PIERS ONLINE, VOL. 4, NO. 2, 2008 276

The Design of a Half-bridge Series-resonant Type Heating Systemfor Magnetic Nanoparticle Thermotherapy

Cheng-Chi Tai and Chien-Chang Chen

Department of Electrical Engineering, National Cheng Kung University, Tainan, Taiwan

Abstract— Application of magnetic materials for hyperthermia of biological tissue has beenknown, in principle, for more than four decades. Many empirical works were done in order toconfirm a therapeutic effect on several types of tumors by performing experiments with animalsor using cancerous cell cultures. The main idea of magnetic nanoparticle (MNP) thermotherapyis to utilize 7- to 50-nm diameter of ferric oxide (Fe3O4) particles which are heated up to 42Cunder AC magnetic field for cancer therapy applications. In order to achieve the goal of killingcancer cells using the AC magnetic field, we designed a heating system to generate magnetic fieldwhich can be focused and frequency adjustable. This study adopts the half-bridge series-resonanttype circuit as the core scheme of the heating system, and utilizes the frequency-adjustable designto conduct the heating experiment. The experiment results show that the designed coil and theheating system can warm up the magnetic nanoparticle 6C in 32 minutes.

1. INTRODUCTION

Electromagnetic inducing of heat has been extensively studied for the treatment of hyperthermia,wherein case deposited magnetic particles are used to locally heat human tissues. The basic ideaof hyperthermia is raising the tissue temperature up to between 41.5 and 46C to kill malignantcells while preserving normal cells [1]. Many empirical works were done in order to confirm thetherapeutic effect on several types of tumors by performing experiments with animals [2] or usingcancerous cell cultures, and poor AC magnetic field parameters [3]. At least two full-sized humanprototypes have been built based on magnetic fluid hyperthermia (MFH) and used shortly for thefirst clinical trials of hyperthermia [3]. But, these systems are too bulky and with poor heatingefficiency. Therefore, we want to develop a more compact, stable and efficient heating applicator.Our research is focused on heating system design and the improvement of applicator [4, 5].

2. SYSTEMATIC CIRCUIT STRUCTURE

The circuit diagram of the heating system is shown in Fig. 1, which includes a rectification circuit, aconverter circuit, gates driver, a half-bridge MOSFET inverter and an applicator. The halfCbridgeseries-resonance circuit [6] basically has two bi-directional switches of power MOSFET (Q1 andQ2) and a resonant circuit. Each power switch (S1 or S2) is composed of a switch of MOSFET (Q1

or Q2), as shown in Fig. 2. According to the combined form of the MOSFETs and the resonantcircuit, it can commonly be classified into the series resonant topology, parallel resonant topology,and a combination series-parallel resonant topology. The power MOSFET is focused on turn-on resistance, reducing conduction losses, operating junction temperatures, and switching speeds.This proposed circuit is adopted to avoid introducing any DC voltage upon the applicator duringoperation. The power MOSFET switches, Q1 and Q2, of the HBSR inverter are gated by twocomplementary signals, vgs1 and vgs2, respectively. Each switch (S1 or S2) is composed of a switchof power MOSFET Q1 (Q2) and its intrinsic anti-parallel diode D1 (D2). To prevent cross condition,the waveforms of vgs1 and vgs2 should be non-overlapping and have a short dead time. AC voltagesupply is given in input voltage 110 Vrms and output voltage. By symmetrically driving two powerMOSFET switches, the output of the HBSR inverter is a sine-wave voltage with a DC term of Vdc/2on the applicator circuit. Thus, a DC-blocking capacitor (CS1 and CS2) must be used for blockingthe DC term of the square-wave. On the other hand, the HBSR inverter outputs a square-wavevoltage without any DC term on the applicator circuit. Hence, there is no DC component acrossthe applicator to increase the current during resonant [7, 8].

3. COIL MODELING AND HEATING SYSTEM

The applicator consisted of two parts, wire coil and ferrite core. Although the structure of theelectromagnetic coil seems quite simple, its high frequency and high current mechanism are very

PIERS ONLINE, VOL. 4, NO. 2, 2008 277

Figure 1: System structure. Figure 2: Half-bridge series-resonant type circuit.

complicated. Fortunately, when the applicator operated in high frequency, it has been demonstratedto be approximately an open circuit, and the coil characteristic is not sensitive as verified from theimpedance analyzer (Agilent 4294A). Fig. 3 shows the comparison of coil models from measurementand simulation (impedance and phase). The operation characteristics of the high-frequency heatingsystem can be calculated by using the applicator model shown in Fig. 4. The measured parametervalues of the coil model are listed in Table 1. The E-type geometry structure of the applicator andthe simulated distribution of magnetic field are shown in Figs. 5 and 6. The MNP heating systemis shown in Fig. 7.

Figure 3: Comparison of coil models from measurement and simulation: (a) impedance and (b) phase.

Figure 4: Coil model. Figure 5: The geometry structure of the applicatorand the simulated distribution of magnetic field.

PIERS ONLINE, VOL. 4, NO. 2, 2008 278

Figure 6: The magnetic coil of the applicator. Figure 7: The MNP heating system.

Table 1: Parameters of the coil model.

Parameter Name Measurement Value

Lp 48.3804 µH

Cp 14.1527 pF

Rp 7.08915 kΩ

Rs 0.02Ω

4. EXPERIMENT RESULTS AND DISCUSSION

The induced electromagnetic heating effect in magnetic materials is mainly attributed to threephysical phenomena, namely, hysteresis, eddy current, and Neel or Brownian relaxation losses.The total power loss in a conductor media is [9]:

Ptotal =(

βhµ + βeµ2f

ρ+ βnb

2πfτ

1 + (2πfτ)2

)fH2, (1)

where βh, β2 and βnb are the geometry constant coefficients of hysteresis, eddy current, and Neelor Brownian relaxation, µ is the permeability of the magnetic material, H is the amplitude of theapplied AC magnetic field, ρ is electrical resistivity, τ is relaxation time and is exponentially sizedependent. However, heating of MNP is inherently due to hysteresis losses and Brownian relaxation

Table 2: SAR experiment results.

ExperimentConcentration

(Fe/mL)Amount(mL)

TemperatureDifference(∆T/∆t)

SAR values(W/g)

A 0.02887 Fe3O4 (2.0mL) 4.9E-3 0.653

B 0.0144Fe3O4 (1.0mL)

+ Water(1.0mL)2.6E-3 0.694

C 0.02887 Fe3O4 (1.5mL) 5.2E-3 0.693D 0.02887 Fe3O4 (1.5mL) 5.1E-3 0.679E 0.02887 Fe3O4 (1.5mL) 4.4E-3 0.586

PIERS ONLINE, VOL. 4, NO. 2, 2008 279

losses [10, 11]. In fact, the experiment results of many studies revealed that the heating power lossis affected by the strength of magnetic field and the characters of the material, such as particlesize, size variation, and magnetization saturation [12]:

SAR = C∆T

∆t

1mFe

, (2)

where C is the specific heat of sample, ∆T/∆t is temperature gradient. The experimental valuesof SARs for five different cases are shown in Table 2.

In experiments A and B (Fig. 8), the results show absolute heating temperature for differentFe3O4 concentration. Furthermore, the heating effects on nanoparticle (Fe3O4) are mainly at-tributed to three physical phenomena, as indicated in (1). The heating curves, C, D, and E, fordifferent initial temperatures are shown in Fig. 9. The results show similar curves which agree withthe expected physical characteristics.

Figure 8: Experimental results for absolute heatingtemperature at different concentration.

Figure 9: Experimental results for absolute heatingtemperature at different initial temperature.

5. CONCLUSION

In this study, we have succeeded in using the half-bridge resonant methods with coil design usingLitz wire to heat iron powder and magnetic nanoparticle. The results show that under differentbackground temperature all the samples can be heated up to 42C in AC magnetic field. There isan absolute temperature increased about 6C in 32 minutes. We have achieved a substantial stepto the final goal of killing cancer cells using the AC magnetic field and have attained the expectedheating effect for the study.

ACKNOWLEDGMENT

This research was supported by the grant from National Science Council, Taiwan (NSC 95-2221-E-006-016). Also, this work made use of Shared Facilities supported by the Program of Top 100Universities Advancement, Ministry of Education, Taiwan.

REFERENCES

1. Ferguson, S. D., J. A. Paulus, R. D. Tucker, S. A. Loening, and J. B. Park, J. of AppliedBiomaterials, 55–60, April, 1993.

2. Lisa, B. P. and J. O. Blanchette, “Nanoparticle and targeted systems for cancer therapy,”Advanced Drug Delivery Reviews, Vol. 56, 1649–1659, 2004.

3. Jordan, A., R. Scholz, P. Wust, H. Fahling, and F. Roland, “Magnetic fluid hyperthermia(MFH): Cancer treatment with AC magnetic field induced excitation of biocompatible super-paramagnetic nanoparticles,” J. of Magnetism and Magnetic Materials, Vol. 201, Nos. 1–3,413–419, 1999.

PIERS ONLINE, VOL. 4, NO. 2, 2008 280

4. Tai, C.-C. and M.-K. Chen, “A compact half-bridge induction heating system for magneticnanoparticle thermotherapy applications,” Biomedical Engineering — Applications, Basis, andCommunication, Vol. 19, No. 1., 27–35, 2007.

5. Chen, C.-C., C.-L. Chen, M.-K. Chen, and C.-C. Tai, “Experimental study on the heatingcharacteristics of magnetic nanoparticles by an AC magnetic field for local hyperthermia,”International Symposium on Nano Science and Technology (2006-ISNST), Southern TaiwanUniversity of Technology, Tainan, Taiwan.

6. Kazimierczuk, M. K. and D. Czarkowski, Resonant Power Converter, John Wiley & Sons, Inc,1995.

7. Steigerwald, R. L., “A comparison of half-bridge resonant converter topologies,” IEEE Trans-actions on Power Electronic, Vol. 3, No. 2, 174–182, April 1988.

8. Mohan, N., T. M. Undeland, and W. P. Robbins, Power Electronics, Applications and Design,John Wiley & Sons, Inc, 2nd Ed., 1995.

9. Hergt, R., “Physical limits of hyperthermia using magnetite fine particles,” IEEE Transactionson Magnetics, Vol. 34, No. 5, 3745–3754, Sept. 1998.

10. Rosensweig, R. E., Journal of Magnetism and Magnetic Materials, Vol. 252, 370–374, 2002.11. Mornet, S., S. Vasseur, F. Grasset, and E. Duguet, Journal of Materials Chemistry, Vol. 14,

2161–2175, 2004.12. Ma, M., “Size dependence of specific power absorption of Fe3O4 particles in AC magnetic

field,” Journal of Magnetism and Magnetic Material, Vol. 268, No. 1, 33–39, Jan. 2004.