Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The fractional crystallization of the rare earth double nitrates Pearce, Denis W. 1930

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_1930_a8_p3_f7.pdf [ 18.84MB ]
JSON: 831-1.0060636.json
JSON-LD: 831-1.0060636-ld.json
RDF/XML (Pretty): 831-1.0060636-rdf.xml
RDF/JSON: 831-1.0060636-rdf.json
Turtle: 831-1.0060636-turtle.txt
N-Triples: 831-1.0060636-rdf-ntriples.txt
Original Record: 831-1.0060636-source.json
Full Text

Full Text

i U.S.C.  LiBRARY  CAT  THE FRACTIONAL CRYSTALLIZATION OF THE RARE EARTH DOUBLE NITRATES.  BY DENIS W. PEARCE, B. A.  A THESIS SUBMITTED FOR THE DEGREE OF MASTER OF ARTS  DEPARTMENT OF CHEMISTRY THE UNIVERSITY OF BRITISH COLUMBIA April, 1930.  ,  CONTENTS. 1.  INTRODUCTION  Page  Table of Series 2.  HISTORICAL (i) (ii) (iii) (iv) (v) (vi) (vii) (viii) (ix)  Double Double " " " " " Simple Double " "  1. 2. 2.  Mg Nitrates Mh Nitrates Co " and Zn " NHi " Fe " Pb " Nitrates Cu Nitrates and Cd " Ni "  3. 4, 5. 5* 5. 7. 7. 7. 9. 9. 9.  3.  MATERIAL  10.  4.  PROCEDURES  11.  (i) (ii) (iii) (iv) (v)  11. 13. 13. 14.  (vi) 5.  Preparation of series Fractionation Sampling Purification of samples Preparation of solutions for spectrophography Spectrometer and other apparatus  DISCUSSION OF PLATES (i) (ii) (iii) (iv) (v) (vi)  Double " " " " "  Magnesium Nitrates Cobalt " Ammonium " Lead " Manganese " Copper "  15. 17. 18. 18. 18. 19. 20. 20. 21.  6.  SUMMARY  22.  7.  CONCLUSIONS  23.  TABLE OF ILLUSTRATIONS  Standard Iron-arc Lines  XV]i  Chart of Absorption Bands  XVtt  Plate 1, series A " C  XV!)<  Plate 2, series D^ „  vv AX  Plate 3, series M " 0  vxt  THE FRACTIONAL CRYSTALLIZATION OF THE RARE EARTH DOUBLE NITRATES.  1. INTRODUCTION 1 The large amount of work involved in the discovery  2  and confirmation of Illinium - by the tedious fractionation of the double magnesium nitrates - followed by the use of the bromates on the neodymium - rich fractions - has led to this study of the comparative efficiencies of various double nitrates of the Rare Earths.  The hope was that some method  might be more efficient for the separation of Illinium from its plentiful neighbor, Neodymium, than was that using magnesium nitrate and that by the employment of it or by its combination with another method we might so effect a concentration (without recurrence to the bromates) as to allow us to recognize the strong lines in the absorption spectrum at A 5816 andX5123. While it was realized that the majority 3 of these methods had been studied by previous investigators it was nevertheless felt that a re-examination of them should be 1. 2. 3.  J.Allen Harris & B.S. Hopkings: Journ.Am.Chem.Soc. 48.1585.(1926) ibid 48.1594.(1926) Meyer,Schumacker, & Kotowski: Naturwiss.l4.771.(1926) Dehlinger,Glocker, & Kaupp: ibid 14.772.(1926) cf. later.  2. made in the light of the recent discoveries. It was decided that the following series should be run: 1.  A  2.  B  n  Mh  n  3.  C  M  Co  t!  4.  D  R  NHg  5.  F  H  6.  H  H  double Mg Nitrates  7.  L  double Pb Nitrates  8.  N  simple nitrates  9.  0  double Cu Nitrates  R  10.  P  M  Cd  M  Fe  M  11.  Q  n  Ni  !t  Ca  " It  12.  Z  n  Zn  *  !!  + A second Mn series was later started as "Series K". it The double calcium nitrates refused to crystallize more than twice. The rare earths were hence removeed from this series by precipitation with oxalic acid in a solution weakly acid with HNOg, thoroughly washed, and were later run as "Series F".  2. HISTORICAL. The use of the double nitrates for fractional crystallization of the Cerium group began in 1885 when Auer von Welsbach, by their use, succeeded in separating 4 dedymium into praseodymium and neodymium. It has been found that the tendency to form double nitrates varies with the basic strength of the Rare Earths in the most positive elements, i.e. the cerium group, the tendency is more pronounced than it is in the less positive yttrium group.  So while there are a large number of stable,  crystalline, double salts with the former elements their 4.  Welsbach: Monats. 6.477. (1885) Sitz.k.Akad.Wiss.Wien. 92.11.317. (1885).  3* stability decreases with an increase in atomic weight and past the terbium group crystallized double nitrates are difficult, if not impossible, to obtain. (i)  Double Magnesium Nitrates.  5  This method, made popular by Drossbach in 1902, has since been used by most investigators mainly to affect a preliminary separation of the Cerium earths into rough 6 divisions.  James advises starting with this method for the  reason that the most soluble portions crystallize much more readily as the 7 magnesium salts than as any others. Muthmann and. Weiss assign the formula 2M(N0g)g. 3Mg(N0g)g. 2HHg0. to the crystals and give the following as the order in which the earths crystallize:  Lanthanum, Praseodymium,  Neodymium, excess Mg(N0g)g, Samarium, Europium, Erbium, Gadolinium, Yttrium, and the remaining Yttrium earths; 8 these observations were checked by Feit and Przibylla. 9 The method has been much used: by Feit in his concentration  10  of Terbium, by Jantsch who compared its efficiency with the dbouble nitrates of Ni, Co, Zn, and Mh; in the discovery 5. Drossbach: Ber.35.2826. (1902). 6. James: Journ.American Chem.Soc. 30.979. ibid 34.757 (1912) ibid 38. 41 (1916) Chem.News. 97.205 (1908) Proc.Nat.Acad.Sci. 12.696 (1926) 7. Muthmann & Weiss: Liebig's Ann.331. 4 (1904) 8. Feit & Frzibylla: Zeit.Anorg.Chem.43.202 (1905) 9. ibid 43.267 (1905) 10. Jantsch: ibid 76.303 (1912)  4.  11  12  of Illinium; by Hopkins and Kremers in their preparation of pure cerium-group derivatives for arc spectraphotography 13 in the Bureau of Standards; and by others. (ii)  Double Manganese Nitrates. While looking for a quicker and more complete  separation of Praseodymium and Neodymium than that obtained by the use of the sparingly soluble alkali nitrates or of 14 the double ammonium nitrates, Urbain and Lacombe investigated the efficiency of the double manganese nitrates.  To  the double salt they assign the formula 2R(N0g)g . 3Mn(N0g)g 24HgO and strongly recommend its use in the above separation 15 has incorporated the use of the manganese compound James in his "Scheme for the Separation of the Rare Earths" and 16 has used it in his preparation of pure Praseodymium. The 17 18 ^ method has been used by Drossbach, and by Jantsch, the latter of whom has prepared the double nitrates of La, Pr, Nd, Sm, and Gd with Mg, Ni, Co, and Zn, and the corresponding salts of La, Pr, Nd, and Sm with Mn and has compared 11. Vide p. 1. 12. Kiess, Hopkings & Kremers: Bur.Stand.Sci.Paper #421, October (1921). 13. Demarpay: Compt.Rend. 130.1019. (1900) Brauner & Pavlieek: Proc.Chem.Soc. 17.63 (1901) ibid 81.1243 (1902) Demarpay: Compt.Rend. 130.1185 (1900) Demarpay: Compt.Rend. 131.343 (1900) ibid 132.1484 (1901) 14. Urbain & Lacombe: Bull.Soc.Chem.31.570 (iii) (1904) 15. James: Journ.Am.Chem.Soc. 34.757 (1912) James & Grant: ibid 38.41 (1916) etc. 16. James: Chem.News. 97.205 (1908) 17. Drossbach: Ber. 35.2826 (1902) 18. Jantsch: Zeit.anorg.Chem. 76.303 (1912)  their solubitities, densities, melting points etc.  He  states that Gadolinium does not form a crystallizable double manganese salt. (iii) Double Cobalt Nitrates & Double Zinc Nitrates. These compounds seem to have been prepared only 19 by Jantsch and in this case not to have been used for fractional crystallization but to have been made up from the pure nitrates of Lanthanum, Praseodymium, Neodymium, Samarium, and Gadolinium.  The double salts correspond to  the formulae 2R(N0g)g . 3M(N0g)g . 24HgO.  The solubilities,  densities, melting-points, and so on, for the various salts prepared hpve been determined by Jantsch. (iv) Double Ammonium Nitrates. The double ammonium nitrates were first used for the fractional crystallization of the Rare Earths by Mendelee^-in his separation of Lanthanium.  Welsbach found  that by crystallizing from nitric acid a much speedier separation resulted, and, by means of this modification,  21  resolved d&dymium into praseodymium and neodymium*  22  James  19. Jantsch: Zeit.anorg.Chem. 76.303 (1912) 20. MendelgefR St.Petersburg Acad.Sci.Bull.16.45.(1871) J.Russ.Chem.Ges. 5.119(1873) Liebig's Ann. 168.45.(1873) Ber. 6.558(1873) 21. Welsbach: Monatsh.Chem.VI. 477 (1885) Sitzber.K.Akad.Wiss.Wien.92 11.317(1885) 22. James: Chem.News. 97.205 (1908).  recommends the use of these salts for the LanthanumPraseodymium separation and calls attention to the fact that Cerium also, if present, concentrates with the former in the head fractions. Use of this observation had been 23 24 made by Welsbach, and by Schottlander, for the preparation of cerium.  Most of the cerium present goes into the  formation of the double eerie ammonium nitrate Ce(NOg)^ * 2NH^N0g which is much less soluble than the cerous salt and the corresponding salts of the other earths. Scheele's 25 26 modification, also used by Heyer and Marckwald, consisted in the addition of cerium - if some was not already presentfor the reason that the double cerous ammonium salt was intermediate in solubility between those of Lanthanum and Praseodymium.  The presence of Cerium greatly aids the  Praseodymium-Neodymium separation as well as in that of Lanthanum-Praseodymium. Since Cerium is so easily removed and since it acts in this way it is quite analogous to 27 28 Telement  separateur" of G. Urbain.  Esposito, by the use  of this method, obtained a practically complete separation 23. Welsbach: Monatsh.Chem. IV 630 (1883) V 508 (1884) 24. Schottlander: Ber. 25.1 378 (1892) 25. von Scheele: Zeit.Anorg.Chem. 17.310 (1898) ibid 18.352 (1898) Ber. 32.409 (1899) 26. Heyer & Marckwald: Ber. 33.3006(1900) 27. Urbain & Lacombe: Chem.News. 88.295.(1903) ibid 52.277 (1904) 28. Esposito: Proc.Chem.Soc. 22.20 (1906) ibid 23.64 (1907)  7 of Lanthanum, Praseodymium, and Neodymium. 29  Baxter and  Stewart used the double ammonium nitrates for the preparation of atomic weight Praseodymium. (v) Double Iron Nitrates. 30 Grant and James have reported their preparation of an unstable Lanthanum Ferous nitrate of the formulae 2La(N0g)g. 3Fe(N0g)2* 24HgO, and of a green color.  When  exposed to the air the green hexagonal crystals turn reddish-brown owing 31 to the formation of a basic ferric salt.  Jantsch in 1912 reported that there was nothing in  the literature previous to that time on the double nitrates of iron, copper or cadmium and that they had not been prepared by him.  No published account of a double iron  nitrate with the rare earths has been found between that time and the present with the exception of the article by Grant and James, who did not use the salt for fractional crystallization. (vi) Double Lead Nitrates. No references to the use of a double nitrate of the rare earths with lead could be found in the literature. (vii) Simple Nitrates. The simple nitrates of the Rare Earths have 29. Baxter & Stewart: Journ.Am.Chem.Soc. 37.516 (1915) 30. Grant & James: Journ.Am.Chem.Soc. 37.2652 (1915) 31. Jantsch: Zeit.Anorg. gRiem. 76.303.(1912)  8 been very greatly used for fractional crystallization.  By  systematic fractional crystallization of the mixed nitrates 32 from acid of a specific gravity of 1.3 - 1.54 Demarcay was able to concentrate "E", later isolated by him and 33 named europium.  According to Spencer the solubility of  the nitrates in nitric acid decreases with increasing atomic weight to a minimum, which is reached with gadolinium nitrate, after which it increases steadily with increasing atomic weight and reaches a maximum with Ytterbium nitrate.  The nitrate usually crystallizes as the  hexahydrate from aqueous solutions and as the pentahydrate from concentrated nitric acid. For the separation of the Terbium elements from one another the nitrate method has been 34 greatly used. For the Gadolinium-Terbium separation Urbain used this method, employing as well Bismuth Nitrate as "l'element separateur". This same investigator used this method for the preparation 35 of gadolinium, proved to be spectroscopically pure by 36 Eberhard. Pure Terbium compounds have been prepared by 37 Urbain by the fractional crystallization of the nitrates. 32. Demarcay: Compt.Rend. 122.728 (1896) 33. Spencer,J.F: "The Metals of the Rare Earths" (1919) p.26. 34. Urbain: Compt.Rend. 139.736 (1904) 35. Urbain & Lacombe:Compt.Rend. 140.583. (1905) 36. Eberhard: Zeit.Anorg.Chem. 54.374 (1905) 37. Urbain: Compt.Rend. 141.521 (1905) ibid 149. 37 (1909)  9 The nitrate crystallization has also been 38 employed by Urbain for the separation of Ytterbium and Lutecium, for the preparation of pure Ytterbium, and in the concentration of Celtium; and by Brocklemann for the 39 resolution of Didymium. (viii) Double Copper Nitrates and Double Cadmium Nitrates. Various double Rare Earth nitrates with copper 40 41 and Cadmium have been prepared by Carobbi who assigns to the Copper salt the formula 2R(N0g)g . 3Cu(N0g)g . 24BgO. The salts do not appear to have been used for fractional crystallization. (ix) Double Nickel Nitrates. The double nickel nitrates of the Rare Earths have been extensively used by Urbain in his identification 42 of the Z^ of de Botsbaadran with Terbium and in his prepar43 ation of atomic weight Gadolinium. 38. Urbain: Bull.Soc.Chim. XXXIII 739 (1905) Compt.Rend. 145.759 (1907) ibid 146.406 (1908) ibid 149.127 (1909) ibid 152.141 (1911) Chem.Zeit. 32.730 (1908) Zeit.Anorg.Chem. 68.236 (1910) Blumenfield & Urbain: Compt.Rend. 159.323 (1914) 39. Brocklemann: Annalen. 265.24 (1891) 40. Carobbi: Atti.r.Accad.Lincei.(V)33(ii)246.(1924) 41. Carobbi: Chem.Abst. 19.2174 (1925) 42. Urbain: Compt.Rend. 139.736 (1904) 141.521 (1905) 43. Urbain: Journ.Chim.Phys. 4.321 (1906) " Compt.Rend. 140.583 (1905)  10 44 Jantsch has prepared the double Nickle nitrates with Lanthanum, Praseodymium, Neodymium, Samarium, and Gadolinium, and has compared solubilities, densities and melting points.  He assigns the formula* 2R(N0g)g. 3Ni(N0g)g  24Hg0. 3. MATERIAL. The original material had its source in monazite sands and has been extracted for cerium and thorium by the Welsbach Mantle Company of Gloucester, New Jersey. All our oxides had been very kindly donated by Professor B. S. Hopkins of  the University of Illinois.  With the  oxides Dr. H. S. Miner, Chief Chemist for the Company, had supplied an outline of the process by which the monazite sands were treated.  The following is the main part of  his report: "Monazite Sand was decomposed with oil of Vitriol giving a mixture of the sulfates and the phosphates of the rare earths.  From an agueous solution of this mixture  the thoria was precipitated as a phosphate, leaving the other rare earths in solution.  This crude thoria, however,  contained approximately 25 per cent of other rare earths and was purified without any separation of these earths, the process being designed for the removal of other impurities.  After purification these associated rare earths  44. Jantsch: Zeit.Anorg.Chem.76.303 (1912).  11 were removed from the thoria and. were employed for the production of Cerium by oxidation with potassium permanganate.  ^he solution remaining, after the removal of Cerium, /  was treated with Magnesia and the Neodymium, Praseodymium, etc. precipitated leaving an impure Lanthanum in solution. "The Neodymium mixture which we sent you was prepared in the  manner above outlined and has a considerable  amount of the Cerium and Lanthanum removed from it.  You  will note from the above outline that this Neodymium material was not obtained from the original rare earth mixture remaining after the primary separation of the thoria, but rather was what accompanied the thoria until removal by the secondary separation."  4. PROCEDURES. (i) Preparation of the Series. The Rare Earth material was weighed out and dissolved in as little hot concentrated nitric acid as possible with the aid of hydrogen peroxide.  To the solution  was added the calculated amount of the divalent metal as solid nitrate or as nitrate in solution etc. (See Table One).  12 TABLE ONE Fractionally Series Crystallized as A  Amount RE20g  double Mg nitrates  Foreign Nitrate added as  5003* 675 gm. Mg(NOg)^  B  "  Mn  "  352  2.25 lbs. 50% Mn(N0g)^solu.  C  "  Co  "  500  640 gm. Co(NOg)g 6HgO  D  "  NBg  "  423  365 gm. NhgNOg  F  "  Fe  "  490  L  "  Pb  "  500  724.3 gm. FeSC4^°681 gm. Ba(NOg) + 736 gm. Pb(N0g)g 3  M  "  Mn  "  500  3.2 lbs. 50% Mn(N0g)g solu.  N  simple nitrates  500  0  double Cu nitrates  500  1110 gm. Cu(N0g)g.3Hg0  .  P  "  Cd  "  500  1380 gm. Cd(N0g)2*4Hg0  Q  "  Ni  "  475  1200 gm. Ni(N0g)^.6H^0  z  "  Zn  "  500  182 gm. Zno diss. HNOg *  *The FeS04.7H20 was dissolved in water and the solution of Ba(N0g)2 slowly added with stirring, keeping the Fe in slight excess. The BaSO^ was removed by filtration, carefully washed and the washings added to the filtrate. Further additions of dilute Ba(N0g)2 were made until no precipitate resulted. Whhn as free from both Ba*+ and S 0 4 ' as possible the Fe(N0g)2 solution was added to that of the Rare Earth nitrates. The Fe*+ solution was immediately oxidized to Fe20g as evidenced by the bulky yellow brown precipitate. This latter was removed by filtration, extracted with hot water, the washings combined with the solution and the whole re-evaporated and recrystallized. "Due to an error in calculation only one half the requisite quantity of zinc IE to form 2R(N0g)g":T°"'24 &g0 was added. Good crystallization was, however, always obtained.  13 (il) Fractionation. The system of fractional crystallization used in each series was identical.  In all cases the most soluble  portion was built up by five recrystallizations and pourings of the earlier fractions (by which time all were of approximately the same size) before it (the soluble end fraction) was crystallized and a small portion of liquor removed to form the bases of the next fraction. Fractions were crystallized in large porcelain dishes, smaller casseroles, and finally in pyrex flasks as their progressively decreasing size demanded for ease in handling. (iii) Sampling. Samples were taken from the various series at regular intervals in the operation.  Care was taken to  see that complete solution was always obtained and the fraction well mixed by shaking before the sample was withdrawn.  Exception to this procedure was only made when there  was danger of the fraction becoming gummy and difficulty crystallisable or when decomposition due to overheating was to be feared.  In these cases amounts of the crystals and  the liquor were secured as nearly as possible proportional to their quantity in the fraction.  14 iiv) Purification of the Samples. In the cases of the double Mg, Mn, Cd, Fe, and Zn salts the Rare Earth oxalates were precipitated in the presence of the ions of the base metal and freed of it by thorough washing. This procedure was not feasible, however, in the presence of Cu, Ni, Co, or Pb either due to occlusion of salts of these metals by the insoluble oxalates or (in the case of lead) to an oxalate which is insoluble under oertain conditions. The solutions were freed of Co and Nb by the precipitation of the latter as sulfides in ammoniacal solution (i.e. in a suspension of the Rare Earth hydroxides). The latter were then dissolved by acidification with hydrochloric acid and the NiS and CoS removed by filtration. The process had to be repeated several times to sufficiently remove Cobalt and Nickel so that they would not interfere with the oxalate pruification. Copper and lead iohs were removed by saturation of the slightly acid solution with H2S and subsequent filtration. In the cases of the simple nitrates and the double ammonium nitrates care was taken to reduce solubility 45 of the oxalates to a minimum by suitable dilution of the 45. Sarver & Brinton: Journ.Am.Chem.Soc.49.943 (1927) ibid 50.950 (1928) Baxter & Dandt: Journ.Am.Chem.Soc. 30.563 (1908)  15 solutions in which they were precipitated. (v) Preparation of Solutions for Spectrophotography of the Absorption. As mentioned previously, representative samples from each series were taken at regular integvals.  When  freed from the base metal, the Rare Earths from these solutions were precipitated as oxalates and thoroughly ignited to oxides. Solutions of 5 c.c. volume, 2.5 Normal with respect to Rare Earths and Normal with respect HNO3 were made up from each sample.  In all cases .683 gm. oxide  was dissolved  in 1.1 c.c. concentrated HNOg (with the aid of a few drops of HgOg.) and made up to 5 c.c.* These solutions were set aside and will form the bases of another paper on the progress of the separations given by the various double nitrates investigated. In addition to the small samples mentioned above larger samples were taken from each fraction of each series after 42 fractionations had been made. investigate the distribution of  It was decided to  the earths over the various  fractions at this point with the idea in mind of discovering whether or not any of the base metal added was acting as *The formation of Pr^O?, SaO and other abnormal oxides was ignored - the effect on the normality of the solutions would be small and would affect the absorption bands to a small degree. The assumption of 140 as an average molecular weight would also result in a small error in some cases.  15 solutions in which they were precipitate!. (v) Preparation of Solutions for Spectrophotography of the Absorption. As mentioned previously, representative samples from each series were taken at regular integvals.  When  freed from the base metal, the Rare Earths from these solutions were precipitated as oxalates and thoroughly ignited to oxides. Solutions of 5 c.c. volume, 2.5 Normal with respect to Rare Earths and Normal with respect HNO3 were made up from each sample.  In all eases .683 gm. oxide  was dissolved  in 1.1 c.c. concentrated HNOg (with the aid of a few drops of HgOg.) and made up to 5 c.c.* These solutions were set aside and will form the bases of another paper on the progress of the separations given by the various double nitrates investigated. In addition to the small samples mentioned above larger samples were taken from each fraction of each series after 42 fractionations had been made. investigate the distribution of  It was decided to  the earths over the various  fractions at this point with the idea in mind of discovering whether or not any of the base metal added was acting as *The formation of Pr^O?, SaO and other abnormal oxides was ignored - fhe effect on the normality of the solutions would be small and would affect the absorption bands to a small degree. The assumption of 140 as an average molecular weight would also result in a small error in some cases.  16 an "element separateur" and in any case the spectroscopic examination would give much information as to the most advantageous points at which to split the fractions in order to get Neodymium-rich or Samarium-rich portions. Accordingly, then, these final fractions were put into a water bath and brought to a temperature of 25°C when 10 c.c. volumes of the saturated solutions were removed with a pipette. These samples were purified in the same ways as mentioned before, the Rare Earth oxalates precipitated and ignited to the oxides.  The oxides were moistened with  distilled water and dissolved in concentrated nitric acid with the aid of a little HgOg.  The solutions were evaporated  just to dryness several times and finally taken up in 5 c.c. distilled water. 46  As pointed out by Quill, Selwood and  Hopkins the nitric acid could not be completely expelled from those samples which contained relatively more of the less basic earths taking place.  Sa,Eu, Gd, etc.  without hydrolyses  The dispersion of the spectrometer used by  ourselves being quite small, the concentration of acid which was,of necessity, left in the solutions should  not affect  the positions or widths of the absorption bands on our plates to any appreciable degree. 46. Quill, Sehwood, and Hopkins: Journ.Am.Chem.Soc. 50.2929 (1928).  17 (vi) Spectrometer, etc. All solutions were photographed through a 5 mm. cell with quartz sides using Wratten and Wainright Panchromatic plates (emulsion no. 5190).  The spectrometer  used was a Hilger D77 constant deviation model with a dispersion of a little over one Angstrom.  A 300 watt tung-  sten filament lamp was used as source for the continuous spectrum. With the apparatus at hand it was found impossible to photograph the iron arc in juxtaposition with each absorption photograph - due mainly to movement of the plate and instrument when the arc was being re-focused each time. For this reason, then, one picture only of the iron arc was included on each plate.  It was realized that for the  accuracy with which the absorption bands were to be measured such an arrangement would give the information desired. Figure 1 shows the 33 lines in the iron arc which were used as standards- as taken from a photograph by Hilger and identified on our plates by the use of a cathetometer. The wave lengths of the absorption bands of the Rare Earths, as shown in Figure 2, were those taken from a 47 48 chart plotted by Harris from a bibliography by Yntema.  47. 48.  Harris & Hopkins: Journ.Am.Chem.Soc.48.1593.(1926). Yntema: ibid 45. 907.(1923).  o a o -H cC o -<-) S-t  tO M *  #  !> >  o O ^O E-C^ rWOOE-tD EO * r-4 * * ** E- CO O <-D L CO H tOc-^o o o M EO M M M tQ tO lOc-c-O <-) M to to to to to to ^  d -w  EQ ^ tO t>-CO cn O r-) <Mto M M M M M M toK) EOCO  o CO <d 0) 03 3 a o o Si o O B3 43 t-< ai a3 o <3) 03 ^ (D 03 3 P^ <D tO -R to (0  M > aj t3 c-M O ccr< r^ ca)M to ^ cr<CO as r-!oO to 03 to H -sf iOtD ^D C-- CO CO o ^ ^ ^ ^ to to ^ tO  M tO to ^o CO o H M r-) r-r t-4 .-) r^<-) r-1M M  O ,3 CO c- toiD O toc-P- to to tO COO tO COr-{r-fto to to M C- tO 00 r-C)O C-M 00 C- COO O r-!r-<cu toto ^ to to  . . .E-CO  H M to to  o r-< r-<  XV!!.  XV!!.  18 5. DISCUSSION OF PLATES. (i) Double Magnesium Nitrates. Plate 1 A  illustrates the way in which the Rare  Earths have distributed themselves - under the influence of fractional crystallization of the double magnesium nitrates - in the eight fractions of the series. The first fraction is almost wholly devoid of absorption bands - showing the removal of the majority of the Praseodymium and Neodymium.  A faint shadow between  arc lines 10 and 12 (which is more clearly seen on the negative) shows the presence of a small amount of Praseodymium.  This strong Praseodymium band atX4440 - which  covers one of the weaker Illinium lines at 3^4420 and also a weak Samarium line - persists throughout the remaining seven fractions although stronger in the intermediate four. Neodymium appears to be mainly concentrating in the sixth fraction - as shown by the strong bands atX5800, and A. 5220.  The main Samarium band at\4000 shows the con-  centration of this element mainly in the three most soluble fractions - six, seven, and eight - to the largest extent in the sixth.  This band - part of which may belong to  Dysprosium can also be seen in the fifth fraction. (ii) Double Cobalt Nitrates. Plate 1 C  shows the absorption spectra obtained  from saturated solutions taken from the various fractions  ' 'i ^ t^t ' I < T( PC"!  S  a  Pfole xvm.  !.  19 in this series. of plates 1 A  As far as can be judged by a comparison and 1 C  there is exceedingly little  difference between the separations given by the two methods. Fraction four of the C series seems to contain a slightly larger amount of Neodymium than does the corresponding fraction in series A - judging by the thickness of the A5800 band.  Another slight difference noticeable is in  the main Neodymium band in fraction three.  In series A  there is a considerable width of very faint absorption toward the red from this position which is lacking in series C. (iii) Double Ammonium Nitrates. Plate 2 D  differs considerably from either of  the others previously mentioned.  The photographs do not  show any very great differences for the efficiencies of the methods for the concentration of Lanthanum. of series D  In the case  however it must be mentioned that the oxides  from fraction one were snow white, whereas, those from the same position in all other series were far from such.  These  differences would probably be strikingly shown if greater thicknesses of solution were used in the observing of the absorption spectra. The double ammonium salts seem to have aided considerably the Neodymium-Samarium separation.  It will be  noticed that the Neodymium bands atA5800 andA.4750 are  20. much decreased in strength in the last two fractions. Praseodymium also seems to be largely removed from these two - the band at A 4420 is almost totally absent here.  The  strong Samarium-Dysprosium band atX4000 is considerably strengthened in the last fraction - showing its concentrati The absorption to the red of the main Neodymium band in fraction three is here again noticeable. (iv) Double Lead Nitrates. In this series the majority of the lead nitrate crystallized by itself in the head fraction.  Forty-t'wo  fractionations removed the Rare Earths almost completely from this position - mone being obtained from a 10 c.c. saturated portion taken from fraction 1.  There is hence  no photograph representing fraction 1 on Plate 2 L. Very little separation appears to be resulting from the use of the lead salts.  Neodymium and Praseodymium  are probably concentrating in the earlier fractions to some extent and are certainly being removed from fraction 8 but little separation other than this appears to be taking place. (v) Double Manganese Nitrates. Plate 3 M  shows the distribution of the Rare  Earths over the eight fractions.  The photographs show that  the use of the manganese salts is having a considerable effect at both the soluble and the insoluble end of the  XX.  21 series.  The separation of Lanthanum from Neodymium and  Praseodymium, however, has not proceeded to the same extent as was the case with the magnesium nitrates but the separation of the latter two from Samarium has gone much farther.  As evidence of these facts it will be seen that  the Praseodymium band in fraction 1 and the Neodymium bands in fractions 2 and 3 are quite strong - while they are much weakened in fraction 8.  Praseodymium is con-  centrating in fraction 3, Neodymium in 4, 5, and 6, and Samarium in 7 and 8. (vi) Double Copper Nitrates. That the use of copper nitrate influences very greatly the solubility of the Rare Earth nitrates is seen by a consideration of plate 3  0.  It will be seen that  much Praseodymium and some Samarium remains in fraction 1; while not shown in the print, the original negative of this series also shows quite plainly the/^5800 Neodymium band in fraction 1.  There is clearly a concentration of Neodymium  and probably also of Praseodymium in fraction 2.  An out-  standing preculiarity of this series is the abundance of copper of fraction 6.  It is probable that in this fraction  a dif^,rent double salt has been crystallizing - differences in crystal habit had been noticed as the series drew near to the stage at which the final samples were drawn.  There  seems to be little separation of Samarium from Praseodymium  Plate 3. XX).  22 and Neodymium - the band at/^4000 due to the former is quite continuous from fractions 2 to 7. Work on the absorption spectra of the various fractions in the remaining series is being continued.  6. SUMMARY. 1. The following series of fractional crystallization of the Rare Earths have been run to 42 fractionations; double nitrates of Mg, Mn, Co, Zn, NH4, Pb, Cu, Cd, Ni; and simple nitrates. 2. Representative samples have been taken at various intervals from the series and have been made into solutions of 5 c.c. volume, 2.5 N  R(NOg)g  and N HNOg.  A study of the absorption spectra of these solutions will form the basis of a later paper. 3. From the final eight fractions of each series 10 c.c. saturated solutions at 25°C have been withdrawn, the Rare Earths purified and made into 5 c.c. solutions of the nitrate - as close to neutral as possible with the prevention of hydrolysis. 4. The absorption spectra of the above solutions have been photographed through a 5 mm. cell in the cases of the following double nitrate series: Mg, Mn, NH^, Co, Pb, Cu.  23 7.  CONCLUSIONS.  1. As far as can be observed at the present time the Rouble cobalt nitrates are as efficient as the double Magnesium nitrates for the fractional crystallization of the Rare Earths. 2. The double ammonium nitrates seem to hasten the Neodymium-Samarium separation more than the other methods studied - up to the point at which this investigation was completed. 3. The double lead nitrates are not very efficient for fractional crystallization.  Excess lead separates out  in the head fraction. 4. The double manganese salts appear to give an excellent separation of the Cerium-group elements. 5. Copper may be acting as an "element separateur". The presence of copper nitrate greatly effects the relative differences in the solubilities of Praseodymium and Neodymium nitrates.  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items