscillations of the electrons in a particle. These oscillating changes form the source of the scattered light. Over the years many features of the scattered light have been used to determine particle size. These include: changes in the average intensity as a function of angle, changes in the polarization, changes in the wavelength and fluctuations about the average intensity. This later phenomenon is the basis for QELS, the technique employed in the Mas Option, and it arises in the following manner: imagine a detector of light fixed as some angle with respect to the direction of the incident light beam and at some fixed distance from the scattering volume which contains a large number of particles. Scattered light from each particle reaches the detector. Since the small particles are moving randomly in the liquid, undergoing diffusive Brownian motion, the distance that the scattered waves travel to the detector varies as a function of time. Electromagnetic waves, like water and sound waves, exhibit interference effects. Scattered waves can interfere constructively or destructively depending on the distances traveled to the detector. The result is an average 81 intensity with superimposed fluctuations. The decay times of the fluctuations are related to the diffusion constants and, therefore, the sizes of the particles. Small particles moving rapidly cause faster decaying fluctuations than larger particles moving slowly. The decay times of these fluctuations may be determined either in the frequency domain - using a spectrum analyzer - or in the time domain -using a correlator. The correlator generally offers the most efficient means for this type of measurement. In QELS the total time over which a measurement is made is divided into small time intervals called delay times. These intervals are selected to be small compared with the time it takes for a typical fluctuation to relax back to the average. The scattered light intensity in each of these intervals, as represented by the number of electrical pulses registered during each delay time, fluctuates about a mean value. The intensity autocorrelation function is formed by averaging the products of the intensities in these small time intervals as a function of the time between the intervals -delay times. A computer automatically controls the buildup of the function including the choice of delay times, the length of the experiment, and display of pertinent information, data analysis, and the printing of results. 3.6 Magnetic Heating Equipment and Instruments Multiple studies [156-158], have used inductor setups and the inductor allowed for a first approximation of heating in ferromagnetic materials. The use of an inductor in the heating apparatus setup used in this study, was similar to the testing measurements done by Jordan, [156]. The appealing qualities in using this equipment were: efficient power transfer to the ferromagnetic load by the capacitor and inductor in a resonant circuit, selective heating in the ferromagnetic materials over surrounding media, and the ability to generate repeatable results. This very simple method creates H-fields via induction from the induced E-fields. These H-fields are induced with the intention of generating hysteresis losses in synthesized ferromagnetic materials. A disadvantage in this type of heating of samples may be in part due to Ohmic losses generated in the wires of the solenoid. Moreover, the 82 solenoid will in theory cause smaller heat generation in contrast to a capacitive setup due to the smaller E-fields that are induced in the coil setup. A capacitor setup was developed but it failed to produce a response in thermocouples, where one would expect one due to radiofrequency interference. The probable cause of this was the method of implementation: two parallel copper plates acted as capacitor electrodes and this capacitor was in parallel with a 50 load. Since the equipment impedance is 50 W, the copper plates acted as open ends. The specific instruments and equipment used for magnetic heating experiments are the following: (I) Coil apparatus. Coil length of 100 mm, diameter of 53 mm, and 68 turns, with ingle- layer inductor that allows for calculation of magnetic field strength, H, due to the relatively large length, L, of the coil, according to: H = nI / l, where n is the number of wire turns in the coil and I is the current in the coil; thus, for an output of 20 W, we can obtain H as power equals voltage times current; coil was wrapped around a ¼” Plexiglas tube and ¼” foam insulation to prevent sample heating by IR radiation and conduction; capacitor was in parallel with the inductor to allow for a resonance frequency and another capacitor was used in series to allow for different values of capacitance and impedance. The resonant frequency, f, relationship is given by: f = 1 / √LC. The resonance frequency was 961 KHz. Relationships of reactance - or impedance - for inductors and capacitors give the theoretical impedance for the circuit - to match the impedance of the input amplifier/signal generator. The 50 impedance was matched and verified for amplifier to coil in all experiment. (II) LG FG-7002C Sweep/Function Generator. Capable of 1MHz in frequency oscillation. (III) ENI AP-400B Controllable Power Amplifier. Capable of supplying a maximum of 500 watts to the load - power was limited to a maximum 40W in all experiments. (IV) Oscilloscope and Probe. Tektronix TDS 3012 & P3010. (V) Omega HH3506R thermocouple/LCD. Measure for 0.1°C difference at all temperatures between two thermocouple probes inserted into the two open slots. (VI) Two copper constantan thermocouples. Used to record temperature - unsheathed at the tips. 83 This setup was shown to have a slim bandpass of resonance with a distinct peak near 961 kHz. Thus, it appeared to be efficient and initial tests showed that the forward power in the amplifier could match the load power well - nearly equal- at 961 kHz. A picture of the coil setup is shown in Figure 3-11. Notice the placement of thermocouple during testing. Temperature gradients that might exist within the apparatus should also be considered as coil is more likely to heat due to Ohmic losses caused by alternating current. When power is supplied temperature should be larger in coil than in insulation at any instance. Since the insulation will absorb heat generated by the coil insulation will have a higher temperature closer to the coil unless the sample generates large amounts of heat very quickly. Due to the air-space between the sample and the insulation, the insulation temperature will likely reach a minimum closer to the sample than to coil. In addition, gradients are unlikely to be linear especially considering inhomogeneous heating within the sample. Figure 3-11. Picture of coil setup and apparatus used for nanoparticles magnetic heating. 84 As mentioned earlier, the apparatus shown above is similar to the coil devised by Jordan’s. A small difference exists in that, the other setups prevented sample heating from the coil by running water through the coils while in this experiment only the foam and Plexiglas prevent heat from transferring from coil to sample. Insulation also served to maintain heat within the sample to prevent excessive air cooling of the sample. This apparatus creates a homogeneous alternating magnetic field within the coil which in turn create E-fields in space perpendicular to the magnetic field - null E-field at the center of the coil. The equation for the B-field, B, has two parts derived from Faraday’s law of magnetic induction: dΦ B −= ∫ EdS (2) dt S where dΦB/dt is the change in magnetic flux with respect to time and S is the path of the E-field - closed in this case. The flux can be expressed in terms of the cross sectional area times the changing magnetic induction. Therefore, as the E-field within the inductor is one of concentric loops centered along the long axis of the coil the B-field inside the coil is: dB 2 −= E (3) dt r For the B-field outside the coil, the following expression applies: dB 2r −= E (4) dt R 2 where R is the diameter of the coil, r is the radial position of the sample, and dB/dt is the change in magnetic induction with respect to time; therein, a linear increase in E-field strength occurs inside the coil and a decrease in it occurs with the relationship of E ~ 1/r outside of the coil, with the maximum occurring at the coil itself. Note that experiments listed in chapter 4 used the materials and equipment here in a variety of methods to establish a clear picture of the temperature rise effect of each sample. The following schematic details the elements the resonant circuit: 85 Cs= 100pf Cp= 300pf The Vertical Coil No. of turns = 60 L = 68 UH IZI = 48 Frequency = 961 KHz Figure 3-12 . Circuit diagram for the resonance circuit. 3.7 DC Superconducting QUantum Interference Device (SQUID), Magnetometer The DC SQUID Magnetometer -at MARTECH; shown in figure 3-13- for characterizing the magnetic properties of bulk, thin film, spin-glass, and super- paramagnetic nanoparticle systems was used. Using this device, ± 5T DC fields can be applied to samples from 5K to 400K, with a sensitivity of 1x10-7 EMU in the detected moment. A SQUID is the most sensitive device available for measuring magnetic fields and although the SQUID in the MPMS is the source of the instrument’s remarkable sensitivity it does not detect directly the magnetic field from the sample. Instead, the sample moves through a system of superconducting detection coils which are connected to the SQUID with superconducting wires allowing the current from the detection coils to inductively couple to the SQUID sensor. When properly configured, the SQUID electronics produces an output voltage which is strictly proportional to the current flowing in the SQUID input coil. Hence the thin film SQUID devic, which is located 86 approximately 11 cm below the magnet inside a superconducting shield, essentially functions as an extremely sensitive current-to-voltage converter. 1.Sample Rod 8.SQUID Capsule with Magnetic Shield 15.Console Cabinet 2.Sample Rotater 9.Superconducting Pick-up Coil 16.Power Distribution Unit 3.Sample Transport 10.Dewar Isolation Cabinet 17.Model 1822 MPMS Controller 4.Probe Assembly 11.Dewar 18.Gas/Magnet Control Unit 5.Helium Lever Sensor 12.HPThinkjet Printer 19.HP Vectra Computer 6.Superconducting Solonoid 13.Magnet Power Supply 20.Monitor 7.Flow Impedance 14.Model I802 Temperature Controller Figure 3-13. Featuring quantum design’s Magnetic Property Measurement System (MPMS); SYSTEM COMPONENTS (Superconducting components shown in blue) 87 A measurement is performed in the MPMS by moving a sample through the superconducting detection coils which are located outside the sample chamber and at the center of the magnet. As the sample moves through the coils the magnetic moment of the sample induces an electric current in the detection coils. Because the detection coils, the connecting wires and the SQUID input coil form a closed superconducting loop, any change of magnetic flux in the detection coils produces a change in the persistent current in the detection circuit which is proportional to the change in magnetic flux. Since the SQUID functions as a highly linear current-to-voltage convertor, the variations in the current in the detection coils produce corresponding variations in the SQUID output voltage which are proportional to magnetic moment of the sample. In a fully calibrated system, measurements of the voltage variations from the SQUID detector as a sample is moved through the detection coils provide a highly accurate measurement of the sample’s magnetic moment. The system can be accurately calibrated using a small piece of material having a known mass and magnetic susceptibility. The sample space is made from a tube with a 9 mm inside diameter, and is maintained at a low pressure with static helium gas. At the top of the sample space is an airlock that can be evacuated and purged with clean helium gas - boil off from the liquid helium bath in the dewar. When the purge airlock button is depressed, this evacuate/purge cycle is repeated several times. The final step of the cycling sequence brings the airlock to the low pressure maintained in the sample space. The airlock and the sample space are separated by a ball valve which after opening makes the airlock a continuous part of the sample space. The lower portion - ~ 30 cm- of the sample space is lined with copper to provide a region of high thermal uniformity. Two thermometrs determine the sample temperature and provide for temperature control. An extensive calibration procedure in which a standard thermometer is placed in the sample position is used to determine temperature controller constants, temperature gradients, and thermometer calibration. The measurement procedure. The sample is mounted in a sample holder that is attached to the end of a rigid sample rod. The sample rod enters the sample space through a special type of double seal -lip seal- designed to allow the rod to be actuated by a drive mechanism located outside the chamber. The component containing the lip seals is 88 clamped onto the top of airlock with standard O-ring seals, forming the top of the sample space. The top of the sample transport rod is attached to a stepper-motor-controlled platform which is used to drive the sample through the detection coil in a series of discrete steps. A change in the sample’s position causes a change in the flux within the detection coil, thereby changing the current in the superconducting circuit. Since the loop is entirely superconducting the current does not decay as it would in normal conductor. During the measurement the sample is stopped at a number of positions over the specified scan length, and at each stop, several readings of the SQUID voltage are collected and averaged. The complete scans can be repeated a number of times and the signals averaged to improve the signal-to-noise ratio. The currents induced in the detection coil are ideally those associated with the movement of a point-source magnetic dipole through a second order gradiometer detection coil. The spatial - position - dependence of the ideal signal is shown in figure 3- 14. Figure 3-14. The output of the SQUID as a magnetic dipole is moved through second-order gradiometer pick up coil. The vertical scale corresponds to an output voltage and the horizontal scale is sample position. 89 To observe this signal, requires that the sample be much smaller than the detection coil and the sample must be uniformly magnetized. Uniform magnetization, however, is often not encountered with high critical-current-density (Jc) superconductors. If a sample is very long- extending well beyond the coil during a scan- its motion in the gradiometer will not be observable since there would be no net charge of the flux in the detection coil. This is why a long uniform tube can be used as a sample holder. In contrast to this, when the sample is short, the current in the detection coil changes with sample position. This is because different amounts of flux - the local induction B - exist in each loop of the detection coil. So, it is important to realize that there is a limit on the length of a sample for which accurate measurements can be made. Some accommodation for length is made, in particular, computer fitting routines used to extract the value of the moment from the SQUID output. However, the safest procedure is to calibrate the MPMS with standards having a size and shape similar to the samples to be measured. A DC SQUID is capable of detecting magnetic fields of around 2 pT. It has however, been demonstrated that fields of around 100 fT are also within the scope of a SQUID. It is typically required to be kept at temperatures of ~ 4.2 Kelvin, [160]. In this study, the SQUID was used to determine the Curie temperature of the particles from the temperature dependence of magnetization plots. It is a plot of the Long Moment -EMU- vs. the temperature -K. In a typical plot of the data obtained from the SQUID, like the one shown in figure 3-15, the point where the extrapolated linear portion of the data meets the X-axis is the Curie temperature for the particle. Figure 3-15. A typical plot of temperature dependence of magnetization obtained using SQUID. 90 This chapter described various instruments and methods used for characterization of the magnetic nanoparticles synthesized in this study. These involved instruments for determining the magnetic properties such as Curie temperature and magnetic saturation, - SQUID, as well as those for determining the approximate chemical contents of the particles, -XRD- and morphological characterization, TEM. Also, the ZetaPals BI-MAS OPTION was used to get an estimate of the average particle size and size distribution. The instruments, XRD and SQUID used in this study were located at MARTECH, FSU whereas the TEM was located at NHMFL, Tallahassee, Florida State University. 91 CHAPTER 4 RESULTS AND OBSERVATIONS In the review of chapter 3, it was shown that physical properties of MnZn-ferrite and other as-synthesized nanoparticles may depend on their size. Prepared nanoparticles demonstrate a certain polydispersity. It seemed interesting to compare properties of small and large particles fractionated from the same polydisperse sample. TEM samples were prepared by placing a drop of the suspensions on a copper grid with a carbon membrane film. Images were taken using a JEOL-2010 TEM operating up to 200 kV. The crystalline structure of the samples was identified from X- ray diffraction -XRD- patterns recorded in the 2θ range 10–90° with a scan step of 0.05° (2θ) for 5 s on a Philips X’pert Pro diffractometer - Cu Kα radiation. The crystallite size was determined from the different diffraction peaks (hkl), using the Scherrer method. 4.1 Powder XRD Pattern and TEM Magnetic Particles Characterization MnZn-Ferrite Particles In the case of MnZn-ferrite magnetic nanoparticles, the XRD pattern is shown in figure 4-1(a) and (b). Determination of phase purity and their identification were done by X-ray diffraction studies using Phillips powder diffractometer with CuKα radiation. The pattern matches the one expected for the spinel phase characteristic of some ferrites, for example, Cobalt ferrite. The analysis confirms the formation of single-phase spinel structure. The average crystallite size was estimated from the XRD line broadening measurement by using the Scherrer formula: D = 9.0 λ / β cosθ (1) 92 where D is the average crystalline size, λ the X-ray wavelength used, β the angular line width of half maximum intensity and θ the Bragg’s angle in degrees. The analysis of the (3 1 1) peak gave a mean diameter of 5.5 nm. Average lattice parameter, a, was 8.415 A0. [2MNZF.MDI] 2MnZnFe 061907 (311)(311) 01-089-7556> Franklinite - (Zn,Mn,Fe)(Mn,Fe)O4 01-076-1470> Fe 21.333O32 - Iron(III) oxide- ≠gamma 125 100 75 (220) Intensity(Counts) (220) (440)(440) 50 (511) (511) (115) (400) 25 (400)(004) (111) (422)(422) (201) (731) (110) (211) (222) (533)(533) (111) (301) (222) (620) (642) (221) (412) (521)(215) (620)(206)(622) (444)(444) (642)(426) (200) (320)(321) (223)(411)(331)(331)(420)(204)(332) (403)(413) (502) (522)(433)(531)(531)(442)(315)(600)(610)(106)(611)(325) (434)(216)(514)(622)(226)(542)(631)(316) (632)(326)(551)(701)(107)(711)(515)(117)(640)(641)(416)(552)(525) (544)(227)(730)(307) 0 10 20 30 40 50 60 70 80 90 Two-Theta (deg) Figure 4-1 (a) & (b). Powder X-ray diffraction pattern of and reference pattern for MnZnFe2O4. Evaluation of the particle size by using the Scherrer formula gave an average diameter of 5.5 nm. The pattern is representative for all the samples of the series. 93 [2MNZF.MDI] 2MnZnFe 061907 125 100 75 Intensity(Counts) 50 25 0 01-089-7556> Franklinite - (Zn,Mn,Fe)(Mn,Fe)O4 01-076-1470> Fe 21.333O32 - Iron(III) oxide- ≠gamma 10 20 30 40 50 60 70 80 90 Two-Theta (deg) Figure 4-1. continued. 94 In figure 4-2, a typical TEM micrograph of the dispersion of MnZn-ferrite nanoparticles and MnZn-ferrite PEG encapsulated nanoparticles in ethanol deposited over a Cu grid, are shown. The sample consists of a dispersion of almost spherical particles with a narrow size distribution. The histogram of the diameter obtained from a statistical analysis over 100 particles is reported in the inset. It can be fit by a Gaussian distribution with mean diameter 6.2 ± 1.0 nm. The obtained value is close to that obtained from XRD measurements indicating a high crystalline degree of all the particles. A similar size distribution was observed for a solid dispersion in PEG. However, images taken at lower magnification revealed the presence of large aggregates containing hundreds of particles. a 20 nm Figure 4-2. TEM micrograph for: (a) MnZn-ferrite nanopaticles and (b) the PEG polymer encapsulated MnZn-ferrite particles formed by emulsion. 95 b Figure 4-2-continued. Figure 4-2 shows TEM micrographs of MnZn-ferrite sample (a): the particles were dispersed in ethanol; on (b) the particle were embedded in a solid matrix of PEG. The two images were taken at different magnifications. Figure 4-3. SEM images for two different resolutions of PEG encapsulated manganese zinc ferrite nanoparticles. Clustering of the nanoparticles can be seen. 96 Figure 4-3. Continued. PEG films doped with MnZn-ferrite nanoparticles were characterized by a JEOL JSM-840 scanning electron microscope -SEM at NHMFL. In figure 4-3, SEM images are shown at two different resolutions for the same sample. It can be observed that the nanoparticles agglomerate to form large clusters with varying sizes. Also, as is evident from the SEM images shown in figure 4-3 while being put in the PEG matrix these particles tend to form clusters of various size ranges. The clustering itself is due to steric forces in the polymer overcoming the natural tendency of the surfactant coated particles to stay away from each other. These mechanisms contribute to the increase in the blocking temperature and broadening of the peak in susceptibility. A similar structural and morphological characterization was performed on the other samples as-synthesized with the co-precipitation approach. The main structural features obtained are depicted in their respective XRD and TEM images. The histogram of the diameter obtained from a statistical analysis over 100 particles is reported in the inset. The main structural features obtained are listed in Tables in the subsequent sections. The data show that the co-precipitation synthetic method is effective for producing high crystalline ferrite nanoparticles with increasing sizes in the 5–25 nm 97 range. A large increase of the hydrodynamic diameter may result. Such increase can be addressed to dipolar magnetic forces between the particles rather than to a real increase of the organic/inorganic shell surrounding the particles. Indeed, dipolar interaction which increases with the particle magnetic moment and thus with the size strongly affects the Brownian motion. On the other hand, the long time stability of the suspension allows excluding the presence of large particle aggregates as those observed in the solid dispersion. ZnGd-Ferrite Particles In the case of ZnGd-ferrite magnetic nanoparticles, the XRD pattern is shown in figure 4-4 (a) and (b), in the next pages. TEM micrograph of the dispersion of ZnGd-ferrite nanoparticles and ZnGd-ferrite HSA encapsulated nanoparticles in ethanol deposited over a Cu grid are shown in figure 4-5. The sample over here also, consists of a dispersion of almost spherical particles with a narrow size distribution. Table 4-1, lists the main structural features of the as- synthesized GdZn-Ferrite nanoparticles. Table 4-1. Main structural features of the prepared GdZn-Ferrite nanoparticles Mean radius Mean radius Crystalline Surface chemistry Sample DXRD (nm) DTEM (nm) structure 1 5 ± 0.5 3 ± 0.5 crystalline hydrophilic 2 5 ± 0.1 4 ± 0.6 crystalline hydrophilic 3 5 ± 0.9 2 ± 0.2 crystalline hydrophilic 4 6 ± 0.5 4 ± 0.6 crystalline hydrophilic Where DXRD is the average particle diameter evaluated by X-ray diffraction using the Scherrer formula, DTEM is the average particle diameter evaluated by TEM. 98 [4ZGF.MDI] 4ZnGdFe 061907 (111) (311)(311) 03-065-8099> Gd - Gadolinium +2 +3 00-019-0629> Magnetite - Fe Fe2 O4 01-087-1231> (Zn0.333Fe0.666)(Fe1.223Zn0.612)O3.914 - Zinc Iron Oxide 500 400 300 (200) (440) Intensity(Counts) 200 (220) (311) (440) (220) (220) (511) (511) (400) 100 (400) (731) (422) (533) (331) (111) (222) (422) (222) (420) (731) (222) (533) (422) (400)(620) (622) (642) (111) (531) (620) (622) (444)(444) (642) (331) (531)(442) (711) 0 10 20 30 40 50 60 70 80 90 Two-Theta (deg) Figure 4-4 (a) &(b). Powder X-ray diffraction pattern of ZnGdFe2O4. The pattern is representative for all the samples of the series. 99 [4ZGF.MDI] 4ZnGdFe 061907 500 400 300 Intensity(Counts)200 100 0 03-065-8099> Gd - Gadolinium +2 +3 00-019-0629> Magnetite - Fe Fe2 O4 01-087-1231> (Zn0.333Fe0.666)(Fe1.223Zn0.612)O3.914 - Zinc Iron Oxide 10 20 30 40 50 60 70 80 90 Two-Theta (deg) Figure 4-4. Continued. 100 20 nm Figure 4-5. HSA encapsulated Gd-Zn-Ferrite under TEM at high magnification. GdZnCe-Ferrite Particles In the case of GdZnCe-ferrite magnetic nanoparticles, the XRD pattern is shown in figure 4-6. TEM micrograph of the dispersion of GdZnCe-ferrite nanoparticles and GdZnCe-ferrite Ethyl cellulose encapsulated nanoparticles in ethanol deposited over a Cu grid, are shown in figure 4-7. The sample over here also consists of a dispersion of almost spherical particles with a narrow size distribution. The histogram of the diameter obtained from a statistical analysis over 100 particles is reported below. Table 4-2 lists the main structural features of the prepared GdZnCe -Ferrite nanoparticles. Figure 4-7. Ethyl cellulose encapsulated GdZnCe-Ferrite by polymer emulsion. 101 [1CEGZF.MDI] 1CeGdZnFe 061907 (111) (311) (111) 00-033-0334> CeO - Cerium Oxide 00-025-1095> Fe 5Gd - Gadolinium Iron 00-022-1012> Franklinite - ZnFe2O4 +2 +3 00-019-0629> Magnetite - Fe Fe2 O4 100 (200) (101) 75 (220) (311) 50 Intensity(Counts) (110) (200) (220) (440) (511) (002) 25 (400) (202)(222) (301) (001) (422) (112)(211) (553) (533) (111) (220) (113) (100) (222) (642) (201) (300) (620)(103)(622) (302)(311) (331) (531) (444)(221)(551) 0 10 20 30 40 50 60 70 80 90 Two-Theta (deg) Figure 4-6 (a) &(b). Powder X-ray diffraction pattern of and reference pattern for GdZnCe- ferrite magnetic nanoparticles. The pattern is representative for all the samples of the series. 102 [1CEGZF.MDI] 1CeGdZnFe 061907 100 75 50 Intensity(Counts) 25 0 00-033-0334> CeO - Cerium Oxide 00-025-1095> Fe 5Gd - Gadolinium Iron 00-022-1012> Franklinite - ZnFe2O4 +2 +3 00-019-0629> Magnetite - Fe Fe2 O4 10 20 30 40 50 60 70 80 90 Two-Theta (deg) Figure 4-6. Continued. 103 Table 4-2. Characteristics of the GdZnCe-ferrite inorganic nanoparticles used in experimentation. Mean Mean Concentration Crystalline Surface radius radius per flask structure chmistry Sample DXRD DTEM (g/mL) (nm) (nm) 1 15 ± 0.4 11 ± 0.4 3.303 × 10-2 crystalline hydrophilic 2 13 ± 0.5 9 ± 0.5 2.180 × 10-3 crystalline hydrophilic 3 10 ± 0.2 6 ± 0.7 1.130 × 10-4 crystalline hydrophilic 4 8 ± 0.5 5 ± 0.2 3.201 × 10-2 crystalline hydrophilic ZnNd-Ferrite Particles In the case of ZnNd-ferrite magnetic nanoparticles, the XRD pattern is shown in figure 4-8. TEM micrograph of the dispersion of ZnNd-ferrite nanoparticles and ZnNd-ferrite polyvinyl alcohol encapsulated nanoparticles in ethanol deposited over a Cu grid are shown in figure 4-9. Table 4-3 lists the main structural features of the prepared ZnNd- Ferrite nanoparticles. 100 nm Figure 4-9. Polyvinyl alcohol encapsulated ZnNd-ferrite particles. 104 [3ZNF.MDI] 3ZnNdFe 061907 (311)(311) 01-087-1231> (Zn0.333Fe0.666)(Fe1.223Zn0.612)O3.914 - Zinc Iron Oxide 350 01-083-1356> Nd 2O3 - Neodymium oxide - HT 01-082-1533> Magnetite - Fe 3O4 300 250 200 150 Intensity(Counts) (440) (220) (440) (220) (110) (333) 100 (511) (400) (400) 50 (111) (731) (222) (211)(422)(422) (731) (222) (533)(533) (200) (111) (220) (310) (620)(620) (622)(622) (444)(444)(321) (642)(642) (331)(331) (531)(531)(442)(442) (222) (711)(711) (400) 0 10 20 30 40 50 60 70 80 90 Two-Theta (deg) Figure 4-8 (a) & (b). Powder X-ray diffraction pattern of and reference pattern for GdZnCe- ferrite magnetic nanoparticles. The pattern is representative for all the samples of the series. 105 [3ZNF.MDI] 3ZnNdFe 061907 350 300 250 200 150 Intensity(Counts) 100 50 0 01-087-1231> (Zn0.333Fe0.666)(Fe1.223Zn0.612)O3.914 - Zinc Iron Oxide 01-083-1356> Nd 2O3 - Neodymium oxide - HT 01-082-1533> Magnetite - Fe 3O4 10 20 30 40 50 60 70 80 90 Two-Theta (deg) Figure 4-8. Continued. 106 Table 4-3. Main structural features of the prepared ZnNd-Ferrite nanoparticles Crystalline Surface chmistry Sample D (nm) D (nm) XRD TEM structure 1 10 ± 0.3 8 ± 0.5 crystalline hydrophilic 2 15 ± 0.1 12 ± 0.6 crystalline hydrophilic 3 18 ± 0.9 14 ± 0.2 crystalline hydrophilic 4 16 ± 0.5 10 ± 0.6 crystalline hydrophilic 4.2 Properties of Particles after Co-precipitation and Heating In the following, and to avoid repetition, discussion will relate mostly to the MnZn-ferrite magnetic nanoparticles as a representative material for the various synthesized nanoparticles in this study. Other materials may be treated in a similar fashion. For MnZn-ferrite particles with x~0.5 that were collected after co-precipitation and heating via natural sedimentation process in mature alkaline solution, three samples were collected according to different times of sedimentation. Sample S3 sedimented for 5 minutes and has weight of 48 % of the total sample. Other two samples, S2 and S1, have weights of 41 and ~11 % of total sample, respectively. Properties of the samples such as chemical composition, mean size from X-ray, are summarized in Table 4-4. It must be mentioned that X-ray diffraction spectra reveal some additional peaks that may be attributed to the hydroxide of Iron. For some, only peaks of spinel are observed. Properties of MnZn-Ferrite Nanoparticles Liquid magnetic solutions at concentrations of 3 mol/l Me (Me - total metals concentration) are centrifuged at 10,000 rpm for 25 minutes. Magnetic solution, (x = 0.5) is separated into 3 samples taking magnetic solution from 3 parts of centrifugal tube - at a 107 different height. The following samples are obtained: S1, S2 and S3. The magnetic solution, (x = 0.2) after centrifugation, is divided into 2 samples: the S4 remaining as liquid and the S5 concentrated on the bottom of the tube as viscous sediment which is then separated and redispersed in distilled water. Table 4-4. Properties of particles after coprecipitation and heating in alkaline medium. Name of Sample's Chemical composition weight w% Zn Mean Name of of total substit- size of sample ution particles sample MnO ZnO Fe2O3 Molar mol.% mol. mol. ratio degree DXR, nm % % X x S1 10.7 23.4 25.9 50.5 0.328 0.525 10.8* S2 41.1 23.8 24.7 51.4 0.321 0.509 10.7 S3 48.2 24.6 23.3 52.1 0.315 0.486 9.5 *X-ray diffraction spectrum of S1 has an additional peaks of hydroxide of FeOOH. Properties of particles, i.e. chemical composition, mean size from X-ray diffraction and mean size from electron microscopy are determined for the different fractions. Data are summarized in Table 4-5. The mean size values determined from the X-ray diffraction exceed those from the electronic microscopy. Tourinho [161], reported a similar difference in size obtained by these two methods for particles of ferrite of Mn and Co. This difference may be connected with a remained polydispersity in size even in obtained fractions. For samples of S1-S3 the difference in chemical composition is insignificant whereas for samples of S4 and S5 a strong difference in metal content is found. In sample S4, content of Fe as well as the Zn substitution degree x are higher than those in sample S5. Comparison of chemical compositions of the co-precipitated precursor with composition of samples of magnetic solution reveals that bigger particles are less chemically changed during the surface treatment than the smaller ones. 108 Table 4-5. Properties of fractions of the surface treated particles separated by centrifugation for two samples: (x=0.5) and (x=0.2). Mean sizes, determined from X-ray diffraction (DXR) and electron microscopy (DEM), chemical composition and associated water content (calculated). For samples of S4 and S5 properties of precursor (particles before surface treatment) are also given. Particles Fraction part of total mean size Chemical composition (nm) Sample Sample D D XR TEM mol % (nm) (nm) Molar Zn subst. MnO:ZnO:Fe O of Me 2 3 ratio degree X x S1 26.5 10.65 8.7 18.6:22.3:59.0 0.257 0.545 S2 33.8 10.5 - 18.4:23:58.7 0.261 0.556 S3 39.7 9.5 8.5 18.5:23.1:58.4 0.263 0.555 S4 13.5 9.5 4.9 29.2:9.7:61.1 0.241 0.249 S5 86.5 12.7 11.5 35.8:11.0:53.2 0.305 0.235 Precursor - 12.0 - 39:10:51 0.325 0.20 For the other synthesized samples in this study, the following tables, table 4-6 and 4-7, summarize their properties: Table 4-6. Properties of ZnNd-ferrite magnetic nanoparticles. Sample's weight w% Mean size Mean size of particles Sample of total NdO ZnO Fe2O3 of particles sample mol. mol. mol. DXR DTEM % % % (nm) (nm) S1 12.3 25.4 27.9 46.5 18.4 16.5 S2 38.5 24.8 25.7 47.4 16.5 14.5 S3 39.2 25.6 26.3 49.1 11.6 10.2 109 Table 4-7. Properties of ZnGd-ferrite magnetic nanoparticles. Sample's Sample weight w% ZnO of total sample GdO Fe2O3 Mean size of particles mol. mol. mol. DXR DTEM % % % (nm) (nm) S1 12.3 27.6 21.2 53.1 20.7 18.5 S2 38.5 24.4 25.9 50.5 17.5 15.5 S3 39.2 24.0 25.7 51.4 21.0 19.2 4.3 Magnetic Properties of Particles and Magnetic Solution Magnetic Properties of the MnZn-Ferrite Nanoparticles As a representative study in magnetization for all synthesized samples, the magnetization curves for as-synthesized MnZn-ferrite nanoparticles after co-precipitation and heating S1 and S3, Table 4-5, and corresponding spectra of magnetic moments are shown on Figures 4-10, and 4-11, respectively. The sample S1 has a less steep magnetization curve and in a spectrum of magnetic moments there is much more pronounced weakly magnetic part. Difference in the magnetization between two samples is higher in low fields. Figure 4-10 shows the magnetization curves for samples S1 and S3 of MnZn- ferrite nanoparticles as obtained from table 4-5, at temperature of T=300 K, where the magnetization is measured in emu/g and the field in Orested. While, Figure 4-11 shows spectra of magnetic moments of samples S1 and S3 of MnZn-ferrite nanoparticles with data obtained from table 4-5, at temperature of T=300 K, where the y-axis is representing initial moment divided by saturation moment and x-axis represents the magnetic moment given in emu. 110 4 1_2 4 sum. xls 50 , MnZnFe σ emu/g 40 30 (emu/g) (emu/g) σ σ σ σ S 3 20 Light fraction Ds1 S Heavy1 fraction Ds3 10 0 0 5000 10000 15000 20000H, Oe H (Oe) Figure 4-10.Magnetization curves for samples S1 and S3 of MnZn-ferrite nanoparticles (Table 4-5), @T=300 K. 4 1_2 4 sum. xls M i/M s 0.08 S Light3 fraction Ds1 S Heavy1 fraction Ds3 MnZnFe 0.06 Ms / 0.04 Mi 0.02 0.00 1.10E+02 1.16E+03 1.24E+04 1.31E+05 M agnetic Magnetic moment, Moment Bohr (emu) magneton Figure 4-11.Spectra of magnetic moments of samples S1&S3 of MnZn-ferrite nanoparticles (Table 4-5), @ T=300 K. 111 Magnetic Properties of the Magnetic Solution For samples of the magnetic solution the magnetization curves are measured at temperatures close to the room temperature as well at the elevated temperature up to 373 K for samples S1 and S3 and up to 473 K for S4 and S5. The magnetization curves for samples S1 and S3 at 300 K are shown on Figure 4-12. The corresponding spectra of magnetic moments are shown on Figure 4-13. The difference in the maximum positions for these two samples is clearly seen. Magnetization curves and the corresponding spectra of magnetic moments -at 303 K- for samples S4 and S5 are shown on Figure 4-14 and 4-15, respectively. For samples of S1-S3 and S4-S5 the smaller particles have a lower magnetization than the bigger ones. MnZnFe md40f_ls.xls (sheet3) MnZnFe 50 , T=294 K σσσ T = 300 K emu/g 40 30 ( emu/g ) ) emu/g ( S 1 σ σ σ σ Ligt fraction Df1 20 SHeavy 3 fraction Df3 10 0 0 5000 10000 15000 20000 H, Oe H (Oe) Figure 4-12. Magnetization curves of sample S1 & S3 (Table 4-5),@ T=300 K. 112 md40f_ls.xls (sheet3) Mi/Ms T=294 K S Light1 fraction Df1 MnZnFe 0.08 S Heavy3 fractionT=300 KDf3 0.06 Mi / Ms /Mi Ms 0.04 0.02 0.00 1.00E-18 1.00E-17 1.00E-16 1.00E-15 1.00E-14 Magnetic moment, emu Magnetic Moment (emu) Figure 4-13. Spectra of magnetic moments of sample S1 & S3 (Table 4-5), @ T=300 K. MnZnFe md4 0 f _ls. xls ( sheet 3 ) T=303 K MnZnFe 50 σσσ, T= 300 K emu/g 40 30 ( emu/g ) ) emu/g ( σ σ σ σ 20 S 5 Heavy fraction Bf2 Light fraction Bf1 10 S 4 0 0 5000 10000 15000 20000 H, Oe H (Oe) Figure 4-14. Magnetization curves of sample S4 & S5 @ T= 300 K. 113 md4 0 f _ls. xls ( sheet 3 ) M i/M s T=303 K MnZnFe T= 300 K 0.20 S Light4 fraction Bf1 0.15 S Heavy5 fraction Bf2 0.10 Mi / Ms Mi 0.05 0.00 1.00E-18 1.00E-17 1.00E-16 1.00E-15 1.00E-14 M agnetic moment, emu Magnetic Moment ( emu ) Figure 4-15. Spectra of magnetic moments of sample S4 & S5 (Table 4-5) , @ T= 300 K. The difference in magnetization is relatively more pronounced in low fields. This is connected with the fact that for smaller particles higher fields are required to orient their magnetic moments. There is a difference for smaller and larger particles also in dependence of magnetization on temperature. The dependence of magnetization on temperature for samples S1 and S3 in magnetic fields of three different intensities of 20, 5 and 1.025 kOe are shown on Figures 4-16 (a), (b) and (c), respectively. In H=20 kOe the relative difference of magnetization for two fractions ∆∆∆σ∆σσσ/σσσHeavy decreases with growing of temperature, whereas in H=1.025 kOe ∆∆∆σ∆σσσ/σσσS3 it grows from 0.20 at 294 K to 0.27 at 373 K. 114 Magnetization vs. Temperature of Samples S1, S2 & S3 in Magnetic a Fields of Different Strengths: (a) H=20KOe, (b) H=5KOe and (c) H=1.025 KOe. H= 20 KOe 45 S3 35 S1 25 15 Magnetization, sigma (emu/g) 5 280 320 360 400 Temperature, T (K) Figure 4-16. Magnetization vs. temperature of samples S1,S2 and S3 in magnetic fields of different strengths: (a) H=20 kOe, (b) H=5 kOe and (c) H=1.025 kOe. 115 Magnetization vs. Temperature of Samples S1, S2 & S3 in b Magnetic Fields of Different Strengths: (a) H=20KOe, (b) H=5KOe and (c) H=1.025 KOe. H = 5 KOe 45 35 S3 25 S1 15 Magnetization, sigma (emu/g) sigma Magnetization, 5 280 310 340 370 400 Temperature, T (K) c Magnetization vs. Temperature of Samples S1, S2 & S3 in Magnetic Fields of Different Strengths: (a) H=20KOe, (b) H=5KOe and (c) H=1.025 KOe. H = 1.025 KOe 45 35 S3 25 15 Magnetization,sigma (emu/g) S1 5 280 310 340 370 400 Temperature, T (K) Figure 4-16.continued. 116 A similar magnetization dependence on temperature for S4 and S5 in the range of 300-473 K are presented in Figures 4-17, (a), (b) and (c). The conclusion that the relative difference of magnetization between Samples ∆∆∆σ∆σσσ/σσσS3 is higher in lower fields holds true for all taken temperature range. It can be seen that the slope of magnetization - temperature curves is lower of both samples S1 to S3 and S4 to S5. This can be quantitatively expressed by values of the thermomagnetic coefficient kT, determined by a linear approximation of experimental dependence in the temperature range of 300-373 K for the samples of S1 to S3, and for the samples of S4 to S5. a Magnetization versus temperature of samples S4 & S5 in magnetic fields of different strength: (a) - H=20 kOe, (b)-H=5 kOe and (c)-H=1.025 kOe. 50 H = 20 KOe 40 S5 30 20 S4 10 Magnetization, (emu/g) 0 280 330 380 430 480 Temperature, T (K) Figure 4-17. Magnetization versus temperature of samples S4 & S5 in magnetic fields of different strength: (a) - H=20 kOe, (b)-H=5 kOe and (c)-H=1.025 kOe. 117 b Magnetization versus temperature of samples S4 & S5 in magnetic fields of different strength: (a) - H=20 kOe, (b)-H=5 kOe and (c)-H=1.025 kOe. H = 5 KOe 50 40 S5 30 20 10 S4 Magnetization, (emu/g) 0 280 330 380 430 480 Temperature, T (K) c Magnetization versus temperature of samples S4 & S5 in magnetic fields of different strength: (a) - H=20 kOe, (b)-H=5 kOe and (c)-H=1.025 kOe. H = 1.025 KOe 40 S5 30 20 10 S4 Magnetization, (emu/g) 0 280 330 380 430 480 Temperature, T (K) Figure 4-17.continued. 118 The thermomagnetic coefficients for samples S1 and S3 as well as for S4 and S5 in the magnetic field of different strength are presented on Figures 4-18, (a) and (b), respectively. The character of the dependence kT(H) for smaller and bigger particles is similar. It should be noted that kT increases with growing of H up to 5 kOe for samples S1, S3 and up to 3 kOe for S4, S5 and in stronger fields remains approximately constant. For S4, kT reaches the maximal value in H~2 kOe and even slightly decreases in stronger fields. The maximal value of kT for S3 exceeds that of S1 for ~ 1.15 times, whereas kT for S5 exceeds that of S4 for ~ 1.4 times. a Thermomagnetic coefficients kT of samples S1 & S3 (a) and S4 & S5 (b) in a temperature range of 294-373 K versus magnetic field strength S3 0.21 0.19 0.17 0.15 S1 0.13 KT, (emu/g*K) KT, 0.11 0.09 100 5100 10100 15100 H, (Oe) Figure 4-18. Thermomagnetic coefficients kT =-δδδσ/δσ/σ/σ/δδδδΤΤ of samples S1 & S3 (a) and S4 & S5 (b) in a temperature range of 294-373 K vs. magnetic field strength. 119 b Thermomagnetic coefficients kT of samples S1 & S3 (a) and S4 & S5 (b) in a temperature range of 294-373 K vs. magnetic field strength 0.21 0.19 S5 0.17 0.15 KT, (emu/g*K) KT, 0.13 S4 0.11 0.09 0 5000 10000 15000 H, (Oe) Figure 4-18. Continued. In conclusion, nanoarticles of the samples S1 and S4 have a lower Fe content and higher Zn substitution degree x than particles of the samples S3 and S5. Samples S1 and S4 has a lower specific magnetization than the samples S3 and S5. Especially this difference is pronounced in low fields. In comparison with the co-precipitated particles separated by sedimentation centrifugation leads to a better separation in size of particles. Properties of particles - chemical composition and magnetic properties- are dependent on size of particles. Smaller particles have higher content of Fe and a higher content of associated water than the bigger ones. Also, smaller particles possess a lower specific magnetization and lower thermomagnetic coefficients than the bigger ones in all the range of magnetic field of 0-20 kOe. In low fields difference in the magnetization for smaller and bigger ones is relatively more pronounced. 120 4.4 Magnetic Nanoparticles SQUID Characterization Results MnZn-Ferrite Magnetic Nanoparticles Characterization Results Ferrites are a group of technologically important magnetic materials of current interest, in particular Gd-, Mn-, Zn-ferrites. For example, Gd-ferrite, which has a low- Curie temperature -298 K- and high-pyromagnetic coefficient [i.e. high (∂M/∂T)H]. Although, most of the earlier discussion was related to MnZn-ferrite nanoparticles as representative for all synthesized nanoparticles in this study, various other magnetic nanoparticles were synthesized with an aim to find a material which will have a Curie temperature of 315 K. Their characterization results are discussed, herein. The magnetic properties of the particles were probed by SQUID in powder and liquid forms prepared as discussed in chapter 3. The weight of each sample and the capsules was taken prior to all tests, e.g. MnZnFe weight was 30.88 mg, etc. Magnetic moment measurements were performed using a physical property measurement system (PPMS) from Quantum Design. Zero-field-cooled -ZFC- and field- cooled -FC- measurements were done at 100Oe magnetic fields. To study the magnetic properties of the as-synthesized MnZn-ferrite particles the particles were dried of solution to form surfactant-coated particles in powder form. This powder was filled in a gelatin capsule in loosely packed form that provided these particles with rotational as well as translational degrees of freedom under applied magnetic field. Note that the magnetic moment values in all figures are for qualitative comparison only because of the varying loading factor of the magnetic nanoparticles into various matrices. M-H curves were measured at 5 K and 300 K from 60 KOe to - 60 KOe, figures 4-19 and 4-20, respectively. The nanoparticles display typical ferromagnetic behavior with hysteresis at both 5 K and 300 K. No significant coercivities were recorded which indicates the superparamagnetic behavior of the synthesized nanoparticles. When superparamagnetic particles are used for localized hyperthermia therapy, only a weak magnetic field is needed to generate heat compared to the strong field required for ferromagnetic particles. The superparamagnetic particles are economical, harmless to the living body, and appropriate for localized hyperthermia therapy. However, in order to have superparamgnetic properties the particles must have 121 nanodomains of 10-20 nm in diameter and must also be coated by organic or inorganic compounds that prevent any interaction between the magnetic particles, [162,163]. Zero-field-cooled (ZFC) and field-cooled (FC) magnetization curve shown in Figure 4-21, was recorded for MnZn-ferrite sample by applying a probe field of 50 Oe, in the temperature range 5- 450 K. All the measurements were performed on particle solid dispersion in PEG. The main magnetic properties are summarized in Table 4-8. Table 4-8. Main magnetic properties of the prepared MnZn-ferrite nanoparticles Pyro- Magnetic M at S magnetic moment S. T /V 300 K T B XRD coefficient Linear per # B (K/nm3) (emu/ (K) 3 (∂∂∂M/∂∂∂T)H Tc particle cm ) 3 (emu/cm K) (K) (µµµB) 1 298 2.0 205 5 315 13819.70 2 305 2.7 203 4.4 319 12780.00 3 289 1.9 200 4.2 325 8837.69 were, TB is the blocking temperature evaluated from the maximum of the ZFC curves, TB/VXRD is the ratio of the blocking temperature to the particle volume evaluated by XRD size measurement, Ms at 300 K is the saturation magnetization at 300 K. As expected for a size-spread collection of superparamagnetic particles the ZFC curves show a peak whose maximum is commonly associated to the mean blocking temperature of the assembly, TB, i.e., the temperature at which the relaxation time of the magnetization equals the characteristic time of the measurement. According to the narrow size distribution observed the blocking temperature spreading, estimated between the temperature where FC and ZFC curves diverge and the temperature where the FC becomes constant, is quite small for all the samples. The mean blocking temperature TB clearly increases with increasing particle size; from 177 K for up to 289 K. It is worthwhile to note that, even a small change in the particle size produces a significant modification of the magnetic properties related to the magnetization relaxation. 122 In figure 4-19, the M–H loops at 300 K for the loosely packed particles do not show any resolvable coercivity but on the other hand, these particles doped in a PEG matrix show a significant difference with an opening up of hysteresis with coercivity of 272 Oe at 300 K. These results are consistent with the ZFC-FC and SEM results that show clustering of particles and an increase in the blocking temperature due to increased inter-particle interactions. While the presence of large clusters with agglomerated particles does favor enhanced magnetic coupling between particles in these clusters, there could be other contributions to the increase in blocking temperature in these polymer nanocomposite samples, such as effects due to shape anisotropy and matrix mediated interactions. Systematic investigations in a variety of nanocomposite materials are needed to understand the various mechanisms that could contribute, and detailed discussions are beyond the scope of this dissertation. In figure 4-20, the low temperature M-H curve is shown. It can be seen that the loosely packed powders of the MnZn-ferrite material did not exhibit any coercivity even at 5 K, which is well below the blocking temperature. This absence of coercivity even in the ferromagnetic regime may be due to mechanical rotation and translation of the nanoparticles in the external magnetic field. This behavior was also shown by some other loosely packed magnetic nanoparticle systems. The mechanism by which the magnetic moment of a loosely packed particle below the blocking temperature may relax after removal of a magnetic field involves the bulk rotation of the particle as the magnetic moments of the particles are fixed relative to the crystal axes of the particles. This mechanism is similar to the Brownian relaxation of the particles in fluid. On switching the field from positive to negative values the particles tend to align with the magnetic field and thus the hysteresis effect is lost. On the other hand, for such a system above the blocking temperature, the magnetic moment vector itself is free to rotate in response to the applied field while the particle remains stationary. This rotation is known as Neel relaxation. The response of the magnetization of such a system to alternating magnetic fields will be quite interesting. In figure 4-20, the 5K M-H loop measurements on MnZn-ferrite particles is shown. It can be seen that no increase in the coercivity is present. The presence of 123 variable size clusters ranging from single domain particles to possible multi-domain agglomerates makes the interpretation of the magnetization data somewhat complicated, as there would be a collective response from these different regions. However, comparative experimental studies with particles in the PEG matrix do clearly correlate the increase in blocking temperature in the case of the Peg-doped particle system to the presence of larger clusters and the associated increase in inter-particle interactions. In figure 4-21, ZFC–FC curves for loosely packed MnZn-ferrite nanoparticles are presented. It can be observed that the loosely packed particles show a well defined peak at 48 C, typical of a superparamagnetic to blocking transition. In this case, an external field can cause rotational and translational motion of individual particles that would effectively contribute to the temperature and field dependence of the overall magnetization. However, in this case the applied field of 100 Oe was sufficiently low for such contributions to be present. In the case of particles in polymer - PEG, of course, any physical motion due to external magnetic field is arrested because of the constraining matrix. 1.25 300 K 1 0.75 0.5 0.25 0 -60000 -40000 -20000 0 20000 40000 60000 -0.25 -0.5 Magnetization M ( emu / g ) / g ) emu Magnetization ( M -0.75 MnZnFe, 30.88mg -1 -1.25 Magnetic Field (KOe) Figure 4-19. Magnetization Curve of MnZn-ferrite nanopaticles taken by SQUID magnetometry at 300 K. 124 2.5 5 K 2 1.5 1 0.5 0 -60000 -40000 -20000 0 20000 40000 60000 -0.5 -1 Magnetization M ( emu / g ) / g ) emu Magnetization ( M -1.5 -2 MnZnFe, 30.88mg -2.5 Magnetic Field (KOe) Figure 4-20. Magnetization Curve of MnZn-ferrite nanopaticles taken by SQUID magnetometry at 5 K. 9 FcZFc at 100 Oe 8 MnZnFe, 30.88mg 7 6 5 4 3 Long Moment ( emu ) emu ( Long Moment 2 Tc = 315 K 1 0 0 50 100 150 200 250 300 350 400 450 Temperature (K) Figure 4-21. ZFC and FC magnetizations as a function of temperature for MnZn-ferrite nanoparticles sample in PEG measured applying a magnetic field of 100 Oe. 125 In figure 4-21 the FC magnetization data show a downturn at low temperature. This signature is often seen in nanoparticle systems when the applied field - typically 50 or 100Oe in ZFC&FC measurements - is not enough to overcome the effective anisotropy field of the system. At higher applied fields this downturn is not seen and one obtains the conventional FC variation in magnetic nanoparticles. For not interacting single domain particles, the relaxation of the magnetic moment can be described by the Neel model which predicts an Arrhenius law: τN = τ0 exp (KV / TkB) (1) where τN is the relaxation time, K the anisotropy constant, V the particle volume and τ0 the attempt time. According to this model, a rough estimation of the expected value of the increasing rate TB/V, can be performed by taking as characteristic time of the instrument tm=100 s and keeping K and τ0 constant and equal to the typical values obtained by AC 3 −20 susceptibility measurements presented in the following (K≈1.5 MJ/m , τ0≈10 s). This 3 evaluation led to TB/V=2.1 K/nm , in good agreement with the average experimental 3 value 2.0 K/nm . It could be also noted that TB/V slightly decreases with particle dimensions suggesting that τ0 and or K decrease with increasing sizes. Isothermal magnetization curves were measured at 300 K up to 6.5 T for all the samples. As an example the M vs. 0H curve of sample 4 is reported in figure 4-19. Due to the low concentration of the samples the diamagnetic contribution of the PEG matrix dominates the high-field part of the magnetization curve and it must be removed for a correct interpretation of the measurement results. The mass magnetization of the MnZn- ferrite particulate was evaluated by subtracting from the measured value the diamagnetic contribution given by: Mdia=−χPEG (1-c) / cH (2) where c is the weight concentration of MnZnFe2O4 of the sample and the matrix −10 3 susceptibility was separately measured and found to be χPEG=4π×6.5.10 m /kg. Once the diamagnetic contribution was removed, the magnetization curves are not saturated 126 suggesting that the contribution of the surface spins plays a relevant role even at the highest fields. Due to the very low concentration of the samples, a large error affects the diamagnetic contribution corrections which, is particularly important in the high-field region. The error propagates on the Ms - values hindering the observation of a trend of the saturation magnetization with particle dimensions. The Curie temperature for the MnZn- ferrite nanoparticles was found to be ~ 315 K. ZnGdFe Magnetic Nanoparticles Characterization Results During the process of synthesizing different materials through this study it was found out that Curie temperature of Zn Ferrite particles is lower than the preferred 430C K. Also its pyromagnetic coefficient is low. It was observed from the study of the MnZn- ferrite particles and Gd substituted MnZn-ferrite nanoparticles that addition of Gd results in an increase in the Curie temperature as well as its pyromagnetic coefficient. Thus, Gd substituted Zn Ferrite nanoparticles were synthesized using chemical co-precipitation method so as to increase the Curie temperature with the preferred 430 C as well as to increase its pyromagnetic coefficient. The characterization results of these particles are presented below in figures 22-24. 5 ZnGdFe, 23.68mg 4.5 FcZFc at 100 Oe 4 3.5 3 2.5 Tc = 315 K ) emu ( Long Moment 2 0 50 100 150 200 250 300 350 400 450 Temperature (K) Figure 4-22. FcZFc at 100 Oe or, temperature dependence of magnetization for ZnGd-ferrite. 127 30 300K 20 10 ( emu / g ) emu ( 0 -60000 -40000 -20000 0 20000 40000 60000 -10 agnetization M M -20 ZnGd-ferrite, 23.68 mg -30 Magnetic Field (KOe) Figure 4-23.Magnetization Curve of ZnGd-ferrite nanopaticles taken by SQUID magnetometry at 300 K. 125 5K 75 25 -60000 -40000 -20000 0 20000 40000 60000 -25 Magnetization M ( emu / g ) / g ) emu Magnetization ( M -75 ZnGd-ferrite, 23.68 mg -125 Magnetic Field (KOe) Figure 4-24. Magnetization Curve of ZnGd-ferrite nanopaticles taken by SQUID magnetometry at 5 K. 128 The Curie temperatures of ZnGd-ferrite samples were found to be as follows: for x = 0.02, 315 K and for x = 0.05, 340 K. The substitution of Gd in Zn Ferrite leads to an increase in Curie temperature and pyromagnetic co-efficient. Also Curie temperature increased with increased Gd substitution. Addition of Gd3+ ions results in their occupancy of the octahedral sites. The preference for octahedral sites maybe attributed to their large ionic radii. Since the ionic radii of the Gd3+ ions is large, there is a decrease in the distance between these and the oxygen ions when adding Gd ions consequently strengthening the B-B interaction [164]. As a result the ions at the octahedral sites no longer have their moments parallel to each other. A part of these ions have moments aligned antiparallel to the other atoms on these octahedral sites. This results in a reduction in the net magnetic moment of the octahedral atoms. As the Gd substitution is increased, more and more octahedral atoms have their moments antiparallel As a result the B-B interaction is strengthened which consequently results in an increase in the Curie temperature. The Curie temperature of the ZnGd-Ferrite nanoparticles with x = 0.02 was found to be the preferred 315 K for hyperthermia application. The pyromagnetic co-efficient was higher than that of Zn Ferrite. ZnNdFe Magnetic Nanoparticles Characterization Results The use of different materials with larger magnetic anisotropy and larger magnetic moment is strongly envisaged since it can allow a significant improvement of the material efficiency, particularly as concerns magnetic hyperthermia. Moreover, the use of such materials may allow reducing the particle size introducing additional benefits, as a reduced toxicity and longer half-life time in the blood with consequently increased taken up by the reticulo-endothelial system. Zinc Nidenium-ferrite, ZnNdFe2O4 was studied as a candidate material for magnetic hyperthermia. It is a magnetic material with almost the same saturation magnetization as magnetite but with a crystalline anisotropy one order of magnitude larger. As a consequence in nanosized materials, the magnetic moment relaxes much slower than in magnetite or maghemite nanoparticles with similar size. Nanoparticles of ZnNd-ferrite were prepared by a co-precipitation method as discussed in chapter 3. 129 The size controlled synthesis and the investigation of static and magnetic properties of highly monodisperse ZnNd-ferrite nanoparticles were studied. The characterization of the physical properties was mainly focused on the correlation between particle size and magnetic properties with particular attention to those determining the hyperthermic behavior. The understanding of such relations is an imperative requirement to attain the capability of tailoring the properties of the nanomaterials and to select the best promising products for the proposed biomedical magnetic hyperthermia application. Using Fc & ZFc at 100 Oe or, temperature dependence of magnetization graph for ZnNd-ferrite magnetic nanoparticles, shown in figure 4-25, Curie temperature was measured to be 315 K. Figure 4-26 and 4-27 show the magnetization curves of ZnNd- ferrite nanopaticles taken by SQUID magnetometry at 300 K and 5K, respectively. 0.0022 ZnNd-Ferrite, liquid 0.002 FcZFc at 100 Oe 0.0018 0.0016 Tc=315 K 0.0014 TB=298 K 0.0012 Long Moment ( emu ) emu ( Long Moment 0.001 0 100 200 300 400 Temperature (K) Figure 4-25.Fc & ZFc at 100 Oe or, temperature dependence of magnetization for ZnNd-ferrite. 130 0.02 300K 0.015 0.01 0.005 / g ) emu ( 0 -60000 -40000 -20000 0 20000 40000 60000 -0.005 -0.01 agnetization M M -0.015 ZnNd-ferrite -0.02 Magnetic Field (KOe) Figure 4-26. Magnetization Curve of ZnNd-ferrite nanopaticles taken by SQUID at 300 K. 0.03 5K 0.02 0.01 ( emu / g ) emu ( 0 -60000 -40000 -20000 0 20000 40000 60000 -0.01 agnetization agnetization M -0.02 M ZnNd-ferrite -0.03 Magnetic Field (KOe) Figure 4-27. Magnetization Curve of ZnNd-ferrite nanopaticles taken by SQUID at 5 K. 131 GdZnCeFe Magnetic Nanoparticles Characterization Results Figures 28-29 follow a similar rational and explanation laid out previously for the rest of the as-synthesized nanoparticles. 5 4.5 4 (emu) 3.5 FcZFc at 100 Oe 3 GdZnCeFe, 30.52mg 2.5 2 Long Moment Long Moment 1.5 1 Tc=315 K 0.5 0 0 50 100 150 200 250 300 350 400 450 Temperature (K) Figure 4-28. FcZFc at 100 Oe or, temperature dependence of magnetization for GdZnCe-ferrite. 80 60 5K 40 20 0 -60000 -40000 -20000 0 20000 40000 60000 -20 -40 -60 GdZnCeFe, 30.52mg Magnetization M ( ( M emuMagnetization / g ) -80 Magnetic Field (KOe) Figure 4-29. Magnetization Curve of GdZnCe-ferrite nanopaticles taken by SQUID at 5 K. 132 20 300K 15 ) 10 5 0 -60000 -40000 -20000 0 20000 40000 60000 -5 -10 Magnetization / g emu Magnetization ( M -15 GdZnCeFe, 30.52mg -20 Magnetic Field (KOe) Figure 4-30. Magnetization Curve of GdZnCe-ferrite nanopaticles taken by SQUID at 300 K. 4.5 Magnetic Heating Tests and Results The results of various heating tests are given below for MnZnFe, ZnGdFe, ZnNdFe and GdZnCeFe. The results show the temperature rise effect of ferromagnetic samples. What one hopes to find is consistently higher temperatures in the ferromagnetic material over ordinary materials. The results here should be stressed in terms of temperature rise over heating, due to at least two reasons: first, the likelihood of RF interference contributing to an apparent increase in temperature and second the inability to completely characterize heat- producing mechanisms in this specific experimental setting. Various procedures will be attempted to show the contributions of the first reason by using the proper control materials. In addition, second reason will be discussed in terms of current research to help justify claims of more pronounced temperature rises in ferromagnetic materials. 133 MnZn-Ferrite Heating and Curie Temperature Testing Preliminary results indicate a constant source of temperature rise in MnZn-ferrite magnetic nanoparticles and similar trend was observed in all synthesized samples. Care was taken of the coil apparatus to ensure that temperatures increases could be attributed in large part to ferromagnetic losses, and not to Ohmic heating from the coil, or infra red absorption. Little could be done with the equipment to prevent all radiofrequency interference, but its influence on temperature rise is tested below. All moving objects, including individuals, were kept away from the magnetic field near the sample. Sample was placed within the coil and insulation helped to prevent heat losses to the atmosphere, giving increased accuracy in the temperature rise of the ferromagnetic material. Equation 3 (from chapter 3), dB 2 −= E (3) dt r indicates that this placement is not as advantageous as the coil center should have zero E- field, but within the coil one finds the largest magnetic fields, so outside the null one would still expect an electric field. Tests show that temperature rises are still possible with this placement. The thermocouples were tested to see the effects of the RF signal that may cause interference and, therefore, inaccurate readings. Experimental setup for the magnetic heating experiments was as follows: one Cu-Const. thermocouple in sample, another thermocouple outside of the sample but within the insulation. Samples were placed in the center and about halfway down the coil. An 18 Watt forward power driven from the at 961 kHz. A peak-to-peak voltage, V(p-p) = 17.0 V near the center of coil - measured by probe and oscilloscope- to give a relative indication of input power. The Apparatus was cooled down between runs by water-filled tube at the insulation and water-filled glove around the coil. The thermocouple test results are given in Table 4-. Temperatures below given in degree Celsius as T(sample)/T(insulation) (°C) format: thus, 20.0/22.0 would mean 20.0°C for the sample, 22.0°C for insulation. 134 Table 4-9. Thermocouple testing without sample. T (0C) Run 1 Run 2 0 25.7/25.7 25.0/24.8 12 26.2/25.8 25.2/25.1 24 - 25.3/25.2 36 26.5/26.1 25.4/25.3 48 - 25.4/25.3 60 26.6/26.2 25.5/25.4 Curie-Temperature Testing It is a well known fact that, it is rather difficult to accurately determine the Curie temperature Tc of a ferromagnetic body, [165]. These difficulties are incurred because the curve of the spontaneous magnetization versus temperature - σ vs. Tc - does not intersect the temperature axis steeply, but rather bends over to form an asymptotic tail. The temperature defined by extrapolation of the main part of the σ (T) curve is called the ferromagnetic Curie temperature, Tc. The disappearance of spontaneous magnetization in the vicinity of the Curie temperature is accompanied by changes in the internal energy. These changes are also associated with the λ-like spike on the curve of the specific heat Cp vs. temperature. Tc, can be determined by measuring both these quantities (σ and Cp) as function of temperature. In a classical Faraday balance method, the relative σ(T) dependence is determined by using a sensitive balance and by placing a sample in a magnetic field gradient while sweeping temperature. This Faraday balance method is essentially the basis for thermogravimetry, and it is widely used when a relative value of σ is sufficient. However, the method is rather expensive and it requires a long measurement time. Differential scanning calorimetry -DSC- methods which are routinely used in laboratory operations may also be used to determine Tc since this technique directly detects changes in internal energy including those associated with magnetic transitions. Unfortunately, the standard -Perkin-Elmer- equipment can reliably detect Tc only in those 135 alloys where the magnetic transition is associated with the evolution of substantial energy. In other materials where the magnetic transition is not as pronounced and is accompanied by smaller changes in enthalpy DSC cannot be used to identify Tc. A good example of such a class of materials is the low magnetostriction cobalt-base amorphous alloys. In addition DSC measurement of Tc requires at least one hour [165]. There is a simple, inexpensive, accurate, and fast susceptibility method to measure Tc which has been in use for a long time. In this method a ferromagnetic sample serves as a core of a linear variable differential transformer - LVDT. Changes in LVDT output due to changes in sample - core - susceptibility may be detected electronically. Electronic circuitry accurately detects when both the ferromagnetic moment of a sample and the sample susceptibility approach zero with increasing temperature. This temperature may be defined as the Curie temperature. Because the magnetic disordering process in the vicinity of Tc is different for each particular magnetic material, differences exist in the recording of the decay of ferromagnetism by the aforementioned methods. Diminishing returns on temperature rise in MnZn-ferrite nanoparticles should occur near Tc, i.e, changes in temperature for MnZn-ferrite nanoparticles temperature should decrease as the starting temperature of tests increases towards Tc due to the ferromagnetic gradual loss of magnetization given that losses and heat production are considered a result of hysteresis. Understanding ferromagnetic behavior above body temperature will help to find if particles can heat to hyperthermia temperatures and self- limit near the Curie temperature. The setup for Curie temperature testing for various nanoparticles was as follows: forward power = 40 W - similar to load power since the equipment is at resonance. Function generator and power amplifier tested at V = 30 V peak-to-peak and frequency of 961 kHz at coil resonance. A one minute break between each run was observed to help prevent dominant coil heating rather than nanoparticles heating. Also, no cool down via water-filled tube or glove was adhered to, to allow higher temperatures to be reached, utilizing remnant heat produced by nanoparticles as well as the coil. The heating pattern for MnZn-ferrite magnetic nanoparticles is shown in figure 4-31. 136 The heating pattern of Mn-Zn-Fe [ Zn = 0.9 conc.] (20mg/ml) using the Alcohol Thermometer at 963 KHz Temp. Rise @ 963 KHz, 20 W 48 45 Temp. Rise @ 961 KHz, 20 W 42 C) 0 39 36 33 Temperature ( Temperature 30 27 24 0 5 10 15 20 25 30 35 40 45 50 55 Time in minutes Figure 4-31 .The heating pattern of MnZn-ferrite [Zn= 0.9 conc.] (20mg/ml) using the Alcohol Thermometer at 961 kHz. Heating Patern of Mn-Zn-Fe (Zn 0.5 conc) [20mg/ml] using Alchol thermometer at 961KHz 49 44 39 34 Temperature (C) Temperature Temp. Rise @ 961 KHz, 20 W 29 24 0 5 10 15 20 25 30 35 40 45 50 55 60 65 Time (minutes) Figure 4-32.The heating pattern of MnZn-ferrite [Zn = 0.5 conc.] (20mg/ml) using the Alcohol Thermometer at 961 kHz. 137 Figure 4-31, shows that the temperature differences decrease as one would expect for MnZn-ferrite as it nears the Curie point. Other reasons for the decrease include water adsorbing on the nanoparticles or the system falling out of resonance over the duration of testing. The latter was unlikely as the system was tested for maximum power to load using an oscilloscope and probe. The relatively small temperature rise at 23.9°C is likely due to the absence of coil heating at the start of the experiment. Thus, the sample must become in equilibrium with its environment to have no heating assistance by the solenoid. Note that zinc concentration in this case is 0.9. The results in Figure 4-32, show diminishing changes in temperature as the sample is heated to higher temperatures which is expected of ferromagnetic behavior. Heating was significantly greater in each run for MnZn-ferrite than that of the insulation, indicating that sample temperature rise can be attributed to nanoparticles rather than the coil. Zinc concentration in this case is 0.5. In addition, when the insulation temperature was below Tc the temperature of the MnZn-ferrite sample showed no increase above the Tc which is 47°C. The sample was heated at two different frequencies as depicted in the above plot. At 963 KHz heating up to Curie temperature occurred and stopped or saturated beyond 470C. Evidence given here suggests that the hyperthermia temperature range could be reached experimentally, although one still needs to consider whether a similar amount (or less) of MnZn-ferrite nanoparticles can be used and concentrated into a tumor site with similar results. Future results that indicate a slowdown at Tc but marginally exceeding this value should not be surprising in this dense powder form, for the material, though in the spinel phase, may still have a tendency to heat due to hysteresis losses. That being said, the possibility of significant heating above 47°C would be unanticipated. If the temperature rise effect is due to ferromagnetic response to H-fields, the realignment of the domains parallel to the H-field are a likely cause as smaller domains not in the direction of the H-field are minimized. This may also play a role in diminishing temperature gains closer to Tc. At higher temperatures the smaller ∆T values from may be accounted since the magnetization is weakened at higher temperatures and the difference in the magnetization between one domain and the next is therefore minimal and a less coercive H-field may be needed to align the domains. 138 The heating patterns, figures 4-33 to -36, for the rest of the particles are listed in their respective figures below. The resonance frequency at ~ 961 KHz proved to be effective for most of the samples to produce the desired magnetic heating in the synthesized nanoparticle systems discussed earlier. ZnGd-Ferrite Heating and Curie Temperature Testing Heating pattern of ZnGdFe (Zn, x=0.02 conc.) [20mg/ml] using alcohol thermometer @961 KHz. 45 40 35 30 25 20 15 10 Temperature (deg. C) (deg. Temperature Temp. Rise @ 5 961 KHz, 20 W 0 0 15 30 45 60 75 90 Time (minutes) Figure 4-33.The heating pattern of ZnGd-ferrite [Zn = 0.02 conc.] (20mg/ml) at 961 kHz. The heating pattern ofZn-Gd-Fe (20mg/ml) using the Alcohol Thermometer at 433 KHz 30 Temp. Rise @ 433 KHz, 20 W C) 0 27 Temperature ( 24 0.00 5.00 10.00 15.00 20.00 25.00 30.00 35.00 40.00 45.00 50.00 Time in minutes Figure 4-34.The heating pattern of ZnGd-ferrite [Zn, x = 0.02 conc.] (20mg/ml) at 433 kHz. 139 GdZnCe-Ferrite Heating and Curie Temperature Testing The heating pattern for GdZnCeFe [Gd, x=0.2 conc.] 49 (20 mg/ml) using alcohol thermometr @963 KHz 44 39 34 Temp. Rise @ 963 KHz, 20 W 29 Temperature, deg. C Temperature, 24 0 5 10 15 20 25 30 35 40 45 50 55 60 65 Time (Minutes) Figure 4-35.The heating pattern of GdZnCe-ferrite [Gd, x = 0.2 conc.] (20mg/ml) at 963 kHz. ZnNd-Ferrite Heating and Curie Temperature Testing Heating pattern of ZnNdFe (Zn, x=0.5 conc.) [20mg/ml] using alcohol thermomater @ 961 KHz. 50 45 40 35 30 25 20 Temp. Rise @ 961KHz, 20 W 15 10 Temperature (deg. Temperature C) 5 0 0 15 30 45 60 75 Time (minutes) Figure 4-36.The heating pattern of ZnNd-ferrite [Zn, x = 0.2 conc.] (20mg/ml) at 961 kHz. 140 Setbacks. A sample placed closely outside the coil should heat by ferromagnetic losses, infra red radiation from the coil, and conductive losses due to the heating of the coil through the air. In many informal tests, a lack of significant temperature rise occurred outside of the coil. Such tests included heating outside of the coil’s insulation while placing metal between the sample and the coil and placing Plexiglas between the sample and the coil, of which both were used to control the onset of coil heating. In comparison, Vp-p outside the sample measured to an average of about 6 V using the probe/oscilloscope at 40W ~ 2cm from the coil, nearly three times less than what can be created inside the insulation. This is nearly the same as what is recorded inside the insulation using the frequency generator without power amplification. As a relative indicator of power output from the coil, this is insufficient for temperature rise effects. On the other hand, a sample placed within the insulation should not strongly heat since the magnetic field strength minimizes to zero at the boundary of the coil, being max. at the center. In addition, no sample was placed perfectly at the center of the coil, and the samples, of course, extended beyond the center. Furthermore, temperature anomalies are difficult to detail, as variability exists if the thermocouple is not placed in the same part of the sample: the Ohmic losses of the coil - resistive losses due to alternating current- might create slight temperature variations in the insulation. Also, the output of the power amplifier and frequency generator may give variable output as both are controlled by analog inputs. To overcome this, multiple tests are needed. The original samples prepared showed very little heating after repeated testing. Initial tests on that sample showed similar results to that of the new sample, but subsequent runs produced little heating. Hypothesized reasons for such diminished results may result from application of too strong of an H-field strength to the sample such that the remanent magnetization was increased to a level that could not be reached again in following runs, or heating changed the structure of the sample or left the sample susceptible to atmospheric effects. 141 CHAPTER 5 ANALYSIS AND ESTIMATE OF MAGNETIC HEATING 5.1 Dominant Magnetic Heating Mechanisms Measurement of the Neel relaxation of Magnetic Nanoparticles Measurements of the AC susceptibility χ(ω) of magnetic particles, where: χ(ω) = χ ' (ω) − iχ " (ω) (1) have been performed to the frequency region of interest - 1MHz. Studies were done on colloidal suspensions of the magnetic particles. The frequency region of particular interest, i.e. where the Nee1 relaxation of the magnetic moments of the particles occurs was found at and above the frequency of 1MHz. For magnetic solutions two relaxation mechanisms can occur, one by rotational Browning diffusion characterized by a relaxation time τB where, 1 τ B = 3V η / kT (2) and the other by Nee1 relaxation with a relaxation time τN [166], −1 − 2/1 τ N = f 0 (KV / kT ) exp(KV / kT) (3) where V1 and V are the hydrodynamic and magnetic volumes respectively, η is the -1 -9 viscosity, K is the magnetic anisotropy constant and f0 has value of approximately 10 seconds. For values of the median radius of the lognormal volume fraction of particles between 2.5 and 5 nm, the frequency fmax at which the maximum in the loss peak 142 [χ”(ωmax)] occurs, assuming Neel relaxation only, is given by ωτN = 1 which occurs approximately at 1 MHz. For solutions containing small particles ~ 5 nm, and using typical values of saturation magnetization per unit volume Ms and K for magnetite, the assumption that only Neel relaxation need be considered is valid. This is not necessarily so for large particles, i.e. larger than 10 nm. For larger particle size distributions the larger particles will undoubtedly relax by the Brownian mechanism but the frequencies at which this would occur are generally well below the frequencies expected for Neel relaxation. In the systems studied herein, namely 3 colloidal suspensions of MnZn-ferrite of mean diameter sizes: 5.6, 5.05 and 4.6 nm, values of Ms and K are known approximately. Ms can be obtained from the knowledge of the saturation moment of the nanoparticle system and K by measurement of the thermal decay of remanent magnetization. A typical value of Ms for particles of magnetite is approximately 0.44 T and typical values of K are in the region l04 - 4 × l04 J m3. The measurement of the complex susceptibility of three magnetic nanoparticles solutions mentioned earlier that contains particles having a volume distribution with median radii for three MnZn-ferrite nanoparticle samples: (1) 5.6 nm, (20 5.05 nm and (3) 4.6 nm in the frequency range l MHz using a measuring technique similar to the one used in [167], are depicted in figure 5-1. Measurements were done at the chemistry department in FSU. Figure 5-1 shows a plot of χ'' (ω) against log f for the three samples. The loss peaks occur at 1MHz, 963 and 961KHz for increasing particle size. The data presented here show virtually complete loss peaks for particles relaxing by the Neel mechanism. The theory developed by Debye to account for the anomalous dielectric dispersion in dipolar fluids has been used [168] to account for the analogous magnetic case of magnetic solutions. Debye's theory holds for spherical particles when the magnetic dipole-dipole interaction energy Um is small compared with the thermal energy kT. This condition is satisfied for the three samples studied here where Um/kT = 0.06 - 0.2. 143 According to Debye's theory: χ(ω) = χ 0 1( + iωτ N ) 2 2 2 2 (4) = [χ 0 1( + ω τ N )] − i[ωτ N χ 0 1( + ω τ N )] where χo is the static susceptibility. This expression gives the theoretical maximum of χ”(ωmax)/χ0 to be 0.5. Note that, Log 961 KHz and Log 963 KHz = ~ 6. Normalized Plots of X" Against Log f Hz for Three Samples of MnZn-ferrite Nanoparticles: 1 (1) 5.6 nm (2) 5.05 nm and (3) 4.6 nm. 1 [Note: Log 961 KHz =6] 0.8 2 ) ω ω ω ω ) ”( ω ω ω ω 0.6 χ χ χ χ ”( χ χ χ χ 0.4 3 0.2 0 0 2 4 6 8 -0.2 Log f (Hz) Figure 5-1. Normalized plots of χ”(ω) against logf (Hz) for three samples of MnZn-ferrite Nanoparticles: (1) 5.6 nm (2) 5.05 nm and (3) 4.6 nm. 144 Superparamagnetic Nanoparticles Dominant Mechanisms Limitations concerning the particle concentration and the field parameters as well as the effect of the heat conductivity of the tissue [169] result in the demand of a very high specific heating power, SHP: " 2 SHP = µ0 π f χ H / ρ (5) where f is the frequency and H the amplitude of the field and ρ the density of the particles. For materials with linear magnetic properties the specific heating power can be calculated from the χ” data [170], according to equation 5. Estimations of SHP in the order of 1000 W/g or higher should be a goal of the material development. Superparamagnetic iron oxide [Fe3O4, γ-Fe2O3] particles commonly shows SHP values of less than 100 W/g , at 400 kHz and 10 kA/m, [171]. One feasible way to obtain a higher SHP is the use of magnetic materials with higher specific magnetization, e.g. MnZn-ferrite, GdZn-ferrite, ZnNd-ferrite, GdZnCe- ferrite or iron. As a first step in this direction investigation of the properties of aforementioned particles was carried out. Samples preparations, SQUID and XRD investigations were discussed earlier in chapter 4. As a representative for all synthesized nanoparticle materials three ZnNd-ferrite samples were used for the measurements. The first was a powder sample. The second one was an original suspension of the particles and the third sample was prepared by drying 20 ml of the fluid on a small piece of paper. The paper was rolled and inserted into the capsule used for the SQUID measurements. In this kind of sample the particles are immobilized but in contrast to the powder the concentration is much lower resulting in much lower demagnetization effects which simplifies the analysis of the data. In order to obtain the specific magnetization of the particles the powder sample was investigated by SQUID. The data are shown in figure 5-2. The specific saturation magnetization was 77.5Am2/kg. Another SQUID measurement was performed with the immobilized sample. Mean particle size obtained from XRD was ~ 6 nm. 145 0.02 0.015 ) 0.01 emu/g ( 0.005 0 -60000 -40000 -20000 0 20000 40000 60000 -0.005 agnetization -0.01 M -0.015 -0.02 Magnetic Field (KOe) Figure 5-2. Specific magnetization of the ZnNd-ferrite powder. Characterization of the AC behavior The characterization of the behavior in AC magnetic fields was performed by two methods: AC susceptometry and calorimetric measurement of the high specific heating power. For AC susceptometry a field coil which is connected to a function generator provides an AC field with an amplitude up to 60 A/m and a frequency in the range from 20 Hz to1 MHz. The sample was placed in a cylindrical container and placed inside the pick-up coil. A compensation coil is used to cancel out the background signal. The output signal is detected by a lock-in amplifier providing the in-phase and out-of-phase signals that correspond to the real (χ’) and imaginary part (χ”) of the susceptibility. Figures 5-3 through 5-7 show the results of the measurements that were performed for the fluid as well as for the immobilized sample. The data show that the fluid sample shows much higher values of χ’ and χ” and a more pronounced frequency dependence than the immobilized sample which is due to the Brownian rotation of the 146 particles in the liquid. In the fluid sample, χ” shows a distinct maximum at ~ 25 kHz -6 which corresponds to a Brownian relaxation time of τB = 6.4 ×10 seconds. Real part (fluid) 150 130 110 90 70 50 Susceptibility 30 10 -10 110 4 7 100 10 13 1K 16 19 2210K 25 28100K 31 34 1M 37 Frequency [Hz] Figure 5-3. ACS spectra of the real part of ZnNd-ferrite fluid sample. Imaginary part (fluid) 150 130 110 90 70 50 Susceptibility 30 10 -10 1 410 7 100 10 13 1K 16 1922 10K 25 28100K 31 34 1M 37 Frequency [Hz] Figure 5-4. ACS spectra of the imaginary part of ZnNd-ferrite fluid sample. 147 Real part (Immob.) 150 130 110 90 70 50 Susceptibility 30 10 -10 1 310 5 7100 9 1113151719212325272931333537 1K 10K 100K 1M Frequency [Hz] Figure 5-5. ACS spectra of the real part of ZnNd-ferrite immobile sample. Imaginary part (immob.) 150 130 110 90 70 50 Susceptibility 30 10 -10 1 3 10 5 7 100 9 11 13 1K 15 17 19 10K 21 23 100K 25 27 29 1M 31 33 35 37 Frequency [Hz] Figure 5-6. ACS spectra of the imaginary part of ZnNd-ferrite immobile sample. 148 ACS spectra of the ZnNd-ferrite fluid and the immobilized samples 150 130 Real part [fluid] 110 Imaginary part [fluid] 90 Real part (immob.) Imaginary part (immob.) 70 Susceptibility 50 30 10 -10 110 3 5 7 100 9 111315 1K 171921 10K 232527 100K 293133 3537 1M Frequency [Hz] Figure 5-7. ACS spectra of the ZnNd-ferrite fluid and the immobilized samples. The calorimetric measurement of the specific heating power was performed using the coil apparatus discussed in chapter 4. The fluid sample was placed into the center of the coil and the temperature T(t) was monitored using a thermocouple and an alcohol thermometer to minimize interference of the electronics. From these data the specific heating power is calculated by [169]: c m dT SHP = fluid fluid (6) m particle dt where cfluid is the specific heat and mfluid or mparticle are the masses of the fluid or the particles, respectively. The square symbols in figure 5-8 show the SHP data obtained from the fluid for different field amplitudes. 149 SHP data of ZnNd-ferrite fluid meausered calorimetrically (Squares) for different field amplitudes and field dependence according to Eq. 5. 1600 From ACS data 1400 Calorimetric (W/g) 1200 1000 800 600 Specific Loss Power Loss Specific 400 121 2 14 3 4 5 16 6 7 18 8 9 20 1011121314 22 24 Field (KA/m) Figure 5-8. Specific heat power data of the ZnNd-ferrite fluid measured by the coil (line) for different field amplitudes and field dependence according to eq. 5. With Eq. 5 and χ” (961 kHz) = 15 as measured for the fluid we obtain the curve in 5-8 which is in good agreement with the calorimetrical data. The χ” data of the immobilized sample are almost independent on the frequency within the range investigated. The absolute value is about 5. As this value corresponds to pure Neel relaxation, whereas the above-mentioned value of the liquid includes Brown and Neel relaxation we can estimate the contribution of the Brown and Neel mechanisms to the total specific heating power to 70% or 30%, respectively. 150 5.2 Power Calculation Model Ferromagnetic hyperthermia utilizes magnetic particles of metallic alloys of various sizes that are placed surgically or intravenously into tumors. Supplying power to a coil creates an electromagnetic field that produces eddy-current heating within the larger size magnetic particles, and relaxation mechanisms and hysteresis losses within the fine ones. Tissues near the particles are then heated via thermal conduction. The temperature reached by particles via eddy current heating is a function of several variables including the strength and frequency of the electromagnetic field, orientation of particles within the coil, proximity to other particles, permeability and electrical conductivity of the particles, and local blood perfusion rate and the thermo- physical properties. Self-regulating particles have the unique characteristic of heating to a temperature that is maintained within a few degrees. The ability of the particles to self-regulate is a consequence of their magnetic properties. Self-regulating particles are composed of various types of alloys. The Curie point can be made different for each particle type by altering the mass fraction of the diluents, e.g. Mn, Gd, Ce, Nd and Zn in alloys of Iron in the particles. The ability of particles to self-regulate is advantageous since it is based upon an intrinsic material property of the injected particles. However, it is difficult to alter the temperature of individual particles during a hyperthermia treatment since there is no physical contact with the particles. There is therefore a need to predict temperature distributions in advance. One step toward satisfying this need is developing a heat- transfer model of ferromagnetic particles. Thermal models of ferromagnetic particles have been developed previously, [l72, 173]. Some particles models have used a constant power modeling assumption [172-174]. Other thermal models employ a constant-temperature modeling assumption [173,175]. Still other thermal models incorporate the self-regulating feature of particles [175,176]. The constant temperature and self-regulating modeling assumptions have been shown to produce better tumor temperature distributions than those using the constant power 151 assumption [173,174]. Some studies have suggested that particles should incorporate the self-regulating feature in the thermal model and include the finite size of particles as well, [175,176]. The goal of this part of the study is to develop a thermal model that incorporates the finite size of the large particles and models the self regulating feature of them. Measurement of heating power using the area of the hystresis loss will be performed. A model for the calculation of the power dissipated from ferromagnetic nanoparticles is constructed utilizing a combination of the Curie-Wiess law and the electromagnetic induction law below Tc. The calculated values of heating powers will be compared with the experimental results. All hyperthermia treatments consist of a unique heat transferal system so that the target cells are heated while the surrounding cells are almost unaffected. The inducing of heat in magnetic materials is mainly attributed to three physical phenomenons: hysteresis, eddy current, and Néel or Brownian relaxation losses. The total loss in a conductor media is [177, 178]: Ptotal = Ph + Pe + Pr (7) Thus the total loss per unit frequency can be expressed as [177,178]: β µ H 2 µ 2 H 2 f β (mH ) 2 P = 01 + β 0 + 3 (8) total τ 2 ρ kTτ where βi denotes the geometry and first order constant coefficient, and 0 are the permeability of the magnetic material and of free space. H is the magnetic field intensity, ρ is electrical resistivity, τ is relaxation time and is exponentially size dependent, to the highest of 10-7 ~ 10-9 for iron [179]. kT is thermal energy and m is the particle moment. Experiments were properly designed to investigate the heating capacities of ferromagnetic materials. Hilger reported an energy deposition capability of hysteresis and Brownian relaxation loss of up to 93W/g of iron oxide (>100 nm in size) in a 6.5×103 A/m, 400 KHz field [180]. The theoretical prediction for the hysteresis loss of similar size Fe3O4 particles could be 10 KW/g (300 KHz and 100 KA/m). However, due to the multi-domain wall 152 existing in the larger grains and the limitation of field amplitude human tissue will stand, accepted experimental energy dissipation capability is less than 100 W/g [181]. Hergt calculated the peak Néel and Brownian relaxation losses of ferromagnetic particles to each are around 1×1010 W/m3 in a field amplitude of 6.5 KA/m and frequency of 2 MHz. Considering submicron size ferromagnetic conductor with an electrical resistivity of 10-5 ~ 10-6 or less, excited by MHz frequency range magnetic field, the eddy current loss will be dominant in the total energy dissipation. To calculate the heating power generated from the ferromagnetic nanoparticles, the magnetic hysteresis loops of each sample at different temperatures below the Tc should be measured. Direct measurements of magnetization M at different magnetic fields H and temperatures T was made using SQUID magnetometer at MARTECH. The area of the hysteresis loop of each sample is decreased with increasing the temperature. According to authors in [182], the heating power can be deduced from the area of the recorded hysteresis loop using: P = f ∫ MdH (9) where, P is the measured heating power dissipated of the sample under applied magnetic field, f is the assumed frequency of the applied AC magnetic field and ∫ MdH is the area of the DC hysteresis loop. The magnetization of the samples is temperature and field dependent. Through the hysteresis loop measurements, the magnetization M can be given by the linear equation M = ± σ + χ(T) H, where, σ is the remnant magnetization and χ(T) is the susceptibility of the sample. The negative sign represents the value of the remnant magnetization when the field is swept from the negative value back towards zero fields. The hystresis loops can be measured using maximum static magnetic field ± 0.8T for the selected samples. Figure 5-9 shows the hysteresis loops of ZnNdFe at temperatures below the Curie point. It can be seen from the figure that the remanent σ, the saturation magnetization Ms and the area of hysteresis loop were found to decrease with increasing temperature. The power loss decreases with increasing temperature. Similar behavior was found for the other prepared particles. 153 From figure 5-9, the area of the hysteresis curve is proportional to the energy dissipated in the form of power heating loss. The power heating loss decreases with decreasing the area of the hysteresis loop and stopped automatically when the Curie temperature is reached. ZnNdFe T=280C 3000 2000 1000 0 M(A/m) -1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1 -1000 -2000 -3000 H(tesla) Figure 5-9. Hysteresis loops of ZnNdFe at temperatures below the Curie point. 154 ZnNdFe T=370C 3000 2000 1000 0 -1 -0.5 0 0.5 1 M[A/m] -1000 -2000 -3000 H [tesla (x107/4π A/m)] ZnNdFe T=430C 3000 2000 1000 0 M(A/m) -1 -0.5 0 0.5 1 -1000 -2000 -3000 H [tesla(x107/4π A/m)] Figure 5-9. continued 155 The area of the hysteresis curve is proportional to the energy dissipated in the form of power heating loss. The power heating loss decreases with decreasing the area of the hysteresis loop and stops automatically when the Curie temperature is reached. The area under the hysteresis loop may be evaluated using a numerical analysis method [183-185] using the following relation: H n ∫M (H )dH = h ∫ f q dq 0 0 n q(q + )1 2 q(q + )(1 q + )2 3 = h fo + q∇fo + ∇ fo + ∇ fo ⋅⋅⋅⋅⋅⋅+ (10) !2 !3 0 where, h is the step of integration between each two points of H (here h=0.1), fq is the magnetization function of H, n is the integral terminator and q is constant near values of n. From Eq. 9, the calculated values of the heating power evaluated from the area of the hysteresis loops for sample as a function of temperature at any specified frequency may be plotted as shown in figure 5-10. It is clear from figure that, the heating power of sample decreases with increasing temperature. This leads to a disappearance of power produced from these particles near the Curie temperature. For this reason, mainly, ferromagnetic particles material is the best localized self-regulated temperature control system. 156 100 80 60 40 P (mw/cm) P 20 0 25 30 35 40 45 50 T (0C) Figure 5-10. Calculated values of heating power evaluated from hysteresis loop area for ZnNdFe particles as a function of temperature at frequency 961 KHz. Calculation of Heating Power In this section, a theoretical model for the calculation of heating power produced from any ferromagnetic nanoparticle material system placed under AC magnetic field is derived. When a high frequency magnetic field is applied on a ferromagnetic nanoparticle’s material system then, hysteresis loss occurs. This loss converts to energy dissipated in the form of heat [186]. From the electromagnetic induction law, the power loss per unit volume is given by: r 2 dM µ 2 P = o ( 2 .) o (11) 2ρ dt χ 2 where ro is the radius of the particle, o is the permeability of air, and ρ is the resistivity of nanoparticle. From Curie-Weiss law: C M = H (12) T −Tc 157 where, C is the Curie constant and Tc is the Curie temperature. Substitute Eq. 3, into Eq. 12: 2 2 µ 2 ro d C o P = ( )H 2 (13) 2ρ dt T − Tc χ From Eq.13, one can write for the maximum total power per unit volume (watt/m3), which is emitted from a ferromagnetic material placed in AC magnetic field: Ptotal = P(H) + P(T) (14) where, P(H) is the part of power at constant temperature, it is a function of H2 and P(T) 2 is the part of power at constant a.c magnetic field, it is a function of T . At H = Ho= constant, then Eq. 13, becomes: 2 2 2 2 r0 µ0 C H 0 1 dT 2 P T )( = 2 . 4 ( ) (15) 2ρχ (T − Tc ) dt By taking the square root of Eq. 15: r µoo CH o 1 dT P T)( = . 2 ( ) (15-a) 2 χρ (T − Tc ) dt By integrating Eq.15-a: T t −2 χ 2ρP T)( ∫ (T − Tc ) dT = ∫ dt (15-b) 0 oCr µo H o 0 Then, −1 χ 2ρP T )( (T − Tc ) = t (15-c) oCr µo H o 158 Squaring Eq. 15-c, we get: 2 2 2 2 µ 0 r0 C H 0 1 1 P(T ) = 2 . 2 . 2 (16) 2ρχ (T − Tc ) t The power P(T) at constant AC magnetic field through one cycle is given by: 2 2 2 2 2 f µ 0 r0 C H 0 1 P(T ) = 2 . 2 (17) 2ρχ (T − Tc ) At constant temperature T and since H=Ho cos ωt, then, from Eq. (12): dM C dH = . (18) dt T − Tc dt Cw = H o sin wt T − Tc The maximum value of dH/dt is given at ωt =90, then, dM C = H 0 w (19) dt T − Tc Substituting Eq. 19, into Eq. 11, the power at constant temperature is given by: 2 2 2 2 2 r0 µ0 C 4π f 2 P(H ) = . 2 2 H 0 (20) 2ρ χ (T − Tc ) Substitute Eq. 17 and Eq. 20 in Eq. 14, one gets for the total power: 2 2 2 2 2 2 2 2 r0 µ0 C 4π f 2 µ0 C f 2 Ptotal = 2 2 H 0 + 2 H 0 2ρχ (T − Tc ) (T − Tc ) 2 2 2 2 r0 2 µ0 C f 2 Ptotal = 2 4( π + )1 2 H 0 2ρχ (T − Tc ) 2 2 2 2 2 40r0 µ0 C f H 0 1 Ptotal ≅ 2 . 2 2ρχ (T − Tc ) 159 2 Multiply the above Equation with the area of the seed (π ro ), the power per unit length becomes: 4 2 2 2 40πr0 µ0 f H0 1 Ptotal ≅ 2 2 . 2χ ργ T 2 ( − )1 Tc (21-a) where, γ is the intermolecular field constant = Tc / C and Ho is the amplitude of the AC -7 3 magnetic field in A/m. For: µ0= 4 x 10 , γ= 0.2045 x 10 , equation 21(a) becomes: 2 4 2 2 ( 2.37254e −15 )r0 f H 0 Tc Ptotal ≅ 2 (21-b) 2χ ρ T − Tc Equation 21 represents the total heating power emitted per unit length of the ferromagnetic seed. The theoretical calculation of the temperature dependence of the heating power for ZnNdFe particles at different frequencies is shown in figure 5-11. It can be seen that the power reaches minimum value near Tc and also at 961 KHz, the heating power is about 200 mW/cm for ZnNdFe which is sufficient for most clinical applications [10]. Figure 5-12 shows the theoretical line representing the power calculated from Eq. 21 together with the experimental points which are recorded from the hysteresios loops and Eq. 9 (figure 5-10) for ZnNdFe particles at 961 KHz. 160 350 f=200f = 961 kHz KHz 300 CuNi 3 70.4 250 γ=0.2045*10 f = 433 KHz 200 f=150 kHz 150 P(mw/cm) 100 ff=100 = 100 KHzkHz 50 f=50f = 50 kHzKHz 0 f=10 kHz P (mW/cm) (mW/cm) P 10 20 30 40 50 60 o T(T(oC)C) Figure 5-11. Theoretical calculations of temperature dependence of heating power for ZnNdFe particles at different frequencies. P d C o 2 2 0 1 0 .8 2 0 0 1 8 0 Experimental 1 6 0 Theoretical 1 4 0 1 2 0 1 0 0 f =961 kHz 8 0 P(mw/cm) P (mW/cm) P 6 0 4 0 2 0 0 10 15 20 25 30 35 40 45 50 55 60 65 70 o T(T(C)T(oC) o C) Figure 5-12. Theoretical line representing power calculated from Eq.21 with experimental points recorded from hysteresis loops and Eq. 9 for ZnNdFe particles at 961 KHz. 161 CHAPTER 6 CONCLUSIONS AND FUTURE WORK It is a fascinating concept that tumor cells could be loaded with magnetic particles, which would become activated only by a specific signal, yielding the death of all particle containing cells as soon as an AC magnetic field is applied Many materials have been studied to synthesize and to control their magnetic properties, mainly due to particle size and Curie temperature. Four magnetic particles systems, namely, MnZn-ferrite, ZnGd-ferrite, GdZnCe-ferrite and ZnNd-ferrite were synthesized in situ and their possible use as nanoparticles for magnetic hyperthermia was investigated. Different characterization techniques were pursued to characterize the newly fabricated particles, including XRD, TEM, SQUID and magnetic heating techniques. Nanoparticles were also, encapsulated with polymers. Magnetic heating apparatus was set up to ensure that the nanoparticles heat up in an AC magnetic field and also to observe their self control heating once they reach their Curie temperature. When superparamagnetic particles are used for localized hyperthermia therapy, only a weak magnetic field is needed to generate heat compared to the strong field required for ferromagnetic particles. The superparamagnetic particles are economical, harmless to the living body, and appropriate for localized hyperthermia therapy. However, in order to have superparamgnetic properties the particles must have nanodomains of 10-20 nm in diameter and must also be coated by organic or inorganic compounds that prevent any interaction between the magnetic particles. Magnetization measurements on purified samples were performed in a SQUID. The magnetization isotherms at room temperature show no coercivity and no remanent magnetization. This is characteristic of superparamagnetic behavior which is in good agreement with the nanometric size of the magnetic particles derived from TEM images 162 and XRD by use of Scherrer method. Therefore we can conclude that the synthesized magnetic nanoparticles fulfill all the requirements to be used as magnetic nanoparticles for future in vivo magnetic hyperthermia applications. The experimental results can be compared to those predicted by the theory of Langevin for superparamagnetism. In all cases, superparamagnetic particles are of interest because they do not retain any magnetism after removal of magnetic field. The effectiveness of the particles depends upon: (I) high magnetic susceptibility for an effective magnetic enrichment. (II) size of particles which should be monosized in the range of 6–15 nm -particles below a critical particle size <15nm- would consist of a single magnetic domain, i.e. a particle that is a state of uniform magnetization at any field with superparamagnetism and high saturation magnetization values. The particles in this size range are rapidly removed through extravasations and renal clearance. (III) superparamagnetic behaviour. (IV) tailored surface chemistry for specific biomedical applications. In X-ray diffraction patterns for all the synthesized ferrites, no other crystalline structures are identified than the ferrite spinel one. However, there is a pronounced noise signal, which may indicate to an X-ray amorphous component present in the particles. At least two reasons may be responsible for that: I) non-complete transformation of hydroxides into ferrite, that is confirmed by the presence of 4-14 w% of associated water in particles -complete elimination of associated water from particles is achieved only by a high temperature heating up to 750 °C. II) possible inhomogeneties in crystalline structure of particles, which may be pronounced in our case of three metals ferrite. Zn substitution degree x has a great influence on magnetic properties of particles. Synthesized samples with x > 0.3 do not reach the magnetic saturation in Hm=20 kOe. Nanoparticles with x ≥ 0.5 (mean size of 6 - 10 nm) are found to be far from magnetic saturation even in a very strong magnetic field due to a significant paramagnetic contribution . This prevents a correct determination of their saturation magnetisation. However, for samples with lower Zn content at elevated temperatures spectra of magnetic moments become with one maximum. 163 Particles of the samples S1 and S4 have a lower Fe content and higher Zn substitution degree x than particles of the samples S3 and S5. S1 and S4 has a lower specific magnetisation than the S3 and S5. Especially this difference is pronounced in low fields. In comparison with the co-precipitated particles separated by sedimentation centrifugation leads to a better separation in size of particles. Properties of particles (chemical composition, content of associated water, magnetic properties) are dependent on size of particles. Smaller particles have higher content of Fe and a higher content of associated water than the bigger ones. Smaller particles possess a lower specific magnetization and lower thermomagnetic coefficients than the bigger ones in all the range of magnetic field of 0 - 20 kOe. In low fields difference in the magnetization for smaller and bigger ones is relatively more pronounced. With regard to the role of Zn substitution degree x, growing of x leads also to higher content of associated water in particles. This may be connected with smaller size of particles with higher x: smaller particles have more developed surface, which may adsorb a higher quantity of water. This suggestion is also confirmed by a higher associated water content present in smaller particles fractionated from the polydispersed samples. Other reason for increased associated water content in particles with higher x may be a higher affinity of remained Zn hydroxide to water in comparison to Mn hydroxide. Three bases give particles with size decreasing in the following sequence: DNaOH>DCH3NH3OH>DNH4OH. This may be connected with a different pH values, which are provided by these bases. Distinguishing feature of Mn-Zn ferrite is that for CH3NH3OH and NH4OH the increase of the initial Zn substitution degree x0 above the definite threshold values leads to formation of a non-magnetic X-ray amorphous product. It is remarkable also that for ferrite particles obtained by CH3NH3OH below this threshold value a deficit in Zn content is found in respect to the initial reagent mixture. These phenomena may be explained by the soluble complex formation for Zn2+ 2+ 3+ and partly Mn with CH3NH3OH or NH4OH. Fe is not complexing with these bases and Fe (III) hydroxide may to some extent initiate the precipitating of Mn2+ and Zn2+. To widen x limits where ferrite is formed with CH3NH3OH or NH4OH nonstoichiometric initial compositions with the excess of Fe3+ may be proposed for further studies. 164 For particles obtained by NaOH a dramatic decrease of magnetisation, growing of paramagnetic part in magnetic moment spectra and a decrease of thermomagnetic coefficient kT was observed for x growing from 0.5 to 0.6. The process of the ferrite formation requires a certain time. It is found that the most relevant changes occur during the first 20 min of heating in alkaline solution. It has been noted that particles contain ~10 w% of associated water even after 90-180 min of heating. Further boiling failed to reduce it. This long heating process, however, may lead to the growing of particles size and requires a special reaction vessel with reversed cooler to prevent evaporation of the base and decreasing the pH of the suspension. Some additional treatment may accelerate the ferrite formation and may lead to a more perfect crystalline structure of particles. Physical motion of particles under applied magnetic field and clustering on the blocking temperature and coercivity at various temperatures was studied. One important observation is that, the particles in a polymer matrix form clusters of various size ranges. Loosely packed particles, free to move about, show no coercivity even below the blocking temperature, due to the instant rotation of the entire particle in response to the switching field. In Conclusion, MnZn-ferrite, ZnGd-ferrite, GdCeZn-ferrite and ZnNd-ferrite superparamagnetic and ferromagnetic particles have been prepared for interstitial hyperthermia in cancer treatment. The MnZn-ferrite, ZnGd-ferrite, GdCeZn-ferrite and ZnNd-ferrite show sharp superparamagnetic behavior with preferred Curie temperature ~ 430C. These particles provided power output at room temperature as measured from the hystersis loops, and loss processes which are in good agreement with the values measured by the calorimetric method. Temperature dependence of particles power has been computed from a combination of curie and induction laws. Computations show good agreement with those calculated from the area under hystersis loop at the AC magnetic field. It is also clear that the power is automatically stopped when Tc is reached, indicating that the particles have self limiting temperature control. Therefore, the above mentioned particles are clinically useful in treating localized tumors pending some toxicity studies to ensure their suitability. 165 It is an exciting challenge for future research to increase the biological efficacy and particle specific absorption rate in order to achieve advanced magnetic particle hyperthermia of which only a few particles are required for selective tumor cell inactivation. 166 APPENDIX (A) Table of Conversions SI Gaussian Quantity Symbol Relation Units Units magnetic flux density, B Tesla (T) Gauss (G) 1T = 104G magnetic induction 8 magnetic flux Weber (Wb) Maxwell (Mx) 1Wb = 10 Mx Volt-second (Vs) magnetomotive force, Fm Ampere (A) Gilbert (Gb) 1A = 4 /10Gb magnetic potential magnetic field strength H A/m Oersted (0e) 1A/m = 4 /10³ magnetic polarization, J,I T emu/cm² 1T = 104/4 2 intensity of magnetization Wb/m (mass) magnetization ,M A×m²/kg emu/g 1A×m²/kg = 7 Wb×m/kg 10 /4 ×emu/g magnetic moment m A×m², emu, 1A×m² = J/T erg/G 10³emu Inductance L Henry (H) sec²/cm 1H = 9 10 /c²×sec²/cm permeability µ H/m, dimensionless 1H/m = 107/4 Wb/(A/m) 7 energy E Joule (J) erg 1J = 10 erg (Volume) energy W J/m³ erg/cm³ 1J/m³ = density, 10erg/cm³ energy product (Courtesy: http://www.drusch.com/help/qmcotb.shtml.en) 167 REFERENCES [1] Cavaliere, R., Ciocatta, E. C., Giovanella, B. C, Heidelberger, C., Johnson, R. D., Margottini, M., Mondovi, B., Moricca, G. and Rossi-Fennelli, A., 1967, Cancer, 20, pp. 1351. [2] Hand, J. W., and ter Haar, G., 1981, Br. J. Radiol., 54, pp. 443. [3] Shinkai, M., 2000, J. Biosci. Bioeng., 94 (6), pp. 606. [4] Cavalier, R., Ciocatto, E. C., Giovanel, B., Heidelbe, C., Johnson, R. O., Margotti, M., Mondovi, B., Moricca, G., Rossifan, A., 1967, “Selective heat sensitivity of cancer cells-biochemical and clinical studies,” Cancer, 20(9), pp.1351. [5] Stauffer, P. R., Cetas, T. C., Fletcher, A. M., Deyoung, D. W., Dewhirst, M. W., Oleson, J. R., Roemer, R. B., 1984, “Observations on the use of ferromagnetic implants for inducing hyperthermia,” IEEE Trans. Biomed. Eng., 31(1), pp.76-90. [6] Lin, J. C., Wang, Y. J., 1987, “Interstitial microwave antennas for thermal therapy,” Int. J. Hyperthermia, 3(1), pp. 37-47. [7] Ikeda, N., Hayashida, O., Kameda, H., Ito, H., Matsuda, T., 1994,”Experimental study on thermal damage to dog normal brain,” Int. J. Hyperthermia, 10(4), pp. 553-561. [8] Van der Zee, J., 2002,”Heating the patient: a promising approach?,” Ann. Oncol., 13, pp. 1173-84. [9] Wust, P., Hildebrandt, B., Sreenivasa, G., Rau, B., Gellermann, J., Riess, H., Felix, R. and Schlag, P. M., 2002, “Hyperthermia in combined treatment of cancer,” Lancet Oncol., 3, pp. 487-97. [10] Moroz P, Jones S K and Gray B N 2001 Status of hyperthermia in the treatment of advanced liver cancer J. Surg. Oncol. 77 259–69 [11] Gilchrist, R. K., Medal, R., Shorey, W. D., Hanselman, R. C., Parrott, J. C. and Taylor, C. B., 1957, “Selective inductive heating of lymph nodes,” Ann. Surg.,146, pp. 596–606. 168 [12] Mosso, J. A., and Rand, R. W., 1972, “Ferromagnetic silicone vascular occlusion,” Ann. Surg., pp. 663–8. [13] Rand, R. W., Snyder, M., Elliott, D. G., and Snow, H. D., 1976, “Selective radiofrequency heating of ferrosilicone occluded tissue: a preliminary report,” Bull. Los Angeles Neurol. Soc., 41, pp. 154–9. [14] Gordon, R. T., Hines, J. R., and Gordon, D., 1979, “Intracellular hyperthermia: a biophysical approach to cancer treatment via intracellular temperature and biophysical alterations,” Med. Hypotheses, 5, pp. 83–102. [15] Rand, R. W., Snow, H. D., Elliott, D. G., and Snyder, M., 1981, “Thermomagnetic surgery for cancer,” Appl. Biochem. Biotechnol., 6, pp. 265–72. [16] Borrelli, N. F., Luderer, A. A., and Panzarino, J. N., 1984, “Hysteresis heating for the treatment of tumors,” Phys. Med. Biol., 29, pp. 487–94. [17] Hase, M., Sako, M., and Hirota, S., 1990, “Experimental study of ferromagnetic induction heating combined with hepatic arterial embolization of liver tumours,” Nippon- Igaku-Hoshasen-Gakkai-Zasshi, 50, pp. 1402–14. [18] Suzuki, S., Arai, K., Koike, T., and Oguchi, K., 1990, “Studies on liposomal ferromagnetic particles and a technique of high frequency inductive heating—in vivo studies of rabbits,” J. Japan. Soc. Cancer Therapy, 25, pp. 2649–58. [19] Chan, D. C. F., Kirpotin, D. B., and Bunn, P. A. Jr., 1993, “Synthesis and evaluation of colloidal magnetic iron oxides for the site-specific radiofrequency-induced hyperthermia of cancer,” J. Magn. Magn. Mater., 122, pp. 374–8. [20] Matsuki, H., Yanada, T., Sato, T., Murakami, K., and Minakawa, S., 1994, “Temperature sensitive amorphous magnetic flakes for intratissue hyperthermia,” Mater. Sci. Eng., A181/A182, pp. 1366–8. [21] Mitsumori, M., et al, 1994, “Development of intra-arterial hyperthermia using a dextran–magnetite complex,” Int. J. Hyperthermia, 10, pp. 785–93. [22] Suzuki, M., Shinkai, M., Kamihira, M., and Kobayashi, T., 1995, “Preparation and characteristics of magnetite-labelled antibody with the use of poly(ethylene glycol) derivatives,” Biotechnol. Appl. Biochem., 21, pp. 335–45. [23] Mitsumori, M., Hiraoka, M., Shibata, T., Okuno, Y., Nagata, Y., Nishimura, Y., Abe, M., Hasegawa, M., Nagae, H., and Ebisawa, Y., 1996, “Targeted hyperthermia using dextran magnetite complex: a new treatment modality for liver tumours,” Hepato- Gastroenterology, 43, pp. 1431–7. 169 [24] Jordan, A., Scholz, R., Wust, P., Fahling, H., Krause, J., Wlodarczyk, W., Sander, B., Vogl, T., and Felix, R., 1997, “Effects of magnetic fluid hyperthermia (MFH) on C3H mammary carcinoma,” in vivo Int. J. Hyperthermia, 13, pp. 587–605. [25] Shinkai, M., Yanase, M., Suzuki, M., Honda, H., Wakabayashi, T., Yoshida, J., and Kobayashi, T., 1999, “Intracellular hyperthermia for cancer using magnetite cationic liposomes,” J. Magn. Magn. Mater., 194, pp. 176–84. [26] Jordan, A., Scholz, R., Wust, P., Fahling, H., and Felix, R., 1999, “Magnetic fluid hyperthermia (MFH): cancer treatment with AC magnetic field induced excitation of biocompatible superparamagnetic nanoparticles,” J. Magn. Magn. Mater., 201, pp. 413– 19. [27] Minamimura, T., Sato, H., Kasaoka, S., Saito, T., Ishizawa, S., Takemori, S., Tazawa, K., and Tsukada, K., 2000, “Tumour regression by inductive hyperthermia combined with hepatic embolization using dextran magnetite incorporated microspheres in rats,” Int. J. Oncol., 16, pp. 1153–8. [28] Moroz, P., Jones, S. K., Winter, J., and Gray, B. N., 2001, “Targeting liver tumors with hyperthermia: ferromagnetic embolization in a rabbit liver tumor model,” J. Surg. Oncol., 78, pp. 22–9. [29] Jones, S. K., Winter, J. W., and Gray, B. N., 2002, “Treatment of experimental rabbit liver tumours by selectively targeted hyperthermia,” Int. J. Hyperthermia, 18, pp. 117–28. [30] Hilger, I., Fruhauf, K., Andra, W., Hiergeist, R., Hergt, R., and Kaiser, W. A., 2002, “Heating potential of iron oxides for therapeutic purposes in interventional radiology,” Acad. Radiol., 9 , pp. 198–202. [31] Gordon, R. T., Hines, J. R., Gordon, D., 1979, “Intracellular hyperthermia: a biophysical approach to cancer treatment via intracellular temperature and biophysical alterations,” Med Hypotheses, 5(1), pp. 83–102. [32] Rosensweig, R. E., 2002, “Heating magnetic fluid with alternating magnetic field,” J Magnet. Magn. Mater., 252(1–3), pp. 370–374. [33] Hergt, R., Andra, W., d’Ambly, C. G., Hilger, I., Kaiser, W. A., Richter, U., Schmidt , H. G., 1998, “Physical limits of hyperthermia using magnetite fine particles,” IEEE Trans. Magn., 34(5), pp. 3745–3754. [34] Moroz, P., Jones, S. K., Gray, B. N., 2002, “Magnetically mediated hyperthermia: current status and future directions,” Int. J. Hyperthermia, 18(4), pp. 267–284. [35] Jordan, A., Scholz, R., Maier-Hauff, K., Johannsen, M., Wust, P., Nadobny, J., Schirra, H., Schmidt, H., Deger, S., Loening, S., Lanksch, W., Felix, R., 2001, 170 “Presentation of a new magnetic field therapy system for the treatment of human solid tumors with magnetic fluid hyperthermia,” J. Magnet. Magn. Mater., 225(1–2), pp. 118- 126. [36] Overgaad, J., “History and heritage. An introduction,” In: Overgaad J. (ed) Hyperthermic Oncology, Taylor and Francis, London, 2, pp. 3-8. [37] Suit, H. D., Gerweck, L. E., 1979, “Potential for hyperthermia and radiation therapy,” Cancer Res., 39, pp. 2290-2298. [38] Moroz, P., Jones, S. K., and Gray, B. N., 2001, “Status of hyperthermia in the treatment of cancer,” J. Surg. Oncol., 77, pp. 259–269. [39] Gel’vich, E. A., and Mazokhin, V. N., 2001, “Simultaneous EM Whole-Body and Local Hyperthermia - an Effective Adjuvant in Curing of Malignant Tumours,” Crit. Rev. Biomed. Eng., 29, pp. 77–97. [40] Moroz, P., Jones, S. K., and Gray, B. N., 2002, “Magnetically mediated hyperthermia: current status and future directions,” Int. J. Hyperthermia, 18 (4), pp. 267- 284. [41] Ohno, T., Wakabayashi, T., Takemura, A., Yoshida, J., Ito, A., Shinkai, M., Honda, H., and Kobayashi, T., 2002, “Augmentation of MHC,” J. Neuro-Oncol., 56, pp. 233- 239. [42] Jordan, A., Scholz, R., Wust, P., Fa¨ hling, H., and Felix, R., 1999, “Magnetic Fluid Hyperthermia (MFH): Cancer treatment with AC magnetic field induced excitation of biocompatible superparamagnetic nanoparticles,” J. Magn. Magn. Mater., 201, pp. 413- 419. [43] Morrish, A. H., 2001, The Physical Principles of Magnetism, New York: IEEE Press. [44] Pankhurst, Q. A., Connolly, J., Jones, S. K., Dobson, J., 2003, “Applications of magnetic nanoparticles in biomedicine,” J. Phys. D: Appl Phy., 36: pp. R167-181. [45] Shinkai, M., 2002, “Functional magnetic particles for medical application,” J. Bioscience and Bioeng., 94(6), pp. 606-613. [46] Ma, M., Wu, Y., Zhou, J., Sun, Y., Zhang, Y., Gu, N., 2004, J. Magn. Magn. Mater., 268, pp. 33. [47] Jiles, D., 1991, Introduction to Magnetism and Magnetic Materials, Chapman and Hall, London. 171 [48] Jagadish, B. R, Prabhu, N., Bahadur, D., 1998, “Studies on synthesis of calcium ferrite-based bio glass ceramics,” Bull. Mater. Sci., 21, pp.133–137. [49] Jianhua, C., Naru,Y. , 2001, J. Chin. Ceram. Soc., 29, pp. 238–244. [50] Kawashita, M., Takaoka, H., Kokuls, T.,Yao, T., Hamada, S., 2001, J. Ceram. Soc. Jpn., 109, pp. 39-44. [51] Kokubo, T., 1999, “Apatite formation on surfaces of ceramics, metals and polymers in body environment,” Mater. Sci. Forum, 293, pp. 65-82. [52] Ohura, K., Lkenaga, M., Nakamura, T., Yamamuro, T., Ebisawa, Y., Kokuko, T., Kotoura, Y., Oka, M., 1990, Bioceramics, (Proc. 3rd Int. Symp. Ceramics in Medicine) Terre Haute, Indiana, pp. 225–229. [53] Brezovich, I. A., Meredith, R. F., 1989, “Practical aspects of ferromagnetic thermoseed hyperthermia,” Radiol. Clin. North Am., 27 (1), pp. 589-602. [54] Cetas, T. C., Gross, E. J., Contractor,Y., 1998, “A ferrite core/metallic sheath thermoseed for interstitial thermal therapies,” IEEE Trans. Biomed. Eng., 45, pp. 68–77. [55] Jordan, A., Scholz, R., Wust, P., F¨ahling, H., Felix, R., 1999, “Magnetic Fluid Hyperthermia (MFH): Cancer treatment with AC magnetic field induced excitation of biocompatible superparamagnetic nanoparticles,” J. Magn. Magn. Mater. 201, pp. 413- 419. [56] Medal, B. S., Shorey, W., Gilchrist, R. K., Barker,W., Hanselman, R., 1959, Am. Med. Assoc. Arch. Surg., 79, pp. 427-431. [57] Gilchrist, R. K., Shorey, W. D., Hanselman, R. C., Depeyster, F. A., Yang, J., Medal, R., 1965, Ann. Surg., 161, pp. 890-896. [58] Gordon, R. T., Hines, J. R., Gordon, D., 1997,“ A biophysical approach to cancer treatment via intracellular temperature and biophysical alteration,” Med. Hypothesis, 5, pp. 83-102. [59] Chan, D. C. F., Kirpotin, D. B., Bunn, P. A. Jr., 1997, In: Scientific and clinical applications of magnetic carriers (eds), H¨afeli, U., Sch¨utt, W., Teller, J., Zborowski, M., (New York, London: Plenum), p. 607. [60] Streffer, C., Van Beuningen, D., 1987, In: “Hyperthermia and the therapy of malignant tumors,” (ed.) , Streffer, J., Berlin: Springer, p. 24-70. [61] Mosbach, K., Anderson, L., 1997, Nature, 270, p. 259. 172 [62] Jordan, A., Scholz, R., Maier-Hauff, K., Johannsen, M., 2001, “Presentation of a new magnetic field therapy system for the treatment of human solid tumors with magnetic fluid hyperthermia,” J. Magn. Magn. Mater., 225, pp. 118–126. [63] Bertotti G 1998 Hysteresis in Magnetism, London: Academic. [64] Kronm¨uller, H., and F¨ahnle, M., 2003, Micromagnetism and the Microstructure of Ferromagnetic Solids, Cambridge, Cambridge University Press. [65] Stoner, E. C., and Wohlfarth, E. P., 1948, “A mechanism of magnetic hysteresis in heterogeneous alloys,” Phil. Trans. R. Soc.A, 240, pp. 599-642. [66] Aharoni, A., 1996, Introduction to the Theory of Ferromagnetism, Oxford, Clarendon. [67] Eagle, D. F., and Mallinson, J. C., 1967, “On the coercivity of γ -Fe2O3 particles,” J. Appl. Phys., 38, pp. 995-7. [68] Heider, F., Dunlop, D. J., and Sugiura, N., 1987, “Magnetic properties of hydrothermally recrystallized magnetite crystals,” Science, 236, pp. 1287-90. [69] Hergt, R., Andr¨a, W., d´Ambly, C. G., Hilger, I., Kaiser, W. A., Richter, U., and Schmidt, H. G., 1998, “Physical limits of hyperthermia using magnetite fine particles,” IEEE Trans. Magn., 34, pp. 3745-54. [70] Jordan, A., Scholz, R., Wust, P., Fahling, H., and Felix, R., 1999, “Magnetic fluid hyperthermia (MFH): cancer treatment with AC magnetic field induced excitation of biocompatible superparamagnetic nanoparticles,” J. Magn. Magn. Mater., 201, pp. 413- 19. [71] Jordan, A., Wust, P., Fahling, H., Johns, W., Hinz, A., and Felix, R., 1993, “Inductive heating of ferrimagnetic particles and magnetic fluids: physical evaluation of their potential for hyperthermia,” Int. J. Hyperthermia, 9, pp. 51-68. [72] Jordan, A., et al, 2001, “Presentation of a new magnetic field therapy system for the treatment of human solid tumours with magnetic fluid hyperthermia,” J. Magn. Magn. Mater., 225, pp. 118-26. [73] Debye, P., 1929, Polar Molecules, New York, Dover. [74] Shliomis, M. I., and Stepanov, V. I., 1994, “Theory of dynamic susceptibility of magnetic fluids,” Relaxation Phenomena in Condensed Matter, ed. W. Coffey, Chichester, Wiley. 173 [75] Delaunay, L., Neveu, S., Noyel, G., and Monin, J., 1995, “A new spectrometric method using magneto-optical effect to study magnetic liquids,” J. Magn. Magn. Mater., 149, pp. L239-49. [76] Landau, L. D., and Lifshitz, E. M., 1960, Electrodynamics of Continuous Media, London, Pergamon. [77] Chan, D. C. F., Kirpotin. D. B., and Bunn, P. A. Jr., 1993, “Synthesis and evaluation of colloidal magnetic iron oxides for the site-specific radiofrequency-induced hyperthermia of cancer,” J. Magn. Magn. Mater., 122, pp. 374-8. [78] Fannin, P. C., and Charles, S. W., 1991, “Measurement of the Neel relaxation of magnetic particles in the frequency range 1 kHz–160MHz,” J. Phys. D: Appl. Phys., 24, pp. 76-7. [79] Fannin, P. C., Scaife, B. K. P., and Charles, S. W., 1993, “Relaxation and resonance in ferrofluids,” J. Magn. Magn. Mater.,122, pp.159-63. [80] Hanson, M., 1991, “The frequency dependence of the complex susceptibility of magnetic liquids,” J. Magn. Magn. Mater.,96, pp.105-13. [81] Jones, S. K., Gray, B. N., Burton, M. A., Codde, J. P., and Street, R., 1992, “Evaluation of ferromagnetic materials for low frequency hysteresis heating of tumours,” Phys. Med. Biol.. 37, pp. 293-9. [82] Maehara, T., Konishi, K., Kamimori, T., Aono, H., Naohara, T., Kikkawa, H., Watanabe, Y., and Kawachi, K., 2002, “Heating of ferrite powder by an AC magnetic field for local hyperthermia,” Japan. J. Appl. Phys., 41, pp. 1620-1. [83] Kirpotin, D., Chan, D. C. F., and Bunn, P. A., 1995, “Magnetic microparticles,” US Patents vol 5,411,730, Tucson, AZ, USA, Research Corporation Technologies, Inc. [84] Hergt, R., Andra, W., d’Ambly, C., Hilger, I., Kaiser, W., Richter, U., and Schmidt , H., 1998, “Physical limits of hyperthermia using magnetite fine particles,” IEEE Trans. Magn., 34, pp. 3745-54. [85] Stauffer, P. R., Cetas, T. C., Jones, R. C.,1984, “Magnetic induction heating of ferromagnetic implant for inducing localized hyperthermia in deep-seated tumors,” IEEE Trans. Biomed. Eng., 31, pp. 76-90. [86] Kittel, C., 1946, “Theory of the structure of ferromagnetic domains in films and small particles,” Phys. Rev., 70, pp. 965-971. [87] Landau, L., D., and Lifshitz, E., M., 1960, Electrodynamics of Continuous Media, Pergamon, London, U.K. 174 [88] Simonds, J., L., K. G. Ashar, 1995, Physics Today, Magnetic Disk Drive Technology: Heads, Media, Channel, Interfaces, and Integration, Institute of Electrical and Electronics Engineers, New York, 48, p. 26. [89] Chouly, C., Pouliquen, D., Lucet, I., Jeune, J., J., Jallet, P., J., 1996, “Development of superparamagnetic nanoparticles for MRI: Effect of particle size, charge and surface nature on biodistribution,” Microencapsulation, 13, pp. 245-255. [90] Raj, K., Rosenzweig, R., E., Aziz, L., M., 1998, In European Patent No. 0,859,379. [91] Rocchiccioli-Deltcheff, C., Franck, R., Cabuil, V., Massart, R., 1987, Journal of Chemical Research, Synopses, 5, pp.126-127. [92] Morais, P., C., DaSilva, S., W., Soler, M., A., G., Buske, N., 2000, “Ramanspectroscopy in oleoylsarcosine-coated magnetic fluids: a surface grafting investigation,” IEEE Trans. Magn., 36, pp. 3712-3714. [93] Lesnikovich, A., I., Shunkevich, T., M., Naumenko, V.. N., Vorobyova, S., A., Baykov, M., V., 1990,“Dispersity of magnetite in magnetic liquids and the interaction with a surfactant,” J. Magn. Magn. Mater., 85, pp. 14-16. [94] Moeser, G., D., Roach, K., A., Green, W., H., Laibinis, P., E., Hatton, T., A., 2002, Ind. Eng. Chem. Res.,41, pp. 4739-4749. [95] Mendenhall, G., D., Geng, Y., Hwang, J., 1996, “Optimization of Long-Term Stability of Magnetic Fluids from Magnetite and Synthetic Polyelectrolytes,” Journal of Colloid and Interface Science, 184, pp. 519-526. [96] Massart, R., 1981, “Preparation of aqueous magnetic liquids in alkaline and acidic media,” IEEE Trans. Magn., MAG-17, pp.1247-1248. [97] Papell, S. S.,1965, U.S. Patent no. 3, 215, 572. [98] Rosensweig, R. E., and Kaiser, R.,1967, NTIS Rep. No. NASW-1219. [99] Rosensweig, R. E.,1985, Ferrohydrodynamics, Cambridge Univ. Press, Cambridge, London, republished by Dover.Publ.Inc., New York ,1997. [100] Berkovski, B., and Bashtovoy, V., (ed.), 1996, Magnetic Fluids and Applications Handbook , Begell House, Wallingford. [101] Blakemore, R., 1975, “Magnetotactic bacteria,” Science, 190, pp. 377-379. [102] Lilly, M. B., Brezovich, I. A., Atkinson, W. J., 1985,“Hyperthermia induction with thermally self-regulated ferromagnetic implants,” Radiology, 154, pp. 243-244. 175 [103] Brezovich, I.,A., and Meredith, R.,F.,1989, “Practical aspects of ferromagnetic thermoseed hyperthermia,” Radiol. Clin. North Am., 27, p. 589. [104] Kobayashi, T., Kida, Y., Ohta, M., Tanaka, T., Kageyama, N., Kobayashi, H., Amemiya, Y., 1986, “Magnetic induction hyperthermia for brain tumor using ferromagnetic implant with low curie temperature. Effect on intracutaneously implanted brain tumor,” Neurol.Med.Chir.(Tokyo), 26, pp. 116-121. [105] Sato, T., 1993, “The development of anticancer agent releasing microcapsule made of ferromagnetic amorphous flakes for intratissue hyperthermia,” IEEE Trans. Magnetics, 29, pp. 3325-3330. [106] Yanada, T., Matsuki, H., Takahashi, M., Satoh, T., Minakawa, S., Kikuchi, S., Murakami, K., 1991,“evaluation of temperature sensitive amorphous metal flakes for self-regulated hyperthermia,” IEEE Trans. Magnetics, 27, pp. 5390-5392. [107] Hand, J., W., 1987, “Heat delivery and thermometry in clinical hyperthermia,” Ret. Res. In Cancer Res., 104 , pp. 1-23. [108] Chen, J., S., Poirier, D., R., Damento, M., A., Demer, L., J., Biancaniello, F., and Cetas, T., C., 1988,“Development of Ni-4 wt% Si thermoseeds for hyperthermia cancer treatment,” Inr. Biomed. Morer. Res., 22, pp. 302-303. [109] Brezovich, I., A., Lilly, M., B.,, Meridith, R., F., Weppelmann, B., Henderson, R., A., Brawner, Jr., W., and Salter, M., M.,1990, “ Hyperthermia of pet animal tumors with self regulating ferromagnetic thermaseeds,” Inr. J. Hyperthermia, 6 , pp. 117-130. [110] Matsuki, H., Satah, T., Murakami, K., Hashino, T., Yanada, T., and Kikuchi, S., 1990, “Local hyperthermia based on soft heating method utilizing temperature- sensitive ferrite rod,” IEEE Trans. Mag., 26, pp. 1551-1554. [111]Hiemenz, P. C., Rajagopalan, R., 1997, Principles of Colloid and Surface Chemistry, editor, New York, Marcel Dekker, Inc. [112] Kim, D. K., Zhang, Y., Voit, W., Rao, K. V., Muhammed, M., 2001a, “Synthesis and characterization of surfactant-coated superparamagnetic monodispersed iron oxide nanoparticles,” J. Magnetism and Magnetic Materials, 225, pp. 30-36. [113] Massart, R., 1981,“Preparation of aqueous magnetic liquids in alkaline and acidic media,” IEEE Transactions on Magnetics, 17(2), pp. 1247-48. [114] Shalafalla, E. K., Reimers G. W., 1980, IEEE Trans Magn. 16. [115] Tourinho, F. A., Frank, R., Massart, R., 1990, J. Mater. Sci., 25, pp. 3249. [116] Bee, A., Massert, R., Neveu, S., 1995, J. Magn. Mag. Mater., 149, pp. 6. 176 [117] Schlesinger, H., J. Am. Chem. Soc.,1953, 75, 215. [118] Yiping, L., et al, 1989, J. Magn.Mag. Mater., 79, pp. 321. [119] Wells, S., et al, 1989, J. phys.: Condens. Mater., 1, pp. 8199. [120] Suslick, K. S., Hyeon, T., Fang, M., 1996, Chem. Mater., 8, pp. 2172. [121] Wang, Q., Xia, H., Zhang, C., 2001, J. Appl. Polym. Sci., 80, pp. 1478. [122] Kernizan, C. F., Klabunde, K. J., Sorensen, C. M., Hadjipanayis, G. C., 1990, Chem. Mater., 2, pp. 70. [123] Chen, J. P., Lee, K. M., Sorensen, C. M., Klabunde, K. J., Hadjipanayis, G. C., 1994, Phys. Review B., 75, pp. 5876. [124] Vollath, D., Szabo, D. V., Taylor, R. D., Wills, J. O., J. Mater. Res., 1997, 12, pp. 2175. [125] McHenry, M. E.., Majetich, S. A., DeGraef, M., Artman, J. O., Staley, S. W., 1994, Phys. Rev. B., 49, pp.11358. [126] Bee, A., Massert, R., Neveu, S., 1995, J. Magn. Mag. Mater., 149, pp. 6. [127] Chatterjee, J., Haik, Y., Chen, C. J., 2002, 246, pp. 382. [128] Leong, Y. K., Seah, U., Chu, S. Y., Ovy, B. C., 2001b, Colloids and Surfaces A , 182, pp. 263-68. [129] Kim, D. K., Zhang, Y., Voit, W., Rao, K. V., Muhammed, M., 2001a, “Synthesis and characterization of surfactant-coated superparamagnetic monodispersed iron oxide nanoparticles,” J. Magnetism and Magnetic Materials, 225, pp. 30-36. [130] Nalwa, H. S., Handbook of Nanostructured Materials and Nanotechnology, Vol 1, pp. 3-4. [131] Auzans, E., Ph.D. Thesis: Mn-Zn Ferrite nanoparticles for water and hydrocarbon based ferrofluids: preparation and properties, Institute of Physics of Latvian University. [132] Nalwa, H. S., Handbook of Nanostructured Materials and Nanotechnology, Vol 1, pp. 17. [133] Chow, G. M., Kurihara, L. K., Chem synth and processing of nano powders and films. [134] Mathiowitz, E., 1990, J. Scan. Micros., 4, pp. 329. 177 [135] Smith, J., and Wijn, H. P. J., Ferrites, 1959, John Wiley and Sons, N.Y. [136] Morrish, A. H., and Clarck, P. E., 1975,“High-field Mossbauer study of Mn-Zn ferrites,” Physical Review B, 11, pp. 278-286. [137] Shannon, R. D., 1974, Acta Cristallogr., A3z, pp. 751. [138] Guillaud, C., and Creveuex, H.,1950, “Proprietes ferromagnetiques des ferrites mixtes de Co et de Zn et de Mn et Zn,” C. R. Acad. Sci ., (Paris), 230, pp. 1458. [139] Guillaud, C., and Roux, 1949, C. R. Acad. Sci., (Paris), 229, pp. 1331. [140] Ohta, K., 1963, J. Phys. Soc. Jap., 18, pp. 685. [141] Galt, J. K., Yager, W. A., Remeika, J. P., and Merritt, F. R.,1951, Phys. Rev., 81, pp. 470. [142] Massart, R., 1981, IEEE Trans. Magn. Mag., 17, pp. 1247. [143] Tourinho, F . A., Franck, R., and Massart, R.,1990, J. Mater. Sci., 25, pp. 3249. [144] Chalyi, V. P., and Novosadova, E. B., 1973, Neorganicheskie Materiali, 9, pp. 2190. [145] Komarneni, S., Fregeau, E., Breval, E., and Roy, R. ,1988, J. Amer. Ceram. Soc. 71(1), pp. C26. [146] Remy, H.,“Lehrbuch der Anorganischen Chemie, band I,” 1960, Leipzig, p. 467. [147] Ueda, M., Shimada, S. H., and Inagaki, M.,1991, J. Mater. Chem., 3, pp. 1199. [148] Upadhyay, R. V., Mehta, R. V., Parekh, K., Srinivas, D., Pant, R. P., 1999, J. of Mag. Mag. Matr., 201, pp. 129-132. [149] Nejad, A. M., Movahedi, A. A. M., Safarian, S., 2002, Thermochimica Acta, 389, pp. 141-151 [150] De Gennes, P. G.,1987, Adv. Colloid Interface Sci., 27, pp.189. [151] Dickson, E., Ericsson, L., 1991, Adv. Colloid Interface Sci., 143, pp. 9. [152] Scintag: Basics of X-Ray Diffraction, http://epswww.unm.edu/xrd/xrdbasics.pdf [153] http://www.unl.edu/CMRAcfem/temoptic.htm 178 [154] http://www.mos.org/sln/SEM/tour15.html [155] Hiemenz, P. C., Rajagopalan, R., 1997,“Principles of Colloid and Surface Chemistry,” editor, New York, Marcel Dekker, Inc. [156]Jordan, A. P., Wust, H., Fahling, W., Johns, A., Hinz, R., Felix, 1993,“Inductive heating of ferrimagnetic particles and magnetic fluids – physical evaluation of their potential for hyperthermia,” Int. J. Hyperthermia, 9(1), pp. 51-68. [157] Park, Byeong-Ho, Bong Sig Koo, Young Kon Kim, Moon Kon Kim, 2002, “The induction of hyperthermia in rabbit liver by means of duplex stainless steel thermoseeds,” Korean Journal of Radiology, 3(2), pp. 98-104. [158 ] Brusentsov, Nikolai, A., V. V. Gogosov, T. N. Brusentsova, A. V. Sergeev, N. Y. Jurchenko, Anatoly A. Kuznetsov, Oleg A. Kuznetsov, L. I. Shumakov, 2001, “Evaluation of ferromagnetic fluids and suspensions for the site-specific radiofrequency- induced hyperthermia of MX11 sarcoma cells in vitro”, J. Magn. Magn. Mater., 225, pp. 113-117. [159] Halliday, David, Robert Resnick, Jearl Walker, 1997, Fundamentals of Physics, John Wiley and Sons, Inc. New York, p. 762. [160] http://www.abdn.ac.uk/physics/case/squids.html [161]Tourinho, F., 1988, Ferrofluides a base de ferrite de Mn et de ferrite de Co, Ph.D. Thesis, Paris. [162] Choa, Y., H-., Yang, J., -K., Yang, W., -J., Auh, K., -H.,2003, J. Magn. Magn. Mater., 266, pp. 20-27. [163] Kim, D., K., Zhang, Y., Voit, W., Rao, K., V., Muhammed, M., 2001, J. Magn. Magn. Mater., 225, pp. 30-36. [164] Smit, J., and Wijn, H. P. J.,1959, Ferrites, John Wiley & Sons, NY, pp. 139-142. [l65] Cullity, B. D.,1972, Introduction to Magnetic Materials, Reading, MA, Addison- Wesley, p. 617. [166] Brown, W. F., 1963, J. Appl. Phys., 34, pp.1319. [167] Fannin, P. C., Scaife, B. K. P., and Charles, S. W., 1988, J. Magn. Magn. Mater. 72 , pp. 95. [168] Maiorov, M. M., 1979, Magnetohydrodynamics, 2, pp. 21 (cover- to-cover translation of Magnitnaia Hidrodinamika). 179 [169] Andra, W., d’ Ambly, C.G., Hergt, R., et al.,1999, J. Magn. Magn. Mater., 194, pp. 197. [170] Landau, L. D., Lifshitz, E. M., 1960, Electrodynamics of Continuous Media, Pergamon Press, London. [171] Neuberger, T., Scho, B., Hofmann, pf, H., 2005, et al., J. Magn. Magn. Mater., 293, pp. 483. [172] Atkinson, W. J., Brezovich, I. A., and Chakraborty, D. P., 1984,“Usable frequencies in hyperthermia with thermal seeds,” IEEE Trans. Biomed. Eng., vol. 31, pp. 70-75. [173] Brezovich, I. A., Atkinson, W. J., and Chakraborty, D. P.,1984, “Temperature distributions in tumor models heated by self-regulating nickel-copper alloy thermoseeds,” Med. Phys., vol. 11, pp. 145-152. [174] Matloubieh, A. Y, Roemer, R. B., and Cetas, T. C.,1984, “Numerical simulation of magnetic induction heating of tumors with ferromagnetic seed implants,” IEEE Trans. Biomed. Eng., vol. 31, pp. 227-234. [175] Haider, S. A., Cetas, T. C., and Roemer, R. B.,1993, “Temperature distribution in tissues from a regular array of hot source implants: An analytical approximation,” IEEE Trans. Biomed. Eng., 40(5), pp. 408-417. [176] Chen, Z. P., Roemer, R. B., and Cetas, T. C.,1992, “Three-dimensional simulations of ferromagnetic implant hyperthermia,” Med. Phys., 19 (4), pp. 989-997. [177] Cheng, K.W.E., Tang, C.Y., Chan, L.C., 2003,“Dynamic modeling of magnetic materials for high frequency applications,” Journal of Magnetism and Magnetic Materials, 139, pp. 578-584. [178] Zhu, J.G., Ramsden, B. V. S.,1993, “Discrete modeling of magnetic cores including hysteresis eddy current and anomalous losses,” IEEE Proc. A, 140 (4), pp. 317-322. [179] Pankhurst, Q. A., Jones, S. K., and Dobson, J.,2003, “Applications of magnetic nanoparticles in biomedicine,” J. Phys. D: Appl. Phys., 36, pp. R167-R181. [180] Ingrid Hilger, K.F., Wilfried Andra, Robert Hiergeist, Rudolf Hergt, Werner A. Kaiser, 2002, “Heating Potential of Iron Oxides for Therapeutic Purposes in Interventional Radiology,” Acad. Radiol., 9, pp. 198-202. [181] Rudolf Hergt, W. A., Carl, G., d'Ambly, Ingrid Hilger, Werner, A., Kaiser, Uwe Richter, and Hans-George Schmidt,1998,“Physical Limits of Hyperthermia Using Magnetite Fine Particles,” IEEE Transactions on magnetics, 34(5). 180 [182] Shimizu,T., Matsui, M.,2003, “New magnetic implant material for interstitial hyperthermia” Science and technology of advanced materials, 4, pp. 469. [183] Fogiel, M., 1986,“The numerical analysis Problem solver”, Staff of research and education association. [184] Burden, R. L., and Faires, J. D.,1988, “Numerical analysis”, 4th edition, Youngstown state university. [185] Atkinson, K. E., 1988,“An introduction to numerical analysis”, 2nd edition, University of Iowa. [186] Chikazumi, S., and Charap, S. H., 1964, “Physics of Magnetism”, John Wiley, pp. 323. [187] Van Wieringen, N., Van Dijk, J. D. P., Nieuwenhuys, G. J., Snel, C. E., and Cetas, T. C.,1996, “Power absorption and temperature control of multi-fillament palladium nickel thermoseeds for interstitial hyperthermia” Phys. Med. Biol., 41, pp.2367. 181 BIOGRAPHICAL SKETCH Saleh Saad Al-Hayek was born on July 14th, 1968, in Zarqa, Jordan, to Saad AL- Hayek and Aminah Abu-Shindi. He received a Cultural Exchange Program Scholarship from the ministry of higher education of Jordan and began his undergraduate studies in 1988 at the Birla Institute of Technology (B.I.T.), Ranchi, India. In 1992, he received his Bachelors degree in Mechanical Engineering. He worked in Jordan for a couple of years as a maintenance and operation engineer at the National Gas Industries, Amman, Jordan. In 1996, he received his Masters degree in Mechanical Engineering from Wayne State University, Detroit, MI. He was a GAANN fellowship recipient from Florida A&M University for 2 years. In 2004, he transferred to Florida State University. He received his PhD in mechanical Engineering under the advisement of Dean/Professor Ching-Jen Chen, in 2007. Saleh Hayek is married and the proud father of two children: Umar and Rhunda Hayek. 182