Session 3P6 Microwave Treatment of Materials

Session 3P6
Microwave Treatment of Materials
Microwave Penetrating and Heating of Metallic Powders
Anton P. Anzulevich, V. D. Buchelnikov, I. V. Bychkov, Dmitri V. Louzguine-Luzgin, . . . . . . . . . . . . .
Effective Medium Approximation for Composite from Three-layered Spherical Particles
D. M. Dolgushin, Anton P. Anzulevich, V. D. Buchelnikov, I. V. Bychkov, Dmitri V. LouzguineLuzgin, . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Metallic Glassy and Composite Samples Produced by Using Microwave Radiation
Dmitri V. Louzguine-Luzgin, V. D. Buchelnikov, G. Xie, S. Li, A. Inoue, N. Yoshikawa, M. Sato,
Full Wave Analysis of Cylindrical Microwave Reactor
Pierre Pribetich, Christophe Lohr, Didier Albert Camill Stuerga, . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Thermal Tuning and Loop Modes within Cylindrical Applicator
Didier Albert Camill Stuerga, Christophe Lohr, Pierre Pribetich, . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Effects of Geometrical Parameters within Microwave Applicator Design
Didier Albert Camill Stuerga, Christophe Lohr, Pierre Pribetich, . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Measurement of Dielectric Properties and Finite Element Simulation of Microwave Pretreatment for
Convective Drying of Grapes
S. R. S. Dev, Y. Gariépy, G. S. Vijaya Raghavan, . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Multiphysics Simulations of Microwave Heating Phenomena in Domestic Ovens
Michal Soltysiak, Malgorzata Celuch, Ulrich Erle, . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Efficiency Optimization for Microwave Thermal Processing of Materials with Temperature-Dependent
Media Parameters
Ethan K. Murphy, Vadim V. Yakovlev, . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Coupled Electromagnetic-thermal 1-D Model of Combined Microwave-convective Heating with Pulsing
Microwave Energy
Erin M. Kiley, Suzanne L. Weekes, Vadim V. Yakovlev, . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Regularities of Semiconductor Powders Dynamics in Chladni Effect
Victor I. Kuzmin, D. L. Tytik, . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
385
386
387
388
389
390
391
392
393
394
396
398
386
Progress In Electromagnetics Research Symposium Abstracts, Moscow, Russia, August 18–21, 2009
Microwave Penetrating and Heating of Metallic Powders
A. P. Anzulevich1 , V. D. Buchelnikov1 ,
I. V. Bychkov1 , and D. V. Louzguine-Luzgin2
1
2
Chelyabinsk State University, Russia
WPI Advanced Institute for Materials Research, Tohoku University, Japan
Abstract— Owing to so-called skin-effect bulk metals reflect microwaves (MWs) and can hardly
be heated. They can undergo only surface heating due to limited penetration of the MW radiation. Whereas metallic powders can be penetrated into itself and absorb such radiation and
efficiently heat. Recently MW heating has been successfully applied to powdered metals and fully
sintered samples were obtained in 1999 in a multimode cavity. Later MW heating in separated
electric (E-) field and magnetic (H-) field of a standing wave was performed. The MW sintering of
various metals powders, steels and non-ferrous alloys helped to produce sintered samples within
tens of minutes at sintering temperature ranges from 1370 K to 1570 K. Moreover, nanomaterials
and some composite materials can also be produced by such a technique.
The reason of heating of metallic powders has not been clarified fully yet. Here for explanation of
MW heating of metallic powders we propose the following model. We consider the metallic powder
as some composite medium. This composite medium consists from the mixture of spherical
metallic particles covered by thin oxide dielectric shell and gas (or vacuum) [1, 2].
Thus eddy currents can penetrate into metallic powders at a depth of the size of metallic particles
due to sphericity of the skin-depth of these particles [3]. Whereas in bulk metals eddy currents can
penetrate into a planar skin-depth only. But eddy currents in metallic powders can be generated
on all surface of conductive particle if allowed a condition of quasistationarity. Condition of
quasistationarity is requirement that a size of conductive domains less than wavelength of incident
MWs.
So, in the present work, we theoretically studied using a model of conductive composite the MW
penetrating mechanisms, the possible MW heating mechanisms of metallic powders and provide
some theoretical explanation of the MW penetrating and MW heating behavior for iron powder
(Fig. 1).
1000
T, K
800
600
400
0
200
400
600
800
t, s
Figure 1: The time dependence of temperature for iron powder. The solid line is the modeling results; the
dark square is the experimental ones.
REFERENCES
1. Buchelnikov, V. D., D. V. Louzguine-Luzgin, G. Xie, S. Li, N. Yoshikawa, A. P. Anzulevich,
I. V. Bychkov, and A. Inoue, “Heating of metallic powders by microwaves: Experiment and
theory,” J. Appl. Phys., Vol. 104, No. 9, 01, November 2008.
2. Anzulevich, A. P., V. D. Buchelnikov, I. V. Bychkov, D. V. Lousguine-Luzgin, N. Yoshikawa,
M. Sato, and A. Inoue, “Penetration of microwave radiation into and through metallic powders,” Solid State Phenomena, Vol. 152–153, 361–364, 2009.
3. Smythe, W. R., Static and Dynamic Electricity, 2nd Edition, New York, Toronto, London,
1950.
Progress In Electromagnetics Research Symposium Abstracts, Moscow, Russia, August 18–21, 2009
387
Effective Medium Approximation for Composite from Three-layered
Spherical Particles
D. M. Dolgushin1 , A. P. Anzulevich1 , V. D. Buchelnikov1 ,
I. V. Bychkov1 , and D. V. Louzguine-Luzgin2
1
Condensed Matter Physics Department, Chelyabinsk State University, Chelyabinsk 454021, Russia
2
WPI Advanced Institute for Materials Research, Tohoku University, Sendai 980-8577, Japan
Abstract— Let us consider the three-layered spherical particles, which randomly distributed
in gas (for example, in air) or vacuum. According to effective medium approximation (EMA)
an average value of electric displacement of effective medium connects with an average value of
electric field strength as
hDi = εeff hEi = εeff E0 ,
(1)
where Rεeff is the effective permittivity of composite, E0 is the external electric field, hDi =
(1/V ) V DdV , V is the volume of the whole composite.
In EMA, we deal with a mixture of two types of spherical particles, which are randomly distributed
in the effective medium. The first type of particles is three-layered particles. As second type
of particles we will consider spherical inclusions of gas (vacuum). It is considered that the
permittivity of such a composite is equal to the permittivity of the effective medium.
After substitution of electrical fields in Eq. (1) and their integration we find the final equation
for calculation of the effective permittivity of composite from three-layered spherical particles
µ
¶
x1
(1 − x1 )
1
(ε1 − εeff ) K1 p + (ε2 − εeff ) K2 p
+ (ε3 − εeff ) K3 p 1 −
x2
x2
x2
(εg − εeff ) (2εeff A − ε3 B)
(2)
+ 3 (1 − p)
= 0,
εg + 2εeff
where
K1
A
B
x1
α1
β1
µ
¶
α1
α1 β2
= 9ε3 x2 β2 1 +
, K2 = 9ε3 x2
, K3 = 3x2 (α1 α2 − 2x1 β1 β2 ) ,
β1
β1
= x1 β1 (2ε2 (1 − x2 ) + ε3 (1 + 2x2 )) − α1 (ε2 (1 − x2 ) − ε3 (1 + 2x2 )) ,
= x2 (2x1 β1 β2 − α1 α2 ) − 2 (α1 β2 − x1 β1 α3 ) ,
µ ¶3
µ ¶3
r1
r3
=
, x2 =
,
r2
r2
= ε1 + 2ε2 , α2 = ε2 + 2ε3 , α3 = ε3 + 2ε2 ,
= ε2 − ε1 , β2 = ε2 − ε3 ,
ε1 , ε2 , ε3 , εg is permittivity of core, first shell, second shell and gas, p is the volume fraction of
solid spherical particles in effective medium.
Figure 1: The dependences of real and imaginary parts of effective permittivity of composite from the volume
fraction of solid spherical particles for one set of parameters.
388
Progress In Electromagnetics Research Symposium Abstracts, Moscow, Russia, August 18–21, 2009
Metallic Glassy and Composite Samples Produced by Using
Microwave Radiation
D. V. Louzguine-Luzgin1 , V. D. Buchelnikov2, 3 , G. Xie2 ,
S. Li2 , A. Inoue1 , N. Yoshikawa4 , and M. Sato5
1
WPI Advanced Institute for Materials Research, Tohoku University, Sendai 980-8577, Japan
2
Institute for Materials Research, Tohoku University, Sendai 980-8577, Japan
3
Condensed Matter Department, Chelyabinsk State University, Chelyabinsk 454021, Russia
4
Graduate School of Environmental Studies, Tohoku University, Sendai 980-8579, Japan
5
National Institute for Fusion Science, 322-6 Oroshi, Toki, Gifu 509-5292, Japan
Abstract— Microwave heating is recognized for its various advantages, such as: time and energy saving, very high heating rates, considerably reduced processing cycle time and temperature,
improved mechanical properties, better product performance, etc. In the present work, we study
heating, phase transformations and sintering behavior of metallic glassy, crystalline and composite samples under microwave (MW) radiation. We developed new metallic glassy alloys and
composites having a large supercooled liquid (SCL) region (temperature range between the glasstransition (Tg ) and crystallization temperatures (Tx )) and significant flow-ability which can be
used for MW treatment and sintering. We prepared powder glassy and nanocrystalline samples
by gas atomization and mechanical milling techniques. We studied heating behavior of metallic
powders by MW radiation and performed numerical fitting of the observed heating curves. We
build a new custom-made MW radiation treatment machine (915 MHz), which allows electrical
and magnetic field separation and pressing. We processed the metallic glassy and crystalline
powders using a single mode MW applicator.
The metallic glassy alloy powders were produced by a high pressure argon gas atomization method
using argon gas. The specimen powders were placed in a position of either E-field or H-field
maximum area in the single-mode wave guide applicator and heated by energy absorption of
MWs having 2.45 GHz or 915 MHz (for some samples) frequency. Among the studied alloys are
Fe73 Si7 B17 Nb3 and Fe65 Co10 Ga5 P12 C4 B4 , Zr55 Cu30 Al10 Ni5 Cu50 Zr45 Al5 and Ni52.5 Zr15 Nb10 Ti15
Pt7.5 alloys sintered samples were obtained. Composite Ni52.5 Zr15 Nb10 Ti15 Pt7.5 /Sn and Cu50 Zr45
Al5 /Fe samples were also produced. Bulk metallic glasses (BMGs) exhibit high thermal stability,
ultra-high strength and good corrosion resistance. The combination of superior properties and
low material cost enhances BMGs to have promising applications as engineering and functional
materials. However, the critical size of various BMGs obtained is much smaller compared to
conventional crystalline alloys. Microwave heating of iron boride, Fe3 C powders, and mixtures
of iron and iron boride powder was performed in the separated E- and H-fields. The heating
mechanisms of metallic powder samples have also been studied and will be discussed in detail.
The heating rate was found to depend upon various factors including electrical conductivity,
thickness of the oxide layer, volume fraction of metallic part etc. which will also be discussed.
We also studied phase transformations and heating behavior of iron based ceramic powders in a
single mode microwave applicator.
Progress In Electromagnetics Research Symposium Abstracts, Moscow, Russia, August 18–21, 2009
Full Wave Analysis of Cylindrical Microwave Reactor
P. Pribetich1 , C. Lohr2 , and D. Stuerga1, 2
1
GERM/ICB, (Institut Carnot de Bourgogne), UMR 5209 CNRS, Université de Bourgogne
BP 47870, 21078 Dijon Cedex, France
2
NAXAGORAS Technology, France
Abstract— Study of high power industrial applicators implies full wave analysis of loaded
device as waveguides and especially cylindrical waveguides. Many interests have been shown in
cylindrical geometry because of the extensive use of this geometry for fluids heating within a
pipe. In case of lossy media, due to high level of dielectric losses, limits of classical perturbations
approaches and modes established for lossless structures can be completely avoided.
Authors describe an original technique for making full wave analysis of an inhomogeneous cylindrical waveguide loaded by a lossy pipe. The mode spectrum of the studied structure can be
obtained by use of analytical and numerical techniques. These matching conditions lead to the
characteristic equation which is expressed by a matrix. The eigenvalue or complex propagation
constant for each mode could be found within complex plane by a numerical procedure based
on the residue theory. This procedure calculates the zeros of the characteristic equation within
complex plane.
The results describe modes available in this kind of microwave applicators very close to industrial
operating devices. TE and TM modes have been studied and all these modes obtained have been
classified according to four classes: the propagative (β > α ≈ 0), the quasi-propagative (β ≈ α),
the attenuated (α > β), and the evanescent (α > β ≈ 0).
389
390
Progress In Electromagnetics Research Symposium Abstracts, Moscow, Russia, August 18–21, 2009
Thermal Tuning and Loop Modes within Cylindrical Applicator
D. Stuerga1, 2 , C. Lohr2 , and P. Pribetich1
1
GERM/ICB, (Institut Carnot de Bourgogne), UMR 5209 CNRS, Université de Bourgogne
BP 47870, 21078 Dijon Cedex, France
2
NAXAGORAS Technology, France
Abstract— The microwave applicator studied is constituted by two coaxial rods. The water
pipe to be heated is described by the central lossy rod and the other medium is air (lossless
dielectric). The cylindrical applicator has been shown because of the extensive use of this geometry for heating pipe. In conventional heating techniques or conduction techniques, high-power
densities at the outer surface of the pipe lead to excessive heating of boundary layers, compared
with higher flow rates along the pipe axis.
Full-wave analysis make by the authors have shown that it is possible to minimize electric field
amplitude on the wall water pipe in order to reduce superheating. The advantage would be the
highest value of electric field at specific regions where the loads are normally inserted. Moreover,
the tuning due to thermal dependency of dielectric properties of water induces consequent change
of the phase (β) and attenuation (α) constants of the propagation constant (γ = α + j) within
the temperature range 10◦ C–140◦ C.
The authors have obtained original modes which exhibit loop within (α, β) complex plane. Despite the strong tuning due to thermal dependency of dielectric properties of water, the mode
guided wavelength has variation close to few millimetres within the temperature range 10◦ C until 140◦ C. According to these results; predictive control and design of optimized travelling wave
applicators could be obtained. According to authors, a viable alternative to the trial and error
methods currently used for designing microwave applicator for industrial heating applications has
been set up.
Progress In Electromagnetics Research Symposium Abstracts, Moscow, Russia, August 18–21, 2009
Effects of Geometrical Parameters within Microwave Applicator
Design
D. Stuerga1, 2 , C. Lohr2 , and P. Pribetich1
1
GERM/ICB, (Institut Carnot de Bourgogne), UMR 5209 CNRS, Université de Bourgogne
BP 47870, 21078 Dijon Cedex, France
2
NAXAGORAS Technology, France
Abstract— The classical industrial design of microwave applicators and specifically the choice
of the geometrical shape are based on a simple similarity principle between the wave propagation
and spatial distribution within the empty and the loaded microwave applicator. The dielectric
load is the object to be heated. Moreover, dielectric tuning due to thermal dependency of
dielectric properties must be taken into account. Hence, this design method will be only valid
if the dielectric perturbation induced by the reactor is negligible. In fact, the magnitude of the
perturbation is proportional to reactor to applicator volume ratio. Hence, it is more efficient
but also more complicated to be guided by a geometrical matching principle. According to this
geometrical matching principle the microwave applicator designer want to ensure a good match
between electric field spatial distribution and geometrical shape of the chemical vessel used.
This geometrical matching principle is easier to apply for monomodes applicators because of the
knowledge of the wave propagation directions and spatial distribution. The limit of this design
method is that it requires the knowledge of the empty applicator modes, but also the of the
loaded applicator modes.
The authors will show effect of geometrical parameters of a microwave applicator constituted by
two coaxial rods. Effects of cylindrical waveguide and load diameters will be discussed in term
of TE and TM modes propagation constants.
391
392
Progress In Electromagnetics Research Symposium Abstracts, Moscow, Russia, August 18–21, 2009
Measurement of Dielectric Properties and Finite Element Simulation
of Microwave Pretreatment for Convective Drying of Grapes
S. R. S. Dev, Y. Gariépy, and G. S. V. Raghavan
Department of Bioresource Engineering, McGill University, QC, Canada
Abstract— In this study measurement and modelling of the dielectric properties of grapes
was conducted on a matrix of frequencies (from 200 MHz to 10 GHz) and temperatures (5◦ C to
80◦ C). There are studies on the measurement and modelling of dielectric properties of grapes at
2.45 GHz for different temperatures. But there is no data available on the dielectric properties of
grapes at different frequency. This gives a better understanding of the behaviour of the grapes
on a broader electromagnetic spectrum and helps further simulation studies at other permitted
frequencies like 915 MHz.
Mass production of dried raisins is often done by convective drying. The main problem in grape
drying has been slow drying rate due to waxy layer at skin. Dipping in hot water or the use of
chemicals such as sulphur, NaOH, and ethyl or methyl oleate emulsions are some of pretreatments
widely used for grape drying to increase drying rate of raisins. While subjecting the grape
berries to microwave heating, the moisture in the berry is heated to a saturation temperature,
the temperature rises with pressure, resulting in volume expansion, causing the berry to rupture.
Research on the possible use of microwave as a pretreatment for the convective drying of grapes
was conducted and found that if the rate of vaporization is controlled by the level of microwave
energy applied, a puffed nature can be achieved by the rupture of different layers. In grapes, this
rupturing is reported to start near the surface and propagate into the interior, giving the raisins
a puffy texture, thus providing the necessary pathways for moisture migration from different
layers of the berry. This enhances the drying rate in further drying process. But there is
poor understanding of the mechanisms involved and actual energy distribution inside the grapes
creating new channels for moisture migration.
In this study, a Finite Element Model (FEM) of the microwave pretreatment of the grapes was
made and simulation studies were conducted for grapes subjected to 5 minutes pretreatment
under 915 MHz and 2450 MHz and power densities of 0.5 W/g, 5 W/g and 50 W/g in order to
visualize and investigate the energy distribution within the berries.
Progress In Electromagnetics Research Symposium Abstracts, Moscow, Russia, August 18–21, 2009
Multiphysics Simulations of Microwave Heating Phenomena in
Domestic Ovens
Michal Soltysiak1 , Malgorzata Celuch1 , and Ulrich Erle2
1
Institute of Radioelectronics, Faculty of Electronics and Information Technology
Warsaw University of Technology, Nowowiejska 15/19, 00-665 Warszawa, Poland
2
Nestlé Product Technology Centre, Lebensmittelforschung GmbH Singen
Lange Str. 21, 78224 Singen, Germany
Abstract— This work presents a multiphysics technique for microwave heating simulations.
Experimental validation for the case of food products in domestic microwave ovens is provided.
Hence, the presented technique facilitates better understanding and more effective design of
systems for microwave treatment of materials.
The multiphysics simulation technique combines three building blocks: a full-wave 3D electromagnetic solver, a set of thermal analysis modules, and temperature-dependent material data
obtained via measurements. A commercial FDTD simulator, QuickWave-3D, constitutes the
electromagnetic part of the system. It serves to calculate electromagnetic power converted
into heat within the treated product. To this end, electromagnetic steady state with initial material parameters is first reached, and average values of power dissipated due to electric, magnetic,
and metal losses are extracted. These are further applied as a 3D source function by a thermal
module. Several thermal modules are available within the system, and an appropriate one
is chosen depending on phenomena relevant to a particular scenario. In the simplest approach,
the temperature pattern is updated from the initial state via a linear solution of the 3D heat
diffusion equation. More typically, nonlinear problems are solved, where dielectric and/or thermal material parameters automatically varying as a function of local temperature or enthalpy
density. These data are generated with the in-house measurement setups that will also be
presented at the Symposium. The effects of load movement, including rotation in popular domestic ovens but also translation along user-defined trajectories, can also be taken into account.
After each thermodynamic solution over a user-defined heating time step is completed, the electromagnetic analysis is resumed from the previous electromagnetic steady state, but with the
modified material parameters.
Essentially, a nonlinear electromagnetic-thermodynamic problem is converted to a multistep
parametric problem, with bilateral coupling between the two solvers. The user decides about
the number of heating time steps to cover the total heating time.
The coupled electromagnetic — thermal simulations become powerful tools for microwave
engineers. They allow one to produce and inspect the temperature patterns within the whole
volume of the heated product. Different shapes, dimensions and initial positions of the sample
inside the oven cavity can easily be considered from the viewpoint of their influence on the final
temperature patterns. Additionally, high costs associated with physical experiments, such as
production of samples, manufacturing of apparatus prototypes, and measurements, are reduced
to the necessary minimum (being a priori investigation of thermal and dielectric properties of the
sample as a function of temperature).
The results of multiphysics simulations are compared to temperature patterns actually measured
in selected food products treated in a domestic microwave oven. The measurements are conducted
with an infrared camera or fibre optic thermal probes. Good overall agreement between
simulations and measurements is noted. Discrepancies are related to uncertainties in material
characterisation, which therefore requires enhancements. Elements crucial for correct mapping
between the laboratory and virtual scenarios are pointed out.
393
394
Progress In Electromagnetics Research Symposium Abstracts, Moscow, Russia, August 18–21, 2009
Efficiency Optimization for Microwave Thermal Processing of
Materials with Temperature-Dependent Media Parameters
Ethan K. Murphy and Vadim V. Yakovlev
Department of Mathematical Sciences, Worcester Polytechnic Institute
Worcester, MA 01609, USA
Abstract— Microwave heating of materials is known to be the technology capable of substantial improvement in the efficiency and quality for a variety of applied thermal processes.
However, corresponding industrial implementations are still quite limited because, as a physical
phenomenon, microwave heating is hard to control. Several years ago, it was suggested that, with
the remarkable progress in efficient numerical techniques allowing for quite accurate computer
simulation of complex microwave systems, the problem of optimization of microwave thermal
processing can be approached through modeling-based techniques [1]. One crucial aspect of this
type of optimization, namely, optimization of microwave energy coupling interpreted as a numerical characteristic of system efficiency, has been discussed and conceptualized, for the first
time, in [1]. Then in [2] an artificial neural network (ANN)-based finite-difference time-domain
(FDTD)-backed algorithm has been introduced as an optimization procedure suitable for viable
multi-parameter optimization of energy efficiency for microwave heating systems.
Recently, the critical upgrade of the algorithm proposed in [2] has been reported [3, 4]. The revised version of the optimization technique deals now with a new objective function and features
a principal improvement of dynamic training of the RBF network by Constrained Optimization
Response Surface (CORS) technique — global response surface type algorithm designed to minimize the number of function evaluations in the process of finding the global minimum. It has
been shown [3, 4] that the new technique substantially outperforms its predecessor [2] by getting
optimal solutions of better “quality” and substantially reducing the number of FDTD analyses (and thus dramatically cutting the optimization’s computational cost) for such systems as a
waveguide band-pass filter, a dielectric resonator antenna, and a loaded microwave oven.
In this contribution, we demonstrate how the CORS-RBF optimization procedure [3, 4] can be
applied for efficiency optimization of the systems of microwave heating of materials whose media
parameters (the dielectric constant ε0 and the loss factor ε00 ) change in the course of heating. The
considered scenario is concerned with a microwave oven (with the dimensions and feed location
of Sanyo EM-N105W ) containing a glass shelf and a cylindrical sample of processed material on
it. The optimization problem is formulated as follows:
Given:
(1) the processed material with temperature characteristics ε0 (T ) and ε00 (T ) for the working
temperature range, and
(2) the fixed dimensions of the cylinder (diameter D and height H);
Find:
(a)
(b)
(c)
(d)
thickness of the glass shelf t,
diameter of the shelf d,
the position of the shelf above the bottom h, and
the position of the cylinder on the shelf with respect to its center, dx and dy
such that the reflection coefficient of the entire system is guaranteed to be less than 0.3 (i.e., less
than 9% of microwave energy is reflected back to the magnetron) in 75% of the frequency range
from 2.4 to 2.5 GHz.
The 5-parameter optimization problem is solved for a particular pair of (ε0 , ε00 ) corresponding to
a certain temperature; the optimization is then repeated, for the same space of design variables,
for the values of the dielectric constant and the loss factor at a number of other temperatures.
In the considered illustration, we work with experimentally determined values of ε0 and ε00 of
resin R498 at T = 30, 80, and 120◦ C [5]. The underlying FDTD model developed for the 3D
conformal FDTD simulator QuickWave-3D [6] consists of 166,000 to 189,000 cells (16 to 18 MB
RAM), so one analysis of the system involving 20,000 time-steps takes 2.2 to 2.5 min of CPU
time on Xeon 3.2-GHz PC operating under Windows XP. It turns out that the CORS-RBF
procedure requires as little as 177, 160, and 185 simulations (i.e., about 10 h total) for each of
these temperatures, respectively, to find an optimal solution satisfying the 75% frequency band
Progress In Electromagnetics Research Symposium Abstracts, Moscow, Russia, August 18–21, 2009
395
constraint. (For comparison, the best solution found by the previous version of this optimization
algorithm [2] corresponds to 52% bandwidth, and it needs 462 analyses to get this solution.)
Finally, the optimal configuration for each temperature is tested for two other pairs of (ε0 ,0 , ε00 ),
and the one demonstrating best bandwidth is chosen as overall optimal.
Due to a fully parameterized underlying FDTD model, the optimization problem can be instantly
formulated for any other set of parameters in accordance with the practical need of the system
designer. Thanks to its computational effectiveness, the presented optimization tool may assist in
fairly practical CAD projects in microwave power engineering easily dealing with several design
variables and performing optimization of regular widely available PCs.
REFERENCES
1. Mechenova, V. A. and V. V. Yakovlev, “Efficiency optimization for systems and components
in microwave power engineering,” J. Microwave Power & Electromag. Energy, Vol. 39, No. 1,
15–29, 2004.
2. Murphy, E. K. and V. V. Yakovlev, “RBF network optimization of complex microwave systems represented by small FDTD modeling data sets,” IEEE Trans. Microwave Theory Tech.,
Vol. 54, No. 7, 3069–3083, 2006.
3. Murphy, E. K. and V. V. Yakovlev, “Reducing a number of full-wave analyses in RBF neural
network optimization of complex microwave structures,” IEEE MTT-S Intern. Microwave
Symp. Dig., Boston, MA, June 2009.
4. Murphy, E. K. and V. V. Yakovlev, “Optimization of complex microwave systems with the
CORS RBF neural network backed by FDTD analysis data,” Progress In Electromagnetics
Research Symposium, Moscow, Russia, August 18–21, 2009.
5. Akhtar, M. J., L. E. Feher, and M. Thumm, “Nondestructive approach for measuring temperaturedependent dielectric properties of epoxy resins,” J. Microwave Power & Electromag.
Energy, Vol. 42, No. 3, 17–26, 2008.
6. QuickWave-3D, QWED Sp. z o. o., Warsaw, Poland, 1998–2009, http://www.qwed.com.pl/.
396
Progress In Electromagnetics Research Symposium Abstracts, Moscow, Russia, August 18–21, 2009
Coupled Electromagnetic-thermal 1-D Model of Combined
Microwave-convective Heating with Pulsing Microwave Energy
Erin M. Kiley1, 2 , Suzanne L. Weekes2 , and Vadim V. Yakovlev2
1
2
Department of Mathematics, University of New Hampshire, Durham, NH 03824, USA
Department of Mathematical Sciences, Worcester Polytechnic Institute, Worcester, MA 01609, USA
Abstract— As a result of the well-known tendency of microwave (MW) heating to develop
hot and cold spots in practically unpredictable locations, special measures must be undertaken
to simultaneously bound the maximum temperature (which occurs at the hot spots) while still
sufficiently heating the rest of the object (in particular, the cold spots). This difficulty has
been ameliorated with use of turntables [1], or mode stirrers [2], or multiple feeds [3], while an
alternative approach used in industry, which has not been systematically studied yet, features
a MW pulsing regime [4] in which periods of relaxation allow the effects of thermal diffusion, a
naturally-occurring mechanism that operates on a vastly different time scale from MW heating,
to make the temperature distribution more uniform.
This contribution presents an algorithm and modeling software allowing for 1D simulation of
thermal processing of dielectrics by pulsed MW energy. The presented technique is a continuation
of our earlier study [5], which first presented software to consider the pulsing regime as a technique
to ensure heat diffusion through the load in the time intervals when the microwave is off, and thus
to evaluate its efficacy as a controlling parameter in making the resulting temperature field more
uniform. The algorithm is also capable, in accordance with industrial practices, of simulating
combined MW-convective heating. Here we report an upgraded version of the algorithm and a
new series of computational experiments which allow us to see the pulsing regime with different
pulsing parameters on the materials with different electromagnetic and thermal properties and
with the new option of adiabatic boundary conditions.
The software is implemented as a MATLAB code executing an analytical-numerical solution of a
1-D fully coupled electromagnetic-thermal problem, with temperature-dependent electromagnetic
parameters (dielectric constant and the loss factor) and thermal parameters (heat conductivity,
heat capacity, and density). We account for these dependencies in the solution of the coupled problem using a special numerical procedure implementing a finite-difference computational
scheme. Similarly to [5], performance of the code was validated by the 3-D conformal FDTD
simulator QuickWave-3D [6].
While a 1D solver cannot be applied to realistic MW heating systems and be considered as a tool
for practical CAD, it is effective in the context of studying the functionality of a MW pulsing
regime, and as it is fully parameterized, can be used to study pulsing in the context of a variety
of scenarios actually used by industry.
A series of performed computational experiments shows that microwave pulsing in combination
with convective heating at a temperature equal to or greater than the minimum temperature
required for the load to be sufficiently heated is more effective than microwave pulsing alone,
because during periods when the microwave is off, diffusion is conditioned by both thermal
conductivity and additional heat introduced to the load. Naturally, when the boundaries are
maintained at a temperature lower than the intended minimum threshold, then truly sufficient
heating can never be achieved; yet, even this kind of convective heating is beneficial for uniformity
in the first stages of heating. We also note the general trend that the greater the number of
pulses over a given time interval, the more quickly uniformity is achieved. The developed model
can therefore be conceptually and specifically instructive in designing practical applicators with
pulsing MW energy.
REFERENCES
1. Kopyt, P. and M. Celuch, “FDTD modeling and experimental verification of electromagnetic
power dissipated in domestic microwave oven,” J. Telecomm. & Information Techn., No. 1,
59–65, 2003.
2. Plaza-Gonzalez, P., J. Monzó-Cabrera, J. M. Catalá-Civera, and D. Sánchez-Hernández, “Effect of mode-stirrer configurations on dielectric heating performance in multimode microwave
applicators,” IEEE Trans. Microwave Theory Tech., Vol. 53, No. 5, 1699–1706, 2005.
Progress In Electromagnetics Research Symposium Abstracts, Moscow, Russia, August 18–21, 2009
397
3. Pitarch, J., A. J. Canós, F. L. Peñaranda-Foix, J. M. Catalá-Civera, and J. V. Balbastre, “Synthesis of uniform electric field distributions in microwave multimode applicators by multifeed
techniques,” Proc. 9th Conf. Microwave & High-frequency Heating, 221–224, Loughborough,
U.K., 2003.
4. Gunasekaran, S. and H.-W. Yang, “Effect of experimental parameters on temperature distribution during continuous and pulsed microwave heating,” J. Food Engineering, Vol. 78,
1452–1456, 2007.
5. Feldman, D. A., E. M. Kiley, S. L. Weekes, and V. V. Yakovlev, “Modeling of temperature
fields in 1D and 2D heating scenarios with pulsing microwave energy,” Proc. 41st Microwave
Power Symp., 130–134, Vancouver, BC, Canada, 2007.
6. QuickWave-3D, QWED Sp. z o.o., Warsaw, Poland, 1998–2009. http//: www.qwed.com.pl/.
398
Progress In Electromagnetics Research Symposium Abstracts, Moscow, Russia, August 18–21, 2009
Regularities of Semiconductor Powders Dynamics in Chladni Effect
V. I. Kuzmin1 and D. L. Tytik2
1
Moscow State Institute of Radio Engineering, Electronics, and Automation
pr. Vernadskogo 78, Moscow 119454, Russia
2
Frumkin Institute of Physical Chemistry and Electrochemistry
Leninskii pr. 31, Moscow 119991, Russia
Abstract— This article is a presentation of powder pattern dynamics (Chladni figures) on the
plates in a variety of shapes and critical dimensions under acoustic and magnetic fields applied
in the vicinity of bifurcation points. The study involved the use of powders with critical size of
particles of diverse composition — semiconductor material B4 C and dielectric material SiO2 . The
study detected the acoustic field frequencies at which powder figures (B4 C) rearrange themselves
on the plane by escaping into the third dimension (forming a vortex above the plane at the point
of bifurcation). Dielectric powders (SiO2 ) at certain frequencies form stationary vortex above
the plane due to the natural lumpiness effect, which is the cause of existence of dominant sizes
of material structures in the nature, regardless of their phase state. They are consistent with
dominant values of time intervals (frequencies) forming the rhythm quantization system. The
natural lumpiness effect serves as technological basis for the transfer of electromagnetic signals
in various media at specific frequencies (transparency windows).
Combined effect of the acoustic and magnetic fields defines the specifics of powder figures (B4 C)
on the plane and brings forth the problem of electromagnetic impact on powder materials with
various physical and chemical properties. These experiments demonstrate that the phase state
of a substance can be controlled through application of alternating fields of diverse origin along
with critical values of wavelengths (frequencies).