Scientific paper
A Facile Route to the Synthesis of Coated Maghemite Nanocomposites for Hyperthermia Applications
Gregor Ferk,1 Irena Ban,1'6'* Janja Stergar,1 Darko Makovec,2 Anton Hamler,3 Zvonko Jaglicic4'5 and Miha Drofenik1'2
1 University ofMaribor, Faculty of Chemistry and Chemical Engineering, Smetanova 17, SI-2000 Maribor, Slovenia
2 Jozef Stefan Institute, Department for Materials Synthesis, Jamova 39, SI-1000 Ljubljana, Slovenia
3 University ofMaribor, Faculty of Electrical Engineering and Computer Science, Smetanova 17,
SI-2000 Maribor, Slovenia
4 Institute of Mathematics, Physics and Mechanics, Jadranska 19, SI-1000 Ljubljana, Slovenia
5 University of Ljubljana, Faculty of Civil and Geodetic Engineering, Jamova 2, SI-1000 Ljubljana, Slovenia
6 Center of Excellence NAMASTE, Jamova 39, SI-1000, Ljubljana, Slovenia
* Corresponding author: E-mail: irena.ban@uni-mb.si Tel: +386 2 2294 417; Fax: +386j 2 2527 774
Received: 06-01-2012
Abstract
CM-dextran-covered maghemite particles for applications in magnetic hyperthermia treatments were synthesized and their physical, magnetic and morphological properties were examined. Magnetic fluids were prepared and their heating properties in an alternating magnetic field were studied. The results reveal that the particle size and the thickness of the carboxy-methyl-dextran (CM-dextran) coatings have a decisive influence on the heating properties: specific absorption rate (SAR). The majority of the magnetic dissipation comes from the Neel relaxation, while the Brown contribution is small. A thermal steady state at the selected temperature (42 °C) can be achieved using synthesized maghemite particles with proper particle morphology and by controlling the magnetic field intensity or the frequency.
Keywords: Magnetite nanoparticles, hyperthermia, magnetic properties.
1. Introduction
Magnetic nanoparticles and their derivatives, magnetic fluids, are being investigated for various applications, including resonance imaging,1 drug delivery,2 magnetic separation3 and magnetic hyperthermia.4,5
The magnetic nanoparticles and/or their derivatives normally consist of polymer-covered ferrimagnetic iron oxides, which are synthesized in situ by precipitation in the presence of polymers that act as a stabilizer. It has been noted that materials based on precipitated ferrimagnetic iron oxides generally give superior results, compared with other types of magnelic materials. In addilion, precipitated iron oxides were shown to produce little, if any, toxicity to humans. This means that iron oxides are,
because of their high magnetization, biocompatibility, simple synthesis and a relatively high magnetic dissipation, among the most attractive materials, with the only weakness being that they possess a relatively high Curie point that may cause the tissue to overheat. However, the generated heat can be effectively controlled by selection of the particle morphology6 and the alternating magnetic field (AMF) intensity and/or frequency.7 When we are able to control all these parameters and reach thermal equilibrium in such a way that the generated heat at the selected temperature and location will be equal to that lost to the environment due to the thermal convection or heat exchange to the veins, no overheating of the surrounding tissue will take place. In this case iron oxide can still be used, in spite of its high transition temperature. For this
reason it is of crucial importance to be able to precisely determine the specific absorption rate (SAR) of the magnetic energy absorbed by the magnetic particles dispersed in the magnetic fluid (MF). This magnetic energy is then transformed into the heat that warms up the MF under consideration.
A variety of nanocomposites (polymer-coated magnetic particles) dispersed in water have been prepared in the past, for example, water-soluble dextran,8 polyvinyl alcohol,9 poly(ethylene) glycol),10 amylose starch11 and chitosan biopolymer.12 Among them, CM-dextran is a polysaccharide with outstanding properties, as well as being biodegradable, biocompatible and bi-oactive, and as such is commonly used for the stabilization of MFs.13
In this work we synthesized MFs using maghemite particles with a proper morphology stabilized by CM-dex-tran. We achieved a steady state at the selected temperature using synthesized maghemite particles with a proper particle morphology and by applying a controlled magnetic power.
2. Experimental
2. 1. Reagents and Materials
For the synthesis of the maghemite particles coated with CM-dextran the following chemicals were used: Fe-Cl2 ■ 4H2O and FeCl3 ■ 6H2O (Acros Organics), sodium carboxymethyl - dextran salt (CM-dextran; Fluka Bioche-mika with M = 14400 g/mol), Tryptic Soy Agar (Fluka) and 25% NH4OH (Alkaloid Skopje). During the synthesis of the magnetic fluid, ultrasound radiation was applied with an ultrasonic generator (Sonics Vibra Cell 750 W inten sity).
2. 2. Synthesis
The detailed procedure for sample A was as follows:
A total of 125 mL of a sotution containing Fe2+ (0.135 mol/L) and Fe3+ (0.113 mol/L) ions was poured into a 250-mL beaker equipped with a magnetic stirrer and then heated to 80 °C. When a temperature of 80 °C was achieved, 50 mL of 25% NH4OH was added (pH > 9). The sotution turned from a red-brown cotor to black, indicating the formation of magnetite, which then oxidized to maghemite. The suspension was then aged at 80 °C for 1 hour and cooled to room temperature. The suspension was magnetically decanted and the magnetic particles were washed several times with distilled water.
The detailed procedure for sample B was as follows:
The same as sample A, but immediately after the addition of NH4OH, 10 ml of the CM-dextran solution (10 g CM-dextran / 20 g H2O) was added in several successive small additions over a period of 30 minutes. The suspen-
sion was then aged at 80 °C for 30 minutes and cooled to room temperature. The suspension was magnetically decanted and the magnetic particles were washed several times with di stil led water. To the mag ne ti cally de can ted partictes, 10 mL of CM-dextran sotution was added and sonicated for 1 hour to obtain the magnetic fluid (MF). In order to sediment the agglomerates, the MF was centrifu-ged at 5000 rpm for 5 min and the supernatant was ultra-filtrated with a UF membrane Mwcut_off = 100 kDa to produce a MF with a large content of iron oxide particles. For the SAR measurement we prepared a MF with 9.75% (wt) of iron oxide.
For the SAR measurements in agar:
Agar was added to the MF with a high content of iron oxide particles and during stirring heated to 70 °C to obtain a very viscous fluid. After that the content was poured into the measuring cuvette and cooled to room temperature, after which a hard gel was obtained. The content of iron oxide in the gel was 9.75% (wt).
2. 3. Analysis and Measurements
The nanoparticles were characterized using X-ray diffraction analyses (Bruker Axs, D 5005) and a transmission electron microscope (TEM) (JEOL 2010F). The partict e size was estabt ished using the Scherrer method, based on the broadening of the X-ray diffraction peaks. The particte size and the size distribution were determined from the TEM images. Dynamic light scattering (Horiba LB-550) was used to estimate the CM-dextran-maghemite hydrodynamical size distribution (dH).
The presence of the CM-dextran was confirmed by FTIR spectroscopy (PERKIN ELMER 1600), white the amount/concentration of CM-dextran adsorbed onto the magnetic particles was determined by a thermogravime-tric analysis (TGA) (TGA/SDTA 851, Mettter Totedo). The magnetization was measured with a helium-cooled superconducting quantum interference device (Quantum Design MPMS-XL-5 SQUID magnetometer).
In our investigation a conventionally built system was used to generate an alternating magnetic field with a nominal field strength of 2 kA/m and a maximum frequency of 104 kHz. For the calorimetric measurements of the magnetic fluid losses a sample of magnetic fluid was placed in a glass vial in the center of the surrounding supply coil. The length of the coil was twice the length of the glass vial in order to guarantee a homogeneous field in the center of the vial. The glass vial was placed in some heat-insulating polystyrene to reduce the heat dissipation from the magnetic fluid to the surroundings and vice versa. The thermal measurements were performed using a fiber-optic thermometer in the presence of the magnetic field. The ca-lorimetric measurements were carried out using a calorimeter, where the heat effects can be determined by an im-
mediate measurement of the temperature. In that case the calorimeter is a measurement system and the heat effects are due to the absorption of magnetic energy when subjected to an alternating magnetic field. The measurements were continued until a steady state is achieved, i.e., until the heat induced equals the heat dissipated. For more see ref..14
3. Results and Discussion
3. 1. X-ray Diffraction Analysis
Fig. 1 shows the X-ray diffraction patterns of the as-prepared maghemite particles (A) and the particles covered with CM-dextran (B). All the diffraction peaks displayed by the diffractogram correspond to the maghemite cubic phase (JCPDS No.39-1346). From the X-ray line broadening and the Sherrer formula it is possible to estimate an average particle size of 16.0 nm for sample (A) and 11.8 nm for sample (B).
Typical TEM images of the as-synthesized sample
(A)	and the CM-dextran-coated maghemite nanoparticles
(B)	are shown in Figs. 2a and 2b, and the corresponding grain-size distributions obtained from the TEM images
Fig. 1. XRD spectra of as-prepared samples (A) and of maghemite particles covered with CM-dextran (B).
are shown in Figs. 2c and 2d. The average particle diameters estimated from the TEM images are 14.5 nm for sample (A) and 12.0 nm in sample (B), in good agreement with X-ray diffraction analysis. The main reason that the coated particles showed a smaller average particle size is because they were centrifuged, which eliminates the larger particles from the MF.
a)
b)
c)
d)
6 8 10 12 14 16 18 20 22 24 Particle diameter (rim)
6 8 10 12 14 16 18 20 Particle diameter (nm)
Fig. 2. TEM images of a) as-synthesized maghemite particles (A), b) CM-dextran-stabilized maghemite particles (B)-insert; smaller agglomerates, c) the grain size distribution of the uncoated particles (A) and d) the CM-dextran-coated particles (B).
3. 2. TGA and SDTA Analysis
The distinctive weight losses established for the magnetic particles covered with CM-dextran can be divided into two temperature regions: 150-500 °C and 500-750 °C, Fig. 3. A total weight loss of 21.2% (sample B) occurred during the thermal heating from room temperature to 750 °C. In the first temperature range, from 150 °C to 500 °C, the oxidation of the CM-dextran occurs, as indicated by an exothermic simultaneous differential thermal analysis (SDTA) peak and a weight loss of 16.4%. The exothermic peak associated with this mass loss is the result of the oxidation of the CM-dextran molecules due to the temperature, which is consistent with the literature data on free polysaccharides at 230-400 °C.15 The last step is associated with a weight loss of 3.1%, which is associated with the removal of the CM-dextran covatently bound to the particle surface, i.e., the very first covatent bound layer of CM-dextran to the particle surface.
Fig. 3. Thermal analysis (TGA) of the maghemite nanoparticles coated with CM-dextran and the corresponding SDTA.
It is well known that CM-dextran dissolved in water is in equilibrium with that attached physically to the magnetic particles. Namely, after the very first layer of CM-dextran, which is covalently bound to the particle surface, the subsequent layers are att ached by physical bonds, where hydrogen bonding prevails. In the water suspension this dextran is in equilibrium with the dextran in the sus-pension.16
3. 3. Fourier Transform Infrared
Spectroscopy (FTIR) and Dynamic Light Scattering Analysis (DLS)
The CM-dextran coatings were investigated using FTIR spectroscopy. In Fig. 4 the FTIR spectra of the mag-
4000 3600 3200 2800 2400 2000 1600 1200 900 400 Wavenumber (cm"1)
Fig. 4. FTIR spectrum of a) maghemite particles b) maghemite particles coated with CM-dextran and c) CM-dextran.
hemite nanoparticles a), maghemite nanoparticles coated with CM-dextran b) and pure CM-dextran c) are shown.
The absorption bands at 450 cm1, 587 cm1 and 626 cm1 in the FTIR spectra of iron oxide nanoparticles with and without a CM-dextran coating are attributed to the multiple lattice absorption bands of partially ordered maghemite.17 The broad band in the region between 3200-3600 cm1 is assigned to O-H stretching. The absorption band at 1596 cm-1 is attributed to asymmetric COO stretching in alkali carboxylic acid salts,18 which indicated the existence of free alkali carboxylate groups on the nanoparticles surface. The observed symmetrical C=O stretching at 1404 cm-1 suggested that the carboxylate was bound symmetrically to the surface of nanoparticles, as one of the possible carboxylate binding modes.19 Mo -reover, adsorption bands at 2921 cm-1 (C-H stretching), 1107 cm-1 (C-O-C stretching) and characteristic peak of CM-dextran at 1011 cm-1 additionally supports the attachment of CM-dextran onto the maghemite nanoparticles surface.
Hydrodynamic diameter and particle size distribution of the CM-dextran-coated maghemite nanoparticles was inspected with measuring the DLS. From the Fig 5 hydrodynamic diameter dH was found to be approximately 77.9 nm with the broad size distribution.
3. 4. Magnetic Properties
The zero field cooling (ZFC) and field cooling (FC) curves in Fig. 6a indicate that the particles in sample (B) are superparamagnetic and show a broad maximum at around 200 K, shifted with respect to the point where the ZFC and FC curves coincide. This is an indication of the particle agglomeration and the spread of the particle size. In this case a unique K using the relation TB = VK/25kB for the assembly of the particles in sample (B) cannot be
Fig. 5: The particle size distribution of the CM-dextran-coated maghemite nanoparticles in sample (B) measured using dynamic light scattering.
given. As such, V and K correspond to the individual particle properties, where K represents the anisotropy constant, kB is the Boltzman constant and V is the particle volu me.
The magnetization vs. the magnetic field of the as-synthesized and CM-dextran-covered maghemite particles are shown in Fig. 6b. The uncoated particles show no hysteresis, exhibiting a saturation magnetization of 65.2 emu/g. This value is smaller than that for the bulk materials Mbulk = 88 emu/g. The reduction of the magnetization of the magnetic particles with the decreasing particle size is usually caused by the incomplete coordination of the atoms on the particte surface, leading to a non-coltinear spin configuration that causes the formation of a surface spin canting forming a nonmagnetic layer, i.e., a "dead layer", which might reduce the particle's magnetization.20,21
We can estimate the thickness of this layer in sample (B) by applying the equation of Zheng et al.22
Msat = MBulkd - 6t / d)
(1)
where t is the thickness of the surface layer, d is the diameter of the particles, and Msat and MBulk are the total sa-
turation of the whote particte and of its bulk, respectively. When Msat is deduced from the experimentally determined saturation magnetization of 65.2 emu/g, MBulk
is taken as the bulk val ue (88 emu/g ) and dT
14.5
nm, as obtained from the TEM images of the as-prepared particles, we obtain a thickness for the "dead layer" of t = 0.62 nm.
The M(H) dependence at 293 K of the samples (A) and (B) was analyzed by considering a log-normal particle size distribution, as described in,23 and using a volumetric particle packing fraction £ =1for solid nanoparticles (non-magnetic fluid) in our case. We calculated the median diameter of the distribution dA = 8.20 nm and the standard deviation cA = 0.37 for sample (A) and dB = 8.17 nm and the standard deviation cB = 0.36 for sample (B).
Considering the estimated "dead layer" 2t = 1.24 nm and the standard deviation we can compare this value with those obtained from the TEM images of the samples (A) and (B). Namely, the magnetic measurement "sees" only the magnetic phase, white that obtained from the TEM images also includes the so-called "dead layer", which is not otherwise registered by the magnetic measurements.
a)
b)
	60-
	40-
	20-
O)	
	
b	0-
0)	
s	-20-
	-40-
	-60-
	-———.
-Sample A * Sample 8	
	
	
———	■
-10
-5
0
H( kOe)
10
Fig. 6. a) ZFC and FC magnetization of coated maghemite particles (B) in the MF, TB = 203 K, b) hysteresis of as-synthesized maghemite particles (A) and CM-dextran-coated particles (B) at 300K.
Fig. 7. Size distributions of sample (A) and sample (B), deduced from the fit of the magnetization curves.
The size distribution of the magnetic particles resembles that determined from the TEM images, albeit with a smaller average particle size. Average particles diameters for our particles determined using various experimental methods are collected in Table I.
Table 1. Average particle diameter of samples (A) and (B) obtained using different methods.
	dXRD (nm)	dTEM(nm)	dM(nm)	dH(nm)
Sample A	16.0	14.5	8.20	/
Sample B	11.8	12.0	8.17	77.9
3. 4.1. Dissipation Mechanism
An AMF magnetic field will normally induce heating in a MF due to the dissipation of heat from the magnetic particles that is caused by the delay in the relaxation of the magnetic moment due to the Neel relaxation (which refers to the rotation of the magnetic moment within the particle) and the Brown relaxation (which refers to the rotation of the particle itself). The characteristic time for the Neel relaxation is:
T = T0 exp (KV/kbT
(2)
where kb is the Boltzmann constant, T is the absolute temperature, T0 is the time constant ~ 10-9 s, K is the aniso-tropy constant and V is the volume of the particle, while the Brown relaxation time is given by the relation:
Tb = 3nVH / kT	(3)
where n is the viscosity of the carrier liquid, VH is the
hydrodynamic volume of the particle, kb is the Boltzmann constant, and T is the absolute temperature.
According to,24 the volumetric power dissipation can be expressed as:
P = n^o/'Ho2/
(4)
where is the imaginary part of the complex AC susceptibility and is related to the material parameters of the magnetic fluid/' = £0[®t/(1 + (arc)2)], ¡u0 = 4n 10-7 (Tm-A-1) is the permeability of free space, Tis the effective relaxation time, and the overall relaxation time is given as 1/t = 1/tn + 1/Tb, where the shortest term governs the dominant mechanism and f = a/2n is the frequency of the applied magnetic field. The above relaxations govern the dissipation of the magnetic energy, which then heats the magnetic fluid.
The morphology of the magnetic particles, their ani-solropy constant and the carrier liquid's viscosity play a crucial role during the absorption of the magnetic energy when the magnetic fluid is subjected to an AMF magnetic field. All the parameters are implicitly involved in the effective relaxation time t. The theoretical estimation of the SAR as a function of the particle diameters and the magnetic field at 104 kHz calculated using the equations above is in Fig. 8.
The theoretical estimation of the SAR at room temperature as a function of the monodispersed particle size shows a maximum that increases with frequency and is centered on a particte size around 15.3 nm. In Table I the average particle diameter for our particles in sample (B) determined using various experimental methods is between 8 and 12 nm. The pronounced dependence of SAR on the particle diameter is mostly related to the Neel retaxation. Here, the Neel retaxation time depends exponentially on the magnetic anisotropy and the particle volume, while the Brownian time depends linearly on the particle volume/size and the solvent visco sity.
3. 4. 2. SAR Measurements
The specific absorption rate (SAR) of the sample was measured calorimetrically in an AMF with a frequency of 104 kHz and a magnetic field amplitude ranging from 1.5 to 3.0 kA/m, which is in the range of biomedical applications for magnetic hyperthermia.26
From the course of the temperature increase, Fig. 9, the SAR was estimated as:
SAR = (CsoIvent/^1ronox1de)(AT/At)
(5)
where Csolvent is the heat capacity of the solvent (C = 4.18 J K-1g-1 for water), xiron oxide is the weight fraction of iron oxide and (AT/At) is the initial slope of the time dependence of the self-heating temperature.
a)
E
<
b)
10 15
d (nm)
65403
I3"
0C
V) 2_
-Total heat dissipation - Neel relaxation
10 15 20 25 d (nm)
Fig. 8. a) Theoretical estimation of the SAR at room temperature as a function of the particle diameter (d) and the AMF magnetic field (H). Here we have K (anisotropy constant) = 16 kJ/m3,6 T = 293 K, n = 1E-03 Pas (viscosity of the media - of water), f = 104 kHz and CM-dextran layer thickness S = 18 nm (calculate as Lmin 25) and b) total heat dissipation through Neel and Brownian relaxation at f = 104 kHz and H = 2 kA/m (difference between total heat dissipation and Neel relaxation losses is the Brownian relaxation losses).
0 10 20 30 40 50 60 70 80 90 100 f (min)
Fig. 9. Time dependence of the self-heating temperature on the magnetic field at 104 kHz for sample (B).
After a certain period of self-heating the steady state was achieved when the generated heat at constant tempe-ralure was equal to that lost to the environment due to thermal convection. The maximum temperature of this steady stat e is import ant when using iron-oxide-based MFs with a high Curie temperature for "in vivo" applications, because this is the only way to use such a MF without damaging the surrounding tissue during the magnetic hyperthermia. In Fig. 10, the temperature of the magnetic fluid at the steady state is plotted vs. the applied magnetic field.
In the system considered the steady stale at, e.g., 42. °C, is achieved at 2.0 kA/m (104 kHz). These condi-
0 10 20 30 40 50 60 70 80 90 100 f (min)
Fig. 10. Temperature of the system of the magnetic fluid in the calorimeter at steady state vs. the applied magnetic field at a constant frequency (f = 104 kHz).
tions can be achieved reproducibly and might serve as a guaranlee that in a silualion when instead of the system being considered a human tissue is considered and/or an "in vivo" application, the overheating of the tissue can be completely avoided. The SAR needed to heat the system to the desired temperature, for these samples at 42 °C was
SAR104 kHz, 2.0 kA/m = 0.854 W/giron oxide. Some typted SAR
data for a MF based on water as a carrying liquid and iron oxide (maghemile) as the magnelic phase are shown in Tab le II.
The theoretical estimation of the SAR shows that the main relaxation mechanism is due to the Neel relaxation, while the Brown relaxation is small and its contribution to
Table 2. Some typical SAR data for a MF based on maghemite particles.
d (nm)	H	f	SAR	SAR	Refe-
	(A/m)	(kHz)	(W/g)	normalized*	rences
8	24800	700	37	0.036	6
10.2	24800	700	275	0.266	6
12	2000	104	0.854	0.854	our
15.3	11000	410	600	5.030	27
16/12.8	27000	700	952	0.776	28
* normalized to H = 2.0 kA/m and f = 104 kHz
the dissipation mechanism must be very limited, Fig. 8. a, b. This can be substantiated by the DLS spectra. The smallest particle diameter dH that would ensure a relaxation time which would follow the fast changes of the magnetic fields at 104 kHz and contribute to the Brown retaxation must show a relaxation time of tb < 10-5 s, thus exhibiting a particle size of dH < 29.5 nm, estimated using equation (2). All the other particles with a larger hydrodynamic diameter, i.e., agglomerates, will be not active in the heating of the MF in a high-frequency magnetic field (HFMF).
Thus, based on the consideration above, only a very small fraction of the composite particles in sample (B) are able to contribute to the Brown relaxation. All the others are agglomerated in clusters and cannot rotate and the heating contribution of the Brown relaxation can be excluded from the total magnetic dissipation.
To reveal the mechanism that governs the major losses in sample (B), samples with immobilized particles were also prepared. The particles of the original magnetic fluid were immobilized by a drastic increase of the viscosity. The magnetic particles were dispersed in an aqueous solution of gelatin (sol).29 An abrupt increase in the visco-
Fig. 11. Dependence of the self-heating temperature on time at a magnetic field frequency of 104 kHz and a magnetic field amplitude of 2.0 kA/m for a magnetic fluid with agar and without it.
sity, by many orders of magnitude, can be achieved, and consequently viscous losses can be eliminated by simply freezing the Brownian motion of the magnetic particles.
In Fig. 11 the time dependence of the self-heating temperature on the magnetic field at 104 kHz for testing the magnetic fluid with agar and without it is shown.
It can be seen that the increase of the self-heating temperature of MF of sample (B) is a littte higher than that of the MF(agar) sample where the Brown relaxation is not possible, Fig.11. We believe that this small difference in the SAR is caused due to the Brown retaxation of particles in the sample (B) where a very small fraction of single particles with a hydrodynamic radii dH < 29.5 nm, Fig. 5, can freely rotate and contribute to the Brown relaxation. One can calculate approximately its separate contribution from the SAR difference at the steady state for both samples using the corresponding (A77Ai)steady state for both curves and eq. (4). This estimated contribution is alike to that obtained directly from the diagram on the Fig. 11 at the steady state. The Brown contribution was confirmed to be about 5%.
4. Conclusions
Our results have shown that the high Curie point of a magnetic particle, which can damage the tissue during a magnetic hyperthermia treatment, can be circumvented by applying a controlled particle morphology. The heat flow is adjusted with the AMF magnetic field intensity and/or frequency so that the generated heat at the selected temperature and location will be equal to that lost to the environment due to heat exchange. Furthermore, the investigation revealed that the major heat dissipation in our sample (B) came from the Neel retaxation losses. Here, the particle size governs the extent of the Neel relaxation process, while the Brown-based dissipation is very small. Here, the number of the particles with a suitable rH which absorb in the Brownian retaxation range is very small and contribute only 5% to the total SAR.
6. Acknowledgements
This work was supported by the Ministry of Education, Science and Sport of Slovenia. The authors also wish to acknowledge Milos Bekovic for his help in SAR measurements.
7. References
1.	A. E. Merbach, E. Toth: The chemistry of contrast agents in medical magnetic resonance imaging, John Wiley & Sons, Chichester, England, 2001.
2.	T. Neuberger, B. Schöpf, H. Hofmann, M. Hofmann, B. Rechenberg, J. Magn. Magn. Mater. 2005, 295, 483-496.
3.	A. Ito, M. Shinkai, H. Honda, T. Kobyashi, J. Biosci. Bioeng. 2005, 100, 1-11.
4.	U. Häfeli, W. Schütt, J. Teller, M. Zborowski (Eds.), Scientific and Clinical applications of Magnetic Carriers, Plenum Press, New York, 1997, 569-595.
5.	A. K. Gupta, M. Gupta, Biomaterials 2005, 26, 3995-4021.
6.	J. P. Fortin, C. Wilhelm, J. Servais, C. Menager, J. C. Bacri, F. Gazeau, J. Am. Chem. Soc. 2007,129, 2628-2635.
7. A. Skumiel, M. Labowski, H. Gojzewski, Mol. Quant. Acoust. 2007, 28, 229-239.
8.	D. Kirpotin, D. C. F. Chan, P. A. Bunn, Magnetic micropartic-les, US Patent Number 5,411,730, date of patent May 2, 1995.
9.	J. Lee, T. Isobe, M. Senna, J. Colloid Interface Sci. 1996, 177, 490-494.
10.	M. Suzuki, M. Shinkai, M. Kamihira, T. Kobayashi, Biotech-nol. Appl. Biochem. 1995, 21, 335-345.
11.	V. Veiga, D. H. Ryan, E. Sourty, F. Llanes, R. H. Marc-hessault, Carbohydr. Polym. 2000, 42, 353-357.
12.	K. Donadel, M. D. V. Felisberto, V. T. Favere, M. Rigoni, N. J. Batistela, M. C. M. Laranjeira, Mater. Sci. Eng., C 2008, 28, 509-514.
13.	S. Dutz, W. Andrä, R. Hergt, R. Müller, C. Oestreich, C. Schmidt, J. Töpfer, M. Zeisberger, M. E. Bellemann, J. Magn. Magn. Mater. 2007, 311, 51-54.
14.	M. Bekovic, I. Ban, A. Hamler, Assessment of magnetic fluid losses out of magnetic properties measurement, J. Phys.: Conf. Ser. 2010, 200, 1-4.
15.	A. Carpov, G. G. Bumbu, G. C. Chitanu, C. Vasile, Cellul. Chem. Technol. 2000, 34, 455-471.
16.	X. Q. Xu, H. Shen, J. R. Xu, J. Xu, X. J. Li, X. M. Xiong, Appl. Surf. Sci. 2005, 252, 494-500.
17.	M. P. Morales, S. Veintemillas-Verdaguer, M. I. Montero, C. J. Serna, Chem. Mater. 1999, 11, 3058-3064.
18.	M. Ibrahim, A. Nada, D. E. Kamal, Indian J. Pure Appl. Phys. 2005, 43, 911-917.
19.	Y. T. Tao, J. Am. Chem. Soc. 1993, 115, 4350-4358.
20.	J. M. D. Coey, Phys. Rev. Lett. 1971, 27, 1140-1142.
21.	R. H. Kodama, A. E. Berkowitz, E. J. McNiff, S. Foner, Phys. Rev. Lett. 1996, 77, 394-397.
22.	M. Zheng, X. C. Wu, B. S. Zou, Y. J. Wang, J. Magn. Magn. Mater. 1998, 183, 152-156.
23.	R. W. Chantrell, J. Popplewell, S. W. Charles, IEEE Trans. Magn. 1978, 14, 975-977.
24.	R. E. Rosensweig, J. Magn. Magn. Mater. 2002, 252, 370-374.
25.	T. Kawaguchi, M. Hasegawa, J. Mater. Sci.: Mater. Med. 2000, 11, 31-35.
26.	A. Jordan, R. Scholz, P. Wust, H. Fähling, R. Felix, J. Magn. Magn. Mater. 1999, 201, 413-419.
27.	R. Hergt, R. Hiergeist, I. Hilger, W. A. Kaiser, Y. Lapatnikov, S. Margel, U. Richler, J. Magn. Magn. Mater. 2004, 270, 345-357.
28.	M. Levy, C. Wilhelm, J. M. Siaugue, O. Horner, J. C. Bacri, F. Gazeau, J. Phys.: Condens. Matter 2008, 20, 204133.
29.	R. Hergt, R. Hiergeist, M. Zeisberger, G. Glöckl, W. Weitsc-hies, L. P. Ramirez, I. Hilger, W. A. Kaiser, J. Magn. Magn. Mater. 2004, 280, 358-368.
Povzetek
V prispevku je predstavljena sinteza in karakterizacij a magnetnih maghemitnih nanodelcev prevlečenih s karboksimetil dekstranom (CM-dekstran), primernih za uporabo v magnetni hipertermiji. Stopnjo specifične absorpcije (SAR) stabilne magnetne tekočine pripravljene iz maghemitnih nanodelcev prevlečenih s CM-dekstranom, smo preverili s pomočjo kalorimetričnih meritev magnetne tekočine izpostavljene zunanjemu izmeničnemu magnetnem polju. Rezultati kažejo, da velikost magnetnih nanodelcev in debelina prevleke CM-dekstrana vplivata na stopnjo specifične absorpcije (SAR), pri čemer večji del toplotnih izgub izhaja iz Neel-ove relaksacije, medtem ko je prispevek Brown-ove relaksacije majhen. Termično ravnotežje pri ciljni temperaturi (42 °C) lahko dosežemo s primernimi morfološkimi lastnostmi sintetizi-ranih magnetnih nanodelcev maghemita in s kontroliranjem j akosti ali frekvence zunanjega izmeničnega magnetnega polja.