close

Вход

Забыли?

вход по аккаунту

?

el%3A19860403

код для вставкиСкачать
Dr. A. Preece of the Oncology Research Unit, Bristol Royal
Infirmary, for observations with the thermographic camera.
R. H. JOHNSON
Wolfson RF Engineering Centre
Royal Military College of Science
Shrivenham, Swindon SN6 8LA, United Kingdom
15th April 1986
References
1
2
KANTOR, G.: 'Evaluation and survey of microwave and radiofrequency applicators', J. Microwave Power, 1981,16, pp. 135-150
PAGLIONE, R., STERZER, F., MENDECKI, J., FRIEDENTHAL, E., a n d BOT-
STEIN, c : '27 MHz ridged waveguide applicator for localised
hyperthermia treatment of deep seated tumours', Microwave J.,
1981, 24, pp. 71-80
3
ANDERSON, J. B., BAUN, A., HARMARK, K., HENZEL, L., RASHMARK, P.,
and OVERGAARD, j . : 'A hyperthermia system using a new type of
inductive applicator', IEEE Trans., 1984, BME-31, pp. 21-27
4
5 MHz nominal resolution. The acoustic beam spreading
from the transducer is such that the centre of the optical
aperture for 5 MHz resolution lies within the near field of the
transducer. For an acoustically anisotropic medium such as
gallium phosphide the acoustic power flow distribution is
modified by the shape of the inverse velocity surface. It may
be either collimated or defocused. The power flow direction
for a given wave vector direction is normal to the inverse
velocity surface for that direction.4 If the inverse acoustic
velocity surface about a particular direction is approximated
by a power series,2-5 such that
FRANCONI, C , TIBERIO, C. A., RAGANELLA, L., a n d BERNOZZI.L.: ' L o w
frequency RF twin dipole applicator for intermediate depth hyperthermia', ibid., 1986, MTT-34 (to be published)
(6)
^
(
bB2
where (1/K)(0) is the value of inverse velocity for a propagation direction at an angle 8, then the spread of acoustic power
flow directions AO^ from a transducer is related to the spread
of wave vector directions by A ^ = (1 — 2b) A0.
The transducer height is adjusted to take account of this
modification.
The diffracted signal level for a particular input acoustic
power level is given by
where Ia is the acoustic power. For small diffracted signal
levels the diffraction efficiency is given by
Indexing terms: Semiconductor devices and materials, Bragg
cells
The design, fabrication and evaluation of efficient wideband
Bragg cells in gallium phosphide is reported. The devices
were fabricated using the <110> and <111> longitudinal
modes of acoustic propagation with design bandwidths of
500 MHz centred on 1-2 GHz. State-of-the-art efficiencies of
up to 115%/W and 220%/W, respectively, were obtained.
Introduction: High-efficiency bulk-wave Bragg cells are being
used in various RF signal processing applications, particularly
in spectrum analysis. In many cases, system dynamic range is
limited by two-tone third-order intermodulation products in
the drive amplifiers used in the spectrum analyser. By the use
of Bragg cells with as high an efficiency as possible the effect
of these nonlinearities may be minimised.
Gallium phosphide Bragg cells have been described by a
number of authors, 1 " 3 having been used both with shear and
longitudinal acoustic modes. The material has been used in
two longitudinal acoustic modes, both of which have high
acousto-optic figures of merit M2. The material, however, has
an acoustic attenuation which limits its use to operation with
moderate acoustic apertures and at frequencies below many
gigahertz. The material exhibits, at present, a fairly high level
of optical scatter. This does not, however, present a problem
in use in devices for interferometric spectrum analysers.
Device design: Bragg cells were designed using the <110> and
<111> longitudinal acoustic modes in gallium phosphide, to
have 500 MHz bandwidths centred on 1-2 GHz, and for operation at 633 nm.
For isotropic cells, the transducer length L is calculated7
from the acoustic wave vector spread from the transducer
required to satisfy phase-matching conditions (wave vector
addition) across the device bandwidth.
The relationship between transducer length L, device bandwidth A/ and centre frequency / is given, to a first approximation, by
Lfl
where n is the refractive index, V is the acoustic velocity and X
is the optical wavelength.
The transducer height H (dimension orthogonal to interaction plane) is determined by the resolution required and
acoustic beam spreading. The devices were designed for a
Vol. 22
e
x[-2H-) \
HIGH-EFFICIENCY BRAGG CELLS IN
GALLIUM PHOSPHIDE
ELECTRONICS LETTERS 22nd May 1986
l
No. 11
M2L
Device design data are shown in Table 1. The values of the
acoustic velocity, acoustic anisotropy factor and M 2 are calculated using data from Reference 8. The value of M 2 is normalised with respect to silica. The calculated diffraction efficiency
assumes no loss in the transducer and negligible acoustic
attenuation.
Table 1
Acoustic
direction
Calculated
velocity
Optical
direction
M2
<110>
<111>
ms" 1
6493
6692
<110>
<112>
28-1
29-3
Calculated
value
(1 - 2b)
L/H
1-432
0-611
400
6-25
Calculated
diffraction
efficiency
%/W
107
176
The transducers are designed using the approach described
in Referenced and are designed to be matched using parallel
bondwire tuning.
Device fabrication: The two types of Bragg cell blank were cut
and oriented from the same slice of a gallium phosphide
boule. These were machined to shape and polished to flatness.
Three of each type of cell blank were assembled into a
polishing/bonding jig and processed as described in
Reference 6. The transducers, which are 36 Y-cut lithium
niobate, were bonded, and machined to a final thickness of
approximately 3 /im. Gold electrodes were evaporated on to
the transducers. The Bragg cells were mounted into packages
and individual transducers were connected by ultrasonic wire
bonding to microstrip feeds.
Testing and results:
(i) Efficiency and bandshapes: The devices were tested using an
HeNe laser with beam-forming optics to waist light into the
Bragg cell, such that the height of the optical beam was equal
to that of the transducer, and the width of the optical beam
was appropriate to that for the required resolution of the
device. The beam input polarisation is in the interaction
plane. The Bragg angle was adjusted such that maximum diffraction efficiency was obtained, and the diffracted beam was
imaged on to the photodetector with a spatial filter in the
Fourier transform plane to remove the zero-order scattered
light.
593
The bandshapes were displayed on an oscilloscope and
recorded photographically. Figs. l(i) to (iii) are the bandshapes for the Bragg cells using the <110> propagation mode.
(ii) Acoustic anisotropy: The acoustic anisotropy factor was
determined by imaging the acoustic beam using schlieren techniques. An image of the acoustic beam in one of the cells is
shown in Fig. 3. From these the values of the acoustic anisotropy factor in Table 3 were obtained.
Table 3
Fig. 1 Bragg cell acousto-optic bandshapes, <//0> acoustic mode
Frequency scale = 200 MHz/division, centre = 1-2 GHz
Figs. 2(i) to (iii) are the bandshapes for the cells using the
< 111 > propagation mode. As the Bragg cells were not optically coated, the high refractive index of gallium phosphide
(n = 3-31) leads to a transmission loss of 1-47 dB at each
optical surface. Also the material has an appreciable optical
absorption at 633 nm (which depends in detail on the doping).
For the material used in this work the optical attenuation was
measured to be 1-88 dB/cm at 633 nm.
(i)
|615/2 |
s
Fig. 2 Bragg cell acousto-optic bandshapes, (III )
acoustic mode
Frequency scale = 200 MHz/division, centre = 1-2 GHz
The apparent device efficiencies were measured at low diffracted signal levels and were normalised to the intensity of
the transmitted zero order to take account of reflection and
optical absorption losses in the Bragg cell. Results of the measurements are summarised in Table 2.
Table 2
Acoustic
mode
<H0>
<111>
<111>
<111>
Bandwidth
MHz
505
533
503
519
520
530
Centre frequency
Normalised
efficiency
MHz
1170
1234
1183
%/W
115
78
81
1163
1160
1165
194
200
210
The Bragg cells are highly efficient with the < 111 > devices,
having approximately twice the efficiency of the <110>
devices. Centre frequencies and bandwidths are easily reproducible. The efficiencies of the <110> devices are comparable
with the calculated value. The efficiencies of the <111) devices
are consistently greater than those calculated, implying that
M 2 for the <111> propagation direction is greater than the
calculated value.
Acoustic
mode
I
Calculated
<no>
<m>
1-432
0-611
- 2b
Measured
1-40
0-60
Conclusions: Very high efficiencies have been obtained for
500 MHz-bandwidth gallium phosphide Bragg cells. Devices
using the <111> acoustic mode exhibited approximately twice
the efficiency of devices using the <110> mode. This advantage
accrues in part from the more favourable acoustic anisotropy
factor of the <111> mode. There are indications that M 2 for
the < 111 > mode is greater than the calculated theoretical
values. The bandshapes of the devices were easily reproducible.
To our knowledge the efficiencies for the < 111 > mode are
the highest reported for 500 MHz bandwidth, single transducer Bragg cells, and the bandwidth efficiency product is the
highest ever measured.
The acoustic anisotropy factors of the material were determined for the two orientations and found to be in good agreement with theory.
Acknowledgments: This work has been carried out with the
support of Procurement Executive, UK Ministry of Defence,
sponsored by DCVD.
J. M. BAGSHAW
24th March 1986
S. E. LOWE
T. F. WILLATS
GEC Research Limited
Marconi Research Centre
West Hanningfield Road
Great Baddow, Chelmsford, Essex CM2 8HN, United Kingdom
References
1
2
3
CHANG, i. c , CADIEUX, R., and PETRIE, c : 'Wideband acousto optic
Bragg cells'. Proc. 1981 ultrasonics symposium, pp. 735-739
HECHT, D. L., and PETRIE, G. w.: 'Acousto optic diffraction from
acoustic anisotropic shear modes in gallium phosphide'. Proc.
1980 ultrasonics symposium, pp. 474-479
BONNEY, R., ZEHL, o., ROSENBAUM, J., and PRICE, M. G.: 'Performance
and optical characterisation of efficient wideband gallium phosphide Bragg cells', Appl. Opt., 1984, 23, pp. 2778-2783
4 AULD, B. A.: 'Acoustic fields and waves in solids, Vol. 1' (John
Wiley & Sons, 1973)
5 COHEN, M. G.: 'Optical study of ultrasonic diffraction and focusing
in anisotropic media', J. Appl. Phys., 1967, 38, pp. 3821-3828
6 BAGSHAW, J. M., and WILLATS, T. F.: 'Anisotropic Bragg cells', GEC
J. Res., 1984, 2, pp. 96-103
7 CHANG, I. c : 'Acousto optic devices and applications', IEEE
Trans., 1976, SU-23, pp. 2-21
8 LANDOLT-BORNSTEIN: 'Elastic, piezoelectric, pyroelectric, piezo
optic and electro optic constants of crystals, Vol. 11' (SpringerVerlag, 1979)
InGaAsP/lnP MONOLITHIC INTEGRATED
CIRCUIT WITH LASERS AND AN OPTICAL
SWITCH
Indexing terms: Integrated circuits, Integrated optics
Fig. 3 Schlieren image of acoustic beam in gallium phosphide Bragg cell
594
InGaAs/InP monolithic integrated circuits composed of a
compact carrier-injection optical switch and distributed feedback laser diodes are fabricated. These integrated circuits
have a variety of functions, such as monolithic modulators,
switches and optical amplifiers for optical communication
systems.
ELECTRONICS LETTERS 22nd May 1986 Vol.22
No. 11
Документ
Категория
Без категории
Просмотров
0
Размер файла
397 Кб
Теги
3a19860403
1/--страниц
Пожаловаться на содержимое документа