University of Ljubljana Faculty of Mathematics and Physics ...mafija.fmf.uni-lj.si/seminar/files/2013_2014/... · University of Ljubljana Faculty of Mathematics and Physics Department

University of Ljubljana

Faculty of Mathematics and Physics

Department for Physics

Sources and detectors

in the microwave region

author:

Tadej Cigler

mentors:

Izred. prof. dr. Denis Arčon,

Dr. Andrej Zorko

Abstract

Electron paramagnetic resonance spectroscopy is a successful technique in determining the

structure and interactions of crystal atoms. It uses microwave radiation to induce

spectroscopic transitions of which it gets information about the local environment around

paramagnetic centres in solids. In more complex crystals we cannot get all the information

from the EPR spectrum while some of them are visible only in higher frequency ranges. With

experiments at high frequencies we can improve resolution of EPR transition curve and so

determine all the local interactions in matter. When we want to perform experiments in higher

frequency ranges (from hundreds of GHz to couple of THz) we need a spectrometer with its

suitable components. Two of most important components are the microwave source and the

microwave detector. First provide us desired high frequency signal and the second must be

capable of detecting the microwave radiation at corresponding frequencies. In this seminar we

will review some types of modern highly applicable sources and detectors which enable us

more extensive spectroscopic researches.

Contents

1. Introduction ............................................................................................................................ 3

1.1 Widespread use of microwaves ........................................................................................ 3

1.2 Motivation for experiments with electron paramagnetic resonance ................................. 4

2. EPR experimental set-up ........................................................................................................ 5

3. Millimetre-wave sources ........................................................................................................ 6

3.1 YIG resonance phenomena ............................................................................................... 7

3.1 Basic mechanism of YIG oscillator .................................................................................. 9

3.4 Modular Transmitter ....................................................................................................... 10

4. Millimetre-wave detectors .................................................................................................... 12

4.1 Indium-antimonide hot electron bolometer .................................................................... 12

5. Conclusion ............................................................................................................................ 14

6. Literature .............................................................................................................................. 15

1. Introduction

Microwave radiation is nowadays advantageously exploited for industrial and domestic

applications and it frequently plays an important role in basic and applied science [2]. Due to

their appropriate energy, they are applied in electron paramagnetic spectroscopy (EPR). The

EPR is sensitive to the materials which possess paramagnetic ions and provides information

about the local magnetic fields at the very high resolution. EPR experiments are performed in

the region from few to [3]. This requires the suitable multi-frequency microwave

source and the corresponding detectors. In applying EPR experiments we need low power

microwave sources ( output power) with stable signal (low phase noise at the input). To

achieve accurate detection we need not only low power signal detection ( ) and

low phase noise, but also high responsiveness ( ). When designing the experiment, one

has several options in selecting appropriate MW sources or detectors. The purpose of this

seminar is to review some modern high performance electronic components that are pushing

the detection limits of EPR spectroscopy to the higher levels.

1.1 Widespread use of microwaves

Microwaves, which are cause of their misleading name rather called millimetre-waves

(MWs), are located in EM spectrum between light waves and radio waves (Figure 1). Due to

their short wavelengths (1 millimetre – 10 centimetres), large bandwidth ( ),

absorption and reflection properties they provide unique opportunities for several useful

applications in communicational industry, basic/applied science, biology and food industry.

Figure 1: The electromagnetic spectrum. Millimetre-waves are located between light waves

and radio waves including frequencies from to [1].

Current relevance of MWs can be seen in their increasing use in modern communication

systems. Some newer systems that operate in millimetre-wave range are: personal

communication system (PCS), wireless local area computer network (WLAN) and global

positioning satellite (GPS) system. The heating ability of millimetre-waves is today broadly

applied in cooking (microwave oven) and industry (microwave drying machines).

However, it is essential to mention involvement of MWs in basic and applied science.

Interaction of electron beam with periodic millimetre-wave structures are used to design high

power linear accelerators for nuclear research. On the other hand, the absorption of MWs in

crystals that contain unpaired electrons is used to study their local features [2]. In these

applications crystal is inserted in magnetic field and absorption is observed when electron

spin states are separated by the quantum of MW energy. Described mechanism represents

basic part of electron paramagnetic resonance (EPR) spectroscopy (Figure 2).

Figure 2: (a) Linear magnetic field dependance of spin energy states representing the Zeeman splitting. In the

simple case of one unpaired electron the magnetic field will split the originally degenerate energy levels in

two. In EPR experiments we observe absorption when MW energy is matched with the splitting energy [4].

(b) Energy diagram where possible EPR transitions are shown (here single electron interacts with four

protons) [17].

1.2 Motivation for experiments with electron paramagnetic resonance

Electron paramagnetic resonance deals with materials which possess paramagnetic ions (ions

with unpaired electrons) and is therefore appropriate technique for studying them.

In chemistry field, EPR can provide us a wide range of information about molecular structure.

From EPR spectrum we can characterise free radicals, study organic reactions and investigate

electronic properties of paramagnetic inorganic reactions. In biology, EPR is mostly used to

study mechanisms and structures in biological cell such as enzyme reactions and complexes in

proteins. In accordance to achieve sufficient detection, high concentration of paramagnetic

ions is preferable (for standard EPR spin density: is sufficient). This is done by

increasing substance’s natural concentration or by spin labelling. Beside spin labelling,

another technique called spin trapping has proven to be very useful in biology. With it we

could provide detailed information about the structure and dynamics of transient radicals and

radical pairs.

In material science EPR is exploited for studying the influence of surroundings on electron

properties in solids. In studying solids without paramagnetic ions, diamagnetic host crystals

are doped with paramagnetic impurities like transition-metal or rare-earth ions. Such, they are

suitable for performing the EPR spectra measurement. Various features can be determined

from the spectrum, such as: magnetic properties, mechanism of conductive electrons, locating

defects in crystal, etc. [3]. To obtain more detailed information about these features the multi-

frequency EPR research should be carried out. Experiments at higher frequencies such as 35

GHz (Q –band) or 95 GHz (W-band) provide us better resolution of EPR active sites what is

very useful in chemical and biological researches (Figure 3). Experiments at multiple

frequencies are also important in studying magnetic materials, where by detection of

(anti)ferromagnetic resonance significant spin interactions could be reconstructed.

However, to perform these experiments we need a suitable spectrometer device. The most

basic spectrometer consists of millimetre-wave source, detector, electromagnet and resonator.

These components are selected according to requirements which we demand for our

experiment and are related to the frequency range, source power, detection sensitivity, noise

level, tuning ability, responsiveness, etc.

(a) (b)

Figure 3: Scheme of Zeeman splitting vs. magnetic field B in simple case of one unpaired electron, where

magnetic field will split the originally degenerated states in two. EPR experiments are carried out at fixed

frequencies from X and Q band where we observe transitions of electrons form one spin state to another. The

three pairs of lines represent the situations where was parallel to the z (solid lines), y (dotted lines) and x

(dashed lines) crystal axis. The right part of figure represents transmission curves (first derivative) where we

can see increase in resolution at higher frequencies (Q band). The constants: , and represent g-

factors [18].

In this seminar we will focus on the source and the detector as well as their functioning.

Particular attention will be given to the YIG solid state oscillator and the indium - antimonide

hot electron bolometer. At the beginning let us start with the quick overview of the EPR basic

principle trough the experimental set-up.

2. EPR experimental set-up

Assume we have a paramagnetic sample in the resonator, which is designed to resonate at a

specific millimetre-wave frequency like in Figure 4. The resonator is surrounded by

electromagnets that generate magnetic field in the sample. That influences on unpaired

electrons possessed by paramagnetic ions. Electron magnetic moments in the sample align

itself either parallel ( ) or antiparallel ( ) to the field direction, which

represents two separated energy states of electrons. The energy difference between separated

states (also called the splitting energy) is due to Zeeman effect directly proportional to the

applied magnetic field (1), (Figure 2.a, Figure 3),

. (1)

Here is the Landé g-factor and is the Bohr magneton. MW source in the Figure 4

generate MW radiation at fixed frequency. Such radiation is through the circulator directed to

the sample. By varying external magnetic field the splitting energy linearly change according

to the equation (1). When the splitting energy is matched with the quantum energy of MW

radiation transition from one spin state to another occurs (2).

Decrease in the amount of the microwave radiation that is being reflected out of the resonator

is observed by the detector. We get the resonance curve which represents absorption at

particular field strength (Figure 2.a).

. (2)

Here is the Planck constant and is the MW frequency. Typically EPR experiments are

performed in frequency range from S-band to W-band ( ) with

corresponding magnitudes of magnetic fields ( – ) [5].

(a) (b)

Figure 4: Sketch of EPR experimental set-up with its basic components [5].

The component named circulator is here used to insure that radiation from the MW source is

directed only to the resonator and that reflected radiation is directed only to the detector. This

is how absorbance spectrum is provided according to the basic principle of EPR.

In order to obtain a great deal of information from measurements, EPR experiments need to

be performed in the wide range of frequencies. One such example is measuring the frequency

dependence under applied magnetic field on Figure 5.

Figure 5: Frequency-field dependence of magnetic excitations in .

In order to observe magnetic excitations measurements are taken from 100 to 700 GHz [6].

In implementing these and other types of experiments we need a source that provides us the

desired multi-frequency range and the detector that operates in this specified range. Various

different microwave sources exist today such as Gunn diodes, backward-wave oscillators,

optically pumped molecular lasers, YIG oscillators and others. Besides the wide frequency

range ( ), suitable source in the EPR spectrometer must meet the following

requirements such as, low output power ( ), low phase noise and easy controlling.

In the majority of EPR experiments we measure the amount of radiation that is reflected out

of the sample. Due to this we need a detector which is capable of measuring the low power

signal ( ).

3. Millimetre-wave sources

Old spectrometer devices used klystrons as a source of millimetre- waves. Klystrons are

oscillators that work on a principle of amplifying the input MW signal in the vacuum tube.

The input MW oscillations are coupled with the beam of electrons accelerated by high-voltage

electrodes. At the output of klystron we get high power MW signal which is connected to the

input creating a feedback loop circuit oscillator. The great advantage of klystrons is that they

provide oscillating signal with significant power [19]. Commercially available klystrons

produce high power oscillations ( – ) in wide frequency range

( – ) with a small bandwidth ( ) [20]. Nowadays klystrons are applied

in several fields where high power MW signal is required (communications, particle

accelerations), but are not so popular in the spectrometry.

In almost all new spectrometer devices a variety of different sources are used. Due to various

experimental purposes we can divide them according to the output frequency range (Table 1).

Table 1: Sources of millimetre-wave signal divide by their output signal frequency.

20 - 200 GHz Up to 700 GHz 30-1300 GHz 0.25-7.5 THz 1.2-75 THz

Diodes Gunn diodes, YIG

- oscillators

Backward-wave

oscillators

Optically pumped

molecular lasers

Free-electron

laser

Selection of source that is suitable for our experiment firstly depends on the frequency range,

and secondly on the characteristic of the source system such as phase noise and ability to

control the output frequency. In EPR studies the X band region ( ) is the most

common because it is commercially available (magnetic fields up to 1 T are highly suitable,

cause they can be easily achieved with electromagnets). Thus the second group of sources in

Table 2 is a matter of interest.

One of suitable sources is YIG based system which provides us signals with frequencies up to

432 GHz and output power around (Figure 10). The great advantage of this source is that

it has low phase noise and linear tuning. YIG oscillator is based on yttrium iron garnet crystal

which oscillates at microwave frequencies when inserted in DC magnetic field. The

oscillation originates from the resonance phenomena of the YIG crystal. In order to get the

stable output signal of selected frequency YIG oscillator is phase locked with the controlling

signal. The phase locked circuit is called synthesizer and provide us primary frequency range.

With the help of the modular transmitter circuit we amplify and multiply the primary signal to

get desired output radiation frequency. The resonance phenomena, basic mechanism of YIG

oscillator and functioning of modular transmitter are explained in next three subchapters.

3.1 YIG resonance phenomena

Yttrium – iron garnet, Y3Fe2(FeO4)3, is a polycrystalline garnet which belongs to a group of

ferrite materials. For oscillating purposes it is design in the shape of sphere which is highly

polished (Figure 6). A crucial property of the YIG sphere is that its magnetisation

resonates at microwave frequencies when immersed in the external DC magnetic field.

Figure 6: Yttrium iron garnet sphere mounted on a rod in the oscillator fabricated by

Micro Lambda Wireless, INC. The diameter of the YIG sphere ranges from 10-30 millimetres.

For oscillation purposes YIG sphere is surrounded by a conductive loop [7].

In proper orientated crystal magnetic field inside of the sphere remains uniform what is

condition for proper resonance. The most ideal sphere with polished surface provides us the

narrowest possible resonance line width. The resonance phenomena are described below [8].

Let as assume that we have a YIG sphere positioned in the external DC magnetic field in

vertical direction (Figure 7.a). In the YIG crystal there are ions with unpaired

electrons. These electrons possess a magnetic moment and under applied external magnetic

field , they precess about with frequency . In equilibrium lies in the opposite

direction of . With a small radio frequency (RF) magnetic field which is polarized

perpendicular to the external field , we can tilt magnetic moment in such a way that it

makes an angle with . The RF disturbance generates a torque exerted on [8],

. (3)

Figure 7: (a) Precession of magnetic moment about the direction of magnetic field [8].

(b) Sketch of sphere where the precession of under applied field with frequency is seen [8].

This results in the precession of about the direction of external field seen in Figure 7.b. The

total magnetisation of crystal is a sum of all magnetic moments ,

∑ . (4)

Thus, in fact the total magnetisation is tilted by the angle and it precesses about the

external field with frequency. The precession of is represented by equation, [8]

. (5)

Here is the gyromagnetic ratio of free electrons. Because electron magnetic

moments interacts with the lattice, the direction of magnetisation vector relaxes back to the

direction (exponentially in time). Due to the precession moves in spiral way until it aligns

itself with . During this process, the circularly polarised MW field is made outside of the

sphere but dies out exponentially cause of damping. Precession could be maintained with

applying already mentioned signal which tilts . When the frequency of , coincides

with the natural precession frequency of the magnetization, the precession angle grows and

we can observe the resonance (Figure 8.a). From equation (6) we can see, that the natural

precession frequency is determined by the field strength .

. (6)

We can tune the desired output oscillating frequency of millimetre-waves by varying .

Relation between and output frequency is almost linear, what makes this kind of

oscillators so convenient and attractive (Figure 8.b). Leading cause for the deviation from

linearity lies in the complexity of the oscillators transforming network and tuning mechanism.

(a) (b)

Figure 8: (a) Resonant curves at various frequencies as a result of the evaluation of equation (5) [8].

(b) Frequency of oscillation vs. electromagnet biasing current. The dependence is very linear and increases at

a rate of 2.8 GHz/T obeying the equation (6). Nonlinearity at the lower part of dependence comes from

oscillator’s circuit and lies in the inductance of coaxial cables [8].

3.1 Basic mechanism of YIG oscillator

To form an oscillator from a resonator we add a conductive element (wire) in the shape of a

loop around the sphere (Figure 6, Figure 9.a). Now the role of external magnetic field

comes into play. does not just cause the precession of magnetisation inside of the sphere,

but it also induce electrical current in the wire which generate the required RF magnetic field

. According to the configuration on Figure 9.a, is orientated perpendicular to the .

The sphere with its surrounding loop in external magnetic field acts like a parallel LC circuit

that provide an oscillating signal with frequency . The loop has a role of inductor with

inductance and the sphere has a role of capacitor with capacitance . The charge flows back

and forth between the plates of the capacitor, through the inductor (Figure 9.b) [9].

Figure 9: (a) Sketch of coupling loop around the YIG sphere where external field is orientated in such

direction that enables proper operating of the LC circuit [10].

(b) Simple scheme of parallel LC circuit which provide an oscillating signal at the output [9].

The resonance occurs when inductive and capacitive impedances are equal in magnitude [9],

. (7)

This means that inductive and capacitive currents and are equal in size and opposite. The

total current is then minimal and impedance is maximal. We can express the

resonance frequency of LC circuit as,

√ . (8)

The output signal oscillates with frequency ,

(a) (b)

(a) (b)

, (9)

. (10)

The total treatment of the YIG coupled circuit is a bit more complicated thus only main

results are mentioned below. The voltage amplitude of a YIG coupled loop is actually

expressed with external RF susceptibility (the ratio between RF magnetic moment and

applied field: ) by equation

(

). (11)

According to equation (11) an impendence of a YIG coupled loop has the form: [8]

[

(

)]. (12)

In above equations (11) and (12), is magnetization precession frequency,

is the natural precession frequency of magnetic dipoles, is volume of the

sphere, is radius of the coupled loop, is current density and is Bloch-Bloembergen

relaxation time (time associated with any processes that disturbs or opposes the processional

motion). The resonance of the conductive loop by obeying equation (11) at various selected

frequencies can be seen in Figure 8a. With selecting the magnitude of in equation (12) the

output oscillating frequency is determined [8].

We must mention here, that LC assembly have its own resistance due to energy loses in the

LC circuit, what results in unstable output signal. This is solved with configuring the suitable

transistor in the circuit. Such transistor is developed with a negative resistance, what means

that that reflection coefficient of transistor is greater than unity: . The transistor

provides energy which compensate loses in the circuit.

YIG coupled loop with transistor’s circuit provide us oscillating signal only at single

frequency, what is not sufficient for our experimental purposes where multi frequencies are

desired. This condition is fulfilled with additional electronic devices: synthesizers, amplifiers

and multipliers, which are described in the next chapter.

3.4 Modular Transmitter

We can achieve multi-frequency range and wide bandwidth with a system which consists of

the primary MW source and components which amplify and multiply the primary radiation.

Such system is the modular transmitter whose components are usually connected by coaxial

cable used for transmission of MW radiation (Figure 10). Its main part is synthesizer which

is actually the source of millimetre-wave signal. The synthesiser provides a range of signal

(usually ~ 2 GHz) with combining operations such as: multiplication, division, sum and

difference on a signal from primary source (YTO). It is designed on a principal of phase-

locked loop which use secondary oscillator as a reference. The phase-locking works under

principle where frequency from YTO is divided and then compared to the reference in the

phase sensitive detector. Frequency is here divided because the reference frequency ( )

is several times smaller than the output frequency ( ).

Figure 10: Scheme of modular transmitter that provides us signal with frequencies up to 432 GHz

fabricated by Virginia Diodes. This model is design in such manner that the output is digitally controlled.

The reference signal is usually derived from a crystal oscillator which is very stable in

frequency (in VDI assembly this is Wenzel crystal oscillator (Figure 10)). Crystal oscillator is

based on quartz crystal (SiO2) which due to its piezoelectric properties vibrates under applied

DC voltage. Vibrations of piezoelectric quartz generate oscillating signal which depends on

its size and the way it is cutted [21]. Wenzel oscillator is the SC type what means that crystal

is doubly rotated and then cutted [22]. Thus with crystal oscillator the synthesizer can provide

us a phase-locked output frequency with 2 GHz tuning range [10]. YTO have very accurate

frequency output and an excellent phase noise moving from to in magnitude

(Figure 12.a). The phase noise represents random and systematic variations in the output

power of oscillator.

Signal with selected frequency is from the Synthesizer transmitted to an amplifier. The

amplifier generates the power output by combining the outputs of several low-power

amplifiers. An individual amplifier usually has a “distributed” or “traveling wave” topology.

A large frequency range is achieved by arraying individual transistors where each represents

capacitances between series of inductances (Figure 11) [12].

Figure 11: Circuit of four cell distributed amplifier [12].

To achieve frequencies from to we add frequency multipliers to the transmission

assembly. Doublers and triplers from VDI assembly are varactor multipliers based on planar

GaAs Schottky diode technology. In general, frequency multipliers exploit non-linearity in

susceptibility to generate higher harmonic signals from the input DC signal. In varactor

multiplier’s circuit the non-linear element is diode with voltage dependent capacitance. We

extract selected double or triple frequency from mixed harmonics via a band-pass filter [11].

The output power verses frequency at the output of varactor doubler and the modular

transmitter can be seen in Figure 12.b and Figure 12.c.

Figure 12: (a) Comparison of phase noise for different transistor’s configurations of YIG coupled assembly.

Phase noise is usually expressed in terms of dBc in a specified bandwidth at a specific frequency [7]. (b) The output power in frequency region of VDI varactor frequency doubler [16] and (c) VDI modular

transmitter [13].

4. Millimetre-wave detectors

The majority of MW detectors forming a part of EPR spectrometers measure power spectrum

of reflected radiation that comes from the paramagnetic sample. Changes in the power

spectrum at various magnetic fields allow observation of EPR spectroscopic transitions (with

power variations: ). Main detection systems currently in use are Schottky diodes

and hot electron bolometers (HEB). The great advantage of both detection systems is that

they can operate at wide frequency range, bolometers up to and Schottky diodes

between and . They differ in bandwidth: Schottky diodes possess a bandwidth

greater than while HEB only [3]. In the next subchapters we will get familiar

with basic functioning of hot electron bolometer based on indium-antimonide semiconductor.

4.1 Indium-antimonide hot electron bolometer

The hot electron bolometer is a detector which measures power spectrum of incident

electromagnetic radiation via resistance change of a temperature depended resistance. It

consists of absorber and measuring system. The incident radiation heats the absorber what

results in resistive change measured by Wheatstone bridge circuit (Figure 13). The bolometer

forms one of the four arms of the Wheatstone bridge. Before the detection a DC bias current

is applied to the circuit to raise the temperature of bolometer via Joule heating, such that the

resistance of bolometer is matched to that of other resistors . In such a manner is

varied with variable resistor until galvanometer obtains the null point. When we achieve

the equality the circuit is calibrated.

(a) (b)

(c)

Figure 13: Wheatstone bridge circuit which measure power of incident radiation. With variable resistor

DC supply current is set. Ampere-meter measures changes in current during exposing to millimetre-waves [2].

Before exposing to the microwave radiation, dissipated power in the bolometer is given by

(

)

, (13)

where is the DC biasing current. When the bolometer is exposed to the microwave

radiation, is adjusted to balance the bridge. The power dissipated in the bolometer is

(

)

. (14)

Here is the changed current trough the bolometer. Thus incident millimetre-wave power

can be calculated as [2]

(15)

Currents and can be read from a connected ampere-meter and so the incident microwave

power is measured (Figure 13).

To sum up, the incident microwave radiation changes the temperature of bolometer. Due to its

resistance change the bridge becomes unstable. The rebalancing of the bridge is done by

varying the DC power from a voltage source. The incident power can raise or reduce

resistance depending on the type of absorber. With varying the DC power (Joule heating)

we achieve that resistance changes back to the initial value: .

Figure 14: (a) Resistance as a function of temperature of the indium-antimonide (InSb) semiconductor in hot

electron bolometer which is fabricated by QMC Instruments Ltd [14].

(b) High purity undoped n-type InSb absorber manufactured as a toaster element [14].

Suitable detector for detecting millimetre-wave electromagnetic radiation is a hot electron

bolometer based on indium antimonide (InSb) semiconductor which was developed in 1963

by Kinch and Rollin (Figure 14.b) [15]. First HEBs were using doped semiconductors

covered with black colour to absorb the radiation. Their thermal response was slow

(a) (b)

and they were very limited in frequency bandwidth. InSb bolometer has

much higher thermal response ( ) due to its beneficial heat-transfer mechanism.

In the InSb semiconductor absorbed millimetre and sub-millimetre wavelength light only

heats free carriers and does not affect the lattice vibrations (phonons). When the carrier

density is sufficiently high, collisions between carriers increase. Collisions create an internal

equilibrium of carrier gas which distributes the radiation energy with its characteristic

temperature above that of the lattice. In other words, free electrons absorb the radiation and

because they are not coupled on the lattice, they can be heated beyond the lattice temperature

[15]. The speed of electrons is much faster than the speed of phonons, so the transfer becomes

faster what results in above mentioned high thermal response of detection.

Another feature that mainly originates from the heat transfer mechanism is high sensitivity.

Free carriers are able to absorb almost all radiated energy because their mobility is strongly

depended of absorbed energy. Due to this the conduction of energy between free carriers and

lattice is negligible what make them capable of detect even small amount of energy (For InSb

bolometer the lower limit is at 8 K ). What also make this detectors so sensitive

is their strongly temperature depended electrical resistance (Figure 14.a). The highest

sensitivity of InSb bolometer is achieved at cryogenic temperatures ( ) where most

rapid changes in resistance may occur and resistance can be measured accurately. Relevant

technical data of InSb HEB are represented in below Table 2.

Table 2: Technical data of InSb cooled hot electron bolometer.

Sensitivity is given in terms of noise equivalent power (NEP)

that represents the limit where signal is no longer detectable (single/noise ratio is one) [14].

Thermal response Sensitivity (NEP) Minimum detectable

power

Useful frequency

range 0.3μs (at 4.2 K)

[14]

2.23 pWHz-1/2

(at 1 kHz)

[14]

(the incident power at 1

kHz must be greater than

2.23*10-9

W)

3*10-13

W (at 8K)

[15]

<500 GHz

(With the help of

magnets the limit can be

raised up to 2.5 THz

[23])

5. Conclusion

In this seminar we have shown that InSb bolometer along with the YIG oscillator is a suitable

assembly for spectrometer measurements. What makes the YIG oscillator so convenient is its

stable millimetre-wave signal which can be linearly tuned. Currently, such modular

transmitters provide us up to frequencies with output power up to 1 W (30 mW – 1

W). That makes them applicable in various experiments that require low power millimetre-

wave source. Due to needs after these sources lots of companies deals with their development

and manufacturing (Giga-tronics, Micro lambda Wireless, Teledyne Microwave, etc).

On the other hand, we have the InSb bolometer, the detector with high sensitivity (minimal

detectable power is ) and high responsiveness ( ). Along with their

detecting circuit they are capable of detecting oscillating radiation up to . Their

interesting feature is that they are sensitive directly to the energy left in absorber. Thus they

can be used not only for charge particle and photon detection, but also for detecting non-

ionizing particles, and any sort of radiation.

6. Literature

[1] http://www.lightsources.org/imagebank/image/electromagnetic-spectrum, 17.4.2013. [2] M. L. Sisodia, V. L. Gupta, Microwave engineering, first edition, New Age International publishers, New

Delhi, 2005.

[3] Sushil K. Misra, Multifrequency Electron Paramagnetic Resonance, Wiley-WCH, Weinheim, 2011.

[4] Andrei L. Kleschyov, Philip Wenzel, Thomas Munzel, Journal of Chromatography B, 851, 12 (2007).

[5]

http://chemwiki.ucdavis.edu/Physical_Chemistry/Spectroscopy/Magnetic_Resonance_Spectroscopies/Electron_

Paramagnetic_Resonance, 18.4.2013.

[6] S. A. Zvyagin, J. Wosnitza, C. D. Batista, M. Tsukamoto, N. Kawashima, J. Krzystek, V. S. Zapf, M. Jaime,

N. F. Oliveira, Jr., and A. Paduan-Filho, Phys. Rev. Lett., 98, 047205 (2007).

[7] http://en.wikipedia.org/wiki/YIG_sphere#cite_ref-2, 8.4.2013.

[8] http://archives.njit.edu/vol01/etd/1970s/1975/njit-etd1975-001/njit-etd1975-001.pdf, 8.4.2013.

[9] http://en.wikipedia.org/wiki/LC_circuit , 8.4.2013.

[10] G. H. Bryant, Principles of Microwave Measurements, Revised edition, Peter Peregrinus Ltd., London,

1993.

[11] R.E. Miles, P. Harrison and D. Lippens, Terahertz Sources and Systems, Kluwer Academic Publishers,

Dordrecht, 2001.

[12] http://cas.web.cern.ch/cas/Denmark-2010/Caspers/AN-GT101A-

Microwave%20Power%20Amplifier%20Fundamentals%2008-10-27%20%20CAS2010.pdf, 9.4.2013.

[13] http://vadiodes.com/index.php?option=com_content&view=article&id=180&Itemid=16, ( VDI user guide),

25.4.2013.

[14] http://www.terahertz.co.uk/index.php?option=com_content&view=article&id=215&Itemid=594, (QMC

Instruments LTD user guide), 25.4.2013.

[15] M. A. Kinch, B. V. Rollin, Br. J. Appl. Phys., 14, 672 (1963).

[16] http://www.vadiodes, 25.4.2013.

[17] http://www.cardiff.ac.uk/chemy/epr/enhancement.html, 1.5.2013.

[18] S. Van Doorslaer, I. Caretti, I.A. Fallis, D.M. Murphy, Coordination Chemistry Reviews, 253, 2116 (2009).

[19] George Caryotakis, The Klystron: A Microwave Source of Surprising Range and Endurance, Linear

Accelerator Center, Stanford University, Stanford, 1998.

[20] http://www.directindustry.com/tab/klystron.html, 2.5.2013.

[21] www.leapsecond.com/pdf/an200-2.pdf, Fundamentals of Quartz Oscillators, Hewlett-Packard, 2.5.2013.

22 www.conwin.com pdfs at or sc for oc o .pdf, Choosing an AT or SC cut for OCXOs,Conor Winfield,

2.5.2013.

[23] Ali Rostami, Hassan Rasooli, Hamed Baghban, Terahertz Technology, Fundamentals and Applications,

Springer-Verlag Berlin Heidelberg, 2011.