SELECTED MIXED HALSDE PHASE: OF EUROPIUM AND mERBIu-M f Thesis for the Degree cm. 8. « MICK-{EGAN STATE UNIVERSETY BEAFR'ICE LOUISE CLINIC 1974 L1 HR A R y L} b-Iichigan State University r7“: -, 3? ‘ 'v w ‘wmn mg‘ v’fluns 5 anS“ BOOK 3mm me.” ‘ LIBRARY BINDERS I. @FSPO‘RT, MICHIGA! ‘ ABSTRACT SELECTED MIXED HALIDE PHASES 0F EUROPIUM AND YTTERBIUM By Beatrice Louise Clink Two mixed halide phases of europium have been prepared. They are EUC1(0.L»6 i 0.02)Br(1.56 i 0.07) and EuCl(0.16 i o.ou)Br(1.85 i 0.04)' A single crystal of the first phase was prepared. The unit cell was determined to be orthorhombic with lattice parameters: 0 3! An attempt was made to solve the structure based on the least squares refinement of the atomic coordinates of EuCl2 (PbCl2 structure type) with various concentrations of bromide ion substituted for chloride ion. An ytterbium mixed halide ammine complex of the empirical formula YbCl(O.31 i 0.03)Br(2.07 : 0.09)(NH3)(2.51 i 0A3) is believed to have been prepared. SELECTED MIXED HALIDE PHASES OF EUROPIUM AND YTTERBIUM By Beatrice Louise Clink A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Chemistry 1974 ACKNOWLEDGMENTS I wish to express my sincere appreciation to Dr. Harry A. Eick for the patience, encouragement and assistance he has so generously given. A deep expression of gratitude is also extended to my husband, Larry, for his support and understanding. My parents are acknowledged, in particular, for their hopes and encouragement in the attainment of this educational goal. Acknowledgement is also given to past and present members of the High Temperature Group for their fruitful discussions and willingness to help. The assistance and discussion which Dr. A. Timnick has offered is most appreciated. The financial support of the United States Atomic Energy Commission under contract number AT (11-1)-716 is gratefully acknowledged. ii I. II. E. III. TABLE OF CONTENTS INTRODUCTION BACKGROUND AND THEORETICAL CONSIDERATIONS Preparation of Anhydrous Lanthanide Halides Structural Information 1. Lanthanide Trichlorides and Tribromides 2. Lanthanide Dichlorides and Dibromides 3. Lanthanide Trichloride and Tribromide Hexahydrates Thermal Decomposition Studies 1. Lanthanide Trichloride Hydrates 2. Lanthanide Trichloride Ammine Complexes Amperometric Determination of the Chloride and Bromide Ion X-Ray Diffraction Analysis EXPERIMENTAL MATERIALS, EQUIPMENT, AND PROCEDURES Chemicals and Materials Handling Procedures Preparative Procedures 1. Anhydrous Lanthanide Trichlorides 2. Anhydrous Lanthanide Chloride Bromide Phases a. Sealed Bomb Preparation b. Pseudo Taylor—Carter Preparation 3. Single Crystal Preparation Elemental Analyses 1. Metal Analysis 2. Halide Analyses a. Total Halide Analysis b. Chloride Analysis 3. Data Reduction X—Ray Diffraction Analyses 1. Powder Diffraction Analysis iii HCDCD CID-PUP? w b0 9.; 16 19 19 19 20 20 22 23 24 24 24 24 25 25 27 30 30 TABLE OF CONTENTS (cont.) a. Sample Preparation b. Equipment 0. Data Reduction 2. Single Crystal X—Ray Diffraction Analysis a. Crystal Mounting b. Equipment and Parameters IV. RESULTS A. Europium Mixed Halides 1. Phase 1 a. Elemental Analyses b. X—Ray Diffraction Analyses €13 Powder Diffraction 2 Single Crystal X-Ray Diffraction Analysis c. Reaction Pathway 2. Phase 2 a. Elemental Analyses b. X-Ray Diffraction Analysis 3. Sealed Bomb B. Ytterbium Mixed Halide V. DISCUSSION VI. CONCLUSIONS AND SUGGESTIONS FOR FUTURE RESEARCH REFERENCES APPENDICES iv Page 30 3O 31 31 31 33 35 35 35 35 36 36 36 37 39 39 39 39 4O 42 49 52 APPENDIX LIST OF APPENDICES o I. Observed and Calculated sin29 (A = 1.54051 A) and Interplanar d-Values U1 b’ 7] PD Cd C) EUC1O.46(2)Br1.56(7) (Observed) EUC1O.46(2)Br1.56(7) (Calculated) EuC13-6H20-NHhBr Matrix Cubic NHuBr (a = 6.91 X)50 Monoclinic EuC13'6H202u EuClO 16(4)Br1 85(4) (Observed) Page 52 52 53 81st 56 2-5. 2—6. LIST OF TABLES Structural information on the lanthanide trichlorides and tribromides. Structural information on the lanthanide dihalides. Structural information on the lanthanide trichloride and tribromide hexahydrates. Temperatures (DC) at which the intermediate hydrates were stable in the thermal decomposition of the hydrated lanthanide chlorides. Temperatures (DC) at which the intermediate hydrates were stable in the thermal decomposition of the lanthanide trichloride hexahydrates. Temperatures (DC) at which the intermediate ammine complexes were found during the thermal decomposition study of LnClB-BNHB. Crystal data for EUC10.46(2)Br1.56(7) Analysis of Phase 1 (EuCler2_x). Analysis of Phase 2 (EuCler2_x). Analysis of the ytterbium mixed halide. Proposed reaction pathways Page 10 12 12 34 35 39 41 2-2. 3-1. 3-2. 3—3. 3—1.. 3-5- 4—1. LIST OF FIGURES Current-applied EMF curve. Reversed L amperometric titration curve. Halide preparation system. Halide titration vessel. Bromide oxidation setup. Powder pattern sample preparation. Single crystal mounting apparatus. R as a function of bromide concentration in Phase 1. vii Page 14 1t. 21 26 28 31 38 CHAPTER I INTRODUCTION Lanthanides and their compounds are used for cracking oil, as additives to increase the ductility of iron and steel, as mischmetal- iron lighter flints, mixed oxide polishing compounds and liners for military shells, as hosts and activators in phosphors, as magnetic and electronic materials, optical devices, alloy additives, carbon arcs, and as control rod materials or diluents in nuclear fuels.1 The anhydrous lanthanide chlorides are of particular importance because they are widely used for the production of metals and for starting materials from which many other anhydrous compounds are prepared. For theoretical reasons there has been interest in the thermo— dynamic properties of the lanthanide halides. The structure types of several of these halides have been determined and methods of"preparation have been described in the literature. The "simple" compounds (MX2 and MX3) of this series of elements have been shown to have both ionic and covalent character. Insight into the thermodynamics and the bonding involved with 4f shell elements would be gained by an analysis of the thermodynamic and structural data available plus a careful study of some lanthanide mixed halide phases. Stability relationships of the transition from MX to 3 M3 and MXZ to MY: (x and Y are nonidentical halide ions), metal-haIOgen bond strengths and the effect of anion size on packing may be established by such studies. It was, therefore, the intent of this work to prepare various lanthanide mixed halides and to study their structural relationship to the pure halide analogue. The europium and ytterbium systems were selected because they are known to exist in both the trivalent and divalent oxidation states, and as such allow the study of both MY2 and MY3 systems for the same metal. Europium and ytterbium halides are also representative of the different structure types which have been found for the lanthanide halides. Thermodynamic data are known for EuClz, EuBrZ, and to a lesser extent for YbCl2 and YbBrZ. For this reason, the chloride ion and bromide ion were selected as the halide ions, X and Y. CHAPTER II BACKROUND AND THEORETICAL CONSIDERATIONS A. Preparation of Anhydrous Lanthanide Halides Preparative methods have been reviewed by Taylor.2 Johnson and Mackenzie have compared many of these methods based on the purity of the product obtained.3 Taylor and Carter have described a general method for the prepara- tion of anhydrous lanthanide trihalides in which the trihalide hydrates are dehydrated in the prescence of an appropriate ammonium halide.“ The preparation of the europium bromides and hydrated bromides has been described by Haschke and Eick.5 Carter and Murray have also presented a method for the preparation of anhydrous rare earth trihalides. This method involves the reaction of excess mercuric halide with the rare earth metal.6 'Preparation or the growth of single crystals of anhydrous lanthanide halides has been described by Mroczkowski7 and Cox and Fong.8 Synthesis of the dichlorides of samarium, europium, and ytterbium from the anhydrous lanthanide trichlorides with zinc as the reducing agent in a zinc chloride melt has been described by DeKock andRadtke.9 Tanguy, Pezat, Fontenit, and Fouassier have prepared the mixed halide EuFCl by mixing the pure dihalides in stoichiometric amounts.10 A study of the reduced halides of most of the rare earth elements has been detailed by Corbett.11 B . Structural Information 1. Lanthanide Trichlorides and Tribromides Structural data for the anhydrous lanthanide trichlorides and bro- mides is summarized in Table 2-1. Of particular importance to this study is the structure of EuCl3 which is isostructural with the lanthanide trichlorides included in the group La-Gd. Typical of the Y(OH)3 or UCl structure type, the metal 3 ion is nine coordinated and is surrounded by six chloride ions which define a trigonal prism. Three other chloride ions lie in the same mirror plane as the metal ion, approximately perpendicular to the rec- tangular faces of the prism. The single crystal structure determination of EuCl3 and other lanthanide chlorides has been reported by Morosen.13 2. Lanthanide Dichlorides and Dibromides Structural information for the known lanthanide dichlorides and dibromides is summarized in Table 2-2. 17 2. The coordination sphere of the Eu is similar in its basic shape to that Baernighausen has reported the single crystal structure of EuCl of the trichloride. The chloride ions may be classified according to the types of holes they occupy. The first type may be described as a chloride ion surrounded by four europium metal ions forming a tetrahedron while the second as a chloride ion centered in a square pyramid of metal ions. In contrast to the metal ion coordination sphere in EuClz, its coordination sphere in EuBr is believed to be only eight coordinate. 2 5 Table 2-1. Structural information on the lanthanide trichlorides and tribromides. Species Structure Lattice Parameters Ref. Type a (2) b (X) c (X) 0L.B LaCi3 Y(OH)3a 7.4779 4.3745 13 LaBr3 Y(OH)3 7.967 3.54 12 Gaol3 Y(OH)3 7.451 4.313 12 CeBr3 Y(0H)3 7.952 4.444 12 PrCl3 Y(OH)3 7.42 4.26 12 PrBr3 Y(OH)3 7.93 4.39 12 NdCl3 Y(OH)3 7.3988 4.2423 13 NdBr3 PuBer 12.65 4.11 9.16 12 PmCl3 Y(OH)3 7.397 4.211 15 SmCl3 Y(OH)3 7.378 4.171 12 SmBr3 PuBr3 12.63 4.04 9.07 12 EuCl3 Y(0H)3 7.3746 4.1323 13 EuBr3 PuBr3 9.12 12.66 4.013 5 GdCl3 Y(0H)3 7.3663 4.1059 13 chr3 Fe013° 7.633 56.400° 14 TbCl3 PuBr3 3.86 11.71 8.48 15 TbBr3 Feel3 7.608 56.133 14 DyCl3 YCle 6.91 11.97 6.40 111.2 12 DyBr3 Feel3 7.592 55.834 14 80013 YCl3 6.85 11.85 6.39 110.80 12 HoBr3 FeCl3 7.576 55.666 14 Table 2-1 (cont'd). Species Structure Lattice Parameters Ref. Type a (X) b (X) c (X) (1.8 ErCl3 ml3 6.80 11.79 6.39 110.700 12 ErBr3 Feel3 7.568 55.466 14 TmCl3 ml3 6.75 11.73 6.39 ' 110.60 12 TmBr3 FeCl3 7.544 55-333 14 YbCl3 YCl3 6.73 11.65 6.38 110.40 12 YbBr3 Feel3 7.540 55.166 14 LuCl3 YCl3 6.72 11.60 6.39 110.40 12 LuBr3 Feel3 7.527 55.000 14 a. Y(OH)3 is isostructural with UC13, of hexagonal symmetry and belongs to the space group P63/m. b. PuBr 3 Ccmm. is of orthorhombic symmetry and belongs to the space group c. FeCl3 is of rhombohedral symmetry and belongs to the space group 83. d. YCl3 is isostructural with AlCl , of monoclinic symmetry and belongs 3 to the space group CZ/m. 7 Table 2-2. Structural information for the lanthanide dihalides. Species Structure Lattice Parameters Ref. Type a (3) b (A) c (A) NdCl2 Pb012a 9.06 7.59 4.50 18 SmCl2 PbCl2 8.99 7.55 4.51 16 Eu012 PbCl2 8.965(2) 7.538(1) 4.511(1) 17 EuBr2 SrBer 11.574 7.098 5 EuFCl PbFClc 4.1181(3) 6.972(1) 23 DyCl2 SrIZd 13.38 7.06 6.76 19 TmCl2 Srl2 13.10(4) 6.93(2) 6.68(2) 20 YbCl2 SrI2 13.150 6.942 6.693 21 YbBr2 CaClZe 6.63(2) 6.93(2) 4.37(2) 22 PbCl2 is of orthorhombic symmetry and belongs to space group anm. SrBr2 is of tetragonal symmetry and belongs to space group P4/n. PbFCl is of tetragonal symmetry and belongs to space group P4/nmm. SrI2 is of orthorhombic symmetry and belongs to space group Pbca. CaCl2 is of orthorhombic symmetry and belongs to space group Pnnm. 8 3. Lanthanide Trichloride and Tribromide Hexahydrates Structural information for the lanthanide trichloride and tri- bromide hexahydrates is tabulated in Table 2-3. The lanthanide trichloride and tribromide hexahydrates are iso- structural with monoclinic symmetry and belong to the space group P2/n. C. Thermal Decomposition Studies 1. Lanthanide Trichloride Hydrates Wendlandt and Bear25-27, Powell and Burkholder28, and Haeseler and Mattbes29 have studied the thermal decomposition of the lanthanide(III) chloride hydrates in air. They disagree over the composition of the intermediate hydrates in the decomposition process, however, they do agree that the lanthanide oxidechloride is the principle final product. A mole of HCl is evolved in favor of the last mole of water, (2-1). -H o -H O 2 2 LnCl3 6H20 T LnCl3 H20 T LnOCl + HCl (2-1) The thermal decomposition of the lanthanide trichloride hydrates in nitrogen was investigated by Ashcroft and Mortimer.3O Using a differ- ential scanning calorimeter, they defined three stable hydrate inter- mediates for the LaCl -EuCl 3 3 3 hydrate. The intermediate complexes contain three, two, and one water hydrates with the exception of SmCl of hydration, (Table 2-4). The stable intermediates found for ErClB- LuCl hydrates contain four, two, and one water of hydration. In all 3 cases, anhydrous lanthanide trichloride is reported as the principle final product. Other investigators, using similar experimental condit- ions, have found significant contamination of the anhydrous lanthanide Table 2-3. Structural information on the lanthanide trichloride and 9 tribromide hexahydrates. Species Space Symmetry Lattice Parameters Ref. Group a (X) b (R) c (3) e NdBr3 P2/n monoclinic 10.074 6.785 8.212 93.510 14 SmCl3 P2/n monoclinic 9.67 6.55 7.96 93.67 24 EuCl3 P2/n monoclinic 9.68 6.53 7.96 93.67 24 GdCl3 P2/n monoclinic 9.64 6.53 7.93 93.67 24 GdBr3 P2/n monoclinic 10.014 6.753 8.149 93.43 14 TbCl3 P2/n monoclinic 9.63 6.51 7.89 93.67 24 TbBr3 P2/n monoclinic 10.000 6.744 8.139 93.35 14 DyCl3 P2/n monoclinic 9.61 6.49 7.87 93.67 24 DyBr3 P2/n monoclinic 9.969 6.733 8.102 93.50 14 HoCl3 P2/n monoclinic 9.58 6.47 7.84 93.67 24 HoBr3 P2/n monoclinic 9.937 6.717 8.085 93.53 14 ErCl3 P2/n monoclinic 9.57 6.47 7.84 93.67 24 ErBr3 P2/n monoclinic 9.925 6.700 8.073 93.54 14 TmCl3 P2/n monoclinic 9.55 6.45 7.82 93.67 24 TmBr3 P2/n monoclinic 9.920 6.698 8.056 93.73 14 YbBr3 P2/n monoclinic 9.920 6.693 8.046 93.72 14 LuBr3 P2/n monoclinic 9.902 6.678 8.024 93.73 14 10 Table 2—4. Temperatures (DC) at which the intermediate hydrates were stable in the thermal decomposition of the hydrated lanthanide chlorides. (ref. 30) Element 7H20 6H20 5H20 4H20 3H20 2H20 1H20 LnCl3 La 40 90 100 130 180 Ce 30 95 115 140 220 PT 55 115 140 160 225 Nd 40 105 125 160 225 Sm 40 120 140 165 250 Eu 50 110 130 145 230 Cd 60 120 130 160 245 Tb 70 115 135 150 250 Dy 70 130 180 275 Ho 70 150 180 260 Er 60 135 155 185 250 Tm 70 140 165 185 240 Yb 70 135 160 180 240 Lu 70 135 160 180 220 11 trichloride product by the oxidechloride and dichloride. Il'in, Krenev, and Evdokimov have carried out a similar thermal decomposition study of selected rare earth halide hexahydrates in vacuum.31 There is a significant difference between the intermediate hydrated phases which they found and those which Ashcroft and Mortimer found30, (Table 2-5). They find the final phase of decomposition consists of a mixture of the anhydrous trihalide, the dichloride, and the oxidechloride. The proportion of oxidechloride found in the final product for the three systems increases in the order Yb—Eu-Sm. 2. Lanthanide Trichloride Ammine Complexes Ephraim and Bloch32, Beck and Gasse33, and Hflttig and DauschanBu have investigated the ammine complexes of selected lanthanide trichlo— rides. The results of their thermal decomposition studies are summa- rized in Table 2—6. It is uncertain from the data listed whether the phases are stable and pure at the temperature given. It appears from a comparison of the data on the ammine complexes to that tabulated on the hydrates that the ammine complexes are stable to decomposition at higher temperatures. Ammoniation energies for europium and ytterbium ammoniate ions [M(NH§)6++] have been reported by Frisbee and Senozan.35 Gaseous ammoniation energies are ~26O kcal/mol for Eli-*-+ and -281 kcal/mol for Yb++, (2-2). me) + 6 m3~m <2-2) 12 Table 2-5. Temperaturesa(oC) at which the intermediate hydrates were stable in the thermal decom osition of the lanthanide trichloride hexahydrates. (ref. 31) Element 6H20 5H20 4H20 3H20 2H20 1H20 Final Product Sm 50 65 100 180 300 400 Eu 4O 60 110 120 280 Yb 35 75 100 280 a. Initial plateau temperatures. Table 2—6. Temperatures (OC) at which the intermediate ammine complexes were found during the thermal decomposition study of LnC13'8NH3. (refs. 32 and 33) Element 8NH3 7NH3 6NH3 5NH3 4NH3 3NH ZNH3 1NH3 pNH3 La 68 7O 86 147 183 >265 Ge 64 70 102 135 244 >281 Pr 67 71 149 220 >310 Nd 49 71 209 283 >293 Cd 18 37 63 100 160 216 Er 70 100 197 259 >360 Y 25 90 180 335 13 D. Amperometric Determination of the Chloride and Bromide Ion Electroanalytical techniques normally give erroneous results due to mixed-crystal formation when the bromide and chloride ion are successively titrated. However, both the total halide and the individ- ual halide content can be determined quite precisely and accurately by amperometric titration with silver nitrate at a rotating platinum electrode. Amperometric titrations are performed by observing changes in the current between two electrodes as the titration is carried out. The principle of such titrations is inherent in current-applied emf curves or polarograms, (Figure 2—1). If one applies an increasing voltage to an polarized electrode immersed in a solution of substance to be examined, the current between the polarized electrode and reference electrode remains essentially zero until the reduction potential of the substance to be determined is reached (A). At this point the current rises rapidly as the emf is increased (B) until a maximum known as the limiting current is reached (C). Once the limiting current is reached, further increases in the emf cause little or no change in the current. The limiting current is proportional to the concentration of the reducible substance in solution. Indifferent electrolytes in the solution suppress current which would be due to the migration of an ion being examined. When titrating, one applies to the polarized electrode a voltage which is sufficient to reduce either one or both of the species involved in.the solution. In the silver nitrate-halide titration performed in this work, an emf which was sufficient to reduce the silver ion in solu- tion was applied to the polarized electrode. Essentially zero current should flow until an excess of silver nitrate is present in the solution. 14 .p c C) E U B A Applied EMF Figure 2-1. Current—applied EMF curve, .A .p a E d) ‘(,/’EndPOint ml of titrant Figure 2-2. Reversed L amperometric titration curve. 15 The current should then increase linearly with the concentration of the silver ion. Because the solubility of the silver halide formed is significant compared to the amount of excess silver nitrate present through the end point region, one must titrate to 80-100 % excess of silver nitrate and extrapolate back to find the true end point36, (Figure 2-2). The use of amperometric titrations for the determination of chloride, bromide, and iodide was first developed by Laitinen, Jennings, and Parks.37 Kolthoff and Kuroda later extended the method for more dilute solutions.38 Stock and Sienkowski apply amperometric titrations to the analysis of bromide-chloride mixtures.39 For completeness, it also should be noted that Rashbrook and Noodger use a dead-stop end- point titration(biamperometric titration) to determine trace amounts of chloride in sodium bromide.)+0 The analyses performed here are based on Stock's work.39 Stock first titrates the total halide amperometrically with silver nitrate. By using a second sample, the bromide ion is selectively oxidized with NaBrO3 and nitric acid according to (2-3). Br03- + 5 Br- + 6 H+ —-~ 3 Br‘2 + 3 H20 (2'3) The chloride content is then analysed amperometrically. A rotating platinum electrode serves as the polarized electrode. Rotating platinum electrodes are known to have high sensitivity, and are easily constructed and maintained. The hydrogen overpotential is smaller with this electrode than with a dropping mercury electrode, and an emf of up to 1 volt may be applied to the platinum electrode before it is subject to oxidative attack. 16 A saturated calomel electrode (SCE) serves as the reference electrode. Large non-polarizable electrodes are usually selected as reference electrodes. The SCE electrode has the advantage that the potential difference between it and the platinum indicator electrode is sufficient to reduce the silver ion. E. X-Ray Diffraction Analysis The phenomenon known as X-ray diffraction may be represented as a "reflection" of X-rays by successive planes of atoms in a crystal. More precisely, the diffraction pattern arises from the constructive inter- ference of waves scattered from adjacent parallel lattice planes. The path difference between waves must be an integral number of wavelengths according to the Bragg equation, (2-4). n). = 2d sin 0 (24+) where A is the wavelength of radiation, and 8, the angle of incidence. The symmetry and size of a unit cell determines the angles at which constructive interference of the diffracted X-rays will occur, while the relative intensities of these reflections depend on the arrangement of the atoms or ions within this cell. X-Ray diffraction patterns are described more easily in terms of reciprocal than real space. The reciprocal lattice may be related to the real lattice of a crystal by erecting normals to the two dimensional planes in the crystal. Given a common origin, the length of the normal is a distance i/dhkl from the origin. The reciprocal lattice points are then the points at the end of the normals. They are indexed according to the plane to which they are normal. 17 Indexing is done by integers (hkl), the Miller indices. The plane with Miller indices h, k, and 1 makes intercepts a/h, b/k, and 0/1 with the unit cell axes a,b, and 0. Because the electrons in a crystal are responsible for the scat- tering of X-rays, it is realistic to think of a crystal as a continuous distribution of electron density, p(x,y,z). The electron density is a function with the periodicity of the lattice and, therefore, can be written as a Fourier series in three dimensions as in equation (2-5). p(x,y,z) = i/V'i E E Fhkl expE—Zni(hx + ky + lz - ahkl)] (2—5) v is the volume of the unit cell, F(h k 1) is the structure factor, andahkl is the phase angle (2-6). (2-6) 2+sin 2n (hx + ky + lz) ahkl = arCtanLzzqcos 2n(hx + ky + 12) The relationship between the electron density and the structure factors is that one is the Fourier transform of the other. The summation is carried out over all values of h, k, and l, in order that there is one term for each set of planes (hkl) and for each diffracted spot. The intensity of scattered radiation is proportional to the square of the amplitude, |F(hkl)|2. One cannot calculate the structure factors directly from intensity data because the phase angle is not known and cannot be measured. The lack of information regarding the phase angle is often referred to as the phase problem. To determine the spatial arrangement of atoms, one may propose a structure by assigning spatial coordinates to all atoms, and compare the calculated structure factors, (Fc)’ to those observed (F0), or one may generate from |F(hkl)l2 a Patterson synthesis which has peaks corresponding to all 18 the interatomic vectors in the cell. From a knowledge of the inter- atomic vectors and the known symmetry of the cell, one may infer an initial set of atomic coordinates. One may also use a direct method to obtain initial values for the atomic coordinates. This initial set of atomic coordinates may be refined by least squares techniques. The R function (reliability or residual factor), as defined in equation (2-7), is used to evaluate the preciseness of the structure. “Fol " IFCI IF R = (2-7) 0| The least squares refinement procedure consists of systematically varying the atomic parameters to minimize the R function. Intensity data are corrected for the following: (a) backround, (b) absorption of the X-rays by the ions in the crystal, (c) polarization due to the monochromator, (d) different sweep rates of the reciprocal lattice points through the sphere boundary (Lorentz factor), and (e) the efficiency of the reflection of the X-ray beam (polarization factor). Anomalous dispersion corrections for phase changes due to the electronic nature of the electron are applied to the scattering factors. Atoms are known to vibrate at their lattice points; therefore, the "electron cloud" describing the atom is considered as dispersed over a volume of space. The correction term to the scattering factor for this effect is defined by B, the thermal parameter. One may define the electron cloud isotropically in terms of a sphere or anisotropically in terms of an ellipse. CHAPTER III EXPERIMENTAL MATERIALS, EQUIPMENT, AND PROCEDURES Ao Chemicals and Materials Chemicals and materials used were: (a) ytterbium and europium oxide, 99.9 %, Michigan Chemical Corp, St. Louis, MI; (b) argon and helium, Air Reduction Co. Inc., New Ybrk, NY; (0) nitrOgen, technical prepurified, Air Reduction Co. Inc., New York, NY; (d) silver nitrate, 99.9 %, J.T. Baker Chemical Co., Phillipsburg, NJ; (e) mercury, hydrochloric acid, and nitric acid, analytical reagent, Mallinckrodt, St. Louis MO; (f) gelatin, practical, Eastman Kodak Co., Rochester, NY; (g) potassium nitrate and ammonium bromide, reagent grade, Matheson Coleman and Bell, Norwood, Ohio; (h) potassium chloride and ammonium chloride, certified A.C.S., Fisher Scientific Co., Fair Lawn, NJ; (1) sodium bromate, reagent grade, Baker and Adamson, Allied Chemical, New York, NY; (j) vitreous carbon boats and crucibles, Beckwith Carbon Corp., Van Nuys, CA; and (k) quartz tubing, Englehard Industries, Inc., Hillsdale, NJ. B. Handling Procedures The storage and handling of air and moisture-sensitive materials 2, YbCl3 and the mixed halide phases were made in the :recirculating argon atmosphere glove box. The glove box has been des— such as EhCl cribed in detail by Hariharan.”1 Because of the difficulty encountered 19 20 in adjusting the internal pressure of the box, exact weighings of samples were generally done by difference outside of it. Care was taken to pour the argon atmosphere out of the weighing bottle prior to weighing, Weighings were made in less than 1 minute. Weight gains due to the hydration of the sample were not noticeable within this time period. Crystalline samples could be viewed under paraffin oil or in the single crystal mounting apparatus described later. C. Preparative Procedures 1. Anhydrous Lanthanide Trichlorides Both europium and ytterbium trichloride were prepared according to the method of Taylor and Carter“ as modified by Hariharan.41 Two grams of 11203 (M = Eu, Yb) were added to 30 m1 of 6 g HCl. The solution was heated on a hot plate and stirred until it became clear. Ammonium chloride was then added in a 6:1 molar ratio of ammoni- um chloride to lanthanide ion along with 50 ml of concentrated HCl. The solution was returned to the hot plate, stirred, and allowed to evaporate to a thick slurry, which was subsequently transferred to a carbon boat and placed in the apparatus as shown in Figure 3—1. The sample was heated at 200 0C under a stream of deoxygenated helium for several hours (x 16h) until excess water and water of hydration appeared to have evaporated. The temperature was then in- creased to 350 0C until all the ammonium chloride has evolved (x 7h). The carbon boat was shifted to the transfer tube; the transfer tube separated from the Vycor tube under a high helium flow, and capped. 21 .smpmam soapmammonm mufifimm .fium endmdm acne. So sammmwem .u poop nopndo .m 6959 Hooh> .m ooncnsh .n wasp Newmcdnfi .0 may» nomoufifiz dandfiq .m anpco mac .< \\\\\\in B \ \— s\\\\.\\s 22 The tube was evacuated and transferred to a dry box, where the sample was stored in a screw capped vial, which was then sealed in plastic under a dry argon atmosphere. 2. Anhydrous Lanthanide Chloride Bromide Phases Based on the known thermodynamics for the europium trichloride and tribromide systems, and thermodynamic functions for a mixed halide phase assumed to lie linearly between those of the pure phases, methods that use C12(g) or Br2(g) to chlorinate or brominate the desired lanthanide tri- halide are unfavorable for the preparation of pure mixed halide phases. However, the reaction (3-1) involving ammonium bromide and europium tri- chloride seemed favorable. 700 OK EuCl3 + 2 NHuBr EuClBr2 + 2 NH3(g) + 2 HCl(g) (3-1) (AG = -9.0 kcal mol'1)L’1"l’3 The sealed bomb preparation was based on these calculations. The acidic nature of ammonium halide is believed to prevent hydrol- ysis of the lanthanide trihalide in the Taylor-Carter preparation.)+ It is my belief that there is an exchange of halide ion between the ammon- ilnm halide and the intermediate phase, but no investigation presently .available defines this reaction pathway. The pseudo Taylor-Carter ;preperation of the mixed halide is based on the assumption of a halide irni exchange between the ammonium halide and the europium halide hydrate or' cofipmnpfie xooomopm anxious a flea cowonvfiz asap hopper coapooccoo manohoz HovoEEd 1 Managuoa one: Monaco Hcpos nsoqoanonam .Hommo> :oapdnpav ovaamz .sz on:Mam 27 02 diffusion into the solution. A large initial surge of current occurs when the electrode is initially rotated, and to protect the meter, the circuit across it is short-circuited (or the meter set to a high range). Aliquots of silver nitrate cannot be added while in a current measuring mode, with the electrode rotating. The total halide is titrated at zero applied potential (potential difference between the rotating platinum electrode and the SCE) until the excess of 0.01 N silver nitrate is 100 %. Stock solutions of unknown were made 0.01 N,in halide concentration and diluted to working solutions of 0.002 Nlin halide. Blanks comprised of the supporting electrolyte diluted to volume with distilled water and standard KCl solution were run with each set of titrations. b. Chloride Analysis An aliquot containing 0.12-0.24 pmol of chloride is measured into a 150 ml beaker. Five ml of 6 % nitric acid and an appropriate volume of 0.45 %isodium bromate solution is added. (An appropriate volume is 1.0 ml per 0.01 mmol of bromide, plus an excess of 1.4 ml.) The solution is heated on a steam bath for twenty minutes while dust free air is bubbled through it, (Figure 3-3). The solution is allowed to cool to room temperature, then transferred quantitatively to the titration vessel. One ml of gelatin solution and enough distilled water to bring the volume to 40 ml is added. The solution is then titrated as (lescribed for the total halide analysis. (Base currents of 0.2-5.0 “A 'were obtained under these experimental conditions.) Blanks and standards were treated as the unknowns with 1.4 ml of 0.45 % sodium bromide added. 28 Figure 3-3. B C A. Air inlet B. Class wool column C. Springs [__4 D. Glass sinter E. Beaker (150 ml) 4 F. Steam bath E (I53) Bromine oxidation setup. 29 3. Data Reduction The titration endpoint was determined by the intersection of the base line and the line formed by the set of points where the current is directly proportional to the silver ion concentration (80-100 %iexcess .Ag+). A linear least squares technique was used to determine the slope and intercept of each line. With each line mathematically described, the two equations were solved for their intersection. The value of X, ml of AgNO3 (Figure 2-2), at this intersection was taken to be the endpoint. Because the current rises rapidly for the blank, the base current before titrating was assumed to be constant and this value was used in place of a set of base line points. Care was taken not to use any points within the region where the concentration of Ag* changes significantly as a result of the solubility of AgCl. To calculate the concentration, the endpoints of the blanks were averaged and the average was used to correct the sample endpoints as described in equation 3-2, in which X is the ml of AgNO at the endpoint 3 of sample 1. tin...) = xi - (é xbkvn <3-2) By knowing the corrected values of X and using the normality of the K01 standard solution, the normality of the silver nitrate solution was calculated. These values were averaged and it was the average value (N) which was used to calculate the millimoles of halide per milligram of sample (H) according to equation 3—3. NX corr H = ._j.-.(___l (3-3) CS 30 where C is the concentration of the sample solution used in mg/ml and S is the sample size in ml. At least three samples of each unknown were titrated. The average value of H is reported along with its standard deviation. The metal analysis results are also reported in terms of moles of metal per gram of sample. Three samples were analyzed for each unknown. The average value and the standard deviation are reported. The empirical formula was calculated using the results of the metal, total halide and chloride analyses. E. X-Ray Diffraction Analyses 1. Powder Diffraction Analysis a. Sample Preparation water and air sensitive samples were prepared by placing a small piece of transparent tape, the size of the mounting disk hole, on a portion of PVC bag, (Figure 3-4). The sample and internal standard, Pt, were placed on the gum side of the tape. With the two flaps of the PVC bag held together with forceps, the sample was sealed with heat in plastic. The sample could then be removed and mounted on the disk. General sample preparation, film measurement, and Guinier techniques have oeen detailed by Stezowski.u5 b. Equipment Diffraction patterns of powdered samples were obtained with a Haegg TPype Cuinier forward focussing camera (radius 80 mm) and Cu Kb.1 radia- 0 tion, Pt internal standard (a0 = 3.9231 A), t = 24 i 2 °c. The X-ray 31 C—EB A. PVC bag " 4__ B. Transparent tape C. Sample Figure 3—4. Powder pattern sample preparation. source was a fine focus tube powered by a Picker 809B generator. 0. Data Reduction With platinum powder as the internal standard, the diffraction data were reduced with the least squares regression program of Lindqvist and Wengelin.l+6 2. Single prstal X-Ray Diffraction Analysis a. prstal Mountipg Crystals were transferred to a special capped transfer tube (Figure 3-5) in the glove box. Under a strong flow of argon, the transfer tube was uncapped and attached to the mounting apparatus. The entire apparatus was alternately evacuated and filled with argon. The last evacuation period was for a minimum of thirty minutes. The tube was then filled with argon and a crystal selected through the viewing tube. Under a heavy flow of argon the closing tube was removed and a 32 .mSpdemgm wedpcsos Hmpmhwo cameam ops» meamoao .H sweaaaaoo .m a seam Loan“ woam> .6 camp :oapooponm awmaaammo .m mspmnwmmw weaveso: . “season woaev ones noamswwy . m xooomopm essom> .n a o m sous“: mcfixoa> . don mmmau .n-m enemas 33 bent glass rod inserted in order that more crystals could be moved into the viewing field. An attempt was made to select a crystal which was clear, colorless, of uniform shape and lacking signs of fracture. Once a crystal was selected and somewhat isolated by using a glass rod, it was transferred to the entrance of the capillary. .A strong tap on the glass rod caused the crystal to fall into the capillary tube. The system was then closed and evacuated once again. Under argon flow the capillary protection tube was removed. The capillary was sealed with a small gas and oxygen flame. It was necessary to seal the capillary {we as close to the crystal as possible in order that it would be locked in ‘ place. The tube was then mounted in clay on the goniometer head. b. Equipment and Parameters Lattice parameters and intensity data measurements were made at a temperature of 23 t_3 0C with a computer-controlled, four circle, Picker goniostat with Mo Ka radiation (A = 0.7093 A), and a graphite monochrom- ator mounted adjacent to the X-ray tube. Thecu-scan technique was used for data collection with a scan range of 1.50 and a scan rate of 1.00/min. Both the hkl and Rkl octants were collected. Cell constants were obtained from a least squares refinement of 12 reflections which were hand-centered on the goniostat. Reflections from the Hkl and hkl octants were centered. PKCORR, an adaption of NUFACS written by Ibersu7, was used to correct the intensity measurements for backround. An empirical absorp— tion correction around the phi curve was applied using EMPASS. Lorentz and polarization corrections were effected with the program INCOR provided 34 by A. Zalkin. This program had been modified to include a correction for the graphite monochromator. The least squares refinement program was also provided by Zalkin. Table 3-1- Crystal Data for EuClo.46(2)Br1.56(7) -1 EuClO.u6(2)Br1.56(7), u = 239(6) g mol g_= 7.8797(35). b.= 4.611(2).,s== 9.198(4) Systematic Absences: Okl, k + 1 = 2n + 1; hk0, h >0 2n + 1 Space Group: ana, No. 62 Z = 4. F(000) = 502(11) e, y = 334.2(3) 33 u = 387.1 cm"1 (Mo Ka) D = 4.75 g cm-3 0 A(Mo Kn, graphite monochromator) = 0.7093 A CHAPTER IV RESULTS A. Egpgpium Mixed Halides 1. Phase 1 a. Elemental Analyses The pseudo Taylor-Carter preparation in which the initial heating period was at a temperature less than 200 oC yielded a white hydro- scopic powder. Halide and metal analysis showed the product to be EuC1O.46Br1.56' Results of the analysis are listed in Table 4-1. Table 4-1. Analysis of Phase 1 (EuCler2_x). Species moles/gram of sample moles of halide/mole of Eu ion Total Halide 7.12 i 0.27 x 10‘3 2.02 i 0.07 Chloride 1.62 i 0.07 x 10"3 0.46 t 0.02 Bromide 5024'9 i 0028 X 1.0-3 1056 t 0.07 Eu ion 3.512 i_0.005 x 10-3 35 36 b. X—Ray Diffraction Analyses (1) Powder Diffragtion The powder pattern is compared to that calculated from single crystal data in Appendices IA and IB. Close correlation between the two can be seen. (2) Single Crystal X-Ray Diffpaction Analysis From precession measurements and intensity data collected by use of a four circle goniostat, the unit cell was determined to be orthorhombic with lattice parameters: a = 7.8797 i_0.0035, b = 9.197 i,0.0043, and 7 c = 4.6111 t'O'OOZO 2. Because of the close similarity between the unit cell and extinctions observed in the intensity data of the mixed halide. phase and that of EuCl2 (a = 7.51, b = 8.93, c = 4.50 I)41, an attempt was made to solve the structure based on least squares refinement of the atomic coordinates of EuCl2 (PbCl2 structure type) with various concen- trations of bromide ion substituted for chloride ion. The positions used for the calculations were those of the space group anm, no. 62, as given by Baernighausen17 with x and y being inter- changed according to the choice of the principle axes. In this setting all atoms are in special positions defined by the symmetry relationships: X:Y.z (4—1) 1/2 - x, 1/2 + y, 1/2 + z (4-2) 1/2 + x, 1/2 - y, z (4-3) -x, -y, 1/2 + z (4-4) The chloride ions have two different points of symmetry. Negative 37 thermal parameters arose when a chloride ion was refined in position two, an indication that the electron density was too low. A bromide ion was substituted for the chloride ion in this position. The electron density was then increased further by progressively replacing the chloride ion scattering curve with that of a bromide ion until the R factor was minimized (and thermal parameters maximized). By plotting R as a function of the mole %.Bromide of the total halide and finding the best least squares parabola to fit the data (4-5), one finds the minimum at 75.1 % Bromide, which corresponds to an empirical formula of EUC10.498Br1.502' (Figure 4-1). y = 70.9 - 1.76x + 0.0117x2 (4-5) G. Reaction Pathway Samples were taken at various stages of the heating period. The samples were divided according to color. Sample (a) was an air-dried portion of the initial thick slurry which was transferred to the carbon boat; (b) a light orange phase which developed first on heating; (c) a dark orange phase which became bright orange on cooling; and (d) the light grey phase which appears after heating at 370 0C. Each sample was examined by powder X-ray diffraction. The powder pattern of sample (a) corresponds to a mixture of Eu013-6H20 and NHuBr (Appendices IE, 10, and ID) and that of (d) to what is normally obtained for Phase 1. However, the powder patterns of (b) and (0) could not be indexed successfully and no known phase was identifiable. 38 L— —( - '1 r ‘ . o r 10 - 7A a L. b ,— _. L. J 35 5 L— -1 s: o "-1 .p o _ -1 :1 61 CI: *— _1 I 4 l l J l l J l O 20 4O 60 80 100 Mole % Bromide of Total Halide Figure 4-1. R as a function of bromide concentration in Phase 1. 39 2. Phase 2 a. Elemental Analyses The white hydroscopic powder obtained from the pseudo Taylor-Carter preparation (initial heating temperature greater than 250 oC) had the empirical formula EuCl Results of the metal and halide 0.16Br1.85' analyses are listed in Table 4—2. Table 4-2. Analysis of Phase 2 (EuCler2_x). Species moles/gram of sample moles of halide/mole of Eu ion Total Halide 6.63 :_o.50 x 10‘3 2.01 :_0.04 CMMMe 5aoin3x1f“ 046idm Bromide 6.10 :_0.51 x 10'3 1.85 :_0.04 Eu ion 3.30 :_O.O83 x 10'3 b. X-Ray Diffraction Analysis The powder pattern of Phase 2 is characteristic of an orthorhombic cell similar to that of Phase 1 with increased lattice parameters. 3. Sealed Bomb Four colored phases including a dark yellow powder, a light yellow powder, a dark grey powder and dark grey crystals were found present in the quartz tube upon cooling. The dark yellow and dark grey phases which were heated under “'31 40 vacuum to remove excess ammonium halide decomposed into a dark grey phase. The X-ray powder pattern of the final dark grey phase was simi- lar to that of Phase 1 and Phase 2, characteristic of an orthorhombic cell with lattice parameters smaller than that of Phase 1. B. Ytterbium Mixed Halide A white hydroscopic powder was obtained from the pseudo Taylor— Carter preparation. The results of the elemental analyses are tabulated in Table 4—3. The results indicate that the bromide, chloride and ytterbium ion are not the sole constituents of the ytterbium mixed halide phase. Because thermal decomposition studies of ytterbium trichloride hydrates have shown the possible existence of a hydrate to be minimal at the experimental temperatures used; and because the presence of elements which have a molecular weight of 67 i_7 (which may be calculated from the results of the elemental analyses, assuming the Yb ion to be in a +3 oxidation state) cannot be justified, it seems likely that the remaining constituent of the system is ammonia. When the remaining fraction is considered to be ammonia, the empirical form- ula for the ytterbium mixed halide phase is mu0.3113"”2.07(NH3)(2.51 i 0.43)' 41 Table 4-3. Analysis of the ytterbium mixed halide phase. Species moles/gram of sample moles of halide/mole of Yb ion Total Halide 6.07 i_- 0.21 x 10'3 2.38 i 0.083 Chloride 7.88 i 0.80 x 10-4 0.31 10.031 Bromide 5.289 T. 0.23 x 10"3 2.07 3; 0.089 Yb ion 2.552 i 0.0067 x 10'3 CHAPTER V DISCUSSION As described earlier, there have been many methods reported for the preparation of anhydrous lanthanide halides. Many investigators, however, have experienced difficulty in preparing these compounds in pure form. Very few of the reaction pathways of the systems used have been investigated. The known methods generally involve either convert- ing the oxide to the halide with a suitable halogenating reagent or dehydrating the wet halide. Reed, Hopkins and Audrieth48 prepared anhydrous rare earth halides by the action of fused and solid "onium" salts in the lanthanide oxides, (5-1). La203 + 6 NHuCl-—*-2 LaCl3 + 6 N83 + 3 H20 (5-1) They believe the ammonium salts are acting as acids according to the BrUnsted definition. Because of its acidic nature, the excess ammon— ium chloride is believed to prevent hydrolysis of the lanthanum tri- chloride. Taylor and Carter“ also used an ammonium halide to prevent hydrolysis of the rare earth halide they had prepared by dehydration of the lanthanide hexahydrate, (5-2). vacuum LnO +HC1+HO LnCl o6H20+NHC1 2 3 2 3 LnCl + H 0-+ NHu01 (5-2) 4 A 3 2 42 43 The results of the thermal decomposition studies of the lanthanide halide hydrates have been outlined in Chapter II. Although the authors differ on which intermediate hydrate phases are formed, they both agreed that EuCl oHZO exists at 200 OC and the principle phase at 3 250 °C is EuClB. The results of thermal decomposition studies of the lanthanide trichloride ammine complexes have also been summarized in Chapter II. h—J Ammine complexes appear to be more stable to decomposition at higher temperatures than the hydrates. With these facts in mind, one might propose several reaction pathways for the pseudo Taylor-Carter preparations used in this work, (Table 5-1). Reaction pathway I is probably the easiest to eliminate. Neither Ascroft and Mortimer30 nor Il'in, Krenev and Evdokimor31 found the oxidechloride to be the principle product when the lanthanide trichloride hydrates are dehydrated in a deoxygenated system. Because the bromide ion is the stronger reducing agent, we would expect the final product to contain a 1:1 ratio of bromide ion to chloride ion rather than the 3.4 and 12 ratio which was found for the europium mixed halide system. The only difference in the experimental conditions for the prep- aration of Phase 1 and Phase 2 is the temperature of the initial heat— ing period. Phase 1 was obtained when the EuC13-6H20-NHhBr matrix was initially dehydrated at a temperature of 200 oC, and Phase 2 when the initial temperature was greater than 250 0C. It has already been determined from the thermal decomposition studies, that EuCl -H 0 3 2 should be the principle phase during this period at 200 oC and EuClB, the principle phase at a temperature greater than 250 oC provided that mmz x + Nam m + H-wwmw-maosm .11. N I N m 6 mm N n Sr x + ONm + mashwwmwASemalll 83% x + o x. 851ml. 0 me. 84a .3 co own so OON am m2 oo OON Hoarz- ONmo- -a s- a a- ANVthm + a um maze ..|I|. Baez s + Hm mace .III. 832 a + mass ill... ONmo.mBaa .HHH 6 com o oaN wmsrz O 0 oo osN a N -s a- do - a a- 0 mm- Amvmnmm + H Hm macaw {8an A + omm. Hm mausmollwmamz a + 09.36% :IIII ommm.maoafi .HH no can so OON nmsrz oo ooN Hum- N N m N a 0 mm- N m 30 r + 3 E N + .886 olllwm m2 N + Bose :lmlo 5. Sam .H uo OON hm m2 oo OON .mmmzzosmm codvooon bomomonm .Hum 638m. 45 there is no interaction with the ammonium halide. Pathway II might then be suitable to describe the preparation of Phase 1 and pathway III for the preparation of Phase 2. Very little is known concerning the crystal structure of the lanthanide trihalide monohydrate. Crystal data for the monohydrate of europium tribromide is given by Haschke and Eick.5 Although the hexahydrates of the lanthanide trichlorides and tribromides are iso- morphic, the anhydrous lanthanide trichloride and tribromide belong to different structure types, (Chapter II)- The anion: cation ratio is more significant for the close packing of the anhydrous compounds than for the hexahydrates. Therefore, correlations drawn between the monohydrate of europium tribromide and that of europium trichloride would be nebulous, and it is difficult to draw any further conclusions concerning the correctness of pathway II. A single crystal structure determination of EuCl was done by 3 Morosin13, (Chapter II). Four of the chloride ions in the metal ion coordination sphere have metal-chloride bond distances of 4.492 2, two have bond distances of 2.835 2, and there is one with a M-Cl bond distance of 2.919 3. The ionic radius of Eu(III) is 0.950 315, while Pauling lists the ionic radius of bromide ion as 1.95 g. The sum of the ionic radii is 2.90 A. A bromide could easily substitute for the four chloride ions with bond distances of 4.492 X, The empirical form- ula for such a compound would be Eu011.28Br1.72. There is the possibil- ity of further substitution of a bromide ion for the chloride ion with a metal bond distance of 2.919 2. The empirical formula of this compound would be Eu610.86Br2.14' In either case, we would expect the bromide ion to be the reducing ion, and the final divalent halide, from 46 the decomposition of these trivalent europium mixed halide phases, to be richer in chloride than either phase prepared, which does not support pathway III. A similar study of the EuCl structure shows one metal-chloride 2 o o 0 bond each of 3.046 A, 2.994 A, and 2.916 A and two metal-chloride bonds o o of 3.440 A and 2.925 A. The ionic radius of Eh+2 is reported to be o between 1.21 and 1.29 A depending on the anion.17 In this structure 56 % of the chloride ions lie in the center of a square pyramid of metal ions and 44 % lie in a tetrahedral hole. As would be expected, the M-Cl bond distances greater than 3.0 2 are for those chloride ions in the tetrahedral holes. If a bromide ion were substituted for seven of the nine chloride ions in the coordination sphere, the empirical form- ula for such a compound would be EuClO “#Br The R function 1.56' in the single crystal analysis of Phase 1 was minimized for a compound with the empirical formula EuCl with bromide ion substituted 0.5Br1.5 entirely for the chloride ion in position two. The elemental analyses showed Phase 1 to be EuCl(O 46 + 0 02)Br(1 56 For bromide ions 1 0.07)' in eight of the nine positions, the empirical formula of such a compound would be EuCl 8’ analogous to Phase 2, 0.223r1.7 8101(046 i 0.04)Br(1.85 i 0.04)’ The reduction of the anhydrous europium trichloride to the dichlo- ride is not known to be significant under the experimental conditions used in this work. One would not expect EuCl to be a major intermedi- 2 ate, yet the structure may be used to account for the particular phases obtained. It is interesting to note that even with the high concentration of bromide ion, the unit cell appears to retain its orthorhombic symmetry. 47 The final proposed pathway, IV, is through an ammonia complex, but not necessarily the one given in IV. The structural chemistry of YbCl is isomorphic with that of YC13, while the structure of EuCl3 is 3 isomorphic with that of NdCl and GdCl (Table 2-1). The thermal 3 3' decomposition temperatures of ammine complexes of NdClB, GdCl3 and Y013 probably are similar to those of EuCl3 and YbCl3 respectively. From the data tabulated in Table 2-6, an ammine complex of europium trichloride would not exist above 360 0C (the final temperature used in the Taylor— Carter preparation), but a ytterbium mixed halide ammine complex may exist. The results of the ytterbium mixed halide preparation supports this contention. CHAPTER VI CONCLUSIONS AND SUGGESTIONS FOR FUTURE RESEARCH One may conclude from this work that the substitution of the bromide ion for the chloride ion was selective rather than random as originally thought. This conclusion suggests that the space group used for the least squares refinement of the atomic parameters was of too high symmetry. Redefining the data in terms of a lower space group may increase the refinement. Further refinement was not attempted pre— viously because it was believed that the empirical absorption correction, which was applied because of the irregular shape of the crystal was not representative of the real absorption problem. The intermediate phase then, should also require selective substitution of the chloride ion. An ammine complex of the lanthanide trihalide is a likely candid- ate for an intermediate phase. If an ammine complex could stabilize the trihalide during dehydration, by using other complexing agents, various anhydrous compounds may be prepared by a more direct route. A Mbssbauer study of the reaction intermediates or a detailed analysis of the structure of the ammine complexes and the monohydrates of the lanthanide trichlorides could better define the reaction pathway. REFERENCES 10. 11. 12. 13. 14. 15. 16. 17. REFERENCES 0. B. Michelsen (ed.), "Analysis and Applications of Rare Earth Materials", p 1, Universitetsforlaget, Oslo, Norway, 1973. M. D. Taylor, Chem. Rev., 62, 503 (1962). ,1 K. E. Johnson and J. R. Mackenzie, J. Inorg. Nucl. Chem., 2. 43 (1970). i M. P.)Taylor and C. P. Carter, J. Inorg. Nucl. Chem., 24, 387 (1962 . %. M.)Haschke and H. A. Eick, J. Inorg. Nucl. Chem., 32, 2153 1970 . F. L. Carter and J. F. Murray, Mat. Res. Bull., 2, 519 (1972). s. Mroczkowski, J. Cryst. Growth, Q, 147 (1970). D. E. Cox and F. K. Fong, J. Cryst. Growth, 20, 233 (1973). %. W.)DeKock and D. D. Radtke, J. Inorg. Nucl. Chem., 3;, 3687 1970 . B. Tanguy, M. Pezat, C. Fontenit and C. Fouassier, C. R. Acad. Sc. Paris, Series C, 222, 25 (1973). J. D. Corbett, Rev. Chimie Min., 19, 239 (1973). R. W. G. Wyckoff, "Crystal Structures", vol 2, 2nd ed, Interscience Publishers, New York (1963). B. Morosin, Jr Chem. Physics, 42, 3007 (1968). D. Brown, S. Fletcher, D. G. Holah, J. Chem. Soc. (A), 1889 (1968). D. Brown, "Halides of the Lanthanides and Actinides", p 154, John Wiley and Sons Ltd., New York (1968). w. Doll and w. Klemm, z. Anor . All . Chem., gg1, 239 (1939). H. Barnighausen, Rev. Chimie Min., 10, 77 (1973). 49 18. 19. 20. 21. 22. 230 24. 25. 26. 27. 28. 29. 30. 31. 32. 33- 34. 36. 37. 38. 39. 50 L. F.)Druding and J. D. Corbett, J. Amer. Chem. Soc., 83, 2462 1961 . J. D. Corbett and B. C. McCollum, Inorg. Chem., 5, 938 (1966). P. E. Caro and J. D. Corbett, J. Less-Common Metals, 18, 1 (1969). N. A. Fishel and H. A. Eick, J. Inorg. Nucl. Chem., 33, 1198 (1971). H. Bdrnighausen, H. P. Beck and H. W. Grueninger, Proc. Ninth Rare Earth Conference, Blacksburg, VA, 1971. L. H. Brixner and J. D. Bierlein, Mat. Res. Bull., 2; 99 (1974)- i ‘ E. J. Graeber, G. H. Conrad and S. F. Duliere, Acta. Cryst., ; . 31, 1012 (1966). w. w. Wendlandt, J. Inorg. Nucl. Chem., 5, 118 (1957). w. w. Wendlandt, J. Inorg, Nucl. Chem., 9, 136 (1959). w. w. Wendlandt and J. L. Bear, Anal. Chem. Acta.. 21, 439 (1959). J. E. Powell and H. R. Burkholder, J. Inorg. Nucl. Chem., 14, 65 (1963). G. Haeseler and F. Matthes, J. Less—Cgmmon Metals, 2, 133 (1965). S. J. Ashcroft and C. T. Mortimer, J. Less-Common Metals, 14, 403 (1968). V. K. Ll'in, V. A. Krenev and V. I. Eudokinov, Russ. J. Inorg. Chem., 12. 1497 (1972). T. Ephrain and R. Bloch, Chem. Ber., 59, 2692 (1926). G. Beck and A. Gasser, Anal. Chem. Acta., 3, 41 (1949). G. F. Huttig and w. Dauschan, Monatsh, 8g, 742 (1951). R. H. Frisbee and N. M. Senozan, J. Chem. Phys., 52, 1248 (1972). J. T. Stock, "Amperometric Titrations", Interscience Publishers, New York, 1965. H. A. Laitinen, W. P. Jennings and T. D. Parks, Ind. Eng. Chem., 18. 355 (1946). I. M. Kolthoff and P. K. Kuroda, Anal. Chem., g3, 1307 (1951). J. T. Stock and R. P. Sienkowski, Microchgm. J., 2, 157 (1965). 40. 41. [4'20 43. 45. 46. 47. 49. 500 51 R. B. Rashbrook and S. C. Woodger, Anal st, 88, 479 (1963). A. V. Hariharan, Ph.D. Thesis, Michigan State University, East Lansing, MI, 1971. "Janaf Thermochemical Tables", D. R. Stull and H. Prophet, Project Directors, Dow Chemical Co., Midland, MI, 1971. C. E. Wicks and F. E. Block, U.S. Bureau of Mines, Thermodynamic Properties of 65 Elements- Their Oxides, Halides, Carbides, and Nitrides, No. 605, U.S. Department of the Interior, washington, Deco, 1963. J. M. Hashke, Ph.D. Thesis, Michigan State University, East Lansing, MI, 1969. J. J. Stezowski, Ph.D. Thesis, Michigan State University, East Lansing, MI, 1968. 0. Lindqvist and F. Wengelin, Ark. Kemi., 88, 179 (1967). J. A. Ibers, Inorg. Chem., 6, 197 (1973). J. B. Reed, B. S. Hopkins and L. F. Audrieth, J. Amer. Chgp. Soc., 57. 1159 (1935). L. Pauling, "The Nature of the Chemical Bond.and the Structure of Molecules and Crystals", 3rd ed, Cornell University Press, Ithaca, N.Y., 1960. G. Bartlett and I. Langmuir, J. Amer. Chem. 500-. £33 84 (1921)- APPENDICES .n’ _L! 0 Appendix I: Observed and Calculated sin28 (A = 1.54051 A) and Interplanar d-Values Appendix IA: EuCIOJ+6<2>Br1 56(7) (Observed) Rela- Rela- 13.15:..- d Iii 1:12.- d I? sity sity m 0.0371 4.000 vvw 0.2074 1.692 w .0378 3.962 vvw .2112 1.676 m .0439 3.684 vw .2215 1.636 m .0648 3.027 w .2243 1.626 w .0655 3.009 vw .2392 1.555 s .0725 2.861 vw .2592 1.513 m .0900 2.567 vw .2634 1.501 w .0924 2.534 vvw .2650 1.494 m .0995 2.440 vw .2724 1.476 m .1107 2.315 vw .2869 1.438 mw .1132 2.289 vw .2942 1.420 w .1201 2.223 vvw .3213 1.359 vw .1485 1.999 vvw .3281 1.345 vvw .1523 1.974 vvw .3348 1.331 vvw .1761 1.836 vvw .3410 1.319 vvw .1800 1.815 VW .3496 1.303 vvw .1828 1.802 vvw .3770 1.255 vvw .2032 1.709 52 53 Appendix IB: EUC10.46(2)Br1.56(7) (Calculated) a . 1;?” a . {3:7 101 0.0375 3.980 151 0.2128 1.670 120 .0376 3.972 042 .2238 1.628 111 .0445 3.653 322 .2257 1.621 030 .0631 3.066 142 .2415 1.568 121 .0655 3.009 500 .2389 1.576 130 .0727 2.857 332 .2607 1.509 031 .0910 2.553 013 .2582 1.516 310 .0930 2.526 402 .2645 1.498 131 .1006 2.429 440 .2651 1.496 002 .1116 2.306 412 .2715 1.478 040 .1122 2.299 511 .2738 1.472 301 .1139 2.282 053 .2869 1.438 311 .1209 2.215 123 .2887 1.433 330 .1491 1.995 521 .2948 1.419 400 .1529 1.970 133 .3238 1.354 331 .1770 1.831 432 .3256 1.346 401 .1808 1.812 450 .3282 1.345 420 .1809 1.811 303 .3371 1.327 132 .1843 1.794 360 .3385 1.324 051 .2032 1.709 502 .3505 1.301 421 .2088 1.686 442 .3767 1.255 49 Appendix 10: EuCl 3 °6H20-NHuBr Matrix Rela- Rela- i:::.. a d 28% 1:12.- ”11’8” sity sity mw 0.0146 6.375 8 0.0731 2.850 mw .0164 6.009 vvw .0956 2.492 w .0228 5.097 vvw .1053 2.374 w .0252 4.853 mw .1094 2.329 w .0282 4.589 vw .1112 2.310 w .0302 4.435 mw .1461 2.015 ms .0365 4.033 mw .1826 1.803 mw .0460 3.591 mw .2191 1.646 mw .0501 3.443 vvw .2917 1.426 w .0613 3.111 vw .3288 1.343 vvw .0643 3 .037 vw .3648 1.275 Appendix ID: Cubic NHuBrSO (a = 6.91 X) 110 .0248 3.989 222 .1491 1.995 111 .0373 3.990 322 .2112 1.676 200 .0497 3.455 422 .2982 1.410 211 .0746 2.821 510 .3231 1.355 220 .0994 2.443 520 .3603 1.283 300 .1183 2.303 Appendix IE: Monoclinic EuCl 3 55 24 -6H20 hkl sinze d :giue hkl sinZG d zgiue -101 0.0148 6.338 -203 0.1041 2.387 101 .0168 5.952 113 .1079 2.345 011 .0233 5.044 320 .1129 2.293 200 .0254 4.830 -313 .1469 2.010 -111 .0287 4.548 330 .1824 1.803 111 .0307 4.399 -521 .2191 1.646 -211 .0468 3.562 -424 .2920 1.426 211 .0507 3.420 701 .3279 1.345 120 .0620 3.093 -711 .3279 1.345 021 .0651 3.020 504 .3292 1.343 -212 .0730 2.851 151 .3646 1.276 -122 .0976 2.465 116 .3647 1.276 103 .0939 2.513 56 Appendix IF: EMC10.16(4)Br1.85(4) (Observed) Rela— Rela— tive d value tive d value Inten- Sin 9 (2) Inten- sin 9 (2) sity sity vw 0.0329 4.247 w 0.1251 2.178 w .0337 4.193 vw .1294 2.141 w .0359 4.066 vvw .1412 2.050 w .0371 3.998 vvw .1450 2.023 w .0564 3.245 vvw .1506 1.985 w .0628 3.073 vvw .1721 1.857 w .0642 3.040 vw .1773 1.829 w .0697 2.919 vvw .2153 1.660 m .0710 2.891 vw .2188 1.647 m .0875 2.604 vvw .2516 1.536 w .0889 2.584 vvw .2584 1.515 w .0911 2.552 vvw .2619 1.505 w .0921 2.538 vw .2687 .1486 vvw .1006 2.429 vw .2795 1.457 w .1070 2.355 W .2849 1.443 w .1114 2.307 vvw .3382 1.324 vw .1179 2.243 it: IN 11117 1111111411 11111111111 3 6 557 4 o 3 o 3 9 2 1 3