Moulson A.J., Herbert J.M. Electroceramics: Materials, Properties, Applications

Подождите немного. Документ загружается.

coincides with the polarization. The eﬀective negative birefringence is speciﬁed

Dn, the magnitude of which depends on the applied ﬁeld strength E.

The birefringence is measured using apparatus of the type shown in

Fig. 8.10(a). He–Ne laser light (l ¼ 0.633 mm) is passed through the polarizer

and then through the electroded specimen. The specimen is in the form of a

polished plate, of thickness t typically 250 mm, carrying gold electrodes with a

gap of approximately 1 mm.

With no voltage applied to the electrodes an unpoled specimen is in the

isotropic state and the light is passed on unchanged to the second polarizer P

where it is extinguished, as measured by the photodiode. Any ﬁeld-induced

relative retardation G between the horizontal and vertical components of the

plane light that takes place during passage through the specimen leads to

ellipticity and so a part of the light is transmitted by P

The Babinet compensator is a calibrated optical device comprising two wedge-

shaped pieces of quartz arranged so that their optic axes, and the light path, are

452 ELECTRO-OPTIC CERAMICS

Fig. 8.10 (a) Optical system for measuring the electro-optic coeﬃcient; (b) reference planes

viewed along the optical axis from the photodiode.

TEAMFLY

Team-Fly

mutually perpendicular. By moving one wedge relative to the other in the sense

indicated, variable and known amounts of relative retardation can be introduced

between the vertical and horizontal components of the plane-polarized light.

Therefore the relative retardation introduced by the specimen can be measured

directly from the adjustment of the calibrated compensator necessary to return

the situation to complete extinction. It follows that

G ¼

D nt ¼

Et ¼

t ð8:44Þ

For h ¼ t and G ¼ l=2, U ¼ V

¼l=n

, i.e. the half-wave voltage referred to

in Section 8.1.3 and Table 8.1.

An expression is now derived for the light intensity transmitted by P

in the

case where the relative retardation introduced by the specimen is G and the

compensator is removed. It is helpful to refer to Fig. 8.10(b) which represents

the system viewed from the photodiode. If the amplitude transmitted by the

polarizer P

is a sin(ot), then the x and y components incident on the PLZT are

x ¼

sinðotÞð8:45Þ

y ¼

sinðotÞð8:46Þ

On passing through the uniaxial negative PLZT the x component is retarded by

an angle d ¼ðG=lÞ2p, where G ¼ t

Dn. The amplitude a

transmitted by P

¼ y cos 45



 x cos 45



ðy  xÞ

i.e.

fsinðotÞsinðot  dÞg

which reduces to

¼ a sin





cos



ot 



ð8:47Þ

The time-averaged transmitted intensity I

is given by

¼ a

sin





cos



ot 



i.e.

LANTHANUM-SUBSTITUTED LEAD ZIRCONATE TITANATE 453

sin





sin





ð8:48Þ

where I

is the intensity incident on P

. I

is a maximum when G ¼(2m+1)

l=2, d ¼ð2m þ 1Þp and a minimum when G ¼ mlðd ¼ 2mpÞ, where m is an

integer.

The total polarization in the specimen for a given applied ﬁeld can be

measured by the voltage developed across the series capacitor, because

capacitors in series each carry the same charge. Closing the switch discharges

the specimen and capacitor; opening it charges the specimen to a voltage close to

U, provided that C is very much greater (at least a factor of 10

) than the

specimen capacitance. This allows the determination of hysteresis loops

corresponding to the birefringence–ﬁeld or birefringence–polarization relations.

8.2.3 Electro-optic characteristics

Depending on composition PLZT ceramics display one of three major types of

electro-optic characteristic, i.e. ‘memory’, ‘linear’ or ‘quadratic’. These are shown

in Fig. 8.11, together with the corresponding hysteresis loops, and are discussed

brieﬂy below.

Memory

A typical PLZT exhibiting memory characteristics has the composition 8/65/35

and a grain size of about 2 mm. The general approach to obtaining a hysteresis

loop is referred to above; the intermediate remanent polarization states (e.g. P

etc. in Fig. 8.12) necessary to generate the Dn versus normalized polarization

plots (Fig. 8.11(a)(ii)) are obtained as follows.

The polarization is saturated by applying a ﬁeld of approximately 3E

, where

is the coercive ﬁeld. When the saturating ﬁeld is removed, the polarization

relaxes to the remanent states P

or P

. Intermediate remanent states, such as

, can be achieved by removing the ﬁeld after bringing the electrical state of the

material to C, applying an appropriate voltage for a limited time and controlling

the charge movement by means of a high series resistance. There is a state D on

Fig. 8.12 from which the polarization relaxes to zero. Intermediate states can be

reached equally well from the positive E quadrants, and the right-hand side of

Fig. 8.11(a)(ii) reﬂects the observed hysteresis between the two approaches. To

achieve zero birefringence, coupled with zero polarization, it is usually necessary

to depole a specimen thermally. Thermal depoling involves heating the specimen

above the Curie temperature and cooling in the absence of a ﬁeld, a procedure

which randomizes the domain conﬁguration.

454 ELECTRO-OPTIC CERAMICS

The reduced polarization states P

, P

etc. are due partly to 1808 domain

reversals, which do not aﬀect the birefringence, and partly to changes in other

domains (718 and 1098 in rhombohedral crystals) which reduce both the

polarization and the birefringence. The relation between birefringence and

remanent polarization is therefore complex and, as indicated in Fig. 8.11, zero

remanent polarization does not necessarily correspond to zero birefringence.

The existence of intermediate polarization states is not of great practical use

because the eﬀect of an applied ﬁeld depends on both its exact mode of

application and the previous treatment of a specimen. Reproducible results can

be achieved by having an electrode and circuit arrangement that allows the

LANTHANUM-SUBSTITUTED LEAD ZIRCONATE TITANATE 455

(b) Linear

(a) Memory

Fig. 8.11 Hysteresis and electro-optic characteristics of the three main types of PLZT: (a)

memory; (b) linear; (c) quadratic (after G.H. Haertling (1971) J. Am. Ceram. Soc., 54, 303).

application of ﬁelds in two approximately perpendicular directions. This permits

the establishment of two optically distinct states by applying brief pulses to the

electrodes. These states are stable in the absence of an external ﬁeld.

Linear

Compositions that exhibit the linear electro-optic eﬀect (Pockels eﬀect) tend to

come from the PbTiO

-rich end of the solid solution range (Fig. 8.8). As

expected, high PbTiO

favours tetragonal distortion from the cubic; for example,

for 8/60/40 c=a ¼ 1.002, whereas for 8/40/60 c=a ¼ 1.01. A consequence of high

tetragonality is high coercivity.

To exhibit the Pockels eﬀect, the ceramic, e.g. 8/40/60, is ﬁrst poled to positive

saturation remanence P

with a ﬁeld of approximately 3 MV m

, i.e. about

twice the coercive ﬁeld (cf. the left-hand side of Fig. 8.11(b)(i)).

Dn is then

measured, at remanence, for positive and negative d.c. ﬁelds ranging from

71.4 MV m

to 2.0 MV m

. Over this range linearity is excellent (right-hand

side of Fig. 8.11(b)(ii)) with a Pockels coeﬃcient r

of approximately 100 pm V

(Table 8.1). Some PLZT compositions have higher r

values, e.g. 8/65/35

(Table 8.1), but at the expense of departures from linearity for small ﬁelds. Grain

size is known to have a very signiﬁcant eﬀect on the linearity of the

Dn–E

characteristic.

Quadratic

Compositions that display the quadratic Kerr eﬀect lie close to the ferroelectric

rhombohedral–tetragonal boundary (Fig. 8.8); 9.5/65/35 is typical. At room

temperature they are essentially cubic, but the application of a ﬁeld enforces a

transition to the rhombohedral or tetragonal ferroelectric phase, and the optical

456 ELECTRO-OPTIC CERAMICS

Fig. 8.12 Establishment of intermediate polarization states P

, P

etc.

anisotropy increases with E

. Because of their zero, or very low, remanence

values, they are known as ‘slim-loop’ materials (left-hand side of Fig. 8.11(c)(i)).

Longitudinal effects

Strain bias

In the examples so far discussed, birefringence has been developed by electrically

inducing a polarization in the plane of the PLZT plate, and this eﬀect is the most

widely exploited. However, it is possible to engineer a longitudinal eﬀect in a

number of ways, one being by strain biasing. A suitable strain can be introduced

by cementing a PLZT plate to an inactive transparent plate, e.g. Perspex, and

ﬂexing the combination so that tensional strain of order 10

is induced in the

PLZT, as illustrated in Fig. 8.17(a) below. This results in an orientation of the

polar axes of the crystallites approximately in the direction of the strain and in an

accompanying birefringence. The polar axes can now be switched through 908 by

the application, via ITO transparent electrodes, of a ﬁeld across the thickness of

the plate, which reduces the birefringence to zero. The PLZT will revert to the

birefringent state when the ﬁeld is removed.

Scatter mode

An 8.2/70/30 composition is antiferroelectric at zero ﬁeld but becomes

ferroelectric when ﬁelds greater than 1 MV m

are applied. As a consequence

the hysteresis loop has a narrow region at low ﬁelds which develops into a

normal saturating characteristic at higher ﬁelds (Fig. 8.13). This behaviour is

temperature sensitive but occurs to a suﬃcient extent for practical application

between 0 and 40 8C.

The change to the ferroelectric state is accompanied by the development of

birefringence (D n 0.05) and a marked scattering of light, particularly when

the grain size lies in the 10–15 mm range. A ferroelectric domain structure only

becomes apparent when the ﬁeld exceeds about 2 MV m

. There is uncertainty

regarding the extents to which domain boundaries and grain boundaries

contribute to the overall scattering eﬀect.

Hot-pressed material has a grain size of about 2 mm which can be increased by

annealing at 1200 8C in a PbO atmosphere provided by burying the specimen in

powder of similar composition. Grain growth follows the relation

G  2.0t

0.4

ð8:49Þ

where G (in microns) and t (in hours) are the grain size and time respectively.

The scattering eﬀect is exploited in the longitudinal mode, i.e. with the light

and applied ﬁeld directions collinear. The major faces of the plate are coated with

ITO electrodes and antireﬂection layers. The arrangement of the device is shown

in Fig. 8.14.

LANTHANUM-SUBSTITUTED LEAD ZIRCONATE TITANATE 457

458 ELECTRO-OPTIC CERAMICS

Fig. 8.13 Variation in P versus E hysteresis with temperature for 8.2/70/30 PLZT.

Fig. 8.14 PLZT used in the scattering mode.

Scattering has its maximum eﬀect in reducing the light ﬂux over a narrow

angle, typically 38; therefore it cannot be exploited in wide-angle devices. The

‘scatter ratio’, i.e. the ratio of the detected intensity in the transparent state to

that in the scattering states, can be as high as 1000 for a plate 1 mm thick and an

applied voltage of 2000 V.

As might be expected, scattering increases with specimen thickness since the

number of scattering interfaces is thereby increased. However, for a given

thickness there is a peak in scattering at a certain crystal size, since larger

crystals, although having greater birefringence, lead to a smaller number of

interfaces in a given thickness.

8.3 Applications

Brief descriptions of a selection of applications illustrating the various electro-

optic eﬀects and demonstrating ways in which they can be exploited are given

below.

8.3.1 Flash goggles

Flash goggles designed to protect the eyes of pilots of military aircraft from the

eﬀects of nuclear ﬂash are in production. Similar devices could also be of value to

arc-welders, to those who suﬀer eye disorders and as part of television stereo-

viewing systems.

The elements of ﬂash goggles are a polarizer, a PLZT plate and a second polarizer

arranged as shown in Fig. 8.15. The PLZT composition, 9.2/65/35, can be prepared

by hot-pressing to a very pale yellow transparent ceramic with a grain size of about

APPLICATIONS 459

Fig. 8.15 (a) Arrangement for optical shutter; (b) detail of the electroding conﬁguration.

2 mm. It is a slim-loop quadratic material with almost zero birefringence under zero

applied ﬁeld. The R value (Table 8.1) is about 4610

716

For normal vision the light incident on the PLZT plate from a polarizer has its

plane of polarization turned through 908 by the birefringence induced by a

voltage (about 800 V) applied to the interdigitated electrode system (Fig. 8.15(b)).

An analyser is so oriented as to allow the emergent light through to the wearer’s

eyes. A ﬂash of intense light is detected by a photodiode which operates a switch

that removes the voltage and closes the circuit between the electrodes. The plane

of polarization is no longer rotated and the analyser now stops most of the light.

As ﬁnally designed, the PLZT is in the form of a plate 0.4 mm thick with

corresponding grooves on both faces containing gold electrodes with gaps of

about 1 mm between the interdigitated arrays. The embedded electrodes improve

the ﬁeld intensity distribution, and therefore the birefringence, across the

thickness of the plate.

Switching times between the ‘on’ and ‘oﬀ’ states are typically 100 ms and the

transmission ratios 1000:1.

8.3.2 Colour ¢lter

The optical shutter described above can function as a voltage-controlled colour

ﬁlter. The PLZT is again of the slim-loop quadratic variety, and the polarizer–

PLZT–analyser conﬁguration as for the shutter.

When no voltage is applied, incident white light is extinguished. As the voltage

is increased, the retardation reaches a value for which the mid-spectrum ‘green’ is

retarded by l/2 and so is fully transmitted; the remainder of the partially

transmitted spectrum, along with the green, give the eﬀect of approximately

‘white’ light. As the voltage is increased further, full-wave retardation for the

shortest-wavelength primary colour, blue, is reached, and so that colour is

extinguished. The other two primaries, red and green, are transmitted, to give the

eﬀect of yellow light. In the same way, as one of the other two primary colours is

fully or partially extinguished, the remainder of it, together with the

complementary two primaries, is transmitted.

8.3.3 Display

A PLZT reﬂective display is similar in appearance to the common liquid crystal

display (LCD). The structure of the device is shown schematically in Fig. 8.16; a

suitable PLZT composition is the slim-loop quadratic 9.5/65/35.

When the polarizers P

and P

are in the parallel position and no ﬁeld is

applied to the ITO electrode pattern, incident unpolarized light passes through

. The plane-polarized light then passes through the ITO electrode–PLZT

460 ELECTRO-OPTIC CERAMICS

combination and the parallel P

, and then, after reﬂection, passes back through

the system to be observed. When a ﬁeld is applied to the electrode pattern so as

to induce a l/2 retardation for a single passage through P

, the light is

extinguished at P

. The activated segments then appear black against a light

background. If P

and P

are arranged to be in the crossed position, the

characters appear white against a dark background. The segmented alpha-

numeric electrode pattern is of the interdigitated form described above.

The disadvantages of the PLZT device, compared with the LCD, are high

operating voltages (typically about 200 V) and high cost; an advantage is the fast

switching times.

8.3.4 Image storage

PLZT can be exploited for image storage in a variety of ways: the ﬁrst of the

following two examples makes use of strain-induced birefringence and the

second light scattering (Fig. 8.14).

A strain-biased antiparallel domain conﬁguration is induced in the plane of a

memory-type PLZT (7/65/35 with a grain size of 1.5 mm) plate approximately

60 mm thick, as shown in Fig. 8.17(a). A strain of 10

is achieved by bonding the

PLZT to a thicker plate of transparent Perspex and bending the combination so

that the ceramic is in tension in the convex surface. Combining a photo-

conductive layer, e.g. polyvinyl carbazole, CdS, ZnS or selenium, with the PLZT

plate allows an optical pattern to be electrically stored, read and erased.

Transparent ITO electrodes sandwich the photoconductor–PLZT combination.

When a light pattern illuminates the front face (Fig. 8.17(b)), the light passes

through the transparent electrode to interact with the photoconductive layer,

raising its conductivity. Simultaneously a voltage pulse (about 200 V) is applied

across the electrodes, the major part of which, in the illuminated regions, drops

across the ferroelectric. As a result the polarization is switched from the strain-

biased conﬁguration to a direction approximately normal to the plate. In this

APPLICATIONS 461

Fig. 8.16 Principle of the PLZT reﬂective display.