Growth of single crystals of lead monoxide

Growth of single crystals of lead monoxide

Journal of Crystal Growth 47 (1979) 568—572 © North-Holland Publlshing Company GROWTH OF SINGLE CRYSTALS OF LEAD MONOXIDE K. OKA, H. UNOKI and T. SAK...

444KB Sizes 0 Downloads 22 Views

Journal of Crystal Growth 47 (1979) 568—572 © North-Holland Publlshing Company

GROWTH OF SINGLE CRYSTALS OF LEAD MONOXIDE K. OKA, H. UNOKI and T. SAKUDO Electrotechnical Laboratory

“,

Tanashi, Tokyo 188, Japan

Received 19 March 1979; manuscript received in final form 17 May 1979

Large single crystals of the orthorhombic phase (yellow) and the tetragonal phase (red) lead monoxide have been grown by the top-seeded solution growth method with titanium oxide as solvent. Single crystals of the lower temperature tetragonal-phase up to 100 g have been obtained from a 6% Ti0 2—PbO solution. The higher temperature orthorhombic-phase crystal, as a boule of about 130 g, has been obtained by simply pulling the crystal from the melt, followed by rapid quenching through the phase transition temperature during cooling. The limiting concentration for growth of these two phases was found to be ‘—4—-S mole% Ti02. Optical transmission, dielectric constant and etch pattern observation experiments have been performed on crystals of both

phases.

1. Introduction

very easy to cleave along the c-plane. Early experiments on the growth of PbO have

LEAD monoxide (PbO) is known to show a polymorphic transition. At room temperature, it is in the tetragonal form with a transparent dark-red colour, called Litharge. Above about 500°C it is in the orthorhombic modification with a yellow.greenish colour, called Massicot. Crystallographic structures of both phases have been estabhshed from X-ray and neutron scattering experiments [1,2]. The phase transformation is very sluggish in nature, and the range of thermal hysteresis seems to depend on the size of the PbO crystallites. Fine powder specimens of the orthorhombic phase are very stable, and do not convert into the tetragonal phase even down to liquid hthAtroote:

been carried oot by the slow cooling method from the melt [3] and by the hydrothermal method [4]. In these experiments, only relatively thin platelet crystals of up to 0.5 mm thickness or relatively small (at most 8 X 10 X 3 mma) crystals, respectively, were synthesized. The Czochralski method does not seem to have been used for the crystal growth of this material, possibly because the lead oxide is rather

--

-

2

--

~

2

I

inga small stress. The two crystallographic structures are given in fig. 1 [1,2]. As seen in the figure, the tetragonal structure is of two.dimensional character in the c-plane, and the orthorhombic structure is composed of chains along the a-axis, which are packed also to form a layer structtire in the c-plane. Both crystals are *

-

I k— --

I

—~



‘-it

—~

-



I

C



o~

a —

(a)

(b)

Fig. 1. The crystal structure of lead monoxide for (a) the tetragonal phase and (b) the orthorhombic phase. The narrow solld llnes indicate the unit cell framework.

The address after December 1979 will be: Electrotechnical Laboratory, Sakura-mura, Ibaraki 300—31, Japan. 568

K. Oka et al.

/ Growth of single crystals of P1,0

volatile at higher temperatures and because the high temperature orthorhombic phase boule is liable to shatter on cooling through the distortion temperature. As investigations of lead monoxide have mostly been aimed at its photoelectric property or at X-ray sensitive photoconductivity, experiments have been done with powders, thin films and pressed specimens. However, for physical measurements such as dielectric permittivity, one requires bulk crystals. We developed new methods of growing large single crystals separately for the two polymorphs of the lead monoxide. In this note, we describe the growth methods and the results. We also present some results of electrical and optical measurements for crystals obtained by the method developed.

2. Experimental

water

Cooling

569

Shaft

~dui~~m Shaft Thermal Insulator ~

0 ~3~

r I Heating Coil

Platinum

~

Crucible

0

Thermocouple Cru cible Susceptor



Stainless Steel Shaft

Fig. 3. Vertical cross section of the crucible arrangement.

containing between 11 and 14 mole% Ti0

2. The

orthorhombic phase crystal should be grown from 2.1. Experimental method The phase diagram of the PbO—Ti02 system reported previously [5] is partly reproduced in fig. 2. In this figure, it may be seen that this mixed system could be utilized as the starting material to selectively grow the PbO crystals of either the tetragonal or orthorhombic phase. According to the phase diagram, the tetragonal phase crystal, being stable below about 500°C should be grown from a PbO—Ti02 solution

1200 Pb055 (TeIr) Pb0os(Orth) / ~<~)1000

PbTIO3 (Cubic)

liquid //

/

quid

/

838°

88 ~ 800

~/

/F~Os,(Tetr)

+

Hquid

489 0 PbO

20

The growth assembly used was an ADL-MP furnace, in which the crucible and the heating coil, as shown in fig. 3, were encased in a metal chamber furnished with a viewing window. The platinum crucible was heated by induction from an rf powder source working at 400 kHz and 30 kW. The seed crystal was connected to a platinum shaft which was in turn connected to the water-cooling shaft. The pulling shaft was rotated during the growth while the crucible was stationary. The upper part of the crucible was folded inside to reduce the loss of heat from the surface of the melt. The melt temperature

+

Pb050(Tetr) PbTiO, Cubic) Pb005(Tetr) i- PbTlO, (for) ___________________________________ 10

2.2. Apparatus and the growing procedure

Pb055(Tefr)

PbOm(Orth)h PbQ05 (TeIr) 600

solutions containing less than 11 mole% Ti02. As will be described below, this technique proved to be effective, though the limiting composition was found not to be 11 mole% but about 5 mole% (see fig. 4).

30

~

• •

I

40

Mole % fjQ2

Fig. 2. Part of the phase diagram of the system PbO—Ti02 (after Jaffe et al. [5]).

MOLE

PERCENT TiO~

Fig. 4. Relation between molar ratio of the hO2 solvent in the PbO—Ti02 solution and the structural phase of the resultant crystal: (.) orthorhonibic, (o) tetragonal.

570

K. Oka eta!.

/ Growth of single crystals of PbO

was controlled within ±0.5°Cby using a platinum thermocouple attached to the bottom of the crucible. The chemicals used were of 99.99% purity. The evaporation of PbO took place during the heating of the melt, though the weight loss never amounted more than 1% of the charge.

______ ________ ______

2.3. The tetragonal phase crystal According to the liquidus line between the 11 and 14 mole% Ti0 2 area in the phase diagram, we can expect to get the tetragonal phase crystal by the topseeded solution growth technique. However, after several attempts to pull crystals from the various Ti02 concentrations, we found that mixing more than 5% of Ti02 yielded the tetragonal phase. In fig. 4, the relation between the concentration of Ti02 in the solution and the structural phase of the resultant crystal is shown. It seems that some modification of the phase diagram is required. The seed for the first run was an orthorhombic phase crystal obtained by flux growth from a 10% PbF2 solvent [6]. The dry materials were mixed together in a polyethylene bottle for 2 h and then put in a 100 ml platinum crucible. The crucible can be filled with approximately 400 g powder. A typical composition for the tetragonal phase crystal growth consisted of 6% Ti02 and 94% PbO. As the melt tended to bubble, the mixture was heated slowly. After the bubbling was almost over, the mixture was further heated substantially higher than the melting temperature of 870°C.It took about 30 mm of soaking time to allow complete mixing. The mixture was then seeded. Since the method is of solution growth and the crystal is produced by lowering the temperature, temperature control is more critical than in the conventional method. A slightly high temperature of the solution causes dissolution of the seed and a temperature below the liquidus line causes freezing of the melt when the seed is brought into contact with the melt surface. In order to find the starting temperature for the growth, the seed was dipped into the melt and then the temperature was raised at a rate of 5°C/huntil the bottom of the seed crystal was melted away. Thus a critical temperature was determined by observmg the dissolution of the seed. The mixture was then seeded again at a temperature about 2—3°Cbelow this temperature.

o rt ho I

I

t et r a ,

Jig. 5. Grown crystals of the tetra~onalphase and the orthorhombic phase lead monoxide.

The temperature bf the melt was lowered along the liquidus line in the phase diagram at a rate of, typically, 0.5—1°C/huntil the desired boule diameter was attained. At the same time the seed crystal was pulled upwards at a rate of the order of approximately 2—5 mm/h. In the final stage of growth, the grown crystal was removed from the melt and kept suspended just above the surface and the temperature was lowered at a rate of 200°C/hdown to room temperature. A crystal weighing ‘--‘80--100 g could be grown in 10—20 h operation, and the temperature was decreased as much as 30°Cduring this operation. Examination by X-ray diffraction proved that diffraction lines from the obtained crystal were completely those of the tetragonal phase structure. However, the obtained crystal, once heated to higher temperatures, never turns into the other phase, even when heated over 700°C,which is much higher than the known transition temperature, suggesting some superheating phenomenon occurred in this crystal. The single crystal boule actually grown is shown in fig. 5. The boule is of the order of 30 mm in diameter, 20 mm in length and 85 g in weight. The crystal is easy to cleave along the (001) plane into thin plates, which appear transparent and dark red. 2.4. The orthorhombic phase crystal The orthorhombic phase crystal was grown by the simple Czochralski method. The pure PhD reagent

K. Oka eta!.

/ Growth of single crystals of PbO

was placed in a 100 ml platinum crucible, which was placed in the colsed chamber with an air atmosphere. The melt temperature was kept constant during the growth at around 890°C. The pulling rate was 5 mm/h with a rotating rate of 40—50 rpm. The grown boule of the orthorhombic phase crystal was found to shatter into thin flakes when it was slowly cooled down to room temperature. Actually, the crumbling of the boule was observed to occur at a temperature around 100°C, which was much lower than the known temperature. It seems that for this crystal the high temperature phase was supercooled and the distortion temperature was shifted down to this temperature. It was found, however, that the crumbling phenomenon could be prevented by quenching the crystal boule after the following procedures: Shortly after the growing was over, the power supply for the furnace was abruptly cut off. Then the boule was removed from the chamber with tongs, probably at about 200 or 300°C, and was quenched by dipping into water. This may be the technique for making a crackless orthorhombic phase crystal at room temperature, where the distortion temperature is again further shifted downwards. Thus an orthorhombic phase crystal of about 130 g has been obtained in about 10 h operation. The single crystal actually grown by the technique described above is also shown in fig. 5. The crystal is also easy to cleave along the (001) plane. Thin plates are transparent and greenish-yellow. A fragment of the crystal, even of the crystal grown from the 4 mole% Ti0 2 melt, again displays the crumbling in a second or further heat treatment near room temperature during the slow cooling. We have tried the solution growth from several different compositions ranging from 0.01 to 11 mole% Ti02 and found that the growth from solutions containing less than 4 rnole% resulted in orthorhombic phase crystals. in fig. 4 we indicated the situation. It should be emphasized that a boundary value of Ti02 between 4 and 5 mole% is a reproducible value, in obtaining either structural phase. Quantitative analyses of the Ti impurity in the grown crystal by X-ray fluorescence spectrum analysis and by electron microprobe analysis have proved that the crystals from 1.0, 4.0, 6.0 and 11.0 mole% Ti02 mixed melt contain Ti impurity of, respectively, 0.02, 0.96, 1.36 and 1.76 mole%.

571

20

00 I

~t

80 ortho

~ ~

I

~ ‘--a

cc I

-

[

5

________

0 400

_____________

500 I

60 0

_______________ ______________

WbVELENGTH In m) Fig. 6. Optical transmission of the tetragoanl phase and the orthorhombic phase crystals along the [0011 direction.

3. Optical and dielectric measurements Optical transmission measurements were made on ‘--200 pm thick platelets for both structural phase crystals, and the results are shown in fig. 6. The tetragonal phase crystal is transparent in the wave-length region above ‘-—620 nm and the orthorhombic one, above ‘-—480 nm. The results are almost coincident

(110)

L

______________________

1

io j

I

1oo~u

Fig. 7. Etch pit pattern on the (001) surface of the tetragonal

phase crystal.

572

K. Oka etal.

/ Growth of single crystals of PbO

Table 1 The static dielectric constant in the two structural phases at room temperature

was equivalent to a slow heat treatment. The measurements were made at a frequency of 1 kHz with an ac amplitude of ‘--10 V/cm, at room tempera-

Structural phase

Direction

ture, and prominent anisotropy was observed. The values obtained are shown in table 1.

Orthorhombic

a-axis b-axis c-axIs

Tetragonal ________________

a-axis c-axis

Dielectric constant ~-200 73 6 43 11

Acknowledgements 20 a) 12 a)

18 b) 18 b)

___________________

We thank H. Yasunaga of the University of

a) Ref. [81. b) Ref. [71.

Electro-Communications for informing us about results of his earlier investigation on the same subject, and also K. Arai, I. Hayashi, N. Koshizuka and T. Yao for their help in the optical transmission measure-

with the earlier observations [3,7]. (The polarization direction of the incident light within the c-plane was not specified.) In fig. 7 the etch pattern on the (001) plane in the tetragonal phase crystal is shown. Following the data given by linurna et al. [3], we observed etch pits by heating the crystal at 650°C 2 Torr for 20 mm. The etch and at a pressure of’-—10 pit lines are seen parallel to (100) directions and opposite at 90°to each other. The availability of single crystals makes it possible to determine three components of the dielectric tensor. Crystals were cut into rectangualr plates of, say, 8mm on edge and with a thickness of approximately 1 mm. Three plates, which are perpendicular to the a-, b- and c-axes of the orthorhombic crystal, and two, perpendicular to the a- and c-axes of the tetragonal one, were prepared. The electrodes were made by gold evaporation for the tetragonal phase crystal and by painting with silver paste for the orthorhombic phase crystal. The metal evaporation caused the specimens to crumble because the process

ments and electron microprobe analysis. We are also grateful to A. Nakamura for attracting our attention to this material and for discussions.

References [1] W.J. Moore, Jr. and L. Pauling, J. Am. Chem. Soc. 63 (1941) 1392. [21 MI. Kay, Acta Cryst. 14 (1961) 80. [31 K. linuma, T. Seki and M. Wada, Mater. Res. Bull. 2 (1967) 527. [41 C.J.M. Rooymans and W.F. Th. Langenhoff, J. Crystal Growth 3/4 (1968) 311. [5J B. Jaffe, W.R. Cook, Jr. and H. Jaffe, Piezoelectric Ceramics cation(Academic from D.E. Press, Rase).London, 1971) (private communi[6[ M. Levin, C.R. Robbins and HF. McMurdie, Eds., Phase Diagrams for Ceramists (American Ceramic Society, 1964) fig. 1674.

[71 J. van den Broek, Philips Res. Rept. 22 (1967) 36. [81 D.A. Adams and D.C. Stevens, J. Chem. Soc. Dalton (1977) 1096.