Journal of Crystal Crowth 57 (1982) 194—196 North-Holland Publishing Company
LETTER TO THE EDITORS FLUX GROWTH OF NdOCI SINGLE CRYSTALS Jilali ARIDE, Jean-Pierre CHAMINADE and Michel POUCHARD Laboratoire de Chimie du Solide du CNRS, 35] C~oursde la Libr~rbtion,F-33405 Talence Cedex, France Received 9 October 1981
3 have been grown from molten neodymium chloride in Neodymium oxide chloride single crystals up to 10 x 10 X 0.1 mm the temperature range 900—700°C. A slow-cooling technique (3°C/h)with a scaled platinum 10% rhodium crucible v,as used. The best crystals are transparent, pale purple in color and have well developed (001) faces. They have been characterized by X-ray diffraction and optical measurements.
In recent years there has been a great deal of interest in rare-earth oxide chlorides and their properties as cathodo-luminescent phosphors or energy converters (IR—visible) [1—51.A cooperation has been developed with the “Laboratoire des Elements de Transition dans les Solides du CNRS” to study the crystal growth and characterization of neodymium compounds in particular by optical absorption spectra and paramagnetic anisotropy measurements [61. Synthesis techniques and structural properties of rare-earth oxide chlorides are well known [7—13]. Nevertheless there have been few studies so far con. cerning growth of single crystals. Wanklyn prepared small LaOC1 crystals by hydrolysis of a dehydrated Lad3 and MgSO4 melt . Brandt and Diehl obtained small YbOC1 crystals for X-ray diffraction studies by slow cooling of a Yb2 03 and YbCl3 melt in an argon atmosphere . In this work NdOC1 growth was carried out by slow cooling in molten NdCl3 in a sealed platinum choice of NdCl3 as flux presents several the absence of foreign cations or anions minimizes crystal impurities; — there is no corrosion of the platinum crucible; there is easy dissolution of the flux in distilled water. Baev and Novikov , then Ngok T’en and Morozov  studied the NdC13-rich part of the NdCl3 —Nd2 03 phase diagram. An eutectic point of
722 °C for a 93 mole% NdC1 3 composition was found by these authors. Decomposition of NdOC1 at 791 °C was announced by Baev and Novikov. Before carrying out growth experiments, it was necessary to control this possible decomposition. NdOC1 powder was introduced into a platinum tube filled with pure dry argon and sealed. Platinum tubes were heated at different temperatures up to 1400 °C and subsequently quenched. X-ray analysis of the
L+ Nd U CL is)
L +Nd Cl3~)
0022 -0248/82/0000—-0000/$02 .75 © 1982 North-Holland
Nd CL3 • Nd~01CL
NdO Ci 0.50)
Nd Cl3 ,,ol,r fractio,,
-fig. 1. NdCl3—Nd203 equilibrium diagram in the NdCI3-rich part, from ref. [16l~ vertical arrows indicate compositions
used for crystal growth runs.
J. Aride et al.
/ Flux growth of NdOC1 single crystals
reaction products showed there to be no decomposition up to this temperature. Several compositions
were controlled by micro-DTA and our results agree well with those of the previous authors (fig. 1). NdC13 appears to be a good solvent for the growth of NdOC1 single crystals at a temperature higher than 722 °C. The furnace used in this work was built with
Kanthal wires, controlled by a Eurotherm regulator and programmer. The performance of the furnace can be characterized as follows: temperature stability better than ±0.5°C, cooling rate in the range 0.5—50°C/h and maximum temperature of about 1350’~C.Cylindrical 3shaped 10%tube rhodium of were platinum used. A thin at the crucibles top of the 30 cm sealed-cap crucible permits filling with powder and then sealing. The starting materials were NdCl 3 and Nd2 03 with a purity better than 99.9% (Cerac). NdCl3 was purified by heating in an atmosphere of HC1 and conserved in a dry box because of its highly hygroscopic nature. The mixture of grounded NdC13 and Nd2 03 was introduced into the platinum cru-
cible and sealed with an arc in a dry argon glove box (02 ~ 10 ppm, H2O ~ 10 ppm). We used the following thermal cycle: heating up to the operational temperature, then a slow decrease down to the solidification flux temperature and finally a rapid decrease to room temperature. After sawing off the upper part of the crucible single crys~ tals were removed by dissolution of the flux in distilled water. Several steps can be involved in the reactional procedure: the reaction of neodymium oxide and chlo-
ride to form NdOC1, the dissolution of NdOCI in the excess of chloride which then acts as flux and finally the precipitation of NdOC1 single crystals by slowcooling of the solution. Table 1 NdOC1 growth conditions; crystal NdOCl, flux NdCI 3) 3, crucible Pt—10%Rh (30 cm Compounds Mass Composition
Fi1,. 2. Examples of NdOC1 single crystals (side of small squares = 0.1 cm).
Various experimental parameters were tested in order to improve the crystal growth of NdOCl: the Nd2 O3—NdC13 composition (the chosen Nd2 O3/NdC13 ratios are indicated in fig. 1); maximum temperature from 900 to 1100°C; holding time from 2 to 15 h; cooling rate between 0.5 and 8 °C/h. The parameters necessary for best growth conditions are collected in table 1. The material obtained grew as transparent, pale purple coloured a sizeinupfig.to2.10A Xtypical 10 X 3. A goodplatelets examplewith is given 0.1 mm —
Table 2 NdOCI lattice parameters; NdOC1, tetragonal system, P4/ nmnm; dcaic = 5.948 g/cm3 dmeas = 5.945 ± 0.03 g/cm3 Z=2
This work [101 [111
4.015 4.04 4.018
Nd 2O3 NdOC1 NdC13 NdCl3
9.6 26.3 90.4 73.7
12.5 21.8 87.5 78.2
0.002 0.03 ± 0.002 ± ±
6.782 6.77 6.782
0.003 0.04 ± 0.004 ± ±
a!. / Flux growth of VdOC1 single crystals
run yields also a large proportion of small poor quality crystals. The precipitated crystals were identified by X-ray diffraction. The lattice parameters of the crystals were refined by X-ray powder diffraction data with a Si internal standard (high purity) arid by using a least square method. The parameters are compared in table 2 with those obtained by previous authors, The measured density of the crystals is in good agreement with the calculated value by assuming two formula units in the cell (table 2). A Laue photograph taken parallel to the large face of the platelets showed clearly a four-fold symmetry axis characterizing (001) planes of atetragonalPhFCllike structure. An examination of the platelets using a polarizing microscope shows no extinction, thus indicating that the tetragonal c-axis is perpendicular to the plane of the platelets. Finally, observation in convergent polarizing light between crossed polars showed an uniaxial negative interference figure. As a consequence the crystal habit of NdOC1 is of well developed (001) planes, as shown in fig. 2. n+ The NdOC1 structure can be described as (NdO)rm layers of edge-linked ONd 4 tetrahedra, alternating with sheets of Cl— anions. It has been pointed out that the plate-like crystal habit is simply related to
the layer structure characterizing the lattice. The concentrations of the major impurities were determined by atomic absorption and emission spectrometry (table 3). Representative crystals which appeared optically homogeneous and free of visible inclusion (after microscopic examination) were ultrasonically cleaned in diluted hydrochloric acid. The si7e and quality of the crystals are appropriate for optical absorption and susceptibility anisotropy measurements. A detailed examnination of the properties will be reported in a separate publication.
References iii LU. Van Uitert, lET. Levinstcin and \V.ll. (,rodkicwicz. Mater. Res. Bull. 4 (1969) 381.
[21 Li. Basile, JR. Ferraro and D. Cronert, J - lnorg. Nod.  Cliem. P. Caro,33(1971)1047. i. Derouct and P. Brun, Bull. So~ ClOm. 8 (1972) 3023. [4! A. Rulinont, j~cad.Roy. BeIg. (lasse Sci. Memo. (‘@11cclion 8,42 (1976)1.  J. FlölsS, P. Porcher, L. NOristO and P. Cam, (‘orlipt. Rend. (Paris) 290 (1980) 201.
161 P. Caro, J. Derouet, L. Beaury, U. Tcste de Sagcy, J.P Chaminadc, J. Aride and M. Pouchard, J. (Them. Plrvs. 74 (1981) 2698. [71 L. Mazza, A. landelli and F. Botti. Gazz. (‘him. hal. 7(1 (1940) 57.
1 8 1).
Table 3 Major impurities in NdOCI crystals
Brown, (lie flalides o I’ the Lan thanides and Ac— tinides (New York, 1968) p. 148. [9[ LU. Sillen and AL. Nylander. Svemssk Kcm. l’idskr. 53
[101 WI-I. Zachariasen, Acta Cryst. 2 (1949) 389. 
Al Si Ti Mg Ca K Na
10 15 20 50 70 10 60
D.H. Templeton and C.H. Dauben, J. Am. (‘hicm. Soc. 75(1953)6069. [121 P. Cam, J. Less-Coinmnon Metals 16 (1968) 367.  U. Brandt and R. Diehl. Mater. Rcs. Bull. 9 (1974) 411. 1141 B.M. ~Vanklyn, J. Muter, Sci. 7 (1972) 813. [151 AK. Baev and UI. Novikov, Russ. J. Inurg. them. 10 (1965) 1337. 116] I’. Ngok T’eii amid IS. Morozov, RLmss. J. Inorg. (‘hem. (1969) 1179.