Phase Equilibrium Relationships in the System Gd₂O₃-TiO₂

Abstract

The phase equilibrium relationships for a major portion of the Gd₂O₃-TiO₂ system were determined in air, from a study of solid state reactions and from fusion characteristics. Three intermediate phases, a 1:2 compound, a 1:1 compound, and a face-centered cubic solid solution occur in the system. The solid solution phase is indicated on the phase diagram as existing from 33 to 40 mole percent TiO₂, at 1750 °C. This phase melts incongruently over a range of temperatures and compositions, from 1840 °C, the peritectic temperature at 35 mole percent TiO₂, to 1775 °C at 40 mole percent TiO₂, the incongruent melting temperature of Gd₂O₃∙TiO₂. The minimum temperature of stability for the phase is 1600 °C at 38 mole percent TiO₂. The compound Gd₂O₃∙2TiO₂ has a cubic pyrochlore structure type with a = 10.181 Å and melts congruently at 1820 °C. This phase apparently accepts up to about 3 mole percent TiO₂ in solid solution at 1460 °C. The compound Gd₂O₃∙TiO₂, melts incongruently at 1775 °C and has a reversible phase transition at 1712 °G. The x-ray powder diffraction pattern of the high temperature modification was indexed on the basis of a hexagonal cell a = 3.683 Å, c= 11.995 Å and is apparently related to the A-type rare earth oxide. The compositions Sm₂O:TiO₂ and Eu₂O₃:TiO₂ gave x-ray diffraction powder patterns with marked similarity one to the other as well as to both the high and low polymorphs of Gd2O₃∙TiO2. The composition Dy₂O₃:TiO₂ formed several phases one of which is apparently related to the high temperature polymorph of Gd₂O₃ •TiO₂.

1. Introduction

The increased availability of more pure rare earth oxides together with a continuing search for new rare earth refractory materials with desirable optical, electrical, and magnetic properties has served as an impetus to investigate the phase equilibrium relationships in rare earth oxide systems. The authors [1]¹ outlined, in part, the subsolidus relationships of the binary combinations of the rare earth sesquioxides with selected oxides of the trivalent cations.

It was decided to extend this rare earth sesquioxide study to incorporate the oxides of the tetravalent cations. For this reason, the system Gd₂O₃-TiO₂ was selected to be studied in detail in air since it is probably representative of other B-type rare earth oxide-TiO2 systems. The only detailed Ln₂O₃-TiO2 phase study² reported previously in the literature was that of the system La₂O₃-TiO₂ by MacChesney and Sauer [2]. Roth [3] and Brixner [4] reported the existence of a number of rare earth titanates of the general type Ln₂Ti₂O₇ having the pyrochlore type structure, i.e., Gd₂Ti₂0₇ (a =10.181 Å). Queyroux [5] reported a tentative phase diagram for the system Gd₂O₃-TiO₂ in which three phases were postulated: a solid solution phase occurring at approximately 40 mole percent TiO₂, a 1:2 compound with limited solid solution, and a 1:1 compound.

Three polymorphs of Gd₂O₃ have been reported. They are the A (hexagonal), B (monoclinic) and C (cubic) type rare earth oxides. The A type was reported [6] to occur metastably. Several workers [7, 8] have reported that the C to B phase transition in Gd₂O₃ is reversible. Roth and Schneider [6] concluded that Gd₂O₃ crystallizes in the C form at low temperatures and transforms irreversibly to the B type at 1225 °C. In this study the B type was considered to be the only stable modification. The unit cell dimensions of B-type Gd₂O₃ were reported by Roth and Schneider [6] as a =14.06 Å, 6 = 3.572 Å, c = 8.75 Å, and β= 100.10°. Those of TiO₂ (rutile) were given by Swanson and Tatge [9] as a = 4.594 Å, c = 2.958 Å.

2. Materials

The starting materials used in this study were found by general quantitative spectrochemical analyses³ to have the following impurities: Gd₂O₃-Ca present in amounts less than 0.01 percent, Fe, Mg, Pb, and Si each present in amounts less than 0.001 percent, and B, Cu, and Mn each present in amounts less than 0.0001 percent. TiO₂−Si present in amounts less than .1 percent, Mg present in amounts less than 0.01 percent, Cu present in less than 0.001 percent, and Ca present in amounts less than 0.0001 percent.

3. Specimen Preparation and Test Methods

Two gram batches of various combinations of Gd₂O₃ and TiO₂ were weighed, mixed in a mechanical shaker for approximately 10 min and pressed into disks at about 10⁴ psi. The disks were placed on platinum setters and calcined in air at 1000 °C for 18 hrs. To achieve physical homogeneity the specimens were ground in an agate mortar, repressed and recalcined for an additional 4 hrs at 1000 °C. Following these preliminary heat treatments portions of the ground specimen were placed in platinum alloy tubes and heated in the quenching furnace to various temperatures for different periods of time.

The tubes containing the specimens were quenched into ice water and examined by x-ray diffraction techniques. A high-angle recording Geiger counter diffractometer and Ni-filtered Cu radiation was used in the study. The Geiger counter traversed the specimen at 1/4 deg/min and radiation was recorded on the chart at 1 deg-2θ/in. The unit cell dimensions reported can be considered accurate to about ±2 in the last decimal place listed. Equilibrium was considered to have been achieved when x-ray patterns showed no change after successive heat treatments of a specimen or when the data were consistent with the results from a previous set of experiments. Solidus and liquidus temperatures were obtained by using both a quenching furnace and an induction furnace. Because of the temperature limitation of the quenching furnace melting points above 1800 °C were determined with the induction furnace. Some duplicate determinations below 1800 °C were made using both furnace types.

The essential features of the furnaces were described previously by the authors [11]. In essence, the quenching furnace consisted of two concentric ceramic tubes wound with platinum-rhodium alloy wire. The inner tube served as the primary winding and the outer one as the booster. Separate power sources were used with each winding.

The power for the outer winding (booster) was supplied from a variable auto transformer. An a-c bridge type controller [12] in which the furnace winding was one arm of the bridge was used to control the temperature of the inner winding. The furnace temperature was controlled to about ±3 °C. Temperature was measured with a 95 percent Pt-5 percent Rh versus 80 percent Pt-20 percent Rh thermocouple used in conjunction with a high precision potentiometer.

The induction furnace consisted of an iridium crucible and cover which acted as the susceptor and specimen container. A small fragment of the calcined material was placed in the iridium crucible on an iridium setter or button and heated to the desired temperature for about 2 min to achieve thermal equilibrium. Apparent temperatures were measured with a calibrated, disappearing filament-type optical pyrometer which was sighted through a 45 deg calibrated prism into the viewing hole at the center of the crucible cover. The temperature measuring system of both the quenching furnace and the induction furnace were calibrated frequently against the melting point of Au (1063 °C), Pd (1552 °C), and Pt (1769 °C)⁴. In addition, the measuring system of the induction furnace was also calibrated against the melting point of Rh (1960 °C). Temperatures reported in the present study were considered accurate to within ± 10 °C below 1650 °C and to within ±20 °C above. The measurements were reproducible to within ±5 °C, or better. The degree of melting was determined by the physical appearance of quenched or rapidly cooled specimens. The first adherence of the specimen to the platinum container or iridium setter generally established the beginning of melting. In quenched specimens, complete melting was established by the formation of a meniscus. Similarly the slumping or loss of shape of the inductively heated specimens established complete melting.

4. Results and Discussion

The equilibrium phase diagram (fig. 1) for a major portion of the Gd₂O₃-TiO₂ system has been constructed from the data listed in tables 1 and 2. Solid circles represent composition and temperatures of experiments conducted in the quenching furnace and solid triangles represent those conducted in the induction furnace. A combination of a solid triangle bound by an open circle represents those conducted in both types of furnaces at the same temperature. Two melting points of Gd₂O₃ 2330 °C and 2350 °C [13, 14] have been reported. Because of the temperature limitations of the present equipment the value was not redetermined. The numerous reported melting points of TiO₂ have been tabulated by Schneider [15]. Because of the apparent problems inherent with the determination of the melting point of TiO₂, a redetermination was not made at this time.

Figure 1. Phase equilibrium diagram for the system Gd₂O₃-TiO₂.

L—1 : 1 —Low temperature form of Gd₂O₃• TiO₂.

H-l : 1 — High temperature form of Gd₂O₃ • TiO₂.

1 : 2 —Gd₂O₃ • 2TiO₂.

FCC —Face centered cubic phase.

ss —Solid solution.

Liq —Liquid.

● —Compositions and temperatures of experiments conducted in the quench furnace.

▲ —Compositions and temperatures of experiments conducted in the iridium crucible induction furnace.

● — Compositions and temperatures of experiments conducted in both the quenching and induction furnace.

For clarity, not all experimental data appearing in tables 1 and 2 are plotted on this diagram.

Table 1.

Experimental quenching data for compositions in the Gd₂O₃—TiO₂ system

Composition		Heat treatmenta		X-ray analysisb	Remarks
Gd2O3	TiO2	Temp.	Time
Mole %	Mole %	°C	hr
85	15	1648	2.5	Gd2O3 + Fccss
70	30	1548	2	L-Gd2O3-TiO2 + Gd2O3
		1607	2.5	FCCss+ Gd2O3
		1648	2.5	FCCss + Gd2O3
		1684	2.5	FCCss + Gd2O3
		1751	2	FCCss+ Gd2O3
66.67	33.33	1601	2	FCCss+ Gd2O3c
		1649	2	FCCss+ Gd2O3
		1688	3	FCCss+ Gd2O3
		1767	2	FCCss + Gd2O3
		1781	4	FCCss
65	35	1444	64	L–Gd203 TiO2 + Gd2O,
		1550	2	L–Gd203 TiO2+Gd2O3
		1567	3	L–Gd203 · TiO2 + Gd2O3
		1584	2	L–Gd203 TiO2 + Gd2O3
		1601	2.5	FCCss + (Gd2O3)	Gd2O3 not detected in x-ray pattern.
		1639	4	FCCss+ Gd2O3c
		1648	3	FCCss + Gd2O3
		1649	19	FCCss + Gd2O3c
		1751	1	FCCss
		1767	2	FCCss
		1548	5	L-Gd2O3 · TiO2 + Gd2O3	Reheat of 1648 °C specimen.
62	38	1606	6	FCCss	No microscopic evidence of second phase.
60	40	1444	64	L–Gd2O3-TiO2 + Gd2O3
		1567	3	L–Gd2O3TiO2 + Gd2O,
		1599	16	L–Gd2O3-TiO2 + Gd2O3
		1601	2.5	FCCss + (L–Gd2O3 · TiO2)	L-Gd2O3 · TiO2 not detected by x-ray diffraction; presumed to be present on basis of other data.
		1613	1	FCCss + L–Gd2Os · TiO2
		1627	4	FCCss+ L–Gd2O3TiO2
		1648	3	FCCss+ (L–Gd2O3-TiO2)	L-Gd2O3 · TiO2 not detected by x-ray diffraction; presumed to be present on basis of other data.
		1672	2.5	FCCss + L–Gd2O3 TiO2c
60	40	1700	1	FCCss
		1751	2	FCCss
		1776	66	FCCss	Specimen partially melted.
		1548	3	L–Gd2O3 TiO2 + Gd2O3	Reheat of 1648 °C, specimen.
56.45	43.55	1645	2.5	FCCss + L — Gd2O3 · TiO2
		1698	6	FCCss+ L-Gd2O3TiO2
		1702	1	FCCss+ L-Gd2O3TiO2
		1724	3	L–Gd2O3 · TiO2 + FCCss + H–Gd2O3 · TiO2	Nonequilibrium mixture.
50	50	1444	64	L–Gd2O3 · TiO2
		1550	2	L–Gd2O3 · TiO2 + Gd2O3 + Gd2O3 · 2TiO2	Nonequilibrium mixture.
		1648	2.5	Gd2O3 · 2TiO2	Do.
		1696	1.5	Gd2O3 · 2TiO2	Do.
		1708	1	L–Gd2O3 TiO2
		1708	2	L–Gd2O3 TiO2
		1718	2.5	L–Gd2O3 · TiO2 + H–Gd2O3 · TiO2	H—Gd2O3-TiO2 present in small amount.
		1738	1	L–Gd2O3 · TiO2 + H–Gd2O3 · TiO2	Do.
		1751	1	H–Gd2O3 · TiO2 + L–Gd2O3 · TiO,
		1760	1	H–Gd203 · TiO, + L–Gd2O3 · TiO2
		1777	.75	H–Gd203 · TiO2 + L–Gd2O3 · TiO2	Specimen partially melted L–Gd2O3 ·TiO2 present in very small amount.
					Specimen partially melted.
		1790	1	H–Gd2O3 · TiO2 + unknown	Extraneous x-ray peak d— 3.0435 Å probably represents nonequilibrium phase.
		1652	7	L–Cd2O3 · TiO2	Reheat of 1777 °C specimen.
		1754	5	H–Gd2O3 · TiO2 + L–Gd2O3 · TiO2	Reheat of 1790 °C specimen.
44	56	1737	3.5	H–Gd2O3 · TiO2 + Gd2O3 · 2TiO2
		1762	.33	H–Gd2O3 · TiO2 + Gd2O3 · 2TiO2
40	60	1508	3	L–Gd2O3 · TiO2 + Gd2O3 · 2TiOz
		1597	2	L–Gd2O, · TiO2 + Gd2O3 · 2TiO2
		1662	5	L–Gd2O3 · TiO2 + Gd2O3 · 2TiO2
		1670	4	L–Gd2O3 · TiO2 + Gd2O3 · 2TiO2
		1720	3.5	Gd2O3 · 2TiO2 + H–Gd2O3 · TiO2
		1730	1.5	Gd2O3 · 2TiO2 + H–Gd2O3 · TiO2
33.33	66.67	1460	19	Gd2O3 · 2TiO2
		1508	1	Gd2O3 · 2TiOz
		1604	0.5	Gd2O3 · 2TiO2
		1702	2	Gd2O3 · 2TiO2
30	70	1460	19	Gd2O3 · 2TiO2ss + TiO2
20	80	1416	25	Gd2O3 · 2TiO2ss −1- TiO2
		1460	19	Gd2O3 · 2TiO2ss + TiO2
10	90	1416	25	TiO2 + Gd2O3 · 2TiO2ss

^aUnless otherwise indicated all specimens were first calcined at a 1000 °C-18 hr. and recalcined at 1000 °C-4 hr

^bPhases identified are given in order of amount present (greatest amount first) at room temperature.

^cSecond phase identified by use of petrographic microscope.

Table 2.

Melting characteristics of the Gd₂O₃-TiO₂ system

Compositiona		Temperatureb	Furnacec	Observation
Gd2O3	TiO2
Mole %	Mole %	°C
85	15	1816	I	Not melted.
		1839	I	Do.
		1860	I	Partially melted.
		1897	I	Do.
		1974	I	Do.
70	30	1911	I	Do.
		1946	I	Do.
		1967	I	Completely melted.
		1987	I	Do.
65	35	1838	I	Not melted.
		1841	I	Partially melted.
		1845	I	Do.
		1862	I	Do.
		1894	I	Completely melted.
		1911	I	Do.
		1974	I	Do.
62	38	1765	I	Not melted.
		1773	I	Do.
		1790	I	Do.
		1804	I	Partially melted.
		1835	I	Do.
		1867	I	Completely melted.
60	40	1751	Q	Not melted.
		1776	Q	Partially melted.
		1816	I	Do.
		1829	I	Do.
		1850	I	Completely melted.
56.45	43.55	1745	I	Not melted.
		1765	I	Do.
		1768	I	Do.
		1779	I	Partially melted.
		1796	I	Do.
		1803	I	Do.
		1811	I	Do.
		1821	I	Completely melted.
		1830	I	Do.
		1841	I	Completely melted.
		1849	I	Do.
		1853	I	Do.
		1866	I	Do.
52.60	47.40	1780	I	Partially melted.
		1805	I	Completely melted.
		1824	I	Do.
50	50	1772	Q	Not melted.
		1777	Q	Partially melted.
		1779	I	Do.
		1782	Q	Do.
		1795	I	Completely melted.
		1790	Q	Do.
		1814	I	Do.
		1832	I	Do.
44	56	1737	I	Not melted.
		1762	Q	Not melted.
		1762	I	Do.
		1773	I	Partially melted.
		1790	I	Do.
		1796	I	Do.
		1797	Q	Do.
		1813	I	Completely melted.
		1830	I	Do.
		1862	I	Do.
		1996	I	Do.
		2050	I	Do.
40	60	1767	Q	Not melted.
		1799	I	Partially melted.
		1806	I	Do.
		1813	I	Completely melted.
		1849	I	Do.
33.33	66.67	1752	I	Not melted.
		1770	I	Do.
		1792	I	Do.
		1806	I	Do.
		1810	I	Do.
		1814	I	Do.
		1823	I	Melted.
		1834	I	Do.
		1843	I	Do.
		1850	I	Do.
		1878	I	Do.
30	70	1550	Q	Not melted.
		1573	Q	Partially melted.
		1592	Q	Do.
		1719	Q	Do.
		1747	Q	Do.
		1781	Q	Do.
		1798	I	Do.
		1840	I	Completely melted.
25	75	1490	Q	Not melted.
		1541	Q	Do.
		1555	Q	Partially melted.
		1704	I	Do.
		1736	I	Do.
		1772	I	Do.
		1794	I	Completely melted.
20	80	1495	Q	Not melted.
		1508	Q	Partially melted.
		1583	Q	Do.
		1606	Q	Do.
		1641	Q	Do.
		1664	Q	Melted.
		1677	Q	Do.
		1683	Q	Do.
		1747	Q	Do.
17	83	1505	Q	Not melted.
		1525	Q	Do.
		1562	Q	Melted.d
		1585	Q	Do.
15	85	1506	Q	Not melted.
		1533	Q	Do.
		1549	Q	‘Melted.d
		1565	Q	Do.
		1575	Q	Do.
		1584	Q	Do.
		1618	Q	Do.
13	87	1542	Q	Not melted.
		1553	Q	Melted.d
		1564	Q	Do.
		1576	Q	Do.
10	90	1465	Q	Not melted.
		1495	0	Do.
		1511	Q	Do.
		1554	Q	Partially melted.
		1561	Q	Do.
		1580	Q	Melted.d
		1608	Q	Do.
		1700	Q	Do.
5	95	1542	Q	Not melted.
		1556	Q	Partially melted.
		1568	Q	Do.
		1694	Q	Melted.d
		1725	Q	Completely melted.

^aSpecimens first calcined at 1000 °C for 18 hr and recalcined at 1000 °C for 4 hr.

^bSpecimens furnace cooled except when indicated.

^cQ-quenching furnace; I-induction furnace. All specimens heated in induction furnace slow cooled rather than quenched.

^dDefinite verification of complete melting could not be obtained.

The following intermediate phases, indicated in figure 1, were found to occur in the system: a solid solution (FCC_SS), existing at about 1750 °C, from approximately 33 to 40 mole percent TiO₂; a compound, Gd₂O₃-TiO₂, and a compound Gd₂O₃∙2TiO₂. These data generally confirm those reported by Queyroux [5].

The solid solution phase (FCC_SS) has a maximum incongruent melting point at 1840 °C at 35 mole percent TiO₂. A minimum dissociation temperature was found to occur at 1600 °C at 38 mole percent TiO₂ as compared to the 1550 °C [5] value reported previously. The existence of a minimum decomposition temperature was evident since the solid solution phase once formed could be decomposed by reheating to 1548 °C for 4 hrs. From x-ray powder diffraction data, the solid solution phase was shown to be face-centered cubic with a maximum unit cell dimension of 5.32 Å and a minimum of 5.28 Å at 1750 °C. Since the x-ray patterns were of rather poor quality with characteristic high background, the polarizing microscope was employed to determine the compositional limits of the phase. Small amounts of anisotropic phases were easily distinguished from the isotropic cubic phase.

The compound, Gd₂O₃∙TiO₂, was found to melt incongruently at 1775 °C and to have a reversible phase transition at 1712 °C. The x-ray powder diffraction patterns for the low and the high temperature polymorphs of the 1:1 compound are given in tables 3 and 4 respectively. Unfortunately, neither of the x-ray patterns of the polymorphs of Gd₂O₃∙TiO₂ could be indexed on a unit cell based on the constants given by Queyroux [5]. Without a more complete reporting of Queyroux’s data these discrepancies cannot be resolved. The x-ray diffraction powder pattern of the low temperature form (L-Gd₂O₃ • TiO₂) was not indexed. However, the x-ray diffraction pattern of the high temperature form (H-Gd₂O₃ • TiO₂) was indexed on the basis of a hexagonal cell, a = 3.683 Å, c = 11.995 Å. The high temperature polymorph is apparently related to the hexagonal A-type rare earth oxide through the relationship $c\left(\text{hexagonal\hspace{0.17em}}{\text{Ln}}_2{\text{O}}_3\right)\cong \frac{1}{2}c\left(\text{H}-{\text{Gd}}_2{\text{O}}_3\cdot {\text{TiO}}_2\right).$ Some difficulty was encountered in determining the inversion temperature since numerous attempts to quench H—Gd₂O₃ • TiO₂ as a single phase below the solidus were unsuccessful. In each instance a small amount of L—Gd,₂O₃•TiO₂ was present. However, the x-ray pattern of the partially melted 1:1 compound quenched from 1790 °C was found to be essentially single phase. One very weak extra line occurred in the x-ray pattern at d = 3.044 Å which was probably representative of a metastable phase formed only in the quenched liquid. Both the 33.33 and the 43.55 percent TiO₂ compositions when quenched from above 1712 °C, the inversion temperature of Gd₂O₃ • TiO₂, contained H-Gd₂O₃ • TiO₂ and an appropriate second phase. The same compositions quenched from below 1712 °C contained L-Gd₂O₃ •TiO₂ and an appropriate second phase. These experimental data seem to indicate that the possibility of quenching H-Gd₂O₃-TiO₂ without residual traces of L-Gd₂O₃ • TiO₂ is only accomplished by the presence of another second phase.

Table 3.

X-ray diffraction powder data for the low temperature modification of Gd₂O₃•TiO₂ (CuK_α radiation)

da	I/I0b
Å
7.70	21
3.56	21
3.066	100
3.060	79
3.054	64
2.976	14
2.947	14
2.667	64
2.624	14
2.566	17
2.554	21
2.334	11
2.130	14
2.113	14
1.925	14
1.896	10
1.869	21
1.661	19
1.648	10
1.607	26
1.601	11
1.597	43
1.586	29
1.583	17
1.569	14

^aInterplanar spacing.

^bRelative intensity.

Table 4.

X-ray diffraction powder data for the high temperature modification of Gd₂O₃ • TiO₂^a (CuKα radiation)

hklb	dc	1/d2		I/I0e
		Obs	Cald
	Å
002	6.02	0.0276	0.0277	10
100	3.189	.0983	.0983	38
101	3.083	.1052	.1056	32
	3.0435	.1080		10
004	3.007	.1105	.1112	26
102	2.817	.1260	.1264	100
104	2.185	.2094	.2098	22
105	1.918	.2720	.2724	8
110	1.839	.2958	.2958	29
112	1.757	.3239	.3236	8
106	1.694	.3484	.3488	12
200	1.645	.3944	.3944	58
201	1.578	.4017	.4014	10
114	1.568	.4070	.4070	59
202	1.538	.4227	.4222	60
212	1.1801	.7183	.7181	8
300	1.0612	.8881	.8875	8

^aSpecimen not single phase. The line which occurs at d= 3.0435 Å probably represents a metastable phase formed only in the quenched liquid.-

^bHexagonal Miller indices.

^clnterplanar Spacing.

^dBased on a hexagonal cell, a = 3.683 Å, r = 11.995 Å.

^eRelative intensity.

Other Ln₂O₃ : TiO₂ compositions were also studied. From x-ray powder diffraction data, it appears that the compositions Sm₂O₃ : TiO₂ and Eu₂O₃ : TiO₂ form phases which are very similar to both L —Gd₂O₃ • TiO₂ and H — Gd₂O₃ • TiO₂. The composition Dy₂O₃: TiO₂ was found to contain a mixture of phases, one of which is apparently similar to H-Gd₂O₃ • TiO₂. These related phases were not detected in the equimolar compositions of the oxides of the smaller rare earth cations with TiO₂.

The compound Gd₂O₃ • TiO₂ was found to melt congruently at 1820 °C. From the change in unit cell constants with composition the compound apparently accepted up to about 3 mole percent TiO₂ in solid solution. The unit cell dimensions of the compositions Gd₂O₃: 2TiO₂ and 30Gd₂O₃: 70TiO₂ quenched from 1460 °C are 10.181 Å and 10.169 Å, respectively. Since the latter composition contained a small amount of TiO₂ the exsolution curve delineating the phase boundary is shown (fig. 1) as dashed at 69 percent TiO₂.

Earlier, the present authors [16] reported the parameters of the following Ln₂O₃ • 2TiO₂ pyrochlore-type phases: Eu₂O₃ • 2TiO₂ (a = 10.195); Ho₂O₃ • 2TiO₂ (a= 10.098 Å); Er₂O₃ • 2TiO₂ (a =10.075 Å); Tm₂O₃ • 2TiO₂ (a = 10.055 Å) and Lu₂O₃ • 2TiO₂ (a = 10.023 Å). Subsequent work by Brixner [4] confirmed these earlier data.

The eutectic between Gd₂O₃ • 2TiO₂ and TiO₂ is shown (fig. 1) as occurring at approximately 86 mole percent TiO₂. Some difficulty was encountered in establishing the liquidus curves near the eutectic composition because it was virtually impossible to distinguish a partially melted from a completely melted specimen. For this reason, the eutectic composition shown in fig. 1 was approximated. Since the melting point of TiO₂ was not redetermined the liquidus was dashed from the eutectic composition to approximately 100 percent TiO₂.

5. Summary

The phase equilibrium diagram for a major portion of the Gd₂O₃-TiO₂ system has been determined from a study of solid state reactions and melting point relations.

A platinum alloy quenching furnace was used to establish all of the subsolidus and some of the liquidus relationships below 1800 °C. An inductively heated iridium crucible was used for the determination of solidus and liquidus relationships above 1800 °C. Phases were identified by x-ray powder diffraction and polarizing microscopic techniques.

Three intermediate phases were formed in the system. A solid solution phase is indicated on the phase diagram as existing from 33 to 40 mole percent TiO₂, at 1750 °C. This phase melts incongruently over a range of temperatures and compositions varying from 1840 °C, the peritectic temperature, at 35 mole percent TiO₂ to 1775 °C, at 40 mole percent TiO₂ the incongruent melting temperature of Gd₂O3 • TiO₂. The minimum temperature of stability for the phase is 1600 °C at 38 mole percent TiO₂. The x-ray diffraction pattern of the solid solution phase was indexed on the basis of a face-centered cubic cell with unit cell dimension varying from about 5.28 Å to 5.32 Å at 1750 °C. The second phase, a 1 : 1 compound, melts incongruently at 1775 °C and has a reversible phase transition at 1712 °C. The high temperature modification was indexed on the basis of a hexagonal cell related to the hexagonal rare earth oxides with a = 3.683 Å, c= 11.995 Å. The third phase, Gd₂O₃ •2TiO₂, melts congruently at 1820 °C. This phase apparently accepted up to about 3 mole percent TiO₂ in solid solution at 1460 °C. The unit cell dimensions of this pyrochlore structure type were found to vary from approximately 10.181 Å to 10.169 Å. From x-ray powder diffraction data the equimolar compositions of either Sm₂O₃ or Eu₂O₃ with TiO₂ were found to form phases which were very similar to both the low and high temperature modifications of Gd₂O₃ • TiO₂. The composition Dy₂O₃: TiO₂ apparently formed several phases one of which was apparently similar to the high temperature form of Gd₂O₃ • TiO₂.