A STUDY OF THE THERMAL DECOMPOSITION OF THE CARBONATES‘ AND OXALATES OF SOME RARE EARTH ELEMENTS BY DIFFERENTIAL THERMAL ANALYSIS Thesls for the Degree of DI}. D. MICHIGAN STATE UNIVERSITY Carl Barnes BishOp 1959 III;JlulgulzuynyllnuznlwflImmxw : IVTESI_J RETURNING MATERIALS: PIace in book drop to LIBRARJES remove this checkout from Ail-(SIIIL. your record. FINES wiII be charged if book is returned after the date stamped below. u: ”mm—«987': ‘1 4" fjflc , MICHIGAN STATE UNIVERSITY EAST LANSING, MICHIGAN A STUDY 0" n3 xxaxAL DECORPOSITION CF I \ ' ‘ “3 CF sous RARE Tl.:l CALSESOZEA’ 2233 AND OIL/MT EARTH ELEKENTS BY DIFFEREHTIAL THERIAL ANRLYSIS By CARL EAREES BISHOP A TfiibIS Submitted to the fiichI ---L School of Graduate Studies of can fitate University of Agriculture and Applied Science in partial fulfillment of the requirements for the degree of DOCTOR OF iv! 5. m ’I ILO”OL"III Department, 02102.71 stI‘y of 1959 A054; ‘TO'JLEDGLJSN T The author wishes to eXpress his appreciation to Dr. L. L. Quill for his stimulating direction and en- couragement throughout this research program. The author would like to express his thanks to Mr. Frank Bette for his advice and able assistance in the construction and assembly of the equipment. The author would 1 he to express his gratitude to the faculty of the Chemistry Department for their helpful suggestions during these studies. The author would like to express his appreciation to the All-University Research Fund for financial assistance in completion of this work. To my Wife VITA The author was born in Banberg, South Carolina, on August 30, 1932. He received his secondary and high school education in the Bamberg schools, gradu- ating from Bamberg High School in June, 1950. The author received the Bachelor of Science degree from Clemson A. and M. College with a major in Chemistry. He entered Graduate School at Michigan State University in September 1954 and held Graduate Teach- ing Assistantships in Chemistry. is also obtained support on his research through the University Re- search Fund. 1 '\ 'rffI’N‘ -._ " If AL Jars-MILL; «L f f "P ‘f‘ ‘~"‘ 3'!" "H ‘ ~' 11’ h ' "If" ‘.‘-(." "I T" "I I"?! PVT' f“ 'r‘-."‘ ""s"‘~." , b - A . ‘> y. :- Ii .' 1». ‘ - .- . - , v . . . A u' a. Li. JA. (1.... .L LA... J. 1.6...-‘ 5. MM” ;)i..«u s“. .4'. {.4 n» a .L\.n.| 1,. 3.4....) \.‘.3sm.»U-uf$.a...-AJ fl . l . 'V- '.' 4'" . (._ h '. ’, a , ~ . r ,_ 'pl .'- . - «. 3‘ « .1... Q .' . .. . . . t. . I"\'~_ .,’ . ‘. a"? ‘ ‘ . . . . V.» Ira. T ’I‘ -.~ ‘ is , ' . '.~- -. , .fi ,~ :~ '5 ; i ‘. .9 I '- , . . , ~ -. . . ., 121.41 J cud a‘p’ pi -."J U.‘ Nu) V '0; ‘u‘u 5....4 AJL h’l-b‘i fi- ~.u.~ e-rcor‘ .5- vs) I ,, v .:c 1L 1- "'5 L..- > IL n'n-- 0‘ ’ buy-.4! c '1'}; a. L 5111.. 35.5.1.1 L; l a . by Carl Barnes Eishoy The the wzeal deconposition of the carbonates and oxalates of lanthanum, cerium, ncodyaiun, praceodyniun. sacarium and yttrium was studied by the "ferentisl thermal analysis technique. A ecciirtioa for the construction ands paration of equipment for differential thernel alalysa s :as .he carbonates were prepared by decomposing so1u~ tions of the tricl.loroacetates of the clcucnts stuJici. Cnly yttrium indicated a stable hyirate. The other ccrzonates were arthrous after air iryirg for forty- cight hours. The carbonates {smelt lly decosposc as Ln icated by the equation 32(1‘3-03):5 The oxalates were precipitated free: tdilute nit. rat e solutions by :Eilute 0: cells aci iecahgdratcs were is uei for all the rare earth are ates studied. Lanthanuu also formed a hexahych etc which woull <5 econ pose to a tetrahyirate. Intermediate hydrates were identified from the decahydrates of neodgsiun, camarius n1 yttrius. The 02 :alates generally deco3; 0161 as Abstract - page 2 Carl Barnes Bishop indicated by the equation R2(0204)3___ R203o2C02_.—_ R2030C02 .......... R203 A reversible crystal conversion at 960°C was found for praseodymium oxide (Pr5011). X-ray powder diffraction patterns, chemical analysis, and microsCOpic photography were used to identify the compounds. TABLE OF CONTENTS I Introduction II History Differential Thermal Analysis. . . . . . . . . 1 Sample Holder . . . . . . . . . . . . . . .3 Thermocouples . . . . . . . . . . . . . . .5 Furnace Design . . . . . . . . . . . . . . 7 Furnace Temperature Control . . . . . . . .7 Thermocouple Potential Recording Equipment 8 Calibration of Reactions . . . . . . . . . 9 Incrt Thermal Standards . . . . . . . . . .9 Applications of DTA . . . . . . . . . . . .9 Oxalates . . . . . . . . . . . . . . . . . . .10 Preparation of Oxalates . . . . . . . . . lO Decomposition of Oxalates . . . . . . . . l2 0 O O O I O O O O O O O O .18 Carbonates . . . III EATERIALS AND HETHODB Construction of Equipment . . . . . . . . . . 20 Sample Holder . . . .'. . . . . . . . . . 20 Thermocouples . . . . . . . . . . . . . . 24 Furnace Base . . . . . . . . . . . . . . .25 Furnace . . . . . . . . . . . . . . . . . 28 Power Supply . . . . . . . . . . . . . . .31 Recorders . . . . . . . . . . . . . . . . 33 Control Panel . . . . . . . . . . . . . . 34 Assembly . . . . . . . . . . . . . . . . .34 Operation.cocoon-0000000035 IV Preparation of Compound . Source of Material . Carbonates . . Oxalates . . . Analytical Techniques . Ignition . . . Oxalate Titration . Carbon Dioxide Content X-Ray Powder Diffraction Patterns Experimental and Results Lanthanum Carbonate . Cerium Carbonate . Praseodymium Carbonate . Neodymium Carbonate . Samarium Carbonate . Yttrium Carbonate Lanthanum Oxalate Hydrates . . . O O O Oxalate Decomposition . Cerium Oxalate . . Praeeodymium Oxalate . Neodymium Oxalate Hydrates . . . O Anhydrous Oxalates . Samarium Oxalate . Yttrium Oxalate . O O . 40 . 43 . 45 . .50 I .65 VI VII VIII Conclusions . . . . . . . . . . . . . . . . . 70 Summary . . . . . . . . . . . . . . . . . . . 76 Literature Cited . . . . . . . . . . . . . . .77 Appendix X-Ray Diffraction Patterns. . . . . . . . . 1 Differential Thermal Analysis Curves . . .111 FIGURES Page Figure I: Typical Differential Thermal Analysis 2 Curve Figure II: Thermocouple Circuits 6 Figure III: A Side View and 3 Sketch of the Parts of the Sample Crucible 23 Figure IV: Jig Used in Making Differential Thermo- 25 couples Figure V: Technique for Mounting Fixtures in Fur- 27 nace Tu?- Figure VI: Collapsible Core for making Furnace Heat- 29 ing Equipment Figure VII: Electrical Wiring Diagram for Equipment 37 Figure VIII: Vacuum Residue Bottle 38 PHOTOS Photo 1: Sample Crucible, Thermocouples, and Fur- 21 nace Base Photo 2: Sample Crucible, Thermocouples, and Fur- 22 nace Base Photo 3: Power Supply Control 32 Photo 4: Assembly of Differential Thermal Analysis 36 Equipment Photo 5: Lanthanum Carbonate 46 Photo 6: Neodymium Carbonate 46 Photo 7: Samarium Carbonate ‘ 46 Photo 8: Lanthanum Oxalate Decahydrate 48 Photo 9: Lanthanum Oxalate Hexahydrate 48 Photo 10: Cerium Oxalate Decahydrate 48 Photo 11: Praseodymium Oxalate Decahydrate 49 Photo 12: Neodymium Oxalate Decahydrate 49 Photo 13: Samarium Oxalate Decahydrate 49 INTRODUCTION The decomposition of oxygen-containing anion salts such as carbonates, oxalates, sulfates, nitrates, etc. of the rare earth elements has been studied for many years. In the past, two techniques, the measurement of pressure created by the reaction products and the deter- mination of weight loss as the temperature of the salt was increased, have been used primarily. These techniques required very slow heating rates to allow equilibrium to be maintained. Differential thermal analysis was used in this study to detect thermal changes which accompany decomposition reactions or crystalline inversion. X—ray powder diffraction patterns were used in conjunction with chemical analyses to identify the intermediate compounds formed during decomposition. The purpose of this study was to determine the inter— mediate products of the decomposition and the temperatures of decomposition of the oxalates and carbonates of yttrium, lanthanum, cerium, praseodymium, neodymium and samarium. HISTORY Differential Thermal Analysis Le Chatelier 1 has been given credit for first using temperature measurements in 1887 to study thermal reactions. Only one thermocouple was used in a sample which was heated at a rather rapid and uniform rate. At short intervals, galvanic readings were taken and compared with readings taken when no sample was present. Variations in the read- ings indicated that a thermal reaction had taxen place in the sample. In 1899 Roberts-Austen2 devised a simple differential thermocouple circuit for measuring the difference in the temperatures of a sample and an inert reference material thereby determining when a thermal reaction occurred. This technique was used extensively in metallurgy. In 1913 in a study of silica minerals, Fenner 3 was the first to use this newer technique to study thermal reactions out- side of metallurgy. This technique, called differential thermal analysis, can be described as a method by which thermal reactions in a material can be studied by determining the differences1 in the temperatures of that material and some thermally inert standard while both are being heated to higher temperatures at a rather rapid constant rate. The differential thermal analysis curves, which are usually plots of the temperature difference between the sample material and the inert standard versus the temperature of the inert standard or the surroundings in the furnace is a record of the dynamic thermal reactions taking place within the sample material. The curves obtained during differential thermal analysis may exhibit peaks on one or both sides of the bass line. (The base line is the line indicating the differential temperature values obtained when no reactions occurred. The base line may drift away from a horizontal line if the sample and the inert standard material differ greatly in specific heat or conductivity of heat). These peaks above or below the base line represent endothermic or exothermic reactions depending upon the arrangement and electrical wiring of the thermocouples. Figure 1 shows a typical curve obtained in this study. 9’ C ICC 200 300 400 500 I End othermic Reacfio ns T ............... . Exotherm'ic Reactions FIGURE 1 TYPICAL DIFFERENTIAL THERHAL ANALYSIS CURVE 3 Gruver I found that heat effects could be caused by the liberation of absorbed water, water of hydration, or water of crystallization; by the dissociation of a radical such as carbonate, sulfate, nitrate, etc.; by crystallographic inversions; or by chemical reactions such as oxidations. nuch study has been made of the components of the equipment needed for the procedure. The equipment includes a specimen holder, thermocouples, a furnace, a controlled power supply and a recording mechanism. Pas: and Warner 5 pointed out two important features of a specimen holder; (1) the sample holder should be large enough to hold enough sample to give thermal activity but small enough to prevent the occurrence of temperature gradients in the sample; (2) the walls of the holder should be this; enough to act as a heat stabilizer but thin enough so as not to absorb all the heat of the reaction. The sample holder should be a material which will not react with the sample to be studied. 1 Le Chatelier and others 4'13 used platinum crucibles as sample holders. More recently nickel or ceramic holders 6 have been used. Norton suggested the use of a heavy nichel bloch to eliminate thermal gradients around the sample. Gruver 7 claimed that thin walled platinum holders had the advantages of possessing a high conductivity of heat which allowed the sample to quickly reach the furnace temperature and also of possessing a lower heat capacity which would not tend to decrease the intensity and sharpness of some thermal reactions. Herold and Flange 8 combined the ceramic block with Special platinum-platinum rhodium cups which served both as the sample holder and as the differential thermocouple. The botton half of the cups and the thermocouple leads from the bottom were platinum (90;) rhodium (10;) alloy and the tops of the cups and the connecting wire were platinur. This construction had the advantage of allowing the thermo- couple to be cleaned easily when certain samples shranx and sintered into hard residues but had the disadvantage of not having the thermocouple in contact wi h the sample after it began to shrink. After comparing the preperties of nickel and alumina sample holders, We b 9 reported that although a nichel bloc: Cave sharper peahs than an alumina bloc; the peaLs were less intense and began at higher temperatures. The porous nature of the alumina bloc; allowed gaseous products to diffuse away more rapidly and thus allowed the reaction to be com- pleted quickly. When a silica liner was uSvd in the alumina bloc: to prevent corrosion, it also cancelled this increase in reaction rate. Gordon and Campbell 10 used a steel block with the sample holes lined with borosilicate tubes. Lerr and Xulp 11 have described a holder of chrome nickel-steel in which six different samples could be analyzed in one run. \j‘l Thermocouples should be made from material which will withstand corrosion by the sample. The wire should be small enough so as not to conduct heat to or from the sample rapidly but yet large enough to have physical strength so it could be easily cleaned without damage. Wire ranging from 24 to 28 B and 3 gauge has proven to be satisfactory. Several thermocouple circuits which have been used have been reported by Grim 12 (Figure 2, page 6). Circuit no. 1 has become the most popular because of the better reproduci- bility of results. A slight variation of this (circuit no. 3) will be used in this study so that all thermocouple Junctid will be equal distance from the heating zone. 5 grounded the thermocouple with a lOO Pea: and Warner m.f.d. electrolytic 50 volt D.C. condenser to filter out any A.C. pickup. Early investigators used platinum-platinum (90$) rhodium (lOfl) thermocouples and these alloy combinations are still being widely used. Chromel-alumel thermocouples have been used satisfactorily with some materials up to 100000. In lower temperature work, iron-constantan thermocouples are used to tahe advantage of the larger e.m.f. characteristic of these thermocouple elements. hraceh 13 used a gold- platinum versus platinum-rhodium thermocouple which produced about fifty microvolts per degree centigrade at high tem- peratures. No. I No. 2 No. 3 a — Sample Temperature —» lne rt Inert Material Material anmple g Sample avg 0 a V b a Sampte Temperature b b T Ditte rent iol Temperature Inert Material l I: ’I “-41, FIGURE 2 vaiOCOU‘BLE CIRCUITS. NUIIBIZR 1 AND HUI-3313?. 2 CIchéTS GIVEN BY AND NUIEBER 3 USED IN THIS STU The furnace must be designed to give a uniform heating zone and have a heating capacity large enough to allow the specimen to be heated at a controlled uniform rate. 5 1-. rash and warner observed that heating elements of nichrome wireunne satisfactory up to 100000, of Kanthal wire up to lBOOOC, and of platinum-rhodium wire up to 150000. Several types of furnaces have been reported. Grin 14 used a horizontal furnace mounted on rollers so that it could be easily rolled over the sample holder. This per- mitted easy access to the sample and reproducible positioning of the sample in the furnace. Gruver 7 has contended that the horizontal furnace reduced thermal currents and thermal gradients in the heating zone. Kraceh 13 used a vertical furnace and lowered the sample and thermocouples through the tep. Baffles were placed in the furnace tube to reduce 8 convection currents. Herold and Planje used a vertical furnace and inserted the sample holder through the bottom. 2nsasss_issnszainzs_flsnizal The control must be tailored for each furnace to give the desired heating rate. Several "homemade“ or commercial automatic program controllers have been made which give reproducible uniform heating rates. Autotransformers driven slowly by motors through gear reducers have been used to continuously increase the voltage. Controls which give heating rates from 6 to 20 degrees per minute have been used but a rate of about 12 degrees per minute is most CL) common. Any sudden change in the heating rates will be reflected in the differential thermocouple potential. Norton 6 has shown that a heating rate which is too slow produces broader peaks which occur at lower temperatures. The heating rate has to be fast enough to cause sharp re- actions but slow enough to prevent the overlapping of the differential thermal peaks. ‘7'“ ,w, an “6‘71"; 1’13 measured at regular intervals the Early workers e.m.f. of the differential thermocouples by a potentiometer and plotted thenadata against time or furnace temperature. The first continuous recorders were photographic re- corders in which a beam of light reflected from the mirror of a galvanometer was focused on photographic paper moving slowly on a drum.. This method required elaborate prepara- tions such as a dark box, a dark room and deve10ping equip- ment and had the disadvantage of not allowing the investigator to watch the recording as it was made. 11 Gruver 7, Kerr and Kulp , and Gordon and Campbell 10 have described automatic recording devices which were satis- factory. Grim 12 was of the opinion that a recording device should give about a two inch deflection for a one tenth millivolt potential. He also suggested placing variable resistances in the thermocouple circuit to vary the sensi- tivity of the recorder when necessary. ‘e The at- ficrystal inversion of quartz which occurs at 15’16 to establish a 575°C has been used by Berhelheimer known temperature in calibrating the sample temperature. Faustl7 described a method of using quartz as an internal temperature standard in a sample. Prasad and Patel18 in their study of manganese dioxide calibrated the sample temperature thermocouple with synthetic pyrolusite which gave peaks at 620° and 990°C. Barshadl9 has made an excellent study of calibrating reactions and has reported fourteen compounds which give sharp peaxs of crystal inversion or melting points which cover temperature ranges from 32°C to 961°C. dards Any material, which will not undergo any thermal re— action over the temperature range being studied, can be used as the standard reference material around one junction of the differential thermocouple. If there is a large dif- ference in the heat capacities of the standard reference material and the sample, the base line, which is the plot of the readings of the differential thermocouple potential when no reaction is taking place, will drift away from zero potential due to the uneven heating of the differential thermocouple junction. Calcined aluminum oxide (A1203) has been used most commonly as the reference mater1a1.11912117920 finalissiiana Q: Differential Ingres] assli°is Speil, Berkelheimer, Pash and Davies21 deve10ped a 10 method which could be used quantitatively (1 5;) to determine the mineral constitution of mixtures whose prime constituhnts could be obtained and analyzed in- dividually to establish standard peaks. The method could be used seniquantitatively (I 10%) or qualita- tively to study the mineral constitution of natural clay or other aluminosilicate formations. Berhelheimerl6 reported that the endothermic peak of the ar-fi inversion of quartz can be detected in a sample containing as little as fl quartz. The area under the peak was reproducible to a maximum variation of 5% in- dependent of the size of the quartz pieces. More than 155 calcium carbonate present in a sample when the quartz was less than 29p in size resulted in loss of quartz by chemical reaction. Barshad19 described a procedure to measure the heats of reaction of compounds by comparing the area under the thermal peaks with peak areas which resulted from reactions of compounds having known heats of reaction. OXALATES W In a study of the precipitation or the oxalates of neo- dymium. preseodymium, yttrium and certain other rare earths, Baxter and Griffin22 reported that in neutral or nearly neu- tral solutions these oxalatcs had high tendencies to carry down ammonium oxalate which was being used as the precipi- tating agent. They suggested adding a strong acid and decreasing the ammonium oxalate concentration as a means of preventing this phenomenon. 11 Later Baxter and Daudt23 found only a trace of ammonium oxalate carried down when the solution was acidified with two equivalents of nitric acid. Sodium oxalate was carried down only by canarius oxalate in neutral solution. Yttrium occluded alkaline oxalates to almost the same degree in acid or neutral solutions. Cxalic acid was not carried down when used as the precipitating agent. Wirth24 reported that the precipitation of the rare earths with oxalic acid was best in a 5.5 M sulfuric acid solution and that excess oxalic acid greatly decreased the solubility of the oxalate. wyliezfi reported two hydrates (hexa- and di-) were formed when lanthanum nitrate solution was precipitated with 103 oxalic acid solution and digested. The hexahydrate was converted to the dihydrate at 180°C. Cerium (Ill) 01d neodymium formed only interstitial hydrates while yttrium formed a hexahydrate when heated for seven hours at 98°C in water. Homogeneous precipitatione of thorium and the rare earths with methyl oxalate were studied by Hillard and Gordon26. Later Gordon, Brandt, Quill and 5alutshy27 used the hydrolysis of methyl oxalate to separate lanthanum and cerium (III) and also lanthanum and preseodymium. sendlandt28 formed the oxalates of lanthanum, cerium (III), praseodymium, neodymium, samarium, eurOpium, gadolinium and holmium by homogeneous precipitation w th methyl oxalate and by drying with 955 ethyl alcohol and reported the decahydrate in all cases. Yttrium formed a nine hydrate while erbium formed a hexahydrate. Lower hydrates were formed when these were heated as indicated in Table I, page 12. II. III. V. VI. VII. ~002 2200-63300 " or ' (C O )- J I‘D-173°C " f) - M 3.5253 Ala—73,J .- ‘N Y '\ f 'A f A“. CO L"r~ u out 7010; "‘ Pris?“ LEM 959... gossm £392.... gottm 22.0 32.6 :20 .3230 Begum 03:32:85... 39.33. 2222.50... 8953.... 0.29.2.8: 3.2.2.26 Siege“; .ozococgo oEEom 3.22.5.0 Tn . . 4% H a J )T—i “ J 1 0333252... 22939:... 2:63.. . W «33383:... .2280 32:3. u M w m m u 5825.3 .n. m m < m a £8.25 / I r. M «>20 305:“. 3.25:4 .m W 35 0095:“. , _ auo I. 5082 5 ml H »4 “ .o.< pl. IT“? .o.< «2-338 33> r... 53.5 asozm II L1 23> .5... Fl .o.< £22322! 3:225 8 5....45 33> r... 52.002 \JJ 0) 5. The connecting nut of the flared joint of the outlet water was unscrewed and the furnace and cooling jacket were lifted off the guide rods. The furnace was left standing upside down. 6. The crucible lid was removed to eXpose the platinum lined sample holes. 7. A residue bottle (Figure VIII) and a gentle vacuum were used to remove the sample residue from the sample hole. —’ To Vacuum Cotton Plug 50 ml. Erlenmeyer Flask Residue FIG n3 VIII VACUUX BCTTLE USED TO R23 VB SAQILE RESIDUES FROH THE S .I.I;: HO I“. If compounds of a different element Jere to be used next and it was important to prevent contamination, the next five steps were included in the procedure. 8. The platinum ring lining in the center section was removed with small tweezers. The hole was cleaned again with the vacuum residue bottle. 9. The differential thermocouple was exposed by removing the center section carefully. 39 10. The differential thernccoujle was loosened by streichtcrirr the leads to rive more length or was uIcco.ncc cd from the leads. The the :r ocouple was ::ved to one side to allow the platinum cup to be renovcd for cleaning the hole with the vacuum bottle. 1. The platinuzn liim are were cleans with 10 N nitric acid, rinsed with distilled water and acetone, and air dried. 12. The crucible block was reasseabled by reversing the preceding et 3s, taxing care to center the junctions of the differential thermocouples in the sample holes, to replace tle insulators to pre~ vent shorting between the ther rccc uple and the la inun liners, and tor repach the £120 I I"; 3 around the sample te pare ture thorn :oceu le and the reference thcrnccouplc. 13. When thv se.nnle crucible was satisfactorily cleaned, the sample to be studle» was packed in the platinua lined can pl e hole usifl.s a small funnel made from an 'ndex card and a glass rod as a tanner. A dierr cnt funnel was used for compounds of each element. 14. The crucible lid was replaced. Then the urnace gWIiIGs were sli13ye " ever the guide rods and the furnace 1., we “ed over the cr‘; c1. :16‘ 13. Ruboer washers cut from one-fourth inch ruthr hose were placed in the flared cepper tube 30 Ints to prev ent lee s and the connecting nuts were tightened by hand. 40 16. The cooling water control valve was Opened slowly until a flow rate of approxinately one quart per minute was obtained. This rate was estimated by the length of the water Jet from the one-fourth inch drain hose. A Jet of approximately three-fourth inch from a hose horizonal on the bottom of the sink was indicative of this flow rate. Any leans in the cooling system were corrected at this time. 17. The furnace power leads were connected to the binding posts on the cooling Jacket and were nods se- cure with the brass nuts. 18. The winged screws which held the drive gear meshed with the gear on the powerstat were loosened to allow the powerstat rotor to he set on zero voltage. 19. Hhile the tenplate rider was held away from the temperature program template'in the-flheelco Chronotrol, the program template uni nicr switch trip arm were ro- tated cloczwise to reset the tenclate rider below 20°C using the ten degree per minute template. 20. The master power switch was turned on. 21. The chart switches were turned on and the charts were allowed to advance until the recording pens were on the same relative time lines of the charts. This aided positioning the charts when the charts were overlaid to determine the temperature of the differential curves. Al 22. The Chronotrol template switch was turned on and the template was allowed to advance to 20°C. *3. The powerstat was reset by turning the drive DJ gear so that the voltmeter read 54-55 volts and was secured again with the winged screws. This voltage setting was always approached from the low voltag» side to remove slack in the gears and to insure re- producibility. 24. The knife switch in the Bodine motor circuit was cheched to be sure it was closed. 25. The furnace power switch, the chart switches, and the template switch were turned on simultaneously to begin the analysis. When the microswitch was positioned behind the hole in the 15 degrees template, the sample would heat to approximately lOSCOC?MHIthG instrument would turn completely off auto- matically. About two and one-half hours were allowed for the furnace to cool. The furnace was cooled below 50°C be~ fore the next analysis. Preparation of Compounds Lanthanum oxide was obtained from the Lindsay Light and Chemical Company, Inc. and was used as euch without any further purification. Heodymiun was obtained from Dr. Laurence L. Quill as the "GS-9" cut from a double magnesium nitrate fractional crystallization series. The use ymium and 42 means siun were separs.ted by the £0 lowing procedure. Fifty mil ilite ers of the consent atcd neodymium no.5- nes Mllm :uitrate solution was diluted to four liters and heated almost to boilin3. Dilute xalic acid which was rce1acrl 12y nixin31 lie epsrts lie Mt lled water with one part of a ‘P‘Vr ted solution of oxalic a;id was added crxnuioe uztil the precicitat on was cs::lete. The precipitate was circsted for several hours, cooled in washed ritL dis.ill ed water. The nrecipitate was dissolved in hot concentra+ ed nitric acid, diluted to four litere,and precipitated ng n tit: dilute oxalic a :_C. 3_.e green ed: 1:; cycle was reyeaten and the product of tIJG third preClr'iathn was used in.the preparation of th e compo inds in this study. The yttrium oxide used as orcwc‘e\ by R. H. Jaquith E ,7; {.4 gr. mas rsportcd to have an s,omic we ht of approximately \0 ‘3 Samarium was obta ned from Dr. Laurence L. Quill as a camarium perchlorate solution. Tlie camarium was rrecipit mt d from a hot dilute solution with dilute oxalic acid. After the precipitate vss digested for several hours on a hot plate -nd cooled in an ice bath, the precipitate was filtered from the solution on No. 50 filter paper and used in the preparation of the compounds in this study. #3 Fresco dynium oxalate decahydrate prepared by M. L. Salutshy was used in the study of praseodymium. Two thorium-free cerium magnesium nitrate fractional crystallize tion cuts ("Ce 25" and "Ce 2?") were combined. The pxwoc edw e of diluting the solution, of precipitating with dilute oxalic acid, of digesting the precipitate on a hot plate several hours, of cooling in an ice bath, of filterine on No. 50 filter paper, and of J‘k.) dissolving the recipitate in hot concentrated nitric {—4 acid was repeated three times. The oxalate precipitate from this was used in this study. The carbonates of lanthanum, need 'sium, yttrium, sasarius, and preseo dymium were prepared by homogeneous precipitation resulting from the decomposition of the trichloroacetate ion. About fifteen grams of each oxide were dissolved in a slight excess of 25$ tri- chloroccetic acid solution. Any solids which would not dissolve in four hours were filtered from solution. The solution we 5 trc.1orrc° to a two liter beaker, diluted to 500 ml., and heated on a hot plate for Six or eight hours to decompose the excess acid and the trichloro~ Wcdiffract10n angles. The Cs \ n ”V" diffraction angles were converted to d values (in- terplanar distances in angstroms) from the "Tables for Conversion of X—ray hiffraction angles to inter- planar Spacings”50. The "d" values of tLa oxices were compared with values listed in "Index to the X-ray _ -- -, r: {'OWC’E‘I’ pitta. L‘ 118 ”J1 o '3.-'- . Av’ . - “ “ ' fixW‘W‘ x‘ " A!“ 7"" -.‘."v mi: £45113." 2..“ LI _. in.“ J. I'LL! Aux) .LL JULJ. A.) Kany hydrates of the carbonates of the rare earth ele- ments have been reported. Their variable compositions have depenisd on the conditions of preparation. Fresh lanthanum carbonate which had been rinsed with acetone and dried on the filter by pulling air through for one hour analyzed to he 54.15§ lanthanum oxide. Upon dif- ferential thermal analysis, a curve lies ETA Ourve B0. 1‘ was obtained. After this material was air dried for three days, the curve was lihe ETA Curve ho. 2 and analyzed to be 72.12% lanthanum oxide. The change from 5#.15% to 72.12fl lanthanum oxide due to air drying indicated that the freshly prepared material may have occluded considerable wa- ter or was an unstable hydrate containing bound molecules of water. Constant relative humidity chambers were prepared in desiccators in an attempt to establish the presence of in- termediate hydrates. Saturated solutions of petassium car- bonate dihydrate (-3; relative humidity), calcium nitrate tetrahydrate (5 E relative humidity), magnesium acetate tetrahydrate (65; relative hunidity), ammonium sulfate (3 s relative humidity) and zinc sulfate heptahydrate (90; relative humidity) with excess solid52 were prepared in the separate desiccators. Approxinately five grams of free} damp lanthanum carbonate and five grams of anhydrous lanthanum carbonate (air driedcnm:year) were placed in each desiccator 57 and given two to three weeks to reach equilibrium. Differential thermal analysis curves of these samples were very similar, and DTA curve No. 2 is typical. Peaks indicative of hydration were not observed. Average analyses were about 72.2% lanthanum oxide. The anhydrous lanthanum carbonate began to decompose about 420~430°c. The peak indicating an endothermic reaction reached a maximum around 510°C (DTA Curves Nos. 1, 2, and 4). A second endothermic peak began slowly around 690°C and.increased rapidly around 780°C. A maximum was reached at approximately 880°C. The de- composition to the oxide was completed by 910°C. When lanthanum carbonate was heated to 540°C for eight hours, the resulting material gave a differential thermal analysis curve like DTA Curve No. 3 and analyzed to be 88.38% lanthanum oxide. Since lanthanum dioxy- carbonate (La203‘002) contains 88.18% lanthanum oxide, the material was postulated to be the dioxycarbonate. Lanthanum carbonate was prepared and stored in the _ mesenoe of water vapor at room temperature for two years to allow any hydrate to reach a stable equili- brium. A second sample was allowed to air dry for about one year. The room relative humidity during the final month of air drying and during all the eXperiments was between fifteen and twenty per cent. The differential thermal analysis curve (DTA Curve No. 4 A) for the material kept in the presence of water 58 vapor for two years showed a group of stall peaks in the temperature range of 50° to 22000. Analysis of this material by igniting to the oxide indicated 72.68% lanthanum oxide present. Theoretically there is 71.1% lanthanum oxide in anhydrous lanthanum carbonate. The material air dried for one year (DTA Curve No. 4 B) analyzed to be 72.75% lanthanum oxide. er“‘ °rfiw a A satisfactory cerium carbonate sample could not be prepared for study (See page 44, Preparation of Com- pounds). as A small sample of praseodymium carbonate air dried for twenty-four hours gave a differential thermal analysis curve like DTA Curve No. 5 which did not in- dicate any water of hydration. The carbonate began to decompose at 430°C as shown by the endothermic peak and reached a maximum at 495°C. Two small endothermic peaks followed with maxima at 630° and 700°C. The decomposition was completed to the oxide (Pr ) by 730°C. Between 960-99000, the oxide 6011 undergoes an endothermic change which is reversible. This peah is probably due to an oxide crystal inversion. T I’r‘ . C b The effect of air drying a sample of neodymium car- bonate is shown in DTA Curve No. 6. The damp carbonate analyzed to have an apparent molecular weight of 591 (molecular weight of Nd2(C03)3o7h20 = 592). After air S9 drying for eighteen hours, the material analyzed to be 73.13% neodymium oxide. Anhydrous neodymium carbonate contains 71.81% neodymium oxide. Neodymium carbonate began to decompose about 420°C and reached a maximum in the endothermic peak around 500°C. A second endothermic pea: began slowly around 655°C and reached a maximum at 800°C.- The reaction was complete by 835°C. when neodymium carbonate was heated to 510°C for twelve hours (DTA Curve ho.7), the material analysed to be 86.57% neodymium oxide. Neodymium dioxycarbonate (Ed203°002) contains ‘3.44§ neodymium oxide. EEKEHQJELJBflflEnEflEi samarium carbonate when spread thinly and air dried for only one and one-half hours showed no hydrate present When subjected to differential thermal analysis DTA Curve No.8). The material analyzed to be 72.91% samarium oxide. Anhy~ drous samarium carbonate contains 72.56% sasarium oxide. Around 410°C the anhydrous carbonate began to decompose endothermally. A second endothermic peak began slowly around 610°C and increased rapidly at 630°C to reach a maximum at 770°C. The conversion to the oxide was com- plated by 810°C. W The source of yttrium was an oxide sample prepared by Jaquith47 in a study of the separation of yttrium from rare earth elements. It was determined that the average atomic weight of the metal ion was about\ 92. The atomic weight of pure yttrium is 88.92. 60 The average atomic weight of this sample will be used in the theoretical calculations in this study. A damp Bangle of yttrium carbonate produced a differential thermal analysis curve similar to DTA Curve no.9 and indicated a possible stable hydrate. After air drying the carbonate for forty-eight hours, the material was found to contain 58.46; yttrium oxide. The dihydrate contains 57.993 yttrium oxide. when the hydrate was heated for thirteen hours at 96°C, a material (01A Curve No.10) contained 63.59% yttrium oxide. The anhydrous carbonate contains 63.74% yttrium oxide. The anhydrous carbonate (sin Curve ho.10) began to decompose endothermally slowly at 100°C and reached a maximum at 200°C. At 580°C a second endothermic peas began and reached a maximum around 620°C. A third posh Which reached a maximum at 670°C began be- fore the second peak was completed. All the reactions were completed by 740°C. A compound (DIA.Curve No. 11) containing 72.85% yttrium oxide resulted when the hydrate was heated to EGO-70°C for eighteen hours. Yttrium oxydicarbonate (1203-2002) contains 72.513 yttrium oxide. A. Hydrates Lanthanum oxalate which had been precipitated from a nitrate solution with dilute oxalic acid was found to 61 form two hydrates Which were stable to air drying and gave two different X-ray powder diffraction patterns. A decahydrate (ETA Curve No. l?) (Ehoto 8) which gave no intermediate hydrate when decomposed was formed when the material was first precipitated. The hexanydrate (DEA Curve No. 1.5 e) (i‘hoto so. 9) was formed by adding dilute oxalic acid to'a nitrate ao~ lution or by hydrolyzing dimethyl oxalate in a nitrate solution and digesting the precipitate on a steam bath for six hours or more. when analyzed by differential ‘ thermal analysis,_the hexahydrate indicated the forma« tion of an intermediate hydrate. After heating a sample .of the hexahydrate to 96°C for twelve hours. the residue analyzed to be a tetrahydrate (Din Curve No. 1% A). The anhydrous oxalate eculd be formed by heating the hex hydrate to 200°C for seven hours (DTA Curve 1%. s). A study of the relative stabilities of the hydrate in water during digestion was made. Approximately 2 gm. of the decahydrate was placed in a 250 ml. wide mouth glass Jar. To this was added 5-10 mg. of the hexahydrate and 100 ml. of distilled water. The Jar was then covered with a ground glass plate to inhibit evaporation of water. A second Jar was prepared similarly with.the hexahydrate. Both hydrates were then digested for seven days at 82-85OC. X-ray powder diffraction patterns of both samples were identical and the compounds analyzed to be the hexahydrate. 62 Azeotropic dehydration by refluxing the hydrates with toluene for twentyéfour hours did not give definite hydrates. The dried material from the decahydrate analyzed to be 57.49fi lanthanum oxide (Approximately 3 1.35 hydrate) while that from the hexahydrate analyzed to be S4.&4fl lanthanum oxide (approximately 3.15 hydrate). Slow dehydration of the hexahydrated oxalate over anhydrous magnesium perchlorate for three months re« sulted in a c mpound which approximated a 4.5 hydrate. B. Oxalate Decomposition The decomposition of the oxalate (DTA Curve No. 14 B) began at 375-3500. no two sharp peaks at approximately 395°C and 405°C indicated exothernic reactions. During the heating process, the white oxalate changed to a pale yellow brown color. it about 500°C a less intense exothermic reaction began and continued for about the next 125 degrees. The material became shiny black from carbon focmed during the reaction. A final peak beginning at 700°C and reaching a maximum at 790-800°C was noted to be endothermic. The reaction was complete to lanthanum oxide by 835°C. X-ray powder diffraction patterns were made of the anhydrous oxa ate, the pale yellow brown material. the shiny black material, and the white oxide which had been rem ved from the differential thereal analysis furnace,but only the oxide gave a diffraction pattern. All the other materials appeared to be amorphous. 63 Attempts were made to reproduce these compounds out- side or the differential thermal analysis furnace by heat- ing lanthanum oxalate at a lower temperaturethan indicated by the analysis curves; for a long period of time. After the hexahydrate of the oxalate had been heated to 335°C for ten hours (ore Curve so. 15) or 300°C for fifty— three hours (Dis durve No. 16), the resulting pale yellow brown compounds analyzed to be 78.66; and 78.53% lanthanum oxide, respectively. Lanthanum oxydicarbonate (La203'2002) contains 178.78% lanthanum oxide. The compound dissolved in 3 N hydrochloric acid with the evolution of carbon di- oxide. A solution of the compound in 10 N sulfuric acid did not decolor permanganate solution. The shiny black material, which could be formed when the oxalate was heated to 405°C for five hours, evolved carbon dioxide when dissolved in 3 N hydrochloric acid and left insoluble carbon. The carbon dioxide content of the material was determined by collecting the evolved carbon dioxide with ascarite (the gain in weight represented the amount of carbon dioxide). The lanthanum was determined ay precipitation with oxalic acid and titratins the preci~ pitated oxalate with a standardized permanganate solution (See page I». The ratio of’expgnguLngfianJ$Umedg .X 100 grams of lanthanum oxide was found to be 13.83. This ratio for lanthanum dioxy- carbonate (La203o002) is 13.51. e - a . Cerium oxalate (Bhoto 10) was precipitated by oxalic 64 acid from a nitrate solution and after air drying ana- lyzed to be 47.6?5 eerie oxide. The deeahydrate contains 47.53; eerie oxide. Differential thermal analysis of this material gave a curve similar to DTA Curve E0. 17 and in- dieated no intermediate hyarateo. Deh dration was slow. After three months over anhydrous _magncaium perdhlorate, the oxalate analyzed to be cerium oxalate 4.14 hydrate. After heating the hydrate at 96°C for fourteen hours, the material was not completely dehydrated (ETA Curve No. 18). The material contained 61.33% ceric oxide which corresponded to a 0.93 hydrate. The anhydrous oxalate began to decompose exothermally at 320°C. Uhen the decomposition was almost complete, a very sharp exothermic peak at about 355°C occurred, indi- cating the ox dation of the eeroue to eerie oxide. The reaction was complete by 39000. The sample studied was a decahydrate (Photo 11) which had been prepared and analyzed by SalutehyhB. The differen- tial thermal analysis curve indicated that no intermediate hydrates were formed (ETA Curve he. 19). The anhydrous oxalate was prepared by heating the hydrate to 105°C for twenty-four hours (ETA Curve ho. 20). The oxalate began decomposing about 375°C and gave two exothermic peaks a- round 330° and 390°C. A second strong exothermic peak began about 410°C and reached a maximum around 450°C. A final enaothermic pea: beginning at 500°C and reaching a 65 maximum aroung 600°C indicated the conversion of the material to a steel black oxide. At about 970°C, a reversible con- version of the oxide was noted. This change is probably due to an oxide crystal inversion. A. fiydrates Only the decahydrate (Photo 12) was found when neodymium Ilalete was precipitated from a nitrate solution with dilute oxalic acid or dimethyl oxalate. When this hydrate was i studied by differential thermal analysis, two endothermic. pealzs were formed as the water of hydration was driven off (DTA Curve No. 21). The first peak began about 120°C and ree.ched a an azimut.a around 190°C. The second peak had a maximum at about 255°C. When the decehedrete was heated at 96°C for thirty-six hours (DTA Curve Ho. 22), the com- pound analyzed to be 56. 23$ neod3mium oxide. The dlhydrate is 57.1; neodymium oxide. VIn the thermogravimetric curve of neod3 ium or alate determined by w endlendtza , a break in the curve corresponded approximately to the dihydrate. The complete dehydration of the oxalate was difficult. After heating the decahydrate to 170°C for fourteen hours or 200°C for ten and one-half hours, the materials analyzed to be 59.7W and 60. 3‘3 neod3nium oxide respectively. Anhydrous neodgrmium oxalate contains 6 .755 neodymium oxide. then tie decehyLErate was kept over anhydrous magnesium perchlorate for two weeks, it was only dehydrated to a compound corresponding to neodymium oxalate nine hydrate. 66 B. Anhydrous Oxalate Keodymium oxalate began to decompose at approximately 3850C and produced two exothermic peaks at about 390°C and #1000. The pale blue oxalate changed to a brownish blue with a tinge of yellow in this region. At about 51000 a large exothermic peak starts and shows double naxina about lO~15 degrees apart in the region of 585°C. The material turned black in this teZperature range. Finally an endothermic peak beginning between 625~ 64000 indicated the conversion of the material to the pale blue neodymium oxide by 750°C. To prepare these intermediate materials the hydrated , oxalate was heated to SASOC for four hours or 335°C for six hours to produce the brownish blue yellow product. By weight lose this material appeared to be neodymium oxydicarbonate (Hd203'ECCg); it gave a differential thermal analysis curve Lsimilar to DEA Curve No. 24. when the hydrate as heated to 300°C for fifty-three hours (DmA Curve he. 25), the product analyzed to be 80.61% neodymium oxide. Hecdysiuu oxydicarbonate con— tains 79.273 neodymium oxide. Since one of the possible structures for the par- tially decomposed oxalate includes a peroxide linkage —, ." O -. 1 '\ .4 ‘ and was reported by Gunther and Rehaag .‘a test for the peroxide group was made. To approximately a tenth gram of this material dissolved in two ml. of l H sulfuric acid was added two ml. of a dilute titanium 67 (IV) sulfate solution. An intense yellow color due to a titanium pe oxys ulfate complex is indicative of the pre- sence of the peroxide group. In the testing no color was noted hence it can be concluded that the material ’0. did not have a neroxid.e linkai2e. u. Samarium oxalate (Ehoto lg) precipitated from a perchlorate solution by dilute oxalic acid and digested over night on a h t plate analyzed to be 46.91% samarium oxide when air d.riei (‘ TR Curve Ho. 27). The decahydrate contains 46.33fi sanarium oxide. hhen the decahydrate was studied by differential thernsl analysis, the curve 1? diccted the formation of two intcruediate hydrates. first or. be an at 63°C was the first endothermic poakr cachel a maximum at l 0°C. The second and third endothermic peaks were at 13000 and ECO-70°C. When the hgdrete lLGS heated to I'TOC for ten hours, all of the water of the first stable hydrate had not been removed (STA Curve Ho. 28}. the material analyzed to contain 50.85fi sanariun. ihe hexahydrate contains 51.3; sanariun oxide. The yroduct, after heating the hydrate at 96°C for twenty-four hours, analyse;1 to be 54.33; samarium oxide. The tetrehydrste contains 54.785 steerium oxide. 30 attcupt was made to prepare the anhydrous oxalate, ~t it prooably couli have been pre pared by heating the hydrate to 150-16090 for twenty- -four hours. 68 At about 330°C, the anhydrous oxalate began to decom- pose ex thereelly and proiucefl two sharp peaks at 390° and 43003. A second exothermic peak began slowly around 520-3000 and reached two maxima about ten to fifteen degrees apart arouid 590°C. xermic pee: was almost com- pletely absent in this decomposition. dhen the oxalate "fl .,. ‘,.. x. . 'r . ,. .0 . ....¢4. .... . .._, l -.1. etc neatel to JEOOt let i ioeen heels, the resulting "'5‘ material (elk Curve Lo. 30) .noljzed to be 79.93% eemerium oxide. tamarium oxydicarbonete (32203-2C05) The average atomic weight of 92 for yttrium Which was deterzined for this sample of yttrium was used for tho 'heoreticel calculations in this study. I triun oxalate precipitated from a nitrate solu- tion by dilute oxalic acid woe air dried for 18 hours and analyzed to be 37.09; ”ttriue oxide. The decehydrate of yttrium oxalate contains 36.953 yttrium oxide. When this 1ydrete was studlei by differential thermal analysis, the urve (DEA Curve No. El) gave evidence of an intermediate hydrate. Eben the hydrate was heated to 96°C for eighteen hours, the material obtained Wee ene_yzed to be 46.#B% yttrium ox fie and produced a curve similar to DTA Curve no. 2d. A yttrium oxalate trihyioete would contain ‘, yttrium oxide. The enhyiroue oxalate was preparei by heating the 69 hydrate to EGO-70° for twenty hours (DTA Curve No. 33). The anhydrous oxalate began to decompose around 395°C and produced two exothermic peaks at 410°C and 425°C. A eecond exothermic peek began around 530°C and reached a maximum around 630~50°C. The reaction was completed to the oxide between 660~80°C. COHCLUSIONS Differential thermal analysis proved to be a useful tool in understanding the decomposition of the carbonates and oxalates of the rare earths. The presence of inter» mediate hydrates and compounds along with crystal inversions 'wore readily detected. Differential thermal analysis pre- bsbly gives the best indication of changes which take place in a bulk of material during decomposition. This type information would be most useful for large scale decompositions as used in commercial productions. The most serious drawback of the technique is the difficulty of analyzing the intermediate products Which formed during the decomposition. The rather small sample required in the analysis does not yield a quantity of pro- duct largc enough for direct analysis for the rare earth elements. The hydrates which have been reported for the carbonates of the rare earth elements appear to have been only varying amounts of water absorbed in the carbonate materials and did not constitute any definite hydrates. Even under conditions of high relative humidity no stable hydrate could be found. Yttrium carbonate may have formed a stable dihydrate although the impurity of the yttrium sample may have resulted in the calculated 1.8# hydrate. All the carbonates studied began to decompose around 420-3000 except yttrium carbonate. Ambroshii and Hikola- evshayas3 also found that the temperatures of the first 71 stage of decomposition of lanthanum. cerium, praseodymiumy and neodymium carbonates were close but they reported a range of 460-51300 for the first step. The carbonates of lanthanum, neodymium, and samarium decomposed to the dioxycarbonate compounds (R203o002) and then to the oxide. Final decomposition of these compounds to the oxide occurred at increasingly lower temperatures as the oxides became more acidic. Fraseodymium and yttrium carbonates behaved different~ ly. Preseodymium carbonate began to decompose at 430°C, perhaps to the oxydicarbonate which in turn decomposed in two steps to the oxide. ‘The tendency of praceodymium to form higher oxides may be responsible for the final decomposition.occurrin3 at a lower temperature than would be expected When compared with the behavior of lanthanum or neodymium carbonates. A reversible reaction occurring at 960°C was probably due to a crystalline inversion of the oxide (Pr5011). This inversion has not been reported in the literature. Yttrium carbonate began to decompose at 100°C and converted to yttrium oxydicarbonate. This low temperature of decomposition may be a result of instability in the crystal lattice due to the small size of the yttrium ion. No reason can be given for the appearance and stability or the oxydicarbcnate for yttrium. Decahydrates for lanthanum. cerium, praseodymium. neodymium, samarium, and yttrium oxalates were found. 'F‘Q ‘ h A hexehydrete was also found for lanthanum oxalate which would decompose to e tetrehydrate. The deoehydrate of lanthanum and preseodymiun had no intermediate hydrates. neodymium oxalate decahydrete formed a dihydrete es an intermediate. The decehydrete of semerium oxalate de- composed to a hexehydrste end then to e tetrehydrete. A trihydrato was the intermediate hydrate formed when yttrium oxalate decehydrete was dehydrated. When the oxalates began to decompose, two exothermic peeks about ten degrees spert were formed in the differential thermal analysis curve. Gilpin end EcCrone54 studied the crystal structure of lanthanum oxalate decahydreie and found that as the water of hydration was removed from the crystal, the physical aopesrence remained unchanged but the material became amorphous to X-ray diffraction. This open skeleton expanded lattice which accommodated the water of hydration becomes unstable around 385°C; collapse of the lattice could liberate energy which would result in the first exothermic peak.‘ When the crystal bonds uhich stabilized the oxalate ion were broken, the oxalate ion ‘ decomposed into carbon dioxide and carbon monoxide and left the compound, fi203'2002. Upon further heating, the oxydicerbonste compound de~ composed exothermelly to the dioxycarbonate and then finally endothermally to the oxide. The oxide or presea- dynium again showed the reversible reaction at 960~70°C. 73 Carbon Which.wss formed during the decomposition resulted from the disprOportionstion or carbon monoxide into carbon dioxide and carbon. The oxidation of this carbon in the sample tended to reduce the size of the last endothermic peak associated with the liberation of the last molecule of csrbamdioxide. It could not be determined definitely in this study whether the formation of the oxydicerbonste and the dioxycsrbonate were a result or partial decomposition of the oxalate ions to the carbonate or were the products of a reaction between an oxide and carbon dioxide formed in the decomposition. The materials were amorphous in X~rsy diffraction studies. As the acidity of the rare carth.oxidc increased, there was a marked decrease in the size of the last endothermic peak which indicated the decomposition of the dioxycarbonste compound. In the study by Wendlmndte8 in Which the oxalate -wss decompdsed as a very thin layer which.sllowed very little interreaction of the decomposition products, the dioxycsrbonete was reported for only lanthanum. the most basic of the rare earths. Also, since this peak was present in the carbonate compounds, one might conclude that in the decomposition of the oxalate. the dioxycarbonate compound was formed by the reaction or the oxide and carbon dioxide. 'i.)-"“Y ‘ 74 X~ray diffraction patterns for all of the carbonates studied (see Appendix) show very close resemblances for the crystal structures or the carbonates. This behavior would be expected because of the similarity of the size and charge on the metal ion of the elements studied. A satisfactory diffraction pattern for yttrium carbonate was not obtained. Two definite patterns for the prepared deeds and hexahydrates of lanthanum oxalate were found. The "d" values for the decahydrate compared closely with the values of Gilpin and McCroneSA. The decahydrates of lanthanum, cerium, praseodymium, neodymium, and samarium are again very similar. The lines of the ery diffraction pattern of the. decal:rz‘rate of Tattriv. U. 2 oxalate were spread slightly further apart and are more intense. Samarium also formed two stable hydrates as shown by the xrray patterns. Vickery55 discussed work by Goldschmidt who stated the rare earth oxides usually form hexagonal. cubic and in some cases pseudotrigonal crystals. The Xeray diffrac- tion patterns or the lanthanum and neodymium oxides are identified as hexagonal. The patterns of prescodymium oxide (PrGCii) below 960°C and or cerium oxide are cubic. Yttrium oxide gave a pattern identified as a body centered cubeSI. The pattern for samarium oxide was not identified but may be the pseudotrigonal structure proposed by Goldschmidt. The ”d" values 75 for the oxides of lanthanum, neodymium, praaeo- dymium and yttrium compared well with the values of Reed56 and the "Index to X-ray Powder Data F119 "51 o 76 SUMKARY 1. An apparatus for observing the thermal changes during thermal decomposition of rare earth carbonates and oxalates was designed and constructed. 2. The decomposition process for rare earth car- -bonates consisted only of endothermic reactions and generally formed only a dioxycarbonate compound as an intermediate. 3. The decomposition processes for the rare earth oxalates were both endothermic and exothermic reactions and usually formed both an oxydicarbonate and a dioxy- carbonate intermediate. 4. The decomposition process for the cerium compounds differs somedhat from that for other rare earth element compounds due to the fact that the cerium (III) ion is easily oxidized to cerium (IV). In general, the cerium compounds were all converted to ceric oxide (0002) by 400°C. . 5. In general, the temperatures for the final stage of decomposition for the carbonates and oxalates decreased with decreasing basicity of the elements. 6. The dehydration of hydrated compounds could be followed, in most cases, by means of differential thermal analysis. 7. A reversible change in the crystal structure of praseodymium oxide (Prsoll) occurs at 960°C. This change has not previously been reported and must be studied further. 1. 2. 3. 5. 6. 7. 9. 10. ll. 12. 13. .14. 15. 16. 1?. 77 LI TLzATURE C I TED Le Chatelier, ‘52.. Bull. Soc. Franc. ;.:in.. 19,. 204 (1887) Roberts-Austin, I-“roc. Inst. Liech. Eng, 25,, (1899) Fenner, C. N., A33, J. Science l'.th Series. m, 331 (1913) ' " Gruver, R. 5.2., J. Am... Ceram. 500., 11. 96 (1950) Peels, J. A. and Warner, H. F.. Am... .Ceram. Soc. . 31111.. 33., 158 (1954) Horton, F. 1%., J. m... Ceram. 300., 22,, 54 (1939) G-ruver, B. 1-1.. J. Art). Ceram. £300.. 31, 323-28 (1948) Herold, P. G. and Planje, T. J.. J. Am. Ceram. Soc.. :1. 20 (1948) ' z-zebb, T. I... Nature, 129... 686 (1954) Gordan, s. and Campbell. 0., Anal. Cher... 21. 1102 (1955) Herr, P. F. and Iiulp, J. L., its». L~Iin., 33,, 387 (1948) Grim, R. 13., Ann. 1‘. x. Acad. sci... 51, 1031 (1951) lirace‘sz, F. C., Jour. Phys. Chem” 1'1. 1281 (1929) Grim, R. :3. and Rowland, R. I... J. A:r:-.: . Cer. Soc.. 21, 65 (1944) Berkelheimer. L. H., U. 3. Bureau of Mines, Rept. Invest. 316%(1944) Eerl-;elheimer, L. H.. U. S. Ia‘ureau of lanes, Rept. Kept. Invest 125.1,(1944) Faust, G. T.. A512“. . .1111... 31. 337 (1948) 18. 19. 20. 421. 22. 23. '2... 25. 26. 27. 28. 29. 30- 31. 32. 33. 78 Prasad, H. S. K. and Patel, C. C.. J. Indian Inst. Science, 36A, 23 (1954) Barshad. 1.. Ans . Min.. 11. 667 (1952) Chiang, K. and Smothers, W. J., J. Chem. Ed.._21. 308-9 (1952) Spoil. 8.. Berkelheimer. L. H.. rash. J. A. and Davies, 3., U. 5. Bureau of Mines Tech. Paper, 664 (1945) Baxter, G. P. and Griffin. 3. 0.. J. Ame . Chem. Soc.. as, 1684 (1907) Baxter, c. P. and Daudt, H. 31.. J. Amofi. Chem. Soc., 29. 563 (1908) Wiruh, F.. 2. users. Chem.. 16, 174 (1912) Wylie, A. W.. J. Chem. Soc.. 1687 (1947) Willard, H. H. and Gordon, L., Anal. Chem.. 29. 165 (1948) Gordon, 1... Brandt,R. A., Quill, L. L. and Salutshy, M. L.. Anal. Chem.. 23, 1811 (1951) Wendlandt, s. W.. Anal. Chem.."1Q, 53 (1958) Hobbs. D. E. and Noyes, s. A. Jr.. J. Amrr. Chem. Soc...&§, 2856 (1926) Herschkowitsch, M.. Z. anorg. allgem. Chem.. 115. 159 (1921) ‘éve’aa. J., Chem. Listy, u. 47. 81. 112 (1923) melee, E. and Villamil, C. D.. Analee soc. aspen. fie. quim., 23, 465‘94 (1925) Ugai. I. A., . lilhur Obshchei Kim” 25, 1315-21 (1954) 34- 35. 36. 37. 38. 39. 41. 42. 43. 79 D'Eye, R. W. :3. and Sellman, P. 0., "The Thermal Decomposition of Thorium Oxalate," Atomic Energy Research Establishment, Ministry of Supply, Harwell Berks (1953) Centola, 0., 1V Concr. intern. quim. pure. aplicada, enema, 1, 23098 (1934) Gflnther, P. L. and Rehaag, H., Ber., 113, 1771 (1938) Fonnoff, F. J.. "The Rare Earth metals and Their Compounds: A Study of Ceric Oxide", Ph. D. Thesis, Ohio State University (1939) Somiya, T. and Hirano, 8., J. Soc. Chem. 1nd., Japan 34, Suppl. binding No. 11, 459 (1931) Preiss, J. and Rainer, N., Z. anorg. allgem. Chem., 111, 287 (1923) Cleve, P. J.. Bull. Soc. Chim.[2], 52, 162, 359 (1885) Salutshy, M. L. and Quill, L. L., Anal. Chem., Ed. 1453 (1952) Praise, J. and Dussik. A., Z. anorg. allgem. Chem.,:;:1. 215 1:23) Salutshy, h. L., "The Rare Earth Elements and Their Compounds: The 2urification and Prepsrties of Praseodymium Oxide,” Ph. D. Thesis, Michigan State College (1950) gallander, L. 0., “A study of Sodium Lanthanum Carbonate," H. S. Thesis, Michigan State College (1954) 45. #6. 47. 48. 49. SO. 51. 53. 80 Spencer, J. F., "The Hotels of the Rare narths". Longmans, Green and 00., London (1919) Paul, H. M., Brinton, P. and James, C., J. Am. . Chem. £00., £5, 1445 (1921) Jaquith, R. H., "The Rare Earth Elements and Their Compounds: Homogeneous Precipitation Studies of the Yttrium berths," Eh. D. Thesis, Hichigan State University (1955) Ehgel, T., Jr., "The Observed Optical Proyerties of the carbonates and Acetylacetonates of Neo- dymium and Lanthanum." private communications. holthoff, I. m. and Sandell, E. 5., “Textbook of Quantitative Inorganic Analysis," 3rd Ed. the mac- nillan Company, New York (1952) p. 373. "Tables for Conversion of X-Ray Diffraction Angles to Interplanar Spacing,“ U. 3. Dept. of Commerce, National Bureau of Standards, U. 3. Government Printing Office, Washington, D. c. (1950) "Index to the X-Ray Powder Data File," American Society for Testing Materials, Ihiladelphia (1958) Inorganic Section. "Handbook of Chemistry and Physics: 37th hd., Chemical Rubber Publishing 60., Cleveland (1955) p. 2309. Ambrozhii, n. N. and Nikolaevshaya, M. 1., Nauch Ezhcgodnik as 1954 g Saratov, Referat. .Zhur.. Lhim, (1956) abstr. No. 50507 81 Gilpin, V. and McCrone, H. C., Anal. Chem..;afl, 225-6 (1952) Vichery, R. C., "Chemistry of the Lanthanons," Academic Press Inc., How York (1953) Reed, w. R., "Density and Structure of Lanthanum and Praeeodymium Oxides," Ph. D. Thesis, Michigan State College (1954) APPENDIX X-Ray Diffraction Patterns. -- In the contact prints of the diffraction patterns, dots were placed on the lines which were mea- sured and tabulated in the following tables. RI Relative Intensity 8 Strong m medium w ' weak vw very weak vvw very very weak Differential_Therma1 Analysis Curves. -- All peaks above the base line are endothermic and all peaks below the base line are exo~ thermic. A peak deflection of one inch indicates about a 0.2 millivolt differen- tial between the sample and the inert standard. 1.1" A.“ (60,), C0,. ((0,)! Pv. (“3): #2) NJ, (cos), > i). $M1(c03)3 La3(003)3 32.1. 2 6 m 1.610 m 2.100 m 2.425 a 2.665 m 3.030 m 3.390 m 3.345 vvw 4.365 vvw b.425 vvw 4.500 w $.655 vw' 4.790 w 4.92 VVW 5.2?5 vw 5.4Ao vw 5.690 vvw 5.810 (1 5.5003 $.2267 3.6671 3.3420 2.9fi72 2.6s30 7 7 . 293J93 2.0719 2.0fi51 2.0128 1.9493 1.89?4 1.8303 1.7338 1.6351 1.6168 1.5863 2 t 3 :i ‘3 it :t a £2. 3 6.239 1. ?65 111 Pr2(CO3)3 2.1. 2 6 m l 597 a 2.062 8 2.404 m 2.668 8 3.033 m 3.378 m 3.603 w 3.712 3 3.863 m 4.383 m 4.536 w 4.678 vw 4.813 w 4.927 vw 5.028 2.6511 2.4906 2.4199 2.3287 2.0637 1.9976 1.9402 1.8889 1.8131 8.1. 3 3 3: E3 =2 m a t 5] WW 2‘6 1.623 2.100 2.447 2.722 3.103 3.422 3.670 3.78 3.902 4.462 4.623 4.772 4.867 5.022 5.117 1.9600 1.9042 1.8692 1.8151 1.7836 Cez(003)3 3.1. 219 w 1.605 m 2.070 m 2.420 m 2.665 m 3.039 w 3.389 vw 3.58 vw 3.710 'm 3.854 w 4.365 vw 4.433 w 4.525 w 4.664 vw 4.805 w 4.935 vvw 4.995 iv d 5.5174 4.2872 3.6?45 3.3420 2.9337 2.6427 2.5000 2.4212 2.3339 2.0718 2.0416 2.0022 1.9457 1.8918 1.8450 1.8242 Ce, (ctoq)J - Io HLO ’0 o (Pv.[l.°v)3‘ "0 ° ), J\ 5‘1(¢x°v)aq”) vi La2(0204)3-10220 R010 2 0 d m 8.530 100394 a 13.30 5.5513 vvw 16.30 5.4333 vw 17.00 5.1212 w 7.60 5.0340 m 18.30 4.8437 vvw 23.30 3.7824 vvw 25.15 3.5384 mw 29.43 3.0303 vw 31.30 ‘2.0115 vw 34.20 2.6195 C83(C204)3010H20 2.1. 333333‘ 2 0 :‘l , (.0 a. 13.1; 17.15 K)! 17073 13.48 25.28 b1 r.) C O O) U 1» \fi «D'UI :- 1)! P.) 34.43 38.88 39.63 40.23 ?r2(C204)3-10H20 d 2.1. 2 0 d 0.297 m 8.57 10.307 6.5873 6 13.42 6.5927 5.1629 W 17.42 5.0864 4.9976 w 17.92 4.9456 4.7970 m 10.67 4.7484 vvw 20.53 4.3224 vvw 21.00 4.2267 VVW 21.91 4.0532 vvw 23.97 3.7092 3.5199_:w 25.42 3.5009 VVW 26.22 3.3958 vw 29.51 3.0243 vw 30.12 2.9644 vvw 31.42 2.8447 2.7919 vw 31.67 2.8228 2.7460 vw 32.8] 2.7224 2.677 vvw 33.68 2.6588 2.602 *W 34.50 2.5974 2.314 2.272 2.239 v11 Nd2(0204)3-1OHQO R.I. 2 €7- C“. m 8.61 10.261 vvw 2.65 6.9861 a 13.25 0.057} m 7.31 5.11(3 m 17.06 4.9021 8 15.35 4.2230 vvw 20.36 4.3331 vvw 20.26 4. vvw 21.75 4.0007 V V W 23.96 3.71.03 1'; 2'34. "‘6 J.63kjl VVW 25.3 3.5150 vvw 29.26 3.0410 vvw 33.71 2.6365 ’- -. ,_' .- vvw 34.13 2.007! Sm2(0204)3.10320 1.1. 2 2- a a 9.52 3.4003 vw 12. 2 7.0001 B 13.32 (.5521 .11 7037 5010:; m 17.32 4.2731 vvw 21.22 4.0513 VV‘J 21". 22 3.631115 3 25047 3-h353 vvw 26.3"; 3.375- vvw 27.22 3.273” vvw 30.37 2-94 3 vw 34.42 3.6033 vvw 33.52 2.5252 vw 33.72 2.3233 12(0204)3610H20 r“. r-o-a r... 8 0“! lOoC’;‘ vw 12.70 6.9642 3 13.57 {.51o9 w 18.90 4.6913 vvw 20.62 4.3244 vvw 21.07 4.2128 w 22.42. 3.9621 {.2 R) I: O C: [3 U3 0 «‘2 U1 viii La%(c204):¥6n2o 5m2(0204)364nzo *4 A 9 9 ’ R. . C3. m 11.33 7.7588 w 14.20 6.2317 3 13.64 (.4353 m 6.23 5.4549 3 15.51 5.3556 m 33.90 4.531} m 21.43 4.1333 w 20.84 4.2538 VVW 22.53 3.9344 vvw 29.75 3.004 m 23.33 3.3015 w 32.60 2.7444 R 27.03 3.2339 vvw 33.90 2.6420 V 27.53 3.2314 vvw 3?.2’ 2.4149 vw 31.43 2.3433 vvw 33.20 2.3539 vv 32.13 2.7792 vvw 42.60 2.1204 vvw 44.15 2.0496 m 46.95 1.9336 g.::ggga..... 0002 Ergoll ("1 ELI. 2 O (‘1 ELI. 2 C (1 3.4023 - vvw 29.25 3.5244 3.0583 8 28.64 3.1142 8 28.10 3.1728 2.9692 m 33.19 2.5909 s 32.? 2.7346 2.2756 a 47.58 1.9095 e 47.00 1.931? 1.9644 3 56.48 1.6279 5 55.72 1.6483 1.7503 W 59.14 1.5503 m 53.37 1.5796 1.6551 w 69.43 1.351? W 63.55 1.3677 1.6401 m 76.69 1.2415 m 75.74 1.2547 1.5296 mw 79.14 1.2091 m 78.10 1.2226 1.4835 m 83.43 1.1045 m (D *1 \4 0 1.1159 1.3961 m 95.35 1.0418 .2931 1.2600 1.2033 1.1338 x1 1511203 3.1. 2 0 m 27.05 m 29.90 e 30.85 m 40.65 m 47.50 m 53.53 vvw 55.45 m 47.05 m 57.75 vvw 61.95 V 64.15 w 63.60 w 74.25 vvw 75.85 131W 77-75 w 01.40 {‘3 K...) k)! U1 ‘1 «I I I . ' 9-. Q ’ \0 [D N {:4 6" KO 195 276 O H 2. 1.9125 1.7093 1.6556 1.6130 .5950 1.4966 1.4505 1.3663 1.2761 n79 loCSQQ 1.2273 1.1312 VVW VW VW W W 41.88 R 1.9363 ‘ 1.8131 1.7672 1.7149 1.5771 ' 1.5893 1.3407 2.054 2.629 2.909 3.590 3.979 4.349 5.909 5.009 4.839 5.314 5.614 5.759 5.899 6.034 6.179 6.314 6.454 6.584 d 4.3203 3.3370 3.0670 2.0T89 1.5420 1.8195 1.8793 1.7220 1.6369 1.5990 1.5645 1.5315 1.5002 1.4713 1.4427 1.4173 111 < 1 < < \ :92 - 2.813 .m 23> at... .303 35.4.6834 .<. _ w T tn :3... m -n.»oov~o._ .8. nor... .n 1SEES: 3:201 oxoom 38.003 5m 0» a: ..000_ :00 98...... - 39.3 .N / _ 0530: .N /\ 955 nice“: ._ \/ _ com o O 0 00b 00w 000 00¢ 00» CON [4| 00?- U 00 . :3: ~_- .06 «1.312,. M1313.302 N 253.. m. 1:5 .3. «:3... d Ito :4 32.23:. cm 9.... . 3.08.; .m 7\|‘ /\/\ _ 232.. cm - to .._< «:96 E: 4- n.» o Sui .0 P _ 000. 00m . 000 00x. 80 000 00¢ 00m 00m 00. coo .. 1 x 1 :1 :3: o. .263. 33.113? Ncommon» .: i /\ 2:31 a. Jam 3 to .131». .o. i /\/1\: 2:5 :68.» .m If I J .32....m. to .2 .1315 .m _ L _ 000. 000 000 00x. 000 00m. 00¢ 00m 00m 00. 0.0 .533. h .- comm 2 a: o.:ln.+o«2«o._ -m an... «7.3 3.5- :oigozfi.-< ”10303.. .3 \II\ >5 :4- «*Eusgorré , / \ neon «#Euagoxozué {.0613 .9 3222.33 {0813. .m. b P 000. 0 00 000 00h 000 00m. 00¢ 00m 00m 00. .0 l 1:71 \ £22233 {enema .t b _ 4 7 2.3: mm coon 63.8: $633... ~oo~.no~3 .m. ms 9: o. .363 8 2:3: £6333 _ «oo~.no~3 .n. _ 000. 000 000 00k 000 000 00.. com 08 00. 0.0 m1 I< 1/ 28: cm .3. 3 2:5 {cues .8 1 1/\ Ill, 222503 n.vouo.«i .m. :3: v. .8 S 3.5 n. u . K a o 830 2 con 09. com 08 8. 0.0 000. 00m 000 00k. 000 xvii: 2...oI....N.0_ oooN .3 «.50: G. 00%. 00:0 {ouovuuz .3 3 :F , \/ g 33:8 .8 35 £2235 £6313 .mm _ / . r. l\ 4}] £822.38 mooaovuuz ._ m _ _ 000. com com 02. com com 00¢ 00» oou oo. o .o :3: .ml. on? 3 use»: .103sz «8.6:: .8 r J / 3301 mm 000” 5 0200! MAVONOoNuZ ~3N .635 .8 ,Iullll‘ \ I. .I / 1/ «.501 V 00¢” 50 23: m .8» 3 2:8: {oaomuz _ ~8m.no_«uz .em 80. com com co.“ can con 00¢ 00» com 00. o .o :30: 32 ES 8:» /\/ VN 000 035 €033...» .3 7\1 :3: o. .2. 35 3225.612 “106%.; .mm 000. \1 L 222533 {orgasm Nu _ . 5! com com 02. com coo 00¢ 00» con 00. o .o l\ i /\ Y i I. /\ a 9.301 O. own. $0.59 2225.» acne? .~n filI/g /‘ (‘11 Tl ‘ £823.33 :63; ._m I. i Ill :3: n. .03 S 3.8: Agueag «oomkoasa .om , _ 08. com com 02. com com ooc com com 8 . o .o n11 fl /1 :3: o~ .2503. 3.5 {as} .3 08. com coo oo» com com - ooc can oou 00. 000 1ICHIGAN STOTE UNIV. LIBRARIES llWWIWIIIWI(WIIIHIWWWI““IIWHIWI 31293200867632