-1 INTRODUCTIONshodhganga.inflibnet.ac.in/bitstream/10603/252/7/07_introduction.pdf · dielectric waveguides, lenses, radomes, dielectric resonators and microwave integrated circuit
Post on 18-Jul-2020
8 Views
Preview:
Transcript
- 1 INTRODUCTION
CHAPTER 1
INTRODUCTION
Microwave technology owes its origin to the design and development of radar and
gained a tremendous progress during the World War 11. In the earlier stages of
development, the invention of microwave generators like klystron, magnetron etc.
opened the gigahertz frequency region of electro-magnetic spectrum to
communication engineers. Hence the major development especially comes in the
field of satellite communication. It can be seen that microwaves constitute only a
small portion of electromagnetic spectrum, but their uses have become increasingly
important in the material characterization for industrial, scientific and medical
applications.
Material characterisation is essential for the proper selection and
implementation of a substance when used in industrial, scientific and medical
applications. The dielectric parameters over a wide range of temperature on low
loss dielectrics are needed to assess their suitability for use in telecommunications,
dielectric waveguides, lenses, radomes, dielectric resonators and microwave
integrated circuit (MIC) substrates; and on lossy materials for estimating their
heating response in microwave heating applications. The dielectric data would also
be required on lossy ceramics for their use as microwave absorbers, lossy pastes for
the design of new food packages, for heating in microwave ovens and on biological
materials for diathermy. The measurement of the dielectric parameters will serve as
a tool for investigating the intermolecular and intramolecular mechanisms of
compounds. The dielectric data have also been used to estimate the amount of
moisture in wcxd, sand and agricultural products. The dielectric properties are
needed for the calculation of internal electric fields resulting from the exposure to
non-ionizing electromagnetic (EM) fields and are thus important in the development
of diagnostic and therapeutic medical applications of this energy and studies of
possible hazards of EM fields. Electromagnetic waves in the radio and microwave
frequencies are effectively used in the treatment of hyperthermia of tumors and other
disorders.
In the industrial, scientific and medical (ISM) band which covers from few
kHz to several GHz, the range of relative permittivity (q) for the materials of
interest is very large. The real part of relative complex permittivity < can vary
from a value of about two to a value of a few thousand while the imaginary part E",
of complex permittivity from a small fraction to a large value of about hundreds for
some materials.
It is found that any single method of complex permittivity measurement is
not suitable over such a wide range of frequency and complex permittivity.
Besides, different materials may be available in different physical states such as gas,
Liquid, powder, paste, solid etc. Even in the case of solids, they are available in
different shapes such as flat sheet, grains and hard-to-grind arbitrary geometrical
shapes such as rocks. Due to the variations in the values of E' and E", frequency,
physical state and shape, different materials need different techniques for complex
permittivity measurements.
The methods for measuring the complex permittivities of materials can be
broadly classified into two categories:
I . Frequency domain methods
2. Time domain methods
The frequency domain methods are well established and have been in use for
over the past 50 years. But the time domain methods are recently developed.
However, time domain methods are growing in acceptance as these provide quick
measurement to estimate the dielectric response of a material over a wide frequency
range. Nowadays, this range covers beyond 10 GHz with the presently available
equipment. Figure 1.1 shows the various categories of methods for complex
permittivity measurements.
TIKE IN H I S C E L ~ S
I I(ETHCOS
I
I SINGLE
I UULTIPLE
I SINGLE
I WLTIPLE
REFLECTION REFLECTION TRANSnISSION l'RANSnISSIOH
CAVITY P E R T ~ A T I O N CAVITY ~XNANCE A ~ T I C HAWORK FREE'SPACE &ISSICH I I ANILYSER PROPAGAT ION LINE
SINGLE mlDE UULTI HODE LENGTH TUNED FRD3ENCY TUNED REFLECTION TRANSHISSION CAVITY CAVITY
I
I1 I
WAVEGUIDE I
COAXIAL I
Fiqure 1 . 1 . Classification of dielectric lneasurement metMs.
1.1 FREQUENCY DOMAIN METHODS
First of all, various categories of frequency domain techniques are
considered.
1.1.1 Free Space Methods
Free space methods are based on the optical type measurements. (Figure
1.2) Generally these methods were considered suitable for frequencies in the
millimeter wave region. Both reflection and transmission methods have been used.
A major drawback of these methods was the requirement of a large sample to avoid
diffraction effect around sample edges for performing measurements in the
centimeter wave region. However it has been shown that with precision horn lens
antennas whch have better far-field focussing ability, it is possible to make accurate
free space measurements in the low frequency bands (below X-band).
Compared to other methods, free space methods have the following
advantages.
(a) These provide contactless methods for the complex permittivity
measurement. Hence they are more suitable for high temperature
measurements.
(b) In the shorted waveguide method and cavity method, it is necessary to
machine the sample so as to fit exactly in the waveguide or cavity. During
the sample preparation, it is very difficult to machine the sample exactly to
the required dimensions. This requirement limits the accuracy of
measurement of hard and brittle materials. Free space methods do not suffer
from the above limitations, because no sample preparation may be required,
if the material is available as a sheet.
HP 8341 B HP 8510 B
SWEEP OSCILLATOR NETWORK ANALYSER
HP 9000/300 S-PARAMETER
INSTRUMENTATION COMPUTER
DIELECTRIC SHEET
TRANSMITTING A W E NNA
-
RECEIVING
-
Figure 1.2 Free space experimental set-up for complex permittivity measurement
Different free space methods are thoroughly discussed by Musil and Zacek
I l l . Ghodgaonkar et al. 121 have given the measurement set-up using network
analyser. They have reported results on some materials that agree very well with the
previous resulls obtained using more established methods.
1.1.2 Transmission Line Methods
In transmission line methods, a sample of the dielecvic material is placed
either between the outer and inner conductors of a coaxial line or inside a
waveguide. The sample when placed at the end of the waveguide may be terminated
with either a short or some other known impedance. The complex permittivity of
the material is then determined by the measurements of the line without the sample
and with the sample. Most important and widely used method was that developed by
Roberts and Von Hippel 13) (Figure 1.3). In this method, a waveguide is
terminated by the sample in physical contact with a short circuit. This niethod has
been extensively used in the past and is recommended as a standard method by the
American Society for Testing and Materials (ASTM). Nelson (41 has made the
measurement more convenient by writing a computer programme which exactly
computes 2 for high loss as well as low lass materials. This programme is
applicable to measurements made in rectangular and circular waveguides as well as
to coaxial lines Certain corrections are also included in the programme for the
influence of the slot in the slotted-waveguide section, and the difference in velocity
of propagation in air and vacuum. Even if these corrections are ignored, it is
claimed that the error in computed results is only 0.5% for t' and 2 - 3% for e".
Using the standard waveguide components, waveguide methtds have been used up
to 140 GHz. Beyond this frequency, waveguide dimensions become very small and
transmission losses very high 151.
PROBE
7
SBORT C I R C U I T
ISOLATION
DIELECTRIC GENERATOR
Figure 1.3 Transmission line method
Various transmission line methods with different designs for holding the
sample have been discussed in the literature and the modification is still going on.
For example, a two port circular cell which requires a sample of cylindrical shape
machined with a hole in the middle to fit hetween the imer and outer conductors of
the cell has been used by Nicolson and Ross [6] , and Weir 171. In another
configuration the sample is placed between a coaxial cable and either a short or open
[8]. Taherian et al. [9] have discussed the use of a two port cell, for measurements
at 1.1 GHz on brine solution, which has a symmetric cylindrical sample holder
comected to a coaxial line section at each end. This cell is different from those
used earlier such that the sample is in the shape of a solid circular disc without a
hole in its center. Scott [ lo] has proposed a new type of fixture in which very little
sample preparation is required for planar materials. In waveguide methods, at low
frequencies (<2GHz) a large sample and big temperature controlled oven are
required. Due to these demerits, the waveguide methods are not considered very
suitable for ISM applications.
1.1.3 Automatic Network Analyzer Methods
In certain aspects, automatic network analyser (ANA) methods can also be
considered as transmission line methods because the sample is held in a transmission
line and helecmc constant is determined by measuring the reflection andlor
transmission coefficient. The scattering parameters are measured with a vector
network analyser (HP 8510 type) consisting of a synthesized sweeper and an S
parameter test set. A PC (90001300 series instrumentation computer) can be used
for automation, data acquisition, printing of S-parameters etc. Even though these
methods have similarity with transmission line methods, they can be considered as
separate category because ANA methods have emerged as powerful modem methods
with which dielectric measurements can he canied out over a wide range of
frequencies in single measurement. A typical measurement set-up is given by
Ghodgaonkar ( 1 1 1 and details are given by Somlo 1121 (Figure 1.4).
HP 8341 B R P 8510 B
SWEEP OSCILLATOR NETWORK ANALYSER
HP 8514 B
S-PARAMETER T E S T S E T I NSTRUIIEUL'ATION
COMPUTER
SAMPLE BOLDER
F i g u r e 1.4 A u t o m a t i c n e t w o r k analyeer m e t h o d
In ANA methods, various measurement errors are significantly reduced by a
calibration procedure. The most general procedure uses a through path connection,
matched termination, a short and an open (TMSO) 1131. With wide-band sweep
oscillators, and under the control of a computer, dielectric measurements have been
done accurately by ANA's over a wide range of frequencies and materials 114-161.
In the case of broad band measurements, the TMSO calibration procedure
suffers in accuracy. Therefore different researchers have suggested the use of other
methods of calibration. In one such method an open ended coaxial probe (sensor) is
immersed in a material of known permittivity [17]. A major development in this
area is the use of open ended waveguide or coaxial line. This is a non-destructive
method of dielectric measurement in which the problem of sample preparation is
greatly reduced. Gardiol et al. [18,19] was the first to propose this method using
rectangular waveguide. Athey et al. 1171 have suggested the use of open ended
elliptical probe instead of a circular coaxial probe. They have pointed out that (i)
fabrication of such a probe is easier and (ii) the sensitivity of measurements is
improved. The open ended transmission line method is quite suitable for lossy
dielectric materials, as the infinite half space condition assumed in the analysis can
he. achieved more readily than with medium or low loss dielectric materials.
However, the ANA methods suffer from the following disadvantages.
( i ) In the case of hard and brittle solids, it is impossible to perfectly fit the
sample hetween outer and inner conductor.
(ii) Unless the sample is filling the entire co-axial line, it is very difficult to
define a reference plane which is necessary for the calibration of the
measurement system.
(iii) The presence of the sample may generate higher order modes in the
transmission line.
(iv) The initial capital cost of Hewlett Packard type ANA set-up is very high.
Therefore, methods are more useful for applications where data is required
over a wide band of frequencies and others are not convenient.
1.1.4 Cavity Perturbation Techniques
It was Bethe and Schwinger (201 who proposed cavity perturbation technique
for the first time. A material sample, when introduced into a cavity, alters its
characteristic parameters namely the resonance frequency (fr) and the loaded quality
factor (Q,). The changes depend upon the real and imaginary parts of the complex
permittivity or permeability and on the geometry of the sample and cavity. (Figure
1.5). In the past, most researchers have preferred simple geometrical shapes like
cylindrical 121) and rectangular cavities 1221. In such cases, the perturbation
equations are easier to derive. However, cavities having strong fields in some
regions have also been used to increase the sensitivity of measurement. For example
Sen et al. 1231 devised a re-entrant cylindrical cavity for complex permittivity
measurement (Figure 1.6).
SAMPLE
COUPLING PROBE
Figure 1.5(a) Cross-sectional view of cylindrical transmission type cavity resonator
SAMPLE -
PROBE
Figure 1.5( b ) Cross-sectional view of cylindrical reflection type cavity resonator
SAMPLE
PROBE
Fiqure 1 .6 R e - e n t r a n t c a v i t y .
1.1.5 Resonant Cavity Methods
ln the past few decades, the closed cavity resonator operated in the low-
attenuation TEol mode, which consists of a hollow metallic cylinder of circular
cross-section terminated with two short-circuited plates, has been used to measure
the complex permittivity of materials (especially low-loss solids) 124). Reviews of
literature on dielectric measuring methods including cavity resonator techniques have
been given by Bussy 1251, Lynch (261, Chamberlain and Chanhy [27] and more
recently by Birch and Clarke (281.
Resonant cavity methods can be classified into two categories.
1 . Length Tuning Method (LTM)
2. Frequency Tuning Method (FTM)
in the case of length tuning method, the real and imaginary parts of complex
permittivity are obtained at a fixed frequency, from the measured changes of the
length and of the Q-factor of the cavity at resonance. In order to tune the
resonator, the short-circuit can be moved in the axial direction to vary the resonator
length by a bearing and positioning device (Figure 1.7).
In frequency tuned method (FTM), no movable mechanical parts are used.
Here the resonator is terminated only by the two fixed short-circuits. At constant
resonator Length, the frequency is altered to tune it to resonance. The real and
imaginary parts of complex permittivity are measured from the changes of
resonance frequency and Q-factor of the cavity 1291.
MICROMETER HEAD
PISTON
DETECTOR PROBE
EXCITER PROBE
DIELECTRIC UNDER TEST
CAP
DETECTOR
DIELECTRIC SAMPLE
MICROMETER PLUNGER
Figure 1.7 Length tuning resonant cavity methods
Figure 1 . 8 Cross - sec t iona l view of t h e c a v i t y
Boifot 1301 devised a coaxial dielectrometer for broad band complex
permittivity measurement (Figure 1.8). In the resonant cavity method suggested by
Seaman et al. [31], the plunger position is adjusted for resonance with and without
the sample. Then the change in position gives dielectric constant and loss tangent is
found from the change in resonant current or from the width of the resonance curve.
In another method, the position of the dielectric sample can be adjusted by a
micrometer plunger. Power is coupled into or out of the cavity through two
coupling loops. Difficulty due to the variation of the minimum position may be
eliminated by using two travelling probes instead of coupling loops. In some better
methods 132,331 the two loops are located at the short circuit and hence are always
at a node. Then the length of the transmission line is changed, but the distance from
the sample to one end remains constant.
1.2 TIME DOMAIN METHODS
The time domain methods are modem techniques for the material
characterization. They are becoming popular very fast as wide band dielectric
measurements are required in many cases such as characterization of components
and in determining dielectric relaxations, in materials like muscle tissue, which
range from 100 Hz to 100 GHz. Till late 1960's, dielectric measurements on
materials were mainly carried out by means of frequency domain methods.
However, since then, due to the availability of fast sampling oscilloscopes and
tunnel diode step generators with picoseconds rise time, it has become possible to
use time domain techniques in the microwave region (Figure 1.9). Time domain
methods offer three distinct advantages over frequency domain methods. First, time
domain techniques are inherently of wide band as compared to the traditional
frequency domain methods. In the case of waveguide techniques, at least four
waveguide systems are required to measure the dielectric properties over 1 GHz to
10 GHz range, along with samples of different shapes and sizes compatible with the
waveguide dimensions for different bands. Second, the time required for
performing a time domain measurement valid over a wide range of frequencies is
considerably less than that required for multiple frequency domain measurements.
Hence time domain methods are more suitable for wide band measurements on
materials whose characteristic may not remain constant over a long period of time
(for example, biological materials). Thud, the instrumentation required for time
domain methods requires less capital investment as compared to an automated vector
network analyzer (ANA).
Time domain methods can be broadly classified into two categories.
(a) Reflection methods
(b) Transmission methods
Initially, reflection methods were used in electrical engineering for locating
the faults in the cables. These techniques are called "Time Domain Reflectometry"
(TDR) 1341. The measurement of reflection coefficient and the time delay gives the
type and distance of the fault from the secondary end. These faults could be a
break in the cable conductor, cable shield, or perhaps water in the cable. Van
Gemert 1351 used transmission methods for dielectric measurements and he discussed
the different variations on the time domain reflection and transmission methods.
Most researchers have preferred reflection methods to transmission methods. Of the
few who have used the transmission method, only Gestblom et al. 1361 have given
quantitative experimental results.
GENERATOR
SAMPLE HOLDER
Figure 1 . 9 Schematic representation of the experimental set-up for time domain tranamiasion measurement method
The reflection and transmission time domain techniques are collectively
called "Time Domain Spectroscopy" (TDS). Time domain measurement methods
have also been used for performing broad hand measurements on filters, directional
couplers, amplifiers and studying the electronic system response to lightning and
EMP [37].
Although TDS is hasically a time domain technique, the data is usually
converted to the frequency domain using a Fast Fourier Transform (FFT) analysis.
Fellner-Feldegg [38] suggested time domain data directly to obtain G , r, and the
dielectric relaxation time. Whittingham [39] and Van Gemert [40] pointed out
certain conceptual difficulties with that approach and indicated that only % and E,
can be accurately determined directly from the time domain data. Later Cole 1411
gave an analysis to compute complex permittivity directly from the time domain
waveforms. But this method can give unacceptable errors for polar materials.
Cormack et al. 1421 have suggested the use of Extended Function Fast Fourier
Transform (EF-FFT) for computation of the spectral components of an infinite
duration step-like waveform. Cormack has shown that the EF-FFT technique is
more accurate than the conventional techniques. Nozaki et al. 1431 have reported
that with the recent improvement it is possible to conduct precise measurements
from about 100 kHz to 25 GHz using this technique.
1.3 MISCELLANEOUS METHODS
For complex permittivity measurements, certain new techniques are
discussed in the literature (44-461. Scott et al. [44] suggested a method based on the
measurement of the input impedance of an antenna over a range of frequencies in a
standard medium of known permittivity like air. The antenna is then immersed into
the medium whose permittivity is to be determined and the impedance of the antenna
is measured over a range of frequencies. The measured impedance is then used to
calculate the complex permittivity of the material. This method has been used over
a frequency range of 50 MHz to 10 GHz and looks more suitable for liquids.
In the method described by Shimin 1451, a rectangular microsbip antenna has
been used for measuring the dielectric constant of thin slab substrates. Neelkanta et
al. 1461 suggested another method in which the sample is put on a planar conducting
surface in the form of a thin film. This structure is excited by a monopole and the
change in directivity due to the inclusion of the dielectric film on the conducting
sheet gives the dielectric constant of the film material. Bernard et al. 1471 suggested
a microstrip ring resonator technique for dielectric constant measurement. This
method is based on the fact that the effective permittivity will change if the alumina
or air boundary is modified by placing a dielectric material above the alumina
substrate, thereby changing the resonant frequency of the ring. The variational
calculation of the line capacitance is used to compute the effective permittivity of the
multilayer microstrip like ring resonator and hence the resonant frequency for
various test materials.
1.4 MOTIVATION OF THE PRESENT STUDY
Among the various methods discussed in the previous sections, the cavity
perturbation techniques are more accurate and precise than other methods. For I to
10 GHz range, rectangular or circular waveguide cavities are generally employed.
Earlier, the measurements were carried out at single frequency due to non-
availability of microwave sources covering wide frequency bands. With the
advancement of microwave technology, the synthesized sweep generators and vector
network analysers are available. Also these instruments can be interfaced with series
insbumentation computers. Although different waveguide cavities are required for
different bands, it is possible to measure the dielectric parameters at different
frequencies in single band with the help of modem equipments mentioned above.
So the major aims of the present work are development of modified cavity
perturbation techniques and their applications to material characterization.
1.5 BRIEF SKETCH OF PRESENT STUDY
The scheme of the work presented in this thesis is given below:
A review of important research work done in the field of cavity perturbation
techniques and its applications is presented in Chapter 2. Special emphasis is given
on the complex permittivity measurements.
In Chapter 3, the details of the design and fabrication of various resonators
employed for measurements are discussed.
In Chapter 4, the first part describes the theoretical analysis for the
measurements of complex permittivity and complex permeability of materials. The
second part deals with the measurement procedures.
The experimental results are discussed in detail in Chapter 5
The conclusions drawn from the sadies and further scope of the present
work are discussed in Chapter 6.
The experimental results of a few investigations in the related fields done by
the author are included in the thesis as four appendices. Appendix A deals with the
application of a cross-iris coupling for enhancing the loaded Q-value of rectangular
waveguide cavities. in Appendix B, the design and development of a Triple Comer
Reflector antenna are discussed. Appendix C describes the investigation of an
asymmetric hollow sectoral dielectric horn antenna. Appendix D deals with the
synthesis and characterisation of a ceramic dielectric resonator.
top related