I JoumaJ oJ 01 tho Nationol B u,onu oJ Slandrud, Vol. 53, No. 4, October 1954 Research Paper 2542 Separation of Titanium, Tungsten, Molybdenum, and Niobium by Anion Exchange John L. Hague , Eric D. Brown, and Harry A. Bright Th e res ul.ts of a s tud y of elution of titanium , tungste n, molyb- de num , and 11l0blUm In vanou s hy drochlonc-hydroflounc acid media ar e given. Th ese data th e behavi or . of th e anion of on an anion- exc han ge r eSIn The po sslb JI Jt y of severallllter es tJJl g quantItatIve separation s is ev iden t, and expe nments ar e des cnb ed t ha t demon s trate that t hese element s can be sepa ra te d from each oth er. I. Introduction One of th e more difficult problems of analyti cal chemistry is the separation of Litanium, niobium, and tantalum . Existing methods [1 , 2, 3, 5, 6]1 for the separation of niobium and tantalum arc not entir ely sati sfactory in the pr esen ce of titanium, and su ch methods ar e further complicat ed by additional separations when molybdenum and Lungsten are present. Krau s and Moor e [4] hav e shown that niobium and tantalum can be separated as mixed chloro-fluoro anion complexes by use of an anion exchange r esin in a hydrochloric-hydrofluoric acid medium. The r es ults of a systemati c s tudy of the elution behavior of titanium, tungsten, molybd e num, a nd niobium in some hydrochloric -hydrofluoric acid media ar e presented h er e, a well as a demons tration of the separation of these elements from each other . 2. Experimental Method 2.1. Pr eparation of Col umn The columnLised ill Lhe se rxperimenLs was pr e- par ed from a l2-in. length or polyst.yrene Lubin g with a %- in . insid e diameter. Th e bottom of the tube is clo sed by a waxed No. 5 rubb er stoppel' with a hole. A 6-in. length of polystyrene tubing with a outside diameter O{6 -in. bore) is inserted into the hole, flush with the upper surfaee of the stopper. Another 6-in . length of thi s tubing is atta ched to the smaller tube with a 2-in. length of T ygon R tubing, and the flow controlled by a pinch- cock on th e Tygon tubing. The bottom of th e large tube is covered with a to %-in. layer of acid- r esistant vinyl chloride plastic "wool ". The column is filled with prepar ed 2 200- to 400-me sh Dowex-l r es in, of 8- to 10-percent crosslinkage, to obt ain a set tled column of the r es in 8 in. high. A measure of th e liquid capa city of the column of resin is det er- mined by was hing with neu t ral ammon i um chloride (150 gjl it er , pH 6) until th e resin darkens, and th en I Fi gu res in br ackets indicate the literature references at the end of this paper. 2 r rhe finest and coarsest materials arc removed from the main portion of tho res in by settli ng in di lu ted llydroehl ori e acid (1 + 19), a nd se veral cycles of sue- ccssi vely decant ing th e fin e RJ ld main po rtions of the res in from the coarse fraction. ::11 47 15- tJ4--5 26 1 determining the volume of dilu ted hydro chloric acid (1 + 3) required to light en th e color of the r es in. The volume so determined was approximat ely 35 ml. 2 .2. Elution Procedure Acid mL,tures to co vel' the range of 5 to 20 pcrcenL by volume of con centrated hydrofluoric acid (48 %) and 10 to 60 p er cent by volume of c oncen tra ted hydrochloric acid (36 %) were used. To expedit e Lhe work:, the elemen ts were run in pairs, molybdenum and titanimn c ons tituting one pair, and tungs ten and niobium the other . As an example of the pro cedur e used, approxi- mat ely 20 mg of high-purity molybdic oxide and 20 mg of high-purit y titanium dioxide were t ran sferr ed to a platinum cru cible and fused with 0.5 g of potas tum bi sulfat e. Th e fusion wa di ss olved in 50 ml of a mixtur e of 1 volume of hydrofluoric acid , 2 volum es of hydrochloric acid, and 17 volmnes of wat er (d esignat ed 5-percen t hydrofluoric- 10-percent hydrochloric acid). Th e pr e pared column was washed with 50 ml of the same hydrofluoric-hydro- chloric acid mixture, and th e eluate discard ed. Th e solution of the fusion wa s trail fen'ed to the r esin , the first 25 ml of cluate being disc ard ed. Elu tion of th e r esin was then c ontinued with the hydrofluoric- hyorochloric acid solut ion, 50-ml fractions being coll ec ted in polystyrene graduat es for analysis. Suitable portions (usually 10 ml) taken from each 50-ml fraction as the el ution procedure was con- . tinued, were examined for titanium and molybdenum . Fluoride wa s eliminated from these portions by evaporation (usually three tim es) with sulfuric acid and hydrog en peroxide. Molybdenum wa s deter- mined by the thiocy anat e photometric me thod and titanium by th e peroxide photometric me thod [2] . Elution was continu ed until examination of th e fractions showed that the titanium and molybdenum were remov ed from the collll1ll. The amount of each eleme nt in each fraction examined was then cal- culat ed to a percentag e of the total amount of each elem ent eluted , and the acc umulat ed percentages were plotted against volume of eluate. The proce- dur e was then repeated wit h a solution of a fusion of tungsten and niobium oxides, the tungsten b eing determin ed by the hydroquinone photometric pro ce - dur e and niobium by the peroxide photometric pro ce dur e.